Molecular Programming of Perivascular Stem Cell Precursors

Pericytes have been shown to act as precursors of resident adult stem cells in stromal tissues in vivo. When expanded in vitro these cells are capable of giving rise to multiple mesenchymal cell types, irrespective of their tissue of origin. This phenomenon of multi-lineage differentiation is only observed in culture, whereas in vivo, stromal stem cell differentiation is restricted to tissue-specific cell types. An important unanswered question is how a single, widely distributed cell type (a pericyte) gives rise to stem cells with tissue-specific functions and attributes. Using a combination of transcriptomics and epigenomics we have compared the molecular status of two populations of stromal stem cell precursors. Using a LacZ transgene insertion that is expressed in pericytes but not in stem cells, we were able to compare pericyte populations from two different tissues, mouse incisors and bone marrow. Pericytes, freshly isolated from mouse incisors and bone marrow, exhibited transcriptomes and epigenetic landscapes that were extensively different, reflecting their tissue of origin and future in vivo differentiation potential. Dspp, an odontoblast differentiation gene, as well as additional odontogenic genes, are shown to be expressed in dental pulp-derived pericytes. These genetic loci are also decorated with histone modifications indicative of a transcriptionally active chromatin state. In bone marrow pericytes, a major osteogenic differentiation gene, Runx2, is not expressed but is marked by both active and repressive histones and therefore primed to be expressed. Polycomb repressor complex 1 analysis showed that key genes involved in the induction of adipogenesis, chondrogenesis, and myogenesis are targeted by Ring1b and therefore stably repressed. This indicates that pericyte populations are molecularly obstructed from differentiating down certain lineages in vivo.