Yeast GMP Kinase Mutants Constitutively Express AMP Biosynthesis Genes by Phenocopying a Hypoxanthine-Guanine Phosphoribosyltransferase Defect

Corresponding author: Bertrand Daignan-Fornier, Institut de Biochimie et Génétique Cellulaires, 1, rue Camille Saint-Saëns, 33077 Bordeaux Cedex, France., b.daignan-fornier{at}ibgc.u-bordeaux2.fr (E-mail)

Communicating editor: A. G. HINNEBUSCH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have characterized a new locus, BRA3, leading to deregulation of the yeast purine synthesis genes (ADE genes). We show that bra3 mutations are alleles of the GUK1 gene, which encodes GMP kinase. The bra3 mutants have a low GMP kinase activity, excrete purines in the medium, and show vegetative growth defects and resistance to purine base analogs. The bra3 locus also corresponds to the previously described pur5 locus. Several lines of evidence indicate that the decrease in GMP kinase activity in the bra3 mutants results in GMP accumulation and feedback inhibition of hypoxanthine-guanine phosphoribosyltransferase (HGPRT), encoded by the HPT1 gene. First, guk1 and hpt1 mutants share several phenotypes, such as adenine derepression, purine excretion, and 8-azaguanine resistance. Second, overexpression of HPT1 allows suppression of the deregulated phenotype of the guk1 mutants. Third, we show that purified yeast HGPRT is inhibited by GMP in vitro. Finally, incorporation of hypoxanthine into nucleotides is similarly diminished in hpt1 and guk1 mutants in vivo. We conclude that the decrease in GMP kinase activity in the guk1 mutants results in deregulation of the ADE gene expression by phenocopying a defect in HGPRT. The possible occurrence of a similar phenomenon in humans is discussed.

MICROORGANISMS modify their metabolism in response to environmental changes. When metabolic precursors are present in the extracellular medium, yeast and bacteria generally use these precursors instead of synthesizing them de novo. Consequently, the synthesis of metabolic enzymes is regulated according to the presence of metabolites in the medium.

Such a regulatory mechanism exists for the purine biosynthesis pathway in Saccharomyces cerevisiae. Indeed, all the genes encoding enzymes required for de novo AMP biosynthesis are repressed at the transcriptional level by the presence of extracellular purines (adenine or hypoxanthine) (MANTSALA and ZALKIN 1984 ; GIANI et al. 1991 ; DAIGNAN-FORNIER and FINK 1992 ; DENIS et al. 1998 ). This regulation process requires two transcription factors, Bas1p and Bas2p (DAIGNAN-FORNIER and FINK 1992 ), and regulation by extracellular purines has been proposed to occur through interactions between these two factors (ZHANG et al. 1997 ).

To gain an insight into the signal transduction pathway between extracellular adenine and the transcription factors, we have isolated mutants in which purine biosynthesis genes are no longer repressed by extracellular adenine (GUETSOVA et al. 1997 ). These mutants are termed bra for bypass of repression by adenine. The identification of some of these mutants has shown that the repression effect of adenine requires the purine base to enter the cell, i.e., bra7 mutants are alleles of the FCY2 gene that codes for the purine cytosine permease (see Fig 1; GUETSOVA et al. 1997 ). Adenine itself is not the effector molecule since it has to be metabolized into AMP and then ADP to exert its regulatory effect. There are two possible routes for the synthesis of AMP from adenine: a direct one catalyzed by adenine phosphoribosyltransferase (APRT) and a more indirect one requiring four enzymatic steps via the formation of hypoxanthine and IMP (see Fig 1). The APRT route does not seem to play a major role in the repression process since disruption of APT1, the APRT encoding gene, had no effect on adenine repression (GUETSOVA et al. 1997 ). We found that mutations in the genes encoding hypoxanthine-guanine phosphoribosyltransferase (HGPRT), adenylosuccinate synthase (Ade12p), and adenylosuccinate lyase (Ade13p) abolished the repression signal (BRA6, BRA9, and BRA1 are HPT1, ADE12, and ADE13, respectively; GUETSOVA et al. 1997 ; see Fig 1). The major route for the repression signal from adenine to AMP is thus via the formation of hypoxanthine and IMP. In addition, we have shown that AMP needs to be phosphorylated into ADP to exert its regulatory role (GUETSOVA et al. 1997 ).

View larger version (18K): In this window In a new window Download PPT slide   Figure 1. Scheme of purine interconversion in yeast. The following abbreviations are used: PRPP, 5-phosphoribosyl-1-pyrophosphate; IMP, inosine 5'-monophosphate; GMP, guanosine 5'-monophosphate; GDP, guanosine 5'-diphosphate. Gene names are indicated in italic and encode the following enzymatic activities: AAH1, adenine deaminase; ADE12, adenylosuccinate synthetase; ADE13, adenylosuccinate lyase; ADK1, AMP kinase; APT1, adenine phosphoribosyltransferase; FCY2, purine cytosine permease; GUK1, GMP kinase; HPT1, hypoxanthine-guanine phosphoribosyltransferase. The thick arrows represent the major route for the repression signal by adenine. For the purpose of simplification, nucleosides are not represented.

The bra mutants define more than 10 complementation groups, indicating that the regulation process is complex and requires several proteins. To identify a new partner in the signal transduction pathway, we have now characterized the bra3 complementation group. We show that BRA3 is GUK1, an essential gene encoding GMP kinase. We document several new phenotypes associated with the GMP kinase defect and present evidence that the guk1 mutations result in a phenocopy of hpt1. The possible implications for human diseases associated with purine overexpression and uric acid excretion are discussed.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains and media:
Yeast strains are listed in Table 1. Yeast media were prepared according to SHERMAN et al. 1986 . Adenine and hypoxanthine were used at a final concentration of 0.15 mM. The XGal synthetic medium was prepared using the methods previously described (DANG et al. 1994 ). The base analog 8-azaguanine (8AG) was added to the medium at a final concentration of 0.2 mg/ml.

Plasmids:
pCG3 (DEELEY 1992 ) is a YEp13 (BROACH et al. 1979 ) derivative carrying the APT1 gene.

P556, the 2µ plasmid carrying the HPT1 gene, was obtained by ligating the EcoRI-HindIII fragment from P385 that contains the HPT1 gene (GUETSOVA et al. 1997 ) into the multicopy plasmid YEpLac181 (GIETZ and SUGINO 1988 ) linearized with HindIII and EcoRI.

P195, the plasmid carrying the GUK1 wild-type gene, was constructed by insertion of the HindIII-BglII fragment from the pGUK1 plasmid (KONRAD 1992 ) into a centromeric LEU2 vector named pRS315 (SIKORSKI and HIETER 1989 ) linearized with HindIII and BamHI.

The P1718 plasmid expressing the Hpt1p-His₆ fusion in bacteria was constructed as follows: a 684-bp fragment carrying the HPT1 coding sequence was amplified by PCR using the following synthetic oligonucleotides: HPT1Ca, 5'-GTGATG CATATGTCGGCAAACGATAAGC-3' and HPT1Cb, 5'-CGAT GCTCGAGATTGCTTGTGTTCCTGCTC-3'. The PCR product was cut with NdeI and XhoI and introduced into a pJC20-HisC expression plasmid linearized with NdeI and XhoI. The resulting plasmid encodes a 26.7-kD Hpt1p-His₆ fusion protein. The pJC20-HisC vector was generated by introducing a double-stranded oligonucleotide linker into pJC20 (KONRAD 1993 ) restricted with BamHI and ApaI: 5'-GATCC CAT CAC CAT CAC CAT CAC TGA GGGCC-3' (sense) and 5'-C TCA GTG ATG GTG ATG GTG ATG G-3' (antisense).

LacZ fusions and ßGal assays:
The lacZ fusions used in this study have been previously described (DAIGNAN-FORNIER and FINK 1992 ; GUETSOVA et al. 1997 ). P115 is a plasmid carrying an ADE1-lacZ fusion in a 2µ URA3 vector YEp356R (MYERS et al. 1986 ). P473 is a plasmid carrying an ADE1-lacZ fusion in a 2µ LEU2 vector YEp367 (MYERS et al. 1986 ).

ßGal assays were performed as described by KIPPERT 1995 on cells grown for 6 hr in the presence or absence of purine base. ßGal units are defined as:

The repression factor is defined as the ratio of ßGal units measured in the absence of purine to those measured in the presence of purine. In each experiment, at least two independent ßGal assays were performed, and each assay was done on three independent transformants. The variation between assays in each experiment was <20%.

Integration of LEU2 at the GUK1 locus:
A HindIII-XbaI fragment carrying the GUK1 gene from plasmid P195 was cloned into the pRS305 integrative LEU2 vector (SIKORSKI and HIETER 1989 ) linearized with HindIII and XbaI. The resulting plasmid named P1015 was linearized at the unique StuI site in the GUK1 coding region and used to transform the Y642 yeast strain. Correct integration of the plasmid at the GUK1 locus was verified by Southern blot analysis of genomic DNA extracted from tranformants and cut with BglII (data not shown). One of these transformants named Y882 was used for linkage analysis.

AMP and GMP kinase enzymatic assays:
AMP and GMP kinase activities in protein extracts were measured using a spectrophotometric assay in a coupled lactate dehydrogenase/pyruvate kinase system according to the method of AGARWAL et al. 1978 . Briefly, yeast strains were grown in 20 ml of rich YPD medium to an OD₆₀₀ of 0.75 ± 0.05. Cells were harvested, washed with water, and resuspended in 0.6 ml breaking buffer (20 mM TrisHCl pH 7.9, 10 mM MgCl₂, 1 mM EDTA, 5% glycerol, 1 mM dithiothreitol (DTT), 0.3 M ammonium sulfate, and 2 mM phenylmethylsulfonyl fluoride). The cells were then broken with glass beads by vortexing four times for 30 sec in the cold. After addition of 0.2 ml of breaking buffer, glass beads and unbroken cells were pelleted in a microfuge for 5 min and 0.5 to 10 µl of the supernatant was used for the enzymatic assay. The assay was done in 100 mM TrisHCl pH 7.5, 100 mM KCl, 10 mM MgCl₂, 0.25 mM NADH, 0.5 mM PEP, 2 mM ATP, 5 units lactate dehydrogenase (Sigma, St. Louis), and 4 units pyruvate kinase (Sigma) in a final volume of 1 ml. Finally, 1 mM AMP or GMP was added to the reaction mix, depending on which nucleotide kinase activity was tested. Conversion of AMP to ADP by AMP kinase or that of GMP to GDP by GMP kinase was monitored by the decrease of absorbance of NADH at 340 nm (extinction coefficient = 6.2 cm² µmol^-1). One unit of enzyme activity is defined as the consumption of 1 µmol of nucleoside triphosphate per minute. Protein concentration was determined using the Bio-Rad (Hercules, CA) Protein Micro Assay system with crystalline bovine serum albumin serving as the reference standard. GMP kinase activity values are the result of two independent enzymatic assays, each performed with three different protein extract concentrations.

Yeast HGPRT expression and purification:
To produce yeast HGPRT, P1718 was expressed in the C41 (DE3) Escherichia coli strain (MIROUX and WALKER 1996 ). Yeast HGPRT was expressed and purified under native conditions using the QIAGEN (Chatsworth, CA) QIAexpressionist kit with the following modifications: in all buffers, NaH₂PO₄ was replaced by Tris, and elution buffer contained 350 mM imidazole.

Determination of yeast HGPRT kinetic parameters:
HGPRT assay was done in a 50-µl mix containing [8-³H]hypoxanthine (20 Ci/mmol, ICN Pharmaceuticals, Irvine, CA), 0.1 mM 5-phosphoribosyl-1-pyrophosphate, 100 mM Tris pH 8.0, 4 mM DTT, 10 mM MgCl₂, and 1.5 ng of purified yeast HGPRT. Initial rate measurements were performed at 30° and reactions were stopped after 90 sec. Hypoxanthine K_M value was determined using 1 to 200 µM hypoxanthine concentrations. Each reaction was stopped by adding 1 ml PRT stop buffer (50 mM Na-acetate and 2 mM Na₂HPO₄ pH 5.0) and lanthanium chloride (200 µl of 2.5 M LaCl₃), allowing precipitation of the nucleotide product. The samples were incubated on ice for 1 hr and the precipitate was collected on GF/C glass filters, washed 6 times with 1.5 ml cold water, and dried at 80° for 45 min. The filters were then placed in scintillation counting vials along with 5 ml Packard Filter Count scintillation liquid (complete LSC-cocktail for counting membrane filters, Packard, Meriden, CT) and counted on a TRI CARB 1500 Packard scintillation counter to determine the amount of [³H]hypoxanthine converted into inosine 5'-monophosphate. To determine the K_i value for GMP, the apparent K_M (K_{M app}) for hypoxanthine in the presence of either 50 µM, 100 µM, or 200 µM GMP was assayed. The inhibition was determined to be competitive since there was no major variation of the yeast HGPRT V_max when increasing GMP concentration. The K_i was thus calculated using the equation

In vivo [¹⁴C]hypoxanthine incorporation:
Wild-type (Y350), guk1 (220), and hpt1 (Y508) strains were grown in 60 ml of minimal medium supplemented with 20 g/liter casamino acids, 0.2 mM uracil, and 0.2 mM tryptophan to an OD₆₀₀ of 1.0. Cells were then harvested and resuspended in 6 ml of the same medium plus 20 µM hypoxanthine containing 1 µCi [8-¹⁴C]hypoxanthine (50 mCi/mmol, ICN). Cells were allowed to grow for 15 or 90 min and then 1 ml of culture was used for extraction of the intracellular purine compounds according to GONZALEZ et al. 1997 . Separation of the purine compounds was then achieved by HPLC using a supelcosil LC-18 5-µm reversed phase column. Gradient was set up with A buffer (0.025 M K₂HPO₄) and B buffer (0.05 M K₂HPO₄, 25% methanol). The following proportions of A and B buffers, respectively (indicated in parentheses), were used at the indicated run time: 0 min (98/2), 5 min (96/4), 10 min (70/30), 20 min (20/80), 24 min (20/80), and 25 min (98/2) and the flow was 1.25 ml/min. Fractions were collected every 30 sec during the 5 first min of the run and then every minute. All fractions were adjusted with water to 1.25 ml, transferred to scintillation counting vials along with 5 ml PCS liquid scintillation cocktail (Amersham, Buckinghamshire, United Kingdom), and counted on a TRI CARB 2100TR Packard scintillation counter to determine the amount of ¹⁴C radioactivity in each vial.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The bra3 mutants are resistant to 8-azaguanine and excrete purines:
In a previous report, we have shown that the bra3 complementation group contains four members, two in each mating type (GUETSOVA et al. 1997 ). All four mutants in addition to their derepression phenotype showed resistance to base analogs such as 8-azaadenine (8AA) and 8-azaguanine (8AG). The bra3-2 mutant was crossed with a wild-type strain and the meiotic products of this cross were scored for adenine repression and 8AG resistance. In 15 analyzed tetrads, we noticed a 2:2 segregation of the colony size on the germination plate (data not shown). Interestingly, all the spores forming smaller colonies were 8AG resistant and deregulated for expression of an ADE1-lacZ fusion (Bra^-), while all the large colonies were 8AG sensitive and Bra⁺. Therefore, adenine derepression, resistance to 8AG, and a growth defect cosegregate in this cross. The bra3-1 and bra3-2 mutants and the isogenic PLY121 wild-type strain were grown in rich YPD medium at 30° and their generation time was measured. The bra3-1, bra3-2, and wild-type generation time during exponential growth was 148, 131, and 109 min, respectively, thus establishing that the bra3 mutants have a slower vegetative growth. Finally, we found that the bra3-1 and bra3-2 mutants excrete purines in the medium, as shown by cross-feeding experiments, using a plate assay based on the rescue of an ade1 mutant on purine-free medium. In this assay, the purine-excreting mutants are surrounded by a halo of growing ade1 colonies. As shown in Fig 2, growth of bra3-1 and bra3-2 mutants on a purine-free medium sustained growth of an adenine auxotrophic strain. Apparently the bra3-1 mutant was a more efficient purine excretor than bra3-2.

View larger version (60K): In this window In a new window Download PPT slide   Figure 2. Purine excretion by the bra3-1 and bra3-2 mutants. A lawn of ade1 (L587) cells was plated on purine-free SC medium. A suspension of wild-type (PLY121) or bra3 mutant (129 and 130) cells was dropped onto this lawn. Purine excretion was monitored after 5 days at 30°.

BRA3 is GUK1:
The BRA3 gene was cloned by complementation of the bra3-3 derepression phenotype. This mutant was transformed with a genomic library carried on a centromeric vector. Candidates for complementation were isolated according to their ability to repress the expression of an ADE1-lacZ fusion in the presence of adenine. Such candidates were scored as pale blue among dark blue colonies on XGal medium plus adenine. The P131 plasmid, isolated as complementing the bra3-3 mutant phenotype, was also shown to complement the derepression phenotype of the other three mutants of the complementation group. This plasmid was further analyzed and shown to contain a yeast DNA insert that hybridized to chromosome IV (data not shown). Sequence analysis of both termini of the yeast DNA insert revealed that this plasmid carried a 9.2-kb fragment from chromosome IV. This fragment contained five complete open reading frames (YDR452w to YDR456w), one of which (YDR454c) corresponded to the previously described GUK1 gene that encodes GMP kinase, a purine metabolism enzyme catalyzing phosphorylation of GMP into GDP (KONRAD 1992 ). Because of its important role in purine metabolism, we then tested whether the GUK1 gene alone could complement the derepression phenotype of bra3-1 and bra3-2 mutants.

A centromeric plasmid carrying a 919-bp HindIII-BglII fragment containing only the GUK1 gene was constructed. This plasmid and a control plasmid were transformed in bra3-1 and bra3-2 mutants carrying an ADE1-lacZ fusion, and the repression by adenine of the ADE1-lacZ fusion expression was tested. The repression factor—defined as the ratio between expression of the ADE1-lacZ fusion under derepression (-ade) and repression (+ade) conditions—was calculated. Results in Table 2 show that whereas the repression factor in bra3-1 and bra3-2 mutants transformed with the control plasmid was very low (1.2 and 1.5, respectively), transformation of the two mutants with the plasmid carrying the GUK1 gene restored a higher repression factor (8.1 and 10.2, respectively), similar to the wild-type level (7.6). This experiment thus clearly established that the 919-bp DNA HindIII-BglII fragment containing the GUK1 gene was sufficient to restore adenine regulation when introduced into the bra3-1 and bra3-2 mutant strains. The GUK1 gene alone was also able to fully complement the 8AG resistance of the bra3 mutants (Fig 3).

View larger version (44K): In this window In a new window Download PPT slide   Figure 3. Complementation of the bra3 mutants' 8AG resistance by the GUK1 gene. Strains were transformed with a control centromeric plasmid or a centromeric plasmid carrying the wild-type GUK1 gene. A serial dilution of the different transformants was dropped onto SC medium containing 8AG and the growth on this medium was observed after 3 days at 30°.

We then tested the linkage between the bra3 mutation and the GUK1 gene. Since disruption of GUK1 is lethal (KONRAD 1992 ), we could not use a GUK1-disrupted strain for linkage analysis. Therefore, we crossed the bra3-2 mutant with a wild-type strain in which the LEU2 marker was integrated at the GUK1 locus (Y882, see MATERIALS AND METHODS). After sporulation of the diploid, 15 tetrads were dissected. In these tetrads, all the Leu^- spores were deregulated for expression of an ADE1-lacZ fusion and were 8AG resistant, while all the Leu⁺ spores were wild type for these phenotypes.

These results strongly suggested that bra3 mutants are alleles of GUK1, the GMP kinase encoding gene. This was further demonstrated by measuring GMP kinase activity in crude extracts from the bra3-1 and bra3-2 mutants. As expected, GMP kinase activity was found to be severely impaired in the bra3 mutants compared to the isogenic wild-type strain (Table 3), while in the same experiment AMP kinase activity used as a control was not significantly affected in the bra3 mutants. It is worth noting that the wild-type AMP and GMP kinase activity in the wild-type strain is in good agreement with those previously presented (KONRAD 1993 ).

From all these data, we conclude that BRA3 is GUK1 and therefore renamed the bra3-1, bra3-2, bra3-3, and bra3-4 alleles guk1-1, guk1-2, guk1-3, and guk1-4, respectively.

On the chromosome IV map, GUK1 is located close to the previously characterized pur5 locus (MORTIMER et al. 1991 ). This locus was initially described as associated with purine excretion, 8AA and 8AG resistance (ARMITT and WOODS 1970 ; WOODS et al. 1983 ). To test the possibility that GUK1 and PUR5 are the same gene, we transformed the pur5 mutant strain with the wild-type GUK1 gene and found that GUK1 on a centromeric plasmid complemented the purine excretion (Fig 4A) and the 8AG-resistance phenotype of pur5 (Fig 4B). Consistently, the pur5 and guk1-2 mutants did not complement for purine excretion and 8AG resistance (data not shown). Finally, we crossed the pur5 mutant with the Y882 strain in which the LEU2 marker is integrated at the GUK1 locus. After sporulation of the diploid, 17 tetrads were dissected. In these tetrads, all the Leu^- spores were 8AG resistant, while the Leu⁺ spores were 8AG sensitive (data not shown). Altogether these data show that pur5 is an allele of the GUK1 gene.

View larger version (31K): In this window In a new window Download PPT slide   Figure 4. The pur5 phenotypes are complemented by GUK1. (A) A lawn of ade1 (L587) cells was plated on purine-free SC medium. The pur5 strain (Y719) transformed with the indicated plasmids was spotted onto the ade1 lawn. Purine excretion was monitored after 4 days at 30°. (B) The pur5 strain (Y719) was transformed with a control centromeric plasmid (pRS315) or a centromeric plasmid carrying the wild-type GUK1 gene. A serial dilution of the pur5 transformants was dropped onto SC medium containing 8AG and the growth on this medium was observed after 3 days at 30°.

Utilization of adenine through the APRT route bypasses the deregulation phenotype of the guk1 mutants:
Why should a mutation in the GUK1 gene induce a deregulation of ADE gene expression in response to adenine? We have previously shown (GUETSOVA et al. 1997 ) that adenine needs to be metabolized into ADP to exert its regulatory role (see Introduction and Fig 1). Furthermore, the production of ADP from adenine preferentially occurs through the HPRT route (formation of hypoxanthine, IMP, and then AMP) rather than directly via APRT (see the Introduction section and GUETSOVA et al. 1997 ). We reasoned that if one of the reactions required for AMP synthesis from adenine is affected in the guk1 mutants, overexpression of the APT1 gene (encoding APRT) should suppress the adenine deregulation in these mutants. Indeed, the direct transformation of adenine into AMP should bypass the HPRT route (see Fig 1). Such a bypass was previously shown for a hpt1 mutant (GUETSOVA et al. 1997 ). Results in Table 4A show that overexpression of APT1 increased the regulation factor by adenine from 1.8 to 3.8 in the guk1-2 strain. As an internal control, overexpression of APT1 had no effect on regulation by hypoxanthine, a purine base that cannot be utilized through the APRT route (see Fig 1). This result suggests that the guk1 mutation affects adenine regulation via the HPRT route (Aah1p, Hpt1p, Ade12p, Ade13p). Consistently, in the guk1 mutant transformed with the control plasmid, hypoxanthine cannot repress ADE1-lacZ expression at all, while adenine has a weak repression effect.

This hypothesis was further tested in an independent approach using an aah1 mutant. The AAH1 gene, encoding adenine deaminase, participates in the transformation of adenine into AMP through the HPRT route but an aah1 mutant is not deregulated. We hypothesized that in such a mutant, more adenine is available for Apt1p, which transforms it directly into AMP (GUETSOVA et al. 1997 ; see Fig 1). Thus, if the guk1 mutation has an effect on AMP synthesis via Hpt1p, Ade12p, or Ade13p, an aah1 mutation should suppress the adenine derepression phenotype in the guk1 mutants. We tested this prediction by crossing an aah1::URA3 disrupted strain (Y663) with the guk1-2 mutant. Several tetrads were dissected from this cross and tetratypes carrying the four combinations of these two mutations were transformed with an ADE1-lacZ fusion to allow evaluation of the repression by adenine. Results presented in Table 4B show that both adenine and hypoxanthine efficiently repressed expression of the fusion in the wild-type and aah1 spores. As expected, adenine and hypoxanthine, respectively, had little or no effect on expression of the ADE1-lacZ reporter gene in the guk1-2 mutant. Finally, in the aah1 guk1-2 double mutant, adenine but not hypoxanthine repressed expression of the reporter fusion. Therefore, repression by adenine is effective in the double mutant because adenine can be metabolized directly into AMP, whereas hypoxanthine, which must be transformed into AMP via the HPRT route, cannot repress the ADE genes. We conclude that the guk1-2 mutation most probably affects the pathway between hypoxanthine and AMP.

guk1 mutations phenocopy a HGPRT defect:
In the guk1 mutants, the severe decrease in GMP kinase activity could lead to GMP accumulation, which could in turn inhibit activity of HGPRT, a key enzyme in the adenine repression process (HPT1 is BRA6, GUETSOVA et al. 1997 ). Indeed, guk1 and hpt1 mutants share several phenotypes (adenine derepression, purine excretion, and 8-azaguanine resistance).

This hypothesis was tested by monitoring the effect of overexpression of HPT1 on adenine regulation in the guk1 mutants. Results (Fig 5A) show that overexpression of HPT1 suppressed the deregulated phenotype of the guk1 mutants. Consistently, overexpression of HPT1 allowed suppression of the purine excretion phenotype of a guk1 mutant (Fig 5B). The guk1 mutation thus seems to affect Hpt1p activity and a simple explanation would be that HGPRT activity could show feedback inhibition by GMP accumulated in the guk1 mutants.

View larger version (39K): In this window In a new window Download PPT slide   Figure 5. HPT1 overexpression suppresses the pur5 phenotypes. (A) Expression of an ADE1-lacZ fusion was monitored by measuring ßGal activity in the wild-type and guk1-2 strains transformed with a control multicopy plasmid (YEpLac181) or a 2µ plasmid overexpressing HPT1 (P556). Transformed strains were grown for 6 hr in the presence or absence of adenine as indicated. (B) A lawn of ade1 (L586) cells was plated onto purine-free SC medium. The pur5 strain (Y719), transformed with either a control plasmid (YEpLac181) or a plasmid overexpressing HPT1 (P556), was spotted onto the ade1 lawn. Purine excretion was monitored after 4 days at 30°.

Feedback inhibition of HGPRT by GMP was assayed in vitro on a Hpt1p-His₆ protein. Hpt1p-His₆ was expressed and purified from E. coli and we then determined its kinetic parameters. The K_M for hypoxanthine was 17 µM, the k_cat was 5.2 sec^-1, and the estimated K_i for GMP was 26 µM (Fig 6). Therefore, GMP feedback inhibits the yeast HGPRT, further suggesting that guk1 mutations could phenocopy a HGPRT defect. This was confirmed by assaying in vivo whether hypoxanthine utilization was affected in the guk1 mutants.

View larger version (15K): In this window In a new window Download PPT slide   Figure 6. Yeast HGPRT is feedback inhibited by GMP. A Hpt1p-His6 fusion protein was purified from E. coli and assayed for hypoxanthine phosphoribosyltransferase activity. The KM and KM app for hypoxanthine were determined in the absence of GMP and in the presence of either 50, 100, or 200 µM GMP. The Ki for GMP was then calculated as described in MATERIALS AND METHODS.

First, we tested whether a double ade2 guk1 mutant would behave like an ade2 hpt1 mutant, i.e., grow on medium supplemented with adenine but not hypoxanthine (GUETSOVA et al. 1997 ). The guk1-2 mutant was crossed with an ade2 isogenic strain (L4364). After sporulation of the diploid, the tetrads were dissected. Among these, a tetratype was characterized by monitoring the ade2 mutation with adenine auxotrophy and the guk1 mutation with 8AG resistance. Results (Fig 7A) show that a double ade2 guk1-2 mutant could not utilize hypoxanthine as a purine source and that it required extracellular adenine for growth. This result confirms that a double ade2 guk1-2 mutant has properties similar to an ade2 hpt1 strain.

View larger version (32K): In this window In a new window Download PPT slide   Figure 7. In vivo hypoxanthine utilization is affected in the guk1-2 mutant. (A) Growth of an ade2 guk1-2 double mutant on various purine sources. The different strains were transformed with a CEN plasmid containing no insert (pRS315), the ADE2 gene (pASZ11, STOTZ and LINDER 1990 ), or the GUK1 gene (P195). Transformants were grown on SC medium or on the same medium supplemented with 8AG, adenine (Ade), or hypoxanthine (Hyp). Growth was monitored after 3 days. (B) [14C]Hypoxanthine accumulation in the hpt1 and guk1 mutants. Isogenic strains that were either wild type (Y350, A), guk1-3 (220, B), or hpt1 (Y508, C) were first grown in minimal medium and then transferred into the same medium containing [14C]hypoxanthine for either 15 min (left) or 90 min (right). Purine derivatives were extracted and separated by HPLC and each fraction was counted as described in MATERIALS AND METHODS. Radioactivity detected in the fractions (expressed in disintegrations per minute) is presented as a function of retention time on the column. Arrows indicate the identified peaks with the following abbreviations: Ade, adenine; Hyp, hypoxanthine; Ino, inosine; NT, mix of all purine nucleotides (mono-, di-, and triphosphate).

Second, we directly assayed whether a guk1 mutant could incorporate radiolabeled hypoxanthine into nucleotides in vivo. Results presented in Fig 7B show that guk1-3 and hpt1 mutants accumulated radiolabeled hypoxanthine. Both mutants had difficulty converting hypoxanthine into nucleotides while the isogenic wild-type strain did not. As expected, this conversion appeared less affected in the partially inactive guk1-3 mutant than in the hpt1 knock-out. We conclude that the guk1 mutations characterized in this report result in a phenocopy of a HGPRT defect.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Consequences of a GMP kinase defect in yeast:
Mutants of the bra3 complementation group were initially isolated for their ability to derepress the purine de novo pathway genes (GUETSOVA et al. 1997 ). Complementation and linkage data identified BRA3 as GUK1, the GMP kinase gene. Here, we show that pur5, a previously described purine excretion mutant (ARMITT and WOODS 1970 ), is also allelic to GUK1. Mutations at the guk1 locus lead to multiple phenotypes. Some of these are likely to be due to the impaired GDP synthesis whereas others are more probably due to the decrease in HGPRT activity resulting from GMP accumulation.

For example, the growth defect of the guk1 mutants most probably results from the GDP starvation since GUK1 is an essential gene (KONRAD 1992 ), while knock-out of HPT1 does not lead to any vegetative growth defect (GUETSOVA et al. 1997 ). On the other hand, the purine excretion and deregulation phenotypes are likely to be due to the HGPRT defect phenocopy since both phenotypes have been reported for hpt1 mutants (WOODS et al. 1983 ; GUETSOVA et al. 1997 ) and are suppressed by overexpression of HPT1 (this work). We have found (K. LECOQ and B. DAIGNAN-FORNIER, unpublished results) that the pur5 mutants excrete a mix of hypoxanthine, guanine, xanthine, inosine, and guanosine. Resistance of the pur5 mutant to 8AA but not 8AG was previously shown to be abolished when purine excretion is suppressed by the su-pur mutation (LOMAX and WOODS 1973 ). This is interpreted as follows: excreted purines compete with 8AA for uptake and lead to an apparent resistance due to inefficient uptake of the analog. In contrast, 8AG resistance is likely to be due to the GMP kinase defect or to the HGPRT phenocopy that would impair transformation of 8AG into its toxic form.

Because guanine nucleotides are involved in multiple cellular processes, it is not surprising that GMP kinase mutants show a variety of phenotypes. Our guk1 mutants could thus prove to be a useful tool to study the effect of GDP and GTP starvation in yeast. Interestingly, a mutation complemented by GUK1 was recently shown to have a defect in glycosylation, most probably because of a lack in GDP-mannose (SHIMMA et al. 1998 ). We have found that our mutants mate and sporulate normally (data not shown) even under conditions where GDP is starting to be limiting for vegetative growth.

guk1: a hpt1 phenocopy that leads to ADE gene derepression and purine excretion:
In this report, we show that mutations in the GUK1 gene cause ADE gene derepression and purine excretion. We present several lines of evidence suggesting that the guk1 mutations affect HGPRT, a key enzyme whose lack causes ADE gene derepression and purine excretion (GUETSOVA et al. 1997 ). We propose that guk1 phenocopies hpt1 as a result of HGPRT feedback inhibition by GMP, which is likely to accumulate in the GMP kinase-deficient mutants. Such a feedback inhibition is documented in this report and has already been reported for human HGPRT (HENDERSON et al. 1968 ).

Could a defect in human GMP kinase lead to purine overproduction and excretion as it does in yeast? It is well known that pathological purine overproduction and excretion lead to accumulation of extracellular hypoxanthine. This, in turn, is converted into uric acid, an insoluble compound that accumulates in joints and causes hyperuricemia and gout. A HGPRT defect in humans induces purine overproduction and excretion (KELLEY et al. 1969 ) as it does in yeast (WOODS et al. 1983 ; GUETSOVA et al. 1997 ). The finding that a defect in GMP kinase phenocopies a HGPRT defect may therefore have implications for human diseases. It is tempting to assume that a defect in human GMP kinase might be the primary cause of, or at least be implicated in, some cases of hyperuricemia. This assumption may have important consequences since all causes of hereditary gout are not yet identified and only two human genes encoding HGPRT and PRPP synthetase have been shown to cause gout when mutated (KELLEY et al. 1969 ; BECKER et al. 1973 ). The hypothesis of a GMP kinase defect in families of patients suffering from hereditary gout could be tested simply by determining the level of GMP kinase activity.

	ACKNOWLEDGMENTS

We are grateful to Drs. R. A. Woods, P. Lunjdall, M. Deeley, and G. Fink for sending strains and plasmids. The authors thank C. Desgranges, P. Gonin, I. Lascu, C. Napias, and O. Spangenberg for helpful advice. We also thank I. Belloc, F. Borne, and U. Welscher-Altschäffel for proficient technical help. This work was supported by grants from Fondation pour la Recherche Médicale, Procope 99016 European Collaboration Programme, Conseil Régional d'Aquitaine, and CNRS (UPR9026). K.L. was supported by a Ministère de l'Education Nationale, de l'Enseignement Supérieur et de la Recherche training fellowship.

Manuscript received January 22, 1999; Accepted for publication July 5, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AGARWAL, K. C., R. P. MIECH, and R. E. PARKS, JR., 1978  Guanylate kinases from human erythrocytes, hog brain and rat liver. Methods Enzymol. 51:483-490[Medline].

ARMITT, S. and R. A. WOODS, 1970  Purine excreting mutants of Saccharomyces cerevisiae. I. Isolation and genetic analysis. Genet. Res. 15:7-17[Medline].

BECKER, M. A., P. J. KOSTEL, L. J. MEYER, and J. E. SEEGMILLER, 1973  Human phosphoribosylpyrophosphate synthetase: increased enzyme specific activity in a family with gout and excessive purine synthesis. Proc. Natl. Acad. Sci. USA 70:2749-2752[Abstract/Free Full Text].

BROACH, J. R., J. N. STRATHERN, and J. B. HICKS, 1979  Transformation in yeast: development of a hybrid cloning vector and isolation of the CAN1 gene. Gene 8:121-133[Medline].

DAIGNAN-FORNIER, B. and G. R. FINK, 1992  Coregulation of purine and histidine biosynthesis by the transcriptional activators BAS1 and BAS2. Proc. Natl. Acad. Sci. USA 89:6746-6750[Abstract/Free Full Text].

DANG, V.-D., M. VALENS, M. BOLOTIN-FUKUHARA, and B. DAIGNAN-FORNIER, 1994  A genetic screen to isolate genes regulated by the yeast CCAAT-box binding protein Hap2p. Yeast 10:1273-1283[Medline].

DEELEY, M. C., 1992  Adenine deaminase and adenine utilization in Saccharomyces cerevisiae.. J. Bacteriol. 174:3102-3110[Abstract/Free Full Text].

DENIS, V., H. BOUCHERIE, C. MONRIBOT, and B. DAIGNAN-FORNIER, 1998  Role of the Myb-like protein Bas1p in Saccharomyces cerevisiae: a proteome analysis. Mol. Microbiol. 30:557-566[Medline].

GIANI, S., M. MANONI, and D. BREVIARIO, 1991  Cloning and transcriptional analysis of the ADE6 gene of Saccharomyces cerevisiae.. Gene 107:149-154[Medline].

GIETZ, R. D. and A. SUGINO, 1988  New yeast: Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534[Medline].

GONZALEZ, B., J. FRANCOIS, and M. RENAUD, 1997  A rapid and reliable method for metabolite extraction in yeast using boiling buffered ethanol. Yeast 13:1347-1356[Medline].

GUETSOVA, M. L., K. LECOQ, and B. DAIGNAN-FORNIER, 1997  The isolation and characterization of Saccharomyces cerevisiae mutants that constitutively express purine biosynthetic genes. Genetics 147:383-397[Abstract].

HENDERSON, J. F., L. W. BROX, W. N. KELLEY, F. M. ROSENBLOOM, and J. E. SEEGMILLER, 1968  Kinetic studies of hypoxanthine-guanine phosphoribosyl-transferase. J. Biol. Chem. 243:2514-2522[Abstract/Free Full Text].

KELLEY, W. N., M. L. GREENE, M. ROSENBLOOM, J. R. HENDERSON, and J. E. SEEGMILLER, 1969  Hypoxanthine-guanine phosphoribosyltransferase deficiency in gout. Ann. Intern. Med. 70:155.

KIPPERT, F., 1995  A rapid permeabilization procedure for accurate quantitative determination of ßgalactosidase activity in yeast cells. FEMS Microbiol. Lett. 128:201-206[Medline].

KONRAD, M., 1992  Cloning and expression of the essential gene for guanylate kinase from yeast. J. Biol. Chem. 267:25652-25655[Abstract/Free Full Text].

KONRAD, M., 1993  Molecular analysis of the essential gene for adenylate kinase from the fission yeast Schizosaccharomyces pombe.. J. Biol. Chem. 268:11326-11334[Abstract/Free Full Text].

LOMAX, C. A. and R. A. WOODS, 1973  A complex genetic locus controlling purine nucleotide biosynthesis in yeast. Mol. Gen. Genet. 120:139-149[Medline].

MÄNTSÄLÄ, P. and H. ZALKIN, 1984  Glutamine nucleotide sequence of Saccharomyces cerevisiae ADE4 encoding phosphoribosylpyrophosphate amidotransferase. J. Biol. Chem. 259:8478-8484[Abstract/Free Full Text].

MIROUX, B. and J. E. WALKER, 1996  Over-production of proteins in Escherichia coli: mutant hosts that allow synthesis of some membrane proteins and globular proteins at high levels. J. Mol. Biol. 260:289-298[Medline].

MORTIMER, R. K., C. R. CONTOPOULOU and J. S. KING, 1991 The molecular and cellular biology of the yeast Saccharomyces. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MYERS, A. M., A. TZAGOLOFF, D. M. KINNEY, and C. J. LUSTY, 1986  Yeast shuttle and integrative vectors with multiple cloning sites suitable for construction of lacZ fusions. Gene 45:299-310[Medline].

SHERMAN, F., G. R. FINK and J. B. HICKS, 1986 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SHIMMA, Y., A. NISHIKAWA, B. BIN KASSIM, A. ETO, and Y. JIGAMI, 1998  A defect in GTP synthesis affects mannose outer chain elongation in Saccharomyces cerevisiae.. Mol. Gen. Genet. 256:469-480.

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.. Genetics 122:19-27[Abstract/Free Full Text].

STOTZ, A. and P. LINDER, 1990  The ADE2 gene from Saccharomyces cerevisiae: sequence and new vectors. Gene 95:91-98[Medline].

WOODS, R. A., D. G. ROBERTS, T. FRIEDMAN, D. JOLLY, and D. FILPULA, 1983  Hypoxanthine:guanine phosphoribosyltransferase mutants in Saccharomyces cerevisiae.. Mol. Gen. Genet. 191:407-412[Medline].

ZHANG, F., M. KIROUAC, N. ZHU, A. G. HINNEBUSCH, and R. J. ROLFES, 1997  Evidence that complex formation by Bas1p and Bas2p (Pho2p) unmasks the activation function of Bas1p in an adenine-repressible step of ADE gene transcription. Mol. Cell. Biol. 17:3272-3283[Abstract].

This article has been cited by other articles:

				A. Breton, B. Pinson, F. Coulpier, M.-F. Giraud, A. Dautant, and B. Daignan-Fornier Lethal Accumulation of Guanylic Nucleotides in Saccharomyces cerevisiae HPT1-Deregulated Mutants Genetics, February 1, 2008; 178(2): 815 - 824. [Abstract] [Full Text] [PDF]

				H. Fang, W. Luo, J. Henkel, J. Barbieri, and N. Green A yeast assay probes the interaction between botulinum neurotoxin serotype B and its SNARE substrate PNAS, May 2, 2006; 103(18): 6958 - 6963. [Abstract] [Full Text] [PDF]

				M. Castellanos, D. B. Wilson, and M. L. Shuler A modular minimal cell model: Purine and pyrimidine transport and metabolism PNAS, April 27, 2004; 101(17): 6681 - 6686. [Abstract] [Full Text] [PDF]

				B. J. Beck, M. Huelsmeyer, S. Paul, and D. M. Downs A Mutation in the Essential Gene gmk (Encoding Guanlyate Kinase) Generates a Requirement for Adenine at Low Temperature in Salmonella enterica J. Bacteriol., November 15, 2003; 185(22): 6732 - 6735. [Abstract] [Full Text] [PDF]

				K. Rebora, C. Desmoucelles, F. Borne, B. Pinson, and B. Daignan-Fornier Yeast AMP Pathway Genes Respond to Adenine through Regulated Synthesis of a Metabolic Intermediate Mol. Cell. Biol., December 1, 2001; 21(23): 7901 - 7912. [Abstract] [Full Text] [PDF]

				M. Escobar-Henriques and B. Daignan-Fornier Transcriptional Regulation of the Yeast GMP Synthesis Pathway by Its End Products J. Biol. Chem., January 5, 2001; 276(2): 1523 - 1530. [Abstract] [Full Text] [PDF]

				R. J. Shaw, J. L. Wilson, K. T. Smith, and D. Reines Regulation of an IMP Dehydrogenase Gene and Its Overexpression in Drug-sensitive Transcription Elongation Mutants of Yeast J. Biol. Chem., August 24, 2001; 276(35): 32905 - 32916. [Abstract] [Full Text] [PDF]