Tools for the rational design of improved odorant ligands and receptors

Unmet Need

Our sense of smell, or olfaction, is integral to myriad aspects of everyday life, from our health and safety to our personal preferences. Yet the mechanisms behind olfaction are not well understood, leading to impacts on a wide range of issues, from therapeutics to consumer products. Olfactory dysfunction (OD) takes a burdensome toll on over 13 million patients in the US alone. Olfactory dysfunction is primarily treated with olfactory training, in which the patient smells 4 essential odors daily to retrain the nose and brain. The essential odors were selected according to the odor prism hypothesis developed in 1916, and clinical outcomes would benefit from an updated understanding of how the properties of odorants relate to their perception. Whether arising from COVID, aging, or neurodegenerative disease, OD is associated with accelerated cognitive decline, depression, and situational hazards and safety concerns like fire or spoiled food. Additionally, olfaction is central to consumer decision-making. Retailers, manufacturers, and advertisers have long recognized the power of fragrances in product design and marketing. Worldwide fragrance sales totaled $58B in 2023. Yet despite olfaction’s commercial, economic, and health importance, it is difficult to design better odorants due to the complex relationship between odorant structure and receptor response. There is a need for tools that enable the rational design of odorant molecules to deliver more effective products, marketing, and healthcare.

Technology

Duke inventors have developed an odorant prediction platform that combines software predictions and in vitro assays for studying how odorants bind their corresponding receptors. Specifically, the software models odorant-receptor interaction based on the molecular properties of the odorant and the amino acid sequence of the receptor. These predictions can be tested empirically using in vitro assays. These tools are intended to be used in a research setting to design improved odorant molecules, identify natural alternatives to synthetic odorants, or design synthetic alternatives to natural ones. This technology has applications in consumer product development, such as fragrances, cleaning products, cosmetics, and therapeutics. Furthermore, it has potential applications in smell retraining therapy for patients with OD by improving therapeutic odorants. This technology has been demonstrated with data generated in live mice. A classifier trained on this data can predict receptor activation based on the molecular properties of a given odorant, even if that odorant was not included in the training data. The dataset used to train the classifier was generated by exposing live mice to a diverse panel of 52 odorants, one odorant at a time while measuring receptor activation.

Other Applications

This technology can also be used to design synthetic odorant receptors for artificial sensing systems used in healthcare diagnostics, product quality control, and chemical monitoring systems. This technology can also aid in the design of food flavorants since flavor is heavily dependent on olfaction.

Advantages

• Rational design more effective, faster, cheaper than blind modification
• Tool consists of modeling and in vitro assays – providing a means to predict and test odorant design strategies without animal testing
• Ability to substitute natural<->synthetic to meet different consumer needs