- Email: [email protected]
Protein A and protein G ELISA for the detection of IgG autoantibodies against tissue transglutaminase in childhood celiac disease
Clinica Chimica Acta 395 (2008) 72–76
Contents lists available at ScienceDirect
Clinica Chimica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n c h i m
Protein A and protein G ELISA for the detection of IgG autoantibodies against tissue transglutaminase in childhood celiac disease Ingrid Dahlbom a,b,⁎,1, Daniel Agardh c,1, Tony Hansson b,d a
Department of Genetics and Pathology, Uppsala University, Uppsala, Sweden Department of Women's and Children's Health, Uppsala University, Uppsala, Sweden Unit of Diabetes and Celiac Disease, University Hospital MAS, Department of Clinical Sciences/Pediatrics, Lund University, Malmö, Sweden d Rheumatology Unit, Department of Medicine at Karolinska Hospital, Karolinska Institutet, Stockholm, Sweden b c
A R T I C L E
I N F O
Article history: Received 1 February 2008 Received in revised form 8 May 2008 Accepted 8 May 2008 Available online 15 May 2008 Keywords: Celiac disease ELISA Protein A Protein G Tissue transglutaminase
A B S T R A C T Objectives: To investigate if the detection of celiac disease (CD) in children was improved by using alternative conjugates for assessment of tissue transglutaminase (tTG) autoantibodies. Methods: Serum samples from 108 biopsy confirmed CD children and 42 control subjects were investigated for the presence of autoantibodies with tTG coated microplates using protein A (PA), protein G (PG), anti-IgG, or anti-IgA as conjugates. Results: Of the 108 CD children, 86 (80%) were IgG-tTG positive, 91 (84%) were positive with the PAconjugate, 94 (87%) were positive with the PG-conjugate, and 103 (95%) were IgA-tTG positive. Among the 42 controls, 4 (10%) were IgG-tTG positive, 5 (12%) were positive with both the PA- and PG conjugates, whereas 3 (7%) were IgA-tTG positive. Compared with IgG-tTG the concordance was 93% for PA and 95% for PG, with a positive correlation between antibody levels (r = 0.967 and r = 0.975, p b 0.0001). All but one CD child were found positive by combining IgG-tTG and IgA-tTG detection. Conclusions: The sensitivity of IgG-tTG detection with ELISA increased by protein A or protein G conjugates, whereas the specificity was reduced as compared with anti-IgG conjugate. The combined measurement of IgA-tTG and IgG-tTG still seems to be the optimal procedure when screening children for CD. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Celiac disease (CD) is a small bowel enteropathy due to intolerance against gliadin and related prolamines in wheat, barley and rye [1]. The diagnosis is confirmed by a biopsy showing typical histopathological features characterized by mucosal villous atrophy of the small intestine and exclusion of gluten from the diet results in restitution of the mucosa and remission of the disease [2]. The clinical presentation has been suggested to be age dependent with more profound symptoms among small children [3]. However, during the last decades it has become evident that the majority of CD patients may present with minor or atypical signs of malabsorption and some patients experience no symptoms, despite a damaged intestinal mucosa [4]. CD is strongly associated with the presence of IgA endomysial autoantibodies (EMA) directed against tissue antigens localized in the extracellular matrix of the small intestine [5]. The enzyme tissue transglutaminase (tTG) has been identified as the main autoantigen of EMA [6] and the expression of tTG is increased in the mucosal lesion of
⁎ Corresponding author. Department of Women's and Children's Health, Pediatrics, SE 751 85 Uppsala University, Sweden. E-mail address: [email protected] (I. Dahlbom). 1 These authors contributed equally to this paper. 0009-8981/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2008.05.005
untreated patients with CD [7,8]. As the majority of IgA-sufficient patients have elevated levels of IgA-tTG these markers are frequently used for the identification and follow-up of patients with CD [9]. Using enzyme-linked immunosorbent assays (ELISA) for the detection of IgA-tTG the reported diagnostic sensitivity and specificity have been estimated at 86–100% and 95–100%, respectively [10–12]. Selective IgA deficiency is a CD associated disorder which occurs in approximately 1/500 of the general population, who will consequently not be detected by conventional IgA based tTG assays [13]. Whereas the sensitivity and specificity of IgG-tTG analyzed with ELISA in IgA-deficient CD patients has proven highly accurate reaching 99% [14], the sensitivity of IgG-tTG among IgA-sufficient individuals is reduced [15]. Conversely, IgG-tTG analyzed with radioligand binding assays (RBA) seems to display a high sensitivity (99%) and specificity (96%) for untreated CD independent of the total serum concentration of IgA [16]. The observed discrepancy for ELISA and RBA suggests that the binding of IgG-tTG might differ between the methods [17]. With the ELISA method, patient serum is brought into contact with immobilized antigen and tTG-specific IgG antibodies are thereafter detected with enzyme-linked anti-IgG conjugate. With the RBA method, serum autoantibodies react with radioactive antigen in solution and tTG-specific IgG antibodies are then separated from the solution by means of protein A (PA) or protein G (PG) coated beads. PA produced by
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
73
2. Materials and method 2.1. Patients This retrospective study encompassed 150 serum samples obtained from children consecutively admitted to the Department of Pediatrics, University Hospital MAS in Malmö with suspected CD for the investigation with intestinal biopsy during the years 2002–2006. All biopsies were examined and evaluated independently by two histopathologists. A total of 108 children (71 females, 37 males), including 3 children with IgA deficiency, had atrophic intestinal findings [20]. They were diagnosed with CD at median 5.3 years of age (range 1.2–19.0 years) according to the revised criteria of European Society of Pediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) [2]. The remaining 42 children (21 females, 21 males) had normal biopsies at median 4.3 years of age (range 1.4–16.4 years) and were included as disease control subjects. Among the disease control subjects, four children had selective IgA deficiency and another child partial IgA deficiency, three had transient EMA, three were diagnosed with wheat or cow's milk protein allergy, one had Helicobacter pylori gastritis, one lipase deficiency, one type 1 diabetes and the remaining 32 had temporary gastrointestinal symptoms or were investigated because of failure to thrive. In addition, 20 of the untreated CD children (13 females, 7 males) were followed with gluten-free diet with serum samples taken after three and six months from diagnosis at median 3.9 years (range 1.3–13.9 years). The Ethics Committee of Medical Faculty, Lund University approved the study. 2.2. Protein A and protein G ELISA
Fig. 1. Receiver operator characteristic (ROC) curves for tissue transglutaminase (tTG) antibodies detected with ELISA using protein A (PA-tTG), protein G (PG-tTG), antihuman IgG (IgG-tTG), and anti-human IgA (IgA-tTG) as conjugates.
Staphylococcus aureus binds with strong affinity all subtypes of IgG except IgG3, and also some IgA and IgM [18], whereas streptococcal PG binds all human IgG subclasses except IgA and IgM [19]. It is therefore possible that the higher clinical performance of IgG-tTG assessed with RBA is due to particular antibody binding properties of PA or PG. Although the IgG-tTG assessed with RBA has proved to be highly accurate for CD screening, the method is based on the use of radioactivity and is therefore difficult to standardize for laboratories not frequently using the method. In contrast, the ELISA technique is handled by most of laboratories and more accessible to all standard laboratories. A clinical easily accessible method for detection of tTG antibodies irrespective of immunoglobulin isotype would overcome the possibility to miss potential CD patients who only have increased levels of IgG-tTG. Few efforts for optimizing current ELISA methods have been performed and the use of PA and PG has not yet been evaluated in the current IgG-tTG based ELISA system. In this study, we hypothesized that the use of PA or PG as conjugates might improve the detection of tTG antibodies also in the ELISA in a similar manner as what has been demonstrated for the RBA method. The aim of the present study was therefore to investigate if PA or PG conjugates in the ELISA format could benefit the detection of tTG antibodies compared to the use of traditional anti-IgA or anti-IgG conjugates.
For quantification of anti-tTG antibodies detected with PA or PG, microtiter plates and reagents included in the Celikey-IgG kit, except for the conjugate, were used. The detection of anti-tTG antibodies was made with horseradish peroxidase conjugated (HRP) recombinant PA, diluted 1:80,000, or HRP conjugated recombinant PG, diluted 1:6000, (Pierce Biotechnology, Rockford, IL, USA). The concentration of anti-tTG antibodies was expressed as arbitrary units (U/mL), calculated from a 6-point calibrator curve containing 0, 3, 7, 16, 40, and 100 U/mL IgG-tTG. The within and between Cv of the PA ELISA was b4.5% and b 3.5% respectively, with a total Cv of b 5.5%. The within and between Cv of the PA ELISA was b3.7% and b11.1% respectively, with a total Cv of b 11.7%. Cut-off values according to ROC curves were 4.0 U/ml for the PA ELISA and 5.0 U/ml for the PG ELISA, (Fig. 1). 2.3. IgA and IgG anti-tTG-ELISA Celikey based on human recombinant tTG was used for quantification of IgA antitTG and Celikey IgG was used for detection of IgG anti-tTG (Phadia, Freiburg, Germany). The relative amount of tTG antibodies was expressed as U/mL calculated from standard calibrator curves constructed to contain 0, 3, 7, 16, 40, and 100 U/mL of respective IgA and IgG anti-tTG [21]. The within and between Cv of Celikey was b 8.0% and b1.5% respectively, with a total Cv of b 8.1%. The within and between Cv of Celikey IgG was b5.7% and b3.2% respectively, with a total Cv of b5.7%. The cut-off levels chosen for the present study were 3.0 U/mL for IgA anti-tTG and 6.0 U/mL for IgG anti-tTG according to ROC curves [22], (Fig. 1). 2.4. Total IgA and EMA Total serum IgA was determined by turbidometry [23], and samples with IgA concentrations less than 0.07 g/L were further analyzed for establishment of IgAdeficiency (b 0.05 g/L) by rocket immunoelectrophoresis [24]. EMA was detected with fluorescein isothiocyanate conjugated goat anti-human IgA antibodies applied to tissue slides of primate esophagus (Euroimmun, Lübeck, Germany) and visualized by immunofluorescence [5]. Results were expressed as the highest dilution factor giving a positive fluorescence pattern in microscope and all sera manifesting fluorescence at a titer N1:10 were considered to be positive.
Table 1 Detection of anti-tissue transglutaminase (tTG) antibodies in children with untreated celiac disease and in disease controls Celiac disease (N =108) ELISA
Disease controls (N =42)
Positive
Sensitivity
Median value
Positive
Specificity
Median value
(n)
(95% CI)
(5–95th centile)
(n)
(95% CI)
(5–95th centile)
84.3% (76.0–90.6) 86.1% (79.2–92.7) 79.6% (70.8–86.8) 95.4% (90.2–98.9)
21 U/ml (1.3–106) 25 U/ml (1.5–240) 26 U/ml (1.8–104) 80 U/ml (3.7–1200)
5
88.6% (75.4–96.2) 88.6% (75.4–96.2) 90.5% (77.4–97.3) 92.9% (80.5–98.5)
1.5 U/ml (0.5–5) 2.1 U/ml (0.7–8) 1.6 U/ml (0.6–6) 0.4 U/ml (0.1–7)
PA-tTG
91
PG-tTG
93
IgG-tTG
86
IgA-tTG
103
N; total number of patients, 95% CI; 95% confidence interval.
5 4 3
74
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
CD children of whom 82 were detected with the IgG-tTG ELISA, 87 were IgA-tTG positive, and 86 had positive EMA titres. All 3 (2.8%) CD children with IgA deficiency were detected with PA, PG as well as with the IgG-tTG ELISA, whereas EMA and IgA-tTG were negative. Another 3 (2.8%) children were positive with the PG ELISA, IgA-tTG and EMA, whereas they were negative with the ELISAs using PA or anti-IgG as conjugates. An additional 12 children had IgA-tTG and EMA only. Of the remaining two children, one was positive for IgA-tTG, IgG-tTG, and EMA and the other child had EMA only. A total of 34/42 (81.0%) disease controls had normal levels of IgGtTG and IgA-tTG according to all ELISAs and were EMA negative as well. Five (11.9%) disease control children were detected with the PA and PG ELISAs of whom four were positive for IgG-tTG, three for EMA and one had also IgA-tTG. The only child detected with both PA and PG ELISA has IgA-deficient. Three (7.1%) additional disease control children were IgA-tTG positive and two of them were EMA positive. The remaining child had type 1 diabetes and was EMA negative at the time of biopsy, but had a previous history of fluctuating EMA. 3.2. The diagnostic performance of the tTG antibody assays The sensitivity for PA ELISA was 84.3% and for PG ELISA 87.0%, while the specificity was 88.6% for both assays (Table 1). Sensitivity of IgG-tTG and IgA-tTG was 79.6% and 95.4%, whereas the specificity was 90.5% and 92.9%, respectively. The concordance for a positive and negative test result was 92.7% (139/150) for PA and 94.7% (142/150) for PG when compared with the IgG-tTG ELISA and the tTG antibody levels correlated (r = 0.967 and r = 0.9759, respectively, p b 0.0001) (Fig. 2A and B). The correlation was even higher between PA-tTG and IgG-tTG (r = 0.976, p b 0.001) (Fig. 2C). The concordance was reduced to 82.0–84.0% (120– 127/150) when the three PA, PG and IgG-tTG assays were compared with the IgA-tTG ELISA, although a positive correlation between tTG antibody levels was still obtained (r = 0.647–0.697, p b 0.001). 3.3. The effect of gluten-free diet on tTG antibody levels
Fig. 2. Correlation between the levels of tissue transglutaminase (tTG) antibodies detected with ELISA using (A) anti-human IgG (IgG-tTG) and protein A (PA-tTG) as conjugates, (B) IgG-tTG and protein G (PG-tTG) as conjugates, and (C) PA-tTG and PGtTG as conjugates. Filled symbols represent celiac disease patients and unfilled symbols represent controls.
In 20 CD patients followed for 6 months during a gluten-free diet the tTG antibody levels decreased (p b 0.01) in all patients except for one child. Two children had elevated IgG-tTG levels only, and one of them was IgA-deficient. For this child, the IgG-tTG levels decreased during the gluten-free diet, although the antibody levels remained elevated after 6 months of treatment. A modest reduction in IgG-tTG levels was noted for most of the CD children already after 3 months, but the median IgG-tTG levels did not decrease significantly until 6 months (Fig. 3). At this point 13/20 (65%) and 14/20 (70%) were still positive with the PA and PG and 15/20 (75%) remained positive with the IgG-tTG ELISA.
2.5. Statistical methods Differences in tTG antibody levels between independent groups were calculated using the Mann–Whitney U-test. The Wilcoxon signed rank test was used to test significant change in autoantibody levels before and after gluten-free diet. Correlations between the assessed tTG antibody levels were evaluated using Spearman rank correlation (r). p-values b 0.05 were considered significant.
3. Results 3.1. tTG antibodies in untreated CD children and disease control subjects The area under the ROC curve was 0.92 (95%CI = 0.880–0.9636) for the PA ELISA, 0.937 (95%CI = 0.900–0.974) for the PG ELISA, 0.92 (95% CI = 0.878–0.962) for IgG-tTG and 0.969 (95%CI = 0.841–0.996) for IgAtTG (Fig. 1). At the respective cut-off estimated from the ROC curves, both the PA and PG ELISAs detected a total of 91/108 (85.8%) untreated
Fig. 3. Median tissue transglutaminase (tTG) antibody levels detected with ELISA using protein A (PA-tTG), protein G (PG-tTG), anti-human IgG (IgG-tTG), and anti-human IgA (IgA-tTG) as conjugates in 20 celiac disease children during a gluten-free diet.
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
In contrast, the IgA-tTG levels decreased rapidly within 3 months for most of the children (p b 0.001) of whom 16/20 (80%) children had normal IgA-tTG levels after 6 months of gluten-free diet. 4. Discussion In this study we evaluated the clinical performance of CD associated anti-tTG antibodies assessed with ELISA where the conventional polyclonal anti-human IgA and anti-human IgG conjugate had been substituted with PA or PG conjugates. We demonstrate that the levels of IgG-tTG were elevated in the majority of children with untreated CD, including those with IgA-deficiency, and that the use of PA or PG slightly increased the sensitivity of the ELISA as compared to the conventional anti-IgG conjugate. Moreover, our results indicate that the use of PA or PG as conjugates in the ELISA technique mainly detected IgG-tTG as the levels of tTG-specific antibodies highly correlated with the antibody levels obtained with the anti-IgG conjugate. However, the IgA-tTG ELISA was by far the most sensitive assay (95%) for the identification of untreated CD, missing only five of the included 108 CD children, of whom three were IgA-deficient. These three children and one additional child were detected with all the anti-tTG ELISAs measuring IgG, whereas one child with minor mucosal alterations was EMA positive only. On the other hand, a considerable fraction of the untreated CD children were not detected with any of the IgG-tTG ELISA, whereas five of the control subjects were detected with the ELISAs using PA or PG as conjugates. Two of these children were also EMA positive, in spite of a normal small intestine mucosa, and it is possible that these children have a latent manifestation with potential risk to develop clinical CD later in life [25]. Our findings are in accordance with previous studies carried out with the ELISA technique, showing that many untreated IgA-sufficient patients are negative for IgG-tTG [15,26]. This contrasts with the results obtained from studies using immunoprecipitation methods or radiobinding assays for the detection of IgG-tTG, which have shown the same diagnostic specificity as IgA-tTG [16,27]. However, with a recent comparison between IgA-tTG and IgG-tTG assessed with ELISA and RBA we observed that the choice of resin for immunoprecipitation with the RBA method appears to be critical for the detection of IgGtTG. No major difference in the prevalence of IgG-tTG as measured with the ELISA or RBA was observed when anti-IgG beads were used for immunoprecipitation, while the use of PA beads displayed a 99% concordance with IgA-tTG [17]. The higher prevalence of IgG-tTG observed with RBA using PA beads might thus depend on the immunoglobulin-binding properties of PA itself, which has a strong binding affinity for the Fc portion of human IgG, but also binds IgA and IgM [18]. One might therefore hypothesize that PA has the ability to detect both IgA-tTG and IgG-tTG captured by the antigen. Another hypothesis is that PA might bind IgG with a higher affinity than anti-IgG conjugates promoting the detection of lower amounts with PA. Moreover, PG binds all human IgG subclasses and proven an even higher binding affinity than PA [19] and might therefore increase the sensitivity of IgG-tTG detection even further. At least from our own observations, the sensitivity of IgG-tTG was increased using PG resins in the RBA method (unpublished results). Still, our results demonstrated that the advantages of using PA or PG for detection of IgG-tTG in the RBA method were not transferable to the ELISA technique. Even if these conjugates improved the sensitivity slightly as compared to the anti-IgG conjugate, the combined detection of IgA-tTG and IgG-tTG identified 99% of all untreated CD children including the CD children with IgA-deficiency. A possible explanation for this technical drawback could be the difference in binding of the antigen in the two systems. In a solid phase ELISA the microtiter wells are coated with a limited amount of tTG, which might cause a competition between antibodies directed against epitopes in close proximity favouring the binding of antibodies with the highest affinity for tTG. In contrast, RBA
75
uses an excess of PA linked to the beads with a higher capacity to bind antibodies irrespective of their affinity for tTG. In the present study, the decrease pattern of the tTG-specific antibodies during six months of gluten-free diet evinced no major differences between the IgG-tTG assays regardless of conjugate. Only few of the children regained normal levels of IgG-tTG. It has previously been reported that the time for normalization of the tTG antibody response may be as long as one year [28]. In this study, levels of IgAtTG decrease most dramatically of all assays within the first 3 months. This is in line with our previous observations showing a more rapidly decline in IgA-tTG than IgG-tTG levels, suggesting that the withdrawal of gluten is first reflected in the IgA antibody response [29]. In summary, our study showed only slightly improvement of the diagnostic sensitivity in the detection of IgG-tTG when PA or PG was used as conjugates instead of anti-IgG tTG, whereas the disease specificity was reduced as compared with the tTG ELISA using anti-IgG conjugate. The optimal procedure to detect untreated CD in paediatric patients and how they respond to gluten-free diet still seems to be the combined measurement of IgA-tTG and IgG-tTG. Acknowledgements The study was funded by Gillbergska stiftelsen, Uppsala, the Faculty of Medicine, Lund University, and the Skåne Council Foundation for Research and Development, University Hospital MAS in Malmö. All reagents and kits for this study were kindly provided by Phadia AB, Uppsala, Sweden. References [1] Trier JS. Celiac sprue. N Engl J Med 1991;325:1709–19. [2] Walker-Smith J, Guandalini S, Schmitz J, Shmerling D, Visakorpi J. Revised criteria for diagnosis of coeliac disease: Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition (ESPGAN). Arch Dis Child 1990;65:909–11. [3] Collin P, Kaukinen K, Maki M. Clinical features of celiac disease today. Dig Dis 1999;17:100–6. [4] Mäki M, Collin P. Coeliac disease. Lancet 1997;349:1755–9. [5] Chorzelski TP, Sulej J, Tchorzewska H, Jablonska S, Beutner EH, Kumar V. IgA class endomysium antibodies in dermatitis herpetiformis and coeliac disease. Ann N Y Acad Sci 1983;420:325–34. [6] Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997;3:797–801. [7] Bruce SE, Bjarnason I, Peters TJ. Human jejunal transglutaminase: demonstration of activity, enzyme kinetics and substrate specificity with special relation to gliadin and coeliac disease. Clin Sci (Lond) 1985;68:573–9. [8] Korponay-Szabo I, Halttunen T, Szala Z, et al. In vivo targeting of intestinal and extraintestinal transglutaminase 2 by coeliac autoantibodies. Gut 2004:641–8. [9] Hopper AD, Cross SS, Hurlstone DP, et al. Pre-endoscopy serological testing for coeliac disease: evaluation of a clinical decision tool. BMJ 2007;334:729. [10] Vitoria JC, Arrieta A, Ortiz L, Ayesta A. Antibodies to human tissue transglutaminase for the diagnosis of celiac disease. J Pediatr Gastroenterol Nutr 2001;33:349–50. [11] Hansson T, Dahlbom I, Hall J, et al. Antibody reactivity against human and guinea pig tissue transglutaminase in children with celiac disease. J Pediatr Gastroenterol Nutr 2000;30:379–84. [12] Wolters V, Vooijs-Moulaert AF, Burger H, et al. Human tissue transglutaminase enzyme linked immunosorbent assay outperforms both the guinea pig based tissue transglutaminase assay and anti-endomysium antibodies when screening for coeliac disease. Eur J Pediatr 2002;161:284–7. [13] Cataldo F, Marino V, Ventura A, et al. Prevalence and clinical features of selective immunoglobulin A deficiency in coeliac disease: an Italian multicentre study. Gut 1998;42:362–5. [14] Korponay-Szabo IR, Dahlbom I, Laurila K, et al. Elevation of IgG antibodies against tissue transglutaminase as a diagnostic tool for coeliac disease in selective IgA deficiency. Gut 2003;52:1567–71. [15] Van Meensel B, Hiele M, Hoffman I, et al. Diagnostic accuracy of ten secondgeneration (human) tissue transglutaminase antibody assays in celiac disease. Clin Chem 2004;50:1856–60. [16] Bazzigaluppi E, Lampasona V, Barera G, et al. Comparison of tissue transglutaminase-specific antibody assays with established antibody measurements for coeliac disease. J Autoimmun 1999;12:51–6. [17] Agardh D, Dahlbom I, Daniels T, et al. Autoantibodies against soluble and immobilized human recombinant tissue transglutaminase in children with celiac disease. J Pediatr Gastroenterol Nutr 2005;41:322–7. [18] Ljungberg UK, Jansson B, Niss U, Nilsson R, Sandberg BE, Nilsson B. The interaction between different domains of staphylococcal protein A and human polyclonal IgG, IgA, IgM and F(ab')2: separation of affinity from specificity. Mol Immunol 1993;30:1279–85.
76
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
[19] Nomellini JF, Duncan G, Dorocicz IR, Smit J. S-layer-mediated display of the immunoglobulin G-binding domain of streptococcal protein G on the surface of caulobacter crescentus: development of an immunoactive reagent. Appl Environ Microbiol 2007;73:3245–53. [20] Marsh M. Gluten, major histocompatibility complex and small intestine. Gastroenterology 1992;102:330–54. [21] Hansson T, Dahlbom I, Rogberg S, et al. Recombinant human tissue transglutaminase for diagnosis and follow-up of childhood coeliac disease. Pediatr Res 2002;51:700–5. [22] Dahlbom I, Olsson M, Forooz N, Sjoholm A, Truedsson L, Hansson T. Immunoglobulin G (IgG) anti-tissue transglutaminase antibodies used as markers for IgAdeficient celiac disease patients. Clin Diagn Lab Immunol 2005;12:254–8. [23] Stiehm ER, Fudenberg HH. Serum levels of immune globulins in health and disease: a survey. Pediatrics 1966;37:715–27. [24] Laurell CB. Electroimmunoassay. Scand J Clin Lab Invest Suppl 1972;124:21–37.
[25] Troncone R, Greco L, Paparo F, et al. Latent and potential coeliac disease. Acta Paediatr 1996;412:10–4. [26] Sblattero D, Berti I, Trevisiol C, et al. Human recombinant tissue transglutaminase ELISA: an innovative diagnostic assay for celiac disease. Am J Gastroenterol 2000;95:1253–7. [27] Seissler J, Boms S, Wohlrab U, et al. Antibodies to human recombinant tissue transglutaminase measured by radioligand assay: evidence for high diagnostic sensitivity for celiac disease. Horm Metab Res 1999;31:375–9. [28] Martin-Pagola A, Ortiz-Paranza L, Bilbao JR, et al. Two-year follow-up of antitransglutaminase autoantibodies among celiac children on gluten-free diet: comparison of IgG and IgA. Autoimmunity 2007;40:117–21. [29] Agardh D. Antibodies against synthetic deamidated gliadin peptides and tissue transglutaminase for the identification of childhood celiac disease. Clin Gastroenterol Hepatol 2007;5:1276–81.
Contents lists available at ScienceDirect
Clinica Chimica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n c h i m
Protein A and protein G ELISA for the detection of IgG autoantibodies against tissue transglutaminase in childhood celiac disease Ingrid Dahlbom a,b,⁎,1, Daniel Agardh c,1, Tony Hansson b,d a
Department of Genetics and Pathology, Uppsala University, Uppsala, Sweden Department of Women's and Children's Health, Uppsala University, Uppsala, Sweden Unit of Diabetes and Celiac Disease, University Hospital MAS, Department of Clinical Sciences/Pediatrics, Lund University, Malmö, Sweden d Rheumatology Unit, Department of Medicine at Karolinska Hospital, Karolinska Institutet, Stockholm, Sweden b c
A R T I C L E
I N F O
Article history: Received 1 February 2008 Received in revised form 8 May 2008 Accepted 8 May 2008 Available online 15 May 2008 Keywords: Celiac disease ELISA Protein A Protein G Tissue transglutaminase
A B S T R A C T Objectives: To investigate if the detection of celiac disease (CD) in children was improved by using alternative conjugates for assessment of tissue transglutaminase (tTG) autoantibodies. Methods: Serum samples from 108 biopsy confirmed CD children and 42 control subjects were investigated for the presence of autoantibodies with tTG coated microplates using protein A (PA), protein G (PG), anti-IgG, or anti-IgA as conjugates. Results: Of the 108 CD children, 86 (80%) were IgG-tTG positive, 91 (84%) were positive with the PAconjugate, 94 (87%) were positive with the PG-conjugate, and 103 (95%) were IgA-tTG positive. Among the 42 controls, 4 (10%) were IgG-tTG positive, 5 (12%) were positive with both the PA- and PG conjugates, whereas 3 (7%) were IgA-tTG positive. Compared with IgG-tTG the concordance was 93% for PA and 95% for PG, with a positive correlation between antibody levels (r = 0.967 and r = 0.975, p b 0.0001). All but one CD child were found positive by combining IgG-tTG and IgA-tTG detection. Conclusions: The sensitivity of IgG-tTG detection with ELISA increased by protein A or protein G conjugates, whereas the specificity was reduced as compared with anti-IgG conjugate. The combined measurement of IgA-tTG and IgG-tTG still seems to be the optimal procedure when screening children for CD. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Celiac disease (CD) is a small bowel enteropathy due to intolerance against gliadin and related prolamines in wheat, barley and rye [1]. The diagnosis is confirmed by a biopsy showing typical histopathological features characterized by mucosal villous atrophy of the small intestine and exclusion of gluten from the diet results in restitution of the mucosa and remission of the disease [2]. The clinical presentation has been suggested to be age dependent with more profound symptoms among small children [3]. However, during the last decades it has become evident that the majority of CD patients may present with minor or atypical signs of malabsorption and some patients experience no symptoms, despite a damaged intestinal mucosa [4]. CD is strongly associated with the presence of IgA endomysial autoantibodies (EMA) directed against tissue antigens localized in the extracellular matrix of the small intestine [5]. The enzyme tissue transglutaminase (tTG) has been identified as the main autoantigen of EMA [6] and the expression of tTG is increased in the mucosal lesion of
⁎ Corresponding author. Department of Women's and Children's Health, Pediatrics, SE 751 85 Uppsala University, Sweden. E-mail address: [email protected] (I. Dahlbom). 1 These authors contributed equally to this paper. 0009-8981/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.cca.2008.05.005
untreated patients with CD [7,8]. As the majority of IgA-sufficient patients have elevated levels of IgA-tTG these markers are frequently used for the identification and follow-up of patients with CD [9]. Using enzyme-linked immunosorbent assays (ELISA) for the detection of IgA-tTG the reported diagnostic sensitivity and specificity have been estimated at 86–100% and 95–100%, respectively [10–12]. Selective IgA deficiency is a CD associated disorder which occurs in approximately 1/500 of the general population, who will consequently not be detected by conventional IgA based tTG assays [13]. Whereas the sensitivity and specificity of IgG-tTG analyzed with ELISA in IgA-deficient CD patients has proven highly accurate reaching 99% [14], the sensitivity of IgG-tTG among IgA-sufficient individuals is reduced [15]. Conversely, IgG-tTG analyzed with radioligand binding assays (RBA) seems to display a high sensitivity (99%) and specificity (96%) for untreated CD independent of the total serum concentration of IgA [16]. The observed discrepancy for ELISA and RBA suggests that the binding of IgG-tTG might differ between the methods [17]. With the ELISA method, patient serum is brought into contact with immobilized antigen and tTG-specific IgG antibodies are thereafter detected with enzyme-linked anti-IgG conjugate. With the RBA method, serum autoantibodies react with radioactive antigen in solution and tTG-specific IgG antibodies are then separated from the solution by means of protein A (PA) or protein G (PG) coated beads. PA produced by
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
73
2. Materials and method 2.1. Patients This retrospective study encompassed 150 serum samples obtained from children consecutively admitted to the Department of Pediatrics, University Hospital MAS in Malmö with suspected CD for the investigation with intestinal biopsy during the years 2002–2006. All biopsies were examined and evaluated independently by two histopathologists. A total of 108 children (71 females, 37 males), including 3 children with IgA deficiency, had atrophic intestinal findings [20]. They were diagnosed with CD at median 5.3 years of age (range 1.2–19.0 years) according to the revised criteria of European Society of Pediatric Gastroenterology Hepatology and Nutrition (ESPGHAN) [2]. The remaining 42 children (21 females, 21 males) had normal biopsies at median 4.3 years of age (range 1.4–16.4 years) and were included as disease control subjects. Among the disease control subjects, four children had selective IgA deficiency and another child partial IgA deficiency, three had transient EMA, three were diagnosed with wheat or cow's milk protein allergy, one had Helicobacter pylori gastritis, one lipase deficiency, one type 1 diabetes and the remaining 32 had temporary gastrointestinal symptoms or were investigated because of failure to thrive. In addition, 20 of the untreated CD children (13 females, 7 males) were followed with gluten-free diet with serum samples taken after three and six months from diagnosis at median 3.9 years (range 1.3–13.9 years). The Ethics Committee of Medical Faculty, Lund University approved the study. 2.2. Protein A and protein G ELISA
Fig. 1. Receiver operator characteristic (ROC) curves for tissue transglutaminase (tTG) antibodies detected with ELISA using protein A (PA-tTG), protein G (PG-tTG), antihuman IgG (IgG-tTG), and anti-human IgA (IgA-tTG) as conjugates.
Staphylococcus aureus binds with strong affinity all subtypes of IgG except IgG3, and also some IgA and IgM [18], whereas streptococcal PG binds all human IgG subclasses except IgA and IgM [19]. It is therefore possible that the higher clinical performance of IgG-tTG assessed with RBA is due to particular antibody binding properties of PA or PG. Although the IgG-tTG assessed with RBA has proved to be highly accurate for CD screening, the method is based on the use of radioactivity and is therefore difficult to standardize for laboratories not frequently using the method. In contrast, the ELISA technique is handled by most of laboratories and more accessible to all standard laboratories. A clinical easily accessible method for detection of tTG antibodies irrespective of immunoglobulin isotype would overcome the possibility to miss potential CD patients who only have increased levels of IgG-tTG. Few efforts for optimizing current ELISA methods have been performed and the use of PA and PG has not yet been evaluated in the current IgG-tTG based ELISA system. In this study, we hypothesized that the use of PA or PG as conjugates might improve the detection of tTG antibodies also in the ELISA in a similar manner as what has been demonstrated for the RBA method. The aim of the present study was therefore to investigate if PA or PG conjugates in the ELISA format could benefit the detection of tTG antibodies compared to the use of traditional anti-IgA or anti-IgG conjugates.
For quantification of anti-tTG antibodies detected with PA or PG, microtiter plates and reagents included in the Celikey-IgG kit, except for the conjugate, were used. The detection of anti-tTG antibodies was made with horseradish peroxidase conjugated (HRP) recombinant PA, diluted 1:80,000, or HRP conjugated recombinant PG, diluted 1:6000, (Pierce Biotechnology, Rockford, IL, USA). The concentration of anti-tTG antibodies was expressed as arbitrary units (U/mL), calculated from a 6-point calibrator curve containing 0, 3, 7, 16, 40, and 100 U/mL IgG-tTG. The within and between Cv of the PA ELISA was b4.5% and b 3.5% respectively, with a total Cv of b 5.5%. The within and between Cv of the PA ELISA was b3.7% and b11.1% respectively, with a total Cv of b 11.7%. Cut-off values according to ROC curves were 4.0 U/ml for the PA ELISA and 5.0 U/ml for the PG ELISA, (Fig. 1). 2.3. IgA and IgG anti-tTG-ELISA Celikey based on human recombinant tTG was used for quantification of IgA antitTG and Celikey IgG was used for detection of IgG anti-tTG (Phadia, Freiburg, Germany). The relative amount of tTG antibodies was expressed as U/mL calculated from standard calibrator curves constructed to contain 0, 3, 7, 16, 40, and 100 U/mL of respective IgA and IgG anti-tTG [21]. The within and between Cv of Celikey was b 8.0% and b1.5% respectively, with a total Cv of b 8.1%. The within and between Cv of Celikey IgG was b5.7% and b3.2% respectively, with a total Cv of b5.7%. The cut-off levels chosen for the present study were 3.0 U/mL for IgA anti-tTG and 6.0 U/mL for IgG anti-tTG according to ROC curves [22], (Fig. 1). 2.4. Total IgA and EMA Total serum IgA was determined by turbidometry [23], and samples with IgA concentrations less than 0.07 g/L were further analyzed for establishment of IgAdeficiency (b 0.05 g/L) by rocket immunoelectrophoresis [24]. EMA was detected with fluorescein isothiocyanate conjugated goat anti-human IgA antibodies applied to tissue slides of primate esophagus (Euroimmun, Lübeck, Germany) and visualized by immunofluorescence [5]. Results were expressed as the highest dilution factor giving a positive fluorescence pattern in microscope and all sera manifesting fluorescence at a titer N1:10 were considered to be positive.
Table 1 Detection of anti-tissue transglutaminase (tTG) antibodies in children with untreated celiac disease and in disease controls Celiac disease (N =108) ELISA
Disease controls (N =42)
Positive
Sensitivity
Median value
Positive
Specificity
Median value
(n)
(95% CI)
(5–95th centile)
(n)
(95% CI)
(5–95th centile)
84.3% (76.0–90.6) 86.1% (79.2–92.7) 79.6% (70.8–86.8) 95.4% (90.2–98.9)
21 U/ml (1.3–106) 25 U/ml (1.5–240) 26 U/ml (1.8–104) 80 U/ml (3.7–1200)
5
88.6% (75.4–96.2) 88.6% (75.4–96.2) 90.5% (77.4–97.3) 92.9% (80.5–98.5)
1.5 U/ml (0.5–5) 2.1 U/ml (0.7–8) 1.6 U/ml (0.6–6) 0.4 U/ml (0.1–7)
PA-tTG
91
PG-tTG
93
IgG-tTG
86
IgA-tTG
103
N; total number of patients, 95% CI; 95% confidence interval.
5 4 3
74
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
CD children of whom 82 were detected with the IgG-tTG ELISA, 87 were IgA-tTG positive, and 86 had positive EMA titres. All 3 (2.8%) CD children with IgA deficiency were detected with PA, PG as well as with the IgG-tTG ELISA, whereas EMA and IgA-tTG were negative. Another 3 (2.8%) children were positive with the PG ELISA, IgA-tTG and EMA, whereas they were negative with the ELISAs using PA or anti-IgG as conjugates. An additional 12 children had IgA-tTG and EMA only. Of the remaining two children, one was positive for IgA-tTG, IgG-tTG, and EMA and the other child had EMA only. A total of 34/42 (81.0%) disease controls had normal levels of IgGtTG and IgA-tTG according to all ELISAs and were EMA negative as well. Five (11.9%) disease control children were detected with the PA and PG ELISAs of whom four were positive for IgG-tTG, three for EMA and one had also IgA-tTG. The only child detected with both PA and PG ELISA has IgA-deficient. Three (7.1%) additional disease control children were IgA-tTG positive and two of them were EMA positive. The remaining child had type 1 diabetes and was EMA negative at the time of biopsy, but had a previous history of fluctuating EMA. 3.2. The diagnostic performance of the tTG antibody assays The sensitivity for PA ELISA was 84.3% and for PG ELISA 87.0%, while the specificity was 88.6% for both assays (Table 1). Sensitivity of IgG-tTG and IgA-tTG was 79.6% and 95.4%, whereas the specificity was 90.5% and 92.9%, respectively. The concordance for a positive and negative test result was 92.7% (139/150) for PA and 94.7% (142/150) for PG when compared with the IgG-tTG ELISA and the tTG antibody levels correlated (r = 0.967 and r = 0.9759, respectively, p b 0.0001) (Fig. 2A and B). The correlation was even higher between PA-tTG and IgG-tTG (r = 0.976, p b 0.001) (Fig. 2C). The concordance was reduced to 82.0–84.0% (120– 127/150) when the three PA, PG and IgG-tTG assays were compared with the IgA-tTG ELISA, although a positive correlation between tTG antibody levels was still obtained (r = 0.647–0.697, p b 0.001). 3.3. The effect of gluten-free diet on tTG antibody levels
Fig. 2. Correlation between the levels of tissue transglutaminase (tTG) antibodies detected with ELISA using (A) anti-human IgG (IgG-tTG) and protein A (PA-tTG) as conjugates, (B) IgG-tTG and protein G (PG-tTG) as conjugates, and (C) PA-tTG and PGtTG as conjugates. Filled symbols represent celiac disease patients and unfilled symbols represent controls.
In 20 CD patients followed for 6 months during a gluten-free diet the tTG antibody levels decreased (p b 0.01) in all patients except for one child. Two children had elevated IgG-tTG levels only, and one of them was IgA-deficient. For this child, the IgG-tTG levels decreased during the gluten-free diet, although the antibody levels remained elevated after 6 months of treatment. A modest reduction in IgG-tTG levels was noted for most of the CD children already after 3 months, but the median IgG-tTG levels did not decrease significantly until 6 months (Fig. 3). At this point 13/20 (65%) and 14/20 (70%) were still positive with the PA and PG and 15/20 (75%) remained positive with the IgG-tTG ELISA.
2.5. Statistical methods Differences in tTG antibody levels between independent groups were calculated using the Mann–Whitney U-test. The Wilcoxon signed rank test was used to test significant change in autoantibody levels before and after gluten-free diet. Correlations between the assessed tTG antibody levels were evaluated using Spearman rank correlation (r). p-values b 0.05 were considered significant.
3. Results 3.1. tTG antibodies in untreated CD children and disease control subjects The area under the ROC curve was 0.92 (95%CI = 0.880–0.9636) for the PA ELISA, 0.937 (95%CI = 0.900–0.974) for the PG ELISA, 0.92 (95% CI = 0.878–0.962) for IgG-tTG and 0.969 (95%CI = 0.841–0.996) for IgAtTG (Fig. 1). At the respective cut-off estimated from the ROC curves, both the PA and PG ELISAs detected a total of 91/108 (85.8%) untreated
Fig. 3. Median tissue transglutaminase (tTG) antibody levels detected with ELISA using protein A (PA-tTG), protein G (PG-tTG), anti-human IgG (IgG-tTG), and anti-human IgA (IgA-tTG) as conjugates in 20 celiac disease children during a gluten-free diet.
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
In contrast, the IgA-tTG levels decreased rapidly within 3 months for most of the children (p b 0.001) of whom 16/20 (80%) children had normal IgA-tTG levels after 6 months of gluten-free diet. 4. Discussion In this study we evaluated the clinical performance of CD associated anti-tTG antibodies assessed with ELISA where the conventional polyclonal anti-human IgA and anti-human IgG conjugate had been substituted with PA or PG conjugates. We demonstrate that the levels of IgG-tTG were elevated in the majority of children with untreated CD, including those with IgA-deficiency, and that the use of PA or PG slightly increased the sensitivity of the ELISA as compared to the conventional anti-IgG conjugate. Moreover, our results indicate that the use of PA or PG as conjugates in the ELISA technique mainly detected IgG-tTG as the levels of tTG-specific antibodies highly correlated with the antibody levels obtained with the anti-IgG conjugate. However, the IgA-tTG ELISA was by far the most sensitive assay (95%) for the identification of untreated CD, missing only five of the included 108 CD children, of whom three were IgA-deficient. These three children and one additional child were detected with all the anti-tTG ELISAs measuring IgG, whereas one child with minor mucosal alterations was EMA positive only. On the other hand, a considerable fraction of the untreated CD children were not detected with any of the IgG-tTG ELISA, whereas five of the control subjects were detected with the ELISAs using PA or PG as conjugates. Two of these children were also EMA positive, in spite of a normal small intestine mucosa, and it is possible that these children have a latent manifestation with potential risk to develop clinical CD later in life [25]. Our findings are in accordance with previous studies carried out with the ELISA technique, showing that many untreated IgA-sufficient patients are negative for IgG-tTG [15,26]. This contrasts with the results obtained from studies using immunoprecipitation methods or radiobinding assays for the detection of IgG-tTG, which have shown the same diagnostic specificity as IgA-tTG [16,27]. However, with a recent comparison between IgA-tTG and IgG-tTG assessed with ELISA and RBA we observed that the choice of resin for immunoprecipitation with the RBA method appears to be critical for the detection of IgGtTG. No major difference in the prevalence of IgG-tTG as measured with the ELISA or RBA was observed when anti-IgG beads were used for immunoprecipitation, while the use of PA beads displayed a 99% concordance with IgA-tTG [17]. The higher prevalence of IgG-tTG observed with RBA using PA beads might thus depend on the immunoglobulin-binding properties of PA itself, which has a strong binding affinity for the Fc portion of human IgG, but also binds IgA and IgM [18]. One might therefore hypothesize that PA has the ability to detect both IgA-tTG and IgG-tTG captured by the antigen. Another hypothesis is that PA might bind IgG with a higher affinity than anti-IgG conjugates promoting the detection of lower amounts with PA. Moreover, PG binds all human IgG subclasses and proven an even higher binding affinity than PA [19] and might therefore increase the sensitivity of IgG-tTG detection even further. At least from our own observations, the sensitivity of IgG-tTG was increased using PG resins in the RBA method (unpublished results). Still, our results demonstrated that the advantages of using PA or PG for detection of IgG-tTG in the RBA method were not transferable to the ELISA technique. Even if these conjugates improved the sensitivity slightly as compared to the anti-IgG conjugate, the combined detection of IgA-tTG and IgG-tTG identified 99% of all untreated CD children including the CD children with IgA-deficiency. A possible explanation for this technical drawback could be the difference in binding of the antigen in the two systems. In a solid phase ELISA the microtiter wells are coated with a limited amount of tTG, which might cause a competition between antibodies directed against epitopes in close proximity favouring the binding of antibodies with the highest affinity for tTG. In contrast, RBA
75
uses an excess of PA linked to the beads with a higher capacity to bind antibodies irrespective of their affinity for tTG. In the present study, the decrease pattern of the tTG-specific antibodies during six months of gluten-free diet evinced no major differences between the IgG-tTG assays regardless of conjugate. Only few of the children regained normal levels of IgG-tTG. It has previously been reported that the time for normalization of the tTG antibody response may be as long as one year [28]. In this study, levels of IgAtTG decrease most dramatically of all assays within the first 3 months. This is in line with our previous observations showing a more rapidly decline in IgA-tTG than IgG-tTG levels, suggesting that the withdrawal of gluten is first reflected in the IgA antibody response [29]. In summary, our study showed only slightly improvement of the diagnostic sensitivity in the detection of IgG-tTG when PA or PG was used as conjugates instead of anti-IgG tTG, whereas the disease specificity was reduced as compared with the tTG ELISA using anti-IgG conjugate. The optimal procedure to detect untreated CD in paediatric patients and how they respond to gluten-free diet still seems to be the combined measurement of IgA-tTG and IgG-tTG. Acknowledgements The study was funded by Gillbergska stiftelsen, Uppsala, the Faculty of Medicine, Lund University, and the Skåne Council Foundation for Research and Development, University Hospital MAS in Malmö. All reagents and kits for this study were kindly provided by Phadia AB, Uppsala, Sweden. References [1] Trier JS. Celiac sprue. N Engl J Med 1991;325:1709–19. [2] Walker-Smith J, Guandalini S, Schmitz J, Shmerling D, Visakorpi J. Revised criteria for diagnosis of coeliac disease: Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition (ESPGAN). Arch Dis Child 1990;65:909–11. [3] Collin P, Kaukinen K, Maki M. Clinical features of celiac disease today. Dig Dis 1999;17:100–6. [4] Mäki M, Collin P. Coeliac disease. Lancet 1997;349:1755–9. [5] Chorzelski TP, Sulej J, Tchorzewska H, Jablonska S, Beutner EH, Kumar V. IgA class endomysium antibodies in dermatitis herpetiformis and coeliac disease. Ann N Y Acad Sci 1983;420:325–34. [6] Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997;3:797–801. [7] Bruce SE, Bjarnason I, Peters TJ. Human jejunal transglutaminase: demonstration of activity, enzyme kinetics and substrate specificity with special relation to gliadin and coeliac disease. Clin Sci (Lond) 1985;68:573–9. [8] Korponay-Szabo I, Halttunen T, Szala Z, et al. In vivo targeting of intestinal and extraintestinal transglutaminase 2 by coeliac autoantibodies. Gut 2004:641–8. [9] Hopper AD, Cross SS, Hurlstone DP, et al. Pre-endoscopy serological testing for coeliac disease: evaluation of a clinical decision tool. BMJ 2007;334:729. [10] Vitoria JC, Arrieta A, Ortiz L, Ayesta A. Antibodies to human tissue transglutaminase for the diagnosis of celiac disease. J Pediatr Gastroenterol Nutr 2001;33:349–50. [11] Hansson T, Dahlbom I, Hall J, et al. Antibody reactivity against human and guinea pig tissue transglutaminase in children with celiac disease. J Pediatr Gastroenterol Nutr 2000;30:379–84. [12] Wolters V, Vooijs-Moulaert AF, Burger H, et al. Human tissue transglutaminase enzyme linked immunosorbent assay outperforms both the guinea pig based tissue transglutaminase assay and anti-endomysium antibodies when screening for coeliac disease. Eur J Pediatr 2002;161:284–7. [13] Cataldo F, Marino V, Ventura A, et al. Prevalence and clinical features of selective immunoglobulin A deficiency in coeliac disease: an Italian multicentre study. Gut 1998;42:362–5. [14] Korponay-Szabo IR, Dahlbom I, Laurila K, et al. Elevation of IgG antibodies against tissue transglutaminase as a diagnostic tool for coeliac disease in selective IgA deficiency. Gut 2003;52:1567–71. [15] Van Meensel B, Hiele M, Hoffman I, et al. Diagnostic accuracy of ten secondgeneration (human) tissue transglutaminase antibody assays in celiac disease. Clin Chem 2004;50:1856–60. [16] Bazzigaluppi E, Lampasona V, Barera G, et al. Comparison of tissue transglutaminase-specific antibody assays with established antibody measurements for coeliac disease. J Autoimmun 1999;12:51–6. [17] Agardh D, Dahlbom I, Daniels T, et al. Autoantibodies against soluble and immobilized human recombinant tissue transglutaminase in children with celiac disease. J Pediatr Gastroenterol Nutr 2005;41:322–7. [18] Ljungberg UK, Jansson B, Niss U, Nilsson R, Sandberg BE, Nilsson B. The interaction between different domains of staphylococcal protein A and human polyclonal IgG, IgA, IgM and F(ab')2: separation of affinity from specificity. Mol Immunol 1993;30:1279–85.
76
I. Dahlbom et al. / Clinica Chimica Acta 395 (2008) 72–76
[19] Nomellini JF, Duncan G, Dorocicz IR, Smit J. S-layer-mediated display of the immunoglobulin G-binding domain of streptococcal protein G on the surface of caulobacter crescentus: development of an immunoactive reagent. Appl Environ Microbiol 2007;73:3245–53. [20] Marsh M. Gluten, major histocompatibility complex and small intestine. Gastroenterology 1992;102:330–54. [21] Hansson T, Dahlbom I, Rogberg S, et al. Recombinant human tissue transglutaminase for diagnosis and follow-up of childhood coeliac disease. Pediatr Res 2002;51:700–5. [22] Dahlbom I, Olsson M, Forooz N, Sjoholm A, Truedsson L, Hansson T. Immunoglobulin G (IgG) anti-tissue transglutaminase antibodies used as markers for IgAdeficient celiac disease patients. Clin Diagn Lab Immunol 2005;12:254–8. [23] Stiehm ER, Fudenberg HH. Serum levels of immune globulins in health and disease: a survey. Pediatrics 1966;37:715–27. [24] Laurell CB. Electroimmunoassay. Scand J Clin Lab Invest Suppl 1972;124:21–37.
[25] Troncone R, Greco L, Paparo F, et al. Latent and potential coeliac disease. Acta Paediatr 1996;412:10–4. [26] Sblattero D, Berti I, Trevisiol C, et al. Human recombinant tissue transglutaminase ELISA: an innovative diagnostic assay for celiac disease. Am J Gastroenterol 2000;95:1253–7. [27] Seissler J, Boms S, Wohlrab U, et al. Antibodies to human recombinant tissue transglutaminase measured by radioligand assay: evidence for high diagnostic sensitivity for celiac disease. Horm Metab Res 1999;31:375–9. [28] Martin-Pagola A, Ortiz-Paranza L, Bilbao JR, et al. Two-year follow-up of antitransglutaminase autoantibodies among celiac children on gluten-free diet: comparison of IgG and IgA. Autoimmunity 2007;40:117–21. [29] Agardh D. Antibodies against synthetic deamidated gliadin peptides and tissue transglutaminase for the identification of childhood celiac disease. Clin Gastroenterol Hepatol 2007;5:1276–81.