Oxaloacetate Hydrolase, the C–C Bond Lyase of Oxalate Secreting Fungi*

Oxalate secretion by fungi is known to be associated with fungal pathogenesis. In addition, oxalate toxicity is a concern for the commercial application of fungi in the food and drug industries. Although oxalate is generated through several different biochemical pathways, oxaloacetate acetylhydrolase (OAH)-catalyzed hydrolytic cleavage of oxaloacetate appears to be an especially important route. Below, we report the cloning of the Botrytis cinerea oahA gene and the demonstration that the disruption of this gene results in the loss of oxalate formation. In addition, through complementation we have shown that the intact B. cinerea oahA gene restores oxalate production in an Aspergillus niger mutant strain, lacking a functional oahA gene. These observations clearly indicate that oxalate production in A. niger and B. cinerea is solely dependent on the hydrolytic cleavage of oxaloacetate catalyzed by OAH. In addition, the B. cinera oahA gene was overexpressed in Escherichia coli and the purified OAH was used to define catalytic efficiency, substrate specificity, and metal ion activation. These results are reported along with the discovery of the mechanism-based, tight binding OAH inhibitor 3,3-difluorooxaloacetate (Ki = 68 nm). Finally, we propose that cellular uptake of this inhibitor could reduce oxalate production.

Numerous filamentous fungi, including the food biotechnology fungus Aspergillus niger, the opportunistic human pathogen Aspergillus fumigatus, the phytopathogenic fungi Botrytis cinerea and Sclerotinia sclerotiorum, as well as many brown-rot and white-rot basidiomycetes, are able to efficiently produce large quantities of oxalate (1,2). It is known that oxalate secretion is associated with fungal pathogenesis (1,(3)(4)(5)(6). In the wood-rotting fungus Fomitopsis palustris oxalate is formed as the product of glucose metabolism (7). We recently initiated investigations of the oxalate biosynthetic pathway to develop a genomic-based method for distinguishing between oxalate producing and non-producing fungi. An additional goal of this effort was to identify enzyme inhibitors that could be used to arrest oxalate formation in targeted fungi.
To attenuate oxalate production in fungi, it is necessary to first identify the major pathway responsible for oxalate formation. There are three potential routes for production of oxalate in fungi: oxidation of glyoxylate (8,9), oxidation of glycolaldehyde (10), and hydrolysis of oxaloacetate (11). The results of studies of [ 14 C]CO 2 incorporation into the metabolite pools of A. niger indicate that oxalate is derived from oxaloacetate (12). This finding parallels the results of earlier work on the purification of an enzyme "oxalacetalase" (now known as oxaloacetate acetylhydrolase or OAH) 4 that catalyzes the hydrolytic cleavage of oxaloacetate to form acetate and oxalate (11). In a subsequent study, a mutant A. niger strain, NW228 (13), was found to be deficient in both oxalate production and in the synthesis of active OAH (14). These observations suggest that oxalate is produced only by the OAH catalyzed process. In the present investigation, we verified the connection between the absence of oxalate formation and the absence of OAH activity in the NW228 mutant by demonstrating that the OAH encoding oahA gene is interrupted by a stop codon. In an independent effort, Pedersen et al. (15) mutated the A. niger oahA gene by recombination with an oahA sequence-based plasmid, to create a metabolically robust A. niger strain deficient in oxalate production.
OAH isolated from A. niger is reported to be a mixture of N-terminal truncates (16). To obtain homogeneous OAH, we first cloned the A. niger oahA gene for overexpression in Escherichia coli. However, we failed to isolate active transformants. Subsequent efforts focused on cloning the oahA gene from the alternate fungal source B. cinerea (previously known as Botryotinia fuckeliana). The cloned B. cinera oahA gene was shown to both restore oxalate production in the A. niger NW228 mutant strain (13,14) that lacks a functional oahA gene and produce OAH in transformed E. coli cells. To provide insight into the mechanism of catalysis and to set the stage for the design of mechanism-based inhibitors, the OAH substrate and metal cofactor specificities were determined. Below we report the results from these studies, which have culminated in the discov-ery that 3,3-difluorooxaloacetate is a novel tight-binding OAH inhibitor.

3,3-Difluorooxaloacetate
This compound was prepared by modifying the protocol described by Saxty et al. (18). A stirred suspension of zinc powder (6.5 g, 0.1 mol) in anhydrous tetrahydrofuran (50 ml) containing Me 3 SiCl (4 ml, 0.04 mol) was stirred at room temperature for 15 min and then at reflux while ethyl bromodifluoroacetate (24.24 g, 0.12 mol) was added. To this solution was added ethyl formylformate (5.2 g, 0.05 mol) at a rate sufficient to maintain gentle reflux. After stirring at reflux for 1 h, the solution was cooled to room temperature and concentrated in vacuo. The resulting residue was dissolved in water and then extracted with ethyl acetate. The extracts were washed with water, dried, and concentrated in vacuo. The resulting residue was subjected to silica gel column chromatography (methylene chloride and 1:5 ethyl acetate/hexane) to give diethyl 2,2-difluoro-3-hydroxysuccinate (7.9 g, 69%).

A. niger OAH Gene Cloning and Sequencing
The oahA genes from the A. niger wild type strain N400 (CBS120.49), and derived mutant strains NW228 (prtF28) and NW229 (prtF29) (13), were amplified by using standard PCR protocol in conjunction with the forward primer CTGGCCCT-TCCTTTCTATC and the reverse primer CCATCCAATG-CAGTTCAAC. The PCR products were cloned using the vector pGEM-T Easy (Promega) and sequenced. The nucleotide sequence of the N400 strain has been deposited in the public data bases under accession number AJ567910.

Cloning of the B. cinerea oahA Gene
The GenomeWalker kit (Clontech) was used to obtain the PCR product of the genomic region containing the OAH encoding gene from B. cinerea strain B05. 10. The sequences of gene-specific primer 1 (ATCAACACAATATCGGAGTTCA-TGG) and primer 2 (GCACGAATTCTCATGT-AGTACTCC-TCT) were derived from B. cinerea EST AL117176. The complete oah gene along with 308 bp of the upstream region was amplified, cloned in the vector pGEM-T Easy (Promega), and sequenced. Standard reverse transcriptase-PCR techniques were used to verify the two introns. The nucleotide sequence has been deposited in the public data bases under accession number AY590264.

Complementation of the A. niger prtF Mutation
Three different constructs were made for complementation of the prtF mutation with the B. cinerea oah gene. These are the full-length B. cinerea oah gene including the 308-bp upstream region and two fusion constructs in which the A. niger pkiA (20) promoter was fused to the two possible start codons. All three constructs where co-transformed into strain NW188 (pyrA6, prtF28) as previously described (21). Transformants where screened for oxalate production by growing single colonies on complete media (22) plates with 10 ng/ml methyl orange as the pH indicator. Only the transformants in which the pkiA promoter was fused to the most upstream start codon showed oxalate production as verified by HPLC analysis. The copy number of OAH B. cinerea of the transformants was determined by Southern analysis using the pkiA promoter as a radiolabeled probe. The copy number was estimated by comparing the intensity of the native A. niger pki promoter band (1 copy) to the intensity of the band of the introduced pkiA-BcoahA construct.

Targeted Mutagenesis of B. cinerea
A gene replacement construct containing fragments originating from either end of the BcoahA gene flanking a hygromycin resistance cassette (GenBank TM accession AJ439603) was prepared. The 5Ј-flanking fragment (466 bp) was amplified using the primers CCCAATCCTCCAAGAGAAGTC and GATTACTAACAGATATCAAGGCTTCAAGCGGGAAGC-AGTGGTAC. The 3Ј-flanking fragment (611 bp) was amplified using primers GGGTACCGAGCTGCAATTCGTTGTGGA-CATCTCCAAGGC and CCAACCAGGTACTGAGATCAG. The flanking fragments were joined to the hygromycin cassette by overlap extension. The PCR mixture contained the three template fragments and primers GACTGCTACTGAGTATT-CGGT and CTACTCAAACACCATCCGCGA. The resulting PCR fragment was excised from the gel and directly transformed to B. cinerea protoplasts using the published procedure (23).

Purification of Recombinant B. cinerea OAH from the OAH-pET3c E. coli Clone
The OAH encoding gene was amplified by using a PCRbased strategy (24) with the OAH-pGem-T Easy vector clone serving as the template. Pfu DNA polymerase (Stratagene) and oligonucleotide primers containing NdeI and BamHI restriction sites were used for the subcloning the OAH gene into the pET-3c vector (Novagen). The recombinant plasmid, OAH-pET3c, was used to transform competent E. coli BL21(DE3) cells (Novagen). The transformed cells were grown at 20°C with mild agitation (180 rpm) in Luria broth containing 50 g/ml carbenicillin. After 19 h of cell growth (A 600 nm ϳ 0.7), induction was initiated with 0.4 mM isopropyl ␤-D-thiogalactopyranoside (RPI Corp.). The culture was incubated for 12.5 h at 20°C under conditions of vigorous mixing (200 rpm). The cells were harvested by centrifugation (6,500 rpm (7,808 ϫ g)) for 15 min at 4°C in a yield of 4 g/liter of culture media. The cell pellet (23 g) was suspended in 230 ml of ice-cold lysis buffer (50 mM K ϩ /Hepes (pH 7.5), 1 mM EDTA, 1 mM benzamide hydrochloride, 0.05 mg/ml trypsin inhibitor, 1 mM 1,10-phenanthroline, 0.1 mM phenylmethylsulfonyl fluoride, and 5 mM DTT). The suspension was passed through a French Press at 1,200 p.s.i., and then centrifuged at 4°C for 60 min at 20,000 rpm (48,384 ϫ g). The supernatant was fractionated by ammonium sulfateinduced protein precipitation. The 30 -40% ammonium sulfate protein precipitate was dialyzed at 4°C against 4 ϫ 2 liters of Buffer A (50 mM triethanolamine (pH 7.5), 5 mM MgCl 2 , and 5 mM DTT) before loading onto a 4.5 ϫ 45-cm DEAE-cellulose column equilibrated with 2 liters of Buffer A. The column was washed with 1 liter of Buffer A, and then eluted with a 2-liter linear gradient of 0 to 0.3 M KCl in Buffer A. The column fractions were analyzed by measuring the absorbance at 280 nm and by carrying out SDS-PAGE analysis. Ammonium sulfate was added to the combined fractions to generate 20% saturation. The resulting solution was loaded onto a 3 ϫ 30-cm Butyl-Sepharose column equilibrated at 4°C with 20% ammonium sulfate in 500 ml of Buffer A. The column was washed with 450 ml of 20% ammonium sulfate in Buffer A and then eluted with a 1-liter linear gradient of 20 to 0% ammonium sulfate in Buffer A. The column fractions were analyzed by measuring the absorbance at 280 nm and by carrying out SDS-PAGE analysis. The OAH eluted at 4% ammonium sulfate. The OAH-containing fractions were combined, concentrated at 4°C with an Amicon concentrator (Amicon), and then dialyzed against Buffer A. The resulting sample was concentrated using a MACROSEP 10K OMEGA for storage at Ϫ80°C. The protein sample was shown to be homogeneous by SDS-PAGE analysis with a yield of 4 mg of protein/g wet cell.

Recombinant B. cinerea OAH N-terminal Sequence Determination
OAH was chromatographed on a SDS-PAGE gel, transferred to a polyvinylidene difluoride membrane (Novex Co.) and subjected to automated protein N-terminal amino acid sequencing by Dr. Brian M. Martin of the National Institutes of Mental Health (Molecular Structure Unit, Department of Neurotoxicolgy, Bethesda, MD) to obtain the sequence PAYSDKVMLT.

Recombinant B. cinerea Molecular Mass Determination
The theoretical subunit molecular mass of recombinant OAH was calculated by using the amino acid composition, derived from the gene sequence, and the EXPASY Molecular Biology Server program Compute pI/MW (25). The subunit size of recombinant OAH was determined by SDS-PAGE analysis with molecular weight standards from Invitrogen. The subunit mass was determined by MS-ES mass spectrometry (University of New Mexico Mass Spectrometry Lab). The molecular weight of native recombinant OAH was first determined using gravity flow gel filtration techniques. The chromatography of OAH was carried out with a 1.5 ϫ 180-cm Sephacryl S-200 column (GE Healthcare) equilibrated with Buffer B (25 mM K ϩ /Hepes, 0.15 M KCl, 0.5 mM DTT, pH 7.5) at 4°C. The Pharmacia Gel Filtration Calibration Kit (catalase, 232,000; aldolase, 158,000; albumin, 67,000; ovalbumin, 43,000; chymotrypsinogen A, 25,000; ribonuclease A, 13,000) was used to calibrate the column according to the manufacturer's instructions. The chromatography was carried out at 4°C using Buffer B as eluant and a peristaltic pump to maintain a constant flow rate of 1 ml/min. The plot of the elution volume of the molecular weight standards versus log molecular weight was found to be linear. The OAH molecular weight was thus derived from the measured elution volume by extrapolation. OAH native mass was also analyzed at the HHMI Biopolymer/Keck Foundation Biotechnology Resource Laboratory at Yale University by size exclusion chromatography coupled with on-line laser scattering, refractive index, and ultraviolet detection.

Oxaloacetate Hydrolase Assay
OAH activity was assayed according to a published procedure (11). Reaction solutions (1 ml) contained 0.06 -1.0 mM oxaloacetate, 0.032 M OAH, and 5 mM MgCl 2 or 0.18 -1.2 mM oxaloacetate, 0.011 M OAH, and 0.3 mM MnCl 2 in 0.1 M imidazole (pH 7.6 and 25°C). The reaction was monitored at 255 nm for the disappearance of the enol tuatomer of oxaloacetate (⌬⑀ ϭ 1.1 mM Ϫ1 cm Ϫ1 ). The rate of oxaloacetate consumption via spontaneous decarboxylation was measured prior to initiating the enzymatic reaction to determine the "background rate." The background rate was subtracted from the reaction rate measured in the presence of OAH. The influence of the buffer properties on the OAH kinetic behavior was tested by replacing the 0.1 M imidazole of the reaction solutions with 50 mM K ϩ /Hepes (pH 7.5) or 50 mM Tris-HCl (pH 7.5).

Steady-state Kinetic Constant Determination for Recombinant B. cinerea OAH Substrates
The steady-state kinetic parameters (K m and k cat ) were determined from the initial velocity data measured as a function of substrate concentration. The initial velocity data were fitted to Equation 1 with KinetAsystI, where [S] is the substrate concentration, V 0 is the initial velocity, V max is the maximum velocity, and K m is the Michaelis-Menten constant for the substrate. The k cat value was calculated from V max and the enzyme concentration using the equation is the protein subunit molar concentration in the reaction calculated from the ratio of measured protein concentration and the protein molecular mass (34,486 Da).

RESULTS
Cloning and Sequencing of the oahA Gene from Oxalate Nonproducing Mutant Strains-The oahA genes from the A. niger mutant strains, prtF28 (NW228) and prtF29 (NW229), were sequenced and their sequences were compared with that of the oahA gene contained in the wild-type parental strain. Previous studies had shown the mutant strains lack both oxalate production and OAH activity (14,26). However, it remained to be demonstrated that the oahA gene is disrupted. The sequence analysis verified that the oahA genes from the mutant strains had acquired a stop codon in the open reading frame (at amino acid positions Arg-94 (NW228) or Gln-159 (NW229)). Furthermore, by using a pH indicator plate assay (Fig. 1) and HPLC analyses (data not shown) we demonstrated that the mutant strains, transformed with the wild type oahA gene, regain their oxalate production capability.
Cloning and Disruption of the B. cinerea oahA Gene-The results from the complementation experiment described above strongly suggest that oxalate production in A. niger is caused solely by OAH-catalyzed cleavage of oxaloacetate. We scrutinized the genomes of other fungal oxalate producers to see if they too encoded OAH. BLAST searches using the A. niger oahA gene (AnoahA) as the query suggested that EST AL117176 from the oxalate producer B. cinerea represents a partial copy of the B. cinerea OAH encoding gene. The complete BcoahA gene, isolated from the B. cinerea genome by using PCR and genome walking techniques, was then used for complementation of the prtF28 mutation. The sequence of the putative BcOAH was shown to be 70% identical to that of the A. niger OAH. Because there are two possible start codons separated by 7 amino acids, three different constructs were made for complementation purposes. These are the full-length B. cinerea oahA gene (including the 308-bp upstream region calculated from the most upstream candidate start codon) and two fusion constructs, in which the constitutive A. niger pkiA promoter (20) is fused to one of the two possible start codons. Complementation of the A. niger NW228 non-oxalate producing strain was achieved with the BcoahA coding region that included the most upstream start codon fused to the strong A. niger pkiA promoter (20). Seven independent pkiA::BcoahA transformants were identified by using an indicator plate assay (Fig. 2). The integration of the construct within the transformants was verified by using Southern analysis (Fig. 3). HPLC analysis was used to demonstrate that each transformant had regained the ability to produce oxalate (data not shown).
To confirm the role played by the BcoahA gene in oxalate biosynthesis in B. cinerea, the gene was deleted from the B. cinerea genome by employing targeted gene replacement. Transformants, in which homologous recombination had occurred, were identified by PCR screening and Southern blot analysis. Five independent transformants were identified as having perfect gene replacement in the absence of additional ectopic integration (data not shown). The pH indicator plate assay (Fig. 2) was used to demonstrate that the BcoahA-deficient mutants do not produce a detectable level of oxalate.
Purification of Recombinant B. cinerea OAH and Size Determination-The DNA sequence of the subcloned gene was found to agree with the published sequence (GenBank accession number AY590264) except that the nucleotide at position 1637 is G not A. Thus, the encoded amino acid is Ala not Thr. The recombinant OAH was purified to homogeneity in an overall yield of 4 mg/g wet cells by using the 4-step protocol summarized in Table 1. The N-terminal sequence of OAH revealed that the N-terminal methionine is lost during posttranslational modification. This was confirmed by a molecular mass determination using mass spectroscopy. The theoretical mass of OAH-Met is 34,355 Da compared with the experimental value of 34,355 Ϯ 1 Da. The SDS-PAGE analysis gave an estimated subunit mass of 35 kDa, whereas the native mass measured by using molecular size gel filtration chromatography is ϳ100 kDa. This result is indicative of a homotrimeric quaternary structure. The ␣,␤-barrels of the enzymes of the PEP mutase/isocitrate lyase superfamily incorporate the C-terminal ␣-helix from an adjacent subunit, a family structural trait known as "helix swapping." The functional unit is therefore a dimer, and the quaternary structure that has been most frequently encountered is the homotetramer (Ref. 17 and references therein). Because a trimeric structure is not consistent with this pattern, we examined the possibility that the association state of OAH is unstable. Native molecular mass determination, by using size exclusion chromatography, coupled with on-line laser scattering, refractive index, and ultraviolet detection provided additional information regarding the OAH quaternary structure. A sample of OAH eluted in a single peak from size exclusion fractionation was polydispersed with a molecular mass range of 50 kDa at 0.05 mg/ml to 114 kDa at 0.3 mg/ml and averaged molecular mass of 97 kDa. At 1.1 mg/ml the sample reached an averaged molecular mass of 118 Da and a hydrodynamic radius of 4.1 nm. The hydrodynamic radius remained at 4.1 nm at 3 mg/ml indicating that the protein does not exist in oligomeric forms higher than a tetramer. The results are consistent with the existence of a monomer-dimer-tetamer equilibrium.
Metal Ion Specificity of B. cinerea OAH-Enzymes of the isocitrate lyase/PEP mutase superfamily require a divalent metal ion for catalysis. The steady-state kinetic constants for metal ion activation of OAH were determined at a saturating concentration of oxaloacetate (1 mM) and varying metal ion concentration using 0.1 M imidazole (pH 7.6), 50 mM K ϩ /Hepes (pH 7.5), or 50 mM Tris-HCl (pH 7.5) as buffer. The k cat and K a values, measured for the Mg 2ϩ , Mn 2ϩ , and Ca 2ϩ activation are listed in Table 2. Mn ϩ2 and Mg 2ϩ are significantly better activators for OAH than is Ca 2ϩ . The k cat values determined using the 0.1 M imidazole buffer do not differ significantly from those measured using the K ϩ /Hepes buffer. The K a values measured using the K ϩ /Hepes buffer (which does not bind divalent metal ions) are slightly smaller than those measured using the imidazole buffer (which does bind metal ions). OAH is not subject to    Table 2. Mn 2ϩ -activated OAH displays a higher turnover rate than does the Mg 2ϩ -activated OAH. On the other hand, the K m of oxaloacetate measured by using Mn 2ϩ as activator is larger than that measured with Mg 2ϩ serving as the activator. Consequently, there is no significant difference in the specificity constant k cat /K m between Mg 2ϩ and the Mn 2ϩ -activated OAH.
The ability of OAH to catalyze C-C bond cleavage in ␣-hydroxycarboxylate metabolites (Scheme 1) was probed to determine whether OAH retains the C-C lyase activity characteristics of the C-C bond lyase branch of the isocitrate lyase/PEP mutase family (viz. petal death protein, isocitrate lyase, and 2-methylisocitrate lyase) (Ref . 17 and references therein). The results are listed in Table 3. The first set of substrates tested are the R-and S-enantiomers of malate. To enhance the sensitivity for detection of product formation, reaction mixtures containing high concentrations of OAH (13 M) were used. In addition, to ensure saturation of catalytic sites high concentrations (1 mM) of the reactants were employed. Under these conditions, no C-C lyase activity is observed, suggesting that the k cat for cleavage of malate is less than the detection limit that is 1 ϫ 10 Ϫ5 s Ϫ1 .
Next, substrate activities of C(2) alkyl malates were probed. Whereas the (2S)-methylmalate is not a substrate for OAH, the (2R)-methylmalate is converted to pyruvate and acetate with a k cat ϭ 0.01 s Ϫ1 and a K m ϭ 1.45 mM by this enzyme. The addition of a methyl group at the C(3) position of the (2R)-methylmalate leads to a further improvement in substrate activity. The K m value of (2R,3S)-dimethylmalate is 10-fold smaller than that for (2R)-methylmalate. However, the k cat values measured for these two substrates are equivalent. The substrate (2R)-ethyl-(3S)-methylmalate displayed a 10-fold larger k cat and a comparably larger K m . The k cat /K m value measured for (2R)-propyl-(3S)-methylmalate is significantly smaller (49 M Ϫ1 s Ϫ1 ), a finding that suggests that the active site has limited space for the C(2) alkyl group.
Inhibition of B. cinerea OAH-The competitive inhibition constants of oxalate, (R)-malate, (S)-malate, and 3,3-difluorooxaloacetate were evaluated to gain information about the structural determinants for ligand binding. Oxalate, a product of oxaloacetate cleavage, binds to OAH with high affinity (K i ϭ 19 Ϯ 1 M). Oxalate is also an analog of the pyruvate enolate anion intermediate formed in the cleavage of (2R,3S)-2,3-dimethylmalate. The petal death protein binds oxalate tightly (K i ϭ 4.3 Ϯ 0.3 M), as does isocitrate lyase and PEP mutase (17).
3,3-Difluorooxaloacetate, which differs from oxaloacetate by replacement of the C(3) hydrogens with fluorine atoms, is not a substrate for OAH (the k cat detection limit is 1 ϫ 10 Ϫ5 s Ϫ1 ). However, this substance is an exceptionally tight binding linear competitive inhibitor: K i ϭ 68 Ϯ 4 nM) (Fig. 4). In contrast, malate is a weak binding competitive inhibitor of OAH. Malate  differs from the oxaloacetate in that its C(2) center is tetrahedral and functionalized with a hydroxyl group. (R)-Malate (K i ϭ 2.5 Ϯ 0.2 mM) binds an order of magnitude tighter than does its S-enantiomer (K i ϭ 22 Ϯ 3 mM), but ϳ5 orders of magnitude less tightly than does 3,3-difluorooxaloacetate.

DISCUSSION
OAH Function and Distribution-The observations described above show that 1) oxalate formation in the fungi A. niger and B. cinerea derives from OAH-catalyzed hydrolytic cleavage of oxaloacetate; and 2) the respective genomes of A. niger and B. cinerea contain a single OAH encoding gene. Although, the two OAHs share 70% sequence identity, they display different behavior. The A. niger OAH is reported to associate to form high order oligomers and to be specific for Mn 2ϩ as the metal ion cofactor. Both enzymes do, however, conserve the active serine residue (Ser-260 of B. cinerea) known to contribute significantly to OAH catalytic efficiency. 5 BLAST searches, in which the A. niger OAH sequence is used as query, have led to the identification of one or more close homologs (as defined by Ͼ50% sequence identity) encoded in the A. niger genome, and in the genomes of other Aspergillus strains. The homologs of A. niger do not posses the active site serine found in authentic OAH. Also, inactivation of the OAH encoding gene results in loss of oxalate formation. Presently, their catalytic functions are unknown.
A. niger and B. cinerea are representatives of different subphyla of ascomycete fungi, separated by hundreds of millions of years of evolution. Our results strongly suggest that both fungi use OAH catalyzed hydrolytic cleavage of oxaloacetate as the main (if not the sole) route for oxalate production. Among the 22 fungi, whose genomes have been sequenced, the presence of the oah gene is strictly correlated with oxalate production. 5 The acquisition of OAH activity in a given fungal species for oxalate production appears to be important for niche adaptation.
We searched for evidence of the occurrence of OAH in organisms from other kingdoms. Earlier reports indicate that oxalate biosynthesis occurs in some plants, especially in plants of the genus Oxalis (17, 28 -31). Studies carried out with plant tissue extracts indicate that enzyme-catalyzed conversion of oxaloacetate to oxalate does take place (30). However, because of the scarcity of sequenced plant genomes, plant oah genes have not yet been identified. One exception is the gene encoding the petal death protein of the flowering plant Dianthus caryophyllus (Swiss-Prot entry Q05957). The petal death protein possesses OAH activity in addition to 2-alkylmalate C-C bond lyase activity (27). A BLAST search of the gene data bank, identified homologs of ϳ60% sequence identity in the plants Arabidopsis thaliana and Oryza sativa. However, neither of these homologs have been isolated or tested for activity and neither appear to possess an active site Ser, which is well correlated with the presence of OAH activity. 5 The presence of OAH activity has been observed in extracts of the bacterium Streptomyces cattleya (32). Unfortunately, the sequence of the gene associated with this activity has not been reported. To identify other bacterial proteins that might possess OAH activity, we carried out a BLAST search of bacterial genomes using the D. caryophyllus petal death protein and the B. cinerea OAH as query. The Bacillus cereus Swiss-Prot entry   3S and 2S,3R) Based on the results of this survey we conclude that OAH is widely used among fungi for the purpose of oxalate formation associated with niche adaptation. Moreover, OAH may also function in oxalate producing plants and perhaps in some specialized bacteria.
OAH and the petal death protein are members of the PEP mutase/ isocitrate lyase enzyme superfamily (17, 34 -40). Together with isocitrate lyase and 2-methylisocitrate lyase, they form a subgroup of enzymes of similar chemical function. Within this subgroup, the superfamily catalytic scaffold is tailored to bind and activate ␣-oxocarboxylate metabolites for C(␣)-C(␤) bond cleavage (17,(35)(36)(37)(38)(39). The ability of OAH and the petal death protein to act on C(␣)OH and C(␣)ϭO substrates is quite remarkable. This dual catalytic activity might also be possible for isocitrate lyase, but to our knowledge this has not been tested.
The catalytic mechanism that has been proposed for the isocitrate and 2-methylisocitrate lyases is depicted in Fig. 5 (39). The key elements of catalysis are 1) electron withdrawal from the C(␣) oxygen via electrostatic interaction with the Mg 2ϩ cofactor and the conserved Arg residue; 2) electron withdrawal from the acicarboxylate leaving group via hydrogen bond formation with the conserved glutamic acid residue; 3) general acid catalysis by the conserved Cys residue; and 4) general base catalysis by a yet to be assigned residue.
These elements are conserved in the petal death protein for which the active site structure is known (27). Although the x-ray structure of OAH is not yet available, it is evident from sequence alignment analysis that these same elements contribute to the OAH catalysis. Thus, the catalytic mechanism proposed for the isocitrate and 2-methylisocitrate lyases could be operative in the petal death protein and OAH-mediated cleavage reactions of the (2R)-alkylmalates. It is tempting to assume that the OAH catalytic scaffold is designed to promote reaction by way of a single catalytic mechanism (Fig. 5, pathway 1). If so, the only adjustment required for operation of a retro-Aldol type mechanism would be the hydration of oxaloacetate to generate the C(␣) gem-diol. The C(␣) gem-diol is expected to bind in the active site in the same manner as do (2R)-alkylmalates with the pro-S OH of the gem-diol located in place of the C(␣) alkyl groups of the malates. Modeling studies indicate that Ser-257 in the petal death protein is properly positioned for interaction with the pro-S OH of the oxaloacetate hydrate. 5 The corresponding residue in OAH is Ser-260. FIGURE 5. A, the catalytic mechanism of the ␣-hydroxycarboxylate lyases of the PEP mutase/isocitrate lyase enzyme superfamily. B, the four pathways leading from oxaloacetate to oxalate and acetate.