- Email: [email protected]
Congener-specific effects of pcbs on contractions of pregnant rat uteri
Reproductive. Toxicology, Vol. 10, No. I, pp. 21-28, 1996 Copyright 0 1996 Elsevier Science Inc. Printed in the USA. All rights reserved 0890-6238/96 $15.00 + 00 ELSEVIER
0890-6238(95)02014-4
CONGENER-SPECIFIC EFFECTS OF PCBS ON CONTRACTIONS OF PREGNANT RAT UTERI MEI-LING TsAI,*‘~ R. CLINTONWEBB,? and RITA LOCH-CARUSO* *Departments of Environmental and Industrial Health and tPhysiology, University of Michigan, Ann Arbor, Michigan Abstract - Both increased and decreased gestation lengths have been reported following exposures to polychlorinated biphenyl (PCB) mixtures and congeners. Because oscillatory uterine contractions are essential for parturition, we hypothesized that the disparate findings on gestation length may be the result of distinct PCB congener-specific actions on oscillatory uterine contractions. This study examined the acute effects of PCB congeners on isometric contractions of isolated pregnant uteri and the structure-activity relationship for individual congeners. After cumulative exposure to individual PCB congeners (0.5 pM to 150 pM), oscillatory contractions were: 1) not altered by 2,4,5,2’,4’,5’-hexachlorobiphenyl, 3,4,5,3’,4’-pentachlorobiphenyl, or 3,4,3’,4’-tetrachlorobiphenyl; 2) significantly inhibited by 4-hydroxy-2’,4’,6’-trichlorobiphenyl; and 3) markedly increased by 2,4,6-trichlorobiphenyl and 2,4,2’,4’-tetrachlorobiphenyl, when compared to solvent controls. The uteri were more sensitive to PCB congeners with or&o-substituted light chlorination than those highly chlori-
nated, or those interacting with the Ah-receptor. Key Words: uterus; uterine contraction;
spontaneous
oscillations;
rats; polychlorinated
INTRODUCTION
biphenyls;
smooth muscle.
ily on carcinogenesis, teratogenesis, immunotoxicity, and the induction of hepatic enzymes. However, effects of PCBs on parturition have been observed in humans and experimental animals. Chronic exposure to the commercial PCB mixture Aroclor 1254 is associated with abortion in rhesus monkeys (4) and preterm delivery in women (5,6). Women experiencing miscarriage have higher blood levels of Fenclor 54, a commercial PCB mixture (7). In contrast, exposure of rats to a PCB coplanar congener, 3,4,3’,4’-tetrachlorobiphenyl (3,4TCB), delays parturition (8). A lightly chlorinated PCB congener, 2,2’-dichlorobiphenyl, prolongs gestation length in NMR mice (9). In brief, the reported effects of PCBs on parturition are diverse and inconsistent. The mechanism for the PCB actions on parturition is not clear. Parturition is a complex process that requires functional integration of the different parts of the reproductive tract. A critical component of parturition is timely and proper oscillatory contraction of the uterus. Occurrence of oscillatory contractions during midgestation is associated with preterm labor (10,l l), and uterine dysfunction is a major factor associated with postterm delivery and cesarean section (12). Consequently, PCB effects on parturition could be related to modulation of uterine contractions. Uterine muscle can be a target organ for toxicant actions (13,14). Even though the effect of PCBs on uterine contractions has not been reported, PCB accumulation in uterine muscle is much higher on a lipid weight
Polychlorinated biphenyls (PCBs) were widely used in transformers and capacitors (1). Because of lipophilicity and stability, PCBs tend to accumulate in the environment and in tissues of animals (2). PCB contamination generally occurs as a mixture of isomers or congeners. The mechanism responsible for the PCB actions is very complex, partly because the degree and pattern of chlorination alter the biochemical properties of PCB congeners. For example, with respect to the induction of aryl hydrocarbon hydroxylase (AHH) in liver, the most active congeners 3,4,5,3’,4’-pentachloroand 3,4,3’,4’tetrachlorobiphenyl are substituted at both paru positions, at two or more meta positions, but not at ortho positions. In contrast, the PCB congener 2,4,5,2’,4’,5’hexachlorobiphenyl with no meta-substitution induces a phenobarbital type of hepatic microsomal enzyme but not AHH activity (3). PCB studies have focused primar-
*A portion of this work was presented at the 33rd Annual Meeting of the Society of Toxicology, March 13-17, 1994, Dallas, TX, and published in abstract form (The Toxicologist, 14, 80, 1994). ‘Present address: Department of Physiology, Medical College, National Cheng Kung University, Tainan, Taiwan, ROC. Address correspondence to: Dr. Rita Loch-Caruso, Toxicology Program, Department of Environmental and Industrial Health, University of Michigan, 1420 Washington Heights, Ann Arbor, MI 481092029, Phone: 313-936-1256, Fax: 313-763-8095, E-mail: [email protected] Received 1 May 1995; Revision received 5 September 1995; Accepted 9 September 1995. 21
22
Reproductive Toxicology
basis than in maternal adipose tissue, fetal blood, or placenta from pregnant women (15). Thus, it is reasonable that the pregnant uterus could be a target for PCB actions. The hypothesis in this study was that PCB congeners act on pregnant uteri in a congener-specific manner. Because we had standardized procedures for working with midgestation uteri and because it was not known which stage of pregnancy may be at greatest risk from exposure to the PCB congeners, we chose to use uteri from gestation day 10 rats in this investigation. The objectives of this study were to examine: 1) the influence of PCB congeners on oscillatory contractions of uteri from gestation day 10 pregnant rats, and 2) the structureactivity relationships for PCB congeners. MATERIALS AND METHODS Chemicals The six PCB congeners used in this study, 2,4,5,2’,4’,5’-hexachlorobiphenyl (HCB), 3,4,5,3’,4’pentachlorobiphenyl (3,4-PCB), 3,4,3’,4’-tetrachlorobiphenyl (3,4-TCB), 2,4,2’,4’-tetrachlorobiphenyl (2,4TCB), 2,4,6-trichlorobiphenyl (2,4,6-TCB), and 4-OH2’,4’,6’-trichlorobiphenyl (4-OH-TCB). were purchased from Chemical Service (West Chester, PA) and are listed in Figure 1. These six PCB congeners were dissolved in dimethyl sulfoxide (DMSO). All other chemicals were obtained from the University of Michigan Chemistry Stores.
2,4,5,2’,4’.5’hexachlorobiphenyl
ucl
3,4,3’,4’tetrachlorobiphenyl
3,4-TCB
2,4,2’,4’tetrachlorobiphenyl
2,4-TCB
Fig. 1. The PCBs selected for study.
Cl Cl
Volume 10, Number I. 1996
Animals Female Sprague-Dawley rats were obtained from the colony of the Reproductive Science Program at the University of Michigan between 60 and 90 days of age and weighing 180 to 220 g. The animals were housed at 24 f 1°C under a 14 h light (0500-1900 h) schedule. They were mated with males and the day of detection of spermatozoa in the vaginal smear was designated as day 0 of pregnancy.
Preparation of uterine strips On day 10 of pregnancy, rats were anesthetized with ether followed by exsanguination. A longitudinal uterine strip 1 mm wide by 20 mm long was taken from the anteriomesometrial side of the mid-portion of a horn that had contained live fetuses. The strips were placed in physiologic salt solution (PSS) composed of 116 mM NaCl, 4.6 mM KCl. 1.16 mM NaH,PO, . H,O, 1.16 mM MgSO, 7H,O, 21.9 mM NaHCO,, 1.8 mM CaCl, . 2H,O, 11.6 mM dextrose, and 0.03 mM CaNa,EDTA, pH 7.4. To minimize variability, each strip contained four implantation sites.
Measurement of spontaneous oscillatory contractions The uterine strips were suspended in an organ bath containing PSS for isometric force measurement as described previously (16). The PSS was maintained at 37°C and was aerated with a mixture of 95% O2 and 5% CO,. All preparations were allowed to equilibrate for at least 60 min under a constant passive force of 1.O g. This level of passive force was determined to be optimal for maximum force development of 60 mM potassium chloride (KCl). After a 40-min equilibration period, all strips were challenged with 60 mM KC1 for determining viability and maximum contraction force. Those strips that did not respond to KC1 were discarded. After another 90-min equilibration, spontaneous oscillatory contractions developed in a predictable manner and lasted for at least 6 h. Oscillatory activity was calculated by multiplying the number of contraction/relaxation cycles by the average force per contraction in a lo-min period. A contraction/relaxation cycle was operationally defined as the generation of force of at least two-thirds of the KC1 (60 mM)-induced tonic contraction initiating and returning to baseline. The oscillatory activity following equilibration and prior to any treatment was termed basal oscillatory activity. Basal oscillatory activity was calculated 10 min before any treatment. Treatment-related responses were normalized with respect to basal oscillatory activity as follows: Y&of basal oscillatory activity = oscillatory treatment / basal oscillatory activity x 100%.
activity
after
PCBs and Uterine Contractions
Treatments A segment from one horn of a pregnant uterus was exposed to a PC3 congener, and the counterpart of the other horn was treated with an equivalent amount of DMSO solvent as a negative control. Uterine strips were exposed to increasing concentrations of individual PCB congeners or solvent added to the muscle bath in a cumulative manner. The concentrations were within the range of 0.3 and 150 PM, and were selected on the basis of preliminary experiments to span from non-effective to maximally effective but nontoxic concentrations. The time interval for each concentration was 10 to 15 min. Five to six concentrations for each congener were used to analyze the concentration dependence and to determine the ED,,. After rinsing, all strips were rechallenged with 60 mM KC1 to re-examine the tissue viability. Those strips that did not respond to KC1 were thought to suffer general cytotoxicity and/or tissue death and were excluded from analysis. This occurred in less than 10% of the uterine strips.
Data analysis and statistical evaluation Data are reported as mean -’ standard error of the mean (SEM). Two-way repeated measures analysis of variance (ANOVA) was used to compare observations between PCB-treated and DMSO-treated (solvent control) groups. If significant, Bonferoni’s correction was applied for pairwise comparison of means. In all cases, a P value less than 0.05 was considered statisticaIIy significant.
RESULTS Spontaneous oscillatory contractions of gestation day 10 uteri After equilibration, spontaneous oscillatory contractions of gestation day 10 uteri developed in a predictable manner (shown in Figure 2A) and lasted for at least 6 h. The frequency of the spontaneous oscillatory contractions was 10.5 -+ 0.5 contraction/relaxation cycles per 10 min. The average amplitude was 1.44 + 0.13 g force (summarized in Figure 2B). The magnitude of the contractions was approximately 90% of maximal force development to 60 mM KCl. These spontaneous oscillations were not altered by 10 FM indomethacin (data not shown), consistent with previous findings (17).
The effects of PCB congeners (150 pM) on spontaneous oscillatory contractions In this experiment, the highest concentration (150 FM) for each congener was used to induce the maximum effect. In the presence of the PCB congeners HCB, 3,4-
l
M.-L.
23
TSAI ET AL.
AverageAmphde Frequency (No. of cyclel/lO min) (%fwc.z) 10 5+0.5 I 14420.13
Bwl Oscilhtory Activity (e/IO mill) 1 IS.282 1.8
Fig. 2. A) Representative polygraph tracing of spontaneous oscillatory contractions of longitudinal uterine strips (1 mm wide x 20 mm long) excised from midgestation rats. B) Characteristics of spontaneous oscillatory contractions tion day 10 uteri (mean f SEM, n = 12).
from gesta-
PCB, and 3,4-TCB, spontaneous uterine oscillations were 90 -+ II%, 107 + 8%, and 100 f 10% of basal oscillatory activity, respectively (Figure 3). Relative to DMSO solvent controls, these three congeners had no significant effects on uterine contraction. To exclude the possibility that lower concentrations of these three congeners might exert effects not observable at high concentration, we also examined the effects of these congeners with lower concentrations (1 to 100 FM). As we expected, the lower concentrations of these three congeners had no effect on uterine contractions (data not shown). The orfho-, para-substituted PCB congeners 2,4TCB, 2,4,6-TCB, and 4-OH-TCB did modify uterine contractions. The nonhydroxylated congeners 2,4-TCB and 2,4,6-TCB significantly increased oscillatory activity to 138 k 14% and 183 + lo%, respectively (Figure 3).
i
I
*
Fig. 3. The effects of PCB congeners (150 FM) on spontaneous oscillatory contractions of pregnant uteri. Each point represents the mean rt SEM of at Ieast four individuai uteri. Asterisks indicate significant differences (P < 0.05) compared with DMSO solvent controls.
24
Reproductive Toxicology
Volume 10, Number l* 1996
4-OH-TCB markedly decreased oscillatory activity to almost zero as shown in Figure 3. The data suggested that hydroxylation and the position of chlorination were involved in the changes of uterine contraction. During a 60-min exposure to each congener, no separation of any congener from the PSS was visually observed in the muscle baths. Sixty minutes after rinsing, all uterine strips responded to KC1 (60 mM), indicating that no strip suffered from general cytotocixity or tissue death. In contrast,
A+
IOmin
4
2,CTCB (150 @f)
C.
B. The concentration-dependent effect of 3,4-TCB on spontaneous oscillatory contractions Exposure of uterine strips to increasing concentrations of 3,4-TCB (1.5 to 150 FM) did not significantly change spontaneous oscillatory activity, nor was a significant difference detected between 3,4-TCB-treated and solvent control groups (Figure 4A and 4C), even though the maximum concentration (1.50 PM) of 3,4-
7 i
A.
% s
50
-c)- DMSO --3%.2,4_TCB
Concentration (PM)
3,4-TCB (150 @I)
B.
C.
Fig. 5. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri after exposure to DMSO (solvent control: top tracing) or 150 FM 2,4-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to treatment with DMSO (solvent control) or 2,4-TCB. C) The concentration-dependent effect of 2,4-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means k SEM of four individual uteri. If no error bar is observed, the standard error from those data is smaller than the size of the symbol. Asterisks indicate significant differences (P < 0.05) compared with DMSO (solvent) controls.
TCB slightly
-I)-
s
1
DMSO
---U-.- 3,4-TCB I I 10 1000 100 Concentration (PM)
Fig. 4. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri in response to DMSO (solvent control; top tracing) and 150 FM 3,4-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to treatment with DMSO (solvent control) or 3,4-TCB. C) The effect of increasing concentrations (I 5 to 150 p.M) of 3,4-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means + SEM of six individual uteri. If no error is observed, the standard error from those data is smaller than the size of the symbol.
increased
oscillatory
activity.
The basal os-
and solvent control groups were compared. Figure 4B indicated no significant difference between the means of these two groups, suggesting that the basal oscillatory activity did not affect the response to 3,4-TCB. cillatory
activities
of the congener-treated
The concentration-dependent effect of 2,4-TCB on spontaneous oscillatory contractions Pregnant uteri responded to 2,4-TCB (0.45 to 150 FM) with increased oscillatory activity in a concentration-dependent manner (Figure 5C). The maximum response was reached at 15 PM (131 + 10%) and continued to be observed at 45 FM (13 1 f 10%) and at 150 FM of 2,4-TCB (138 ? 14%). Pregnant uteri did not respond to lower concentrations (0.45 to 1.5 FM) of 2,4-TCB.
PCBs
25
and Uterine Contractionsl M.-L. TSAI ET AL.
The ED,, value was about 3 PM. A representative tracing shown in Figure 5A illustrates a typical response to 150 pM 2,4-TCB. The basal oscillatory activities of the PCB-treated and solvent control group were compared. As Figure 5B indicates, no significant difference was observed between these two groups, allowing us to exclude the possibility that different basal oscillatory activities might have contributed to the uterine response. The concentration-dependent effect of 2,4,6-TCB on spontaneous oscillatory contractions The exposure of uterine strips to increasing concentrations (0.3 to 100 pM) of 2,4,6-TCB increased oscillations in a concentration-dependent manner (Figure 6C). The ED,, was not less than 30 PM. Relative to DMSO solvent controls, 30 ~.LMand 100 pM 2,4,6-TCB significantly increased uterine contractions to 147 f 16 and 184 + lo%, respectively, of basal oscillatory activity, but
_ a d: cw o\
A
C.
2.WTCB (loo pi) * 200-
.9 >
F
1 m
pm g
-? U Nl-
‘;;j SO3 m “0 o-i s 0.1
* p
-Cl-
DMSO
.--a--. 2,4,6-TCB I 1 IO
I 100
Concentration (PM)
Fig. 6. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri after exposure to DMSO (solvent control; top tracing) or 100 p,M 2,4,6-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to exposure to DMSO (solvent control) or 2,4,6-TCB. C) The concentrationdependent effect of 2,4,6-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means f SEM of five individual uteri. If no error bar is observed, the standard error from those data is smaller than the size of the symbol. Asterisks indicate significant differences (P < 0.05) compared with DMSO (solvent control).
4o
...X.** 9: ‘?,
+DMSO
--*Q-*0 4_OH_TCB o0.1
I 1
‘i; *
1 10 Concentration (PM)
I 100
Fig. 7. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri after exposure to DMSO (solvent control; top tracing) or 30 pM 4-OH-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to treatment with DMSO (solvent control) or 4-OH-TCB. C) The concentration-dependent effect of 4-OH-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means f SEM of five individual uteri. If no error bar is observed, the standard error from those data is smaller than the size of the symbol. Asterisks indicate significant differences (P < 0.05) compared with DMSO (solvent control).
lower concentrations (0.3 to 10 pM) of 2,4,6-TCB did not have any significant effects. Representative tracings show the responses of uterine strips to 100 pM 2,4,6TCB or DMSO solvent control (Figure 6A). Figure 6B shows that the basal oscillatory activity was not significantly different in 2,4,6-TCB-treated and solvent control groups, suggesting that basal oscillatory activity did not affect the response to 2,4,6-TCB. The concentration-dependent effect of 4-OH-TCB on spontaneous oscillatory contractions of pregnant uteri Uterine strips exhibited decreased oscillatory activity with increasing concentrations of 4-OH-TCB (Figure 7C). Relative to DMSO solvent controls, 3, 10, and 30 (LM 4-OH-TCB significantly decreased uterine contractions to 73 + 11, 52 f 10, and 8.5 f 5.6% of basal oscillatory activity, respectively. The ED,, value was less than 10 PM. Representative tracings show the in-
Reproductive
26
Toxicology
hibitory effect of 30 p,M 4-OH-TCB compared to solvent controls (Figure 7A). Because of no significant difference between 4-OH-TCB-treated and DMSO-treated solvent control groups, the inhibition of spontaneous activity by 4-OH-TCB did not result from a difference of basal oscillatory activity.
DISCUSSION In vivo or in vitro, uterine oscillatory contractions can arise spontaneously as a myogenic response (18) or can be induced by physiologic chemicals such as oxytotin (19). In the present work, spontaneously contracting uterine strips isolated from midgestation rats were used to examine PCB congener effects on uterine oscillatory activity. Consistent with a previous report (17) these spontaneous oscillatory contractions could not be blocked by indomethacin, a cyclooxygenase inhibitor. The data demonstrated that PCB effects on spontaneous oscillatory uterine activity depended on the structure of the PCB congener. With acute exposure, uterine strips isolated from midgestation pregnant rats exhibited increased activity in the presence of 2,4-TCB and 2,4,6TCB, whereas oscillatory contractions were inhibited by 4-OH-TCB. The coplanar PCB congeners 3,4-TCB and 3,4-PCB, as well as the non-coplaner congener HCB. had no significant acute effects on spontaneous uterine contractions. Certain patterns emerge from an examination of the data with respect to the PCB congener structure. First, the lightly chlorinated congeners with ortho-, parusubstitution (4-OH-TCB, 2,4,6-TCB, and 2,4-TCB) modified uterine contractions. Second, the replacement of ortho-substitution (2,4-TCB) with meta-substitution (3,4-TCB) changed the uterine response from observable to nonobservable. Together, these observations suggest that ortho-chlorine substitution may be necessary for acute PCB actions on oscillatory contractions of pregnant uteri. Third, the addition of a chlorine residue to the metu-positions of 2,4-TCB to form HCB shifted the response from observable to nonobservable. This finding suggests that meta-substitution may interfere with the activity of ortho-substituted congeners on uterine contractions. Fourth, 3,4-TCB and 3,4-PCB did not alter uterine contractions, yet they are potent inducers of aryl hydrocarbon hydroxylase, function as analogs of 2,3,7,8tetrachlorodibenzodioxin (TCDD) and activate the Ah receptor (3). This observation suggests that dioxin-like compounds are not involved in acute alterations of uterine contractions. Fifth, the maximum efficacy for 2,4TCB is less than that of 2,4,6-TCB, suggesting that its bulky size may interfere with PCB activity in the uterus. In summary, because of the structural specificity of the
Volume IO. Number
I, lYY6
PCB congener actions, it is suggested that the PCB congeners interact with specific molecules in the uterus to alter uterine contractions. Even though the analogs 2,4,6-TCB and 4-OH-TCB both altered uterine contractions, the response to 2,4,6TCB was opposite to that of 4-OH-TCB: 2,4,6-TCB stimulated uterine contractions but 4-OH-TCB inhibited uterine contractions. Because of their structural similarity. 2,4,6-TCB and 4-OH-TCB may act on the same pathway, one as an antagonist and the other as an agonist. However, it is also possible that 2,4,6-TCB and 4-OHTCB act on separate pathways. PCBs can be metabolized by many different pathways. The most important pathway is through hydroxylation and subsequent conjugation (3). Hydroxylation of dichloro-, trichloro-, tetrachloro-, pentachloro-, and hexachlorobiphenyl is favored at the pm-u position in the least chlorinated phenyl ring (20). Higher chlorinated PCBs tend to be metabolized more slowly and accumulate to a greater extent (21,22). In our design, 4-OH-TCB was used as a hydroxylated analog of 2,4,6-TCB. If hydroxylation were required for activity and the uterine tissue had significant metabolic capacity to hydroxylate 2,4,6-TCB, then the response to 4-OH-TCB and 2,4,6TCB would be similar, but the potency of 4-OH-TCB would be greater than that of 2,4,6-TCB. In fact, our observation was that 2,4,6-TCB stimulated oscillations and 4-OH-TCB suppressed oscillations, inconsistent with that model. Considering these results, together with the very short period of exposure, it is unlikely that metabolites produced by the uterus contributed to the alteration of uterine contractions in our system. Korach et al. (23) demonstrated that 4-OH-TCB is estrogenic in that it competitively inhibits estrogen binding to estrogen receptors. 4-OH-TCB also increases the weight and cell proliferation of immature rat uteri, further supporting its estrogenicity (24). The effect of 4-OH-TCB on uterine contractions in our study is consistent with previous reports that estrogenic compounds rapidly inhibit uterine contractions (14,25). In a previous report, the rapid inhibitory effect of estrogens was proposed to be nongenomic and related to inhibition of extracellular calcium entry (25). In our laboratory, we have also observed that 17@estradiol rapidly inhibits spontaneous oscillatory uterine contractions in a manner similar to that observed with 4-OH-TCB (26). However, in our experiments, the calcium ionophore A23187 (up to 2 FM) was unable to reverse the inhibitory action of either 17&estradiol(26) or 4-OH-TCB (27) on uterine contractions, Rather, our results suggest that inhibition of gap junctional communication may be involved in the inhibition of spontaneous oscillations by 17P-estradiol (26) and 4-OH-TCB (27). Garfield and his colleagues have shown that cell-to-cell coupling via gap junctions is as-
PCBs and Uterine Contractions
sociated with the development of synchronized oscillatory contractions in pregnant uteri at term (28). The mechanism by which the PCB congeners 2,4,6TCB and 2,4-TCB increase spontaneous oscillatory activity is not clear. However, it has been shown that 2,4-TCB increases the production of superoxide in neutrophils, which then causes the activation of the phosphoinositol signal transduction pathway (29). A lightly chlorinated PCB congener, 2,2’-dichlorobiphenyl, increases intracellular calcium by inhibition of mitochondrial calcium uptake in cerebellar granule cells (30). It is known that increased phosphoinositide turnover and calcium accumulation are associated with increased uterine contractions (31). Thus, a possible mechanism for 2,4TCB and 2,4,6-TCB actions may include elevated production of superoxide and increased intracellular calcium, which in turn increase uterine contractions. Interestingly, PCB congeners with light chlorination at the or&-position are the most active neurotoxicants (32) a pattern similar to our observations with uterine contractions. In contrast, the highly chlorinated PCB congeners such as 3,4,5,3’,4’-pentachloroand 3,4,3’,4’tetrachlorobiphenyl tend to be more potent inducers of aryl hydrocarbon hydroxylase (AHH) in liver. Thus, we speculate that the mechanism responsible for the lightly chlorinated PCB action may be different from that for the highly chlorinated PCBs. In summary, our data show that PCB congeners modify in vitro contractions of pregnant uteri in an acute, congener-specific manner. Because of the relationship between PCB structure and activity, it is suggested that the uterine response to PCBs requires congeners with or&o-substituted light chlorination that can interact with specific molecules, but not with the Ah-receptor in the uterus. The uterine responses were divergent and even opposite in nature depending on the structure of the PCB congener. PCB contamination of the environment usually occurs as a mixture, and synergistic, additive or antagonistic interactions may occur among the different congeners. However, mixtures of congeners were not evaluated in this study, so no definate conclusions can be drawn regarding such interactions and uterine contractility. Finally, these experiments examined only the midgestation uterus. Consequently, it is not known if the responses observed are specific for this stage of pregnancy or if they would also be observed at other stages of pregnancy or in the nonpregnant uterus. Acknowledgments - The authors thank Shelly Coe and Howard Listopad for their assistance with these experiments and Dr. Craig Harris for providing uterine tissue. The project described was conducted as partial fulfillment of M.-L. Tsai’s doctoral dissertation and was supported by grant number HL18575 to R. C. Webb and P42 ES0491 1 to R. Loch-Caruso from the National Heart, Lung, and Blood Institute and the National Institute of Environmental Sciences, respectively, of the NIH. Additional support was provided by the Laboratory Animal Cores
l
M.-L. TSAI ET AL.
21
of the Center for the Study of Reproduction, which is supported by grant number P30 HD18258 from the National Institute of Child Health and Human Development, NIH.
REFERENCES 1. Hutzinger 0, Safe S, Zitko V, eds. The chemistry of PCBs. Cleveland: CRC Press; 1974. 2. Erickson MD, ed. Analytical chemistry of PCBs. Boston: Butterworth; 1986. 3. Safe S. Polychlorinated biphenyls (PCBs) and polybrominated biphenyls (PBBs): biochemistry, toxicology, and mechanism of action. CRC Crit Rev Toxicol. 1984;13:319-95. 4. Altman NH, New AE, McConnell EE, Ferrell TL. A spontaneous outbreak of polychlorinated biphenyl (PCB) toxicity in rhesus monkeys (Macaca m&m): clinical observations. Lab Anim Sci. 1979;29:661-5. 5. Wassermann M, Ron M, Bercovici B, Wassermann D, Cucos S, Pines A. Premature delivery and organochlorine compounds: polychlorinated biphenyls and some organochlorine insecticides. Environ Res. 1982;28:106-12. 6. Taylor PR, Lawrence C, Hwang HL, Paulson AS. Polychlorinated biphenyls: influence on birthweight and gestation, Am J Public Health. 1984;74: 1153-4. I. Leoni V, Fabiani L, Marinelli G, et al. PCB and other organochlotine compounds in blood of women with or without miscarriage: a hypothesis of correlation. Ecotoxicol Environ Safety. 1989;17: l11. 8. White RD, Allen SD, Bradshaw WS. Delay in the onset of parturition in the rat following prenatal administration of developmental toxicants. Toxic01 Lett. 1983;18: 185-92. 9. T&ok P. Delayed pregnancy in NMRI mice treated with PCB: 2,2’-dichlorobiphenyl. Bull Environ Contam Toxicol. 1976; 16: 33-6. 10. Bell R. The prediction of preterm labor by recording spontaneous antenatal uterine activity. Br J Obstet Gynecol. 1983;90:88&7. 11. Morgan MA, Silavin SL, Wentworth RA. et al. Different patterns of myometrial activity and 24-h rhythms in myometrial contractility in the gravid baboon during the second half of pregnancy. Biol Reprod. 1992;46:1158864. 12. Boyd ME, Usher RH, McLean FH, Kramer MS. Obstetric consequences of postmaturity. Am J Obstet Gynecol. 1988;158:334-8. 13. Modzelewski P, Szamatowicz J, Laudanski T, MoniuszukoJakoniuk J, Akerlund M. The influence of lead ions on uterine activity in the rat. Int J Gynaecol Obstet. 1990;32:169-73. 14. Juberg DR. Loch-Caruso R. Investigation of the role of estrogenic action and prostaglandin E, in DDT-stimulated rat uterine contractions ex vivo. Toxicology. 1992;74:161-72. 15. Polishuk ZW, Wassermann Organochlorine compounds viron Res. 1977;13:278-84.
D, Wassermann M, Cucos S, Ron M. in mother and fetus during labor. En-
16. Juberg DR, Webb RC, Loch-Caruso R. Characterization of o,p’DDT-stimulated contraction frequency in rat uterus in vitro. Fundam Appl Toxicol. 1991;17:543-9. 17. Poli E, Merialdi A, Coruzzi G. Characterization of the spontaneous motor activity of the isolated human pregnant myometrium. Pharmacol Res. 1990;22: 115-24. 18. Ruegg JC. Smooth muscle tone. Physiol Rev. 197 I;5 I :20148. 19. Honnebier MB, Myers T, Figueroa JP, Nathanielsz PW. Variation in myometrial response to pulsatile intravenous oxytocin administration at different times of the day in the pregnant rhesus monkey. Endocrinology. 1989; 125: 1498-503. 20. Borlakoglu
JT, Wilkins
JP. Metabolism
of di-, tri-, tetra-, and
28
21.
22.
23.
24.
25.
26.
Reproductive
Toxicology
hexachlorobiphenyls by hepatic microsomes isolated from control animals and animals treated with Aroclor 1254, a commercial mixture of polychlorinated biphenyls (PCBs). Comp Biochem Physiol C. 1993;105:107-12. Lutz RJ, Dedrick RL, Matthews HB, Eling TE, Anderson MW. A preliminary pharmacokinetic model for several chlorinated biphenyls in the rat. Drug Metab Dispos. 1977;5:386-96. Kaley RG, Hicks 0, Mees WM, et al. Tissue residues from subacute oral feeding of polychlorinated biphenyl dielectric fluids. Bull Environ Contam Toxicol. 1976;15:699-707. Korach KS, Sarver P, Chae K, McLachlan JA, McKinney JD. Estrogen receptor-binding activity of polychlorinated hydroxybiphenyls: conformationally restricted structural probes. Mol Pharmacol. 1988;33:120-6. Jansen HT, Cooke PS, Porcelli J, Liu T-C, Hansen LG. Estrogenic and antiestrogenic actions of PCBs in the female rat: in vitro and in vivo studies. Reprod Toxicol. 1993;7:237-48. Femandez AI, Martinez V, Cantabrana B, Hidalgo A. Differential effect of calcium and Bay K 8644 on the inhibitory action of estrogens in the rat uterus. Gen Pharmacol. 1992;23:549-54. Tsai M-L, Webb RC, Loch-Caruso R. Inhibition of spontaneous oscillatory uterine contractions by 17B-estradiol is associated with
Volume 10, Number
1, 1996
disruption of gap junctional 1995;52(Suppl 1):115.
communication.
Biol
Reprod.
27. Tsai M-L, Webb RC, Loch-Caruso R. Acute inhibition of uterine contraction by 4-hydroxy-2’,4’,6’-trichlorobiphenyl (40HTCB) may result from disruption of gap junctional cologist. 1995;15:275.
communication.
Toxi-
28. Garfield RE, Sim S, Daniel EE. Gap junctions: their presence and necessity in myometrium during parturition. Science. 1977; 198: 958860. 29. Ganey PE, Sirois JE, Denison M, Robinson JP, Roth RA. Neutrophi1 function after exposure to polychlotinated biphenyls in vitro. Environ Health Perspect. 1993;101:430-4. 30. Kodavanti PRS, Shin DS, Tilson HA, Harry GJ. Comparative effects of two polychlorinated biphenyl congeners on calcium homeostasis in rat cerebellar granule cells. Toxic01 Appl Pharmacol. 1993;123:97-106. 31. Garfield RE, ed. Uterine contractility. posia: 1990.
Norwell, MA: Serono Sym-
32. Shain W, Bush B, Seegal R. Neurotoxicity of polychlorinated biphenyls: structure-activity relationship of individual congeners. Toxic01 Appl Pharmacol. 1991;111:3342.
0890-6238(95)02014-4
CONGENER-SPECIFIC EFFECTS OF PCBS ON CONTRACTIONS OF PREGNANT RAT UTERI MEI-LING TsAI,*‘~ R. CLINTONWEBB,? and RITA LOCH-CARUSO* *Departments of Environmental and Industrial Health and tPhysiology, University of Michigan, Ann Arbor, Michigan Abstract - Both increased and decreased gestation lengths have been reported following exposures to polychlorinated biphenyl (PCB) mixtures and congeners. Because oscillatory uterine contractions are essential for parturition, we hypothesized that the disparate findings on gestation length may be the result of distinct PCB congener-specific actions on oscillatory uterine contractions. This study examined the acute effects of PCB congeners on isometric contractions of isolated pregnant uteri and the structure-activity relationship for individual congeners. After cumulative exposure to individual PCB congeners (0.5 pM to 150 pM), oscillatory contractions were: 1) not altered by 2,4,5,2’,4’,5’-hexachlorobiphenyl, 3,4,5,3’,4’-pentachlorobiphenyl, or 3,4,3’,4’-tetrachlorobiphenyl; 2) significantly inhibited by 4-hydroxy-2’,4’,6’-trichlorobiphenyl; and 3) markedly increased by 2,4,6-trichlorobiphenyl and 2,4,2’,4’-tetrachlorobiphenyl, when compared to solvent controls. The uteri were more sensitive to PCB congeners with or&o-substituted light chlorination than those highly chlori-
nated, or those interacting with the Ah-receptor. Key Words: uterus; uterine contraction;
spontaneous
oscillations;
rats; polychlorinated
INTRODUCTION
biphenyls;
smooth muscle.
ily on carcinogenesis, teratogenesis, immunotoxicity, and the induction of hepatic enzymes. However, effects of PCBs on parturition have been observed in humans and experimental animals. Chronic exposure to the commercial PCB mixture Aroclor 1254 is associated with abortion in rhesus monkeys (4) and preterm delivery in women (5,6). Women experiencing miscarriage have higher blood levels of Fenclor 54, a commercial PCB mixture (7). In contrast, exposure of rats to a PCB coplanar congener, 3,4,3’,4’-tetrachlorobiphenyl (3,4TCB), delays parturition (8). A lightly chlorinated PCB congener, 2,2’-dichlorobiphenyl, prolongs gestation length in NMR mice (9). In brief, the reported effects of PCBs on parturition are diverse and inconsistent. The mechanism for the PCB actions on parturition is not clear. Parturition is a complex process that requires functional integration of the different parts of the reproductive tract. A critical component of parturition is timely and proper oscillatory contraction of the uterus. Occurrence of oscillatory contractions during midgestation is associated with preterm labor (10,l l), and uterine dysfunction is a major factor associated with postterm delivery and cesarean section (12). Consequently, PCB effects on parturition could be related to modulation of uterine contractions. Uterine muscle can be a target organ for toxicant actions (13,14). Even though the effect of PCBs on uterine contractions has not been reported, PCB accumulation in uterine muscle is much higher on a lipid weight
Polychlorinated biphenyls (PCBs) were widely used in transformers and capacitors (1). Because of lipophilicity and stability, PCBs tend to accumulate in the environment and in tissues of animals (2). PCB contamination generally occurs as a mixture of isomers or congeners. The mechanism responsible for the PCB actions is very complex, partly because the degree and pattern of chlorination alter the biochemical properties of PCB congeners. For example, with respect to the induction of aryl hydrocarbon hydroxylase (AHH) in liver, the most active congeners 3,4,5,3’,4’-pentachloroand 3,4,3’,4’tetrachlorobiphenyl are substituted at both paru positions, at two or more meta positions, but not at ortho positions. In contrast, the PCB congener 2,4,5,2’,4’,5’hexachlorobiphenyl with no meta-substitution induces a phenobarbital type of hepatic microsomal enzyme but not AHH activity (3). PCB studies have focused primar-
*A portion of this work was presented at the 33rd Annual Meeting of the Society of Toxicology, March 13-17, 1994, Dallas, TX, and published in abstract form (The Toxicologist, 14, 80, 1994). ‘Present address: Department of Physiology, Medical College, National Cheng Kung University, Tainan, Taiwan, ROC. Address correspondence to: Dr. Rita Loch-Caruso, Toxicology Program, Department of Environmental and Industrial Health, University of Michigan, 1420 Washington Heights, Ann Arbor, MI 481092029, Phone: 313-936-1256, Fax: 313-763-8095, E-mail: [email protected] Received 1 May 1995; Revision received 5 September 1995; Accepted 9 September 1995. 21
22
Reproductive Toxicology
basis than in maternal adipose tissue, fetal blood, or placenta from pregnant women (15). Thus, it is reasonable that the pregnant uterus could be a target for PCB actions. The hypothesis in this study was that PCB congeners act on pregnant uteri in a congener-specific manner. Because we had standardized procedures for working with midgestation uteri and because it was not known which stage of pregnancy may be at greatest risk from exposure to the PCB congeners, we chose to use uteri from gestation day 10 rats in this investigation. The objectives of this study were to examine: 1) the influence of PCB congeners on oscillatory contractions of uteri from gestation day 10 pregnant rats, and 2) the structureactivity relationships for PCB congeners. MATERIALS AND METHODS Chemicals The six PCB congeners used in this study, 2,4,5,2’,4’,5’-hexachlorobiphenyl (HCB), 3,4,5,3’,4’pentachlorobiphenyl (3,4-PCB), 3,4,3’,4’-tetrachlorobiphenyl (3,4-TCB), 2,4,2’,4’-tetrachlorobiphenyl (2,4TCB), 2,4,6-trichlorobiphenyl (2,4,6-TCB), and 4-OH2’,4’,6’-trichlorobiphenyl (4-OH-TCB). were purchased from Chemical Service (West Chester, PA) and are listed in Figure 1. These six PCB congeners were dissolved in dimethyl sulfoxide (DMSO). All other chemicals were obtained from the University of Michigan Chemistry Stores.
2,4,5,2’,4’.5’hexachlorobiphenyl
ucl
3,4,3’,4’tetrachlorobiphenyl
3,4-TCB
2,4,2’,4’tetrachlorobiphenyl
2,4-TCB
Fig. 1. The PCBs selected for study.
Cl Cl
Volume 10, Number I. 1996
Animals Female Sprague-Dawley rats were obtained from the colony of the Reproductive Science Program at the University of Michigan between 60 and 90 days of age and weighing 180 to 220 g. The animals were housed at 24 f 1°C under a 14 h light (0500-1900 h) schedule. They were mated with males and the day of detection of spermatozoa in the vaginal smear was designated as day 0 of pregnancy.
Preparation of uterine strips On day 10 of pregnancy, rats were anesthetized with ether followed by exsanguination. A longitudinal uterine strip 1 mm wide by 20 mm long was taken from the anteriomesometrial side of the mid-portion of a horn that had contained live fetuses. The strips were placed in physiologic salt solution (PSS) composed of 116 mM NaCl, 4.6 mM KCl. 1.16 mM NaH,PO, . H,O, 1.16 mM MgSO, 7H,O, 21.9 mM NaHCO,, 1.8 mM CaCl, . 2H,O, 11.6 mM dextrose, and 0.03 mM CaNa,EDTA, pH 7.4. To minimize variability, each strip contained four implantation sites.
Measurement of spontaneous oscillatory contractions The uterine strips were suspended in an organ bath containing PSS for isometric force measurement as described previously (16). The PSS was maintained at 37°C and was aerated with a mixture of 95% O2 and 5% CO,. All preparations were allowed to equilibrate for at least 60 min under a constant passive force of 1.O g. This level of passive force was determined to be optimal for maximum force development of 60 mM potassium chloride (KCl). After a 40-min equilibration period, all strips were challenged with 60 mM KC1 for determining viability and maximum contraction force. Those strips that did not respond to KC1 were discarded. After another 90-min equilibration, spontaneous oscillatory contractions developed in a predictable manner and lasted for at least 6 h. Oscillatory activity was calculated by multiplying the number of contraction/relaxation cycles by the average force per contraction in a lo-min period. A contraction/relaxation cycle was operationally defined as the generation of force of at least two-thirds of the KC1 (60 mM)-induced tonic contraction initiating and returning to baseline. The oscillatory activity following equilibration and prior to any treatment was termed basal oscillatory activity. Basal oscillatory activity was calculated 10 min before any treatment. Treatment-related responses were normalized with respect to basal oscillatory activity as follows: Y&of basal oscillatory activity = oscillatory treatment / basal oscillatory activity x 100%.
activity
after
PCBs and Uterine Contractions
Treatments A segment from one horn of a pregnant uterus was exposed to a PC3 congener, and the counterpart of the other horn was treated with an equivalent amount of DMSO solvent as a negative control. Uterine strips were exposed to increasing concentrations of individual PCB congeners or solvent added to the muscle bath in a cumulative manner. The concentrations were within the range of 0.3 and 150 PM, and were selected on the basis of preliminary experiments to span from non-effective to maximally effective but nontoxic concentrations. The time interval for each concentration was 10 to 15 min. Five to six concentrations for each congener were used to analyze the concentration dependence and to determine the ED,,. After rinsing, all strips were rechallenged with 60 mM KC1 to re-examine the tissue viability. Those strips that did not respond to KC1 were thought to suffer general cytotoxicity and/or tissue death and were excluded from analysis. This occurred in less than 10% of the uterine strips.
Data analysis and statistical evaluation Data are reported as mean -’ standard error of the mean (SEM). Two-way repeated measures analysis of variance (ANOVA) was used to compare observations between PCB-treated and DMSO-treated (solvent control) groups. If significant, Bonferoni’s correction was applied for pairwise comparison of means. In all cases, a P value less than 0.05 was considered statisticaIIy significant.
RESULTS Spontaneous oscillatory contractions of gestation day 10 uteri After equilibration, spontaneous oscillatory contractions of gestation day 10 uteri developed in a predictable manner (shown in Figure 2A) and lasted for at least 6 h. The frequency of the spontaneous oscillatory contractions was 10.5 -+ 0.5 contraction/relaxation cycles per 10 min. The average amplitude was 1.44 + 0.13 g force (summarized in Figure 2B). The magnitude of the contractions was approximately 90% of maximal force development to 60 mM KCl. These spontaneous oscillations were not altered by 10 FM indomethacin (data not shown), consistent with previous findings (17).
The effects of PCB congeners (150 pM) on spontaneous oscillatory contractions In this experiment, the highest concentration (150 FM) for each congener was used to induce the maximum effect. In the presence of the PCB congeners HCB, 3,4-
l
M.-L.
23
TSAI ET AL.
AverageAmphde Frequency (No. of cyclel/lO min) (%fwc.z) 10 5+0.5 I 14420.13
Bwl Oscilhtory Activity (e/IO mill) 1 IS.282 1.8
Fig. 2. A) Representative polygraph tracing of spontaneous oscillatory contractions of longitudinal uterine strips (1 mm wide x 20 mm long) excised from midgestation rats. B) Characteristics of spontaneous oscillatory contractions tion day 10 uteri (mean f SEM, n = 12).
from gesta-
PCB, and 3,4-TCB, spontaneous uterine oscillations were 90 -+ II%, 107 + 8%, and 100 f 10% of basal oscillatory activity, respectively (Figure 3). Relative to DMSO solvent controls, these three congeners had no significant effects on uterine contraction. To exclude the possibility that lower concentrations of these three congeners might exert effects not observable at high concentration, we also examined the effects of these congeners with lower concentrations (1 to 100 FM). As we expected, the lower concentrations of these three congeners had no effect on uterine contractions (data not shown). The orfho-, para-substituted PCB congeners 2,4TCB, 2,4,6-TCB, and 4-OH-TCB did modify uterine contractions. The nonhydroxylated congeners 2,4-TCB and 2,4,6-TCB significantly increased oscillatory activity to 138 k 14% and 183 + lo%, respectively (Figure 3).
i
I
*
Fig. 3. The effects of PCB congeners (150 FM) on spontaneous oscillatory contractions of pregnant uteri. Each point represents the mean rt SEM of at Ieast four individuai uteri. Asterisks indicate significant differences (P < 0.05) compared with DMSO solvent controls.
24
Reproductive Toxicology
Volume 10, Number l* 1996
4-OH-TCB markedly decreased oscillatory activity to almost zero as shown in Figure 3. The data suggested that hydroxylation and the position of chlorination were involved in the changes of uterine contraction. During a 60-min exposure to each congener, no separation of any congener from the PSS was visually observed in the muscle baths. Sixty minutes after rinsing, all uterine strips responded to KC1 (60 mM), indicating that no strip suffered from general cytotocixity or tissue death. In contrast,
A+
IOmin
4
2,CTCB (150 @f)
C.
B. The concentration-dependent effect of 3,4-TCB on spontaneous oscillatory contractions Exposure of uterine strips to increasing concentrations of 3,4-TCB (1.5 to 150 FM) did not significantly change spontaneous oscillatory activity, nor was a significant difference detected between 3,4-TCB-treated and solvent control groups (Figure 4A and 4C), even though the maximum concentration (1.50 PM) of 3,4-
7 i
A.
% s
50
-c)- DMSO --3%.2,4_TCB
Concentration (PM)
3,4-TCB (150 @I)
B.
C.
Fig. 5. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri after exposure to DMSO (solvent control: top tracing) or 150 FM 2,4-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to treatment with DMSO (solvent control) or 2,4-TCB. C) The concentration-dependent effect of 2,4-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means k SEM of four individual uteri. If no error bar is observed, the standard error from those data is smaller than the size of the symbol. Asterisks indicate significant differences (P < 0.05) compared with DMSO (solvent) controls.
TCB slightly
-I)-
s
1
DMSO
---U-.- 3,4-TCB I I 10 1000 100 Concentration (PM)
Fig. 4. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri in response to DMSO (solvent control; top tracing) and 150 FM 3,4-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to treatment with DMSO (solvent control) or 3,4-TCB. C) The effect of increasing concentrations (I 5 to 150 p.M) of 3,4-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means + SEM of six individual uteri. If no error is observed, the standard error from those data is smaller than the size of the symbol.
increased
oscillatory
activity.
The basal os-
and solvent control groups were compared. Figure 4B indicated no significant difference between the means of these two groups, suggesting that the basal oscillatory activity did not affect the response to 3,4-TCB. cillatory
activities
of the congener-treated
The concentration-dependent effect of 2,4-TCB on spontaneous oscillatory contractions Pregnant uteri responded to 2,4-TCB (0.45 to 150 FM) with increased oscillatory activity in a concentration-dependent manner (Figure 5C). The maximum response was reached at 15 PM (131 + 10%) and continued to be observed at 45 FM (13 1 f 10%) and at 150 FM of 2,4-TCB (138 ? 14%). Pregnant uteri did not respond to lower concentrations (0.45 to 1.5 FM) of 2,4-TCB.
PCBs
25
and Uterine Contractionsl M.-L. TSAI ET AL.
The ED,, value was about 3 PM. A representative tracing shown in Figure 5A illustrates a typical response to 150 pM 2,4-TCB. The basal oscillatory activities of the PCB-treated and solvent control group were compared. As Figure 5B indicates, no significant difference was observed between these two groups, allowing us to exclude the possibility that different basal oscillatory activities might have contributed to the uterine response. The concentration-dependent effect of 2,4,6-TCB on spontaneous oscillatory contractions The exposure of uterine strips to increasing concentrations (0.3 to 100 pM) of 2,4,6-TCB increased oscillations in a concentration-dependent manner (Figure 6C). The ED,, was not less than 30 PM. Relative to DMSO solvent controls, 30 ~.LMand 100 pM 2,4,6-TCB significantly increased uterine contractions to 147 f 16 and 184 + lo%, respectively, of basal oscillatory activity, but
_ a d: cw o\
A
C.
2.WTCB (loo pi) * 200-
.9 >
F
1 m
pm g
-? U Nl-
‘;;j SO3 m “0 o-i s 0.1
* p
-Cl-
DMSO
.--a--. 2,4,6-TCB I 1 IO
I 100
Concentration (PM)
Fig. 6. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri after exposure to DMSO (solvent control; top tracing) or 100 p,M 2,4,6-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to exposure to DMSO (solvent control) or 2,4,6-TCB. C) The concentrationdependent effect of 2,4,6-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means f SEM of five individual uteri. If no error bar is observed, the standard error from those data is smaller than the size of the symbol. Asterisks indicate significant differences (P < 0.05) compared with DMSO (solvent control).
4o
...X.** 9: ‘?,
+DMSO
--*Q-*0 4_OH_TCB o0.1
I 1
‘i; *
1 10 Concentration (PM)
I 100
Fig. 7. A) Representative polygraph tracing of oscillatory contractions of pregnant uteri after exposure to DMSO (solvent control; top tracing) or 30 pM 4-OH-TCB (bottom tracing). B) Basal oscillatory activity of uterine strips prior to treatment with DMSO (solvent control) or 4-OH-TCB. C) The concentration-dependent effect of 4-OH-TCB on spontaneous oscillatory contractions of pregnant uteri. Data graphed in B and C represent the means f SEM of five individual uteri. If no error bar is observed, the standard error from those data is smaller than the size of the symbol. Asterisks indicate significant differences (P < 0.05) compared with DMSO (solvent control).
lower concentrations (0.3 to 10 pM) of 2,4,6-TCB did not have any significant effects. Representative tracings show the responses of uterine strips to 100 pM 2,4,6TCB or DMSO solvent control (Figure 6A). Figure 6B shows that the basal oscillatory activity was not significantly different in 2,4,6-TCB-treated and solvent control groups, suggesting that basal oscillatory activity did not affect the response to 2,4,6-TCB. The concentration-dependent effect of 4-OH-TCB on spontaneous oscillatory contractions of pregnant uteri Uterine strips exhibited decreased oscillatory activity with increasing concentrations of 4-OH-TCB (Figure 7C). Relative to DMSO solvent controls, 3, 10, and 30 (LM 4-OH-TCB significantly decreased uterine contractions to 73 + 11, 52 f 10, and 8.5 f 5.6% of basal oscillatory activity, respectively. The ED,, value was less than 10 PM. Representative tracings show the in-
Reproductive
26
Toxicology
hibitory effect of 30 p,M 4-OH-TCB compared to solvent controls (Figure 7A). Because of no significant difference between 4-OH-TCB-treated and DMSO-treated solvent control groups, the inhibition of spontaneous activity by 4-OH-TCB did not result from a difference of basal oscillatory activity.
DISCUSSION In vivo or in vitro, uterine oscillatory contractions can arise spontaneously as a myogenic response (18) or can be induced by physiologic chemicals such as oxytotin (19). In the present work, spontaneously contracting uterine strips isolated from midgestation rats were used to examine PCB congener effects on uterine oscillatory activity. Consistent with a previous report (17) these spontaneous oscillatory contractions could not be blocked by indomethacin, a cyclooxygenase inhibitor. The data demonstrated that PCB effects on spontaneous oscillatory uterine activity depended on the structure of the PCB congener. With acute exposure, uterine strips isolated from midgestation pregnant rats exhibited increased activity in the presence of 2,4-TCB and 2,4,6TCB, whereas oscillatory contractions were inhibited by 4-OH-TCB. The coplanar PCB congeners 3,4-TCB and 3,4-PCB, as well as the non-coplaner congener HCB. had no significant acute effects on spontaneous uterine contractions. Certain patterns emerge from an examination of the data with respect to the PCB congener structure. First, the lightly chlorinated congeners with ortho-, parusubstitution (4-OH-TCB, 2,4,6-TCB, and 2,4-TCB) modified uterine contractions. Second, the replacement of ortho-substitution (2,4-TCB) with meta-substitution (3,4-TCB) changed the uterine response from observable to nonobservable. Together, these observations suggest that ortho-chlorine substitution may be necessary for acute PCB actions on oscillatory contractions of pregnant uteri. Third, the addition of a chlorine residue to the metu-positions of 2,4-TCB to form HCB shifted the response from observable to nonobservable. This finding suggests that meta-substitution may interfere with the activity of ortho-substituted congeners on uterine contractions. Fourth, 3,4-TCB and 3,4-PCB did not alter uterine contractions, yet they are potent inducers of aryl hydrocarbon hydroxylase, function as analogs of 2,3,7,8tetrachlorodibenzodioxin (TCDD) and activate the Ah receptor (3). This observation suggests that dioxin-like compounds are not involved in acute alterations of uterine contractions. Fifth, the maximum efficacy for 2,4TCB is less than that of 2,4,6-TCB, suggesting that its bulky size may interfere with PCB activity in the uterus. In summary, because of the structural specificity of the
Volume IO. Number
I, lYY6
PCB congener actions, it is suggested that the PCB congeners interact with specific molecules in the uterus to alter uterine contractions. Even though the analogs 2,4,6-TCB and 4-OH-TCB both altered uterine contractions, the response to 2,4,6TCB was opposite to that of 4-OH-TCB: 2,4,6-TCB stimulated uterine contractions but 4-OH-TCB inhibited uterine contractions. Because of their structural similarity. 2,4,6-TCB and 4-OH-TCB may act on the same pathway, one as an antagonist and the other as an agonist. However, it is also possible that 2,4,6-TCB and 4-OHTCB act on separate pathways. PCBs can be metabolized by many different pathways. The most important pathway is through hydroxylation and subsequent conjugation (3). Hydroxylation of dichloro-, trichloro-, tetrachloro-, pentachloro-, and hexachlorobiphenyl is favored at the pm-u position in the least chlorinated phenyl ring (20). Higher chlorinated PCBs tend to be metabolized more slowly and accumulate to a greater extent (21,22). In our design, 4-OH-TCB was used as a hydroxylated analog of 2,4,6-TCB. If hydroxylation were required for activity and the uterine tissue had significant metabolic capacity to hydroxylate 2,4,6-TCB, then the response to 4-OH-TCB and 2,4,6TCB would be similar, but the potency of 4-OH-TCB would be greater than that of 2,4,6-TCB. In fact, our observation was that 2,4,6-TCB stimulated oscillations and 4-OH-TCB suppressed oscillations, inconsistent with that model. Considering these results, together with the very short period of exposure, it is unlikely that metabolites produced by the uterus contributed to the alteration of uterine contractions in our system. Korach et al. (23) demonstrated that 4-OH-TCB is estrogenic in that it competitively inhibits estrogen binding to estrogen receptors. 4-OH-TCB also increases the weight and cell proliferation of immature rat uteri, further supporting its estrogenicity (24). The effect of 4-OH-TCB on uterine contractions in our study is consistent with previous reports that estrogenic compounds rapidly inhibit uterine contractions (14,25). In a previous report, the rapid inhibitory effect of estrogens was proposed to be nongenomic and related to inhibition of extracellular calcium entry (25). In our laboratory, we have also observed that 17@estradiol rapidly inhibits spontaneous oscillatory uterine contractions in a manner similar to that observed with 4-OH-TCB (26). However, in our experiments, the calcium ionophore A23187 (up to 2 FM) was unable to reverse the inhibitory action of either 17&estradiol(26) or 4-OH-TCB (27) on uterine contractions, Rather, our results suggest that inhibition of gap junctional communication may be involved in the inhibition of spontaneous oscillations by 17P-estradiol (26) and 4-OH-TCB (27). Garfield and his colleagues have shown that cell-to-cell coupling via gap junctions is as-
PCBs and Uterine Contractions
sociated with the development of synchronized oscillatory contractions in pregnant uteri at term (28). The mechanism by which the PCB congeners 2,4,6TCB and 2,4-TCB increase spontaneous oscillatory activity is not clear. However, it has been shown that 2,4-TCB increases the production of superoxide in neutrophils, which then causes the activation of the phosphoinositol signal transduction pathway (29). A lightly chlorinated PCB congener, 2,2’-dichlorobiphenyl, increases intracellular calcium by inhibition of mitochondrial calcium uptake in cerebellar granule cells (30). It is known that increased phosphoinositide turnover and calcium accumulation are associated with increased uterine contractions (31). Thus, a possible mechanism for 2,4TCB and 2,4,6-TCB actions may include elevated production of superoxide and increased intracellular calcium, which in turn increase uterine contractions. Interestingly, PCB congeners with light chlorination at the or&-position are the most active neurotoxicants (32) a pattern similar to our observations with uterine contractions. In contrast, the highly chlorinated PCB congeners such as 3,4,5,3’,4’-pentachloroand 3,4,3’,4’tetrachlorobiphenyl tend to be more potent inducers of aryl hydrocarbon hydroxylase (AHH) in liver. Thus, we speculate that the mechanism responsible for the lightly chlorinated PCB action may be different from that for the highly chlorinated PCBs. In summary, our data show that PCB congeners modify in vitro contractions of pregnant uteri in an acute, congener-specific manner. Because of the relationship between PCB structure and activity, it is suggested that the uterine response to PCBs requires congeners with or&o-substituted light chlorination that can interact with specific molecules, but not with the Ah-receptor in the uterus. The uterine responses were divergent and even opposite in nature depending on the structure of the PCB congener. PCB contamination of the environment usually occurs as a mixture, and synergistic, additive or antagonistic interactions may occur among the different congeners. However, mixtures of congeners were not evaluated in this study, so no definate conclusions can be drawn regarding such interactions and uterine contractility. Finally, these experiments examined only the midgestation uterus. Consequently, it is not known if the responses observed are specific for this stage of pregnancy or if they would also be observed at other stages of pregnancy or in the nonpregnant uterus. Acknowledgments - The authors thank Shelly Coe and Howard Listopad for their assistance with these experiments and Dr. Craig Harris for providing uterine tissue. The project described was conducted as partial fulfillment of M.-L. Tsai’s doctoral dissertation and was supported by grant number HL18575 to R. C. Webb and P42 ES0491 1 to R. Loch-Caruso from the National Heart, Lung, and Blood Institute and the National Institute of Environmental Sciences, respectively, of the NIH. Additional support was provided by the Laboratory Animal Cores
l
M.-L. TSAI ET AL.
21
of the Center for the Study of Reproduction, which is supported by grant number P30 HD18258 from the National Institute of Child Health and Human Development, NIH.
REFERENCES 1. Hutzinger 0, Safe S, Zitko V, eds. The chemistry of PCBs. Cleveland: CRC Press; 1974. 2. Erickson MD, ed. Analytical chemistry of PCBs. Boston: Butterworth; 1986. 3. Safe S. Polychlorinated biphenyls (PCBs) and polybrominated biphenyls (PBBs): biochemistry, toxicology, and mechanism of action. CRC Crit Rev Toxicol. 1984;13:319-95. 4. Altman NH, New AE, McConnell EE, Ferrell TL. A spontaneous outbreak of polychlorinated biphenyl (PCB) toxicity in rhesus monkeys (Macaca m&m): clinical observations. Lab Anim Sci. 1979;29:661-5. 5. Wassermann M, Ron M, Bercovici B, Wassermann D, Cucos S, Pines A. Premature delivery and organochlorine compounds: polychlorinated biphenyls and some organochlorine insecticides. Environ Res. 1982;28:106-12. 6. Taylor PR, Lawrence C, Hwang HL, Paulson AS. Polychlorinated biphenyls: influence on birthweight and gestation, Am J Public Health. 1984;74: 1153-4. I. Leoni V, Fabiani L, Marinelli G, et al. PCB and other organochlotine compounds in blood of women with or without miscarriage: a hypothesis of correlation. Ecotoxicol Environ Safety. 1989;17: l11. 8. White RD, Allen SD, Bradshaw WS. Delay in the onset of parturition in the rat following prenatal administration of developmental toxicants. Toxic01 Lett. 1983;18: 185-92. 9. T&ok P. Delayed pregnancy in NMRI mice treated with PCB: 2,2’-dichlorobiphenyl. Bull Environ Contam Toxicol. 1976; 16: 33-6. 10. Bell R. The prediction of preterm labor by recording spontaneous antenatal uterine activity. Br J Obstet Gynecol. 1983;90:88&7. 11. Morgan MA, Silavin SL, Wentworth RA. et al. Different patterns of myometrial activity and 24-h rhythms in myometrial contractility in the gravid baboon during the second half of pregnancy. Biol Reprod. 1992;46:1158864. 12. Boyd ME, Usher RH, McLean FH, Kramer MS. Obstetric consequences of postmaturity. Am J Obstet Gynecol. 1988;158:334-8. 13. Modzelewski P, Szamatowicz J, Laudanski T, MoniuszukoJakoniuk J, Akerlund M. The influence of lead ions on uterine activity in the rat. Int J Gynaecol Obstet. 1990;32:169-73. 14. Juberg DR. Loch-Caruso R. Investigation of the role of estrogenic action and prostaglandin E, in DDT-stimulated rat uterine contractions ex vivo. Toxicology. 1992;74:161-72. 15. Polishuk ZW, Wassermann Organochlorine compounds viron Res. 1977;13:278-84.
D, Wassermann M, Cucos S, Ron M. in mother and fetus during labor. En-
16. Juberg DR, Webb RC, Loch-Caruso R. Characterization of o,p’DDT-stimulated contraction frequency in rat uterus in vitro. Fundam Appl Toxicol. 1991;17:543-9. 17. Poli E, Merialdi A, Coruzzi G. Characterization of the spontaneous motor activity of the isolated human pregnant myometrium. Pharmacol Res. 1990;22: 115-24. 18. Ruegg JC. Smooth muscle tone. Physiol Rev. 197 I;5 I :20148. 19. Honnebier MB, Myers T, Figueroa JP, Nathanielsz PW. Variation in myometrial response to pulsatile intravenous oxytocin administration at different times of the day in the pregnant rhesus monkey. Endocrinology. 1989; 125: 1498-503. 20. Borlakoglu
JT, Wilkins
JP. Metabolism
of di-, tri-, tetra-, and
28
21.
22.
23.
24.
25.
26.
Reproductive
Toxicology
hexachlorobiphenyls by hepatic microsomes isolated from control animals and animals treated with Aroclor 1254, a commercial mixture of polychlorinated biphenyls (PCBs). Comp Biochem Physiol C. 1993;105:107-12. Lutz RJ, Dedrick RL, Matthews HB, Eling TE, Anderson MW. A preliminary pharmacokinetic model for several chlorinated biphenyls in the rat. Drug Metab Dispos. 1977;5:386-96. Kaley RG, Hicks 0, Mees WM, et al. Tissue residues from subacute oral feeding of polychlorinated biphenyl dielectric fluids. Bull Environ Contam Toxicol. 1976;15:699-707. Korach KS, Sarver P, Chae K, McLachlan JA, McKinney JD. Estrogen receptor-binding activity of polychlorinated hydroxybiphenyls: conformationally restricted structural probes. Mol Pharmacol. 1988;33:120-6. Jansen HT, Cooke PS, Porcelli J, Liu T-C, Hansen LG. Estrogenic and antiestrogenic actions of PCBs in the female rat: in vitro and in vivo studies. Reprod Toxicol. 1993;7:237-48. Femandez AI, Martinez V, Cantabrana B, Hidalgo A. Differential effect of calcium and Bay K 8644 on the inhibitory action of estrogens in the rat uterus. Gen Pharmacol. 1992;23:549-54. Tsai M-L, Webb RC, Loch-Caruso R. Inhibition of spontaneous oscillatory uterine contractions by 17B-estradiol is associated with
Volume 10, Number
1, 1996
disruption of gap junctional 1995;52(Suppl 1):115.
communication.
Biol
Reprod.
27. Tsai M-L, Webb RC, Loch-Caruso R. Acute inhibition of uterine contraction by 4-hydroxy-2’,4’,6’-trichlorobiphenyl (40HTCB) may result from disruption of gap junctional cologist. 1995;15:275.
communication.
Toxi-
28. Garfield RE, Sim S, Daniel EE. Gap junctions: their presence and necessity in myometrium during parturition. Science. 1977; 198: 958860. 29. Ganey PE, Sirois JE, Denison M, Robinson JP, Roth RA. Neutrophi1 function after exposure to polychlotinated biphenyls in vitro. Environ Health Perspect. 1993;101:430-4. 30. Kodavanti PRS, Shin DS, Tilson HA, Harry GJ. Comparative effects of two polychlorinated biphenyl congeners on calcium homeostasis in rat cerebellar granule cells. Toxic01 Appl Pharmacol. 1993;123:97-106. 31. Garfield RE, ed. Uterine contractility. posia: 1990.
Norwell, MA: Serono Sym-
32. Shain W, Bush B, Seegal R. Neurotoxicity of polychlorinated biphenyls: structure-activity relationship of individual congeners. Toxic01 Appl Pharmacol. 1991;111:3342.