INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE DEVELOPING ORGANISM differs from the adult in many aspects. One such aspect is related to the activation of the hypothalamic-pituitary-adrenal (HPA) axis. Whereas adults secrete glucocorticoids (GCs) in a circadian fashion and in response to stressful stimuli, the infant rat's HPA axis appears to lack such activities. Approximately on postnatal day (PND) 4, the infant rat enters the so-called stress-hyporesponsive period (SHRP). This period lasts until PND 14 and is characterized by reduction of adrenal sensitivity, evidenced by the fact that exogenous administration of adrenocorticotropic hormone (ACTH) does not result in a significant adrenal output (for review, see Refs. 26 and 28). It appears that low levels of GCs during this critical period are essential for a normal development (19), because exposure to high levels of GCs has widespread deleterious effects on the developing central nervous system (2) and on some functional parameters of the adrenocortical axis (29).

The underlying mechanism of the adrenal insensitivity to ACTH was investigated recently. Both in vivo (41) and in vitro (1) data point to a nadir of adrenal sensitivity around PND 10. Examination of steroidogenesis during development shows that immunoreactive levels of cytochrome P-450 enzymes do not change during neonatal development. Immunoreactive levels of 3--hydroxysteroid dehydrogenase, however, are lower in PND 1 and 10 rat pups compared with adults, although the enzyme activity does not parallel changes in immunoreactivity; this parameter is similar in PND 10 and adult rats, and both are higher than that of PND 1 pups. The authors suggest that enzymes involved in steroidogenesis are differentially regulated during development, with no apparent correlation between enzyme activity and immunoreactivity levels of the proteins, probably due to the presence of different isoforms of these enzymes during this period in life (21).

Development of circadian rhythmicity (15) and negative feedback capacity occurs much later in development (8). The first appearance of circadian rhythmicity of plasma corticosterone (Cort) can be detected at PND 18 in overfed and standard-fed rats, with a slight delay in underfed animals, i.e., the rhythm is present at PND 21. In young animals, the peak of plasma Cort levels is observed during the dark period, whereas in adult rats it shifts to the end of the dark period (15). Both Goldman and co-workers (8) and Vázquez and Akil (37) showed that the negative feedback mechanism in the rat is active from PND 25 on. Although return to basal levels is somehow delayed at this age, it more closely resembles that of the adult rat (8, 37).

The theories formulated to explain the SHRP include factors of intrinsic origin, such as an enhanced negative feedback at the brain and pituitary levels (39), and immaturity of neural pathways that project to the hypothalamus (5). There is a growing body of evidence, however, indicating that inhibition of the infant's HPA axis also occurs as a result of maternal care (3, 14, 25, 32).

Mammalian development is characterized by a prolonged period of dependence on maternal care. The dam provides the infant with nutrients, warmth, tactile stimulation, and protection, to name only a few of the more obvious maternal behaviors. Less obvious, however, is the regulatory role the mother exerts on the pup's physiology. For instance, separation from the mother results in a decrease in heart rate, respiratory rate, and growth hormone secretion (9-12). Specifically in the case of HPA axis activity, 24 h of maternal separation induces an immediate increase of basal and stress- and ACTH-induced levels of Cort, in addition to increased stress-induced ACTH plasma levels (14, 32). Release of Cort is mainly, but not entirely, regulated by feeding (24), whereas ACTH secretion is completely inhibited by anogenital stroking (34).

The prolonged effects of maternal deprivation on Cort secretion have not been fully examined. Reunion of 12-day-old pups, previously maternally deprived (Dep) for 24 h, with their mothers for 4 days is capable of resetting Cort basal levels, but not the response to novelty nor to a saline or hypertonic saline injection. This study, however, investigated only one age after maternal deprivation (27). Therefore, it remains to be determined whether or not these effects are long lasting. The present study sought to examine the prolonged effects of maternal deprivation on Cort secretion in response to a mild stress.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects

Deprivation Procedure

Testing Procedure

Blood Sample Procedure

Data Analysis

	RESULTS

Top Abstract Introduction Methods Results Discussion References

PND 12

View larger version (16K): [in this window] [in a new window]   Fig. 1.   A: corticosterone plasma levels (means ± SE) of 12-day-old rat pups, either nondeprived (open bars) or deprived of their mothers for 24 h (filled bars). Pups were sampled before (0 min) or at different time intervals (5, 15, 30, 60 min) after a saline injection. Number of animals was 9-12/group. * Differs from nondeprived pups, P < 0.05. B: corticosterone plasma levels (means ± SE) of 16-day-old rat pups, left undisturbed (open bars) or deprived of their mothers for 24 h (filled bars), on postnatal day 11. Pups were sampled before (0 min) or after a saline injection at different time intervals (5, 15, 30, 60 min). Number of animals was 6-11/group.

PND 16

PND 22

View larger version (20K): [in this window] [in a new window]   Fig. 2.   A: corticosterone plasma levels (means ± SE) of 22-day-old rat pups, left undisturbed (open bars) or deprived (filled bars) of their mothers for 24 h, on postnatal day 11. Pups were sampled before (0 min) or after a saline injection at different time intervals (5, 15, 30, 60 min). Number of animals was 6-14/group. B: corticosterone plasma levels (means ± SE) of 30-day-old rats, left undisturbed (open bars) or deprived (filled bars) of their mothers for 24 h, on postnatal day 11. Animals were sampled before (0 min) or at different time intervals (5, 15, 30, 60 min) after a saline injection. Number of animals was 8-13/group. * Differs from nondeprived pups, P < 0.05.

PND 30

PND 60

Males. An interaction between time and group was observed [F(4,61) = 3.57; P < 0.02]. Post hoc analysis of this interaction revealed that Cort levels of non-Dep males were higher at 60 min than basal. Dep male rats showed augmented hormone levels at 5, 15, and 30 min over basal. Non-Dep values were higher than Dep levels 60 min after saline injection (Fig. 3A).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   A: corticosterone plasma levels (means ± SE) of 60-day-old male rats, left undisturbed (open bars) or deprived (filled bars) of their mothers for 24 h, on postnatal day 11. Animals were sampled before (0 min) or after a saline injection at different time intervals (5, 15, 30, 60 min). Number of animals was 5-8/group. * Differs from nondeprived pups, P < 0.05. B: corticosterone plasma levels (means ± SE) of 60-day-old female rats, left undisturbed (open bars) or deprived (filled bars) of their mothers for 24 h, on postnatal day 11. Rats were sampled before (0 min) or at different time intervals (5, 15, 30, 60 min) after a saline injection. Number of animals was 5-10/group. * Differs from nondeprived pups, P < 0.05. For further differences between time points or genders, refer to RESULTS.

Females. Similarly, an interaction between time and group was observed [F(4,56) = 5.67; P < 0.001]. Non-Dep females showed increased Cort levels at 5, 15, and 30 min, compared with 0 and 60 min. Cort response of Dep females was higher at 5 and 15 min than levels at 0 min. In addition, Cort values at 5 min were higher than those at 30 and 60 min after saline. Comparison between groups showed higher levels of Cort in non-Dep females than their Dep counterparts at 30 min (Fig. 3B).

Body Weight

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The results of the present study indicate the effects of maternal deprivation are long lasting. Contrary to its immediate effects, however, maternal deprivation renders the adult offspring less responsive to a mild stress. This effect is more clearly observed at 30 days of age. At PND 60, hormone levels of non-Dep males and females were higher than those of their Dep counterparts at some time points after saline. Moreover, a closer inspection of the characteristics of the Cort response revealed that 30 day-old Dep animals responded faster to the stimulus: the onset of response of Dep animals occurred 5 min after the stimulus, whereas this increased response was only detected at 15 min in non-Dep animals. Similarly, Dep 60-day-old male rats responded to the stimulus with augmented Cort levels at 5 min, and this response remained so elevated until 60 min after the stimulus, when it became undistinguishable from nontreated levels. Differences in Cort secretion in non-Dep male rats were only observed at the 60 min time point. This persistent elevation may not reflect solely the response to the saline injection; rather, it may be a summation of this stimulus and the transport of animals to the decapitation room. The time course of Cort secretion of 60-day-old females was not so distinct between non-Dep and Dep animals. The turnoff of the Cort response in 60-day-old female Dep rats, however, did not appear to be as efficient as that of non-Dep females. The reason for this isolated difference is not evident. Whether this is a result of maternal deprivation-induced changes in female hormones leading to alteration on Cort clearance requires additional investigation.

The results of 12-day-old pups, maternally deprived for 24 h, are in agreement with previous findings (14, 25, 33, 34), although rat strain and conditions of cage cleaning were different between those and our study. At PND 16, both non-Dep and Dep animals exhibited increased Cort response to the stimulus, in accordance with the well-known fact that, at this age, pups are emerging from the SHRP (26, 28). Recently, van Oers et al. (36) reported that 20-day-old rat pups, separated from their mothers for 24 h on PND 3, exhibit higher ACTH response to stress, whereas this manipulation on 7-day-old pups does not result in altered ACTH release, and on PND 11, ACTH response of 20-day-old pups to stress is lower than that presented by normally reared pups. These groups, however, showed no differences of Cort levels compared with non-Dep pups. These findings evidence that maternal deprivation persistently alters the activation of the HPA axis. Moreover, they indicate that depending on the age when maternal deprivation is imposed, adrenal sensitivity is altered in different ways. If the reduced ACTH response to saline injection observed in animals deprived on PND 11 is long lasting, it would be legitimate to speculate that our results on 30-day-old rats, i.e., reduced Cort response to the saline injection, were a consequence of this effect.

Interestingly, high endogenous Cort levels induced by maternal deprivation did not appear to be detrimental for the adrenocortical response of adult rats to stress, contrary to the effects of exogenous administration of GCs on some adrenal parameters, such as adrenal relative weight (29). Several studies that examined the effects of neonatal administration of GCs show that 20- to 25-day-old pups present lower plasma ACTH levels in response to ether and increased feedback sensitivity in response to dexamethasone, without change of Cort-binding globulin binding capacity (7). When tested at 45-48 days of age, hydrocortisone-treated female rats show decreased plasma Cort response to novelty (mild stress) and to ether (intense stress), whereas reduced Cort levels in males is seen in response to novelty, but not ether. The authors suggest an impairment of the adrenocortical response to stress at this age (6). The adrenocortical response to ACTH and to immobilization is also preserved in 30-day-old rats, neonatally treated with hydrocortisone, although the circadian rhythm of plasma Cort levels is obliterated at this age, but not in 80-day-old male rats (13). It is important to bear in mind, however, that these steroids are administered at different ages and in pharmacological doses, and they probably do not mimic accurately a more physiological situation, such as maternal deprivation.

The possibility that maternal reunion is capable of reverting or attenuating possible harmful effects of increased Cort levels during infancy is a matter of speculation and requires further investigation. The ability of a lactating anesthetized dam to suppress the adrenocortical response of Dep pups to stress was previously demonstrated. Thus 12-, 16-, and 20-day-old Dep pups exposed to novelty in the presence of a lactating dam showed a suppressed Cort response, with or without a milk infusion. This effect was not obtained when pups were tested in the presence of a male or kept alone (30, 31). However, these studies only examined the effects of the presence of a dam during the testing procedure and not the consequences of the mother-infant reunion. Rosenfeld and co-workers (27) recently showed that only pups deprived of their mother for 24 h remain hyperresponsive to several stressors, after 4 days of reunion with their mothers, although basal levels are undistinguishable from those of non-Dep pups. The same results were not observed after 1 or 3 periods of 8 h of maternal deprivation, suggesting that there is a critical period, between 8 and 24 h, in which maternal deprivation effects appear to be long lasting. Therefore, the length, in addition to the regimen of exposure to maternal separation appears to be of critical importance. Although our results do not completely replicate their PND 16 data, a direct comparison is not possible because they used a different strain of rats and we examined different ages.

Apparently, the length of maternal separation is of vital importance in determining the long-term effects of this manipulation. Thus pups subjected to a 2-h period of maternal separation for 28 days show hyperactivity and increased grooming in the open field, growth retardation, and immune suppression, suggesting an alteration of the endogenous opioid system (38). Maternal separation for 4.5 h during the first 3 wk of life, on the contrary, induces a reduction of Cort response to a 2-h restraint stress and increased inhibition of Cort response by dexamethasone, indicating increased negative feedback efficiency in separated pups. No difference in hippocampal GC and in cerebral cortex serotonin and adrenaline receptor densities is observed between separated and control animals. According to the authors, these results suggest that periodic maternal separation for 4.5 h renders the animals less sensitive to environmental stimuli in adulthood (22). These effects were shown to persist until week 92 of life (20). A comparison between the effects of early handling and 180 min of maternal separation on some variables related to corticotropin-releasing factor (CRF) showed that handled rats presented lower basal hypothalamic CRF mRNA and content than maternally separated and control animals. Conversely, the latter groups showed higher Cort response to a restraint stress than handled rats (23).

Another interesting finding from our study is that gender differences in Cort response to stress were attenuated in Dep pups. Whereas non-Dep females showed a higher magnitude of Cort response than males at all time points except basal and 60 min, Cort levels of Dep females were higher than those of males only at the 5 min time point. We are unable, at this moment, to explain such attenuation in sex difference of Cort response in our Dep pups. Because we did not anticipate such an effect, possible alterations in estrous cycle were not followed. Therefore, any alteration in sexual hormones that might have been induced by maternal deprivation requires a deeper investigation. An alternative explanation may come from a recent report in which 24 h of maternal deprivation on PND 3 was shown to differentially alter the population of Cort receptors in the hippocampus of 48-day-old males and females. Whereas males show a downregulation of GC and mineralocorticoid receptors (GRs and MRs, respectively), females exhibit an upregulation of GRs at PND 48. These alterations were intensified when Dep pups received an ACTH injection at the end of the deprivation period, suggesting that increased circulating levels of either ACTH (from the injection) and/or Cort (as a consequence of the injection) appear to have a long-lasting influence on GRs and MRs that is dependent on the gender (35).

The fact that reductions of the Cort response in Dep pups were more consistent at PND 30 indicates that this effect of maternal deprivation, although long lasting, may not be permanent. Similar results were observed with the pups' weight gain, in which Dep pups presented a smaller weight than their non-Dep counterparts until PND 30. It is not possible to affirm that reduced body weight of Dep pups is a function of reduced food consumption. On the basis of data showing the influence of feeding on the HPA axis activity, i.e., increased activation of the HPA axis as a consequence of starvation (4), and the dependence of diurnal Cort response to ACTH on time of feeding (40), it appears unlikely these animals were in any way underfed compared with non-Dep rats. Our results are, to a certain extent, in agreement with previous studies reporting that neonatal cortisol treatment results in depressed body weight gain even at 60- or 80-day-old rats (16, 29).

Contrary to the early handling paradigm, in which old handled rats present decreased Cort response to stress in addition to a more efficient negative feedback and diminished GR loss than nonhandled animals (17, 18), the effects of maternal deprivation at much older ages is unknown. At present, our results appear to indicate that maternal deprivation results in long-lasting effects on Cort response to a mild stress.

Perspectives

It would be interesting to examine both ACTH and Cort responses of Dep rats to different kinds of stress. In addition, the behavioral response of Dep rats to adverse situations, such as the open-field and the elevated plus-maze could shed some light on possible maternal deprivation-induced alterations on brain mechanisms, other than the HPA axis itself.

Finally, on the basis of the early handling literature, it appears worth verifying the consequences of 24-h maternal deprivation on the hippocampal function of aged rats, because this brain region contains both GRs and, therefore, is a target area of Cort action.