Breakthrough in Taste Receptor Research: Virtual Screening Uncovers Steroid Candidates for TAS2R46
Researchers at Kyushu University have published new findings that advance understanding of the human bitter taste receptor TAS2R46. The study, titled Structure-based virtual screening identifies potential endogenous ligands of the human bitter taste receptor TAS2R46, appears in the Journal of Biological Chemistry. Lead authors Yuki Nagasato, Keisuke Sanematsu, Yuko Kawabata, Shingo Takai, and Noriatsu Shigemura combined computational methods with laboratory validation to identify steroid hormones as likely activators of this receptor outside the mouth.
The work builds directly on the 2022 determination of the TAS2R46 structure. By applying ensemble docking across multiple receptor conformations generated through molecular dynamics simulations, the team screened the Human Metabolome Database and narrowed candidates to nine steroid compounds. Functional assays confirmed activation by eight of them, highlighting a potential new role for TAS2R46 in sensing internal body chemistry.
Background on Bitter Taste Receptors and Extra-Oral Expression
Bitter taste receptors, known collectively as TAS2Rs, form a family of G protein-coupled receptors that help detect potentially harmful compounds in food. Humans express around 25 such receptors. While their primary function occurs in taste cells on the tongue, scientists have documented TAS2R expression in many other tissues, including the airways, intestine, pancreas, brain, and reproductive organs. This distribution raises questions about additional physiological functions beyond taste perception.
For receptors like TAS2R46, which responds to compounds such as strychnine, the search for natural activators inside the body has been ongoing. Extra-oral locations are not exposed to dietary bitters, so endogenous molecules likely serve as signals. The new study addresses this gap by leveraging the recently solved three-dimensional structure of TAS2R46 to perform targeted virtual screening.
Methods: Ensemble Docking and Machine Learning Integration
The research team started with the known cryo-electron microscopy structures of TAS2R46. They ran multiple 500-nanosecond molecular dynamics simulations of the receptor bound to strychnine in a lipid bilayer environment. From these trajectories, they extracted diverse conformations focused on the binding-site residues.
Principal component analysis helped map the conformational space. The group then performed ensemble docking of compounds from the Human Metabolome Database against these conformations. A machine learning step aggregated scores from the different docking runs, improving the reliability of predictions compared with single-structure approaches. This hybrid strategy identified nine steroid hormones or derivatives as top candidates.
Key Findings: Eight Steroids Activate TAS2R46
Laboratory tests confirmed that eight of the nine predicted compounds activate TAS2R46. The active molecules include 17-hydroxyprogesterone, testosterone, dihydrotestosterone, dehydroepiandrosterone, androstenedione, corticosterone, deoxycorticosterone, and cortexolone. Estrone did not show activation under the tested conditions.
Further computational modeling and mutational analysis pinpointed key amino acid residues involved in steroid binding and receptor activation. The steroid backbone appears to interact with specific sites that stabilize the active conformation. These details provide a mechanistic framework for how TAS2R46 might respond to circulating hormones.
Photo by Google DeepMind on Unsplash
Implications for Physiology and Extra-Oral Functions
The identification of steroid hormones as potential endogenous ligands expands the known scope of TAS2R46. In extra-oral tissues, these receptors may monitor metabolic or hormonal states rather than solely serving as toxin detectors. For example, activation in the airways or gut could influence local responses tied to steroid levels.
Such findings align with broader interest in TAS2Rs as chemical sensors throughout the body. They also open avenues for exploring connections between bitter taste pathways and endocrine or immune functions, though the precise downstream effects remain subjects for future investigation.
Broader Context in GPCR Ligand Discovery
Structure-based virtual screening has gained traction for G protein-coupled receptors as more experimental structures become available. The approach used here, combining molecular dynamics, ensemble docking, and machine learning, offers a template that could apply to other TAS2Rs or related receptors where traditional ligand-based methods fall short.
By focusing on the Human Metabolome Database, the study prioritized biologically relevant molecules over synthetic libraries. This choice increases the likelihood that identified hits represent genuine physiological signals.
Research Team and Institutional Setting
All authors are affiliated with Kyushu University in Fukuoka, Japan. Their primary base is the Section of Oral Neuroscience within the Graduate School of Dental Science, with additional ties to the Oral Health/Brain Health/Total Health Research Center and the Research and Development Center for Five-Sense Devices. This interdisciplinary environment supports integrated studies of taste receptors from structural biology to functional assays.
The collaboration reflects ongoing strengths at the university in sensory neuroscience and computational biology.
Future Directions and Potential Applications
The study establishes a computational pipeline that can be refined and extended to other receptors. Follow-up work may examine dose-response relationships, tissue-specific effects, and interactions with known bitter compounds. Understanding how steroids modulate TAS2R46 could inform research into metabolic disorders or respiratory conditions where these receptors play roles.
Longer term, the framework may assist in designing selective modulators for taste receptors, with possible relevance to nutrition, pharmacology, or even flavor science.
Photo by Compagnons on Unsplash
Accessing the Original Publication
The full paper is available at https://www.sciencedirect.com/science/article/pii/S0021925826021678. It credits Yuki Nagasato, Keisuke Sanematsu, Yuko Kawabata, Shingo Takai, and Noriatsu Shigemura as authors and was published online in June 2026 in the Journal of Biological Chemistry.