Search for: All records

Abstract Wastewater-based epidemiology (WBE) is a powerful tool for monitoring community disease occurrence, but current methods for bacterial detection suffer from limited scalability, the need fora prioriknowledge of the target organism, and the high degree of genetic similarity between different strains of the same species. Here, we show that surface-enhanced Raman spectroscopy (SERS) can be a scalable, label-free method for detection of bacteria in wastewater. We preferentially enhance Raman signal from bacteria in wastewater using positively-charged plasmonic gold nanorods (AuNRs) that electrostatically bind to the bacterial surface. Transmission cryoelectron microscopy (cryoEM) confirms that AuNRs bind selectively to bacteria in this wastewater matrix. We spike the bacterial speciesStaphylococcus epidermidis, Staphylococcus aureus, Serratia marcescens, andEscerichia coliand AuNRs into filter-sterilized wastewater, varying the AuNR concentration to achieve maximum signal across all pathogens. We then collect 540 spectra from each species, and train a machine learning (ML) model to identify bacterial species in wastewater. For bacterial concentrations of 10⁹cells/mL, we achieve an accuracy exceeding 85%. We also demonstrate that this system is effective at environmentally-realistic bacterial concentrations, with a limit of bacterial detection of 10⁴cells/mL. These results are a key first step toward a label-free, high-throughput platform for bacterial WBE. 

Tuberculosis (TB) is the world’s deadliest infectious disease, with over 1.5 million deaths and 10 million new cases reported anually. The causative organismMycobacterium tuberculosis(Mtb) can take nearly 40 d to culture, a required step to determine the pathogen’s antibiotic susceptibility. Both rapid identification and rapid antibiotic susceptibility testing of Mtb are essential for effective patient treatment and combating antimicrobial resistance. Here, we demonstrate a rapid, culture-free, and antibiotic incubation-free drug susceptibility test for TB using Raman spectroscopy and machine learning. We collect few-to-single-cell Raman spectra from over 25,000 cells of the Mtb complex strain Bacillus Calmette-Guérin (BCG) resistant to one of the four mainstay anti-TB drugs, isoniazid, rifampicin, moxifloxacin, and amikacin, as well as a pan-susceptible wildtype strain. By training a neural network on this data, we classify the antibiotic resistance profile of each strain, both on dried samples and on patient sputum samples. On dried samples, we achieve >98% resistant versus susceptible classification accuracy across all five BCG strains. In patient sputum samples, we achieve ~79% average classification accuracy. We develop a feature recognition algorithm in order to verify that our machine learning model is using biologically relevant spectral features to assess the resistance profiles of our mycobacterial strains. Finally, we demonstrate how this approach can be deployed in resource-limited settings by developing a low-cost, portable Raman microscope that costs <$5,000. We show how this instrument and our machine learning model enable combined microscopy and spectroscopy for accurate few-to-single-cell drug susceptibility testing of BCG. 

Colloidal single-walled carbon nanotubes (SWCNTs) oer a promising platform for the nanoscale engineering of molecular recognition. Optical sensors have been recently designed through the modification of noncovalent corona phases (CPs) of SWCNTs through a phenomenon known as corona phase molecular recognition (CoPhMoRe). In CoPhMoRe constructs, DNA CPs are of great interest due to the breadth of the design space and our ability to control these molecules with sequence specificity at scale. Utilizing these constructs for metal ion sensing is a natural extension of this technology due to DNA’s well-known coordination chemistry. Additionally, understanding metal ion interactions of these constructs allows for improved sensor design for use in complex aqueous environments. In this work, we study the interactions between a panel of 9 dilute divalent metal cations and 35 DNA CPs under the most controlled experimental conditions for SWCNT optical sensing to date. We found that best practices for the study of colloidal SWCNT analyte responses involve mitigating the eects of ionic strength, dilution kinetics, laser power, and analyte response kinetics. We also discover that SWCNT with DNA CPs generally oers two unique sensing states at pH 6 and 8. The combined set of sensors in this work allowed for the dierentiation of Hg2+, Pb2+, Cr2+, and Mn2+. Finally, we implemented Hg2+ sensing in the context of portable detection within fish tissue extract, demonstrating nanomolar level detection.