Scaffold Fusion and SAR Transfer with a Chemical Language Model Generates Novel Liver X Receptor Modulators

Area of Science:

• Medicinal Chemistry
• Molecular Pharmacology
• Computational Chemistry

Background:

• Liver X receptors (LXRs) are key regulators of lipid metabolism.
• Dysregulation of LXRs is implicated in atherosclerosis and metabolic dysfunction-associated steatotic liver disease (MASLD).
• Novel therapeutic strategies targeting LXRs are needed for these metabolic disorders.

Purpose of the Study:

• To explore the use of a chemical language model (CLM) for the de novo design of LXR modulators.
• To generate novel LXR modulators with diverse activity profiles and selectivity.
• To identify promising candidates for further development, particularly for MASLD.

Main Methods:

• Utilized a chemical language model (CLM) for generative design of LXR modulators.
• Integrated structural features from existing LXR modulator templates.
• Transferred structure-activity relationship (SAR) knowledge during the design process.
• Evaluated generated compounds for LXR modulation and selective properties in vitro.

Main Results:

• The CLM successfully merged structural features and transferred SAR knowledge.
• Generated computational designs exhibited diverse LXR modulation activities.
• Selective modulator properties were observed among the designed compounds.
• An inverse LXR agonist demonstrated significant lipolytic activity in an in vitro MASLD model.

Conclusions:

• Chemical language models are effective tools for designing novel LXR modulators.
• The study identified a promising inverse LXR agonist with potential therapeutic application for MASLD.
• Further optimization of ADME properties is warranted for the identified lead compound.