Chemoproteomic Strategies for Natural Product Target Identification Between 2017 and 2025 and Their Contribution to Mechanistic Drug Discovery

Download PDF

Natural products frequently enter drug discovery through phenotypic activity, yet translating an active molecule into a defensible molecular mechanism requires more than identifying proteins that bind, enrich, label, or change stability. Chemoproteomics has expanded the experimental routes available for natural-product target identification, but different platforms interrogate different forms of molecular engagement and impose distinct interpretive limitations. This systematic review examined peer-reviewed literature published from 2017 through 2025 concerning natural-product target identification by affinity capture, photoaffinity labeling, covalent and activity-based protein profiling, thermal and other probe-free proteome-wide approaches, and integrative validation. Evidence was organized according to target nomination, competition or specificity testing, site-level information, orthogonal target-engagement validation, functional dependence, and mechanistic interpretation. Methodological quality was assessed using a proposed qualitative appraisal appropriate to heterogeneous chemoproteomic designs rather than a numerical certainty score. The literature shows a transition from enrichment-centered target fishing toward increasingly proteome-wide, site-aware, probe-free, and perturbation-enabled strategies. Across platforms, target nomination was substantially easier to establish than functional mechanism. Probe derivatization, electrophile selectivity, indirect thermal effects, target-class rather than single-target engagement, and natural-product polypharmacology repeatedly complicated interpretation. Mechanistically stronger studies combined chemoproteomic discovery with independent biochemical, structural, genetic, pharmacological, or phenotypic evidence. Chemoproteomics contributes most strongly to mechanistic drug discovery when treated as an evidence-generating system rather than a target-listing technology. A proposed evidence-convergence interpretation is developed while preserving uncertainty, method-specific observability, and the distinction between engagement and causal mechanism.