Association between functional polymorphisms in IL‐33/ST2 pathway and risk of osteosarcoma

Abstract Interleukin (IL)‐33/ST2 pathway plays crucial roles in tumour growth and metastasis. The aim of this study was to investigate the association of two functional polymorphisms (IL‐33 rs7025417 and ST2 rs3821204) with osteosarcoma (OS) risk. The rs7025417 and rs3821204 were genotyped by Taqman assay. IL‐33 mRNA and protein levels were measured by real‐time PCR or enzyme‐linked immunosorbent assay. The luciferase activity was measured by a dual luciferase reporter gene assay. The allele‐specific transcription factor binding for rs7025417 was examined by ChIP‐seq. The IL‐33 rs7025417 CC genotype was significantly associated with a decreased risk of OS (CC vs TT: OR = 0.59, 95% CI, 0.41‐0.85; recessive model: OR = 0.68, 95% CI, 0.49‐0.94; C vs T: OR = 0.76, 95% CI, 0.63‐0.91). Combined analysis showed that the IL‐33 rs7025417CT/CC‐ST2 rs3821204CG/CC and the IL‐33 rs7025417CT/CC‐ST2 rs3821204GG genotypes also had a decreased risk of OS. IL‐33 mRNA and protein levels in OS patients were significantly higher than controls. Patients with the rs7025417 CC genotype exhibited lower levels of IL‐33 (P = .03). The rs7025417 C allele presented a lower transcriptional activity by disrupting the binding site to c‐Myb (P < .01). Moreover, the rs3821204 G/C influences the transcriptional activity and ST2 mRNA expression by altering the binding site of miR‐202‐3p. These findings suggest that the rs7025417 and rs3821204 may have a combined effect to protect against the development of OS by decreasing the expression levels of IL‐33 or ST2.

The luciferase activity was measured by a dual luciferase reporter gene assay. The allele-specific transcription factor binding for rs7025417 was examined by ChIP-seq.
The IL-33 rs7025417 CC genotype was significantly associated with a decreased rs7025417CT/CC-ST2 rs3821204GG genotypes also had a decreased risk of OS.
IL-33 mRNA and protein levels in OS patients were significantly higher than controls. Patients with the rs7025417 CC genotype exhibited lower levels of IL-33 (P = .03). The rs7025417 C allele presented a lower transcriptional activity by disrupting the binding site to c-Myb (P < .01). Moreover, the rs3821204 G/C influences the transcriptional activity and ST2 mRNA expression by altering the binding site of miR-202-3p. These findings suggest that the rs7025417 and rs3821204 may have a combined effect to protect against the development of OS by decreasing the expression levels of IL-33 or ST2.

K E Y W O R D S
IL-33, osteosarcoma, plasma, polymorphism, ST2

| INTRODUCTION
Osteosarcoma (OS) is the most common primary malignant neoplasm in bone and accounts for about 60% of all cancer-related death in children and adolescents. [1][2][3] Despite improvements of therapeutic strategies, the overall 5-year survival rate remains 60%-70%. [4][5][6][7] Treatment often fails because of recurrent OS and/or metastasis, especially pulmonary metastasis that is the major cause of death. 5,7 Therefore, it is vital to better understand the biological aetiology underlying the development and progression of OS. Previous studies have demonstrated some risk factors for the carcinogenesis of OS such as radiation and chemicals exposure. 8,9 However, not all individuals develop OS when exposed to these risk factors, indicating that genetic factors may be involved in the development of OS. Our previous work showed that interleukin (IL)-12A rs568408 GA and IL-16 rs11556218 TG genotype may confer the susceptibility to OS risk. 10,11 Besides these cytokines' polymorphisms, IL-1B-31CC and -511TT were reported to be protective factors against OS. 12 IL-33, a member of the IL-1 family, is secreted by a wide variety of cell types, including fibroblasts, mast cells, dendritic cells, macrophages, osteoblasts, endothelial cells and epithelial cells. 13 Binding of   IL-33 to its receptor ST2 induces the activation of NF-jB and Th-2 pro-inflammatory cytokines, such as IL-4, IL-5 and IL-13, 14,15 which play important roles in the development of OS. 16,17 Epithelialderived IL-33 can promote tumorigenesis and metastasis both in vivo and in vitro. [18][19][20] These findings implied that IL-33 may be tumour-associated pro-inflammatory cytokine.
IL-33, located on chromosome 9p24.1 in the human genome, contains several single nucleotide polymorphisms (SNPs) within the full sequence. Most of them, however, are located in the intron of IL-33 and have no function. In this study, we focused on a functional polymorphism rs7025417 that are located in the promoter of IL-33 (À1611 bp from the transcription start site), with the different genotypes affecting the levels of circulating IL-33. 21 Because IL-33 exerts functions via ST2, another SNP rs3821204 was selected, which was located within the 3 0 untranslated region (UTR) of ST2 mRNA. The rs3821204 can disrupt the binding site of miR-202-3p and finally affect plasma-soluble ST2 levels. 22 To date, no study has reported on the association of the 2 functional SNPs with OS risk. The aim of this study was to investigate whether the 2 SNPs influenced the susceptibility to OS in a Chinese population. Moreover, whether the SNPs influence on IL-33 expression levels and transcriptional activities were also examined.

| Ethics, consent and permissions
The hospital-based case-control study was approved by the Review Boards of Affiliated Hospital of Youjiang Medical College for Nationalities. All individuals agreed to participate in the study and provided informed consent.

| Consent to publish
The participants signed the consent to publish the data. Detailed information of the study participants was described in our previous study. 11 Briefly, OS patients were diagnosed by histological examination. The following information was collected: age of diagnosis, gender, family history of cancer, tumour location and metastasis status. Patients who had a history of familial cancer were excluded from this study. The healthy controls were recruited from the same hospital with the following selection criteria: no cardiovascular diseases, hypertension, diabetes mellitus and other inflammatory diseases; no history of any cancer and no family history of any cancer.

| Study population
All the participants in this study were unrelated Han Chinese.

| SNPs selection
It is well known that SNPs in the non-coding region of genes contribute to the susceptibility to OS. In this study, we selected SNPs according to the following criteria: (i) SNPs located in the non-coding region of IL-33 or ST2; (ii) functional SNPs with known biological mechanism based on published literatures.

| Genotyping
Fasting venous blood was obtained from peripheral vein of each participant. After centrifugation at 1000 g for 10 minutes, plasma was aliquoted and stored at À80°C. DNA was extracted using a TIA-Namp Blood DNA Kit (Tiangen Inc., Beijing, China). The rs7025417 and rs3821204 were genotyped by Taqman assay (Applied Biosystems, Foster City, CA, USA). The assay ID for the 2 SNPs was C_31940410_20 and C_1226153_10 respectively. The genotyping results were verified by Sanger sequencing and the concordant rate was 100%.

| Real-time PCR
Total RNA was isolated from peripheral blood cells using a commercial kit (Tiangen, Beijing, China). A total of 500 ng RNA was used for the reverse transcription reaction. A real-time PCR assay was per-

| Chromatin immunoprecipitation (ChIP) assay
The chromatin immunoprecipitation assay (ChIP) assay was performed with a commercial kit from ThermoFisher Scientific (Waltham, MA, USA) following the manufacturer's protocol. MG-63 and Saos-2 cells were cross-linked in 1% formaldehyde for 10 minutes; DNA was then subjected to immunoprecipitation using antibodies against c-Myb or non-specific rabbit IgG (Abcam). Purified DNA was amplified by PCR with the primers: 5 0 -TTGTGTCTCCTTTCCCCTACA-3 0 (forward) and 5 0 -ATGCACACCAACCACTTTGA-3 0 (reverse). PCR product was then analysed by Sanger sequencing.

| Statistical analysis
All statistical analyses were performed with the statistical software package SPSS 13.0 (SPSS Inc., Chicago, IL, USA). Continuous data, including age distribution and dual-luciferase reporter gene activities, were compared using a Student's t test. HWE and the genotype distributions of the rs7025417 and rs3821204 in cases and controls were compared using chi-squared test. Odds ratios (ORs) and 95% confidence intervals (CIs) were computed to assess the association of the rs7025417 and rs3821204 with OS risk. Differences in plasma IL-33 levels in patients with OS and controls were analysed using a Mann-Whitney rank-sum test. Two-sided P values <.05 were considered statistically significant.

| Characteristics of the study subjects
The characteristics of controls and patients with OS are shown in Table 1. The median age of diagnosis in controls and cases was 18.0

IL-33 mRNA and protein levels
The relative expression of IL-33 mRNA was higher in OS patients than controls (P < .001, Figure 1A). We then analysed the correlation between the IL-33 rs7025417 and IL-33 mRNA levels. We found that the rs7025417 CC genotype was associated with a reduced IL-33 mRNA levels (P = .03, Figure 1B). Moreover, the median levels of plasma IL-33 in OS patients was 405.3 pg/mL, which was higher than that in controls (206.9 pg/mL) (P = .01, Figure 2A).
As shown in Figure 2B, patients with the rs7025417 CC genotype exhibited lower levels of IL-33 compared to those with the TT genotype (P = .03).

| Effect of the IL-33 rs7025417 on the transcriptional activity
To identify whether the -1611 bp polymorphism (rs7025417) in the promoter of IL-33 influenced the transcriptional activity, we constructed plasmids containing rs7025417T and rs7025417C allele and analysed the luciferase activity. As shown in Figure 3

| IL-33 rs7025417 T allele-specific binding to transcription factor c-Myb
In silico analysis was used to predict allele-specific transcription factor binding for rs7025417, and we found that rs7025417 T to C base change might affect the binding site of c-Myb. Chip-seq was then performed to assess the validity of allele-specific transcription factor binding. As shown in Figure 4A, the DNA fractions were immunoprecipitated specifically with the c-Myb antibody but not non-specific IgG in MG-63 and Saos-2 cells. Further sequencing analysis revealed that the region contained rs7025417 TT genotype ( Figure 4B,C).

| The rs3821204 G/C influences the transcriptional activity and ST2 mRNA expression by altering the binding site of miR-202-3p
Previous study showed that the rs3821204 disrupted a miR-202-3p seeding site, resulted in an allelic difference of sST2 mRNA stability and expression, and finally modified the susceptibility of hypertension. 22 In this study, we cotransfected rs3821204 G or C plasmid with miR-202-3p mimics (50 lmol/L) or negative control into HEK293 cells and confirmed the alteration of miR-202-3p binding site and a significantly decreased luciferase activity of rs3821204-G but not rs3821204-C ( Figure 5A,B). Overexpression of miR-202-3p caused a decrease of ST2 mRNA levels in Saos-2 (rs3821204 GG) cells but not MG-63 (rs3821204 CC) cells ( Figure 5C).

| DISCUSSION
The hospital-based case-control study showed that the rs7025417 CC  real-time PCR and ELISA showed for the first time that IL-33 mRNA and protein levels in OS patients were higher than controls. Notably, IL-33 mRNA and protein levels in patients with the rs7025417 CC genotype were lower than those with the rs7025417 TT genotype.
The dual-luciferase reporter assay showed that the rs7025417C allele presented a lower luciferase activity. ChIP assay revealed that the rs7025417 T allele can specifically bind to transcription factor c-Myb.
Additionally, the rs3821204 G/C influences the transcriptional activity and ST2 mRNA expression by altering the binding site of miR-202-3p.
These findings suggest that the rs7025417 and rs3821204 may reduce the transcriptional activity and IL-33/ST2 expression, and eventually result in the decreased risk of OS.
A possible explanation for the positive results of the IL-33CC genotype decreasing OS risk is that the rs7025417 located in the promoter region of IL-33 and exerted influences on the transcriptional activity. In this study, the dual-luciferase gene reporter assay was performed and we found that the rs7025417 T to C shift decreased the luciferase activity. These findings were in agreement with previous results reported by Tu 28 and squamous cell carcinoma of tongue. 29 The higher IL-33 expression predicted a worse prognosis and progression-free survival after chemotherapy. 27,29,30 Downregulation of IL-33 expression can inhibit cell growth, metastasis and prolong the survival time. 28 In contrast, some authors reported reduced IL-33 levels in patients with multiple myeloma 31 and lung cancer. 32 No   F I G U R E 5 The rs3821204 G/C influences the transcriptional activity and ST2 mRNA expression by altering the binding site of miR-202-3p. A, In silico prediction of miR-203-3p binding site to ST2 3 0 UTR containing rs3821204 G or C allele. B, The rs3821204 G or C plasmid was cotransfected with miR-202-3p mimics or negative control into HEK293 cells. Relative luciferase activity was measured using the dual luciferase reporter assay. C, miRNA-202-3p mimics or negative control was transfected into Saos-2 (rs3821204 GG) and MG-63 (rs3821204 CC) cells. Relative expression of ST2 mRNA was analysed using real-time PCR. Data were expressed by mean AE standard error (*P < .05) secretion of interferon-gamma, IL-17 and tumour necrosis factor-a, which attenuated tumour growth. 36 These findings suggest the crosstalk of IL-33 and ST2 in tumorigenesis. In this study, we found that the ORs of the combined genotypes of the 2 SNPs were lower than those of a single one in the association with OS. Further functional analysis revealed that the rs3821204 influences the transcriptional activity and ST2 mRNA levels by altering the binding site of miR-202-3p. miR-202-3p was a tumour suppressor in several cancer types, such as gastric cancer 37 and colorectal carcinoma, 38 which may be used as a target to develop novel therapeutic approaches.
Based on the results above, a conclusion was made that the IL-33 rs7025417 and ST2 rs3821204 may affect the occurrence of OS by interaction with each other in a biological way.
We have to acknowledge some limitations of this study. As is well known, genetic variants may have different effect in diverse ethnicities. In the present study, only Chinese Han was recruited, and thus our results cannot be extended directly to other ethnic groups until confirmation analysis was performed. In addition to genetic factors, environmental factors are crucial in the development and progression of OS. The study design, however, did not take them into consideration, which prevented our further interaction analysis of IL-33 and environmental factors with the risk of OS.
In conclusion, we found a functional polymorphism in the promoter of IL-33 (rs7025417). The rs7025417 CC genotype protected against OS tumorigenesis and reduced IL-33 expression levels by decreasing the transcriptional activity. Our findings further supported the key role of IL-33/ST2 pathway in OS and highlighted IL-33/ST2 as valuable targets for prevention and treatment of OS.

CONF LICT OF I NTEREST
None declared.