Mitochondria and autophagy in adult stem cells: proliferate or differentiate

Abstract

Adult stem cells are undifferentiated cells that are found in many different tissues after development. They are responsible for regenerating and repairing tissues after injury, as well as replacing cells when needed. Adult stem cells maintain a delicate balance between self-renewal to prevent depletion of the stem cell pool and differentiation to continually replenish downstream lineages. The important role of mitochondria in generating energy, calcium storage and regulating cell death is well established. However, new research has linked mitochondria to stem cell maintenance and fate. In addition, efficient mitochondrial quality control is critical for stem cell homeostasis to ensure their long-term survival in tissues. In this review, we discuss the latest evidence linking mitochondrial function, remodeling and turnover via autophagy to regulation of adult stem cell self-renewal and differentiation.

Introduction

Adult stem cells are uncommitted cells that are found in many different tissues after development. Adult stem cells have been identified in a wide range of tissues including skeletal muscle, heart, brain, liver, intestine, and bone marrow (Chen et al. 2008; Goncalves et al. 2016; Ito et al. 2004; Korski et al. 2019; Miyajima et al. 2014; Orogo et al. 2015; Xie et al. 2018; Yang et al. 2017). These cells are usually restricted to differentiate into the types of cell in the tissue or organ where they reside. Adult stem cells exist throughout the organism’s life span and function to replace lost cells and regenerate damaged tissues as needed. Adult stem cells possess two key properties. First, they must have the ability to undergo self-renewal where they divide while maintaining an undifferentiated state. Second, they must have the ability to generate progeny of several distinct cell lineages. These are both highly regulated processes and an imbalance in either can lead to depletion of the stem cell pool. For instance, mature blood cells are short lived and hematopoietic stem cells (HSCs) reside in the bone marrow to replenish the cells (Crisan and Dzierzak 2016). The HSCs possess the ability to undergo self-renewal and to differentiate into the various blood cells. Thus, HSCs continuously provide differentiated progenitors while properly maintaining the HSC pool by precisely balancing self-renewal and differentiation (Ito et al. 2004; Jang and Sharkis 2007). Similarly, muscle stem cells, also called satellite cells, are localized between the sarcolemma and the basal membrane of skeletal muscle fibers and play a key role in skeletal muscle regeneration (Yin et al. 2013). After muscle damage, these cells proliferate, fuse and differentiate to form new myofibers or to repair existing injured fibers.

Because of their therapeutic potential, an immense interest exists in understanding the molecular mechanisms underlying self-renewal and differentiation in the adult stem cells. Recent studies have demonstrated the importance of mitochondria in regulating these processes in adult stem cells (Brown et al. 2013; Chen et al. 2008; Chung et al. 2007; Norddahl et al. 2011; Orogo et al. 2015; Rodriguez-Colman et al. 2017; Spitkovsky et al. 2004; Wahlestedt et al. 2014). Mitochondria are specialized double membrane organelles of bacterial origin that are well known for their role in generating energy for the cell via oxidative phosphorylation (Baines 2010). However, mitochondria are also important in many other processes, including calcium homeostasis, steroid metabolism, and apoptosis (Baines 2010). Recent research has demonstrated that mitochondria also participate in the regulation of stem cell maintenance and fate. Mitochondria in adult stem cells are sparse and have poorly developed cristae, while the mature cell possess mitochondria with well-developed cristae and denser matrix and rely on mitochondrial oxidative phosphorylation for ATP generation (Chung et al. 2007; Orogo et al. 2015). Thus, lineage commitment, stem cell fate, and differentiation are associated with metabolic changes which includes significant changes in mitochondrial structure, mass and function.

Adult stem cells are long-lived in tissues and rigorous quality control mechanisms are essential for their long-term survival. Autophagy is a cellular degradation pathway that has been found to play a crucial role in regulating quiescence, self-renewal, and differentiation. Autophagy is a major cellular quality control pathway and is important in eliminating unwanted or damaged mitochondria in cells (Gustafsson and Dorn 2019). In this review, we discuss the latest evidence that supports mitochondrial function, remodeling and turnover via autophagy in regulating adult stem cell homeostasis and function.

Glycolytic energy metabolism and hypoxia

At baseline, adult stem cells exist in a quiescent state with reduced metabolic activity which helps preserve their self-renewal capacity needed for long-term tissue and cell maintenance (Chen et al. 2008; de Meester et al. 2014; Ito et al. 2004; Ito and Suda 2014). They rely primarily on glycolytic metabolism for maintenance and self-renewal (Figure 1). Moreover, the quiescent adult stem cells reside within hypoxic niches in tissues and glycolysis allows them to survive in the low oxygen environment. For instance, the oxygen concentration in the bone marrow niche where HSCs and MSCs reside is reportedly less than 5% (Spencer et al. 2014) and the glycolytic metabolism allows the quiescent HSCs and mesenchymal stem cells (MSCs) to adapt to the hypoxic environment (Chen et al. 2008; de Meester et al. 2014; Ito and Suda 2014). The low oxygen environment is also important for stem cell function and promotes self-renewal and prevents senescence. Studies evaluating stem cells in hypoxic versus normoxic conditions have demonstrated the important influence of oxygen concentrations on stem cell function. For instance, culturing of MSCs under ambient oxygen levels (21%) led to decreased proliferation and increased cellular senescence compared to when cells were cultured under hypoxic conditions (Estrada et al. 2012; Hofig et al. 2016). Similarly, isolation and subsequent culturing of human c-kit positive human cardiac progenitor cells from left ventricular tissue explants at 1% oxygen led to increased self-renewal, reduced ROS levels as well as suppression of senescence compared to cells that were isolated and cultured at 21% oxygen (Korski et al. 2019).

Figure 1.

Mitochondrial structure and function in quiescent and lineage committed adult stem cells.

Hypoxia inducible factor (HIF) signaling provides the link between oxygen levels and cellular processes such as metabolism, proliferation and survival (Semenza 2011). The HIFs form activated heterodimers consisting of an O₂-sensitive α-subunit and an O₂-insensitive β-subunit (Mathieu et al. 2014). During normoxic conditions, the α-subunit of HIF is subjected to ubiquitination and proteasomal degradation. During hypoxia, degradation of the α-subunit is abrogated and the α and β subunits form a stable heterodimeric complex. This complex then binds to hypoxia-responsive elements on DNA to activate transcription of genes involved in the adaptation to hypoxia. Hence, activation of the HIF complex ensures that energy demands are met under hypoxic conditions by increasing levels of glycolytic enzymes and inhibiting oxygen consumption (Kim et al. 2006; Papandreou et al. 2006). HIF signaling plays an important role in the survival and self-renewal of HSCs in the hypoxic bone marrow niche (Rouault-Pierre et al. 2013; Takubo et al. 2010). Moreover, satellite cells (SCs) are muscle-resident stem cells that maintain homeostasis and mediate skeletal muscle regeneration (Yin et al. 2013). SCs cells exist in an intrinsic hypoxic state in vivo and the hypoxic condition promoted SC self-renewal and enhances the efficiency of myoblast transplantation (Liu et al. 2012). The HIF proteins are highly expressed in the SCs and HIF2α has been implicated in promoting stemness and long-term homeostatic maintenance of SCs by maintaining their quiescence, increasing their self-renewal, and blocking their myogenic differentiation (Xie et al. 2018). Loss of HIF2α led to the depletion of the SCs pool which correlated with regenerative failure of skeletal muscle. In contrast, transient pharmacological inhibition of HIF2α accelerated muscle regeneration by increasing SC proliferation and differentiation. Similarly, specific knockout of HIF1α/HIF2α in postnatal SCs was also demonstrated to affect their numbers and muscle regeneration (Yang et al. 2017). Interestingly, mice with myoblast specific HIF-1α or HIF1α/HIF2α knockouts had normal muscle development and no effect on myofiber size and number (Majmundar et al. 2015; Yang et al. 2017). Overall, these findings suggest that HIF1α and HIF2α are dispensable for normal muscle development but essential for maintaining the pool of SCs in skeletal muscle. The studies also suggest that HIF signaling regulates SC proliferation and that disrupting the HIF signaling leads to enhanced proliferation with potential depletion of the SC pool and reduced regenerative potential of the muscle.

Although reliance on glycolysis is considered an adaptation to the hypoxic environment of the stem cell niche and reflects the low energetic demands of stem cells, another benefit of glycolytic metabolism is that it minimizes production of reactive oxygen species (ROS). Actively respiring mitochondria are a major source of ROS which are generated as a byproduct of oxidative phosphorylation (Kowaltowski et al. 2009). It has been reported that HSCs are susceptible to elevated levels of ROS which can lead to the exit of HSCs from quiescence, impair their differentiation capacity, and induce uncontrolled proliferation (Ito et al. 2004; Jang and Sharkis 2007). This ultimately leads to loss of self-renewal and depletion of the HSC pool. Excessive ROS also contribute to stem cell aging and senescence (Brown et al. 2013; Korski et al. 2019; Naka et al. 2008). For instance, Sirtuin 3 (SIRT3) is a mitochondrial localized nicotinamide adenine dinucleotide (NAD)-dependent deacetylase that can enhance mitochondrial antioxidant activity (Liu et al. 2017). SIRT3 is highly expressed in HSCs and plays an important role in adapting to cellular stress (Brown et al. 2013). However, SIRT3 levels were found to decline with age in HSCs which correlated with increased oxidative stress and a reduction in the HSC pool. More importantly, the authors demonstrated that the aged HSCs could be rejuvenated by reducing ROS levels (Brown et al. 2013), confirming that the effects of ROS on HSC function are reversible and could potentially represent a therapeutic target to restore stem cell function in aged tissues.

Mitochondria and Differentiation

Despite their quiescent state, adult stem cells possess a metabolic flexibility that allows them to quickly activate proliferation and differentiation. This facilitates efficient regeneration of tissue in response to injury or replenishment of downstream blood lineages when needed. Moreover, differentiation is associated with extensive remodeling of the mitochondria network and metabolic reprogramming as they exit quiescence (Figure 1). Differentiation is an energy consuming cellular process and a robust increase in energy demand is associated with its activation. Also, higher amounts of energy are required to sustain the specialized functions of the mature cell. Therefore, stem cells undergo a transition from glycolysis to mitochondrial oxidative phosphorylation that is accompanied by induction of mitochondrial biogenesis to increase mitochondrial content in the cell (Chen et al. 2008; Chung et al. 2007; Orogo et al. 2015; Rodriguez-Colman et al. 2017). Several studies have reported that defects in mitochondrial function also impact stem cell differentiation (Norddahl et al. 2011; Orogo et al. 2015; Spitkovsky et al. 2004; Wahlestedt et al. 2014). For example, suppression of oxidative phosphorylation in HSCs by deleting Ptpmt1, a PTEN-like mitochondrial phosphatase, led to defective hematopoiesis due to impaired differentiation of HSCs (Yu et al. 2013). Additionally, a more recent study reported that complete disruption of mitochondrial respiration by genetic deletion of the mitochondrial complex III subunit Rieske iron-sulfur protein (RISP) in HSCs led to defects in differentiation and subsequent depletion of the HSC pool (Anso et al. 2017). Conversely, the reprogramming of somatic cells to inducible pluripotent stem cells (iPSCs) requires the reverse transition from oxidative phosphorylation to glycolysis (Folmes et al. 2011; Prigione et al. 2010). While the metabolic switch from glycolysis to mitochondrial oxidative phosphorylation is required to meet the robust energy and metabolic demands imposed by differentiation, the precise mechanism underlying this switch remains elusive. Further work is needed to determine how glycolysis and/or mitochondria regulate the exit from the quiescent state to differentiation. It is likely that this metabolic reprogramming is more complex than a simple switch from one form of energy production to another.

Autophagy, Mitophagy and Stem Cells

Autophagy is an evolutionary conserved degradation pathway where cellular components are delivered to the lysosome for degradation and recycling (Leidal et al. 2018) (Figure 2). Autophagy is rapidly activated in response to starvation and provides the cell with nutrients needed to survive. However, autophagy is also an important cellular quality control mechanism and maintains cellular homeostasis by regulating both quantity and quality of organelles. Moreover, the selective autophagy of mitochondria, also known as mitophagy, is the process by which damaged or unwanted mitochondria are sequestered by autophagosomes for delivery to the lysosomes (Gustafsson and Dorn 2019). A growing body of evidence suggest key roles for autophagy and mitophagy in the function and maintenance of adult stem cells where they preserve quiescence, maintain stemness and self-renewal, and mediate differentiation (Figure 2). Activation of autophagy during differentiation serves at least two roles. First, differentiation of stem cells involves extensive cellular remodeling and autophagy ensures the elimination of cellular components that are no longer needed (Lampert et al. 2019; Sin et al. 2016). Second, autophagy also helps provide the cell with essential building materials (i.e. amino acids and fatty acids) needed during the cell remodeling by recycling of unnecessary cell components.

Figure 2.

Mitochondrial autophagy (mitophagy) in stem cells. Elimination of respiring mitochondria in uncommitted adult stem cells functions to prevent differentiation and maintain stemness.

Recent studies have begun to explore the function of autophagy in muscle SC homeostasis and reduced autophagy levels in SCs directly contribute to decreased regenerative potential and muscle atrophy (Garcia-Prat et al. 2016; Solanas et al. 2017; Sousa-Victor et al. 2014). For instance, Garcia-Prat and colleagues uncovered that SCs relied on autophagy to maintain quiescence (Garcia-Prat et al. 2016). Disrupting autophagy in young SCs accelerated their aging phenotype and promoted their entry into senescence with increased mitochondrial dysfunction and oxidative stress. In contrast, enhancing autophagy in old satellite cells reversed senescence and restored their regenerative properties (Garcia-Prat et al. 2016). In addition, a recent study reported that SCs expressed genes involved in autophagy in a circadian rhythm, where key autophagy genes peaked late at night or early in the morning in adult SCs with autophagic activity highest during the day (Solanas et al. 2017). Interestingly, this cycling of autophagy genes was reduced in aged SCs. It is known that SCs transition to irreversible senescence with age which leads to a functional decline and reduced muscle regeneration (Sousa-Victor et al. 2014), and these studies suggest that a decline in autophagy contributes to the decreased function of SCs.

Similarly, autophagy of mitochondria plays a key role in maintaining stemness and regenerative potential of HSCs (Ho et al. 2017). Specifically, autophagy was found to be responsible for suppressing oxidative metabolism in HSCs by removing “activated” mitochondria. This study also found that defects in autophagy in HSCs led to an accumulation of actively respiring mitochondria which contributed to accelerated differentiation. These results demonstrate that mitochondrial oxidative phosphorylation can drive differentiation and that mitophagy functions to suppress differentiation by removing actively respiring mitochondria to maintain the stem cell pool. Although the study by Ito et al. confirmed the importance of mitophagy in maintaining stemness of HSCs, their findings demonstrated a specific role for mitophagy in HSC expansion and self-renewal rather than suppressing differentiation (Ito et al. 2016). It is possible that the different results of these two studies are due to the heterogeneity of HSCs used in these studies. While damaged mitochondria are known to be degraded by the PINK1/Parkin pathway (Gustafsson and Dorn 2019), it is still unclear how activated mitochondria are removed in stem cells. It is possible that this process involves mitophagy receptors in the outer mitochondrial membrane that can directly tether mitochondria to the autophagosome. Further research is necessary to identify the molecular mechanism(s) of homeostatic mitophagy in stem cells.

Mitophagy is also activated during differentiation but its function under these conditions is less clear. Differentiation of myoblasts into mature myotubes involves metabolic remodeling with a shift from glycolysis to oxidative phosphorylation. Sin et al. observed that myogenic differentiation was associated with activation of mitophagy and that suppression of autophagy interfered with mitochondrial biogenesis and myogenic differentiation (Sin et al. 2016). More recently, Lampert et al. reported that mitophagy was induced during differentiation of adult cardiac progenitor cells (Lampert et al. 2019). However, the mitophagy was not linked to mitochondrial biogenesis or lineage commitment, but to the formation of a functional interconnected mitochondrial network. Abrogation of mitophagy in cardiac progenitor cells led to development of a fragmented mitochondrial network with suppressed mitochondrial function and increased susceptibility to oxidative stress (Lampert et al. 2019). Mitophagy has also been shown to be activated during developmental and programmed differentiation. For instance, the lens is a transparent tissue and maturation of the lens fiber cells involves the elimination all organelles, including mitochondria. Thus, mitophagy contributes to the formation of transparent organelle-free lens fibers (Costello et al. 2013). Similarly, during maturation to red blood cells, reticulocytes eliminate their mitochondria through mitophagy (Sandoval et al. 2008; Schweers et al. 2007). In these instances, the function of mitophagy is to eliminate the entire population of mitochondria in the cell during differentiation, while in adult stem cells, only a subset of mitochondria are subjected to mitophagy during differentiation. Overall, these studies implicate mitophagy in the differentiation process, but its exact function(s) might be cell specific (complete vs selective elimination).

Alternative Mechanisms of Mitochondrial Quality Control in Stem Cells

Although mitophagy is clearly an important quality control process, alternative mechanisms to ensure a healthy population of mitochondria in stem cells exist. Stem cells can divide symmetrically to generate two new stem cells which serve to increase stem cell number or to compensate for stem cell loss, or asymmetrically to give rise to two daughter cells with different fates (i.e. one stem cell and one differentiating cell) (Tajbakhsh et al. 2009). During symmetrical cell division, there is an equal segregation of organelles including mitochondria. However, Katajisto and colleagues observed that during asymmetrical division, there was a disproportionate division of young and aged mitochondria into the two daughter cells (Katajisto et al. 2015). Using an in vitro model of low (young) or high (aged) replication number cells, this study found that the daughter cell that received fewer aged mitochondria maintained the stem cell characteristics while the daughter cell with more aged mitochondria was fated for differentiation. This demonstrates that in vitro, inhibiting the asymmetrical sorting of mitochondria caused loss of stemness, confirming the importance of this process in maintaining stem cells. Further studies are necessary to confirm whether a similar process occurs in vivo in a more physiologically relevant aging model.

There is also emerging evidence implicating the mitochondrial unfolded protein response (UPR^mt) in stem cell maintenance (Berger et al. 2016; Mohrin et al. 2015; Mohrin et al. 2018; Zhang et al. 2016). The UPR^mt is a cellular quality control pathway that is activated by mitochondrial protein folding stress to promote repair and cell survival (Callegari and Dennerlein 2018). Activation of the UPR^mt involves transcriptional activation of nuclear-encoded mitochondrial chaperones and proteases. Stem cells that are transitioning from quiescence to proliferation induce mitochondrial biogenesis to enable the metabolic reprogramming from glycolysis to oxidative phosphorylation. Mohrin et al. observed that the UPR^mt was activated upon HSC transition from quiescence to proliferation which correlated with increased mitochondrial biogenesis (Mohrin et al. 2018). The increase in mitochondrial mass coincided with increased expression of mitochondrial chaperones and proteases, indicative of UPR^mt activation (Mohrin et al. 2018). Increased mitochondrial respiration also leads to increased production of ROS in lineage committed cells, and it is likely that activation of the UPR^mt during differentiation functions to facilitate repair of mitochondrial components that have been damaged by the ROS.

Dysregulation of the UPR^mt has also been linked with compromised stem cell self-renewal and a reduced stem cell pool (Berger et al. 2016; Mohrin et al. 2015; Zhang et al. 2016). For instance, SIRT7 is a nuclear histone deacetylase that interacts with various transcription factors to repress transcription (Kiran et al. 2015). Mohrin et al. discovered that SIRT7 is a key component of the UPR^mt and alleviates mitochondrial protein folding stress by limiting mitochondrial biogenesis and metabolism (Mohrin et al. 2015). Specifically, SIRT7 interacted with NRF1 to repress transcription of mitochondrial ribosomal subunits and translation factors. SIRT7 levels decreased in HSC with age which correlated with chronic activation of UPR^mt, increased proliferation, and depletion of the stem cell pool (Mohrin et al. 2015). In contrast, a study comparing young and aged muscle SCs observed upregulation of senescence pathways, increased mitochondrial dysfunction and decreased UPR^mt in aged SCs (Zhang et al. 2016). Several regulators of the UPR^mt were downregulated in aged SCs which could contribute to the mitochondrial dysfunction observed with age in these cells. More importantly, activation of the UPR^mt led to rejuvenation of the aged SCs (Zhang et al. 2016), suggesting that the UPR^mt preserves mitochondrial homeostasis and self-renewal in SCs. Additional studies are needed to investigate the function of the UPR^mt in other adult stem cell populations and to determine how the UPR^mt is coordinated with mitophagy. If the UPR^mt is impaired, is there increased compensation in mitophagy?

Mitochondrial DNA mutations and stem cell aging

Mitochondria contain their own genome, which encodes 37 genes including 13 subunits of the respiratory chain complexes. Each mitochondrion contains multiple copies of mitochondrial DNA (mtDNA) and replication of the genome occurs during mitochondrial biogenesis. The mtDNA is more prone to damage and mutations compared to nuclear DNA because the mitochondrial genome lack protective histones and is located close to the respiratory chain where ROS are generated during oxidative phosphorylation. MtDNA mutations are known to accumulate in cells and tissues with age which contributes to mitochondrial dysfunction (Sharma and Sampath 2019). Interestingly, a specific threshold of mtDNA mutations must be reached before a functional effect can be observed on mitochondrial respiration. This so called “threshold effect” varies between tissues with heart and muscle having high thresholds before an effect on mitochondrial function is observed (Rossignol et al. 2003; Woodall et al. 2019).

Studies using a mouse model carrying a proofreading defective mitochondrial DNA polymerase (POLG) have provided important insights into how mtDNA mutations affect stem cell renewal and function. These mice accumulate mtDNA mutations at an accelerated rate which leads to a progeroid phenotype. Many adult stem cell populations are affected in these mice. Ahlqvist et al. observed that neural stem cells (NSCs) and HSCs in POLG mice were impaired (Ahlqvist et al. 2012). Interestingly, while the mtDNA mutations affected self-renewal of NSCs (Ahlqvist et al. 2012), they had little effect on the HSC pool but affected the differentiation of downstream progenitors (Norddahl et al. 2011). In adult cardiac progenitor cells, accumulation of mtDNA mutations led to an inability to undergo the metabolic switch from glycolysis to oxidative phosphorylation, resulting in activation of cell death during differentiation (Orogo et al. 2015). Interestingly, although the mtDNA mutations in the cardiac progenitor cells did not disrupt baseline autophagy and mitophagy, these cells failed to activate the programmed mitophagy during differentiation (Lampert et al. 2019). This suggest that a link exists between mtDNA integrity, oxidative phosphorylation and induction of programmed mitophagy. In addition, using a mouse model with transient mtDNA damage, Wang and colleagues found that this led to a significant decline in muscle SCs, which decreased the muscle’s capacity to regenerate and repair during aging. Thus, the primary cause of the muscle wasting that occurs in the mouse model with mtDNA damage might be related to alterations in the satellite cell pool (Wang et al. 2013).

Conclusion

Recent studies provide persuasive evidence that mitochondria and autophagy play key roles in stem cells and can influence stem cell functions and survival. Regenerative therapies using adult stem cells represent a very promising strategy to repair and regenerate various tissues in many different diseases. Looking ahead, it is important to increase our understanding of the biology of these cells and utilize this information to enhance their therapeutic potential in vivo. Several questions need clarification. First, it is clear that stem cells undergo a metabolic switch from glycolysis to oxidative phosphorylation. However, the molecular mechanisms involved in the switch remain elusive. Future studies need to investigate how mitochondria regulate the metabolic transition upon initiation of differentiation. It is very likely that the metabolic reprogramming and initiation of differentiation is more complex than a simple switch from one form of energy production to another. Second, most of our knowledge of adult stem cell biology comes from in vitro studies of cells in culture under ambient oxygen levels. However, adult stem cells exist in a hypoxic environment and increasing oxygen levels has profound effects on their proliferation and differentiation in culture. To fully understand the biology of these cells, additional experiments performed under hypoxic conditions are needed. Finally, expectedly, cellular quality control through autophagy and mitophagy are important to preserve the stemness and self-renewal of adult stem cells. However, the exact function of mitophagy during differentiation is still intriguing as only a small number of mitochondria are eliminated while mitochondrial mass is substantially increased. Why does a failure to remove this small pool of mitochondria have such a profound effect on the mitochondria? Once we have a better understanding of the relationship between metabolism, mitochondria, and autophagy, and through it stem cell function and differentiation, we will hopefully be able to utilize this knowledge into modulating stem cell function and improving regenerative therapies for various diseases.