Generative Artificial Intelligence Optimization of Albumin Binders: Coumarin and Fatty Acid Derivatives

Clinical Snapshot

Key Findings

• P Value: Not reported in abstract

• Effect Size: Not quantified in abstract; described qualitatively as 'enhanced affinity' and 'nanomolar HSA binding'

• Primary Outcome: Optimised coumarin derivative (CD1) and fatty acid derivative achieved nanomolar HSA binding affinity by SPR, with enhanced affinity compared to parent compounds (4-hydroxycoumarin and dodecanedioic acid)

• Nnt Or Sensitivity: Not applicable at this preclinical stage; KD values (nanomolar range) are the relevant metric but specific values not provided in abstract. CD1 showed no detectable anticoagulant activity in acute clotting time assay (qualitative negative result)

• Confidence Interval: Not reported in abstract

Clinical Application

The AI-guided synthesis pipeline is feasible within a well-resourced pharmaceutical or academic medicinal chemistry setting. SPR instrumentation and computational infrastructure are required. Clinical translation would require extensive additional development including ADMET profiling, in vivo pharmacokinetic studies, and regulatory preclinical safety packages — a timeline of many years. There is no immediate relevance to Australian PBS listings, TGA regulatory decisions, or RACGP clinical guidelines. HSA-binding drug optimisation is a foundational drug discovery activity. If the approach eventually yields clinically viable compounds, TGA evaluation under the Australian Regulatory Guidelines for Prescription Medicines (ARGPM) would apply. The research aligns with NHMRC priorities for translational medicine and may be of interest to Australian medicinal chemistry groups at institutions such as Monash Institute of Pharmaceutical Sciences or the University of Queensland's Institute for Molecular Bioscience. No PBS or prescribing implications exist at this stage. No direct clinical applicability at present. The research is relevant to the preclinical development of HSA-binding drug conjugates, albumin-targeted drug delivery systems, or compounds designed to exploit HSA as a pharmacokinetic modulator. Future applications may include oncology (albumin-drug conjugates), long-acting injectables, or anti-inflammatory therapeutics.

Abstract

Previously, we reported a dual combination based on 4-hydroxycoumarin and dodecanedioic acid that could synergistically bind to human serum albumin (HSA). However, optimizing this combination remains challenging and could often be guided by empirical selection and extensive experimental screening, which may limit the chemical diversity and suboptimal affinity. In this study, we established a systematic artificial intelligence framework that integrates computational optimization with wet-lab synthesis and experimental validation, enabling improvement of the dual combination while preserving the core chemotypes. We first trained a machine learning classifier on curated HSA binding data and used it as an external scoring function to guide reinforcement learning-driven scaffold decoration with LibINVENT, enabling goal-directed generation of coumarin derivatives and fatty acid derivatives. Candidate molecules were prioritized through multiparameter filtering and diversity-aware selection, followed by synthesis and experimental validation using surface plasmon resonance. The optimized representatives show nanomolar HSA binding and enhanced affinity compared to the original ligands. Molecular docking and molecular dynamics simulations further provide a mechanistic rationale for the affinity improvements by revealing additional stabilizing interactions and more favorable binding energetics at the corresponding HSA sites. Besides, the optimized coumarin derivative (CD1) is a warfarin-derived coumarin analogue, yet it did not show detectable anticoagulant activity in an acute clotting time assay, whereas warfarin did. Overall, this work demonstrates a practical AI-guided route to expand chemical diversity and improve affinity for a synergistic HSA binding combination.

References

1. 1.Zhang, Y., Deng, Q., Yang, X., Yue, X., Zhang, H., Ding, S., Lei, J., Zhang, B., Yu, S., & Zhang, G. (2026). Generative artificial intelligence optimization of albumin binders: Coumarin and fatty acid derivatives. Journal of Chemical Information and Modeling. Advance online publication. https://doi.org/10.1021/acs.jcim.6c00296