Comparative effects of prenatal and postnatal undernutrition on learning and memory in rats

Indian Journal of Experimental Biology Vol. 37, January 1999, pp. 17-22 Comparative effects of prenatal and postnatal undernutrition on learning and memory in rats Arun K Jaiswal, S N Upadhyay, K S Satyan & S K Bhattacharya' Neuropharmacology Laboratory, Department of Pharmacology, Institute of Medical Sciences, Banaras Hindu University, Varanasi 221 005, India Received 4 July \996; revised \ 0 September 1998 Effects of pre- and post-natal undernutrition on learning and memory parameters were studied in albino rats. Prenatal undernutrition was induced in rat pups by restricting the mother's diet by 50% during the entire gestation period, whereas postnatal undernutrition was induced in rat pups by restriction of their diet by rotating them between lactating and non-lactating maternalised females for 12hr each day during suckling period from 2nd day to 18th day after birth. At 2.5 to 3 months of age all the rat offsprings were subjected to (i) original and reversal discrimination learning, (ii) passive avoidance, and (iii) active avoidance and its retention tests. The results indicate that both pre- and post-natal undernutrition in rat pups caused significant deficits in original and reversal discrimination learning, retention of passive avoidance after one week retention interval, and retention of active of avoidance learning. However, both pre- and post-natal undernutrition did not show significant effect on acquisition of active avoidance and retention of passive avoidance after 24hr retention interval. Psychobiological deficits due to undernutrition during the early period in rats are accompanied, and possibly due to the significant and often permanent morphological changes in CNS I - 3. Extensive literature on the effects of prenatal and postnatal undernutrition on learning and memory are available 4.5. While some of the studies indicate no straight-forward "hippocampal syndrome" in prenatally undernourished rats on at-maze and operrant behaviour 6, others report, alterations in the rate of learning and extinction of avoidance response to painful stimuli, and changes in spontaneous behaviour in adult nutritionally rehabilitated rats which had been earlier been nutritionally deprived during postweaning period by restriction of protein diets 7 The present investigation was planned to have a comparative assessment of the effects of prenatal and postnatal undernutrition on original and reversal discrimination learning, passive avoidance behaviour and conditioned avoidance response acquisition and retention in rats. The method of undernutrition is unique in the current protocol in that ; prenatal undernutrition was induced in pups born to the underfed dams during the gestation period and *Correspondent author these pups born were foster nursed by normal lactating mothers and thus the effect of undernutrition of the mother was eliminated. In postnatal undernutrition the pups were rotated between lactating dam and non-lactating maternalised females for 12hr each day. Thus, only lactational undernutrition was induced during the suckling period while maternal care was not compromised. Materials and Methods The study was conducted on inbred littermate Charles Foster albino rats of either sex. Two different techniques were employed to induce undernutrition: (1) Prenatal undernutrition: Pregnant rats (as evidenced by presence of sperm in vaginal swab) were given half the amount of stock diet (Brook Bond Lipton) until parturition. This regimen is known to induce undernourishment in pups born to the underfed dams during gestation periods. The pups born from the underfed mothers were removed and placed with normal lactating dams (8 pups per dam) until weaning at 25 days. (2) Postnatal undernutrition: A nutritional deprivation procedure was employed wherein pups

18 INDIAN J EXP BIOL, JANUARY 1999 were removed from the lactating dam for l2hr each day and placed with non-lactating maternalised females. Thus, the pups were deprived of milk without being deprived of maternal care and environmental stimuli. Deprivation was initiated when pups were 2 days old and continued for 16 days. Thereafter, the pups were allowed to rejoin the control dams (non-deprived) and allowed to stay with the dam for another 14 days. After reaching the age of 30 days, the pups were removed from the dam and received laboratory chow thereafter. The behavioural experiments began when the rats were 2.5 to 3 months old. Behavioural testing (I) Original and reversal discrimination learning: The apparatus used for original and reversal discrimination learning was a single unit black/white T-maze. A modified single unit T maze, fabricated locally following the design of Crocket and Noble 9 was used. It consists of a start box (22.8x lox I2 cm), a lane (72.4x IOx I2 cm) ending in two goal arms (22.8x lox 12 cm each) and two goal boxes (22.8xlOxI2 cm each). The wooden maze had a wire mesh top for observation and was painted brown with the exception of two arms, one of which had a white and the other a black coating on the inside walls and floor. Half of the animals of each of the groups were made to run to black+ ve vs white- Ve and the other half to white +ve vs black- ve discriminations in the single unit T-maze. Therefore, all the animals were tested and their response to +ve cue, irrespective of its position, was always rewarded by food while response to the negative cue went unrewarded. The response to the positive alley led to the access of the animal to the goal box with the food reward. The position of the black and white arms was randomly varied lo Change in positions of the positive and negative arms was made in each trial. The rat was placed in the start box with its head towards the exit and the intervening door between the start box and runway was closed as soon as the animal left the start box. The runway had a choice point at its end from where the positive and negative alleys led to the door of the goal boxes, which were kept open. The rat was permitted to make one or many errors before entering the rewarded goal box. Thereafter, the goal box door was closed and the rat was placed back in the start box and randomization of the positions of the arms was done for the next trial. Thus, each animal was trained on a ten trial session schedule per day following a 23hr food deprivation. The criterion of maximal learning was eight errorless consecutive trials in a single day's session. A correction method. was tonowed in determining a test trial where a trial was lerminated with a ~ing le correct response irrespective of the number of errors the animal might have committed ll. After the original discrimination learning described above, the rats, were subjected to reversal discrimination learning by following the same procedure after reversal of the cues. Thus, an animal black ve which had learnt to discriminate white +ve vs cues was now tested on reversed cues of black+ ve vs white- Ve and vice-versa. The total number of errors committed and trials required to learn the given task were regarded as response measures of the original and subsequent reversal discrimination learning tests. (2) Passive avoidance behaviour: The apparatus and method used was essentially the same as described earlier l2. Briefly, the rat was placed on the elevated platform in the centre of the floor of the test box and the latency to step down was recorded. Immediately after stepping down, each rat received 3 sec of electric shock (0.5 rna) through the grid floor and were then returned to the home cage. On the following day (24hr retention interval) each rat was again placed on the raised platform 'and given a 5 min inhibition period. Latency to step down was recorded. Electric shock was not given on day 2. If the animals remained on the raised platform for 5 min period, the maximum score of 300 sec was assigned. At day 9 (after a gap of one week) latency to step down was again recorded to test the retention of the passive avoidance learning. (3) Active avoidance learning and its retention: The rats were singly placed on the right compartment of the shuttle bo~ and al lowed to adapt for IS sec and were then exposed to an acoustic (buzzer) stimulus (conditioned stimulus) for IS sec which was followed by both the acoustic stimulus and electric shock (unconditioned stimulus, 1.5 rna, 50 Hz) through the grid fl oor of the right compartment for 30 sec. Jumping to the adjacent left compartment, which was

JAISWAL et al. : EFFECT OF UNDERNUTRlTION OF LEARNING & MEMORY 19 unelectrified, during or before the conditioning stimulus, was regarded as correct conditioned response. The number of trials required to reach the criterion of two consecutive correct responses represented the learning rate. A 60 min inter-trial interval period was maintained. In order to make statistical analysis possible, rats not reaching the criterion within eight trials were arbitrarily assigned the score of 9. All the rats were subjected to this training schedule and were retested 24hr and 1 week later. Total trials, total time taken and number of shocks received by each rat were recorded. Statistical analysis Means ± SO values were calculated for various response measures of original and reversal discrimination, passive avoidance and active avoidance learning and their retention tests. Further, the obtained data were subjected to Kruskal-Wallis one way ANOYA and posthoc group comparisons were made by Mann- Whiteny V-test l3 Results The data pertaining to body weight and growth of rat pups of different nutritional groups (Table I & Fig. 1) clearly indicate that prenatally undernourished (PRUN) rat pups had significantly lower body weight in comparison to both control and postnatally undernourished (POUN) rat pups at birth. But later on rats of both PRUN and POUN groups exhibited significantly stunted body growth. However, the rats of PRUN and POUN groups did not differ from each other with respect to their body growth after birth. The data pertaining to error and trial scores of the original and reversal discrimination learning tests are given in Table 2. The analysis of the data indicates significant effects for error score of original discrimination learning (H=6.64, P < 0.05) and error (H=6.61, P < 0.05) and trial (H=9.14, 250.--.. Q rn -H C! ro Q) ~ '----' 0 C! 200 150 o Conlrol P RUN 'V PO UN T ~...C! Ql)... Q) ~ >, " III 0 100 50 Birth 2 4 6 8 10 12 14 Age in weeks Fig. l--body growth curves of different groups of rats

20 INDIAN J EXP BlOL, JANUARY 1999 Table l~ody weights (g) of rats at different stages of development [Values are mean ± SD] Groups n Control 16 PRUN 8 POUN 21 At birth After 7 week After 14 week 6.31±0.68 98.44±19.35 182.88±32.98 4.69±0.46" 57.75±8.55" 141.25±31.41 5.95±O.42bb 65.24±14.67" 129.29±31.66" PRUN = Prenatal undernutrition ; POUN = Postnatal undernutrition. Superscripts.. and bb represent statistical significance respectively in comparison to control and PRUN at P<O.OI (Mann-Whitney U-test, two tailed). Table 2--Error and trial scores of original and reversal discrimination learning tests [Values are mean ±SD; abbreviations are same as in Table I] Groups n Error Trial Original discrimination learning Control 16 7.94±3.23 34.69±11.18 PRUN 8 16.00±9.37" 47.J3±20.00 POUN 7 13.29±5.44" 48.14±17.61 Reversal discrimination learning Control 16 6.50±2.48 27.88±9.34 PRUN 8 13.25±8.78.. 46.25± 16.76" POUN 7 7.57±1.62 b 30.14±2.46 bb Superscripts" and b represent statistical significance respectively in comparison to control and PRUN. a.b and... bb denote P<0.05 and <0.01 respectively (Mann-Whitney U-test, two tailed). P < 0.0 I) scores of reversal discrimination learning tests. However, trial score of original discrimination learning test yielded insignificant H-value (4.73, P > 0.05). Further, multiple group comparisons revealed that both prenatal and postnatal undernutrition induced significant learning acquisition deficits in the offsprings. PRUN rats have also been found to be significantly deficient in the reversal discrimination learning test in comparison to controls. However, POUN rats did not differ significantly from the: control rats in learning the reversal discrimination problem. Values of step down latency scores of the passive avoidance retention test after 24 hr and I week intervals are given in Table 3 and the results indicate significant treatment effects both at 24 hr (H=6.81, P < 0.05) and 1 week (H=7.19, P < 0.05) retention intervals. Further analysis indicated that both PRUN and POUN rats did not differ from control rats in the retention tests after 24 hr, however, after 1 week retention interval PRUN rats showed significant retention deficits in comparison to control as well as POUN rats. Significant treatment effects have also been observed for trial (H=7.1O, P < 0.05), time (H=7.34, P < 0.05) and shocked trial (1-1=7.92, P < 0.05) response measures of CAR acquisition tests and trial (H=7.09, P < 0.05) and shocked trial (H = 13.43, P < 0.01) scores of CAR retention test after 24hr. Further, none of the response measures yielded significant effects for CAR retention test after 1 week retention interval. Further, multiple group comparisons indicated that PRUN rats required significantly more trials and time and they also received more number of shocks in learning the CAR task in comparison to control and POUN rats (Table 4). PRUN rats also showed significant Table 3--Step-down latency scores (in sec) in passive avoidance test [Values are mean ±SD; abbreviations are same as in Table I] Groups n Retention interval 24 hr I week Control 13 113.99±95.27 226.30± I 00.70 PRUN 8 53.79±32.44 55.35±35.35" POUN 12 147.95±115.19 b 21 4.50±116.20 Superscripts " and b represent statistical significance respectively in comparison to control and PRUN... and b denote P<O.OI and <0.05 respectively (Mann-Whitney U-test, two tailed). Table 4--Total trials, time (in sec) and shocked trial scores of active avoidance learning acquisition and retention tests. [Values ar,e mean ±SD; abbreviations are same as in Table 1] Groups n. Control 8 PRUN 5 POUN 2 1 Control 8 PRUN 5 POUN 21 Control 8 PRUN 5 POUN 21 Trials Time Shocked trials CAR Acquisition 3.63±0.74 100.55± 19.23 1.50±0.53 6.60±O.55" 191.70±55.18" 4.20±1.64" 4.10±1.18 bb 140.90±88.49a. b 1.9O±0.89 b Retention after 24 hr 2.00±0.00 34.33±12.50 0.00 3.20±1.30" 84.1 0±60. 78" 3.20±1.30" 2.29±0.46 42.08±24.10 0.29±O.46 bb 2. 13±O.35 2.80±tO. 84 2.30±0.66 Retention after 1 week 37.30±17.45 59.36±24.86 41.80±32.20 0.I3±O.35 O.60±0.55" 0.20±0.4Ib CAR = Conditioned avoidance response. Superscripts " and b represent statistical significance respectively in comparison to control and PRUN..,b and... bb denote P=<0.05 and <0.01 respectively (Mann-Whitney U test, two tailed).

JAISW AL et al.: EFFECT OF UNDERNUTRITION OF LEARNING & MEMORY 21 retention deficits after 24 hr whereas performance of POUN rats on CAR retention after 24 hr did not differ significantly from that of control rats with respect to total trials and time scores. Further, performance of both PRUN and POUN rats on CAR retention task after 1 week interval also did not differ significantly from control rats. Discussion Early insults of undernutrition on brain development give rise to altered neurophysiological function and anatomy of the hippocampal formation, resulting in reduced ability to maintain the synaptic component of long-term potentiation (LTP)14. Increased resistance to extinction of a food reward alternation task on an elevated T-maze 6 and retarded acquisition of a DRL 18 sec operant task l5 are in conformity with the view that hippocampal function may be compromised. It is obvious from the results of the present study that both pre- and post-natal undernutrition produced significant origipal learning deficits while only prenatal undernutrition induced reversal learning deficits. These findings indicate that the prenatal undernutrition mak~s the animal less efficient to shift from a previous strategy (which was successful to learn original discrimination task) to a new strategy to learn the reversal discrimination task reflecting likely compromised hippocampal function. These results are in agreement with an earlier report l6 wherein prenatally malnourished animals made significantly greater number of errors in learning to alternate in a 5-unit multiple T-maze than did well nourished controls. Elevated T-maze has proven reliable in revealing performance deficits in animals with septohippocampal damage I7. 18. Interestingly, on passive avoidance task, both prenatal and postnatal undernourishment did not affect the 24 hr retention performance but PRUN induced significant retention deficits after one week as compared to control rats. These findings lend further credence to the hypothesis that previously undernourished rats show diminished ability to potentiate the synaptic component of L Tp14. It is evident from our results on active avoidance test, that the PRUN rats were more resistant to acquisition on various parameters and also showed significantly poorer retention of CAR task in comparison to control rats in retention test after 24 hr interval. On the contrary, in POUN rats, the acquisition and retention were comparable with that of controls on CAR task. This indicates that the nutritional insults during the gestation period, when brain growth takes place l9, may significantly affect the learning and retention on. the active avoidance paradigm. Early malnutrition has been known to alter both GABAergic 20 and seretonergic activilfl. Neonatal undernutrition has been shown to result in an increase in the brain stem concentrations of 5-HT and its metabolite 5-HIAA and heightened activity of brain seretonergic system 22. It is now generally accepted that stimulation of seretonergic neurotransmitter system impairs learning and memory functions 23. It has also been shown that increase in central seretonergic activity invariably leads to anxietylheightened emotionalilf4 and in a recent study both prenatally and postnatally undernourished rats exhibited increased anxiety and depression on vanous behavioural paradigms25. It is also' known that increased serotonin in brain may mediate the decreased release of acetylcholine via the 5-HT receptor activation 23. Decreased cholinergic activity is implicated in various types of dementi as associated with aging, Alzheimers's disease, Downs syndrome and other neurodegenerative disorders. Further, ascending seretonergic protection from the brain exert a powerful control over generalized electrical activity of the hippocampal formation and neocortex 26. These, neurotransmitter systems, known to be involved in memory, appear to be particularly vulnerable to nutritional insults, provided that such insults are inflicted at a time when foetal brain development and differentiation.. kl927 IS at Its pea '. It is also notable that PRUN and POUN rats showed lower body weight at birth and stunted body growth later during postnatal period despite access to normal nutrition after birth to PRUN rat pups and after suckling period to POUN rat pups. Prenatal nutritional deprivation has been reported to result in lower birth weight, reduced brain weight, brain DNA, protein and reduced ' cells in cerebellum in rats l9. On the other hand, undernutrition during lactation i.e., during the period of growth spurt of brain in rats, has also been documented to cause decrease in cerebellum, cerebral and ventricular cell size and cell

22 INDIAN J EXP BIGL, JANUARY 1999 division 28. Another possibility is that the procedures used to rear PRUN rat pups with foster mothers and also to underfed POUN rat pups by rotating them between lactating and non-lactating mothers are.iikely to disrupt the mother-infantenvironment relationship and this may affect the acceptance of pups by the dams and/or of the dams by the pups, and these factors may further result in undernutrition of rat pups during suckling period and consequently may lead to compromised cognitive development in rat pups. Therefore, the results of the present study indicate that pre- and post-natal undernourishment induced in rats by the models used in the present study, caused poorer learning and memory functions in rats. These findings may have clinical relevance in the developing country like, India where pre- and post-natal undernutrition still prevails. However, further studies are needed to elucidate the role of 17 specific neurotransmitter systems and functional elements of the developing brain underlying pre- 18 and post-natal undernutrition induced behavioural alterations. Acknowledgment This work was supported by grant-in-aid from Roussel Research Institute, Bombay, India. References 1 1 Altman J, Das G D &. Sudarshan K, Dev Fsychobiol, 3 (1970) 281. 2 Morgane P J, Miller M, Kemper T, Ster W, Forbes W Hall R; Bronzino J, Kissane J, Hawrylewicz E &. Res~ick 0, Neurosci Biobehave Rev, 2 (1978) 137. 3 Dobbing J &. Smart J L, Br Med Bull, 30 (1974) 164. 4 Smart J L, in Genetics, environment & intelligence, edited by A Oliverio (Elsevier, North-Holland, Amsterdam) 1977,215. 5 Smart J L, in Proceedings of XIllth congress of nutrition, edited by TG Taylor &. NK Jenkins (Libby, London) 1986, 74. 6 Tonkiss J &. Galler R J, Behav Brain Res, 40 (1990) 95. 7 Frankova S &. Barnes R H, } Nutr, 96 (1968) 485. 8 Jaiswal A K, Upadhyay S N &. Bhattacharya S K, Indian} Exp Bioi, 27 ( 1989) 269. 9 Crocket W &. Noble M N, J Genet Psychol, 103 (1963) 105. 10 Gellerman L W &. Spence M N, J Genet Psychol, 42 (1933) 356. II Hull C L &. Spence K W, J Comp Psychol, 25 (1938) 127. 12 Iaiswal A K, Upadhyay S N &. Bhattacharya S K. Indian J Exp Bioi, 26 (1990) 609. 13 Seigel, S, Nonparametric statistics for behavioural sciences (McGraw-Hili, NY) 1956. 14 Austin K B, Bronzino I &. Morgane P J, Dev Brain Res, 29 (1988) 267. 15 Tonkiss I, Galler' I R, Formica R N, Shukitt-Hab B &. Timm R R, Physiol Behave, 48 ( 1990) 73. 16 Hsueh A M, Simonson M, Chow B F &. Hanson H M, J Nutri, 104 (1974) 37. Olton D S, Becker IT&. Handleman G E, Behav Brain Res, 2 (1979) 315. Rawkins J N P &. Olton D S, Behav Brain Res, 5 (1 982) 331. 19 Nowak T S Ir &. Munro H N, in Nutrition and brain, edited by R J Wurtman &. I I Wurtman (Raven Press NY) 1977, 193. 20 Wiggins R, Fuller G &. Enna S, Life Sci, 35 (1984) 2085. 21 Resnick &. Morgane P I, Brain Res, 303 (1984) 163. 22 Sobotka J T, Cook PM&. Brodie E R, Brain Res, 65 (1974) 443. 23 Costall B, Domeney A M, Kelly M E & Naylor R I, Adv in Biosci, 45, (1992) 147. 24 Kahn R S, van Prag H M, Wtzler S, Asnis G M &. Baar G, Bioi Pychiat, 23 (1988) 189. 25 Iaiswal A K, Upadhyay S N, Satyan K S & Bhattacharya S K, Indian} Exp Bioi, 34 (1996) 1216. 26 V anderwolf C H, Int Rev Neurobiol, 30 (1 988) 225. 27 Kellog C K &. Guillet R, Transplancental effect on foetal health (Alan R Liss, NY) 1988, 265. 28 Winick M, Malnutrition and brain development (Oxford Univ Press) NY, 1976.