Maintaining low glucose concentrations in airway surface liquid (ASL) is essential to reduce the risk of airway infection. In normoglycemia (5mM) ASL glucose is maintained at ~0.4mM (12x lower than blood glucose), but can rise to 4mM during hyperglycaemia (15mM) and inflammation1. Monitoring ASL glucose may be a useful pre-emptive tool for detecting a risk of airway infection. However, in situ monitoring is difficult due to accessibility and the depth of the fluid (~8µm)2. To overcome this we designed a fluorescent glucose biosensor to measure glucose concentrations in the ASL. Point mutations were generated in the periplasmic glucose binding protein (GBP). Environmentally sensitive long wavelength fluorophores were attached (for future in vivo imaging) and the affinity and fluorescence range was measured in vitro. The protein exhibiting the largest fluorescence change over predicted ASL glucose concentrations was used to measure ASL glucose in bronchoalveolar lavage (BAL) from mice in a proof of concept study. In vivo studies were approved by the local Ethical committee in Gothenburg (29/2013). Data are shown as mean±SD and analysed using unpaired T-test. Of the mutated forms of GBP created, E149C/A213R GBP had an affinity to glucose within the range found in the ASL. Analysis of E149C/A213R GBP labelled with IANBD revealed the Kd for glucose was 0.84±0.26mM (n=5) (range: 25µM to 3mM), and fluorescence range; fmax/f0= 5.8. This was used to measure glucose in BAL samples from 9-week-old normoglycemic (C57BL/6) and hyperglycaemic (dbdb) mice. Blood glucose measurements were taken prior to BAL. C57BL/6 blood glucose was measured as 8.52±0.73mM (n=6), and for dbdb mice, this was 31.86±0.07mM (p≤0.0001). Glucose concentration in the BAL of dbdb mice, as measured by E149C/A213R GBP was 170.4±95.4µM. This was verified with a colorimetric glucose oxidase assay. For dbdb BALS the values did not differ (p=0.063; n=6). However, BAL glucose from C57BL/6 was not detected with the biosensor (detection limit:25µM), thus, glucose concentration in BAL from C57BL/6 was assumed to be lower than this, but the assay measured 48.67 ± 5.58µM. This discrepancy in values could be that glucose oxidase kits are more sensitive, or that the generation of colour can be affected by endogenous protease/oxidase activity and antioxidants. This study indicates that glucose concentrations measured in BAL using E149C/A213R GBP-IANBD were higher in hyperglycaemic compared to normoglycaemic mice. The biosensor exhibited a good fluorescence range and sensitivity and considering the ~17-fold dilution of the ASL from the BAL process, ASL glucose in dbdb was ~12x lower than blood glucose, as expected. This is promising data that if the sensor were applied directly to ASL rather than diluted BAL, it could possibly detect glucose in the ASL of both mice, providing a useful tool to monitor ASL glucose concentrations in situ.