Current Status and Future Applications of MALDI Imaging

MALDI imaging is a technique with increasing importance for marker discovery and in clinical research. The sample preparation, especially the application of matrix onto the sample is of utmost importance on the quality of the results. The main parameters in judging the results are the achievable spatial resolution and the spectra quality-unfortunately these parameters seem to be negatively correlated. Here, various preparation methods, such as pneumatic spray, robotic and manual spotting were evaluated. The images are usually analysed visualizing the spatial resolution of selected peaks on an image. The data however contain more complex information, such as overall changes in the proteomic pattern. Here we discuss how such information can be tapped. The data quality is directly correlated with the time of solvent exposure and the amount of matrix. To achieve good quality of spectra at maximal spatial resolution it proved necessary to optimize preparation parameters such as the number of repeated matrix applications, solvent composition and matrix concentration.

When tissues were compared that contained only homogenous tumor or non-tumor regions the simple presence or absence of a peak was not sufficient to classify the tissue because of the lack of an “internal control”. Therefore the data evaluation of the several thousands of spectra was performed by chemometric, multivariate statistical models to distinguish between various tissue areas and to locate specific outlier compounds.

It was not only possible to distinguish tumor from non-tumor areas, but also between tumors of different types. Moreover, specific known biomarkers could be allocated which may de¬termine the further patient therapy.

Human tissue sections of different types of cancer were pre¬pared on a microtome, transferred onto electrically conductive glass slides were used for the experiments. Images were acquired with a MALDI-TOF/TOF instrument equipped with a 200 Hz laser with changeable focus size (~120 µm to 10 µm). Data evaluation was done using dedicated software packages.