Research ArticleG Protein Signaling

Dual phosphorylation of Ric-8A enhances its ability to mediate G protein α subunit folding and to stimulate guanine nucleotide exchange

See allHide authors and affiliations

Science Signaling  29 May 2018:
Vol. 11, Issue 532, eaap8113
DOI: 10.1126/scisignal.aap8113
  • Fig. 1 Ric-8A is constitutively phosphorylated in cells.

    (A) Gel mobility shift assays of endogenous Ric-8A immunoprecipitated (IP) from cytosolic and detergent-extracted membrane fractions (5× material loaded) of HEK293 cells, and recombinant Ric-8A purified from E. coli or insect cells before and after alkaline phosphatase treatment. Top: Standard SDS-PAGE. Bottom: Phos-tag PAGE. Ric-8A* may have phosphosite(s) partially resistant to alkaline phosphatase. The molecular mass markers (75 and 50 kDa) do not accurately reflect the true masses of Ric-8A proteins on the Phos-tag PAGE. Data are representative of more than three independent experiments. (B) Anion exchange chromatography resolution of recombinant Ric-8A purified from E. coli or from insect cells before and after alkaline phosphatase treatment. Data are representative of more than three independent experiments. (C) Mass spectra of Ric-8A proteins were obtained through whole-protein ESI/MS analysis. Spectra are of insect cell–purified recombinant rat Ric-8A (blue trace) and alkaline phosphatase–treated Ric-8A (red trace). (D) The ESI-MS/MS spectrum of E. coli–purified Ric-8A revealed a completely unmodified protein (mass, 60156.2 Da) with a series of Hepes buffer adducts. MW, molecular weight.

  • Fig. 2 Identification of the Ric-8A sites phosphorylated by CK2.

    (A) GPS 3.0 was used to evaluate sites within Ric-8A with potential for CK2-mediated phosphorylation (41). The prediction was run at the high-stringency setting using a maximal false-positive rate threshold cutoff of 2%. Higher relative scores represent greater potential for phosphorylation. (B) Phosphosite identification in rat Ric-8A purified from insect cells and from E. coli–produced Ric-8A treated with CK2 was made by MS analysis of peptides from tryptic and chymotryptic digests. Red boxed residues were phosphorylated sites in insect cell–produced Ric-8A, yellow boxed residues were phosphorylated sites in E. coli–produced Ric-8A treated with CK2, and blue boxed residues were phosphorylated sites obtained for both proteins. All phosphosites were identified by Mascot software and confirmed by analysis of CID spectra. Peptide coverage of the three C-terminal phosphosites (Ser522, Ser523, and Ser527) was not attained for insect cell–produced Ric-8A. The 10-residue acidic sequence that contains the two consecutive CK2 sites of Ser435 and Thr440 is underlined. (C) The 10-residue acidic sequence is highly conserved across vertebrate Ric-8 proteins. The phosphosites corresponding to rat Ric-8A Ser435 and Thr440 are invariantly serine or threonine residues in all homologs examined. Dictyostelium Ric-8 is a shortened protein that lacks the acidic sequence stretch altogether. (D) Insect cell–produced rat wild-type (WT), S435A, and T440A Ric-8A proteins were treated with or without CK2 and then analyzed by Western blotting with an antibody specific for the CK2 consensus site (pS/pT)-D-X-E, which recognizes the phosphorylated Thr440 site (p-Thr440), the 6383 Ric-8A antiserum that recognizes the phosphorylated Ser435 site (p-Ser435), and the 1184 antiserum that recognizes Ric-8A. (E) Insect cell–produced mouse Ric-8B was treated with and without alkaline phosphatase and analyzed by Western blotting with the Ric-8A phosphosite-specific antibodies that also detect mouse Ric-8B p-Ser468 and p-Ser472, respectively. Ric-8B antiserum 2413 was used to detect the Ric-8B alkaline phosphatase–dependent gel shift after protein resolution by Phos-tag PAGE. (F) Purified Ric-8A from insect cells was treated with or without alkaline phosphatase, whereas E. coli–produced Ric-8A was treated with or without CK2. The proteins were analyzed by Western blotting with the p-Thr440, p-Ser435, and 1184 antibodies. (G) Insect-produced WT Ric-8A and the indicated Ric-8A alanine substitution mutant proteins produced in E. coli were treated with or without CK2 and analyzed by Western blotting with the p-Thr440 and 1184 Ric-8A antibodies. The proteins were also resolved by Phos-tag PAGE and analyzed by Western blotting with the 1184 Ric-8A antibody, as indicated. (H) WT, T440A, and S435A/T440A Ric-8A-ΔCT purified from E. coli were treated with or without CK2, resolved by SDS-PAGE and Phos-tag PAGE, and analyzed by Western blotting with the p-Thr440, p-Ser435, and 1184 Ric-8A antibodies, as indicated. Data in (D) to (H) are representative of more than three independent experiments.

  • Fig. 3 i1 binds phosphorylated Ric-8A with a higher affinity than it has for unphosphorylated Ric-8A.

    A concentration series of purified Gαi1-CFP (1 to 1000 nM) was incubated with insect cell– or E. coli–produced GST-tagged Ric-8A or GST to determine the nonspecific component. Glutathione-coated microspheres were used to adsorb the GST-Ric-8A–Gαi1-CFP complexes before flow cytometric measurement of bead-associated fluorescence. Data were gated to include only singlet beads and then were analyzed for median fluorescence intensity (MFI). Nonspecific binding (GST only) was subtracted from total to yield the percentage of Gαi1-CFP bound specifically to Ric-8A, with 100% indicating the maximum fluorescence of saturated binding. The plotted data are the result of three independent experiments, and error bars indicate the SEM. The data were fitted to one-site specific binding curves with GraphPad Prism software.

  • Fig. 4 Requirements of the Thr440 and Ser435 phosphosites for Ric-8A–stimulated Gαq steady-state GTP hydrolysis activity.

    (A to F) Purified Gαq (50 nM) was incubated with the indicated concentrations of purified Ric-8A proteins and [ γ-32P]GTP. The linear rate of GTP hydrolysis was determined by measuring the production of free 32Pi. (A) Full-length WT Ric-8A proteins purified from insect cells (Hi5) and treated with or without alkaline phosphatase or purified from E. coli and treated with or without CK2 were tested for their ability to stimulate the steady-state GTPase activity of Gαq. (B) WT Ric-8A-ΔCT protein purified from E. coli and treated with or without CK2 was tested for its ability to stimulate the steady-state GTPase activity of Gαq. (C to F) WT Ric-8A purified from insect cells (Hi5) was used as a positive standard of phosphorylated Ric-8A activity for comparison to the activities of (C) Ric-8A–T440A purified from insect cells and Ric-8A–T440A purified from E. coli and treated with or without CK2, (D) Ric-8A–S435A purified from insect cells and Ric-8A–S435A purified from E. coli and treated with or without CK2, (E) Ric-8A–T440D purified from insect cells and Ric-8A–T440D purified from E. coli and treated with or without CK2, and (F) Ric-8A–S435D purified from insect cells and Ric-8A–S435D purified from E. coli and treated with or without CK2. Data were plotted on semilog graphs and fitted to one-phase exponential association functions using GraphPad Prism. Experiments were performed in triplicate. Error bars indicate the SEM and were sometimes smaller than the size of the plotted symbols.

  • Fig. 5 Efficient Gα subunit folding in a WGE/Ric-8A reconstitution assay is dependent on Ric-8A Ser435 and Thr440 phosphorylation.

    (A) Top: The mRNA encoding a Gαi1 fusion protein with an internal GFP tag was translated in WGE that had been reconstituted with purified rat Ric-8A. Bottom: Folding of the fusion protein was monitored by evolution of GFP fluorescence. (B) Samples for SDS-PAGE were attained at the conclusion of the kinetic Gαi1-GFP translation/folding reactions and analyzed by Western blotting (WB) to detect Gαi1, GFP, and Ric-8A. Data are representative of more than three experiments. (C) Gαi1-GFP mRNA was introduced into WGE translation/folding reactions reconstituted with the indicated concentrations of insect cell–produced WT Ric-8A or E. coli–produced WT Ric-8A with or without CK2 pretreatment. Maximal Gαi1-GFP relative fluorescence units (RFUs) at 535 nm were plotted versus Ric-8A concentration on semilog plots. The data were fitted to variable Hill slope, four-parameter concentration response functions using the following equation in GraphPad Prism: Y = Ymin + (YmaxYmin)/(1 + 10((LogEC50X)*Hillslope)). EC50 values were estimated from the fitted line functions. Experiments were performed in triplicate, and data are means ± SEM. (D) E. coli– and insect cell–purified Ric-8A proteins were incubated for 5 hours in WGE, resolved by SDS-PAGE and Phos-tag PAGE, and then analyzed by Western blotting. Data are representative of more than three experiments. (E to G) Insect cell–produced WT Ric-8A was used as phosphorylated Ric-8A standard in the WGE/Gαi1-GFP folding assay to compare to the actions of (E) E. coli–produced Ric-8A S435A or T440A, (F) E. coli–produced Ric-8A S435A/T440A treated with or without CK2, or (G) E. coli–produced Ric-8A-ΔCT treated with or without CK2. Data were processed as described in (C). Experiments were performed in triplicate, and data are means ± SEM.

  • Fig. 6 Ric-8A–T440 is required for efficient G protein chaperoning activity and signaling in cells.

    (A to D) A HEK293T cell line (G7) lacking RIC-8A was developed using CRISPR-Cas9 technology. (A to C) Crude membrane preparations from RIC-8A–null cells stably expressing WT or the indicated mutant rat Ric-8A proteins with single alanine or aspartic acid point mutations at the regulatory phosphosites were subjected to quantitative Western blotting for Ric-8A, Gα13, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). (D) Relative Gα13 abundances were quantified by pixel densitometry analysis and normalized to the GAPDH signal. Data are means ± SEM of three independent experiments. Statistical significance was determined by one-way analysis of variance (ANOVA), Dunnett’s multiple comparison to WT: *P < 0.05, **P < 0.005, ***P < 0.0001; n.s., not significant. (E) The effects of Ric-8A on GPCR-mediated G13 signaling activity were measured by dual SRE-Luc assay. RIC-8A–null cells were transiently transfected with plasmids expressing constitutively active GPR56, the SRE-Luc reporter, Renilla luciferase, and the indicated amounts of plasmids encoding WT and phosphosite mutant Ric-8A proteins. The accumulated firefly luciferase signal was measured 24 hours after transfection and normalized to the Renilla luciferase signal. Data were normalized to the signal generated from RIC-8A–null cells expressing the luciferase plasmids alone. Data are means ± SEM of three experiments.

  • Fig. 7 Locomotor and postural defects in ric-8 S467A or S472A C. elegans mutants.

    (A to D) WT (N2) and CRISPR-modified C. elegans strains expressing ric-8 with alanine mutations at sites Ser467 or Ser472 were treated with 10 μM phorbol ester or ethanol (vehicle) for 60 min. (A) Representative images show the rod-like body posture phenotype of S467A and S472A mutants on vehicle plates, as compared to the sinusoidal body posture of N2 controls. Phorbol ester–exposed animals had a hyperflexive phenotypes in N2 controls and S467A and S472A mutants, denoted by the omega-shaped body postures. Red arrowheads denote eggs laid by ric-8 S467A animals in response to phorbol ester. Scale bar, 0.5 mm. (B) The numbers of body bends were quantified in freely moving animals over a 2-min period. (C) The maximum bending amplitude was quantified as a measure of hyperflexion. (D) Track length (forward + reverse movement) was quantified in freely moving animals for 2 min before and after 60-min phorbol ester treatment using WormLab acquisition and analysis software. For all experiments, ten 1-day-old adults were assayed per plate with 10 replicated plates per strain for all experiments. Statistical significance was determined by one-way ANOVA with Tukey’s post hoc test. **P < 0.005, ***P < 0.0001.

  • Fig. 8 Predicted orientation of the CK2 phosphorylation sites in Ric-8A.

    A structural prediction of Ric-8A was made using the I-TASSER server to depict four views of the protein with 90° rotation around a vertical axis (I to IV) (67, 68). The experimentally identified CK2 phosphorylation sites, Ser435, Thr440, Ser522, Ser523, and Ser527, are labeled and colored yellow. The N terminus of Ric-8A is located at the top in each representation. The surface area is colored according to local pKa (logarithmic acid dissociation constant) from 0 (red) to 14 (blue). Rendering was produced using Discovery Studio 4.0 (Accelrys Software Inc.) to visualize structure coordinates. Importin-β bound to Ran small GTPase [left; Protein Data Bank (PDB): 1IBR] is composed of repeated HEAT α-helical elements and has a crescent shape that is similar to the predicted structure of the ~430 N-terminal residues of Ric-8A (73). An acidic surface important for Ran binding in the importin-β crescent is denoted with an arrow. A negatively charged surface predicted to face into the Ric-8A crescent is composed of acidic residues and has a negative charge that is contributed by the experimentally verified CK2-phosphorylated residues.

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/11/532/eaap8113/DC1

    Fig. S1. Rat Ric-8A sequence and MS coverage of tryptic and chymotryptic peptides.

    Fig. S2. Enrichment of the IgG fraction from rabbit Ric-8A p-Ser435 antiserum.

    Fig. S3. Stimulation of GTPγS binding to Gαi, Gαq, and Gα13 by phosphorylated and unphosphorylated Ric-8A.

    Fig. S4. Western blotting analysis of Ric-8A aspartic acid phosphorylation site mutants.

    Fig. S5. Validation of the CRISPR-Cas9–generated RIC-8A–null HEK293T cell line.

    Fig. S6. Locomotion and body posture defects in C. elegans ric-8A–S472A mutants are rescued by phorbol ester.

    Table S1. MS/MS analysis of purified recombinant Ric-8A.

    Table S2. Representative CID analysis of Ric-8A to identify specific phosphorylation sites after tryptic digest and MS analysis.

    Movie S1. C. elegans ric-8A–S472A mutant movements.

    Movie S2. C. elegans ric-8A–S472A mutant movements after phorbol ester treatment.

  • Supplementary Materials for:

    Dual phosphorylation of Ric-8A enhances its ability to mediate G protein α subunit folding and to stimulate guanine nucleotide exchange

    Makaía M. Papasergi-Scott, Hannah M. Stoveken, Lauren MacConnachie, Pui-Yee Chan, Meital Gabay, Dorothy Wong, Robert S. Freeman, Asim A. Beg, Gregory G. Tall*

    *Corresponding author. Email: gregtall{at}med.umich.edu

    This PDF file includes:

    • Fig. S1. Rat Ric-8A sequence and MS coverage of tryptic and chymotryptic peptides.
    • Fig. S2. Enrichment of the IgG fraction from rabbit Ric-8A p-Ser435 antiserum.
    • Fig. S3. Stimulation of GTPγS binding to Gαi, Gαq, and Gα13 by phosphorylated and unphosphorylated Ric-8A.
    • Fig. S4. Western blotting analysis of Ric-8A aspartic acid phosphorylation site mutants.
    • Fig. S5. Validation of the CRISPR-Cas9–generated RIC-8A–null HEK293T cell line.
    • Fig. S6. Locomotion and body posture defects in C. elegans ric-8A–S472A mutants are rescued by phorbol ester.
    • Table S1. MS/MS analysis of purified recombinant Ric-8A.
    • Legend for table S2
    • Legends for movies S1 and S2

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Table S2 (Microsoft Excel format). Representative CID analysis of Ric-8A to identify specific phosphorylation sites after tryptic digest and MS analysis.
    • Movie S1 (.mp4 format). C. elegans ric-8A–S472A mutant movements.
    • Movie S2 (.mp4 format). C. elegans ric-8A–S472A mutant movements after phorbol ester treatment.

    © 2018 American Association for the Advancement of Science

Stay Connected to Science Signaling

Navigate This Article