Fig 2: Sym004 promotes ubiquitin-independent, ESCRT-dependent lysosomal degradation of EGFR. (a) SCC47 cells were serum-starved for 1 h, then MG132 (10 µM) or bafilomycin A1 (600 nM) were added, followed 1 h later by cisplatin (50 µM), cetuximab, Sym004 or EGF. The cells were incubated overnight and lysed with RIPA buffer. (b) SCC47 cells were serum-starved for 1 h, then bafilomycin A1 (600 nM) was added followed 1 h later by Sym004-AF568 or EGF-AF647. The cells were incubated overnight, fixed and stained with EGFR-AF488 antibody. (c) Serum-starved SCC47 were treated with cetuximab, Sym004 or EGF as indicated, then lysed. EGFR was immunoprecipitated with anti-EGFR antibody (normal mouse IgG used in Control). For quantifications, ubiquitylated EGFR was normalised to total EGFR. (d) Serum-starved, Hrs- or Tsg101-depleted (or control) SCC47 cells were treated with Sym004 (3 µg/ml or 30 µg/ml) or EGF for 6 h, then lysed. (e) The cells treated as in (d) were fixed and stained with EGFR-AF488. Nuclei were stained with Hoechst 33342. All immunoblots were cropped for clarity.

Fig 3: ZNRF1 mediates the recruitment of ESCRT machinery to internalized EGFR. (A–D) Wild type or ZNRF1–/– A549 cells were serum-starved overnight and stimulated with 100 ng/mL EGF for 0, 5, 10, and 20 min. (A) Cells were co-stained with EGFR (Green) and HRS (Red) antibodies. (B) Quantification of the fraction of EGFR co-localized with HRS in (A). (C) Cells were co-stained with EGFR (Green) and TSG101 (Red) antibodies. (D) Quantification of the fraction of EGFR co-localized with TSG101 in (C). (E) Cell lysates from wild type and ZNRF1–/– A549 cells were prepared and the levels of HRS and TSG101 were analyzed by immunoblotting. (F) Quantification of the size distribution of EGFR puncta in (A). White arrows indicate non-colocalized signals. Scale bar, 10 µm. Data are presented as mean ± SEM. n > 1000 puncta from at least 50 cells for each group; *P < 0.05; **P < 0.01; ***P < 0.001. All experiments were repeated two to three times with similar results.

Fig 4: IFN-? and Ca2+ promote sEV PD-L1 secretion in H1299 cells. (a) A representative transmission electron microscopy (TEM) image of a purified sEV from H1299 cells. Scale bar, 100 nm. (b) sEVs were isolated by sucrose gradient density ultracentrifugation, followed by the immunoblotting for PD-L1 and sEVs markers CD63 and HRS. (c) Nanoparticle tracking analysis (NTA) was conducted on H1299 cells to determine the size distribution of sEVs isolated by ultracentrifugation. Traces represent mean values in units of million particles per ml from 3 experiments. (d) Immunoblots for PD-L1 and sEVs markers ALIX, CD63 and CD9 in whole-cell lysates and sEVs from H1299 cells treated with or without IFN-? (10 ng/ml). (e) Immunoblotting analysis of PD-L1, EVs markers and organelle makers performed on whole-cell lysates (WC), m/lEVs and sEVs from H1299 cells. (f) Effect of IFN-? pretreatment on thapsigargin-(Tg 2 µM) and ATP-(100 µM) mediated Ca2+ dynamics. Values of fluorescent intensity represent mean ± SEM, which is normalized to the basic fluorescence intensity. All data is derived from more than 4 wells in 3 individual experiments. (g) Schematic illustration of PD-L1_pHluorin plasmid (upper) and proposed sEVs releasing model (below) through visualization of PD-L1_pHluorin. sEV PD-L1_pHluorins are quenched while resident in the acidic lumen of the multivesicular body (MVB). Immediately after sEVs secretion, the pHluorin is exposed to the neutral extracellular medium, causing a sudden increase in fluorescence intensity. (h) Total projection of secretion events over a 180 s was recorded from a cell before (top) and after (bottom) stimulation with Tg (2 µM). (i) Counts of fusion events within a time window of 180 s in individual H1299 cells (n = 11) transfected with PD-L1_pHluorin. Two points connected by a line represent the fusion events numbers from the same cell but under different conditions (before Tg vs. after Tg). For all of Figure 1, data are representative of at least three independent experiments. ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001.

Fig 5: HRS depletion substantiates increased STING responses AdCas9-KRAB expressing iBMDMs, without (i.e. dCas9 alone) or with HRS-targeting sgRNAs (i.e., sg1, sg2 or sg3) were treated for 48 h with 1 µg/ml doxycycline (Dox) to induce sgRNA expression. Cells were then left untreated (UT), or treated with 20 µg/ml DMXAA or 10 µg/ml 2'3'-cGAM(PS)2 for 6 h. Cells were lysed for immunoblot with the indicated antibodies. Data shown are representative of three independent experiments.B, CdCas9-KRAB expressing iBMDMs, without (i.e. dCas9 alone) or with HRS-targeting sgRNAs (i.e. sg1, sg2 or sg3) were treated for 48 h with 1 µg/ml Dox to induce sgRNA expression. Cells were then left UT, or treated with 20 µg/ml DMXAA or 5 µg/ml 2'3'-cGAM(PS)2 for 6 h. Cell supernatant was collected and assayed for secreted IFNß (B) and IL6 (C) by ELISA. Data are shown as mean ± SEM combined from N = 3 independent experiments. Statistical analysis was performed using two-way ANOVA using Bonferroni's multiple comparisons test, where *P < 0.05, ***P < 0.001, ****P < 0.0001. n.d., not detected.D–FdCas9-KRAB expressing iBMDMs, without (i.e. dCas9 alone) or with HRS-targeting sgRNAs (i.e. sg1, sg2 or sg3) were treated for 24 h with 1 µg/ml Dox to induce sgRNA expression. iBMDMs were further treated with 1 µg/ml (i.e. 1,000 ng/ml) or 100 ng/ml H-151 as indicated and incubated for a further 24 h. Cells were lysed for RNA purification and the expression of Ifnb1 (D), Isg15 (E) and Irf7 (F) was analysed by qPCR. Data are shown as mean ± SEM combined from N = 3 independent experiments. Statistical analysis was performed using two-way ANOVA using Tukey's multiple comparisons test, where *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Bars display significant differences between dCas9 and sgRNA knockdowns in the absence of H-151. Additional asterisks represent significant differences between the absence or presence of H-151 across the same cell line. Source data are available online for this figure.