Binding of polychlorinated biphenyls classified as either phenobarbitone-, 3-methylcholanthrene- or mixed-type inducers to cytosolic Ah receptor

Chem.-Biol. Interactions, 39 (1982) 259--277 Elsevier/North-Holland Scientific Publishers Ltd.

259

BINDING OF POLYCHLORINATED BIPHENYLS CLASSIFIED AS EITHER PHENOBARBITONE-, 3-METHYLCHOLANTHRENEOR MIXED-TYPE INDUCERS TO CYTOSOLIC Ah RECEPTOR

S. BANDIERA a, S. S A F E b'* and A.B. OKEY c

aThe Guelph-Waterloo Centre for Graduate Work in Chemistry, Department of Chemistry, University of Guelph, Ontario NIG 2W1 (Canada), bDepartment o f Physiology and Pharmacology, College of Veterinary Medicine, Texas A & M University, College Station, TX 77843 (U.S.A.) and CDivision of Clinical Pharmacology, Department of Paediatrics, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario M5G 1X8 (Canada) (Received February 6th, 1981) (Revision received October 9th, 1981 ) (Accepted October 13th, 1981)

SUMMARY

It has been postulated that reversible, high-affinity binding of 3-methylcholanthrene(MC)-type inducers to a receptor protein (the Ah receptor) in hepatic cytosol is essential for induction of aryl hydrocarbon hydroxylase (AHH) enzymic activity. To test this postulate, the binding affinities of 16 highly-purified, synthetic chlorinated biphenyl (PCB) congeners, which have been categorized either as phenobarbitone(PB)-, MC- or mixed (PB + MC)-type inducers of cytochrome P-450-dependent monooxygenases have been examined. The affinity of individual biphenyl congeners for the receptor was determined by t h e i r c o m p e t i t i o n with 2,3,7,8-[3H]tetrachlorodibenzo-pdioxin ( [ 3H] TCDD) for specific cytosolic binding sites as measured by sucrose density gradient analysis following dextran-charcoal treatment. This assay demonstrates (a) that the receptor binds with highest affinity *To whom correspondence should be addressed. Abbreviations: Ah, The Ah locus represents a complex cluster of genes controlling the induction of numerous drug-metabolizing enzyme activities by polycyclic aromatic compounds. The cytosolic receptor for TCDD is viewed as the major regulatory gene product of the Ah complex; AHH, aryl hydrocarbon hydroxylase; p,p'-DDE, 1,1-dichloro-2,2-bis (p-chlorophenyl) ethylene; EC~0, effective concentration of competitor necessary to reduce specific binding of [3H]TCDD to 50% of the maximal value obtained in the absence of competitor; gay, average gravitational force; GLC, gas liquid chromatography; HEDG buffer, Hepes 25 raM, EDTA 1.5 raM, dithiothreitol 1 raM, glycerol, 10% v/v; HPLC, high pressure liquid chromatography; log P, logarithm of the n-octanol/water partition coefficient; MC, 3-methylcholanthrene; PCB, polychlorinated biphenyl; PB, phenobarbitone; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; TLC, thin-layer chromatography. 0009--2797/82/0000--0000/$02.75 © 1982 Elsevier/North-Holland Scientific Publishers Ltd.

260 to 3,3',4,4',5-pentachlorobiphenyl and 3,3',4,4'-tetrachlorobiphenyl ('pure MC-like' inducers); (b) mixed-type PCB inducers also bind to the receptor b u t with an affinity (average ECs0-value of 8.6 /aM) lower than that for 3,3' ,4,4' ,5-pentachlorobiphenyl; this corresponds with their relatively lower potencies as .~MH inducers; (c) the receptor binds 2,2',4,4'-tetra-, 2,3,4,5tetra-, 2,2',4,4',5,5'-hexa- and 2,3',4,4',5',6-hexachlorobiphenyl at high concentrations (0.1 mM), whereas PB fails to bind, even at a concentration of 10 mM. All PCBs tested competed with [3H]TCDD for A h receptor, but there was a great variation in their relative binding affinities. The fact that two chlorinated biphenyls classed as 'PB-like' inducers and t w o chlorinated biphenyls which are neither 'PB- nor MC-type' inducers competed, coupled with the fact that PB did not bind to the receptor suggest that chemicals other than 'pure MC-type' inducers can bind to the cytosolic receptor. Affinity of the binding dictates the relative p o t e n c y of given PCB congeners as inducers o f c y t o c h r o m e P-448.

INTRODUCTION

PCBs are widespread environmental contaminants which, because of their relative chemical stability, tend to accumulate in food chains. PCBs produce a variety of toxic effects in laboratory animals and humans. Difficulties arise, however, in measuring the toxic effect. Assessment of the biologic significance of PCBs is more easily accomplished by measuring a biochemical response which is dose-related and can be used as an indicator of potential toxicity of these compounds. Commercial mixtures of PCBs are well k n o w n inducers of hepatic microsomal drug-metabolizing enzymes [1--3]. These enzymes are involved in the metabolism of xenobiotics and endogenous chemicals, and in metabolicallymediated cellular toxicosis. The most thoroughly studied of the microsomal monooxygenase activities is AHH [ 4,5 ]. It has been shown that individual, purified, synthetic PCB isomers and congeners differ in their ability to induce AHH [6--8]. Based on these differences, the PCB isomers and congeners have recently been classed into three major categories [9--12]. One category, typified by PB, induced c y t o c h r o m e P-450
_.a_

* F o r s i m p l i c i t y in this p a p e r we refer t o t h e f o r m ( s ) of c y t o c h r o m e i n d u c e d b y MC collectively as ' c y t o c h r o m e P - 4 4 8 ' a n d t h e f o r m ( s ) i n d u c e d b y PB as ' c y t o c h r o m e P - 4 5 0 ' . T h e r e is evide~nce t h a t M C t r e a t m e n t i n d u c e s m o r e t h a n o n e f o r m of c y t o c h r o m e a n d t h a t t h e s e f o r m s differ in spectral, i m m u n o c h e m i c a l a n d c a t a l y t i c p r o p e r t i e s (reviewed in Ref. 13).

261 summation of the relative concentrations and activities of individual components of the mixture. Recently, evidence has been presented for the reversible, high-affinity binding of TCDD, a p o t e n t inducer of AHH, to a receptor protein in the hepatic cytosol of mice, rats and in cell lines in continuous culture [14--16]. The TCDD-receptor complex has been shown to translocate into the nucleus [ 16,17 ] and it has been suggested that nuclear uptake is an essential event in the induction of cytochrome P-448-dependent monooxygenases [16,17]. Other polycyclic aromatic hydrocarbons, such as MC, b e n z o [ a ] p y r e n e and ~-naphthoflavone, which induce AHH, also bind to the TCDD receptor [14--17]. Recently, evidence has been presented for the non-covalent binding of MC to a cytosolic receptor protein in rat liver [18]. Based on the binding characteristics of the MC binding protein, it may be that this protein is the same as the TCDD receptor. There appears to be a good correlation b e t w e e n the p o t e n c y of these c o m p o u n d s as inducers of hepatic AHH activity and their affinity for the cytosolic binding species [4--6,14,16]. Furthermore, the ontogeny of the receptor in rat liver corresponds well with the o n t o g e n y of hepatic, nuclear A H H in rats [19,20]. In the present report, we examine the binding of 16 highly purified, synthetic polychlorinated biphenyl congeners and isomers, as well as Aroclor 1254, to the cytosolic receptor protein from rat liver. M A T E R I A L S AND METHODS

Synthesis and purification o f PCB congeners Individual PCB isomers and congeners were synthesized from the appropriate precursors and purified by column chromatography on alumina/Florisil and silica gel thin-layer chromatography (TLC) as previously described [9,11,12,21,22; summarized in Table I]. The purities were determined by gas liquid chromatography (GLC) using a Hewlett-Packard model 5710 chromatograph equipped with a 63Ni electron capture (EC) detector and a 0.6 cm X 1.2 m glass colum packed with 3% OV 101 on Ultrabound Carbowax 20 M, 80--100 mesh ( R F R Corporation, Hope, RI). Structures of purified PCB isomers and congeners were confirmed by their 200 MHz NMR and mass spectra (determined on a VG Micromass 7070 mass spectrometer). Octanol/water partition coefficients o f PCB congeners The logarithm of the n-octanol/water partition coefficient (log P) for the chlorinated biphenyl isomers and congeners (Table I) was measured by high pressure liquid chromatography (HPLC) using the m e t h o d of Vieth et al. [ 23]. It has been shown that the logarithm of the retention time on a bonded (C-18) reverse-phase HPLC system is linearly related to log P [23--25]. Retention times of the PCBs were obtained on a Beckman Isocratic HPLC Model 110A, using an Ultrasphere-ODS Column (5 pm, 25 cm X 4.6 mm i.d.), with 85% methanol/15% water as eluting solvent at a flow rate of 1.5 ml/min at ambient temperature. A calibration mixture consisting of benzene, bromo-

Cl

Cl

cl~--~cl

C!

Cl

Cl

Cl

Cl

Cl

CI

c ~ ~ C ~

Cl Cl

cl-~~cl

Cl Cl

cl-~~cl

Cl Cl

c~-~~c~

Cl Cl

cl c~-~~c~

c , ~ ~ -

Ci

Structure

2,3',4,4',5-Pentachlorobiphenyl

2,3,3',4,4'-Pentachlorobiphenyl

2,3,4,4',5-Pentachlorobiphenyl

3,3',4,4',5,5'-Hexachlorobiphenyl

3,3',4,4',5-Pentachlorobiphenyl

3,4,4',5-Tetrachlorobiphenyl

c, 3,3',4,4'-Tetrachlorobiphenyl

Nm~ne

2,4,5-Trichloroaniline

2,3,4-Trichloroaniline

2,4-Dichloroaniline

3,4,5-Trichloroaniline 1,2,3-Trichlorobenzene

3,4,5-Trichloroaniline o-dichlorobenzene

4-Chloroaniline 1,2,3-Trichlorobenzene

3,3' 4,4'-Tetraaminobiphenyl

Synthetic precursors

TABLE I S Y N T H E S I S A N D P R O P E R T I E S O F PCB I S O M E R S A N D C O N G E N E R S

>99%

> 99%

>99%

88%

97%

>99%

>99%

GLC purity

6.39

6.44

6.89

7.71

6.91

5.90

5.73

Log P

11

11

12

21

Unpublished

12

21

Ref. No.

b~ b~

cI

Cl

Cl

CI

Cl Cl

Cl

Cl

Cl

c,

c,

CI

CI

Cl

Cl

CI

Cl

CI

Cl

cl-~~~ cI-~~cl

Cl

c e - ~ ~ c l

Cl

CI

ci-~~c,

c l - ~ ~ c l cl c~

Cl

C! Cl

c l - ~ ~ c ,

Cl Cl

c l ~ ~ } -

CI

c,~---~c~

Cl CI

c,~--~

Cl

2,3' ,4,4' ,5',6-Hexachlorobiphenyl

2,3,4,5-Tetrachlorobiphenyl

2,2',4,4',5,5'-Hexachlorobiphenyl

2,2',4,4'-Tetrachlorobiphenyl

2,3,3',4,4',5,5'-Heptachlorobiphenyl

2,3,3',4,4',5'-Hexachlorobiphenyl

2,3',4,4',5,5'-Hexaehlorobiphenyl

2,3,3',4,4',5-Hexachlorobiphenyl

2,3,4,4',5-Pentachlorobiphenyl

3,4,5-Trichloroaniline 1,3,5-Trichlorobenzene

2,3,4-Tetrachloroaniline benzene

2,2',5,5'-Tetrachlorobenzidine

2,2' -Dichlorobenzidine

3,4,5-Trichloroaniline 1,2,3,4-Tetrachlorobenzene

2,3,4-Trichloroaniline 1,2,3-Trichlorobenzene

2,4,5-Trichloroaniline 1,2,3-Trichlorobenzene

1,2,3,4-Tetrachlorobenzene

3,4-Dichloroaniline

1,2,3,4-Tetrachlorobenzene

4-Choroaniline

>98%

>99%

>99%

>99%

>99%

>99%

> 99%

>99%

>99%

7.26

6.26

6.67

5.29

8.14

7.27

7.37

6.43

22

Unpublished

21

21

11

11

11

12

264 benzene, biphenyl, bibenzyl, 1,1-dichloro-2,2-bis (p-chlorobiphenyl) ethylene (p,p'-DDE) and 2,2',4,5,5'-pentachlorobiphenyl was employed as reference standards. Materials [3H]TCDD (spec. act., 50--52 Ci/mmol) and non-radiolabelled TCDD were purchased from KOR Isotopes (Cambridge, MA). The radiolabelled TCDD contained approx. 80% of the TCDD congener, as determined by the manufacturer and confirmed in our laboratory by GLC. The remaining 20% consisted of trichloro- and pentachlorodibenzo-p~lioxin congeners, which have been reported to be considerably less active as inducers of AHH activity and as competitors for specific cytosolic binding sites [14,30]. Dextran (average Mr = 82 000) and Hepes were purchased from Sigma Chemical Company, (St. Louis, MO). Dithiothreitol was obtained from Eastman Kodak Company (Rochester, NY); MC from Pfaltz and Bauer Inc. (Stanford, CN); decolorizing charcoal (Norit A) from BDH Chemicals (Toronto, Ontario}; sucrose (SDS grade} from Beckman Instruments (Fullerton, CA), dimethyl sulfoxide, glycerol (A.C.S. grade) and disodium ethylenediamine tetraacetate (EDTA) from Fischer Scientific Company (Toronto, Ontario); Aroclor 1254 from Monsanto Company (St. Louis, MO). Sodium phenobarbitone was a generous gift from Dr. G.D. Sweeney, McMaster University Health Sciences Centre (Hamilton, Ontario). Animals One-month~ld male Wistar rats, average weight 100 g, were purchased from Woodlyn Laboratories, Ltd. (Guelph, Ontario). They were housed in wire cages, allowed free access to Purina Rat Chow #5002 and water and were maintained on a diurnal cycle of 13 h of light/ll h of darkness. Preparation of cy tosol Rats were killed by cervical dislocation and the liver was perfused with ice~cold HEDG buffer (Hepes 25 raM, EDTA 1.5 mM, dithiothreitol 1 mM, glycerol, 10% v/v) pH 7.6. After perfusion in situ, the liver was removed and transferred to pre.weighed, ice-cold solutions of HEDG buffer, weighed, rinsed with HEDG buffer, minced, and homogenized in ice-cold HEDG buffer (3 ml/g of liver}. The homogenate was centrifuged at 10 000 × g for 20 min and the resulting supernatant fraction was further centrifuged at 105 000 × g for 1 h. Surface lipid was carefully removed with a Pasteur pipette and the cytosol was carefully removed without disturbing the microsomal pellet. The resultant cytosol usually contained about 20 mg protein/ ml. All protein concentrations were determined by the method of Lowry eta!. [26], with bovine serum albumin as the standard and using appropriate buffer blanks. All cytosols representing a pool from five different animals of the same strain and age, were stored in liquid nitrogen for periods up to two months with no loss of specific binding under these conditions.

265 C y t o s o I re c e p to r b in d i n g

The procedure used to quantitate the cytosol receptor was sucrose density gradient analysis following dextran-charcoal treatment as previously described [16]. This assay is more reliable in separating a class of high-affinity, lowcapacity sites from non-saturable binding than DEAE-cellulose column chromatography or dextran-charcoal adsorption methods. All procedures were done at 0--5°C unless stated otherwise. Samples for sucrose density gradient analysis were prepared b y incubating 1 ml of cytosol (5--6 mg protein) with 10 nM [3H]TCDD for 1 h at 0 - 5 ° C . [3H]TCDD was added in 10 pl of dimethyl sulfoxide/ml of cytosol. In competitor experiments, an equal amount of dimethyl sulfoxide containing the desired concentration of competitor was added. A control containing [3H]TCDD plus an equal volume of solvent (20 t~l), but no competitor was also prepared. After incubation, the u n b o u n d and loosely bound [3H]TCDD and/or competitor were removed by agitating the cytosol with a dextrancharcoal pellet (10 mg of charcoal/mg of dextran, pelleted from HEDG buffer). Following a 15 min incubation at 0--5°C, the dextran-charcoal was removed by centrifugation at 4000 X g for 15 min. Aliquots of cytosol were taken both before and after dextran-charcoal treatment for determination of 'total' and ' b o u n d ' radioactivity. Three hundred microlitres of the charcoaltreated cytosol were layered onto 5--20% sucrose gradients prepared in HEDG buffer. Gradients were centrifuged in a Beckman SW 50.1 rotor at 48 000 rev./min for 16 h (gay. = 216 000) at 2°C. After centrifugation, 40 fractions of 0.12 ml each were collected per gradient into 7 ml plastic mini-scintillation vials, using an ISCO model 184 gradient fractionator (Instrumentation Specialities Company, Lincoln, NE) and a LKB model 2070 fraction collector (LKB-Produkter AB, Bromma, Sweden). Radioactivity in each fraction was determined by liquid scintillation counting and corrected for counting efficiency. RESULTS Incubation of rat hepatic cytosol with 10 nM [3H]TCDD for 1 h at 5°C produced a specific binding peak detectable by sucrose gradient analysis. The peak sedimented at approx. Fraction 25 under our experimental conditions (Fig. 1). The concentration of the receptor in hepatic cytosol was found to be approx. 30 fmol/mg of cytosolic protein. This value was remarkably constant over more than 30 experiments and with different preparations of cytosol from animals of the same age. In the presence of 10 pM non-radiolabelled TCDD or 10 pM MC, this peak was completely abolished. In the presence of 10 pM PB, however, this peak was unaltered (Fig. 1A). Of the chlorinated biphenyl congeners, 10 pM 3,3',4,4',5-pentachlorobiphenyl, which {like MC and TCDD) is a p o t e n t inducer of c y t o c h r o m e P-448 and AHH [27 ], completely inhibited specific binding of [ 3H] TCDD to cytosolic receptor. In contrast, 10 t~M 2,2',4,4'-tetrachlorobiphenyl, which is

400

800

1200

TOP

0

FRACTION

20

C~(~

NUMBER

H

10 nM [3H]TCDD

3J0

4 0I

w i t h 1 0 u M PHENOBARB

w i t h 1 0 p M 3-METHYLCHOL

TOP

B

10

20 FRACTION

NUMBER

H

30

with 10pM

40

2.2:4.4LTCBP

with 10pM 2.3:4.4:5-PCBP

3,3:4.4:5- P C B P

10 nM [3H]TCDD

~ ' ~ with 10pM

O--O

Fig. 1. Sucrose density gradient detection of specific high affinity binding of [ 3H ]TCDD to a c o m p o n e n t in rat hepatic cytosol. A: cytosol (5--6 mg protein/ml) from immature male Wistar rats was incubated with 10 nM [~H]TCDD in the absence of c o m p e t i t o r (c~()) and in the presence of 10 pM MC (m =) or 10 pM PB (A A). Following dextran-charcoal treatment, gradients were centrifuged and fractionated as described (see Materials and Methods). B: chlorinated biphenyls as competitors: cytosol (5--6 mg of p r o t e i n / m l ) w a s incubated with 10 nM [3H ]TCDD in the absence of competitor (c) o) and in the presence of 10 pM 3,3',4,4',5-pentachlorobiphenyl ( ~ - - - ~ ) or 10 pM 2,3',4,4',5-pentachlorobiphenyl (v v) or 2,2',4,4'-tetrachlorobiphenyl (4~ #). The amount of [3H]TCDD specifically bound to the receptor, either in the absence or presence of competitors, was measured from the area under the peak which sedimented at approx. Fraction 25. Based on radioactivity determinations, approx. 30 fmol of [3H]TCDD were specifically bound/rag protein, in the absence of competitor, under our experimental conditions.

-r

Q e, U

~D

z

o

£

1600

2000r

I

A

10 -9 M

10 -s M

0

3 18 40 42

10 -s M

IO-~M 10 -~ M

No competitor 48 + 3,3',4,4',5-Pentachlorob iphenyl 5 37 88 87

0

100 2 1 13 39 39

37

f m o l [ 3H ] T C D D b o u n d • m g p r o t e i n -~

f m o l [ 3H ] T C D D b o u n d • mg p r o t e i n -~ Spec. b i n d i n g o f [ 3H]TCDD (% c o n t r o l )

5 × 10 9 M [ 3 H ] T C D D

10 -~ M [ 3 H ] T C D D

5 3 35 106 107

100

Spec. b i n d i n g o f [3 H ] T C D D (% c o n t r o l )

OF 3,3',4,4',5-PENTACHLORO-

0 0 3 11 13

16

0 1 18 65 83

100

fmol [3H]TCDD Spec. b i n d i n g b o u n d • m g p r o t e i n - ' o f [ 3H]TCDD (% c o n t r o l )

10 -9 M [ 3 H ] T C D D

S P E C I F I C B I N D I N G O F [ 3 H ] T C D D TO H E P A T I C C Y T O S O L A T V A R Y I N G C O N C E N T R A T I O N S B I P H E N Y L , IN T H E P R E S E N C E O F T H R E E D I F F E R E N T C O N C E N T R A T I O N S O F [ 3 H ] T C D D

T A B L E II

268

A

H H ~ ~

100

nonlabelled TCOD 3-METHYLCHOL 3,3~4.4"-TCB P 3,4,4".5-TC B P . 3,3~4,4'5- PC BP

; ;.;

80 60 40 "6

20 lO-e

10-6

0-4

10-:2 2,3,3;4,4'- PC 8p ~ 2.3~44~5-PCBP ~ 2"3'4'4~5-PCBP ~ - - ~ 2.3.3~4 4~5- HC BP 2.3 ~4.4.'5,5"H C BP <~--<] 2.~ 3~4,4~5'-H C BP H 2.3 3~ 4.'5.5'- HpC Be

~. 1 0 0

t'~

~O eo

0

i .

.

10-4

l

10080

~

P

C

J 10- 2

~

B

2'2:44~'TC B P

P

~

2.3.4,5-TCBP 2.3~4.4;5;6 - H C B P

60 o

20 0

10"8 10-6 10.4 COMPETITOR CONCENTRATION (M)

10.2

269 classed as a PB-type inducer [11,12] competed very little with [3H]TCDD for the specific binding peak. On the other hand, 10 t~M 2,3',4,4',5-pentachlorobiphenyl, a mixed-type inducer of AHH activity [ 11 ], inhibited greater than 50% of the [3H]TCDD binding to the cytosolic binding site, as measured b y sucrose density gradient analysis. It can be seen that this cytosolic binding peak is greatly reduced relative to that of 2,2',4,4'-tetrachlorobiphenyl yet is n o t totally eliminated as observed with 3,3',4,4',5-pentachlorobiphenyl. The data suggest that only those c o m p o u n d s which are p o t e n t inducers of c y t o c h r o m e P-448, totally eliminate [3H]TCDD binding to the cytosolic receptor. PB which induces a form of c y t o c h r o m e P-450 that is separate and distinct from P-448 [28,29], has no effect on [3H]TCDD binding. In addition, raixed-type inducers, which are less p o t e n t as inducers of c y t o c h r o m e P-448 also c o m p e t e with [ 3H]TCDD for specific binding sites. In order to determine if [3H]TCDD and the chlorinated biphenyl congeners bind to the same site on the receptor protein, competition experiments were carried o u t in which cytosol was incubated with 10 nM, 5 nM and 1 nM [3H]TCDD in the presence of various concentrations of 3,3',4,4',5-pentachlorobiphenyl (Table II). The results were evaluated by means of a Dixon plot. Unfortunately, sucrose density gradient analysis is not suitable for rigorous kinetic analysis [16]. Two [3H]TCDD binding peaks were also observed, which sedimented at approx. Fractions 12 and 38 under our experimental conditions. The peaks were unaltered b y treatment with large amounts of all competitors (Fig. 1). This suggests that they represent non-specific binding.

Competitive binding by chlorinated biphenyl congeners Sixteen highly purified synthetic chlorinated biphenyl congeners, as well as the commercial mixture, Aroclor 1254, were tested in competition experiments. The results of this competitive binding between 10 nM [3H]TCDD and varying concentrations of the biphenyl congeners are shown in Fig. 2. All of the PCBs tested inhibited specific binding of [3H]TCDD to the binding sites. Of the c o m p o u n d s which are considered 'pure MC-like' inducers [6,11,12,26] (Fig. 2A), MC was the best competitor with [3H]TCDD for the receptor, next to non-radiolabelled TCDD and completely abolished the binding peak at a concentration of 1 t~M. Both MC and TCDD b o u n d more avidly than 3,3',4,4',5-pentachlorobiphenyl, which was found to be the most Fig. 2. Comparative competition by chlorinated biphenyl congeners. A: competition by the 'pure MC-like' inducers for the specific binding of [3H]TCDD to hepatic cytosol: rat hepatic cytosol (5--6 mg protein/ml) was incubated with 10 nM [~H]TCDD in the presence of varying concentrations (given on the abscissa) of non-labelled TCDD, MC and four chlerinated biphenyl congeners. Cytosol incubated with 10 nM [3H]TCDD in the absence of competitor was used as control. B: competition by 'mixed-type' inducers; conditions as in (A). C: competition by 'PB-like' inducers and 'non-inducers'; conditions as in (A). For PB, concentrations greater than 10 mM were not attempted. To test the reproducibility of the assay, the competition experiments for some compounds were carried out in triplicate. S.D. is shown by error bars.

270 potent competitor among the chlorinated biphenyl congeners tested and completely inhibited [3H]TCDD binding at a concentration of 10/~M. The next most effective competitor in the PCB series was 3,3',4,4'-tetrachlorobiphenyl. The other two 'MC-like' biphenyls, 3,3',4,4',5,5'-hexachlorobiphenyl and 3,4,4',5-tetrachlorobiphenyl, also competed with [3H]TCDD for specific binding sites, but at higher concentrations than 3,3',4,4'-tetrachlorobiphenyl. Figure 2B summarizes the competitive binding data for the PCBs which are classed as mixed-type inducers [ 9 - 11]. All eight of the PCB congeners, as well as the commercial mixture, Aroclor 1254, were very similar in their ability to inhibit binding to the Ah receptor. At concentrations ranging between 0.1 mM and 1.0 mM, these chlorinated biphenyls eliminated the cytosolic binding peak, although for one of the congeners higher concentrations were required but could not be achieved because of the limited solubility of the biphenyl in aqueous solution. Of the compounds which are classed as 'PB-type' inducers [6,11,12] (Fig. 2C), the two polychlorinated biphenyls, 2,2',4,4'-tetra- and 2,2',4,4',5,5'-hexachlorobiphenyl bound to the receptor as evidenced by the competition with [3H]TCDD, at high concentrations (1.0 mM). Similarly, 2,3,4,5-tetra- and 2,3',4,4',5',6-hexachlorobiphenyl, bound to the receptor at comparable concentrations (1.0 mM). Based on the enzymic, spectral and electrophoretic properties of microsomes prepared from rats pretreated with 2,3,4,5-tetra- and 2,3',4,4',5',6-hexachlorobiphenyl, these compounds have been classified as non-inducers of cytochrome P-450
Relative affinities o f TCDD, MC and the chlorinated biphenyl congeners The equilibrium dissociation constant, Kd, is routinely employed as a measure of the affinity of binding to the receptor. However, as reported earlier [16], Kd-values measured by Scatchard plot analyses were difficult to quantitate accurately. For this reason, ECs0-values derived from competitive binding experiments were chosen as a measure of the relative affinities of the chlorinated biphenyls for the receptor. Although all of the PCBs inhibited specific binding of [3H]TCDD to cytosolic receptor when tested at high concentrations, there was a great variation in their relative binding affinities, as shown in Fig. 3. The in vivo binding affinities, expressed as the ECs0-value, represent the effective concentration of non-labelled competitor necessary to reduce specific binding

COMPETITOR

EC S0

VALUES

Jncreas,ng phenobar bitone

CO

2~3.4,5 - t e t rac h I o r o b i p h e n y l

1.4 x l O - 4 M

2,2~4 4 ' - t e t r a c h l o r o b i p h e n y l

1 3 ~10 "4 M

2.3~4.4~5:6 - h e x a c h I o r o bi p h e n y I

affinity

9.9 ~10 -s M

2.2:4,4:5.S" h e x a c h i o ro b iphe n y l

7.9 x 10 - s M

2,3.3;4.4~5.5"- h e p t a c h I or o blptt e n y I

not d e t e r m m e d due 1o l i m i t e d

_ _

_ _c o m p e t e s

solubility 2.3;4,4~5.5 "-he x a c h I o r o b i p h e n y I

1.6 x l O - $ M

2~3,4.4"5 - p e n t a c h I o r o b; p h e n y I

1,4 x l O - s M

2,3~4,4"5 - p e n t ac h I o rob; p h e n y I

5.1 x l O - e M

2.3,3;4.4~5-he x a c h Ior o b i p h e n y l

Aroclor

1254

I I

6 0 x 10-6 M

5.0 X 10 - 6 M

2.3.3:4 4"- p e n t a c h Ior o b i p h e n y l

4.3 xlO -6 M

2,3,4.4~5-pent ac h l o r o b i p h e n y l

4.1 x 10- e M

3 3:4,4~5 5 " h e x a c h l o r o b i p h e n y l

ot d e t e r m i n e =e to l i m i t e d )lubilitv

344~5 -tet rachlorobiphenyl

2 . 8 x IO "s M

3.3~4 4 ' - t e t r ac h lot o b i p h e n y l

7.1 x 10-7 M

3 314415 ° p e n t a c h l o r o b i p h e n ¥ I

1.3 x l O " z M

n o n l a b e ~;ed T C D D

I

7.1 x 10 - e M

2,3,3:44~5 s-he x a c h i o r o b i p h e n y I

3- methylc holant hrene

I

2 . 5 x 10 - 8

competes

I , J i

M

]

1 0 x 1 0 "8 M I 1 0 -9

I

COMPETITOR

lt0_7

I

CONCENTRATION

10-5

I

10-3

(M)

Fig. 3. Comparative EC~0-values for non-labelled TCDD, MC and chlorinated biphenyl congeners. The in vitro binding affinities of the chlorinated biphenyl congeners, MC and non-labelled T C D D , expressed as ECs0-values , were determined as the effective concentration of c o m p e t i t o r necessary to reduce specific binding o f [3H]TCDD to 50% of the maxim~! value obtained in the absence of c o m p e t i t o r (control). The smaller the ECs0-value , the greater the apparent affinity o f the competitor for [ 3 H ] T C D D binding sites. All values are expressed as molar concentrations. For MC, 3,3',4,4',5-pentachloro- and 2,3,4,4',5-pentachlorobiphenyl, the EC~0-values represent the mean of 3 determinations ± S.D. as s h o w n by the error bars.

272 of [3H]TCDD to 50% the maximal value obtained in the absence of competitor. The effective concentration is defined here as the amount of each PCB which was added to the incubation mixture. As shown earlier (Fig. 2A), non-labelled TCDD had the greatest affinity for the receptor. With 10 nM [3H]TCDD, it was found that l 0 nM nonlabelled TCDD displaced approx. 50% of the radioligand, as expected. MC was found to have an apparent affinity for the receptor within 5-fold that of non-labelled TCDD as reported previously [14,16], and approx. 10 times greater than that of 3,3',4,4',5-pentachlorobiphenyl. 3,3',4,4'Tetrachlorobiphenyl also b o u n d the receptor with high avidity. Its ECs0-value of 0.71 aM was almost three times that of 3,3',4,4',5-pentachlorobiphenyl (ECso = 0.13 p M), indicating a lower affinity. However, its apparent affinity was approx. 40-fold greater than that of another MC-type inducer, namely 3,4,4',5-tetrachlorobiphenyl. The binding affinities for the fourth 'MC-type' inducer, 3,3',4,4',5,5'-hexachlorobiphenyl (and for 2,3,3',4,4',5,5'-heptachlorobiphenyl, a mixed-type inducer) could n o t be estimated because the limited aqueous solubility of these congeners made it impossible to achieve a soluble concentration sufficient to c o m p e t e for at least half the specific binding of [3H]TCDD. The variation among the ECs0-values for the four 'MC-type' congeners is great and cannot be merely attributed to their degree of lipophilicity. The n-octanol/water partition coefficients (expressed as log P) of the chlorinated biphenyls used in the study are given in Table I. Log P-values reflect the degree of lipophilicity or h y d r o p h o b i c i t y of the PCBs and can be used as a measure of the aqueous solubilities of the c o m p o u n d s [23--25]. If the differences in the ECs0-values, which have been observed for the four 'MCt y p e ' inducers, or similarly, for the entire series of PCB congeners in this report, were simply reflections of solubility differences, a correlation between ECs0-values and log P would be expected. This was not observed. For example, the n-octanol/water partition coefficients for 2,2',4,4',5,5'-hexachlorobiphenyl and 3,3',4,4',5-pentachlorobiphenyl are very similar, yet, there is approx, a 600-fold difference in their ECs0-values. More likely, the disparity between the ECs0-value of 3,4,4',5-tetrachlorobiphenyl and those o f the other two MC-like biphenyls, for which ECs0 were calculated (results in Fig. 3), is due to the lower biological potency of 3,4,4',5-tetrachlorobiphenyl as an inducer o f c y t o c h r o m e P-448 [12]. The binding affinities of those PCBs classified as mixed-type inducers were very consistent, with ECs0-values ranging from 4.3 t~M to 16 pM. Because of their similarity, an average value of 8.6 t~M could be calculated. Comparison of this average with the value obtained for 3,3',4,4',5-pentachlorobiphenyl reveals that there is approx, a 65-fold difference between the mixed-type congeners and 3,3',4,4',5-pentachlorobiphenyl in the apparent affinity for the receptor. However, this average value is almost three times less than that of 3,4,4',5-tetrachlorobiphenyl, indicating a greater affinity. The two congeners which act as PB-like inducers were relatively poor competitors for the [3H]TCDD specific binding sites, with apparent affini-

273 ties of 6 0 0 - . 1 0 0 0 times less than that of 3,3',4,4',5-pentachlorobiphenyl. Similarly, the ECs0-values for 2,3,4,5-tetra- and 2,3',4,4',5',6-hexachlorobiphenyl (1.4 × 10 -4 M and 9.9 × 10 -s M, respectively), which are neither 'PB-' nor 'MC-type' inducers, were comparable to those of 2,2',4,4'-tetraand 2,2',4,4',5,5'-hexachlorobiphenyl. The ECs0-value for PB could not be calculated because of the failure of this c o m p o u n d to compete with [3H]TCDD for the A h receptor even at a concentration of 10 mM. DISCUSSION

In this report we examined the in vitro binding affinities of a series of chlorinated biphenyls for the cytosolic [ 3HI TCDD receptor protein. At high concentrations, all of the PCBs tested b o u n d to A h receptor as judged by competition with [3H]TCDD for specific binding sites..41though none of the PCB congeners competed as effectively as MC, 3,3',4,4',5-pentachlorobiphenyl had an apparent affinity only 10-fold less than that of MC and was clearly the best competitor o f the PCBs tested. 3,3',4,4'-Tetrachlorobiphenyl also demonstrated high affinity for the receptor. The t w o congeners, 3,3',4,4',5,5'-hexa- and 3,4,4',5-tetrachlorobiphenyl are 'pure MC-like' inducers, devoid of PB-like activity [6]. 3,3',4,4',5-Penta-, 3,3',4,4'-tetraand 3,3',4,4',5,5'-hexachlorobiphenyl have been shown to be at least as p o t e n t as MC as inducers of hepatic AAH [6,27]. Furthermore, MC is 30 000 times less p o t e n t than TCDD at inducing hepatic hydroxylase activity in the rat and responsive mouse [5,6,14], yet the data show that it binds to the receptor with an affinity near that of non-radiolabelled TCDD, as reported previously [14,16]. This finding along with the fact that 3,3',4,4',5-pentaand 3,3',4,4'-tetrachlorobiphenyl do not c o m p e t e for the specific [ 3H] TCDD binding sites to the same extent as MC suggests that perhaps it is MC that exhibits an unusually high avidity, in light of its biological potency. Since AHH metabolizes MC and other polycyclic hydrocarbons, the diminished in vitro p o t e n c y of MC as an AHH inducer, relative to TCDD and the PCBs, could be d u e to metabolic inactivation [14]. Although 3,3',4,4',5,5'-hexachlorobiphenyl did not c o m p e t e for more than 40% of the specific binding of [3H]TCDD, and its ECs0-value could not be estimated under our experimental conditions, this result should not be taken as an indication that the congener has a poor affinity for the receptor. In vivo, it acts as a full agonist and is a potent inducer of c y t o c h r o m e P-448 dependent monooxygenase activities [6--8]. On the other hand, 3,4,4',5-tetrachlorobiphenyl is a poor competitor for the specific binding sites as demonstrated by its ECs0-value, and indeed, it is a relatively weak inducer of hepatic hydroxylase activity in vivo. On a quantitative basis in terms of the dose required for maximum AHH induction, 3,4,4',5-tetrachlorobiphenyl is comparable to some PCBs which are mixed-type inducers [12; unpublished results]. It had been suggested that any given PCB congener produced either an 'MC-like' response or a 'PB-like' response or was devoid of both activities [6]. It was only recently that some chlorobiphenyl isomers, apart from the com-

274 mercial mixture, Aroclor 1254, were shown to induce a pattern of microsomal monooxygenase activities resembling that resulting from the combined administration of MC and PB [9--12]. The results reported herein represent the first competitive binding experiments that have been carried out employing 'mixed-type' compounds. All eight congeners, which are mixed-type inducers [9--11], bound to the cytosolic receptor with very similar binding affinities. On the average, they were 65 times weaker in their capacity to displace [3H]TCDD from the receptor than 3,3',4,4',5-pentachlorobiphenyl. ~ince these congeners do act as AHH inducers, it could be postulated that they would act on the same receptor. Therefore, it is not surprising to find that they compete with [3H]TCDD for A h receptor. However, these congeners also act like PB inducers, and it has been found that both their capacity to induce AHH in vivo and their binding affinity for the hepatic cytosol species in vitro, was much less than that of the pure MC-like PCBs, with the exception of 3,4,4',5-tetrachlorobiphenyl. As expected, Aroclor 1254 also behaved very similarly to the biphenyls which are classified as mixed-type inducers, as illustrated by its ECs0-value (6.0 pM, based on an average molecular weight 326.4 for Aroclor 1254). It has been postulated that the hepatic cytosol protein that binds [3H]TCDD is the induction receptor, that is the binding site that initiates the expression of cytochrome P-448 and AHH. Accordingly, it has been predicted that the relative binding affinities of halogenated aromatic hydrocarbons, such as dibenzo-p-dioxins, dibenzofurans and biphenyls, should closely correspond to their biological potencies in inducing AHH activity [4--6, 19,30] and, possibly, their capacity to produce a toxic effect [4,27]. Competitive experiments with 2,2',4,4'-tetra- and 2,2',4,4',5,5'-hexachlorobiphenyl clearly demonstrate that these two 'PB-like' congeners do, in fact, inhibit [~H]TCDD binding to A h receptor, albeit at relatively high concentrations. However, 2,3,4,5-tetra- and 2,3',4,4',5',6-hexachlorobiphenyl, which are not inducers of cytochrome P-450-dependent monooxygenase, also inhibited [3H]TC,DD binding to the receptor at very similar concentrations, whereas PB did not. ~qince PB is water soluble while these two biphenyls are lipid soluble, it must be concluded that the binding to the receptor at this high concentration represents a non-specific interaction. It is not surprising that 'PB-type' inducers and the 'non-inducers' interact with the A h receptor when tested at very high concentrations. Specificity for many receptors is not absolute, but, rather, expresses degrees of affinity between receptor and various ligands. It is well-known that hormone specificity for binding to steroid receptors involves some 'overlap' among various classes: e.g., androgens, at high concentration, are capable of binding to the estrogen receptor and translocating the receptor into uterine nuclei in vitro [31]; similarly, progestins bind not only to the progesterone receptor, but also bind to the glucocorticoid receptor [32]. An earlier study examined the binding of 2,2',4,4',5,5'-hexachlorobiphenyl to the cytosolic TCDD receptor and found that this hexachlorobiphenyl was biologically inactive in inducing AHH activity and failed to compete with

275 [3H ] TCDD for specific binding sites when present at 1000 times the concentration of the radioligand [6]. We find that a concentration of 10 pM (1000 times [3H]TCDD concentration), 2,2',4,4',5,5'-hexachlorobiphenyl showed little competition for the A h receptor while at a concentration 100 000 times that of the radioligand, complete inhibition occurred. One conclusion which may be made in light of these results is that for chlorinated biphenyls, and possibly, for all halogenated aromatic hydrocarbons, it is not t h e simple presence or absence of binding of the congener to the cytosolic receptor that is required for the induction of c y t o c h r o m e P-448, b u t it is the avidity of t h e binding that correlates with AHH induction. ACKNOWLEDGEMENTS

The financial assistance of the Research Programs Branch, Health and Welfare Canada, the Natural Sciences and Engineering Research Council of Canada and the U.S. Environmental Protection Agency are gratefully acknowledged. A.B.O. acknowledges the National Cancer Institute of Canada for financial support. The assistance of Dr. A. Parkinson, Dr. L. Robertson and Mrs. L. Safe were also appreciated. REFERENCES 1 A.P. Alvares and A. Kappas, The inducing properties of polychlorinated biphenyls on hepatic monooxygenases, Clin. Pharmacol. Ther., 22 (1977) 807. 2 C.L. Litterist, T.M. Farber, A.M. Baker and E.J. Van Loon, Effect of polychlorinated biphenyls on hepatic microsomal enzymes in the rat, Toxicol. Appl. Pharmacol., 23 (1972) 112. 3 D.J. Ecobichon and A.M. Comeau, Comparative effects of commercial arochlors on rat liver enzyme activities, Chem.-Biol. Interact., 9 (1974) 341. 4 A. Poland and E. Glover, 2,3,7,8-Tetrachlorodibenzo-p-dioxin: Segregation of toxicity with the A h locus, Mol. Pharmacol., 17 (1980) 86. 5 A. Poland and A. Kende, The genetic expression of aryl hydrocarbon hydroxylase activity: Evidence for a receptor mutation in nonresponsive mice, in: H.H. Hiatt, J.D. Watson and J.A. Winston (Eds.), Origins of Human Cancer, Book B, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1977, p. 847. 6 A. Poland and E. Glover, Chlorinated biphenyl induction of aryl hydrocarbon hydroxylase activity: a study of the structure-activity relationship, Mol. Pharmacol., 13 (1977) 924. 7 J.A. Goldstein, P. Hickman, H. Bergman, J.D. McKinney and M.P. Walker, Separation of pure polychlorinated biphenyl isomers into two types of inducers on the basis of cytochrome P-450 and P-448, Chem.-Biol. Interact., 17 (1977) 69. 8 H. Yoshimura, N. Ozawa and S. Saeki, Inductive effects of polychlorinated biphenyls mixture and individual isomers on the hepatic microsomal enzymes, Chem. Pharm. Bull., 26 (1978) 1215. 9 A. Parksinson, R. Cockerline and S. Safe, Induction of both 3-methylcholanthreneand phenobarbitone-type microsomal enzyme activity by a single polychlorinated biphenyl isomer, Biochem. Pharmacol., 29 (1980) 259. 10 A. Parkinson, R. Cockerline, L. Robertson and S. Safe, Induction of cytochrome P-448 and P-450 by PCB isomers and congeners, in: M.J. Coon, A.H. Conney, R.W. Estabrook, H.V. Gelboin, J.R. Gillette and P.J. O'Brien (Eds.), Microsomes, Drug Oxidations, and Chemical Carcinogenesis. Academic Press, New York, 1980, p. 579.

276 11 A. Parkinson, R. Cockerline and S. Safe, Polychlorinated biphenyl isomers and congeners as inducers of both 3-methylcholanthrene- and phenobarbitone-type microsomal enzyme activity, Chem.-Biol. Interact., 29 (1980) 277. 12 A. Parkinson, L. Robertson, L. Safe and S. Safe, Polychlorinated biphenyls as inducers of hepatic microsomal enzymes: structure-activity rules, Chem.-Biol. Interact., 30 (1980) 271. 13 D.W. Nebert, H.J. Eisen, J. Negishi, M.A. Lang, L.M. Hjelmeland and A.B. Okey, Genetic mechanisms controlling the induction of polysubstrate monooxygenase (P-450) activities, Annu. Rev. Pharmacol. Toxicol., 21 (1981) 431. 14 A. Poland, E. Glover and A.S. Kende, Stereospecific, high affinity binding of 2,3,7,8tetrachlorodibenzo-p-dioxin by hepatic cytosol, J. Biol. Chem., 251 (1976) 4936. 15 A.B. Okey, G.P. Bondy, M.E. Mason, D.W. Nebert, C.J. Forster-Gibson, J. Muncan and M.J. Dufresne, Temperature-dependent cytosol-to-nucleus translocation of the A h receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin in continuous cell culture lines, J. Biol. Chem., 255 (1980) 11415. 16 A.B. Okey, G.P. Bondy, M.E. Mason, G.F. Kahl, H.J. Eisen, T.M. Guenthner and D.W. Nebert, Regulatory gene product of the A h locus, J. Biol. Chem., 254 (1979) 11636. 17 W.F. Greenlee and A. Poland, Nuclear uptake of 2,3,7,8-tetrachlorodibenzo-p-dioxin in C57BL/6J and DBA/2J mice, J. Biol. Chem., 254 (1979) 9814. 18 B. Tierney, D. Weaver, N.H. Neitz, W.I. Schaeffer and E. Bresnick, The identity and nuclear uptake of a cytosolic binding protein for 3-methylcholanthrene, Arch. Biochem. Biophys., 200 {1980) 513. 19 J.M.B. Carlstedt-Duke, G. Elfstr6m, B. H6gberg and J.-A. Gustafsson, Ontogeny of the rat hepatic receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin and its endocrine independence, Cancer Res., 39 {1979) 4653. 20 J.C. Nunnink, A.H.L. Chuang and E. Bresnick, The ontogeny of nuclear aryl hydrocarbon hydroxylase, Chem.-Biol. Interact., 22 (1979) 225. 21 O. Hutzinger, S. Safe and V. Zitko, Polychlorinated biphenyls: synthesis of some individual chlorobiphenyls, Bull. Environ. Contain. Toxicol., 6 (1971 ) 209. 22 A. Parkinson, L.W. Robertson, L. Safe and S. Safe, Polychlorinated biphenyls as inducers of hepatic microsomal enzymes: effects of d i - o r t h o substitution, Chem.-Biol. Interact., 35 (1981) 1. 23 G.D. Vieth, N.M. Austin and R.T. Morris, A rapid method for estimating log P for organic chemicals, Water Res., 13 (1979) 43. 24 B. McDuffie, Estimation o f octanol/water partition coefficients for organic pollutants using reverse-phase HPLC, Chemosphere, 10 (1981) 73. 25 L. Renberg, G. SundstrSm and K. Sundh-Nygard, Partition coefficients of organic chemicals derived from reversed phase thin layer chromatography. Evaluation of methods and application on phosphate esters, polychlorinated paraffins and some PCB-substitutes, Chemosphere, 9 {19801 683. 26 O.H. Lowry, N.J. Rosebrough, A.L. Fan" and R.J. Randall, Protein measurement with folin phenol reagent, J. Biol. Chem., 193 (1951) 265. 27 H. Yoshimura, S. Yoshimura, N. Ozawa and M. Miki, Possible correlation between induction modes of hepatic enzymes by PCBs and their toxicity in rats, Ann. N.Y. Acad. Sci., 320 (1979) 179. 28 D.A. Haugen, M.J. Coon and D.W. Nebert, Induction of multiple forms of mouse liver cytochrome P-450, J. Biol. Chem., 251 (1976) 1817. 29 A.H. Conney, Pharmacological implications of microsomal enzyme induction, Pharmacol. Rev., 19 (1967) 317. 30 A. Poland, W.F. Greenlee and A.S. Kende, Studies on the mechanisms of action of the chlorinated dibenzo-p-dioxins and related compounds, Ann. N.Y. Acad. Sci., 320 (1979) 214. 31 W.N. Schmidt, M.A. Sadler and B.S. Katzenellenbogen, Androgen-uterine interaction: Nuclear translocation of the estrogen receptor and induction of the uterine-

277 induced protein (IP) by high concentration of androgens in vitro but not in vivo, Endocrinology, 98 (1976) 702. 32 K.B. Horwitz and W.L. McGuire, Estrogen and progesterone: their relationship in hormone-dependent breast cancer, in: W.L. McGuire, J.-P. Raynaud and E.-E. Baulieu (Eds.), Progesterone Receptors in Normal and Neoplastic Tissues, Raven Press, New York, 1977, p. 103.