Structural and functional study of the serotonin 5-HT3A/B receptor binding site

M. Lochner & S. C. R. Lummis, Dept. of Biochemistry, University of Cambridge, Tennis Court Road, Cambridge CB2 1QW, UK.

The Cys-loop family, of which the 5-HT3 receptor is a member, includes nicotinic acetylcholine (nACh), GABAA and glycine receptors. The 5-HT3 receptor can function as a homopentamer of A subunits, but native receptors may exist as heteropentamers consisting of A and B subunits (Fletcher et al., 1998), which may affect kinetics, voltage dependence, pharmacology and single channel conductance (Dubin et al., 1999; Davies et al., 1999). To understand molecular details of the B subunit we have examined models of the 5-HT3A/B binding site to determine potentially important residues, which were examined using radioligand binding and functional assays.

Amino acid residues important for 5-HT binding in the mouse 5-HT3A receptor subunit (Reeves et al., 2003), were mutated to the corresponding 5-HT3B subunit residues using the Kunkel method (Kunkel, 1985) in pcDNA3.1 (Clontech), and then transfected into HEK 293 cells. They were then examined using radioligand binding assays (Table 1) with [ 3H]granisetron, a 5-HT3 receptor antagonist (Lummis et al., 1993). Selected mutants were subcloned into pGEMHE, mRNA synthesised, injected into Xenopus oocytes, and voltage clamp recordings performed 36 h post-injection. Data was analysed using PRISM software.

Table 1. [3H]granisetron binding affinities and LogEC50s for 5-HT3 receptor mutants. Values are mean ± s.e.m (n ≥ 4). NB = no binding, NR = no response, * = significant difference to WT (P < 0.05), - = not attempted.

The data (Table 1) show differential effects on binding and gating. Thus T181N has a similar Kd to WT but a 10 fold difference in EC50, and Y153H failed to bind antagonist but nonetheless yielded functional receptors. F226R and I228S have similar effects on binding, but I228S results in a 100 fold change in EC50 while F226R produces non-functional receptors. Y234F (Price et al., 2004) and E236Q bind antagonist, but the double mutation disrupts this ability. The results have been fitted onto a model of the binding pocket and suggest that most of the residues studied here play specific roles in 5-HT binding in AB receptors.

Davies et al. (1999) Nature 397: 359.
Dubin et al. (1999) J Biol Chem 274: 30799.
Fletcher et al. (1998) Trends Pharmacol Sci 19: 212.
Kunkel (1985) Proc Natl Acad Sci USA 82: 488.
Lummis et al. (1993) Eur J Pharmacol 243: 7.
Price et al. (2004) J Biol Chem 274: 30799.
Reeves et al. (2003) Biophys J 84: 2338.