Rapid Publication An ELISA for Antibodies to Recombinant Glutamic Acid Decarboxylase in IDDM HENRY J. DEAIZPURUA, LEONARD C. HARRISON, AND DAVID S. CRAM

To detect serum antibodies to GAD in subjects with IDDM, three recombinant mBGAD 67 peptides encompassing the full-length protein were used in an ELISA. In this study 7 of 9 (78%) preclinical IDDM subjects (ICA + first-degree relatives of a person with IDDM) and 6 of 13 (46%) recent-onset IDDM subjects, but no subjects with Graves' disease (n = 10) or scleroderma (n = 10), nor healthy nondiabetic control subjects (n= 10) had antibodies that reacted with one or more of the recombinant mBGAD peptides. We found no preferential reactivity with any recombinant peptide. Although only 3 preclinical subjects and 1 recent-onset subject had antibodies to all three mBGAD peptides, the results indicate that mBGAD 67 contains at least three B-cell autoepitopes. Compared with an immunoprecipitation assay of native human brain GAD, the ELISA detected 5 of 6 (83%) preclinical and 6 of 6 (100%) recent-onset IDDM subjects. The ELISA should facilitate screening to evaluate the role of autoimmunity to GAD in the development of IDDM. Diabetes 41:1182 - 87, 1992


DDM is an organ-specific autoimmune disease that culminates in the destruction of pancreatic islet p-cells and insulin deficiency (1). Antibodies to insulin (2), ICA (3), and an MT 64,000 antigen (4) have been detected in the sera of subjects in the preclinical and

From the Burnet Clinical Research Unit, the Walter and Eliza Hall Institute of Medical Research, Parkville, Australia. Address correspondence and reprint requests to Dr. Henry J. DeAizpurua, Burnet Clinical Research Unit, the Walter and Eliza Hall Institute, P.O. Royal Melbourne Hospital, Parkville 3050 Australia. Received for publication 20 April 1992 and accepted in revised form 28 May 1992. ELISA, enzyme-linked immunosorbent assay; IDDM, insulin-dependent diabetes mellitus; GAD, glutamic acid decarboxylase; mBGAD, mouse brain GAD; ICA, islet cell antibody; JDF U, Juvenile Diabetes Foundation unit; PCR, polymerase chain reaction; bp, base pair; GST, glutathione-S-transferase; PBS, phosphate-buffered saline; RT, room temperature; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.


clinical phases of IDDM. The Mr 64,000 antigen has been reported to be GAD (5), the enzyme that synthesizes the inhibitory neurotransmitter 7-aminobutyric acid. Three distinct isomeric forms of mammalian brain GAD with Mr 80,000 (6), 67,000 (GAD 67) (7-9), and 65,000 (GAD 65) (10) have been cloned and sequenced, in addition to GAD 65 from islets (9,11). Except for allelic variations, the GAD 65 and GAD 67 isoforms are each identical between brain and islets in different species, but share only 65% homology (9; D.S.C. et al., unpublished observations). We have shown that antibodies that precipitate native human brain GAD are found in as many as 80% of preclinical IDDM subjects but in only 40% of recentonset IDDM subjects (12), which suggests they may be an important early marker of IDDM. This assay of antiGAD antibodies by immunoprecipitation and enzymatic assay is relatively time-consuming and requires a supply of purified, enzymatically active native GAD (12,13). A rapid assay for antiGAD antibodies would facilitate screening and prospective studies to determine whether their presence predicts progression to IDDM. We have cloned and expressed three overlapping cDNA fragments of mBGAD 67, and we have developed an ELISA, with the recombinant proteins, that detects anti-GAD antibodies. RESEARCH DESIGN AND METHODS Experimental procedures were approved by the Human Ethics Committee of Royal Melbourne Hospital. Subjects with preclinical IDDM were asymptomatic, first-degree relatives of IDDM patients who had circulating ICA > 20 JDF U. Recent-onset IDDM subjects (mean ± 2 SD of results with control sera. We observed that 5 of 9 (55%) preclinical IDDM sera reacted significantly with mBGAD 12 or 34, whereas 4 of 9 (44%) reacted with mBGAD 56. A lower percentage of sera from the recentonset IDDM subjects reacted with each of the peptides, i.e., 2 of 13 (15%) with mBGAD 12, 4 of 13 (30%) with mBGAD 34, and 3 of 13 (23%) with mBGAD 56. None of 10 Graves' disease or 10 scleroderma sera reacted significantly with any of the mBGAD peptides (Fig. 2). The interassay precision of the ELISA, expressed as the coefficient of variation over three assays with each GAD peptide, ranged from 8-20% for IDDM sera. Results that were just above or below the cutoff were placed consistently in the three assays. The ELISA results, together with clinical and other laboratory data, are summarized in Table 1. Although 7 of



by immunoprecipitation of native human brain GAD enzymatic activity was high (Table 1). Thus, of the 10 sera (both preclinical and recent-onset) that were negative in the GAD enzymatic assay, 8 (80%) also were negative in the ELISA assay; whereas of the 12 sera that were positive in the GAD enzymatic assay, 11 (91%) were also positive in the ELISA assay. We found no apparent relationship between the presence of antiGAD antibodies and ICA, although the numbers in each group are too small to allow definite conclusions.

A 100 - • 0-80 060



d* o 040


•• *


• •

s *



Since Baekkeskov et al. (5) found that GAD could account for the previously uncharacterized Mr 64,000 100 r B •• islet protein recognized by antibodies in IDDM sera, GAD has been cloned successfully from brain and islets (6-10). At least two distinct genes code for two isotypes 080 of GAD are termed GAD 65 and GAD 67 (10). These two forms of GAD (which share - 6 5 % amino acid identity) in 060 o are found on different chromosomes (20), possess :. .. d* slightly different biochemical and cellular properties, and o 040 could possibly be recognized differentially by IDDM sera (21). We cloned and expressed mBGAD as three overt •• lapping peptides. The full-length sequence of this clone 020 represents the GAD 67 form of the enzyme and is - 9 6 % identical to human brain GAD 67 (8) and human islet GAD 67, recently cloned in our laboratory (K. Yamashita 100 et al., unpublished observations). Using the ELISA format, we found that the majority of 080 sera from preclinical IDDM subjects (78%) and a minority from recent-onset IDDM subjects (46%) contain antibodies that react with at least one epitope in mBGAD 67. 060 in o These frequencies agree with those found with the less Q convenient immunoprecipitation assay of GAD activity. O 0-40 The difference in frequency of antibodies between the preclinical and recent-onset IDDM subjects could, as discussed previously (12), reflect loss of antigenic drive 020 s • for antibody production concomitant with near total p-cell destruction. Alternatively, it could mean that antibodies to GAD are a marker of preclinical subjects at low risk for *y. ^ -a clinical disease. The small sample numbers and the cross-sectional nature of this study, designed to establish the ELISA, do not allow us to distinguish these Subject groups possibilities. However, the ELISA will facilitate longitudiFIG. 2. ELISA reactivity of preclinical and recent-onset IDDM sera with nal studies of preclinical subjects to answer these quesmBGAD 12 (A), 34 (5), and 56 (Q. The mean ± 2 SD of 10 healthy tions. controls (indicated by the horizontal line) was used as the lower limit of posltivlty. The three fragments of mBGAD were recognized by antibodies in both preclinical and recent-onset IDDM sera, indicating that at least three major (3-cell epitopes 9 (78%) preclinical IDDM and 6 of 13 (46%) recent-onset exist. Antibodies in other autoimmune diseases whose IDDM sera reacted with at least one of the mBGAD antigenic epitopes have been mapped display a similar peptides, only 3 of 9 (33%) and 1 of 13 (8%) preclinical polyclonal pattern of reactivity (22,23). Karlsen et al. (11) recently reported the cloning and and recent-onset IDDM sera, respectively, reacted with expression of the human islet GAD 65 isoform, which all three mBGAD fragments. None of the three GAD peptides was recognized preferentially. These findings shares - 6 5 % homology with rat brain GAD 67. Most of indicate that patterns of sera reactivity with recombinant the sequence differences are in the amino terminal mBGAD are heterogenous and that at least three major region. Using four IDDM sera, Kaufman et al. (21) reported no reactivity to the amino terminal region of rat epitopes exist in the GAD 67 isoform. brain GAD 65, whereas three of the four sera reacted Despite different substrates and techniques, the concordance between the ELISA results and those obtained with at least one of either the central or carboxy terminal i





•v- i

re -




regions. Our results indicate that the amino terminal region of GAD 67 and the conserved central and carboxy terminal regions are recognized by autoantibodies in IDDM sera. The use of recombinant proteins in an ELISA format theoretically could preclude detection of antibodies that react with conformation-dependent epitopes. However, comparison of the ELISA results with those from the immunoprecipitation of native human brain GAD (Table 1) shows that this is not the case; the ELISA is not only specific but at least as sensitive. An ELISA format incorporating separate smaller recombinant proteins, ideally from the islet isoforms of GAD, should allow the fine mapping of GAD epitopes. It then should be possible to ascertain whether differential reactivity with specific epitopes exists that might be clinically relevant, for example, in subdividing subjects operationally defined as having preclinical IDDM on the basis of ICA positivity. ACKNOWLEDGMENTS We thank Dr. Peter Colman and Mrs. Jan Caudwell for ICA assays, Dr. Brian Tait for HLA typing, Mrs. Jacqueline Slattery and Ms. Louise Barnett for excellent technical assistance, and Mrs. Margaret Thompson for secretarial assistance. H.J.D. and D.S.C. are supported by Diabetes Australia and Amrad Kaneka Autoimmune Disease Pty. Ltd. L.C.H is a senior principal research fellow of the National Health and Medical Research Council of Australia. REFERENCES 1. Harrison LC, Campbell IL, Colman PG, Chosich N, Kay TWK, Tait B, Bartholomeusz RK, DeAizpurua H, Joseph JL, Chu S, Kielczynski WE: Type 1 diabetes: immunology and immunotherapy. Adv Endocrinol Metab 1:35-94, 1990 2. Colman PG, Nayak RC, Campbell IL, Eisenbarth GS: Binding of cytoplasmic islet cell antibodies is blocked by human pancreatic glycolipid extracts. Diabetes 37:645-52, 1988 3. Palmer JP, Asplin CM, Clemons P, Lyen K, Tatpati 0, Raghu PK, Paquette TL: Insulin antibodies in insulin-dependent diabetes before insulin treatment. Science 222:1337-39, 1983 4. Baekkeskov S, Nielsen JH, Marner B, Bilde T, Ludevigsson J, Lemmark A: Autoantibodies in newly diagnosed diabetic children immunoprecipitate human pancreatic islet cell proteins. Nature (Lond) 298:167-69, 1982 5. Baekkeskov S, Aanstoot H-J, Christgau S, Reetz A, Solimena A, Cascalho M, Folli F, Richter-Olesen H, De Camilli P: Identification of the 64K autoantigen in insulin-dependent diabetes as the GAGAsynthesizing enzyme glutamic acid decarboxylase. Nature (Lond)


347:151-56, 1990 6. Huang WM, Fourquest LR, Wu E, Wu JY: Molecular cloning and amino acid sequence of brain L-glutamate decarboxylase. Proc Natl Acad Sci USA 87:8491-95, 1990 7. Wyborski RJ, Bond RW, Gottlieb Dl: Characterization of a cDNA coding for rat glutamic acid decarboxylase. Mol Brain Res 8:193— 98, 1990 8. Cram DS, Barnett LD, Joseph JL, Harrison LC: Cloning and partial nucleotide sequence of human glutamic acid decarboxylase (GAD) cDNA from brain and pancreatic islets. Biochem Biophys Res Comm 176:1239-44, 1991 9. Michelsen BK, Petersen JS, Boel S, Moldrup A, Dryberg T, Madsen O: Cloning, characterization and autoimmune recognition of rat islet glutamic acid decarboxylase in insulin-dependent diabetes mellitus. Proc Natl Acad Sci USA 88:8854-58, 1991 10. Erlander MG, Tillakaratne NJK, Feldblum S, Patel N, Tobin AJ: Two genes encode distinct glutamate decarboxylases. Neuron 7:91100,1991 11. Karlsen AE, Hagopian WA, Grubin CE, Dube S, Disteche CM, Adler DA, Barmeier H, Mathewes S, Grant DJ, Foster D, Lernmark A: Cloning and primary structure of a human islet isoform of glutamic acid decarboxylase from chromosome 10. Proc Natl Acad Sci USA 88:8337-41, 1991 12. DeAizpurua HJ, Wilson Y, Harrison LC: Glutamic acid decarboxylase (GAD) autoantibodies in islet cell antibody (ICA) positive relatives of insulin-dependent diabetics. Proc Natl Acad Sci USA. In press 13. Rowley MJ, Mackay IR, Chen Q-Y, Knowles WJ, Zimmet PZ: Antibodies to glutamic acid decarboxylase discriminate major types of diabetes mellitus. Diabetes 41:548-51, 1992 14. Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning. A Laboratory Manual. Vol. 1. Cold Spring Harbor, NY, Cold Spring Harbor, 1989 15. Sanger F, Nicklen S, Coulson AR: DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74:5463-66, 1977 16. Smith DB, Johnson KS: Single step purification of polypeptides expressed in Escherichia coli as fusions with glutathione-S-transferase. Gene 67:31-36, 1988 17. Hochuli E, Babbwarth W, Dobeli H, Gentz R, Stuber D: Genetic approach to facilitate purification of recombinant proteins with a novel metal chelate adsorbent. Biotechnology 6:1321-25, 1988 18. Wu J-Y, Matsuda T, Roberts E: Purification and characterization of glutamate decarboxylase from mouse brain. J Biochem 248:302934, 1973 19. Albers RW, Brady RO: The distribution of glutamic decarboxylase in the nervous system of the Rhesus monkey. J Biol Chem 234:92628, 1958 20. Bu DF, Erlander MG, Hitz BC, Tillakaratne NJK, Kaufman DL, Wagner-McPherson CB, Evans GA, Tobin AJ: Two human glutamate decarboxylases, 65-kDa GAD and 67-kDaGAD, are each encoded by a single gene. Proc Natl Acad Sci USA 89:2115-19, 1992 21. Kaufman DL, Erlander MG, Clare-Salzer M, Atkinson MA, Maclaren NK, Tobin AJ: Autoimmunity to two forms of glutamate decarboxylase in insulin-dependent diabetes mellitus. J Clin Invest89:283-92, 1992 22. Cram DS, Fisicaro N, Coppel RL, Whittingham S, Harrison LC: Mapping of multiple B cell epitopes on the 70-kDa autoantigen of the U1 ribonucleoprotein complex. J Immunol 145:630-35, 1990 23. McNeilage LJ, Macmillan EM, Whittingham SF: Mapping of epitopes on the La(SS-B) autoantigen of primary Sjogren's syndrome: identification of a cross-reactive epitope. J Immunol 145:3829-35,1990


An ELISA for antibodies to recombinant glutamic acid decarboxylase in IDDM.

To detect serum antibodies to GAD in subjects with IDDM, three recombinant mBGAD 67 peptides encompassing the full-length protein were used in an ELIS...
1MB Sizes 0 Downloads 0 Views