Effects of phosphate limitation on expression of genes involved in pyrimidine synthesis and salvaging in Arabidopsis

Plant Physiology and Biochemistry 43 (2005) 91–99 www.elsevier.com/locate/plaphy

Original article

Effects of phosphate limitation on expression of genes involved in pyrimidine synthesis and salvaging in Arabidopsis Matthew M. Hewitt, Jessica M. Carr, Cynthia L. Williamson, Robert D. Slocum * Department of Biological Sciences, Goucher College, Baltimore, MD 21204-2794, USA Received 17 August 2004; accepted 3 January 2005 Available online 24 February 2005

Abstract Arabidopsis seedlings grown for 14 d without phosphate (P) exhibited stunted growth and other visible symptoms associated with P deficiency. RNA contents in shoots decreased nearly 90%, relative to controls. In shoots, expression of Pht1;2, encoding an inducible highaffinity phosphate transporter, increased threefold, compared with controls, and served as a molecular marker for P limitation. Transcript levels for five enzymes (aspartate transcarbamoylase, ATCase, EC 2.1.3.2; carbamoyl phosphate synthetase, CPSase, EC 6.3.5.5); UMP synthase, EC 2.4.1.10, EC 4.1.1.23; uracil phosphoribosyltransferase, UPRTase, EC 2.4.2.9; UMP kinase, EC 2.7.1.14) increased 2–10-fold in response to P starvation in shoots. These enzymes, which utilize phosphorylated intermediates at putative regulated steps in de novo synthesis and salvaging pathways leading to UMP and pyrimidine nucleotide formation, appear to be coordinately regulated, at the level of gene expression. This response may facilitate pyrimidine nucleotide synthesis under P limitation in this plant. Expression of P-dependent and P-independent phosphoribosyl pyrophosphate (PRPP) synthases (PRS2 and PRS3, respectively) which provide PRPP, the phosphoribosyl donor in UMP synthesis via both de novo and salvaging pathways, was differentially regulated in response to P limitation. PRS2 mRNA levels increased twofold in roots and shoots of P-starved plants, while PRS3 was constitutively-expressed. PRS3 may play a novel role in providing PRPP to cellular metabolism under low P availability. © 2005 Elsevier SAS. All rights reserved. Keywords: Arabidopsis; Gene expression; Phosphate starvation; Pyrimidine nucleotide biosynthesis

1. Introduction Phosphate availability is one of the major factors affecting plant growth and development and plants exhibit common morphological, physiological and biochemical adaptations to P limitation [34]. Morphological changes in response to P deficiency include slowing of overall growth, increased lateral root production [25,47], an increase in the length and density of root hairs [7], and accumulation of anthocyanins [33]. P limitation also results in an increased root to shoot

Abbreviations: ATCase, aspartate transcarbamoylase; CarAsp, carbamoyl aspartate; CPSase, carbamoyl phosphate synthetase; DHO, dihydroorotate; FOA, 5-fluoroorotic acid; OMP, orotate-5’-monophosphate; P, phosphate; PALA, N-(phosphonacetyl)-L-aspartate; Pi, orthophosphate; PRPP, 5-phosphoribosyl-1-pyrophosphate; UMP, uridine-5’-monophosphate; UPRTase, uracil phosphoribosyltransferase.. * Corresponding author. Fax: +1 410 337 6508. E-mail address: [email protected] (R.D. Slocum). 0981-9428/$ - see front matter © 2005 Elsevier SAS. All rights reserved. doi:10.1016/j.plaphy.2005.01.003

biomass ratio [17] and increased Pi uptake by and retention within roots [33]. A large number of biochemical changes, which serve to increase acquisition and tissue availability of P, are also seen in response to P starvation. These include the production of intracellular and extracellular RNases [5,24,29] and phosphodiesterases [1], which release P from RNA and nucleotide pools, a variety of phosphatases [17], secretion of organic acids, which acidify the rhizosphere and maximize uptake of phosphorous as H2PO4– [17,22], and induction of numerous high-affinity phosphate transporters [23,39]. The synthesis of nucleotides has been reported to be strongly influenced by P availability (see [45] for review). The main P flux through nucleotide pools is toward nucleic acids, representing 45% of the total P in cultured plant cells [4]. In Catharanthus cell cultures, NTP pools decreased three to fivefold within 24 h after inoculation into P-deficient medium [3]. Decreased nucleotide pools associated with depleted cellular P in Datura cell cultures results in a pro-

92

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

longed cell division phase, during which most cellular P is consumed. After addition of P to these cultures, uptake was rapid (complete within 2 d of inoculation into fresh medium) and preceded nucleotide biosynthesis [49]. We are interested in the regulation of pyrimidine biosynthesis in plants. Pyrimidines are derived from UMP, which is formed either by de novo synthesis or the “salvaging” of preformed nucleobases or nucleosides ([19,40] see Fig. 1). In the de novo pathway, ATP is the phosphoryl donor in the synthesis of carbamoyl-P, which is utilized by ATCase in the committed step in the pathway. However, ATP is later resynthesized (step 3 of de novo pathway; Fig. 1) by capturing the energy from dihydroorotate oxidation to orotate via the mitochondrial flavoenzyme dihydroorotate dehydrogenase, transferring reducing equivalents to the proximal quinone and final molecular oxygen acceptor in the respiratory chain [44]. Other ATP-utilizing steps include the salvaging of uridine to UMP, via uridine kinase, and phosphorylation of UMP to UDP by UMP kinase, which is representative of numerous other NMP/NDP kinase-mediated transformations in which P is incorporated into pyrimidine nucleotides. The formation of phosphoribosyl pyrophosphate (PRPP), which donates the phosphoribosyl moiety in salvaging of uracil to UMP and in the synthesis of OMP by the orotate phosphoribosyltransferase (EC 2.4.2.10) activity of the bifunctional UMP synthase [35], also requires ATP. In these later PRPP-requiring steps, phosphate is directly incorporated into newly-formed nucleotides. It is known that P limitation results in markedly reduced cellular levels of ATP [45] and Ukaji and Ashihara [43] reported that Pi addition to

P-starved cells resulted in a 10-fold increase in PRPP levels. Thus, it is not surprising that P limitation would negatively impact pyrimidine nucleotide synthesis by limiting the availability of the essential phosphorylated intermediates ATP and PRPP. While metabolic studies have provided important insights into the regulation of pyrimidine synthesis under P limitation, little is known about the regulation key enzymes in these pathways, at the level of gene expression. Genes encoding ATCase (PyrB) and UMP synthase (PyrE-F), which catalyze the committed and rate-limiting steps of the de novo pathway, respectively, have been shown to be regulated in response to changes in pyrimidine availability [6,35] and we expected that they might similarly be regulated by P availability. Metabolic regulation of plant CPSase, which produces the carbamoyl-P intermediate used in both pyrimidine and arginine synthesis ([37]; see Fig. 1) has been characterized [30], but nothing is known regarding the expression of the CarA and CarB genes encoding the small (glutamine amidohydrolase [9]) and large (synthetase [46]) subunits of this enzyme. Similarly, expression of genes for UMP kinase, which catalyzes the committed step in the synthesis of pyrimidines from UMP [50], and the salvage enzyme UPRTase [39], does not appear to have been investigated. Recent studies of spinach PRPP synthases [20,21] have demonstrated that, in addition to two isoforms that display Pi dependence for activity (“class I”; PRS1, PRS2), two additional isoforms do not require Pi for activity (“class II”; PRS3, PRS4). The second class of enzymes might enable a plant experiencing limited P availability to continue synthesizing PRPP for nucleotide synthe-

Fig. 1. Synthesis of UMP in plants via the de novo pathway (steps 1–6) or salvaging pathways (steps 7, 8). Synthesis of UDP from UMP (step 9) is the committed step leading to the synthesis of other pyrimidine nucleotides. Phosphate is required for the synthesis of PRPP (step 10), the phosphoribosyl donor in UMP synthesis, and ATP, the phosphoryl donor in reactions catalyzed by NMP and NDP kinases and carbamoyl phosphate synthetase (step 1). Carbamoyl phosphate is a common intermediate in the de novo synthesis of UMP and arginine. Numbers indicate enzymes catalyzing each step. Circled numbers represent enzymes for which gene expression was investigated in the present study. 1, Carbamoyl phosphate synthetase (CPSase, EC 6.3.5.5); 2, aspartate transcarbamoylase (ATCase, EC 2.1.3.2); 3, dihydroorotase (DHOase, EC 3.5.2.3); 4, dihydroorotate dehydrogenase (EC 1.3.3.1); 5, 6, orotate phosphoribosyltransferase (OPRTase, EC 2.4.2.10) and OMP decarboxylase (EC 4.1.1.23) activities residing within the bifunctional UMP synthase; 7, uracil phosphoribosyltransferase (UPRTase, EC 2.4.2.9); 8, uridine/cytidine kinase (EC 2.7.1.48); 9, UMP kinase (EC 2.7.1.14); 10, PRPP synthase (EC 2.7.6.1).

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

93

2.2. Effects of phosphate starvation on seedling growth and development Seedling growth was normal on both complete and -P medium for the first 9 d, after which growth was arrested on -P medium (Fig. 2). After 14 d, -P seedlings exhibited an accumulation of anthocyanins, had a shorter primary root and an increased number of short lateral roots, and an increase in density and length of root hairs. An increase in the root to shoot biomass ratio, doubling from 0.27 in controls to 0.52 in the phosphate-starved seedlings, was also observed.

2.3. Tissue RNA contents and Pht1.2 expression as indicators of phosphate deficiency in seedlings

Fig. 2. Effects of phosphate starvation on growth and development of Arabidopsis seedlings. A, control plants; B, minus phosphate seedlings; C, backlighted image showing well-developed fibrous root system in control plants; D, increased production of short lateral roots in phosphate-deficient seedlings.

sis, to the extent that ATP, or other diphosphoryl donors are available. We investigated the effects of decreased pyrimidine availability, associated with severe P limitation, on expression of genes representing each of these enzymes in Arabidopsis. Our goals were to identify potentially-regulated steps and to characterize coordination of de novo and salvage pathway activities, which is poorly understood. 2. Results 2.1. Pathways for synthesis of UMP and pyrimidines Enzymes involved in the synthesis of UMP via the de novo pathway, or salvaging of preformed nucleobases or nucleosides, are shown in Fig. 1. Reactions utilizing ATP or PRPP are indicated.

Although Arabidopsis seedlings exhibited growth and developmental responses typical of P deficiency, we wanted to have biochemical and molecular markers for the relative P status of root and shoot tissues. We determined total RNA contents of seedlings as an indirect indicator of pyrimidine availability in control and -P seedlings. The RNA content of roots increased 41%, from 223 to 314 µg RNA g–1 fresh wt. In contrast, RNA contents decreased 87% in shoots of -P plants, from 93 to 12 µg RNA g–1 fresh wt. Smith et al. [39] reported that APT1 and APT2 (=Pht1;2 and Pht1.1, respectively [33]), two genes encoding nearly identical membrane orthophosphate transporters, were coordinately induced in root tissues in response to phosphate starvation in Arabidopsis. We used gene-specific primers (Table 1) to amplify the Pht1;2 transcript from total RNA of root and shoot tissues of control and -P seedlings, using a semiquantitative RT-PCR assay to profile its expression. As is seen in Fig. 3, Pht1;2 mRNA increased about threefold in shoots of seedlings grown for 14 d in phosphate-deficient medium, compared with controls. Shoot Pht1;2 transcript levels were approximately eight and fivefold lower, respectively, than those in roots of both control and -P plants, which did not differ significantly.

Table 1 Gene-specific primer sets used in RT-PCR expression analyses. Amplified products (bp) are indicated. The Arabidopsis Genome Initiative (AGI) locus and common gene names are given for each enzyme (see Fig. 1 legend)

94

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

2.5. Phosphate limitation and expression of genes encoding other enzymes utilizing phosphorylated intermediates in pyrimdine synthesis

Fig. 3. Semi-quantitative RT-PCR analysis of gene expression in root and shoot tissues of control and phosphate-starved (-P) seedlings. Abbreviations for enzyme names are as indicated in Fig. 1. PHT1;2 is an inducible orthophosphate transporter encoded by Pht1;2. The constitutively-expressed Act-2 isoform [16] served as the internal standard for normalization of expression data. Template amplification was linear: 26-cycles for Act-2, 28–32 cycles for other transcripts.

2.4. Effects of phosphate starvation on PyrB expression and ATCase protein and enzyme activity levels We previously reported that pyrimidine starvation in Arabidopsis seedlings treated with the ATCase inhibitor PALA resulted in markedly increased PyrB transcript levels [6]. In the present study, PyrB mRNA levels were twofold higher in both roots and shoots of -P seedlings, compared with control plants (Fig. 3). ATCase protein levels in -P seedlings increased very slightly (27%) over controls (Fig. 4) and ATCase enzyme activity was 36% higher than in tissues of control plants, increasing from 489 to 666 pkat carbamoyl aspartate mg–1 protein.

Fig. 4. SDS-PAGE immunoblot profile showing 35.8 kDa ATCase monomer protein levels in control (A) and -P seedlings (B). Relative amounts of each protein, determined by scanning densitometric analysis, are shown below each band.

Nucleotide sequences for genes encoding other enzymes of interest in this study were obtained from the NCBI GenBank repository (http://www.ncbi.nlm.nih.gov). As is shown in Table 1, gene-specific primer sets were used in RT-PCR analyses to examine relative transcript levels for each enzyme. Genes encoding the small (CarA) and large (CarB) subunits of CPSase appeared to be coordinately regulated, in response to changes in P availability (Fig. 3). CarA and CarB mRNA was barely detectable in shoots of control plants, and increased about fivefold in -P plants. Levels of both transcripts were approximately 10-fold higher in roots than in shoots of control plants. CarA and CarB transcript levels did not change in roots of either control or -P plants. UMP synthase (PyrE-F) transcript levels were about fivefold higher in shoots of -P vs. control plants and, as for CPSase, mRNA levels were about 10-fold higher in roots vs. shoots of control plants and changed little in response to P starvation (Fig. 3). UMP kinase (Uck) expression increased threefold in shoots. Uck transcript levels were unchanged in roots of control and -P plants, and were about fivefold and threefold higher than in shoots of these plants, respectively (Fig. 3). Expression of the Uprt gene encoding the salvage enzyme UPRTase increased about threefold in shoots of -P seedlings. In contrast to expression patterns seen for other genes, root Uprt mRNA levels decreased about twofold in -P plants, vs. controls. In control plants, Uprt transcript levels were about fivefold higher in roots than in shoots (Fig. 3). We identified Arabidopsis orthologs of the spinach Prs1-4 genes, and an additional Prs-2 like gene (At2g35390) that produces two different transcripts (accession numbers NM_179913 and NM_129091, the latter encoding a truncated PRS2-like protein which is missing the N-terminal putative chloroplast transit peptide sequence of the protein product of NM_179913, to which it is otherwise identical). The phylogenetic relationships between the Arabidopsis and spinach PRS enzymes, which exhibit high sequence identities, are shown in Fig. 5. Expression of Prs3 was constant in root and shoot tissues of both control and minus P plants. In contrast, P limitation increased Prs2 expression twofold in root tissues, while shoot expression was unchanged (Fig. 3). Expression of most genes examined in this study does not appear to have been investigated previously. Expression data for root and shoot (leaf) tissues from untreated, 3-week-old Arabidopsis plants were retrieved from the Arabidopsis MPSS database (http://mpss.udel.edu/at/java.html; [10]) for each of these genes, except Uck, and are summarized in Table 2. MPSS expression data for Pht1;2, PyrB and Prs2 and Prs3 were similar to expression data obtained from RT-PCR analyses. RT-PCR analyses showed markedly higher root vs. shoot expression profiles for CarA, CarB, PyrE-F, Uck and Uprt transcripts, compared with MPSS data.

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

95

Fig. 5. Phylogenetic relationship of PRPP synthases. Arabidopsis and spinach P-independent (class II) PRPP synthases form a distinct group from the remaining P-dependent PRPP synthases (class I). Amino acid sequences were aligned using CLUSTAL X v. 1.83 [41]. An unrooted tree was constructed using the neighbor-joining tree option in CLUSTAL X and displayed using the TreeView v. 1.6.6 drawing program [31]. Bootstrap values (for 1000 replicates) are shown at nodes. Accession numbers: spinach PRS 1 (CAB43599), PRS2 (CAB43600), PRS3 (CAB43601), and PRS4 (CAB43602); Arabidopsis PRS1 (NP_181981), PRS2 (NP_174516), PRS3 (NP_172540), PRS4 (NP_181819); Bacillus subtilis (NP_387932), E. coli (KIECRY).

3. Discussion 3.1. Effects of phosphate limitation on growth, expression of Pht1;2 and tissue RNA contents in Arabidopsis Seed P reserves appeared to be sufficient to support normal growth and development until about 9 d after planting on minus P medium. Control plants continued to grow normally until harvesting on day 14, when -P seedlings exhibited typical responses to P limitation, including arrested growth, alterations in root architecture, and an increased root to shoot biomass ratio. The threefold increase in expression of Pht1;2 in -P shoot tissues indicates that they were experiencing significant P deficiency. Smith et al. [39] reported a twofold increase in both APT1 and APT2 mRNA levels (Pht1;2, Pht1;1, respectively) in roots of Arabidopsis plants after 3 d of growth in a -P medium. Neither gene appeared to be expressed in shoot or leaf tissues in either control or P-deficient plants. In tomato,

the LePT1 and LePT2 orthologs of Pht1;1 and Pht1;2 showed similar responses to P starvation [23], with the exception that LePT1 was also weakly expressed in leaf palisade parenchyma after only 5 d of phosphate starvation. The observed shoot expression of Pht1;2 in our studies may be due to the fact that we used a highly-sensitive RT-PCR assay to detect Pht1;2 transcripts, compared with less sensitive RNA blot analyses employed by [39]. A Blast search [2] revealed that sequences for the gene-specific primers used to amplify a 5′-UTR region of this transcript are not present in any of the nine paralogs that comprise the Pht1 family [33], thus, the non-specific amplification of other Pht1 transcripts seems unlikely. Pht1;2 expression in control shoots may, therefore, represent a low basal level expression of this gene. Alternatively, these tissues may have begun to experience some phosphate limitation, leading to Pht1;2 induction, although no visible symptoms were associated with P deficiency were observed. Two recent studies [18,48] have shown that expression patterns for a large number of genes change within 3–4 d

96

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

Table 2 Massively-parallel signature sequencing (MPSS) expression profiles for genes analyzed by RT-PCR in Fig. 3. Signature sequences (17 bp) are located within exons or 3’-UTR regions of mRNAs. Values represent averages of expression data obtained from leaf and root libraries in which MPSS analyses were carried out by two different methods (LEAF = ″LEF″ + ″LES″, ROOT = ″ROF″ + ″ROS″; see http://mpss.udel.edu/at/java.html for details). For two of the transcripts, values for two different signatures were further averaged. Normalized transcript abundance values are expressed in parts per million (MPSS standard) or on the basis of equal numbers of Act-2 transcripts, for direct comparison with RT-PCR analyses Protein

Signatures

PHT1;2 CPSase (large) CPSase (small) ATCase UMP Syn UMP/CMP Kin UPRTase PRPP Syn 2 PRPP Syn 3 Actin-2

15 45 4 3 5 None 4 9, 13 23 33, 35

Normalized transcript abundance LEAF ROOT LEAF ROOT (PPM) (Act-2) 5 507 10 507 65 148 134 148 45 60 93 60 36 108 74 108 30 88 62 88 20 36 35 248

18 23 60 513

41 74 72 513

18 108 60 513

after P withdrawal, before effects on growth are noticeable. However, RNA contents of control shoot tissues were not significantly different from those seen in plants grown under similar conditions in other experiments, so any P limitation would have been minimal. In contrast, the marked reduction in total RNA contents of shoots of -P plants is consistent with previous reports that P starvation depresses synthesis of NTPs and nucleic acids [4,45,48] and increases their turnover as a result of increased production of phosphodiesterases and RNases [1,5,24,29]. The increased RNA content of roots versus shoots of -P plants presumably reflect differences in P availability for nucleotide and nucleic acid biosynthesis in these organs. Trull and Deikman [42] reported that although the total P in plants decreased, in response to P limitation, the proportion of total P found in the root actually increased. Under P starvation conditions, P is known to be redistributed from old source leaves to young sink organs and P-starved roots [33]. The high Pht1;2 expression levels in the roots would both facilitate this redistribution of P within the plant and increase uptake of available P from the growth medium. Such high-affinity Pi transporters permit plant roots to accumulate P to millimolar tissue concentrations from soil Pi concentrations that are typically in the µmolar range [33]. Indeed, overexpression of Pht1;1 in cultured tobacco cells increased Pi uptake and growth, even under P limitation [27]. 3.2. Effects of phosphate limitation on expression of genes involved in pyrimidine synthesis and salvaging ATCase catalyzes the committed step in the de novo synthesis of UMP and is highly-regulated in plants and other organisms [26]. We previously showed that PyrB expression increases in response to PALA-mediated pyrimidine starva-

tion in Arabidopsis seedlings [6]. We expected a similar response to P-starvation, which is known to inhibit nucleotide synthesis and salvaging. The approximately twofold increase in PyrB expression in -P seedlings was accompanied by a smaller increase in ATCase protein and enzyme activity. Decreased efficiency of translation of the PyrB transcript in PALA-treated seedlings, which also exhibit markedly decreased RNA contents (including tRNA, rRNA of the ribosome translational machinery), and a higher rate of turnover for the non-UMP-ligated enzyme in pyrimidine-starved tissues, may explain the lack of a more direct correlation between transcript and enzyme protein levels [6]. Expression of genes encoding other enzymes of de novo pyrimidine synthesis (CPSase, UMP synthase, UMP kinase) and salvaging (UPRTase) was markedly higher in roots than in shoots of control plants. In comparison, MPSS expression profiles for these genes showed approximately equal transcript levels, except for Uprt, which was actually about twofold higher in shoots than in roots. These differences may reflect different ages and growth conditions for plants used in the two studies (2 weeks on MS agar medium, present study; 3-week-old plants potted in soil, MPSS study), or the different methodologies used to estimate transcript levels. The fact that root expression of these genes did not further increase in response to P limitation, as it did in shoots, might indicate high constitutive expression levels in this organ or, conversely, that roots of control plants may have been experiencing initial stages of P limitation, resulting in induction of these genes. Elevated Pht1;2 transcript levels in the root would suggest this. However, MPSS data also indicate markedly higher (50-fold) Pht1;2 expression in roots versus shoots of (presumably) non-P-limited plants, even though expression of genes of pyrimidine metabolism was basically unchanged. The fact that RNA contents of roots actually increased in response to P limitation in the present study, coupled with the observation that P is redistributed from shoot to root tissues in such plants [33,42], further argue against P being limiting for nucleotide synthesis in these roots. We plan to conduct additional studies, examining earlier time points, in order to further characterize root expression patterns for these genes. In shoots, P starvation uniformly increased expression of pyrimidine genes. UMP synthase (PyrE-F) was previously demonstrated to be transcriptionally up-regulated in response to thymidylate starvation in FOA-treated tobacco tissues [35] or general pyrimidine starvation in PALA-treated seedlings (unpublished data). Thus, increased expression of this transcript was expected in -P seedlings. Regulation of CarA/CarB and Uck genes encoding CPSase and UMP kinase has not been previously investigated in plants. Carbamoyl-P production is considered to be the overall rate-limiting step in pyrimidine synthesis [32] and CPSase represents an important site of regulation. In plants, as in bacteria, a single Gln-dependent CPSase provides carbamoyl-P for both pyrimidine and arginine synthesis and is allosterically regulated by the UMP end-product of de novo pyrimidine synthesis, and by the ornithine intermediate in the argi-

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

97

nine pathway [13,30,37]. In bacteria, expression of the CarAB operon is regulated by binding of repressors to tandem pyrimidine-specific (P1) and arginine-specific (P2) promoter elements [11]. Whether a similar mechanism regulates transcription of the individual CarA and CarB genes in plants is not known. Levels of both mRNAs increase in response to decreased pyrimidine availability under P limitation (present study) or PALA-inhibition (unpublished observations), and decrease in response to exogenous pyrimidines, supporting a model in which a pyrimidine-inducible repressor may regulate their expression. UMP kinase commits UMP to the general pyrimidine nucleotide pool and the P starvation-induced increase in Uck expression may facilitate pyrimidine nucleotide synthesis. In addition to At3g60180, two additional Uck genes (At4g25280, At5g26667) are also expressed in Arabidopsis and represent a potential functional redundancy at this step. Nothing is known regarding regulation of these Uck genes, but the enzyme encoded by At4g25280 has been functionally characterized [50]. In shoots, up-regulation of the single Uprt gene by P limitation suggests that increased UPRTase-mediated salvaging of uracil to UMP may accompany the general up-regulation of genes encoding de novo pathway enzymes. We have identified four other expressed genes which may encode functionally-redundant UPRTase activities in Arabidopsis. Each of these genes (At1g55810, At3g27190, At4g26510, At5g40870) encodes a putative bifunctional uridine kinase/ UPRTase protein, although none appear to have been functionally characterized. The uridine kinase domains of these proteins could potentially catalyze the other major salvaging activity in plants (uridine to UMP; step 8, Fig. 1). Two additional genes encoding monofunctional uridine kinase proteins are also expressed in Arabidopsis (At1g26190, At1g73980), representing further potential redundancy in this enzyme activity. We chose not to study these genes, given the complexity of their gene families and general lack of information about their enzyme products, although one or more of these may be important in regulating nucleobase/nucleoside salvaging to UMP under P limitation. Sasamoto et al. [36] reported that uridine kinase activity is about twofold higher than UPRTase activity in Catharanthus, and uridine was salvaged more efficiently than uracil, suggesting that uridine kinase activity may be the more important of the two salvaging activities.

ies have led to the discovery of a novel type of PRPP synthase in plants, which may function in the synthesis of PRPP under conditions of low intracellular P availability. In spinach, genes encoding four different PRPP synthase isozymes have been identified (Prs1-4), and their protein products have been functionally characterized in an Escherichia coli Dprs mutant [20]. The PRS1 and PRS2 isozymes exhibit classical (class I) PRPP synthase activities (ATP diphosphoryl donor, Pi required for maximal activity, allosteric inhibition by ADP), while the PRS3 and PRS4 isozymes belong to a novel group of PRPP synthases (class II) found only in plants [21]. The class II enzymes exhibit a low specificity for the diphosphoryl donor (accepting dATP, GTP, CTP and UTP, in addition to ATP), are competitively inhibited by NDPs, indicating a lack of allosteric regulation, and exhibit activity which is Pi-independent. We investigated the effects of P limitation on the expression of Arabidopsis genes At1g32380 and At1g10700, encoding the PRS2 and PRS3 isoforms, respectively, for the following reasons. First, subcellular fractionation evidence or targeting prediction, based upon amino acid sequence analyses, places PRS2 and PRS3, along with the PRPP-requiring enzymes UMP synthase and UPRTase, within the plastid [14,15,20]. Second, the biochemical properties of PRS3 suggested that it might be particularly well-suited to produce PRPP under conditions of P limitation. In Arabidopsis, P limitation resulted in induction of Prs2 in roots, but not shoots, although the potential contribution of any increase in PRS2 enzyme activity to overall PRPP synthesis would be expected to be minimal at low intracellular concentrations of Pi and ATP. In contrast, Prs3 expression was not significantly different in root and shoot tissues of control and -P plants. Heterologously-expressed PRS3 displays an approximately eightfold higher specific activity than PRS2 at optimal Pi concentrations, and approximately 50-fold higher activity in the absence of Pi [20]. The unique attributes of this PRS (high activity, Pi-insensitivity, and ability to utilize a range of phosphoribosyl donors in PRPP synthesis) may confer a selective advantage on plants for growth under conditions of low P availability. The Arabidopsis PRS2 and PRS3 enzymes appear to represent a functional redundancy which would permit PRPP production in the plastid, supporting the synthesis of pyrimidines and other PRPP-requiring processes, over a wide range of intracellular Pi concentrations.

3.3. Potential role for PRS3 in maintaining chloroplastic PRPP synthesis under P limitation

4. Methods

PRPP is the phosphoribosyl donor in the synthesis of pyrimidine nucleotides at the UMP synthase (i.e. OPRTase) and UPRTase steps of the de novo and salvage pathways, respectively. Ashihara et al. [3] reported that P limitation reduced nucleotide titers and synthesis via de novo and salvage pathways in Catharanthus cell cultures, and concluded that this was due primarily to reduced synthesis of PRPP. Recent stud-

4.1. Plant tissues Arabidopsis thaliana L. seeds (Col-2, Columbia ecotype; stock number CS907, Arabidopsis Biological Resource Center, Columbus, OH) were sterilized as follows. Seeds were soaked in 10 ml of sterile water + 0.1% Tween-20 for 30 min, followed by 10 ml of 95% EtOH for 15 min, 10 ml of 20%

98

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99

bleach for 5 min, and five rinses in 10 ml of sterile water. Approximately 10–15 seeds were plated on 150 mm petri dishes containing 50 ml of standard MS medium ([28]; Gibco BRL, Gaithersburg, MD), 0.2% Gelrite (Scott Laboratories, Carson, CA) and 1% sucrose, or on phosphate-minus MS medium (MS macronutrients, minus KH2PO4; medium typically contains < 30 µM Pi, compared with 1.25 mM Pi in standard MS medium). Plates were placed at 4 °C for 2 d, in order to synchronize seed germination, then removed to a growth table under a 16 l/8D photoperiod, 25 °C and a light intensity of approximately 100 µE m–2 s–1. Seedlings were harvested after 14 d of growth.

conditions were optimized for each template and the number of cycles required for linear amplification of target sequences was empirically determined. The constitutively-expressed Act-2 transcript [16] was amplified as an internal reference. Aliquots of the PCR reactions were run on agarose gels and the products were visualized by ethidium bromide staining. Minus RT reactions were run as controls and RT-PCR samples showed no evidence of contamination by genomic DNA. Band densities were quantitated by scanning densitometry [38].

4.2. Tissue extraction and protein determination

This work was supported by National Science Foundation grant MCB-0076881.

Seedlings were briefly rinsed with water to remove attached solid medium from roots, then blotted dry and weighed. Shoot and root tissues were homogenized at 0.5 g tissue per ml of Extraction Buffer (100 mM Hepes, pH 7.5, 10% (v/v) glycerol, 5 mM EDTA, 5 mM 2-mercaptoethanol, 1 mM PMSF). The extract was centrifuged at 16,000 × g for 10 min at 4 °C. and the soluble protein content of the supernatant was determined using the dye-binding assay of [8]. 4.3. Quantitation of ATCase protein and enzyme activity ATCase protein in tissue homogenates was estimated by scanning densitometric quantitation of band densities on SDSPAGE immunoblots, and ATCase activity in the extracts was determined, as was previously described [6]. Enzyme specific activity was expressed as katals of carbamoylaspartate (CarAsp) per milligram of protein at 30 °C, pH 8.5. 4.4. RT-PCR analyses of transcript levels Total RNA was isolated from seedling shoot and root tissues using a modification of the method described by [12]. Steady-state transcript levels were estimated by semiquantitative RT-PCR, as was previously described [6]. Briefly, 40 µg of total RNA was treated with RNase-free DNase I, then phenol-chloroform extracted and ethanol precipitated. The RNA pellet was dried and rehydrated in DEP-treated (Rnase-free) water and quantitated by OD260. Approximately 5 µg of each sample was electrophoresed in 1.5% formaldehyde denaturing gels in order to evaluate the quality of isolated RNA. Two micrograms of the RNA was reverse transcribed in 20 µl reactions containing 0.5 µg random hexamer primers, 0.8 mM dNTPs, 10 mM DTT, 20 units RNasin, and 200 units of M-MLV reverse transcriptase in 1 × RT buffer, according to the manufacturer’s instructions (Gibco BRL). The RT samples were diluted 1:5 with water and 5 µl of each diluted sample was added to 20 µl PCR reaction mixture (1.2 pmol primers, 0.2 mM dNTPs, 3 mM Mg2+, and 0.625 U of AmpliTaq Gold DNA Polymerase in 1× PCR buffer (PE Applied Biosystems, Foster City, CA) in 25 µl final volume) and amplified, using gene-specific primers (see Table 1). PCR

Acknowledgements

References [1]

S. Abel, T. Nürnberger, V. Ahnert, G.-J. Krauss, K. Glund, Induction of an extracellular cyclic nucleotide phosphodiesterase as an accessory ribonuclolytic activity during phosphate starvation of cultured tomato cells, Plant Physiol. 122 (2000) 543–552. [2] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignment search tool, J. Mol. Biol. 215 (1990) 403–410. [3] H. Ashihara, X.-N. Li, T. Ukaji, Effect of inorganic phosphate on the biosynthesis of purine and pyrimidine nucleotides in suspensioncultured cells of Catharanthus roseus, Ann. Bot. (Lond.) 61 (1988) 225–232. [4] H. Ashihara, T. Tokoro, Metabolic fate of inorganic phosphate absorbed by suspension cultured cells of Catharanthus roseus, J. Plant Physiol. 118 (1985) 227–235. [5] P.A. Bariola, C.J. Howard, C.P. Taylor, M.T. Verburg, V.D. Jaglan, P.J. Green, The Arabidopsis ribonuclease gene RSN1 is tightly controlled in response to phosphate limitation, Plant J. 6 (1994) 673–685. [6] E.V. Bassett, B.Y. Bouchet, J.M. Carr, C.L. Williamson, R.D. Slocum, PALA-mediated pyrimidine starvation increases expression of aspartate transcarbamoylase (pyrB) in Arabidopsis seedlings, Plant Physiol. Biochem. 41 (2003) 695–703. [7] T.R. Bates, J.P. Lynch, Stimulation of root hair elongation in Arabidopsis thaliana by low phosphorous availability, Plant Cell Environ. 19 (1996) 529–538. [8] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dyebinding, Anal. Biochem. 72 (1976) 248–254. [9] S.A. Brandenburg, C.L. Williamson, R.D. Slocum, Characterization of a cDNA encoding the small subunit of Arabidopsis carbamoyl phosphate synthetase (Accession No. U73175) (PGR 98-087), Plant Physiol. 117 (1998) 717. [10] S. Brenner, M. Johnson, J. Bridgham, G. Golda, D.H. Lloyd, D. Johnson, et al., Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays, Nat. Biotechnol. 18 (2000) 630–634. [11] D. Charlier, G. Hassandzadeh, A. Kholti, D. Gigot, A. Pierard, N. Glansdorff, carP, involved in pyrimidine regulation of the Escherichia coli carbamoylphosphate synthetase operon, encodes a sequence-specific DNA-binding protein identical to XerB and PepA, also required for resolution of ColEI multimers, J. Mol. Biol. 250 (1995) 392–406. [12] P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction, Anal. Biochem. 162 (1987) 156–159.

M.M. Hewitt et al. / Plant Physiology and Biochemistry 43 (2005) 91–99 [13] R. Cunin, N. Glansdorff, A. Pierard, V. Stalon, Biosynthesis and metabolism of arginine in bacteria, Microbiol. Rev. 50 (1986) 314– 352. [14] H.D. Doremus, A.T. Jagendorf, Subcellular localization of the pathway of de novo pyrimidine nucleotide biosynthesis in pea leaves, Plant Physiol. 79 (1985) 856–861. [15] O. Emanuelsson, H. Nielsen, S. Brunak, G. von Heijne, Predicting subcellular localization of proteins based on their N-terminal amino acid sequence, J. Mol. Biol. 300 (2000) 1005–1016 (TargetP URL). [16] L.U. Gilliland, M.K. Kandasamy, L.C. Pawloski, R.B. Meagher, Both vegetative and reproductive actin isovariants complement the stunted root hair phenotype of the Arabidopsis act2-1 mutation, Plant Physiol. 130 (2002) 2199–2209. [17] A. Grossman, H. Takahashi, Macronutrient utilization by photosynthetic eukaryotes and the fabric of interactions, Annu. Rev. Plant Physiol. Plant Mol. Biol. 52 (2001) 163–210. [18] J.P. Hammond, M.J. Bennett, H.C. Bowen, M.R. Broadley, D.C. Eastwood, S.T. May, et al., Changes in gene expression in Arabidopsis shoots during phosphate starvation and the potential for developing smart plants, Plant Physiol. 132 (2003) 578–596. [19] C. Kafer, L. Zhou, D. Santoso, A. Guirgis, B. Weers, S. Park, et al., Regulation of pyrimidine metabolism in plants, Front. Biosci. 9 (2004) 1611–1625. [20] B.N. Krath, B. Hove-Jensen, Organellar and cytosolic localization of four phosphoribosyl diphosphate synthase isozymes in spinach, Plant Physiol. 119 (1999) 497–505. [21] B.N. Krath, B. Hove-Jensen, Class II recombinant phosphoribosyl diphosphate synthase from spinach. Phosphate independence and diphosphoryl donor specificity, J. Biol. Chem. 276 (2001) 17851– 17856. [22] D.S. Lipton, R.W. Blancher, D.G. Blevins, Citrate, malate and succinate concentrations in exudates from P-sufficient and P-starved Medicago sativa L. seedlings, Plant Physiol. 85 (1987) 315–317. [23] C. Liu, U.S. Muchhal, M. Uthappa, A.K. Kononowicz, K.G. Raghothama, Tomato phosphate transporter genes are differentially regulated in plant tissues by phosphorous, Plant Physiol. 116 (1998) 91–99. [24] A. Löffler, S. Abel, W. Jost, K. Glund, Phosphate-regulated induction of intracellular ribonucleases in cultured tomato (Lycopersicon esculentum) cells, Plant Physiol. 98 (1992) 1472–1478. [25] J. López-Bucio, E. Hernández-Abreu, L. Sánchez-Calderón, M.F. Nieto-Jacobo, J. Simpson, L. Herrera-Estrella, Phosphate availability alters architecture and causes changes in hormone sensitivity in the Arabidopsis root system, Plant Physiol. 129 (2002) 244–256. [26] C.J. Lovatt, A.H. Cheng, Aspartate carbamyltransferase. Site of endproduct inhibition of the orotate pathway in intact cells of Cucurbita pepo, Plant Physiol. 75 (1984) 511–515. [27] N. Mitsukawa, S. Okumura, Y. Shirana, S. Sato, T. Kato, S. Hrashima, et al., Overexpression of an Arabidopsis thaliana high-affinity phosphate transporter gene in tobacco cultured cells enhances cell growth under phosphate-limited conditions, Proc. Natl. Acad. Sci. USA 94 (1997) 7098–7102. [28] T. Murashige, F. Skoog, A revised medium for rapid growth and bioassay with tobacco tissue cultures, Physiol. Plant. 15 (1962) 473– 497. [29] T. Nürnberger, S. Abel, W. Jost, K. Glund, Induction of an extracellular ribonuclease in cultured tomato cells upon phosphate starvation, Plant Physiol. 92 (1990) 970–976. [30] T.D. O’Neal, A.W. Naylor, Some regulatory properties of pea leaf carbamoyl phosphate synthetase, Plant Physiol. 57 (1976) 23–28. [31] R.D.M. Page, TREEVIEW: an application to display phylogenetic trees on personal computers, Comput. Appl. Biosci. 12 (1996) 357– 358.

99

[32] N.F. Parker, J.F. Jackson, Control of pyrimidine biosynthesis in synchronously dividing cells of Helianthus tuberosus, Plant Physiol. 67 (1981) 363–366. [33] Y. Poirier, M. Bucher, Phosphate transport and homeostasis in Arabidopsis, in: C.R. Somerville, E.M. Meyerowitz (Eds.), The Arabidopsis Book, American Society of Plant Biologists, Rockville, MD, 2002 (doi/10.119/tab.0024, http://www.aspb.org/publications/arabidopsis). [34] K.G. Ragothama, Phosphate acquisition, Annu. Rev. Plant Physiol. Plant Mol. Biol. 50 (1999) 665–693. [35] D. Santoso, R. Thornburg, UMP synthase is transcriptionally regulated by pyrimidine levels in Nicotiana plumbaginifolia, Plant Physiol. 116 (1998) 815–821. [36] H. Sasamoto, K. Saito, H. Ashihara, Metabolism of pyrimidines in protoplasts from cultured Catharanthus roseus cells, Ann. Bot. (Lond.) 60 (1987) 417–420. [37] H. Sasamoto, H. Ashihara, Biosynthesis of pyrimidine nucleotides and arginine in a suspension culture of Catharanthus roseus, Int. J. Biochem. 20 (1988) 87–92. [38] T.B. Shea, An inexpensive densitometric analysis system using a Macintosh™ computer and a desktop scanner, Biotechniques 16 (1994) 1126–1128. [39] F.W. Smith, P.M. Ealing, B. Dong, E. Delhaize, The cloning of two Arabidopsis genes belonging to a phosphate transporter family, Plant J. 11 (1997) 83–92. [40] C. Stasolla, R. Katahira, T.A. Thorpe, H. Ashihara, Purine and pyrimidine nucleotide metabolism in higher plants, J. Plant Physiol. 160 (2003) 1271–1295. [41] J.D. Thompson, T.J. Gibson, F. Plewniak, F. Jeanmougin, D.G. Higgins, The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools, Nucleic Acids Res. 25 (1997) 4876–4882. [42] M.C. Trull, J. Deikman, Arabidopsis thaliana: a model system for examining plant responses to phosphorous starvation, in: H.E. Flores, J.P. Lynch, D. Eissenstat (Eds.), Radical Biology: Advances and Perspectives on the Function of Plant Roots, American Society of Plant Physiologists, Rockville, MD, 1997, pp. 59–66. [43] T. Ukaji, H. Ashihara, Effect of inorganic phosphate on synthesis of 5-phosphoribosyl-1-pyrophosphate in cultured plant cells, Int. J. Biochem. 19 (1987) 1127–1131. [44] A. Ullrich, W. Knecht, J. Piskur, M. Löffler, Plant dihydroorotate dehydrogenase differs significantly in substrate specificity and inhibition from animal enzymes, FEBS Lett. 529 (2002) 346–350. [45] K.G. Wagner, A.I. Backer, Dynamics of nucleotides in plants studied on a cellular basis, Int. Rev. Cytol. 134 (1992) 1–84. [46] C.L. Williamson, M.R. Lake, R.D. Slocum, A cDNA encoding carbamoyl phosphate synthetase large subunit (carB) from Arabidopsis (Accession No. U40341) (PGR 96-055), Plant Physiol. 111 (1996) 1354. [47] L.C. Williamson, S.P.C.P. Ribrioux, A.H. Fitter, H.M.O. Leyser, Phosphate availability regulates root system architecture in Arabidopsis, Plant Physiol. 126 (2001) 875–882. [48] P. Wu, L. Ma, S. Hou, M. Wang, Y. Wu, F. Liu, et al., Phosphate starvation triggers distinct alterations of genome expression in Arabidopsis roots and leaves, Plant Physiol. 132 (2003) 1260–1271. [49] C. Wylegalla, R. Meyer, K.G. Wagner, Nucleotide pools in suspension-cultured cells of Datura innoxia II. Correlation with nutrient uptake and macromolecular synthesis, Planta 166 (1985) 446– 451. [50] L. Zhou, F. Lacroute, R. Thornburg, Cloning, expression in Escherichia coli and characterization of Arabidopsis thaliana uridine 5′-monophosphate/cytidine 5′-monophosphate kinase, Plant Physiol. 117 (1998) 245–253.