Fast and Inexpensive Detection of Bacterial Viability and Drug Effectiveness through Metabolic Monitoring

Conventional methods for the detection of bacterial infection such as DNA or immunoassays are expensive, time consuming, or not definitive and thus may not provide all the information sought by medical professionals. In particular, it is difficult to obtain information about viability or drug effectiveness, which is crucial to formulate a treatment. Bacterial culture tests are the “gold standard” because they are inexpensive and do not require extensive sample preparation, and most importantly, provide all the necessary information sought by healthcare professionals, such as bacterial presence, viability and drug effectiveness. These conventional culture methods, however, have a long turnaround time, anywhere between 1 day and 4 weeks. Here, we solve this problem by monitoring the growth of bacteria in thousands of nanowells simultaneously to more quickly identify their presence in the sample and their viability. The segmentation of a sample with low bacterial concentration into thousands of nanoliter wells digitizes the samples and increases the effective concentration in those wells that contain bacteria. We monitor the metabolism of aerobic bacteria by using an oxygen-sensitive fluorophore, ruthenium tris (2,2’-diprydl) dichloride hexahydrate (RTDP), which allows us to monitor the dissolved oxygen concentration in the nanowells. Using E. coli K12 as a model pathogen, we demonstrate that the detection time of E. coli can be as fast as 35–60 min with sample concentrations varying from 10(4) (62 min for detection), 10(6) (42 min) and 10(8) cells/mL (38 min). More importantly, we also demonstrate that reducing the well size can reduce the detection time. Finally we show that drug effectiveness information can be obtained in this format by loading the wells with the drug and monitoring the metabolism of the bacteria. The method that we have developed is low cost, simple, requires minimal sample preparation and can potentially be used with a wide variety of samples in a resource-poor setting to detect bacterial infections such as tuberculosis.