Debayan Datta

Supplementary Tables Supplementary Table S1. Clinical and Tumoral Data from CCHD-PPGL patients. Supplementary Table S2. Selection inference in the PPGL (TCGA, non-CCHD) and CCHD-PPGL cohort. Supplementary Table S3. Cancer Cell Fraction of EPAS1 mutations in CCHD-PPGL and PPGL patients. Supplementary Table S4. Clonal Clustering of CCHD-PPGL tumor mutations, highlighting EPAS1 mutation, using Conipher. Supplementary Table S5. Clonal Clustering of PPGL tumor mutations (TCGA), highlighting EPAS1 mutation, using Conipher. Supplementary Table S6. Parallelism of EPAS1 genetic variant selection in high-altitude populations and human tumors. Supplementary Table S7. EPAS1 Frequencies of Denisovan AGGAA haplotype from 4,000 Han Chinese Population (https://www.biosino.org/pgghan2/index) compared with gnomAD cohort around the world. Supplementary Table S8. Genetic Instability Characteristics (MSI, CNA, TMB) from CCHD-PPGL and PPGL (TCGA) tumors. Supplementary Table S9. Data of 354 differentially expressed mitochondrial genes (from Human Mitocarta Database) between EPAS1MUT and EPAS1WT samples in patients with PPGL (TCGA). Supplementary Table S10. Gene expression profiling analysis from patients with hemoglobin disorders and 52 paragangliomas with various genotypes including 19 EPAS1 (including one tumor in duplicate), 6 SDHx, 2 SLC25A11, 6 VHL, 3 RET, 3 NF1, 1 MAX, 1 MAML3 and 11 unmutated paragangliomas (Mancini M et al. Br J Haematol, 2023). Supplementary Table S11. RNAseq Raw Counts from PC12 Cell line cultured under normoxia/hypoxia conditions. Supplementary Table S12. Tumor type and hypoxia onset in patients with CCHD-PPGL. Supplementary Table S13. O2 Saturation in CCHD patients with or without PPGL tumors. Supplementary Table S14. Heart surgery influence in tumor development in patients with CCHD-PPGL. Supplementary Table S15. Clinical features of tumors developed under hypoxia (CCHD-PPGL) and normoxia (non-CCHD-PPGL). Supplementary Table S16. Representation of clinical variables in patients with PPGL and CCHD-PPGL.

Supplementary Figure S1. Description of the CCHD-PPGL cohort of 27 individuals with CCHD who developed 38 PPGL tumors. Supplementary Figure S2. Summary of the clinical features of patients with CCHD-PPGL. Supplementary Figure S3. EPAS1 mutations are enriched in the hypoxic cohort of patients with CCHD-PPGL (88.8%) compared to the normoxic cohort of patients without CCHD-PPGL (4.5%). Supplementary Figure S4. EPAS1 clonality in PPGL tumor samples from TCGA(1–4) (patients without CCHD with PPGL in normoxic conditions). Supplementary Figure S5. No EPAS1 mutations were detected in carotid body PGL tumors. Supplementary Figure S6. Hierarchical cluster analysis of RNA sequencing data from EPAS1-mutated CCHD-PPGL tumors. Supplementary Figure S7. Convergent Evolution in Natural Populations and Human Tumors. Supplementary Figure S8. Evolutionary speed of EPAS1 selection in different cohorts. Supplementary Figure S9. Hematoxylin and eosin staining and immunohistochemistry for Chromogranin A (neuroendocrine tumors maker) and for mismatch repair (MMR) proteins in PPGL tumors. Supplementary Figure S10. Acquisition of EPAS1 mutation and subsequent selection in sympathetic PPGL tumors developed under systemic hypoxia. Supplementary Figure S11. Assessment of apoptosis marker in pheochromocytoma-derived cells upon expression of mutant EPAS1 genes or exposure to hypoxic conditions. Supplementary Figure S12. Increase levels of fumarate metabolite in EPAS1MUT PPGL tumors and increase levels of succinate metabolite in SDHxMUT PPGL tumors. Supplementary Figure S13. Decrease of fumarate hydratase/fumarase (FH) mRNA expression and increment of SDHB mRNA expression in EPAS1MUT/VHLMUT PPGL tumors when compared to EPAS1WT/VHLWT PPGL tumors, as well as to SDHxMUT PPGL tumors. Supplementary Figure S14. Decrease of FH mRNA expression in missense EPAS1-mutated tumors (n = 8) when compared to EPAS1-wild-type tumors (n = 170). Supplementary Figure S15. Elevated fumarate and succinate levels in EPAS1-Overexpressing PPGL cells. Supplementary Figure S16. Oxygen consumption and supply balance in the mitochondrial electron transport chain as an advantageous survival and proliferative mechanism. Supplementary Figure S17. Gene-level extremely positive selection on EPAS1 gene in systemic hypoxia. Supplementary Figure S18. Time-dependent effects of systemic hypoxia on oxygen-sensing tissues. Supplementary Figure S19. Specialization process of the HIF2α gene for adaptation to systemic hypoxia. Supplementary Figure S20. Proliferative advantage and increased sensitivity to HIF2α inhibition in HEK293T cells expressing the EPAS1 P405A/P531A double mutant. Supplementary Figure S21. Authentication of PPGL-derived PC12 cell line. Supplementary Figure S22. Full western blots shown in Fig. 3B. Supplementary Figure S23. Full western blots shown in Supplementary Fig S11.

Supplementary Figure S1. Description of the CCHD-PPGL cohort of 27 individuals with CCHD who developed 38 PPGL tumors. Supplementary Figure S2. Summary of the clinical features of patients with CCHD-PPGL. Supplementary Figure S3. EPAS1 mutations are enriched in the hypoxic cohort of patients with CCHD-PPGL (88.8%) compared to the normoxic cohort of patients without CCHD-PPGL (4.5%). Supplementary Figure S4. EPAS1 clonality in PPGL tumor samples from TCGA(1–4) (patients without CCHD with PPGL in normoxic conditions). Supplementary Figure S5. No EPAS1 mutations were detected in carotid body PGL tumors. Supplementary Figure S6. Hierarchical cluster analysis of RNA sequencing data from EPAS1-mutated CCHD-PPGL tumors. Supplementary Figure S7. Convergent Evolution in Natural Populations and Human Tumors. Supplementary Figure S8. Evolutionary speed of EPAS1 selection in different cohorts. Supplementary Figure S9. Hematoxylin and eosin staining and immunohistochemistry for Chromogranin A (neuroendocrine tumors maker) and for mismatch repair (MMR) proteins in PPGL tumors. Supplementary Figure S10. Acquisition of EPAS1 mutation and subsequent selection in sympathetic PPGL tumors developed under systemic hypoxia. Supplementary Figure S11. Assessment of apoptosis marker in pheochromocytoma-derived cells upon expression of mutant EPAS1 genes or exposure to hypoxic conditions. Supplementary Figure S12. Increase levels of fumarate metabolite in EPAS1MUT PPGL tumors and increase levels of succinate metabolite in SDHxMUT PPGL tumors. Supplementary Figure S13. Decrease of fumarate hydratase/fumarase (FH) mRNA expression and increment of SDHB mRNA expression in EPAS1MUT/VHLMUT PPGL tumors when compared to EPAS1WT/VHLWT PPGL tumors, as well as to SDHxMUT PPGL tumors. Supplementary Figure S14. Decrease of FH mRNA expression in missense EPAS1-mutated tumors (n = 8) when compared to EPAS1-wild-type tumors (n = 170). Supplementary Figure S15. Elevated fumarate and succinate levels in EPAS1-Overexpressing PPGL cells. Supplementary Figure S16. Oxygen consumption and supply balance in the mitochondrial electron transport chain as an advantageous survival and proliferative mechanism. Supplementary Figure S17. Gene-level extremely positive selection on EPAS1 gene in systemic hypoxia. Supplementary Figure S18. Time-dependent effects of systemic hypoxia on oxygen-sensing tissues. Supplementary Figure S19. Specialization process of the HIF2α gene for adaptation to systemic hypoxia. Supplementary Figure S20. Proliferative advantage and increased sensitivity to HIF2α inhibition in HEK293T cells expressing the EPAS1 P405A/P531A double mutant. Supplementary Figure S21. Authentication of PPGL-derived PC12 cell line. Supplementary Figure S22. Full western blots shown in Fig. 3B. Supplementary Figure S23. Full western blots shown in Supplementary Fig S11.

Supplementary Tables Supplementary Table S1. Clinical and Tumoral Data from CCHD-PPGL patients. Supplementary Table S2. Selection inference in the PPGL (TCGA, non-CCHD) and CCHD-PPGL cohort. Supplementary Table S3. Cancer Cell Fraction of EPAS1 mutations in CCHD-PPGL and PPGL patients. Supplementary Table S4. Clonal Clustering of CCHD-PPGL tumor mutations, highlighting EPAS1 mutation, using Conipher. Supplementary Table S5. Clonal Clustering of PPGL tumor mutations (TCGA), highlighting EPAS1 mutation, using Conipher. Supplementary Table S6. Parallelism of EPAS1 genetic variant selection in high-altitude populations and human tumors. Supplementary Table S7. EPAS1 Frequencies of Denisovan AGGAA haplotype from 4,000 Han Chinese Population (https://www.biosino.org/pgghan2/index) compared with gnomAD cohort around the world. Supplementary Table S8. Genetic Instability Characteristics (MSI, CNA, TMB) from CCHD-PPGL and PPGL (TCGA) tumors. Supplementary Table S9. Data of 354 differentially expressed mitochondrial genes (from Human Mitocarta Database) between EPAS1MUT and EPAS1WT samples in patients with PPGL (TCGA). Supplementary Table S10. Gene expression profiling analysis from patients with hemoglobin disorders and 52 paragangliomas with various genotypes including 19 EPAS1 (including one tumor in duplicate), 6 SDHx, 2 SLC25A11, 6 VHL, 3 RET, 3 NF1, 1 MAX, 1 MAML3 and 11 unmutated paragangliomas (Mancini M et al. Br J Haematol, 2023). Supplementary Table S11. RNAseq Raw Counts from PC12 Cell line cultured under normoxia/hypoxia conditions. Supplementary Table S12. Tumor type and hypoxia onset in patients with CCHD-PPGL. Supplementary Table S13. O2 Saturation in CCHD patients with or without PPGL tumors. Supplementary Table S14. Heart surgery influence in tumor development in patients with CCHD-PPGL. Supplementary Table S15. Clinical features of tumors developed under hypoxia (CCHD-PPGL) and normoxia (non-CCHD-PPGL). Supplementary Table S16. Representation of clinical variables in patients with PPGL and CCHD-PPGL.

Supplementary Tables Supplementary Table S1. Clinical and Tumoral Data from CCHD-PPGL patients. Supplementary Table S2. Selection inference in the PPGL (TCGA, non-CCHD) and CCHD-PPGL cohort. Supplementary Table S3. Cancer Cell Fraction of EPAS1 mutations in CCHD-PPGL and PPGL patients. Supplementary Table S4. Clonal Clustering of CCHD-PPGL tumor mutations, highlighting EPAS1 mutation, using Conipher. Supplementary Table S5. Clonal Clustering of PPGL tumor mutations (TCGA), highlighting EPAS1 mutation, using Conipher. Supplementary Table S6. Parallelism of EPAS1 genetic variant selection in high-altitude populations and human tumors. Supplementary Table S7. EPAS1 Frequencies of Denisovan AGGAA haplotype from 4,000 Han Chinese Population (https://www.biosino.org/pgghan2/index) compared with gnomAD cohort around the world. Supplementary Table S8. Genetic Instability Characteristics (MSI, CNA, TMB) from CCHD-PPGL and PPGL (TCGA) tumors. Supplementary Table S9. Data of 354 differentially expressed mitochondrial genes (from Human Mitocarta Database) between EPAS1MUT and EPAS1WT samples in patients with PPGL (TCGA). Supplementary Table S10. Gene expression profiling analysis from patients with hemoglobin disorders and 52 paragangliomas with various genotypes including 19 EPAS1 (including one tumor in duplicate), 6 SDHx, 2 SLC25A11, 6 VHL, 3 RET, 3 NF1, 1 MAX, 1 MAML3 and 11 unmutated paragangliomas (Mancini M et al. Br J Haematol, 2023). Supplementary Table S11. RNAseq Raw Counts from PC12 Cell line cultured under normoxia/hypoxia conditions. Supplementary Table S12. Tumor type and hypoxia onset in patients with CCHD-PPGL. Supplementary Table S13. O2 Saturation in CCHD patients with or without PPGL tumors. Supplementary Table S14. Heart surgery influence in tumor development in patients with CCHD-PPGL. Supplementary Table S15. Clinical features of tumors developed under hypoxia (CCHD-PPGL) and normoxia (non-CCHD-PPGL). Supplementary Table S16. Representation of clinical variables in patients with PPGL and CCHD-PPGL.

Supplementary Figure S1. Description of the CCHD-PPGL cohort of 27 individuals with CCHD who developed 38 PPGL tumors. Supplementary Figure S2. Summary of the clinical features of patients with CCHD-PPGL. Supplementary Figure S3. EPAS1 mutations are enriched in the hypoxic cohort of patients with CCHD-PPGL (88.8%) compared to the normoxic cohort of patients without CCHD-PPGL (4.5%). Supplementary Figure S4. EPAS1 clonality in PPGL tumor samples from TCGA(1–4) (patients without CCHD with PPGL in normoxic conditions). Supplementary Figure S5. No EPAS1 mutations were detected in carotid body PGL tumors. Supplementary Figure S6. Hierarchical cluster analysis of RNA sequencing data from EPAS1-mutated CCHD-PPGL tumors. Supplementary Figure S7. Convergent Evolution in Natural Populations and Human Tumors. Supplementary Figure S8. Evolutionary speed of EPAS1 selection in different cohorts. Supplementary Figure S9. Hematoxylin and eosin staining and immunohistochemistry for Chromogranin A (neuroendocrine tumors maker) and for mismatch repair (MMR) proteins in PPGL tumors. Supplementary Figure S10. Acquisition of EPAS1 mutation and subsequent selection in sympathetic PPGL tumors developed under systemic hypoxia. Supplementary Figure S11. Assessment of apoptosis marker in pheochromocytoma-derived cells upon expression of mutant EPAS1 genes or exposure to hypoxic conditions. Supplementary Figure S12. Increase levels of fumarate metabolite in EPAS1MUT PPGL tumors and increase levels of succinate metabolite in SDHxMUT PPGL tumors. Supplementary Figure S13. Decrease of fumarate hydratase/fumarase (FH) mRNA expression and increment of SDHB mRNA expression in EPAS1MUT/VHLMUT PPGL tumors when compared to EPAS1WT/VHLWT PPGL tumors, as well as to SDHxMUT PPGL tumors. Supplementary Figure S14. Decrease of FH mRNA expression in missense EPAS1-mutated tumors (n = 8) when compared to EPAS1-wild-type tumors (n = 170). Supplementary Figure S15. Elevated fumarate and succinate levels in EPAS1-Overexpressing PPGL cells. Supplementary Figure S16. Oxygen consumption and supply balance in the mitochondrial electron transport chain as an advantageous survival and proliferative mechanism. Supplementary Figure S17. Gene-level extremely positive selection on EPAS1 gene in systemic hypoxia. Supplementary Figure S18. Time-dependent effects of systemic hypoxia on oxygen-sensing tissues. Supplementary Figure S19. Specialization process of the HIF2α gene for adaptation to systemic hypoxia. Supplementary Figure S20. Proliferative advantage and increased sensitivity to HIF2α inhibition in HEK293T cells expressing the EPAS1 P405A/P531A double mutant. Supplementary Figure S21. Authentication of PPGL-derived PC12 cell line. Supplementary Figure S22. Full western blots shown in Fig. 3B. Supplementary Figure S23. Full western blots shown in Supplementary Fig S11.

This study explores parallels between systemic hypoxia adaptation in high-altitude populations and tumorigenesis. We identified EPAS1, a gene critical for hypoxia adaptation in populations such as Tibetans and Sherpas, as playing a similar adaptive role in tumors arising under hypoxic conditions. Tumors from patients with chronic hypoxia displayed impaired DNA repair and frequent emergence of EPAS1 variants, with frequencies reaching up to 90%, echoing the positive selection seen in high-altitude dwellers. Mechanistically, EPAS1 gain-of-function mutations promote COX4I2 expression, reducing cellular oxygen consumption and supporting tumor proliferation in hypoxia. Analysis of clinical data from patients with hypoxia revealed tissue-specific and time-sensitive tumorigenic effects, particularly impacting oxygen-sensitive cells in the postnatal period. Our findings suggest that EPAS1-driven adaptation mechanisms in high-altitude populations provide a model for understanding tumor evolution under hypoxic stress, highlighting how genetic adaptations to diverse stressors in natural populations may yield insights into tumorigenesis and cancer progression. Significance: This study reveals a broad convergence in genetic adaptation to hypoxia between natural populations and tumors, suggesting that insights from natural populations could enhance our understanding of cancer biology and identify novel therapeutic targets. See related commentary by Lee, p. 875

SUMMARY: Metastasis is responsible for nearly 90% of all cancer-related deaths. Despite global efforts to prevent aggressive tumours, cancers such as pancreatic ductal adenocarcinoma (PDAC) are poorly diagnosed in the primary stage, resulting in lethal metastatic disease. RAS mutations are known to promote tumour spread, with mutant KRAS present in up to 90% of cases. Until recently, mutant KRAS remained untargeted and, despite the recent development of inhibitors, results show that tumour cells develop resistance. Another strategy for targeting mutant KRAS-dependent PDAC proliferation and metastasis may come from targeting the downstream effectors of KRAS. One such axis, which controls tumour proliferation, invasiveness and immune evasion, is represented by ARF6-ASAP1. Here we show that targeting ARF6 results in adaptive rewiring that can restore proliferation and invasion potential over time. Using time-series RNA and ATAC sequencing approaches, we identified TLR-dependent NFκB, TNFα and hypoxia signalling as key drivers of adaptation in ARF6-depleted KRAS-dependent PDAC. Using in vitro and in vivo assays, we show that knocking down TLR2 with ARF6 significantly reduces proliferation, migration and invasion. Taken together, our data shed light on a novel co-targeting strategy with the therapeutic potential to counteract PDAC proliferation and metastasis. GRAPHICAL SUMMARY: