MicroRNA-orchestrated pathophysiologic control in gut homeostasis and inflammation

Abstract

The intestine represents the largest and most elaborate immune system organ, in which dynamic and reciprocal interplay among numerous immune and epithelial cells, commensal microbiota, and external antigens contributes to establishing both homeostatic and pathologic conditions. The mechanisms that sustain gut homeostasis are pivotal in maintaining gut health in the harsh environment of the gut lumen. Intestinal epithelial cells are critical players in creating the mucosal platform for interplay between host immune cells and luminal stress inducers. Thus, knowledge of the epithelial interface between immune cells and the luminal environment is a prerequisite for a better understanding of gut homeostasis and pathophysiologies such as inflammation. In this review, we explore the importance of the epithelium in limiting or promoting gut inflammation (e.g., inflammatory bowel disease). We also introduce recent findings on how small RNAs such as microRNAs orchestrate pathophysiologic gene regulation.

THE INTESTINE AS A HOTSPOT FOR HOMEOSTASIS AND INFLAMMATION

The intestine is a unique organ where multiple communications between the immune system and commensal microbiota take place. It consists of diverse specialized cell types that function in distinct compartments, such as the epithelial layer, lamina propria, and gut-associated lymphoid tissue (GALT). Maintenance of intestinal homeostasis depends on elaborate interactions within and between these compartments; a breakdown of host homeostasis can lead to inflammatory disorders such as inflammatory bowel disease (IBD). IBD, defined as chronic, uncontrolled mucosal inflammation within the gastrointestinal tract, is accompanied by diarrhea and abdominal pain (1). IBD is generally divided into two clinical phenotypes: Crohn’s disease (CD), which affects the whole gastrointestinal tract (from the mouth to the anus), and ulcerative colitis (UC), which is limited mainly to the colon (1). The precise etiology of IBD is not completely understood, but it is clear that dynamic crosstalk between some genetic and environmental factors and the host immune system is critical for IBD pathogenesis (1). In this review, we introduce some key players (e.g., epithelial cells) in IBD pathophysiology and focus on the roles of microRNAs (miRNAs) in aggravating or ameliorating this inflammatory disorder.

 

GENERAL FEATURES OF INTESTINAL EPITHELIAL CELLS

The intestinal epithelium is composed of a single layer of intestinal epithelial cells (IECs) that covers the very large surface of the intestines (approximately 100 m2) and plays the foremost role as a physical barrier. IECs are important for the absorption of nutrients and they actively contribute to mucosal homeostasis by sensing antigens (e.g., commensal microbiota, food antigens, and pathogens) and eliciting immune responses. IECs include specialized cell types with distinct functions: intestinal stem cells (tissue self-renewal), transit amplifying cells (cell proliferation), Paneth cells (antimicrobial peptide secretion), enterocytes (nutrient absorption), goblet cells (mucus secretion), tuft cells (opioid release), enteroendocrine cells (endocrine signaling), and microfold (M) cells (antigen sampling) (2, 3). One of the basic functions of IECs is to maintain intestinal homeostasis. IECs contribute to the barrier function by providing a first-line defense against continuous attack by invasive pathogens and heavy colonization by opportunistic bacteria. IECs also constitute a communication interface between bacteria in the lumen and leukocytes in intraepithelial and lamina propria compartments to mount local immune responses (2, 3).

Turnover and shedding are important features of IECs (4). Their proliferation and differentiation from intestinal stem cells are regulated by key genes related to Wnt signaling (e.g., leucine rich repeat-containing G-protein coupled receptor 5 [Lgr5] and achaete-scute complex homolog 2 [ASCL2]) and the Notch pathway (e.g., Notch1 and Notch2) (5-7). The tight junction (TJ) proteins such as zonula occludens (ZO)-1 and integrins of aged IECs are altered to facilitate their detachment from the tip of the villus (small intestine) or crypt (colon), and removal (8). Under pathological conditions, the p53-PUMA (p53-upregulated modulator of apoptosis)/Noxa and immune cell-derived TNF-α/FasL pathways are thought to be involved in promoting IEC apoptosis (9-11), whereas receptor-interacting protein kinase 1 (RIPK1) is crucial for preventing IEC apoptosis (12). Thus, numerous intracellular signaling molecules maintain or alter gut homeostasis by regulating IEC apoptosis.

 

ROLES OF IECs IN MAINTAINING GUT HOMEOSTASIS

Antimicrobial peptides are essential components of the host immune system. IECs constitutively produce antimicrobial peptides to regulate microbial colonization and penetration of intestinal epithelium (13, 14). The primary function of Paneth cells, which are long-lived cells located at the base of the crypts of Lieberkühn in the small intestine, is to secrete (in an inducible manner) various antimicrobial peptides and proteins such as lysozyme, α-defensins (e.g., human α-defensins 5 [HD5] and 6 [HD6] and cryptdins in mice), C-type lectins of the Reg III (regenerating islet-derived protein 3) family, cathelicidins, and lipocalins (15). They play important roles in the susceptibility to IBD; this susceptibility is conferred by mutations in NOD2 (nucleotide-binding oligomerization domain-containing protein 2) (16, 17). For instance, reduced expression of defensins in Paneth cells, which may impair barrier function, is one of the characteristics of IBD, especially ileal CD (16). Insufficient amounts of Wnt ligands in ileal CD patients’ monocytes, but not in lymphocytes, are associated with decreased defensin formation (17). ATG16L1 (autophagy-related 16-like 1) has been suggested to control granule exocytosis (18) and XBP1 (X-box binding protein 1) has been suggested to maintain Paneth cell function (19). Other defensins such as human neutrophil peptides (HNPs) are secreted from neutrophil granules (20, 21). Beta-defensins (e.g., hBD-1) are expressed mainly in goblet cells, enterocytes, and Paneth cells (22). LL37/hCAP-18 (human cathelicidin LL-37/human cationic antimicrobial peptide 18) expressed in mature enterocytes plays a principal role in protecting the intestinal mucosal surface (23). Various antimicrobial peptides produced by intestinal epithelium are thus considered to be important for homeostasis in the digestive tract.

 

INFLAMMATORY MEDIATORS IN IBD

Genetic, immunologic, and environmental factors are implicated in IBD. Genome-wide association studies (GWAS) have demonstrated the genetic architecture of pathways related to IBD; for example, some of the IBD-related loci that have been identified include IBD1 (NOD2) and IBD3 (HLA-region) (24-26). However, enormous heterogeneity in genetic or epigenetic etiology observed among IBD patients has indicated the importance of experimental models that recapitulate pathologic and symptomatic features of human IBD. In this section, we introduce a few key players that mediate pathophysiologic exacerbation of IBD.

TNF-α

TNF-α is produced mainly by leukocytes (e.g., lymphocytes, activated monocytes, and macrophages) (27) and is a crucial pro-inflammatory cytokine. Binding of TNF-α to its receptors (TNFRs) elevates the levels of pro-inflammatory molecules such as adhesion molecules (e.g., intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and mucosal addressin cell adhesion molecule-1), cytokines (e.g., IL-1β and IL-6), and chemokines (e.g., CCL19, CCL21, CXCL12, and CXCL13) (28). TNF-α also reduces the barrier function of IECs in IBD by inducing TNFR1-mediated IEC apoptosis (29) and increasing the permeability of TJ through TNFR2 signaling (30). TNF-α modulates immunological self-tolerance in IBD, because it inhibits the suppressive function of regulatory T cells (Treg) by activating the NF-κB pathway (31). Consequently, many anti-TNF-α therapies have been suggested for the treatment of IBD, including infliximab (IFX), adalimumab (ADA), golimumab (GLM), and certolizumab pegol (CZP) (32, 33). However, these treatments have some adverse effects (including malignancy); therefore, novel or modified therapies with increased selectivity are very much needed (34).

IL-6

IL-6 is a central cytokine in IBD pathology. A remarkable increase in IL-6 in the sera (35) and inflamed colonic mucosa (36) of IBD patients has been reported. The ameliorating effect of anti-IL-6 receptor mAb administration has been reported in a model of CD45RBhigh CD4＋ T cell-induced colitis in SCID mice (37). Blockade of IL-6-mediated activation reduced the expression of the inflammatory cytokines, IFN-γ, TNF-α, and IL-1β, suggesting its therapeutic potential for the treatment of CD (37).

STAT3

The transcription factor, signal transducer and activator of transcription 3 (STAT3), acts downstream of IL-6 (38, 39). IL-6-induced T-cell proliferation is significantly reduced by STAT3 deficiency; in the presence of IL-6, STAT3-deficient T cells undergo apoptosis (40). Hyper-activation of STAT3 occasionally results in epithelial cell apoptosis (41). Among several mem bers of the STAT family (e.g., STAT1, STAT3, STAT5, and STAT6), phosphorylation of STAT3 is prominent in the colon tissue of IBD patients and in mice with dextran sulfate sodium (DSS)-induced colitis (42). STAT3 phosphorylation was also observed in chronic models of T cell-dependent colitis (42). Accordingly, STAT3 is referred to as a ‘double-edged sword’ in the development of inflammation.

 

CHARACTERISTICS OF SMALL RNA BIOGENESIS AND FUNCTION

In this section, we focus on recent findings on the role of small RNAs, especially miRNAs, in intestinal inflammation. Small RNAs of 20-30 nucleotides (nt) are grouped into three major categories: endogenous small interfering RNA (endo-siRNA), Piwi-interacting RNA (piRNA), and miRNA (43). They are present in eukaryotes from yeast to humans and they regulate homeostasis by targeting chromatin and transcripts (44, 45). endo-siRNA are double-strand RNA-derived small RNAs that share some features with miRNAs, including the use of Dicer and Argonaute proteins in their processing (44). piRNAs are RNase III-independent small RNAs that function in transposon silencing; they are processed by Zucchini and PIWI proteins (45). miRNAs are predominant small RNAs in most tissues of multicellular organisms (46).

miRNAs are small, endogenous, non-coding, single-stranded RNAs that modulate gene expression during cell development, differentiation, and apoptosis (47). In comparison with other types of small RNAs, their biogenesis pathways and functional roles have been extensively investigated (43). Mature miRNAs post-transcriptionally regulate protein-coding genes by base pairing between their 5'-end seed regions (nucleotide positions 2-7) and the 3'-untranslated regions (UTR) of target mRNAs, which leads to translational repression or mRNA degradation (48). The canonical multi-step miRNA biogenesis is performed mainly by two processing enzymes, the nuclear RNase III Drosha (49) and the cytoplasmic RNase III Dicer (50). The involvement of miRNAs in IBD—not only their functional targets and mechanisms, but also their biomarker discovery and therapeutic use—has been extensively investigated.

 

miRNAs AS TRIGGERS OF IBD PROGRESSION

Diverse factors affect the development of IBD (see above), but its exact etiology and pathophysiology remain elusive. However, a few microRNAs have shown consistent results in multiple studies and may be useful as IBD biomarkers. Here, we introduce the well-known miRNAs (miR-21, miR-155, and miR-223) and novel miRNAs (miR-1224-5p, miR-3473a, and miR-5128) that were predicted to have colitogenic ability in the gut epithelium (Table 1).

Table 1.aUC, ulcerative colitis. bLarge-IECs, epithelial cells of the large intestine.

miR-21

miR-21 is one of the most studied miRNAs that are implicated in many inflammatory diseases, including IBD. miR-21 is significantly elevated in the blood of patients with UC, together with other signature miRNAs (e.g., miR-16, miR-28-5p, miR-151-5p, miR-155, and miR-199a-5p) (48). However, miR-21 is not increased in the blood of CD patients (51). Given that UC and CD are phenotypically distinct, this circulating miRNA turned out to be useful in distinguishing these two major phenotypes of IBD at diagnosis. The importance of miR-21 in UC has been emphasized in other reports (52-54). Microarray analysis and quantitative PCR showed miR-21 upregulation in the inflamed colon tissue of patients with active UC (52). Another study focused on the role of miR-21 in epithelial barrier function and reported that miR-21 accumulated in the intestinal epithelium of UC patients (53). This study also showed that overexpression of miR-21 reduced trans-epithelial electrical resistance and increased inulin permeability by disrupting intestinal epithelial TJ. Finally, RhoB (Ras homolog gene family, member B) was suggested as a putative target of miR-21 (53). The unique role of miR-21 was recently described in a murine colitis model: miR-21 knockout protected mice from acute experimental colitis (54). miR-21 was found to regulate the expression of RhoB and Cdc42 (cell division cycle 42), which are associated with intestinal epithelial function (54). Taken together, the available data indicate that miR-21 plays a pivotal role in regulation of gene expression in IBD.

miR-155

miR-155 is important in hematopoietic-derived immune cells (55-57). Both the frequency and the absolute number of Treg cells are diminished in the lymphoid organs (such as spleen, thymus, and lymph nodes) of miR-155-/- mice, although the intrinsic suppressive function of Treg cells is not impaired (55). As Foxp3 is a target of miR-155 (56), the miR-155/Foxp3 axis may be important in transcriptional programming for Treg cell development (55). In line with the role of miR-155 in T cells, miR-155 deficiency ameliorates the pathogenesis of experimental colitis by downregulating the Th1/Th17 response (57). miR-155 also has a modulatory function in mesenchymal cells from colon tissues of patients with UC: miR-155 upregulation in intestinal myofibroblasts decreases the level of SOCS1 (suppressor of cytokine signaling 1); and SOCS1 is a direct target of miR-155 (58). An important observation is that both miR-21 and miR-155 are critical for controlling IBD: they promote microsatellite instability (MSI) (59), which results from impaired DNA mismatch repair (MMR), and may induce K-ras mutations in colonic mucosa and thus aggravate UC (60). These two miRNAs act as a link between IBD and colorectal cancer by targeting MMR (59). Therefore, miR-21 and miR-155 can be considered as targets for IBD treatment.

miR-223

Dendritic cells (DCs) and macrophages are important for orchestrating both host immune tolerance and protective immune responses (61). Recently, a novel function of miR-223 was identified in DCs and macrophages of the small intestine (62). miR-223 is highly expressed in myeloid cells and regulates granulocyte production and function (63, 64). Depletion of miR-223 significantly reduces the number of CX3CR1high macrophages at steady state and enhances the pro-inflammatory phenotype of macrophages, CD103＋ conventional DCs (cDCs), and monocyte-derived DCs (moDCs). miR-223 KO mice had more severe colitis than wild-type (62). miR-223 expressed in myeloid cells is thought to be an essential regulator of intestinal homeostasis.

The level of miR-223-3p is increased in inflamed epithelial cells of the small intestine (small-IECs) and large intestine (large-IECs) of mice with DSS-induced colitis (65). Intriguingly, miRNA profiling and network analysis showed that miR-223-3p regulates its target genes differentially in small- and large-IECs (Fig. 1) (65). Thus, cell-based bioinformatics proved to be a powerful tool for the identification of etiologic or regulatory factors that determine distinct pathophysiology of small and large intestines.

Fig. 1.Networks between miR-223-3p and its putative target genes distinctive to inflamed small- (pink) or large-IECs (green). miR-223-3p was up-regulated in both small- and large-IECs of the mice with DSS-induced colitis (65). Microarray and TargetScan database analyses suggest that miR-223-3p exclusively down-regulates different target genes between both inflamed intestines except for SLC4A4 (solute carrier family 4 anion exchanger) (underlines): HIST4H4 (histone cluster 4, H4), SLC2A5 (solute carrier family 2 member 5), CML2 (camello-like 2), CREB3L3 (cAMP responsive element binding protein 3-like 3), BRPF3 (bromodomain and PHD finger containing 3), ATP1B1 (ATPase, Na＋/K＋ transporting, beta 1 polypeptide), HSP90B1 (heat shock protein 90, beta, member 1), OGDH (oxoglutarate dehydrogenase), DUOXA2 (dual oxidase maturation factor 2), ACE2 (angiotensin I converting enzyme 2), XPNPEP2 (X-prolyl aminopeptidase 2), and CYP4V3 (cytochrome P450, family 4, subfamily v, polypeptide 3) were markedly reduced in inflamed, compared with normal, small-IECs; GOLGB1 (golgin subfamily b macrogolgin 1), TMIGD1 (transmembrane and immunoglobulin domain containing 1), CHMP4C (chromatin modifying protein 4C), MLL3 (myeloid/lymphoid or mixed-lineage leukemia 3), TMEM140 (transmembrane protein 140), ARFRP1 (ADP-ribosylation factor related protein 1), TNNI1 (troponin I, skeletal, slow 1), and RAP2A (RAS related protein 2a) were significantly down-regulated in inflamed, compared with normal, large-IECs.

miR-1224-5p, miR-3473a, and miR-5128

miRNAs are considered to be a reliable genetic source for understanding the pathogenesis of IBD. Most studies have focused on the markedly altered expression of miRNAs and their target genes in tissues or cell lines (66, 67). Live IECs sampled from IBD models could be useful for revealing the basis for pathophysiological differences between different regions of intestines (65). Furthermore, establishing putative networks of gene expression regulation by miRNAs in either normal or inflamed IECs helps to comprehensively understand IBD pathophysiology, especially in relation to the barrier function and communication. In a mouse colitis model, a more complex network in large-IECs than in small-IECs could demonstrate a correlation between UC severity and the complexity of miRNA networks. Among putative biomarkers in inflamed large-IECs, miRNAs (e.g., miR-1224-5p, miR-3473a, or miR-5128) and their targets ABCG2 (ATP-binding cassette transporter G2) or AQP8 (aquaporin 8) appear to be involved in IBD development and progression (65). These biomarkers remain to be functionally characterized in vivo and in vitro to further clarify their contributions to IBD pathology. Taken together, current elaborate studies will provide clues for understanding the pathophysiologic mechanisms in the gut and for the discovery of miRNA biomarkers and their specific gene regulatory networks in IBD.

 

CONCLUSIONS AND PERSPECTIVES

Considerable progress has been made in our understanding of IBD pathophysiology. It is currently accepted that one of the main causes of IBD is imbalanced immunological communication between enteric microbiota and effector immune cells across the epithelial interface. Research on miRNAs has focused on alteration of their expression at the tissue level in human IBD patients or in animal models and elucidation of the role of target gene silencing in disruption of gut homeostasis to trigger the development of IBD. miRNAs and their target genes can be considered as promising targets for IBD therapeutics. Despite the potential applicability of miRNAs or their related molecules in treating IBD, substantial limitations remain, including insufficiency of comprehensive profiles of miRNA expression or established networks specific to relevant cell types; variability of miRNA expression among patients or within tissues; and lack of adequate methods for in vivo RNAi delivery into target cells or tissues. Validation of miRNA functions in vivo and in vitro will help to acquire further valid scientific knowledge of IBD pathophysiology that is needed for diagnostics and therapy.