Fig 2: UNC119 KD in HT-29 CRC cells inhibits SFK activity and cell proliferation. a Representative western blot showing UNC119 expression level, total SFK (tSFK), total FAK (tFAK), tubulin (as loading control) and SFK phosphorylated on Y419 (pY419-SFK) and FAK phosphorylated on Y925 (pY925-FAK) in HT-29 cells expressing Cas9 and a doxycycline-inducible sgRNA vector, containing either a non-targeting guide RNA (c), or UNC119A (a) or UNC119B (b) targeting guide RNAs, induced with doxycycline for 72 h. b Bar graph depicting the mean change in expression upon sgRNA induction for pYSrc/tSrc, pYFAK/tFAK and UNC119, compared to the non-targeting sgRNA control (n = 5 independent experiments; data are mean ± SE. P values determined by one-way ANOVA with Bonferroni post hoc test). c HT-29 cells were transfected with UNCA-mCh or UNCB-mCh and probed for SFK phosphorylation by western blotting. Plot depicts pYSFK/tSFK normalised to non-transfected control against UNC119-mCherry expression normalised to GAPDH loading control from western blots (see also Supplementary Fig. 8b). d Fluorescence micrographs of HT-29 Cas9 cells expressing the doxycycline-inducible UNC119B sgRNA and treated with or without doxycycline for 72 h, transfected with Src-mCh, Yes-mCh or Fyn-mKate2 (see also Supplementary Fig. 8c). Dot plot depicts the endomembrane/total integrated fluorescence intensity for each SFK (n = 28 or 29 cells for Src-mCh, n = 28 or 34 cells for Yes-mCh and n = 12 or 14 cells for Fyn-mKate2; data are mean ± SD from two independent experiments, ***P < 0.001; ****P < 0.0001, as determined by Student’s t-test). e Clonogenic assay of HT-29 Cas9 cells containing the indicated sgRNAs and induced with doxycycline for the duration of the assay. f Bar graph depicting the mean colony area coverage upon induction of sgRNA, normalised to the non-targeting sgRNA control well. g Bar graph depicting the mean colony number change upon induction of sgRNA, normalised to the pre-doxycycline control wells. h Bar graph depicting the mean individual colony size change upon induction of sgRNA, normalised to the pre-doxycycline control wells (n = 4 independent experiments; data are mean ± SE; **P < 0.01; ***P < 0.001; ****P < 0.0001; n.s., not significant as determined by one-way ANOVA with Bonferroni post hoc test). Scale bars, 10 μm

Fig 3: Src released from UNC119 by Arl2/3 is trapped at the RE to be transported to the PM. a Representative time-lapse sequences of fluorescence distribution of HeLa cells co-expressing Src-mCh (first and third rows) and SrcWT-paGFP (second and fourth rows) in the absence (first and second rows) or presence (third and fourth rows) of UNC119A-TagBFP ectopic expression. Src-paGFP was photoactivated at 0 s in the perinuclear region and subsequently Src-mCh was photobleached at 400 s in the same area (photoactivation/photobleach area: red outline). b Segmentation of cells in (a): RE (red), PM (blue outline) and communicating cytoplasm (green). The ratio of Src-paGFP fluorescence per segment (RE, red; PM, blue; cytoplasm, green) over total fluorescence in the cell and normalised to the respective ratio of Src-mCh fluorescence was plotted against time. The PM ratio was fitted to a mono-exponential recovery (black line). c Average traces of FLAP and FRAP curves fitted to the compartmental model described in Methods. Loss of Src-paGFP fluorescence at the perinuclear region normalised to whole-cell Src-paGFP fluorescence (blue and cyan curves) and corresponding gain of Src-mCh fluorescence after photobleaching at the same ROI, normalised to whole-cell Src-mCh fluorescence (red and orange curves). Control cells: mean ± SEM; n = 5 cells (blue and red curves), UNC119 ectopic expression: mean ± SEM; n = 6 cells (cyan and orange curves). Scale bars, 10 µm

Fig 4: Spatial cycles of Src and Fyn counter entropic equilibration to endomembranes to maintain their PM localisation. SFKs are displaced by endocytic processes and spontaneous dissociation from the plasma membrane and equilibrate with the vast endomembrane system. Depalmitoylation by APT activity enhances dissociation of Fyn from membranes, whereas Src rapidly dissociates from the non-charged endomembrane surfaces. In the cytosol, the GDI-like solubilisation factor UNC119 binds to and enhances diffusion of the myristoylated SFKs. Arl2/3-GTP activity on and proximal to the recycling endosome discharges Src from UNC119 and concentrates it onto the RE where it gets trapped by electrostatic interaction with the negatively charged membranes (scheme on the left). In contrast, Arl2/3-GTP-mediated discharge of depalmitoylated Fyn from UNC119 on the recycling endosome cannot lead to electrostatic trapping because of the lack of a polybasic stretch in Fyn, resulting in rapid dissociation and resolubilisation by UNC119. Spontaneous dissociation of Fyn from UNC119 will result in membrane ‘hopping’ in this perinuclear area until kinetic trapping of Fyn at the Golgi by palmitoyl acyl-transferases catalysed palmitoylation (scheme on the right). The concentrated SFKs are then trafficked along the constitutive anterograde vesicular transport from these organelles, back to the PM. The blue arrows denote entropic equilibration to endomembranes; red arrows represent the countering, energy-driven relocalisation process

Fig 5: UNC119 solubilises SFKs in a myristoyl-dependent manner. a, b FRET-FLIM of the interaction between either mCitrine fused to Src a or Fyn b and UNCA-mCh in HeLa cells (see also Supplementary Figs. 3 and 4). For each sample, the fluorescence intensity of SFK-mCit (donor), the fluorescence intensity of UNCA-mCh (acceptor), the spatial distribution of the mean fluorescence lifetime (t) in nanoseconds and the molar fraction of interacting molecules (a) are shown per the false-colour look-up tables. The top rows depict the steady-state localisation of wild type (WT) SFK-mCit when co-expressed with UNCA-mCh. The second rows show representative measurements with the polybasic stretch mutant Src6Q-mCit a or non-palmitoylatable mutant FynC3S-C6S-mCit (b), and the bottom row shows the non-myristoylatable mutant SFKG2A-mCit proteins when co-expressed with UNCA-mCh. Bar graphs on the right depict average t (c, e) and a-values (d, f) for different SFK proteins without (-), with UNCA-mCh (a), or UNCB-mCh co-expression (b; n = 10–20 cells/condition, data are mean ± SD from three independent experiments; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 as determined by one-way analysis of variance (ANOVA) with Bonferroni post–hoc test). For a significance determination, samples were compared to the donor only control. Cells were sectioned into 10 segments of equal radius from the PM to the centre of the nucleus, and the mean intensity for each channel and a calculated for each segment (g–j), for the Src-mCit profile angular masking was performed to look at the profile change across the perinuclear RE compartment. Line diagrams depict the normalised intensity for each channel (SFK-mCit in green and UNCA-mCh in magenta) and the mean a (black trace) in each segment ± SEM; the vertical dotted line indicates the average localisation of the nuclear membrane (n = 13 cells for Src-mCit, Src6Q-mCit and FynC3S-C6S-mCit, and n = 11 for Fyn-mCit). Scale bars, 10 µm