We have applied cDNA microarrays containing 13 627 clones to analyze gene expression changes that take place during the in vitro differentiation of neural progenitor cells. The rapid production of arrays was facilitated by applying a robotic platform combined with a cooled microtiter plate storage system. To deal with the limited amount of RNA available, we applied a signal amplification method (Genisphere Inc. (see Protocol 5.3)). This allowed us to minimize cultivation time of primary cells, hybridize each sample twice with dyes swapped, and exclude potential biases in relative transcript abundance that might be introduced by an RNA amplification procedure.

Using a variance estimation we determined that a cut off in expression change for each individual clone set at twofold resulted in 2-5% false positives (see Protocol 5.4). This rate of false positives was further reduced by considering only such clones as relevant, whose expression changed more than twofold in two experimental series and represents another data filtering step.

To gain a more sophisticated insight into the molecular processes underlying the differentiation of neural progenitor cells we collected samples at different time points during the experiment. We followed the changes in gene expression in the cells over a 4-day period. This approach gives a more detailed insight into the molecular mechanisms involved in cellular development in our model system. For an individual gene or a group of genes a time course illustrates the onset and dynamics of expression changes that take place after induction of differentiation. A cluster analysis helps to identify such groups of coregulated genes which might also be involved in shared biological processes (see Protocol 5.5).

The goal of microarray experiments like this one is to learn more about gene expression changes that occur during the switch of cell states. Genes that are down-regulated during differentiation are preferentially expressed in the progenitor cells. They might be relevant for maintaining their self-renewing and differentiation capacity. For example, cluster 9 contains genes with the strongest and the earliest expression changes, and most of them encode cell-cycle proteins, indicating that one of the first events in this in vitro differentiation is cell cycle exit. In contrast, up-regulated genes are likely relevant for the process of differentiation or necessary for a specific function of the differentiated cell. Also, the timing of gene expression changes is definitely important for proper differentiation. Genes that change early, especially those related to transcriptional regulation or signal transduction, are likely to have regulatory functions in the differentiation process. Many genes whose expression changes later encode for products that are important in the differentiated cell.

The heterogeneity of the neurosphere culture complicates the evaluation of the microarray results. Changes in expression of individual genes are unlikely to take place in all cells equally. Instead, the fold changes measured can be due to much stronger changes in gene expression in subpopulations of cells. Therefore, it is desirable to attribute individual gene expression changes to certain cell types, which can be done by immuno-fluorescence using antibodies against corresponding gene products. The purpose of this analysis was to discover genes that are relevant for the maintenance of neural progenitor cells, as well as for the migration and differentiation of their progeny. Having identified candidate genes relevant for these processes and knowing the dynamics of their expression is crucial for further functional analyses that will enhance our understanding of adult neural progenitor cells.

0 0

Post a comment