A naturally occurring membrane-anchored Gαs variant, XLαs, activates phospholipase Cβ4

Abstract

Extra-large stimulatory Gα (XLα_s) is a large variant of G protein α_s subunit (Gα_s) that uses an alternative promoter and thus differs from Gα_s at the first exon. XLα_s activation by G protein-coupled receptors mediates cAMP generation, similarly to Gα_s; however, Gα_s and XLα_s have been shown to have distinct cellular and physiological functions. For example, previous work suggests that XLα_s can stimulate inositol phosphate production in renal proximal tubules and thereby regulate serum phosphate levels. In this study, we show that XLα_s directly and specifically stimulates a specific isoform of phospholipase Cβ (PLCβ), PLCβ4, both in transfected cells and with purified protein components. We demonstrate that neither the ability of XLα_s to activate cAMP generation nor the canonical G protein switch II regions are required for PLCβ stimulation. Furthermore, this activation is nucleotide independent but is inhibited by Gβγ, suggesting a mechanism of activation that relies on Gβγ subunit dissociation. Surprisingly, our results indicate that enhanced membrane targeting of XLα_s relative to Gα_s confers the ability to activate PLCβ4. We also show that PLCβ4 is required for isoproterenol-induced inositol phosphate accumulation in osteocyte-like Ocy454 cells. Taken together, we demonstrate a novel mechanism for activation of phosphoinositide turnover downstream of G_s-coupled receptors that may have a critical role in endocrine physiology.

Keywords: G protein; G protein–coupled receptor; Gαs; adenylate cyclase; inositol phosphate; phospholipase C; plasma membrane targeting; signal transduction.

Figure 1

XLα_sactivates PLCβ4.A, splicing of the XL exon to exons 2 to 13 at the GNAS locus results in XLα_s. The XL amino-terminal domain contains a proline-rich region (PRR) followed by a highly charged domain (HCD). Red asterisks denote two cysteine residues (C287 and C318). B, COS-7 cells were transfected with indicated plasmid constructs. About 24 h post-transfection, cells were incubated with F-10 media containing 1.5 mCi/well myo[2-³H(N)] inositol and assayed the next day for total [³H]inositol phosphate (IP) accumulation, using Dowex AGX8 anion exchange columns as detailed in the Experimental procedures section. ∗∗∗∗ One-way ANOVA test, Bonferroni post hoc test, p < 0.0001. C, concentration-dependent activation of PLCβ4 by XLα_s in COS-7 cells. COS-7 cells were transfected with indicated plasmid constructs. Coexpression of an increasing amount of XLα_s (0–200 ng) and a fixed amount of PLCβ4 (200 ng) results in increasing IP accumulation. ∗, ∗∗, ∗∗∗, and ∗∗∗∗ one-way ANOVA, Bonferroni post hoc test, p < 0.05, p < 0.01, p < 0.001, p < 0.0001, respectively. D, Western blot of XLα_s (HA tagged) shows increased XLα_s protein expression in corresponding to the amount of XLα_s plasmid transfected, whereas PLCβ4 expression is unchanged. E, IP accumulation in COS-7 cells transiently transfected with different Gα_s variants and PLCβ4. ∗, ∗∗, and ∗∗∗∗ One-way ANOVA, compared with PLCβ4, Bonferroni post hoc test, p < 0.05, p < 0.01, p < 0.0001, respectively. F, Western blot of XLα_s, Gα_slong, Gα_sshort, and Gα_sQL shows similar protein expression, whereas PLCβ4 expression is unchanged. Data combined from at least three independent experiments are shown as mean ± SEM. Alpha Stimulating activity polypeptide; GNAS, Guanine Nucelotide binding protein, HA, hemagglutinin; PLCβ4, phospholipase Cβ4; XLα_s, extra-large stimulatory Gα.

Figure 2

Direct and specific activation of PLCβ4 by XLα_sin a reconstituted enzyme assay.A, PLCβ4 and XLα_s were purified to ∼95% and ∼30%, respectively. B, titration of PLCβ4 activity with XLα_s and (C) Gα_q. D, recombinant XLα_s stimulates PLCβ4 in a concentration-dependent manner, whereas recombinant Gα_s does not increase PLCβ4 enzymatic activity. E, recombinant Gα_q stimulates PLCβ3 in a concentration manner, whereas XLα_s does not. Data combined from at least three independent experiments are shown as mean ± SEM. PLCβ4, phospholipase Cβ4; XLα_s, extra-large stimulatory Gα.

Figure 3

Activation of PLCβ4 by XLα_sis not nucleotide state dependent but is inhibited by Gβγ.A, specific activity of PLCβ4 in the presence of different concentrations of XLα_s with or without 30 μM AlCl₃ and 10 mM NaF (AlF₄⁻). B, COS-7 cells cotransfection with PLCβ4 and XLα_s or GTPase-deficient XLα_s (XLα_s R543H) results in higher IP accumulation. XLα_s R543H does not lead to a higher IP accumulation. ∗∗ and ∗∗∗∗ One-way ANOVA, Dunnett post test, p < 0.01, p < 0.0001, respectively. C, Western blot of XLα_s, XLα_sR543H shows similar protein expression, whereas PLCβ4 expression is unchanged. D, effect of addition of purified Gβ₁γ₂ on XLα_s-activated PLCβ4 in reconstituted assay. ∗∗∗ and ∗∗∗∗ Two-way ANOVA, Dunnett post test, p < 0.001, p < 0.0001, respectively, or in E, cellular assay, ∗∗∗∗ one-way ANOVA, Dunnett post test, p < 0.0001. F, Western blot from COS-7 cells coexpressing PLCβ4 with or without Gβ₁ and Gγ₂ as indicated in E. Data combined from three to four independent experiments are shown as mean ± SEM. IP, inositol phosphate; PLCβ4, phospholipase Cβ4; XLαs, extra-large stimulatory Gα.

Figure 4

Loss-of-function mutation in XLα_sdoes not affect its ability to activate PLCβ4.A, mutations at glutamate-268 and glycine-226 (blue) in Gα_s (red) render Gα_s protein that has impaired agonist-induced cAMP generation. Green helix denotes the switch II region that engages adenylyl cyclase (tan). B, Gα_sKO HEK293 cells transfected with β2 adrenergic receptor, cAMP Glo, and indicated plasmid constructs (YFP, Gα_s, XLα_s, and XLα_s^Mut) and cAMP content following isoproterenol addition was assayed as described in the Experimental procedures section. C, COS-7 cells were transfected with indicated plasmid constructs, and total IP accumulation was assayed as described for Figure 1A. ∗∗∗∗ and ∗∗∗ One-way ANOVA test, Dunnett post hoc test, p < 0.0001, p < 0.001, respectively. D, Western blots show similar protein expression of XLα_s and XLα_s^Mut, whereas PLCβ4 expression is unchanged. Data combined from three to four independent experiments are shown as mean ± SEM. HEK293, human embryonic kidney 293 cell line; IP, inositol phosphate; PLCβ4, phospholipase Cβ4; XLα_s, extra-large stimulatory Gα.

Figure 5

Plasma membrane localization of XLα_sis important for PLCβ4 activation.A, schematic diagram of XLα_s domain structure, WT XLα_s consists of an XL domain, which contains a highly charged domain (HCD), and a proline-rich region (PRRP). Asterisks depict the two conserved cysteines in the XL domain. Immunocytochemical analysis of subcellular distribution for WT and Cys-to-Ser mutants of XLα_s in HEK293 cells by using an anti-HA antibody. HEK293 cells were transiently transfected with expression constructs encoding HA-tagged WT or Cys-to-Ser mutants of XLα_s (Cys-287 and Cys-318). Twenty-four hours after transfection, subcellular localizations of these XLα_s mutants were investigated. The scale bar represents 5 μM. B, total IP accumulation in COS-7 cells expressing WT XLα_s or Cys-to-Ser mutants of XLα_s and PLCβ4. ∗∗∗∗p < 0.0001 compared with PLCβ4, one-way ANOVA, Tukey post test, ### and #### p < 0.001, p < 0.0001, respectively, compared with XLα_s + PLCβ4, one-way ANOVA, Tukey post test. Western blots show expression of PLCβ4 and different XLα_s constructs tested in the IP accumulation assays. Data combined from three to four independent experiments are shown as mean ± SEM. HA, hemagglutinin; HEK293, human embryonic kidney 293 cell line; IP, inositol phosphate; PLCβ4, phospholipase Cβ4; XLαs, extra-large stimulatory Gα.

Figure 6

Identifying the region in XLα_sthat activates PLCβ4. Membrane-targeting intact Gα_s activates PLCβ4. A, sequence alignment of XLα_s and the amino terminus of Gα_s long (through Gα_s amino acid 138. The remainder of Gα_s is identical to XLα_s). Red asterisks denote XLα_s cysteine 287 and 318. Arrow marks the beginning of the post PRR-XLα_s and the QMR-XLα_s sequence that follow immediately after Lyn membrane targeting motif (Lyn-QMR-XLα_s). Red box denotes the N-terminal α-helical Gβγ interaction domains in XLα_s and Gα_s. B–D, COS-7 cells were transfected with indicated plasmid constructs, and total IP accumulation was assayed as described for Figure 1A; protein expression was examined by SDS-PAGE and immunoblotting. For D, above the graph is a schematic depicting the sequence at Lyn-N-terminal Gα_s junction in Lyn-Gα_s. ∗∗ and ∗∗∗∗ One-way ANOVA test, Dunnett post hoc test, p < 0.01, p < 0.0001, respectively. Data combined from three to four independent experiments are shown as mean ± SEM. IP, inositol phosphate; PLCβ4, phospholipase Cβ4; PRR, prOline-rich region; XLαs, extra-large stimulatory Gα.

Figure 7

Isoproterenol-induced IP accumulation in Ocy454 cells is mediated by PLCβ4.A, isoproterenol induces a concentration-dependent increase in IP accumulation in osteocyte-like Ocy454 cells. ∗p < 0.05 and ∗∗p < 0.01, one-way ANOVA, Dunnett post test. B, representative immunoblot showing reduced PLCβ4 expression in Ocy454 cells transiently transfected with PLCβ4 siRNA or control scramble siRNA (SmartPool) (Scrm, scrambled). C, IP1 concentrations were significantly diminished at baseline and after isoproterenol stimulation in PLCβ4 knockdown-Ocy454 cells. ∗ and ∗∗∗ Two-way ANOVA test, Sidak post hoc test, p < 0.05, p < 0.001, respectively. Data combined from five independent experiments are shown as mean ± SEM. IP, inositol phosphate; PLCβ4, phospholipase Cβ4.