Combining Surface-Enhanced Raman Spectroscopy and Electrokinetics for Bacterial Monitoring in Wastewater

Wastewater-based epidemiology can monitor population-level infections, provide early warnings about disease outbreaks, and help to control disease spread at the community level [1]. However, bacterial identification in wastewater presents outstanding challenges; notably, current methods [2] to identify bacteria are slow and costly, not suitable for high-throughput screening of diverse bacterial species, and may not work well in the complex wastewater matrix. Here, we develop a new electro-optical method that has the promise to detect a wide range of pathogenic bacteria in wastewater.<br/><br/>We combine surface-enhanced Raman spectroscopy (SERS) [3] and electrified interfaces, further assisted with machine learning models [4] to realize rapid and amplification-free detection of bacteria, at cell concentrations as low as 10⁵ cells/mL in filtered wastewater. First, we collect SERS from bacteria spiked into filtered wastewater, including <i>Staphylococcus aureus, Staphylococcus epidermidis</i>, and <i>Escherichia coli</i>. Here, plasmonic gold nanorods electrostatically bind to the bacteria surface, allowing for rapid biomolecular recognition of the cell surfaces. Additionally, two-dimensional gold microelectrodes are patterned on the sensor surface to apply electrical fields to wastewater, which selectively attracts and enriches bacteria present in the complex wastewater matrices within minutes, driven by dielectrophoretic effects [5]. The enrichment of bacteria is directly observed through microscopy, and Raman signal intensities are increased by five to ten fold under electrical fields for bacterial concentrations ranging from 100 cfu/mL up to 10⁸ cfu/mL. Such enhancement enables the detection sensitivity to reach environmentally-relevant concentrations (typically below 10⁷ cfu/mL) with just ten seconds of integration times. Finally, using machine learning, we identify biologically-relevant “fingerprint” Raman peaks describing proteins, nucleic acids, and lipids from bacteria surfaces, and achieve rapid and accurate identification of bacteria species in wastewater. The method has the promise for future label-free detection and prediction of bacterial outbreaks through wastewater. It also opens up new directions for generalized pathogen detection and molecular recognition in a broader range of complex liquid samples including wastewater, blood, and seawater.<br/><br/>[1] Hellmér, et al. Applied and environmental microbiology. (2014).<br/>[2] Jahn, et al. Nature Microbiology. (2022).<br/>[3] Tadesse, et al. Nano Lett. (2020).<br/>[4] Ho, et al. Nat. Comm. (2019).<br/>[5] Pethig. John Wiley & Sons (2010).