Fig 2: PRV infection modifies LC3 lipidation after infection in permissive cells.(a) Vero, 3T3 and PK cells were mock infected or infected with PRV at MOI of 0.1. At 2, 6, 12, 24 or 36 h post-infection (hpi), the cells were lysed, separated by reducing SDS-PAGE, and subjected to a Western blot using antibodies against LC3, ACTB (ß -actin), SQSTM1 or PRV gE protein as indicated. Densitometry was performed for quantification. The ratio of LC3-II to ACTB from three independent experiments is expressed as the mean ± SD. One-way ANOVA test; *P < 0.05; **P < 0.01, as compared with the mock infection group. Vero cells (b) and PK cells (c) were mock infected or infected with PRV at MOI of 10. At 2, 4, 6, 8 or 10 hpi, the cells were lysed, separated by reducing SDS-PAGE, and subjected to a Western blot using antibodies against LC3, ACTB, SQSTM1, AKT, p-AKT (p-Ser473), PRV gE protein or US3 protein as indicated. Densitometry was performed for quantification. The ratio of LC3-II to ACTB from three independent experiments is expressed as the mean ± SD. One-way ANOVA test; *P < 0.05, as compared with the mock infection group. Gels were run under the same experimental conditions. For better clarity and concise presentation, cropped blots are shown. The raw uncropped images can be found in Supplementary Fig. S1.

Fig 3: Imaging autophagy by conventional EM and immunofluorescence. (A) Conventional EM of control U2OS cells. Endo-lysosomal compartments are identified by morphology. (B) Conventional EM of U2OS cells starved for 2.5 h in the presence of BafA1 showing an accumulation of different types of autophagic compartments. Autophagosomes (AP) are recognized by their double-membrane that encapsulates part of the cytoplasm with the same density as the surrounding cytoplasm and containing multiple membrane structures. Autolysosomes (AL) are vacuoles lined by a single membrane. Their lumen is heterogenous in content (membranes, vesicles, amorphous material) and electron density (from light to dense). Both AP and AL contain recognizable endoplasmic reticulum (ER) cisternae (arrows). Asterisks indicate autophagic content. (C-E) IF of semi-thin cryosections of control and starved, BafA1-treated U2OS cells, fixed either with 4% PFA ON (PFA) or 2% PFA+0.2% GA for 2 h (PFA+GA) and labeled with Cosmo Bio anti-LC3B (1:10) and Alexa Fluor 488-conjugated goat anti-mouse IgG (1:300). Images are recorded with identical settings and represented with equal intensity ranges. PFA and PFA+GA fixed cells contain comparably intense LC3B puncta. AL, autolysosome; AP, autophagosome; E, endosome; Ly, lysosome; M, mitochondrion; PM, plasma membrane. Scale bars: 500 nm (A-B); 20 µm (C-E).

Fig 4: Palmitate induces ER stress and autophagy at early time points, but prolonged palmitate treatment impairs autophagy in a hypothalamic cell line. a and b Immunoblotting analysis (a) and quantification (b) of ER stress markers (p-PERK, PERK, p-IRE1, IRE1, ATF6, ATF4, and CHOP) in cells treated with different concentrations of palmitate (25, 50, and 100 µM) for the indicated times (n = 3 per each group). Data are mean ± SEM; p-PERK/PERK; *p = 0.0252, ***p = 0.0003 vs. control at 12 h, **p = 0.0065 vs. control at 24 h, ATF6; **p = 0.0040 vs. control at 12 h, **p = 0.0074 vs. control at 24 h, ATF4; *p = 0.0454, **p = 0.0017 vs. control at 12 h, *p = 0.00139 vs. control at 24 h, CHOP; *p = 0.0240, ****p < 0.0001 vs. control at 12 h, ***p = 0.0002 vs. control at 24 h. c and d Immunoblotting analysis (c) and quantification (d) of ER stress markers in cells treated with 0.1 mM palmitate for the indicated times (n = 3 per each group). Data are mean ± SEM; p-PERK/PERK; **p = 0.0050, ***p = 0.0007, ****p < 0.0001 vs. control, ATF6; **p = 0.0091 vs. control, ATF4; **p = 0.0033 at PA 6 h vs. control, **p = 0.0067 at PA 12 h vs. control, ***p = 0.0009 vs. control, CHOP; ****p < 0.0001 vs. control. e and f Immunoblotting analysis (e) and quantification (f) of autophagy markers (p62 and LC3-II) in cells treated with 0.1 mM palmitate for the indicated times. Baf A1 (200 nM) was added for 3 h before harvest (p62; n = 7 for control, control + Baf A1, PA 3 h, PA 3 h + Baf A1, n = 3 for PA 6 h, PA 6 h + Baf A1, PA 12 h, PA 12 h + Baf A1, LC3-II; n = 8 for control, n = 7 for control + Baf A1, PA 3 h, PA 3 h + Baf A1, n = 4 for PA 6 h, PA 6 h + Baf A1, n = 3 for PA 12 h, PA 12 h + Baf A1. Data are mean ± SEM; p62; *p = 0.0295 at PA 6 h vs. control, *p = 0.0250 at 12 h vs. control, ****p < 0.0001 vs. control, ###p = 0.0008, LC3-II; *p = 0.0444, ***p = 0.0003, ****p < 0.0001 vs. control, ####p < 0.0001. g and h Cells transiently expressing mRFP-GFP-LC3 were treated with 0.1 mM palmitate for the indicated times. Representative micrographs (g) and quantification (h) of the average number of mRFP-GFP-LC3 puncta per cell (n = 18 per each group). Scale bar, 10 µm. Data are mean ± SEM; Red puncta; *p = 0.0158, ****p < 0.0001 vs. control, ####p < 0.0001, Yellow puncta; *p = 0.0257, ****p < 0.0001 vs. control, †p = 0.0174. n.s., no significant difference

Fig 5: UBE2QL1 abrogates recruitment of VCP/p97, reduces accumulation of p62, and compromises association of LC3 with damaged lysosomes HeLa cells stably expressing p97-GFP were control or UBE2QL1-depleted for 48 h, and expression of p97-GFP was induced for the last 24 h. Cells were LLOMe or control treated for 1 h, fixed after a chase of 2 h, and stained with antibodies to K48 ubiquitin chains and LAMP1 as indicated. Note the reduction of the K48 and p97-GFP signals on damaged lysosomes in UBE2QL1-depleted cells. Arrows indicate colocalizing or non-colocalizing vesicles. Scale bars: 20 µm, 10 µm for inlays. See Fig EV4A for images of cells treated with an alternative siRNA.Automated quantification of (A) and Fig EV4A. Shown are fold increases of Pearson correlation coefficients (P.C.C.) of p97-GFP and LAMP1 signals, normalized to untreated siCtrl. Data are from four independent experiments with = 30 cells per condition.HeLa cells were control or UBE2QL1-depleted for 60 h, fixed after 1 h of LLOMe or control treatment and stained for LAMP1 and p62, and analyzed by confocal microscopy. Arrows indicate colocalizing or non-colocalizing vesicles. Scale bars: 20 µm, 5 µm for inlays. See Fig EV4B for images of cells treated with an alternative siRNA and control cells.Quantification of (C) and Fig EV4B. Shown are fold increases of Pearson correlation coefficients (P.C.C.) of p62 on LAMP1 signals, normalized to untreated siCtrl. Graph represents data from three independent experiments with = 30 cells per condition.HeLa cells were treated as in (C), but fixed after additional 2 h of chase. Samples were stained for galectin-3 (Gal3) to detect damaged lysosomes, and LC3B as marker for autophagosomes. Note the reduction of LC3B colocalization with Gal3 indicating impaired autophagosome formation. Arrows indicate colocalizing or non-colocalizing vesicles. Scale bars: 20 µm, 5 µm for inlays.Quantification of (E). Shown are fold increases of Pearson correlation coefficients (P.C.C.) of LC3B and Gal3 signals, normalized to untreated siCtrl, representing LC3B colocalization with Gal3. Graph represents data from three independent experiments with = 50 cells per condition.Western blot analysis of lysates of cells transfected with UBE2QL1 or control siRNA for 72 h, treated with 200 nM Bafilomycin A1 for 5 h or 250 µM LLOMe for 3 h as indicated and probed with an antibody specific for LC3A/B. GAPDH was probed as loading control.Data information: In (B, D, and F) data are presented as mean ± SD. *P < 0.05; ***P < 0.001 (one-way ANOVA with Bonferroni's multiple comparison test).Source data are available online for this figure.