An Enduring Role for Quiescent Stem Cells

Abstract

The intestine’s ability to recover from catastrophic injury requires quiescent intestinal stem cells (q-ISCs). While rapidly cycling (Lgr5+) crypt base columnar (CBC) ISCs normally maintain the intestine, they are highly sensitive to pathological injuries (irradiation, inflammation) and must be restored by q-ISCs in order to sustain intestinal homeostasis. Despite clear relevance to human health, virtually nothing is known regarding the factors that regulate q-ISCs. A comprehensive understanding of these mechanisms would likely lead to targeted new therapies with profound therapeutic implications for patients with gastrointestinal conditions. We briefly review the current state of the literature highlighting homeostatic mechanisms important for q-ISC regulation listing key questions in the field and offer strategies to address them.

INTRODUCTION

The mammalian intestine turns over faster than any other tissue in the body. The capacity of this tissue to function relies upon two highly specialized, long-lived populations of intestinal stem cells (ISCs) (Fig. 1). Rapidly cycling crypt base columnar (CBC) ISCs function to maintain the intestinal epithelium under normal baseline conditions (Barker et al., 2007; Barker et al., 2008; Jaks et al., 2008; Roche et al., 2015). Perhaps due to their highly proliferative and active metabolic state, CBC ISCs are especially sensitive to apoptosis in response to physiological and pathological injuries (e.g., irradiation, inflammation and starvation) and readily undergo neoplastic transformation and adenoma formation in response to oncogenic mutations (Barker et al., 2009). In contrast, quiescent intestinal stem cells (q-ISCs) are required for the intestine’s ability to recover from catastrophic injury. Q-ISCs are largely dormant, located in the “+4” supra-Paneth cell position (Sangiorgi and Capecchi, 2008; Montgomery et al., 2011; Takeda et al., 2011; Tian et al., 2011; Powell et al., 2012; Metcalfe et al., 2014; Ritsma et al., 2014; Roche et al., 2015), and are highly resistant to the effects of physiological and pathological injuries (Montgomery et al., 2011; Yan et al., 2012; Richmond et al., 2015). In fact, we and others have shown that q-ISCs are activated upon injury to contribute to tissue regeneration, including restoration of the CBC ISC population (Montgomery et al., 2011; Takeda et al., 2011; Tian et al., 2011) (Fig. 1). Moreover, recent studies from our laboratory indicate that q-ISCs are resistant to neoplastic transformation and adenoma formation (unpublished observations). Because there has been much debate as to whether q-ISCs are truly quiescent or merely slowly cycling, for simplicity we will refer to them as q-ISCs in this review. Despite clear relevance to human health, surprisingly little is known regarding how q-ISCs are regulated. Thus, a comprehensive understanding of these cells may help inform targeted new therapies for patients with gastrointestinal conditions including: inflammatory bowel disease, intestinal cancer, malnutrition, and short bowel syndrome as well as those at risk for radiation and chemotherapy-related intestinal toxicity. Here, we briefly review the current state of the literature highlighting homeostatic mechanisms important for q-ISC regulation. We also list key questions in the field and offer strategies to address them.

Fig 1. Model of Intestinal Homeostasis.

Two ISC models of intestinal homeostasis showing quiescent (green) ISCs and rapidly cycling (orange) CBC ISCs during steady state, injury and recovery from injury. Modified from Carlone and Breault, Cell Stem Cell 2012.

THE ROLE OF QUIESCENT STEM CELLS IN TISSUE HOMEOSTASIS

Lessons Learned from the HSC

To gain insight into potential mechanisms that may underlie q-ISC regulation, scientists have turned to the hematopoietic stem cell (HSC), the most extensively characterized adult tissue stem cell to date. In blood, a hierarchical paradigm exists whereby self-renewing quiescent (q)-HSCs give rise to cycling multi-potent progenitor cells and subsequent differentiated lineages. The quiescent nature of HSCs has been interpreted as a trait that serves to protect their genome from accumulating deleterious mutations, thus preventing their premature exhaustion and replicative senescence. Many of the lessons learned from studying q-HSCs have been subsequently applied to other tissues (Guasch and Blanpain, 2004; Tumbar et al., 2004; Bjornson et al., 2012). As a result, quiescent adult stem cells have since been described in a wide-range of tissues (Tumbar et al., 2004; Cheung and Rando, 2013; Hsu et al., 2014; White et al., 2014), including the intestine.

Identification of LRCs in the Intestine

The notion that q-ISCs might exist within the intestinal crypt was initially put forth by Potten and colleagues (Potten, 1977; Potten et al., 2009). They predicted that such cells could be identified based on their infrequent cell cycling status, which would lead to long-term retention of DNA labeling agents. Such long-term DNA label-retaining cells (LRCs) were initially identified using ³H-thymidine (and later Bromodeoxyuridine, BrdU), administered following cytotoxic injury or during intestinal development (Bjerknes and Cheng, 1999). The identification of single LRCs in the “+4” crypt position, in a tissue whose epithelium turns over every ~4 days, was a landmark achievement in the field. Despite this, the lack of a functional ISC assay left the true identity of these cells unknown for more than 3 decades. However, with the development of functional lineage-tracing methods, remarkable advances have been made in our understanding of q-ISCs. For example, expression of a number of genes (e.g., mTert, Bmi1, Lrig1, HopX, Sox9^hi and DCKL1) (Sangiorgi and Capecchi, 2008; Montgomery et al., 2011; Takeda et al., 2011; Tian et al., 2011; Powell et al., 2012; Metcalfe et al., 2014; Ritsma et al., 2014) has been shown to mark cells in the “+4” position that contribute to all the differentiated cell lineages within the intestinal epithelium and persist long-term (e.g., > 1 year), confirming their identity as q-ISCs. Given the number of marker genes identified to date, it is likely that these cells represent a heterogeneous population of ISCs with varying degrees of quiescence, lineage-commitment and self-renewal potential. Establishing the precise homeostatic role for each subpopulation and the factors and pathways involved in their regulation remain essential unanswered questions.

Questioning the Role of Quiescent Stem Cells (q-SCs) in Tissue Maintenance

Despite the available data supporting the existence and functional potential of both q-HSCs and q-ISCs, significant debate continues regarding their role in tissue maintenance and regeneration. A central question is whether the hierarchical HSC paradigm remains relevant today and whether or not similar paradigms exist in other tissues. For example, the veracity of the hierarchical HSC paradigm was recently questioned based on the observation that multi-potent progenitor cells alone were sufficient for maintaining all the differentiated blood lineages, calling into question whether HSCs are needed at all (Sun et al., 2014). In addition, the observation that rapidly cycling CBC ISCs are sufficient for maintenance of the intestinal epithelium has led to similar questions in the gut (Clevers, 2015). When considering these arguments, however, it is important not to overlook the role that q-SCs play during tissue regeneration when cycling cells are vulnerable to injury and death and selective pressure is at its highest.

The Role of q-SCs in Tissue Regeneration

It has long been held that quiescent stem cells serve to restore homeostasis following catastrophic tissue injury, when actively cycling progenitor/stem cells are lost. In fact, the strongest evidence that q-HSCs represent bona-fide stem cells comes from their capacity for functional multi-lineage engraftment following bone marrow transplantation (Morrison et al., 1995), which exerts strong selective pressure. In the intestine, despite the heterogeneous nature of the q-ISC population, it is clear that q-ISCs have the capacity to respond to tissue injury (Montgomery et al., 2011; Takeda et al., 2011; Powell et al., 2012; van Es et al., 2012; Van Landeghem et al., 2012; Yan et al., 2012; Barker, 2014). On the other hand, recent evidence suggests that the mechanism of cellular plasticity, wherein differentiated or differentiating cells revert back to a stem cell state, may support tissue regeneration following injury (Mills and Sansom, 2015). For example, it has been reported that committed progenitor cells, label-retaining secretory precursors, and even “differentiated” cells are highly plastic in response to tissue injury (van Es et al., 2012; Buczacki et al., 2013; Tetteh et al., 2016). However, the use of fluorescent histone marking as a strategy can lead to the marking of a heterogeneous population of crypt cells, which in this case, included secretory progenitor cells and Lgr5-expressing cells, amongst others (Buczacki et al., 2013). Additionally, it remains possible that “differentiation” markers (e.g., Alpi) may be more broadly expressed, making them less specific than originally believed. Precisely how each q-SC population contributes to tissue regeneration and the role cellular plasticity plays are major issues still to be addressed by the field.

Natural Selection and the Role of q-SCs in Divergent Tissues

Debate remains regarding whether the rules governing a q-SC in one tissue can be applied globally. For example, some have argued that natural selection would not necessarily favor utilizing the same self-renewal strategy in all tissues (Clevers, 2015). That is, given Nature’s ability to evolve independent strategies when necessary, it is likely that the diversity inherent in each tissue would have led to divergent regenerative mechanisms (Clevers, 2015). However, the opposite scenario, i.e., evolving a single strategy rather than multiple strategies, is equally possible, as Occam’s Razor suggests. Countless examples exist in nature where the evolution of a successful strategy that favors natural selection is continually repeated in highly divergent settings throughout the phylogenetic tree as described by the concepts of “parallel” and “convergent” evolution (Orr, 2005). Thus, a priori, it remains entirely possible that multiple organ systems may share common regenerative strategies, such as a reliance on quiescent stem cells as a reserve stem cell population. Establishing the factors that regulate these essential cells will provide fundamental insight into their precise role in tissue homeostasis.

FACTORS REGULATING Q-ISCS

Because q-ISCs are phenotypically and functionally distinct from CBC ISCs, they are likely regulated by different mechanisms. In fact, increasing evidence indicates that quiescence is an actively imposed state, rather than a passive phenomenon (Cheung and Rando, 2013). While the factors that regulate the quiescent state are largely unknown, a range of mechanisms have been implicated, including those involved in cell cycle control, injury response, developmental signaling pathways, cellular metabolism, and epigenetic modulation (Cheung and Rando, 2013). ISCs reside within intestinal crypts in a specialized niche, which regulates the balance between self-renewal and lineage-commitment divisions (Moore and Lemischka, 2006). Three signaling pathways, canonical Wnt/β-catenin (cWnt) (Korinek et al., 1998; Barker et al., 2009), Bone Morphogenetic Protein (BMP) (Auclair et al., 2007), and Insulin/IGF1 (IIS) (Richmond et al., 2015), have been identified as major regulators of ISCs and the intestinal crypt. cWnt signaling is a critical regulatory pathway in the intestine (Morin et al., 1997; Barker et al., 2007) but its role in q-ISC regulation is largely unknown. The BMP signaling pathway is known to regulate growth, apoptosis and differentiation of intestinal epithelial cells and over-expression of Noggin, a BMP inhibitor, leads to ectopic crypt formation suggesting an important role for this pathway in the regulation of the ISC niche (Haramis et al., 2004). Finally, q-ISCs in particular have recently been found to be specifically regulated by IIS signaling (Richmond et al., 2015; Van Landeghem et al., 2015). Previously, Li and Clevers proposed a “two-cell model” of ISC homeostasis, which posits that at baseline, CBC ISCs are maintained in a “Wnt on” and “BMP off” state and q-ISCs are maintained in a “Wnt off” and “BMP on” state (Li and Clevers, 2010), though definitive evidence in support of this model is lacking. Based on our recent findings that the IIS pathway and its negative regulators, e.g., PTEN, play an important role in ISC crypt regulation (Richmond et al., 2015), we now propose an updated two-cell model of ISC homeostasis (Fig. 2).

Fig 2. Two Cell Intestinal Stem Cell Model.

Schematic showing the potential role of three canonical regulatory pathways (Wnt, Bone Morphogenic Protein (BMP) and Insulin/IGF-1 Signaling (IIS)) in the regulation of q-ISCs and CBC-ISCs. At baseline, q-ISCs are maintained in a Wnt Off, BMP On & IIS Off inhibitory zone while CBC ISCs are maintained in a Wnt On, BMP Off & IIS On stimulatory zone. Modified from Li and Clevers, Science 2010.

Signaling Pathways

cWnt Signaling

cWnt signaling is essential for intestinal homeostasis and mediates cellular proliferation and differentiation (Fig. 3). Consistent with this, expression of the Wnt-target gene, Lgr5 is highest in actively cycling CBC ISCs and lowest in q-ISCs (Munoz et al., 2012). Moreover, patients with germ-line mutations in key components of this pathway develop Familial Adenomatous Polyposis (FAP) (Kay et al., 2015). Consistent with this, mice with gain-of-function mutations in the Wnt pathway develop intestinal neoplasia (Barker et al., 2009), whereas loss-of-function mutations result in intestinal failure (Korinek V, 1998). While CBC ISCs are Wnt-responsive and readily transformed following activation of this pathway (Barker et al., 2007; Barker et al., 2009), conflicting data exist for q-ISCs. For example, Bmi1⁺ q-ISCs were originally reported to form adenomas following stabilization of β-catenin (Sangiorgi and Capecchi, 2008); however, more recently Bmi1⁺ q-ISCs were found to be relatively unresponsive to cWnt signaling (Yan et al., 2012) (Table 1). This discrepancy may be explained by inherent differences in the mouse models utilized to mark this ISC population (Li et al., 2014). The critical question of whether other q-ISCs populations, such as mTert⁺ q-ISCs, are Wnt-responsive remains to be formally established.

Fig 3. Schematic of Canonical Wnt/β-Catenin and BMP Signaling in q-ISCs.

Together these signaling pathways are thought to mediate the balance between quiescence, proliferation and survival. In q-ISCs, BMP signaling may antagonize Wnt signaling, helping to maintain quiescence.

Table 1.

Proposed ISC classification scheme

Cycling Frequency	Quiescent	Rapidly Cycling
•Cells Per Crypt	~1	~15
•Long-Term Persistence	Yes	Yes
•Multilineage Contribution	Yes	Yes
•Contribution To CBC	Yes	Yes
•Wnt Responsive (at baseline)	No	Yes
•Response To Stress
➢Radiation (>10gy)
➢Fasting

BMP Signaling

Another pathway critical for intestinal homeostasis is the BMP signaling pathway, which regulates quiescence, proliferation and cell survival within the crypt (Auclair et al., 2007) (Fig. 3). It is known, for example, that patients with germ-line mutations in BMP signaling components develop Juvenile Polyposis Syndrome which confers increased lifetime cancer risks, albeit lower than those individuals with FAP (Scoville et al., 2008). Though, precisely how defects in BMP signaling result in this phenotype remains largely unknown (Goodell et al., 2015). It has been proposed that BMP signaling blocks proliferation in q-ISCs by antagonizing canonical Wnt signaling (He et al., 2004), though conclusive evidence has yet to be provided (Goodell et al., 2015) (Fig. 3). Whether BMP signaling directly regulates q-ISCs requires further investigation.

Insulin/IGF1 signaling

Insulin/IGF1 signaling, together with PTEN (phosphatase and tensin homolog), an essential negative regulator of the IIS pathway (Fig. 4), plays an important role in the regulation of cellular metabolism, survival and proliferation (Sadagurski and White, 2013). Patients with germ-line deletion of PTEN (PTEN Hamartoma Tumor Syndrome, Cowden syndrome, and Bannayan-Riley-Ruvalcaba Syndrome) experience unrestrained IIS and develop intestinal polyps (Scoville et al., 2008). Consistent with this, gain-of-function mutations in IIS are also associated with colorectal cancer (Cancer Genome Atlas, 2012) indicating that tight control of this pathway is required for normal intestinal homeostasis. Within the crypt, PTEN specifically marks q-ISCs and functions as an important negative regulator of their activation (He et al., 2007; Montgomery et al., 2011; Richmond et al., 2015) (Fig. 4). Moreover, PTEN is dynamically regulated within these cells as demonstrated by transient and reversible de-repression in response to acute nutrient deprivation (Richmond et al., 2015). Furthermore, PTEN loss leads to an impaired regenerative response following high dose radiation (Richmond et al., 2015). Precisely how IIS and PTEN modulate q-ISC behavior at baseline and in response to intestinal injury is an important area for ongoing study.

Fig 4. Schematic of Insulin/IGF-1 Signaling (IIS) in q-ISCs.

PTEN negatively regulates IIS in q-ISCs under baseline maintenance conditions. PTEN is transiently inactivated to allow q-ISC activation during regeneration.

Environmental Factors

The behavior and function of q-ISCs are additionally modified by their micro- and macro-environments, which impact both the niche as well as the q-ISC itself. Examples of external environmental factors that profoundly impact this population include ionizing radiation (Potten 2004), the nutrient status of the organism (Richmond et al., 2015), and oxidative stress (Tothova et al., 2007). In contrast, an example of an intrinsic factor that may impact ISC activity is its specific metabolic profile (Shyh-Chang et al., 2013). Below, we address how these various factors influence ISCs behavior.

Response to Radiation-Induced injury

To date, most studies examining the regenerative response of ISCs have focused on radiation-induced injury (a potent but pathological insult) (Potten, 2004; Kirsch et al., 2010), which remains the gold standard for studying crypt regeneration by stem cells. Generally, CBC ISCs are radio-sensitive while q-ISCs are radio-resistant (Barker et al., 2007; Montgomery et al., 2011; Takeda et al., 2011; Powell et al., 2012; van Es et al., 2012; Van Landeghem et al., 2012; Yan et al., 2012; Barker, 2014). In addition, the q-ISC population expands in response to irradiation (Montgomery et al., 2011), suggesting activation of these cells. Transcriptional profiling studies reveal that the expression pattern (e.g., genes involved in cell cycle, DNA replication/repair, cellular assembly) of irradiated q-ISCs is comparable to non-injured rapidly cycling ISCs, further supporting the notion that quiescent ISCs are activated during the regenerative response (Van Landeghem et al., 2012). Along these lines, recent studies have shown that q-ISCs isolated from irradiated mice exhibit significantly increased ability to form organoids in culture as compared with non-irradiated controls (Van Landeghem et al., 2012). Moreover, when these cells are depleted using genetically modified mice, the intestine’s ability to regenerate following irradiation is severely compromised (Richmond et al., 2015; Roche et al., 2015). Thus, radiation-induced injury remains a robust model to test the regenerative capacity of the stem cell compartment in the intestine.

Challenges, however, remain in interpreting and integrating the available data from multiple irradiation studies to delineate the role of q-ISCs. Difficulties include reconciling various radiation dosages, endpoints following recovery and methods of evaluation. For example, lineage-tracing strategies designed to look at the ISC response need to take into account the timing of tamoxifen administration and factor in the long biological half-life of this commonly used compound as it can remain biologically active for >1 week following a single dose (Lien et al., 1989; Kisanga et al., 2005; Reinert et al., 2012). In this regard, it is important to consider the degree to which tamoxifen remains bioavailable when studying a complex process such as intestinal regeneration, where various ISC populations may be activated at different times to contribute to the regenerative response (Metcalfe et al., 2014).

Response to Nutrient Availability

Acute and chronic changes in nutrient availability have major physiological effects on many systems, including on the ISC compartment. For example, over-nutrition and obesity are known risk factors for intestinal neoplasia (Font-Burgada et al., 2016), and gain-of-function mutations in the IIS pathway typically lead to intestinal neoplasia (Cancer Genome Atlas, 2012). In contrast, the effect of under-nutrition on ISC behavior is less clear, raising key questions about the impact of nutrient availability on ISC activity. In Drosophila, ISCs have little or no response to acute nutrient deprivation, but are activated to proliferate in response to local insulin signaling during re-feeding (Choi et al., 2011; O’Brien et al., 2011). In contrast, work focusing on the CBC ISC population showed that long-term caloric restriction in mice leads to activation of CBC ISC self-renewal mediated indirectly by decreased mammalian target of rapamycin complex 1 (mTORC1) signaling in neighboring Paneth cells (Yilmaz et al., 2012). This discrepancy may be due to the significant physiological differences between long-term caloric restriction and acute fasting as more recent work from our laboratory has shown that in response to an acute fast, CBC ISCs exit cell cycle while q-ISCs increase in number (Richmond et al., 2015). Recent progress has led to the understanding that PTEN is an essential negative regulator of q-ISC number and activity. For example, in mice, fasting leads to transient PTEN phosphorylation (inactivation) within q-ISCs and a corresponding increase in their number. The loss of PTEN inhibition then renders q-ISC functionally poised to contribute to the regenerative response during re-feeding via cell-autonomous activation of the PI3K→AKT→mTORC1 pathway (Richmond et al., 2015). These results highlight a PTEN-dependent mechanism for q-ISC maintenance and raise the question of what other metabolic regulatory pathways regulate q-ISC behavior.

Intrinsic Metabolic Profiles

In addition to being responsive to nutrient availability on an organismal level, stem cells appear to exhibit metabolic profiles distinct from their differentiated progeny, which may have important implications for how they respond to environmental changes. Recent work in the field of stem cell metabolomics has demonstrated that cellular metabolism plays a pivotal role in determining whether a cell proliferates, differentiates or remains quiescent (Shyh-Chang et al., 2013). These findings have been made possible by important technical advances that allow for the analysis of key metabolic parameters, such as O₂ consumption, glycolytic rate, ATP production and respiratory capacity, in very small assay volumes, allowing for the study of rare populations. In the intestine, two recent papers have reported methods to evaluate the metabolic profile of both primary enteroids and isolated CBC ISCs (Bas and Augenlicht, 2014; Fan et al., 2015). In these studies, Lgr5^high colonic stem cells demonstrated decreased mitochondrial metabolism and increased glycolytic flux when compared with Lgr5^low daughter cells (Fan et al., 2015). These studies open up the possibility of visualizing, in real time, the metabolic changes taking place in ISCs in response to bioactive dietary agents, toxins, probiotics and other environmental stimuli.

Response to Environmental Stress

Recent studies have shown that q-SCs, especially those in the bone marrow, reside in a hypoxic (low oxygen) micro-environment, which contributes to their overall survival (Suda et al., 2011). In general, such q-SCs express high levels of hypoxia inducible factor 1 alpha (HIF1α) and rely on anaerobic glycolysis, rather than on mitochondrial oxidative phosphorylation for energy (Chen et al., 2008). Whether the q-ISC niche represents an hypoxic environment remains to be established. In addition to HIF1α, other factors associated with metabolic regulation of q-SCs include transcription factors, e.g., PPARδ and MEIS1 (Simsek et al., 2010; Ito et al., 2012), and tumor suppressors, e.g., liver kinase B1 (LKB1) (Gurumurthy et al., 2010). Q-SCs are also sensitive to increased levels of reactive oxygen species (ROS) and can either undergo differentiation or apoptosis in the presence of excessive ROS (Tothova et al., 2007; Rossi et al., 2008; Renault et al., 2009; Suda et al., 2011; Shyh-Chang et al., 2013). Along these lines, it has been suggested that a ‘ROS rheostat’ may help to monitor stem cell fate decisions (e.g., whether to undergo self-renewal or undergo lineage commitment) (Tothova and Gilliland, 2007; Paik et al., 2009). Additionally, the Forkhead box protein O (FOXO) family of transcription factors have been shown to protect q-SCs from high levels of ROS (Tothova and Gilliland, 2007; Tothova et al., 2007; Paik et al., 2009; Renault et al., 2009). A detailed understanding of the metabolic profiles of q-ISCs, their response to environmental stress, and the associated implications for physiology are important future goals.

Defining the q-ISC Molecular Signature

A number of molecular factors likely function to maintain q-ISCs in their quiescent state. Below we review what is known about some of these molecular and epigenetic factors and discuss how recent advancements in high throughput analyses of small numbers of stem cell populations will lead to valuable new insights regarding the q-ISCs molecular signature.

Transcriptional Profiling

Given their diverse functions, it is reasonable to assume that CBC ISCs and q-ISCs have unique transcriptional profiles/identities. Attempts to define the transcriptional signature of q-ISCs, vis-à-vis CBC ISCs, have led to some confusion in the field (Itzkovitz et al., 2012; Munoz et al., 2012; Li et al., 2014; Metcalfe et al., 2014; Roche et al., 2015). For example, analysis of the q-ISC populations marked by Lrig1-, Hopx-, Bmi1- and Sox9-expression, as well as q-ISCs defined using the label-retention assay, have revealed conflicting expression levels of mRNA transcripts associated with CBC ISCs and progenitor cells (Munoz et al., 2012; Li et al., 2014; Roche et al., 2015). While it is likely that some overlap in gene expression may be present between the two populations given that both function as SCs, it is possible that these data may be confounded by the fact that each ISC population has been harvested using reporter mice. In theory, this could lead to q-ISCs and/or their immediate progeny being marked at different stages, resulting in heterogeneous populations of labeled cells. Such heterogeneity has been described with several of these markers (Itzkovitz et al., 2012; Li et al., 2014). Additional studies are needed to further define the degree of heterogeneity within the q-ISC population.

Epigenetic Regulation and Identification of Gene Regulatory Networks

Epigenetic modifications can serve as a durable mechanism to regulate gene expression. Recent work has begun to identify the role of key epigenetic modifications (such as DNA methylation and histone modifications, as well as microRNA expression) in the maintenance of the quiescent state and cell fate decisions in various adult SC populations (Liu et al., 2013). In the intestine, however, it was recently shown that chromatin structure is broadly permissive between CBC ISCs and transit-amplifying progenitor cells (Kim et al., 2014). That is, under normal conditions, rapidly cycling progenitors and stem cells essentially share a common epigenome, raising questions about what role modification of chromatin structure will play in such a system that differentiates within only a few days. It remains possible, however, that chromatin structure and/or other epigenetic factors underlie key differences that are important for the regulation of q-ISCs. One strategy to address the question of whether q-ISCs differentially utilize such mechanisms involves (1) the identification of unique regions of open chromatin and (2) the putative transcription factor binding sites within these domains. Performing such an analysis at genome-wide resolution has the potential to discover gene regulatory networks that are unique to q-ISCs. ATAC-seq (Assay for Transposase-Accessible Chromatin using sequencing) (Buenrostro et al., 2013; Buenrostro et al., 2015), a recent technical advance, enables genome-wide analysis on small numbers of primary cells (e.g., ~5,000 q-ISCs), making it possible to define their gene regulatory networks. This innovative method will facilitate the identification of accessible chromatin and transcription factor binding sites using the in vitro transposition of sequencing adaptors into native chromatin. This technique provides a fast and sensitive method to pursue integrative epigenomic analyses, and comparison with CBC ISCs and other defined populations of intestinal crypt cells.

FUTURE DIRECTIONS

Moving forward, the ISC field has many remaining avenues of investigation. Chief among these is the comprehensive analysis of q-ISCs, which includes defining their metabolic profile and regulatory pathways, and defining their heterogeneity and relationship to CBC ISCs. In addition, optimizing enteroid culture conditions and transplantation assays to facilitate the functional analysis of q-ISCs will also be important.

Metabolic Profiling and Pathways Mediating Stress Responsiveness and Cell Fate

It will be important to identify which metabolic pathways are essential for the regulation of q-ISCs both at baseline and in response to injury. That is, do q-ISCs utilize anaerobic glycolysis or mitochondrial oxidative phosphorylation for energy? Do the pathways that regulate q-ISC metabolism also dictate their cell fate (e.g., maintenance of quiescence, proliferation or differentiation)? How do q-ISCs respond to reactive oxygen species (ROS) and what are the downstream transcription factors, epigenetic regulators, and microRNAs that regulate pluripotency, differentiation and self-renewal?

Optimized Culture Conditions

The advent of the enteroid culture system in the last several years has dramatically changed the field by allowing for the in vitro culture of primary intestinal epithelial cells (Sato et al., 2009), opening the door to a wide range of potential applications and translational opportunities. For example, 3-dimensional enteroids are being used in 2-dimensional microwells, organ-on-chips, and bioactive scaffolds (Keller, 1989; Bhatia and Ingber, 2014; Wang et al., 2014). These new bioengineering techniques raise the promise of being able to readily access the apical epithelial surface and more closely recapitulate the intestinal microenvironment in culture. While early studies have largely utilized epithelial-only cultures, the addition of different cell types (e.g., endothelial and mesenchymal cells) to these systems offers hope of increased complexity, functional activity and translational potential. A potential limitation of the most widely used enteroid culture protocols, which have been tailored to support the growth of CBC ISCs, is the requirement for large amount of Wnt ligands (Sato et al., 2009; Miyoshi and Stappenbeck, 2013), which may not be optimal for propagation of q-ISCs. This notion is supported by the observation that q-ISCs have dramatically reduced, or lack completely, the ability to form enteroids using current methods. It is also possible that current protocols may contain functional inhibitors or be lacking critical factors required to optimally support q-ISCs. Consistent with this, Lund and colleagues recently demonstrated that the addition of IGF-1 to the “standard” EGF/Noggin/R-spondin enteroid growth media facilitated the growth of single Sox9^high q-ISCs into enteroids (Van Landeghem et al., 2012; Van Landeghem et al., 2015). The development of a more “universal” media will both advance our understanding of the factors critical to the survival and regulation of q-ISCs as well as add physiologic fidelity to this important culture system.

Development of Transplantation Models

The ISC field currently lacks a truly robust transplantation assay to assess functional activity of ISCs in vivo. A few groups have been able to demonstrate engraftment of CBC ISCs, derived from enteroid cultures, into isolated patches of the distal colon following dextran sodium sulfate-induced colitis (Yui et al., 2012; Fordham et al., 2013). While providing important proof-of-principle that engraftment is possible, these experiments have been difficult to replicate elsewhere in the intestine and have not been successfully performed using q-ISCs. Alternative approaches involve transplantation of epithelial grafts into blind loops of small intestine (Avansino et al., 2006) or into a subcutaneous pocket (Montgomery et al., 1983), as well as expansion of full thickness intestinal grafts within the omentum (Barthel et al., 2012). Ultimately, the functional heterogeneity of ISCs will need to be established using robust functional assays that allow for assessment of a cell’s transplantation potential.

Understanding q-ISC Heterogeneity

To date, a number of genes have been identified that mark q-ISCs, raising the possibility that together these genes define a heterogeneous population of cells with a range of functional activity. Moreover, it remains possible that heterogeneity also exists within each subpopulation. Single cell mRNA sequencing has emerged as a powerful tool to measure cell-to-cell variability in gene expression (Li et al., 2014; Grun et al., 2015). An integrated understanding of a single cell’s gene expression profile can provide important insight into which signaling pathways are active in a given cell in addition to providing clues regarding the cell’s activity and functional potential (Shapiro et al., 2013; Treutlein et al., 2014). Using newly developed techniques like Drop-Seq (Klein et al., 2015; Macosko et al., 2015) to capture and sequence large numbers of single cells, it is now possible to generate single-cell transcriptional and ATAC-seq maps, which will, no doubt, provide important insight regarding the nature of q-ISCs (Buenrostro et al., 2015; Cusanovich et al., 2015). These studies can then be combined with both traditional and newer functional assays, such as the Microraft Assay, which facilitates the study of functional characteristics at single and/or small cell numbers (Gracz and Magness, 2014). Combining epigenetic, transcriptomic, proteomic, and surface marker profiling from single cell approaches with more functional analysis techniques will help establish a better understanding of q-ISCs and their distinct cellular identity.

Many questions remain to be explored in this rapidly evolving field but significant translational applications of solid organ stem cell research are on the horizon. A comprehensive understanding of q-ISCs may help inform targeted new therapies for millions of patients with gastrointestinal conditions including inflammatory bowel disease, intestinal cancer, malnutrition, chemotherapy-associated enteritis, and short bowel syndrome. The gastrointestinal scientific community has an important role to play in upcoming years to provide the framework for these necessary bio-medical advances.