Discussion:
In this study we highlighted the potential of bioinformatic, and systems
biology tools and approaches to provide a better insight into the
interaction of biomaterials with stem cells. Bioinformatic analyses can
unravel subtle molecular interactions between biomaterials and stem
cells. Using such information will allow for the optimization of stem
cell differentiation in regard to time, expense, and quantity of
properly differentiated cells. Here, we briefly discuss the impact of
biomaterials and of cells of origin on stem cell differentiation.
Recent studies have shown that the microenvironment in which stem cells
grow has a dramatic effect on the efficiency of generated cells (Tibbitt
& Anseth, 2012). For example, it has been shown that hyaluronic
acid-based 3D hydrogels can successfully be used to culture hippocampal
neural progenitor cells (Tarus et al., 2016). However, use of a 3D
system has limitations, as this dimensionality can negatively impact the
size and shape of 3D spheres. For instance, Hydrogel films of
3-hydroxybutyrate and 3-hydroxyhexanoate copolymers has been used in 2D
form to drive NSCs toward neurons (Xu et al., 2010). This is one example
in which the ideal culturing techniques have yet to be completely
elucidated, stressing the importance to better understand 2D versus 3D
culturing systems. In our study, we delved deeper into the importance of
biomaterial dimensionality for NPC differentiation and found that NPCs
cultured on 2D porous polystyrene scaffolds appear to produce more
transcriptionally mature neuronal cells than those cultured on 3D
biomaterials based on their transcriptome profile. Our data also
indicates the importance of specific biomaterial use, as NPCs were
differentiated more efficiently on a 3D hyaluronic acid compared to 2D,
but both of these results were overshadowed by the efficiency of NPC
differentiation on 2D porous polystyrene scaffolds.
In addition to the effect of biomaterials on stem cell differentiation,
the cell of origin has been shown to be an important factor in
determining the quality of the differentiated cell. For example, Wu et
al. compared biological characteristics of different adult multipotent
stem cells, including stem cells isolated from human bone marrow,
placental decidua basalis and urine, based on their ability in colony
formation, morphology, proliferation and differentiation into different
cell types (Wu et al., 2018). They showed that BM-MSCs have a higher
potential to differentiate into chondrogenic and osteogenic cell
lineages than other multipotent stem cells (Wu et al., 2018). In another
study, Rim et al. took a similar approach, but they aimed to determine
the effect of the cell of origin on iPSC differentiation. iPSCs were
generated from four different cell of origins, including dermal
fibroblasts (DF), peripheral blood mononuclear cells (PBMC), cord blood
mononuclear cells (CBMC), and osteoarthritis fibroblast-like
synoviocytes (OAFLS) (Rim et al., 2018). Differentiation of these cells
toward chondrogenic fate showed iPSCs derived from CBMC had the highest
differentiation potential (Rim et al., 2018). These groups found that
cell of origin played an important role in the functionality of
differentiated cells. However, we believed that there are more subtle
differences between cells differentiated from separate cells of origin
that may be difficult to interpret with some functional assays.
Therefore, we investigated differences at the transcriptomic level that
would give a more precise similarity/difference readout between the
different cells of origin as they were differentiated.
Therefore, we attempted to determine whether MSCs generated from
different cell sources, including hiPSCs, hESCs, and BM, have different
capacities during osteogenesis. Our results show that osteoblasts
generated from hiPSC-MSCs and hESC-MSCs clustered together, while BM-MSC
derived osteoblasts displayed a greater variability in their
transcriptomic profile. Furthermore, our results clearly showed more
genes involved in cell cycle and extracellular matrix organization in
differentiated osteoblasts from MSCs originated from ESCs and iPSCs than
BM-MSCs. The transcriptomic profile of each separately derived
osteoblast lineage was mapped on the regulatory network specific for
ESC-MSCs, and large differences between each lineage were evidently
observed. In this regulatory network, MYC was significantly decreased
exclusively in the ESC-MSC generated osteoblast and was the main
regulator of the entire network in this lineage. Ontology analysis of
MYC target genes also showed that cell cycle and mitochondrial terms
were the most significant affected biological terms. This finding is of
interest because Hanna et al and Persson et al showed that the
proliferative capacity of osteogenic differentiating cells reduces in
later stages of this process (Hanna, Mir, & Andre, 2018; Persson et
al., 2018). Therefore, according to the expression pattern of MYC in
osteoblasts generated across the three cells of origin, the ESC-MSC
derived osteoblasts may be the most biologically relevant differentiated
cell. Although more analysis is required to get the full picture, our
analysis shows the importance of looking at the transcriptomic level of
different cells of origin to understand which lineage may be the best
for a particular differentiation.