Metabolic Regulation of Pluripotent Stem Cell Expansion, Differentiation, and Reprogramming

from OXPHOS to glycolysis [6]. The pluripotent marker OCT4 affects several target genes (e.g. ATP5D) associated with energy metabolism which are in favor of glycolysis [1]. Various new hPSC culture systems have been developed in recent years, including defined serum-free media (e.g., mTESR, StemPro, and E8) and synthetic substrates (e.g., synthemax) [7,8]. How the subtle differences in these culture conditions affect energy balance and the differentiation efficiency of hPSCs needs to be explored.


Undifferentiated State
Metabolic modulation has been used to develop new culture systems function using antimycin A, which decreases ATP production from OXPHOS and Reactive Oxygen Species (ROS) formation, can support from OXPHOS to glycolysis [6]. The pluripotent marker OCT4 affects several target genes (e.g. ATP5D) associated with energy metabolism which are in favor of glycolysis [1]. Various new hPSC culture systems have been developed in recent years, including defined serum-free media (e.g., mTESR, StemPro, and E8) and synthetic substrates (e.g., synthemax) [7,8]. How the subtle differences in these culture conditions affect energy balance and the differentiation efficiency of hPSCs needs to be explored.

IPSC Reprogramming
Metabolic plasticity of PSCs has been recently demonstrated in the method development of iPSC reprogramming [4]. The somatic cells switch the metabolism from OXPHOS to the enhanced glycolysis and become pluripotent when reprogramed with pluripotent genes and/or small molecules [9,10]. Compared to early passages of iPSCs, the levels of metabolites from late passages of iPSCs were found more close to those of ESCs. The slow transition in the metabolic profile changes also suggested the effect of "metabolic memory" during reprogramming [9]. Consequently, the somatic cells with a higher glycolytic and lower OXPHOS capacity showed higher reprogramming efficiency [4,9]. Therefore, promotion of glycolic metabolism by drugs, hypoxia, or supplementation of glycolic intermediates (e.g. PS48, a small molecule activator of 3'-phosphoinositide-dependent kinase-1 (PDK1)) has been found to augment the reprogramming efficiency, while inhibition of glycolysis and/or stimulation of OXPHOS impaired reprogramming [9][10][11].

Specific Lineage Differentiation: Monolayer or Embryoid Body (EB)?
Most current hPSC differentiation protocols either use the formation of EBs or through monolayer induction by growth factors. Higher differentiation efficiency has been observed for some monolayer protocols for several lineages (cardiomyocytes, hematopoietic progenitors, etc) compared to EB protocols [12,13]. Due to the differences in oxygen and nutrient transport in EBs compared to monolayer cultures, distinct different metabolic fluxes were observed during spontaneous differentiation with EBs displaying a stronger lactogenic pattern [14]. However, the energy needs for lineage-specific differentiation may be different from spontaneous differentiation.

The Case of Neural Differentiation
Although neural stem/progenitor cells can be derived from hPSCs, little is known about the bioenergetic requirements during this cellular process. Contrary to the increased utilization of OXPHOS during hPSC differentiation, results also have been reported for the shift from OXPHOS to glycolysis during neural differentiation of hPSCs [15,16]. The conflicting results may be due to the variations in neural differentiation protocols, the culture conditions, and the stage of differentiation. For example, proliferating neural progenitor cells showed higher levels of glycolysis compared to neural stem cells [2]. Evaluation of metabolic shift for differentiation into different subtypes of neural cells has direct impact in PSC engineering. Most neural differentiation protocols from hPSCs are lengthy. For example, the derivation of Oligodendrocyte Progenitor Cells (OPCs) from hESCs needs 41 days in the production process involving both suspension and adherent cultures [17]. The failure of the production process will cause huge economical loss due to the scale of production. The ability to predict the differentiation outcome by monitoring the metabolic activity will greatly benefit the production of hPSC-derived OPCs.
In recent years, metabolic plasticity of human Pluripotent Stem Cells (hPSCs), including human Embryonic Stem Cells (hESCs) and induced Pluripotent Stem Cells (iPSCs), emerges as an important area to regulate hPSC expansion, differentiation, and the iPSC reprogramming [1,2]. PSCs have unique energy and biosynthetic requirements and utilize predominantly glucose through glycolysis rather than Oxidative Phosphorylation (OXPHOS) at undifferentiated state [3]. Upon differentiation, the metabolic pathway is observed to shift from glycolysis to OXPHOS [2]. The reverse process is also observed during the reprogramming of somatic cells to iPSCs, where the metabolic pathway shifts from OXPHOS to glycolysis when the cells gain pluripotency [4]. Metabolic pathways of PSCs (e.g. glucose metabolism, mitochondrial function, redox status, etc.) can be regulated by environmental conditions such as oxygen and the biomolecules that act in the intermediate steps of glycolysis and/or OXPHOS [5]. Understanding the metabolic status during different stages of PSC development can benefit various aspects of PSC engineering in expansion, differentiation, and reprogramming.

Expansion and Maintenance of hPSC in
Studying the effect of 3-D organization on metabolic fluxes during lineage-specific differentiation would be required.
for undifferentiated hPSCs. For example, modulation of mitochondria undifferentiated hPSC propagation by stimulating the metabolic shift

The Case of Cardiomyocyte Differentiation
Cardiomyocyte differentiation has been significantly improved in on the status of the cell population before the growth factor induction, even though the cells do express high levels of pluripotent markers. The ability to detect the subtle differences in hPSCs contributing to differentiation variations is needed for robust production of hPSCderived cardiomyocytes. The endogenous ROS level, which is promoted in high glucose medium, was found to enhance cardiomyocyte differentiation from PSCs [20]. In addition, downstream purification is also critical to the isolation of cardiomyocytes because the differentiation efficiency is low (20-50%) in most of the protocols. Based on distinct metabolic flow, it has been demonstrated that hPSCderived cardiomyocytes can be enriched under glucose-depleted and lactate abundant conditions [21]. Providing the medium that can induce adult-like energy metabolism was also found to accelerate the

Current Challenges and Perspectives
Although the metabolic shift correlates with the status during PSC expansion, differentiation, and iPSC reprogramming, applications of metabolic regulation in hPSC-derived cell production are still illusive and needs further studies and better understanding of the effects of various metabolites and substrates. Glucose and oxygen are the two most important substrates in cell metabolism, and their concentrations in the culture media can be used to modulate metabolic status of cells. Design the feeding regime and the exposure to hypoxia condition can help to regulate the cellular composition. In addition, small molecules affecting the mitochondrial function and glycolysis/OXPHOS pathways can also be used to modulate PSC fate decisions. Understanding the metabolic plasticity can help design the process in PSC engineering in the following aspects: 1) developing well-defined culture conditions for hPSCs; 2) process monitoring based on changes in the metabolic status of the cells; 3) developing novel downstream separation process for purifying the cells of specific lineages differentiated from hPSCs. These developments depend on if the following questions can be answered: 1) Can metabolic status be used to predict the specific lineage differentiation efficiency? 2) Can metabolic status be used to modulate the composition of cell population? 3) Can metabolic regulation be used for process integration of iPSC reprogramming, expansion, and differentiation? Taken together, translation the findings in metabolic plasticity of hPSCs to the manufacturing process should accelerate the advancements in hPSC-derived cell production for drug screening, disease modeling, and tissue engineering.