FDR for synergy score paralog combinations. Table of FDR corresponding to synergy scores calculated using GEMINI.

P value for synergy score paralog combinations. Table of P value corresponding to synergy scores calculated using GEMINI.

A-K, Fragmentation spectrum supporting peptide identity and phosphosites on YAP, TAZ, and NF2. Precursor ion chromatogram (XIC) and corresponding ions are provided in supplementary materials.

A,B, CRISPR screening results from 22 cancer cell lines. A, Abundance fold-change of positive controls (dgRNAs targeting essential genes n=28 paired with control) and negative controls (dgRNAs targeting non-coding regions n=54 and nontargeting dgRNAs n=97). Data are shown as mean ± SD B, Abundance fold-change of dgRNAs targeting MARK2+3. Each dot represents a single dgRNA. Data are shown as mean ± SD n=24 dgRNAs. C, Pearson correlation analysis of CRISPR screening data comparing log2 fold-changes of single and double gene knockout with published and replicate experiments. Functional domain class libraries are indicated. Data are shown as mean R2 ± SD n=2-8. D, CRISPR screening data from indicated Cas9-expressing cancer cell lines after lentiviral knockout of indicated genes (abundance was read out using next-generation sequencing). Heatmap color indicates the log2(fold-change) of normalized dgRNA abundance (normalized to doubling time 0). E, Competition-based fitness assays in Cas9-expressing cancer cells after lentiviral knockout of indicated genes with independent dgRNAs (expression of dgRNAs was linked with GFP) (Data shown are an extension of Fig. 1D). Data shown are the mean ± SD of %GFP+ (normalized to day 3 after infection). n=3. F, Western blot analysis of Cas9+ K-562 cells. G, Analysis of apoptosis assay using Annexin-V and DAPI in Cas9+ MDA-MB231 cells. Indicated genes were knocked out using lentiviral dgRNAs linked to GFP. Data are shown as mean ± SD. n=3-6. P value was calculated on change in viability compared to control with one-way ANOVA and Dunnett’s correction. H, EdU incorporation assays following indicated gene knock out using lentiviral dgRNAs linked to GFP in Cas9+ indicated cells. Data are shown as mean ± SD. n=3. P value was calculated on change in S-phase population to control with one-way ANOVA and Dunnett’s correction. I, Crystal violet stain of indicated cells following lentiviral knockout of indicated genes. Data shown are representative of three independent experiments and an extension of Fig. 1G.

YAP:TEAD4 bound YAP/TAZ dependent enhancers. Peak location file for enhancers with YAP,TEAD4 binding, and reduction of signal upon YAP/TAZ knockout.

OBJECTIVE:To investigate the feasibility of using implantable microdevices (IMDs) in pancreatic ductal adenocarcinoma (PDAC). BACKGROUND:Pancreatic ductal adenocarcinoma (PDAC) remains a highly lethal malignancy with limited treatment options. IMDs permit localized delivery of multiple therapies with simultaneous in vivo assessment of tumor response. Although IMDs have previously been evaluated in multiple malignancies, their application in PDAC has not been reported. METHODS:Xenograft tumors were generated by injecting PDAC patient-derived organoids (PDO) into mouse pancreata. IMDs loaded with standard-of-care chemotherapeutic drugs including gemcitabine, paclitaxel, SN38 (irinotecan metabolite), oxaliplatin, and 5-fluorouracil were then inserted into the tumor during laparotomy. Mice were sacrificed at 4 or 24 hours after insertion. Immunofluorescence was performed to assess biomarkers of DNA damage (γH2AX), apoptosis (cleaved caspase-3; CC3), and proliferation (Ki67). RESULTS:Fourteen mice underwent IMD placement with successful device retrieval and analysis. CC3 and γH2AX analyses revealed treatment specific biologic response, with the most pronounced response at 24 hours. Ki67 analysis demonstrated reduction of the proliferation within treated regions at both 4 and 24 hours compared to controls. Overall, the IMD enabled localized chemotherapy delivery and concurrent assessment of responses to multiple chemotherapeutic agents in an in vivo setting. CONCLUSION:Intra-pancreatic IMD deployment in orthotopic PDAC organoid tumors presents a feasible approach to assess chemotherapy sensitivities. Changes in proliferation revealed drug specific response at all timepoints, while apoptotic markers required longer incubation. These data support further investigation into the intra-operative or endoscopic deployment of IMDs as a platform for precision medicine.

Mass spectrometric identification and quantification of MARK2 substrate peptides This document contains mass spectrometric data supporting phosphorylation of specific phosphorylation sites on CDC25C, YAP, TAZ, and NF2. For each considered site, the fragmentation spectrum supporting peptide identity and phosphorylation localization, and the extracted precursor ion chromatogram (XIC) of the corresponding ions in samples treated with MARK2 and untreated (used for label-free quantification) are provided. All spectral intensity values are without normalization. For each phosphorylation site considered, both the phosphorylated and unmodified peptides are reported. To control for unspecific MARK2 activity, a peptide containing S and/or T residues and not detected as phosphorylated is considered for each protein. For these peptides, XIC for the theoretical m/z of the unmodified and phosphorylated peptides is reported to confirm that missed identification by MS2 was not caused by precursor under-sampling. To control for differences in the abundance of MARK2 substrate proteins, a peptide containing no S and T residues was reported, when available.

Imaging reporter genes are essential tools in cancer research for monitoring tumor evolution, cell dissemination, gene activation and treatment response. However, germline reporter models typically rely on single imaging modalities operating over limited spatial scales. To address these limitations, we developed an inducible triple-reporter mouse model (Rosa26LSL-NRL) that integrates three reporters for complementary imaging modalities, fluorescence, bioluminescence and positron emission tomography (PET), along with inducible Cre-lox functionality for precise spatiotemporal control of reporter expression in genetically engineered mouse models. Using a multiscale, multimodal approach, we visualize deep tissue oncogenesis in models of hepatocellular carcinoma and lung adenocarcinoma and, guided by whole-body imaging with bioluminescence and [18F]tetrafluoroborate PET/magnetic resonance imaging (MRI), resolve cell-cell interactions within tumor microenvironments using in situ microscopy. This triple-reporter system enables multiscale investigation of biological processes within whole animals, facilitating sensitive, tissue-specific, in vivo cell tracking from whole-body to cellular resolution.

Paralog CRISPR screening data. Table for sgRNA sequences and raw sequencing counts of paralog CRISPR screens in 22 cancer cell lines.

Phosphosite identification using mass spectrometry. Combined table of peptides identified using mass spectrometry.

A, Growth kinetics of subcutaneous YAPC xenografts implanted in immunodeficient mice. Indicated genes were knocked out just before injection. Data are shown as mean ± SEM n=5 per group. P values are calculated using a mixed effects model (considering the interaction of experimental groups over time) compared to Ctrl group and corrected with Bonferroni-Holm (BH). B, Tumor imaging at the end-point of the xenograft experiments shown in A. C, mRNA expression of MARK1-4 in different tissues. Data from Genotype-Tissue Expression (GTEx) database. D, Competition-based fitness assays for Ctrl (dgRNA targeting hROSA26 locus) and knockout of essential gene CDK1 corresponding to experiments shown in Fig. 1H and J. Data shown are the mean ± SD of %GFP+ (normalized to day 3 after infection). n=3. E, Western blot analysis of MARK1-4 overexpression in YAPC cells. F, Rescue experiment in YAPC cells using lentiviral expression of MARK1, MARK4 cDNAs or empty vector control (Ctrl). Data shown are the mean ± SD of %GFP+ (normalized to day 3 after infection). n=3. P values are calculated using a mixed effects model (considering the interaction of experimental groups over time) compared to Ctrl group and corrected with Bonferroni-Holm (BH). G¬¬, Rescue experiment in Cas9+ YAPC cells using lentiviral overexpression cDNA of CRISPR resistant (CR) analog sensitive mutant MARK2M129G, kinase-dead mutant MARK2K82H or empty vector control (Ctrl). Data shown are the mean ± SD of %GFP+ (normalized to day 3 after infection). n=3. P values are calculated using a mixed effects model (considering the interaction of experimental groups over time) compared to Ctrl group and corrected with Bonferroni-Holm (BH).

A, Western blot analysis of YAP localization following doxycycline (Dox) induced empty vector control, MKIWT or MKIMUT expression for 24h in YAPC cells. Nuclear (Nuc) and cytosolic (Cyto) fractionation are indicated. (Data shown are an extension of Fig. 6E). B, Western blot analysis in YAPC cells. C, Gene set enrichment analysis (GSEA) of RNA-seq data from MKIWT compared to MKIMUT expressing MDA-MB231 cells. Normalized enrichment score (NES) and P value are shown. D, E, Western blot analysis of MARK2,3 specific substrates phosphorylation following MKI/MARK2,3 analog sensitive mutant co-overexpression in HEK-293T cells. F, Competition-based fitness assays in dCas9VPR-expressing YAPC cells after lentiviral expression of sgRNA targeting promoters of indicated genes and expression of MKI (expression of MKI was linked to GFP). Data are shown as mean ± SD of normalized %GFP+ (to day 3 after infection). n=3-9. G, qPCR of MARK2 mRNA following CRISPR activation in dCas9VPR expressing YAPC cells. Housekeeping gene and Ctrl normalized expression is shown. n=3. P value was calculated using a two-tailed parametric t-test with Welch’s correction. H, Tumor volume quantification using NIS-SPECT imaging at indicated time points. n=10 per group. P value was calculated using a two-tailed parametric t-test with Welch’s correction. I, Growth kinetics of subcutaneous YAPC xenografts implanted in immunodeficient mice. Expression of MKIWT from doxycycline (Dox)-inducible lentiviral construct was induced on day 10 post-injection of the cells. Data are shown as mean ± SD. n=5 per group. P values are calculated using a mixed effects model (considering the interaction of experimental groups over time) compared to Ctrl group (-Dox) and corrected with Bonferroni-Holm (BH). J, Tumor image at the end-point of the xenograft experiments shown in I.

A, IP–western blot analysis evaluating the interaction between 14-3-3e and YAPWT and YAP5D mutants or TAZWT and TAZ4D mutants in HEK-293T cells. Aspartate mutant reversion to wild-type S/T is indicated. Data are representative of two independent experiments. B, mRNA expression log2 fold-change of YAP/TAZ target genes (n=78 genes significantly downregulated upon YAP/TAZdKO and upregulated upon YAPWT expression) following YAP mutant overexpression compared to Ctrl in RH-30 cells. n=2 C, Rescue experiment of MARK2+3dKO following lentiviral YAP mutant overexpression in RH-30 cells. Data shown are the mean ± SD of %GFP+ (normalized to day 3 after infection). n=3. P values are calculated using a mixed effects model (considering the interaction of experimental groups over time) compared to Ctrl group and corrected with Bonferroni-Holm (BH). Data are an extension of Figure 4E. D, Western blot analysis in RH-30 cells following cDNA overexpression of YAP mutants. E,F, Representative immunofluorescence (IF) images of HA-YAP or HA-TAZ mutants, DAPI, and alpha-tubulin following lentiviral overexpression of indicated mutants in HEK-293T cells. scale bar: 30µm.

YAP/TAZ signature. Table of genes used as YAP/TAZ signature.

A, Mutational two-class comparison between 19 MARK2/3-dependent and 12 MARK2/3-independent human cancer cell lines (shown in B). Mutation data were obtained from the CCLE database. The effect size and P values were calculated for every mutated gene of the cell lines using a linear association model. B, C, Competition-based fitness assays in Cas9-expressing cancer cells after lentiviral knockout of indicated genes (expression of double guide RNAs (dgRNA) was linked to GFP). Heatmap color indicates the log2(fold-change) of normalized GFP (%GFP+ normalized to day 3 or 6 after infection). n=3. D, E, Western blot analysis in Cas9+ YAPC cells. F, Crystal violet stain of YAPC and CHL-1 (MARK2/3 independent) cells following dgRNA assisted lentiviral knockout of indicated genes. Data are representative of three independent experiments. G, CUT&RUN density profile of YAP/TAZ sensitive H3K27ac marked enhancer loci (n=7,896) following YAP+TAZdKO. Profiles shown are an average of 50bp bins around the summit of the enhancers. H, I, Occupancy profiles of public Chromatin immunoprecipitation sequencing (ChIP-seq) (TEAD4, YAP) (GSE66083) and CUT &RUN (H3K27ac) upon indicated gene knockout at YAP/TAZ target gene loci. (Three different dgRNAs for MARK2+3) (Data shown are an extension of Fig. 2I).

Paralog CRISPR screening data 2. Table of calculated gene level log2(fold-changes) of Paralog

Cancer cell line model information. Table of mutations and cancer driver of cell lines used in this study.

RNA-seq results. Combined tables for DESeq2 output of Ctrl, MARK2/3, YAP/TAZ double knockout in 20 cancer cell lines.

A-E, Western blot analysis of MARK2 specific substrates phosphorylation. Labeling as described in Fig 3D. Data are representative of two independent experiments. F, Lolli-pop illustration of MARK2-dependent phosphorylation sites on MST1/2 and MAP4K1-4,6,7 identified using mass spectrometry-based phosphoproteomics. SARAH= Sav/Rassf/Hpo domain, CNH= Citron homology domain. G, IP–western blot analysis evaluating the phosphorylation p-LATS2 (T1041) in the presence or absence of MARK2 or MARK3 following NF2 overexpression in HEK-293T cells. Data are representative of two independent experiments. H, IP–western blot analysis evaluating the phosphorylation p-LATS2 (T1041) after NF2 mutant overexpression in HEK-293T cells. Data are representative of two independent experiments. I, IP–western blot analysis evaluating p-LATS1 (T1079) in the presence of NF2 together with MARK2 overexpression in HEK-293T cells. Data are representative of two independent experiments. J, IP–western blot analysis evaluating the interaction of NF2 and MAP4K4,6,7 in the presence or absence of MARK2 overexpression in HEK-293T cells. Data are representative of two independent experiments. K, IP–western blot analysis evaluating the phosphorylation p-LATS1 (T1079) following indicated gene overexpression in HEK-293T cells. L, IP–western blot analysis evaluating the phosphorylation p-LATS1 (T1079) following indicated gene overexpression in HEK-293T cells. Data are representative of two independent experiments. M, IP–western blot analysis evaluating p-LATS1 (T1079) in the presence of MAP4K4 or MAP4K6 together with MARK2,3, kinase-dead MARK2K82H or empty vector control overexpression in HEK-293T cells. Data are representative of two independent experiments. N, O, IP–western blot analysis evaluating p-LATS1 (T1079) in the presence of MAP4K4 or MST1 mutants. Data are representative of two independent experiments. P, Coomassie stain of recombinant proteins used in in vitro kinase assays (Fig.4D, 3E), purified from bacteria (GST-YAP, GST-TAZ) and insect cells. Q, R, In vitro kinase assay using recombinant full-length GST-YAP (P) or synthetically synthesized S128-phosphorylated peptide (Q) together with recombinant LATS2 followed by Mass spectrometry analysis of double phosphorylated YAP peptides. F was created with BioRender.com.

Colorectal cancer remains a major cause of cancer mortality, and most microsatellite stable tumors derive little benefit from immune checkpoint blockade. Here, we identify a microbiome-dependent mechanism that converts immune-refractory colorectal cancer into a more immunologically responsive state. Using orthotopic mouse models spanning distinct genetic and immunologic contexts, we show that a Helicobacter-containing microbiome suppresses primary tumor growth and limits metastasis. This protective state is associated with increased intratumoral lymphocyte infiltration and stronger effector programs. Mechanistically, microbial exposure induces MHC class II expression in colon cancer cells to promote anti-tumor immunity. Tumor-intrinsic loss of CIITA abrogates microbial protection, whereas enforced CIITA expression is sufficient to increase intratumoral T cell accumulation, restrict progression and metastasis, and sensitize microsatellite-stable tumors to PD-1 and CTLA-4 blockade. In human microsatellite-stable patient-derived organoids, increased cancer-cell MHC-II enhanced interactions with autologous immune cells and increased tumor cell apoptosis. Together, these findings define a microbiome-cancer cell antigen presentation axis that restrains metastasis and overcomes immunotherapy resistance in colorectal cancer.