A single ring of charged amino acids at one end of the pore can control ion selectivity in the 5-HT3 receptor

Abstract

1. To determine the mechanisms by which cation- or anion-specific channels select between these ions, we have examined the role of electrostatic factors in a typical ligand-gated ion channel, the 5-hydroxytryptamine3 (5-HT3) receptor, by removal and/or insertion of rings of conserved charged residues at either end of the pore. 2. Neutralization of the negatively charged ring at the intracellular end of the channel (E-1'A) results in a nonselective channel (PNa/PCl=0.89). 3. Insertion of positively charged residues at the extracellular end of the pore either results in a nonfunctional receptor (A24'K) or one that remains cation-selective (PNa/PCl=110; S19'R). 4. Combining the removal of a negatively charged ring (E-1'A) with the insertion of a positively charged ring (S19'R), however, results in a channel that is predominantly anion-selective (PNa/PCl=0.37). 5. The data suggest that for the cation-selective 5-HT3 receptor, the control of selectivity exerted by charged rings at either end of the pore is dominated by the ring of negatively charged residues at the intracellular side of the channel. As changing the charge at this position has also been shown to change ionic selectivity in anion-selective receptors, these data suggest that electrostatic factors can control selectivity in the whole Cys-loop family.

Figure 1

Alignment of the amino-acid sequences of M2 and its flanking regions for representative cation- and anion-selective ligand-gated ion channel subunits. Sequences are shown in single-letter code for the murine 5-HT_3A receptor, the α7 subunit of the chicken nACh receptor, the α1 subunit of the rat glycine receptor and the α1 subunit of the bovine GABA_A receptor (North, 1995). Bold type indicates conserved charged amino-acid residues. Boxes indicate amino-acid residues that were mutated in the present study. The cytoplasmic (−4′), intermediate (−1′) and extracellular (20′) rings of charged residues bordering M2 in the cation-selective channels (Θ) as well as the conserved central leucine residue (^*) are also indicated.

Figure 2

Expression of WT and mutant 5-HT_3A receptors. (a) Normalized whole-cell recordings of the responses to 10 μM 5-HT (solid bar, 30 s) in HEK293 cells expressing WT, EA, SR or EASR mutant receptors. (b) Typical immunofluorescence labelling of nonpermeabilised HEK293 cells expressing the AK mutant receptor with an antiserum specific for the N-terminal region of 5-HT₃ receptors. Scale bar, 10 μM. (c) Typical maximal increases in the intracellular Ca²⁺ concentration induced by 10 μM 5-HT (solid bar) in individual HEK293 cells loaded with Fura-2 expressing WT or mutant receptors.

Figure 3

Properties of WT and mutant 5-HT_3A receptors. Concentration–response curves (a) and current–voltage plots (b) for WT and mutant 5-HT_3A receptors. Currents were recorded in symmetrical solutions and normalized curves plotted as mean±s.e.m. The EC₅₀ values for 5-HT derived from the normalized current (I/I_max) curves were 1.56, 1.68, 3.1 and 3.4 μM, and the Hill coefficients were 2.2, 3.1, 3.1 and 2.3, for WT, EA, SR and EASR receptors, respectively (n=50, 6, 9 and 15).

Figure 4

Determination of cation/anion permeability ratios of WT and mutant 5-HT_3A receptors. (a) Current–voltage curves of WT and mutant receptors in different solutions. (b) Plots of reversal potential versus extracellular Na⁺ (Na_o) activity for determination of relative cation/anion permeability ratios of WT and mutant 5-HT_3A receptors. The expected lines for P_Na/P_Cl=0 or ∞ are also shown. Data are means±s.e.m. for n values of 25 (WT), 6 (EA), 9 (SR) and 15 (EASR).

Figure 5

Model showing the location of E−1′ in M2. The model is based on the Pashkov NMR data for the nACh receptor, with electronic mutations and some rotamer corrections. All atoms are coloured CPK (N=blue, O=red, C=white) except for E−1′, which is green. The likely location of the membrane is indicated by pink bars.