Catalysis by Orotidine 5′-Monophosphate Decarboxylase: Effect of 5-Fluoro and 4′-Substituents on the Decarboxylation of Two-Part Substrates

Abstract

The syntheses of two novel truncated analogs of the natural substrate orotidine 5′-monophosphate (OMP) for orotidine 5′-monophosphate decarboxylase (OMPDC) with enhanced reactivity towards decarboxylation are reported: 1-(β-D-erythrofuranosyl)-5-fluoroorotic acid (FEO) and 5′-deoxy-5-fluoroorotidine (5′-dFO). A comparison of the second-order rate constants for the OMPDC-catalyzed decarboxylations of FEO (10 M⁻¹ s⁻¹) and 1-(β-D-erythrofuranosyl)orotic acid (EO, 0.026 M⁻¹ s⁻¹) shows that the vinyl carbanion-like transition state is stabilized by 3.5 kcal/mol by interactions with the 5-F substituent of FEO. The OMPDC-catalyzed decarboxylations of FEO and EO are both activated by exogenous phosphite dianion (HPO₃²⁻), but the 5-F substituent results in only a 0.8 kcal stabilization of the transition state for the phosphite-activated reaction of FEO. This provides strong evidence that the phosphite-activated OMPDC-catalyzed reaction of FEO is not limited by the chemical step of decarboxylation of the enzyme-bound substrate. Evidence is presented that there is a change in rate-limiting step from the chemical step of decarboxylation for the phosphite-activated reaction of EO, to closure of the phosphate gripper loop and an enzyme conformational change at the ternary E·FEO·HPO₃²⁻ complex for the reaction of FEO. The 4′-CH₃ and 4′-CH₂OH groups of 5′-dFO and orotidine, respectively, result in identical destabilizations of the transition state for the unactivated decarboxylation of 2.9 kcal/mol. By contrast, the 4′-CH₃ group of 5′-dFO and the 4′-CH₂OH group of orotidine result in very different 4.7 and 8.3 kcal/mol destabilizations of the transition state for the phosphite-activated decarboxylation. Here, the destabilizing effect of the 4′-CH₃ substituent at 5′-dFO is masked by the rate-limiting conformational change that depresses the third-order rate constant for the phosphite-activated reaction of the parent substrate FEO.

Orotidine 5′-monophosphate decarboxylase (OMPDC) employs no metal ions or other cofactors yet effects an enormous 10¹⁷-fold acceleration of the decarboxylation of enzyme-bound orotidine 5′-monophosphate (OMP, Scheme 1A) to give uridine 5′-monophosphate (UMP).^1,2 We have shown that the enzymatic reaction proceeds by a stepwise mechanism through a vinyl carbanion intermediate,^3-6 and the overall rate acceleration corresponds to a 31 kcal/mol stabilization of the transition state by interactions of OMP with the protein catalyst.^1,2 The functionalities present in the substrate OMP are, in principle, sufficient to account for this large transition state binding energy, but the mechanism for the increase in the stabilizing binding interactions with the protein from ca. 8 kcal/mol at the ground state Michaelis complex (K_m = 1.4 μM)^7,8 to 31 kcal/mol at the transition state for decarboxylation is poorly understood.^9,10 This specificity in transition state binding is a feature of enzyme-catalyzed reactions that exhibit large transition state binding energies, and is required to prevent the irreversible formation of tight Michaelis complexes with the reactant and/or product.

Scheme 1.

We have developed novel experimental protocols for study of the mechanism of action of OMPDC and have used these to characterize the enzyme specificity in binding the transition state as follows. First, we first characterized the interactions between the phosphodianion group of OMP and OMPDC, by evaluating the effect of removal of the 4′-CH₂OPO₃²⁻ group of OMP to give the truncated neutral substrate 1-(β-D-erythrofuranosyl)orotic acid (EO, Scheme 1B). The second-order rate constant for OMPDC-catalyzed decarboxylation of EO is 4 × 10⁸-fold smaller than that for OMP, which shows that interactions of the protein with the 4′-CH₂OPO₃²⁻ group of OMP provide a ca. 12 kcal/mol stabilization of the transition state for decarboxylation.¹¹ However, these interactions do not simply anchor OMP to OMPDC, because ca. 8 kcal/mol of this phosphodianion binding energy is recovered as transition state stabilization from the binding of phosphite dianion (HPO₃²⁻) to the transition state in the phosphite-activated decarboxylation of EO (Scheme 1B).^7,11-13 The activation of OMPDC by phosphite dianion results from the high specificity of the dianion for binding to the transition state for the OMPDC-catalyzed decarboxylation of EO, compared with the weak binding of phosphite to the free enzyme (K_d ≈ 0.1 M).¹¹ We propose that OMPDC exhibits the same specificity for the tethered 4′-CH₂OPO ²⁻₃ group of OMP so that the binding interactions of this substrate moiety are utilized for transition state stabilization.^10,13

Second, OMPDC was shown to provide effective catalysis of the exchange for deuterium of the C-6 protons of UMP and 5-fluorouridine 5′-monophosphate (FUMP) in D₂O (Scheme 2A).^3,5,14 The kinetic data for the enzymatic and nonenzymatic deuterium exchange reactions of UMP and FUMP show that OMPDC stabilizes the vinyl carbanion intermediate relative to the bound substrate by at least 13 kcal/mol.^3,5 This strong stabilization of the UMP and FUMP vinyl carbanion intermediates (Scheme 2A) is reflected in the high specificity of OMPDC for binding the carbanion-like transition state of the enzyme-catalyzed deuterium exchange reactions.^15,16

Scheme 2.

Third, these protocols were combined in a study of the OMPDC-catalyzed exchange for deuterium of the C-6 proton of the truncated substrate 1-(β-D-erythrofuranosyl)-5-fluorouracil (FEU) in D₂O (Scheme 2B).¹⁷ The observed activation of this reaction by phosphite dianion shows that there is a large ca. 6 kcal/mol stabilization of the vinyl carbanion-like transition state by interactions with HPO₃²⁻, which is similar to the ca. 8 kcal/mol stabilization of the transition state for the phosphite-activated decarboxylation of EO.¹¹ We concluded that phosphite dianion exhibits the same high specificity for binding to the vinyl carbanion-like transition states that are common to the OMPDC-catalyzed decarboxylation and C-6 deuterium exchange reactions (Schemes 1B and 2B).

The addition of a 5-fluoro substituent to UMP to give FUMP results in a 3400-fold increase in the rate constant for the OMPDC-catalyzed C-6 deuterium exchange reaction of the enzyme-bound nucleotide in D₂O, as a result of stabilization of the negative charge at the vinyl carbanion intermediate by interaction with the electron-withdrawing 5-F substituent.^3,5 By contrast, the stabilization of the vinyl carbanion-like transition state for the OMPDC-catalyzed decarboxylation of OMP by a 5-F substituent is not easily quantified, because the decarboxylation of 5-fluoroorotidine 5′-monophosphate (FOMP) is limited by substrate binding (k_cat/K_m) or product release (k_cat).¹⁸ We report here that the second-order rate constant for the OMPDC-catalyzed decarboxylation of the novel truncated substrate 1-(β-D-erythrofuranosyl)-5-fluoroorotic acid (FEO, Scheme 1B) is 400-fold larger than that for the decarboxylation of EO. Surprisingly, we find that the third-order rate constant for the phosphite!activated decarboxylation of FEO is only 4-fold larger than that for the decarboxylation of EO. The obliteration of the 5-F substituent effect is attributed to a change in rate-limiting step for the phosphite-activated reaction, from the chemical step of decarboxylation for EO to an obligate conformational change for the reaction of FEO.

We also report the effect of 4′-substituents on the kinetic parameters for the OMPDC-catalyzed decarboxylation reactions of EO and FEO. The addition of a 4′-CH₂OH group to EO to give orotidine or of a 4′-CH₃ group to FEO to give 5′-deoxy-5-fluooroorotidine (5′-dFO) results in identical destabilizations of the transition state for OMPDC-catalyzed decarboxylation of 2.9 kcal/mol. These surprisingly large 4′-CH₂OH and 4′-CH₃ substituent effects provide striking evidence that OMPDC-catalyzed decarboxylation proceeds through a transition state that is stabilized by a highly organized, but easily disrupted, network of protein-ligand interactions that extend from the substrate phosphodianion group to the pyrimidine binding loci at OMPDC.

EXPERIMENTAL SECTION

Materials

Orotidine 5′-monophosphate and 1-(β-D-erythrofuranosyl)-5-fluorouracil (FEU) were available from our earlier studies.^4,7,17,19 5′-Deoxy-5-fluorouridine (5′-dFU, 98%) was purchased from AK Scientific. Orotidine, 5-fluorouridine (≥ 99.0%), ethyl cyanoformate (99%), diisopropylamine (redistilled, 99.95%), butyllithium (1.6 M solution in hexanes) and 3-(N-morpholino)propanesulfonic acid (MOPS, ≥ 99.5%) were purchased from Sigma-Aldrich. Sodium phosphite (dibasic, pentahydrate) was purchased from Riedel-de Haën (Fluka). Acetone was dried over calcium sulfate (CaSO₄) and distilled prior to use. Water was from a Milli-Q Academic purification system. All other chemicals were reagent grade or better and were used without further purification.

Wildtype orotidine 5′-monophosphate decarboxylase from Saccharomyces cerevisiae (OMPDC) was prepared as described previously.^12,18 The protein sequence differs from the published sequence for wildtype yeast OMPDC by the following mutations: S2H, C155S, A160S and N267D.^12,18 The C155S mutation does not affect the kinetic parameters or the overall structure of the enzyme.²⁰

Chemical Syntheses

Our syntheses of the novel compounds 1-(β-D-erythrofuranosyl)-5-fluoroorotic acid (FEO) and 5′-deoxy-5-fluoroorotidine (5′-dFO) are summarized in Scheme 3. Lithium diisopropylamide (LDA) was prepared from redistilled diisopropylamine and butyllithium (BuLi, 1.6 M solution in hexanes) by the following procedure: BuLi (2 mL, 3.2 mmol) was added slowly to a solution of diisopropylamine (0.5 mL, 3.6 mmol) in dry tetrahydrofuran (THF, 30 mL) under argon at −78 °C. The solution was stirred under argon for 1 h at −78 °C and the whole was used immediately as described below.

Scheme 3.

Reagents and conditions: (a) H₂SO₄, acetone, room temperature, 1.5 h, ~ 90%; (b) (i) LDA, THF, −78 °C, 2 h; (ii) EtOC(O)CN, THF, −78 °C, 50 min, 70–75%; (c) (i) NaOH/THF, room temperature, 3 h; (ii) Dowex 50WX8-100 (H⁺), H₂O, 45 °C, 90–100%.

¹H and ¹⁹F NMR spectra were acquired on a Varian Unity Inova 500 spectrometer operating at 500 MHz for proton. Proton chemical shifts (δ ppm) were referenced to the appropriate solvent peak: 4.69 ppm for HOD in D₂O, 2.54 ppm for d₅-DMSO in d₆-DMSO, and 7.26 ppm for CHCl₃ in CDCl₃. Proton chemical shift assignments were confirmed using two-dimensional ¹H-¹H correlated spectroscopy (¹H-¹H COSY).

2′,3′-O-Isopropylidene-1-(β-D-erythrofuranosyl)-5-fluorouracil (1a)

The ribosyl hydroxyl groups of FEU and 5′-dFU were protected using adaptations of published procedures as follows.²¹ FEU (0.31 g, 1.34 mmol) was dried in vacuo over activated molecular sieves (4 Å) for 24 h at 50 °C and then dissolved in dry acetone (22 mL). The reaction was initiated by the slow addition of concentrated H₂SO₄ (0.16 mL) under argon at 0 °C. After 80 min at room temperature the solution was poured into saturated aqueous NaHCO₃ (10 mL), stirred for 25 min at room temperature, and extracted with chloroform (4 × 30 mL). The combined organic extracts were dried (MgSO₄) and the solvent was removed by evaporation under reduced pressure to give 1a as a white solid (0.34 g, 1.25 mmol, 93%) which was used without further purification. ¹H NMR (500 MHz, CDCl₃) δ ppm: 9.16 (s, 1H, NH), 7.34 (d, 1H, J_HF = 6 Hz, C6-H), 5.39 (broad s, 1H, C1′-H), 5.18 (d, 1H, J = 6 Hz, C2′-H), 5.06 (dd, 1H, J = 6, 4 Hz, C3′-H), 4.33 (dd, 1H, J = 10, 4 Hz, C4′-H_A), 4.24 (d, 1H, J = 10 Hz, C4′-H_B), 1.56 (s, 3H, CH₃), 1.39 (s, 3H, CH₃).

5′-Deoxy-2′,3′-O-isopropylidene-5-fluorouridine (1b)

The procedure described above for the synthesis of 1a was followed for the preparation of 1b from 5′-dFU (0.67 g, 2.72 mmol). The yellow solid obtained from chloroform extraction was purified by flash chromatography on silica gel (hexanes/ethyl acetate 1:1) to give 1b as a white solid (0.70 g, 2.44 mmol, 90%): ¹H NMR (500 MHz, CDCl₃) δ ppm: 9.42 (s, 1H, NH), 7.39 (d, 1H, J_HF = 6 Hz, C6-H), 5.70 (d, 1H, J = 2.5 Hz, C1′-H), 4.9 (dd, 1H, J = 6.5, 2.5 Hz, C2′-H), 4.52 (dd, 1H, J = 6.5, 4.5 Hz, C3′-H), 4.27 (dq, 1H, J = 6.5, 4.5 Hz, C4′-H), 1.59 (s, 3H, CCH₃), 1.43 (d, 3H, J = 6.5 Hz, 4′-CH₃), 1.37 (s, 3H, CCH₃).

2′,3′-O-Isopropylidene-1-(β-D-erythrofuranosyl)-5-fluoroorotic acid ethyl ester (2a)

Freshly prepared LDA (3.2 mmol) was added slowly to a solution of 1a (0.34 g, 1.23 mmol) in dry THF (10 mL) under argon at −78 °C. The mixture was stirred for 2 h at −78 °C after which ethyl cyanoformate (0.5 mL, 5.06 mmol) was added. After 50 min the reaction was quenched with water (20 mL), stirred for 20 min at room temperature, followed by neutralization of residual HCN by the addition of solid Na₂CO₃ and extraction with ethyl acetate (4 × 30 mL). The combined organic extracts were dried (MgSO₄) and the solvent was removed by evaporation under reduced pressure to give a brown oil. Purification by flash chromatography on silica gel (hexanes/ethyl acetate 2:1) gave a yellow solid (0.33 g, 78%) which was recrystallized from hexanes/diethyl ether to give 2a as a white solid (0.30 g, 70%): mp 159–161 °C. ¹H NMR (500 MHz, CDCl₃) δ ppm: 8.68 (s, 1H, NH), 5.45 (d, 1H, J = 0.5 Hz, C1′-H), 5.26 (dd, 1H, J = 6.5, 0.5 Hz, C2′-H), 5.09 (m, 1H, C3′-H), 4.52 (q, 2H, J = 7 Hz, CH₂CH₃), 4.44 (dd, 1H, J = 10, 4 Hz, C4′-H_A), 4.15 (d, 1H, J = 10 Hz, C4′-H_B), 1.53 (s, 3H, CCH₃), 1.45 (t, 3H, J = 7 Hz, CH₂CH₃), 1.38 (s, 3H, CCH₃). HRMS (ESI): calculated (M + Na)⁺ C₁₄H₁₇O₇N₂F₁Na₁, 367.0912; found, 367.0908.

5′-Deoxy-2′,3′-O-isopropylidene-5-fluororotidine ethyl ester (2b)

The procedure described above for the preparation of 2a was followed for the preparation of 2b from 1b (0.34 g, 1.33 mmol). The crude product, a dark brown oil, was purified by flash chromatography on silica gel (hexanes/ethyl acetate 2:1) to give a yellow solid (0.39 g, 82%). Recrystallization from hexanes/diethyl ether gave 2b as a white solid (0.36 g, 75%): mp 146–147 °C. ¹H NMR (500 MHz, CDCl₃) δ ppm: 9.23 (s, 1H, NH), 5.75 (d, 1H, J = 2.5 Hz, C1′-H), 5.17 (dd, 1H, J = 6.5, 2.5 Hz, C2′-H), 4.52 (m, 2H, C3′-H and CH₂CH₃), 4.42 (m, 1H, CH₂CH₃), 4.07 (m, 1H, C4′-H), 1.58 (s, 3H, CCH₃), 1.44 (t, 3H, J = 7 Hz, CH₂CH₃), 1.38 (d, 3H, J = 6.5 Hz, 4′-CH₃), 1.37 (s, 3H, CCH₃). HRMS (ESI): calculated (M + Na)⁺ C₁₅H₁₉O₇N₂F₁Na₁, 381.1069; found, 381.1079.

1-(β-D-Erythrofuranosyl)-5-fluoroorotic acid (FEO)

The hydrolyses of the ethyl ester groups of 2a and 2b were carried out in THF containing NaOH. The failure to allow the hydrolysis of 2a to proceed to completion resulted in poor product yields in the subsequent removal of the isopropylidene group. Therefore, the reaction time was optimized by monitoring a small scale reaction in d₈-THF by ¹H NMR, for which the ester hydrolysis was judged to be complete after 2.5 h at room temperature.

2a (132 mg, 0.38 mmol) was dissolved in 2.7 mL of THF and 0.54 mL of 3 M NaOH was added with stirring at room temperature.²² After 3 h at room temperature the solution was poured into a mixture of 4 mL water and 10 mL of wet Dowex 50WX8-100 (H⁺) and stirred for 3 h at room temperature followed by 3 h at 45 °C. The Dowex resin was removed by filtration and washed with water (3 × 20 mL), and the combined aqueous extracts were lyophilized to give FEO as a white solid (104 mg, 0.38 mmol, 100 %): mp 212-214 °C. ¹H NMR (500 MHz, D₂O) δ ppm: 5.60 (d, 1H, J = 6 Hz, C1′-H), 4.57 (dd, 1H, J = 6, 5 Hz, C2′-H), 4.29 (m, 1H, C3′-H), 4.24 (dd, 1H, J = 10, 3.5 Hz, C4′-H_A), 3.84 (dd, 1H, J = 10 Hz, 1.5 Hz, C4′-H_B). ¹⁹F NMR (470 MHz, D₂O) δ ppm: -169.46 (s, 1F, C5-F). HRMS (ESI): calculated (M + 2Na)²⁺ C₉H₈O₇N₂F₁Na₂, 321.0105; found, 321.0100.

5′-Deoxy-5-fluoroorotidine (5′-dFO)

The procedure described above for the preparation of FEO was followed for the preparation of 5′-dFO from 2b (0.040 g, 0.11 mmol), except that removal of the isopropylidene protecting group using Dowex (H⁺) employed a 16 h reaction time at 45 °C. This gave 5′-dFO as a white solid (0.045 g, 0.10 mmol, 90%). ¹H NMR (500 MHz, D₂O) δ ppm: 5.33 (d, 1H, J = 2.5 Hz, C1′-H), 4.57 (dd, 1H, J = 6, 2.5 Hz, C2′-H), 3.99 (m, 1H, C3′-H), 3.80 (m, 1H, C4′-H), 1.26 (d, 3H, J = 6.5 Hz, 4′-CH₃). ¹⁹F NMR (470 MHz, D₂O) δ ppm: −169.61 (s, 1F, C5-F). HRMS (ESI): calculated (M - 1)⁻ C₁₀H₁₀O₇N₂F₁, 289.0478; found, 289.0478.

Kinetic Studies

Preparation of solutions

Solution pH was determined at 25 °C using an Orion Model 720A pH meter equipped with a Radiometer pHC4006-9 combination electrode that was standardized at pH 7.00 and 10.00 at 25 °C. Stock solutions of OMP, FEO, 5′-dFO and orotidine were prepared by dissolution in water followed by adjustment to pH ≈ 7 using 1.0 M NaOH. The stock solution of 5′-dFO was freed of small amounts of contaminating 5′-dFU (≤ 1%) by passage over Supelclean™ LC-18 solid phase extraction tubes (3 mL, Supelco), eluting with water. The solutions were stored at −20 °C. The concentrations of FEO and 5′-dFO were determined from the absorbance in 0.1 M HCl at 271 nm using ε = 10,200 M⁻¹ cm⁻¹ reported for FOMP.⁴ The concentrations of OMP and orotidine were determined from the absorbance in 0.1 M HCl at 267 nm using ε = 9430 M⁻¹ cm⁻¹ for OMP and ε = 9570 M⁻¹ cm⁻¹ for orotidine.²³

Stock solutions of sodium phosphite at pH 7.0 (80% dianion) were prepared by addition of a measured amount of 1 M HCl to the sodium salt to give the desired acid/base ratio. MOPS buffers at pH 7.0 (45% free base) or pH 7.1 (50% free base) were prepared by addition of measured amounts of 1 M NaOH and solid NaCl to give the desired acid/base ratio and ionic strength.

Samples of yeast OMPDC that had been stored at −80 °C were defrosted and exhaustively dialyzed at 7 °C against 50 mM MOPS (45% free base) at pH 7.0 and I = 0.14 (NaCl) for the experiments with FEO and 5′-dFO, or 100 mM MOPS (50% free base) at pH 7.1 and I = 0.28 (NaCl) for the experiments with orotidine. The concentration of OMPDC was determined from the absorbance at 280 nm using an extinction coefficient of 29,910 M⁻¹ cm⁻¹, calculated using the ProtParam tool available on the ExPASy server.²⁴

Standard enzyme assays

The activity of OMPDC in the enzyme stock solutions and in the decarboxylation reaction mixtures was determined by monitoring the decrease in absorbance at 279 nm accompanying the enzyme-catalyzed decarboxylation of OMP (Δε = −2400 M⁻¹ cm⁻¹ at 25 °C).^3,7,8,12 Prior to assay, the stock solutions of OMPDC were diluted with 10 mM MOPS, 50% free base (pH 7.1) containing 100 mM NaCl to give a final concentration of ca. 20 μM OMPDC. Assays in a total volume of 1 mL were conducted in 10 mM MOPS (50% free base) at pH 7.1, 25 °C and I = 0.105 (NaCl), with 40 - 50 μM OMP (25 - 30K_m). Reactions were initiated by the addition of 1 μL of the diluted stock solution of OMPDC or an aliquot of the decarboxylation reaction mixture to give a final enzyme concentration of 20 - 40 nM, and the initial velocity of the ensuing decarboxylation of OMP was determined within 1 min.

Yeast OMPDC-catalyzed decarboxylation of FEO

The decarboxylation of FEO catalyzed by OMPDC at pH 7.0, 25 °C and I = 0.14 (NaCl) was followed spectrophotometrically by monitoring the decrease in absorbance at 282 nm.¹¹ Reactions in the absence of phosphite were initiated by the addition of 100 μL of enzyme to give a reaction mixture (1 mL) containing 25 mM MOPS (45% free base, pH 7.0), 0.14 mM FEO and 30 μM OMPDC at I = 0.14 (NaCl). Reactions in the presence of phosphite were initiated by the addition of 10 - 100 μL of enzyme to give reaction mixtures (1 mL) at pH 7.0 containing 5 mM MOPS (45% free base), 2 - 36 mM HPO₃²⁻, 0.14 mM FEO and 2.7 - 29 μM OMPDC at I = 0.14 (NaCl). The reactions were monitored for 200 – 800 min (≥ 10 halftimes), after which standard assay (see above) showed that OMPDC maintained full activity towards decarboxylation of OMP. In several cases, after completion of the reaction, the enzyme was removed by ultrafiltration (Amicon, 10K MWCO) and the pH of the filtrate was determined. No change in pH was observed during these reactions.

Observed first-order rate constants for enzyme-catalyzed decarboxylation, k_obs (s⁻¹), were obtained from the fits of the progress curves to an equation for a single exponential decay.¹¹ The observed second-order rate constants (k_cat/K_m)_obs (M⁻¹ s⁻¹) were calculated using the relationship (k_cat/K_m)_obs = k_obs/[E].

Yeast OMPDC-catalyzed decarboxylation of 5′-dFO

The decarboxylation of 5′-dFO catalyzed by OMPDC at pH 7.0, 25 °C and I = 0.14 (NaCl) was followed in a discontinuous assay in which the initial velocity of formation of the product 5′-dFU was monitored by HPLC analysis with peak detection at 270 nm, as described previously for the decarboxylation of EO.¹³ Reaction mixtures (200 μL) were prepared by mixing stock solutions of the enzyme in MOPS buffer with stock solutions of MOPS buffer, sodium chloride and, if needed, sodium phosphite (100 mM, 80% dianion), and the reactions were initiated by the addition of an aliquot of the stock solution of 5′-dFO in water. The reaction conditions in the absence of phosphite dianion were 25 mM MOPS (45% free base, pH 7.0), 0.9 – 9 mM 5′-dFO and 65 μM OMPDC at I = 0.14 (NaCl). The reaction conditions in the presence of phosphite dianion were 5 mM MOPS (45% free base, pH 7.0), 4 - 32 mM HPO₃²⁻, 4.4 mM 5′-dFO and 10 - 50 μM OMPDC at I = 0.14 (NaCl). The reactions were followed for up to 3 hours, during which time there was up to ca. 5% reaction. At various times an aliquot (20 μL) was withdrawn and the reaction was quenched by the addition of 180 μL of a quench solution of formic acid (2.4 mM) containing ca. 60 μM 5-fluorouracil that served as an internal standard. The enzyme was removed by ultrafiltration using an Amicon Ultra device (10K MWCO) that had been prewashed 2 to 3 times with 500 μL of water. Failure to prewash the filtration device resulted in loss of retention and/or erratic HPLC retention times. The filtrate (100 μL) was analyzed by HPLC using a Waters Atlantis dC₁₈ 3 Gm column (3.9 × 150 mm) with a linear gradient from 10 mM NH₄OAc pH 4.2 to 60/40 NH₄OAc/MeOH over 10 min, with a flow of 1 mL/min and peak detection at 270 nm. Under these conditions the unreacted 5′-dFO eluted at 2.5 min, 5-fluorouracil eluted at 4.2 min and the product 5′-dFU eluted at 11.8 min. The observed HPLC peak area for 5′-dFU was normalized using the observed peak area for 5-fluorouracil and the standard HPLC peak area for 5-fluorouracil determined by direct HPLC analysis of the formic acid/5-fluorouracil quench solution. This procedure was necessary because there is a small variable dilution of the sample upon its passage through the prewashed filtration device. The concentration of the product 5′-dFU in the reaction mixture at time t, [5′-dFU]_t, was then obtained from its normalized HPLC peak area by interpolation of a standard curve that was constructed using 5-fluorouridine. The concentration of 5-fluorouridine in the stock solutions used for this calibration was determined from the absorbance in 0.1 M HCl at 270 nm using ε = 9660 M⁻¹ cm⁻¹ reported for FEU.¹⁷ Periodic standard assay of the reaction mixture using OMP as substrate (see above) showed that there was no significant decrease in the activity of OMPDC during these reactions.

Initial velocities of the decarboxylation of 5′-dFO, v_o (M s⁻¹), were determined as the slopes of linear plots of [5′-dFU]_t against time that covered up to 5% reaction. The observed second-order rate constants (k_cat/K_m)_obs (M⁻¹ s⁻¹) were determined using the relationship (k_cat/K_m)_obs = v_o/[E][5′-dFO]_o, where [5′-dFO]_o is the concentration of 5′-dFO at zero time.

Yeast OMPDC-catalyzed decarboxylation of orotidine

The decarboxylation of orotidine catalyzed by OMPDC at pH 7.1, 25 °C and I = 0.14 or 0.21 (NaCl) was followed in a discontinuous assay in which the initial velocity of formation of the product uridine was monitored by HPLC analysis with peak detection at 262 nm.¹¹ Reaction mixtures (160 μL) were prepared by mixing stock solutions of the enzyme in MOPS buffer with water and, if needed, sodium phosphite (200 μM, 80% dianion), and the reactions were initiated by the addition of an aliquot of the stock solution of orotidine in water. The reaction conditions in the absence of phosphite dianion were 50 mM MOPS (50% free base, pH 7.1), 10.3 mM orotidine and 137 μM OMPDC at I = 0.14 (NaCl). The reaction conditions in the presence of phosphite dianion were 25 mM MOPS (50% free base, pH 7.1), 40 mM HPO₃²⁻, 10.3 mM orotidine and 73 μM OMPDC at I = 0.21 (NaCl). The reactions were followed for ca. 30 h, during which time there was ca. 0.5% reaction. At various times an aliquot (20 μL) was withdrawn and the reaction was quenched by the addition of 180 μL of a quench solution of formic acid (6 – 12 mM). The enzyme was removed by ultrafiltration and the sample analyzed by HPLC, as described previously for the decarboxylation of EO.¹¹ The concentration of the product uridine in the reaction mixture at time t, [uridine]_t, was obtained from the HPLC peak area by interpolation of a standard curve. Periodic standard assay of the reaction mixture using OMP as substrate (see above) showed that there was no significant decrease in the activity of OMPDC during these reactions.

Initial velocities of the decarboxylation of orotidine were determined as the slopes of linear plots of [uridine]_t against time that covered up to 0.5% reaction. The observed second-order rate constants (k_cat/K_m)_obs (M⁻¹ s⁻¹) were determined using the relationship (k_cat/K_m)_obs = v_o/[E][orotidine]_o, where [orotidine]_o is the concentration of orotidine at zero time.

RESULTS

The novel compounds FEO and 5′-dFO were synthesized in good yields using the procedures outlined in Scheme 3. These syntheses represent a novel approach to nucleosides containing an orotate or a 5-fluoroorotate base.

The relatively rapid yeast OMPDC-catalyzed decarboxylation reactions of FEO (0.14 mM) to give FEU at pH 7.0, 25 °C and I = 0.14 (NaCl) in the absence and presence of phosphite dianion were followed spectrophotometrically by monitoring the decrease in absorbance at 282 nm. These reactions were cleanly first-order for the entire reaction time course (≥ 10 halftimes). Observed first-order rate constants for enzyme-catalyzed decarboxylation, k_obs(s⁻¹), were obtained from the fits of the data to a single exponential.¹¹ The observed second-order rate constants, (k_cat/K_m)_obs (M⁻¹ s⁻¹), were calculated using the relationship (k_cat/K_m)_obs = k_obs/[E]. The second-order rate constant for the OMPDC-catalyzed decarboxylation of FEO in the absence of phosphite dianion was determined as (k_cat/K_m)_E = 10 M⁻¹ s⁻¹ (Scheme 4).

Scheme 4.

Figure 1A shows the dependence of the observed second-order rate constant (k_cat/K_m)_obs for OMPDC-catalyzed decarboxylation of FEO on the concentration of added phosphite dianion at pH 7.0, 25 °C and a constant ionic strength of 0.14 (NaCl). There is weak downward curvature at high concentrations of phosphite, which is consistent with partial saturation of the enzyme by HPO₃²⁻ to give the E·HPO₃²⁻ complex (Scheme 4). The solid line shows the nonlinear least squares fit of the data to eq 1, derived for Scheme 4, with (k_cat/K_m)_E = 10 M⁻¹ s⁻¹, which gave K_d = 0.1 M for binding of phosphite to OMPDC, and (k_cat/K_m)_E·HPi = 5000 M⁻¹ s⁻¹ for decarboxylation catalyzed by the phosphite-liganded enzyme E·HPO₃²⁻ (Scheme 4). However, it is also possible that the downward curvature results from a specific salt effect arising from the replacement of Cl⁻ with HPO₃.^2-7 Therefore, the third-order rate constant for the phosphite-activated reaction of FEO was determined as the slope of the linear correlation of the data at [HPO₃²⁻] ≤ 12 mM, as (k_cat/K_m)_E·HPi/K_d = 4.8 × 10⁴ M⁻² s⁻¹. Table 1 reports the rate constants for the unactivated and phosphite-activated reactions of FEO that were determined in these experiments.

$${\left(\frac{k_{\text{cat}}}{K_{\mathrm{m}}}\right)}_{\mathrm{obs}}=\left(\frac{K_{\mathrm{d}}}{K_{\mathrm{d}}+\left[{\mathrm{HPO}}_3^{2-}\right]}\right){\left(\frac{k_{\text{cat}}}{K_{\mathrm{m}}}\right)}_{\mathrm{E}}+\left(\frac{\left[{\mathrm{HPO}}_3^{2-}\right]}{K_{\mathrm{d}}+\left[{\mathrm{HPO}}_3^{2-}\right]}\right){\left(\frac{k_{\text{cat}}}{K_{\mathrm{m}}}\right)}_{\mathrm{E}\cdot \mathrm{HPi}}$$ (1)

Figure 1.

Dependence of the observed second-order rate constant for the yeast OMPDC-catalyzed decarboxylation of truncated substrates on the concentration of added phosphite dianion at pH 7.0, 25 °C and I = 0.14 (NaCl). A. Decarboxylation of FEO. The solid line shows the fit of the data to eq 1, derived for Scheme 4, with (k_cat/K_m)_E = 10 M⁻¹ s⁻¹ and K_d = 0.1 M. The dashed line is the linear correlation at [HPO₃²⁻] ≤ 12 mM, the slope of which gives the third-order rate constant (k_cat/K_m)_E·HPi/K_d = 4.8 × 10⁴ M⁻² s⁻¹. B. Decarboxylation of 5′-dFO. The solid line shows the fit of the data to eq 1 with (k_cat/K_m)_E = 0.078 M⁻¹ s⁻¹ and K_d = 0.05 M. The dashed line is the linear correlation at [HPO₃²⁻] ≤ 8 mM, the slope of which gives the third-order rate constant (k_cat/K_m)_E·HPi/K_d = 16 M⁻² s⁻¹.

Table 1.

Kinetic parameters for the decarboxylation of truncated substrates catalyzed by yeast OMPDC and for activation by phosphite dianion at 25 °C.^a

	EO	FEO	Orotidine	5′-dFO
(kcat/Km)Eb (M−1 s−1)	0.026 c	10	1.8 × 10−4 d	0.078
(kcat/Km)E·HPi/Kd e (M−2 s−1)	1.2 × 104 f	4.8 × 104	9.3 × 10−3 g	16
Activation by 1 M HPO32−	5 × 105-fold	5 × 103-fold	50-fold	200-fold
Intrinsic Phosphite Binding Energy (kcal/mol)	7.7	5.0	2.3	3.2

^aAt pH 7.0 and I = 0.14 (NaCl), unless noted otherwise.

^bSecond-order rate constant for the unactivated OMPDC-catalyzed decarboxylation of the truncated substrate in the absence of phosphite. The estimated error is ± 10%.

^cData from Ref. 13.

^dAt pH 7.1.

^eThird-order rate constant for the phosphite-activated OMPDC-catalyzed decarboxylation of the truncated substrate. The estimated error is ± 20%.

^fData from Ref. 11.

^gAt pH 7.1 and I = 0.21 (NaCl).

The initial velocities, v_o(M s⁻¹), of the yeast OMPDC-catalyzed decarboxylation of 5′-dFO (up to 9 mM) at pH 7.0, 25 °C and I = 0.14 (NaCl) were determined by following the formation of up to 5% of the product 5′-dFU by HPLC. Observed second-order rate constants were determined using the relationship (k_cat/K_m)_obs = v_o/[E][5′-dFO]_o, where [5′-dFO]_o is the concentration of 5′-dFO at zero time. The second-order rate constant for the OMPDC-catalyzed decarboxylation of 5′-dFO in the absence of phosphite dianion was determined as (k_cat/K_m)_E = 0.078 M⁻¹ s⁻¹ (Scheme 4). There was no significant change in the value of (k_cat/K_m)_E when the concentration of 5′-dFO was increased from 0.9 to 9 mM, which shows that there is no significant saturation of OMPDC by 9 mM 5′-dFO.¹¹

Figure 1B shows the dependence of the observed second-order rate constant (k_cat/K_m)_obs for OMPDC-catalyzed decarboxylation of 5′-dFO on the concentration of added phosphite dianion at pH 7.0, 25 °C and a constant ionic strength of 0.14 (NaCl). The modest downward curvature is consistent with K_d = 0.050 M (Scheme 4). The third-order rate constant for the phosphite-activated reaction of 5′-dFO was determined as the slope of the linear correlation of the data at [HPO₃²⁻] ≤ 8 mM, as (k_cat/K_m)_E·HPi/K_d = 16 M⁻² s⁻¹. Table 1 reports the rate constants for the unactivated and phosphite-activated reactions of 5′-d-FO that were determined in these experiments.

The second-order rate constant for the yeast OMPDC-catalyzed decarboxylation of orotidine to give uridine at pH 7.1, 25 °C and I = 0.14 (NaCl) in the absence of phosphite dianion was determined by our HPLC assay as (k_cat/K_m)_E = 1.8 × 10⁻⁴ M⁻¹ s⁻¹ (Table 1). This is in good agreement with the earlier value of 2 × 10⁻⁴ M⁻¹ s⁻¹ determined under similar conditions by following the release of ¹⁴CO₂ that accompanies the decarboxylation of orotidine labeled at the 6-CO₂⁻ group.²⁰ The observed second-order rate constant for the decarboxylation of orotidine in the presence of 40 mM phosphite dianion at pH 7.0, 25 °C and I = 0.21 (NaCl) was determined as (k_cat/K_m)_obs = 5.6 × 10⁻⁴ M⁻¹ s⁻¹. This was used to estimate the third-order rate constant for the phosphite-activated reaction of orotidine as (k_cat/K_m)_E·HPi/K_d = [(k_cat/K_m)_obs - (k_cat/K_m)_E]/0.04 = 9.3 × 10⁻³ M⁻² s⁻¹ (Table 1).

DISCUSSION

We showed previously that a comparison of the effects of site-directed mutations of OMPDC on the kinetic parameters for decarboxylation of the whole substrate OMP and on the decarboxylation of the truncated substrate EO activated by the second substrate piece HPO₃²⁻ allows for a clean distinction between the amino acid side chains at OMPDC that interact with the substrate phosphodianion and those that interact with the pyrimidine ring.^12,13 However, the severity of the mutations that can be examined is limited by the low reactivity of the truncated substrate EO, for which the second-order rate constant for unactivated decarboxylation catalyzed by wildtype yeast OMPDC is (k_cat/K_m)_E = 0.026 M⁻¹ s⁻¹.^11,13 The desire to characterize an expanded range of mutations prompted us to develop a synthesis of the more reactive 5-fluoro substituted truncated substrate FEO (Scheme 3). The high reactivity of this substrate in enzyme-catalyzed decarboxylation has enabled the study of severely impaired site-directed mutants of OMPDC from Methanothermobacter thermautotrophicus, which show extremely low activities for the decarboxylation of EO.²⁵

5-Fluoro Substituent Effects

The addition of a 5-F substituent to EO to give FEO results in a 390-fold increase in the second-order rate constant for the unactivated yeast OMPDC-catalyzed decarboxylation through a carbanion intermediate (Scheme 1B), from (k_cat/K_m)_E = 0.026 M⁻¹ s⁻¹ for EO to 10 M⁻¹ s⁻¹ for FEO (Tables 1 and 2). Therefore, the 5-F substituent provides a 3.5 kcal/mol stabilization of the carbanion-like transition state for enzyme-catalyzed decarboxylation (Table 2), which suggests that the chemical step of decarboxylation of the enzyme-bound substrate to give the enzyme-bound product EU or FEU is cleanly rate-determining for these reactions. Similarly, the addition of a 5-F substituent to UMP to give FUMP results in a 3400-fold increase in the rate constant for the OMPDC-catalyzed C-6 deuterium exchange reaction of the enzyme-bound nucleotide in D₂O (Scheme 2A),^3,5 so that the negative charge in the transition state for formation of the vinyl carbanion intermediate is stabilized by 4.8 kcal/mol by interactions with the 5-F substituent (Table 2). These substituent effects on decarboxylation and deuterium exchange provide a measure of the change in the interactions of the reacting center with the 5-F substituent on proceeding from the reactant to the carbanion-like transition state in each case.²⁶ The effect of a 5-F substituent on the OMPDC-catalyzed decarboxylation of OMP is ca. 1 kcal/mol smaller than that on the deuterium exchange reaction of UMP (Table 2). This suggests that the gain in the stabilizing interactions of the developing negative charge with the 5-F substituent on proceeding to the transition state for decarboxylation is partly offset by the loss of ground state stabilizing interactions of the 5-F substituent with the anionic 6-CO₂⁻ group of the substrate.

Table 2.

Effect of a 5-fluoro substituent on decarboxylation and C-6 deuterium exchange reactions catalyzed by yeast OMPDC at 25 °C.

Substrate	Reaction	Rate Constant	Effect of 5-F Substituent	ΔΔG‡ (kcal/mol)
OMP	Decarboxylation	kcat/Km (M−1 s−1)	None a	0
EO	Decarboxylation	(kcat/Km)E (M−1 s−1)	+ 390-fold b	−3.5
EO + HPO32−	Decarboxylation	(kcat/Km)E·HPi/Kd (M−2 s−1)	+ 4-fold b	−0.8
UMP	Exchange	kex (s−1)	+ 3400-fold c	−4.8

^aCalculated from the values of k_cat/K_m for the decarboxylation reactions of OMP (1.1 × 10⁷ M⁻¹ s⁻¹, Ref. 7) and FOMP (1.2 × 10⁷ M⁻¹ s⁻¹, Ref. 27) at pH 7.1 and I = 0.105 (NaCl).

^bCalculated from the data in Table 1 for the unactivated and phosphite-activated reactions of EO and FEO at pH 7.0 and I = 0.14 (NaCl).

^cCalculated from the pD-independent values of k_ex for the C-6 deuterium exchange reactions of enzyme-bound FUMP (0.041 s⁻¹) and UMP (1.2 × 10⁻⁵ s⁻¹) in D₂O at I = 0.10 (NaCl) (data from Ref. 3).

The very similar kinetic parameters for the yeast OMPDC-catalyzed decarboxylation reactions of the natural substrate OMP (k_cat/K_m = 1.1 × 10⁷ M⁻¹ s⁻¹, k_cat = 15 s⁻¹)⁷ and of the 5-F substituted substrate FOMP (k_cat/K_m = 1.2 × 10⁷ M⁻¹ s⁻¹, k_cat = 95 s⁻¹)²⁷ show that there is little or no effect of a 5-F substituent on the stability of the transition state for decarboxylation of the natural substrate (Scheme 1A). The formation of the Michaelis complex is partly rate-determining for k_cat/K_m for turnover of OMP by yeast OMPDC,⁸ so that these essentially identical large values of k_cat/K_m for decarboxylation of OMP and the chemically more reactive FOMP are consistent with cleanly rate-determining formation of the productive Michaelis complex for FOMP, with a second-order rate constant k_d = k_cat/K_m ≈ 10⁷ M⁻¹ s⁻¹ (Scheme 5). The observed 390-fold effect of a 5-F substituent on the decarboxylation of EO can be combined with k_cat = 15 s⁻¹ for decarboxylation of enzyme-bound OMP to estimate a rate constant of k_chem ≈ 6000 s⁻¹ for decarboxylation of enzyme-bound FOMP to give enzyme-bound FUMP (Scheme 5). This is much larger than the observed value of k_cat = 95 s⁻¹ for FOMP,²⁷ and it is consistent with a change in rate-determining step for k_cat from the chemical step of loss of CO₂ for OMP (k_chem, Scheme 5), to product release for the much more reactive substrate FOMP (k_off).¹⁸ These conclusions are supported by the observations of a substantially smaller primary ¹³C isotope effect on k_cat/K_m for the decarboxylation of FOMP than for OMP labeled at the 6-CO₂⁻ group,²⁸ and of larger dependencies on solvent viscosity of the kinetic parameters k_cat/K_m and k_cat for the decarboxylation of FOMP than for OMP.¹⁸

Scheme 5.

By contrast with the relatively large 390-fold difference in the second-order rate constants for the unactivated OMPDC-catalyzed decarboxylation reactions of EO and FEO, for which the chemical step of the loss of CO₂ from the substrate is cleanly rate-determining, there is only a minimal 4-fold increase in the third-order rate constant (k_cat/K_m)_E·HPi/K_d for the phosphite-activated decarboxylation of EO upon the addition of a 5-F substituent to give FEO (Table 1). This provides strong evidence that, by analogy with the decarboxylation of FOMP, the phosphite-activated decarboxylation of FEO is not limited by the chemical step of the loss of CO₂ from the substrate. We conclude that the OMPDC-catalyzed reactions of FOMP and of the corresponding substrate pieces FEO and HPO₃²⁻ are limited by the formation of the reactive binary E·FOMP and ternary E·FEO·HPO₃²⁻ complexes, respectively.

Decarboxylation of FEO Limited by a Conformational Change

We have proposed that OMPDC exists in two forms: a predominant inactive loop open form, E_O, and a rare active loop closed form E_C .^13,17 This conformational change is proposed to result in an environment at the active site that is optimized for catalysis of decarboxylation and deuterium exchange reactions.^13,17 The conversion of E_O to E_C involves, at a minimum, closure of the mobile active site loop (loop 7), which sequesters the bound substrate from bulk solvent.^29-32 There is a large difference between the small observed binding energy of HPO₃²⁻ at the free ground state enzyme (K_d ≈ 0.1 M, Scheme 4 and Figure 1) and the much larger intrinsic phosphite binding energy of 7.7 kcal/mol for the phosphite-activated reaction of the truncated substrate EO (Table 1). This difference is the binding energy that is specifically utilized to stabilize the transition state for enzyme-catalyzed decarboxylation of EO.^11,13 We have proposed that the large intrinsic phosphite dianion binding energy is utilized to drive the unfavorable conformational change at OMPDC, and at other enzymes, which converts E_O to E_C and activates the enzyme for catalysis.^13,17,33-35 Loop closure and the conformational change is driven by the formation of favorable interactions between phosphite dianion and the closed enzyme E_C, and this dianion binding energy is expressed later at the transition state for decarboxylation.^11,13,33

Scheme 6 shows our proposed kinetic pathway for the phosphite-activated OMPDC-catalyzed decarboxylation of FEO. The binding of FEO to the ground state open enzyme E_O is assumed to precede the binding of HPO₃²⁻, since dianion binding triggers loop closure over the active site and presumably occludes the binding of the second substrate piece.^29-32 The subsequent binding of phosphite dianion (k_on) results in the conformational change and loop closure to give the productive ternary Michaelis complex E_C·FEO·HPO₃²⁻. The conclusion that the OMPDC-catalyzed decarboxylation of the substrate pieces FEO + HPO₃²⁻ is limited by the formation of the reactive ternary E_C·FEO·HPO₃²⁻ complex (see above) shows that the chemical step for decarboxylation of FEO is faster than release of phosphite dianion from the enzyme, k_chem > k_off (Scheme 6), so that the binding of phosphite is essentially irreversible. However, this must be reconciled with the observed low affinity of OMPDC for phosphite dianion, K_d = k_off/k_on ≈ 0.1 M, because weak binding is usually associated with fast ligand release to solution, which would tend to strongly favor a rate-determining chemical step.

Scheme 6.

Scheme 6 shows that the overall binding of phosphite involves at least two steps: the initial encounter-limited formation of a weak complex of HPO₃²⁻ with the E_O·FEO complex (k_d), followed by a conformational change (k_c) to give the productive ternary Michaelis complex E_C·FEO·HPO₃²⁻. The first-order rate constant for decarboxylation of enzyme-bound FEO at the ternary complex can be estimated as k_chem ≤ 6000 s⁻¹, with the assumption that the decarboxylation of FEO should be no faster than the decarboxylation of enzyme-bound FOMP (see above). Therefore, the requirement k_chem > k_off and the value of K_d ≈ 0.1 M (we assume that phosphite binds to E_O and E_O·FEO with similar affinities), give k_on ≤ 6 × 10⁴ M⁻¹ s⁻¹ for binding of phosphite dianion to E_O·FEO to give the reactive E_C·FEO·HPO₃²⁻ complex. This slow irreversible formation of the reactive ternary complex strongly suggests that the binding of phosphite is limited not by diffusional encounter (k_d ≈ 10⁷ M⁻¹ s⁻¹), but rather by the conformational change that results in closure of the loop over the active site (k_c, Scheme 6).

The conclusion that the phosphite-activated decarboxylation of FEO is limited by the conformational change that converts the inactive enzyme at the ternary E_O·FEO·HPO₃²⁻ complex to the active enzyme at the E_C·FEO·HPO₃²⁻ complex contrasts the observed encounter-limited formation of the reactive E_C·FOMP complex with k_cat/K_m = k_d ≈ 1 × 10⁷ M⁻¹ s⁻¹ (Scheme 5). These results suggest that there is a larger barrier to loop motion and the conformational change at the E_O·FEO·HPO₃²⁻ complex than at the physiological E_O·FOMP complex, resulting in a change in rate-limiting step for the overall decarboxylation from k_d for the reaction of FOMP (Scheme 5) to k_c for the phosphite-activated reaction of FEO (Scheme 6). We suggest that the slower closure of the loop at E_O·FEO·HPO₃²⁻ than at E_O·FOMP results from a higher entropy of activation for the former process, arising from a greater requirement for “freezing” the conformation of weakly bound HPO₃²⁻ at E_O·FEO·HPO₃²⁻ on proceeding to the tighter E_C·FEO·HPO₃²⁻ complex.

4′-Substituent Effects

Table 3 summarizes the effects of 4′-CH₂OPO₃²⁻, 4′-CH₂OH and 4′-CH₃ substituents on the unactivated and the phosphite-activated OMPDC-catalyzed decarboxylation reactions of the truncated substrates EO and/or FEO.

Table 3.

Effect of 4′-substituents on the decarboxylation reactions of the truncated substrates EO and FEO catalyzed by yeast OMPDC at 25 °C.

Substrate	4′-Substituent	Effect of 4′-Substituent	ΔΔG‡ (kcal/mol)
EO	CH2OPO32−	+ 4.2 × 108-fold a	−11.7
FEO	CH2OPO32−	+ 1.2 × 106-fold a	−8.3
EO	CH2OH	−140-fold b	+2.9
EO + HPO32−	CH2OH	−1.3 × 106-fold c	+8.3
FEO	CH3	−130-fold b	+2.9
FEO + HPO32−	CH3	−3000-fold c	+4.7

^aCalculated from the values of k_cat/K_m for the decarboxylation of OMP (1.1 × 10⁷ M⁻¹ s⁻¹, Ref. 7) and FOMP (1.2 × 10⁷ M⁻¹ s⁻¹, Ref. 27) at pH 7.1 and I = 0.105 (NaCl) and the values of (k_cat/K_m)E in Table 1 for the decarboxylation of EO and FEO at pH 7.0 and I = 0.14 (NaCl).

^bCalculated from the values of (k_cat/K_m)E in Table 1 for the unactivated decarboxylation reactions of EO, orotidine, FEO, and 5′-dFO at pH 7.0 or pH 7.1 (for orotidine) and I = 0.14 (NaCl).

^cCalculated from the values of (k_cat/K_m)_E·Hpi/K_d in Table 1 for the phosphite-activated decarboxylation reactions of EO, FEO, and 5′-dFO at pH 7.0 and I = 0.14 (NaCl) and the reaction of orotidine at pH 7.1 and I = 0.21 (NaCl).

The large 4 × 10⁸-fold difference in the second-order rate constants for OMPDC-catalyzed decarboxylation of the natural substrate OMP (k_cat/K_m = 1.1 × 10⁷ M⁻¹ s⁻¹) ⁷ and the truncated substrate EO [(k_cat/K_m)_E = 0.026 M⁻¹ s⁻¹]¹³ shows that the 4′-CH₂OPO₃²⁻ substituent stabilizes the transition state for decarboxylation by 11.7 kcal/mol.^7,11,13 However, this large intrinsic phosphate binding energy³⁶ is not fully expressed for the decarboxylation of FOMP. Here the encounter-limited decarboxylation of FOMP (see above) results in only a 1 × 10⁶-fold difference in the second-order rate constants for the OMPDC-catalyzed decarboxylation reactions of FOMP and FEO. The observed transition state stabilization from the 4′-CH₂OPO₃²⁻ group of FOMP is only 8.3 kcal/mol, which suggests that the barrier to the decarboxylation of FOMP is ca. 3.4 kcal/mol lower than that expected if the chemical step were rate-determining. This is consistent with a rate constant ratio of k_chem/k_−d ≈ 300 for partitioning of enzyme-bound FOMP between decarboxylation and release to solution (Scheme 5). The dependence of k_cat/K_m for the decarboxylation of OMP on solvent viscosity is consistent with a partitioning ratio of k_chem/k_-d ≈ 0.6 for the reaction of OMP,¹⁸ which is similar to the reported commitment to catalysis of k_cat/k_-d = 0.3.⁸ If there is little or no effect of a 5-F substituent on the rate constant k_-d for release of the enzyme-bound nucleotide to solution, these observations suggest that there is a ca. 500-fold effect of a 5-F substituent on k_chem for decarboxylation of the enzyme-bound nucleotide (Scheme 5). This is very similar to the 400-fold effect of a 5-F substituent on decarboxylation determined directly from the unactivated OMPDC-catalyzed decarboxylation reactions of FEO and EO, for which the chemical step is cleaning rate-limiting (Table 2).

We expected that the phosphodianion binding pocket of OMPDC would easily accommodate the 4′-CH₃ group of 5′-dFO at the transition state for enzyme-catalyzed decarboxylation, because the enzyme has evolved to accommodate the presumably even larger 4′-CH₂OPO₃²⁻ group of OMP. We were therefore very surprised to find that the addition of a 4′-CH₃ group to FEO to give 5′-dFO results in a substantial 130-fold decrease in the rate constant for the unactivated decarboxylation and a 2.9 kcal/mol destabilization of the transition state (Table 3). There is an even larger 3000-fold effect (4.7 kcal/mol transition state destabilization) of the addition of a 4′-CH₃ group to FEO to give 5′-dFO on the third-order rate for the phosphite-activated reactions of these substrates. Therefore, not surprisingly, the steric effect of the 4′-CH₃ group on decarboxylation at the ternary E·5′-dFO·HPO₃²⁻ complex is larger than that at the binary E·5′-dFO complex.

Interestingly, the 4′-CH₂OH group of orotidine results in the same 2.9 kcal/mol destabilization of the transition state for decarboxylation observed for the 4′-CH₃ group of 5′- dFO, and there are similar stabilizations of the transition states for these reactions by the binding of phosphite dianion of 2.3 and 3.2 kcal/mol, respectively (Table 1). This suggests that there is little contribution of the terminal 5′-OH group of orotidine to the stability of the transition state for decarboxylation. By contrast, the 4′-CH₂OH group of orotidine results in a very large 8.3 kcal/mol destabilization of the transition state for phosphite-activated decarboxylation, exceeding the destabilization observed for the 4′-CH₃ group of 5′-dFO by 3.6 kcal/mol (Table 3). The destabilization of the transition state for the latter is masked because the reactivity of the parent two-part substrate FEO + phosphite is attenuated by the rate-limiting conformational change (see above).

These surprisingly large 4′-CH₃ and 4′-CH₂OH substituent effects on the decarboxylation reactions of FEO and EO are consistent with destabilization of the transition state for decarboxylation by steric effects of these groups that disrupt the required precise positioning of the peptide backbone and amino acid side chains at the reactive closed form of OMPDC. This is consistent with our earlier proposal that the binding energy of the substrate phosphodianion group is utilized to stabilize a rare active closed form of OMPDC that is required for the decarboxylation of both whole and truncated substrates.^13,17,25,35

Abbreviations

OMPDC

orotidine 5′-monophosphate decarboxylase

OMP

orotidine 5′-monophosphate

UMP

uridine 5′-monophosphate

FOMP

5-fluoroorotidine 5′-monophosphate

FUMP

5-fluorouridine 5′-monophosphate

EO

1-(β-D-erythrofuranosyl)orotic acid

EU

1-(β-D-erythrofuranosyl)uracil

FEO

1-(β-D-erythrofuranosyl)-5-fluoroorotic acid

FEU

1-(β-D-erythrofuranosyl)-5-fluorouracil

5′-dFO

5′-deoxy-5-fluoroorotidine

5′-dFU

5′-deoxy-5-fluorouridine

LDA

lithium diisopropylamide

BuLi

butyllithium

THF

tetrahydrofuran

MOPS

3-(N-morpholino)propanesulfonic acid

NMR

nuclear magnetic resonance