Molecular basis for SH3 domain regulation of F-BAR-mediated membrane deformation - PubMed (original) (raw)

Molecular basis for SH3 domain regulation of F-BAR-mediated membrane deformation

Yijian Rao et al. Proc Natl Acad Sci U S A. 2010.

Abstract

Members of the Bin/amphiphysin/Rvs (BAR) domain protein superfamily are involved in membrane remodeling in various cellular pathways ranging from endocytic vesicle and T-tubule formation to cell migration and neuromorphogenesis. Membrane curvature induction and stabilization are encoded within the BAR or Fer-CIP4 homology-BAR (F-BAR) domains, alpha-helical coiled coils that dimerize into membrane-binding modules. BAR/F-BAR domain proteins often contain an SH3 domain, which recruits binding partners such as the oligomeric membrane-fissioning GTPase dynamin. How precisely BAR/F-BAR domain-mediated membrane deformation is regulated at the cellular level is unknown. Here we present the crystal structures of full-length syndapin 1 and its F-BAR domain. Our data show that syndapin 1 F-BAR-mediated membrane deformation is subject to autoinhibition by its SH3 domain. Release from the clamped conformation is driven by association of syndapin 1 SH3 with the proline-rich domain of dynamin 1, thereby unlocking its potent membrane-bending activity. We hypothesize that this mechanism might be commonly used to regulate BAR/F-BAR domain-induced membrane deformation and to potentially couple this process to dynamin-mediated fission. Our data thus suggest a structure-based model for SH3-mediated regulation of BAR/F-BAR domain function.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.

Fig. 1.

Syndapin 1 F-BAR but not full-length syndapin 1 forms membrane tubules in living cells. (A) Schematic representation of the domain organization of syndapin 1 and of the various truncations and mutants used in this study. (B_–_G) Live-cell spinning-disk confocal imaging of Cos7 cells expressing full-length (FL) eGFP-syndapin 1 wild-type or mutants. (B) Expression of full-length eGFP-syndapin 1 did not cause membrane tubulation. (C_–_G) Membrane tubules induced by expression of eGFP-syndapin 1 F-BAR wild-type (C) or (I122F/M123F) (D). Mutational inactivation of the amphipathic wedge loop I122T/M123Q (E) or the basic belt (K127E/K130E) (F) and (K145E/K146E/K148E) (G) eliminated the tubulating activity. Representative images from at least three independent experiments are shown. (Scale bars, 10 μm.) (H and I) Representative electron micrographs of Cos7 cells expressing full-length eGFP-syndapin 1 (H) or eGFP-syndapin 1 F-BAR (I). Note the presence of numerous F-BAR-induced membrane tubules in I. (Scale bar, 0.5 μm.) (J) Quantification of the number of eGFP-positive tubules in Cos7 cells expressing syndapin 1 variants. Given are the mean number of tubules (±SE) per cell (n = 3 for each syndapin 1 variant) averaged for at least 20 cells.

Fig. 2.

Fig. 2.

Structure of full-length syndapin 1. (A) Secondary-structure representation of syndapin 1. Protein coloration: gold, chain A; purple, chain B; green, chain D (the SH3 domain). The SH3 domain may be associated with either chain A or chain B of the F-BAR domain, but the lacking electron density for the tether disallows a definite chain assignment. A second SH3 domain found in the asymmetric unit which interacts with the F-BAR domain in an irrelevant manner is not shown in the model. N and C termini, the secondary-structure assignment, and the wedge loop protruding from the long helical region are labeled only on chain A. (B) Close-up view of the interface of the SH3 and F-BAR domains. β-Strands of the SH3 domain are labeled. Residues involved in the interaction are shown as a stick model. Polar contacts are presented as red dashed lines. (C) A proline-rich peptide (blue) is modeled by hand to the putative PxxP-binding groove in the syndapin 1 SH3 domain, corresponding to the peptide position in the SH3-peptide complex structures PDB ID codes 1W70 and 2DRK. (D) The protein surface is colored according to the electrostatic potential: red (negative) through white (neutral) to blue (positive). The lower panel shows an open-up view of the surface charge distribution between F-BAR and SH3. Residues mutated in functional studies are labeled.

Fig. 3.

Fig. 3.

Complex formation between syndapin 1 F-BAR and SH3 regulates syndapin 1-induced membrane tubulation. (A) GST-syndapin 1 SH3 (379–441) or GST-syndapin 1 ΔF-BAR (340–441) bind to their F-BAR domain in vitro, whereas GST does not. (B) Mutation of a basic patch on the F-BAR domain (K145E/K146E/K148E) eliminates binding to SH3. (C) Mutations within the SH3 domain (D394R/E400R) and (Q396R/E397R) selectively impair binding to the F-BAR domain. By contrast, mutation of the proline-rich motif-binding site on SH3 (P434L) does not affect F-BAR association. Binding assays were done as described in Materials and Methods. Samples were analyzed by SDS/PAGE and staining with Coomassie blue. Input, 15% (A and B) or 10% (C) of the total amount of purified F-BAR used for the binding assay. (D) Cotransfection of wild-type (WT) but not mutant SH3 inhibits eGFP-syndapin 1 F-BAR-induced membrane tubulation. Shown are the mean number of tubules (±SE) per cell (n = 3 for each syndapin 1 variant) averaged for at least 20 cells. (E_–_G) In vitro binding assays. GST or GST-SH3 fusion proteins were immobilized on glutathione beads and incubated with purified His6-tagged endophilin 1 BAR (E), amphiphysin 1 BAR (F), or syndapin 1 F-BAR (G) domains. Input, 5% (E and F) or 3.5% (G) of the total amount of purified BAR domains used for the assay. Samples were analyzed by SDS/PAGE and staining with Coomassie blue.

Fig. 4.

Fig. 4.

Association of syndapin 1 SH3 with dynamin 1 releases its tubulating activity from autoinhibition. (A) The proline-rich domain of dynamin 1 (Dyn I-PRD) and syndapin 1 F-BAR show partially overlapping binding sites on syndapin 1 SH3. GST pull-down assays were done as described in Materials and Methods. (B) A dynamin 1-derived syndapin 1-binding wild-type (PxxP) but not an inactive mutant (AxxA) peptide (each at 100 μM) competes with F-BAR for binding to GST-SH3. A peptide derived from the syndapin 1 linker region (NPF) was taken as a further negative control. Samples were analyzed by SDS/PAGE and staining with Coomassie blue. Input, 15% (A and B) of the total amount of purified F-BAR domain used in the assay. (C) Expression of a GTPase-defective dynamin 1 mutant (K44A) unlocks a potent membrane-tubulating activity of eGFP-syndapin 1. Quantification of eGFP-syndapin 1-induced membrane tubules in Cos7 cells coexpressing either mRFP or mRFP-tagged dynamin 1 variants (wild-type or GTPase-defective mutant K44A). Shown are the mean number of tubules (±SE) per cell (n = 3 for each syndapin 1 variant) averaged for at least 20 cells.

Fig. 5.

Fig. 5.

Syndapin 1-induced membrane tubulation reconstituted on liposomes in vitro. (A_–_D) Negative-stain electron microscopy. (A) Transmission electron microscope image of liposomes incubated with full-length (FL) syndapin 1. Inset shows liposomes without the protein. (B_–_D) Representative micrographs of liposomes incubated with syndapin 1 F-BAR (B), FL syndapin 1 with 170 μM dynamin peptide (C), and FL syndapin 1 (Q396R, E397R) (D). (Scale bar, 500 nm.) (E) Quantification of liposome tubulation in the experiments illustrated above. Note the significant increase in tubulation efficiency (P < 0.001, Student's t test) of FL syndapin 1 after addition of the dynamin peptide or upon mutation of its SH3 domain. (F) Hypothetical model for the cooperative role of BAR-SH3 domain-containing proteins and dynamin 1 in membrane deformation and fission. See text for details.

Similar articles

Cited by

References

    1. Hatzakis NS, et al. How curved membranes recruit amphipathic helices and protein anchoring motifs. Nat Chem Biol. 2009;5:835–841. - PubMed
    1. Hui E, Johnson CP, Yao J, Dunning FM, Chapman ER. Synaptotagmin-mediated bending of the target membrane is a critical step in Ca(2+)-regulated fusion. Cell. 2009;138:709–721. - PMC - PubMed
    1. Frost A, Unger VM, De Camilli P. The BAR domain superfamily: Membrane-molding macromolecules. Cell. 2009;137:191–196. - PMC - PubMed
    1. Frost A, et al. Structural basis of membrane invagination by F-BAR domains. Cell. 2008;132:807–817. - PMC - PubMed
    1. Peter BJ, et al. BAR domains as sensors of membrane curvature: The amphiphysin BAR structure. Science. 2004;303:495–499. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources