Differentiation between Acetylcholinesterase and the Organophosphate-inhibited Form Using Antibodies and the Correlation of Antibody Recognition with Reactivation Mechanism and Rate*

Two types of polyclonal antibodies were generated from ( a ) a decapeptide sequence that includes the active site serine of acetylcholinesterase (anti-AChE 10S ) and ( b ) the identical decapeptide sequence phosphorylated at the active site serine of acetylcholinesterase (anti-AChE 10SP ). The anti-AChE 10S antiserum was found to specifically recognize native, control, and vehicle-treated recombinant mouse AChE (rMoAChE) but did not recognize rMoAChE that was phosphorylated by the four organophosphate (OP) compounds tested. Conversely the anti-AChE 10SP antiserum recognized phos- phoserine rMoAChE that resulted from reaction with phosphorous oxychloride (POCl 3 ) but did not recognize native or vehicle-treated rMoAChE. Anti-AChE 10SP also did not recognize OP-AChE conjugates that resulted from the reaction of rMoAChE with other OP compounds that afford neutral or monoanionic phosphoserine groups thereby indicating a high specificity for a precise OP conjugate. Antisera recognition correlated well with the rates of enzyme inhibition, aging, and oxime-induced reactivation indicating these antisera can both quantify

Two types of polyclonal antibodies were generated from (a) a decapeptide sequence that includes the active site serine of acetylcholinesterase (anti-AChE 10S ) and (b) the identical decapeptide sequence phosphorylated at the active site serine of acetylcholinesterase (anti-AChE 10SP ). The anti-AChE 10S antiserum was found to specifically recognize native, control, and vehicletreated recombinant mouse AChE (rMoAChE) but did not recognize rMoAChE that was phosphorylated by the four organophosphate (OP) compounds tested. Conversely the anti-AChE 10SP antiserum recognized phosphoserine rMoAChE that resulted from reaction with phosphorous oxychloride (POCl 3 ) but did not recognize native or vehicle-treated rMoAChE. Anti-AChE 10SP also did not recognize OP-AChE conjugates that resulted from the reaction of rMoAChE with other OP compounds that afford neutral or monoanionic phosphoserine groups thereby indicating a high specificity for a precise OP conjugate. Antisera recognition correlated well with the rates of enzyme inhibition, aging, and oxime-induced reactivation indicating these antisera can both quantify the extent and type of inhibition and also differentiate between select mechanisms of inhibition. The ability to discern mechanistic differences between native AChE and OP-AChE conjugates suggests that these antisera can be used to identify biomarkers of OP exposure in a mechanism-based approach.
Organophosphate esters 1 (OPs) 1 are a diverse class of compounds ( Fig. 1 and Table I) with the most well known applications as insecticides (e.g. malathion and parathion), as protease inhibitors (e.g. diisopropyl fluorophosphate (DFP)), in glaucoma treatment (e.g. echothiophate), and as chemical warfare agents (e.g. sarin and VX). The acute toxic effects of OP compounds correlate well with their ability to inhibit acetylcholinesterase (AChE) by reaction with an essential serine hy-droxyl to form a relatively stable phosphoserine ester bond or an OP-AChE conjugate ( Fig. 1 and Table I, 2) (1, 2). AChE inactivation results in an accumulation of acetylcholine in cholinergic synapses and, depending on the OP structure and persistence of the OP-AChE conjugate, excessive stimulation of acetylcholine receptors. Excess acetylcholine leads to hyperstimulation of skeletal muscle endplates, smooth muscle, and secretory glands as well as altered central nervous system and cardiac activity, which can result in cardiovascular and respiratory compromise and, in extreme cases, death (3).
The biochemical mechanism of OP inactivation of AChE is a widely accepted process and initiated by precursory phosphorylation at the catalytic serine residue ( Fig. 1) (4). The phosphorylation of AChE by an OP is synchronous with the ejection of a leaving group (Z) to yield a stable, covalent phosphoserine ester bond ( Fig. 1 and Table I, 2).
The phosphorylated AChE can usually undergo two possible postinhibition fates: (a) reactivation, which is cleavage of the phosphoester-serine bond either spontaneously (water) or mediated by oxime antidotes or (b) "aging," in which a phosphoester bond other than the phosphoserine is cleaved to produce a phosphate oxyanion ( Fig. 1 and Table I, 3). An "aged" OP-AChE conjugate is considered irreversible inhibition and typically unable to regain enzymic activity. The inhibition and postinhibition rates and mechanisms are dependent on the structure and reactivity of the phosphorus ester ligands (4,5). For example, OP-AChE conjugates that contain branched dimethyl esters generally reactivate readily, whereas OP-AChE conjugates that contain larger or branched alkyl groups (e.g. isopropyl, isobutyl, etc.) are more prone to undergo an aging mechanism. Overall OP compounds react with AChE to afford structurally precise phosphoserine esters suggesting that these products can serve as selective indicators of mechanism and possibly a direct correlation with the original OP structure.
Human exposure to OPs is assessed clinically by measuring the depression in serum or erythrocyte cholinesterase activity utilizing a colorimetric assay (6). The colorimetric assay is considered rapid and capable of high throughput analysis, however, several limitations to this method exist: (a) control levels of cholinesterase activity must be obtained prior to analysis of OP exposure, (b) cholinesterase activity and exposure level are not directly correlated, and most importantly (c) this assay does not identify the structure of the OP-AChE conjugates. As indicated, different OP structures can form distinct OP-AChE conjugates that can yield different toxic outcomes, and therefore the need to determine the structure of the OP-AChE conjugate is essential to understand the reactivation and/or aging profile for a given OP (3,(7)(8)(9).
In this study, we sought to utilize the exquisite selectivity of antibody-antigen interactions to differentiate between AChE and AChE modified by reactive phosphorus compounds. These mechanism-based antibodies can then serve as probes of the phosphorylation/dephosphorylation process and possible indicators of enzyme reaction rate. And since the resultant OP conjugate typically correlates directly with the OP structure, an antibody-based analysis could improve upon the current, end point-based colorimetric method. To test the hypothesis that OP-AChE conjugates can be distinguished from native, uninhibited AChE, polyclonal antiserum (anti-AChE 10S ) was raised against a linear decapeptide corresponding to the native mouse and human AChE sequence (Fig. 1, 4a) that includes the catalytic serine residue. To represent OP-modified AChE and to serve as an antigen distinct from native protein, a phosphoserine (Ser-OPO 3 2Ϫ ) decapeptide ( Fig. 1, 4b) was used to generate a second polyclonal antiserum (anti-AChE 10SP ). The selection of decapeptide serine phosphate as an antigen is consistent with the proposed inhibition of AChE by phosphorus oxychloride (POCl 3 ) (10). The initial OP-AChE conjugate resulting from reaction of POCl 3 with AChE is a chlorophosphoryl oxyanion ( Fig. 1 and Table I, 2: X ϭ Cl, Y ϭ O Ϫ ) leading to a putative phosphate group ( Fig. 1 and Table I Anti-AChE 10SP was hypothesized to better react with aged AChE than did anti-AChE 10S , and overall, the two antisera should differentiate between native, inhibited, and aged AChE. A critical aspect of biomolecular recognition of the catalytic serine of AChE is its location in the protein. The catalytic serine is located at the base of a deep 18-Å gorge within the protein (11) and is largely inaccessible to many traditional forms of analysis unless the protein is denatured. The experiments described here involve reacting the intact, catalytically active enzyme with OP, measuring the activity of the enzyme, and then denaturing the enzyme to expose the active site to antibodies. This approach permits the study of an "inaccessible" catalytic peptide sequence as well as the corresponding covalent phosphorylation by OP that will enable the determination of mechanism and reaction progress.

EXPERIMENTAL PROCEDURES
Reagents-Recombinant mouse AChE (rMoAChE) was purified as described previously (12). Electric eel AChE (Type V-S) and horse butyrylcholinesterase were used as purchased (Sigma). Anti-mAChE antiserum is an affinity-purified polyclonal antiserum raised against monomeric rMoAChE enzyme. POCl 3 and DFP were purchased from Aldrich, and paraoxon was purchased from Chem Service Inc. (West Chester, PA). The S,S-stereoisomer of isomalathion was synthesized as described previously (13). Reagents for immunoaffinity purification were purchased from Bio-Rad.
Polyclonal Antisera Generation and Purification-Anti-AChE 10S and anti-AChE 10SP antisera were generated by conjugating peptides AChE 10S 4a and AChE 10SP 4b ( Fig. 1) to keyhole limpet hemocyanin. Rabbits were immunized using standard procedures (SynPep Corp., Dublin, CA). Antisera were equilibrated into TSSA buffer (20 mM Tris-HCl, pH 8.0, 25 mM NaCl, 0.02% NaN 3 ) with Econopac 10DG columns. Contaminants were removed by chromatography over a DEAE Affi-Gel Blue column. Peptides 4a and 4b were conjugated to Affi-Gel 15 beads in anhydrous isopropanol for 72 h at 4°C. Antisera were purified over the peptide-conjugated column by eluting less specific antisera with 1 M NaSCN. Peptide-specific antisera were eluted with 0.1 M glycine-HCl (pH 1.8) and immediately buffered to pH 8.0. All eluted fractions were analyzed by enzyme-linked immunosorbent assay for recognition of keyhole limpet hemocyanin, AChE 10S , and AChE 10SP to ensure high specific recognition (14).
Reaction of Organophosphates with rMoAChE-rMoAChE was diluted in 0.1 M phosphate buffer (pH 7.6) (final concentration, 110 nM) prior to OP addition. OP compounds were diluted in appropriate vehicle, and either vehicle or OP were added. Final concentrations of OP and their respective vehicles used in experiments were as follows: DFP (136 M in isopropanol), POCl 3 (3 mM in tetrahydrofuran), isomalathion (526 M in ethanol), and paraoxon (750 M in acetone). All vehicles were added to Ͻ5% final volume and had no significant effect on rMoAChE activity or reactivity with anti-mAChE antiserum (data not shown).
Western Immunoblot Assays-Western immunoblot assays were performed as described (15) with the following minor modifications. Primary antibodies were diluted 1:100 in blocking buffer (2% bovine serum albumin, phosphate-buffered saline, 0.1% Tween 20). Secondary antibodies were diluted 1:10,000 in blocking buffer. Densitometric measurements were performed using a Bio-Rad GS800 densitometer and Quantity One software.
benzoic acid) (final concentration, 0.32 mM) and initiated with acetylthiocholine iodide (final concentration, 0.75 mM). Changes in absorbance were measured at A 412 nm in 10-s intervals over a 3-min period using a Softmax Pro microplate reader.
Competition Experiment with DFP and POCl 3 to Test Specificity of Anti-AChE 10SP -rMoAChE was treated with DFP or POCl 3 at final concentrations of 500 M and 2.7 mM, respectively, and allowed to incubate for 5 days at 4°C when AChE activity was measured. Aliquots of samples were treated with pralidoxime (2-PAM) (final concentration, 1.4 mM) for 15 min to reactivate any available rMoAChE and to determine the percentage of aging. Aliquots of POCl 3 -treated and DFPtreated rMoAChE were subjected to a second exposure of the other OP or vehicle overnight at 4°C. These samples were analyzed in duplicate by immunoblot and probed with anti-AChE 10SP .
Reactivation Assays-rMoAChE was inhibited with 5 M paraoxon for 12 min, and the AChE activity was measured. Inhibited rMoAChE was aliquoted into different final concentrations of 2-PAM, and the percent reactivation was measured (in triplicate) for various time points (17,18). Immediately after reactivation was measured, the reaction was stopped by addition of Laemmli SDS-PAGE dye, and the samples were boiled. Immunoblots were performed in quadruplicate, bands were quantitated on a densitometer, and the data points shown represent the average of quadruplicate immunoblots with standard deviations. After detection with anti-AChE 10S , the membrane was stripped with 10 mM glycine-HCl (pH 3) and reprobed with anti-mAChE antiserum to confirm that equivalent amounts of protein were loaded. Control samples were treated with acetone (vehicle for paraoxon) followed by 2-PAM at the highest concentration. Percent reactivation was measured as 100 Ϫ (( , where E C is the control enzyme (no OP), E I is the inhibited enzyme (fully inhibited, no 2-PAM added), and S is the sample being measured.
Statistics-t test (paired) analyses were performed on data in Fig. 4 using Microsoft Excel software. Regression parameters, including the slope, 95% confidence interval, and correlation coefficient (R 2 ) were determined for Fig. 5 using SigmaPlot Version 4.01 software.

RESULTS
Anti-AChE 10S Recognition of Cholinesterases-Anti-AChE 10S was tested for reactivity with cholinesterases from several different sources. The decapeptide used as the antigen to generate anti-AChE 10S is 100% homologous with rMoAChE and horse butyrylcholinesterase (Fig. 1, 4a) and differs from electric eel AChE by one amino acid (Leu 3 Ile). Anti-AChE 10S recognized rMoAChE, electric eel AChE, and horse butyrylcholinesterase ( Fig. 2A), and the molecular mass of the band in the rMoAChE sample was ϳ60 kDa, which corresponds with the correct molecular mass due to truncation at the carboxyl terminus of the cDNA at amino acid 548 (12). Anti-AChE 10S detected two bands in the electric eel AChE sample. One band at ϳ65 kDa corresponds to the correct mass for intact protein (19); however, the major band recognized by anti-AChE 10S migrates at ϳ43 kDa. This has been observed by our laboratory in other antibody analyses of eel AChE (15) and presumably corresponds to partially degraded enzyme. The band detected in the horse butyrylcholinesterase samples migrates at ϳ65 kDa, which corresponds to the predicted size (20).
Anti-AChE 10S Antiserum Recognition of Native, Uninhibited rMoAChE and Anti-AChE 10SP Recognition of POCl 3 -treated rMoAChE-To characterize the specificity of the antisera, rMoAChE was first treated with POCl 3 to reduce the AChE activity to less than 5% of control activity (Table II). The inhibited rMoAChE samples were then analyzed by immunoblot. Anti-AChE 10S clearly recognized native and vehicletreated rMoAChE but only weakly recognized the inactive, POCl 3 -treated rMoAChE (Fig. 2B, top panel), the latter recognition likely due to partially reversible or incomplete interaction of POCl 3 with rMoAChE. This result demonstrated that anti-AChE 10S recognized the native active site and that phosphorylation of the active site serine is sufficient to disrupt anti-AChE 10S recognition. Anti-AChE 10SP recognized only POCl 3 -treated rMoAChE (Fig. 2B, bottom panel), which indicates that the phosphorylation of serine is required for anti-AChE 10SP recognition.

FIG. 2. Demonstration of cholinesterase recognition by anti-AChE 10S
and differential antibody recognition of OP-AChE conjugates by immunoblot analysis. A, cholinesterases from different sources were analyzed for recognition by anti-AChE 10S . The molecular mass of rMoAChE is indicated. B, rMoAChE was either not treated (nt) or reacted with tetrahydrofuran (THF, vehicle) or POCl 3 and detected with anti-AChE 10S and anti-AChE 10SP . C, rMoAChE was reacted with OP (as indicated above immunoblot) to generate OP-AChE conjugates. Samples were analyzed for recognition by anti-AChE 10S , anti-AChE 10SP , and anti-mAChE. Density of bands was quantitated and is shown below the immunoblots in B and C. Density is expressed as the adjusted density (OD/mm 2 ) after local background subtraction. PX, paraoxon; BuChE, butyrylcholinesterase.
Anti-AChE 10SP Recognition Is Specific for POCl 3 -treated rMoAChE-To further characterize the specificity of anti-AChE 10SP , OP-AChE conjugates were generated from the reaction of rMoAChE with three additional OP compounds, and the recognition was tested. rMoAChE was reacted with isomalathion, paraoxon, or DFP, and rMoAChE activity was determined to be less than 5% of control (Table II). Inhibited rMoAChE samples were analyzed for antibody recognition by immunoblot analysis. Anti-AChE 10S again recognized the vehicle-treated rMoAChE but did not recognize any of the OP-AChE conjugate samples (Fig. 2C, top panel), although it slightly recognized rMoAChE treated with POCl 3 (as indicated previously). These results demonstrate that regardless of the OP ester ligands, the covalently OP-modified active sites of rMoAChE disrupted anti-AChE 10S recognition. Anti-AChE 10SP recognized the rMoAChE treated with POCl 3 (Fig. 2C, middle panel) but not native rMoAChE or rMoAChE inhibited by the other OPs. As a positive control to determine the relative protein amount and integrity, immunoblots were probed with a polyclonal anti-mAChE antiserum (Fig. 2C, bottom panel). The anti-mAChE antiserum showed equivalent recognition of all samples and also supports the conclusion that OP treatment did not affect the integrity of the enzyme.
Anti-AChE 10SP Recognition Is Specific for the Phosphorylated (Dianionic) Active Site of rMoAChE-Because POCl 3 can act as a non-selective phosphorylating agent and potentially react with AChE at multiple residues (21), the possibility existed that anti-AChE 10SP recognized phosphorylated residues other than the catalytic serine. To address this, a competition experiment was conducted in which rMoAChE was reacted first with DFP, which phosphorylates only the catalytic serine (22,23). Next POCl 3 was added to the DFP-inhibited rMoAChE to phosphorylate any other reactive residues. If anti-AChE 10SP is capable of recognizing phosphoserine residues other than the active site, recognition of this sample preparation should occur. Conversely rMoAChE was reacted stepwise with POCl 3 and then DFP to form the expected phosphoserine group at the active site. For DFP to fully protect the active site serine from POCl 3 , it was necessary to allow the diisopropyl phosphorylated rMoAChE, the DFP-AChE conjugate, to age (4, 5). Therefore, the length of time of the pretreatment was increased to allow a maximum amount of DFP-AChE conjugates to age as determined colorimetrically. After the initial exposure, DFP and POCl 3 inhibited rMoAChE activity to less than 5% of either control (data not shown). After reactivation of the remaining DFP-AChE population with 2-PAM, the DFP-treated rMoAChE reactivated slightly (Ͻ5%), and the POCl 3 -treated rMoAChE did not reactivate, indicating maximum aging (data not shown). Immunoblots of these pairwise competition experiments showed that anti-AChE 10SP recognized both rMoAChE treated with POCl 3 followed by vehicle and rMoAChE treated with POCl 3 followed by DFP (Fig. 3). However, anti-AChE 10SP did not recognize rMoAChE treated with DFP followed by vehicle or, more importantly, rMoAChE treated with DFP fol-lowed by POCl 3 (Fig. 3). This result supports our hypothesis that anti-AChE 10SP is specific for the phosphoserine (3, Y ϭ O Ϫ ) in the active site and not a random phosphoserine or other phosphorylated group resulting from reaction with POCl 3 .
Anti-AChE 10S Recognition of Oxime-reactivated rMoAChE-When possible, OP-AChE conjugates can reactivate, a process resulting in restored AChE activity. Reactivation occurs via scission of the phosphoserine ester bond either spontaneously or promoted chemically and kinetically with oxime antidotes (e.g. 2-PAM). Since anti-AChE 10S antibody specifically recognizes uninhibited rMoAChE and not OP-modified rMoAChE, it was of interest to determine whether reactivated enzymatic activity could be correlated with anti-AChE 10S recognition.
To study the reactivation of OP-AChE conjugates, rMoAChE was treated with paraoxon, and the inhibition-reactivation processes were monitored versus time using the anti-AChE 10S antibody. Paraoxon was used for this experiment instead of POCl 3 because the reactivation of paraoxon-inhibited AChE is known and the postinhibitory rate of aging for paraoxon-inhibited AChE is slow (t1 ⁄2 ϭ 24 h) compared with POCl 3 (t1 ⁄2 ϭ 2.5 min) (5,10,24). To achieve the maximum reactivation of the paraoxon-AChE conjugate and test the specificity of anti-AChE 10S , rMoAChE was inhibited to less than 5% of the original activity with paraoxon (as described previously) and treated with increasing concentrations (100 -400 M) of 2-PAM versus time. The addition of 2-PAM at 100 M showed only a slight effect on reactivating paraoxon-inhibited AChE, whereas the addition of 2-PAM at 200 -400 M concentrations afforded proportionally greater percent reactivation to a maximum of 70% of the original activity. The time course of the paraoxon-AChE conjugate reactivation shows a large initial reactivation within 10 min that accounts for about 75% of the reactivation and another modest increase after 70 min (Fig. 4). The return of over 70% of the original activity at 90 min shows that a significant population of uninhibited rMoAChE can be regenerated following near complete inhibition of the enzyme. Although the addition of 100 M 2-PAM did not restore much reactivity, the possibility exists that 2-PAM removed the OP group from the OP-AChE conjugate but did not return the enzyme to its native active form due to denaturation or other consequence of the experiment. Therefore, in these experiments a small percentage of rMoAChE may be inactive yet contain the native sequence that is recognized by anti-AChE 10S .
To test the hypothesis that anti-AChE 10S is capable of quantifying total populations of active and reactivated AChE enzyme, rMoAChE was inhibited by paraoxon, and the reactivation process was monitored at 15 and 90 min kinetically via the colorimetric assay and probed mechanistically via recognition with anti-AChE 10S . After inhibition with paraoxon, samples  . 3. Anti-AChE 10SP recognition is specific for the phosphoserine in the context of the active site of rMoAChE. rMoAChE was treated stepwise as indicated (Treatment 1/Treatment 2), and the OP-AChE conjugates were detected with anti-AChE 10SP . Density of bands was quantitated and is shown below the immunoblot panels. Density is expressed as the adjusted density (OD/mm 2 ) after local background subtraction. THF, tetrahydrofuran; IPA, isopropanol.
were reactivated with 2-PAM (100 -300 M) for 15 min and simultaneously measured for AChE activity and recognition by anti-AChE 10S . The relative band density as recognized by anti-AChE 10S was plotted against the percent reactivation and, as evaluated by linear regression, showed an excellent correlation (R 2 ϭ 0.902) with slope ϭ 2.12. (Fig. 5A). The paraoxon-inhibited rMoAChE samples reactivated to 8 -24% of control activity when treated with 2-PAM, but without 2-PAM no reactivation occurred. Recognition by anti-AChE 10S of these same samples increased from near zero for inhibited rMoAChE (no 2-PAM treatment) to 21-52% of control when treated with 2-PAM. Analysis of the samples with anti-mAChE was conducted to ensure equal sample loading (Fig. 5A).
The correlation between the colorimetric assay and anti-AChE 10S was next examined after a 90-min reactivation period anticipating an increase in percent reactivation and a corresponding correlation with anti-AChE 10S recognition. rMoAChE was inhibited with paraoxon and treated with 2-PAM (100 -350 M), and the reactivation was allowed to proceed for 90 min whereupon the percent reactivation increased to 56 -103% of control. Recognition by anti-AChE 10S also increased to 49 -158% of control (Fig. 5B). Similar to the 15-min reactivation data, a strong correlation (R 2 ϭ 0.819) with slope ϭ 2.10 was found linking the kinetic data and anti-AChE 10S recognition. DISCUSSION In this report, we describe the rationale for and characterization of two polyclonal antisera raised against native and phosphate-modified active sites of AChE. As presented previously, OPs can vary in chemical composition leading to a panel of possible OP-inhibited AChE structures. Moreover each individual OP compound can lead to more than one OP-AChE conjugate structure by virtue of aging or other postinhibitory processes. On this basis, a number of possible OP-AChE structures are possible, yet little work has been done to develop methods or tools capable of measuring or differentiating between native AChE and the OP-AChE conjugate structures. Critical questions in this study were whether or not antibodies could be generated against a catalytic active site sequence of an enzyme, whether these antibodies would differentiate between native and modified enzyme, and finally would analysis on a denaturing gel allow for investigation of active site modifications that are buried deep within a protein structure. In sum, the development of these mechanism-based antibodies was envisioned to help identify OP-AChE conjugate structures, differentiate OP-AChE conjugates, and monitor reaction (inhibition/ reactivation) progress.
Antibodies raised against native and OP-inhibited AChE have been used to probe biochemical properties of AChE including (a) the three-dimensional structure of AChE (25), (b) allosteric influences over catalytic activity (26 -28), and (c) structural characteristics between isoforms of AChE (29,30). In one of these studies, antibodies were raised against a synthetic peptide corresponding to the active site sequence of Torpedo californica AChE for the purpose of determining whether the active site was present on external or internal domains of the enzyme (25). Anti-AChE 10S antiserum was found to differentiate between native, uninhibited rMoAChE and rMoAChE that had been reacted with several types of OP compounds. The only "nonspecific" interaction noted was in the modest recognition of POCl 3 -treated rMoAChE by anti-AChE 10S . Among the many reasons for this nonspecific interaction, one possibility is that the phosphate group may have been ejected, thereby reactivating the AChE and allowing recognition by anti-AChE 10S . Alternatively the phosphoserine group could have been eliminated to form a dehydroalanine, which may allow recognition by anti-AChE 10S . In support of the latter outcome, anti-AChE 10S recognition of peptide 4a was compared with a decapeptide that has the identical sequence of 4a except that it contains an alanine substitution for the serine residue (4a S/A ). By enzyme-linked immunosorbent assay, anti-AChE 10S was capable of significant recognition of 4a S/A , albeit less than the recognition of 4a (data not shown). This indicates that anti-AChE 10S may be capable of recognizing AChE with a dehydroalanine in the active site.
Anti-AChE 10S is also capable of recognizing rMoAChE that has been previously inhibited and reactivated with an oxime. Importantly the restoration of enzyme function and anti-AChE 10S recognition of the reactivated enzyme correlated very well. The slopes of both linear regression plots (Fig. 4) are greater than 1, suggesting that the antibody-based analysis better evaluates the molecular state of the active site than the colorimetric activity assays measure. This may be due to the fact that although scission of the phosphoryl group from the active site serine (reactivation) affords the correct epitope for antibody recognition, removal of the phosphoryl group may not necessarily restore the enzyme activity due to protein denaturation or other disruptive pathway. These results demonstrate that a strong correlation can be achieved between AChE activity and antibody recognition and also that antibodies generated to modified active site sequences can complement colorimetric enzyme assays in understanding the nature and extent of the inhibition process.
Although the colorimetric assay used to evaluate OP exposure provides rapid and reliable data on cholinesterase activity, it fails to identify the structure of the insulting OP agent. Further the colorimetric assay is not routinely used to ascertain the reactivation potential of the enzyme-OP complex although it can be used to evaluate reactivation rate. The reactivation of OP-AChE conjugates (Fig. 1) is significant because this time-dependent process varies greatly with both OP structure and stereochemistry (5,31). Antibody assays that can discriminate between populations of native, inhibited, or reactivated AChE molecules have the potential to provide more specific mechanistic evidence that the reactivation event occurred in part or in full. Further the ability of the antisera to recognize AChE that had previously been inhibited provides structural information of the reactivated enzyme active site.
The anti-AChE 10SP antiserum specifically recognized rMoAChE that had been inhibited to yield a dianion phosphoserine in the active site but not rMoAChE that had been inhibited with other OP compounds including those that age to afford monoionic phosphoryl groups (DFP and isomalathion). This result originally surprised us because we had hypothesized that anti-AChE 10SP would recognize rMoAChE containing any phosphate ion group at the catalytic serine even though the antibody was generated against the phosphate dianion. We hypothesize that the differences in charge on the phosphate and/or presence of an alkyl ester group may negate anti-AChE 10SP recognition. Based on this result, it is anticipated that antibodies can be raised to specifically recognize individual OP-AChE conjugates based on the precise chemical modification imposed by a specific reactive OP compound (Table I).
The anti-AChE 10S and anti-AChE 10SP antibodies generated for this study aided the measurement of active and inactive (covalent modification by OPs) rMoAChE. Moreover these antibodies were helpful in identifying the structure of the OP group attached to the active site serine of rMoAChE. This level of structural analysis and mechanistic detail is essential to understanding the direction and extent of postinhibitory processes, namely, reactivation and/or aging. For example, the presence of a charged phosphate group at the catalytic serine of AChE is far less likely to reactivate than an active site serine bearing a neutral phosphorus ester group. Such key mechanistic information could be useful in the clinical setting in which the identity of the OP-AChE conjugate is needed to evaluate therapeutic choices including the probability of oxime efficacy.
Our results establish native and modified enzyme active sites of AChE as differential epitopes for antibody recognition. Even subtle differences in the OP modifying group conferred clear differences in the selectivity and specificity of these mechanism-based antibodies toward native and modified rMoAChE. Antibody recognition also correlated well with existing methods of enzyme analysis (colorimetric assay of activity) thereby allowing for a more powerful combined approach to better understand the mechanism of action of OP compounds.