In the emerging landscape of drug discovery, Chinese herbal medicines (CHMs) have long represented both a rich resource and a complex challenge. Despite their historical importance in traditional Chinese medicine (TCM) and their contribution to groundbreaking drugs like artemisinin and ephedrine, CHMs have struggled to fully integrate into modern pharmaceutical pipelines. Only a fraction—approximately 23.5%—of new drugs approved by the US FDA over the past four decades can be traced back to botanical drugs or natural products, highlighting a gap rooted in the multifaceted nature of CHMs. The intricate chemical makeup, the convoluted pharmacokinetics of multiple compounds, and the multifactorial mechanisms of action contribute to a therapeutic “black-box” that complicates characterization and regulatory acceptance.
Traditional target-based drug discovery (TDD) approaches, empowered by advances in artificial intelligence (AI) and three-dimensional molecular target analyses, have become the mainstay of modern pharmaceutical innovation. Nevertheless, this target-centric paradigm falls short in deconvoluting the nuanced, multitarget effects characteristic of CHMs due to their unresolved active constituents and unknown molecular targets. This mismatch underscores the urgent need for new methodologies that can reconcile the complexity of natural products with the rigor of contemporary drug development.
Addressing this challenge head-on, a pioneering study published in Engineering introduces the Phenotype–Target Coupled Drug Screening (PTDS) framework, a high-efficiency, stepwise workflow designed to harness the therapeutic potential of CHMs. PTDS bridges phenotypic drug discovery (PDD) with target-based drug discovery, leveraging the strengths of both strategies in a complementary fashion. Unlike the rigid target-focused methods, phenotypic screening identifies drug candidates based on observed biological effects without prior assumptions about molecular targets, a methodology credited with enabling the development of most FDA-approved first-in-class drugs.
The PTDS process is hierarchical and multifaceted, starting with phenotypic screening at multiple biological scales—from molecular and cellular systems to tissues and whole organisms. This staged approach allows for the precise localization of bioactive compounds and insights into their putative mechanisms. Following this functional mapping, target deconvolution techniques elucidate the molecular targets, further enabling the retargeting and refinement of drug candidates. This cyclical interplay between phenotypic readouts and molecular targets enhances the reliability and efficiency of drug discovery workflows, particularly in the context of multi-component natural products.
Technological advancements have been instrumental in enabling the scalable implementation of PTDS. Longitudinal multiomic integration facilitates the comprehensive analysis of dysregulated molecular networks over time, while dynamic network biomarker algorithms computationally spotlight critical nodes amenable to therapeutic intervention. AI-powered drug–target interaction (DTI) prediction models then assess the ability of compounds to restore these networks towards homeostasis. Such integrative computational-experimental pipelines dramatically optimize target identification and validation.
Moreover, cutting-edge experimental platforms complement these computational strides. High-resolution mass spectrometry-based metabolomics ascertain the chemical composition of CHM constituents at target organs, enabling spatially resolved mass spectrometry imaging that visualizes the tissue-level distribution of key metabolites. In parallel, AI-enhanced cell painting assays quantify subtle subcellular phenotypic alterations—capturing complex physiological phenomena such as mitochondrial dynamics, endoplasmic reticulum stress responses, and DNA damage—thereby elucidating cellular mechanisms even when targets remain unidentified.
The introduction of three-dimensional microphysiological systems—organoids and organ-on-a-chip technologies—adds another dimension to PTDS. These platforms mimic the architecture and physiological function of tissues and organs, providing more human-relevant environments for drug testing and mechanistic studies. The integration of multi-organ interconnected systems further advances in vitro absorption, distribution, metabolism, excretion, and toxicity (ADMET) evaluations, enabling more predictive preclinical data to inform candidate selection.
After preliminary phenotypic selection, the identification of drug targets is achieved via affinity chromatography and affinity mass spectrometry, building on the molecular-level interactions revealed by phenotypic methodologies. This target elucidation enables the rational screening of compound libraries with high throughput virtual screening platforms, streamlining the discovery process for CHM-derived candidates with mechanistic clarity.
Despite its promise, the PTDS framework is not without current limitations. Challenges persist in integrating multimodal datasets that span molecular, cellular, and organismal scales, and in the interpretability of AI models which often operate as “black boxes” themselves. Furthermore, organoid systems currently lack complete vascularization, limiting their capacity to fully replicate in vivo physiology. The synchronization and alignment of dynamic spatiotemporal data remain an unsolved hurdle that will be critical for future refinement.
Still, PTDS represents a transformative paradigm in natural product drug discovery. By narrowing the candidate landscape through rigorous phenotypic and target-based screening cycles, it clarifies the compound basis underlying CHM efficacy and opens avenues for mechanistic investigation. This approach has the potential to reduce attrition in preclinical and clinical phases, expediting the translation of botanical therapeutics into safe, efficacious medicines that meet modern regulatory standards.
The convergence of systems biology, AI, high-resolution analytics, and microphysiological modeling embodied in PTDS demonstrates a novel, resource-efficient pipeline. It invites a reevaluation of how complex natural products can be scientifically harnessed, potentially catalyzing a renaissance in botanical drug discovery that respects both the wisdom of traditional medicine and the rigor of molecular pharmacology.
This study, titled “Phenotype–Target Coupled Drug Screening: A High-Efficiency Framework for Innovative Drug Discovery from CHMs,” authored by Wei Zhou and Yue Gao, provides a detailed blueprint for this innovative workflow, establishing a practical framework for bridging phenotypic observation with molecular mechanism elucidation. It is accessible as an open-access publication in Engineering, offering valuable insights for researchers and pharmaceutical developers aiming to unlock the therapeutic potential of Chinese herbal medicines through advanced integrative methodologies.
Subject of Research: Innovative drug discovery leveraging Chinese herbal medicines (CHMs) through an integrated phenotypic and target-based screening framework.
Article Title: Phenotype–Target Coupled Drug Screening: A High-Efficiency Framework for Innovative Drug Discovery from CHMs
News Publication Date: 18-Feb-2026
Web References:
- Full article: https://doi.org/10.1016/j.eng.2025.11.019
- Journal homepage: https://www.sciencedirect.com/journal/engineering
Image Credits: Wei Zhou, Yue Gao
Keywords
Chinese herbal medicines, drug discovery, phenotypic screening, target-based drug discovery, phenotype–target coupled drug screening, PTDS, artificial intelligence, multiomic integration, mass spectrometry imaging, cell painting, organoids, microphysiological models, natural product pharmacology
