Infrared microspectroscopy used in the search for spectral markers of liver disease

When a cross-section of biological tissue is illuminated with different wavelengths in the infrared domain, each of the molecular groups constituting the tissue will respond to the light excitation in enhancing its vibrational motion when one specific wavelength comes into resonance with it. A spectrum is then recorded in which all the vibration signatures of the various molecules appear. Infrared spectroscopy thus gives access to the overall composition of the tissue in its components: proteins, lipids, carbohydrates, and nucleic acids (DNA and RNA).
By combining microscopy with spectroscopy, it also becomes possible to determine these compositions on a few microns scale, providing some idea of the heterogeneity of the tissue. In addition to these microscopic analyses, biochemical imaging of the tissue can be performed: by taking spectral measurements at every point of the cross-section (where a point corresponds to the diameter of the incident infrared beam) and compiling the data obtained, it is possible to reconstruct the distribution of the various tissue constituents on the surface being analysed.

Pathology generates modifications of the tissue in terms of its molecular composition and its metabolism. If the existence of a 'spectral signature' reflecting the specific changes caused by a particular disease could be identified, it is easy to realize why it would be useful to carry out a systematic search for this signature, which would have diagnostic value.
That is the idea that Marie-Pierre Bralet (*) and François Le Naour of the Paul Brousse hospital in Villejuif worked on with Christophe Sandt, Frédéric Jamme and Paul Dumas, of the SMIS beamline. They applied for beam time at SOLEIL in the first call for projects, in November 2006. Paul Dumas already had some expertise in the field of analysing biological tissue, acquired at LURE, and so he was able to help the future SMIS users to evaluate the feasibility of their project.
Their first series of experiments on SMIS made it possible to evaluate the protocol for the preparation of these samples: untreated cross-sections 6 microns thick obtained from previously frozen clinically characterised liver tissue.
A second series of experiments, carried out in 2008, confirmed the validity of the method and the advantages of a synchrotron source for this technique: the brilliance (quantity of light) of the infrared radiation reduces the spectrum acquisition time and increases the lateral resolution. This series of experiments was carried out on samples from patients afflicted with steatosis. The tests revealed significant changes in the biochemical composition. Moreover, statistical analyses carried out on a large number of spectra showed how heterogeneous steatosic tissue is compared to healthy tissue. Interestingly, the morphologically normal regions of steatosic livers had many spectral characteristics that distinguished them from a healthy liver. This means that infrared microspectroscopy can reveal early changes in the metabolism of hepatocytes.

These studies should continue with research on cellular and animal models so that the early stages of the pathology can be analysed. The medical interest of spectral markers will be studied using series of clinical samples.
 

(*) Marie-Pierre Bralet, during the preparation of this report, passed away. She was the driving scientist of this project, and her competence, human quality were very much appreciated. We will try to carry out this study as far as possible in memory of Marie-Pierre.