This report examines lipophilic extracts containing mycolic acids isolated from tuberculosis (MTB) and non-tuberculosis (NTM) mycobacterial strains using chromatography, mass spectrometry (MS), nuclear magnetic resonance (NMR), and Raman spectroscopy. Gas chromatography-MS was used to identify major fatty acid mycolate components, while proton NMR confirmed the presence of characteristic cis- and trans-cyclopropane rings within different mycolic acid sub-types. Surface-enhanced Raman (SERS) spectra were obtained from the mycolic acids extracted from the bacterial cell envelopes of the MTB or NTM mycobacterial species. The Raman spectral profiles were used to develop a classification method based on chemometrics for identification of the mycobacterial species. Multivariate statistical analysis methods, including principal component analysis (PCA), hierarchical cluster analysis (HCA), and partial least squares discriminant analysis (PLS-DA) of the SERS spectra enabled differentiation of NTM mycobacteria from one another with 100% accuracy. These methods are also sensitive enough to differentiate clinically-isolated MTB strains that differed only by the presence or absence of a single extracytoplasmic sigma factor with 83-100% sensitivity and 80-100% specificity. The current work is the first report on discrimination of mycobacteria strains based on the SERS spectra of the constituent mycolic acids in lipophilic extracts. These results suggest that SERS can be used as an accurate and sensitive method for species and strain discrimination in mycobacteria.

Sigma factor C (SigC) contributes to Mycobacterium tuberculosis virulence in various animal models, but the stress response coordinated by this transcription factor was undefined. The results presented here indicate that SigC prevents copper starvation. Whole genome expression studies demonstrate short-term (4-hour) induction of sigC, controlled from a tetracycline-inducible promoter, upregulates ctpB and genes in the nonribosomal peptide synthase (nrp) operon. These genes are expressed at higher levels after 48-hour sigC induction, but also elevated are genes encoding copper-responsive regulator RicR and RicR-regulated copper toxicity response operon genes rv0846–rv0850, suggesting prolonged sigC induction results in excessive copper uptake. No growth and global transcriptional differences are observed between a sigC null mutant relative to its parent strain in 7H9 medium. In a copper-deficient medium, however, growth of the sigC deletion strain lags the parent, and 40 genes (including those in the nrp operon) are differentially expressed. Copper supplementation reverses the growth defect and silences most transcriptional differences. Together, these data support SigC as a transcriptional regulator of copper acquisition when the metal is scarce. Attenuation of sigC mutants in severe combined immunodeficient mice is consistent with an inability to overcome innate host defenses that sequester copper ions to deprive invading microbes of this essential micronutrient.