Identification, characterization and comparative analysis of a novel chorismate mutase gene in Arabidopsis thaliana

Gene 240 (1999) 115–123 www.elsevier.com/locate/gene

Identification, characterization and comparative analysis of a novel chorismate mutase gene in Arabidopsis thaliana Evelyn M. Mobley a, Barbara N. Kunkel c, Brian Keith b, * a Department of Biochemistry and Molecular Biology, University of Chicago,Chicago, IL, USA b Department of Medicine, University of Chicago, Chicago, IL 60637, USA c Department of Biology, Washington University, St. Louis, MO, USA Received 16 April 1999; received in revised form 22 August 1999; accepted 10 September 1999 Received by W. Martin

Abstract Phenylalanine, tyrosine, and tryptophan have a dual biosynthetic role in plants; they are required for protein synthesis and are also precursors to a number of aromatic secondary metabolites critical to normal development and stress responses. Whereas much has been learned in recent years about the genetic control of tryptophan biosynthesis in Arabidopsis and other plants, relatively little is known about the genetic regulation of phenylalanine and tyrosine synthesis. We have isolated, characterized and determined the expression of Arabidopsis thaliana genes encoding chorismate mutase, the enzyme catalyzing the first committed step in phenylalanine and tyrosine synthesis. Three independent Arabidopsis chorismate mutase cDNAs were isolated by functional complementation of a Saccharomyces cerevisiae mutation. Two of these cDNAs have been reported independently (Eberhard et al., 1993. FEBS 334, 233–236; Eberhard et al., 1996. Plant J. 10, 815–821), but the third (designated CM-3) represents a novel gene. The different organ-specific expression patterns of these cDNAs, their regulation in response to pathogen infiltration, as well as the different enzymatic characteristics of the proteins they encode are also described. Together, these data suggest that each isoform may play a distinct physiological role in coordinating chorismate mutase activity with developmental and environmental signals. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Gene families; Metabolic regulation; Phenylalanine; Tyrosine

1. Introduction The aromatic amino acids phenylalanine, tyrosine, and tryptophan have a dual biosynthetic role in plants; they are required for protein synthesis and are also precursors to a number of aromatic secondary metabolites, including the plant growth regulator auxin (indole 3-acetic acid), antimicrobial alkaloids, UV-absorbing flavonoids, and polyphenolic compounds such as lignin, which is required for structural support of mature xylem Abbreviations: ARO7, yeast chorismate mutase; cDNA, complementary DNA; CM, Arabidopsis chorismate mutase; GAL1, galactokinase; HIS3, imidazoleglycerolphosphate dehydratase; Mata, yeast mating type a; PCR, polymerase chain reaction; PGK, phosphoglycerate kinase; UV, ultraviolet. * Corresponding author. Present address: Department of Medicine, Abramson Cancer Research Institute, University of Pennsylvania, 477 BRBII/III, 421 Curie Boulevard, Philadelphia, PA 19104, USA. Tel.: +1-215-746-5533. E-mail address: [email protected] (B. Keith)

(Lewis and Yamamoto, 1990; Chapple et al., 1994; Radwanski and Last, 1995; Schmid and Amrhein, 1995; Whetten and Sederoff, 1995). Unlike primary metabolites, secondary metabolites are not considered necessary for cell survival per se, but are essential to the normal development of the multicellular plant. Some of these compounds are produced in response to environmental signals as part of a general defense response. A large series of experiments has demonstrated that increased production of aromatic secondary metabolites in irradiated, wounded or infected tissue correlates to increased activity of enzymes in both aromatic amino acid and secondary metabolic pathways (Dyer et al., 1989; Lamb et al., 1989; McCue and Conn, 1989; Keith et al., 1991; Henstrand et al., 1992; Muday and Herrmann, 1992; Bohlmann and Eilert, 1994). Moreover, this increase reflects the induced expression of genes encoding these biosynthetic enzymes, suggesting that aromatic amino acid biosynthesis and secondary metabolite production are coordinately regulated at the

0378-1119/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S0 3 7 8 -1 1 1 9 ( 9 9 ) 0 0 42 3 - 0

116

E.M. Mobley et al. / Gene 240 (1999) 115–123

genetic level to meet complex and varying requirements for aromatic compounds in plants. All three aromatic amino acids are synthesized from chorismic acid, the last common intermediate of the shikimate biosynthetic pathway (Fig. 1). Chorismate is converted either into anthranilic acid by anthranilate synthase, the first step in tryptophan biosynthesis, or into prephenate by chorismate mutase, the committed step in phenylalanine and tyrosine biosynthesis. Flow through the shikimate pathway accounts for up to 20% of a plant’s photosynthetically fixed carbon (Herrmann, 1995), most of which is shuttled through phenylalanine and tyrosine to generate abundant phenylpropanoid secondary metabolites (Haslam, 1993; Kutchan, 1998). The complex demand for aromatic secondary metabolites in specific cell types, and in response to multiple environmental stimuli, suggests that the regulation of phenylalanine and tyrosine biosynthesis in plants may differ fundamentally from that observed in microorganisms. Whereas much has been learned in recent years about the genetic control of tryptophan biosynthesis in Arabidopsis and other plants, relatively little is known about the genetic regulation of phenylalanine and tyrosine synthesis. Biochemical analyses in several plant species have typically revealed two separable isoenzymes, one localized to plastids and the other apparently cytosolic (Herrmann, 1995; Radwanski and Last, 1995; Schmid and Amrhein, 1995). To begin investigating the genetic regulation of phenylpropanoid metabolism and phenylalanine and tyrosine synthesis, we have isolated, characterized and determined the expression of genes encoding chorismate mutase in the flowering plant Arabidopsis thaliana. We describe here the isolation of three independent Arabidopsis chorismate mutase cDNAs by functional

complementation of a Saccharomyces cerevisiae mutation. Two of these cDNAs have been reported independently ( Eberhard et al., 1993, 1996b), but the third represents a novel gene. The different organ-specific expression patterns of these cDNAs, their regulation in response to pathogen infiltration, as well as the different enzymatic characteristics of the proteins they encode are also described. Together, these data suggest that each isoform may play a distinct physiological role in coordinating chorismate mutase activity with developmental and environmental signals.

2. Materials and methods 2.1. Chorismate mutase null yeast strain YEMCM13 (Mata ARO7::HIS3 leu2-d1, lys2-801am, ade2-101oc, trp1-d63, ura3-52) was created by direct gene deletion of ARO7 from parental strain YPH499 (Mata histidine3-d200 leu2-d1, lys2-801am, ade2-101oc, trp1-d63, ura3-52) (Baudin et al., 1993). The HIS3 gene was PCR amplified using primers containing 5∞ and 3∞ flanking sequences from the published ARO7 gene (Schmidheini et al., 1989). The 5∞ primer sequence was: 5∞-CAAATAGCACTCAGCATCCTGCATAAAATTGGTATAAGATGATCCTGCCTCGGTAATG-3∞ and the 3∞ primer sequence was 5∞-ACTCGGCAATGTGGAATTGTTACCGTGATAGCCTTCATGCTCGT TCAGAATGACACG-3∞. This linear DNA product was integrated into the yeast genome according to Gietz et al. (1992). Replacement of the ARO7 gene by HIS3 sequences in the resulting His+, Aro− YEMCM13 strain was confirmed by genomic DNA blot analysis and chorismate mutase activity assays. 2.2. Yeast extracts Yeast cells were pelleted, washed once with water and twice with grinding buffer (50 mM Tris–Cl (pH 8.0), 1 mM benzamidine, 1 mM EDTA and 1 mM PMSF ). An equal volume of 425–600 mm glass beads were added to the mixture and vortexed until approx. 80% of the cells were lysed. The mixture was centrifuged at 12 000×g for 20 min. The supernatant was collected and desalted on Sephadex G-25 Nap 5 (Pharmacia) columns equilibrated with grinding buffer. Samples were assayed for protein concentration (Bradford, 1976) and stored at −70°C.

Fig. 1. Aromatic amino acid biosynthetic pathway, showing position of chorismate mutase (CM ) in regulating phenylalanine and tyrosine biosynthesis. The conversion of chorismate to anthranilate by anthranilate synthase (AS ) in the biosynthesis of tryptophan ( W ) is also shown.

2.3. Chorismate mutase assays Chorismate mutase activity was assayed as described (Gilchrist and Connelly, 1987). Briefly, 100 ml of yeast extracts were added to 150 ml of 1.0 mM chorismic acid

E.M. Mobley et al. / Gene 240 (1999) 115–123

117

(barium salt; Sigma No. C1259) in 50 mM Tris–Cl (pH 8.0) and incubated at 30°C. Conversion of chorismate to prephenate was determined by nonenzymatic conversion of prephenate to phenylpyruvate by addition of 250 ml 0.2 N HCl for 10 min at room temperature. Following neutralization with 0.5 ml 3.2 N NaOH, phenylpyruvate was measured by absorbance at 320 nm (E=17 500 M−1 cm−1) in either a Perkin Elmer Lambda 3-A or a Gilford 2400 spectrophotometer. Apparent K values were determined by measuring chorismate m mutase activity in extracts of yeast strains expressing CM-1, CM-2 or CM-3 as a function of substrate concentration. Briefly, enzymatic rates were determined at fixed chorismate concentrations (0.05, 0.1, 0.25, 0.5, 1.0, 2.0, 3.0, and 5.0 mM chorismate for CM-2 and CM-3; 0.05, 0.25, 1.0, 2.0, 3.0 and 5.0 mM chorismate for CM-1), and K values obtained by linear regression using double m reciprocal Lineweaver–Burke plots.

approx. 2 weeks old as described (Ausubel et al., 1990). Both DNA and RNA blot hybridization conditions were as follows: prehybridization and hybridization were at 65°C in hybridization buffer (7% SDS, 1% BSA, 1 mM EDTA, 250 mM NaPO , pH 7.2) (Church and Gilbert, 4 1984). After hybridization, filters were washed four times for 20 min with (0.2×SSC, 0.1% SDS) at 55°C before exposure. For wounding experiments, leaves from two independent 3-week-old plants were cut with scissors and maintained in a moist chamber for the indicated time prior to RNA isolation. Arabidopsis plants were vacuum-infiltrated with Pseudomonas syringae pv. tomato strain DC3000 expressing the avrRpt2 gene according to Boch et al. (1998). Control plants were mock-infiltrated with 10 mM MgCl . 2

2.4. Arabidopsis cDNA libraries

3.1. Isolation of Arabidopsis chorismate mutase cDNA clones

The PFL61 plasmid library (a kind gift from F. Lacroute) contains Arabidopsis cDNAs expressed under the control of the constitutive PGK promoter (Minet et al., 1992). The l YES plasmid library (kind gift from S. Elledge, Baylor College of Medicine) contains Arabidopsis cDNAs that are regulated by the yeast GAL1 promoter. cDNAs are expressed in transformed yeast cells grown on galactose medium, and repressed in cells grown on glucose medium ( Elledge et al., 1991). 2.5. Yeast complementation The PFL61 cDNA library (Minet et al., 1992) was transformed into either YBK10 (ura3-52, aro7, ade2, his3, lys1, cdc6) or YEMCM13 (Mata ARO7::HIS3 leu2-d1, lys2-801am, ade2-101oc, trp1-d63, ura3-52) as described (Gietz et al., 1992). Transformants were selected on synthetic complete medium lacking uracil ( Ura−), and subsequently replica plated onto SC, Ura− medium lacking phenylalanine and tyrosine ( Ura−, Aro−) to select for functional complementation of the Aro7 mutations. Transformants from the lYES cDNA library were selected by plating on SC, Ura− medium and replica plated onto SC, Ura− medium containing galactose in order to induce expression of Arabidopsis cDNAs from the GAL1 promoter. These plates were subsequently replica plated onto SC, Ura−, Aro− medium containing galactose to select for complementation of the Aro7 mutations. 2.6. DNA and RNA hybridization DNA and RNA blot analysis were performed using standard protocols as described (Ausubel et al., 1990). Arabidopsis DNA was purified using a Puregene kit. RNA was extracted from Columbia ecotype plants of

3. Results

A S. cerevisiae strain containing a non-revertible aro7 mutation was generated as a first step toward isolating Arabidopsis chorismate mutase cDNAs. The single-copy S. cerevisiae chorismate mutase gene ARO7 (Ball et al., 1986) was replaced with the HIS3 gene in strain YPH499 as described in Materials and methods, creating YEMCM13. Replacement of the ARO7 coding region with HIS3 sequences was confirmed by hybridizing genomic DNA from YPH499 and YEMCM13 on Southern blots to gene-specific probes (data not shown). As expected, the YEMCM13 strain is incapable of growth on synthetic complete media lacking phenylalanine and tyrosine (SC, Ura−, Aro−). Extracts from YPH499 and YEMCM13 strains were assayed directly for chorismate mutase activity. YPH499 contained measurable chorismate mutase activity (0.8 nmol/mg protein/min) comparable to previously reported values (1.5 nmol/mg protein/min) (Schmidheini et al., 1989). Chorismate mutase activity in these extracts was increased to 19.6 nmol/mg protein/min by the addition of 0.5 mM tryptophan, which is also consistent with previously reported values for allosteric activation of yeast chorismate mutase by tryptophan (15 nmol/ mg/min) (Schmidheini et al., 1989). YEMCM13 extracts had no detectable chorismate mutase activity in the absence or presence of tryptophan. To identify Arabidopsis chorismate mutase cDNA clones, we used two independently constructed Arabidopsis cDNA expression libraries to complement the aro7 mutation in YEMCM13, as well as in another chorismate mutase-deficient strain, YBK10. The PFL61 library, in which Arabidopsis cDNAs are expressed from the yeast phosphoglycerate kinase (PGK ) promoter (Minet et al., 1992), was transformed into both strains and cells plated on complete medium lacking phenylala-

118

E.M. Mobley et al. / Gene 240 (1999) 115–123

Fig.2. Southern blot analysis of Arabidopsis genomic DNA digested with BamHI (B), EcoRI (R) or HindIII (H ), and hybridized to CM1, CM-2, and CM-3 cDNA probes.

nine and tyrosine. Plasmids were recovered from a total of 30 Aro+ colonies, and each retested for its ability to complement YEMCM13. In a similar fashion, five cDNA clones were obtained from the l YES Arabidopsis cDNA library, in which cDNAs are expressed under the control of the inducible GAL1 promoter ( Elledge et al., 1991). A combination of sequencing, dot blot analysis, restriction endonuclease digestion and DNA hybridization analysis indicated that the 35 cDNAs represent three genes, denoted CM-1, CM-2 and CM-3. Of the 35 cDNAs identified, 10 were derived from CM-1, 23 from CM-2 and two from CM-3. Sequence analysis revealed that CM-1 and CM-2 were identical to the independently reported sequences for CM-1 and CM-2 ( Eberhard et al., 1993, 1996b), whereas CM-3 is a unique chorismate mutase gene (GenBank accession No. AF131219). Analysis of the Arabidopsis expressed-sequence-tag database revealed a single EST (GenBank accession No. AA404859) corresponding to CM-3. Conceptual translation of the cDNA sequences indicates that CM-1 encodes a peptide of 344 amino acids, whereas CM-2 encodes a peptide of 265 amino acids and CM-3 encodes a peptide of 315 amino acids. Analysis of Arabidopsis genomic DNA was consistent with the presence of three singlecopy genes (Fig. 2). Amino acid alignment of the three Arabidopsis chorismate mutase protein sequences is shown in Fig. 3 along with the consensus sequences indicating conserved amino acids. As shown in Table 1, CM- 1 is 53.0 and 68.0% similar to CM-2 and CM-3, respectively. CM-1 and CM-3 display relatively unrelated amino-terminal Table 1 Similarity of CM-1, CM-2 and CM-3 amino acid sequences

CM-1 CM-2

CM-2

CM-3

51.2%

67.7% 52.2%

extensions of 76 and 65 amino acids, respectively. Similar extensions are observed in many plant amino acid biosynthetic proteins, and have been shown in many cases to function as transit peptides mediating uptake into plastids (Radwanski et al., 1995; Schmid and Amrhein, 1995; Bischoff et al., 1996). In contrast to CM-1 and CM-3, CM-2 does not appear to contain a putative plastid transit peptide, consistent with previous reports ( Eberhard et al., 1996a, b). A comprehensive survey of available prokaryotic and eukaryotic chorismate mutase protein sequences indicates that CM-1, CM-2 and CM-3 are most highly related to a chorismate mutase cloned from tomato, and to the Aro7 protein from S. cerevisiae, whereas bacterial isoforms show much lower levels of similarity (data not shown). The alignment shown in Fig. 3 reveals extensive regions of sequence similarity in Arabidopsis, tomato and yeast. 3.2. Biochemical properties of CM-1, CM-2 and CM-3 To investigate the enzymatic activity of CM-1, CM-2 and CM-3 isoforms, we assayed chorismate mutase activity in extracts from yeast strains expressing Arabidopsis CM-1, CM-2 or CM-3 cDNAs ( Table 2). Both CM-1 and CM-3 enzyme activities were allosterically regulated by free aromatic amino acids: free tryptophan increased specific activity of both CM-1 and CM-3; however, phenylalanine and tyrosine inhibited both enzymes. In contrast, CM-2 activity was insensitive to free aromatic amino acids. Apparent K values for chorismate were determined m for CM-1, CM-2 and CM-3 expressed in yeast, and are shown in Table 3. The apparent K value observed for m CM-1 (2.9±0.76 mM ) is similar to values reported for CM-1 isoforms in extracts of tobacco leaves (1.7 mM ) and cell cultures (2.5 mM ) (Goers and Jensen, 1984). Similarly, the value measured for CM-2 (0.23±0.042 mM ) is close to reported CM-2 K values m of 0.24 mM in tobacco (Goers and Jensen, 1984), 0.202 mM in potato leaf ( Kuroki and Conn, 1989) and 0.33 mM in mungbean (Gilchrist et al., 1972). The Table 2 Regulation of CM-1, CM-2, and CM-3 enzymatic activity by free amino acidsa

No addition Phe Trp Tyr

CM-1

CM-2

CM-3

100 32.5 207 31.9

100 90.1 93.5 93.5

100 35.3±5.1 198±17.2 43.3±7.6

a Activity listed as percentage of chorismate mutase activity in untreated extracts from YEMCM13 expressing CM-1, CM-2 or CM-3 cDNAs. Values for CM-1 and CM-2 are averages of two experiments, whereas values for CM-3 were calculated from three experiments. Free amino acids listed were added to reactions to final concentration of 1.0 mM.

E.M. Mobley et al. / Gene 240 (1999) 115–123 Table 3 Chorismate K values for CM-1, CM-2 and CM-3 expressed in m YEMCM13 K (Mm) m CM-1 CM-2 CM-3

2.9±0.76 0.23± 0.042 0.42±0.10

apparent K value for the CM-3 isoform is m 0.41±0.098 mM, although the corresponding CM-3 K value from Arabidopsis or other plant extracts has m not yet been determined.

119

3.3. CM-1, CM-2 and CM-3 gene expression To investigate the expression of the three chorismate mutase genes in Arabidopsis plants, blot analysis was performed on RNA isolated from different organs of adult plants. CM-1, CM-2 and CM-3 show different tissue-specific expression patterns. CM-1 RNA was most highly expressed in floral organs, and less highly expressed in stems, roots, cauline leaves, siliques and rosettes leaves (Fig. 4). In contrast, CM-2 RNA was most abundant in root tissue, and progressively less abundant in secondary stems, cauline leaf and floral organs, rosette leaves, siliques and primary stems. CM-

Fig. 3. Amino acid alignment of chorismate mutase protein sequences. Arabidopsis CM-1, CM-2 and CM-3 mature protein sequences were aligned to those from tomato (Lypersicon esculentum) and Saccharomyces cerevisiae using the MegAlign feature of the DNAStar sequence analysis program.

120

E.M. Mobley et al. / Gene 240 (1999) 115–123

Fig. 4. Organ-specific expression pattern of Arabidopsis chorismate mutase genes. Blots containing total RNA from cauline leaves ( lanes 1, 8, 15), flowers ( lanes 2, 9, 16), roots ( lanes 3, 10, 17), rosette leaves ( lanes 4, 11, 18), siliques ( lanes 5, 12, 19), secondary inflorescence stems ( lanes 6, 13, 20) and primary inflorescence stems ( lanes 7, 14, 21) was hybridized to gene-specific probes from CM-1, CM-2, and CM-3. Lower panels show ethidium bromide staining of RNA gel.

3 RNA was most abundant in floral organs followed by root and secondary stems, old stems, cauline leaf, siliques and rosette leaves. To determine whether CM-1, CM-2 or CM-3 RNA levels are differentially regulated by environmental stress, RNA blot analysis was performed on tissues subjected to either wounding or pathogen infiltration. RNA was isolated from wounded leaves of two independent plants and hybridized to gene-specific probes from CM-1, CM-2 and CM-3. The data in Fig. 5 show that CM-1 RNA levels in both plants increased noticeably by approx. 1 h, decreased by 2.5 h after wounding, and then increased again by 4–6 h after wounding. In contrast, CM-2 RNA levels decreased after 1 h, but returned to starting levels in both plants by 2–3 h after wounding. CM-3 RNA levels were not noticeably altered by this treatment. Similar results were obtained upon exposure to the bacterial pathogen P. syringae. Pathogens expressing a given avirulence (avr) gene trigger a complex defense response in plants, including rapidly induced expression of many defense-related genes, in resistant plants carrying the corresponding resistance gene. As a result,

growth of the invading pathogen is restricted, and disease symptoms fail to develop in the plant. In contrast, if the infected plant lacks the specific resistance gene corresponding to the avirulence gene expressed in the pathogen, then the defense response is not activated, allowing significant proliferation of the pathogen. In this case, defense gene expression is induced only late in infection, probably in response to signals released from diseased or dead plant cells. Mutations in the Arabidopsis RPS2 gene render the plant incapable of recognizing P. syringae strains expressing the avrRpt2 avirulence gene. Infiltration of rps2 mutants with P. syringae harboring the avrRpt2 gene therefore fails to trigger the defense response, and results in disease progression ( Kunkel, 1996). To investigate whether CM-1, CM-2 or CM-3 RNAs are induced during interactions with avirulent pathogens, leaves of wild-type and rps2-201C mutant plants were infiltrated with P. syringae pv. tomato strain DC3000 expressing avrRpt2. A Northern blot sequentially hybridized with gene-specific probes from CM-1, CM-2 and CM-3 is shown in Fig. 6. CM-1 transcript levels were induced rapidly in wild-type leaves in response to the bacteria,

Fig. 5. Expression of Arabidopsis chorismate mutase genes in response to wounding. Blots containing total RNA extracted from rosette leaves at different times after physical wounding (see Materials and methods) were hybridized to gene-specific probes from CM-1, CM-2, and CM-3. (A), (B) Results obtained for two independent plants. Numbers indicated hours post-wounding. Lower panels show ethidium bromide staining of RNA gel.

E.M. Mobley et al. / Gene 240 (1999) 115–123

Fig. 6. Expression of Arabidopsis chorismate mutase genes in response to pathogenic infiltration. Leaves from wild-type or rps2-201C mutant plants were vacuum-infiltrated with 10 mM MgCl containing 2 106 CFU/ml of P. syringae pv. tomato strain DC3000 expressing the avrRpt2 gene (Boch et al., 1998). Control samples indicates wild-type leaves infiltrated with MgCl alone. Total RNA was subsequently 2 extracted at the indicated times (numbers in hours). The same RNA blot was hybridized successively to probes from CM-1, CM-2 and CM3. Lower panels show ethidium bromide staining of 28S RNA.

peaking approx. 12 h after infiltration ( Fig. 6). In contrast, CM-2 RNA levels remained flat, and CM-3 RNA levels decreased over this time course in wild-type leaves. Interestingly, CM-3 RNA levels increased late (24 h) in infiltrated rps2-201C leaves, whereas both CM-1 and CM-2 transcript levels show a transient, more modest increase 12 h after infiltration. The hybridization patterns for all three genes were qualitatively similar for RNAs from the mock-infiltrated control. The wounding and pathogen infiltration data indicate that the Arabidopsis chorismate mutase genes are differentially regulated in response to environmental stresses, and suggest that the full induction of CM-1 expression in wild-type leaves is part of a genetically regulated defense response.

4. Discussion The identification of a novel chorismate mutase isoform, encoded by CM-3, indicates that the regulation of chorismate mutase activity in Arabidopsis is more complex than previously thought. Interestingly, both CM-1 and CM-3 proteins contain putative plastid transit peptide sequences, suggesting that these isoforms are targeted to plastids, whereas CM-2 appears to be cytosolic ( Eberhard et al., 1996b). In order to characterize the activities of the gene products encoded by CM-1, CM-2 and the newly identified CM-3, we assayed chorismate mutase activity in

121

extracts from yeast cells expressing the Arabidopsis proteins. We showed that CM-3 is, like CM-1, allosterically regulated by free amino acids, whereas CM-2 activity is not. These results confirm and extend those from a previous report ( Eberhard et al., 1996b). In addition, we measured the apparent K values for all three m isoforms expressed in yeast. Given the putative plastid transit peptides on both CM-1 and CM-3, it is intriguing that two allosterically regulated isoforms with considerably different apparent K values (2.9 mM for CM-1, m and 0.41 mM for CM-3) may exist in the same cellular compartment. It should be noted that the K values m obtained from CM proteins expressed in yeast could be affected by the putative transit peptide sequences ( Wiersma et al., 1990); however, the similarity of our data to K values for CM-1 and CM-2 from plant m extracts suggests that this is not the case for these isoforms. We also cannot rule out the possibility that CM-1 or CM-3 may be targeted to non-plastid organelles, such as the mitochondria, but there is no direct evidence of chorismate mutase activity in cellular compartments other than the plastids and cytosol. Chloroplast uptake experiments using radiolabeled CM-1 and CM-3 proteins could be used to address this question in the future. It will be particularly interesting to determine whether CM-3 and CM-1 isoforms are expressed in the same cell types. If not, each isoform may have a unique and essential role in the tissues in which it is specifically expressed. If, however, both are expressed in the same cell types, then the different apparent K values observed m for CM-3 and CM-1 may be informative about their physiological role. For example, it may be that CM-3 activity is sufficient to provide requisite levels of free phenylalanine and tyrosine under normal growing conditions, but that it becomes saturated when flux through the pathway is rapidly increased, perhaps in response to environmental stress. In this case, CM-1 activity could be regulated over a much wider range of substrate concentrations. These possibilities can be distinguished in the future by generating promoter-reporter gene constructs in transgenic plants, and by RNA in situ hybridization analysis. We observed differential expression of CM-1, CM-2 and CM-3 genes in all tissue types examined, in keeping with results observed for other Arabidopsis aromatic amino acid biosynthetic genes (Radwanski et al., 1995; Eberhard et al., 1996a; Pruitt and Last, 1993). Differential expression was also observed in response to environmental stresses. CM-1 RNA levels are increased in response to wounding and pathogenic attack, suggesting that CM-1 may have a particularly important role in supplying phenylalanine and tyrosine for stressrelated changes in metabolism ( Eberhard et al., 1996b). The biphasic nature of CM-1 RNA induction in response to wounding (Fig. 5) also suggests that CM-1 expression

122

E.M. Mobley et al. / Gene 240 (1999) 115–123

is induced by temporally distinct signals following wounding. The nature of these signals, and their importance to successful wound responses is unknown, but CM-1 RNA induction could be a useful marker in their identification. The significance of the transient decrease in CM-2 RNA levels in wounded leaf tissue, as well as the specific role of CM-3 in these responses, is not clear at present. We also provide evidence that the rapid and more pronounced induction of CM-1 RNA in wild-type plants, in comparison to rps2-201C mutants, is part of a genetically determined plant defense response regulated by plant resistance genes and corresponding pathogen avirulence genes. In contrast, the later induction of CM-3 expression in the rps2-201C mutant may be a response to signals generated by progressing disease. Similar patterns of CM-1 and CM-2 RNA induction by fungal pathogens has been previously reported ( Eberhard et al., 1996b), adding credence to the idea that CM-1 may have a primary role in rapid responses to environmental stimuli that affect aromatic amino acid metabolism. Interestingly, the expression patterns of all three genes is different in wild-type and rps2-201C mutant leaves, suggesting that specific metabolic requirements may be different in these different genetic backgrounds. Taken together, the different properties of CM-1, CM-2 and CM-3 (apparent K values, regulation by m free aromatic amino acids, distinct tissue-specific expression patterns and wounding and pathogen induction) suggest that each isoform fulfills distinct physiological roles in specific tissues under different environmental conditions. To clearly define these roles, the isolation of loss-of-function mutations in each chorismate mutase gene, and construction of double and triple mutants, will be required. There is precedence for obtaining mutations in specific members of multigene families in Arabidopsis: for example, mutations in one of two Arabidopsis tryptophan synthase b genes (TSB2) confer tryptophan auxotrophy under high light, apparently due to the inability of the less highly expressed TSB1 gene to compensate for loss of TSB2 function (Barczak et al., 1995). The phenotype of mutations in any of the Arabidopsis chorismate mutase genes is not obvious a priori, but a screen based on tissue-specific expression data could be performed. For example, as CM-2 is the most highly expressed chorismate mutase RNA in roots, loss of function mutations in CM-2 could produce a perceptible loss of overall enzyme activity in roots. The downstream effects of such mutations on phenylpropanoid biosynthesis and stress responses in Arabidopsis will be of particular interest.

Acknowledgements The authors thank Eva Rosen and Amy Matsumura for experimental input and assistance, and Drs. Celeste

Simon and Laurens Mets for critical review of the manuscript. We also thank Dr. Elizabeth McNally for access to graphics equipment. This work was supported in part by NIH grant R01-GM50408 to B. Keith.

References Ausubel, F.M., Brent, R., Kingston, R.E.Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K. ( Eds.), Current Protocols in Molecular Biology 1990. Greene Publishing Associates/Wiley-Interscience, New York. Ball, S.G., Wickner, R.B., Cottarel, G., Schaus, M., Tirtiaux, C., 1986. Molecular cloning and characterization of ARO7-OSM2, a single yeast gene necessary for chorismate mutase activity and growth in hypertonic medium. Mol. Gen. Genet. 205, 326–330. Barczak, A.J., Zhao, J., Pruitt, H.D., Last, R.L., 1995. 5-Fluoroindole resistance identifies tryptophan synthase beta subunit mutants in Arabidopsis thaliana. Genetics 140, 303–313. Baudin, A., Ozier-Kalogeropoulos, O., Denouel, A., Lacroute, F., Cullin, C., 1993. A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae. Nucleic Acid Res. 21, 3329–3330. Bischoff, M., Rosler, J., Raesecke, H.R., Gorlach, J., Amrhein, N., Schmid, J., 1996. Cloning of a cDNA encoding a 3-dehydroquinate synthase from a higher plant, analysis of the organ-specific and elicitor-induced expression of the corresponding gene. Plant Mol. Biol. 31, 69–76. Boch, J., Verbsky, M.L., Roberston, T.L., Larkin, J.C., Kunkel, B.N., 1998. Analysis of resistance gene-mediated defense responses in Arabidopsis thaliana plants carrying a mutation in CPR5. Mol. Plant–Microbe Interact. 12, 1196–1206. Bohlmann, J., Eilert, U., 1994. Elicitor induced secondary metabolism in Ruta gravelens L. Role of chorismate utilizing enzymes. Plant Cell Tiss. Organ Culture 38, 189–198. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal. Biochem. 72, 248–254. Chapple, C.C.S.Shirley, B.W., Zook, M., Hammerschmidt, R., Somerville, S.C. ( Eds.), Secondary Metabolism in Arabidopsis 1994. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Church, G., Gilbert, W., 1984. Genomic sequencing. Proc. Natl. Acad. Sci. USA 81, 1991–1995. Dyer, W.E., Henstrand, J.M., Handa, A.K., Herrmann, K.M., 1989. Wounding induces the first enzyme of the shikimate pathway in Solanaceae. Proc. Natl. Acad. Sci. USA 86, 7370–7373. Eberhard, J., Bischoff, M., Raesecke, H.R., Amrhein, N., Schmid, J., 1996a. Isolation of a cDNA from tomato coding for an unregulated, cytosolic chorismate mutase. Plant Mol. Biol. 31, 917–922. Eberhard, J., Ehrler, T.T., Epple, P., Felix, G., Raesecke, H.R., Amrhein, N., Schmid, J., 1996b. Cytosolic and plastidic chorismate mutase isozymes from Arabidopsis thaliana: molecular characterization and enzymatic properties. Plant J. 10, 815–821. Eberhard, J., Raesecke, H.R., Schmid, J., Amrhein, N., 1993. Cloning and expression in yeast of a higher plant chorismate mutase. FEBS 334, 233–236. Elledge, S.J., Mulligan, J.T., Ramer, S.W., Spottswood, M., Davis, R., 1991. lYES: A multifunctional cDNA expression vector for the isolation of genes by complementation of yeast and Escherichia coli mutations. Proc. Natl. Acad. Sci. USA 88, 1731–1735. Gietz, D., St. Jean, A., Woods, R.A., Schiestl, R.H., 1992. Improved method for high efficiency transformations of intact yeast cells. Nucleic Acid Res. 20, 1425. Gilchrist, D.G., Woodin, T.S., Johnson, M.L., Kosuge, T., 1972.

E.M. Mobley et al. / Gene 240 (1999) 115–123 Regulation of aromatic amino acid biosynthesis in higher plants. Plant Physiol. 49, 52–57. Gilchrist, D.G., Connelly, J.A., 1987. Methods Enzymol. 142, 450–463. Goers, S.K., Jensen, R.A., 1984. Separation and characterization of two chorismate-mutase isoenzymes from Nicotiana silvestris. Planta 162, 109–116. Haslam, E., 1993. Shikimic Acid: Metabolism and Metabolites. John Wiley, Chichester. Henstrand, J.M., McCue, K.F., Brink, K., Handa, A.K., Herrmann, K.M., Conn, E.E., 1992. Light and fungal elicitor induces 3-deoxy-arabino-heptulosonate 7-phosphate synthase mRNA in suspension cultured cells of parsley (Petroselium crispum L.). Plant Physiol. 98, 761–763. Herrmann, K.M., 1995. The shikimate pathway — early steps in the biosynthesis of aromatic compounds. Plant Cell 7, 907–919. Keith, B., Dong, X., Ausubel, F.M., Fink, G.R., 1991. Differential induction of 3-deoxy--arabino-heptulosonate 7-phosphate synthase genes in Arabidopsis thaliana by wounding and pathogenic attack. Proc. Natl. Acad. Sci. USA 88, 8821–8825. Kunkel, B.N., 1996. A useful weed put to work: genetic analysis of disease resistance in Arabidopsis thaliana. Trends Genet. 12, 63–69. Kuroki, G.W., Conn, E.E., 1989. Differential activities of chorismate mutase isozymes in tubers and leaves of Solanum tuberosum L. Plant Physiol. 89, 472–476. Kutchan, T.M., 1998. Molecular Genetics of Plant Alkaloid Biosynthesis. Academic Press, San Diego. Lamb, C.J., Lawton, M.A., Dron, M., Dixon, R.A., 1989. Signals and transduction mechanisms for activation of plant defenses against microbial attack. Cell 56, 215–224. Lewis, N.G., Yamamoto, E., 1990. Lignin: occurrence, biogenesis and

123

biodegradation. Annu. Rev. Plant Phys. Plant Mol. Biol. 41, 455–496. McCue, K.F., Conn, E.E., 1989. Induction of 3-deoxy--arabino-heptulosonate-7-phosphate synthase activity by fungal elicitor in cultures of Petroslinum crispum. Proc. Natl. Acad. Sci. USA 86, 7374–7377. Minet, M., Dufour, M.-E., Lacroute, F., 1992. Complementation of Saccharomyces cerevisiae auxotrophic mutants by Arabidopsis thaliana cDNAs. Plant J. 2, 417–422. Muday, G.K., Herrmann, K.M., 1992. Wounding induces one of two isoenzymes of 3-deoxy--arabino-heptulosonate 7-phosphate synthase in Solanum tuberosum L. Plant Physiol. 98, 496–500. Pruitt, K.D., Last, R.L., 1993. Expression patterns of duplicate tryptophan synthase B genes in Arabidopsis thaliana. Plant Physiol. 102, 1019–1026. Radwanski, E.R., Last, R.L., 1995. Tryptophan biosynthesis and metabolism: biochemical and molecular genetics. Plant Cell 7, 921–934. Radwanski, E.R., Zhao, J., Last, R.L., 1995. Arabidopsis thaliana synthase alpha: gene cloning and expression analysis. Mol. Gen. Genet. 248, 657–667. Schmid, J., Amrhein, N., 1995. Molecular organization of the shikimate pathway in higher plants. Phytochemistry 39, 737–749. Schmidheini, T., Sperisen, P., Paravicini, G., Hutter, R., Braus, G., 1989. A single point mutation results in a constitutively activated and feedback-resistant chorismate mutase of Saccharomyces cerevisiae. J. Bacteriol. 171, 1245–1253. Whetten, R., Sederoff, R., 1995. Lignin biosynthesis. Plant Cell 7, 1001–1013. Wiersma, P.A., Hachey, J.E., Crosby, W.L., Moloney, M.M., 1990. Specific truncations of an acetolactate synthase gene from Brassica napus efficiently complement ILVB/ILVG mutants of Salmonella typhimurium. Mol. Gen. Genet. 224, 155–159.