Phosphoacetylcholinesterase: Toxicity of Phosphorus Oxychloride to

Dan Xiao , Zhi Li , Uwe Gohs , Udo Wagenknecht , Brigitte Voit , De-Yi Wang ... Gary B. Quistad , Susan E. Sparks , Yoffi Segall , Daniel K. Nomura , ...
41 downloads 0 Views 154KB Size
Download PDF
652

Chem. Res. Toxicol. 2000, 13, 652-657

Phosphoacetylcholinesterase: Toxicity of Phosphorus Oxychloride to Mammals and Insects That Can Be Attributed to Selective Phosphorylation of Acetylcholinesterase by Phosphorodichloridic Acid Gary B. Quistad, Nanjing Zhang, Susan E. Sparks, and John E. Casida* Environmental Chemistry and Toxicology Laboratory, Department of Environmental Science, Policy and Management, 114 Wellman Hall, University of California, Berkeley, California 94720-3112 Received February 11, 2000

Phosphorus oxychloride (POCl3) is an intermediate in the synthesis of many organophosphorus insecticides and chemical warfare nerve gases that are toxic to insects and mammals by inhibition of acetylcholinesterase (AChE) activity. It was therefore surprising to observe that POCl3, which is hydrolytically unstable, also itself gives poisoning signs in ip-treated mice and fumigant-exposed houseflies similar to those produced by the organophosphorus ester insecticides and chemical warfare agents. In mice, POCl3 inhibits serum butyrylcholinesterase (BuChE) at a sublethal dose and muscle but not brain AChE at a lethal dose. In houseflies, POCl3-induced brain AChE inhibition is correlated with poisoning and the probable cause thereof. POCl3 in vitro is selective for AChE (IC50 ) 12-36 µM) compared with several other serine hydrolases (BuChE, carboxylesterase, elastase, R-chymotrypsin, and thrombin) (IC50 ) 88-2000 µM). With electric eel AChE, methylcarbamoylation of the active site with eserine reversibly protects against subsequent irreversible inhibition by POCl3. Most importantly, POCl3-induced electric eel AChE inhibition prevents postlabeling with [3H]diisopropyl phosphorofluoridate; i.e., both compounds phosphorylate at Ser-200 in the catalytic triad. Pyridine2-aldoxime methiodide does not reactivate POCl3-inhibited AChE, consistent with an anionic phosphoserine residue at the esteratic site. The actual phosphorylating agent is formed within seconds from POCl3 in water, has a half-life of ∼2 min, and is identified as phosphorodichloridic acid [HOP(O)Cl2] by 31P NMR and derivatization with dimethylamine to HOP(O)(NMe2)2. POCl3 on reaction with water and HOP(O)Cl2 have the same potency for inhibition of AChE from either electric eel or housefly head as well as the same toxicity for mice. In summary, the acute toxicity of POCl3 is attributable to hydrolytic activation to HOP(O)Cl2 that phosphorylates AChE at the active site to form enzymatically inactive [O-phosphoserine]AChE.

Introduction The mechanism of toxicity of phosphorus oxychloride (POCl3) is of both practical relevance and academic interest. POCl3 is used in the synthesis of gasoline additives, hydraulic fluids, fire retardants, and organophosphorus pesticides (1), leading to potential human exposure. It is regarded by the Environmental Protection Agency as an extremely hazardous substance (1). POCl3 is intensely irritating to skin, eyes, and mucous membranes, and inhalation may cause pulmonary edema. It is a chlorinating agent and readily releases HCl on exothermic reaction with water, yet the toxicity (at least in lung) is not related to HCl release (2). POCl3 also reacts directly with a variety of cellular components ranging from serine in preparation of phosphoserine (3) to proteins such as β-lactoglobulin (4), casein, and lysozyme (5). Therefore, a potential alternative mechanism of toxicity, not previously considered, involves action as a phosphorylating agent. * To whom correspondence should be addressed: Environmental Chemistry and Toxicology Laboratory, Department of Environmental Science, Policy and Management, 115 Wellman Hall, University of California, Berkeley, CA 94720-3112. Telephone: (510) 642-5424. Fax: (510) 642-6497. E-mail: [email protected].

Our investigation of POCl3 started with the unexpected observation that its poisoning signs in both mice and houseflies are similar to or the same as those produced by triester phosphates that form dialkylphosphoryl derivatives of acetylcholinesterase (AChE).1 This led to the hypothesis that POCl3 is an anticholinesterase poison in mammals and insects due to selective phosphorylation of AChE by POCl3 or a derivative thereof. Although dialkylphosphoryl-AChE and its aged form (monoalkylphosphoryl-AChE) are well-known (6, 7), the possible formation of phospho-AChE (phosphoryl-AChE) would be a novel poisoning mechanism.

Materials and Methods Caution: POCl3 is volatile, corrosive, and extremely toxic and therefore must be used under containment conditions. The rat oral LD50 is 380 mg/kg (8). Chemicals. The sources were as follows: POCl3 (>99% pure), POBr3 (>95% pure), PSCl3 (>98% pure), and PCl3 (>95% pure) 1 Abbreviations: AChE, acetylcholinesterase; BuChE, butyrylcholinesterase; [3H]DFP, [1,3-3H]diisopropyl phosphorofluoridate; IC50, median inhibitory concentration; LC50, median lethal concentration; 2-PAM, pyridine-2-aldoxime methiodide; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TMS, trimethylsilyl.

10.1021/tx000028o CCC: $19.00 © 2000 American Chemical Society Published on Web 06/17/2000

Phosphoacetylcholinesterase from POCl3 from Aldrich Chemical (Milwaukee, WI), [1,3-3H]diisopropyl phosphorofluoridate ([3H]DFP) (8.4 Ci/mmol) from NEN Life Sciences Products (Boston, MA), dichlorvos (99% pure) (a commercial insecticide) from Chem Service (West Chester, PA), and other chemicals, biochemicals, enzymes, and plasma from Sigma Chemical (St. Louis, MO) or Aldrich. The preparation and characterization of phosphorodichloridic acid [HOP(O)Cl2] are described later. Toxicity and Poisoning Signs of POCl3. (1) Mice. Male Swiss-Webster mice (ca. 30 g, Charles River Laboratories, Wilmington, MA) were treated ip with POCl3 and HOP(O)Cl2 (0-100 mg/kg) using corn oil as the carrier solvent (50 µL). Dimethyl sulfoxide is not an acceptable vehicle for POCl3 because it is unstable in this solvent, in which it undergoes a rapid, exothermic reaction. Poisoning signs were observed, and mortality was recorded at 1 h (n ) 7-9 per dose). (2) Houseflies. Adult female houseflies (Musca domestica L., SCR strain, ca. 20 mg, 10-20 for each dose, 3-5 replicates) were added to an all-glass cylinder (3.8 cm diameter × 10.7 cm; 120 mL volume). A 1 cm2 filter paper was treated (3-fold difference in concentration) with POCl3 or dichlorvos (added neat or in 5 µL of THF) and attached to a glass top used to close the cylinder. The treated filter paper was separated from the houseflies by 1 cm using a single layer of cheesecloth. After exposure for 15 min at 25 °C, mortality was recorded and the houseflies were frozen on dry ice and removed for AChE activity assays (below). Relationship between Toxicity of POCl3 and in Vivo Cholinesterase Inhibition. (1) Mice. Tissue samples were removed 1 or 24 h after treatment or at the time of death for cholinesterase assays (n ) 3-5). Blood was obtained by cardiac puncture and serum recovered by centrifugation for the butyrylcholinesterase (BuChE) assay with butyrylthiocholine as the substrate (9). Brain, skeletal muscle (quadriceps), and diaphragm were removed at death (100 mg/kg) or after 1 h (0 mg/kg) and homogenized (1% w/v for brain and 10% w/v for quadriceps and diaphragm) in 100 mM phosphate buffer (pH 7.4) for an AChE activity assay of 50-100 µL aliquots (10). For diaphragm and quadriceps from control mice, eserine (10 µM) inhibits 90 ( 3 and 69 ( 6% (n ) 3) of the apparent hydrolytic activity for acetylthiocholine and this substrate is cleaved twice as fast as butyrylthiocholine, indicating that most of the observed hydrolysis is attributable to AChE. (2) Houseflies. Flies exposed to vapors of POCl3 or dichlorvos for 15 min were frozen and subjected to brain AChE activity assays using heads cut from the frozen insects. Ten heads per concentration (3-5 replicates) were homogenized in 100 mM phosphate buffer (pH 7.4, 625 µL), and 10 µL aliquots were assayed for AChE activity (11). The THF carrier (up to 5 µL per assay chamber) did not affect the AChE activity. Specificity of POCl3 for in Vitro Inhibition of Cholinesterases and Other Serine Hydrolases. AChE from electric eel (Electrophorus electricus) (0.02 unit), bovine erythrocytes (0.04 unit), or fly head or mouse brain as described above in 100 mM sodium phosphate buffer (pH 7.4, 500 µL) was reacted for 15 min at 25 °C with POCl3 added in THF (5 µL). Acetylthiocholine hydrolysis rates were then determined (10). Similar inhibitor sensitivity assays were carried out with BuChE (purified or blood plasma), carboxylesterase, elastase, R-chymotrypsin, and thrombin as described previously except here using THF as the carrier solvent for POCl3 and in the control (9, 12). The inhibitory potency of POCl3 was also compared with those of HOP(O)Cl2 and several other phosphorus derivatives in 15 min preincubation assays with electric eel AChE. Four concentrations up to 2000 µM were used at 2-3fold dose differentials, giving 15-85% inhibition (n ) 3-4 per concentration). The median inhibitory concentration (IC50) values that are reported are means of at least two independent experiments, and the relative SE values are the average for the two data points nearest the IC50 (n ) 4). AChE Inhibition by POCl3 Involves Reaction at the Esteratic Site. (1) Protection by Eserine and Acetylcho-

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 653 line. Electric eel AChE (0.2 unit) in 100 mM sodium phosphate buffer (pH 7.4, 500 µL) was treated with eserine (0 or 5 µM final concentration) added in the same phosphate buffer (10 µL) and incubated for 30 min at 25 °C. POCl3 (220 µM final concentration) in THF (5 µL) or THF alone (control) was added with incubation for 15 min at 25 °C. Alternatively, acetylcholine chloride (10 or 100 mM final concentration) was added just prior to POCl3. AChE was recovered by size-exclusion chromatography (10DG column packed with P-6DG gel, Bio-Rad, Hercules, CA) with all activity collected in a 1 mL fraction. Aliquots (100 µL) were assayed for AChE activity (10). Reproducible recovery of high levels of AChE activity requires pretreatment of the column with 1% bovine serum albumin (0.5 mL) and rinsing with buffer (8 mL) before each run (column reused). (2) Inhibition of Postlabeling by [3H]DFP. Electric eel AChE (9.3 units) in 100 mM phosphate at pH 7.4 (500 µL) was reacted with POCl3 (0, 11, 22, 73, and 220 µM final concentrations, n ) 3 each) or with carrier solvent alone as described above. AChE was radiolabeled with [3H]DFP (1 µCi added in 10 µL of propylene glycol) for 1 h at 25 °C, and then recovered by size-exclusion chromatography for analysis of hydrolytic activity (1% aliquot) and 3H content by liquid scintillation counting. A larger-scale reaction was used to verify that the sizeexclusion chromatographic fraction contained [3H]DFP-labeled AChE. Hence, AChE (28 units) in phosphate buffer (330 µL) was reacted with [3H]DFP (3 µCi added in 30 µL of propylene glycol) for 2.5 h at 25 °C. Bovine serum albumin (25 µg) was added to the excluded protein fraction collected from the BioRad 10DG column, and the sample was evaporated to dryness (Speed Vac concentrator, Savant, Farmingdale, NY). Half (73 000 dpm) of the protein fraction was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) (11% polyacrylamide gel), stained with Coomassie blue, destained, and cut into 2 mm slices with a gel slicer (model DE 113, Hoefer Scientific, San Francisco, CA) for quantification of 3H by liquid scintillation counting. Rates of AChE Inhibition by POCl3 and Subsequent Reactivation. To determine the rate of AChE inhibition, the electric eel assay as described above was varied by comparing AChE activity after incubation for 1, 5, 15, and 30 min with POCl3 (22 µM). In a reactivation assay, electric eel AChE was inhibited with 220 µM POCl3 as described above. Spontaneous recovery of AChE activity was assayed at 1-7 h. Pyridine-2aldoxime methiodide (2-PAM) (500 µM final concentration) was added for possible chemically induced reactivation. Potential enzyme-enhanced recovery of AChE activity utilized mouse liver cytosol (10-50 µL from a 20% homogenate) and horse blood plasma (20-50 µL), both previously treated for 30 min with phenyl saligenin cyclic phosphonate at 100 µM to inhibit liver esterases and plasma BuChE, respectively (12). Hydrolysis of POCl3 and Stability of Intermediates. (1) Synthesis of HOP(O)Cl2 [see also Hudson and Moss (13)]. A solution of water (6.6 mmol) in THF (10 mL) was added dropwise to POCl3 (6.6 mmol) in dry THF (10 mL) with stirring at -78 °C. The reaction mixture was warmed to room temperature and stirred for an additional 15 min. 31P NMR (121.5 MHz with a Bruker AM-300 spectrometer) showed a single major peak (δ -3.8 ppm relative to 85% phosphoric acid as the external standard) corresponding to HOP(O)Cl2 (>95% pure) which was confirmed by addition to excess dimethylamine in THF, giving quantitative conversion to HOP(O)(NMe2)2 (δ 18.0 ppm). HOP(O)Cl2 is stable in THF. The corresponding pyrophosphate [Cl2P(O)OP(O)Cl2] was also considered as a possible product, but it is less likely to form under dilute conditions, gives a 31P NMR chemical shift of δ -9.5 ppm, reacts with excess dimethylamine to give (Me2N)2P(O)OP(O)(NMe2)2 (δ 11.0 ppm), and is not stable in a 2:1 water/THF mixture, hydrolyzing instantly to phosphate. (2) POCl3 Hydrolysis and Reaction with Serine. POCl3 (6 µL, 64 µmol) was added to 200 mM borate buffer (pH 7.4, 500 µL) and the mixture shaken (5 s), and then the reaction progress was followed by 31P NMR. Alternatively, the buffer contained serine (100 mg, 950 µmol) or was replaced by water.

654

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

Figure 1. Toxicity of POCl3 administered ip to mice involves typical anticholinesterase poisoning signs and mortality with serum BuChE inhibition serving as a marker at sublethal doses. Inhibition data at 1 h posttreatment are means ( SE (n ) 3).

Quistad et al.

Figure 3. Potency of POCl3 and HOP(O)Cl2 as inhibitors of electric eel and housefly head AChE with preincubation for 15 min. Table 1. Specificity of POCl3 for in Vitro Inhibition of Cholinesterases and Other Serine Hydrolases serine hydrolase and source

Figure 2. Toxicity of POCl3 and dichlorvos vapors to houseflies correlated with inhibition of brain AChE activity. The exposure time was 15 min prior to noting the poisoning signs and mortality (n, no observed effect; p, paralysis; d, death) and assessing AChE inhibition. Inhibition data are means ( SE (n ) 3-6). Ethyl phosphorodichloridate [EtOP(O)Cl2] (6 µL) also replaced POCl3. (3) Hydrolysis Rates in Buffer (pH 7.4) for POCl3 and HOP(O)Cl2. The electric eel AChE inhibition assay allowed estimation of hydrolysis rates. Thus, POCl3 and HOP(O)Cl2 in THF (5 µL) were added to 100 mM phosphate buffer (pH 7.4, 495 µL) (73 µM final concentration). After 0-9.5 min at 25 °C, AChE (0.02 unit) was added (10 µL of buffer) and residual AChE activity was determined following incubation for 15 min. Amounts of POCl3 and HOP(O)Cl2 were calculated from inhibition curves. (4) Hydrolysis Rate in a 2:1 Water/THF Mixture for HOP(O)Cl2 Determined by 31P NMR and AChE Inhibition. Water (6.6 mmol) in THF (5 mL) was added dropwise to POCl3 (3.3 mmol) in THF (5 mL) at -78 °C to give HOP(O)Cl2. Water (20 mL) was added, and the water/THF (2:1) solution was warmed to 25 °C. Aliquots were removed simultaneously over the course of 4 h for 31P NMR analysis (500 µL) and the electric eel AChE inhibition assay (5 µL of a 1:7 dilution with THF).

Results Toxicity and Poisoning Signs of POCl3. (1) Mice. The mouse ip LD50 of POCl3 administered in corn oil is

AChE housefly (head) mouse (brain) electric eel bovine (erythrocyte) BuChE mouse (plasma) dog (plasma) horse (plasma, purified) human (plasma, purified) horse (plasma) carboxylesterase porcine (liver) elastase porcine (pancreas) R-chymotrypsin bovine (pancreas) thrombin bovine (plasma, purified) a

IC50 (µM)a 12 ( 2 32 ( 7 33 ( 6 36 ( 4 88 ( 23 160 ( 14 550 ( 127 1200 ( 120 1700 ( 140 140 ( 11 1030 ( 84 2000 >2000

Preincubation for 15 min; means ( SE (n ) 4).

40-60 mg/kg. The poisoning signs involve muscle fasciculation typical of nicotinic anticholinesterases. The dose-response relationship and poisoning signs for HOP(O)Cl2 are similar to those for POCl3 within the first hour after treatment. (2) Houseflies. POCl3 is a vapor toxicant for houseflies with a 15 min median lethal concentration (LC50) of ∼20 mg/L of air. Under the same assay conditions, dichlorvos gives an LC50 of ∼0.1 mg/L. The poisoning signs are the same for POCl3 and dichlorvos, proceeding from hyperactivity to paralysis typical of anticholinesterase poisons. Relationship between Toxicity of POCl3 and in Vivo Cholinesterase Inhibition. (1) Mice. Apparent AChE activity in diaphragm and quadriceps muscle at the time of death from a lethal dose of POCl3 (100 mg/ kg) is inhibited by 93 ( 9 and 75 ( 15%, respectively; i.e., the eserine-sensitive portion is completely inhibited. However, the AChE activity in brain is unaffected. Serum BuChE is much more sensitive and serves as a marker of poisoning (Figure 1). The 1 h IC50 dose is 12 mg/kg POCl3, about one-fourth of the LD50 dose. The inhibited BuChE appears to recover a portion of its activity within 24 h; i.e., 45 ( 17% inhibition at 1 h becomes 24 ( 7% inhibition at 24 h (n ) 4). (2) Houseflies. In contrast to the lack of brain AChE inhibition in mice treated ip with POCl3, there is com-

Phosphoacetylcholinesterase from POCl3

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 655

Table 2. POCl3 and Related Compounds as in Vitro Inhibitors of Electric Eel AChE compound

IC50 (µM)a

POCl3 HOP(O)Cl2 POBr3 PCl3 PSCl3 P(O)(OTMS)3 P(OTMS)3 phosphorylcholine acetyl phosphate

33 ( 6 25 ( 6 95 ( 22 400 ( 24 >200b >500 >500 >500 >500

a Preincubation for 15 min; means ( SE (n ) 4). b Highest concentration tested because PSCl3 hydrolysis products react with 5,5′-dithiobis(2-nitrobenzoic acid) in the Ellman assay procedure (10).

Figure 5. POCl3 inhibition of electric eel AChE activity and [3H]DFP postlabeling. Data points are means ( SE (n ) 3). [3H]DFP-labeled AChE (excluded fraction from the Bio-Rad 10DG column) is identified as a 55 kDa protein by SDS-PAGE. Figure 4. Eserine protection of electric eel AChE from POCl3 inhibition. AChE was incubated with eserine (0 or 5 µM) for 30 min followed by POCl3 (0 or 220 µM) for 15 min. AChE was then separated from inhibitors by size-exclusion chromatography prior to the assay of residual activity at various times for possible reactivation. Activity data are means ( SE (n ) 3).

plete inhibition in houseflies dying from exposure to POCl3 vapors (Figure 2). The IC50 for POCl3 is 6 mg/L versus 0.04 mg/L for dichlorvos, in each case with little or no sign of poisoning at 0-60% AChE inhibition but paralysis and death at g90% inhibition. Specificity of POCl3 for in Vitro Inhibition of Cholinesterases and Other Serine Hydrolases. AChE is the most sensitive serine hydrolase tested. POCl3 is about 3-fold more potent for housefly head AChE (IC50 ) 10-12 µM) than for AChE from electric eel (IC50 ) 2533 µM) (Figure 3), mouse brain, or bovine erythrocyte (Table 1). BuChE also is inhibited by POCl3 but less potently and with considerable species variability. Mouse plasma BuChE is most sensitive (IC50 ) 88 µM), consistent with its utility as a biomarker for poisoning from POCl3 in mice. Carboxylesterase is inhibited at relatively low levels of POCl3 (IC50 ) 140 µM), whereas other serine hydrolases (elastase, R-chymotrypsin, and thrombin) are less affected (IC50 > 1000 µM). POCl3 and Related Compounds as in Vitro Inhibitors of AChE. POCl3 and HOP(O)Cl2 (discussed later) are equipotent inhibitors of electric eel and housefly AChE (Figure 3) and are much more potent than other phosphorus halides that were examined (Table 2). Two trimethylsilyl (TMS) esters, phosphorylcholine, and acetyl phosphate were essentially inactive. These findings

Figure 6. Loss of electric eel AChE inhibitory activity for POCl3 and HOP(O)Cl2 (73 µM) in phosphate buffer (pH 7.4) at 25 °C. The main figure shows AChE inhibition data, and the inset presents the data as the amount of inhibitor based on the inhibition curve for POCl3 and HOP(O)Cl2 shown in Figure 3. Data points for AChE inhibition are the mean of three determinations with SE values averaging 5% of the mean.

identify HOP(O)Cl2 as the candidate inhibitor on hydrolytic activation of POCl3. AChE Inhibition by POCl3 Involves Reaction at the Esteratic Site. (1) General. Electric eel AChE was used to test the possibility that the site at which POCl3 reacts to inhibit enzyme activity is the same as that at which acetylcholine, eserine, and DFP bind or derivatize the enzyme. (2) Protection by Eserine and Acetylcholine. Eserine (5 µM) inhibits >90% of the AChE activity, and its removal allows recovery of 85% of the AChE activity

656

Chem. Res. Toxicol., Vol. 13, No. 7, 2000

Figure 7. Hydrolysis rate for HOP(O)Cl2 (110 mM) in a 2:1 water/THF mixture at 25 °C determined by 31P NMR and AChE inhibition. Data points for AChE inhibition (n ) 3) and the amount of HOP(O)Cl2 determined by 31P NMR are from a typical experiment.

at 2 h with a recovery t1/2 of about 30 min (Figure 4). Treatment of eserine-protected AChE with POCl3 (220 µM) followed by removal of excess eserine gave 74% of the AChE activity at 2 h compared to eserine alone. Likewise, AChE activity was 13 ( 2% of the control for 10 mM acetylcholine protection against POCl3 (220 µM) and was unchanged at 100 mM (12 ( 2%) compared to 1.0 ( 0.2% activity in the unprotected control (POCl3 alone). (3) Inhibition of Postlabeling by [3H]DFP. Treatment of AChE (9.3 units) with [3H]DFP gives 6140 ( 687 dpm (n ) 6) as radiolabeled AChE. Pretreatment with POCl3 at 0-220 µM (same conditions as eserine protection), and then incubation with [3H]DFP gives a concentration-dependent inhibition of [3H]DFP postlabeling which correlates with the decrease in AChE activity (Figure 5). The radioactivity is verified as [3H]DFPlabeled AChE by SDS-PAGE with predominant labeling of a 55 kDa protein (14). Rates of AChE Inhibition by POCl3 and Subsequent Reactivation. The reaction of POCl3 (22 µM) with electric eel AChE (IC50 ) 33 µM) is very fast with 28 ( 1% (n ) 5) inhibition at 1 min, 40 ( 7% (n ) 4) at 5 min, and not increasing thereafter (15-30 min). At 220 µM POCl3, 99% of the AChE activity is inhibited and does not recover spontaneously in 1-7 h. Recovery is assisted neither chemically by 2-PAM nor enzymatically with mouse liver cytosol or blood plasma.

Quistad et al.

Hydrolysis of POCl3, Stability of Intermediates, and Reaction with Serine. The initial hydrolysis product from POCl3 is HOP(O)Cl2, identified by 31P NMR, directly and after trapping with dimethylamine. HOP(O)Cl2 forms instantly in both buffer (pH 7.4) and water and converts to phosphate without 31P NMR evidence for (HO)2P(O)Cl [which has been identified from hydrolysis of PCl5 (15)]. In the presence of serine, HOP(O)Cl2 gives an about 3% yield of phosphoserine observed by 31P NMR (shoulder of a much larger phosphate peak) and confirmed by spiking with the authentic standard. EtOP(O)Cl2 also hydrolyzes within 1 min to O-ethyl phosphoric acid with no intermediates observed by 31P NMR. HOP(O)Cl2 Is Proposed To Be the Ultimate Inhibitor from POCl3. HOP(O)Cl2 has a potency (IC50) similar to that of POCl3 for AChE from electric eel (2533 µM) and housefly head (10-12 µM) (Figure 3). The AChE inhibitory activity from both POCl3 and HOP(O)Cl2 (each at 73 µM) decreases at the same rate in phosphate buffer (pH 7.4), and when expressed as the amount of inhibitor, the t1/2 is 2 min (Figure 6). The hydrolysis rate for HOP(O)Cl2 to H3PO4 is much slower in a 2:1 water/THF mixture as determined by 31P NMR and loss of AChE-inhibiting activity (Figure 7). The same curve is obtained with both end points, so the loss in inhibitory activity is due to hydrolysis of HOP(O)Cl2 with a t1/2 of 50 min.

Discussion The poisoning signs of POCl3 in ip-treated mice and fumigated houseflies indicate an anticholinesterase action. AChE inhibition by POCl3 in mouse quadriceps muscle and diaphragm (but not brain) is consistent with neuromuscular block, and BuChE inhibition in serum is a potentially useful marker for toxicity. Mouse serum BuChE is much more sensitive than mouse brain AChE to in vivo inhibition by POCl3, but the opposite relationship is observed for mouse plasma versus brain in vitro, suggesting that the active agent is present at critical levels in blood but not brain. On the other hand, in houseflies POCl3 inhibition of brain AChE activity correlates with the toxicity of this fumigant as noted earlier with many organophosphorus insecticides and chemical warfare agents applied topically to these insects (16). The action of POCl3 in a housefly is therefore very similar to that of the more potent commercial fumigant insecticide dichlorvos (17).

Figure 8. Proposed mechanism for electric eel AChE inhibition by POCl3 following hydrolytic activation to HOP(O)Cl2. Phosphorochloridic acids shown in brackets are not observed but are presumed to be very reactive intermediates. The phosphoric acids are predominantly present as anions. Eserine protects against phosphorylation by forming methylcarbamoyl-AChE which can be subsequently reactivated. The inhibited phosphoenzyme is not reactivated by H2O or 2-PAM. POCl3 [HOP(O)Cl2] inhibition of [3H]DFP postlabeling establishes Ser-200 as the probable site of phosphorylation.

Phosphoacetylcholinesterase from POCl3

POCl3 is a selective inhibitor of AChE compared with other serine hydrolases. The average IC50s for POCl3 observed here are 28 µM for four AChEs, 740 µM for five BuChEs, 140 µM for the carboxylesterase, and >1000 µM for three other hydrolases. The most sensitive enzyme that was assayed was housefly head AChE with an IC50 of 12 µM. Clearly, considerable specificity is involved in the esterase-POCl3 reaction to confer the selectivity as an inhibitor both in vitro and apparently in vivo. Three types of evidence suggest that POCl3 reacts at Ser-200 within the active catalytic triad of electric eel AChE. (1) The concentration-dependent inhibition of [3H]DFP postlabeling of AChE (18) correlates to inhibition of AChE hydrolytic activity. (2) Protective carbamoylation of AChE with eserine at the active-site serine prevents 74% of the inhibition by POCl3 as determined from enzyme recovery after hydrolytic removal of the carbamoyl moiety. (3) Excess (10 mM) acetylcholine substrate prevents 12% of the AChE inhibition by POCl3, although this same level totally protects AChE from [3H]DFP radiolabeling (14). The active agent for selective phosphorylation of AChE is proposed to be HOP(O)Cl2 rather than POCl3 itself (Figure 8). The AChE-inhibiting potency and toxicity are the same for the precursor and ultimate phosphorylating agent. (HO)2P(O)Cl cannot be detected by 31P NMR as an intermediate in this investigation and was not reported previously for POCl3 hydrolysis (13). POCl3 phosphorylates free serine but via HOP(O)Cl2 and not (HO)2P(O)Cl as previously purported (3). We propose that the anion of HOP(O)Cl2 has high affinity for the active site of AChE, positioning it for phosphorylation. The absence of detectable intermediates in the hydrolysis of HOP(O)Cl2 to phosphate and conversion of EtOP(O)Cl2 to EtOP(O)(OH)2 suggests that AChE-OP(O)(OH)(Cl) (if present) is rapidly hydrolyzed to AChE-OP(O)(OH)2. Although amino acids other than Ser-200 in electric eel AChE may be sensitive to phosphorylation, those sites seem to have little effect on the enzymatic activity. The inhibited enzyme is probably [O-phosphoserine200]AChE, analogous to the proposed [O-phosphoserine198]BuChE derived from human plasma BuChE inhibition by ethephon (19). Although ethephon is much less reactive, it probably also gives [O-phosphoserine]AChE in vitro since its IC50 is 270 µM for rat brain cholinesterase (20). [O-Phosphoserine200]AChE of electric eel could not be reactivated with 2-PAM (not surprising since it is similar to aged, organophosphate-inhibited AChE), liver cytosol (a protein phosphatase source), or blood plasma. Slow reactivation of [O-phosphoserine]BuChE was suggested in vivo for mice where blood serum BuChE activity inhibited with ethephon (9) or POCl3 (this study) undergoes partial recovery by 24 h. This investigation establishes that the acute toxicity of POCl3 to mammals and insects is attributable to hydrolytic activation to HOP(O)Cl2 which then phosphorylates AChE to form the enzymatically inactive phosphoAChE.

Acknowledgment. This work was supported by Grant R01 ES08762 from the National Institute of

Chem. Res. Toxicol., Vol. 13, No. 7, 2000 657

Environmental Health Sciences (NIEHS), NIH, and its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIEHS, NIH. We thank Franz Schuler of this laboratory for assistance.

References (1) Pohanish, R. P., and Greene, S. A., Eds. (1997) Hazardous Substances Resource Guide, 2nd ed., Gale Research, Detroit, MI. (2) Barbee, S. J., Stone, J. J., and Hilaski, R. J. (1995) Acute inhalation toxicology of oxalyl chloride. Am. Ind. Hyg. Assoc. J. 56, 74-76. (3) Neuhaus, F. C., Korkes, S., and Smith, R. A. (1958) Phosphoserine. Biochem. Prep. 6, 75-79. (4) Sitohy, M., Chobert, J. M., and Haertle, T. (1995) Phosphorylation of β-lactoglobulin using amino acids as the sole base and nucleophile of the reaction. J. Protein Chem. 14, 145-150. (5) Matheis, G., Penner, M. H., Feeney, R. E., and Whitaker, J. R. (1983) Phosphorylation of casein and lysozyme by phosphorus oxychloride. J. Agric. Food Chem. 31, 379-387. (6) Aldridge, W. N., and Reiner, E. (1972) Enzyme Inhibitors as Substrates: Interactions of Esterases with Esters of Organophosphorus and Carbamic Acids, North-Holland Publishing, Amsterdam. (7) Ballantyne, D. B., and Marrs, T. C., Eds. (1992) Clinical and Experimental Toxicology of Organophosphates and Carbamates, Butterworth Heinemann, Oxford. (8) Lewis, R. J., Sr. (1996) Sax’s Dangerous Properties of Industrial Materials, 9th ed., Van Nostrand Reinhold, New York. (9) Sparks, S. E., Quistad, G. B., and Casida, J. E. (1999) Organophosphorus pesticide-induced butyrylcholinesterase inhibition and potentiation of succinylcholine toxicity in mice. J. Biochem. Mol. Toxicol. 13, 113-118. (10) Ellman, G. L., Courtney, K. D., Andres, V., Jr., and Featherstone, R. M. (1961) A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 7, 88-95. (11) Mahajna, M., Quistad, G. B., and Casida, J. E. (1997) Acephate insecticide toxicity: safety conferred by inhibition of the bioactivating carboxyamidases by the metabolite methamidophos. Chem. Res. Toxicol. 10, 64-69. (12) Quistad, G. B., and Casida, J. E. (1999) Sensitivity of bloodclotting factors and digestive enzymes to inhibition by organophosphorus pesticides. J. Biochem. Mol. Toxicol. 14, 51-56. (13) Hudson, R. F., and Moss, G. (1962) The mechanism of hydrolysis of phosphorochloridates and related compounds. Part IV. Phosphoryl chloride. J. Chem. Soc., 3599-3604. (14) Keshavaraz-Shokri, A., Suntornwat, O., and Kitos, P. A. (1999) Identification of serine esterases in tissue homogenates. Anal. Biochem. 267, 406-411. (15) Mitchell, R. A. (1997) Kinetics of hydrolysis of PCl5 in situ as evaluated from the partial hydrolysis products formed in [18O]water. J. Chem. Soc., Dalton Trans., 1069-1073. (16) Mengle, D. C., and Casida, J. E. (1958) Inhibition and recovery of brain cholinesterase activity in house flies poisoned with organophosphate and carbamate compounds. J. Econ. Entomol. 51, 750-757. (17) Tomlin, C. D. S., Ed. (1997) The Pesticide Manual, 11th ed., British Crop Protection Council, Farnham, Surrey, U.K. (18) Salih, E., Chishti, S. B., Vicedomine, P., Cohen, S. G., Chiara, D. C., and Cohen, J. B. (1994) Active-site peptides of acetylcholinesterase of Electrophorus electricus: labelling of His-440 by 1-bromo[2-14C]pinacolone and Ser-200 by tritiated diisopropyl fluorophosphate. Biochim. Biophys. Acta 1208, 324-331. (19) Haux, J. E., Quistad, G. B., and Casida, J. E. (2000) Phosphobutyrylcholinesterase: phosphorylation of the esteratic site of butyrylcholinesterase by ethephon [(2-chloroethyl)phosphonic acid] dianion. Chem. Res. Toxicol. 13, 646-651. (20) Hennighausen, G., and Tiefenbach, B. (1978) U ¨ ber die Mechanismen der akuten toxischen Wirkung von Chlorcholinchlorid und 2-Chlora¨thylphosphonsa¨ure (Ethephon). Arch. Exp. Veterinaermed. 32, 609-621.

TX000028O