EFRI ELiS: Desiccatable living cell-based sensors to monitor pollutants and pathogens in built environments

University Of Maryland, College Park, College Park MD

Investigators

Abstract

The sense of smell is amazingly versatile. Birds and insects can sense and seek out specific flowers. Hounds can follow scent trails that are hours and days old. Even humans can detect about 10,000 different odors. Current sensing technology, in comparison, is poor at detecting and discriminating between different odorants. Environmental monitoring would benefit greatly from a dramatically improved odor sensing technology. The objective of this project is to create an artificial “e-nose” technology based on engineered biological cells. These cells will emit fluorescent signals when compounds are detected. In addition to monitoring the safety of human environments, a general-purpose e-nose would have a range of applications including food safety, non-invasive medical diagnostics, and health monitoring; detecting explosives and illicit drugs; and mapping agricultural pests. Ethical issues will be assessed in public engagement exercises with focus groups. The research will utilize an insect cell line expressing an odorant receptor and a GCaMP reporter for Ca2+, giving a fluorescence signal upon odorant binding. The cell line will be able to be desiccated and rehydrated, while retaining functionality. Desiccation addresses the problem of the impracticality of device storage that has prevented wide-spread deployment of cell-based sensors. Beginning with an existing cell line expressing the Drosophila receptor Or47a, responses to odorants, singly and in mixtures, will be characterized to understand the requirements for reliable recognition, given biological variability. Odorant introduction has largely been confined to liquids, so methods for introducing airborne odorants will be investigated. At least three new stable cell lines expressing different odorant receptors will be developed for a prototype array by optimizing the expression system; reliable creation of new cell lines is required for the large number of array elements needed for an eventual general-purpose e-nose. The spatio-temporal patterns generated by the cell arrays to various odors will be complex, posing a challenge for odor recognition even using machine learning (ML), since there are no ML models that handle combinatorial interactions and nonlinear dynamical phenomena. The integration of new ML algorithms will lead to the possibility of implementing combinatorial coding in an e-nose. A miniaturized optical detection system will be developed, generating significant knowledge concerning requirements for a portable device. The project will expand knowledge about how to practically achieve artificial olfaction from multiple angles: biology, hardware, and data interpretation; it will also shed light on ethical concerns and community involvement around engineered living systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.