Multicolour 3D-FISH in combination with confocal microscopy, 3D image reconstruction and quantitative image analysis is an efficient tool for the analysis of the 3D genome structure and of the spatial relationship of defined nuclear targets comprising entire chromosome territories down to the level of single gene loci. Until a few years ago the drawback of confocal microscopy was its limitation to three or at maximum to four different fluorochromes that could be visualized simultaneously. Recent developments of a "new generation" of confocal microscopes allow the simultaneous excitation and distinct visualization of five different fluorochromes (the number can be increased if colour unmixing software is used) within one experiment, opening the way for a simultaneous delineation of numerous differently labeled intranuclear targets.
Here we provide protocols for the preparation of complex DNA-probe sets suitable for 3D-FISH with up to six different fluorochromes, protocols for 3D-FISH on cultured mammalian cells (growing in suspension or adherently growing), and protocols for an efficient 3D-FISH on tissue sections, that have all been used successfully by our group. We restrict to protocols describing the labeling of a given DNA probe (such as chromosome specific probes, BACs or plasmids etc.) with a single hapten or fluorochrome. We should mention here that the term M-FISH (which is originally the abbreviation for multiplex (!) FISH and not for multicolour FISH) is often related to the combinatorial labeling of a probe with different, usually two or three fluorochromes/haptens in order to increase the number of distinguishable targets. While this approach has been widely used as a tool for the complex analysis of metaphase chromosomes and interphase cytogenetics, its successful application for 3D-FISH on 3D preserved nuclei in combination with confocal microscopy has been shown only in a few studies (see e.g. Bolzer et al. (2005)). This is mainly due to the fact that analysis of confocal image stacks containing combinatorial labeled probes is highly demanding and requires specialized skills. For the special aspects with regard to the generation of DNA probes by combinatorial labeling we kindly ask you to refer to the papers of Bolzer et al., (2005) and Fauth et al., (2001).
Finally we want to emphasize that multicolour FISH on 3D preserved nuclei is a somewhat delicate method where minor deviations or experimental mistakes can easily change the quality of an experiment. For readers that are interested to concern this technique in more detail we refer to our previous and recent publications (Solovei et al. 2002a; Solovei et al. 2002b; Walter et al. 2006)
AG Thomas Cremer
Department Biology II
Grosshadernerstr. 2 - 82152 Martinsried-Planegg, Germany