15
A nucleosome assembly protein-like polypeptide binds to chloroplast group II intron RNA in Chlamydomonas reinhardtii Stephanie Glanz, Astrid Bunse, Andrea Wimbert, Carsten Balczun and Ulrich Ku ¨ ck* Lehrstuhl fu ¨ r Allgemeine und Molekulare Botanik, Ruhr-Universita ¨t Bochum, D-44780 Bochum, Germany Received May 2, 2006; Revised and Accepted August 4, 2006 ABSTRACT In the unicellular green alga Chlamydomonas rein- hardtii, the chloroplast-encoded tscA RNA is part of a tripartite group IIB intron, which is involved in trans-splicing of precursor mRNAs. We have used the yeast three-hybrid system to identify chloroplast group II intron RNA-binding proteins, capable of interacting with the tscA RNA. Of 14 candidate cDNAs, 13 encode identical polypeptides with sig- nificant homology to members of the nuclear nucleosome assembly protein (NAP) family. The RNA-binding property of the identified polypeptide was demonstrated by electrophoretic mobility shift assays using different domains of the tripartite group II intron as well as further chloroplast transcripts. Because of its binding to chloroplast RNA it was designated as NAP-like (cNAPL). In silico analysis revealed that the derived poly- peptide carries a 46 amino acid chloroplast leader peptide, in contrast to nuclear NAPs. The chloro- plast localization of cNAPL was demonstrated by laser scanning confocal fluorescence microscopy using different chimeric cGFP fusion proteins. Phylogenetic analysis shows that no homologues of cNAPL and its related nuclear counterparts are present in prokaryotic genomes. These data indicate that the chloroplast protein described here is a novel member of the NAP family and most probably has not been acquired from a prokaryotic endosymbiont. INTRODUCTION Group II introns have been detected in eukaryotic organelles as well as in eubacterial genomes as intervening sequences of protein, tRNA or rRNA coding genes. Common to all group II introns is a conserved secondary structure that consists of six double-helical domains (DI-DVI) radiating from a central wheel (1). Group II introns splice via two sequential transesterifications that in vitro occur in some cases auto- catalytically. In vivo, however, splicing is dependent on protein co-factors as was shown by mutant analyses (2,3). In Chlamydomonas reinhardtii chloroplasts, two trans- splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case of the first intron of the psaA gene, three independently tran- scribed RNAs associate via tertiary interactions to form a functional group II intron, resulting in trans-splicing of the flanking exon 1 and exon 2 (5). Part of this tripartite group II intron is the tscA RNA that is processed from a chloroplast- encoded precursor RNA. Secondary structure predictions revealed that tscA contains domain DII and DIII as well as partial domains DI and DIV of the conserved secondary group II intron structure (6). Mutant work in C.reinhardtii has shown that >14 nuclear genes affect the chloroplast trans-splicing reaction, and complementation of some of the characterized mutants led to the identification of the molecular determinants. Although some of the identified polypeptides are related to proteins involved in nucleic acid metabolisms, this function seems to be non-essential for intron splicing (7,8). Similar to nuclear mRNA splicing, these protein components might be part of a chloroplast splicing complex, because they were found in large stromal or membrane-bound RNA-containing complexes (9–11). To define further components of a putative ‘chloroplast spliceosome’ in C.reinhardtii, we used the tscA RNA as bait to identify novel intron RNA-binding proteins. Using the yeast three-hybrid system, we isolated a polypeptide that seems to be a novel member of the multifunctional nucleosome assembly protein (NAP) family. These well- conserved eukaryotic histone chaperones facilitate, for example, the nucleosome assembly and remodelling of chromatin, and are implicated in transcriptional regulation and cell cycle regulation (12,13). Here, we assign a novel function to a NAP-like protein. Laser scanning confocal fluorescence microscopy (LSCFM) and in vitro binding *To whom correspondence should be addressed. Tel: +49 234 3226212; Fax: +49 234 3214184; Email: [email protected] Present address: Astrid Bunse, Qiagen GmbH, Qiagen Straße 1, D-40724 Hilden, Germany Ó 2006 The Author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Published online 29 September 2006 Nucleic Acids Research, 2006, Vol. 34, No. 18 5337–5351 doi:10.1093/nar/gkl611

A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

A nucleosome assembly protein-like polypeptidebinds to chloroplast group II intron RNA inChlamydomonas reinhardtiiStephanie Glanz Astrid Bunse Andrea Wimbert Carsten Balczun and Ulrich Kuck

Lehrstuhl fur Allgemeine und Molekulare Botanik Ruhr-Universitat Bochum D-44780 Bochum Germany

Received May 2 2006 Revised and Accepted August 4 2006

ABSTRACT

In the unicellular green alga Chlamydomonas rein-hardtii the chloroplast-encoded tscA RNA is part ofa tripartite group IIB intron which is involved intrans-splicing of precursor mRNAs We have usedthe yeast three-hybrid system to identify chloroplastgroup II intron RNA-binding proteins capable ofinteracting with the tscA RNA Of 14 candidatecDNAs 13 encode identical polypeptides with sig-nificant homology to members of the nuclearnucleosome assembly protein (NAP) family TheRNA-binding property of the identified polypeptidewas demonstrated by electrophoretic mobility shiftassays using different domains of the tripartitegroup II intron as well as further chloroplasttranscripts Because of its binding to chloroplastRNA it was designated as NAP-like (cNAPL)In silico analysis revealed that the derived poly-peptide carries a 46 amino acid chloroplast leaderpeptide in contrast to nuclear NAPs The chloro-plast localization of cNAPL was demonstrated bylaser scanning confocal fluorescence microscopyusing different chimeric cGFP fusion proteinsPhylogenetic analysis shows that no homologuesof cNAPL and its related nuclear counterpartsare present in prokaryotic genomes These dataindicate that the chloroplast protein described hereis a novel member of the NAP family and mostprobably has not been acquired from a prokaryoticendosymbiont

INTRODUCTION

Group II introns have been detected in eukaryotic organellesas well as in eubacterial genomes as intervening sequences ofprotein tRNA or rRNA coding genes Common to all groupII introns is a conserved secondary structure that consists of

six double-helical domains (DI-DVI) radiating from acentral wheel (1) Group II introns splice via two sequentialtransesterifications that in vitro occur in some cases auto-catalytically In vivo however splicing is dependent onprotein co-factors as was shown by mutant analyses (23)

In Chlamydomonas reinhardtii chloroplasts two trans-splicing introns were found in the psaA gene encoding themajor P700 chlorophyll ab-binding protein (4) In the caseof the first intron of the psaA gene three independently tran-scribed RNAs associate via tertiary interactions to form afunctional group II intron resulting in trans-splicing of theflanking exon 1 and exon 2 (5) Part of this tripartite groupII intron is the tscA RNA that is processed from a chloroplast-encoded precursor RNA Secondary structure predictionsrevealed that tscA contains domain DII and DIII as well aspartial domains DI and DIV of the conserved secondarygroup II intron structure (6)

Mutant work in Creinhardtii has shown that gt14 nucleargenes affect the chloroplast trans-splicing reaction andcomplementation of some of the characterized mutants ledto the identification of the molecular determinants Althoughsome of the identified polypeptides are related to proteinsinvolved in nucleic acid metabolisms this function seemsto be non-essential for intron splicing (78) Similar to nuclearmRNA splicing these protein components might be part ofa chloroplast splicing complex because they were found inlarge stromal or membrane-bound RNA-containingcomplexes (9ndash11)

To define further components of a putative lsquochloroplastspliceosomersquo in Creinhardtii we used the tscA RNA asbait to identify novel intron RNA-binding proteins Usingthe yeast three-hybrid system we isolated a polypeptidethat seems to be a novel member of the multifunctionalnucleosome assembly protein (NAP) family These well-conserved eukaryotic histone chaperones facilitate forexample the nucleosome assembly and remodelling ofchromatin and are implicated in transcriptional regulationand cell cycle regulation (1213) Here we assign a novelfunction to a NAP-like protein Laser scanning confocalfluorescence microscopy (LSCFM) and in vitro binding

To whom correspondence should be addressed Tel +49 234 3226212 Fax +49 234 3214184 Email ulrichkueckruhr-uni-bochumdePresent addressAstrid Bunse Qiagen GmbH Qiagen Straszlige 1 D-40724 Hilden Germany

2006 The Author(s)This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (httpcreativecommonsorglicensesby-nc20uk) which permits unrestricted non-commercial use distribution and reproduction in any medium provided the original work is properly cited

Published online 29 September 2006 Nucleic Acids Research 2006 Vol 34 No 18 5337ndash5351doi101093nargkl611

assays demonstrate that chloroplasts of Creinhardtii containa NAP-like protein that specifically binds to organellar groupII intron RNA and U-rich chloroplast transcripts Our datafurther support the view that during evolution chloroplastshave acquired novel nuclear components that most probablywere not delivered by endosymbiont gene transfer (14)

MATERIALS AND METHODS

Strains culture conditions and transformation

Creinhardtii cell wall depleted arginine auxotroph strainargb-ecw15 as well as wild-type strain CC406 cw15 mt-were obtained from the Chlamydomonas Center (Duke Uni-versity Durham NC USA) and were grown as describedon Trisndashacetatendashphosphate (TAP) medium (15) Whenrequired the medium was supplemented with 50 mg arginineper ml Nuclear transformation was carried out using theglass bead method (16) and was performed according to (8)

with the following modifications A total of 1ndash3 middot 106

cellsml were incubated with 1ndash5 mg supercoiled or SpeIdigested pMF59 (17) plasmid DNA After transformationcells were spread on solid TAP medium and incubated in con-tinuous light at 25C For selection of transformants carryingpMF59 medium contained 5 mgml of zeocin (InvitrogenKarlsruhe Germany)

Recombinant plasmids

All recombinant plasmids used for in vitro transcription PCRanalysis protein synthesis yeast three-hybrid screeninghybridization or generation of transgenic algal strains arelisted in Table 1 Sequences of all oligonucleotides aregiven in Table 2

Molecular biological techniques

Procedures for standard molecular techniques were per-formed as described previously (18) Escherichia coli strain

Table 1 Plasmids used in this work

Plasmid Characteristics Reference

Yeast three-hybridpADRevM10 pACTII derivative expression of fusion protein GAL4-AD and HIV1 RevM10 (21)pDBRevM10 Yeast shuttle vector (TRP1 LEU2) cloning of hybrid RNA transcription unit cassette

of pPGKRRE and expression cassette of fusion protein GAL4-DB and HIV1 RevM10(21)

pGAD10 derivatives cDNA fragments of Creinhardtii CC406 cw15 in pGAD10 (BD Biosciences Clontech) (20)pPGKRRE Intermediate to clone hybrid RNAs as a transcription unit cassette into pDBRevM10 (21)pRev30Lhcb Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and Lhcb1 30-UTR (pBQ24) (37) (8)pRevpsaADV+VI Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and domains DV+VI of psaA intron 1 This workpRevR2 RREndashRRE transcription unit cassette (21)pRevrI1DIV-VI Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and domains DIV-VI of intron rI1

of Sobliquus (36)(8)

pRevtscA1 Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and processed tscA RNA (6) (8)cNAPL cDNA

pGAD6-19 16 kb cNapl cDNA fragment in pGAD10 This workEMSA

pQE70-cNAPL 1 kb SphIBglII coding sequence of cNAPL in pQE70 with His6 tag (Qiagen) This workpdI 347 bp fragment of domain DI of intron rI1 of Sobliquus in pT3T7BMSmaI This workpdII+III 139 bp fragment of domain DII and DIII of intron rI1 of Sobliquus in pT3T7BMSmaI This workpdIV 57 bp fragment of domain DIV of intron rI1 of Sobliquus complementary oligonucleotides

in pT3T7BM(22)

pdV+VI 73 bp fragment of domain DV and DVI of intron rI1 of Sobliquus in pT3T7BMSmaI (22)p734 211 pb fragment of rps4 50-UTR in plasmid pBIIKS+ (22)pT3T730UTRLhcb1T7 109 nt EcoRIBamHI fragment of Lhcb1 30-UTR in pT3T7BM This workpT3T7psaADV+VIT7 400 nt MluI fragment of pRevpsaADV+VI in pT3T7BMSmaI (Boehringer

Mannheim Germany)(23)

pT3T7Rps18T7 136 nt EcoRIBamHI fragment of exon 2 of Rps18 (46) in pT3T7BM This workpT3T7Tba1AT7 127 nt EcoRIBamHI fragment of exon 3 of Tba1A in pT3T7BM This workpT3T7tscADIIT7 156 nt EcoRIBamHI fragment of domain DII of tscA RNA in pT3T7BM This workpT3T7tscADIIIT7 193 nt EcoRIBamHI fragment of domain DIII of tscA RNA in pT3T7BM This workpT3T7tscADII+IIIT7 338 nt EcoRIBamHI fragment of domain DII and DIII of tscA RNA in pT3T7BM This work

LSCFMpCr1 HSP70ARBCS2 promoter (pCB740) (49) and Lhcb1 30-UTR terminator ASL for selection This workpCr1g 07 kb NheIBglII fragment of cgfp (47) in pCr1 This workpCr1g_cNaplN84 03 kb NheI fragment N-terminal 84 amino acids of cNAPL in pCr1g encoding for

fusion protein N84-cGFPThis work

pCr1g_cNaplN84D8-39 02 kb NheI fragment N-terminal 84 amino acids of cNAPL without targeting signalin pCr1g encoding for fusion protein N84D8-39-cGFP

This work

pCr1g_Rps18 05 kb NheI fragment of Rps18 coding sequence in pCr1g encoding for fusionprotein Rps18-cGFP

This work

pGAD-Rps18 Rps18 cDNA fragment in pGAD424 (BD Biosciences Clontech) This workpMF59 Encoding for fusion protein Ble-cGFP (17)

AD DNA activation domain ASL arginiosuccinate lyase DB DNA-binding domain EMSA electromobility shift assay LSCFM laser scanning confocalfluorescence microscopy

5338 Nucleic Acids Research 2006 Vol 34 No 18

XL1-blue MRF0 served as host for general plasmid cons-truction and maintenance (19) The Creinhardtii cDNAlibrary contains double-stranded cDNA fragments withEcoRI adaptors cloned into the EcoRI site of pGAD10 inDH10B (BD Biosciences Clontech Heidelberg Germany)(20) pGAD10 derivatives were used in the yeast three-hybridscreen Creinhardtii total RNA was prepared according to(4) and RNA blot experiments were carried out accordingto (8) Northern analyses were performed with a radioactivelylabelled 1690 bp pGAD6-19EcoRI cNapl-specific probeand a 460 bp pGAD-Rps18EcoRIndashBamHI Rps18-specificprobe cDNA was prepared using Omniscript ReverseTrancriptase (Qiagen Hilden Germany) and purified withNucAway Spin Columns (Ambion Austin TX USA)according to the manufacturerrsquos protocols For PCRs Taqpolymerase (Eppendorf Hamburg Germany) Hot MasterTaq DNA Polymerase (Eppendorf) and the Expand LongTemplate PCR System (Roche Mannheim Germany) wereused Oligonucleotide primers applied for conventional PCRare listed in Table 2

The yeast three-hybrid system

The yeast three-hybrid analysis was performed as describedpreviously (8) Plasmids are listed in Table 1Plasmids pADRevM10 and pRevR2 were generous giftsfrom Dr U Putz (University of Hamburg) (21) PlasmidpRevtscA1 (8) was used to screen a Creinhardtii cDNAlibrary (Table 1)

Electrophoretic mobility shift assays (EMSAs)

Recombinant plasmid pQE70-cNAPL (see above) was usedfor synthesis of cNAPL in Ecoli Recombinant His-taggedcNAPL protein was purified from Ecoli with Ni2+-NTA-agarose resin according to the manufacturerrsquos protocols(Qiagen) For RNA mobility shift assays uniformly32P-UTP-labelled run-off transcripts served as substrateRNAs and were generated by in vitro transcription of plas-mids as given in Table 1 In vitro transcription and EMSAswere performed as previously reported (2223) Unlabelledcompetitor RNAs and non-specific competitor RNA deri-ved from plasmid pBIIKS+ were synthesized as described(22) DNA mobility shift assays were done essentially asdescribed previously (24) with the following modificationsThe probes were amplified by PCR with appropriate plasmidDNA or genomic DNA of strain arg-cw15 in the presence of[32P]dATP The labelled probes were purified by denaturingPAGE Binding reactions were carried out by incubating30 fmol of the probes with 15 mg of recombinant cNAPLThe reaction mixtures were separated on a 5 non-denaturing polyacrylamide gel

Laser scanning confocal fluorescence microscopy

The fluorescence emissions of transformed Creinhardtiicells were analysed by LSCFM using a Zeiss LSM510 META microscopy system (Carl Zeiss JenaGermany) based on an Axiovert inverted microscopecGFP and plastids were excited with the 488 nm line of an

Table 2 Oligonucleotide primer pairs used in PCR and RTndashPCR experiments

Oligonucleotide Specific sequence (5030) Specificity

2428_NheIBglIISmaI GCT AGC AGA TCT CCC GGG TTT GTA TGA GCG TAT AAG CTC TGC 50-region of Lhcb1 30-UTR2429_EcoRV GAT ATC ATC GAT GGC CTA CCC AAA CCC C 30-region of Lhcb1 30-UTRfor_6-19_BspHI TAT TCA TGA TGG CAC TAG CTC TGC TCA CTC G cNapl RTndashPCRrev_6-19_2xMet_HindIII ATA AAG CTT CTA CAT CAT GTA CTC GCC CTC CTC GTA CTC G cNapl RTndashPCRfor_6-19_ SphI GCA TGC GCA TGG CAC TAG CTC TGC TCA C cNaplrev_6-19_BglII AGA TCT GTA CTC GCC CTC CTC GTA C cNaplfor_ATG_cgfp CTA GCT AGC ATG GCC AAG GGC GAG GAG cgfprev_cgfp GAA GAT CTT TAC TTG TAC AGC TCG TCC ATG C cgfpfor_cNapl_N84 CTA GCT AGC ATG GCA CTA GCT CTG CTC AC cNaplN84rev_cNapl_N84 CTA GCT AGC GGC GGT GTT GAG CTC CTC cNaplN84 cNaplN84D8-39for_cNapl_N84D8-39 CTA GCT AGC ATG GCA CTA GCT CTG CTC ACT TTC CTG

GCA GTG GTT GGC GCCcNaplN84D8-39

for_30Lhcb_EcoRI CCA TCA AGG ACC AGG TCA TC Fragment of Lhcb1 30-UTRrev_30Lhcb_BamHI AGG AAT TCA AGA GCC CAT TCC AAA GTC Fragment of Lhcb1 30-UTRLhcbm6 T7-1 GTA ATA CGA CAT CAC TAT AGG GCC TGT TTG GCG CTT Fragment of Lhcb1 50-UTRLhcbm6-2 CAT TCC TGC ACG CTC GCT TG Fragment of Lhcb1 50-UTRT7-psbA50 GTA ATA CGA CTC ACT ATA GGG TAC CAT GCT TTT AAT Fragment of psbA 50-UTRpsbA30-2054 GAT CCA TGG TCA TAT GTT AAT TTT TTT AAA G Fragment of psbA 50-UTRsu3131 TGT GCG TTT CTC TTG ATA TGT ACC G Fragment of psbD 50-UTR2126 TAA TAC GAC TCA CTA TAG GGA CAC AAT GAT TAA AAT TAA A Fragment of psbD 50-UTRT7-rbcL50 GTA ATA CGA CTC ACT ATA GGG TAT GCT CGA CTG ATA AGA C Fragment of rbcL 50-UTRrbcL30 CTG CTT TAG TTT CTG TTT GTG GAA CC Fragment of rbcL 50-UTRfor_Rps18_NheI CTG CTA GCA TGG GCT CTC TCG TCC A Rps18rev_Rps18_NheI GAG CTA GCC TTC TTC TTG GCG ACA CC Rps18for_Rps18_EcoRI AGG AAT TCT CTG CGT CTG CTG AAC ACT AAC Fragment of exon 2 of Rps18rev_Rps18_BamHI CGG GAT CCA GGT CAA CCT CAG CCT TCT T Fragment of exon 2 of Rps18for_tscADII AGG AAT TCA AAT TTT TAG TAA TGT TAA AG Domain DII of tscA RNArev_tscADII CGG GAT CCC CTT AAT TTT TAG AAA ATG Domain DII of tscA RNAfor_tscADIII AGG AAT TCT AAA CAA CAA GTA ATG CC Domain DIII of tscA RNArev_tscADIII CGG GAT CCT AAA CAA TAA AGT TTT TCA A Domain DIII of tscA RNAfor_a1Tub_EcoRI AGG AAT TCC GGC TTC AAG TGC GGT ATC AAC Fragment of exon 3 of Tba1Arev_a1Tub_BamHI CGG GAT CCT CGC CGA TAG CAG TGC TGT T Fragment of exon 3 of Tba1A

Nucleic Acids Research 2006 Vol 34 No 18 5339

argon-ion laser The fluorescence emission was selected bybandpass filter BP505-530 and longpass filter LP560 respec-tively using beam splitters HFT UV488543633 andNFT545

Sequences alignments and phylogenetic analysis

Sequences were retrieved from GenBank (NCBI) at httpwwwncbinlmnihgov the Creinhardtii JGI database v20(DOE Joint Genome Institute) at httpgenomejgi-psforgchlre2chlre2homehtml the Thalassiosira pseudonana JGIdatabase v10 at httpgenomejgi-psforgthaps1thaps1homehtml the Porphyra yezoensis EST index at httpwwwkazusaorjpenplantporphyraEST (2526) and theCyanoBase genome database at httpwwwkazusaorjpcyano Multiple sequence alignments of NAPs were per-formed using ClustalW software (27) Sequence alignmentswere displayed with GENEDOC (28) and refined manuallyThe phylogenetic tree was generated based on trimmedprotein sequences with programs contained in the programpackage PHYLIP version 363 (29) An alignment withsequences from mature proteins was impossible since all pro-teins differ largely in length and some were only availablefrom partial cDNAs (AU191247 BM448458) obtainedfrom EST libraries Analyses with the PHYLIP packagewere performed using PROTDISTNEIGHBOR and thePROTPARS algorithms for distance matrix and maximumparsimony analysis respectively Support for the brancheswas estimated by bootstrap analysis of 2000 replicatesgenerated with SEQBOOT and a majority rule consensustree was generated using the program CONSENSEThe resulting tree topology was then visualized inTREEVIEW (30) For analyses of putative targeting signalsequences ChloroP (31) LOCtree (32) TargetP (33)GENOPLANTETM PREDOTAR (httpwwwgenoplantecom) and PSORT (34) were used Isoelectric pointand sequence masses were calculated by the programWinPep 301 (35)

RESULTS

Identification of a tscA-specific RNA-binding proteinusing the yeast three-hybrid system

To identify RNA-binding proteins most probably involved inorganellar group II intron splicing we performed an in vivoscreen using the yeast three-hybrid system In this systeman RNAndashprotein interaction is detected by the reconstitutionof the yeast transcriptional activator GAL4 and subsequentactivation of reporter genes The system is based on onehybrid RNA and two recombinant fusion proteins containingthe GAL4 DNA-binding and GAL4 transcription activationdomains respectively To isolate chloroplast intron RNA-binding proteins of Creinhardtii a cDNA library wasscreened using the yeast three-hybrid system developed by(21) In this system the chimeric protein RevM10 fused toa gene fragment encoding the DNA-binding domain ofGAL4 (GAL4-DB RevM10) binds to RRE which is partof a chimeric RNA The 30 part of this RNA consists of thetscA RNA which functions as bait during the screen (RRE-tscA) Chimeric proteins serve as prey and are encoded bycDNA clones of a Creinhardtii cDNA library All cDNAswere fused to a gene fragment encoding the transcriptionactivation domain of GAL4 (GAL4-AD) The reliability ofthis three-hybrid screen was tested with a recombinantRNA carrying two copies of RRE (RREndashRRE) co-expressedwith the two RevM10 fusion proteins (AD-RevM10DB-RevM10) RRE and RevM10 are known to interactwith each other (21) As depicted in lane 1 of Figure 1 theinteraction resulted in reconstitution of GAL4 activity

For a global Creinhardtii cDNA screen the yeast strainCG1945 was co-transformed with pRevtscA1 expressingboth the hybrid protein DB-RevM10 and the recombinantRNA This strain served as host for transformation of thelibrary of pGAD10-derived plasmids harbouring cDNAsequences of Creinhardtii A total of 21 middot 107 clones wereanalysed and 14 clones were identified to bind to tscA RNAInterestingly of these 14 candidate cDNAs 13 encoded

Figure 1 Schematic overview of selected constructs used in the yeast three-hybrid analyses In each case five representative yeast transformants were tested fortheir ability to activate b-galactosidase by a colony colour assay Abbreviations Lhcb30 fragment of 30-UTR of the Lhcb1 transcript (37) cNAPL chloroplastNAP-like GAL4-AD activation domain of GAL4 transcription factor GAL4-DB DNA-binding domain of GAL4 transcription factor psaADV+VI domainsDV and DVI of the first psaA intron (4) RevM10 RNA-binding protein RevM10 (21) rI1DIV-VI domains DIV to DVI of intron rI1 (36) RRE Rev responsiveelement that is specifically bound by RevM10 (21) tscA processed tscA RNA (6)

5340 Nucleic Acids Research 2006 Vol 34 No 18

identical polypeptides with significant homology to NAPsBecause of its binding to chloroplast intron RNA theNAP-like polypeptide was designated as cNAPL

To assess the specificity of the three-hybrid interactionbetween tscA RNA and cNAPL several controls were carriedout The specificity of the tscA interaction was tested againsttwo further hybrid RNAs Recombinant control RNAscontaining sequences of domains DIV to DVI of the hetero-logous intron rI1 (36) (RRE-rI1DIV-VI) and a fragment ofthe 30-untranslated region (30-UTR) of the Lhcb1 transcript(37) (RRE-Lhcb30) were fused separately to the RREsequence respectively As shown in lanes 5 7 and 8 ofFigure 1 activity of the reporter b-galactosidase is dependenton the presence of the RRE-tscA hybrid RNA and thecNAPL-AD fusion protein These results indicate a specificinteraction of cNAPL with the tscA transcript In additionwe were able to demonstrate that cNAPL also binds todomains DV and DVI of the first psaA intron (RRE-psaADV+VI) as seen in lane 6 (Figure 1) To exclude one-or two-hybrid interaction or unspecific RNA interactionappropriate negative controls were also performed in the filterassay including different combinations of the three-hybridcomponents represented in lanes 2ndash4 (Figure 1) In theabsence of any one of the hybrid components transformantsdisplayed no b-galactosidase activity Results of the three-hybrid analysis therefore clearly demonstrate the specificinteraction of cNAPL with tscA RNA and the 30 part of thefirst intron of psaA RNA

In vitro binding of cNAPL to chloroplast transcripts

To confirm the specific interaction between cNAPL and tscARNA in the yeast three-hybrid system we performed in vitrobinding studies using EMSA For synthesis of cNAPL thecNapl gene was fused with six histidine residues forexpression in Ecoli After purification of the recombinantHis-tagged protein with Ni2+-NTA chromatography cNAPLwas incubated with in vitro transcribed radioactively labelledRNAs corresponding to domains DII and DIII of tscA RNAAs shown in Figure 2a domains DII and DIII range from nt75 to 218 and from 220 to 400 respectively Together theycover nt 75ndash400 RNAs were incubated in the absence orpresence of recombinant cNAPL and fractionated on nativegels (Figure 2bndashi) In Figure 2bndashd increasing amounts ofcNAPL (2ndash20 mg) were incubated with 30 fmol of in vitrosynthesized transcripts of tscA intron RNA fragmentscomprising either domain DII domain DIII or both domainsDII+III Arrows indicate RNAndashprotein complexes as multipleshifted bands that can be observed when the RNA isincubated with increasing amounts of cNAPL polypeptide

To investigate the binding specificity of cNAPL competi-tion analyses were performed The above-mentioned radio-labelled RNAs were incubated with a constant amount of15 mg of cNAPL protein in the presence of increasingamounts of unlabelled specific and non-specific competitorRNAs respectively As shown in Figure 2fndashh the presenceof just a 2-fold molar excess of unlabelled specific competitorRNA almost completely abolished the formation of acomplex between cNAPL and the labelled target tscA RNAIn contrast the presence of the same amount of non-specific competitor RNA namely pBIIKS+ (121 nt) had no

significant effect on the formation of the complex with thelabelled target RNAs In addition EMSA was performedwith in vitro synthesized RNA of domains DV and DVI ofthe first psaA intron As shown in Figure 2a the two domainsrange from nt 108 to 192 EMSAs with increasing amountsof cNAPL protein (Figure 2e) and competition analysis(Figure 2i) indicate that cNAPL binds specifically to domainsDV and DVI of the first psaA intron albeit with lower affinitythan to tscA RNA

Chloroplast transcripts are rich in A and U residues Tostudy the sequence preferences we compared the ability ofA C G or U RNA homopolymers to compete for bindingof cNAPL to the labelled domain DII of the tscA RNAand domain DV+VI of the first psaA intron respectivelyAs depicted in Figure 3a excess amounts of poly(U)abolished cNAPL binding while the same amounts ofpoly(C) or poly(G) have no significant effect on the formationof the complex Using poly(A) resulted in a weak reductionof the binding specificity As depicted in Figure 3b bindingof cNAPL to domain DV and DVI of the first psaA intronwas also competed by poly(U) and in addition by poly(A)and to a lesser extent by higher amounts of poly(G)(02 and 05 mg) The competition assays suggest thatcNAPL preferentially interacts with U-rich sequences Tofurther investigate the specificity of cNAPL for the tscARNA total wild-type RNA was used as competitor inEMSAs As shown in Figure 3a and b 05 mg of total RNAcompeted the binding of cNAPL to the tscA and psaA intronRNA respectively suggesting that there are other transcriptsbound by cNAPL Simultaneously we examined the bindingof cNAPL to transcripts of four chloroplast genes and threenuclear genes As given in Figure 3c cNAPL binds to 50-UTRs of chloroplast psbA psbD rbcL and rps4 RNA andto a lesser extent to nuclear Lhcb1 50-UTR and Rps18 Nobinding is observed to nuclear Lhcb1 30-UTR and Tba1ATo investigate the binding specificity of cNAPL to the tscARNA unlabelled RNA of the above-mentioned chloroplastand nuclear transcripts was used in competition assaysAs shown in Figure 3d a 50-fold molar excess of Lhcb1 30-UTR Tba1A and a 15-fold molar excess of Lhcb1 50-UTRhad no significant effect on the formation of the complexwhereas a 50-fold molar excess of psbA 50-UTR psbD 50-UTR and Rps18 resulted in a slight decrease in tscA RNAndashprotein complex formation Interestingly similar results incompetition experiments were obtained when binding ofcNAPL to the labelled tscA RNA was competed by a50-fold molar excess of 50-UTRs of rbcL or rps4 or by a2-fold molar excess of unlabelled tscA RNA Thus cNAPLshows specificity for the tscA RNA and to a lesser extent toother U-rich chloroplast transcripts as well The overallcomparison of all tested in vitro transcripts revealed thatbinding of cNAPL depends on the U-content and stretchesof Us (data not shown)

We further tested whether the typical secondary structuresof group II intron domains mediate cNAPL bindingIn Figure 3e only domains with U-rich sequences are ableto shift cNAPL indicating that the structure is not relevantfor binding For example the U-rich domain DV+VI of thefirst psaA intron (Figure 2e) show cNAPL binding whilethe corresponding domains of the heterologous intron rI1 ofScenedesmus obliquus do not Furthermore the experiment

Nucleic Acids Research 2006 Vol 34 No 18 5341

in Figure 3f shows that cNAPL do not have DNA-bindingactivity

Taken together these results indicate that cNAPL bindsspecifically the first psaA group II intron and thus confirm

data from the yeast three-hybrid analyses FurthermorecNAPL has affinities to U-rich chloroplast 50-UTRs Hencethe data imply that the RNA-binding protein cNAPLmay play a more general role in the RNA metabolism ofCreinhardtii

Characterization of a chloroplast NAP-like polypeptide

The nucleotide sequence of the cNapl cDNA (pGAD6-19)consists of 1642 bp with a TGTTA polyadenylation sitelocated 139 bp upstream of the poly(A) tail in the 50-UTRThe 30-UTR region is 588 bp in length and a 50 rapid ampli-fication of cDNA ends (RACE) analysis revealed 22 bp forthe 50-UTR (data not shown) Sequence analysis identifiedan open reading frame of 1053 bp and 350 amino acidsencoding a polypeptide with a predicted size of 394 kDaThe amino acid composition of the protein revealed 206acidic and 135 basic residues with an estimated pI of433 The hydropathy profile calculated according to (38)indicates mostly hydrophilic regions suggesting thatcNAPL is a soluble protein The neural-network based pre-dictors for subcellular localization of proteins ChloroP (31)and LOCtree (32) revealed a putative transit sequence forimport into the chloroplast at the N-terminal end ofcNAPL whereas TargetP proposes a mitochondrial localiza-tion However for chloroplast transit peptides of Chlamydo-monas it was shown that they are more similar tomitochondrial targeting peptides than to chloroplast transitpeptides of higher plants (39) The N-terminal amino acidsequence of cNAPL possesses features typical of chloroplasttransit peptides such as an alanine as the second residue TheN-terminal end is an uncharged region with an abundance ofarginine (136) valine (136) and alanine (182) resi-dues This transit peptide could be 46 amino acids longsince a possible cleavage site sequence Val-Gly-Ala ispresent (39) The cleavage would give rise to a predictedmature protein of 304 amino acids with a molecular massof 346 kDa

Figure 2 In vitro binding of cNAPL to representative intron domains of tscAand psaA RNA (a) Secondary structure model of the first psaA intron (6)The tripartite group II intron consists of three separate RNA molecules (tscA50 intron exon 1 30 intron exon 2) Arrows indicate 50ndash30 strand polarity of thethree separate transcripts Roman letters denote the six conserved domains ofgroup II introns (DI-DVI) Black numerals correspond to nucleotides of theprocessed tscA RNA sequence and to the 30 intron sequence of psaA exon 2The gndashg 0 tertiary interaction is symbolized by a dashed line (79) The hndashh0

interaction is indicated (72) The peripheral structures of domains DI and DIVcould not be unambiguously determined and are not represented modifiedaccording to (6) (bndashe) RNA EMSAs of domain DII and DIII of tscA RNAand domain DV+VI of the first psaA intron with increasing amounts ofcNAPL EMSAs were performed by incubating 30 fmol internally labelledintron RNA (tscADII tscADIII tscADII+III psaADV+VI) with increasingamounts of affinity-purified cNAPL polypeptide (numbers indicate theamounts of protein in mg) or without protein (0) Affinity-purified CEFEFprotein of Acremonium chrysogenum was used as a negative control (C)RNAndashprotein complexes were resolved from free RNA (F) by electrophoresisin 5 native polyacrylamide gels Arrows indicate shifted bands (fndashi)Competition assays of cNAPL using 30 fmol radioactive probes of internallylabelled intron RNA (tscADII tscADIII tscADII+III psaADV+VI) andexcess of cold specific (tscADII tscADIII tscADII+III psaADV+VI) andnon-specific competitors (pBIIKS+) Lanes covered with triangle on top are0- 2- 5- and 10-fold molar excess of competitor respectively in the case oftscA RNA and 0- 25- 5- 10- and 20-fold molar excess of competitorrespectively in the case of psaA Free RNA negative control and shiftedbands are indicated as in (bndashe)

5342 Nucleic Acids Research 2006 Vol 34 No 18

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 2: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

assays demonstrate that chloroplasts of Creinhardtii containa NAP-like protein that specifically binds to organellar groupII intron RNA and U-rich chloroplast transcripts Our datafurther support the view that during evolution chloroplastshave acquired novel nuclear components that most probablywere not delivered by endosymbiont gene transfer (14)

MATERIALS AND METHODS

Strains culture conditions and transformation

Creinhardtii cell wall depleted arginine auxotroph strainargb-ecw15 as well as wild-type strain CC406 cw15 mt-were obtained from the Chlamydomonas Center (Duke Uni-versity Durham NC USA) and were grown as describedon Trisndashacetatendashphosphate (TAP) medium (15) Whenrequired the medium was supplemented with 50 mg arginineper ml Nuclear transformation was carried out using theglass bead method (16) and was performed according to (8)

with the following modifications A total of 1ndash3 middot 106

cellsml were incubated with 1ndash5 mg supercoiled or SpeIdigested pMF59 (17) plasmid DNA After transformationcells were spread on solid TAP medium and incubated in con-tinuous light at 25C For selection of transformants carryingpMF59 medium contained 5 mgml of zeocin (InvitrogenKarlsruhe Germany)

Recombinant plasmids

All recombinant plasmids used for in vitro transcription PCRanalysis protein synthesis yeast three-hybrid screeninghybridization or generation of transgenic algal strains arelisted in Table 1 Sequences of all oligonucleotides aregiven in Table 2

Molecular biological techniques

Procedures for standard molecular techniques were per-formed as described previously (18) Escherichia coli strain

Table 1 Plasmids used in this work

Plasmid Characteristics Reference

Yeast three-hybridpADRevM10 pACTII derivative expression of fusion protein GAL4-AD and HIV1 RevM10 (21)pDBRevM10 Yeast shuttle vector (TRP1 LEU2) cloning of hybrid RNA transcription unit cassette

of pPGKRRE and expression cassette of fusion protein GAL4-DB and HIV1 RevM10(21)

pGAD10 derivatives cDNA fragments of Creinhardtii CC406 cw15 in pGAD10 (BD Biosciences Clontech) (20)pPGKRRE Intermediate to clone hybrid RNAs as a transcription unit cassette into pDBRevM10 (21)pRev30Lhcb Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and Lhcb1 30-UTR (pBQ24) (37) (8)pRevpsaADV+VI Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and domains DV+VI of psaA intron 1 This workpRevR2 RREndashRRE transcription unit cassette (21)pRevrI1DIV-VI Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and domains DIV-VI of intron rI1

of Sobliquus (36)(8)

pRevtscA1 Fusion protein GAL4-DB and RevM10 hybrid RNA RRE and processed tscA RNA (6) (8)cNAPL cDNA

pGAD6-19 16 kb cNapl cDNA fragment in pGAD10 This workEMSA

pQE70-cNAPL 1 kb SphIBglII coding sequence of cNAPL in pQE70 with His6 tag (Qiagen) This workpdI 347 bp fragment of domain DI of intron rI1 of Sobliquus in pT3T7BMSmaI This workpdII+III 139 bp fragment of domain DII and DIII of intron rI1 of Sobliquus in pT3T7BMSmaI This workpdIV 57 bp fragment of domain DIV of intron rI1 of Sobliquus complementary oligonucleotides

in pT3T7BM(22)

pdV+VI 73 bp fragment of domain DV and DVI of intron rI1 of Sobliquus in pT3T7BMSmaI (22)p734 211 pb fragment of rps4 50-UTR in plasmid pBIIKS+ (22)pT3T730UTRLhcb1T7 109 nt EcoRIBamHI fragment of Lhcb1 30-UTR in pT3T7BM This workpT3T7psaADV+VIT7 400 nt MluI fragment of pRevpsaADV+VI in pT3T7BMSmaI (Boehringer

Mannheim Germany)(23)

pT3T7Rps18T7 136 nt EcoRIBamHI fragment of exon 2 of Rps18 (46) in pT3T7BM This workpT3T7Tba1AT7 127 nt EcoRIBamHI fragment of exon 3 of Tba1A in pT3T7BM This workpT3T7tscADIIT7 156 nt EcoRIBamHI fragment of domain DII of tscA RNA in pT3T7BM This workpT3T7tscADIIIT7 193 nt EcoRIBamHI fragment of domain DIII of tscA RNA in pT3T7BM This workpT3T7tscADII+IIIT7 338 nt EcoRIBamHI fragment of domain DII and DIII of tscA RNA in pT3T7BM This work

LSCFMpCr1 HSP70ARBCS2 promoter (pCB740) (49) and Lhcb1 30-UTR terminator ASL for selection This workpCr1g 07 kb NheIBglII fragment of cgfp (47) in pCr1 This workpCr1g_cNaplN84 03 kb NheI fragment N-terminal 84 amino acids of cNAPL in pCr1g encoding for

fusion protein N84-cGFPThis work

pCr1g_cNaplN84D8-39 02 kb NheI fragment N-terminal 84 amino acids of cNAPL without targeting signalin pCr1g encoding for fusion protein N84D8-39-cGFP

This work

pCr1g_Rps18 05 kb NheI fragment of Rps18 coding sequence in pCr1g encoding for fusionprotein Rps18-cGFP

This work

pGAD-Rps18 Rps18 cDNA fragment in pGAD424 (BD Biosciences Clontech) This workpMF59 Encoding for fusion protein Ble-cGFP (17)

AD DNA activation domain ASL arginiosuccinate lyase DB DNA-binding domain EMSA electromobility shift assay LSCFM laser scanning confocalfluorescence microscopy

5338 Nucleic Acids Research 2006 Vol 34 No 18

XL1-blue MRF0 served as host for general plasmid cons-truction and maintenance (19) The Creinhardtii cDNAlibrary contains double-stranded cDNA fragments withEcoRI adaptors cloned into the EcoRI site of pGAD10 inDH10B (BD Biosciences Clontech Heidelberg Germany)(20) pGAD10 derivatives were used in the yeast three-hybridscreen Creinhardtii total RNA was prepared according to(4) and RNA blot experiments were carried out accordingto (8) Northern analyses were performed with a radioactivelylabelled 1690 bp pGAD6-19EcoRI cNapl-specific probeand a 460 bp pGAD-Rps18EcoRIndashBamHI Rps18-specificprobe cDNA was prepared using Omniscript ReverseTrancriptase (Qiagen Hilden Germany) and purified withNucAway Spin Columns (Ambion Austin TX USA)according to the manufacturerrsquos protocols For PCRs Taqpolymerase (Eppendorf Hamburg Germany) Hot MasterTaq DNA Polymerase (Eppendorf) and the Expand LongTemplate PCR System (Roche Mannheim Germany) wereused Oligonucleotide primers applied for conventional PCRare listed in Table 2

The yeast three-hybrid system

The yeast three-hybrid analysis was performed as describedpreviously (8) Plasmids are listed in Table 1Plasmids pADRevM10 and pRevR2 were generous giftsfrom Dr U Putz (University of Hamburg) (21) PlasmidpRevtscA1 (8) was used to screen a Creinhardtii cDNAlibrary (Table 1)

Electrophoretic mobility shift assays (EMSAs)

Recombinant plasmid pQE70-cNAPL (see above) was usedfor synthesis of cNAPL in Ecoli Recombinant His-taggedcNAPL protein was purified from Ecoli with Ni2+-NTA-agarose resin according to the manufacturerrsquos protocols(Qiagen) For RNA mobility shift assays uniformly32P-UTP-labelled run-off transcripts served as substrateRNAs and were generated by in vitro transcription of plas-mids as given in Table 1 In vitro transcription and EMSAswere performed as previously reported (2223) Unlabelledcompetitor RNAs and non-specific competitor RNA deri-ved from plasmid pBIIKS+ were synthesized as described(22) DNA mobility shift assays were done essentially asdescribed previously (24) with the following modificationsThe probes were amplified by PCR with appropriate plasmidDNA or genomic DNA of strain arg-cw15 in the presence of[32P]dATP The labelled probes were purified by denaturingPAGE Binding reactions were carried out by incubating30 fmol of the probes with 15 mg of recombinant cNAPLThe reaction mixtures were separated on a 5 non-denaturing polyacrylamide gel

Laser scanning confocal fluorescence microscopy

The fluorescence emissions of transformed Creinhardtiicells were analysed by LSCFM using a Zeiss LSM510 META microscopy system (Carl Zeiss JenaGermany) based on an Axiovert inverted microscopecGFP and plastids were excited with the 488 nm line of an

Table 2 Oligonucleotide primer pairs used in PCR and RTndashPCR experiments

Oligonucleotide Specific sequence (5030) Specificity

2428_NheIBglIISmaI GCT AGC AGA TCT CCC GGG TTT GTA TGA GCG TAT AAG CTC TGC 50-region of Lhcb1 30-UTR2429_EcoRV GAT ATC ATC GAT GGC CTA CCC AAA CCC C 30-region of Lhcb1 30-UTRfor_6-19_BspHI TAT TCA TGA TGG CAC TAG CTC TGC TCA CTC G cNapl RTndashPCRrev_6-19_2xMet_HindIII ATA AAG CTT CTA CAT CAT GTA CTC GCC CTC CTC GTA CTC G cNapl RTndashPCRfor_6-19_ SphI GCA TGC GCA TGG CAC TAG CTC TGC TCA C cNaplrev_6-19_BglII AGA TCT GTA CTC GCC CTC CTC GTA C cNaplfor_ATG_cgfp CTA GCT AGC ATG GCC AAG GGC GAG GAG cgfprev_cgfp GAA GAT CTT TAC TTG TAC AGC TCG TCC ATG C cgfpfor_cNapl_N84 CTA GCT AGC ATG GCA CTA GCT CTG CTC AC cNaplN84rev_cNapl_N84 CTA GCT AGC GGC GGT GTT GAG CTC CTC cNaplN84 cNaplN84D8-39for_cNapl_N84D8-39 CTA GCT AGC ATG GCA CTA GCT CTG CTC ACT TTC CTG

GCA GTG GTT GGC GCCcNaplN84D8-39

for_30Lhcb_EcoRI CCA TCA AGG ACC AGG TCA TC Fragment of Lhcb1 30-UTRrev_30Lhcb_BamHI AGG AAT TCA AGA GCC CAT TCC AAA GTC Fragment of Lhcb1 30-UTRLhcbm6 T7-1 GTA ATA CGA CAT CAC TAT AGG GCC TGT TTG GCG CTT Fragment of Lhcb1 50-UTRLhcbm6-2 CAT TCC TGC ACG CTC GCT TG Fragment of Lhcb1 50-UTRT7-psbA50 GTA ATA CGA CTC ACT ATA GGG TAC CAT GCT TTT AAT Fragment of psbA 50-UTRpsbA30-2054 GAT CCA TGG TCA TAT GTT AAT TTT TTT AAA G Fragment of psbA 50-UTRsu3131 TGT GCG TTT CTC TTG ATA TGT ACC G Fragment of psbD 50-UTR2126 TAA TAC GAC TCA CTA TAG GGA CAC AAT GAT TAA AAT TAA A Fragment of psbD 50-UTRT7-rbcL50 GTA ATA CGA CTC ACT ATA GGG TAT GCT CGA CTG ATA AGA C Fragment of rbcL 50-UTRrbcL30 CTG CTT TAG TTT CTG TTT GTG GAA CC Fragment of rbcL 50-UTRfor_Rps18_NheI CTG CTA GCA TGG GCT CTC TCG TCC A Rps18rev_Rps18_NheI GAG CTA GCC TTC TTC TTG GCG ACA CC Rps18for_Rps18_EcoRI AGG AAT TCT CTG CGT CTG CTG AAC ACT AAC Fragment of exon 2 of Rps18rev_Rps18_BamHI CGG GAT CCA GGT CAA CCT CAG CCT TCT T Fragment of exon 2 of Rps18for_tscADII AGG AAT TCA AAT TTT TAG TAA TGT TAA AG Domain DII of tscA RNArev_tscADII CGG GAT CCC CTT AAT TTT TAG AAA ATG Domain DII of tscA RNAfor_tscADIII AGG AAT TCT AAA CAA CAA GTA ATG CC Domain DIII of tscA RNArev_tscADIII CGG GAT CCT AAA CAA TAA AGT TTT TCA A Domain DIII of tscA RNAfor_a1Tub_EcoRI AGG AAT TCC GGC TTC AAG TGC GGT ATC AAC Fragment of exon 3 of Tba1Arev_a1Tub_BamHI CGG GAT CCT CGC CGA TAG CAG TGC TGT T Fragment of exon 3 of Tba1A

Nucleic Acids Research 2006 Vol 34 No 18 5339

argon-ion laser The fluorescence emission was selected bybandpass filter BP505-530 and longpass filter LP560 respec-tively using beam splitters HFT UV488543633 andNFT545

Sequences alignments and phylogenetic analysis

Sequences were retrieved from GenBank (NCBI) at httpwwwncbinlmnihgov the Creinhardtii JGI database v20(DOE Joint Genome Institute) at httpgenomejgi-psforgchlre2chlre2homehtml the Thalassiosira pseudonana JGIdatabase v10 at httpgenomejgi-psforgthaps1thaps1homehtml the Porphyra yezoensis EST index at httpwwwkazusaorjpenplantporphyraEST (2526) and theCyanoBase genome database at httpwwwkazusaorjpcyano Multiple sequence alignments of NAPs were per-formed using ClustalW software (27) Sequence alignmentswere displayed with GENEDOC (28) and refined manuallyThe phylogenetic tree was generated based on trimmedprotein sequences with programs contained in the programpackage PHYLIP version 363 (29) An alignment withsequences from mature proteins was impossible since all pro-teins differ largely in length and some were only availablefrom partial cDNAs (AU191247 BM448458) obtainedfrom EST libraries Analyses with the PHYLIP packagewere performed using PROTDISTNEIGHBOR and thePROTPARS algorithms for distance matrix and maximumparsimony analysis respectively Support for the brancheswas estimated by bootstrap analysis of 2000 replicatesgenerated with SEQBOOT and a majority rule consensustree was generated using the program CONSENSEThe resulting tree topology was then visualized inTREEVIEW (30) For analyses of putative targeting signalsequences ChloroP (31) LOCtree (32) TargetP (33)GENOPLANTETM PREDOTAR (httpwwwgenoplantecom) and PSORT (34) were used Isoelectric pointand sequence masses were calculated by the programWinPep 301 (35)

RESULTS

Identification of a tscA-specific RNA-binding proteinusing the yeast three-hybrid system

To identify RNA-binding proteins most probably involved inorganellar group II intron splicing we performed an in vivoscreen using the yeast three-hybrid system In this systeman RNAndashprotein interaction is detected by the reconstitutionof the yeast transcriptional activator GAL4 and subsequentactivation of reporter genes The system is based on onehybrid RNA and two recombinant fusion proteins containingthe GAL4 DNA-binding and GAL4 transcription activationdomains respectively To isolate chloroplast intron RNA-binding proteins of Creinhardtii a cDNA library wasscreened using the yeast three-hybrid system developed by(21) In this system the chimeric protein RevM10 fused toa gene fragment encoding the DNA-binding domain ofGAL4 (GAL4-DB RevM10) binds to RRE which is partof a chimeric RNA The 30 part of this RNA consists of thetscA RNA which functions as bait during the screen (RRE-tscA) Chimeric proteins serve as prey and are encoded bycDNA clones of a Creinhardtii cDNA library All cDNAswere fused to a gene fragment encoding the transcriptionactivation domain of GAL4 (GAL4-AD) The reliability ofthis three-hybrid screen was tested with a recombinantRNA carrying two copies of RRE (RREndashRRE) co-expressedwith the two RevM10 fusion proteins (AD-RevM10DB-RevM10) RRE and RevM10 are known to interactwith each other (21) As depicted in lane 1 of Figure 1 theinteraction resulted in reconstitution of GAL4 activity

For a global Creinhardtii cDNA screen the yeast strainCG1945 was co-transformed with pRevtscA1 expressingboth the hybrid protein DB-RevM10 and the recombinantRNA This strain served as host for transformation of thelibrary of pGAD10-derived plasmids harbouring cDNAsequences of Creinhardtii A total of 21 middot 107 clones wereanalysed and 14 clones were identified to bind to tscA RNAInterestingly of these 14 candidate cDNAs 13 encoded

Figure 1 Schematic overview of selected constructs used in the yeast three-hybrid analyses In each case five representative yeast transformants were tested fortheir ability to activate b-galactosidase by a colony colour assay Abbreviations Lhcb30 fragment of 30-UTR of the Lhcb1 transcript (37) cNAPL chloroplastNAP-like GAL4-AD activation domain of GAL4 transcription factor GAL4-DB DNA-binding domain of GAL4 transcription factor psaADV+VI domainsDV and DVI of the first psaA intron (4) RevM10 RNA-binding protein RevM10 (21) rI1DIV-VI domains DIV to DVI of intron rI1 (36) RRE Rev responsiveelement that is specifically bound by RevM10 (21) tscA processed tscA RNA (6)

5340 Nucleic Acids Research 2006 Vol 34 No 18

identical polypeptides with significant homology to NAPsBecause of its binding to chloroplast intron RNA theNAP-like polypeptide was designated as cNAPL

To assess the specificity of the three-hybrid interactionbetween tscA RNA and cNAPL several controls were carriedout The specificity of the tscA interaction was tested againsttwo further hybrid RNAs Recombinant control RNAscontaining sequences of domains DIV to DVI of the hetero-logous intron rI1 (36) (RRE-rI1DIV-VI) and a fragment ofthe 30-untranslated region (30-UTR) of the Lhcb1 transcript(37) (RRE-Lhcb30) were fused separately to the RREsequence respectively As shown in lanes 5 7 and 8 ofFigure 1 activity of the reporter b-galactosidase is dependenton the presence of the RRE-tscA hybrid RNA and thecNAPL-AD fusion protein These results indicate a specificinteraction of cNAPL with the tscA transcript In additionwe were able to demonstrate that cNAPL also binds todomains DV and DVI of the first psaA intron (RRE-psaADV+VI) as seen in lane 6 (Figure 1) To exclude one-or two-hybrid interaction or unspecific RNA interactionappropriate negative controls were also performed in the filterassay including different combinations of the three-hybridcomponents represented in lanes 2ndash4 (Figure 1) In theabsence of any one of the hybrid components transformantsdisplayed no b-galactosidase activity Results of the three-hybrid analysis therefore clearly demonstrate the specificinteraction of cNAPL with tscA RNA and the 30 part of thefirst intron of psaA RNA

In vitro binding of cNAPL to chloroplast transcripts

To confirm the specific interaction between cNAPL and tscARNA in the yeast three-hybrid system we performed in vitrobinding studies using EMSA For synthesis of cNAPL thecNapl gene was fused with six histidine residues forexpression in Ecoli After purification of the recombinantHis-tagged protein with Ni2+-NTA chromatography cNAPLwas incubated with in vitro transcribed radioactively labelledRNAs corresponding to domains DII and DIII of tscA RNAAs shown in Figure 2a domains DII and DIII range from nt75 to 218 and from 220 to 400 respectively Together theycover nt 75ndash400 RNAs were incubated in the absence orpresence of recombinant cNAPL and fractionated on nativegels (Figure 2bndashi) In Figure 2bndashd increasing amounts ofcNAPL (2ndash20 mg) were incubated with 30 fmol of in vitrosynthesized transcripts of tscA intron RNA fragmentscomprising either domain DII domain DIII or both domainsDII+III Arrows indicate RNAndashprotein complexes as multipleshifted bands that can be observed when the RNA isincubated with increasing amounts of cNAPL polypeptide

To investigate the binding specificity of cNAPL competi-tion analyses were performed The above-mentioned radio-labelled RNAs were incubated with a constant amount of15 mg of cNAPL protein in the presence of increasingamounts of unlabelled specific and non-specific competitorRNAs respectively As shown in Figure 2fndashh the presenceof just a 2-fold molar excess of unlabelled specific competitorRNA almost completely abolished the formation of acomplex between cNAPL and the labelled target tscA RNAIn contrast the presence of the same amount of non-specific competitor RNA namely pBIIKS+ (121 nt) had no

significant effect on the formation of the complex with thelabelled target RNAs In addition EMSA was performedwith in vitro synthesized RNA of domains DV and DVI ofthe first psaA intron As shown in Figure 2a the two domainsrange from nt 108 to 192 EMSAs with increasing amountsof cNAPL protein (Figure 2e) and competition analysis(Figure 2i) indicate that cNAPL binds specifically to domainsDV and DVI of the first psaA intron albeit with lower affinitythan to tscA RNA

Chloroplast transcripts are rich in A and U residues Tostudy the sequence preferences we compared the ability ofA C G or U RNA homopolymers to compete for bindingof cNAPL to the labelled domain DII of the tscA RNAand domain DV+VI of the first psaA intron respectivelyAs depicted in Figure 3a excess amounts of poly(U)abolished cNAPL binding while the same amounts ofpoly(C) or poly(G) have no significant effect on the formationof the complex Using poly(A) resulted in a weak reductionof the binding specificity As depicted in Figure 3b bindingof cNAPL to domain DV and DVI of the first psaA intronwas also competed by poly(U) and in addition by poly(A)and to a lesser extent by higher amounts of poly(G)(02 and 05 mg) The competition assays suggest thatcNAPL preferentially interacts with U-rich sequences Tofurther investigate the specificity of cNAPL for the tscARNA total wild-type RNA was used as competitor inEMSAs As shown in Figure 3a and b 05 mg of total RNAcompeted the binding of cNAPL to the tscA and psaA intronRNA respectively suggesting that there are other transcriptsbound by cNAPL Simultaneously we examined the bindingof cNAPL to transcripts of four chloroplast genes and threenuclear genes As given in Figure 3c cNAPL binds to 50-UTRs of chloroplast psbA psbD rbcL and rps4 RNA andto a lesser extent to nuclear Lhcb1 50-UTR and Rps18 Nobinding is observed to nuclear Lhcb1 30-UTR and Tba1ATo investigate the binding specificity of cNAPL to the tscARNA unlabelled RNA of the above-mentioned chloroplastand nuclear transcripts was used in competition assaysAs shown in Figure 3d a 50-fold molar excess of Lhcb1 30-UTR Tba1A and a 15-fold molar excess of Lhcb1 50-UTRhad no significant effect on the formation of the complexwhereas a 50-fold molar excess of psbA 50-UTR psbD 50-UTR and Rps18 resulted in a slight decrease in tscA RNAndashprotein complex formation Interestingly similar results incompetition experiments were obtained when binding ofcNAPL to the labelled tscA RNA was competed by a50-fold molar excess of 50-UTRs of rbcL or rps4 or by a2-fold molar excess of unlabelled tscA RNA Thus cNAPLshows specificity for the tscA RNA and to a lesser extent toother U-rich chloroplast transcripts as well The overallcomparison of all tested in vitro transcripts revealed thatbinding of cNAPL depends on the U-content and stretchesof Us (data not shown)

We further tested whether the typical secondary structuresof group II intron domains mediate cNAPL bindingIn Figure 3e only domains with U-rich sequences are ableto shift cNAPL indicating that the structure is not relevantfor binding For example the U-rich domain DV+VI of thefirst psaA intron (Figure 2e) show cNAPL binding whilethe corresponding domains of the heterologous intron rI1 ofScenedesmus obliquus do not Furthermore the experiment

Nucleic Acids Research 2006 Vol 34 No 18 5341

in Figure 3f shows that cNAPL do not have DNA-bindingactivity

Taken together these results indicate that cNAPL bindsspecifically the first psaA group II intron and thus confirm

data from the yeast three-hybrid analyses FurthermorecNAPL has affinities to U-rich chloroplast 50-UTRs Hencethe data imply that the RNA-binding protein cNAPLmay play a more general role in the RNA metabolism ofCreinhardtii

Characterization of a chloroplast NAP-like polypeptide

The nucleotide sequence of the cNapl cDNA (pGAD6-19)consists of 1642 bp with a TGTTA polyadenylation sitelocated 139 bp upstream of the poly(A) tail in the 50-UTRThe 30-UTR region is 588 bp in length and a 50 rapid ampli-fication of cDNA ends (RACE) analysis revealed 22 bp forthe 50-UTR (data not shown) Sequence analysis identifiedan open reading frame of 1053 bp and 350 amino acidsencoding a polypeptide with a predicted size of 394 kDaThe amino acid composition of the protein revealed 206acidic and 135 basic residues with an estimated pI of433 The hydropathy profile calculated according to (38)indicates mostly hydrophilic regions suggesting thatcNAPL is a soluble protein The neural-network based pre-dictors for subcellular localization of proteins ChloroP (31)and LOCtree (32) revealed a putative transit sequence forimport into the chloroplast at the N-terminal end ofcNAPL whereas TargetP proposes a mitochondrial localiza-tion However for chloroplast transit peptides of Chlamydo-monas it was shown that they are more similar tomitochondrial targeting peptides than to chloroplast transitpeptides of higher plants (39) The N-terminal amino acidsequence of cNAPL possesses features typical of chloroplasttransit peptides such as an alanine as the second residue TheN-terminal end is an uncharged region with an abundance ofarginine (136) valine (136) and alanine (182) resi-dues This transit peptide could be 46 amino acids longsince a possible cleavage site sequence Val-Gly-Ala ispresent (39) The cleavage would give rise to a predictedmature protein of 304 amino acids with a molecular massof 346 kDa

Figure 2 In vitro binding of cNAPL to representative intron domains of tscAand psaA RNA (a) Secondary structure model of the first psaA intron (6)The tripartite group II intron consists of three separate RNA molecules (tscA50 intron exon 1 30 intron exon 2) Arrows indicate 50ndash30 strand polarity of thethree separate transcripts Roman letters denote the six conserved domains ofgroup II introns (DI-DVI) Black numerals correspond to nucleotides of theprocessed tscA RNA sequence and to the 30 intron sequence of psaA exon 2The gndashg 0 tertiary interaction is symbolized by a dashed line (79) The hndashh0

interaction is indicated (72) The peripheral structures of domains DI and DIVcould not be unambiguously determined and are not represented modifiedaccording to (6) (bndashe) RNA EMSAs of domain DII and DIII of tscA RNAand domain DV+VI of the first psaA intron with increasing amounts ofcNAPL EMSAs were performed by incubating 30 fmol internally labelledintron RNA (tscADII tscADIII tscADII+III psaADV+VI) with increasingamounts of affinity-purified cNAPL polypeptide (numbers indicate theamounts of protein in mg) or without protein (0) Affinity-purified CEFEFprotein of Acremonium chrysogenum was used as a negative control (C)RNAndashprotein complexes were resolved from free RNA (F) by electrophoresisin 5 native polyacrylamide gels Arrows indicate shifted bands (fndashi)Competition assays of cNAPL using 30 fmol radioactive probes of internallylabelled intron RNA (tscADII tscADIII tscADII+III psaADV+VI) andexcess of cold specific (tscADII tscADIII tscADII+III psaADV+VI) andnon-specific competitors (pBIIKS+) Lanes covered with triangle on top are0- 2- 5- and 10-fold molar excess of competitor respectively in the case oftscA RNA and 0- 25- 5- 10- and 20-fold molar excess of competitorrespectively in the case of psaA Free RNA negative control and shiftedbands are indicated as in (bndashe)

5342 Nucleic Acids Research 2006 Vol 34 No 18

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 3: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

XL1-blue MRF0 served as host for general plasmid cons-truction and maintenance (19) The Creinhardtii cDNAlibrary contains double-stranded cDNA fragments withEcoRI adaptors cloned into the EcoRI site of pGAD10 inDH10B (BD Biosciences Clontech Heidelberg Germany)(20) pGAD10 derivatives were used in the yeast three-hybridscreen Creinhardtii total RNA was prepared according to(4) and RNA blot experiments were carried out accordingto (8) Northern analyses were performed with a radioactivelylabelled 1690 bp pGAD6-19EcoRI cNapl-specific probeand a 460 bp pGAD-Rps18EcoRIndashBamHI Rps18-specificprobe cDNA was prepared using Omniscript ReverseTrancriptase (Qiagen Hilden Germany) and purified withNucAway Spin Columns (Ambion Austin TX USA)according to the manufacturerrsquos protocols For PCRs Taqpolymerase (Eppendorf Hamburg Germany) Hot MasterTaq DNA Polymerase (Eppendorf) and the Expand LongTemplate PCR System (Roche Mannheim Germany) wereused Oligonucleotide primers applied for conventional PCRare listed in Table 2

The yeast three-hybrid system

The yeast three-hybrid analysis was performed as describedpreviously (8) Plasmids are listed in Table 1Plasmids pADRevM10 and pRevR2 were generous giftsfrom Dr U Putz (University of Hamburg) (21) PlasmidpRevtscA1 (8) was used to screen a Creinhardtii cDNAlibrary (Table 1)

Electrophoretic mobility shift assays (EMSAs)

Recombinant plasmid pQE70-cNAPL (see above) was usedfor synthesis of cNAPL in Ecoli Recombinant His-taggedcNAPL protein was purified from Ecoli with Ni2+-NTA-agarose resin according to the manufacturerrsquos protocols(Qiagen) For RNA mobility shift assays uniformly32P-UTP-labelled run-off transcripts served as substrateRNAs and were generated by in vitro transcription of plas-mids as given in Table 1 In vitro transcription and EMSAswere performed as previously reported (2223) Unlabelledcompetitor RNAs and non-specific competitor RNA deri-ved from plasmid pBIIKS+ were synthesized as described(22) DNA mobility shift assays were done essentially asdescribed previously (24) with the following modificationsThe probes were amplified by PCR with appropriate plasmidDNA or genomic DNA of strain arg-cw15 in the presence of[32P]dATP The labelled probes were purified by denaturingPAGE Binding reactions were carried out by incubating30 fmol of the probes with 15 mg of recombinant cNAPLThe reaction mixtures were separated on a 5 non-denaturing polyacrylamide gel

Laser scanning confocal fluorescence microscopy

The fluorescence emissions of transformed Creinhardtiicells were analysed by LSCFM using a Zeiss LSM510 META microscopy system (Carl Zeiss JenaGermany) based on an Axiovert inverted microscopecGFP and plastids were excited with the 488 nm line of an

Table 2 Oligonucleotide primer pairs used in PCR and RTndashPCR experiments

Oligonucleotide Specific sequence (5030) Specificity

2428_NheIBglIISmaI GCT AGC AGA TCT CCC GGG TTT GTA TGA GCG TAT AAG CTC TGC 50-region of Lhcb1 30-UTR2429_EcoRV GAT ATC ATC GAT GGC CTA CCC AAA CCC C 30-region of Lhcb1 30-UTRfor_6-19_BspHI TAT TCA TGA TGG CAC TAG CTC TGC TCA CTC G cNapl RTndashPCRrev_6-19_2xMet_HindIII ATA AAG CTT CTA CAT CAT GTA CTC GCC CTC CTC GTA CTC G cNapl RTndashPCRfor_6-19_ SphI GCA TGC GCA TGG CAC TAG CTC TGC TCA C cNaplrev_6-19_BglII AGA TCT GTA CTC GCC CTC CTC GTA C cNaplfor_ATG_cgfp CTA GCT AGC ATG GCC AAG GGC GAG GAG cgfprev_cgfp GAA GAT CTT TAC TTG TAC AGC TCG TCC ATG C cgfpfor_cNapl_N84 CTA GCT AGC ATG GCA CTA GCT CTG CTC AC cNaplN84rev_cNapl_N84 CTA GCT AGC GGC GGT GTT GAG CTC CTC cNaplN84 cNaplN84D8-39for_cNapl_N84D8-39 CTA GCT AGC ATG GCA CTA GCT CTG CTC ACT TTC CTG

GCA GTG GTT GGC GCCcNaplN84D8-39

for_30Lhcb_EcoRI CCA TCA AGG ACC AGG TCA TC Fragment of Lhcb1 30-UTRrev_30Lhcb_BamHI AGG AAT TCA AGA GCC CAT TCC AAA GTC Fragment of Lhcb1 30-UTRLhcbm6 T7-1 GTA ATA CGA CAT CAC TAT AGG GCC TGT TTG GCG CTT Fragment of Lhcb1 50-UTRLhcbm6-2 CAT TCC TGC ACG CTC GCT TG Fragment of Lhcb1 50-UTRT7-psbA50 GTA ATA CGA CTC ACT ATA GGG TAC CAT GCT TTT AAT Fragment of psbA 50-UTRpsbA30-2054 GAT CCA TGG TCA TAT GTT AAT TTT TTT AAA G Fragment of psbA 50-UTRsu3131 TGT GCG TTT CTC TTG ATA TGT ACC G Fragment of psbD 50-UTR2126 TAA TAC GAC TCA CTA TAG GGA CAC AAT GAT TAA AAT TAA A Fragment of psbD 50-UTRT7-rbcL50 GTA ATA CGA CTC ACT ATA GGG TAT GCT CGA CTG ATA AGA C Fragment of rbcL 50-UTRrbcL30 CTG CTT TAG TTT CTG TTT GTG GAA CC Fragment of rbcL 50-UTRfor_Rps18_NheI CTG CTA GCA TGG GCT CTC TCG TCC A Rps18rev_Rps18_NheI GAG CTA GCC TTC TTC TTG GCG ACA CC Rps18for_Rps18_EcoRI AGG AAT TCT CTG CGT CTG CTG AAC ACT AAC Fragment of exon 2 of Rps18rev_Rps18_BamHI CGG GAT CCA GGT CAA CCT CAG CCT TCT T Fragment of exon 2 of Rps18for_tscADII AGG AAT TCA AAT TTT TAG TAA TGT TAA AG Domain DII of tscA RNArev_tscADII CGG GAT CCC CTT AAT TTT TAG AAA ATG Domain DII of tscA RNAfor_tscADIII AGG AAT TCT AAA CAA CAA GTA ATG CC Domain DIII of tscA RNArev_tscADIII CGG GAT CCT AAA CAA TAA AGT TTT TCA A Domain DIII of tscA RNAfor_a1Tub_EcoRI AGG AAT TCC GGC TTC AAG TGC GGT ATC AAC Fragment of exon 3 of Tba1Arev_a1Tub_BamHI CGG GAT CCT CGC CGA TAG CAG TGC TGT T Fragment of exon 3 of Tba1A

Nucleic Acids Research 2006 Vol 34 No 18 5339

argon-ion laser The fluorescence emission was selected bybandpass filter BP505-530 and longpass filter LP560 respec-tively using beam splitters HFT UV488543633 andNFT545

Sequences alignments and phylogenetic analysis

Sequences were retrieved from GenBank (NCBI) at httpwwwncbinlmnihgov the Creinhardtii JGI database v20(DOE Joint Genome Institute) at httpgenomejgi-psforgchlre2chlre2homehtml the Thalassiosira pseudonana JGIdatabase v10 at httpgenomejgi-psforgthaps1thaps1homehtml the Porphyra yezoensis EST index at httpwwwkazusaorjpenplantporphyraEST (2526) and theCyanoBase genome database at httpwwwkazusaorjpcyano Multiple sequence alignments of NAPs were per-formed using ClustalW software (27) Sequence alignmentswere displayed with GENEDOC (28) and refined manuallyThe phylogenetic tree was generated based on trimmedprotein sequences with programs contained in the programpackage PHYLIP version 363 (29) An alignment withsequences from mature proteins was impossible since all pro-teins differ largely in length and some were only availablefrom partial cDNAs (AU191247 BM448458) obtainedfrom EST libraries Analyses with the PHYLIP packagewere performed using PROTDISTNEIGHBOR and thePROTPARS algorithms for distance matrix and maximumparsimony analysis respectively Support for the brancheswas estimated by bootstrap analysis of 2000 replicatesgenerated with SEQBOOT and a majority rule consensustree was generated using the program CONSENSEThe resulting tree topology was then visualized inTREEVIEW (30) For analyses of putative targeting signalsequences ChloroP (31) LOCtree (32) TargetP (33)GENOPLANTETM PREDOTAR (httpwwwgenoplantecom) and PSORT (34) were used Isoelectric pointand sequence masses were calculated by the programWinPep 301 (35)

RESULTS

Identification of a tscA-specific RNA-binding proteinusing the yeast three-hybrid system

To identify RNA-binding proteins most probably involved inorganellar group II intron splicing we performed an in vivoscreen using the yeast three-hybrid system In this systeman RNAndashprotein interaction is detected by the reconstitutionof the yeast transcriptional activator GAL4 and subsequentactivation of reporter genes The system is based on onehybrid RNA and two recombinant fusion proteins containingthe GAL4 DNA-binding and GAL4 transcription activationdomains respectively To isolate chloroplast intron RNA-binding proteins of Creinhardtii a cDNA library wasscreened using the yeast three-hybrid system developed by(21) In this system the chimeric protein RevM10 fused toa gene fragment encoding the DNA-binding domain ofGAL4 (GAL4-DB RevM10) binds to RRE which is partof a chimeric RNA The 30 part of this RNA consists of thetscA RNA which functions as bait during the screen (RRE-tscA) Chimeric proteins serve as prey and are encoded bycDNA clones of a Creinhardtii cDNA library All cDNAswere fused to a gene fragment encoding the transcriptionactivation domain of GAL4 (GAL4-AD) The reliability ofthis three-hybrid screen was tested with a recombinantRNA carrying two copies of RRE (RREndashRRE) co-expressedwith the two RevM10 fusion proteins (AD-RevM10DB-RevM10) RRE and RevM10 are known to interactwith each other (21) As depicted in lane 1 of Figure 1 theinteraction resulted in reconstitution of GAL4 activity

For a global Creinhardtii cDNA screen the yeast strainCG1945 was co-transformed with pRevtscA1 expressingboth the hybrid protein DB-RevM10 and the recombinantRNA This strain served as host for transformation of thelibrary of pGAD10-derived plasmids harbouring cDNAsequences of Creinhardtii A total of 21 middot 107 clones wereanalysed and 14 clones were identified to bind to tscA RNAInterestingly of these 14 candidate cDNAs 13 encoded

Figure 1 Schematic overview of selected constructs used in the yeast three-hybrid analyses In each case five representative yeast transformants were tested fortheir ability to activate b-galactosidase by a colony colour assay Abbreviations Lhcb30 fragment of 30-UTR of the Lhcb1 transcript (37) cNAPL chloroplastNAP-like GAL4-AD activation domain of GAL4 transcription factor GAL4-DB DNA-binding domain of GAL4 transcription factor psaADV+VI domainsDV and DVI of the first psaA intron (4) RevM10 RNA-binding protein RevM10 (21) rI1DIV-VI domains DIV to DVI of intron rI1 (36) RRE Rev responsiveelement that is specifically bound by RevM10 (21) tscA processed tscA RNA (6)

5340 Nucleic Acids Research 2006 Vol 34 No 18

identical polypeptides with significant homology to NAPsBecause of its binding to chloroplast intron RNA theNAP-like polypeptide was designated as cNAPL

To assess the specificity of the three-hybrid interactionbetween tscA RNA and cNAPL several controls were carriedout The specificity of the tscA interaction was tested againsttwo further hybrid RNAs Recombinant control RNAscontaining sequences of domains DIV to DVI of the hetero-logous intron rI1 (36) (RRE-rI1DIV-VI) and a fragment ofthe 30-untranslated region (30-UTR) of the Lhcb1 transcript(37) (RRE-Lhcb30) were fused separately to the RREsequence respectively As shown in lanes 5 7 and 8 ofFigure 1 activity of the reporter b-galactosidase is dependenton the presence of the RRE-tscA hybrid RNA and thecNAPL-AD fusion protein These results indicate a specificinteraction of cNAPL with the tscA transcript In additionwe were able to demonstrate that cNAPL also binds todomains DV and DVI of the first psaA intron (RRE-psaADV+VI) as seen in lane 6 (Figure 1) To exclude one-or two-hybrid interaction or unspecific RNA interactionappropriate negative controls were also performed in the filterassay including different combinations of the three-hybridcomponents represented in lanes 2ndash4 (Figure 1) In theabsence of any one of the hybrid components transformantsdisplayed no b-galactosidase activity Results of the three-hybrid analysis therefore clearly demonstrate the specificinteraction of cNAPL with tscA RNA and the 30 part of thefirst intron of psaA RNA

In vitro binding of cNAPL to chloroplast transcripts

To confirm the specific interaction between cNAPL and tscARNA in the yeast three-hybrid system we performed in vitrobinding studies using EMSA For synthesis of cNAPL thecNapl gene was fused with six histidine residues forexpression in Ecoli After purification of the recombinantHis-tagged protein with Ni2+-NTA chromatography cNAPLwas incubated with in vitro transcribed radioactively labelledRNAs corresponding to domains DII and DIII of tscA RNAAs shown in Figure 2a domains DII and DIII range from nt75 to 218 and from 220 to 400 respectively Together theycover nt 75ndash400 RNAs were incubated in the absence orpresence of recombinant cNAPL and fractionated on nativegels (Figure 2bndashi) In Figure 2bndashd increasing amounts ofcNAPL (2ndash20 mg) were incubated with 30 fmol of in vitrosynthesized transcripts of tscA intron RNA fragmentscomprising either domain DII domain DIII or both domainsDII+III Arrows indicate RNAndashprotein complexes as multipleshifted bands that can be observed when the RNA isincubated with increasing amounts of cNAPL polypeptide

To investigate the binding specificity of cNAPL competi-tion analyses were performed The above-mentioned radio-labelled RNAs were incubated with a constant amount of15 mg of cNAPL protein in the presence of increasingamounts of unlabelled specific and non-specific competitorRNAs respectively As shown in Figure 2fndashh the presenceof just a 2-fold molar excess of unlabelled specific competitorRNA almost completely abolished the formation of acomplex between cNAPL and the labelled target tscA RNAIn contrast the presence of the same amount of non-specific competitor RNA namely pBIIKS+ (121 nt) had no

significant effect on the formation of the complex with thelabelled target RNAs In addition EMSA was performedwith in vitro synthesized RNA of domains DV and DVI ofthe first psaA intron As shown in Figure 2a the two domainsrange from nt 108 to 192 EMSAs with increasing amountsof cNAPL protein (Figure 2e) and competition analysis(Figure 2i) indicate that cNAPL binds specifically to domainsDV and DVI of the first psaA intron albeit with lower affinitythan to tscA RNA

Chloroplast transcripts are rich in A and U residues Tostudy the sequence preferences we compared the ability ofA C G or U RNA homopolymers to compete for bindingof cNAPL to the labelled domain DII of the tscA RNAand domain DV+VI of the first psaA intron respectivelyAs depicted in Figure 3a excess amounts of poly(U)abolished cNAPL binding while the same amounts ofpoly(C) or poly(G) have no significant effect on the formationof the complex Using poly(A) resulted in a weak reductionof the binding specificity As depicted in Figure 3b bindingof cNAPL to domain DV and DVI of the first psaA intronwas also competed by poly(U) and in addition by poly(A)and to a lesser extent by higher amounts of poly(G)(02 and 05 mg) The competition assays suggest thatcNAPL preferentially interacts with U-rich sequences Tofurther investigate the specificity of cNAPL for the tscARNA total wild-type RNA was used as competitor inEMSAs As shown in Figure 3a and b 05 mg of total RNAcompeted the binding of cNAPL to the tscA and psaA intronRNA respectively suggesting that there are other transcriptsbound by cNAPL Simultaneously we examined the bindingof cNAPL to transcripts of four chloroplast genes and threenuclear genes As given in Figure 3c cNAPL binds to 50-UTRs of chloroplast psbA psbD rbcL and rps4 RNA andto a lesser extent to nuclear Lhcb1 50-UTR and Rps18 Nobinding is observed to nuclear Lhcb1 30-UTR and Tba1ATo investigate the binding specificity of cNAPL to the tscARNA unlabelled RNA of the above-mentioned chloroplastand nuclear transcripts was used in competition assaysAs shown in Figure 3d a 50-fold molar excess of Lhcb1 30-UTR Tba1A and a 15-fold molar excess of Lhcb1 50-UTRhad no significant effect on the formation of the complexwhereas a 50-fold molar excess of psbA 50-UTR psbD 50-UTR and Rps18 resulted in a slight decrease in tscA RNAndashprotein complex formation Interestingly similar results incompetition experiments were obtained when binding ofcNAPL to the labelled tscA RNA was competed by a50-fold molar excess of 50-UTRs of rbcL or rps4 or by a2-fold molar excess of unlabelled tscA RNA Thus cNAPLshows specificity for the tscA RNA and to a lesser extent toother U-rich chloroplast transcripts as well The overallcomparison of all tested in vitro transcripts revealed thatbinding of cNAPL depends on the U-content and stretchesof Us (data not shown)

We further tested whether the typical secondary structuresof group II intron domains mediate cNAPL bindingIn Figure 3e only domains with U-rich sequences are ableto shift cNAPL indicating that the structure is not relevantfor binding For example the U-rich domain DV+VI of thefirst psaA intron (Figure 2e) show cNAPL binding whilethe corresponding domains of the heterologous intron rI1 ofScenedesmus obliquus do not Furthermore the experiment

Nucleic Acids Research 2006 Vol 34 No 18 5341

in Figure 3f shows that cNAPL do not have DNA-bindingactivity

Taken together these results indicate that cNAPL bindsspecifically the first psaA group II intron and thus confirm

data from the yeast three-hybrid analyses FurthermorecNAPL has affinities to U-rich chloroplast 50-UTRs Hencethe data imply that the RNA-binding protein cNAPLmay play a more general role in the RNA metabolism ofCreinhardtii

Characterization of a chloroplast NAP-like polypeptide

The nucleotide sequence of the cNapl cDNA (pGAD6-19)consists of 1642 bp with a TGTTA polyadenylation sitelocated 139 bp upstream of the poly(A) tail in the 50-UTRThe 30-UTR region is 588 bp in length and a 50 rapid ampli-fication of cDNA ends (RACE) analysis revealed 22 bp forthe 50-UTR (data not shown) Sequence analysis identifiedan open reading frame of 1053 bp and 350 amino acidsencoding a polypeptide with a predicted size of 394 kDaThe amino acid composition of the protein revealed 206acidic and 135 basic residues with an estimated pI of433 The hydropathy profile calculated according to (38)indicates mostly hydrophilic regions suggesting thatcNAPL is a soluble protein The neural-network based pre-dictors for subcellular localization of proteins ChloroP (31)and LOCtree (32) revealed a putative transit sequence forimport into the chloroplast at the N-terminal end ofcNAPL whereas TargetP proposes a mitochondrial localiza-tion However for chloroplast transit peptides of Chlamydo-monas it was shown that they are more similar tomitochondrial targeting peptides than to chloroplast transitpeptides of higher plants (39) The N-terminal amino acidsequence of cNAPL possesses features typical of chloroplasttransit peptides such as an alanine as the second residue TheN-terminal end is an uncharged region with an abundance ofarginine (136) valine (136) and alanine (182) resi-dues This transit peptide could be 46 amino acids longsince a possible cleavage site sequence Val-Gly-Ala ispresent (39) The cleavage would give rise to a predictedmature protein of 304 amino acids with a molecular massof 346 kDa

Figure 2 In vitro binding of cNAPL to representative intron domains of tscAand psaA RNA (a) Secondary structure model of the first psaA intron (6)The tripartite group II intron consists of three separate RNA molecules (tscA50 intron exon 1 30 intron exon 2) Arrows indicate 50ndash30 strand polarity of thethree separate transcripts Roman letters denote the six conserved domains ofgroup II introns (DI-DVI) Black numerals correspond to nucleotides of theprocessed tscA RNA sequence and to the 30 intron sequence of psaA exon 2The gndashg 0 tertiary interaction is symbolized by a dashed line (79) The hndashh0

interaction is indicated (72) The peripheral structures of domains DI and DIVcould not be unambiguously determined and are not represented modifiedaccording to (6) (bndashe) RNA EMSAs of domain DII and DIII of tscA RNAand domain DV+VI of the first psaA intron with increasing amounts ofcNAPL EMSAs were performed by incubating 30 fmol internally labelledintron RNA (tscADII tscADIII tscADII+III psaADV+VI) with increasingamounts of affinity-purified cNAPL polypeptide (numbers indicate theamounts of protein in mg) or without protein (0) Affinity-purified CEFEFprotein of Acremonium chrysogenum was used as a negative control (C)RNAndashprotein complexes were resolved from free RNA (F) by electrophoresisin 5 native polyacrylamide gels Arrows indicate shifted bands (fndashi)Competition assays of cNAPL using 30 fmol radioactive probes of internallylabelled intron RNA (tscADII tscADIII tscADII+III psaADV+VI) andexcess of cold specific (tscADII tscADIII tscADII+III psaADV+VI) andnon-specific competitors (pBIIKS+) Lanes covered with triangle on top are0- 2- 5- and 10-fold molar excess of competitor respectively in the case oftscA RNA and 0- 25- 5- 10- and 20-fold molar excess of competitorrespectively in the case of psaA Free RNA negative control and shiftedbands are indicated as in (bndashe)

5342 Nucleic Acids Research 2006 Vol 34 No 18

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 4: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

argon-ion laser The fluorescence emission was selected bybandpass filter BP505-530 and longpass filter LP560 respec-tively using beam splitters HFT UV488543633 andNFT545

Sequences alignments and phylogenetic analysis

Sequences were retrieved from GenBank (NCBI) at httpwwwncbinlmnihgov the Creinhardtii JGI database v20(DOE Joint Genome Institute) at httpgenomejgi-psforgchlre2chlre2homehtml the Thalassiosira pseudonana JGIdatabase v10 at httpgenomejgi-psforgthaps1thaps1homehtml the Porphyra yezoensis EST index at httpwwwkazusaorjpenplantporphyraEST (2526) and theCyanoBase genome database at httpwwwkazusaorjpcyano Multiple sequence alignments of NAPs were per-formed using ClustalW software (27) Sequence alignmentswere displayed with GENEDOC (28) and refined manuallyThe phylogenetic tree was generated based on trimmedprotein sequences with programs contained in the programpackage PHYLIP version 363 (29) An alignment withsequences from mature proteins was impossible since all pro-teins differ largely in length and some were only availablefrom partial cDNAs (AU191247 BM448458) obtainedfrom EST libraries Analyses with the PHYLIP packagewere performed using PROTDISTNEIGHBOR and thePROTPARS algorithms for distance matrix and maximumparsimony analysis respectively Support for the brancheswas estimated by bootstrap analysis of 2000 replicatesgenerated with SEQBOOT and a majority rule consensustree was generated using the program CONSENSEThe resulting tree topology was then visualized inTREEVIEW (30) For analyses of putative targeting signalsequences ChloroP (31) LOCtree (32) TargetP (33)GENOPLANTETM PREDOTAR (httpwwwgenoplantecom) and PSORT (34) were used Isoelectric pointand sequence masses were calculated by the programWinPep 301 (35)

RESULTS

Identification of a tscA-specific RNA-binding proteinusing the yeast three-hybrid system

To identify RNA-binding proteins most probably involved inorganellar group II intron splicing we performed an in vivoscreen using the yeast three-hybrid system In this systeman RNAndashprotein interaction is detected by the reconstitutionof the yeast transcriptional activator GAL4 and subsequentactivation of reporter genes The system is based on onehybrid RNA and two recombinant fusion proteins containingthe GAL4 DNA-binding and GAL4 transcription activationdomains respectively To isolate chloroplast intron RNA-binding proteins of Creinhardtii a cDNA library wasscreened using the yeast three-hybrid system developed by(21) In this system the chimeric protein RevM10 fused toa gene fragment encoding the DNA-binding domain ofGAL4 (GAL4-DB RevM10) binds to RRE which is partof a chimeric RNA The 30 part of this RNA consists of thetscA RNA which functions as bait during the screen (RRE-tscA) Chimeric proteins serve as prey and are encoded bycDNA clones of a Creinhardtii cDNA library All cDNAswere fused to a gene fragment encoding the transcriptionactivation domain of GAL4 (GAL4-AD) The reliability ofthis three-hybrid screen was tested with a recombinantRNA carrying two copies of RRE (RREndashRRE) co-expressedwith the two RevM10 fusion proteins (AD-RevM10DB-RevM10) RRE and RevM10 are known to interactwith each other (21) As depicted in lane 1 of Figure 1 theinteraction resulted in reconstitution of GAL4 activity

For a global Creinhardtii cDNA screen the yeast strainCG1945 was co-transformed with pRevtscA1 expressingboth the hybrid protein DB-RevM10 and the recombinantRNA This strain served as host for transformation of thelibrary of pGAD10-derived plasmids harbouring cDNAsequences of Creinhardtii A total of 21 middot 107 clones wereanalysed and 14 clones were identified to bind to tscA RNAInterestingly of these 14 candidate cDNAs 13 encoded

Figure 1 Schematic overview of selected constructs used in the yeast three-hybrid analyses In each case five representative yeast transformants were tested fortheir ability to activate b-galactosidase by a colony colour assay Abbreviations Lhcb30 fragment of 30-UTR of the Lhcb1 transcript (37) cNAPL chloroplastNAP-like GAL4-AD activation domain of GAL4 transcription factor GAL4-DB DNA-binding domain of GAL4 transcription factor psaADV+VI domainsDV and DVI of the first psaA intron (4) RevM10 RNA-binding protein RevM10 (21) rI1DIV-VI domains DIV to DVI of intron rI1 (36) RRE Rev responsiveelement that is specifically bound by RevM10 (21) tscA processed tscA RNA (6)

5340 Nucleic Acids Research 2006 Vol 34 No 18

identical polypeptides with significant homology to NAPsBecause of its binding to chloroplast intron RNA theNAP-like polypeptide was designated as cNAPL

To assess the specificity of the three-hybrid interactionbetween tscA RNA and cNAPL several controls were carriedout The specificity of the tscA interaction was tested againsttwo further hybrid RNAs Recombinant control RNAscontaining sequences of domains DIV to DVI of the hetero-logous intron rI1 (36) (RRE-rI1DIV-VI) and a fragment ofthe 30-untranslated region (30-UTR) of the Lhcb1 transcript(37) (RRE-Lhcb30) were fused separately to the RREsequence respectively As shown in lanes 5 7 and 8 ofFigure 1 activity of the reporter b-galactosidase is dependenton the presence of the RRE-tscA hybrid RNA and thecNAPL-AD fusion protein These results indicate a specificinteraction of cNAPL with the tscA transcript In additionwe were able to demonstrate that cNAPL also binds todomains DV and DVI of the first psaA intron (RRE-psaADV+VI) as seen in lane 6 (Figure 1) To exclude one-or two-hybrid interaction or unspecific RNA interactionappropriate negative controls were also performed in the filterassay including different combinations of the three-hybridcomponents represented in lanes 2ndash4 (Figure 1) In theabsence of any one of the hybrid components transformantsdisplayed no b-galactosidase activity Results of the three-hybrid analysis therefore clearly demonstrate the specificinteraction of cNAPL with tscA RNA and the 30 part of thefirst intron of psaA RNA

In vitro binding of cNAPL to chloroplast transcripts

To confirm the specific interaction between cNAPL and tscARNA in the yeast three-hybrid system we performed in vitrobinding studies using EMSA For synthesis of cNAPL thecNapl gene was fused with six histidine residues forexpression in Ecoli After purification of the recombinantHis-tagged protein with Ni2+-NTA chromatography cNAPLwas incubated with in vitro transcribed radioactively labelledRNAs corresponding to domains DII and DIII of tscA RNAAs shown in Figure 2a domains DII and DIII range from nt75 to 218 and from 220 to 400 respectively Together theycover nt 75ndash400 RNAs were incubated in the absence orpresence of recombinant cNAPL and fractionated on nativegels (Figure 2bndashi) In Figure 2bndashd increasing amounts ofcNAPL (2ndash20 mg) were incubated with 30 fmol of in vitrosynthesized transcripts of tscA intron RNA fragmentscomprising either domain DII domain DIII or both domainsDII+III Arrows indicate RNAndashprotein complexes as multipleshifted bands that can be observed when the RNA isincubated with increasing amounts of cNAPL polypeptide

To investigate the binding specificity of cNAPL competi-tion analyses were performed The above-mentioned radio-labelled RNAs were incubated with a constant amount of15 mg of cNAPL protein in the presence of increasingamounts of unlabelled specific and non-specific competitorRNAs respectively As shown in Figure 2fndashh the presenceof just a 2-fold molar excess of unlabelled specific competitorRNA almost completely abolished the formation of acomplex between cNAPL and the labelled target tscA RNAIn contrast the presence of the same amount of non-specific competitor RNA namely pBIIKS+ (121 nt) had no

significant effect on the formation of the complex with thelabelled target RNAs In addition EMSA was performedwith in vitro synthesized RNA of domains DV and DVI ofthe first psaA intron As shown in Figure 2a the two domainsrange from nt 108 to 192 EMSAs with increasing amountsof cNAPL protein (Figure 2e) and competition analysis(Figure 2i) indicate that cNAPL binds specifically to domainsDV and DVI of the first psaA intron albeit with lower affinitythan to tscA RNA

Chloroplast transcripts are rich in A and U residues Tostudy the sequence preferences we compared the ability ofA C G or U RNA homopolymers to compete for bindingof cNAPL to the labelled domain DII of the tscA RNAand domain DV+VI of the first psaA intron respectivelyAs depicted in Figure 3a excess amounts of poly(U)abolished cNAPL binding while the same amounts ofpoly(C) or poly(G) have no significant effect on the formationof the complex Using poly(A) resulted in a weak reductionof the binding specificity As depicted in Figure 3b bindingof cNAPL to domain DV and DVI of the first psaA intronwas also competed by poly(U) and in addition by poly(A)and to a lesser extent by higher amounts of poly(G)(02 and 05 mg) The competition assays suggest thatcNAPL preferentially interacts with U-rich sequences Tofurther investigate the specificity of cNAPL for the tscARNA total wild-type RNA was used as competitor inEMSAs As shown in Figure 3a and b 05 mg of total RNAcompeted the binding of cNAPL to the tscA and psaA intronRNA respectively suggesting that there are other transcriptsbound by cNAPL Simultaneously we examined the bindingof cNAPL to transcripts of four chloroplast genes and threenuclear genes As given in Figure 3c cNAPL binds to 50-UTRs of chloroplast psbA psbD rbcL and rps4 RNA andto a lesser extent to nuclear Lhcb1 50-UTR and Rps18 Nobinding is observed to nuclear Lhcb1 30-UTR and Tba1ATo investigate the binding specificity of cNAPL to the tscARNA unlabelled RNA of the above-mentioned chloroplastand nuclear transcripts was used in competition assaysAs shown in Figure 3d a 50-fold molar excess of Lhcb1 30-UTR Tba1A and a 15-fold molar excess of Lhcb1 50-UTRhad no significant effect on the formation of the complexwhereas a 50-fold molar excess of psbA 50-UTR psbD 50-UTR and Rps18 resulted in a slight decrease in tscA RNAndashprotein complex formation Interestingly similar results incompetition experiments were obtained when binding ofcNAPL to the labelled tscA RNA was competed by a50-fold molar excess of 50-UTRs of rbcL or rps4 or by a2-fold molar excess of unlabelled tscA RNA Thus cNAPLshows specificity for the tscA RNA and to a lesser extent toother U-rich chloroplast transcripts as well The overallcomparison of all tested in vitro transcripts revealed thatbinding of cNAPL depends on the U-content and stretchesof Us (data not shown)

We further tested whether the typical secondary structuresof group II intron domains mediate cNAPL bindingIn Figure 3e only domains with U-rich sequences are ableto shift cNAPL indicating that the structure is not relevantfor binding For example the U-rich domain DV+VI of thefirst psaA intron (Figure 2e) show cNAPL binding whilethe corresponding domains of the heterologous intron rI1 ofScenedesmus obliquus do not Furthermore the experiment

Nucleic Acids Research 2006 Vol 34 No 18 5341

in Figure 3f shows that cNAPL do not have DNA-bindingactivity

Taken together these results indicate that cNAPL bindsspecifically the first psaA group II intron and thus confirm

data from the yeast three-hybrid analyses FurthermorecNAPL has affinities to U-rich chloroplast 50-UTRs Hencethe data imply that the RNA-binding protein cNAPLmay play a more general role in the RNA metabolism ofCreinhardtii

Characterization of a chloroplast NAP-like polypeptide

The nucleotide sequence of the cNapl cDNA (pGAD6-19)consists of 1642 bp with a TGTTA polyadenylation sitelocated 139 bp upstream of the poly(A) tail in the 50-UTRThe 30-UTR region is 588 bp in length and a 50 rapid ampli-fication of cDNA ends (RACE) analysis revealed 22 bp forthe 50-UTR (data not shown) Sequence analysis identifiedan open reading frame of 1053 bp and 350 amino acidsencoding a polypeptide with a predicted size of 394 kDaThe amino acid composition of the protein revealed 206acidic and 135 basic residues with an estimated pI of433 The hydropathy profile calculated according to (38)indicates mostly hydrophilic regions suggesting thatcNAPL is a soluble protein The neural-network based pre-dictors for subcellular localization of proteins ChloroP (31)and LOCtree (32) revealed a putative transit sequence forimport into the chloroplast at the N-terminal end ofcNAPL whereas TargetP proposes a mitochondrial localiza-tion However for chloroplast transit peptides of Chlamydo-monas it was shown that they are more similar tomitochondrial targeting peptides than to chloroplast transitpeptides of higher plants (39) The N-terminal amino acidsequence of cNAPL possesses features typical of chloroplasttransit peptides such as an alanine as the second residue TheN-terminal end is an uncharged region with an abundance ofarginine (136) valine (136) and alanine (182) resi-dues This transit peptide could be 46 amino acids longsince a possible cleavage site sequence Val-Gly-Ala ispresent (39) The cleavage would give rise to a predictedmature protein of 304 amino acids with a molecular massof 346 kDa

Figure 2 In vitro binding of cNAPL to representative intron domains of tscAand psaA RNA (a) Secondary structure model of the first psaA intron (6)The tripartite group II intron consists of three separate RNA molecules (tscA50 intron exon 1 30 intron exon 2) Arrows indicate 50ndash30 strand polarity of thethree separate transcripts Roman letters denote the six conserved domains ofgroup II introns (DI-DVI) Black numerals correspond to nucleotides of theprocessed tscA RNA sequence and to the 30 intron sequence of psaA exon 2The gndashg 0 tertiary interaction is symbolized by a dashed line (79) The hndashh0

interaction is indicated (72) The peripheral structures of domains DI and DIVcould not be unambiguously determined and are not represented modifiedaccording to (6) (bndashe) RNA EMSAs of domain DII and DIII of tscA RNAand domain DV+VI of the first psaA intron with increasing amounts ofcNAPL EMSAs were performed by incubating 30 fmol internally labelledintron RNA (tscADII tscADIII tscADII+III psaADV+VI) with increasingamounts of affinity-purified cNAPL polypeptide (numbers indicate theamounts of protein in mg) or without protein (0) Affinity-purified CEFEFprotein of Acremonium chrysogenum was used as a negative control (C)RNAndashprotein complexes were resolved from free RNA (F) by electrophoresisin 5 native polyacrylamide gels Arrows indicate shifted bands (fndashi)Competition assays of cNAPL using 30 fmol radioactive probes of internallylabelled intron RNA (tscADII tscADIII tscADII+III psaADV+VI) andexcess of cold specific (tscADII tscADIII tscADII+III psaADV+VI) andnon-specific competitors (pBIIKS+) Lanes covered with triangle on top are0- 2- 5- and 10-fold molar excess of competitor respectively in the case oftscA RNA and 0- 25- 5- 10- and 20-fold molar excess of competitorrespectively in the case of psaA Free RNA negative control and shiftedbands are indicated as in (bndashe)

5342 Nucleic Acids Research 2006 Vol 34 No 18

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 5: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

identical polypeptides with significant homology to NAPsBecause of its binding to chloroplast intron RNA theNAP-like polypeptide was designated as cNAPL

To assess the specificity of the three-hybrid interactionbetween tscA RNA and cNAPL several controls were carriedout The specificity of the tscA interaction was tested againsttwo further hybrid RNAs Recombinant control RNAscontaining sequences of domains DIV to DVI of the hetero-logous intron rI1 (36) (RRE-rI1DIV-VI) and a fragment ofthe 30-untranslated region (30-UTR) of the Lhcb1 transcript(37) (RRE-Lhcb30) were fused separately to the RREsequence respectively As shown in lanes 5 7 and 8 ofFigure 1 activity of the reporter b-galactosidase is dependenton the presence of the RRE-tscA hybrid RNA and thecNAPL-AD fusion protein These results indicate a specificinteraction of cNAPL with the tscA transcript In additionwe were able to demonstrate that cNAPL also binds todomains DV and DVI of the first psaA intron (RRE-psaADV+VI) as seen in lane 6 (Figure 1) To exclude one-or two-hybrid interaction or unspecific RNA interactionappropriate negative controls were also performed in the filterassay including different combinations of the three-hybridcomponents represented in lanes 2ndash4 (Figure 1) In theabsence of any one of the hybrid components transformantsdisplayed no b-galactosidase activity Results of the three-hybrid analysis therefore clearly demonstrate the specificinteraction of cNAPL with tscA RNA and the 30 part of thefirst intron of psaA RNA

In vitro binding of cNAPL to chloroplast transcripts

To confirm the specific interaction between cNAPL and tscARNA in the yeast three-hybrid system we performed in vitrobinding studies using EMSA For synthesis of cNAPL thecNapl gene was fused with six histidine residues forexpression in Ecoli After purification of the recombinantHis-tagged protein with Ni2+-NTA chromatography cNAPLwas incubated with in vitro transcribed radioactively labelledRNAs corresponding to domains DII and DIII of tscA RNAAs shown in Figure 2a domains DII and DIII range from nt75 to 218 and from 220 to 400 respectively Together theycover nt 75ndash400 RNAs were incubated in the absence orpresence of recombinant cNAPL and fractionated on nativegels (Figure 2bndashi) In Figure 2bndashd increasing amounts ofcNAPL (2ndash20 mg) were incubated with 30 fmol of in vitrosynthesized transcripts of tscA intron RNA fragmentscomprising either domain DII domain DIII or both domainsDII+III Arrows indicate RNAndashprotein complexes as multipleshifted bands that can be observed when the RNA isincubated with increasing amounts of cNAPL polypeptide

To investigate the binding specificity of cNAPL competi-tion analyses were performed The above-mentioned radio-labelled RNAs were incubated with a constant amount of15 mg of cNAPL protein in the presence of increasingamounts of unlabelled specific and non-specific competitorRNAs respectively As shown in Figure 2fndashh the presenceof just a 2-fold molar excess of unlabelled specific competitorRNA almost completely abolished the formation of acomplex between cNAPL and the labelled target tscA RNAIn contrast the presence of the same amount of non-specific competitor RNA namely pBIIKS+ (121 nt) had no

significant effect on the formation of the complex with thelabelled target RNAs In addition EMSA was performedwith in vitro synthesized RNA of domains DV and DVI ofthe first psaA intron As shown in Figure 2a the two domainsrange from nt 108 to 192 EMSAs with increasing amountsof cNAPL protein (Figure 2e) and competition analysis(Figure 2i) indicate that cNAPL binds specifically to domainsDV and DVI of the first psaA intron albeit with lower affinitythan to tscA RNA

Chloroplast transcripts are rich in A and U residues Tostudy the sequence preferences we compared the ability ofA C G or U RNA homopolymers to compete for bindingof cNAPL to the labelled domain DII of the tscA RNAand domain DV+VI of the first psaA intron respectivelyAs depicted in Figure 3a excess amounts of poly(U)abolished cNAPL binding while the same amounts ofpoly(C) or poly(G) have no significant effect on the formationof the complex Using poly(A) resulted in a weak reductionof the binding specificity As depicted in Figure 3b bindingof cNAPL to domain DV and DVI of the first psaA intronwas also competed by poly(U) and in addition by poly(A)and to a lesser extent by higher amounts of poly(G)(02 and 05 mg) The competition assays suggest thatcNAPL preferentially interacts with U-rich sequences Tofurther investigate the specificity of cNAPL for the tscARNA total wild-type RNA was used as competitor inEMSAs As shown in Figure 3a and b 05 mg of total RNAcompeted the binding of cNAPL to the tscA and psaA intronRNA respectively suggesting that there are other transcriptsbound by cNAPL Simultaneously we examined the bindingof cNAPL to transcripts of four chloroplast genes and threenuclear genes As given in Figure 3c cNAPL binds to 50-UTRs of chloroplast psbA psbD rbcL and rps4 RNA andto a lesser extent to nuclear Lhcb1 50-UTR and Rps18 Nobinding is observed to nuclear Lhcb1 30-UTR and Tba1ATo investigate the binding specificity of cNAPL to the tscARNA unlabelled RNA of the above-mentioned chloroplastand nuclear transcripts was used in competition assaysAs shown in Figure 3d a 50-fold molar excess of Lhcb1 30-UTR Tba1A and a 15-fold molar excess of Lhcb1 50-UTRhad no significant effect on the formation of the complexwhereas a 50-fold molar excess of psbA 50-UTR psbD 50-UTR and Rps18 resulted in a slight decrease in tscA RNAndashprotein complex formation Interestingly similar results incompetition experiments were obtained when binding ofcNAPL to the labelled tscA RNA was competed by a50-fold molar excess of 50-UTRs of rbcL or rps4 or by a2-fold molar excess of unlabelled tscA RNA Thus cNAPLshows specificity for the tscA RNA and to a lesser extent toother U-rich chloroplast transcripts as well The overallcomparison of all tested in vitro transcripts revealed thatbinding of cNAPL depends on the U-content and stretchesof Us (data not shown)

We further tested whether the typical secondary structuresof group II intron domains mediate cNAPL bindingIn Figure 3e only domains with U-rich sequences are ableto shift cNAPL indicating that the structure is not relevantfor binding For example the U-rich domain DV+VI of thefirst psaA intron (Figure 2e) show cNAPL binding whilethe corresponding domains of the heterologous intron rI1 ofScenedesmus obliquus do not Furthermore the experiment

Nucleic Acids Research 2006 Vol 34 No 18 5341

in Figure 3f shows that cNAPL do not have DNA-bindingactivity

Taken together these results indicate that cNAPL bindsspecifically the first psaA group II intron and thus confirm

data from the yeast three-hybrid analyses FurthermorecNAPL has affinities to U-rich chloroplast 50-UTRs Hencethe data imply that the RNA-binding protein cNAPLmay play a more general role in the RNA metabolism ofCreinhardtii

Characterization of a chloroplast NAP-like polypeptide

The nucleotide sequence of the cNapl cDNA (pGAD6-19)consists of 1642 bp with a TGTTA polyadenylation sitelocated 139 bp upstream of the poly(A) tail in the 50-UTRThe 30-UTR region is 588 bp in length and a 50 rapid ampli-fication of cDNA ends (RACE) analysis revealed 22 bp forthe 50-UTR (data not shown) Sequence analysis identifiedan open reading frame of 1053 bp and 350 amino acidsencoding a polypeptide with a predicted size of 394 kDaThe amino acid composition of the protein revealed 206acidic and 135 basic residues with an estimated pI of433 The hydropathy profile calculated according to (38)indicates mostly hydrophilic regions suggesting thatcNAPL is a soluble protein The neural-network based pre-dictors for subcellular localization of proteins ChloroP (31)and LOCtree (32) revealed a putative transit sequence forimport into the chloroplast at the N-terminal end ofcNAPL whereas TargetP proposes a mitochondrial localiza-tion However for chloroplast transit peptides of Chlamydo-monas it was shown that they are more similar tomitochondrial targeting peptides than to chloroplast transitpeptides of higher plants (39) The N-terminal amino acidsequence of cNAPL possesses features typical of chloroplasttransit peptides such as an alanine as the second residue TheN-terminal end is an uncharged region with an abundance ofarginine (136) valine (136) and alanine (182) resi-dues This transit peptide could be 46 amino acids longsince a possible cleavage site sequence Val-Gly-Ala ispresent (39) The cleavage would give rise to a predictedmature protein of 304 amino acids with a molecular massof 346 kDa

Figure 2 In vitro binding of cNAPL to representative intron domains of tscAand psaA RNA (a) Secondary structure model of the first psaA intron (6)The tripartite group II intron consists of three separate RNA molecules (tscA50 intron exon 1 30 intron exon 2) Arrows indicate 50ndash30 strand polarity of thethree separate transcripts Roman letters denote the six conserved domains ofgroup II introns (DI-DVI) Black numerals correspond to nucleotides of theprocessed tscA RNA sequence and to the 30 intron sequence of psaA exon 2The gndashg 0 tertiary interaction is symbolized by a dashed line (79) The hndashh0

interaction is indicated (72) The peripheral structures of domains DI and DIVcould not be unambiguously determined and are not represented modifiedaccording to (6) (bndashe) RNA EMSAs of domain DII and DIII of tscA RNAand domain DV+VI of the first psaA intron with increasing amounts ofcNAPL EMSAs were performed by incubating 30 fmol internally labelledintron RNA (tscADII tscADIII tscADII+III psaADV+VI) with increasingamounts of affinity-purified cNAPL polypeptide (numbers indicate theamounts of protein in mg) or without protein (0) Affinity-purified CEFEFprotein of Acremonium chrysogenum was used as a negative control (C)RNAndashprotein complexes were resolved from free RNA (F) by electrophoresisin 5 native polyacrylamide gels Arrows indicate shifted bands (fndashi)Competition assays of cNAPL using 30 fmol radioactive probes of internallylabelled intron RNA (tscADII tscADIII tscADII+III psaADV+VI) andexcess of cold specific (tscADII tscADIII tscADII+III psaADV+VI) andnon-specific competitors (pBIIKS+) Lanes covered with triangle on top are0- 2- 5- and 10-fold molar excess of competitor respectively in the case oftscA RNA and 0- 25- 5- 10- and 20-fold molar excess of competitorrespectively in the case of psaA Free RNA negative control and shiftedbands are indicated as in (bndashe)

5342 Nucleic Acids Research 2006 Vol 34 No 18

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 6: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

in Figure 3f shows that cNAPL do not have DNA-bindingactivity

Taken together these results indicate that cNAPL bindsspecifically the first psaA group II intron and thus confirm

data from the yeast three-hybrid analyses FurthermorecNAPL has affinities to U-rich chloroplast 50-UTRs Hencethe data imply that the RNA-binding protein cNAPLmay play a more general role in the RNA metabolism ofCreinhardtii

Characterization of a chloroplast NAP-like polypeptide

The nucleotide sequence of the cNapl cDNA (pGAD6-19)consists of 1642 bp with a TGTTA polyadenylation sitelocated 139 bp upstream of the poly(A) tail in the 50-UTRThe 30-UTR region is 588 bp in length and a 50 rapid ampli-fication of cDNA ends (RACE) analysis revealed 22 bp forthe 50-UTR (data not shown) Sequence analysis identifiedan open reading frame of 1053 bp and 350 amino acidsencoding a polypeptide with a predicted size of 394 kDaThe amino acid composition of the protein revealed 206acidic and 135 basic residues with an estimated pI of433 The hydropathy profile calculated according to (38)indicates mostly hydrophilic regions suggesting thatcNAPL is a soluble protein The neural-network based pre-dictors for subcellular localization of proteins ChloroP (31)and LOCtree (32) revealed a putative transit sequence forimport into the chloroplast at the N-terminal end ofcNAPL whereas TargetP proposes a mitochondrial localiza-tion However for chloroplast transit peptides of Chlamydo-monas it was shown that they are more similar tomitochondrial targeting peptides than to chloroplast transitpeptides of higher plants (39) The N-terminal amino acidsequence of cNAPL possesses features typical of chloroplasttransit peptides such as an alanine as the second residue TheN-terminal end is an uncharged region with an abundance ofarginine (136) valine (136) and alanine (182) resi-dues This transit peptide could be 46 amino acids longsince a possible cleavage site sequence Val-Gly-Ala ispresent (39) The cleavage would give rise to a predictedmature protein of 304 amino acids with a molecular massof 346 kDa

Figure 2 In vitro binding of cNAPL to representative intron domains of tscAand psaA RNA (a) Secondary structure model of the first psaA intron (6)The tripartite group II intron consists of three separate RNA molecules (tscA50 intron exon 1 30 intron exon 2) Arrows indicate 50ndash30 strand polarity of thethree separate transcripts Roman letters denote the six conserved domains ofgroup II introns (DI-DVI) Black numerals correspond to nucleotides of theprocessed tscA RNA sequence and to the 30 intron sequence of psaA exon 2The gndashg 0 tertiary interaction is symbolized by a dashed line (79) The hndashh0

interaction is indicated (72) The peripheral structures of domains DI and DIVcould not be unambiguously determined and are not represented modifiedaccording to (6) (bndashe) RNA EMSAs of domain DII and DIII of tscA RNAand domain DV+VI of the first psaA intron with increasing amounts ofcNAPL EMSAs were performed by incubating 30 fmol internally labelledintron RNA (tscADII tscADIII tscADII+III psaADV+VI) with increasingamounts of affinity-purified cNAPL polypeptide (numbers indicate theamounts of protein in mg) or without protein (0) Affinity-purified CEFEFprotein of Acremonium chrysogenum was used as a negative control (C)RNAndashprotein complexes were resolved from free RNA (F) by electrophoresisin 5 native polyacrylamide gels Arrows indicate shifted bands (fndashi)Competition assays of cNAPL using 30 fmol radioactive probes of internallylabelled intron RNA (tscADII tscADIII tscADII+III psaADV+VI) andexcess of cold specific (tscADII tscADIII tscADII+III psaADV+VI) andnon-specific competitors (pBIIKS+) Lanes covered with triangle on top are0- 2- 5- and 10-fold molar excess of competitor respectively in the case oftscA RNA and 0- 25- 5- 10- and 20-fold molar excess of competitorrespectively in the case of psaA Free RNA negative control and shiftedbands are indicated as in (bndashe)

5342 Nucleic Acids Research 2006 Vol 34 No 18

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 7: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

Comparison of the genomic and cDNA sequences of cNaplusing the Creinhardtii JGI database (DOE Joint GenomeInstitute) reveals the presence of 9 large introns and 10exons (Figure 4) As mentioned above database searchesdetected significant sequence similarity with NAPs Usingthe Creinhardtii JGI database we could also identify aChlamydomonas homologue of a NAP that does not possessa chloroplast targeting signal Furthermore another algalNAP homologue was detected in the marine centric diatomTpseudonana using the Tpseudonana JGI database (DOEJoint Genome Institute) Figure 5 shows a multiple alignmentof cNAPL with NAP polypeptides of algae and higher plantsthat displayed the highest homology using the BLASTP

algorithm (40) Regions of sequence homology betweenthese proteins were found over the entire length of theamino acid sequence including several blocks of sequenceidentity A total of 50 amino acids are identically positionedwithin all 8 sequences Overall there is 29 identitybetween cNAPL and the other NAP proteins indicatingthat cNAPL is a member of the NAP family With 46identical amino acids the Chlamydomonas NAP polypeptideshows a higher degree of homology to plant NAPs thancNAPL However Chlamydomonas cNAPL and Thalas-siosira NAP share common features atypical for NAPsThey possess for example an N-terminal extension that isnot found in plant NAPs In cNAPL this extension was

Figure 3 Binding specificity of cNAPL (a and b) RNA EMSAs of domain DII+III of tscA RNA (30 fmol) and domain DV+VI of the first psaA intron (40 fmol)with cNAPL in the presence of no competitors (-) the unlabelled RNA homopolymer competitors poly(U) poly(A) poly(C) poly(G) or the unlabelled wt totalRNA Numbers indicate the amount of competitor in mg (c) Chloroplast transcripts (30 fmol) corresponding to fragments of 50-UTRs (50) 30-UTR (30) and toregions of coding sequences were tested as indicated (d) Competition assay of cNAPL using 30 fmol of labelled domain DII+III of tscA RNA and excess of coldspecific or unspecific competitors as indicated (e) RNA EMSA of domain DI (20 fmol) DII+III (40 fmol) DIV (60 fmol) and DV+VI (60 fmol) of heterologousintron rI1 of Sobliquus with cNAPL (f) DNA EMSA of cNAPL using 30 fmol of labelled PCR fragments as indicated DNA was incubated in the absence (0) orpresence (15 mg) of recombinant cNAPL RNA (30 fmol) serves as a control for shift conditions RNA in all experiments was incubated in the absence (0) orpresence (15 mg) of affinity-purified cNAPL Free RNA controls and shifted bands are denoted as in Figure 2bndashe

Nucleic Acids Research 2006 Vol 34 No 18 5343

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 8: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

shown in silico to correspond to a chloroplast targeting sig-nal whereas an unambiguously prediction is currentlyimpossible for the extension in the sequence of ThalassiosiraNAP because of its complex plastids and multipartite plastidtargeting signals (41) Interestingly no chloroplast transit sig-nal could be identified in all currently available NAPsequences of higher plants

When we compared the protein sequences of NAPs fromdifferent sources several evolutionarily conserved domainssuch as stretches of acidic residues as well as nuclear trans-port signal sequences were detected (1242) Recently Dongand co-worker (42) identified a 16 amino acid sequence in theN-terminal regions of NAP-like sequences from rice andtobacco that is similar to the nuclear export signal (NES) ofthe yeast NAP1 protein From the comparison in Figure 5 itis evident that this 16 amino acid stretch is absent in theCreinhardtii cNAPL sequence However there are stretchesof amino acid residues in the central and C-terminal part ofthe sequences of Arabidopsis maize pea rice soybean andChlamydomonas NAP that have characteristics of nuclearlocalization signals (43) The NLS sequences are similar toPKKKKRKV the NLS found in SV40 T antigen (44) In con-trast cNAPL and Thalassiosira NAP do not possess any NLSsequences The alignment in Figure 5 reveals two clusters ofhighly acidic amino acids at the C-terminus of higher plantsand Chlamydomonas NAP both these clusters can also befound in similar positions in the amino acid sequences ofyeast NAP1 and HeLa hNRP Interestingly however neitherof these clusters is present in cNAPL and Thalassiosira NAP

Furthermore Arabidopsis maize pea rice soybean andChlamydomonas NAP contain the CKQQ sequence at theirC-terminal ends a motif suggested as a potential prenylationsignal This signal is supposed to promote membrane interac-tions and appears to play a major role in several proteinndashprotein interactions The CKQQ sequence was also identifiedin NAP polypeptides of rice and tobacco (45) The prenyla-tion signal is missing in cNAPL and Thalassiosira NAP

To analyse the expression of cNapl in the wild-type strainRTndashPCR analysis was carried out with specific primersdesigned to amplify cNapl from the start (ATG) to the stop(TAG) codon (Figure 4) As shown in Figure 6a a singleband of around the expected size of 1 kb was obtainedThis result indicates the specific expression of cNapl inCreinhardtii wild-type strain To examine cNapl expressiontotal RNA was prepared from wild-type strains CC406 cw15and arg-cw15 and photosystem I mutants T11-1 59 and T11-339 (S Glanz unpublished data) Northern blot analysis wascarried out using the 32P-labelled 17 kb fragment of thecNapl cDNA as probe As shown in Figure 6b the probe

detected a specific transcript of 16 kb which is consistentwith the expected size of the cNapl transcript An equal loadcheck using Rps18 (46) revealed no difference between tran-script accumulation in wild-type versus photosystem I mutantstrains Thus pleiotropic phenotypes that are often associatedwith photosynthetic defects do not affect transcriptionalexpression of the cNapl gene

Cellular localization of the cNAPL

The N-terminal 46 amino acids of cNAPL were predicted tobe a signal sequence because of certain structural features Todetermine the cellular localization of the cNAPL polypeptidewe performed LSCFM A map of the constructs is shown inFigure 7a As can be seen in Figure 7b non-transformedarg-cw15 cells (wt) do not show any fluorescence whengreen fluorescent protein (GFP) was excited The completeN-terminal signal sequence (46 amino acids) and thesequence for the first 38 amino acids of the mature cNAPLwere fused to a synthetic GFP sequence (47) (N84-cGFP)in frame keeping the putative N-terminal transit peptidefunctional Previous work on diatom protein localizationestablished that the complete pre-sequence of plastid prepro-teins is sufficient to drive the import of GFP into plastids(41) A similar strategy was successfully applied recentlyfor localization of the phage-type organellar RNA poly-merases of Arabidopsis thaliana and Chenopodium album(48) To determine the cellular localization of cNAPL inthe algal cell the cNaplN84-cgfp fusion gene was expressedin Creinhardtii cells under the control of the HSP70ARBCS2promoter (49) As shown in Figure 7b LSCFM confirmed aplastidic localization of cNAPL The fluorescent imagesreveal a co-localization of the chimeric N84-cGFP polypep-tide (green fluorescence) with the chloroplast (red autofluo-rescence of plastids) Right panels show the merged imagesof cGFP fluorescence and chlorophyll autofluorescence ofthe same cell To rule out the possibility that use of thenon-native promoter HSP70ARBCS2 leads to mis-targetingof cNAPL a gene for a fusion protein with a 31 aminoacid deletion in the chloroplast targeting sequence ofcNAPLN84 was constructed (N84D8-39-cGFP) As shownin Figure 7b this protein clearly localizes in the cytoplasmand to spot-like structures outside of the autofluorescingchloroplast As further control for nuclear localization weused cGFP in fusion with the Ble polypeptide (Ble-cGFP)The bacterial ble gene has been successfully expressed inCreinhardtii by fusing the coding sequence to the 50 and 30

non-coding sequence of the RBCS2 gene containing addi-tional intron sequences (50) For the Ble polypeptide it has

Figure 4 Physical map of the cNapl gene Exons (1ndash10) and introns are represented by closed and open boxes respectively with their corresponding size in basepairs The arrow indicates transcriptional direction Oligoucleotides used for RTndashPCR analysis are given as arrowheads

5344 Nucleic Acids Research 2006 Vol 34 No 18

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 9: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

Figure 5 Amino acid sequence comparison of Creinhardtii cNAPL NAP and NAPs of higher plants and Tpseudonana Black shading denotes distinctsimilarities of the tscA-binding protein cNAPL to NAPs Grey shading shows the higher degree of homology of the NAP protein from Creinhardtii to NAPproteins of higher plants in contrast to cNAPL Arrows delimit the NAP domain A putative prenylation motif is underlined putative nuclear localization signalsare framed and a putative NES is underlined with lsquorsquo A putative chloroplast targeting signal sequence is represented by lsquo+rsquo on top of the alignment The twoacidic amino acid regions are underlined with dashed lines An asterisk indicates identical residues Similar residues are indicated by one or two dots dependingon the degree of similarity of the relevant amino acids The following protein sequences were used for comparison At Athaliana AAL47354 Cr CreinhardtiiC_1660026 (JGI) Gm Glycine max AAA88792 Os Oryza sativa XP_475795 Ps Pisum sativum T06807 Tp Tpseudonana 115707 (JGI) Zm Zea maysAY100480

Nucleic Acids Research 2006 Vol 34 No 18 5345

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 10: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

been shown previously that the bacterial protein is localizedin the nucleus (51) This localization was also confirmed byour experiments as demonstrated in Figure 7b and therewas no congruence of cGFP fluorescence with the positionof the plastid as observed by fluorescence microscopy Inaddition we used the C-terminal cGFP-fused Rps18 polypep-tide (Rps18-cGFP) as a marker for cytoplasmic localizationThe corresponding cGFP fluorescence signals appeared toaccumulate in spots within the cytoplasm To the best ofour knowledge this is the first demonstration of a cytoplas-mic localization of a Creinhardtii protein using GFP In con-clusion our results from microscopic investigations clearlydemonstrate that the N-terminal extension of cNAPL func-tions as a chloroplast targeting signal that is able to directcNAPL to the organelle

Phylogenetic analyses

The members of the NAP family are considered as histonechaperones which are conserved from yeast to man Toexamine the evolutionary relationships between cNAPL andthe above-mentioned homologues we performed a phyloge-netic analysis using the neighbour-joining method A treewas constructed from multiple alignments of the NAPdomains of cNAPL and of NAP polypeptides from variouseukaryotes In addition to the NAP homologue in the diatomTpseudonana a partial sequence of a predicted protein withhomology to NAPs was found in the EST database of the redalga Porphyra yezoenis (Kazusa DNA Research Institute)(2526) Similarly we detected another partial EST homo-logue (GenBank) from the green alga Dunaliella salinaSo far genomic sequences of these algae are not availableyet For prokaryotes however we were unable to detect

any NAP or cNAPL homologues in the publicly availablegenome sequences of bacteria (GenBank) and cyanobacteria(CyanoBase)

The unrooted tree in Figure 8 shows four main clustersanimals algae fungi and higher plants Trees generatedwith maximum parsimony and Bayesian analysis (52) hadan overall similar topography (data not shown) FromFigure 8 it becomes evident that the Chlamydomonas NAPpolypeptide is related to the group of higher plant NAP pro-teins which all lack a chloroplast targeting sequencewhereas cNAPL clusters together with the algal NAP homo-logues For the NAP domain the amino acid sequence ofChlamydomonas NAP shows 54 identity to higher plantNAPs whereas the NAP domain of cNAPL exhibits 35identity to higher plant NAPs The corresponding valuesfor the algal NAPs are 35 for Thalassiosira NAP26 for Dunaliella NAP and 15 for Porphyra NAPOverall we propose that cNAPL with its chloroplast transitpeptide represents a new type of NAP-like polypeptide inphotoautotrophic organisms

DISCUSSION

cNAPL is an tscA-specific RNA-binding protein

The splicing chemistry and RNA structures of self-splicinggroup II introns and nuclear pre-mRNA introns are strikinglysimilar Thereby suggesting some evolutionary relationshipbetween group II introns and the nuclear spliceosomal intron(53) Moreover there is compelling evidence to suggest thatchloroplast spliceosomes catalyse intron splicing using asimilar mechanism to that reported for nuclear spliceosomesThis suggestion is the result of data obtained form a combina-tion of different experimental approaches that have identifiedcomponents of a putative chloroplast spliceosome inCreinhardtii (7ndash11) and higher plants (54) Additional sup-port comes from studies with tobacco chloroplasts Nakamuraand co-workers (55) identified chloroplast ribonucleoproteins(cpRNPs) that are associated in vivo with various species ofchloroplast mRNAs and intron-containing precursor tRNAsThey suggested that these stromal RNAndashprotein complexespromote for example RNA splicing processing or editingPreviously the yeast three-hybrid system was used to demon-strate specific binding of proteins to organellar intron RNA(856) In this work we used the yeast three-hybrid systemsuccessfully to screen a Creinhardtii cDNA library and wewere able to detect the novel organellar protein cNAPL thatspecifically binds to the first psaA group II intron The bind-ing of cNAPL to domains DII+III and DV+VI of the firstpsaA intron was further shown in electromobility shift assays

In several other studies using defined intron domains sepa-rated molecules were found to fold into a structure corre-sponding to their conformation in the complete activeintronic RNA (5758) Recent studies have demonstratedthat fragmented intron molecules are appropriate targets forgel retardation assays (22) In EMSA we observed multipleshifted bands after incubating domains DII+III of tscARNA with increasing amounts of cNAPL As was shownfor example for the mutated SxlN1 protein of Drosophilamultiple shifted bands probably correspond to the sequentialfilling of binding sites when protein concentration is

Figure 6 Analyses of cNapl transcript (a) RTndashPCR analysis of cNapltranscripts in wild-type arg-cw15 RTndashPCR was carried out by using eitherDNase treated poly(A) RNA as a control (cw15 RNA) cDNA (cw15 cDNA)or genomic DNA (cw15 DNA) as template Oligonucleotides used for RTndashPCR are for_6-19_BspHI and rev_6-19_2xMet_HindIII Their position isshown in Figure 4 Oligonucleotides rpsF2 and rpsR2 specifically amplify afragment of the Rps18 transcript and serve as a control for cDNA preparationAbbreviation Rps18 cytoplasmic ribosomal protein S18 of Creinhardtii(46) (b) RNA blot analysis of cNapl transcripts in wild-type and non-photosynthetic mutant strains Total RNA (20 mg) of wild-type strains CC406cw15 and arg-cw15 and photosystem I mutants T11-3 39 and T11-1 59(S Glanz unpublished data) were separated on a 1 agaroseformaldehydegel transferred onto a nylon membrane and hybridised separately with acNapl- and Rps18-specific probe as a loading control

5346 Nucleic Acids Research 2006 Vol 34 No 18

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 11: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

increased (59) Hence the two shifted bands observed inFigure 2bndashd shows non-cooperative binding of cNAPL todomains DII and DIII thus indicating at least two bindingsites However we were not able to detect a known RNA-binding motif [httpwwwelsnet (60)] in cNAPL such asthe RNP motif the K homology motif (KH) or the RGG(Arg-Gly-Gly) box Further binding studies with truncatedcNAPL constructs could help to determine the exact RNA-binding domain

cNAPL is phylogenetically related to nuclear localizedNAPs and most probably has not been acquired from acyanobacterial endosymbiont

cNAPL is a member of the multifunctional family of NAPsthat are involved in processes which normally take place inthe nucleus or the cytoplasm of various eukaryotes

(124261) Besides their distinct functions in mitotic eventsand cytokinesis NAPs act as histone chaperones binding toall core histones with a preference for H2A and H2B andworking as a deposition factor by transferring NAP1-boundhistones to double-stranded DNA In addition to its nucleo-some assembly and histone-binding activity NAP1 as wellas plant NAP1-like proteins may be involved in regulatinggene expression and therefore cellular differentiation(456263) NAP1 has also been shown to facilitate transcrip-tion factor binding by disruption of the histone octamerthrough the binding of H2A and H2B (1364) NAP1 also shut-tles histones H2A and H2B as well as a complex of H2A H2BH3 and H4 between the template DNA and nascent RNA dur-ing transcription (65) To date however it is still unknownwhether NAP1 interacts directly also with RNA

Phylogenetic analysis indicates that cNAPL is related tonuclear localized NAP polypeptides The clustering of

Figure 7 Confocal fluorescence microscopy to demonstrate chloroplast localization of the cNAPL polypeptide (a) Schematic representation of fusion proteinconstructs used for cGFP-assays (b) Laser scanning confocal microscopy of Creinhardtii arg-cw15 transformants Transformants were assayed by differentialinterference contrast microscopy (DIC) or by confocal fluorescence microscopy DIC cGFP fluorescence (green) and chlorophyll autofluorescence (red) weremerged as indicated The DIC and cGFP images are representatives of five independent experiments Scale bar represents 5 mm Abbreviations ble phleomycinresistance gene of Streptoalloteichus hindustanus (50) cgfp synthetic GFP (47) N84 the N-terminal 84 amino acids of cNAPL N84D8-39 the N-terminal 84amino acids of cNAPL lacking 31 amino acids in the chloroplast targeting sequence P HSP70ARBCS2 promoter (49) Rps18 cytoplasmic ribosomal proteinS18 of Creinhardtii T 30-UTR of Lhcb1 or of RBCS2 gene

Nucleic Acids Research 2006 Vol 34 No 18 5347

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 12: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

cNAPL with algal NAPs and not with NAPs of higher plantscould be explained by its chloroplast targeting signal In thecase of Thalassiosira NAP TargetP predicted a 69 aminoacid signal sequence which could not be assigned clearlyto any organelle Diatoms such as Tpseudonana possessso-called lsquocomplex plastidsrsquo delineated by four distinct mem-branes For a nuclear-encoded plastid polypeptide to bedirected to the plastids multiple targeting signals are neces-sary Thus determining a proteinrsquos localization in diatomsusing todayrsquos prediction programs is difficult

It is now generally accepted that double-membrane-boundchloroplasts are the result of an endosymbiotic event invol-ving a cyanobacterial-like organism early on in evolutionDuring evolution the cyanobacterial endosymbiont has lostits autonomy and this has been accompanied by significantchanges of the chloroplast proteome (1466) A key factorin this process of adoption was the loss of genetic materialresulting from gene transfer to the cell nucleus The vastmajority of proteins in present day chloroplasts are encodedby the host nucleus and require N-terminal presequencesthat target them back to the chloroplast Therefore in orderto establish a functional organellendashnucleus interactionchloroplasts have had to import a set of new proteins to

adapt their metabolism to the new conditions Interestinglya proportion of these proteins do not seem to have beenacquired from cyanobacteria (14) Phylogenetic analysisand database research revealed that no NAP homologuescould be identified in prokaryotic organisms We thereforesuggest that cNAPL does not originate from an endosymbi-otic gene transfer event and that the chloroplast of Crein-hardtii has gained cNAPL as a novel nuclear-encodedfactor The phylogenetic analysis suggests that cNapl origi-nate from a nuclear Nap gene and is probably the result ofa gene duplication event that occurred after separation ofhigher plants from green algae We thus propose thatcNAPL with its chloroplast transit peptide represents a newtype of NAP-like polypeptide

Is cNAPL a multifunctional protein

Many group II introns of plants have lost their self-splicingactivity during evolution and thus recruited host-encodedproteins as splicing factors (167) The participation ofnuclear-encoded proteins in chloroplast splicing couldprovide a means to control the biogenesis of the photosyn-thetic apparatus It has been suggested that these recruited

Figure 8 Phylogenetic tree of 31 NAPNAP-like proteins The tree was generated using PHYLIP (29) as indicated in Material and Methods Numbers atbranches indicate bootstrap support (2000 bootstrap replicates) in percent Abbreviations Af Aspergillus fumigatus XP_754077 At1 Athaliana AAL47354At2 Athaliana BAA97025 At3 Athaliana NP_194341 Cn Cryptococus neoformans AAW43807 Dd Dictyostelium discoideum EAL71995 DmDrosophila melanogaster AAB07898 Ds Dsalina BM448458 Gm Gmax AAA88792 Hs1 Homo sapiens BAA08904 Hs2 Hsapiens NP_004528MgMagnaporthe grisea AAX076 Nc Neurospora crassa CAD70974 Nt1 Nicotiana tabacum CAD27460 Nt2 Ntabacum CAD27461 Nt3 NtabacumCAD27462 Os1 Osativa AAV88624 Os2 Osativa CAD27459 Os3 Osativa P0456F08 Os4 Osativa XP_475795 Pc Paramecium caudatumBAB79235 Ps Psativum T06807 Py Pyezoensis AU191247 Rn1 Rattus norvegicus AAC67388 Rn2 Rnorvegicus NP_446013 Sc ScerevisiaeP25293 Sp Schizzosaccharomyces pombe T41330 Tb Trypanosoma brucei XP_826968 Tp Tpseudonana 115707 (JGI)

5348 Nucleic Acids Research 2006 Vol 34 No 18

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 13: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

host-encoded proteins probably had other functions in thecell Many of these proteins show some degree of homologyto proteins involved in RNA metabolism For example Raa2of Creinhardtii shows homology to pseudouridine syn-thetase however such enzymatic activity is not requiredfor trans-splicing of psaA RNA (7) Another example isMss116p of Saccharomyces cerevisiae which influences thesplicing of all nine group I and all four group II introns inmitochondria (68) Mss116p is related to DEAD-box proteinswith RNA chaperone function and may facilitate splicing byresolving misfolded introns However it is generally dis-cussed that proteins involved in splicing and interactingwith intron RNAs seem to support RNA folding or to stabi-lize the active conformation whereas the catalytic potentialis clearly located in the RNA itself (69) FurthermoreMss116p is a bifunctional protein that influences in additionto splicing of group I and group II introns some RNA end-processing reactions and translation of a subset of mitochon-drial mRNAs Indeed for several nuclear-encoded host pro-teins it is known that they have additional cellularfunctions and seem to have been recruited for splicing duringevolution (1) Therefore it has been questioned whethercNAPL is also a multifunctional protein The analysed inter-actions of cNAPL with tscA RNA and domains DV and DVIof the first psaA intron implicate a role for cNAPL as a poten-tial tscA RNA processing factor and also as a splicing factorof the first psaA intron An example for such a bifunctionalprotein comes from recent work where Rat1 is required forboth processing of tscA RNA as well as for splicing of thefirst psaA intron (8) Another factor involved in tscA RNAprocessing and splicing of both introns is Raa1-L137H (9)allelic to HN31 (70)

cNAPL binds specifically to domains DII and DVI of thefirst psaA intron Domain DII is a phylogenetically less-conserved region and was assigned as functionally unimpor-tant because of its high degree of sequence and structurevariations (71) However Chanfreau and Jacquier (72)could show that the so-called tertiary hndashh0 interaction (seealso Figure 2a) between DII and DVI of group IIA intronsis responsible for a conformational change between the twocatalytic transesterification steps Interestingly only a fewof the known group IIB introns can potentially form thisinteraction It is discussed that the majority of the groupIIB introns interact with protein factors to change the con-formation between the first and the second step (69) Thefact that cNAPL interacts with both domains DII and DVIimplies a role for cNAPL in this conformational change

Our experiments further demonstrated the binding ofcNAPL to domain DV of the first psaA intron Domain DVis the phylogenetically most conserved sequence and essen-tial for the catalytic splicing reaction DV shows numerousimportant tertiary interactions and contains important con-served sequence motifs (69) Overall we assume thatcNAPL presumably has a function in the stabilization andcorrect folding of the catalytic core of the first psaA intronby interacting with tscA RNA and domains DV and DVI

In addition our analyses showed that the binding ofcNAPL to tscA RNA was competed efficiently by poly(U)in contrast to other RNA homopolymers Moreover cNAPLbinds the 50-UTR of U-rich chloroplast transcripts psbApsbD rbcL and rps4 considerably better than the nuclear

transcripts Lhcb1 50-UTR Lhcb1 30-UTR Rps18 andTba1A It is proven that the 50-UTRs of chloroplast transcriptscontain the cis-acting determinants for the stabilization ofplastid transcripts and are essential for initiation of chloro-plast translation by binding trans-acting factors (mostlyencoded by nuclear genes) and by interaction with theribosomal preinitiation complex (73) For example the stabi-lity of the psbD mRNA depends on a nucleus-encoded TPRprotein which is part of a high-molecular weight complexmediating its function via the psbD 50-UTR (74) For thepsbA mRNA it is assumed that a multiprotein complex specif-ically regulates translation of this mRNA through interactionwith a U-rich sequence in the 50-UTR (75) Furthermorethere are several proteins in Chlamydomonas binding to 50-UTRs of chloroplast mRNAs eg to 50-UTRs of atpBrbcL rps7 and rps12 that so far have not been characterizedand are discussed to be required for translational regulation(7677) However the cpRNP complexes seem to be notonly involved in regulation of translation but also in distinctsteps of post-transcriptional gene expression eg by facili-tating proper RNA folding for intron-containing RNA (78)Recently it was demonstrated that there are two RNP com-plexes in Creinhardtii that are involved in trans-splicing ofpsaA one in the soluble fraction of the chloroplast and theother one associated with chloroplast membranes whichcould provide a link to translation (11)

In conclusion we have isolated a novel type of group IIintron RNA-binding protein using the yeast three-hybridsystem Phylogenetic analysis indicates that the chloroplastlocated cNAPL protein is encoded by a nuclear gene thatprobably has evolved from duplication of an ancestral Napgene Moreover it is proposed that cpRNP complexes areglobal mediators of chloroplast RNA metabolism whichconnect transcription and translation in the chloroplast (78)Because of cNAPLrsquos binding to U-rich sequences of chloro-plast 50-UTRs a more general role for this RNA-bindingprotein is conceivable eg in translation initiation of severalchloroplast transcripts Thus cNAPL might represent acomponent of the cpRNP complex

ACKNOWLEDGEMENTS

We wish to express our thanks to Profs Peter Hegemann(Berlin) Christoph Beck (Freiburg) and Dr Ulrich Putz(Hamburg) for the gift of plasmids and Drs Birgit Hoff andStephan Pollmann for advice in LSCFM and Patricia Cebulaand Franziska Thorwirth for help with some experimentsThis work was financially supported by the DeutscheForschungsgemeinschaft (SFB480 B3) Funding to pay theOpen Access publication charges for this article was providedby DFG (Bonn Bad Godesberg Germany)

Conflict of interest statement None declared

REFERENCES

1 LambowitzAM and ZimmerlyS (2004) Mobile group II intronsAnnu Rev Genet 38 1ndash35

2 BarkanA and Goldschmidt-ClermontM (2000) Participation ofnuclear genes in chloroplast gene expression Biochimie 82 559ndash572

Nucleic Acids Research 2006 Vol 34 No 18 5349

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 14: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

3 NickelsenJ (2003) Chloroplast RNA-binding proteins Curr Genet43 392ndash399

4 KuckU ChoquetY SchneiderM DronM and BennounP (1987)Structural and transcriptional analysis of two homologous genes for theP700 chlorophyll a-apoproteins in Chlamydomonas reinhardtiievidence for in vivo trans-splicing EMBO J 6 2185ndash2195

5 ChoquetY Goldschmidt-ClermontM Girard-BascouJ KuckUBennounP and RochaixJD (1988) Mutant phenotypes support atrans-splicing mechanism for the expression of the tripartite psaA genein the C reinhardtii chloroplast Cell 52 903ndash913

6 Goldschmidt-ClermontM ChoquetY Girard-BascouJ MichelFSchirmer-RahireM and RochaixJD (1991) A small chloroplast RNAmay be required for trans-splicing in Chlamydomonas reinhardtii Cell65 135ndash143

7 PerronK Goldschmidt-ClermontM and RochaixJD (1999) A factorrelated to pseudouridine synthases is required for chloroplast group IIintron trans-splicing in Chlamydomonas reinhardtii EMBO J 186481ndash6490

8 BalczunC BunseA HahnD BennounP NickelsenJ and KuckU(2005) Two adjacent nuclear genes are required for functionalcomplementation of a chloroplast trans-splicing mutant fromChlamydomonas reinhardtii Plant J 43 636ndash648

9 MerendinoL PerronK RahireM HowaldI RochaixJD andGoldschmidt-ClermontM (2006) A novel multifunctional factorinvolved in trans-splicing of chloroplast introns in ChlamydomonasNucleic Acids Res 34 262ndash274

10 RivierC Goldschmidt-ClermontM and RochaixJD (2001)Identification of an RNAndashprotein complex involved in chloroplastgroup II intron trans-splicing in Chlamydomonas reinhardtii EMBO J20 1765ndash1773

11 PerronK Goldschmidt-ClermontM and RochaixJD (2004) Amultiprotein complex involved in chloroplast group II intron splicingRNA 10 704ndash711

12 KrudeT and KellerC (2001) Chromatin assembly during S phasecontributions from histone deposition DNA replication and the celldivision cycle Cell Mol Life Sci 58 665ndash672

13 ItoT IkeharaT NakagawaT KrausWL and MuramatsuM (2000)p300-Mediated acetylation facilitates the transfer of histone H2A-H2Bdimers from nucleosomes to a histone chaperone Gene Dev 141899ndash1907

14 TimmisJN AyliffeMA HuangCY and MartinW (2004)Endosymbiotic gene transfer organelle genomes forge eukaryoticchromosomes Nature Rev Genet 5 123ndashU116

15 HarrisEH (1989) The Chlamydomonas Sourcebook AComprehensive Guide to Biology and Laboratory Use Academic PressSan Diego CA

16 KindleKL (1990) High-frequency nuclear transformation ofChlamydomonas reinhardtii Proc Natl Acad Sci USA 871228ndash1232

17 ZorinB HegemannP and SizovaI (2005) Nuclear genetargeting by using single-stranded DNA avoids illegitimate DNAintegration in Chlamydomonas reinhardtii Eukaryotic Cell 41264ndash1272

18 SambrookJ and RusselDW (2001) Molecular Cloning - ALaboratory Manual Cold Spring Harbor Laboratory Press Cold SpringHarbor NY USA

19 JerpsethB GreenerA ShortJM ViolaJ and KretzPL (1992)XL1-Blue MRFrsquo E coli cells McrA- McrCB- McrF- Mrr- HsdR-derivative of XL1-Blue cells Strateg Mol Biol 5 81ndash83

20 BalczunC BunseA NowrousianM KorbelA GlanzS andKuckU (2005) DNA macroarray and real-time PCR analysis of twonuclear photosystem I mutants from Chlamydomonas reinhardtii revealdownregulation of Lhcb genes but different regulation of Lhca genesBiochim Biophys Acta 1732 62ndash68

21 PutzU SkehelP and KuhlD (1996) A tri-hybrid system for theanalysis and detection of RNAndashprotein interactions Nucleic Acids Res24 4838ndash4840

22 BunseAA NickelsenJ and KuckU (2001) Intron-specific RNAbinding proteins in the chloroplast of the green alga Chlamydomonasreinhardtii Biochim Biophys Acta 1519 46ndash54

23 BalczunC BunseA SchwarzC PiotrowskiM and KuckU (2006)Chloroplast heat shock protein Cpn60 from Chlamydomonasreinhadrtii exhibits a novel function as a group II intron-specificRNA-binding protein FEBS lett 580 4527ndash4532

24 KuchoK YoshiokaS TaniguchiF OhyamaK and FukuzawaH(2003) Cis-acting elements and DNA-binding proteins involved inCO2-responsive transcriptional activation of Cah1 encoding aperiplasmic carbonic anhydrase Chlamydomonas reinhardtii PlantPhysiol 133 783ndash793

25 AsamizuE NakajimaM KitadeY SagaN NakamuraY andTabataS (2003) Comparison of RNA expression profiles betweenthe two generations of Porphyra yezoensis (Rhodophyta)based on expressed sequence tag frequency analysis J Phycol 39923ndash930

26 NikaidoI AsamizuE NakajimaM NakamuraY SagaN andTabataS (2000) Generation of 10 154 expressed sequence tags from aleafy gametophyte of a marine red alga Porphyra yezoensis DNA Res7 223ndash227

27 ThompsonJD HigginsDG and GibsonTJ (1994)ClustalWmdashimproving the sensitivity of progressive multiple sequencealignment through sequence weighting position-specific gap penaltiesand weight matrix choice Nucleic Acids Res 22 4673ndash4680

28 NicholasKB NicholasHBJ and DeerfieldDW (1997) GeneDocanalysis and visualization of genetic variation EMBNET News 4 1ndash4

29 FelsensteinJ (1989) PHYLIP phylogeny inference packageCladistics 5 164ndash166

30 PageRDM (1996) TreeView an application to display phylogenetictrees on personal computers Comput Appl Biosci 12 357ndash358

31 EmanuelssonO NielsenH and Von HeijneG (1999) ChloroP aneural network-based method for predicting chloroplast transit peptidesand their cleavage sites Protein Sci 8 978ndash984

32 NairR and RostB (2005) Mimicking cellular sorting improvesprediction of subcellular localization J Mol Biol 348 85ndash100

33 EmanuelssonO NielsenH BrunakS and von HeijneG (2000)Predicting subcellular localization of proteins based on their N-terminalamino acid sequence J Mol Biol 300 1005ndash1016

34 NakaiK and KanehisaM (1991) Expert system for predicting proteinlocalization sites in gram-negative bacteria Proteins 11 95ndash110

35 HennigL (1999) WinGeneWinPep user-friendly software for theanalysis of amino acid sequences Biotechniques 26 1170ndash1172

36 HerdenbergerF HollanderV and KuckU (1994) Correct in vivoRNA splicing of a mitochondrial intron in algal chloroplasts NucleicAcids Res 22 2869ndash2875

37 JasperF QuednauB KortenjannM and JohanningmeierU (1991)Control of cab gene expression in synchronized Chlamydomonasreinhardtii cells J Photochem Photobiol B Biol 11 139ndash150

38 KyteJ and DoolittleRF (1982) A simple method for displaying thehydropathic character of a protein J Mol Biol 157 105ndash132

39 FranzenLG RochaixJD and von HeijneG (1990) Chloroplasttransit peptides from the green alga Chlamydomonas reinhardtii sharefeatures with both mitochondrial and higher plant chloroplastpresequences FEBS Lett 260 165ndash168

40 AltschulSF GishW MillerW MyersEW and LipmanDJ (1990)Basic local alignment search tool J Mol Biol 215 403ndash410

41 AptKE ZaslavkaiaL LippmeierJC LangM KilianOWetherbeeR GrossmanAR and KrothPG (2002) In vivocharacterization of diatom multipartite plastid targeting signals J CellSci 115 4061ndash4069

42 DongAW LiuZQ ZhuY YuF LiZY CaoKM andShenWH (2005) Interacting proteins and differences in nucleartransport reveal specific functions for the NAP1 family proteins inplants Plant Physiol 138 1446ndash1456

43 BoulikasT (1993) Nuclear localization signals (NLS) Crit RevEukaryot Gene Expr 3 193ndash227

44 KalderonD RobertsBL RichardsonWD and SmithAE (1984) Ashort amino acid sequence able to specify nuclear location Cell 39499ndash509

45 DongAW ZhuY YuY CaoKM SunCR and ShenWH (2003)Regulation of biosynthesis and intracellular localization of rice andtobacco homologues of nucleosome assembly protein 1 Planta 216561ndash570

46 HahnD and KuckU (1995) cDNA nucleotide sequences andexpression of the genes encoding the cytoplasmic ribosomal proteinsS18 and S27 from the green alga Chlamydomonas reinhardtii PlantSci 111 73ndash79

47 FuhrmannM OertelW and HegemannP (1999) A synthetic genecoding for the green fluorescent protein (GFP) is a versatile reporter inChlamydomonas reinhardtii Plant J 19 353ndash361

5350 Nucleic Acids Research 2006 Vol 34 No 18

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351

Page 15: A nucleosome assembly protein-like polypeptide binds to ......splicing introns were found in the psaA gene encoding the major P700 chlorophyll a/b-binding protein (4). In the case

48 HedtkeB MeixnerM GillandtS RichterE BornerT andWeiheA (1999) Green fluorescent protein as a marker to investigatetargeting of organellar RNA polymerases of higher plants in vivo PlantJ 17 557ndash561

49 SchrodaM BlockerD and BeckCF (2000) The HSP70A promoteras a tool for the improved expression of transgenes in ChlamydomonasPlant J 21 121ndash131

50 LumbrerasV StevensDR and PurtonS (1998) Efficient foreigngene expression in Chlamydomonas reinhardtii mediated by anendogenous intron Plant J 14 441ndash447

51 CalmelsTPG MistryJS WatkinsSC RobbinsPD McguireRand LazoJS (1993) Nuclear localization of bacterialStreptoalloteichus hindustanus bleomycin resistance protein inmammalian cells Mol Pharmacol 44 1135ndash1141

52 RonquistF and HuelsenbeckJP (2003) MRBAYES 3 Bayesianphylogenetic inference under mixed models Bioinformatics 191572ndash1574

53 ValadkhanS (2005) snRNAs as the catalysts of pre-mRNA splicingCurr Opin Chem Biol 9 603ndash608

54 OstheimerGJ RojasM HadjivassiliouH and BarkanA (2006)Formation of the CRS2-CAF2 group II intron splicing complex ismediated by a 22-amino acid motif in the COOH-terminal region ofCAF2 J Biol Chem 281 4732ndash4738

55 NakamuraT OhtaM SugiuraM and SugitaM (1999) Chloroplastribonucleoproteins are associated with both mRNAs andintron-containing precursor tRNAs FEBS Lett 460437ndash441

56 JaegerS ErianiG and MartinF (2004) Results and prospectrsquos of theyeast three-hybrid system FEBS Lett 556 7ndash12

57 FranzenJS ZhangMC and PeeblesCL (1993) Kinetic analysis ofthe 50 splice junction hydrolysis of a group II intron promoted bydomain 5 Nucleic Acids Res 21 627ndash634

58 PyleAM and GreenJB (1994) Building a kinetic framework forgroup II intron ribozyme activity quantitation of interdomain bindingand reaction rate Biochemistry 33 2716ndash2725

59 BlackDL ChanR MinH WangJ and BellL (1998) Theelectrophoretic mobility shift assay for RNA binding proteinsIn SmithCWJ (ed) RNA Protein Interactions A PracticalApproach The Practical Approach Series IRL Press Oxford UKVol 192 pp 109ndash136

60 Houser-ScottF and EngelkeDR (2001) Protein-RNA InteractionsEncyclopedia of Life Sciences John Wiley amp Sons Ltd ChichesterUK

61 SteerWM Abu-DayaA BrickwoodSJ MumfordKLJordanairesN MitchellJ RobinsonC ThomeAW and GuilleMJ(2003) Xenopus nucleosome assembly protein becomes tissue-restrictedduring development and can alter the expression of specific genesMech Develop 120 1045ndash1057

62 KawaseH OkuwakiM MiyajiM OhbaR HandaH IshimiYFujiiNakataT KikuchiA and NagataK (1996) NAP-I is a functionalhomologue of TAF-I that is required for replication and transcription ofthe adenovirus genome in a chromatin-like structure Genes Cells 11045ndash1056

63 ShikamaN ChanHM Krstic-DemonacosM SmithL LeeCWCairnsW and La ThangueNB (2000) Functional interaction betweennucleosome assembly proteins and p300CREB-binding protein familycoactivators Mol Cell Biol 20 8933ndash8943

64 WalterPP OwenhughesTA CoteJ and WorkmanJL (1995)Stimulation of transcription factor-binding and histone displacement bynucleosome assembly protein-1 and nucleoplasmin requires disruptionof the histone octamer Mol Cell Biol 15 6178ndash6187

65 LevchenkoV and JacksonV (2004) Histone release duringtranscription NAP1 forms a complex with H2A and H2B andfacilitates a topologically dependent release of H3 and H4 from thenucleosome Biochemistry 43 2359ndash2372

66 LeisterD (2003) Chloroplast research in the genomic age TrendsGenet 19 47ndash56

67 BarkanA (2004) Intron splicing in plant organelles In DaniellH andChaseC (eds) Molecular Biology and Biotechnology of PlantOrganelles Kluwer Academic Publishers Dordrecht The Netherlandspp 281ndash308

68 MohrS MatsuuraM PerlmanPS and LambowitzAM (2006) ADEAD-box protein alone promotes group II intron splicing and reversesplicing by acting as an RNA chaperone Proc Natl Acad Sci USA103 3569ndash3574

69 LehmannK and SchmidtU (2003) Group II introns structure andcatalytic versatility of large natural ribozymes Crit Rev BiochemMol 38 249ndash303

70 HahnD NickelsenJ HackertA and KuckU (1998) A single nuclearlocus is involved in both chloroplast RNA trans-splicing and 30 endprocessing Plant J 15 575ndash581

71 MichelF UmesonoK and OzekiH (1989) Comparative andfunctional anatomy of group II catalytic intronsmdasha review Gene 825ndash30

72 ChanfreauG and JacquierA (1996) An RNA conformational changebetween the two chemical steps of group II self-splicing EMBO J 153466ndash3476

73 HauserCR GillhamNW and BoyntonJE (1998) Regulation ofchloroplast translation In RochaixJD Goldschmidt-ClermontM andMerchantS (eds) The Molecular Biology of Chloroplasts andMitochondria in Chlamydomonas Kluwer Academic PublishersDordrecht The Netherlands Vol 7 pp 197ndash217

74 BoudreauE NickelsenJ LemaireSD OssenbuhlF andRochaixJD (2000) The Nac2 gene of Chlamydomonas encodes achloroplast TPR-like protein involved in psbD mRNA stability EMBOJ 19 3366ndash3376

75 BarnesD CohenA BruickRK KantardjieffK FowlerS EfuetEand MayfieldSP (2004) Identification and characterization of a novelRNA binding protein that associates with the 50-untranslated region ofthe chloroplast psbA mRNA Biochemistry 43 8541ndash8550

76 FargoDC BoyntonJE and GillhamNW (2001) Chloroplastribosomal protein S7 of Chlamydomonas binds to chloroplast mRNAleader sequences and may be involved in translation initiation PlantCell 13 207ndash218

77 HauserCR GillhamNW and BoyntonJE (1996) Translationalregulation of chloroplast genesmdashproteins binding to the 50-untranslatedregions of chloroplast mRNAs in Chlamydomonas reinhardtii J BiolChem 271 1486ndash1497

78 NakamuraT SchusterG SugiuraM and SugitaM (2004)Chloroplast RNA-binding and pentatricopeptide repeat proteinsBiochem Soc T 32 571ndash574

79 JacquierA and MichelF (1990) Base-pairing interactions involvingthe 50 and 30-terminal nucleotides of group II self-splicing intronsJ Mol Biol 213 437ndash447

Nucleic Acids Research 2006 Vol 34 No 18 5351