Films of Candida albicans, Candida dubliniensis and Candida parapsilosis were prepared and the infrared spectra of these films were obtained in the region 4000 to 1000 cm-1, with resolution of 4 cm-1, in the transmission mode, at 20 ºC. Fifty four spectra were obtained, 18 of each microorganism, with the PerkinElmer Spotlight 400 FT-IR, which has a microscope attached to a FT-IR spectrophotometer. The spectra were analyzed through three methods: (1) mere visual inspection; (2) multivariate statistical analysis; (3) curve-fitting for determining secondary structures of proteins. In the region 1200 to 1000 cm-1, the spectral bands show differences that can be seen by a mere visual inspection. On the other hand, the amide I bands, in the region 1710 to 1590 cm-1, have the same visual aspect for the three microorganisms. Multivariate statistical analysis was applied to analyze these amide I bands of all the 54 spectra. Principal component analysis (PCA) and techniques of hierarchical cluster analysis (HCA, Hierarchical Clustering Analysis) according to Ward's method were applied using the software MINITAB 15. The results show a clear discrimination of the three microorganisms. The average spectrum of each microorganism was obtained in the amide I band. Each average spectrum was analyzed by curve-fitting for the determination of secondary structures of proteins. The software used was the ORIGIN 7.5 and the results confirm the discrimination obtained through multivariate statistical analysis. This result shows that multivariate statistical analysis can be useful to discriminate infrared spectra of different microorganisms. Furthermore, this work shows that the amide I bands of the infrared spectra of Candida albicans, Candida dubliniensis, and Candida parapsilosis provide a set of data of known group structure that can be useful to test statistical algorithms of cluster analysis.
Candida albicans; Candida dubliniensis; Candida parapsilosis; Multivariate statistical analysis; Infrared microespectroscopy; Secondary structures of proteins
