Effects of Aroclors and individual PCB congeners on activation of the human androgen receptor in vitro

Reproductive Toxicology 17 (2003) 15–23

Effects of Aroclors and individual PCB congeners on activation of the human androgen receptor in vitro Timothy J. Schrader a , Gerard M. Cooke a,b,∗ a

Toxicology Research Division, Health Products and Foods Branch, Food Directorate, Health Canada, Sir Frederick G. Banting Research Centre, 2202D1 Tunney’s Pasture, Ottawa, Ont., Canada K1A 0L2 b Reproductive Biology Unit, and Departments of Cellular and Molecular Medicine and Obstetrics and Gynecology, University of Ottawa, 451 Smyth Road, Ottawa, Ont., Canada K1A 8M5 Received 27 April 2002; received in revised form 23 July 2002; accepted 11 August 2002

Abstract To investigate possible interactions between the human androgen receptor and PCBs in vitro, we have used a previously characterized human androgen receptor reporter gene assay in which PC-3 LUCAR+ cells respond to 5␣-dihydrotestosterone (DHT, 50 pM) with enhanced luciferase activity. The effects of Aroclors, commercial mixtures of PCBs, or polychlorinated terphenyls (PCTs) (0, 0.1, 1.0, and 10.0 ␮M) on luciferase activity in PC-3 LUCAR+ cells were determined after exposure for 18 h in the presence and absence of DHT (50 pM). In the absence of DHT, none of the Aroclors induced luciferase activity but, in the presence of DHT (50 pM), Aroclors 1016, 1221, 1232, 1242, 1248, 1254, 1260, 5432, and 5442 acted antagonistically at concentrations that did not affect cell viability. Aroclor 5460 was without effect. Similarly, when PCBs found as human milk contaminants were assessed as individual congeners (each at 1 ␮M, where no cytotoxic effects were observed), none activated luciferase expression in the absence of DHT but PCBs 49, 66, 74, 105, and 118 completely antagonized the stimulation by DHT (50 pM) and PCBs 138, 153, and 156 were less effective antagonists, reducing the DHT stimulation by about 50%. Thus, 30% (by weight) of the PCBs in human milk are androgen antagonists (PCBs 66, 74, 105, and 118) and a further 25% are partial antagonists (PCBs 138, 153, and 156). A proportionally representative mixture of PCBs that contaminate human milk also caused the DHT-mediated activation of luciferase activity in PC-3 LUCAR+ cells to be reduced by more than 50%. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Androgen; Receptor; Polychlorinated biphenyl; Human

1. Introduction Male reproductive function requires testicular testosterone for spermatogenesis and 5␣-dihydrotestosterone (DHT) for androgen target tissue function [1,2]. Testosterone or DHT interacts with the androgen receptor causing transcription of specific genes [2]. Chemicals that interfere with the binding of androgens to the androgen receptor result in abnormalities in sexual development [3–7]. Recently, concern has been raised regarding the possible endocrine disruptive capability of chemicals that persist in nature and accummulate in body tissues. Some chemicals are considered to act as estrogen mimics because they antagonize the effects of estradiol in vitro by interacting with the estrogen receptor, and because their toxicity in vivo resembles the effects of premature estrogen exposure [8–11]. While it is clear that in vitro and in vivo assays of estrogenicity have ∗ Corresponding author. Tel.: +1-613-957-0990; fax: +1-613-941-6959. E-mail address: gerard [email protected] (G.M. Cooke).

shown that certain chemicals act as estrogens, the possibility that some chemicals may exert their effects by interfering with the binding of androgens to the androgen receptor has received comparatively little attention. However, the fungicide vinclozolin was reported to mediate its developmental effects as an antiandrogen [12] and p,p -DDE, which is one of the most abundant organochlorine chemicals in the environment and which does not bind to the estrogen receptor (or at least, binds very poorly) has been shown to act as an antiandrogen [13,14]. More recently, other “environmental estrogens” have been shown to be antiandrogenic in vitro [15]. Consequently, the developing male reproductive tract is a target for inappropriate organochlorine exposure that may cause such reproductive pathologies as cryptorchidism, testicular cancer, and prostate inflammation, and could lead to infertility later in adulthood. Under normal circumstances, the stage at which the developing human receives the greatest amount of organochlorine chemicals is during breast feeding due to the high content of these compounds in lipid of human milk [16]. Later in life,

0890-6238/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 8 9 0 - 6 2 3 8 ( 0 2 ) 0 0 0 7 6 - X

16

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

humans are exposed to organochlorine chemicals from food and possibly from direct exposure to commercial preparations [16]. A major component of the organochlorines to which humans are exposed are polychlorinated biphenyls (PCBs). For a breast-fed human infant, the average daily intake of PCBs will be in the range of 1.8 to 14 ␮g/kg body weight ([17] for review), although the intake may be lower for second and third children from the same mother [18] (for review, [19]). PCBs have been shown to be reproductive toxicants in animal models [20–23], to be embryotoxic in vitro [24], and to cause behavioural changes and intellectual impairment in primates [22,25,26]. PCBs have been considered to be either estrogenic, antiestrogenic, or nonestrogenic, depending upon their interaction with the estrogen receptor [27]. In addition to the estrogen receptor, some methylsulphonyl PCBs have been shown to interact with the glucocorticoid receptor [28] and some hydroxylated-PCBs have been shown to interact with the thyroid receptor [29] but to our knowledge, there have been no studies of the interaction between PCBs and the androgen receptor. Therefore, to investigate the intriguing possibility that some PCBs and polychlorinated terphenyls (PCTs) may be acting as antiandrogens, an androgen-responsive reporter gene assay that had been developed by cotransfecting human PC-3 androgen-insensitive prostatic carcinoma cells with a human wild type androgen receptor cDNA expression vector and a mammary tumour virus (MMTV)–luciferase reporter construct [30] was used to assess the interaction between the human androgen receptor and PCBs and PCTs.

2. Materials and methods 2.1. Chemicals and tissue culture materials The human PC-3 and 22Rv1 prostate cell lines were purchased from the American Type Culture Collection, Rockville, MD, USA. The human androgen receptor expression vector, pCMV5–hAR, which contains the full cDNA coding sequence for the human androgen receptor constitutively transcribed through the cytomegalovirus (CMV) immediate-early enhancer/promoter region, was the kind gift of Dr. T.R. Brown, School of Hygiene and Public Health, Johns Hopkins University, Baltimore, MD, USA. The reporter plasmid pMAMneo-luc, containing the luciferase gene under transcriptional control of the androgen- and glucocorticoid-responsive MMTV long terminal repeat was purchased from Clontech, Palo Alto, CA, USA. The PRL–CMV vector encoding Renilla luciferase, luciferase assay and lysis buffer, and dual luciferase buffer were purchased from Promega, Madison, WI, USA. Sterile tissue culture plasticware was purchased from Corning Costar Corporation, Cambridge, MA, USA and Becton Dickinson, Franklin Lakes, NJ, USA, while phenol red-free RPMI medium containing glutamine, and trypsin/EDTA were obtained from Gibco, Gaithersburg,

MD, USA. Fetal bovine serum was purchased from ICN, Costa Mesa, CA, USA and charcoal/dextran, as well as 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Sigma, St. Louis, MO, USA. 5␣-DHT and estradiol-17␤ were purchased from Steraloids Inc., Newport, RI, USA and were recrystallized before use. Aroclors were purchased from Monsanto Chemical Company, St. Louis, MO, USA. Individual PCB congeners were from Ultra Scientific, N. Kingstown, RI, USA except for the 2,3 ,4,4 ,5-pentachlorobiphenyl (PCB 118, 95% pure) which was a gift from Dr. W.H. Newsome (Health Products and Food Branch, Ottawa, Ont., Canada) and 2,4,4 ,5-tetra-, 2,3,3 ,4,4 -penta-, 2,2 ,3,4,4 ,5,5 -hepta-, and 2,2 ,3,4 ,5,5 ,6-heptachlorobiphenyls (PCBs 74, 105, 180, and 187) which were gifts from Dr. S. Safe (Texas A&M University, College Station, Texas, TX, USA). The mixture of PCBs proportionally representing human milk PCB contamination was prepared according to the concentrations reported by Mes et al. [31]. 2.2. Cell culture and transfection Cotransfection and characterization of the PC-3 LUCAR+ cell line produced by cotransfection of human PC-3 cells with human androgen receptor expression vector pCMV5–hAR and androgen responsive reporter vector pMAMneo-luc has been described previously [30]. In this system, the Kd for DHT activation of the androgen receptor is ∼50 pM and, unless otherwise stated, this concentration was used in the present studies in order to ensure that activation and inhibition of androgen receptor activation by test chemicals was observed. The specificity of the system was also examined by testing the ability of nonandrogenic steroids to activate the system and also by the use of flutamide to inhibit the activation by DHT [30]. PC-3 LUCAR+ cells were maintained in 75 cm2 tissue culture flasks containing 15 ml RPMI medium supplemented with 10% fetal bovine serum stripped of steroids by charcoal/dextran [32], and gentamycin. Cells were passaged by trypsinization as required. For assaying PCB effects, PC-3 LUCAR+ cells (80,000 in 1 ml medium) were cultured in 96-well plates at 20,000 cells (250 ␮l medium) per well for 24 h prior to the addition of DHT (50 pM) and/or chemicals of interest at 0, 0.1, 1.0, or 10.0 ␮M. Chemicals and steroids were added in 1 ␮l DMSO such that cultures were never exposed to greater than 0.4% DMSO. There were no indications of any problem with solubility, possibly because the assays contained 10% fetal calf serum and the organochlorines were added in DMSO. Another human prostatic cell line, 22Rv1, was transiently cotransfected with pCMV5–hAR and pMAMneoluc vectors, as well as PRL–CMV to correct for transfection efficiency, and used to confirm the results obtained with the stable PC-3 LUCAR+ cells. For these experiments, 22Rv1 cells, grown in supplemented phenol red free RPMI medium, were seeded in 24-well dishes at 50,000 cells/well and allowed to attach

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

overnight. The medium was then removed, replaced with 150 ␮l/well fresh supplemented medium, and 100 ␮l/well transfection mixture added [400 ng/well pCMV5–hAR and pMAMneoluc, 200 ng/well PRL–CMV using GeneJammer transfection reagent (Stratagene, La Jolla, CA, USA)]. After incubation (37 ◦ C, 3 h), 250 ␮l/well of supplemented medium was added and the wells were dosed with chemical or solvent for control and incubated as above.

17

nescence). Each of the chemicals of interest was examined in at least three separate assays. 22Rv1 cells were lysed with 50 ␮l lysis buffer, 20 ␮l aliquots transferred to white 96-well plates. Luciferase activity was determined using dual luciferase reagent by first injecting 100 ␮l firefly luciferase reagent (10-s read period) to determine inducible firefly luciferase activity followed by 100 ␮l “stop and glow” reagent (using a 2-s delay period followed by 10-s read period) to determine Renilla luciferase activity.

2.3. Firefly luciferase assay procedure 2.4. MTT assay for toxicity After an 18-h incubation for the PC-3 LUCAR+ or a 72-h incubation for the 22Rv1 transient transfectants, the medium was gently removed from the plates by blotting on filter paper. For PC-3 LUCAR+ cells, lysis buffer (20 ␮l) was then added and the plates agitated gently for 20 min. The plates were read on a multiwell plate reading luminometer (Model MLX, Dynex, Winooski, VA, USA; 50 ␮l luciferase reagent injected/well, luminescent signal quantitated over a 10-s read interval following a 0.5-s delay to avoid nonspecific lumi-

The effects of PCBs and PCTs on cell viability were assessed by staining parallel cultures with MTT using a modified technique of Denizot and Lang [33]. Briefly, cells were resuspended at 80,000 cells/ml in supplemented RPMI medium lacking phenol red, seeded at 20,000 cells/well (in 250 ␮l medium) in 96-well tissue culture grade multiwell plates, and allowed to attach overnight. Test chemical dissolved in DMSO or DMSO alone was added (1 ␮l/well, or

Fig. 1. The effect of Aroclors 1016, 1221, 1232, and 1242 on luciferase activity in PC-3 LUCAR+ cells in the presence and absence of DHT. PC-3 LUCAR+ cells were incubated for 18 h with Aroclors (0, 0.1, 1.0, and 10.0 ␮M) in the presence or absence of DHT (50 pM) as described in Section 2. (∗) Significantly different compared with control (100%, no DHT, no Aroclor). (†) Significantly different compared with DHT (50 pM, no Aroclor) using arcsine transformed proportionalized data from three separate studies.

18

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

0.4%) and the cultures incubated for 72 h. After the incubation period, 50 ␮l of medium was removed from each well and replaced with 50 ␮l/well MTT (5 mg/ml in PBS, filter-sterilized). The plates were then incubated for an additional 2 h. After the incubation period, the plates were inverted to gently remove the medium and blotted against filter

paper. Acidified isopropanol (100 ␮l/well) was then added and the plates incubated in the dark with shaking to dissolve the coloured formazan product. The absorbance of the converted dye was then measured at 570 nm with a reference wavelength of 630 nm on a ␮Quant microplate spectrophotometer (Biotek, Winooski VT, USA).

Fig. 2. The effect of Aroclors 1248, 1254, and 1260 on luciferase activity in PC-3 LUCAR+ cells in the presence and absence of DHT. PC-3 LUCAR+ cells were incubated for 18 h with Aroclors (0, 0.1, 1.0, and 10.0 ␮M) in the presence or absence of DHT (50 pM) as described in Section 2. (∗) Significantly different compared with control (100%, no DHT, no Aroclor). (†) Significantly different compared with DHT (50 pM, no Aroclor) using arcsine transformed proportionalized data from three separate studies.

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

2.5. Statistical analyses

3. Results

Significant effects of the chemicals of interest on luciferase activity were determined by one-way ANOVA (Sigmastat 2.0, Jandel Scientific, 1992, 1995, San Rafael, CA, USA) using arcsine transformed proportionalized data. EC50 values were determined from the equations for linear regression analysis.

3.1. The effect of Aroclors on androgen receptor activation

19

Aroclors 1016, 1221, 1232, and 1242 in the absence of DHT had no effect on the PC-3 LUCAR+ cell response but at 1.0 ␮M, completely ablated the activation of the androgen

Fig. 3. The effect of Aroclors 5432, 5442, and 5460 (which contain PCTs) on luciferase activity in PC-3 LUCAR+ cells in the presence and absence of DHT. PC-3 LUCAR+ cells were incubated for 18 h with Aroclors (0, 0.1, 1.0, and 10.0 ␮M) in the presence or absence of DHT (50 pM) as described in Section 2. (∗) Significantly different compared with control (100%, no DHT, no Aroclor). (†) Significantly different compared with DHT (50 pM, no Aroclor) using arcsine transformed proportionalized data from three separate studies.

20

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

Table 1 Estimates of EC50 values for the inhibition of DHT-mediated activation of luciferase activity in PC-3 LUCAR+ cells by Aroclors containing PCBs (10xx and 12xx) and Aroclors containing PCTs (54xx) Aroclor

EC50 (␮M)

r2

1016 1221 1232 1242 1248 1254 1260 5432 5442 5460

0.51 0.29 0.27 0.25 0.25 0.42 0.58 1.11 0.85 ∞

0.99 0.64 0.73 0.77 0.77 0.95 0.80 0.97 0.63 0.07

The data shown in Figs. 1–3 (with the exception of the data for 10 ␮M since cytotoxicity was observed with this concentration of Aroclors) were analysed using linear regression analysis and then the EC50 values were obtained by substitution. The symbol (∞) indicates no reliable value determinable.

receptor by DHT (50 pM). Aroclors 1221 and 1232 were also inhibitory at 0.1 ␮M (P < 0.05, Fig. 1). At concentrations of 0.1 and 1.0 ␮M, these Aroclors were not cytotoxic (cell survival >97%) but at 10 ␮M, they reduced viability to below 15%. Similarly, Aroclors 1248, 1254, and 1260, in the absence of DHT, were without effect but in the presence of DHT (50 pM) these Aroclors ablated the stimulation by DHT (P < 0.05, Fig. 2). Cell survival was identical to controls at Aroclor concentrations of 0.1 and 1.0 but at 10 ␮M, cell survival was 5, 39, and 56%, respectively. Aroclors 5432, 5442, and 5460 were also without effect in the absence of DHT. In the presence of DHT (50 pM), Aroclors 5432 and 5442 were partially inhibitory at 1 ␮M but without effect at 0.1 ␮M and Aroclor 5460 was without effect at any concen-

Table 2 PCBs that were inhibitory with PC-3 LUCAR+ cells (Figs. 4–6) were incubated at 0.1 and 1.0 ␮M with 22Rv1 cells transiently transfected with the human androgen receptor, the luciferase reporter, and renilla luciferase (to enable correction for transfection efficiency) PCB

0.1 ␮M

49 66 74 105 118 138 153 156 HM

75.0 86.4 91.5 52.1 52.1 100.0 81.4 91.0 84.2

± ± ± ± ± ± ± ± ±

1.0 ␮M 11.0 4.7a 1.4a 2.7a 11.0a 5.9 11.3 8.1 5.8a

80.9 86.6 73.9 68.5 76.1 93.1 82.7 86.5 64.8

± ± ± ± ± ± ± ± ±

4.9a 4.1a 10.0a 1.5a 19.1a 4.7 3.8a 16.9 5.9a

Data from three separate assays are presented as mean percentages ± S.E.M. of the activity in controls stimulated with 100 pM DHT. HM is a PCB mixture that is proportionally representative of the PCB contamination of human milk. a Significantly different compared with controls (100%) by one-way ANOVA of arcsine transformed proportionalized data (P < 0.05).

tration (Fig. 3). Cytotoxicity was only seen at 10 ␮M (cell survival was 59, 87, and 73% respectively compared with controls). Estimates of EC50 values for most of the Aroclors were similar (between 0.2 and 0.6 ␮M) (Table 1) but for Aroclors 5432 and 5442 the EC50 values were higher (1.1 and 0.85 ␮M) and for Aroclor 5460, where no inhibition was observed, no reliable value for the EC50 could be determined. 3.2. The effect of individual PCB congeners on androgen receptor activation When individual PCB congeners (1.0 ␮M) were assessed for their effects on the androgen receptor in the absence

Fig. 4. The effect of PCB congeners 28, 49, 66, 74, 99, 105, 118, 137, 138, 153, and 156 on luciferase activity in PC-3 LUCAR+ cells in the presence and absence of DHT. PC-3 LUCAR+ cells were incubated for 18 h with PCBs (1 ␮M) in the presence or absence of DHT (50 pM) as described in Section 2. (∗) Significantly different compared with control (100%, no DHT, no PCB) (†) Significantly different compared with DHT (50 pM, no PCB) using arcsine transformed proportionalized data from three separate studies.

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

21

Fig. 5. The effect of PCB congeners 157, 170, 180, 183, 187, 194, 201, and 203 on luciferase activity in PC-3 LUCAR+ cells in the presence and absence of DHT. PC-3 LUCAR+ cells were incubated for 18 h with PCBs (1 ␮M) in the presence or absence of DHT (50 pM) as described in Section 2. (∗) Significantly different compared with control (100%, no DHT, no PCB). (†) Significantly different compared with DHT (50 pM, no PCB) using arcsine transformed proportionalized data from three separate studies.

of DHT, none were able to activate (Figs. 4 and 5). In the presence of DHT (50 pM), PCBs 28, 49, 66, 74, 105, and 118 antagonized the action of DHT, reducing the luciferase activity to levels that were not significantly different from the activity in the absence of DHT. PCBs 138, 153, and 156 were partially antagonistic reducing the DHT activation by about 50%. None of the PCBs were cytotoxic (cell survival ranging from 90 to 100%). Confirmation of these findings

Fig. 6. The effect of a proportional mixture of PCBs that are found in human milk on luciferase activity in PC-3 LUCAR+ cells in the presence and absence of DHT. PC-3 LUCAR+ cells were incubated for 18 h with the PCB mixture (1 ␮M) in the presence or absence of DHT (50 pM) as described in Section 2. Bars represent the mean and S.D. from three separate studies. (a, b, c) Bars with different superscripts are significantly different (P < 0.05).

was obtained using transiently transfected 22RV1 cells where PCBs not only inhibited DHT-mediated activation of the androgen receptor at 1.0 ␮M concentration but also at 0.1 ␮M (Table 2).

Fig. 7. The effect of a proportional mixture of PCBs that are found in human milk on luciferase activity in 22Rv1 cells transiently transfected with the human androgen receptor and renilla luciferase. 22Rv1 cells were incubated for 18 h with the PCB mixture (0, 0.1, 1.0 ␮M) in the presence or absence of DHT (100 pM) as described in Section 2. Bars represent the mean and S.D. from quadruplicate estimations. (a, b, c) Bars with different superscripts are significantly different by one way ANOVA of arcsine transformed proportionalized data (P < 0.05).

22

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23

3.3. The effect of a mixture of PCB congeners found in human milk on androgen receptor activation When the proportionally representative mixture of PCBs (at 1 ␮M) that are found in human milk (HM) was assessed for effects on the androgen receptor, the activation by DHT (50 pM) was attenuated by more than 50% (Fig. 6). This mixture had no effect in the absence of DHT nor was it cytotoxic at this concentration. Confirmation of these findings was obtained using transiently transfected 22RV1 cells where HM not only inhibited androgen receptor activation by DHT at 1 ␮M concentration but also at 0.1 ␮M (Table 2, Fig. 7).

4. Discussion The possibility that some industrial chemicals may act as antiandrogens was first reported for vinclozolin [12] and p,p -DDE [13] and more recently for DDT and butyl benzyl phthalate [15]. In the present studies, we used PC-3 LUCAR+ cells to investigate the possibility that some PCBs may interact with the androgen receptor in vitro. It was found that several Aroclors act as androgen antagonists at concentrations that are not cytotoxic and that some individual PCB congeners ablate the activation of the androgen receptor by DHT. These findings led to the investigation of the effects of a proportionally representative mixture of PCBs that contaminate human milk, which demonstrated that activation of the androgen receptor by the natural ligand could be antagonized by this mixture. Comparison of the PCB components in human milk and the effects of the individual PCB congeners observed above, indicates that 30% (by weight) of the PCBs in human milk would antagonize the effect of DHT and none would activate. The individual PCB congeners that were used in the present studies make up most of the weight of PCBs in human milk [31]. Of the 18 congeners tested, 15 contain chlorine at positions 2,4,4 and, therefore, any consideration of structure–activity relationships regarding the interaction of PCBs with the androgen receptor is limited by this restriction. However, some tentative conclusions are possible. The inhibitory PCBs 66, 74, 105, 118, and 156 are mono-ortho-substituted and are considered coplanar. Of those exhibiting little or no inhibition, PCBs 99, 137, 170, 180, 183, 187, 194, 201, and 203 are di-ortho-substituted and are considered noncoplanar. PCBs 28 and 157 were the only coplanar PCBs that were poor inhibitors of androgen receptor activation and PCBs 49, 138, and 153 were the only noncoplanar PCBs that were considered to be inhibitory. Thus in general, coplanar PCBs with 4, 5, or 6 chlorines interact with the androgen receptor, in contrast to the glucocorticoid receptor where interaction was seen with methylsulphonyl PCBs that were tri-ortho-substituted and therefore, noncoplanar [28]. Estimates of the average daily PCB intake of breast-fed infants range from 1.8 to

14 ␮g/kg body weight/day [17]. Thus, the PCB concentration in infant tissues would be approximately 5–40 nM assuming no accumulation and an even distribution throughout the body. In reality, infants are probably accumulating PCBs and thus, the concentration in certain tissues may be much higher. In our studies, inhibition of androgen receptor activation in vitro was seen with some PCBs at 100 nM. There have been few studies of human neonatal reproductive development but it is known that human male infants have high serum androgen levels during the first 3 months of postnatal life and the levels are approximately 50% of the adult level [34,35]. In addition, the androgen receptor is clearly present in at least one androgen target tissue, the foreskin, in the first month or so of life at a concentration equivalent to the adult level (although not as high as the adolescent level) [36]. It is not fully understood why human male babies exhibit such androgenic physiology in the first few months of life but it is thought that the subsequent behaviour of the individual is imprinted at this time [37]. Bearing in mind that interference in androgen dependent processes at this early age may have significant consequences for later sexual development, some diminution of androgen dependent processes may be expected during the months of breast-feeding in cases where the mother has been exposed to high levels of PCB contamination. It is thus encouraging that recent reports show that the body burdens and human milk levels of PCBs are decreasing in populations living in areas where PCB food contamination is high and where the human plasma, tissue, and human milk levels were much higher than in the general population [38–40]. The reasons for the decline probably include restriction of PCB release into the environment and the heeding of health advisories recommending limits on the intake of contaminated foods.

Acknowledgments The authors gratefully acknowledge the excellent technical assistance of Irene Langlois, Carole Whalen, and Georges Angers. References [1] Sharpe RM. Regulation of spermatogenesis. In: Knobil E, Neill JD, editors. The physiology of reproduction, vol. 1. 2nd ed. New York: Raven Press Ltd.; 1994, p. 1363–1434. [2] Luke MC, Coffey DS. The male sex accessory tissues. In: Knobil E, Neill JD, editors. The physiology of reproduction, vol. 1. 2nd ed. New York: Raven Press Ltd.; 1994, p. 1435–87. [3] Husmann DA, McPhaul MJ. Time-specific androgen blockade with flutamide inhibits testicular descent in the rat. Endocrinology (Baltimore) 1991;129:1409–16. [4] Spencer JR, Torrado T, Sanchez RS, Darracot Vaughn Jr E, Imperato-McGinley J. Effects of flutamide and finasteride on rat testicular descent. Endocrinology (Baltimore) 1991;129:741–8.

T.J. Schrader, G.M. Cooke / Reproductive Toxicology 17 (2003) 15–23 [5] van der Schoot P. Disturbed testicular descent in the rat after prenatal exposure to the antiandrogen flutamide. J Reprod Fertil 1992;96:483– 96. [6] Silversides DW, Price CA, Cooke GM. Effects of short-term exposure to hydroxyflutamide in utero on the development of the reproductive tract in male mice. Can J Physiol Pharmacol 1995;73:1582–8. [7] Gray Jr LE, Wolf C, Lambright C, Mann P, Price M, Cooper RL, et al. Administration of potentially antiandrogenic pesticides (procymidone, linuron, iprodione, chlozolinate, p,p -DDE and ketoconazole) and toxic substances (dibutyl- and diethylhexyl phthalate, PCB 169 and ethane dimethane sulphonate) during sexual differentiation produces diverse profiles of reproductive malformations in the male rat. Toxicol Ind Health 1999;15:94–118. [8] Colborn T, vom Saal FS, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993;101:378–84. [9] Safe S. Enviromental and dietary estrogens and human health: is there a problem? Environ Health Perspect 1995;103:346–51. [10] Cooper RL, Kavlok RJ. Endocrine disrupters and reproductive development: a weight-of-evidence overview. J Endocrinol 1997;152: 159–66. [11] Turner KJ, Sharpe RM. Environmental oestrogens—present understanding. Rev Reprod 1997;2:69–73. [12] Gray Jr LE, Ostby JM, Kelce WR. Developmental effects of an environmental antiandrogen: the fungicide vinclozolin alters sex differentiation of the male rat. Toxicol Appl Pharmacol 1994;129: 46–52. [13] Kelce WR, Stone CR, Laws SC, Gray Jr LE, Kemppainen JA, Wilson EM. Persistent DDT metabolite p,p -DDE is a potent androgen receptor antagonist. Nature (Lond) 1995;375:581–5. [14] Kelce WR, Lambright CR, Gray Jr LE, Roberts KP. Vinclozolin and p,p -DDE alter androgen-dependent gene expression: in vivo confirmation of an androgen receptor-mediated mechanism. Toxicol Appl Pharmacol 1997;142:192–200. [15] Sohoni P, Sumpter JP. Several environmental oestrogens are also anti-androgens. J Endocrinol 1998;158:327–39. [16] Ryan JJ, Norstrom RJ. Occurrence of polychlorinated dibenzo-pdioxins and dibenzofurans in humans and major exposure routes. In: Rappe C, Buser HR, Dodet B, O’Neill IK, editors. Environmental carcinogens: methods of analysis and exposure measurement, vol. 11: polychlorinated dioxins and dibenzofurans. WHO-IARC Scientific Publications and Oxford, UK: Oxford University Press; 1991, p. 51–104 [chapter 4]. [17] Jensen AA, Slorach SA. Assessment of infant intake of chemicals via breast milk. In: Jensen AA, Slorach SA, editors. Chemical contaminants in human milk. Boca Raton, Ann Arbor, Boston: CRC Press Inc.; 1991, p. 215–22 [chapter 8]. [18] Jensen AA, Slorach SA. Factors affecting the levels of residues in human milk. In: Jensen AA, Slorach SA, editors. Chemical contaminants in human milk. Boca Raton, Ann Arbor, Boston: CRC Press Inc.; 1991, p. 199–207 [chapter 6]. [19] Abraham K, Papke O, Gross A, Kordonouri O, Wiegand S, Wahn U, et al. Time course of PCDD/PCDF/PCB concentrations in breast-feeding mothers and their infants. Chemosphere 1998;37: -1731–41. [20] Sager DB, Girard DM, Nelson D. Early postnatal exposure to PCBs: sperm function in rats. Environ Toxicol Chem 1991;10:737–46. [21] Sager DB, Girard DM. Long-term effects on reproductive parameters in female rats after translactational exposure to PCBs. Environ Res 1994;66:52–76. [22] Brouwer A, Ahlborg UG, van der Berg M, Birnbaum LS, Boersma ER, Bosveld B, et al. Functional aspects of developmental toxicity

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

23

of polyhalogenated aromatic hydrocarbons in experimental animals and human infants. Eur J Pharmacol 1995;293:1–40. Faqi AS, Dalsenter PR, Mathar W, Heinrich-Hirsch B, Chahoud I. Reproductive toxicity and tissue concentrations of 3,3 ,4,4 -tetrachlorobiphenyl (PCB 77) in male adult rats. Hum Exp Toxicol 1998;17:151–6. Krogenæs AK, Nafstad I, Skåre U, Farstad W, Hafne A-L. In vitro reproductive toxicity of polychlorinated biphenyl congeners 153 and 126. Reprod Toxicol 1998;12:575–80. Tilson HA, Jacobson JL, Rogan WJ. Polychlorinated biphenyls and the developing nervous system: cross-species comparisons. Neurotoxicol Teratol 1990;12:239–48. Jacobson JL, Jacobson SW. Intellectual impairment in children exposed to polychlorinated biphenyls in utero. N Engl J Med 1996;335:783–9. Sonnenschein C, Soto AM. An updated review of environmental estrogen and androgen mimics and antagonists. J Steroid Biochem Mol Biol 1998;65:143–50. Johansson M, Nilsson S, Lund B-O. Interactions between methylsulphonyl PCBs and the glucocorticoid receptor. Environ Health Perspect 1998;106:769–72. Cheek AO, Kow K, Chen J, McLachlan JA. Potential mechanisms of thyroid disruption in humans: interaction of organochlorine compounds with thyroid receptor, transthyretin and thyroid-binding globulin. Environ Health Perspect 1999;107:273–8. Schrader TJ, Cooke GM. Examination of selected food additives and organochlorine food contaminants for androgenic activity in vitro. Toxicol Sci 2000;53:278–88. Mes J, Turton D, Davies D, Sun WF, Lau PY, Weber D. The routine analysis of some specific isomers of polychlorinated biphenyl congeners in human milk. Int J Environ Anal Chem 1987;28:197– 205. Borras M, Hardy L, Lempreur F, El khissiin AH, Legros N, Gol-Winkler R, et al. Estradiol-induced down-regulation of estrogen receptor: effect of various modulators of protein synthesis and expression. J Steroid Biochem Mol Biol 1994;48:325–36. Denizot F, Lang R. Rapid colorimetric assay for cell growth and survival—modification to the tetrazolium dye procedure giving improved sensitivity and reliability. J Immunol Methods 1986;89:271–7. Forest MG, Sizonenko PC, Cathiard AM, Bertrand J. Hypophysogonadal function in humans during the first year of life: 1. Evidence for testicular activity in early infancy. J Clin Invest 1974;53:819–28. Bolton NJ, Tapanainen J, Koivisto M, Vihko R. Circulating sex hormone-binding globulin and testosterone in newborns and infants. Clin Endocrinol 1989;31:201–7. Roehrborn CG, Lange JL, George FW, Wilson JD. Changes in amount and intracellular distribution of androgen receptor in human foreskin as a function of age. J Clin Invest 1987;79:44–7. Sachs BD, Meisel RL. The physiology of male sexual behaviour. In: Knobil E, Neill J, editors. The physiology of reproduction, vol. 2. 1st ed. New York: Raven Press Ltd.; 1988, p. 1444–7 [chapter 34]. Atuma SS, Hansson L, Johnsson H, Slorach S, de Wit CA, Lindstrom G. Organochlorine pesticides, polychlorinated biphenyls and dioxins in human milk from Swedish mothers. Food Addit Contam 1998;15:142–50. Fitzgerald EF, Hwang SA, Bush B, Cook K, Worswick P. Fish consumption and breast milk PCB concentrations among Mohawk women at Akwesasne. Am J Epidemiol 1998;148:164–72. Newsome WH, Ryan JJ. Toxaphene and other chlorinated compounds in human milk from northern and southern Canada: a comparison. Chemosphere 1999;39:519–26.