| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Studies |
Department of Medicine, King’s College School of Medicine, London, United Kingdom SE5 9PJ; Central Laboratory of The Netherlands Red Cross Blood Transfusion Service (A.V.), Amsterdam, The Netherlands; the Division of Endocrinology, Vanderbilt University, Department of Veterans Affairs (A.C.P.), Nashville, Tennessee 37232-6303, and the Department of Internal Medicine III, University of Leipzig (W.A.S.), Leipzig, Germany
Address all correspondence and requests for reprints to: Dr. J. P. Banga, Department of Medicine, King’s College School of Medicine, Bessemer Road, London, United Kingdom SE5 9PJ.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results indicate that antibodies to GAD65 present in nondiabetic patients with multiple autoantibodies to endocrine organs show similarities to those in IDDM (by targeting IDDM-E1 and IDDM-E2 regions of GAD65) as well as subtle differences in epitope recognition (such as binding to denatured and reduced GAD65 and by protein footprinting). Thus, the GAD65 epitopes recognized by autoantibodies in different autoimmune diseases may overlap and be more heterogeneous than previously recognized.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Differences in antibody recognition of GAD between different patient groups have been demonstrated with a variety of approaches, including denaturation in SDS and Western blotting (9, 10, 11, 12, 13), deletion mutants of GAD65 (8, 14), and chimeric proteins of GAD65 and GAD67 (15). One of these studies has led to the recognition of two distinct conformation-dependent epitopes, termed IDDM-E1 (amino acids 240–435) and IDDM-E2 (amino acids 451–570), recognized by autoantibodies in IDDM patients (16). Although the GAD65 autoantibodies in IDDM are usually low titer, patients with SMS and polyendocrine autoimmune disease are characterized by very high levels of antibodies to GAD without necessarily the development of diabetes (16, 17, 18). There are other differences in the spectrum of autoantibodies to GAD in IDDM and other autoimmune disease patients. Serum antibodies in SMS and polyendocrine autoimmune disease recognize denatured GAD in Western blots, whereas the autoantibodies in IDDM are dependent upon the conformation of the molecule (reviewed in 1 . Like autoantibodies to a number of other enzyme autoantigens in organ-specific autoimmune diseases (19), all patients with SMS and a proportion of antibodies in polyendocrine autoimmune disease patients inhibit the enzymatic activity of GAD, suggesting that the epitopes recognized are at or near the catalytic site (20). Although a large number of studies have focused on the recognition of GAD epitopes in IDDM and SMS patients (10, 11, 12), there has been little progress on identifying epitope specificity in other autoimmune endocrinopathies (20, 21, 22).
Recently, Richter and colleagues reported the first isolation of human IgG monoclonal antibodies (hmAbs) to GAD65 from patients with new-onset IDDM as well as patients with IDDM and another organ-specific autoimmune disease, Graves’ disease (23, 24). The IgG hmAbs, designated MICA1 to MICA10, all have islet cell activity and are specific for GAD65. Using a combination of approaches, including GAD65 deletion mutants, GAD65/GAD67 chimeras, and blocking studies, five distinct conformational epitopes were revealed, which included the middle and C-terminal regions together with the large region encompassed by residues 39–585 of GAD65 (23, 24). Furthermore, one of the hmAbs, MICA2, which showed binding to GAD65 by Western blot analysis was dependent on binding to an SDS-resistant, miniconformational epitope residing in amino acids 506–531 (25). Other human IgG hmAbs to GAD65 from an established IDDM patient have been described, although the precise epitopes remain to be characterized (26).
The exact determination of the autoantigenic epitopes on GAD65 in autoimmune diseases such as IDDM (and SMS) may allow the disease-specific epitopes to be elucidated. Furthermore, antibodies to different regions of autoantigens may be relevant in determining the spectrum of peptides available to autoreactive T cells (27) that play a pivotal role in IDDM. In this study, we describe IgG antibodies to GAD65 from a nondiabetic patient with multiple autoantibodies to endocrine organs and Graves’ disease (28) who has autoantibodies with similar specificities, but also differences, as those present in IDDM patients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Venous blood was obtained with informed consent from a nondiabetic, 28-yr-old man (PV) who had been treated with radioiodine for Graves’ disease complicated by ophthalmopathy and pretibial myxoedema. He was positive for multiple autoantibodies, including TSH receptor (TSH binding inhibitory activity), thyroid peroxidase, gastric parietal cells, and islet cell autoantibody (ICA), and was strongly positive for GAD65 and GAD67 autoantibodies.
Peripheral blood lymphocytes were isolated from 50 mL blood by density gradient centrifugation on Ficoll-Hypaque. The mononuclear cells were immortalized with Epstein-Barr virus (EBV) supernatant (23, 29). Briefly, cells were infected by overnight culture with infectious EBV supernatant from the B-95 marmoset cell line, and the IgG-secreting B cells were isolated by magnetic separation on Dynal beads (Wirral, Merseyside, United Kingdom) coated with antihuman IgG. On this occasion, 50 mL blood yielded 0.5 x 106 IgG-positive B cells. These were plated on a feeder monolayer of irradiated mononuclear cells in complete medium (RPMI 1640 containing 15% FCS, 2 mmol/L glutamine, 2 mmol/L oxaloacetate, and 10 IU insulin) in 96-well plates, with 4 plates each at concentrations of 250 or 500 cells/well (cpw) and with two plates at 1000 cpw. Two or 3 weeks later, the wells containing acidified medium were screened for antibodies to GAD65 by radioligand binding assay (see below). Pools of supernatants from 8 wells were initially screened, which led to the identification of 9 positive pools; screening of individual wells from the positive pools identified 27 positive wells. After further culture and expansion for 4–6 weeks of the individual wells, 7 wells continued to be positive for antibody; these were expanded into 24-well Costar plates (Cambridge, MA) and subsequently into T25 flasks. Three wells, designated b35, b78, and b96, continued to be positive for antibody to GAD65, and cells from the wells were cloned and expanded by limiting dilution (29).
Immunoprecipitation with in vitro translated GAD65 and GAD65/67 chimeras
Screening for antibody to GAD65 was performed by radioligand binding assay using human (h) GAD65 complementary DNA (cDNA) cloned in the vector pB1882 (a gift from Dr. Thomas Dyrberg) (30). Samples that precipitated mean ± 3 SD greater radioactivity than the counts per min in negative controls (culture medium alone) were taken as positive for subsequent expansion and cloning of the cell lines.
GAD65/GAD67 chimeric proteins
The following nomenclature for GAD chimera was used: GAD65 or GAD67 (amino acid number of that GAD species present in the GAD chimera)/GAD65 or GAD67 (amino acid number of that GAD species present in the GAD chimera). The GAD65 protein has 585 amino acids and the GAD67 protein has 594 amino acids. Chimeric GAD proteins were prepared from chimeric GAD cDNAs as previously described (15). Three chimeric proteins were used: 1) GAD65/67/67 = GAD65-(1–195)/GAD67-(205–594); 2) GAD67/65/67 = GAD67-(1–230)/GAD65-(221–442)/GAD67-(452–594) (this contains IDDM-E1); and 3) GAD67/67/65 = GAD67-(1–441)/GAD65-(443–585) (this contains IDDM-E2). For immunoprecipitations, tissue culture supernate diluted 1:5,000 to 1:10,000 or control medium was used, and immunoprecipitations were performed as previously described (15).
Recombinant insect cell-expressed GAD and Western blotting
Recombinant hGAD65 and hGAD67 were expressed in insect cells and used to determine binding of the IgG mAbs by Western blotting. Recombinant hGAD65 baculovirus supernatant was a gift from Dr. Thomas Dyrberg (31). hGAD67 cDNA was cloned by PCR from total human islet cDNA in our laboratory, sequenced, and subcloned into pVL1393 transfer vector. A plasmid with cDNA in the correct orientation was used to obtain recombinant baculovirus by cotransfection with linearized BaculoGold baculovirus DNA (PharMingen, San Diego, CA) using lipofectin (Banga, J. P., and W. A. Scherbaum, unpublished observations). The expression of hGAD67 was carried out by infecting Sf9 monolayer cultures in TC100 medium containing 10% heat-inactivated FCS with 5–10 plaque-forming units/cell of cloned virus; sodium glutamate (2 mmol) was added every day (31), and the cells were harvested 72 h postinfection, pelleted, snap-frozen in liquid nitrogen, and stored at -70 C. Cell extracts were used in Western blotting after SDS-PAGE (11). After transfer to nitrocellulose membrane, the filters were blocked with 5% fat-free milk powder in Tris-borate-saline buffer containing 0.05% Tween-20 (TBST/milk) and blotted with antibody supernatants (1:100 in TBST/milk) for 2 h at room temperature; after washing, the filters were probed with alkaline phosphatase-labeled antihuman IgG in TBST/milk for 2 h at room temperature. The washed filters were developed with tetrazolium blue.
Protein footprinting
In vitro translated [35S]methionine-labeled GAD65 complexed with 100 µl undiluted supernatant containing human IgG antibody or with 10 µL undiluted or diluted serum (to give equivalent amounts of labeled GAD65 in the immunoprecipitate) and protein A-Sepharose were treated with different concentrations of trypsin at 1, 0.5, and 0.1 mg/mL/sample. After incubation at 37 C for 20 min, the samples were washed, resuspended in 30 µL loading buffer and analyzed in 15% acrylamide gels. The gels were processed for fluorography and autoradiographed for 2 weeks.
ICA
Undiluted serum or culture supernatant was screened for ICA by indirect immunofluorescence on unfixed sections of blood group O human pancreas (32).
Determination of IgG subclass
Total IgG and the IgG subclasses in the tissue culture supernatant were determined by nephelometry (33). When the concentration of IgG was below the sensitivity of nephelometric measurement (<10 ng/mL), the more sensitive method of enzyme-linked immunosorbent assay was used (34).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both b78 and b96 IgG were positive by indirect immunofluorescence for
islet cell staining of pancreatic sections (Fig. 1
, a
and b). The antibodies did not bind sections of stomach, thyroid,
adrenal gland, or liver tissue (not shown). All three antibodies were
IgG1, where an antibody level of 15 µg/mL was present in 14-day
cultures (not shown).
|
). The regions on GAD65 recognized were
investigated by immunoprecipitation with a series of chimeras of
hGAD65/rat GAD67 (15). The b35 mAb immunoprecipitated
GAD67-(1–230)/GAD65-(221–442)/GAD67-(452–594), which contains the
IDDM-E1 region (15) (Fig. 2b). Interestingly, both b78 and b96 bound
GAD67-(1–451)/GAD65-(443–585), which contains the IDDM-E2 region, and
did not react with the IDDM-E1 region (Fig. 2b). None of the mAbs
targeted the amino-terminus of GAD65.
|
). This indicates that although both
b78 and b96 target the large IDDM-E2 region of GAD65, the epitopes
targeted by the antibodies differ. Neither of the two IgG showed
binding to hGAD67 under these conditions (Fig. 3).
|
). Thus, anti-GAD65 antibodies in patients with
SMS and multiple autoantibodies lead to the protection of three high
molecular mass bands in the 42- to 50-kDa region (arrowed in
Fig. 4a) that were not protected in four of the five antibodies present
in IDDM sera. Antibodies in one IDDM patient (patient 4, Fig. 4a)
showed a similar pattern as the SMS and multiple
autoantibody-containing sera by protecting the three large fragments,
probably due to the presence of antibodies to GAD67, which are
occasionally present in IDDM patients.
|
). mAb b78 (Fig. 4b, lane 1) showed only one of the
three large protected fragments, whereas b96 and b35 mAb did not
protect any of these three large fragments (Fig. 4b, lanes 2 and 3,
arrowed). Furthermore, the footprint patterns of b78 and b96
were strikingly different; almost all of the smaller (below the three
large fragments) protected footprint bands observed in the PV serum
(Fig. 4b, arrowed) were present in b35, b78, or b96, showing
that each of these antibodies contributes to the total number of bands
observed in the PV serum footprint (Fig. 4b). Interestingly, the
footprint pattern of the murine mAb to GAD65, GAD6, was similar to that
of the b78 hmAb (Fig. 4b). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The IgG antibodies b35, b78, and b96 to GAD65 that were generated
belong to the 
1 subclass of IgG, the major subclass of ICA present
in IDDM patients (35). All three antibodies were specific for GAD65,
and none showed binding to GAD67. Examination of the regions on GAD5
recognized by the hmAbs showed that b35 recognized the central
(IDDM-E1) region, whereas b78 and b96 recognized the carboxyl (IDDM-E2)
region of the molecule. Thus, the repertoire of the three hmAbs to
GAD65 generated from a patient with polyendocrine autoimmune disease
mirrors the repertoire present in IDDM patients (8, 15). Differences in
the antibodies directed to the IDDM-E2 region of GAD65 were apparent;
b78, but not b96, recognized denatured and reduced GAD65 in Western
blot analysis, and they demonstrated different protein footprinting
patterns. It is interesting that b78 shows a similar footprint pattern
as the murine mAb, GAD6, which recognizes a linear epitope in region
529–589 of GAD65 (8); recent data also indicate that the GAD6 epitope
is in IDDM-E2, but the epitope is not identical to b78 (Powers, A. C.,
and K. Daw, unpublished).
Although sera from polyendocrine autoimmune disease and SMS patients also exhibit autoantibodies to GAD, their reactivities are different. For example, serum from SMS individuals has a selected pattern of immunofluorescence staining on pancreas sections, where the ICA activity can be abolished by incubation with GAD proteins; in contrast, serum from new-onset IDDM patients exhibit additional ICA-reactive autoantibodies (36, 37). In comparing the serum anti-GAD65 antibodies in IDDM and SMS patients, it has become clear that SMS sera have higher titers of antibody to GAD65 (18) than IDDM sera and contain antibodies that recognize denatured and reduced GAD65 in Western blots as well as those that react with GAD67 (10, 12, 13). Epitope mapping studies with SMS sera have localized linear epitopes to residues 1–8 (12), 354–368, and 390–402 (13) and to two carboxyl-terminal regions containing residues 475–484 and 571–585 (10). However, recently some antibodies to GAD65 in SMS patients have also been shown to be conformation dependent (10, 38) and thus are similar in some respects to those present in IDDM. Studies of the regions on GAD65 recognized by anti-GAD65 antibodies in polyendocrine autoimmune disease have not been analyzed as extensively as those in SMS and IDDM sera. In one study, antibodies to GAD65 in seven patients with polyendocrine autoimmune disease were reported to inhibit the enzymatic activity of the GAD enzyme, in contrast to a proportion of SMS and no IDDM sera showing this inhibitory activity (20). The study reported herein with the hmAbs to GAD65 shows that patients with multiple autoantibodies to endocrine organs and Graves’ disease also have antibodies to GAD65 that overlap with those present in IDDM and SMS patients.
The b78 and b96 mAbs show binding to pancreatic islet cells and are similar in this respect to the other reported GAD65 mAbs with ICA activity (23, 24, 26). It is difficult to compare the epitope specificities of the b35, b78, and b96 antibodies to the specificities reported for the MICA antibodies (14, 24, 26) because different methods have been employed for this purpose. Thus, it is not known whether the hmAbs MICA1 and -3 and MICA7, recognizing two conformation-specific epitopes in the C-terminal region (amino acids 450–570) of GAD65, are similar to the IDDM-E2 region; similarly, MICA4 and -6 and MICA10, which recognize another set of two conformation-specific epitopes in the middle region (amino acids 245–449) of GAD65, are similar to the IDDM-E1 region (24). On the same note, binding of the low abundance MICA2 to denatured GAD65 and directed to a SDS-resistant miniconformational region within residues 506–531 of GAD65 (25) may be similar to the binding of b78 antibodies, which also recognizes a SDS-resistant epitope within the IDDM-E2 region of GAD65.
Taken together, our results indicate that b35, b78, and b96 show similarities by targeting specifically GAD65 and recognizing either the IDDM-E1 or the IDDM-E2 region of the molecule. However, the epitopes recognized by the IDDM-E2-specific antibodies b78 and b96 are different, as b78 shows binding to GAD65 by Western blotting. Further confirmation of the recognition of distinct epitopes on GAD65 by the three antibodies was obtained by protein-footprinting patterns. Although antibodies to GAD65 may not play a pathogenic role in the autoimmune islet ß-cell destruction in type 1 diabetes, it is possible that such antibodies may play a significant role in the processing and presentation of T cell epitopes to pathogenic T cells. Professional antigen-presenting cells, such as B cells, are potent presenters by virtue of their ability to capture low abundance antigens by their surface Ig and are rich in the human leukocyte antigen-DM-containing endosomal compartments where the processing and loading of peptides to major histocompatibility complex class II antigens takes place. Thus, antibody binding to an antigen has previously been shown to influence the generation of T cell epitopes, leading to either enhanced or suppressed immune response (27). Using a variety of different epitope-specific, GAD65-specific, B cells generated from patients with autoimmune endocrinopathies, including IDDM, may allow studies of the processing of GAD65 that leads to the recruitment of autoreactive T cells responsible for islet ß-cell destruction.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Present address: Department of Internal Medicine III, University of
Leipzig, Philip Rosenthal Strasse 27, D-04103 Leipzig, Germany.
Received December 11, 1996.
Revised March 7, 1997.
Revised May 6, 1997.
Accepted May 15, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. Fenalti, C. S. Hampe, Y. Arafat, R. H.P. Law, J. P. Banga, I. R. Mackay, J. C. Whisstock, A. M. Buckle, and M. J. Rowley COOH-Terminal Clustering of Autoantibody and T-Cell Determinants on the Structure of GAD65 Provide Insights Into the Molecular Basis of Autoreactivity Diabetes, May 1, 2008; 57(5): 1293 - 1301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Oak, L. K. Gilliam, M. Landin-Olsson, C. Torn, I. Kockum, C. R. Pennington, M. J. Rowley, M. R. Christie, J. P. Banga, and C. S. Hampe The lack of anti-idiotypic antibodies, not the presence of the corresponding autoantibodies to glutamate decarboxylase, defines type 1 diabetes PNAS, April 8, 2008; 105(14): 5471 - 5476. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Hampe, R. Nalini, M. R. Maldonado, T. R. Hall, G. Garza, D. Iyer, and A. Balasubramanyam Association of Amino-Terminal-Specific Antiglutamate Decarboxylase (GAD65) Autoantibodies with {beta}-Cell Functional Reserve and a Milder Clinical Phenotype in Patients with GAD65 Antibodies and Ketosis-Prone Diabetes Mellitus J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 462 - 467. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Raju, J. Foote, J. P. Banga, T. R. Hall, C. J. Padoa, M. C. Dalakas, E. Ortqvist, and C. S. Hampe Analysis of GAD65 Autoantibodies in Stiff-Person Syndrome Patients J. Immunol., December 1, 2005; 175(11): 7755 - 7762. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Padoa, J. P. Banga, A.-M. Madec, M. Ziegler, M. Schlosser, E. Ortqvist, I. Kockum, J. Palmer, O. Rolandsson, K. A. Binder, et al. Recombinant Fabs of Human Monoclonal Antibodies Specific to the Middle Epitope of GAD65 Inhibit Type 1 Diabetes-Specific GAD65Abs Diabetes, November 1, 2003; 52(11): 2689 - 2695. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, GAD65; 
,
GAD67. b, Immunoprecipitation of human IgG mAbs with
[35S]methionine-labeled chimeric GAD proteins.
Immunoprecipitation was performed with either human IgG mAbs b35, b78,
and b96 or control medium (15). The nomenclature for the three chimeric
proteins is: 1) GAD65/67/67 = GAD65-(1–195)/GAD67-(205–594); 2)
GAD67/65/67 = GAD67-(1–230)/GAD65-(221–442)/GAD67-(452–594);
and 3) GAD67/67/65 = GAD67-(1–441)/GAD65-(443–585). The amount
of GAD protein immunoprecipitated (counts per min) is the mean of
duplicate determinations. 
