OUP user menu

OXP1/YKL215c encodes an ATP-dependent 5-oxoprolinase in Saccharomyces cerevisiae: functional characterization, domain structure and identification of actin-like ATP-binding motifs in eukaryotic 5-oxoprolinases

Akhilesh Kumar, Anand Kumar Bachhawat
DOI: http://dx.doi.org/10.1111/j.1567-1364.2010.00619.x 394-401 First published online: 1 June 2010


OXP1/YKL215c, an uncharacterized ORF of Saccharomyces cerevisiae, encodes a functional ATP-dependent 5-oxoprolinase of 1286 amino acids. The yeast 5-oxoprolinase activity was demonstrated in vivo by utilization of 5-oxoproline as a source of glutamate and OTC, a 5-oxoproline sulfur analogue, as a source of sulfur in cells overexpressing OXP1. In vitro characterization by expression and purification of the recombinant protein in S. cerevisiae revealed that the enzyme exists and functions as a dimer, and has a Km of 159 μM and a Vmax of 3.5 nmol h−1 μg−1 protein. The enzyme was found to be functionally separable in two distinct domains. An ‘actin-like ATPase motif’ could be identified in 5-oxprolinases, and mutation of key residues within this motif led to complete loss in ATPase and 5-oxoprolinase activity of the enzyme. The results are discussed in the light of the previously postulated truncated γ-glutamyl cycle of yeasts.

  • glutathione
  • 5-oxoprolinase
  • 5-oxoproline
  • γ-glutamyl cycle
  • actin-like ATPase domain
  • OXP1


The γ-glutamyl cycle that describes the synthesis and degradation of glutathione consists of six enzymatic steps and has been shown to be present in both mammals and plants (Meister, 1973; Mazelis & Creveling, 1978). In yeasts, biochemical studies directed toward investigating the role of the γ-glutamyl cycle concluded that it was incomplete in yeast as the activities of the last two enzymes, γ-glutamyl cyclotransferase (γ-GCT) and 5-oxoprolinase could not be detected (Jaspers, 1985). The enzyme γ-GCT cyclizes γ-glutamyl amino acids to form 5-oxoproline (a dehydrated cyclized form of glutamate), while 5-oxoprolinase hydrolyzes 5-oxoproline to yield glutamate (Connell & Hanes, 1956; Van der Werf, 1971). Although a yeast ORF (YKL215c) homologous to the rat 5-oxoprolinase has been identified in the Saccharomyces cerevisiae genome, a functional activity has not been demonstrated for the encoded protein (Ye, 1996). In the case of γ-GCT, homologues of the mammalian enzyme seem to be absent in S. cerevisiae (Oakley, 2008). In the absence of a source of 5-oxoproline, a role for 5-oxoprolinase in yeasts thus has seemed unlikely. A fungal-specific glutathione degradation pathway that has been discovered in S. cerevisiae has been thought to compensate for the truncated nature of the γ-glutamyl cycle in yeasts (Ganguli, 2007). However the recent detection of significant levels of 5-oxoproline in yeast in a metabolomic study (Mohler, 2006) and the observation that 5-oxoproline has also been shown to play a role as a secondary messenger in the upregulation of amino acid transporters in mammals, and as an osmoprotectant in certain bacteria (Lee, 1996; Trotsenko & Khelenina, 2002), have prompted us to initiate investigations into this protein and to determine whether YKL215c/OXP1 encodes a functional 5-oxoprolinase. Our studies, facilitated by development of a simple plate-based in vivo assay and our ability to purify for the first time recombinant eukaryotic 5-oxoprolinase, reveal that YKL215c/OXP1 encodes a functional 5-oxoprolinase that has activity and kinetics comparable to the mammalian 5-oxoprolinase. We have also initiated the first structure–function studies on 5-oxoprolinases and we demonstrate that 5-oxoprolinases contain two separable domains and have identified the ‘actin-like ATPase motifs’ responsible for ATPase activity in this enzyme.

Materials and methods

All reagents were of analytical grade. Glutamate/glutamate oxidase assay kit was obtained from Invitrogen. Malachite green phosphate assay kit was obtained from BioChain. Protein molecular weight markers were obtained from MBI Fermentas. Molecular weight standards for gel filtration were purchased from Amersham Biosciences. 5-Oxoproline was obtained from Sigma. Oligonucleotide primers were synthesized from Biobasic Inc., Canada, and Sigma (Bangalore, India). Medium components were purchased from BD (Difco). Restriction enzymes, Vent DNA polymerase and other DNA-modifying enzymes were obtained from NEB. Plasmid miniprep kit, gel extraction kit and Ni-NTA agarose resin were obtained from Qiagen. Dithiothreitol was obtained from USB.

Strains, growth media and growth assays

Escherichia coli strain DH5α was used as the cloning host. Yeast strains ABC733 (BY4741) (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0), MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 ykl215cΔ∷kanMX4 and ABC1603 (aco1Δ) (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0 YLR304ckanMX4) were obtained from Euroscarf. Yeast media, handling of yeast, bacteria and all the molecular techniques used in the study were according to standard protocols. Growth assays were carried out by serial dilutions on different media as described earlier (Kaur & Bachhawat, 2009).

Cloning of yeast 5-oxoprolinase and different domains

YKL215c was initially cloned in p416TEF yeast expression vector at Spe1 and Xho1 by Bangalore Genei, India. For the purification of recombinant yeast 5-oxoprolinase from S. cerevisiae, YKL215c was subcloned from p416TEF vector to p426GPD vector. N-terminal His-8 tagging was carried out by amplifying a part of the gene using YKLSpe1HisF forward and YKLEcoR1-1369R reverse primers and subcloning in the p42GPD-YKL215c at Spe1 and EcoR1 sites.

For cloning of the HyuA and HyuB domains of yeast 5-oxoprolinase, the N-terminal 2.208-kb part of the yeast 5-oxoprolinase was amplified using HyuASpe1F and HyuAXho1R primers and cloned at SpeI and XhoI sites of the pTEF416 yeast expression vector, while the C-terminal part of the yeast 5-oxoprolinase was cloned by amplifying 1.632 kb using HyuBBamh1F and HyuBXho1R primers and cloning at BamHI and XhoI sites of the pTEF413 yeast expression vector. The oligonucleotides used here are shown in Supporting Information, Table S1.

Construction of putative ATPase motifs mutants

Site-directed mutagenesis was performed using the splice overlap extension method. The final PCR products were subcloned in 8 × His-tagged p426GPD-YKL215c using appropriate restriction sites. The mutants were sequenced to confirm the presence of the desired nucleotide changes and to rule out any undesired mutations introduced during the mutagenic procedure. Primers used for mutagenesis and sequencing are given in Table S1.

Purification of recombinant 5-oxoprolinase from yeast

Yeast cells harboring p426GPD-YKL215c were grown to one OD600 nm in 1 L of synthetic defined medium. Cells were harvested at 7000 g for 7 min and pellet was resuspended in lysis buffer (50 mM Tris-HCl, pH 8, 300 mM NaCl, 20% glycerol, 0.5 mM 5-oxoproline, 100 μM dithiothreitol and 2 mM PMSF). Cells were lysed by the glass bead lysis method (20 cycles of on and off) at 4 °C. The lysed cell suspension was centrifuged at 17 000 g for 15 min at 4 °C to separate the cell debris and the supernatant was recovered. The yeast 5-oxoprolinase protein with the 8 × His tag at the N-terminus was purified using Ni-NTA affinity chromatography. The protein sample was loaded onto the Ni-NTA column and column was washed with 20 column volumes of the wash buffer (50 mM Tris-HCl, pH 8, 300 mM NaCl, 20% glycerol, 0.5 mM 5-oxoproline, 100 μM dithiothreitol and 30 mM imidazole) and the protein eluted with elution buffer (50 mM Tris-HCl, pH 8, 300 mM NaCl, 20% glycerol, 0.5 mM 5-oxoproline, 100 μM dithiothreitol and 250 mM imidazole). The purified protein was dialyzed against buffer 50 mM Tris-HCl, pH 8, 300 mM NaCl and 20% glycerol, 0.5 mM 5-oxoproline and 100 μM dithiothreitol. Yeast 5-oxoprolinase was further purified by gel filtration chromatography (Superdex 200) that was equilibrated with 50 mM Tris-HCl, pH 8, 300 mM NaCl, 50 μM dithiothreitol and 20% glycerol. The elution was monitored by running different fractions on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) followed by Coomassie brilliant blue R-250 staining.

Enzyme assays

5-Oxoprolinase activity was assayed by measuring glutamate using the sensitive Amplex red-based glutamate/glutamate oxidase fluorescent assay that we have modified to make it applicable for assay of 5-oxoprolinase. The reaction mixture of the enzyme reaction (total volume 50 μL) typically consists of 50 mM Tris-HCl (pH 8), 5 mM ATP, 10 mM MgCl2, 150 mM KCl and 0.5 mM 5-oxoproline. The reaction was initiated by addition of 1 μg of purified protein. 5-Oxoproline concentration was varied from 10 μM to 1 mM for enzyme kinetics studies. Incubation was carried out routinely for 1 h at 37 °C. The reaction mixture was then placed at 99 °C for 3 min to inactivate the enzyme. After inactivation, 50 μL of glutamate assay kit solution was added in the reaction mixtures and incubated for 30 min at 37 °C. The reaction mixture was then transferred to fluorescent 96-well plates and the fluorescence was measured using a fluorescence microplate reader with an excitation at 544 nm and an emission at 584 nm.

ATPase activity was assayed using a malachite-based phosphate colorimetric assay kit obtained from BioChain. Readings were taken at 620 nm using ELISA plate reader.


YKL215c/OXP1 overexpression cleaves 5-oxoproline and l-2-oxothiazolidine-4-carboxylic acid (OTC) in vivo

Comparison of the growth of ykl215cΔ deletion strains with wild-type strains (YKL215c) in a variety of growth conditions did not yield any discernible phenotypes (data not shown). YKL215c/OXP1 was therefore cloned downstream of the constitutive GPD promoter in the multicopy yeast vector p426GPD. This construct was transformed in an S. cerevisiae aco1Δ strain that is a glutamate auxotroph. 5-Oxoproline was provided as a source of glutamate. Cells expressing OXP1 were able to grow on 5-oxoproline, although the growth was very slow and visible growth could be seen only after 10 days of incubation (Fig. 1a). This suggested that OXP1 was encoding a functional enzyme. The slow growth could be a consequence of either poor transport of 5-oxoproline or, alternatively, limited activity of 5-OPase, or both, leading to inadequate levels of glutamate being formed. We also examined growth on OTC, a sulfur-containing analogue of 5-oxoproline (Williamson & Meister, 1982b). 5-Oxoprolinase cleaves OTC to yield cysteine, which can then be efficiently utilized as a sulfur source by S. cerevisiae (Kaur & Bachhawat, 2007). Using a met15ΔS. cerevisiae stain, which is an organic sulfur auxotroph, we observed that OXP1 could enable efficient utilization of OTC as sole source of sulfur (Fig. 1b). This result further confirmed that OXP1 encodes a functional 5-oxoprolinase.

Fig. 1

Overexpression of OXP1 allows utilization of (a) 5-oxoproline as a source of glutamate in the glutamate auxotrophic yeast strain (aco1Δ); and (b) utilization of OTC as an organic sulfur source in organic sulfur auxotroph yeast strain (met15Δ). The respective strains bearing either OXP1 or vector alone were harvested, washed and resuspended in water and serially diluted to give 0.2, 0.02, 0.002 and 0.0002 OD600 nm of cells. A 10-μL aliquot of each dilution was spotted on minimal medium containing (a) 0.2 mM glutamate and 10 mM 5-oxoproline (photographs taken after 10 days), and (b) 0.2 mM methionine and 5 mM OTC (photographs taken after 3 days of growth).

Purification and kinetic properties of recombinant yeast 5-oxoprolinase

Initial attempts at expressing recombinant Oxp1p from E. coli were unsuccessful (data not shown), so we examined the native host, S. cerevisiae, for expression. The N-terminally His-tagged S. cerevisiae 5-oxoprolinase was purified from crude extracts by passing through Ni-NTA columns followed by gel filtration chromatography with the help of S200 Superdex column (as described in Materials and methods). The protein obtained was >95% pure as seen on Coomassie-stained SDS-PAGE and was of the expected subunit size (140 kDa) (Fig. 2a). The enzyme existed as a dimer in solution, as the molecular weight determined in gel filtration studies was 280 kDa (Fig. 2b).

Fig. 2

(a) Purified recombinant Oxp1p-His8 using Ni-NTA chromatography. Lane 1: molecular weight marker, Lane 2: Ni-NTA purified fraction, Lane 3: gel-filtered fraction. (b) Gel filtration analysis. Molecular weight standard graph prepared using different proteins of known molecular weight obtained from Amersham Biosciences (ferritin, 440 kDa; β-amylase, 200 kDa; alcohol dehydrogenase, 150 kDa and BSA, 66 kDa). The arrow indicates the elution volume and size of native Oxp1p. (c) Michaelis–Menten plot for 5-oxoproinase against the substrate 5-oxoproline. Gel-filtered fraction was used to determine the kinetic parameters that were carried out using a saturating concentration of ATP (5 mM) and different concentrations of 5-oxoproline ranging from 10 μM to 1 mM. Data were analyzed using nonlinear regression. Each data point represents the mean value of three independently assayed samples.

Mammalian 5-oxoprolinases have been shown to be sulfhydryl enzymes, as they are inactivated in the presence of thiol-reacting reagents and also require dithiothreitol for its optimum activity (Williamson & Meister, 1982a). We observed that the yeast 5-oxoprolinase activity was inactivated in the presence of thiol-reacting reagents NEM, DTNB and AgNO3 but were not affected significantly by the absence of dithiothreitol. This suggested that although the yeast enzyme is also a sulfhydryl enzyme similar to mammalian 5-oxoprolinase, it had far less sensitivity to the absence of dithiothreitol in the buffers. We also observed that yeast 5-oxoprolinase significantly stabilized in the presence of 20% glycerol and that the presence of 5-oxoproline and dithiothreitol in the purification and storage buffers only contributed minimally toward the stability in contrast to glycerol (data not shown).

Yeast 5-oxoprolinase required ATP, a monovalent cation (K+/NH4+) and a divalent cation (Mg2+/Mn2+) for its 5-oxoprolinase activity, similar to the mammalian enzyme, as the absence of any of these components led to no 5-oxoproline hydrolysis (data not shown) (Van der Werf, 1971). Thus the yeast 5-oxoprolinase is also an ATP-dependent amidohydrolase.

To determine the kinetic parameters, yeast 5-oxoprolinase was assayed in the presence of a saturating concentration of ATP (5 mM), 10 mM of MgCl2, 150 mM KCl and various concentrations of 5-oxoproline ranging from 10 μM to 1 mM. The enzyme displayed Michaelis–Menten kinetics (Fig. 2c). The kinetic parameters of 5-oxoprolinase with respect to substrate 5-oxoproline were determined and found to have a Km of 159 μM and a Vmax of 3.5 nmol h−1 μg−1 protein.

Eukaryotic 5-oxoprolinase contains two separable domains

The sequence of eukaryotic 5-oxoprolinases (both yeast and mammalian) reveals two distinct domains (HyuA and HyuB) in the enzyme. In the case of yeast 5-oxoprolinase, the N-terminal 736 amino acids showed similarity to hydantoinase A (HyuA), whereas the C-terminal 543 amino acids were similar to hydantoinase B (HyuB) (Fig. 3). To investigate whether these putative domains are functionally separable, the HyuA domain and the HyuB domain were cloned separately in yeast expression vectors p416TEF and p413TEF, respectively. These constructs were transformed either singly or together in the met15Δ and growth on OTC as a sole source of sulfur was examined. We observed that cells expressing either the HyuA or the HyuB domains alone could not utilize OTC. However, cells coexpressing both HyuA and HyuB domains were able to grow on OTC as a sole source of sulfur, similar to cells expressing full-length yeast 5-oxoprolinase gene (Fig. 4). This result clearly indicates that eukaryotic 5-oxoprolinase contains two domains which can be separated.

Fig. 3

A schematic diagram showing the domain structure of yeast 5-oxoprolinase and an alignment of the relevant regions of Oxp1p with bovine Hsc70 to indicate the motifs and residues involved in ATP hydrolysis in the ‘actin-like ATPase domain’. The residues shaded grey have been mutated.

Fig. 4

5-Oxoprolinase contains two separable domains. Constructs pTEF416-HyuA and pTEF413-HyuB were cotransformed or transformed along with vectors in the met15Δ strain of Saccharomyces cerevisiae as indicated in the figure. The functionality of these constructs was checked using OTC-based in vivo assay with the help of serial dilution spotting as described earlier.

5-Oxorolinasecontain ‘actin-like ATPase motifs’ responsible for its ATPase activity

Examination of the 5-oxoprolinase sequence for possible motifs or residues that may be involved in ATP binding and hydrolysis revealed the presence of a phosphate-binding loop (P-loop) or Walker A motif from residues 21 to 27, GNIGTGK, which could be responsible for the ATP binding and hydrolysis, as it corresponded to the consensus P-loop motif (GXXGXGK) (Moller & Amons, 1985). However, mutation of this putative P-loop motif by creation of a G26A K27A double mutant did not affect the functionality of yeast 5-oxoprolinase (based on in vivo functionality assays, Fig. S1), suggesting that this putative P-loop motif was unlikely to be involved in the ATPase activity of 5-oxoprolinase.

The N-terminal half of 5-oxoprolinase also revealed the presence of an actin-like ATPase domain. This domain has been observed in proteins of different function and sequence such as actin, HSP70 and hexokinase, and these share a similar fold and three-dimensional structure in relation to ATP binding (Kabsch & Holmes, 1995). Moreover, it has been found that despite the absence of any significant similarity in their primary sequences, they all contain five conserved motifs: Phosphate1 (DXG), Connect1 (Q/AXXXS/A), Phosphate2 (DXGXG/T), Adenosine (GG) and Connect2 (G). The motif sequences, and the distance between these motifs, appeared to be more or less conserved in these members (Bork, 1992). Alignment of the N-terminal half of Oxp1p with the actin fold ATPase part of bovine HSc70 revealed that the Oxp1 protein contains all the above signature motifs except Connect1 motif (Q/AXXXS/A), although HSc70 and Oxp1p do not share any significant similarity in their primary sequences (Figs 3 and S2). To discover whether these motifs in Oxp1p are really involved in the ATPase activity of 5-oxoprolinase, the D residues of Phosphate1 (DXG), and Phosphate2 (DXGXG/T), and both the G residues of Adenosine (GG) motifs were mutated to alanine. These mutants were evaluated in vivo. Mutation of the aspartyl (D) residue in either Phosphate motif resulted in complete loss of in vivo 5-oxoprolinase activity. However, the G to A mutations in the Adenosine motif led to only a partial loss of in vivo activity (Fig. 5). Previous structural studies on the ‘actin fold ATPase superfamily’ proteins have revealed that the charged aspartyl residues of the phosphate motifs are involved in interactions with the metal phosphate complex and the glycine residues of these motifs are conserved to avoid steric clashes between ATP and the side chains of amino acids (Bork, 1992). We therefore also created a mutant where we mutated the first glycine (of the Adenosine motif) to alanine and the second glycine residue of the Adenosine motif to leucine (GG to AL). This mutant was completely nonfunctional and in agreement with structural studies on these motifs where a longer side chain amino acid (like leucine) in the Adenosine-binding motif is not tolerated (Fig. 5). These results strongly indicated that the 5-oxoprolinase contains an actin-like ATPase motif.

Fig. 5

(a) Mutations in ‘actin-like ATPase’ motifs led to complete loss of activity. met15Δ yeast strain was transformed with vector, wild-type gene and its different mutants were assayed using an OTC-based in vivo assay with the help of serial dilution spotting.

To further confirm this observation, both the ATPase activity and the 5-oxoprolinase activity of the different mutants were evaluated in vitro after purification of the different mutant proteins. As shown in Fig. 6, mutants of Phosphate1 and Adenosine motifs completely lost ATPase activity; however, Phosphate2 motif retained nearly 20% activity of wild type, whereas all the mutants lacked 5-oxoprolinase activity.

Fig. 6

Assay of ATPase and 5-oxoprolinase activity of different mutants. Ni-NTA purified mutants were dialysed, and assayed for (a) ATPase and (b) 5-oxoprolinase activity. One microgram of each protein was added in 5-oxoprolinase reaction mixture. Experiments in both cases were performed twice in triplicate independently; one of the representative data is shown.

To rule out the possibility of misfolding and inactivation of the HyuA domain ATPase motif mutants, these mutants were coexpressed with a functionally inactive HyuB domain mutant of yeast 5-oxoprolinase in which two C-terminally conserved residues were converted into alanine (H867AG872A), leading to loss of functionality in vivo. As can be seen from Fig. 7, ATPase activity-deficient mutants were able to cleave OTC in the presence of C-terminal half mutant, suggesting that the proper folding of the ATPase motif mutants was still occurring, as it allowed the intra-allelic complementation and the formation of a functional homodimer.

Fig. 7

A functionally inactive HyuB domain mutant of 5-oxoprolinase complements the loss of the activity phenotype of HyuA domain ATPase motif mutants. Different nonfunctional ATPase motif mutants that are restricted in HyuA domain of 5-oxoprolinase were either cotransformed with a nonfunctional HyuB domain mutant (H867A-G872A) or singly transformed and their functionality examined using an OTC-based in vivo assay with the help of serial dilution spotting.

The N-terminal 536 amino acids of Oxp1p were also modeled using the phyre server (http://www.sbg.bio.ic.ac.uk/phyre/) and produced a structure similar to the actin fold proteins (Kelley & Sternberg, 2009). Moreover the motifs involved in ATP hydrolysis were observed to form the ATP-binding pocket similar to the actin fold proteins (Fig. 8) (Kabsch, 1990).

Fig. 8

Homology model of actin-like ATPase domain of Oxp1p. N-terminal 536 amino acids of Oxp1p were modeled using the phyre server. The ATP-binding pocket is shown. Residues D11, D423 and G498 lining the ATP-binding pocket similar to actin are highlighted by ball and stick representation.


The studies described in this article demonstrate that YKL215c/OXP1, an uncharacterized ORF of S. cerevisiae, encodes functional 5-oxoprolinase, an enzyme previously considered to be absent owing to a truncated γ-glutamyl cycle being postulated in this yeast (Jaspers, 1985). This follows our recent identification of Dug1p as a Cys-Gly peptidase, another enzyme of the degradative part of the γ-glutamyl cycle in yeasts that had not been previously identified (Kaur, 2009). We also provide the first structure–function insights into 5-oxoprolinases, by demonstrating, first, that this large protein is made up of two separable domains. Secondly, we have identified and demonstrated that the ATPase domain of eukaryotic 5-oxoprolinases is formed through an actin-like ATPase fold in the N-terminal region.

Surprisingly, this is the first description of recombinant expression of ATP-dependent 5-oxoprolinase despite the cloning of the mammalian gene several years ago (Ye, 1996). Previous attempts and our current attempts to use E. coli as an expression host were unsuccessful. Recombinant expression in yeast allowed reasonable levels of recombinant protein facilitating studies on this interesting enzyme. This is particularly important in the light of 5-oxoproline now being recognized as playing a role beyond that of a simple intermediate metabolite of the γ-glutamyl cycle (Lee, 1996; Trotsenko & Khelenina, 2002). 5-Oxoprolinases are also important clinically, as deficiency of this enzyme leads to overaccumulation of 5-oxoproline (5-oxoprolinuria) and metabolic acidosis (Ristoff & Larsson, 2007) and has recently been proposed to play a role in being part of a futile ATP-depleting cycle leading to ATP depletion, a key event in the pathogenesis of nephrotic cystinosis (Kumar & Bachhawat, 2010).

The inability to detect any phenotypes with OXP1 deletion strains seems to be a consequence of the enzyme being expressed at low levels under normal growth conditions. We confirmed this using promoter fusions to a β-galactosidase reporter where barely any activity was observed above background (data not shown). However, OXP1 overexpression clearly displayed phenotypes and the OTC growth assay should assist further structure–function studies of this protein.

The demonstration of a 5-oxoprolinase in S. cerevisiae indicates that this enzyme, earlier thought to be absent in yeasts, may be ubiquitous in eukaryotes (Jaspers, 1985). Most prokaryotes lack homologues of the ATP-dependent 5-oxoprolinase. Interestingly, though, the prokaryotic 5-oxoprolinase studied from Pseudomonas putida has two different protein components, A and B, involved in 5-oxoproline hydrolysis. Component A phosphorylates the enzyme-bound 5-oxoproline, but component B is required to hydrolyze the phosphorylated 5-oxoproline along with component A (Seddon & Meister, 1986). In this work, we have shown for the first time that eukaryotic 5-oxoprolinase contains two separable domains that are likely to correspond to the separate components A and B of P. putida. The Hyu A domain of the yeast 5-oxoprolinase clearly seems to be responsible for the ATPase activity but more studies are needed to evaluate how these domains combine together to carry out the catalysis.

The yeast 5-oxoprolinases displayed a Km for 5-oxoproline that was threefold higher than the mammalian enzyme. The results are interesting in the light of the apparent absence of γ-GCT in yeast. Some intriguing questions that are immediately raised are, how, in the absence of γ-GCT, 5-oxoproline might be generated in the yeast cell, and secondly, the possible role that 5-oxoproline might play in the physiology of this yeast. The studies described in this report should greatly facilitate further studies on 5-oxoprolinases and in the area of 5-oxoproline metabolism, an extremely unexplored area of research.

Supporting Information

Fig. S1. OXP1 mutant G26K27-A26A27 does not lead to loss in activity.

Fig. S2. Multiple sequence alignment of N-terminal part of Ykl215cp (Oxp1p) and bovine HSc70 revealed actin-like ATPase motifs.

Table S1. Oligomer used for cloning, site-directed mutagenesis and sequencing.


We thank Purva Vats, Siddharth Jaitley and Rajkumar for help with some of the experiments. We acknowledge Manish Datt for the model prediction. A.K. is the recipient of a Senior Research fellowship from CSIR, India. The work was partly funded by a grant-in-aid from the Department of Science and Technology, Government of India.


  • Editor: Ian Dawes


View Abstract