Testosterone, but not nonaromatizable dihydrotestosterone, improves working memory and alters nerve growth factor levels in aged male rats.

Recent studies have suggested that testosterone levels are lower in men with Alzheimer's disease and that testosterone treatment improves cognition in older men. Since testosterone can be aromatized to estrogen, testosterone's effects could be due to conversion into estrogen. We treated aged male rats with either testosterone or dihydrotestosterone (DHT), the latter of which is not aromatized to estrogen, in order to determine whether these treatments improve spatial working and reference memory as assessed in the water radial arm maze. We also tested whether such effects are related to beta-amyloid levels in the hippocampus or neurotrophin levels in the hippocampus, entorhinal cortex, frontal cortex, or striatum. Aged rats made more errors than young rats on all memory measures. Testosterone, but not DHT, improved working memory and decreased hippocampal NGF protein in aged rats, while having no effect on beta-amyloid. However, higher beta-amyloid levels were correlated with poorer working memory performance in young rats. Neurotrophin levels in entorhinal cortex were positively correlated with errors for all memory measures in androgen-treated rats. Similar to findings in human studies, in our study androgen treatment lowered circulating estradiol levels in aged rats, suggesting that androgen treatment exerts feedback to the hypothalamic pituitary axis and that conversion to estrogen may not be the underlying biological mechanism of testosterone's effects on memory and growth factor levels. The ratio of estradiol to testosterone, or the actions of the aromatase enzyme itself, may be responsible for the observed effects. These data support the hypothesis that testosterone therapy in aging men may provide positive effects on cognition and that neural regions that are linked to cognition, such as the hippocampus and/or entorhinal cortex, may be involved in such effects.