A Proteome-Wide Gain-of-Function Atlas Mapping the Landscape of Chemotherapy Response

Background: Conventional chemotherapies remain the core therapeutic backbone of cancer treatment, yet the rapid emergence of chemoresistance often compromises their clinical efficacy. Although large-scale functional genomics has transformed precision oncology, existing reference maps rely predominantly on loss-of-function paradigms and lack systematic data on gain-of-function (GOF) events that commonly drive clinical resistance. Methods: To address this gap, we engineered a site-specific Bxb1 integrase landing-pad system in HeLa cells, enabling single-copy integration of a human proteome-wide library [18,388 open reading frames (ORFs)] at the AAVS1 locus under a uniform EF1 promoter. We systematically profiled cellular fitness across 16 chemotherapeutics. We integrated this GOF perturbation screen with bulk and single-cell RNA sequencing, FACS-based lysosomal functional screening, clinical cohort survival analyses, and preclinical xenograft modeling. Results: Stratified modeling of over 230,000 gene-drug interactions defined distinct functional categories that are strongly enriched for clinical cancer biomarkers and prognostic relevance. Global network analysis and organelle-specific lysosomal functional screening uncovered that chemotherapeutic agents, most prominently the nucleoside analog TAS-102 (Lonsurf), converge upon a shared survival hub centered on the BCL-2 family-associated autophagy-endolysosomal network. Mechanistically, TAS-102 triggered lysosomal activation as an adaptive survival response. Co-treatment with the BCL-2 family inhibitor ABT-263 (Navitoclax) attenuated this TAS-102-induced lysosomal response and synergistically reversed TAS-102 resistance across colorectal and pancreatic cancer models in vitro and in colorectal cell line- and patient-derived xenograft (CDX, PDX) models in vivo. Conclusions: Our study establishes an open-access, proteome-wide GOF reference atlas that systematically maps chemotherapeutic responses, recapitulates canonical resistance biology, and nominates rational drug combinations. This systems-genomics resource bridges massive-scale perturbation mapping with actionable target discovery, providing a translational framework to overcome therapeutic resistance and reprogram cancer drug responses.