mTrs130 is a component of a mammalian TRAPPII complex, a Rab1 GEF that binds to COPI-coated vesicles

Abstract

The GTPase Rab1 regulates endoplasmic reticulum-Golgi and early Golgi traffic. The guanine nucleotide exchange factor (GEF) or factors that activate Rab1 at these stages of the secretory pathway are currently unknown. Trs130p is a subunit of the yeast TRAPPII (transport protein particle II) complex, a multisubunit tethering complex that is a GEF for the Rab1 homologue Ypt1p. Here, we show that mammalian Trs130 (mTrs130) is a component of an analogous TRAPP complex in mammalian cells, and we describe for the first time the role that this complex plays in membrane traffic. mTRAPPII is enriched on COPI (Coat Protein I)-coated vesicles and buds, but not Golgi cisternae, and it specifically activates Rab1. In addition, we find that mTRAPPII binds to gamma1COP, a COPI coat adaptor subunit. The depletion of mTrs130 by short hairpin RNA leads to an increase of vesicles in the vicinity of the Golgi and the accumulation of cargo in an early Golgi compartment. We propose that mTRAPPII is a Rab1 GEF that tethers COPI-coated vesicles to early Golgi membranes.

Figure 1.

A mammalian TRAPPII complex is detected in NIH3T3 cells. (A) Cytosol from NIH3T3 cells was fractionated on a Superdex-200 gel filtration column. The column fractions were concentrated as described in the Materials and Methods and mTrs130, mTrs120, mTrs23, mBet3, and mTrs20 were detected by Western blot analysis. (B) Fractions 9 and 10 were pooled and immunoprecipitated with anti-mBet3 (left, lane 2) or anti-mTrs120 IgG (right, lane 4) to isolate the mammalian TRAPP complex. Coprecipitated TRAPP subunits were detected by Western blot analysis. Immunoprecipitation using nonspecific rabbit IgG was performed as a negative control (lanes 1 and 3). (C) mTrs120 interacts with mTrs130. A 10-cm plate of COS-7 cells, grown to 40% confluence, was transfected with 3 μg of plasmid DNA. The next day, the cells were washed with PBS and resuspended in 1 ml of lysis buffer (20 mM Tris, pH 7.5, 100 mM NaCl, 0.1% Triton X-100, 0.1% NP-40, and protease inhibitor cocktail). The lysed cells were spun at 16,000 × g for 10 min at 4°C in a microfuge tube. Lysates of COS-7 cells overexpressing GFP (SFNB 1520) and GFP-mTrs130 (SFNB 1173, left top two panels, lanes 1 and 2), or Myc-mTrs120 (SFNB 1517, left bottom panel, lanes 1 and 2) were processed for Western blot analysis using anti-GFP (top two panels) or anti-Myc (bottom) antibodies. The same lysates were also precipitated with anti-Myc antibody and blotted with either anti-GFP (right, top two panels) or anti-myc 9E10 (right, bottom) antibodies. Myc-mTrs120 only interacted with GFP-mTrs130 (right top, lane 2) and not with GFP (right middle, lane 1).

Figure 2.

mTrs130 and mTrs120 are components of a TRAPPII complex that is a GEF for Rab1, but not Rab11. mTRAPPII was immunoprecipitated from NIH3T3 cytosol with anti-mTrs120 IgG, and its ability to stimulate guanine nucleotide exchange activity on Rab1a (A) or Rab11a (GeneID 8766) (B) was tested. Equal amounts of Superdex-200 column fractions 9 and 10 (Figure 1A), containing mTRAPPII, were immunoprecipitated with anti-mBet3 IgG, nonspecific rabbit IgG, or anti-mTrs120 IgG (also see Supplemental Figure S2). An equal volume of the protein A-Sepharose beads were then assayed 12 min for Rab1a (C) or Rab 11a (D) guanine nucleotide exchange activity. Similar results were obtained when the complex was precipitated with anti-mTrs130 antibody (data not shown). Error bars are SEM.

Figure 3.

mTrs130 largely colocalizes with early Golgi markers. Colocalization experiments were performed in COS-7 cells with antibodies directed to mTrs130 (red) and YFP-Rab1a (green) (A), mTrs130 (green) and GM130 (red) (B), mTrs130 (green) and COPI (red) (C), and mTrs130 (green) and the trans-Golgi marker GalT (red) (D). The images were captured on a scanning laser confocal microscope using a 63× objective lens. The merged images are shown on the right, and the insets are magnified in the next row. Bar, 20 μm.

Figure 4.

mTrs130 colocalizes with GM130 and COPI in nocodazole treated cells. COS-7 cells were treated with 10 μg/ml nocodazole for 1 h before the samples were fixed and processed for staining with antibodies directed against mTrs130 (red) and YFP-Rab1a (green) (A), mTrs130 (green) and GM130 (red) (B), mTrs130 (green) and COPI (red) (C), and mTrs130 (green) and GalT (red) (D). The merged images are shown on the right, and the insets are magnified in the next row. Bar, 20 μm.

Figure 5.

The mTRAPPII complex is bound to COPI-coated vesicles. mTrs23 (green) colocalizes with GM130 (red) (A) and COPI (red) in COS-7 cells (B). (C) Immuno-EM was performed in NRK cells by using anti-mTrs23 and anti-β′COP antibodies as described in Materials and Methods. The arrow points to a COPI coated bud that contains mTrs23 (15-nm gold particles) and β′COP (10-nm gold particles). G denotes Golgi cisternae. Bar, 200 nm. (D and E) Multiple grids dual-labeled for mTrs23 and β'COP were quantitated as described in Materials and Methods. The data for mTrs23 is shown in D and β′COP in E.

Figure 6.

mTRAPPII interacts with γ1COP. mTrs130 (A) and mTrs120 (B) coprecipitate with HA-γ1COP. The protocol used to immunoprecipitate HA-γ1COP was the same as described in the legend to Figure 1C only the cells were lysed in 50 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 1% NP-40 (plus protease inhibitor cocktail). HA-γ1COP was precipitated by incubating 1.8–2.5 mg of lysate with anti-HA–conjugated beads (10 μl). The beads were washed three times with 50 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, 0.5% NP-40 and processed for Western blot analysis. The precipitate (40% of the sample) and lysate (1% of the sample) were blotted with anti-GFP antibody (top left), anti-myc antibody (top right) or anti-HA antibody (bottom). GFP-mTrs130 interacted with γ1COP but not εCOP (Figure 6A, lanes 2 and 3, compare to lysate in lanes 5 and 6). Myc-mTrs120 interacted with γ1COP but not εCOP (Figure 6B, lanes 2 and 3, compare to lysate in lanes 5 and 6). The interaction with γ1COP was specific, because mBet3 did not interact with γ1COP (Figure 6, A and B, lane 1, compared with lysate in lane 4). (C) Binding assays were done as described in Cai et al. (2007b), with the following modifications. Sixty microliters of beads (2 μg of fusion protein) were incubated overnight at 4°C with 0.8 mg of lysate in a reaction volume of ∼200 μl. Half of the binding reaction was loaded onto the gel.

Figure 7.

The depletion of mTrs130 disrupts the early Golgi. (A) COS-7 cells were transfected with the pSilencer 1.0-U6 vector (mock) or vector1 (shRNA). The cells were harvested 48 h after transfection and immunoblotted with anti-Trs130 (top) and anti-ER60 antibodies (bottom). The ER protein ER60 was used as a loading control. (B) COS-7 cells were transfected with pSilencer 1.0-U6 vector (mock) or vector1 (shRNA) and immunostained with anti-mTrs130 (green) and anti-GM130 antibodies (red). (C) COS-7 cells were transfected with the pSilencer 1.0-U6 vector (mock) or vector1 (shRNA) together with vectors that express either GFP or GFP-mTrs130* (see Materials and Methods). The cells were immunostained with anti-GM130 antibody and the number of fragmented Golgi was quantitated in greater than 30 cells. SEM was calculated from three independent experiments. Bar, 20 μm.

Figure 8.

The depletion of mTrs130 by shRNA leads to the accumulation of vesicles in the vicinity of the Golgi. (A) After depleting cells of mTrs130 by shRNA expression vector1, the cells were fixed and processed for EM as described in the Materials and Methods. A representative EM micrograph of mTrs130-depleted cells is shown at the top of the figure, and the mock-treated cells are shown at the bottom. Empty arrows point to Golgi cisternae and filled arrows to vesicles in the vicinity of the Golgi. (B) Quantitation of the number of vesicles per square μm of Golgi area in the shRNA and mock-treated cells. (C) Quantitation of Golgi cisternal membranes in shRNA and mock-treated cells. p < 0.0001. Bar, 1 μm.

Figure 9.

The depletion of mTrs130 leads to the accumulation of VSV-G in an early Golgi compartment that is marked by GM130. (A) To express tsO45VSV-G-GFP, pEGFPN1-VSV-G was cotransfected with the pSilencer 1.0-U6 (mock) vector or vector1 (shRNA). The cells were incubated at 40°C for 20 h, cycloheximide was added, and 30 min later the cells were shifted to 32°C for 30, 60, and 90 min. To detect cell surface VSV-G-GFP, the cells were fixed without permeabilization and immunostained with anti-VSV-G antibody. The data shown was quantitated from four independent experiments as described in the Materials and Methods. Greater than 30 cells were quantified in each experiment. The 60-min time point was set to one. Error bars are the SEM. (B) The protocol was the same as in A, only the 90-min samples were fixed and permeabilized with methanol. The cells were immunostained with anti-GM130 antibody (red). VSV-G was detected with a GFP tag (green). Greater than 58 cells were quantified in three independent experiments.