Fig 2: Phosphorylation of TNK1 at S502 within the proline rich domain is required for TNK1 binding to 14-3-3.a Schematic diagram of wild type (WT) TNK1 and our panel of TNK1 truncations. FLAG-TNK1 (WT or indicated mutants) were immunoprecipitated from HEK-293T cells and immunoblotted for 14-3-3. The graph shows quantitation from n = 3 biological replicates with signals normalized to WT. Error bars represent SEM. P-values were calculated using a two-tailed student t-test. b HEK-293A cells expressing TNK1-GFP (WT or ?UBA) were analyzed by confocal imaging for TNK1 localization. The graph shows the number of cells, counted manually, with diffuse or punctate TNK1 localization (n = 50). The scale bar is 10 µm. c Schematic diagram of phosphorylations identified by TiO2 phospho-proteomics within the proline-rich domain. The lower diagram shows an alignment of the TNK1 sequence encompassing pS502 (human numbering) from human to zebrafish. d Graph (green bars) shows the number of recorded observations of S/T phosphorylations on TNK1 via high-throughput methods (data derived from phosphosite.org) and the predicted disorder (cyan lines: ANCHOR score derived from https://iupred2a.elte.hu) across the sequence of TNK1. e WT and mutant FLAG-TNK1 were immunoprecipitated from HEK-293T cells and immunoblotted for 14-3-3 and with a custom phospho-specific antibody to pS502 of TNK1. Graphs show quantitation of infrared signals from multiple biological replicates (n = 3) with signals normalized to WT. Error bars represent SEM. P-values were calculated using a two-tailed student t-test. f WT and mutant FLAG-TNK1 were immunoprecipitated from HEK-293T cells and immunoblotted for 14-3-3. Graphs show quantitation of infrared signals from multiple replicates (n = 3) with signals normalized to WT. Error bars represent SEM. P-values were calculated using a two-tailed student t-test. g Bio-layer interferometry (BLI) was performed to characterize the binding between 14-3-3? and TNK1 peptides. Graphs show representative BLI sensorgram and fit for 14-3-3 ? and corresponding TNK1 peptide. Kds indicated with standard deviation from multiple replicates (n = 3). Source data are provided as a Source Data file.

Fig 3: A model of TNK1 regulation showing the roles of MARK-mediated 14-3-3 binding and the contribution of the UBA to TNK1 kinase activation.Our data suggest that MARK-mediated phosphorylation on TNK1 allows 14-3-3 binding, which restrains TNK1 activity by preventing its interaction with ubiquitin-rich clusters, in which TNK1 is active. Once released from 14-3-3, interactions between TNK1 and ubiquitin are essential for full TNK1 activity. Mutations that disrupt 14-3-3 binding activate TNK1, rendering it capable of driving growth factor independence and tumor growth in vivo. Also shown is the small molecule TP-5801, which inhibits TNK1 at low nM levels and blocks TNK1-driven proliferation.

Fig 4: Quantitative phospho-tyrosine proteomics reveals a TNK1-mediated network of phospho-substrates and pro-growth signaling via STAT3.a Schematic of experimental design for phospho-tyrosine proteomics. The heat map shows the hierarchical clustering of all 2294 phospho-tyrosine peptides from three biological replicates of each sample. b A focused heat map showing the relative phospho-tyrosine signal of selected significant substrates across the different samples. Z scores were calculated by (value-mean)/standard deviation. c Volcano plots of phosphorylation fold changes between the indicated samples. Y axis = -log10 of p value. X axis = fold change. Gray dashed lines represent 2-fold changes and p value = 0.05. Selected phospho-substrates are highlighted in red. d Ba/F3 cells transformed with TNK1 (WT or AAA), BCR-ABL, or parental Ba/F3s were immunoblotted for phospho-PLC-? (Y783), total PLC-?, phospho-STAT3 (pY705), total STAT3, phospho-STAT5 (Y694), total STAT5 and ß-actin. Graphs show quantification from n = 3 biological replicates. Phosphorylation signals were normalized to respective total protein signals. pSTAT3/STAT3 signals were normalized to TNK1 WT. pSTAT5/STAT5 signals were normalized to BCR-ABL. Error bars represent SEM. P-values were calculated using a two-tailed student t-test. e Source data are provided as a Source Data file.

Fig 5: Full activation of TNK1 requires release from 14-3-3 and interactions between the TNK1 UBA and ubiquitin.a Ba/F3 or FDCP1 cells were transduced and analyzed as in Fig. 3, with either pMIG-empty vector, TNK1 (WT or indicated mutants), or BCR-ABL. IL-3 independent growth was monitored as in Fig. 3g. The graph shows average cell confluency with error bars representing SEM from 3 biological replicates. b Ba/F3 cells from panel a were immunoblotted for global phospho-tyrosine and GFP (expressed from virus). Graphs show average global tyrosine phosphorylation quantitation from multiple biological replicates (n = 6) with signals normalized to WT. Error bars represent SEM. P-values were calculated using a two-tailed student t-test. c 1 × 106 Ba/F3-luc cells transduced with WT TNK1 or the indicated TNK1 mutants were injected into the tail vein of NOD/SCID mice (n = 5). Luminescent signal was imaged and quantified using IVIS imaging and ROI analysis. Images shown are mice in prone position 28 days after injection. d Corrected Flux (raw luminescent signal minus background luminescent signal) of mice from panel c. The box plot shows the median, first and third quartile of corrected flux (n = 5). P-values were calculated using a two-tailed student t-test. e Survival plot of mice from panel c (n = 5). Source data are provided as a Source Data file.