Stimulatory Effects of Serum for Ex-Vivo Culture of Mesenchymal Stromal Cells Can be a Parameter for Heterogeneity in Hematopoietic Supporting Activity

Although hematopoietic supporting activity of mesenchymal stromal cells (MSCs) have been widely used in clinical trials to facilitate the hematopoietic recovery, variations have been observedin supporting effects. In the present study, we show that the niche activity of MSCs is regulated by stimulatory effects of fetal bovine serum during ex-vivo culture of MSCs.First, we observed extensive variations in the numbers of colonies (CFU-F)(up to 40-folds) differences in numbers when human bone marrow (BM)-derived MSCs were cultured under different batches of fetal bovine serum (FBS),and two batches with high (SS) and low (NSS) stimulating effects were selected. 5 independent BM-MSCs cultured with stimulating serum exhibitedsignificantly lower doubling times, higher numbers of CFU-F and higher % of CD146+ cells. When umbilical cord blood-derived CD34+ cells were co-cultured on each MSC group, significantly higher expansion of CD34+CD90+ cells and higher numbers of colonies were observed than those cultured with non-stimulatory serum. Similarly, numbers of primitive hematopoietic cells analyzed by long-term cultureinitiating cells (LTC-IC) were significantly higher in stimulating serum conditions than in non-stimulating condition. Furthermore, MSCs under stimulatory serum exhibited higher level expression of niche cross-talk molecules (CXCL-12 and Jagged-1) than MSCs under non-stimulatory condition, indicating higher support of HSC self-renewal. Whenexamined for murine model, the stimulatory serum caused similar difference in colonogenic activity (2.5-folds higher CFU-F) and phenotypic marker for mesenchymal progenitors (PDGFR-alpha+ cells). Moreover, when each group MSCs were co-transplanted with HSCs into congenic recipient, significantly higher reconstitution of donor-derived cells were observed without changes in the lympho-myeloid lineage distribution in the MSCs with stimulatory serum group than in MSCs with non-stimulatory group (51.2% vs 8.2%, respectively, p<0.005). In contrast, priming of HSCs with 2hr pre-incubation with each group MSCs before transplantation or co-transplantation of MSCs cultured under hypoxic conditions did not exhibit additional increase in the reconstitution level by donor-derived cells.

Of note, the enhancing effects of stimulatory serum on MSCs or on the co-cultured HSCs were reversed when the MSC culture conditions were switched to non-stimulatory serum, indicating that the enhancing effects were caused by extrinsic stimulating effects rather than clonal difference of MSC subpopulation. To explore the differential signals between MSCs grown under stimulatory and non-stimulatory condition, microarray-based analysis of gene expression were analyzed. The gene ontology (GO) analysis showed that extensive signals related to cell growth and survival was enriched in the MSCs grown in stimulatory conditions compared to those in non-stimulatory conditions. Furthermore, upstream signal analysis showed that signals for P53, TRIB3, TGF-beta1, RABL6 or ATF4 were significantly up-regulated in the MSCs under stimulatory serum condition. Taken together, our study shows that MSCs can exhibit significant heterogeneity in the hematopoietic niche activity depending on the stimulatory effects of the serum during ex-vivo culture to cause difference in hematological recovery, and suggest that optimization of CFU-F-based screening of MSCs can produce more predictable outcomes in clinical trials.