Metabolic Regulation of the Gene Encoding Glutamine ... · The pBluescript SKI1 plasmids (Strategene) containing the cDNA clones were excised from the XZapII vector, and double-stranded

Plant Physiol. (1994) 106: 1347-1357

Metabolic Regulation of the Gene Encoding Glutamine- Dependent Asparagine Synthetase in Arabidopsis thaliana’

Hon-Ming lam, Sheila S.-Y. Peng’, and Gloria M. Coruzzi*

Department of Biology, New York University, 1 O09 Main Building, Washington Square East, New York, New York 10003

Here, we characterize a cDNA encoding a glutamine-dependent asparagine synthetase (ASN7) from Arabidopsis thaliana and assess the effects of metabolic regulation on ASNl mRNA levels. Sequence analysis shows that the predicted ASNl peptide contains a purF- type glutamine-binding domain. Southern blot experiments and cDNA clone analysis suggest that ASNl is the only gene encoding glutamine-dependent asparagine synthetase in A. thaliana. The ASNl gene i s expressed predominantly in shoot tissues, where light has a negative effect on its mRNA accumulation. This negative effect of light on ASNl mRNA levels was shown to be mediated, at least in part, via the photoreceptor phytochrome. We also investi- gated whether light-induced changes in nitrogen to carbon ratios might exert a metabolic regulation of the ASNl mRNA accumulation. These experiments demonstrated that the accumulation of ASNl mRNA in dark-grown plants i s strongly repressed by the presence of exogenous sucrose. Moreover, this sucrose repression of ASNl expression can be partially rescued by supplementation with exogenous amino acids such as asparagine, glutamine, and glutamate. These findings suggest that the expression of the ASNl gene is under the metabolic control of the nitrogen to carbon ratio in cells. This i s consistent with the fact that asparagine, synthesized by the ASNl gene product, i s a favored compound for nitrogen storage and nitrogen transport in dark-grown plants. We have put forth a working model suggesting that when nitrogen to carbon ratios are high, the gene product of ASNl functions to re-direct the flow of nitrogen into asparagine, which acts as a shunt for storage and/or long-distance transport of nitrogen.

Physiological studies have shown that certain amino acids play key roles in plant nitrogen metabolism. The amides Gln and Asn serve as major nitrogen transport compounds in most higher plants and especially in nitrogen-fixing temperate legumes such as pea and alfalfa (Lea and Miflin, 1980; Urquhart and Joy, 1981). Nitrogen assimilated into Gln may be converted into Asn for long-distance transport (Urquhart and Joy, 1981). In certain legume species, Asn can account for as high as 86% of transported nitrogen (Lea and Miflin, 1980). Asn is an ideal compound for nitrogen transport and storage, because of its high nitrogen:carbon ratio and its

stability (Lea and Miflin, 1980; Urquhart and Joy, 1981; Sieciechowicz et al., 1988). The role that Asn plays in nitrogen transport and storage is less well understood in nonlegume species. For Arabidopsis, recent studies of nitrogen assimilation have focused on the process of NOa- reduction to ammonia by nitrate and nitrite reductase (Crawford and Arst, 1993) and on its subsequent assimilation into Gln by GS (Peterman and Goodman, 1991). Here, we report the regulation of nitrogen metabolism downstream of Gln in Arabi- dopsis.

Standard biochemical studies of Asn biosynthesis in plants have been hampered by the instability of the AS enzyme (Huber and Streeter, 1985), the presence of enzyme inhibitors (Streeter, 1977), and the rapid tumover of Asn by asparaginase (Joy et al., 1983). Molecular biological approaches have been used to directly study AS gene structure and gene regulation (Tsai and Coruzzi, 1990, 1991). Two cDNA clones encoding Gln-dependent AS from Pisum sativum (AS2 and AS2) were obtained based on DNA sequence similarity with the human AS gene (Tsai and Coruzzi, 1990, 1991). The combined efforts of biochemical and molecular analyses have led to several important observations in the regulation of AS by environmental and physiological conditions. Expression pattem analyses revealed that transcription of both the AS2 and AS2 genes of P. sativum is repressed by light in leaves. In roots, AS2 is expressed constitutively, whereas AS1 is negatively regulated by light (Tsai and Coruzzi, 1991). Con- sistent with these molecular data, dark treatment has been reported to enhance the Asn content in phloem exudates of pea (Urquhart and Joy, 1981), as well as to increase AS enzyme activities in pea (Joy et al., 1983). The light repression of the pea AS2 gene is at least partly mediated via the photoreceptor phytochrome (Tsai and Coruzzi, 1990). In contrast to pea, the expression of a gene for Gln-dependent AS in asparagus is insensitive to light treatment (Davis and King, 1993). Levels of Asn and AS transcripts in asparagus do, however, increase in the tips of spears during the post- harvest period (Davis and King, 1993). This pattem of AS gene expression parallels the decline of SUC content in these spears (D.E. Irving, D.L. Hurst, unpublished data cited by Davis and King [1993]). The effects of sugars on AS enzyme activity in plants was also reported at the physiological level in sycamore cell cultures (Geri. et al., 1994) and root tips of com (Stulen and Oaks, 1977). Asn levels were found to

This work was supported by Department Of Energy pant DE- FG02-92-ER20071. Genetics Computer Group computer software for sequence analysis was supported by the National Science Foun- dation under grant No. DIR-8908095.

Present address: Department of Biochemistry, University of Texas Health Science Center at Houston, Houston, TX 77030. Abbreviations: AS, asparagine synthetase; GOGAT, glutamate

* Corresponding author; fax 1-212-995-4204. synthase; GS, glutamine synthetase; MS, Murashige and Skoog. 1347

www.plantphysiol.orgon August 30, 2020 - Published by Downloaded from Copyright © 1994 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

1348 Lam et al. Plant Physiol. Vol. 106, 1994

increase during SUC starvation of sycamore cell cultures (Ge- nix et al., 1994), and the level of AS activity in com root tips was reduced when plants were grown in the presence of Glc (Stulen and Oaks, 1977).

In this report, we examine the regulation of Asn biosynthesis by light and metabolic control at the molecular level in Arabidopsis thaliana. Based on this analysis, we have put forth a working model suggesting that when nitrogen:carbon ratios are high, the gene product of ASNZ functions to re- direct the flow of nitrogen into Asn, which acts as a shunt for storage and/or long-distance transport of nitrogen.

MATERIALS AND METHODS

Nomenclature of Genes

The ASNZ gene of Arabidopsis thaliana was named according to the recommendations of North American Arabidopsis Research Steering Committee. This nomenclature differs from the nomenclature for previously reported genes for plant AS (eg. AS1 and AS2 in pea [Tsai and Coruzzi, 19901; A S in asparagus [Davis and King, 19931). Genes names herein are italicized to distinguish them from their peptide products. For bacterial genes, standard nomenclature rules are applied.

Plant Materials and Growth Conditions

Mature plant materials of A. thaliana (ecotype Columbia) were harvested from plants grown on Metro-Mix Soil (Hum- mert, St. Louis, MO). Young seedlings were grown on MS agar plates (MS basal medium [Sigma], pH adjusted to 5.7 with KOH, 0.9% [w/v] agar). Where indicated, MS media were supplemented with sugars and/or other chemicals. Pla.nts in soil or tissue culture plates were grown in an environmental growth chamber (EGC, Chagrin Falls, OH) with daylength of 16 h, illumination of 80 WE m-’ s-I, and temperatures of 22OC (day) and 18OC (night). In some experiments, the plants were grown in continuous light (80 FE m-’ si-’,, 22OC) or in total darkness (18OC).

Isolation and DNA Sequence Analyses of the Arabidopsis ASNl cDNA Clone

Arabidopsis AS cDNA clones were isolated from a XZapII (Stratagene, La Jolla, CA) cDNA library generated from 3-d- old ethylene-treated dark-grown A. thaliana (ecotype Colum- bia) (Schindler et al., 1992). A DNA fragment containing the entire pea AS1 gene (Tsai and Coruzzi, 1990) was used to generate a heterologous DNA probe for screening the Arabi- dopsis cDNA library. Low-stringency hybridization (see below) was used for isolation of the Arabidopsis ASNZ cDNA clones. The pBluescript SKI1 plasmids (Strategene) containing the cDNA clones were excised from the XZapII vector, and double-stranded DNA sequencing was performed on plasmid clones using the Sequenase kit (United States Biochemical). Both DNA strands were sequenced, and all ambiguities were resolved. The DNA sequence was analyzed by the GCG Sequence Analysis software package (Genetics Computer Group, Inc., Madison, WI).

DNA and RNA Analyses

Genomic DNA was isolated from plant leaves using ELU- QUIK DNA purification kit (Schleicher & Schuell). T2tal RNA was obtained by a phenol extraction protocol (Jackson and Larkins, 1976).

Digoxigenin-labeled probes for northem blot and Southem blot analyses were generated by the digoxigenin DNA-labeling kit and digoxigenin RNA-labeling kit (Boehringer-hYIannheim Biochemicals, Indianapolis, IN) according to the manufacturer’s instructions. Radioactive probes were random-prime labeled with [cY-~’P]~AW (Boehringer-Mannheim Biochemical;). North- em blot analyses were performed with high-stringency hybridization conditions at a temperature of 65OC in Q W M 3 (Stra- tagene) solution for 1 h and at 42OC (DNA probes) or 65OC (RNA probes) in 50% (v/v) formamide hybridization solution overnight. Washing and chemiluminescent detection were performed according to the Boehringer-Mannheim Genius System User’s Guide.

Genomic Southern blot analyses were performed by run- ning 10 pg of digested DNA on a 1% (w/v) Tris-phosphate- EDTA (Sambrook et al., 1989) agarose gel. The DNA was transferred after the depurination, denaturation, m d neu- tralization steps as recommended by the Boehringer-Mann- heim Genius System User’s Guide, followed by low-stringency hybridization in QUIKHYB solution.

Low-stringency hybridization for the cDNA library screen and genomic Southern blot experiments were perfoimed at a temperature of 5OoC for 1 h in QUIKHYB solution, and the filters were then washed twice in 2X SSC (sodium chloride- sodium citrate buffer [Sambrook et al., 1989]), O . l % [w/v] SDS) solution at room temperature for 5 min and then l x SSC, 0.1% (w/v) SDS solution at 5OoC for 30 min. Chemil- uminescent detection was as described above for northem blots.

Phytochrome Experiments

Etiolated seedlings were grown as described in the legend of Figure 5. Both red and far-red light treatments were performed in wooden boxes contained in a dark room. Red- light treatment (a total of 58,000 pE m-*) was done by placing the seedlings under red fluorescent light bulbs. Far-red light treatment (a total of 800 pE m-’) was achieved by exposing the seedlings to light from incandescent light bulbs through an FRF 700 plexiglass filter (AIN Plastics, Mount Vemon, NY).

RESULTS

Cln-Dependent AS of Arabidopsis Contains a purl:-Type Gln-Binding Domain

The cDNA clone encoding the Gln-dependent AS2 gene of pea (Tsai and Coruzzi, 1990) was used as a hetmologous probe to isolate 16 independent cDNA clones from an Ara- bidopsis cDNA library (Schindler et al., 1992). Paríial DNA sequence data were obtained from the untranslatetl regions of these 16 independent cDNA clones (data not shown). These results revealed that all of the cDNAs encoded over- lapping portions of a single mRNA species. One of the longest



Metabolic Regulation of Arabidopsis ASNl Gene 1349

cDNA clones (pcAt-ASN1 ) containing the Arabidopsis ASNl gene was chosen for further analyses. DNA sequence data (GenBank accession No. L29083) showed that pcAt-ASN1 encodes a single, long, open reading frame. The translational start was assigned based on the strong sequence similarity between the Arabidopsis ASNl peptide and AS of other plants ( e g pea and asparagus [Fig. 11). The predicted ASNl peptide, which consists of 584 amino acids, has a M, of 65,546 and an isoelectric point of 6.63. This encoded ASNl enzyme also contains a purF-type Gln-binding domain (consisting of a Cys'-His'02-Asp29 triad [Mei and Zalkin, 19891) found at the N terminus of various AS peptides (Andrulis et al., 1987,

1989; Gong and Basilico, 1990; Richards and Schuster, 1992; Davis and King, 1993) (Fig. 1). This putative Gln-binding consensus is completely conserved at the N terminus of the Arabidopsis ASNl peptide (Fig. 1) and is evidence that ASNl encodes a Gln-dependent AS enzyme.

Structural Features of Cln-Dependent versus Ammonia- Dependent AS Enzymes in Several Different Organisms

The cloning of the Arabidopsis ASNl gene has added an extra piece of information to our understanding of the Gln- dependent AS enzyme in plants. The homology among the

At-ASN1 AO-AS Ps-AS1 Ps-AS2 ECO-AsnB Hum-AS

At-ASN1 AO-AS PS-AS1 PS-AS2 Eco-AsnF Hum-AS

At-ASN1 AO-AS Ps-AS1 Ps-AS2 Eco-AsnE Hum-AS

At-ASN1 AO-AS Ps-AS1 Ps-AS2 Eco-AsnB Hum-AS

At-ASN1 AO-AS Ps-AS1 Ps-AS2 Eco-AsnB Hum-AS

At-ASN1 Ao-AS Ps-AS1 Ps-AS2 Eco-AsnB Hum-AS

M C G I L A V L G C S D D S Q A K R V R V L E L S R R L R H R G P D - - - W S G L Y Q N G D N Y L A H Q R L A V I D P A S G D Q P L F N E D K T I V - V H - K F R T G S D C E V I A H L 105 ............................ K C.H . . CF.S ..... 105 .................... I . . . . . . . . . . H. H. Q.P..-..F.QC..D . . . . . . . . . . . . P.R . . . . . . . . . . . . . K .. S.FG.FD1KT.AVEL.K KA I.ASDNA1 ... E .. SIV .VNA. A...Y.QQ..H.-LA........QA..AEYGDRYQ.Q........IJL . 106 .... W.LF.SD.CLSVQCLSAMXIA---- . . . . . AFRFENVNGYTNCCFGFH .... V . . LF.M .. IRVKKYPYLWLCY ....... KKMQQHF--EFEYQ.KV.G.I.L.. 104

YEEYG-VDFVDMLDGIFSFLDTRDNSF~~AIG~SLYIGWGLDGSWISS~G~-----------------DDCEHFETFPPGHFYSSKLGGFKQWY" 197 . . . H.-E . . . . . . . . . . . H.-EN ..................... I... ..A . . 197

E. C . . . . . L....DS..RR.... S.YS 197

VTLKHSATPFLKVEPFLPGHY.VLDLK.N.KVA.VENVKYHHCRDV.LHA 214 . Q. K. -PE .Y.SEKDAYLIG..HL.IIP..M.YDEH.QLYVA... V-----------------PV.RTIKE..A.SYLW.QD.EIRSY.HRD..D 198 .DKG.IEQTIC .... V.A ...... ANKKVFLG..TY..RP.F.TE..FLAVC.. A

ES-VPSTPYEPLA-------I~F~AVI~~DVPFGVLLSGGLDSSLVASIT~LAGT-QWG----PQLHSFCVGLEGSPDLKAGKEV~YLG~EFHFS 295 .T-I .. ..AV . . . . . . . S ... T 295 .AI1 . . . . . L.N ....................... V . . . Y.. P......K.A..........DF........ E.T 296 .A-I .. T.T 295 YDA.KDNVTDKNE-------L.Q.L.DS.KSH . . S . . . Y...........IISA..KKYA.RRVEDQERSEAWW......A...P.......AQ...NH.......I.. T 301 LYDNVEKLFPGFEIETVK"L.1L.N . . . K . . . . . . RR1.C . . . . . . . . . . . . ATLLKQ.--KE.QV.YP------.QT.AI. M.D L.ARK..DHI.SE.Y.VL.N 316

VQDGIDAIEDVIYHVETYDVTTIRASTPMFLMSRKI-KSLGVKMVLSGEGADEIFGGYLYFHKAPNKKEFHQETCRKIKALHKYDCtRANKSTS~GLEARVPFLDKDFI 404 . . . . . . . . . . . . F.1 . . . . . . . . . . . . . . . . . A I....S................E...H..........Q........A...W.... E.M 404 . . . . . . . . . . . . . 405 ......... R . . . . . . . . . . Y.W .... . . . I

..................... s W.1 . . . . S . . . . . . . . . . . . . . . . E...E..........Q...Q......Y.W...........A.. 404 R . . . . . I..............Y......-.AM.I........S..V............A..L.E..V..LL...M...A....AM..W.V......... K.L 410

SEE .. Q.LDE . . FSL . . . . I .. V...VG.Y.I.KY.R.NTDSWIF....S..LTQ..I......SPEKAEE.SE.LLRE.YLF.V...DRT.A.H...L......H R.F 426

NTAMSLDPESKMIKPEEGRIEKWVLRRAFDDEERPYLPKHILYRQKEQFSDGVG---YSWIDGLKDHAAQNVNDKSNAGHIFPHNTPNTKEAYWRMIFERFFPQNSA 511 DV . . . I . . . . . . . . . DL . . . . . . . . . 511 KV . . DI . . . F . . . . HD . . . . . . . 512 .V .. NI ... 511 DV .. RIN.QD . . C--GN . K M . . HI . . EC.E----A ... ASVAW . . . . . . . . . . T..EV...Q.S.QQLET.RFR..Y...TS....L..E...EL..LP.. 511 SYYL . . P..MRIP.NG---...HL..ET.E.SN--LI..E..W.P..A....ITSVKN..FKI.QEWEHQ.D.A..A..AQK..F...K...G....QV...HY.GRAD 5 3 1

_ _ _

R L T V P G G A T V A C S T A K A V E S W S ~ P S G ~ I G ~ L S A Y D - - - - - - - - - G ~ ~ T I P P L K A I D ~ P ~ ~ G W I Q S .F . . . . . PSI . . . . . . . I....R....L.......L...D....PPLPSSISA..GA.MITNKKPR.VDVA---TP.... ST . . . . . . . PS . . . . . E .. I.........L.......L...V... EHQINP.KGVEPEKIIPKIGVSPLGVA1QT . . . . . . . P . . . . . . . . . . . . . . A....L.......L...D...ENH-"KTVEFEKIIP-LEA .PVELAI. AEC .... PS . . . . S . . . I...EAFKKMD.....-.V...Q... K W.S--------------HY.MPK.I.AT . . . A,-TLTHYK..VKA

Figure 1. Amino acid alignment of AS peptides. Amino acid alignment was performed using the Genetics Computer Group PILEUP program and then edited for display by the Genetics Computer Group MALINGED program. The sources of AS peptide sequences used in this comparison are: At-ASN1, Arabidopsis (this work); Ao-AS, asparagus (Davis and King, 1993); Ps-AS1, pea AS1 (Tsai and Coruzzi, 1990); Ps-AS2, pea AS2 (Tsai and Coruzzi, 1990); Eco-AsnB, E . coli AsnB (Scofield et al., 1990); Hum-AS, human AS (Andrulis et al., 1987). AS peptides of Chinese hamster (Andrulis et al., 1989) and Syrian hamster (Gong and Basilico, 1990) share high homology to human AS and have a conserved purf-type Gln- binding triad. For simplicity, only the human AS sequence is shown; other animal AS peptide sequences are highly homologous to human AS. Identical amino acids are represented by dots and gaps are represented by dashes. The asterisks indicate the residues of the putative purf-type Gln-binding triad: C, Cys'; D, Asp"; H, His'". Numbering of the conserved residues was according to the studies by Mei and Zalkin (1989). The conserved Cys residue is considered as in position 1, since the N-terminal Met residue is cleaved off in mature protein (Mei and Zalkin, 1989).

584 590 586 583 554 561




known AS peptide sequences was determined by the Genetics Computer Group PILEUP (Fig. 1) and FASTA (Table I) pro- grams. Homologous regions can be found throughout the entire sequence of Gln-dependent AS enzymes such as plant AS, animal AS, and Escherichia coli AsnB (Fig. 1). However, the amino acid sequences near the C-terminal regions of the plant AS peptides (about 30-40 amino acids) were found to be hypervariable when compared to each other (Fig. 1). In addition, this C-terminal region is absent in the animal AS and E. coli AsnB peptides (Fig. 1). This comparison also shows that the Gln-dependent AsnB from E. coli has a stronger homology to plant AS (53-57% identity) than to animal AS (42-47% identity) (Table I). This suggestion, that the evolutionary pathway of bacterial Gln-dependent AS is closer to plant than to animal AS, is supported by evolutionary trees generated using the Genetics Computer Group PILEUP program (data not shown). By contrast, the asnA gene of E. coli (Nakamura et al., 1981) encodes an ammonia-dependent AS enzyme that shares no homology with the Gln-dependent AS peptides when analyzed by the Genetics Computer Group FASTA program (data not shown). However, when we performed the Genetics Computer Group FASTA comparison after the amino acids were grouped according to their func- tional groups using the Genetics Computer Group SIMPLIFY program, homology was found between the AsnA protein arid the C-terminal half of the Gln-dependent AS peptides (data not shown). This observation is consistent with

the idea that the unique region of the N terminus of Gln- dependent AS contains the Gln-binding domain (Richards and Schuster, 1992). It is possible that the C-terminal half of AS, which is conserved between Gln- and ammonia- dependent AS enzymes, contains the Asp-binding domain (see “Discussion”).

Arabidopsis Contains a Single Gene for Cln-Dependent AS

Southern blot analysis was used to examine the total number of genes homologous to the Gln-dependent ASN1 gene in Arabidopsis. Arabidopsis genomic DNA digested with three different restriction enzymes (Fig. 2, lanes 4-6) wa- f raction- ated on a 1.0% (w/v) agarose gel. An internal PCR fragment, which covers the N terminus of the ASNI-coding region (including the purF Gln-binding domain), was used to probe the Southem blots. Low-stringency hybridization (conditions were used that enabled the Arabidopsis pcAt-ASN1 cDNA probe to cross-hybridize to a DNA fragment containing the heterologous pea AS1 cDNA (Fig. 2, lane 3). Under these low-stringency conditions, a single DNA band hyt>ridized to the ASNl probe in all three different genomic restriction digests of Arabidopsis genomic DNA (Fig. 2, lanes 4-6). Together with the fact that there was only one mRPJA species identified during the cDNA library screen (see above), these results strongly suggest that Arabidopsis contains a single gene for Gln-dependent ASNl.

Table 1. Comparison of different AS peptide sequences The predicted peptide sequence of Arabidopsis ASNl was compared to the Gin-dependent AS

peptide sequences from asparagus (Davis and King, 19931, pea (Tsai and Coruzzi, 1990), E . coli (AsnB [Scofield et al., 19901, human (Andrulis et ai., 1987), Chinese hamster (Andrulis et ai., 1989), Syrian hamster (Gong and Basilico, 1990), and the ammonia-dependent AS from E . coli (AsnA [Nakamura et ai., 19811). The numerical values represent t h e percentage identity between two peptides computed by the Genetics Computer Group FASTA program. The values in parentheses represent the computation done after the amino acid data were transformed with the Genetics Computer Group SIMPLIFY program (A = P, A, G, S, T; D = Q, N, E, D, 6, Z; H = H, K, R; I = L, I, V, M; F = F, Y, W; C = C ) . The - indicates that without prior data transformation, homology cannot be identified between t h e ammonia-dependent E. coli AsnA and t h e Gln-dependent AS enzymes. Homology between Gin- and ammonia-dependent AS was found only in the C-terminal half of the peptides after data transformation with SIMPLIFY (data not shown). By contrast, all of t h e Gln-dependent AS enzymes show identities throughout the sequence (see Fig. 1 ) .

Pea Pea E. coli Chinese Syrian AS2 AsnB Human Hamster Hamster Arabidopsis Asparagus AS,

Asparagus 84.3 (92.3)

Pea-AS1 81.4 81.7 (90.4) (89.8)

Pea-AS2 84.9 84.5 85.1 (90.7) (90.4) (92.3)

E . coli AsnB 55.0 56.6 53.0 54.1 (73.6) (73.1) (71.5) (72.0)

Human 47.2 46.3 45.0 45.6 43.3 (55.2) (54.3) (53.8) (52.8) (53.6)

Chinese Hamster 46.9 47.9 46.3 46.9 41.0 95.0 (54.9) (54.7) (55.1) (54.5) (55.0) (97.7)

Syrian Hamster 45.6 46.6 45.6 46.3 43.3 94.7 98.6 (55.1) (54.8) (55.1) (54.4) (53.6) (98.0) (99.3)

E . coli AsnA (44.9) (44.3) (37.7) (42.0) (41.4) (39.9) (38.5) (37.0) .

- - - - - - - -



Metabolic Regulation of Arabidopsis ASN1 Gene 1351

c!•en

S

<u§O<

oo

1

oozQ0

Pea

AS I |OQ Uj

3

1 2 3 4 5 6

Figure 2. Genomic Southern blot analyses of Arabidopsis ASN1gene. Lane 1, Digoxigenin-labeled H/ndlll-digested X-DNA used asmolecular weight (M.W.) marker (from Boehringer-Mannheim); lane2, an EcoRI fragment containing the Arabidopsis A5N7 cDNA clone;lane 3, an EcoRI fragment containing pea AS/ cDNA clone; lanes 4,5, and 6, Arabidopsis genomic DNA digested with SamHI, EcoRI,and H/ndlll, respectively. The Southern blot was probed with aDNA probe that spans the N-terminal purf-type Gin-binding do-main of ASN1.

shoot tissues as compared to roots of dark-adapted plants(Fig. 4). We failed to detect ASN1 mRNA in roots of plantsgrown under a variety of conditions (data not shown). Thelack of detectable ASN1 mRNA in Arabidopsis roots is incontrast to pea, in which both AS1 and AS2 genes are ex-pressed in roots (one constitutively and one light regulated)(Tsai and Coruzzi, 1991).

Control of Arabidopsis ASN1 mRNA Levels byPhytochrome

To test whether light exerts a direct effect on the expressionof the ASN1 gene in Arabidopsis, we monitored whetherphytochrome was involved in mediating the negative effectsof light on ASN1 mRNA accumulation. ASN1 mRNA levelswere monitored in etiolated plants exposed to white light,red light, or red/far-red light treatment (Fig. 5). Etiolatedseedlings showed high levels of ASN1 mRNA (Fig. 5, lane 1).White light treatment of these etiolated seedlings caused astrong repression of ASN1 mRNA accumulation (Fig. 5, lane2). A treatment of red light reduced the level of ASN1transcript (Fig. 5, lane 3) compared to etiolated plants. Thisred light effect could be at least partially reversed by asubsequent pulse of far-red light (Fig. 5, lane 4). As a control,the transcript levels of the gene for cytosolic GS (GSR2) ofArabidopsis (Peterman and Goodman, 1991) were also mon-itored, and light was found to have no negative effect on the

Light Dark

GS2(GSL1)

Expression of the Arabidopsis ASN1 Gene Is Affected byLight and Tissue Type

Previous studies in pea and tobacco have shown that lightrepressed the transcription of AS1 gene in leaves (Tsai andCoruzzi, 1990, 1991). In pea, the AS1 mRNA levels are alsonegatively regulated by light in roots, whereas the AS2 mRNAis constitutively expressed in roots (Tsai and Coruzzi, 1991).By contrast, the level of mRNA for AS in asparagus spears isinduced during harvesting and is insensitive to light treat-ments (Davis and King, 1993). Our results indicate that theASN1 mRNA of Arabidopsis accumulates to high levels inshoots of dark-adapted plants (Fig. 3, lane 2) and at very lowlevels in shoots of light-grown plants (Fig. 3, lane 1). As areciprocal control, the mRNA for the nuclear gene (GSL1)encoding chloroplast GS2 of Arabidopsis (Peterman andGoodman, 1991) was shown to be light induced (Fig. 3, lane1), consistent with the results reported by Peterman andGoodman (1991). These opposite effects of light on thetranscription of ASN1 and GS2 genes of Arabidopsis mayreflect the preference for Gin or Asn as nitrogen carriers inArabidopsis grown under different light conditions (McGrathand Coruzzi, 1991).

With regard to organ specificity, our results indicated thatthe Arabidopsis ASN1 gene is expressed predominantly in

ASN1

rRNA

1

Figure 3. Light has a negative effect on A5N/ mRNA levels inArabidopsis. The effects of light repression on ASN? expressionwere tested in mature plants. Plants were grown on soil under a16-h light/8-h dark cycle for 2 weeks and transferred to continuouslight (lane 1) or continuous darkness (lane 2) for 5 d. Total RNA (10Mg) was used for each of the lanes. Hybridization was performedwith [a-32P]dATP-labeled CSU (Arabidopsis chloroplastic CS2 [Pe-terman and Goodman, 1991]) or digoxigenin-labeled A5N/ DNAprobes. The nitrocellulose filter was first hybridized with GS2 probe,then stripped, and re-hybridized with the ASN1 probe. www.plantphysiol.orgon August 30, 2020 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.



Shoot Root

ASN1

rRNA

Figure 4. Tissue-specific gene expression of ASN1 in Arabidopsis.Arabidopsis seeds were grown on MS plates plus 3% (w/v) Sueunder a 16-h light/8-h dark cycle for 2 weeks. The plants were thentransferred to medium without Sue and grown in the dark for 2.5d. Total RNA (10 ^g) from shoot (lane 1) or root (lane 2) was usedfor northern analyses and probed with ASN7 gene.

taming amino acid supplements (used as endogenous reducednitrogen source) such as Asn, Gin, and Glu (Fig. 7). Todemonstrate that these amino acids are taken up by theplants, we first determined the concentrations of these aminoacids required to reverse the effects of an AS and GOGATinhibitor albizziin (a Gin analog) (Lea and Fowden, 1975).We have shown that albizziin inhibits Arabidopsis root growthand that the concentrations of amino acids used in Figure 7could substantially reverse the growth inhibition by albizziin(Table II). These levels of amino acids did not significantlyaffect the normal growth of plants when supplied in theabsence of the inhibitor (data not shown). The fact that theexogenously applied amino acids can reverse the inhibitoryeffect of albizziin supports the notion that the exogenouslysupplied amino acids were affecting the internal pools ofamino acids. The results shown in Figure 7 indicate that ASN1expression in dark-grown plants was at a maximum whenplants were grown in the absence of exogenously appliedSue (Fig. 7, lanes 1-4). Under these conditions of 'low* carbonavailability, ASN1 expression is high and treatment with anyof the amino acids has no observable effects on ASN1 expres-

accumulation of CSR2 mRNA (Fig. 5, lanes 1-4). Althoughlight treatment did slightly induce the expression of GSR2mRNA (Fig. 5, lanes 2-4), a far-red pulse did not reverse theeffect of red light treatment. These results strongly supportthe hypothesis that in Arabidopsis, light can directly affectthe expression of ASN1, at least in part, via the photoreceptorphytochrome.

Metabolic Control of ASN1 Transcription in Arabidopsisby Carbon and Nitrogen Sources

To test whether the transcription of ASN1 is also subject tometabolic regulation, we first examined the effect of Sue onASN1 mRNA levels in etiolated seedlings. It was found thatSue strongly repressed the level of ASN1 transcripts in etio-lated seedlings (Fig. 5, cf. lanes 1 and 5). The levels of acontrol mRNA (GSR2) were not affected by Sue (Fig. 5, cf.lanes 1 and 5). Further studies of Sue effects were carried outusing the shoot and root tissues of 2-week-old dark-adaptedseedlings (Fig. 6). In shoot tissue, the level of ASN1 mRNAin plants grown on MS media (Fig. 6, lane 2) was stronglyrepressed by the addition of Sue (Fig. 6, lanes 3 and 5) butnot by the nonmetabolizable sugar mannitol of the samew/v concentration (Fig. 6, lanes 4 and 6). This control dem-onstrates that the Sue repression of ASN1 mRNA accumula-tion is not due to osmotic effects. Thus, the expression ofASN1 in Arabidopsis shoots of dark-adapted plants is under astrong metabolic control by carbon availability. That is, highcarbon levels repress the expression of ASN1. In root tissue,no ASN1 expression was detected with any of the treatments(data not shown).

Since Asn serves as a nitrogen transport compound inplants (Lea and Miflin, 1980; Urquhart and Joy, 1981; Siecie-chowicz et al., 1988), we also tested whether manipulatingthe levels of endogenous reduced nitrogen had effects onASN1 expression in Arabidopsis. The levels of ASN1 transcriptswere measured in dark-adapted plants grown in media con-

rRNA

Figure 5. Phytochrome mediates the effects of light on ASN/ tran-scription in Arabidopsis. Etiolated Arabidopsis seedlings were grownon MS plates either without Sue (lanes 1 -4) or with 3% (w/v) Sue(lane 5) for 5 d in complete darkness. The seedlings were dividedinto groups that were treated with different light conditions priorto harvesting. Lanes 1 and 5, Seedlings harvested in total darkness;lane 2, dark-grown seedlings transferred to continuous white lightfor 3 h prior to harvesting; lanes 3 and 4, dark-grown seedlingssubjected to red light treatment (lane 3) or red light treatmentfollowed by a subsequent pulse of far-red light (lane 4). Followingred/far-red treatments, plants were exposed to a dark period of 6h prior to harvesting. The expression of both ASN) and a gene forArabidopsis cytosolic GS (CSR2) (Peterman and Goodman, 1991)was detected by northern analyses on a replicate RNA blot (10 /igof total RNA per lane). www.plantphysiol.orgon August 30, 2020 - Published by Downloaded from



Metabolic Regulation of ArabidopsisASNl Gene 1353

2U

CO

«CO CO

2 2c«

ccU

c3 2v? vPo^ o^U) U)

CO CO

rRNA

1 6

Figure 6. Sue effects on ASN1 transcription in Arabidopsis. Arabi-dopsis seeds were grown on MS media plus 3% (w/v) Sue under a16-h light/8-h dark cycle for 2 weeks. The plants were then trans-ferred to different media described below and continued to growin complete darkness for 2.5 d. Lane 1, Control of soil-grown dark-adapted plants (see legend to Fig. 3); lane 2, MS medium with nosugar; lane 3, MS medium with 3% (w/v) Sue; lane 4, MS mediumwith 3% (w/v) mannitol; lane 5, MS medium with 4.5% (w/v) Sue;lane 6, MS medium with 4.5% (w/v) mannitol. Northern analysiswas performed under high-stringency conditions in 50% (v/v) form-amide solution using the ASN1 gene as probes.

% Sucrose 0AsparaglneGlutamlneGlutamate

Figure 7. Effects of amino acid supplementation on ASN1 transcrip-tion in Arabidopsis. Arabidopsis seeds were grown on plates con-taining MS medium plus 3% (w/v) Sue under a 16-h light/8-h darkcycle for 2 weeks. The plants were then transferred to mediadescribed below and grown in complete darkness for 2.5 d. Lanes1 to 4, MS medium with no sugar; lanes 5 to 8, MS medium with3% (w/v) Sue. MS was supplemented with 0.4 mM Asn (lanes 2 and6), 3.4 mM Gin (lanes 3 and 7), or 3 mM Clu (lanes 4 and 8). Theexpression of both ASN1 and a cytosolic CS (C5R2) (Peterman andGoodman, 1991) was detected by northern analyses (under high-stringency conditions in 50% [v/v] formamide solution) on replicateRNA blots (10 n% of total RNA per lane).

Table II. Rescue of albizziin inhibition of root growth by Asn, Clu,and Cln

Albizziin (a Gin analog) inhibits the growth of Arabidopsis roots(see B and F) compared to Arabidopsis grown on MS media with noadditions (see A and E). Supplemental additions of Asn (C); gluta-mate (Glu) (D), and Gin (G) were able to reverse in part theinhibitory effect of albizziin on root growth. These amino acid (Asn,Glu, Gin) concentrations supplied in the media, therefore, affectendogenous pools of amino acids and thus establish the concentra-tions of amino acid used in Figure 7. Plants were grown on MStissue culture plates supplemented with 3% Sue and other constit-uents as indicated. Alb, 0.5 mM albizziin; Asn, 0.4 mM Asn; Glu,3 mM glutamate; Gin, 3.4 mM glutamine. Root lengths were meas-ured as the parameter of plant growth. N, Number of plants meas-ured. Statistical analyses (t test) showed that additions of Asn, Glu,and Gin significantly (P < 0.0001) increased the root lengths ofalbizziin-treated plants.

Treatment

Experiment 1 :A. No additionsB. AlbC. Alb plus AsnD. Alb plus Glu

Experiment 2:E. No additionsF. AlbG. Alb plus Gin

Growth

d

9999

121212

N

10981

107105

242630

Mean ± SD

cm

2.89 ± 0.330.53 ± 0.201.08 ±0.181.44 + 0.44

3.88 ± 0.570.82 ± 0.271.99 + 0.50

sion (Fig. 7, cf. lane I to lanes 2-4). However, if ASN1transcription is repressed by Sue (Fig. 7, lane 5), this effectcould be partially relieved by supplementation with aminoacids (Fig. 7, lanes 6-8). As a negative control, the levels ofGS.R2 mRNA (cytosolic GS of Arabidopsis) were found to berelatively constant under the conditions tested (Fig. 7). Theimplication of these results on the possible relations amongnitrogen:carbon ratio, ASN1 expression, and nitrogen flux,will be discussed below.

DISCUSSION

Comparison of the Putative Arabidopsis ASN1 Peptidewith Other AS Enzymes

In E. co/i, there are two different genes for AS encodingtwo different enzymes. The asnA gene encodes an ammonia-dependent AS enzyme (Nakamura et al., 1981) and the asnBgene encodes a Gin-dependent AS enzyme (Scofield et al.,1990). Although not yet rigorously established, it is thoughtthat the enzyme encoded by asnA may be more tightlyregulated by Asn pools in cells, since the asnA gene isnegatively regulated by its end product (de Wind et al., 1985).This notion is further supported by the fact that there arevery few Asn residues in the AsnA protein (3 per 330 aminoacid residues) (Nakamura et al., 1981). This feature is a typicalcharacteristic of a bacterial amino acid biosynthetic enzyme.In animals and plants, the Gin-dependent AS enzyme is theonly form of AS that has been identified to date at both thebiochemical level (Sieciechowicz et al., 1988) and the genelevel (Tsai and Coruzzi, 1990, 1991; Davis and King, 1993). www.plantphysiol.orgon August 30, 2020 - Published by Downloaded from




One interesting observation is that a11 Gln-dependent AS genes of plants encode proteins with a relatively high leve1 o1 Asn residues (e.g. 24 per 584 amino acid residues in Arabidopsis ASNZ protein). Since the ASNl peptide contains a relatively high content of Asn residues, it is likely that this peptide will be made efficiently only under conditions of njtrogen excess, when Asn supplies are not limiting.

The major biosynthetic pathway for Asn in plants is via the dimeric enzyme Gln-dependent AS (Van Heeke and Schuster, 1989a), which transfers the amido group of Gln to Asp and thus leads to the formation of Asn (Milman and Cooney, 1979). Peptide sequence analyses suggested that Gln-dependent AS in both animals and plants contain a purF-type Gln-binding domain, which was first reported in prokaryotes (Tso et al., 1982). Site-directed mutagenesis studies in purF-type Gln amidotransferase of E. coli showed that a Cy~'-His'~'-Asp*~ catalytic triad located at the N terminus of' the peptide is involved in the Gln-amide transfer function (Mei and Zalkin, 1989). The cleavage of the N-formylme- thionine seems to be necessary to expose the catalytic Cys (Tso et al., 1982). In the animal AS enzyme, experiments using monoclonal antibodies also revealed separate Gln- and Asp-binding sites (Pfeiffer et al., 1986, 1987). Furthermore, site-directed mutagenesis studies in the human AS gene also showed that the N-terminal Cys is essential for its Gln- dependent activity (Van Heeke and Schuster, 1989b). A11 purF-type enzymes except glucosamine-6-phosphate synthase can utilize ammonia as a nitrogen source in the absence of Gln (Badet et al., 1987; Zalkin, 1993), although plant AS enzymes in general have a lower K, for Gln than for ammonia (Lea and Miflin, 1980). Mutations resulting in a change of the conserved Cys, His, or Asp residue will result in the loss of Gln-dependent but not of ammonia-dependent activity in purF-type enzymes (Mei and Zalkin, 1989).

We have cloned the single gene (ASNZ) coding for a Gln- dependent AS enzyme from Arabidopsis. The sequence of this gene combined with the DNA sequence of the E. coli Gln-dependent asnB gene, which has recently become available (Scofield et al., 1990), has allowed us to examine the relationship among Gln- and ammonia-dependent AS peptide sequences. One model of Gln-dependent enzymes of the purF-type proposed that the Cys-His-Asp triad is involved in Gln binding (Mei and Zalkin, 1989). This triad is also conserved in Arabidopsis ASN1. The N-terminal Cys is conserved in a11 Gln-dependent AS enzymes, indicating that this residue is absolutely essential. However, there is one exception each for the conserved His and Asp residues. The conserved His and Asp residues are missing from E. coli AsnB (Scofield et al., 1990) and pea AS2 (Tsai and Coruzzi, 1990), respectively (Fig. 1). Surprisingly, this deviation from the purF-type Gln- binding consensus does not weaken the binding affinity of Gln to the enzyme AsnB encoded by E. coli. In fact, E. coli AsnB has a K, for Gln that is an order of magnitude smaller when compared to human AS (Richards and Schuster, 1992). The deviation of the normally conserved Asp residue in the pea AS2 peptide also complicates the picture. A recent study of the Gln-dependent nitrogen transfer in E. coli AsnB showed that the conserved Asp residue does not appear to mediate the nitrogen reaction (Boehlein et al., 1994). Site- directed mutagenesis experiments of other putative His resi-

I

dues also suggested that none of them plays a catídytic role (Boehlein et al., 1994). The role of N-terminal C p in Gln binding is also challenged (Boehlein et al., 1994). It seems that either E. coli AsnB represents an exceptional c(3s.e or the mechanism of Gln binding in AS is more complicated then was previously thought.

The peptide sequence comparison of AS enzymes shows that E. coli AsnB shares a much higher sequence liomology to Gln-dependent AS in eukaryotes than it does to its ammonia-dependent counterpart in E. coli (AsnA) (Tatde I). This suggests that the E. coli asnA and asnB genes may come from different evolutionary origins. Another interesting, observation is that the homology between Gln-dependent AS enzymes is higher between plants and bacteria thari between plants and animals. This analysis suggests that the wolution- ary pathways among eukaryotes is not always closer than that observed between eukaryotes and prokaryotes. To study the relationship between the Gln- and ammonia-tlependent AS enzymes, the peptide sequence data were compared after transformation with the Genetics Computer Gmup SIM- PLIFY program. Consistent with the studies using monoclonal antibodies (Pfeiffer et al., 1986, 1987), it appears that Gln-dependent AS enzymes contain a separate Asp-binding domain in the C-terminal half (see "Results"). A functionally homologous Asp-binding domain may exist in the xn"mnia- dependent AsnA of E. coli. One distinct feature of plant Gln- dependent AS peptides of plants is the hypervariable region located at their C-terminal ends (Fig. 1). This hypervariable region is absent from any of the other AS peptides n bacteria and animals. Whether this region plays a regulatory role in the expression and function of plant Gln-depenclent AS is still unknown.

Light Regulation of Arabidopsis ASNl Transcription

In a series of experiments, we examined how P.SN1 transcription is regulated in Arabidopsis. These results reveal comparisons and contrasts with AS genes of different plant species. Arabidopsis appears to contain a single nuclear gene for AS (ASNZ), whereas peas contain two genes f c r AS. The single Arabidopsis ASNl gene is expressed predorrinantly in shoots, and its mRNA is undetectable in roots. Tkus finding contrasts with AS expression in pea roots and suggests that the expression of AS1 and AS2 genes in roots of p?a may be related to the nodulation processes. For instance, a significant amount of nitrogen fixed in roots of temperate legumes is transported as Asn (Scott et al., 1976). By contrast, it appears that the primary site for Asn synthesis in Arabidopsis is in the shoot tissue, where photosynthesis is taking place.

The ASNZ gene of Arabidopsis is expressed mainly in the aerial parts of the plants, and accumulation of AS transcript is subject to light repression. It was anticipated that light could exert its effects on ASNl transcription directly via a photoreceptor. Altematively, light might exert its effects on ASNZ transcription indirectly by altering the meta bolic pool of carbon skeletons generated during photosynthesis. To distinguish these effects of light (direct or indirect) on the control of Arabidopsis Gln-dependent ASNl, we tested the involvement of phytochrome on ASNZ transcription in etiolated seedlings. These studies showed that phftochrome




plays a role in the transmission of light signals to repress the accumulation of the Arubidopsis A S N l mRNA. Thus, light exerts a direct effect on A S N l transcription via phytochrome. This effect occurs within a short time (3-6 h) and, hence, is likely to be physiologically significant. How phytochrome mediates the light regulation of gene expression is still an open question for genes positively regulated by light (e.g. rbcS) (Quail, 1991). A recent report suggests that a guanine nucleotide-binding protein may be involved (Romero and Lam, 1993). It is noteworthy that ASNl represents one of a few examples of light-repressed plant genes, which include those encoding phytochrome (Otto et al., 1984; Lissemore and Quail, 1988; Kay et al., 1989) and Pchlide reductase (Mosinger et al., 1985). Further genetic and molecular analyses of the Gln-dependent A S N l gene in the model plant Arubidopsis may shed light on the factors that mediate the negative effects of light on plant gene expression.

Metabolic Regulation of Arabidopsis ASNl Transcription

Since light significantly affects the physiology and metabolic flow in plants, we explored whether light may also exert effects on gene regulation indirectly through metabolic control. For instance, light stimulates photosynthesis and, hence, increases the cellular sugar content. Metabolic repression of gene expression by SUC, Glc, and acetate has been reported for genes involved in carbon fixation using a maize protoplast transient assay system (Sheen, 1990). It appears that metabolic control may even ovemde the direct effects of light and other forms of regulation (Sheen, 1990). Previous physiological studies suggested that carbon metabolites may affect Asn biosynthesis. In asparagus, AS transcription is induced in spears during harvest (Davis and King, 1993). This timing of mRNA accumulation coincides with a decline in Suc content of spears (D.E. Irving, D.L. Hurst, unpublished data cited by Davis and King [1993]). In corn root tips, AS enzyme levels were reduced in the presence of exogenous Glc (Stulen and Oaks, 1977).

In Arubidopsis, we tested the effects of supplying exogenous carbon on A S N l mRNA levels. These studies showed that Suc repression of A S N l transcription overshadows the induc- tion effect of dark treatment in both etiolated seedlings and in mature green plants. Hence, it appears that when the levels of carbon skeletons are low (such as in dark growth conditions) the Arubidopsis A S N l gene is induced. This may reflect the fact that Asn has a higher nitrogen:carbon ratio relative to Gln and, hence, is a more economical storage compound for cellular nitrogen and is the favored transport compound when carbon levels are low (Lea and Miflin, 1980; Urquhart and Joy, 1981; Sieciechowicz et al., 1988). To further establish the relationship between cellular nitrogen:carbon levels and the transcription of the ASNl gene in Arubidopsis, we tested the effects of amino acid supplementation in the presence and absence of Suc (Fig. 7). In the absence of Suc (and under dark growth conditions), A S N l in Arubidopsis is maximally expressed, and there are no effects of amino acid supplementation to the media. However, A S N l transcription is repressed by SUC, and the effect of Suc repression can be partially relieved by amino acid supplementation. We are confident that the exogenous supplements

of amino acids in the media affect the endogenous pools, since we have shown that the concentrations of amino acids used were able to reverse the toxic effects of the Gln analog albizziin (Table 11). Thus, we conclude that the level of Arubidopsis A S N l transcript accumulation is regulated by the nitrogen:carbon ratio rather than the absolute amount of any of the amino acids tested. Since the studies reported here are based on treatment of plants under extended dark period, it cannot be concluded whether this phenomenon occurs diur- nally. However, it has been previously shown for pea AS that AS1 mRNA accumulates in a diurnal cycle and that this effect is due to the light-dark cycle and not to circadian rhythm (Tsai and Coruzzi, 1991).

It should be pointed out that the nuclear gene for chloroplast GS2 in Arubidopsis (Peterman and Goodman, 1991) and other plants (Tingey et al., 1988; Edwards and Coruzzi, 1989) is induced by light treatment and is transcribed at very low levels under dark growth conditions. Although the effects of metabolic control on GS2 expression have not yet been tested, it seems that GS2 and AS enzymes in plants should work in a coordinated manner to control the flow of nitrogen into Gln versus Asn (McGrath and Coruzzi, 1991). This regulation is known to be directly affected by light (Tingey et al., 1988; Tsai and Coruzzi, 1990; Peterman and Goodman, 1991) and may also include indirect effects such as a change in nitrogen:carbon ratio. Moreover, the gene expression of nitrate reductase is also positively regulated by light and responds to changes in nitrogen:carbon ratio (Crawford and Arst, 1993; Vincentz et al., 1993).

We have put forth a working model to integrate the above observations into a scheme for the molecular regulation of Asn biosynthesis with regard to the carbon and nitrogen metabolic status of a plant (Fig. 8). In a nonleguminous plant such as Arubidopsis, soil Nos- is the primary nitrogen source. After NOs- is transported to the aerial part of the plant through xylem, NO3- is reduced to ammonia through the combined action of nitrate reductase in cytosol and nitrite reductase in chloroplasts (Crawford and Arst, 1993). The ammonia generated in chloroplasts is then incorporated into Gln by chloroplastic GS2 (Lea, 1993). Gln thus generated will be used as the amide donor for Glu biosynthesis (via GOGAT) or for the biosynthesis of other amino acids and nucleic acids. Under light growth condltions, in which the nitrogen:carbon ratio is low, the GS2/GOGAT cycle will be activated to generate Gln and Glu for active anabolic processes while Asn biosynthesis is suppressed. This hypothesis is supported by the observations that the transcription of both GS2 (Peterman and Goodman, 1991) and a Fd-dependent GOGAT (K. Coschigano, G. Coruzzi, unpublished data) in Arubidopsis are induced by light, whereas A S N l expression is repressed by light. Under these circumstances, the ammonia pool is maintained by the combined actions of nitrate and nitrite reductases. The expression of nitrate reductase has been found to be induced by both light and a low nitrogen:carbon ratio (Crawford and Arst, 1993; Vincentz et al., 1993). Like nitrate reductase, nitrite reductase is positively regulated by light (Crawford and Arst, 1993). In dark growth conditions (low carbon availability), however, nitrogen is transported and/or stored as the more economic molecule Asn, biosynthesized from Gln and Asp. Thus, Gln-dependent




LIGHT (Low N:C) ASNl gene expre~sion OFF

DARK (High N:C) I ASNl gene expression ON

I I T A S P

I G in I I

o I

I I I ROOT + 1 G l n

t Asn

SOIL NOS-

Figure 8. A working model depicting the metabolic control of nitrogen assimilation into Gln and Asn and effects on ASNl gene expression in Arabidopsis. NOs- taken up from the soil is subse- quently reduced into NO2-, and then NH,+, by the combined actions of nitrate reductase ( N R ) and nitrite reductase (NiR). NH,+ molecules generated in the chloroplasts then enter GS/COGAT cycle (using chloroplastic GS2 and Fd-dependent COGAT) to generate Gln and Glu. Gln and Glu serve as nitrogen donors for the synthesis of other amino acids (AA) and nucleic acids (NA). This Cln-generating cycle continues to operate under light growth conditions (low nitrogen:carbon ratio). Under dark growth conditions (high nitrogen:carbon ratio), nitrogen assimilated into Cln is converted to Asn by AS where Asn is synthesized for storage of nitrogen. The model predicts that the synthesis of Asn and ASNl gene expression only occur when the nitrogen:carbon ratio is high. aKG, a-Ketoglutarate; Asp, aspartate.

AS may act as an important switching enzyme in the metabolic flow of nitrogen in Arubidopsis under changes of physiological conditions (such as light versus dark). Here we show that this regulation is at the level of ASNl transcription. Although this hypothesis is waiting to be tested by more rigorous biochemical and genetic studies, it is consistent with the available data and future studies will be directed toward this effort.

ACKNOWLEDGMENTS

The Arabidopsis cDNA library was a generous gift from Dr. Joseph Ecker. GSLl and GSR2 cDNA clones were generously provided by Dr. T.K. Peterman and H. Goodman. We thank Brett Bialer and Oliver Pihlar for sample and data collections. We also thank Nora Ngai and Dr. Albert0 Mancinelli for their guidance and assistance with phytochrome experiments. We also acknowledge Dr. Gabrielle Tjaden for helpful discussions.

Received March 22, 1994; accepted July 22, 1994. Copyright Clearance Center: 0032-0889/94/l06/1347/11.

LITERATURE CITED

Andrulis IL, Chen J, Ray PN (1987) Isolation of human cDNAs for asparagine synthetase and expression in Jensen rat sarcoma cells. Mol Cell Biol7: 2435-2443

Andrulis IL, Shotwell M, Evans-Blacker S, Zalkin H, Siminovitch

L, Ray PN (1989) Fine structure analysis of the Chinese hamster AS gene encoding asparagine synthetase. Gene 8 0 75-85

Badet B, Vermoote P, Haumont PY, Lederer F, LeGoffil: F (1987) Glucosamine synthetase from Escherichia coli. Purification, properties and glutamine-utilizing site location. Biochernistry 2 6

Boehlein SK, Richards NGF, Schuster SM (1994) Glutamine-dependent nitrogen transfer in Escherichia coli asparagine synthetase 8. Searching for the catalytic triad. J Biol Chem 269 7450-7457

Crawford NM, Arst HN Jr (1993) The molecular genetics, of nitrate assimilation in fungi and plants. Annu Rev Genet 27: 1 15-146

Davis KM, King GA (1993) Isolation and characterization Jf a cDNA clone for a harvest induced asparagine synthetase from Asparagus officinalis L. Plant Physiol102 1337-1340

de Wind N, de Jong M, Meijer M, Stuitje AR (1985) Siíe-directed mutagenesis of the Escherichia coli chromosome near oriC identification and characterization of asnC, a regulatory eleinent in E. coli asparagine metabolism. Nucleic Acids Res 1 3 8797-8811

Edwards JW, Coruzzi GM (1989) Photorespiration and light act in concert to regulate the expression of the nuclear gene j'or chloroplast glutamine synthetase. Plant Cell 1: 241-248

Genix P, Bligny R, Martin J-B, Douce R (1994) Transient accumulation of asparagine in sycamore cells after a long period of sucrose starvation. Plant Physiol94 717-722

Gong SS, Basilico C (1990) A mammalian temperature-sensitive mutation affecting G1 progression results from a single ,imino acid substitution in asparagine synthetase. Nucleic Acids Res 18:

Huber TA, Streeter JG (1985) Purification and propertie; of asparagine synthetase from soybean root nodules. Plant Sci 4 2 9-17

Jackson AO, Larkins BA (1976) Influence of ionic strength, pH, and chelation of divalent metals on isolation of polyribosomes from tobacco leaves. Plant Physiol57: 5-10

Joy KW, Ireland RJ, Lea PJ (1983) Asparagine synthmis in pea leaves, and the occurrence of an asparagine synthetasc inhibitor. Plant Physiol73 165-168

Kay SA, Keith B, Shinozaki K, Chye ML, Chua NH (1989) The rice phytochrome gene: structure, autoregulated expression, and binding of GT-1 to a conserved site in the 5' upstredm region. Plant Cell 1: 351-360

Lea PJ (1993) Nitrogen metabolism. In PJ Lea, RC Leegood, eds, Plant Biochemistry and Molecular Biology. John Wiley & Sons, London, pp 155-180

Lea PJ, Fowden L (1975) The purification and properties of glutamine-dependent asparagine synthetase isolated from Idupinus al- bus. Proc R SOC Lond B Biol Sci 192 13-26

Lea PJ, Miflin BJ (1980) Transport and metabolism of asparagine and other nitrogen compounds within the plant. In PK Stumpf EE Conn, eds, The Biochemistry of Plants Academic Press. New York,

Lissemore JL, Quail PH (1988) Rapid transcription regulation by phytochrome 06 the genes for phytochrome and chlorophyll a b - binding protein in Avena sativa. Mol Cell Biol8: 4840-4850

McGrath RB, Coruzzi GM (1991) A gene network conh,olling glutamine and asparagine biosynthesis in plants. Plant J 1: 275-280

Mei B, Zalkin H (1989) A cysteine-histidine-aspartate catalytic triad is involved in glutamine amide transfer function in purF-type glutamine amidotransferases. J Biol Chem 264 16613-1 6619

Milman HA, Cooney DA (1979) Partial purification and properties of L-asparagine synthetase from mouse pancreas. Biochem J 181:

Mosinger E, Batschauer A, Schafer E, Ape1 K (1985) Phytochrome control of in vitro transcription of specific genes in isohted nuclei from barley (Hordeum vulgare). Eur J Biochem 147: 137-142

Nakamura M, Yamada M, Hirota Y, Sugimoto K, Oka A, Takanami M (1981) Nucleotide sequence of the asnA gene coding for asparagine synthetase of E. coli K-12. Nucleic Acids Res 9 41569-4676

Otto V, Schafer E, Nagatani A, Yamamoto KT, Furuya M (1984) Phytochrome control of its own synthesis in Pisum satiuum. Plant Cell Physiol25 1579-1584

Peterman TK, Goodman HM (1991) The glutamine synthetase gene family of Arabidopsis thaliana: light-regulation and differential

1940-1948

3509-3513

pp 569-607

51-59




expression in leaves, roots and seeds. Mol Gen Genet 230

Pfeiffer NE, Mehlhaff PM, Wylie DE, Schuster SM (1986) Mono- clonal antibodies specific for bovine pancreatic asparagine synthetase. J Biol Chem 261: 1914-1919

Pfeiffer NE, Mehlhaff PM, Wylie DE, Schuster SM (1987) Topo- graphical separation of the catalytic sites of asparagine synthetase explored with monoclonal antibodies. J Biol Chem 262

Quail PH (1991) Phytochrome: a light-activated molecular switch that regulates plant gene expression. Annu Rev Genet 2 5 389-409

Richards NGJ, Schuster SM (1992) An alternative mechanism for the nitrogen transfer reaction in asparagine synthetase. FEBS Lett

Romero LC, Lam E (1993) Guanine nucleotide binding protein involvement in early steps of phytochrome-regulated gene expression. Proc Natl Acad Sci USA 90: 1465-1469

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual, Ed 2. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Schindler U, Menkens AE, Ahmad M, Ecker JR, Cashmore AR (1992) Heterodimerization between light-regulated and ubiqui- tously expressed Arabidopsis GBF-like bZip proteins. EMBO J 11:

Scofield MA, Lewis WS, Schuster SM (1990) Nucleotide sequence of Escherichia coli asnB and deduced amino acid sequence of asparagine synthetase B. J Biol Chem 265 12895-12902

Scott DB, Farnden KJF, Robertson JG (1976) Ammonia assimilation in lupin nudules. Nature 263: 703-705

Sheen J (1990) Metabolic repression of transcription in higher plants. Plant Cell 2 1027-1038

Sieciechowicz KA, Joy KW, Ireland RJ (1988) The metabolism of asparagine in plants. Phytochemistry 27: 663-671

145-154

11565-11570

313 98-102

1261-1273

Streeter JM (1977) Asparaginase and asparagine transaminase in soybean leaves and root nodules. Plant Physiol60 235-239

Stulen I, Oaks A (1977) Asparagine synthetase in com roots. Plant Physiol60: 680-683

Tingey SV, Tsai F-Y, Edwards JW, Walker EL, Coruzzi GM (1988) Chloroplast and cytosolic glutamine synthetase are encoded by homologous nuclear genes which are differentially expressed in vivo. J Biol Chem 263 9651-9657

Tsai F-Y, Coruzzi GM (1990) Dark-induced and organ-specific expression of two asparagine synthetase genes in Pisum sativum.

Tsai F-Y, Coruzzi GM (1991) Light represses the transcription of asparagine synthetase genes in photosynthetic and non-photosynthetic organs of plants. Mol Cell Biol 11: 4966-4972

Tso JY, Hermodsen MA, Zalkin H (1982) Glutamine phosphoribo- sylpyrophosphate amidotransferase from cloned Escherichia coli purF NH2-terminal amino acid sequence, identification of the glutamine site, and trace metal analysis. J Biol Chem 257:

Urquhart AA, Joy KW (1981) Use of phloem exudate technique in the study of amino acid transport in pea plants. Plant Physiol 6 8

Van Heeke G, Schuster SM (1989a) Expression of human asparagine synthetase in Escherichia coli. J Biol Chem 264 5503-5509

Van Heeke G, Schuster SM (1989b) The N-terminal cysteine of human asparagine synthetase is essential for glutamine-dependent activity. J Biol Chem 264 19475-19477

Vincentz M, Moureaux T, Leydecker M-T, Vaucheret H, Caboche M (1993) Regulation of nitrate and nitrite reductase expression in Nicotiana plumbaginifolia leaves by nitrogen and carbon metabolites. Plant J 3 315-324

Zalkin H (1993) The amidotransferases. Adv Enzymol Relat Area Mol Biol 6 6 203-309

EMBO J 9 323-332

3532-3536

750-754



Documents

Metabolic Regulation of the Gene Encoding Glutamine ... · The pBluescript SKI1 plasmids (Strategene) containing the cDNA clones were excised from the XZapII vector, and double-stranded