Document KxjbzMjywe3wEOz05JggY7k6

00M-tiX//02On9-C99O1.00/0 Copynfflt <. tyy Th. A a-ne Soc y fot Pham*** .nd Espanmafitai Th.rep~t.re AUn |flU of reproduction in any lorm rewrred. MOLXCLLa A PMAAMACOLOOY, 3 4 22 - 23? Kinetic and Equilibrium Studies of Ah Receptor-Ligand Binding: Use of [125l]2-lodo-7,8-dibromodibenzo-p-dioxin CHRISTOPHER A. BRAOFIELO. ANDREW S. KENDE. and ALAN POLAND McAraie Laboratory tor Cancer Reseerch. University ot Wisconsin. Medlson, Wisconsin 53706 (C.A.B.. A.P) and Department ot Chemistry, University ot Rocnester. Rochester New York 14627 (A.S.K.) Received Decernoer 14, 1987: Accepted Apr* 12. 1988 SUMMARY In this report, we have used the radioligand (,29l]2-iodo-7.8dibromo-dibenzo-p-dioxin to desenbe the kinetics of ligand bind ing to the A h receptor prepared from C57BL/6J mouse liver. The higher specific activity of this radioligand (2176 Ci/mmol). com pared with the usuai tntiated ligand (1,6-3H]2.3.7,8-tetrachiorodibenzo-p-dioxin (58 Ci/mmol) permitted the study of ligandreceptor interactions at much lower component concentrations. For this radioiodinated ligand. Scatchard analysis of saturation binding curves, determined at six different protein concentra tions. indicated that the apparent equiiibnum dissociation con stant, K0, was directly related to the dilution of the receptor preparation; for example, at 1160 ug of protetn/mi, K0 - l . 6 x 10" m; at 36 ug of protein/ml. K0 - 1.2 x 10"" m. Extrapolation of this function to infinite receptor dilution yielded K0 - 6 x 10"'2 m. The addition of 70 ug/ml of bovine serum albumin to a receptor preparation of 30 ug of protein/ml produced a 10-fold decrease m the slope of the Scatchard plot (i.e.. 10-fold increase in the apparent K0). Conversely, enrichment of the receptor by high performance liquid chromatography led to an increased slope and thus decreased estimate of K0. The association rate constant (ft,), calculated from the integrated second-order rate equation, was 2.8 x 10,o m'' hr-' and. from the initial velocity equation, had a value of 5.25 x 10' m*' hr"'. The dissociation rate constant was Chphasic, consisting of a predominant fast com ponent with a rate constant of 0.36 hr" (k_,) and a slower component with a rate constant of 4 2-9.4 x 10'3 hr" (A_2). Higher protein concentrations produced a decrease in estimates of it, but not /r_, or k_2. The K0 determined from the ratio of the kinetic rate constant. k_,/k, = 6.9 x 10"2 m, is in excellent agreement with that derived from the results of equilibrium, binding expenments extrapolated to infinite dilution. K0 = 6 x 1 0 "2 m. The decrease in K0. observed in equiiibnum binoing studies upon dilution of the receptor preparation, is best ex plained by a more accurate classification ot `free' radioligand at lower protein concentrations. Finally, ligand binding to the Ah receptor is best described by a two-step process, the formation of an initial complex, charactenzed by rapid ligand dissociation, which undergoes transformation to a second distinct complex displaying a much slower ligand dissociation rate. 783948 Planar haiogenated aromatic hydrocarbons licit a variety of biological effects, including induction of cytochrom P-450 isozymes, epidermal hyperplasia and metaplasia, thymic invo lution, teratogenesis, tumor promotion, and lethality (1). These diverse biological responses are believed to arise from the altered gene expression that is consequent to the high affinity, stereospecific binding of planar halogenated aromatic hydro carbons to a soluble protein known as the Ah receptor (1, 2). Like the steroid hormone receptors, the Ah receptor is rou tinely found in the high speed supernatant fraction ("cytosol") T h u paper u dadicatad to Or. Elizabeth Miller. This work was supported in pan by the National Institute oi Environmental Health Science Grant 3-01884. National Cancer Institute Core Grant-07173, and National Cancer Institute Postdoctoral Training Grant T32-CA0902Q. of tissue homogenates. After ligand binding and a poorly de fined "transformation"1step, the ligand-receptor complex dis plays an increased affinity for polyanions, such as DNA (3-3), and functions as a transcriptional activator (7, 8). The A h receptor also bears many physicochemical similarities to steroid hormone receptors, including a high molecular weight "aggre gate" form in low salt buffers (Af, 2.5-2.8 x 104) that dissociates to a low molecular weight "monomeric subunit" form in high salt buffers (Af, approximately 1 x 104) (9). For the Ah receptor present in the hepatic cytosol of the C57BL/6 mouse and the Sprague-Dawley rat estimates of sedimentation coefficients. 1Racaptor tranaformauoa u dafinad in thu report aa an alteration in the physiochamical nature of tha ligand-recaptor complex, which reeulta in an tncreaaad affinity for ONA or other poiyamoruc matncaa. ABBREVIATIONS: TCOO. 2 .3 .7 ,8-tetrach*orooi0enzo-D-dBxei; M O PS. 3KA#-morpftolino)propaneaufonc sod; K0. equRbrtum dtsaoaatm.'i ccnsts-tu (LJ. concentration of unoound ligand: (/?], concentration of unigandsd recep to r [LR], concentration of receptor-kgand complex; (L R U necsetcv- iigand com plex at squRtmum; (L R ']. concentration of activated RL com plex : (L ]r. total raotofegsnd concentration; (R ]f . total receptor concsnuxocvt: receptor concentration per ml: second-order rate constant of association; fast component ckssooaoon rate constant; k . t . sievv com ponent ckssoaatxxi rate co n stan t HPLC. htgn perform ance kqud enrom etograpny; TEA. tnetnyiemme; f,, , natf-We of dissociation; ASFFT. 4 0 - 5 5 % ammonium sulfate preoprtete: BCA. bonenorw se aod: BSA. txjveie serum aldumei; DM SO . dimetnyl sulfoxide.____________________________ C t F .N P 011137 \ 230 SradfteM a/. Stokes radii, molecular weights, and frictional and axial ratios (9) are all within the ranges described for steroid hormone receptors (e.g.. for the low salt aggregate form of steroid hor mone receptors the ranges reported are: sedimentation coeffi cient. 8-10 S; Stokes radius. 7-10 nm; molecular weight, 2-3.5 x 10'; frictional ratio. 1.45-2; and axial ratio 8-20; for a complete listing see Ref. 10). We recently reported the synthesis and use of an 1-uI-labeled azidodibenzo-p-dioxin as a photoaifinity ligand for the A h receptor (11). The use of '-'Mabeled dioxin congeners is ap pealing because of their higher specific activity and counting efficiency compared with tritiated ligands. We now report the synthesis of a reversible radioligand, (l*sI]2-iodo-7.8-dibromodibenzo-p-dioxm. and its use to characterize the ligand binding kinetics of the munne hepatic A h receptor. Materials and Methods R eag en ts TCDD was a gift of Dow Chemical Co. (Midland, MI); 2.3,7,8tetrachlorodibenzoturan was a gift from Dr. David Firestone (Food and Drug Administration. Washington. D.C.). Activated charcoal, grade PX-21. was a gift from Amoco Research Corp. (Chicago, IL). BactoGelatm was from Difco Laboratories IDetroit, MI). Magnesium sulfate was from Fisher Scientific Co. (Fair Lawn. NJ). Ammonium sulfate (ultra-pure was from Schwarz/Mann (Cambridge. MA). Methanol (HPLC grade) was from Burdick and Jackson Laboratories, Inc. (Muskegan. MI). Glycerol was from J. T. Baker (Phillipaburg, NJ). EDTA was from EM Scientific (Cherry Hill. NJ). Camer-free N a`" I (NE2033L) was from New England Nuclear (North Billenca, MA). BSA, dithiothreitol, J-mercaptoethanoi. sodium azide, chloramine T. MOPS (free acid and sodium salt). TEA. and TEA hydrochlonde were pur chased from Sigma Chemical Co. (St. Louis. MO). Dichioromethane (reagent grade), tetrahydrofuran (anhydrous. 99.9% pure), n-butyl ni trite 97% pure), p-dioxane (anhydrous 99-*% pure), and dimethyl sulfoxide anhydrous. 99% pure, stored under N? gas) were purchased from Aldrich Chemical Co. (Milwaukee. WI). BCA Protein Assay Reagent was purchased from Pierce Chemical Co. (Rockford. IL). B u ffe rs MN represents the stock buffer which contains 25 mss MOPS and 0.02% sodium azide. pH 7.5 (at 4*). MEN is the stock buffer plus 1 mM EDTA. MdENG and MDENG represent MEN with the addition of 10% (w/vi glycerol, plus the addition of either 10 mM d-mercaptoethanol or 1 mM dithiothreitol. Radiosynttiaaia of ( ,I*l]2-4odo-7,8^1ib romodib f iz o -p-dioxin The synthesis was performed in a room dedicated to radioeynthesis, equipped with a high flow chemical hood, with the use of sealed vials and airtight synnges to minimize escape of radioiodine. The enure procedure was monitored with a y-radiation detector. S ynthesis o f (l u T]2>amino>3iodo>7,&<iibroaKMiibeiizo>/>* dioxin (Fig. I). To 5 mCi of carrier-free Nam I (2350 mCi/ml in 6 mM NaOH. 2L5 ul of water) in e septum-sealed conical vial was added 2.5 nmol of 2-amino-7.8-dibromodibenzo-p-dioxin (II) in 25 m1 of methanol. 1.13 Mmol of sulfuric acid in 10 m1 of methanol/water (9:1), and 25 nmol of chloramine T in 5 m! of methanol (12). Tha iodination reaction was complete m 30 min and terminated by the addition of 500 Mg of sodium metehiaulfite and 5 Mmol of wxiini^ hydroxide in 35 m1 of water. The reaction product, (l,tSl]2-amino-3-iodo-7,a-dibromodibnxop-dioxin, was extracted with 250 m! of dichioromethane with the use of a vortex mixar (30 sec). The organic phaaa was aspirated into a synnge (gas syringe, pressure-locked plunger tip, P010032; Pierce Chemical Co.. Rockford. IL) transferred to a l-ml sealed conical vial (Reacti-vial 13221, Piarca Chemical Co.) containing 10 mg of magnesium sulfate. 75 m1of dichioromethane. and a microstimng bar. and tha mixture w&s stirred for 45 mm to dry the organic phase. (l,AIl2-iodo-7.8-dibromodibenzo-p-<iioxm was formed from [ll,I]2amino-3-iodo*7,8.dibromodibenzo-p-dioxm by generation of the aryldiazonmm salt, which produces the aryl radical with subsequent ab straction of hydrogen from tetrahydrofuran to form tha daamm&ted product (13). This was accomplished as follows. The dried dichlcromethane solution was aspirated and transferred to a 1-ml conical vial containing 250 ul of anhydrous tetrahydrofuran and 67 Mmol of n-butyl nitnte. Tha reaction mix was then heated at 55* with stirring. After 15 mm. the vial was opened and the solvent evaporated, with heating, undar a stream of nitrogen gas. venting the vial into a funnel 2 cm diameter) attached to two serially connected charcoal filters in line with negative airflow. This system trapped most of the volatile radioiodine in the first charcoal filter and minimized release into the labo ratory hood ventilation system. The reaction mix solute was dissolved in 50 m1of methanol/water (92:8) and the ``JI-labeied product punned by HPLC aa detailed below. Analysis and Purification of ttia Radioligand Unlabeled 2-iodo-7,8-dibromodibenzo-p-dioxin. synthesized as above on a milligram seals, was used aa an analytical standard. The unlabelec compound was charactenzed by 1) low resolution mass spectrometry: m/e (more than 40% of the baa peak) 467.9 (100% M*), 469.9 (46%), 466 (50%), 71 (55%). and 43.1 (60%); and 2) reverse pnase HPLC; using a C-L8 column, 4.6 x 250 mm. 5 Mm particle (Altex Scientific Inc.. Berkeley, CA). and an isocratic solvent system (methanol/water. 92:8) with a flow rata at 1 ml/min. The compound was detected by its UV absorbance at 254 nm and eluted with a retention time of 24 mm. Tha radiolabeled reaction mix was subjected to HPLC separation as above. Approximately 60% of the radioactivity injected eluted with a UV absorption peak that had a retention time identical to that of the unlabeled compound (i.e., 24 mm). In this methanol/water HPLC system, tha potentially significant reaction product contaminants. 2ammo-7.8-dibromodibenzo-p-dioxm, 2-iodo-3-amino-7,8-dibromodibenxo-p-dioxin. and 2.3-dibromodibenzo-p-dioxin. elute with retention times of 6.8. 11.6. and 13.0 min, respectively. Reinjection of the 2<-tmr. tiuate (from above) also cochromatographs with authentic 2-ioao-T.Sdibromodibenxo-p-dioxin using an 9:1 acetonitrila/watar isocratic sol vent system (retention time, 22 min; 98% radiochemical punty). The overall yield of radioligand, judged by incorporation of `" I into tb.\= purified product, rangea from 40 to 50%. The specific activity was aaeumed to be equivalent to that of earner-free NalMI, (2178 Ci/mmel. 4.831 dpm/fmol). The radioligand was stored at 6 x 10* dpm/ml in pdioxane in a capped teat tube protected from light. For use in binding experiments, an aliquot of tha stock solution was dried under a stream of nitrogen gas and dissolved in DMSO at tha appropriate concentra tion. Rcaptor Preparation Cytoaoi. C57BL/6J mica were purchased from The Jackson Labo ratory (Bar Harbor. ME) and bred in our laboratory. Adult male ta d femaia mice were killed by cervical dislocation and their livers were removed, rinsed with ice-cold KC1 (150 mM), homogenized in 9 volume*, of MdENG buffer, and centrifuged at 10.000 x g for 20 min at 4*. The poetmitochondrial supernatant was carefully removed to avoid contam ination by tha surface lipid layer and subjected to centriftigerion tu 105,000 x g. The supernatant (approximately 10 mg of protein/ml) w*s 1. Radtoaymnesi* of ['**1)2-080 7.8-diPrornod*)enzo-p-0oxjn. THF. tatraftydrofuran. Ah Receptor Kinetics 2 separated from surface lipids and the microsomal pellet and stored at- Software -80* for use in binding experiments or until processed further. A S P P T . The frozen cytosolic fraction was chawed and placed in an ice bach with slow stirring. A saturated solution of ammonium sulfate in MEN buffer waa added to the cytosolic preparation, to a final Weighted nonlinear curve-fitting estimates of binding peramet were determined as described (15, 16) using a commercially avails software package for the (BM-PC (Kinetic, EBDA, Ligand. Low Elsevier Biosoft, Cambridge, England). concentration o f 40%, over a period of 30 min and stirred for an additional 30 min. The solution was centrifuged at 10,000 x g for 20 Results min. and the supernatant fraction waa removed, slowly brought to an Equilibrium binding analysis. Equilibrium saturati ammonium sulfate concentration of 55%. stirred for an additional 30 binding of {wsI2-iodo-7f8*dibromodibenzo-p-dioxin to the l min, and centriAigcd at 10,000 x g, and tha supernatant waa discarded. patic Ah receptor of CS7BL/8J mice, determined at thi The ASPPT waa resuspended in 55% ammonium sulfate in MdEN, different protein concentrations, is shown in Fig. 2. Analysis and aliquots equivalent to 0.5 g wet weight of liver (15 mg of protoin) saturation binding curves by the method of Scatchard (17) were placed in 12 x 75 mm borosilicate tubes (VWR Scientific, San presented in Figure 3 and Table 1. With progressive 2-fc Francisco, CA) and spun at 5000 x g. Tha supernatant waa removed dilutions of the receptor preparation, the estimate of maxim by careful aspiration, and the pellets were stored In stoppered tubes at binding sites per ml, Bmtt (x axis intercept), shows appro; -80* until use, mately 2-fold decreases, i.e., Bm/rag of protein remains co HPLC-anion exchange* nriched receptor preparation. HPLC-anion exchange chromatography waa performed on a Mono-Q HR 5/5 column (5 x 50 mm; Pharmacia. Uppsala, Sweden). All column parts, external, tubing, injection loop, and buffers were placed on ice during the entire run. The HPLC instrumentation was identical to that outlined in the purification of the radioligand (above). The sample (ASPPT fraction, 200-900 Mg of protein) was dissolved in 1-2 ml of 20 d m TEA buffer pH 7.5 (4*) and injected into a 2-ml sample loop. The elution buffets used Were the following: buffer A, 20 d m TEA. pH 7.5; and buffer B, 20 d m TEA, pH 7.5, plus 500 OM NaCl. The linear gradient program ran from l(XhO (A:B) to 40:60 over 8 min, then to 10:90 over 10 min, followed by isocratic elution (10:90) for 5 min. The 16-18-min fractions were collected and pooled for use in binding studies. These fractions appear at the beginning of a large (JV-absorb ing peak and'thus some variation in purification waa observed between runs. Protein concentrations were determined by the method of War burg and Christian (14) or by the use of BCA Protein Assay using BSA as standard. 1 I Binding Experiments stant (Table 1; Fig. 3). Unexpectedly, the slopes of the Sea chard plots (--l/Ko) increased from 2.6- to t.2-fold with ea< 2-fold dilution (i.e., the estimate of Kn decreased from 1.6 10~10 to 1.2 x 10-u M, as the protein concentration decreast from 1160 to 36 ttg/ml). Hill coefficients were 1.0 at all receptor dilutions, indicator that aite^aite cooperatively did not contribute to the increa; in the slope of Scatchard plots observed with dilution 113 Scatchard and HiU analyses of saturation binding data we; reproducible and of good linear fit, with coefficients of variatic below 30% and with correlation coefficients of at least Oi (Table 1). Regression analysis of the relationship between the procei concentration of the ASPPT and the estimate of K0 as detet mined by Scatchard analysis suggested a simple linear functio; (Fig. 4). Extrapolation of the least squares fit of the line to th y intercept (infinite dilution of protein) yielded an estimate c K0 of 6 3 x 10'12M. ' Two experiments were conducted to describe more clear! the relationship between protein concentration and the in Binding experiments were carried out at 4*. The frozen ASPPT creosf' in apparent binding affinity that was observed witl pellets or cytosolic fractions were dissolved and diluted in ice-cold dilution. First, when the Ah receptor concentration mi MOENG buffer to the appropriate protein concentration. T he radioli of protein) in the ASPPT fraction was enriched 5- to 9-fold b gand and the! competing unlabeled ligand (2,3.7,8-tetrachlorodibenzo- HPLC-anion exchange chromatography, the.estimated Kn furan) were dissolved in DMSO and added to the bufftr/receptor i approximately 4-fold lower than the K0 determined on tht solution so that the final concentration of solvent waa 6 pi o f DMSO per ml of buffer. Binding reactions were performed in 1.0-mi volum es/ (12 x 75 mm borosilicate test tubes), or in 50-mi volumes (Ehrlenmeyer flasks), from which 1.0-ml samples w e n transferred to the smaller tubes. The binding reaction waa terminated by the addition of a OJS-mi volume o f a s ice-cold charcoal/geletin suspension (3%/0.3% in MN buffer). The suspension eras then stirred vigorously on a vonax mixer corresponding ASPPT fraction (Table I; compare,,20jag /m wet weight values). Second* saturation binding was describe on an ASPPT preparation (30 pg/mi) in the absence or piesenci of 70 Mg/ml of a highly purified protein, BSA (Fig. 5). Scatchsrx analysis of the binding data described a 10-fold increase in the apparent K0 when BSA waa present. Despite the large effect 01 (2 sec), followed by incubation far 10 min at 4*. The charcoal/gviatin BSA on the apparent K0, the estimate of STM. was only slightly waa then sedimented at 2000 x g for 10 min at 4*. 1.0-ml aliquot of increased in the presence of the additional protein, approxi the supernatant fraction o f each tube waa transferred to 12 x 75 mm mately 1.3-fold. It should be noted that this ASPPT preparatior polypropylene tube* (for higher counting efficiency, 75%), end the waa from a different batch than that used to generate the data bound radioligand waa quantified in a MINAX1 Series-5000 ^-counter In Table 1 and Figs. 2, 3, and 4. This preparation yielded e (United Technologies/Packard Instrument Co., Downers Grove. IL). slightly greater receptor concentration per milligram of protein. Sample counting times were adjusted to achieve a counting error of Binding kinetics. For a simple bimolecular reaction gov 3%. \ erned by the law of mass action, the Ko of ligand binding is Total radioligand, [LJt, waa defined aa tha concentration of radioli equal to the rate constant of ligand-receptor dissociation di gand in solution after the 16-hr incubation. Total radioligand bound' was defined aa the radioligand in solution after charcoal adsorption. vided by tha second-order rate constant of ligand-receptor association: Nonspecific binding waa defined aa the amount of radioligand bound R + L*=tRL (l) in the presence of a 200-fold molar excese o f 2A?3-tetrachlorodiben* toftiran. Specific binding wae defined aa the difference between total and radioligand bound and radioligand oonspecifically bound. Unbound ("free'*) radioligand waa defined ea the difference between total radi oligand and total bound radioligand. , Co m-r1i (2) O P A T T > n h i* > 232 B ra d fM d * 783951 o t t n i p m 1 1 4 .0 Tout RadkttiQsnd (QPMfml tn Thousands) Fig. 2. Equilibrium binding of [11M]2*iodo*7,8-dIbramodibenzo^i-dioxin. The receptor preparation (ASPPT. 36 (A), 145 (8). or 580 (C) ng of protein/ mi) was incubated at 4* tor 16 hr, with concentrations of radioligand ranging from 2.5 to 250 fmol/mi. Total, specifically bound, and nonspeaficaily bound radioligand were determined as outlined in Materials and Methods. Each point represents the average of duplicate determinations. Fig. 3. Scatchard analysis of equilibrium binding data generated at vanous dilutions of receptor. Equilibrium saturation binding curves of ['" l]2-iodo-7.8-dibromodibenzo-0-dioxin were determined at vanous di lutions of the ASPPT fraction (as descnbed in Fig. 2). The binding parameters K0 (-1 /slope of the line) and S-. (receptor concentration/ mi; the i intercept) were calculated from a weighted least squares fit of a plot of speaficaily bound radiotigand/free radioligand versus specifically bound radioligand (18, 17). When more than two plots were generated at a given protan concentration, me plot that yielded the median Ka value was chosen tor presentation. Eq. 2 provides an estimate of tha K& independent of that calculated from equilibrium binding studies. Association ra te constant. Tha rate constant of associa tion was calculated from tha tin co u n t of ligand-receptor binding, via the integrated second-order rata equation, aa da- scribed by Weiland and Molinoff (19): 1J [ L R l [ [ L I r - ( L R ( L R M R l r i l 1" [L M IL R , - iL R J ] J (3) Substitution of InV for the left side of the equation and re arrangement yields the equation of a straight line, with slope In Y/time. Thus, kt can be calculated bom the relationship: \m This method allows determination of kx from estimates of the concentrations of total ligand, [L]r, total receptor, (R]r (ob tained from Scatchard analysis). Ligand-receptor complex at TABLE) Summary of saturation binding experim ents perform ed on th e Ah receptor at vanous protein conc e ntrations The saturation bnding isotherms for the radioligand and Aft receptor in hepatic cytosol, the ASPPT fraction of cytosol, and the HPLC-enncned preparation ot the ASPPT fraction were determn ed over a range of protest concentrations. The binding parameters K0. and were calculated by Scatenard analysts of binding data as descnbed in the legend to Fig. 2 or by HHI analysis of boding data as descnbed (IS ). Values represent the mean the standard davattan. except m the case of a angie data pant or duplicate cats points, for which onty me mean is given. Correlation coefficients were at least 0.99 for bath Scatchard and H i plots. The standard deviation of the Ha coefficient did not exeeed 0.02. EquviMnl wet wwgit PlUtfll Ke flU hi covftawit Nixnoer of eioenments mg/n> ASPPT 1.25 2.5 5 10 20 40 HPLC-Enricried 6.7 14 20 30 Cytosol 0.62 1.25 2.5 Mfl/fflf p* 36 12 72 14 4 145 19 3 290 49 9 580 92 1160 160 8 11 9 12 25 22 46 22 42 7 85 13 170 24 fmor/mgofprawn 141 114 21 110* 14 121 7 121 131 780 811 664 1100 95 113 80 1.0 1.00 0.97 0.98 0.98 0.98 1.00 0.99 0.98 0.96 1.00 1.00 0.99 1 5 4 3 1 2 1 1 1 1 1 2 1 equilibrium, [#], and ligand-receptor complex, [ZJZ], at var ious time intervals, t. The advantage of this method is that no constraints are placed upon the concentrations of ligand or receptor used in the binding studies. By this method, dilution of the protein concentration by 18-fold resulted in a 37-fold increase In the calculated rate constant of association (kx- 7.6 x 10* to 3.0 x 10' M"1 hr"*; Fig. 8, A and B, and Table 2). In contrast, at s fixed protein concentration a 13-fold change in total radioligand concentration did not alter the association rate constant (Len kx m 3.0 end 2.6 x 1010 M"1 hr"1; Fig. 6, E and C, and Table 2). The association rate constant was also estimated from the initial binding velocities (vo) under condition! in which [L)r and (R]r ^ [LR], so the free radioligand and free receptor concentrations are approximated by [Jr and [R]t, respec tively. 783952 Ah Acceptor Kinetic 233 [A1t * I.SxlO^M; [L]t a 3.1X1" M Rg. 4. The apparent Kg as a function of protein concentration. The apparent Ka. calculated from Scatchard analysis of radioligand binding to the ASPPT fraction, plotted as a function of the protein concentration (Mg/ml). The data were fit to the equation of a straight line by the method of (east squares. The equation of the line is inset, r. correlation coefficient. [AL, 8-axl" M; IU1T- 3.1x10" M IA1t - 3^x10'] M; IL)t s 4.U 1 0 i1M BOUND (fmol/mi) Rg. 5. The effect of BSA an the slope of the Scatchard plot. EqusUbhurn saturation binding curves were generated, as described in Rg. 3, for an ASPPT sample of 30 ug of protem/ml vwth and without the addition of 70 Mg/ml BSA. Calculated binding parameters; piua BSA. K0 - 6.3 x 10'" M, 8mm 9.2 fmoi/mf; without BSA. Ko m 7.3 x 10"* m. 8mm 7.4 fmoi/rrx. where: Fig. 8. Time course of aasocuition of [,Mf]2-todo-7,8dibramodibn;>t^ diown to the hepatic Ah receptor at 4s: estimate of the association rate constant. The time course of ligand receptor association was mannered at receptor concentrations of 1.5 x I0"'a u and 8.2 x 10~'S w (ASPPT fraction equal to protein concentrations of 1160 and 72 Mg/ml) and radioligand concentrations of S.1 x ,i0 "Mm and 4.1 x 10"** m (250.000 and 20.000 dpm/ml). Each point represents me average of two deter minations. The data are presented as plot of bound radioligand as a function of time (A1. B1, and C1), and as a plot of inY (left side of Eq. 3) as s function of time (A2, B2. and C2). The rats constant of association wee detemxned from the slope of tnT versus time plots as described in Eq. 4 and is inset (units of u " h r'). ,o 4^ rearranging gives: Bo * Ai [ R l r i L r l (5) At v 0/ [ R ] r ( L ] r (6) Plotting initial binding velocity aa a function of receptor or radioligand concentration, with the other variable fixed, allows the determination of kt. A plot of initial binding velocity versus receptor concentra tion yielded a hyperbolic plot (Fig. 7A). The correlation coef ficient of the plot (for a linear fit) using all data points pre sented is 0.94. By elimination of the values obtained above a receptor concentration of 1 x 10"11 m and inclusion of the theoretical y intercept of (0,0),-s correlation coefficient of 0.99 is obtained and an association rate constant of 5.5 x 10` M*1 hr"1is calculated (i.e,, calculation of kx from the asymptote of the hyperbole). The initial binding velocity aa a fraction of radioligand concentration, at a fixed receptor concentration of 6.8 x I0~u M, is shown in Fig. 7B. This plot yielded a good linear fit and an estimate of the second order rate constant of association equal to 5 x 10' M"1hr"1. D issociation ra te constants. As shown in Figs. 8 and 9, the dissociation of the ligand-receptor complex does not follow a simple first-order process. The disaociation profile is inde pendent of the receptor concentration or the type of prepara tion. The data is best fit by s two-component exponential curve (y A*-'* + Be"-*).3The initial rapidly dissociating component comprises approximstsly 75% of the total complex snd 'has a 1Our reculta indie* (hat a atatiatieally batter lit waa obtained by csinp t biexpoaantial modal ovar a monoexponentiai ana ( p < 0.001). We used the* partial F taat aa daacribad by Munaon and Rodbaid (IS). No further improvement was obtained whan attetapta warn made to lit cha data to a triaxponantiai moda* i p < 0.98), altboufh chia and higher order exponential modala cannot hi ruled cut by cha data pmaatad 234 B redfleM H /. TABLE 2 binding has appeared. The kinetic and equilibrium studies in Summary of kinetic data this report challenge the following two assumptions implicit in Rat* constants war* caicuiataa as oescnoea n me f**t ana exoresaaa as me mean ana. wnerw 501* 0* me stanaara arror Tha equwonum association con* stant. *0. was aatarmnaa Dy 1) Scatcnara analysis of tna saturation omcrvj sotnarm at mfimta anution iF<g. 3) or 2) as a ratio of tna ansooaoon rata constant. (0 36 nr") ana tna association rata constant from aimer tna mtagrataa rata equat:on or tna initial anqing vewaty rnamoa The dissocstion rata constants prasantad aoova nava not oaan corractaa for recsotor-ngana comtsax oestrueoor rates, tnus tnay are tna sum of dissociation ana destruction ritas isaa Results). past binding studies: l) that both specifically bound radioligand and free radioligand concentrations can be determined accu rately at high protein concentrations and 2) that ligand-A/i receptor binding can be described by a simple reversible bimo- lecular model (Eq. 1). The most common method of analyzing equilibrium satura i. tion binding data is the linear transformation of Scatchard 9# **' nrm< (17): Association rate constant integrated 2nd-order initial velocity method Ugand Receptor 4 1 x 10"'* 8.2 x 1 0 "* 5.1 x 1 0 "' 8 .2 x 1 0 "* 5.1 x 1 0 '" 1 5 x 1 0 " 2.1 x 10_" Varied Varied 6 .8 x 1 0 "* zrctrn 2 .8 x 1 0 ' 3.0 x 10' 7.8 x 10 5.5 x 10' 5 .0 x 1 0 ' Pwcant or [LB] _ [R]r - [LB] (LI * Kd [Bound] [Free] 77" [Bound] + A.0 (8 ) Dissociation rate constant ASPPT Cytosol Warmeo cytosol nr" 0.37 0.02 0.38 0.05 0.36 r 0.04 78 4 77 7 72 5 nr" x 10* 4 3 0.4 9 4 0.8 7 2 0.5 22 1 23 2 28 1 *0 A plot of the ratio of specifically bound ligand/free ligand as a function of specifically bound permits one to estimate K0 and total binding sites, [B]r- Analysis of equilibrium binding data by the Scatchard equation requires that a number of criteria be met as follows: 1) the concentration of specifically bound and free ligand are measured accurately at equilibrium; 2) both Equilibnum dissociation constant Scatchard (infinite dilution) Kinetic (initial velocity) Kinetic (integrateo 2nd order) 8.0 3 x 1 0 "* 8.9 x 1 0 "* 13.0 x 1 0 "* ligand and receptor are homogenous species; 3) binding obeys the law of mass action as described in Eq. 1 and 2; and 4) no cooperativity exists. C riterion 1: M easurement of free and specifically bound ligand is accurate. The accuracy of K0 and B*,, dissociation rate of 0.36 hr"' (f., * 1.9 hr). The second slower values generated from Eq. 8 is dependent upon the accuracy of dissociating component (25%) has a rate constant of 4.3 to 9.4 the determination of bound and free. Estimates of bound radi x 10"' h r '1 (f,, 74-161 hr) (Table 2). The rate constant for oligand are fairly straightforward and can be supported by "degradation"' of the receptor-ligand complex at 4* is -2 x physicochemical characterization of the ligand-receptor com 10" h r '1, monoexponential for over 120 hr. and not signifi plex (e.g., density gradient centrifugation, size exclusion chro cantly affected by protein concentration. Thus, the second matography, anion exchange chromatography, etc.). Con component of "dissociation" is a composite, representing both versely, the concentration of free radioligand is most often ligand-receptor dissociation and degradation. The rate constant determined indirectly as the difference between total radioli for degradation of the unoccupied receptor is - 4 x 10"J hr"1 gand in solution and total bound radioligand. In Ah receptor- and is monoexponential for over 100 hr. ligand binding studies, free is defined operationally aa the As shown in Table 2, Kn values calculated aa the ratio of amount of radioactivity that can be adsorbed to charcoal (20), kinetic constants. k .J k x (where It-, is the dissociation rate that can be removed from hydroxyapatite by extensive deter constant of the major rapid dissociating pool and kxis estimated gent washing (21), or that does not sediment in a sucrose from the integrated second-order equation or initial velocity gradient (22). By these methods, radioligand, bound to low method), are 6.9 x lO"* and 13.0 x 10"`* M, which agree very affinity nonspecific binding sites, that dissociates during char well with the Kn determined by equilibrium saturation data coal adsorption, hydroxyapatite washing, or long-term centrif extrapolated to infinite dilution, 6.0 x 10"'3 M. ugation will be misclaaaified aa free radioligand. This systematic Discussion misclaaaification will decrease the ratio of bound/free ligand and hence decrease the slope of the Scatchard plot (i.e., -1 / Due to its high receptor affinity, high biological potency, commercial availability, and environmental importance, TCDD has become the most studied Ah receptor ligand. Despite exten sive characterization of the Ah receptor using (3H]TCDD or other ligands, no detailed kinetic analysis of ligand-receptor Ko). although it may not significantly compromise estimates of Bm, (18, 23, 24). One would expect the misclaaaification of nonapacifically bound radioligand as free radioligand to be greatest, and thus the overestimation of K0 to be greatest, at higher protein concentrations. We believe this is the most plausible explanation of the data presented in Figs. 3-5 and 1TK* degradation rata of tha ligand-receptor complex waa determined by ncubeuon of the ASPPT fraction (72, US. and 290 *4 of protein/mi) with radioligand with or without 00-fold eiceaa J.7,9-i*trachioredib*nxoi\iran at 4* for 16 hr (aa deecnbed in tha legend to Fig. 9). SpwnAe binding waa determined on aliquot* removed at time intervale from 16 to 136 hr. Tha degradation rata of tha unocr- pied receptor aa eat 1mated by sampling a receptor preparation tored at 4* attar various time interval*. The receptor sample waa incubated with radioligand with or without 00-fold escea* 2.3.7.8-tetrachlorodibentofuran at 4* for 16 hr. and pacifically bound radioligand waa determined. The degradation rates were estimated from the plots of the log of specific binding versus urn*. Table 1. Using [3H]TCDD, Farrel and Safe (25) have noted a rela tionship between protein concentration and K0 estimates gen erated from Scatchard analysis of saturation binding <kus_ Cheracterixation of Ah receptor ligand binding with a radioli gand of modest specific activity, such as [3H]TCDD, necessi tates the use of protein concentrations greater than 1 mg/rol (20-22, 25). Thus, Ko values generated previously for (3K? OFTsTP 0 1 i i/io Ah R eceptor K in etic* 235 -4 CP CO CcDn -> Fig. 7. Imtiai binding velocity of [,15l]2-iodo-7 8-dibromodibenzo-o-dwxm to tne heoatic Ah receptor estimate of the association rate constant. The ASPPT fraction m MDENG buffer was incubated with radtoiigand m 50-mi Ehrtenmeyer flasks. At the indicated times intervals, i 0-mi aiiguots were transferred to borosilcate tuOes containing 1000-fold moiar excess of 2.3,7.8-tetrachkxodibenzofuran (800-foW for nonspecific binding) to prevent further binding of the radioligand and total, specifically, and nonspecifically pound radioligand was determined. Receotor concentrations w ere determined by duplicate Scatchard analyses. A. initial binding velocity as a function of Ah receptor concentration. Receptor concentrations of 1 7 x 10*'* m (G3). 3.4 x 10" 'a wi (), 8.1 x 1 0 ''* u (If). 8.5 x 10"'* m (0). 1.7 x 10"" u <), and 2.4 x 10*" m () were incuPated with a fixed radioligand concentration. 2.7 x 10*" m. The mitial binding velocity was determined for each receptor concentration from the swoe of the plot of specifically bound radioligand as a function of time (m stt). The initial binding velocities were then potted as a function of receotor concentration. The rate constant of association was caJculated from tne stope of the linear portion of the line (asymptote of the hyperbola) divided by the ligand concentration (see Materials and Methods). B. Initial boding velocity as a function of radioligand concentration. A fixed concentration of receptor. 6.8 x 10*'* u. was incubated with radioligand concentrations of 5.6 x 10*'* m >3). 1.1 x 10*" m (O). 2.0 x 10~" m (), and 3.1 x 10*" m ( ). The initial binding veKxnty was calculated from the slope of speaficaify bound radioligand as a function of time (<nser). The initial velocity was blotted as a function of radip<igand concentration. The rate constant of association was calculated from the slope of this ime divided by the receptor concentration. TCDD are probably overestimates. Using a competitive binding assay with (l,:''I]2-iodo-7.8-dibromodibnzo-p-<iioxin as the ra dioligand and low protein concentration, we estimate the Ko for TCDD to be approximately 4 x 10"l* M.4 Misclassification of nonspecificaily bound radioligand as free radioligand has been described in a number of other receptor systems that bind hydrophobic ligands: estrogen (23), proges terone (24), and thyroxine (26). In these systems, misclassifi cation errors have been minimized by the use of methods that better estimate free and bound radioligand, e.g,, equilibrium dialysis (23, 24) or centrifugal ultrafiltrmtion-dialysia (27). Be cause our radioligand adsorbs strongly to dialysis tubing and filter membranes,5 we provided support for radioligand misclassification by the following: 1) demonstrating that addition of a purified protein (BSA) decreases, whereas receptor enrich ment increases, the slope of the Scatchard plot (Fig. 5 and Table 1), and 2) eliminating the possibility that the rate con stants of association or dissociation were altered by dilution. We observed that, like /Co, the apparent association rate constant also varies as a function of the concentration of receptor, or more likely, protein (Figs. 8 and 7A). Association rata constant estimates are also dependent upon accurate de termination of free radioligand. From the initial velocity rate equation (Eq. 6), it follows that for a given ligand concentration a plot of u0 as a function of receptor concentration will yield a linear plot, with a slope equal to k,(L]r. The hyperbolic shape | C. A. BradlWId tad A. Poland, manuscript in preparation. C. A. BradAaki. tad A. Poland, unpublished observations. of the curve generated (Fig. 6A) indicates that ki[L]r is decreas ing as the concentration of the receptor preparation increases and supports the idea that free is being overestimated at higher receptor concentrations. An alternative explanation is that k, is actually increasing with dilution of the receptor. Because k x. is a function of diffusion rate and the energy of activation for the binding step (28), an increase in k x with dilution implies that a decrease in the energy of activation for ligand binding is also occurring with dilution. A decrease in the energy of acti vation for the forward reaction should also result in a decrease in the energy of activation for the dissociation reaction and lead to a increasing k_, with dilution. Because k_, is not affected by protein concentratipn (Fig. 7 and discussion below), the hyperbolic nature of the curve depicted in Fig. 6A is best explained by an overestimation of free radioligand at the higher protein concentrations. C riterion 2: Both ligmnd and receptor are homogenous binding species. The time course of ligand-receptor dissocia tion is biphaaic, suggesting two ligand-receptor species that have dissociation rate constants differing by a factor of at lease 60-fold (Figs. 8 and 9 and Table 2). The dissociation of pH] TCDD from the rat Ah receptor has previously been described as ``irreversible'' (25). Our preliminary comparisons bc-cweet the dissociation kinetics of ths Ah receptor-radioligand com plex in the Harlan-Spragua Dawley rat and C57BL/6J mouseindicata significant differences.5 Criterion 3: Binding obeys the law of masc action as described in Eq. 1. The linear functions generated from plots of initial binding velocity ( v0) versus either the ligand c . 783955 230 BradfMd * 0 20 40 60 80 HOURS Fig. 8. Dissociation of tna ('" i]2-KX>o-7.8-dibromoditoenzo-p--dioxin Ah 'eceptor complex at 4* m effect of protein concentration. Radtongand'eceptor complexes were preformed by incuoatxxi of radioligand and receptor for 16 hr at 4* m 50-ml Ehrlenmeyer flasks. Fractional receptor occupancy of rad<xgand was between 15 and 18% for ail protein concentrations. This was achieved by the following proten (receotor) and radioligand concentrations: 1160 **g/mi (150 pm ) and 62 p m ; 290 Mg/mi (35 pM) and 15 pM; and 72 g/mf (8.2 pM) and 4 pM. After eguilibnum binding had been obtained, a 1000-fold excess of 2.3.7.8tetracniorodibenzofuran was added to prevent reOmdmg of radioligand (time zero). Bound radioligand was determined, at the indicated times, on 1 0-mi aliquots. Nonspecific binding was determined by the addition of a 200-fold excess of 2.3.7 8-tetracmorodiOenzofuran during the prebinotng stage, followed by the additon of an 800-fold excess at time zero. All values were determined m duplicate. Results are presented as specifically bound radwsgano/radioiigand specifically bound at time zero, plotted as a function of time. 1.0 < 1 3az 0.1 03 O3z 03 0.01 ` a CYTOSOL . * HEATED CYTOSOL - ASPPT 1l ' i 0 50 100 a ' 150 HOURS Fig. 9. Oissoaatwns of f^ -io d o -T .S -d lO ro m o d ib e n zo o ^ to ja n from different receptor preparation at 4*. Dissociation rates w ere detsrm ned as described m Fig. 6. Cyroeof (85 ng of proten/rm ) and ASPPT (72 *g / mi) w ere premcuoated with 10 pM radwagand for 15 hr at 4* before dissociation rate determn a oons . beared cytotoi was prmncuPated for 18 hr at 4# as aoove. foeowed by a 30-m m incubation at 3 0 *. then recoded to 4* (30 m n) before (association rate detemsnatxsn. Kinetic rate constants w ere calculated by the nonsneer c u v e fitting technique, constrained to a dexponerm ai m odel, as descnbed (16). receptor concentrations (Fig. 7) indicate that the binding re action is first order with respect to both radioligand and recep tor and thus second order overall (formation of the ligand receptor complex) (28). This lends support to th# assumption that one molecule of ligand binds to one molecule of receptor. Biphasic ligand-receptor dissociation is not in agreement with the equilibrium binding scheme presented in Eq. I. Bi phasic ligand-receptor dissociation curves have been observed for ligands bound to the estrogen (29, 30). progesterone (31), glucocorticoid (32), and androgen (33) receptors. For these steroid hormone receptors, the slow dissociating species (be lieved to represent a "transformed" receptor species) has an increased affinity for polyanionic matrices such as ONA (3335) and an altered molecular mass aa compared with the fast dissociating species (36, 37). In these systems, in which the ratio of the fast to slow component can be altered by increasing the ionic strength, heating, etc., we did not observe a significant change in the ratio of the fast and slow components by heating a 30* for 30 min. dilution, or ammonium suifate precipitation.' Ligand binding to the Ah receptor appears to conform to the model proposed for the steroid hormone receptor (32) in which; L + R *=c LR LR' *=x L + R' 9( ) and thus at equilibrium: [L] [R] 0 m [LR] + [LR'] ( 10) C riterion 4: No site-site cooperatively exists. Hill coef ficients were 1.0 at all dilutions; there is no indication of cooperativity in this system. Hill coefficients of approximately 1.0 have previously been reported for a number of mammalian species (25, 38). In this report, analysis of the Ah receptor-ligand binding, facilitated by a new high specific activity radioligand has re vealed important findings. 1) Estimates of the apparent K0, as determined by Scatchard analysis, are a function of protein concentration and attributable to a systematic overestim&ticn of free ligand. The true K0 (from equilibrium saturation at infinite dilution and ratio of kinetic constants k . J k x) is ap proximately 100 timet lower than previous estimates. 2) The Ah receptor ligand complex can exist as two distinct species, which have different diaeociation rates, and thus ligand binding does not conform to a simple reversible mass action model, as has been assumed previously. B efew oea 1. Poland. A., and J. C. Knutaon. 2.3.7,8-Tetrachlofodibenzo-p-<iioxin and relatad haiofenated aromatic hydrocarbons: examination of the mechanism of toxicity. Annu. Rtv. Pharmocoi. Toxicol 22:517-554 (1982). 2. Whitlock, J. P. Tha regulation of cytochrome P-450 yens expression. Anne. Rtv. Pharmacol ToxicoL 28:333-389 (1988). 3. Greenlee, W. F . and A. Poland. Nuclear uptaka of the 2J.7,8-tetrachlorod>benso-p-dioxin in C57BL/6J and DBA/2J mice. J. Biol C htm 264:98149821 (1979). 4. Okay, A. B., G. P. Bondy. M. E. Mason. G. F. KahL H. J. Eiaen. T. M. Guenther, and D. W. Nebert. Refuiatory (te e product of tha Ah Ioova: characterization of the cyiooolic inducer-receptor complex and rndanev for its nuclear translocation. J. Biol Chtm. 264:11838-11648 <1979). 5. Hannah. R. R.. J. Lund. L Poellinser, M. Giilner. and J.-A Gu&caHuwu. Characterisation of the DNA-binding properties of the receptor for 2.3.7.Etetrachlofodibenxo-p-dioxin. Eur J. Biochtm 168)237-242 119861. 8. Gaatewicx. T. A . and P. A Bauman. Heteroftnetcy of the rat Ah reetpter and evidence for transformation in vitro and ui vivo. J. Biol C htm 262:21162120 (1987). 7. Jones, P. B. C.. L K. Durrln. J. M. Fisher, and J. P. Whitlock. Control of gens expression by 2J.7.&-tetrachlorodibenxo-p-dioxin: multiple dioun-rt-- rrc T sjp n ^ ^ ^ A A Ah Rpcpqtor Kin#bcs 237 sponaive doiMUU 5'-w*rd of the cytochrome Pi-450 gene. J. BioL Chsm 201:6647-6650 (1966). 8. Giguere. V.. S. M. Hollenberg. M. G. Roeen/eki. ta d R- M. Evan*. Function*! domain* of the human glucocorticoid receptor. Ceil 46:646--652 (1986). 9 Daniaon. M. S.. L M. VU* ta d A. B. Okay. Structure and function of tht Ah recaptor for 2J.7,8-utrechlorodibeiuo-p-<iioiin../. BioL Chsm 261:39873995 (1986). 10. Yedeckis. W V. Steroid hormone receptor itructur* in normal and neoplastic cell*, in Hormonally Responsive Turnon (V. P. Hollander, td). Academic Pr**a. New York. 4-55 (1985). 11. Poland. A.. E. Glover. F H. Ebitino. and A S. Kende. Photoa/flnity labeling of the Ah receptor. J BioL Cham. 201:8352-6365 (1986). 12. Denny. J. B.. and G. Blobel. '"(-Labeled croeelinking reagent that ia hydrophylic. photoectivatable. and cleevable through an axo linkage. Proc. S a il Acad. Set CSA 81:5286-5290 (1964). 13. Cedogan. J. (. G.. and G. A. Molina. A aimpla and convenient deamination of aromatic amine*. J. Chsm Soe. Perkins Trans l 541-542 (1973). 14. Warburg. 0. and. W Chnatian. Iaoliemng und knatalliaation dee garungafermenta enolaae. Biochem Z. 3 10:364--421 (1942). 15. Munaon. P. J.. and D. Rodbard. Ligand: a versatile computerized approach for characterization of ligand-binding system* Anal Biochem 107:220-239 11980) 16. McPherson. G. A. Analyai* of radioligand binding experiment*: a collection of computer progTam* for the IBM PC. J. PharmacoL Methods 14:213-228 H985). 17. Scatehard. G. The attraction* of protein* for small molecule* and ion* Ann. .V Y Acad. S cl 51:660-672 (1949). 18 Bennet. J. P.. and H. I. Yamamur* Neurotranamitter. hormone, or drug receptor binding method*, in Seurotransmutsr Receptor Binding iH. I. Ya mamur* S. J. Enn*. and M. J. Kuhar. eda). Raven Pre*a. New York. 61-90 (1985). 19 Wetland. G. A., and P B. Molinoff. Quantitative analyst* of drug-receptor interaction*. I. Determination of kinetic and equilibrium properties. Li/t Set 29:313-330(1981). 20. Poland. A., and E. Glover. 2J.7.8-Tetrachlorodib*nzo-p-dioxin: segregation of toxicity with the AA locu* MoL PharmacoL 17:86-94 (1979). 21. Gaaiewicx. T. A. and R. A. Neal. The examination and quantitation of tiaeue cytosolic receptor* for 2.3.7.8-tetrechiorodibenzo-p-dioxin using hydroxylapatite. AnaL du x hem. 124:1-11 (1982). 22. Manchester. 0. K.. S. K. Gordon. S. L Cola* E. A. Robert* and A. B. Okey. Ah receptor in human placenta: stabilization by molybdate and characteriza tion of binding of 2^.7.8-tetrachlorodibenxo-p-dioxin. 3-methylcholanthrene. and benzol a) pyrene. Cancer Res. 47:4861-4868 (1987). 23. Suten. P. K. Receptor binding studie* Science (Wash. D. C.J 223:191-193 (1984). 24. Seiduddin. 5.. and H-. P Zeaaenheu* In vitro and in uivo enhancement of progesterone binding to the uterine progesterone receptor by cortisol. Bio chemistry 18:2829-2834 (1977). 25. Farrel. K-. and S. Safe. Abaence of positive cooperative in the binding of 2.3.7.8-tetrachlorodibenzo-p-dioxin to iu cytosolic receptor protein. Biocnem J 24 4:539-546(1987). 26. Seelig. S.. H. L Schwartz, and J. H. Oppenheimer. Limitations in the conventional analysis of tha interaction of triiodothyronine with solubilized nuclear receptor sitae. J. BioL Chem 256:2154-2161 (1981). 27. Hammond. G. L-. J. A Niaker. L. A. Jones, and P K. Suun. Estimation of the percentage of free steroid tn undiluted serum by centrifugal ulirafiltration dialysis. J. B ioL Chsm. 255:5023-5026 (1980). 28. Meitee. L An introduction to chemical kinetics, m An Introduction to Chemical Equilibrium and Kinetics. Pergamon Pres*. New York. 33-76 i ISS11. 29. Best-Belpomme. M.. J. Fries, and T. Erdoa. Interaction* entre I'oeatrtiiol et des sites recepceur* uterine. Eur J Biochem 17:425-432 (1970). 30. Best-Belpomme, M.. J. Master. H. Wetntraub. and E.-E. Baulieu. Oestrogen receptors in chick oviduct: characterization and subcetlular fractionation. Eur. J. Biocnem. 57:537-547 (1975). 31. Wolfson. A . J. Master. Y. Cheng-Ren. and E.-E. Baulieu. Non-acuveted form of the progesterone receptor from chick oviduct: characterization. Biochem Biophys Ret. Common. 95:1577-1584 (1980). 32. Pratt, W. B.. J. L. Kama, and D. V Pratt. The kinetics of glucocorticoid binding to the soluble specific binding protein of mouse fibroblasts. J. Bioi. Chem 250:4584-4591 (1975). 33. de Boer. W.. M. Lindh. J. Bolt. A. Bnnkmann. and E. Muldar Characteriza tion of the calf uterine androgen receptor and its activation to the deoxvribonucleic acid-binding state Endocrinology 118:851-861 (1986). 34. McBlein. W A.. D. 0. Toft, and G. Shyam*i* Transformation of mammarv cytoplaamic glucocorticoid receptor under cell-free conditions. Buxnemistrv 20:6790-6798 (1981). 35. Yang. C. R.. J. Me*tar. A. Wolfson. J. M. Renoir, and E.-E. Baulieu. Activation of the chick oviduct progesterone receptor by heparin m the presence or absence of hormone. Buxhem. J 208:399--406 11982). 36. Weichman B. M.. and A C. Notide*. Estradiol-binding kinetics of tne activated and nonactivated estrogen receptor. J BioL Chem. 252:8856-8862 (1977). 37. Muller. R. E-. D. M. Beebe. E. Bercel. A. M. Traish. and H. H. Wotiz. Eatnol and estradiol interactions with the estrogen receptor in uuio and in vitro. J Steroid Biochem 20:1039-1046 (19841. 38. Gaaiewicx. T. A. and G. Rucci. Cytosolic receptor for 2.3.7.8-tetrachiorodibenzo-p-dioxut: evidence for a homologous nature along mammalian species MoL PharmacoL 26:90-98 (1984). Bead re p rin t request to: Alan Poland. McArdia Laboratory for Cancer Re search. University of Wisconsin. Madiaon. WI 53706. G EN P 011145 783956