Full text of DOI:10.1007/BF03343819

J. Endocrinol. Invest. 24: 88-91, 2001
HLA DQA1*0501 and DRB1*0301 antigens do not
independently convey susceptibility to Graves’ disease
G. Philippou*, A. Krimitzas*, G. Kaltsas*, E. Anastasiou*, A. Souvatzoglou*,
and M. Alevizaki**
*
1st Endocrine Section, and **Department of Medical Therapeutics, Athens University School of
Medicine, Alexandra Hospital, Athens, Greece
ABSTRACT. Genes of, or closely associated to, the
HLA complex are assumed to contribute to the ge-
netic predisposition of Graves’ disease. The aim of
this study was to investigate the presence of the
HLA DQA1*0501 and DRB1*0301 antigens in
Greek patients with Graves’ disease. In addition,
we tried to establish if there is any association be-
tween these antigens and any of the clinical mani-
festations of the disease. We examined 117 pa-
tients with Graves’ disease and 104 healthy con-
trols. DNA was extracted from peripheral lympho-
cytes and the HLA DQA1*0501 and DRB1*0301
genomic regions were amplified by PCR and char-
acterized by hybridization with sequence specific
oligonucleotides (SSO). Two of the patients had a
positive family history for Graves’ disease and 46
had clinical thyroid eye disease (TED). The fre-
quencies of both DQA1*0501 and DRB1*0301 anti-
gens were significantly increased in patients com-
pared to controls (relative risk [RR] 4.2 and 4.5 for
each antigen respectively). Neither of these two
antigens was an independent risk factor for Graves’
disease. However, the combination of both these
HLA antigens resulted in a striking increase in the
RR for development of Graves’ disease especially in
females (RR/F=27, RR/M=8.4). No association was
found between these antigens and positive family
history or the presence of TED. These data suggest
that HLA DQA1*0501 and DRB1*0301 antigens are
not independent risk factors for the development
of Graves’ disease. On the contrary, the presence
of both these alleles results in a significant increase
in the RR for the development of Graves’ disease in
the Greek population, particularly in females.
(J. Endocrinol. Invest. 24: 88-91, 2001)
©
2001, Editrice Kurtis
INTRODUCTION
probably in linkage with DR3 (5). However, there are
conflicting results in the literature as to whether this
represents an independent risk factor or even more
whether it is associated with the sex of patients (6-8).
The aim of the present study was to examine the
frequency of DQA1*0501 and of one of the poly-
morphisms of the previously known as DR3 antigen
(DRB1*0301) in Greek patients with Graves’ dis-
ease. We also examined if the DQA1*0501 is an in-
dependent risk factor for the disease and whether
it is associated with gender, TED and positive fam-
ily history for the disease.
Graves’ disease is common in the general popula-
tion; the frequency of the disease is 2.7% in females
and 0.3% in males (1). A significant proportion of pa-
tients have a positive family history of the disease. Its
pathogenesis remains largely unknown, however it is
believed to be a multifactorial disease, since a com-
bination of environmental and genetic factors appear
to contribute to its manifestation. It has long been
recognized that both Graves’ disease as well as thy-
roid eye disease (TED) are associated with specific
HLA antigens. HLA-DR3 and its polymorphisms have
been consistently associated with the disease in
Caucasians (2-4). More recently, an association with
the antigen DQA1*0501 has been reported which is
MATERIALS AND METHODS
A total of 117 patients with Graves’ disease were ex-
amined (93 F and 24 M). Diagnosis was based on the
presence of hyperthyroid syndrome associated with
elevated T3 and T4 levels and suppressed TSH levels.
One of the following criteria was used to confirm the
autoimmune nature of the hyperthyroidism: diffuse
thyroid enlargement with diffuse radionucleotide up-
Key-words: Graves’ disease, HLA, autoimmunity, thyroid eye disease.
Correspondence: Dr. George Philippou, 1st Endocrine Section, Alexandra
Hospital, 80 Vass. Sofias Ave, 115 28 Athens, Greece.
E-mail: geophi@hol.gr
Accepted 24 July, 2000.
88

HLA antigens and Graves’ disease
Table 1 - HLA DQA1*0501 and DRBI*0301 antigens in Graves’ disease patients and controls.
Graves’ disease
Normal controls
DQAI*0501 (+)
30 (44)
N
24
DQAI*0501 (+)
N
X2
8.7
p
RR
4.8
3.1
4.2
Male
19 (79)
74 (80)
68
36
<0.01
<0.01
<0.001
Female
Total
93
20 (55)
7.6
117
93 (79.5)
104
50 (48.1)
DRBI*0301 (+)
24 (35)
23.8
DRB1*0301 (+)
Male
24
93
15 (62)
68
36
5.3
<0.05
<0.001
<0.001
3.1
7.3
4.5
Female
Total
63 (68)
78 (67)
8 (22)
21.7
28.4
117
104
32 (30)
Percentage in parenthesis. RR: relative risk; (+) positive.
take in the thyroid scan, presence of TED, or elevat-
ed thyrotropin receptor antibodies (TRAb). The TRAb
antibodies were determined by a radioreceptor as-
say (Biocode biotechnology, Belgium). Our results
were compared to those in 104 random healthy
blood donors (36 F/68 M) who did not have a positive
family history for autoimmune thyroiditis and did not
have antithyroid autoantibodies in their serum.
CAACTAC (5). Results were obtained after an over-
night exposure to autoradiography film.
The results were analyzed using 2x2 contingency
tables and the Fisher exact test for calculating the
chi-square values and odds ratio and the corre-
sponding confidence intervals (C.I.) for evaluation of
the relative risk (RR).
Genomic DNA was extracted from peripheral lym-
phocytes and the second exon of the DQA1*0501
gene was amplified by PCR using the sequence
specific primers F: GGTGTAAACTTGTACCAG and
R: GGTAGCAGCGGTAGAGTTG (5). The following
PCR conditions were used, 30 cycles of 1 min de-
naturation at 94 C, 1 min annealing at 58 C and 1
min extension at 72 C. One microlitre of PCR reac-
tion was dot-blotted on nylon membranes and was
hybridized to the oligonucleotide probe TCAGA-
CAATTTAGATTTG (9). Hybridization conditions
were according to Yanagawa et al., 1993. For
DRB1*0301 genotyping the second exon was am-
plified by PCR using the following primers F: CCC-
CACAGCACGTTTCTTG and R: CCGCTGCACT-
GTGAAGCTCT. The PCR conditions were the same
as described above for DQA1* 0501. The probe
used for hybridization was GGCCGGGTGGA-
RESULTS
Twenty-two patients had a positive family history of
Graves’ disease and 46 patients had signs of TED
(class II-IV, Werner classification) (10). HLA DQA1*
0501 antigen was detected in 79.5% of patients vs
2
48.1% of controls (Table 1) (X =23.8, p<0.001, RR
4.2). The presence of the antigen was significantly in-
creased both in male and female patients, the RR be-
ing higher in males (Table 1). An increased frequen-
cy in both male and female patients was also found
for the HLA DRB1*0301 antigen (67% in patients vs
2
30% in controls) (X =28.4, p<0.001, RR 4.5) (Table 1).
Neither of the two antigens was able to confer sus-
ceptibility to Graves’ disease in the absence of the
other, i.e. as an independent risk factor (Table 2).
On the contrary, when both of these antigens were
present, the relative risk was even higher (13.8). This
Table 2 - Association of the presence of one of the two HLA haplotypes in the absence of the other with the development of Graves’
disease.
Graves’
25
Controls
38
X2
p
DQA1*0501 (+) DRB1*0301 (-)
DQA1*0501 (-) DRB1*0301 (-)
DRB1*0301 (+) DQA1*0501 (-)
DQA1*0501 (-) DRB1*0301 (-)
NS: not significant; (+) positive; (-) negative.
1.3
NS
14
34
10
20
0.1
NS
14
34
89

G. Philippou, A. Krimitzas, G. Kaltsas, et al.
Table 3 - Association of the presence of both DRB1*0301 and DQAI*0501 HLA antigens with the development of Graves’ disease.
Graves’
DQA1*0501 (+)
Controls
DQA1*0501 (-)
DRB1*0301 (-)
DQA1*0501 (+)
DQA1*0501 (-)
DRB1*0301 (-)
X2
p
RR
DRB1*0301 (+)
DRB1*0301 (+)
Male
14 (58)
4
10 (42)
24
10
34
11.1 <0.001 8.4
24.0 <0.001 27.0
40.6 <0.001 13.8
Female
Total
54 (96)
10
14
2 (4)
68 (85)
12 (15)
Percentage in parenthesis. RR: relative risk; (+) positive; (-) negative.
was particularly evident in female patients (RR=27
for females and 8.4 for males) (Table 3).
No associations were found between these two
HLA polymorphisms and either TED or a family his-
tory of Graves’ disease (Table 4).
analyzing the data in the study of Cuddihy et al.,
(7) where again there was no patient who was DR
positive but DQ negative. These results show the
close linkage between these antigens in agreement
with what has previously been suggested (16).
We found no association of these antigens with the
presence of a family history of Graves’ disease
which suggests that the loci responsible for the pre-
disposition to familial disease may be multiple and
heterogeneous (16). Furthermore, there was no as-
sociation of these two polymorphisms with the
presence of TED, which suggests that other loci
may be linked to the extrathyroidal manifestations
of Graves’ disease (17, 18).
An interesting result of our study was that the pres-
ence of both these antigens, rather than each indi-
vidually, greatly increased the relative risk to de-
velop Graves’ disease, which in the female popu-
lation was as high as 27, indicating that these two
loci are probably part of a haplotype which does
predispose to Graves’ disease.
In recent years the genetic component is believed
to be less important for the development of Graves’
disease than previously considered (19). This no-
tion is further supported by the recent observation
that the discordance rate for the disease in monozy-
gotic twins is higher than expected (20). Never-
theless, our results indicate that if the genetic com-
ponent is of some importance for the development
of Graves’ disease, then the combination of the
HLA DRB1*0304 and DQA1*0501 loci may be part
of a combined haplotype which is important for the
DISCUSSION
This study in a Greek population confirms the associ-
ation of HLA DRB1*0301, previously known as DR3, as
well as of DQA1*0501 with Graves’ disease. These as-
sociations have previously been observed in other
populations as well (8, 11, 12). We also found, as pre-
viously reported, that the presence of the DRB1*0301
antigen confers a higher RR in women (2, 4, 11). In ad-
dition, we found that the presence of the DQA1*0501
antigen confers a relatively higher RR in male compa-
red to female patients. This is in agreement with some
(
6) but not all of relevant previous studies (7, 8) which
found a higher female RR. Our results, as well as those
reported by Yanagawa et al., (6) support the hypoth-
esis that perhaps male patients, who are normally pro-
tected from autoimmune diseases, need a stronger
genetic factor to manifest the disease (13, 14).
Neither of these two antigens, in the absence of
the other, was able to confer susceptibility to
Graves’ disease. This finding is in agreement with
the results of the recent study by Heward et al., (15)
who also found that all Graves’ patients who were
DR3 positive (albeit with a different polymorphism
-
DRB1*0304) were also DQA1*0501 positive.
Interestingly, similar results can be obtained after
Table 4 - HLA DQAI*0501 and DRB1*0301 in Graves’ disease according to family history and the presence of TED (FH=family his-
tory, TED=thyroid eye disease).
DQA1*0501 (+)
DQA1*0501 (-)
p
DRB1*0301 (+)
DRB1*0301 (-)
p
FH (+)
FH (-)
16 (85)
51 (84)
6 (15)
10 (16)
NS
13 (59)
42 (69)
9 (41)
19 (31)
NS
TED (+)
TED (-)
39 (85)
39 (75)
7 (15)
13 (25)
NS
34 (74)
31 (60)
12 (26)
21 (40)
NS
Percentage in parenthesis. (+) positive; (-) negative; NS: not significant.
90

HLA antigens and Graves’ disease
development of this disease, possibly through link-
age with an unknown locus.
10. Werner S.
Modification of the classification of the eye changes
of Graves’ disease.
In summary, this study demonstrates that the pres-
ence of neither of the DRB1*0304 and DQA1*0501
antigens are independent risk factors for Graves’ dis-
ease. However, when both are present they may be
of importance for the development of Graves’ dis-
ease, probably as part of a combined haplotype, in
some way predisposing to this autoimmune disease.
Am. J. Ophthalmol. 1977, 83: 725-727.
11. Ratanachaiyavong S., Lloyd L., Darke C., McGregor A.
MHC-extended haplotypes in families with Graves’
disease.
Hum. Immunol. 1993, 36: 99-111.
1
1
1
2. Cuddihy M., Dutton M., Bahn S.
A polymorphism in the extracellular domain in the
thyrotropin receptor is highly associated with au-
toimmune thyroid disease.
REFERENCES
Thyroid 1995, 5: 89-95.
3. Ahmed A., Young P., Penhale W.
The effects of female sex steroids on the develop-
ment of autoimmune thyroiditis in thymectomized
and irradiated rats.
1
. Tunbridge W.M., Evered D.E, Hall R., Appleton D.,
Brewis M., Clark F., Evans J., Young E., Bird T. , Smith P.
The spectrum of thyroid disease in a community: the
Wickham Survey.
Clin. Exp. Immunol. 1983, 54: 351-358.
Clin. Endocrinol. 1977, 7: 481-493.
4. Cavan A., Penny A., Jacobs H.
The HLA association with Graves’ disease is sex spe-
cific in Hong-Kong Chinese.
2
. Farid N.R., Stone E., Johnson G.
Graves’ disease and HLA: Clinical and epidemiolog-
ical associations.
Clin. Endocrinol. 1994, 40: 63-66.
Clin. Endocrinol. 1980, 13: 535-544.
3
. Ratanachaiyavong S., Gunn A., Bidwell A., Darke C.,
Hall R., McGregor A.
DQA2 U allele: a genetic marker for relapse of
Graves’ disease.
Clin. Endocrinol. 1991, 32: 241-251.
15. Heward J.M., Allahabadia A., Daykin J., Carr-Smith
J., Daly A., Armitage M., Dodson P.M., Sheppard
M.C., Barnett A.H., Franklyn J.A., Gough S.C.
Linkage disequilibrium between the human leuko-
cyte antigen class of the major histocompatibility
complex and Graves’ disease: Replication using a
population case control and family-based study.
J. Clin. Endocrinol. Metab. 1998, 83: 3394-3397.
4. Farid N.R.
Editorial: Understanding the genetics of autoimmune
thyroid disease - still an elusive goal.
J. Clin. Endocrinol. Metab. 1992, 74: 495A-495B.
. Yanagawa T., Mangklabruks A., Chang Y., Okamoto
Y., Fisfalen M., Curran P., DeGroot L.
Human Histocompatibility Leukocyte Antigen-
DQA1*0501 allele associated with genetic suscepti-
bility to Graves’ disease in a Caucasian population.
J. Clin. Endocrinol. Metab. 1993, 76: 1569-1574.
. Yanagawa T., Mangklabruks A., DeGroot L.
Strong association between HLA - DQA1*0501 and
Graves’ in a male population.
J. Clin. Endocrinol. Metab. 1994, 79: 227-229.
. Cuddihy R.M., Bahn R.S.
16. Tomer Y., Barbesino G., Greenberg D., Conception
E., Davies T.
5
Mapping the major susceptibility loci for familial
Graves’ and Hashimoto’s diseases: Evidence for ge-
netic heterogeneity and gene interactions.
J. Clin. Endocrinol. Metab. 1999, 84: 4656-4664.
17. Weetman A., So A., Warner C., Foroni P., Fells P.,
Shine B.
6
Immunogenetics of Graves’ ophthalmopathy.
Clin. Endocrinol. 1988, 28: 619-628.
1
8. Badenhoop K., Walfish P.G., Rau H., Fischer S.,
Nicolay A., Bogner U., Schleusener H., Usadel K.H.
Susceptibility and resistance alleles of human leuko-
cyte antigen (HLA) DQA1 and HLA DQB1 are shared
in endocrine autoimmune disease.
7
Lack of an independent association between the hu-
man leukocyte antigen allele DQA1*0501 and
Graves’ disease.
J. Clin. Endocrinol. Metab. 1996, 81: 847-849.
J. Clin. Endocrinol. Metab. 1995, 80: 211-217.
8
. Barlow A., Wheatcroft N., Watson P., Weetman A.
Association of HLA-DQA1*0501 with Graves’ disease
in English Caucasian men and women.
19. Philippou G., McGregor A.
The aetiology of Graves’ disease: what is the genet-
ic contribution?
Clin. Endocrinol. 1996, 44: 73-77.
Clin. Endocrinol. 1998, 48: 393-395.
9
. Reed A., Pachman L., Ober C.
20. Brix T.H, Christensen K., Holm N.V., Harvald B.,
Hegedus L.
Molecular genetic studies of major histocompatibility
complex genes in children with juvenile dermatom-
yositis: increased risk associated with HLA-DQA1*0501.
Hum. Immunol. 1991, 32: 235-240.
A population - based study of Graves’ disease in
Danish twins.
Clin. Endocrinol. 1998, 48: 397-400.
91