Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

OP insecticides are widely used worldwide and can pose a threat to human and animal populations because many of them are highly toxic, with rat oral LD₅₀s in the low mg/kg range for some compounds (Gaines, 1960). Our laboratories have developed an interest in the phosphorothionate insecticides parathion (rat oral LD₅₀, 3-14 mg/kg; Gaines, 1960) and chlorpyrifos (rat oral LD₅₀, 135-163 mg/kg; Worthing and Walker, 1987) because of their previous or current widespread use, and because of their roughly 10-fold relative difference in acute toxicity level. Pxn and Cpxn are the active metabolites of parathion and chlorpyrifos, respectively. These cytochrome P450-generated metabolites produce acute toxicity through inhibition of AChE (E.C. 3.1.1.7). The OP insecticides are readily metabolized by several oxidative and hydrolytic pathways.

A-esterases (E.C. 3.1.1.2, also paraoxonase) can hydrolyze organophosphates such as Pxn and Cpxn into their constituent acids and alcohols and are potentially one of the most important detoxication mechanisms for OP compounds. To date, the majority of A-esterase research has been conducted with serum enzymes and recently serum A-esterases have been purified from humans (Furlong et al., 1991; Gan et al., 1991) and rabbits (Furlong et al., 1991). Evidence strongly suggested that A-esterases exist as two isozymes (Adkins et al., 1993) which have similar molecular masses: 43 kDa each (Smolen et al., 1991) or 44.7 and 47.9 kDa (Gan et al., 1991) in humans and 35 and 38 kDa in rabbits (Furlong et al., 1991). Reports have also indicated that the isozymes vary in pH activity profiles, activity response to salts, buffer type and concentration and turnover rates for Pxn although no difference in turnover rate was apparent for PA or Cpxn (Gan et al., 1991; Smolen et al., 1991). We have recently reported the presence and isolation of two rat serum A-esterases, which have approximate molecular masses of 200 and 250 kDa (Pond et al., 1996).

A recent survey of rat tissues detected the highest level of A-esterase activity in liver (Pond et al., 1995); however, relatively little work has been done with hepatic A-esterases and purification of these proteins has proven challenging. Huang et al. (1994) reported purification of an A-esterase from mouse hepatic microsomes. Certain column chromatographic procedures detected more than one peak of A-esterase activity; however, 10% recovery of only one protein having A-esterase activity toward Pxn and methyl paraoxon was attained. Its molecular mass was about 40 kDa. The researchers suggested that multiple forms of A-esterase likely existed. Gil et al. (1993a, b) reported that rat liver A-esterase (paraoxonase) activity is mainly microsomal and achieved a partial purification of this activity. More recently, Rodrigo et al. (1997) reported purification of paraoxonase from rat liver. The purification protocol yielded 6% recovery of a protein exhibiting EDTA-sensitive hydrolytic activity toward Pxn, an apparent M_r of 45,000 and an isoelectric point of 4.7 to 4.8. A second peak of activity was detected using Cibacron Blue 3GA chromatography, but this activity was not purified.

Our laboratory has previously reported that although rat liver possesses high A-esterase activity toward high concentrations of PA, Cpxn and Pxn, it possesses substantial activity toward low, toxicologically relevant concentrations (10⁵ M) of Cpxn but not Pxn (Chambers et al., 1994; Pond et al., 1995). This dramatic difference in A-esterase reactivity between Cpxn and Pxn may be an important factor in the 10-fold lower toxicity of chlorpyrifos compared to parathion. It is possible that there is a separate protein with high affinity for Cpxn but low affinity for Pxn which is responsible for hydrolysis of Cpxn at low concentrations; however, our previous studies with hepatic whole homogenates or subcellular fractions could not determine the number of A-esterases, or, if more than one, their relative importance in the differential hydrolysis of Cpxn and Pxn. Our purpose was to further explore A-esterase activities in rat liver by determining: 1) the number of protein fractions expressing A-esterase activity; 2) methods for solubilizing and purifying these proteins and 3) which of the active proteins is/are responsible for the hydrolysis of oxon at high concentrations and which is/are responsible for the hydrolysis toward low oxon concentrations.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. Biochemicals were obtained from Sigma Chemical Company (St. Louis, MO) or Bio-Rad Company (Hercules, CA). The gel filtration medium was procured from Pharmacia, Inc. (Piscataway, NJ), Metofane (methoxyflurane) was purchased from Pitman-Moore, Inc. (Mundelein, IL) and organophosphate compounds were synthesized as previously described (Chambers et al., 1990).

Animal care. Male Sprague-Dawley rats [Crl:CD(SD)BR] (200-250 g), derived from an original Charles River stock, were housed at 22 ± 2°C in a 12:12 hr light cycle and given tap water and Purina Laboratory Rat Chow (Nashville, TN) ad libitum. All procedures were previously approved by the Mississippi State University Animal Care and Use Committee.

Sample preparation. Animals were euthanized by decapitation. Tris-HCl buffered saline (0.05 M, pH 7.7, 4°C) was injected into the liver via the hepatic portal vein and then the dorsal aorta to clear the organ of residual blood. The liver was then removed, rinsed with cold saline (4°C) and stored at 70°C for no more than 2 mo. Thawed liver samples were homogenized on ice in 0.05 M Tris-HCl buffer (pH 7.4, buffer I) using a motor-driven glass-Teflon Potter-Elvejehem tissue grinder driven by a motorized Wheaton stirrer.

Preparation of microsomes. Liver (45 g) homogenized at 350 mg/ml was centrifuged at 1000 × g (15 min, 4°C) to remove cell debris and nuclei. Calcium chloride (0.1 M) was added to the supernate to produce a final concentration of 1 mM. The supernate was re-homogenized and centrifuged at 17,000 × g (15 min, 4°C) to remove mitochondria. This supernate was then centrifuged (100,000 × g, 1 hr, 4°C) to sediment a microsomal pellet that was resuspended in buffer I containing 1 mM calcium chloride. The suspension was gently stirred with 0.5% Triton X-100 (30 min, room temperature) to solubilize the membrane-associated A-esterase. Nonsoluble proteins were then removed by centrifugation (100,000 × g, 1 hr, 4°C).

Purification of A-esterases. Ammonium sulfate was added to the solubilized microsome suspension to 20% saturation and stirred on ice for 10 min. The sample was centrifuged (20,000 × g, 15 min, 4°C) and the resulting pellet was discarded. Ammonium sulfate was then added to the supernate to yield a final saturation level of 45% and the sample was centrifuged (20,000 × g, 15 min, 4°C) to sediment a pellet containing A-esterase activity. The pellet was dissolved in buffer I containing 1 mM calcium chloride. The addition of ammonium sulfate to dissolved pellets (45% final saturation), stirring and centrifugation were repeated three times to remove extraneous proteins.

SDS PAGE. SDS- and native-PAGE were performed using either 7.5 or 10% acrylamide resolving gels and 4% acrylamide stacking gels on a Mini-Protean II Electrophoresis Cell (Bio-Rad) or on a LKB 2001 Vertical Electrophoresis Unit (Laemmli, 1970). The gels for SDS-PAGE contained 1% SDS, and sample preparation buffer consisted of 0.125 M Tris-HCl buffer (pH 6.8) containing 20% glycerol, 0.05% bromphenol blue and 4% SDS. Some samples and standards (as indicated in figure legends) were boiled (10 min) in sample preparation buffer containing 10% (volume/volume) 2-mercaptoethanol prior to SDS-PAGE analysis. A 0.05 M Tris buffer (pH 8.3) with 0.384 M glycine and 0.2% SDS was used as electrode buffer. Development by electrophoresis was conducted at 200 V constant voltage for 45 min. Electrophoresed gels were stained for protein with Coomassie Brilliant Blue R-250 (Weber and Osborn, 1969) or the Silver Stain Plus system (Bio-Rad) (Gottlieb and Chavko, 1987).

Native PAGE. Samples for native-PAGE (Laemmli, 1970) were diluted in 0.062 M Tris-HCl buffer (pH 6.8) containing 10% glycerol and 0.001% bromphenol blue. Pxn was added to a final concentration of 10⁵ M to inhibit carboxylesterase (E.C. 3.1.1.1) activity. The electrode buffer was the same as above without SDS. Electrophoresis was conducted at constant voltage (200 V) for 1 hr at 4°C. Developed gels were incubated in 200 ml of buffer containing 1.0 mM CaCl₂ or 1.0 mM disodium EDTA and incubated in a shaking water bath (37°C, 5 min). Gels were stained for A-esterase activity (Gomori, 1953) by addition of 2-NAc in ethanol and Triton X-100 (final concentrations of 5.0 mM, 0.25 and 0.005%, respectively) in 0.05 M Tris-HCl (pH 7.4, 37°C) containing either 0.1% Fast Garnet GBC or 0.05% Fast Blue RR. Substrate development was performed at an incubation temperature of 37°C until bands appeared (7-15 min). The reaction was terminated by transfer of gels to a stop solution consisting of methanol:acetic acid:water (5:1:5, vol/vol) for 15 min. The gels were then dried.

Direct A-esterase assay for activity toward chlorpyrifos-oxon. Direct assay for activity toward high concentrations of Cpxn was modified from Furlong et al. (1989). Samples containing either 1 mM EDTA (to correct for non-A-esterase hydrolysis) or 1 mM CaCl₂ were incubated at 37°C in a shaking water bath (15 min). The reaction was initiated with Cpxn in ethanol (320 µM final concentration). Reaction mixtures were incubated for an additional time period (within the linear range of the reaction) and then stopped with 0.5 ml 2% SDS. Triplicate subsamples were run. Absorbances were determined at 315 nm in a spectrophotometer (Perkin Elmer Lambda 5; Norwalk, CT) and the amount of 3,5,6-trichloropyridinol produced was calculated using a molar extinction coefficient of 7.704 × 10³ M¹ cm¹.

Direct A-esterase assay for activity toward paraoxon. Direct assay for activity toward high concentrations of Pxn was modified from the methods of Furlong et al. (1988). Samples containing either 1 mM CaCl₂ or 1 mM EDTA were incubated at 37°C in a shaking water bath (15 min). Pxn in ethanol was added to the samples to initiate the reaction (5 mM final concentration). Reaction mixtures were incubated for an additional time period (within the linear range of the reaction) and then stopped with 0.5 ml of a mixture of 2% SDS in 2% Tris base to denature protein and alkalinize the sample for 4-nitrophenol detection. Triplicate subsamples were run. Absorbances were read at 400 nm and the amount of 4-nitrophenol produced was calculated using a molar extinction coefficient of 1.7408 × 10⁴ M¹ cm¹.

Direct A-esterase assay for activity toward phenyl acetate. The assay for activity toward PA was modified from Johnson (1977). Samples containing either 1 mM CaCl₂ or 1 mM EDTA were incubated at 37°C in a shaking water bath (15 min). PA in ethanol was added to initiate the reaction (1 mM final concentration). The reaction mixtures were incubated for an additional time period (within the linear range of the reaction) and stopped with 0.5 ml of 0.4% aminoantipyrine in 5% SDS. An aliquot of 0.25 ml 1% potassium ferricyanide was added to develop color. Triplicate subsamples were run. Absorbances were read at 510 nm and the amount of phenol produced was calculated using a molar extinction coefficient of 1.4162 × 10⁴ M¹ cm¹.

Indirect A-esterase assay for activity toward chlorpyrifos-oxon. The indirect assay of activity, which allows assessment of the existence of significant hydrolysis of low toxicologically relevant concentrations of Cpxn and Pxn (10⁶-10⁸ M), was performed as described earlier (Chambers et al., 1994), with the exception that volumes were reduced to one fourth of those described. A selective carboxylesterase inhibitor, 4-nitrophenyl diphenylphosphinate (0.67 µM final concentration), was added to each sample and allowed to incubate at 4°C for 10 min before assay. The indirect assay assessed the residual oxon not hydrolyzed by the A-esterases based on the ability of the oxon remaining to inhibit an exogenous source of AChE. The concentrations of Cpxn which produced 50% AChE inhibition with EDTA were subtracted from the concentrations of Cpxn which produced 50% AChE inhibition with calcium. This difference represented the amount of organophosphate degraded and, therefore, the amount of product formed. This method allowed detection of A-esterase activity at substrate concentrations below the detection limits of direct spectrophotometric methods.

Protein assay. Protein concentrations were determined in triplicate using bovine serum albumin as the standard by the method of Lowry et al. (1951).

Kinetic analyses. Kinetic parameters for the hydrolysis of Pxn and Cpxn using the direct assays were determined from Eadie-Hofstee plots of velocity versus velocity/substrate concentration (Eadie, 1942; Hofstee, 1959) and Lineweaver-Burk plots of inverse velocity versus inverse substrate concentration (Lineweaver and Burk, 1934). The range of substrate concentrations tested was 1 to 10 mM for Pxn and 0.2 to 1 mM for Cpxn. The kinetic constants K_m and V_max were calculated from the plots by linear regression and the method of least squares applying the SAS statistical software package (SAS Institute, 1988).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

A-esterase activities toward high concentrations of Pxn (5 mM) and PA (1 mM) and toward high (0.32 mM) and low (10⁵ M) concentrations of Cpxn were detected in rat liver, with substantially greater activity toward Cpxn than Pxn apparent (table 1). The homogenate concentrations assayed are also indicated to further illustrate the difference in activities between the two substrates. A number of bands from whole liver homogenates stained for A-esterase activity on native-PAGE gels (fig. 1, lane 1). Three high molecular mass proteins stained under these conditions in the presence of 10⁵ M Pxn (fig. 1, lane 2); only two of these, however, stained more intensely in the presence of calcium. These latter two proteins were detected by this A-esterase staining procedure at all steps of the purification through the gel filtration in fractions exhibiting A-esterase activity toward PA, Pxn and Cpxn (fig. 2). The third, lower molecular mass band was eliminated from fractions having A-esterase activity by the purification process.

View larger version (40K): [in this window] [in a new window]   Fig. 1.   Native-PAGE gel of whole liver homogenate stained for A-esterase activity with Fast Blue RR using 2-naphthylacetate as the substrate in the presence of 1 mM calcium chloride. Lane 1 is a liver homogenate sample incubated for 5 min with 1 mM calcium chloride. Lane 2 is the same homogenate incubated for 5 min with 105 M paraoxon and 1 mM calcium chloride. After electrophoresis, activity was developed in the presence of calcium chloride.

View larger version (88K): [in this window] [in a new window]   Fig. 2.   Native-PAGE gel of rat liver fractions from purification stained for A-esterase activity with Fast Garnet GBC. Lane 1 is a microsomal fraction sample. Lane 2 is a solubilized microsomal fraction after 100,000 × g centrifugation. Lane 3 is a sample after ammonium sulfate fractionation. Lane 4 is a sample after gel filtration.

The solubilization of microsomal A-esterases using 0.5% Triton X-100 typically resulted in about 90 to 95% recovery (tables 2 and 3). The ammonium sulfate fractionation procedure used here, however, resulted in very low recovery (tables 2 and 3). This was a result of a physical loss of A-esterase and not to loss of catalytic activity.

Gel filtration chromatography of the dissolved ammonium sulfate pellet resulted in two major protein peaks (fig. 3). However, peak A-esterase activity occurred between the protein peaks. Samples within this area were pooled and retained (samples 69-100).

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Representative gel filtration chromatography fractionation.

Final purification and separation of the two A-esterase proteins was achieved by PSIEF although a large loss of activity was experienced (tables 2 and 3). In the final purification step with ampholytes of pH range 3 to 10, activity was typically found in wells 4 through 10 (pH range 4.5-7.5) with the peak activities occurring in wells 8 and 9 (pH 5.6 and 6.5, respectively). Typically, samples from PSIEF wells 4, 5, 6 and 7 (pH 4.6, 4.8, 5.1 and 5.2, respectively) contained one high molecular mass protein as detected by Coomassie Blue R-250 or silver staining for protein (fig. 4). These samples exhibited activity toward all three substrates measured directly. Wells 8 and 9 contained the two A-esterase proteins. With ampholytes of pH range 5 to 8, activity was typically found in wells 2 through 8 (pH range 4.0-6.4) with the peak well being 5 (pH 5.3). These ampholytes resolved the two proteins, with the smaller protein occurring in wells 6 and 7 (pH 5.4 and 5.9, respectively; fig. 5, lanes 3 and 4) and the larger one in wells 2, 3 and 4 (pH 3.9, 4.7 and 5.0, respectively). Well 5 sometimes contained both isozymes (fig. 5, lane 2).

View larger version (51K): [in this window] [in a new window]   Fig. 4.   SDS-PAGE gel of a representative PSIEF (preparative scale isoelectric focusing) procedure with ampholytes pH range 3 to 10 stained for protein via silver stain (Bio-Rad). Lane 1 contains these broad spectrum molecular mass standards (Bio-Rad): a, myosin; 206 kDa; b, -galactosidase; 137 kDa; c, bovine serum albumin, 83 kDa. Lanes 2 to 5 are PSIEF wells 4 through 7, respectively (pH 4.6, 4.8, 5.1 and 5.2, respectively).

View larger version (93K): [in this window] [in a new window]   Fig. 5.   SDS-PAGE gel of a representative PSIEF (preparative scale isoelectric focusing) procedure with ampholytes pH range 5 to 8 stained for protein via Coomassie Brilliant Blue. Lane 1 contains these broad spectrum molecular mass standards (Bio-Rad): a, myosin, 206 kDa; b, -galactosidase, 137 kDa; c, bovine serum albumin, 83 kDa. Lanes 2 to 4 are PSIEF wells 5 through 7, respectively (pH 5.3, 5.4 and 5.9, respectively).

The two protein fractions exhibited A-esterase activity toward all substrates assayed directly and toward Cpxn assayed indirectly. A protein that comigrated with albumin standards appeared in well 5; minute amounts of this protein sometimes appeared in well 6 (fig. 5). Although pH 5 to 8 ampholytes resolved the two proteins, their activities were more adversely affected than with the pH 3 to 10 ampholytes.

To ensure that no proteins with A-esterase activity were lost in the PSIEF procedure, wells from two PSIEF (ampholyte 3-10) procedures were pooled according to well number and pH followed by direct assay for activity toward Pxn and Cpxn and indirectly for activity toward Cpxn (fig. 6). This was done on three separate preparations with very similar results. PSIEF procedures suggested that the isoelectric point range for these proteins was 4.5 to 6.5. 

View larger version (17K): [in this window] [in a new window]   Fig. 6.   A-esterase activity toward paraoxon (direct assay) and chlorpyrifos-oxon (direct and indirect assays) in preparative scale isoelectric focusing unit wells: a representative sample set.

Purification using the pH 3 to 10 ampholytes was performed twice; wells 4 to 7 from both PSIEF procedures were subsequently pooled. A sample from this pooled extract was boiled in buffer with SDS and 2-mercaptoethanol as described earlier and analyzed by SDS-PAGE. No protein was detected in the boiled sample with Coomassie Blue staining; however, silver staining procedures detected a diffuse band (fig. 7, lane 1) which migrated to the same point as did a fraction in the non-boiled sample (fig. 7, lane 2). Its molecular mass was estimated to be 250 kDa using linear regression of the logs of the molecular masses of protein standards vs. migration distances.

View larger version (34K): [in this window] [in a new window]   Fig. 7.   SDS-PAGE gel of larger (250 kDa) A-esterase stained for protein via silver stain (Bio-Rad). Numbers at left represent molecular mass (in Da) of protein standards and their migration distance. Lane 1 is a purified sample which was boiled with -mercaptoethanol. Lane 2 is a purified sample which was not boiled or exposed to -mercaptoethanol.

The purification process using the ampholytes with pH range 5 to 8 was repeated four times and all number 6 wells were pooled. Applying reducing conditions (boiling, 2-mercaptoethanol) the second protein was detected with Coomassie Blue R-250 after SDS-PAGE development. The protein migrated with a protein standard of molecular mass 200 kDa (fig. 8).

View larger version (69K): [in this window] [in a new window]   Fig. 8.   SDS-PAGE gel of smaller (200 kDa) A-esterase stained for protein via Coomassie Brilliant Blue. Lanes 1 to 3 are varying concentrations of pure A-esterase. Lane 4 contains these broad spectrum molecular mass standards (Bio-Rad): a, myosin, 206 kDa; b, -galactosidase, 137 kDa; c, bovine serum albumin, 83 kDa.

Kinetic analyses of Cpxn hydrolysis (0.2-1.0 mM) were performed with three separately purified samples of the 250-kDa protein. Eadie-Hofstee plots indicated a K_m of 0.93 ± 0.04 mM (S.E.) and a V_max of 368.6 ± 25.6 nmol product formed/mg protein-min (mU/mg protein) in the presence of 1 mM calcium chloride at 37°C. The mean r² was 0.94 ± 0.02. Kinetic analyses using Eadie-Hofstee plots of Cpxn (0.2-1.0 mM) and Pxn (1-10 mM) performed with resuspended microsomal fractions (n = 6) gave a K_mapp of 1.53 ± 0.04 mM and V_max of 1197 ± 41 mU/mg protein for Cpxn, and a K_mapp of 4.48 ± 0.12 mM and V_max of 36.30 ± 1.01 mU/mg protein for Pxn. Lineweaver-Burk plots yielded calculations similar to those for Eadie-Hofstee plots with the 250-kDa protein and the microsomal fractions (data not shown). Kinetic analyses were not conducted with the 200-kDa protein because of insufficient quantities of pure material.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

A-esterase activities measured by direct assays in rat liver are about 10 to 11 times higher toward Cpxn than toward Pxn using approximately 16-fold more Pxn as substrate. Activity can be detected toward low (10⁵ M) concentrations of Cpxn but not Pxn. These results concur with previous reports concerning rat, human and rabbit sera (Gil et al., 1993b; Gomori, 1953; Furlong et al., 1989).

Native-PAGE gels of rat liver detected three proteins which stained for A-esterase activity (as described) in the presence of Pxn (10⁵ M). The two largest proteins stained more intensely than the third smaller protein and were activated in the presence of calcium whereas the third protein was not. It is likely that the third band was a `C'-esterase (i.e., a Pxn-resistant esterase which can hydrolyze 2-NAc but is not calcium-activated). This lightly staining third band (which was not always detected) disappeared from the A-esterase fractions during purification procedures. The purification procedures described in this report were developed to allow the purification of both of these major proteins so that the ability of each to hydrolyze low concentrations of Cpxn could be assessed. These two proteins have similar characteristics and are not easily separable.

About 55 to 65% of the whole homogenate activity was microsomal. Solubilization of microsomal fractions with 0.5% Triton X-100 resulted in little loss of A-esterase activity; this finding concurs with Gil et al. (1993b). The ammonium sulfate precipitation of A-esterase from this solubilized sample resulted in a large loss of A-esterase activity; a large quantity of A-esterase activity remained in the supernate which was discarded. Ammonium sulfate fractionation was attempted here because it is a common technique to remove extraneous protein. However, A-esterase, solubilized as we described, apparently is not easily precipitated by ammonium sulfate. A single fractionation with 20% saturated ammonium sulfate increased the final percent A-esterase yield. When the purification (as we described) was attempted with this sample, however, the 200-kDa protein was not purified. Purification of A-esterases from the sample yielded by the single ammonium sulfate fractionation would require modification of the procedures we described. Perhaps ion-exchange or nonspecific affinity chromatography (Furlong et al., 1991; Huang et al., 1994) could be included.

Preparative scale isoelectric focusing separated and purified two proteins of similar molecular mass and similar isoelectric point. The isoelectric point range of the 250-kDa protein (about 4.0-5.5) is similar to that of the protein purified by Rodrigo et al. (1997). It appears that the isoelectric point range of the 250-kDa protein is lower than that of the 200-kDa protein (about 5.0-7.0) although some overlap of proteins typically occurred in two or three wells with the pH range 3 to 10 ampholytes and often in one well with the pH range 5 to 8 ampholytes. Percent recovery of both proteins was low, which is consistent with other hepatic A-esterase purifications (Gil et al., 1993b; Rodrigo et al., 1997). Yield was sacrificed here to increase purity and achieve separation of the two isozymes so that characterization of their abilities to hydrolyze low concentrations of Cpxn could be performed. Also, the PSIEF step appeared to negatively affect A-esterase activity. Perhaps this step could be eliminated by eluting the proteins from native-PAGE gels of pooled gel filtration samples. Development of an immunoaffinity column from antibodies to the proteins purified here could improve recovery of these enzymes. However, such affinity columns (generated with antibody for one protein) may actually complex both A-esterases on each column if the proteins are as antigenically homologous as are the isozymes of human and rabbit sera (Adkins et al., 1993; Furlong et al., 1993; Hassett et al., 1991). Theoretically, separation of the two could perhaps also be achieved by elution from native-PAGE gels.

The detection of two A-esterase enzymes in rat liver correlates well with literature reporting the existence of two isozymes in rabbit and human sera (Gan et al., 1991; Furlong et al., 1991) and multiple forms of A-esterases in mouse liver (Huang et al., 1994). The molecular masses reported here (about 250 and 200 kDa), however, are considerably higher than those described in other reports (Gan et al., 1991; Furlong et al., 1991; Rodrigo et al., 1997). These molecular masses concur well, however, with molecular masses reported for rat serum A-esterases by our laboratory which were isolated by procedures similar to those described here for liver (Pond et al., 1996). Perhaps, because serum A-esterases have been reported to be associated with the high density lipoprotein complex (Mackness et al., 1985), the proteins of this complex remained intact during these purification procedures, thereby resulting in the high molecular masses. This purification method may have been more gentle and allowed the complex to be purified without disruption. It is possible that the two proteins purified here are actually two different forms of very similar protein complexes (with different molecular masses) which contain the same or different A-esterase enzyme(s). However, these rat A-esterases may actually be different proteins than those described in other species (Furlong et al., 1993) and rat (Rodrigo et al., 1997). When greater quantities of the purified proteins are obtained in the future, further characterization of these proteins (e.g., amino acid sequence analysis) can be conducted to determine their similarity with other proteins displaying A-esterase activity.

The purified 250-kDa protein was capable of calcium activated hydrolysis of Pxn and PA by direct assay and of Cpxn by both direct and indirect assay methods. It appears likely, therefore, that A-esterase and arylesterase activities are attributable to the same enzymes; this conclusion concurs with current literature (Gan et al., 1991; Furlong et al., 1991; La Du et al., 1993; Smolen et al., 1991). Although the 200-kDa protein could not be thoroughly characterized, it did hydrolyze both high and low concentrations of Cpxn, as indicated by the indirect assay results. It appears, therefore, that both proteins have similar activity toward Cpxn, including the ability to hydrolyze low toxicologically relevant concentrations of Cpxn, as indicated by the indirect assay results.

The K_mapp for Pxn in rat liver microsomes reported here (4.48 mM) is similar to other K_mapp values (0.5-9.9 mM) reported for rat liver preparations (Gil et al., 1993a; Pellin et al., 1990; Lauwerys and Murphy, 1969). It appears that only one report concerning A-esterase activity K_mapp for Cpxn in liver is available. The source reports a K_mapp of 1.87 mM in mouse liver microsomes (Sultatos and Murphy, 1983) similar to the K_mapp (1.53 mM) calculated here in rat liver microsomes. The 250-kDa protein was determined here to have a K_m of 0.93 mM for Cpxn which is roughly half the K_mapp for Cpxn reported for mouse liver microsomes (Sultatos and Murphy, 1983) and the K_mapp calculated here in rat liver microsomes; this is expected because the purified protein was free of any additional membrane components which could sequester some of the substrate.

The K_mapp determined here in microsomes for Cpxn is lower (approximately 3-fold) than that for Pxn suggesting that A-esterase activity has a slightly greater affinity for Cpxn. Further, the V_maxapp for Cpxn is 33-fold higher than that for Pxn. Therefore, the higher V_max for Cpxn is probably more important than the slightly greater affinity for Cpxn in contributing to the measurable hydrolysis of Cpxn and not Pxn at low (10⁵ M) concentrations.

To ascertain if the direct and indirect assays of Cpxn hydrolysis are measuring the same activity, the mean K_mapps and V_maxapps for Cpxn and Pxn were each fitted to Lineweaver-Burk plots and then velocities at low (10⁵ M) substrate concentrations were extrapolated from this using linear regression. Theoretically, this should indicate whether the enzymes responsible for the activity toward high substrate concentrations are capable of hydrolyzing the substrates at the low concentrations used in the indirect assays. Extrapolations predicted that the enzymes responsible for the direct activity should produce an activity of about 7777 pmol/mg protein-min with 10⁵ M Cpxn and that activity would decrease over time to 8 pmol/mg protein-min with loss of substrate to about 10⁸ M in the 15-min incubation time. The decay curve for substrate loss would be expected to be inversely logarithmic in shape, yielding a relatively low enzyme activity during the majority of the incubation period. Indirect assays of rat liver microsomal fractions typically yielded activities toward Cpxn of 1041 ± 183 pmol/mg protein-min (n = 3). It appears, therefore, that the direct and indirect assays are measuring the same activity. To further strengthen this postulate, when similar extrapolations to low substrate concentrations were performed for Pxn, the data predicted that at 10⁵ M Pxn velocity should be about 81 pmol/mg protein-min, about 1% of that for Cpxn, a value below the limits of detectability.

In summary, two proteins of similar molecular mass and isoelectric point which exhibited A-esterase activity toward high and low concentrations of Cpxn and high concentrations of Pxn were separated and purified from rat liver microsomes. Recoveries were low, however, and only enough of the higher molecular mass protein was obtained for further characterization. Kinetic studies indicated that this protein had a K_m of 0.93 mM toward Cpxn and a V_max of 368.6 nmoles/mg protein-min. It exhibited hydrolytic activity toward Cpxn, Pxn and PA suggesting that, at least in the case of rat liver, arylesterase and A-esterase activities are attributable to the same protein(s). Actual assays of purified proteins and kinetic studies of microsomes suggest that the activity toward high and low concentrations of Cpxn are due to the same protein(s) and that the high activity toward Cpxn at 320 µM is attributable to a higher affinity and a higher V_max (but primarily the latter) for Cpxn than for Pxn which is not detectably hydrolyzed at low concentrations. The higher A-esterase activity with Cpxn than Pxn may be a major determinant in the observed lower toxicity of chlorpyrifos relative to parathion.