Noncoding RNAs as Promising Diagnostic Biomarkers and Therapeutic Targets in Intestinal Fibrosis of Crohn’s Disease: The Path From Bench to Bedside

Abstract

Fibrosis is a major pathway to organ injury and failure, accounting for more than one-third of deaths worldwide. Intestinal fibrosis causes irreversible and serious clinical complications, such as strictures and obstruction, secondary to a complex pathogenesis. Under the stimulation of profibrotic soluble factors, excessive activation of mesenchymal cells causes extracellular matrix deposition via canonical transforming growth factor-β/Smads signaling or other pathways (eg, epithelial-to-mesenchymal transition and endothelial-to-mesenchymal transition) in intestinal fibrogenesis. In recent studies, the importance of noncoding RNAs (ncRNAs) stands out in fibrotic diseases in that ncRNAs exhibit a remarkable variety of biological functions in modulating the aforementioned fibrogenic responses. In this review, we summarize the role of ncRNAs, including the emerging long ncRNAs and circular RNAs, in intestinal fibrogenesis. Notably, the translational potential of ncRNAs as diagnostic biomarkers and therapeutic targets in the management of intestinal fibrosis is discussed based on clinical trials from fibrotic diseases in other organs. The main points of this review include the following:

• Characteristics of ncRNAs and mechanisms of intestinal fibrogenesis

• Wide participation of ncRNAs (especially the emerging long ncRNAs and circular RNAs) in intestinal fibrosis, including transforming growth factor-β signaling, epithelial-to-mesenchymal transition/endothelial-to-mesenchymal transition, and extracellular matrix remodeling

• Translational potential of ncRNAs in the diagnosis and treatment of intestinal fibrosis based on clinical trials from fibrotic diseases in other organs

BACKGROUND

Fibrosis stands out as a global medical challenge accounting for more than one-third of deaths worldwide.¹ Specifically, intestinal fibrosis is a common and refractory pathology that leads to bowel strictures, perforation, fistula formation, and organ failure in many alimentary diseases, such as inflammatory bowel disease (IBD) and radiation enteritis. For example, approximately 50% of patients with Crohn's disease (CD) develop clinically relevant strictures and fistulas, and experience a nearly lifetime risk of surgery and heavy cost burden.² More than two-thirds of patients with CD have endoscopic recurrence at 1 year, and nearly half still need treatment at 4 years.³ Nevertheless, there are few effective and reliable methods of identifying the early stages of fibrosis or reversing existing intestinal strictures. Current studies mainly focus on the sophisticated network of intestinal fibrosis including inflammatory cascades, extracellular matrix (ECM), profibrotic mediators, and gut microbiota. Thus far, noncoding RNAs (ncRNAs) have been proved to participate in the fibrotic diseases of multiple organs (eg, liver diseases, myocardial fibrosis, and renal fibrosis). The ncRNAs involved in fibrotic diseases mainly consist of microRNAs (miRNAs), long noncoding RNAs (lncRNAs), and circular RNAs (circRNAs). NcRNAs modulate the function of mesenchymal cells, inflammatory cascades, ECM, and microbiota via mechanisms of endogenous RNA competition, RNA transcription regulation, protein sponges, and translation regulation.^4-6 In this review, we introduce the complicated roles of various ncRNAs, including the emerging lncRNAs and circRNAs, in intestinal fibrosis and explore their clinical value as biomarkers and therapeutic targets.

PATHOGENESIS OF INTESTINAL FIBROSIS

Similar to the fibrogenesis of other organs, intestinal fibrosis is triggered by autocrine and paracrine factors, pathogen- or damage-induced inflammation, and subsequent dysregulation of bowel mucosal healing.⁷ As a crucial factor in intestinal fibrogenesis, mesenchymal cells (eg, fibroblasts, myofibroblasts, and smooth muscle cells) are activated by multiple profibrotic soluble factors.⁸ Transforming growth factor-β (TGF-β), a major cytokine in intestinal fibrosis, is mainly secreted by macrophages in response to interleukin-4 (IL-4) and IL-13.⁹ Research has shown that TGF-β can transdifferentiate α-smooth muscle actin (α-SMA)-negative fibroblasts into α-SMA-positive myofibroblasts¹⁰ and activate the proliferation, migration, and contraction of myofibroblasts by a series of signaling pathways, including Smad2/3/4, ERK/JNK/p38/AKT, and rho/ROCK/actin/MRTF/SRF.^11-14

In addition, other profibrotic cytokines, such as connective tissue growth factor (CTGF), platelet-derived growth factor, fibroblast growth factor, insulin-like growth factor (IGF), endothelin, IL-36, and tumor necrosis factor-like cytokine 1A (TL1A) also promote myofibroblast proliferation and ECM production.^15-20 Because the balance between ECM production and degradation is disrupted, collagen-rich ECM is produced and excessively accumulates via fibrogenic responses,⁹ along with a significant upregulation of the collagens fibronectin and tenascin C,²¹ thereby ultimately leading to the pathologic thickening of all layers of the intestinal wall from the mucosa to the muscularis propria.

Meanwhile, microbiota dysbiosis is also associated with intestinal fibrosis. Gut infection by pathogens (eg, adherent-invasive Escherichia coli and Salmonella typhi), contributes to the pathogenesis of intestinal fibrosis.²² In bacterial infection, flagellin binds to Toll-like receptor 5 (TLR5) of the intestinal epithelium, induces the expression of IL-33 and its receptor, and therefore promotes IL-13 and TGF-β.²³ The flagellin-induced MyD88 activation elicits increased collagen I and fibronectin production in intestinal myofibroblasts, and MyD88 deletion in α-SMA-positive cells alleviates fibrosis in a mouse model of chronic colitis.²⁴

Epithelial-to-mesenchymal transition (EMT) is another potential fibrogenic mechanism in intestinal fibrosis in that it promotes epithelial-derived fibroblasts and ECM deposition in the fibrogenesis of many organs. Although the occurrence of EMT in intestinal fibrosis has been proved, its functional mechanism is still warranted.²⁵

ncRNAS

The MiRNAs, lncRNAs, and circRNAs are the major entities of ncRNAs. Most of them selectively bind to other nucleic acids by base pairing and regulate gene transcription, RNA processing, and translation in various pathophysiological processes such as fibrosis.²⁶ Defined as small fragments of RNA that comprise 20~25 nucleotides, miRNAs bind to the complementary sequences of targeted mRNAs and degrade them via cleavage, destabilization, or inhibition of mRNA translation, which finally represses the expression of target genes.²⁷ NcRNAs with >200 nucleotides are classified as lncRNAs, which not only control gene transcription but also modulate regulate mRNA processing, stability, and translation via posttranscriptional regulation by acting as sponges for miRNAs or sources of other small RNAs.²⁸ As another subclass of ncRNAs, circRNAs are generally produced by backsplicing and are highly stable, resulting from the formation of a covalently closed loop. CircRNAs have a similar function to lncRNAs, such as sponging miRNAs, sequestering RNA-binding proteins, and regulating mRNA transcription.²⁹

Until now, many studies have focused on the relationship between ncRNAs and CD by means of high-throughput sequencing and microarray.^30,31 Some ncRNAs have recently been developed as biomarkers of CD (eg, miR-146b-5p).³⁰ Research has reported that ncRNAs participate in the inflammatory response by modulating the relevant cytokines or chemokines, activation, and differentiation of immune cells (eg, Th1 and Th17 cells).^32,33 On the other hand, ncRNAs regulate tight junctions (eg, the claudin family) of the intestinal epithelium, mucus barrier, and immune homeostasis, therefore widely manipulating intestinal epithelial barrier function.³⁴ In addition, the significance of ncRNAs in gut microbiota and fibrogenesis is gradually unveiled in the etiology of CD. In this review, we mainly elucidate the importance of miRNAs, along with lncRNAs and circRNAs, in the process of intestinal fibrosis in CD (Fig. 1).

FIGURE 1.

Schematic diagram of ncRNAs involved in intestinal fibrosis. In intestinal fibrogenesis, excessive activation of mesenchymal cells causes ECM deposition via canonical TGF-β/Smads signaling or other pathways (eg, EMT/EndMT), and ncRNAs contribute to the aforementioned mechanisms.

DYSREGULATION OF NCRNAS IN INTESTINAL FIBROSIS

In intestinal fibrotic diseases, ncRNAs are often dysregulated. Lewis et al³⁵ compared the serum level of 372 miRNAs of patients with stricturing CD (defined as Montreal criteria, n = 6) with those of patients with nonstricturing CD (n = 11) and healthy control patients (n = 5) and detected 94 differentially expressed miRNAs, such as miR-19-3p (miR-19a-3p and miR-19b-3p), miR-29a-3p, and miR-29c-3p. In accordance with the results of the aforementioned miRNA serum array, decreased levels of miR-19-3p and the miR-29 family have been further verified in the serum or tissues of intestinal strictures in patients with CD.^35,36 Similarly, Zhou, Liang, et al³⁷ analyzed the differential expression of lncRNAs in tissue samples from patients with radiation-induced intestinal fibrosis and reported that 76 lncRNAs (54 upregulated and 22 downregulated) exhibited 10-fold or more differences in comparison with nonradiation-induced intestinal fibrosis controls, such as lncRNA WWC2-AS1, lncRNA RP1-65 J11·1, lncRNA XLOC-004117, and lncRNA RP11-63P12·7. The changes of miRNA and lncRNA expression profiles suggest their underlying roles in modulating fibrogenic responses in different types of intestinal fibrotic diseases.

ncRNAs in TGF-β Signaling Modulation

TGF-β signaling modulates a wide spectrum of biological processes, such as tumor metastasis, tissue fibrosis, immune response, and cell proliferation and differentiation.³⁸ Because TGF-β signaling not only alleviates inflammation but also drives organ fibrosis,³⁹ miRNAs modulating TGF-β signaling are found dysregulated in inflammatory diseases (eg, miR-4448) and fibrotic diseases (eg, miR-21).^38,40 For example, miR-155 increases in the inflamed duodenal mucosa and inhibits TGF-β signaling by targeting and downregulating Smad2 in inflammation.^41,42 However, it decreases in the primary duodenal fibroblasts of pediatric patients with CD under TGF-β stimulation.⁴¹ The dual function of miR-155 partially reveals the sophisticated modulating network of ncRNAs in inflammation and fibrosis by modulating TGF-β signaling.

In canonical TGF-β signaling (Smad-dependent pathways), TGF-β triggers the phosphorylation of Smad2 and Smad3 by binding to TGF-β receptor (TGFBR) 1. Smad4 binds phosphorylated Smad2/3 and enables the nuclear translocation of the Smad2/3 complex, therefore activating the transcription of fibrosis-relevant genes. Smad7 competes with the Smad2/3 complex for TGFBR1 and exerts negative regulation on TGF-β signaling.³⁸ As important transcription factors of TGF-β signaling, Smads are often targeted by ncRNAs. When treated with TGF-β, miR-21 expression is elevated in fibroblasts and epithelial cells depending on phosphorylated-Smad2/Smad3.^43-45 MiR-21 also directly targets Smad7 and increases collagen expression in TGF-β activation,^46,47 whereas the miR-21/Smad7 pathway can be further regulated by lncRNA COL1A2-AS1.⁴⁸

Other ncRNAs have also been proven to manipulate every step of TGF-β signaling. MiR-503 modulates Smad2 differently because it mediates the ubiquitination of Smad2. It is known to upregulate Smad2 by directly targeting Smad ubiquitin regulatory factor 2, an E3 ubiquitin ligase that promotes the ubiquitination and degradation of phosphorylated Smad2.⁴⁹ The miR-503-induced activation of TGF-β/Smad2 signaling further promotes downstream CTGF and collagen production.^49,50 The antifibrotic role of miR-29b is attributed to its inhibition of the phosphorylation of Smad3 and the expression of collagen I and collagen III via the Sp1/TGF-β1/Smad/CTGF pathway.^36,51 Different from miR-29b, lncRNA HOX transcript antisense RNA (HOTAIR) targets and downregulates the antifibrotic factor peroxisome proliferator-activated receptor γ (PPARγ) in fibrosis because it antagonizes Smad3 and interferes with TGF-β signaling.^52,53 Regarding Smad4, miR-34 upregulation provides positive feedback under TGF-β stimulation and reciprocally activates TGF-β signaling by upregulating Smad4.⁵⁴ Nevertheless, the function of miR-34 in fibrosis may be tissue-specific because the miR-34 downregulation triggered by lncRNA HOTAIR de-represses Notch signaling and elicits the enhanced expression of collagen I and α-SMA in dermal fibroblasts.⁵⁵ In addition, the TGF-β receptor is another important target site for ncRNAs to modulate TGF-β signaling. Downregulated miR-20a-5p leads to the de-repression of TGFBR2 and activates TGF-β signaling in fibrogenesis⁵⁶; Similarly, let-7b/c targets TGFBR1 and downregulates TGF-β signaling.⁵⁷

Potential Involvement of ncRNAs in EMT and Endothelial-to-Mesenchymal Transition

Research has shown that EMT is a common pathological process of cellular transdifferentiation in fibrosis and cancer. Through EMT, epithelial cells acquire mesenchymal features, such as fibroblast-like morphology, downregulated epithelial markers (eg, E-cadherin, tight junction, and cytoskeleton proteins), upregulated mesenchymal markers (eg, α-SMA, vimentin, and collagens), and upregulated EMT transcription factors (eg, Twist, Snail, Slug, and zinc finger E-box binding homebox 1/2).²⁵ The Wnt/β-catenin pathway is one of the most important signaling pathways that positively modulates the transcription of EMT-promoting genes.⁵⁸ Because accumulating evidence has revealed the contribution of EMT to ECM deposition, EMT is believed to play a role in intestinal fibrogenesis.⁵⁹ As a special form of EMT, endothelial-to-mesenchymal transition (EndMT) refers to the transdifferentiation of endothelium into mesenchymal cells, which exhibits a loss of endothelial markers and an upregulation of transcription factors and mesenchymal markers similar to EMT.⁶⁰

Mainly modulating EMT/EndMT transcription factors, the miR-200 family takes on an antifibrotic role in intestinal fibrosis. Members of this family (miR-141, miR-200a, miR-200b, miR-200c, and miR-429) are all downregulated in the stricture-overlying mucosa of patients with CD.^61,62 They inhibit TGF-β-induced EMT/EndMT by targeting zinc finger E-box binding homeobox 1 and 2,^63–67 whereas lncRNA activated by TGF-β (lncRNA ATB) abrogates the antifibrotic function of the miR-200 family.^68,69 MiR-200b-3p regulates microfibrial-associated glycoprotein 2 and the downstream expression of Slug, Snail, matrix metalloproteinase (MMP)-2, and MMP-9.⁷⁰ The regulatory role of miR-200b has been further verified in in vivo experiments: miR-200b-containing microvesicles alleviate 2,4,6-Trinitrobenzenesulphonic acid–induced intestinal fibrosis in rats by inhibiting EMT.⁷¹ Because it plays an inhibitory role in EndMT, miR-200a may directly decrease the expression of growth factor receptor-bound 2.⁷²

Different from the miR-200 family, lncRNA HOTAIR acts extensively in EMT and fibrosis mainly by regulating the expression of epithelial/mesenchymal markers and the Wnt/β-catenin pathway. Under the stimulation of TGF-β, HOTAIR targets antifibrotic miR-124 and upregulates Notch1 signaling, resulting in increased α-SMA, MMP-2, and MMP-9 in vitro.^73,74 In addition, HOTAIR maintains the expression of IGF2 binding protein 2, therefore promoting IGF signaling–induced EMT.⁷⁵ Furthermore, HOTAIR has an indirect function in the epigenetic regulation of fibrosis by inhibiting miR-29b because miR-29b is identified as targeting DNA methyltransferases in the methylation of EMT-relevant genes.^76,77 MiR-29b-3p targets progranulin, a Wnt/β-catenin-signaling downstream adaptor, and significantly increases E-cadherin expression but downregulates vimentin and Snail.⁷⁸ Other ncRNAs also form the sophisticated modulating network of EMT/EndMT and are listed in Table 1.

TABLE 1.

Potential Targets and Mechanisms of ncRNAs in EMT/EndMT

Changes in EMT/EndMT	ncRNA	Target	Organ or Cell	Mechanism	Reference
EMT
Down	miR-101	ZEB1	Liver	Suppresses ZEB1-induced EMT	79
Down	miR-147	ZEB1	Colon and lung	Reverses ZEB1-induced EMT and represses AKT phosphorylation	80
Down	miR-186-5p	ZEB1	Colon	Reduces ZEB1 expression	81
Down	miR-205-5p	ZEB1	Colon	Directly targets ZEB1 and raises E-cadherin and CDH1 levels	82
Down	miR-132	ZEB2	Colon	Suppresses ZEB2 expression	83
Down	miR-3653	ZEB2	Colon	Down-regulates ZEB2 level	84
Down	miR-34	SNAIL	Colon, breast, and lung	Targets SNAIL and key SNAIL regulators (eg, CTNNB1, LEF1, and Axin2)	85
Down	miR-203	TGF-β2, SNAIL	Colon	Reduces TGF-β2 and SNAIL expression	86
Down	miR-451	SNAIL	Colon	Suppresses SNAIL expression	87
Down	miR-17~92	CTNNB1	Colon	Inhibits Wnt/β-catenin signaling and EMT	88
Down	miR-371-5p	SOX2	Colon	Directly targets SOX2 and represses Wnt/β-catenin signaling	89
Down	miR-378	SDAD1	Colon	Directly targets SDAD1 and represses Wnt/β-catenin signaling	90
Down	miR-503	p85	Lung	Attenuates EMT by inhibiting PI3K/Akt/mTOR/Snail pathway	91
Up	lncRNA TUG1	TWIST1	Colon	Participates in the positive loop of TGF-β/TUG1/TWIST1 and activates EMT pathway	92,93
Up	lncRNA MALAT1	TWIST, SNAIL, SLUG	Colon	Sponges miR-126-5p and promotes EMT genes (eg, SNAIL, SLUG, and TWIST)	94,95
Up	lncRNA SNHG15	SLUG	Colon	Directly interacts with SLUG and impedes ubiquitination-induced SLUG degradation	96
Up	miR-675-5p	DDB2	Colon	Targets DDB2, an EMT repressor of VEGF, ZEB1, and SNAIL	97
Up	lncRNA ZFAS1	ZEB1	Colon	Upregulates ZEB1 and induces EMT	98
Up	lncRNA CYTOR	CTNNB1	Colon	Enables β-catenin nuclear translocation by impeding phosphorylation and is reciprocally upregulated by β-catenin activation, finally activates c	99
Up	lncRNA TCF7	Unknown	Colon	Activates Wnt/β-catenin signaling	100
Up	lncRNA CASC21	Unknown	Colon	Activates Wnt/β-catenin signaling	101
Up	hsa_circRNA_102610	SMAD4	Colon	Sponges miR-130a-3p and promotes TGF-β1-induced EMT	102
EndMT
Down	miR-148b	TGFBR2, SMAD2	Skin	Inhibits EndMT via diretly targeting TGFBR2 and Smad2	103
Down	miR-18a-5p	NOTCH2	Heart	Downregulates α-SMA, fibronectin, and vimentin via binding to Notch2	104
Down	lncRNA H19	Unknown	Retina	Suppresses TGF-β1-mediated EndMT through MAPK-ERK1/2 pathway	105
Down	let-7	Unknown	Endothelial cells	Anti-EndMT effects	106
Up	lncRNA MALAT1	TGFBR2, SMAD3	Endothelial progenitor cells and human umbilical vein endothelial cells	Regulates TGFBR2 and Smad3 via sponging miR-145 and therefore modulates TGF-β1-induced EndMT of EPCs	107,108
				Promotes nuclear translocation of CTNNB1 and activates Wnt/β-catenin signaling
Up	lncRNA TUG1	ATG5	Liver	Sponges miR-142-3p and promotes autophagy and EndMT by upregulating ATG5	109
Up	miR-199a-5p	Unknown	Human umbilical vein endothelial cells	Promotes radiation-induced EndMT and subsequent transdifferentiation of fibroblasts into myofibroblasts	110
Up	miR-21	Unknown	Endothelial cells	Activates endothelial Akt pathway and increases SNAIL and MMP-2/-9	111,112

ATG5, autophagy-related 5; CDH1, cadherin 1; CTNNB1, catenin beta 1; DDB2, damage-specific DNA binding protein 2; EPC, endothlial progenitor cell; LEF1, lymphoid enhancer binding factor 1; SDAD1, SDA1 domain containing 1, Sox2, SRY-box transcription factor 2; TWIST1, twist family bHLH transcription factor 1; VEGF, vascular endothelial growth factor; ZEB1, zinc finger E-box binding homeobox 1; ZEB2, zinc finger E-box binding homeobox 2.

ncRNA-Associated ECM Remodeling

Both EMT/EndMT and dysregulated TGF-β signaling finally lead to excessive ECM remodeling, which is crucial in fibrogenesis. The 2 entities of ECM, interstitial matrix and basement membrane, are different from each other in molecular composition and biological function.¹¹³ Consisting of proteins, glycosaminoglycans, proteoglycans, and enzymes, the heterogeneous ECM structure provides a dynamic microenvironment for collagen-producing cells (eg, fibroblasts, myofibroblasts, and smooth muscle cells).¹¹⁴ Fibronectin bridges ECM components (eg, collagens and cell surface integrins) to modulate ECM structural changes and signaling pathways.¹¹⁴ The collagen family, especially (myo-)fibroblast-produced collagen I/III, represents a major part of interstitial ECMs.¹¹⁵ Collagen-producing cells further organize the alignment of collagens under the stimulation of profibrotic cytokines or growth factors (eg, TGF-β and IL-13).¹¹⁶ In addition, ECM proteases, such as MMPs, tissue inhibitors of metalloproteinase, neutrophil elastases, and meprins, are the major mediators of ECM degradation.¹¹⁶ Because of the imbalance between ECM degradation and deposition, excessive ECM remodeling leads to intestinal fibrosis.

MiR-16 plays different parts in ECM remodeling because of different etiologies. In a mouse model resembling postsurgical intestinal inflammation and fibrosis, miR-16-1 increases at the site of anastomosis and exacerbates ileocolonic anastomotic fibrosis by de-repressing myofibroblast differentiation.^117,118 In contrast, miR-16 is downregulated by lncRNA WWC2-AS1 in radiation-induced intestinal fibrosis. As a result, reduced miR-16 gives rise to the production of fibroblast growth factor 2, α-SMA, and collagen I, and therefore promotes fibroblast proliferation and fibrosis.³⁷ Unlike miR-16, miR-210 is proven to be a profibrotic miRNA in radiation-induced intestinal fibrosis because it promotes collagen Iα1 expression in fibrotic smooth muscle cells.^119,120 In addition, ECM deposition and degradation in IBD fibrogenesis are orchestrated by ncRNAs. Bioinformatic analysis indicates that miR-192 may participate in ECM remodeling in CD.¹²¹ Experiments have further shown that miR-192 is upregulated by TGF-β signaling and promotes the accumulation of matrix collagens.¹²² Apart from manipulating ECM components, the cotranscribed miR-143/145 functions as an upstream process and promotes the transdifferentiation of smooth muscle cells into myofibroblasts in that the knockout of miR-143/145 leads to morphological abnormality and dysfunction of myofibroblasts in a mouse model of chemically induced colitis.¹²³

A few miRNAs also play a part in regulating the production of ECM proteins via similar targets. MiR-150 suppresses the expression of α-SMA, TGF-β1, and collagen fibers in ECM.¹²⁴ MiR-101 suppresses the production of ECM components (eg, α-SMA, collagen I) in fibrosis by inhibiting PI3K/AKT/mTOR signaling.¹²⁵ In addition, the negative correlation between lncRNA growth arrest–specific transcript 5 (lncRNA GAS5) and MMP-2/MMP-9 reveals the potential modulating mechanism of ncRNAs in ECM degradation.¹²⁶

PERSPECTIVES

Challenges

Over recent years, research on ncRNA-modulated intestinal fibrosis has made substantial progress, but there are still challenges. First, although most studies reveal correlations instead of causal relationships between dysregulated ncRNAs and intestinal fibrotic diseases, whether these correlations differ across segments of gut and reflect the stages of fibrosis remains a question because there is no gold standard for diagnosis in radiology, pathology, or endoscopy. Second, few studies elaborate the underlying molecular mechanisms thoroughly, such as ncRNA localization via RNAscope or BaseScope and functional verification based on in vivo and in vitro experiments. Third, many ncRNAs have significant function in modulating EMT and profibrotic factors. For example, hsa_circRNA_102610 promotes TGF-β1-induced EMT by sponging miR-130a-3p.¹⁰² However, EMT in intestinal fibrogenesis and the contribution of EndMT should be further studied in in vivo experiments and clinical studies.⁹ Fourth, although some research reveals that circRNAs act in intestinal fibrosis, more convincing evidence is still warranted. Learning from studies on miRNAs and lncRNAs in intestinal fibrogenesis, researchers and clinicians could collect and analyze gut biopsy, serum, and fecal samples from patients with stricturing CD and those with nonstricturing CD. Based on screening from samples of large cohorts, potential circRNAs could be first profiled and subsequently validated in vitro and in vivo.

Finally, as a useful tool to unveil intestinal fibrogenic responses, spontaneous, induced, and gene-targeted animal models are developed and widely utilized,¹²⁷ whereas the ncRNA-targeted model is rarely applied in studies.

Diagnosis

There is an urgent need for efficient and accurate biomarkers to diagnose and prognosticate intestinal fibrosis, especially for those that can be detected by noninvasive methods, such as blood and fecal tests. NcRNAs have been newly developed as diagnostic biomarkers for various diseases, including fibrosis diseases, in clinical trials and other studies (Tables 2 and 3). For example, miR-29c in urinary exosomes indicates early renal fibrosis in lupus nephritis (AUC = 0·946).¹²⁸ Given the promising role of ncRNAs, certain ncRNAs (eg, miR-200 family, miR-29 family, and miR-19 family) could have clinical significance in patients. However, their diagnostic values should be further validated in larger cohort studies or multicenter clinical trials. In addition, even though intestinal fibrogenesis may be in part independent of inflammatory signal cascades, fibrogenic responses are indeed triggered by certain types of chronic intestinal inflammation (eg, IBD).¹⁶ NcRNAs in intestinal inflammation may present a unique opportunity for predicting the early stage of colitis-induced fibrosis.

TABLE 2.

ncRNAs as Diagnostic Biomarkers for Diseases in Registered Clinical Trials

ncRNA	Disease	Reference/Clinical Trial
miR-10b	Gliomas	NCT01849952
miR-100	Breast cancer	NCT02950207
miR-107	Alzheimer disease	NCT01819545
miR-122*	Chronic hepatitis C	NCT00980161/NCT03687229
miR-122	Drug-induced liver injury by chemotherapy	NCT03039062
miR-126*	Postmyocardial infarction remodeling	NCT01875484
miR-126	Allergic contact dermatitis	NCT04365140
miR-138	Oral lichen planus	NCT02834520
miR-146a	Chronic periodontitis and coronary heart disease	NCT03721159
SNP rs2910164 in pre-miR-146a gene	Cancer	NCT04038996
miR-142-3p	Synaptopathy in multiple sclerosis	NCT03999788
miR-150/miR-155	Multiple sclerosis	NCT04300543
miR-155	Preeclampsia	NCT04277390
miR-155	Nonmuscle invasive bladder cancer	NCT03591367
miR-155	Oral lichen planus	NCT03871114
miR-192/miR-25*	Diabetic kidney disease	NCT04176276
miR-200b/miR-21*	Diabetic wounds	NCT02581098
miR-204	Capillarization in limb muscles of patients with chronic obstructive pulmonary disease	NCT02903043
miR-210	Preeclampsia	NCT03193554
miR-210*	Wound healing	NCT02024243
miR-221/miR-222	Hepatocellular carcinoma	NCT02928627
miR-25	Pancreatic cancer	NCT03432624
miR-29 family*	Shoulder stiffness	NCT02534558
miR-29 family	Head-and-neck squamous cell carcinoma	NCT01927354
miR-29b	Oral squamous cell carcinoma	NCT02009852
miR-30a	Childhood nephrotic syndrome	NCT03235128
miR-30 family	Schizophrenia	NCT02650102/NCT03007303
miR-31-3p/miR-31-5p	Colon cancer	NCT03362684
miR-452	Preeclampsia	NCT03258125
miR-494	Cerebral ischemia	NCT03577093
lncRNA CCAT1	Colorectal cancer	NCT04269746
lncRNA HOTAIR	Thyroid cancer	NCT03469544
lncRNA NBR2	Sepsis	NCT04427371
circRNA Uck2	Acute myocardial infarction	NCT03170830

*Fibrotic diseases.

TABLE 3.

Validated ncRNAs as Biomarkers in Clinical Studies on Fibrotic Diseases

ncRNA	Disease	Sample	Result	Reference
miR-29 family	Hepatic fibrosis	Serum	↓	129-132
miR-122	Hepatic fibrosis	Liver tissue and serum	↓	130,133-136
miR-34a-5p	Hepatic fibrosis	Serum	↑	137-139
miR-378 family	Hepatic fibrosis	Liver tissue	↑	140,141
let-7	Hepatic fibrosis	Serum	↓	142,143
miR-223	Hepatic fibrosis	Serum	↑	144,145
miR-21	Hepatic fibrosis	Liver tissue and serum	↑	132,139,146,147
lncRNA H19	Hepatic fibrosis	Liver tissue and serum	↑	148-150
lncRNA MALAT1	Hepatic fibrosis	Liver tissue and serum	↑	151-153
lncRNA HOTAIR	Hepatic fibrosis	Liver tissue	↑	76,77
lincRNA p21	Hepatic fibrosis	Liver tissue and serum	↓	154-156
lncRNA APTR	Hepatic fibrosis	Serum	↑	157,158
lncRNA ATB	Hepatic fibrosis	Liver tissue and serum	↑	68,159
miR-21	Renal fibrosis	Renal tissue, urine, and serum	↑	160-166
miR-214	Renal fibrosis	Renal tissue	↑	165,167
miR-29 family	Renal fibrosis	Urine	↓	128,164,166,168
miR-29 family	Cardiac fibrosis	Cardiac tissue and serum	↓	169-175
miR-21	Cardiac fibrosis	Cardiac tissue and serum	↑	173,176-182
miR-208	Cardiac fibrosis	Cardiac tissue and serum	↑	173,183,184
miR-133	Cardiac fibrosis	Cardiac tissue and serum	↑	185-187
miR-155	Cardiac fibrosis	Cardiac tissue and serum	↑	174,187
miR-146	Cardiac fibrosis	Cardiac tissue and serum	↑	176,188,189
miR-21	Pulmonary fibrosis	Lung tissue and serum	↑	190-195
miR-200 family	Pulmonary fibrosis	Lung tissue and serum	↓	191,196
miR-155	Pulmonary fibrosis	Lung tissue and serum	↑	197,198
miR-101	Pulmonary fibrosis	Lung tissue	↑	199,200
miR-31	Pulmonary fibrosis	Serum and bronchoalveolar lavage fluid	↓	191,201
miR-21	Skin fibrosis	Skin tissue	↑	202,203
miR-29	Skin fibrosis	Skin tissue	↓	202,204,205
miR-145	Skin fibrosis	Skin tissue	↑	205,206
lncRNA HOXA11-AS	Skin fibrosis	Skin tissue	↑	207,208
lncRNA CACNA1G-AS1	Skin fibrosis	Skin tissue	↑	207,209
miR-29	Intestinal fibrosis	Gut tissue	↓	36,210
miR-200	Intestinal fibrosis	Gut tissue and serum	↓	61-63
miR-19	Intestinal fibrosis	Serum	↓	35

We only include studies on specific ncRNAs in fibrosis diseases screened and validated by multicenter studies or no less than 2 studies.

Therapeutic Potential

Owing to the extensive participation of ncRNAs in intestinal fibrogenesis, more attention should be paid to their potential role as therapeutic targets to prevent early-stage fibrosis or reverse existing fibrosis.²¹¹ Since Miravirsen (SPC3649, a miR-122 inhibitor) was first used for hepatitis C in clinical trials, ncRNA-based therapies have become feasible and attractive^212,213 (Table 4). As for fibrotic diseases, Remlarsen (MRG-201), an anti-fibrotic miR-29 mimic,²⁰⁴ has been applied to keloids in phase 2 clinical trials (eg, ClinicalTrials.gov identifier: NCT03601052). Because miR-29 is also an inhibitor in intestinal fibrosis because it inhibits TGF-β signaling, whether Remlarsen can be used to treat intestinal fibrosis is worth a discussion. However, the design of ncRNA-based drugs needs further consideration for optimized curative effects. First, a successful delivery system (eg, nanoparticles and liposome-bearing microvesicles) of artificial miRNAs is prerequisite because of their vulnerability to degradation, especially in the alimentary tract. Second, effectiveness and efficiency should be considered in the choice of drug administration method for patients with intestinal obstruction. Because of varied tissue enrichment among different organs, choosing a gut-specific ncRNA for alimentary fibrotic diseases is important. In addition, an intestine-targeting delivery system for ncRNA-based drugs remains to be developed. Third, because of the intricate network of ncRNA, 1 ncRNA usually plays multiple roles in different organs and needs to be thought over as a whole system. For example, in spite of its antifibrotic role, miR-200 has been shown to promote the malignant transformation of tumors by inducing EMT in hepatocellular carcinoma²¹⁴ and to enhance the proliferative and invasive capacities of ovarian cancer cells.²¹⁵ Therefore, whether the application of ncRNA-based drugs may cause adverse effects remains a problem.

TABLE 4.

ncRNAs as Therapeutic Targets for Diseases in Registered Clinical Trials

Drug	Disease	Target	Phase	Reference/Clinical Trial
RG-125 (AZD4076)	Type 2 diabetes with nonalcoholic fatty liver disease	miR-103/miR-107	Phase 1/2a	NCT02826525
RG-125 (AZD4076)	Non-alcoholic steatohepatitis	miR-103/miR-107	Phase 1	NCT02612662
Miravirsen (SPC3649)*	Hepatitis C	miR-122	Phase 2	NCT01200420/ NCT01727934/NCT01872936/NCT01646489/NCT02452814/NCT02508090/NCT00979927/ NCT00688012
Cobomarsen (MRG-106)	Mycosis fungoides	miR-155	Phase 2	NCT03713320/NCT03837457
TargomiRs	Malignant pleural mesothelioma, Non-small cell lung cancer	miR-16	Phase 1	NCT02369198
Lademirsen (SAR339375, RG-012)	Alport syndrome	miR-21	Phase 2	NCT02855268/NCT03373786
Remlarsen (MRG-201)*	Keloids	miR-29	Phase 2	NCT03601052/NCT02603224
MRX34	Primary liver cancer, Lymphoma, melanoma, non–small cell lung cancer, small cell lung cancer	miR-34a	Phase 1	NCT01829971
	Multiple myeloma, renal cell carcinoma
MRX34	Melanoma	miR-34a	Phase 1/2; withdrawn	NCT02862145
MRG-110*	Heart failure	miR-92	Phase 1	NCT03603431

*Fibrosis diseases.

CONCLUSIONS

As the biology of ncRNA-modulated intestinal fibrosis is gradually unveiled, ncRNAs may present as promising biomarkers and therapeutic targets in the future.

Glossary

Abbreviations

α-SMA

α-smooth muscle actin

CD

Crohn’s disease

circRNA

circular RNA

CTGF

connective tissue growth factor

ECM

extracellular matrix

EMT

epithelial-to-mesenchymal transition

EndMT

endothelial-to-mesenchymal transition

EPC

endothlial progenitor cell

HOTAIR

HOX transcript antisense RNA

IBD

inflammatory bowel disease

IGF

insulin-like growth factor

IL

interleukin

lncRNA

long noncoding RNA

miRNA

microRNA

MMP

matrix metalloproteinase

ncRNA

noncoding RNA

TGF-β

transforming growth factor-β

TGFBR

transforming growth factor-β receptor

VEGF

vascular endothelial growth factor

Author contributions: R. Mao had the idea for the article. Y. Zhou and S.N. Lin performed the literature search. L.Y. Zhou drew the figures and tables. L.Y. Zhou and S.N. Lin drafted the article, and M.H. Chen, F. Rieder, S.H. Zhang, and R. Mao critically revised the manuscript. All authors approved the final version of the manuscript.

Supported by: This study was supported by National Natural Science Foundation of China (81970483).

Conflict of interest: All authors declared no conflict of interest.