Previous studies have shown that NCAM is a modulator of various cellular functions such as cell–cell and cell–matrix interactions, neuronal migration, neurite outgrowth, and synapse formation [3, 38, 39]. Furthermore, it is an important regulator of cell migration and condensation during skeletal development [1618]. It is noteworthy that NCAM is not merely an adhesion molecule, but initiates the formation of intracellular membrane-proximal signaling complexes, thereby activating a complex network of signal transduction [40]. In addition, polySia affects cellular functions such as migration, cytokine response, cell contact-dependent differentiation, and immune response modulation [4144].

To date, definite information about NCAM and polySia expression in hBM-MSCs has been lacking. Because of the intrinsic heterogeneity of the MSC populations, donor variation, and diversity in culture conditions and analysis methods, biomolecular and cytometric characterization of MSCs from different laboratories is not easy to compare [2]. Furthermore, transcriptomic profiling has become a popular method for characterizing MSCs, but protein and mRNA expression levels do not always correlate. Analysis at the protein level is therefore needed before conclusions about the MSC phenotype can be drawn.

To gain more insight into the characteristics of xenofree hBM-MSCs, we have utilized an established clinical-grade culture protocol based on human platelet lysate [32]. Platelet lysate has been approved as a safe and effective supplement for MSC cultivation in vitro [45]. Conventional cell culture methods involve many animal-derived components; however, they are not desirable for clinical-grade cell production because of the increased risk of cross-contamination and host immune reactions [46]. In addition, xenogeneic additives, like bovine sera, may negatively alter the self-renewal and stemness of hBM-MSCs [47]. Because MSCs are very rare in the bone marrow, isolation and in-vitro expansion of the cells is usually required prior to their use. We analyzed noncryopreserved cells at passage 2 or 3, which is the time point of choice for most MSC applications because it offers the minimum required number of cells that still hold functional potency [48, 49]. NCAM and polySia expression is possibly altered during in-vitro culture. However, analysis of the properties of these cells directly from the bone marrow would be challenging because reliable markers for their identification are lacking. Culture of MSCs for at least two passages is commonly used to attain population purity [48].

Our findings regarding NCAM expression in hBM-MSCs differ from those reported previously [1925], and show—in contrast to the generally held view—that the cells in fact do express NCAM. Furthermore, we conducted a more detailed analysis regarding NCAM gene and protein expression. Our data show that all five known NCAM isoforms are transcribed in hBM-MSCs. In particular, the main isoforms are detected at the protein level. NCAM protein is expressed throughout the cell surface in a clustered manner. However, flow cytometric analysis revealed quite broad donor-specific variation in expression levels between the hBM-MSC lines. This is not unexpected even in our relatively homogeneous donor population (healthy females, age 21–31 years), because it has been reported previously that hBM-MSC cultures are heterogeneous mixtures of cells, the properties and potency of which vary greatly between individual donors independent of age or gender [5052]. In this study, the donor population was not selected based on any specific characteristic, but samples were obtained from the volunteer donors in the order they came in based on the national guidelines for bone marrow donation eligibility. In Finland, as in most other western countries, the majority of bone marrow donors are female [53]. In general, gender has little effect on hBM-MSC features [50, 54, 55].

There are many possible reasons for the conflicting reports about NCAM expression. MSCs are cultured under various conditions (e.g., FBS, human serum, platelet lysate, or serum free) and diverse procedures are employed in their management. Donor-dependent variation may also occur. Expression patterns may be regulated temporally and the cells may be in different stages and passages at the time of analysis, or have undergone replicative senescence. Also, the cellular phenotype may alter between fresh and cryopreserved cells. Technical variation may also be responsible; for example, the variable sensitivity and isoform specificity of anti-NCAM antibodies may give rise to misleadingly low or lacking protein expression. Furthermore, the expression of NCAM mRNA transcripts does not necessarily correlate with the expression of protein on the cell surface.

For comparison, the hBM-MSC lines were also analyzed by flow cytometry for the surface expression of CD44 and TNAP. CD44, a receptor for hyaluronic acid and a common MSC marker, is involved in the contact between stem cells and the niche for stemness maintenance, as well as MSC homing [56, 57]. All five hBM-MSC lines expressed high levels of CD44 (>99 %), as expected. TNAP is an early osteogenic marker that is expressed in a stage-specific manner during skeletal development [58]. Furthermore, TNAP deficiency causes bone hypomineralization, abnormalities in brain development, cortical malformations, as well as epileptic seizures [59]. Thus, TNAP and NCAM are developmentally involved in many of the same processes. However, no correlation between TNAP and NCAM expression was observed in the hBM-MSC lines. Also, it has been reported previously that TNAP expression may vary greatly between individual donors [50, 60] and our results further support this finding.

To our knowledge, polySia expression in hBM-MSCs has not been reported previously. Our results show that both polysialyltransferases, ST8SIA2 and ST8SIA4, catalyzing polySia synthesis are transcribed in these cells. However, very few cells express polySia on the cell surface. On closer examination it was perceived that polySia is expressed mostly on the cell extensions, in accordance with its natural role as a promoter of cell projection outgrowth and targeting [61]. Difference in polysialyltransferase expression and polysialylation levels is an interesting finding, because traditionally it is thought that expression of polySia correlates with transcription of polysialyltransferases [62, 63]. However, it has been previously shown that other, calcium-dependent, nontranscriptional regulatory pathways also exist [64]. Such nontranscriptional regulation may be due to the spatiotemporal nature of polySia, requiring specific cues for prompt and selective expression on the cell surface [65, 66].

Different cell types express different glycan signatures, a property which has also been utilized to identify and purify stem cells [67]. For example, the glycolipids SSEA-3 and SSEA-4 are amongst the most commonly used markers to identify embryonic stem cells; however, they are not necessary for the maintenance of pluripotency [68]. It is well known that expression of polysialylated NCAM decreases during postnatal development and mostly unpolysialylated NCAM is expressed in adult tissues, where it regulates cell interactions independent of polySia [69, 70]. In addition, a recent study shows that polysialylation is regulating human pluripotent stem cell differentiation into the three germ layers [63]. In the mesoderm, ST8SIA4 is the principle polysialyltransferase under normal conditions, but this switches to ST8SIA2 when ST8SIA4 activity is eliminated [63]. The observed expression pattern of polysialyltransferases and restricted polysialylation may thus indicate that hBM-MSCs are different from their prenatal pluripotent counterparts. However, our differentiation results evidently demonstrate that these cells still possess multilineage differentiation capacity. Furthermore, the uncovered polySia and NCAM expression may provide novel targets to modify MSC function [71].