Role of Rho GTPases in stem cell regulation

Abstract

The future of regenerative medicine relies on our understanding of stem cells which are essential for tissue/organ generation and regeneration to maintain and/or restore tissue homeostasis. Rho family GTPases are known regulators of a wide variety of cellular processes related to cytoskeletal dynamics, polarity and gene transcription. In the last decade, major new advances have been made in understanding the regulatory role and mechanism of Rho GTPases in self-renewal, differentiation, migration, and lineage specification in tissue-specific signaling mechanisms in various stem cell types to regulate embryonic development, adult tissue homeostasis, and tissue regeneration upon stress or damage. Importantly, implication of Rho GTPases and their upstream regulators or downstream effectors in the transformation, migration, invasion and tumorigenesis of diverse cancer stem cells highlights the potential of Rho GTPase targeting in cancer therapy. In this review, we discuss recent evidence of Rho GTPase signaling in the regulation of embryonic stem cells, multiple somatic stem cells, and cancer stem cells. We propose promising areas where Rho GTPase pathways may serve as useful targets for stem cell manipulation and related future therapies.

Introduction

Mammalian Rho family GTPases are members of the Ras superfamily including over 20 gene products which can be further subdivided into six groups based on sequence identity, domain structure and function: RhoA subfamily proteins, Rac proteins, CDC42 proteins, Rnd proteins, Rho BTB proteins and Miro proteins [1,2]. Rho family GTPases play critical roles in multiple cell processes including gene expression, cytoskeletal dynamics, survival, cell division, cell adhesion, polarity and vesicle trafficking [3,4]. Dysfunction of Rho GTPase regulated signaling can lead to multiple cancers, neurodegenerative disorders, immunological disorders, and developmental abnormalities [5]. Most Rho family members act as molecular switches that cycle between an active GTP-bound form and an inactive GDP-bound form and active Rho GTPases turn on a cascade of signaling events leading to cellular responses.

Stem cells are characterized by their ability to perpetuate themselves through self-renewal and giving rise to mature specialized cell types [6]. Based on their potency and origins, naturally occurring stem cells can be categorized into embryonic and somatic. Embryonic stem cells (ESCs) originate from inner cell mass of the blastocyst and are pluripotent with an indefinite replicative capability. Somatic or adult stem cells, on the other hand, are less totipotent with limited replicative life span than ESCs, and exist in mature tissues such as hematopoietic, neural, gastrointestinal and mesenchymal to sustain life-long tissue homeostasis. The ability of adult stem cells to divide and regenerate mature cells of a particular tissue to replace damaged tissues provides the organism with an internal repair system in response to tissue damage and stress. Tremendous efforts have been made to understand how diverse signaling pathways and regulators coordinate stem cell functions during development, homeostasis and disease/injury. The future of regenerative medicine lies in developing cell-replacement therapies to treat debilitating diseases founded on stem cell research [7].

A major focus of stem cell biology is to understand the molecular and cellular mechanisms underlying stem cell self-renewal and differentiation. In homeostasis, stem cells must strike a balance between self-renewal and differentiation, and recent studies using mouse gene targeting and small molecule inhibitors of individual Rho GTPases such as RhoA, Rac1, and Cdc42 have established that the Rho GTPase family is one of the essential regulators of embryonic and adult stem cells by controlling tissue-specific stem cell migration, niche interaction, proliferation/survival, self-renewal and differentiation (Table 1). In this review, we summarize and update recent discoveries of Rho GTPase signaling in stem cell regulation.

Table 1.

Summary of the reported roles of Rho GTPases in various stem cells

Cell type	Rho GTPase	Function and Mechanism	Reference
Embryonic Stem Cells (ESCs)	Rho/ROCK	Cause dissociation-induced apoptosis of human ESCs and dramatically decrease cloning efficiency	[1]
		Promote colony formation and maintain them at undifferentiated state at high seeding density	[2]
	RhoA	Mediates fibronectin-regulated mouse ESCs proliferation and migration	[3]
		Maintain cell–cell contacts in mouse ESCs	[4]
		Associated with enhanced self-renewal and reduced differentiation in ZO-1 null mouse ESCs	[5]
	RhoC	Required for human ESCs survival and long-term growth in vitro via modulating actin microfilament organization and YAP/TAZ signaling	[6]
	Rac1/Cdc42	Increases mouse ESCs migration via PAK1 mediated E-cadherin complex disruption	[7]
Hematopoie-tic stem cells (HSCs)	RhoA	Critical for multilineage hematopoiesis by regulates actomyosin signaling, cytokinesis, and programmed necrosis of the hematopoietic progenitors	[8]
	RhoC	Regulated by RhoGAP Arhgap21, it reduces HSCs’ ability to form progenitor colonies in vitro and engraftment ability in vivo, along with a decrease in LSK cell frequency during serial bone marrow transplantation	[9]
	Cdc42	Elevated in aging HSCs and mediates non-canonical Wnt5A induced aging phenotypes in HSCs	[10]
		Control HSCs asymmetric division and fate by forming complex with Scribble/Yap1	[11]
	Rac1/Rho	Enhanced Rac/Rho/integrin signaling by MLL1 induction increases hematopoietic potential	[12]
Neural stem cells (NSCs)	RhoA/Rac1	RhoA activation and Rac1 inhibition inhibit NSCs migration by reducing actin filament polymerization	[13]
	RhoA	Activation in hippocampal progenitors suppresses neurogenesis by regulating cellular stiffness	[14]
		Inhibition of RhoA leads to neurite outgrowth of NSCs and improves neuronal differentiation of NSCs	[15]
		Loss of RhoA impairs neural progenitor cell differentiation and cause delay during embryonic neurogenesis due to disrupted cell cycle gene expression	[16]
		Deletion of RhoA in midbrain and forebrain neural progenitors cause massive dysplasia of the brain due to accelerated proliferation, reduction in cell-cycle exit, and increased expression of downstream target genes of the hedgehog pathway	[17]
	Rac1	Promote NSC migration by regulating actin cytoskeleton	[18]
		Required for the normal proliferation and differentiation of SVZ progenitors and for survival of both VZ and SVZ progenitors	[19]
	Cdc42	Maintains NSC quiescence by inducing expression of molecules involved in NSCs identity and anchorage to the niche	[20]
	Rnd3/RhoE	Maintain radial fiber scaffold responsible for guided neurons migration by inhibiting Notch signaling	[21]
Intestinal stem cell (ISCs)	RhoA	RhoA activate canonical Wnt and YAP signaling, and RhoA deficiency resulted in reduced cell proliferation, increased apoptosis, and a loss of ISCs	[22]
	Cdc42	Required for the functioning of the Lgr5+ ISCs by activating Rab8a and its multiple effectors	[23]
		Loss of CDC42 caused intestinal hyperplasia and ISCs to TA cells fate switch by regulating YAP-Ereg-mTOR signaling	(author’s unpublished data)
		Ectopic CDC42 enhances ISCs regeneration by mediating EGF-stimulated, receptor-mediated endocytosis and sufficient to promote MAPK signaling	[24]
		Loss of CDC42 promotes regenerative potential of aged ISCs	[25]
	Rac1	Necessary and sufficient to drive ISC proliferation and regeneration in an ROS-dependent manner	[26,27]
Mesenchymal stem cells (MSCs)	RhoA/ROCK	MSCs differentiate into myofibroblasts when RhoA/ROCK is turned on, into endothelial cells when turned off, and RhoA promotes formation of an extracellular matrix complex consisting of connective tissue growth factor and vascular endothelial growth factor	[28]
		Regulate fluid-flow-induced osteogenic differentiation via the regulation of Runx2, Sox9 and PPARgamma	[29]
		Inhibiting RhoA signaling promotes chondrogenic differentiation by increasing the expression of SOX-9	[30]
		Suppression of ROCK signaling, concomitant with EGF and bFGF stimulation enhances the differentiation of bone marrow-derived MSCs into neurons and neuroglial cells	[31]
		Mediates the VEGF-induced differentiation of human and rat bone marrow-derived MSCs into endothelial cells.	[32]
	RhoA/Cdc42/Rac1	Mediates TGFbeta-induced myofibroblast differentiation of MSCs via actin polymerization	[33]
	Rac1/Cdc42	Regulate migration of human MSCs via E-cadherin reduction	[34]
	Cdc42	Elevated CDC42 activity in aged rat adipose-derived MSC leads to senescence	[35]
Cancer Stem Cells (CSCs)	RhoA/ROCK	Promotes breast CSC formation by YAP/TAZ activity in breast CSCs	[36]
		Targeting Rho/ROCK Pathway is important for radiation resistance of CSCs	[37]
	RhoC	Regulates metastatic potential of breast CSCs and tightly associated with the breast CSCs marker ALDH1	[38]
	Cdc42/Rac1	Associated with cancer stem cell properties including holoclone, sphere formation, invasion	[39]
	Rac1	Promotes the self-renewal of breast CSCs and mediates TAZ activation by VEGF signaling	[40]
		Regulate migration and promote chemotherapeutic sensitivity in Leukemia stem cells via regulating the expression of several cell intrinsic cell cycle inhibitors, as well as the extrinsic molecules that mediated the interaction of Leukemia stem cells with osteoblastic niche	[41]
		Involved in the glioma stem-like cell maintenance and tumorigenicity via the Rac1-Pak signaling	[42]
		Promotes the EMT in gastric adenocarcinoma and the acquisition of a CSC state	[43]
		Regulates proliferation, invasion, sphere formation and lung colonization in in human NSCLA	[44]
Epidermal stem cells	Cdc42	Essential for epidermal stem cells and epidermal development via Cdc42-SPRR pathway, which may correlate with epidermal barrier function	[45]
	RhoA/Rac1	Required for epidermal stem cell migration by regulating F-actin polymerization	[46]
	Rac1	Indispensable for the fate of epidermal stem cells by controlling their exit from the stem cell niche	[47–49]
Human dental pulp stem cells	Rho/ROCK	Implicated in the differentiation of human dental pulp stem cells by down-regulation of Runx2 expression	[50]

Rho GTPases in embryonic stem cells

ESCs are cells derived from the early embryo that can be maintained in the primitive undifferentiated state indefinitely while remaining pluripotent [8]. Rho GTPases have been implicated in multiple physiological and pathological processes in the ESC regulation (Figure 1).

Figure 1. Schematics of the recognized role of Rho GTPases in embryonic stem cells.

RhoA regulates fibronectin mediated ESC migration, while Rac1 and Cdc42 are activated by laminin to enhance E-cadherin complex disruption and increase ESC migration. ROCK inhibition via Y-27632 significantly decreases ESC apoptosis while RhoA activity is associated with enhanced self-renewal and reduced differentiation. RhoC signaling is required for human ESC survival and long-term growth.

The Rho-associated kinase (ROCK) inhibitor, Y-27632, could markedly reduce dissociation-induced apoptosis of human ESCs and significantly increase cloning efficiency [9], as a ‘Rho-high/Rac-low’ state appears to initiate dissociation-induced actomyosin hyperactivation and apoptosis in the ESCs [10]. Y-27632 has also been used to improve human ESC pluripotency [11] where ROCK activation is thought to affect cellular motility [12]. More recent study found that RhoC signaling is required for human ESC survival and long-term growth in vitro by modulating actin microfilament organization and sustaining the nuclear function of the transcriptional cofactors YAP and TAZ [13]. Furthermore, it was reported that ABR, a modulator of Rho GTPase activities, is an upstream factor controlling the survival vs. death decision of dissociated human ESCs [14].

Murine ESC adhesion to fibronectin stimulated cell proliferation and migration in a RhoA dependent manner [15,16]. Functional analysis of the promoters of the RhoA and the RhoB genes in mouse ESCs showed significantly higher RhoA mRNA levels than RhoB, and expression of both genes was inhibited by their own protein products [17]. In addition, RhoA is also required in early embryogenesis for the maintenance of intercellular junctions in mouse ESCs [18]. Separately, Rac1/Cdc42 can be activated by laminin, the first extracellular matrix component expressed in the developing mammalian embryo, to enhance E-cadherin complex disruption leading to increased ESCs migration [19].

The role of Rho GTPases in the differentiation of ESCs is relatively under studied. Due to the similarity of ESCs and induced pluripotent stem cells (iPSCs) which is undergoing fast-evolving studies, understanding how Rho GTPases regulate the controlled differentiation of iPSCs towards different tissue specific progenitor lineages will help gain insights into ESC differentiation mechanisms in the future.

Rho GTPases in hematopoietic stem cells

While adult stem cells from a variety of organs have the potential to be useful for therapeutic applications in the future, the hematopoietic stem cells (HSCs) have already been extensively used for clinical purposes, mainly by bone-marrow or cord blood-related transplantation. HSCs and hematopoietic progenitor cell (HPCs) in the bone marrow or cord blood initiate the production of multilineage hematopoietic cells via two critical features of HSCs: indefinite self-renewal and multilineage differentiation potential. Rho GTPases are required for most aspects of HSC functions including lineage determining gene transcription, cell survival, cell cycle progression and homing and migration. Alterations of HSC function through Rho GTPases contribute to various human diseases including leukemogenesis, bone marrow failure syndromes, immunodeficiencies, and blood aging phenotypes [1] (Figure 2). The function of Rho GTPases especially Rac1 and Rac2 in the hematopoietic system has been well studied in HSCs and mature hematopoietic cells, and earlier work has been summarized [20]. Research focusing on their role in HSCs have revealed the essential contributions of RhoA, Rac1/2, Cdc42 and other Rho GTPases/regulators in HSC adhesion, migration, bone marrow retention, proliferation, survival, senescence and oncogenic transformation [1,21].

Figure 2. Schematics of the role of Rho GTPases in hematopoietic stem cells.

RhoA regulates actomyosin signaling, cytokinesis, and programmed necrosis of the hematopoietic progenitors. Cdc42 is a master regulator of HSC engraftment, quiescence, niche interaction, and aging associated loss of polarity phenotypes, while Rac1/Rac2 play critical roles in the HSC engraftment and migration.

Earlier studies of the role of RhoA in HSC regulation utilized mouse models expressing dominant negative RhoA (RhoAN19) and showed that partial inhibition of RhoA caused an enhanced HSC self-renewal activity and long-term engraftment capability [22]. Whereas loss of p190-B Rho GTPase activating protein, a negative regulator of RhoA, induced elevated RhoA activity and the associated increased HSC self-renewal activity and long-term engraftment [23]. Subsequent RhoA conditional knockout in a mouse model demonstrated directly that RhoA deficiency in HSCs caused multilineage hematopoietic failure with a blockage of hematopoiesis at the multipotent progenitor stage [24]. RhoA-null HSCs still possess long-term engraftment potential but are incapable of producing multipotent progenitors and differentiated blood cells, as RhoA regulates actomyosin signaling, cytokinesis, and programmed necrosis of the hematopoietic progenitors [24]. Later studies in human patients found that vasculopathy-associated hyperangiotensinemia induces a mobilization of HSCs and progenitors through endothelial AT2R and RhoA signaling [25].

Studies of the function of other Rho GTPases have found that CDC42 is a master regulator of HSC engraftment, quiescence, niche interaction, and aging associated loss of polarity phenotypes [26,27]. Expression of Wnt5a and Cdc42 are elevated in aged primitive hematopoietic cells and Wnt5a activated Cdc42 is involved in the induction of ageing-like myeloid skewage and HSC engraftment defects of HSCs [28]. The recently identified Cdc42 activity-specific inhibitor, CASIN, could inhibit intracellular Cdc42 activity transiently and cause mouse HSCs and progenitor cell egress from bone marrow by suppressing actin polymerization, adhesion, and directional migration of HSCs, which is useful to study biological and pathological roles of Cdc42 in HSCs as well as other tissue cells [29]. More recent work have shown that the cell planar polarity molecule Scribble and the Hippo pathway component Yap1 coordinate to control cytoplasmic Cdc42 activity and HSC fate in vivo [30]. On the other hand, increased RhoC activity in HSCs by loss of Arhgap21, a member of the Rho GAP family, led to impaired ability to form progenitor colonies in vitro and decreased engraftment ability in vivo, along with a decrease in multiple potential progenitor cell frequency in serial bone marrow transplantation and reduced erythroid commitment and differentiation [31]. Studies in zebrafish and iPSCs indicate that members of the Rho GTPase family act upstream of YAP to regulate HSCs and progenitor cell production [32]. Induction of MLL1, a gene critical for development and maintenance of HSCs, in mouse ESCs enhances Rac1/RhoA/integrin signaling and selectively affects c-Kit+/CD41+ hematopoietic progenitor function and increases the ESC differentiated HPC hematopoietic potential [33].

The role of Rho GTPases in HSCs is one of the most studied among stem cell subtypes. RhoA, Rac and Cdc42 all have been demonstrated to have essential roles in multiple aspects of HSC function. Ongoing studies of Rho GTPases and their regulators/inhibitors in HSCs and hematological diseases are moving toward translation in HSC transplantation, HSC aging, and myeloid dysplasia/transformation arenas.

Rho GTPases in neural stem cells

Neural stem cells (NSCs) reside in the ventricular and subventricular zones (VZ and SVZ) and give rise to the highly elongated radial glial progenitors as the central nervous system (CNS) matures during development [34,35] (Figure 3). NSCs play key roles in the developing CNS, including generating neurons and macroglial cells and guiding the migration of newborn neurons to their proper destinations [36]. In adult, NSCs are largely quiescent during homeostasis, and upon CNS injury, NSCs begin proliferating and differentiating to three major CNS cell types — neurons, astrocytes, and oligodendrocytes — while migrating towards the lesions to repair injury [37]. The molecular mechanisms that trigger the activation of NSCs for neural tissue repair were unclear until a recent study found that NSC quiescence is maintained by Cdc42 which is in turn regulated by non-canonical Wnt signaling [38]. Cdc42 induces the expression of molecules involved in NSC identity maintenance and anchor the NSCs to the niche, while during demyelination injuries, down-regulation of Cdc42 activity is required for the activation and lineage progression of quiescent NSCs for tissue repair [38].

Figure 3. Schematics of the key roles of Rho GTPases in neural stem cells.

Cdc42 maintains NSC quiescence, and down-regulation of Cdc42 activity results in the activation of quiescent NSCs for tissue repair. NSC migration relies heavily on Rho GTPase (Rac1, RhoA and Rnd3/RhoE)-controlled actin and microtubule cytoskeleton, whereas RhoA has diverse roles during neurogenesis. The blue and brown tracts indicate the blood vessels.

NSC migration relies heavily on the regulation of actin and microtubule cytoskeleton, and Rho GTPases are critical regulators of key steps during NSC migration [39,40]. Fasudil, a ROCK inhibitor, drives mobilization of adult neural stem cells after hypoxia/reoxygenation injury in mice [41], and overexpression of Rho-GDIγ (RhoGDI-3/ARHGDIG), mainly a down-regulator of Rho GTPases while also having a positive role as chaperones in certain cell-context, maintains NSCs in the stem cell state, prevents NSC migration and inhibit of Rac1 expression [42]. A recent study found that the maintenance of radial fiber scaffold responsible for guided neurons migration, and consequently axon pathfinding, are dependent on an atypical Rho-GTPase, Rnd3/RhoE and its binding partner ARHGAP35/p190A, a RHO GTPase-activating protein [36], and genetic deletion of Rnd3 in NSCs promotes their proliferation via up-regulation of Notch signaling [43]. In cultured primary NSCs, RhoA activation and Rac1 inhibition caused by the loss of actin alpha 2 result in reduced actin filament polymerization and an inhibition of NSC migration [37], suggesting that the regulation of actin and microtubule cytoskeleton by Rho GTPases is important for NSC migration.

Self-renewal and differentiation of NSCs depend not only on the intrinsic biochemical and genetic factors but also on the mechanical cues in their microenvironment such as extracellular matrix (ECM) stiffness [44]. ECM-derived mechanical signals act through Rho GTPases to activate the cellular contractility machinery and regulate NSC lineage commitment, as NSCs cultured on stiff ECMs showed elevated RhoA and Cdc42 activities and suppressed neurogenesis, whereas RhoA activation in hippocampal progenitors also suppresses neurogenesis [45]. During in vitro culturing of human NSCs, it is found that YAP-mediated mechano-sensing regulates neurogenesis and lineage commitment of the NSCs and the intracellular localization of YAP is regulated by RhoA activity [46]. A later study found that RhoA inhibition can also affect neuritogenesis and differentiation of NSCs by promoting neuronal differentiation and neurite outgrowth in NSCs isolated from mouse SVZ [47]. A recent study identified a Rho GTPase-activating protein (RhoGAP), α2-chimaerin, in regulating the transition of NSCs into intermediate progenitor cells, and conditional knockout of α2-chimaerin resulted in a loss of NSC population by premature differentiation with impaired neuron production [48]. On the contrary, suppression of RhoA activity by a dominant-negative RhoA mutant impaired neural progenitor cell differentiation and caused a delay of embryonic neurogenesis [49], suggesting that RhoA has diverse roles at different developmental stages. Interestingly, spontaneous neural differentiation from mouse ESCs can be induced by Noggin, and RhoA activity in the ESCs prevents neural differentiation by inhibiting Noggin transcription and limiting Noggin secretion [50]. Furthermore, ROCK inhibitor Y-27632 treatment of ESCs promotes the differentiation of the cells into neurons [51].

More detailed analyses of various populations of neural stem/progenitor cells have seen more selective roles of various Rho GTPases in the regulation of self-renewal and differentiation. Conditional deletion of RhoA in midbrain and forebrain neural progenitors disrupted apical adhesion junctions and caused a massive dysplasia of the brain due to accelerated proliferation, reduction in cell-cycle exit, and increased expression of downstream target genes of the hedgehog pathway, resulting in an expansion of neural progenitor cells and exencephaly-like protrusions [52]. Conditional deletion of Cdc42 at different developmental stages of neurogenesis in mouse telencephalon also caused decay of adhesion junctions leading to a loss of the neuroepithelial polarity and failure of apically directed interkinetic nuclear migration [53]. On the other hand, Cdc42 deficiency in the developing forebrain stem cells disrupted apical-basal polarity and led to disorganized and expanded neural progenitors throughout the entire neuroepithelium [54].

Thus, Rho GTPases are important for the activation, migration and differentiation of NSCs and neural progenitors (Figure 3). Due to their important function in regulating polarity and cell adhesions, it is not surprising that individual Rho GTPases are critical for migration-related functions of NSCs in injury recovery and posits them as valuable candidate targets in the treatments of neuronal injuries and neuronal degenerative diseases.

Rho GTPases in intestinal stem cells

Intestinal epithelium is one of the most vigorously self-renewing tissues in adult mammals; the entire lining of the human gut replenishes approximately every 7 days [55]. The function of several Rho GTPases in intestinal stem cell regulation (ISCs) has become more clear recently (Figure 4). Genetic deletion of Cdc42 in mice results in massive proliferation, disrupted epithelium morphogenesis, impaired intestinal epithelium polarity, and defective Paneth cell localization [56]. Another Cdc42 knockout study found that Cdc42 deficiency perturbs crypt homeostasis and reduced the function of the Lgr5+ ISCs [57]. An ISC-specific deletion of Cdc42 in mice found that Cdc42 controls apical/basal lateral polarity which regulates ISC to transit amplifying (TA) cell transition and proliferation at the expense of ISC population via a canonical Hippo signaling of YAP/TAZ and the consequent Ereg mediated mTOR pathway (author’s unpublished data), positioning Cdc42 regulated epithelial polarity at the center of ISC/TA cell proliferation and cell fate control. Recent studies have also found that pharmacological suppression of elevated Cdc42 activity restores the regenerative potential of aged ISCs [58], and extends the general lifespan of mice [59].

Figure 4. Schematics of the role of Rho GTPases in intestinal stem cells.

Rac1/RhoA/Cdc42 regulate ISC proliferation and regeneration via different downstream effectors. RhoA and Cdc42 are required for intestinal epithelial polarity but distinct downstream signaling through the Hippo signaling effecting on ISC proliferation. Cdc42 uniquely contributes to the ageing effect of ISCs and ISC into TA cell transition. TA: Transit amplifying cells; PC; Paneth cells; ISC: Intestinal stem cells; GB: Goblet cells; EE: Enteroendocrine cells; TC: Tuft cells.

Aside from Cdc42, RhoA GTPase also controls epithelial polarity as well as YAP-mediated Ereg signaling in ISCs as RhoA deficiency results in a reduced proliferation, increased apoptosis, and loss of ISCs that mimic effects of radiation damage [60]. On the other hand, Rac1 works downstream of canonical Wnt signaling and is necessary and sufficient to drive ISC proliferation and regeneration in an ROS-dependent manner [61,62]. Recent studies revealed further that the Wnt target Lgr5 primarily functions via the IQGAP1-Rac1 pathway to strengthen cell-cell adhesion in ISCs [63]. It remains to be seen how the interactions of each Rho GTPase with other known regulators of ISCs such as Wnt, Hippo, and Notch pathways are integrated together in the ISC maintenance, particularly after a stress in the intestine.

Rho GTPases in other somatic stem cells

While the available literature remains relatively sparse, Rho GTPases have also been implicated in the regulation of other somatic stem cell types including epidermal, teeth, cardiac and mesenchymal stem cells. In vivo study showed that Cdc42 deficiency leads to the loss of epidermal stem cells [64], whereas Rac1 and RhoA, but not Cdc42, are required for epidermal stem cell migration [65]. Furthermore, both in vivo and in vitro studies have found that loss of Rac1 causes epidermal stem cell depletion [66–68]. The ROCK inhibitor Y-27632 promotes proliferation of stem cell from human exfoliated deciduous teeth [69], and the Rho/ROCK signaling is implicated in the differentiation of human dental pulp stem cells [70]. In the growing mouse incisor, Cdc42 was found to activate YAP which drives the proliferation of transit-amplifying cells and affects stem cell-based tissue renewal in mice [71]. Additionally, Y-27632 significantly increased cell survival rate, adhesion, and migration in cardiac stem cells [72].

It has been shown that TGFβ-activated RhoA/ROCK signaling functions as a molecular switch regarding the fate of mesenchymal stem cells (MSCs) in arterial repair/remodeling after injury, and MSCs can differentiate into myofibroblasts when RhoA/ROCK is turned on but to endothelial cells when turned off [73]. In bone marrow, RhoA and its effector ROCKII regulate fluid-flow-induced osteogenic differentiation [74], while RhoA GEFs and GAPs work through RhoA to influence MSC fate [75]. The function of Rho GTPases in skeletal myogenesis was summarized previously [76]. Interestingly, Vav1, a Rho/Rac GEF highly expressed in MSCs, regulates MSC differentiation decision between adipocyte and chondrocyte [77]. Suppression of ROCK signaling enhances the efficacy of bone marrow-derived MSC differentiation into neurons and neuroglial cells [78], and pharmacologic experiments found that elevated Cdc42 activity in aged rat adipose-derived MSCs leads to senescence and reduced differentiation potential while inhibition of Cdc42 activity can, at least partially, rejuvenate aged MSCs [79].

Rho GTPases in cancer stem cells

The cancer stem cell (CSC) theory proposes that a tumor is hierarchical in cellular components and there is a subpopulation of tumor cells with enhanced self-renewal and tumor seeding activities upon experimental implantation [80] (Figure 5). In many myeloid leukemia and solid tumor types including acute myeloid leukemia, glioblastoma, breast cancer, colorectal cancer and skin squamous-cell cancers [81], CSCs are active architects of their microenvironment to maintain a supportive niche [82]. In many aforementioned tumor types, only a small subset of the cancer cells contains the capability of tumor invasiveness, therapy resistance and tumor initiation, positing CSCs in the central scheme of cancer biology with great clinical potentials [81,83].

Figure 5. Schematics of the function of individual Rho GTPases in select cancer stem cells.

Rho GTPases contribute to the initiation and progression of multiple types of tumors, and play major roles of the metastatic process. ROCK and Rac1 are implicated in the chemo/radiation therapy resistance, and together with other pro-oncogenic Rho GTPases, are emerging as novel therapeutic targets.

In human breast cancer, the CSC marker ALDH1 is tightly correlated with RhoC expression, and in vitro analysis and xenograft studies further suggest that RhoC regulates ALDH1 and endows a metastatic potential of breast CSCs [84]. Recent studies found that a STARD13-correlated ceRNA network, which is inhibitory to breast CSCs, is negatively correlated with RhoA-ROCK signaling mediated YAP/TAZ activity, suggesting that inactivating RhoA GTPase/F-actin pathway might be a useful approach to suppress YAP/TAZ activity and inhibit breast CSC formation [85]. Accumulating evidence have shown that YAP/TAZ of the hippo pathway are essential for the function of CSCs and other aspects of aggressive tumor behavior [86], and Rac1 is critical in TAZ activation by VEGF signaling which promotes the self-renewal of breast CSCs [87]. Using selective pharmacologic inhibitors, it is found that in gastric CSC-like cells, activation of PAR1 inhibits Lats1/2 kinase and activates YAP/TAZ activity via a ROCK-independent Rho GTPase pathway [88], consistent with a key role of Rho family GTPases in promoting YAP/TAZ activation observed in endothelial cells [89].

CSCs derived from melanoma and breast cancer cell lines exhibit increased contractility as well as ECM remodeling capacity, and more importantly, a blockade of ROCK signaling completely abolishes the contractility and collagen degradation capacity of both CSCs and non-CSCs, suggesting a potential of pharmacological targeting of ROCK pathway in CSCs and overall tumor population in the treatment of metastasis of these cancers [90]. Interestingly, forced expression of miR-141 in prostate CSCs and progenitor cells directly inhibits Rho GTPase family members/signaling partners Cdc42, CDC42EP3 (BORG2/CEP3), Rac1 and ARPC5 (a component of the Arp2/3 complex), causing reduced CSC-related activities including holoclone, sphere formation, invasion, as well as tumor regeneration and metastasis [91]. On the other hand, RhoA knockout fibroblasts lose their normal tumor-inhibitory capacity and gain the tumor stem cell-promoting properties including induced cancer cell migration and proliferation in vitro and tumor growth in vivo [92], whereas ROCK inhibitor Y-27632 and fasudil can sensitize pancreatic CSCs to chemotherapy response to Gemcitabine [93]. There is also evidence that the Rho/ROCK Pathway are involved in radio-resistance and targeting Rho/ROCK Pathway offers a potential solution to overcoming radiation resistance of CSCs [94].

Inactivation of Rac1 causes an impaired migration and enhanced chemotherapeutic sensitivity in leukemia stem cells [95]. In human glioma cell lines, Rac1 is involved in the glioma stem-like cell maintenance and tumorigenicity [96]. Rac1 activity is also significantly higher in spheroid-forming or CD44+ gastric adenocarcinoma CSCs, and a cell line study using a xenograft model indicated that Rac1 promotes the epithelial-to-mesenchymal transition (EMT) in gastric adenocarcinoma and the acquisition of a CSC state [97]. Furthermore, knockdown of Rac1 in human non-small cell lung adenocarcinoma (NSCLA) cells decreases proliferation, invasion, sphere formation and lung colonization [98].

While solid tumors require Rho GTPases for migration, dissemination and metastasis, hematological malignancies depend on Rho GTPases for survival and proliferation. In acute myeloid leukemia, elevated Cdc42 activity was associated with a polarized transformation of HSPCs to acute myeloid leukemia (AML) with altered cell division symmetry whereas elevated Rac1/Rac2 activity is crucial for Bcl-2 mediated leukemia initiating cell survival; a knockdown or small molecule targeting of Cdc42 or Rac in primary murine or human AML cells suppressed leukemia progression in xenograft models [99,100], implicating Cdc42 and Rac as potential targets for inhibiting leukemia stem cell. In solid tumors, hyperactive Rac1 and Cdc42 are implicated in EMT, cell-cycle progression, migration/invasion, tumor growth, angiogenesis, and oncogenic transformation, and an up-to-date review of Rac1 and Cdc42-targeting small molecules has been summarized elsewhere [101]. On the other hand, the RhoA gene is often deleted or mutated rather than amplified in many cancer types, suggesting a tumor suppressor role in defined cancer types [2]. However, RhoA inhibitors still have clinic potential for certain cancers where RhoA is activated. For example, a small molecule Y16 which selectively suppressed RhoA GEF signaling was able to inhibit mammary sphere formation in MCF7 breast cancer cells [102], and more recently, JK-136 and JK-139, two of the anti-RhoA hydrazide derivatives, were found to inhibit gastric cancer activities in mice [103]. So far, the majority of Rho GTPase inhibitors remain at the pre-clinical testing stage, but pharmaceutical development and research of Rho GTPase inhibitors will continue to hold promise as a therapeutic strategy.

Last but not the least, most of our understanding of Rho GTPase function in CSCs were obtained using cancer cells in vitro or in xenografts in immunocompromised mice which lack consideration of the tumor microenvironment. The development of improved preclinical models such as 3D co-culture, organoids and patient-derived xenografts in humanized animal models will increase our understanding for the role of Rho GTPases in CSCs.

Overall, Rho GTPases play diverse roles (either pro-oncogenic or tumor-suppressive) in cancer stem cells depending on the tissue of origin, progression stage, and degree of differentiation (Figure 5), presenting a challenge for rational design of anti-cancer therapies based on targeting Rho GTPase signaling.

Conclusion

Rho family GTPases are known to regulate diverse cellular responses to morphogens, adhesion molecules, growth factors/cytokines, and chemokines involved in stem cell proliferation, polarity and cytoskeletal dynamics [104]. During the last decade, chemical inhibitors and mouse genetics have allowed rapid discoveries of the essential roles of individual Rho GTPase family members in self-renewal, differentiation, migration, survival and/or niche interaction of embryonic and various somatic stem cells (Table 1). Related studies of the large arrays of Rho GTPase regulators/effectors in stem cell regulation, which are not extensively covered here due to page limit, also add to our ever-growing knowledge of Rho GTPase mediated tissue specific signaling in stem cell subtypes presents new opportunities for stem cell manipulation and therapeutics. Understanding how Rho GTPase signaling is controlled in vivo remains a challenge and important future directions include: in-depth mechanistic investigation of the interaction between Rho GTPase regulated polarity/adhesion and other intracellular signaling pathways elicited from diverse stem cell microenvironments in temporal/spatial regulation of subtypes of stem cells, and the discovery of novel therapeutic opportunities using selective pharmaceutical activation/inhibition of Rho GTPases in physiologic or pathologic settings. Such further studies of Rho GTPase function will shine light on the basic mechanisms of stem cell biology and their translational potential in stem cell-related diseases.

Perspectives.

• Importance of the field: Stem cell derived tissue/organ development and regeneration provide great potentials for the future of regenerative medicine. A comprehensive understanding of regulatory mechanisms in embryonic/adult stem cells as well as in cancer stem cell is crucial for stem cell biology translation to future therapies. Rho family GTPases are important regulators for diverse cell biological functions, and their role in stem cell function and as potential therapeutic targets have been actively investigated in the past decade, yielding important insights into tissue specific stem cell regulatory mechanisms.

• Current thinking: Rho family GTPases play critical regulatory roles in multiple intracellular signaling functions of various stem cells to regulate embryonic development, adult tissue homeostasis, and tissue regeneration upon stress. In addition, Rho GTPases and their upstream regulators or downstream effectors are also involved in the transformation, migration, invasion and tumorigenesis of cancer stem cells, and recognized to be potential targets for cancer therapy.

• Future directions: Future in-depth studies into the mechanisms on how Rho GTPase-regulated polarity/adhesion may coordinate with intracellular signaling pathways for the temporal/spatial regulation of stem cells and the discovery of novel therapeutic approaches using selective pharmaceutical activation/inhibition of Rho GTPases, will shine a light on the mechanisms of stem cell biology and reveal their translational potential in stem cell-related diseases.

Abbreviations

AML

acute myeloid leukemia

CNS

central nervous system

CSC

cancer stem cell

ECM

extracellular matrix

EMT

epithelial-to-mesenchymal transition

ESCs

embryonic stem cells

HSCs

hematopoietic stem cells

ISC

intestinal stem cell

MSCs

mesenchymal stem cells

NSCLA

non-small cell lung adenocarcinoma

TA

transit amplifying