Perinatal exposure to low levels of the environmental antiandrogen vinclozolin alters sex-differentiated social play and sexual behaviors in the rat

Environmental Health Perspectives, June, 2005 by Nathan K.W. Colbert, Nicole C. Pelletier, Joyce M. Cote, John B. Concannon, Nicole A. Jurdak, Sara B. Minott, Vincent P. Markowski

Because the penile erection data showed a clear dose-response relationship with evidence that even the lowest dose of Vz disrupted the behavior, we further examined the data with Benchmark Dose Modeling Software (BMDS; version 1.3.2; U.S. EPA National Center for Environmental Assessment, Washington, DC). BMDS is a useful alternative to the no observed adverse effect level (NOAEL) approach because it uses the entire dose-response relationship and does not involve extrapolations far below experimental observations. We used the BMDS continuous model to calculate benchmark doses that represented the model-estimated control mean, minus proportional deviations equivalent to a 10% (E[D.sub.10]) or 1% (E[D.sub.01]) decrement in behavior. BMDS also provided a 95% lower bound that can be divided by a standard uncertainty factor to calculate a reference dose or generate a margin of exposure.

Results

Maternal and postpartum data. We found no evidence of gross maternal or neonatal toxicity (Table 1), nor were there any exposure-related changes in maternal body weight, pup body weight, or AGD (males, Table 2; females, Table 3). The percentage of male pups that possessed at least one visible areola on PND 12 (control, 18%; 1.5 mg/kg, 50%; 3 mg/kg, 44%; 6 mg/kg, 45%; 12 mg/kg, 33%) was not significant (p = 0.367).

Play behavior. For this procedure, the primary dependent variable was the total number of play behaviors per session. Although we found no significant main effects of the exposure, sex, or PND factors on total play behaviors, there was a significant exposure x PND interaction [F(1,21) = 7.72, p = 0.01]. We also examined the total number of behaviors separately for PND22 and PND34. There was a significant effect of sex on PND22 [males > females; F(1,21) = 9.91, p < 0.01] and a significant effect of exposure on PND34 [F(1,37) = 16.38, p < 0.001]. Probe tests revealed that the male 12-mg/kg and 6-mg/kg Vz groups produced significantly more play behaviors than the did controls on PND34 (Figure 1A). There were no differences between the female exposure groups (Figure 1B).

[FIGURE 1 OMITTED]

Nape contact and pounce variables made the greatest contribution to the significant exposure-related effects on total play behaviors. For nape contacts, there was a significant exposure x PND interaction [F(1,21) = 5.51, p = 0.03]. We also examined the number of nape contacts separately for PND22 and PND34. As with the total play behavior variable, there was a significant main effect of sex on PND22 [males > females; F(1,21) = 11.13, p < 0.01] and a significant main effect of exposure on PND34 [F(1,37) = 16.09, p < 0.001]. Probe tests indicated that the male 12-mg/kg Vz group produced significantly more nape contacts than did the 0- and 3-mg/kg groups on PND34 (Figure 2). For the pounce variable, there was a significant main effect of exposure [F(1,21) = 6.44, p = 0.02]. Data were averaged across sex and age, and probe tests indicated that the 12-mg/kg group pounced more than did controls (Figure 3). There were no exposure-related differences for pin, wrestle, or mount behaviors.

[FIGURES 2-3 OMITTED]

Penile reflex. We found a significant exposure-related decline in total erections per session [F(4,40) = 4.62, p < 0.01; Figure 4] as each of the Vz groups produced significantly fewer erections than controls. The decline in total erections was due primarily to a dose-related decline of E1 or low-intensity erections [F(4,40) = 10.07, p < 0.01] as well as the number of reflex clusters per session [F(4,40) = 3.23, p = 0.02; Figure 5]. The latency to the first penile reflex and the frequency of E2 and E3 responses were not significantly different. Surprisingly, there was a significant increase in seminal emissions [F(4,40) = 7.37, p < 0.01; Figure 6] as the 12-mg/kg group expelled more often than did any of the other groups. This effect was unanticipated because rats do not usually emit seminal fluid during the ex copula procedure.