Chemical imaging of articular cartilage sections with Raman mapping, employing uni- and multi-variate methods for data analysis.

Raman mapping in combination with uni- and multi-variate methods of data analysis is applied to articular cartilage samples. Main differences in biochemical composition and collagen fibers orientation between superficial, middle and deep zone of the tissue are readily observed in the samples. Collagen, non-collagenous proteins, proteoglycans and nucleic acids can be distinguished on the basis of their different spectral characteristics, and their relative abundance can be mapped in the label-free tissue samples, at so high a resolution as to permit the analysis at the level of single cells. Differences between territorial and inter-territorial matrix, as well as inhomogeneities in the inter-territorial matrix, are properly identified. Multivariate methods of data analysis prove to be complementary to the univariate approach. In particular, our partial least squares regression model gives a semiquantitative mapping of the biochemical constituents in agreement with average composition found in the literature. The combination of hierarchical and fuzzy cluster analysis succeeds in detecting variations between different regions of the extra-cellular matrix. Because of its characteristics as an imaging technique, Raman mapping could be a promising tool for studying biochemical changes in cartilage occurring during aging or osteoarthritis.