| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Endocrinology and Metabolism, Department of Medicine (G.B., Y.T., E.C., T.F.D.), and Department of Psychiatry (D.A.G.), Mount Sinai School of Medicine, New York, New York 10029
Address all correspondence and requests for reprints to: Giuseppe Barbesino, M.D., Mount Sinai Medical Center, New York, New York 10128. E-mail: gb{at}doc.mssm.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
and Vß gene complexes, located on 14q11 and
7q35, respectively; and the Ig gene complex (IgH), located on 15q11. We
used polymorphic microsatellite markers located within these genes, or
gene complexes, to test for linkage (rather than association), to each
of these candidates. Using markers within the loci allowed us to assume
a fixed recombination fraction of 0.01 in the tested model. Three hundred eight subjects from 48 multiplex families were studied, with 142 affected subjects. Using this set of families, we have previously shown evidence of linkage with a major susceptibility locus for Graves’ disease (GD-1) on 14q24.3–31, with a maximum lod (logarithm + odds) score of 2.1, at a penetrance of 80% and with a dominant mode of inheritance. In the present study, we obtained consistently negative lod scores for each of the candidate genes, assuming either dominant or recessive modes of inheritance.
These data, therefore, showed evidence against linkage with all the candidate genes. Unlike association studies, linkage analyses detect major genetic influences on disease susceptibility exerted by the linked loci. The lack of linkage for the immunoregulatory genes that were studied indicated, therefore, that they were not major contributors to disease etiology.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
and Vß chains, and the
Ig heavy chain region (IGH). Some studies have found positive
associations, whereas others have not (see Ref. 3 for an extensive
review of the available data). Discrepancies between studies may be
explained by genetic differences between populations and/or by the use
of different polymorphisms for the same genes, making comparisons among
studies difficult. However, even in the reported positive studies, the
relative risk never exceeded 3.2, suggesting a limited influence on
disease expression for the genes analyzed.
In the present study, we present linkage analyses of a number of
candidate immunoregulatory genes in a large series of multiplex AITD
families. Genes studied included the T-cell receptor V and Vß
chains, the Ig heavy chain gene, and CTLA-4. Highly polymorphic
microsatellite markers within or close to each selected locus were used
in the analyses to be certain we were looking at effects associated
with the genes in question. Negative lod (logarithm + odds) scores were
observed with all the candidate genes studied, under all the
inheritance models analyzed, with both GD and HT. Our results allowed
us to exclude linkage with all the studied genes. These findings do not
contradict previous reports of a positive association of GD and/or HT
with some of the candidate genes we have analyzed, but the findings do
indicate that these genes are not major determinants of genetic
susceptibility to AITD.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 48 AITD multiplex families were analyzed. Forty-six families were of Caucasian descent, collected in New York, Italy, Britain, and Israel; 2 families were of Caribbean origin. Eleven families had at least 2 members with GD, whereas 20 families had at least 2 HT members. The remaining 17 families had at least two affected members, one of which had HT and the other had GD (mixed families). There were 29 families with 2 affected members, 14 families with 3, and 5 families with 4 or more, for a total of 142 patients. The diagnosis of GD was determined on the basis of documented clinical and laboratory evidence of present or past hyperthyroidism and the presence of at least one of the following: a diffuse goiter, positive TSH receptor antibody tests, or the presence of exophthalmos. The diagnosis of HT was based on past or present evidence of thyroid hormone-replaced primary hypothyroidism with any of the following: a diffuse, firm goiter, positive antithyroid peroxidase, or antithyroglobulin tests. Blood samples were collected from all available affected and nonaffected members of each family, after informed consent.
Allele determinations and typing
Whole blood was collected in tubes containing EDTA. Total
genomic DNA was extracted using a commercial method (Puregene; Gentra
Systems Inc., Minneapolis, MN). PCR reactions were conducted using
fluorescent label primers, following the procedure of Weber (8).
Briefly, each PCR reaction was performed in 10 µl of a mixture
containing 50 ng total DNA, 0.12 U Taq polymerase
(Perkin-Elmer, Foster City, CA), 5 pg fluorescent-labeled primers,
deoxyribose dinucleotide triphosphate, and 1.5 mmol MgCl2
buffer. The primer sequences and marker characteristics are described
in Table 1
. Fluorescent labeled PCR
products were denatured and separated on an ABI-310 (Applied
Biosystems, Foster City, CA) automated sequencer. Allele typing was
performed using Genotyper software in a semiautomated fashion.
|
Linkage analysis. Computer simulations were performed under
different models, simulating a dataset of 50 families (9). The results
showed maximum lod scores for a linked Mendelian locus ranging from 2.1
(at a penetrance of 20% and recessive mode of inheritance) to 19.25
(at 80% penetrance and dominant mode), suggesting that our dataset is
sufficient to detect linkage in a wide range of models. Experimental
family data were then analyzed using LIPED (10). The hypothesis of
positive linkage was tested in different models, assuming a dominant or
a recessive mode of inheritance. For each mode of inheritance,
different levels of penetrance (20, 40, 50, 60, and 80%) were tested.
Furthermore, to avoid the problem of age-dependent or reduced
penetrance, an affected-only analysis also was performed on the
dataset. Because all the markers studied were close to or within the
respective candidate genes, a low (0.01) recombination fraction (
)
was assumed in the model. Nonetheless, lod scores were obtained also
for up to 0.5 to examine the possibility of linkage to other genes
in the vicinity of the markers studied. Also, a positive lod score at a
distance from the analyzed marker could indicate heterogeneity (see
below). The hypothesis of linkage was rejected when lod scores were
-2.0 or less. lod scores were first obtained considering all affected
subjects, i.e. with either GD or HT (AITD affectedness). In
a second step, to verify whether the studied genes could have a role in
determining the type of AITD inherited, analyses were run with the same
parameters as above, but classifying only persons with GD as affected
or only with HT as affected. In each of these analyses, mixed families
were included; and when the GD affectedness was examined, HT patients
were considered as nonaffected and vice-versa. Because the total lod
scores obtained with this method represented the algebraic summation of
the lod scores obtained in each individual family, any results close to
zero could have resulted from the coexistence of different families
with very negative and very positive lod scores. This situation could
have resulted from the phenomenon of genetic heterogeneity, in which
the same phenotype was attributable to different genes in different
families. Therefore, lod scores obtained for individual families were
always examined in each of our analyses. In this situation, a test for
heterogeneity was also applied in the linkage analysis.
Transmission disequilibrium test (TDT). The TDT (11, 12) was
applied to the case of the CTLA-4 gene, for which the 106-bp allele of
the marker used has been previously associated with GD and HT (13, 14, 15, 16).
Briefly, all available parents heterozygous for the 106-bp allele and
any other of the alleles, with one or more affected offspring, were
included. The deviation of the transmission of the 106-bp allele from
the expected 50% ratio was then examined in the affected offsprings by
-square analysis.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At least 1 affected first-degree relative of the proband was present in each of the families, making our set of families a potentially highly informative one for analyzing linkage. Of the 48 families analyzed, 26 were nuclear families, whereas the others had more complex structure, with affected subjects in 3 generations and/or in parallel branches. A total of 308 subjects were analyzed, with 54 GD patients and 70 HT patients. The female/male ratio was 4.4 in GD patients and 10.1 in HT patients. To reduce the chance of mislabeling carrier subjects, because of age-dependent penetrance, children less than 18 years old were not included, unless affected.
AITD
Table 2 shows the maximum and
minimum lod scores obtained for the four markers at = 0.01, when
all AITD patients were considered affected. None of the markers showed
a maximum lod score more than 1 in either the recessive or the dominant
model and at any of the penetrances analyzed. Figure 1 shows total lod scores obtained in AITD
families with all four markers at = 0.01, considering different
levels of penetrance, in both the recessive and dominant mode of
inheritance. lod scores less than -2 were observed at all levels of
penetrance, in both the dominant and recessive model, the only
exception being CTLA-4, when assuming a penetrance of 20% and a
recessive mode of inheritance. Even in this extreme model, the lod
score was still negative (-1.6). Similar results were obtained in the
affected-only analysis, where all the candidates showed lod scores les
than -2 in both the recessive and dominant mode, with the only
exception being CTLA-4 in the recessive mode (-1.4). In none of the
models did any of the families show a lod score of +1.0 or higher,
suggesting that heterogeneity could not explain our results. When
examining lod scores obtained at higher , no evidence for linkage at
loci in the vicinity of any of the candidate genes was observed. In the
case of CTLA-4, however, a lod score of 1.008 was observed, at a of
0.2, penetrance 80%, in the dominant model (Fig. 2). A low positive lod score may be found
in cases of positive association and in the absence of linkage, at
relatively high (17). This could be explained by the fact that, in
the presence of an association, affected subjects are more likely to
carry the associated allele (therefore, yielding higher lod scores). At
the same time, the presence of many other affected subjects not
carrying the allele is interpreted in the linkage calculations as a
high frequency of recombinations, placing the maximum lod score at a
higher than expected distance from the marker itself. To verify the
hypothesis of positive association, we carried out a TDT on the
available data, as described in Subjects and Methods.
Another possible explanation for the results observed for CTLA-4 is the
presence of heterogeneity. However, analysis of the data for CTLA-4, in
a model assuming heterogeneity, showed a maximum lod score of 0.87 in
only 25% of the families, making heterogeneity an unlikely explanation
for our results.
|
|
|
Twenty-two of the parents of 1 or more AITD-affected members were heterozygous for the 106-bp allele of CTLA-4, with a total of 31 affected offsprings. Of these 31 affected family members, 16 inherited the 106-bp allele from 1 or 2 of the heterozygous parents, whereas 15 did not, a distribution which was very close to the 50% expected by chance alone (P, not significant). It must be noted, however, that our sample (not designed for the TDT specifically) is small and therefore able to detect only major associations.
GD and HT
Results similar to the combined AITD grouping were obtained when
the affectedness status of GD or HT were separately considered (Table 3). In general, the data set was less
informative, because of the smaller number of families available for
each analysis. Mainly negative results were observed; although, in all
the models studied, the lod score did not reach a sufficiently negative
level to be able to reject linkage with total certainty.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Many association studies have looked at immunoregulatory genes. The HLA
locus has been most intensely studied, and several haplotypes have been
shown to increase the risk for GD and/or for HT, with a maximum
relative risk of 3.8 for Caucasian GD patients with the HLA-DQA110501
haplotype (13, 24). Among the non-HLA genes, the T-cell receptor
and ß genes have been deemed particularly interesting because of the
central role of their products in the normal and aberrant immune
response. A polymorphism in the V chain gene was first noted to be
associated with HT in a small series of patients (25), but subsequent
studies have failed to confirm this finding (26). A similar situation
was found in GD, in which some studies have shown association of the
disease with the T-cell receptor ß chain (27), whereas others have
not (28, 29). The Ig heavy chain gene also has yielded positive results
in association studies (30) but not in linkage analyses (6). Recently,
much attention has been given to CTLA-4, a protein involved in antigen
presentation. CTLA-4 has emerged as a central factor in the homeostasis
and down regulation of the immune response (31) and, therefore,
represents an important candidate gene. The CTLA-4 locus has been shown
to be linked to IDDM and associated with GD (14) and, subsequently,
also with HT, with similar risk ratios (32, 33, 15, 16). Thus, CTLA-4
represents a significant susceptibility gene in addition to HLA.
We addressed the role of some of these genes (IGH, CTLA-4, TCR V,
and TCR Vß) by the method of linkage analysis. Linkage analysis has
been used successfully in mapping a large number of genes that cause
monogenic disease. However, it can also be used in locating
susceptibility genes for complex multigenic diseases, as shown by the
example of IDDM (34). Linkage has several advantages over association:
it allows the identification of genes that may be necessary for disease
development and does not involve possible biases caused by mismatched
control populations. Moreover, the markers employed may be many million
base pairs away from the disease locus and still show linkage. The
marker loci need only be in the genetic neighborhood of the disease
gene. Furthermore, linkage analysis allows whole genome scanning
without any previous knowledge of the locus sought (35). However,
linkage yields no information on disease mechanisms and becomes less
sensitive as the number of disease loci increases. Also, linkage
requires a large amount of data and is less sensitive, in the sense
that it fails to detect loci with small overall effects on the total
susceptibility to a multigenic disease.
In the present study, we used the candidate gene approach and chose highly polymorphic microsatellite markers from the noncoding sequence of each candidate gene or gene complex. This allowed us to assume low recombination fractions in the linkage model, thus adding a great deal of power to the analysis. By these means, we were able to fully exploit the intrinsic informativeness of the family data set, and we explored a wide range of inheritance models. In a subset of these families, we have previously shown evidence suggestive of linkage to a locus on chromosome 14q31, which we have termed GD-1, thus demonstrating that our set of families was adequate to screen the genome for candidate loci (36, 37).
The data presented here allowed the rejection of linkage with
AITD for a number of important immunoregulatory genes that have yielded
conflicting results in the earlier association studies, such as the TCR
V and Vß chains and the Ig heavy chain. The same results applied
to CTLA-4, which has been consistently associated with AITD in a number
of independent laboratories. We, therefore, tested the hypothesis of
association for CTLA-4 by the TDT in our set of families. Because our
series was not designed for such an approach, the data available
(mainly dependent on the number of available heterozygous parents) was
limited and was likely to detect only a large effect of CTLA-4.
However, the 106-bp allele was transmitted to almost 50% of the 31
affected offspring, providing no evidence for an association. It should
be remembered that in the earlier published association studies,
affected cases of GD and HT were randomly selected, independently of
their family history. For our linkage approach, probands were selected
based on the high incidence of thyroid autoimmune diseases in their
families. It is, therefore, possible that different genes have
different roles in familial AITD, as compared with sporadic AITD,
explaining the discrepancy observed in our data with CTLA-4. Indeed, an
association may only be present in a nonfamilial phenotype.
In conclusion, the linkage approach reported here initiates the process of excluding minor contributing genes on the way to finding the major genes that cause the human AITDs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Additional members of the International Consortium for the Genetics
of Autoimmune Thyroid Disease included: Drs. Rhoda Cobin (New York,
NY), Luca Chiovato and Aldo Pinchera (Pisa, Italy), Sandra McLachlan
(San Francisco, CA), Bernard Rees Smith (Cardiff, Wales), Fred Clark
and Eric Young (Newcastle upon Tyne, UK), and Meir Berezin (Tel
Hashomer, Israel). These contributors were not responsible for the
content of the present manuscript.
Received December 11, 1997.
Revised January 23, 1998.
Accepted February 2, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Matsuda, M. Tomita, J.-N. Uchihara, T. Okudaira, K. Ohshiro, T. Tomoyose, T. Ikema, M. Masuda, M. Saito, M. Osame, et al. Human T Cell Leukemia Virus Type I-Infected Patients with Hashimoto's Thyroiditis and Graves' Disease J. Clin. Endocrinol. Metab., October 1, 2005; 90(10): 5704 - 5710. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Tomer and T. F. Davies Searching for the Autoimmune Thyroid Disease Susceptibility Genes: From Gene Mapping to Gene Function Endocr. Rev., October 1, 2003; 24(5): 694 - 717. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Dittmar and G. J. Kahaly Polyglandular Autoimmune Syndromes: Immunogenetics and Long-Term Follow-Up J. Clin. Endocrinol. Metab., July 1, 2003; 88(7): 2983 - 2992. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Jin, W. Teng, S. Ben, X. Xiong, J. Zhang, S. Xu, Y. Y. Shugart, L. Jin, J. Chen, and W. Huang Genome-Wide Scan of Graves' Disease: Evidence for Linkage on Chromosome 5q31 in Chinese Han Pedigrees J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1798 - 1803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Anjos, A. Nguyen, H. Ounissi-Benkalha, M.-C. Tessier, and C. Polychronakos A Common Autoimmunity Predisposing Signal Peptide Variant of the Cytotoxic T-lymphocyte Antigen 4 Results in Inefficient Glycosylation of the Susceptibility Allele J. Biol. Chem., November 22, 2002; 277(48): 46478 - 46486. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ando, M. Imaizumi, P. N. Graves, P. Unger, and T. F. Davies Intrathyroidal Fetal Microchimerism in Graves' Disease J. Clin. Endocrinol. Metab., July 1, 2002; 87(7): 3315 - 3320. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Sakai, S. Shirasawa, N. Ishikawa, K. Ito, H. Tamai, K. Kuma, T. Akamizu, M. Tanimura, K. Furugaki, K. Yamamoto, et al. Identification of susceptibility loci for autoimmune thyroid disease to 5q31-q33 and Hashimoto's thyroiditis to 8q23-q24 by multipoint affected sib-pair linkage analysis in Japanese Hum. Mol. Genet., June 1, 2001; 10(13): 1379 - 1386. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Imrie, B. Vaidya, P. Perros, W. F. Kelly, A. D. Toft, E. T. Young, P. Kendall-Taylor, and S. H. S. Pearce Evidence for a Graves' Disease Susceptibility Locus at Chromosome Xp11 in a United Kingdom Population J. Clin. Endocrinol. Metab., February 1, 2001; 86(2): 626 - 630. [Abstract] [Full Text]
|
|
|
|
|
|
Y. Ban, M. Taniyama, and Y. Ban Vitamin D Receptor Gene Polymorphism Is Associated with Graves' Disease in the Japanese Population J. Clin. Endocrinol. Metab., December 1, 2000; 85(12): 4639 - 4643. [Abstract] [Full Text]
|
|
|
|
|
|
T. H. Brix, K. O. Kyvik, and L. Hegedüs A Population-Based Study of Chronic Autoimmune Hypothyroidism in Danish Twins J. Clin. Endocrinol. Metab., February 1, 2000; 85(2): 536 - 539. [Abstract] [Full Text]
|
|
|
|
|
|
Mapping the Major Susceptibility Loci for Familial Graves' and Hashimoto's Diseases: Evidence for Genetic Heterogeneity and Gene Interactions J. Clin. Endocrinol. Metab., December 1, 1999; 84(12): 4656 - 4664. [Abstract] [Full Text]
|
|
|
|
|
|
B. Vaidya, H. Imrie, P. Perros, E. T. Young, W. F. Kelly, D. Carr, D. M. Large, A. D. Toft, M. I. McCarthy, P. Kendall-Taylor, et al. The cytotoxic T lymphocyte antigen-4 is a major Graves' disease locus Hum. Mol. Genet., July 1, 1999; 8(7): 1195 - 1199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Heward, A. Allahabadia, M. Armitage, A. Hattersley, P. M. Dodson, K. Macleod, J. Carr-Smith, J. Daykin, A. Daly, M. C. Sheppard, et al. The Development of Graves' Disease and the CTLA-4 Gene on Chromosome 2q33 J. Clin. Endocrinol. Metab., July 1, 1999; 84(7): 2398 - 2401. [Abstract] [Full Text]
|
|
|
|
|
|
Y. Tomer, G. Barbesino, D. A. Greenberg, E. Concepcion, and T. F. Davies Linkage Analysis of Candidate Genes in Autoimmune Thyroid Disease. III. Detailed Analysis of Chromosome 14 Localizes Graves' Disease-1 (GD-1) Close to Multinodular Goiter-1 (MNG-1) J. Clin. Endocrinol. Metab., December 1, 1998; 83(12): 4321 - 4327. [Abstract] [Full Text]
|
|
|
|
|
|
T. F. Davies Autoimmune Thyroid Disease Genes Come in Many Styles and Colors J. Clin. Endocrinol. Metab., October 1, 1998; 83(10): 3391 - 3393. [Full Text]
|
|
|
|
|
|
G. Barbesino, Y. Tomer, E. S. Concepcion, T. F. Davies, and D. A. Greenberg Linkage Analysis of Candidate Genes in Autoimmune Thyroid Disease. II. Selected Gender-Related Genes and the X-Chromosome J. Clin. Endocrinol. Metab., September 1, 1998; 83(9): 3290 - 3295. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
