The cargo receptors Surf4, endoplasmic reticulum-Golgi intermediate compartment (ERGIC)-53, and p25 are required to maintain the architecture of ERGIC and Golgi

Abstract

Rapidly cycling proteins of the early secretory pathway can operate as cargo receptors. Known cargo receptors are abundant proteins, but it remains mysterious why their inactivation leads to rather limited secretion phenotypes. Studies of Surf4, the human orthologue of the yeast cargo receptor Erv29p, now reveal a novel function of cargo receptors. Surf4 was found to interact with endoplasmic reticulum-Golgi intermediate compartment (ERGIC)-53 and p24 proteins. Silencing Surf4 together with ERGIC-53 or silencing the p24 family member p25 induced an identical phenotype characterized by a reduced number of ERGIC clusters and fragmentation of the Golgi apparatus without effect on anterograde transport. Live imaging showed decreased stability of ERGIC clusters after knockdown of p25. Silencing of Surf4/ERGIC-53 or p25 resulted in partial redistribution of coat protein (COP) I but not Golgi matrix proteins to the cytosol and partial resistance of the cis-Golgi to brefeldin A. These findings imply that cargo receptors are essential for maintaining the architecture of ERGIC and Golgi by controlling COP I recruitment.

Figure 1.

Surf4 localizes mainly to the ERGIC. (A) Localization of endogenous Surf4 in HeLa cells by confocal immunofluorescence microscopy using affinity-purified antibodies against Surf4 in combination with antibodies against ERGIC-53, giantin, and BAP31. Right, HeLa cells were treated with 10 μg/ml BFA for 90 min (+BFA) and labeled with antibodies against Surf4 and ERGIC-53. (B) HeLa cells were transfected with HA-Surf4 or HA-Surf4SSS. The tagged versions of Surf4 were stained with anti-HA and costained with anti-ERGIC-53 and anti-giantin antibodies. Arrowheads indicate colocalization of HA-Surf4 with ERGIC-53. Bars, 10 μm.

Figure 2.

Surf4 forms protein complexes with p24 family members and ERGIC-53. (A) ERGIC membranes were isolated from parent HepG2 cells (lane 1) and HepG2 cells stably expressing HA-Surf4 (lane 2) (see Supplemental Figure 1). Isolated membranes were separated by Blue Native-PAGE followed by Western blotting with antibodies against Surf4 and the HA epitope. (B) ERGIC membranes of HepG2 cells stably expressing HA-Surf4 were separated by Blue Native-PAGE as described in A, followed by SDS-PAGE in a second dimension. Proteins were stained with Coomassie Blue and excised for mass spectrometry analysis. Black circles indicate the protein complex containing HA-Surf4 and the p24 family members p23, p24, and p25. Open circles indicate complexes of identified proteins that were not further analyzed. Asterisks, proteins not identified by mass spectrometry. (C) Coimmunoprecipitation experiments: ERGIC membranes of parent HepG2 cells (−) and HepG2 cells stably expressing HA-Surf4 (+) were isolated, lysed, and subjected to immunoprecipitation with anti-HA, anti-ERGIC-53, anti-p23, and anti-p24 followed by Western blotting by using antibodies against the HA-epitope, p23, p24, and p25. (D) HeLa cell lysates were immunoprecipitated with anti-ERGIC-53 antibodies coupled to beads or with beads alone. The total lysate (1/20) was loaded as indicator for protein amount in the cell. Proteins were visualized by Western blotting with antibodies to ERGIC-53 or p23.

Figure 3.

Double knockdown of Surf4/ERGIC-53 or single knockdown of p25 leads to Golgi dispersal. (A) HeLa cells were transfected with control siRNA, Surf4/ERGIC-53 siRNA or p25 siRNA. The Golgi was visualized by immunofluorescence microscopy using anti-giantin. (B) Quantitative analysis of the Golgi phenotype in Surf4/ERGIC-53 and p25 knockdowns. More than 100 cells of three independent experiments were counted for each condition, and the percentage of cells with fragmented Golgi plotted. Results are means ± SD. Bar, 10 μm. (C) Cells treated with control, Surf4/ERGIC-53, or p25 siRNA were processed for electron microscopy and sections of 10,000× and 20,000× magnifications are shown. Arrows, Golgi ribbon. Asterisks, dispersed Golgi stacks. Bars, 0.5 μm.

Figure 4.

ERGIC structures are reduced in cells depleted of Surf4/ERGIC-53 or p25. (A) HeLa cells transfected with control, Surf4/ERGIC-53, or p25 siRNA were immunostained with antibodies against KDEL-receptor and giantin, and then they were analyzed by confocal microscopy. The giantin staining was used as indication for efficient knockdown of Surf4/ERGIC-53 and p25. The cell borders are outlined in white. Bars, 10 μm. (B) Quantitative analysis of the ERGIC structures. More than 18 cells per condition of three independent experiments were analyzed. KDEL-receptor–positive ERGIC structures were counted after removal of the Golgi area defined by giantin staining (see Materials and Methods). Results are means ± SD.

Figure 5.

ER exit site formation and anterograde transport are not affected in Surf4/ERGIC-53 or p25 knockdown cells. (A) HeLa cells transfected with control, Surf4/ERGIC-53 or p25 siRNAs were processed for immunofluorescence microscopy using antibodies against Sec31 and giantin. The giantin staining was used as indication for efficient knockdown of Surf4/ERGIC-53 and p25. The cell borders are outlined in white. Bars, 10 μm. (B) Quantitative analysis of ER exit sites. More than 25 cells per condition of three independent experiments were analyzed. ER exit sites were counted according to the Sec31 staining (see Materials and Methods). Results are means ± SD. Bars, 10 μm. (C) HeLa cells were transfected with control, Surf4/ERGIC-53 and p25 siRNA and subjected to pulse-chase analysis using [³⁵S]methionine. Media from cells were collected and assayed for incorporated radioactivity. Results are means ± SD of at least three independent experiments.

Figure 6.

Live imaging of GFP-ERGIC-53 reveals a shorter life span of ERGIC structures in p25 knockdown cells. (A) Time series from Supplemental Movies 1 and 2. Cells were transfected with control or p25 siRNA and imaged with an interval of ∼2 s. Representative frames from a control cell show stationary ERGIC structures that hardly move throughout the imaging period (top, arrowheads). In p25 knockdown cells the stationary ERGIC structures do not move either, but they disappear with time (bottom, arrowheads). (B) Life span of ERGIC structures in p25-depleted cells. Quantification of the relative life span of GFP-ERGIC-53 structures presented in Figure 6. The average life span is plotted in percentage. Note that in p25 knockdown cells the life span of ERGIC structures is reduced by ∼35%. This difference is statistically significant (Student's t test, p ≤ 0.05). Results are means ± SD (n = 8).

Figure 7.

Golgi matrix proteins remain associated with the dispersed Golgi. HeLa cells in which Surf4/ERGIC-53 or p25 was silenced by siRNA were processed for immunofluorescence microscopy by using anti-GM130, anti-GRASP65, and anti-p115 and colabeled with anti-giantin antibodies. The giantin staining was used as indication for efficient knockdown of Surf4/ERGIC-53 and p25. Bars, 10 μm.

Figure 8.

β-COP is dispersed in Surf4/ERGIC-53 and p25 knockdown cells. (A) HeLa cell treated with Surf4/ERGIC-53 or p25 siRNA were immunostained for β-COP and giantin. Shown are representative images of three independent experiments. (B) Quantification of β-COP and giantin intensities in the Golgi region. The Golgi region was defined by giantin staining. Shown are intensity ratios of β-COP and giantin normalized to 100%. +BFA indicates control siRNA-transfected HeLa cells treated with 10 μg/ml BFA for 5 min. Results are means ± SD (n = 3). Bars, 10 μm.

Figure 9.

Cis-Golgi remains partially resistant to BFA in Surf4/ERGIC-53– and p25-depleted cells. HeLa cells transfected with Surf4/ERGIC-53, p25 siRNA, or control siRNA were treated with 10 μg/ml BFA for 30 min. Cells were processed for immunofluorescence microscopy by using anti-GM130 and anti-KDEL-receptor (A), anti-GM130 and anti-p115 (C), or anti-giantin and anti-GPP130 (D) antibodies. (B) Quantification of cells showing BFA resistant cis-Golgi according to GM130 staining. Results are means ± SD (n = 3). Bars, 10 μm.

Figure 10.

Cargo receptors are required for tubulation of the cis-Golgi. HeLa cells treated with control, Surf4/ERGIC-53, or p25 siRNA were incubated in hypotonic medium for 5 min. Cells were processed for immunofluorescence microscopy by using anti-GM130 and anti-giantin antibodies. Note that the knockdown conditions prevented the tubulation of the cis-Golgi indicated by GM130. Bars, 10 μm.

Figure 11.

Model depicting the effect of silencing Surf4/ERGIC-53 or p25 on the early secretory pathway. In the presence of cargo receptors (+cargo receptors), the architecture of the organelles is guaranteed by balanced anterograde and retrograde trafficking indicated by arrows. Depletion of cargo receptors such as Surf4/ERGIC-53 or p25 (−cargo receptors) dissociate COP I coats from cis-Golgi and ERGIC membranes, impairing retrograde transport from cis-Golgi to ERGIC and ERGIC to ER. The sum of this reaction results in dispersal of the Golgi apparatus and reduction of ERGIC structures.