The nematode Caenorhabditis elegans is a widely used model system for developmental studies. The major tissues of complex metazoans, (muscle, intestine, nervous system, skin, and so on) are represented in the worm, but the entire animal is composed of fewer than 1,000 somatic cells. Owing to this simplicity and to the rapid development of the C. elegans body plan, the anatomy of every adult cell has been described and the patterns of division giving rise to each one are known [1,2]. The C. elegans genome is fully sequenced [3,4] and encodes over 20,000 predicted genes. Thus, C. elegans offers a unique opportunity to identify specific combinations of genes that define the differentiation and structure of specific cell types. In principle, microarray profiles can provide this information. In order to implement this strategy, however, the small size of C. elegans (length = 1 mm) has required the development of specialized methods for extracting mRNA from specific cell types. In one approach, MAPCeL (micro-array profiling of C. elegans cells), green-fluorescent protein (GFP)labeled cells are isolated by fluorescence activated cell sorting (FACS) from preparations of dissociated embryonic cells [5]. This method has now been used to profile global gene expression in specific subsets of neurons and muscle cells [5-10] (RMF, DMM, unpublished data). An alternative technique, mRNA-tagging [11], can be utilized to profile larval cells, which are not readily accessible for FACS [12]. In this approach, an epitopetagged mRNA binding protein (FLAG-PAB) is expressed transgenically with a specific promoter (Figure 1). FLAG-PAB-bound transcripts are then immunoprecipitated for microarray analysis. mRNA-tagging profiles have been reported for two major tissues, body wall muscles and the intestine [11,13].
Here we apply the MAPCeL and mRNA-tagging strategies to provide a comprehensive picture of gene expression in the embryonic and larval nervous systems. This analysis reveals approximately 2,500 transcripts that are significantly elevated in neurons versus other C. elegans cell types during these developmental periods. The enrichment in these datasets of transcripts known to be expressed in neurons, as well as newly created GFP reporters from previously uncharacterized genes in these lists, confirmed the tissue specificity of our results. The 'pan-neural' transcripts detected in these datasets encode proteins with a wide array of molecular functions, including ion channels, neurotransmitter receptors and transcription factors. Overall, 56% of these C. elegans genes are conserved in humans. The discovery of 27 uncharacterized human homologs enriched in both embryonic and larval neurons suggests that these profiles have uncovered novel genes with potentially conserved function in the nervous system.
In order to identify transcripts that are selectively expressed in a specific neural cell type, we used the mRNA-tagging strategy to fingerprint a subset of motor neurons (A-class) in the ventral nerve cord of L2 stage larvae. This A-class dataset contains around 400 significantly enriched genes. Approximately 25% of these transcripts are not detected in the profile of the entire nervous system. This finding suggests that individual neurons may express rare transcripts that are likely to be restricted to specific neuron types. The application of the mRNA-tagging strategy to profile a specific class of larval neurons complements earlier work in which this method was used to profile larval ciliated neurons [14] and also experiments in which MAPCeL and other FACS-based approaches have been applied to selected embryonic neurons [5-10]. Thus, this work demonstrates the utility of complementary profiling strategies that can now be applied to catalog gene expression in specific C. elegans neurons throughout development.
Answers to all your Biology Questions
