Decreased plasma gonadotropin and testosterone levels in arthritic rats: are corticosteroids involved?

Abstract

Infectious and inflammatory diseases are often accompanied by abnormal reproductive functions, and the present working hypothesis is that proteins (called cytokines or interleukins, ILs) released by activated immune cells are at least in part responsible for these neuroendocrine changes. In order to test this hypothesis, we need paradigms of immune pathologies in which concentrations of cytokines are increased, and those of hormones of the hypothalamic-pituitary-gonadal (HPG) axis are blunted. We chose a rodent model of arthritis, adjuvant-induced arthritis (AIA), in which rats show elevated plasma IL-6 and decreased testosterone (T) concentrations. We describe here the first phase of our studies, in which we determined whether gonadotropin release was also altered, whether this change was responsible for the low T levels, and whether elevated corticosterone participated in the decreased activity of the HPG axis.AIA is induced by the intramuscular injection ofMycobacterium butyricum (MBB) into the tail base of the rat, with swelling of the limbs occurring 11-12 days later. We observed significant decreases in LH and FSH secretion of castrated AIA male rats, suggesting that altered gonadotropin output was independent of the gonads. The absence of significant alterations in GnRH gene expression in the hypothalamus of AIA rats, as well as only modest declines in pituitary responsiveness to GnRH, indicate that these mechanisms are not primarily responsible for the blunted gonadotropin concentrations. Intact AIA rats exhibited a dramatic decline in T levels, but no concimitant rise in LH concentrations. The observation that gonadotropin secretion does not increase despite significantly reduced T levels suggests the presence of an unidentified defect within the GnRH neuronal circuitry that prevents the gonadotrophs to respond to decreased steroid feedback. Testicular responsiveness to hCG was significantly blunted in AIA rats, and this decrease was not reversed by acute blockade of nitric oxide formation or of prostaglandin synthesis. Interestingly, the onset of these hormonal changes preceded the appearance of symptoms (limb swelling), as well as the decrease in body weight that accompanies visible joint enlargement. On the other hand, blunted T secretion coincided with rising levels of ACTH and corticosterone. This suggested that adrenal steroids might be responsible for the decrease in LH and T values, but this hypothesis did not prove valid. Indeed, we observed that adrenalectomized AIA animals implanted with corticosterone pellets retained their low T levels. Furthermore, clamping corticosterone levels was only moderately effective in reversing the inhibitory influence of the arthritic process on LH secretion.In the absence of significant alterations in GnRH gene expression, it is possible that low Gn levels are secondary to an abnormal pattern in GnRH pulse amplitude and/or frequency. While the decrease in plasma LH concentrations may play a role in the dramatically lowered plasma T values, it is more likely that the inability of the testes to respond to gonadotropin is of significance. While we cannot rule out the participation of perceived stress at the onset of the changes in pituitary and testicular function of AIA rats, we hypothesize that cytokines released by the inflamed tissues, an event that may well precede the appearance of overt swelling, are responsible for the activation of the HPA axis and independently, for the decreased activity of the HPG axis. The AIA model may therefore provide an experimental paradigm in which to test hypotheses related to the cross-talk between the immune system and reproductive parameters.