Peptides, Vol. 10, pp. 1117-1120. © Pergamon Press plc, 1989. Printed in the U.S.A. 0196-9781/89 $3.00 + .00

Effect of Posttraining and Pretest [3-Endorphin and ACTH Administration

in Normal and Protein Malnourished Rats

M A R C O S L. S. P E R R Y A N D I V A N I Z Q U I E R D O

Centro de Memoria, Departamento de Bioquimica, Instituto de Biociencias U.F.R.G.S . (centro), 90049 Porto Alegre, RS, Brazil

Rece ived 27 October 1987

PERRY, M. L. S. AND I. IZQUIERDO. Effect of posttraining and pretest B-endorphin and ACTH administration in normal and protein malnourished rats. PEPTIDES 10(6) 1117-1120, 1989. --Rats were submitted to a normal (25% casein) or a low protein diet (8% casein) from the day of birth until the age of 110 to 120 days. Hypothalamic 13-endorphin-like immunoreactivity was lower in the animals raised and maintained with the low protein diet, and, in addition, it did not respond to training in a step-down inhibitory avoidance task with or without footshock with a depletion, as was the case with the normal diet animals. In the animals submitted to the normal protein diet posttraining ACTH (0.2 ~g/kg) and 13-endorphin (1.0 ~g/kg) caused retrograde amnesia of a step-down inhibitory avoidance task,and pretest administration of these substances had no effect of its own, but was able to reverse the amnesia induced by their previous posttraining administration. In the animals submitted to the low protein diet, results were similar except that pretest 13-endorphin caused amnesia on its own. On the basis of previous findings which suggest that pretest actions of ACTH and 13-endorphin depend on their endogenous release at the time of training, the present results are compatible with a malfunction of the brain 13-endorphin system in the undernourished animals.

13-Endorphin ACTH Posttraining effects Pretest effects Memory Learning Protein malnutrition

POSTTRAINING 13-endorphin, over a very wide dose range (6, 7, 9, 11), or ACTH administration, at high doses (4--6, 12, 17), disrupts retention test performance of avoidance tasks in rats. The effect is reversed by another injection of the peptide(s) prior to testing (6-8, 17). For reasons described elsewhere in detail-- mainly, that the pretest effect of the drugs is strictly dependent on their posttraining effect (6, 7, 12, 17)--this cannot be explained by the postulation of two separate effects, one on consolidation and another on retrieval (6). It has, therefore, been interpreted as due to an accretion of information to the tasks by the 13-endorphin injected in the posttraining period (6,8).

The effects of both 13-endorphin and ACTH on memory may reflect a physiological phenomenon. Pituitary ACTH is secreted in response to arousal or stress both at the time of training and at the time of testing (21). Activation of the brain (but not pituitary) 13-endorphin system, on the other hand, occurs only at the time of training, since it is a response to novelty and test sessions are, by definition, usually not novel (7, 8, 10). Injected ACTH, therefore, adds to endogenously released ACTH both after training and at the time of testing whereas exogenous 13-endorphin sums with endog- enous 13-endorphin only after training. The possibility exists, therefore, that posttraining and pretest effects of ACTH and 13-endorphin may be different in animals with a selective long-term impairment of these two systems.

In a previous paper (16), we reported that rats reared and maintained under a low protein diet (8% casein) do not respond

with a reduction of brain 13-endorphin immunoreactivity levels to the exposure to a variety of training situations, like normal animals do. This is in contrast to the pituitary ACTH-mediated cortico- sterone response to footshock stress, which is normal or near- normal in rats raised on a low protein (22), or a low protein-low calorie diet (1).

The present paper extends our previous observation on the lack of effect of behavioral training on brain 13-endorphin-like immu- noreactivity to other training situations, and studies the influence of posttraining and pretest ACTH and 13-endorphin administration on retention of an inhibitory avoidance task, using the classic 2 x 2 design (6, 15, 23), in protein malnourished rats as compared with normal rats.

Animals and Diets

METHOD

Pregnant Wistar rats from our own breeding stock were housed individually. On the day of delivery they were divided into two groups: One was fed a control diet (25% casein) and the other a low protein diet (8% casein). Both diets were isocaloric, contained a normal amount of salts and vitamins, and were similar to those used by Shoemaker and Wurtman (20), or Marichich et al. (13), or in a previous paper (16) (Table 1). Litters were restricted to 8 pups and were weaned at the age of 21 days. After weaning, each group was maintained on the same diet that had been fed to the mothers.

1117

1118 PERRY AND IZQUIERDO

TABLE 1 COMPOSITION OF THE NORMAL AND OF THE LOW PROTEIN DIET

g/kg Diet

Normal Protein Low Protein

Casein* 312 100 Cornstarch 468 680 Celulose 10 10 Vegetable oil 150 150 Salt mix1- 50 50 Vitamin mixt 10 10

*Casein contained 80% protein. ~As used by Shoemaker and Wurtman (18) and Perry et al. (16).

Behavioral and biochemical studies were carried out at the age of 110 to 120 days using male animals.

13-Endorphin-Like Immunoreactivity

Animals were sacrificed by decapitation immediately after the end of the behavioral procedures (free exploration of the training apparatus during 30 sec, or step-down inhibitory avoidance training, see below), or, in the case of the control groups, were taken out of their home cages and sacrificed right away. The brain was quickly removed and the hypothalamus plus preoptic area were dissected on an inverted Petri dish placed on salted ice at a temperature of - 5 ° C or less. Dissection land marks were as follows: Frontally, the caudal border of the optic chiasma; cau- daily, the mammillary bodies; laterally, the lateral borders of the hypothalamus; vertically, approximately 3 mm. The tissue sample was extracted in 1 N acetic acid at 98°C and subsequently processed for 13-endorphin radioimmunoassay as recommended by Rossier et al. (16) and as extensively used in this laboratory (7-10). Cross-reacting molecules [13-1ipotropin, proopiomelano- cortin, 13-endorphin(1-27) and its N-acetylated and N-detyrosynated metabolites] account for a small fraction of hypothalamic 13- endorphin immunoreactivity, and there is reason to believe that they are not released by training (8). However, since no effort was made to separate these molecules, the results wilt be expressed as representing 13-endorphin-like immunoreactivity (16). The radio- immunoassay procedure involved rabbit-raised antibodies against

125 human 13-endorphin. A 50% displacement of ( )l-labelled human 13-endorphin was obtained with 51 pg of human 13-endorphin. Recovery was 95%. The assay cross-reacted with 13-1ipotropin (33-35%) and with proopiomelanocortin (less than 1%) but not with Met-enkephalin, Leu-enkephaiin or dynorphin(1-13).

As suggested by Cohen and Wurtman (2), since the brain is heterogeneous as to cell types, the different cell types are differently affected by malnutrition, and 13-endorphin-like immu- noreactivity is restricted to one cell type, immunoreactivity values are expressed per organ and not per gram of tissue (15).

Behavioral Procedures

Animals were trained and tested in a step-down inhibitory avoidance task as described elsewhere in detail (3, 4, 8, 10, 14). Briefly, a 50 x 25 × 25 cm acrylic box was used. The left half of the floor was covered by a 5-cm high wood platform; its right side was a grid of parallel 1-mm caliber bronze bars spaced 10 mm apart. In the training session, animals were placed on the platform and their latency to step down onto the grid was measured with a stopwatch; when they placed their four paws on the grid they

TABLE 2 B-ENDORPHIN LIKE IMMUNOREACT1VITY OF HYPOTHALAMUS

(ng/ORGAN) OF 110- TO 120-DAY-OLD RATS SUBMIIrED TO A NORMAL OR A LOW PROTEIN DIET FROM THE DAY OF BIRTH: EFFECT OF THE EXPOSURE TO AN INHIBITORY AVOIDANCE TRAINING SESSION WITH

OR WITHOUT FOOTSHOCK

Normal Protein Low Protein Diet Diet

Control 58.7 + 4.5 40.6 + 3.8 (9) (8)

Training with footshock 40.4 + 6.0* 36.0 + 6.4 (8) (9)

Training without 43.3 + 3.5* 33.2 + 2.4 footshock (7) (6)

*Significant difference from control animals of the same diet group at p<0.01 level in Duncan multiple range test. The difference between the normal and the low protein controls was significant at a p<0.001 level in a t-test. Differences between control and trained animals in the low protein diet group are not significant at a p<0.2 level in a Duncan multiple range test. Data expressed as means + SE; n is shown in parentheses below each figure.

received a 0.8 mA, 60 Hz, 2-sec footshock. Immediately after this, they received an IP injection of saline, human 13-endorphin (Sigma) (1 ~g/kg), or ACTH(1-24) (Organon) (0.2 ~,g/kg); the drugs were dissolved in saline. Injection volume was 1 ml/kg in all cases. Twenty-four hours later, the animals received another injection of saline, ACTH, or 13-endorphin (see Table 3). Five min after the injections they were placed again on the platform where their step-down latency was measured once more (test session). Test minus training step-down latency was considered as a measure of retention. A ceiling of 180 sec was imposed on this measure: Values equal to or higher than 180 sec were counted as 180 sec. Retention score differences among groups were therefore evaluated by nonparametric statistics: A Kruskal-Wallis analysis of variance followed by individual two-tailed Mann-Whitney U-tests.

RESULTS

In confirmation of the findings of many others (13, 19, 22), as well of our own (16), the animals in the 8% protein group had a significantly lower body (62+3 g, n=87 ) and hypothalamic (45 + 2 mg, n = 23) weight than those of the 25% group (171 + 7 g, n = 8 8 , and 59+2 mg, n = 2 4 , respectively, p<0.001 in both cases). (The n was much lower for hypothalamic than for body weight values because the latter includes all the animals in the present study, i.e., those shown in Tables 2 and 3; whereas the hypothalamus was dissected and weighed only in those that were submitted to radioimmunoassay, Table 2.)

Radioimmunoassay data are shown in Table 2. Differences among groups were significant in a one-way analysis of variance, F(5,39)= 3.85, p<0.01. As previously reported (16), hypotha- lamic 13-endorphin-like immunoreactivity values were lower in the 8% protein group than in the 25% protein group. In addition, both exposure to the training apparatus and inhibitory avoidance train- ing were followed by a decrease of hypothalamic 13-endorphin-like immunoreactivity in the control, but not in the prea~in malnour- ished group.

Training step-down latency differences among groups were not different [mean+ SE of all groups, 11.1 +0.8, n --- 140; F(13,126) = 1.01, p>0.2].

Retention scores are shown in Table 3. In the 25% protein

PEPTIDE EFFECTS IN MALNOURISHED RATS 1119

TABLE 3

RETENTION SCORES [MEDIAN (INTERQUARTILE RANGE) TEST MINUS TRAINING STEP-DOWN LATENCY, IN SEC] OF 110- TO 120-DAY-OLD

RATS SUBMITFED TO A NORMAL OR A LOW PROTEIN DIET FROM THE DAY OF BIRTH, TREATED WITH SALINE, ACTH(I-24) (0.2 ttg/kg), OR 13-

ENDORPHIN (1.0 o.g/kg), GIVEN 1P 1 MIN AFTER TRAINING OR 5 MIN PRIOR TO TESTING

Treatment Retention Scores

Normal Protein Low Protein Posttraining Pretest Diet Diet

Saline Saline 170 (145/180) 180 (143/180) [3-Endorphin Saline 0 (-3/12)* - 1 (-4/11)* 13-Endorphin 13-Endorphin 180 (166/180) 180 (157/180) Saline [3-Endorphin 180 (171/180) - 1 (-6/17)*

ACTH Saline 64 ( - 1/91)* 3 (-3/44)* ACTH ACTH 180 (151/180) 180 (104/180) Saline ACTH 180 (120/180) 180 (82/180)

*Significant difference from both saline-saline control groups at p<0.02 in a two-tailed Mann-Whitney U-test.

Data expressed as medians (interquartile ranges); n= 10 per group. A ceiling of 180 sec was imposed on the retention latency measure.

group, both ACTH and 13-endorphin given posttraining showed reduced memory scores, and the pretest administration of the substances reversed this effect and had no influence of its own on retention test performance. In the 8% group, the effects of ACTH were similar to those observed in control animals, but those of [3-endorphin were not. Posttraining [3-endorphin did reduce mem- ory scores, but so did pretest 13-endorphin except in the animals that had received a posttraining injection of the substance, in which case it reversed the retrograde amnesia.

DISCUSSION

The literature indicates that the pituitary ACTH-adrenal axis functions normally or near-normally in the malnourished animals (1,22). Both the present and our previous findings (16) suggest that the brain 13-endorphin system is, instead, deficient in the malnourished group: It does not respond to novel experiences, such as the training experience, with a reduction of 13-endorphin- like immunoreactivity. Evidence commented on elsewhere in detail (7-10) shows that this reduction is best explained by a

release and subsequent degradation of 13-endorphin mainly be- cause of the impossibility of explaining it by synthesis inhibition or by intragranular destruction (8,10). Therefore, it can be assumed that protein malnourished rats are able to develop, through their lifetime, behavioral responses to systems that depend on ACTH release, but not to systems that depend on brain 13-endorphin release.

Accordingly, the behavioral response to posttraining and pre- test ACTH treatments was similar in the control and malnourished groups: Posttraining administration of the hormone was amnestic, pretest administration was without effect on its own, but reversed the amnestic effect of the posttraining treatment. This combination of effects suggests that administration of the hormone in the posttraining period adds information to the tasks, which then becomes necessary at the time of retrieval (8,15). The behavioral response to posttraining and pretest 13-endorphin was similar to that of ACTH in the 25% protein group, as previously shown in many other experiments using similar doses in animals submitted to a standard diet (6-11), but was clearly different in the protein malnourished animals. In these, retention scores were low in animals submitted to posttraining or pretest 13-endorphin, but not in animals submitted to both posttraining and pretest I~-endorphin. The results in this group are consistent with the idea that the action of pretest ~3-endorphin is strictly dependent on a previous influence of this peptide at the posttraining period (6,8); the difference with normal animals is quite simply explainable by the fact that, in the normal group, endogenous 13-endorphin is released at the time of training, whereas in the malnourished group it is not. It is likely that malnourished animals are unable to develop a relation between posttraining and pretest 13-endorphin effects because their brain 13-endorphin system has a chronic malfunction.

There were no other differences in behavior between these two groups of animals: Training step-down latencies in all groups were similar; retention scores in saline-saline groups were also similar; the overt behavior while training or testing appeared to be the same. A previous study using similar diet groups also showed no difference in the performance of shuttle avoidance or intertrial crossings between the two groups (16). There is no evidence that any of these behavioral variables depend on the brain 13-endorphin system [see (7,8)].

ACKNOWLEDGEMENTS

Supported by financial help from Financiadora de Estudos e Projetos (FINEP), grant No. 43.86,0371.00. We are indebted to Maria A. Carrasco, Carlos A. Netto, Norma Volkmer and Renato D. Dias for their collabo- ration in some of the experiments.

REFERENCES

1. Adlard, B. P. F.; Smart, J. L. S. Adrenocortical function in rats subjected to nutritional deprivation in early life. J. Endocrinol. 54:99-105; 1972.

2. Cohen, E. L.; Wurtman, R. J, Nutrition and brain neurotransmitters. In: Winick, M., ed. Nutrition: Pre- and postnatal development. New York: Plenum Press; 1979:103-132.

3. de Almeida, M. A. M. R.; Izquierdo, I. Effect of the intraperitoneal and intracerebroventricular administration of ACTH, epinephrine, or B-endorphin on retrieval or an inhibitory avoidance task in rats. Behav. Neural Biol. 40:119-122; 1983.

4. de Almeida, M. A. M. R.; Kapczinski, F. P.; Izquierdo, I. Memory modulation by posttraining intraperitoneal, but not intracerebroven- tricular, administration of ACTH or epinephrine. Behav. Neural Biol. 39:277-283; 1984.

5. Gold, P. E.; van Buskirk, R. B. Effect of posttrial hormone injections on memory processes. Horm. Behav. 7:509-517; 1976.

6. Izquierdo, I. Endogenous state dependency: Memory depends on the relation between the neurohumoral and hormonal states present after

training and at the time of testing. In: Lynch, G.; McGaugh, J. L.; Weinberger, N. M., eds. Neurobiology of learning and memory. New York: Guilford Press; 1984:333-350.

7. Izquierdo, I.; Netto, C. A. The role of B-endorphin in memory modulation. Ann. NY Acad. Sci. 444:162-177; 1985.

8. Izquierdo, I.; Netto, C. A.; Chaves, M. L. F.; Dalmaz, C.; Pereira, M. E.; Siegfried, B. Construction and reconstruction of memories. Braz. J. Med. Biol. Res. 21:9-16; 1987.

9. Izquierdo, I.; Souza, D. O.; Dias, R. D.; Carrasco, M. A.; Perry, M. L.; Eisinger, S.; Elisabetsky, E.; Vendite, D. A. Beta-endorphin causes retrograde amnesia and is released from the rat brain by various forms of training and stimulation. Psychopharmacology (Berlin) 70:173-177; 1980.

10. Izquierdo, I.; Souza, D. O.; Dias, R.D.; Carrasco, M. A.; Volkmer, N.; Perry, M. L. S.; NeRo, C. A. Effect of various behavioral training and testing procedures on brain B-endorphin-like immunoreactivity, and the possible role of B-endorphin in behavioral regulation. Psy- choneuroendocrinology 9:381-389; 1984.

1120 PERRY AND IZQUIERDO

11. Lucion, A. B.; Rosito, G.; Sapper, D.; Palmini, A.; Izquierdo, I. Intracerebroventricular administration of nanogram amounts of B- endorphin and Met-enkephalin causes retrograde amnesia in rats. Behav. Brain Res. 4:111-115; 1982.

12. McGaugh, J. L. Involvement of hormonal and neuromodulatory systems in the regulation of memory storage. Annu. Rev. Neurosci. 12:in press; 1989.

13. Marichich, E. S.; Molina, V. A.; Orsingher, O. A. Persistent changes in central catecholaminergic system after recovery of perinatally undernourished rats. J. Nutr. 109:1045-1050; 1979.

14. NeRo, C. A.; Cavalheiro, E. A.; Carrasco, M. A.; Volkmer, N.; Dias, R. D.; Izquierdo, I. Response of the rat brain B-endorphin system to novelty: Importance of the fomix connection. Behav. Neural Biol. 43:37-46; 1985.

15. Overton, D. A. Experimental methods for the study of state depen- dency. Fed. Proc. 33:1900-1913; 1974.

16. Perry, M. L. S.; Carrasco, M. A.; Dias, R. D.; Izquierdo, I. B-Endorphin like immunoreactivity of brain, pituitary gland and plasma of rats submitted to postnatal protein malnutrition: Effect of behavioral training. Peptides 5:15-20; 1984.

17. Riccio, D. C.; Concannon, J. T. ACTH and the reminder phenome- non. In: Martinez, J. L., JR.; Jensen, R. A.; Messing, R. B.; Rigter, H.; McGaugh, J. L., ed. Endogenous peptides and learning and

memory processes. New York: Academic Press; 1981:117-142. 18. Rossier, J.; Vargo, T. M.; Minick, S.; Ling, N.; Bloom, F. E.;

Guillemin, R. Regional dissociation of B-endorphin and enkephalin content in rat brain and pituitary. Proc. Natl. Acad. Sci. USA 74:5162-5165; 1977.

19. Shoemaker, W. J.; Wurtman, R. J. Effect of perinatal undernutriiton on the metabolism of catecholamines in the rat brain. J. Nutr. 103:1537-1547; 1973.

20. Spear, N. E.; Mueller, C. W. Consolidation as a function of retrieval. In: Weingartner, H.; Parker, E. S., eds. Memory consolidation-- Psychobiology of cognition. Hillsdale: Lawrence Erlbaum; 1984: 111-147.

21. Van Wimersma Greidanus, Tj. B.; Rees, L. H.; Scott, A. P.; Lowry, P. J.; de Wied, D. ACTH release during passive avoidance behavior. Brain Res. Bull 2:101-104; 1977.

22. Wiener, S. G.; Levine, S. Perinatal malnutrition and early handling: interactive effects on the development of the pituitary-adrenal system. Dev. Psychobiol. 11:335-352; 1978.

23. Zornetzer, S. F. Neurotransmitter modulation and memory: A new psychopharmacologic phrenology? In: Lipton, M. A.; DiMascio, A.; Killam, K. F., eds. Psycbopharmacology: A generation of progress. New York: Raven Press; 1978:637-649.