Abstract
Aim
To understand the etiology of childhood-onset diabetes, we examined genetic risk markers, autoantibodies (AB), and β-cell function in a mixed-race group of young patients.
Methods
145 patients aged 0–17 at diagnosis (54% African American, 22% Caucasian, 16% Latino, 8% Mixed-Other) were studied at mean duration 6.9 ± 5.7 (range 0.1–28.5) years, including HLA-DQA1-DQB1 genotyping, stimulated C-peptide (CP), GAD and IA-2 antibodies (AB). Based on no residual β-cell function (CP−) and islet autoantibodies (AB+), 111 patients were classified with type 1 diabetes (T1DM); 22 were CP+ and AB− and thus considered to have type 2 diabetes (T2DM); 12 patients had features of both T1DM and T2DM, or mixed phenotype.
Results
Based on the presence of two high-risk HLA-DQA1/B1 haplotypes, 39% of African Americans, 81% of Caucasians, 70% of Latinos, and 67% of Mixed-Others, were at high genetic risk. In T1DM patients, 41% of African Americans, 80% of Caucasians, 73% of Latinos, and 63% of Mixed-Others were genetically susceptible. 31% of African Americans, including 29% of those with T1DM, could not be characterized because their haplotypes had unknown T1DM associations. These unusual haplotypes comprised 11% in T1DM, 14% in T2DM, and 8% of haplotypes in mixed phenotype patients.
Conclusions
59% of childhood-onset T1DM patients were identified with high genetic risk based on known HLA-DQA1/B1 associations. Many non-Caucasian patients carry HLA-DQ alleles whose association with T1DM is undetermined. Genetic approaches can provide insights into the etiology and appropriate treatment of childhood-onset diabetes, but only if sufficient data are available in diverse ethnic groups.
Keywords: Diabetes Type 1, Diabetes Type 2, Children and adolescents, Ethnic differences, African Americans, Caucasians, Latinos, Epidemiology, Phenotypes, HLA-DQA1-DQB1
Background - Introduction
Diabetes is one of the most common chronic diseases of childhood. Although in recent years it has become obvious that childhood-onset diabetes is heterogeneous in etiology, it is primarily comprised of the autoimmune-mediated, insulin-deficient type 1 phenotype. By far the strongest genetic determinants of type 1 diabetes risk are located in the major histocompatibility region (MHC) on chromosome 6p21, with the most robust associations with the human leukocyte antigen (HLA) class II genes DRB1 and DQB1. Historically, cross-population comparisons have allowed major insights into the genetic basis for disease, including type 1 diabetes (1,2). While MHC genes account for approximately 40% of type 1 diabetes risk, it has long been recognized that other regions of the genome also confer susceptibility or resistance to beta-cell autoimmunity and type 1 diabetes (3–5). Currently, studies comparing type 1 diabetes genetics across ethnic lines have focused on these non-HLA associations (2,6,7), confirming associations with insulin gene (INS) polymorphisms and variants at CTLA4, and PTPN22, among others. Unfortunately, few data, even on the older HLA associations, are currently available for African-origin or indigenous American groups.
Over the past 1–2 decades there has been a steady increase in childhood diabetes incidence of approximately 3% per annum across the world (8,9). Major secular changes in the environmental-behavioral determinants of diabetes etiology, specifically overnutrition and physical inactivity, are also occurring at this time. Furthermore, recent epidemiologic data show that the risk for type 1 diabetes in non-Caucasian children is roughly equivalent to that in non-Hispanic whites (9,10), while the risk for non-type 1 phenotypes is roughly double that for Caucasian young people (9, 11–13). However, important disparities in diabetes outcomes have been demonstrated between Caucasian and non-Caucasian children in the United States (14–16). This is probably rooted in social, not biological, causes, i.e. lack of adequate health care, but if the etiology of diabetes in non-Caucasian children is not well understood, then inappropriate treatment could also be contributing to the documented ethnic disparities. It is essential to decipher the genetic and nongenetic contributions to the etiology of childhood diabetes, not only to develop appropriate treatments, but also to correctly identify patients for inclusion in future research.
We used samples from 145 patients derived from a family study in the Chicago metropolitan area to evaluate genetic associations with apparent type 1, type 2, and mixed diabetes phenotypes. All patients were initially diagnosed between the ages of 0–17, and 54% were African American in ancestry (Table 1). Phenotyping was based on physical findings after, on average, seven years’ duration of diabetes. This work was undertaken to address the hypothesis that the genetic profile of children with type 1 diabetes is different than those of children with type 2 or mixed forms of diabetes, among both Caucasian and non-Caucasian populations. Specifically, we hypothesized that a) The prevalence of type 1 diabetes-associated HLA haplotypes will be lower among patients with youth-onset diabetes exhibiting mixed features or those of type 2 diabetes, than among patients with “true” type 1 diabetes, irrespective of ethnicity; and b) The frequency of high-risk HLA-DQA1/B1 haplotypes would be similar among Caucasian type 1 patients and those of other ethnic backgrounds.
Table 1.
Characteristics of 145 Childhood-onset Diabetes Patients with HLA DQA1-DQB1 Typing, by Patients’ Ethnicity
| African American | Caucasian | Latino | Mixed-Other Race | P-value | |
|---|---|---|---|---|---|
|
|
|||||
| N (males/females) | 78 (29/49) | 32 (20/12) | 23 (10/13) | 12 (3/9) | |
| Diabetes Phenotype*, n (%) | 0.011 | ||||
| Type 1 | 51 (65.4) | 30 (93.8) | 22 (95.7) | 8 (66.7) | |
| Type 2 | 17 (21.8) | 1 (3.1) | 1 (4.3) | 3 (25.0) | |
| Mixed | 10 (12.8) | 1 (3.1) | 0 | 1 (8.3) | |
| Age at diag., yrs, mean (SD) | 9.0 (4.4) | 7.7 (4.0) | 7.7 (3.8) | 9.2 (4.5) | NS** |
| DM duration, yrs, mean (SD) | 7.5 (5.4) | 5.4 (4.4) | 6.8 (4.5) | 7.0 (4.6) | NS |
| Parent +/or sib with DM^, n (%) | 26 (33.3) | 5 (16.1) | 5 (23.8) | 3 (27.3) | NS |
| IA-2 +/or GAD Ab pos, n (%) | 45 (57.7) | 21 (65.6) | 13 (56.5) | 8 (66.7) | NS |
| IA-2 Ab pos, n (%) | 29 (37.2) | 13 (40.6) | 12 (52.2) | 3 (25.0) | NS |
| GAD Ab pos^, n (%) | 34 (43.6) | 15 (48.4) | 11 (50.0) | 6 (54.5) | NS |
| Overall type 1 risk, based on two HLA-DQA1/B1 haplotypes, n (%) | 0.005 | ||||
| Resistant | 5 (6.4) | 2 (6.2) | 1 (4.3) | 0 | |
| Neutral | 19 (24.4) | 3 (9.4) | 1 (4.3) | 2 (16.7) | |
| Susceptible | 30 (38.5) | 26 (81.2) | 16 (69.6) | 8 (66.7) | |
| Unable to determine | 24 (30.8) | 1 (3.1) | 5 (21.7) | 2 (16.7) | |
| In Type 1 patients only, | n=51 | 30 | 22 | 8 | |
| IA-2 +/or GAD Ab pos, n (%) | 37 (72.5) | 20 (66.7) | 13 (59.1) | 7 (87.5) | NS |
| IA-2 Ab pos, n (%) | 26 (51.0) | 12 (40.0) | 12 (54.5) | 3 (37.5) | NS |
| GAD Ab pos^, n (%) | 29 (56.9) | 15 (51.7) | 11 (52.4) | 5 (71.4) | NS |
| Overall type 1 risk, based on two HLA-DQA1/B1 haplotypes, n (%) | 0.020 | ||||
| Resistant | 3 (5.9) | 2 (6.7) | 1 (4.5) | 0 | |
| Neutral | 12 (23.5) | 3 (10.0) | 0 | 2 (25.0) | |
| Susceptible | 21 (41.2) | 24 (80.0) | 16 (72.7) | 5 (62.5) | |
| Unable to determine | 15 (29.4) | 1 (3.3) | 5 (22.7) | 1 (12.5) | |
Phenotype: Type 1, absent stimulated C-peptide with >2 years’ diabetes duration or positive antibodies and insulin monotherapy with <2 years’ duration; Type 2, C-peptide present with >2 years’ diabetes duration and no islet autoantibodies; Mixed, either a) residual β-cell function and autoantibodies (n=10); or b) residual β-cell function after >5 years, on insulin monotherapy since diagnosis, and reports developing DKA when discontinuing insulin after the first 6 months of diabetes (n=2).
NS, not statistically significant
Information available for 141 subjects (family history); 142 subjects (GAD antibodies); 108 type 1 subjects (GAD antibodies)
Methods
Subjects
Patients (n=150) were recruited from clinics or by mail, if they were aged 0–17 years at initial diagnosis of diabetes, and if the diabetes was not secondary to another medical condition. Patients were excluded based on insufficient HLA data (n=2) or inability to determine their phenotype at the time of examination (n=3), leaving a final study sample of 145 patients. Clinical studies were conducted in the Clinical Research Centers at the University of Illinois at Chicago and the University of Chicago, and in participants’ homes. Human Subjects Research committees at the University of Illinois at Chicago and the University of Chicago approved the study protocol. Written informed consent was obtained from participants prior to the interview and clinical studies; those currently aged 12–17 were asked to give written assent, and parental consent was obtained.
Patients were examined 6.9±5.7 (mean±SD); range 0.1–28.5; years after their initial diagnosis. Information on family history and clinical characteristics was gathered in detailed interviews. Ethnicity was defined as having at least three grandparents with the same ethnic origin according to the proband or a family informant. In addition to genetic studies, physical measures were taken and islet cell autoantibodies (insulinoma-associated antigen 2, or IA-2; glutamic acid decarboxylase, or GAD) and β-cell function were tested.
Phenotyping
We determined the phenotypes of patients using information from their physical examinations and questionnaires. Patients with <2 years’ duration who were autoantibody positive and using insulin alone, or those with no detectable post-challenge C-peptide and a disease duration >2 years, i.e. no endogenous insulin production, were classified as having ‘true’ type 1 diabetes (n=111). Patients with residual β-cell function, no islet autoantibodies, and either not using insulin at all, or using ≤2 insulin shots/day (with or without oral antidiabetic agents), were classified as having type 2 diabetes (n=22). Fifteen of these individuals were using oral agents or diet alone. We observed a substantial fraction of patients who had to be categorized as having ‘mixed’ phenotype (n=12); these patients had either a) residual β-cell function and autoantibodies (n=10); or b) measurable C-peptide, but reporting an acute outcome, i.e. DKA, when stopping insulin therapy (n=2).
Autoantibodies
Antibodies to radiolabelled recombinant human GAD65 (whole) and human IA-2 (349 AA cytoplasmic portion) were quantified by fluid-phase immunoprecipitation assay (17). These antigens were cloned from human islets and human glioblastoma, respectively, and are expressed in recombinant form by coupled in-vitro transcription and translation in the presence of 35S-methionine. After incubation with subject serum in separate triplicate wells, immune complexes are precipitated using Protein A, washed, and counted for bound radioactivity. Using the mean of triplicates on each index serum and each unknown serum, an antibody “index” is calculated as: Index = (Unknown Serum-Index Negative)/(Index Pos-Index Neg). The WHO/JDF standard serum for GAD antibody and IA-2 were used in each assay. The assays were 94% and 80% sensitive, and 94% and 95% specific, for GAD and IA-2, respectively, in the 2005 Diabetes Autoantibody Standardization Program sponsored by the Centers for Disease Control & Prevention and the Immunology of Diabetes Society.
HLA Genotyping
HLA-DQA1 and -DQB1 typing was conducted in the laboratory of Dr. William Hagopian at the Pacific Northwest Research Institute, using extracted DNA (n=39) or dried bloodspot samples applied to Schleicher & Schuell 903 paper (n=109). Genotyping at HLA-DQB1 (high resolution) and HLA-DQA1 (moderate resolution) used published PCR primers (18). and standard SSOP techniques, supplemented by direct sequencing on an ABI310 automated sequencer as needed. Results were used to generate extended haplotypes for assignment of type 1 diabetes risk according to published multi-ethnic data (19–21). We considered the following HLA-DQA1-DQB1 haplotypes to confer susceptibility to type 1 diabetes: 0201-0302, 0301-0201, 0301-0302, 030x-0302, 0302-0303, 030x-0303, 0401-0301, 0401-0402 (in Caucasians), 0501-0201/2, 050x-0201/2. Neutral haplotypes were identified as: 0101-0501, 010x-0501, 0101/4-0503, 0101-0604, 0101-0609, 0102-0502, 0102-0604, 0102-0609; haplotypes that were classified as resistant were: 0102-0602, 0103-0603, 0201-0201, 0201-0201/2, 0201-0303, 0301-0301, 030x-0301, 0401-0402 (in African Americans), 0501-0301, 0601-0301. There were several additional haplotypes whose association with type 1 diabetes risk has not yet been decisively identified. Based on both HLA-DQA1-DQB1 haplotypes, we considered those with one susceptible haplotype paired with either a susceptible, neutral, or unknown haplotype to be at high genetic risk for type 1 diabetes. Similarly, those with at least one resistant haplotype paired with either a resistant, neutral, or unknown haplotype to have a low genetic risk for type 1 diabetes. Intermediate type 1 risk was assumed for those with a neutral haplotype paired with either an unknown or a neutral haplotype, or those with one susceptible and one resistant haplotype. There were 113 of the 145 subjects with HLA typing who were able to be classified in this manner; the remaining 32 subjects carried two haplotypes with undetermined risk status.
C-Peptide
All assays were conducted in the Diabetes Research and Training Center Ligand Assay Core laboratory at the University of Chicago. For participants examined before 2004 (n=39), plasma C-peptide was measured using a previously described immunoassay (22). The lower limit of sensitivity of this assay was 0.02 nmol/L and the intra-assay CV averaged 6%. C-peptide for participants seen after 2004 (n=106) was measured with a solid-phase, competitive chemiluminescent enzyme immunoassay (Immulite 2000; Diagnostic Products Corporation, Biermann GmBH, Bad Nauheim, Germany) in the same laboratory (23). A modified assay was substituted by the manufacturer in October 2007 (Immulite 1000). The lower limit of detection for both Immulite assays was 0.17 nmol/L and the intra-assay coefficient of variation was 8% on average. For purposes of this study, we defined residual β-cell function as a C-peptide value ≥0.05 nmol/L using the pre-2004 assay, and as ≥0.17 nmol/L in the newer assays (22–24).
Statistical Methods
Univariate analyses, t-tests or Chi-square tests as appropriate, were conducted to compare variables across ethnic groups. P-values were not corrected for multiple comparisons. Due to small numbers, we chose not to construct multivariable models. All analyses were conducted using SPSS version 16 (Chicago, IL).
Results
The study group included 83 females, with 78 patients of non-Hispanic black (African American) origin, 32 non-Hispanic whites (Caucasians), 23 Latinos and 12 patients of other or mixed ethnicity, primarily Asian (Table 1). Caucasians had somewhat shorter disease duration at the time of examination, 5.4 years vs. 7.5, 6.8, and 7.0 years for African Americans, Latinos, and Mixed-Other groups, respectively. Fewer Caucasians, 16%, than African Americans (33%), Latinos (24%), or Mixed-Others (27%) had a first-degree relative with diabetes, though neither of these comparisons was statistically significant (Table 1). Islet autoantibodies (GAD +/or IA-2) were detectable in 87 (60%) of the patients, including 69% of type 1 and 83% of mixed phenotype patients; by definition type 2 diabetes patients were antibody negative. Among the type 1 patients there were no ethnic differences in the prevalence of antibody positivity (Table 1); there were too few Caucasian and Latino mixed phenotype patients to allow significance testing of antibody positivity by ethnicity in that subgroup.
We defined an overall level of genetic risk for type 1 diabetes based on two HLA-DQA1/B1 haplotypes with known risk status. We found that 39% of African Americans, 81% of Caucasians, 70% of Latinos, and 67% of Mixed-others carried a high level of genetic risk; while 24%, 9%, 4% and 17% were at intermediate risk, and 6%, 6%, 4%, and none were at low risk, respectively (p=0.005; Table 1). Among type 1 patients, 41% of African Americans, 80% of Caucasians, 73% of Latinos, and 63% of Mixed-others carried a high level of genetic risk (p=0.02). 31% of African American patients, including 29% of those with type 1 diabetes, could not be characterized in terms of overall genetic risk based on HLA because they carried HLA-DQA1/B1 haplotypes of unknown association. This was also the case for Latino and mixed race patients, where 22% and 17%, respectively, could not be assigned an HLA-based genetic risk, including 23% of Latino type 1 patients (Table 1). In contrast, only one Caucasian subject was not able to be characterized because this type 1 patient carried undefined haplotypes.
Comparing overall HLA-based genetic risk across phenotypes, we found that 59% of type 1 patients displayed a susceptible (high risk) profile, compared with 32% of type 2 and 58% of mixed phenotype patients. Among African Americans, 41% of type 1 patients displayed a susceptible HLA profile, as did 24% of type 2 and 50% of mixed phenotype patients. The proportions of African Americans who displayed a resistant HLA profile were 6% of type 1, 12% of type 2, and none of the mixed phenotype patients. Among Caucasians, 80% of type 1 patients carried a susceptible HLA profile, as did the single type 2 patient and the single mixed phenotype patient. Among Latinos, 73% of type 1 patients displayed a susceptible HLA-DQA1/DQB1 profile; the single Latino type 2 patient displayed a neutral risk profile. Among the 12 mixed/other race patients who were studied, the majority of both type 1 patients (5/8) and type 2 patients (2/3) displayed a susceptible genetic risk profile based on two HLA-DQA1/B1 haplotypes.
There appear to be relatively large deviations from predicted frequencies for the type 1-associated HLA-DQA1/B1 haplotypes (Table 2). There were 290 haplotypes for evaluation in the 145 patients, of which 167 (58%) were defined as type 1 diabetes susceptibility haplotypes, 42 (15%) were neutral, 50 (17%) have been reported to confer resistance to type 1 diabetes, and 33 (11%) were undefined in terms of their risk association.
Table 2.
HLA DQA1-DQB1 Haplotypes and Type 1 Diabetes Associations, in 145 Childhood-onset Diabetes Patients, by Ethnicity and Phenotype
| A. Type 1 Diabetes Patients (n=111)
| |||||
|---|---|---|---|---|---|
| Afr Amer | Caucasian | Latino | Mixed-Other | Total | |
| Total haplotypes: | 102 | 60 | 44 | 16 | 222 |
| Type 1 diabetes association: | |||||
| Resistant | |||||
| 0102-0602 | 1 | 1 | 0 | 1 | 3 |
| 0103-0603 | 1 | 1 | 0 | 0 | 2 |
| 0201-0201/2 | 11 | 1 | 2 | 0 | 14 |
| 0201-0303 | 1 | 0 | 0 | 0 | 1 |
| 0301/x-0301 | 1 | 1 | 0 | 1 | 3 |
| 0401-0402 (African American) | 3 | - | - | - | 3 |
| 0501-0301 | 2 | 2 | 0 | 0 | 4 |
| Resistant haplotypes, n(%) | 20(19.6) | 6 (10.0) | 2 (4.5) | 2 (12.5) | 30 (13.5) |
| Neutral | |||||
| 0101/x-0501 | 5 | 1 | 4 | 0 | 10 |
| 0101/4-0503 | 2 | 3 | 0 | 0 | 5 |
| 0101-0604 | 1 | 0 | 0 | 0 | 1 |
| 0102-0502 | 2 | 0 | 0 | 0 | 2 |
| 0102-0604/9 | 3 | 5 | 0 | 0 | 8 |
| Neutral haplotypes, n(%) | 13 (12.7) | 9 (15.0) | 4 (9.1) | 0 | 26 (11.7) |
| Susceptible | |||||
| 0201-0302 | 0 | 0 | 1 | 0 | 1 |
| 0301-0201 | 5 | 0 | 2 | 0 | 7 |
| 0301/x-0302 | 22 | 19 | 13 | 5 | 59 |
| 0302/x-0303 | 2 | 1 | 1 | 1 | 5 |
| 0401-0402 (Caucasian) | - | 2 | 0 | 1 | 3 |
| 0501/x-0201/2 | 24 | 22 | 16 | 6 | 68 |
| Susceptible haplotypes, n(%) | 53 (52.0) | 44 (73.3) | 33 (75.0) | 13 (81.3) | 143 (64.4) |
| Undetermined | |||||
| 010x-0604 | 0 | 0 | 0 | 1 | 1 |
| 0302-0202 | 6 | 0 | 3 | 0 | 9 |
| 030x-0201/2 | 11 | 0 | 0 | 0 | 11 |
| 030x-0301/9 | 0 | 1 | 0 | 0 | 1 |
| 030x-0304 | 0 | 0 | 1 | 0 | 1 |
| 040x-0402 | 0 | 0 | 1 | 0 | 1 |
| 050x-0301/9 | 1 | 0 | 0 | 0 | 1 |
| Undetermined haplotypes, n(%) 18 (17.6) | 1 (1.7) | 5 (11.4) | 1 (6.3) | 25 (11.3) | |
| B. Type 2 Diabetes Patients (n=22)
| |||||
|---|---|---|---|---|---|
| Afr Amer | Caucasian | Latino | Mixed-Other | Total | |
| Total haplotypes: | 34 | 2 | 2 | 6 | 44 |
| Type 1 diabetes association: | |||||
| Resistant | |||||
| 0102-0602 | 4 | 0 | 0 | 0 | 4 |
| 0103-0603 | 0 | 0 | 1 | 0 | 1 |
| 0201-0201/2 | 1 | 0 | 0 | 1 | 2 |
| 0401-0402 (African American) | 6 | - | - | - | 6 |
| 0501-0301 | 3 | 0 | 0 | 0 | 3 |
| 0601-0301 | 1 | 0 | 0 | 0 | 1 |
| Resistant haplotypes, n(%) | 15 (44.1) | 0 | 1 (50.0) | 1 (16.7) | 17 (38.6) |
| Neutral | |||||
| 0101-0501 | 4 | 1 | 0 | 0 | 5 |
| 0101/4-0503 | 1 | 0 | 0 | 0 | 1 |
| 0101-0609 | 2 | 0 | 0 | 0 | 2 |
| 0102-0502 | 1 | 0 | 0 | 1 | 2 |
| Neutral haplotypes, n(%) | 8 (23.5) | 1 (50.0) | 0 | 1 (16.7) | 10 (22.7) |
| Susceptible | |||||
| 0301-0201 | 1 | 0 | 0 | 1 | 2 |
| 0301/x-0302 | 0 | 0 | 1 | 1 | 2 |
| 0302-0303 | 1 | 0 | 0 | 0 | 1 |
| 0401-0301 | 1 | 0 | 0 | 0 | 1 |
| 0501/x-0201 | 3 | 1 | 0 | 1 | 5 |
| Susceptible haplotypes, n(%) | 6 (17.6) | 1 (50.0) | 1 (50.0) | 3 (50.0) | 11 (25.0) |
| Undetermined | |||||
| 0101-0502 | 1 | 0 | 0 | 0 | 1 |
| 0101-0602 | 1 | 0 | 0 | 0 | 1 |
| 0102-0501 | 1 | 0 | 0 | 0 | 1 |
| 0501-0304 | 1 | 0 | 0 | 0 | 1 |
| 0601-0402 | 1 | 0 | 0 | 0 | 1 |
| 060x-0301/9 | 0 | 0 | 0 | 1 | 1 |
| Undetermined haplotypes, n(%) | 5 (14.7) | 0 | 0 | 1 (16.7) | 6 (13.6) |
| C. Mixed Phenotype Diabetes Patients (n=12)
| |||||
|---|---|---|---|---|---|
| Afr Amer | Caucasian | Latino | Mixed-Other | Total | |
| Total haplotypes: | 20 | 2 | 0 | 2 | 24 |
| Type 1 diabetes association: | |||||
| Resistant | |||||
| 0102-0602 | 1 | 0 | - | 0 | 1 |
| 0401-0402 (African American) | 1 | 1 | |||
| 0501-0301 | 1 | 0 | - | 0 | 1 |
| Resistant haplotypes, n(%) | 3 (15.0) | 0 | - | 0 | 3 (12.5) |
| Neutral | |||||
| 0101-0501 | 2 | 1 | - | 0 | 3 |
| 0101/4-0503 | 2 | 0 | - | 0 | 2 |
| 0101-0604 | 1 | 0 | - | 0 | 1 |
| Neutral haplotypes, n(%) | 5 (25.0) | 1 (50.0) | - | 0 | 6 (25.0) |
| Susceptible | |||||
| 0301-0201 | 1 | 0 | - | 0 | 1 |
| 0301/x-0302 | 3 | 0 | - | 1 | 4 |
| 0501/x-0201/2 | 6 | 1 | - | 1 | 8 (33.3) |
| Susceptible haplotypes, n(%) | 10 (50.0) | 1 (50.0) | - | 2 (100) | 13 (54.2) |
| Undetermined | |||||
| 0101-0602 | 1 | 0 | - | 0 | 1 |
| 030x-0201/2 | 1 | 0 | - | 0 | 1 |
| Undetermined haplotypes, n(%) | 2 (10.0) | 0 | - | 0 | 2 (8.3) |
|
| |||||
| Total Haplotypes | 156 | 64 | 46 | 24 | 290 |
| Number of Patients | 78 | 32 | 23 | 12 | 145 |
Among the 222 haplotypes in those patients who had type 1 diabetes, 143 haplotypes (64%) carried type 1 susceptibility, 26 (12%) were neutral, 30 (14%) were resistant, and 25 (11%) were undefined. Among the 102 haplotypes in African American type 1 patients, 52% were susceptible, 13% were neutral, 20% were resistant, and fully 18% were not yet defined in terms of their type 1 diabetes risk association. In the 60 haplotypes from Caucasian patients with type 1 diabetes, 73% carried type 1 diabetes susceptibility, 15% were neutral, 10% were resistant, and one was undefined. In 44 haplotypes of Latino type 1 patients, 75% were susceptible, 9% were neutral, 5% were resistant, and 11% were undefined. Thirteen of the 16 haplotypes carried by mixed-other race type 1 patients were susceptible, while just one was undefined in terms of its type 1 diabetes association.
Discussion
We explored the extent of genetic variation at HLA-DQA1 and -DQB1 among patients diagnosed in childhood, in whom the inborn components of diabetes risk are likely to be of greatest importance. Our sample represents a diversity of ethnic origins, and phenotype was determined as precisely as possible within the constraints of research among otherwise healthy, free-living individuals. We found that the prevalence of type 1 diabetes-associated HLA haplotypes was lower among patients with youth-onset diabetes exhibiting features of type 2 diabetes, than among patients with “true” type 1, irrespective of ethnicity. However, the frequency of high-risk HLA-DQA1/B1 haplotypes was considerably lower among African American patients, compared to those of Caucasian ancestry, even among those who clearly demonstrated a type 1 phenotype. This effect was largely accounted for by the much higher frequency among African Americans and Latinos of haplotypes that have not yet been characterized in terms of their type 1 diabetes risk. This observation is consistent with the report of Antal, et al. (25) who examined HLA class I supertypes in minority type 1 patients, and found that 65% did not belong to the A2 supertype, which is the most commonly associated with type 1 in Caucasians.
The relatively small number of non- type 1 patients, particularly among Caucasians and Latinos, limited the statistical power of this analysis. In fact, the value of this manuscript is that it is the first description of frequencies of these variants in an ethnically diverse sample with well-characterized diabetic phenotypes, rather than its statistical robustness. Clearly, if we had conducted HLA-DR typing in addition to the DQA1-DQB1 analysis, it would have allowed greater insight into the type 1 diabetes risk associations (20). We hope that further studies of the HLA-DR-DQ loci in non-European type 1 patients will soon emerge in response to this manuscript which highlights the dearth of such information that now exists.
We used a straightforward approach to determining the phenotype of patients at follow-up, yet the possibility of misclassification cannot be altogether ruled out. We did not test for the zinc transporter 8 autoantibody which was relatively recently identified as a marker of type 1 risk, and thus may have allowed us to more precisely classify phenotype. Certainly, some “true” type 1 patients may continue to demonstrate residual β-cell function for many years. It is also possible that glucose toxicity associated with extremely poor glycemic control could mask the ability to produce insulin in those who would otherwise have been classified as type 2 patients. Among the 95 type 1 patients in our study who had no detectable C-peptide, there were 28 (30%) in very poor metabolic control (HbA1c ≥10%), a proportion not different among those with mixed or type 2 phenotype. Thus, we remain confident that the type 1 diabetes group in this analysis is largely homogeneous. Finally, this cohort, though grossly representative of the ethnic and socioeconomic distribution of young people in the Chicago metropolitan area, was not randomly selected, but instead consisted of patients whose families were able to be contacted under recently-adopted stringent legal constraints, and who agreed to be examined. Nonetheless, this study does represent a large group of young diabetes patients with careful attention to phenotyping and HLA results, and it includes a substantial number of minority patients.
In summary, fewer than expected of the HLA-DQA1/B1 haplotypes in patients with childhood-onset diabetes conferred high risk for type 1 diabetes, particularly among those from non-Caucasian backgrounds. Studies across racial groups using sufficently large numbers of samples carry the potential to identify causal variants. Future exploration of the role of genetic variation in diabetes risk among young people should include determination of HLA class II associations in a variety of ethnic groups, as well as meticulous attention to determining phenotype. This is a crucial step in preparing for future etiologic and prevention research, as well as in identifying the appropriate treatment plan for all young diabetes patients.
Acknowledgments
Funding: NIH – R01-DK44752; P60-DK20595; M01-RR13987; UL1-RR024999. Rose Briars, Irwin Brodsky, Deborah Burnet, Paula Butler, Rachel Caskey, Carmela Estrada, Brigid Gregg, Latrisha Hampton, Elizabeth Littlejohn, Maureen Mencarini, Jennifer Miller, Monica Mortensen, Aida Pourbovali, Barry Rich, Lydia Rodriguez, Paul Rue, Tracie Smith, Frank Thorp, Christine Yu; participating patients and families.
Abbreviations
- AB
antibodies
- CP
C-peptide
- DKA
diabetic ketoacidosis
- GAD
glutamic acid decarboxylase
- HbA1c
hemoglobin A1c
- HLA
human leukocyte antigen
- IA-2
insulinoma-associated antigen 2
- MDM
mixed diabetic phenotype
- MHC
major histocompatibility complex
- PCOS
polycystic ovarian syndrome
- T1DM
type 1 diabetes
- T2DM
type 2 diabetes
References
- 1.Todd JA, Mijovic C, Fletcher J, Jenkins D, Bradwell AR, Barnett AH. Identification of susceptibility loci for insulin-dependent diabetes mellitus by trans-racial gene mapping. Nature. 1989;338:587–89. doi: 10.1038/338587a0. [DOI] [PubMed] [Google Scholar]
- 2.Ikegami H, Kawabata Y, Noso S, Fujisawa T, Ogihara T. Genetics of type 1 diabetes in Asian and Caucasian populations. Diabetes Res Clin Pract. 2007;77 (Suppl 1):S116–21. doi: 10.1016/j.diabres.2007.01.044. [DOI] [PubMed] [Google Scholar]
- 3.Concannon P, Rich SS, Nepom GT. Genetics of type 1A diabetes. N Engl J Med. 2009;360:1646–54. doi: 10.1056/NEJMra0808284. [DOI] [PubMed] [Google Scholar]
- 4.Cooper JD, Walker NM, Healy BC, Smyth DJ, Downes K, Todd JA Type I Diabetes Genetics Consortium. Analysis of 55 autoimmune disease and type II diabetes loci: further confirmation of chromosomes 4q27, 12q13.2 and 12q24.13 as type I diabetes loci, and support for a new locus, 12q13.3-q14.1. Genes Immun. 2009;10(Suppl 1):S95–120. doi: 10.1038/gene.2009.98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bjørnvold M, Undlien DE, Joner G, Dahl-Jørgensen K, Njølstad PR, Akselsen HE, Gervin K, Rønningen KS, Stene LC. Joint effects of HLA, INS, PTPN22 and CTLA4 genes on the risk of type 1 diabetes. Diabetologia. 2008;51:589–96. doi: 10.1007/s00125-008-0932-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Steck AK, Zhang W, Bugawan TL, Barriga KJ, Blair A, Erlich HA, Eisenbarth GS, Norris JM, Rewers MJ. Do non-HLA genes influence development of persistent islet autoimmunity and type 1 diabetes in children with high-risk HLA-DR, DQ genotypes? Diabetes. 2009;58:1028–33. doi: 10.2337/db08-1179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Berka N, Nunlee-Bland G, Erabhaoui E, Belmamoun M, Dunston GM. IDDM2 and the polymorphism of the human tyrosine hydroxylase (hTH) gene in African Americans with type-1 diabetes. J Natl Med Assoc. 2004;96:1042–46. [PMC free article] [PubMed] [Google Scholar]
- 8.Soltesz G, Patterson CC, Dahlquist G on behalf of EURODIAB Study Group. Worldwide childhood type 1 diabetes incidence – what can we learn from epidemiology? Pediatric Diabetes. 2007;8(Suppl 6):6–14. doi: 10.1111/j.1399-5448.2007.00280.x. [DOI] [PubMed] [Google Scholar]
- 9.Smith TLS, Drum ML, Lipton RB. Incidence of childhood type 1 and non-type 1 diabetes in a diverse population: The Chicago Childhood Diabetes Registry, 1994 to 2003. J Pediatr Endocrinol Metab. 2007;20:1093–1107. doi: 10.1515/jpem.2007.20.10.1093. [DOI] [PubMed] [Google Scholar]
- 10.Libman IM, Pietropaolo M, Arslanian SA, LaPorte RE, Becker DJ. Evidence for heterogeneous pathogenesis of insulin-treated diabetes in black and white children. Diab Care. 2003;26:2876–82. doi: 10.2337/diacare.26.10.2876. [DOI] [PubMed] [Google Scholar]
- 11.Duncan GE. Prevalence of diabetes and impaired fasting glucose levels among US adolescents: National Health and Nutrition Examination Survey, 1999–2002. Arch Pediatr Adolesc Med. 2006;160:523–28. doi: 10.1001/archpedi.160.5.523. [DOI] [PubMed] [Google Scholar]
- 12.Scott CR, Smith JM, Cradock MM, Pihoker C. Characteristics of youth-onset noninsulin-dependent diabetes mellitus and insulin-dependent diabetes mellitus at diagnosis. Pediatrics. 1997;100:84–91. doi: 10.1542/peds.100.1.84. [DOI] [PubMed] [Google Scholar]
- 13.Urakami T, Kubota S, Nitadori Y, Harada K, Owada M, Kitagawa T. Annual incidence and clinical characteristics of type 2 diabetes in children as detected by urine glucose screening in the Tokyo metropolitan area. Diab Care. 2005;28:1876–81. doi: 10.2337/diacare.28.8.1876. [DOI] [PubMed] [Google Scholar]
- 14.Akinbami LJ, Saydah SH, Eberhardt MS, Polakowski LL. Racial Disparities in Diabetes Mortality Among Persons Aged 1–19 Years — United States, 1979–2004. CDC Morbidity and Mortality Weekly Report. 2007;56:1184–87. [PubMed] [Google Scholar]
- 15.Lipton R, Good G, Mikhailov T, Freels S, Donoghue E. Ethnic differences in mortality from insulin-dependent diabetes mellitus among people less than 25 years of age. Pediatrics. 1999;103:952–56. doi: 10.1542/peds.103.5.952. [DOI] [PubMed] [Google Scholar]
- 16.Estrada CL, Danielson KK, Drum ML, Lipton RB. Hospitalization subsequent to diagnosis in young patients with diabetes in Chicago. Pediatrics. 2009;124:926–34. doi: 10.1542/peds.2008-3826. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Woo W, LaGasse JM, Zhou Z, Patel R, Palmer JP, Campus H, Hagopian WA. A novel high-throughput method for accurate, rapid, and economical measurement of multiple type 1 diabetes autoantibodies. J Immunol Meth. 2000;244:91–103. doi: 10.1016/s0022-1759(00)00259-3. [DOI] [PubMed] [Google Scholar]
- 18.Sanjeevi CB, Hagopian WA, Landin-Olsson M, Kockum I, Woo W, Palmer JP, Lernmark A, Dahlquist G. Association between autoantibody markers and subtypes of DR4 and DR4-DQ in Swedish children with insulin-dependent diabetes reveals closer association of tyrosine pyrophosphatase autoimmunity with DR4 than DQ8. Tissue Antigens. 1998;51:281–86. doi: 10.1111/j.1399-0039.1998.tb03103.x. [DOI] [PubMed] [Google Scholar]
- 19.Tsuji K, et al. HLA. 11th International Histocompatiblity Workshop; Oxford. 1991. [Google Scholar]
- 20.Erlich H, Valdes AM, Noble J, et al. Type 1 Diabetes Genetics Consortium. HLA DR-DQ haplotypes and genotypes and type 1 diabetes risk: analysis of the type 1 diabetes genetics consortium families. Diabetes. 2008;57:1084–92. doi: 10.2337/db07-1331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Heward JM, Mijovic CH, Kelly MA, Morrison E, Barnett AH. HLA-DQ and DRB1 polymorphism and susceptibility to type 1 diabetes in Jamaica. Eur J Immunogenetics. 2002;29:47–52. doi: 10.1046/j.0960-7420.2001.00276.x. [DOI] [PubMed] [Google Scholar]
- 22.Faber OK, Binder C, Markussen J, Heding LG, Naithani VK, Kuzuya H, Blix P, Horwitz DL, Rubenstein AH. Characterization of seven C-peptide antisera. Diabetes. 1978;27(Suppl 1):170–77. doi: 10.2337/diab.27.1.s170. [DOI] [PubMed] [Google Scholar]
- 23.Berger B, Stenström G, Sundkvist G. Random C-peptide in the classification of diabetes. Scand J Clin Lab Invest. 2000;60:687–93. doi: 10.1080/00365510050216411. [DOI] [PubMed] [Google Scholar]
- 24.Törn C, Landin-Olsson M, Lernmark A, Palmer JP, Arnqvist HJ, Blohmé G, Lithner F, Littorin B, Nyström L, Scherstén B, Sundkvist G, Wibell L, Ostman J. Prognostic factors for the course of β-cell function in autoimmune diabetes. J Clin Endocrinol Metab. 2000;85:4619–23. doi: 10.1210/jcem.85.12.7065. [DOI] [PubMed] [Google Scholar]
- 25.Antal Z, Jarchum I, DiLorenzo TP. HLA class I supertypes in type 1 diabetic children in an urban children’s hospital. Ann N Y Acad Sci. 2008;1150:86–89. doi: 10.1196/annals.1447.005. [DOI] [PMC free article] [PubMed] [Google Scholar]