Role of microRNAs in gastrointestinal smooth muscle fibrosis and dysfunction: novel molecular perspectives on the pathophysiology and therapeutic targeting

Abstract

MicroRNAs (miRNAs) belong to a group of short noncoding RNA molecules with important roles in cellular biology. miRNAs regulate gene expression by repressing translation or degrading the target mRNA. Recently, a growing body of evidence suggests that miRNAs are implicated in many diseases and could be potential biomarkers. Fibrosis and/smooth muscle (SM) dysfunction contributes to the morbidity and mortality associated with several diseases of the gastrointestinal tract (GIT). Currently available therapeutic modalities are unsuccessful in efficiently blocking or reversing fibrosis and/or SM dysfunction. Recent understanding of the role of miRNAs in signaling pathway of fibrogenesis and SM phenotype switch has provided a new insight into translational research. However, much is still unknown about the molecular targets and therapeutic potential of miRNAs in the GIT. This review discusses miRNA biology, pathophysiology of fibrosis, and aging- associated SM dysfunction in relation to the deregulation of miRNAs in the GIT. We also highlight the role of selected miRNAs associated with fibrosis and SM dysfunction-related diseases of the GIT.

revolution in computational biology led to the identification of ∼2,800 novel human miRNAs, of which more than 1,000 have been validated (data from miRNABase 21; http://www.mirbase.org). miRNAs are a group of endogenous, short (∼22 nucleotides), noncoding RNAs (ncRNAs) with important roles in cellular biology via targeting mRNAs and thereby causing translational repression (2, 6, 44). The first discovered miRNA, lin-4, influenced the development of the nematode Caenorhabditis elegans by repressing the protein expression of its target gene lin-14 (66). Subsequently, miRNAs were found to have a ubiquitous presence among living organisms including human beings; however, their expression profile and biological function could be species or tissue specific (72). miRNAs regulate the expression of over 60% of all mammalian protein coding genes responsible for diverse physiological processes, such as cell growth and differentiation, fibrogenesis, smooth muscle (SM) phenotype switch, and response to stress and aging (7, 39). In addition, miRNA deregulation is being increasingly implicated in several pathophysiological outcomes underlying diseases of various organ systems, such as gastrointestinal tract (GI) dysmotility (75, 91).

In health, GI motility arises from coordinated contractions of smooth muscle cells (SMCs). Although GI SMCs may have spontaneous electrical activity, their contractile function is also regulated by inputs from enteric nervous system including enteric motor neurons and interstitial cells of Cajal (114). Fibrosis, the final outcome in many chronic inflammatory diseases, can result in permanent scarring, organ damage, and even death. In the GIT, fibrosis and/or SM dysfunction contributes to dysmotility, which can lead to intestinal obstruction or rectoanal incontinence (RI). Historically, fibrosis has been viewed as an irreversible process, but recent evidence indicates that it may be reversible (47). Similarly, SM dysfunction is also potentially reversible. The aim of this review is to discuss miRNA biology and its regulation of the pathophysiology of fibrosis and SM dysfunction in the GIT. Regulation of GI SMCs by enteric nervous system is not the focus of our review and is not discussed here. In addition, we highlight the role of selected miRNAs implicated in fibrosis and SM dysfunction underlying diseases of GIT, such as scleroderma and Crohn's disease.

miRNA Biogenesis and Mechanism of Action

ncRNAs can have a regulatory role over the expression of coding genes. miRNA can be coded by its own gene or produced from introns of other genes. The majority of the characterized miRNA genes are intergenic or oriented antisense to their neighboring genes in the chromosome and are therefore transcribed as independent units. However, in certain situations a miRNA gene can be transcribed along with its host gene; this provides a coregulation of miRNA and the protein-coding mRNA gene. Canonical pathway of miRNA biogenesis (Fig. 1) begins in the nucleus where RNA polymerase II transcribes long molecules called primary miRNAs (pri-miRNAs), which are subsequently processed by the RNase III endonuclease, Drosha, and its cofactor DiGeorge syndrome critical region gene 8 (DGCR8)/Pasha, to yield a shorter (∼70 nucleotides) stem-looped structure called precursor miRNA (pre-miRNA) (48, 68). Subsequently, pre-miRNA is actively transported by exportin 5 to the cytoplasm (137), where a second RNase III enzyme, Dicer (57), and its cofactor transactivating response RNA-binding protein (TRBP) cleave it to form a double-stranded miRNA duplex. Later, the strands are unwound and one of the strands is degraded by Argonaute (Ago) 2 (104), whereas the remaining mature miRNA strand is loaded onto the miRNA-induced silencing complex (RISC). Ago2 and glycine-tryptophan protein (GW182) form the core elements of miRNA-RISC, which are vital for miRNA-mRNA binding. A 6- to 8-nucleotide-long sequence (seed sequence) at the 5′ end of a miRNA binds to complementary sequence in the 3′ untranslated regions (UTR) of target mRNA (102). Finally, translational inhibition or, in some cases, mRNA degradation (5, 87) leads to downregulation of the target protein.

Fig. 1.

Biogenesis of miRNA 29a and its regulation of Col3A1 mRNA inside an activated fibroblast in scleroderma. Schematics illustrate the transcription of pri-miR-29a being regulated through TGF-β, PDGF, CTGF, ET-1, IL-4, and LPS signaling via SMAD and non-SMAD pathways, and by epigenetic silencing. Transcription of Col3A1 mRNA is regulated by the overactive SMAD and non-SMAD signaling pathways. Extracellular exchange of miR 29a via exosomes (small vesicles produced by budding of plasma membrane) is consequently reduced, resulting in low circulating miR 29a levels in the very early stage of scleroderma, which could potentially serve as a disease biomarker (60). In human beings, imperfect complementarity of miR-mRNA binding is more common than perfect complementarity; therefore, miR-induced translational repression (thick black line) is more common that target mRNA degradation (thin black line). *Mechanical stress also activates fibroblasts to synthesize excess collagen (34). Red lines and arrows indicate inhibition or decrease; green lines and arrows indicate stimulation or increase. AGO2, Argonaute2; CTGF, connective tissue growth factor; DGCR8, DiGeorge syndrome critical region gene 8; ET-1, endothelin-1; IGF-1, insulin like growth factor-1; IL-4, interleukin-4; LPS, lipopolysaccharide; ORF, open reading frame; P body, processing body; PDGF-β, platelet-derived growth factor-β; Pre-miR, precursor miR; Pri-miR, primary miR; RAN-GTP, RAS-related nuclear protein-guanosine triphosphate; RISC, RNA-induced silencing complex; TRBP, transactivating response RNA-binding protein; TGF-β, transforming growth factor-β; UTR, untranslated region.

Current evidence points to translational repression occurring mostly at the initiation step of protein synthesis, whereby miRNA-RISC prevents the association of ribosomes with mRNA (56). By contrast, in postinitiation translational repression, ribosomes associate with the mRNAs but are unable to complete translation because of either early dissociation or physical obstruction by associated miRNA. Degradation of mRNA involves deadenylation and decapping by the 5′-3′ exonuclease XRN1 pathway (12, 31).

Although miRNAs bind most frequently to the 3′UTR region of mRNA, other reported sites of binding include the open reading frame (ORF) or the 5′UTR of the target mRNA (81). However, it is seen that binding of miRNA to the 3′UTR of mRNA leads to a greater degree of translational repression compared with binding at the ORF or 5′UTR of target mRNA (4). Interestingly, on rare occasions, miRNAs can also activate mRNA translation (127). Furthermore, a mature miRNA can reenter the nucleus to interact directly with transcription factors or the DNA itself, thereby modulating gene expression (9). Besides the canonical pathway, there are alternative miRNA biogenesis pathways that do not require the ribonuclease Drosha to generate pre-miRNA, such as the splicing-derived miRNAtrons (17, 94).

A single miRNA can bind to multiple mRNAs. Similarly, multiple miRNAs can target a single mRNA (4, 71). However, the role of a miRNA is decided not only by its target specificity but also by its tissue-expression levels. Specific cell types or tissues display a unique miRNA expression profile. Moreover, since one miRNA could regulate a set of mRNAs involved in a specific signaling pathway, the final outcome depends on the complex molecular interactions between the set of miRNAs and their target mRNAs.

Fibrosis Pathophysiology: Cellular and Epigenetic Mechanisms

Fibrogenesis during the physiology of wound healing and tissue repair is well coordinated and tightly regulated. In contrast, pathological fibrosis occurs when extracellular matrix (ECM) synthesis exceeds degradation due to deregulation of the wound healing or tissue repair process (59). Mesenchymal cells, such as fibroblasts and myofibroblasts, are the chief cells responsible for ECM production during fibrosis. As will be discussed later, under certain conditions even nonmesenchymal cells, such as epithelial cells or endothelial cells, can undergo transdifferentiation into ECM-producing fibroblasts or myofibroblasts via epithelial-mesenchymal transition (EMT) (42) or endothelial-mesenchymal transition (EndMT), respectively (138). Fibrosis develops through a set of complex interactions between mesenchymal cells, cytokines and growth factors, and the surrounding ECM. In addition, epigenetic (heritable changes in gene expression or cell phenotype without altering the DNA sequence) modifications within fibroblasts/myofibroblasts are also implicated in the pathobiology of fibrosis (135). As fibrosis progresses, there is increased ECM accumulation and atrophy of neighboring tissues, resulting in loss of normal function and finally organ failure.

Herein, we review the roles of cellular mediators and cellular plasticity and epigenetic regulation in fibrosis of GIT. Specific details of liver fibrosis and molecular mechanisms in fibrosis, such as oxidant stressors, growth factors and cytokines, and immune response, are discussed elsewhere (41, 67, 132).

Cellular players in fibrosis.

Myofibroblast, a phenotypic intermediate cell type between a fibroblast and a SMC, is considered the key cellular player in most fibrotic disorders (14, 40, 52). Myofibroblast arises from transdifferentiation of fibroblasts, fibrocytes, epithelial cells, and endothelial cells. In the gastrointestinal tract, a special group of myofibroblasts called intestinal subepithelial myofibroblasts (SEMFs) are located near the basement membrane of the small and large intestines. Myofibroblasts express their characteristic marker protein α-smooth muscle actin (α-SMA) and synthesize ECM components (mainly collagen) and tissue inhibitors of metalloproteases (14). Myofibroblasts are also a major source of secreted TGF-β during the fibrotic response. In early wound healing physiology, myofibroblasts help in contraction of granulation tissue. Later, myofibroblasts disappear by the process of apoptosis leading to wound resolution. However, prolonged presence of myofibroblasts in pathological conditions contributes to excessive deposition and contraction of ECM as seen in liver cirrhosis (8).

Fibroblasts, once activated, overproduce ECM components: the pathogenetic hallmark of scleroderma. Activated fibroblasts also secrete growth factors, cytokines, and chemokines and express receptors for growth factors, resulting in feed-forward amplification and sustenance of the fibrotic process. In addition, fibroblasts can convert into myofibroblasts (103).

Pericytes are mesenchymal cells located adjacent to endothelial cells of small blood vessels. In fibrotic disorders, activated pericytes undergo hyperplasia and express growth factor receptors (51, 88) in addition to transdifferentiating into fibroblasts and myofibroblasts.

Fibrocytes constitute circulating bone marrow-derived cells with fibroblast-like properties. Roderfeld et al. (112) hypothesized that bone marrow-derived circulating fibrocytes represent key mediators of liver fibrogenesis in mice.

SMCs are also implicated in the pathophysiology of GIT fibrosis in Crohn's disease and ulcerative colitis by producing collagen and matrix metalloproteinases (MMPs) in response to several inflammatory mediators, such as TGF-β and IL-1β. In Crohn's disease, SMC contraction, hyperplasia, and hypertrophy (64, 80) along with increased transmural ECM deposition lead to thickening of bowel wall, stricture formation, and, ultimately, intestinal obstruction. In scleroderma, vascular SMCs can contribute toward fibrosis by transdifferentiating into myofibroblasts.

Hepatic stellate cells (HSCs) are the main effector cells responsible for deposition of ECM during liver fibrosis (67). Stellate cells are also found in the GIT, where they contribute to fibrosis in Crohn's disease and ulcerative colitis (43, 109).

Although mesenchymal stem cells (MSCs) are believed to contribute to fibrosis, their exact role is still controversial (37, 74). A detailed discussion on the role of MSCs and fibrosis can be found elsewhere (37, 74, 77, 126).

Phenotypic plasticity of cells: EMT and EndMT.

Switching of cells from an epithelial to mesenchymal phenotype is important during embryogenesis (65). A similar process, EndMT, is also known to contribute toward fibrosis (111). The role of EMT/EndMT in intestinal fibrosis has been shown in animal models and Crohn's disease in human beings (36, 111).

Epigenetics and fibrosis.

Studies have shown an upregulation of genes in fibrotic myofibroblasts, consistent with activation of signaling pathways involved in fibrosis (63). In addition, collagen genes in scleroderma fibroblasts become autonomous and upregulated (30, 53). Mechanisms of epigenetic regulation in fibrotic disorders include DNA methylation, histone modification, and regulation of miRNAs.

Role of miRNAs in Fibrosis of Gastrointestinal Tract

microRNA databases, validation of databases, and challenges.

MicroRNAs impact gene expression primarily by targeting 3′UTR of the target messenger RNAs. Based on the experimental evidences, thousands of miRNA-target interactions (MTIs) are publically available at http://www.targetscan.org; http://zmf.umm.uni-heidelberg.de/apps/zmf/mirwalk2/custom.html; http://www.microrna.org/microrna/home.do; and http://c1.accurascience.com/miRecords/.

To date, miRTarBase provides ∼3,000 MTIs, TarBase provides detailed information on each miRNA-gene interaction, and miRecords includes excellent quality database from experimentally proven miRNA targets obtained from manual literature. In addition, miRWalk database provides human, mouse, and rat validated binding sites on their target genes with different applications for target prediction. These databases are readily available and time saving (11, 24, 33, 58, 70, 73, 119, 136).

Although these databases are being constantly updated by the institutes/organizations, the perspective investigators should be vigilant about the miRs of their interest for the specific mRNA target and use further caution for nonspecific off-targets. It is further challenging considering the milieu of miR targets for mRNAs in different model systems (rodent, human, explant, the organ used, or specific phenotype switching). For example, miR-143 and miR-145 may be highly expressed with an important role in the airway SM contractile dysfunction, which may not be the case in the urinary bladder SM (28). Therefore, there is a dire need for the buildup and regular update of these databases in a model/cell/tissues/organ-specific validation manner.

miRNAs as potential biomarkers.

Being key regulators of cellular survival, proliferation, and SM differentiation (21) and of gain or loss of function under physiological and pathophysiological conditions, miRNAs not surprisingly may also serve as novel biomarkers for a number of diseases involving heart, cancer, kidney, autism, Parkinson's disease, depression, peritoneal fibrosis, autoimmunity, and arthritis (3, 25, 32, 35, 50, 54, 96, 116, 140). Recently, this list has grown to include the role of miRNAs as potential biomarkers in the pathophysiology of different SMs. For example, miR-204 is a major regulator of vascular SMCs (VSMCs) calcification, miR-221, and miRs143/145 are modulators of VSMCs phenotype and contractility (27, 29, 125). Similarly, a large number of miRNAs have been suggested to play important role as biomarkers and therapeutic targets in pulmonary arterial hypertension (16, 115).

Involvement of microRNAs in the remodeling of GI SM in the rodent model suggests a significant impact of microRNAs on phenotypic and functional aspects in the GIT (97, 99). Currently, because of the limited data in the GI SM fibrosis and dysfunction, in this review we have limited our discussion to the following miRNAs.

miRNA-29 family.

The human miRNA-29 family, or simply miRNA-29, includes miRNA-29a, miRNA-29b1, miRNA-29b2, and miRNA-29c, which are encoded by two separate gene clusters. However, all members have consensus seed-sequence that allows them to target the same mRNA. To date, miRNA-29 is the most well-studied regulator of ECM production among all miRNAs. Members of the miRNA-29 family target several genes involved in ECM production (such as collagen, elastin, and fibrillin) and are implicated in multiple fibrosing disorders of the human body, such as scleroderma and liver fibrosis (Table 1). The role of miRNA-29 has been more extensively researched in the hepatology literature compared with the GIT. However, there is mounting evidence to suggest that the pathophysiological role of miRNA-29 in fibrosis of liver and GIT is similar.

Table 1.

Selected miRNAs associated with fibrotic diseases involving the GIT

Disease Associations	Deregulated Human miRNAs; Chromosome Location	Deregulation in Disease	Validated miR Targets	Affected Pathways and Final Effect in Disease Due to miR Deregulation	Sites Affected and Complications/Clinical Consequences of Fibrosis in Disease	References
Scleroderma (GIT involvement)	29a and 29b1; chromosome 7	Downregulated	Col 1A1, Col 1A2	PDGF-β and TGF-β, resulting in increased production of type I and type III collagen	Esophagus: Most common site of GIT involvement in SSc. Leads to GERD, stenosis, strictures, and Barrett's esophagus (due to chronic GERD)	53, 60, 101, 120, 124, 139
					Stomach: Watermelon stomach (gastric antral vascular ectasia) and delayed gastric emptying
	29b2 and 29c; chromosome 1		Col 3A1, ELN, FBN1		Small intestine: pseudo-obstruction, malabsorption syndrome (due to bacterial overgrowth), and PCI, constipation, and diarrhea
					Colon: Pseudo-obstruction, intestinal perforation, constipation, and diarrhea
	92a; chromosome 13	Upregulated	Integrin αv	Integrin overexpression leading to activation of TGF-β. This results in increased ECM synthesis.	Rectum and anal canal: 2nd most common site of GIT involvement. Constipation (due to impaired relaxation of fibrotic internal anal sphincter). Fecal incontinence and rectal prolapse
	150; chromosome 19	Downregulated	Integrinβ3	Integrin overexpression leading to activation of TGF-β. This results in increased ECM synthesis.
	196a; chromosomes 12 and 17	Downregulated	Col1A1, Col1A2	TGF-β pathway leading to increased collagen production
	21; chromosome 17	Upregulated	Smad7	TGF-β pathway leading to increased collagen production
Crohn's disease (GIT fibrosis)	200b; chromosome 1	Downregulated	ZEB1 and ZEB2	EMT resulting in increased ECM production	Esophagogastroduodenum: fistulae>strictures Jejunum:  strictures > fistulae Ileum:  fistulae>strictures Colon:  fistulae>>>strictures Anoperineum:  fistulae>>>strictures	20, 69, 92
	19a-3p; chromosome 13	Downregulated	CTGF and TSP-1
	19b-3p; chromosome13			Increased collagen production
	29b; chromosomes 1 and 7	Downregulated	Col 1A1,	PDGF-β and TGF-β, resulting in increased production of type I and type III collagen
			Col 1A2, Col 3A1

ECM, extracellular matrix; ELN, elastin; FBN1, fibrillin; GIT, gastrointestinal tract; GERD, gastroesophageal reflux disease; miR, miRNA; PCI, pneumatosis cystoides intestinalis; SSc, scleroderma; ZSB, zinc finger E-box-binding homeobox.

Studies have shown that miRNA-29 is a direct posttranscriptional repressor of collagen gene expression in scleroderma, thereby acting as an antifibrotic miRNA (Fig. 1) (15, 49, 83, 133). Similarly, in the liver, miRNA-29b is antifibrotic by suppressing the genes involved in the activation of HSCs (118). Downregulation of the miRNA-29 family has been reported in patients with advanced hepatic fibrosis. In addition, miRNA-29b overexpression in HSCs resulted in decreased collagen synthesis (112). Studies have shown that TGF-β stimulation produces a reduction in miRNA-29 expression and derepression of collagen synthesis whereas stimulation with hepatocyte growth factor is associated with increased miRNA-29 levels and consequently decreased collagen synthesis (62). Recent studies implicate Toll-like receptors (TLRs) signaling in the pathogenesis of fibrosis in scleroderma by suppressing miRNA-29 expression (13). Although TLR4 polymorphisms are associated with Crohn's disease, their role in fibrosis via interaction with miRNA-29 is unknown (38). In a study by Nijhuis et al. (92), there was significant downregulation of the miR-29 family in the mucosa overlying strictures in Crohn's disease.

miRNA-200 family.

Similar to the miRNA-29 family, members of the human miRNA-200 family are transcribed from two distinct gene clusters. They include miRNA-200a, miRNA-200b, miRNA-200c, miRNA-141, and miRNA-429. On the basis of their seed sequence, the miRNA-200 family is categorized into two functional groups. The first group comprises miRNA-200b, miRNA-200c, and miRNA-429, and the second group includes miRNA-200a and miRNA-141. Because the functional groups have dissimilar seed sequence, they target different mRNAs. The miRNA-200 family participates in a signaling nexus involving the E-cadherin transcriptional repressors zinc finger E-box-binding homeobox (ZEB)1 and ZEB2, and TGF-β, which allows switching of cells between epithelial and mesenchymal phenotypes (45). Expression of the miRNA-200 family inhibits EMT via repressing ZEB1/ZEB2 and TGF-β. Conversely, downregulation of the miRNA-200 family promotes fibrosis by inducing EMT. Studies have established the role of the miRNA-200 family in EMT during fibrosis of several organs, such as lungs and kidneys (128, 135). Recently, deregulation of the miRNA-200 family has been implicated in the pathophysiology of EMT leading to GIT fibrosis (Table 1) (22).

Phenotypic Plasticity of GI Smooth Muscle

In health, GI motility is brought about by a series of coordinated contractions and relaxations of muscles of the gut that enable food to pass from mouth to anus. Except for a part of the esophagus and the external anal sphincter, the rest of the gut musculature is made up of SM. Moreover, not all the SMs of the gut are alike; there are significant differences in their phenotypic makeup. The SMs that form the sphincters (tonic SMs) have a unique contractile mechanism resulting in a higher basal tone compared with phasic SMs. It is well established that the major molecular determinant of basal tone in sphincteric GI SMs, such as internal anal sphincter (IAS), is a RhoA-associated kinase (RhoA)/Rho kinase (ROCK) pathway within the specialized SMCs (105–108).

SMs, unlike cardiac or skeletal muscles, are not terminally differentiated; therefore, they are capable of phenotypic modulation (plasticity). Normally, the GI SMCs exhibit a contractile phenotype characterized by expression of contractile proteins as well as ion channels and signaling molecules involved in contractile function (95). SMC phenotype switch is believed to contribute to the development of fibrosis in Crohn's disease (110). Recently, the role of gut microenvironment in SMC phenotype has received a great deal of attention. It is postulated that gut microbiota can induce phenotypic changes in GI SMCs, leading to GI dysmotility (117). TLRs mediate the interaction between the GIT and microbiota. Moreover, TLRs can regulate miRNAs or mRNAs in humans (82, 123). Changes to the phenotype can affect the “smooth” functioning of SMCs, and the resulting alteration in SM tone may synergize with fibrosis-induced changes to GIT, leading to discordant contractions that are the hallmark of GI dysmotility. Table 2 lists the diseases associated with altered tone due to GI SM phenotype switch.

Table 2.

Implications of altered tone due to smooth muscle phenotype switch in disorders of GIT

Site of GIT Affected	Disorder
Esophageal body	• Diffuse esophageal spasm
	• Achalasia
	• Hypotensive peristalsis
	• Hypertensive peristalsis (nutcracker esophagus)
LES	• Achalasia
	• Hypertensive LES, hypercontracting LES
	• Gastroesophageal reflux due to transient LES relaxation
	• Hypotensive LES, decreased reflex contraction of LES
Stomach	• Gastric visceral hypersensitivity
	• Gastric dysrhythmias
	• Gastroparesis
	• Pylorospasm
Gall bladder	• Biliary dyskinesia
Small intestine	• Adynamic ileus
	• CIPO
Colon and rectosigmoid region	• Colonic inertia, Hirschsprung's disease
	• CIPO or congenital human MMIHS
Anal canal and internal anal sphincter	• Recurrent anal fissures • Hemorrhoids • Constipation • Rectoanal incontinence

CIPO, Chronic intestinal pseudo-obstruction; LES, lower esophageal sphincter; MMIHS, megacystic-microcolon-intestinal hypoperistalsis syndrome.

Role of Selected miRNAs in GI Smooth Muscle Phenotype Switch

Phenotypic changes in SMCs occur because of major alteration in gene expression controlled chiefly by serum response factor (SRF). Genes determining SM phenotype are regulated by serum response elements (CArG boxes), to which the transcription factor SRF binds (86). In addition, SRF is regulated by transcriptional coactivators myocardin (MYOCD) and ELK1. RNA-binding protein for multiple spicing 2 (RBPMS2), a marker of SMC precursors, is progressively lost during SMC differentiation. Moreover, it has been found that expression of RBPMS2 in primary cultures of differentiated SMCs changes their phenotype from contractile to proliferative (93). Recently, studies have shown the importance of miRNAs in regulating SM differentiation and proliferation; some of these miRNAs are regulated by SRF via at least one CArG box in promoter regions of miRNAs. Studies have also identified genome-wide miRNAs in human and mouse intestinal SM (97).

Of all the miRNAs identified in intestinal SM, two miRNA clusters, miRNA-143/miRNA-145 and miRNA-199a/miRNA-214, induced by SRF, regulate SMC differentiation and proliferation, respectively (Fig. 2). miRNA-143 and miRNA-145 are highly expressed in differentiated SMCs but repressed in proliferating SMCs, whereas miRNA-199a and miRNA-214 are highly expressed in proliferating SMCs but repressed in differentiated SMCs. Additionally, gain-of-function or loss-of-function studies with these miRNAs demonstrated that each miRNA could switch SMC phenotype between the proliferating and differentiated state (18, 86, 97).

Fig. 2.

Role of miRNAs in the regulation of GI smooth muscle phenotype switch. Schematics illustrate miRNA-mediated GI smooth muscle switching into proliferative, differentiated, and senescent phenotypes. miRNAs 143/145 control GI SMC phenotype via targeting MYOCD whereas miRNAs 199a/214 target ELK1 and regulate GI SMC phenotype. Upregulation of miRNAs 143/145 and corresponding downregulation of miRNAs 199a/214 lead to differentiation of GI SMCs into a contractile phenotype (97). RhoA/ROCK levels correlate inversely with tissue expression of miRNAs 139 5p/133a in the GI SMCs, resulting in a phasic or tonic GI SMC phenotype (23, 121). Downregulation of miRNAs 143/145 and corresponding upregulation of miRNAs 199a/214 results in GI SMC proliferation that can progress into SMC hypertrophy culminating in GI dysmotility. Senescence of GI SMCs is mediated by miRNAs 29, 200, and 205 via targeting ZEB1, ZEB2, and SHIP1. Green boxes indicate high; red boxes indicate low. GI, gastrointestinal; MYOCD, myocardin; ROCK, rho kinase; SMC, smooth muscle cell; ZEB, zinc finger E-box-binding homeobox.

Phenotype remodeling of SMs occur in various diseases of GIT; however, there is a paucity of literature regarding the exact underlying molecular mechanisms involving SRF signaling and regulation by miRNAs. Deletion of Dicer, the endonuclease responsible for production of mature miRNAs (10), caused developmental failure of the GI SMCs and downregulation of contractile genes and transcription factors (99). Recently, it was shown that loss of SRF-dependent miRNAs resulted apoptosis of GI SMCs (98). Thus miRNAs are essential for the development and establishment of the SMC phenotype, and modulating these molecules might have therapeutic implications in the GIT.

Aging and Oxidative Stress-Related Fibrosis and SM Dysfunction

A prototype of age-related change in the GIT is loss of tone and fibroelastic properties of IAS leading to RI (46, 48). Because of the paucity of literature on age-related changes in the IAS, precise pathogenic mechanisms of aging-associated decrease in the IAS tone are unknown (55, 90, 122). However, data from other SMs, such as vascular SM, have shown an aging-associated increase in oxidative stress and decrease in nitric oxide production (113), which can lead to changes in cellular phenotype and intracellular signaling.

Our earlier studies have shown an oxidative stress-induced decrease in the IAS tone via RhoA/ROCK downregulation (122). Paradoxically, a number of studies have revealed that RhoA/ROCK may be initiators of oxidative stress during disease states based on the findings of ROCK inhibitor-induced halt in the progression of renal fibrosis (89). Recently, it was shown that miRNAs 19b, 20a, and 106a were differentially regulated in aged human cells, such as endothelial cells, renal epithelial cells, and skin fibroblasts (46). In another study, upregulation of miRNA-22 was found to induce senescence in human fibroblasts (134). More recently, Mahavadi et al. (78) have shown that increases in the oxidative stress in gastric SMCs of diabetic rats lead to upregulation of RhoA expression and increase in SMC contraction via decrease in expression of miRNA-133a; however, it is not known whether diabetic status of rats had contributed to their observed findings. Nevertheless, the role of miRNAs in RhoA/ROCK-mediated oxidative stress and/or fibrosis in the GIT awaits further investigation.

Diagnostic and Therapeutic Opportunities

miRNAs as novel diagnostic biomarkers.

In the year 2008, the discovery of miRNAs' presence extracellularly in the serum highlighted the exciting potential of their use as noninvasive disease biomarkers (19). As of today, miRNAs have been found in most body fluids, such as plasma, serum, saliva, tears, breast milk, and urine (130). miRNAs exhibit striking extracellular stability that makes them exciting candidates for use as biomarkers. Although miRNA profiles have been well studied in the cardiovascular, cancer, and neurological fields, there is limited data in the GI literature.

Compared with the GIT, the role of miRNAs in liver fibrosis has been more extensively studied. However, caution should be exercised when extrapolating the roles of miRNAs in liver fibrosis to GIT because of the possibility of tissue-specific differences in the roles of miRNAs. It has been found that serum levels of the miR-29 family were lower in patients with strictures in Crohn's disease compared with those without strictures, suggesting the potential of using miRNAs as biomarkers (92). Similarly, circulating miRNAs in scleroderma have a diagnostic and prognostic value (131) because their expression can be upregulated by using synthetic miRNA mimics. Recent studies have also reported age- and gender-specific influences on miRNA patterns in blood (61, 85). Moreover, the jury is still out on whether miRNAs should be obtained from serum or plasma (84, 129).

miRNA-based therapeutic strategies.

Therapeutic strategies involve the use of miRNA mimics or miRNA antagonists. So far, miRNA-based therapeutic trials in animal models have produced encouraging results with minimal toxicity. Although the miRNA-based therapeutic strategy is promising, certain challenges and issues need to be kept in mind, such as that miR mimics can have off-target effects and miRNA-based drug delivery to the target cells/tissues can pose difficulties because of the presence of RNAses in the blood (100). Novel methods of delivery of miRNAs include use of lenti- or adenovirus vectors, although safety concerns exist. Exosomal delivery of miRNA also appears promising (1).

In a study by Cordes et al. (26), introduction of miRNA-145 via lentiviral-mediated gene transfer into the ligated carotid arteries of mouse led to suppression of VSMC proliferation and promoted differentiation. Conversely, miRNA antagonists, also called antagomirs, can be useful for miRNA silencing. Liu et al. (76) used antagomirs to knock down miRNA-221 and -222 in cultured VSMCs. Recently, Makino et al. (79) described a novel method of miRNA let-7a overexpression in bleomycin-induced fibrosis in mice by intraperitoneal injection of miRNA let-7a. Although miRNA-based therapeutic targeting has yielded promising results in pulmonary and cardiovascular literature, much of the data on the role of miRNAs in GIT is still preliminary. More studies in GIT are warranted.

Concluding Remarks

Fibrosis and SM dysfunction correlate with the morbidity and mortality associated with several diseases of the gastrointestinal tract. However, there is an unmet need of novel markers that reflect the true extent of fibrosis-related gastrointestinal SM dysfunction. In such a scenario, miRNAs have emerged as promising biomarkers for diagnosis and assessment of disease activity and severity. In vitro and in vivo animal models of miRNA-based therapeutic success have given us a ray of hope that miRNA-based therapy for human fibrotic disorders and GI dysmotility could be a reality in the not-so-distant future.

GRANTS

The work was supported by National Institutes of Diabetes and Digestive and Kidney Diseases Grant RO1DK035385 and an institutional grant from Thomas Jefferson University.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

C.V.K., J.S., and S.R. conception and design of research; C.V.K. and J.S. prepared figures; C.V.K., J.S., and S.R. drafted manuscript; C.V.K., J.S., C.T., and S.R. edited and revised manuscript; C.V.K., J.S., C.T., and S.R. approved final version of manuscript.