Counting Gene Expressing Nuclei in Whole Drosophila Blastoderm Embryos.
Life Sciences and Genomics Divisions, Lawrence Berkeley National Laboratory, Berkeley, CA.
Animal development is dependent on complex temporal and spatial patterns of gene expression that specificy morphology and tissue identity. To fully understand this process, we need to characterize the expression of large number of genes during development in three-dimensions with cellular resolution. To make such analyses feasible, gene expression patterns must be quantitative and reduced to a computable form. The Berkeley Drosophila Transcription Network Project is developing a set methods to do this. Our goal is to build an expression atlas that will record the expression of 1,000 genes in wild type pregastrula embryos and up to 200 genes in a series of mutant embryos, each mutant for one of 34 early acting transcription factor. We have adapted fluorescent in situ hybridization protocols for this purpose. Stained and mounted blastoderm embryos are imaged whole by multiphoton confocal microscopy. One fluorescent channel is reserved for a DNA-stain to detect nuclei; the other two channels contain different gene expression patterns detected with tyramide-reactions. The confocal image stacks are then analyzed to yield a 3D computer representation of gene expression around each nucleus (see Luengo Hendriks et al.). By producing data on the numbers and percentages of cells expressing specific genes, we hope to understand how gene expression correlates with the total number of nuclei. As proof of principle, here we report the use of these methods to count the blastoderm nuclei expressing and not expressing the essentially binary ftz and sna mRNA expression patterns in a set of late stage 5 embryos. We are now expanding our analysis and methodology to describe the spatial co-expression between other pairs of genes, including eve, hb and kni.
Last modified April 9, 2005.