Fig 2: Synergistic Lethality of AML Cells by Dual Targeting of PAK4 andMCL-1. (A) PAK4 mRNA and protein were detected by qPCR (left) and immunoblots (right) in HL-60 cells. Data shown represent the means (±SD) of technical triplicates, and these experiments were independently repeated at least three times with the similar results. (B) Live HL-60 cells with or without PAK4 depletion were counted at the indicated time points and cell growth was plotted as shown. Data shown are means (±SD) of technical triplicates. (C) Live HL-60 cells treated with PAK4 inhibitor KPT-9274 were counted at the indicated time points and cell growth was plotted as shown. Data shown are means (±SD) of technical triplicates. (D–F) HA-MYC was overexpressed in HL-60 cells with PAK4 depletion. PAK4 and MYC mRNA and protein were detected by qPCR (D) and immunoblots (E). Live cells treated as described were counted at the indicated time points and cell growth was plotted as shown (F). Data shown are means (±SD) of technical triplicates. (G) Assessment of cell death in PAK4-depleted HL-60 cells. Data shown are means (±SD) of technical triplicates. (H) Immunoblots of MYC, MCL-1, BCL2 and Bcl-XL in PAK4-depleted HL-60 cells. (I) Immunoblots of MYC and MCL-1 in HL-60 cells treated with KPT-9274 for 72 h. (J) Assessment of cell death in HL-60 cells treated with 100 nM KPT-9274 and/or 50 nM S63845 for 48 h. Combo, combination treatment. Data shown are means (±SD) of technical triplicates.

Fig 3: PAK4 depletion Promotes FBXW7 mediated MYCubiquitylation-proteasomaldegradation. (A) PAK4-depleted HL-60 and OCI-AML3 cells were treated with MG132 (10 μM) for 6 h before harvest. PAK4, MYC and p-S67 MYC were analyzed by immunoblot, with ACTIN as a loading control. (B) Analysis of endogenous MYC polyubiquitylation. Denaturing protein IP with anti-MYC antibody and immunoblots were performed in HL-60 and OCI-AML3 cells with or without PAK4 depletion. (C) Analysis of endogenous MYC polyubiquitylation in PAK4-depleted HL-60 cells with or without enforced expression of HA-PAK4 or HA-PAK4 KI. Denaturing protein IP with anti-MYC antibody and immunoblots were performed. (D) FBXW7 was depleted by shRNA in HL-60 and OCI-AML3 cells with or without PAK4 depletion. MYC protein was analyzed by immunoblot. (E) Immunoblot analysis in HEK293FT cells co-expressing HA-MYC, Myc-FBXW7α and Flag-PAK4 as indicated. (F) Analysis of MYC polyubiquitylation. Denaturing proteins from HEK293FT cells expressing indicated constructs were pulled down with streptavidin beads and subjected to immunoblot analysis. MG132 (10 μM) was added in HEK293FT cells for 6 h before harvest. SBP: streptavidin-binding peptide. (G) Analysis of MYC polyubiquitylation. HA-tagged ubiquitin was co-transfected with Myc-FBXW7α and Flag-MYC (WT, S67A or S67D) into 293T cells. The resulting cell lysates were subjected to co-IP by anti-Flag, followed by immunoblotting of ubiquitin.

Fig 4: PAK4 mediated S67 Phosphorylation Promotes AML cells proliferation. (A) Immunoblots of MYC and p-S67 MYC in primary T-ALL cells, T-ALL cell lines and AML cell lines. (B) Immunoblots of MYC in MYC depleted HL-60 cells with or without enforced expression of HA-MYC, HA-MYC S67D or HA-MYC S67A. (C) Live HL-60 cells treated in (B) were counted at the indicated time points and cell growth was plotted as shown. Data shown are means (±SD) of technical triplicates. (D) Immunoblots of MYC and p-S67 MYC in various cells treated with 10 nM AZD1152 for 24 h or 0.5 μM BMS-345541 for 12h as indicated. (E) Potential kinases phosphorylating MYC at serine 67 predicted from GPS 5.0, PhosphoSitePlus and NetPhos 3.1. (F) Immunoblots of MYC and p-S67 MYC in HL-60 cells treated with TC-DAPK6 (DAKi), Rock inhibitor-2 (ROCKi), HA-1004 (PKA/G/Ci) and KPT-9274 (PAK4i) for 2 h or 4 h. (G) Immunoblots of MYC and p-S67 MYC in HL-60 cells treated with LCH-7749944 and PF-3758309 for 1 h or 2 h. (H–I) In vitro kinase assay of recombinant GST-MYC. Active human PAK4 proteins were incubated with GST-MYC for kinase reaction. Phosphorylation was detected by immunoblots using phosphorylated protein antibodies as shown. Loading controls were shown as Coomassie blue staining in the bottom panels (H). The product of kinase reaction from (H) were treated with calf-intestinal alkaline phosphatase (CIP) for 30 min before immunoblot (I).

Fig 5: PAK4 Directly Interacts with MYC at MBⅡ domain. (A) HL-60 cell lysates were subjected to reciprocal co-IP and immunoblot to detect endogenous PAK4 and MYC interaction. IP: immunoprecipitation. (B) Lysates of HEK293FT cells overexpressing (HA-MYC and Flag-PAK4) or (HA-PAK4 and Flag-MYC) were subjected to co-immunoprecipitation (co-IP) to detect protein interaction. (C) GST pull-down to detect direct PAK4 and MYC interaction. Recombinant His-tagged PAK4 was incubated with GST-MYC in vitro before pull-down assay with GST beads, and interaction was analyzed by immunoblot. GST: glutathione-S-transferase. (D) Schematic presentation of various human MYC truncations used in PAK4-binding assays. (E) Lysates from HEK293FT cells overexpressing Flag-PAK4 and HA-MYC truncations (amino acids 1–221 and 220–439) were subjected to co-IP and immunoblot. (F) Characterization of the MB motif required for MYC and PAK4 interaction. Lysates from HEK293FT cells overexpressing HA-PAK4 and Flag-SBP-MYC truncations (amino acids 1–221 or amino acids 1–221 with Myc box domain deletion mutants (ΔMB) as shown) were subjected to co-IP and immunoblot.