PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Biochemistry. Author manuscript; available in PMC 2010 June 23.
Published in final edited form as:
PMCID: PMC2697262
NIHMSID: NIHMS117285

Mechanism of the Orotidine 5’-Monophosphate Decarboxylase-Catalyzed Reaction: Evidence for Substrate Destabilization

Abstract

The reaction catalyzed by orotidine 5’-monophosphate decarboxylase (OMPDC) involves a stabilized anionic intermediate, although the structural basis for the rate acceleration (kcat/knon, 7.1 × 1016) and proficiency [(kcat/KM)/knon, 4.8 × 1022 M−1] is uncertain. That the OMPDCs from Methanothermobacter thermautotrophicus (MtOMPDC) and Saccharomyces cerevisiae (ScOMPDC) catalyze the exchange of H6 of the UMP product with solvent deuterium allows an estimate of a lower limit on the rate acceleration associated with stabilization of the intermediate and its flanking transition states (≥ 1010). The origin of the “missing” contribution, ≤ 107 (~1017 total - ≥ 1010), is of interest. Based on structures of liganded complexes, unfavorable electrostatic interactions between the substrate carboxylate group and a proximal Asp (Asp 70 in MtOMPDC and Asp 91 in ScOMPDC) have been proposed to contribute to the catalytic efficiency [Wu, N., Mo, Y., Gao, J., and Pai, E. F. (2000) Proc Natl Acad Sci U S A 97, 2017−2022]. We investigated that hypothesis by structural and functional characterization of the D70N and D70G mutants of MtOMPDC and the D91N mutant of ScOMPDC. The substitutions for Asp 70 in MtOMPDC significantly decrease the value of kcat for decarboxylation of FOMP (a more reactive substrate analog) but have little effect on the value of kex for exchange of H6 of FUMP with solvent deuterium; the structures of wild type MtOMPDC and its mutants are superimposable when complexed with 6-azaUMP. In contrast, the D91N mutant of ScOMPDC does not catalyze exchange of H6 of FUMP; the structures of wild type ScOMPDC and its D91N mutant are not superimposable when complexed with 6-azaUMP, with differences in both the conformation of the active site loop and the orientation of the ligand vis á vis the active site residues. We propose that the differential effects of substitutions for Asp 70 of MtOMPDC on decarboxylation and exchange provide additional evidence for a carbanionic intermediate as well as the involvement of Asp 70 in substrate destabilization.

As Nature's most proficient enzyme, orotidine 5’-monophosphate decarboxylase (OMPDC) has attracted considerable attention amongst mechanistic enzymologists (Scheme 1). In 1995, Wolfenden quantitated the extraordinary rate acceleration, 1.4 × 1017 (kcat = 39 sec−1; knon = 2.8 × 10−16 sec−1), and proficiency, [(kcat/KM)/knon], 2.0 × 1023 M−1 (kcat/KM = 5.6 × 107 M−1 sec−1), for the OMPDC from Saccharomyces cerevisiae (ScOMPDC) (2).2 Since that time, the challenge has been to understand the catalytic prowess in terms of active site structure (3). However, despite the availability of high resolution structures and limited kinetic data for mutant enzymes (vide infra), little progress has been made in achieving that goal.

Despite significant sequence divergence, the active site structures of OMPDCs are highly conserved (4). OMPDCs are obligate dimers of (β/α)8-barrels, with the active sites located at the dimer interface. Four divergent subfamilies of OMPDCs can be identified in the sequence databases that differ in the sequence and structure of a loop at the end of the seventh β-strand that closes when substrate binds, “clamping” the substrate within the solvent-sequestered active site (J. A. Gerlt, P. C. Babbitt, and S. Brown, unpublished observations).

In 2000, structures were reported for four OMPDCs (5-8), including ScOMPDC and the OMPDC from Methanothermobacter thermautotrophicus (formerly Methanobacterium thermoautotrophicum; MtOMPDC). Subsequently, Wolfenden reported kinetic studies of mutants of ScOMPDC (9-11), although the only structures of ScOMPDC have been those of the wild type enzyme in the absence and presence of the competitive inhibitor barbiturate 5’-monophosphate (BMP, an analogue of a carbanionic transition state/intermediate with negative charge localized proximal to C6 of the pyrimidine moiety; Scheme 2) (5). Pai reported structures of wild type MtOMPDC as well as of several mutants of active site residues in the absence and presence of either BMP or 6-azaUMP (also an analogue of a carbanionic transition state/intermediate; Scheme 2), although the kinetic constants of the mutants were not reported (6, 12). In virtually all of the liganded structures that have been reported, C6/N6 of the pyrimidine base is proximal to a hydrogen-bonded Asp-Lys-Asp triad. The active site structure of the 6-azaUMP complex of MtOMPDC (vide infra) is shown in Figure 1 (panel A, front view; panel B, side view, with Asp 70, Lys 72, and Asp 75 (from the neighboring polypeptide in the biological dimer) forming the catalytic triad.3 Lys 72 is thought to deliver a solvent-derived hydrogen to C6 to form the UMP product. But, the roles of the Asp residues are both uncertain and controversial (13, 14), although early speculation focused on destabilization of the OMP carboxylate group by unfavorable electrostatic interactions (6, 15).

Figure 1
Active site of wild type MtOMPDC complexed with 6-azaUMP. Panel A, front view, with Asp 75 from the second polypeptide marked with an asterisk. Panel B, side view.

Although several mechanisms have been proposed, we recently established that the mechanism is stepwise, involving the formation of a stabilized vinyl carbanion intermediate (Scheme 1): 1) no discrimination between solvent protium and deuterium is observed in the formation of the UMP product (16); and 2) H6 of the UMP product undergoes slow exchange with solvent deuterium (17). The rate of exchange of H6 of the UMP product with solvent deuterium can be used to estimate its pKa within the active site. In solution, the pKa of 1,3-dimethyluracil is estimated as 34 ± 2 (18) and 30 (19); in the active sites of both ScOMPDC and MtOMPDC, the pKa of UMP is ≤ 22 (17). The difference establishes that the active site stabilizes the vinyl carbanion by ≥ 14 kcal/mol (producing a rate acceleration ≥ 1010). The structural features that provide this stabilization remain uncertain: the proximity of the protonated Lys that delivers the solvent-derived hydrogen to C6 to generate UMP provides electrostatic stabilization, but additional stabilization by interactions of O4 of a carbene resonance structure with active site hydrogen bond donors has been proposed (20, 21) but not experimentally investigated.

The simultaneous reports of the structures of ScOMPDC (5), MtOMPDC (6), and the OMPDCs from Bacillus subtilis (7) and Escherichia coli (8) prompted speculation that the proximity of the substrate carboxylate group to an active site Asp (Asp 70 in MtOMPDC; Asp 91 in ScOMPDC; Figure 1) contributes to the rate acceleration by electrostatic repulsion and/or geometric distortion of the substrate (6, 13, 14). In support of this proposal, Raman spectroscopic studies of 6-cyanoUMP complexed with ScOMPDC have been interpreted in terms of distortion of the nitrile group out of the plane of the pyrimidine ring (22). Perhaps such “substrate destabilization”4 provides the “missing” factor of ≤ 107 to fully explain the rate acceleration (1017 total - ≥ 1010 from intermediate stabilization). However, the importance of substrate destabilization has not been experimentally evaluated.

Substitutions for an active site residue that participates only in stabilization of the vinyl carbanion intermediate should have similar effects on the rate constants for formation of the carbanion from OMP by decarboxylation and deprotonation of UMP (exchange) in the reverse direction. However, if the effects on the rate constants for decarboxylation and exchange were substantially different, the conclusion would be that the residue plays different roles in stabilizing the transitions states for carbanion formation by the different pathways.

We now report kinetic and structural studies that evaluate the mechanistic roles of Asp 70 in MtOMPDC and its homologue Asp 91 in ScOMPDC. For MtOMPDC, the D70N and D70G substitutions significantly and progressively decrease the value of kcat for decarboxylation of OMP and FOMP (a more reactive substrate analog; Scheme 2) but have little effect on the value of kex for exchange of H6 of the FUMP product with solvent deuterium; the structures of the active sites of wild type MtOMPDC and its D70N and D70G mutants are virtually superimposable when complexed with 6-azaUMP. These results establish that the Asp-Lys-Asp triad functions differently in the decarboxylation and exchange reactions, providing additional evidence for a carbanion intermediate and suggesting the involvement of Asp 70 in destabilization of the substrate.

In unexpected contrast, the D91N mutant of ScOMPDC is not an effective catalyst for either decarboxylation of FOMP or exchange of H6 of FUMP. Also, the structures of the active sites of wild type ScOMPDC and its D91N mutant are not superimposable when complexed with 6-azaUMP, with striking differences in both the conformation of the active site loop and the orientation of the ligand vis á vis the active site residues. These structural differences show that ScOMPDC can form two different complexes with ligands: 1) a “crowded,” catalytically productive complex in which the substrate carboxylate group and the Asp-Lys-Asp triad are in forced proximity; and 2) a more “commodious,” loop-open complex that can accommodate bulky ligands with binding constants comparable to those for OMP and UMP (23).

MATERIALS AND METHODS

Materials

1H and 19F NMR spectra were recorded using a Varian INOVA 500NB NMR spectrometer (500 MHz for 1H and 470 MHz for 19F). OMP and FOMP were prepared as described in the preceding manuscript (1). 6-AzaUMP was purchased from R.I. Chemical Inc. All other reagents were purchased from Sigma Aldrich at the highest purity available. All solutions were prepared with Millipore Ultrapure filtered water.

Synthesis of 5,6-Dihydro-Uridine 5′-Monophosphate (H2UMP)

UMP (500 mg, 1.36 mmoles) was dissolved in 10 mL of H2O, and 10 mL of methanol was added. After addition of 5% palladium on activated carbon (200 mg), the reduction was performed in a Parr shaker under 35 psi pressure of H2. The reaction was allowed to continue for 24 hours; complete reduction of the pyrimidine moiety was confirmed by disappearance of the absorbance at 268 nm. The reaction mixture was filtered to remove the catalyst, and the solvent was removed by rotary evaporation to yield a clear, colorless syrup. The material was dissolved in a small volume of H2O; the concentration was determined by Ames phosphate analysis (24). Purity was confirmed by 1H NMR.

H2UMP: 1H NMR (D2O, 500 MHz): δ (ppm) 5.7 (d, H-1’), 4.3 (dd, H-2’), 4.2 (dd, H-3’), 4.0 (m, H-4’), 3.7 (t, H-6), 3.6−3.4 (ABX, H-5’), and 2.6 (t, H-5).

Synthesis of 5,6-Dihydro-Orotidine 5′-Monophosphate (H2OMP)

OMP (40 mg, 0.1 mmoles) was dissolved in 6 mL of H2O, and 6 mL of methanol was added. After addition of 5% palladium on activated carbon (21 mg), the reduction was performed in a Parr shaker under 40 psi pressure of H2. After 24 hrs, complete reduction was confirmed by disappearance of the absorbance at 279 nm. The reaction mixture was filtered and concentrated to dryness. The residue was dissolved in a small volume of H2O; the concentration was determined by Ames phosphate analysis.

Inspection of the 1H NMR spectrum revealed production of only one diastereomer of H2OMP (a chiral center is generated at C-6 of the dihydropyrimidine ring). Given the preference for OMP to adopt the syn conformation (25), assignment of the 6-R configuration to C-6 was determined by a standard one-dimensional NOESY experiment. A strong NOE was observed between H-6 and H-2’, and weak NOEs were observed between H-6 and both H-1’ and H-2’, consistent with the 6-R configuration. No NOEs were observed between H-6 and the remaining protons.

H2OMP: 1H NMR (D2O, 500 MHz): δ (ppm) 5.8 (d, H-1’), 4.2 (dd, H-6), 4.1 (dd, H-2’), 4.0 (dd, H-3’), 3.9 (m, H-4’), 3.8−3.6 (m, H-5’), and 3.0−2.7 (ABX, H-5).

Synthesis of FUMP

The procedure of Fujita et al. was followed with modifications (26). Phosphorus oxychloride (330 L) was added under dry nitrogen to a stirred mixture of 5-fluorouridine (1.9 mmoles) and 5 mL of anhydrous trimethylphosphate. After 16 hrs at 4 °C, the reaction was quenched with 100 mL H2O. Sodium hydroxide (18 nmmoles) was added, and solution was loaded onto a DEAE Sepharose Fast Flow column (2.5 × 25 cm; HCO 3 ); unreacted 5-fluorouridine was washed from the column with 500 mL of water. FUMP was eluted with a 1 L linear gradient from 0 to 0.25 M triethylammonium bicarbonate, pH 8.4. The product-containing fractions were identified by OD at 269 nm (Σ269 = 9200 M−1 cm−1); 1H NMR spectra of aliquots of fractions were recorded to assess purity. Fractions containing pure FUMP were evaporated; the solid was redissolved in water and taken to dryness several times to remove excess triethylammonium bicarbonate. The product was dissolved in water and lyophilized to yield the final product (26% yield).

MtOMPDC and ScOMPDC

Wild type MtOMPDC, the D70N mutant of MtOMPDC, and wild type ScOMPDC were prepared as described in the preceding manuscript (1). The D70G mutant of MtOMPDC were constructed using the QuikChange II method (Stratagene). DNA sequencing at the University of Illinois Core Sequencing Facility confirmed the expected sequences. The D70G mutants was purified from a pyrF strain of E. coli as described in the preceding manuscript for the D70N mutant (1).

Crystallization and Data Collection

For wild type MtOMPDC, five different crystal forms were grown by the hanging drop method at room temperature (Table 1): 1) in the absence of a ligand WT•apo), 2) complexed with 6aza-UMP (WT•azaUMP), 3) complexed with UMP (WT•UMP), 4) complexed with H2OMP (WT• H2OMP), and 5) complexed with H2UMP (WT•H2UMP). The crystallization conditions utilized the following conditions:

Table 1
Data collection and refinement statistics for complexes of wild type MtOMPDC

For MtOMPDC•apo, the protein solution contained wild type MtOMPDC (38 mg/mL) in 20 mM HEPES, pH 7.5, 150 mM NaCl, and 3 mM DTT; the precipitant contained 30% PEG ME 550, 0.1 M HEPES, pH 7.5, and 0.05 M MgCl2. Crystals appeared in 2−3 days and exhibited diffraction consistent with the space group P41, with two polypeptides per asymmetric unit.

For MtOMPDC•azaUMP, the protein solution contained wild type MtOMPDC (38 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM azaUMP, 150 mM NaCl, and 3 mM DTT; the precipitant contained 25% PEG 3350, 0.1 M Tris-HCl, pH 8.5, and 0.2 M MgCl2. Crystals appeared in 4 days and exhibited diffraction consistent with space group P21, with two polypeptides per asymmetric unit.

For MtOMPDC•UMP, the protein solution contained wild type MtOMPDC (38 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM UMP, 150 mM NaCl, and 3mM DTT; the precipitant contained 30% PEG 4000, 0.1 M sodium citrate, pH 5.6, and 0.2 M ammonium acetate. Crystals appeared in 4−5 days and exhibited diffraction consistent with space group P21 with two polypeptides per asymmetric unit.

For MtOMPDC•H2OMP, the protein solution contained wild type OMPDC (38 mg/mL) in 20 mM HEPES, pH 7.5, 20 mM H2OMP, 150 mM NaCl, and 3mM DTT; the precipitant contained 30% PEG 4000, 0.1 M sodium citrate, pH 5.6, and 0.2 M ammonium acetate. Crystals appeared in 8 days and exhibited diffraction consistent with the space group C2, with 13 polypeptides per asymmetric unit.

For MtOMPDC•H2UMP, the protein solution contained wild type OMPDC (38 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM H2UMP, 150 mM NaCl, and 3 mM DTT; the precipitant contained 30% M PEG 4000, 0.1 M sodium citrate, pH 5.6, and 0.2 M ammonium acetate. Crystals appeared in 7−8 days and exhibited diffraction consistent with the space group C2 with 13 polypeptides per asymmetric unit.

For the D70N mutant of MtOMPDC, three different crystal forms were grown by the hanging drop method at room temperature (Table 2): 1) complexed with SO 2−4 (D70N• SO 2−4), 2) complexed with UMP (D70N•UMP), and 3) complexed with azaUMP (D70N•azaUMP). The crystallization conditions utilized the following conditions:

Table 2
Data collection and refinement statistics for complexes of the D70N and D70G mutants of MtOMPDC

For D70N•SO 2−4, the protein solution contained the D70N mutant of MtOMPDC (46 mg/mL) in 20 mM HEPES, pH 7.5, 100 mM NaCl, and 3 mM DTT; the precipitant contained 2.0 M (NH4)2SO4, 0.1 M HEPES, pH 7.5, and 2% PEG 400. Crystals appeared in 4 days and exhibited diffraction consistent with the space group P41, with two polypeptides per asymmetric unit.

For D70N•UMP, the protein solution contained the D70N mutant of MtOMPDC (46 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM UMP, 100 mM NaCl, and 3 mM DTT; the precipitant contained 25% PEG 3350, 0.1 M Tris-HCl, pH 8.5, and 0.2 M MgCl2. Crystals appeared in 6 days and exhibited diffraction consistent with space group P21, with two polypeptides per asymmetric unit.

For D70N•6azaUMP form, the protein solution contained the D70N mutant of MtOMPDC (46 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM 6-azaUMP, 100 mM NaCl, and 3 mM DTT; the precipitant contained 25% PEG 3350, 0.1 M Tris-HCl, pH 8.5, and 0.2 M MgCl2. Crystals appeared in 6 days and exhibited diffraction consistent with space group P21 with two molecules of D70N OMPDC per asymmetric unit.

For the D70G mutant of MtOMPDC, three different crystal forms were grown by the hanging drop method at room temperature (Table 2) : 1) in the absence of a ligand (D70G•apo), 2) complexed with UMP (D70G•UMP), and 3) complexed with FUMP (D70G•FUMP). The crystallization conditions utilized the following conditions:

For D70G•apo, the protein solution contained D70G OMPDC (35 mg/mL) in 20 mM HEPES (pH 7.5), 150 mM NaCl, and 3 mM DTT; the precipitant contained 4.0 M sodium formate, pH 7.0. Crystals appeared in 2 days and exhibited diffraction consistent with the space group P41, with two polypeptides per asymmetric unit.

For D70G•UMP, the protein solution contained the D70G mutant of MtOMPDC (35 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM UMP, 150 mM NaCl, and 3 mM DTT; the precipitant contained 1.8 M tri-ammonium citrate, pH 7.0. Crystals appeared in 6 days and exhibited diffraction consistent with space group P41, with two polypeptides per asymmetric unit.

For D70G•FUMP, the protein solution contained the D70G mutant of MtOMPDC (35 mg/mL) in 20 mM HEPES, pH 7.5, 40 mM FUMP, 150 mM NaCl, and 3 mM DTT; the precipitant contained 0.3 M magnesium formate and 0.1 M Bis-Tris, pH 5.5. Crystals appeared in 9 days and exhibited diffraction consistent with space group P41, with two polypeptides per asymmetric unit.

For wild type ScOMPDC, two different crystal forms were grown by the hanging drop method at room temperature (Table 3): 1) in the absence of a ligand (ScOMPDC•apo), and 2) and complexed with 6-azaUMP (ScOMPDC•azaUMP).

Table 3
Data collection and refinement statistics for complexes of wild type and the D91N mutant of ScOMPDC

For ScOMPDC•apo, the protein solution contained wild type OMPDC (29 mg/mL) in 20 mM Tris, pH 7.9, 100 mM NaCl, and 20% glycerol; the precipitant contained 30% PEG 4000, 0.1 M Tris, pH 8.5, and 0.2 M sodium acetate. Crystals appeared in 4−5 days and exhibited diffraction consistent with the space group P212121, with four polypeptides per asymmetric unit.

For ScOMPDC•azaUMP, the protein solution contained wild type OMPDC (29 mg/mL) in 20 mM Tris, pH 7.9, 100 mM NaCl, 20% glycerol, and 40 mM 6-azaUMP; the precipitant contained 30% PEG 2000, 0.1 M sodium acetate, pH 4.6, and 0.2 M ammonium sulfate. The crystals appeared in 9 days and exhibited diffraction consistent with the space group C2, with 2 polypeptides per asymmetric unit.

For the D91N mutant of ScOMPDC, two different crystal forms were grown by the hanging drop method at room temperature (Table 3): 1) in the absence of a ligand (D91N•apo), and 2) and complexed with 6-azaUMP (D91N•azaUMP).

For D91N•apo, the protein solution contained the D91N mutant of ScOMPDC (27 mg/mL) in 50 mM Tris, pH 7.0, and 5% glycerol; the precipitant contained 25% PEG 3350, 0.1 M Bis-Tris, pH 6.5, and 0.2 M MgCl2. Crystals appeared in 4 days and exhibited a diffraction pattern consistent with space group P212121, with four polypeptides per asymmetric unit.

For D91N•azaUMP, the protein solution contained the D91N mutant of ScOMPDC (27 mg/mL) in 50 mM Tris, pH 7.0, 5% glycerol, and 40 mM 6-azaUMP; the precipitant contained 25% M PEG 3350, and 0.1 M Bis-Tris, pH 5.5. Crystals appeared in two weeks and exhibited diffraction consistent with the space group P212121 with four polypeptides per asymmetric unit.

Prior to data collection, crystals were transferred to cryoprotectant solutions composed of their mother liquids containing 20% glycerol and flash-cooled in a nitrogen stream. All data sets were collected at the NSLS X4A beamline (Brookhaven National Laboratory) with an ADSC CCD detector. Diffraction intensities were integrated and scaled with programs DENZO and SCALEPACK (27). The data collection statistics are given in Tables 1 and and22.

Structure Determination and Model Refinement

The structures of MtOMPDC and its mutants were solved by molecular replacement with a fully automated molecular replacement pipeline BALBES (28), using only diffraction and sequence data as input; partially refined structures were obtained without any manual intervention. Subsequent several iterative cycles of refinement were performed using manual model rebuilding with TOM (29), refinement with CNS (30), automatic model rebuilding with ARP (31), and solvent building with the CCP4 suite (32).

In the unliganded structures of wild type MtOMPDC and the D70N and D70G mutants, the active site loop containing residues 181−189 is disordered. The various liganded structures of wild type MtOMPDC and its mutants have well-defined density for this loop, which contacts the ligand. The liganded structures have well-ordered density for the ligands in each active site. No residue (other than Gly) lies in disallowed regions of the Ramachandran plot in the structures of wild type MtOMPDC and its mutants. In each structure, the polypeptides form closed dimers, with the monomers connected by non crystallographic two-fold axes.

For ScOMPDC and the D91N mutant, the structures were solved by molecular replacement with fully automated molecular replacement pipeline BALBES (28), using only input diffraction and sequence data. Partially refined structures were obtained from BALBES without any manual intervention. Subsequent iterative cycles of refinement were performed that included manual model rebuilding with TOM (29), refinement with CNS (30), automatic model rebuilding with ARP (31), and solvent building with the CCP4 suite (32).

In the unliganded structures of wild type ScOMPDC and the D91N mutant, residues 205−218 in the active site loop are disordered in two of the four molecules in asymmetric unit and are not included in the final models of these polypeptides; in the polypeptides in which the loop is ordered, this loop is in an “open” conformation. In the ScOMPDC•azaUMP structure, this loop is well-defined in all molecules in the asymmetric unit and is in a “closed” conformation that contacts the 6-azaUMP ligand; the electron density for the ligand is well-defined. In the D91N•azaUMP structure, this loop is well-defined in the “open” conformation in two of the four polypeptides in the asymmetric unit and is disordered in the remaining polypeptides; in the polypeptides in which the loop is ordered, the electron density for the ligand is well-defined, except that it bound in a nonproductive conformation that is described in the Results and Discussion section.

No residue (other than Gly) lies in disallowed regions of the Ramachandran plot in the structures of wild type ScOMPDC and the D91N mutant. In each structure, the polypeptides form closed dimers, with the monomers connected by noncrystallographic two-fold axes.

Final crystallographic refinement statistics are provided in Tables 1--33.

Enzymatic Assays

OMPDC activity was measured at 25 °C as described in the accompanying manuscript (1), except that the buffer concentration was 50 mM and the ionic strength was maintained at 0.1 M with NaCl. The pH dependence of the kinetic constants was determined using the following buffers: acetate, pH 5.0; MES, pH 6.0; MOPS, pH 7.1 and 7.6; HEPES, pH 7.1, 7.6, and 8.2; Tris-HCl, pH 7.6, 8.2, and 8.4; glycylglycine, pH 8.0, 8.2, 8.4, and 8.7; glycine, pH 9.0, 9.3, 9.6, 10.0, and 10.5; and CAPS, pH 11.0.

Exchange of H6 of FUMP

Experiments to measure the rate of exchange of H6 of FUMP with solvent deuterium were performed as described by Amyes et al. with modifications (17). Spectra (150 scans) were acquired at 25 °C using a sweep width of 1000 Hz, a 90° pulse angle, an acquisition time of 6 s, and a 2 s relaxation delay. Instead of measuring exchange by integrating the intensity of the remaining proton on C6, exchange was measured by integrating the well-resolved 19F resonances associated with [6-1H]-FUMP (doublet at −164.1 ppm; J = 6.7 Hz, coupled to H6) and [6-2H]-FOMP (singlet at −164.4 ppm). The logarithm of the fraction of [6-1H]-FUMP remaining at time t was plotted as a function of time; the negative slope is the observed exchange rate, kobs. The exchange rate, kex, = kobs × [FUMP]total/[E].

The pH dependence of kex was measured using different buffers at the following values of pD: MOPS, pD 7.1 and 7.6; glycylglycine, pD 8.1 and 9.3; glycine, pD 10 and 10.5. The concentration of FUMP was determined from its absorbance at 270 nm (ε270 = 8914 M−1 cm−1, pH 7.1).

Product Isotope Effect (PIE)

The reaction (800 μL) was performed in 50/50 (v/v) H2O/D2O at 25 °C with 2 mM FOMP in 50 mM MOPS, pL 7.3, with the ionic strength maintained at 0.1 M with NaCl (16). When the reaction was completed, the enzyme was removed by centrifugation (Amicon Microcon YM-10), and the solvent was removed by lyophilization. The residue was dissolved in 800 μL D2O, and the pD was adjusted to 7.3. A 19F NMR spectrum was recorded; the PIE was calculated from the ratio of the integrations for the resonances associated with [6-2H]- and [6-1H]-FUMP (vide supra).

Molecular Graphics Images

The images in Figures 2--55 and and88 were constructed using Chimera (33).

Figure 2
Representative electron density for the active site of wild type MtOMPDC complexed with 6-azaUMP and contoured at 1.5σ. The details of the interactions between 6-azaUMP and the active site are described in the text.
Figure 5
Structures of mutants with substitutions for Asp 70. Panel A, wild type MtOMPDC complexed with 6-azaUMP. Panel B, D70N complexed with 6-azaUMP. Panel C, D70N complexed with UMP. Panel D, D70G complexed with 6-azaUMP. Panel E, D70G complexed with FUMP. ...
Figure 8
Panel A, comparison of the structures of wild type ScOMPDC in the absence of ligand (gray), wild type ScOMPDC complexed with 6-azaUMP (red), the D91N mutant of ScOMPDC in the absence of ligand (magenta), and the D91N mutant of ScOMPDC complexed with 6-azaUMP ...

RESULTS AND DISCUSSION

The studies reported in this manuscript focus on the roles of homologous Asp residues in the active site hydrogen-bonded networks in MtOMPDC (Asp 70) and ScOMPDC (Asp 91). The mechanisms of the reactions catalyzed by these homologous OMPDCs have been controversial, so we studied both enzymes to assess whether we would obtain consistent mechanistic inferences. The residues that form the active site are strictly conserved despite the low sequence identity shared by these homologous proteins (~20% from a sequence-based alignment). However, this expectation was not realized, reinforcing the limitations of kinetic and mechanistic studies in the absence of structural information (vide infra).

In addition to the natural OMP substrate and UMP product, we used FOMP and FUMP to facilitate our kinetic measurements (Scheme 2). These substrate/product analogs are more reactive than the natural OMP substrate for decarboxylation and the UMP product for exchange of H6 because the electronegative 5-fluoro substituent stabilizes the carbanionic intermediate obtained by decarboxylation of the substrate or abstraction of H6 from the product, respectively. Shostak and Jones reported that the value of kcat for decarboxylation of FOMP by ScOMPDC exceeds that for OMP by 30-fold, although the values of kcat/KM are the equivalent for OMP and FOMP (34). Cleland and coworkers confirmed that FOMP is a better substrate for ScOMPDC, with the increase in kcat reported as 11-fold (35). Although not discussed by either Jones or Cleland, the equivalent values for kcat/KM for decarboxylation of OMP and FOMP (~1 × 107 M−1 sec−1) by ScOMPDC are consistent with rate-limiting substrate binding for one or both substrates. Also, that the values of kcat for OMP and FOMP differ by ~11-fold is consistent with rate-limiting product dissociation, not decarboxylation, for one or both substrates. A measure of the intrinsic effect of 5-fluorine substitution on decarboxylation may be provided by the 350-fold difference in the values of kcat for decarboxylation of OMP and FOMP catalyzed by the D70N mutant of MtOMPDC (vide infra); for D70N the values of kcat for both OMP and FOMP are independent of solvent viscosity, so kcat measures the rate of the decarboxylation reaction.

As described in the accompanying manuscript, the values of kcat measured with FOMP for MtOMPDC and with both OMP and FOMP for ScOMPDC are dependent on solvent viscosity so they are influenced by product dissociation and, therefore, do not directly measure the rate of decarboxylation (1). For both enzymes and substrates the values of kcat/KM also are dependent on solvent viscosity so these kinetic constants include contributions from substrate binding. Recognition that the values of the kinetic constants for both OMP and FOMP can depend on the physical steps of substrate binding and product dissociation is important for interpreting the effects of substitutions of active site residues on mechanism.

Structural Characterization of Wild Type MtOMPDC

Pai and coworkers reported structures of unliganded wild type MtOMPDC (1DV7) and the complex with 6-azaUMP (1DVJ), a transition state/intermediate analog (12). We also determined structures of unliganded wild type MtOMPDC (1.6 Å resolution; 3G18) and its complex with 6-azaUMP (1.5 Å resolution; 3G1A); representative electron density of the 6-azaUMP ligand is presented in Figure 2. Although our structures are essentially identical to those determined by Pai [0.53 Å r.m.s.d. for 209 Cα pairs in the unliganded structures (1DV7) and 0.37 Å r.m.s.d. for 215 Cα pairs in the liganded structures (1DVJ)], we discovered that the Pai's unliganded structure was solved assuming an incorrect space group (P41212), with the correct space (P41) the same as that we determined. The major difference between our unliganded structure and that determined by Pai is the geometry of the contact between Phe 100 in the two polypeptides at the interface in the functional dimer.

We also determined structures of wild type MtOMPDC complexed with the UMP product (1.5 Å resolution; 3G1D), (6R)-H2OMP (a distorted substrate analog; 2.5 Å resolution; 3G1F), and H2UMP (a distorted product analog; 2.3 Å resolution; 3G1H). (6R)-H2OMP, with a tetrahedral geometry at C6 (and C5), might be considered an analog of a destabilized OMP substrate, although the mechanistic importance of out-of-plane distortion of the carboxylate group is uncertain because the electron pair of the carbanionic intermediate that resides on C6 is located in an sp2 orbital in the plane of the pyrimidine ring.

In the unliganded structure, the active site loop (residues 181−189) is disorderd; in all four liganded structures (WT•azaUMP, WT•UMP, WT•H2OMP, and WT•H2UMP), the loop is ordered and closed over the ligand to sequester the ligand from solvent (Figure 3, panel A is the unliganded structure; panels B-E are the structures of the WT•azaUMP, WT•UMP, WT•H2OMP, and WT•H2UMP complexes, respectively). The structures of the four complexes superimpose well, although the structures of the backbones of the polypeptides of the liganded structures show small, but significant, differences when compared with that of the unliganded enzyme. Although coupling of substrate binding to catalysis has not been investigated with MtOMPDC as it has with ScOMPDC (9, 36-38), conformational changes that accompany substrate binding presumably transmit the substrate's binding energy to the active site where it can be used for substrate destabilization and/or transition state/intermediate stabilization. In each complex, O2 of the pyrimidine ring of the ligand is hydrogen-bonded to the carboxamide group of Gln 185, O4 is hydrogen-bonded to the backbone amide of Ser 127, and N3 is hydrogen-bonded to the OH group of Ser 127.

Figure 3
Structures of wild type MtOMPDC in the absence and complexed with ligands. Panel A, in the absence of ligands. Panel B, complexed with 6-azaUMP. Panel C, complexed with UMP. Panel D, complexed with H2OMP. Panel E, complexed with H2UMP.

Structural Characterization of Wild Type MtOMPDC Liganded with (6R)-H2OMP

Wolfenden reported that a mixture of diastereomers of H2OMP (of unspecified diastereomeric composition) is an effective inhibitor of ScOMPDC (Ki = 2.5 × 10−7 M) (3). Within the limits of our NMR analyses, we obtained only the (6R)-diastereomer by catalytic hydrogenation of OMP (configuration assessed by NOEs between H6 of the reduced pyrimidine ring and the ribose protons; data not shown).

We determined that (6R)-H2OMP is a weak inhibitor of both ScOMPDC and MtOMPDC (Ki, 80 μM and 100 μM, respectively; data not shown). We also were able to determine the structure of its complex with MtOMPDC but not ScOMPDC. Perhaps the (6S)-diastereomer is a more potent inhibitor, although we regard this as unlikely because the 6-carboxylate would be located in a spatially more congested region of the active site, i.e., juxtaposed to the hydrogen-bonded triad (vide supra). In the absence of the experimental procedure for Wolfenden's preparation of H2OMP and, also, his procedures for measurement of the kinetic constants, we cannot explain this apparent kinetic discrepancy. H2UMP was analogously prepared from UMP; the structure of its complex with MtOMPDC provides an independent assessment of the effect of reduction of the pyrimidine ring on the active site structure.

Both (6R)-H2OMP and H2UMP are accommodated without perturbation of active site residues relative to their positions in complexes with UMP and 6-azaUMP (Figure 3, panels D and E, respectively). The carboxylate group of (6R)-H2OMP is displaced from the “plane” of the dihydropyrimidine base, as expected from sp3 hybridization of C6, and is located in a hydrophobic pocket on the opposite face of the active site from Lys 72 (Figure 4, panel A). The carboxylate group does not participate in any hydrogen bonding/ionic interactions, although disordered water molecules could be located in the hydrophobic pocket. This structure can be compared with that of the inactive D70A/K72A mutant of MtOMPDC complexed with OMP (1KM6) that has been used as evidence for substrate destabilization (Figure 4, panel B). In 1KM6, the carbon of the carboxylate group of the substrate is 1) displaced 0.3 Å out of the plane of the pyrimidine ring, also away from Lys 72, and 2) twisted so that one carboxylate oxygen can be hydrogen bonded to Lys 42. In the absence of the anionic carboxylate group of Asp 70 and the side chain of Lys 72, no potential unfavorable steric/electrostatic interactions are apparent that would produce the observed out-of-plane distortion of the OMP carboxylate group.

Figure 4
Panel A, front view of active site of wild type MtOMPDC complexed with H2OMP, with the side chains of the hydrophobic residues colored in green. Panel B, side view of the superposition of the active site of wild type OMPDC complexed with H2OMP (gray) ...

Structural Characterization of the D70N and D70G Mutants of MtOMPDC

We determined structures of the D70N and D70G mutants. Structures were obtained for D70N in the absence of ligands and also complexed with UMP and 6-azaUMP; structures were obtained for D70G in the absence of ligands and complexed with UMP and FUMP. Pai reported structures of D70A complexed with UMP (1KLZ) and D70G complexed with 6-azaUMP (1KLY). As in Pai's structures of D70A (12), our structures of D70G contain a chloride ion from the crystallization buffer within the active site, occupying the position of the δ-carboxylate group of Asp 70 in wild type MtOMPDC.

In four of the five available structures for D70N and D70G, the nucleotide ligands superimpose closely, with O2 of the pyrimidine ring hydrogen-bonded to Gln 185 [Figure 5, panel A is the wild type MtOMPDC liganded with 6-azaUMP; panels B-F are the D70N•azaUMP, D70N•UMP, D70G•6-azaUMP (1KLY), D70G•FUMP, and D70G•UMP complexes, respectively]. The exception is our structure of D70N complexed with 6-azaUMP, in which O2 is hydrogen-bonded to the carboxamide group of Asn 70, i.e., the base is rotated 180° about the glycosidic bond. With the exception of Lys 72, the positions of the active site residues also superimpose closely. Because Asp 70 is hydrogen bonded to Lys 72 in the active site hydrogen-bonded network, substitution of the Asp carboxymethyl side chain with a neutral and/or sterically less demanding side chain allows Lys 72 to adopt alternate conformations, although in each structure a hydrogen bond is maintained to Asp 75. However, the side chain of Lys 42 remains fixed in the various liganded structures, presumably because of its hydrogen bonding interaction with O3’ of the ribose moiety of the ligand.

The active sites of D70G complexed with FUMP or UMP but not 6-azaUMP contain a chloride ion located near the position of the carboxylate group of Asp 70 in the wild type enzyme (vide infra). We performed assays of D70G as a function of chloride ion concentration and observed no activation or inhibition, suggesting that the chloride ion is not a functional surrogate of the Asp carboxylate group.

On the basis of these structures, we conclude that significant conformational changes do not accompany the substitutions, thereby allowing the impact of the substitutions on the values of kcat and kcat/KM to be used to deduce mechanism.

Kinetic Characterization of the D70N and D70G Mutants of MtOMPDC Decarboxylation of OMP and FOMP

Decarboxylation of OMP is the rate-determining energy barrier on the reaction coordinate for wild type MtOMPDC (the value of kcat is independent of viscosity; the value of kcat/KM has a small inverse dependence on viscosity) (1). However, for FOMP, both kcat and kcat/KM are inversely proportional to solvent viscosity, so a physical step (substrate binding and product dissociation, respectively) dominates the values of kcat and 0kcat/KM. These differences are important for interpreting the kinetic constants.

The pH dependencies of the values of kcat and kcat/KM for decarboxylation of both OMP and FOMP catalyzed by wild type MtOMPDC its D70N and D70G mutants were determined. The data for OMP are displayed in Figure 6, panels A and B; the data for FOMP are displayed in panels C and D. The values of the kinetic constants at pH 9.3 (the pH optimum for D70G with FOMP) are summarized in Table 4.

Figure 6
Kinetic constants (kcat, panel A; kcat/KM, panel B) using OMP as substrate for wild type OMPDC (blue) and its D70N (red) and D70G (green) mutants. Kinetic constants (kcat, panel C; kcat/KM, panel D) using FOMP as substrate for wild type OMPDC (blue) and ...
Table 4
Kinetic constants for decarboxylation of OMP and FOMP at pH 9.3 and exchange of H6 of FUMP.

With OMP, the values of kcat and kcat/KM decrease in the order wild type > D70N > D70G (the kinetic constants for D70A were not quantitated because enzyme concentrations comparable to substrate concentration were required to observe decarboxylation). The decreases in the values of kcat, 200-fold for D70N and 1,800-fold for D70G, provide estimates of the increase in the activation energy barrier as the charge/size of the side chain is neutralized/reduced.

The same qualitative behavior is observed for kcat and kcat/KM when FOMP is used as substrate, although the values of kcat are pH-dependent for the D70G mutant. The values of kcat/KM decrease less than the values of kcat, because the values of KM for the mutants are less than that for wild type. However, with FOMP, the decreases in the value of kcat relative to that for wild type, 35-fold for D70N and 490-fold for D70G, do not provide estimates for the increases in activation energy barrier for decarboxylation, because kcat for wild type MtOMPDC with FOMP is dominated by production dissociation, not decarboxylation (1). Assuming that the ratio of the viscosity-independent values of kcat for decarboxylation of OMP and FOMP catalyzed by D70N (7.7 sec−1/0.022 sec−1 = 350) reflects the “intrinsic” effect of the 5-fluoro substitution in stabilizing the carbanion obtained by decarboxylation, we estimate the rate of decarboxylation of FOMP in the active site by wild type MtOMPDC as 4.4 sec−1 × 350 ≈ 1500 sec−1. With this assumption, the decreases in the rates of decarboxylation are 190-fold for D70N and 2,700-fold for D70G. The reductions for D70N and D70G are similar to those measured when OMP is used as substrate (vide supra).

We conclude that, irrespective of whether OMP or FOMP is the substrate, large reductions in the rate of decarboxylation are produced by substitution of Asp 70 with neutral and/or smaller residues.

That the values of KM for both OMP and FOMP are virtually the same for D70N and D70G substitutions is inconsistent with the classical definition of “ground state destabilization”, i.e., the substitutions for Asp 70 decrease the affinity of the enzyme for the substrate in the absence of a destabilizing interaction.4 A possible explanation is that conformational changes accompany the binding of substrate and subsequent formation of transition state/intermediate/product complexes (vide infra); these will contribute to and prevent quantitative interpretation of the value of KM in the context of changes expected if “ground state destabilization” of a static Michaelis complex were assumed.

Kinetic Characterization of the D70N and D70G Mutants of MtOMPDC Exchange of H6 of FUMP

Both wild type ScOMPDC and MtOMPDC catalyze the slow exchange of H6 of UMP with solvent deuterium (17). For ScOMPDC, the rate constant for the exchange reaction, kex, increases with pD, giving values of 8.0 for the pKa of the base (Lys 93) and 1.2 × 10−5 sec−1 for the pH-independent maximum rate (kex)max. Measurement of the slow rate of exchange reaction using UMP requires both high concentrations of enzyme and long incubation times (≤ 250 hrs). We observed that H6 of FUMP undergoes exchange with solvent deuterium ~3,200-fold faster with FUMP than UMP in the presence of wild type MtOMPDC (vide infra) and wild type ScOMPDC.5 Therefore, we used FUMP to investigate the exchange reactions catalyzed by wild type and the D70N and D70G mutants of MtOMPDC.

We determined kex as a function of pD for the wild type and the D70N and D70G mutants of MtOMPDC using 5 mM FUMP. The value of kex increases with pD and becomes pD independent at high values of pD (Figure 7). The dependence of kex on [FUMP] was measured at pD 9.3 for all three proteins (where the values of kcat are maximal) and observed to be independent of the concentration of FUMP at 5 mM; the values of (kex)max are collected in Table 4. In contrast to the values of kcat for decarboxylation of FOMP, the values of (kex)max are not significantly influenced by substitutions for Asp 70.

Figure 7
Dependence of kex on pH for wild type MtOMPDC (blue), D70N (red) and D70G (green) with FUMP.

The value of kex is a composite of the rates of abstraction of H6 to generate the vinyl carbanion intermediate and subsequent delivery of solvent deuterium from Lys 72. Because the solvent product isotope effect (PIE) is 1.0 for wild type MtOMPDC using both OMP and FOMP, hydron transfer from Lys 72 each of the carbanion intermediates must be faster than rotation of the Cε-N bond that exchanges the positions of the hydrons (16). In wild type MtOMPDC, hydrogen bonds between Lys 72 and both Asp 70 and Asp 75 may restrict this rotation. Substitutions for Asp 70 that remove or weaken the hydrogen bond to Lys 72 could allow enhanced rotation so that it could be faster than hydron transfer; as a result, the same value of kex could be obtained if the stability of the carbanion were decreased by the substitution. If rotation about the Cε-N bond were sufficiently fast, the PIE may increase if the carbanion is then able to discriminate between hydrogen and deuterium. Therefore, we determined the solvent PIEs using FOMP for wild type MtOMPDC and the D70N and D70G mutants (data not shown). Like wild type MtOMPDC, the PIEs are 1.0, consistent with the interpretation that the values of kex measured for the mutants are best explained by the stability of the vinyl carbanion intermediate being independent of the identity of residue 70 (Asp, Asn, or Gly).

Differential Effect of Substitutions for Asp 70 on Decarboxylation and Exchange

In contrast to large decreases in the rate of decarboxylation that accompany substitutions for Asp 70, formation of the FUMP carbanion by abstraction of H6 is unaffected. The differing effects of the substitutions on kcat for decarboxylation and kex for exchange of H6 provide additional evidence for formation of a carbanion intermediate in a two-step mechanism.

That the substitutions affect the rate of formation of the intermediate from FOMP (kcat) but not from FUMP (kex)max is consistent with Asp 70 enhancing the rate of decarboxylation by substrate destabilization.

Kinetic and Structural Characterization of the D91N Mutant of ScOMPDC

We constructed and both kinetically and structurally characterized the homologous D91N mutant of ScOMPDC. Because the active site residues are conserved, we expected that the D91N mutant of ScOMPDC also would provide similar evidence for the importance of Asp 91 in destabilizing the enzyme-substrate complex.

However, using FOMP as substrate, we were unable to measure values for kcat and KM; the value of kcat/KM is 15 M−1 sec−1 at pH 7.1 (data not shown). Using FUMP as substrate, we observed no exchange of H6 of FUMP using high concentrations of the D91N mutant and long incubation times. In contrast to the D70N mutant of MtOMPDC, the D91N mutant of ScOMPDC is severely impaired in both the decarboxylation and exchange reactions.

Structures of wild type ScOMPDC and its D91N mutant were determined in the absence and presence of 6-azaUMP. In the unliganded structures, the 19-residue active site loops adopt the same open conformation (Figure 8, panel A, top row). In the structure of wild type ScOMPDC complexed with 6-azaUMP (Figure 8, panel A, left structure in bottom row), the loop is closed, sequestering the ligand from the active site, with the phosphate group hydrogen bonded to Gln 215, Tyr 217, and Arg 235, and enforcing the juxtaposition of N6 to the active site hydrogen bonding network that is reinforced by the Ser 154-Gln 215 “clamp” (38). Our structures for wild type ScOMPDC are similar to those reported previously for the unliganded enzyme and its complex with BMP (5).

In the structure of the D91N mutant complexed with 6-azaUMP (Figure 8, panel A, right structure in bottom row), the ligand is bound in the active site. However, the loop in the open conformation and, remarkably, the phosphate and pyrimidine “ends” of the molecule occupy “opposite” positions in the active site relative to the 6-azaUMP ligand in wild type ScOMPDC (Figure 8, panel B, wild type; panel C, D91N mutant): the phosphate group is hydrogen-bonded to both Lys 59 and Lys 93 in the active site hydrogen-bonded network and the pyrimidine base is proximal to Tyr 217 in the open loop and, also, exposed to solvent. To the extent that this structure reflects the geometry of binding of FOMP and FUMP to the D91N mutant, this structure provides an explanation for the inability for the D91N mutant to catalyze decarboxylation of FOMP and exchange of H6 of FUMP.

This structure also provides a possible explanation for Wolfenden's observation that derivatives of OMP and UMP obtained by addition of sulfite to the 5,6-double bond of the pyrimidine bases are effective inhibitors of ScOMPDC, with the values of their Kis lower than that for UMP (Scheme 4) (23).

Inhibition by these bulky, anionic substrate/product analogs was used as evidence that substrate destabilization by enforced electrostatic interactions between the substrate carboxylate group and the Asp-Lys-Asp triad can not be important in producing the large rate acceleration in a “commodious” active site. However, our structure of the D91N mutant with 6-azaUMP strikingly reveals that the active site of ScOMPDC can bind ligands in nonproductive conformations as a result of opportunistic interactions with the conformationally flexible active site loop that provide a structural strategy for avoiding otherwise unfavorable interactions with the Asp-Lys-Asp triad. Perhaps the 6R-diastereomer of UMP-SO3 can be accommodated in the active site of ScOMPDC as (6R)-H2OMP is accommodated in the active site of MtOMPDC, but it is unlikely that the diastereomers of OMP-SO3 can be similarly accommodated because of enforced “crowding” of the anionic substituents vis á vis the Asp-Lys-Asp triad.

Role of Asp 70 in Catalysis

The substantial effects of substitutions for Asp 70 on k2 for decarboxylation (as much as ~2,700-fold for D70G with FOMP) relative to the insensitivity of the value of kex is most simply explained by a two-step mechanism involving the formation of a carbanion intermediate. In this mechanism, the carboxylate side chain of Asp 70 influences the rate of formation of the intermediate (decarboxylation) but not its stability (exchange). This differential effect is consistent an unfavorable interaction between the carboxylate group of Asp 70 with the orotate carboxylate group results that is relieved, in part, by a neutral carboxamide group but, more substantially, by a smaller, uncharged group.

A structural description of the substrate destabilization is uncertain, although the OMP-liganded structure of the D70A/K72A mutant of MtOMPDC has been used as evidence that displacement of the carboxylate group out of the plane of the pyrimidine base may be important for catalysis (35). As noted previously and illustrated in Figure 4, in this mutant distortion occurs despite the absence of the side chains of Asp 70 and Lys 72 that might be expected to provide the electrostatic and steric destabilizing interactions. A hydrogen bond is formed between the out-of-plane OMP carboxylate group and the conserved Lys 42; perhaps this interaction is sufficient to stabilize the observed distorted structure for the substrate.

An unfavorable electrostatic interaction between the carboxylate groups of Asp 70 and OMP may destabilize the enzyme-substrate complex as it proceeds to the transition state for decarboxylation; this interaction is not possible in the exchange reaction. However, that 1) the D70N and D70G substitutions cause small increases in the value of KM for OMP (Table 4); and 2) the Ki for UMP (~0.4 mM) is ~400-fold larger than the Kd for OMP (~1.0 μM) (39) show that destabilizing interactions that produce an increase in the value of kcat are not expressed in the measured binding constants.

The destabilizing interactions that promote decarboxylation may develop as the transition state for decarboxylation is approached, perhaps as a result of conformational changes that are coupled to substrate binding (Jiali Gao, personal communication). This strategy for catalysis would require sufficient stabilization of the transition state to overcome the electrostatic destabilization that results from coordinated motion of the anionic carboxylate groups of the OMP substrate and the active site. Both computational and dynamic (NMR) studies of the reaction coordinate can be expected to provide further insights into the structural bases for the substrate destabilization that is suggested by the differential effects of substitution for Asp 70 on decarboxylation of the substrate and exchange of H6 of the product with solvent hydrogen.

Mechanism and Rate Acceleration for the OMPDC-Catalyzed Reaction

Substantial evidence exists for the formation of a carbanion intermediate in the OMPDC-catalyzed reaction: 1) the solvent deuterium product isotope effect is 1.0 (16); 2) H6 of the UMP product undergoes exchange with solvent hydrogen (17), and 3) as documented in this manuscript, substitutions for Asp 70 in MtOMPDC influence decarboxylation but not exchange. Thus, the two-step mechanism is secure.

The rate acceleration for the OMPDC-catalyzed reaction is 1017. The rate of exchange of H6 of the UMP product with solvent hydrogen provides an upper limit on the pKa of the UMP product in the active site (≤ 22) and, therefore, a lower limit on the stabilization of the carbanion intermediate (≥ 14 kcal/mol). Thus, stabilization of the carbanion intermediate and the transition state(s) involved in its formation account for a substantial part of the rate acceleration (≥ 1010). In principle, all of the rate acceleration could result from stabilization of anionic species in the active site. But, the kinetic and structural data described in this manuscript provide evidence that substrate destabilization also contributes to the rate acceleration, although whether the contribution is quantitated by the decrease in the value of k2 observed with D70G (~4.7 kcal/mol) is uncertain.

The structural bases for both stabilization of the vinyl carbanion intermediate and substrate destabilization also are not yet certain. In the case of stabilization of the intermediate, the general acidic Lys 72 in MtOMPDC (Lys 91 in ScOMPDC) has been invoked as providing substantial, and perhaps all of the, stabilization on the basis of the structures of liganded complexes (35). However, other factors, including the hydrogen bonding environment of O4 of the pyrimidine base that could stabilize an anionic carbene intermediate have not been experimentally investigated (20, 21). In the case of substrate destabilization, Asp 70 has been implicated as important, but whether geometric distortion the OMP substrate is catalytically relevant remains unknown. These and other issues are under investigation in our laboratories.

Abbreviations

FOMP
5-fluoroorotidine 5’-monophosphate
FUMP
5-fluorouridine 5’-monophosphate
azaUMP
6-azauridine 5’-monophosphate
BMP
1-(5’-phospho-β-D-ribofuranosyl)barbituric acid
H2OMP
6R-5,6-dihydroorotidine 5’-monophosphate
H2UMP
5,6-dihydrouridine 5’-monophosphate
MtOMPDC
OMPDC from Methanothermobacter thermautotrophicus
OMP
orotidine 5’-monophosphate
OMPDC
orotidine 5’-monophosphate decarboxylase
PIE
product isotope effect
ScOMPDC
OMPDC from Saccharomyces cerevisiae
WT
wild type

Footnotes

This research was supported by NIH GM039754 (to J.P.R.) and GM065155 (to J.A.G.). Molecular graphics images were produced using the UCSF Chimera package from the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco (supported by NIH P41 RR-01081). The X-ray coordinates and structure factors for the following structures have been deposited in the Protein Data Bank with the indicated accession codes: wild type MtOMPDC in the absence of ligands (PDB accession code 3G18), wild type MtOMPDC complexed with 6-azaUMP (3G1A), wild type MtOMPDC complexed with UMP (3G1D), wild type MtOMPDC complexed with H2OMP (3D1F), wild type MtOMPDC complexed with H2UMP (3D1H), the D70N mutant of MtOMPDC in the absence of ligands (3G1Y), the D70N mutant of MtOMPDC complexed with 6-azaUMP (3G24), the D70N mutant of MtOMPDC complexed with UMP (3G22), the D70G mutant of MtOMPDC in the absence of ligands (3G1S), the D70G mutant of MtOMPDC complexed with UMP (3G1X), the D70G mutant of MtOMPDC complexed with FOMP (3G1V), wild type ScOMPDC in the absence of ligands (3GDK), wild type ScOMPDC complexed with 6-azaUMP (3GDL), the D91N mutant of ScOMPDC in the absence of ligands (3GDR), and the D91N mutant of ScOMPDC complexed with 6-azaUMP (3GDT).

2Our values for wild type ScOMPDC: kcat = 20 sec−1, rate acceleration = 7.1 × 1016; kcat/KM = 1.3 × 107 M−1 sec−1, proficiency = 4.6 × 1022 M−1 (1).

3As shown in Figure 1, Lys 42 is hydrogen-bonded to Asp 70 in the triad that is proximal to C6/N6 of pyrimidine nucleotide ligands. Lys 42 is also hydrogen-bonded to the 3’-OH group of nucleotide ligands; the position and conformation of Lys 42 are independent of the identity of residue 70 (vide infra).

4The term “ground state destabilization” is used to describe an interaction between the active site and substrate that destabilizes the Michaelis complex and, therefore, increases the value of KM relative to an environment in which the unfavorable interaction is not present. If the unfavorable interaction is relieved in the transition state, the amount of destabilization will be expressed as an equivalent increase in the value of kcat. Therefore, the value of kcat/KM will be the same in the presence and absence of the interaction if “ground state destabilization” is important. The data presented in this manuscript do not conform to this definition of “ground state destabilization”, perhaps because the destabilizing interactions are not fully expressed in a “Michaelis complex” but develop as the reaction coordinate from an initial encounter complex is traversed. This suggests the importance of conformational changes/dynamics in catalysis, evidence for which is obtained by comparing the structures of both MtOMPDC and ScOMPDC in the absence and presence of transition state/intermediate analogs, as described in this manuscript. We use the term “substrate destabilization” instead of “ground state destabilization” to describe the effects of substitutions for Asp 70 in MtOMPDC on the kinetic constants.

5Tsang, W.-Y., Amyes, T. L., Wood, B. M., Chan, K., Wu, W., Gerlt, J. A., and Richard, J. P., manuscript in preparation.

REFERENCES

1. Wood BM, Chan KK, Amyes TL, Richard JP, Gerlt JA. Mechanism of the Orotidine 5’ Monophosphate Decarboxylase Catalyzed Reaction: Effect of Solvent Viscosity on Kinetic Constants. Biochemistry. 2009;48 preceding manuscript. [PMC free article] [PubMed]
2. Radzicka A, Wolfenden R. A proficient enzyme. Science. 1995;267:90–93. [PubMed]
3. Miller BG, Wolfenden R. Catalytic proficiency: the unusual case of OMP decarboxylase. Annu Rev Biochem. 2002;71:847–885. [PubMed]
4. Traut TW, Temple BR. The chemistry of the reaction determines the invariant amino acids during the evolution and divergence of orotidine 5'-monophosphate decarboxylase. J Biol Chem. 2000;275:28675–28681. [PubMed]
5. Miller BG, Hassell AM, Wolfenden R, Milburn MV, Short SA. Anatomy of a proficient enzyme: the structure of orotidine 5'- monophosphate decarboxylase in the presence and absence of a potential transition state analog. Proc Natl Acad Sci U S A. 2000;97:2011–2016. [PubMed]
6. Wu N, Mo Y, Gao J, Pai EF. Electrostatic stress in catalysis: structure and mechanism of the enzyme orotidine monophosphate decarboxylase. Proc Natl Acad Sci U S A. 2000;97:2017–2022. [PubMed]
7. Appleby TC, Kinsland C, Begley TP, Ealick SE. The crystal structure and mechanism of orotidine 5'-monophosphate decarboxylase. Proc Natl Acad Sci U S A. 2000;97:2005–2010. [PubMed]
8. Harris P, Navarro Poulsen JC, Jensen KF, Larsen S. Structural basis for the catalytic mechanism of a proficient enzyme: orotidine 5'-monophosphate decarboxylase. Biochemistry. 2000;39:4217–4224. [PubMed]
9. Miller BG, Snider MJ, Short SA, Wolfenden R. Contribution of enzyme-phosphoribosyl contacts to catalysis by orotidine 5'-phosphate decarboxylase [In Process Citation]. Biochemistry. 2000;39:8113–8118. [PubMed]
10. Miller BG, Butterfoss GL, Short SA, Wolfenden R. Role of enzyme-ribofuranosyl contacts in the ground state and transition state for orotidine 5'-monophosphate decarboxylase: A role for substrate destabilization? Biochemistry. 2001;40:6227–6232. [PubMed]
11. Miller BG, Snider MJ, Wolfenden R, Short SA. Dissecting a charged network at the active site of orotidine-5'- phosphate decarboxylase. Biochemistry. 2001;30:15174–15176. [PubMed]
12. Wu N, Gillon W, Pai EF. Mapping the active site-ligand interactions of orotidine 5'-monophosphate decarboxylase by crystallography. Biochemistry. 2002;41:4002–4011. [PubMed]
13. Warshel A, Strajbl M, Villa J, Florian J. Remarkable rate enhancement of orotidine 5'-monophosphate decarboxylase is due to transition-state stabilization rather than to ground-state destabilization. Biochemistry. 2000;39:14728–14738. [PubMed]
14. Gao J. Catalysis by enzyme conformational change as illustrated by orotidine 5'-monophosphate decarboxylase. Curr Opin Struct Biol. 2003;13:184–192. [PubMed]
15. Feng WY, Austin TJ, Chew F, Gronert S, Wu W. The mechanism of orotidine 5'-monophosphate decarboxylase: catalysis by destabilization of the substrate. Biochemistry. 2000;39:1778–1783. [PubMed]
16. Toth K, Amyes TL, Wood BM, Chan K, Gerlt JA, Richard JP. Product deuterium isotope effect for orotidine 5'-monophosphate decarboxylase: evidence for the existence of a short-lived carbanion intermediate. J Am Chem Soc. 2007;129:12946–12947. [PMC free article] [PubMed]
17. Amyes TL, Wood BM, Chan K, Gerlt JA, Richard JP. Formation and stability of a vinyl carbanion at the active site of orotidine 5'-monophosphate decarboxylase: pKa of the C-6 proton of enzyme-bound UMP. J Am Chem Soc. 2008;130:1574–1575. [PMC free article] [PubMed]
18. Sievers A, Wolfenden R. Equilibrium of formation of the 6-carbanion of UMP, a potential intermediate in the action of OMP decarboxylase. J Am Chem Soc. 2002;124:13986–13987. [PubMed]
19. Yeoh FY, Cuasito RR, Capule CC, Wong FM, Wu W. Carbanions from decarboxylation of orotate analogs: stability and mechanistic implications. Bioorg Chem. 2007;35:338–343. [PMC free article] [PubMed]
20. Lee JK, Houk KN. A proficient enzyme revisited: the predicted mechanism for orotidine monophosphate decarboxylase. Science. 1997;276:942–945. [PubMed]
21. Houk KN, Lee JK, Tantillo DJ, Bahmanyar S, Hietbrink BN. Crystal structures of orotidine monophosphate decarboxylase: does the structure reveal the mechanism of nature's most proficient enzyme? Chembiochem. 2001;2:113–118. [PubMed]
22. Callahan BP, Bell AF, Tonge PJ, Wolfenden R. A Raman-active competitive inhibitor of OMP decarboxylase. Bioorg Chem. 2006;34:59–65. [PubMed]
23. Lewis CA, Jr., Wolfenden R. Indiscriminate binding by orotidine 5'-phosphate decarboxylase of uridine 5'-phosphate derivatives with bulky anionic c6 substituents. Biochemistry. 2007;46:13331–13343. [PMC free article] [PubMed]
24. Ames BN. Assay of Inorganic Phosphate, Total Phosphate and Phosphatases. 1966. pp. 115–118.
25. Hruska FE. Molecular conformation of orotidine, a naturally occurring nucleoside, in syn conformation in aqueous solution. J. Am. Chem. Soc. 1971;93:1795–1797. [PubMed]
26. Fujita K, Matsukawa A, Shibata K, Tanaka T, Taniguchi M, Oi S. Synthesis of 5-fluorouridine 5'-diphosphate galactose from 5-fluorouridine by chemical phosphorylation and microbial uridylyl transfer. Carbohydr Res. 1994;265:299–302. [PubMed]
27. Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. In: Carter CWJ, Sweet RM, Abelson JN, Simon MI, editors. Methods in Enzymology. Academic Pres; New York: 1997. pp. 307–326.
28. Long F, Vagin AA, Young P, Murshudov GN. BALBES: a molecular-replacement pipeline. Acta Crystallogr D Biol Crystallogr. 2008;64:125–132. [PubMed]
29. Jones AT. Interactive computer graphics: FRODO. Methods Enzymol. 1985;115:157–171. [PubMed]
30. Brunger AT, Adams PD, Clore GM, DeLAno WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, Read RJ, Rice LM, Simonson T, Warren GL. Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr. 1998;D54:905–921. [PubMed]
31. Lamzin VS, Wilson KS. Automated refinement of protein models. Acta Crystallogr D Biol Crystallogr. 1993;49:129–147. [PubMed]
32. CCP4 The CCP4 suite: Programs for protein crystallography. Acta Crystallogr. 1994;D50:760–763. [PubMed]
33. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem. 2004;25:1605–1612. [PubMed]
34. Shostak K, Jones ME. Orotidylate decarboxylase: insights into the catalytic mechanism from substrate specificity studies. Biochemistry. 1992;31:12155–12161. [PubMed]
35. Van Vleet JL, Reinhardt LA, Miller BG, Sievers A, Cleland WW. Carbon isotope effect study on orotidine 5'-monophosphate decarboxylase: support for an anionic intermediate. Biochemistry. 2008;47:798–803. [PubMed]
36. Sievers A, Wolfenden R. The effective molarity of the substrate phosphoryl group in the transition state for yeast OMP decarboxylase. Bioorg Chem. 2005;33:45–52. [PubMed]
37. Amyes TL, Richard JP, Tait JJ. Activation of orotidine 5'-monophosphate decarboxylase by phosphite dianion: the whole substrate is the sum of two parts. J Am Chem Soc. 2005;127:15708–15709. [PubMed]
38. Barnett SA, Amyes TL, Wood BM, Gerlt JA, Richard JP. Dissecting the total transition state stabilization provided by amino acid side chains at orotidine 5'-monophosphate decarboxylase: a two-part substrate approach. Biochemistry. 2008;47:7785–7787. [PMC free article] [PubMed]
39. Porter DJ, Short SA. Yeast orotidine-5'-phosphate decarboxylase: steady-state and pre-steady- state analysis of the kinetic mechanism of substrate decarboxylation. Biochemistry. 2000;39:11788–11800. [PubMed]