HomeHypertensionVol. 57, No. 5Response to Is GPR30 the Membrane Aldosterone Receptor Postulated 20 Years Ago? Free AccessReplyPDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessReplyPDF/EPUBResponse to Is GPR30 the Membrane Aldosterone Receptor Postulated 20 Years Ago? Ross D. Feldman and Robert Gros Ross D. FeldmanRoss D. Feldman Search for more papers by this author and Robert GrosRobert Gros Search for more papers by this author Originally published21 Mar 2011https://doi.org/10.1161/HYPERTENSIONAHA.111.171157Hypertension. 2011;57:e17Other version(s) of this articleYou are viewing the most recent version of this article. Previous versions: January 1, 2011: Previous Version 1 We appreciate the correspondents' comments regarding the significance of our recent findings.1 Furthermore, we are appreciative of the work of Prof Wehling and his colleagues,2 whose landmark studies have been an important stimulus for our investigations.With regard to any conclusions regarding the specificity of previously believed estrogen receptor and mineralocorticoid receptor (MR)–selective antagonists, we would still suggest caution. The previously believed estrogen receptor–specific antagonist, ICI-182780, has also been shown to interact with GPR30,3,4 although these findings have not been universal.5 Our findings, most clearly with eplerenone, but also with spironolactone, should further undermine one's confidence in ascribing “rapid” steroid effects to “classic” steroid receptors, in the absence of consideration of their possible interactions with GPR30. Therefore, we have advocated that the determination of receptor specificity of these steroid receptor ligands mediating rapid effects should be based on a combination of pharmacological and genetic approaches, namely, by use of small-interfering RNA/knockdown and heterologous expression models.With regard to the correspondents' specific concerns, the actions of eplerenone to, at least, partially block the GPR30-dependent effects of aldosterone in the stimulation of extracellular signal–regulated kinase phosphorylation were most unambiguously demonstrated in the rat aortic endothelial cell system (ie, with endogenous GPR30 expression but in the absence of detectable MR). The correspondents have suggested that the partial inhibitory effects of eplerenone on aldosterone's actions seen with GPR30 gene transfer in vascular smooth muscle cells represented eplerenone's effects on a residual population of MRs. However, as we demonstrated, with GPR30 expression, MR was downregulated. In this essentially, MR-null system, the GPR30 antagonist, G15, fully attenuated the actions of aldosterone. In contrast, G15 had no effect in these cells in the absence of GPR30 expression. As we discussed, the most supportable explanation for these findings is that eplerenone is a ligand for GPR30.Our findings are not in direct opposition to Prof Wehling's long-standing assertion that resistance to the effects of spironolactone and related drugs could be used as an index of a non–MR-mediated aldosterone effect. First, we have yet to test the effects of these agents in many of the model systems used by Prof Wehling in his studies, that is, cAMP generation, calcium mobilization, and aldosterone radioligand binding assays. In addition, we have been able to see only partial inhibitory effects of spironolactone and eplerenone. These partial effects may not be expected to be universally detectable across all assay systems and would be expected to depend on the specifics of the coupling efficiency of each individual signaling pathway. Thus, we would suggest that the ability of eplerenone to block the possible GPR30-dependent effects in the assay systems on which Prof Wehling developed his hypothesis remains an open question.Ross D. FeldmanRobert Gros University of Western Ontario Robarts Research Institute London, Ontario, CanadaSources of FundingThe studies from our laboratories cited in this correspondence were supported by a grant-in-aid from the Heart and Stroke Foundation.DisclosuresNone.FootnotesLetters to the Editor will be published, if suitable, as space permits. They should not exceed 1000 words (typed double-spaced) in length and may be subject to editing or abridgment.
Read full abstract