rOXICOLoCY
AND
APPLIED
104,246-258
PHARMACOLOGY
(
1990)
Molecular Determinants of the Species-Selective inhibition of Brain Acetylcholinesterase JON R. KEMP Department
of Pharmacology,
University
Received
August
AND KENDALL of Minnesota,
School
18, 1989; accepted
B. WALLACE’ of Medicine,
March
Duluth,
Minnesota
55812
6, 1990
Molecular Determinants of the Species-Selective Inhibition of Brain Acetylcholinesterase. KEMP, J. R.. AND WALLACE, K. B. (1990). Toxicol. Appl. Pharmacol. 104, 246-258. The objective of this investigation was to distinguish which of the catalytic features of enzyme action is principally responsible for conferring the observed insensitivity of trout brain acetylcholinesterase (AChE, EC 3.1.1.7) to in vitro inhibition by organophosphates. The experimental design consisted of comparing the kinetic constants for the hydrolysis of a series of acylthiocholine substrates as well as the inhibition constants for a homologous series of dialkyl pnitrophenyl phosphates among AChE from rats, hens, and trout. Chicken and rat brain AChE failed to distinguish between acetyl- and propionylthiocholine as inferred from the comparable Michaelis-Menten constants (K,), whereas trout brain AChE exhibited a much higher affinity for acetylthiocholine than for either of the two larger analogs. Diethyl pnitrophenyl phosphate was the most potent inhibitor toward chicken and rat brain AChE, whereas the IC50 for the trout enzyme increased progressively between dimethyl and di-n-propyl pnitrophenyl phosphate. The kinetic constants revealed that a significant determinant of inhibitor potency in the chicken and rat is steric exclusion as reflected by changes in the dissociation constant (&) which paralleled the changes in IC50 and k,. Conversely, Kdwas 120-to 1450-fold higher and did not vary significantly for trout brain AChE. Instead, the phosphorylation rate constant (I$,) for trout brain AChE decreased with progressive methylene substitutions. The kinetic data suggest that trout brain AChE not only possessesless steric tolerance, but also has a weaker nucleophile at the esteratic subsite, both ofwhich may be important factors in conferring the observed insensitivity of trout to acute organophosphate intoxication. o 1990 Academic PKSS, IW.
The relative insensitivity of most species of fish to acute intoxication by organophosphorus compounds such as Parathion (O,O-diethyl p-nitrophenyl phosphorothionate) is well established; the acute intraperitoneal LD50 for the biologically active oxygen analog, Paraoxon, varies from 1.1 and 2.6 pmol/ kg in hens and rats, respectively, to 27.3 rmol/kg in sunfish (Benke et al., 1974; Chattopadhyay et al., 1986). Attempts to elucidate the principal factor which confers this resistance to fish have focussed on species-related differences in pharmacokinetic variables, in’ To whom correspondence should be addressed. 0041-008X/90
$3.00
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form resewed.
eluding differences in the absorption, distribution and elimination of an acute dose, and species-related differences in metabolic transformation (activation or detoxification) of the organophosphates. Despite obvious anatomical and physiological differences, it has been concluded that the species-selective toxicity of organophosphates reflects more than simply subtle differences in kinetic disposition (Benke et al., 1974; Hodson, 1985). Studies designed to assessthe significance of species-related differences in metabolic activation or detoxification as major determinants of species-selectivity are complicated by the fact that the steady-state concentration 246
INHIBITION
OF
BRAIN
of the biologically active intermediate is a function of the relative activities of a group of nonspecific esterases acting both in series and in parallel to inactivate the reactive oxygen analog formed by the cytochrome P450mediated oxidative desulfuration of Parathion. Consequently, despite large differences among species in the activities of the mixed function oxidases and the nonspecific esterases responsible for the activation and detoxification of organophosphonates, respectively, the significance of metabolic transformation as a major determinant of species selectivity in organophosphorus poisoning remains questionable (Potter and O’Brien, 1964; Murphy, 1966; Hitchcock and Murphy, 197 1; Benke et al., 1974; Wang and Murphy, 1982b; Wallace and Dargan, 1987). Recent evidence suggests that the insensitivity of fish to organophosphate toxicity, rather than being a pharmacokinetic phenomenon, may reflect pharmacodynamic differences between species, manifested at the level of inhibition of acetylcholinesterase (AChE) by the active intermediate. The IC50 values for the inhibition of brain or plasma cholinesterase by Paraoxon vary over a 2000fold range between species, with chickens being the most sensitive, rats intermediate, and fish the least sensitive species (Dauterman and O’Brien, 1964; Murphy et al., 1968; Ecobichon and Comeau, 1973; Benke et al., 1974; Andersen et al., 1977; Wang and Murphy, 1982a,b; Johnson and Wallace, 1987). These differences are consistent with the differences in the in vivo sensitivity among species. Several groups of investigators (Becker et al., 1964; Darlington et al., 1971; Lotti and Johnson, 1978) reported direct correlations between the in vitro IC50 for AChE and the acute LD50 for a wide array of organophosphates in rodents, insects, and chickens. Accordingly, the sensitivity of AChE to inhibition in vitro appears to be a principal determinant of the species-selective toxicity of anticholinesterase agents and may be employed as a convenient screen for interspecies comparisons.
ACETYLCHOLINESTERASE
247
On a molecular basis, these differences in enzyme sensitivity may reflect any one or combination of several catalytic properties of AChE; (1) the dimensions of the active site, (2) the strength of the nucleophilic center responsible for interacting with the phosphoryl group, (3) the influence of a possible hydrophobic subsite, and (4) the distance separating the esteratic and anionic subsites. Structure-activity studies have revealed isolated differences in some of these properties between selected enzymes. Zahavi and coworkers (197 1) suggested that resistance in certain strains of mites to Malaoxon is conferred by a sterically smaller esteratic site, thereby excluding the large phosphoryl esters. Alternatively, species-related differences in sensitivity to inhibition may reflect differences in the proximity of the anionic site to the esteratic site (O’Brien, 1963; Hollingworth et al., 1967). Moss and Famey (1978) reported that although fish AChE possesses both anionic and esteratic subsites, the positive cooperativity between these regions is less than that exhibited by mammalian AChE, which is consistent with the resistance of the fish enzyme to inhibition by organophosphates. The objective of the present investigation was to elaborate on the observed differences between rats, chickens, and rainbow trout in in vitro sensitivity of AChE in an attempt to identify the principal catalytic feature of enzyme action responsible for the large idiosyncrasies between species. The investigation was limited to assessing speciesrelated differences in only those variables associated directly with the esteratic site of AChE (steric dimensions, nucleophilic strength, hydrophobic subsite). The paper expands beyond previous reports (Wang and Murphy, 1982a) to include a larger series of five structurally related analogs of Paraoxon, allowing for more definitive deductions regarding molecular and catalytic distinctions between different enzymes. Furthermore, the reactions were conducted at the respective physiological temperatures and, in order to eliminate possible confounding influences of
248
KEMP AND WALLACE
isozyme heterogeneity and allosterism between species, a homogeneous preparation of detergent-solubilized membrane-associated AChE was employed. Finally, the investigation employs a novel kinetic method (Hart and O’Brien, 197 3) to circumvent important restrictions imposed by the extreme rapidity of the inhibition reaction. METHODS Triton X- 100, acetyl-, propionyl-, and butyrylthiocholine iodide were purchased from Sigma Chemical Co. (St. Louis, MO). All other chemicals were reagent grade as supplied by Aldrich Chemical Co. (Milwaukee, WI). Dimethyl, diethyl, di-n-propyl, diisopropyl, and di-n-butyl chlorophosphates were synthesized by a modification of the method of Fiszer and Michalski ( 1952) and M&or et al. ( 1956). Briefly, three equivalents of the appropriate alcohol were added dropwise to one equivalent of cold (5-1o’C) phosphorus trichloride in hexane; an icebath was used to maintain the reaction temperature below 10°C. Immediately after the addition of alcohol, one equivalent of sulfuryl chloride was introduced dropwise to the newly formed trialkylphosphite. The hexane was removed under reduced pressure (water pump aspiration) and the dialkyl chlorophosphate purified by vacuum distillation. The various dialkyl pnitrophenyl phosphates were then prepared from the respective dialkyl chlorophosphates and sodium pnitrophenoxide by a modification of the methods of Fletcher et al. ( 1950) Schrader ( 1952) and DeRoos and Toet (1958). Specifically, sodium pnitrophenoxide was suspended in ethanol (I .5 M) and neat dialkyl chlorophosphate was added dropwise, producing a faint exothermic reaction. After the addition ofthe dialkyl chlorophosphate, the mixture was heated for 90 min at 87°C under a reflux apparatus. The reaction was cooled to room temperature then filtered to remove sodium chloride. Ethanol was removed by rotoevaporation under reduced pressure (water pump aspirator) and the resulting product was dissolved in methylene chloride and washed with a 5% aqueous solution of sodium carbonate to remove unreacted sodium pnitrophenoxide. The methylene chloride was dried with anhydrous magnesium sulfate, filtered, and then removed by rotoevaporation. The resulting dialkyl pnitrophenyl phosphate was then heated under reduced pressure (0.2 mm Hg) to remove traces of unreacted dialkyl chlorophosphate (Fletcher et al., 1950). Purity of the isolated dialkyl pnitrophenyl phosphates were greater than 95% as determined by GC-FID, and confirmed by GC-MS and HPLC-uv. The only detectable impurity was the corresponding chlorophosphate which was found to be inac-
tive in inhibiting AChE. Every 2 months neat dialkyl p nitrophenyl phosphates were dissolved in methylene chloride and washed with 5% sodium carbonate to remove hydrolysis products (pnitrophenol). From this oreanoohosphate reserve. stock solutions were orepared in absolute ethanol (stable for 1 week at -20”C)br in buffer containing 10% ethanol and 20% Triton X-100 (stable for 8 hr at -20°C). Final concentrations of ethanol and Triton X-100 in the reaction medium did not exceed 1 and 2.25%, respectively, neither of which affected the activity of AChE. White Leghorn hens (2-2.5 years old: 1300-2000 g) were supplied by Morterud’s Eggs, Inc. (Duluth, MN), and were killed by cervical dislocation. Male SpragueDawley rats (175-250 g) were purchased from Harlan, Inc. (Indianapolis, IN), and were housed under climatecontrolled conditions on a I2-hr light cycle and provided food and water ad libitum. All rats were fasted for at least 12 hr prior to their use and were killed by decapitation. Unsexed rainbow trout (750-1250 g) were purchased from Cedar Bend Trout Farm (Scandia, MN) and maintained at the U.S. EPA Environmental Research Laboratory (Duluth, MN) under regulated conditions and killed by a blow to the head. Membrane-associated AChE from brain homogenates was solubilized in detergent to recover the majority of the enzyme. Solubilization of 100,OOOgpellet of brain homogenates with Triton X-100 yields 90-95% of the total enzyme activity which, for rats and chickens, is almost exclusively in the form ofa globular 10 S or I 1 S tetramer of AChE (Rakonczay et al., 198 1; Vallette et al., 1983; Rotundo, 1984). Such a homogeneous enzyme preparation allows for comparisons of AChE among species without concern for possible differences in the composition of isozymes present and eliminates any possible differences in allosteric regulation of the enzyme due to species-related differences in membrane lipid composition (Ott, 1985; Stieger and B&beck. 1985). For each experiment, brain homogenates from the corresponding species were prepared from a single animal, with the exception of rainbow trout (two brains were pooled for each experiment). The brain was quickly transferred to ice-cold pH 8.0 buffer [IO0 mM sodium phosphate, 400 IIIM sodium chloride, and 3 mM sodium ethylenediaminetetraacetic acid (EDTA)] and weighed. For hens and rats a 20% (w/v) brain homogenate was prepared in a Wheaton glass homogenizer with Teflon pestel; a 12.5% brain homogenate was prepared from trout brain. The homogenate was centrifuged at lOO,OOOgfor 60 min (4’C). The resulting supematant was discarded and the pellet resuspended in an equal volume of pH 8.0 buffer containing 100 mM sodium phosphate, 3 mM sodium EDTA, and 0.5% Triton X-100 and recentrifuged at 100,OOOgfor 60 min. Sodium chloride was dissolved in the resulting supematant to a final concentration of 400 mrvr and the solution used that same day for all kinetic analyses and IC50 determinations.
INHIBITION
Cholinesterase assays were performed spectrophotometrically utilizing the method of Ellman et al. (196 1) as previously described (Johnson and Wallace, 1987). Acetyl-, propionyl-, and butyrylthiocholines have proven to be representative substrates for a variety ofAChE sources (Andersen and Mikalsen, 1978), with Michaelis-Menten constants (K,,,) corresponding to the oxygen analogs. Protein content for each enzyme preparation was determined by the technique described by Lowry et al. (195 1) using bovine serum albumin as standard. To determine K,,, and I’,,,,, for the entire series of substrates within each species, an excess of 5,5’-dithiobis(2nitrobenzoic acid) (DTNB) was added to AChE in buffer. The reactions were initiated by adding small aliquots of varying concentrations of the respective substrate to yield a final volume of 2 ml. Concentrations of substrate were within a range of one-fifth to five times the estimated Km and the amount of AChE included in the assay was adjusted to yield a maximum rate of0.05 absorbance unit per minute (A&min) at the highest substrate concentration employed. Corresponding blanks lacking AChE were subtracted to yield the enzymatic rate. Kinetic incubations for rat and hen brain AChE were conducted at 37°C. The reactions for the trout enzyme were performed at 11°C which is the temperature at which the fish were acclimated. Finally, determinations of rat AChE kinetics were repeated at I 1“C in order to compare the values with those for trout. The maximum velocity of substrate hydrolysis (V,,,,,) and the Michaelis-Menten constant (K,,,) were estimated by the double-reciprocal method of Lineweaver and Burk (1934), the line being calculated by a least-squares regression. ICSO values were determined as described in detail by Johnson and Wallace (1987), wherein the enzyme was preincubated for 30 min with differing concentrations of the inhibitor prior to measuring residual AChE activity. For all reactions, the enzyme was diluted such that in the absence of inhibitor, the rate of acetylthiocholine hydrolysis was approximately 0.1 A,,,/min. The data were analyzed by linear regression of the natural log (In) of inhibitor concentration versus percentage remaining enzyme activity following incubation for 30 min in the presence of inhibitor. A value of 50% was substituted into the resulting equation to solve for ICSO. The bimolecular inhibition constants for the organophosphates were estimated in one of two ways. The first method was employed to determine the inhibitor kinetic constants for rainbow trout brain AChE as described by Johnson and Wallace (1987),
The bimolecular inhibition constant, k,, is equal to k,/ Kd. Ten microliters of the appropriate concentration of inhibitor was incubated with 90 ~1 of the AChE preparation (which at Time zero yielded anA.&min ofO.l). The reactions were stopped at the indicated times by adding 1.5 ml of 1 mM acetylthiocholine- 1 mM DTNB and the A&min was recorded immediately over 5 min. Addition of excess substrate was effective at preventing the continued inhibition of AChE and provided reliable kinetic measurements. This procedure was repeated with six different concentrations of each dialkyl p-nitrophenyl phosphate. Although the above kinetic method worked well for the trout enzyme, it was unsatisfactory for rat or hen AChE. With these species, the reaction rates were so rapid that it was virtually impossible to employ inhibitor concentrations between one-fifth and five times the Kd as required. According to the above equation, when [I] B Kd, kam becomes a direct function of [I]; k., = k, [I]/ Kd = k;[I]. As expected, double-reciprocal plots of the data intersected very near the origin yielding spuriously high estimates of both k, and Kd. Accordingly, it was necessary to modify the kinetic method for rat and hen AChE in order to approach the Kd for the enzyme and to obtain accurate estimates of the kinetic constants. This was accomplished by including a known concentration of acetylthiocholine in the reaction mixture to compete with, and thus retard, the inhibition of rat or hen AChE by the organophosphates (Hart and O’Brien, 1973). Ten microliters of the appropriate concentration of inhibitor was added to 1.5 ml of buffer containing 1 mM DTNB and 600 or 400 pM acetylthiocholine for the chicken or rat, respectively. This solution was equilibrated to 37°C then 90 ~1 of the corresponding AChE preparation (added at a concentration which in the absence of inhibitor yielded a rate of0. I Aalz/min) was added and the progressive decrease in the rate of substrate hydrolysis was recorded over a IO-min period. This was repeated with at least six concentrations of each dialkyl pnitrophenyl phosphate. The reaction can be represented by [E] + [I] 2 [E.] 2 [E-I] where,
kl k-*
It
[E] + [I] 2 [EI] 1 [EI’]
+ 14)
where k.pp is the observed velocity of inhibition; kp, the first-order phosphorylation rate constant; and Kd, the dissociation constant for the enzyme-inhibitor complex.
Kds = Kd(l
+ [S]/K,)
kz
tES1 k,
[El + [PI li,,,, = WI/(&
249
OF BRAIN ACETYLCHOLINESTERASE
therefore,
k.,, =
kp’[ll Kis + [II
1 Kds 1 and G=k,‘cI]+k,’
1
where Kds is the apparent Kd for the enzyme-inhibitor complex determined in the presence of substrate, KS is the Michaelis-Menten constant (k-,/k,) determined for acetylthiocholine in the species under study, and [S] is the concentration of acetylthiocholine added to the reaction.
KEMP AND WALLACE
250
TABLE I
Acetylthiocholine V,, (nmol/min .mg protein) Hen (37°C) Rat (37°C) Rat(ll”C) Trout ( 11“C)
350.1 289.5 159.0 148.9
Km (FM) Hen (37°C) Rat (37°C) Rat(ll”C) Trout ( 11‘C)
105.2 * 77.5 * 99.1 * 197.2 i
f 44.4” + 6.7 f 3.7 f 11.1 7.0* 3.3* 4.0* 51.3’
Propionylthiocholine 266.8 + 185.8 + 86.2 f 15.1 f
28.7h 5.3* 2.7 b 1.0*
117.7+4.9* 83.3 f 4.2’ 92.3 f 9.3b 2078.5 2 203.2b
Butyrylthiocholine 16.7 f 1.2’ 8.7 + 0.5’ 2.4 kO.lb n.d.’ 23.0 k 1.2* 337.8 + 40.5’ 183.3 f 22.8* n.d.’
’ Values represent the means ? SE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p < 0.05). * Significantly different from the corresponding values for all other enzyme preparations (p < 0.05). ’ n.d. signifies that the reaction proceeded too slow to allow for reliable estimation ofthe kinetic constants.
In this scheme, as with the method used for trout brain AChE, the k,, for each inhibitor concentration was determined by regressing residual enzyme activity @[El,/ [El,) against time (Fig. I): then the double-reciprocal plot was formed with the inverse of each inhibitor concentration regressed against the inverse of -kapp (Fig. 2). The reciprocal of the y-intercept is equal to the k, and the negative reciprocal of the x-intercept is equal to I&; however, a correction factor must be intejected to determine the actual Kd and thus the k, (Kd = K&/( I + [S]/K,) and k, = kp/Kd). Statistical analyses of the log transformed data were accomplished by analysis of variance and the individual means compared by the method of Scheffefor all possible comparisons both within and across species (Sokal and Rohlf, I98 I). A probability of p < 0.05 was used as the criterion for statistical significance.
RESULTS The kinetic constants (V,,, and K,,J describing the hydrolysis of the three acylthiocholine substrates by AChE from each of the three species (including the rat at 11°C) are presented in Table 1. Within each species the V,,,, for the individual thiocholine substrates were significantly different from each other; however, in the trout it was impossible to measure either V,,,,, or K,,, for butyrylthiocholine as the limits of solubility for this com-
pound were reached before any appreciable rate of hydrolysis could be measured. For each species studied, acetylthiocholine was hydrolyzed at the fastest rate, followed by propionyl- and then butyrylthiocholine; however, there were large differences in the degree of substrate preference among species (Table 1). For rat and chicken AChE, the V,,,,, for propionylthiocholine was between 64 and 76% of that for acetylthiocholine. In contrast, the trout enzyme exhibited a more profound preference for the smaller analog, hydrolyzing propionylthiocholine at only 10% of the maximum rate for acetylthiocholine. This idiosyncrasy of the trout brain enzyme was not due to the different temperature at which the reaction was conducted since the rat enzyme incubated at 11 “C exhibited a substrate specificity comparable to that observed at 37°C. Comparing substrate specificity among species, there was no significant difference between the maximum rate of acetylthiocholine hydrolysis for the hen or rat AChE or between trout and rat when both experiments are performed at 11°C. For any given substrate, all other possible comparisons across species were significantly different.
INHIBITION
251
OF BRAIN ACETYLCHOLINESTERASE TABLE 2
IC~OFORTHEINHIBITIONOFBRAINACETYLCHOLINESTERASEBYVARIOUS DIALKYL-p-NITROPHEN~LPHOSPHATES Dialkyl-p-nitrophenyl Methyl Species Hen (37°C) Rat (37°C)
Rat(ll”C) Trout (11°C)
0.037 + 0.003” 0.029 f 0.00 1 0.108~0.005b 3.3 1 + 0.20”
Ethyl
0.028 0.026 0.109 70.04
-+ 0.000
+ 0.001 k o.002b * 4.95b
phosphate IC50
n-Propyl
(pM)
n-Butyl
Isopropyl
0.336 -+0.013’ 0.228 f 0.004b 3.280 t 0. I 16”
0.762 _+ 0.053 0.453 + 0.008 8.400 f 0.740’
3.30+0.10b 2.12 k 0.056 35.64 + 1.76’
963.0-t
763.6 + 142.7’
12,779 rt 1002’
137.1b
’ Values represent the means -CSE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p < 0.05). b Significantly different from the corresponding value for all other species (p < 0.05).
The chicken and rat enzyme failed to dis- species studied. Repeating the IC50 determinations for rat AChE at 11°C yielded a trend tinguish between acetyl- and propionylthiocholine as indicated by the comparable val- similar to that observed at 37°C; IC50 inues for K, for the two substrates (Table I), creased with increasing methylene substituwhereas the trout was far (IO-fold) more se- tion. The individual IC50 values for each inlective for acetylthiocholine than for the pro- hibitor were, however, greater at the lower pionyl analog. For rat and trout AChE, buty- temperature. For the chicken and rat AChE, rylthiocholine yielded a significantly greater all possible comparisons among the different K,,, compared to the two smaller acylthiochoinhibitors were significantly different (except lines; however, the K,,, for the butyryl analog when the rat enzyme was incubated at 11°C in the chicken was 4.8 times lower than the dimethyl and diethyl g-nitrophenyl phosphate were not statistically different). K,,, for either the acetyl or propionyl derivaThe rainbow trout brain AChE was most tive. All possible comparisons of K,,, across by dimethyl pnitrospecies for a given acylthiocholine were sig- sensitive to inhibition phenyl phosphate (Table 2). However, the nificantly different. Diethyl p-nitrophenyl phosphate was the IC50 was approximately 1OO-fold higher than most potent inhibitor for both chicken and that for rat or chicken (IC50 = 3 PM). Unlike rat brain AChE, with an IC50 of 28 and 26 the rat or chicken enzyme, inhibitor potency toward trout AChE decreased 23-fold as the nM, respectively (Table 2). Dimethyl p-nitrophenyl phosphate was almost as potent in the result of including a single additional methyterrestrial species, displaying an IC50 of 37 lene group (diethyi analog IC50 = 70 PM). nM for chicken AChE and 29 nM for rat This trend continued in a stepwise fashion for the remaining inhibitors; however, di-n-butyl AChE. For rat and chicken AChE, a dramatic transition in potency occurred between di- pnitrophenyl phosphate (IC50 = 764 PM) ethyl and di-n-propyl pnitrophenyl phos- was found to be equipotent with di-n-propyl phate, the IC50 increasing 12- and 9-fold, p-nitrophenyl phosphate (IC50 =963 PM). respectively. Potency continued to de- All other possible inhibitor IC50 comparicrease with increasing methylene substitution sons within the trout were significantly through di-n-butyl p-nitrophenyl phosphate; different from each other. When comparing diisopropyl p-nitrophenyl phosphate being each inhibitor across species, there was no the least potent inhibitor of the series for all difference in the IC50 for dimethyl, diethyl,
252
KEMP AND WALLACE Minutes
-0.5 0 iii v iii C
-1.5
-2.5
FIG. 1. Log-linear progression plot for the inhibition of rat brain AChE by increasing concentrations of diethylp-nitrophenyl phosphate at 37°C. The reactions were performed in the presence of known concentrations of acetylthiocholine, functioning as a competitive substrate to retard the rate of inhibition of AChE as described in the text. The ordinate represents the fraction of enzyme activity remaining at time f. Each inhibitor concentration is depicted by a single line, the slope of which equals the respective -Y&,~. The results are those of a typical experiment.
The bimolecular inhibition constants (k;) for the series of dialkyl p-nitrophenyl phosphates across species are summarized in Table 3. Describing the overall reaction between inhibitor and enzyme, kj is a measure of inhibitor potency and can be related mathematically to the IC50 (Johnson and Wallace, 1987). Whereas a low IC50 indicates a potent inhibitor, the inverse is true for k;; the larger the k, the more potent the inhibitor. As with the IC50 data, dimethyl and diethyl p-nitrophenyl phosphate were the most potent inhibitors toward chicken and rat brain AChE as reflected by ki. The comparable k, values for these two inhibitors suggest that neither the rat nor the chicken AChE distinguished between dimethyl and diethyl p-nitrophenyl phosphate. However, there was a stepwise decrease in k, from diethyl through diisopropyl p-nitrophenyl phosphate for these two species of AChE. Di-n-butyl p-nitrophenyl phosphate was 14-fold less potent than diethyl p-nitrophenyl phosphate toward chicken AChE and 12-fold less potent toward rat brain AChE. Similar to that observed for the IC50 data, diisopropyl p-nitrophenyl phosphate was the least potent inhibitor toward all species studied, ki being 72- and 63-fold lower
and di-n-butyl p-nitrophenyl phosphate between chicken and rat AChE when both were determined at 37°C. All other comparisons between species were significantly different. Figures 1 and 2 depict typical kinetic plots for the inhibition of rat brain AChE by diJ ethyl p-nitrophenyl phosphate employing the 2.0 modified kinetic model wherein acetylthio1.6 choline was present during the course of enzyme inhibition, Each reaction was allowed to proceed until approximately 90% of the enzyme activity was inhibited (Fig. 1). Least.i/ squares regression of the progression plots yielded lines of increasing slope (-k,,,) corre0.4 sponding to increasing inhibitor concentrai/l 0.04 .‘# ! . , . I , , tions. In all cases for each species and all in-0.2 -0.1 0.0 0.1 0.2 0.3 0.4 hibitors, r2 3 0.95. The inhibition constants 1Ml (PM) were derived from double-reciprocal plots of kappregressed against inhibitor concentration FIG. 2. Double-reciprocal plot for the inhibition of rat (Fig. 2). Once again, r2 3 0.95 for all cases. brain AChE by diethyl-pnitrophenyl phosphate at 37°C. The values for Knappwere derived from Fig. 1. LeastThe slope of this line is equal to l/ki, intersquares linear regression yielded a line which intersected secting the ordinate at l/k, and the abscissa the ordinate at (&J’ and the abscissa at -(&)-I, where at l/Kds. The dissociation constant (Kd) is Kdr = KAI + [S]/K,). In this example, [S]/K, was 5 and calculated by dividing K~s by ( 1 + [S]/K& thus K*= 6%.
INHIBITION
253
OF BRAIN ACETYLCHOLINESTERASE TABLE 3
BIMOLECULARRATECONSTANTS(~~)FORTHEINHIBITIONOFACE~LCHOLINESTERASEBYASERIES OFDIALKYL-p-NITROPHENYLPHOSPHATES Dialkyl-pnitrophenyl Methyl Species Hen (37’C) Rat (37°C) Trout ( 11“C)
828k34".b 1238+59' 5.37 IL 0.08'
Ethyl
n-Propyl
1138+44
18324
1463+20
282k12
0.375
+o.030b
phosphate k, (mM . min)-’
0.042+0.015b
n-Butyl
79*56 1242
Isopropyl
16k3
lib
0.027 +O.OOlb
23&l 0.0013
+ 0.0001~
0 Values represent the means -t SE ofthree separate enzyme preparations. Means underscored by the same line are not significantly different (p < 0.05). b Significantly different from the corresponding value for the other two species (p < 0.05).
than that for diethyl pnitrophenyl phosphate in chicken and rat, respectively. For rat and chicken AChE, all possible comparisons of ki for the individual inhibitors were significantly different, with the only exception being dimethyl and diethyl p-nitrophenyl phosphate. The kj data indicate that the entire series of dialkyl p-nitrophenyl phosphates were less potent toward the trout AChE than toward the other species (Table 3). Trout brain AChE exhibited the greatest sensitivity to dimethyl pnitrophenyl phosphate with a ki of 5.4 (mM. min))‘. From dimethyl pnitrophenyl phosphate there was a progressive decrease in k; with increasing methylene substitution: the diethyl analog having a ki of 0.38 (mM. min))‘, di-n-propyl and di-n-butyl possessing a k, of 0.042 and 0.027 (mM.min)-‘, respectively, while diisopropyl p-nitrophenyl phosphate had the lowest ki- of 0.0013 (mM. min)-‘. Other than di-n-propyl and di-n-butyl p-nitrophenyl phosphate, the individual ki values for the inhibitors against trout AChE were all statistically different from one another. An analysis across species for each inhibitor showed no significant difference between the ki observed in the chicken and rat for diethyl, di-n-propyl, or diisopropyl p-nitrophenyl phosphate; all other possible comparisons for each inhibitor across all three species were statistically different.
Table 4 illustrates the dissociation constants (&) for all inhibitors in the three species studied. Brain AChE from chicken and rat possessed the lowest Kd for dimethyl and diethyl pnitrophenyl phosphate, the values not being statistically different. There was, however, a two- to sixfold increase in the Kd between diethyl and di-n-propyl p-nitrophenyl phosphate in the chicken and rat and this trend continued in a stepwise fashion with the Kd increasing with successive methylene substitutions through di-n-butyl pnitrophenyl phosphate for both species of AChE. Both the chicken and rat enzyme failed to distinguish between the Kd for the dimethyl and diethyl analogs; similarly, no difference was observed when comparing the Kd of chicken AChE for di-n-propyl and di-n-butyl p-nitrophenyl phosphate. Rat brain AChE also did not distinguish between the Kd values for diethyl and di-n-propyl p-nitrophenyl phosphate or between the di-n-propyl and the din-butyl analog. All other comparisons of Kd for the inhibitors of rat or chicken AChE were statistically significant. Rainbow trout brain AChE possessed Kd values which were two to three orders of magnitude larger than the corresponding Kd for the rat or chicken enzyme (Table 4). However, in contrast to rat and chicken AChE, Kd for the trout enzyme did not vary significantly between the five inhibitors examined. When comparing across spe-
254
KEMP AND WALLACE TABLE 4 DISS~CIATIONCONSTANTS( Kd) FORTHEINHIB~TIONOFACETYLCHOLINESTERASEBYASERIES OFDIALKYL-p-NITROPHENYLPHOSPHATES Dialkyl-p-nitrophenyl
Species Hen (37°C) Rat (37°C) Trout(!
1°C)
phosphate Kd (PM)
Methyl
Ethyl
n-Propy!
n-Buty!
Isopropyl
2.48 rt: 0.20” 1.26 +0.19 1830t_360*
2.21 2 0.37 1.62 t 0.67 851 +_ t39b
12.15 20.54 4.02 f 0.4 1
35.8 -t 12.6 9.03 +_ 1.58
121.1 + 35.6 56.24 + 4.43
1924 +- 547 b
10363 t 14761~
14852 t- t2668b
y Value represent the means f SE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p -c 0.05). b Significantly different from the corresponding value for the other two species (p < 0.05).
ties, the Kd for the entire series of inhibitors phosphates: dimethyl p-nitrophenyl phosdid not differ between rat and chicken AChE. phate possessed the largest kp (9.80 min-‘); Conversely, all comparisons of the & estab- with diethyl, di-n-propyl, and di-n-butyl p-mlished for trout brain AChE were significantly trophenyl phosphate possessing intermediate different from the corresponding values ob- kp values (0.3 11,0.064, and 0.270 mm-‘, retained for the enzyme from rats or chickens. spectively); while diisopropyl pnitrophenyl Data describing the phosphorylation rate phosphate had a kp of only 0.0053 min-‘. All constants (k,) are provided for all inhibitors comparisons of kp among inhibitors of trout and species studied in Table 5. For both AChE, except for diethyl versus di-n-butyl p chicken and rat AChE, the kp values for all nitrophenyl phosphate, were statistically sigfive inhibitors ranged between 1.08 and 2.7 1 nificant. The kp for each inhibitor of trout min-’ and did not differ significantly. In di- AChE was significantly different from the rect contrast, the kp values for AChE from corresponding k,, for either the chicken or the trout brain varied among the five organorat enzyme.
TABLE 5 PHOSPHORYLATIONRATECONSTANTS (k,) FORTHEINHIBITIONOFACETYLCHOLINESTERASEBYASERIES OFDIALKYL-~NITROPHENYLPHOSPHATES Dialkyl-p-nitrophenyl
Species Hen (37°C) Rat (37°C) Trout (I 1°C)
phosphate k,, (min)) ’
Methyl
Ethyl
n-Propy!
n-Buty!
Isopropyl
2.04 + 0.08” 1.54t0.15 9.80 f l.81b
2.48 + 0.31 2.13f0.60 0.3 1 2 0.02b
2.23 ST0.08 1.12*0.08 0.064 -+ 0.003 * _----_---_
2.7 1 + 0.78 1.08 kO.12 0.270 + 0.042’
1.75 + 0.35 1.30 + 0.06 0.0053 +- 0.00206
LIValues represent the means f SE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p c 0.05). b Significantly different from the corresponding value for the other two species (p c 0.05).
INHIBITION
OF BRAIN ACETYLCHOLlNESTERASE
DISCUSSION The linearity of the kinetic plots is consistent both with a first-order process with respect to inhibitor concentration and with the existence of a single major isozyme of AChE (Chemnitius, 1982). The preferential hydrolysis of acetylthiocholine by rat, chicken, and trout brain AChE as well as the relative velocities using propionyl- (60-75% for rats and hens, and 10% for trout) and butyryl- (10% for rats and hens, and
AND
APPLIED
104,246-258
PHARMACOLOGY
(
1990)
Molecular Determinants of the Species-Selective inhibition of Brain Acetylcholinesterase JON R. KEMP Department
of Pharmacology,
University
Received
August
AND KENDALL of Minnesota,
School
18, 1989; accepted
B. WALLACE’ of Medicine,
March
Duluth,
Minnesota
55812
6, 1990
Molecular Determinants of the Species-Selective Inhibition of Brain Acetylcholinesterase. KEMP, J. R.. AND WALLACE, K. B. (1990). Toxicol. Appl. Pharmacol. 104, 246-258. The objective of this investigation was to distinguish which of the catalytic features of enzyme action is principally responsible for conferring the observed insensitivity of trout brain acetylcholinesterase (AChE, EC 3.1.1.7) to in vitro inhibition by organophosphates. The experimental design consisted of comparing the kinetic constants for the hydrolysis of a series of acylthiocholine substrates as well as the inhibition constants for a homologous series of dialkyl pnitrophenyl phosphates among AChE from rats, hens, and trout. Chicken and rat brain AChE failed to distinguish between acetyl- and propionylthiocholine as inferred from the comparable Michaelis-Menten constants (K,), whereas trout brain AChE exhibited a much higher affinity for acetylthiocholine than for either of the two larger analogs. Diethyl pnitrophenyl phosphate was the most potent inhibitor toward chicken and rat brain AChE, whereas the IC50 for the trout enzyme increased progressively between dimethyl and di-n-propyl pnitrophenyl phosphate. The kinetic constants revealed that a significant determinant of inhibitor potency in the chicken and rat is steric exclusion as reflected by changes in the dissociation constant (&) which paralleled the changes in IC50 and k,. Conversely, Kdwas 120-to 1450-fold higher and did not vary significantly for trout brain AChE. Instead, the phosphorylation rate constant (I$,) for trout brain AChE decreased with progressive methylene substitutions. The kinetic data suggest that trout brain AChE not only possessesless steric tolerance, but also has a weaker nucleophile at the esteratic subsite, both ofwhich may be important factors in conferring the observed insensitivity of trout to acute organophosphate intoxication. o 1990 Academic PKSS, IW.
The relative insensitivity of most species of fish to acute intoxication by organophosphorus compounds such as Parathion (O,O-diethyl p-nitrophenyl phosphorothionate) is well established; the acute intraperitoneal LD50 for the biologically active oxygen analog, Paraoxon, varies from 1.1 and 2.6 pmol/ kg in hens and rats, respectively, to 27.3 rmol/kg in sunfish (Benke et al., 1974; Chattopadhyay et al., 1986). Attempts to elucidate the principal factor which confers this resistance to fish have focussed on species-related differences in pharmacokinetic variables, in’ To whom correspondence should be addressed. 0041-008X/90
$3.00
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form resewed.
eluding differences in the absorption, distribution and elimination of an acute dose, and species-related differences in metabolic transformation (activation or detoxification) of the organophosphates. Despite obvious anatomical and physiological differences, it has been concluded that the species-selective toxicity of organophosphates reflects more than simply subtle differences in kinetic disposition (Benke et al., 1974; Hodson, 1985). Studies designed to assessthe significance of species-related differences in metabolic activation or detoxification as major determinants of species-selectivity are complicated by the fact that the steady-state concentration 246
INHIBITION
OF
BRAIN
of the biologically active intermediate is a function of the relative activities of a group of nonspecific esterases acting both in series and in parallel to inactivate the reactive oxygen analog formed by the cytochrome P450mediated oxidative desulfuration of Parathion. Consequently, despite large differences among species in the activities of the mixed function oxidases and the nonspecific esterases responsible for the activation and detoxification of organophosphonates, respectively, the significance of metabolic transformation as a major determinant of species selectivity in organophosphorus poisoning remains questionable (Potter and O’Brien, 1964; Murphy, 1966; Hitchcock and Murphy, 197 1; Benke et al., 1974; Wang and Murphy, 1982b; Wallace and Dargan, 1987). Recent evidence suggests that the insensitivity of fish to organophosphate toxicity, rather than being a pharmacokinetic phenomenon, may reflect pharmacodynamic differences between species, manifested at the level of inhibition of acetylcholinesterase (AChE) by the active intermediate. The IC50 values for the inhibition of brain or plasma cholinesterase by Paraoxon vary over a 2000fold range between species, with chickens being the most sensitive, rats intermediate, and fish the least sensitive species (Dauterman and O’Brien, 1964; Murphy et al., 1968; Ecobichon and Comeau, 1973; Benke et al., 1974; Andersen et al., 1977; Wang and Murphy, 1982a,b; Johnson and Wallace, 1987). These differences are consistent with the differences in the in vivo sensitivity among species. Several groups of investigators (Becker et al., 1964; Darlington et al., 1971; Lotti and Johnson, 1978) reported direct correlations between the in vitro IC50 for AChE and the acute LD50 for a wide array of organophosphates in rodents, insects, and chickens. Accordingly, the sensitivity of AChE to inhibition in vitro appears to be a principal determinant of the species-selective toxicity of anticholinesterase agents and may be employed as a convenient screen for interspecies comparisons.
ACETYLCHOLINESTERASE
247
On a molecular basis, these differences in enzyme sensitivity may reflect any one or combination of several catalytic properties of AChE; (1) the dimensions of the active site, (2) the strength of the nucleophilic center responsible for interacting with the phosphoryl group, (3) the influence of a possible hydrophobic subsite, and (4) the distance separating the esteratic and anionic subsites. Structure-activity studies have revealed isolated differences in some of these properties between selected enzymes. Zahavi and coworkers (197 1) suggested that resistance in certain strains of mites to Malaoxon is conferred by a sterically smaller esteratic site, thereby excluding the large phosphoryl esters. Alternatively, species-related differences in sensitivity to inhibition may reflect differences in the proximity of the anionic site to the esteratic site (O’Brien, 1963; Hollingworth et al., 1967). Moss and Famey (1978) reported that although fish AChE possesses both anionic and esteratic subsites, the positive cooperativity between these regions is less than that exhibited by mammalian AChE, which is consistent with the resistance of the fish enzyme to inhibition by organophosphates. The objective of the present investigation was to elaborate on the observed differences between rats, chickens, and rainbow trout in in vitro sensitivity of AChE in an attempt to identify the principal catalytic feature of enzyme action responsible for the large idiosyncrasies between species. The investigation was limited to assessing speciesrelated differences in only those variables associated directly with the esteratic site of AChE (steric dimensions, nucleophilic strength, hydrophobic subsite). The paper expands beyond previous reports (Wang and Murphy, 1982a) to include a larger series of five structurally related analogs of Paraoxon, allowing for more definitive deductions regarding molecular and catalytic distinctions between different enzymes. Furthermore, the reactions were conducted at the respective physiological temperatures and, in order to eliminate possible confounding influences of
248
KEMP AND WALLACE
isozyme heterogeneity and allosterism between species, a homogeneous preparation of detergent-solubilized membrane-associated AChE was employed. Finally, the investigation employs a novel kinetic method (Hart and O’Brien, 197 3) to circumvent important restrictions imposed by the extreme rapidity of the inhibition reaction. METHODS Triton X- 100, acetyl-, propionyl-, and butyrylthiocholine iodide were purchased from Sigma Chemical Co. (St. Louis, MO). All other chemicals were reagent grade as supplied by Aldrich Chemical Co. (Milwaukee, WI). Dimethyl, diethyl, di-n-propyl, diisopropyl, and di-n-butyl chlorophosphates were synthesized by a modification of the method of Fiszer and Michalski ( 1952) and M&or et al. ( 1956). Briefly, three equivalents of the appropriate alcohol were added dropwise to one equivalent of cold (5-1o’C) phosphorus trichloride in hexane; an icebath was used to maintain the reaction temperature below 10°C. Immediately after the addition of alcohol, one equivalent of sulfuryl chloride was introduced dropwise to the newly formed trialkylphosphite. The hexane was removed under reduced pressure (water pump aspiration) and the dialkyl chlorophosphate purified by vacuum distillation. The various dialkyl pnitrophenyl phosphates were then prepared from the respective dialkyl chlorophosphates and sodium pnitrophenoxide by a modification of the methods of Fletcher et al. ( 1950) Schrader ( 1952) and DeRoos and Toet (1958). Specifically, sodium pnitrophenoxide was suspended in ethanol (I .5 M) and neat dialkyl chlorophosphate was added dropwise, producing a faint exothermic reaction. After the addition ofthe dialkyl chlorophosphate, the mixture was heated for 90 min at 87°C under a reflux apparatus. The reaction was cooled to room temperature then filtered to remove sodium chloride. Ethanol was removed by rotoevaporation under reduced pressure (water pump aspirator) and the resulting product was dissolved in methylene chloride and washed with a 5% aqueous solution of sodium carbonate to remove unreacted sodium pnitrophenoxide. The methylene chloride was dried with anhydrous magnesium sulfate, filtered, and then removed by rotoevaporation. The resulting dialkyl pnitrophenyl phosphate was then heated under reduced pressure (0.2 mm Hg) to remove traces of unreacted dialkyl chlorophosphate (Fletcher et al., 1950). Purity of the isolated dialkyl pnitrophenyl phosphates were greater than 95% as determined by GC-FID, and confirmed by GC-MS and HPLC-uv. The only detectable impurity was the corresponding chlorophosphate which was found to be inac-
tive in inhibiting AChE. Every 2 months neat dialkyl p nitrophenyl phosphates were dissolved in methylene chloride and washed with 5% sodium carbonate to remove hydrolysis products (pnitrophenol). From this oreanoohosphate reserve. stock solutions were orepared in absolute ethanol (stable for 1 week at -20”C)br in buffer containing 10% ethanol and 20% Triton X-100 (stable for 8 hr at -20°C). Final concentrations of ethanol and Triton X-100 in the reaction medium did not exceed 1 and 2.25%, respectively, neither of which affected the activity of AChE. White Leghorn hens (2-2.5 years old: 1300-2000 g) were supplied by Morterud’s Eggs, Inc. (Duluth, MN), and were killed by cervical dislocation. Male SpragueDawley rats (175-250 g) were purchased from Harlan, Inc. (Indianapolis, IN), and were housed under climatecontrolled conditions on a I2-hr light cycle and provided food and water ad libitum. All rats were fasted for at least 12 hr prior to their use and were killed by decapitation. Unsexed rainbow trout (750-1250 g) were purchased from Cedar Bend Trout Farm (Scandia, MN) and maintained at the U.S. EPA Environmental Research Laboratory (Duluth, MN) under regulated conditions and killed by a blow to the head. Membrane-associated AChE from brain homogenates was solubilized in detergent to recover the majority of the enzyme. Solubilization of 100,OOOgpellet of brain homogenates with Triton X-100 yields 90-95% of the total enzyme activity which, for rats and chickens, is almost exclusively in the form ofa globular 10 S or I 1 S tetramer of AChE (Rakonczay et al., 198 1; Vallette et al., 1983; Rotundo, 1984). Such a homogeneous enzyme preparation allows for comparisons of AChE among species without concern for possible differences in the composition of isozymes present and eliminates any possible differences in allosteric regulation of the enzyme due to species-related differences in membrane lipid composition (Ott, 1985; Stieger and B&beck. 1985). For each experiment, brain homogenates from the corresponding species were prepared from a single animal, with the exception of rainbow trout (two brains were pooled for each experiment). The brain was quickly transferred to ice-cold pH 8.0 buffer [IO0 mM sodium phosphate, 400 IIIM sodium chloride, and 3 mM sodium ethylenediaminetetraacetic acid (EDTA)] and weighed. For hens and rats a 20% (w/v) brain homogenate was prepared in a Wheaton glass homogenizer with Teflon pestel; a 12.5% brain homogenate was prepared from trout brain. The homogenate was centrifuged at lOO,OOOgfor 60 min (4’C). The resulting supematant was discarded and the pellet resuspended in an equal volume of pH 8.0 buffer containing 100 mM sodium phosphate, 3 mM sodium EDTA, and 0.5% Triton X-100 and recentrifuged at 100,OOOgfor 60 min. Sodium chloride was dissolved in the resulting supematant to a final concentration of 400 mrvr and the solution used that same day for all kinetic analyses and IC50 determinations.
INHIBITION
Cholinesterase assays were performed spectrophotometrically utilizing the method of Ellman et al. (196 1) as previously described (Johnson and Wallace, 1987). Acetyl-, propionyl-, and butyrylthiocholines have proven to be representative substrates for a variety ofAChE sources (Andersen and Mikalsen, 1978), with Michaelis-Menten constants (K,,,) corresponding to the oxygen analogs. Protein content for each enzyme preparation was determined by the technique described by Lowry et al. (195 1) using bovine serum albumin as standard. To determine K,,, and I’,,,,, for the entire series of substrates within each species, an excess of 5,5’-dithiobis(2nitrobenzoic acid) (DTNB) was added to AChE in buffer. The reactions were initiated by adding small aliquots of varying concentrations of the respective substrate to yield a final volume of 2 ml. Concentrations of substrate were within a range of one-fifth to five times the estimated Km and the amount of AChE included in the assay was adjusted to yield a maximum rate of0.05 absorbance unit per minute (A&min) at the highest substrate concentration employed. Corresponding blanks lacking AChE were subtracted to yield the enzymatic rate. Kinetic incubations for rat and hen brain AChE were conducted at 37°C. The reactions for the trout enzyme were performed at 11°C which is the temperature at which the fish were acclimated. Finally, determinations of rat AChE kinetics were repeated at I 1“C in order to compare the values with those for trout. The maximum velocity of substrate hydrolysis (V,,,,,) and the Michaelis-Menten constant (K,,,) were estimated by the double-reciprocal method of Lineweaver and Burk (1934), the line being calculated by a least-squares regression. ICSO values were determined as described in detail by Johnson and Wallace (1987), wherein the enzyme was preincubated for 30 min with differing concentrations of the inhibitor prior to measuring residual AChE activity. For all reactions, the enzyme was diluted such that in the absence of inhibitor, the rate of acetylthiocholine hydrolysis was approximately 0.1 A,,,/min. The data were analyzed by linear regression of the natural log (In) of inhibitor concentration versus percentage remaining enzyme activity following incubation for 30 min in the presence of inhibitor. A value of 50% was substituted into the resulting equation to solve for ICSO. The bimolecular inhibition constants for the organophosphates were estimated in one of two ways. The first method was employed to determine the inhibitor kinetic constants for rainbow trout brain AChE as described by Johnson and Wallace (1987),
The bimolecular inhibition constant, k,, is equal to k,/ Kd. Ten microliters of the appropriate concentration of inhibitor was incubated with 90 ~1 of the AChE preparation (which at Time zero yielded anA.&min ofO.l). The reactions were stopped at the indicated times by adding 1.5 ml of 1 mM acetylthiocholine- 1 mM DTNB and the A&min was recorded immediately over 5 min. Addition of excess substrate was effective at preventing the continued inhibition of AChE and provided reliable kinetic measurements. This procedure was repeated with six different concentrations of each dialkyl p-nitrophenyl phosphate. Although the above kinetic method worked well for the trout enzyme, it was unsatisfactory for rat or hen AChE. With these species, the reaction rates were so rapid that it was virtually impossible to employ inhibitor concentrations between one-fifth and five times the Kd as required. According to the above equation, when [I] B Kd, kam becomes a direct function of [I]; k., = k, [I]/ Kd = k;[I]. As expected, double-reciprocal plots of the data intersected very near the origin yielding spuriously high estimates of both k, and Kd. Accordingly, it was necessary to modify the kinetic method for rat and hen AChE in order to approach the Kd for the enzyme and to obtain accurate estimates of the kinetic constants. This was accomplished by including a known concentration of acetylthiocholine in the reaction mixture to compete with, and thus retard, the inhibition of rat or hen AChE by the organophosphates (Hart and O’Brien, 1973). Ten microliters of the appropriate concentration of inhibitor was added to 1.5 ml of buffer containing 1 mM DTNB and 600 or 400 pM acetylthiocholine for the chicken or rat, respectively. This solution was equilibrated to 37°C then 90 ~1 of the corresponding AChE preparation (added at a concentration which in the absence of inhibitor yielded a rate of0. I Aalz/min) was added and the progressive decrease in the rate of substrate hydrolysis was recorded over a IO-min period. This was repeated with at least six concentrations of each dialkyl pnitrophenyl phosphate. The reaction can be represented by [E] + [I] 2 [E.] 2 [E-I] where,
kl k-*
It
[E] + [I] 2 [EI] 1 [EI’]
+ 14)
where k.pp is the observed velocity of inhibition; kp, the first-order phosphorylation rate constant; and Kd, the dissociation constant for the enzyme-inhibitor complex.
Kds = Kd(l
+ [S]/K,)
kz
tES1 k,
[El + [PI li,,,, = WI/(&
249
OF BRAIN ACETYLCHOLINESTERASE
therefore,
k.,, =
kp’[ll Kis + [II
1 Kds 1 and G=k,‘cI]+k,’
1
where Kds is the apparent Kd for the enzyme-inhibitor complex determined in the presence of substrate, KS is the Michaelis-Menten constant (k-,/k,) determined for acetylthiocholine in the species under study, and [S] is the concentration of acetylthiocholine added to the reaction.
KEMP AND WALLACE
250
TABLE I
Acetylthiocholine V,, (nmol/min .mg protein) Hen (37°C) Rat (37°C) Rat(ll”C) Trout ( 11“C)
350.1 289.5 159.0 148.9
Km (FM) Hen (37°C) Rat (37°C) Rat(ll”C) Trout ( 11‘C)
105.2 * 77.5 * 99.1 * 197.2 i
f 44.4” + 6.7 f 3.7 f 11.1 7.0* 3.3* 4.0* 51.3’
Propionylthiocholine 266.8 + 185.8 + 86.2 f 15.1 f
28.7h 5.3* 2.7 b 1.0*
117.7+4.9* 83.3 f 4.2’ 92.3 f 9.3b 2078.5 2 203.2b
Butyrylthiocholine 16.7 f 1.2’ 8.7 + 0.5’ 2.4 kO.lb n.d.’ 23.0 k 1.2* 337.8 + 40.5’ 183.3 f 22.8* n.d.’
’ Values represent the means ? SE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p < 0.05). * Significantly different from the corresponding values for all other enzyme preparations (p < 0.05). ’ n.d. signifies that the reaction proceeded too slow to allow for reliable estimation ofthe kinetic constants.
In this scheme, as with the method used for trout brain AChE, the k,, for each inhibitor concentration was determined by regressing residual enzyme activity @[El,/ [El,) against time (Fig. I): then the double-reciprocal plot was formed with the inverse of each inhibitor concentration regressed against the inverse of -kapp (Fig. 2). The reciprocal of the y-intercept is equal to the k, and the negative reciprocal of the x-intercept is equal to I&; however, a correction factor must be intejected to determine the actual Kd and thus the k, (Kd = K&/( I + [S]/K,) and k, = kp/Kd). Statistical analyses of the log transformed data were accomplished by analysis of variance and the individual means compared by the method of Scheffefor all possible comparisons both within and across species (Sokal and Rohlf, I98 I). A probability of p < 0.05 was used as the criterion for statistical significance.
RESULTS The kinetic constants (V,,, and K,,J describing the hydrolysis of the three acylthiocholine substrates by AChE from each of the three species (including the rat at 11°C) are presented in Table 1. Within each species the V,,,, for the individual thiocholine substrates were significantly different from each other; however, in the trout it was impossible to measure either V,,,,, or K,,, for butyrylthiocholine as the limits of solubility for this com-
pound were reached before any appreciable rate of hydrolysis could be measured. For each species studied, acetylthiocholine was hydrolyzed at the fastest rate, followed by propionyl- and then butyrylthiocholine; however, there were large differences in the degree of substrate preference among species (Table 1). For rat and chicken AChE, the V,,,,, for propionylthiocholine was between 64 and 76% of that for acetylthiocholine. In contrast, the trout enzyme exhibited a more profound preference for the smaller analog, hydrolyzing propionylthiocholine at only 10% of the maximum rate for acetylthiocholine. This idiosyncrasy of the trout brain enzyme was not due to the different temperature at which the reaction was conducted since the rat enzyme incubated at 11 “C exhibited a substrate specificity comparable to that observed at 37°C. Comparing substrate specificity among species, there was no significant difference between the maximum rate of acetylthiocholine hydrolysis for the hen or rat AChE or between trout and rat when both experiments are performed at 11°C. For any given substrate, all other possible comparisons across species were significantly different.
INHIBITION
251
OF BRAIN ACETYLCHOLINESTERASE TABLE 2
IC~OFORTHEINHIBITIONOFBRAINACETYLCHOLINESTERASEBYVARIOUS DIALKYL-p-NITROPHEN~LPHOSPHATES Dialkyl-p-nitrophenyl Methyl Species Hen (37°C) Rat (37°C)
Rat(ll”C) Trout (11°C)
0.037 + 0.003” 0.029 f 0.00 1 0.108~0.005b 3.3 1 + 0.20”
Ethyl
0.028 0.026 0.109 70.04
-+ 0.000
+ 0.001 k o.002b * 4.95b
phosphate IC50
n-Propyl
(pM)
n-Butyl
Isopropyl
0.336 -+0.013’ 0.228 f 0.004b 3.280 t 0. I 16”
0.762 _+ 0.053 0.453 + 0.008 8.400 f 0.740’
3.30+0.10b 2.12 k 0.056 35.64 + 1.76’
963.0-t
763.6 + 142.7’
12,779 rt 1002’
137.1b
’ Values represent the means -CSE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p < 0.05). b Significantly different from the corresponding value for all other species (p < 0.05).
The chicken and rat enzyme failed to dis- species studied. Repeating the IC50 determinations for rat AChE at 11°C yielded a trend tinguish between acetyl- and propionylthiocholine as indicated by the comparable val- similar to that observed at 37°C; IC50 inues for K, for the two substrates (Table I), creased with increasing methylene substituwhereas the trout was far (IO-fold) more se- tion. The individual IC50 values for each inlective for acetylthiocholine than for the pro- hibitor were, however, greater at the lower pionyl analog. For rat and trout AChE, buty- temperature. For the chicken and rat AChE, rylthiocholine yielded a significantly greater all possible comparisons among the different K,,, compared to the two smaller acylthiochoinhibitors were significantly different (except lines; however, the K,,, for the butyryl analog when the rat enzyme was incubated at 11°C in the chicken was 4.8 times lower than the dimethyl and diethyl g-nitrophenyl phosphate were not statistically different). K,,, for either the acetyl or propionyl derivaThe rainbow trout brain AChE was most tive. All possible comparisons of K,,, across by dimethyl pnitrospecies for a given acylthiocholine were sig- sensitive to inhibition phenyl phosphate (Table 2). However, the nificantly different. Diethyl p-nitrophenyl phosphate was the IC50 was approximately 1OO-fold higher than most potent inhibitor for both chicken and that for rat or chicken (IC50 = 3 PM). Unlike rat brain AChE, with an IC50 of 28 and 26 the rat or chicken enzyme, inhibitor potency toward trout AChE decreased 23-fold as the nM, respectively (Table 2). Dimethyl p-nitrophenyl phosphate was almost as potent in the result of including a single additional methyterrestrial species, displaying an IC50 of 37 lene group (diethyi analog IC50 = 70 PM). nM for chicken AChE and 29 nM for rat This trend continued in a stepwise fashion for the remaining inhibitors; however, di-n-butyl AChE. For rat and chicken AChE, a dramatic transition in potency occurred between di- pnitrophenyl phosphate (IC50 = 764 PM) ethyl and di-n-propyl pnitrophenyl phos- was found to be equipotent with di-n-propyl phate, the IC50 increasing 12- and 9-fold, p-nitrophenyl phosphate (IC50 =963 PM). respectively. Potency continued to de- All other possible inhibitor IC50 comparicrease with increasing methylene substitution sons within the trout were significantly through di-n-butyl p-nitrophenyl phosphate; different from each other. When comparing diisopropyl p-nitrophenyl phosphate being each inhibitor across species, there was no the least potent inhibitor of the series for all difference in the IC50 for dimethyl, diethyl,
252
KEMP AND WALLACE Minutes
-0.5 0 iii v iii C
-1.5
-2.5
FIG. 1. Log-linear progression plot for the inhibition of rat brain AChE by increasing concentrations of diethylp-nitrophenyl phosphate at 37°C. The reactions were performed in the presence of known concentrations of acetylthiocholine, functioning as a competitive substrate to retard the rate of inhibition of AChE as described in the text. The ordinate represents the fraction of enzyme activity remaining at time f. Each inhibitor concentration is depicted by a single line, the slope of which equals the respective -Y&,~. The results are those of a typical experiment.
The bimolecular inhibition constants (k;) for the series of dialkyl p-nitrophenyl phosphates across species are summarized in Table 3. Describing the overall reaction between inhibitor and enzyme, kj is a measure of inhibitor potency and can be related mathematically to the IC50 (Johnson and Wallace, 1987). Whereas a low IC50 indicates a potent inhibitor, the inverse is true for k;; the larger the k, the more potent the inhibitor. As with the IC50 data, dimethyl and diethyl p-nitrophenyl phosphate were the most potent inhibitors toward chicken and rat brain AChE as reflected by ki. The comparable k, values for these two inhibitors suggest that neither the rat nor the chicken AChE distinguished between dimethyl and diethyl p-nitrophenyl phosphate. However, there was a stepwise decrease in k, from diethyl through diisopropyl p-nitrophenyl phosphate for these two species of AChE. Di-n-butyl p-nitrophenyl phosphate was 14-fold less potent than diethyl p-nitrophenyl phosphate toward chicken AChE and 12-fold less potent toward rat brain AChE. Similar to that observed for the IC50 data, diisopropyl p-nitrophenyl phosphate was the least potent inhibitor toward all species studied, ki being 72- and 63-fold lower
and di-n-butyl p-nitrophenyl phosphate between chicken and rat AChE when both were determined at 37°C. All other comparisons between species were significantly different. Figures 1 and 2 depict typical kinetic plots for the inhibition of rat brain AChE by diJ ethyl p-nitrophenyl phosphate employing the 2.0 modified kinetic model wherein acetylthio1.6 choline was present during the course of enzyme inhibition, Each reaction was allowed to proceed until approximately 90% of the enzyme activity was inhibited (Fig. 1). Least.i/ squares regression of the progression plots yielded lines of increasing slope (-k,,,) corre0.4 sponding to increasing inhibitor concentrai/l 0.04 .‘# ! . , . I , , tions. In all cases for each species and all in-0.2 -0.1 0.0 0.1 0.2 0.3 0.4 hibitors, r2 3 0.95. The inhibition constants 1Ml (PM) were derived from double-reciprocal plots of kappregressed against inhibitor concentration FIG. 2. Double-reciprocal plot for the inhibition of rat (Fig. 2). Once again, r2 3 0.95 for all cases. brain AChE by diethyl-pnitrophenyl phosphate at 37°C. The values for Knappwere derived from Fig. 1. LeastThe slope of this line is equal to l/ki, intersquares linear regression yielded a line which intersected secting the ordinate at l/k, and the abscissa the ordinate at (&J’ and the abscissa at -(&)-I, where at l/Kds. The dissociation constant (Kd) is Kdr = KAI + [S]/K,). In this example, [S]/K, was 5 and calculated by dividing K~s by ( 1 + [S]/K& thus K*= 6%.
INHIBITION
253
OF BRAIN ACETYLCHOLINESTERASE TABLE 3
BIMOLECULARRATECONSTANTS(~~)FORTHEINHIBITIONOFACE~LCHOLINESTERASEBYASERIES OFDIALKYL-p-NITROPHENYLPHOSPHATES Dialkyl-pnitrophenyl Methyl Species Hen (37’C) Rat (37°C) Trout ( 11“C)
828k34".b 1238+59' 5.37 IL 0.08'
Ethyl
n-Propyl
1138+44
18324
1463+20
282k12
0.375
+o.030b
phosphate k, (mM . min)-’
0.042+0.015b
n-Butyl
79*56 1242
Isopropyl
16k3
lib
0.027 +O.OOlb
23&l 0.0013
+ 0.0001~
0 Values represent the means -t SE ofthree separate enzyme preparations. Means underscored by the same line are not significantly different (p < 0.05). b Significantly different from the corresponding value for the other two species (p < 0.05).
than that for diethyl pnitrophenyl phosphate in chicken and rat, respectively. For rat and chicken AChE, all possible comparisons of ki for the individual inhibitors were significantly different, with the only exception being dimethyl and diethyl p-nitrophenyl phosphate. The kj data indicate that the entire series of dialkyl p-nitrophenyl phosphates were less potent toward the trout AChE than toward the other species (Table 3). Trout brain AChE exhibited the greatest sensitivity to dimethyl pnitrophenyl phosphate with a ki of 5.4 (mM. min))‘. From dimethyl pnitrophenyl phosphate there was a progressive decrease in k; with increasing methylene substitution: the diethyl analog having a ki of 0.38 (mM. min))‘, di-n-propyl and di-n-butyl possessing a k, of 0.042 and 0.027 (mM.min)-‘, respectively, while diisopropyl p-nitrophenyl phosphate had the lowest ki- of 0.0013 (mM. min)-‘. Other than di-n-propyl and di-n-butyl p-nitrophenyl phosphate, the individual ki values for the inhibitors against trout AChE were all statistically different from one another. An analysis across species for each inhibitor showed no significant difference between the ki observed in the chicken and rat for diethyl, di-n-propyl, or diisopropyl p-nitrophenyl phosphate; all other possible comparisons for each inhibitor across all three species were statistically different.
Table 4 illustrates the dissociation constants (&) for all inhibitors in the three species studied. Brain AChE from chicken and rat possessed the lowest Kd for dimethyl and diethyl pnitrophenyl phosphate, the values not being statistically different. There was, however, a two- to sixfold increase in the Kd between diethyl and di-n-propyl p-nitrophenyl phosphate in the chicken and rat and this trend continued in a stepwise fashion with the Kd increasing with successive methylene substitutions through di-n-butyl pnitrophenyl phosphate for both species of AChE. Both the chicken and rat enzyme failed to distinguish between the Kd for the dimethyl and diethyl analogs; similarly, no difference was observed when comparing the Kd of chicken AChE for di-n-propyl and di-n-butyl p-nitrophenyl phosphate. Rat brain AChE also did not distinguish between the Kd values for diethyl and di-n-propyl p-nitrophenyl phosphate or between the di-n-propyl and the din-butyl analog. All other comparisons of Kd for the inhibitors of rat or chicken AChE were statistically significant. Rainbow trout brain AChE possessed Kd values which were two to three orders of magnitude larger than the corresponding Kd for the rat or chicken enzyme (Table 4). However, in contrast to rat and chicken AChE, Kd for the trout enzyme did not vary significantly between the five inhibitors examined. When comparing across spe-
254
KEMP AND WALLACE TABLE 4 DISS~CIATIONCONSTANTS( Kd) FORTHEINHIB~TIONOFACETYLCHOLINESTERASEBYASERIES OFDIALKYL-p-NITROPHENYLPHOSPHATES Dialkyl-p-nitrophenyl
Species Hen (37°C) Rat (37°C) Trout(!
1°C)
phosphate Kd (PM)
Methyl
Ethyl
n-Propy!
n-Buty!
Isopropyl
2.48 rt: 0.20” 1.26 +0.19 1830t_360*
2.21 2 0.37 1.62 t 0.67 851 +_ t39b
12.15 20.54 4.02 f 0.4 1
35.8 -t 12.6 9.03 +_ 1.58
121.1 + 35.6 56.24 + 4.43
1924 +- 547 b
10363 t 14761~
14852 t- t2668b
y Value represent the means f SE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p -c 0.05). b Significantly different from the corresponding value for the other two species (p < 0.05).
ties, the Kd for the entire series of inhibitors phosphates: dimethyl p-nitrophenyl phosdid not differ between rat and chicken AChE. phate possessed the largest kp (9.80 min-‘); Conversely, all comparisons of the & estab- with diethyl, di-n-propyl, and di-n-butyl p-mlished for trout brain AChE were significantly trophenyl phosphate possessing intermediate different from the corresponding values ob- kp values (0.3 11,0.064, and 0.270 mm-‘, retained for the enzyme from rats or chickens. spectively); while diisopropyl pnitrophenyl Data describing the phosphorylation rate phosphate had a kp of only 0.0053 min-‘. All constants (k,) are provided for all inhibitors comparisons of kp among inhibitors of trout and species studied in Table 5. For both AChE, except for diethyl versus di-n-butyl p chicken and rat AChE, the kp values for all nitrophenyl phosphate, were statistically sigfive inhibitors ranged between 1.08 and 2.7 1 nificant. The kp for each inhibitor of trout min-’ and did not differ significantly. In di- AChE was significantly different from the rect contrast, the kp values for AChE from corresponding k,, for either the chicken or the trout brain varied among the five organorat enzyme.
TABLE 5 PHOSPHORYLATIONRATECONSTANTS (k,) FORTHEINHIBITIONOFACETYLCHOLINESTERASEBYASERIES OFDIALKYL-~NITROPHENYLPHOSPHATES Dialkyl-p-nitrophenyl
Species Hen (37°C) Rat (37°C) Trout (I 1°C)
phosphate k,, (min)) ’
Methyl
Ethyl
n-Propy!
n-Buty!
Isopropyl
2.04 + 0.08” 1.54t0.15 9.80 f l.81b
2.48 + 0.31 2.13f0.60 0.3 1 2 0.02b
2.23 ST0.08 1.12*0.08 0.064 -+ 0.003 * _----_---_
2.7 1 + 0.78 1.08 kO.12 0.270 + 0.042’
1.75 + 0.35 1.30 + 0.06 0.0053 +- 0.00206
LIValues represent the means f SE of three separate enzyme preparations. Means underscored by the same line are not significantly different (p c 0.05). b Significantly different from the corresponding value for the other two species (p c 0.05).
INHIBITION
OF BRAIN ACETYLCHOLlNESTERASE
DISCUSSION The linearity of the kinetic plots is consistent both with a first-order process with respect to inhibitor concentration and with the existence of a single major isozyme of AChE (Chemnitius, 1982). The preferential hydrolysis of acetylthiocholine by rat, chicken, and trout brain AChE as well as the relative velocities using propionyl- (60-75% for rats and hens, and 10% for trout) and butyryl- (10% for rats and hens, and
