**1. Introduction**

The mesenchymal stem/stromal cells (MSCs) represent an adult stem cell (SCs) population that exhibits tri-lineage di fferentiation potential [1]. The MSCs can be isolated from various tissues, including bone marrow (BM), umbilical cord blood and Wharton's jelly of the umbilical cord (UC), adipose tissue, peripheral blood, synovium or dental pulp [2]. For distinguishing better between MSCs of various origins, the International Society for Cellular Therapy (ISCT) proposed minimal criteria for defining MSCs in 2006 that included the following: (i) an ability to adhere to plastic surfaces under standard culture conditions in vitro; (ii) a specific antigenic profile including at least 95% positivity for CD105, CD73 and CD90 antigen expression with parallel negativity (less than 2%) for CD45, CD34, CD14 or CD11b, CD79 α or CD19 and HLA-DR), and (iii) a capacity to di fferentiate into the cells of mesodermal lineages, such as osteoblasts, chondroblasts and adipocytes in vitro [3]. Moreover, MSCs derived from various tissues may cross the boundaries of the germ layers and di fferentiate into cardiomyocyte- or endothelial-like cells (with parallel ability to form capillary-like structures on Matrigel) upon exposure to lineage-specific growth factor cocktails [4–7]. Recent evidence indicates that murine BM-MSCs may di fferentiate into hepatocytes, lung epithelial cells, myofibroblasts, and renal tubular cells in vivo [8]. Thus, based on their wide di fferentiation potential and their pro-regenerative potential in injured tissues, MSCs are promising "candidates" as medicinal products employed in in vivo applications for advanced therapy in humans. MSCs have been widely evaluated in preclinical studies as well as with a multitude of clinical trials focused on several tissue injuries, including bone defects [9] and other defects such as cartilage injury [10]. Of note, recent evidence demonstrates the biological variability (including di fferentiation capacity) within the MSCs of various origins [11]. Thus, the selection of the appropriate source of MSCs might be critical for the e ffective regeneration of injured tissues following their application in vivo.

BM-derived MSCs represent the gold standard for MSCs used in research and clinical applications (predominantly in autologous transplantations) in recent years [12]. However, the application of autologous MSCs may have some limitations. BM-derived MSCs isolated from elderly patients exhibit decreased biological activity, including osteogenic potential [13], which may result in limited outcomes of the treatment with these cells. Moreover, MSCs isolated from the tissues of patients su ffering from chronic conditions like diabetes or systemic lupus erythematosus may also exhibit altered biological properties impairing their pro-reparative functions in autologous applications [14,15]. Nowadays, growing evidence indicates that umbilical cord tissue represents an attractive alternative source of MSCs for allogenic applications, which eliminates the limitations of autologous MSCs as well as the need for the invasive procedure of tissue harvesting that is required for BM [16,17]. Moreover, UC tissue possesses higher accessibility and lower ethical concerns in comparison with BM. Importantly, UC-MSCs have been demonstrated to exhibit similar transcriptomic profiles with respect to the expression of stemness- and bone development-related genes to BM-MSCs (harvested from healthy donors), demonstrating the grea<sup>t</sup> promise for umbilical cord tissue as an alternative source of MSCs [17].

In 2000, Gronthos et al. described a unique population of SCs residing in human dental pulp referred to as dental pulp stem cells (DPSCs). They exhibit mesenchymal characteristics, including adhesion to plastic surfaces, fibroblast-like morphology, lack of expression of CD14, CD34, and CD45, as well as the potential to di fferentiate into osteoblasts [18]. Moreover, DPSCs show antigenic profile similar to BM-MSCs, along with greater proliferation capacity when compared to that of BM-MSCs [18]. Although DPSCs predominantly originate from the neural crest, which is an ectodermal structure, they are considered as MSCs or MSC-like cells [19–21]. Thus, the unique developmental origin of DPSCs may indicate their ability to produce neural cells and their impact on neural tissue regeneration, conferring certain advantages in comparison with other sources of MSCs [22–26]. DPSCs exhibit capability not only to di fferentiate into neural-like cells [25] but also to secret factors enhancing endogenous repair mechanisms [27]. The unique features of DPSCs resulted in the first clinical application for the treatment of neurodegenerative diseases [28]. Moreover, DPSCs, when cultured under appropriate conditions, are known to exhibit osteogenic potential [18,29]. While long bones

develop from mesoderm during the process of endochondral ossification, flat bones of the skull, similar to the mandible, are formed during the intramembranous ossification from ectomesenchymal cells derived from neural crest SCs [30,31]. Once DPSCs were identified as MSCs derived from the neural crest, they were recognized as a potential promising candidate for the repair of defects of the jaws [20]. Interestingly, DPSCs may also be used in the innovative strategies for dentin regeneration [32].

Following numerous clinical trials employing MSCs of various origins, some concerns about their e fficacy and safety have been raised [33]. In many cases, results of clinical applications were less favourable than expected, with the observation of certain side e ffects upon the administration of MSCs into a suboptimal niche [33]. One of the reasons for the limited pro-regenerative capacity of MSCs may be the lack of well-established protocols for MSC isolation, expansion, and preparation of a cell therapy product for in vivo applications [33] as well as the use of an inappropriate cell source or di fferentiation strategy [32]. Importantly, the standard culture conditions with plastic or glass surfaces (representing two-dimensional culture, 2D) do not fully mimic the three-dimensional (3D) microarchitecture of SC niches [34]. In this context, the use of novel 3D sca ffolds (created by f.e. hydrogels) for cells ex vivo propagation, di fferentiation, or as a carrier for the administration of cells, provides a new exciting approach to enhance the biological potential of MSCs for the repair of the injured tissue. MSCs encapsulated in alginate beads exhibit greater osteogenic di fferentiation potential than MSCs in traditional cell culture, which was confirmed by the higher expression of mRNAs corresponding to osteoblast-specific genes (e.g., *Runx2, Col1A1*, sclerostin or dental matrix protein-1) [35]. Similar observations have been described for DPSCs di fferentiated in sca ffolds, such as Matrigel or collagen-based sponges, in vitro [36].

Oxygen (O2) concentration in the culture environment represents another important factor that should be taken into consideration during the di fferentiation of MSCs both in vitro and in vivo. It has been established that the normoxic environment does not reflect the oxygen concentration in SC niches [37]. Importantly, in numerous reports, MSCs were shown to proliferate faster when cultured in hypoxic conditions with 1% [38], 2% [39], or 3% O2 [40], compared to that in the normoxic atmosphere. In contrast, a decrease in the proliferation of rat BM-MSCs in the presence of 1% O2 was reported owing to the upregulation of p27 protein correlating with the downregulation of cyclin D expression [41]. Hsu et al. demonstrated that the ex vivo culture of human MSCs in the hypoxic (1% O2) environment decreases their osteogenic di fferentiation. The hypoxic environment at the site of cell transplantation in the injured tissues may be a critical factor impairing the e fficacy of MSCs to di fferentiate following their transplantation [42], which, along with the spatial niche organization, provides a unique environment for functional behaviour of the cells in vivo.

Thus, this study aimed to examine the di fferentiation potential of ectoderm-derived human DPSCs in vitro, in comparison with other human MSC fractions, such as the Wharton's jelly-derived MSCs from the umbilical cord (UC-MSCs). We also evaluated osteogenic di fferentiation of human DPSCs in 3D cultures and hypoxic microenvironment in vitro as well as examined their selected functional properties, including proliferation and metabolic activity.
