Fig 2: DRAK1 is significantly underexpressed in chemoresistant cervical cancers.A Representative IHC image showing the expression of DRAK1 in chemosensitive or chemoresistant cervical cancer patient TMAs. Original magnification 200x. Scale bar, 50 μm. B Scatter plots showing the IHC score of DRAK1 from A. P values were calculated by paired two-tailed Student’s t tests. C Representative IHC image showing the expression of DRAK1 and TRAF6 in matched samples from TMAs. Original magnification 200x. Scale bar, 50 μm.

Fig 3: DRAK1 is degraded through K48-linked polyubiquitination induced by CUL3/SPOP E3 ubiquitin ligase.A Immunoprecipitation assay showing endogenous DRAK1 ubiquitination in HeLa and HeLa/PTX cells upon MG132 (10μM) treatment for 6 h. B Immunoprecipitation assay showing the interaction of Myc-DRAK1 with HA-CUL3 and Flag-SPOP in 293 T cells. C Immunoprecipitation assay showing the endogenous interaction of DRAK1 with CUL3 and SPOP in HeLa and HeLa/PTX cells. D In vivo ubiquitination assay showing Myc-DRAK1 ubiquitination by cotransfection of HA-CUL3 and Flag-SPOP in 293 T cells with MG132 (10 μM) treatment for 6 h. E In vivo ubiquitination assay showing Myc-DRAK1 ubiquitination by cotransfection of HA-CUL3 and Flag-SPOP in 293 T cells using Ni-NTA pull down of His-Ubiquitin WT or His-Ubiquitin 7KR mutant (mutating all seven lysines to arginines) with MG132 (10μM) treatment for 6 h. F Immunoprecipitation assay showing endogenous DRAK1 ubiquitin chain linkages. HA-CUL3 ectopically transfected in HeLa/PTX and indicated samples were immunoprecipitated with anti-DRAK1 antibody, followed by immunoblotting with anti-K48 Ub (K48-linkage specific Ubiquitin) or anti-K63 Ub (K63-linkage specific Ubiquitin) antibodies upon MG132 (10 μM) treatment for 6 h.

Fig 4: DRAK1 protein is degraded by CUL3/SPOP E3 ubiquitin ligase in paclitaxel-resistant cells.A Immunoblot analysis showing DRAK1 expression in HeLa and HeLa/PTX cells with or without MG132 (10μM) treatment for 6 h and β-actin was used for internal control. B List of DRAK1 binding protein candidates analyzed from the mass spectrometry-based crosslinking assay. C RT-PCR (left) and immunoblot analysis (right) showing CUL3 and DRAK1 expression in HeLa and HeLa/PTX cells. 18 s or β-actin was used for internal control, respectively. D Immunoblot analysis showing Myc-DRAK1 expression by cotransfection of HA-CUL3 and its adaptor proteins, Flag-KLHL20, Flag-KEAP1, and Flag-SPOP in 293 T cells. E Immunoblot analysis showing the expression of Myc-DRAK1 by dose-dependent changes in the expression levels of Flag-SPOP with HA-CUL3. F Immunoblot analysis showing Myc-DRAK1 expression by cotransfection of HA-CUL3 and Flag-SPOP with or without MG132 (10 μM) treatment for 4 h. G Immunoblot analysis showing endogenous DRAK1 expression in HeLa and HeLa/PTX cells transfected with CUL3-siRNAs and SPOP-siRNAs. H qRT-PCR showing the expressions of downregulated target genes by DRAK1 overexpression validated from Fig. 4C in samples from G.

Fig 5: Loss of DRAK1 protein is associated with paclitaxel resistance in cervical cancer cells.A Immunoblot analysis of parental HeLa and paclitaxel-resistant HeLa (HeLa/PTX) cells with indicated antibodies and β-actin was used for internal control. B qRT-PCR of DRAK1 and TRAF6 expressions in HeLa and HeLa/PTX cells and 18 s was used for internal control. The data represent the mean ± S.D. of three independent experiments. ns nonsignificant *p < 0.05; **p < 0.01; ***p < 0.001. C Cell viability assay of DRAK1-knockdowned HeLa cells upon paclitaxel treatment ranging from 2.5 to 40 nM. The data represent the mean ± S.D. of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001. D Cell viability assay of DRAK1-knockdowned HeLa cells and HeLa/PTX with parental HeLa cells upon paclitaxel treatment ranging from 2.5 to 40 nM. The data represent the mean ± S.D. of three independent experiments. ns nonsignificant *p < 0.05; **p < 0.01; ***p < 0.001. E Spheroid formation assay of DRAK1-knockdowned HeLa cells upon paclitaxel treatment (10 nM). Original magnification 100x. Scale bar, 20 μm. The data represent the mean ± S.D. of three independent experiments. ns non-significant *p < 0.05; **p < 0.01; ***p < 0.001. F Spheroid formation assay of HeLa and HeLa/PTX cells. Original magnification 100x. Scale bar, 20 μm. The data represent the mean ± S.D. of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001. G Cell doublings of HeLa and HeLa/PTX cells. The data represent the mean ± S.D. of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001. H In vivo tumor-formation assay of HeLa and HeLa/PTX cells subcutaneously injected into the flanks of NRGA mice (n = 8 per group). Primary tumor volumes were measured weakly starting 3 weeks after injection and mice were sacrificed at 8weeks. Representative primary tumor images (top) and tumor volumes (bottom) are shown. Error bars represent the mean ± S.D. of independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001. I Representative IHC image showing H&E, DRAK1, TRAF6, and Ki67 expression in primary tumor tissues from H. Original magnification 100x. Scale bar, 100 μm. J qRT-PCR of TRAF6 expressions showing TRAF6-knockdowned efficiency in HeLa and HeLa/PTX cells (left) and cell doublings of TRAF6-knockdowned HeLa and HeLa/PTX cells (right). The data represent the mean ± S.D. of three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001.