Distinct patterns of IgG and IgA against food and microbial antigens in serum and feces of patients with inflammatory bowel diseases

Abstract

Background: Inflammatory bowel disease (IBD) is associated with a defective intestinal barrier and enhanced adaptive immune responses against commensal microbiota. Immune responses against food antigens in IBD patients remain poorly defined.

Methods: IgG and IgA specific for food and microfloral antigens (wheat and milk extracts; purified ovalbumin; Escherichia coli and Bacteroides fragilis lysates; mannan from Saccharomyces cerevisiae) were analyzed by ELISA in the serum and feces of patients with Crohn's disease (CD; n = 52 for serum and n = 20 for feces), ulcerative colitis (UC; n = 29; n = 17), acute gastroenteritis/colitis (AGE; n = 12; n = 9) as well as non-inflammatory controls (n = 61; n = 39).

Results: Serum anti-Saccharomyces cerevisiae antibodies (ASCA) and anti-B. fragilis IgG and IgA levels were increased in CD patients whereas antibody (Ab) levels against E. coli and food antigens were not significantly different within the patient groups and controls. Subgroup analysis revealed that CD patients with severe diseases defined by stricturing and penetrating lesions have slightly higher anti-food and anti-microbial IgA levels whereas CD and UC patients with arthropathy have decreased anti-food IgG levels. Treatment with anti-TNF-α Abs in CD patients was associated with significantly decreased ASCA IgG and IgA and anti-E. coli IgG. In the feces specific IgG levels against all antigens were higher in CD and AGE patients while specific IgA levels were higher in non-IBD patients. Anti-food IgG and IgA levels did not correlate with food intolerance.

Summary: In contrast to anti-microbial Abs, we found only minor changes in serum anti-food Ab levels in specific subgroups of IBD patients. Fecal Ab levels towards microbial and food antigens show distinct patterns in controls, CD and UC patients.

Figure 1. Serum IgG and IgA levels specific for food and microbial antigens in IBD patients and controls.

Serum IgG (A–F) and IgA (G–L) specific for ovalbumin (A/G), wheat (B/H), milk (C/I), mannan from S. cerevisiae  =  ASCA (D/J), and lysates from E. coli K12 (E/K) and B. fragilis ATCC 25285 (F/L) were quantified by ELISA in control patients/healthy controls (Con; n = 61) and patients suffering from CD (n = 52), UC (n = 29) and acute gastroenteritis/colitis (AGE; n = 12). Each dot represents one patient. Medians with interquartile ranges are indicated. P values (*<0.05; **<0.01; ***<0.001) were determined by Kruskal-Wallis test followed by a Dunn's post hoc test.

Figure 2. Anti-food and anti-microbial serum IgG and IgA levels in CD patients with or without stricturing/penetrating disease and controls.

Specific serum IgG (A) and IgA (B) were quantified by ELISA in control patients/healthy controls (n = 61) and CD patients without (n = 17) and with (n = 34) stricturing and/or penetrating complications. Boxes indicate median and 25/75 percentiles and whiskers indicate 10/90 percentiles. P values were determined by Kruskal-Wallis test followed by a Dunn's post hoc test (*<0.05; **<0.01; ***<0.001) or by Mann Whitney U test (# <0.05). Mann Whitney U test was only applied between CD subgroups and results are only shown if the Dunn's post hoc test did not show any significance.

Figure 3. Anti-food and anti-microbial serum IgG and IgA levels in CD patients with or without arthropathy and controls.

Specific serum IgG (A) and IgA (B) were quantified by ELISA in control patients/healthy controls (n = 61) and CD patients without (n = 39) and with (n = 12) current arthropathy. Boxes indicate median and 25/75 percentiles and whiskers indicate 10/90 percentiles. P values were determined by Kruskal-Wallis test followed by a Dunn's post hoc test (*<0.05; **<0.01; ***<0.001).

Figure 4. Anti-food and anti-microbial serum IgG and IgA levels in CD patients with or without anti-TNFα treatment and controls.

Specific serum IgG (A) and IgA (B) were quantified by ELISA in control patients/healthy controls (n = 61) as well as in CD patients without (n = 17) and with (n = 34) current anti-TNFα treatment. Boxes indicate median and 25/75 percentiles and whiskers indicate 10/90 percentiles. P values were determined by Kruskal-Wallis test followed by a Dunn's post hoc test (*<0.05; **<0.01; ***<0.001) or by Mann Whitney U test (# <0.05). Mann Whitney U test was only applied between CD subgroups and results are only shown if the Dunn's post hoc test did not show any significance.

Figure 5. Fecal IgG and IgA levels specific for food and microbial antigens in IBD patients and controls.

Specific fecal IgG (A/B) and IgA (C/D) were quantified by ELISA in control patients/healthy controls (Con; n = 39) and patients suffering from CD (n = 20), UC (n = 17) and acute gastroenteritis/colitis (AGE; n = 9). (A/C) Percentage of measurements above the detection limit and (B/D) geometric means ± 95% confidential interval are shown. (A/C) P values (*<0.05; **<0.01) were determined by Chi-square test. (B/D) P values (*<0.05; **<0.01; ***<0.001) were determined by Kruskal-Wallis test followed by a Dunn's post hoc test. The results of the Kruskal-Wallis test are indicated above the graph and the results of the Dunn's post hoc test are indicated directly above the bars.