Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Sep 26;114(39):E8224-E8233.
doi: 10.1073/pnas.1712176114. Epub 2017 Sep 11.

RABIF/MSS4 Is a Rab-stabilizing Holdase Chaperone Required for GLUT4 Exocytosis

Affiliations
Free PMC article

RABIF/MSS4 Is a Rab-stabilizing Holdase Chaperone Required for GLUT4 Exocytosis

Daniel R Gulbranson et al. Proc Natl Acad Sci U S A. .
Free PMC article

Abstract

Rab GTPases are switched from their GDP-bound inactive conformation to a GTP-bound active state by guanine nucleotide exchange factors (GEFs). The first putative GEFs isolated for Rabs are RABIF (Rab-interacting factor)/MSS4 (mammalian suppressor of Sec4) and its yeast homolog DSS4 (dominant suppressor of Sec4). However, the biological function and molecular mechanism of these molecules remained unclear. In a genome-wide CRISPR genetic screen, we isolated RABIF as a positive regulator of exocytosis. Knockout of RABIF severely impaired insulin-stimulated GLUT4 exocytosis in adipocytes. Unexpectedly, we discovered that RABIF does not function as a GEF, as previously assumed. Instead, RABIF promotes the stability of Rab10, a key Rab in GLUT4 exocytosis. In the absence of RABIF, Rab10 can be efficiently synthesized but is rapidly degraded by the proteasome, leading to exocytosis defects. Strikingly, restoration of Rab10 expression rescues exocytosis defects, bypassing the requirement for RABIF. These findings reveal a crucial role of RABIF in vesicle transport and establish RABIF as a Rab-stabilizing holdase chaperone, a previously unrecognized mode of Rab regulation independent of its GDP-releasing activity. Besides Rab10, RABIF also regulates the stability of two other Rab GTPases, Rab8 and Rab13, suggesting that the requirement of holdase chaperones is likely a general feature of Rab GTPases.

Keywords: CRISPR screen; GLUT4; Rab GTPase; exocytosis; vesicle transport.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
Dissection of insulin-stimulated GLUT4 exocytosis using CRISPR genetic screens. (A) Diagram of the GFP-GLUT4-HA reporter used to monitor insulin-dependent GLUT4 trafficking. After the translocation of the reporter, the HA epitope is exposed to the cell exterior (60). (B) Illustration of the genome-wide genetic screen of insulin-stimulated GLUT4 exocytosis. (C) Ranking of genes in the CRISPR screen based on the P value. Each dot represents a gene. Genes above the horizontal line were tested in the pooled secondary screen. A gene is shown as a large dot if it was validated in the secondary screen. Other genes are shown as small gray dots. Selected known or validated regulators of GLUT4 exocytosis are labeled. (D) Selected genes from the screen were individually mutated in mouse adipocytes using CRISPR-Cas9. Effects of the mutations on insulin-stimulated GLUT4 exocytosis were measured by flow cytometry. Error bars indicate SD. ***P < 0.001.
Fig. S1.
Fig. S1.
Comparative analyses of insulin-stimulated GLUT4 exocytosis in adipocytes and HeLa cells. HeLa cells (A) and mouse adipocytes (B) expressing the GFP-GLUT4-HA reporter were either untreated or treated with 100 nM insulin for 30 min before the surface reporters were stained with anti-HA antibodies and APC-conjugated secondary antibodies. To inhibit insulin signaling, the cells were incubated with 100 nM wortmannin for 10 min before insulin stimulation. The ratio of APC and GFP fluorescence was measured by flow cytometry, and the mean fluorescence was normalized to that of untreated WT cells. Error bars indicate SD. RAB10- and TBC1D4-KO cell lines were generated using CRISPR-Cas9 genome editing.
Fig. S2.
Fig. S2.
CRISPR-Cas9 genetic screens of insulin-stimulated GLUT4 exocytosis. (A) Flow cytometry analysis of the starting mutant population and the final sorted population in the genome-wide primary screen (Top) and the secondary screen (Bottom). HeLa cells were treated with 100 nM insulin for 30 min before the cells were stained and analyzed by flow cytometry. (B) Abundance of sgRNAs in the genome-wide screen and the passage control. (C) Summary of genes validated in the secondary screen.
Fig. 2.
Fig. 2.
RABIF plays a critical role in insulin-stimulated GLUT4 exocytosis. (A) WT or Rabif-KO adipocytes were either untreated or treated with 100 nM insulin for 30 min before the localization of the GLUT4 reporter was visualized by confocal microscopy. (Scale bars: 10 µm.) (B) Normalized surface levels of the GLUT4 reporter in WT or mutant adipocytes. After serum starvation, the cells were treated with 200 µM dynasore for 5 min at 37 °C before 100 nM insulin was added. The cells were harvested for analysis at the indicated time points. Error bars indicate SD. (C) Normalized 2-deoxy-d-glucose uptake into WT or mutant adipocytes. ***P < 0.001. Error bars indicate SEM from four independent experiments. (D) Normalized surface levels of insulin receptor in WT or mutant adipocytes. n.s., not significant. (E) Immunoblots showing the expression levels of PPARγ and α-tubulin in WT or mutant adipocytes.
Fig. 3.
Fig. 3.
RABIF directly interacts with Rab10. (A) Diagram of the liposome coflotation assay. (B) Coomassie blue-stained denaturing gels showing the interaction between recombinant RABIF and liposome-anchored Rab10 proteins in the liposome coflotation assay. PF, protein free.
Fig. S3.
Fig. S3.
Rab10 and RABIF interact with each other and partially colocalize in the cell. (A) RABIF-FLAG was transiently expressed in RABIF-KO HeLa cells. RABIF-FLAG was immunoprecipitated using anti-FLAG antibodies, and the presence of Rab10 in the precipitates was detected by immunoblotting. (B) Representative confocal images showing the localization of Rab10-Myc and RABIF-FLAG in HeLa cells with or without 30 min of insulin treatment (100 nM). (Scale bars: 10 µm.) (Magnification: B, Right, 40×.)
Fig. 4.
Fig. 4.
RABIF does not function as a GEF in GLUT4 exocytosis. (A) Diagram showing the RABIF point mutations predicted to impair its interaction with Rab10. (B) Coomassie blue-stained gel showing purified WT and mutant RABIF proteins. (C) Kinetics of fluorescence changes resulting from RABIF-catalyzed mant-GDP release. The reactions were carried out in the presence of WT or mutant RABIF, using prenylated Rab10 as the substrate. (D) Initial rates of the reactions in C. Data are shown as percentage of fluorescence change within the first 3 min of the reactions. Error bars indicate SD. (E) Normalized surface levels of the GLUT4 reporter in the indicated adipocytes. n.s., not significant. ***P < 0.001. Error bars indicate SD. (F) Normalized surface levels of the GLUT4 reporter in the indicated HeLa cells. ***P < 0.001. Error bars indicate SD.
Fig. S4.
Fig. S4.
Overexpression of DENND4C does not rescue GLUT4 translocation defects in RABIF-KO cells. (A) Immunoblots showing the expression of DENND4C, Rab10, and α-tubulin in WT or RABIF-KO HeLa cells. (B) Effects of DENND4C overexpression on GLUT4 reporter translocation were measured by flow cytometry. Error bars indicate SD.
Fig. 5.
Fig. 5.
RABIF regulates the stability of Rab10. (A) Immunoblots showing the expression levels of endogenous Rab10 and α-tubulin in the indicated adipocytes. (B) Lentiviral expression of FLAG-tagged Rab10 in WT or Rabif-KO adipocytes. The levels of tagged Rab10 and endogenous α-tubulin were probed by immunoblotting. (C) Rabif-KO adipocytes were either untreated or treated with 10 µM MG132 or 100 nM PS341 for 24 h. The expression levels of endogenous Rab10 and α-tubulin were probed by immunoblotting. (D) WT and Rabif-KO adipocytes were treated with 10 µM MG132 or 100 nM PS341 for 24 h in the absence or presence of 100 ng/mL cycloheximide. The expression levels of endogenous Rab10 and α-tubulin were probed by immunoblotting. (E) WT or RABIF-KO HeLa cells were either untransfected or transiently transfected with cerulean-tagged Rab10 or Rab13. The expression levels of Rab10, Rab13, and α-tubulin were probed by immunoblotting. (F) Flow cytometry measurements of the surface levels of the GLUT4 reporter using cells from E. **P < 0.01. n.s., not significant. Error bars indicate SD.
Fig. S5.
Fig. S5.
RABIF regulates the stability of Rab10 in HeLa cells. (A) Immunoblots showing the expression levels of endogenous Rab10 and α-tubulin in the indicated HeLa cells. (B) RABIF-KO HeLa cells were either untreated or treated with the indicated proteasome inhibitors as described in Fig. 5C. The expression levels of endogenous Rab10 and α-tubulin were probed by immunoblotting. (C) WT or RABIF-KO HeLa cells were treated with the indicated inhibitors as described in Fig. 5D. The expression levels of endogenous Rab10 and α-tubulin were probed by immunoblotting. Ctrl, DMSO treatment.
Fig. S6.
Fig. S6.
The GEF mutations of RABIF reduce but do not abolish the RABIF–Rab10 interaction. Coomassie blue-stained gel shows the interaction of Rab10 with WT RABIF or RABIF mutants in a liposome coflotation assay.
Fig. S7.
Fig. S7.
RABIF stabilizes Rab10 protein in E. coli. The RABIF gene was cloned into the pET28a-based SUMO vector (1), while Rab10 was cloned into the pGEX-4T3 vector. GST-Rab10 was expressed in BL21 E. coli cells either alone or together with RABIF. After the addition of 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG), the cells were incubated at 37 °C for 3.5 h. After cell lysis and centrifugation, Rab10 proteins in the supernatant were precipitated using glutathione beads. Rab10 proteins from the supernatant and whole cells were probed by immunoblotting using anti-Rab10 antibodies.
Fig. 6.
Fig. 6.
RABIF regulates a subset of related Rab GTPases. (A) Proteomic analysis of Rab expression in RABIF-KO HeLa cells. Data are presented as percentage of expression levels in WT cells. Average values of two technical replicates are shown (Rab1b levels were identical in the replicates, while Rab5a was quantified in only one replicate). (B) Immunoblots showing the expression levels of selected Rabs in the WT or Rabif-KO adipocytes. (C) Dendrogram showing the relatedness of human Rab proteins to each other. Proteins influenced by RABIF are highlighted in red. Rabs that were not influenced by RABIF are labeled in green. Rabs not detected by MS are shown in black.
Fig. 7.
Fig. 7.
Model illustrating the Rab-stabilizing holdase chaperone function of RABIF.

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback