SEARCH

SEARCH BY CITATION

Keywords:

  • Chronic enteropathy;
  • IgA;
  • Intestinal lymphoma;
  • Mucosal immunity

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Background

Although immunoglobulin A (IgA) plays a key role in regulating gut homeostasis, its role in canine inflammatory bowel disease (IBD) is unknown.

Hypothesis

IgA expression may be altered in dogs with IBD, unlike that observed in healthy dogs and dogs with other gastrointestinal diseases.

Animals

Thirty-seven dogs with IBD, 10 dogs with intestinal lymphoma, and 20 healthy dogs.

Methods

Prospective study. IgA and IgG concentrations in serum, feces, and duodenal samples were measured by ELISA. IgA+ cells in duodenal lamina propria and IgA+ CD21+ peripheral blood mononuclear cells (PBMCs) were examined by immunohistochemistry and flow cytometry, respectively. Duodenal expression of the IgA-inducing cytokine transforming growth factor β (TGF-β), B cell activating factor (BAFF), and a proliferation-inducing ligand (APRIL) was quantified by real-time RT-PCR.

Results

Compared to healthy dogs, dogs with IBD had significantly decreased concentrations of IgA in fecal and duodenal samples. The number of IgA+ CD21+ PBMCs and IgA+ cells in duodenal lamina propria was significantly lower in dogs with IBD than in healthy dogs or dogs with intestinal lymphoma. Duodenal BAFF and APRIL mRNA expression was significantly higher in IBD dogs than in the healthy controls. Duodenal TGF-β mRNA expression was significantly lower in dogs with IBD than in healthy dogs and dogs with intestinal lymphoma.

Conclusions and Clinical Importance

IBD dogs have decreased IgA concentrations in feces and duodenum and fewer IgA+ PBMCs, which might contribute to development of chronic enteritis in dogs with IBD.

Abbreviations
APRIL

a proliferation-inducing ligand

BAFF

B cell activating factor

CCECAI

canine chronic enteropathy clinical activity index

IBD

inflammatory bowel disease

Ig

immunoglobulin

PARR

PCR for antigen receptor gene rearrangement

PBMCs

peripheral blood mononuclear cells

TGF-β

transforming growth factor β

Inflammatory bowel disease (IBD) in dogs is a heterogeneous group of chronic enteropathies characterized by persistent or recurrent gastrointestinal clinical signs.[1, 2] Although the etiology and pathogenesis of canine IBD remain unclear, it is thought to develop from disruption of the mucosal barrier, dysregulation of the mucosal immune response, and alteration in the bacterial flora, as has been proposed for IBD in humans.[3] This concept is supported by the results of various investigations. For example, mRNA expression of several cytokines and chemokines is upregulated in the intestinal mucosa of dogs with IBD.[4-8] Similar to human patients with IBD,[9-11] increased Toll-like receptor (TLR) expression also is observed in intestinal lesions of dogs with IBD.[12, 13] Moreover, 16S rRNA gene sequencing studies identified alterations in the abundance of specific bacterial groups in dogs with IBD compared with healthy dogs.[14-16] These results suggest that alteration in the bacterial flora and uncontrolled TLR activation may lead to chronic enteritis in dogs with IBD through cytokine and chemokine overexpression. However, although the gastrointestinal mucosa is exposed to numerous and diverse bacteria, the intestine normally provides mucosal immune protection to prevent invasion of bacteria and aberrant TLR activation. This implies that abnormalities of immunologic molecules involved in immune protection may be crucial in the mucosal inflammation of canine IBD.

A key strategy of intestinal immune protection is the production of immunoglobulin A (IgA), the most abundant antibody isotype produced in the body, although it is the second most dominant isotype in the circulation after IgG.[17] IgA is largely produced in mucosal lymphoid tissues and plays important roles in mucosal immunity.[18] Mice that lack IgA or have impaired IgA secretion are more susceptible to intestinal toxins and pathogens,[19, 20] suggesting that IgA provides a first line of immune protection at mucosal surfaces. Furthermore, IgA deficiency alters the gut microbiota in mice,[21, 22] supporting a role for IgA in shaping the microbial composition of the gut. In humans, IgA deficiency is the most common primary immunodeficiency, and although many patients are asymptomatic, some suffer from recurrent infections, allergic disorders, autoimmune diseases, and gastrointestinal diseases including IBD.[23, 24] Indeed, the frequency of IgA deficiency among IBD patients is significantly higher than that in the healthy population.[25] Furthermore, decreased mucosal IgA concentrations are reported in human IBD patients.[26-30] These observations suggest that mucosal IgA deficiency may contribute to the pathogenesis of IBD in humans. In contrast, the involvement of IgA in IBD in dogs remains unclear, although IgA deficiency has been reported in several breeding colonies of dogs.[31-33] German Shepherd Dogs are known to be susceptible to canine IBD and have abnormalities in IgA production.[34, 35] Therefore, we speculated that IgA expression may be decreased in dogs with IBD compared with healthy dogs or dogs with other gastrointestinal diseases. The aim of this study was to analyze the concentrations of IgA in serum, feces, and the duodenum, and determine the frequency of IgA+ peripheral blood mononuclear cells (PBMCs) in dogs with IBD, dogs with intestinal lymphoma (as controls for non-IBD gastrointestinal disease), and healthy control dogs.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Study Population

Forty-seven dogs with clinical signs of chronic gastrointestinal disease that underwent endoscopic examination in the Veterinary Medical Center of the University of Tokyo (VMC-UT) between November 2009 and February 2012 were included in this prospective study. Informed consent was obtained from all owners, and the study protocol was approved by the animal care committee of VMC-UT. Case selection criteria of chronic enteropathies were as described previously.[36, 37] Dogs that had been treated with corticosteroids in the 2 weeks before the study were excluded. The dogs then were divided into 2 groups, IBD and intestinal lymphoma, based on the criteria described below.

A diagnosis of IBD was based on histopathologic evidence of intestinal inflammatory cell infiltrates, absence of clonal gene rearrangement of IgH or TCRγ by PCR for antigen receptor gene rearrangement (PARR), and exclusion of food-responsive and antibiotic-responsive diarrhea. Food-responsive diarrhea was ruled out in the absence of a complete response to a minimum period of 2 weeks on an elimination diet. Antibiotic-responsive diarrhea was ruled out if a complete response was not achieved after 2 weeks of treatment with metronidazole1 (10 mg/kg PO q12h) or tylosin2 (20 mg/kg PO q12h).

A diagnosis of intestinal lymphoma was based on histopathologic examination and the presence of clonal IgH or TCRγ gene rearrangements. All dogs diagnosed with IBD and intestinal lymphoma were scored for severity according to the canine chronic enteropathy clinical activity index (CCECAI).[2] All dogs were scored by a single-blinded investigator (S.M.), and the clinical severity of each case was categorized by the total CCECAI score: not clinically relevant (0–3), mild (4–5), moderate (6–8), severe (9–11), or very severe (≥12).

Twenty Beagles were used as healthy controls. This group included 10 females (1 intact and 9 spayed) and 10 males (all intact), with median age of 64 months (range, 31–92 months), and median body weight of 11.2 kg (range, 8.6–15 kg). These dogs were healthy with no clinical signs of gastrointestinal disease and had received no drugs. No abnormalities were observed on urinalysis and blood examinations, including CBC and measurements of blood urea nitrogen and creatinine concentrations, and alanine aminotransferase and alkaline phosphatase enzyme activity. Moreover, there were no abnormalities in the parasitic and bacterial analyses of the fecal samples.

Sample Collection

Fresh feces, serum, and EDTA blood samples were obtained from all of the dogs. Fecal and serum samples were stored at −80°C. EDTA blood samples were used immediately for isolation of PBMCs as described below. Duodenal tissue samples were collected by endoscopic biopsy.3 Dogs were prepared for endoscopy by fasting for 12–18 h. Six mucosal biopsy specimens were obtained from the stomach, duodenum, ileum, and colon or some combination of them for histopathology and immunohistochemistry with biopsy forceps.4 At least 1 biopsy specimen from each site was used for PARR.[38] Duodenal samples for RNA extraction were immediately submerged in RNAlater5 and stored at −20°C. Duodenal specimens for ELISA were rinsed twice in PBS and stored at −80°C.

Histopathology

Six mucosal biopsy specimens from each site were fixed in 10% formalin for 48 h, processed for histopathology, and stained with hematoxylin and eosin (HE). A histopathologic diagnosis of gastrointestinal inflammation was made according to the World Small Animal Veterinary Association criteria.[7, 36, 39] A histopathologic diagnosis of intestinal lymphoma was made from HE-stained sections and immunohistochemical sections stained for T cells6 (1 : 100 dilution) and B cells7 (1 : 400 dilution).[37, 40] Lymphocyte epitheliotropism, heterogeneity, and nuclear size were evaluated to differentiate between intestinal lymphoma and inflammation.

ELISA

Fecal and duodenal samples were placed in 500 μL of homogenization buffer (PBS containing 0.05% Tween 20 and Protease Inhibitor Cocktail8) and homogenized for 30 seconds with a TissueRuptor.9 Homogenates were briefly centrifuged and the supernatants frozen at −20°C for subsequent assays. After centrifugation, coarse particles were removed from fecal supernatants by filtration with a 0.8-μm syringe filter. IgA and IgG concentrations were determined by commercial ELISA kits specific for canine IgA10 and IgG.11 Absorbance was read at 450 nm with a microplate reader12 and results were analyzed by Microplate Manager software.13 All samples were tested in duplicate and the mean optical density (OD) was calculated. Total protein concentrations of the fecal and duodenal samples were determined by the Lowry method;14 IgA and IgG concentrations were normalized per milligram of total protein. Serum IgA and IgG concentrations were expressed as mg/dL.

Immunohistochemistry

Immunohistochemistry was conducted on paraffin-embedded sections (4 μm thick) of duodenal samples obtained by endoscopic biopsy. Briefly, heat-induced antigen retrieval was performed by autoclaving for 5 minutes at 121°C in 10 mM sodium citrate buffer (pH 6.0). Endogenous peroxidase was blocked with a blocking solution15 for 10 minutes at room temperature. The sections were blocked with 5% skimmed milk in TBS for 60 minutes at room temperature, and then incubated with a goat polyclonal antidog IgA antibody16 at 4°C overnight. After washing with TBS, samples were incubated with a biotinylated antigoat IgG antibody17 for 40 minutes at 37°C. Sections were washed again and incubated with HRP-labeled streptavidin18 for 40 minutes at room temperature. Reaction products were visualized with 3,3′-diaminobenzidine. Negative control slides were processed using an isotype-matched (normal goat IgG19) antibody as the primary antibody. Lamina propria IgA+ cells were enumerated by a single investigator (S.M.) with a microscope20 and video camera.21 Digitized images were transferred to a computer with a software package.22 Five areas were chosen at random for each standardized area within the base of villus and crypt (area 2 and 3),[41] and positively stained cells were counted. Results were expressed as positive cells per 10,000 μm2.

Flow Cytometry

PBMCs were collected from fresh EDTA blood samples by density gradient centrifugation.23 The frequency of IgA+ CD21+ PBMCs was examined by flow cytometry. Briefly, PBMCs were resuspended in staining medium (PBS supplemented with 5% FCS), and stained simultaneously with a PE-conjugated mouse monoclonal antidog CD21 antibody24 (1 : 10) and a FITC-conjugated goat polyclonal antidog IgA antibody25 (1 : 100). Appropriate isotypes were used as negative controls. Single staining samples were used for fluorescence compensation. After washing twice with PBS, fluorescence intensities were examined by subjecting the samples to flow cytometry.26 The lymphocyte population was gated based on the forward- and side-scatter characteristics to exclude other cell populations and cell debris. The data were processed by FlowJo software.27

Quantitative Real-Time RT-PCR

Total RNA was extracted from the duodenal samples using by RNAspin Mini RNA Isolation Kit.28 Genomic DNA was removed with the Turbo DNA-free Kit29 and the samples were stored at −20°C for subsequent assays. Expression of cytokines critical for IgA production (transforming growth factor β [TGF-β], B cell activating factor [BAFF], and a proliferation-inducing ligand [APRIL])[18, 42] in duodenal mucosa was quantified by 2-step real-time RT-PCR as described previously.[7, 37] Primer sequences used for quantitative PCR are shown in Supplementary Table 1. TBP and GAPDH were used as reference genes.[7, 43] After PCR, reaction products were sequenced directly by the dideoxy chain termination method30 to confirm amplification of the specific target genes.

Table 1. Summary of histopathologic scores based on the WSAVA standards in the duodenal mucosa of the 37 dogs with IBD
ScoreVillous StuntingEpithelial InjuryCrypt DistensionLacteal DilatationMucosal FibrosisIELsLPLsLamina Propria NeutrophilsTotal WSAVA Score
  1. IBD, inflammatory bowel disease; IELs, intraepithelial lymphocytes; LPLs, lamina propria lymphocytes; WSAVA, World Small Animal Veterinary Association.

  2. The numbers in each cell represent the number of dogs.

0 (normal)7581202726≤4 (insignificant)1
1 (mild)241118151151485–9 (mild)14
2 (moderate)51610811148310–14 (moderate)20
3 (marked)15122568015–19 (severe)2
         ≥20 (very severe)0
Total3737373737373737 37

Statistics

Statistical analyses were performed by a software package.31 Fisher's exact test was used to compare sex distribution among healthy dogs, dogs with IBD, and dogs with intestinal lymphoma. The Mann-Whitney U-test was used to compare serum albumin concentration and CCECAI score between dogs with IBD and dogs with intestinal lymphoma. Survival rates were compared by the log-rank test. The Kruskal–Wallis test was used to test for overall differences among dogs with IBD, dogs with intestinal lymphoma, and healthy dogs. The Steel–Dwass test was used to analyze between-group differences. The relationship between IgA concentrations and CCECAI scores was evaluated by the Spearman rank correlation coefficient. Statistical significance was defined as P < .05.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Dogs with IBD

Thirty-seven dogs diagnosed with IBD were included in this study. All of the dogs had evidence of inflammation within the intestinal mucosa and a histopathologic diagnosis of lymphocytic-plasmacytic enteritis. A summary of histopathologic scores of the IBD dogs is shown in Table 1. The breeds were Miniature Dachshund (n = 5), Toy Poodle (n = 4), Boston Terrier (n = 3), Cavalier King Charles Spaniel (n = 3), Pembroke Welsh Corgi (n = 2), Shetland Sheepdog (n = 2), Chihuahua (n = 2), Pug (n = 2), Shiba Inu (n = 2), Maltese (n = 2), Jack Russell Terrier (n = 2), Beagle (n = 1), Yorkshire Terrier (n = 1), Miniature Schnauzer (n = 1), Cairn Terrier (n = 1), French Bulldog (n = 1), Bichon Frise (n = 1), Pomeranian (n = 1), and Boxer (n = 1). The CCECAI scores of the dogs were 9/37 not clinically relevant, 11/37 mild, 10/37 moderate, 4/37 severe, and 3/37 very severe. Hypoalbuminemia (≤2.5 g/dL) was observed in 23/37 (62%) dogs. All of the dogs received prednisolone32 (0.5–2 mg/kg PO q24h) after endoscopy. Two dogs were treated with cyclosporine33 (8 mg/kg PO q24h) in combination with prednisolone.

Dogs with Intestinal Lymphoma

Ten dogs had intestinal lymphoma. All of the dogs were histopathologically diagnosed with lymphocytic (low-grade) type. Of these, 9 (90%) neoplasms were positive for TCRγ and 1 (10%) was positive for IgH by PARR. The breeds were Boston Terrier (n = 2), Shiba Inu (n = 2), Maltese (n = 1), French Bulldog (n = 1), Shetland Sheepdog (n = 1), Miniature Dachshund (n = 1), Papillon (n = 1), and mixed breed (n = 1). The CCECAI scores were 0/10 not clinically relevant, 0/10 mild, 4/10 moderate, 5/10 severe, and 1/10 very severe. Hypoalbuminemia was observed in 7/10 (70%) dogs. All of the dogs received prednisolone (2 mg/kg PO q24h) and 6 were treated with chlorambucil34 (2 mg/m2 PO q24h) in combination with prednisolone.

Comparison of Clinical Findings

There was no significant difference in sex among healthy dogs, dogs with IBD, and dogs with intestinal lymphoma (Table 2). Significant differences in age and body weight were observed among the 3 groups (P < .0001). Healthy dogs were significantly younger (P < .0001 versus IBD; P = .0003 versus intestinal lymphoma) and heavier (P < .0001 versus IBD and intestinal lymphoma) than the diseased dogs. However, there were no significant differences in age, body weight, and serum albumin concentration between dogs with IBD and those with intestinal lymphoma (Table 2). Dogs with intestinal lymphoma had significantly higher CCECAI scores than dogs with IBD (P = .0018). The overall survival time was significantly shorter for dogs with intestinal lymphoma than for dogs with IBD (P = .0027).

Table 2. Comparison of characteristics of healthy dogs, dogs with IBD, and dogs with intestinal lymphoma
VariableHealthyIBDIntestinal LymphomaP-Value
  1. CCECAI, canine chronic enteropathy clinical activity index; IBD, inflammatory bowel disease.

  2. The age, body weight, serum albumin, clinical score, and survival time data are expressed as median (range).

Sex   .111
Male10 (0 castrated)20 (4 castrated)4 (2 castrated) 
Female10 (9 spayed)17 (11 spayed)6 (3 spayed) 
Age (months)64 (31–92)94 (38–170)110 (80–170)<.0001
Body weight (kg)11.2 (8.6–15.0)6.2 (1.7–22.0)5.5 (2.7–9.9)<.0001
Serum albumin (g/dL)1.8 (1.2–3.7)2.3 (1.1–3.4).335
Clinical score (CCECAI)5 (1–16)10 (6–17).0018
Survival time (days)348 (26–618)170 (15–391).0027

IgA and IgG Concentrations in Serum, Feces, and Duodenal Samples

IgA concentrations in the fecal and duodenal samples were significantly lower in dogs with IBD (feces, median 77.6 μg/mg total protein, range 24–251.8; duodenum, median 152 μg/mg total protein, range 27.9–531.4) than in the healthy dogs (feces, median 116.2 μg/mg total protein, range 55.3–353.4, P = .0068; duodenum, median 322.3 μg/mg total protein, range 24.7–759.1, P = .0037). In addition, duodenal IgA concentrations were significantly lower in dogs with IBD (median 152 μg/mg total protein, range 27.9–531.4) than in dogs with intestinal lymphoma (median 260.5 μg/mg total protein, range 99.1–637.5, P = .0181). However, there were no significant differences in fecal and duodenal IgG concentrations, or in serum IgA or IgG concentrations, among the 3 groups (Fig 1). There was no significant correlation between IgA or IgG concentrations and the CCECAI score in dogs with IBD or intestinal lymphoma (data not shown).

image

Figure 1. IgA and IgG concentrations in serum, fecal, and duodenal samples of healthy dogs (n = 20), dogs with IBD (n = 37), and dogs with intestinal lymphoma (n = 10). IgA and IgG concentrations in fecal and duodenal samples were normalized to the total protein concentration in each sample. Data are presented as the median with 25th and 75th quartiles in each box plot. Whiskers indicate the highest and lowest data within 1.5 times the length of the quartiles. Circles represent outliers. Asterisks indicate statistical differences (*P < .05, **P < .01, ***P < .005).

Download figure to PowerPoint

Lamina Propria IgA+ Cells in Duodenal Mucosa

Canine IgA was strongly expressed in lamina propria cells (Fig 2A–C). The cells did not stain positive for the isotype-matched negative control antibody (Fig 2D). There were significantly fewer lamina propria IgA+ cells in the duodenal mucosa of dogs with IBD (median 11 cells/10,000 μm2, range 3–26) than in healthy controls (median 19 cells/10,000 μm2, range 9–36, P < .0001) or dogs with intestinal lymphoma (median 16 cells/10,000 μm2, range 10–21, P = .0303; Fig 2E). However, the number of lamina propria IgA+ cells in dogs with IBD and intestinal lymphoma was not significantly correlated with the CCECAI score (data not shown).

image

Figure 2. Detection of IgA in duodenal mucosa by immunohistochemistry. Samples are from (A) a healthy dog, (B) a dog with IBD, (C) a dog with intestinal lymphoma. (D) Isotype-matched negative control staining in a sample from a healthy dog. Bar = 50 μm. (E) The number of lamina propria IgA+ cells in duodenal mucosa of healthy dogs (n = 20), dogs with IBD (n = 37), and dogs with intestinal lymphoma (n = 10). Data are presented as the median with 25th and 75th quartiles in each box plot. Whiskers indicate the highest and lowest data within 1.5 times the length of the quartiles. Circles represent outliers. Asterisks indicate statistical differences (*P < .05, ***P < .005).

Download figure to PowerPoint

IgA and CD21 Expression on Peripheral Blood Lymphocytes

IgA expression on peripheral blood B cells was examined by flow cytometric analysis of PBMCs dually stained for canine IgA and CD21, a B cell marker (Fig 3A). The proportion of IgA+ CD21+ PBMCs was significantly lower in dogs with IBD (median 4.1%, range 0.2–16) than in healthy controls (median 11.9%, range 7.2–26.3, P < .0001) or dogs with intestinal lymphoma (median 11.3%, range 2.8–13.9, P = .0043; Fig 3B). However, there was no significant correlation between the proportions of IgA+ CD21+ PBMCs and the CCECAI score in dogs with IBD or intestinal lymphoma (data not shown).

image

Figure 3. IgA+ CD21+ peripheral blood lymphocytes by flow cytometry. (A) Representative results of the expression of IgA and CD21 on PBMCs. Quadrants were set with isotype control-stained cells (data not shown). The numbers indicate the proportions of cells in each quadrant. (B) Proportions of IgA+ CD21+ PBMCs in healthy dogs (n = 20), dogs with IBD (n = 37), and dogs with intestinal lymphoma (n = 10). Data are presented as the median with 25th and 75th quartiles in each box plot. Whiskers indicate the highest and lowest data within 1.5 times the length of the quartiles. Circles represent outliers. Asterisks indicate statistical differences (***P < .005).

Download figure to PowerPoint

Expression of IgA-Inducing Cytokines in Duodenal Mucosa

To investigate possible causes of the decreased IgA concentrations in dogs with IBD, we examined mRNA expression of cytokines critical for IgA class switching in duodenal tissue. Each primer pair specifically amplified cytokine mRNA, as confirmed by sequencing analysis. The relative expression of BAFF and APRIL mRNA was significantly higher in dogs with IBD than in healthy dogs (BAFF, P = .0068; APRIL, P = .0037; Fig 4). In contrast, TGF-β expression was significantly lower in dogs with IBD than in healthy dogs (P < .0001) or intestinal lymphoma cases (P < .0001; Fig 4).

image

Figure 4. mRNA expression of IgA-inducing cytokines in duodenal mucosa of healthy dogs (n = 20), dogs with IBD (n = 37), and dogs with intestinal lymphoma (n = 10). TBP and GAPDH were used as internal controls and similar results were obtained with both genes. Only the data standardized by TBP are shown. Data are presented as the median with 25th and 75th quartiles in each box plot. Whiskers indicate the highest and lowest data within 1.5 times the length of the quartiles. Circles represent outliers. Asterisks indicate statistical differences (*P < .05, ***P < .005). TGF, transforming growth factor; BAFF, B cell-activating factor; APRIL, a proliferation-inducing ligand.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

In this study, we showed that canine IBD is associated with decreased concentrations of IgA in the duodenum and feces, without any change in serum IgA concentrations. IgG expression was not significantly different among the dogs with IBD, healthy dogs, and dogs with intestinal lymphoma. These results are consistent with those from previous studies on German Shepherd Dogs, the most common breed predisposed to canine IBD.[34, 35] In addition, humans with IBD, mostly ulcerative colitis, also have been reported to have decreased IgA secretion by colonic mononuclear cells and a decreased population of colonic IgA containing cells.[26-30] Taken together, these findings suggest that canine IBD is associated with an IgA deficiency specifically in the gut, and this decrease may be involved in the chronic enteritis observed in this disease. Our study shows that dogs with intestinal lymphoma, used here as controls for chronic gastrointestinal inflammatory disease, have similar IgA and IgG concentrations as do healthy dogs in serum, duodenum, and feces. This result implies that intestinal IgA deficiency is not likely to be important in dogs with intestinal lymphoma. For further work, it is necessary to investigate IgA concentrations in various gastrointestinal diseases to confirm whether IgA deficiency is specific to canine IBD.

The decreased concentration of intestinal IgA in dogs with IBD may be caused by impaired secretion of IgA or failure of B cells to undergo class switching to IgA. We observed a significant decrease in the number of lamina propria IgA+ cells in the duodenal mucosa of dogs with IBD, unlike that observed in the other dogs, suggesting that the defective IgA expression may be caused by impaired IgA class switching rather than decreased secretion. However, the decreased number of duodenal IgA+ cells also may result from a relative increase in the number of other leukocytes. Previous immunohistochemical studies showed that dogs with IBD had increased numbers of IgG+ cells, CD3+ T cells, macrophages, and neutrophils in duodenal lamina propria.[44, 45] In this study, marked infiltrations of neutrophils, macrophages, and eosinophils were not observed in the duodenal lesions of dogs with IBD. Although immunohistochemistry for IgG and CD3 was not performed, a relative increase in lamina propria lymphocytes was evident in IBD dogs. These observations suggest that the decreased number of duodenal IgA+ cells may not result from a relative increase in other leukocytes.

To analyze whether IgA class switching is defective in dogs with IBD, we examined the proportions of IgA+ cells in CD21+ PBMCs. B cells undergo IgA class switching in response to interactions with CD4+ T cells and dendritic cells. In the gut, the resulting IgA+ B cells migrate from Peyer's patches to the lamina propria via the draining mesenteric lymph nodes, thoracic duct, and circulating blood.[17, 42] Therefore, if a defect in IgA class switching is present in intestinal B cells, the proportion of IgA+ B cells in peripheral blood also would be expected to decrease. Indeed, we found this to be the case. The proportion of IgA+ CD21+ PBMCs was significantly lower in dogs with IBD than in healthy dogs or those with intestinal lymphoma. This result supports the possibility that the observed decrease in intestinal IgA concentrations in dogs with IBD results from a failure of IgA class switching.

Class switch recombination to IgA occurs via both T cell-dependent and T cell-independent pathways.[18, 42] T cell-dependent IgA class switching is triggered by ligation of CD40 on B cells with CD40 ligand (CD40L) on CD4+ T cells and TGF-β cytokine signaling. T cell-independent IgA class switching is mediated by dentritic cells through BAFF and APRIL. These 2 molecules are soluble B cell-stimulating factors that are structurally and functionally related to CD40L.[46] In this study, we observed that TGF-β mRNA expression was significantly decreased in the duodenal mucosa of dogs with IBD, whereas BAFF and APRIL mRNA were increased. Interestingly, serum BAFF and APRIL concentrations are significantly higher in humans with IgA deficiency than in healthy donors.[47, 48] Because BAFF and APRIL promote IgA production, their overexpression may represent a physiologic compensatory mechanism in an attempt to overcome the decrease in IgA concentrations. Also consistent with our study, serum TGF-β concentrations are decreased in humans with IgA deficiency, unlike healthy controls.[23, 24] In addition, IgA concentrations in both serum and mucosal secretions are significantly decreased in TGF-β1-deficient mice.[49] Moreover, TGF-β receptor-deficient mice are virtually devoid of IgA.[50] Collectively, these results strongly suggest that decreased TGF-β expression in dogs with IBD is involved in the accompanying impaired IgA expression.

One limitation of this study is that the healthy control dogs were not well matched to the clinical cases by age and breed, although the sex, age, body weight, and serum albumin concentration did not differ significantly between dogs with IBD and those with intestinal lymphoma. Additional research will be necessary to determine if there are age- and breed-related differences in IgA concentrations.

In this study, we failed to identify a significant correlation between IgA concentrations and clinical severity in dogs with IBD or intestinal lymphoma. We can exclude observer variation in CCECAI scoring as a source of error in our study, because all of the cases were scored by the same blinded investigator (S.M.) who was not the clinician for the cases. Possible explanations for the lack of correlation are that IgA concentrations may not correlate with clinical severity, or that the sample size was too small to detect a significant correlation. More extensive studies will be necessary to determine if there is an association between IgA concentrations and clinical severity in canine IBD.

In summary, we detected decreased IgA concentrations in feces and duodenum, and a lower proportion of IgA+ PBMCs, in dogs with IBD. However, the present results do not allow a conclusion as to whether the decreased IgA concentrations play a role in disease pathogenesis or simply may have occurred as a consequence of disease. Additional work will be required to investigate the molecular mechanisms underlying these findings.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

This study was supported by a Grant-in-Aid for Science Research (KAKENHI no. 21380193) and JSPS Fellows of Japan Society for the Promotion of Science.

Conflict of Interest: Authors disclose no conflict of interest.

Footnotes
  1. 1

    Flagyl, Shionogi & Co, Osaka, Japan

  2. 2

    Tylan, Intervet, Tokyo, Japan

  3. 3

    VQ-8143B flexible videoendoscope, Olympus Medical Systems, Tokyo, Japan

  4. 4

    FB-54Q-1 biopsy forceps, Olympus Medical Systems

  5. 5

    RNAlater, Qiagen, Valencia, CA

  6. 6

    Rabbit polyclonal anti-human CD3, Dako, Glostrup, Denmark

  7. 7

    Rabbit polyclonal anti-human CD20, Thermo Fisher Scientific, Waltham, MA

  8. 8

    Protease Inhibitor Cocktail, Sigma-Aldrich, St. Louis, MO

  9. 9

    TissueRuptor, Qiagen

  10. 10

    Dog IgA ELISA Quantitation Set, Bethyl Laboratories

  11. 11

    Dog IgG ELISA Quantitation Set, Bethyl Laboratories

  12. 12

    Microplate reader, Bio-Rad Laboratories, Hercules, CA

  13. 13

    Microplate Manager software version 5.2.1, Bio-Rad Laboratories

  14. 14

    DC Protein Assay, Bio-Rad Laboratories

  15. 15

    REAL Peroxidase-Blocking Solution, Dako

  16. 16

    Goat anti-Dog IgA Antibody, Bethyl Laboratories

  17. 17

    Biotinylated Anti-Goat IgG Antibody, Kirkegaard & Perry Laboratories, Gaithersburg, MD

  18. 18

    HRP-labeled streptavidin, Dako

  19. 19

    Normal Goat IgG, Santa Cruz Biotechnology, Santa Cruz, CA

  20. 20

    BX51 Research System Microscope, Olympus

  21. 21

    DP71 Microscope Camera, Olympus

  22. 22

    DP Controller version 3.3.1.292, Olympus

  23. 23

    Ficoll-Paque PLUS, GE Healthcare, Buckinghamshire, UK

  24. 24

    PE-conjugated mouse monoclonal anti-dog CD21 antibody (CA2.1D6), AbD Serotec, Oxford, UK

  25. 25

    FITC-conjugated Goat anti-Dog IgA Antibody, Bethyl Laboratories

  26. 26

    FACSCalibur, BD Biosciences, Franklin Lakes, NJ

  27. 27

    FlowJo, Tree Star, Ashland, OR

  28. 28

    RNAspin Mini RNA Isolation Kit, GE Healthcare

  29. 29

    Turbo DNA-free Kit, Applied Biosystems, Foster City, CA

  30. 30

    ABI prism BigDye Terminator v3.1 Cycle Sequencing Kit, Applied Biosystems

  31. 31

    JMP version 9, SAS Institute, Cary, NC

  32. 32

    Predonine, Shionogi & Co

  33. 33

    Atopica, Novartis Animal Health, Tokyo, Japan

  34. 34

    Leukeran, GlaxoSmithKline, Tokyo, Japan

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
jvim12023-sup-0001-TableS1.docxWord document17KTable S1. Sequences of oligonucleotide primers used for quantitative real-time PCR.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.