Selective exhaled breath condensate collection and competitive fluorescent biosensor for non-invasive glucose detection

Two thirds of patients with diabetes avoid regularly monitoring their blood glucose levels because of the painful and invasive nature of current blood glucose detection. As an alternative to blood sample collection, exhaled breath condensate (EBC) has emerged as a promising non-invasive sample from which to monitor glucose levels. However, the inconsistency in the methods used to collect EBC significantly impacts the reliability of reported analyte concentrations in EBC. Furthermore, this dilute sample matrix requires a highly sensitive glucose biosensor to enable robust and accurate glucose detection at the point-of-care. Together, a reliable collection method and sensitive detection system can enable accurate modeling of glucose transport from blood to breath that is reflective of airway glucose homeostasis.

I address this research gap by simultaneously designing a standardized EBC collection method that allows for separation of dead space and alveolar air and developing a competitive fluorescent biosensor that can resolve micromolar glucose concentrations changes. First, I develop a low-cost, automated condenser that selectively collects exhaled breath that has been exchanged with lung fluid based on the detection of higher breath temperatures that are characteristic of the lower respiratory regions. Using this device, I investigate the relationship between blood and EBC glucose in diabetic and normoglycemic human subjects. Next, I engineer the exquisitely sensitive E. coli glucose binding protein (GBP) with a chemo-enzymatic tag to selectively conjugate it to highly photostable quantum dots (QDs). Finally, I take advantage of the competitive binding of glucose (K_D=0.35 µM) and galactose (K_D=1.4 µM) to GBP to develop a fluorescent glucose biosensor using the GBP-QD conjugate.