Combinatorial targeting of Hippo-STRIPAK and PARP elicits synthetic lethality in gastrointestinal cancers

The striatin-interacting phosphatase and kinase (STRIPAK) complexes integrate extracellular stimuli that result in intracellular activities. Previously, we discovered that STRIPAK is a key machinery responsible for loss of the Hippo tumor suppressor signal in cancer. Here, we identified the Hippo-STRIPAK complex as an essential player in the control of DNA double-stranded break (DSB) repair and genomic stability. Specifically, we found that the mammalian STE20-like protein kinases 1 and 2 (MST1/2), independent of classical Hippo signaling, directly phosphorylated zinc finger MYND type–containing 8 (ZMYND8) and hence resulted in the suppression of DNA repair in the nucleus. In response to genotoxic stress, the cyclic GMP-AMP synthase/stimulator of IFN genes (cGAS/STING) pathway was determined to relay nuclear DNA damage signals to the dynamic assembly of Hippo-STRIPAK via TANK-binding kinase 1–induced (TBK1-induced) structural stabilization of the suppressor of IKBKE 1– sarcolemma membrane–associated protein (SIKE1-SLMAP) arm. As such, we found that STRIPAK-mediated MST1/2 inactivation increased the DSB repair capacity of cancer cells and endowed these cells with resistance to radio- and chemotherapy and poly(ADP-ribose)polymerase (PARP) inhibition. Importantly, targeting the STRIPAK assembly with each of 3 distinct peptide inhibitors efficiently recovered the kinase activity of MST1/2 to suppress DNA repair and resensitize cancer cells to PARP inhibitors in both animal- and patient-derived tumor models. Overall, our findings not only uncover what we believe to be a previously unrecognized role for STRIPAK in modulating DSB repair but also provide translational implications of cotargeting STRIPAK and PARP for a new type of synthetic lethality anticancer therapy.

. Hippo-containing STRIPAK complex is essential for DSB repair and genomic stability  Tables   Table S1. List of siRNA sequence used in this manuscript
At 48 h after transfection, the supernatants were filtered (0.45 µm) and were applied to recipient cells in the presence of 8 µg/ml polybrene (H9268, Sigma Aldrich).

CRISPR-Cas9 approach for the generation of KO cells
All of the KO cell lines used in this study were generated using the CRISPR-Cas9 gene- selected by applying 1 µg/ml puromycin for one week before validation was performed by carrying out western blotting using indicated antibodies. secondary antibodies were used for the immunoprecipitation assays, with these secondary antibodies including mouse anti-rabbit IgG LCS (A25022, 1:5000) and goat anti-mouse IgG LCS (A25012, 1:5000), which were obtained from Abbkine.

DSB reporter assay
293A cells were electroporated with the I-SceI expression construct (pCBASce) together with GR-GFP or EJ5-GFP reporter plasmid at 150 V, 975 microfarads using an NEPA21 Super Electroporator (NEPA GENE, Japan). Cells were further recovered for 48 h after electroporation and were subjected to flow cytometric analysis using a BD FACSCantoII analyzer.

Cell viability assay
For cell death and viability assays, cells were seeded into 96-well plates at a density of 1000 per well overnight to have them become attached to the wells, and then treated with siRNAs, PARPi or PDPA-delivered peptides at indicated concentrations for 48 h.
The PDPA micelles were incubated with peptides at 4℃ for at least 1 h. The final concentrations of PDPA and peptides were 500 mg/ml and 100 mg/ml, respectively.
The cells were incubated for the indicated length of time, and then used with an ATPbased CellTiter-LumiTM Plus kit (Beyotime) according to the manufacturer's instructions. The intracellular ATP contents were measured using a BioTek Synergy TM NEO multi-detector microplate reader (Thermo). Cell viability was calculated using the -1×100.

Neutral comet assay
Briefly, for each experiment, cells were subjected to 10 Gy of IR and collected at either 0.5 h or 4.0 h post IR. Cells were harvested and rinsed twice with ice-cold PBS, and about 1 x 10 5 cells/ml were mixed with molten LMAgarose (4250-050-02, Enzo) at a ratio of 1:10 (v/v) and immediately a volume of 75 μL of the mixture was pipetted onto a Comet Slide apparatus (4250-050-03, Enzo). Each slide with the mixture was put into a 4°C refrigerator for 30 minutes and then treated with neutral lysis buffer (2.5 M NaCl, 100 mM Na2EDTA, 1% sodium lauroyl sarcosine, 10 mM Tris, pH 7.5, 1% Triton-X100 and 10% DMSO) overnight. Next, each treated slide was subjected to electrophoresis at 25 V for 30 minutes and stained in CYGREEN® Nucleic Acid Dye agent (ENZ-GEN105-0020, Enzo) for 30 minutes. Images were captured using a fluorescence microscope (Olympus BX51) and analyzed with CometScore software. For mice radio-sensitivity assay, WT and Stk3 -/mice were exposed to 6Gy wholebody irradiation and were breed with normal diet. Two mice from each group were sacrificed for immunofluorescence assay and western blotting analysis to examine the residual γ-H2AX levels. The rest of mice were continued to culture for survival analysis and eventually sacrificed at 6 weeks post irradiation.
For the tumor formation assay, mice were first injected with 5 × 10 6 BGC-823 cells/per mouse into their flanks to induce tumor formation. Once the tumor volume reached about 100 mm 3 , mice were injected intratumorally with indicted inhibitors for five successive days. Tumor length and width were measured every 3 days to calculate the tumor volume (= width 2 x length x 0.523). Two weeks later, the mice were euthanized and then their tumors were harvested and weighed.

Immunohistochemistry (IHC) staining
All clinical samples used in this study were derived from patients approved by the

Statistical analysis
The presented data were calculated as mean ± SEM from three independent experiments unless otherwise indicated and were analyzed with GraphPad Prism 8.0 statistical software. Continuous data were compared using Student's t tests (comparing two variables) or one-way ANOVA analysis with Dunnett's post hoc test (comparing multiple variables). For correlation, the Spearman rank correlation was used for continuous variables. Survival curves were calculated according to the Kaplan-Meier method; survival analysis was performed using the log-rank test. p < 0.05 was considered statistically significant.