Highly sensitive CMOS bioluminescent biosensor for quantitative, high dynamic range detection of total surface ATP and selective detection of E. coli.

Rapid and reliable microbial contamination detection is essential for hygiene monitoring in food production, healthcare, and environmental settings. Although ATP-based bioluminescence assays are widely used for rapid cleanliness assessment, conventional luminometer-based systems are limited by high cost, large sample volume requirements, limited dynamic range, and lack of bacterial specificity. In this work, we present an improved CMOS-based bioluminescent biosensing platform for quantitative ATP detection and species-selective detection of Escherichia coli. The platform integrates SuperSnap ATP chemistry and MicroSnap E. coli enzyme-specific assays with a CMOS image sensor housed in a light-tight optical enclosure, enabling chemiluminescence detection using sample volumes as low as 10 μL. Quantitative ATP measurements in SuperSnap buffer exhibited strong linearity and good agreement with a commercial luminometer, achieving a limit of detection of 3.51 nM ATP. When normalized to absolute ATP amount, the CMOS platform demonstrated sensitivity comparable to that of the luminometer despite operating with a 30-fold smaller sample volume. In contrast to the luminometer, which saturated at approximately 75 nM ATP, the CMOS platform maintained linear response up to 200 nM ATP, demonstrating a substantially broader dynamic range without requiring sample dilution. Selective E. coli detection achieved a limit of detection of approximately 412 CFU mL-1 following 6-h enrichment and 10 min detection, while maintaining selective response in mixed bacterial samples containing Staphylococcus epidermidis. Performance was further validated through representative surface sampling and controlled surface contamination experiments. Overall, the CMOS-based platform provides a low-cost and portable alternative to conventional luminometers for both rapid hygiene monitoring and targeted microbial detection at the point-of-need.