Abstract Adaptive resistance limits immune checkpoint blockade therapy (ICBT) response duration and magnitude. Interferon γ (IFNγ), a critical cytokine that promotes cellular immunity, also induces adaptive resistance to ICBT. Using syngeneic mouse tumor models, we confirmed that chronic IFNγ exposure confers resistance to anti-Programmed cell death protein 1 (α-PD-1) therapy. We identified consistent upregulation of poly-ADP ribosyl polymerase 14 (PARP14) in both chronic IFNγ-treated cancer cells and patient melanoma with elevated IFNG expression. Knockdown or pharmacological inhibition of PARP14 increased effector T cell infiltration into tumors derived from cells pre-treated with IFNγ and decreased the presence of regulatory T cells, leading to restoration of α-PD-1 sensitivity. Finally, we determined that tumors which spontaneously relapsed following α-PD-1 therapy could be re-sensitized upon receiving PARP14 inhibitor treatment, establishing PARP14 as an actionable target to reverse IFNγ-driven ICBT resistance. Citation Format: Chun Wai Wong, Christos Evangelou, Kieran N. Sefton, Rotem Leshem, Kleita Sergiou, Macarena Lucia Fernandez Carro, Erez Uzuner, Holly Mole, Brian A. Telfer, Daniel J. Wilcock, Michael P. Smith, Kaiko Kunii, Nicholas R. Perl, Paul Lorigan, Kaye J. Williams, Patricia E. Rao, Raghavendar T. Nagaraju, Mario Niepel, Adam F. Hurlstone. PARP14 inhibition restores PD-1 immune checkpoint inhibitor response following IFNγ-driven adaptive resistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 625.

Supplemental Table S1. Genome-wide RNAi toxicity screen z-scores, related to Supplemental Figure S1, Supplemental Table S2. Percent DNA copy number gain in melanoma tumor samples compared to normal skin, related to Supplemental Figure S1, Supplemental Table S3. Melanoma panel siRNA toxicity, related to Supplemental Figure S1, Supplemental Table S4. Whole genome transcript profiles, related to Supplemental Figure S1, Supplemental Table S5. GISTIC analysis and expression correlation of candidate melanoma hits, related to Supplemental Figure S1, Supplemental Table S6. SOX10 ChIPSeq, related to Supplemental Figure S3, Supplemental Table S7. SOX10 ChIPSeq motif targets, related to Supplemental Figure S3, Supplemental Table S8. SOX10 ChIPSeq motif targets hypergeometric distribution analysis, related to Supplemental Figure S3, Supplemental Table S9. Whole genome transcript arrays with siRNAs, related to Supplemental Figure S3, Supplemental Table S10. Candidate SOX10 targets, related to Supplemental Figure S3, Supplemental Table S11. Expression correlations in patient samples, related to Supplemental Figure S3, Supplemental Table S12. Elastic net prediction scores, related to Figures 1, 2 and 3, Supplemental Figures S4, S5 and S7, Supplemental Table S13. Elastic net prediction scores, related to Figure 2, Supplemental Table S14. S2N analysis of gene transcripts between TBK1i-sensitive and -resistant BRAF wild-type melanoma cell lines, related to Figure 3, Supplemental Table S15. TMT-phosphoproteomics peptide quantification, related to Figure S12.

Supplemental Figure S1. Integrative Analysis of Functional Genomics and Copy Number Variation in Melanoma Cells and Tissues, Supplemental Figure S2. Elastic Net Derived Biomarker Results for Melanoma Survival Genes and Detection of These Biomarkers in TCGA SKCM, Supplemental Figure S3. The SOX10 Regulatory Network Supporting Cell Autonomous Melanoma Cell Growth and Survival, Supplemental Figure S4. SOX10 Addiction Specifies Sensitivity of BRAF Mutant Melanomas to BRAF and MEK Inhibitors In Vitro, Related to Figure 1, Supplemental Figure S5. SOX10 Addiction Specifies Sensitivity of BRAF Mutant Melanomas to BRAF and MEK Inhibitors In Patients, Related to Figure 2, Supplemental Figure S6. Bicluster of melanoma cell lines and chemical compounds in McDermott/Benes GDSC dataset, Related to Figure 3, Supplemental Figure S7. Nomination of TBK1 as a Therapeutic Target for Drug-Resistant Melanoma, Related to Figure 3, Supplemental Figure S8. TBK1/IKKε-Addiction is Conserved In Vivo, Related to Figure 4, Supplemental Figure S9. TBK1/IKKε-Addiction Corresponds to a Cell Autonomous Innate Immune Melanoma Subtype, Related to Figure 5, Supplemental Figure S10. TBK1/IKKε Activate AKT and YAP to Support Survival of the Cell-autonomous Immune Melanoma Subtype, Related to Figure 6, Supplemental Figure S11. 13C glucose and 13C glutamine metabolism at 30m and 2h, Related to Figure 7 and Supplemental Fig. S7, Supplemental Figure S12. Distinct Epigenetic Cell Fate Programs Specify TBK1/IKKε Addiction, Related to Figure 7.

From microarray analysis, genes from VGP melanocyte neoplasia whose transcripts are significantly up- or down-regulated compared to wild-type tissue that are associated with lipid metabolism and are not differentially expressed also in BRAFV600E or V12RAS RGP samples.

Resistance mechanisms to immune checkpoint blockade therapy (ICBT) limit its response duration and magnitude. Paradoxically, Interferon γ (IFNγ), a key cytokine for cellular immunity, can promote ICBT resistance. Using syngeneic mouse tumour models, we confirm that chronic IFNγ exposure confers resistance to immunotherapy targeting PD-1 (α-PD-1) in immunocompetent female mice. We observe upregulation of poly-ADP ribosyl polymerase 14 (PARP14) in chronic IFNγ-treated cancer cell models, in patient melanoma with elevated IFNG expression, and in melanoma cell cultures from ICBT-progressing lesions characterised by elevated IFNγ signalling. Effector T cell infiltration is enhanced in tumours derived from cells pre-treated with IFNγ in immunocompetent female mice when PARP14 is pharmacologically inhibited or knocked down, while the presence of regulatory T cells is decreased, leading to restoration of α-PD-1 sensitivity. Finally, we determine that tumours which spontaneously relapse in immunocompetent female mice following α-PD-1 therapy upregulate IFNγ signalling and can also be re-sensitised upon receiving PARP14 inhibitor treatment, establishing PARP14 as an actionable target to reverse IFNγ-driven ICBT resistance.

Supplementary Figure 3. Western-blot showing the expression levels of the MMR enzymes MSH2, MLH1 and MSH6 in 12 melanoma cell lines.

Supplementary Figure 5. A, Colony formation assay on WM98.1 clones expressing an empty vector (vector) or MGMT (MGMT) treated with TMZ with or without veliparib (ABT). B, GI50s of the same WM98.1 clones treated with MMS with our without olaparib veliparib. PF50 values are also shown. D, GI50s of the 10 melanoma cell lines treated with MMS with our without olaparib veliparib. PF50 values are also shown. E, MGMT in vitro activity assays. Recombinant MGMT was incubated with increasing concentrations of olaparib, veliparib or lomeguatrib

Extended Experimental Methods

Supplementary Figure 1. A, Colony formation assay 12-well plate layout. B, MGMT promoter methylation assay.

Breast cancer remains a leading cause of mortality, predominantly due to the development of metastases to vital organs. At present, predictive biomarkers of organ specific metastasis and therapies targeted to the metastatic niche are limited. Here, to identify the molecular determinants of breast cancer metastasis we analysed patient-derived breast tumours by combining quantitative proteomics, bioinformatics, and functional assays in vitro and in vivo. We identified elevated levels of the protein Osteomodulin (OMD) associated with breast cancer bone metastases in patient-derived samples. OMD overexpression in the breast cancer MDA-MB-231 cell model significantly increases cell migration in vitro and promotes the formation of bone metastases in vivo . Phosphoproteomics analysis of MDA-MB-231 cells expressing OMD identifies active Cyclin-dependent kinase 1 (CDK1) downstream of OMD. The importance of the OMD-CDK1 axis was validated using two independent phosphoproteomics datasets analysing patient-derived breast cancer samples. We also show that the OMD-CDK1 axis drives cell migration and cell viability in vitro and the formation of bone metastases in vivo . Finally, CDK1 inhibition reduces in vitro cell viability of an independent cohort of metastatic patient samples showing high CDK1 activity. Therefore, the OMD-CDK1 axis will determine which breast cancer patients develop bone metastases and is a therapeutic target to treat or prevent breast cancer bone metastases.

Supplementary Figure 2. A,Effect 0.25μM olaparib on the proliferation of 12 melanoma cell lines. B, GI50s for TMZ of 12 melanoma cell lines treated with or without olaparib. PF50s values are also shown. C, Colony formation of 6 MGMT+ve and 6 MGMT-ve melanoma cell lines treated with TMZ in the presence or absence of olaparib.

Supplementary Figure 4. A, Western-blot showing the expression levels of the MMR enzymes MSH2, MLH1 and MSH6 in NET cell lines. B, GI50s for glioblastoma (GBM), pancreatic neuroendocrine (NET) and colorectal (CRC) cancer cell lines treated with TMZ with our without olaparib. PF50 values are also shown.

Dysregulated cellular metabolism is a cancer hallmark for which few druggable oncoprotein targets have been identified. Increased fatty acid (FA) acquisition allows cancer cells to meet their heightened membrane biogenesis, bioenergy, and signaling needs. Excess FAs are toxic to non-transformed cells but surprisingly not to cancer cells. Molecules underlying this cancer adaptation may provide alternative drug targets. Here, we demonstrate that diacylglycerol O-acyltransferase 1 (DGAT1), an enzyme integral to triacylglyceride synthesis and lipid droplet formation, is frequently up-regulated in melanoma, allowing melanoma cells to tolerate excess FA. DGAT1 over-expression alone transforms p53-mutant zebrafish melanocytes and co-operates with oncogenic BRAF or NRAS for more rapid melanoma formation. Antagonism of DGAT1 induces oxidative stress in melanoma cells, which adapt by up-regulating cellular reactive oxygen species defenses. We show that inhibiting both DGAT1 and superoxide dismutase 1 profoundly suppress tumor growth through eliciting intolerable oxidative stress.

Background Immune system response to cancer therapies can indicate whether a patient will have positive outcomes following therapy. Understanding how the tumor microenvironment (TME) evolves during tumorigenesis and therapeutic response is crucial to developing personalized treatments with the goal of improving cancer therapy. With robust and comprehensive multiplexed imaging technologies, immune biomarkers can be used to interrogate myeloid and lymphoid cell lineages and structures, and when combined with specific oncology biomarkers, can capture the immune response within the TME in a variety of neoplasms. The availability of cell type specific biomarkers, combined with the ability to interrogate using multiplexed tissue imaging, provides unprecedented and novel insights into immune cell populations and spatial cell interactions with many cell types in the TME. Cell DIVETM Multiplex Imaging Solution allows probing and imaging of dozens of biomarkers on a whole single tissue section with an iterative staining and dye inactivation workflow. At its core, Cell DIVE is a precise and adaptable open multiplexing solution that enables flexibility in antibody selection of biomarker panels used in a multiplexed imaging study. Cell Signaling Technology (CST) has a broad portfolio of IHC-validated antibodies to detect key proteins in the TME, enabling immune cell detection and phenotyping in tissue. CST offers off the shelf (OTS), ready-to-ship antibody conjugates that have been verified to work on Cell DIVE and offers custom conjugation of IHC-validated antibodies to fluorophores and other detection reagents. CST employs a rigorous approach to IHC validation, followed by verification on the Cell DIVE platform to ensure successful detection of proteins. Here, we demonstrate multiplexed Cell DIVE imaging using a novel panel of dozens of CST biomarkers across multiple tissue types. Methods Sections were stained using conjugated antibodies (Cell Signaling Technology) to various biomarkers in 4 channels plus DAPI and imaged using Cell DIVE. Multiple rounds of staining and imaging were accomplished using the Cell DIVE workflow (Leica Microsystems). Results Development of the multiplexed panel required minimal optimization, enabled the identification of complex cell types and revealed their cell-to-cell interactions within the tumor microenvironment. Conclusions Multiplexed whole slide imaging allows deep analysis of immune cell lineages and provided new insights into immune and tumor cell-to-cell interactions within the tumor microenvironment. Ethics Approval Stained samples were commercially available.

Receptor Tyrosine Kinase (RTK) endocytosis-dependent signalling drives cell proliferation and motility during development and adult homeostasis, but is dysregulated in diseases, including cancer. The recruitment of RTK signalling partners during endocytosis, specifically during recycling to the plasma membrane, is still unknown. Focusing on Fibroblast Growth Factor Receptor 2b (FGFR2b) recycling, we reveal FGFR signalling partners proximal to recycling endosomes by developing a Spatially Resolved Phosphoproteomics (SRP) approach based on APEX2-driven biotinylation followed by phosphorylated peptides enrichment. Combining this with traditional phosphoproteomics, bioinformatics, and targeted assays, we uncover that FGFR2b stimulated by its recycling ligand FGF10 activates mTOR-dependent signalling and ULK1 at the recycling endosomes, leading to autophagy suppression and cell survival. This adds to the growing importance of RTK recycling in orchestrating cell fate and suggests a therapeutically targetable vulnerability in ligand-responsive cancer cells. Integrating SRP with other systems biology approaches provides a powerful tool to spatially resolve cellular signalling.

Cells contain intracellular compartments, including membrane-bound organelles and the nucleus, and are surrounded by a plasma membrane. Proteins are localised to one or more of these cellular compartments; the correct localisation of proteins is crucial for their correct processing and function. Moreover, proteins and the cellular processes they partake in are regulated by relocalisation in response to various cellular stimuli. High-throughput 'omics experiments result in a list of proteins or genes of interest; one way in which their functional role can be understood is through the knowledge of their subcellular localisation, as deduced through statistical enrichment for Gene Ontology Cellular Component (GOCC) annotations or similar. We have designed a bioinformatics tool, named SubcellulaRVis, that compellingly visualises the results of GOCC enrichment for quick interpretation of the localisation of a group of proteins (rather than single proteins). We demonstrate that SubcellulaRVis precisely describes the subcellular localisation of gene lists whose locations have been previously ascertained. SubcellulaRVis can be accessed via the web (http://phenome.manchester.ac.uk/subcellular/) or as a stand-alone app (https://github.com/JoWatson2011/subcellularvis). SubcellulaRVis will be useful for experimental biologists with limited bioinformatics expertise who want to analyse data related to protein (re)localisation and location-specific modules within the intracellular protein network.

Cutaneous melanoma is one of the most aggressive human malignancies and shows increasing incidence. Mast cells (MCs), long-lived tissue-resident cells that are particularly abundant in human skin where they regulate both innate and adaptive immunity, are associated with melanoma stroma (MAMCs). Thus, MAMCs could impact melanoma development, progression, and metastasis by secreting proteases, pro-angiogenic factors, and both pro-inflammatory and immuno-inhibitory mediators. To interrogate the as-yet poorly characterized role of human MAMCs, we have purified MCs from melanoma skin biopsies and performed RNA-seq analysis. Here, we demonstrate that MAMCs display a unique transcriptome signature defined by the downregulation of the FcεRI signaling pathway, a distinct expression pattern of proteases and pro-angiogenic factors, and a profound upregulation of complement component C3. Furthermore, in melanoma tissue, we observe a significantly increased number of C3+ MCs in stage IV melanoma. Moreover, in patients, C3 expression significantly correlates with the MC-specific marker TPSAB1, and the high expression of both markers is linked with poorer melanoma survival. In vitro, we show that melanoma cell supernatants and tumor microenvironment (TME) mediators such as TGF-β, IL-33, and IL-1β induce some of the changes found in MAMCs and significantly modulate C3 expression and activity in MCs. Taken together, these data suggest that melanoma-secreted cytokines such as TGF-β and IL-1β contribute to the melanoma microenvironment by upregulating C3 expression in MAMCs, thus inducing an MC phenotype switch that negatively impacts melanoma prognosis.

High-throughput ‘omics methods result in lists of differentially regulated or expressed genes or proteins, whose function is generally studied through statistical methods such as enrichment analyses. One aspect of protein regulation is subcellular localization, which is crucial for their correct processing and function and can change in response to various cellular stimuli. Enrichment of proteins for subcellular compartments is often based on Gene Ontology Cellular Compartment annotations. Results of enrichment are typically visualized using bar-charts, however enrichment analyses can result in a long list of significant annotations which are highly specific, preventing researchers from gaining a broad understanding of the subcellular compartments their proteins of interest may be located in. Schematic visualization of known subcellular locations has become increasingly available for single proteins via the UniProt and COMPARTMENTS platforms. However, it is not currently available for a list of proteins (e.g. from the same experiment) or for visualizing the results of enrichment analyses. To generate an easy-to-interpret visualization of protein subcellular localization after enrichment we developed the SubcellulaRVis web app, which visualizes the enrichment of subcellular locations of gene lists in an easy and impactful manner. SubcellulaRVis projects the results of enrichment analysis on a graphical representation of a eukaryotic cell. Implemented as a web app and an R package, this tool is user-friendly, provides exportable results in different formats, and can be used for gene lists derived from multiple organisms. Here, we show the power of SubcellulaRVis to assign proteins to the correct subcellular compartment using gene list enriched in previously published spatial proteomics datasets. We envision SubcellulaRVis will be useful for cell biologists with limited bioinformatics expertise wanting to perform precise and quick enrichment analysis and immediate visualization of gene lists. Author Summary Cells contain intracellular compartments, such as membrane-bound organelles and the nucleus, and are surrounded by a plasma membrane. Proteins can be found in different cellular compartments; depending on the subcellular compartment they are localized to, they can be differentially regulated or perform location-specific functions. High-throughput ‘omics experiments result in a list of proteins or genes of interest; one way in which their functional role can be understood is through their subcellular localization, as deduced through statistical enrichment for Gene Ontology Cellular Component (GOCC) annotations or similar. We have designed a bioinformatic tool, named SubcellulaRVis, that compellingly visualizes the results of protein localization after enrichment and simplifies the results for a quick interpretation. We demonstrate that SubcellulaRVis precisely describes the subcellular localization of gene lists whose locations have been previously ascertained in publications. SubcellulaRVis can be accessed via the web or as a stand-alone app. SubcellulaRVis will be useful for experimental biologists with limited bioinformatics expertise analyzing data related to cellular signalling, protein (re)localization and regulation, and location-specific functional modules within the cell.

Integration of signalling downstream of individual receptor tyrosine kinases (RTKs) is crucial to fine tune cellular homeostasis during development and in pathological conditions, including breast cancer. However, how signalling integration is regulated and whether the endocytic fate of single receptors controls such signalling integration remains poorly elucidated. Combining quantitative phosphoproteomics and targeted assays, we generated a detailed picture of recycling-dependent fibroblast growth factor (FGF) signalling in breast cancer cells, with a focus on distinct FGF receptors (FGFRs). We discovered reciprocal priming between FGFRs and epidermal growth factor (EGF) receptor (EGFR) that is coordinated at recycling endosomes. FGFR recycling ligands induce EGFR phosphorylation on threonine 693. This phosphorylation event alters both FGFR and EGFR trafficking and primes FGFR-mediated proliferation but not cell invasion. In turn, FGFR signalling primes EGF-mediated outputs via EGFR threonine 693 phosphorylation. This reciprocal priming between distinct families of RTKs from recycling endosomes exemplifies a novel signalling integration hub where recycling endosomes orchestrate cellular behaviour. Therefore, targeting reciprocal priming over individual receptors may improve personalized therapies in breast and other cancers.

Increasing evidence indicates that success of targeted therapies in the treatment of cancer is context-dependent and is influenced by a complex crosstalk between signaling pathways and between cell types in the tumor. The Fibroblast Growth Factor (FGF)/FGF receptor (FGFR) signaling axis highlights the importance of such context-dependent signaling in cancer. Aberrant FGFR signaling has been characterized in almost all cancer types, most commonly non-small cell lung cancer (NSCLC), breast cancer, glioblastoma, prostate cancer and gastrointestinal cancer. This occurs primarily through amplification and over-expression of FGFR1 and FGFR2 resulting in ligand-independent activation. Mutations and translocations of FGFR1-4 are also identified in cancer. Canonical FGF-FGFR signaling is tightly regulated by ligand-receptor combinations as well as direct interactions with the FGFR coreceptors heparan sulfate proteoglycans (HSPGs) and Klotho. Noncanonical FGFR signaling partners have been implicated in differential regulation of FGFR signaling. FGFR directly interacts with cell adhesion molecules (CAMs) and extracellular matrix (ECM) proteins, contributing to invasive and migratory properties of cancer cells, whereas interactions with other receptor tyrosine kinases (RTKs) regulate angiogenic, resistance to therapy, and metastatic potential of cancer cells. The diversity in FGFR signaling partners supports a role for FGFR signaling in cancer, independent of genetic aberration.