RNASET2,GPR174, and PTPN22 gene polymorphisms are related to the risk of liver damage associated with the hyperthyroidism in patients with Graves’ disease

Abstract

Objectives

This study was designed to unveil the association of GPR174 rs3827440, PTPN22 rs3789604, and RNASET2 rs9355610 with the onset of liver damage (LD) among the Graves’ disease (GD) patients.

Methods

A total of 120 GD patients were divided into the none‐LD and LD groups. Several indicators were detected for assessing liver functions, and genotypes of single nucleotide polymorphisms (SNPs) were identified. Logistic regression was introduced for investigating the relationship between risk SNPs and LD‐associated hyperthyroidism in GD patients.

Results

Significant differences were identified between LD and none‐LD groups regarding genotype distributions of rs3827440, rs3789604, and rs9355610. Results from logistic regression indicted that among the GD patients, C carriers of PTPN22 rs3789604 were associated with a higher risk of LD‐associated hyperthyroidism, while C carriers of rs3827440 (GPR174) and G carriers of rs9355610 (RNASET2) were associated with a reduced risk of LD‐associated hyperthyroidism.

Conclusions

The C allele of rs3789604 (PTPN22) was a significant risk factor for LD‐associated hyperthyroidism in GD patients, whereas C allele of GPR174 rs3827440 and G allele of RNASET2 rs9355610 appeared to be a protective factor for this disease.

1. INTRODUCTION

Graves’ disease (GD) is the most usual cause of hyperthyroidism, and its prevalence is high at 0.25%~1.09% in China and up to 1.3% in the US.1, 2 The GD patients produce auto‐antibodies which bound to the thyroid‐stimulating hormone receptor (TSHR), leading to hyperthyroidism and diffuse hyperplasia of the thyroid gland.3, 4, 5 Approximately 70%‐80% of untreated GD patients are estimated to have biochemical liver damage (LD).5 LD will add difficulties to the selection of appropriate GD treatments, particularly for assessing the anti‐thyroid drugs (ATDs).5 What's more, a large number of studies have confirmed that metabolites of ATDs, particularly propylthiouracil, might contribute to severe hepatotoxicity within GD patients.6

To date, some evidence has revealed that GD is a multi‐factorial disease, which may result from interactions between susceptible genes and environmental hazards.3, 7 As suggested by family and twin studies, about 79% of predisposition to GD could be attributed to genetic factors.3 Thus, identifying susceptible genes and loci that are related to hepatic dysfunctions may be critical to discovering novel interventions for GD. Recent studies have screened out a series of genes as potential thyroid disease susceptibility loci, including Ribonuclease T2 (RNASET2), G protein‐coupled receptor 174 (GPR174), and protein tyrosine phosphatase non‐receptor type 22 (PTPN22).4, 8, 9

RNASET2 is discovered as the member of the Rh/T2/S family and is able to encode extra‐cellular RNases.10 The abnormal RNASET2 expression is probably due to its sub‐cellular distribution and the presence of splice variants.11 Previous studies have implied that RNASET2 may participate in the immune response involved in GD.12 Investigations on Chinese Han populations have dug out evidence indicating that RNASET2 polymorphisms especially the loci rs9355610 is significantly associated with the susceptibility to GD.4 GPR174 is an Xq21 hypothetical purinergic receptor, and it is extensively expressed in immune and thyroid tissues.9, 13 Furthermore, T allele of GPR174 rs3827440 has been suggested to be associated with a higher risk of GD in both Chinese and Polish populations.14 PTPN22 is located on chromosome 1 (1p13.1−lp13.3) and it encodes human lymphoid tyrosine phosphatase (LYP).8, 15 More importantly, PTPN22 has been discovered to have significant associations with a series of autoimmune diseases, such as GD, rheumatoid arthritis (RA), systemic lupus erythematosus, and type 1 diabetes.8, 16 Aside from that, PTPN22 is highly expressed in immune and hematopoietic cells, and it has a negative impact on both T‐cell activation and signal transduction of Csk proteins.3, 16 As suggested by Ichimura et al., rs3789604 was a functional SNP of R620W within PTPN22, and it was related to GD among the Japanese population.17 Besides, it has been reported that the gain‐of‐function variants of PTPN22 might be associated with the rejection in liver transplantation, suggesting that PTPN22 could also function to regulate the liver damages derived from hyperthyroidism of GD.18

Since few studies have clarified the association between genetic risk factors and LD‐associated hyperthyroidism in GD patients, this prospective cohort study was specifically designed in order to provide solid evidence for such a relationship. Additionally, investigation of gene polymorphism within RNASET2, GPR174, and PTPN22 may contribute to discovering new interventions for LD‐associated hyperthyroidism in GD patients.

2. MATERIALS AND METHODS

2.1. Subjects

Altogether 120 GD patients were recruited from the First Affiliated Hospital of Nanchang University from March 2014 to April 2015. They were further divided into the none‐LD group (n=79) and the LD group (n=41). All the participants were individuals of Han ethnicities, and they all have signed the informed consents. This study was approved by the ethics committee of the First Affiliated Hospital of Nanchang University.

2.2. Diagnostic criteria of GD

GD was diagnosed based on the following criteria: (1) symptoms or signs of high metabolism were observed; (2) diffuse enlargement of thyroid gland was existent; (3) laboratory examination showed increased free triiodothyronine (FT₃) and free thyroxine (FT₄), declined thyroid stimulating hormone (TSH) and positive thyroid stimulating hormone receptor (TSHR) antibody; (4) potential symptoms of exophthalmos and other invasive diseases, such as pretibial mucous edema, were present; (5) nodular goiter toxicity, highly functional thyroid adenoma and other diseases that could also cause hyperthyroidism were excluded.

2.3. Diagnostic criteria of LD in GD patients

The diagnosis of GD‐induced LD was based on the following diagnostic criteria: (1) with untreated GD hyperthyroidism; (2) presence of at least one indicator of LD including increase in alanine aminotransferase (ALT) and aspartate transaminase (AST), total bilirubin (TBIL), gamma glutamyltranspeptidase (GGT), and alkaline phosphatase (ALP), declined total protein (TP), serum albumin (ALB), and symptoms of hepatomegaly and jaundice.

2.4. Collection of specimens

Venous blood samples were collected from patients prior to the therapy. The anticoagulant citrate dextrose (ACD) solution was used for anti‐coagulation. The extraction kit for DNA (Teck biological technology limited corporation, Beijing, China) was applied to extract the genome of subjects. Nanodrop2000 (Thermo Scientific, Waltham, MA, USA) was applied for detecting the concentration and purity of DNA, and the purity of sample DNA (A₂₆₀/A₂₈₀) was between 1.8 and 2.0. The automatic biochemical analyzer (Sinothinker Technology, Shenzhen, China) was used to evaluate several GD and LD indicators for assistance of diagnosis and patient characterization. Genomic DNA was extracted from samples based on the instructions of DNA extraction kit (Thermo Fisher Scientific, Waltham, MA, USA) and then preserved in an −20°C refrigerator.

2.5. Genotyping

SNPs related to GD were chosen and investigated in this study. Three SNPs (rs3827440 within GPR174, rs3789604 within PTPN22 and rs9355610 within RNASET2) were included according to their association with GD pathogenicity. Samples were sent to Microread Genetics Go, Ltd (Beijing, China) for direct sequencing and genotyping. They used TaqMan probes synthesized by American ABI corporation, and genotyping was conducted based on the Fluidigm EP1 platform of Fluidigm 96×96 Dynamic Array. The results were shown in Figure 1.

Figure 1.

Direct sequencing and genotyping for GPR174,PTPN22 and RNASET2 among the GD patients with LD. (A) Direct sequencing results of rs3827440 polymorphisms in GPR174: T, C/T and C. (B) Direct sequencing results of rs3789604 polymorphisms (A/C) in PTPN22: A, A/C and C. (C) Direct sequencing results of rs9355610 polymorphisms (A/G) in RNASET2: A, A/G and G

2.6. Statistical analysis

All statistics in this study were analyzed by SPSS 21.00 statistical software. Mann‐Whitney test was performed in order to analyze continuous data that were expressed as the mean±SD between two groups. Chi‐square test was carried out in order to analyze categorical data. Logistic regression which incorporated the OR and the corresponding 95% confidence interval was specifically used to assess the influence of risk loci for LD associated with the hyperthyroidism in GD patients. The power calculation was performed with the software of ps power. P‐value of less than .05 provided evidence for statistical significance.

3. RESULTS

3.1. Characteristics of study participants

The sample size was found to be robust after power calculation with the software of ps power. As shown in Table 1, 120 GD patients were enrolled, including 41 LD cases. The LD group consisted of 11 males and 30 females with an average age of (38.1±10.9) years old. The none‐LD group included 20 males and 59 females with the average age of (41.6±11.4) years old. In addition, gender, age, as well as the levels of GD indicators, including FT₃, FT₄ and TSH, showed little discrepancy between the LD and non‐LD groups (P>.05). As for LD indicators, ALT, AST, GGT, TBIL and ALP presented significantly increased levels in LD patients, while TP and ALB displayed declined levels in the LD group in comparison with those in the non‐LD group (P<.05).

Table 1.

The basic characteristics of GD patients in none‐LD group and LD group

Group	None‐LD group (n=79)	LD group (n=41)	P value
Age	41.6±11.4	38.1±10.9	.328a
Gender
Male	20	11	.858b
Female	59	30
ALT (0~40 U/L)	25.2±6.5	68.4±35.7	<.001a
AST (29~35 U/L)	30.0±7.0	50.5±19.1	<.001a
GGT (3~50 U/L)	37.0±20.1	62.9±27.2	<.001a
TBIL (3.4~17.1 μmol/L)	14.7±5.3	19.9±8.2	<.001a
ALP (40~150 U/L)	127.3±58.7	171.2±57.1	<.001a
TP (60~80 g/L)	60.4±1.8	54.8±5.2	<.001a
ALB (35~55 g/L)	38.7±2.1	33.7±2.9	<.001a
FT3 (3~9 pmol/L)	14.3±6.8	16.8±8.9	.089a
FT4 (9~25 pmol/L)	52.6±20.7	59.6±18.9	.073a
TSH (0.04‐2.5 mIU/L)	0.016±0.003	0.015±0.004	.126a

GD, Graves’ disease; none‐LD group, GD patients without LD were included; LD group, GD patients with LD were included; ALT, Alanine aminotransferase; AST, Aspartate transaminase; GGT: Gamma‐glutamyl transferase; TBIL, Total bilirubin; ALP, Alkaline phosphatase; TP, Total protein; ALB, Serum albumin; FT₃, Free triiodothyronine; FT₄, Free thyroxine; TSH, thyroid stimulating hormone.

^aRepresents that P value was related to none‐LD and LD group and was analyzed by Mann‐Whitney test.

^bRepresents that P value was related to none‐LD and LD group and was analyzed by Chi‐square test.

3.2. Association between risk SNPs and LD in GD patients

The detailed genotype and allele frequency distribution of GPR174 rs3827440, PTPN22 rs3789604, and RNASET2 rs9355610 were shown in Table 2, and the approach of logistic regression was applied to identify the risk factors for the onset of LD‐associated hyperthyroidism among GD patients. After adjusting age, gender and other confounding factors, the results of logistic regression suggested that the frequency of rs3827440 C allele in LD patients was significantly lower than that in the non‐LD group (TC vs TT: OR=0.13; 95% CI: 0.04‐0.48; P=.002; CC vs TT: OR=0.12; 95% CI: 0.02‐0.69; P=.018; TC+CC vs TT: OR=0.24; 95% CI: 0.08‐0.67, P=.007; C vs T: OR=0.31; 95% CI: 0.14‐0.68; P=.004), which provided an evidence that C allele of rs3827440 might act as a protective factor in preventing GD patients from the onset of LD. As for rs3789604 of PTPN22, it was indicated in the form of raw ORs that the C allele might increase the risk of LD in GD patients (AC vs AA: OR=1.36, 95% CI: 0.57‐3.25, P=.483; CC vs AA: OR=8.81; 95% CI: 2.49‐31.25, P<.001; AC+CC vs AA: OR=2.30, 95% CI: 1.06‐4.97, P=.033; C vs A: OR=2.79, 95% CI: 1.57‐4.94, P<.001). In addition, the adjusted ORs also attained a similar trend as the raw ORs (AC vs AA: OR=1.80, 95% CI: 0.57‐5.69, P=.315; CC vs AA: OR=35.15, 95% CI: 5.41‐228.56, P<.001; AC+CC vs AA: OR=3.09, 95% CI: 1.13‐8.47, P=.028; C vs A: OR=5.82, 95% CI: 2.66‐12.77, P<.001). Under all genetic models, it was also revealed that the G allele of RNASET2 rs9355610 might play a protective role against LD in GD patients (GA vs GG: OR=0.10, 95% CI: 0.03‐0.31, P<.001; AA vs GG: OR=0.09, 95% CI: 0.02‐0.52, P=.007; GA+AA vs GG: OR=0.17, 95% CI: 0.07‐0.42, P<.001; A vs G: OR=0.25, 95% CI: 0.13‐0.50, P<.001).

Table 2.

The genotype distribution of three SNPs in none‐LD group and LD group

Gene	Genotype	LD group (n=41)	None‐LD group (n=79)	P valuea	Raw OR (95% CI)	P valueb	Adjusted OR (95% CI)
GPR174	rs3827440
	TT	23	34	Ref.	Ref.	Ref.	Ref.
	TC	12	28	.296	0.63 (0.27‐1.50)	.002	0.13 (0.04‐0.48)
	CC	6	17	.230	0.52 (0.18‐1.53)	.018	0.12 (0.02‐0.69)
	TC+CC	18	45	.276	0.59 (0.28‐1.27)	.007	0.24 (0.08‐0.67)
	T	58	96	Ref.	Ref.	Ref.	Ref.
	C	24	62	.127	0.64 (0.36‐1.50)	.004	0.31 (0.14‐0.68)
PTPN22	rs3789604
	AA	16	47	Ref.	Ref.	Ref.	Ref.
	AC	13	28	.483	1.36 (0.57‐3.25)	.315	1.80 (0.57‐5.69)
	CC	12	4	<.001	8.81 (2.49‐31.25)	<.001	35.15 (5.41‐228.56)
	AC+CC	25	32	.033	2.30 (1.06‐4.97)	.028	3.09 (1.13‐8.47)
	A	45	122	Ref.	Ref.	Ref.	Ref.
	C	37	36	<.001	2.79 (1.57‐4.94)	<.001	5.82 (2.66‐12.77)
RNASET2	rs9355610
	GG	27	20	Ref.	Ref.	Ref.	Ref.
	GA	12	41	<.001	0.22 (0.09‐0.52)	<.001	0.10 (0.03‐0.31)
	AA	2	18	<.001	0.08 (0.02‐0.40)	.007	0.09 (0.02‐0.52)
	GA+AA	14	59	<.001	0.18 (0.08‐0.40)	<.001	0.17 (0.07‐0.42)
	G	66	81	Ref.	Ref.	Ref.	Ref.
	A	16	77	<.001	0.26 (0.14‐0.48)	<.001	0.25 (0.13‐0.50)

GD, Graves’ disease; LD, Liver damage; OR, Odds ratios; CI, Confidence interval; Ref., Reference.

^a P value for raw OR.

^b P value for adjusted OR.

4. DISCUSSION

GD is an autoimmune disorder where autoantibodies target antigens expressed in thyroid tissues and hence hyperthyroidism is promoted.14, 19, 20 Consensus has been reached that several genetic and environmental factors would affect GD.2, 14 On the other hand, there has been evidence suggesting that part of untreated GD patients tend to suffer from LD.5 Thus, we attempted to investigate the relation of genetic polymorphisms and the onset of LD‐associated hyperthyroidism among GD patients.

It has been suspected that GD may be affected by various susceptible genes, 3 and sequencing enabled us to discover that different genotypes were identified in rs3827440 (GPR174), rs3789604 (PTPN22) and rs9355610 (RNASET2). More importantly, the frequency of GPR174 rs3827440 C/T polymorphism, PTPN22 rs3789604 A/C polymorphism and RNASET2 rs9355610 G/A polymorphism significantly differed among the LD and non‐LD groups. For rs3827440, T allele carriers were associated with a higher risk of hyperthyroid LD compared with C allele carriers. Similarly, C allele carriers of rs3789604 may experience a higher risk of hyperthyroid LD when compared with the A allele carriers, whereas G allele carriers of rs9355610 were related with an increased risk of hyperthyroid LD in comparison to the A allele carriers. As suggested by recent studies, lysophosphatidylserine (LysoPS), a ligand of GPR174, had an important role in regulating the immune system.21, 22 Moreover, LysoPS was able to stimulate intracellular cyclic adenosine monophosphate (cAMP) by interacting with GPR174.22 The elevated cAMP could inhibit the production of T‐helper 1 (Th1) cytokines, and stabilize or enhance the level of T‐helper 2 (Th2) cytokines.23 Graves's disease was a Th2‐related disorder, and increased cAMP may contribute to the polarization of Th2 and hence facilitation of GD‐derived LD pathogenesis.9 Another research revealed that GPR174 was extensively expressed in immune‐related tissues, and was moderately expressed in thyroid tissues.9 Besides that, variants of PTPN22 have been considered as a significant risk factor for GD, since they triggered irregular antigen presentation and T‐cell activation in immunological synapses.24

In this study, C allele of rs3789604 was a significant risk factor for LD, while C allele of rs3827440 and G allele of rs9355610 appeared to be a protective factor for LD‐associated hyperthyroidism in GD. Several susceptible genes that are related with GD have been discovered in different ethnic populations.3 PTPN22 rs2476601 was initially discovered as a susceptible locus for autoimmune diabetes. Studies on Asian populations have suggested that rs2476601 may play a significant role in GD.25, 26, 27, 28 However, genome wide association study (GWAS) has provided contradictory results that SNPs of PTPN22 were not significantly associated with the autoimmune diseases among Asians.3

So far, a large number of researches on GD were devoted to unveiling the correlation between Graves’ disease and rs2476601.29, 30 Some studies indicated that rs2476601 rarely influenced the onset of GD in the Chinese Han population,3 while others provided evidence that rs3789604 also had an association with GD.31 Meanwhile, rs3789604 was strongly associated with GD among a Japanese population and such an association was confirmed by a study on the Chinese population.25, 26 Since this study investigated GD‐derived LD, it was reasonable to believe that the aforementioned SNPs were associated with GD‐derived LD.

Several limitations in this study should be addressed carefully. This study only indicated that the distribution of GPR174 rs3827440 polymorphisms significantly differed among the three groups. However, whether this trend was related with GD‐derived LD remained vague, hence studies on the mechanisms underlying the correlation between genotypes and GD‐derived LD were encouraged in the future.

In summary, GPR174 rs3827440, PTPN22 rs3789604, and RNASET2 rs9355610 were significantly associated with altered GD‐derived LD risk. Ongoing researches should be further devoted to this so that potential interventions for LD resulted from GD may be discovered.

5. CONCLUSION

The C allele of rs3789604 (PTPN22) was a significant risk factor for LD associated with the hyperthyroidism in GD patients, whereas C allele of rs3827440 (GPR174) and G allele of rs9355610 (RNASET2) appeared to be a protective factor for this disease. The result of this research would provide several new loci for the diagnosis and targeted therapy of LD associated with the hyperthyroidism in GD in clinical application.

Zhang Q, Liu S, Guan Y, Chen Q, Zhang Q, Min X. RNASET2, GPR174, and PTPN22 gene polymorphisms are related to the risk of liver damage associated with the hyperthyroidism in patients with Graves’ disease. J Clin Lab Anal. 2018;32:e22258 10.1002/jcla.22258