Integrating Surface-Enhanced Raman Spectroscopy and Electrokinetics for Bacterial Identification in Wastewater

Antibiotic-resistant bacterial infections claim over 1.2M lives annually and are projected to be the main cause of death in 30 years [1]. Elevated pathogen levels in wastewater are one of the first indicators of disease outbreaks, making wastewater a powerful tool for surveilling the infections present in a community [2]. However, bacterial WBE presents outstanding challenges; current culturing or fluorescence-based methods [3] to identify bacteria are slow and costly, and not suitable for high-throughput screening of diverse bacterial species. Further, the complex substances in wastewater can interfere with chemical probes or cause false negative results.<br/>In this study, we develop a generalized enrichment with Raman-machine learning spectroscopy (GERMS), employing surface-enhanced Raman spectroscopy (SERS) [4] and integrating it with electric fields and machine learning models [5], to enable rapid and amplification-free bacteria identification in filtered wastewater, even at low concentrations down to 10⁴ cells/mL. To achieve this, we first synthesize gold nanorods designed to electrostatically bind with bacteria, enabling SERS measurements from the cell surfaces. We perform SERS measurements on various bacteria, including Staphylococcus aureus, Staphylococcus epidermidis, and Escherichia coli, across a concentration range spanning from 10⁹ down to 10⁴ cells/mL. Spectral clustering analysis reveals that as the concentration decreases, bacterial signals become progressively more challenging to distinguish from the background wastewater. Furthermore, we incorporate electrokinetic effects into SERS by employing gold microelectrodes to apply external electric fields and leveraging dielectrophoresis [6] to rapidly displace and enrich bacteria within minutes. Microscopy directly visualizes the enrichment of bacteria with nanorods, leading to a remarkable increase in Raman signal intensities by up to tenfold under external electric fields for bacterial concentrations from 10⁶ down to 10⁴ cells/mL. This enhancement has the potential to extend the detection sensitivity to environmentally-relevant concentrations. In addition, through data science approaches, we identify biologically significant "fingerprint" Raman peaks that characterize proteins, nucleic acids, and lipids on bacterial surfaces. This discovery holds promise for the rapid identification of bacterial species for wastewater-based epidemiology.<br/> <br/>[1] Thompson, T. Nature. (2022) <br/>[2] Keshaviah, et al. The Lancet Global Health. (2023)<br/>[3] Jahn, et al. Nature Microbiology. (2022) <br/>[4] Tadesse, et al., Nano Lett. (2020).<br/>[5] Ho, et al., Nat. Comm. (2019).<br/>[6] Pethig. John Wiley & Sons (2010).