Antibodies Studied in Ulcerative Colitis

The first studies on antibodies in patients who had IBD focused on autoantibodies to colonic mucosa. These antibodies against a fraction of the crude colon mucosa were found to be specific for patients who had UC . Cross-reactivity with the Escherichia coli bacterial antigen 0:14 had already been reported, suggesting a link with bacterial epitopes. In the following years, antigoblet cell antibodies were described and were very specific for UC, although their sensitivity was low . It is still unclear whether both antibodies are directed to the same epitopes.

A subset of pANCA on IIF has been associated with colonic IBD: for UC, a sensitivity and a specificity of 60% to 80% and 90%, respectively, were reported, as was 10% sensitivity for UC-like CD . A high prevalence of pANCA has also been associated with primary sclerosing cholangitis and autoimmune hepatitis . The role of pANCA in the pathogenesis of UC is not clear; one of the reasons is the target antigen against which pANCA is directed is not known. A number of nuclear and cytoplasmic antigens have been proposed as the target epitopes for the atypical pANCA pattern in UC (eg, myeloperoxidase, proteinase-3, lactoferrin, cathepsin G and C, elastase, lysozyme, bactericidal permeability increasing protein, glucosidase, and galactosidase); however, none has proved to be the causing antigen . In addition, alpha-enolase has been proposed as a target antigen. In a recent study, the authors evaluated the prevalence and diagnostic value of anti-alpha-enolase antibodies in IBD and related disorders . By means of Western blot analysis, a cohort of 525 subjects was screened for the presence of anti-alpha-enolase antibodies. Anti-alpha-enolase antibodies were detected in 49.0% of patients who had UC, in 50.0% of patients who had CD, in 30.5% of patients who had primary sclerosing cholangitis, in 37.8% of patients who had autoimmune hepatitis, in 34.0% of patients who had ANCA-positive vasculitis, in 31.0% of non-IBD gastrointestinal control subjects, and in 8.5% of healthy control subjects. Gene array experiments on RNA extracted from colonic mucosal biopsies from 35 IBD and 6 control subjects showed a significant up-regulation of alpha-enolase mRNA in colonic mucosal biopsy samples from patients who had IBD, but not from control subjects. There was no association, however, between the presence of pANCA and anti-alpha-enolase antibodies. In addition, preabsorption with alpha-enolase did not eliminate the pANCA pattern on IIF. From these results, it seems that alpha-enolase is not the target epitope for the atypical pANCA in UC. Although anti-alpha-enolase antibodies are present in a substantial proportion of patients who have IBD, they are not specific for this disease and are also found in patients who have various inflammatory/autoimmune disorders and in non-IBD gastrointestinal control subjects. Therefore, anti-alpha-enolase antibodies are of limited diagnostic value for the diagnosis of IBD.

Because the exact epitope for pANCA has not been found, the current assays to detect these antibodies consist of an IIF technique, with an atypical perinuclear fluorescence on ethanol-fixed neutrophil slides. Although pANCA refers to a perinuclear fluorescence, it is the inner border of the nucleus that is stained. Because a 50-kd nuclear envelope protein was identified as the target for the pANCA in UC , it was proposed that the terminology of pANCA should be changed to “perinuclear antineutrophil antinuclear antibodies (pANNA).” Further identification of this 50-kd protein is anticipated and may lead to a better understanding of the nature of these antibodies.

A comparative study using five different pANCA IIF assays showed very heterogeneous results among laboratories, with sensitivities among UC patients ranging from 0% to 63% ( Table 2 ). Methodological differences (preparation of neutrophils) were the most important determinants for the heterogeneity in results . When the epitope of pANCA is identified, future studies using standardized ELISAs may overcome the methodological problems as they are now seen.

Antibodies Studied in Crohn’s Disease

The search for antibodies in CD has been limited by the lack of specificity. Although lymphocytotoxic antibodies have been reported in approximately 40% of CD patients and in 30% of unaffected relatives, unaffected spouses of IBD patients also express seroreactivity . Up to 70% of CD patients carry antibodies to epithelial cell components, but these antibodies lack specificity and are detected in other gastrointestinal diseases (30%) and systemic autoimmune diseases (10%) . Anti-erythrocyte antibodies are directed against the heavy chain of the Vh3-15 IgG gene product . They occur at a high frequency (90%) in CD patients, in 30% of UC patients and healthy control subjects, and in a high percentage of patients who have infectious gastroenteritis.

Pancreatic antibodies (PAB) are very specific (95%) for CD and are found at a frequency of 30% in these patients. They are absent in healthy relatives of CD patients . PAB are very stable over time and do not correlate with disease activity, pattern of disease, or extraintestinal manifestations. Patients who have CD and concomitant pancreatic insufficiency, however, are significantly more frequently PAB positive than CD patients who do not have pancreatic insufficiency . On the contrary, in patients who have CD and acute pancreatitis, PAB do not occur more often than in CD patients who do not have acute pancreatitis. These antibodies are directed against a not-yet-further-identified 1300-kd antigen localized within the pancreatic secretions. Several studies focused on defined pancreatic proteins but could not identify them as the causative antigens. Further studies are needed to identify the antigen, which in turn could lead to increased utility of this marker.

ASCA, an antibody against the bakers’ and brewers’ yeast Saccharomyces cerevisiae , has been associated with CD . The reported prevalence is 60% to 70% in patients who have CD, 10% to 15% in patients who have UC, and 0% to 5% in control subjects. The epitope against which ASCA is directed consists of oligomannosidic sequences in the yeast’s cell wall . It is not clear, however, whether ASCA are truly directed against this particular yeast or whether they represent cross-reactivity with other, yet-unidentified microorganisms that also carry mannan sequences. A number of other microorganisms such as Candid a and Mycobacteria have been studied in this context but results were negative . How these antibodies are formed is unclear, but immunization by yeasts is one hypothesis. Dietary yeasts and yeast species that are natural inhabitants of the gastrointestinal tract must be considered immunogen candidates. Most CD patients have increased intestinal permeability, therefore exposing the epithelium to food antigens such as yeasts, consequently leading to increased exposure of yeast to immune reactive cells. Some investigators have postulated that abnormal permeability is a primary defect. The fact that patients who have celiac disease also show ASCA may point toward an increased small bowel permeability as a common etiology. In patients who have CD, ASCA are found especially in small bowel disease. Finally, it could also be that the oligomannosides share structural homologies with oligomannosides expressed on human glycoconjugates as autoantigens .

Accuracy data on ASCA reported a sensitivity for CD of 45% to 60% and a specificity of 85% to 95% for differentiating CD from control subjects in adults . A similar study performed in a pediatric IBD population showed similar results .

In previous years, more antibodies directed against bacterial epitopes have been discovered in patients who have CD. One is the antibody against outer membrane protein C of E coli (anti-OmpC) and an epitope of Pseudomonas fluorescens (anti-I2) . The most recent antibody is cBir1, directed against bacterial flagellin epitopes .

Two years ago, a new panel of antiglycan antibodies was described. These markers include antilaminaribioside antibodies (ALCA), antichitobioside antibodies (ACCA), and antimannobioside antibodies (AMCA) .

Antibodies Studied in Crohn’s Disease

The search for antibodies in CD has been limited by the lack of specificity. Although lymphocytotoxic antibodies have been reported in approximately 40% of CD patients and in 30% of unaffected relatives, unaffected spouses of IBD patients also express seroreactivity . Up to 70% of CD patients carry antibodies to epithelial cell components, but these antibodies lack specificity and are detected in other gastrointestinal diseases (30%) and systemic autoimmune diseases (10%) . Anti-erythrocyte antibodies are directed against the heavy chain of the Vh3-15 IgG gene product . They occur at a high frequency (90%) in CD patients, in 30% of UC patients and healthy control subjects, and in a high percentage of patients who have infectious gastroenteritis.

Pancreatic antibodies (PAB) are very specific (95%) for CD and are found at a frequency of 30% in these patients. They are absent in healthy relatives of CD patients . PAB are very stable over time and do not correlate with disease activity, pattern of disease, or extraintestinal manifestations. Patients who have CD and concomitant pancreatic insufficiency, however, are significantly more frequently PAB positive than CD patients who do not have pancreatic insufficiency . On the contrary, in patients who have CD and acute pancreatitis, PAB do not occur more often than in CD patients who do not have acute pancreatitis. These antibodies are directed against a not-yet-further-identified 1300-kd antigen localized within the pancreatic secretions. Several studies focused on defined pancreatic proteins but could not identify them as the causative antigens. Further studies are needed to identify the antigen, which in turn could lead to increased utility of this marker.

ASCA, an antibody against the bakers’ and brewers’ yeast Saccharomyces cerevisiae , has been associated with CD . The reported prevalence is 60% to 70% in patients who have CD, 10% to 15% in patients who have UC, and 0% to 5% in control subjects. The epitope against which ASCA is directed consists of oligomannosidic sequences in the yeast’s cell wall . It is not clear, however, whether ASCA are truly directed against this particular yeast or whether they represent cross-reactivity with other, yet-unidentified microorganisms that also carry mannan sequences. A number of other microorganisms such as Candid a and Mycobacteria have been studied in this context but results were negative . How these antibodies are formed is unclear, but immunization by yeasts is one hypothesis. Dietary yeasts and yeast species that are natural inhabitants of the gastrointestinal tract must be considered immunogen candidates. Most CD patients have increased intestinal permeability, therefore exposing the epithelium to food antigens such as yeasts, consequently leading to increased exposure of yeast to immune reactive cells. Some investigators have postulated that abnormal permeability is a primary defect. The fact that patients who have celiac disease also show ASCA may point toward an increased small bowel permeability as a common etiology. In patients who have CD, ASCA are found especially in small bowel disease. Finally, it could also be that the oligomannosides share structural homologies with oligomannosides expressed on human glycoconjugates as autoantigens .

Accuracy data on ASCA reported a sensitivity for CD of 45% to 60% and a specificity of 85% to 95% for differentiating CD from control subjects in adults . A similar study performed in a pediatric IBD population showed similar results .

In previous years, more antibodies directed against bacterial epitopes have been discovered in patients who have CD. One is the antibody against outer membrane protein C of E coli (anti-OmpC) and an epitope of Pseudomonas fluorescens (anti-I2) . The most recent antibody is cBir1, directed against bacterial flagellin epitopes .

Two years ago, a new panel of antiglycan antibodies was described. These markers include antilaminaribioside antibodies (ALCA), antichitobioside antibodies (ACCA), and antimannobioside antibodies (AMCA) .