Taste, food and health

Taste is a major driver of food choice and consumption. Ligands of taste GPCRs are numerous, chemically diverse and often have multiple biotargets. Extraoral expression of taste receptors suggests that yet unknown, endogenous ligands and modulators may be essential for their physiological roles. We integrate machine learning and modeling with experimental testing to obtain a deeper understanding of the bitter and sweet chemical space and its biological implications.

I will present the BitterDB database of bitter molecules(1) and machine learning approaches BitterIntense(2) for intense bitterness prediction, and BitterMatch(3) that matches molecules to bitter taste receptors. I will share insights from applying these tools to large chemical datasets, and then introduce an iterative data-driven approach that lead us towards discovery of several T2R14 antagonists.

While bitter taste recognition is achieved by multiple T2R subtypes, typically via their orthosteric binding sites, the versatility of T1R2/T1R3 heterodimeric sweet taste receptor is facilitated by multiple binding sites. I will present our recent findings on T1R2 and T1R3 roles in recognition of sweet molecules.

The implications of the findings will be discussed in the context of food and drugs.

1. A. Dagan-Wiener et al., BitterDB: taste ligands and receptors database in 2019. Nucleic Acids Res 47, D1179-D1185 (2019).

2. E. Margulis et al., Intense bitterness of molecules: Machine learning for expediting drug discovery. Comput Struct Biotechnol J 19, 568-576 (2021).

3. E. Margulis et al., BitterMatch: recommendation systems for matching molecules with bitter taste receptors. J Cheminform 14, 45 (2022).