Fig 2: Identification of the CAIX Interactome using the BioID proximal ligation strategy and validation of high confidence proximal CAIX-associating proteins by co-immunoprecipitation. (a) Schematic showing the domain structure of the CAIX-BirA* fusion protein used for the BioID studies. SP, signal peptide; PG, proteoglycan-like domain; CA, catalytic domain; TM, transmembrane domain; IC, intracellular domain. (b) FACS analysis showing levels of constitutive expression of the CAIX-BirA* fusion protein expressed by MDA-MB-231 cells (MDA-MB-231-CAIX-BirA*) cultured in normoxia compared to levels of endogenous CAIX expression by WT MDA-MB-231 cells cultured in either normoxia or hypoxia. The filled (gray) histograms indicate specific CAIX staining whereas the dotted lined histograms indicate staining by isotype-specific control IgG. (c) Image showing exogenous CAIX (red) localized at the cell surface of MDA-MB-231-CAIX-BirA* cells grown in normoxia. Scale bar, 10 µM. (d–g) Co-IP of high confidence proximal CAIX-associating proteins (d) ITGB1, (e) ITGA2 (f) CD98hc and (g) MMP14 with exogenous CAIX expressed by MDA-MB-231-CAIX-BirA* cells cultured in normoxia. (h–j) Co-IP of (h) ITGB1, (i) ITGA2 and (j) MMP14 with endogenous CAIX expressed by WT MDA-MB-231 cells cultured in either normoxia (N) or hypoxia (H). Isotype-specific IgG was used as a control for co-IPs and data in (d) to (j) are representative of at least two independent experiments. IgG, immunoglobulin G.

Fig 3: Cancer cell-matrix adhesion modulates the tapering of strands.(a) Principal component analysis plots show the metrics derived from over 2000 simulations in the presence of fibroblasts covering variation in cancer cell–matrix adhesion with values indicated by the intensity of green, cancer cell proteolysis (not colour coded), and cancer cell–cancer cell adhesion (not colour coded). (b) Heatmaps show how varying the cancer cell–matrix adhesion value (x axis) impacts on different metrics when fibroblasts are included in all simulations. WT indicates the ‘wild-type’ value based on experimental parameterisation using A431 cancer cells. Yellow indicates a high value, dark blue a low value. (c) Images show the effect of modulating matrix adhesion via Crispr KO of ITGB1 in cancer cells (green) in both organotypic and spheroid assays including fibroblasts (magenta). Scale bar = 100 µm. (d) Quantification of three biological replicates of the experiment shown in panel (c) with strand length, strand width, and tapering metric shown – 1 unit is equivalent to 0.52 µm. Unpaired t-test was performed. 95% confidence intervals are shown, one dot represents one strand. Figure 4—source data 1.Quantification of invading strand length, width, and tapering in A431 cells with/without ITGB1 manipulation.