help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maioli, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maioli, M.
The Journal of Clinical Endocrinology & Metabolism Vol. 89, No. 11 5675-5682
Copyright © 2004 by The Endocrine Society

Epitope-Restricted 65-Kilodalton Glutamic Acid Decarboxylase Autoantibodies among New-Onset Sardinian Type 2 Diabetes Patients Define Phenotypes of Autoimmune Diabetes

Mario Maioli, Emilyn Alejandro, Giancarlo Tonolo, Lisa K. Gilliam, Lynn Bekris, Christiane S. Hampe, Domenica A. Obinu, Alberto Manconi, Loreta Puddu, Kristian Lynch and Åke Lernmark for the Study Group for the Genetics of Diabetes in Sardinia

Metabolic Disease Unit (M.M., G.T., D.A.O., A.M., L.P.), Istituto di Clinica Medica, University of Sassari, Sassari 07100, Italy; and Departments of Medicine (E.A., L.K.G., L.B., C.S.H., .L.) and Biostatistics (K.L.), University of Washington, Seattle, Washington 98195

Address correspondence and requests for reprints to: Åke Lernmark, University of Washington, Department of Medicine, R. H. Williams Laboratory, Box 357710, 1959 N.E. Pacific Street, Seattle, Washington 98195-7710. E-mail: ake{at}u.washington.edu.


   Abstract
Top
 Abstract
Introduction
 Subjects and Methods
 Results
 Discussion
 References
 
The 65-kDa glutamic acid decarboxylase (GAD65) autoantibodies (GAD65Abs), commonly found in type 1 diabetes mellitus (T1DM) patients, are also found at lower frequencies in type 2 diabetes mellitus (T2DM) patients. GAD65Abs in T1DM patients are epitope specific, in contrast to those found in other GAD65Ab-positive individuals, including T2DM patients. Our aim was to assess whether epitope-specific GAD65Abs, or the additional presence of islet antigen 2 (IA-2) autoantibodies, better define T1DM phenotypes among T2DM patients. GAD65 and IA-2 autoantibodies were analyzed in 1436 Sardinian subjects classified with T2DM and in 384 nondiabetic patient controls. Autoantibody binding specificity to the N-terminal, middle (M), and C-terminal (C) portions of the GAD65 molecule was evaluated. Among the T2DM patients, 5.1% had GAD65 (P < 0.001) and 2.4% had IA-2 autoantibodies, compared with 1.3 and 1.6%, respectively, among the controls. GAD65Ab-positive T2DM patients with M+C (epitope-specific) reactivity were found to have the lowest body mass index (P < 0.001), followed by GAD65Ab/IA-2Ab-positive patients (P < 0.01), and non-M+C-reactive (non-epitope-specific) patients (P < 0.02). In GAD65Ab-positive T2DM patients, c-peptide levels were lower in M+C-reactive compared with non-M+C-reactive patients. Sardinian T2DM patients with M+C-predominant GAD65Ab reactivity have clinical features more similar to those of T1DM patients. Thus, GAD65Ab epitope analysis may help to define T1DM phenotypes among newly diagnosed GAD65Ab-positive patients classified with T2DM.


   Introduction
Top
 Abstract
 Introduction
Subjects and Methods
 Results
 Discussion
 References
 
TYPE 1 DIABETES MELLITUS (T1DM) is an autoimmune disease characterized by the specific destruction of the insulin-producing pancreatic ß-cells. Most T1DM patients have circulating autoantibodies (Abs) directed against one or more islet cell autoantigens: 65-kDa glutamic acid decarboxylase (GAD65) (1, 2, 3), insulin (4), and islet antigen 2 (IA-2) (5), also known as islet cell Ab 512 (ICA512) (6, 7). GAD65Ab and IA-2Ab are readily detected by standardized (8, 9), precise, and reproducible RIAs (3, 8, 10) suitable for large-scale analysis and population screening (11, 12).

ICA-positive type 2 diabetes mellitus (T2DM) patients were first reported in 1977 (13), and ICA positivity in these patients was found to predict secondary failure to oral hypoglycemic agents (14). The presence of ICA in adult type 2 patients was associated with T1DM-associated high-risk human leukocyte antigen-DR alleles (15), appeared to identify patients with features of T1DM (16, 17), and predicted earlier insulin dependency (18). Similarly, GAD65Ab found in T2DM patients (19, 20) identified patients with insulin deficiency and clinical features of T1DM (21, 22, 23, 24) and predicted loss of ß-cell function (25, 26, 27). These GAD65Ab-positive T2DM patients are referred to as latent autoimmune diabetes in adults (LADA) (28) or type 1.5 diabetes (29), and the interest in these patients is increasing in the wake of an ensuing worldwide epidemic of T2DM (30). Screening patients classified with T2DM for GAD65Ab has been advocated to improve diagnostic accuracy and to treat patients based on disease etiology, rather than clinical symptoms (25).

It is widely recognized that the ethnic diversity may influence the prevalence of GAD65Ab in different T2DM populations. GAD65Ab prevalence in T2DM patients varies from 1–21%, depending on the population studied. American native populations have a low prevalence of GAD65Ab, with rates of 1% in Alaskan native populations (31) and 2% in Pima Indians (32). European and Asian populations have a higher prevalence, with reported rates of 2.8% in northern Italy (33), 3.7% in Southern Spain (34), 3.8% in Japan (35), and 10% in Sweden (19, 25) and Finland (20, 22, 36). In the United Kingdom Prospective Diabetes Study, the prevalence of GAD65Ab positivity was age dependent, with 4% of older subjects found to be GAD65Ab positive, in contrast to 21% of younger subjects (27). It is likely that genetic factors that confer susceptibility to autoimmunity play an important role in the differences in prevalence of T1DM-associated autoantibodies in different populations.

Although approximately 1% of the general population has GAD65Ab, only a fraction of these subjects will develop T1DM, which supports the notion that GAD65Ab alone is not a good predictor for the disease. In support of this concept, population-based screening studies have indicated that not all types of GAD65Ab predict T1DM (37, 38). As we (39, 40) and others (23, 41, 42) have shown, T1DM patients recognize disease-specific GAD65Ab epitopes, and we found that GAD65Ab epitope patterns can be used to differentiate LADA from T1DM patients (39, 43). We recently reported a high frequency of GAD65Ab and IA-2Ab among type 2 diabetes siblings from Sardinian multiplex families (44). This observation prompted us in the present study to determine the prevalence and levels of GAD65Ab and IA-2Ab in consecutively diagnosed T2DM patients ascertained from all provinces of Sardinia, a country known for its high incidence of T1DM (45). In addition, within the cohort of GAD65Ab-positive T2DM patients, we tested the hypothesis that epitope-specific GAD65Ab or the additional presence of IA-2Ab better define T1DM phenotypes.


   Subjects and Methods
Top
 Abstract
 Introduction
 Subjects and Methods
Results
 Discussion
 References
 
We studied all consecutive unrelated individuals, 35–75 yr of age (Table 1Go), who for the first time attended a Diabetic Unit in any of the four provinces of Sardinia: Cagliari, Oristano, Nuoro, and Sassari from 1998–1999. During this study period, 1820 subjects were enrolled in the present study. The majority of patients diagnosed with T2DM were seen in these diabetic referral centers; therefore, this cohort provides a fairly representative view of the overall population with newly diagnosed diabetes in Sardinia. All subjects had a complete health examination, which included fasting blood glucose determinations on two different occasions, and if blood glucose was in the nondiabetic (ND) range (<140 mg/dl), a standard 75-g oral glucose tolerance test (OGTT). Blood glucose was measured by standard clinical assays. A total of 1436 subjects were diagnosed with diabetes. Of these, 785 had clinical symptoms of diabetes at their first visit, and confirmation of their diagnosis was based on fasting blood glucose greater than or equal to 140 mg/dl on two separate occasions. Another 651 subjects underwent OGTT testing and were diagnosed with diabetes based on blood glucose levels greater than or equal to 200 mg/dl after 2 h. In these subjects, the OGTT was repeated for confirmation within 2 wk. Only subjects classified with T2DM according to World Health Organization (WHO) criteria or with normal glucose tolerance (ND) were enrolled in the study. Subjects with impaired glucose tolerance were excluded. Blood samples for evaluation of Abs and epitope analyses were collected during this initial period at the time of diabetes diagnosis. In addition, the T2DM patients enrolled and subsequently followed in the diabetes clinics had to fulfill these additional criteria: no insulin requirement within 6 months after diagnosis, no clinical symptoms of T1DM such as marked weight loss or acidosis, and no family history of T1DM among first-degree relatives. In the ND subjects, diabetes was excluded based on a normal OGTT, according to WHO criteria (blood glucose level at 120 min < 140 mg/dl). These ND subjects did represent a higher-risk population, based on the fact that they presented or were referred to the diabetes units to rule out a diagnosis of diabetes. According to these criteria, we enrolled 1436 T2DM patients and 384 ND subjects (Table 1Go). In patients positive for GAD65Ab, plasma fasting C-peptide was measured using a commercial kit (C-peptide-ctk; DiaScrip S.r.l., Saluggo, Italy). Informed consent was obtained from all study subjects. The investigations were approved by the institution’s human research committee and carried out in accordance with the principles of the Declaration of Helsinki, as revised in 2000.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Age and gender of Sardinian T2DM patients and ND control subjects

 
GAD65Ab and IA-2Ab assays

Serum samples obtained at the first visit were analyzed for GAD65Ab (3, 46) and IA-2Ab (7) in radioligand binding assays. [35S]Methionine-labeled GAD65, IA-2, GAD67 (47), and GAD65/67 fusion proteins (46) were produced by in vitro coupled transcription/translation with SP6 RNA polymerase and nuclease-treated rabbit reticulocyte lysate (Promega, Madison, WI) as previously described (3). The in vitro translated proteins were kept at –80 C and used within 2 wk of preparation. The upper level of normal was established by analyzing sera obtained from ND Sardinian subjects (n = 1184) used as controls for both GAD65Ab and IA-2Ab (44). The subjects represent 493 schoolchildren from Nuoro, 204 individuals without a family history of diabetes but older than 50 yr of age (blood glucose < 140 mg/dl at 120 min during OGTT), and 487 nondiabetic individuals, 12–80 yr of age who also were found to have a normal OGTT (44).

The interassay coefficient of variation for the positive control sample (WHO standard 97/550) (8) was 15% for GAD65Ab and 9% for IA-2Ab. The intraassay coefficient of variation for triplicate determinations was 10% for GAD65Ab, 8% for IA-2Ab, 10% for GAD67Ab, and 12% for the different GAD65/67 fusion proteins. In the third International Combined Autoantibody Workshop (10), our assay showed 70% sensitivity and 98% specificity for GAD65Ab as well as 47% sensitivity and 98% specificity for IA-2Ab.

Ab levels

The Ab levels are expressed as a relative index, as described in detail elsewhere (3) using the WHO standard 97/550 (8) as the positive control. Briefly, the relative index of each sample was calculated using the mean of two negative controls and the WHO 97/550 standard (8). The 99th percentile of the control group (n = 1184), previously shown not to be normally distributed for GAD65Ab and IA-2Ab levels, determined the cutoff for positivity (44). These indices were 0.1 and 0.034 for GAD65Ab and IA-2Ab, respectively. Samples identified as GAD65Ab positive were further analyzed with GAD67 and GAD65/67 fusion molecules to determine Ab binding characteristics to different epitopes.

Construction of fusion molecules

The GAD65, GAD67, and GAD65/67 cDNA fusion molecules used in the present study are summarized in Table 2Go. Full-length GAD65 (2, 3) and GAD67 (47) cDNA clones were used to prepare fusion proteins by substituting GAD65 sequences with corresponding regions of GAD67 (39, 46). Type 2 diabetes patients with a GAD65Ab index of greater than 0.1 were analyzed for Abs against GAD67 and three GAD65 variants (N, M, and C) (Table 2Go). It should be noted that the designation N, M, and C are particular to the present fusion proteins, which differ in design from constructs used in previous studies performed by our group (40, 48) and others (23, 49, 50).


View this table:
[in this window]
[in a new window]
  		TABLE 2. GAD65, GAD67, and GAD65/67 fusion proteins used to analyze GAD65Ab epitopes

 
Statistical evaluation

Differences in age-specific prevalence between T2DM patients and ND subjects were tested using {chi}2 statistics with Yates correction. To increase power in detecting higher antibody prevalence among T2DM patients, antibody frequencies among ND subjects were assumed constant across age. This assumption was verified using exploratory tables and {chi}2 tests. Confidence intervals were calculated using large-sample normal approximation.

The goal of the epitope analysis was to divide GAD65Ab-positive T2DM patients into different epitope pattern groups. Patients who showed significant binding to GAD65 and each fusion protein were classified together as one group. Separation of fusion protein binding into discrete positive or negative samples was avoided as it reduced the power and resolution to detect differences in epitope pattern. Instead, binding index distributions were normalized by shifting distributions to ensure index values were above zero and by taking the natural logarithm transformation. The relationship between Ab binding to GAD65 and each fusion protein, after adjustment for GAD67, was examined using partial correlation coefficients (51). Scatterplots were used to check that all significant associations were not influenced by outliers. When epitope pattern groups were identified, distribution plots were used to examine the body mass index (BMI) and log C-peptide within each group. Wilcoxon two-sample test was used to test for differences in BMI or log C-peptide between groups.

Multiple regression models were used to estimate age-adjusted changes in average BMI in Ab-positive type 2 diabetic subgroups. Statistical analysis was performed using Sigma STAT and S-PLUS 6.1.


   Results
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
Discussion
 References
 
Prevalence of Abs

The prevalence of GAD65Ab and IA-2Ab was determined in 1436 T2DM patients and 384 ND subjects, and these values were evaluated in different age groups in the diabetic subjects (Table 3Go). Cutoffs for GAD65 and IA-2 positivity were based on the 99th percentile of the 1184 Sardinian control subjects. The overall prevalence of GAD65Ab in T2DM subjects was 5.1% [95% confidence interval (CI), 3.9–6.2%] and did not vary significantly between age groups. The prevalence of GAD65Ab in T2DM subjects (5.1%) was significantly higher than that found in ND subjects (1.3%). Index level distribution revealed a wide spectrum of GAD65Ab levels in the patients classified with T2DM (Fig. 1AGo).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Number of Ab-positive individuals (n) and prevalence (%) (95% CI) of GAD65Ab alone, IA-2Ab alone, and GAD65-IA-2 double-antibody-positive by age in 1436 newly diagnosed Sardinian adults classified with T2DM

 


View larger version (21K):
[in this window]
[in a new window]
  		FIG. 1. Levels of GAD65Ab (GAD65Ab index) (A) and IA-2Ab (IA-2Ab index) (B) plotted against BMI in 1436 patients classified with T2DM and 384 ND subjects. The vertical dotted line marks a BMI of 25 kg/m2.

 
Of the 1436 T2DM patients, 34 (2.4%) were IA-2Ab positive. The IA-2Ab frequencies in ND (1.6%) and T2DM subjects (2.4%) were not significantly different, and the IA-2Ab prevalence was not affected by age at diagnosis of the T2DM patients (Table 3Go). Similar to that seen with GAD65Ab, the distribution of IA-2Ab indices was variable in these subjects (Fig. 1BGo).

None of the ND subjects were GAD65Ab and IA-2Ab double positive. In contrast, nine T2DM patients were positive for both GAD65Ab and IA-2Ab. IA-2Ab index was significantly higher in these double-antibody-positive patients [0.87 (95% CI, 0.05–1.56)] compared with patients who were positive only for IA-2Ab [0.07 (95% CI, 0.04–0.58)] (P < 0.001).

GAD65Ab epitope patterns

A total of 52 of 73 sera from GAD65Ab-positive T2DM patients were available for the epitope analysis. Serum samples that bound GAD65 also bound the fusion proteins N, M, or C (P < 0.0001). Binding to both GAD65 and GAD67 was strongly correlated (r = 0.62; P < 0.0001). The median (range) index for GAD67 binding was 0.35 (0.03–5.94), for N was 0.37 (0.04–6.7), for M was 0.70 (0.04–10.4), and for C was 0.63 (0.03–20.5). Correction for the level of GAD67 binding showed that autoantibody binding to fusion proteins M and C still correlated to GAD65 binding but not to N (Table 4Go). A comparison of binding between fusion proteins confirmed that a linear relationship between binding to the M and C fusion proteins was observed but that binding to the N-fusion protein was independent from the binding to fusion proteins M and C (Table 4Go).


View this table:
[in this window]
[in a new window]
  		TABLE 4. Relationship between GAD65Ab binding to GAD65 and N-terminal, middle, or C-terminal, fusion proteins in 52 T2DM patients positive for GAD65Ab

 
The distribution binding of each fusion protein relative to GAD65 showed higher binding of sera to the M (P < 0.0001) and the C (P < 0.0001) fusion proteins compared with the N-fusion protein. Based on these findings, and based on previously published findings that GAD65Ab binding to the middle region correlates with the binding to the C-terminal region of the GAD65 protein (46), it was decided to group together patients whose sera bound predominantly to the M and C fusion proteins, reporting the two epitopes together as M+C. Sera were considered M+C reactive if they bound the M and C fusion proteins better than the N fusion protein. Two patients bound GAD67 similar to GAD65 and were not considered GAD65 M+C reactive. The degree of M+C binding over N binding, within the M+C-reactive group (n = 32), varied from patient to patient. A comparison of binding levels showed the N fusion protein and GAD67 binding levels to be similarly low but M+C binding levels to be significantly higher. The degree of M+C reactivity was thus dependent on M and C binding and not on binding levels to the N fusion protein or to GAD67.

By a similar analysis of the remaining 20 (non-M+C-reactive) patients, only the N-fusion protein was shown to be related to GAD65Ab binding after adjusting for GAD67 binding. Scatterplot analysis revealed the relationship to be strongly influenced by a few patients showing high binding to the N fusion protein, but little or no binding to the M or C fusion proteins. For simplicity, all these patients were grouped together and considered non-M+C-reactive.

IA-2Abs in relation to GAD65Ab epitopes

Among 73 GAD65Ab-positive patients, seven were IA-2Ab positive and M+C GAD65Ab reactive, two were IA-2Ab positive and non-M+C GAD65Ab reactive, 25 were IA-2 negative and M+C GAD65Ab reactive, and 18 were IA-2Ab negative and non-M+C reactive. Another 21 T2DM subjects were GAD65Ab positive, but epitope data were not available. None of these GAD65Ab-positive patients lacking epitope data were IA-2Ab positive, and 81% (17 of 21) had low levels of GAD65Ab (index < 0.2), precluding an epitope analysis with GAD67 and all three fusion proteins. Among the GAD65Ab-negative patients (n = 1363), 23 T2DM subjects were positive for IA-2Ab alone.

Abs and BMI

BMI was available on all 1436 T2DM patients. A linear regression analysis of each Ab on BMI demonstrated that IA-2Ab-positive patients had a lower average BMI compared with IA-2-negative patients (Table 5Go). Univariate analysis also showed a lower BMI in GAD65Ab M+C (P < 0.001) and non-M+C-reactive patients (P < 0.009). A multivariate analysis showed that both IA-2Ab and GAD65Ab were independently associated with lower BMI, even after adjusting for age and gender (Table 5Go). BMI was lowest in GAD65Ab M+C-reactive patients. A significantly lower BMI was observed in the IA-2Ab/GAD65Ab double-positive patients, compared with patients who were positive for only IA-2Ab (P < 0.01) or positive for only GAD65Ab (P < 0.05).


View this table:
[in this window]
[in a new window]
  		TABLE 5. Linear regression analysis of IA-2Ab and GAD65Ab, respectively, on BMI (kg/m2), is shown for Ab-positive compared with Ab-negative patients

 
A bimodal distribution in BMI was observed within the group of GAD65Ab-positive patients with available epitope data. Among the M+C-reactive patients, the median BMI was close to normal (25.0 kg/m2); however, the BMI distribution was skewed to the right, with a small subgroup having a BMI greater than 30 kg/m2 (Fig. 2AGo). Among the non-M+C-reactive group, the median BMI was higher (28.9 kg/m2), but the BMI distribution was skewed to the left (Fig. 2BGo). The median BMI of the M+C-reactive patients (25.0 kg/m2) was significantly different from that of the non-M+C-reactive patients (28.9 kg/m2) (P = 0.02).



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 2. Distribution of BMI and log C-peptide in 52 newly diagnosed T2DM patients who exhibited different GAD65Ab epitope binding. These patients were all positive for GAD65Ab. IA-2Ab were also determined. The patients shown in A (n = 32) and in C (n = 23) had predominant M+C GAD65Ab reactivity. Patients in B (n = 20) and in D (n = 17) were non-M+C reactive. The vertical line in A and C marks a BMI of 25 kg/m2 and in B and D marks a median log C-peptide of 0.4 log ng/ml.

 
Abs and C-peptide

C-peptide was available on 40 of 52 GAD65Ab-positive patients with available epitope data and on all nine patients with both GAD65Ab and IA-2Ab. Patients with available C-peptide did not differ in age, gender, or BMI from the patients with missing C-peptide. Distribution of log C-peptide within the epitope groups showed very low C-peptide among a subgroup of M+C-reactive patients (Fig. 2CGo) but normal levels and a normal distribution among the non-M+C-reactive patients (Fig. 2DGo). Among the M+C-reactive patients, the subgroup with very low C-peptide (n = 6) did not differ from the group with higher C-peptide (n = 17) by age, BMI, or gender. A significant difference between M+C and non-M+C patients was observed in the median log C-peptide (Table 6Go) both in the more overweight patients (BMI > 28 kg/m2) and in the less overweight and normal weight patients (BMI < 28 kg/m2). No difference in C-peptide was observed between the GAD65-IA-2Ab double-positive (n = 9; all had BMI < 28) and the GAD65Ab-positive (IA-2Ab-negative) (n = 31) patients. Therefore, regardless of the patient’s BMI, C-peptide was lower in M+C-reactive GAD65Ab patients compared with non-M+C-reactive patients, but no difference was found in C-peptide levels based on IA-2Ab status (Table 6Go). Although some IA-2Ab-negative but GAD65Ab M+C-positive patients were obese, they had lower C-peptide levels, suggestive of a T1DM phenotype.


View this table:
[in this window]
[in a new window]
  		TABLE 6. Median log C-peptide by IA-2Ab, epitope, and overweight status among 40 GAD65Ab-positive T2DM patients

 

   Discussion
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
References
 
The results of our study suggest that the prevalence of GAD65Ab but not IA-2Ab is higher in newly diagnosed diabetic Sardinian adults classified clinically with T2DM, compared with a control group of nondiabetic subjects. Although the frequency of IA-2Ab was not higher among the T2DM patients compared with the ND group, it is noted that some GAD65Ab-positive T2DM patients were also positive for IA-2Ab, suggesting a disease pathogenesis more consistent with type 1 rather than T2DM. Double-antibody-positive first-degree relatives have previously been shown to be at high risk for T1DM (7, 52), but to our knowledge this has not previously been reported in a large group of adult patients clinically classified with T2DM. Consistent with a T1DM phenotype, these patients had a lower BMI compared with Ab-negative patients.

The prevalence of GAD65Ab positivity in our Sardinian T2DM patient population was intermediate between low-frequency T2DM populations such as Alaskan natives (31), Pima Indians (32), northern Italy (33), and southern Spain (34) and high-incidence Northern Europeans (19, 20, 25, 27, 36). The reason for a lower frequency of GAD65Ab in Sardinian T2DM patients in relation to that seen in other T2DM populations remains unclear. In part, it may be explained by the age range of the patients and the careful exclusion of subjects with features of T1DM, such as marked weight loss, acidosis, first-degree relatives with T1DM, and insulin treatment started within 6 months after diagnosis. It is also possible that inter-laboratory variation in assay sensitivity and the choice of cutoffs for assay positivity across laboratories is another potential reason for the difference in prevalence of GAD65Ab positivity noted between this study and others. The 1436 newly diagnosed T2DM patients were consecutively diagnosed during 1998–1999 and compared with 384 ND subjects identified during the same observational period. Subjects with diagnosis of impaired glucose tolerance were excluded from the study. It should be noted that the ND subjects were not a true population-based control group, as they had presented to the diabetes units for evaluation, presumably either referred by their physicians, or self-referred, because of concern or risk for diabetes. However, these subjects do serve as a good control group in this study, as they had similar risk factors prompting referral (family history of T2DM, cardiovascular disease, hypertension, dyslipidemia, and obesity), and differed only in the fact that they had normal glucose tolerance. Of note, the prevalence of GAD65Ab and IA-2Ab among the ND subjects was not higher compared with the larger reference group of 1184 control subjects. As expected, patients diagnosed with diabetes had overall higher GAD65Ab frequencies, but not IA-2Ab frequencies, compared with the ND subjects. We had expected a higher frequency of GAD65Ab also in the ND group because GAD65Ab have been shown to be associated with a higher BMI in adult nondiabetic individuals (11, 53). Other factors than T2DM risk may therefore influence the appearance of GAD65Ab in the adult population. It is important to note that GAD65Ab may appear after the clinical diagnosis of diabetes (25, 54), they may be transient (25), and their presence in the general adult population does not predict the development of diabetes (37). We therefore asked whether the epitope pattern of GAD65Ab would distinguish GAD65Ab with little or no T1DM risk from GAD65Ab associated with risk for this disease.

T2DM patients with GAD65Ab, IA-2Ab, or ICA, referred to as LADA or type 1.5 diabetes, are now well recognized and often show distinct phenotypes indicating a slow progression to T1DM (25, 27, 55). Although T1DM is associated with specific genetic and Ab markers (52, 56, 57), the strength of the association between the disease and these markers has previously been shown to vary with age (58). The age of our T2DM patients varied between 35 and 75 yr of age. Of note, in this study the frequency of GAD65Ab in the 35- to 55-yr-old patients did not differ from the patients older than 55 (Table 3Go). In addition, we studied new-onset T2DM patients (<6 months of disease), and our data on GAD65Ab epitope patterns can therefore be compared with previous studies by our group (39, 46) and by others (23) analyzing both younger and hospital-based T2DM patients.

Similar to T1DM patients, type 1.5 diabetes patients have GAD65Ab that bind to the middle and C-terminal Ab epitope of GAD65, but these patients also have a higher frequency of Abs that bind to the N-terminal end of GAD65 as well as to GAD67 (23, 39). Therefore, in this study we used GAD65/67 fusion proteins along with GAD67 to allow us to distinguish a pattern of GAD65 epitope binding in relation to the clinical phenotype of our new-onset T2DM patients. The fusion proteins used in this study were designed to represent the major GAD65 epitopes (50) and differed considerably from those used by others (23). We defined sera as predominantly M+C when their relative recognition of M- and C-fusion proteins was greater than their relative binding to the N-fusion protein and GAD67. The univariate and multivariate regression models of changes in average BMI showed that the epitope pattern of GAD65Ab predict a lower BMI compared with patients who were non-M+C reactive or had IA-2Ab alone. These data support the observation that the presence of GAD65Ab binding to C-terminal epitopes in individuals clinically classified with T2DM is strongly associated with a T1DM phenotype (23). GAD65Ab binding to the N-fusion protein is comparable to the binding pattern found in ND GAD65Ab-positive adults as well as in some LADA patients (39, 46). Our data therefore confirm the notion that GAD65Ab reactive with M+C epitope would be consistent with a T1DM phenotype (39, 59). At present we are following all of our patients over time to evaluate the progression to insulin treatment in the different GAD65Ab-positive and -negative groups and whether there will be a predictive value of or a change in GAD65Ab epitope pattern associated with the disease course.

We also demonstrated that the presence of IA-2Ab predicts a lower BMI, particularly in the IA-2Ab/GAD65Ab double-positive patients. However, unlike GAD65Ab M+C reactivity, IA-2 positivity was not associated with a lower median C-peptide. These data suggest that, like GAD65Ab M+C reactivity, the additional presence of IA-2Ab better defines T1DM phenotypes among T2DM patients.

We conclude that a significant proportion (5.1%) of newly diagnosed Sardinian diabetes patients older than 35 yr and clinically classified with T2DM have GAD65Ab. GAD65Ab epitope analysis showed variable reactivity to GAD65, GAD67, and GAD65/67 fusion proteins that define binding to the N-terminal, middle (M) and the C-terminal epitopes of GAD65. Statistical modeling of the binding data from all test antigens revealed that patients with predominant M+C reactivity predicted a T1DM phenotype with lower C-peptide levels, regardless of age, BMI, and IA-2Ab status. These results suggest that all GAD65Ab-positive adults with clinical T2DM may not undergo the same underlying autoimmune process. It is possible that there is a different, nonpathogenic, non-ß-cell destructive process generating GAD65Abs in type 2 patients who do not exhibit M+C epitope specificity, compared with a pathogenic, more T1DM-like process occurring in the T2DM patients who do exhibit GAD65Ab M+C epitope specificity. Outcome measures, such as time to insulin requirement and rate of loss of C-peptide, will be necessary to support this hypothesis. In conclusion, our results support the notion that new-onset diabetes patients in different populations will likely benefit from evaluation for epitope-specific GAD65Ab. Although the prevalence of M+C-specific GAD65Ab is low among non-T1DM patients, the evaluation of epitope-specific GAD65Ab will permit correct disease classification and may help the treating physician to optimize initial treatment and predict disease course. Furthermore, should effective immunotherapies become available, this will likely be an important target population to identify for treatment trials.


   Acknowledgments
 
We acknowledge the members of The Study Group for the Genetics of Diabetes in Sardinia-Sassari: G. Tonolo (coordination), M. Ciccarese, M. G. Melis, M. F. Angius, A. Carboni, G. Secchi, and M. Maioli; Nuoro: G. Pala and N. Pintori; Lanusei: A. Massidda and G. Meloni; Cagliari-1: M. Manai and R. M. Pilosu; Cagliari-2: R. Cirillo and E. Cossu; Cagliari-3: S. Lostia and G. Tocco; Bosa: G. Idda; Isili: M. Pisano; Sorgono: A. Gigante; Oristano: M. Mastinu; Olbia: F. Sanciu; and Ozieri: E. Secchi and S. Loddoni. We thank Sue Blaylock and Terri Daniels for assistance.


   Footnotes
 
This work was supported in part by grants from Assessorato Regionale della Sanita’, Regione Sardegna for the project entitled "Il Diabete mellito non insulino dipendente in Sardegna," the National Institutes of Health (DK26190, DK53004, and DK17047), and the Swedish Research Council, and by the American Diabetes Association (Career Development Award to C.S.H.).

M.M. and E.A. contributed equally to this study.

Abbreviations: Ab, Autoantibody(ies); BMI, body mass index; CI, confidence interval; GAD65, 65-kDa glutamic acid decarboxylase; IA, islet antigen; ICA, islet cell Ab; LADA, latent autoimmune diabetes in adults; ND, nondiabetic; OGTT, oral glucose tolerance test; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Received May 14, 2004.

Accepted August 16, 2004.


   References
Top
 Abstract
 Introduction
 Subjects and Methods
 Results
 Discussion
 References
 

  1. Baekkeskov S, Aanstoot HJ, Christgau S, Reetz A, Solimena M, Cascalho M, Folli F, Richter-Olesen H, De Camilli P 1990 Identification of the 64K autoantigen in insulin-dependent diabetes as the GABA-synthesizing enzyme glutamic acid decarboxylase. Nature 347:151–156[CrossRef][Medline]
  2. Karlsen AE, Hagopian WA, Grubin CE, Dube S, Disteche CM, Adler DA, Bärmeier H, Mathewes S, Grant FJ, Foster D, Lernmark Å 1991 Cloning and primary structure of a human islet isoform of glutamic acid decarboxylase from chromosome 10. Proc Natl Acad Sci USA 88:8337–8341[Abstract/Free Full Text]
  3. Grubin CE, Daniels T, Toivola B, Landin-Olsson M, Hagopian WA, Li L, Karlsen AE, Boel E, Michelsen B, Lernmark Å 1994 A novel radioligand binding assay to determine diagnostic accuracy of isoform-specific glutamic acid decarboxylase antibodies in childhood IDDM. Diabetologia 37:344–350[Medline]
  4. Palmer JP, Asplin CM, Clemons P, Lyen K, Tatpati O, Raghu PK, Paguette TL 1983 Insulin antibodies in insulin-dependent diabetics before insulin treatment. Science 222:1337–1339[Abstract/Free Full Text]
  5. Lan MS, Lu J, Goto Y, Notkins AL 1994 Molecular cloning and identification of a receptor-type protein tyrosine phosphatase, IA-2, from human insulinoma. DNA Cell Biol 13:505–514[Medline]
  6. Rabin DU, Pleasic SM, Shapiro JA, Yoo-Warren H, Oles J, Hicks JM, Goldstein DE, Rae PM 1994 Islet cell antigen 512 is a diabetes-specific islet autoantigen related to protein tyrosine phosphatases. J Immunol 152:3183–3187[Abstract]
  7. Verge CF, Gianani R, Kawasaki E, Yu L, Pietropaolo M, Jackson RA, Chase HP, Eisenbarth GS 1996 Prediction of type I diabetes in first-degree relatives using a combination of insulin, GAD, and ICA512bdc/IA-2 autoantibodies. Diabetes 45:926–933[Abstract]
  8. Mire-Sluis AR, Das RG, Lernmark Å 2000 The World Health Organization International Collaborative Study for Islet Cell Antibodies. Diabetologia 43:1282–1292[CrossRef][Medline]
  9. Bingley PJ, Bonifacio E, Mueller PW 2003 Diabetes Antibody Standardization Program: first assay proficiency evaluation. Diabetes 52:1128–1136[Abstract/Free Full Text]
  10. Verge CF, Stenger D, Bonifacio E, Colman PG, Pilcher C, Bingley PJ, Eisenbarth GS 1998 Combined use of autoantibodies (IA-2 autoantibody, GAD autoantibody, insulin autoantibody, cytoplasmic islet cell antibodies) in type 1 diabetes: Combinatorial Islet Autoantibody Workshop. Diabetes 47:1857–1866[Abstract]
  11. Rolandsson O, Hägg E, Hampe C, Sullivan EP, Nilsson M, Jansson G, Hallmans G, Lernmark Å 1999 Levels of glutamate decarboxylase (GAD65) and tyrosine phosphatase-like protein (IA-2) autoantibodies in the general population are related to glucose intolerance and body mass index. Diabetologia 42:555–559[CrossRef][Medline]
  12. Williams AJ, Bingley PJ, Moore WP, Gale EA 2002 Islet autoantibodies, nationality and gender: a multinational screening study in first-degree relatives of patients with type I diabetes. Diabetologia 45:217–223[CrossRef][Medline]
  13. Irvine WJ, McCallum CJ, Gray RS, Campbell CJ, Duncan LJP, Farquhar JW, Vaughan H, Morris PJ 1977 Pancreatic islet cell antibodies in diabetes mellitus correlated with the duration and type of diabetes, coexistent autoimmune disease and HLA-type. Diabetes 26:138–147[Abstract]
  14. Irvine WJ, Sawers JS, Feek CM, Prescott RJ, Duncan JP 1979 The value of islet cell antibody in predicting secondary failure of oral hypoglycemic agent therapy in diabetes mellitus. J Clin Immunol 2:23–26
  15. Gleichmann H, Zörcher B, Greulich B 1984 Correlation of islet cell antibodies and HLR-DR phenotypes with diabetes mellitus in adults. Diabetologia 27:90–92
  16. Groop LC, Bottazzo GF, Doniac D 1986 Islet cell antibodies identify latent type 1 diabetes in patients aged 35–75 years at diagnosis. Diabetes 35:237–241[Abstract]
  17. Niskanen L, Karjalainen J, Sarlund H, Siitonen O, Uusitupa M 1991 Five-year follow-up of islet cell antibodies in type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 34:402–408[CrossRef][Medline]
  18. Landin-Olsson M, Nilsson KO, Lernmark Å, Sundkvist G 1990 Islet cell antibodies and fasting C-peptide predict insulin requirement at diagnosis of diabetes mellitus. Diabetologia 33:561–568[CrossRef][Medline]
  19. Hagopian WA, Karlsen AE, Gottsater A, Landin-Olsson M, Grubin CE, Sundkvist G, Petersen JS, Boel E, Dyrberg T, Lernmark Å 1993 Quantitative assay using recombinant human islet glutamic acid decarboxylase (GAD-64) shows that 64K autoantibody positivity at onset predicts diabetes type. J Clin Invest 91:368–374
  20. Tuomi T, Groop LC, Zimmet PZ, Rowley MJ, Knowles W, Mackay IR 1993 Antibodies to glutamic acid decarboxylase reveal latent autoimmune diabetes mellitus in adults with a non-insulin-dependent onset of disease. Diabetes 42:359–362[Abstract]
  21. Lohmann T, Kellner K, Verlohren HJ, Krug J, Steindorf J, Scherbaum WA, Seissler J 2001 Titre and combination of ICA and autoantibodies to glutamic acid decarboxylase discriminate two clinically distinct types of latent autoimmune diabetes in adults (LADA). Diabetologia 44:1005–1010[CrossRef][Medline]
  22. Tuomi T, Carlsson A, Li H, Isomaa B, Miettinen A, Nilsson A, Nissen M, Ehrnstrom BO, Forsen B, Snickars B, Lahti K, Forsblom C, Saloranta C, Taskinen MR, Groop LC 1999 Clinical and genetic characteristics of type 2 diabetes with and without GAD antibodies. Diabetes 48:150–157[Abstract]
  23. Falorni A, Gambelunghe G, Forini F, Kassi G, Cosentino A, Candeloro P, Bolli GB, Brunetti P, Calcinaro F 2000 Autoantibody recognition of COOH-terminal epitopes of GAD65 marks the risk for insulin requirement in adult-onset diabetes mellitus. J Clin Endocrinol Metab 85:309–316[Abstract/Free Full Text]
  24. Pietropaolo M, Barinas-Mitchell E, Pietropaolo SL, Kuller LH, Trucco M 2000 Evidence of islet cell autoimmunity in elderly patients with type 2 diabetes. Diabetes 49:32–38[Abstract]
  25. Borg H, Gottsater A, Fernlund P, Sundkvist G 2002 A 12-year prospective study of the relationship between islet antibodies and ß-cell function at and after the diagnosis in patients with adult-onset diabetes. Diabetes 51:1754–1762[Abstract/Free Full Text]
  26. Gottsäter A, Landin-Olsson M, Lernmark Å, Fernlund P, Sundkvist G, Hagopian WA 1995 Glutamate decarboxylase antibody levels predict rate of ß-cell decline in adult onset diabetes. Diabetes Res Clin Pract 27:133–140[CrossRef][Medline]
  27. Turner R, Stratton I, Horton V, Manley S, Zimmet P, Mackay IR, Shattock M, Bottazzo GF, Holman R 1997 UKPDS 25: autoantibodies to islet-cell cytoplasm and glutamic acid decarboxylase for prediction of insulin requirement in type 2 diabetes. UK Prospective Diabetes Study Group. Lancet 350:1288–1293[CrossRef][Medline]
  28. Zimmet PZ, Tuomi T, Mackay IR, Rowley MJ, Knowles W, Cohen M, Lang DA 1994 Latent autoimmune diabetes mellitus in adults (LADA): the role of antibodies to glutamic acid decarboxylase in diagnosis and prediction of insulin dependency. Diabet Med 11:299–303[Medline]
  29. Juneja R, Hirsch IB, Naik RG, Brooks-Worrell BM, Greenbaum CJ, Palmer JP 2001 Islet cell antibodies and glutamic acid decarboxylase antibodies, but not the clinical phenotype, help to identify type 1(1/2) diabetes in patients presenting with type 2 diabetes. Metabolism 50:1008–1013[CrossRef][Medline]
  30. Zimmet P, Alberti KG, Shaw J 2001 Global and societal implications of the diabetes epidemic. Nature 41:782–787
  31. Mohatt J, Gilliam LK, Bekris L, Ebbesson S, Lernmark Å 2002 Type 1 diabetes-related autoantibodies are rare in Alaska native populations. Int J Circumpolar Health 61:21–31[Medline]
  32. Dabelea D, Palmer JP, Bennett PH, Pettitt DJ, Knowler WC 1999 Absence of glutamic acid decarboxylase antibodies in Pima Indian children with diabetes mellitus. Diabetologia 42:1265–1266[CrossRef][Medline]
  33. Bosi E, Garancini M PF, Bonifacio E, Gallus G 1999 Low prevalence of islet autoimmunity in adult diabetes and low predictive value of islet autoantibodies in the general population of northern Italy. Diabetologia 42:840–844[CrossRef][Medline]
  34. Soriguer-Escofet F, Esteva I, Rojo-Martinez G, Ruiz de Adana S, Catala M, Merelo MJ, Aguilar M, Tinahones F, Garcia-Almeida JM, Gomez-Zumaquero JM, Cuesta-Munoz AL, Ortego J, Freire JM 2002 Prevalence of latent autoimmune diabetes of adults (LADA) in Southern Spain. Diabetes Res Clin Pract 56:213–220[CrossRef][Medline]
  35. Takeda H, Kawasaki E, Shimizu I, Konoue E, Fujiyama M, Murao S, Tanaka K, Mori K, Tarumi Y, Seto I, Fujii Y, Kato K, Kondo S, Takada Y, Kitsuki N, Kaino Y, Kida K, Hashimoto N, Yamane Y, Yamawaki T, Onuma H, Nishimiya T, Osawa H, Saito Y, Makino H 2002 Clinical, autoimmune, and genetic characteristics of adult-onset diabetic patients with GAD autoantibodies in Japan (Ehime Study). Diabetes Care 25:995–1001[Abstract/Free Full Text]
  36. Niskanen LK, Tuomi T, Karjalainen J, Groop LC, Uusitupa MI 1995 GAD antibodies in NIDDM. Ten-year follow-up from the diagnosis. Diabetes Care 18:1557–1565[Abstract]
  37. Rolandsson O, Hägg E, Nilsson M, Hallmans G, Lernmark Å 2001 Prediction of diabetes by screening with body mass index, oral glucose tolerance test (OGT) and islet cell autoantibodies in a regional population. J Intern Med 249:279–288[CrossRef][Medline]
  38. Batstra MR, Petersen JS, Bruining GJ, Grobbee DE, de Man SA, Molenaar JL, Dyrberg T, Aanstoot HJ 2001 Low prevalence of GAD and IA2 antibodies in schoolchildren from a village in the southwestern section of The Netherlands. Hum Immunol 62:1106–1110[CrossRef][Medline]
  39. Hampe CS, Kockum I, Landin-Olsson M, Torn C, Ortqvist E, Persson B, Rolandsson O, Palmer J, Lernmark A 2002 GAD65 antibody epitope patterns of type 1.5 diabetic patients are consistent with slow-onset autoimmune diabetes. Diabetes Care 25:1481–1482[Free Full Text]
  40. Falorni A, Ackefors M, Carlberg C, Daniels T, Persson B, Robertson J, Lernmark Å 1996 Diagnostic sensitivity of immunodominant epitopes of glutamic acid decarboxylase (GAD65) autoantibodies epitopes in childhood IDDM. Diabetologia 39:1091–1098[Medline]
  41. Bonifacio E, Lampasona V, Bernasconi L, Ziegler AG 2000 Maturation of the humoral autoimmune response to epitopes of GAD in preclinical childhood type 1 diabetes. Diabetes 49:202–208[Abstract]
  42. Teoh KL, Fida S, Rowley MJ, Mackay IR 1998 Autoantigenic reactivity of diabetes sera with a hybrid glutamic acid decarboxylase GAD67–65 molecule GAD67(1–101)/GAD65(96–585). Autoimmunity 28:259–266[Medline]
  43. Kobayashi T, Tanaka S, Okubo M, Nakanishi K, Murase T, Lernmark A 2003 Unique epitopes of glutamic acid decarboxylase autoantibodies in slowly progressive type 1 diabetes. J Clin Endocrinol Metab 88:4768–4775[Abstract/Free Full Text]
  44. Maioli M, Tonolo G, Bekris L, Cirillo R, Schranz D, Cossu E, Ciccarese M, Lernmark Å 2002 GAD65 and IA-2 autoantibodies are common in siblings of Sardinia type 2 diabetic families. Diabetes Res Clin Pract 56:41–47[CrossRef][Medline]
  45. Muntoni S, Songini M 1992 High incidence rate of IDDM in Sardinia. Sardinian Collaborative Group for Epidemiology of IDDM. Diabetes Care 15:1317–1322[Abstract]
  46. Hampe CS, Hammerle LP, Bekris L, Ortqvist E, Kockum I, Rolandsson O, Landin-Olsson M, Torn C, Persson B, Lernmark A 2000 Recognition of glutamic acid decarboxylase (GAD) by autoantibodies from different GAD antibody-positive phenotypes. J Clin Endocrinol Metab 85:4671–4679[Abstract/Free Full Text]
  47. Michelsen BK, Petersen JS, Boel E, Møldrup A, Dyrberg T, Madsen OD 1991 Cloning, characterization, and autoimmune recognition of rat islet glutamic acid decarboxylase in insulin-dependent mellitus. Proc Natl Acad Sci USA 88:8754–8758[Abstract/Free Full Text]
  48. Hampe CS, Hammerle LP, Bekris L, Ortqvist E, Persson B, Lernmark A 2002 Stable GAD65 autoantibody epitope patterns in type 1 diabetes children five years after onset. J Autoimmun 18:49–53[CrossRef][Medline]
  49. Daw K, Ujihara N, Atkinson M, Powers AC 1996 Glutamic acid decarboxylase autoantibodies in stiff-man syndrome and insulin-dependent diabetes mellitus exhibit similarities and differences in epitope recognition. J Immunol 156:818–825[Abstract]
  50. Schwartz HL, Chandonia JM, Kash SF, Kanaani J, Tunnell E, Domingo A, Cohen FE, Banga JP, Madec AM, Richter W, Baekkeskov S 1999 High-resolution autoreactive epitope mapping and structural modeling of the 65 kDa form of human glutamic acid decarboxylase. J Mol Biol 287:983–999[CrossRef][Medline]
  51. Fisher LD, Van Belle G 1993 Biostatistics: a methodology for the health sciences. New York: John Wiley, Sons
  52. Notkins AL, Lernmark Å 2001 Autoimmune type 1 diabetes: resolved and unresolved issues. J Clin Invest 108:1247–1252[CrossRef][Medline]
  53. Weets I, Van Autreve J, Van der Auwera BJ, Schuit FC, Du Caju MV, Decochez K, De Leeuw IH, Keymeulen B, Mathieu C, Rottiers R, Dorchy H, Quartier E, Gorus FK 2001 Male-to-female excess in diabetes diagnosed in early adulthood is not specific for the immune-mediated form nor is it HLA-DQ restricted: possible relation to increased body mass index. Diabetologia 44:40–47[CrossRef][Medline]
  54. Landin-Olsson M, Arnqvist HJ, Blohme G, Littorin B, Lithner F, Nystrom L, Schersten B, Sundkvist G, Wibell L, Ostman J, Lernmark A 1999 Appearance of islet cell autoantibodies after clinical diagnosis of diabetes mellitus. Autoimmunity 29:57–63[Medline]
  55. Zimmet P, Turner R, McCarty D, Rowley M, Mackay I 1999 Crucial points at diagnosis. Type 2 diabetes or slow type 1 diabetes. Diabetes Care 22(Suppl 2):B59–B64
  56. Schranz D, Lernmark Å 1998 Immunology in diabetes: an update. Diabetes Metab Rev 14:3–29[CrossRef][Medline]
  57. Redondo MJ, Eisenbarth GS 2002 Genetic control of autoimmunity in type I diabetes and associated disorders. Diabetologia 45:605–622[CrossRef][Medline]
  58. Graham J, Hagopian WA, Kockum I, Li LS, Sanjeevi CB, Lowe RM, Schaefer JB, Zarghami M, Day HL, Landin-Olsson M, Palmer JP, Janer-Villanueva M, Hood L, Sundkvist G, Lernmark A, Breslow N, Dahlquist G, Blohme G 2002 Genetic effects on age-dependent onset and islet cell autoantibody markers in type 1 diabetes. Diabetes 51:1346–1355[Abstract/Free Full Text]
  59. Richter W, Shi Y, Bækkeskov S 1993 Autoreactive epitopes defined by diabetes-associated human monoclonal antibodies are localized in the middle and C-terminal domains of the smaller form of glutamate decarboxylase. Proc Natl Acad Sci USA 90:2832–2836[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maioli, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maioli, M.

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