- Email: [email protected]
Animal models for small-for-gestational-age (SGA) neonates and infants-at-risk (IAR)
Developmental Brain Research, 10 (1983) 221-225
221
Elsevier
Animal Models for Small-For-Gestational-Age (SGA) Neonates and Infants-At-Risk (IAR) O. RESNICK and P. J. MORGANE
WorcesterFoundation for Experimental Biology, 222 Maple Ave., Shrewsbury, MA 01545 (U.S.A.) (Accepted May 31st, 1983)
Key words: protein malnutrition- - small-for-gestational-age- - infants-at-risk - - brain sparing - - minimal brain dysfunction - retardation of brain growth - - retardation of body growth - - brain development - - rehabilitation following malnutrition- maternal nutrition and pregnancy - - 'failure to thrive' syndrome
In this study, the body weight changes seen in rat dams and brain and body weights seen in pups as sequelae of either an overt or a hidden form of chronic protein deprivation have been examined. In the overt model, the 6% casein-diet pups showed SGA brain and body weight deficits and irreversible central and peripheral chemical changes at birth. In contrast, the hidden form of prenatal deprivation did not result in brain and body weight deficits in the 8% casein-diet pups at birth, but their SGA-like irreversible peripheral and central chemical profiles categorizes them as IAR-neonates. In cross-fostering experiments it was shown that pups from dams fed an 8% casein-diet cross-fostered onto 25% casein-diet dams at birth showed no significant differences in brain or body weight at weaning, whereas pups from 6% casein-diet dams cross-fostered onto 25% casein-diet dams at birth continued to show significantly lower brain and body weights at weaning. These studies show that the 8% casein-diet rat at birth is an IAR model because of its irreversible peripheral and central chemical changes dating form birth, even though the body and brain weights are essentially the same as the pups from 25% casein-diet dams. On the other hand, the pups from 6% casein-diet dams are SGA animals at birth and these brain and body weight deficits at birth cannot be completely reversed at weaning even when the pup is cross-fostered to a 25% casein-diet dam at birth. In both the IAR (8%) and SGA (6%) neonates, the following central and peripheral chemical changes were demonstrated: markedly elevated plasma non-esterified fatty acids (NEFA); markedly elevated regional brain levels of phenylalanine, norepinephrine (NE), tryptophan, 5-hydroxytryptamine (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA). These chemical changes were not reversed by cross-fostering the 6% or 8% casein-diet neonates by 25% casein-diet dams. Such chemical data at birth were not necessary to characterize the 6% casein-diet pups as SGA, because of their marked decrease in body and brain weights. This is also true for the 6% casein diet pups cross-fostered by 25% casein-diet dams. The possible implications of these findings in humans is discussed.
INTRODUCTION Although the deleterious effects of prenatal malnutrition on the developing central nervous system of humans have been k n o w n for a long time, until fairly recently relatively few studies have used the rat as a model for these brain deficits. In 1970, however, Winick 15 reported that in utero malnourished rats displayed many of the same brain alterations as those found in the prenatally deprived small-for-gestational-age (SGA) h u m a n infant. Other investigators have since extended his findings, but few of the studies have closely paralleled the h u m a n condition. Since most animal studies begin the dietary restrictions at conception, these tend to depict an acute stage of deprivation rather than the chronic u n d e r n u trition typically seen all through childhood and adult0165-3806/83/$03.00 (~) 1983 Elsevier Science Publishers B.V.
hood in women from low socioeconomic groups. In order to simulate the nutritional status of these women, we have initiated in the rat dietary restrictions (8% casein-diet) to the females well prior to conception 14. However, because the protein deficits of this diet are relatively mild and are compensated for in calories by the addition of excess carbohydrates, the 8% pups do not show the typical S G A weight losses at birth. This appearance of normalcy is, however, highly deceptive since we have reported irreversible central and peripheral chemical changes in the 8% n e w b o r n pups 7. Some of these chemical changes are similar to those found in the S G A h u m a n neonate m. This 'hidden' form of abnormality relates to the 8% pups displaying a reduced capacity to learn new tasks in adolescence 8 which appears to be similar to some forms of minimal brain dysfunction (MBD)
222 found in children. Because m a n y ' n o r m a l ' birth weight children show signs of M B D t3, it is possible
TABLE I
that part of their mental disabilities may be due to the
The effects of different levels of dietary protein on female weight gain before and during pregnancy
same type of presumably mild fetal nutritional deficits occurring in the 8% rats. Thus, this apparent lack
Values given are mean weights ± S.E. Diet groups
of overt symptoms in these pups (and probably many children) places them in a different category of malnutrition, i.e. they are infants at risk (IAR).
Time interval
In this paper, we describe the weight changes that
Initial weights (g) Pregravid weights (g) 1 week 2 weeks 3 weeks 4 weeks 5 weeks Pregnancy weights (g) 1 week 2 weeks 3 weeks
accompany this I A R syndrome in the rat to show how it remains hidden in the dams and offspring through parturition. To show that the 8% casein-diet may represent the borderline whereby caloric compensation can maintain the appearance of dietary adequacies, we have recently devised a malnutritional paradigm in which the protein content of the maternal diet was reduced to a level (6% casein) where extra calories provided by the isocaloric diet could not sustain the nutritional requirements of the dam. Placing female rats on this diet 5 weeks prior to pregnancy resulted in their offspring being S G A at birth. In depicting the differences between the overt and hidden form of maternal malnutrition, we have also described parallels from these animal models to the human condition in order to emphasize the potential consequences of either on the in utero growth and de-
6% Casein
8% Casein
25% Casein
(n = 20)
(n = 13)
(n = 28)
193 + 2
192 ± 2
190 + 2
201 ± 208 + 214 ± 224 ± 233 ±
210 ± 219 + 227 + 240 ± 252 +
3 3c 3c 3b 3b
219 ± 3 234 + 4 242 ± 4 255 + 6 267 + 6
(n = 10) 280 + 5 317 ± 6 370 + 7
(n = 21) 287 + 6 321 ± 6 379 ± 6
3d 3d 3d 3d 3d
(n = 15) 238 + 5a 258 + 6a 293 ± 8d
~P < 0.05; cp < 0.01; dp < 0.001 for 6% or 8% females vs 25% females, 2-tailed, t-tests. were cross-fostered at birth to dams fed the 25% casein-diet. These pups were designated as 6/25 or 8/25, respectively. RESULTS
velopment of the h u m a n neonate. The relationship between the maternal nutritional METHODS The diets and rearing procedures were the same as those previously described7, 9. Briefly, virgin female Sprague-Dawley rats were fed isocaloric (4.3 kcal/g diets containing either normal (25% casein), low (8% casein), or very low (6% casein) amounts of protein. This dietary paradigm was started 5 weeks prior to mating and continued through gestation and lactation. Body weights of the dams were recorded once a week during the pregravid period and pregnancy. Following birth, the weights of the dams, the n u m b e r of pups and their body and brain weights were recorded. Litters were then culled to 8 pups and redistributed to dams of the same diet giving birth on that day. O n e litter of pups from each diet group was sacrificed and their brains weighed. To assess the influence of adequate lactational nutrition on pups born to dams fed either the 6% or 8% casein-diets, two litters of pups from each of these dietary treatments
status on the 3 diets is shown in Table I. From this table it can be seen that in the pregravid period the body weights of the dams on the 6% or 8% caseindiet are significantly lower than those on the 25% diet with the 6% diet group being significantly lower in body weight than the 8% diet group at all pregra-
TABLE II Brain and body weights at birth for pups from dams fed the 3 different protein diets Values given are means + S.E. Diet Body weight (g) Brain weight (mg)
6% Casein
8% Casein
25% Casein
4.95 ± 0.2c (n = 154)
6.00 ± 0.2 (n = 119)
6.20 ± 0.1 (n = 253)
241 ± 12b (n = 8)
280 ± 11 (n = 8)
275 ± 10 (n = S)
bp < 0.05, cp < 0.001 for 6% vs 8% or 25% pups, 2-tailed ttests.
223 TABLE III Influence o f maternal gestational and lactational diets on growth o f pups at weaning (day 21) Dietary reversal
Body weight (g) Brain weight (rag)
6/25 a (n = 16)
8/25 (n = 16)
25/25 (n = 16)
40.9 -i- 0.5b
65.0 _+ 1.0
64.9 + 0.3
1128 4- 25b
1393 4- 32
1410 + 50
aThe slash mark separates the prenatal/postnatal nutritional status of the pups, i.e. pups cross-fostered at birth to 25% casein diet. Values given are mean weights _+S.E. bp < 0.001 for 6/25 pups vs 8/25 or 25/25 rats, 2-tailed t-tests.
vid weeks. However, during pregnancy there is no significant difference between the 8% and 25% groups but there is a highly significant difference between the 6% and the 8% and the 25% diet groups. The effect of the 3 diets on the subsequent body and brain weights of the pups is shown in Table II. From this table it can be seen that on the 8% casein-diet the body and brain weights of the pups do not differ significantly from those on the 25% casein-diet. However, pups from mothers fed the 6% casein-diet show significantly lower body and brain weights at birth. The effects of the dietary reversals at birth on brain and body weights at weaning (21 days) are shown in Table III. Cross-fostering pups from 8% casein-diet mothers to normal 25% casein (8/25) diet mothers resulted in these pups having body and brain weights not significantly different from pups of 25% dams fed a 25% casein-diet during lactation (25/25). On the other hand, cross-fostering pups from 6% casein-diet dams at birth onto dams fed a 25% casein diet (6/25) did not produce pups at weaning with normal body or brain weights. Rather, both body and brain weights were significantly lower than either the 8/25 or 25/25 groups (Table III). DISCUSSION Although the diets are isocaloric, the inadequate protein content of the 6% casein-diet was responsible for the 48% decreases in pregnancy weight gains of these dams compared to those fed the 25% diet. Thus, this condition of maternal malnutrition was responsible for the SGA birth weights of their pups be-
ing almost 20% less than normal and brain weights of some 14% less than normal. In humans, pregnancy weight gains that are about one-half of the normal values are usually accompanied by an increased number of SGA births. This would be expected since fetal growth and development depends on access to maternal nutrients and on the amounts of reserves amassed during early pregnancy that can be mobilized later to ensure fetal needs. Because the latter is determined by the maternal nutritional status, the poorly nourished, low pregnancy weight gain females like the 6% rat dams will have inadequate reserves available to maintain normal fetal growth. Another important, but often ignored, factor in fetal growth retardation is the pregravid nutritional status of the female. By starting the dietary restrictions at conception in most animal studies of malnutrition, the dams have pregravid stores available to support some degree of fetal development which is a situation that is considerably different from that of chronically undernourished women. For such women, however, the pregravid nutritive status makes an important contribution to the subsequent birth weight of their infants. For example, Eastman and Jackson 3 showed that among women with small pregnancy weight gains (5 kg) the incidence of SGA births was 1.5% for those with pregravid weights of 72 kg as compared to the 10% SGA births for women with pregravid weights of 54 kg. Thus, high pregravid weights can furnish reserves to support a good birth weight even when the pregnancy weights are low. The lack of high pregravid weights in the 6% dams (Table I, some 33% less weight than the dams on the 8% or 25% casein-diets) coupled with their small pregnancy weights makes them similar to a large group of chronically undernourished women. Interestingly, in common with SGA humans, the 6% rat pups also showed smaller deficits in brain weights compared to body weights at birth. This brain 'sparing' effect, however, does not prevent losses in neurons, glia, and myelin, as well as impaired dendritic differentiation that occurs in the in utero deprived offspring of either species~0. Although all of these brain deficits can affect intelligence, it is, perhaps, the damage to cortical dendritic development which may have a greater impact on the learning potential of SGA progeny rather than the
224 concurrent decreases in myelin or cell numbers~. In this regard, quantitative Golgi analyses of neurons in several neocortical areas in the 6% pups are presently underway in our laboratory as an extension of our earlier studies of neuronal development in subcortical areas (nucleus raphe dorsalis and locus coeruleus) 1,2 in 8% casein-diet rats. In addition, the SGA offspring display peripheral chemical alterations which can affect brain metabolism. These include hypoalbuminemia and a hyperlipolysis of adipose tissue resulting in increased plasma non-esterified fatty acids (NEFA). We have previously reported that this is the mechanism whereby the levels of tryptophan and serotonin are increased in the brains of the animals on the 6% and 8% casein-diets 4-5,9. While the peripheral and central chemical changes are quantitatively similar in both the 6% and 8% pups, they are much more pronounced in the 6% pups 6. Lastly, like many SGA infants, the 6% pups showed postnatal growth deficits (or stunting) as long-term sequelae of the in utero deprivation. Although they received adequate lactational nutrition by being fostered at birth to 25% dams (labeled as 6/25 rats in Table III), they had markedly lower than normal brain and body weights (20% and 37% decreases, respectively) at weaning (Table III). Thus, the insults of severe in utero deprivation in the rat do not end at birth. In spite of the appearance of both a normal pregnancy and normally appearing neonates for the 8% rats, the pups can be categorized as I A R at birth. Not only do they show many of the peripheral chemical alterations seen in SGA humans and in the 6% pups, but they also have markedly higher than normal brain serotonin and norepinephrine and their precursor amino acids 4,7. Moreover, this I A R syndrome can continue to remain hidden in the 8% pups if they receive adequate postnatal nutrition. While the poor milk production of their dams resulted in the 8% pups having markedly lower than normal brain and body weights (13% and 60% decreases, respectively) at weaning 9, fostering of these rats at birth to 25% dams assured their normal weaning growth indices (labeled as 8/25 rats in Table III). However, these postnatal nutritional adequacies were unable to effect rehabilitation of their in utero determined SGAlike peripheral chemical imbalances and impaired brain neurochemistry 5.
On the whole, these data show the fallacy of assesing the nutritional status of a female and its effect on the offspring by her pregravid weight, her pregnancy weight gain, or by the birth weight of the neonate. Although there is a strong association between birth weights and pregravid weights, they are not necessarily a reliable indicator of good maternal health or the intake of essential nutrients. This is especially true among mildly overweight women from low socioeconomic groups in this country whose pregravid weights may reflect a type of caloric compensation as a replacement for much of their essential protein needs due to the escalating price of the latter. Also, a high pregnancy weight gain in women gives no assurance that the fetus is receiving the proper nutrients to ensure adequate growth and development. As in the non-gravid state, a high weight gain can result from a high caloric diet which may be deficient in protein. For example, not only do these low protein deficits cause pregnant women to show a rapid accumulation of fluid (up to 7 kg) in a short period of time, but may also account for the high percentage (6--7%) of their children who show abnormal growth and neurological functioning at one year of age tl . Our findings of abnormal brain metabolism in the 8% pups as sequelae of mild maternal protein deftcits 7 may possibly be applicable to many normal birth weight infants. Human birth weights above 2.5 kg are defined as normal even though they are below the optimum of 3.5 kg. Because mild fetal deficits may occur in infants with birth weights between these values, this may account for the 6-13% of them who later showed signs of MBD or mental retardation~2. Thus, many children who are considered 'normal' at birth because their birth weights are within the so-called normal range may be at risk. Lastly, the lack of breast-feeding in this country may allow this IAR syndrome to remain hidden during infancy. Using our 8% pups as a model, it was noted that: (1) this mild form of chronic protein lack only becomes noticeable during early lactation when the body weights of these pups become markedly lower than normal; and (2) if they were nursed by 25% dams, their postnatal body weight gains were normal. Thus, the use of infant formulas rather than breast-feeding in our society would allow for the 'normal' postnatal growth of the infant and would miss any signs of low-level maternal protein lacks
225 which could appear during early lactation if the mother nursed her baby. Since the normal weight gain 8/25 pups still showed the altered brain and peripheral chemical changes of IAR offspring at weaning, it becomes less surprising to find many 'normal' birth weight children who have low IQs or show signs
REFERENCES 1 Diaz-Cintra, S., Cintra, L., Kemper, T., Resnick, O. and Morgane, P. J,, The effects of protein deprivation on the nucleus raphe dorsalis: a morphometric Golgi study in rats of three age groups, Brain Research, 221 (1981) 243-255. 2 Diaz-Cintra, S., Cintra, L., Kemper, T. and Morgane, P. J., The effects of protein deprivation on the nucleus locus coeruleus: a morphometric Golgi study in rats of three ages groups, Soc. Neuroscience, (12th Annual Meeting), Minneapolis, MN, 1982, p. 640. 3 Eastman, N. J. and Jackson, E., Weight relationships in pregnancy. 1. The bearing of maternal weight gain and prepregnancy weight on birth weight in full term pregnancies, Obstet. Gynecol. Survey, 23 (1968) 1003-1025. 4 Miller, M., Leahey, J. P., Stern, W. C., Morgane, P. J. and Resnick, O., Tryptophan availability: relation to elevated brain serotonin in developmentally protein-malnourished rats, Exp. Neurol., 57 (1977) 142-157. 5 Miller, M. and Resnick, O., Tryptophan availability: the importance of prepartum and postpartum dietary protein on brain indoleamine metabolism in rats, Exp. Neurol., 67 (1980) 298-314. 6 Miller, M., Hasson, R. and Resnick, O., Developmental protein malnutritionin rats: adaptive changes in brain indoleamine metabolism in small-for-gestational-age offspring, Soc. Neuroscience, (11th Annual Meeting), Los Angeles, CA (1981) p. 72. 7 Morgane, P. J., Miller, M., Kemper, T., Stern, W., Forbes, W., Hall, R., Bronzino, J., Kissane, J., Hawrylewicz, E. and Resnick, O., The effects of protein malnutritionon the
of MBD in the absence of perinatal brain damage. ACKNOWLEDGEMENTS
These studies were supported by Grant HD-06364 from the National Institute of Child Health and Human Development.
developing central nervous system in the rat. Neurosci. Biobehav. Rev., 2 (1978) 137-230. 8 Resnick, O., Miller, M., Forbes, W., Hall, R., Kemper, T., Bronzino, J. and Morgane, P. J., Developmental protein malnutrition: influences on the central nervous system of the rat, Neurosci. Biobehav. Rev., 3: 233-246. 9 Resnick, O., Morgane, P. J., Hasson, R. and Miller, M., Overt and hidden forms of chronic malnutrition in the rat and their relevance to man, Neurosci. Biobehav. Rev., 6 (1982) 55--75. 10 Ross, P. and Cramoy, C., Nutrition and pregnancy. In M. Winick (Ed.), Nutrition Pre- and Postnatal Development, Plenum Press, New York, 1979, pp. 133-228. 11 Shanklin, D. R. and Hodin, J., Maternal Nutrition and Child Health, Charles C. Thomas, Springfield, 1979. 12 Sinclair, J. C., Saigal, S. and Yeung, C. Y., Early postnatal consequences of fetal malnutrition. In M. Winick (Ed.), Nutrition and Fetal Development, John Wiley, New York, 1974, pp. 147-174. 13 Singer, J. E., Westpal, M. and Niswander, K., Relationship of weight gain during pregnancy to birth weight and infant growth and development in the first year of life, Obstet. Gynecol., 31 (1968)417-423. 14 Stern, W. C., Miller, M., Forbes, W. B., Morgane, P. J. and Resnick, O., Ontogeny of the levels of biogenic amines in various parts of the brain and in peripheral tissues in normal and protein malnourished rats, Exp. Neurol., 49 (1975) 314-326. 15 Winick, M., Nutrition and nerve cell growth, Fed. Proc., 29 (1970) 1510-1515.
221
Elsevier
Animal Models for Small-For-Gestational-Age (SGA) Neonates and Infants-At-Risk (IAR) O. RESNICK and P. J. MORGANE
WorcesterFoundation for Experimental Biology, 222 Maple Ave., Shrewsbury, MA 01545 (U.S.A.) (Accepted May 31st, 1983)
Key words: protein malnutrition- - small-for-gestational-age- - infants-at-risk - - brain sparing - - minimal brain dysfunction - retardation of brain growth - - retardation of body growth - - brain development - - rehabilitation following malnutrition- maternal nutrition and pregnancy - - 'failure to thrive' syndrome
In this study, the body weight changes seen in rat dams and brain and body weights seen in pups as sequelae of either an overt or a hidden form of chronic protein deprivation have been examined. In the overt model, the 6% casein-diet pups showed SGA brain and body weight deficits and irreversible central and peripheral chemical changes at birth. In contrast, the hidden form of prenatal deprivation did not result in brain and body weight deficits in the 8% casein-diet pups at birth, but their SGA-like irreversible peripheral and central chemical profiles categorizes them as IAR-neonates. In cross-fostering experiments it was shown that pups from dams fed an 8% casein-diet cross-fostered onto 25% casein-diet dams at birth showed no significant differences in brain or body weight at weaning, whereas pups from 6% casein-diet dams cross-fostered onto 25% casein-diet dams at birth continued to show significantly lower brain and body weights at weaning. These studies show that the 8% casein-diet rat at birth is an IAR model because of its irreversible peripheral and central chemical changes dating form birth, even though the body and brain weights are essentially the same as the pups from 25% casein-diet dams. On the other hand, the pups from 6% casein-diet dams are SGA animals at birth and these brain and body weight deficits at birth cannot be completely reversed at weaning even when the pup is cross-fostered to a 25% casein-diet dam at birth. In both the IAR (8%) and SGA (6%) neonates, the following central and peripheral chemical changes were demonstrated: markedly elevated plasma non-esterified fatty acids (NEFA); markedly elevated regional brain levels of phenylalanine, norepinephrine (NE), tryptophan, 5-hydroxytryptamine (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA). These chemical changes were not reversed by cross-fostering the 6% or 8% casein-diet neonates by 25% casein-diet dams. Such chemical data at birth were not necessary to characterize the 6% casein-diet pups as SGA, because of their marked decrease in body and brain weights. This is also true for the 6% casein diet pups cross-fostered by 25% casein-diet dams. The possible implications of these findings in humans is discussed.
INTRODUCTION Although the deleterious effects of prenatal malnutrition on the developing central nervous system of humans have been k n o w n for a long time, until fairly recently relatively few studies have used the rat as a model for these brain deficits. In 1970, however, Winick 15 reported that in utero malnourished rats displayed many of the same brain alterations as those found in the prenatally deprived small-for-gestational-age (SGA) h u m a n infant. Other investigators have since extended his findings, but few of the studies have closely paralleled the h u m a n condition. Since most animal studies begin the dietary restrictions at conception, these tend to depict an acute stage of deprivation rather than the chronic u n d e r n u trition typically seen all through childhood and adult0165-3806/83/$03.00 (~) 1983 Elsevier Science Publishers B.V.
hood in women from low socioeconomic groups. In order to simulate the nutritional status of these women, we have initiated in the rat dietary restrictions (8% casein-diet) to the females well prior to conception 14. However, because the protein deficits of this diet are relatively mild and are compensated for in calories by the addition of excess carbohydrates, the 8% pups do not show the typical S G A weight losses at birth. This appearance of normalcy is, however, highly deceptive since we have reported irreversible central and peripheral chemical changes in the 8% n e w b o r n pups 7. Some of these chemical changes are similar to those found in the S G A h u m a n neonate m. This 'hidden' form of abnormality relates to the 8% pups displaying a reduced capacity to learn new tasks in adolescence 8 which appears to be similar to some forms of minimal brain dysfunction (MBD)
222 found in children. Because m a n y ' n o r m a l ' birth weight children show signs of M B D t3, it is possible
TABLE I
that part of their mental disabilities may be due to the
The effects of different levels of dietary protein on female weight gain before and during pregnancy
same type of presumably mild fetal nutritional deficits occurring in the 8% rats. Thus, this apparent lack
Values given are mean weights ± S.E. Diet groups
of overt symptoms in these pups (and probably many children) places them in a different category of malnutrition, i.e. they are infants at risk (IAR).
Time interval
In this paper, we describe the weight changes that
Initial weights (g) Pregravid weights (g) 1 week 2 weeks 3 weeks 4 weeks 5 weeks Pregnancy weights (g) 1 week 2 weeks 3 weeks
accompany this I A R syndrome in the rat to show how it remains hidden in the dams and offspring through parturition. To show that the 8% casein-diet may represent the borderline whereby caloric compensation can maintain the appearance of dietary adequacies, we have recently devised a malnutritional paradigm in which the protein content of the maternal diet was reduced to a level (6% casein) where extra calories provided by the isocaloric diet could not sustain the nutritional requirements of the dam. Placing female rats on this diet 5 weeks prior to pregnancy resulted in their offspring being S G A at birth. In depicting the differences between the overt and hidden form of maternal malnutrition, we have also described parallels from these animal models to the human condition in order to emphasize the potential consequences of either on the in utero growth and de-
6% Casein
8% Casein
25% Casein
(n = 20)
(n = 13)
(n = 28)
193 + 2
192 ± 2
190 + 2
201 ± 208 + 214 ± 224 ± 233 ±
210 ± 219 + 227 + 240 ± 252 +
3 3c 3c 3b 3b
219 ± 3 234 + 4 242 ± 4 255 + 6 267 + 6
(n = 10) 280 + 5 317 ± 6 370 + 7
(n = 21) 287 + 6 321 ± 6 379 ± 6
3d 3d 3d 3d 3d
(n = 15) 238 + 5a 258 + 6a 293 ± 8d
~P < 0.05; cp < 0.01; dp < 0.001 for 6% or 8% females vs 25% females, 2-tailed, t-tests. were cross-fostered at birth to dams fed the 25% casein-diet. These pups were designated as 6/25 or 8/25, respectively. RESULTS
velopment of the h u m a n neonate. The relationship between the maternal nutritional METHODS The diets and rearing procedures were the same as those previously described7, 9. Briefly, virgin female Sprague-Dawley rats were fed isocaloric (4.3 kcal/g diets containing either normal (25% casein), low (8% casein), or very low (6% casein) amounts of protein. This dietary paradigm was started 5 weeks prior to mating and continued through gestation and lactation. Body weights of the dams were recorded once a week during the pregravid period and pregnancy. Following birth, the weights of the dams, the n u m b e r of pups and their body and brain weights were recorded. Litters were then culled to 8 pups and redistributed to dams of the same diet giving birth on that day. O n e litter of pups from each diet group was sacrificed and their brains weighed. To assess the influence of adequate lactational nutrition on pups born to dams fed either the 6% or 8% casein-diets, two litters of pups from each of these dietary treatments
status on the 3 diets is shown in Table I. From this table it can be seen that in the pregravid period the body weights of the dams on the 6% or 8% caseindiet are significantly lower than those on the 25% diet with the 6% diet group being significantly lower in body weight than the 8% diet group at all pregra-
TABLE II Brain and body weights at birth for pups from dams fed the 3 different protein diets Values given are means + S.E. Diet Body weight (g) Brain weight (mg)
6% Casein
8% Casein
25% Casein
4.95 ± 0.2c (n = 154)
6.00 ± 0.2 (n = 119)
6.20 ± 0.1 (n = 253)
241 ± 12b (n = 8)
280 ± 11 (n = 8)
275 ± 10 (n = S)
bp < 0.05, cp < 0.001 for 6% vs 8% or 25% pups, 2-tailed ttests.
223 TABLE III Influence o f maternal gestational and lactational diets on growth o f pups at weaning (day 21) Dietary reversal
Body weight (g) Brain weight (rag)
6/25 a (n = 16)
8/25 (n = 16)
25/25 (n = 16)
40.9 -i- 0.5b
65.0 _+ 1.0
64.9 + 0.3
1128 4- 25b
1393 4- 32
1410 + 50
aThe slash mark separates the prenatal/postnatal nutritional status of the pups, i.e. pups cross-fostered at birth to 25% casein diet. Values given are mean weights _+S.E. bp < 0.001 for 6/25 pups vs 8/25 or 25/25 rats, 2-tailed t-tests.
vid weeks. However, during pregnancy there is no significant difference between the 8% and 25% groups but there is a highly significant difference between the 6% and the 8% and the 25% diet groups. The effect of the 3 diets on the subsequent body and brain weights of the pups is shown in Table II. From this table it can be seen that on the 8% casein-diet the body and brain weights of the pups do not differ significantly from those on the 25% casein-diet. However, pups from mothers fed the 6% casein-diet show significantly lower body and brain weights at birth. The effects of the dietary reversals at birth on brain and body weights at weaning (21 days) are shown in Table III. Cross-fostering pups from 8% casein-diet mothers to normal 25% casein (8/25) diet mothers resulted in these pups having body and brain weights not significantly different from pups of 25% dams fed a 25% casein-diet during lactation (25/25). On the other hand, cross-fostering pups from 6% casein-diet dams at birth onto dams fed a 25% casein diet (6/25) did not produce pups at weaning with normal body or brain weights. Rather, both body and brain weights were significantly lower than either the 8/25 or 25/25 groups (Table III). DISCUSSION Although the diets are isocaloric, the inadequate protein content of the 6% casein-diet was responsible for the 48% decreases in pregnancy weight gains of these dams compared to those fed the 25% diet. Thus, this condition of maternal malnutrition was responsible for the SGA birth weights of their pups be-
ing almost 20% less than normal and brain weights of some 14% less than normal. In humans, pregnancy weight gains that are about one-half of the normal values are usually accompanied by an increased number of SGA births. This would be expected since fetal growth and development depends on access to maternal nutrients and on the amounts of reserves amassed during early pregnancy that can be mobilized later to ensure fetal needs. Because the latter is determined by the maternal nutritional status, the poorly nourished, low pregnancy weight gain females like the 6% rat dams will have inadequate reserves available to maintain normal fetal growth. Another important, but often ignored, factor in fetal growth retardation is the pregravid nutritional status of the female. By starting the dietary restrictions at conception in most animal studies of malnutrition, the dams have pregravid stores available to support some degree of fetal development which is a situation that is considerably different from that of chronically undernourished women. For such women, however, the pregravid nutritive status makes an important contribution to the subsequent birth weight of their infants. For example, Eastman and Jackson 3 showed that among women with small pregnancy weight gains (5 kg) the incidence of SGA births was 1.5% for those with pregravid weights of 72 kg as compared to the 10% SGA births for women with pregravid weights of 54 kg. Thus, high pregravid weights can furnish reserves to support a good birth weight even when the pregnancy weights are low. The lack of high pregravid weights in the 6% dams (Table I, some 33% less weight than the dams on the 8% or 25% casein-diets) coupled with their small pregnancy weights makes them similar to a large group of chronically undernourished women. Interestingly, in common with SGA humans, the 6% rat pups also showed smaller deficits in brain weights compared to body weights at birth. This brain 'sparing' effect, however, does not prevent losses in neurons, glia, and myelin, as well as impaired dendritic differentiation that occurs in the in utero deprived offspring of either species~0. Although all of these brain deficits can affect intelligence, it is, perhaps, the damage to cortical dendritic development which may have a greater impact on the learning potential of SGA progeny rather than the
224 concurrent decreases in myelin or cell numbers~. In this regard, quantitative Golgi analyses of neurons in several neocortical areas in the 6% pups are presently underway in our laboratory as an extension of our earlier studies of neuronal development in subcortical areas (nucleus raphe dorsalis and locus coeruleus) 1,2 in 8% casein-diet rats. In addition, the SGA offspring display peripheral chemical alterations which can affect brain metabolism. These include hypoalbuminemia and a hyperlipolysis of adipose tissue resulting in increased plasma non-esterified fatty acids (NEFA). We have previously reported that this is the mechanism whereby the levels of tryptophan and serotonin are increased in the brains of the animals on the 6% and 8% casein-diets 4-5,9. While the peripheral and central chemical changes are quantitatively similar in both the 6% and 8% pups, they are much more pronounced in the 6% pups 6. Lastly, like many SGA infants, the 6% pups showed postnatal growth deficits (or stunting) as long-term sequelae of the in utero deprivation. Although they received adequate lactational nutrition by being fostered at birth to 25% dams (labeled as 6/25 rats in Table III), they had markedly lower than normal brain and body weights (20% and 37% decreases, respectively) at weaning (Table III). Thus, the insults of severe in utero deprivation in the rat do not end at birth. In spite of the appearance of both a normal pregnancy and normally appearing neonates for the 8% rats, the pups can be categorized as I A R at birth. Not only do they show many of the peripheral chemical alterations seen in SGA humans and in the 6% pups, but they also have markedly higher than normal brain serotonin and norepinephrine and their precursor amino acids 4,7. Moreover, this I A R syndrome can continue to remain hidden in the 8% pups if they receive adequate postnatal nutrition. While the poor milk production of their dams resulted in the 8% pups having markedly lower than normal brain and body weights (13% and 60% decreases, respectively) at weaning 9, fostering of these rats at birth to 25% dams assured their normal weaning growth indices (labeled as 8/25 rats in Table III). However, these postnatal nutritional adequacies were unable to effect rehabilitation of their in utero determined SGAlike peripheral chemical imbalances and impaired brain neurochemistry 5.
On the whole, these data show the fallacy of assesing the nutritional status of a female and its effect on the offspring by her pregravid weight, her pregnancy weight gain, or by the birth weight of the neonate. Although there is a strong association between birth weights and pregravid weights, they are not necessarily a reliable indicator of good maternal health or the intake of essential nutrients. This is especially true among mildly overweight women from low socioeconomic groups in this country whose pregravid weights may reflect a type of caloric compensation as a replacement for much of their essential protein needs due to the escalating price of the latter. Also, a high pregnancy weight gain in women gives no assurance that the fetus is receiving the proper nutrients to ensure adequate growth and development. As in the non-gravid state, a high weight gain can result from a high caloric diet which may be deficient in protein. For example, not only do these low protein deficits cause pregnant women to show a rapid accumulation of fluid (up to 7 kg) in a short period of time, but may also account for the high percentage (6--7%) of their children who show abnormal growth and neurological functioning at one year of age tl . Our findings of abnormal brain metabolism in the 8% pups as sequelae of mild maternal protein deftcits 7 may possibly be applicable to many normal birth weight infants. Human birth weights above 2.5 kg are defined as normal even though they are below the optimum of 3.5 kg. Because mild fetal deficits may occur in infants with birth weights between these values, this may account for the 6-13% of them who later showed signs of MBD or mental retardation~2. Thus, many children who are considered 'normal' at birth because their birth weights are within the so-called normal range may be at risk. Lastly, the lack of breast-feeding in this country may allow this IAR syndrome to remain hidden during infancy. Using our 8% pups as a model, it was noted that: (1) this mild form of chronic protein lack only becomes noticeable during early lactation when the body weights of these pups become markedly lower than normal; and (2) if they were nursed by 25% dams, their postnatal body weight gains were normal. Thus, the use of infant formulas rather than breast-feeding in our society would allow for the 'normal' postnatal growth of the infant and would miss any signs of low-level maternal protein lacks
225 which could appear during early lactation if the mother nursed her baby. Since the normal weight gain 8/25 pups still showed the altered brain and peripheral chemical changes of IAR offspring at weaning, it becomes less surprising to find many 'normal' birth weight children who have low IQs or show signs
REFERENCES 1 Diaz-Cintra, S., Cintra, L., Kemper, T., Resnick, O. and Morgane, P. J,, The effects of protein deprivation on the nucleus raphe dorsalis: a morphometric Golgi study in rats of three age groups, Brain Research, 221 (1981) 243-255. 2 Diaz-Cintra, S., Cintra, L., Kemper, T. and Morgane, P. J., The effects of protein deprivation on the nucleus locus coeruleus: a morphometric Golgi study in rats of three ages groups, Soc. Neuroscience, (12th Annual Meeting), Minneapolis, MN, 1982, p. 640. 3 Eastman, N. J. and Jackson, E., Weight relationships in pregnancy. 1. The bearing of maternal weight gain and prepregnancy weight on birth weight in full term pregnancies, Obstet. Gynecol. Survey, 23 (1968) 1003-1025. 4 Miller, M., Leahey, J. P., Stern, W. C., Morgane, P. J. and Resnick, O., Tryptophan availability: relation to elevated brain serotonin in developmentally protein-malnourished rats, Exp. Neurol., 57 (1977) 142-157. 5 Miller, M. and Resnick, O., Tryptophan availability: the importance of prepartum and postpartum dietary protein on brain indoleamine metabolism in rats, Exp. Neurol., 67 (1980) 298-314. 6 Miller, M., Hasson, R. and Resnick, O., Developmental protein malnutritionin rats: adaptive changes in brain indoleamine metabolism in small-for-gestational-age offspring, Soc. Neuroscience, (11th Annual Meeting), Los Angeles, CA (1981) p. 72. 7 Morgane, P. J., Miller, M., Kemper, T., Stern, W., Forbes, W., Hall, R., Bronzino, J., Kissane, J., Hawrylewicz, E. and Resnick, O., The effects of protein malnutritionon the
of MBD in the absence of perinatal brain damage. ACKNOWLEDGEMENTS
These studies were supported by Grant HD-06364 from the National Institute of Child Health and Human Development.
developing central nervous system in the rat. Neurosci. Biobehav. Rev., 2 (1978) 137-230. 8 Resnick, O., Miller, M., Forbes, W., Hall, R., Kemper, T., Bronzino, J. and Morgane, P. J., Developmental protein malnutrition: influences on the central nervous system of the rat, Neurosci. Biobehav. Rev., 3: 233-246. 9 Resnick, O., Morgane, P. J., Hasson, R. and Miller, M., Overt and hidden forms of chronic malnutrition in the rat and their relevance to man, Neurosci. Biobehav. Rev., 6 (1982) 55--75. 10 Ross, P. and Cramoy, C., Nutrition and pregnancy. In M. Winick (Ed.), Nutrition Pre- and Postnatal Development, Plenum Press, New York, 1979, pp. 133-228. 11 Shanklin, D. R. and Hodin, J., Maternal Nutrition and Child Health, Charles C. Thomas, Springfield, 1979. 12 Sinclair, J. C., Saigal, S. and Yeung, C. Y., Early postnatal consequences of fetal malnutrition. In M. Winick (Ed.), Nutrition and Fetal Development, John Wiley, New York, 1974, pp. 147-174. 13 Singer, J. E., Westpal, M. and Niswander, K., Relationship of weight gain during pregnancy to birth weight and infant growth and development in the first year of life, Obstet. Gynecol., 31 (1968)417-423. 14 Stern, W. C., Miller, M., Forbes, W. B., Morgane, P. J. and Resnick, O., Ontogeny of the levels of biogenic amines in various parts of the brain and in peripheral tissues in normal and protein malnourished rats, Exp. Neurol., 49 (1975) 314-326. 15 Winick, M., Nutrition and nerve cell growth, Fed. Proc., 29 (1970) 1510-1515.