- Email: [email protected]
Effect of Postnatal Exposure to a PCB Mixture in Monkeys on Multiple Fixed Interval–Fixed Ratio Performance
Neurotoxicology and Teratology, Vol. 19, No. 6, pp. 429–434, 1997 Crown Copyright © 1997. Published by Elsevier Science Inc. Printed in the USA. All rights reserved 0892-0362/97 $0.00 1 .00
PII S0892-0362(97)00029-9
Effect of Postnatal Exposure to a PCB Mixture in Monkeys on Multiple Fixed Interval–Fixed Ratio Performance DEBORAH C. RICE Toxicology Research Division, Bureau of Chemical Safety, Food Directorate, Health Protection Branch, Health Canada, Tunney’s Pasture 2202D1, Ottawa, Ontario K1A 0L2, Canada Received 17 October 1996; Accepted 3 March 1997 RICE, D. C. Effect of postnatal exposure to a PCB mixture in monkeys on multiple fixed interval–fixed ratio performance. NEUROTOXICOL TERATOL 19(6) 429–434, 1997.—Behavioral impairment as a consequence of PCB exposure beginning in utero has been reported in both humans and animals. The present study assessed the behavioral consequences of postnatal exposure to PCBs. Male monkeys (Macaca fascicularis) were dosed from birth to 20 weeks of age with 7.5 mg/kg/day of a PCB mixture representative of the PCBs typically found in human breast milk (eight monkeys) or vehicle (four monkeys). At 4 years of age, performance under a multiple fixed interval (FI)-fixed ratio (FR) schedule of reinforcement was assessed. The FI component was more sensitive to disruption as a result of PCB exposure than was the FR component. PCB-exposed monkeys displayed shorter mean interresponse times (IRTs) than controls, particularly during the earlier sessions of the experiment. Similarly, the increase in pause time characteristic of the acquisition of typical FI performance emerged more slowly across sessions in the PCB-treated group. However, the number of short IRTs (less than 5 s) remained greater in the treated group compared to controls over the 48-session duration of the experiment. On the FR component, control monkeys decreases the mean pause time across sessions whereas the PCB-treated group did not; there were no differences between groups for absolute value of average IRT or pause time. The results of this study extend previous research in this cohort of monkeys, and provide further evidence that PCB exposure limited to the early postnatal period and resulting in environmentally relevant body burdens produces long-term behavioral effects. Crown Copyright © 1997. Published by Elsevier Science Inc. PCB exposure
Fixed interval
Fixed ratio
Behavioral deficits
Monkeys
concurrently to more toxic polychlorinated dibenzofurans (PCDFs). A prospective study of children born to women who consumed PCB-contaminated fish from Lake Michigan (6,17– 21,42) revealed decreased visual memory at 7 months of age associated with umbilical cord blood levels and increased fish consumption in the mothers. There was no similar association with postnatal exposure through nursing. Umbilical cord PCB levels and maternal milk PCB levels were associated with poorer short-term memory at 4 years of age (18). Concurrent serum PCB levels in the children or quantity of breast milk consumed were not associated with measures of cognitive ability. A composite measure of prenatal exposure, derived from PCB concentrations in umbilical cord serum and mater-
POLYCHLORINATED biphenyls (PCBs) are a family of chlorinated hydrocarbons persistent in the environment and can be detected in biological tissues of most residents of industrialized countries. Attention focused on the neurotoxic potential of these agents to the developing organism following an incident in Japan in which women who ingested PCBs in contaminated rice oil gave birth to infants with hypotoxic reflexes and low IQs (13). Children born after a similar incident in Taiwan were followed for up to 12 years (24,36,45): they exhibited delayed developmental milestones, cognitive deficits, and behavioral problems. These effects were observed at exposure levels that produced overt physical signs including gum hypertrophy, deformed or pigmented nails, acne, hyperpigmentation, and hair loss. These children were also exposed
Requests for reprints should be addressed to Deborah C. Rice, Toxicology Research Division, Tunney’s Pasture 2202D1, Ottawa, Ontario, K1A 0L2, Canada. Tel: (613) 957-0967; Fax: (613) 941-6959; E-mail: [email protected]
429
430 nal serum and milk, was associated with decreased IQ scores, particularly those for memory and attention, and poorer reading ability at 11 years of age in this same cohort (16). Again there was no apparent effect of exposure through nursing on outcome. In a cohort of breast-fed infants followed prospectively for 60 months in a North Carolina study (10,11,37,38) maternal milk fat PCB levels were associated with hypotonicity and hyporeflexia on the Brazelton Neonatal Behavioral Assessment Scale. Higher maternal milk fat levels were associated with lower Bayley Scale scores at 6 and 12 months (10) and 24 months (35). There was no association between either prenatal or concurrent PCB levels and a negative outcome on the McCarthy Scale of Children’s Ability at 3–5 years (10). Behavioral impairment has been observed in monkeys exposed in utero and through breast milk until 4 months of age to either Aroclor 1248 or 1016. Aroclor 1016 exposure produced impairment on spatial discrimination reversal and facilitation on a shape discrimination reversal during infancy (40,41) with no effect on a spatial delayed alternation task at 4–6 years of age (25,40). Monkeys exposed developmentally to Aroclor 1248 exhibited impairment, no impairment, or facilitated performance on spatial and nonspatial discrimination problems as infants (2,41), and impairment on a spatial delayed alternation task at 4–6 years (40). Development exposure of rats to one of three individual ortho-substituted congeners (congeners 118, 153, or 28: 2,39,4,49,5; 2,29,4,49,5,59; 2,4,49) resulted in impairment on delayed alternation performance in females only, with no effect on radial arm maze performance (39). Exposure to a coplanar congener (congener 126: 3,39, 4,49,5), however, produced inconsistent effects on performance of a visual discrimination task (1,14). In a study assessing behavior in male rat offspring exposed to the commercial mixture Clophen A30, pre- and postnatal exposure produced effects on active avoidance and fixed interval performance (27). In a subsequent study, prenatal or pre- and postnatal exposure produced deficits on active avoidance and retention of a visual discrimination task, whereas postnatal exposure produced effects of smaller magnitude that did not reach statistical significance (26). An issue highly relevant to neurotoxicity produced by developmental PCB exposure in humans concerns the relative contribution of pre- vs. postnatal exposure to observed impairment. Results from the Michigan and North Carolina epidemiological studies found no association between any neurological or behavioral outcome and postnatal exposure as assessed by the amount of PCBs received through breast milk (16,20,21,35,38), despite the fact that infants receive a much greater amount of PCBs through breast milk than across the placenta (44). In most animal studies, exposure began in utero, making it impossible to differentiate prenatal from postnatal effects. The exception to this is the cross-fostering study by Lilienthal and Winneke (26), in which postnatal exposure had minimal effect, and certainly less effect than prenatal exposure. The current study extends an examination of the behavioral consequences of postnatal exposure of monkeys to a PCB congener mixture representative of PCBs found in human breast milk. These monkeys exhibited impairment in spatial delayed alternation performance, with some evidence for retarded acquisition of a nonspatial discrimination reversal task (31). The current study examined the performance of these monkeys under a multiple fixed interval (FI)–fixed ratio (FR) schedule of reinforcement. Performance under intermittent reinforcement schedules have proven sensitive to influence by exposure to toxicants (33); combining two schedules
RICE in an assessment procedure allows more than one behavioral baseline (which may be differentially sensitive to the effects of a toxicant) to be examined in the same test session. The FI was chosen in part because it proved sensitive to developmental exposure to PCBs in rats; the FR was included because of the marked difference in baseline performance between the FI and the FR (33), and because a multiple FI-FR schedule had been used previously with other neurotoxicants in monkeys (9,32,34). METHODS
Subjects, Dosing, and PCB Levels A total of 12 male monkeys (Macaca fascicularis) born in the Health Protection Branch, Health Canada breeding colony were separated from their mothers at birth and reared by hand as part of a study designed to determine the distribution of PCBs in infant monkeys (Arnold, Bryce, Mes, Hayward, Tryphonas, and Iverson, unpublished). They were dosed from birth to 20 weeks of age with 0 (four monkeys) or 7.5 mg/kg/ day (eight monkeys) of a mixture of PCBs representative of the pattern found in human breast milk (Table 1). The mixture represented 80% of the congeners present in breast milk in Canadian women, in a proportional mixture (29). To prepare the dosing solution, the congener mixture in isooctane was placed in a beaker and evaporated under a stream of N2 to “near dryness.” One hundred milliliters of freshly made Primalac infant monkey formula (Bio-serv, Frenchtown, N.J.) was placed in a beaker and mixed for 90 min using a Teflon stir bar and stir plate. The solution was divided into 10-ml aliquots and frozen until needed. Concentrations of the dosing solutions were 5 or 10 mg/ml such that the volume given to the monkey was no more than 0.5 ml at a time. The dose was drawn into a syringe and inserted directly into the infant’s mouth. Each daily dose was divided into three equal parts, and delivered at 0800, 1300, and 1500 h 7 days per week. Controls were dosed with Primalac vehicle. Infants were weighed daily, and doses adjusted accordingly. Blood was drawn at 20 weeks of age from all monkeys and analyzed for PCBs as described previously (30). Samples were
TABLE 1 CONGENER MIXTURE OF DOSING SOLUTION
PCB No.
Chlorine Substitution
52 66 74 105 118 138 153 156 157 180 183 187 189 194 203 15
2,29,5,59 2,39,4,49 2,4,49,5 2,3,39,4,49 2,39,4,49,5 2,29,3,4,49,59 2,29,4,49,5,59 2,3,39,4,49,5 2,3,39,4,49,59 2,29,3,4,49,5,59 2,29,3,4,49,59,6 2,29,3,49,5,59,6 2,3,39,4,49,5,59 2,29,3,39,4,49,5,59 2,29,3,4,49,5,59,6 Congener total
% in Canadian Milk
% in Monkey Mixture
1.0 2.0 8.8 3.0 10.3 14.8 15.1 3.3 1.0 10.3 1.8 3.8 0.3 2.0 1.8 79.3
1.5 2.9 10.4 4.4 12.8 17.5 18.6 4.7 1.5 12.8 2.3 4.7 0.5 2.9 2.3 99.8
PCBs AND MULTIPLE FI-FR IN MONKEYS
431
also taken from the nuchal fat pad (scapular area) at 20 weeks of age from all monkeys, and again between 90 and 99 weeks of age from two control and four treated monkeys. Analysis was restricted to the congeners administered to the monkeys; results will be published separately (Arnold, Bryce, Mes, Hayward, Tryphonas, and Iverson, unpublished). Total PCB levels are presented in Table 2. Congener 66 was excluded from the total for both blood and fat due to unexpected high spikes in some control samples. When these monkeys were 3 years of age, they were transferred to the Neurotoxicology Primate Colony for behavioral assessment. Monkeys had previously been tested on nonspatial discrimination reversal and delayed spatial alternation tasks. Testing for the current study began when monkeys were 4 years old. Behavioral Methods Schedule control and data acquisition were accomplished using the SKED system (State Systems, Kalamazoo, MI) run on a PDP 11/93 (Digital Equipment Corp., Maynard, MA). Data were collected as interevent times for all responses and schedule events, so that the session could be completely reconstructed from the raw data. Resolution was 0.01 s. Monkeys sat in a primate restraining chair inside a sound-attenuating cubicle, facing a Plexiglas panel on which was mounted a clear plastic push-button. Monkeys were tested under a multiple fixed interval (FI)–fixed ratio (FR) schedule of reinforcement for a total of 48 sessions. Each session consisted of four cycles of a FI followed by an FR; components were separated by 20-s time out (TO) period, during which responses had no scheduled consequences. The FI was signalled by a white light on the button, FR by a yellow light, and the TO by a dark button. Reinforcement was approximately 1.0 ml of apple juice. The FI value was 6 min for all sessions. The FR value was increased sequentially from 2 (sessions 1–5) to 5 (sessions 6–15), and finally 10 (sessions 16–48). Monkeys were tested 5 days per week, between approximately 0800 and 1300 h. The following variables were calculated: 1. Interresponse time (IRT), which was defined as the time between successive responses, or from the beginning of the
2. 3.
4.
5. 6.
schedule component to the first response, calculated separately for FI and FR mean and median. Pause time, defined as the time between the beginning of the schedule component and the first response, calculated for FI and FR mean and median. FI index of curvature (IOC) mean and median (8), which is a measure of the degree and direction of deviation from linearity of FI responding over the interval; intervals with fewer than eight responses were excluded from analysis. The standard deviation (SD) of the FI IRT distribution for the session divided by the FI IRT mean. The SD was normalized because it appeared to vary linearly with the absolute value of the mean IRT based on visual inspection of the data. The number of FI IRTs less than 5 s. The number of TO responses.
All measures except the number of TO responses were compared by orthogonal polynomial analyses (43) across sessions including the intercept, linear, and quadratic terms, using the SAS GLM procedure on a SUN computer. The parameters for the FR 2, 5, and 10 sessions were analyzed separately. All measures except IOC and FI pause underwent log transformation, which stabilized variance. Parameters were compared by t-tests. TO responses were compared by the Wilcoxin rank sums test. RESULTS
There were a number of differences in FI performance between control and PCB-exposed monkeys. Treated monkeys had shorter mean IRTs as measured by the intercept term of the orthogonal polynomial analysis ( p 5 0.024) (Fig. 1), as well as a difference in the development of performance across sessions as revealed by the quadratic term ( p 5 0.015). The difference between the control and treated group was most marked at the beginning of the experiment, and was less marked by the end. In contrast to the mean, the FI median IRT was only marginally significantly different ( p 5 0.093), suggesting that the IRT distribution may differ between groups. This was also indicated by a difference in the number of IRTs less than 5 s, which was greater in the PCB-treated monkeys as measured by the intercept term ( p 5 0.020) (Fig.
TABLE 2 BLOOD AND ADIPOSE PCB LEVELS (CONGENER 66 EXCLUDED) Blood (ppb Wet Weight)
Control
Treated
Adipose (ppb Lipid)
Week 20
Week 20
(Week)
0.37 0.30 nd* 0.32 1.84 2.52 2.19 nd* 1.85 2.58 2.75 2.84
198 62 50 53 3560 2368 1694 2716 2001 2520 2421 2369
211 (99)
*Below the detection limit of the analysis.
147 (94)
539 (98)
547 (94) 553 (92) 473 (90)
FIG. 1. Mean interresponse time (IRT) over the 48 FI sessions for control and PCB-treated monkeys. Group mean 6 SEM.
432
RICE
The observed differences between control and PCBtreated monkeys on the FI component of the multiple FI-FR
schedule may be interpreted as retarded acquisition. This is perhaps clearest for FI pause time, which increased more slowly over the course of the 48 sessions for the treated group compared to controls, but did not differ from control values over the last 10 sessions. A similar pattern was observed for the mean FI IRT, which was lower in treated monkey for most of the experiment, and increased to a value similar to that of controls near the end of the experiment. However, the greater number of short IRTs (less than 5 s) emitted by the treated monkeys did not decrease during later sessions, suggesting that there was a persistent (at least 48 session) difference between the performance of the control and PCBexposed groups. In fact, if anything this difference appeared to be getting larger as the experiment progressed, as control monkeys decreased their short IRTs across sessions and treated monkeys did not. The results of the current study are generally consistent with data from a study of developmental exposure to a commercial PCB mixture in rats exposed prenatally and continuing postnatally through testing at one year of age on an FI 30 s schedule of reinforcement (27). PCB-treated rats exhibited higher response rates, which is consistent with the shorter average IRT observed in the current study (response rate is the inverse of mean IRT). PCB-exposed rats also made a greater proportion of responses early in the interval compared to controls, which is consistent with the decreased pause time observed in the current study. FR performance was less sensitive to alteration by PCB exposure than was performance on the FI. This differential effect on FI and FR performance has also been observed following developmental exposure to lead in monkeys (32,34) and rats (3,4). In addition, the low FR response requirements in the current study were probably less sensitive to disruption by toxicant exposure than high FR values would have been. The monkeys in the current study had previously exhibited impaired performance on a delayed spatial alternation task (31); PCB exposure resulted in a deficit in acquisition of the task, and increased errors during early sessions of the experiment. In addition, PCB-treated monkeys made more perseverative errors even at the end of the experiment when the total error rate was not higher than that of controls. There were no delay-related effects. There was also some evidence that this group of monkeys displayed retarded acquisition of a nonspatial discrimination reversal task (31). Some PCBtreated individuals made more errors than controls over the
FIG. 2. The number of IRTs less than 5 s over the 48 FI sessions for control and PCB-treated monkeys. Group mean 6 SEM.
FIG. 3. The mean pause time over the 48 FI sessions for control and PCB-treated monkeys. Group mean 6 SEM.
2). Although the linear term was not different between groups, it appeared that the number of IRTs less than 5 s was decreasing for the control group over the last 20 sessions or so, whereas this was not true for the PCB-treated groups. However, the normalized SD of the IRT, which was a measure of response distribution, did not differ between groups ( p 5 0.291 for the intercept term of the orthogonal polynomial analysis). The FI pause time was greater for the control compared to the PCB-treated group ( p 5 0.034 for the intercept term of the orthogonal polynomial analysis) (Fig. 3). In addition, the increase in FI pause time typical of acquisition of FI performance developed more slowly in PCB-exposed monkeys, confirmed by a difference in the quadratic term ( p 5 0.045). In contrast, there was no evidence of an effect on IOC ( p 5 0.715 for the intercept). It was hypothesized that intervals with very low response rates (less than eight responses), which were included in the pause data but not the IOC calculation, might be responsible for the apparent difference. Comparison of the number of excluded intervals by t-test revealed that control monkeys had more sessions with excluded intervals (27.75 vs. 11.50, p 5 0.021) and more excluded intervals (58 vs. 19 out of a total of 192 intervals, p 5 0.022). This difference probably accounts for the apparent dichotomy in effect on FI pause time and IOC. FR performance was less affected by PCB exposure than was FI performance. There was no effect over the first five sessions, which consisted of FR 2. For the 10 FR 5 sessions, there was a difference in the quadratic term for both the IRT mean and median ( p 5 0.047 and p 5 0.037, respectively), but no difference in the average values. Visual inspection of the data revealed nothing remarkable over these 10 sessions. For the 33 FR 10 sessions, control monkeys decreased their mean pause time across sessions whereas PCB-treated monkeys did not, which was reflected in a difference in the linear term of the orthogonal polynomial analysis ( p 5 0.009). There were no other differences between control and PCB-exposed groups or FR performance. There was no difference between groups in the number of TO responses. DISCUSSION
PCBs AND MULTIPLE FI-FR IN MONKEYS
433
first several reversals at the beginning of the experiment, although the treated group as a whole was not different form controls. Combined with the results of the current study, these data suggest learning deficit, perseveration, and/or inability to inhibit inappropriate responding as a result of postnatal PCB exposure in monkeys. The fat levels in the current study were lower than in the previous monkey studies. Peak levels in the present experiment were 1.7–3.5 ppm at 20 weeks of age. In contrast, peak fat levels obtained at weaning at 4 months of age in the Aroclor 1016 study described above were 10 and 27 ppm for the low- and high-dose groups, respectively (40). The estimated peak levels at weaning for three infants whose mothers were exposed to 2.5 ppm of Aroclor 1248 during pregnancy and nursing were 21–123 ppm (2); levels from the 1.5 and 3 years postexposure groups were unavailable. Fat levels in the human population are usually reported to be between 1 and 6 ppm for unexposed populations [(12), review (23,28)], although a recent study from Germany reported average fat levels for nonoccupationally exposed adults of 300 ppb (22). The peak fat levels observed in the PCB-exposed monkeys in the current study (1.7–3.5 ppm) were in the middle of the range for environmentally exposed adult humans, whereas those of the control monkeys were lower than average reported human levels. Average whole blood PCB levels in the human general population range from 2–10 ppb [(7,22,30), reviews (23,44)], whereas levels in serum were 4 ppb based on five congeners (5). Hovinga et al. (15) re-
ported serum levels of 6 ppb in a control population and 15 ppb in a sample of fish eaters. In the North Carolina infant study, maternal serum levels averaged 9 ppb (37), whereas in the Michigan study maternal serum PCB levels were 5 ppb, with umbilical cord levels of 2.5 ppb. The whole blood PCB levels of exposed monkeys in the present study are toward the low end of results from the general human population, whereas those of the control monkeys were below most reported averages. It may be that the dose administered to the treated monkeys represents an approximation of typical intake by humans. In summary, monkeys were exposed from birth to 20 weeks of age to a low dose of a PCB mixture representative of the PCB congeners found in human breast milk. When tested at 4 years of age, they exhibited performance different from that of controls under a multiple FI-FR schedule of reinforcement. These data extend previous behavioral assessments in these monkeys and provide further evidence that exposure to an environmentally relevant level of a PCB mixture for a short time postnatally results in long-lasting behavioral effects. ACKNOWLEDGEMENTS
The author thanks Stephen Hayward of the Bureau of Biostatistics and Computer Applications for statistical analyses, and Virginia Liston, Bruce Martin, Wendy Cherry, Michelle Warankie, and Gary Vickers for expert technical assistance.
REFERENCES 1. Bernhoft, A.; Nafstad, I.; Engen, P.; Skaare, J. U.: Effects of preand postnatal exposure to 3,39,4,49,5-penthachlorobiphenyl on physical development, neurobehavior and xenobiotic metabolizing enzymes in rats. Environ. Toxicol. Chem. 13:1589–1597; 1994. 2. Bowman, R. E.; Hieronimus, M. P.; Allen, J. R.: Correlation of PCB body burden with behavioral toxicology in monkeys. Pharmacol. Biochem. Behav. 9:49–56; 1978. 3. Cory-Slectha, D. A.: Prolonged lead exposure and fixed ratio performance. Neurobehav. Toxicol. Teratol. 8:237–244; 1986. 4. Cory-Slechta, D. A.; Weiss, B.; Cox, C.: Delayed behavioral toxicity of lead with increasing exposure concentration. Toxicol. Appl. Pharmacol. 71:342–352; 1983. 5. Dar, E.; Kanarek, M. S.; Anderson, H. A.; Sonzogni, W. C.: Fish consumption and reproductive outcomes in Green Bay, Wisconsin. Environ. Res. 59:189–201; 1992. 6. Fein, G. G.; Jacobson, J. L.; Jacobson, S. W.; Schwartz, P. H.; Dowler, J. K.: Prenatal exposure to polychlorinated biphenyls: Effects on birth size and gestational age. J. Pediatr. 105:315–320; 1984. 7. Frank, R.; Braun, H. E.; Thorpe, B.: Comparison of DDE and PCB residues in the general diet and in human blood—Ontario 1986–87. Bull. Environ. Contam. Toxicol. 51:146–152; 1993. 8. Fry, W.; Kelleher, R.; Cook, L.: A mathematical index of performance on fixed-interval schedules of reinforcements. J. Exp. Anal. Behav. 3:193–199; 1960. 9. Gilbert, S. G.; Rice, D. C.; Burbacher, T. M.: Fixed interval-fixed ratio performance in adult monkeys exposed in utero to methylmercury. Neurotoxicol. Teratol. 18:539–546; 1996. 10. Gladen, B. C.; Rogan, W. J.: Effect of perinatal polychlorinated biphenyls and dichlorodiphenyl dichloroethene in later development. J. Pediatr. 119:58–63; 1991. 11. Gladen, B. C.; Rogan, W. J.; Hardy, P.; Thullen, J.; Tingelstad, J.; Tully, M.: Development after exposure to polychlorinated biphenyls and dichlorodiphenyl dichloroethene transplacentally and through human milk. J. Pediatr. 113:991–995; 1988. 12. Gómez-Catalán, J.; Lezaun, M.; To-Figueras, J.; Corbella, J.: Organochlorine residues in the adipose tissue of the population
13. 14.
15.
16. 17. 18.
19.
20. 21. 22.
of Navarra (Spain). Bull. Environ. Contam. Toxicol. 54:534–540; 1995. Harada, M.: Intrauterine poisoning: clinical and epidemiological studies of the problem. Bull. Instit. Constit. Med. (Kumamoto Univ.) 25 (suppl.): 1–60; 1976. Holene, E.; Nafstad, I.; Skaare, J. U.; Bernhoft, A.; Engen, P.; Lagvolden, T.: Behavioral effects of pre- and postnatal exposure to individual polychlorinated biphenyl congeners in rats. Environ. Toxicol. Chem. 14:967–976; 1995. Hovinga, M. E.; Sowers, M.; Humphreys, H. E.: Environmental exposure and lifestyle predictors of lead, cadmium, PCB, and DDT levels in Great Lakes fish eaters. Arch. Environ. Health. 48:98–104; 1993. Jacobson, J. L.; Jacobson, S. W.: Intellectual impairment in children exposed to polychlorinated biphenyls in utero. N. Engl. J. Med. 335:783–789; 1996. Jacobson, J. L.; Jacobson, S. W.; Humphrey, H. E. B.: Effects of exposure to PCBs and related compounds on growth and activity in children. Neurotoxicol. Teratol. 12:319–326; 1990. Jacobson, J. L.; Jacobson, S. W.; Humphrey, H. E. B.: Effects of in utero exposure to polychlorinated biphenyls (PCBs) and related contaminants on cognitive functioning in young children. J. Pediatr. 116:38–45; 1990. Jacobson, J. L.; Humphrey, H. E. B.; Jacobson, S. W.; Schantz, S. L.; Mullin, M. D.; Welch, R.: Determinants of polychlorinated biphenyls (PCBs), polybrominated biphenyls (PBBs), and dichlorodiphenyl trichloroethane (DDT) levels in the sera of young children. Am. J. Public Health 79:1401–1404; 1989. Jacobson, S. W.; Fein, G. G.; Jacobson, J. L.; Schwartz, P. M.; Dowler, J. K.: The effect of intrauterine PCB exposure on visual recognition memory. Child. Dev. 56:853–860; 1985. Jacobson, S. W.; Fein, G. G.; Schwartz, P. M.; Dowler, J. K.: Perinatal exposure to an environmental toxin: a test of multiple effects model. Dev. Physiol. 20:523–532; 1984. Kannan, N.; Schulz-Bull, D. E.; Petrick, G.; Duinker, J. C.; Macht-Hausmann, M.; Wasserman, O.: Toxic chlorobiphenyls in adipose tissue and whole blood of an occupationally/accidentally
434
23.
24. 25. 26. 27. 28.
29. 30.
31.
32. 33. 34.
RICE exposed man and the general population. Arch. Environ. Health 49:375–382; 1994. Kocan, A.; Petrik, J.; Drobná, B.; Chovancová, J.: Levels of PCBs and some organochlorine pesticides in the human population of selected areas of the Slovak Republic. Chemosphere 29:2315– 2325; 1994. Lai, T. J.; Guo, Y. L.; Yu, M. L.; Ko, H. C.; Hsu, C. C.: Cognitive development in Yucheng children. Chemosphere 29:2405–2411; 1994. Levin, E. D.; Schantz, S. L.; Bowman, R. E.: Delayed spatial alternation deficits resulting from perinatal PCB exposure in monkeys. Arch. Toxicol. 62:267–273; 1988. Lilienthal, H.; Winneke, G.: Sensitive periods for behavioral toxicity of polychlorinated biphenyls: Determination by cross-fostering in rats. Fundam. Appl. Toxicol. 17:368–375; 1991. Lilienthal, H.; Neuf, M.; Munoz, C.; Winneke, G.: Behavioral effects of pre plus postnatal exposure to a mixture of low chlorinated PCBs in rats. Fundam. Appl. Toxicol. 15:457–467; 1990. Loganathan, B. G.; Tanabe, S.; Hidaka, Y.; Kawano, M.; Hidaka, H.; Tatsukawa, R.: Temporal trends of persistent organochlorine residues in human adipose tissue from Japan, 1928–1985. Environ. Pollution. 81:31–39; 1993. Mes, J.; Marchand, L.: Comparison of some specific polychlorinated biphenyl isomers in human and monkey milk. Bull. Environ. Contam. Toxicol. 39:736–742; 1987. Mes, J.; Doyle, J.; Adams, B.; Davies, D.; Turton, D.: Polychlorinated biphenyls and organochlorine pesticides in milk and blood of Canadian women during lactation. Arch. Environ. Contam. Toxicol. 13:217–223; 1984. Rice, D. C.; Hayward, S.: Effects of postnatal exposure to a PCB mixture in monkeys on nonspatial discrimination reversal and delayed alternation performance. Neurotoxicology 18:479–494; 1997. Rice, D. C.: Lead exposure during different developmental periods produces different effects on FI performance in monkeys tested as juveniles and adults. Neurotoxicology 13:757–770; 1992. Rice, D. C.: Quantification of operant behavior. Toxicol. Lett. 43:361–379; 1988. Rice, D. C.: Schedule-controlled behavior in infant and juvenile
35. 36.
37.
38.
39.
40. 41.
42.
43. 44. 45.
monkeys exposed to lead from birth Neurotoxicology 9:75–88; 1988. Rogan, W. J.; Gladen, B. C.: PCBs, DDE, and children’s development at 18 and 24 months. Ann. Epidemiol. 1:407–413; 1991. Rogan, W. J.; Gladen, B. C.; Hung, K. L.; Koong, S. L.; Shia, L. Y.; Taylor, J. S.; Wu, Y. C.; Yang, D.; Ragan, N. B.; Hsu, C.C.: Congenital poisoning by polychlorinated biphenyls and their contaminants in Taiwan. Science 241:334–336; 1988. Rogan, W. J.; Gladen, B. C.; McKinney, J. D.; Carreras, N.; Hardy, P.; Thullen, J.; Tingestad, J.; Tylly, M.: Polychlorinated biphenyls (PCBs) and dichlorodiphenyl dichloroethene (DDE) in human milk: Effects of maternal factors and previous lactation. Am. J. Public Health 76:172–177; 1986. Rogan, W. J.; Gladen, B. C.; McKinney, J. D.; Carreras, N.; Hardy, P.; Thullen, J.; Tingelstad, J.; Tylly, M.: Neonatal effects of transplacental exposure to PCBs and DDE. J. Pediatr. 109: 335–341; 1986. Schantz, S. L.; Moshtaghian, J.; Ness, D. K.: Spatial learning deficits in adult rats exposed to ortho-substituted PCB congeners during gestation and lactation. Fundam. Appl. Toxicol. 26:117– 126; 1995. Schantz, S. L.; Levin, E. D.; Bowman, R. E.: Long-term neurobehavioral effects of perinatal polychlorinated biphenyl (PCB) exposure in monkeys. Environ. Toxicol. Chem. 10:747–756; 1991. Schantz, S. L.; Levin, E. D.; Bowman, R. E.: Hieronimus, M.; Laughlin, N. K.: Effects of perinatal PCB exposure on discrimination-reversal learning in monkeys. Neurotoxicol. Teratol. 11;243– 250; 1989. Schwartz, P. M.; Jacobson, S. W.; Fein, G. G.; Jacobson, J. L.; Price, H. A.: Lake Michigan fish consumption as a source of polychlorinated biphenyls in human cord serum, maternal serum, and milk. Am. J. Public Health 73:293–296; 1983. Snedecor, G. W.: Statistical methods. Ames: Iowa State University Press; 1980. WHO (World Health Organization). Environmental Health Criteria 140: Polychlorinated Biphenyls and Terphenyls. Geneva; 1993. Yu, M. L.; Hsu, C. C.; Guo, Y. L.; Lai, T. J.; Chen, S. J.; Luo, J. M.: Disordered behavior in the early-born Taiwan Yucheng Children. Chemosphere. 29:2413–2422; 1994.
PII S0892-0362(97)00029-9
Effect of Postnatal Exposure to a PCB Mixture in Monkeys on Multiple Fixed Interval–Fixed Ratio Performance DEBORAH C. RICE Toxicology Research Division, Bureau of Chemical Safety, Food Directorate, Health Protection Branch, Health Canada, Tunney’s Pasture 2202D1, Ottawa, Ontario K1A 0L2, Canada Received 17 October 1996; Accepted 3 March 1997 RICE, D. C. Effect of postnatal exposure to a PCB mixture in monkeys on multiple fixed interval–fixed ratio performance. NEUROTOXICOL TERATOL 19(6) 429–434, 1997.—Behavioral impairment as a consequence of PCB exposure beginning in utero has been reported in both humans and animals. The present study assessed the behavioral consequences of postnatal exposure to PCBs. Male monkeys (Macaca fascicularis) were dosed from birth to 20 weeks of age with 7.5 mg/kg/day of a PCB mixture representative of the PCBs typically found in human breast milk (eight monkeys) or vehicle (four monkeys). At 4 years of age, performance under a multiple fixed interval (FI)-fixed ratio (FR) schedule of reinforcement was assessed. The FI component was more sensitive to disruption as a result of PCB exposure than was the FR component. PCB-exposed monkeys displayed shorter mean interresponse times (IRTs) than controls, particularly during the earlier sessions of the experiment. Similarly, the increase in pause time characteristic of the acquisition of typical FI performance emerged more slowly across sessions in the PCB-treated group. However, the number of short IRTs (less than 5 s) remained greater in the treated group compared to controls over the 48-session duration of the experiment. On the FR component, control monkeys decreases the mean pause time across sessions whereas the PCB-treated group did not; there were no differences between groups for absolute value of average IRT or pause time. The results of this study extend previous research in this cohort of monkeys, and provide further evidence that PCB exposure limited to the early postnatal period and resulting in environmentally relevant body burdens produces long-term behavioral effects. Crown Copyright © 1997. Published by Elsevier Science Inc. PCB exposure
Fixed interval
Fixed ratio
Behavioral deficits
Monkeys
concurrently to more toxic polychlorinated dibenzofurans (PCDFs). A prospective study of children born to women who consumed PCB-contaminated fish from Lake Michigan (6,17– 21,42) revealed decreased visual memory at 7 months of age associated with umbilical cord blood levels and increased fish consumption in the mothers. There was no similar association with postnatal exposure through nursing. Umbilical cord PCB levels and maternal milk PCB levels were associated with poorer short-term memory at 4 years of age (18). Concurrent serum PCB levels in the children or quantity of breast milk consumed were not associated with measures of cognitive ability. A composite measure of prenatal exposure, derived from PCB concentrations in umbilical cord serum and mater-
POLYCHLORINATED biphenyls (PCBs) are a family of chlorinated hydrocarbons persistent in the environment and can be detected in biological tissues of most residents of industrialized countries. Attention focused on the neurotoxic potential of these agents to the developing organism following an incident in Japan in which women who ingested PCBs in contaminated rice oil gave birth to infants with hypotoxic reflexes and low IQs (13). Children born after a similar incident in Taiwan were followed for up to 12 years (24,36,45): they exhibited delayed developmental milestones, cognitive deficits, and behavioral problems. These effects were observed at exposure levels that produced overt physical signs including gum hypertrophy, deformed or pigmented nails, acne, hyperpigmentation, and hair loss. These children were also exposed
Requests for reprints should be addressed to Deborah C. Rice, Toxicology Research Division, Tunney’s Pasture 2202D1, Ottawa, Ontario, K1A 0L2, Canada. Tel: (613) 957-0967; Fax: (613) 941-6959; E-mail: [email protected]
429
430 nal serum and milk, was associated with decreased IQ scores, particularly those for memory and attention, and poorer reading ability at 11 years of age in this same cohort (16). Again there was no apparent effect of exposure through nursing on outcome. In a cohort of breast-fed infants followed prospectively for 60 months in a North Carolina study (10,11,37,38) maternal milk fat PCB levels were associated with hypotonicity and hyporeflexia on the Brazelton Neonatal Behavioral Assessment Scale. Higher maternal milk fat levels were associated with lower Bayley Scale scores at 6 and 12 months (10) and 24 months (35). There was no association between either prenatal or concurrent PCB levels and a negative outcome on the McCarthy Scale of Children’s Ability at 3–5 years (10). Behavioral impairment has been observed in monkeys exposed in utero and through breast milk until 4 months of age to either Aroclor 1248 or 1016. Aroclor 1016 exposure produced impairment on spatial discrimination reversal and facilitation on a shape discrimination reversal during infancy (40,41) with no effect on a spatial delayed alternation task at 4–6 years of age (25,40). Monkeys exposed developmentally to Aroclor 1248 exhibited impairment, no impairment, or facilitated performance on spatial and nonspatial discrimination problems as infants (2,41), and impairment on a spatial delayed alternation task at 4–6 years (40). Development exposure of rats to one of three individual ortho-substituted congeners (congeners 118, 153, or 28: 2,39,4,49,5; 2,29,4,49,5,59; 2,4,49) resulted in impairment on delayed alternation performance in females only, with no effect on radial arm maze performance (39). Exposure to a coplanar congener (congener 126: 3,39, 4,49,5), however, produced inconsistent effects on performance of a visual discrimination task (1,14). In a study assessing behavior in male rat offspring exposed to the commercial mixture Clophen A30, pre- and postnatal exposure produced effects on active avoidance and fixed interval performance (27). In a subsequent study, prenatal or pre- and postnatal exposure produced deficits on active avoidance and retention of a visual discrimination task, whereas postnatal exposure produced effects of smaller magnitude that did not reach statistical significance (26). An issue highly relevant to neurotoxicity produced by developmental PCB exposure in humans concerns the relative contribution of pre- vs. postnatal exposure to observed impairment. Results from the Michigan and North Carolina epidemiological studies found no association between any neurological or behavioral outcome and postnatal exposure as assessed by the amount of PCBs received through breast milk (16,20,21,35,38), despite the fact that infants receive a much greater amount of PCBs through breast milk than across the placenta (44). In most animal studies, exposure began in utero, making it impossible to differentiate prenatal from postnatal effects. The exception to this is the cross-fostering study by Lilienthal and Winneke (26), in which postnatal exposure had minimal effect, and certainly less effect than prenatal exposure. The current study extends an examination of the behavioral consequences of postnatal exposure of monkeys to a PCB congener mixture representative of PCBs found in human breast milk. These monkeys exhibited impairment in spatial delayed alternation performance, with some evidence for retarded acquisition of a nonspatial discrimination reversal task (31). The current study examined the performance of these monkeys under a multiple fixed interval (FI)–fixed ratio (FR) schedule of reinforcement. Performance under intermittent reinforcement schedules have proven sensitive to influence by exposure to toxicants (33); combining two schedules
RICE in an assessment procedure allows more than one behavioral baseline (which may be differentially sensitive to the effects of a toxicant) to be examined in the same test session. The FI was chosen in part because it proved sensitive to developmental exposure to PCBs in rats; the FR was included because of the marked difference in baseline performance between the FI and the FR (33), and because a multiple FI-FR schedule had been used previously with other neurotoxicants in monkeys (9,32,34). METHODS
Subjects, Dosing, and PCB Levels A total of 12 male monkeys (Macaca fascicularis) born in the Health Protection Branch, Health Canada breeding colony were separated from their mothers at birth and reared by hand as part of a study designed to determine the distribution of PCBs in infant monkeys (Arnold, Bryce, Mes, Hayward, Tryphonas, and Iverson, unpublished). They were dosed from birth to 20 weeks of age with 0 (four monkeys) or 7.5 mg/kg/ day (eight monkeys) of a mixture of PCBs representative of the pattern found in human breast milk (Table 1). The mixture represented 80% of the congeners present in breast milk in Canadian women, in a proportional mixture (29). To prepare the dosing solution, the congener mixture in isooctane was placed in a beaker and evaporated under a stream of N2 to “near dryness.” One hundred milliliters of freshly made Primalac infant monkey formula (Bio-serv, Frenchtown, N.J.) was placed in a beaker and mixed for 90 min using a Teflon stir bar and stir plate. The solution was divided into 10-ml aliquots and frozen until needed. Concentrations of the dosing solutions were 5 or 10 mg/ml such that the volume given to the monkey was no more than 0.5 ml at a time. The dose was drawn into a syringe and inserted directly into the infant’s mouth. Each daily dose was divided into three equal parts, and delivered at 0800, 1300, and 1500 h 7 days per week. Controls were dosed with Primalac vehicle. Infants were weighed daily, and doses adjusted accordingly. Blood was drawn at 20 weeks of age from all monkeys and analyzed for PCBs as described previously (30). Samples were
TABLE 1 CONGENER MIXTURE OF DOSING SOLUTION
PCB No.
Chlorine Substitution
52 66 74 105 118 138 153 156 157 180 183 187 189 194 203 15
2,29,5,59 2,39,4,49 2,4,49,5 2,3,39,4,49 2,39,4,49,5 2,29,3,4,49,59 2,29,4,49,5,59 2,3,39,4,49,5 2,3,39,4,49,59 2,29,3,4,49,5,59 2,29,3,4,49,59,6 2,29,3,49,5,59,6 2,3,39,4,49,5,59 2,29,3,39,4,49,5,59 2,29,3,4,49,5,59,6 Congener total
% in Canadian Milk
% in Monkey Mixture
1.0 2.0 8.8 3.0 10.3 14.8 15.1 3.3 1.0 10.3 1.8 3.8 0.3 2.0 1.8 79.3
1.5 2.9 10.4 4.4 12.8 17.5 18.6 4.7 1.5 12.8 2.3 4.7 0.5 2.9 2.3 99.8
PCBs AND MULTIPLE FI-FR IN MONKEYS
431
also taken from the nuchal fat pad (scapular area) at 20 weeks of age from all monkeys, and again between 90 and 99 weeks of age from two control and four treated monkeys. Analysis was restricted to the congeners administered to the monkeys; results will be published separately (Arnold, Bryce, Mes, Hayward, Tryphonas, and Iverson, unpublished). Total PCB levels are presented in Table 2. Congener 66 was excluded from the total for both blood and fat due to unexpected high spikes in some control samples. When these monkeys were 3 years of age, they were transferred to the Neurotoxicology Primate Colony for behavioral assessment. Monkeys had previously been tested on nonspatial discrimination reversal and delayed spatial alternation tasks. Testing for the current study began when monkeys were 4 years old. Behavioral Methods Schedule control and data acquisition were accomplished using the SKED system (State Systems, Kalamazoo, MI) run on a PDP 11/93 (Digital Equipment Corp., Maynard, MA). Data were collected as interevent times for all responses and schedule events, so that the session could be completely reconstructed from the raw data. Resolution was 0.01 s. Monkeys sat in a primate restraining chair inside a sound-attenuating cubicle, facing a Plexiglas panel on which was mounted a clear plastic push-button. Monkeys were tested under a multiple fixed interval (FI)–fixed ratio (FR) schedule of reinforcement for a total of 48 sessions. Each session consisted of four cycles of a FI followed by an FR; components were separated by 20-s time out (TO) period, during which responses had no scheduled consequences. The FI was signalled by a white light on the button, FR by a yellow light, and the TO by a dark button. Reinforcement was approximately 1.0 ml of apple juice. The FI value was 6 min for all sessions. The FR value was increased sequentially from 2 (sessions 1–5) to 5 (sessions 6–15), and finally 10 (sessions 16–48). Monkeys were tested 5 days per week, between approximately 0800 and 1300 h. The following variables were calculated: 1. Interresponse time (IRT), which was defined as the time between successive responses, or from the beginning of the
2. 3.
4.
5. 6.
schedule component to the first response, calculated separately for FI and FR mean and median. Pause time, defined as the time between the beginning of the schedule component and the first response, calculated for FI and FR mean and median. FI index of curvature (IOC) mean and median (8), which is a measure of the degree and direction of deviation from linearity of FI responding over the interval; intervals with fewer than eight responses were excluded from analysis. The standard deviation (SD) of the FI IRT distribution for the session divided by the FI IRT mean. The SD was normalized because it appeared to vary linearly with the absolute value of the mean IRT based on visual inspection of the data. The number of FI IRTs less than 5 s. The number of TO responses.
All measures except the number of TO responses were compared by orthogonal polynomial analyses (43) across sessions including the intercept, linear, and quadratic terms, using the SAS GLM procedure on a SUN computer. The parameters for the FR 2, 5, and 10 sessions were analyzed separately. All measures except IOC and FI pause underwent log transformation, which stabilized variance. Parameters were compared by t-tests. TO responses were compared by the Wilcoxin rank sums test. RESULTS
There were a number of differences in FI performance between control and PCB-exposed monkeys. Treated monkeys had shorter mean IRTs as measured by the intercept term of the orthogonal polynomial analysis ( p 5 0.024) (Fig. 1), as well as a difference in the development of performance across sessions as revealed by the quadratic term ( p 5 0.015). The difference between the control and treated group was most marked at the beginning of the experiment, and was less marked by the end. In contrast to the mean, the FI median IRT was only marginally significantly different ( p 5 0.093), suggesting that the IRT distribution may differ between groups. This was also indicated by a difference in the number of IRTs less than 5 s, which was greater in the PCB-treated monkeys as measured by the intercept term ( p 5 0.020) (Fig.
TABLE 2 BLOOD AND ADIPOSE PCB LEVELS (CONGENER 66 EXCLUDED) Blood (ppb Wet Weight)
Control
Treated
Adipose (ppb Lipid)
Week 20
Week 20
(Week)
0.37 0.30 nd* 0.32 1.84 2.52 2.19 nd* 1.85 2.58 2.75 2.84
198 62 50 53 3560 2368 1694 2716 2001 2520 2421 2369
211 (99)
*Below the detection limit of the analysis.
147 (94)
539 (98)
547 (94) 553 (92) 473 (90)
FIG. 1. Mean interresponse time (IRT) over the 48 FI sessions for control and PCB-treated monkeys. Group mean 6 SEM.
432
RICE
The observed differences between control and PCBtreated monkeys on the FI component of the multiple FI-FR
schedule may be interpreted as retarded acquisition. This is perhaps clearest for FI pause time, which increased more slowly over the course of the 48 sessions for the treated group compared to controls, but did not differ from control values over the last 10 sessions. A similar pattern was observed for the mean FI IRT, which was lower in treated monkey for most of the experiment, and increased to a value similar to that of controls near the end of the experiment. However, the greater number of short IRTs (less than 5 s) emitted by the treated monkeys did not decrease during later sessions, suggesting that there was a persistent (at least 48 session) difference between the performance of the control and PCBexposed groups. In fact, if anything this difference appeared to be getting larger as the experiment progressed, as control monkeys decreased their short IRTs across sessions and treated monkeys did not. The results of the current study are generally consistent with data from a study of developmental exposure to a commercial PCB mixture in rats exposed prenatally and continuing postnatally through testing at one year of age on an FI 30 s schedule of reinforcement (27). PCB-treated rats exhibited higher response rates, which is consistent with the shorter average IRT observed in the current study (response rate is the inverse of mean IRT). PCB-exposed rats also made a greater proportion of responses early in the interval compared to controls, which is consistent with the decreased pause time observed in the current study. FR performance was less sensitive to alteration by PCB exposure than was performance on the FI. This differential effect on FI and FR performance has also been observed following developmental exposure to lead in monkeys (32,34) and rats (3,4). In addition, the low FR response requirements in the current study were probably less sensitive to disruption by toxicant exposure than high FR values would have been. The monkeys in the current study had previously exhibited impaired performance on a delayed spatial alternation task (31); PCB exposure resulted in a deficit in acquisition of the task, and increased errors during early sessions of the experiment. In addition, PCB-treated monkeys made more perseverative errors even at the end of the experiment when the total error rate was not higher than that of controls. There were no delay-related effects. There was also some evidence that this group of monkeys displayed retarded acquisition of a nonspatial discrimination reversal task (31). Some PCBtreated individuals made more errors than controls over the
FIG. 2. The number of IRTs less than 5 s over the 48 FI sessions for control and PCB-treated monkeys. Group mean 6 SEM.
FIG. 3. The mean pause time over the 48 FI sessions for control and PCB-treated monkeys. Group mean 6 SEM.
2). Although the linear term was not different between groups, it appeared that the number of IRTs less than 5 s was decreasing for the control group over the last 20 sessions or so, whereas this was not true for the PCB-treated groups. However, the normalized SD of the IRT, which was a measure of response distribution, did not differ between groups ( p 5 0.291 for the intercept term of the orthogonal polynomial analysis). The FI pause time was greater for the control compared to the PCB-treated group ( p 5 0.034 for the intercept term of the orthogonal polynomial analysis) (Fig. 3). In addition, the increase in FI pause time typical of acquisition of FI performance developed more slowly in PCB-exposed monkeys, confirmed by a difference in the quadratic term ( p 5 0.045). In contrast, there was no evidence of an effect on IOC ( p 5 0.715 for the intercept). It was hypothesized that intervals with very low response rates (less than eight responses), which were included in the pause data but not the IOC calculation, might be responsible for the apparent difference. Comparison of the number of excluded intervals by t-test revealed that control monkeys had more sessions with excluded intervals (27.75 vs. 11.50, p 5 0.021) and more excluded intervals (58 vs. 19 out of a total of 192 intervals, p 5 0.022). This difference probably accounts for the apparent dichotomy in effect on FI pause time and IOC. FR performance was less affected by PCB exposure than was FI performance. There was no effect over the first five sessions, which consisted of FR 2. For the 10 FR 5 sessions, there was a difference in the quadratic term for both the IRT mean and median ( p 5 0.047 and p 5 0.037, respectively), but no difference in the average values. Visual inspection of the data revealed nothing remarkable over these 10 sessions. For the 33 FR 10 sessions, control monkeys decreased their mean pause time across sessions whereas PCB-treated monkeys did not, which was reflected in a difference in the linear term of the orthogonal polynomial analysis ( p 5 0.009). There were no other differences between control and PCB-exposed groups or FR performance. There was no difference between groups in the number of TO responses. DISCUSSION
PCBs AND MULTIPLE FI-FR IN MONKEYS
433
first several reversals at the beginning of the experiment, although the treated group as a whole was not different form controls. Combined with the results of the current study, these data suggest learning deficit, perseveration, and/or inability to inhibit inappropriate responding as a result of postnatal PCB exposure in monkeys. The fat levels in the current study were lower than in the previous monkey studies. Peak levels in the present experiment were 1.7–3.5 ppm at 20 weeks of age. In contrast, peak fat levels obtained at weaning at 4 months of age in the Aroclor 1016 study described above were 10 and 27 ppm for the low- and high-dose groups, respectively (40). The estimated peak levels at weaning for three infants whose mothers were exposed to 2.5 ppm of Aroclor 1248 during pregnancy and nursing were 21–123 ppm (2); levels from the 1.5 and 3 years postexposure groups were unavailable. Fat levels in the human population are usually reported to be between 1 and 6 ppm for unexposed populations [(12), review (23,28)], although a recent study from Germany reported average fat levels for nonoccupationally exposed adults of 300 ppb (22). The peak fat levels observed in the PCB-exposed monkeys in the current study (1.7–3.5 ppm) were in the middle of the range for environmentally exposed adult humans, whereas those of the control monkeys were lower than average reported human levels. Average whole blood PCB levels in the human general population range from 2–10 ppb [(7,22,30), reviews (23,44)], whereas levels in serum were 4 ppb based on five congeners (5). Hovinga et al. (15) re-
ported serum levels of 6 ppb in a control population and 15 ppb in a sample of fish eaters. In the North Carolina infant study, maternal serum levels averaged 9 ppb (37), whereas in the Michigan study maternal serum PCB levels were 5 ppb, with umbilical cord levels of 2.5 ppb. The whole blood PCB levels of exposed monkeys in the present study are toward the low end of results from the general human population, whereas those of the control monkeys were below most reported averages. It may be that the dose administered to the treated monkeys represents an approximation of typical intake by humans. In summary, monkeys were exposed from birth to 20 weeks of age to a low dose of a PCB mixture representative of the PCB congeners found in human breast milk. When tested at 4 years of age, they exhibited performance different from that of controls under a multiple FI-FR schedule of reinforcement. These data extend previous behavioral assessments in these monkeys and provide further evidence that exposure to an environmentally relevant level of a PCB mixture for a short time postnatally results in long-lasting behavioral effects. ACKNOWLEDGEMENTS
The author thanks Stephen Hayward of the Bureau of Biostatistics and Computer Applications for statistical analyses, and Virginia Liston, Bruce Martin, Wendy Cherry, Michelle Warankie, and Gary Vickers for expert technical assistance.
REFERENCES 1. Bernhoft, A.; Nafstad, I.; Engen, P.; Skaare, J. U.: Effects of preand postnatal exposure to 3,39,4,49,5-penthachlorobiphenyl on physical development, neurobehavior and xenobiotic metabolizing enzymes in rats. Environ. Toxicol. Chem. 13:1589–1597; 1994. 2. Bowman, R. E.; Hieronimus, M. P.; Allen, J. R.: Correlation of PCB body burden with behavioral toxicology in monkeys. Pharmacol. Biochem. Behav. 9:49–56; 1978. 3. Cory-Slectha, D. A.: Prolonged lead exposure and fixed ratio performance. Neurobehav. Toxicol. Teratol. 8:237–244; 1986. 4. Cory-Slechta, D. A.; Weiss, B.; Cox, C.: Delayed behavioral toxicity of lead with increasing exposure concentration. Toxicol. Appl. Pharmacol. 71:342–352; 1983. 5. Dar, E.; Kanarek, M. S.; Anderson, H. A.; Sonzogni, W. C.: Fish consumption and reproductive outcomes in Green Bay, Wisconsin. Environ. Res. 59:189–201; 1992. 6. Fein, G. G.; Jacobson, J. L.; Jacobson, S. W.; Schwartz, P. H.; Dowler, J. K.: Prenatal exposure to polychlorinated biphenyls: Effects on birth size and gestational age. J. Pediatr. 105:315–320; 1984. 7. Frank, R.; Braun, H. E.; Thorpe, B.: Comparison of DDE and PCB residues in the general diet and in human blood—Ontario 1986–87. Bull. Environ. Contam. Toxicol. 51:146–152; 1993. 8. Fry, W.; Kelleher, R.; Cook, L.: A mathematical index of performance on fixed-interval schedules of reinforcements. J. Exp. Anal. Behav. 3:193–199; 1960. 9. Gilbert, S. G.; Rice, D. C.; Burbacher, T. M.: Fixed interval-fixed ratio performance in adult monkeys exposed in utero to methylmercury. Neurotoxicol. Teratol. 18:539–546; 1996. 10. Gladen, B. C.; Rogan, W. J.: Effect of perinatal polychlorinated biphenyls and dichlorodiphenyl dichloroethene in later development. J. Pediatr. 119:58–63; 1991. 11. Gladen, B. C.; Rogan, W. J.; Hardy, P.; Thullen, J.; Tingelstad, J.; Tully, M.: Development after exposure to polychlorinated biphenyls and dichlorodiphenyl dichloroethene transplacentally and through human milk. J. Pediatr. 113:991–995; 1988. 12. Gómez-Catalán, J.; Lezaun, M.; To-Figueras, J.; Corbella, J.: Organochlorine residues in the adipose tissue of the population
13. 14.
15.
16. 17. 18.
19.
20. 21. 22.
of Navarra (Spain). Bull. Environ. Contam. Toxicol. 54:534–540; 1995. Harada, M.: Intrauterine poisoning: clinical and epidemiological studies of the problem. Bull. Instit. Constit. Med. (Kumamoto Univ.) 25 (suppl.): 1–60; 1976. Holene, E.; Nafstad, I.; Skaare, J. U.; Bernhoft, A.; Engen, P.; Lagvolden, T.: Behavioral effects of pre- and postnatal exposure to individual polychlorinated biphenyl congeners in rats. Environ. Toxicol. Chem. 14:967–976; 1995. Hovinga, M. E.; Sowers, M.; Humphreys, H. E.: Environmental exposure and lifestyle predictors of lead, cadmium, PCB, and DDT levels in Great Lakes fish eaters. Arch. Environ. Health. 48:98–104; 1993. Jacobson, J. L.; Jacobson, S. W.: Intellectual impairment in children exposed to polychlorinated biphenyls in utero. N. Engl. J. Med. 335:783–789; 1996. Jacobson, J. L.; Jacobson, S. W.; Humphrey, H. E. B.: Effects of exposure to PCBs and related compounds on growth and activity in children. Neurotoxicol. Teratol. 12:319–326; 1990. Jacobson, J. L.; Jacobson, S. W.; Humphrey, H. E. B.: Effects of in utero exposure to polychlorinated biphenyls (PCBs) and related contaminants on cognitive functioning in young children. J. Pediatr. 116:38–45; 1990. Jacobson, J. L.; Humphrey, H. E. B.; Jacobson, S. W.; Schantz, S. L.; Mullin, M. D.; Welch, R.: Determinants of polychlorinated biphenyls (PCBs), polybrominated biphenyls (PBBs), and dichlorodiphenyl trichloroethane (DDT) levels in the sera of young children. Am. J. Public Health 79:1401–1404; 1989. Jacobson, S. W.; Fein, G. G.; Jacobson, J. L.; Schwartz, P. M.; Dowler, J. K.: The effect of intrauterine PCB exposure on visual recognition memory. Child. Dev. 56:853–860; 1985. Jacobson, S. W.; Fein, G. G.; Schwartz, P. M.; Dowler, J. K.: Perinatal exposure to an environmental toxin: a test of multiple effects model. Dev. Physiol. 20:523–532; 1984. Kannan, N.; Schulz-Bull, D. E.; Petrick, G.; Duinker, J. C.; Macht-Hausmann, M.; Wasserman, O.: Toxic chlorobiphenyls in adipose tissue and whole blood of an occupationally/accidentally
434
23.
24. 25. 26. 27. 28.
29. 30.
31.
32. 33. 34.
RICE exposed man and the general population. Arch. Environ. Health 49:375–382; 1994. Kocan, A.; Petrik, J.; Drobná, B.; Chovancová, J.: Levels of PCBs and some organochlorine pesticides in the human population of selected areas of the Slovak Republic. Chemosphere 29:2315– 2325; 1994. Lai, T. J.; Guo, Y. L.; Yu, M. L.; Ko, H. C.; Hsu, C. C.: Cognitive development in Yucheng children. Chemosphere 29:2405–2411; 1994. Levin, E. D.; Schantz, S. L.; Bowman, R. E.: Delayed spatial alternation deficits resulting from perinatal PCB exposure in monkeys. Arch. Toxicol. 62:267–273; 1988. Lilienthal, H.; Winneke, G.: Sensitive periods for behavioral toxicity of polychlorinated biphenyls: Determination by cross-fostering in rats. Fundam. Appl. Toxicol. 17:368–375; 1991. Lilienthal, H.; Neuf, M.; Munoz, C.; Winneke, G.: Behavioral effects of pre plus postnatal exposure to a mixture of low chlorinated PCBs in rats. Fundam. Appl. Toxicol. 15:457–467; 1990. Loganathan, B. G.; Tanabe, S.; Hidaka, Y.; Kawano, M.; Hidaka, H.; Tatsukawa, R.: Temporal trends of persistent organochlorine residues in human adipose tissue from Japan, 1928–1985. Environ. Pollution. 81:31–39; 1993. Mes, J.; Marchand, L.: Comparison of some specific polychlorinated biphenyl isomers in human and monkey milk. Bull. Environ. Contam. Toxicol. 39:736–742; 1987. Mes, J.; Doyle, J.; Adams, B.; Davies, D.; Turton, D.: Polychlorinated biphenyls and organochlorine pesticides in milk and blood of Canadian women during lactation. Arch. Environ. Contam. Toxicol. 13:217–223; 1984. Rice, D. C.; Hayward, S.: Effects of postnatal exposure to a PCB mixture in monkeys on nonspatial discrimination reversal and delayed alternation performance. Neurotoxicology 18:479–494; 1997. Rice, D. C.: Lead exposure during different developmental periods produces different effects on FI performance in monkeys tested as juveniles and adults. Neurotoxicology 13:757–770; 1992. Rice, D. C.: Quantification of operant behavior. Toxicol. Lett. 43:361–379; 1988. Rice, D. C.: Schedule-controlled behavior in infant and juvenile
35. 36.
37.
38.
39.
40. 41.
42.
43. 44. 45.
monkeys exposed to lead from birth Neurotoxicology 9:75–88; 1988. Rogan, W. J.; Gladen, B. C.: PCBs, DDE, and children’s development at 18 and 24 months. Ann. Epidemiol. 1:407–413; 1991. Rogan, W. J.; Gladen, B. C.; Hung, K. L.; Koong, S. L.; Shia, L. Y.; Taylor, J. S.; Wu, Y. C.; Yang, D.; Ragan, N. B.; Hsu, C.C.: Congenital poisoning by polychlorinated biphenyls and their contaminants in Taiwan. Science 241:334–336; 1988. Rogan, W. J.; Gladen, B. C.; McKinney, J. D.; Carreras, N.; Hardy, P.; Thullen, J.; Tingestad, J.; Tylly, M.: Polychlorinated biphenyls (PCBs) and dichlorodiphenyl dichloroethene (DDE) in human milk: Effects of maternal factors and previous lactation. Am. J. Public Health 76:172–177; 1986. Rogan, W. J.; Gladen, B. C.; McKinney, J. D.; Carreras, N.; Hardy, P.; Thullen, J.; Tingelstad, J.; Tylly, M.: Neonatal effects of transplacental exposure to PCBs and DDE. J. Pediatr. 109: 335–341; 1986. Schantz, S. L.; Moshtaghian, J.; Ness, D. K.: Spatial learning deficits in adult rats exposed to ortho-substituted PCB congeners during gestation and lactation. Fundam. Appl. Toxicol. 26:117– 126; 1995. Schantz, S. L.; Levin, E. D.; Bowman, R. E.: Long-term neurobehavioral effects of perinatal polychlorinated biphenyl (PCB) exposure in monkeys. Environ. Toxicol. Chem. 10:747–756; 1991. Schantz, S. L.; Levin, E. D.; Bowman, R. E.: Hieronimus, M.; Laughlin, N. K.: Effects of perinatal PCB exposure on discrimination-reversal learning in monkeys. Neurotoxicol. Teratol. 11;243– 250; 1989. Schwartz, P. M.; Jacobson, S. W.; Fein, G. G.; Jacobson, J. L.; Price, H. A.: Lake Michigan fish consumption as a source of polychlorinated biphenyls in human cord serum, maternal serum, and milk. Am. J. Public Health 73:293–296; 1983. Snedecor, G. W.: Statistical methods. Ames: Iowa State University Press; 1980. WHO (World Health Organization). Environmental Health Criteria 140: Polychlorinated Biphenyls and Terphenyls. Geneva; 1993. Yu, M. L.; Hsu, C. C.; Guo, Y. L.; Lai, T. J.; Chen, S. J.; Luo, J. M.: Disordered behavior in the early-born Taiwan Yucheng Children. Chemosphere. 29:2413–2422; 1994.