162
Chem. Res. Toxicol. 1991, 4 , 162-167
Recognition of Tetramethyienedisuifotetramine and Related Sulfamides by the Brain GABA-Gated Chloride Channel and a Cyclodiene-Sensitive Monoclonal Antibody Thomas Esser,? Alexander E. Karu,* Robert F. Toia,*pt and John E. Casidat Pesticide Chemistry and Toxicology Laboratory, Department of Entomological Sciences, and Hybridoma Facility, College of Natural Resources, University of California, Berkeley, California 94720 Received October 2, 1990
Aldrin and many other cyclodiene and polychlorocycloalkane insecticides interact with both the [35S]-tert-butylbicyclopho~phorothionate ([Y3]TBPS) binding site of the mammalian brain y-aminobutyric acid (GABA) gated chloride channel and several cyclodiene monoclonal antibodies (MAbs) a t concentrations ranging from 0.06 to 8.7 pM. A survey of other classes of GABAA receptor antagonists (including picrotoxinin and several trioxabicyclooctanes) for possible interactions with the cyclodiene MAbs revealed only one potent inhibitor, the heteroadamantane tetramethylenedisulfotetramine(TETS) [mouse intraperitoneal LDm 0.24 mg/kg; TBPS binding site ICw 0.5 pM as a competitive inhibitor (Scatchard analysis); cyclodiene MAb ICw 3 pM]. These findings prompted comparative studies on the structure-activity relationships of other sulfamides as they apply to both the ligand-nerve and ligand-MAb interactions. TETS is active on only one (MAb 8 H l l ) of four cyclodiene MAbs. Several hetero(homo)adamantanes were synthesized and compared with TETS for neurotoxicity and recognition by the TETS-sensitive cyclodiene MAb. The toxicity to mice and/or houseflies decreases in the following order: TETS >> the heterotetracyclic compound hexamethylenetrisulfohexamine (HEXS) and two TETS analogues in which one sulfamide group is replaced with o-phenylenediamine or 1,l-dimethyl-1,2-diaminoethane>> seven other hetero(hom0)adamantanes. The TETS-sensitive cyclodiene MAb recognizes HEXS (ICm 0.4 pM) and, to a lesser extent, two related sulfamides. However, the cross-reactivity noted for the cyclodiene insecticides and TETS relative to the GABA-gated chloride channel (inhibition of TBPS binding) and the cyclodiene MAb does not extend to several TETS analogues including HEXS. Moreover, many compounds active at the TBPS binding site are not recognized by the antibody, e.g., picrotoxinin and the trioxabicyclooctanes. These unusual cross-reactivity patterns suggest that the sulfamide moiety may be equated with the planar Cl-C-CCl,-C-Cl group of the cyclodienes as a minimum requirement for ligand recognition.
Introductlon Cyclodienes and other polychlorocycloakane insecticides act as GABAA receptor antagonists and thereby inhibit GABA-induced chloride flux (1-4). Their recognition site in brain membranes is conveniently assayed as competitive inhibition of [36S]-tert-butylbicyclophosphorothionate ( [35S]TBPS)binding (2, 3). Cyclodienes (e.g., a-endosulfan) and related insecticides are also recognized and analyzed by cyclodiene monoclonal antibodies (MAbs) with an aldrin hapten conjugate (5,6) (Figure 1). A preliminary survey of other GABAA receptor antagonists revealed that tetramethylenedisulfotetramine (TETS) (2,6-dithia1,3,5,7-tetraazatricy~lo[3.3.1.1~~~]decane 2,2,6,6-tetraoxide) (Figure 1) is also recognized by both the TBPS binding site and the cyclodiene antibody, thereby rekindling interest in this heteroadamantane and related compounds. TETS is highly toxic to man (7, 8) and laboratory mammals (9). It is a noncompetitive GABA antagonist on the basis of electrophysiological experiments (IO,11), and it inhibits both GABA-sensitive [36S]TBPSbinding in t Pesticide Chemistry and Toxicology Laboratory, Department of Entomological Sciences. t Hybridoma Facility, College of Natural Resources.
0893-228x/91/2704-0162$02.50/0
brain membranes (12) and GABA-stimulated chloride flux in brain microvesicles ( 4 ) . However, there are also differences in the neurotoxicity of TETS and the cyclodienes. Those molecules with a small or compact cage structure (such as TETS and TBPS) are more toxic to mice than would be expected from their TBPS receptor ICWsand have lower insecticidal activity than other molecules which have larger or more extended polycyclic systems (e.g., the cyclodiene insecticides and 1-substituted-phenyl-2,6,7trioxabicyclo[2.2.2]octanes) (13). We therefore prepared a series of TETS analogues to establish the effects of steric bulk and symmetry on their interactions with the nerve and antibody recognition sites. This study compares these new as well as known TETS analogues (Figure 2) with other established GABAA receptor antagonists as a means to define similarities and differences in the cyclodienesensitive nerve and MAb binding sites. Materials and Methods Chromatography and Spectroscopy. TLC used silica gel chromatoplates developed with 20% cyclohexane and 1% triethylamine in chloroform (for TETS and HEXS),dichloromethane (for 1 and 3-6), 1% triethylamine in dichloromethane (for 7), and ethyl acetate (for 2). UV light was used to visualize 2 and its analogue (SO2replaced by CH,) and the chromotropic
0 1991 American Chemical Society
Recognition of Sulfamides by Chloride Channel a n d MAbs CI
0
N,
4
f/ F0 Cl
0 a-endosulfan
TETS
CI
0 aldrin hapten conjugate
Figure 1. Structures of two GABAAreceptor antagonists (aendosulfan and TETS) and of the aldrin hapten conjugate used for immunization and screening antigens. heteroadamantanes
n
Chem. Res. Toxicol., Vol. 4, No. 2, 1991 163 TETS was prepared by a modification of the procedure of Kang e t al. (19). 1,3,5-Trioxane (0.72 g, 8 mmol) and sulfamide (1.16 g, 12 mmol) were dissolved in trifluoroacetic acid (30 mL) at 0 "C. The reaction mixture was stirred a t 0 "C for 3 h and then a t room temperature overnight. The solution was cooled (ice bath), water (20 mL) added, and the resulting precipitate filtered off. The amorphous product (1.25 g) was dissolved in acetone and then crystallized by the slow addition of hexane, yielding T E T S as colorless cubes: yield 73% (1.10 g); mp >270 "C; 'H NMR 6 5.55 (8, 8 H); 13C NMR 6 71.02; MS m/z 240 (54, M'), 212 (loo), 149 (4), 132 (14), 121 (14), 92 (16), 76 (6). HEXS was obtained according to Kang et al. (method 4 in ref 19) and further purified by washing with acetone: yield 30%; mp >270 "C; lH NMR 6 5.22 and 5.59 (AB, J = 15 Hz, 12 H); 13C NMR 6 61.29; MS m/z 360 (55, M'), 267 (18), 212 (42), 174 (43), 148 (77), 121 (37), 92 (loo), 76 (65), 57 (81). Compound 1 was prepared by dropwise addition of aqueous NH40H solution (58%, 1.2 mL) to a solution of sulfamide (0.96 g, 10 "01) in aqueous formaldehyde (37%, 4 g, 50 "01) at mom temperature with stirring for 4 h (20). The precipitate formed was filtered off and dried a t high vacuum: yield 37% (0.71 g); mp 220-221 "C [lit. (20) 224-225 "C]; 'H NMR 6 4.50 (s, 2 H), 4.90 and 5.18 (AB, J = 15 Hz, 8 H); 13C NMR 6 71.99,72.83; MS m/z 190 (100, M+), 162 (241, 97 (81, 92 (3). Compound 2 was prepared according to Misiti and Chiavarelli (21): yield 42%; mp 248-249 "C; 'H NMR (CDC13)6 4.03 and NMR 6 48.72, 4.78 (AB,J = 15 Hz, 8 H), 7.15-7.42 (m, 10 H); 61.54,126.72,128.22,128.64,133.85,202.92; MS m/z 354 (57, M+), 289 (4), 261 (ll),247 (49), 219 (22), 130 (23), 117 (55), 103 (loo), 77 (45), 51 (9). l-Phenyl-l,2-diaminoethane, a precursor for 5, was prepared as follows: (+)-2-phenylglycinonitrile hydrochloride (3.37 g, 20 mmol) was dissolved in absolute ethanol (100 mL) in a Parr hydrogenator bottle and cooled to 5 "C. HCl(37% aqueous, 20 mmol) was added in ethanol (100 mL) while maintaining the temperature. The solution was allowed to warm to room temperature, platinum(1V) oxide (0.15 g) was added, and the mixture was hydrogenated a t 3 atm for 8 h. The catalyst was filtered off and the ethanol evaporated, yielding the crude product (3.8 g) as its hydrochloride salt. Portions of the free base (0.13 g) were prepared as required, immediately prior to use, by dissolving the salt (0.2 g) in aqueous KOH (2 N, 10 mL) and extracting into ether (3 X 20 mL). Hetero(hom0)adamantanes 3-7 were prepared by dropwise addition of an aqueous solution of the appropriate 1,2-diamine (10 mmol) to a solution of sulfamide (10 mmol) and aqueous formaldehyde (37%, 50 mmol) a t room temperature. Product formation was characterized by a colorless precipitate, which formed either immediately or after stirring for several hours followed by ice cooling. This was filtered off and dried a t high vacuum, and the product was purified by column chromatography on aluminum oxide with dichloromethane for 4 and 6, on silica gel with dichloromethanefor 5, and on silica gel with 1% methanol in chloroform for 7. Compound 3 (pure without chromatography): yield 50%; mp 223 "C dec [lit. (20) 195-196 "C dec and (19) 220 "C]; 'H NMR 6 3.15 (s, 4 H), 4.25 and 4.85 (AB, J = 12 Hz, 8 H); 13CNMR 6 56.98,72.20; MS m/z 204 (100, M'), 176 (53), 162 (64), 140 (75), 121 (4), 112 (85), 97 (4), 85 (31), 71 (14). Compound 4: yield 24%; mp 159-161 "C;'H NMR 6 1.32 (s, 6 H), 3.03 (e, 2 H), 4.41 and 4.79 (AB, J = 15 Hz, 4 H), 4.67 and 4.84 (AB, J = 15 Hz, 4 H); 13C NMR 6 29.41, 62.83, 66.31, 70.08, 71.99; MS m / z 232 (13, M'), 176 (8), 168 (361, 140 (22), 112 (loo), 99 (52), 85 (29). Compound 5: yield 40%; mp 184-186 "C; 'H NMR 6 3.95-5.11 (m, 11 H), 7.35-7.37 (m, 5 H); 13C NMR 6 62.58, 66.69, 68.69, 70.10, 74.15, 126.65, 126.83, 128.57, 141.27; MS m/z 280 (2, M'), 216 (12), 147 (19), 118 (1001, 91 (551, 77 (81, 65 (8). Compound 6: yield 85%;mp 198-199 "C;'HNMR 6 1.3-1.8 (m, 8 H), 2.81 (d, J = 9 Hz, 2 H), 4.3-4.9 (m, 8 H); 13C NMR 6 25.65, 32.75,66.18,72.02,75.16; MS m/z 258 (24, M+), 230 (61, 194 (6), 166 (77), 152 (8), 137 (loo), 123 (35), 109 (47), 96 (97), 81 (63). Compound 7: yield 4%; mp 196-198 "C; 'H NMR 6 4.67 and 5.10 (AB, J = 15 Hz, 8 H), 7.28 (s, 4 H); lYCNMR 6 70.41, 127.64, 128.63, 151.41; MS m / z 252 (21, M+), 131 (100). [%ITBPS Assays. Mouse brain P2membranes were dialyzed against EDTA and water (12). Each incubation mixture consisted of 200 pg of P2 membrane protein in 1 mL of 200 mM NaCl-50
'w
HEXS
1
2
heterohomoadamantanes
3 R1-Rz-H 4 R1 - R Z = C H s 5 R, = C6Hg, Rz = H
8
7
Figure 2. Structures of eight sulfamide analogues of TETS. acid/sulfuric acid spray reagent (14) for the others. 'H and 13C NMR spectra were obtained a t 300 and 75 MHz, respectively, with a Bruker WM-300 spectrometer for compounds dissolved in DMSO-d, unless otherwise specified. 'H NMR data are quoted as chemical shifts (ppm), multiplicity, coupling constants (Hz), and number of protons. 13Cchemical shifts are given in ppm. Electron impact mass spectra were recorded with a Hewlett Packard 5985 system a t 70 eV. The purity of compounds synthesized was determined on a Hewlett Packard 5830A gas chromatograph fitted with a 100% methylsilicone Durabond capillary column DB1 (30 m, 0.25 mm i.d., 0.25-pm film thickness) and a flame ionization detector. Hydrogen was the carrier gas (17 psi, 65 cm/s). The temperature program (150 "C, 10 min; 10 "C/min to 250 "C; 250 "C, 10 min) gave tR values (min) as follows: TETS, 3.28; 1, 2.85; 2,21.97; 3,5.23; 4,8.23; 5, 17.38; 6, 14.45; 7,11.64; and the analogue of 2 with SO2 replaced by CH2, 19.39. Chemicals. The trioxabicyclooctanes were available from earlier studies (15-17). Sources for picrotoxinin and the polychlorocycloalkane insecticides were as previously described (4). [%SITBPS(60-105 Ci/mmol; >99% radiochemical purity) was from New England Nuclear Corp. Hexamethylenetetramine was purchased from Aldrich Chemical Co., and the analogue of 2 (SO2 replaced by CH2) was synthesized according to Stetter et al. (18): yield 20%; mp 262-263 OC; 'H NMR 6 3.66 and 3.81 (AB, J = 12 Hz, 8 H), 4.14 (s,2 H), 7.22-7.36 (m, 10 H); 13C NMR 6 52.65, 64.30,72.96, 126.51, 127.23, 128.19,137.44,206.60; MS m / z 304 (73, M'), 261 (77), 247 (21), 233 (30), 159 (15), 144 (20), 131 (37), 103 (loo), 77 (291, 58 (6).
164 Chem. Res. Toxicol., Vol. 4, No. 2, 1991 mM sodium phosphate, pH 7.4, buffer containing [%SITBPS (2 nM) alone or with unlabeled TBPS (2 pM) to correct for nonspecific binding (12% relative to total binding) (22). Assays were incubated for 30 min a t 37 "C and then filtered. Toxicity t o Mice a n d Insects. Acute (24-h) toxicity (LDW) to male albino Swiss Webster mice (18-22 g) (Simonsen Laboratories, Gilroy, CA) was determined by intraperitoneal (ip) administration of the test compound in dimethyl sulfoxide (DMSO) (50-100 pL) except for hexamethylenetetramine and 2, which were administered in water and methoxytriglycol, respectively. Insect toxicity determinations (24 h) utilized adult female houseflies (Musca domestica L., SCR strain, -20 mg each). Flies were held a t 25 "C after injection of the test compound in acetone (0.22 pL) into the thorax. Monoclonal Antibody Competition Enzyme Immunoass a y s (EIA). The EIA measured the ability of an analyte in solution to compete with an immobilized aldrin conjugate (absorbed to microplate wells) for binding a limiting amount of MAb. Synthesis of the aldrin hapten and protein conjugates and derivation of the cyclodiene-specific MAbs have been briefly described previously (6). Generally, 6,7-dihydro-6-endo-hydroxyaldrin was conjugated to bovine serum albumin (aldrin-BSA) with a hapten density of approximately 12 molecules of cyclodiene/ molecule of BSA. The MAbs, designated 8Hll,6A12,4E3, and 12A9, were all of the IgGIK immunoglobulin subclass. Solutions of the test compounds (>99% pure) in DMSO were stored in tightly sealed Teflon vials and used within 2 days. The EIAs were based on procedures of and utilized buffers described by Voller et al. (23). Conjugates were absorbed to EIA plates in 'coating buffer" (15mM Na2C03-35 mM NaHC03, pH 9.6-3 mM NaNJ. Antibodies and antibody-analyte mixtures were diluted in 'PBS-Tween" (10 mM KH2PO4/K2HPO4,pH 7.4-150 mM NaCl-3 mM NaN3-0.05% Tween 20) containing 10% (v/v) DMSO. The substrate solution for color development was p nitrophenyl phosphate (1 mg/mL) in aqueous 10% (w/v) diethanolamine hydrochloride, pH 9.8, containing MgC1, (0.4 nM) and NaN3 (3 mM). Dilutions of the analyte or aldrin reference standard in DMSO were mixed with a limiting amount of hybridoma culture fluid (made up in 0.1 mL of PBS-Tween to give a final DMSO concentration of 10%)and incubated overnight a t 22 OC in tightly sealed polypropylene tubes. Optimal dilutions of the hybridoma culture fluid in these mixtures (1:50 for MAb 8 H l l ; 1:200 for the others) were determined in previous EIAs. Wells of Dynatech Immulon 2 EIA plates were coated overnight a t 4 "C with aldrin-BSA (0.2 pg for MAb 8 H l l and 0.05 pg for MAbs 6A12,4E3, and 12A9) in 0.1 mL of coating buffer. The wells were washed three times with PBS-Tween and then incubated for 2 h a t 22 "C with the mixtures of analyte and hybridoma culture fluid. The wells were again washed three times with PBS-Tween, incubated for 2 h a t 22 OC with alkaline phosphatase conjugated goat anti-mouse IgG (Sigma Chemical Co., St. Louis, MO), and washed a final three times. The substrate solution (0.1 mL/well) was added, and the rate of color development at 405 nm (AA,,/min) was recorded by using a Flow Multiskan EIA reader interfaced with a Macintosh computer. Dose-response curves were fitted and ICWvalues obtained by iterative regression to the four-parameter logistic equation (24)using Passage I1 (Passage Software, Inc., Fort Collins, CO) on a Macintosh computer.
Results Potency of Polychlorocycloalkane Insecticides and Other GABAAReceptor Antagonists as Inhibitors of the GABA-Gated Chloride Channel and Four Cyclodiene MAbs (Table I, Figure 3). Four cyclodienes and t w o o t h e r polychlorocycloalkane insecticides are recognized, as expected, b y both the [35S]TBPS receptor and cyclodiene MAb 8Hll. The channel and antibody recognition sites are similar to each other i n their stereospecificity for lindane relative to the a-,0-, and &isomers of hexachlorocyclohexane. W i t h the notable exception of a-endosulfan, the cyclodienes and polychlorocycloalkanes are recognized at lower concentrations i n the antibody assay than i n the chloride channel assay.
Esser et al.
- - - - - ---
0.03 1.0
0.3
1
10
3
30
Competitor (FM) Figure 3. Inhibition of cyclodiene MAb 8 H l l by aldrin, lindane, HEXS, and TETS shown as doseresponse curves for competition EIA. Assays were conducted as described under Materials and Methods. The data are for individual samples of aldrin a t each dilution and for the mean i standard deviation (error bar) for triplicate samples of lindane, TETS, and HEXS at each dilution. The rates of color development (AAm/min) were normalized for each of the analytes, i.e., the rates at limiting low dose and limiting high dose (RLand RH)were set to 1.0 and 0.0, respectively, and the normalized rates for each point Ri were then calculated as
(Ri - RH)/(RL- RH). Table I. Potency of Polychlorocycloalkane Insecticides and Other GABAA Receptor Antagonists as Inhibitors of the GABA-Gated Chloride Channel and of Cyclodiene MAL 8Hll chloride channel compound cyclodiene insecticides a-endosulfan 6-endosulfan dieldrin aldrin other polychlorocycloalkane insecticides toxaphene lindane 2,6,7-trioxabicyclo[2.2.2]octanesc t-Bu-C(CH,O)&-Ph (TBOB) t-Bu-C(CH20)3P=S (TBPS)
antibody
0.1' 1.5" 1.4" 8.70
0.5
2.9" 1.7"~~
0.06 0.7b
0.05d
0.06d ~ - B U - C ( C H ~ O ) ~ C - P ~ - ~ - S O6.8' ~M~ other GABAAreceptor antagonists picrotoxinin (PTX) 0.2e TETS 0.5eJ
0.5
0.1 0.1
>lo00
> 1000 >1000 > 1000 3
'Rat brain (2). bThe a-, 8-, and &isomers are less active in the chloride channel (2) and antibody (6)assays. 'Similar results [ICm lo00 pM in the chloride channel (mouse brain) and antibody assays, respectively] are obtained with R-C(CH,O),C-R', wherein R = s-Bu, R' = Ph-4-CN; or R = n-Pr, R = Ph-4-C=CH. Human brain (25). e Mouse brain (new determinations). 'Hill number 0.96.
An interesting pattern was revealed b y examining three o t h e r types of GABAA receptor antagonists (trioxabicyclooctanes, PTX, and T E T S ) for possible recognition b y the channel and antibody sites. Of these, the sulfamide TETS is recognized by both the chloride channel and MAb 8 H l l with IC,s of 0.5 and 3 pM,respectively, b u t five trioxabicyclooctanes and PTX, which are very potent at the [35S]TBPS binding site, are inactive in the a n t i b o d y assay. The cross-reactivity of MAb 8Hll to the cyclodiene and polychlorocycloalkane insecticides and the sulfamide TETS is not observed with the other t h r e e MAbs tested. Thus, MAb 8Hll is more sensitive than MAbs 6A12,4ES, and 12A9 to each of the six chlorinated insecticides considered here, and each M A b shows a different pattern of specificity for these compounds (6). MAb 8Hll is also unique i n being t h e only one of the four MAbs which recognizes a GABAA receptor antagonist other than the
Recognition of Sulfamides by Chloride Channel and MAbs Table 11. Structure-Activity Relationships of TETS and Related Sulfamides as Toxicants and Inhibitors of the GABA-Gated Chloride Channel and of Cyclodiene MAb 8Hll
LD50,m g / k compound hexamethylenetrisulfohexamine
HEXS
mouse
1 2
hetero(hom0)adamantanes 3 4 5 6 7
chloride channel
antibody
0.03
1
1
I
I
1
Kd 'max o control 46 6.0 A TETS 125 6.5
0
-
-
13
heteroadamantaned
TETS
housefly
Chem. Res. Toxicol., Vol. 4 , No. 2, 2991 165
0.24c
>250
>10 (14)O
.- 90
>250 >100
>250 >250
0.5 >10 >10
>125 4.0 >250 >250 125
>250
>10
>250 >250 200 >250
-
0.43 3 28 >lo00
>1000 >10 (17) >lo00 >10 >loo0 >10 >1000 >10 (20) 41
"Inhibition ( 7 0 )at 10 pM is given in parentheses. 6The analogue of 2 with SOz replaced by CH2 and hexamethylenetetramine (methenamine) are inactive, i.e., >loo, >250, >lo, and >1000, respectively, in the tabulated assays. The mouse ip LDw for the former compound is reported to be 67 mg/kg (26). 'Ref 27.
cyclodienes and polychlorocycloalkanes;Le., the other three MAbs do not recognize the trioxabicyclooctanes, PTX, or TETS (or HEXS discussed below) even at 1000 pM. Structure-Activity Relationships of TETS and Related Sulfamides as Toxicants and Inhibitors of the GABA-Gated Chloride Channel and of Cyclodiene MAb 8 H l l (Table 11). TETS differs in structural properties from the other classes of GABAAreceptor antagonists considered: it has two planes of symmetry whereas the others have no more than one; it is also unique in having a sulfamide functionality. To ascertain the contribution and relevance of these structural features to recognition by the chloride channel and antibody sites, a series of analogues was prepared. TETS is >20-fold more potent than eight other sulfamides and two related nonsulfamide compounds in blocking the GABA-gated chloride channel. TETS is also the most toxic compound to mice and houseflies, followed by 4 and HEXS for mice and 6 for houseflies. MAb 8 H l l recognizes four of the sulfamides with a potency order that is not correlated to their neurotoxicity. Although HEXS is >20-fold less active at the chloride channel, its apparent affinity for MAb 8 H l l is 7-fold better than that of TETS. In view of the toxicity observed for HEXS and since TETS can occur as a byproduct in its synthesis (191, the sample of HEXS was specifically analyzed by gas chromatography for impurities; no trace of TETS was observed. The TETS analogue with methylene replacing SO2 (i.e., 1) is also recognized but with a 9-fold lower potency than TETS. The phenylenediamine analogue 7 also shows some activity. Clearly, the structureactivity pattern differs greatly for the channel and antibody recognition sites. TETS as a Competitive Inhibitor of TBPS Binding. Scatchard analysis (Figure 4) and the Hill number of 0.96 for TETS competition with TBPS binding (Table I) indicate that TETS and TBPS act at the same site or closely coupled sites.
Discussion The toxicity of the cyclodienes results from their blocking the GABA-gated chloride channel. As environmental contaminants, they can be analyzed with
-
OO
0.5
0
1.o
1.5
Figure 4. Scatchard plot of specific [%ITBPS binding to m o w brain membranes alone or with 0.5 pM TETS. Average of two experiments for TBPS and three for TETS. B and Fare bound and free [35S]TBPS.B is given as pmol/0.2 mg of protein, B,, as pmol/mg of protein, and Kd as nM.
"cyclodiene-specific" MAbs. This study utilizes a series of chemical probes to determine possible similarities in the recognition sites of the chloride channel and of four cyclodiene-specific antibodies. The sulfamides proved to be of particular interest because of their differing specificity patterns for each of the recognition sites. This cross-reactivity of cyclodienes and some sulfamides does not detract from the practical application of MAb 8 H l l for residue analysis of cyclodienes and other polychlorocycloalkanes since neither TETS nor HEXS would appear in environmental samples. The structure of TETS has been related to the conformations of a diverse series of convulsant drugs in proposing a model for the chloride channel binding site (B), but the polychlorocycloalkane insecticides were not included in this comparison. The sulfamides examined include the highly symmetrical TETS molecule and less symmetrical, larger, or more extended analogues, with modifications that in the trioxabicyclooctane series might be expected to enhance GABAAreceptor potency and insecticidal activity (13). However, this did not prove to be the case, and for now, TETS remains the optimal structure for neuroactivity and insecticidal activity among the hetero(homo)adamantanes. It appears that, in addition to the sulfamide functionality, the symmetry properties of TETS are critical to its unique toxicological properties. The hapten used to prepare the MAbs is an ether derivative of aldrin, specifically designed to evoke specificity for the hexachlorocyclopentenemoiety of the molecule. It is thus expected that the resulting MAbs might recognize most, if not all, cyclodienes and related polychlorocycloalkanes. MAb 8 H l l binds chlordane and endosulfan isomers with different specificities than MAbs 6A12,4E3, and 12A9 (6). It is the rule, rather than the exception, that a small molecule hapten (such as a cyclodiene) will elicit MAbs with different affinities and specificities for the molecule and its analogues; e.g., Leahy et al. (29) characterized different amino acid sequences from the combining sites of 1 2 MAbs that are all specific for dinitrophenol. Thus, it is not surprising that only MAb 8 H l l and not the other MAbs recognize HEXS, TETS, and two other sulfamides. The cyclodiene insecticides, TETS, and TBPS act at the same site or closely coupled sites within the mammalian GABA-gated chloride channel. The cyclodiene-TETS cross-reactivity for the chloride channel extends to the assays with MAb 8 H l l . However, the structural modifications of TETS considered to date result only in drastically reduced potency for the chloride channel: in contrast, MAb 8 H l l binds HEXS better than it binds TETS.
166 Chem. Res. Toxicol., Vol. 4, No. 2, 1991
Figure 5. Common structural features of some of the polychlorocycloalkanes and sulfamides active a t the MAb 8 H l l recognition site. These partial structures for the most active compounds shown in Figures 1 and 2 and for lindane depict two electron-rich centers, positioned diametrically opposite each other in a cyclic system which has a plane of symmetry as shown. One electron-rich region is the planar Cl-C-CClX-C-Cl in the cyclodienes (X = C1) and lindane (X = H) or is N-S02-N in the sulfamides. The opposite electron-rich region is C 4 , epoxide, or OS(0,)O in the cyclodienes and CH-Cl in lindane (Y)or is N-S02-N in the sulfamides (Z). It appears that the sulfamide functionality is a minimum requirement for recognition by MAb 8Hll. Common structural features of some of the cyclodienes and sulfamides that are recognized by MAb BHll, including a plane of symmetry and diametrically opposed centers of high electron density, are illustrated in Figure 5. In this comparison the planar Cl-C-CCl,-C-Cl group of the cyclodienes is equated with the sulfamide group of TETS. However, these features are not sufficient in themselves to confer activity (i.e., 2 is inactive). It is possible to derive MAbs that bind small-molecule ligands with a preference similar to that of a receptor combining site (30-32). Such MAbs could facilitate an understanding of ligand-receptor interactions and structure-activity relationships. Recent advances in the expression of antibody combining sites in recombinant DNA systems may facilitate the derivation and use of receptor-like MAbs (33,34). The present study constitutes the first of many steps required for such developments with the "convulsant binding site" of the GABA-gated chloride channel.
Acknowledgment. We thank our laboratory colleagues Loretta Cole for the receptor assays, Jon Hawkinson for helpful discussions, Thomas Class and Mark Sanders for recording the mass spectra, Weiching Wang and Rick Grendell for the insect assays, and Judith Engel for the mouse LD,, determinations. We acknowledge Allison Lim and Joanna Liliental for excellent technical assistance performing the competition EIAs. This study was presented in part as a poster at the International Conference on Pesticides and Alternatives, Colymbari, Crete, September 1989. This research was supported by California Department of Food and Agriculture Contract 3586 (to A.E.K.) and NIH Grant 5P01 ES00049 (to J.E.C.).
References (1) Matsumura, F. (1987) Advances in understanding insecticide modes of action. In Pesticide Science and Biotechnology (Greenhalgh, R., and Roberts, T. R., Eds.) pp 151-159, Blackwell, Oxford, England. (2) Lawrence, L. J., and Casida, J. E. (1984) Interactions of lindane, toxaphene and cyclodienes with brain-specific t-butylbicyclophosphorothionate receptor. Life Sci. 35, 171-178. (3) Casida, J. E., and Lawrence, L. J. (1985) Structure-activity correlations for interactions of bicyclophosphorus esters and some polychlorocycloalkane and pyrethroid insecticides with the brain-specific t-butylbicyclophosphorothionatereceptor. Enuiron. Health Perspect. 61, 123-132. (4) Obata, T., Yamamura, H. I., Malatynska, E., Ikeda, M., Laird, H., Palmer, C. J., and Casida, J. E. (1988) Modulation of yaminobutyric acid-stimulated chloride influx by bicycloorthocarboxylates, bicyclophosphorus esters, polychlorocycloalkanes and other cage convulsants. J. Pharmacol. Exp. Ther. 244, 802-806.
Esser et al. (5) Jung, F., Gee, S, J., Harrison, R. O., Goodrow, M. H., Karu, A. E., Braun, A. L., Li, Q. X., and Hammock, B. D. (1989) Use of immunochemical techniques for the analysis of pesticides. Pestic. S C ~26, . 303-317. (6) Karu, A. E., Liliental, J. E., Schmidt, D. J., Lim, A. K., Carlson, R. E., Swanson, T. A., Buirge, A. W., and Chamerlik, M. A. (1990) Monoclonal antibody-based immunoassay of cyclodienes. Proceedings of the 6th Annual EPA Waste Testing and Quality Assurance Symposium, Washington, DC, July 18-20,1990, Vol. I, pp 237-245. (7) Hecht, G., and Henecka, H. (1949) Uber Bin hochtoxisches Kondensationsprodukt von Sulfamid und Formaldehyd. Angew. Chem. 61, 365-366. (8) Hagen, J . (1950) Schwere Vergiftungen in einer Polstermobelfabrik durch einen neuartigen hochtoxischen Giftstoff (Tetramethylendisulfotetramin).Dtsch. Med. Wochenschr. 75, 183-184. (9) Haskell, A. R., and Voss, E. (1957) The pharmacology of tetramine (tetraethylenedisulfotetramine). J. Am. Pharm. Assoc. 46, 239-242. (10) Bowery, N. G., Brown, D. A., and Collins, J. F. (1975) Tetramethylenedisulphotetramine: an inhibitor of y-aminobutyric acid induced depolarization of the isolated superior cervical ganglion of the rat. Br. J. Pharmacol. 53,422-424. (11) Dray, A. (1975) Tetramethylenedisulphotetramineand amino acid inhibition in the rat brain. Neuropharmacology 14,703-705. (12) Squires, R. F., Casida, J. E., Richardson, M., and Saederup, E. (1983) [35S] t-Butylbicyclophosphorothionatebinds with high affinity to brain-specific sites coupled to y-aminobutyric acid-A and ion recognition sites. Mol. Pharmacol. 23, 326-336. 13) Palmer, C. J., and Casida, J. E. (1988) Two types of cage convulsant action at the GABA-gated chloride channel. Toxicol. Lett. 42, 117-122. 14) Beroza, M. (1963) Identification of 3,4-methylenedioxyphenyl synergists by thin-layer chromatography. J. Agric. Food Chem. 11,51-54. 15) Milbrath, D. S., Engel, J. L., Verkade, J. G., and Casida, J. E. (1979) Structure-toxicity relationships of 1-substituted-4-alkyl2,6,7-trioxabicyclo[2.2.2]octanes.Toxicol. Appl. Pharmacol. 47, 287-293. (16) Palmer, C. J., and Casida, J. E. (1985) 1,4-Disubstituted 2,6,7trioxabicyclo[2.2.2]octanes: a new class of insecticides. J. Agric. Food Chem. 33, 976-980. (17) Palmer, C. J., and Casida, J. E. (1989) 1-(4-Ethynylphenyl)2,6,7-trioxabicyclo[2.2.2]octanes: a new order of potency for insecticides acting at the GABA-gated chloride channel. J . Agric. Food Chem. 37, 213-216. (18) Stetter, H., Schafer, J., and Dieminger, K. (1958) Uber die bildung des 1,3-Diaza-adamantan-Ringsystems durch MannichKondensation. Chem. Ber. 91, 598-604. (19) Kang, J.-B., Thyagarajan, B. S., Gilbert, E. E., and Siele, V. (1971) A new condensation product from sulfamide and paraformaldehyde. Int. J. Sulfur Chem., A 1, 261-268. (20) Paquin, A. M. (1948) Neue Verbindungen und Reaktionen des Sulfamids. Angew. Chem. A 60, 316-320. (21) Misiti, D., and Chiavarelli, S. (1966) Reactivity of 3,7-diazaadamantanes. Synthesis of 1,5-dipheny1-3,7-diaza-lO-thioadamantan-9-one, 10-oxideand 10,lO-dioxide. Catz. Chim. Ital. 96, 1696-1714; (1967) Chem. Abstr. 66, 857771.1. (22) Cole, L. M., Lawrence, L. J., and Casida, J. E. (1984) Similar properties of [35S]t-butylbicyclophosphorothionatereceptor and coupled components of the GABA-receptor-ionophore complex in brains of human, cow, rat, chicken and fish. Life Sci. 35, 1755-1762. (23) Voller, A., Bidwell, D., and Bartlet, A. (1976) Microplate immunoassays for the immunodiagnosis of virus infections. In Manual of Clinical Immunology (Rose, N., and Friedman, H., Eds.) pp 506-512, American Society of Microbiology, Washington, DC. (24) Canellas, P., and Karu, A. E. (1981) Statistical package for analysis of competitive ELISA results. J. Immurrol. MPthods 47, 375-385. (25) Casida, J. E., Palmer, C. J., and Cole, L. M. (1985) Bicycloorthocarboxylate convulsants. Potent CABAAreceptor antagonists. Mol. Pharmacol. 28, 246-253. (26) Chiavarelli, S., Fennoy, L. V., Settimj, G., and De Baran, L. (1962) The effect of methoxyphenyl substitutions on the strychnine-like activity of aryldiazaadumentanones and aryldiazaadamantanols. J. Med. fJharm.Chem. 5, 1293-1297.
Recognition of Sulfamides by Chloride Channel and MAbs
Chem. Res. Toxicol., Vol. 4, No. 2, 1991 167
(27) Casida, J. E., Eto, M., Moscioni, A. D., Engel, J. L., Milbrath, D. S., and Verkade, J. G. (1976)Structure-toxicity relationships of 2,6,7-trioxabicyclo[2.2.2]octanesand related compounds. Toricol. Appl. Pharmacol. 36, 261-279. (28) Wong, M. G., and Andrews, P. R. (1989)Conformational requirements for convulsant compounds. Eur. J. Med. Chem. 24, 323-334. (29) Leahy, D.,Rule, G., Whittaker, M., and McConnell, H. (1988) Sequences of 12 monoclonal antidinitrophenyl spin-label antibodies for NMR studies. Proc. Natl. Acad. Sci. U.S.A. 85, 3661-3665. (30) Haber, E. (1982)Monoclonal antibodies to drugs: new diagnostic and therapeutic tools. In Monoclonal Antibodies in Clinical Medicine (McMichael, A. J., and Fabre, J . W., Eds.) pp 494-497, Academic Press, New York.
(31) Chatenoud, L., Villemain, F., Hoebeke, J., Garbarg, M., Komer, M., Gros, C., Ruat, M., Cazenave, P. A., Ganellin, C. R., and Bach, J. F. (1988)Polyclonal and monoclonal antibodies directed against SK&F 94461, a specific H1 histamine receptor ligand. Mol. Pharmacol. 34, 136-144. (32) Nahmias, C., Strosberg, A. D., and Emorine, L. (1988)The immune response toward beta-adrenergic ligands and their receptors. VIII. Extensive diversity of VH and VL genes encoding anti-alprenolol antibodies. J.Immunol. 140,1304-1311. (33) Marx, ?J. (1989)Learning how to bottle the immune system. Science 246, 1250-1251. (34) Huse, W. D.,Sastry, L., Iverson, S. A., Kang, A. S., Alting-Mew, M., Burton, D. R., Benkovic, S. J., and Lerner, R. A. (1989)Generation of a large combinatorial library of the immunoglobulin repertoire in phage lambda. Science 246, 1275-1281.
