FimY of Salmonella enterica serovar Typhimurium functions as a DNA-binding protein and binds the fimZ promoter

G

M

FD

KJa

b

Tc

Sd

a

ARRAA

KSTD

1

mis1fifif

t(p(

k

0h

ARTICLE IN PRESS Model

ICRES-25644; No. of Pages 8

Microbiological Research xxx (2014) xxx– xxx

Contents lists available at ScienceDirect

Microbiological Research

jo ur nal ho me p age: www.elsev ier .com/ locate /micres

imY of Salmonella enterica serovar Typhimurium functions as aNA-binding protein and binds the fimZ promoter

e-Chuan Wanga, Yuan-Hsun Hsua,b, Yi-Ning Huangc,iunn-Horng Linc,d, Kuang-Sheng Yehc,∗

Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, 250 Wu-Hsing Street, Taipei 11031, TaiwanDepartment of Microbiology and Immunology, School of Medicine, College of Medicine, Taipei Medical University, 250 Wu-Hsing Street, Taipei 11031,aiwanDepartment of Veterinary Medicine, School of Veterinary Medicine, College of Bioresources and Agriculture, National Taiwan University, 1 Roosevelt Road,ection 4, Taipei 10617, TaiwanDivision of Animal Medicine, Animal Technology Institute Taiwan, Chunan, Miaoli 35053, Taiwan

r t i c l e i n f o

rticle history:eceived 7 August 2013eceived in revised form 2 December 2013ccepted 8 December 2013vailable online xxx

eywords:almonellaype 1 fimbriaeNA binding protein

a b s t r a c t

Salmonella enterica serovar Typhimurium produces type 1 fimbriae with binding specificity to mannoseresidues. Elements involved in fimbrial structural biosynthesis, transport, and regulation are encoded bythe fim gene cluster. FimZ, FimY, FimW, STM0551, and an arginine transfer RNA (fimU) were previouslydemonstrated to regulate fimbrial expression. The amino acid sequences of the C-terminal portion ofFimY revealed similarity with those of LuxR-like proteins. Electrophoretic mobility shift assays indicatedthat FimY possessed DNA-binding capacity and bound a 605-bp DNA fragment spanning the intergenicregion between fimY and fimZ, while a FimY protein harboring a double mutation in the C-terminal helix-turn-helix region containing a glycine (G) to aspartate (D) substitution at residue 189 and isoleucine (I)to lysine (K) substitution at residue 195 lost its ability to bind this DNA fragment. A lux box sequence (5′-

′
TCTGTTATTACATAACAAATACT-3 ) within the fimZ promoter was required for binding. None of the DNAfragments derived from the promoters for fimA, fimY, or fimW was shifted by FimY. Pull-down assaysshowed that there were physical protein/protein interactions between FimY and FimZ. We propose thatin the regulatory circuit of type 1 fimbriae, FimY functions as a DNA-binding protein to activate fimZ, anda FimY-FimZ protein complex may form to regulate other fim genes. Confirming these proposals requires further study.
. Introduction

Fimbriae are proteinaceous hair-like structures on the outerembranes of many bacteria, the primary function of which was

mplied to be adherence (Duguid et al., 1955). Whole-genomeequencing of Salmonella enterica serovar Typhimurium identified3 fimbrial gene clusters that may have the potential to encodembrial proteins (McClelland et al., 2001). Among these, type 1mbriae encoded by the fim gene cluster were the most commonly

ound type in S. Typhimurium (Duguid and Gillies, 1957).Interactions among and cooperation of fim genes contribute

o phenotypic expressions of type 1 fimbriae by S. Typhimurium

Please cite this article in press as: Wang K-C, et al. FimY of Salmonella entbinds the fimZ promoter. Microbiol Res (2014), http://dx.doi.org/10.1016/j

Clegg et al., 1985)(Fig. 1). The fimbrial shaft is primarily com-osed of the major fimbrial subunit, FimA, while other subunitsi.e., FimI, FimF, and FimH) are also incorporated in assembling

∗ Corresponding author. Tel.: +886 2 33441689.E-mail addresses: [email protected], [email protected],

[email protected] (K.-S. Yeh).

944-5013/$ – see front matter © 2013 Elsevier GmbH. All rights reserved.ttp://dx.doi.org/10.1016/j.micres.2013.12.006

© 2013 Elsevier GmbH. All rights reserved.

the fimbrial structure. FimC and FimD are chaperone and molec-ular usher proteins, respectively; they assist in the transportationand anchorage of fimbrial subunits to the outer membrane. FimZ,FimY, STM0551, FimW, and an arginine transfer RNA (fimU) reg-ulate fimbrial expression through a complicated circuit involvingboth transcriptional and translational levels. Both fimZ and fimYmutants exhibited nonfimbriate phenotypes, suggesting that bothgene products are positive regulators of type 1 fimbriae (Tinkerand Clegg, 2000; Yeh et al., 1995). FimZ was shown to bind thefimA promoter to activate fimbrial expression (Yeh et al., 2002).A fimW mutant, on the contrary, produced a hyperfimbriate phe-notype, indicating its repressive role in regulation (Tinker et al.,2001). FimW inhibition of type 1 fimbrial production may be medi-ated by consumption of the positive regulator, FimZ or by physicalinteractions, or may be mediated by repression of the promoteractivity of fimY (Saini et al., 2009; Tinker et al., 2001). A fimU

erica serovar Typhimurium functions as a DNA-binding protein and.micres.2013.12.006

mutant was nonfimbriate due to the inefficient translation of FimY(Tinker and Clegg, 2001). Recently, STM0551 was demonstratedto function as a phosphodiesterase to degrade cyclic-di-GMP, anda stm0551 mutant resulted in upregulation of type 1 fimbriae

dx.doi.org/10.1016/j.micres.2013.12.006


http://www.sciencedirect.com/science/journal/09445013

http://www.elsevier.com/locate/micres

mailto:[email protected]




ARTICLE IN PRESSG Model

MICRES-25644; No. of Pages 8

2 K.-C. Wang et al. / Microbiological Research xxx (2014) xxx– xxx

F peptido k boxef

(lwC

ddobT2cialagfiftbr

2

2

lrEs

TB

A

ig. 1. The fim gene cluster of S. Typhimurium. The predicted sizes of the Fim polyf those genes involving structure and biosynthesis are shown as a small filled blacunctions of the genes are indicated in the text.

Wang et al., 2012). A global regulator, leucine-responsive regu-atory protein (Lrp), and other genes outside the fim gene cluster

ere also implicated in regulating type 1 fimbriae (Baek et al., 2009;huang et al., 2008; McFarland et al., 2008).

Previously, FimY was reported to possess no recognizableomains and exhibited little similarity to any proteins in availableatabases (Clegg et al., 2011). However, the amino acid sequencef the C-terminal part of FimY does share some homology to mem-ers of LuxR family regulators, such as FimZ, SdiA, and CsgD of S.yphimurium, and LuxR of Vibrio fischeri (Patankar and Gonzalez,009). FimY is classified as an orphan LuxR homolog since no asso-iated acylated homoserine lactone synthase has been identifiedn close proximity (Fuqua, 2006). In the present study, fimA, fimZ,nd fimY expressions of fimZ and fimY mutant strains were ana-yzed; in addition, the DNA-binding ability of FimY was tested byn electrophoretic mobility shift assay (EMSA). Our results sug-ested that FimY may function as a DNA-binding protein to activatemZ. A lux box sequence within the fimZ promoter was requiredor such binding. In addition, a pull-down assay also suggestedhat FimY and FimZ may have protein/protein interactions. To theest of our knowledge, these findings have not previously beeneported.

. Materials and methods

.1. Bacterial strains, plasmids, and culture conditions

All of the bacterial strains and plasmids used in this study are


isted in Table 1. Salmonella Typhimurium and Escherichia coli wereoutinely cultured at 37 ◦C in Luria–Bertani (LB) broth or on LB agar.. coli strains for cloning, plasmid purification, and protein expres-ion were maintained with 100 �g ampicillin ml−1. For expression

able 1acterial strains and plasmids used in the study.

Strain or plasmid Description

Bacterial strainsSalmonella Typhimurium

LB5010 Wild type; fimbriated with a complete fim geneLB5010 fimY �fimY::EZ-Tn5 < KAN-2 > Tnp

LB5010 fimZ �fimZ::EZ-Tn5 < KAN-2 > Tnp

Escherichia coliDH5� Molecular cloning and blue/white screening

One Shot® TOP10 Competent cells for the TOPO® cloning proceduBL21 StarTM (DE3) Competent cells for fusion protein expression

PlasmidspET101/D-TOPO® Cloning vector, AmpR, includes a C-terminal 6×pET101-FimY A fimY coding sequence cloned into the pET101pET101-FimY-2aa A fimY coding sequence with G189D and I195 KpET17-b Cloning vector, AmpR, includes an N-terminal T

pET17-b-FimZ A fimZ coding sequence cloned into the pET17-

mpR, ampicillin resistant.

es are given in kilodaltons (kDa) with Arabic numbers. The signal peptide regionss. The arrows indicate the transcriptional directions. The established or postulated

of type 1 fimbriae, all S. Typhimurium strains were cultured in LBbroth without shaking at 37 ◦C for 48 h.

2.2. RNA isolation and reverse-transcription polymerase chainreaction (RT-PCR) analysis

Bacteria were harvested by centrifugation at 9200 × g andwashed with 1× phosphate-buffered saline (PBS). Total RNA wasisolated using an RNeasy Mini Kit (Qiagen, Hilden, Germany)according to the protocol provided by the manufacturer. PurifiedRNA was treated with RNase-free DNase I (1 unit �g−1 RNA) (Fer-mentas, St. Leon-Rot, Germany) to remove contaminated genomicDNA. An RT-PCR was subsequently performed with a Fast-RunHotStart RT-PCR (AMV) kit (Protech, Taipei, Taiwan). For first-strand complementary (c)DNA synthesis, total RNA (0.1 mg) wasdenatured at 58 ◦C for 5 min, and cDNA was synthesized at 42 ◦Cfor 30 min. Then, RNA/cDNA was denatured at 94 ◦C for 2 min.The RT-PCR consisted of 35 cycles of denaturation at 94 ◦C for30 s, annealing at 50 ◦C for 30 s, and extension at 72 ◦C for 1 min.All samples generated by the RT-PCR were separated on 1.3%agarose gels. Comparisons of messenger (m)RNA levels were calcu-lated by ImageQuant software (Amersham/GE Healthcare, Uppsala,Sweden).

2.3. Construction of FimY and FimZ fusion proteins

Genomic DNA of S. Typhimurium LB5010 was used as a tem-plate to amplify fimY and fimZ. The fimY gene was amplified with


FimY-His-F and FimY-His-R primers. The PCR product was purifiedand cloned into the pET101/D-TOPO vector (Invitrogen, Carlsbad,CA, USA). This recombinant plasmid was transformed into an E. coliBL21 StarTM (DE3) strain (Invitrogen) and cultured in LB broth

Source

cluster (Bullas and Ryu, 1983)(Chuang et al., 2008)(Chuang et al., 2008)

RBC Biosciencere Invitrogen

Invitrogen

His epitope Invitrogen vector; AmpR Present study

cloned into the pET101 vector; AmpR Present study7 epitope Novagenb vector; AmpR Present study




K.-C. Wang et al. / Microbiological Research xxx (2014) xxx– xxx 3

Table 2Oligodeoxynucleotides used in the study.

Primer Sequence 5′ → 3′ Description

fimA-F ACTATTGCGAGTCTGATGTTTG RT-PCRfimA-R CGTATTTCATGATAAAGGTGGC RT-PCRfimY-F GAGTTACTGAACCAACAGCT RT-PCRfimY-R GCCGGTAAACTACACGATGA RT-PCRfimZ-F ACTTATCCTGTTGACCTT RT-PCRfimZ-R ATTCGTGTGATTTGGCGT RT-PCRFimY-His-F AAAAATGTCGTGGAAAGTAACGT TOPO cloningFimY-His-R CACCATGCGCAGCGTACCACGCA TOPO cloningFimY-2aa-His-F AGACGCCTCTTTAAAAGAAAGAGCACGTCAGCA TOPO cloningFimY-2aa-His-R CGTGCTCTTTCTTTTAAAGAGGCGTCTTG TOPO cloningT7-FimZ-BamHI CGTCTCGCGGATCCCATGAAACCTGCATCTGTTATCATTATGG The underlined portion denotes a BamHI restriction site.T7-FimZ-EcoRI AGAACAGAGAATTCCAATAATTCGTGTGATTTGGCGTAATCG The underlined portion denotes an EcoRI restriction site.fimZp-F ACGAAGTAACGTTTTGGTGA Promoter DNA synthesisfimZp-F2 TCCAGCGTCAACATCTTTCT Promoter DNA synthesisfimZp-F3 AAATCAGACCAGGTATAATCCA Promoter DNA synthesisfimZp-R TTGTTGTATGCCTCAGACTGTT Promoter DNA synthesisfimZp-R1 AACAGAAAGATGTTGACGCTGG Promoter DNA synthesisfimZp-R2 AAAACAATATGGATTATACCTGGTC Promoter DNA synthesisfimZp-R4 GTGGTGAGCTATTTTCTTT Promoter DNA synthesisfimZp-del-lux2-F GTCTCCAGCGTCAACATCTTAAACAATTAGTCTTTCATAT Promoter DNA synthesisfimZp-del-lux2-R ATATGAAAGACTAATTGTTTAAGATGTTGACGCTGGAGAC Promoter DNA synthesisfimAp-F TTAATCTTTGCGAATATGAC Promoter DNA synthesis

aotTsmdSgeam91fidaawdg5ci((mid

EiRp(aaeat

fimAp-R TATGGTTACCGTAATCCCTC

fimWp-F TAAAAAAATATCCTACACGGCAG

fimWp-R CTAAGGGCGCCTTGTGAA

t 37 ◦C with 100 �g ampicillin ml−1 until an optical density (OD)f 0.5 was reached, then 0.5 mM IPTG was added for an addi-ional 4 h to induce the His6-tagged form of FimY expression.he FimY protein was affinity-purified by a ProBond purificationystem (Invitrogen) according to the protocol provided by theanufacturer. A mutant allele of fimY was constructed by site-

irected mutagenesis using an overlapping-extension PCR of an. Typhimurium LB5010 strain genomic DNA template and muta-enic oligonucleotides FimY-2aa-His-F and FimY-2aa-His-R (Hot al., 1989). Briefly, FimY-His-F and FimY-2aa-His-R were used tomplify the first DNA fragment using Pfu DNA polymerase (Fer-entas). The PCR conditions consisted of initial denaturation at

4 ◦C for 3 min, followed by 35 cycles at 94 ◦C for 1 min, 50 ◦C for min, and 72 ◦C for 1 min. The second DNA fragment was ampli-ed using FimY-2aa-His-F and FimY-His-R by the same procedureescribed above. These two DNA fragments were purified with

Montage Gel Extraction Kit (Millipore, Billerica, MA, USA). Lig-tion of these two DNA fragments with two overlapping endsas achieved with FimY-His-F and FimY-His-R primers as follows:enaturation at 94 ◦C for 3 min, ligation at 50 ◦C for 1 min, and elon-ation at 72 ◦C for 1 min, followed by 35 cycles at 94 ◦C for 1 min,0 ◦C for 1 min, and 72 ◦C for 1 min. Amplified fragments wereloned into the pET101/D-TOPO vector and sequenced to determinef the codons encoding glycine (G) at amino acid 189 and isoleucineI) at amino acid 195 had respectively been replaced by aspartateD) and lysine (K) codons. The resulting recombinant double-

utant plasmid was designated pET101-FimY-2aa. Further proteinnduction and purification were performed using the same proce-ure (described above) for generating the FimY-His fusion protein.

The fimZ gene was amplified with T7-FimZ-BamHI and T7-FimZ-coRI primers, purified, digested with EcoRI and BamHI, and clonednto the corresponding sites of the pET17-b vector (Novogen, Northyde, NSW, Australia) to obtain pET17-b-FimZ. This recombinantlasmid was transformed into the E. coli BL21 StarTM (DE3) strainInvitrogen) and cultured in LB broth at 37 ◦C with 100 �g ml−1

mpicillin until an OD of 0.8–1 was reached, then 0.5 mM IPTG was


dded for an additional 4 h to induce the His6-tagged form of FimZxpression. The FimZ protein was affinity-purified by a T7·Tag®

ffinity purification kit (Merck, Darmstadt, Germany) according tohe protocol provided by the manufacturer.

Promoter DNA synthesisPromoter DNA synthesisPromoter DNA synthesis

2.4. Electrophoretic mobility shift assay (EMSA)

The promoter sequences of fimA, fimZ, and its derivatives, fimYand fimW, were generated by a PCR with specific primers listedin Table 2. We used an overlap extension PCR method to con-struct a DNA fragment with a deleted lux box-like sequence (Hoet al., 1989). Briefly, a fimZp-del-lux2-F and fimZp-R primer set wasused to amplify the first DNA fragment using Pfu DNA polymerase(Fermentas). The PCR conditions consisted of initial denaturationat 94 ◦C for 3 min, followed by 35 cycles at 95 ◦C for 45 s, 48 ◦Cfor 45 s, and 72 ◦C for 45 s. The second DNA fragment was ampli-fied using fimZp-del-lux2-R and fimZp-F by the same proceduredescribed above. Ligation of these two DNA fragments with twooverlapping ends was achieved with fimZp-F and fimZp-R primersas follows: denaturation at 94 ◦C for 3 min, ligation at 48 ◦C for1 min, and elongation at 72 ◦C for 1 min, followed by 35 cyclesat 94 ◦C for 45 s, 48 ◦C for 45 s, and 72 ◦C for 45 s. Promoter DNAfragments were purified using a DNA Clean/Extraction kit (Gen-eMark, Taipei, Taiwan) and labeled with digoxigenin-11-ddUTP(DIG-ddUTP) at the 3′ termini using a 2nd generation DIG gelshift kit (Roche, Mannheim, Germany). The binding reaction mix-ture contained 100 ng of DIG-labeled DNA fragments, 1× bindingbuffer (10 mM Tris at pH 7.4, 1 mM EDTA, 100 mg ml−1 bovineserum albumin (BSA), 10 �g ml−1 xylene cyanol FF, and 5% glyc-erol), and 0.6 �g of protein (or 2-folded dilution of 0.6 �g) in a finalvolume of 10 �l in double-distilled (dd)H2O. The reaction was incu-bated at 23 ◦C for 15 min and set in ice, and then 2 �l of loadingdye was added. The sample mixture was loaded onto a 5% nativepolyacrylamide gel prepared in 0.5× TBS buffer (40 mM Tris atpH 8.3 and 45 mM boric acid) and electrophoresed at 100 V for1.5 h. Next, the gel was transferred to a nylon membrane by anelectroblotting method and fixed by UV cross-linking. The mem-brane was incubated with washing buffer (100 mM maleic acid,150 mM NaCl at pH 7.5, and 0.3% Tween-20) for 5 min, block-ing solution for 30 min, and an anti-DIG AP-conjugated antibodyfor another 30 min, then washed twice with washing buffer. The


color reaction was detected with NBT/BCIP substrates. To testwhether binding, if any, was specific, the promoter containingDNA fragments without labeling was used as the cold competi-tor.





F e useL strain( of diff

2

(dNNt4(iTapapaTs

3

3

mT

ig. 2. RT-PCR analysis of fimA, fimY, and fimZ transcription. (A). RT-PCR assays werB5010, LB5010 fimY, and LB5010 fimZ strains. The intensities of the bands for eachB), fimY (C), and fimZ (D) of LB5010 (WT). Asterisks indicate statistical significance

.5. Pull-down assay

FimY was used as a bait protein and mixed with ProBond resin6× His tag) at 4 ◦C for 1 h. After the resin had been washed withenaturing binding buffer (8 M urea, 20 mM Na3PO4, and 500 mMaCl; pH 7.8) and the native wash buffer (50 mM NaH2PO4, 0.5 MaCl, and 20 mM imidazole; pH 8.0) to remove any unbound pro-

ein, the prey protein, FimZ, was added to the resin and incubated at◦C overnight. Proteins were eluted by adding native elution buffer

50 mM NaH2PO4, 0.5 M NaCl, and 250 mM imidazole; pH 8.0). Sim-larly, when FimZ was used as the bait protein, it was mixed with7·Tag antibody agarose and incubated at 4 ◦C for 1 h. The T7·Tagntibody agarose immobilized with FimZ was incubated with therey protein, FimY, at 4 ◦C overnight, and the protein was eluted bydding 1× T7·Tag elute buffer (0.1 M citric acid at pH 2.2). The elutedrotein fraction was then analyzed by sodium dodecylsulfate poly-crylamide gel electrophoresis (SDS-PAGE) and Western blotting.he colorimetric blotting was performed by adding BCIP/NBT sub-trate (Amresco, Salon, OH, USA).

. Results

.1. RT-PCR analysis


Salmonella Typhimurium LB5010 fimY and LB5010 fimZ are twoutant strains generated in a previous study (Chuang et al., 2008).

he former has a transposon inserted in fimY and the latter in fimZ.

d to determine fimA, fimY, fimZ, and 16S rRNA transcription by the S. Typhimurium were determined by densitometry and are expressed relative to the values of fimAerences between datasets: *p < 0.05.

Fig. 2A demonstrates the agarose gel electrophoresis of the RT-PCR.Both mutants demonstrated barely detectable fimA mRNA expres-sion (Fig. 2B). LB5010 fimZ had lower fimZ mRNA levels; however, itstill retained about 80% of the fimY mRNA level compared to that ofthe LB5010 parental strain (Fig. 2 C). A transposon inserted in fimYabolished both fimY and fimZ mRNA expressions (Fig. 2D).

3.2. EMSA

A 561-bp fimA promoter-containing region was amplified byfimAp-F and fimAp-R. Primers fimYp-F and fimYp-R were used toamplify the fimY promoter-containing region, resulting in a 497-bpDNA fragment. A 266-bp fimW promoter-containing DNA fragmentwas generated by fimWp-F and fimWp-R. The length of the DNAproduct possessing the fimZ promoter was 605 bp and was yieldedby fimZp-F and fimZp-R. These DNA fragments were respectivelydesignated fimA p, fimY p, fimW p, and fimZ p in the present study.EMSAs were performed to determine if the FimY protein was ableto bind to the target DNA fragments and cause a mobility shift.Serial, 2-fold dilutions of the FimY protein starting from 0.6 �gwere added in the EMSA. The undiluted (0.6 �g) and 2-fold diluted(0.3 �g) FimY exhibited the ability to cause fimZ p to shift, whilefurther dilutions of FimY did not cause a shift (Fig. 3A). The DNA


fragments possessing the fimA, fimY, and fimW promoter were notshifted by 0.6 �g of FimY (Fig. 3B). Since there are two lux-likebox sequences within the fimZ p region (Fig. 4), we wanted to testif these sequences were required for FimY binding. Five different





Fig. 3. Electrophoretic mobility shift assay of the FimY-His fusion protein with DNA fragments possessing different promoters. Panel A, a series of two-folded dilution ofF moterw or 0.6

DPts1tf5bdo

t

imY-His fusion protein starting with 0.6 �g (1×) were incubated with the fimZ proere incubated with no protein (lane 1), 0.6 �g of FimY-His fusion protein (lane 2),

NA fragments derived from fimZ p DNA were generated by theCR. The fimZ p2 and fimZ p4 DNA fragments, both of which con-ained the lux box 2 (5′-TCTGTTATTACATAACAAATACT-3′), werehifted by the FimY protein. Neither fimZ p1 possessing lux box

(5′-AATGTAAATATTTCACATAAA-3′) nor fimZ p3 or fimZ p5 (nei-her of which contained a lux box) were shifted by FimY (Fig. 5). Tourther investigate the role the lux box2 plays in FimY binding, a82-bp DNA fragment, designated fimZ p� lux2 with only the luxox 2 sequence deleted, was constructed, and we found that FimY


id not cause this fragment to shift (Fig. 5). Fig. 6 summarizes resultsf the EMSAs of the FimY and fimZ derivative DNA fragments.

The C-terminal portion of FimY contains a helix-turn-helix motifhat shares similarity to that of the well characterized DNA-binding

-containg DNA. Panel B, fimA, fimY, and fimW promoter-containing DNA fragments �g of FimY-His fusion protein and an unlabeled DNA fragment (lane 3).

protein such as �-repressor, trp repressor, or 434 Cro protein(Brennan and Matthews, 1989). To test if changing this regionabolished the binding activity, we constructed FimY-His-2aa thatcontained a glycine (G) to aspartate (D) substitution at residue189 and an isoleucine (I) to lysine (K) substitution at residue 195.Fig. 7 demonstrates that FimY-His-2aa did not cause fimZ p toshift.

3.3. Pull-down assay


In order to test if FimZ interacts with FimY, a pull-down assaywith FimY-His and FimZ-T7 was performed. After the T7·Tagantibody agarose was immobilized with FimZ-T7 and incubated


ARTICLE ING Model


6 K.-C. Wang et al. / Microbiological R

Stop(fimY ) -35 0 TAACGAAGTAACGTTTTGGTGAATAAGATTTTTTAAGAAATTACTTATTT

Puta�ve lux box 1 TCACTTCCATTAAATGT AAATATTTC ACATAAAATTAATATTTACAACTC

-250 GTCCTGGTAAGCGATAAAAATTTTTTTCTTACATCGTCCCCTGTTCCTCC

Puta�ve lux box 2 TCGTTTTTCGAGGTCTCCAGCGTCAACATCTTTCTGTTATTACATAACAA -150 ATACTAAACAATTAGTCTTTCATATTATTTTTATACTTAAAAACAACAGT

TTAAACTACTCAATGTCAACTCTAAAGAAAA TAGC TCC ACC ACGAA CAAA -50

TTTTCTATAACGTCACAGAATATCCT ACCCGCTATCCATAATTCGCATTT Ti

CGCAAGACTTTTCCGAGAAATCAGACCAGGTATAATCCATATTGTTTTTA

CTTAATCAGAAATTTCTTTTTTGTTGGTGCGAAATGGATATTCCACATAT

TTATTACTCTTACGCAAAATAGCCAGTCAACAGGGAGGTCTCATTCTTTA

TTATTGC TCCACTGTGTGGTGTGTAGATCTGGCTATTTCACAATCTTTAA

GGTGTCTGACGCTTATTATAAAACGAA GGACGCATAACAGTCTGAGGCAT Met(fimZ )

ACAACAATG

Fig. 4. Sequence analysis of the region between fimZ and fimY. The transcriptioninitiation site of fimZ is 227 bp upstream of the FimZ initiation codon and is in boldfc

wtrptwt(

4

st

Fw

ace and denoted by Ti. Two lux box sequences are indicated by gray boxes. The stopodon (TAA) of fimY and the start codon (ATG) of fimZ are underlined.

ith FimY-His overnight, the protein eluted was recognized byhe anti-His antibody. T7·Tag antibody beads incubated sepa-ately with FimY-His or FimZ-T7 did not yield a signal whenrobed with the anti-His antibody. Similarily, the protein mix-ure eluted from the ProBond resin exhibited signal when probedith anti-T7 antibody. The anti-T7 antibody did not recognize

he elutant when adding FimZ-T7 or FimY-His alone to the resinFig. 8).

. Discussion


The S. Typhimurium LB5010 fimZ and fimY mutants both demon-trated barely detectable levels of fimA, which is in line withheir roles as positive regulators of type 1 fimbrial production

ig. 5. Electrophoretic mobility shift assay with the FimY-His protein and different lengtith no protein (lane 1), 0.6 �g of FimY (lane 2), or 0.6 �g of FimY and an unlabeled DNA

PRESSesearch xxx (2014) xxx– xxx

(Tinker and Clegg, 2000; Yeh et al., 1995). A previous report indi-cated that FimY and FimZ independently activated each other’sexpression (Saini et al., 2009). Our results revealed that the LB5010fimZ strain still retained 80% of fimY expression, while we couldbarely detect the mRNA of fimZ in the LB5010 fimY strain. Zeiner(2012) demonstrated that fimZ, cloned in a multicopy plasmid, wasable to overcome the lack of fimY, whereas fimY was unable to over-come the lack of fimZ. This suggests that fimY may act upstreamof fimZ. It is possible that the primary function of fimY is to acti-vate fimZ, and fimZ, in turn, activates fimA expression to producetype 1 fimbriae. This activating activity of FimY may be associ-ated with its binding activity to the fimZ promoter as discussedbelow.

Protein sequences of LuxR-type proteins share only 18–25% end-to-end identity, yet their functional domains exhibit much-highersequence conservation (Nasser and Reverchon, 2007; Whiteheadet al., 2001). FimY was previously reported to activate fimZ and itsown promoters, but no DNA-binding activity was demonstrated(Saini et al., 2009). Tinker and Clegg (Tinker and Clegg, 2000) usedpurified FimY to test its binding activity in vitro, but no bind-ing activity was detected when FimY was mixed with either fimAor fimY promoter DNA; nonetheless, fimZ promoter DNA was notincluded in the test. Since FimY possesses some of the conservedresidues associated with DNA binding as seen with other LuxR-typeproteins (Vannini et al., 2002), it was interesting to further inves-tigate the binding ability of FimY with more fim promoter DNAfragments in vitro. No binding activity was observed for fimA orfimY, which correlated with a previous report (Tinker and Clegg,2000). Saini et al. (2009) demonstrated that FimY activated expres-sion by the fimW promoter according to a promoter activity assay.Our result showed no binding activity between FimY and fimW pro-moter DNA. It is possible that some accessory proteins other thanFimY may also participate in activating in vivo fimW gene expres-sion. The EMSAs indicated that the FimY protein bound a 605-bpDNA fragment that spanned the entire intergenic region of fimY andfimZ. This DNA fragment contains the fimZ promoter region (Yehet al., 2002). This region contains two elements with sequence sim-ilarities to the lux box reported previously (Antunes et al., 2008).Although some LuxR-type proteins do not even bind to the DNAregion with an identifiable lux box (Antunes et al., 2007), FimY didbind to the DNA fragments containing one lux box sequence (lux box2) located −197 to −175 relative to the transcription start site offimZ. In addition, a DNA fragment derived from the 605-bp fimZ pro-moter DNA with only the lux box 2 deleted was not shifted by FimY,which underscores the importance of this site for FimY binding.There are 12 identical nucleotides at the same positions between


lux box 1 and lux box 2. Both lux boxes possess the most conservednucleotides of TGT (Antunes et al., 2008), but we do not know thereason why a DNA fragment containing lux box 1 was not mobi-lized by FimY. The intergenic region of fimZ and fimY contains two

hs of DNA fragments derived from 605-bp fimZ p DNA. A DNA fragment incubated fragment (lane 3). The arrow indicates the protein/DNA complex.





Fig. 6. Summary of the electrophoretic mobility shift assays with FimY and fimZ promoter-containing DNA fragments. The ability of FimY to bind to the DNA is indicated aspositive, while negative indicates that FimY did not cause a mobility shift of these DNA fragments. The positions of two lux boxes and the Lrp binding region are indicated, andthe Arabic numerals above these three regions designate their positions relative to the transcriptional initiation site of fimZ. TGT, which are underlined, are highly conservedd

leiea2

mpihsmipDp3ss

pFuFsapmaLl

plex to regulate bgl (aryl-�,d-glucoside) operon expression in E. coli(Venkatesh et al., 2010). Nonetheless, the actual in vivo roles of FimZand FimY warrant further investigation.

eoxynucleotides within the lux box.

ux box-like sequences and also the Lrp-binding region (McFarlandt al., 2008). This possibly indicates that the expression of fimZs important and requires a delicate regulatory mechanism. Forxample, besides type 1 fimbriae, fimZ is also involved in motilitynd invasion functions (Baxter and Jones, 2005; Clegg and Hughes,002).

Changing two amino acids at the C-terminal helix-turn-helixotif of the FimY protein abolished its binding activity to fimZ

romoter DNA. The two residues we changed, glycine 189 andsoleucine 195, are respectively located at the turn and the secondelix region of FimY. These two residues are the most strongly con-erved (Brennan and Matthews, 1989) among the helix-turn-helixotifs of most DNA binding protein, and our result confirmed the

mportance of these two conserved residues to DNA binding. Theresent study provided evidence that FimY of S. Typhimurium is aNA-binding protein, and its binding region resides within the fimZromoter. A lux box sequence (5′-TCTGTTATTACATAACAAATACT-′) is required for FimY binding. The DNase I footprint andite-directed mutagenesis of the binding site may warrant furthertudy.

Tinker et al. (2001) reported that FimZ and FimW had physicalrotein/protein interactions, suggesting that FimW may consumeimZ, the activator of fimA, to achieve its repressive role in reg-lating type 1 fimbriae in S. Typhimurium. However, FimW andimY exhibited no physical interactions in their study. The presenttudy indicated that FimY and FimZ interact. Since FimY and FimZre both required to activate fimA (Swenson and Clegg, 1992), it isossible that these two proteins form a complex to bind the pro-


oter of fimA. On the other hand, the FimY-FimZ complex mightlso bind the fimZ and fimY promoters to regulate these genes. SomeuxR-like proteins were reported to form heterodimers and regu-ate downstream gene activation. For example, both BglJ and RcsB

proteins in E. coli belong to the LuxR-type transcription regulator;and they could interact with each other to form a heterodimer com-


Fig. 7. Electrophoretic mobility shift assay with FimY-His and FimY-His-2aa. A 605-bp fimZ p DNA with no protein (lane 1), 0.6 �g of protein (lane 2), or 0.6 �g of proteinand an unlabeled fimZ p DNA fragment (lane 3). The arrow indicates the protein/DNAcomplex.





F pull-dT nalysi

A

f

A

i2

R

A

A

B

B

B

B

C

C

C

C

D

D

region and may regulate its own expression with FimY. Microbiol Immunol

ig. 8. Protein-protein interactions of FimZ and FimY were determined by in vitro

7·Tag antibody agarose or ProBond resin column and detected by a Western blot a

cknowledgment

This work was supported by a grant (NSC101-2313-B-002-018)rom the National Science Council, Taiwan to K.S. Yeh.

ppendix A. Supplementary data

Supplementary material related to this article can be found,n the online version, at http://dx.doi.org/10.1016/j.micres.013.12.006.

eferences

ntunes LCM, Ferreira RBR, Lostroh CP, Greenberg EP. A mutational analysis definesVibrio fischeri LuxR binding sites. J Bacteriol 2008;190:4392–7.

ntunes LCM, Schaefer AL, Ferreira RBR, Qin N, Stevens AM, Ruby EG, GreenbergEP. A transcriptome analysis of the Vibrio fischeri LuxR-LuxI regulon. J Bacteriol2007;189:8387–91.

aek CH, Wang S, Roland KL, Curtiss R III. Leucine-responsive regulatory protein(Lrp) acts as a virulence repressor in Salmonella enterica serovar Typhimurium.J Bacteriol 2009;191:1278–92.

axter MA, Jones BD. The fimYZ genes regulate Salmonella enterica serovarTyphimurium invasion in addition to type 1 fimbrial expression and bacterialmotility. Infect Immun 2005;73:1377–85.

rennan R, Matthews BW. The helix-turn-helix DNA binding motif. J Biol Chem1989;264:1903–6.

ullas LR, Ryu JI. Salmonella typhimurium LT2 strains which are r− m+ for all threechromosomally located systems of DNA restriction and modification. J Bacteriol1983;156:471–4.

huang YC, Wang KC, Chen YT, Yang CH, Men SC, Fan CC, Chang LH, Yeh KS. Identifi-cation of the genetic determinants of Salmonella enterica serotype Typhimuriumthat may regulate the expression of the type 1 fimbriae in response to solid agarand static broth culture conditions. BMC Microbiol 2008;8:126.

legg S, Hughes KT. FimZ is a molecular link between sticking and swimming inSalmonella enterica serovar Typhimurium. J Bacteriol 2002;184:1209–13.

legg S, Purcell BK, Pruckler J. Construction and comparison of recombinant plasmidsencoding type 1 fimbriae of members of the family Enterobacteriaceae. InfectImmun 1985;48:275–9.

legg S, Wilson J, Johnson J. More than one way to control hair growth: regulatorymechanisms in Enterobacteria that affect fimbriae assembled by the chaper-one/usher pathway. J Bacteriol 2011;193:2081–8.


uguid JP, Gillies RR. Fimbriae and adhesive properties in dysentery bacilli. J PatholBacteriol 1957;74:397–411.

uguid JP, Smith IW, Dempster G, Edmunds PN. Non-flagellar filamentousappendages (fimbriae) and haemagglutinating activity in Bacterium coli. J PatholBacteriol 1955;70:335–48.

own assays. The FimY-His/FimZ-T7 mixture was collected and purified by either as using anti-His or anti-T7 antibody.

Fuqua C. The QscR quorum-sensing regulon of Pseudomonas aeruginosa: an orphanclaims its identity. J Bacteriol 2006;188:3169–71.

Ho SN, Hunt HD, Horton RM, Pullen JK, Pease LR. Site-directed mutagenesis byoverlap extension using the polymerase chain reaction. Gene 1989;77:51–9.

McClelland M, Sanderson KE, Spieth J, Clifton SW, Latreille P, Courtney L, PorwollikS, Ali J, Dante M, et al. Complete genome sequence of Salmonella enterica serovarTyphimurium LT2. Nature (Lond) 2001;413:852–6.

McFarland KA, Lucchin S, Hinton JCD, Dorman CJ. The leucine-responsive reg-ulatory protein, Lrp, activates transcription of the fim operon in Salmonellaenterica serovar Typhimurium via the fimZ regulatory gene. J Bacteriol 2008;190:602–12.

Nasser W, Reverchon S. New insights into the regulatory mechanisms of theLuxR family of quorum sensing regulators. Anal Bioanal Chem 2007;387:381–90.

Patankar AV, Gonzalez JE. Orphan LuxR regulator of quorum sensing. FEMS MicrobRev 2009;33:739–56.

Saini S, Pearl JA, Rao CV. Role of FimW, FimY, and FimZ in regulating the expres-sion of type 1 fimbriae in Salmonella enterica serovar Typhimurium. J Bacteriol2009;191:3003–10.

Swenson DL, Clegg S. Identification of ancillary fim genes affecting fimA expressionin Salmonella typhimurium. J Bacteriol 1992;174:7697–704.

Tinker JK, Clegg S. Characterization of FimY as a coactivator of type 1 fim-brial expression in Salmonella enterica serovar Typhimurium. Infect Immun2000;68:3305–13.

Tinker JK, Clegg S. Control of FimY translation and type 1 fimbrial production by thearginine tRNA encoded by fimU in Salmonella enterica serovar Typhimurium. MolMicrobiol 2001;40:757–68.

Tinker JK, Hancox LS, Clegg S. FimW is a negative regulator affecting type 1fimbrial expression in Salmonella enterica serovar Typhimurium. J Bacteriol2001;183:435–42.

Vannini A, Volpari C, Gargioli C, Muraglia E, Cortese R, De Francesco R, NeddermannP, Marco SD. The crystal structure of the quorum sensing protein TraR bound toits autoinducer and target DNA. EMBO J 2002;21:4393–401.

Venkatesh GR, Kembou Koungni FC, Paukner A, Stratmann T, Blissenbach B, SchnetzK. BglJ-RcsB heterodimers relieve repression of the Escherichia coli bgl operonby H-NS. J Bacteriol 2010;192:6456–64.

Wang KC, Hsu YH, Huang YN, Yeh KS. A previously uncharacterized gene stm0551plays a repressive role in the regulation of type 1 fimbriae in Salmonella entericaserotype Typhimurium. BMC Microbiol 2012;12:111.

Whitehead NA, Barnard AM, Slater H, Simpson NJ, Salmond GP. Quorum-sensing inGram-negative bacteria. FEMS Microb Rev 2001;25:365–404.

Yeh KS, Hancox LS, Clegg S. Construction and characterization of a fimZ mutant ofSalmonella typhimurium. J Bacteriol 1995;177:6861–5.

Yeh KS, Tinker JK, Clegg S. FimZ binds the Salmonella typhimurium fimA promoter


2002;46:1–10.Zeiner SA. [Ph.D. Thesis] Type 1 fimbrial structure and regulation in Salmonella

enterica serovar Typhimurium [Ph.D. Thesis]. Iowa, USA: University of Iowa;2012.


http://dx.doi.org/10.1016/j.micres.2013.12.006

http://dx.doi.org/10.1016/j.micres.2013.12.006

http://refhub.elsevier.com/S0944-5013(13)00211-5/sbref0005










































































































































































































































































































































































































































































































































































































Documents

FimY of Salmonella enterica serovar Typhimurium functions as a DNA-binding protein and binds the fimZ promoter