Gene polymorphisms of VEGF and VEGFR2 are associated with the severity of Hashimoto’s disease and the intractability of Graves’ disease, respectively

Abstract

Vascular endothelial growth factor (VEGF) is one of main regulators of angiogenesis that functions by binding to its receptors, including VEGF receptor (VEGFR) 2. There are few data available regarding the association between VEGF and VEGFR polymorphisms and the susceptibility to and prognosis of autoimmune thyroid diseases (AITDs). To elucidate this association, we genotyped four functional VEGF and two VEGFR2 polymorphisms and measured serum VEGF levels. In the four functional VEGF polymorphisms, the frequencies of the I carrier and I allele of VEGF –2549 I/D, which has lower activity, were higher in patients with severe HD than in those with mild HD. In the two functional VEGFR2 polymorphisms, the frequency of the rs2071559 CC genotype, which has higher activity, was higher in patients with intractable GD than in controls, and the proportion of GD patients with larger goiters was higher in those with the CC genotype. Moreover, the frequency of the rs1870377 TT genotype with higher activity was higher in patients with intractable GD than in those with GD in remission. Combinations of VEGF and VEGFR2 polymorphisms with stronger interactions were associated with the intractability of GD. Serum VEGF levels were higher in HD and AITD patients than those in controls. In conclusion, VEGF polymorphisms with lower activity were associated with the severity of HD, while VEGFR2 polymorphisms and the combinations of VEGF and VEGFR2 polymorphisms, which have stronger interactions, were associated with the intractability of GD. VEGF and VEGFR2 polymorphisms were associated with HD severity and GD intractability, respectively.

AUTOIMMUNE THYROID DISEASES (AITDs), such as Graves’ disease (GD) and Hashimoto’s disease (HD), are organ-specific autoimmune diseases [1, 2]. Their prevalences are reported to be relatively high at approximately 10% in various regions around the world including Japan [3]. Autoimmune diseases, including AITDs, are believed to develop by a combination of genetic and environmental factors [4, 5]. In twin studies, it was reported that the influence of genetic factors on AITD development is approximately 75% and that of environmental factors is approximately 25% [4]. The HLA-DR, CTLA-4, CD40, and TSHR genes were major susceptibility genes for AITD [6], and some environmental factors such as iodine and medications were reported [6].

The intractability of GD and the severity of HD are different among individuals. Some patients with GD achieve remission after treatment with anti-thyroid drugs, whereas others do not. Further, some patients with HD develop hypothyroidism in early life, while some maintain a euthyroid state. However, the intractability of GD and the severity of HD are very difficult to predict at diagnosis. We have already reported that genetically higher expressions of Th1 cytokines and/or lower expressions of Th2 cytokines were associated with the severity of HD [7-9]. On the other hand, genetically higher expressions of Th17 cytokines were associated with intractability of GD [9-11].

In the thyroid gland of patients with GD, since both hypervascularity and increased blood flow are observed, angiogenesis is thought to be related to the development of goiter [12-15]. In addition, it is well known that it is difficult to achieve remission through medical treatment in GD patients with large goiters [16-22], and goiter size is thus considered important to predict the intractability of GD. In this study, we therefore focused on angiogenesis-related genes as one factor affecting goiter size.

Vascular endothelial growth factor A (VEGFA) is a key regulator of angiogenesis and is commonly called VEGF [23, 24]. VEGF induces the proliferation, migration, and survival of endothelial cells and also enhances vascular permeability [23, 24]. VEGF is expressed in many tissues and cells, including endothelial cells [25], macrophages, and some types of tumor cells [26]. In various cancer types, serum VEGF levels are higher relative to controls and show a positive correlation with cancer progression and patient survival [27, 28]. Moreover, serum VEGF levels are high and correlated with disease activity in various autoimmune diseases such as rheumatoid arthritis (RA), psoriasis vulgaris, and systemic lupus erythematosus (SLE) [29-32]. In the case of GD, increased serum VEGF levels can be found in treated euthyroid GD patients with larger goiters as well as untreated hyperthyroid GD patients [15]. In HD, serum VEGF levels are also higher in untreated hypothyroid HD patients relative to controls [15], whereas there is no difference in serum VEGF levels between treated euthyroid patients and controls [15]. On the other hand, another report showed that plasma VEGF levels are decreased in treated euthyroid HD patients compared to controls [33]. Taken together, these studies indicate that VEGF may be associated with the pathology or clinical condition of GD and HD.

Vascular endothelial growth factor receptor (VEGFR) mainly consists of VEGFR1 (fms-like tyrosine kinase-1 (Flt-1)), VEGFR2 (kinase insert domain receptor (KDR)/fetal liver kinase-1 (Flk-1)) and VEGFR3 (Flt-4) [34]. VEGF binds to VEGFR1 and VEGFR2 and activates the signaling pathways for angiogenesis [23, 24]. VEGFR2 plays a major role in angiogenesis [35, 36], regulates tumor angiogenesis [37, 38], and its overexpression is found in many types of tumors [39, 40].

Some gene polymorphisms are functional and can affect the expression or structure of a protein [41, 42]. In the VEGF and VEGFR2 genes, we focused on four functional VEGF polymorphisms (2549 I/D, rs1570360 G/A, rs2010963 C/G, and rs3025039 C/T) and two functional VEGFR2 polymorphisms (rs2071559 C/T and rs1870377 A/T), which may affect angiogenesis-related diseases [43-51].

The D allele of VEGF –2549 I/D (18 bp insertion/deletion) polymorphism is associated with susceptibility to Kawasaki disease (KD) [43] and has 1.95-fold higher transcription activity than the I allele [44]. The G allele of VEGF rs1570360 G/A polymorphism is associated with the risk of renal allograft rejection and higher VEGF production from peripheral blood mononuclear cells (PBMC) [45]. The C allele of VEGF rs2010963 C/G polymorphism is associated with susceptibility to GD [46]. This allele also enhances serum VEGF levels [47] and VEGF mRNA expression [48]. The T allele of VEGF rs3025039 C/T polymorphism is associated with low risk of breast cancer [49] and KD [43], and this allele carrier shows lower levels of plasma VEGF [49].

In the case of the VEGFR2 gene, the CC genotype of 2071559 C/T polymorphism shows decreased serum VEGFR2 levels [50]. The rs1870377 A/T polymorphism changes the sequence of the amino acids from His (T allele) to Gln (A allele), and the T allele shows higher binding affinity of VEGF to VEGFR2 [50]. This allele is also associated with susceptibility to astrocytomas [51].

In this study, we examined these polymorphisms in the VEGF and VEGFR2 genes and compared them between disease (AITD, GD, and HD) and control and between the different prognostic groups (intractable GD and GD in remission; severe HD and mild HD) to elucidate the association with the development and prognosis of AITD, respectively.

Material and Methods

Thyroid function and autoantibodies

Serum concentrations of free T4 (FT4), free T3 (FT3), and thyrotropin (TSH) were measured using ECLIA kit (Roche Diagnostics Ltd., Tokyo, Japan). Normal ranges of serum FT4, FT3, and TSH are 0.9–1.7 ng/dL, 2.3–4.3 pg/mL, and 0.5–5.0 μIU/mL, respectively. TgAb and McAb were measured using a particle agglutination kit (Fujirebio Inc., Tokyo, Japan). A reciprocal titer >1:100 was considered positive. Serum TRAb at onset was measured by radioreceptor assay using a commercial kit (Cosmic Corp., Tokyo, Japan) as part of routine studies. Serum TRAb at the time of sampling was determined by ECLIA kit (third-generation) (Roche Diagnostics Ltd., Tokyo, Japan) using preserved samples. Normal value of TRAb was less than 10% in the radioreceptor assay and 2.0 IU/L in ECLIA. Goiter size in this study was defined as the longitudinal length of the thyroid gland as measured using a vernier caliper.

Subjects for genotyping

We genotyped each polymorphism in 462 AITD patients and 117 healthy controls. Among the 214 GD patients who had a clinical history of thyrotoxicosis and were positive for anti-thyrotropin receptor antibody (TRAb), 97 patients who had been treated with methimazole for at least 5 years were still positive for TRAb (intractable GD), 58 patients with GD who had maintained a euthyroid state were negative for TRAb for more than 2 years without medication (GD in remission), and 59 patients could not be classified into either intractable GD or GD in remission groups at the time of analysis.

Among 248 HD patients who were positive for anti-thyroglobulin antibody (TgAb) and/or anti-microsomal antibody (McAb), we genotyped 118 patients who developed moderate to severe hypothyroidism before 50 years of age and were treated with thyroxine (severe HD), 81 untreated euthyroid patients who were over 50 years of age (mild HD), and 49 patients who could not be classified into either severe HD or mild HD groups at the time of analysis. Healthy controls were euthyroid and negative for any thyroid autoantibodies. Table 1 shows the clinical characteristics of all subjects in this study. Goiter size was larger in patients with mild HD than that in patients with severe HD (p = 0.0295). FT4 levels were higher in patients with severe HD than those in patients with mid HD (p = 0.0001). The titers of TgAb were significantly higher in patients with severe HD than those in patients with mild HD (p = 0.0004), as per our previous report [52].

Table 1 Clinical characteristics of the subjects at the time of sampling in this study

	Controls	GD		HD
		Past clinical history of thyrotoxicosis with elevated TRAb		Diffuse goiter and positive TgAb and/or McAb
		Intractable	In remission	Severe	Mild
n (female/male)	117 (80/37)	97 (83/14)	58 (51/7)	118 (97/21)	81 (67/14)
The age of onset (years)		34.4 ± 14.0	33.6 ± 14.6	38.5 ± 9.9	58.8 ± 12.2
Age of the time of sampling (years)	33.7 ± 14.6	49.0 ± 15.0	44.8 ± 15.3	49.1 ± 14.1	62.7 ± 10.0
Goiter size (cm)	ND	4.9 ± 1.5	4.2 ± 0.6	4.1 ± 1.0*	4.6 ± 1.3
FT4 (ng/dL)	1.2 ± 0.3	1.2 ± 0.3	1.2 ± 0.2	1.3 ± 0.3**	1.2 ± 0.2
TSH (μIU/mL)	1.6 ± 1.0	1.7 ± 1.4	1.8 ± 1.2	4.1 ± 10.3	2.8 ± 1.9
TRAb (IU/L)	<2.0	13.6 ± 60.1	<2.0	<2.0	<2.0
TgAb (2n × 100)a	Negative	2.8 ± 3.0	2.9 ± 3.2	6.6 ± 3.2***	1.3 ± 2.4
McAb (2n × 100)a	Negative	4.4 ± 2.7	5.5 ± 2.0	5.4 ± 3.0	3.5 ± 2.8
Current treatment	None	MMI or PTU	None	L-thyroxine	None
Current dose of anti-thyroid drug (mg/day)b (range)	None	14.5 ± 20.8 (2.5–150)	None	None	None
Current dose of L-thyroxine (μg/day) (range)	None	None	None	80.0 ± 37.2 (12.5–250)	None

Data are expressed as mean ± standard deviation

ND, not determined; PTU, propylthiouracil; MMI, methimazole

^a When the titer of TgAb or McAb was 25,600, it was expressed as 2⁸ × 100, ^b Doses were expressed as the comparable dose of MMI (50 mg of PTU was conveted to 5 mg of MMI)

Analyzed by Mann-Whitney U test, * p = 0.0295 (vs. mild HD), ** p = 0.0001 (vs. mild HD), *** p = 0.0009 (vs. mild HD)

Genomic DNA was isolated from ethylenediamine tetraacetic acid (EDTA)-treated whole blood cells with a commercially available kit (QIAamp^® DNA Blood Mini Kit, QIAGEN, Tokyo, Japan) according to manufacture’s protocol. Briefly, 200 μL blood sample and 20 μL protease were added into a 1.5 mL microcentrifuge tube. These samples were then incubated at 56°C for 10 min and centrifuged. We pipetted these mixtures onto the QIAamp Mini spin column and centrifuged them at 6,000 x g (8,000 rpm) for 1 min. Finally, we eluted the DNA in a new 1.5 mL microcentrifuge tube by centrifuging them at 6,000 x g (8,000 rpm) for 1 min. Written informed consent was obtained from all patients and control subjects, and the study protocol was approved by the Ethics Committee of Osaka University (564).

Genotyping of polymorphisms

Genotyped polymorphisms in this study are shown in Table 2. VEGF –2549 I/D polymorphism was analyzed using the polymerase chain reaction (PCR) method. The other polymorphisms (VEGF rs1570360 A/G, rs2010963 C/G and rs3025039, and VEGFR2 rs2071559 C/T and rs1870377 A/T) were analyzed by the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method. The target sequences of each gene were amplified using PCR, and the PCR products were digested by the addition of restriction enzymes. The primers, the PCR conditions, restriction enzymes used in this study, and electrophoresis image obtained are shown in Table 3 and Fig. 1.

Table 2 Functions of VEGF and VEGFR2 gene polymorphisms targeted in this study

Gene	Polymorphism	Region	Effect to expression
VEGF	–2549 I/D	promoter	D allele > I allele (transcriptional activity)
	rs1570360	promoter	GG genotype > AA genotype (production levels)
	rs2010963	5’-UTR	CC genotype > GG genotype (serum levels and expression levels)
	rs3025039	3’-UTR	CC genotype > T carrier (plasma levels)
VEGFR2	rs2071559	promoter	CC genotype < TT genotype (serum soluble VEGFR2 levels)
	rs1870377	exon 11	T allele > A allele (binding efficiency)

Table 3 Primers, PCR conditions, and methods (restriction enzymes) used in this study

Gene	Polymorphism	PCR primer	PCR conditions	Methods (Restriction enzymes)	PCR prodcuts or Restriction fragments
VEGF	–2549 I/D	5’-GGCCTTAGGACACCATACCG-3’ 5’-AGGGAACAAAGTTGGGGCTC-3’	95°C for 5 min (95°C for 30 sec, 60.2°C for 30 sec, 72°C for 20 sec) × 29 cycle 72°C for 5 min	PCR	II genotype: 296 bp ID genotype: 296 bp + 278 bp DD genotype: 278 bp
	rs1570360	5’-TCCTGCTCCCTCCTCGCCAATG-3’ 5’-GGCGGGGACAGGCGAGCATC-3’	95°C for 5 min (95°C for 30 sec, 61.1°C for 30 sec, 72°C for 30 sec) × 35 cycle 72°C for 5 min	PCR-RFLP (MnlІ)	AA genotype: 206 bp AG genotype: 206 bp + 187 bp + 19 bp GG genotype: 187 bp + 19 bp
	rs2010963	5’-GGAGATTGCTCTACTTCCCCAA-3’ 5’-GTCTGTCTGTCTGTCCGTCA-3’	95°C for 5 min (95°C for 30 sec, 58°C for 30 sec, 72°C for 30 sec) × 30 cycle 72°C for 5 min	PCR-RFLP (BsmFI)	CC genotype: 256 bp GC genotype: 256 bp + 145 bp + 111 bp GG genotype: 145 bp + 111 bp
	rs3025039	5’-GATACAGAAACCACGCTGCC-3’ 5’-CGGGCTCGGTGATTTAGCAG-3’	95°C for 5 min (95°C for 30 sec, 60.6°C for 30 sec, 72°C for 30 sec) × 35 cycle 72°C for 5 min	PCR-RFLP (NlaIII)	CC genotype: 219 bp CT genotype: 219 bp + 160 bp + 59 bp TT genotype: 160 bp + 59 bp
VEGFR2	rs2071559	5’-CAAACTTTCACTAGGGCTCTTCGT-3’ 5’-AGCCACAAGGGAGAAGCGGATA-3’	95°C for 5 min (95°C for 30 sec, 61°C for 30 sec, 72°C for 30 sec) × 30 cycle 72°C for 5 min	PCR-RFLP (BsmI)	TT genotype: 290 bp CT genotype: 290 bp + 174 bp + 116 bp CC genotype: 174 bp + 116 bp
	rs1870377	5’-GCCTCACATATTATTGTACCATCC-3’ 5’-CCTCCTGTATCCTGAATGAATCT-3’	95°C for 5 min (95°C for 30 sec, 60.5°C for 30 sec, 72°C for 30 sec) × 29 cycle 72°C for 5 min	PCR-RFLP (AluI)	AA genotype: 404 bp AT genotype: 404 bp + 213 bp + 191 bp TT genotype: 191 bp + 213 bp

Fig. 1

Representative electrophoresis image to genotype VEGF and VEGFR polymorphisms in this study.

(A) VEGF –2549 I/D, (B) VEGF rs1570360, (C) VEGF rs2010963, (D) VEGF rs3025039, (E) VEGFR rs2071559, and (F) VEGFR rs1870377 polymorphisms. Detailed information on each genotype is shown in Table 3.

Serum VEGF levels

We also selected age-matched subjects for measurement of serum VEGF levels. Serum VEGF levels were examined in 32 GD patients (16 intractable GD patients and 16 GD patients in remission), 32 HD patients (16 severe HD patients and 16 mild HD patients), and 14 healthy controls. We examined the difference in serum VEGF levels in each genotype of VEGF polymorphism by using healthy controls, untreated GD patients in remission, and untreated mild HD patients to avoid the effect of medical treatment. Serum VEGF levels were measured using commercially available enzyme-linked immunosorbent assay (ELISA) kits (Quantikine Human VEGF Immunoassay, R&D Systems, Inc., Minneapolis, MN, USA). Table 4 shows the clinical characteristics of all subjects used in this analysis. Goiter size was larger in intractable GD patients than that in GD patients in remission (p = 0.0004), as well as in mild HD patients relative to severe HD patients (p = 0.0052). FT4 levels were higher in patients with severe HD than those in patients with mid HD (p = 0.0012).

Table 4 Clinical characteristics of the subjects obtained by analyzing the serum VEGF levels

	Controls	GD		HD
		Past clinical history of thyrotoxicosis with elevated TRAb		Diffuse goiter and positive TgAb and/or McAb
		Intractable	In remission	Severe	Mild
n (female/male)	14 (11/3)	16 (14/2)	16 (14/2)	16 (14/2)	16 (14/2)
The age of onset (years)		34.4 ± 15.2	35.4 ± 13.7	38.3 ± 9.9	54.9 ± 6.1
Age at the time of sampling (years)	51.1 ± 16.7	54.1 ± 11.9	54.5 ± 10.7	59.1 ± 14.2	60.9 ± 6.1
Goiter size (cm)	ND	5.0 ± 1.0*	3.8 ± 0.4	3.6 ± 0.5**	4.2 ± 0.7
Free T4 (ng/dL)	1.2 ± 0.3	1.1 ± 0.1	1.2 ± 0.2	1.3 ± 0.1***	1.1 ± 0.2
TSH (μU/mL)	1.6 ± 1.0	1.9 ± 1.1	1.6 ± 0.6	2.0 ± 0.8	2.2 ± 1.1
TRAb (IU/L)	<2.0	13.9 ± 22.4	<2.0	<2.0	<2.0
Current treatment	None	MMI or PTU	None	L-thyroxine	None
Current dose of anti-thyroid drug (mg/day)a (range)	None	14.1 ± 11.8 (1.25–30)	None	None	None
Current dose of L-thyroxine (μg/day) (range)	None	None	None	84.3 ± 31.1 (37.5–150)	None

Data are expressed as mean ± standard deviation

ND, not determined; PTU, propylthiouracil; MMI, methimazole

^a Doses were expressed as the comparable dose of MMI (50 mg of PTU was conveted to 5 mg of MMI)

Analyzed by Mann-Whitney U test, * p = 0.0004 (vs. GD in remission), ** p = 0.0052 (vs. mild HD), ***p = 0.0026 (vs. mild HD)

Statistical analysis

We used the χ² test and Fisher’s exact test to evaluate the significance of the differences in the frequencies of genotypes and alleles and goiter size among the groups. The Mann-Whitney U test was used to analyze the differences between the titers of McAb and TgAb and the levels of TRAb among each genotype. Student’s t test, the Mann-Whitney U test, and the χ² test were used to analyze differences in serum VEGFA levels. The data were analyzed with JMP 13 software (SAS Institute Inc., Tokyo, Japan). Probability values of <0.05 were considered significant. The sample number in this study was required to detect approximately 20% difference in VEGF –2549DD genotype frequency with 80% power (α = 0.05).

Results

VEGF polymorphisms

We could not find any significant differences in the genotype and allele frequencies of each examined polymorphism in VEGF genes between each patient group and control subjects (Table 5). However, the frequencies of the I carrier (II + ID genotypes) and the I allele of the functional VEGF –2549 I/D polymorphism were higher in patients with severe HD than in those with mild HD (p = 0.0430 and 0.0338, respectively) (Table 6). In the VEGF rs3025039 polymorphism, the proportion of patients with larger goiters (>5.6 cm) was higher in GD patients with the CC genotype than in T carriers (p = 0.0239) (Fig. 2).

Table 5 Genotype and allele frequencies of the VEGF polymorphisms in patients with AITD and in control subjects

		Control	All patients with AITD		All patients with GD		All patients with HD
VEGF –2549 I/D	II	11 (11.22%)	33 (7.91%)	NSb	15 (7.94%)	NSb	18 (7.9%)	NSb
	ID	35 (35.72%)	157 (37.65%)		73 (38.62%)		84 (36.84%)
	DD	52 (53.06%)	227 (54.44%)		101 (53.44%)		126 (55.26%)
	II + ID	46 (46.94%)	190 (45.56%)	NSb	88 (46.56%)	NSb	102 (44.74%)	NSb
	DD	52 (53.06%)	227 (54.44%)		101 (53.44%)		126 (55.26%)
	II	11 (11.22%)	33 (7.91%)	NSb	15 (7.94%)	NSb	18 (7.9%)	NSb
	ID + DD	87 (88.78%)	384 (92.09%)		174 (92.06%)		210 (92.1%)
	I allele	57 (29.08%)	223 (26.74%)	NSb	103 (27.25%)	NSb	120 (26.32%)	NSb
	D allele	139 (70.92%)	611 (73.26%)		275 (72.75%)		336 (73.68%)
VEGF rs1570360	AA	3 (3%)	2 (0.49%)	NSa,b	1 (0.54%)	NSa,b	1 (0.45%)	NSa,b
	AG	24 (24%)	88 (21.57%)		43 (23.37%)		45 (20.09%)
	GG	73 (73%)	318 (77.94%)		140 (76.09%)		178 (79.46%)
	AA + AG	27 (27%)	90 (22.06%)	NSb	44 (23.91%)	NSb	46 (20.54%)	NSb
	GG	73 (73%)	318 (77.94%)		140 (76.09%)		178 (79.46%)
	AA	3 (3%)	2 (0.49%)	NSa,b	1 (0.54%)	NSa,b	1 (0.45%)	NSa,b
	AG + GG	97 (97%)	406 (99.51%)		183 (99.46%)		223 (99.55%)
	A allele	30 (15%)	92 (11.27%)	NSb	45 (12.23%)	NSb	47 (10.49%)	NSb
	G allele	170 (85%)	724 (88.73%)		323 (87.77%)		401 (89.51%)
VEGF rs2010963	CC	18 (16.36%)	95 (22.78%)	NSb	43 (23.12%)	NSb	52 (22.51%)	NSb
	GC	56 (50.91%)	202 (48.44%)		95 (51.07%)		107 (46.32%)
	GG	36 (32.73%)	120 (28.78%)		48 (25.81%)		72 (31.17%)
	CC + GC	74 (67.27%)	297 (71.22%)	NSb	138 (74.19%)	NSb	159 (68.83%)	NSb
	GG	36 (32.73%)	120 (28.78%)		48 (25.81%)		72 (31.17%)
	CC	18 (16.36%)	95 (22.78%)	NSb	43 (23.12%)	NSb	52 (22.51%)	NSb
	GC + GG	92 (83.64%)	322 (77.22%)		143 (76.88%)		179 (77.49%)
	C allele	92 (41.82%)	392 (47%)	NSb	181 (48.66%)	NSb	211 (45.67%)	NSb
	G allele	128 (58.18%)	442 (53%)		191 (51.34%)		251 (54.33%)
VEGF rs3025039	CC	67 (65.69%)	275 (66.1%)	NSb	121 (64.02%)	NSb	154 (67.84%)	NSb
	CT	29 (28.43%)	123 (29.57%)		58 (30.69%)		65 (28.64%)
	TT	6 (5.88%)	18 (4.33%)		10 (5.29%)		8 (3.52%)
	CC + CT	96 (94.12%)	398 (95.67%)	NSa,b	179 (94.71%)	NSb	219 (96.48%)	NSa,b
	TT	6 (5.88%)	18 (4.33%)		10 (5.29%)		8 (3.52%)
	CC	67 (65.69%)	275 (66.1%)	NSb	121 (64.02%)	NSb	154 (67.84%)	NSb
	CT + TT	35 (34.31%)	141 (33.9%)		68 (35.98%)		73 (32.16%)
	C allele	163 (79.9%)	673 (80.89%)	NSb	300 (79.37%)	NSb	373 (82.16%)	NSb
	T allele	41 (20.1%)	159 (19.11%)		78 (20.63%)		81 (17.84%)

Analyzed by χ² test except ^a analyzed by Fisher’s exact test. ^b vs. control, NS: not significant

Table 6 Genotype and allele frequencies of VEGF polymorphisms genotyped in this study in patients with HD and GD

		Control	GD			HD
			Intractable	In remission		Severe	Mild
VEGF –2549 I/D	II	11 (11.22%)	9 (9.89%)	3 (5.26%)	NSb	11 (9.82%)	4 (5.26%)	NSc
	ID	35 (35.72%)	38 (41.76%)	21 (36.84%)		44 (39.29%)	22 (28.95%)
	DD	52 (53.06%)	44 (48.35%)	33 (57.9%)		57 (50.89%)	50 (65.79%)
	II + ID	46 (46.94%)	47 (51.65%)	24 (42.1%)	NSb	55 (49.11%)	26 (34.21%)	0.0430c
	DD	52 (53.06%)	44 (48.35%)	33 (57.9%)		57 (50.89%)	50 (65.79%)
	II	11 (11.22%)	9 (9.89%)	3 (5.26%)	NSa,b	11 (9.82%)	4 (5.26%)	NSc
	ID + DD	87 (88.78%)	82 (90.11%)	54 (94.74%)		101 (90.18%)	72 (94.74%)
	I allele	57 (29.08%)	56 (30.77%)	27 (23.68%)	NSb	66 (29.46%)	30 (19.74%)	0.0338c
	D allele	139 (70.92%)	126 (69.23%)	87 (76.32%)		158 (70.54%)	122 (80.26%)
VEGF rs1570360	AA	3 (3%)	1 (1.12%)	0 (0%)	NSa,b	1 (0.9%)	0 (0%)	NSa,c
	AG	24 (24%)	23 (25.84%)	11 (20%)		26 (23.42%)	11 (15.07%)
	GG	73 (73%)	65 (73.04%)	44 (80%)		84 (75.68%)	62 (84.93%)
	AA + AG	27 (27%)	24 (26.96%)	11 (20%)	NSb	27 (24.32%)	11 (15.07%)	NSc
	GG	73 (73%)	65 (73.04%)	44 (80%)		84 (75.68%)	62 (84.93%)
	AA	3 (3%)	1 (1.12%)	0 (0%)	NSa,b	1 (0.9%)	0 (0%)	NSa,c
	AG + GG	97 (97%)	88 (98.88%)	55 (100%)		110 (99.1%)	73 (100%)
	A allele	30 (15%)	25 (14.04%)	11 (10%)	NSb	28 (12.61%)	11 (7.53%)	NSc
	G allele	170 (85%)	153 (85.96%)	99 (90%)		194 (87.39%)	135 (92.47%)
VEGF rs2010963	CC	18 (16.36%)	19 (21.11%)	12 (21.43%)	NSb	22 (19.3%)	24 (31.17%)	NSc
	GC	56 (50.91%)	43 (47.78%)	31 (55.36%)		54 (47.37%)	29 (37.66%)
	GG	36 (32.73%)	28 (31.11%)	13 (23.21%)		38 (33.33%)	24 (31.17%)
	CC + GC	74 (67.27%)	62 (68.89%)	43 (76.79%)	NSb	76 (66.67%)	53 (68.83%)	NSc
	GG	36 (32.73%)	28 (31.11%)	13 (23.21%)		38 (33.33%)	24 (31.17%)
	CC	18 (16.36%)	19 (21.11%)	12 (21.43%)	NSb	22 (19.3%)	24 (31.17%)	NSc
	GC + GG	92 (83.64%)	71 (78.89%)	44 (78.57%)		92 (80.7%)	53 (68.83%)
	C allele	92 (41.82%)	81 (45%)	55 (49.11%)	NSb	98 (42.98%)	77 (50%)	NSc
	G allele	128 (58.18%)	99 (55%)	57 (50.89%)		130 (57.02%)	77 (50%)
VEGF rs3025039	CC	67 (65.69%)	55 (61.11%)	35 (60.34%)	NSa,b	76 (66.67%)	52 (71.23%)	NSa,c
	CT	29 (28.43%)	28 (31.11%)	22 (37.93%)		35 (30.7%)	19 (26.03%)
	TT	6 (5.88%)	7 (7.78%)	1 (1.73%)		3 (2.63%)	2 (2.74%)
	CC + CT	96 (94.12%)	83 (92.22%)	57 (98.27%)	NSa,b	111 (97.37%)	71 (97.26%)	NSa,c
	TT	6 (5.88%)	7 (7.78%)	1 (1.73%)		3 (2.63%)	2 (2.74%)
	CC	67 (65.69%)	55 (61.11%)	35 (60.34%)	NSb	76 (66.67%)	52 (71.23%)	NSc
	CT + TT	35 (34.31%)	35 (38.89%)	23 (39.66%)		38 (33.33%)	21 (28.77%)
	C allele	163 (79.9%)	138 (76.67%)	92 (79.31%)	NSb	187 (82.02%)	123 (84.25%)	NSc
	T allele	41 (20.1%)	42 (23.33%)	24 (20.69%)		41 (17.98%)	23 (15.75%)

Analyzed by χ² test except ^a analyzed by Fisher’s exact test. ^b intractable GD vs. GD in remission, ^c severe HD vs. mild HD, NS: not significant

Fig. 2

Goiter size of GD patients in each genotype of VEGF rs3025039 polymorphism.

Size of a goiter that could not be measured by palpation was defined as 3.5 cm. Difference in the proportion of patients with larger goiter size (>5.6 cm) was analyzed by χ² test.

VEGFR2 polymorphisms

We could not find any significant differences in the genotype and allele frequencies of each examined polymorphism in the VEGFR2 gene between each patient group and control subjects (Table 7). However, the frequency of the CC genotype of VEGFR2 rs2071559 was higher in patients with intractable GD than that in control subjects (p = 0.0362) (Table 8). The proportion of GD patients with larger goiters (>5.6 cm) was higher in subjects with CC genotype than those with T carriers (Fig. 3). The frequency of the TT genotype of the VEGFR2 rs1870377 polymorphism was higher in intractable GD than GD in remission (p = 0.0438) (Table 8).

Table 7 Genotype and allele frequencies of the VEGFR2 polymorphisms in patients with AITD and in control subjects

		Control	All patients with AITD		All patients with GD		All patients with HD
VEGFR2 rs2071559	CC	3 (2.89%)	28 (6.68%)	NSa	13 (6.99%)	NSa	15 (6.44%)	NSa
	CT	35 (33.65%)	149 (35.56%)		65 (34.95%)		84 (36.05%)
	TT	66 (63.46%)	242 (57.76%)		108 (58.06%)		134 (57.51%)
	CC + CT	38 (36.54%)	177 (42.24%)	NSa	78 (41.94%)	NSa	99 (42.49%)	NSa
	TT	66 (63.46%)	242 (57.76%)		108 (58.06%)		134 (57.51%)
	CC	3 (2.89%)	28 (6.68%)	NSa	13 (6.99%)	NSa	15 (6.44%)	NSa
	CT + TT	101 (97.11%)	391 (93.32%)		173 (93.01%)		218 (93.56%)
	C allele	41 (19.71%)	205 (24.46%)	NSa	91 (24.46%)	NSa	114 (24.46%)	NSa
	T allele	167 (80.29%)	633 (75.54%)		281 (75.54%)		352 (75.54%)
VEGFR2 rs1870377	AA	12 (11.88%)	52 (12.59%)	NSa	24 (12.7%)	NSa	28 (12.5%)	NSa
	AT	54 (53.47%)	220 (53.27%)		107 (56.61%)		113 (50.45%)
	TT	35 (34.65%)	141 (34.14%)		58 (30.69%)		83 (37.05%)
	AA + AT	66 (65.35%)	272 (65.86%)	NSa	131 (69.31%)	NSa	141 (62.95%)	NSa
	TT	35 (34.65%)	141 (34.14%)		58 (30.69%)		83 (37.05%)
	AA	12 (11.88%)	52 (12.59%)	NSa	24 (12.7%)	NSa	28 (12.5%)	NSa
	AT + TT	89 (88.12%)	361 (87.41%)		165 (87.3%)		196 (87.5%)
	A allele	78 (38.61%)	324 (39.23%)	NSa	155 (41.01%)	NSa	169 (37.72%)	NSa
	T allele	124 (61.39%)	502 (60.77%)		223 (58.99%)		279 (62.28%)

Analyzed by χ² test, ^a vs. control, NS: not significant

Table 8 Genotype and allele frequencies of VEGFR2 polymorphisms genotyped in this study in patients with HD and GD

		Control	GD			HD
			Intractable	In remission		Severe	Mild
VEGFR2 rs2071559	CC	3 (2.89%)	9 (10.23%)	1 (1.75%)	NSb	8 (7.02%)	6 (8.11%)	NSc
	CT	35 (33.65%)	29 (32.95%)	18 (31.58%)		41 (35.96%)	24 (32.43%)
	TT	66 (63.46%)	50 (56.82%)	38 (66.67%)		65 (57.02%)	44 (59.46%)
	CC + CT	38 (36.54%)	38 (43.18%)	19 (33.33%)	NSb	49 (42.98%)	30 (40.54%)	NSc
	TT	66 (63.46%)	50 (56.82%)	38 (66.67%)		65 (57.02%)	44 (59.46%)
	CC	3 (2.89%)	9 (10.23%)d	1 (1.75%)	NSa,b	8 (7.02%)	6 (8.11%)	NSc
	CT + TT	101 (97.11%)	79 (89.77%)	56 (98.25%)		106 (92.98%)	68 (91.89%)
	C allele	41 (19.71%)	47 (26.7%)	20 (17.54%)	NSb	57 (25%)	36 (24.32%)	NSc
	T allele	167 (80.29%)	129 (73.3%)	94 (82.46%)		171 (75%)	112 (75.68%)
VEGFR2 rs1870377	AA	12 (11.88%)	10 (10.75%)	8 (14.03%)	NSb	11 (9.73%)	10 (14.08%)	NSc
	AT	54 (53.47%)	47 (50.54%)	36 (63.16%)		62 (54.87%)	33 (46.48%)
	TT	35 (34.65%)	36 (38.71%)	13 (22.81%)		40 (35.4%)	28 (39.44%)
	AA + AT	66 (65.35%)	57 (61.29%)	44 (77.19%)	0.0438b	73 (64.6%)	43 (60.56%)	NSc
	TT	35 (34.65%)	36 (38.71%)	13 (22.81%)		40 (35.4%)	28 (39.44%)
	AA	12 (11.88%)	10 (10.75%)	8 (14.03%)	NSb	11 (9.73%)	10 (14.08%)	NSc
	AT + TT	89 (88.12%)	83 (89.25%)	49 (85.97%)		102 (90.27%)	61 (85.92%)
	A allele	78 (38.61%)	67 (36.02%)	52 (45.61%)	NSb	84 (37.17%)	53 (37.32%)	NSc
	T allele	124 (61.39%)	119 (63.98%)	62 (54.39%)		142 (62.83%)	89 (62.68%)

Analyzed by χ² test except ^a analyzed by Fisher’s exact test. ^b Intractable GD vs. GD in remission, ^c severe HD vs. mild HD, ^d p = 0.0362 (vs. control), NS: not significant

Fig. 3

Goiter size of GD patients in each genotype of VEGFR2 rs2071559 polymorphism.

Size of a goiter that could not be measured by palpation was defined as 3.5 cm. Difference in the proportion of patients with larger goiter size (>5.6 cm) was analyzed by χ² test.

Combined analysis of VEGF and VEGFR2 polymorphisms

We performed combined analysis of the polymorphisms in the VEGF and VEGFR2 genes between intractable GD patients and patients in remission. The frequencies of the GG/CC genotypes of VEGF rs1570360/VEGFR2 rs2071559 polymorphisms, C carrier/CC genotype of VEGF rs2010963/VEGFR2 rs2071559 polymorphisms, C carrier/CC genotype of VEGF rs3025039/VEGFR2 rs2071559 polymorphisms, D carrier/TT genotype of VEGF –2549 I/D/VEGFR2 rs1870377 polymorphisms, and G carrier/TT genotype of VEGF rs1570360/VEGFR2 rs1870377 polymorphisms were higher in patients with intractable GD than that in patients with GD in remission (p = 0.0427, 0.0222, 0.0491, 0.0326, and 0.0342, respectively) (Table 9).

Table 9 Combined analysis to study the association between polymorphisms in VEGF and VEGFR2 among GD

Genotype of gene polymorphisms		GD
		Intractable	In remission
VEGF rs1570360/VEGFR2 rs2071559	GG/CC others	7 (8.14%) 79 (91.86%)	0 (0%) 55 (100%)	0.0427a
VEGF rs2010963/VEGFR2 rs2071559	CC + CG/CC others	8 (9.3%) 78 (90.7%)	0 (0%) 56 (100%)	0.0222a
VEGF rs3025039/VEGFR2 rs2071559	CC + CT/CC others	9 (10.71%) 75 (89.29%)	1 (1.75%) 56 (98.25%)	0.0491a
VEGF –2549 I/D/VEGFR2 rs1870377	DD + ID/TT others	33 (37.93%) 54 (62.07%)	12 (21.05%) 45 (78.95%)	0.0326
VEGF rs1570360/VEGFR2 rs1870377	GG + GA/TT others	36 (40.91%) 52 (59.09%)	13 (23.64%) 42 (76.36%)	0.0342

Analyzed by χ² test except ^a analyzed by Fisher’s exact test.

Serum VEGF levels

Serum VEGF levels were higher in HD and AITD patients than those in control subjects (p = 0.0154, 0.0208, respectively) (Fig. 4) and tended to be higher, but not significantly so, in GD patients than those in control subjects (p = 0.0507) (Fig. 4). However, there were no significant differences in serum VEGF levels between intractable GD patients and GD patients in remission or between severe HD patients and mild HD patients (data not shown).

Fig. 4

Serum VEGF levels in AITD patients and controls.

Analyzed by ^a Student’s t test and ^b Mann-Whitney U test.

To clarify the effect of VEGF polymorphism on serum VEGF levels, we analyzed the association between serum VEGF levels and genotypes of VEGF polymorphism. The levels of serum VEGF were higher in the DD genotype of the VEGF –2549 I/D polymorphism than in I carriers (DI + II genotypes) (Fig. 5A), in the GG genotype of the VEGF rs1570360 polymorphism than in the GA genotype (Fig. 5B), in the CC genotype of VEGF rs2010963 than that in G carriers (GG + GC genotypes) (Fig. 5C), and in the CC genotype of VEGF rs3025039 than that in T carriers (CT + TT genotypes) (Fig. 5D), but not significantly so.

Fig. 5

Serum VEGF levels in each genotype of VEGF polymorphism.

We examined the difference in serum VEGF levels in each genotype of VEGF polymorphism by using healthy controls, untreated GD patients in remission, and untreated mild HD patients to avoid the effect of medical treatment. We analyzed the differences of serum VEGF levels (A and B) and the proportion of subjects with high serum VEGF levels (C and D). Analyzed by Student’s t test except ^a analyzed by χ² test.

However, we found that the proportion of subjects with higher serum VEGF levels (>325 pg/mL) was higher in the CC genotype of VEGF rs2010963 than that in G carriers (GG + GC genotypes) (p = 0.0430) (Fig. 5C) and that the proportion of subjects with higher serum VEGF levels (>300 pg/mL) was higher in the CC genotype of VEGF rs3025039 than that in T carriers (CT + TT genotypes) (p = 0.0244) (Fig. 5D).

Discussion

In this study, we expected that enhanced angiogenesis in thyroid glands of GD patients would be associated with larger goiters and intractability of GD. In HD, we also expected that enhanced angiogenesis would cause severe infiltration of immune cells, such as cytotoxic T lymphocytes (CTL) and helper T (Th) 1 cells, to the thyroid and lead to the destruction of the thyroid glands.

In VEGF –2549 I/D polymorphism, we hypothesized that the frequency of the D allele would be increased in severe HD patients relative to mild HD patients because this allele has higher transcriptional activity [44]. Contrary to our expectation, we found that the frequencies of the I carrier and I allele, which are associated with lower VEGF expression [44], were higher in severe HD patients relative to mild HD patients (Table 6). Previous reports have shown that the AA genotype of rs699947, which is linked to the II genotype of –2549 I/D [44, 53], is associated with susceptibility to and early onset of type 1 diabetes, suggesting that VEGF may play a protective role in diabetes [54]. Moreover, other reports show that VEGF enhances survival of brown adipocytes [55], as well as endothelial cells [23, 24]. VEGF also has neuroprotective effects in mice with experimental autoimmune encephalomyelitis [56]. On the other hand, serum level of IFN-γ was significantly lower in a group of colorectal carcinoma patients with positive staining of VEGF [57]. We have already reported that genetically higher expressions of IFN-γ and high Th1/Th2 ratio were associated with the severity of HD [7, 9]. Therefore, we suggested that VEGF might protect thyrocytes from CTL and Th1 cells and suppress IFN-γ production in HD.

In the case of GD, a previous study has shown that the C allele of the VEGF rs2010963 polymorphism was associated with susceptibility to GD [46]. In the present study, the frequency of this allele also tended to be higher in GD patients relative to controls, but not significantly so (Table 5). We also found that the proportion of GD patients showing higher serum VEGF levels was increased in individuals with the rs3025039 CC genotype (Fig. 5D), was also shown in a previous study [49]. Furthermore, we found that the proportion of GD patients with larger goiters was increased in individuals with the same genotype (Fig. 2). Therefore, in subjects with the CC genotype, higher serum VEGF levels might increase angiogenesis in the thyroid glands of GD patients and result in enlargement of goiters.

In the case of VEGFR2, we also hypothesized that the frequency of the rs2071559 CC genotype would be higher in intractable GD patients than in GD patients in remission because individuals with this genotype showed lower serum VEGFR2 levels [50]. In accordance with this hypothesis, we found that the frequency of the CC genotype was higher in patients with intractable GD than control subjects (Table 8), and that the proportion of GD patients with larger goiters (>5.6 cm) was higher in individuals with the CC genotype than T carriers (Fig. 3). It is well known that VEGFR2 is expressed on endothelial cells [58] and can be detected in its soluble form in sera as soluble VEGFR2 (sVEGFR2). Although sVEGFR1 is known as an inhibitor of angiogenesis [59-61], recent studies have indicated that sVEGFR2 may trap VEGF in the peripheral blood and play a role as a natural cell regulator [62, 63]. Therefore, we suggested that individuals with the CC genotype of VEGFR2 rs2071559 may promote angiogenesis in thyroid glands through insufficient suppression of VEGF and induce enlargement of goiters to enhance the intractability of GD. However, there is the possibility that individuals with the CC genotype of this polymorphism also show lower full-length VEGFR2 expression. Further study is needed on this aspect.

In the VEGFR2 rs1870377 polymorphism, it has been reported that the T allele relates to higher binding affinity of VEGF to VEGFR2 [50]. As expected, we found that the frequency of the TT genotype was higher in patients with intractable GD than those in remission (Table 8). Therefore, we suggested that individuals with the TT genotype of this polymorphism may exhibit enhanced angiogenesis associated with the intractability of GD.

Combined analysis showed that each genotype related to higher VEGF expression [44, 45, 47-49], risk of GD [46], lower serum VEGFR2 levels [50], and higher VEGF-VEGFR2 binding affinity [50] were more frequent in intractable GD patients than in GD patients in remission (Table 9). These findings suggest that the genetic combination of the interaction of VEGF-VEGFR2 ligation, especially combinations of gene polymorphisms with higher VEGF expression and with higher VEGF-VEGFR2 binding affinity, may be more important for the intractability of GD than each individual VEGF and VEGFR2 polymorphism. Serum level of IL-6, which is a Th17-inducing cytokines, was significantly higher in patients with ulcerative colitis showing positive staining of VEGF [57]. We have previously reported that high proportion of Th17 cells may be associated with GD intractability. Therefore, VEGF-VEGFR2 ligation may be associated with GD intractability through Th17 differentiation.

On the other hand, serum VEGF levels were higher in AITD and HD patients and tended to be higher in GD patients than those in controls (Fig. 4). However, we also found that the frequencies of the I carrier and I allele of VEGF –2549 I/D polymorphism, which are associated with lower VEGF expression [44], were higher in severe HD patients relative to mild HD patients (Table 6). Since VEGF has a protective role against autoimmunity [56, 58], our findings suggest that HD patients who are genetically disposed to lower rates of VEGF production might be susceptible to more severe HD than those with higher rates of production. Therefore, increase of serum VEGF in AITD patients might be a secondary phenomenon against autoimmune reaction. Furthermore, serum VEGF levels were higher in the CC genotype of VEGF rs2010963 than in G carriers (GG + GC genotypes) (Fig. 5C) and were higher in the CC genotype of VEGF rs3025039 than in T carriers (CT + TT genotypes) (Fig. 5D). Although they were not significant, serum VEGF levels were also higher in the DD genotype of the VEGF –2549 I/D polymorphism than in I carriers (DI + II genotypes) (Fig. 5A) and were higher in the GG genotype of the VEGF rs1570360 polymorphism than in the GA genotype (Fig. 5B). These results were generally consistent with previous reports that showed significant differences in VEGF measurements among each genotype of the four VEGF polymorphisms [44, 45, 47-49].

There are some limitations in study. We could not compare genotypes/alleles between groups with ophthalmopathy. This is because the prevalence of GD ophthalmopathy is too low in the Japanese population to analyze. We also could not compare genotypes/alleles relative to goiter size and family history because of sample numbers. In this study, goiter size was defined as the longitudinal length of the thyroid gland as measured using a vernier caliper. Therefore, we could not show calculated thyroid volumes because of the difficulty of measuring this for all GD patients. Furthermore, despite the possibility that young controls could become patients in the future, we were unable to compare VEGF and VEGFR2 polymorphisms between controls and GD or HD patients using age-matched controls. However, there was no significant difference between controls and patients in this study.

In conclusion, VEGF polymorphisms with lower activity were associated with the severity of HD, and VEGFR2 polymorphisms and the combination of VEGF and VEGFR2 polymorphisms, with stronger interactions, were associated with the intractability of GD. VEGF and VEGFR2 polymorphisms with higher activity were associated with goiter size at the onset of GD. VEGF and VEGFR2 polymorphisms were associated with HD severity and GD intractability, respectively.

Abbreviations

AITDs, autoimmune thyroid diseases; GD, Graves’ disease; HD, Hashimoto’s disease; IFN, interferon; IL, interleukin; VEGFA, Vascular endothelial growth factor A; RA, rheumatoid arthritis; SLE, systemic lupus erythematosus; VEGFR, Vascular endothelial growth factor receptor; KD, Kawasaki disease; PBMC, peripheral blood mononuclear cells; FT4, free T4; FT3, free T3; TSH, thyrotropin; TRAb, anti-thyrotropin receptor antibody; TgAb, anti-thyroglobulin antibody; McAb, anti-thyroid microsomal antibody; EDTA, ethylenediamine tetraacetic acid; PCR, polymerase chain reaction; RFLP, restriction fragment length polymorphism; ELISA, enzyme-linked immunosorbent assay; CTL, cytotoxic T lymphocytes; Th, helper T; sVEGFR2, soluble forms of VEGFR

Acknowledgements

This work was supported by JSPS KAKENHI grant nos. JP17H04111, JP17K15774, and JP19H04048.

Disclosure

The authors declare that they do not have any conflicts of interest.

References

• 1   Weetman  AP (2013) The immunopathogenesis of chronic autoimmune thyroiditis one century after hashimoto. Eur Thyroid J 1: 243–250.
• 2   Menconi  F,  Marcocci  C,  Marino  M (2014) Diagnosis and classification of Graves’ disease. Autoimmun Rev 13: 398–402.
• 3   McLeod  DS,  Cooper  DS (2012) The incidence and prevalence of thyroid autoimmunity. Endocrine 42: 252–265.
• 4   Brix  TH,  Hegedus  L (2012) Twin studies as a model for exploring the aetiology of autoimmune thyroid disease. Clin Endocrinol (Oxf) 76: 457–464.
• 5   Coppede  F (2017) Epigenetics and autoimmune thyroid diseases. Front Endocrinol (Lausanne) 8: 149.
• 6   Eschler  DC,  Hasham  A,  Tomer  Y (2011) Cutting edge: the etiology of autoimmune thyroid diseases. Clin Rev Allergy Immunol 41: 190–197.
• 7   Ito  C,  Watanabe  M,  Okuda  N,  Watanabe  C,  Iwatani  Y (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.
• 8   Nanba  T,  Watanabe  M,  Akamizu  T,  Iwatani  Y (2008) The –590CC genotype in the IL4 gene as a strong predictive factor for the development of hypothyroidism in Hashimoto disease. Clin Chem 54: 621–623.
• 9   Nanba  T,  Watanabe  M,  Inoue  N,  Iwatani  Y (2009) Increases of the Th1/Th2 cell ratio in severe Hashimoto’s disease and in the proportion of Th17 cells in intractable Graves’ disease. Thyroid 19: 495–501.
• 10   Yamada  H,  Watanabe  M,  Nanba  T,  Akamizu  T,  Iwatani  Y (2008) The +869T/C polymorphism in the transforming growth factor-beta1 gene is associated with the severity and intractability of autoimmune thyroid disease. Clin Exp Immunol 151: 379–382.
• 11   Hayashi  F,  Watanabe  M,  Nanba  T,  Inoue  N,  Akamizu  T, et al. (2009) Association of the –31C/T functional polymorphism in the interleukin-1beta gene with the intractability of Graves’ disease and the proportion of T helper type 17 cells. Clin Exp Immunol 158: 281–286.
• 12   Castagnone  D,  Rivolta  R,  Rescalli  S,  Baldini  MI,  Tozzi  R, et al. (1996) Color Doppler sonography in Graves’ disease: value in assessing activity of disease and predicting outcome. AJR Am J Roentgenol 166: 203–207.
• 13   Ralls  PW,  Mayekawa  DS,  Lee  KP,  Colletti  PM,  Radin  DR, et al. (1988) Color-flow Doppler sonography in Graves disease: “thyroid inferno”. AJR Am J Roentgenol 150: 781–784.
• 14   Nagura  S,  Katoh  R,  Miyagi  E,  Shibuya  M,  Kawaoi  A (2001) Expression of vascular endothelial growth factor (VEGF) and VEGF receptor-1 (Flt-1) in Graves disease possibly correlated with increased vascular density. Hum Pathol 32: 10–17.
• 15   Iitaka  M,  Miura  S,  Yamanaka  K,  Kawasaki  S,  Kitahama  S, et al. (1998) Increased serum vascular endothelial growth factor levels and intrathyroidal vascular area in patients with Graves’ disease and Hashimoto’s thyroiditis. J Clin Endocrinol Metab 83: 3908–3912.
• 16   Laurberg  P,  Buchholtz Hansen  PE,  Iversen  E,  Eskjaer Jensen  S,  Weeke  J (1986) Goitre size and outcome of medical treatment of Graves’ disease. Acta Endocrinol (Copenh) 111: 39–43.
• 17   Vitti  P,  Rago  T,  Chiovato  L,  Pallini  S,  Santini  F, et al. (1997) Clinical features of patients with Graves’ disease undergoing remission after antithyroid drug treatment. Thyroid 7: 369–375.
• 18   Winsa  B,  Dahlberg  A,  Jansson  R,  Agren  H,  Karlsson  FA (1990) Factors influencing the outcome of thyrostatic drug therapy in Graves’ disease. Acta Endocrinol (Copenh) 122: 722–728.
• 19   Schleusener  H,  Schwander  J,  Fischer  C,  Holle  R,  Holl  G, et al. (1989) Prospective multicentre study on the prediction of relapse after antithyroid drug treatment in patients with Graves’ disease. Acta Endocrinol (Copenh) 120: 689–701.
• 20   Kim  TY,  Park  YJ,  Park  DJ,  Chung  HK,  Kim  WB, et al. (2003) Epitope heterogeneity of thyroid-stimulating antibodies predicts long-term outcome in Graves’ patients treated with antithyroid drugs. J Clin Endocrinol Metab 88: 117–124.
• 21   Allahabadia  A,  Daykin  J,  Holder  RL,  Sheppard  MC,  Gough  SC, et al. (2000) Age and gender predict the outcome of treatment for Graves’ hyperthyroidism. J Clin Endocrinol Metab 85: 1038–1042.
• 22   Glaser  NS,  Styne  DM (1997) Predictors of early remission of hyperthyroidism in children. J Clin Endocrinol Metab 82: 1719–1726.
• 23   Hicklin  DJ,  Ellis  LM (2005) Role of the vascular endothelial growth factor pathway in tumor growth and angiogenesis. J Clin Oncol 23: 1011–1027.
• 24   Ferrara  N (2004) Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev 25: 581–611.
• 25   Namiki  A,  Brogi  E,  Kearney  M,  Kim  EA,  Wu  T, et al. (1995) Hypoxia induces vascular endothelial growth factor in cultured human endothelial cells. J Biol Chem 270: 31189–31195.
• 26   Berse  B,  Brown  LF,  Van de Water  L,  Dvorak  HF,  Senger  DR (1992) Vascular permeability factor (vascular endothelial growth factor) gene is expressed differentially in normal tissues, macrophages, and tumors. Mol Biol Cell 3: 211–220.
• 27   Kondo  S,  Asano  M,  Matsuo  K,  Ohmori  I,  Suzuki  H (1994) Vascular endothelial growth factor/vascular permeability factor is detectable in the sera of tumor-bearing mice and cancer patients. Biochim Biophys Acta 1221: 211–214.
• 28   Poon  RT,  Fan  ST,  Wong  J (2001) Clinical implications of circulating angiogenic factors in cancer patients. J Clin Oncol 19: 1207–1225.
• 29   Carvalho  JF,  Blank  M,  Shoenfeld  Y (2007) Vascular endothelial growth factor (VEGF) in autoimmune diseases. J Clin Immunol 27: 246–256.
• 30   Ozgonenel  L,  Cetin  E,  Tutun  S,  Tonbaklar  P,  Aral  H, et al. (2010) The relation of serum vascular endothelial growth factor level with disease duration and activity in patients with rheumatoid arthritis. Clin Rheumatol 29: 473–477.
• 31   Meki  AR,  Al-Shobaili  H (2014) Serum vascular endothelial growth factor, transforming growth factor beta1, and nitric oxide levels in patients with psoriasis vulgaris: their correlation to disease severity. J Clin Lab Anal 28: 496–501.
• 32   Kuryliszyn-Moskal  A,  Klimiuk  PA,  Sierakowski  S,  Ciolkiewicz  M (2007) Vascular endothelial growth factor in systemic lupus erythematosus: relationship to disease activity, systemic organ manifestation, and nailfold capillaroscopic abnormalities. Arch Immunol Ther Exp (Warsz) 55: 179–185.
• 33   Vural  P,  Degirmencioglu  S,  Erden  S,  Gelincik  A (2009) The relationship between transforming growth factor-beta1, vascular endothelial growth factor, nitric oxide and Hashimoto’s thyroiditis. Int Immunopharmacol 9: 212–215.
• 34   Ferrara  N,  Gerber  HP,  LeCouter  J (2003) The biology of VEGF and its receptors. Nat Med 9: 669–676.
• 35   Shalaby  F,  Rossant  J,  Yamaguchi  TP,  Gertsenstein  M,  Wu  XF, et al. (1995) Failure of blood-island formation and vasculogenesis in Flk-1-deficient mice. Nature 376: 62–66.
• 36   Bernatchez  PN,  Soker  S,  Sirois  MG (1999) Vascular endothelial growth factor effect on endothelial cell proliferation, migration, and platelet-activating factor synthesis is Flk-1-dependent. J Biol Chem 274: 31047–31054.
• 37   McMahon  G (2000) VEGF receptor signaling in tumor angiogenesis. Oncologist 5 Suppl 1: 3–10.
• 38   Millauer  B,  Shawver  LK,  Plate  KH,  Risau  W,  Ullrich  A (1994) Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1 mutant. Nature 367: 576–579.
• 39   Herold-Mende  C,  Steiner  HH,  Andl  T,  Riede  D,  Buttler  A, et al. (1999) Expression and functional significance of vascular endothelial growth factor receptors in human tumor cells. Lab Invest 79: 1573–1582.
• 40   Guo  S,  Colbert  LS,  Fuller  M,  Zhang  Y,  Gonzalez-Perez  RR (2010) Vascular endothelial growth factor receptor-2 in breast cancer. Biochim Biophys Acta 1806: 108–121.
• 41   Chorley  BN,  Wang  X,  Campbell  MR,  Pittman  GS,  Noureddine  MA, et al. (2008) Discovery and verification of functional single nucleotide polymorphisms in regulatory genomic regions: current and developing technologies. Mutat Res 659: 147–157.
• 42   Albert  PR (2011) What is a functional genetic polymorphism? Defining classes of functionality. J Psychiatry Neurosci 36: 363–365.
• 43   Breunis  WB,  Biezeveld  MH,  Geissler  J,  Ottenkamp  J,  Kuipers  IM, et al. (2006) Vascular endothelial growth factor gene haplotypes in Kawasaki disease. Arthritis Rheum 54: 1588–1594.
• 44   Yang  B,  Cross  DF,  Ollerenshaw  M,  Millward  BA,  Demaine  AG (2003) Polymorphisms of the vascular endothelial growth factor and susceptibility to diabetic microvascular complications in patients with type 1 diabetes mellitus. J Diabetes Complications 17: 1–6.
• 45   Shahbazi  M,  Fryer  AA,  Pravica  V,  Brogan  IJ,  Ramsay  HM, et al. (2002) Vascular endothelial growth factor gene polymorphisms are associated with acute renal allograft rejection. J Am Soc Nephrol 13: 260–264.
• 46   Vural  P,  Baki  M,  Dogru-Abbasoglu  S,  Ozderya  A,  Karadag  B, et al. (2012) Vascular endothelial growth factor polymorphisms increase the risk of developing Graves’ disease. Int Immunopharmacol 14: 133–137.
• 47   Awata  T,  Inoue  K,  Kurihara  S,  Ohkubo  T,  Watanabe  M, et al. (2002) A common polymorphism in the 5’-untranslated region of the VEGF gene is associated with diabetic retinopathy in type 2 diabetes. Diabetes 51: 1635–1639.
• 48   Vailati  FB,  Crispim  D,  Sortica  DA,  Souza  BM,  Brondani  LA, et al. (2012) The C allele of –634G/C polymorphism in the VEGFA gene is associated with increased VEGFA gene expression in human retinal tissue. Invest Ophthalmol Vis Sci 53: 6411–6415.
• 49   Krippl  P,  Langsenlehner  U,  Renner  W,  Yazdani-Biuki  B,  Wolf  G, et al. (2003) A common 936 C/T gene polymorphism of vascular endothelial growth factor is associated with decreased breast cancer risk. Int J Cancer 106: 468–471.
• 50   Wang  Y,  Zheng  Y,  Zhang  W,  Yu  H,  Lou  K, et al. (2007) Polymorphisms of KDR gene are associated with coronary heart disease. J Am Coll Cardiol 50: 760–767.
• 51   Gao  Y,  Ma  P,  He  Y,  Liu  Y,  Jiang  Y (2016) Genetic Variations of Kinase Inserts Domain Receptor (KDR) gene are associated with the risk of astrocytomas. Mol Neurobiol 53: 2541–2549.
• 52   Watanabe  M,  Yamamoto  N,  Maruoka  H,  Tamai  H,  Matsuzuka  F, et al. (2002) Independent involvement of CD8+ CD25+ cells and thyroid autoantibodies in disease severity of Hashimoto’s disease. Thyroid 12: 801–808.
• 53   Brogan  IJ,  Khan  N,  Isaac  K,  Hutchinson  JA,  Pravica  V, et al. (1999) Novel polymorphisms in the promoter and 5’ UTR regions of the human vascular endothelial growth factor gene. Hum Immunol 60: 1245–1249.
• 54   Del Bo  R,  Scarlato  M,  Ghezzi  S,  Maestroni  A,  Sjolind  L, et al. (2006) VEGF gene variability and type 1 diabetes: evidence for a protective role. Immunogenetics 58: 107–112.
• 55   Bagchi  M,  Kim  LA,  Boucher  J,  Walshe  TE,  Kahn  CR, et al. (2013) Vascular endothelial growth factor is important for brown adipose tissue development and maintenance. FASEB J 27: 3257–3271.
• 56   Lin  W (2017) Neuroprotective effects of vascular endothelial growth factor A in the experimental autoimmune encephalomyelitis model of multiple sclerosis. Neural Regen Res 12: 70–71.
• 57   Zdravkovic  ND,  Jovanovic  IP,  Radosavljevic  GD,  Arsenijevic  AN,  Zdravkovic  ND, et al. (2014) Potential dual immunomodulatory role of VEGF in ulcerative colitis and colorectal carcinoma. Int J Med Sci 11: 936–947.
• 58   Quinn  TP,  Peters  KG,  De Vries  C,  Ferrara  N,  Williams  LT (1993) Fetal liver kinase 1 is a receptor for vascular endothelial growth factor and is selectively expressed in vascular endothelium. Proc Natl Acad Sci U S A 90: 7533–7537.
• 59   Ye  C,  Feng  C,  Wang  S,  Wang  KZ,  Huang  N, et al. (2004) sFlt-1 gene therapy of follicular thyroid carcinoma. Endocrinology 145: 817–822.
• 60   Kendall  RL,  Thomas  KA (1993) Inhibition of vascular endothelial cell growth factor activity by an endogenously encoded soluble receptor. Proc Natl Acad Sci U S A 90: 10705–10709.
• 61   Toi  M,  Bando  H,  Ogawa  T,  Muta  M,  Hornig  C, et al. (2002) Significance of vascular endothelial growth factor (VEGF)/soluble VEGF receptor-1 relationship in breast cancer. Int J Cancer 98: 14–18.
• 62   Kou  B,  Li  Y,  Zhang  L,  Zhu  G,  Wang  X, et al. (2004) In vivo inhibition of tumor angiogenesis by a soluble VEGFR-2 fragment. Exp Mol Pathol 76: 129–137.
• 63   Ebos  JM,  Lee  CR,  Bogdanovic  E,  Alami  J,  Van Slyke  P, et al. (2008) Vascular endothelial growth factor-mediated decrease in plasma soluble vascular endothelial growth factor receptor-2 levels as a surrogate biomarker for tumor growth. Cancer Res 68: 521–529.