Fig 2: (A) Fold changes of expression levels for VCP/p97 mRNA in two TNBC cell lines (MDAMB468 and MDAMB231) compared to the non-cancerous MCF-10A cells, evaluated by the 2-??Cq method and reported as log2 scale. We consider values higher than +1statistically significant. (B) Western blot validation of VCP/p97 and SELENOS expression on MCF-10A, MDAMB231 and MDAMB468 cell lines. ß-actin was used as a loading control. The numbers indicate the quantifications performed for VCP/p97 and SELENOS by IMAGEJ software.

Fig 3: SVIP-dependent VCP recruitment to lysosomes is required for lysosome integrity and function.a Schematic of SVIPKO mutant generated by CRISPR (top) and western blot analysis of SVIP protein levels in wild type (WT), SVIPKO heterozygous and SVIPKO homozygous mutants (bottom). b Representative images of Spin-RFP in 3rd instar larval muscles for the indicated genotypes. Quantification (right) for indicated genotype for tubule complexity (junctions per tubule) and tubule density (tubules/100 µm2). (n = 10 muscle cells examined over five independent animals). c Representative images of Spin-RFP in 1-week-old adult abdominal muscles for the indicated genotypes. Quantification as in b. (n = 8 muscle cells examined over four independent animals). d Representative images of mCherry-GFP-Atg8a (autophagy sensor) in wild type (WT) and SVIPKO larval muscles. Quantification at right. (n = 7 muscle cells examined over three independent animals). e Cartoon diagram of FKBP-FRP dimerization system (components as indicated). f Representative images of FKBP-Spin-mCherry and FRB-VCP-Venus localization in the absence (control) or presence of B/B dimerizer. Quantification at right as in b. (n = 6 muscle cells examined over three independent animals). Scale bars, 5 µm. See also Supplementary Fig. S2. Data presented as mean and SEM. Student’s t-test for individual comparisons (*p < 0.05; **p < 0.01; ***p < 0.001). ANOVA for multiple comparisons (*p < 0.05; **p < 0.01; ***p < 0.001).

Fig 4: (A) Immunohistochemical expression of human SELENOS and VCP/p97 at 200× magnification in normal mammary tissues and TNBC patients with grade 2 and 3. In grade 2 (G2) and grade 3 (G3) tissues the neoplastic cells form solid clusters separated by bands of connective tissue. (B) Correlation between SELENOS and VCP/p97 expression and tumor grading. (C) Pearson correlation between SELENOS and VCP/p97 scores obtained by IHC (R, correlation coefficient).

Fig 5: (A) Functional analysis performed by the STRING tool (https://string-db.org/, accessed on 1 May 2020) on the five more significant selenoproteins (GPX1, GPX4, SELENOS, TXNRD1 and TXNRD3). The bars indicate the number of selenoproteins involved in each Molecular Function and Biological Process Gene Ontology (GO), and Reactome and KEGG Pathways. (B) First-order interaction network obtained by extracting the selenoproteins from the TNBC network. We report selenoproteins in orange, HUB nodes in red and other nodes in gray. (C) Sub-network related to the links between more significant selenoproteins and two HUB nodes, CUL3 and VCP/p97. We report DIO2, GPX1, GPX4, SELENOF, SELENOK, SELENOS and TXNRD1 in orange, CUL3 and VCP/p97 in red and other nodes in gray.