| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Pediatrics, Seoul National University (J.Y.), Seoul, Korea; Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center ( L.Y., K.B., M.H., M.R., G.S.E.), Denver, Colorado 80262; and Roche Molecular Systems (T.L.B., H.A.E.), Alameda, California 94501
Address all correspondence and requests for reprints to: George S. Eisenbarth, M.D., Ph.D., Barbara Davis Center for Childhood Diabetes, University of Colorado Health Sciences Center, Campus Box B140, 4200E 9th Avenue, Denver, Colorado 80262. E-mail: george.eisenbarth{at}uchsc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Eight relatives of 478 (1.7% of SOC) expressed a transient autoantibody, and none had the high risk genotype DR3/4(DQ2/8). In contrast, 28 relatives (5.9%) had persistent antiislet autoantibodies, and 14 (50%) were DR3/4(DQ2/8) heterozygotes. Thirteen children of 793 (1.6% of NEC) expressed a transient autoantibody, and none had the high risk genotype DR3/4(DQ2/8). Seven of the NEC (0.9%) had persistent antiislet autoantibodies, and 4 (57.1%) were DR3/4(DQ2/8) heterozygous.
Expression of persistent autoantibodies was strongly related to human leukocyte antigen status and family history of type 1 diabetes. In contrast, the expression of transient antiislet autoantibodies did not differ by family history of diabetes, and none of the DR3/4(DQ2/8) relatives and DR3/4(DQ2/8) newborns expressed transient autoantibodies.
Our results indicate that children can express transient antiislet autoantibodies, but such transient autoantibodies are relatively infrequent and are not correlated with known genetic risk factors for type 1 diabetes.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Given the above progress in assay determination combined with rigid
quality control protocols, we thought that it would be possible to
evaluate determinants of both persistent and transient autoantibody
expression. We hypothesized that genetic determinants of the expression
of persistent antiislet autoantibodies would similarly influence the
expression of transient autoantibodies. In the current study based on
the DAISY (Diabetes Autoimmunity Study in the Young) population (both
relatives of patients with type 1 diabetes and newborns from the
general population identified by human leukocyte antigen (HLA) typing
of 
20,000 newborns), we disprove the above hypothesis. HLA status
and family history of diabetes dramatically influence the expression of
persistent, but not transient, antiislet autoantibodies.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We have prospectively evaluated 481 siblings or offspring of
patients with type 1 diabetes (SOC group) and 796 newborns from the
general population (NEC group). The general population children were
selected for expression of specific HLA haplotypes (DR3, DR4, or both)
and did not have a relative with diabetes (NEC: newborn nonrelative
cohort). The number of subjects enrolled according to the age of
initial study and follow-up duration are presented in the Table 1
. Among them, 478 relatives and 793
newborns could be analyzed because subjects categorized as suggested
transplacental (?TP) or last positive (LP) group were excluded from
analysis as we could not determine whether they have transient or
persistent autoantibodies.
|
) of the SOC were first studied before 1 yr of age
(124 of 481), whereas by protocol, with cord blood analysis of HLA, the
great majority (743 of 796, or 93%) of the NEC were first studied
before 1 yr of age. The median age of relatives at initial assay was
2.82 yr (range, 0–8.03 yr), and that of the nonrelative newborns was
0.76 yr (range, 0–2.66 yr). Among 481 relatives the median length of
follow-up from the initial date of the serum tested was 2.01 yr or 3
clinic visits (maximum, 4.83 yr or 14 clinic visits), and that in the
NEC was 1.26 yr or 3 clinic visits (maximum, 4.11 yr or 8 clinic
visits). One thousand five hundred and eighty-five sera from 481
relatives and 2,320 sera from 796 newborns from the general population
were assayed for 3 antiislet autoantibodies (3,905 sera and 11,715
autoantibody determinations). We measured random blood glucose and
hemoglobin A1C whenever we evaluated individuals
at 3 month follow-up (those with persistent and those with transient
autoantibodies). The diagnosis of diabetes was made using National
Diabetes Data Group criteria (12). Subjects or their parents gave
informed consent to be studied, and the protocol was approved by the
institutional review board of the University of Colorado. Autoantibody assays
Serum samples were stored at -20 C before testing. Glutamic acid decarboxylase autoantibodies (GAAs) were measured in duplicate by a radioassay using in vitro transcribed and translated recombinant human GAD (65-kDa isoform) and precipitation with protein A-Sepharose (13). The interassay coefficient of variation in our laboratory is 6.5% (n = 10). ICA512bdc autoantibodies (ICA512AAs) were measured in duplicate using a similar assay format, but with in vitro transcribed and translated ICA512bdc. The interassay coefficient of variation in our laboratory is 11.7% (n = 9). Insulin autoantibodies (IAAs) were measured previously by a fluid phase radioassay incorporating competition with cold insulin and precipitation with polyethylene glycol (14). The interassay coefficient of variation for the IAA assay is 10.3% at low positive levels. All antiinsulin autoantibodies positive in the polyethylene glycol-based assay were retested with a protein A/G-based assay (15). This protein A/G-based IAA assay is called a micro-IAA (mIAA) assay and is performed in a 96-well plate. Human [125I]insulin of 20,000 cpm is incubated with 5 µL serum with and without nonlabeled human insulin, respectively, at a 1:5 dilution of serum for 3 days at 4 C. We washed with the same methods used in the GAA and ICA512bdcAA assays, precipitated with protein A/G-Sepharose, and counted with a TopCount (96-well plate ß-counter, Packard, Downers Grove, IL) scintillation counter. We measured mIAAs in duplicate. The interassay coefficient of variation is 11% (n = 8). Only samples confirmed positive with the protein A/G assay were considered positive.
The results of the GAD65, ICA512bdc, and microinsulin autoantibody assays were expressed as an index calculated from the counts per min for the test sample compared to those for a positive and a negative control sample. The upper limits of the normal range for IAAs (42 nU/mL; standard assay) and GAAs (index of 0.032) were established as the 99th percentile of the levels in 198 healthy control subjects, and those of mIAAs (index of 0.01) were established as the 99th percentile of the levels in 106 healthy control subjects. For ICA512bdcAAs, the limit of normal (index of 0.071) was established as the 100th percentile in 198 healthy controls because increasing the cut-offs from the 99th percentile in control subjects (index of 0.048) to 0.071 increased specificity from 99% to 100%, with no loss of sensitivity. In the Immunology of Diabetes Society’s Combined Autoantibody Workshop of 1995, sensitivity (for diabetes at less than age 30 yr) for the GAA assay was 82%, and specificity was 99%. Sensitivity for the ICA512AA assay was 73%, and specificity was 100%. For the mIAA assay reported in our laboratory, sensitivity was 66% and specificity was 99% for 106 healthy control subjects and 105 patients with recent-onset diabetes (Yu et al., unpublished data).
Sera found to be positive for a given autoantibody as well as 10% of negative sera were reassayed in a blinded manner for antiislet autoantibodies and were confirmed positive if two of three measurements were positive. If in the repeat assay the sample was negative, the serum was assayed again; if the result was again negative, then the sample was considered negative.
HLA typing
HLA DQA1 alleles were typed using HLA DQ
PCR amplification
and a reverse dot blot typing kit (AmpliType produced by
Perkin-Elmer Corp., Norwalk, CT). HLA-DRB1 and DQB1
alleles were typed using PCR and sequence-specific oligonucleotide
probes (16, 17). HLA typing was available for 468 relatives and 796 of
the NEC.
Statistical analysis
Fisher’s exact test was used for comparison of the proportion of individuals who were autoantibody positive. Levels of positive autoantibodies were compared using the rank sum test.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
illustrates the age at initial testing and the age at last
test for the SOC and NEC. As illustrated in Table 2, children who expressed autoantibodies
were divided into seven categories (negative, persistent, a single
transient positive sample, multiple samples positive but transient,
only the last sera sample tested positive, transplacental antibody, and
possible transplacental antibody). The categories were defined by the
pattern of autoantibody expression on follow-up. A serum sample
positive for any of the three measured autoantibodies was considered
positive for determination of categories. It was uncommon for
individuals with persistent autoantibodies not to be positive for the
same autoantibody(ies) at subsequent visits. Only five children in the
entire study (all in the SOC group) lost a single autoantibody but
expressed other autoantibodies. These five children were positive for
all three autoantibodies, lost one of three autoantibodies at the last
follow-up sample (one IAA and four GAA lost), and were defined as
persistent.
|
Among the SOC group, 8 (1.7%) children had transient antiislet
autoantibodies, including only 3 (0.6%) with transient autoantibodies
that were present on more than 1 occasion, and 5 relatives (1.1%) had
a single transient antiislet autoantibody (Fig. 1). None of the 8 relatives with
transient antibodies had autoantibodies reacting with more than 1 islet
autoantigen, and none had the high risk genotype DR3/4(DQB1*0302)
(Figs. 2 and 3). The autoantibody levels of 1 of these
SOC children with transient antiislet autoantibodies is shown in Fig. 4. Among the NEC group, 13 (1.6%)
children had transient antiislet autoantibodies including 2 subjects
(0.25%) with a transient antibody which was present on more than 1
occasion and 11 (1.4%) with a single positive sample (Fig. 1). None of
the 13 newborns with transient antibodies had autoantibodies reacting
with more than 1 islet autoantigen, and none had the high risk
genotype DR3/4(DQB1*0302) (Figs. 2 and 3). The autoantibodies levels of
1 of these NEC children with transient antiislet autoantibodies are
shown in Fig. 4.
|
|
|
|
Twenty-eight relatives (5.9% of the SOC group) had persistent
antiislet autoantibodies (Fig. 1). Fourteen were DR3/4(DQB1*0302), and
the majority (22 of 28, or 78.6%) expressed multiple
autoantibodies (Tables 3 and 4 and Figs. 2, 3, and 5). Among the NEC group, 7 (0.9%) had
persistent antiislet autoantibodies, and 4 of 7 (57.1%) were
DR3/4(DQB1*0302) heterozygous. The majority (5 of 7, or 71.4%)
expressed multiple autoantibodies (Tables 3 and 4 and Figs. 2, 3, and 5). Expression of persistent autoantibodies was strongly related to HLA
in the SOC group [28.6% of DR3/4(DQB1*0302) relatives were positive
vs. 3.3% not DR3/4; P < 0.0001; Fig. 2].
A family history of type 1 diabetes was also strongly associated with
expression of persistent autoantibodies even among DR3/4(DQB1*0302)
individuals (DR3/4 SOC, 28.6%; DR3/4 NEC, 1.9%; P <
0.0001; Fig. 2). Among the NEC group, persistent autoantibodies may
also be related to HLA [DR3/4(DQB1*0302) NEC, 1.9%; not DR3/4 NEC,
0.5%; P = 0.08]. When analyzing only SOC and NEC
children, who were first studied at less than 1 yr of age, the pattern
of expression of persistent autoantibodies remained the same [DR3/4
SOC, 6 of 17 positive (35%); DR3/4 NEC, 4 of 199 positive (2%;
P < 0.0001); not DR3/4 SOC, 4 of 100 positive
(4%); not DR3/4 NEC, 3 of 544 positive (0.6%; P
< 0.05; Table 5].
|
|
|
|
).
Progression to diabetes was associated with persistent expression of
autoantibodies reacting with more than one islet antigen. Of the SOC
group, six have developed overt diabetes, and of these, three were
first studied before 1 yr of age. Of the NEC group, two have developed
overt diabetes. All six prediabetic relatives and two prediabetic NECs
expressed multiple autoantibodies and were categorized as persistent
and had HLA DR3/4 (DQ2/8) type (Table 3).
Autoantibody levels at first sample positive for an autoantibody
The levels of positive GAD autoantibodies were significantly
higher in the individuals with persistent autoantibody(ies) compared to
the levels in those with transient autoantibody expression (by rank sum
test, P < 0.01; Fig. 6).
The levels of positive ICA512 or insulin autoantibodies in individuals
with persistent autoantibody(ies) were not significantly different from
the levels in individuals with transient autoantibody expression (Fig. 6).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The above assays we believed would aid in evaluation of the controversial question of whether antiislet autoantibodies frequently remitted. In addition, we hypothesized that transient antiislet autoantibodies might be influenced by genetic and familial risk factors in a manner similar to persistent antiislet autoantibodies. A number of studies have described frequent fluctuations in antiislet autoantibodies, particularly with the cytoplasmic ICA assay (18, 19, 20). In one report, fluctuations in such autoantibodies occurred in parallel in multiple members of the same family (21). Palmer and co-workers described frequent remission of autoantibodies, in particular cytoplasmic ICA, and hypothesized that the type 1 diabetes results from multiple immunological events, with a waxing and waning course (18). The occurrence of such transient autoantibodies might be related to environmental factors either increasing or decreasing risk of diabetes.
In the current study we found that expression of transient antiislet autoantibodies was not associated with the major risk factors associated with persistent autoantibodies. In particular, neither having a relative with type 1 diabetes nor expression of the HLA DR3/4 (DQ2/8) genotype increased the proportion of individuals expressing transient antiislet autoantibodies. In contrast persistent antiislet autoantibodies were markedly influenced by both of these factors, with 29% of relatives of patients with type 1 diabetes with DR3/4 (DQ2/8) expressing persistent autoantibodies compared to 3.3% of relatives without this HLA genotype and 1.9% of DR3/4 (DQ2/8) children from the nonrelative population. The association of DR3/4 (DQ2/8) with a high risk of relatives to develop antiislet autoantibodies is similar to the findings of Schenker and colleagues for the BABYDIAB study (22). Transient antiislet autoantibodies also differed markedly from persistent antiislet autoantibodies, in that sera with transient autoantibodies had antibodies reacting with a single islet antigen, whereas the great majority of individuals with persistent antiislet autoantibodies had sera reacting with multiple antiislet autoantibodies (for the SOC group, 22 of 28; for the NEC group, 5 of 7 with multiple antigens recognized). Transient GAD antiislet autoantibodies usually, although not always, had a low index value and most often occurred on only a single serum sample. Finally, persistent antiislet autoantibodies were associated with a high risk of progression to diabetes [8 of 35 (23%) developed diabetes to date with less than 2 yr mean follow-up], whereas none of the children with transient autoantibodies have progressed to diabetes (0 of 21).
Transient antiislet autoantibodies occurred in less than 1 of 50 relatives or individuals from the general population. Given assays with cut-offs set at the first percentile we believe that this is relatively infrequent, although compared to the number of persistent positives in the NEC group, it is a significant occurrence. Of note no individual with the high risk genotype (DR3/4, DQ2/8) expressed transient autoantibodies. We did observe transient antiislet autoantibodies in 8 of 478 relatives and 13 of 793 general population children. Four individuals in the transient category expressed an autoantibody on more than one occasion. We cannot be certain that such individuals will not be at higher risk of developing persistent antiislet autoantibodies or progressing to diabetes. It is also not clear whether the transient antiislet autoantibodies observed are associated with antiislet autoimmunity. The lack of multiple islet antigens recognized argues somewhat against specific antiislet autoimmunity. The hypothesis we favor is that a subset of the transient autoantibodies observed may represent cross-reactivity of antibodies generated for example to environmental factors that by chance also bind to a single epitope of GAD, insulin, or ICA512. Studies of the epitopes recognized by transient and persistent autoantibodies should allow initial testing of this hypothesis.
A practical result of this investigation of antiislet autoantibodies of young children is the recognition that transient anti-slet autoantibodies can occur, but sera reacting with multiple autoantigens and autoantibody(ies) present on more than one occasion increase the probability of persistent autoantibody expression and progression to diabetes. In addition, the general stability of autoantibody expression may provide a surrogate marker to test therapies aimed at preventing ß-cell destruction. In particular, with more than 90% of individuals with high risk antibody patterns retaining autoantibody expression, relatively small studies could determine whether a therapy suppresses antibody expression. A caveat is that a given therapy may suppress autoantibody expression but not favorably influence T cell-mediated islet destruction, or a therapy may prevent islet destruction but not alter autoantibodies. At present we have no therapy to suppress or prevent the development of antiislet autoantibodies. If we had such a therapy, it would be a likely candidate to evaluate for prevention of type 1 diabetes.
|
Footnotes |
|---|
2 Consultant to Quest and Immco Diagnostics Laboratory, companies
that have interest in developing autoantibody assays for use in the
diagnosis and prediction of type 1 diabetes.
Received October 6, 1999.
Revised March 28, 2000.
Accepted March 29, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Makinen, T. Harkonen, J. Ilonen, M. Knip, and the Finnish Pediatric Diabetes Register Characterization of the humoral immune response to islet antigen 2 in children with newly diagnosed type 1 diabetes. Eur. J. Endocrinol., July 1, 2008; 159(1): 19 - 26. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mayr, M. Schlosser, N. Grober, H. Kenk, A. G. Ziegler, E. Bonifacio, and P. Achenbach GAD Autoantibody Affinity and Epitope Specificity Identify Distinct Immunization Profiles in Children at Risk for Type 1 Diabetes Diabetes, June 1, 2007; 56(6): 1527 - 1533. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kondrashova, H. Viskari, P. Kulmala, A. Romanov, J. Ilonen, H. Hyoty, and M. Knip Signs of {beta}-Cell Autoimmunity in Nondiabetic Schoolchildren: A comparison between Russian Karelia with a low incidence of type 1 diabetes and Finland with a high incidence rate Diabetes Care, January 1, 2007; 30(1): 95 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Achenbach, E. Bonifacio, K. Koczwara, and A.-G. Ziegler Natural History of Type 1 Diabetes Diabetes, December 1, 2005; 54(suppl_2): S25 - S31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kukko, T. Kimpimaki, S. Korhonen, A. Kupila, S. Simell, R. Veijola, T. Simell, J. Ilonen, O. Simell, and M. Knip Dynamics of Diabetes-Associated Autoantibodies in Young Children with Human Leukocyte Antigen-Conferred Risk of Type 1 Diabetes Recruited from the General Population J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2712 - 2717. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Barker, K. J. Barriga, L. Yu, D. Miao, H. A. Erlich, J. M. Norris, G. S. Eisenbarth, and M. Rewers Prediction of Autoantibody Positivity and Progression to Type 1 Diabetes: Diabetes Autoimmunity Study in the Young (DAISY) J. Clin. Endocrinol. Metab., August 1, 2004; 89(8): 3896 - 3902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Fronczak, A. E. Baron, H. P. Chase, C. Ross, H. L. Brady, M. Hoffman, G. S. Eisenbarth, M. Rewers, and J. M. Norris In Utero Dietary Exposures and Risk of Islet Autoimmunity in Children Diabetes Care, December 1, 2003; 26(12): 3237 - 3242. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kimpimaki, P. Kulmala, K. Savola, A. Kupila, S. Korhonen, T. Simell, J. Ilonen, O. Simell, and M. Knip Natural History of {beta}-Cell Autoimmunity in Young Children with Increased Genetic Susceptibility to Type 1 Diabetes Recruited from the General Population J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4572 - 4579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Knip Can We Predict Type 1 Diabetes in the General Population? Diabetes Care, March 1, 2002; 25(3): 623 - 625. [Full Text] [PDF]
|
|
|
|
|
|
K. Savola, E. Laara, P. Vahasalo, P. Kulmala, H. K. Akerblom, and M. Knip Dynamic Pattern of Disease-Associated Autoantibodies in Siblings of Children With Type 1 Diabetes: A Population-Based Study Diabetes, November 1, 2001; 50(11): 2625 - 2632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kimpimaki, A. Kupila, A.-M. Hamalainen, M. Kukko, P. Kulmala, K. Savola, T. Simell, P. Keskinen, J. Ilonen, O. Simell, et al. The First Signs of {beta}-Cell Autoimmunity Appear in Infancy in Genetically Susceptible Children from the General Population: The Finnish Type 1 Diabetes Prediction and Prevention Study J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4782 - 4788. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. E. Naserke, E. Bonifacio, and A.-G. Ziegler Prevalence, Characteristics and Diabetes Risk Associated with Transient Maternally Acquired Islet Antibodies and Persistent Islet Antibodies in Offspring of Parents with Type 1 Diabetes J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4826 - 4833. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
