Fig 2: A Schematic of protein synthesis machinery. Representative 3D confocal images of adult cardiac myocytes showing B ribosomal protein Rpl22 (associated with actively translating ribosomes), C protein transport protein Sec61b, part of the translocon complex, D protein transport protein Sec23a, component of COPII, E trans-Golgi network protein TGN38, and F Cis-Golgi protein GM130. In all cases, proteins are shown in grayscale with nuclei in blue (3–9 cells per protein from 2 hearts). Quantitative analysis of these data and additional control experiments (along with n’s for specific experiments) are included in Supplementary Fig. 6. Antibodies used for these studies are listed in the methods

Fig 3: Decreased protein levels and mislocation of SEC23B result from E109K missense mutation.A, immunoblotting analysis was performed to detect protein levels of endogenous and exogenous SEC23B, its endogenous paralog SEC23A, and total SEC23 in Nthy-ori 3-1 stable cell lines with GFP-tagged WT SEC23B, SEC23BE109K mutant, or SEC23BR14W mutant. Asterisk denotes a nonspecific band. Experiments were repeated three times. B, immunofluorescence staining of Nthy-ori 3-1 stable cell lines was performed to detect the intracellular localization of SEC23B. Nthy-ori 3-1 cells were stained with rabbit anti-Sec16A (red) for ER exit sites and DAPI (blue) for nuclei. Exogenous SEC23B-GFP fusion protein was stained green. The scale bar represents 10 µm. Experiments were repeated three times. DAPI, 4',6-diamidino-2-phenylindole; ER, endoplasmic reticulum.

Fig 4: The absence of Sar1 destabilizes ERGIC membranes(A, B, D, E, and F) Cells lacking Sar1a (mock transfected or depleted of Sar1b for 72 h) were immunostained using antibodies directed against Sec16a (A), Tango1(B), ERGIC-53 (and Sec16a shown in insets, with arrows highlighting their juxtaposed distribution) (D), TFG (E), or COPB1 (and GM130 shown in insets) (F). Representative confocal images (maximum intensity projections) are shown. Scale bars, 5 µm; inset bars, 2 µm.(C) Quantification of the fold change in the fluorescence intensities of Sec31a, Sec16a, and Tango1 in the absence of Sar1 (relative to mock siRNA treatment). Error bars represent mean ± SEM (n = 15 cells each; at least three biological replicates each). **p < 0.01, calculated using an ANOVA followed by a Tukey post hoc test.(G) Representative immunoblots of extracts generated from control and CRISPR-modified cell lines lacking Sar1a and either in the presence or in the absence ofSar1b, using antibodies directed against COPB1 and actin.(H) Quantification of the fold change in COPB1 levels in the absence of Sar1 (relative to mock siRNA treatment of cells lacking only Sar1a). Error bar represents mean ± SEM (n = 4 biological replicates). See also Figure S8.

Fig 5: COPII condensates associate with secretory cargoes that leave the ER in the absence of Sar1(A) Cells expressing ss-DsRed and lacking both Sar1 isoforms were immunostained using antibodies directed against Sec24a and GM130 following treatment with SLF (50 µM) and imaged using STED microscopy. Representative images are shown (n = 15 cells; at least three biological replicates). Scale bar, 2 µm.(B) Cells expressing EGFP-Sec61ß and HaloTag-Sec23a in the absence of Sar1 were transfected with a construct encoding ss-DsRed and imaged live using spinning disk confocal microscopy following cargo release and labeling with JFX646-HaloTag ligand. Violin plots show the relative displacement of each marker over time (n = 10 cells; three biological replicates). **p < 0.01, calculated using an ANOVA followed by a Tukey post hoc test, compared with Sec61ß displacement.(C) Cells lacking both Sar1 isoforms and expressing native EGFP-Sec61ß and HaloTag-Sec23a were transfected with a construct encoding ss-DsRed and imaged live using spinning disk confocal microscopy following treatment with SLF (50 µM) and labeling with JFX646-HaloTag ligand. Representative images are shown (n = 7 cells; three biological replicates). Arrowheads indicate accumulation of ss-DsRed with COPII condensates, which ultimately move away from their site of origin. Scale bar, 2 µm.(D) Relative fluorescence intensities of cargo (ss-DsRed) and COPII (HaloTag-Sec23a) were measured at ER subdomains over time. Error bars represent mean ± SEM. An asterisk highlights the time point at which cargo and COPII undergo a >1 µm displacement, and all measurements are aligned with respect to this time point (n = 10 cells; three biological replicates).(E) Expression of 4×FM-HaloTag-L1CAM was transiently induced in control cells co-expressing Sar1 (H79G) or in cells lacking Sar1a and depleted of Sar1b, each labeled with JFX646-HaloTag ligand, and subjected to photobleaching after treatment with DDS for 60 min. Normalized fluorescence recovery in each case is shown (n = 10 cells each; at least three biological replicates each), and error bars represent mean ± SEM. An asterisk highlights the time point at which individual sites exhibiting elevated 4×FM-HaloTag-L1CAM fluorescence were bleached.(F) Cells lacking Sar1a and natively co-expressing EGFP-Sec61ß and HaloTag-Sec23a in the presence and absence of Sar1b were transfected with a construct encoding ss-DsRed and imaged as described for (B). The length of time ss-DsRed remained associated with HaloTag-Sec23A prior to undergoing displacement (more than 1 µm) was determined in each case. **p < 0.01, calculated using a t test, compared with cells lacking only Sar1a. See also Figure S7.