Somatodendritic autoreceptor regulation of serotonergic neurons: dependence on l-tryptophan and tryptophan hydroxylase-activating kinases Rong-Jian Liu, 1 Evelyn K. Lambe 1 and George K. Aghajanian 1,2 1 Department of Psychiatry, Yale School of Medicine, New Haven, CT, USA 2 Department of Pharmacology , Yale School of Medicine, New Haven, CT, USA Keywords: 5-HT 1A autoreceptor, CaMKII, dorsal raphe nucleus, serotonin, WAY 100635 Abstract The somatodendritic 5-HT 1A autoreceptor has been considered a major determinant of the output of the serotonin (5-HT) neuronal system. However, recent studies in brain slices from the dorsal raphe nucleus have questioned the relevance of 5-HT autoinhibition under physiological conditions. In the present study, we found that the difficulty in demonstrating 5-HT tonic autoinhibition in slice results from in vitro conditions that are unfavorable for sustaining 5-HT synthesis. Robust, tonic 5-HT 1A autoinhibition can be restored by reinstating in vivo 5-HT synthesizing conditions with the initial 5-HT precursor l-tryptophan and the tryptophan hydroxylase co- factor tetrahydrobiopterin (BH 4 ). The presence of tonic autoinhibition under these conditions was revealed by the disinhibitory effect of a low concentration of the 5-HT 1A antagonist WAY 100635. Neurons showing an autoinhibitory response to l-tryptophan were confirmed immunohistochemically to be serotonergic. Once conditions for tonic autoinhibition had been established in raphe slice, we were able to show that 5-HT autoinhibition is critically regulated by the tryptophan hydroxylase-activating kinases calcium ⁄ calmodulin protein kinase II (CaMKII) and protein kinase A (PKA). In addition, at physiological concentrations of l-tryptophan, there was an augmentation of 5-HT 1A receptor-mediated autoinhibition when the firing of 5-HT cells activated with increasing concentrations of the a 1 adrenoceptor agonist phenylephrine. Increased calcium influx at higher firing rates, by activating tryptophan hydroxylase via CaMKII and PKA, can work together with tryptophan to enhance negative feedback control of the output of the serotonergic system. Introduction The somatodendritic 5-HT 1A autoreceptor has been implicated in the pathogenesis of mood disorders and in the action of antidepressant drugs. It is generally accepted that activation of 5-HT 1A autoreceptors, through activating an inwardly rectifying K + conductance (Aghajanian & Lakoski, 1984; Williams et al., 1988), results in negative-feedback regulation of the firing of 5-HT neurons in the raphe nuclei (for review, see Aghajanian & Sanders-Bush, 2002). However, uncertainty remains about whether tonic activation of somatodendritic 5-HT 1A autoreceptors in the dorsal raphe nucleus is functional under physiological conditions as the potent and selective 5-HT 1A antagonist WAY 100635 (Fletcher et al., 1996) has not been consistently reported to have a disinhibitory effect on the firing of dorsal raphe 5-HT neurons. Some in vivo studies have reported statistically significant disinhibitory effects of low doses of WAY 100635 (Fornal et al., 1996; Mundey et al., 1996; Hajos et al., 2001) or another 5-HT 1A antagonist NAD-299 (Arborelius et al., 1999), while others have found only a trend or no disinhibitory effect (Forster et al., 1995; Gartside et al., 1995; Fletcher et al., 1996; Martin et al., 1999; Haddjeri et al., 2004). It has been even more difficult to demonstrate tonic 5-HT 1A autoreceptor regulation of 5-HT cell firing in vitro in brain slice. With the exception of one study (Corradetti et al., 1996), 5-HT 1A receptor antagonists have reported to have little or no effect on cell firing in brain slices (Craven et al., 1994; Fletcher et al., 1996; Johnson et al., 2002). The latter results have led to the proposal that 5-HT 1A autoreceptors are not operative under normal physiological conditions (see Johnson et al., 2002). We hypothesize that the difficulty in demonstrating 5-HT 1A receptor-mediated autoinhibition in brain slice is that standard artificial cerebrospinal fluid (ACSF) is deficient in substances such as l-tryptophan needed to maintain normal rates of 5-HT synthesis. The consequence of this deficiency is underlined by the fact that l-tryptophan has extremely low affinity for the rate-limiting enzyme in 5-HT synthesis, tryptophan hydroxylase (TPH, K M ¼ 50 lm, Friedman et al., 1972), and that depletion of l-tryptophan in vivo leads to a large drop in extracellular levels of 5-HT and its metabolite 5-hydroxyindole acetic acid in the dorsal raphe nucleus (Bel & Artigas, 1996). By the same token, it is likely that 5-HT would become depleted in brain slice as the standard slice medium is deficient in l-tryptophan. If that is the case, reestablishment of 5-HT synthesizing conditions by addition of l-tryptophan should reinstate 5-HT 1A receptor-mediated autoinhibition. In the present study, we reexamined the disinhibitory effects of WAY 100635 on the firing rate of 5-HT in raphe slices incubated in ACSF with or without added l-tryptophan. These electrophysiological experiments were performed in parallel with an assessment of 5-HT levels by immunohistochemistry. Based on these studies, which showed robust tonic autoinhibition in the presence of added l-tryptophan, we explored the role of Correspondence: Dr R.-J. Liu, as above. E-mail: Rong-Jian.Liu@yale.edu Received 15 September 2004, revised 3 November 2004, accepted 2 December 2004 European Journal of Neuroscience, Vol. 21, pp. 945–958, 2005 ª Federation of European Neuroscience Societies doi:10.1111/j.1460-9568.2005.03930.x