Adult glycine receptors are pentamers of α1 and β subunits but their stoichiometry is still controversial. Knowing the stoichiometry of heteromeric glycine receptors is an essential starting point for any functional and kinetic study, since the number of α subunits provides an upper limit to the number of binding sites present. In order to assess the stoichiometry we mutated the 9Ô residue of the pore lining domain of α1 and β subunits from a conserved hydrophobic leucine into a hydrophilic threonine. Such mutation results in a leftward shift in the agonist dose-response curves for all channels in the nicotinic superfamily tested (Boorman et al. 2000).

HEK 293 cells were transfected with α1 and β cDNA in a 1:40 ratio in order to minimize contamination by homomeric α1 receptors. Glycine responses were recorded in whole-cell patch clamp from cells transfected with wild-type α1β, α1β^LT and α1^LTβ subunits. EC₅₀ and Hill slope values were obtained by fitting dose-response curves with the Hill equation. EC₅₀ values of mutant receptors were lower than the wild-type value of 92 ± 4 µM (mean ± S.E.M., n = 7), particularly for α1^LTβ (1.3 ± 0.1 µM, n = 11; cf. 18 ± 4 µM, n = 12 for α1β^LT).

This change is accompanied by a decrease in the Hill slope, from 2.0 ± 0.1 for the wild-type to 1.5 ± 0.1 and 1.2 ± 0.1 for α1β^LT and α1^LTβ, respectively. In order to compare EC₅₀ changes, dose-response curves were normalized to their maxima and averaged. When the three curves were fitted with the constraint of parallelism, the common Hill slope was 1.5, while the EC₅₀ decreased 10.1- and 74.5-fold for α1β^LT and α1^LTβ, respectively.

If the effect of the mutation is independent of which subunit carries it, our data show that there are more copies of α1 than β. This is in contrast with nicotinic and GABA receptors, which have only two copies of the main agonist binding subunit. Either a 4:1 or a 3:2 stoichiometry is possible. However, the former is not in agreement with the observation that in other channels in this superfamily each copy of this mutation produces the same amount of shift in the EC₅₀ and that such shifts are additive, namely:

n Í{special}(EC_50,wt/EC_50,n) = m Í{special}(EC_50wt/EC_50,m).

Thus if n is the number of α1 and m the number of β subunits, for a 4:1 stoichiometry, the values observed would give (74.5)^1/4 = 2.9 and (10.1)¹ = 10.1. On the contrary, for a stoichiometry of 3:2 our data yield (74.5)^1/3 = 4.2 and (10.1)^1/2 = 3.2. This finding is also in agreement with the conclusions of Kuhse et al. (1993) based on the expression of chimaeric receptor subunits.

This work was supported by the MRC and The Wellcome Trust (project grant 064652).