Developing new compounds and therapeutics requires in-depth understanding of an organism’s metabolism – something which can also be very useful for diagnosing cancer. But traditional techniques such as computerized tomography and histological staining methods only produce anatomical images, with little or no information on metabolism. The best way to fully understand an organism’s metabolism is to use a complete protein spectrum, obtained via mass spectrometry (MS). However, conventional MS methods provide no information on the spatial distribution of proteins. The Steinbeis Innovation Center SCiLS (Scientific Computing in Life Sciences) in Bremen is working on the development of mass spectrometry methods that deliver 3D images.
The development of matrix-assisted laser desorption/ionization (MALDI) almost ten years ago was a major step forward in expanding mass spectrometry to a spatial imaging method. This technique made it possible to select individual points in tissue to a high degree of accuracy and record their mass spectrum. MALDI imaging was the first mass spectrometry method capable of delivering information on the protein composition of individual tissue sections in 2D spatial resolution, thus providing detailed spatial information on metabolism.
The data set that makes up a single MALDI image contains around 108 recorded values. Processing these values requires highly specialized automatic visualization and evaluation routines. In partnership with the company Bruker Daltonik, the Steinbeis Innovation Center SCiLS has developed a new method that divides 2D cross-sections into segments where similar metabolic processes are occurring. This makes it possible to identify proteins in the tissue and consider them during cancer diagnosis, for example. The new method is based on a denoising technique that takes local data into account – a mathematical method for image processing.
The next challenge being addressed by the researchers at SCiLS is extending the 2D MALDI technique to three spatial dimensions. To do this, the center is currently developing technical process chains in a joint project with the Fraunhofer MEVIS Institute for Medical Image Computing and Bruker Daltonik. 3D MALDI imaging will make it possible to record and analyze the protein spectrum of an entire organ or lesion of diseased tissue, in all its complexity. This will enable specialists to directly investigate key oncological issues that can only be understood in the context of highly complex (heterogeneous) 3D tissue. This includes the distribution and metabolism of active ingredients in tumor tissues that have undergone highly complex pathological change, plus the response of these tissues to a treatment. For the first time, 3D MALDI imaging in organs and tissues would allow direct systematic analysis in these areas.
Adding a third dimension to the MALDI technique results in data sets of approximately 1010 recorded values. From a technical perspective, visualizing this 3D metabolic information is highly complex – while from a medical perspective, it still does not provide enough information for diagnosis. To make sense of the data, it must first be correlated with 3D anatomical information (such as data obtained via computerized tomography). However, superimposing these two data sets, which were generated using entirely different measurement techniques, is complicated by the issue of image registration. The final result is a high dimensional image that combines the data from both imaging methods and visualizes both anatomy and metabolism.
The project is set to run until June 2012 and is funded by Bremen Economic Development (WFB). As well as the three main partners – MEVIS Fraunhofer Institute, Bruker Daltonik and the Steinbeis Innovation Center SCiLS – doctors from Helmholtz Zentrum Munich are also acting as an external advisory board for the project.
Dr. Dennis Trede
Prof. Dr. Peter Maass
Steinbeis Innovation Center SCiLS (Scientific Computing in Life Sciences) (Bremen)
su1412@stw.de