Effect of cefazolin treatment on the nonresonant Raman signatures of the metabolic state of individual Escherichia coli cells.

Laser tweezers Raman spectroscopy (LTRS) was used to characterize the Raman fingerprints of the metabolic states of Escherichia coli (E. coli) cells and to determine the spectral changes associated with cellular response to the antibiotic Cefazolin. The Raman spectra of E. coli cells sampled at different time points in the bacterial growth curve exhibited several spectral features that enabled direct identification of the growth phase of the bacteria. Four groups of Raman peaks were identified based on similarities in the time-dependent behavior of their intensities over the course of the growth curve. These groupings were also consistent with the different biochemical species represented by the Raman peaks. Raman peaks associated with DNA and RNA displayed a decrease in intensity over time, while protein-specific Raman vibrations increased at different rates. The adenine ring-breathing mode at 729 and the 1245 cm(-1) vibration peaked in intensity within the first 10 h and decreased afterward. Application of principal component analysis (PCA) to the Raman spectra enabled accurate identification of the different metabolic states of the bacterial cells. The Raman spectra of cells exposed to Cefazolin at the end of log phase exhibited a different behavior. The 729 and 1245 cm(-1) Raman peaks showed a slight decrease in intensity from 4 to 10 h after inoculation. Moreover, a shift in the spectral position of the adenine ring-breathing mode from 724 to 729 cm(-1), which was observed during normal bacterial growth, was inhibited during antibiotic drug treatment. These results suggest that potential Raman markers exist that can be used to identify E. coli cell response to antibiotic drug treatment.