Fig. 6: Molecular pathway analysis of cell cycle regulators indicated CDK1 as therapeutic biomarker for Chk1 response.

a Levels of cell cycle regulating proteins of five HNSCC cell lines and one primary oral fibroblast culture were examined on Western blot. Cells were harvested 12 h and 24 h after treatment with 2 µM LY2603618/Rabusertib. Phosphorylated γH2Ax Ser139 was found in all cell lines after 24 h Chk1 inhibition indicating DNA damage, but was almost absent in primary cells. Cyclin B1 protein levels did not correlate to sensitivity for Chk1 inhibition. b Baseline proteins levels of CDK1 are shown for HNSCC cell lines UM-SCC-22A, UM-SCC-38, VU-SCC-120, FaDu, VU-SCC-096, and primary oral fibroblasts. The levels of CDK1, correlated borderline significantly (two-sided p-value = 0.057) reverse correlation to Chk1 inhibition response (Fig. S5b). Furthermore, protein levels CDK1 were low in primary oral fibroblasts compared to the tumor cell lines. c Quantification of basal protein levels (Fig. 6b) of CDK1 expression levels showed a borderline significantly (two-sided p-value = 0.057) reverse correlation with Chk1 inhibition response (Fig. S5b). CDK1 levels might be applicable as a clinical biomarker for Chk1 inhibition response. d Microarray gene expression data of 22 tumors revealed a significant upregulation of CDK1 and Cyclin B1 (CCNB1) in tumors when compared to the paired primary mucosa (both p < 0.0001, paired (two-sided) t-test). CDK1 and Cyclin B1 expression varied within the patient cohort with a factor eight, but the role of stromal percentage was not taken into consideration. On the y-axis the relative expression level is displayed. See also legend of Fig. 1e. e The dose–response (y-axis, shown in relative cell viability) of siCDK1 dilution range (x-axis) for untreated (in gray) and 750 nM LY2603618 treated (in black) conditions. Chk1 inhibition was started 24 h post transfection. Complete knockdown (mRNA < 10%, Fig. S5f) of CDK1 resulted in resistance to Chk1 inhibition, indicating that lowering the CDK1 levels in a high expressing cell line does not increase responsiveness to Chk1 inhibition. f Combination treatment of CDK4/6 inhibitor Palbociclib (EC10 and EC20) and a serial dilution of Chk1 inhibitor LY2603618/Rabusertib (x-axis). CDK4/6 was inhibited for 8 h (typical length of mammalian S-phase) before a serial dilution of LY2603618/Rabusertib was added. CDK4/6 inhibition partially reversed the Chk1 effects on viability. g The cell cycle distribution, analyzed by DNA content (PI), confirmed a partial G1/G0-arrest with the EC10 of Palbociclib and a total G1/G0-arrest with Palbociclib EC20, for both single treatment as well as in combination with LY2603618/Rabusertib. h, i Combining EC10 concentrations of Wee1 inhibitor Adavosertib (formerly known as AZD1775 or MK-1775) and Chk1 inhibitor LY2603618/Rabusertib, induced an additive effect in HNSCC cell lines UM-SCC-22A and VU-SCC-096. An additional EC40 concentration of LY2603618/Rabusertib was tested in combination with the same Adavosertib concentration as well for UM-SCC-22A. Results of combining Chk1 with Wee1 inhibition in cell line VU-SCC-120 is shown in Fig. S5j. These findings support the hypothesis that combination therapies that facilitate cell cycle progression magnify the toxicity of each of the inhibitor alone