Characterisation of bacteria by surface enhanced Raman spectroscopy (SERS)

Bacterial infections are common and usually require antimicrobial therapy; however, as antibiotic resistance is increasing, mortality and morbidity rates are also increasing, placing significant economic burden on healthcare systems worldwide. Therefore, rapid detection of bacterial species causing infection could help inform clinicians regarding appropriate antibiotic treatment. Numerous techniques are available for the detection of bacteria causing infection including culture-based, molecular, and spectroscopic techniques. Surface-enhanced Raman spectroscopy (SERS) has received considerable attention as a potential method for the characterisation of bacteria. The aims of this thesis were to (i) develop a SERS method that would characterise and distinguish between different bacterial species, (ii) investigate variation in the SERS spectra of Pseudomonas aeruginosa and determine if this was related to genotypic or phenotypic differences and (iii) determine the effect on the SERS spectra of altering the P. aeruginosa phenotype in vitro. Results from this study showed that a reproducible SERS method could be developed that removed multiple sources of experimental and microbiological variation. This standardised SERS method, incorporating principal component analysis (PCA), was not suitable to distinguish between the SERS spectra of large numbers of bacterial isolates and species. Variation was present in the SERS spectra of bacterial isolates grown on different days and between different bacterial isolates from the same bacterial species. Variation in the SERS spectra of P. aeruginosa could not be explained solely by differences in the strain tested or differences in the phenotype. Differences were observed in the SERS spectra of P. aeruginosa isolates obtained under anaerobic conditions compared to aerobic conditions. The variation in the SERS spectra of P. aeruginosa isolates may result from adaption of these isolates to different micro-environments in the CF lung, resulting in genotypic and phenotypic diversity between and within strains. In conclusion, this SERS method may be of limited value characterising bacterial isolates cultured from the sputum from people with CF.