Stereo-specific Synthesis of Analogs of Nerve Agents And Their Utilization For Selection And Characterization of Paraoxonase (PON1) Catalytic Scavengers

Abstract

Fluorogenic organophosphate inhibitors of acetylcholinesterase (AChE) homologous in structure to nerve agents provide useful probes for high throughput screening of mammalian paraoxonase (PON1) libraries generated by directed evolution of an engineered PON1 variant with wild-type like specificity (rePON1). Wt PON1 and rePON1 hydrolyze preferentially the less-toxic R_P enantiomers of nerve agents and of their fluorogenic surrogates containing the fluorescent leaving group, 3-cyano-7-hydroxy-4-methylcoumarin (CHMC). To increase the sensitivity and reliability of the screening protocol so as to directly select rePON1 clones displaying stereo-preference towards the toxic S_P enantiomer, and to determine accurately K_m and k_cat values for the individual isomers, two approaches were used to obtain the corresponding S_P and R_P isomers: (a) stereo-specific synthesis of the O-ethyl, O-n-propyl, and O-i-propyl analogs; (b) enzymic resolution of a racemic mixture of O-cyclohexyl methylphosphonylated CHMC. The configurational assignments of the S_P and R_P isomers, as well as their optical purity, were established by X-ray diffraction, reaction with sodium fluoride, hydrolysis by selected rePON1 variants, and inhibition of AChE. The S_P configuration of the tested surrogates was established for the enantiomer with the more potent anti-AChE activity, with S_P/R_P inhibition ratios of 10–100, whereas the R_P isomers of the O-ethyl and O-n-propyl were hydrolyzed by wt rePON1 about 600- and 70-fold faster, respectively, than the S_P counterpart. Wt rePON1-induced R_P/S_P hydrolysis ratios for the O-cyclohexyl and O-i-propyl analogs are estimated to be ≫1000. The various S_P enantiomers of O-alkyl-methylphosphonyl esters of CHMC provide suitable ligands for screening rePON1 libraries, and can expedite identification of variants with enhanced catalytic proficiency towards the toxic nerve agents.

1. Introduction

Organophosphate ester hydrolases (OPH) displaying multiple turnover, such as bacterial phosphotriesterases (PTE) and mammalian serum paraoxonases (PON1), have been demonstrated to have the capacity to serve as potential prophylactic drugs to counteract in vivo intoxication by nerve agents and organophosphate-based pesticides [1–5]. A significant obstacle in applying OPHs as antidotes against nerve agent intoxication is their preference for hydrolysis of the less toxic isomers of nerve agents with the R_P absolute configuration at the phosphorous center [6–9]. All nerve agents are racemic mixtures due to the chiral phosphorous atom, with the more potent anti-AChE agents, i.e., the toxic isomers, being assigned the absolute configuration S around the P atom (S_P) [10]. It should be pointed out, however, that the assignment of the S_P configuration to the toxic isomers of sarin (GB), soman (GD), cyclosarin (GF) and VX (all liquids) has been suggested on the basis of deductive stereo-chemistry, and only indirect structural evidence exists to support this contention [11].

The combined application of computational design and directed evolution of mammalian PON1 has the potential to evolve a variant with reversed stereo-preference and kinetic properties that will qualify it to serve as a realistic candidate for prophylactic of nerve agent intoxications. The need to screen huge libraries of mutants, and to rapidly identify and select variants evolved to hydrolyze the toxic isomers, as well as the call for accurate biochemical characterization of k_cat and K_m of both the S_P and R_P enantiomers, necessitate the synthesis of the individual fluorogenic optical isomers. The synthesis and biochemical properties of racemic O-alkyl methylphosphonates containing coumarin analogs as the fluorogenic leaving group have already been described [12–14]. In order to qualify as catalytic scavenger drug candidates, rePON1 variants should be capable of hydrolyzing the toxic OP isomers at k_cat/K_m = 5x107 M⁻¹min⁻¹ (unpublished calculations). To achieve the desired specificity and efficacy towards the toxic isomers when reacting with racemic nerve agents, it was, therefore, necessary to develop stereo-specific methods for the synthesis of appropriate S_P and R_P fluorescent surrogates for screening of libraries, for improved selection, and for accurate biochemical characterization of promising catalytic variants.

In this report we describe the synthesis, purity, structural determination and biochemical properties of optical isomers of nerve agents in which the leaving groups fluoride (G agents) and N,N-dialkylaminoethanthiolo (V agents) were replaced by the fluorescent moiety, 3-cyano-7-hydroxy-4-methylcoumarin (CHMC) (Fig. 1). The S_P enantiomers were utilized to screen libraries by fluorescence–activated cell sorting and by direct monitoring of the release of the CHMC leaving group by crude lysates in 96-well plates.

Fig.1.

Stereo-specific synthesis of R_P- and S_P-O-alkyl methylphosphonyl-CHMC esters from methylphosphonyl dichloride [CH₃P(O)Cl₂], L-proline methyl ester, and 3-cyano-7-hydroxy-4-methylcoumarin (CHMC).The indicated absolute configuration was determined by X-ray diffraction of the corresponding crystals (Tables1&2) and by enzymatic hydrolysis (Fig. 3). Alcoholysis was catalyzed by 1 M H₂SO₄.

2. Materials and Methods

2.1 Chemicals and enzymes

3-cyano-7-hydroxy-4-methylcoumarine (CHMC) and L-proline methyl ester hydrochloride were purchased from Aldrich (USA) and used as obtained. Methylphosphonyl dichloride (CH₃POCl₂) was prepared according to Moedritzer and Miller [15]. Recombinant human AChE (hAChE) was purchased from Sigma (USA) and Torpedo californica AChE (TcAChE) was purified as previously described [16]. Wt rePON1 is variant G3C9 of the gene shuffling products of 4 mammalian PON1s [17], and its H115W mutant were produced and characterized by Khersonsky and Tawfik [18]. Variants 3B3 and 3D8 were generated by directed evolution [19] using variant G3C9 as the lead rePON1 [Devi -Gupta et al., unpublished].

2.2 Synthesis of fluorescent racemic O-alkyl methylphosphonylated CHMCs

The syntheses of racemic O-alkyl methylphosphonylated CHMCs (for structures see Fig. 1a&b) were carried out essentially by a modified protocol based on the general procedure described by Amitai et al. [13]. The modified protocol replaced chromatography by crystallization after the removal of free CHMC and of the O-alkyl methylphosphonic acid by extraction of the chloroform solution containing the crude product, with 2% Na ₂CO₃, pH 9.0, at 4^oC. The purity (>95%, containing <3% free CHMC) and homogeneity of the crystallized OP was established by tlc, ³¹P- and ¹H-nmr spectroscopy, and absorption at 400 nm, following hydrolysis with (a) rePON1 variants capable of hydrolyzing either R_P alone (3B3) or both R_P and S_P enantiomers (3D8), and by monitoring the release of total OP-bound CHMC in the presence of NaF at pH 8.0 (see below). Calculations were based on a value of 37,000 M⁻¹cm⁻¹ for the molar absorbance of CHMC at pH 8.0.

2.3 Reactions with NaF in dilute aqueous solution

(Note: the reader should be aware that this substitution reaction may contain non-hazardous low levels of G-type agent residues in dilute aqueous solution). The purity of the surrogate OPs was chemically determined as follows: 1 ml of 5–10 μM O-alkyl methylphosphonylated CHMC in 50 mM phosphate, pH 8.0, was reacted with 0.05–0.1 M NaF, and the release of CHMC was monitored until no further change was observed in the absorption at 400 nm.

2.4 Preparation of S_P-O-cyclohexyl O-(3-cyano-4-methyl-7-coumarinyl) methylphopsphonate(S_P-CMP-CHMC; Fig. 1b)

The S_P isomer was prepared by partial enzymatic degradation of racemic CMP-CHMC, using a different protocol from that earlier described by Amitai et al. [13]. Hundred mg of racemic CMP-CHMC in 15 ml methanol were added drop-wise to 200 ml of 50 mM Tris-1 mM CaCl₂, pH 8.0, and the clear solution was spiked with concentrated rePON1 variant 3B3 to produce a final concentration of 0.3 μM protein. Variant 3B3 is expected to hydrolyze only the R_P isomer, and the progress of the reaction was monitored at room temperature (RT) by following the release of CHMC at 400 nm. After 30 min, when no further change in OD was observed, and the amount of CHMC approached half the theoretical value, 40 g NaCl were added, the aqueous solution was extracted with 3x75 ml CHCl₃, and the combined organic layers were washed with 3x75 ml 2% Na₂CO₃, pH 9.0, pre-cooled to 4^oC. The CHCl₃ solution was dried over MgSO₄, filtered and dried under vacuum.

The crude product was crystallized from ethyl acetate-hexane to give a 58% yield of S_P-CMP-CHMC. ³¹P-nmr (124.1 MHz, CDCl₃), δ 27.98 (9 lines). ¹H-nmr (300 MHz, CDCl₃) δ 7.72 (d, J=6.5 Hz, 1H, 7.35 (dq J=7.5, 0.8 Hz, 1H), 7.26 (m, 1H), 4.60 (m, 1H), 2.80 (s, 3H), 1.95 (m, 2H), 1.85 (m, 2H), 1.70 (d, J=17 Hz, CH₃-P), 1.6-1.3 (m, 7H).

2.5 Synthesis and resolution of diastereoisomers of (S_P/R_P)-N-(S_C)-(proline methylester)- O-(3-cyano-4-methyl-oxo-2H-coumarin -7-yl ) methylphosphonate (I-S_PS_C and I-R_PS_C, see Fig.1a)

A solution of L-proline methyl ester hydrochloride ( 3.3. g, 20 mmol) and triethylamine (4.4 g, 40 mmol) in 60 ml CHCl₃ was added drop-wise to a stirred pre-cooled (−10^oC) solution of methylphosphonyl dichloride (2.66 g, 20 mmol) in 80 ml CHCl₃. The progress of the reaction was monitored by ³¹P-nmr spectroscopy, and the reaction was completed within 45 min, as apparent from the full conversion of methylphosphonyl dichloride to the corresponding methyl ester of L-proline monochloridoamidate.

To the forgoing preparation, a slurry of CHMC (4.02 g, 20 mmol) and triethylamine (2.5 g, 25 mmol) in 80 ml CHCl₃ was added, and the mixture was stirred at RT overnight. The resulting clear and homogenous solution was washed twice with 100 ml cold water (4^oC), followed by washing twice with 50 ml cold 2% Na₂CO₃ solution. It was then dried over Na₂SO₄, and the CHCl₃ removed under vacuum to give a yellow-green semi-solid residue which, upon trituratiion with ethyl acetate, turned into a crystalline suspension of a ~1:1 crude mixture of I-S_PS_C and I-R_PS_C. Yield, 6.0 g, 80%. ³¹P-nmr (124.1 MHz, CDCl₃), showed as expected, two distinct signals indicating the presence of both the I-R_PS_C and I-S_PS_C diastereoisomers. Further, ¹H-nmr (300 MHz, CDCl₃) indicated 2 sets of signals for the CH₃ of the methyl proline ester, and two CH₃-assigned doublets of the CH₃-P moiety. The CH₃ at the 4 position of the coumarin ring was found to overlap for the two diastereoisomers to give a single peak.

The separation of the two diastereoisomers was achieved by repeated fractional crystallization using boiling benzene in which I-R_PS_C displayed greater solubility than I-S_PS_C. The soluble fractions of I-R_PS_C from the cumulative benzene mother liquors were recrystallized from ethyl acetate: hexane, m.p. 143-5^oC, and the collected precipitates of I-S_PS_C recrystalized from benzene, m.p 181-3^oC. The absolute configuration was determined by X-ray crystallography (Table 1). TLC (silica, 5% methanol/95% ethyl acetate) confirmed the homogeneity of the two diastereoisomers. The ³¹P-and ¹H-nmr characteristics of the purified diastereoisomers are: I-R_PS_C: ³¹P (124.1 MHz, CDCl₃), δ 35.7 ppm (¹H: ³¹P coupled, 4 lines, q). ¹H (300 MHz, CDCl₃), δ 7.70 (d, J=7.5 Hz, 1H), 7.40 (dd, J=7.5, 0.5 Hz, 1H), 7.25 (bs, 1H), 4.40 (m,1H), 3.75 (s,3H), 3.40 (m, 1H), 3.20 (m, 1H), 2.75 (s,3H), 2.1-1.8 (m, 4H), 1.85 (d, J= 18Hz, 3H, CH₃-P). [α]_D²⁰ =−76.5^o (c=2, CHCl₃). I-S_PS_C: ³¹P (124.1 MHz, CDCl₃), δ 34.1 ppm (¹H: ³¹P coupled, 4 lines, q). ¹H (300 MHz, CDCl₃), δ 7.70 (d, J=7.0 Hz, 1H), 7.45 (d, J=7.0 Hz, 1H), 7.32 (bs, 1H), 4.30 (m,1H), 3.65 (s,3H), 3.35 (m, 1H), 2.75 (s, 3H), 2.1 (m, 2H), 1.9 (m, 2H), 1.70 (d, J= 18Hz, 3H, CH₃-P).. [α]_D²⁰ =−47.5^o (c=2, CHCl₃).

Table 1.

Crystallographic data for I-S_PS_C, I-R_PS_C, S_P-EMP-CHMC and R_P –EMP-CHMC

	I-SPSC	I-RPSC	SP-EMP-CHMC	RP –EMP-CHMC
Formula	C18H19N2O6P	C18H19N2O6P+0.5(C6H6)	C14H14NO5P	C14H14NO5P
Crystal description	Colorless plate	Colorless needle	Colorless plate	Colorless plate
Crystal size, mm3	0.15×0.08×0.03	0.40×0.15×0.15	0.50×0.40×0.10	0.57×0.13×0.04
FW, gr mol−1	390.32	429.38	307.23	307.23
Space group	P21	P21212	P21	P21
Crystal system	Monoclinic	Orthorhombic	Monoclinic	Monoclinic
a, Å	10.8945(3)	12.5780(4)	7.3072(4)	7.3015(3)
b, Å	7.4485(2)	24.3800(7)	25.2207(13)	25.2282(12)
c, Å	11.3426(4)	6.7055(2)	7.7947(4)	7.7996(4)
α, °
β, °	96.594(1)		99.079(1)	99.033(3)
γ, °
Cell volume, Å3	914.34(5)	2056.25(11)	1417.38(13)	1418.90(12)
Z	2	4	4	4
Density(calc, gr cm−3)	1.418	1.387	1.440	1.438
μ, mm−1	0.189	0.175	0.215	0.215
No. of reflections	26282	15120	20660	12791
No. of unique reflections	4183	4236	9264	6375
2Θ max	54.90	52.92	64.42	55.78
Rint	0.054	0.041	0.027	0.032
No. of parameters (restraints)	257(0)	274(0)	385(1)	385(1)
Final R for data with I > 2σ(I)	0.0454	0.0395	0.0475	0.0382
wR for data with I > 2σ(I)	0.1142	0.0855	0.1184	0.0967
Goodness of fit	1.091	1.037	1.035	1.014

2.6 Conversion of the diastereoisomers to the corresponding S_P and R_P O-alkyl methylphosphonate esters of CHMC (Fig. 1a)

Two hundred mg of either I-R_PS_C or I-S_PS_C were added to 10 ml of 1 M H₂SO₄ in either ethanol, n-propanol or i-propanol, and the mixture stirred at RT for 24 h to complete the reaction, as evidenced from the disappearance of the starting OP amidate ³¹P-nmr signal and the appearance of the new signal of the product. The alcoholysis medium was then vortexed with 100 ml CHCl₃, and the organic layer was washed first with cold water (2x50 ml), and then with cold 2% Na₂CO₃ (20 ml). The organic solution was dried over Na₂SO₄, the solvent evaporated, and the residual oily material triturated with ethyl acetate to give 30–40 mg of solid material. The two optical isomers of the O-ethyl analog (Fig. 1a, S_P-EMP-CHMC, and R_P-EMP-CHMC) were recrystalized from ethyl acetate and displayed the same nmr characteristics: ³¹P (124.1 MHz, CDCl₃), δ 29.0 (¹H^{: 31P} coupled; 9 lines). ¹H (300 MHz, CDCl₃), δ 7.72 (d, J=7.0 Hz, 1H), 7.35(dq, J=8.0, 0.5 Hz, 1H), 7.26 (bs, 1H), 4.24 (m, 2H), 2.75 (s, 3H), 1. 72 (d, J=17.5 Hz, 3H, CH₃-P), 1.35 (t, J= 7.0 Hz, 3H). It should be noted that the limited amounts of purified enantiomers of EMP-CHMC and the very low reading on the polarimeter precluded accurate determination of [α] _D. However, it is clear that the individual enantiomers displayed negative and positive rotations.

The R_P and S_P enantiomers of the O-n-propyl methylphosphonate ester of CHMC (Fig. 1a, S_P-nPrMP-CHMC, and R_P-nPrMP-CHMC) were obtained as viscous oils, and could not be crystallized. The ³¹P- and ¹H-nmr spectra of the two antipodes were identical: ³¹P (124.1 MHz, CDCl₃), δ 31.8 (¹H:³¹P coupled; 9 lines). ¹H (300 MHz, CDCl₃), δ 7.70 (d, J=7.5 Hz, 1H), 7.40 (dd., J=7.5, 0.5 Hz, 1H), 7.20 (bs, 1H), 4.15 (m, 2H), 2.75 (s, 3H), 1.75 (d, J=18 Hz, CH₃-P), 1.70 (m, 2H), 0.95 (t, J=7.0 Hz). [α]_D²⁰, S_P-nPrMP-CHMC = +8.6^o (c=3.7, CHCl₃)., R_P- nPrMP-CHMC = −8.4^o (c=2.5, CHCl₃).

In the case of the O-isopropyl analog, due to lack of sufficient material only the I-R_PS_C was available to react with isopropanol, and the absolute configuration determined by X-ray crystallography verified, as predicted, the R_P-iPrMP-CHMC antipode. The ³¹P- and ¹H-nmr spectra of R_P-iPrMP-CHMC were found identical to that of racemic iPrMP-CHMC which was prepared by the same protocol as described for racemic surrogates (see 2.2). ³¹P (121.4 MHz, CDCl₃) δ 28.0 (¹H:³¹P coupled: 8 lines, dq). 1H (300 MHz, CDCl₃) δ 7.70 (d, J=7.0 Hz, 1H), 7.35 (d, J= 6.5 Hz, 1H), 7.26 (bs,1H), 4.80 (m, 1H), 2.70 (s, 3H), 1.65 (d, J=16 Hz, CH₃-P), 1.40 [d, J= 6 Hz, 3H, one methyl of -CH(CH₃)₂], 1.30 [d, J= 6.0 Hz, 3H, the 2nd methyl of -CH(CH₃)₂]. It should be noted that the two CH₃ groups of isopropyl group are magnetically non-equivalent due to the pro-chirality of the carbon atom, –CH (CH₃)₂

2.7 Determination of K_m and k_cat. values for rePON1 variants

The enzymatic hydrolyses of the OP surrogates of nerve agents by purified rePON1 mutants were followed by monitoring the initial velocity of the release of the CHMC leaving group at 400 nm (in 50 mM Tris-1mM CaCl₂, pH 8.0, 25^oC). Data analysis was processed in accordance with the double reciprocal plot of Lineweaver-Burk using GraphPad Prism, version 5.0a, and assuming classical Michaelis-Menten behavior. Background non-specific hydrolysis of the CHMC-containing substrates was subtracted. k_cat was calculated using a molar absorbance coefficient of 3.7x10⁴ M⁻¹cm⁻¹ at 400 nm (pH 8.0) and a molecular weight of 40 kDa for rePON1 variants.

2.8 Inhibition of acetylcholinesterase

0.2–1.0 nM of either hAChE or TcAChE in 50 mM phosphate, pH 8.0, were incubated at 25^oC with a ≥10-fold excess of OP, and the rate of loss of enzyme activity was monitored by 20–50-fold dilution of the inhibition mixture into Ellman’s reaction assay medium [20] containing 1 mM acetylthiocholine as substrate.

3.0 Results and Discussion

3.1 Stereo-specific synthesis via construction of pairs of diastereoisomers

The separated pair of diastereoisomers which was obtained by reacting CH₃POCl₂ in a one-pot synthesis with L-proline methyl ester, followed by reaction with CHMC (Fig. 1a), was found by X-ray diffraction to possess the same S_C configuration at the chiral proline carbon. The S_P and R_P configurations at the phosphorous atom were established, and the purified diastereoisomers designated as I-S_PS_C and I-R_PS_C, respectively (Table 1). X-ray diffraction showed that reaction with either primary or secondary alcohols under acidic conditions leads to the formation of the S_P enantiomer from I-S_PS_C, and of the R_P isomer from I-R_PS_C. Similar conclusions with respect to the stereochemistry of the products of ethanolysis of homologous diastereoisomers were reached by Leader and Casida [21], who used ¹H- and ³¹-P-nmr considerations to propose inversion of the configuration at the phosphorous center of the pesticide profenofos. The finding that the absolute configuration at the P atom of I-S_PS_C and I-R_PS_C was maintained in the EMP-CHMC S_P and R_P enantiomers is attributed to inversion at the phosphorus center that accompanied the displacement of the proline moiety and to the change in substituent priority that governs the assignment of the absolute configuration of atoms of a chiral molecule, due to the replacement of the P-N bond of the OP-bound proline by a P-O-alkyl moiety. Similarly, structural analysis by X-ray diffraction of the reaction product between I-R_PS_C and isopropanol resulted in R_P-iPrMP-CHMC, which is consistent with the contention that inversion occurred during the replacement of the proline by an O-alkyl group, presumably via an S_N2 substitution mechanism. No attempts were made to identify side products of the alcoholysis reaction.

When the same reaction was repeated with n-propanol, the transformation of the I-S_PS_C and I-R_PS_C to the S_P and R_P enantiomers resulted in a viscous oil and X-ray crystallography could not be used to detemine the absolute configuration. However, production of the assumed S_P and R_P isomers, based on analogy with the stereochemical course of the reaction with ethanol and isopropanol was confirmed by use of a rePON1 mutant (3B3) that hydrolyzes almost exclusively the R_P isomer, and of the 3D8 mutant that hydrolyzes both enantiomers. This is demonstrated in Fig. 2 that compares the stereo-selectivities of wt rePON1 and of 3B3 (Fig. 2A) with those of variants H115W and 3D8 (Fig. 2B) in hydrolyzing racemic EMP-CHMC. Fig.3 shows the optical purity of the various enantiomers. The OP-bound CHMC was released quantitavely (>95% of the calculated value of the stock solution) by NaF, in a concentration-dependent manner (Fig. 2A), while wt rePON1 and 3B3 released only half of the available OP-bound CHMC. Mutant H115W also displayed a marked preference towards the same isomer, presumably the R_P enantiomer of the racemic EMP-CHMC; however, it also displayed an experimentally detectable reactivity on the second isomer (Fig. 2B). Mutant 3D8 clearly showed enhanced reactivity towards the wt-resistant enantiomer and, together with 3B3, was utilized to determine: (a) the optical purity of the S_P and R_P enantiomers, and (b) to verify the configuration of the assumed S_P and R_p enantiomers of nPrMP-CHMC. Fig. 3 (panels A & B) shows the kinetic behavior of the X-ray-established S_P and R_p enantiomers of EMP-CHMC in the presence of 3B3 and 3D8. Each isomer contains about 5% of the contaminating enantiomer. Similar time-course patterns and levels of optical purity were recorded for the deduced S_P and R_p of nPrMP-CHMC (Fig 3C&D). Thus, results suggest that propanolysis of the two diastereoisomers was also accompanied by inversion. It is assumed that the minor presence of the counter isomer in each preparation stems from impurities in the parent diastereoisomers I-S_PS_C and I-R_PS_C, and is not due to racemization during the course of alcoholysis [21].

Fig.2.

Hydrolysis of 0.014 mM racemic EMP-CHMC. Panel A: □- □, 0.1 M NaF; ○-○, 0.05 M NaF; •-•, 0.03 μM wt rePON1; ▪- ▪, 0.012 μM 3B3. Panel B: ○-○, 0.05 M, NaF; ▪-▪, 0.048 μM 3D8; •-•, 0.035 μM H115W

Fig. 3.

Determination of the optical purity of the enantiomers of EMP-CHMC (Panels A & B) and of nPrMP-CHMC (Panels C & D) by use of mutants 3B3 (○-○) and 3D8 (▪-▪). The optical purities of S_P-CMP-CHMC and R_P-iPrMP-CHMC are illustrated in panels E and F, respectively. In all cases, 3D8 was added after termination of CHMC release in presence of 3B3. The latter hydrolyzes exclusively the R_P isomer, whereas 3D8 hydrolyzes both the S_P and R_P enantiomers (Fig. 2B)

The structural analysis, together with results of the enzymic hydrolysis data, provided direct evidence for the marked preference of wt rePON1 and the H115W variant for the three R_P isomers of the O-alkyl methylphosphonyl-CHMC analogs, regardless of the size of the alkyl group. Variant 3B3 displayed even greater stereo-specificity towards the R_P enantiomers (Devi-Gupta et al., unpublished). Thus, assuming that the CH₃, the P=O, and the CHMC groups of the tested stereoisomers are accommodated by the same binding regions at the active site, the various O-alkyl moieties of the R_P isomers are probably projected towards the same relatively wide pocket that does not impose steric constraints on the alkyl residue. This conclusion is further supported by the kinetic data for hydrolysis of the cyclohexyl analog (see below).

3.2. Isolation of S_P-CMP-CHMC by enzymic degradation of racemic CMP-CHMC

The free CHMC in S_P-CMP-CHMC did not exceed 2% of the bound CHMC, and the contamination of S_P-CMP-CHMC by the R_P isomer, as judged by the release of CHMC in presence of 3B3, was ~ 3% (Fig. 3E).

The absolute configuration of the O-cyclohexyl analog isolated after partial hydrolysis of racemic CMP-CHMC by mutant 3B3 was shown by X-ray crystallography to conform to the S_P configuration (Table 2). The S_P was found to be an extremely poor substrate of wt rePON1 and of the 3B3 mutant, whereas the 3D8 mutant released the theoretical amount of CHMC quite rapidly (not shown). These results are consistent with the stereo-preferences of the two mutants as revealed from their reaction with the individual enatiomers of EMP-CHMC, nPrMP-CHMC, and iPrMP-CHMC.

Table 2.

Crystallographic data for R_p isomer of iPrMP-CHMC and S_P enantiomer of CMP-CHMC

	Rp-iPrMP-CHMC	SP-CMP-CHMC
Formula	2(C15H16N1O5P) + O	C18H20N1O5P
Crystal description	Colorless needle	Colorless plate
Crystal size, mm3	0.80×0.30×0.10	0.50×0.40×0.10
FW, gr mol−1	658.52	361.32
Space group	C2	P21
Crystal system	Monoclinic	Monoclinic
a, Å	31.302(7)	6.5200(5)
b, Å	7.0663(14)	12.9097(10)
c, Å	28.358(6)	10.5795(9)
α, °
β, °	90.30(3)	95.090(3)
γ, °
Cell volume, Å3	6272(2)	886.98(12)
Z	8	2
Density(calc, gr cm−3)	1.395	1.353
μ, mm−1	0.202	0.183
No. of reflections	15673	24739
No. of unique	9361	6818
reflections
2Θ max	52.74	66.90
Rint	0.1339	0.028
No. of parameters (restraints)	846(73)	228(1)
Final R for data with I > 2σ(I)	0.0756	0.0327
wR for data with I > 2σ(I)	0.1855	0.0830
Goodness of fit	0.953	1.039

Hydrolysis induced by NaF showed that the S_P-CMP-bound CHMC corresponds to >95% of the calculated value. S_P-CMP-CHMC displayed high optical purity, with a low fluorescence background, which permitted meaningful screening of huge libraries for positive clones capable of hydrolyzing the toxic isomer of the OP surrogate both by use of a FACS protocol, and by monitoring the release of chromophore by lysates in a 96-well-plate format.

3.3 Inhibition of acetylcholinesterase by the S_P and R_P enantiomers

For all the S_P isomers acting on AChE, pseudo-first order inhibition rate constants (k_obs) were obtained by fitting the data points for residual activity to a mono-exponential decay function, and the bimolecular rate constants for inhibition (k_i) were determined from the slopes of the linear plots of k_obs vs. inhibitor concentration. In the case of the R_P isomers, that were contaminated with ~ 5% of the more potent anti-AChE, the S_P form, the experimental data were best fitted to a bi-exponential decay curve with the constraint of 5% and 95% amplitude distribution for the S_P and R_P isomers, respectively (not shown)..

The k_i values for the inhibition of hAChE and TcAChE by the various optical isomers are summarized in Table 3. The faster rates of inhibition by the S_P isomers of the O-ethyl and On-propyl analogs, compared to their R_P enantiomers, confirmed the previously deduced S_P configuration of the toxic isomers of nerve agents [10].

Table 3.

Bimolecular rate constants (k_i) for the inhibition of hAChE and TcAChE

Enantiomer	human AChE (ki, x107 M−1 min−1)a	Torpedo AChE (ki,x107 M−1 min−1)a
Sp-EMP-CHMC	2.80(SP/RP =23)	21.0(SP/RP =11)
RP-EMP-CHMCb	0.12	1.9
SP-nPrMP-CHMC	15.8(SP/RP =26)	77.0(SP/RP =97)
RP-nPrMP-CHMCb	0.61	0.79
SP-CMP-CHMC	8.2	3.7

^aS.D. < 20% (n=3–5)

^bFigures corrected for the contamination with 5% S_P isomer

The X-ray-determined absolute configuration of the O-alkyl methylphosphonates provides an important set of nerve agent surrogates that serve to unequivocally establish the stereo-preference of AChEs. The simultaneous binding of the CH₃-P, O-alkyl-P, and the P=O groups in the acyl pocket, the alkoxy binding site, and the oxyanion hole, respectively, [11], dictates the preferred accommodation of the S_P configuration that stabilizes both the ground-state complex between the approaching inhibitor and the chiral AChE (analogous to the Michaelis-Menten complex), and the transition state that facilitates the inhibition reaction. In all these cases the leaving group is oriented towards the opening of the active-site gorge, ready to be expelled by an in-line attack of the active-site serine [11]. Thus, for O-alkyl methylphosphonates with a wide variety of leaving groups such as fluoride (G agents), N, N-dialkylaminoiethanthiolo (V-agents), and the bulky leaving group of the CHMC surrogates, the S_P enantiomer is the more potent anti-AChE form.

TcAChE was found significantly more sensitive than hAChE to inhibition by the S_P enantiomers of the ethyl (EMP) and the n-propyl (nPrMP) esters of CHMC, and also by the R_P-EMP-CHMC isomer, but not by S_P-CMP-CHMC. These findings contribute new data to an accumulated body of evidence showing significant differences in the susceptibilities of the two AChEs to various inhibitors [22]. The kinetics of inhibition of hAChE and TcAChE by the O-i-propyl enantiomers will be reported elsewhere after optically pure S_P-iPrMP-CHMC is obtained.

3. 4 Enzymatic hydrolysis of the S_P and R_P enantiomers by rePON1 variants

The kinetic parameters for the reaction of wt rePON1 and mutant H115W with the enantiomers of EMP-CHMC and nPrMP-CHMC are summarized in Table 4. In the case of the S_P enantiomer that contains about 3–5% of the preferred substrate with the R_P configuration, the data points for the Lineweaver-Burke plots were taken after 5–10 min of pre-incubation so as to minimize a possible contribution by the R_P contaminant, and the analytical concentration of the remaining S_P substrate (~95%) was adjusted on the basis of the released chromophore monitored at 400 nm. Compared to the relatively inactive S_P isomers, the two R_P enantiomers were found to be very good substrates for both the wt and the H115W variants of rePON1, with k_cat values of 300–700 min⁻¹ and k_cat/K_m values of 3.7–5.6 x10⁷ M⁻¹min⁻¹. It should be noted that for the O-alkyl methylphosphonate series the stereo-specificity is not affected by the size of the O-alkyl group, nor by the size and electronic features of the leaving group, as can be judged from the fact that the same preferences are reported for the corresponding fluoridates [6-9]. Results are the first direct structural evidence, based on X-ray crystallography, that wt mammalian paraoxonases hydrolyze preferentially the R_P-CHMC surrogates of nerve agents. Similar conclusions were reported with respect to the reaction of bacterial PTE with the R_P isomers of p-nitrophenolate analogs of nerve agents [6].

Table 4.

k_cat (min⁻¹) and k_cat/K_m (x10⁶ M⁻¹min⁻¹) for the hydrolysis of the S_P and R_P enantiomers of EMP- and nPrMP-CHMC, and for S_P-CMP-CHMC. by rePON1 variants

	wt (G3C9)		H115W
	kcat	kcat/Km	kcat	kcat/Km
SP-EMP-CHMC	9	0.063	30	0.4
RP-EMP-CHMC	703	37	33	46
RP/SP		587		115
SP-nPrMP-CHMC	110	0.82	388	1.5
RP-nPrMP-CHMC	297	56	10	53
RP/SP		68		35
SP-CMP-CHMC	NDa	<0.002b	NDa	~0.005b

^aNot determined. K_m is estimated at > 1 mM

^bEstimates based on the assay limits of the chromophore absorption at 400 nm and the protein concentration in the reaction mixture.

Table 4 also demonstrates the appearance of a variant (mutant H115W) that is characterized by 2- to 6-fold enhanced k_cat/K_m values for hydrolysis of the S_P isomers of O-ethyl, O-n-propyl, and O-cyclohexyl analogs as compared to the wt enzyme. The fact that H115W displays accelerated hydrolysis of the S_P isomer, regardless of the size of the O-alkyl moiety, suggests that directed evolution can readily utilize lead variants, such as H115W, for development of mutants capable of effective hydrolysis of the toxic isomers of nerve agents. The use of the S_P isomers to construct a repertoire of mutants with k_cat/K_m>1x10⁷ M⁻¹min⁻¹ when reacting both with nerve agents analogs and with the actual ‘live agents’ will be published elsewhere (Devi-Gupta et al., ms in preparation).

4. Conclusions

The individual stereoisomers of several O-alkyl methylphosphonylated coumarin analogs of nerve agents were prepared by a stereo-specific synthesis protocol, and by an enzymatic procedure that utilized the racemic mixture of the fluorogenic surrogates and a rePON1 mutant that hydrolyze preferentially the R_P isomer. The high enantiomeric purity, and the low level of free chromophore suggests that the S_P chiral analogs of nerve agents are suitable for screening of large rePON1 libraries for variants capable of hydrolyzing the toxic isomers of nerve agents. Of particular importance is the successful alcoholysis of the pair of diastereoisomers with a secondary alcohol, such as i-propanol, that opens the way to the synthesis of chiral O-pinacolyl and O-cyclohexyl analogs of soman and cyclosarin, respectively. The X-ray crystal structures, together with in vitro measurements of the rates of inhibition of AChE, and of the proficiency (k_cat/K_m) of rePON1 variants, confirmed that the S_P isomers as the more potent anti-AChE agents, and the corresponding R_P antipodes as the preferred substrates of wt rePON1. The moderate, yet detectable, enhancement of k_cat/K_m of H115W, relative to the wt rePON1, suggests that directed evolution of rePON1 lead mutants such as H115W, together with designed molecular engineering guided by the 3D structures of rePON1 variants, and k_cat and K_m values, is likely to evolve rePON1 variants with catalytic efficacy that will qualify them as potential drug candidate for prophylactic treatment of nerve agent intoxication.

Abbreviations

OPH

organophosphate hydrolase

PTE

phosphotriesterase

PON

paraoxonases

CHMC

3-cyano-7-hydroxy-4-methylcoumarin

hAChE

recombinant human AChE

TcAChE

Torpedo californica AChE

EMP

O-ethyl methylphosphonyl

nPrMP

O-n-propyl methylphosphonyl

iPrMP

O-i-propyl methylphosphonyl

CMP

O-cyclohexyl methylphosphonyl