Islet Cell Autoantibodies in Cord Blood from Children with Blood Group Incompatibility or Hyperbilirubinemia

Transcription

1 Autoimmunity, 2003 Vol. 36 (2), pp Islet Cell Autoantibodies in Cord Blood from Children with Blood Group Incompatibility or Hyperbilirubinemia A. MARIA ELFVING a, BENGT A. LINDBERG b, *, M. LANDIN-OLSSON c, CHRISTINE S. HAMPE d,åke LERNMARK d and STEN-A. IVARSSON b a Department of Pediatrics, Lund University, Lund, Sweden; b Department of Pediatrics, Malmö University Hospital, Lund University, S , Malmö, Sweden; c Department of Medicine, Lund University, Lund, Sweden; d Department of Medicine, University of Washington, Seattle, WA, USA (Submitted 16 October 2002; Accepted with revisions 25 November 2002) Blood group incompatibility is a risk factor for type 1 diabetes. Our aim was to test the hypothesis that islet cell autoantibodies, as markers for beta cell autoimmunity, are increased in cord blood from newborns with a diagnosis of blood group incompatibility. Using the diagnosis register of the Malmö University Hospital we obtained cord blood from 151 children with ABO immunization, 311 children with hyperbilirubinemia and a control group of 320 other children born during the same time period. The cord blood samples were analyzed for islet cell antibodies (ICA) by indirect immunofluorescence and autoantibodies against the Islet Cell Antigen-2 (IA-2Ab) and the 65 kda isoform of glutamic acid decarboxylase (GAD65Ab) by standard radioligand binding assays. The prevalence of ICA was increased compared to controls (0.6%) not only in children with ABO immunization (4.0%, p ¼ 0:02), but also in newborn children with hyperbilirubinemia (4.2%, p ¼ 0:003). The prevalence of IA2Ab, but not of GAD65Ab, was increased in children with ABO immunization (3.3%) compared to the hyperbilirubinemia group without incompatibility (0.6%, p ¼ 0:04), or the controls (0.3%, p ¼ 0:02). Our findings that hyperbilirubinemia is associated with an increased prevalence of ICA, and blood group incompatibility with both ICA and IA-2, suggests that intra-uterine factors may be associated with islet cell autoimmunity. Keywords: Autoimmunity; GAD65 antibodies; IA-2 antibodies; Islet cell antibodies; Type 1 diabetes Abbreviations: GAD65Ab, antibodies to the 65 kda isoform of glutamate decarboxylase; IA-2Ab, antibodies to the cytoplasmic portion of ICA512; IA-2, islet cell antigen-2; ICA, islet cell antibodies detected by immunofluorescence on frozen sections INTRODUCTION Several studies suggest that intrauterine events are risk factors for type 1 diabetes. Children with the congenital rubella syndrome [1,2] or with intrauterine enterovirus infections are at increased risk. [3,4] There are also indications of increased risk among children born to mothers with eclampsia or pre-eclampsia. [5,6] Perinatal events including blood group incompatibility between mother and child, leading to immunization and jaundice are also associated with a higher risk for childhood-onset type 1 diabetes. [7,8] Such early immunological events may disturb the development of tolerance to autoantigens during fetal life. Mothers who had islet autoantibodies during pregnancy developed diabetes long after but passed the autoantibodies on to the fetus. [9] Dependent on the HLA-type of the offspring this may or may not pose a risk for diabetes. [10,11] On the other hand, children later developing type 1 diabetes between the ages of 2 15, but born to healthy mothers, were retrospectively shown to have had an increased prevalence of cord blood islet autoantibodies. [12] These studies suggested that gestational events may be associated with an increased risk for type 1 diabetes in the off-spring. Blood group incompatibility has been associated with an increased risk for type 1 diabetes in two separate epidemiological investigations, but immunological indications of this association has so far been missing. [7,8] We tested the hypothesis that cord *Corresponding author. Tel.: þ Fax: þ bengt.lindberg@pediatrik.mas.lu.se ISSN print/issn X online q 2003 Taylor & Francis Ltd DOI: /

2 112 A.M. ELFVING et al. blood sera from newborn children with blood group incompatibility would exhibit increased frequency of autoantibodies, to islet cell antigen-2 (IA-2Ab), the 65 kda isoform of glutamic acid decarboxylase (GAD65Ab) or islet cell antibodies (ICA). RESEARCH DESIGN AND METHODS Study Subjects Malmö, a Swedish city with a population of 250,000, is served by a single tertiary-level university hospital. The diagnosis register at this hospital was used to identify 153 children with ABO immunization, and 312 children with hyperbilirubinemia without blood group incompatibility, where also cord blood sera had been stored. These children had all been treated with phototherapy because of the diagnosis of hyperbilirubinemia. After excluding three children born to mothers with type 1 diabetes, the remaining 151 children (84 males, 67 females) with ABO immunization and 311 children (187 males, 124 females) with hyperbilirubinemia were subjected to the analysis of islet cell autoantibodies. Control Subjects The control group comprised 320 randomly selected cord blood sera (170 males, 150 females) from children born during the same time period as the study subjects. There was enough cord blood to analyze all three islet cell autoantibodies (IA-2Ab, GAD65Ab and ICA) in sera from 288 of the control children, but only enough sera for GAD65Ab and ICA analysis for the remaining 32 samples. All sera were coded and analyzed in triplicate. The study was approved by the Human Research Ethics Committee of the Faculty of Medicine, University of Lund. GAD65Ab GAD65Ab was analyzed using a radioligand assay described elsewhere, [13,14] with minor modifications, [15] including the use of a Beta-Plate reader (Wallac Instruments, Turku, Finland) to count the radioactivity directly in 96-well plates. A GAD65Ab index was calculated to measure antibody levels as described above, [16] using the WHO standard 97/550 as the positive control [16,17] and two normal sera as negative controls. The 320 control cord blood sera were used to define cutoffs at the 97.5 and 99th percentiles. IA-2Ab IA-2Ab was analyzed in a radioligand binding assay similar to the one described for GAD65Ab, [18] using ICA512 cdna [19] kindly provided by George Eisenbarth of the University of Colorado Health Sciences Center (Denver, CO) to prepare the 35 S-methionine-labeled IA-2. The IA-2 index was calculated as described elsewhere. [14,20] The 288 control cord blood sera available for the IA-2Ab analysis were used to define the 97.5 and 99th percentiles. ICA ICA was determined in a two-color indirect immunofluorescence assay performed on sections of frozen human pancreas, as described previously. [21,22] Levels of ICA are expressed in Juvenile Diabetes Foundation Units (JDF-U), using the world reference standard curve based on the international JDF reference sera sample. [23] The Malmö laboratory participated in the 13th Immunology of Diabetes workshop standardization. [24] The children in the control group were used to identify the cut-off for the upper level of normal. Neither the 97.5 nor the 99th percentiles proved applicable, since only two out of 320 control children (0.6%) had values above zero. Statistical Analysis Comparisons between groups were done with Fisher s Exact Test (two-tailed). Values of p, 0:05 were considered significant. RESULTS ABO Immunization In children with ABO immunization, there was a significantly elevated prevalence of IA-2Ab as compared with control children (Table I). The prevalence of IA-2Ab was 10.6% (16/151), compared to 1.4% (4/288) for the control group ð p ¼ 0:0001Þ; when using the 97.5th percentile as the cutoff level. At the more stringent 99th percentile, the significance remained, since the prevalence of IA-2Ab in the group of children with ABO immunization was 3.3% (5/151), compared to 0.3% (1/288) in the control group ð p ¼ 0:02Þ: There was also a significant difference for cord blood ICA in the ABO immunization group since the prevalence was 4.0% (6/151) compared to 0.6% (2/320) in the control group ð p ¼ 0:02Þ: The frequency of GAD65Ab was not different between the ABO immunized children and controls neither at the 97.5th, nor at the more stringent 99th percentile ( p ¼ n.s.). Hyperbilirubinemia without Blood Group Incompatibility At the 97.5th percentile, the prevalence of IA2-Ab in the group with hyperbilirubinemia was higher (7.1%) compared to the controls (1.4% p, 0:001) (Table I). There was no significant difference at the 99th percentile cut-off.

3 HYPERBILIRUBINEMIA AND ICA 113 TABLE I Prevalence of islet cell autoantibodies in ABO immunization, hyperbilirubinemia without blood group incompatibility and control group IA-2Ab GAD65Ab ICA Percentile 97.5th 99th 97.5th 99th NA Antibody index level for positivity (.) ABO immunization % positive 10.6% 3.3% 4.0% 2.0% 4.0% (N) (16/151)*** (5/151)* (6/151) (3/151) (6/151)* Hyperbilirubinemia % positive 7.1% 0.6% 1.9% 1.0% 4.2% (N) (22/311)*** (2/311) (6/311) (3/311) (13/311)*** Controls % positive 1.4% 0.3% 1.9% 0.6% 0.6% (N) (4/288) (1/288) (6/320) (2/320) (2/320) NA ¼ Not Applicable. Difference compared to controls: *p, 0:05; **p, 0:01; ***p, 0:001: GAD65Ab did not differ between the groups neither at the 97.5th, nor at the 99th percentile ( p ¼ n.s.). The prevalence of ICA was higher in the group of children with hyperbilirubinemia (4.2%) as compared to control children (0.6%, p ¼ 0:003). ABO Immunization Compared to Hyperbilirubinemia Patients with ABO immunization had a significantly higher prevalence of IA-2Ab at the 99th percentile, compared to the group with hyperbilirubinemia ð p ¼ 0:045Þ: For GAD65Ab and ICA, no difference was found between the groups. The results of autoantibody tests revealed no correlation to gender, nor were there any significant differences in antibody levels found between children born prematurely and those born at term. DISCUSSION Our major observation is that newborn children with ABO immunization or hyperbilirubinemia have an increased prevalence of IA-2Ab and ICA but not GAD65Ab. The increase of IA-2Ab in ABO immunization was observed whether the 97.5 or 99th percentile was used, while the difference was significant in hyperbilirubinemia only when using the 97.5th percentile. There are compelling reasons to analyze autoantibody levels at different cut offs. First, the GAD65Ab and IA-2Ab are not normally distributed. It was therefore, appropriate to define upper level of normal as percentile. Second, we have previously shown in competition experiments with human recombinant GAD65 that low level GAD65Ab are present in human serum. [25] The interpretation of our data indicate that ABO immunization or hyperbilirubinemia without blood group incompatibility may induce or be associated with islet cell autoimmunity. The subgroup with hyperbilirubinemia may include some subjects with undiagnosed ABO immunization or immunization against other blood groups. The mechanism of autoimmunity induction is not understood. It is possible that the cord blood autoantibodies may come from the mother since the fetus produce little IgG and the transport of IgG over the placenta barrier is an active process resulting in an IgG concentration in the newborn that is 150% of the mother s. [26] Although sera from the mother at the time of delivery were not available, our cord blood measurements would provide a sensitive measure of maternal autoantibodies formed at the end of pregnancy. These observations support the view that the mother-child pair has been sensitized to develop autoantibodies binding to islet cell autoantigens similar to what we have reported from cord blood in children who developed type 1 diabetes later in life. [12] It would have been of interest to contact those individuals to learn retrospectively whether mother, child, or both later developed type 1 diabetes. The Swedish Human Research Ethics Committee requires, however, that the identity of individuals (both mother and child) be deleted from the database of serum samples. Only gender and age are therefore available for the subjects in this study. It is, therefore, not possible to contact those individuals to learn retrospectively whether mother, child, or both, have developed type 1 diabetes. Prospective studies will, therefore, be required to determine whether ABO immunization and hyperbilirubinemia are associated not only with an increased risk of developing IA-2Ab and ICA, but also type 1 diabetes later in life. Epidemiological studies have shown an increased risk for developing type 1 diabetes for children with ABO-immunization. [7,8] The reason for such risk is not clear. Other events during the neonatal period has also been associated with an increased risk type 1 diabetes. [6] Premature termination of breastfeeding due to early separation of mother and child has been suggested, since several studies have indicated that breastfeeding of newborns decreases the risk for developing type 1 diabetes. [5,27,28] Immunological isolation in infancy may be another factor of importance, since the hygiene hypothesis postulates that early exposure to infection may be protective against type 1 diabetes. [29,30]

4 114 A.M. ELFVING et al. Phototherapy has been cited by previous research as a possible risk factor for type 1 diabetes. [8] In this study, however, the reason for the elevated prevalence of islet autoantibodies has been sought among prenatal events, since cord blood was obtained minutes after birth, so that no subsequent neonatal procedures can have had any impact on the samples. We have demonstrated in an earlier investigation [12] that children who develop type 1 diabetes before age 15 show an increased prevalence of islet autoantibodies in their cord blood. [31] This emphasizes the importance of early immunological events in the development of type 1 diabetes. Finally, it has been reported that Rh immunization is more severe in subjects with HLA DQB1*0201 alleles. [32] DQB1*0201 is important to diabetes risk and in prospective studies we plan to investigate whether IA- 2Ab formation in ABO incompatible children is associated with HLA DQB1*0302 as is the case in new onset type 1 diabetes children. [33] In conclusion, our study supports the hypothesis that ABO immunization and possibly hyperbilirubinemia may contribute to the risk for type 1 diabetes provided that the child has a susceptible HLA type. [10,11] Another possibility may be a direct effect on the islet b cells by anti-a or anti-b antibodies. The association between IA- 2Ab and AB0 immunization or hyperbilirubinemia suggest that gestational autoimmunity may in part explain the type 1 diabetes risk observed in previous epidemiological studies. Acknowledgements The technical assistance by Lisa P. Hammerle is greatly appreciated. This study was supported by grants from the Faculty of Medicine, University of Lund; the Health Services Administration, Malmö University Hospital; the Novo Nordisk Foundation; the Malmö Branch of the Swedish Diabetic Association; the Childhood Diabetic Fund; the Swedish Diabetes Association; the Sven Jerring Fund; Lions Club International, District 101-S; the Trygg- Hansa Research Fund; the Swedish Medical Research Council, Project and the National Institutes of Health (DK26190). References [1] Forrest, J.M., Menser, M.A. and Burgess, J.A. (1971) High frequency of diabetes mellitus in young adults with congenital rubella, Lancet ii, [2] Menser, M.A., Forrest, J.M. and Bransby, R.D. (1978) Rubella infection and diabetes mellitus, Lancet i, [3] Dahlquist, G., Ivarsson, S., Lindberg, B. and Forsgren, M. (1995) Maternal enteroviral infection during pregnancy as a risk factor for childhood IDDM, Diabetes 44, [4] Hyöty, H., et al. (1995) A prospective study of the role of coxsackie B and other enterovirus infections in the pathogenesis of IDDM. Childhood Diabetes in Finland (DiMe) Study, Diabetes 44, [5] Jones, M., Swerdlow, A., Gill, L. and Goldacre, M. (1998) Pre-natal and early life risk factors for childhood onset diabetes mellitus: a record linkage study, Int. J. Epidemiol. 27, [6] McKinney, P.A., et al. (1999) Perinatal and neonatal determinants of childhood type 1 diabetes. A case-control study in Yorkshire, U.K, Diab. Care 22, [7] Dahlquist, G. and Källen, B. (1992) Maternal-child blood group incompatibiity and other perinatal events increased the risk for early-onset type 1 (insulin-dependent) diabetes melllitus, Diabetologia 35, [8] Dahlquist, G.G., Patterson, C. and Soltesz, G. (1999) Perinatal risk factors for childhood type 1 diabetes in Europe. The EURODIAB Substudy 2 Study Group, Diab. Care 22, [9] Ivarsson, S.A., et al. (1997) Glutamate decarboxylase antibodies in non-diabetic pregnancies precedes insulin-dependent diabetes in the mother but not necessarily in the offspring, Autoimmunity 26, [10] Atkinson, M.A. and Maclaren, N.K. (1994) The pathogenesis of insulin-dependent diabetes mellitus, N. Engl. J. Med. 331, [11] Kockum, I., et al. (1999) Complex interaction between HLA DR and DQ in conferring risk for childhood type 1 diabetes, Eur. J. Immunogenet. 26, [12] Lindberg, B., et al. (1999) Islet autoantibodies in cord blood from children who developed type I (insulin-dependent) diabetes mellitus before 15 years of age [see comments], Diabetologia 42, [13] Grubin, C.E., et al. (1994) A novel radioligand binding assay to determine diagnostic accuracy of isoform-specific glutamic acid decarboxylase antibodies in childhood IDDM, Diabetologia 37, [14] Falorni, A., Örtqvist, E., Persson, B. and Lernmark, Å. (1995) Radioimmunoassays for glutamic acid decarboxylase (GAD65) and GAD65 autoantibodies using 35 Sor 3 H recombinant human ligands, J. Immunol. Methods 186, [15] Hampe, C.S., et al. (1999) Species-specific autoantibodies in type 1 diabetes, J. Clin. Endocrinol. Metab. 84, [16] Mire-Sluis, A.R., Das, R.G. and Lernmark, Å. (1999) The development of a World Health Organisation international standard for islet cell antibodies: the aims and design of an international collaborative study, Diab. Metab. Res. Rev. 15, [17] Mire-Sluis, A.R., Das, R.G. and Lernmark, Å. (2000) The World Health Organization International Collaborative Study for islet cell antibodies, Diabetologia 43, [18] Vandewalle, C.L., et al. (1997) Associations of GAD65- and IA-2- autoantibodies with genetic risk markers in new-onset IDDM patients and their siblings, Diab. Care 20, [19] Kawasaki, E., et al. (1997) Evaluation of islet cell antigen (ICA) 512/IA-2 autoantibody radioassays using overlapping ICA512/IA-2 constructs, J. Clin. Endocrinol. Metab. 82, [20] Verge, C.F., et al. (1998) Combined use of autoantibodies (IA-2 autoantibody, GAD autoantibody, insulin autoantibody, cytoplasmic islet cell antibodies) in type 1 diabetes: Combinatorial Islet Autoantibody Workshop, Diabetes 47, [21] Madsen, O.D., et al. (1986) A two-colour immunofluorescence test with a monoclonal human proinsulin antibody improves the assay for islet cell antibodies, Diabetologia 29, [22] Landin-Olsson, M., Sundkvist, G. and Lernmark, Å. (1987) Prolonged incubation in the two-colour immunofluorescence test increases the prevalence and titres of islet cell antibodies in type 1 (insulin-dependent) diabetes mellitus, Diabetologia 30, [23] Bonifacio, E., Lernmark, A. and Dawkins, R.L. (1988) Serum exchange and use of dilutions have improved precision of measurement of islet cell antibodies, J. Immunol. Methods 106, [24] Greenbaum, C.J., Palmer, J.P., Kuglin, B. and Kolb, H. (1992) Insulin autoantibodies measured by radioimmunoassay methodology are more related to insulin-dependent diabetes mellitus than those measured by enzyme-linked immunosorbent assay: results of the Fourth International Workshop on the Standardization of Insulin Autoantibody measurement, J. Clin. Endocrinol. Metab. 74, [25] Rolandsson, O., Hägg, E.M.N., Hallmans, G. and Lernmark, Å. (2001) Prediction of diabetes by screening with body mass index, oral glucose tolerance test (OGT) and islet cell autoantibodies in a regional population, J. Intern. Med. 249,

5 HYPERBILIRUBINEMIA AND ICA 115 [26] Soothill, J.F., Hayward, A.R. and Wood, C.B.S. (1981) Development of immunity mechanism, Paediatric Immunology (Blackwell, Oxford), pp [27] Borch-Johnsen, K., Mandrup-Poulsen, T. and Zachau-Christiansen, B.Z. (1984) Relation between breast-feeding and incidence rates of insulin-dependent diabetes mellitus, Lancet ii, [28] Blom, L., Dahlquist, G., Nystrom, L., Sandstrom, A. and Wall, S. (1989) The Swedish childhood diabetes study social and perinatal determinants for diabetes in childhood, Diabetologia 32, [29] Kolb, H. and Elliot, R.B. (1994) Increasing incidence of IDDM as a consequence of improved hygiene, Diabetologia 37, [30] Gibbon, C., Smith, T., Egger, P., Betts, P. and Phillips, D. (1997) Early infection and subsequent insulin dependent diabetes, Arch. Dis. Child. 77, [31] Otonkoski, T., Beattie, G.M., Mally, M.I., Ricordi, C. and Hayek, A. (1993) Nicotinamide is a potent inducer of endocrine differentiation in cultured human fetal pancreatic cells, J. Clin. Investig. 92, [32] Hilden, J.O., Gottvall, T. and Lindblom, B. (1995) HLA phenotypes and severe Rh(D) immunization, Tissue Antigens 46, [33] Graham, J., et al. (2002) Genetic effects on age-dependent onset and islet cell autoantibody markers in type 1 diabetes, Diabetes 51,

6

7

8 Copyright of Autoimmunity is the property of Taylor & Francis Ltd and its content may not be copied or ed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or articles for individual use.