TSHR Gene Polymorphisms in the Enhancer Regions Are Most Strongly Associated with the Development of Graves’ Disease,Especially Intractable Disease,and of Hashimoto's Disease

Abstract

Background:

Graves' disease (GD) and Hashimoto's disease (HD) are autoimmune thyroid disorders distinguished by the presence or absence of antithyrotropin receptor (TSHR) antibodies (TRAb). TSHR gene polymorphisms determine the amount of TSHR expressed, which may in turn influence TRAb production. The FANTOM5 project identified six GD-associated single nucleotide polymorphisms (SNPs) within the enhancer regions of the TSHR and unknown genes. This study examined the association of 11 TSHR and unknown gene polymorphisms, five of which are located in TSHR enhancer regions, with the development and prognosis of GD and HD.

Methods:

SNPs of the TSHR and unknown genes were genotyped in 180 GD patients, including 62 patients with intractable GD and 48 patients with GD in remission; 151 HD patients, including 65 patients with severe HD and 40 patients with mild HD; and 111 healthy controls.

Results:

The rs4411444 GG genotype and G allele, the rs2300519 AA genotype, and the rs179247 AA genotype and A allele were more frequent in GD patients than they were in controls. These same genotypes and alleles, in addition to the rs2300519 A allele and rs4903961 GG genotype and G allele, were more frequent in patients with intractable GD than they were in controls and patients with GD in remission. Interestingly, the rs2300519 TT genotype and T allele, rs4903961 CC genotype and C allele, and rs179247 GG genotype, all of which are minor genotypes and alleles among the evaluated SNPs, were more frequent in HD patients than they were in controls, but there were no differences in the frequencies of these genotypes and alleles between patients with severe HD and mild HD. Among the evaluated SNPs, the rs4411444 GG genotype and the rs4903961 C allele in the enhancer regions of the TSHR gene were most strongly associated with the development of GD, especially intractable disease, and that of HD, respectively.

Conclusions:

Among the evaluated TSHR gene SNPs, the rs4411444 GG genotype and the rs4903961 C allele in the enhancer regions of the TSHR gene were most strongly associated with the development of GD, especially intractable disease, and that of HD, respectively.

Introduction

Autoimmune thyroid diseases (AITDs), including Hashimoto's disease (HD) and Graves' disease (GD), are frequent organ-specific autoimmune diseases with varying severity and intractability (1,2). Some HD patients develop hypothyroidism at a young age, while others remain in a euthyroid state, even into old age. When HD is in an asymptomatic state, it is difficult to predict disease prognosis (3,4). Some GD patients reach remission through medical treatment, but others do not, indicating that intractable GD is also difficult to predict. Several genetic factors are associated with AITDs, including genes encoding human leukocyte antigen (HLA), cytotoxic T lymphocyte-associated antigen (CTLA-4), protein tyrosine phosphatase non-receptor 22 (PTPN22), and thyroid-specific antigens, in addition to other chromosomal regions (5 –10). The association of AITD development and prognosis with candidate gene polymorphisms that are not specific for thyroid function, such as CTLA-4, CD40, Fc receptor-like 3 (FCRL3), interferon-gamma (IFN-γ), tumor necrosis factor alpha (TNF-α), and others, has been previously reported (7,11,12).

AITDs can be characterized by lymphocytic infiltration in the thyroid and the production of autoantibodies to thyroid-specific antigens, such as thyrotropin receptor (TSHR), thyroglobulin (Tg), and thyroid peroxidase (TPO) (13 –17). GD is diagnosed by the presence of an anti-TSHR antibody (TRAb), including thyroid-stimulating antibodies (TSAb), which stimulate thyroid epithelial cells to produce thyroid hormones and cause hyperthyroidism (18 –20). HD is diagnosed by the absence of TRAb and the presence of other thyroid autoantibodies, including anti-Tg antibodies (TgAb) and/or anti-TPO antibodies (TPOAb) (21). Sometimes HD transitions to GD, as determined by the appearance of TRAb, and GD remission with antithyroid drug therapy is associated with the disappearance of TRAb. Therefore, the propensity to develop GD is determined by poor immunological tolerance of the TSHR antigen and production of a TSHR autoantibody (22). A genome-wide association study identified strong associations of the TSHR and the major histocompatibility complex (MHC) class II variants with persistently positive TRAb and GD, and the authors suggested that shedding the A subunit of the TSHR may result in TSAb production (23,24).

Recently, the FANTOM5 project identified 43,011 enhancers within the majority of human cells and tissues (http://fantom.gsc.riken.jp/5/) and identified many disease-associated single nucleotide polymorphisms (SNPs), including six GD-associated SNPs within the enhancer regions of TSHR and unknown genes, which are shown in the FANTOM5 enhancer atlas (http://enhancer.binf.ku.dk) (25). Other SNPs in the intron 1 and intron 7 regions of the TSHR gene have been associated with susceptibility to GD (26 –29).

This study aimed to clarify the association of genetic differences in the TSHR with the development and prognosis of AITDs by genotyping previously described putative GD-associated SNPs: six within the enhancer regions of the TSHR, unknown genes, and five SNPs in the intron 1 and intron 7 regions of the TSHR. To analyze the function of these polymorphisms, the expression level of TSHR mRNA and serum levels of TRAb were examined.

Materials and Methods

Subjects

The presence/absence of each polymorphism was determined in 151 HD patients, 180 GD patients, and 111 healthy volunteers (control subjects). HD patients were positive for TgAb and/or antithyroid microsomal antibodies (McAb). GD patients were positive for TRAb at diagnosis and had a clinical history of thyrotoxicosis during HD. Healthy volunteers were euthyroid and negative for thyroid autoantibodies.

Among HD patients, 65 patients who developed moderate to severe hypothyroidism before 50 years of age and who were treated with thyroxine (severe HD) and 40 patients older than 50 years of age who were euthyroid and left untreated (mild HD) were genotyped. Among GD patients, 62 euthyroid patients who had been treated with methimazole for at least five years and who were still positive for TRAb (intractable GD) and 48 patients who had maintained a euthyroid state and who were negative for TRAb for more than two years without medication (GD in remission) were genotyped. Seventy GD patients and 46 HD patients who could not be categorized by the abovementioned parameters were defined as the “GD unclear group” and “HD unclear group,” respectively.

All patients with AITD and the control subjects were Japanese and were unrelated to each other. Genomic DNA was isolated from EDTA-treated peripheral blood mononuclear cells (PBMCs) using commercially available kits (QIAamp DNA Blood Mini Kit; Qiagen, Tokyo, Japan). Written informed consent was obtained from all subjects, and the study protocol was approved by the Ethics Committee of Osaka University, Osaka, Japan.

Genotyping GD-associated SNPs

The SNPs genotyped in this study are described in Table 1 and Figure 1. SNPs in enhancer regions were named SNP E1-6, and those in the intron 1 and intron 7 regions of the TSHR gene were named SNP1-5. The linkage of SNP E1-3 was analyzed using HaploView v4.2 software (Broad Institute, Cambridge, MA), and it was found that these SNPs were strongly linked (r ² = 1.0). Therefore, only SNP E2 was genotyped. A polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) method was used to genotype TSHR SNP E2, SNP E5, and SNP1. Briefly, the target sequence of each gene was amplified using PCR, and the PCR products were digested by restriction enzymes. A direct sequencing method was used for genotyping TSHR SNP E4, SNP2, SNP3, and SNP4. Briefly, the target sequence of each gene was amplified using PCR, and then the purified PCR products were sequenced using Big Dye thermosequenase (Bid Dye Terminator v1.1 cycle Sequencing Kit; Applied Biosystems, Carlsbad, CA). The dye-labeled purified PCR products were analyzed using a sequencer (Applied BioSystems 3100 Genetic Analyzer; Applied Biosystems). An amplification refractory mutation system-PCR (ARMS-PCR) method was used to genotype TSHR SNP5. Briefly, the PCR products containing only each target allele were amplified using four sequence-specific PCR primers. The sequences of PCR primer pairs and sequence primers, PCR conditions, and restriction enzymes used in this study are summarized in Table 2. TaqMan SNP genotyping assays (Applied Biosystems, Tokyo, Japan) were used to genotype SNP E6.

FIG. 1.

Locations of TSHR gene polymorphisms.

Table 1.

TSHR and Chromosome 6 Gene Polymorphisms Targeted in This Study

SNP name	Polymorphism	Chromosome	Gene	Region
SNP E1	rs2215981G/A
SNP E2	rs4411444 G/A
SNP E3	rs1035144T/C	14	TSHR	Intron 1	Enhancer
SNP E4	rs2300519A/T
SNP E5	rs4903961G/C
SNP E6	rs6941355G/A	6	Unknown		Enhancer
SNP1	rs179247 A/G
SNP2	rs541680762C/G
SNP3	rs137877366A/G	14	TSHR	Intron 1
SNP4	rs201206147A/G
SNP5	rs2268475T/C	14	TSHR	Intron 7

Table 2.

Primers, PCR Conditions, and Restriction Enzymes Used in This Study

Polymorphism	PCR primer and sequence primer	PCR conditions	Methods (reaction enzymes)
SNP E2	5′‐CTG ACA AAA TCT TGG CCA AA‐3′5′‐CAG ATG TGC AAA CAG AGA AG‐3′	95°C for 5 min(95°C for 30 sec, 55°C for 30 sec, 72°C for 30 sec) × 35 cycles72°C 5 min	PCR-RFLP (RsaI)
SNP E4	5′‐TAC TAG CCC CAT TTA TTT ACC AG‐3′	95°C for 3 min
SNP2	5′‐CTG TAA TCC CAG CAC TTT GG‐3′	(95°C for 30 sec, 62°C for 25 sec, 72°C for 17 sec) × 35 cycles72°C 5 min	Direct sequence
SNP3	5′‐CTC ACA TTC CAG TTG GGA AAC‐3′^*
SNP4
	5′‐TGA ATG CAT GAG CCG TAA CC‐3′	95°C for 3 min
SNP E5	5′‐TAG ACT CCC TGT ACA CTG AGA TG‐3′	(95°C for 30 sec, 58°C for 30 sec, 72°C for 20 sec) × 35 cycles72°C 5 min	PCR-RFLP (BtsCI)
SNP E6			TaqMan SNP genotyping assays
	5′‐CAT GAA GCT TTT GGC TTA TAT TTT‐3′	95°C for 5 min
SNP1	5′‐CTA GTT TCT TGG CTT AAA AAA ATA‐3′	(95°C for 30 sec, 45°C for 30 sec, 72°C for 30 sec) × 35 cycles72°C 5 min	PCR-RFLP (SspI‐HF)
	5′‐AGC AGT TAG AGA TGA AGA GG‐3′	95°C for 5 min
	5′‐TAA GAT GCT AAA GCA GCC AAT CT‐3′	(95°C for 30 sec, 62°C for 30 sec, 72°C for 30 sec) × 35cycle
SNP5	5′‐CTT CTC TGC AGT TGT TTT TTT GTC‐3′	72°C 5 min	ARMS-PCR
	5′‐AGA GTG GGC AAA ATC CAC CTT TAA‐3′

Asterisk indicates sequence primer.

PCR polymerase chain reaction; PCR-RFLP, PCR–restriction fragment length polymorphism; ARMS-PCR, amplification refractory mutation system-PCR.

mRNA extraction from PBMCs

The TSHR mRNA expression levels in PBMCs from 30 control subjects were analyzed to exclude the effect of medical treatment. Briefly, peripheral blood samples were collected in EDTA-treated tubes and PBMCs were isolated by density gradient centrifugation with Lymphoprep (Axis-Shield PoC As, Oslo, Norway) at 400 g for 30 min at room temperature. PBMCs were washed in sterile phosphate-buffered saline (PBS) and preserved in RNAlater solution (Ambion, Austin, TX) at –80°C until analysis. Total RNA was extracted from PBMCs with the mirVana^™ PARIS^™ KIT (Ambion), according to the manufacturer's protocol.

Quantitative real-time PCR

cDNAs were generated using the High Capacity cDNA Reverse Transcription Kit (Life Technologies, Tokyo, Japan). Real-time PCR was performed with TaqMan Gene Expression Assays (Life Technologies). The expression levels of TSHR mRNA were normalized to GAPDH. All reactions were performed in triplicate. The relative expression level of each mRNA was calculated using the ΔΔCt method and shown as the relative value to a particular sample.

Statistical analysis

A chi-square test was used to evaluate the significance of differences in the frequencies of genotypes and alleles among the groups. The Mann–Whitney U-test was used to analyze the difference between the serum levels of TRAb. Data were analyzed with JMP11 software (SAS Institute, Inc., Tokyo, Japan). Probability values of <0.05 were considered significant.

Results

Genotypes of TSHR gene polymorphisms

TSHR rs4411444 G/A polymorphism (SNP E2)

The frequencies of the GG genotype and G allele were significantly higher in GD patients than they were in controls (p < 0.0001 and p = 0.0008, respectively; Table 3). These frequencies were also significantly higher in GD patients than they were in HD patients (p < 0.0001 and p < 0.0001, respectively; Table 3). The GG genotype and G allele were significantly more frequent in intractable GD patients compared with controls (p < 0.0001 and p = 0.0003, respectively; Table 4) and were also significantly more frequent in intractable GD patients compared with GD patients in remission (p = 0.0227 and p = 0.0427, respectively; Table 4).

Table 3.

Genotype and Allele Frequencies of the TSHR Polymorphisms in Patients with AITD and in Control Subjects

		Control	All patients with GD	GD vs. control	All patients with HD	HD vs. control	GD vs. HD
TSHR	GG	40 (36.0%)	107 (59.4%)	0.0004	51 (33.8%)	NS	<0.0001
Intron 1	AG	58 (52.3%)	57 (31.7%)		74 (49.0%)
rs4411444	AA	13 (11.7%)	16 (8.9%)		26 (17.2%)
(SNP E2)
	GG + AG	98 (88.3%)	164 (91.1%)	NS	125 (82.8%)	NS	0.0234
	AA	13 (11.7%)	16 (8.9%)	NS	26 (17.2%)	NS	0.0234
	GG	40 (36.0%)	107 (59.4%)	<0.0001	51 (33.8%)	NS	<0.0001
	AG + AA	71 (64.0%)	73 (40.6%)	<0.0001	100 (66.2%)	NS	<0.0001
	G allele	138 (62.2%)	271 (75.3%)	0.0008	176 (58.3%)	NS	<0.0001
	A allele	84 (37.8%)	89 (24.7%)	0.0008	126 (41.7%)	NS	<0.0001
TSHR	AA	44 (39.6%)	96 (53.3%)	0.0363	49 (32.4%)	0.0418	0.0004
Intron 1	AT	57 (51.4%)	65 (36.1%)		72 (47.7%)
rs2300519	TT	10 (9.0%)	19 (10.6%)		30 (19.9%)
(SNP E4)
	AA + AT	101 (91.0%)	161 (89.4%)	NS	121 (80.1%)	0.0132	0.0176
	TT	10 (9.0%)	19 (10.6%)	NS	30 (19.9%)	0.0132	0.0176
	AA	44 (39.6%)	96 (53.3%)	0.0228	49 (32.4%)	NS	0.0001
	AT + TT	67 (60.4%)	84 (46.7%)	0.0228	102 (67.6%)	NS	0.0001
	A allele	145 (65.3%)	257 (71.4%)	NS	170 (56.3%)	0.0366	<0.0001
	T allele	77 (34.7%)	103 (28.6%)	NS	132 (43.7%)	0.0366	<0.0001
TSHR	GG	48 (43.2%)	97 (53.9%)	NS	45 (29.8%)	0.014	<0.0001
Intron 1	GC	52 (46.9%)	67 (37.2%)		74 (49.0%)
rs4903961	CC	11 (9.9%)	16 (8.9%)		32 (21.2%)
(SNP E5)
	GG + GC	100 (90.1%)	164 (91.1%)	NS	119 (78.8%)	0.0126	0.0015
	CC	11 (9.9%)	16 (8.9%)	NS	32 (21.2%)	0.0126	0.0015
	GG	48 (43.2%)	97 (53.9%)	NS	45 (29.8%)	0.0249	<0.0001
	GC + CC	63 (56.8%)	83 (46.1%)	NS	106 (70.2%)	0.0249	<0.0001
	G allele	148 (66.7%)	261 (72.5%)	NS	164 (54.3%)	0.0042	<0.0001
	C allele	74 (33.3%)	99 (27.5%)	NS	138 (45.7%)	0.0042	<0.0001
TSHR	AA	40 (36.0%)	95 (52.8%)	0.0196	48 (31.8%)	NS	<0.0001
Intron 1	AG	57 (51.4%)	67 (37.2%)		68 (45.0%)
rs179247	GG	14 (12.6%)	18 (10.0%)		35 (23.2%)
(SNP1)
	AA + AG	97 (87.4%)	162 (90.0%)	NS	116 (76.8%)	0.0274	0.0011
	GG	14 (12.6%)	18 (10.0%)	NS	35 (23.2%)	0.0274	0.0011
	AA	40 (36.0%)	95 (52.8%)	0.0052	48 (31.8%)	NS	0.0001
	AG + GG	71 (64.0%)	85 (47.2%)	0.0052	103 (68.2%)	NS	0.0001
	A allele	137 (61.7%)	257 (71.4%)	0.0158	164 (54.3%)	NS	<0.0001
	G allele	85 (38.3%)	103 (28.6%)	0.0158	138 (45.7%)	NS	<0.0001
TSHR	TT	37 (33.3%)	80 (44.4%)	0.0112	68 (45.0%)	NS	NS
Intron 7	TC	53 (47.8%)	86 (47.8%)		64 (42.4%)
rs2268475	CC	21 (18.9%)	14 (7.8%)		19 (12.6%)
(SNP5)
	TT + TC	90 (81.1%)	166 (92.2%)	0.0052	132 (87.4%)	NS	NS
	CC	21 (18.9%)	14 (7.8%)	0.0052	19 (12.6%)	NS	NS
	TT	37 (33.3%)	80 (44.4%)	NS	68 (45.0%)	NS	NS
	TC + CC	74 (66.7%)	100 (55.6%)	NS	83 (55.0%)	NS	NS
	T allele	127 (57.2%)	246 (68.3%)	0.0068	200 (66.2%)	0.0355	NS
	C allele	95 (42.8%)	114 (31.7%)	0.0068	102 (33.8%)	0.0355	NS

AITD, autoimmune thyroid disease; GD, Graves' disease; HD, Hashimoto's thyroiditis; NS, not significant.

Table 4.

Genotype and Allele Frequencies of the TSHR Intron1 and Intron7 Polymorphisms in Patients with GD and HD

				GD		HD
		Control	Intractable GD vs. control	Intractable	In remission	Intractable vs. in remission	Severe	Mild	Severe vs. mild
TSHR	GG	40 (36.0%)	<0.0001	43 (69.3%)	23 (47.9%)	NS	19 (29.2%)	15 (37.5%)	NS
Intron 1	AG	58 (52.3%)		14 (22.6%)	20 (41.7%)		35 (53.9%)	18 (45.0%)
rs4411444	AA	13 (11.7%)		5 (8.1%)	5 (10.4%)		11 (16.9%)	7 (17.5%)
(SNP E2)
	GG + AG	98 (88.3%)	NS	57 (91.9%)	43 (89.6%)	NS	54 (83.1%)	33 (82.5%)	NS
	AA	13 (11.7%)	NS	5 (8.1%)	5 (10.4%)	NS	11 (16.9%)	7 (17.5%)	NS
	GG	40 (36.0%)	<0.0001	43 (69.3%)	23 (47.9%)	0.0227	19 (29.2%)	15 (37.5%)	NS
	AG + AA	71 (64.0%)	<0.0001	19 (30.7%)	25 (52.1%)	0.0227	46 (70.8%)	25 (62.5%)	NS
	G allele	138 (62.2%)	0.0003	100 (80.7%)	66 (68.8%)	0.0427	73 (56.1%)	48 (60.0%)	NS
	A allele	84 (37.8%)	0.0003	24 (19.3%)	30 (31.2%)	0.0427	57 (43.9%)	32 (40.0%)	NS
TSHR	AA	44 (39.6%)	0.0099	39 (62.9%)	20 (41.7%)	NS	17 (26.1%)	14 (35.0%)	NS
Intron 1	AT	57 (51.4%)		18 (29.0%)	21 (43.7%)		36 (55.4%)	17 (42.5%)
rs2300519	TT	10 (9.0%)		5 (8.1%)	7 (14.6%)		12 (18.5%)	9 (22.5%)
(SNP E4)
	AA + AT	101 (91.0%)	NS	57 (91.9%)	41 (85.4%)	NS	53 (81.5%)	31 (77.5%)	NS
	TT	10 (9.0%)	NS	5 (8.1%)	7 (14.6%)	NS	12 (18.5%)	9 (22.5%)	NS
	AA	44 (39.6%)	0.0032	39 (62.9%)	20 (41.7%)	0.0263	17 (26.1%)	14 (35.0%)	NS
	AT+TT	67 (60.4%)	0.0032	23 (37.1%)	28 (58.3%)	0.0263	48 (73.9%)	26 (65.0%)	NS
	A allele	145 (65.3%)	0.0173	96 (77.4%)	61 (63.5%)	0.0243	70 (53.9%)	45 (56.2%)	NS
	T allele	77 (34.7%)	0.0173	28 (22.6%)	35 (36.5%)	0.0243	60 (46.1%)	35 (43.8%)	NS
TSHR	GG	48 (43.2%)	0.045	39 (62.9%)	19 (39.6%)	0.0489	16 (24.6%)	11 (27.5%)	NS
Intron 1	GC	52 (46.9%)		19 (30.6%)	23 (47.9%)		34 (52.3%)	21 (52.5%)
rs4903961	CC	11 (9.9%)		4 (6.4%)	6 (12.5%)		15 (23.1%)	8 (20.0%)
(SNP E5)
	GG + GC	100 (90.1%)	NS	58 (93.6%)	42 (87.5%)	NS	50 (76.9%)	32 (80.0%)	NS
	CC	11 (9.9%)	NS	4 (6.4%)	6 (12.5%)	NS	15 (23.1%)	8 (20.0%)	NS
	GG	48 (43.2%)	0.0127	39 (62.9%)	19 (39.6%)	0.0147	16 (24.6%)	11 (27.5%)	NS
	GC + CC	63 (56.8%)	0.0127	23 (37.1%)	29 (60.4%)	0.0147	49 (75.4%)	29 (72.5%)	NS
	G allele	148 (66.7%)	0.0214	97 (78.2%)	61 (63.5%)	0.0166	66 (50.8%)	43 (53.7%)	NS
	C allele	74 (33.3%)	0.0214	27 (21.8%)	35 (36.5%)	0.0166	64 (49.2%)	37 (46.3%)	NS
TSHR	AA	40 (36.0%)	0.0011	40 (64.5%)	19 (39.6%)	0.0157	18 (27.7%)	15 (37.5%)	NS
Intron 1	AG	57 (51.4%)		19 (30.7%)	21 (43.7%)		31 (47.7%)	15 (37.5%)
rs179247	GG	14 (12.6%)		3 (4.8%)	8 (16.7%)		16 (24.6%)	10 (25.0%)
(SNP1)
	AA + AG	97 (87.4%)	NS	59 (95.2%)	40 (83.3%)	0.0395	49 (75.4%)	30 (75.0%)	NS
	GG	14 (12.6%)	NS	3 (4.8%)	8 (16.7%)	0.0395	16 (24.6%)	10 (25.0%)	NS
	AA	40 (36.0%)	0.0003	40 (64.5%)	19 (39.6%)	0.009	18 (27.7%)	15 (37.5%)	NS
	AG + GG	71 (64.0%)	0.0003	22 (35.5%)	29 (60.4%)	0.009	47 (72.3%)	25 (62.5%)	NS
	A allele	137 (61.7%)	0.0027	99 (79.8%)	59 (61.5%)	0.0027	67 (51.5%)	45 (56.3%)	NS
	G allele	85 (38.3%)	0.0027	25 (20.2%)	37 (38.5%)	0.0027	63 (48.5%)	35 (43.7%)	NS
TSHR	TT	37 (33.3%)	NS	27 (43.5%)	17 (35.4%)	NS	29 (44.6%)	23 (57.5%)	NS
Intron 7	TC	53 (47.8%)		29 (46.8%)	26 (54.2%)		27 (41.5%)	13 (32.5%)
rs2268475	CC	21 (18.9%)		6 (9.7%)	5 (10.4%)		9 (13.9%)	4 (10.0%)
(SNP5)
	TT + TC	90 (81.1%)	NS	56 (90.3%)	43 (89.6%)	NS	56 (86.1%)	36 (90.0%)	NS
	CC	21 (18.9%)	NS	6 (9.7%)	5 (10.4%)	NS	9 (13.9%)	4 (10.0%)	NS
	TT	37 (33.3%)	NS	27 (43.5%)	17 (35.4%)	NS	29 (44.6%)	23 (57.5%)	NS
	TC + CC	74 (66.7%)	NS	35 (56.5%)	31 (64.6%)	NS	36 (55.4%)	17 (42.5%)	NS
	T allele	127 (57.2%)	NS	83 (66.9%)	60 (62.5%)	NS	85 (65.4%)	59 (73.7%)	NS
	C allele	95 (42.8%)	NS	41 (33.1%)	36 (37.5%)	NS	45 (34.6%)	21 (26.3%)	NS

TSHR rs2300519 A/ T polymorphism (SNP E4)

The frequency of the AA genotype was significantly higher in GD patients than it was in controls (p = 0.0228; Table 3). The proportions of the AA genotype and A allele were also significantly increased in GD patients compared with those in HD patients (p = 0.0001 and p < 0.0001, respectively; Table 3). On the other hand, the TT genotype and T allele were significantly more frequent in HD patients than they were in controls (p = 0.0132 and p = 0.0366, respectively; Table 3). The frequencies of the AA genotype and A allele were significantly higher in intractable GD patients than they were in controls (p = 0.0032 and 0.0173, respectively; Table 4) and were also significantly higher in intractable GD patients compared with GD patients in remission (p = 0.0263 and p = 0.0243, respectively; Table 4).

TSHR rs4903961 G/C polymorphism (SNP E5)

The frequencies of the CC genotype and C allele were significantly higher in HD patients than they were in controls (p = 0.0126 and p = 0.0042, respectively; Table 3). The GG genotype and G allele were significantly more frequent in GD patients than they were in HD patients (p < 0.0001 and p < 0.0001, respectively; Table 3). The proportions of the GG genotype and G allele were significantly increased in intractable GD patients compared with those in controls (p = 0.0127 and p = 0.0214, respectively; Table 4) and were also significantly increased in intractable GD patients compared with GD patients in remission (p = 0.0147 and p = 0.0166, respectively; Table 4).

TSHR rs179247 A/G polymorphism (SNP1)

The frequencies of the AA genotype and A allele were significantly higher in GD patients than they were in controls (p = 0.0052 and p = 0.0158, respectively; Table 3), and were also significantly higher in GD patients than they were in HD patients (p = 0.0001 and p < 0.0001, respectively; Table 3). On the other hand, the GG genotype was significantly more frequent in HD patients than it was in controls (p = 0.0274; Table 3). The frequencies of the AA genotype and A allele were significantly higher in intractable GD patients compared with those in controls (p = 0.0003 and p = 0.0027, respectively; Table 4), and were also significantly higher in intractable GD patients compared with GD patients in remission (p = 0.0090 and p = 0.0027, respectively; Table 4).

TSHR rs541680762 C/G polymorphism (SNP2), rs137877366 A/G polymorphism (SNP3), and rs201206147 A/G polymorphism (SNP4)

No minor allele was found in genotyped subjects (data not shown).

TSHR rs2268475 T/C polymorphism (SNP5)

The frequency of the TT + TC genotype and T allele were significantly higher in GD patients than they were in controls (p = 0.0052 and p = 0.0068, respectively; Table 3). Furthermore, the population of the T allele was significantly increased in HD patients compared with those in controls (p = 0.0355; Table 3). No associations were found between this polymorphism and the intractability or severity of AITDs.

Genotypes of chromosome 6 polymorphism

rs6941355G/A polymorphism (SNP E6)

No difference was found in the genotype and allele frequencies of this polymorphism between each group of patients and controls (Tables 5 and 6).

Table 5.

Genotype and Allele Frequencies of SNP E6 in Patients with AITD and in Control Subjects

		Control	All patients with GD	GD vs. control	All patients with HD	HD vs. control
Chromosome 6	AA	31 (27.9%)	35 (19.5%)	NS	32 (21.2%)	NS
rs6941355	AG	46 (41.5%)	89 (49.4%)		67 (44.4%)
(SNP E6)	GG	34 (30.6%)	56 (31.1%)		52 (34.4%)
	AA + AG	77 (69.4%)	124 (68.9%)	NS	99 (65.6%)	NS
	GG	34 (30.6%)	56 (31.1%)	NS	52 (34.4%)	NS
	AA	31 (27.9%)	35 (19.5%)	NS	32 (21.2%)	NS
	AG + GG	80 (72.1%)	145 (80.6%)	NS	119 (78.8%)	NS
	A allele	108 (48.7%)	159 (44.2%)	NS	131 (43.4%)	NS
	G allele	114 (51.3%)	201 (55.8%)	NS	171 (56.6%)	NS

Table 6.

Genotype and Allele Frequencies of SNP E6 in Patients with GD and HD

			GD			HD
		Control	Intractable	In remission	Intractable vs. in remission	Severe	Mild	Severe vs. mild
Chromosome 6	AA	31 (27.9%)	11 (17.7%)	12 (25.0%)	NS	12 (18.5%)	7 (17.5%)	NS
rs6941355	AG	46 (41.5%)	31 (50.0%)	19 (39.6%)		29 (44.6%)	13 (32.5%)
(SNP E6)	GG	34 (30.6%)	20 (32.3%)	17 (35.4%)		24 (36.9%)	20 (50.0%)
	AA + AG	77 (69.4%)	42 (67.7%)	31 (64.6%)	NS	41 (63.1%)	20 (50.0%)	NS
	GG	34 (30.6%)	20 (32.3%)	17 (35.4%)	NS	24 (36.9%)	20 (50.0%)	NS
	AA	31 (27.9%)	11 (17.7%)	12 (25.0%)	NS	12 (18.5%)	7 (17.5%)	NS
	AG + GG	80 (72.1%)	51 (82.3%)	36 (75.0%)	NS	53 (82.5%)	33 (82.5%)	NS
	A allele	108 (48.7%)	53 (42.7%)	43 (44.8%)	NS	53 (40.8%)	27 (33.8%)	NS
	G allele	114 (51.3%)	71 (57.3%)	53 (55.2%)	NS	77 (59.2%)	53 (66.2%)	NS

mRNA expression levels of TSHR in PBMCs

The expression of TSHR mRNA was not detected in PBMCs, and the association between the TSHR mRNA expression levels and the examined SNPs could not be analyzed.

Serum levels of TRAb and genotyped polymorphisms

No associations were found between each of the genotyped SNPs and the serum levels of TRAb at the onset of the disease in GD patients (data not shown).

Discussion

This is the first report that has comprehensively examined the association of previously described putative GD-associated SNPs, including those within the enhancer regions of the TSHR gene and unknown genes, with the development and prognosis of GD and HD. It has been clarified for the first time that SNPs within the enhancer regions of the TSHR gene are most strongly associated with the development of GD, especially intractable disease, and HD.

In the case of SNP1, it was found that the frequencies of the AA genotype and the A allele were significantly higher in GD patients than they were in controls (Tables 3 and 4), as shown in previous studies (26,30). It was also found that those were significantly higher in intractable GD patients than they were in patients with GD in remission, and that the frequency of the GG genotype was higher in HD patients than it was in controls (Tables 3 and 4). These results are consistent with those of a previous study (31). Expression of ST4, which is a TSHR mRNA splice variant, was higher in subjects with the AA genotype than it was in those with the GG genotype (26), and increased ST4 expression may result in a greater level of shedding the A subunit and thus increased production of TSAb (23). Therefore, SNP1 may be associated with GD development and prognosis based on differences in the TSAb levels caused by changes in ST4 expression.

In the case of SNP E1–5 in the TSHR gene enhancer region, it was found for the first time that the frequencies of both major genotypes and alleles of these SNPs were higher in GD patients, especially in intractable GD patients, than they were in controls, and that the frequencies of both minor genotypes and alleles were higher in HD patients than they were in controls (Tables 3 and 4). Because these SNPs are near SNP1, their linkage was analyzed using HaploView v4.2 software, and it was found that SNP E2 is strongly linked to SNP1 (r ² = 0.902). Among all of the genotyped SNPs in intron 1 of the TSHR (SNP1, E2, E4, and E5), SNP E2 was most strongly associated with the pathogenesis of GD. Thus, SNP E2 may directly affect the expression of the TSHR, ST4, and TSAb by altering transcription factor binding. GD is mainly an autoimmune thyroid disease in which most patients are positive for TgAb and TPOAb, in conjunction with the appearance of TRAb, including TSAb. GD remission can be achieved with antithyroid drug therapy in association with the disappearance of TRAb. Therefore, major genotypes and alleles of SNP E1–5, which were more frequent in GD patients, especially with intractable GD, than they are in controls, may induce higher TSHR gene expression and higher production of TSHR antigen for presentation to TSHR-specific T helper cells and B cells. Thus, TSAb would be produced more efficiently or abundantly, making it easier to break immunological tolerance of the TSHR antigen. By contrast, minor genotypes and alleles of these SNPs, which were more frequent in HD patients than they were in controls, may induce lower TSHR gene expression and lower production of the TSHR antigen for a weaker presentation to TSHR-specific cells and a lower impairment of immunological tolerance. The pathological condition of HD is usually characterized by an absent break of immunological tolerance to the TSHR antigen, and therefore the minor genotypes and alleles of these SNPs were considered to be more frequent in HD.

To test this hypothesis, transcription factor binding to TSHR enhancer regions, including SNP E1-6, was simulated using P-match software (www.gene-regulation.com/cgi-bin/pub/programs/pmatch/bin/p-match.cgi). However, no transcription factors were found that bind to these regions with this in silico approach. Thus, some unknown transcription factors and/or mechanisms may regulate the transcription of the TSHR gene.

It has been suggested that TSHR mRNA is expressed not only on thyroid epithelial cells but also in PBMCs (32). Because normal thyroid tissue could not be obtained, the TSHR mRNA levels in PBMCs were examined from controls to clarify the effect of polymorphisms in TSHR expression. However, TSHR mRNA could not be detected in PBMCs. This is consistent with another study, which was unable to find relevant TSHR mRNA expression in PBMCs (26). These findings indicate that TSHR mRNA expression is limited in PBMCs, and for this reason, differences in TSHR expression among genotypes of TSHR SNPs could not be clarified.

Next, the association between SNPs in this study and the serum levels of TRAb at the onset of GD were analyzed. However, no associations were found between each of these SNPs and TRAb expression levels. In a previous study, the TSAb levels were increased, but the TRAb levels were not changed in experimental hyperthyroid mice (23). Therefore, it is supposed that some association may be found between the examined SNPs and TSAb levels.

Contrary to expectation, in the case of SNP5 in the intron 7 region, the major T allele was more prevalent not only in GD patients but also in HD patients compared with controls. Although SNP5 in GD has already been reported (29), this is the first study indicating the association between this SNP and susceptibility for HD. In other words, this SNP was associated with the development of AITD, including both GD and HD, through some unknown mechanism. This SNP is also located in a different gene, LOC101928462, whose function has not yet been characterized. SNP5 may affect the expression of this gene, and one could speculate that it could be involved in the development of AITD.

In conclusion, the major genotype and allele of each SNP (SNP1 and SNP E1-5) in the TSHR gene is associated with the development of GD, especially intractable GD. Among the SNPs, the rs441144 GG genotype (SNP E2) shows the most significant association. The minor genotype and allele of these polymorphisms is associated with the development of HD. Among them, the rs4903961 C allele (SNP E5) shows the most significant association. SNP5 in the TSHR gene is also associated with the development of GD and HD.

Footnotes

Acknowledgments

This work was supported by JSPS KAKENHI grant number 26293128.

Author Disclosure Statement

The authors declare that they have no conflict of interest.

References

Menconi

, Oppenheim

, Tomer

. 2008. Graves's disease. In: Shoenfeld

, Cervera

, Gershwin

(eds) Diagnostic Criteria in Autoimmune Diseases. Humana Press, Totowa, NJ, pp 231–235.

Weetman

. 2000. Chronic autoimmune thyroiditis. In: Braverman

, Utiger

(eds) Werner and Ingbar's The Thyroid: A Fundamental and Clinical Text. Eighth edition. Lippincott Williams & Wilkins, Philadelphia, PA, pp 721–732.

Amino

, Hagen

, Yamada

, Refetoff

. 1976. Measurement of circulating thyroid microsomal antibodies by the tanned red cell haemagglutination technique: its usefulness in the diagnosis of autoimmune thyroid diseases. Clin Endocrinol (Oxf), 5:115–125.

Yoshida

, Amino

, Yagawa

, Uemura

, Satoh

, Miyai

, Kumahara

. 1978. Association of serum antithyroid antibodies with lymphocytic infiltration of the thyroid gland: studies of seventy autopsied cases. J Clin Endocrinol Metab, 46:859–862.

Akamizu

, Hiratani

, Ikegami

, Rich

, Bowden

. 2003. Association study of autoimmune thyroid disease at 5q23–q33 in Japanese patients. J Hum Genet, 48:236–242.

Ban

, Greenberg

, Concepcion

, Skrabanek

, Villanueva

, Tomer

. 2003. Amino acid substitutions in the thyroglobulin gene are associated with susceptibility to human and murine autoimmune thyroid disease. Proc Natl Acad Sci U S A, 100:15119–15124.

Inoue

, Watanabe

, Yamada

, Takemura

, Hayashi

, Yamakawa

, Akahane

, Shimizuishi

, Hidaka

, Iwatani

. 2012. Associations between autoimmune thyroid disease prognosis and functional polymorphisms of susceptibility genes, CTLA4, PTPN22, CD40, FCRL3, and ZFAT, previously revealed in genome-wide association studies. J Clin Immunol, 32:1243–1252.

Skorka

, Bednarczuk

, Bar-Andziak

, Nauman

, Ploski

. 2005. Lymphoid tyrosine phosphatase (PTPN22/LYP) variant and Graves' disease in a Polish population: association and gene dose-dependent correlation with age of onset. Clin Endocrinol (Oxf), 62:679–682.

Ueda

, Howson

, Esposito

, Heward

, Snook

, Chamberlain

, Rainbow

, Hunter

, Smith

, Di Genova

, Herr

, Dahlman

, Payne

, Smyth

, Lowe

, Twells

, Howlett

, Healy

, Nutland

, Rance

, Everett

, Smink

, Lam

, Cordell

, Walker

, Bordin

, Hulme

, Motzo

, Cucca

, Hess

, Metzker

, Rogers

, Gregory

, Allahabadia

, Nithiyananthan

, Tuomilehto-Wolf

, Tuomilehto

, Bingley

, Gillespie

, Undlien

, Ronningen

, Guja

, Ionescu-Tirgoviste

, Savage

, Maxwell

, Carson

, Patterson

, Franklyn

, Clayton

, Peterson

, Wicker

, Todd

, Gough

. 2003. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature, 423:506–511.

10.

Vaidya

, Kendall-Taylor

, Pearce

. 2002. The genetics of autoimmune thyroid disease. J Clin Endocrinol Metab, 87:5385–5397.

11.

Inoue

, Watanabe

, Nanba

, Wada

, Akamizu

, Iwatani

. 2009. Involvement of functional polymorphisms in the TNFA gene in the pathogenesis of autoimmune thyroid diseases and production of anti-thyrotropin receptor antibody. Clin Exp Immunol, 156:199–204.

12.

Ito

, Watanabe

, Okuda

, Watanabe

, Iwatani

. 2006. Association between the severity of Hashimoto's disease and the functional +874A/T polymorphism in the interferon-gamma gene. Endocr J, 53:473–478.

13.

Baloch

, Livolsi VA 2005 Pathology. In: Braverman

, Utiger

(eds) Werner and Ingbar's The Thyroid: A Fundamental and Clinical Text. Eighth edition. Lippincott Williams & Wilkins, Philadelphia, PA, pp 422–449.

14.

Brand

, Gough

. 2010. Genetics of thyroid autoimmunity and the role of the TSHR. Mol Cell Endocrinol, 322:135–143.

15.

Davies

. 1996. The pathogenesis of Graves' disease. In: Davies

, Braverman

, Utiger

(eds) The Thyroid: A Fundamental and Clinical Textbook. Seventh edition. Lippincott Press, Philadelphia, PA, pp 525–536.

16.

Davies

. 2005. Pathogenesis of Graves' disease. In: Braverman

, Utiger

(eds) Werner and Ingbar's The Thyroid: A Fundamental and Clinical Text. Eighth edition. Lippincott Williams & Wilkins, Philadelphia, PA, pp 457–473.

17.

Iwatani

, Watanabe

. 1999. Normal mechanisms for self-tolerance. In: Volpe

(ed) Autoimmune Endocrinopathies. Humana Press, Totowa, NJ, pp 1–30.

18.

Rapoport

, McLachlan

. 2007. The thyrotropin receptor in Graves' disease. Thyroid, 17:911–922.

19.

Smith

, Sanders

, Furmaniak

. 2007. TSH receptor antibodies. Thyroid, 17:923–938.

20.

Weetman

. 2000. Graves' disease. N Engl J Med, 343:1236–1248.

21.

Levine

. 1983. Current concepts of thyroiditis. Arch Intern Med, 143:1952–1956.

22.

Colobran

, Armengol Mdel

, Faner

, Gartner

, Tykocinski

, Lucas

, Ruiz

, Juan

, Kyewski

, Pujol-Borrell

. 2011. Association of an SNP with intrathymic transcription of TSHR and Graves' disease: a role for defective thymic tolerance. Hum Mol Genet, 20:3415–3423.

23.

Chen

, Pichurin

, Nagayama

, Latrofa

, Rapoport

, McLachlan

. 2003. The thyrotropin receptor autoantigen in Graves disease is the culprit as well as the victim. J Clin Invest, 111:1897–1904.

24.

Chu

, Pan

, Zhao

, Liang

, Gao

, Zhang

, Yuan

, Li

, Xue

, Shen

, Liu

, Xie

, Yang

, Wang

, Shi

, Sun

, Du

, Zuo

, Shi

, Liu

, Guo

, Zhan

, Gu

, Zhang

, Sun

, Wang

, Song

, Zou

, Sun

, Guo

, Cao

, Ma

, Han

, Li

, Jiang

, Huang

, Liang

, Liu

, Chen

, Su

, Peng

, Zhao

, Ning

, Chen

, Huang

, Song

. 2011. A genome-wide association study identifies two new risk loci for Graves' disease. Nat Genet, 43:897–901.

25.

Andersson

, Gebhard

, Miguel-Escalada

, Hoof

, Bornholdt

, Boyd

, Chen

, Zhao

, Schmidl

, Suzuki

, Ntini

, Arner

, Valen

, Li

, Schwarzfischer

, Glatz

, Raithel

, Lilje

, Rapin

, Bagger

, Jorgensen

, Andersen

, Bertin

, Rackham

, Burroughs

, Baillie

, Ishizu

, Shimizu

, Furuhata

, Maeda

, Negishi

, Mungall

, Meehan

, Lassmann

, Itoh

, Kawaji

, Kondo

, Kawai

, Lennartsson

, Daub

, Heutink

, Hume

, Jensen

, Suzuki

, Hayashizaki

, Muller

, Forrest

, Carninci

, Rehli

, Sandelin

. 2014. An atlas of active enhancers across human cell types and tissues. Nature, 507:455–461.

26.

Brand

, Barrett

, Simmonds

, Newby

, McCabe

, Bruce

, Kysela

, Carr-Smith

, Brix

, Hunt

, Wiersinga

, Hegedus

, Connell

, Wass

, Franklyn

, Weetman

, Heward

, Gough

. 2009. Association of the thyroid stimulating hormone receptor gene (TSHR) with Graves' disease. Hum Mol Genet, 18:1704–1713.

27.

Davies

, Yin

, Latif

. 2010. The genetics of the thyroid stimulating hormone receptor: history and relevance. Thyroid, 20:727–736.

28.

Pujol-Borrell

, Gimenez-Barcons

, Marin-Sanchez

, Colobran

. 2015. Genetics of Graves' disease: special focus on the role of TSHR gene. Horm Metab Res, 47:753–766.

29.

Hiratani

, Bowden

, Ikegami

, Shirasawa

, Shimizu

, Iwatani

, Akamizu

. 2005. Multiple SNPs in intron 7 of thyrotropin receptor are associated with Graves' disease. J Clin Endocrinol Metab, 90:2898–2903.

30.

Ploski

, Brand

, Jurecka-Lubieniecka

, Franaszczyk

, Kula

, Krajewski

, Karamat

, Simmonds

, Franklyn

, Gough

, Jarzab

, Bednarczuk

. 2010. Thyroid stimulating hormone receptor (TSHR) intron 1 variants are major risk factors for Graves' disease in three European Caucasian cohorts. PLoS One, 5:e15512.

31.

Inoue

, Watanabe

, Katsumata

, Hidaka

, Iwatani

. 2013. Different genotypes of a functional polymorphism of the TSHR gene are associated with the development and severity of Graves' and Hashimoto's diseases. Tissue Antigens, 82:288–290.

32.

Stefan

, Wei

, Lombardi

, Li

, Concepcion

, Inabnet

3rd , Owen

, Zhang

, Tomer

. 2014. Genetic-epigenetic dysregulation of thymic TSH receptor gene expression triggers thyroid autoimmunity. Proc Natl Acad Sci U S A, 111:12562–12567.