The differentiation potential of stem cells from different origins: Osteogenic differentiation from iPSCs, ESCs, and BMSC-derived mesenchymal stem cells.
Cellular context affects the way that cells respond to different biomaterials. Although stem cells from different sources retain the capacity to divide and differentiate endlessly, each source imprints subtle changes to cells that affect their differentiation process. To address how stem cells from different sources respond to the same biomaterial, we used data generated from mesenchymal stem cells (MSCs) under differentiation on osteoinductive scaffolds in perfusion bioreactors (de Peppo et al., 2013). These MSCs arose from three different origins: they were differentiated from human induced pluripotent stem cells (hiPSCs) or human embryonic stem cells (hESCs), or they were directly acquired from human bone marrow (BM) (de Peppo et al., 2013). All these different cells were induced to MSCs using serum-supplemented medium for 7 days and then more 10-11 passages (de Peppo et al., 2013). The main focus of that study was proving the feasibility of using hiPSCs for generation of osteoblasts.
We used this dataset to more deeply analyze the transcriptomic differences between osteoblast differentiation of different cells of origin. For hiPSC, hESC and bone marrow cells, we compared gene expression profiles of MSCs before and after differentiation towards osteoblasts. In summary, using the same criteria (FC>2 and adjusted p-value <0.05) we found 416, 869, and 341 differentially expressed genes respectively for hiPSCs, hESCs, and BM-derived MSCs differentiated into osteogenic fate (Fig. S1). Analysis of the common genes in transcriptomic level from all differentiation experiments using osteoinductive scaffolds in perfusion bioreactors showed an up regulation of an osteoblast specific profile as the most significant cell signature (Fig. 3a and 3b, Table S4). Many of these genes are involved in extracellular matrix organization, including COL12A1, COL8A2, and COMP. It has been shown that COL12A1 is involved in regulation of osteoblast differentiation and bone matrix formation (Izu et al., 2011). This suggests that MSCs generated from different stem cells can successfully be committed into osteogenic fate. Despite the similarities between all induced osteoblasts, we found that differentially expressed genes during osteogenic differentiation of ESC derived and iPCS derived MSCs clustered together but were markedly different from osteogenic differentiation with BM origin (Fig. S2).
To get a better understanding of osteogenic differentiation from different cells of origin, we restricted our list to genes that were exclusively upregulated in cells differentiated from hESC, hiPSC, or BM. There were 549, 197, and 134 genes significantly upregulated (FC>2 and p-values <0.05) from hESC-MSC, hiPSC-MSC, and BM-MSC derived osteoblasts, respectively. Gene set enrichment analysis showed genes involved in extracellular matrix organization enriched significantly in iPSC-MSC and ESC-MSC derived osteoblasts, but not in BM-MSC derived osteoblasts (Fig. 3c and 3d). In comparing genes involved in extracellular matrix organization between iPSCs and ESCs, we found higher number of genes with more significantp-value related to this biological term in iPSCs than ESCs. Extracellular matrix as a local microenvironment of osteoblast has significant role in cell motility, communication and shape by providing an anchorage site for cells and storing and presenting cytokines for cell growth (Rozario & DeSimone, 2010).
To provide a better understanding of the different differentiation potentials of MSCs from each origin, we made a gene regulatory network for genes exclusively involved in the differentiation of ESC-MSCs into osteoblasts. We then mapped gene expression profiles of osteoblasts generated from all three cells of origin onto the ESC-MSC constructed gene regulatory network (Fig. 3e, 3f, and 3g). Our results clearly indicated these three cells of origin showed significantly different responses in the gene expression level upon osteogenic differentiation. For example, based on the gene regulatory network analysis, we found the transcription factor MYC to be exclusively down-regulated in the ESC-MSC osteogenic differentiation according to adjusted p-value of the genes. MYC is a master regulator of cell proliferation and stemness, so this suggests that the ESC-MSC derived osetoblasts were more differentiated than the iPSC and BM derived MSCs. Additionally, 76% of down-regulated genes in the ESC-MSC network are controlled by MYC, further suggesting that the entire MYC pathway is dampened in the ESC-MSC derived osteoblasts. Network based ontology revealed biological processes related to cell cycle, mitochondrial and metabolism as most significant processes (Fig. 3h).
Our results clearly showed that MSCs originated from different cells of origin were transcriptionally differently respond to the osteogenic differentiation. Despite the fact that all three stem cell populations could differentiate into osteogenic lineage, it is evident that each cell of origin resulted in a moderately distinct transcriptional profile of the differentiated cell. Overall, through bioinformatic analysis of biomaterial-specific responses, as well as analysis of cell-of-origin based response, we highlight that both of these modalities are instrumental in the optimization of targeted stem cell differentiation.