The Effects of Litter Size during Gestation and Lactation on Rat Development Prior to Weaning

FRANCINE WEHMER

Department o f Psychology Wayne State University

Detroit, Michigan

KAI-LIN CATHERINE JEN

Physiological and behavioral development of rats was affected by prenatal nutrition and postnatal litter size. Prenatal nutrition was manipulated by combining differences in maternal nutrition with variation of prenatal litter size produced by pre-mating isolation of 1 uterine horn. The pups were reared in postnatal litters of either 4 or 12. During the 1st postnatal week, development and free behavior were affected only by the prenatal treatments. During the 2nd postnatal week, both prenatal and postnatal effects were observed. Development and behavior during the 3rd postnatal week reflected only postnatal litter size. Brain weight at weaning was influenccd by both the prenatal and postnatal treatments.

Experimental studies with laboratory animals have revealed the consequences of inadequate diet during the developmental period on many physiological and behavioral variables (Chow & Lee, 1964; Davis, Hargen, & Chow, 1972; Dobbing, Hopewell, & Lynch, 1973; Vore & Ottinger, 1970). Considerably less research exists on the effects of overnutrition on animal development.

One approach used to produce prenatal overnutrition has been to reduce prenatal litter size, analogous to the technique used in postnatal research in which litters are reduced to small sizes by culling at birth. Pre-mating isolation of one uterine horn reduces the number of fetuses by half through prevention of fertilization of eggs released by one ovary. With only half the number of fetuses competing for available nutrients, the pups are heavier at birth with increased brain parameters (van Marthens & Zamenhof, 1969). (In our laboratory, pups from prenatal half-litters are from 19 to 24% heavier than base line controls, consisting of normally sized prenatal litters whose dams were on ad lib food throughout pregnancy [Jen, 1977; Ryan, 19761 .)

By simultaneously manipulating prenatal litter size and maternal nutrition during pregnancy, the researcher can produce large differences in prenatal nutrition between groups of pups without the unwanted pup debility and mortality which is frequently

Reprint requests should be sent to Dr. Francine Wehmer, Department of Psychology, Wayne State University, Detroit, Michigan 48202, U.S.A.

Received for publication 14 January 1977 Revised for publication 23 May 1977 Developmental Psychobiology, 11 (4):353-360 (1978) @ 1978 by John Wiley & Sons, Inc. 0012-1630/78/0011-0353$01 .OO

354 WEHMER AND JEN

seen when such differences are produced by severe restriction of maternal diet during pregnancy (Altman, Das, Sudarshan, & Anderson, 1971). The aim of this sludy was to examine the effects of prenatal litter size and maternal nutrition during pregnancy on pup development during the preweaning period, and to assess the role postnatal litter size contributes towards eliminating or maintaining prenatal effects.

Method

Subjects

Seventy-five subjects were the progeny of 24 female rats (Rattus nowegicus) of the Sprague-Dawley strain (Spartan Research Institute, Haslett, Mich.). The females weighed 250 to 275 g when they arrived at this laboratory. They were housed individually in stainless steel cages (27 x 23 x 30 cm) and kept in a colony room with a 12-hr light:12-hr dark cycle. Unless otherwise noted, they were fed and watered ad lib. On Day 15 of pregnancy, the females were placed in plastic breeding cages (23 x 45 x 20 cm) containing wood chips and strips of paper toweling.

Surgery

Two weeks before mating, 12 females were anesthetized with sodium pentobarbital supplemented with atropine. An incision about 2 cm long was made 1 cm off midline in the lower abdomen. The left utero-tuba1 junction was pulled out onto a sterile cloth. This junction was cut with a cold cautery scalpel and was then gently replaced in the abdominal cavity. The other uterine horn was untouched. The remaining females were sham operated. After surgery, the females were returned to their honie cages and tetracycline (Polyotic; American Cyanamid, Princeton, N.J.) was added to their drinking water for 1 week.

Apparatus

Open field I : A 30.5 x 30.5-cm square board, marked with black. concentric circles, served as a small open field for the infant rats. The outside circle was 23.3 cm in diameter. Each adjacent circle was reduced .65 cm in diameter.

Open field / I : A large wooden gray field, 78 x 91.5 x 28 cm, was used. Its floor was partitioned into 80 rectangles by thin black lines.

Lashley III water maze: The maze measured 44 x 76 x 17.8 cm with a start box and a goal box (10 x 21 x 17.8 cm) attached to the 1st and last alleys, respectively (Jen & Wehmer, 1977). A guillotine door separated the start box from the maze The subjects could excape the maze by climbing a ladder in the goal box. The water was 10 cm deep and kept at room temperature. The water was changed every other day.

Procedure

Two weeks after surgery, male rats of the same strain were caged with each female. The day of mating was determined by the presence of a vaginal plug and this

LITTER SIZE AND DEVELOPMENT 355

day was designated Day 1 of preganancy. Starting on Day 10 of pregnancy, sham-operated females were put on a dietary restriction of 75% of the food eaten by ad lib fed females. Thus, we produced 2 prenatal nutritional conditions, the prenatal small litter group composed of subjects from small in utero litters whose dams were fed ad lib (Group Pre-ON) and a group of subjects from normal-sized litters whose dams were undernourished (Group Pre-UN). At birth, all subjects were randomly fostered within prenatal conditions to ad lib fed dams who had delivered their own litters within the previous 24 hr. Postnatal litter sizes were adjusted to form 2 groups: small litters of 4 or larger litters of 12, forming the Post-SL and Post-LL groups, respectively. The result was a 2 x 2 factorial design varying prenatal nutrition and postnatal litter size. The pups were sexed, weighed, and individually identified by toe clipping.

An additional 21 animals were sacrificed at birth, nose-anus length measured, and the cerebrum and cerebellum removed and weighed to the nearest .1 mg.

Dams and pups were weighed every day. The day of appearance of the developmental landmarks of unfolding of the external pinnae, eye opening, ear opening, and incisor eruption were recorded. The animals were tested as follows: (1) When the pups were 5 to 7 days old, they were individually placed in the center of the small field for two min. Activity level, head lifting, and pivoting were recorded as described in Jen and Wehmer (1977). (2) When the pups were 14-18 days old, free-fall righting was observed by dropping the subject, back-downwards from a height of 30 cm over a cotton padded container. (3) When the subjects were 18-20 days old, they were placed in the large open field for 2 min. Head lifting and rearing with or without support were recorded. (4) When the subjects were 21-25 days old, they were placed in the Lashley I11 water maze for 1 trial a day and time and errors were recorded. (See Table 1.)

TABLE 1. Chronological Schedule of Tests and Main Effect of Nutrition on Postnatal Development.

1 1

2-4 5 -7

6-1 1 14-18 14-18 16-21 17

18-20

21 21

21-25

~~ ~ ~~ ~~

Observation Main Effect of Nutrition

Birth weight Prenatal Body length Prenatal Unfolding of pinnae Prenatal Open field

Head lifting Prenatal Pivoting Prenatal Activity level Postnatal

Incisor eruption Pre- and postnatal Eye opening Pre- and postnatal

Ear opening Postnatal Body weight Postnatal Open field

Standing with support Postnatal Cerebellar weight Post natal Body length Postnatal Lashley 111 water maze Postnatal

Free-fall righting Postnatal

356 WEHMER AND JEN

At 21 days of age, 30 pups were sacrificed and body weight, body length, cerebral weight, and cerebellar weight were recorded.

Results

Prenatal Litter Size

Isolation of 1 uterine horn successfully reduced prenatal litter size to half o E the potential number of fetuses. At birth, average litter size in the Pre-ON group was 6 pups (range: 3-9 pups), whereas in the Pre-UN group it was 12 pups (range: 8-17 PUPS).

Birth Weight and Length

At birth subjects from the Pre-ON group were 29% heavier and 13% longer than subjects from the Pre-UN group ( t = 11.71, d f = 3 8 , p < .001 and t = 6.56, d f = 19, p < .001, respectively).

Brain Weight a t Birth and Weaning

At birth the brains (minus cerebellum) of the Pre-ON group were 267:) heavier than the brains from the Pre-UN group ( t = 13.42, d f = 19, p < .001). The cerebella from the Pre-ON group were 24% heavier than the cerebella from the Pre-U’N group ( t = 4.72, df= 19, p < .001).

Analysis of variance (ANOVA) revealed significant effects of weaning of prenatal (F= 114.3,c/f’= 1/26,p <.001)and postnatal (F=43.O5,df= 1 / 2 6 , p < . O O l ) treatment on weight of the brain (minus cerebellum), as well as a Prenatal x Postnatal treatment interaction (F = 10.54, df = 1/26, p < .001). Post hoc analysis revealed that subjects in the Pre-ON-Post-SL Group had considerably heavier brains than subjects in the Pre-UN-Post-LL Group ( p < .001) and their brains were also heavier than :subjects froin the Pre-ON-Post-LL Group ( p < .OOl) and the Pre-UN-Post-SL Group ( p < .001). Subjects in the Pre-ON-Post-LL Group also had heavier brains than subjects in the Pre-UN-Post-SL Group ( p < .Ol) . Of the total variance in cerebral weight, 58.35%. was due to the prenatal treatment, 21.65% to the postnatal treatment, and 4.9%:. was contributed by the interaction (w2 ; Winer, 1971). The group means in grams were: Pre-ON-Post-SL, 1.63 ; Pre-ON-Post-LL, 1.37; Pre-UN-Post-SL, 1.26; Pre-UN-Post-LL, 1.18.

Cerebellar weight at weaning demonstrated only a postnatal treatment effect (F = 12.03, df = 1/26, p < .OOl). The cerebella of the Post-SL groups were 15% heavier than the cerebella of the Post-LL groups.

Body Weight and Length a t Day 17

Body weight at Day 17 was affected by postnatal treatment, with subjects in the Post-SL groups being 28% heavier than subjects in the Post-LL groups (F = 22.56, d j =

LITTER SIZE AND DEVELOPMENT 357

1/26, p < .001). Body length of the Post-SL groups were 10% greater than that of the Post-LLgroups ( F = 19.66,df= 1/26,p < .001).

Rate of Growth

An ANOVA with a repeated measure factor demonstrated that rate of growth was different between groups. Body weights on Days 1, 5 , 9, 13, and 17 were analyzed. Both prenatal treatment and postnatal treatment affected growth (F = 7.82, d f = 1/40, p < .01 and F = 7.33, df= 1/40, p < .01, respectively). Subjects in the Post-SL groups grew faster than subjects in the Post-LL groups, as revealed by a significant Postnatal treatment x Days interaction (F = 38.44, df = 4/160, p < .001). All the subjects gained weight over days (F = 1285.1 1, df = 4/160, p < .OOl).

Infant Development

Unfolding of the external pinnae occurred significantly earlier in the Pre-ON group (F = 194.42, df = 1/40, p < .OOl). No postnatal treatment or interaction effects were observed.

The age at which both upper incisors erupted was affected by the Prenatal and Postnatal treatments (F = 46.89, df = 1/40, p < .001 and F = 4.14, df = 1/40, p < .05, respectively). Subjects in the Pre-ON groups cut their incisors earlier than subjects in the Pre-UN groups and subjects in the Post-SL groups cut their incisors earlier than subjects in the Post-LL groups. The prenatal treatment contributed to 49.4% of the total variance; postnatal treatment contributed only 3.4% of the variance.

Eye opening was also affected by both prenatal and postnatal treatments (F = 15.56, df = 1/40, p < .001 and F = 12.97, d f = 1/40, p < .001, respectively). Subjects in the Pre-ON and Post-SL groups opened their eyes earlier than the subjects in the Pre-UN and Post-LL groups, respectively. Of the total variance, 21% was associated with prenatal treatment, whereas 17.2% was associated with postnatal treatment.

Ear opening was solely determined by the postnatal treatment (F = 20.83, dJ”= 1/40, p < .001). Subjects from the Post-SL groups had their ears opened earlier than pups from the Post-LL groups.

Behavioral Development

Behavior in the open field over the 1st 3-day testing period (Days 5 through 7 postnatal) revealed significant prenatal treatment effects on head lifting (F = 5.94, df = 1/40, p < .025) and pivoting (F = 4.04, df = 1/40, p < .06). Subjects from the Pre-ON group performed at a higher level than subjects in the Pre-UN group. Activity level was only affected by the postnatal treatment. Subjects from the Post-SL groups were more active than the subjects from the Post-LL groups (F = 4.91, df = 1/40, p < .005). Pivoting and exploratory activity increased for all groups over the 3 testing days (F = 10.41, d f = 2/80, p < .005 and F = 5.80, d f = 2/80, p < .025, respectively).

Free-fall righting increased over days (F = 19.66, df 2/80, p < .OOl) and was achieved more frequently by subjects in the Post-SL groups (F = 14.76, df = 1/40, p < .OOl). A significant Prenatal Treatment x Days interaction (F = 4.56, df = 2/80, p <

358 WEHMER AND JEN

.005) showed that subjects from the Pre-ON groups showed a greater increasc in frce-fall righting over days than the othcr groups.

The open field test conducted on Days 18-20 showed that subjccts from the Post-SL groups reared with support more frequently than subjects fi-om the Post-LL groups ( I ; = 18.53, d,f = 1 /40, p < .001). Rearing with support decreased over days (F = 9.44, d f = 2/80, p < .005). Rearing without support showed a triple interaction between Prenatal treatment, Postnatal treatment, and testing Days (F = 4.33, u'f = 2/80, p < .025 j, which is not interpretable. Grooming in the open field revealed a n interaction between prenatal and postnatal treatment (F = 4.48, df = 1/40, p < . O M ) . Subjects from the Prc-UN-Post-SL group groomed more than subjects from the other groups. Grooniing in the open field increased over days (F = 21.85, df = 2/80, p < .00 1 ).

Learning d t i r i q the 5 days of trials in the Lashley 111 water maze was affected by postnatal treatinen!. Subjects from the Post-SL groups required less time (F = 9.71, d'f = 1/20, p < .001) and made fewer errors (F = 11.75, d f = 1/20, p .005) than subjects from the Post-LL groups. Time and errors decreased for all groups over Days (fi' = 40.45. df' = 4/80, p < ,001 and F = 23.02, df' = 4/80, p < .001). (Table 1 sutnniarizes the tests and the nutritional main effect associated with it.)

Discussion

Tlic data indicate that body weight was affectcd by both the prenatal and postnatal treatmcnts, but the influence of these treatments was differcnt depending on the chronological age of the subjects. Prior t o 2 weeks of age, prenatal nutrition was the rnajor factor influencing body weight, but after the 2nd week only postnatal nutrition affectcd growth up t o weaning. The treatments not only influenced body fat stores, thcy also affectcd skelctal growth as measured by body length. At birth, the Pre-ON group was significantly longer tlian the Pre-UN group. At weaning, skeletal growth was soleiy influenced by postnatal factors, indicating that catch-up growth occurred in the Pre-UN-Post-SL group. These data support and extend the work of Willianis and Hughes (1975) who examined catch-up growth following postnatal undernutrition. As seen in this study, catch-up growth can also be achieved in subjects experiencing different nutritional treatments during the prenatal period.

Similar trends were also observed in the developmental mcasurcs. Excluding thc brain weight a t weaning, the prenatal main effects of nutritional treatments were present for the 1st 2 weeks after birth, but were not observed after that time.

The only long-term effect of the prenatal treatment used in this study was seen in weight of the brain (minus cerebellum), which was heavier in weanlirigs from the Pre-ON groups than in wcanlings from the Pre-UN groups. Of considerable interest was the effect of placing the Pre-ON group in small postnatal litters. The results indicate that brain weight in the Pre-ON-Post-SL group was considerably heavier than in any other group, as reflected in the prcnatal x postnatal interaction. These data indicate that cerebral weight a t weaning is very much dependent upon prenatal nutritional factors, although postnatal factors play some role in the process of brain growth magiifying group differences when postnatal nutritional treatments continue the subject along the same directions established by prenatal nutritional conditions and

LITTER SIZE AND DEVELOPMENT 359

attenuating group differences when postnatal treatments move the subjects onto a nutritional plane different from that experienced during the prenatal period. Unlike cerebral weight, cerebellar weight at weaning was affected only by postnatal nutritional treatment, although significant weight differences were found at birth. Because the rat cerebellum undergoes its major growth spurt during the postnatal period (Dobbing, Hopewell, & Lynch, 1971), the cerebellum may have undergone the same type of catch-up growth that was reflected in body weight and body length at weaning.

Whether the increased cerebral weight in the Pre-ON-Post-SL group is functionally related to increased adaptive behavior is not clear. No prenatal effect was evident in the analysis of Lashley 111 maze learning. Possibly, as for the growth data reported by Chow and Stephan (1971) and Ryan (1976), the effects of the prenatal treatments may reemerge when the subjects are adult. Prenatal nutritional effects of adult maze learning have been reported p o r e & Ottinger, 1970), whereas others have found only a postnatal effect (Smart, Dobbing, Adlard, Lynch, & Sands, 1973). Small postnatal litter size may produce inferior learning (Frafikovi, 1970). The possible confounding of postnatal nutrition with differences in maternal care should not be ignored in any discussion of such data (Grota & Ader, 1969). An alternate hypothesis is that the brain weight of these subjects is predictive only of the larger adult size these animals d 1 achieve (Ryan, 1976).

In summary, the data indicate that relatively large differences in prenatal nutrition do not affect development beyond the 1st postnatal week independent of postnatal litter size. Brain weight at weaning is the only variable which reflects prenatal influences beyond the 1st 2 postnatal weeks.

Notes

This research was supported by a Wayne State University Faculty Research Award to the 1st author in 1974. Some of these data were presented at the annual meeting of the American Psychological Association, Chicago, Illinois, 1975.

References

Altman, J . , Das, G. D., Sudarshan, K., and Anderson, J. B . (1971). The influence of nutrition on neural and behavioral development. 11. Growth of body and brain in infant rats using different techniques of undernutrition. Dev. Psychobiol., 4 : 55-70.

Chow, B. F., and Lee, C. J. (1964). Effect of dietary restriction of pregnant rats on body weight gain of the offspring. J. Nutr., 82: 10-18.

Chow, B . F., and Stephan, J . D. (1971). Fetal undernourishment and growth potential. Nutr. Rep . Int., 4 : 245-255.

Davis, R. L., Hargen, S. M., and Chow, B. F. (1972). The effect of maternal diet on the growth and metabolic patterns of progeny. Nutr. Rep . Int., 6 : 1-7.

Dobbing, J., Hopewell, J. W., and Lynch, A. (1971). Vulnerability of developing brain. VII. Permanent deficit of neurons in cerebral and cerebellar cortex following early mild undernutrition. Exp . Neurol., 32: 439447.

Fraiikova’, S. (1970). Behavioral responses of rats to early overnutrition. Nutr. Metabol., 12:

Grota, L. J. , and Ader, R . (1969). Continuous recording of maternal behavior in Rattus norvegicus. 228-239.

Anim. Behav., 17 : 722-129.

360 WEHMER AND JEN

Jen, K.-L. C. (1977). Selected behavioral patterns of dietary obese rats. Unpublished doctoral dissertation, Wayne State University.

Jen, K.-L. C., and Wehmer, F. (1977). Effects of thyroxine on development of undernourl\hed infant rats. Dev. PsychobioL, 10: 87-95.

v m Marthens, E., and Zanenhof , S. (1969). Deoxyribonucleic acid of neonatal rat cerebrum increased by operative restriction of litter size. Exp. Neurol., 23: 214-219.

Ryan, V. (1976). Effect of prenatal and postnatal nutrition on the developtncnt, behavior, and physiology of the rat. Unpublished doctoral dissertation, Wayne State University.

Smart, J. L., Dobbing, .1., Adlard, B. P. F., Lynch, A., and Sands, J . (1973). Vulnerability o f developing brain: Relative effects of growth restriction during the fctal and suckling pcriods on behavior and brain composition of adult rats. J. Nufr., 103: 1327-1338.

Vore, D. A., and Ottinger, D. R. (1970). Maternal food restriction: Effects on offspring development, learning and a program of therapy. Dcv. Psycho/., 3 : 337-342.

Williams, J. P. G., and Hughes, P. C. R. (1975). Catch-up growth in rats undernourished for different periods during the suckling period. Growrh, 39: 179-193.

Wincr, R . J . (1971). Sraristical Pvitzciples in Expcrimeritd Design. (2nd ed.) New York: MtSraw-Ilill.