Activity of Yeast Orotidine-5′-phosphate Decarboxylase in the Absence of Metals*

Yeast orotidine-5′-phosphate decarboxylase was recently shown to contain zinc and to be inhibited by zinc-complexing agents. When the gene for the yeast enzyme was expressed inEscherichia coli, the gene product was devoid of metal atoms but exhibited a specific activity and molecular mass similar to those of the enzyme obtained directly from yeast. This invalidates the hypothesis that zinc is involved in substrate decarboxylation. The zinc-free enzyme undergoes thermal inactivation at a somewhat lower temperature than does the zinc-containing enzyme isolated from yeast.

Enzyme activity was routinely measured using a continuous assay based on the decrease in absorbance at 285 nm, where ⌬⑀ M ϭ Ϫ1650 cm Ϫ1 (6), in MOPS buffer (10 mM, pH 7.2) containing substrate OMP (0.1 mM). Because the K m value for OMP is very low, determination of k cat and K m values (Table I) required the use of radioisotopically labeled substrate following the release of 14 CO 2 at 25°C (7); reaction mixtures contained MOPS (10 mM, pH 7.2) and varying concentrations of [7-14 C]OMP (0.5-8 M). Yeast ODCase was dialyzed extensively against HEPES buffer (2 mM, pH 7.0) that had been freed of metal ions by treatment with Chelex 100. The specific activity, measured before and after dialysis, showed that no significant loss of enzyme activity occurred during dialysis.
Enzyme Expression and Purification-ODCase was expressed in yeast using strain BJ5424 containing plasmid pGU2, which contains the ura3 gene under control by a galactose-inducible promoter (2). Purification of the enzyme made by the yeast expression system was performed as described earlier (2), except for the addition of an anion exchange chromatography step to increase the purity of the product.
For expression of ODCase in Escherichia coli, the Saccharomyces cerevisiae ura3 gene of plasmid pRS306 (8) was subcloned into the EcoRV site of Stratagene vector pBCSKϩ following polymerase chain reaction amplification. This procedure yielded a ura3 gene with a unique NdeI site at the 5Ј-end and a BamHI site 3Ј to the ura3 termination codon that are compatible with unique expression vector endonuclease cleavage sites. The nucleotide sequence of the amplified ura3 gene was determined for both DNA strands using overlapping primers. To express yeast ODCase in E. coli, the NdeI, BamHI-ended ura3 coding sequence was isolated from a preparative agarose gel, purified using Jetsorb gel extraction reagents (Genomed), and inserted 3Ј to the cdd promoter of cytidine deaminase expression plasmid pCDA6022. This construction resulted in replacement of the pCDA6022 cdd gene by the ura3 gene and yielded plasmid pBGM88. ODCase expression vector pBGM88 was then introduced into E. coli SS6130 (cytR, ⌬cdd) by electroporation. In this strain, transcription of the plasmid-borne ura3 gene is completely derepressed (9).
ODCase was purified from E. coli SS6130(pBGM88) that had been grown at 27°C for 16 -18 h in 2ϫ YT medium (10) supplemented with Vogel and Bonner salts (11), 1% yeast extract, ZnCl 2 (10 M), and ampicillin (150 g/ml). The bacteria were collected by centrifugation (10,000 ϫ g for 10 min) and suspended to 1 g of cells, wet weight/4.8 ml of buffer containing potassium phosphate (50 mM, pH 6.0), glycerol (20%), and ␤-mercaptoethanol (5 mM). Cell extracts were prepared by passage of the suspension through a French press at 10,000 p.s.i. Intact cells and cell debris were removed by centrifugation at 100,000 ϫ g for 70 min. The supernatant was filtered through a 0.2-m filter and applied to an anion exchange column (Poros II Q, Perspective Biosystems) equilibrated with Buffer A (Tris-HCl (50 mM, pH 7.0), glycerol (5%, w/v), and ␤-mercaptoethanol (5 mM)). Those fractions in the column flow-through with ODCase activity were pooled and dialyzed against Buffer A for 16 h at 4°C. The dialyzed protein solution was applied to a second anion exchange column (HR 10/10 Mono Q, Amer-sham Pharmacia Biotech) that had been equilibrated with Buffer A. ODCase was recovered from the Mono Q resin by elution with a linear gradient (0 -0.3 M) of NaCl. The protein obtained from the Mono Q resin was at least 95% homogeneous as judged by SDS-polyacrylamide gel electrophoresis analysis (Fig. 2).
Thermal Stability-Before analysis of heat inactivation, enzymes were dialyzed against Tris-HCl (50 mM, pH 7.0) containing glycerol (15% w/v) and ␤-mercaptoethanol (5 mM) at 4°C. Enzymes were then incubated for 5 min in a Lauda RC-6 circulating water bath, an aliquot was removed, and the decarboxylase activity was measured using the spectrophotometric assay described above.
Zinc Binding Studies-In attempts to add zinc to metal-free enzyme made in bacteria, ODCase (9.1 M in subunits) was treated on ice with urea, guanidine HCl, or guanidine thiocyanate. Incubation with the latter two chaotropic agents resulted in an irreversible loss of activity. Following incubation of enzyme for 1 min in 2 M urea containing ZnS0 4 (2-24 M), the enzyme was diluted 50-fold into MOPS (10 mM, pH 7.2) containing equivalent concentrations of ZnS0 4 and dialyzed for 18 h at 4°C against Tris-HCl (10 mM, pH 7.0) containing glycerol (5% w/v). On average, 50% of the initial ODCase activity was recovered after dialysis. The urea/ZnSO 4 -treated enzyme was analyzed for its thermal stability and metal content by atomic absorption spectroscopy following metalfree dialysis.

RESULTS AND DISCUSSION
To allow expression of yeast ODCase in E. coli, two residues of the native yeast enzyme were replaced; the penultimate N-terminal serine was changed to histidine, and the C-terminal asparagine was changed to aspartate. These replacements showed no significant effect on the activity of the bacterially produced enzyme, compared with the values obtained for the enzyme isolated from yeast ( Table I).
Expression of the yeast ura3 gene in bacteria yielded 75-100 mg of highly purified ODCase per liter of culture and produced a fully active enzyme that contained less than 0.1 molar eq of zinc, as indicated by atomic absorption spectroscopy. In contrast, ODCase purified from yeast contained 0.89 mol of zinc/ subunit ( Table I). Analysis of concentrated samples of the bac-terially produced enzyme by inductively coupled plasma emission spectroscopy (Garratt-Callahan Co., Millbrae, CA) revealed the presence of less than 0.12 molar eq of a variety of other metals including cadmium, cobalt, copper, iron, manganese, molybdenum, nickel, and lead. These results appear to invalidate the hypothesis that metals directly participate in the mechanism of OMP decarboxylation. Amino acid sequencing of the native yeast ODCase revealed that the N terminus is blocked, presumably as a result of posttranslational modification. As expected from our previous experience with bacterial expression systems, the enzyme purified from E. coli was found to be devoid of posttranslational modifications, as judged by amino acid sequencing and electrospray mass spectrometry. SDS-polyacrylamide gel electrophoresis analysis of the yeast and bacterially generated enzymes showed that these proteins have similar electrophoretic mobilities.
On the basis of this present evidence, we do not understand why the enzymes from E. coli and from yeast, prepared by the same purification procedure, resemble each other in catalytic activity but differ in their metal content. The two-residue difference between yeast and bacterially expressed ODCase seems unlikely to be the source of the difference in metal content, because the introduced amino acids, histidine and aspartate, might be expected to assist rather than interfere with the coordination of zinc. That source of difference cannot be excluded, however, in view of the possibility that other   interactions might be affected by these replacements in such a way as to reduce the enzyme's affinity for zinc. To determine whether there might be other observable differences between enzymes obtained from the two expression systems, we determined the susceptibilities of the yeast-expressed and bacterially expressed enzymes to thermal inactivation. Fig. 3 shows that the extent of inactivation of both enzymes exhibited a sigmoidal dependence on temperature. When the inflection points were compared, the zinc-containing enzyme isolated from yeast showed a T inact of 58.0°C, whereas the zinc-deficient enzyme expressed in bacteria was markedly less stable, with a T inact of 46.2°C. The extent to which these differences might arise from differences in posttranslational modification remains to be determined. Experiments designed to interconvert these species were not successful. Removal of zinc from the enzyme expressed in yeast by dialysis in the presence of EDTA (0.01 M) resulted in loss of activity as described (3). Attempts to add ZnSO 4 to bacterially expressed ODCase, both in the absence and presence of 2 M urea, failed to produce a zinc-liganded enzyme.
Consistent with the lack of zinc in the bacterially produced enzyme, expression of the yeast ura3 gene in E. coli yielded an enzyme that, unlike the enzyme expressed in yeast, was insensitive to inactivation by 1,3-dimercaptopropanol and EDTA. However, the enzyme expressed in bacteria remained highly sensitive to reversible competitive inhibition by 6-CSNH 2 -UMP (K i ϭ 3.5 ϫ 10 Ϫ9 M), like the enzyme expressed in yeast. That seems surprising because the design of 6-CSNH 2 -UMP was predicated on the possibility that zinc, with its high affinity for sulfur ligands, might be present at the active site (7). The basis of this inhibitor's remarkable binding affinity and of the catalytic process itself remains to be determined by structural studies that are now in progress.