Wavelength-dependent label-free identification of isolated nontuberculous mycobacteria using surface-enhanced Raman spectroscopy

Abstract

We investigated the effect of Raman excitation wavelengths on the surface-enhanced Raman spectroscopy (SERS)-based identification of isolated nontuberculous mycobacteria (NTM). The SERS spectra with 3 commonly used excitation wavelengths, 532, 638, and 785 nm, were compared across 6 representative NTM species that primarily cause human NTM infections in Korea and the United States; these species were identified. The statistical differences among NTM SERS spectra at each Raman excitation wavelength were verified using 1-way analysis of variance, and the 6 NTM species were identified using principal components-linear discriminant analysis with leave-one-out cross validation. The identification accuracies with aromatic amino acid biomarkers were 99.3%, 91.3%, and 90.7% for 532, 638, and 785 nm, respectively. We believe that the proposed SERS protocol with aromatic amino acid biomarkers at the 532-nm Raman excitation wavelength will enable fast and accurate identification of NTM compared to previous identification methods.