Steroid interaction with a single potentiating site is sufficient to modulate GABA-A receptor function

Abstract

Neuroactive steroids are efficacious potentiators of GABA-A receptors. Recent work has identified a site in the alpha1 subunit of the GABA-A receptor, that is essential for potentiation by steroids. However, each receptor contains two copies of the alpha1 subunit. We generated concatemers of subunits so that the alpha1 subunits could be mutated separately and examined the consequences of mutations that remove potentiation by most neurosteroids (alpha1 Q241L, alpha1 Q241W). Concatemers were expressed in Xenopus laevis oocytes, and activation by GABA, potentiation by neurosteroids, and the agonist activity of piperidine-4-sulfonic acid (P4S) were determined. When the alpha1 Q241L mutation is present in alpha1 subunits the EC(50) for activation by GABA is shifted to higher concentration and potentiation by neurosteroids is diminished. When the alpha1 Q241W mutation is expressed, the EC(50) for GABA is shifted to lower concentration, the relative efficacy of P4S is increased, and potentiation by neurosteroids is diminished. Mutation of only one alpha1 subunit does not produce the full effect of mutating both sites. Overall, the data demonstrate that at a macroscopic level, the presence of a single wild-type steroid-binding site is sufficient to mediate responses to steroid, but both must be mutated to completely remove the effects of steroids. Differences between the two sites seem to be relatively subtle.

Fig. 1.

The structures of the concatemers and the steroids used. A, the concatemers used and the approximate location of the mutations (arrows). B, structures of the steroids used (left column as flat structures and right column as three-dimensional views).

Fig. 2.

Normalized concentration-response curves for responses to GABA. A, data for wild-type concatemers and concatemers containing the α1 Q241L mutation. The presence of the mutation in the βα concatemer or in both concatemers shifts the curve to higher [GABA]. B, data for wild-type concatemers and concatemers containing the α1 Q241W mutation. The presence of the mutation in either or both concatemers shifts the curve to lower [GABA]. The curves show fits of eq. 1 (values shown in Table 1). For easier comparison of the curves, the data have been normalized to the fit maximal amplitude for display in the figure (fit maximal amplitudes are shown in Table 1). Points show mean ± S.E. for 3 to 15 oocytes (small symbols in A show data for single oocytes).

Fig. 3.

Potentiation of wild-type concatemers by neurosteroids. The figure shows the potentiation ratio plotted against the fraction of maximal response for the control response for wild-type concatemers. A, potentiation produced by 1 μM 3α5αPOH for responses elicited by different concentrations of GABA. B, potentiation by 1 μM 3α5αPOH, 1 μM 3α5βPOH, and 1 μM 3α5αP (data for responses elicited by all concentrations of GABA). C, potentiation by 1 μM 3α5αPOH for responses elicited by a low concentration of GABA (•) and by 1 mM P4S (○). The data for the two sets overlap. Each point is the response of an oocyte. The dashed lines show a potentiation ratio of one (that is, no potentiation) and a fraction of maximal response of 0.11 (the cutoff for the means presented in Table 3). Mean potentiation ratios are shown in Table 3.

Fig. 4.

Potentiation of concatemers containing the α1 Q241L mutation by neurosteroids. Data are shown as in Fig. 3. A, data for potentiation by 1 μM 3α5αPOH. Note that the points for oocytes injected with the βα γβαL constructs overlap the data for wild-type concatemers. The presence of the mutation in the βαL or both concatemers reduces the potentiation ratio at similar levels of control activation. Similar data were obtained for potentiation by 1 μM 3α5βPOH and 1 μM 3α5αP (Supplementary Fig. 2; mean data in Table 3). B, potentiation by 1 μM 3α5αPOH for responses elicited by a low concentration of GABA (filled symbols, defined in A) and by 1 mM P4S (empty symbols). Potentiation for responses to GABA and P4S is affected essentially equivalently by the mutations.

Fig. 5.

Potentiation of concatemers containing the α1 Q241W mutation by neurosteroids. Data are shown as in Fig. 3. A, data for potentiation by 1 μM 3α5αPOH. Similar data were obtained for potentiation by 1 μM 3α5βPOH (Supplementary Fig. 3; mean data in Table 3). B, potentiation by 1 μM 3α5αPOH for responses elicited by a low concentration of GABA (filled symbols, defined in A) and by 1 mM P4S (empty symbols). Potentiation for responses to GABA and P4S is affected essentially equivalently by the mutations.

Fig. 6.

Fits of a simple model for potentiation. Data are shown for potentiation of responses to low concentrations of GABA by 1 μM 3α5αPOH. A, data for wild-type concatemers and concatemers containing the α1 Q241L mutation. The fitted curves are shown by the lines (solid, wild-type; long dash, βα γβαL; short dash, βαL γβα; dash-dot, βαL γβαL). Note that the lines for βα γβαL and wild-type concatemers overlap. B, similar data for wild-type concatemers and concatemers containing the α1 Q241W mutation. The ordinal scales in A and B are different.

Fig. 7.

The concentration-effect relationship for 3α5αPOH potentiation of responses elicited by GABA. The data are means for control fractional activation less than 0.11. Points show mean ± S.E. for 3 to 27 oocytes. Data points at 0.1 and 10 μM for βαl-γβα and βαW-γβαW constructs are the means ± S.D. for responses from two oocytes only. The lines simply connect the points.

Fig. 8.

The effects of the mutations in specific α1 subunits. This figure shows the ratios of parameters to the value for wild-type concatemers, expressed on a logarithmic scale, so that the overall pattern of effects can be readily compared. Data for concatemers containing the α1 Q241L mutation (A) and for concatemers containing the α1 Q241W mutation (B). The parts of each panel show the relative EC₅₀ for activation by GABA (part 1; values in Table 1), the mean potentiation ratios for responses elicited by low concentrations of GABA (parts 2-4, Tables 3 and 4), relative gating by 1 mM P4S (part 5, Table 1) and potentiation of responses to 1 mM P4S (part 6, Tables 2 and 4). The effect ratios for potentiation are calculated using the average of the potentiation ratio and the inverse of Z. Two qualitative points are illustrated in the Figure. First, the effect of a mutation in the γβα concatemer (black bars) is consistently less than the effect of the same mutation in the βα concatemer (hatched bar). Second, for the α1 Q241W mutation (B), the product of the effect ratios for mutations in single α1 subunits (the sum of the logarithms of the effect ratios, shown by •) is close to the effect seen when the mutation is present in both concatemers. However, this is not seen for the Q241L mutation (A), for which the presence of the mutation in both concatemers leads to a larger effect than expected.