¹Division of Psychiatry, NEUROTEC, Karolinska Institute, Stockholm, Sweden

Correspondence: Dr M Heilig, M57 Huddinge University Hospital, S-14186 Stockholm, Sweden. E-mail: markus.heilig@neurotec.ki.se

Received 24 February 2003; Revised 16 July 2003; Accepted 24 July 2003; Published online 10 September 2003.

Abstract

Chronic treatment with cannabinoid agonists leads to tolerance. One possible mechanism for this is receptor internalization, but tolerance has also been reported with compounds that only cause internalization to a low degree. Furthermore, cannabinoid antagonist administration precipitates a characteristic withdrawal syndrome in tolerant subjects, accompanied by neuronal activation and enhanced release of corticotropin-releasing hormone (CRH) in the central amygdala. The underlying molecular mechanisms are unknown. We examined the role of cannabinoid tolerance and withdrawal for the expression of the cannabinoid 1 (CB1) receptor and of CRH in rats. Tolerance was first established functionally. An acute dose (100 g/kg) of the CB1 agonist HU-210 suppressed locomotor activity, and had an anxiogenic-like effect on the elevated plus-maze. Both effects were absent following daily treatment with the same agonist or a lower (40 g/kg) dose for 14 days. Next, withdrawal was reliably precipitated by a single dose (3 mg/kg) of the CB1 antagonist SR141716A in rats treated subchronically with 14-day HU-210. Using in situ hybridization, a robust suppression of CB1 mRNA expression was found in the caudate-putamen, indicating a downregulation of CB1 expression levels as one mechanism for tolerance to the locomotor suppressant effects of HU-210. The CRH transcript was upregulated in the central amygdala in precipitated withdrawal compared to nonwithdrawn tolerant subjects, suggesting that increased gene expression contributes to the previously reported CRH release in withdrawal. Most importantly, this increase occurred from a suppressed level in tolerant subjects, and behavioral signs of withdrawal, presumably mediated by CRH, were seen at the CRH expression that had only returned to normal nontolerant levels. This suggests the possibility of an allostatic shift, as previously proposed on theoretical grounds. The expression of CRH-R1, CRH-R2, NPY, and its Y1 receptor mRNA was analyzed in search of neural substrates for the allostatic shift observed, but did not seem to contribute to the dysregulated state.