|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
| ||||||||||||||||||||
Files in this Data Supplement:
Fig. S1. The BAR domain of SNX9 is required for its localization to the plasma membrane. (A) COS-7 cells were transfected with GFP-tagged SNX9 or various deletion mutants and tubule formation and cellular localization were observed. Note that BAR domain deletion mutant (GFP-SNX9-ΔBAR) exhibited an aggregated pattern without showing any plasma membrane localization while other mutants mutants are expressed in cytosol as well as the plasma membrane. Scale bars: 30 µm; insets, 10 µm. (B-D) The BAR deletion mutant is largely insoluble and formed aggregates which colocalized neither with endosomal markers EEA1(B), GFP-2×FYVEHrs(C), nor lysosomal marker Lysotracker (D). Scale bars: 30 µm; insets, 5 µm.
Fig. S2. Antagonistic effects of N-WASP and dynamin-2 on SNX9-induced tubulation. (A, B) When high levels of GFP-tagged N-WASP or mRFP-tagged dynamin-2 were coexpressed with GFP- or mRFP-tagged SNX9 in COS-7 cells, tubular invaginations were significantly reduced or even absent in many cases. (C, D) Coexpression with N-WASP-ΔVCA, a mutant that cannot interact with the Arp2/3 complex, failed to affect tubulation by SNX9. Likewise, coexpression of SNX9 with dynamin-2 K44A, a GTPase-deficient mutant, abolished the antagonistic effect of dynamin-2 on tubulation. Scale bars: 30 µm; insets, 5 µm. (E-H) Representative images from various combinations of SNX9 or SNX9-ΔSH3 with dynamin-2 or N-WASP. Scale bars: 30 µm; insets, 5 µm.
Fig. S3. An antagonistic effect of dynamin-2 on SNX9-induced tubulation was abolished by the treatment of latrunculin B. Latrunculin B was added to cells cotransfected with RFP-SNX9 and GFP-dynamin-2. Note dynamin-2 and SNX9 staining on latrunculin-B-induced tubules. Scale bars: 30 µm.
Fig. S4. The tubulating activity of SNX9 is regulated by PtdIns(4,5)P2 metabolizing enzymes. (A) Low magnification images from cells that were cotransfected with GFP-tagged SNX9 and HA-tagged PtdIns(4)P-5-kinase Iα, PtdIns(4)P-5-kinase Iβ, or empty vector. When HA-tagged PtdIns(4)P-5-kinase Iα or HA-tagged PtdIns(4)P-5-kinase Iβ was coexpressed with mildly expressed GFP-tagged SNX9 in COS-7 cells, tubular invaginations were significantly increased around the cell periphery. (B) The kinase-dead mutant of PtdIns(4)P-5-kinase Iα (D309N/R427Q) failed to increase tubulation by mildly expressed GFP-tagged SNX9 in COS-7 cells. (C) Coexpression of mRFP-tagged SNX9 with GFP-tagged 5-phosphatase domain of synaptojanin inhibits tubule formation even with high levels of SNX9 expression. Scale bars: 30 µm; insets, 5 µm.
Fig. S5. SNX9 is required for transferrin uptake. (A) Texas-Red transferrin uptake was measured in cells overexpressing GFP-SNX9 and in control cells. Transferrin uptake was completely inhibited in SNX9-expressed cells. Scale bar: 30 µm. (B) Knockdown efficiency of endogenous SNX9 by SNX9 siRNA was confirmed by immunoblotting and quantified. (C) HeLa cells were transfected with SNX9 siRNA and uptake of Texas-Red−transferrin was performed 48 hours after transfection. Transferrin uptake was largely impaired in cells expressing SNX9 siRNA. Scale bar: 50 µm. (D) Quantitative transferrin internalization assay of cells expressing SNX9 siRNA. Error bars indicate s.e.m.
Fig. S6. The deletion of a distinct domain of SNX9 prevents its binding to certain protein(s) but retains its ability to bind others. SNX9 binds dynamin and N-WASP via its SH3 domain, clathrin/AP-2 and Arp2/3 via its LC domain, and PtdIns(4)P-5-kinases via its PX domain. (A) The deletion of SH3 domain (SNX9-ΔSH3-GFP) prevented interaction of SNX9 with dynamin and N-WASP but retained its ability to bind PtdIns(4)P-5-kinase Iα. (B) The PX domain deletion mutant (SNX9-ΔPX-GFP) failed to bind PtdIns(4)P-5-kinase Iα but its binding to N-WASP and dynamin was not affected.
Fig. S7. Summary of the regulatory mechanisms of membrane tubulation by SNX9. During early steps of clathrin-mediated endocytosis, SNX9 is recruited to the plasma membrane through its PX domain and BAR domain. The BAR domain of SNX9 deforms the membrane and induces membrane tubular invagination whereas the SH3 domain of SNX9 recruits dynamin and N-WASP to the tubulating sites of endocytic pits. Dynamin severs the tubules by GTP hydrolysis with the supporting force of actin cytoskeleton regulated by N-WASP. By these mechanisms, the increase in actin dynamics and dynamin activity would inhibit the tubulation and vice versa. Calcium influx exerts various effects including activation of gelsolin, which severs actin filaments. EGF increases recruitment of SNX9 to the plasma membrane.
| ||||||||||||||||||||