Tales from the Crypt: Mesenchymal Stem Cells for Replenishing the Intestinal Stem Cell Pool

Studies have shown that Mesenchymal Stem Cells (MSC) contribute to replenishing the radiation-depleted intestinal stem cell (ISC) compartment after abdominal/pelvic radiotherapy. Coupling these findings with other encouraging results showing that MSC promotes healing in inflammatory bowel disease suggests that MSC may represent an attractive source for regeneration of gastrointestinal tissue. Nevertheless, controversy regarding the true contribution of MSC to the cellular pool of an injured organ versus promotion of healing through production of growth factors/cytokines raised questions regarding the true potential of these cells. Since MSC are known to be heterogeneous and some populations may represent tissue-specific stem cells or cells which have already committed to differentiation to a specific phenotype, it is important to find a means of identifying and enriching for subpopulations that are best suited for intestinal cell therapy. Thus, we examined the in vivo ability of clonally-derived MSC populations to contribute to the stem or mature epithelial pool of the intestine, and correlated these results with genomic analysis in an attempt to find a specific marker that enriches for such an MSC population. BM-derived Stro-1+ MSC clones were established by single cell deposition. Phenotypic characterization of the various clones showed that, despite many similarities, there was phenotypic variance between clones for some commonly expressed stem cell proteins, such as CD44 and c-kit. 6 of the clonally-derived MSC were transplanted into 55–60 day-old fetal sheep at a concentration of 10⁶cells/fetus, and evaluated the recipient intestine at 2.5–3 months post-transplant for the presence of human organ-specific cells using fluorescent in situ hybridization and immunofluorescence. We found that MSC localized preferentially to the mucosal layer, distributed equally between the crypts and the villi, with no evidence of cell fusion. While 4 of the 6 clones had a similar level of intestinal engraftment of 1.5±0.4%, 2 of the clones exhibited enhanced engraftment of 5±0.65%. Dual staining with antibodies to cytokeratin and the ISC marker Musashi showed a direct correlation between the levels of donor contribution to the ISC pool and the levels of mature enterocytes within the villi, such that 1–4% of the Musashi+ cells within the crypt were donor-derived with the 4 clones exhibiting overall intestinal engraftment levels of 1.5%, whereas 12–14% of the total Musashi+ ISC population was donor-derived with the 2 clones exhibiting enhanced levels of intestinal engraftment. Since no correlation was found between the enhanced levels of engraftment and any of the phenotypic markers tested prior to transplant, transcriptome analysis using Affymetrix microarrays, and GeneSifter software was done in the hopes of identifying a marker that would allow selection of these cells. MSC did not express Mushashi, cytokeratin 2,18,19,20, or GPR49 prior to transplant, precluding their use for selection for transplant. MSC expressed calmodulin, caldesmon, B-catenin and cytovillin, proteins expressed by intestinal crypt cells. Unfortunately, the cytoplasmic location of these antigens prevents their use for selection. Nevertheless, EphrinB-2, a receptor expressed at high levels on ISC, was also found to be expressed on MSC. Studies are underway to confirm whether levels of EphrinB-2 correlate with the differentiation potential of the clones. Our studies demonstrate that MSC subsets have the ability to repopulate the ISC pool and generate differentiated intestinal cells and that combining in vivo functionality studies with genomic analysis may lead to identification of markers for enrichment of MSC that exhibit desired traits upon transplantation.