In this study, we report that the use of a selective, XF culture medium, specifically designed for MSCs, led to the isolation from human WJ explants and enzymatically dissociated WJ of cell populations with MSC characteristics as assessed at gene, protein and functional levels, which were differently expressed according to the cell isolation method used. Furthermore, we demonstrated that the initial culture conditions, used for the isolation and expansion of WJ-MSCs, resulted in lack of subsequent endothelial differentiation potential. Therefore, the use of defined MSC culture media is a prerequisite for the establishment of the real capacity of WJ-MSCs to differentiate towards the endothelial lineage.
The WJ-derived cells isolated and expanded in a MSC XF medium proved to present a robust MSC profile through expression of CD44/CD73/CD90/CD105 surface markers, multipotent differentiation potential into adipogenic, osteogenic and chondrogenic cell lineages, as well as a high proliferation rate. When BM-MSCs [26, 42, 43], which expressed less growth-related genes as compared to WJ-MSCs , and adipose stem cells  were manipulated in vitro by using XF media, they proliferated more rapidly than the cells cultured by using conventional media. In addition, MSCs derived from enzymatically-digested whole human umbilical cord segments  and umbilical cord matrix  had a higher expansion potential when cultured in serum-free/XF media, as compared to FBS-based media cultured counterparts and BM-MSCs/adipose stem cells, respectively. Although we did not compare in this study XF versus conventional culture media, our results corroborated with the literature data further indicate that the XF conditions described here might bring an advantage over the conventional media, in respect of proliferation properties of WJ-MSCs.
We also demonstrated significant differences in CD44, CD73 and CD90 surface marker expression in WJ-MSCs cultured in XF conditions, as well as in WJ-MSCs and WJ-derived adherent cells exposed to endothelial differentiation conditions, depending on the cell isolation method used or the passage number. It is well known that both in vitro cell manipulation technique  and passage number  exert a prominent impact on the protein expression profile of MSCs. The issue of heterogeneity in MSC markers expression has also been reported for BM-MSCs and has been correlated with their proliferation and differentiation potential . It has been shown that CD44, CD73 and CD90 markers were expressed at higher levels in umbilical cord-derived MSC subpopulations that exhibited a higher proliferation capacity and a reduced amount of aging cells . In addition, CD44, a cell-surface receptor for hyaluronic acid, facilitated MSCs migration through interaction with extracellular hyaluronic acid, suggesting that such migratory mechanism could be critical for MSCs recruitment to tissue regeneration sites . As CD44 and CD90 are involved in cell-to-cell interactions, cell adhesion and migration, and CD73 plays an important role in cell cycle progression and apoptosis, they might differently control these processes in WJ explant- versus pellet-derived cells, nevertheless by passage propagation. Therefore, the biological benefits of such heterogeneity in the level of surface markers’ expression may be related to the proposed functions of these cells. Further insights into these markers’ expression by WJ-MSCs manipulated in XF culture conditions are worthwhile in order to modulate biological processes, such as cell adhesion, migration and proliferation, to optimize clinical cell therapy approaches. Thus, the development of protocols for isolation of distinct MSC subpopulations, based also on the expression levels of their surface markers, may provide improved vectors for the treatment of specific diseases .
When assessing multipotent differentiation outcomes, we demonstrated that in WJ explant- and pellet-derived MSCs, the chondrogenic and osteogenic differentiation potential increased and decreased, respectively, by passage progression. This could be due to the fact that the MSCs isolation method used (via WJ explants versus enzymatic dissociation of WJ) led to the generation of different MSC populations in respect of their multipotent differentiation capacity. It has also been documented that the passage number has an important impact on the differentiation capacity of MSCs [48, 53, 54] due to a consistent pattern of changes in the global gene expression signature of MSCs at different passages . In this respect, MSCs derived from other sources than WJ, either exhibited a decreased osteogenic differentiation capacity by passage progression  or calcium deposition transiently decreased from P4 to P6, but returned to levels near or above those of primary cells by P10 . Therefore, further work is needed in order to elucidate the optimum isolation protocol and passage number of WJ-MSCs, grown in defined, XF conditions, to optimize their multipotent differentiation potential according to the clinical cell therapy foreseen.
Although it has been shown that WJ-MSCs cultured in conventional media presented a lower adipogenic differentiation potential as compared to other MSC sources, such as the BM , periodontal ligament  and chorionic-plate , we found that WJ-MSCs isolated and expanded in the XF medium presented different degrees of adipogenic differentiation, according to the isolation method and passage number; interestingly, the adipogenic differentiation capacity of WJ pellet-derived MSCs was higher at P5 as compared to P2 and was inverse proportional with the osteogenic differentiation potential of these cells. This may be due to the involvement of MicroRNA-22 that was found to regulate adipogenic and osteogenic differentiation of human adipose tissue-derived MSCs in opposite directions , a finding that deserves to be further investigated in WJ-MSCs, too.
While it has been reported that both the osteogenic gene expression pattern  and osteogenic differentiation rate  were lower in WJ-MSCs as compared to BM-MSCs, isolated by using conventional media, we demonstrated a very robust osteogenic differentiation potential of WJ explant-derived MSCs, which increased by passage progression. Furthermore, the influence of the culture conditions used for MSCs isolation and expansion on gene and protein expression profile of BM-MSCs , as well as on their osteoblastic differentiation has been demonstrated. In this regard, XF conditions used for BM-MSCs culture, through FBS substitution for allogeneic human platelet lysate, enhanced their osteogenic differentiation . Hence, the WJ-MSCs isolated and expanded by using the selective, XF conditions described here might be better cell candidates for bone regeneration, as compared to WJ-MSCs manipulated in conventional media, suggesting potential superior in vivo osteogenic regeneration outcomes. In respect of the impact of the XF conditions on chondrogenic differentiation potential, it has been shown that BM-MSCs that were isolated, stored and expanded using XF materials, including the XF medium used in our study, had a higher gene expression of aggrecan than cells cultured in conventional media . These observations indicate that WJ-MSCs, manipulated in vitro by using XF media, might also represent superior cell candidates for cartilage repair.
Upon exposure to endothelial differentiation signals, we showed that both WJ explant- and pellet-derived cells from Group A - isolated and expanded until P2 in MSC XF medium, and then subcultured for five passages in endothelial differentiation conditions - did not exhibit endothelial differentiation potential. By contrast, pellet- but not explant-derived cells belonging to Group B - isolated and expanded in endothelial differentiation conditions from P0 to P5 - presented endothelial cell characteristics, being different in respect to surface and gene marker expression, as well as functional properties. This might be due to the fact that the enzymatically dissociated, freshly isolated WJ cell populations contained stem/progenitor cells or other contaminating endothelial cell types able to give rise to an endothelial progeny, when seeded directly into endothelium medium for five passages; in the case of explant-derived cells of group B, these cells may not migrate out of the explants. Furthermore, they may be lost when initially cultured in XF conditions.
Interestingly, we also demonstrated that, besides the endothelial markers expression at gene and protein levels, the WJ pellet-derived cells belonging to group B presented CD73 and CD44 MSC surface markers. Although there are no data yet documenting CD44 marker expression on endothelial cell outgrowth derived from WJ, it has been reported the involvement of CD44 molecule in endothelial cell proliferation, migration and angiogenesis , contributing to the organization and/or stability of developing endothelial tubular networks . While there is also no evidence of CD73 expression on endothelial progenitors derived from any adult tissues, Choi KD et al.  identified a novel population of CD73+ endothelial progenitors derived from human embryonic stem cells. These data and the fact that the umbilical cord is closer in development to embryonic than adult tissues [44, 57], reinforce the speculation that the isolated endothelial progeny within pellet-derived cells belonging to group B resulted from a different stem/progenitor cell type than MSCs, residing in the matrix of the umbilical cord, whose isolation might have been facilitated by the use of enzymatic dissociation and initial endothelial culture conditions.
Although some reports have demonstrated endothelial differentiation of WJ-derived cells in vitro or in vivo[13, 16–19], endothelial differentiation of WJ-MSCs is still very controversial. Furthermore, it has been shown that MSCs derived from other tissues than WJ, such as BM and adipose tissue, did not display functional properties of endothelial cells, but promoted neovascularization via paracrine mechanisms [63, 64]. Choi M et al. also demonstrated the lack of endothelial differentiation capacity of WJ-MSCs, isolated and expanded in a conventional FBS-based medium, but paying careful attention to the removal of the umbilical cord blood vessels and amnion. Upon exposure to endothelial differentiation media, these WJ-MSCs neither expressed endothelial markers, nor directly participated to angiogenesis/vasculogenesis in vitro and in vivo; rather, the cells improved perfusion recovery and neovascularization by secreting paracrine factors and by functioning as perivascular precursor cells . Our data, corroborated to this report, suggest that the MSCs’ isolation method has a tremendous impact on the homogeneity of the cells and their true endothelial differentiation capacity. By contrast with other authors that did not use a selective MSC medium for WJ-MSCs isolation prior to their exposure to endothelial differentiation conditions  or cultured freshly isolated cells directly in endothelial differentiation media , we further documented, by using a defined, XF medium for MSCs, that the initial culture conditions have a strong impact on the progeny outgrowth resulted upon exposure of WJ-MSCs to endothelial differentiation signals. Based on our observations, we speculate that in the case of previous reports, describing WJ-MSCs differentiation into endothelial phenotypes, the endothelial outgrowth was derived from other types of stem cells than WJ-MSCs, circulating endothelial progenitors, or umbilical vein endothelial cells, contaminating the primary cultures during the dissection process, performed to remove the umbilical cord arteries and the vein.
It is well known that gene expression profiling is significantly different among MSCs from distinct sources and that intrinsic gene expression has an important impact on MSCs overall differentiation potential . It has been postulated that BM-MSCs constitutively expressed genes related to immunomodulation, adipose tissue derived-MSCs highly expressed genes implicated in tissue development , whereas the transcriptome profiling of WJ-MSCs revealed an increased expression of genes involved in liver development  and an inherent bias towards the neuro-ectoderm lineage [14, 67]. These observations suggest that WJ-MSCs might be better cell candidates for differentiation into hepatocyte and neuronal rather than endothelial phenotypes. As MSCs from different sources exhibit distinct and unique gene expression signatures, which make them competent to give rise to specific lineages rather than others , at this moment there are no well-defined master MSCs that are suitable for vascular regeneration. Therefore, according to the isolation and culture methods used, MSCs should be rigorously characterized, especially at the gene and functional levels, and much caution needs to be given when choosing the best MSC type for a particular therapeutic application.