It has been documented that stress or glucocorticoids have conflicting effects on memory under different conditions. However, it is not fully understood why stress can either impair or enhance memory. Here, we have examined the performance of six age groups of Wistar rats in a water maze spatial task to evaluate the effects of stress under different conditions. We found that the impairment or enhancement effect of an 'elevated platform' (EP) stress on memory was dependent on previous stress experience and on age. EP stress impaired memory retrieval in water maze naive animals, but enhanced rather than impaired memory retrieval in young water maze stress-experienced animals. Furthermore, exogenously applied corticosterone or foot shock stress before water maze training prevented the impairment of memory retrieval that should be induced by treatment with corticosterone or foot shock before the 'probe trial'. Again, memory retrieval was enhanced in young animals under these conditions, and this enhancement can be prevented by the glucocorticoid receptor antagonist RU 38486. Thus, glucocorticoid receptor activation not only induced impairment of memory but also increased the capacity of young animals to overcome a later stress. The present findings suggest that the effect of stress on memory can be switched from impairment to enhancement dependent on both stress experience and age.

ABSTRACT

Tissue and plasma levels of insulin-like growth factor-I (IGF-I), and relative levels of liver IGF-I RNA, were measured in 6-month-old ewe lambs which were well fed (n = 10) or starved (n = 10) for 5 days. Half of each nutrition group was given daily (09.00 h) injections of human GH (hGH; 0·15 mg/kg body weight per day). Blood was sampled daily from 09.00 to 12.00 h at 15-min intervals through jugular vein catheters and the lambs were slaughtered 24 h after the fifth injection of hGH.

Tissue and plasma IGF-I was extracted using an acid-ethanol-cryo-precipitation technique and estimated by radioimmunoassay. Tissue IGF-I was corrected for retained plasma IGF-I using tissue and blood haemaglobin levels. Liver IGF-I RNA levels were monitored by in-situ hybridization.

Plasma IGF-I (nmol/l) was higher in both the fed group and the fed group given GH treatment. Tissue IGF-I from kidneys (nmol/kg) was also higher (P < 0·001) in the fed group. There was no significant difference in IGF-I concentrations in the muscle biceps femoris or liver between fed and starved lambs. Although GH treatment did not increase IGF-I levels in tissues significantly, IGF-I RNA levels in liver were increased (P = 0·02) in both fed and starved animals. The relative liver IGF-I RNA levels positively correlated with their corresponding tissue IGF-I levels in the fed group and the fed group given GH treatment. The lack of a significant IGF-I response to GH in tissues may be due to either the time at which tissues were sampled after the GH treatment or the dose of GH administered. However, the higher IGF-I concentrations in plasma and kidney from fed compared with starved animals and the positive correlations between liver IGF-I and IGF-I RNA levels suggest that tissue and plasma IGF-I is regulated by nutrition and GH, with nutrition playing a critical role in the regulation of tissue and plasma IGF-I in normal lambs.

Journal of Endocrinology (1993) 136, 217–224