Inflammatory bowel disease (IBD) is a debilitating autoimmune disease characterised by chronic inflammation along the GI tract. Patients diagnosed with IBD are symptomatic for recurrent intestinal inflammation, diarrhoea, abdominal pain, rectal bleeding, weight loss and anaemia. Due to its complexity, a number of factors are attributed to IBD aetiology, including patients’ genetics and makeup of microbiota, food and pharmaceutical consumption, and even limiting antigen exposure due to excessive sanitation ^[1][2][19,20]. All these aspects further contribute to changes in miRNA expression.

IBD is caused by the overactivation of the mucosal immune system driven mainly by increased exposure to the gut microbiota as a result of compromised gut permeability. Importantly, host genetics are a major factor attributed to the manifestation of disease and linked to multiple regions of the genome ^[2][20]. For example, it is more likely for IBD patients to possess variants of the NOD2/CARD15 gene on chromosome 16 than healthy individuals, with this gene encoding a specific pattern-recognition receptor (PRR) for bacterial lipopolysaccharide that regulates macrophage activation of nuclear factor-κB (NF-κB) ^[3][21]. Additionally, researchers consistently observe that the IBD3 gene on chromosome 6, which encodes the major histocompatibility complex (MHC), has a genetic linkage to IBD ^[3][21]. Other environmental factors that increase IBD incidence include the use of non-steroidal anti-inflammatory drugs, antibiotics and smoking, which are all known to alter the gut microbiota ^[4][22].

Clinically, IBD is segregated into two main types known as Crohn’s disease (CD) and ulcerative colitis (UC). CD can impact any part of the GI tract with patchy regions of inflammation, whereas UC has inflammation typically localised to the colon or rectum ^[5][23]. Incidence of CD or UC can also lead to an increased risk of other diseases such as colorectal cancer ^[6][24]. The activation of central immune cell populations leads to the recruitment of non-specific mediators of the inflammatory response, such as the formation of metabolites such as prostaglandins and leukotrienes, along with damaging compounds such as reactive oxygen species (ROS) ^[2][20]. All these factors can compromise the gut epithelial barrier where the majority of host and gut microbiota interactions occur. In such a disease state, luminal antigens gain access to the lamina propria, triggering a response by innate and adaptive immune cells via various PRRs, causing professional (i.e., dendritic cells) and non-professional (i.e., intestinal epithelial cells, IECs) antigen presenting cells (APCs) to further activate central effector immune cells as well as other pro-inflammatory mediators. This cascade thereby perpetuates a positive feedback loop of leukocyte recruitment and increasing tissue damage in both types of IBD ^[1][19].

Research on IBD has highlighted changes in gut immunity and the overall functioning of cells that participate in the characteristic overactive immune response. Immune cells can be a part of innate immunity, which serves as the initial rapid defence upon the recognition of foreign pathogens, or adaptive immunity, which leads to slow but long-lasting defensive measures ^[7][25]. An effective inflammatory response to any pathogenic invasion is conducted mainly by innate immune cells, which include neutrophils, dendritic cells, monocytes, macrophages and natural killer cells. For an immunological response to specific pathogens, adaptive immune cells are more critical and include effector T-cells, regulatory T-cells and B-cells ^[7][25]. The function within, and communication between, the two arms of immunity contribute to chronic inflammation associated with IBD, leading to tissue damage through ROS production, fibrosis and continuous feedback loops of pro-inflammatory cytokine signalling ^[1][19].

There are notable immunological differences between CD and UC. Previously accepted notions were that CD typically has a major CD4+ lymphocyte population with a type-1 helper T-cell (Th1) phenotype, while UC has a type-2 helper T-cell (Th2) phenotype ^[8][26]. Due to these differences in lymphocyte populations, CD is driven by interferon-γ (IFN-γ) and interleukin-12 (IL-12) expression, while UC is driven by transforming growth factor (TGF)-β, IL-4, IL-5 and IL-13 expression. More recently, this paradigm has been expanded to incorporate the IL-23/Th17 axis to further distinguish the two types ^[8][26]. The interactions between immune cells and cells that constitute the gut epithelial barrier are paramount in understanding permeability within CD and UC disease states.

The archetypical method of diagnosis for IBD is endoscopy in conjunction with biopsies ^[9][27]. Despite the established belief that CD is mainly attributed to the overactivation of the inflammatory response throughout the GI tract while UC is mainly confined to the rectum and colon, there remain issues in the proper diagnosis of the two conditions. Serological biomarkers include the presence of perinuclear staining anti-neutrophil cytoplasmic antibodies in 70% of UC patients, or the presence of anti-Saccharomyces cerevisiae antibodies in 50% of CD patients ^[10][28]. Other biomarkers include C-reactive protein in serum and granulocyte proteins lactoferrin and calprotectin in faeces ^[11][29]. Recently, a panel of 51 protein biomarkers in serum was determined to be predictive of CD within up to five years before diagnosis that included associated changes to complement cascade, lysosomes, innate immune response and glycosaminoglycan metabolism ^[12][30]. Further development on IBD biomarkers is still required for differentiation between CD and UC, along with indication of their severity and active state.

Due to their high stability in bodily fluids, miRNAs have been studied as potential biomarkers of disease ^[13][31]. Research has expanded novel diagnostic tools to determine differences in the presence of miRNAs within various samples, with miRNAs being observed in serum, urine, tears, saliva and breast milk ^[14][32]. Total serum RNA from active and inactive CD and UC patients demonstrated that miR-595 and miR-1246 were significantly upregulated in IBD and could serve as non-specific biomarkers ^[15][33]. In plasma samples, only significant downregulation of miR-16 was validated for diagnosing CD ^[16][34]. Whole blood samples obtained from IBD patients showed that CD4+ T-cell expression of miR-1307-3p, miR3615 and miR-4792 predicted disease progression of IBD ^[17][35]. In testing miRNA profiles of saliva samples taken from IBD patients, there was a significantly altered expression of miR-101 in CD patients and miR-21, miR-31, miR-142-3p and miR-142-5p in UC ^[18][36]. Interestingly, freshly frozen colonic mucosa tissues from IBD patients displayed high levels of miR-31, miR-146a, miR-206 and miR-424, with miR-31 also highly expressed in formalin-fixed, paraffin-embedded tissues ^[19][37]. A recent study investigated the mucosal and serum expression of miRNAs in the colon of a canine IBD model. In canine IBD, miR-16, miR-21, miR-122 and miR-147 were elevated in serum and the colonic mucosa, while miR-146a, miR-192 and miR-223 were upregulated in the serum only compared to their controls ^[20][38]. In human blood or biopsy samples of IBD patients, among others, an elevation of miR-16, miR-21, miR-106a, miR-122, miR-151-5p, miR-155, miR-199a-5p, miR-320 and miR-362-3p was observed ^{[21][22][23][24]}[16,39,40,41]. The variety of samples in which miRNAs exist provide several options for studying biomarkers in IBD.

There remains an emphasis on biomarker studies that show distinguishable aspects between CD and UC for novel diagnostic tools. Some research has looked at developing miRNA panels for CD and UC diagnosis with great accuracy. One study identified an 11-miRNA panel for CD using serum samples ^[25][42], while another used platelet-derived miRNAs to determine a 31-miRNA panel for UC ^[26][43]. A further study used a six-miRNA panel to distinguish between CD and UC from colon biopsies ^[18][36]. Using peripheral blood, an eight-miRNA panel was found to distinguish between active UC and CD ^[27][44]. In addition, differentiation between types of IBD and intestinal colitis is also crucial in furthering diagnostic methods. Differential expression of miR-24 allowed researchers to distinguish between UC and L2 CD within rectal biopsies ^[28][45]. These advances are necessary for identifying specific treatments tailored uniquely to the patient.

Determining IBD activity is also crucial in prescribing treatments and predicting patient health impacts. One study found that miR-150, miR-196b, miR-199a-3p, miR-199b-5p, miR-223 and miR-320a displayed significant differential expression in non-inflamed UC compared to non-inflamed CD colonic tissues ^[29][46]. Another study demonstrated that miR-20b, miR-26b, miR-98, miR-99a and miR-203 were significantly upregulated in colonic mucosal pinch biopsies obtained from patients with active UC compared to quiescent UC ^[30][47]. Significant downregulation of miR-192, miR-375 and miR-422b and a significant upregulation in miR-16, miR-21, miR-23a, miR-24, miR-29a, miR-126, miR-195 and let-7f have been observed in sigmoid colon pinch biopsies ^[31][48]. Levels of miR-192 were significantly upregulated and miR-16 significantly downregulated in active UC ^[31][32][48,49]. A downregulation in miR-4284 in colonic tissue samples from active UC patients was also observed in a separate study ^[33][50]. Moreover, levels of miR-142-5p, miR-595 and miR-1246 in serum samples could differentiate active and non-active CD with high accuracy ^[15][33]. Interestingly, miR-31-5p and miR-203 were identified as inflammation-independent diagnostic markers for CD in colonic tissue samples, while miR-215 predicted a specific penetrating/fistulising CD phenotype in the ileum ^[34][51]. An overview of altered expression patterns of miRNAs in the context of IBD can be found in Table 1.

Treatments for IBD include pharmaceuticals, antibody therapies and full surgical procedures. One frequently prescribed treatment is 5-aminosalicylate, which was suggested to block the production of prostaglandins and leukotrienes, inhibit bacterial peptide-induced neutrophil chemotaxis, scavenge circulating ROS and further inhibit the activation of NF-κB ^[2][20]. Others include the use of corticosteroids, though it is important to note that continuous monitoring of prevalent side effects such as corticosteroid-induced osteoporosis, hypertension or diabetes must be undertaken ^[2][20]. Immunosuppressants such as azathioprine are also prescribed; however, this often puts patients at risk of opportunistic infections and toxic side effects such as neutropenia, pancreatitis and drug-induced hepatotoxicity ^[2][20]. Distinguishing between types of IBD is crucial as immunoregulatory treatments could sometimes be beneficial for one and not the other. This is indeed the case for cyclosporin in treating UC, or anti-tumour necrosis factor (TNF) therapy via infliximab combined with thiopurines for CD ^[2][9][20,27]. Even appendectomies have been associated with an increased risk of developing strictures in CD patients while having improved effects in UC patients ^[1][19].

MiRNAs have been investigated as new therapeutic targets and indicators of drug treatment suitability. For example, utilising miR-200c-3p mimics may reduce levels of inflammation caused by IL-8 or NF-κB response to TLR4 activation in IBD ^[35][52]. Overexpression of miR-122 was suggested to downregulate NOD2 in IECs, inhibiting apoptosis and destruction of the intestinal barrier ^[36][53]. Inhibiting miR-155 expression may be a therapeutic avenue due to its targeting of SOCS1 in UC or the inhibition of the hypoxia-inducible factor (HIF)-1α/trefoil factor (TFF)-3 axis observed in animal models ^[37][38][54,55]. Moreover, it was suggested that using miR-195 as a biomarker may help track therapeutic steroid resistance in IBD patients ^[39][56]. Other miRNAs such as let-7d, let-7e, miR-28-5p, miR-221 and miR-224 were found to be significantly increased in patients after six weeks of infliximab treatment, with let-7d and let-7e also showing upregulation in patients undergoing remission ^[40][57]. Further details on other miRNAs studied for their regulation in response to specific therapies and as treatment biomarkers are described elsewhere ^[41][58].

Research on IBD and gut immunity uses multiple in vivo models. For example, the dextran sulfate sodium (DSS) mouse model pertains more closely to UC with colitis manifesting due to destruction of the intestinal barrier, while the T-cell adoptive transfer mouse model more closely resembles CD with a focus on early immunological factors ^[13][31]. Interestingly, drastic impacts occur upon the deletion of Dicer1 within the intestinal epithelium of mice, one of the major enzymes in miRNA biogenesis, including the spontaneous development of colitis and a dramatic increase in epithelial cell apoptosis ^[42][59], highlighting the relevance of miRNA-mediated regulation of the GI barrier. Other chemically induced or knockout (KO) mouse models, along with alternative model organisms used in the study of IBD pathogenicity, are well-described elsewhere ^[43][44][60,61].

A difficulty in studying miRNA relevance in IBD and innate immunity is the species- and cell-specific features. For instance, Toll-like receptors (TLRs) are key for IECs to identify pathogen-associated molecular pathways (PAMPs). TLR4-activated NF-κB induction of microRNA-9 (miR-9) occurs primarily in human monocytes and neutrophils, while miR-210 plays a regulatory role in the NF-κB feedback pathway typically in murine macrophages ^[45][46][62,63]. Additionally, nucleotide-binding and oligomerization domain (NOD)-like receptors (NLRs) are important in the surveillance of the intracellular environment for signs of possible infection. Modulations by miR-29 in general dendritic cells, miR-122 in HT-29 cells, miR-146a in muramyl dipeptide (MDP)-activated macrophages and miR-192/miR-495/miR-512/miR-671 in HCT116 cells were shown to play critical roles in the regulation of NLRs ^{[36][47][48][49]}[53,64,65,66]. It is essential that researchers understand the limitations to their study models and explore all alternatives to study miRNA relevance in IBD.

One of the most widely studied miRNAs regarding health and disease is miR-21. Among the most abundantly expressed miRNAs in various mammalian cell types, miR-21 is considered an oncomiR within the intronic region of the protein coding gene TMEM49 ^[50][67]. The regulation of miR-21 is still not fully understood, as there are multiple layers towards maturity that can be regulated, including several transcription factors that bind to its promoter region or bind to the pri-miR-21 form ^[51][68]. Elevated levels of miR-21 are suggested to be pathological in IBD ^[52][53][69,70]. Epigenome-wide whole blood DNA methylation profiles of paediatric CD treatment-naïve patients showed that hypomethylation of the miR-21 locus correlated with increased expression in leukocytes and inflamed intestinal tissue ^[54][71]. Importantly, several studies showed that the ablation of miR-21 in mice led to protection against DSS-induced colitis ^[55][56][72,73]. For UC patients in remission, miR-21 was found to be downregulated while known target programmed cell death protein (PDCD)-4 was upregulated in CD3+ T-cells compared to active disease and healthy controls ^[57][74]. Further work is required to understand the full scope of miR-21 influence within the inflamed gut.

Other miRNA KO models demonstrated amelioration during DSS-induced colitis. Like miR-21, the deletion of miR-155 in mice protected against DSS-induced colitis ^[58][75]. Additionally, the deletion of miR-301a also protects mice against DSS-induced colitis by rescuing BTG anti-proliferation factor 1 (BTG1) expression and is associated with lowering levels of pro-inflammatory markers such as IL-1β, IL-6, IL-8 and tumour necrosis factor (TNF) ^[59][76]. Genetic studies on the consequences of combined KO models could elaborate more on the negative roles these miRNAs have in IBD.

Another hallmark factor in those afflicted by IBD is the presence of oxygen, either as increased levels of ROS through continuous activation of macrophages or the sensing of oxygen in the gut environment. Several miRNAs were shown to be involved in regulating nitric oxide synthase-2 (NOS2) in IBD tissues. Induction of the nitric oxide pathway by miR-21, miR-126, miR-146a, miR-221 and miR-223 led to senescence among adjacent epithelial cells via the upregulation of HP1γ ^[60][77]. Regarding the sensing of environmental oxygen in the gut, HIF was demonstrated to be a key regulator of barrier integrity and induced expression of miR-320a to improve barrier function in T84 cells ^[61][78]. Developing methods for measuring levels of oxygen as damaging free radicals and in its gaseous state within the inflamed gut may help researchers track the progression of the disease.

Numerous miRNAs demonstrated relevance to IBD and adaptive immunity. T-cells have significant roles in the genesis and development of IBD. The deletion of miR-21 exacerbates CD4+ T-cell-mediated models of colitis, while loss of miR-155 tends to decrease Th1/Th17, showing that these are key regulators in regulatory T-cell (Treg) homeostasis ^[58][62][75,79]. Continued work has demonstrated other miRNAs in Treg regulation such as miR-10a, miR-17-92 cluster, miR-146a and miR-212/132 ^{[63][64][65][66]}[80,81,82,83]. MiRNAs implicated in Th1 and Th2 differentiation include miR-17-92 cluster, miR-27b, miR-29, miR-128, miR-146a, miR-155 and miR-340 ^{[65][67][68][69]}[82,84,85,86]. For Th17 differentiation, miRNAs that show impact when imbalanced are miR-10a, miR-155 and miR-326, as well as miR-301a as an indirect inducer ^{[70][71][72][73]}[87,88,89,90]. Overexpression of miR-210 may negatively impact Th17 differentiation due to targeting hypoxia-induced inhibitor HIF1α ^[74][91]. Finally, B-cell maturation was shown to be regulated by miR-10a, miR-17-92 cluster and miR-181a ^[70][75][76][87,92,93]. The coordinated interplay between immunity regulation and IECs is essential in controlling barrier permeability. Further information on relevant research models used in the study of IBD and their conclusions regarding miRNAs’ impact on pathogenesis has been summarised elsewhere ^[77][94].