Regulation of Src tumor activity by its N-terminal intrinsically disordered region

Abstract

The membrane-anchored Src tyrosine kinase is involved in numerous pathways and its deregulation is involved in human cancer. Our knowledge on Src regulation relies on crystallography, which revealed intramolecular interactions to control active Src conformations. However, Src contains a N-terminal intrinsically disordered unique domain (UD) whose function remains unclear. Using NMR, we reported that UD forms an intramolecular fuzzy complex involving a conserved region with lipid-binding capacity named Unique Lipid-Binding Region (ULBR), which could modulate Src membrane anchoring. Here we show that the ULBR is essential for Src's oncogenic capacity. ULBR inactive mutations inhibited Src transforming activity in NIH3T3 cells and in human colon cancer cells. It also reduced Src-induced tumor development in nude mice. An intact ULBR was required for MAPK signaling without affecting Src kinase activity nor sub-cellular localization. Phospho-proteomic analyses revealed that, while not impacting on the global tyrosine phospho-proteome in colon cancer cells, this region modulates phosphorylation of specific membrane-localized tyrosine kinases needed for Src oncogenic signaling, including EPHA2 and Fyn. Collectively, this study reveals an important role of this intrinsically disordered region in malignant cell transformation and suggests a novel layer of Src regulation by this unique region via membrane substrate phosphorylation.

Fig. 1. Evolutionary conservation of the UD in SFKs.

a Analysis of dN/dS ratios (ω) of the SH4, UD, SH3, SH2 and kinase domains in primate SFKs. Nucleic sequences coding the different domains were aligned based on translation and processed by codeml to estimated ω-values. The phylogenetic tree on the left was generated by PhyML analysis of a multiple alignment of full-length SFK sequences. b Multiple alignment of UD sequences of primate SFKs. Sequences were aligned with MAFFT. The most conserved residues are indicated above the alignment. The strictly conserved FGG of the ULBR is framed. Numbering corresponds to human c-Src sequence. H sap Homo sapiens, P tro Pan troglodytes, G gor Gorilla gorilla, P abe Pongo abelii, M mul Macaca mulatta, C jac Callithrix jacchus, O gar Otolemur garnettii, G gal Gallus gallus, X tro Xenopus tropicalis, D rer Danio rerio.

Fig. 2. ULBR inactivation affects Src oncogenic activity.

a Strategy of ULBR inactivation. Molecular properties of Src-ULBR including amino acids involved in this process are highlighted. b the level of Src expression in NIH3T3 cells transduced with indicated Src constructs. Anchorage-independent cell growth (c) and invasion (d) of indicated Src-transformed NIH3T3 cells. e The level of Src expression and activity in SW620 colon cancer cells transduced with indicated Src constructs. Anchorage-independent growth (f) and invasion (g) of SW620 cancer cells expressing indicated Src constructs. The histograms show the percentage of colonies in soft agar normalized to the maximal condition set a 100% (colonies) and the percentage of migrating cells in the matrigel matrix normalized to control condition set at 100% (cell invasion). Is shown the mean ± SD; n = 3; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; Student’s t test.

Fig. 3. ULBR inactive mutations affects Src tumor activity in nude mice.

a, b Time-course of tumor development in nude mice subcutaneously inoculated with SW620 tumor cells that were transduced with control (mock) or indicated Src construct. Analysis of tumor cell proliferation (c), apoptosis (d) and angiogenesis (e) from indicated tumor sections. Representative sections and quantification of immunohistochemical analysis showing tumor cell proliferation (anti-ki67), apoptosis (anti-cleaved Caspase 3) and angiogenesis (anti-CD31; length of blood vessels) in xenograft tumors derived from SW620 cells transduced with indicated Src construct. Is shown the mean ± SEM (a) and the mean ± SD (b–e); n > 8 mice per cohort; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; Student’s t test.

Fig. 4. ULBR inactive mutation inhibits MAPK signaling without affecting Src localization and kinase activity.

a, b ULBR regulates Src-induced protein tyrosine phosphorylation and of p42/44 MAPK activation. Immunoblots and relative band intensity quantification of whole-cell lysates showing cellular protein tyrosine phosphorylation and of p42/44 MAPK activity in HEK293T cells (a) and SW620 cells (b) transduced with indicated Src constructs (n = 5). c Src dimerization is not affected by ULBR inactivation. HEK293T cells were transfected with the indicated constructs. Src-Myc proteins were immunoprecipitated (IP) from cell lysates and immunoblotted with the indicated antibodies. Immunoblots of whole-cell lysates were also performed as indicated (n = 2). d Representative confocal image of direct fluorescence of HEK293T cells co-expressing Src-mCherry and indicated Src-GFP ULBR mutants. The overlay is also shown. e In vitro kinase assay of purified Src-GFP and Src3A-GFP that were expressed in HEK293T cells using indicated substrate. The level of immunoprecipitated Src-GFP proteins and tyrosine phosphorylation of indicated substrate is shown (n = 3).

Fig. 5. phospho-proteomic analysis of Src-ULBR signaling in SW620 cancer cells.

a A label-free quantitative phospho-proteomic analysis centered on tyrosine phosphorylation. A Veen diagram where quantified phospho-peptides were sorted as differentially phosphorylated from the control condition (mock) (log2FC ≥ 1) in the indicated Src (or Src3A) conditions. b A phospho-RTK array approach. c A phospho-signaling kinase array approach. Comparison of Src (gray boxes) and Src3A (white boxes) induced tyrosine phosphorylation of RTKs and phosphorylation of signaling kinases. Is shown the phosphorylation level of selected kinases relative to the mock condition (fold control; duplicates from 2 independent experiments). d–f Fyn and EPHA2 are important mediators of ULBR-Src signaling in SW620 cancer cells. d, e Biochemical analysis and relative band intensity quantification of p42/44 MAPK and Akt activity in SW620 expressing or not Src or Src3A mutant as shown and transfected with indicated siRNA (n = 3). The level of EPHA2 and Fyn is also shown (n = 2). f Cell invasion of SW620 expressing or not Src or Src3A mutant and transfected with indicated siRNA. The histograms show the percentage of migrating cells in the matrigel matrix normalized to control condition set at 100% (cell invasion). Is shown the mean ± SD; n = 4; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001; Student’s t test.