Prenatal Cocaine But Not Prenatal Malnutrition Affects ... - Science Direct

0 downloads 0 Views 1MB Size Report
S. S. ALMEJDA. AND J. R. CALLER. Prenatal cocaine buf not prenatal malnutrition affects foster mother-pup interactions in rats. NEUROTOXICOL. TERATOL.
Neurotoxicology

Pergamon

and Teratology, Vol. 17, No. 5. pp. 60-608, 1995 Copyright o 1995 Elsevier Science Inc. Printed in the USA. All rights reserved 0892.0362195 $9.50 + .OO

0892-0362(95)00021-6

Prenatal Cocaine But Not Prenatal Malnutrition Affects Foster Mother-Pup Interactions in Rats JOHN

TONKISS,*

JED

S. SHUMSKY,* PENNY L. SHULTZ,* AND JANINA R. GALLER*

SEBASTIAO

S. ALMEIDA’I

*Center for Behavioral Development and Mental Retardation, M923, Boston University School of Medicine, 80 E. Concord St., Boston, MA 02118 TLaboratory of Nutrition and Behavior, FFCLRP, Campus of the University of Sdo Paulo, 14040-901 Ribeirco Preto, SP, Brazil Received

19 December

1994; Accepted

9 May 1995

TONKISS, J., J. S. SHUMSKY, P. L. SHULTZ, S. S. ALMEJDA AND J. R. CALLER. Prenatal cocaine buf not prenatal malnutrition affects foster mother-pup interactions in rats. NEUROTOXICOL TERATOL 17(5) 601-608, 1995. The separate and combined effects of prenatal cocaine exposure and malnutrition on mother-pup interactions in rats were assessed daily from postnatal day 2 to day 21. Sprague-Dawley dams were fed a diet of low protein content (6% casein), an isocaloric diet of adequate protein content (25% casein, control), or a laboratory chow diet prior to mating and throughout pregnancy. Within each diet group, rats received either cocaine injections (30 m&kg IP two times per week prior to mating and then 30 mg/kg SC daily from days 3 to 18 of pregnancy) or saline injections. Litters were fostered on the day of birth to control mothers (i.e., nondrug-exposed dams fed the control or chow diet). Foster mothers fed the 25% casein diet showed increased contact with cocaine-exposed pups compared with nondrug-exposed pups in the second postnatal week but lower levels as the pups approached weaning. Passive nursing was increased in dams caring for prenatally malnourished, cocaineexposed pups compared with those caring for similar pups with no drug exposure. Chow-fed mothers did not differ in their behavior towards pups with or without prenatal cocaine treatment. Prenatal cocaine and malnutrition independently compromised birth weight and various reflexive milestones but the attainment of physical milestones was affected only by prenatal cocaine. There were no additive effects of the two prenatal insults on any measure of mother-pup interaction or pup development. Pregnancy

Gestation

Reflexive development

Cocaine

Malnutrition

Physical development

Mother-pup

interaction

Maternal

behavior

Rat

THE PREVALENCE of cocaine use over the last decade has stimulated intense investigation of the effects of prenatal coCaine exposure on behavioral development [e.g., (1,5,26)]. Polydrug abuse and other confounding factors (e.g., the presence or absence of prenatal care, socioeconomic status of the individual, and the home environmental conditions) in human cocaine abusers has promoted the use of animal models in which the variables may be more easily controlled. However, animal models may also have variables that are not readily subject to experimental control. It has long been recognized that environmental insults can give rise to alterations in mother-pup interactions [e.g., (12,25)], and that such disruption of the social milieu has the potential to influence behavioral development and brain function of the young independent of any direct effects of the insult on the brain [e.g., (11,23)]. Recent animal studies have sought to document the

effects of cocaine treatment during pregnancy or the suckling period on maternal behavior (16,20,27,33,40). Because pup characteristics play a significant role in determining maternal behavior (29,31), assessment of the behavior of untreated mothers towards cocaine-exposed offspring is required to fully evaluate the influence of prenatal cocaine on mother-pup interaction. Only one such investigation has been conducted (16), and it concluded that there were no significant differences in home cage behavior on postnatal day 4, in pup retrieval on day 5, or maternal aggression towards an intruder female on postnatal day 10 between foster mothers rearing prenatally cocaine-exposed pups or control pups. However, because only selected time points were used in that study, a difference in maternal behavior over the course of the suckling period may have been overlooked. The aim of the present study was to obtain a daily profile of mother-pup 601

TONKISS

602 interactions over the course of the entire preweaning period. In a companion paper (36) the development of a model of prenatal cocaine and prenatal malnutrition was described. To determine whether the effects of prenatal cocaine and prenatal malnutrition may be additive or synergistic with respect to disturbances in the postnatal social environment, the behavior of well-nourished, drug-free foster mothers towards pups with either prenatal cocaine exposure, prenatal malnutrition, or a combination of the two insults was examined. METHOD

Dietary and Drug Treatment One hundred and twenty female Sprague-Dawley, viral and antibody-free (VAF plus) rats (175-200 g) were obtained from Charles River breeding laboratories (Kingston, MA) and housed in polycarbonate cages (26 x 47 x 21 cm, Lab Products, Maywood, NJ) with 2-3 cm of “Beta-chip” hardwood bedding material (Northeastern Products Corp., Warrensburg, NY). The animal rooms were maintained at a temperature of 73 -t 3OF, monitored by a centralized computer system, and humidity was maintained within the range 45555%. A reverse 12-h day (2000-0800 h)/12-h night (0800-2000 h) light cycle was employed, with continuous dim illumination provided by red fluorescent lighting during the night phase of the cycle. The rats were divided into three groups and allocated to different diets. One group of 24 rats received a purified diet (Teklad, Madison, WI) that was restricted in protein content (6% casein). A second group of 60 rats received a purified diet that had an adequate protein content (25% casein) and was isocaloric with the 6% casein diet. These two diets had the same vitamin and mineral content. A third group of 36 rats was placed onto a regular chow diet (Purina formula 5001). This group was included to allow comparisons with other publications on prenatal cocaine. The specific composition of the casein diets is given in a previous paper (13). Within each diet group, rats were subdivided into those that would receive cocaine injections (n = 12 per diet group) and those that would receive saline injections. Larger numbers of rats were included in the 25% casein/saline (n = 48) and chow/saline groups (n = 24) to accommodate their use as foster mothers, enabling the postnatal development of the offspring to be followed. A mixed route of administration of cocaine (IP prior to mating and SC during pregnancy), which we have described in detail in a previous paper (36), was employed. Briefly, for 5 weeks prior to mating, drug-treated rats received two (IP) injections per week (Monday and Thursday, or Tuesday and Friday) of 30 mg/kg cocaine in a volume of 1 ml/kg of 0.9% saline. At the beginning of week 6 the females were mated with males that had been acclimated to the appropriate diet for a period of 1 week. The day of conception (day 0) was considered to be the day on which sperm was detected in a vaginal smear obtained between 0830 and 1030 h. From day 3 through day 18 of pregnancy, the mated females in the drugexposed groups received daily SC injections of 30 mg/kg cocaine in a volume of 3 ml/kg. Control rats were injected with saline prior to and during pregnancy using the same injection regime as the drug-exposed rats. At parturition (assigned as postnatal day 0), all litters were culled to eight pups (six males and two females, where possible) and fostered as whole litters to dams that had undergone saline injections. Pups were selected for fostering based upon their proximity to the litter mean body weight for each sex. Pups born to 6% or 25% casein-fed mothers (whether drug treated or saline treated)

ET AL.

were fostered to 25% casein/saline mothers, and pups born to chow-fed mothers (whether drug treated or saline treated) were fostered to chow/saline mothers. All fostering was performed within 24 h of birth. The following abbreviations are used to identify the six groups of pups and are based upon their prenatal dietary and drug treatment: 6% SAL (6% casein/saline); 6% COC (6% casein/cocaine); 25% SAL (25% casein/saline); 25% COC (25% casein/cocaine); Chow-SAL (chow/saline); Chow-COC (chow/cocaine). Foster motherpup interactions and the attainment of physical and reflexive milestones were assessed during the suckling period. All pups were weaned onto a standard chow diet (Purina formula 5001) on day 21. Mother-Pup

Interaction

The interaction between the foster mothers and their assigned pups was assessed daily (between 0900 and 1030 h) from postnatal day 2 to day 21, according to the procedure of Caller and Tonkiss (13). These observations were taken prior to any disturbance of the litter for assessment of developmental measures. Briefly, observations were recorded over a 30-min period using 10 spot checks at 3-min intervals. The following behaviors were measured: active nursing- mother crouched over pups in an active nursing position while the pups were suckling; passive nursing-pups suckling while the mother was not in an active nursing position; groom pup-mother picking up the pups in her mouth and licking one or more of the pups; groom self-mother licks her nipples or fur; location of the mother relative to the nest - mother located in or out of the nest region; mother-pup contact-values assigned as follows: 0, no pups in contact with the mother; 1, fewer than half of the pups in contact with the mother; 2, half or more than half of the pups in contact with the mother; pups in the nest -0, no pups in the nest; 1, fewer than four pups in the nest; 2, four or more pups in the nest. At the beginning of each daily observation period, the quality of the nest was evaluated, according to the following scale (30): 0, no nest; 1, shavings trampled down in one corner; 2, shavings pushed aside to make a bare spot in corner; 3, low ring of shavings around bare spot in corner; and 4, all shavings in cage piled into high ring around an indentation in one corner. After day 15 the pups became much more likely to follow the mother around the cage, leading to an apparent frequent relocation of the nest. Nest rating was discontinued at this time. Behavior of the dams that were not considered maternal in nature were also measured: eating, lying/sleeping, and rearing/exploring. Low-incidence behaviors (e.g., nestbuilding, drinking, digging, retrieving pups, and tail-chasing) were noted but not subject to statistical analysis. Pup Weight and Developmental

Milestones

On postnatal day 1 (the day after cross-fostering), day 21 (weaning), and day 62 all of the pups within the litter were weighed to the nearest 0.1 g using an Ohaus CT4000 averaging balance. The impact of cocaine and malnutrition on offspring development and maturation was assessed for three physical milestones: pinna unfolding (from birth), tooth eruption (from day 7), and eye opening (from day 10); and for seven reflexive milestones: righting (from birth), cliff avoidance

PRENATAL

COCAINE

AND

603

MALNUTRITION

(from birth), negative geotaxis (from birth), vibrissae placing (from day 7), auditory startle (from day 9), visual placing (from day 14), and free-fall righting (from day 14). Two male and two female pups, that were the closest to the litter mean for their sex at the time of fostering, were assessed daily between 1030 and 1500 h. Eye opening was assessed in two ways: first, as the earliest signs of membrane disjunction, and second, as fully open, defined as > 75% membrane disjunction for both eyes. The following procedures for assessment of the reflexive milestones were derived from those described for mice (10) and rat pups (37):

Righting: When the animal was placed on its side, it turned over to rest in the normal position with all four feet on the ground (principally labyrinthine and body righting mechanisms). Pups were tested at the same time each day until righting occurred within the 30-s trial, frequently in < 2 s once movement was initiated. Cliff avoidance: When the pup was placed on the edge of a table top, it turned and crawled away from the cliff drop. Pups were tested until retraction and backward (or sideways) movement was begun within 10 s of placement at the edge, and required no longer than 30 s to complete. Negative geotaxis: When the subject was placed on a slope (in this case ZOO) with its head pointing down the incline, it turned around and crawled up the slope. A cutoff of 30 s to begin to turn around, and 60 s to complete a 180° manoeuvre to a head-up position was imposed. Pups meeting both requirements were considered to have exhibited negative geotaxis; all others continued to be tested on the following day. Vibrissa placing response: When the pup was suspended by the tail and lowered so that the vibrissae made contact with a solid object (a pencil), the head was raised and the forelimbs extended to grasp the object. Auditory startle: A loud, sharp noise (created by striking two pieces of wood together, approximately 18 in. above the pup) caused an immediate startle response, seen as a sudden extension of the head and fore- and hindlimbs, which were then withdrawn and a crouching position was assumed, or there was flight (escape-avoidance) behavior at a certain age. Startle was assessed visually by two independent observers. Visualplacing response: When the pup was suspended by the tail and lowered toward a solid object (a foam rubber pad) it raised its head and extended the forelimbs in a pIacing response. Free-fall righting: When the subject was suspended by the hind- and forelimbs with the abdomen uppermost, and was suddenly released, it turned in midair to land on all fours (statokinetic labyrinthine response).

period for each behavioral measure and these values were subjected to log (X + 1) transformation (38) to normalize the data. Minor violations of sphericity remained for three measures (contact with pups, dam in nest, and passive nursing); hence, Huynh-Feldt probabilities are quoted throughout for all interactions involving the repeated measure. Pup data. On postnatal days 1, 21, and 62 litter mean values for body weight were derived for each sex. These data were analyzed using three-way (diet x drug x sex) ANOVAs. Analyses were performed separately for each age because of vastly dissimilar variances across the three ages. Preliminary analysis of the age of attainment of the physical and reflexive milestones using three-way (diet x drug x sex) ANOVAs revealed no significant main effects of sex or first-order interactions. Consequently, the data were collapsed across sex for all milestone measures by deriving litter mean values and subjecting these data to two-way ANOVAs (diet x drug). RESULTS

Mother-Pup

Interaction

Significant alterations in mother-pup interactions were observed for the 25% casein-fed foster mothers rearing the offspring of 25% casein or 6% casein-fed mothers that had been prenatally treated with cocaine or saline. In contrast, chowfed foster mothers did not differ in their behavior towards the offspring of chow-fed dams that had been prenatally treated with cocaine or saline, nor were there any alterations in the three nonmaternal behaviors in these foster mothers. Hence, the results reported below appertain only to foster mothers of the 25% casein group.

Active and Passive Nursing With respect to active nursing, no significant differences were found in the behavior of 25% casein-fed foster mothers towards pups with different prenatal drug and dietary histories. However, a significant diet x drug interaction was demonstrated for passive nursing, F(l, 27) = 4.78, p < 0.05. No significant age x diet or age x drug interactions were indicated by ANOVA. Hence, the data were collapsed across age to show the mean incidence of passive nursing for each pup treatment group (Fig. 1). Passive nursing was observed significantly more often in dams rearing 6% COC pups than in dams rearing 6% SAL pups (p < 0.05, Newman-Keuls). There was no significant difference in the incidence of passive nursing between dams caring for 25% SAL or 25% COC pups, nor was there a difference between dams caring for 6% SAL or 25% SAL pups. The occurrence of both active nursing, F(9, 243) = 5.48, p < 0.001, and passive nursing, F(9, 243) = 2.85, p < 0.01, decreased with pup age.

Data Analysis

Nest Rating, Contact With Pups, and Dam in Nest

Maternal data. Analysis of interactions between the foster mothers and their pups was performed separately for the two types of diet groups (i.e., those consuming the casein diets and those consuming the chow diet). Thus, three-way (diet x drug x age) ANOVAs were applied to the behavioral data of 25% casein foster mothers (two levels of diet and two levels of drug pertaining to the prenatal treatment of their pups: 25% SAL, 25% COC, 6% SAL, and 6% COC), and two-way (drug x age) ANOVAs were applied to the data of chow-fed foster mothers (two levels of drug pertaining to the prenatal treatment of their pups: Chow-COC and Chow-SAL). Age was a repeated measure. Mean values were derived for each 2-day

Significant age x drug interactions were indicated for nest rating, F(6, 162) = 2.34, p < 0.05, and for contact with pups, F(9, 243) = 2.51, p < 0.05. There were no significant effects of diet nor were there significant interactions involving diet for either of these measures. Hence, the data for both casein diet groups have been combined for illustration (Fig. 2). Figure 2 (upper) shows that the quality of the nest declined as the pups matured, F(6, 162) = 11.65, p < 0.001, but the pattern of decline differed according to drug exposure of the pups. Early in the postnatal period nest quality tended to be high in dams rearing cocaine-exposed pups, but nest quality of these dams was significantly lower than that of dams caring

604

TONKISS PASSIVE

pared with that of dams caring for saline-exposed pups (M = 0.353). No other differences were indicated by ANOVA.

NURSING

Pup Body Weights and Developmental

PUP GROUP

FIG.

1. Mean

(log transformed)

daily

incidence

ET AL.

of passive

nursing

during the suckling period exhibited by foster dams (fed a 25% casein diet) rearing pups of four prenatal treatment groups: 6%SAL (6% casein diet/saline), 6%COC (6% casein diet/cocaine), 25%SAL (25% casein diet/saline), 25%COC (25% casein diet/cocaine). *Significantly greater than 6% SAL (p < 0.05, Newman-Keuls).

for saline-exposed pups on postnatal days 10-l 1 (p < 0.05, Newman-Keuls). Physical contact between the foster mothers and their pups showed a U-shaped curve (Fig. 2, lower), with mother-pup contact being high both early in the postnatal period and again as the pups approached weaning age, but low in the second postnatal week, F(9, 243) = 6.47, p < 0.001. Contact with cocaine-exposed pups was significantly higher than contact with saline-exposed pups on postnatal days 8-9 (p < 0.01) and 14-15 (p < 0.05), but was significantly lower than contact with saline-exposed pups on postnatal days I& 19 (17 < 0.05, Newman-Keuls). The mothers were observed to be in the nest less often as the pups matured, F(6, 162) = 19.20, p < 0.001, but nest occupation did not differ according to prenatal drug exposure or the dietary group of the pups.

Milestones

Table 1 shows the litter mean pup weight at postnatal days 1, 21, and 62. On day 1, the day after fostering, there were significant effects of diet, F(2, 92) = 25.23, p < 0.001, and of drug, F(l, 92) = 4.06, p < 0.05. As expected, pups born to chow-fed and to 25% casein-fed mothers weighed significantly more than those born to mothers fed the low protein (6% casein) diet (p < 0.01, Newman-Keuls). More surprisingly, pups of chow-fed dams were significantly heavier than those of 25% casein dams (p < 0.01). Prenatal cocaine gave rise to a small but significant body weight deficit at this age (7070) compared with the saline-treated controls (overall means: saline = 7.29 g, cocaine = 6.76 g), but there was no significant sex difference and none of the interaction terms were significant. By weaning (day 21) an effect of diet, F(2, 92) = 6.01, p < 0.01, was demonstrated comparable to that observed on day 1, and indicated the same pattern of results. Thus, prenatally malnourished pups weighed significantly less than those of the chow (p < 0.01, Newman-Keuls) and 25% casein pups (p < 0.05); and, 25% casein pups weighed less than those of the chow pups (p < 0.01). However, there was no longer a significant effect of drug, indicating that the body weights of the prenatally cocaine-exposed pups had caught up

NEST RATING

Groom Pup and Groom Self CONTACT

WITH PUPS

The incidence of pup grooming, F(9, 243) = 7.54, p < 0.001, and of self-grooming, F(9, 243) = 13.99, p < 0.001, declined as the pups matured. A diet x drug interaction for pup grooming approached significance, F( 1, 27) = 2.90, p < 0.1, and this trend was due to the 6% COC pups being groomed less often than 6% SAL pups, with the 25% SAL pups being groomed more often than 25% COC pups. No other differences were indicated by ANOVA. Lying/Sleeping,

Eating, and Rearing/Exploring

The incidence of the three nonmaternal behaviors of lying/ sleeping, F(9, 243) = 9.17, p < 0.001, eating, F(9, 243) = 2.43, p < 0.05, and rearing/exploring, F(9, 243) = 2.49, p < 0.01, all showed significant changes as the pups matured. However, only lying/sleeping showed a systematic increase; eating and rearing/exploring were subject to nonspecific fluctuations. A significant effect of drug was indicated for rearing/exploring, F(l, 27) = 4.47, p < 0.05. This was due to a lower incidence of rearing/exploring in the dams caring for cocaine-exposed pups [mean log(rear + I), M = 0.281) com-

FIG. 2. Top panel: mean (log transformed) nest rating (for postnatal days 2-15) of dams fed a 25% casein diet and rearing prenatally cocaine-exposed or saline-exposed pups. Lower panel: mean (log transformed) contact of dams fed a 25% casein diet with prenatally cocaine-exposed or saline-exposed pups, during postnatal days 2-21. Cocaine different from saline: **p < 0.01, *p < 0.05 (NewmanKeuls).

PRENATAL

COCAINE

AND

605

MALNUTRITION TABLE

POSTNATAL BODY WEIGHTS DURING PREGNANCY 6%

AS (days)

SCf

Casein

Saline

1

i SEM OF RATS WHOSE MOTHERS AND WERE EXPOSED TO COCAINE 25%

Casein

Saline

Cocaine

CONSUMED DIFFERENT OR SALINE INJECTIONS

DIETS

Chow

COCaioe

Saline

Cocaine

1

M F

6.7 + 6.3 f

0.2 0.2

5.8 k 0.3 5.6 k 0.4

7.2 + 0.3 6.9 + 0.3

7.0 * 6.5 k

0.5 0.5

8.1 + 0.2 7.7 + 0.3

7.8 f 7.7 t

0.3 0.3

21

M F

67.2 k 63.0 k

2.2 2.5

61.2 +- 2.8 59.1 + 3.1

68.7 k 2.2 66.2 k 2.0

68.8 f 64.0 +

4.2 5.1

63.7 f 1.9 59.0 f 2.3

60.1 -t 58.6 k

1.6 1.1

62

M F

400.8 k 11.0 390.3 t 4.8 254.6 k 13.2 251.9 f 4.0

406.0 k 5.7 266.7 + 1.4

Day 1 =

day

after cross-fostering;

409.7 f 10.9 411.3 -+ 8.3 260.4 f 9.5 264.3 k 8.7

383.5 k 16.3 270.4 f 6.5

day 21 = weaning.

to

those of the saline-treated pups. At weaning a sex difference had emerged, F(1, 92) = 4.09, p < 0.05, with the male offspring weighing more than the females. None of the interaction terms were significant. At day 62, there was no longer a significant difference between the three diet groups, nor was there a difference between the drug groups, but the male rats remained significantly heavier than the females, F(1, 92) = 604.75, p < 0.001. The mean age at which the pups attained their physical milestones is given in Table 2. Prenatal cocaine gave rise to significant delays in the unfolding of the pinnae, F( 1, 46) = 4.34, p < 0.05, and in eye opening [first signs: F(1, 46) = 5.15, p < 0.05; fully open: F(1, 46) = 4.14, p < 0.051. A tendency for prenatal cocaine to delay tooth eruption failed to attain statistical significance, F(1, 46) = 2.23, p = 0.1. No significant differences were observed between the three diet groups, nor were there any significant diet x drug interactions for any of these measures. The mean age of attainment of reflexive milestones is given in Table 3. Three of the seven measures showed significant differences due to diet or to prenatal cocaine exposure. Development of the auditory startle response was significantly delayed by prenatal cocaine exposure, F(1, 46) = 6.48, p < 0.02, but not prenatal malnutrition, whereas vibrissae placing was delayed by prenatal malnutrition, F(2, 46) = 3.41, p < 0.05, but not prenatal cocaine. Newman-Keuls tests revealed that prenatally malnourished pups were significantly older than chow-fed pups when vibrissae placing emerged (p < 0.05) and tended to be older than pups born to 25% casein-fed dams (p = 0.05). The age at

which the pups demonstrated the righting reflex differed among the six treatment groups, as evidenced by a significant diet x drug interaction, F(2, 46) = 4.21, p < 0.05. This interaction was due to prenatal cocaine exposure delaying the development of the righting response in both of the casein diet groups but tending to accelerate development of righting in the chow group. An ANOVA that excluded the chow group confirmed a significant delay in the righting reflex due to cocaine, F(1, 27) = 4.61, p < 0.05, in the casein diet groups, with no significant diet x drug interaction. Cliff avoidance, negative geotaxis, visual placing, and free-fall righting were not affected by either prenatal cocaine or prenatal dietary conditions. DISCUSSION

Mother-pup interactions were significantly affected by prenatal cocaine exposure but not by prenatal malnutrition. More specifically, the behavior of one group of foster dams (i.e., those fed the 25% casein diet) towards pups with prenatal cocaine exposure was different to that exhibited towards nondrug-exposed pups. These foster dams showed increased contact with cocaine-exposed pups in the second postnatal week compared with nondrug-exposed pups (days 8-9 and days 1415), but lower levels of contact with the cocaine-exposed pups as the pups approached weaning (days 18-19). The extent to which these differences in mother-pup contact can be attributed to differences in nursing must be considered. The amount of active nursing of pups with prenatal cocaine exposure was

TABLE 2 MEAN k SEM AGE OF ATTAINMENT

OF DEVELOPMENTAL

6% Casein Saline

MILESTONES

25% Casein Cocaine

Saline

(DAYS) Chow

Cocaine

Saline

Cocaine

Pinna unfolding

2.28 + 0.21

2.82 k 0.29

2.60 k 0.22

3.08 + 0.38

2.38 f 0.14

2.60 + 0.19

Tooth eruption

9.25 c 0.36

9.33 f 0.35

8.97 + 0.27

9.57 * 0.39

8.91 + 0.16

9.27 + 0.17

Eye opening (First signs)

13.13 + 0.18

13.45 + 0.30

13.21 f 0.23

13.64 + 0.26

13.35 f 0.21

13.90 + 0.13

Eye opening (Fully open)

13.73 + 0.18

13.93 f 0.33

13.66 f 0.20

14.06 t 0.30

13.85 * 0.17

14.37 + 0.11

No gender

differences

were present;

hence the data have been collapsed

across gender.

606

TONKISS TABLE MEAN

? SEM AGE

OF ATTAINMENT

3

OF REFLEXIVE

6% Casein Saline

Righting reflex Cliff avoidance Negative geotaxis Vibrissae placing Auditory startle Visual placing Free-fall righting

1.17 3.63 4.71 8.46 12.63 15.08 15.17

No gender differences

* k k + f k i

0.17 0.40 0.35 0.57 0.28 0.20 0.15

were present;

MILESTONES

(DAYS)

25% Casein COCaiIlC

1.41 3.93 4.88 9.15 13.80 15.43 15.68

* t k * & f +

0.21 0.50 0.60 0.37 0.40 0.25 0.35

Saline

1.11 3.81 4.36 8.01 12.89 15.31 15.61

+ 0.06 -t 0.34 + 0.32 +_ 0.26 i- 0.34 * 0.31 + 0.22

hence the data have been collapsed

similar to that displayed towards nondrug-exposed offspring. However, passive nursing was increased in those dams caring for prenatally malnourished pups with concomitant cocaine exposure compared with that displayed towards prenatally malnourished pups with no drug exposure. It follows that only in the malnourished group could the observed increase in contact between the foster mothers and the cocaine-exposed pups in the midsuckling period be attributed (in part) to increased levels of suckling. However, contact between the mother and the pup represents more than simply nursing contact. Because the pups are born without fur, the mother also serves as an important source of heat until the pups are able to regulate their own temperature. Elevation of maternal body temperature is well established as an important determinant of mother-young contact (l&22). During periods of contact with the pups the mother eventually suffers an acute elevation in body temperature, forcing her to break contact with the young, which in turn enables a return to thermal homeostasis (22). Ultrasonic vocalization by the pups, which may be elicited by cooling (3) or by separation from the mother (17), can stimulate the dam to resume contact (18). Nevertheless, contact is maintained only if the pups are cool or if they attach to the nipples (18). It is not possible to identify the reason for alterations in the amount of contact between pups with prenatal cocaine exposure and their foster mothers in the present investigation, but alterations in the pattern of ultrasonic vocalizations or a slower rise in maternal body temperature are two feasible mechanisms. Previous investigations have not identified differences in ultrasonic vocalizations between prenatally cocaine-exposed and nondrug-exposed pups in the absence of further drug challenge (14,24). However, these studies were conducted on the cocaine-treated offspring of chow-fed dams for which differences in mother-pup interaction have generally not been observed (see below). An analysis of ultrasonic vocalizations in the cocaine-treated offspring of casein-fed dams appears to be warranted. A slow rise of maternal temperature associated with the small body size of pups with postnatal malnutrition has been proposed as a likely explanation for increased contact with such pups [e.g., (31)]. However, size of the pups alone cannot account for the present findings. Prenatal malnutrition gave rise to a larger and more enduring body weight deficit than that observed with prenatal cocaine, yet only prenatal cocaine was observed to alter mother-pup contact. Thus, the involvement of an alteration in pup characteristics other than physical size is indicated. Late in the suckling period (days 18- 19) the lower levels of

ET AL.

Chow Cocaine

1.25 3.22 4.59 8.31 13.03 15.57 15.64

?I 0.09 rf- 0.24 + 0.47 ~fr0.48 t 0.39 t 0.32 k 0.29

Saline

1.29 3.94 5.48 7.67 12.29 15.42 15.35

f 0.08 f 0.27 + 0.25 f 0.17 +I 0.23 k 0.20 k 0.18

COCt3iW?

1.08 3.97 5.36 8.17 12.86 15.50 15.17

-t 0.06 f 0.44 f 0.70 IL 0.27 + 0.18 + 0.38 + 0.24

across gender.

mother-pup contact observed with prenatally cocaine-exposed pups compared with nondrug-exposed pups were probably related more to the behavior of the pups than to the behavior of the dam herself. This is because the pups were highly active by the third postnatal week and were well-equipped to follow the dam around the cage. Contact was characterized most often by one or more pups clambering over the dam while she was sleeping or engaged in some nonmaternal activity, such as grooming. It may be of relevance that deficits in play behavior following prenatal cocaine exposure have recently been documented (39). However, the amount of contact with the cocaine-treated pups only underwent a transient reduction at this age and returned to the same levels as for the saline-treated pups at the subsequent age. Consequently, the reliability of this finding remains uncertain. Maternal behavior was not significantly affected in the chow-fed group of foster dams caring for pups with prenatal cocaine exposure. This finding was somewhat surprising in view of the differences observed with 25% casein dams, and it may be that the smaller sample size of the chow-fed group of foster mothers was too limited to reveal a significant impact of cocaine on these measures. Nevertheless, the present findings are consistent with those of previous studies (16,33). When caring for their own pups, chow-fed animals exposed to cocaine during pregnancy have been shown to exhibit similar levels of pup retrieval and similar nest-building latencies to those of nondrug-exposed dams (33). In a more extensive study, untreated surrogate foster dams were found to show similar behavior in the home cage towards prenatally cocaineexposed and saline-exposed pups on postnatal day 4, similar pup retrieval on days 5-9, and similar levels of aggression towards an adult female intruder on postnatal day 10 (16). One study, in which chow-fed dams were treated with cocaine daily until the weaning of their pups, showed an overall increase in the amount of nursing (27). However, such postnatal cocaine treatment of the dams would be expected to have a more profound effect upon mother-infant interaction [e.g., (40)] than that observed in untreated or saline-injected foster mothers caring for prenatally cocaine exposed pups. It is worth noting that in a previous study (13) foster mothers fed a 25% casein diet showed small but significant differences in the amount of passive nursing and grooming of prenatally malnourished pups compared with well-nourished control pups. In the current experiment no differences were observed between the diet groups in these measures. The main difference between the two studies is the presence (current study) or absence (previous study) of prenatal injections of saline or cocaine. Chronic SC saline injections during the pre-

PRENATAL

COCAINE

AND

607

MALNUTRITION

natal period have been shown to produce subtle but lasting effects on offspring behavior (1S), suggesting that even saline injections are mildly stressful. Thus, the additional stress to both the foster mothers and the pups experienced as a result of SC injections may have served to obscure any small differences in mother-pup interaction associated with the nutritional history of the pups. The issue of prenatal stress must also be addressed when interpreting the effects of prenatal cocaine on mother-pup interaction. It is well established that rat dams will respond differently to pups that have experienced prenatal stress than to nonstressed pups [e.g., (24,28)]. Because SC cocaine injections induce skin ulcerations, such experience during pregnancy may have proven to be more stressful to the dams and their pups than saline injections. However, the severity of cocaine-induced skin lesions was very low and did not differ between the three diet groups [see companion article (36)]. Had skin lesions in the treated dams been unduly stressful to the pups it would have been predicted that the foster mothers would have interacted differently with cocaine-exposed pups in all of the diet groups. It will be recalled that differences in maternal behavior were demonstrated only towards those cocaine-treated pups born to dams fed the casein diets. Thus, these data argue against the notion that subsequent alterations in mother-pup interaction are simply related to greater prenatal stress resulting from skin lesions. With the exception of a delay in vibrissae placing, prenatally malnourished pups did not differ from well-nourished pups in the age of attainment of physical and reflexive milestones. In contrast, prenatal cocaine delayed pinna detachment, eye opening, and the emergence of auditory startle. Hence, changes in these pup characteristics following prenatal cocaine exposure could have contributed to significant alterations in the pup-directed behaviors of the foster mothers. However, the finding that chow-fed dams showed similar maternal behavior towards pups with prenatal cocaine exposure and pups with no drug exposure, in spite of differences in the physical and reflexive development of these pups, argues against this possibility. Previous studies of the effect of prenatal cocaine on the attainment of physical and reflexive milestones have produced somewhat mixed results in rats (2,4,6-9,19,21,32-35). With the exception of a study by Kunko et al. (21), which used IV cocaine administration, the majority of these studies used the SC route of administration and cocaine doses that ranged from 10 mg/kg daily (32) to 100 mg/kg daily (7) beginning either on the day of conception (19) or as late as day 15 (33) and continuing until days 18-21 of pregnancy. Across these investigations the trend has been towards a delay in develop-

ment with the effect being most reliable at higher doses of cocaine [e.g., (7)). Those physical milestones that appear to be most sensitive to developmental delay are: pinna detachment (7), ear opening (7-9), and eye opening (7,8). Hence, the present findings are consistent with these trends. Incisor eruption (7,9,33) and fur emergence (2,4,7-9) remain unaffected by cocaine. The most intensively studied reflexive milestones have been negative geotaxis (4,6,21,32-35) and surface righting (19,21,32,34,35). No significant impact of prenatal cocaine has been observed in any of these investigations except for a delay in negative geotaxis in the offspring of dams given IV cocaine (21) and a trend (p < 0.1) toward a delay in surface righting in the offspring of rats given twice daily SC cocaine injections from conception to parturition (19). Taken together, these findings suggest that the higher the exposure to cocaine (whether it be through dosage, route of administration, or timing of drug exposure relative to the stage of pregnancy), the greater is the likelihood of observing significant developmental delays in the offspring. Thus, the fact that physical and reflexive milestones were delayed in the progeny of cocaine-exposed mothers in the current investigation, despite the use of a moderate cocaine dose, is probably related to the model in which cocaine was given prior to mating and throughout pregnancy. The significance of these findings is twofold. First, the observation that prenatal cocaine altered the interaction between the pups and the 25% casein-fed foster dams raises the possibility that any long-term behavioral changes exhibited in these offspring may be unrelated to the direct effects of cocaine on brain development. Second, prenatal malnutrition did not produce a similar effect to that of prenatal cocaine on mother-pup interaction or on the attainment of milestones, nor was there an additive effect of the two insults when administered concomitantly (with the exception perhaps of passive nursing). In all cases, the independent effects of prenatal cocaine were greater than those of prenatal malnutrition. The lack of additivity of the two insults is particuIarly intriguing because cocaine produces significantly greater anorexia during pregnancy in dams on a 6% casein diet compared with dams on either a 25% casein diet or a laboratory chow diet (36).

ACKNOWLEDGEMENTS

This work was supported by grant #ROl-DA07934 from the National Institute on Drug Abuse. S. S. Almeida was the recipient of a research fellowship from Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP, Proc. 94/1538-g). We thank Eva Sabo, Sandra Fleming, and Gary Landry for excellent technical assistance during the course of this study.

REFERENCES I.

2.

3.

4.

5.

Azuma, S. D.; Chasnoff, I. R. Outcome of children prenatally exposed to cocaine and other drugs: A path analysis of three-year data. Pediatrics 92:396-402; 1993. Bilitzke, P. _I.; Church, M. W. Prenatal cocaine and alcohol exposures affect rat behavior in a stress test (The Posolt swim test). Neurotoxicol. Teratol. 14:359-364; 1992. Blumberg, M. S.; Efimova, I. V.; Alberts, J. R. Thermogenesis during ultrasonic vocalization by rat pups isolated in a warm environment: A thermographic analysis. Dev. Psychobiol. 25: 497-510; 1992. Church, M. W.; Holmes, P. A.; Overbeck, G. W.; Tilak, J. P.; Zajac, C. S. Interactive effects of prenatal alcohol and cocaine exposures on postnatal mortality, development and behavior in the Long-Evans rat. Neurotoxicol. Teratol. 13:377-386; 1991. Church, M. W.; Kaufmann, R. A.; Keenan, J. A.; Martier, S. S.;

Savoy-Moore, R. T.; Ostrea, E. M.; Subramanian, M. G.; Welch, R. A.; Zajac, C. S. Effects of prenatal cocaine exposure. In: Watson, R. R., ed. Biochemistry and physiology of substance abuse. Boca Raton: CRC Press Inc.; 1991:179-204. 6. Church, M. W.; Overbeck, G. W. Prenatal cocaine exposure in the Long-Evans rat: II. Dose-dependent effects on offspring behavior. Neurotoxicol. Teratol. 12:335-343; 1990. 7. Church, M. W.; Overbeck, G. W.; Andrzejczak, A. L. Prenatal cocaine exposure in the Long-Evans rat. I. Dose-dependent effects on gestation, mortality, and postnatal maturation. Neurotoxicol. Teratol. 12:327-334; 1990. 8. Church, M. W.; Rauch, H. C. Prenatal cocaine exposure in the laboratory mouse: Effects on maternal water consumption and offspring outcome. Neurotoxicol. Teratol. 14:313-319; 1992. 9. Ferrari, C. M.; Riley, E. P. Effect of prenatal cocaine on the

608 acquisition of cocaine-induced taste aversions. Neurotoxicol. Teratol. 16:17-23; 1994. and behavioral development of the IO. Fox, W. M. Reflex-ontogeny mouse. Anim. Behav. 13:234-241; 1965. M. A.; Kucharski, L. T. 11. Caller, J. R.; Ricciuti, H. N.; Crawford, The role of the mother-infant interaction in nutritional disorders. In: Galler, J. R., ed. Nutrition and behavior. New York: Plenum Press; 1984:269-304. M. The effects of intergenerational mal12. Caller, J. R.; Rosenthal, nutrition on the mother-pup relationship in rats. Infant Behav. Dev. 2:1-12; 1979. and ma13. Galler, _I. R.; Tonkiss, J. Prenatal protein malnutrition ternal behavior in Sprague-Dawley rats. .I. Nutr. 121:762-769; 1991. G. A.; Moody, C. A.; Spear, L. P. Prenatal cocaine 14. Goodwin, exposure increases the behavioral sensitivity of neonatal rat pups to ligands sensitive at opiate receptors. Neurotoxicol. Teratol. 15: 425-431; 1993. stress 15. Grimm, V. E.; Frieder, B. The effects of mild maternal during pregnancy on the behavior of rat pups. lnt. J. Neurosci. 35165-72; 1987. 16. Heyser, C. J.; Molina, V. A.; Spear, L. P. A fostering study of the effects of prenatal cocaine exposure: I. Maternal behaviors. Neurotoxicol. Teratol. 14:415-421; 1992. 17. Hofer, M. A.; Brunelli, S. A.; Shair, H. N. Ultrasonic vocalization responses of rat pups to acute separation and contact comfort do not depend on maternal thermal cues. Dev. Psychobiol. 26:X1-95; 1993. of mother-young contact in 18. Jans, J. E.; Leon, M. Determinants Norway rats. Physiol. Behav. 30:919-935; 1983. 19. Johns, J. M.; Means, L. W.; Means, M. J.; McMillen, B. A. Prenatal exposure to cocaine I: Effects on gestation, development, and activity in Sprague-Dawley rats. Neurotoxicol. Teratol. 14:337-342; 1992. I.; Li, L.; Pedersen, 20. Johns, J. M.; Noonan, L. R.; Zimmerman, C. A. Effects of chronic and acute cocaine treatment on the onset of maternal behavior and aggression in Sprague-Dawley rats. Behav. Neurosci. 108:107-l 12; 1994. S. E. Intravenous gesta21. Kunko, P. M.; Moyer, D.; Robinson, tional cocaine in rats: Effects on offspring development and weanling behavior. Neurotoxicol. Teratol. 15:335-344; 1993. P. G.; Smith, G. K. Thermal control of 22. Leon, M.; Croskerry, mother-young contact. Physiol. Behav. 21:793-811; 1978. 23. Levine, S.; Wiener, S. A critical analysis of data on malnutrition and behavioral deficits. Adv. Pediatr. 22:l 13-136; 1976. N. R. Effects of pre24. Meyer, J. S.; Sherlock, J. D.; MacDonald, natal cocaine on behavioral responses to a cocaine challenge on postnatal day 11. Neurotoxicol. Teratol. 14:183-189; 1992.

TONKISS 25 26 27

28

29 30 31

32

33

34

35

36

37

38 39

40

ET AL.

Moore, C. L.; Power, K. L. Prenatal stress affects mother-infant interactions in Norway rats. Dev. Psychobiol. 19:235-245; 1986. Neuspiel, D. R.; Hamel, S. C. Cocaine and infant behavior. J. Dev. Behav. Pediatr. 12:55-64; 1991. Peeke, H. V. S.; Dark, K. A.; Salamy, A.; Salfi, M.; Shah, S. N. Cocaine exposure prebreeding to weaning: Maternal and offspring effects. Pharmacol. Biochem. Behav. 48:403-410; 1994. Power, K. L.; Moore, C. L. Prenatal stress eliminates differential attention to male offspring in Norway rats. Physiol. Behav. 38: 667-67 1; 1986. Rosenblatt, J. S. Prepartum and postpartum regulation of maternal behavior in the rat. Ciba Found. Symp. 33:17-37; 1975. Seitz, P. F. D. The maternal instinct in animal subjects, I. Psychosom. Med. 22:214-226; 1958. Smart, J. L. Undernutrition, maternal behaviour, and pup development. In: Elwood, R. W., ed. Parental behaviour of rodents. New York: John Wiley & Sons; 1983:205-234. Smith, R. F.; Mattran, K. M.; Kurkjian, M. F.; Kurtz, S. L. Alterations in offspring behavior induced by chronic cocaine dosing. Neurotoxicol. Teratol. 11:35-38; 1989. Sobrian, S. K.; Burton, L. E.; Robinson, N. L.; Ashe, W. K.; James, H.; Stokes, D. L.; Turner, L. M. Neurobehavioral and immunological effects of prenatal cocaine exposure in rat. Pharmacol. Biochem. Behav. 35:617-629; 1990. Spear, L. P.; Kirstein, C. L.; Bell, J.; Yoottanasumpun, V.; Greenbaum, R.; O’Shea, J.; Hoffmann, H.; Spear, N. E. Effect of prenatal cocaine exposure on behavior during the early postnatal period. Neurotoxicol. Teratol. 11:57-63; 1989. Spear, L. P.; Kirstein, C. I,.; Frambes, N. A. Cocaine effects on the developing central nervous system: Behavioral, psychopharmacological, and neurochemical studies. Ann. NY Acad. Sci. 562:290-307; 1989. Tonkiss, J.; Shultz, P. L.; Shumsky, J. S.; Blease, S.; Kemper, T.; Galler, J. R. The effects of cocaine exposure prior to and during pregnancy in rats fed low or adequate protein diets. Neurotoxicol. Teratol. 17:571-575; 1995. Vorhees, C. V.; Butcher, R. E.; Brunner, R. I..; Sobotka, T. J. A developmental test battery for neurobehavioral toxicology in rats: A preliminary analysis using monosodium glutamate, calcium carrageenan, and hydroxyurea. Toxicol. Appl. Pharmacol. 50: 267-282; 1979. Winer, B. J. Statistical principles in experimental design. 2nd ed. New York: McGraw-Hill: 1971. Wood, R. D.; Bannoura, M. D.; Johanson, I. B. Prenatal cocaine exposure: Effects on play behavior in the juvenile rat. Neurotoxicol. Teratol. 16:139-144; 1994. Zimmerberg, B.; Gray, M. S. The effects of cocaine on maternal behaviors in the rat. Physiol. Behav. 52:379-384; 1992.