Site-specific effects of neurosteroids on GABAA receptor activation and desensitization

  1. Yusuke Sugasawa
  2. Wayland WL Cheng
  3. John R Bracamontes
  4. Zi-Wei Chen
  5. Lei Wang
  6. Allison L Germann
  7. Spencer R Pierce
  8. Thomas C Senneff
  9. Kathiresan Krishnan
  10. David E Reichert
  11. Douglas F Covey
  12. Gustav Akk
  13. Alex S Evers

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

This study examines how site-specific binding to three identified neurosteroid-binding sites in the α 1 β 3 GABA A receptor (GABA A R) contributes to neurosteroid allosteric modulation. We found that the potentiating neurosteroid, allopregnanolone, but not its inhibitory 3β-epimer epi-allopregnanolone, binds to the canonical β 3 (+)–α 1 (-) intersubunit site that mediates receptor activation by neurosteroids. In contrast, both allopregnanolone and epi-allopregnanolone bind to intrasubunit sites in the β 3 subunit, promoting receptor desensitization and the α 1 subunit promoting effects that vary between neurosteroids. Two neurosteroid analogues with diazirine moieties replacing the 3-hydroxyl (KK148 and KK150) bind to all three sites, but do not potentiate GABA A R currents. KK148 is a desensitizing agent, whereas KK150 is devoid of allosteric activity. These compounds provide potential chemical scaffolds for neurosteroid antagonists. Collectively, these data show that differential occupancy and efficacy at three discrete neurosteroid-binding sites determine whether a neurosteroid has potentiating, inhibitory, or competitive antagonist activity on GABA A Rs.

Article activity feed

  1. Version published to 10.7554/elife.55331 on eLife
  2. eLife

    ###This manuscript is in revision at eLife

    The decision letter after peer review, sent to the authors on April 22, 2020, follows.

    Summary

    This report provides significant new information about the mechanisms of neurosteroid enhancement and inhibition of GABA-A receptor (GABAR) function. This study builds on an earlier investigation by the same group (Chen et al. PLOS Biology 2019) showing that photoactive NS ligands can bind to three distinct sites on α1β3 GABARs - the canonical intersubunit site at the interface between the transmembrane domains (TMDs) of adjacent subunits and additional intrasubunit sites located within the TMDs of the alpha and beta subunits. In the current study, combining [3H]muscimol radioligand binding assays, site identification by photoaffinity labeling, and electrophysiological analyses of steroid modulation of wildtype and mutant α1β3 GABARs, the authors suggest that the overall functional effect of a given NS molecule is dependent upon which binding sites are targeted, with binding to the intersubunit site causing positive allosteric modulation (PAM), whilst occupancy of the intrasubunit sites appear to promote desensitization and negative allosteric modulation (NAM). Given the physiological significance of neurosteroids, elucidating how these structurally similar compounds can act as positive, negative or null modulators is clearly important.

    Essential Revisions

    1. The electrophysiological data presented (changes in steady state desensitization current magnitudes) is insufficient to conclude that NAM steroids inhibit GABAR function by stabilizing a desensitized state. Additional experiments such as co-application of agonist + NS and monitoring desensitization kinetics would be informative. Measuring the rate of recovery from agonist-induced desensitization in the presence of neurosteroids might also be helpful. While the data presented can be interpreted as changes in desensitization, the authors should discuss that alternative models are also possible. For example, it has been proposed that selectively stabilizing a pre-active state can result in changes in macroscopic desensitization (Gielan and Corringer, J. Physiol. 2018).

    2. Mutant receptors were not assayed for their sensitivities to agonist before measuring effects of neurosteroids. The functional assays and binding experiments need to be done at a consistent fractional EC value for each mutant construct being analyzed. For example, if the apparent Kd for muscimol has shifted substantially, the observed potentiation of muscimol binding by a neurosteroid will be artificially high or low. The is also true for experiments measuring neurosteroid potentiation/inhibition of functional activation by GABA.

    3. In the result section, there are concerns about quantitatively comparing electrophys data and [3H]muscimol data (measured at different agonist concentrations and time periods). Are the methods reliable enough to infer that the small changes in Popen and Pdesensitized are real? In some cases, data are not shown. Inherent methodological limitations of two-electrode voltage clamping (e.g. slow ligand exchange) raises concerns that authors are over interpreting the data. As it stands, the comparison seems to be a bit of a reach and in this reviewers' opinion does not significantly add to the paper.

    4. While having three distinct sites for NS binding to GABARs does fit with aspects of the data, it's noteworthy that with the suggested model, there are three ligands that bind to all three sites, 3a5aP, KK148 & KK150, but each has a distinct functional profile, PAM, NAM via stabilizing desensitization, and competitive antagonist, respectively. This implies that divergence in function is dependent upon differential binding/efficacy at these three sites, presumably due to the ligand sitting in each site in a different orientation. While the observation from the [3H]muscimol binding experiments suggests that 3a5aP binds to the b3 intrasubunit site with lower affinity, the data presented in Fig 6B also suggest that binding of 3a5aP to the intersubunit and a1 intrasubunit sites works synergistically to increase muscimol binding. The reasoning being because with both sites intact, the Emax for muscimol binding is 374%, whereas mutating these sites individually causes similar decreases in Emax (to 159% and 146%). This implies an allosteric interaction between these binding sites, a conclusion which the authors also reach in their previous publication (Chen et al 2019). This makes interpretation of the effects of mutations in these two sites (and possibly also the beta intrasubunit site) difficult to interpret and to use to specifically dissociate a mutations effects on NS actions to binding to one particular site. The authors need to thoroughly discuss this concern/limitation.

    5. The demonstration that steroids apparently enhance [3H]muscimol binding affinity without changing the number of sites (Fig 6 supplement 1) is in contrast to past reports from multiple labs that [3H]muscimol binding (to brain membranes) is characterized by high and low affinity components and that steroids and other GABAR positive allosteric modulators increase the number of high affinity sites with little effect on their binding affinity. Please discuss. In addition, we would like to see presented in supplementary material representative experimentally determined [3H]muscimol binding curves (total and non-specific vs [ 3H]musc concentration, not just the calculated Bspec of fig6 supp fig 1). In their methods (p.25) they say that they determined [3H]muscimol binding isotherms from 0.3 nM to 1 uM [3H]muscimol at a radiochemical specific activity of 2 Ci/mmol. It is surprising that they can go to micromolar concentrations with such small uncertainties, and it is crucial to their claim steroids produce only shifts of affinity, not shifts of Bmax.

    6. [3H]muscimol binding is measured on cell homogenates over a time scale of hours. There seems no reason to "infer" that 3α5αβP increases [3H]muscimol binding by stabilizing an active state while 3β5αP stabilizes a desensitized state. By my reading, the previous studies (13,14) report that the αV256S mutation removes the "inhibitory" effects of sulfated steroids and 3β5αP, not the "desensitizing" effects, this should be more clearly articulated in this manuscript. This report will be strengthened by avoiding unnecessary overinterpretation, and leave it for future studies to determine whether there is any measurable quantity of receptors in an active state under the conditions of the [3H]muscimol equilibrium binding assay.

    7. Given the expectations that some of the neurosteroids stabilize a desensitized state, do they "fit" in the proposed intrasubunit sites in, for example, one of the published presumed desensitized-state structures of the α1β3γ2 receptor?

    8. A more thorough discussion of why recently solved GABAR structures have not resolved intrasubunit neurosteroid binding sites is warranted.

  3. Version published to 10.1101/2020.04.27.063404v1 on bioRxiv