Canine chronic inflammatory enteropathy (CIE) is a term used for gastrointestinal diseases present for 3 weeks or longer, after extraintestinal diseases and intestinal diseases of infectious origin or neoplastic conditions have been ruled out [1]. As in humans, a few years ago, the term IBD was used for canine CIE. However, nowadays there is consensus among experts that CIE is the most appropriate term ^[1][2][3][1,2,3], since, clinically, there are fundamental differences between dog and human diseases.

Human IBD includes at least two different chronic disorders characterized by different patterns of inflammation of the intestinal wall: Crohn’s disease (CD) and ulcerative colitis (UC) [4]. Instead, CIE in dogs is defined by the response to medical treatment: food-responsive enteropathy (FRE); antibiotic-responsive enteropathy (ARE); immunosuppressant-responsive enteropathy (IRE/SRE); and non-responsive enteropathy (NRE) [1]. Specifically for ARE, Cerquetella et al. recently provided evidence establishing that the empirical use of antibiotics in dogs with IBD can have detrimental effects. Thus, the researcheuthors suggested the use of antibacterials only after histopathological evaluation of gastrointestinal biopsies, when endoscopy is not possible, after other therapeutic trials have been unsuccessful, or when there is evidence of adherent-invasive bacteria [5].

Several studies have described differences in the gastrointestinal microbiota among dogs with various gastrointestinal diseases ^[6][7][12,13]. Thus, intestinal dysbiosis has been proposed as an important factor involved in CIE pathogenesis ^[8][14]. In addition, the use of a dysbiosis index (DI) to assess changes in the intestinal microbiota has been shown to be useful in evaluating microbial changes in fecal samples from dogs with CIE. ^[9][15].

Each intestinal segment has a specific microbiota; the colon and rectum contain the most diverse populations ^[8][14]. Bacteroides, Clostridium, Lactobacillus, Bifidobacterium spp., and Enterobacteriaceae are predominant genera. By 16S rRNA sequencing, the phyla Firmicutes, Bacteroidetes, and Fusobacteria have been identified as 95% of the total bacterial population, followed by Proteobacteria and Actinobacteria (1–5%) ^[10][16].

Phylogenetic studies have identified a decrease in the proportion of Clostridia and an increase in Proteobacteria in the duodenum of dogs with CIE ^[6][12]. Data concerning the genera present in the large intestine are limited. Suchodolski et al. (2012) concluded that dogs with active CIE present with a decrease in Faecalibacterium spp., which produces anti-inflammatory peptides in vitro, and Fusobacterium phyla. There are no differences in the Proteobacteria members ^[11][8].

Gastrointestinal mucus of the intestinal epithelia is the first physical barrier to reduce the exposure to aggressors ^[12][17]. In the small intestine, mucus forms a single removable layer and, in the colon, a double layer. Here, mucins are the major barrier with the transmembrane and gel-forming mucins ^[13][14][15][18,19,20], which have direct immunological effects by binding to the numerous lectin-like proteins found in immune cells ^[16][21].

Homeostatic maintenance of the barrier is central to preventing the entry of bacteria and toxins from the lumen ^{[17][18][19][20][21]}[22,23,24,25,26]. In dogs with CIE, it is possible that the breakdown of barrier integrity and the immunological tolerance against intestinal symbionts lead to deregulated inflammation and disease. In turn, this breakdown may also be amplified by CIE. Additionally, pathophysiological or environmental factors may induce loss of mucus barrier integrity ^[22][23][27,28].

Goblet cells, crucial for epithelial restitution, produce small peptides called trefoil factors (TFFs) which protect and repair the epithelial surfaces. The expression of TFFs is upregulated in human IBD ^[24][25][29,30]. In dogs with CIE, TFF1 expression is elevated in the duodenum, where TFF3 expression is down-regulated in the colon, suggesting they may contribute to the deterioration of the epithelial barrier ^[26][31].

Another epithelial barrier component is P-glycoprotein (P-gp), a membrane-bound efflux pump involved in the transport of a wide range of small molecules, whose abnormal expression is observed in dogs with lymphoplasmacytic enteritis (LPE). Some LPE patients have increased P-gp expression in the apical surface membrane of villus epithelial cells in the duodenum, jejunum, and/or ileum. In other patients, P-gp expression is decreased ^[27][32]. An upregulation in P-gp expression has been identified in lymphocytes from lamina propria after prednisolone treatment in dogs with CIE, which may be considered a predictor of response to therapy ^[28][33].

The barrier between blood and endothelial cells is tightly controlled physiologically. The barrier establishes the type and numbers of inflammatory cells that migrate to the interstitial space where dendritic cells (DCs), macrophages and mast cells in the lamina propria, and intraepithelial lymphocytes (IELs) monitor tissues, contributing to intestinal homeostasis ^[29][30][34,35].

In human IBD, when the barrier is disrupted, an uncontrolled transfer of inflammatory cells from the blood to the intestinal tissue occurs ^[31][36]. The extravasation is mediated mainly by integrins that bind their counterpart receptors on the endothelial cells. The molecules mediating normal endothelial–leukocyte interaction are the same as the molecules engaged in human IBD (α4β1 (VLA-4), α4β7, αDβ2, JAM-A, E-selectin, P-selectin, CD31, and CD99), although their expression levels are upregulated by inflammation ^[32][33][34][37,38,39]. There is not much information about specific integrins overexpressed in dogs with CIE. However, a reduced expression of the β-integrin CD11c has been described. This finding suggests that canine CIE may have an imbalance in the intestinal CD11c+ DCs. However, further studies are needed to determine whether CD11c could be a useful diagnostic biomarker for canine IBD ^[35][40].

In canine CIE, IL-8 may stimulate transmigration of neutrophils to the mucosa and luminal contents to eliminate microbes during intestinal inflammation ^[23][32][28,37]. In a study on German shepherd dogs with CIE, the mRNA expression of many cytokines such as IL-2, IL-5, IL-12p40, interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), and transforming growth factor-β1 (TGF-β1) was higher in diseased animals compared to controls ^[36][41]. However, Jergens et al. previously described through a meta-analysis that healthy dogs showed mRNA expression for most cytokines including IL-2, IL-4, IL-5, IL-10, IL-12, IFN-γ, TNF-α, and TGF-β. They determined that only IL-12 mRNA expression was increased consistently in small-intestinal CIE, whereas CIE lacked consistent patterns of expression ^[37][42]. Additionally, it remains unclear whether epithelial cell-derived cytokines such as IL-25, IL-33, and thymic stromal lymphopoietin (TSLP) contribute to the development of canine CIE ^[38][43].

Matrix metalloproteinases (MMPs) 2 and 9 are endopeptidases that play an important role in the turnover of extracellular matrix and cell migration and activate and degrade chemokines, cytokines, growth factors, and junction proteins ^[39][44]. MMP-2 is produced by stromal cells ^[40][41][45,46] and MMP-9 mainly by neutrophils, followed by eosinophils, monocytes/macrophages, lymphocytes, and epithelial cells ^{[40][42][43][44][45]}[45,47,48,49,50]. Both MMPs could be involved in the pathogenesis of canine CIE. Although their role in this pathology has not been completely elucidated, they are upregulated in dogs with CIE ^[46][51].

In the duodenal mucosa of dogs with CIE, an increase in neutrophils is associated with disease severity ^[43][48]. Serum perinuclear anti-neutrophilic cytoplasmatic autoantibodies (pANCA) ^[47][52] and blood neutrophil-to-lymphocyte ratio (NLR) ^[48][53] have been proposed as biomarkers of canine CIE severity. NLR has also been proposed as a useful marker to differentiate FRE from IRE, with clinical utility to subclassify the canine CIE. However, it is important to note that NLR may not be useful for NRE subclassification ^[49][54].

In addition, calgranulin-C, a protein secreted by activated neutrophils and monocytes/macrophages, and myeloperoxidase (MPO) activities increase in the mucosa of the duodenum and colon of dogs with CIE, and MPO also increases in the ileum and cecum. However, none have been related to the clinical outcome of patients ^[50][55].

Calprotectin, another protein released by activated mononuclear cells, has increased expression in canine intestinal mucosa ^[51][56] and is used as a diagnostic and prognostic factor in human IBD ^[52][53][57,58]. Determination of fecal calprotectin concentration is a useful screening test for human IBD diagnosis, reducing the need for colonoscopy by 66.7% ^[54][59]. Serum calprotectin concentrations may also be a useful biomarker for the detection of inflammation in dogs, but the use of certain drugs such as glucocorticoids could limit clinical usefulness ^[55][60]. The researcheuthors also showed that fecal calprotectin could be used as a possible marker for assessing the severity of gastrointestinal inflammation in dogs with CIE ^[56][61]. Additionally, a recent meta-analysis concluded that fecal calprotectin concentration is one of the most promising biomarkers of gastrointestinal functionality in dogs ^[57][62].

Macrophages participate in the host defense against infections and also remove apoptotic cells and remodel the extracellular matrix ^[58][63]. These cells are differentiated into two subtypes, termed M1 and M2. M1 macrophages initiate and maintain inflammatory processes, whereas M2 are associated with the resolution of chronic inflammation and the promotion of tissue repair ^[59][64].

Macrophages are present in large amounts in the intestine, primarily the colon, which has a high bacterial load. Intriguingly, these macrophages release mediators that promote homeostasis and thereby do not contribute to a proinflammatory environment ^{[60][61][62][63]}[65,66,67,68]. This selective inertia is important to maintain the homeostasis and epithelial integrity. Disturbances in this condition may be involved in the pathogenesis of UC and CD in humans ^[64][65][66][69,70,71].

An IBD murine model determined that under healthy conditions, macrophages display an anti-inflammatory phenotype (M2) with expression of MHC II, CD163, and IL-10 production. Under pathological conditions, monocytes differentiate into pro-inflammatory macrophages (M1), characterized by the expression of inducible NO synthase (iNOs), CD64⁺HLA-DR^hi CD14^lo, producing pro-inflammatory cytokines and chemokines such as TNFα, IL-1β, IL-6, IL-12, IL-23, and CCL11 ^[64][67][69,72].

Increased numbers of macrophages have also been identified in the duodenal mucosa of dogs with CIE ^[68][73]. Similar to humans with UC, boxer breed dogs with histiocytic ulcerative colitis (HUC) have higher infiltration of periodic acid-Schiff (PAS)-positive macrophages in the lamina propria in colonic and non-colonic affected regions, with a decrease in Goblet cells and an increase in MHC class II expression in enterocytes ^[69][74]. However, a later study revealed that HUC in boxer dogs is caused by enteroinvasive E. coli and can be successfully treated with fluoroquinolones such as enrofloxacin. Therefore, canine HUC is considered rather an infectious disease than belonging to the canine idiopathic CIE complex ^[11][70][8,75]. Similarly, granulomatous colitis in young French bulldogs has been also associated with the presence of invasive E. coli ^[70][75].

A recent study on dogs of different breeds and with or without CIE determined a reduced number of total macrophages but a slightly increased number of CD64⁺ macrophages, contributing to CIE pathogenesis ^[71][76]. Another study characterizing macrophages in the duodenum by immunohistochemistry, evaluated calprotectin (CAL) as a marker of early differentiated macrophages (M1) and CD163 expression as a marker of duodenal resident macrophages (M2), before and after treatment. This study demonstrated that macrophages play an important role in dogs with CIE. In particular, in dogs with FRE and IRE, the CD163⁺/CAL ratio is lower than in healthy dogs at diagnosis, and it is normalized after treatment in dogs with FRE. No significant differences were observed in dogs with ARE ^[72][77].

A study analyzing transcription nuclear factor (NF-κB) activation during mucosal inflammation in situ in dogs with CIE, identified significantly more macrophages/mm² with increased activity of the NF-κB pathway in the lamina propria ^[73][78], suggesting a role of NF-κB and derived pro-inflammatory cytokines in CIE.

Eosinophils are granulated cells that contribute to the host defense against parasites and play an important role in local immune regulation. In humans with IBD, eosinophils are increased in number along with IL-5 production, prompting circulation and activation ^[74][75][79,80]. Eosinophils infiltrating the intestinal mucosa release granules composed of crystalloid-containing core encapsulating cationic proteins and leukotriene C4 ^[75][76][77][80,81,82].

Canine CIE classification is based on the predominant type of inflammatory cells. The second most diagnosed form of CIE is in the small intestine where lymphoplasmacytic and eosinophilic enteritis are observed ^[78][79][80][83,84,85]. A study demonstrated that dogs with CIE have a significantly higher number of degranulated eosinophils in the lower region of the lamina propria, while the upper region has a significantly higher number of degranulated and intact eosinophils ^[81][6]. Comparing human and canine eosinophilic gastroenteritis showed that both pathologies have clinical and histopathological similarities ^[80][85], and canine eosinophilic gastroenteritis could be a good model for its human counterpart.

Various noninvasive tools and markers have been studied to identify eosinophil activation in the GI tract, including peripheral eosinophil counts and serum 3-bromotyrosine concentrations (3-BrY) ^[82][86]. Serum 3-BrY concentrations were also higher in dogs with SRE/IRE than in those with FRE or healthy control dogs. Thus, 3-BrY may serve as a noninvasive biomarker for CIE diagnosis and prognosis ^[83][87].

Another potential marker of eosinophil activity is the soluble epoxide hydrolase (sEH), a molecule with a proinflammatory role by metabolizing anti-inflammatory epoxyeicosatrienoic acid to proinflammatory diols. In a murine model, the use of a specific inhibitor of sEH significantly inhibited eosinophil migration, suggesting that sEH plays an important role in the migration of eosinophils to the gastrointestinal system ^[84][88].

Interestingly, a study of 30 dogs with CIE found that a significant number of dogs with CIE showed severe (n = 8) or moderate and mixed eosinophilic inflammation (n = 12). Future studies should be performed to further characterize the role of these eosinophils in canine CIE ^[85][89].

Mast cells, involved in the immediate and delayed defense against foreign antigens, also release mediators that affect the mucosal barrier ^[86][90]. More recently, a possible relationship between mast cells and host microbiota in human IBD pathogenesis has been proposed. This interaction is crucial to prevent mast cell hyper-reactivity. However, when microbiota genera are expanded, the interaction increases, favoring permeability and release of immunomodulatory molecules that promote inflammation ^[87][91].

In dogs with CIE, an increase in mast cells in the area of the eosinophilic gastroenterocolitis has been described, suggesting a role of type I hypersensitivity ^[80][85]. Thus, dogs with CIE have significantly more cells positive for IgE protein and mast cells in the mucosa, but their main location is mesenteric lymph nodes ^[88][92]. Moreover, a study defining the distribution and types of mast cells in the normal gastrointestinal tract of canines detected fewer mast cells in the villus area compared to the crypt areas; tryptase-positive mast cells (MC_T) were the most abundant cell type, followed by chymase- and a few tryptase- and chymase-positive mast cells (MC_C, MCT_C) ^[89][93]. However, in dogs with lymphocytic-plasmacytic or eosinophilic gastroenterocolitis, there was a decrease in the number of metachromatically stained granule-containing mast cells and a decrease in the number of the three types of mast cells identified (MC_{T, -C, -TC}), suggesting a mast cell degranulation or a Th1 predominant pattern ^[90][94].

N-Methylhistamine (NMH) is a stable metabolite of histamine and may be used as a marker of mast cell degranulation and gastrointestinal inflammation ^[91][95]. A study by Berghoff et al. showed that some dogs with CIE have increased fecal and/or urinary NMH concentrations, which could indicate increased mast cell activity. However, they were unable to definitively demonstrate such an association. In the same study, the researcheuthors suggest that urinary NMH concentrations could have clinical utility as a biomarker of chronic gastrointestinal inflammation, but this area remains to be explored further ^[92][96].

Natural killer lymphocyte (NKT) and natural killer (NK) cells also have a role in human IBD. Th2 cytokines such as IL-13, IL-5, and IL-4, involved in UC, are partly produced by NKTs ^{[93][94][95][96]}[97,98,99,100]. However, NKTs have a dual role as they play a protective role in a dextran sulfate sodium-induced colitis model ^[97][101] and a detrimental role in an oxazolone-induced model ^[98][102]. An increase in the cytotoxic CD56⁺CD16⁺ NK cell subset in the lamina propria in human IBD patients ^[99][103] and a decrease in the NKp44⁺/NKp46⁺ ratio in biopsies of CD patients have been demonstrated ^[100][104]. Recently, a meta-analysis evaluating the role of killer-cell immunoglobulin-like receptor (KIR) genes of IBD susceptibility in humans found that 2DL5 and 2DLS1 genes are associated with an increased risk of UC, while the 2DS3 gene is associated with a decreased risk of CD development ^[101][105]. Additionally, experimental treatment with monoclonal antibodies against NK Group 2D (NKG2D), a constitutively expressed receptor whose ligand is highly expressed in human IBD, has resulted in remission of CD in some patients ^[102][106].

The role of NK and NKT cells in CIE development in dogs has not been studied. Due to the relevance in human IBD, further investigation in dogs could provide relevant information.

Natural antibodies (Nabs), IgM and other pre-existent classes of immunoglobulins circling in plasma, are essential components of innate immunity reacting against foreign antigens and microbe-derived substances and activating the classical pathway of complement activation ^[103][104][107,108]. Its role in canine CIE has not been established but should be considered because, in a murine model, homeostatic intestinal IgAs are natural polyreactive antibodies with innate specificity to microbiota ^[105][109].

S100/calgranulins are a group of three phagocyte-specific damage-associated molecular pattern molecules (DAMPs) ^[106][107][110,111] that include S100A12 (calgranulin C) and the S100A8/A9 (calprotectin or calgranulin A/B) complex. The proteins are produced by activated macrophages and neutrophils and accumulate at sites of inflammation [3]. Recently, it has been suggested that fecal S100A12 and fecal calprotectin concentrations are clinically useful markers of gastrointestinal inflammation in dogs [3]. Fecal canine S100A12 concentrations are increased in dogs with CIE, associated with clinical disease activity, the severity of endoscopic lesions, and the severity of colonic inflammation in dogs with CIE ^[108][109][112,113].

Additionally, a recent study evaluated the expression of gastrointestinal mucosal receptor for advanced glycation end products (RAGE), which are considered molecular pattern receptors with relevance to inflammation in dogs with CIE, and its binding to canine S100/calgranulin ligands. CIE in dogs is associated with decreased serum sRAGE concentrations ^[110][114] and an increase in epithelial RAGE expression in the duodenum and colon ^[109][113] suggesting a dysregulated sRAGE/RAGE axis ^[110][114]. The epithelial RAGE expression in the duodenum and colon was significantly higher in dogs with CIE than in healthy controls, with a pattern of overexpression in the ileum and underexpression in the stomach. Thus, although the role of this axis in canine CIE is not completely understood, this axis might be a possible therapeutic target for dogs with CIE, with utility as a therapeutic model for humans ^[111][115].

Both commensals and pathogenic bacteria express pathogen-associated molecular patterns (PAMPs) on their surface, which are recognized by host pattern recognition receptors PRRs ^[112][116]. Among the best characterized PRRs are Toll-like receptors (TLRs) and NOD-like receptors (NOD) ^[113][114][117,118]. Under normal conditions, PRRs recognize antigens from food and commensal bacteria inducing tolerogenic responses. In canine CIE, these antigens, which normally induce immune tolerance, trigger an inflammatory response, with proliferation of T lymphocytes and production of several pro-inflammatory cytokines ^[112][116]. Thus, an increased TLR2, TLR4, and TLR9 mRNA expression in dogs with CIE has been described ^[115][116][119,120].

In human IBD, some mutations in these PRRs have been associated with its development ^[117][121]. Similarly, several mutations in PRRs have been also associated with the development of canine CIE. Single-nucleotide polymorphisms (SNPs) associated with TLR4 and TLR5 ^[118][122] and NOD2 ^[119][123] have been identified in German shepherd dogs with CIE. Additionally, a genetic component has been established in canine CIE, with a predisposition of certain breeds. Among those predisposed breeds are German shepherd dogs, Weimaraners, Rottweilers, border collies, and boxers ^[120][124].

Subsequently, a TLR5 haplotype has been identified, which is associated with a hypersensitivity to flagellin, exacerbating inflammatory pathways in dogs carrying this haplotype, increasing the risk of developing CIE ^[121][125].

While all these results provide valuable information for the development of possible genetic markers of CIE, it should be considered that CIE is a polygenetic disorder. Furthermore, these potential genetic markers would in many cases be expected to be breed-dependent ^[112][116].