OUP user menu

Journal of Crohn's and Colitis: 10 (12)

Editor-in-Chief

Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium

6.585
5.586

Published on behalf of

Loss of tolerance to one or two major targets in Crohn's disease or just cross-reactivity?

Dirk Roggenbuck, Dimitrios Bogdanos, Karsten Conrad
DOI: http://dx.doi.org/10.1016/j.crohns.2012.12.013 e273-e274 First published online: 1 August 2013

Dear Sir,

We read with interest the paper of Komorowski et al. reporting the discovery of two autoantigenic targets, CUB and zona pellucida-like domain-containing protein 1 (CUZD1) and glycoprotein 2 (GP2), in Crohn's disease (CD).1 The identification of GP2 as CD-specific target and the role of antibodies to GP2 (anti-GP2) in the serology and pathophysiology of CD have been reported extensively.2,3 Unfortunately, data regarding antibodies to CUZD1 (anti-CUZD1) as a CD-specific marker are sparse. Thus, we speculated about the relevance of anti-CUZD1 as a CD-specific autoantibody. Investigating 103 pediatric patients with CD and 74 with UC, we did not find significantly different prevalences for combined anti-CUZD1 and anti-GP2 IgG/IgA testing by indirect immunofluorescence (IIF), employing human embryonic kidney cell (HEK293)-expressed antigens compared with anti-GP2 IgG/IgA testing by enzyme-linked immunosorbent assay (ELISA) (P = 0.91; Table 1). We found, that the prevalence of anti-GP2 detected by IIF is significantly lower than that detected by ELISA in patients with CD (30/103 vs 46/103, respectively; P = 0.0423). In our experience, sensitive determination of CD-specific anti-GP2 requires post-translationally modified GP2 released from the cell surface. Conversely, recombinant human GP2 purified directly from infected cells was readily reactive with rabbit polyclonal anti-GP2 antibodies by ELISA, but did not interact with the specific antibodies of CD patients. We therefore postulate that the GP2 expressed in transfected cells might not provide all autoantigenic CD epitopes required.4

View this table:
Table 1

Prevalence of IgG and IgA to glycoprotein 2 (anti-GP2) and CUB and zona pellucida-like domain containing protein 1 (anti-CUZD1) in 103 pediatric patients with Crohn's disease (CD) and 74 pediatric patients with ulcerative colitis (UC). Anti-GP2 anti-CUZD1 were detected by a commercial indirect immunofluorescence (IIF) assay (Euroimmun AG, Lübeck, Germany) employing recombinant CUZD1 as well as GP2 expressed in transfected human kidney cells (HEK293).1 Furthermore, IgG and IgA antibodies to GP2 were determined by commercial enzyme-linked immunosorbent assay (ELISA) employing recombinant GP2 expressed in the baculovirus system (GA Generic Assays GmbH, Dahlewitz/Berlin, Germany).4

IIF on transfected cellsELISA
Anti-GP2 IgG/A n (%)Anti-CUZD1 IgG/A n (%)Anti-GP2/CUZD1 IgG/A n (%)Anti-GP2 IgG n (%)Anti-GP2 IgA n (%)Anti-GP2 IgG/A n (%)
CD (n = 103)30 (29.1)29 (28.2)45 (43.7)45 (43.7)26 (25.2)46 (44.7)
UC (n = 74)7 (9.5)11 (14.9)13 (17.6)14 (18.9)3 (4.0)14 (18.9)
Total (n = 177)37 (20.9)40 (29.2)58 (32.8)59 (43.1)29 (21.2)60 (43.8)

Furthermore, CUZD1 synthesis has not yet been demonstrated at the site of CD inflammation, as has been done for GP2.2 Remarkably, GP2 appears to modulate innate and acquired immune intestinal responses and has been shown to be a specific receptor on microfold cells in the follicle-associated epithelium, presumably the site of CD inflammation onset.3,5 Additionally, there are currently no reports regarding the amount of CUZD1 secreted together with zymogens by the pancreas into the intestine. Interestingly, the two main protein fractions purified by affinity chromatography employing Ulex europaeus agglutinin I (UEA-I) contained GP2 according to Komorowsi et al.1 The monoclonal antibodies utilized by these authors to identify the CD-specific autoantigenic targets have been generated on the basis of reactivity to pancreatic juice and different IIF patterns on pancreatic tissue sections.1 So far, the authors have not reported their reactivity with purified CUZD1 and GP2. Thus, we presume that there may be cross-reactivity between both autoantigenic targets. Indeed, CUZD1 and GP2 share zona pellucida-like domains which could bear putative cross-reactive epitopes.3 In our present study, 60% of anti-CUZD1 positive sera were also positive by testing in the anti-GP2 ELISAs, and in 25% of those cases even in the absence of anti-GP2 determined by IIF.

In conclusion, we suggest CD-specific autoantibody testing by molecular tests employing purified recombinant proteins like in particular GP2 and presumably CUZD1 for performance evaluation thereof.

References

View Abstract