Neonatal alcohol exposure and early development of motor skills in alcohol preferring and nonpreferring rats

Neurotoxicology and Teratology, Vol. 17, No. 2, pp. 103-110, 1995 Copyright o 1995 Elsevier Science Ltd Printed in the USA. All rights reserved 0892-0362/95 $9.50 + .oO

Pergamon 0892-0362(94)00058-l

Neonatal Alcohol Exposure and Early Development of Motor Skills in Alcohol Preferring and Nonpreferring Rats TED

MELCER,’

DAVID

GONZALEZ,

CHRIS

SOMES

AND

EDWARD

P. RILEY

Department of Psychology, Centerfor Behavioral Teratology, San Diego State University, 6363 Alvarado Court, f209, San Diego, CA 92120 Received MELCER,

T.,

D. GONZALEZ,

30 August

1993; Accepted

12 August

1994

C. SOMES AND E. P. RILEY. Neonatal alcohol exposure and early development of

motor skills in alcoholpreferring and nonpreferring rats. NEUROTOXICOL

TERATOL 17(2) 103-I 10, 1995. -It has been suggested that differential sensitivity to alcohol might influence the severity of effects seen in offspring following gestational alcohol exposure and data exist to support this contention. Previously, we found that neonatal alcohol treatment produced greater increases in activity at the time of weaning in alcohol preferring (P) than by alcohol nonpreferring (NP) rat lines. Whereas these lines were genetically selected for extremes in alcohol preference they also differ on “sensitivity” to alcohol. Neonatal exposure in rats is used to model human third trimester alcohol exposure and the present study examined motor skills in P and NP ra’ts following such exposure. On postnatal days 4 through 7, P and NP rats received a daily dose of 6 g/kg in four administrations 2 h apart. The alcohol was delivered in a milk solution through an indwelling intragastric cannula. Artificially reared and normally reared controls were included in the study. At 21 and 43 days of age, rats were tested for abnormalities in gait by walking an inclined runway and for dysfunction in balance using the parallel bar test. Neonatal alcohol exposure increased falling from the bars and altered gait and these effects were similar in the P and NP lines. The parallel bar test was generally too difficult for the NP rats, limiting the utility of this test in trying to determine the effects of differential alcohol sensitivity. Thus, the present results suggest that neonatal alcohol exposure had equivalent effects on gait and balance, as measured by falling from parallel bars, in P and NP rats. These results contrast with our previous findings on activity in which P rats were more affected and argue that differential alcohol sensitivity may only influence certain behavioral outcomes. P and NP rats

Neonatal alcohol

Motor development

PRENATAL alcohol exposure can produce profound behavioral and physical problems in children and, in extreme cases, these children are often diagnosed with the Fetal Alcohol Syndrome (FAS). FAS includes physical effects such as facial abnormalities, general growth retardation, and brain damage that may manifest itself through behavioral effects such as hyperactivity, slow learning, and limited attention (24,25). Most important for the present research is the observation that only a minority of women classified as chronic alcohol abusers during pregnancy have children with full blown FAS

It has been proposed that genetic factors, including differential sensitivity to alcohol, might contribute to the variability in outcomes following gestational alcohol exposure (6,22). Work with animal models using strains or lines differing in initial sensitivity to alcohol have indeed demonstrated the interactive effects of perinatal alcohol exposure and genotype in determining fetal alcohol effects. For example, Long and Short Sleep (LS/SS) mice, bred for extremes in loss of the righting reflex after acute ethanol treatment experienced different effects of prenatal alcohol exposure. The LS mice had greater mortality, growth deficits, and weight reduction following prenatal ethanol exposure than SS mice (7, lo), indicat-

(1). Thus, women with relatively similar patterns of alcohol consumption during pregnancy have children that experience very different consequences.

ing some factor in the dam or fetus associated

with the original

’ Requests for reprints should be addressed to Ted Melter, Center for Behavioral Teratology, 6363 Alvarado Court, #209, San Diego State University, San Diego, CA 92120. 103

104 selection criterion had an influence on the outcome following prenatal alcohol exposure. Gilliam and Irtenkauf (6) utilized a reciprocal cross between C57BL/6J mice and the LS line, and indeed found that maternal genotype was a significant factor in determining fetal alcohol effects. There was a greater weight deficit and malformation rate in Fl litters carried by the C57 mothers than by LS mothers indicating a significant influence of maternal genotype on the teratogenicity of ethanol. Differential susceptibility to neonatal alcohol exposure has also been demonstrated when comparing Maudsley Reactive (MR) and M520 inbred rats (10). Specifically, neonatal alcohol produced a greater reduction in cerebellar weight in the MR strain than in the M520 strain. The M520 strain is initially more sensitive to acute alcohol treatment (as measured by blood alcohol level at regaining aerial righting reflex) than the MR strain. However, the M520 strain rapidly develops tolerance and becomes less sensitive to alcohol treatment than the MR strain (26). Recently, utilizing a neonatal alcohol exposure paradigm, we reported that alcohol preferring (P) rats were more susceptible to neonatal alcohol exposure than alcohol nonpreferring (NP) rats when assessed for locomotor activity as weanlings (21). The P and NP rats were selectively bred by Drs. T.-K. Li and L. Lumeng at Indiana University (15). These lines are derived from Wistar rats and were selectively bred based on the amount of 10% v/v ethanol consumed voluntarily when water was available simultaneously in a free choice test. In addition to the selected differences in alcohol preference, the P line is less susceptible than the NP line to the acute hypnotic and sedative effects of alcohol. The P line also develops alcohol tolerance more rapidly than the NP line (16,28). In our previous study (21), when testing occurred at the time of weaning, we found that neonatal alcohol exposure produced greater overactivity in P rats than in NP rats and that the P rats showed this overactivity at a lower level of alcohol exposure than that which affected NP rats. These findings were contrary to our expectations because we felt that the P rats, who show less susceptibility within a single session of testing and develop tolerance to alcohol more quickly than NP rats, would be less affected by the neonatal alcohol exposure. In the present study, offspring of P and NP rats were exposed to alcohol during the peak of the brain growth spurt that occurs between postnatal days (PND) 4 to 10 in the rat (5). In the human fetus, the brain growth spurt occurs during the third trimester of pregnancy and is thought to be a critical period of early brain development. This neonatal exposure paradigm in the rat thus models the effects of third trimester alcohol exposure in humans in terms of CNS development. Also, while most of the work on the P and NP rats has been done in adult animals, our personal observation is that they do differ in their behavioral response to alcohol at the age at which they are exposed using this neonatal alcohol treatment. Neonatal alcohol exposure produces substantial deficits in brain growth, particularly in the cerebellum (11,12,14). Treatments such as irradiation that selectively retard early cerebellar development produce impairments in activity, gait, and balance (2,3,13,19). Neonatal alcohol exposure in rats also produces locomotor, gait, and balance deficits (12,17,18). Problems in motor control are also seen in children with prenatal alcohol histories (4). The present experiment determined the effects of neonatal alcohol exposure on tests of gait and balance in P and NP rats. The results of the Riley et al. (21) study suggest that P rats might show greater deficits in these tasks than NP rats.

MELCER

ET AL.

METHOD

Subjects P and NP rat pups were utilized as subjects. Parent animals were supplied by Drs. T.-K. Li and L. Lumeng of the Indiana University School of Medicine Alcohol Research Center (Indianapolis, IN) and offspring were born in the San Diego State University laboratory. All rats were maintained on a 12L : 12D cycle throughout the experiment (lights on at 0700). Mating Breeder females were group housed and breeder males were individually housed in metal hanging cages. For mating, 1 female was placed in the home cage of each male in the evening and left overnight. A check for seminal plugs in the morning was used as evidence of mating and Gestation Day (CD) 0 was defined as the day a plug was found. Females that did not mate were returned to group housing. Females that mated were housed in plastic breeding cages in a temperature (70°F) and humidity (40% to 60%) controlled nursery with ad lib rat chow (Wayne Rodent-Blox) and water during the remainder of gestation. Breeding cages were checked at 8:30 a.m. and 5:00 p.m. daily for births. Twenty-four hours after a birth, litters were weighed and culled to the 10 heaviest pups (5 males; 5 females whenever possible). The surgery to implant the stomach cannula for artificial rearing (AR) was conducted on the morning of CID 26 which was almost always PND 4. Surgery and Artificial Rearing The procedure for surgical implantation of the intragastric cannula has been described in great detail previously (23). Each pup was anesthetized with halothane mixed with oxygen (50% halothane/50% oxygen) and then a stainless steel surgical wire sheathed in polyethylene tubing was maneuvered down the esophagus into the stomach. The lower tip of the wire was pushed out through the stomach and abdominal wall. The feeding cannula, made of Clay Adams polyethylene tubing (PE lo), was attached to the end of the wire protruding from the pup’s mouth. This connection was coated with an antibiotic and corn oil, and the cannula was pulled into the stomach and out through the abdominal wall. A thin plastic disc attached to the end of the cannula formed a seal against the inside of the stomach wall that prevented milk from leaking out. The cannula was then pulled through loose skin about the back of the pup’s neck and washers were attached to the cannula where it entered and exited the back of the neck. The washers absorbed any pull on the cannula at the point of the neck and prevented direct pull on the cannula in the pup’s stomach. The entire surgical procedure required about 2 min per animal. Pups recovered from the anesthetic within 3 min. The sham surgery group received identical surgical procedures except that the cannula was not installed and the pups were returned to a Sprague-Dawley dam after surgery. From PND 4 until PND 12, the artificial rearing pups were individually housed and maintained in Styrofoam cups containing hardwood chips and lined with a piece of fur. The fur reportedly minimizes behavioral depression associated with maternal deprivation (27). The cups in which pups lived were placed inside Styrofoam cups partially filled with sand to keep the housing cup dry and upright. All cups floated in an aquarium tank filled with 49OC aerated water. Every 2 days, the water temperature was decreased by 2OC until the tank tem-

NEONATAL

ALCOHOL

105

EXPOSURE

perature was 42V. Unpublished data from independent animals showed that this temperature schedule allowed the AR pups to maintain body temperatures similar to those observed in sham control pups in the litter. Each pup’s stomach cannula was connected to a tygon tubing lead attached to a syringe containing a milk-based formula which substituted for mother’s milk (23). The syringes were mounted on an infusion pump (Harvard Instruments) operated on an automated-timed schedule of one 20-min period of milk delivery every 2 h. This schedule resulted in a total of 12 feeds of milk formula daily for each pup. Neonatal Treatment Groups

Prior to surgery, pups were removed from the litter, weighed, and the heaviest pups were randomly assigned to one of the following 2 treatment groups: (a) artificial rearing with ethanol exposure (6 g/kg/day) or (b) artificial rearing without ethanol exposure. In the artificially reared control group, maltose calories were substituted isocalorically for ethanol derived calories. The remaining pups received sham surgery (no stomach cannula installed) and normal rearing with a Sprague-Dawley female that had delivered her pups at about the same time the P and NP rats were born. The use of Sprague-Dawley surrogate mothers precluded any confounding that might result from differences in maternal care between the natural P and NP mothers. Litter sizes were maintained at 8 pups using Sprague-Dawley pups where necessary. All pups received the same base milk formula during the 8 nighttime feeds (5:30 p.m. to 7:50 a.m.). Ethanol or isocaloric maltose was added to the base diet during the 4 daytime feeds (9:30 a.m. to 3:50 p.m.) for pups in the alcohol and maltose groups. The amount of milk administered in milliliters was equivalent to 33% of the weight of the AR pups within each line. Because pups from the P line weighed more than pups from the NP line, it was necessary that two separate artificial rearing apparatus be employed. Ethanol administration was limited to the 4 daytime feeds to represent a “binge drinking” model (14). On PNDs 4 through 7 pups in the ethanol group received a dosage of 6 g/kg/day. To minimize the severity of withdrawal, the ethanol-exposed pups received a dosage of 3g/kg/ day on PN days 8 and 9. On PNDs 10 and 11, all AR pups received the base milk formula without alcohol prior to being placed back into the litter, to allow recovery from mild withdrawal symptoms that were occasionally observed. Pilot studies indicated that, without this tapering procedure, mortality rates increased above 15%. The ethanol and maltose groups received isocaloric milk formula throughout the artificial rearing period (PNDs 4-l 1). Pups were weighed daily during the artificial rearing period. On the morning of PND 12, pups from all groups were earmarked for group identification and weighed. Each cannula was cut where it exited the pup’s side and all pups were bathed in a slurry of maternal feces and warm water before being returned to a dam. This slurry procedure virtually eliminated pup rejection by the dam. Pups were returned to Sprague-Dawley dams that were not their biological mothers. Each of these surrogate dams received a new litter of 8 pups including equivalent numbers of pups from the 3 treatment groups whenever possible. Behavioral Tests Parallel bars. Pups were tested for balance on PND 21 and 43. The sample sizes were as follows: P line: 6 g/kg-male =

7, female = 11; Maltose-male = 10, female = 9; Shammale = 8, female = 8; NP line: 6 g/kg-male = 14, female = 14; Maltose-male = 11, female = 9; Sham-male = 12, female = 16. The apparatus consisted of two parallel horizontal bars (0.3 cm in diameter and 60 cm long) fastened into grooved slots on two end platforms (19 x 16.5 cm) set in a frame 60 cm above a cushioned floor. The initial gap between bars was 2 cm for 21-day-old rats and 4 cm for 43-dayold animals. Each animal was first placed on the end platform for 60 s to familiarize it with the safe area. The pup was then placed on the bars halfway between the end platforms with its left paws on one bar and its right paws on the other bar. The pup’s task was to walk toward the end platform with its left paws stepping along one bar and right paws stepping along the opposite bar. An animal reached criterion at each gap if it made 4 consecutive steps with alternate hindpaws. On trials where criterion was reached, the pup was allowed to continue walking along the bars until it reached the end platform. Falling off the bars, swinging under the rods, or any other behavior such as placing left and right paws on the same bar terminated the trial. The trial also terminated if the rat remained immobile for 10 consecutive s. This occurred on 4% of trials for all groups on PND 21 whereas on PND 43, NP rats remained immobile on 17% of trials versus 9% for P rats. These results were not influenced by sex or neonatal treatment at either age. After reaching criterion by making 4 successful steps on any trial, the gap was increased 1 cm on the following trial. If pups made fewer than 4 successful steps on a trial then the gap width remained the same on the following trial. Each pup received 5 such trials per day and an additional 1 minute long 6th trial. The 1-min trial was included to give the animals additional opportunity to learn the task, because even if the animal fell, it was replaced on the bars to obtain one full minute of practice. The pups were placed on the end platform for 10 s between trials. When pups failed to successfully traverse any gap during a day’s testing they were tested for only 1 additional day. Pups were tested for no more than 5 consecutive days. We recorded the number of gaps successfully traversed (0,1,2, . . . ), the total number of successful steps with hind limbs, as just defined, and the number of falls and swings by each animal. The latter two measures were calculated as a proportion of total trials because some animals qualified for more test trials than others. Pups were weighed following testing on PNDs 23 and 43. Gait Test

The same pups that received the parallel bar test also received a test for gait or walking patterns on PNDs 21 and 43. The sample sizes were as follows for PND 21/PND 43 (sample sizes differ between PND Zl/PND 43 because some pups did not readily walk up runway and were discarded from analysis at that age): P line: 6 g/kg-male = 7/7, female = 7/10; Maltose-male = 10110, female = 9/8; Sham-male = 7/ 8, female = 6/8; NP line: 6 g/kg-male = 11113, female = 13/12; Maltose-male = lO/ll, female = 8/9; Sham-male = 9/12, female = 13/16. The apparatus was a 10 x 10 x 92 cm runway inclined 10 degrees and the top was covered with wire mesh to prevent escape. To familiarize pups with the runway on PND 20, they were placed at the bottom of the incline and allowed to explore the runway. The entrance to the runway at the bottom of the incline was closed off with a barrier after the rat was placed inside. The rat could exit by

106

MELCER

walking up the length of the runway. There was no barrier at the top of the inclined runway and the rat could walk out onto a platform overlooking its home cage. Most pups readily walked up the runway, out the open end and then were placed back in the home cage. Pups that did not reach the end of the runway after 5 min were placed at the end and allowed to walk out and then returned to their home cage. Prior to the gait assessment the experimenter rubbed vaseline on the pup’s hind paws and placed the pup at the bottom of the runway now lined with a strip of paper. Pups walked up the runway and barriers were dropped at several points along the runway to prevent backtracking. If pups did not walk on the first trial they received a second trial. After a successful run, the experimenter removed the paper from the runway and dusted it with charcoal to outline footprints. A blind observer selected 4 to 6 consecutive footprints from each record for scoring. The focal point for measurements of footprints was a notch in the footpad between the second and third digits on each hind foot (Fig. 1). A computer scoring system (Jandel Scientific, Model 2210, Montgomeryville, PA) was used to identify the location of each notch and measured 3 parameters of gait: (a) Stridelength: the distance in cm between consecutive footprints of same hind paw (Fig. 1; Line A); (b) Stride width: the distance in cm between a right footprint and a straight line formed by consecutive left footprints. This measure begins at the notch in one footprint and extends at a right angle to the line formed by consecutive footprints of opposite paws. Step angle, the angle (Fig. 1, angle b) formed by the intersection of lines C and A in Fig. 1 was also assessed. The scores used in the analysis of variance (ANOVA) were the sine of the measured angles. One observer scored 5 randomly selected records twice. The correlation coefficient between first and second observations for all 3 measures was at least: r = .99. These measures were sensitive to neonatal alcohol exposure in previous research using Sprague-Dawley rats (17). RESULTS

Pup Data Table 1 shows pup body weights from PNDs 4 through 10. The means shown include both sexes (n’s = 7 to 16 of each

PATH DIRECTION 4

STRIDEWIDTH -

TABLE

ET AL.

1

MEAN AND SE FOR BODY WEIGHTS (GRAMS) THE ARTIFICIAL REARING PERIOD

DURING

Neonatal Treatment Postnatal Day P

6 g/kg

Maltose

Sham

rats 4 5 6 7 8 9

10 NP rats 4 5 6 7 8 9 10

11.01 12.11 13.64 15.07 16.41 17.39 18.58

(.56) (.27) (.29) (.35) (.39) (.38) (.38)

10.92 12.36 13.78 15.47 16.82 18.17 19.29

(.23) (.20) (.18) (.17) (.26) (.24) (.32)

10.05 11.45 13.71 15.78 17.82 19.66 22.54

(.37) (.36) (.52) (.63) (.69) (.72) (.56)

9.34 10.60 12.06 13.55 14.76 15.95 17.93

(.ll) (.12) (.16) (.21) (.24) (.28) (.29)

9.65 10.75 12.04 13.77 15.39 16.87 18.95

(.18) (. 19) (. 19) (.19) (.25) (.45) (.39)

9.00 9.91 11.93 13.77 15.69 17.34 20.34

(.22) (.18) (.29) (.34) (.42) (.38) (.62)

sex per group). As shown, the AR pups weighed more than the sham pups at the start of the artificial rearing period due to the selection criterion for surgery, but by the end of the AR procedure the sham animals weighed more than the AR animals. An ANOVA on these data with sex, neonatal treatment (alcohol, maltose or sham), and line (P or NP) as between-subjects factors and postnatal age (PND 4-10) as a within-subjects factor indicated a significant day x treatment x line interaction, F(12, 702) = 2.13, p < 0.05. We tested for simple effects of dose and line on each day separately to breakdown the interaction and used Newman-Keuls (p < 0.05) to isolate the source of any significant simple effects. There were significant effects of line on body weights on all days as P rats weighed more than NP rats. Neonatal treatment produced significant effects on PNDs 4, 5, 8, 9, and 10. On PND 4 and 5 the AR groups (6 g/kg and maltose) were equivalent and both AR groups weighed more than the sham group. On PND 8 the maltose and sham groups had equivalent weights and both of these control groups weighed more than the 6 g/kg group. Finally, on PND 9 and 10 the 6 g/kg groups weighed less than both control groups and the maltose groups weighed less than the sham group. None of the line x dose interactions were significant in any of the individual test days. The overall ANOVA also indicated a significant main effects of line, F( 1, 117) = 42.72, p < 0.01, with P pups weighing more than NP pups, and sex, F(l, 117) = 8.83 p < 0.01, with males weighing more than females. Finally, pup weight increased between PNDs 4 and 10, F(6, 702) = 1706.82, p < 0.01. Test Weights

STRIDE LENGTH FIG.

1. Illustration of footprint record from gait test. The 3 parameters of gait are shown: stridelength (line A), stridewidth (line at right angle to line A), and step angle (angle formed by intersection of lines C and A).

Body weights were recorded after behavioral testing on PND 23 and 43. An ANOVA was performed on body weights including test day (PND 23/PND 43) as a within-subjects factor and sex, neonatal treatment and line as between-subjects factors. This analysis showed a test day x treatment interaction, F(2, 117) = 4.16, p < 0.05. Separate analyses were conducted to test for main effects of treatment on PND 23 and 43

NEONATAL

ALCOHOL

EXPOSURE

107

and Newman-Keuls test were used where necessary to isolate the source of treatment effects. There was a significant treatment effect on PN 23 [sham, M (mean score) = 44.63 g, SE = 1.21; maltose, M = 40.43 g, SE = 0.93; 6 g/kg, M = 39.85 g, SE = 0.821 and sham pups weighed (p’s < 0.05) more than AR pups (maltose and 6 g/kg) and there were no differences between the AR groups. On PND 43 there was also a treatment effect and the sham group (M = 174.11 g, SE = 4.61) weighed more (p’s < 0.05) than the maltose (M = 162.72 g, SE = 4.21) and the 6 g/kg groups (M = 156.72 g, SE = 3.88). The AR groups (maltose and 6 g/kg) were equivalent in weight. The overall ANOVA also revealed a significant day x sex interaction, F(l, 117) = 73.61 g, p < 0.01. There was no effect of sex on PND 23 (males, M = 42.23 g, SE = 0.99; females, M = 41.12 g, SE = 0.71) but there was a substantial sex effect on PND 43 (males, M = 182.09 g, SE = 3.72; females, M = 148.09 g, SE = 1.84). Finally, there was a significant effect of line, F(l, 117) = 11.53, p < 0.01, as P rats weighed more than NP rats on both PND 23 (P, M = 44.88 g, SE = .99; NP, M = 39.41 g, SE = 0.65) and on PND 43 (P, M = 168.42 g, SE = 3.68; NP, M = 161.66 g, SE = 3.39). Behavioral Tests Parallel bars. The sample sizes for analyses were 7 to 16 of each sex per group. Figures 2, 3, and 4 show the results of the parallel bar tests. Data were combined across sex and test days (PND 21 and 43) because there were no systematic effects of sex or age on performance. As shown in Fig. 2, there was a rather large effect of artificial rearing on the mean number of gaps traversed in the P rats. This artificial rearing effect appears much smaller in the NP animals. Also, there does not appear to be any substantial differences between the alcohol and maltose groups on this measure. The ANOVA on these data revealed a significant line ic treatment interaction, F(2, 117) = 3.86, p < 0.05 and a subsequent Newman-Keuls test (p < 0.05) indicated that the P sham group traversed more

PARALLEL

Wkg

BAR

TEST

Maltose NEONATAL

Sham

TREATMENT

FIG. 2. Mean (+ SE) gaps traversed during the parallel bar tests as a function of rat line and neonatal treatment. The means shown are combined across sexes and test ages (PND 21 and 43).

PARALLEL

BAR

TEST

1.5

P E v) z

1.0

i=

-

E 8 E

0.5

z :

0.0

I

2 Wh

Maltose

NEONATAL

Sham

TREATMENT

FIG. 3. Mean (+SE) proportion steps (successful steps/trials) on parallel bars as a function of rat line and neonatal treatment. The means shown are combined across sexes and test ages (PND 21 and 43).

gaps than all other groups. The overall ANOVA also indicated that P rats traversed more gaps than did NP rats, F(l, 117) = 20.46,~ < 0.01. Figure 3 illustrates the mean step scores (successful steps/ trials). In the P rats, the alcohol group had fewer steps than maltose and sham groups. In the NPs, the step scores of the AR groups appear similar to the NP sham scores. The ANOVA indicated a significant treatment x line interaction in step scores, F(2, 117) = 3.14, p < 0.05. A Newman-Keuls test (p < 0.05) showed that both the P sham and P maltose groups differed from the P alcohol group and that the two control groups (maltose and sham) did not differ from each other. There were no differences among the NP groups. The overall ANOVA also indicated a significant effect of line, F(l, 117) = 9.52, p < 0.01. There were no main effects or interactions involving sex or age in the analysis of step scores. Figure 4 shows mean proportion falls + swings per trial. Data were combined across sex and age, as these variables did not interact with treatment. As shown, the alcohol groups fell/swung off the bars more often than controls. An ANOVA on these data indicated a significant effect of treatment, F(2, 117) = 11.01, p < 0.01, and the subsequent Newman-Keuls test (p c 0.05) indicated that the alcohol groups had significantly higher scores than controls. The overall ANOVA also indicated a sex x line x age interaction, F(1, 117) = 5.02,~ c 0.05. The Newman-Keuls test showed that P rats (M = 0.17) had a higher proportion falls/swings than NP rats (M = 0.10) on PND 21, whereas on PND 43, P females (M = 0.21) had more falls swings than P males and NP animals of both sexes (M = 0.12). Walking/gait. The sample sizes for analysis were 6 to 13 of each sex per group for the analyses of PND 21 scores. Figure 5 shows these results. The upper graph shows that alcohol pups had shorter stride lengths than controls and this was supported by a significant effect of treatment, F(2, 98) = 4.91, p c 0.01, and a subsequent Newman-Keuls test (p < 0.05) confirmed that the alcohol groups had shorter stride-

MELCER ET AL. PARALLEL

BAR

TEST

n P 0

WW

Maltose NEONATAL

NP

I

Sham

TREATMENT

coordination. Previous work from our lab (21) has indicated that the two lines have equivalent blood alcohol levels following the neonatal alcohol exposure regime used in this study so we know that the two lines were equally exposed to alcohol. Although neonatal alcohol exposure did not have a differential effect on the P and NP lines, it did cause alterations in gait replicating the general findings of Meyer et al. (17). Alcohol-exposed animals of both lines had a shorter stridelength and a smaller stridewidth than artificially reared (maltose groups) and normally reared controls (sham groups) at 21 days of age. These alterations did not appear to be due to the artificial rearing procedure, because the maltose groups did not differ from the sham group on any gait measure. Another possibility for the observed differences in gait might be related to differences in body weight. However, this also seems unlikely because the alcohol group and the maltose group were similar in weight on PND 21 yet were significantly different on the gait measure. Furthermore, the maltose group weighed less than the sham animals, yet showed equivalent gait measures. Neonatal alcohol exposure produced no effect on step

FIG. 4. Mean (+ SE) proportion falls/swings ({falls + swings}/trials) on the parallel bars as a function of rat line and neonatal treatment. The means shown are combined across sexes and test ages (PND 21 and 43).

a 00

lengths than the control groups. This effect of alcohol did not interact with line, nor did sex have any significant effect. The lower graph shows the results for stride width on PND 21. Again, there was a significant effect of treatment on stridewidth, F(2, 98) = 3.51, p < 0.05. A Newman-Keuls test (p < 0.05) showed than alcohol pups had significantly shorter stride widths than both sham and maltose control pups. The ANOVA also showed that the P rats (M = 3.12 cm, SE = 0.04) had greater stride width than NPs (M = 2.82 cm, SE = 0.05), F(1, 98) = 19.39, p < 0.01. The ANOVA on the sine of step angle scores revealed no significant effects of sex, line, or treatment. The walking data for PND 43 (sample sizes 7 to 16 per group) were analyzed separately from PND 21 because a number of subjects failed to walk the runway on PND 21 but did walk on PND 43. In contrast to PND 21, there were no significant effects of neonatal treatment on any of the 3 walk measures. The mean stridewidth of males (M = 4.26 cm, SE = 0.055) was less than that for females (M = 4.48 cm, SE = 0.058), F(1, 112) = 5.1, p < 0.05, and the sine of step angle of males (M = 42.32, SE = 0.61) was greater than for females (M = 39.9, SE = .56), F(1, 112) = 8.6,~ < 0.01. Finally, the P rats (M = 42.15, SE = 0.60) had larger step angles than NP rats (M = 40.39, SE = 0.58), F(1, 112) = 5.09,p < 0.05. DISCUSSION

The results indicate that neonatal alcohol exposure produced deficits in motor function in both the P and NP rats. The alcohol exposure produced equal deficits in P and NP rats in the test of gait and in balance as measured by falling/ swinging from the parallel bars. The other measures of balance (stepping, gaps traversed) were problematic because the extremely low scores in the NP line may have resulted in a floor effect obscuring any effects due to alcohol. Thus, previous findings of greater susceptibility of the P rats to neonatal alcohol exposure did not generalize to these tasks of motor

II

6.00

Maltose

Wkg 3.50

w

3.00

P K k f

275

$

2.50

Sham

WP

‘1

0

I

NP

t

Wkg

Maltose

NEONATAL

Sham

TREATMENT

FIG. 5. Mean (+SE) stride length (upper graph) and stride width (lower graph) as a function of rat line and neonatal treatment in 21-day-old rats. The means shown are combined across sexes. See Fig. 1 and text for explanation of gait measures of stridelength and stridewidth.

NEONATAL

ALCOHOL

EXPOSURE

109

angle on PND 21 and no differences were found between treatment groups on any of the gait measures at PND 43. Although these findings support the general conclusion reached by Meyer et al. (17), there are some inconsistencies between that study and the current study. In the Meyer et al. study, differences in gait between the groups were not apparent at 21 days of age but only became evident when testing was done at 45 and 65 days of age. In the present study, differences were apparent at 21 but not at 43 days of age. There are a number of significant methodological differences between these two studies that might account for these differences. These include the breed of rat (Long Evans vs. P and NP rats derived from Wistar rats), differences in the alcohol administration protocol (continuous exposure vs. binge exposure), and the dose of ethanol administered daily [rats in the Meyer et al. study (17) received the high alcohol dose for 2 more days than rats in the present study]. Both lines of rats were also tested for the ability to walk across two horizontal, parallel bars at PND 21 and 43. Alcohol exposure equally increased the rate of falling/swinging off the bars in P and NP rats at both test ages. The alcohol groups fell/swung off the bars about twice as often as the controls. However, it is difficult to reach any significant conclusions about the differential effects of alcohol in this task, as it was a relatively difficult one, and most groups successfully traversed only 1 gap on average. The P alcohol group did have fewer successful steps on the bars than P control groups whereas no differences in the step scores were found among the NP treatment groups. Again, the step scores of the NP rats were too low to detect a possible alcohol effect and the absence of an alcohol effect in the NP line was probably due to a floor effect. Therefore, the step scores cannot be used to determine differential susceptibility of P and NP rats to neonatal alcohol exposure. In this task, there was also no substantial improvement between PND 21 and 43, but this finding is difficult to interpret because the initial gap width at PND 43 was greater than at PND 21. As mentioned earlier, Riley et al. (21) found that neonatal alcohol exposure produced greater overactivity in P rats than in NP rats, indicating that some characteristic associated or

linked with the selection criterion of voluntary ethanol consumption (e.g., differential alcohol sensitivity, differential ontogeny of the serotonergic system) might be responsible for this difference. However, the present findings indicate equivalent susceptibility of the P and NP lines to neonatal alcohol effects in tests involving motor coordination. It may be that whatever characteristic that interacted with ethanol exposure in the previous study to cause this differential sensitivity of the P and NP lines as measured in the open field does not influence gait or coordination to any great extent. For example, Riley et al. (21) speculated that because the development of the serotonergic system occurs at different rates in the P and NP rats, that this system might be differentially affected by the neonatal alcohol exposure. If such was the case, alcohol might have profound affects on activity but leave behaviors not controlled by the affected serotonergic systems unaffected or equally compromised between the lines. Thus, one could find differential effects for certain behaviors but not for others. It also warrants mentioning that because replicate lines do not exist for the P and NP lines that the differential effect on locomotor activity might simply be a serendipitous finding and one not related to any characteristic associated with the genes controlling differences in voluntary ethanol consumption. Previous studies comparing selectively bred lines or inbred strains indicate that differences in alcohol-related phenotypes may indeed be important determinants of prenatal or neonatal alcohol effects. Additional research needs to explore whether recent findings in the P and NP rats with regard to neonatal alcohol exposure generalize from young to mature and aged rats and from measures of motor performance to more complicated tasks involving learning and attention. It would also be critical to determine whether neonatal alcohol exposure differentially affects brain development in the P and NP lines. ACKNOWLEDGEMENTS

This work was supported in part by Grants AA0692 to E.P.R. and AA07611 to T.K. and Alcoholism.

Li from

the National

Institute

on Alcohol

Abuse

REFERENCES 1. Abel, E. L.; Sokol, R. J. Incidence of fetal alcohol syndrome and economic impact of FAS-related anomalies. Drug Alcohol Depend. 1951-70; 1987. 2. Altman, J. Effects of interference with cerebellar maturation on the development of locomotion. An experimental model of mental retardation. In: Buchwald, N. A., Brazier, M. A. B., eds. Brain mechanisms in mental retardation. New York: Academic Press; 1975:41-91. 3. Bruner, R. L.; Altman, J. Locomotor deficits in adult rats with moderate to massive retardation of cerebellar development during infancy. Behav. Biol. 9:169-188; 1973. 4. Conry, J. Neuropsychological deficits in fetal alcohol syndrome and fetal alcohol effects. Alcohol: Clin. Exp. Res. 14:650-655; 1990. 5. Dobbing, J.; Sands, J. Quantitative growth and development of the human brain. Arch. Dis. Child. 48:757-767; 1973. 6. Gilliam, D.; Irtenkauf, K. Maternal genetic effects on ethanol teratogenesis and dominance of relative embryonic resistance to malformations. Alcohol: Clin. Exp. Res. 14:539-545: 1990. 7. Gilliam, D. M.; Ketch, L.; Dudek, B. C.; Riley, E: P. Ethanol teratogenesis in selectively bred long-sleep and short-sleep mice: A comparison to inbred C57BL/6J mice. Alcohol Clin. Exp. Res. 13:667-672; 1989a. 8. Gilliam, D. M.; Ketch, L.; Dudek, B. C.; Riley, E. P. Ethanol

9.

10.

11.

12.

13.

14.

15.

teratogenesis in mice selected for differences in alcohol sensitivity. Alcohol 5:513-519; 1989b. Gilliam, D. M.; Stilman, A.; Dudek, B. C.; Riley, E. P. Fetal alcohol effects in long- and short-sleep mice: Activity, passive avoidance and in utero ethanol levels. Neurotoxicol. Teratol. 9: 349-357; 1987. Goodlett, C. R.; Gilliam, D. M.; Nichols, J.; West, J. Genetic influences on brain growth restriction induced by developmental exposure to alcohol. Neurotoxicol. 10:321-324; 1989. Goodlett, C. R.; Mahoney, J. C.; West, J. R. Brain growth deficits following a single day of alcohol exposure in the neonatal rat. Alcohol 6:121-126; 1989. Goodlett, C. R.; Thomas, J. D.; West, J. R. Long-term deficits in cerebellar growth and rotarod performance of rats following “binge-like” alcohol exposure during the neonatal brain growth spurt. Neurotoxicol. Teratol. 13:69-74;1991. Gruner, J. A.; Altman, J.; Spivak, N. Effects of arrested cerebellar development on locomotion in the rat. Cinematographic and electromyographic analysis. Exp. Brain Res. 40:361-373; 1980. Kelly, S. J.; Pierce, D. R.; West, J. R. Microencephaly and hyperactivity in adult rats can be induced by neonatal exposure to high blood alcohol concentrations. Exp. Neurol. 96:580-593; 1987. Lumeng, L.; Hawkins, T. D.; Li, T.-K. New strains of rats with

110

16.

17.

18.

19. 20.

21.

22.

MELCER alcohol preference and nonpreference. In: Thurman, R. G., Williamson, J. R., Drott, H. R., Chance, B., eds. Alcohol and aldehyde metabolizing systems, vol. 3. New York: Academic Press; 1977:531-544. Lumeng, L.; Waller, M. B.; McBride, W. J.; Li, T.-K. Different sensitivities to ethanol in alcohol preferring and nonpreferring rats. Pharmacol. Biochem. Behav. 16:125-130; 1982. Meyer, L. S.; Ketch, L. E.; Riley, E. P. Alterations in gait following ethanol exposure during the brain growth spurt in rats. Alcohol: Clin. Exp. Res. 14:23-27; 1990a. Meyer, L. S.; Ketch, L. E.; Riley, E. P. Neonatal ethanol exposure: Functional alterations associated with cerebellar growth retardation. Neurotoxicol. Teratol. 12:15-22; 1990b. Mullenix, P.; Norton, S.; Culver, B. Locomotor damage in rats after x-irradiation. Exp. Neurol. 48:310-324; 1975. Murphy, J. M.; McBride, A. W.; Wlodarek, C. A.; McBride, W. J.; Lumeng, L.; Li, T.-K. Differences in the ontogeny of locomotor activity between the alcohol-preferring P and nonpreferring NP rats. Alcohol: Clin. Exp. Res. 16:406; 1992. Riley, E. P.; Barron, S.; Melter, T.; Riley, E. P. Alterations in activity following alcohol administration during the third trimester equivalent in P and NP rats. Alcohol: Clin. Exp. Res. 17: 1240-1246; 1993. Riley, E. P.; Lochry, E. A. Genetic influences in the etiology of fetal alcohol syndrome. In: Abel, E. L. ed. Fetal alcohol syn-

23.

24.

25.

26.

27.

28.

ET

AL.

drome, 3, animal studies. Boca Raton, FL: CRC Press; 1982:113130. Samson, H. H.; Diaz, J. Effects of neonatal ethanol exposure on brain development in rodents. In: Abel, E. L. ed. Fetal alcohol syndrome, III Animal studies. Boca Raton, FL: CRC Press; 1982: 131-150. Streissguth, A. P. The behavioral teratology of alcohol: Performance, behavioral, and intellectual deficits in prenatally exposed children. In: West, J. R., ed. Alcohol and brain development. New York: Oxford University Press; 1986:3-44. Streissguth, A. P.; Barr, H. M.; Martin, D. C. Alcohol exposure in utero and functional deficits in children during the first four years of life. In: Porter, R.; G’Conner, M.; Whelan, J. eds. Mechanism of alcohol damage in utero. Ciba Foundation Symposium 105; London, Pittman; 1984:176-196. Tabakoff, B.; Culp, S. G. Studies on tolerance development in inbred and heterogeneous stock national institutes of health rats. Alcohol: Clin. Exp. Res. 5:495-499; 1984. Terry, L. M.; Scheinman, J. A.; Hall, W. G. Response deficits in isolate reared rats. Paper presented at the annual meetings of the International Society for Developmental Psychobiology; 1987. Waller, M. B.; McBride, W. J.; Lumeng, L.; Li, T.-K. Initial sensitivity and acute tolerance to ethanol in the P and NP lines of rats. Pharmacol. Biochem. Behav. 19:683-686; 1983.