Paradoxical effects of lithium on serotonergic receptor function: an immunocytochemical, behavioural and autoradiographic study.

Lithium is the preferred treatment for bipolar affective disorder, yet its mechanism of action is poorly understood. Our study was designed to investigate the effect of lithium on the 5-HT2A or 5-HT2C (5-HT2A/2C) receptor subtypes, by comparing the consequences of chronic pre-treatment of rats with lithium on 5-HT2A/2C receptor-mediated behavioural responses, Fos expression, and the density of these receptors in the brain. In addition, the time-course and persistence of the effect of chronic lithium on 5-HT2A/2C receptor-mediated Fos expression was examined. Furthermore, the acute action of lithium on Fos expression was also examined. In an investigation of the dose response of Fos to the 5-HT2A/2C agonist 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI), rats received saline or 1, 2, 4, 8, 12, 16, 24 or 32 mg/kg DOI, then were sacrificed 3 h later for immunocytochemical localisation of Fos. In a chronic lithium study, rats received either control or lithium-containing (0.1% LiCO3) chow for 3 weeks prior to challenge with 8 mg/kg DOI. DOI-induced locomotor activity was measured for 30 min immediately following the drug challenge, then 150 min later, the animals were sacrificed for Fos immunocytochemistry. The brains of another group of rats, also receiving either control or lithium-containing diet for 3 weeks, were analysed for the distribution and density of 5-HT2A receptor binding sites by quantitative [3H]ketanserin autoradiography. One group of chronic lithium treated rats received ritanserin (0.4 mg/kg), a 5-HT2A/2C receptor antagonist, 40 min before DOI challenge and were sacrificed 3 h later for Fos localisation. In the time-course experiment, rats received lithium-containing diet for 3 weeks followed by normal, control diet for 48 h, 1, 2 or 4 weeks prior to DOI or saline challenge. A further group of animals received an injection of LiCl (3 mM/kg) before being challenged with DOI or saline 12, 24, 36 or 48 h later. The dose-response experiment revealed that little Fos-like immunoreactivity was evident above basal levels following administration of 1 mg/kg DOI. However, at all other doses examined, Fos-like immunoreactivity was elevated in a number of brain areas, particularly in cerebral cortex, olfactory tubercle and amygdala. Following 24 mg/kg DOI, the number of Fos-positive nuclei appeared to have reached a plateau level. Treatment of rats with chronic lithium significantly enhanced DOI-induced locomotor activity and Fos-like immunoreactivity throughout the cerebral cortex. This elevation in Fos-like immunoreactivity was completely abolished by prior treatment with ritanserin. In contrast, chronic lithium treatment had no effect on the density of [3H]ketanserin binding to 5-HT2A receptors in any brain region examined. The results of the time-course experiment demonstrated that the enhancing effect of lithium on 5-HT2A/2C receptor-mediated Fos expression was short-lived such that Fos-like immunoreactivity returned to untreated levels within 48 h. In the acute lithium experiment, administration of lithium to rats 12 or 24 h before DOI resulted in a similar elevation of Fos-like immunoreactivity to that seen in chronically treated animals. Administration of acute lithium 36 or 48 h before DOI had no effect. The effects of lithium on 5-HT2A/2C receptor function thus appear to be complex. In particular, the results of this study indicate that the enhancing effects of lithium on DOI-induced locomotor activity and Fos-like immunoreactivity are not accompanied by any alteration in the density of 5-HT2A receptor binding sites. If changes in receptor numbers therefore do not account for the physiological effect of chronic lithium, other explanations must be sought. The study also suggests that the inositol depletion hypothesis of lithium's therapeutic action does not adequately explain the mechanism of action of lithium in man.