Evaluation of Systemic and Secretory IgA Concentrations and Immunohistochemical Stains for IgA-Containing B Cells in Mucosal Tissues of an Irish Setter With Selective IgA Deficiency
Immunoglobulin A is the predominant secretory antibody at mucosal surfaces. In the dog, immunoglobulin A deficiency (IgAD) is characterized by low to absent serum IgA and normal to elevated serum immunoglobulin G (IgG) and immunoglobulin M (IgM) concentrations. However, studies comparing serum and secretory IgA in dogs have often documented a poor correlation, suggesting that serum concentrations should not be used to estimate mucosal secretion of this antibody. This report demonstrates the concurrent use of serum IgA, IgG, and IgM; secretory IgA (from bronchoalveolar lavage fluid); and immunohistochemical stains on bronchial and duodenal mucosa for IgA-containing B cells in a young Irish setter with recurrent respiratory and gastrointestinal signs.
Case Report
A 1-year-old, spayed female Irish setter was referred to the Veterinary Medical Teaching Hospital at the University of California, Davis (VMTH-UCD) for evaluation of recurrent bacterial pneumonia and failure to gain weight appropriately. The dog was noted to have a moist cough at 2 months of age when acquired from a breeder. Veterinary attention was not sought until the dog was 4 months of age, at which time the referring veterinarian diagnosed pneumonia based on a fever and the presence of alveolar infiltrates in the caudal portion of the left cranial lung lobe evidenced on thoracic radiography. Clinical signs and radiographic lesions resolved on a 2-week course of amoxicillin with clavulanic acid (10 mg/kg body weight, per os [PO] q 12 hours). Clinical signs recurred at 8 months and 11 months of age, and each episode responded to 2-week courses of amoxicillin with clavulanic acid. The dog was being fed a commercially available adult maintenance diet and had a variable appetite. The dog had no history of regurgitation or vomiting, but she occasionally had diarrhea. At the time of referral to the VMTH-UCD, the dog was asymptomatic. The physical examination was unremarkable, except the dog had a small stature, was thin with a body weight of 23 kg, and had a body condition score of 2/9.
The major differential diagnoses for the recurrent pneumonia included repeated aspiration (e.g., secondary to megaesophagus or esophageal dysmotility, oronasal fistula, or other anatomical defects), foreign body pneumonia, or primary or secondary immunodeficiency (e.g., defects in the innate arm including physiological barriers and phagocytic cells, or in the adaptive arm including humoral and cell-mediated immunity). Differential diagnoses for failure to gain weight included organ insufficiency (e.g., renal, hepatic, cardiac, exocrine pancreatic), endocrinopathy, primary gastrointestinal disease (e.g., food allergy, parasites, inflammatory bowel disease), and chronic pneumonia.
To look for evidence of a current infection or inflammatory condition, or an organ insufficiency or endocrinopathy, a complete blood count, serum biochemical profile, and urinalysis were performed. The only significant abnormality noted was a mild peripheral eosinophilia (1,460/μL; reference range, 100 to 1,250/μL). Survey lateral and dorsoventral radiography of the thorax was performed to determine if there was current evidence of pneumonia; radiographs were unremarkable except for a small amount of air in the cervical esophagus. To rule out a megaesophagus or esophageal dysmotility, an esophagram was performed with fluoroscopic guidance and showed normal esophageal motility with no evidence of esophageal dilatation. A fecal flotation and direct smear, looking for a parasitic infection as a cause of the failure to gain weight, revealed no parasites. Empirical therapy was started, pending results of serum trypsin-like immunoreactivity, cobalamine, folate, and quantitative serum immunoglobulin G (IgG), immunoglobulin M (IgM), and immunoglobulin A (IgA), which were done to rule out exocrine pancreatic insufficiency and malabsorptive syndromes, and immunoglobulin deficiencies, respectively. Therapy consisted of fenbenda-zole (50 mg/kg body weight, PO q 24 hours for 3 days, to be repeated in 3 weeks) as an anthelmintic, and oxytetracycline (20 mg/kg body weight, PO q 12 hours) for treatment of possible small intestinal bacterial overgrowth (SIBO). Trypsin-like immunoreactivity (TLI) was in the low normal range (7.1 μg/L; reference range, 5 to 35 μg/L), and both cobalamine and folate were normal (397 ng/L; reference range, 249 to 733 ng/L; and 8.5 μg/L; reference range, 6.5 to 11.5 μg/L, respectively). Quantitative serum IgG, IgM, and IgA performed by single radial immunodiffusion (SRID) revealed elevated IgG and IgM concentrations (30.1 mg/mL; reference range, 10 to 20 mg/mL; and 2.5 mg/mL; reference range, 1 to 2 mg/mL, respectively) and an IgA concentration that was not detectable on SRID (reference range, 0.11 to 0.93 mg/mL).a
To further determine if the dog had selective IgA deficiency with low to undetectable amounts of IgA in secretions from mucosal surfaces, and to determine the cause for the dog’s failure to gain weight, the owners elected to have bronchoscopy and gastroduodenoscopy performed. They returned for these procedures 1 month later. At that time, there had been no change in body weight, and the physical examination was unchanged from the first visit. Gross bronchoscopic examination was unremarkable, and a total yield of 87 out of 120 mL of sterile saline was retrieved from bronchoalveolar lavage, with multiple airways being sampled. Five bronchial mucosal biopsies were obtained. Several abnormalities were noted on gastroduodenoscopy. The stomach had patchy edematous regions, and there were pinpoint erosions in the pyloric antrum. The duodenal mucosa was erythematous, friable, and bled excessively. Tissue easily sloughed off in the duodenum when the endoscope was maneuvered through this area. Duodenal juice was aspirated using an endoscopy aspiration kit.b At the time of collection, the qualities of the biopsy specimens (six from the stomach and 12 from the duodenum) were considered poor.
One aliquot of bronchoalveolar lavage fluid (BALF) was used for aerobic and anaerobic culture, and a small number of Flavobacterium spp. grew. A second aliquot was submitted for cytopathological examination but was inadvertently lost. The third aliquot of BALF was concentrated 100 times for quantitative IgA analysis, and no IgA was detected using SRID.a As a control, BALF collected from two additional dogs seen at the VMTH-UCD by the same clinician (Norris) as the dog of this case report, also had the same concentration procedure and IgA analysis by SRID performed. The volume of saline instilled during each of the lavage procedures was 120 mL, and the total yield of lavage fluid was similar in all three dogs (±8 mL). The control dogs each had IgA concentrations of 1.6 mg/mL.
Duodenal juice was used for aerobic and anaerobic culture, and the results were not consistent with SIBO (total viable bacterial cell counts, <105 organisms/mL). The quantity of duodenal juice was too small to also allow for quantitative IgA analysis. Routine histopathological examination of tissue from the bronchial, gastric, and duodenal mucosa, using hematoxylin and eosin stain, showed no significant lesions. The samples from the gastrointestinal tract were of poor quality on microscopic examination. Bronchial mucosal biopsies and duodenal mucosal biopsies were also embedded in OCT Compoundc for frozen sectioning, and then, using immunofluorescence, they were stained for IgA-containing B cells.d In both the bronchial and duodenal mucosa, several IgA-containing B cells were seen. The causes for the gross abnormalities on visual examination of the duodenal mucosa and the dog’s inability to gain weight were not definitively determined; full-thickness biopsies of the intestinal tract were offered but declined by the owners.
The dog was sent home on a 1-month trial of pancreatic enzyme supplementatione (2 teaspoons), mixed with low-fat cottage cheese and boiled rice q 12 hours. Although the TLI was within the reference range, it was near the low end of the reference range and may have been indicative of subclinical exocrine pancreatic insufficiency (EPI) as a cause of poor ability to gain weight. The dog refused to eat the cottage cheese and rice at home. The owners mixed the pancreatic enzyme supplement in the previous commercial adult maintenance diet, and the dog ate normally. At a 1-month recheck, the dog had gained 0.8 kg body weight. The owner was instructed to continue with the pancreatic supplement and return for another recheck in 1 month. Subsequently, the owners moved, and the dog was lost to follow-up.
Discussion
In the dog, immunoglobulin A deficiency (IgAD), also termed relative or selective IgA deficiency, is characterized by low to absent serum IgA concentrations and normal to elevated serum IgG and IgM concentrations.1–4 More rigorous criteria for the disease in humans includes a deficiency of secretory IgA, normal cellular immunity, and normal humoral antibody production in response to antigenic stimulation.56 Selective IgA deficiency is reportedly the most common primary immunodeficiency in humans, and it has been described in German shepherd dogs, beagles, and Chinese shar peis, among others.347 Since many humans and dogs with IgAD are clinically asymptomatic,28–10 it has recently been suggested in human medicine that the term “selective IgAD” should be reserved for patients who have low IgA concentrations in conjunction with clinical signs attributable to the deficiency.11 The clinical diseases associated with IgAD include recurrent upper and lower respiratory tract infections, gastrointestinal infections (including SIBO), skin infections, allergies, and immune-mediated diseases.1–367
Based on this Irish setter’s recurrent respiratory infections and gastrointestinal signs, undetectable serum IgA and elevated serum IgG and IgM concentrations, undetectable IgA from the BALF, and the presence of IgA-containing B cells in the bronchial tissue and duodenal mucosa, the dog was diagnosed as having selective IgAD. Most studies in the veterinary literature have evaluated serum IgA concentrations alone.247–9 Rarely have concurrent secretory IgA levels10 or staining for IgA-containing cells in mucosal surfaces been performed in dogs with serum IgAD.13 To the authors’ knowledge, this is the first report documenting the absence of both serum and secretory IgA, and the presence of IgA-containing B cells in mucosal tissue, in a dog with clinical signs compatible with IgAD.
Small intestinal bacterial overgrowth was suspected in the dog of this report, but it was not definitively proven. A lack of increased bacterial numbers in the duodenal juice could have been secondary to the recent empirical oxytetracycline therapy, and the cobalamine and folate concentrations that were within the reference ranges have been previously reported in dogs with SIBO.1 The current definition of SIBO is under debate, with some reports using a value of >105 organisms/mL of duodenal juice112 and others recommending a value of >109 organisms/mL of duodenal juice.13 Regardless, and using either definition, the Irish setter of this report did not have SIBO at the time of gastroduodenoscopy. Inflammatory bowel disease could also not be ruled out based on results of histopathological examination of biopsies obtained endoscopically. It is suspected that surgically obtained biopsies would have provided more insight on any gastrointestinal lesions.
Immunoglobulin A is the predominant secretory antibody at mucosal surfaces, and it is produced locally by plasma cells.101415 The normal function of secretory IgA is to inhibit adherence and colonization of mucosal surfaces by pathogenic organisms and to modulate absorption of macromolecular antigens.137 The presence of IgA in secretions from mucosal surfaces helps explain why specific immune resistance to local infection can exist in the absence of demonstrable serum antibody;14 this emphasizes the importance of determining secretory IgA concentrations in patients suspected to have selective IgAD. Evaluation of canine secretory IgA has been performed in tears, saliva, bile, and duodenal juice.1101617 In clinically healthy dogs without IgAD, there was a poor correlation between serum IgA and secretory IgA from saliva, tears, and bile; this suggests that serum concentrations should not be used to estimate mucosal secretion of this antibody.16 In dogs with low serum IgA concentrations but lacking clinical signs, secretory IgA levels in the tears were within the control range, implying that low serum levels do not necessarily reflect a reduction in secretory levels.10 However, in a study of clinically symptomatic dogs (German shepherd dogs with SIBO), mean serum IgA concentrations and duodenal juice IgA concentrations were both significantly lower than concentrations in the various breeds from the control group.1 In the dog of this report, no secretory IgA was detected in the BALF, suggesting a lack of mucosal IgA as a possible contributing factor to the recurrent bacterial pneumonias.
The underlying defect in selective IgAD is believed to result from an arrest in the B cell differentiation pathway, increases in T suppressor cell function, or decreases in T helper cell function.18 The precursors to IgA-secreting plasma cells, B lymphocytes possess surface IgA that can be detected by immunofluorescent staining; in a study in humans with IgA deficiency, lymphocytes bearing membrane-bound IgA were present in all patients evaluated.19 The authors concluded that B cells from IgAD patients had undergone an isotype switch toward IgA, but they had not differentiated into IgA-producing cells. In the veterinary literature, immunoperoxidase staining has been employed to evaluate for the presence of IgA-containing cells in duodenal or jejunal mucosal specimens.120 In German shepherd dogs with low serum and secretory (duodenal juice) IgA concentrations, the densities of IgA-containing cells were not significantly different from the control group, which had normal serum and secretory IgA concentrations.1 This suggests that in dogs with selective IgAD, defective synthesis or secretion of IgA from B cells in the mucosa, rather than decreased or absent local IgA-producing cells, is a more likely cause of the clinical signs. The findings in the patient of this report are in agreement with the previous study, in that IgA-containing cells were present in the bronchial and duodenal mucosa.
The underlying immunological defect that gives rise to IgAD has yet to be fully elucidated, but evidence supports a role for cytokines and T helper cells in IgA production.111518 Interleukin 10, added in vitro to peripheral blood lymphocytes from humans with IgAD, was shown to enhance B cell proliferation and induce B cell differentiation into IgA-secreting plasma cells.18 A study in mice using Peyer’s patch T lymphocytes activated with the mitogen con-canavalin A, showed a 20-fold increase in B cell IgA production.15 Maximal B cell secretory activity was attributed to increased T helper cell activity and elimination of T suppressor cell activity. The molecular basis of this disease is likely to have both genetic and environmental influences.61118
There is currently no available specific treatment for selective IgAD. Even if there were a source of purified IgA for replacement therapy, parenteral administration would likely not be reflected as an increase in local secretory IgA.5 In humans, development of anti-IgA antibodies is known to occur, especially with prior treatment with commercial gamma globulins, and can result in anaphylaxis to transfused blood products.511 This complication has not yet been described in dogs with IgAD. The mainstay of therapy is to treat infections (if or when they develop) with the appropriate antimicrobial, ideally based on culture and sensitivity.
Rabbit anti-canine IgG (H+L); The Binding Site, San Diego, CA and Goat anti-canine C3, goat anti-canine IgM, and goat anti-canine IgA; Bethyl Laboratories, Montgomery, TX
Product ±64830 (for use with scopes having a 2.6-mm diameter channel, 1.4-m long); Har-Vet Inc., Spring Valley, WI
Tissue-Tek OCT Compound; Lab Tek Products, Miles Inc., Elkhart, IN
Goat anti-canine IgA; Bethyl Laboratories, Montgomery, TX
Viokase; Fort Dodge Laboratories, Fort Dodge, IA
