ID-1-P-3247 Correlative Light and Electron Microscopy – on the way from 2D towards 3D

More and more researchers have been wanting to interlink complementary information using correlative microscopy to gain insights into the interdependency of function and structure. This can involve the combination of any microscopical methods, but usually the term refers to light microscopy (LM) and electron microscopy (EM). The interest in correlative microscopy has been rapidly growing in the last decades and in 2 dimension this workflow is well established because of the ease of use of the provided solutions. For correlation of two microscopical images, regions of interest are specified and imaged in one microscope, which can be relocated easily in a different microscope by using Zeiss “Shuttle & Find”. Afterwards, the images are overlaid. However, there are still a number of challenges that have to be addressed in order to realize the full potential of correlative microscopy. One major challenge is the correlation of 3D data sets. To achieve this, it is necessary to exactly define volumes of interest (VOI) in the data of the first microscope. Further, the precise relocation of the identical VOI in the second microscope is essential as well as the registration of the 3D object in all spatial directions. Even if the correlation of 3-dimensional data from different microscopes (e.g. LSM and FIB-SEM) is feasible due to cross correlation methods it has to be stated that this workflow is not yet fully-automated [1,2]. The 3D workflow can be simplified by reducing the scale of the object in one dimension. One popular approach is to cut the sample into serial sections (correlative array tomography) [3]. Thus, the segmentation in one dimension is done mechanically and only 2-dimensional microscope images have to be correlated. Correlative array tomography allows the detection of fluorescent labels as well as the investigation of the ultrastructure of ultrathin serial sections. Regions of interest can be marked and automatically imaged within all the individual sections building up a long ribbon using a procedure according to the “Shuttle & Find” approach. The challenge of this approach is on one hand the alignment of the consecutive 2D images taken with a light microscope and a scanning electron microscope and on the other hand their subsequent registration to a correlative 3D data set. A comparison of the features in the single sections followed by an alignment of the features results in an accurate alignment of the single sections. Finally, the full volume can be reconstructed by a similar slice-to-slice stack alignment.

References [1] M Lucas et al, Imaging & Microscopy. 10(3) (2008), pp. 30-31. [2] L Blazquez-Llorca et al, J Alzheimers Dis, 34(4) (2013), pp. 995-1013. [3] KD Micheva and SJ Smith, Neuron 55 (2007), pp. 25-36.