Functional polymorphisms of the TET1 gene increase the risk of neuroblastoma in Chinese children

Abstract Common genetic mutations are absent in neuroblastoma, one of the most common childhood tumours. As a demethylase of 5‐methylcytosine (m5C) modification, TET1 plays an important role in tumourigenesis and differentiation. However, the association between TET1 gene polymorphisms and susceptibility to neuroblastoma has not been reported. Three TET1 gene polymorphisms (rs16925541 A > G, rs3998860 G > A and rs12781492 A > C) in 402 Chinese patients with neuroblastoma and 473 cancer‐free controls were assessed using TaqMan. Multivariate logistic regression analysis was used to evaluate the association between TET1 gene polymorphisms and susceptibility to neuroblastoma. The GTEx database was used to analyse the impact of these polymorphisms on peripheral gene expression. The relationship between gene expression and prognosis was analysed using Kaplan–Meier analysis with the R2 platform. We found that both rs3998860 G > A and rs12781492 A > C were significantly associated with increased neuroblastoma risk. Stratified analysis further showed that rs3998860 G > A and rs12781492 A > C significantly increased neuroblastoma risk in certain subgroups. In the combined risk genotype model, 1–3 risk genotypes significantly increased risk of neuroblastoma compared with the 0 risk genotype. rs3998860 G > A and rs12781492 A > C were significantly associated with increased STOX1 mRNA expression in adrenal and whole blood, and high expression of STOX1 mRNA in adrenal and whole blood was significantly associated with worse prognosis. In summary, TET1 gene polymorphisms are significantly associated with increased neuroblastoma risk; further research is required for the potential mechanism and therapeutic prospects in neuroblastoma.


| INTRODUC TI ON
Neuroblastoma is an embryonic tumour of the sympathetic nervous system, but its origin is unclear. 1 Currently, it is generally believed that it originates from incomplete precursors of neural crest tissue. [2][3][4] Neuroblastoma is the most common solid tumour and the second most common extracranial tumour in children. 5,6 In developed countries, the incidence of neuroblastoma in children aged 0-14 years is approximately 10.1-15.0 per 1 million. 7 In China, the incidence of neuroblastoma is 7.7 cases per 1 million children. 8 Neuroblastoma exhibits a high degree of heterogeneity, from biological characteristics to clinical processes. 1 Some neuroblastomas can spontaneously subside. 1 Even after a series of intensive treatments, such as surgery, immunotherapy and radiation therapy, high-risk neuroblastomas still have a relatively poor prognosis and high risk of recurrence. [9][10][11] Patients with high-risk neuroblastoma will also face serious psychological problems such as anxiety. 12 Identifying biological characteristics associated with high-risk neuroblastoma is a research focus. In addition to MYCN amplification, some important other genetic changes have been found in neuroblastomas, such as ATRX, 13 TERT, 14 ALK 15 and RAS 16 mutations. However, data thus far are insufficient to explain the genetic variations associated with neuroblastoma risk.
As the most common genetic variation in DNA, single-nucleotide polymorphisms (SNPs) are closely related to tumour susceptibility, prognosis and immunity. [17][18][19] Different tumour types have different associated SNPs. 20 Genome-wide association studies (GWASs) have enabled large-scale exploration of SNPs truly related to tumours in the entire genome. With the application of GWAS to neuroblastoma, many SNPs have been proven to be related to its susceptibility. [21][22][23][24] John et al. 24 have shown that rs2168101 G > T located in the LMO1 gene super-enhancer is closely related to genetic susceptibility to neuroblastoma; the rs2168101 T allele is associated with reduced LMO1 expression and tumour suppression in primary neuroblastoma tumours. However, the genetic variations associated with neuroblastoma susceptibility in different populations are yet to be elucidated.
Our previous studies have shown that the BER pathway and ERCC1, XPF, NRAS and ALKBH5 gene polymorphisms affect neuroblastoma risk in Chinese children. 16,[25][26][27] However, the genetic polymorphisms associated with the risk of neuroblastoma in Chinese children are not fully understood.
Epigenetic modifications commonly found in genomes and transcriptomes have been shown to play important biological roles in growth and development, aging and various diseases. [28][29][30][31] In recent years, RNA modifications have been shown to play an important role in cancer. 28,32 Common RNA modifications include N6-methyladenosine (m6A), N1-methyladenosine (m1A), 5-methylcytosine (m5C), 5-hydroxymethylcytosine (hm5C), pseudouridine (Ψ) and N7-methylguanosine (m7G). As the most common DNA modification and important RNA modification, the role of m5C modification in cancer cannot be ignored. 29,[33][34][35] As m5C demethylases, the ten-eleven translocation (TET) family of enzymes oxidize m5C in RNA to form hm5C. 36 TET further oxidizes 5hmC to 5fC (5-formylcytosine) and 5caC (5-carboxycytidine) in DNA 37 ; thymidine DNA glycosylase recognizes 5fC and 5caC and converts them to unmethylated cytosine. [38][39][40] A member of the TET (ten-eleven translocation) family, TET1 (tet methylcytosine dioxygenase 1) has been shown to play an important biological role in cancer, including neuroblastoma, in recent years. [41][42][43][44] Fragile X mental retardation protein (FMRP) promotes demethylation of m5C by interacting with TET1 and induces transcription-coupled homologous recombination, which is an important mechanism for mRNA repair and cell survival in cancer. 41 The oncoprotein YAP induces expression of TET1, which promotes transcriptional activation and induces tumourigenesis through epigenetics, such as DNA demethylation caused by its interaction with TEAD. 42 Gao et al. 44 found that TET1 correlated negatively with neuronal differentiation in neuroblastoma cells. TET1 mediates negative regulation of neuronal differentiation by srGAP3 through non-catalytic action. 44 The TET1 gene is repeatedly mutated in lung cancer, gastrointestinal cancer, skin cancer and urinary system cancer. 45 In cancer patients treated with immune checkpoint inhibitors (ICIs), the TET1 mutation is associated with better therapeutic efficacy. 45 This suggests a potential therapeutic target for the TET1 gene. However, the important role of TET1 gene mutations in neuroblastoma needs to be further explored. Currently, there is no research indicating that TET1 gene polymorphisms are associated with susceptibility to neuroblastoma.
Based on the importance of the TET1 gene in neuroblastoma, we hypothesize that TET1 gene polymorphisms have an impact on risk of neuroblastoma. To test our hypothesis, we conducted a casecontrol study in Jiangsu, China, to explore whether TET1 gene polymorphisms are associated with susceptibility to neuroblastoma.

| Study subjects
This case-control study included 402 children with neuroblastoma and 473 noncancer children (Table S1)

| Polymorphisms selection and genotyping
We first selected potential functional SNPs located in the 5′ flanking region, 5′ untranslated region (UTR), exon, intron and 3′ UTR regions of the TET1 gene from the dbSNP database and SNPinfo website. The secondary allele frequency of all SNPs in the Han Chinese population reported in 1000 Genomes is >5%.
Linkage disequilibrium (LD) between the selected SNPs is less than 0.8. The final SNPs selected were rs16925541, rs3998860 and rs12781492. Both rs16925541 and rs3998860 are missense variants located in the coding region of the TET1 gene. The variant rs12781492 is located in the 3′ UTR of the TET1 gene and is predicted to bind to miRNA. We used TIANamp Blood DNA Kit (TianGen Biotech Co. Ltd.) to extract genomic DNA from tissue or blood samples from all subjects. The concentration and purity of the extracted DNA were measured using a UV spectrophotometer (Nano Drop Technologies, Inc.). The DNA sample was diluted and transferred to a 96-well plate. For all samples, TaqMan® SNP Genotyping Assays (Applied Biosystems) were used for genotyping. We randomly selected 10% of the samples from the case and the control groups for repeated genotyping, and the results were 100% consistent.

| Statistical analysis
Differences in genotype frequency distribution and demographic characteristics between the case group and the control group were analysed using a bilateral chi-square test. The goodness of fit χ 2 test was used for Hardy-Weinberg equilibrium (HWE). To explore the association between TET1 gene polymorphisms and susceptibility to neuroblastoma, we used multiple logistic regression analysis to analyse the odds ratio (OR) and 95% confidence interval (CI) of TET1 gene polymorphisms in the case group and the control group. The OR calculates the crude OR and adjusted OR before and after adjustment for age and sex, respectively.
Subsequently, further stratification analysis was conducted based on age, sex, tumour location and clinical stage for significant SNPs.
The above statistical analyses were performed using SAS V9.4 (SAS Institute). Gene expression quantitative trait loci (eQTLs) were analysed through the Genotype-Tissue Expression (GTEx) official website (https://www.gtexp ortal.org/home/). Analysis of the relationship between gene expression and prognosis of neuroblastoma cases derived from the GSE62564 dataset using Kaplan-Meier analysis through R2: Genomics Analysis and Visualization Platform (http://r2.amc.nl). All statistics were conducted using a two-sided test with a significance level of 0.05.

| Associations of TET1 gene polymorphisms with neuroblastoma susceptibility
We successfully obtained the genotypes of the three SNPs in 400 cases and 473 controls. As shown in Table 1, all SNPs in the control group conformed to HWE (p > 0.05). After adjusting for age and sex, we found that subjects with the rs3998860 AA genotype had a 1.5-fold increased neuroblastoma risk compared with those with the GG genotype (AA vs. GG: adjusted OR = 2.51, 95% CI = 1.27-4.95, p = 0.008). We also found that the recessive model of the SNP rs3998860 was significantly associated with increased neuroblastoma risk (AA vs. GG/GA: adjusted OR = 2.69, 95% CI = 1.37-5.27, p = 0.004). Compared with the rs12781492 AA genotype, the rs12781492 CC genotype was significantly associated with

| Stratification analysis of significant SNPs
We further analysed the impact of TET1 gene polymorphisms on susceptibility to neuroblastoma by stratified analysis of age, sex, site of origin and clinical stage ( Table 2). Our results suggested that the

| Functional effect of rs3998860 G > A and rs12781492 A > C on surrounding genes
Based on the impact of TET1 rs3998860 G > A and rs12781492 A > C on susceptibility to neuroblastoma, we further explored the impact of these two loci on expression of nearby genes. We conducted cis-eQTL analysis on rs3998860 ( Figure 1) and rs12781492 ( Figure 2) using the GTEx database. The results showed that the rs3998860 A allele was significantly associated with increased STOX1 mRNA expression in the adrenal gland ( Figure 1A) and whole blood ( Figure 1B) compared to the G allele. In cultured fibroblasts ( Figure 1C), the A allele was significantly associated with increased KIF1BP mRNA expression compared to the rs3998860 G allele. In the tibial artery, the rs3998860 A allele was significantly associated with increased mRNA expression of SLC25A16 ( Figure 1D) and RUFY2 ( Figure 1E).
In skin not exposed to sun, the rs3998860 A allele was significantly associated with decreased CCAR1 mRNA expression ( Figure 1F).
In the adrenal gland ( Figure 2A) and whole blood ( Figure 2B), the rs12781492 C allele was significantly associated with increased STOX1 mRNA expression compared to the rs12781492 A allele. The rs12781492 C allele significantly enhanced mRNA expression of CCAR1 in the oesophageal mucosa ( Figure 2C) and skin not exposed to sun ( Figure 2D) compared to the A allele.

| Relationship between STOX1 mRNA expression and prognosis of neuroblastoma
Based on the significant impact of rs3998860 and rs12781492 on STOX1 mRNA expression in the adrenal gland and whole blood, TA B L E 1 Association of TET1 gene polymorphisms with neuroblastoma risk in children from Jiangsu province.  was significantly (p = 0.016) associated with low EFS ( Figure 3B).

| DISCUSS ION
Neuroblastoma is the most common tumour in children, and its high-risk type has a high risk of recurrence even after multiple treatments. 3,[9][10][11] Fundamentally understanding the genetic variations associated with neuroblastoma is key to early diagnosis and targeted treatment. Based on the important role of m5C modification in cancer, we focus on polymorphisms of m5C modification core genes. Our previous study showed that rs13181449 C > T in the m5C methyltransferase gene NSUN2 confers reduced risk of neuroblastoma. 46 In recent years, the m5C demethylase gene TET1 has been proven to mediate the occurrence and development of some cancers. [41][42][43][44] However, the important role of TET1 gene polymorphisms in cancer, including neuroblastoma, has not been revealed.
Therefore, we explored the impact of TET1 gene polymorphisms on risk of neuroblastoma through a case-control study. Our results suggest that TET1 gene polymorphisms (rs3998860 G > A and rs12781492 A > C) are associated with increased neuroblastoma risk.
We used a case-control study to explore the association between risk of neuroblastoma and TET1 gene polymorphisms in To further explore the functions of rs3998860 and rs12781492, we used the GTEx database to analyse their impact on nearby gene expression. Importantly, we found that both the rs3998860 A allele and the rs12781492 C allele were significantly associated with increased STOX1 mRNA expression in the adrenal gland and whole blood. Kaplan-Meier analysis showed that high mRNA expression of STOX1 was associated with poor prognosis. Interestingly, our findings precisely indicate that the rs3998860 A allele and the rs12781492 C allele are associated with increased neuroblastoma risk. The STOX1 (storkhead Box 1) gene has been implicated in preeclampsia. 47 Expression of the largest isoform of STOX1 (STOX1A) activates the PI3K-Akt-FOX pathway in the nucleus and inhibits the pathway in the cytoplasm. 47  Our study is the first to clarify the association between TET1 gene polymorphisms and susceptibility to neuroblastoma in children in Jiangsu, China, and its possible mechanism. Our sample size was relatively large, and comparability between the case group and the control group was high. Nevertheless, there are shortcomings in this study. First, the subjects included were only recruited from one hospital in Nanjing, China, limiting the extension of the conclusion.
Second, the predicted functional SNPs examined may be biased, and functional SNPs that affect neuroblastoma may be missed. Finally, the specific mechanism of TET1 rs3998860 and rs12781492 in neuroblastoma remains to be elucidated.
In summary, our study demonstrates that TET1 gene rs3998860 G > A and rs12781492 A > C significantly increase neuroblastoma risk. The potential mechanisms of TET1 gene polymorphisms in neuroblastoma need to be further elucidated. funding acquisition (equal); investigation (equal); methodology (equal); supervision (equal); writing -review and editing (equal).

ACK N O WLE D G E M ENTS
This study was supported by grants from the National Natural

CO N FLI C T O F I NTE R E S T S TATE M E NT
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
All the data are available upon request from the corresponding authors.