Use of differentiating adult stem cells (marrow stromal cells) to identify new downstream target genes for transcription factors.

We developed a strategy for use of microarray data to rapidly identify new downstream targets of transcription factors known to drive differentiation by following the time courses of gene expression as a relatively homogeneous population of stem/progenitor cells are differentiated to multiple phenotypes. Microarray assays were used to follow the differentiation of human marrow stromal cells (MSCs) into chondrocytes or adipocytes in three different experimental conditions. The steps of the analysis were the following: (a) hierarchical clustering was used to define groups of similarly behaving genes in each experiment, (b) candidates for new downstream targets of transcription factors that drive differentiation were then identified as genes that were consistently co-expressed with known downstream target genes of the transcription factors, and (c) the list of candidate new target genes was refined by identifying genes whose signal intensities showed a highly significant linear regression with the signal intensities of the known targets in all the data sets. Analysis of the data identified multiple new candidates for downstream targets for SOX9, SOX5, CCAAT/enhancer binding protein (C/EBP)-alpha, and peroxisome proliferator-activated receptor (PPAR)-gamma. To validate the analysis, we demonstrated that PPAR-gamma protein specifically bound to the promoters of four new targets identified in the analyses. The same multistep analysis can be used to identify new downstream targets of transcription factors in other systems. Also, the same analysis should make it possible to use MSCs from bone marrow to define new mutations that alter chondogenesis or adipogenesis in patients with a variety of syndromes.