- Email: [email protected]
Behavioral changes in mice following lead administration during several stages of development
Physiology &Behavior,Vol. 30, pp. 583-589. Pergamon Press Ltd., 1983. Printed in the U.S.A.
Behavioral Changes in Mice Following Lead Administration During Several Stages of Development I Z E L I G S. D O L I N S K Y , 2 R I C H A R D G. B U R R I G H T A N D P E T E R J. D O N O V I C K 3
Center for Neurobehavioral Sciences and Department o f Psychology State University o f New York at Binghamton, Binghamton, N Y 13901 R e c e i v e d 6 J a n u a r y 1982 DOLINSKY, Z. S., R. G. BURRIGHT AND P. J. DONOVICK. Behavioralchanges in micefollowing lead administration during several stages of development. PHYSIOL BEHAV 30(4) 583-589, 1983.--How lead ingestion during different developmental periods influences activity of mice was investigated. Binghamton Heterogenous Stock (HET) mice were assigned to one of four groups defined by a 2 × 2 factorial in which the only available drinking fluid was either lead (L) or water (C) from the time of mating to birth or from birth to the end of the experiment. Thus, a control group (CC) received water throughout, whereas the other three groups received a 0.5% lead acetate solution at the time they were mated (LC) or when pups were discovered (group CL), or throughout the experiment (group LL). The effects of these exposure regimes on activity were assessed in an open field or a running wheel when the mice were 25 days of age and then again when they were 55 days old. Aspects of agonistic behavior were also examined in these animals. Mice that received lead only following birth (CL group) appeared most affected in the open field and in running wheels. But, both the direction and degree of this effect were influenced by the specific test situation and measures as well as by the age of the mouse when tested. For example, the CL group crossed the most squares in the open field at both ages. However, the CL group was less active during their first day in the running wheels when 25 days old, but not when 55 days old. In general, activity of the LC group was least affected by administration of lead, but the effects of continued exposure to this toxic substance (LL group) were not simply additive. In contrast to measures of activity, agonistic testing at 60 days of age showed that all groups that had been or were being exposed to lead (CL, LC, LL) displayed a shorter latency to fight when compared to the control group (CC). Lead
Developmental periods
Activity
Aggression
WITH industrialization, lead in the earth's environment has increased dramatically. World production of lead currently exceeds 3.5 million metric tons per year, a value greater than that for any other heavy metal [18]. Because of its widespread presence in the environment and its known toxicity at high levels, there has been considerable concern with the possible behavioral effects of " l o w - l e v e l " lead exposure. In 1973, Silbergeld and Goldberg [21] observed that mice ingesting 0.5% lead acetate postnatally were more active than controls, and these researchers suggested that lead may play an etiological role in the attention deficit disorder previously termed childhood hyperkinesis [22]. However, since that time there have been conflicting results regarding the effects of lead on activity of rodents. Various investigators have reported hyperactivity [4, 11, 16, 20, 21, 22], hypoactivity [5, 12, 19], and no change in activity [4, 12, 21, 22] following exposure of animals to lead. While various methodological
HET Mice
differences across studies may contribute to the failure to obtain consistent results, the pattern of data from the reports listed above does suggest that lead administered after birth results in hyperactivity. However, systematic examination(s) of the differential impact of period of lead administration has not been extensive (but see [8]). The present study specifically examined the influence of the developmental period during which lead was administered on activity in mice. The dependent measures used to assess activity and the age at which we tested the mice are compatible with previous investigations. We chose to utilize open-field and running-wheel behavior as two measures which have been commonly used to assess activity. A cold water swim was also employed to determine the differential effect o f " stress" on subsequent open field behavior of our mice. Animals were tested twice, once when they were 25 days old and again when they were 55 days of age to determine the effect
~This research was supported in part by NSF (DAR 7911233) and BRSG grant awarded by the Division of Research Resources, NIH (5S07RR07149-04). zCurrent address: Department of Psychiatry, University of Connecticut Health Center, Farmington, CT 06032. 3Requests for reprints should be addressed to Peter J. Donovick, Department of Psychology, State University of New York, Binghamton, NY 13901.
Copyright © 1983 Pergamon Press Ltd.--0031-9384/83/040583-07503.00
584
DOLINSKY ET AL.
of lead on preadolescent and young adult mice. Thus, the present study systematically manipulates some of the potential influencing variables which have been represented in previous investigations. METHOD
Subjects were 98, male, Binghamton Heterogenous (HET) stock mice [9]. These animals were derived over a period of five months from 34 mating pairs in our laboratory. The pairs were randomly assigned to four treatment groups, two groups with eight mating pairs and two with nine pairs. These groups differed with respect to when, and if, lead (0.5% lead acetate) was presented in the drinking fluid of either the parents and/or their offspring. All mice tested in this experiment were killed after the last behavioral test when they were 60 days of age. The four group (2x2 factorial) assignments were as follows: (1) CC (control)---water (C) throughout the study; (2) LC (prenatal lead presentation)---i.e., the lead (L) solution was the only source of fluid for the mother (and father, but see below) from the time of mating until parturition, water was available thereafter; (3) CL (postnatal lead presentation)----mating pairs received water prior to birth of their litter, at which time a lead solution was presented as the only available fluid source for the duration of the experiment; LL (pre- and postnatal lead presentation)--breeder pairs received a lead solution as the only source of fluid at the time of mating, and exposure via the drinking water continued for the duration of the experiment. Obviously, it is difficult to separate pre- from postnatal lead exposure (as opposed to presentation) since a body burden of lead (cf, [12]) acquired by the dam during gestation will continue to contribute some lead to her milk following parturition, even if she is no longer exposed to this toxic contamination. Mating pairs and subsequent offspring to be tested were housed under standard laboratory conditions in a vivarium maintained at 21-22°C with white lights on between 8:00 a.m. and 8:00 p.m. All mice received Charles River Mouse Chow and their assigned drinking fluid ad lib. During the mating period fathers had access to the same drinking solution as the mothers; the fathers were removed from the cages when vaginal plugs were visible in the females. At parturition litters were culled to a maximum of six pups, retaining equal numbers of males and females whenever possible. Culled litters remained with their mothers until weaning at 21 days of age. The three male pups from each litter that were used in this study remained group housed throughout the experiment except during the running wheel test. All male offspring used in the study remained on the assigned drinking fluid throughout the experiment. Two males from each litter were tested twice in an open field when they were 25 and again when they were 55 days of age. There were 15 to 18 mice per treatment group, with representatives from 8 to 9 litters. Animals were placed in a 28 x 28 x 20 cm white Plexiglas box with the floor divided into a grid of twenty-five, 5.6 cm squares and sixteen, 1.2 cm diameter holes. The box itself rested on four 1.5 cm high legs. The open field box was covered with a clear, Plexiglas top through which animals were observed. Open field tests were carried out between 0800 and 1200 hr in the same room in which animals were housed; the number of squares crossed and stand ups, as well as the number of times mice poked their noses through the holes in the bottom of the field ("hole/nose pokes") were recorded for a five minute period.
At each testing age, immediately following this first open field testing, one of the two animals tested in the open field was placed in a dry basin and the other mouse was placed in a similar basin that contained 17°C water. This 60x60x35 cm basin was divided into 36, 7 cm squares. Mice were observed in the basin for one minute and the number of squares through which they ran or swam was recorded. Fourteen minutes following the one minute basin observation mice from both the wet and dry conditions were returned to the original open field for a second open field test. The conditions of the second test were identical to those during the first exposure to the open field. This second open field test, at each testing age, was used to assess the differential effects of the exposure to a dry versus wet basin on activity. This second test provided information concerning how age at the time of testing and exposure to lead influenced subsequent activity in an open field as a function of the intervening, presumably differentially stressful experience in the basin. The remaining (third) male from each litter was tested at both 25-30 and 55-60 days of age in a running wheel. These animals continued drinking their assigned solution during their six day stay in the testing cages with available wheels. At 0800 hr each of these mice was removed from their home cage and litter mates and placed singly into a wire cage (26x 18x 17 cm) with an attached running wheel (16.5 cm dia. x 6.9 cm width) which was always available through a 4 cm (dia.) hole at the back of the living cage. Wheel revolutions were monitored via mechanical counters for a total of six days. Revolutions were recorded for the first three hours after both white light onset and offset, as well as the remaining nine hours in each of the light and dark periods. Mice remained in the test cages with attached wheels for the entire six day period during which they received ground Charles River Mouse Chow and their assigned fluid (lead or water) ad lib. The mice used in the running wheel test were isolated from their littermates for six days; thus two measures of agonistic behavior [9] were observed between the three littermates when the animal from the running wheel was returned to its original cage. Latency to initiate fighting, as well as the number of such bouts among members of the triad, were observed for a five-minute period immediately following return of the mouse tested in the running wheel to its home cage. These agonistic observations were conducted when the mice were both 30 and 60 clays of age. Data Analysis In general, data from each of the dependent measures were analyzed separately using an unweighted means analysis of variance (ANOVA) with period of lead presentation (CC, CL, LC, LL) as a between subjects factor and age (25, 55) a within subjects factor. When appropriate, additional factors (i.e., dry vs. wet basin tests) also were included. Note that animals had differential experience after the first open field test at age 25 because half of them subsequently received a dry and half a wet basin test. Thus, any group differences in the data obtained during the first openfield test when mice were 55 days of age may have been influenced by differential group response(s) to the previous basin experience. To assess this possibility, a basin variable (dry, wet) was included as a between subjects factor in the statistical analyses of the open field data. This overall ANOVA, including basin condition, failed to reveal any significant interaction involving basin condition on any of the
LEAD, D E V E L O P M E N T A L PERIOD AND BEHAVIOR dependent measures concerning the first open field test at each age. Therefore, data from both wet and dry conditions were combined for the first open field measures obtained when mice were 25 or 55 days of age. Data from two animals in the CL group, and one from the LL group were not included because of apparent illness of these mice. To assess the relative influence of the wet vs. dry basin experience on the re-exposure to the open field at each age, the following ratio (2nd open field - initial open field/2nd open field + initial open field) was employed. A negative ratio score would reflect a decline, a positive score an increase, and a zero no change in relative activity from the first to second test in the open field on a given day of testing. Because variability was considerably greater in the running wheels when animals were tested at 55--60 days of age as compared to when they were 25-30 days old, separate analyses were performed at each age. Furthermore, day one activity was analyzed alone to assess the initial reaction of the mice to being placed individually in the living cages with attached running wheels. In addition, data from days 2-6 were analyzed (with days as a repeated, within subjects measure) to assess the development of wheel-running activity. Mechanical problems occasionally were encountered with the wheels, and in such cases, missing data were replaced with the appropriate mean from the remaining members of the group. This replacement procedure was employed only in the repeated measures analyses, and involved less than 2% of the data points. We deemed this procedure preferable to dropping all of the data for a given subject if any single data point was missing. When the mice were 25 days old returning the individual from the running wheel to his home cage and littermates resulted in little or no fighting among the social triad. However, agonistic behavior clearly did occur when the mice were 60 days old; thus, only results of agonistic testing at 60 days of age are presented below. RESULTS
Body Weight As can be seen in Table i, all groups of mice increased their body weight from 25 to 55 days of age, as expected, Age: F(1,88)=3989.55, p<0.001. However, at 25 days of age those mice that received either lead only postnatally (CL), or both pre- and postnatally (LL), were comparable in weight and lighter than those whose mothers drank the lead solution only prior to giving birth (LC) or had not been exposed to the lead solution at all (CC). These group differences were still present, but were not as marked at 55 days of age, Group x Age: F(3,88)=9. l 1, p<0.001.
585 TABLE 1 MEAN BODY WEIGHT (g) -+SEM
Age (Days) Group CC CL LC LL
25 16.0 -+ 0.3 12.5 - 0.4 15.4 -+ 0.4 11.4 _+ 0.3
55 27.5 _+ 0.4 26.3 _+ 0.5 27.4 _+ 0.4 25.8 _+ 0.4
difference (Duncan's p<0.05) was between CL (most squares crossed) and LL (fewest crossed) groups, but at 55 days of age the CL group crossed significantly more squares than either of the other lead exposed groups (LL or LC). Hole pokes. Figure 1B presents nose-poke data in the first open field test conducted when the mice were 25 days old and subsequently when they were 55 days of age. Mice that received lead postnatally only (CL) poked the most holes when tested at 25 days of age, while the other three groups were comparable. However, when the mice were 55 days old those either receiving lead during the postnatal period (CL), or both prior to and following birth (LL), explored more holes than either the CC or the LC groups, Group x Age: F(3,59)=4.19, p<0.01. In the context of this interaction it should be noted that all groups except the LL group showed a substantial decrease in the number of holes poked from 25 to 55 days of age. The LL group actually increased their exploration of holes somewhat between the two tests. Stand ups. Similar to the data previously described, all groups increased the number of times they stood up during the first open field test from when they were 25 days of age to the first open field test when they were 55 days of age, Age: F(1,59)=107.66, p<0.001 (see Fig. 1C). In addition the LL group stood up less than any of the other groups, particularly when they were 55 days old, Group: F(3,59)=2.74, p =0.05.
Basin Test Squares crossed. Generally, mice tested in the 17°C water swam across more squares than littermates ran across when tested in the same basin without water, Basin: F(1,58)=208.59, p<0.001 at ages 25 and 55 days. However, there were no significant effects involving either age or group. In addition, while we did not attempt to quantify aspects of swimming performance (cf, [1]), there did not appear to be any gross abnormalities in the swimming ability of mice exposed to lead as compared to control animals.
Open Field~First Test at Each Age Squares crossed. All groups increased the number of squares crossed during the first open field test (Fig. 1A) from 25 to 55 days of age, Age: F(1,59)=93.01, p<0.001. Furthermore, across ages, the CL group crossed more squares than any of the other three groups which were comparable to each other, Group: F(3,59)=3.60, p<0.02. Although the overall Group x Age interaction did not reach statistical significance, a Duncan-multiple-range test was performed at each age to assess the apparently reduced slope in the squares crossed data of the LC group from age 25 to 55 days of age relative to the other groups. These specific contrasts indicated that at 25 days of age, the only statistically significant
Open Field: Pre-Basin vs. Post-Basin Test at Each Age Hole poke ratios. As reflected in the hole-poke ratio described in the data analyses section above, mice tended to decrease the number of holes they nosed in the second (post-basin) test relative to the first (pre-basin) test at both ages of testing; that is, all group mean ratio scores are negative or near zero (see Fig. 2). However, there were differences between groups in the degree to which these ratio scores were altered as a function of both age and the intervening basin condition, Group × Age x Basin: F(3,57)=3.70, p<0.02. Following testing in the dry basin (left panel, Fig. 2) this
DOLINSKY ET AL.
586
F I R S T OPEN F I E L D A. SQUARE CROSSINGS I CL
30C
B. NOSE POKES
C. STAND UPS
70
~
40 I
I\
~E
CC
26( Q Z I
/
/
I
t
3E
18(
~
30
JLC
•
50
20
"'~'-~CC
,o
30
14C I
I
25
55
eS
LL
",,N~
;.;:/
~
,,,,,"T L c
T"".,,
/
Z Z
/,
\
LL
/
C
#
I II
"
nr aD 22C
60
L
I( 1
I
I
I
I
25
55
25
55
DAYS OF AGE FIG. 1. First Open Field: Mean number (_SEM) of square crossings (panel A), nose pokes (panel B), and standups (panel C) for each of the four groups when tested prior to wet or dry basin exposure at 25 and again at 55 days of age,
ratio increased toward zero (a zero means there was no difference in relative numbers of hole pokes on first and second tests) for both the CC and LC groups from testing at 25 to testing at 55 days of age. But this trend was reversed for those mice still drinking lead (CL and LL groups). However, following wet basin exposure (right panel, Fig. 2) all groups, regardless of lead history, tended to show an increase toward zero in terms of this hole poke ratio score from testing at age 25 to 55 days; here however, that increase was not as marked for the LC group. Ratios for squares crossed and stand ups. In the dry basin condition at 25 days of age all groups tended to cross more squares in the second test in the open field relative to the initial open field test. This effect was attenuated when mice were retested at 55 days of age. The trend observed in the dry basin was reversed from that of the wet basin condition where the mice at 25 days of age crossed fewer squares in the second open field test relative to the first exposure. This decrease in activity was less apparent when mice were retested when they were 55 days of age, Age × Basin: F(1,57)=26,48, p<0.001. A similar trend was observed for stand ups, Age × Basin: F(1,57)= 11.80, p<0.01. There were
no statistically significant effects involving groups with respect to the influence of the intervening basin condition on the ratios for either squares crossed or the number of times mice stood up in the open field. Thus, in contrast to the hole poke ratios, the squares crossed and stand up ratios suggest that mice exposed to lead during several developmental periods are not differentially responsive to the presumed disparity in stress of the two basin conditions at either testing age.
Running Wheel Activity Trends in the running wheel data for the 12-hour light and dark periods across the six days of running wheel testing were similar to those for data from the first three hours after light onset and offset. However, there was somewhat less variability, and the more immediate responses to change in illumination were better reflected in the data from the three hour periods. Thus, while providing a similar picture to that seen when each of the entire 12 hr periods were examined, data from the shorter, three-hour periods are presented. Thus, only means for each of the three hour periods follow-
L E A D , D E V E L O P M E N T A L PERIOD A N D B E H A V I O R
587
OPEN F I E L D HOLE POKE CHANGE RATIO DRY BASIN BETWEEN ¢,i c~ Z ,,¢
WET BASIN BETWEEN OPEN FIELD TESTS I 81 2
OPEN FIELD TESTS I 8, 2 0,1
ILl nO.. "1I-U') 0
(+i
0
(-)
J
~-0°1
P
W
S
J
/
o.
i
LC
P
I
I-- - 0 . 2
~
-0.3
IE
.~ - 0 , 4
25
55
25
55
DAYS OF AGE FIG. 2. Open Field Hole Poke: Ratios (2rid O . F . - l s t O.F./2nd O.F.+lst O.F.) for mice from each of the four groups when tested at 25 and again at 55 days of age. Either wet or dry basin exposure occurred between the 1st and 2nd open field test at each age.
ing light onset (a.m.) and offset (p.m.) across the six days of testing in the wheel available cages are presented (Fig. 3). During the first three hours of the light and dark phases of the first day, the 25-30 day old CL mice were less active in the wheels than any other group of comparable age (top panels, Fig. 3), Group: F(3,27)=4.06, p<0.02. All groups increased activity in the dark period across the subsequent five days. There were no reliable differences between groups with respect to either day one activity or the development of activity across the remaining five days when the mice were retested at 55-60 days of age (bottom panels, Fig. 3).
Agonistic Behavior As might be expected in preadolescent male mice, there was relatively little fighting among littermates after the 30-day old, wheel-tested animals were returned to their home cages. However, as shown in Fig. 4 when these males were 60 days old and then reintroduced to their home cages (after their second, 6 day running wheel test), all groups with any lead-exposure history (LC, CL, LL) were comparable and had markedly shorter mean latencies to initiate fighting than the CC mice with no history of exposure to lead, Group: F(3,27)=4.68, p<0.01; Duncan's, p<0.05. No differences however, were observed among the groups in terms of the number of bouts of fighting during the five minute observation period following return of the 60 day old animals to their home cage.
DISCUSSION
There has been a lack of consensus concerning the effects of lead on activity levels in rodents. In part, this may b e due to differences in experimental protocols utilized. Our present results demonstrate the complexity of the task before us. Factors such as developmental period when lead is delivered, age at testing, and specific test(s) employed influence experimental findings. However, data from our aggression testing at 60 days of age suggest that lead, regardless of when it was presented, may alter some aspects of social behavior. It thus appears that changes in agonistic behavior may be associated with lead ingestion [3, 12, 20]; however, the relationship is undoubtedly complex. That is, when triads of lead treated mice were observed after a period of isolation, their latency to aggress was reduced. Interestingly, Finch and Reiter [7] found decreased aggression when young lead treated rats were matched with same aged control animals. The latter findings are congruent with our observation that young lead treated males are submissive to control mice; this pattern was reversed in older pairs [3]. In the present study, lead caused a comparable decrease in body weight in both groups receiving lead postnatally (CL, LL). It has been suggested [17] that the effects of lead on activity levels are influenced by body weight changes associated with, but not unique to, lead exposure. However, many of the measures of activity reflected differences between these two groups. F o r example, during the first open field test, mice that were presented lead only postnatally
588
DOLINSKY ET A L .
CC
1400
1200
f
LC
CL
LL
_..oA M •---- PM
I000
I AGE 25 -301
800
f
600
O3
Z 0
400
\
.L \
200
\
I
Io.~O-'O-"O,,
b-.o--'°- °
"
T-
I
,ll.
0--"0
0~.
0 - -0~
I
•
|
I
i
1500 1400
Z LLI
1200 I000
~o"
%
0
.J
W
0
U,t
"I
-i
1
800
.t I I t
600
t
400
I
I
! I 'I
T
!
o
1
I
0 ~"
200
,I
,IJ.
I ~0
~.
"0..0o0 "O
~O--o •
|
•
•
2 3 4 5 6
•
•
•
•
0--0
|
23456
23456
23456
DAYS FIG. 3. Running Wheel: Mean number of revolutions and average SEM, (vertical bars) during the first 3 hours after light onset (AM, dashed lines) and after light offset (PM solid lines) on each of six consecutive days for mice from each of the four groups (not tested in the open field) when they were 25-30 days of age (top four panels) and when they were 55-60 days of age (bottom four panels).
(CL) crossed more squares than mice which had received lead both before and after birth (LL). In addition, activity levels were not similar in these two groups on day one in the running wheel when the mice were 25-30 days old. Thus, while one of the effects of postnatal lead administration was to decrease body weight, weight alone does not appear to predict changes in either the open field or running wheel behavior. Previously we reported a similar failure of bodyweight to predict the impact of lead on seizure susceptibility of several stocks of 21 day old mice [2]. Not surprisingly, the specific test employed in the present
investigation differentially reflected our manipulations (see also [15,24]). For instance, 25 day old mice exposed to lead postnatally only (CL) made relatively few revolutions in the running wheels, but were highly active in the open field at this age. These test conditions apparently are assessing different facets of behavior and lead toxicity. The development of a more unified schema of the impact of this toxic element on behavior may call for new approaches such as those which have been utilized in the analysis of changes in behavior following specific brain lesions [6]. Silbergeld and Goldberg [21,22] reported increased hori-
LEAD, DEVELOPMENTAL
l
E
200
T
POST RUNNINGWHEEL AGE 60 AGRESSION- 5 MINUTE TEST
CC
mm
I00
o
z
50
.J
Ix
PERIOD AND BEHAVIOR
CL LC LL FIG. 4. Agonistic Behavior: Mean latency to first fight (seconds) during the five minute aggression test for 60 day old mice in each of the four groups after the mice were tested in the wheel-available cages and were returned to their home cage which contained two littermates.
589 suggested an increase in activity for mice who had r e c e i v e d lead from birth to weaning but w e r e then returned to water. The latter findings, h o w e v e r , were not consistent. It is interesting to note that in the present study, squares crossed during the first open field test at both ages also suggested that mice that r e c e i v e d lead postnatally only (CL) m o v e d across more squares than any o f the o t h e r groups. Finally, our results suggest that the nature and severity o f lead's behavioral effects are not simply or monotonically related to the total duration o f e x p o s u r e , but are critically related to the develo p m e n t a l period(s) during which this toxic element is present and the behavioral task e m p l o y e d [15,24]. Thus in several instances perinatal (LC) and postnatal alone (CL) lead exposure yield opposite effects on behavior, while the effects o f lead administration during both periods (LL) is not explained by an additive m o d e l based on perinatal and postnatal exposure. The underlying m e c h a n i s m s for these results warrants further investigation.
ACKNOWLEDGEMENTS zontal activity, as d e t e c t e d by activity monitors, in 40-60 day old mice that had r e c e i v e d lead from birth until the time o f testing (similar to our C L group). T h e i r data also
We thank Wayne Kashinsky for his help with the word processing system and Dr. Richard Pastore who allowed us access to his printer.
REFERENCES
1. Adams, P. M., J. D. Fabricant and M. S. Legator. Cyclophosphamide-induced spermatogenic effects detected in the Ft generation by behavioral testing. Science 211: 80-82, 1981. 2. Burright, R. G., P. J. Donovick, K. Micheis, R. J. Fanelli and Z. Dolinsky. Effect of amphetamine and cocaine on seizure in lead treated mice. Pharmacol Biochem Behav 16: 631-635, 1982. 3. Burright, R. G,, W. J. Engellenner and P. J. Donovick. Lead differentially affects agonistic behavior of adult mice of two ages. Physiol Behav 30: 285-288, 1983. 4. Dolinsky, Z., E. Fink, R. G. Burright and P. J. Donovick. The effect of lead, d-amphetamine, time of day on activity levels in the mouse. Pharmacol Biochem Behav 14: 877-880, 1981. 5. Driscoll, J. W. and S. E. Stegner. Behavioral effects of chronic lead ingestion on laboratory rats. Pharmacol Biochem Behav 4: 411-417, 1976. 6. Fanelli, R. J., R. G. Burright and P. J. Donovick. A multivariate approach to the analysis of genetic and septal lesion effects on maze performance in mice. Behav Neurosci, 1983, in press. 7. Finch, A. A. and L. W. Reitter. Alterations in social behavior in the rat during chronic low-level exposure to lead and tritum. Toxicol Appl Pharmacol 37: 160, 1976. 8. Fiynn, J. C., E. R. Flynn and J. H. Patton. Effects of pre- and postnatal lead on affective behavior and learning in the rat. Neurobehav Toxicol l: Suppl l, 93-103, 1979. 9. Fuller, J. L. Fuller BWS lines: History and results. In: Development and Evolution o f Brain Size, edited by M. E. Hahn, C. Jensen and B. C. Dudek. New York: Academic Press, 1979. 10. Fuller, J. L. and M. E. Hahn. Issues in the genetics of social behavior. Behav Genetics 6: 391-406, 1976. I I. Gotler, M. and A. L. Michaelson. Growth, behavior and brain catecholamines in lead exposed neonatal rats: A reappraisal. Science 187: 359-360, 1975. 12. Hastings, L., G. P. Cooper, R. L. Boruschein and A. Michaelson. Behavioral effect of low level neonatal lead exposure. Pharmacol Biochem Behav 7: 37--42, 1977.
13. Hejtmancik, M. R. Jr., E. B. Dawson and B. J. Williams. Tissue distribution of lead in rat pups nourished by lead-poisoned mothers. J Toxicol Environ Health 9: 77-86, 1982. 14. Kostas, J. D., D. J. McFarland and W. G. Drew. Lead-induced behavioral disorders in the rat: Effects of amphetamine. Pharmacology 16: 226--236, 1978. 15. Mullenix, P. Effect of lead on spontaneous behavior. In: Low Level Lead Exposure: The Clinical Implications o f Current Research, edited by H. L. Needleman. New York: Raven Press, 1980. 16. Overman, S. R. Behavioral effects of asymptomatic lead exposure during neonatal development in rats. Toxicol Appl Pharmacol 41: 459-471, 1977. 17. Rafales, L. S., R. L. Bornschein, I. A. Michaelson, R. K. Loch and G. F. Barker. Drug induced activity in lead-exposed mice. Pharmacol Biochem Behav 10: 95--101, 1979. 18. Ratcliffe, J. M. Lead in Man and the Environment. New York: Ellis Horwood Limited, 1981. 19. Reiter, L. W., G. E. Anderson, J. W. Laskey and D. F. Cahill. Developmental and behavioral changes in the rat during chronic exposure to lead. Environ Health Perspect 12:119-123, 1975. 20. Sauerhoff, M. W. and A. I. Michaelson. Hyperactivity and brain catecholamines in lead exposed developing rats. Science 182: 1022-1024, 1973. 21. Silbergeld, E. K. and A. M. Goldberg. A lead induced behavior disorder. Life Sci 13: 1275-1282, 1973. 22. Silbergeld, E. K. and A, M. Goldberg. Lead induced behavioral dysfunction: An animal model of hyperactivity. Exp Neurol 43: 146-147, 1974. 23, Sobotka, T. J. and M. P. Cook. Postnatal lead acetate exposure in rats: Possible relationship to minimal brain dysfunction. A m J Merit Defic 79: 5-9, 1974. 24. Zimering, R. T., R. G. Burright and P. J. Donovick. Effects of pre-natal and continued lead exposure on activity in the mouse. Neurobehav Toxicol Teratol 4: 9-14, 1982.
Behavioral Changes in Mice Following Lead Administration During Several Stages of Development I Z E L I G S. D O L I N S K Y , 2 R I C H A R D G. B U R R I G H T A N D P E T E R J. D O N O V I C K 3
Center for Neurobehavioral Sciences and Department o f Psychology State University o f New York at Binghamton, Binghamton, N Y 13901 R e c e i v e d 6 J a n u a r y 1982 DOLINSKY, Z. S., R. G. BURRIGHT AND P. J. DONOVICK. Behavioralchanges in micefollowing lead administration during several stages of development. PHYSIOL BEHAV 30(4) 583-589, 1983.--How lead ingestion during different developmental periods influences activity of mice was investigated. Binghamton Heterogenous Stock (HET) mice were assigned to one of four groups defined by a 2 × 2 factorial in which the only available drinking fluid was either lead (L) or water (C) from the time of mating to birth or from birth to the end of the experiment. Thus, a control group (CC) received water throughout, whereas the other three groups received a 0.5% lead acetate solution at the time they were mated (LC) or when pups were discovered (group CL), or throughout the experiment (group LL). The effects of these exposure regimes on activity were assessed in an open field or a running wheel when the mice were 25 days of age and then again when they were 55 days old. Aspects of agonistic behavior were also examined in these animals. Mice that received lead only following birth (CL group) appeared most affected in the open field and in running wheels. But, both the direction and degree of this effect were influenced by the specific test situation and measures as well as by the age of the mouse when tested. For example, the CL group crossed the most squares in the open field at both ages. However, the CL group was less active during their first day in the running wheels when 25 days old, but not when 55 days old. In general, activity of the LC group was least affected by administration of lead, but the effects of continued exposure to this toxic substance (LL group) were not simply additive. In contrast to measures of activity, agonistic testing at 60 days of age showed that all groups that had been or were being exposed to lead (CL, LC, LL) displayed a shorter latency to fight when compared to the control group (CC). Lead
Developmental periods
Activity
Aggression
WITH industrialization, lead in the earth's environment has increased dramatically. World production of lead currently exceeds 3.5 million metric tons per year, a value greater than that for any other heavy metal [18]. Because of its widespread presence in the environment and its known toxicity at high levels, there has been considerable concern with the possible behavioral effects of " l o w - l e v e l " lead exposure. In 1973, Silbergeld and Goldberg [21] observed that mice ingesting 0.5% lead acetate postnatally were more active than controls, and these researchers suggested that lead may play an etiological role in the attention deficit disorder previously termed childhood hyperkinesis [22]. However, since that time there have been conflicting results regarding the effects of lead on activity of rodents. Various investigators have reported hyperactivity [4, 11, 16, 20, 21, 22], hypoactivity [5, 12, 19], and no change in activity [4, 12, 21, 22] following exposure of animals to lead. While various methodological
HET Mice
differences across studies may contribute to the failure to obtain consistent results, the pattern of data from the reports listed above does suggest that lead administered after birth results in hyperactivity. However, systematic examination(s) of the differential impact of period of lead administration has not been extensive (but see [8]). The present study specifically examined the influence of the developmental period during which lead was administered on activity in mice. The dependent measures used to assess activity and the age at which we tested the mice are compatible with previous investigations. We chose to utilize open-field and running-wheel behavior as two measures which have been commonly used to assess activity. A cold water swim was also employed to determine the differential effect o f " stress" on subsequent open field behavior of our mice. Animals were tested twice, once when they were 25 days old and again when they were 55 days of age to determine the effect
~This research was supported in part by NSF (DAR 7911233) and BRSG grant awarded by the Division of Research Resources, NIH (5S07RR07149-04). zCurrent address: Department of Psychiatry, University of Connecticut Health Center, Farmington, CT 06032. 3Requests for reprints should be addressed to Peter J. Donovick, Department of Psychology, State University of New York, Binghamton, NY 13901.
Copyright © 1983 Pergamon Press Ltd.--0031-9384/83/040583-07503.00
584
DOLINSKY ET AL.
of lead on preadolescent and young adult mice. Thus, the present study systematically manipulates some of the potential influencing variables which have been represented in previous investigations. METHOD
Subjects were 98, male, Binghamton Heterogenous (HET) stock mice [9]. These animals were derived over a period of five months from 34 mating pairs in our laboratory. The pairs were randomly assigned to four treatment groups, two groups with eight mating pairs and two with nine pairs. These groups differed with respect to when, and if, lead (0.5% lead acetate) was presented in the drinking fluid of either the parents and/or their offspring. All mice tested in this experiment were killed after the last behavioral test when they were 60 days of age. The four group (2x2 factorial) assignments were as follows: (1) CC (control)---water (C) throughout the study; (2) LC (prenatal lead presentation)---i.e., the lead (L) solution was the only source of fluid for the mother (and father, but see below) from the time of mating until parturition, water was available thereafter; (3) CL (postnatal lead presentation)----mating pairs received water prior to birth of their litter, at which time a lead solution was presented as the only available fluid source for the duration of the experiment; LL (pre- and postnatal lead presentation)--breeder pairs received a lead solution as the only source of fluid at the time of mating, and exposure via the drinking water continued for the duration of the experiment. Obviously, it is difficult to separate pre- from postnatal lead exposure (as opposed to presentation) since a body burden of lead (cf, [12]) acquired by the dam during gestation will continue to contribute some lead to her milk following parturition, even if she is no longer exposed to this toxic contamination. Mating pairs and subsequent offspring to be tested were housed under standard laboratory conditions in a vivarium maintained at 21-22°C with white lights on between 8:00 a.m. and 8:00 p.m. All mice received Charles River Mouse Chow and their assigned drinking fluid ad lib. During the mating period fathers had access to the same drinking solution as the mothers; the fathers were removed from the cages when vaginal plugs were visible in the females. At parturition litters were culled to a maximum of six pups, retaining equal numbers of males and females whenever possible. Culled litters remained with their mothers until weaning at 21 days of age. The three male pups from each litter that were used in this study remained group housed throughout the experiment except during the running wheel test. All male offspring used in the study remained on the assigned drinking fluid throughout the experiment. Two males from each litter were tested twice in an open field when they were 25 and again when they were 55 days of age. There were 15 to 18 mice per treatment group, with representatives from 8 to 9 litters. Animals were placed in a 28 x 28 x 20 cm white Plexiglas box with the floor divided into a grid of twenty-five, 5.6 cm squares and sixteen, 1.2 cm diameter holes. The box itself rested on four 1.5 cm high legs. The open field box was covered with a clear, Plexiglas top through which animals were observed. Open field tests were carried out between 0800 and 1200 hr in the same room in which animals were housed; the number of squares crossed and stand ups, as well as the number of times mice poked their noses through the holes in the bottom of the field ("hole/nose pokes") were recorded for a five minute period.
At each testing age, immediately following this first open field testing, one of the two animals tested in the open field was placed in a dry basin and the other mouse was placed in a similar basin that contained 17°C water. This 60x60x35 cm basin was divided into 36, 7 cm squares. Mice were observed in the basin for one minute and the number of squares through which they ran or swam was recorded. Fourteen minutes following the one minute basin observation mice from both the wet and dry conditions were returned to the original open field for a second open field test. The conditions of the second test were identical to those during the first exposure to the open field. This second open field test, at each testing age, was used to assess the differential effects of the exposure to a dry versus wet basin on activity. This second test provided information concerning how age at the time of testing and exposure to lead influenced subsequent activity in an open field as a function of the intervening, presumably differentially stressful experience in the basin. The remaining (third) male from each litter was tested at both 25-30 and 55-60 days of age in a running wheel. These animals continued drinking their assigned solution during their six day stay in the testing cages with available wheels. At 0800 hr each of these mice was removed from their home cage and litter mates and placed singly into a wire cage (26x 18x 17 cm) with an attached running wheel (16.5 cm dia. x 6.9 cm width) which was always available through a 4 cm (dia.) hole at the back of the living cage. Wheel revolutions were monitored via mechanical counters for a total of six days. Revolutions were recorded for the first three hours after both white light onset and offset, as well as the remaining nine hours in each of the light and dark periods. Mice remained in the test cages with attached wheels for the entire six day period during which they received ground Charles River Mouse Chow and their assigned fluid (lead or water) ad lib. The mice used in the running wheel test were isolated from their littermates for six days; thus two measures of agonistic behavior [9] were observed between the three littermates when the animal from the running wheel was returned to its original cage. Latency to initiate fighting, as well as the number of such bouts among members of the triad, were observed for a five-minute period immediately following return of the mouse tested in the running wheel to its home cage. These agonistic observations were conducted when the mice were both 30 and 60 clays of age. Data Analysis In general, data from each of the dependent measures were analyzed separately using an unweighted means analysis of variance (ANOVA) with period of lead presentation (CC, CL, LC, LL) as a between subjects factor and age (25, 55) a within subjects factor. When appropriate, additional factors (i.e., dry vs. wet basin tests) also were included. Note that animals had differential experience after the first open field test at age 25 because half of them subsequently received a dry and half a wet basin test. Thus, any group differences in the data obtained during the first openfield test when mice were 55 days of age may have been influenced by differential group response(s) to the previous basin experience. To assess this possibility, a basin variable (dry, wet) was included as a between subjects factor in the statistical analyses of the open field data. This overall ANOVA, including basin condition, failed to reveal any significant interaction involving basin condition on any of the
LEAD, D E V E L O P M E N T A L PERIOD AND BEHAVIOR dependent measures concerning the first open field test at each age. Therefore, data from both wet and dry conditions were combined for the first open field measures obtained when mice were 25 or 55 days of age. Data from two animals in the CL group, and one from the LL group were not included because of apparent illness of these mice. To assess the relative influence of the wet vs. dry basin experience on the re-exposure to the open field at each age, the following ratio (2nd open field - initial open field/2nd open field + initial open field) was employed. A negative ratio score would reflect a decline, a positive score an increase, and a zero no change in relative activity from the first to second test in the open field on a given day of testing. Because variability was considerably greater in the running wheels when animals were tested at 55--60 days of age as compared to when they were 25-30 days old, separate analyses were performed at each age. Furthermore, day one activity was analyzed alone to assess the initial reaction of the mice to being placed individually in the living cages with attached running wheels. In addition, data from days 2-6 were analyzed (with days as a repeated, within subjects measure) to assess the development of wheel-running activity. Mechanical problems occasionally were encountered with the wheels, and in such cases, missing data were replaced with the appropriate mean from the remaining members of the group. This replacement procedure was employed only in the repeated measures analyses, and involved less than 2% of the data points. We deemed this procedure preferable to dropping all of the data for a given subject if any single data point was missing. When the mice were 25 days old returning the individual from the running wheel to his home cage and littermates resulted in little or no fighting among the social triad. However, agonistic behavior clearly did occur when the mice were 60 days old; thus, only results of agonistic testing at 60 days of age are presented below. RESULTS
Body Weight As can be seen in Table i, all groups of mice increased their body weight from 25 to 55 days of age, as expected, Age: F(1,88)=3989.55, p<0.001. However, at 25 days of age those mice that received either lead only postnatally (CL), or both pre- and postnatally (LL), were comparable in weight and lighter than those whose mothers drank the lead solution only prior to giving birth (LC) or had not been exposed to the lead solution at all (CC). These group differences were still present, but were not as marked at 55 days of age, Group x Age: F(3,88)=9. l 1, p<0.001.
585 TABLE 1 MEAN BODY WEIGHT (g) -+SEM
Age (Days) Group CC CL LC LL
25 16.0 -+ 0.3 12.5 - 0.4 15.4 -+ 0.4 11.4 _+ 0.3
55 27.5 _+ 0.4 26.3 _+ 0.5 27.4 _+ 0.4 25.8 _+ 0.4
difference (Duncan's p<0.05) was between CL (most squares crossed) and LL (fewest crossed) groups, but at 55 days of age the CL group crossed significantly more squares than either of the other lead exposed groups (LL or LC). Hole pokes. Figure 1B presents nose-poke data in the first open field test conducted when the mice were 25 days old and subsequently when they were 55 days of age. Mice that received lead postnatally only (CL) poked the most holes when tested at 25 days of age, while the other three groups were comparable. However, when the mice were 55 days old those either receiving lead during the postnatal period (CL), or both prior to and following birth (LL), explored more holes than either the CC or the LC groups, Group x Age: F(3,59)=4.19, p<0.01. In the context of this interaction it should be noted that all groups except the LL group showed a substantial decrease in the number of holes poked from 25 to 55 days of age. The LL group actually increased their exploration of holes somewhat between the two tests. Stand ups. Similar to the data previously described, all groups increased the number of times they stood up during the first open field test from when they were 25 days of age to the first open field test when they were 55 days of age, Age: F(1,59)=107.66, p<0.001 (see Fig. 1C). In addition the LL group stood up less than any of the other groups, particularly when they were 55 days old, Group: F(3,59)=2.74, p =0.05.
Basin Test Squares crossed. Generally, mice tested in the 17°C water swam across more squares than littermates ran across when tested in the same basin without water, Basin: F(1,58)=208.59, p<0.001 at ages 25 and 55 days. However, there were no significant effects involving either age or group. In addition, while we did not attempt to quantify aspects of swimming performance (cf, [1]), there did not appear to be any gross abnormalities in the swimming ability of mice exposed to lead as compared to control animals.
Open Field~First Test at Each Age Squares crossed. All groups increased the number of squares crossed during the first open field test (Fig. 1A) from 25 to 55 days of age, Age: F(1,59)=93.01, p<0.001. Furthermore, across ages, the CL group crossed more squares than any of the other three groups which were comparable to each other, Group: F(3,59)=3.60, p<0.02. Although the overall Group x Age interaction did not reach statistical significance, a Duncan-multiple-range test was performed at each age to assess the apparently reduced slope in the squares crossed data of the LC group from age 25 to 55 days of age relative to the other groups. These specific contrasts indicated that at 25 days of age, the only statistically significant
Open Field: Pre-Basin vs. Post-Basin Test at Each Age Hole poke ratios. As reflected in the hole-poke ratio described in the data analyses section above, mice tended to decrease the number of holes they nosed in the second (post-basin) test relative to the first (pre-basin) test at both ages of testing; that is, all group mean ratio scores are negative or near zero (see Fig. 2). However, there were differences between groups in the degree to which these ratio scores were altered as a function of both age and the intervening basin condition, Group × Age x Basin: F(3,57)=3.70, p<0.02. Following testing in the dry basin (left panel, Fig. 2) this
DOLINSKY ET AL.
586
F I R S T OPEN F I E L D A. SQUARE CROSSINGS I CL
30C
B. NOSE POKES
C. STAND UPS
70
~
40 I
I\
~E
CC
26( Q Z I
/
/
I
t
3E
18(
~
30
JLC
•
50
20
"'~'-~CC
,o
30
14C I
I
25
55
eS
LL
",,N~
;.;:/
~
,,,,,"T L c
T"".,,
/
Z Z
/,
\
LL
/
C
#
I II
"
nr aD 22C
60
L
I( 1
I
I
I
I
25
55
25
55
DAYS OF AGE FIG. 1. First Open Field: Mean number (_SEM) of square crossings (panel A), nose pokes (panel B), and standups (panel C) for each of the four groups when tested prior to wet or dry basin exposure at 25 and again at 55 days of age,
ratio increased toward zero (a zero means there was no difference in relative numbers of hole pokes on first and second tests) for both the CC and LC groups from testing at 25 to testing at 55 days of age. But this trend was reversed for those mice still drinking lead (CL and LL groups). However, following wet basin exposure (right panel, Fig. 2) all groups, regardless of lead history, tended to show an increase toward zero in terms of this hole poke ratio score from testing at age 25 to 55 days; here however, that increase was not as marked for the LC group. Ratios for squares crossed and stand ups. In the dry basin condition at 25 days of age all groups tended to cross more squares in the second test in the open field relative to the initial open field test. This effect was attenuated when mice were retested at 55 days of age. The trend observed in the dry basin was reversed from that of the wet basin condition where the mice at 25 days of age crossed fewer squares in the second open field test relative to the first exposure. This decrease in activity was less apparent when mice were retested when they were 55 days of age, Age × Basin: F(1,57)=26,48, p<0.001. A similar trend was observed for stand ups, Age × Basin: F(1,57)= 11.80, p<0.01. There were
no statistically significant effects involving groups with respect to the influence of the intervening basin condition on the ratios for either squares crossed or the number of times mice stood up in the open field. Thus, in contrast to the hole poke ratios, the squares crossed and stand up ratios suggest that mice exposed to lead during several developmental periods are not differentially responsive to the presumed disparity in stress of the two basin conditions at either testing age.
Running Wheel Activity Trends in the running wheel data for the 12-hour light and dark periods across the six days of running wheel testing were similar to those for data from the first three hours after light onset and offset. However, there was somewhat less variability, and the more immediate responses to change in illumination were better reflected in the data from the three hour periods. Thus, while providing a similar picture to that seen when each of the entire 12 hr periods were examined, data from the shorter, three-hour periods are presented. Thus, only means for each of the three hour periods follow-
L E A D , D E V E L O P M E N T A L PERIOD A N D B E H A V I O R
587
OPEN F I E L D HOLE POKE CHANGE RATIO DRY BASIN BETWEEN ¢,i c~ Z ,,¢
WET BASIN BETWEEN OPEN FIELD TESTS I 81 2
OPEN FIELD TESTS I 8, 2 0,1
ILl nO.. "1I-U') 0
(+i
0
(-)
J
~-0°1
P
W
S
J
/
o.
i
LC
P
I
I-- - 0 . 2
~
-0.3
IE
.~ - 0 , 4
25
55
25
55
DAYS OF AGE FIG. 2. Open Field Hole Poke: Ratios (2rid O . F . - l s t O.F./2nd O.F.+lst O.F.) for mice from each of the four groups when tested at 25 and again at 55 days of age. Either wet or dry basin exposure occurred between the 1st and 2nd open field test at each age.
ing light onset (a.m.) and offset (p.m.) across the six days of testing in the wheel available cages are presented (Fig. 3). During the first three hours of the light and dark phases of the first day, the 25-30 day old CL mice were less active in the wheels than any other group of comparable age (top panels, Fig. 3), Group: F(3,27)=4.06, p<0.02. All groups increased activity in the dark period across the subsequent five days. There were no reliable differences between groups with respect to either day one activity or the development of activity across the remaining five days when the mice were retested at 55-60 days of age (bottom panels, Fig. 3).
Agonistic Behavior As might be expected in preadolescent male mice, there was relatively little fighting among littermates after the 30-day old, wheel-tested animals were returned to their home cages. However, as shown in Fig. 4 when these males were 60 days old and then reintroduced to their home cages (after their second, 6 day running wheel test), all groups with any lead-exposure history (LC, CL, LL) were comparable and had markedly shorter mean latencies to initiate fighting than the CC mice with no history of exposure to lead, Group: F(3,27)=4.68, p<0.01; Duncan's, p<0.05. No differences however, were observed among the groups in terms of the number of bouts of fighting during the five minute observation period following return of the 60 day old animals to their home cage.
DISCUSSION
There has been a lack of consensus concerning the effects of lead on activity levels in rodents. In part, this may b e due to differences in experimental protocols utilized. Our present results demonstrate the complexity of the task before us. Factors such as developmental period when lead is delivered, age at testing, and specific test(s) employed influence experimental findings. However, data from our aggression testing at 60 days of age suggest that lead, regardless of when it was presented, may alter some aspects of social behavior. It thus appears that changes in agonistic behavior may be associated with lead ingestion [3, 12, 20]; however, the relationship is undoubtedly complex. That is, when triads of lead treated mice were observed after a period of isolation, their latency to aggress was reduced. Interestingly, Finch and Reiter [7] found decreased aggression when young lead treated rats were matched with same aged control animals. The latter findings are congruent with our observation that young lead treated males are submissive to control mice; this pattern was reversed in older pairs [3]. In the present study, lead caused a comparable decrease in body weight in both groups receiving lead postnatally (CL, LL). It has been suggested [17] that the effects of lead on activity levels are influenced by body weight changes associated with, but not unique to, lead exposure. However, many of the measures of activity reflected differences between these two groups. F o r example, during the first open field test, mice that were presented lead only postnatally
588
DOLINSKY ET A L .
CC
1400
1200
f
LC
CL
LL
_..oA M •---- PM
I000
I AGE 25 -301
800
f
600
O3
Z 0
400
\
.L \
200
\
I
Io.~O-'O-"O,,
b-.o--'°- °
"
T-
I
,ll.
0--"0
0~.
0 - -0~
I
•
|
I
i
1500 1400
Z LLI
1200 I000
~o"
%
0
.J
W
0
U,t
"I
-i
1
800
.t I I t
600
t
400
I
I
! I 'I
T
!
o
1
I
0 ~"
200
,I
,IJ.
I ~0
~.
"0..0o0 "O
~O--o •
|
•
•
2 3 4 5 6
•
•
•
•
0--0
|
23456
23456
23456
DAYS FIG. 3. Running Wheel: Mean number of revolutions and average SEM, (vertical bars) during the first 3 hours after light onset (AM, dashed lines) and after light offset (PM solid lines) on each of six consecutive days for mice from each of the four groups (not tested in the open field) when they were 25-30 days of age (top four panels) and when they were 55-60 days of age (bottom four panels).
(CL) crossed more squares than mice which had received lead both before and after birth (LL). In addition, activity levels were not similar in these two groups on day one in the running wheel when the mice were 25-30 days old. Thus, while one of the effects of postnatal lead administration was to decrease body weight, weight alone does not appear to predict changes in either the open field or running wheel behavior. Previously we reported a similar failure of bodyweight to predict the impact of lead on seizure susceptibility of several stocks of 21 day old mice [2]. Not surprisingly, the specific test employed in the present
investigation differentially reflected our manipulations (see also [15,24]). For instance, 25 day old mice exposed to lead postnatally only (CL) made relatively few revolutions in the running wheels, but were highly active in the open field at this age. These test conditions apparently are assessing different facets of behavior and lead toxicity. The development of a more unified schema of the impact of this toxic element on behavior may call for new approaches such as those which have been utilized in the analysis of changes in behavior following specific brain lesions [6]. Silbergeld and Goldberg [21,22] reported increased hori-
LEAD, DEVELOPMENTAL
l
E
200
T
POST RUNNINGWHEEL AGE 60 AGRESSION- 5 MINUTE TEST
CC
mm
I00
o
z
50
.J
Ix
PERIOD AND BEHAVIOR
CL LC LL FIG. 4. Agonistic Behavior: Mean latency to first fight (seconds) during the five minute aggression test for 60 day old mice in each of the four groups after the mice were tested in the wheel-available cages and were returned to their home cage which contained two littermates.
589 suggested an increase in activity for mice who had r e c e i v e d lead from birth to weaning but w e r e then returned to water. The latter findings, h o w e v e r , were not consistent. It is interesting to note that in the present study, squares crossed during the first open field test at both ages also suggested that mice that r e c e i v e d lead postnatally only (CL) m o v e d across more squares than any o f the o t h e r groups. Finally, our results suggest that the nature and severity o f lead's behavioral effects are not simply or monotonically related to the total duration o f e x p o s u r e , but are critically related to the develo p m e n t a l period(s) during which this toxic element is present and the behavioral task e m p l o y e d [15,24]. Thus in several instances perinatal (LC) and postnatal alone (CL) lead exposure yield opposite effects on behavior, while the effects o f lead administration during both periods (LL) is not explained by an additive m o d e l based on perinatal and postnatal exposure. The underlying m e c h a n i s m s for these results warrants further investigation.
ACKNOWLEDGEMENTS zontal activity, as d e t e c t e d by activity monitors, in 40-60 day old mice that had r e c e i v e d lead from birth until the time o f testing (similar to our C L group). T h e i r data also
We thank Wayne Kashinsky for his help with the word processing system and Dr. Richard Pastore who allowed us access to his printer.
REFERENCES
1. Adams, P. M., J. D. Fabricant and M. S. Legator. Cyclophosphamide-induced spermatogenic effects detected in the Ft generation by behavioral testing. Science 211: 80-82, 1981. 2. Burright, R. G., P. J. Donovick, K. Micheis, R. J. Fanelli and Z. Dolinsky. Effect of amphetamine and cocaine on seizure in lead treated mice. Pharmacol Biochem Behav 16: 631-635, 1982. 3. Burright, R. G,, W. J. Engellenner and P. J. Donovick. Lead differentially affects agonistic behavior of adult mice of two ages. Physiol Behav 30: 285-288, 1983. 4. Dolinsky, Z., E. Fink, R. G. Burright and P. J. Donovick. The effect of lead, d-amphetamine, time of day on activity levels in the mouse. Pharmacol Biochem Behav 14: 877-880, 1981. 5. Driscoll, J. W. and S. E. Stegner. Behavioral effects of chronic lead ingestion on laboratory rats. Pharmacol Biochem Behav 4: 411-417, 1976. 6. Fanelli, R. J., R. G. Burright and P. J. Donovick. A multivariate approach to the analysis of genetic and septal lesion effects on maze performance in mice. Behav Neurosci, 1983, in press. 7. Finch, A. A. and L. W. Reitter. Alterations in social behavior in the rat during chronic low-level exposure to lead and tritum. Toxicol Appl Pharmacol 37: 160, 1976. 8. Fiynn, J. C., E. R. Flynn and J. H. Patton. Effects of pre- and postnatal lead on affective behavior and learning in the rat. Neurobehav Toxicol l: Suppl l, 93-103, 1979. 9. Fuller, J. L. Fuller BWS lines: History and results. In: Development and Evolution o f Brain Size, edited by M. E. Hahn, C. Jensen and B. C. Dudek. New York: Academic Press, 1979. 10. Fuller, J. L. and M. E. Hahn. Issues in the genetics of social behavior. Behav Genetics 6: 391-406, 1976. I I. Gotler, M. and A. L. Michaelson. Growth, behavior and brain catecholamines in lead exposed neonatal rats: A reappraisal. Science 187: 359-360, 1975. 12. Hastings, L., G. P. Cooper, R. L. Boruschein and A. Michaelson. Behavioral effect of low level neonatal lead exposure. Pharmacol Biochem Behav 7: 37--42, 1977.
13. Hejtmancik, M. R. Jr., E. B. Dawson and B. J. Williams. Tissue distribution of lead in rat pups nourished by lead-poisoned mothers. J Toxicol Environ Health 9: 77-86, 1982. 14. Kostas, J. D., D. J. McFarland and W. G. Drew. Lead-induced behavioral disorders in the rat: Effects of amphetamine. Pharmacology 16: 226--236, 1978. 15. Mullenix, P. Effect of lead on spontaneous behavior. In: Low Level Lead Exposure: The Clinical Implications o f Current Research, edited by H. L. Needleman. New York: Raven Press, 1980. 16. Overman, S. R. Behavioral effects of asymptomatic lead exposure during neonatal development in rats. Toxicol Appl Pharmacol 41: 459-471, 1977. 17. Rafales, L. S., R. L. Bornschein, I. A. Michaelson, R. K. Loch and G. F. Barker. Drug induced activity in lead-exposed mice. Pharmacol Biochem Behav 10: 95--101, 1979. 18. Ratcliffe, J. M. Lead in Man and the Environment. New York: Ellis Horwood Limited, 1981. 19. Reiter, L. W., G. E. Anderson, J. W. Laskey and D. F. Cahill. Developmental and behavioral changes in the rat during chronic exposure to lead. Environ Health Perspect 12:119-123, 1975. 20. Sauerhoff, M. W. and A. I. Michaelson. Hyperactivity and brain catecholamines in lead exposed developing rats. Science 182: 1022-1024, 1973. 21. Silbergeld, E. K. and A. M. Goldberg. A lead induced behavior disorder. Life Sci 13: 1275-1282, 1973. 22. Silbergeld, E. K. and A, M. Goldberg. Lead induced behavioral dysfunction: An animal model of hyperactivity. Exp Neurol 43: 146-147, 1974. 23, Sobotka, T. J. and M. P. Cook. Postnatal lead acetate exposure in rats: Possible relationship to minimal brain dysfunction. A m J Merit Defic 79: 5-9, 1974. 24. Zimering, R. T., R. G. Burright and P. J. Donovick. Effects of pre-natal and continued lead exposure on activity in the mouse. Neurobehav Toxicol Teratol 4: 9-14, 1982.