Genetic predisposition to papillary thyroid carcinoma is mediated by a long non-coding RNA TINCR enhancer polymorphism

Single nucleotide polymorphisms (SNPs) in the enhancer region have been demonstrated to confer to altered enhancer activities, aberrant gene expression, and cancer susceptibility. In this study, we aimed to examine the association between an SNP, rs8101923, within terminal differentiation-induced non-coding RNA (TINCR) and the risk of papillary thyroid carcinoma (PTC). Blood samples from 559 patients with PTC and 445 healthy individuals were collected. The rs8101923 was genotyped by using polymerase chain reaction-restriction fragment length polymorphism assay. The impact of the rs8101923 on TINCR expression and enhancer activity was evaluated by quantitative real-time PCR and dual-luciferase reporter assay. The binding of AP-2α to TINCR enhancer was determined by chromatin immunoprecipitation. The rs8101923 G allele was signicantly associated with a higher risk of PTC (adjusted OR = 1.37; 95% CI: 1.15–1.64). Mechanistically, the rs8101923 was related to increased transcriptional levels and enhancer activities (P < 0.05). Transcription factor AP-2α binds to the enhancer region of TINCR containing the rs8101923 locus, and promotes cell proliferation in PTC. These ndings suggest the rs8101923 as a risk factor in the pathogenesis of PTC, which provides evidence for explaining the mechanism of the rs8101923 risk allele predisposing to PTC. Our ndings conrmed this hypothesis. We found that the rs8101923 within the enhancer region of TINCR was associated with a 1.37-fold increased risk of PTC and the risk G allele correlated to a higher enhancer activity and TINCR expression levels. These ndings indicate the rs8101923 as an enhancer SNP that affects the occurrence of PTC by modulating the expression of TINCR.


Introduction
Thyroid cancer is the most common head and neck cancer. With the development of diagnostic technology, the incidence of thyroid cancer, especially papillary thyroid carcinoma (PTC), has increased signi cantly worldwide in the past few decades 1,2 . In 2020, the incidence of thyroid cancer ranks 9th among all cancers globally, with more than 580,000 cases 3 . A clinical study including 15,000 thyroid cancer patients from China showed that PTC is the most common histological subtype, accounting for about 99% of all cases 4 . It is evident that the occurrence of thyroid cancer is related to multiple risk factors, such as genetic variants, radiation exposure, and iodine intake 2,3,5 . However, the speci c molecular mechanism of its initiation and progression is not completely clear.
Long non-coding RNAs (lncRNAs), a type of non-coding transcripts with the length over 200 nucleotides, are most easily overlooked in research at the beginning due to its non-coding feature. Recent studies have shown that lncRNAs can perform a variety of functions in organisms, such as regulating chromatin function, modulating the assembly and function of membraneless nucleosomes, changing the stability and translation of cytoplasmic mRNA, and interfering with signal pathways 6 . Currently, lncRNAs and lncRNArelated single nucleotide polymorphisms (SNPs) have been proven to contribute to the susceptibility and aggressiveness of various cancers, including PTC 7-9 . Kim et al. reported that a lncRNA, LOC100507661, was highly expressed in human thyroid cancer, and the over-expression promoted proliferation, migration, and invasion of thyroid cancer cells 10 . Jendrzejewski et al. reported that a lncRNA, PTCSC3 (PTC susceptibility candidate gene 3), in uenced PTC predisposition and carcinogenesis by modulating S100A4 gene expression 11 . Gou et al. reported that a tumor-suppressive lncRNA, AB074169, inhibited cell proliferation by binding KHSRP (KH-type splicing regulatory protein) and increasing p21 expression in PTC 12 . Maruei-Milan et al. reported that SNPs within ANRIL (antisense non coding RNA in the INK4 locus) were associated not only with the risk but also with the severity of PTC 13 . More importantly, enhancer-related SNPs were found to confer to altered enhancer activities, aberrant expression, and cancer susceptibility 8,14−16 .
Terminal differentiation-induced non-coding RNA (TINCR), a lncRNA with a length of 3.7k bases, was rst discovered in 2012 with an important role in human epidermal differentiation 17 . Subsequently, TINCR has been found to be associated with the occurrence and progression of a variety of malignant tumors, such as gastric cancer, esophageal cancer, and breast cancer 18-20 . As mentioned above, SNPs in the enhancer region may play an essential role in the predisposition to PTC 8,14−16 . In this study, we hypothesized that SNPs within the enhancer region of TINCR may contribute to the risk of PTC. To test this hypothesis, we carried out a case-control study to evaluate the relationship between SNP within the enhancer region of TINCR and the susceptibility of PTC in a Chinese population.

Study subjects
The study included 559 patients with PTC and 445 healthy controls. The subjects were enrolled from the West China Hospital of Sichuan University and were all unrelated Han nationality. Participants' information such as age, gender, and TNM staging were recorded in detail. Tumor stages were determined according to American Joint Committee on Cancer. The healthy volunteers were selected from the same hospital during the same period as the PTC cases were recruited. All controls were frequency matched to PTC cases based on age and gender. Among them, individuals suffering from thyroid disease or other tumors were excluded.
Peripheral blood samples of all subjects were taken before undergoing any medical treatment such as surgery and radiotherapy. Furthermore, the fresh PTC clinical specimens and adjacent normal tissues were collected immediately after the operation and placed in liquid nitrogen for later use. Informed consent was obtained from all individuals included in the study. The research protocol was approved by the ethics committees of West China Second University Hospital, Sichuan University.

SNPs selection
We selected SNPs according to the following criteria: (1) SNPs within TINCR enhancer that are annotated by ensemble database (http://www.ensembl.org/index.html); (2) SNPs with minor allele frequency more than 10% in Han Chinese. Only one SNP, rs8101923, followed the criteria and was selected for further analysis.

DNA extraction and SNP genotyping
According to the manufacturer's protocol, we used TIANamp Genomic DNA Kit (TianGen Biotech Co. Ltd., Beijing, China) to extract genomic DNA from blood samples, and used NanoDrop spectrophotometer (ND-1000) to determine the purity and concentration of DNA. The DNA fragment containing the SNP site (i.e., rs8101923) was ampli ed by polymerase chain reaction (PCR). The ampli cation primers used in this study were: 5'-AAGGTTGTGTGTGGGAGAGG-3' (forward), 5'-CATGACCTCTGGGGTGTCTT-3' (reverse). The PCR system was set to 10 μl mix: 5 μl 2X Power Taq PCR MasterMix, 10 μM for each primer, 0.5 μl DNA template, and Milli-Q water as a supplement to 10 μl. The PCR reaction conditions were: 94°C for 4 min, 94°C for 30 s, 60°C for 30 s, 72°C for 30 s, 35 cycles, and 72°C for a nal extension of 10 minutes. Subsequently, the PCR product was digested by Sty I (New England BioLabs, MA, USA), and genotyped by PCR-restriction fragment length polymorphism assay. After digestion, the G allele produced a 216 bp band, and the A allele produced 123 bp and 93 bp bands. In order to verify the genotyping results, we randomly selected the samples for Sanger sequencing, and the agreement rate was 100%.
Quantitative real-time RT-PCR (qRT-PCR) TRIzol reagent (Invitrogen, Carlsbad, CA, USA) was used to extract total RNA from 43 pairs of PTC tissues and adjacent normal tissues, and RevertAid First Strand cDNA Synthesis Kit (Thermo Scienti c Scienti c, Rockford, IL, USA) was used to reverse transcribed total RNA into cDNA. LightCycler® 480 System (Roche Indianapolis, IN, USA) and QuantiNova™ SYBR® Green PCR Kit (Qiagen, Hilden, Germany) were used for determining the relative expression levels of TINCR. The TINCR primers used for qRT-PCR were described as follows 17 : forward 5'-TGTGGCCCAAACTCAGGGATACAT -3' and reverse 5'-AGATGACAGTGGCTGGAGTTGTCA-3'. 18s was used as an internal control and ampli ed with the following primers: forward 5'-GCAATTATTCCCCATGAACG-3' and reverse 5'-GGCCTCACTAAACCATCCAA-3'. Each ampli cation reaction was completed in a total volume of 10 μl, which contained 0.7 μl primers, 5 μl Master mix, and 100 ng cDNA. The reaction conditions were set as follows: 95°C for 2 min, 95°C for 5 s, and 60°C for 10 s, 40 cycles.
Cell culture, plasmid construction, and luciferase reporter assay TPC-1 and BCPAP cells were purchased from GuangZhou Jennio Biotech Co. Ltd. (Guangzhou, China). All cell lines were maintained in RPMI1640 medium supplemented with 10% fetal bovine serum and 1% ampicillin/streptomycin and cultured under 5% CO 2 . The TINCR fragment containing the rs8101923 AA or GG was ampli ed by PCR, and the ampli cation primers were forward 5'-AGATGACAGTGGCTGGAGTTGTCA-3' and reverse 5'-TGTGGCCCAAACTCAGGGATACAT-3'. The PCR products were inserted into pmirGLO miRNA target expression vector (Promega) to construct pmirGLO-rs8101923A and pmirGLO-rs8101923G plasmids.
For the luciferase reporter gene detection, we used the dual luciferase reporter gene detection system (Promega). Fire y and renilla luciferase activity was measured at 48 h after transfection according to the manufacturer's instructions.

Chromatin immunoprecipitation (ChIP)
Cells were grown to approximately 80% to 90% con uence, cross-linked with 1% paraformaldehyde at 37°C for 10 minutes, and quenched in 125 mM glycine. The cell lysates were sonicated under conditions that produced fragments of 200 bp to 1000 bp. The material was clari ed by centrifugation, diluted 10 times in dilution buffer, and pre-clari ed with protein A-Sepharose beads. The pre-cleared supernatant containing chromatin was used for the immunoprecipitation reaction with the antibody against AP-2α (Abcam, Cambridge, MA, USA). Anti-mouse IgG was used as a control. After digestion with proteinase K and RNase, the immunoprecipitated genomic DNA was extracted with phenol: chloroform: isoamyl alcohol and precipitated with ethanol. For gene-speci c ChIP analysis, PCR and Sanger sequencing were used to determine the DNA sequence enriched by immunoprecipitation.
Statistical analysis SPSS 20.0 statistical software (SPSS Inc, IL, USA) was used for data analysis. The rs8101923 genotype frequency was obtained by direct counting. The Student's t-test used to compare the age of cases and controls. The gender comparison and Hardy-Weinberg equilibrium (HWE) were calculated using Chi-square (χ 2 ) test. Results of qPCR were analyzed using Wilcoxon rank sum test. The unpaired Student's t-test was used to analyze the luciferase data. The difference of the rs8101923 genotype distribution between cases and controls was analyzed using the χ 2 check test. Odds ratio (OR) and 95% con dence interval (CI) were used to assess the correlation between the rs8101923 and PTC risk. P <0.05 was considered statistically signi cant.

Characteristics of the study population
Totally, 559 patients with PTC and 445 healthy individuals were included in this study. The age of the PTC group and the control group was 43.2 ± 9.4 and 44.5 ± 13.3 years, respectively; while the difference was not statistically signi cant (P = 0.08). The proportion of male subjects in the two groups was 29.7% and 24.7%, respectively, and the difference was not statistically signi cant (P = 0.08). Among PTC patients, there were 42 cases of Tx (7.5%), 173 cases of T1 and T2 (31%), and 344 cases of T3 and T4 (61.5%); there were 343 cases with lymph node metastasis (61.4%) and 216 cases without lymph node metastasis (38.6%); there were 11 patients (2%) who developed distant metastasis (Table 1).   The rs8101923 G increased TINCR levels through increasing enhancer activity qRT-PCR was used to examine the expression levels of TINCR in 43 pairs of clinical PTC tissue samples. We found that the expression of TINCR in PTC tissues was signi cantly increased compared to normal tissues (P < 0.001) (Fig. 1A). We then examined whether the rs8101923 had an allele-speci c effect on the expression of TINCR. As shown in Fig. 1B, the TINCR levels in PTC tissues with the rs8101923 GG genotype were signi cantly higher than those with the AA genotype (P < 0.05). Next, we used reporter gene analysis to evaluate whether the rs8101923 in uenced enhancer activity. As shown in Fig. 1C, TPC-1 and BCPAP cells transfected with the rs8101923 G allele presented a signi cantly increased enhancer activity compared to the rs8101923 A allele (P < 0.05).

AP-2α enriched the TINCR enhancer region containing the rs8101923 and promoted cell proliferation
Using the TRANSFAC® 7.0 software, we predicted that the transcription factor AP-2α may be enriched in the enhancer region of TINCR containing the SNP, rs8101923. We then used the CHIP assay to verify the binding site. As shown in Fig. 2A and 2B, AP-2α speci cally bound to DNA sequences in BCPAP cells and PTC-1 cells compared to the IgG control. Sanger sequencing con rmed that AP-2α biding site contained the rs8101923 ( Fig. 2C and 2D). Next, we examined whether knock-down of AP-2α in uenced proliferation in BCPAP and PTC-1 cells (Fig. 2E). As shown in Fig. 2F and 2G, cell viability was signi cantly reduced at 72h after transfection in both BCPAP cells (P < 0.01) and PTC-1 cells (P < 0.05). These ndings suggested that AP-2α enriched the enhancer region of TINCR containing the rs8101923, and promoted cell proliferation in PTC.

Discussion
Previous studies have demonstrated that various lncRNAs were differentially expressed in PTC, such as PTCSC3, PTCSC2, maternally expressed gene 3 (MEG3), and BRAF-activated non-protein coding RNA (BANCR), serving as tumor suppressors or oncogenic lncRNAs 8,21−24 . However, there is no report investigating the role of TINCR in the occurrence of PTC. In this study, we found for the rst time that TINCR was highly expressed in PTC tissues. Mechanistically, an SNP, rs8101923, within the enhancer region of TINCR was found to be associated with increased transcriptional levels and enhancer activities, and contributed to the risk of PTC in the Chinese population. Transcription factor AP-2α bound to the enhancer region of TINCR containing the SNP (i.e., rs8101923) and promoted cell proliferation in PTC. These ndings suggest the rs8101923 as a risk factor in the pathogenesis of PTC, which provides evidence for explaining the mechanism of the rs8101923 risk allele predisposing to PTC. Enhancers are cis-regulatory elements that can be bound by transcription factors to activate gene expression 42 . In this study, the transcription factor AP-2α was predicted to bind the enhancer region of TINCR containing the rs8101923 by using the TRANSFAC® 7.0 software. ChIP results con rmed the prediction. AP-2α was found to be highly expressed in PTC tissues, and the high expression of AP-2α mRNA was signi cantly correlated with advanced tumor stage and shorter overall survival 43 . Knock-down of AP-2α inhibited proliferation in PTC cells. Previous ndings together with our current work indicate that AP-2α may enrich the TINCR enhancer region containing the rs8101923, and promote cell proliferation.
There are some limitations in this study. All patients with PTC included in this study were collected from the same hospital for a continuous period of time, and thus the selection bias might be inevitable. There was no signi cant association between the rs8101923 and clinical features of PTC in strati ed analysis. This negative result may be caused by limited sample size and insu cient statistical power. What's more, we did not consider the patient's living environment, such as ionizing radiation and daily iodine intake, when collecting samples. The interaction of TINCR and environmental factors on the risk PTC cannot be assessed.

Conclusion
Page 11 /15 This study provides the rst evidence that the rs8101923 within the enhancer region of TINCR was a risk factor for the development of PTC in the Chinese population. Mechanistically, the rs8101923 might increase transcriptional levels and enhancer activities of TINCR. In the future, further research on the correlation between the rs8101923 and other thyroid tumors will help to understand the role of SNP in the treatment, diagnosis, and prognosis of thyroid cancer. Figure 1 The