High-resolution autoreactive epitope mapping and structural modeling of the 65 kDa form of human glutamic acid decarboxylase

Structure and dynamics of GAD65 in complex with an autoimmune polyendocrine syndrome type 2-associated autoantibody

The enzyme glutamate decarboxylase (GAD) produces the neurotransmitter GABA, using pyridoxal-5'-phosphate (PLP). GAD exists as two isoforms, GAD65 and GAD67. Only GAD65 acts as a major autoantigen, frequently implicated in type 1 diabetes and other autoimmune diseases. Here we characterize the structure and dynamics of GAD65 and its interaction with the autoimmune polyendocrine syndrome type 2-associated autoantibody b96.11. Using hydrogen-deuterium exchange mass spectrometry (HDX), X-ray crystallography, cryo-electron microscopy, and computational approaches, we examine the conformational dynamics of apo- and holoGAD65 and the GAD65-autoantibody complex. HDX reveals local dynamics accompanying autoinactivation, with the catalytic loop promoting collective motions at the CTD-PLP domain interface. In the GAD65-b96.11 complex, heavy chain CDRs dominate the interaction, with a long CDRH3 bridging the GAD65 dimer via electrostatic interactions with the 260PEVKEK265motif. This bridging links structural elements controlling GAD65's conformational flexibility to its autoantigenicity. Thus, intrinsic dynamics, rather than sequence differences within epitopes, appear to be responsible for the contrasting autoantigenicities of GAD65 and GAD67. Our findings elucidate the structural and dynamic factors that govern the varying autoantibody reactivities of GAD65 and GAD67, offering a revised rationale for the autoimmune response to GAD65.

Two putative glutamate decarboxylases of Streptococcus pneumoniae as possible antigens for the production of anti-GAD65 antibodies leading to type 1 diabetes mellitus

Type 1 diabetes mellitus (T1DM) has been increasing in prevalence in the last decades and has become a global burden. Autoantibodies against human glutamate decarboxylase (GAD65) are among the first to be detected at the onset of T1DM. Diverse viruses have been proposed to be involved in the triggering of T1DM because of molecular mimicry, i.e., similarity between parts of some viral proteins and one or more epitopes of GAD65. However, the possibility that bacterial proteins might also be responsible for GAD65 mimicry has been seldom investigated. To date, many genomes of Streptococcus pneumoniae (the pneumococcus), a prominent human pathogen particularly prevalent among children and the elderly, have been sequenced. A dataset of more than 9000 pneumococcal genomes was mined and two different (albeit related) genes (gadA and gadB), presumably encoding two glutamate decarboxylases similar to GAD65, were found. The various gadASpn alleles were present only in serotype 3 pneumococci belonging to the global lineage GPSC83, although some homologs have also been discovered in two subspecies of Streptococcus constellatus (pharyngis and viborgensis), an isolate of the group B streptococci, and several strains of Lactobacillus delbrueckii. Besides, gadBSpn alleles are present in > 10% of the isolates in our dataset and represent 16 GPSCs with 123 sequence types and 20 different serotypes. Sequence analyses indicated that gadA- and gadB-like genes have been mobilized among different bacteria either by prophage(s) or by integrative and conjugative element(s), respectively. Substantial similarities appear to exist between the putative pneumococcal glutamate decarboxylases and well-known epitopes of GAD65. In this sense, the use of broader pneumococcal conjugate vaccines such as PCV20 would prevent the majority of serotypes expressing those genes that might potentially contribute to T1DM. These results deserve upcoming studies on the possible involvement of S. pneumoniae in the etiopathogenesis and clinical onset of T1DM.

Reduced display of conformational epitopes in the N-terminal truncated GAD65 isoform: relevance for people with stiff person syndrome or DQ8/8-positive Type 1 diabetes mellitus

AIMS

To investigate whether the N-terminal truncated glutamic acid decarboxylase 65 (GAD65) isoform is as well recognized by people with stiff person syndrome as it is by people with Type 1 diabetes, and whether conformational GAD65 antibody epitopes are displayed properly by the isoform.

METHODS

GAD65 antibody-positive healthy individuals (n=13), people with stiff-person syndrome (n=15) and children with new-onset Type 1 diabetes (n=654) were analysed to determine binding to full-length GAD65 and the N-terminal truncated GAD65 isoform in each of these settings. GAD65 autoantibody epitope specificity was correlated with binding ratios of full-length GAD65/N-terminal truncated GAD65.

RESULTS

The N-terminal truncated GAD65 isoform was significantly less recognized in GAD65Ab-positive people with stiff-person syndrome (P=0.002) and in healthy individuals (P=0.0001) than in people with Type 1 diabetes. Moreover, at least two specific conformational GAD65Ab epitopes were not, or were only partially, presented by the N-terminal truncated GAD65 isoform compared to full-length GAD65. Finally, an N-terminal conformational GAD65Ab epitope was significantly less recognized in DQ8/8 positive individuals with Type 1 diabetes (P=0.02).

CONCLUSIONS

In people with stiff person syndrome preferred binding to the full-length GAD65 isoform over the N-terminal truncated molecule was observed. This binding characteristic is probably attributable to reduced presentation of two conformational epitopes by the N-terminal truncated molecule. These findings support the notion of disease-specific GAD65Ab epitope specificities and emphasize the need to evaluate the applicability of novel assays for different medical conditions.

DPD epitope-specific glutamic acid decarboxylase (GAD)65 autoantibodies in children with Type 1 diabetes

AIM

To study whether DPD epitope-specific glutamate decarboxylase autoantibodies are found more frequently in children with milder forms of Type 1 diabetes.

METHODS

We prospectively evaluated 75 children with new-onset autoimmune Type 1 diabetes, in whom we collected demographic, anthropometric and clinical data and measured islet autoantibodies. Glutamate decarboxylase 65 autoantibody-positive samples were analysed for epitope specificities using recombinant Fab against the DPD-defined epitope of glutamate decarboxylase 65.

RESULTS

After adjustment for age, positive DPD epitope recognition was significantly associated with higher C-peptide levels at onset (P = 0.02, r² =0.21, n = 35), and high DPD recognition in the highest quartile tended to be associated with HbA_1c ≤ 53 mmol/mol (7%) at the last follow-up [mean (sd) follow-up 1.3 (0.4) years; P = 0.07; for the model, P = 0.044, n = 30)]. Age- and sex-adjusted BMI percentile was significantly correlated with recognition of the DPD-defined epitope (P < 0.03, r² =0.14, n = 34), but this correlation was driven by the older age group (age ≥ 10 years; P = 0.016, r² =0.27, n = 21) and was not significant in younger children (P = 0.93, n = 13). There were no independent associations with sex, race/ethnicity, diabetic ketoacidosis, HbA_1c , HLA DR3-DQ2/DR4-DQ8 or autoantibody number.

CONCLUSIONS

Our findings suggest that recognition of the DPD-defined glutamate decarboxylase 65 autoantibody epitope at Type 1 diabetes onset is directly associated with β-cell function, BMI and age, which supports the hypothesis that immunological factors contribute to the clinical heterogeneity of Type 1 diabetes. Larger studies relating epitope-specific glutamate decarboxylase 65 autoantibody to clinical phenotype in children with Type 1 diabetes are warranted.

Mapping autoantigen epitopes: molecular insights into autoantibody-associated disorders of the nervous system

BACKGROUND

Our knowledge of autoantibody-associated diseases of the central (CNS) and peripheral (PNS) nervous systems has expanded greatly over the recent years. A number of extracellular and intracellular autoantigens have been identified, and there is no doubt that this field will continue to expand as more autoantigens are discovered as a result of improved clinical awareness and methodological practice. In recent years, interest has shifted to uncover the target epitopes of these autoantibodies.

MAIN BODY

The purpose of this review is to discuss the mapping of the epitope targets of autoantibodies in CNS and PNS antibody-mediated disorders, such as N-methyl-D-aspartate receptor (NMDAR), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR), leucine-rich glioma-inactivated protein 1 (Lgi1), contactin-associated protein-like 2 (Caspr2), myelin oligodendrocyte glycoprotein (MOG), aquaporin-4 (AQP4), 65 kDa glutamic acid decarboxylase (GAD65), acetylcholine receptor (AChR), muscle-specific kinase (MuSK), voltage-gated calcium channel (VGCC), neurofascin (NF), and contactin. We also address the methods used to analyze these epitopes, the relevance of their determination, and how this knowledge can inform studies on autoantibody pathogenicity. Furthermore, we discuss triggers of autoimmunity, such as molecular mimicry, ectopic antigen expression, epitope spreading, and potential mechanisms for the rising number of double autoantibody-positive patients.

CONCLUSIONS

Molecular insights into specificity and role of autoantibodies will likely improve diagnosis and treatment of CNS and PNS neuroimmune diseases.

Optimized purification strategies for the elimination of non-specific products in the isolation of GAD65-specific monoclonal autoantibodies

Autoantibodies against antigens expressed by insulin-producing β cells are circulating in both healthy individuals and patients at risk of developing Type 1 diabetes. Recent studies suggest that another set of antibodies (anti-idiotypic antibodies) exists in this antibody/antigen interacting network to regulate auto-reactive responses. Anti-idiotypic antibodies may block the antigen-binding site of autoantibodies or inhibit autoantibody expression and secretion. The equilibrium between autoantibodies and anti-idiotypic antibodies plays a critical role in mediating or preventing autoimmunity. In order to investigate the molecular mechanisms underlying such a network in autoimmunity and potentially develop neutralizing reagents to prevent or treat Type 1 diabetes, we need to produce autoantibodies and autoantigens with high quality and purity. Herein, using GAD65/anti-GAD65 autoantibodies as a model system, we aimed to establish reliable approaches for the preparation of highly pure autoantibodies suitable for downstream investigation.

Zinc transporter 8 (ZnT8) autoantibody epitope specificity and affinity examined with recombinant ZnT8 variant proteins in specific ZnT8R and ZnT8W autoantibody-positive type 1 diabetes patients

Variant-specific zinc transporter 8 autoantibodies (ZnT8A) against either arginine (R) or tryptophan (W) at amino acid (aa) position 325 of the zinc transporter 8 (ZnT8) has been identified in type 1 diabetes (T1D) patients. Reciprocal cross-over tests revealed differences in half-maximal binding to indicate variable affinity of patient ZnT8 autoantibodies. Insufficient recombinant ZnT8 variant proteins have precluded detailed analyses of ZnT8 autoantibody affinity. The aims in the present study were to (i) generate recombinant ZnT8R- and ZnT8W-aa275-369 proteins; (ii) test the ZnT8R- and ZnT8W-aa275-369 proteins in reciprocal competitive radiobinding assays (RBA) against ZnT8R- and ZnT8W-aa268-369 labelled with (35) S-methionine; and (iii) determine the specificity and affinity of sera specific for either ZnT8 arginine (R) or ZnT8 tryptophan (W) autoantibodies in newly diagnosed T1D patients. The results demonstrate, first, that it was possible to produce recombinant human MBP-ZnT8-aa275-369 protein purified to homogeneity for RBA reciprocal competition experiments. Secondly, high-titre ZnT8WA sera diluted to half maximal binding showed significant specificity for respective variants of either ZnT8R or ZnT8W. Thirdly, ZnT8WA-positive sera showed high affinity for ZnT8W compared to ZnT8RA for ZnT8R. These data demonstrate that T1D patients may have single amino acid-specific autoantibodies directed against either ZnT8R or ZnT8W and that the autoantibody affinity to the respective variant may be different. Further studies are needed to assess the mechanisms by which variant-specific ZnT8A of variable affinity develop and their possible role in the pathogenic process leading to the clinical onset of T1D.

Cofactor-dependent conformational heterogeneity of GAD65 and its role in autoimmunity and neurotransmitter homeostasis

The human neuroendocrine enzyme glutamate decarboxylase (GAD) catalyses the synthesis of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) using pyridoxal 5'-phosphate as a cofactor. GAD exists as two isoforms named according to their respective molecular weights: GAD65 and GAD67. Although cytosolic GAD67 is typically saturated with the cofactor (holoGAD67) and constitutively active to produce basal levels of GABA, the membrane-associated GAD65 exists mainly as the inactive apo form. GAD65, but not GAD67, is a prevalent autoantigen, with autoantibodies to GAD65 being detected at high frequency in patients with autoimmune (type 1) diabetes and certain other autoimmune disorders. The significance of GAD65 autoinactivation into the apo form for regulation of neurotransmitter levels and autoantibody reactivity is not understood. We have used computational and experimental approaches to decipher the nature of the holo → apo conversion in GAD65 and thus, its mechanism of autoinactivation. Molecular dynamics simulations of GAD65 reveal coupling between the C-terminal domain, catalytic loop, and pyridoxal 5'-phosphate-binding domain that drives structural rearrangement, dimer opening, and autoinactivation, consistent with limited proteolysis fragmentation patterns. Together with small-angle X-ray scattering and fluorescence spectroscopy data, our findings are consistent with apoGAD65 existing as an ensemble of conformations. Antibody-binding kinetics suggest a mechanism of mutually induced conformational changes, implicating the flexibility of apoGAD65 in its autoantigenicity. Although conformational diversity may provide a mechanism for cofactor-controlled regulation of neurotransmitter biosynthesis, it may also come at a cost of insufficient development of immune self-tolerance that favors the production of GAD65 autoantibodies.

A versatile protein microarray platform enabling antibody profiling against denatured proteins

PURPOSE

We aim to develop a protein microarray platform capable of presenting both natural and denatured forms of proteins for antibody biomarker discovery. We will further optimize plasma screening protocols to improve detection.

EXPERIMENTAL DESIGN

We developed a new covalent capture protein microarray chemistry using HaloTag fusion proteins and ligand. To enhance protein yield, we used HeLa cell lysate as an in vitro transcription translation (IVTT) system. Escherichia coli lysates were added to the plasma blocking buffer to reduce nonspecific background. These protein microarrays were probed with plasma samples and autoantibody responses were quantified and compared with or without denaturing buffer treatment.

RESULTS

We demonstrated that protein microarrays using the covalent attachment chemistry endured denaturing conditions. Blocking with E. coli lysates greatly reduced the background signals and expression with IVTT based on HeLa cell lysates significantly improved the antibody signals on protein microarrays probed with plasma samples. Plasma samples probed on denatured protein arrays produced autoantibody profiles distinct from those probed on natively displayed proteins.

CONCLUSIONS AND CLINICAL RELEVANCE

This versatile protein microarray platform allows the display of both natural and denatured proteins, offers a new dimension to search for disease-specific antibodies, broadens the repertoire of potential biomarkers, and will potentially yield clinical diagnostics with greater performance.

Stiff-person syndrome (SPS) and anti-GAD-related CNS degenerations: protean additions to the autoimmune central neuropathies

Stiff Person Syndrome (SPS) is a rare autoimmune neurological disease attributable to autoantibodies to glutamic acid decarboxylase (anti-GAD) more usually associated with the islet beta cell destruction of autoimmune type 1 diabetes (T1D). SPS is characterized by interference in neurons with the synthesis/activity of the inhibitory neurotransmitter gamma amino butyric acid (GABA) resulting in the prototypic progressive spasmodic muscular rigidity of SPS, or diverse neurological syndromes, cerebellar ataxia, intractable epilepsy, myoclonus and several others. Remarkably, a single autoantibody, anti-GAD, can be common to widely different disease expressions, i.e. T1D and SPS. One explanation for these data is the differences in epitope engagement between the anti-GAD reactivity in SPS and T1D: in both diseases, anti-GAD antibody reactivity is predominantly to a conformational epitope region in the PLP- and C-terminal domains of the 65 kDa isoform but, additionally in SPS, there is reactivity to conformational epitope(s) on GAD67, and short linear epitopes in the C-terminal region and at the N-terminus of GAD65. Another explanation for disease expressions in SPS includes ready access of anti-GAD to antigen sites due to immune responsiveness within the CNS itself according to intrathecal anti-GAD-specific B cells and autoantibody. Closer study of the mysterious stiff-person syndrome should enhance the understanding of this disease itself, and autoimmunity in general.

Autoantigen discovery with a synthetic human peptidome

Immune responses targeting self-proteins (autoantigens) can lead to a variety of autoimmune diseases. Identification of these antigens is important for both diagnostic and therapeutic reasons. However, current approaches to characterize autoantigens have, in most cases, met only with limited success. Here we present a synthetic representation of the complete human proteome, the T7 peptidome phage display library (T7-Pep), and demonstrate its application to autoantigen discovery. T7-Pep is composed of >413,000 36-residue, overlapping peptides that cover all open reading frames in the human genome, and can be analyzed using high-throughput DNA sequencing. We developed a phage immunoprecipitation sequencing (PhIP-Seq) methodology to identify known and previously unreported autoantibodies contained in the spinal fluid of three individuals with paraneoplastic neurological syndromes. We also show how T7-Pep can be used more generally to identify peptide-protein interactions, suggesting the broader utility of our approach for proteomic research.

An analysis of the cross-reactivity of autoantibodies to GAD65 and GAD67 in diabetes

Background

Autoantibodies to GAD65 (anti-GAD65) are present in the sera of 70–80% of patients with type 1 diabetes (T1D), but antibodies to the structurally similar 67 kDa isoform GAD67 are rare. Antibodies to GAD67 may represent a cross-reactive population of anti-GAD65, but this has not been formally tested.

Methodology/Principal Findings

In this study we examined the frequency, levels and affinity of anti-GAD67 in diabetes sera that contained anti-GAD65, and compared the specificity of GAD65 and GAD67 reactivity. Anti-GAD65 and anti-GAD67 were measured by radioimmunoprecipitation (RIP) using ¹²⁵I labeled recombinant GAD65 and GAD67. For each antibody population, the specificity of the binding was measured by incubation with 100-fold excess of unlabeled GAD in homologous and heterologous inhibition assays, and the affinity of binding with GAD65 and GAD67 was measured in selected sera. Sera were also tested for reactivity to GAD65 and GAD67 by immunoblotting. Of the 85 sera that contained antibodies to GAD65, 28 contained anti–GAD67 measured by RIP. Inhibition with unlabeled GAD65 substantially or completely reduced antibody reactivity with both ¹²⁵I GAD65 and with ¹²⁵I GAD67. In contrast, unlabeled GAD67 reduced autoantibody reactivity with ¹²⁵I GAD67 but not with ¹²⁵I GAD65. Both populations of antibodies were of high affinity (>10¹⁰ l/mol).

Conclusions

Our findings show that autoantibodies to GAD67 represent a minor population of anti-GAD65 that are reactive with a cross-reactive epitope found also on GAD67. Experimental results confirm that GAD65 is the major autoantigen in T1D, and that GAD67 per se has very low immunogenicity. We discuss our findings in light of the known similarities between the structures of the GAD isoforms, in particular the location of a minor cross-reactive epitope that could be induced by epitope spreading.

Two distinct mechanisms target GAD67 to vesicular pathways and presynaptic clusters

The inhibitory neurotransmitter gamma-amino butyric acid (GABA) is synthesized by two isoforms of the enzyme glutamic acid decarboxylase (GAD): GAD65 and GAD67. Whereas GAD67 is constitutively active and produces >90% of GABA in the central nervous system, GAD65 is transiently activated and augments GABA levels for rapid modulation of inhibitory neurotransmission. Hydrophobic lipid modifications of the GAD65 protein target it to Golgi membranes and synaptic vesicles in neuroendocrine cells. In contrast, the GAD67 protein remains hydrophilic but has been shown to acquire membrane association by heterodimerization with GAD65. Here, we identify a second mechanism that mediates robust membrane anchoring, axonal targeting, and presynaptic clustering of GAD67 but that is independent of GAD65. This mechanism is abolished by a leucine-103 to proline mutation that changes the conformation of the N-terminal domain but does not affect the GAD65-dependent membrane anchoring of GAD67. Thus two distinct mechanisms target the constitutively active GAD67 to presynaptic clusters to facilitate accumulation of GABA for rapid delivery into synapses.

Characterisation of peptide microarrays for studying antibody-antigen binding using surface plasmon resonance imagery

Background

Non-specific binding to biosensor surfaces is a major obstacle to quantitative analysis of selective retention of analytes at immobilized target molecules. Although a range of chemical antifouling monolayers has been developed to address this problem, many macromolecular interactions still remain refractory to analysis due to the prevalent high degree of non-specific binding. We describe how we use the dynamic process of the formation of self assembling monolayers and optimise physical and chemical properties thus reducing considerably non-specific binding and allowing analysis of specific binding of analytes to immobilized target molecules.

Methodology/Principal Findings

We illustrate this approach by the production of specific protein arrays for the analysis of interactions between the 65kDa isoform of human glutamate decarboxylase (GAD65) and a human monoclonal antibody. Our data illustrate that we have effectively eliminated non-specific interactions with the surface containing the immobilised GAD65 molecules. The findings have several implications. First, this approach obviates the dubious process of background subtraction and gives access to more accurate kinetic and equilibrium values that are no longer contaminated by multiphase non-specific binding. Second, an enhanced signal to noise ratio increases not only the sensitivity but also confidence in the use of SPR to generate kinetic constants that may then be inserted into van't Hoff type analyses to provide comparative ΔG, ΔS and ΔH values, making this an efficient, rapid and competitive alternative to ITC measurements used in drug and macromolecular-interaction mechanistic studies. Third, the accuracy of the measurements allows the application of more intricate interaction models than simple Langmuir monophasic binding.

Conclusions

The detection and measurement of antibody binding by the type 1 diabetes autoantigen GAD65 represents an example of an antibody-antigen interaction where good structural, mechanistic and immunological data are available. Using SPRi we were able to characterise the kinetics of the interaction in greater detail than ELISA/RIA methods. Furthermore, our data indicate that SPRi is well suited to a multiplexed immunoassay using GAD65 proteins, and may be applicable to other biomarkers.

Long-lived plasma cells and memory B cells produce pathogenic anti-GAD65 autoantibodies in Stiff Person Syndrome

Stiff person syndrome (SPS) is a rare, neurological disorder characterized by sudden cramps and spasms. High titers of enzyme-inhibiting IgG autoantibodies against the 65 kD isoform of glutamic acid decarboxylase (GAD65) are a hallmark of SPS, implicating an autoimmune component in the pathology of the syndrome. Studying the B cell compartment and the anti-GAD65 B cell response in two monozygotic twins suffering from SPS, who were treated with the B cell-depleting monoclonal anti-CD20 antibody rituximab, we found that the humoral autoimmune response in SPS is composed of a rituximab-sensitive part that is rapidly cleared after treatment, and a rituximab-resistant component, which persists and acts as a reservoir for autoantibodies inhibiting GAD65 enzyme activity. Our data show that these potentially pathogenic anti-GAD65 autoantibodies are secreted by long-lived plasma cells, which may either be persistent or develop from rituximab-resistant memory B lymphocytes. Both subsets represent only a fraction of anti-GAD65 autoantibody secreting cells. Therefore, the identification and targeting of this compartment is a key factor for successful treatment planning of SPS and of similar autoimmune diseases.

GAD65 as a prototypic autoantigen

The repertoire of known autoantigens is limited to a very small proportion of all human proteins, and the reason why only some proteins become autoantigens is unclear, but is likely associated with structural features. The 65kDa isoform of the enzyme glutamic acid decarboxylase (GAD65) is a major autoantigen in type I diabetes, and in various neurological diseases, whereas the closely related isoform, GAD67, is rarely antigenic. Conformational epitopes of GAD65 have been mapped using human monoclonal antibodies to GAD65 and GAD mutated by GAD65/67 sequence exchanges or point mutations, but these studies have been limited by a lack of structural information. The recent publication of crystal structures for the two isoforms has shown that the N-, C- and middle domains that have been identified previously as likely epitope regions are closely associated within the GAD dimer. Two major epitope regions, ctc1 and ctc2, have been identified in the C-terminal domain of GAD65, that encompass N- and C-terminal residues, and middle and C-terminal residues respectively. These regions are highly flexible compared with the equivalent regions in GAD67, and T cell epitopes have been localized to the same surface region of GAD65. Comparative analysis of these two structurally similar isoforms, GAD65 and GAD67, only one of which is autoantigenic should provide new insights into the provocations to autoimmunity.

COOH-terminal clustering of autoantibody and T-cell determinants on the structure of GAD65 provide insights into the molecular basis of autoreactivity

OBJECTIVE

To gain structural insights into the autoantigenic properties of GAD65 in type 1 diabetes, we analyzed experimental epitope mapping data in the context of the recently determined crystal structures of GAD65 and GAD67, to allow "molecular positioning" of epitope sites for B- and T-cell reactivity.

RESEARCH DESIGN AND METHODS

Data were assembled from analysis of reported effects of mutagenesis of GAD65 on its reactivity with a panel of 11 human monoclonal antibodies (mAbs), supplemented by use of recombinant Fab to cross-inhibit reactivity with GAD65 by radioimmunoprecipitation of the same mAbs.

RESULTS

The COOH-terminal region on GAD65 was the major autoantigenic site. B-cell epitopes were distributed within two separate clusters around different faces of the COOH-terminal domain. Inclusion of epitope sites in the pyridoxal phosphate-and NH(2)-terminal domains was attributed to the juxtaposition of all three domains in the crystal structure. Epitope preferences of different mAbs to GAD65 aligned with different clinical expressions of type 1 diabetes. Epitopes for four of five known reactive T-cell sequences restricted by HLA DRB1*0401 were aligned to solvent-exposed regions of the GAD65 structure and colocalized within the two B-cell epitope clusters. The continuous COOH-terminal epitope region of GAD65 was structurally highly flexible and therefore differed markedly from the equivalent region of GAD67.

CONCLUSIONS

Structural features could explain the differing antigenicity, and perhaps immunogenicity, of GAD65 versus GAD67. The proximity of B- and T-cell epitopes within the GAD65 structure suggests that antigen-antibody complexes may influence antigen processing by accessory cells and thereby T-cell reactivity.

Modulation of diabetes in NOD mice by GAD65-specific monoclonal antibodies is epitope specific and accompanied by anti-idiotypic antibodies

Type 1 diabetes is caused by the autoimmune destruction of pancreatic beta cells. Here we show that administration of a human monoclonal antibody (b96.11) specific to the 65-kDa isoform of glutamate decarboxylase (GAD65) to prediabetic non-obese diabetic (NOD) mice significantly delays the onset of autoimmune diabetes. We found this effect to be epitope-specific, as only b96.11 showed this therapeutic property, while a GAD65-specific human monoclonal control antibody (b78) derived from the same patient, but specific to a different determinant of GAD65, had no significant effect on the progression of disease. Administration of b96.11 or b78 to NOD mice was accompanied by the generation of anti-idiotypic antibodies. Importantly, the induced anti-idiotypic antibodies were specific for the immunizing antibody and blocked the binding of GAD65 by the respective antibody. These findings suggest a potential role for the internal image of the GAD65 determinant recognized by b96.11 in the anti-idiotypic antibody, supporting an immunomodulatory role for GAD65-specific autoantibodies, as originally postulated by Jerne.

Autoantibody epitopes to the smaller isoform of glutamate decarboxylase do not differ in Swedish and Japanese type 1 diabetes patients and may be associated with high-risk human leucocyte antigen class II alleles

Type 1 diabetes (T1D) is an autoimmune disease with a strong human leucocyte antigen (HLA) class II association. Depending on geographic locations, the disease-associated HLA class II alleles vary. We evaluated the beta cell-specific autoimmunity reflected in autoantibodies directed to the smaller isoform of glutamate decarboxylase (GAD65) in Japanese and Swedish T1D patients. GAD65Ab epitope specificities were assessed using GAD65-specific recombinant Fab. GAD65Ab epitope specificities did not differ between Swedish and Japanese patients. Only recognition of the MICA-4-defined middle epitope was significantly stronger in the Japanese T1D patient group compared to the Swedish T1D patients (P = 0.001). Binding to the b96.11-defined middle epitope was substantial in both groups and showed significant associations with high-risk HLA class II haplotypes. In the Japanese T1D group the association was with haplotype DRB1*0802-DQB1*0302 (P = 0.0008), while in the Swedish T1D patients binding to the b96.11-defined epitope as associated with the presence of high-risk HLA genotypes DR3-DQB1*0201 and/or DR4-DQB1*0302 (P = 0.02). A significant association between reduction in binding in the presence of recombinant Fab (rFab) DPD and high-risk allele DQB1*0201 was found (P = 0.008) in the Swedish T1D patients only. We hypothesize that epitope-specific autoantibodies effect the peptide presentation on HLA class II molecules by modulating antigen uptake and processing. Molecular modelling of the high-risk HLA class II molecules will be necessary to test whether these different molecules present similar peptide-binding specificities.

Longitudinal changes in epitope recognition of autoantibodies against glutamate decarboxylase 65 (GAD65Ab) in prediabetic adults developing diabetes

We analysed the beta cell-specific autoimmunity reflected in autoantibodies to the smaller isoform of glutamate decarboxylase (GAD65Ab) in the prediabetic period of GAD65Ab-positive healthy adults who developed Type 2 diabetes (T2D) during a follow-up period of 10 years. We found that of the adults that tested GAD65Ab-positive at baseline (n=25), six developed T2D and one developed Type 1 diabetes (T1D). Of the subjects that tested GAD65Ab-negative at baseline (n=2209), 81 developed T2D, one developed T1D and four developed unclassified diabetes, indicating that the risk for GAD65Ab-positive healthy adults to develop diabetes is increased sixfold. The GAD65Ab epitopes were characterized in a competition radioligand binding assay using recombinant Fab derived of GAD65-specific monoclonal antibodies. We observed that the GAD65Ab epitope specificities in the prediabetic period changed dynamically. Specifically, the binding to a middle and a C-terminal epitope increased during the follow-up period (P=0 x 03), causing a significant increase in the number of epitopes recognized (P=0 x 03). These findings are similar to previous observations of dynamic changes in the prediabetic period of schoolchildren at high risk for T1D development. However, the character of the epitopes differs between the two populations, suggesting differences in the beta cell-specific autoimmune response in the prediabetic period of patients with latent autoimmune diabetes in adults (LADA) and T1D.

GAD65 autoantibody epitopes in adult patients with latent autoimmune diabetes following GAD65 vaccination

AIMS

Subcutaneous injection of recombinant human GAD65 (rhGAD65) in patients with latent autoimmune diabetes in adults (LADA) correlates with an increase in C-peptide levels. In this study we analysed the effect of rhGAD65 administration on the GAD65-specific autoimmune response.

METHODS

Longitudinal serum samples obtained from LADA patients (n = 47) who received 4, 20, 100 or 500 microg alum-formulated rhGAD65 or placebo by subcutaneous injection twice (4 weeks apart) were analysed for their epitope recognition using GAD65-specific recombinant Fab and GAD65/67 fusion proteins.

RESULTS

Overall, minor changes in the epitope pattern were observed using either approach. Only in the 500-microg dosage group was an increase in GAD65Ab level associated with a significant increase in the binding to a conformational epitope located at the middle part of GAD65.

CONCLUSIONS

Our data suggest that the apparent beneficial effects of 20 microg alum-formulated recombinant human GAD65 is not explained by changes in the GAD65Ab epitope pattern.

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

CONTEXT

We previously characterized patients presenting with diabetic ketoacidosis prospectively into four subgroups of ketosis-prone diabetes mellitus (KPDM), based on the presence or absence of beta-cell autoimmunity (A+ or A-) and beta-cell functional reserve (B+ or B-). The A+B- KPDM subgroup comprises patients with classic, autoimmune type 1 diabetes, whereas the A+B+ KPDM subgroup has only partial beta-cell loss and a distinct clinical phenotype.

OBJECTIVE

We hypothesized that epitope specificity of autoantibodies directed against the 65-kDa isoform of glutamate decarboxylase (GAD65) reflects differences in beta-cell destruction.

DESIGN

Sera of sequential GAD65Ab-positive KPDM patients admitted for diabetic ketoacidosis (n = 36) were analyzed for their epitope recognition using five GAD65-specific recombinant Fab and their ability to inhibit GAD65 enzymatic activity. All patients were followed longitudinally to assess beta-cell functional reserve and insulin dependence.

RESULTS

Binding to an amino-terminal epitope defined by monoclonal antibody DPD correlated positively with fasting serum C-peptide levels at baseline (P = 0.0008) and after 1 yr (P = 0.007). Binding to the DPD-defined epitope also correlated positively with area under the curve for C-peptide after glucagon stimulation (P = 0.007) and with homeostasis model assessment percent B at 1 yr (P = 0.03). Binding to the DPD-defined epitope was significantly stronger in A+B+ than in A+B- patients (P = 0.001). Sera of 16 patients (44%) significantly inhibited GAD65 enzymatic activity, but this did not correlate with beta-cell function.

CONCLUSION

DPD-defined epitope specificity is correlated directly with preserved beta-cell functional reserve in GAD65Ab-positive patients and is associated with the milder clinical phenotype of A+B+ KPDM.

Molecular characterization of a disease associated conformational epitope on GAD65 recognised by a human monoclonal antibody b96.11

Autoantibodies to the 65kDa isoform of glutamate decarboxylase (GAD65) are associated with type I diabetes and recognise highly conformational epitope(s) that remain to be defined. The human recombinant Fab from mAb b96.11 inhibits binding of most GAD65 antibody positive sera from patients and its epitope has previously been localized to the middle region of GAD65. Recent studies indicate that b96.11 antibody specificity predicts the risk of developing type 1 diabetes in prediabetic individuals. We describe the use homology modelling, protein-protein docking simulations and biopanning of random peptide phage displayed libraries with b96.11 to predict contact amino acids on the interface of GAD65/Fab b96.11 complex. Further analysis by in vitro mutagenesis of GAD65 followed by radioimmunoprecipitation refined the amino acids contributing to the b96.11 epitope. Our studies show an interface characterized by a protruding antibody-combining site centered on the long heavy chain CDR3 loop of Fab b96.11 establishing interactions with the critical residue Phe(344) in the core of the epitope on GAD65, surrounded by charged sites within (375)RK(376) and (305)DER(307). The epitope requires residues from both middle and the C-terminal domains, and is the first precise definition of an epitope on GAD65. The nature of the b96.11 epitope leads to considerations of potential structural variations for differences in antigenicity between the isoforms GAD65 and GAD67. The study shows the utility of using a combination of in silico techniques and experimental data for molecular characterization and localization of conformational epitopes for which crystal structures are lacking.

Species and epitope specificity of two 65 kDa glutamate decarboxylase time-resolved fluorometric immunoassays

The 65 kDa isoform of human glutamate decarboxylase (GAD65) is a major autoantigen in type 1 diabetes (T1D). In the present study, we have developed a sensitive sandwich time-resolved fluorescence immunoassay (TRFIA) for the quantification of GAD65 in cell extracts, cell media and serum. The monoclonal antibody GAD-6 is used to selectively capture GAD65 but not the slightly larger isoform GAD67, and the utilization of different detecting antibodies with distinct GAD65 epitope specificity allows modulating the specificity of the assay. To this effect we have biotinylated a recombinant antigen-binding fragment (rFab) with epitope specificity for the N-terminal region of rat and human GAD65 (rFab N-GAD65) and another rFab that selectively binds to the middle part of human GAD65 (rFab b96.11). In the assay the biotinylated rFabs are recognized by Europium labeled streptavidin. The obtained time-resolved fluorescence (TRF) is directly proportional to the concentration of GAD65 over a large measuring range (0.1 to >100 ng/mL). Based on total error estimation including both bias and imprecision, the lower limit of quantitation (LLOQ) of GAD65 in cell extracts is 0.33 ng/mL with the N-GAD65 TRFIA, and 0.10 ng/mL with the b96.11 TRFIA, but the latter is suitable for human GAD65 only, whereas the N-GAD65 TRFIA has equal sensitivity with rat and human GAD65. Specificity was further checked with GAD65/67 fusion proteins, confirming that the presence of intact capture as well as detection epitope on the analyte is a prerequisite for recognition in both assays. We show that the beta cell-specific marker GAD65 can be quantified in pancreatic cell extracts and in serum, allowing studies on discharge during cell death in vitro as well as in vivo.

Activity-dependent cleavage of brain glutamic acid decarboxylase 65 by calpain

Previously, we reported that l-glutamic acid decarboxylase isoform 65 (GAD65) could be cleaved in vitro to release a stable truncated form which lacks amino acid 1-69 from the N-terminus, GAD65(Delta1-69). However, whether such a truncated form is also present under certain physiological conditions remains elusive. In the present study, we showed that, upon sustained neuronal stimulation, GAD65 could be cleaved into a truncated form in a rat synaptosomal preparation. This truncated form had similar electrophoretic mobility to purified recombinant human GAD65(Delta1-69). Furthermore, we demonstrated that this conversion was calcium dependent. Calcium-chelating reagents such as EDTA and 1,2-bis-(o-aminphenoxy)-ethane-N,N,N',N'-tetra-acetic acid tetra-acetoxy-methyl ester prevented the cleavage of GAD65. In addition, our data suggested that calpain, a calcium-dependent cysteine protease, is activated upon neuronal stimulation and could be responsible for the conversion of full-length GAD65 to truncated GAD65 in the brain. Moreover, calpain inhibitors such as calpain inhibitor I or calpastatin could block the cleavage. Results of our in vitro cleavage assay using purified calpain and immunopurified rat GAD65 also supported the idea that GAD65 could be directly cleaved by calpain.

A highly sensitive immunoassay resistant to autoantibody interference for detection of the diabetes-associated autoantigen glutamic acid decarboxylase 65 in blood and other biological samples

BACKGROUND

Glutamic acid decarboxylase-65 (GAD65) is a major autoantigen in autoimmune diabetes and is discharged from injured islet beta cells. GAD65 may also be released by transplanted islets undergoing immunological rejection. To test hypotheses regarding the utility of GAD65 as a biomarker for transplant rejection or diabetes-associated islet damage and also regarding the timing and instigators of GAD65 release in humans or animal models, a sensitive assay capable of measuring GAD65 in serum or plasma will be necessary. Ideally, this assay would also be resistant to interference by anti-GAD65 autoantibodies.

METHODS

A novel, magnetic bead-based assay was developed based on GAD65 capture by a monoclonal antibody directed to the only region of the protein known not to be significantly targeted by autoantibodies. A subsequent denaturation step allows sensitive immunodetection to proceed using anti-GAD65 polyclonal antibodies that would otherwise potentially be blocked by bound autoantibodies.

RESULTS

The GAD65 assay worked equally well with serum and plasma as with a solution of bovine serum albumin (BSA). The limit of blank was 31 pg/mL and did not differ significantly in the BSA solution (27 pg/mL). Mean recovery of GAD65 from the plasma of control subjects and GAD65 autoantibody-positive and -negative subjects with type 1 diabetes was 101 +/- 4.6%, 88 +/- 7.8%, and 99 +/- 7.0% (+/- SEM), respectively. The assay was used to quantify both recombinant GAD65 and the GAD65 content of human and rodent islets and other tissue extracts that were added to human plasma samples.

CONCLUSIONS

A sensitive, autoantibody-resistant GAD65 assay has been developed that is compatible with detection in serum and plasma and therefore will likely also work with a variety of other biologic fluids. This assay may enable the use of circulating GAD65 as a biomarker of islet damage or transplant rejection and will facilitate in vivo studies of the pathogenesis of anti-GAD65 autoreactivity.

Characterisation of an autoreactive conformational epitope on GAD65 recognised by the human monoclonal antibody b78 using a combination of phage display, in vitro mutagenesis and molecular modelling

Autoantibodies to the diabetes autoantigen, the 65kDa isoform of glutamic acid decarboxylase (GAD65), react with conformational epitopes defined according to linear sequences but not according to structural information, or contact sites with the antibody paratope. To ascertain such information for an exemplary human monoclonal antibody (mAb) to GAD65, b78, we combined antibody screening of phage-displayed peptide libraries, alanine mutagenesis of selected motifs, homology modelling of the PLP and C-terminal regions of GAD65, and molecular dynamics to examine for structural effects of mutagenesis. By phage display, mAb b78 selected phagotopes containing acidic residues (D, E), hydrophobic residues (Y, F or W) and LRS that localised to a possible surface-exposed conformational epitope on the combined homology model. Alanine mutants of GAD65 based on deduced contact residues were examined for binding with b78 and control sera. Mutation of (524)SRL(526), (572)DF(573) and (498)KPQ(500) reduced reactivity of b78 with mutant GAD65 > 50%. Molecular dynamics indicated that mutation of (498)KPQ(500) caused structural changes that could account for effects of this mutation. Thus phage display in combination with molecular modelling identified contact residues within a highly conformational epitope for mAb b78 in the C-terminus of GAD65. These techniques should have broad applicability to definition of epitope structure.

Structural model of human GAD65: prediction and interpretation of biochemical and immunogenic features

The 65 kDa human isoform of glutamate decarboxylase, GAD65, plays a central role in neurotransmission in higher vertebrates and is a typical autoantigen in several human autoimmune diseases, such as insulin-dependent diabetes mellitus (IDDM), Stiff-man syndrome and autoimmune polyendocrine syndrome type I. In autoimmune diabetes, an attack of inflammatory cells to endocrine pancreatic beta-cells leads to their complete destruction, eventually resulting in the inability to produce sufficient insulin for the body's requirements. Even though the etiology of beta-cell destruction is still a matter of debate, the role and antigenic potency of GAD65 are widely recognized. Herein a model of GAD65 is presented, which is based on the recently solved crystal structures of mammalian DOPA decarboxylase and of bacterial glutamate decarboxylase. The model provides for the first time a detailed and accurate structure of the GAD65 subunit (all three domains) and of its dimeric quaternary assembly. It reveals the structural basis for specific antibody recognition to GAD65 as opposed to GAD67, the other human isoform, which shares 81% sequence similarity with GAD65 and is much less antigenic. Literature data on monoclonal antibody binding are perfectly consistent with the detailed features of the model, which allows explanation of several findings on GAD65 immunogenicity. Importantly, by analyzing the active site, we identified the residues most likely involved in catalysis and substrate recognition, paving the way for rational mutagenesis studies of the GAD65 reaction mechanism, specificity and inhibition.

Amino acids critical for binding of autoantibody to an immunodominant conformational epitope of the pyruvate dehydrogenase complex subunit E2: Identification by phage display and site-directed mutagenesis

The E2 subunit of the mitochondrial multienzyme pyruvate dehydrogenase complex (PDC-E2) is the major autoantigen in the liver disease, primary biliary cirrhosis (PBC). An epitope region which has been localized to amino acids 91-227 is believed to include the residue K173 to which is attached the lipoyl cofactor. We investigated structural features of this epitope region by screening random peptide phage-displayed libraries and identified prevalent phagotopes that contained likely contact amino acids in separate regions of the linear sequence, H132M133, and F178, V180. These were confirmed by site-directed alanine mutagenesis singly or in combination of the HM and FV residues in wild-type (wt) PDC-E2, and by immunization of rabbits with phage that expressed peptides MHLNTPP or FVLPWRI. The lipoyl lysine K173 also was mutated. Reactivities of mutants and wild-type (wt) PDC-E2, compared by ELISA using 12 PBC sera, showed decremental reactivity of mutant versus wt PDC-E2 (normalized to 100%): wt PDC-E2 (100%)>>PDC-E2(F178A,V180A) (mean+/-S.D., 59+/-17%)>PDC-E2(M133A) (50+/-13%)>PDC-E2(H132A) (36+/-13%)>PDC-E2(H132A,M133A) (28+/-8%)>PDC-E2(H132A,M133A,F178V,M180A) (18+/-13%). Notably PDC-E2(K173A) retained full reactivity (93+/-21%). Rabbits immunized with phage peptides generated antibodies reactive with entire PDC-E2. Our data convincingly validate phage library technology for defining spatially disparate contact residues for conformational epitopes. Ensuing data could be generally applicable to search for occult extrinsic agents as initiators of autoimmunity.

Analysis of GAD65 autoantibodies in Stiff-Person syndrome patients

Autoantibodies to the 65-kDa isoform of glutamate decarboxylase GAD65 (GAD65Ab) are strong candidates for a pathological role in Stiff-Person syndrome (SPS). We have analyzed the binding specificity of the GAD65Ab in serum and cerebrospinal fluid (CSF) of 12 patients with SPS by competitive displacement studies with GAD65-specific rFab-derived from a number of human and mouse mAbs specific for different determinants on the Ag. We demonstrate considerable differences in the epitope specificity when comparing paired serum and CSF samples, suggesting local stimulation of B cells in the CSF compartment of these patients. Moreover, these autoantibodies strongly inhibit the enzymatic activity of GAD65, thus blocking the formation of the neurotransmitter gamma-aminobutyric acid. The capacity of the sera to inhibit the enzymatic activity of GAD65 correlated with their binding to a conformational C-terminal Ab epitope. Investigation of the inhibitory mechanism revealed that the inhibition could not be overcome by high concentrations of glutamate or the cofactor pyridoxal phosphate, suggesting a noncompetitive inhibitory mechanism. Finally, we identified a linear epitope on amino acids residues 4-22 of GAD65 that was recognized solely by autoantibodies from patients with SPS but not by serum from type 1 diabetes patients. A mAb (N-GAD65 mAb) recognizing this N-terminal epitope was successfully humanized to enhance its potential therapeutic value by reducing its overall immunogenicity.

A look to the future: prediction, prevention, and cure including islet transplantation and stem cell therapy

Type 1 diabetes mellitus (T1DM) is characterized by the almost complete absence of insulin secretion, which is secondary to an autoimmune destruction or dysfunction of the insulin-producing cells of the pancreatic islets of Langerhans. Because T1DM is an autoimmune disease with a long preclinical course, the predictive testing of individuals before the clinical onset of the disease has provided a real opportunity for the identification of risk markers and the design of therapeutic intervention. With such a high degree of predictability using a combination of immunologic markers, strategies to prevent T1DM may become possible. A number of novel therapeutic strategies are under investigation in newly diagnosed T1DM patients and might ultimately be applied to prevent T1DM.

Humoral autoimmune responses to glutamic acid decarboxylase have similar target epitopes and subclass that show titer-dependent disease association

Glutamic acid decarboxylase (GAD) is an autoantigen in stiff man syndrome (SMS) and type 1 diabetes (T1DM). Different GAD autoantibody characteristics in these disorders have suggested distinct underlying mechanisms of autoimmunity. Here, it is shown that increased prevalence of autoantibodies to GAD65 amino terminal and GAD67 epitopes and autoantibodies of IgG2, IgG3, or IgG4 subclass in patients with SMS (P < 0.001 vs. T1DM) are secondary to the markedly higher autoantibody titers in SMS patients (P < 0.0001) and that autoantibody epitopes and subclasses were similar when patients were matched for autoantibody titer. Exposure to autoantigen in the disorders is likely to involve similar humoral antigenic determinants, but different B cell regulation.

Epitope analysis of insulin autoantibodies using recombinant Fab

Autoantibodies to insulin are often the first autoantibodies detected in young children with type 1 diabetes and can be present before the onset of clinical diabetes. These autoantibodies and their epitopes are, however, not well characterized. We explored the use of monoclonal antibodies and their recombinant Fab as reagents for epitope analysis. In this study we cloned and characterized the recombinant Fab of the insulin-specific monoclonal antibody CG7C7. We found the epitope of this antibody to be located predominantly at the A-chain loop of the insulin molecule. The recombinant Fab was then used to compete for insulin binding against insulin autoantibodies present in sera from patients with type 1 or type 1.5 diabetes. In competition experiments with sera positive for autoantibodies to insulin the recombinant Fab significantly reduced the binding to [125I]-insulin by sera of type 1 (n = 35) and type 1.5 diabetes [latent autoimmune diabetes in adults (LADA)] (n = 14) patients (P < 0.0001). We conclude that competition between insulin-specific monoclonal antibodies or their recombinant Fab and insulin autoantibodies should prove useful in the epitope analysis of autoantibodies to insulin.

Structural and functional analysis of cysteine residues in human glutamate decarboxylase 65 (GAD65) and GAD67

Previously, we have shown that brain glutamate decarboxylase (GAD) is greatly inhibited by sulfhydryl reactive reagent suggesting cysteine residue(s) may play an important role in GAD function. In this report, we determined the role of cysteine residues in the recombinant human 65-kDa GAD isoform (hGAD65) and 67-kDa GAD isoform (hGAD67), using a combination of matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) mass spectrometry and site-directed mutagenesis. Here, we report that cysteine 446 (C446) in hGAD65 is important for its activity and is present as free sulfhydryl group. This conclusion is based on the following observations: (i) mutation of C446 in hGAD65 to alanine reduced hGAD65 activity by more than 90%, (ii) MALDI-TOF analysis of the non-reduced, trypsin-digested GAD65 revealed that C446 is present as a free sulfhydryl group as indicated by a peak at m/z (mass/charge) 647.3446 (peptide 443-448) and, when GAD65 was treated with sulfhydryl reagent, N-ethylmaleimide (NEM), the peak is shifted to m/z 772.3702,a mass increase of 125.1 daltons (Da) as a result of modification of cysteine by NEM. Parallel studies have also been conducted with hGAD67. Cysteine 455 was found to be important for GAD67 activity.

Dynamic changes of GAD65 autoantibody epitope specificities in individuals at risk of developing type 1 diabetes

AIMS/HYPOTHESIS

Progression to type 1 diabetes is associated with intramolecular epitope spreading to disease-specific antibody epitopes located in the middle region of glutamic acid decarboxylase 65 (GAD65).

METHODS

The relationship between intramolecular epitope spreading of autoantibodies specific to GAD65 in relation to the risk of developing type 1 diabetes was tested in 22 high-risk individuals and 38 low-risk individuals. We determined the conformational epitopes in this longitudinal study by means of competition experiments using recombinant Fab of four GAD65-specific monoclonal antibodies.

RESULTS

Sera from high-risk children in the preclinical stage recognise a specific combination of GAD65 antibody epitopes located in the middle and the C-terminus of GAD65. High risk of progressing to disease is associated with the emergence of antibodies specific for conformational epitopes at the N-terminus and the middle region. Binding to already established antibody epitopes located in the middle and at the N-terminus increases and shows a significant relation (p=0.005) with HLA, which confers risk of developing diabetes.

CONCLUSIONS/INTERPRETATION

In type 1 diabetes, GAD65 antibodies are initially generated against the middle and C-terminal regions of GAD65. In genetically predisposed subjects the autoimmune response may then undergo intramolecular epitope spreading towards epitopes on the N-terminus and further epitopes located in the middle. These findings clearly demonstrate that the GAD65 autoantibody response in the preclinical stage of type 1 diabetes is dynamic and related to the HLA genotypes that confer risk of diabetes. GAD65-specific Fab should prove useful in predicting progression from islet autoimmunity to clinical onset of type 1 diabetes.

Autoimmune epitopes: autoepitopes

The identity of reactants for autoantibodies has been successively refined from whole cellular organelles (immunofluorescence), identified molecules (immunoblot; gene expression libraries), epitope regions (truncated cDNAs; peptide scanning) to contact residues, as described here. Most autoantibodies react with conformational epitopes, in which amino acids distant in the linear sequence come into contiguity by protein folding. Identification of contact sites with the antibody paratope requires particular technologies, crystallography, or antibody screening of phage-displayed random peptide libraries. The latter is illustrated by our studies on the autoepitope for anti-PDC-E2 (AMA) in primary biliary cirrhosis (PBC), anti-GAD65 in type 1 diabetes, and anti-C1 of type II collagen in collagen-induced arthritis. More precise definition of the structure of conformational autoepitopes could (a) clarify controversial aspects of autoimmunity including epitope mimicry, epitope spreading, and molecular spatial relationships between B and T cell autoepitopes, and (b) impact on novel diagnostic and therapeutic (vaccine) molecules.

Absence of avidity maturation of autoantibodies to the protein tyrosine phosphatase-like IA-2 molecule and glutamic acid decarboxylase (GAD65) during progression to type 1 diabetes

Immunoglobulin G avidity assays are used to distinguish between the acute and chronic phase of several infectious diseases, and there is evidence of autoantibody affinity maturation also in autoimmune diseases. To assess whether the analysis of the avidity of autoantibodies against the protein tyrosine phosphatase-like IA-2 molecule and glutamic acid decarboxylase (GAD65) could improve the accuracy of risk assessment of progression to clinical type 1 diabetes, we established methods for the determination of the autoantibody avidity based on our previously developed time-resolved fluorometric IA-2 and GAD65 autoantibody (IA-2A and GADA) assays. The avidity indices of sequential plasma samples from six IA-2A-positive and seven GADA-positive prediabetic children were analysed applying elution with urea and diethylamine (DEA). For comparison, corresponding avidity indices of control children, who have remained non-diabetic for at least 3 years after seroconversion to IA-2A and GADA positivity, were analysed. For most of the children, only a slight fluctuation in the avidity index values was observed over time, although the titres for IA-2A and GADA varied substantially in some cases. The avidity indices of the prediabetic children remained within the same range as those of the control group throughout the follow-up. Our results indicate that the analysis of the avidity index levels of IA-2A and GADA does not improve the accuracy of the prediction of type 1 diabetes based on autoantibody detection.

Multiplicity of the antibody response to GAD65 in Type I diabetes

Type I diabetes (TID) is an autoimmune disease characterized in part by the presence of autoantibodies directed against glutamic acid decarboxylase 65 (GAD65), among other pancreatic islet antigens. We investigated the independent epitope specificities of these GAD65 antibodies (GAD65Ab) and their combinations in the sera of new onset TID patients and first-degree relatives positive for GAD65Ab. For our analysis, we used four GAD65-specific recombinant Fabs (rFabs) that recognize different conformational determinants of GAD65 located throughout the molecule, including the N-terminal, the middle and the C-terminal regions. We used these epitope-specific rFabs in competition assays to determine the binding specificity of the autoantibodies found in patient sera. Among the 61 sera from newly diagnosed GAD65Ab-positive TID patients GAD65 binding was competed for 23 sera by all four rFabs, 29 by at least two rFabs, and in nine sera were displaced by one or no rFab. In contrast, none of the 24 sera from GAD65Ab-positive first-degree relatives of TID patients were displaced by all four rFabs. When using all four rFabs simultaneously to compete with GAD65Ab binding, binding of sera from TID patients was reduced by an average of 70%. A significantly weaker competition was observed when evaluating sera of GAD65Ab-positive first-degree relatives (P < 0.0001).

Epitope analysis of GAD65Ab using fusion proteins and rFab

The identification of disease-specific autoantibodies to the 65-kDa isoform of glutamate decarboxylase (GAD65Ab) epitopes in type 1 diabetes has been hampered by their conformational nature. Here, we compared two methods of GAD65Ab epitope analysis: GAD65/67 fusion proteins and competition assays using GAD65-specific recombinant fraction antigen binding (rFab). Sera from newly diagnosed type 1 diabetes patients (n=61) were studied using both approaches. Competition of GAD65 binding by an rFab to a specific epitope did not correlate with binding to the fusion protein that represented this epitope. Conversely, samples that bound to specific fusion proteins were not necessarily competed with rFab specific to determinants in the same region. We conclude that epitopes of different characteristics are detected by fusion proteins and by competition with rFab. Fusion proteins allow the definition of large epitope regions; however, some conformational GAD65Ab epitopes, especially those residing in the middle region, are destroyed or distorted in the fusion proteins. Competition studies using rFab allow the identification of conformational epitopes. However, monoclonal rFab may only reflect a limited proportion of the epitopes recognized by polyclonal sera. A combined analysis using both approaches may therefore be necessary to gain best understanding of autoantibody characteristics and affinity maturation.

Conformation-dependent GAD65 autoantibodies in diabetes

AIMS/HYPOTHESIS

Conformation-dependent autoantibodies directed against GAD65 are markers of Type 1 diabetes. In this study we aimed to determine whether the substitution of GAD65 with GAD67 amino acids would affect the binding of conformation-dependent GAD65 autoantibodies.

METHODS

We used PCR-based site-directed mutagenesis to generate a series of mutated GAD65 cDNA constructs in which specific GAD65 coding sequences for regions of the protein critical for autoantibody binding were replaced with GAD67 coding sequences.

RESULTS

The introduction of a point mutation at position 517, substituting glutamic acid with proline, markedly reduced the binding of disease-associated GAD65 antibodies. The binding of GAD65 antibodies to the E517P mutant was reduced in the sera of all newly diagnosed Type 1 diabetes patients ( n=85) by a mean of 72% ( p<0.0001) compared with binding to wild-type GAD65. Patients with latent autoimmune diabetes in adults ( n=24) showed a similar reduction in binding (79% reduction, p<0.0001). First-degree relatives who subsequently progressed to Type 1 diabetes ( n=12) showed a reduction in binding of 80% compared with a reduction of only 65% among relatives who had not progressed to disease ( n=38; p=0.025). In healthy GAD65Ab-positive individuals who did not progress to diabetes during a 9-year follow-up period ( n=51), binding to GAD65-E517P was reduced by only 28% compared with binding to wild-type GAD65.

CONCLUSIONS/INTERPRETATION

Differences in autoantibody binding to wild-type GAD65 versus GAD65-E517P may provide predictive information about Type 1 diabetes risk beyond that provided by the presence or absence of GAD65 autoantibodies. Lack of binding to mutant GAD65-E517P defines GAD65-positive individuals who are at higher risk of developing diabetes.

Mapping of epitopes for autoantibodies to the type 1 diabetes autoantigen IA-2 by peptide phage display and molecular modeling: overlap of antibody and T cell determinants

IA-2 is a major target of autoimmunity in type 1 diabetes. IA-2 responsive T cells recognize determinants within regions represented by amino acids 787-817 and 841-869 of the molecule. Epitopes for IA-2 autoantibodies are largely conformational and not well defined. In this study, we used peptide phage display and homology modeling to characterize the epitope of a monoclonal IA-2 Ab (96/3) from a human type 1 diabetic patient. This Ab competes for IA-2 binding with Abs from the majority of patients with type 1 diabetes and therefore binds a region close to common autoantibody epitopes. Alignment of peptides obtained after screening phage-displayed peptide libraries with purified 96/3 identified a consensus binding sequence of Asn-x-Glu-x-x-(aromatic)-x-x-Gly. The predicted surface on a three-dimensional homology model of the tyrosine phosphatase domain of IA-2 was analyzed for clusters of Asn, Glu, and aromatic residues and amino acids contributing to the epitope investigated using site-directed mutagenesis. Mutation of each of amino acids Asn(858), Glu(836), and Trp(799) reduced 96/3 Ab binding by >45%. Mutations of these residues also inhibited binding of serum autoantibodies from IA-2 Ab-positive type 1 diabetic patients. This study identifies a region commonly recognized by autoantibodies in type 1 diabetes that overlaps with dominant T cell determinants.

A comprehensive guide to antibody and T-cell responses in type 1 diabetes

Type 1 diabetes (T1D) is an organ-specific autoimmune disease in which the insulin-producing beta cells in the pancreatic islets are selectively eliminated. T cells specific for beta-cell antigens are the mediators of this precise cellular destruction. However, antibodies to beta-cell proteins are also generated and may be used for predicting disease in at-risk populations. Over the past two decades, numerous beta-cell proteins and lipids have been implicated as autoantigens in patients or in non-obese diabetic (NOD) mice, a well-studied animal model of T1D. Here, we present a review of these antigens, accompanied by their T-cell epitopes, where known, and a discussion of our current understanding of why particular self-proteins become disease-inciting antigens. Although two dozen beta-cell antigens have been identified to date, few of these have been confirmed to be recognized by pathogenic T cells early in the disease process. Further identification and characterization of initiating beta-cell antigens targeted by pathogenic T cells should be a priority for future studies.

Antibodies to GAD65 epitopes at diagnosis and over the first 10 years of clinical type 1 diabetes mellitus

Antibodies to glutamate decarboxylase (GAD65Ab) may persist, and their titres even increase after the clinical onset of type 1 diabetes. To characterize this phenomenon in detail, we analysed sequentially antibodies to GAD65 epitope clusters in a radio-binding assay in patients with type 1 diabetes. Serum samples were taken at diagnosis and 2, 5 and 10 years later from 50 young patients who had tested positive for GAD65Ab at least once during observation. The levels of GAD65Ab peaked in 21 patients after diagnosis. Antibodies to the middle region of GAD65 (GAD65-M-Ab, 88%) were more common at diagnosis than antibodies to the C-terminal (GAD65-C-Ab, 68%, P < 0.05) or N-terminal region (4%, P < 0.001). Antibodies to middle and especially to C-terminal epitopes decreased in those with decreasing levels of GAD65Ab (P < 0.001), whereas the frequencies of GAD65-M-Ab and GAD65-C-Ab remained quite stable among the subjects with increasing levels. Lower exogenous insulin dose and HbA(1) levels and stronger humoral immune response to islet cells were observed in those with increasing levels of GAD65-M-Ab than in those with decreasing levels (P < 0.05). The present observation supports the view that the middle region of GAD65 comprises immunodominant epitopes. An enhanced humoral immune response to GAD65 after diagnosis is related to persistent immune reactivity to the middle and C-terminal regions.