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 (CH3
) was prepared according to Moedritzer and Miller [15
]. Recombinant human AChE (hAChE) was purchased from Sigma (USA) and Torpedo californica
AChE) 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] using variant G3C9 as the lead rePON1 [Devi -Gupta et al., unpublished].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 ) 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 2
, pH 9.0, at 4o
C. The purity (>95%, containing <3% free CHMC) and homogeneity of the crystallized OP was established by tlc, 31
P- and 1
H-nmr spectroscopy, and absorption at 400 nm, following hydrolysis with (a) rePON1 variants capable of hydrolyzing either RP
alone (3B3) or both RP
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−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 SP-O-cyclohexyl O-(3-cyano-4-methyl-7-coumarinyl) methylphopsphonate(SP-CMP-CHMC; )
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 RP
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% Na2
, pH 9.0, pre-cooled to 4o
C. 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 SP-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 (SP/RP)-N-(SC)-(proline methylester)- O-(3-cyano-4-methyl-oxo-2H-coumarin -7-yl ) methylphosphonate (I-SPSC and I-RPSC, see )
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 (−10oC) 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 (4oC), 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-SPSC and I-RPSC. Yield, 6.0 g, 80%. 31P-nmr (124.1 MHz, CDCl3), showed as expected, two distinct signals indicating the presence of both the I-RPSC and I-SPSC 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-RPSC displayed greater solubility than I-SPSC. The soluble fractions of I-RPSC from the cumulative benzene mother liquors were recrystallized from ethyl acetate: hexane, m.p. 143-5oC, and the collected precipitates of I-SPSC recrystalized from benzene, m.p 181-3oC. The absolute configuration was determined by X-ray crystallography (). 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-RPSC: 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 =−76.5o (c=2, CHCl3). I-SPSC: 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 =−47.5o (c=2, CHCl3).
Crystallographic data for I-SPSC, I-RPSC, SP-EMP-CHMC and RP –EMP-CHMC
2.6 Conversion of the diastereoisomers to the corresponding SP and RP O-alkyl methylphosphonate esters of CHMC ()
Two hundred mg of either I-RPSC or I-SPSC 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 (, SP-EMP-CHMC, and RP-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 RP and SP enantiomers of the O-n-propyl methylphosphonate ester of CHMC (, SP-nPrMP-CHMC, and RP-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, SP-nPrMP-CHMC = +8.6o (c=3.7, CHCl3)., RP- nPrMP-CHMC = −8.4o (c=2.5, CHCl3).
In the case of the O-isopropyl analog, due to lack of sufficient material only the I-RPSC was available to react with isopropanol, and the absolute configuration determined by X-ray crystallography verified, as predicted, the RP-iPrMP-CHMC antipode. The 31P- and 1H-nmr spectra of RP-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 Km and kcat. 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, 25oC). 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. kcat 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 Tc
AChE in 50 mM phosphate, pH 8.0, were incubated at 25o
C 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.