AURKA gene polymorphisms and central nervous system tumor susceptibility in Chinese children

Introduction Central nervous system (CNS) tumors comprise 15–20% of all malignancies occurring in childhood and adolescence. Previous researches have shown that overexpression and amplification of the AURKA gene could induce multiple human malignancies, with which the connection of CNS tumor susceptibility has not been extensively studied. Material and methods In this study, we assessed whether and to what extent AURKA gene single nucleotide polymorphisms (SNPs) (rs1047972 C > T, rs2273535 T > A, rs8173 G > C) were associated with CNS tumor susceptibility, based on a case–control analysis in 191 CNS tumor patients and 248 controls. We determined this correlation using odds ratios (ORs) and 95% confidence intervals (CIs). Results AURKA gene rs8173 G > C exhibited a crucial function to CNS tumor susceptibility fall-off (GC/CC vs. GG: adjusted OR = 0.68, 95% CI = 0.46–0.998, P = 0.049). In addition, the combined effect of lowering the risk of developing CNS tumors was more pronounced in carriers with 3 protective genotypes than others (adjusted OR = 0.55, 95% CI = 0.31–0.98, P = 0.044). Further stratification analysis illustrated that the existence of rs8173 GC/CC and three protective genotypes lowered CNS tumor risk in some subgroups. Conclusions Our research suggested that the AURKA gene rs8173 G > C could significantly reduce CNS tumor susceptibility in Chinese children. More functional experiments are needed to explore the role of the AURKA gene rs8173 G > C. Supplementary Information The online version contains supplementary material available at 10.1007/s12672-021-00459-w.


Sample selection
The cases and controls were frequency-matched for age, gender and ethnicity. They were enrolled between 2005 and 2019 from Guangzhou Women and Children's Medical Center, the Second Affiliated Hospital, and Yuying Children's Hospital of Wenzhou Medical University. Every patient in the case group was confirmed with CNS tumors. The inclusion criteria for CNS tumors patients were biopsy verified or histologically confirmed glioma. The biopsy requires ample tumor tissue derived from the contrast-enhancement margin of the lesion instead of necrotic core and subsequently histological diagnose [30]. Histologically classification is served as the gold standard to determine CNS tumors. Briefly, CNS tumors are diagnosed histologically by WHO classification. Grade I CNS tumors grow slowly and have well-demarcated tumor cells. Grade II CNS tumors also develop slowly but usually exhibit brain-invasive growth, which cannot be resected entirely. Grade III CNS tumors are characterized by anaplasia histologically, showing rapid growth, particularly high cellularity, cellular pleomorphism, and elevated mitotic activity. Grade IV CNS tumors are glioblastomas, microscopically different from grade III CNS tumors, which display microvascular proliferation and necrosis [11]. The control group with no history of malignancy was derived from those undergoing conventional check-up during the same period. The frequency distribution of chosen variables between the case and controls was listed in Additional file 1:

Polymorphism selection and genotyping
We selected single nucleotide polymorphisms based on the established criteria in the previous studies [31,32]. In general, the underlying SNPs were chosen in the 5'-flanking region, 3'-and 5'-untranslated regions, and the exons of the AURKA gene. Finally, we decided rs1047972 C > T, rs2273535 T > A, and rs8173 G > C in the AURKA gene to investigate their association with CNS tumor susceptibility in Chinese children. Using the TIANamp Blood DNA Kit (TianGen Biotech, Beijing, China), genomic DNA was isolated from the peripheral blood specimens of participants. As mentioned before [31,32], the TaqMan real-time PCR was utilized in genotyping through a 7900 HT sequence detection system (Applied Biosystem, Foster City, CA).

Statistical analysis
The differences between genotypes frequencies in the controls and Hardy-Weinberg equilibrium (HWE) were measured via goodness-of-fit χ 2 test. The two-sided χ 2 test was conducted to compare the discrepancies in the distributions of participants' selected variables. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using logistic regression analysis. Association of AURKA gene polymorphisms with CNS tumor risk was evaluated using ORs and 95% CIs adjusted for age and gender. P-value was less than 0.05, implying statistically significant. We performed the combined effect of genotypes using logistic regression analysis, subsequently evaluated the impact on CNS tumor risk by agegender adjusted OR, 95% CI, and P-value. The statistical analysis was staged on SAS v10.0 (SAS Institute, Cary, NC, USA).

Effect of AURKA gene SNPs on CNS tumor risk
The chosen AURKA SNPs in our research are all located on chromosome 20. The rs1047972 is localized in exon 3: 56, 386, 407, in which the nucleotide substitution usually exists between T and C. The rs2273535 is localized in exon 3: 56, 386, and its nucleotide substitution is A/C, A/G, and A/T. The rs8173 is localized in exon 9: 56, 369, 735, and the nucleotide substitution is C/G and C/T. 191 cases and 248 controls were successfully genotyped to investigate the impact of rs1047972, rs2273535, and rs8173 on CNS tumor risk in this study. All genotype frequency distributions of these SNPs conformed Vol:.(1234567890)

Research
Discover Oncology (2021) 12:62 | https://doi.org/10.1007/s12672-021-00459-w 1 3 to Hardy-Weinberg equilibrium (HWE = 0.567 for rs1047972; HWE = 0.328 for rs2273535; HWE = 0.332 for rs8173). Table 1 included detailed results of these three SNPs' genotype frequencies and their association with CNS tumor susceptibility. Of note, rs8173 G > C exhibited a prominently correlation with decreased CNS tumor risk under the dominant model (GC/CC vs. GG: adjusted OR = 0.68, 95% CI = 0.46-0.998, P = 0.049), which means the effect of GC/CC genotypes was more protective than GG genotype after the C allele was regarded as mutant type. No significant associations with CNS tumor risk were observed in the rest of SNPs (rs1047972 C > T, rs2273535 T > A). The combined effect of protective genotypes showed that participants with 3 protective genotypes had a 0.55-fold (adjusted OR = 0.55, 95% CI = 0.31-0.98, P = 0.044) reduction in the development of CNS tumors. Subjects with 1, 2 protective genotypes could not impact CNS tumor risk. A similar negative result was detected for 3 protective genotypes in comparison to 0-2 protective genotypes.

Stratification analysis
Our stratified analysis discovered the relationship between rs8173 G > C polymorphism and CNS tumor susceptibility based on age, gender, subtypes, and clinical stage ( Table 2). In comparison to GG genotype carriers, GC/CC genotype carriers were less susceptible to CNS tumors in males (adjusted OR = 0.47, 95% CI = 0.28-0.80, P = 0.005) and clinical

Discussion
This case-control study was performed to analyze the relationship between AURKA gene SNPs and CNS tumor risk in 191 patients and 248 cancer-free controls. Vast documents have indicated that AURKA ensured normal mitosis by maintaining centrosomes and spindle poles' stability. If the AURKA gene was abnormally expressed, the process was defective, promoting the transformation and progression of several human malignancies. Thus, amplification and overexpression of AURKA were strongly associated with several tumor risks [33][34][35], particularly evident in certain gene polymorphisms. For example, Wang et al. indicated that genetic variations in AURKA played a predictive role in the early clinical status of hepatocellular carcinoma (HCC). The HCC development was more affected by AURKA gene rs1047972 (V57I) polymorphism (C > T) than other SNPs. The rs1047972 variant TT genotypes conferred higher HCC risk than CC genotypes [25]. Gastric cancer (GC) was influenced by AURKA rs1047972. The rs1047972 TT and CC + CT genotypes contributed to elevated GC development risk, especially intestinal, diffuse, and both types of GC, reported by Mesic et al. in 2016 [36]. The AURKA gene rs2273535 (T91A) comes from nucleotide position 91, boosting the transformation of I1e to Phe. The research by Lee et al. revealed that the AURKA rs2273535 variant was susceptible to oral cancer. Additionally, if this SNP were exposed to alcohol, betel quid, and cigarettes, it would combine with them, known as gene-environmental interaction, which could produce synergistic effects to increase oral cancer risk [37]. Furthermore, the correlation between AURKA overexpression and GC poor prognosis was identified by Zhou et al. The AURKA rs2273535 TT genotypes increased higher risk of GC tumorigenesis than AA genotypes [38]. These two polymorphisms (rs1047972 G > A, rs2273535 T > A) were detected in exon 3 of the AURKA gene. It has been studied that rs1047972 variants modified rs2273535 polymorphism function and protein secondary structure to reduce carcinogenesis risk. Of note, the protective effect of rs1047972 was obvious in Caucasians and breast cancer, while not among Asians [39]. In terms of literature on the AURKA gene rs8173 is rather lacking. Our previous research has reported that rs8173 G > C was conducive to decreasing children's Wilms tumor susceptibility, especially individuals with rs8173 GC/CC genotypes in subgroups of > 18 months, male, and clinical stages III + IV [40]. Regrettably, the AURKA polymorphisms were not associated with neuroblastoma risk by another previous study [41]. Other AURKA SNPs such as rs2064863 and rs6024836 have been found to be related to cancer susceptibility [42,43]. Thus, although the investigations of how these three AURKA SNPs impact tumor risk are still preliminary, the probability that these SNPs affect CNS tumors cannot be excluded. Given that AURKA SNPs are frequently observed in multiple human malignancies and the implication of AURKA SNPs in CNS tumors has not been investigated intensively. Therefore, we are the first to speculate that the susceptibility of CNS tumors may be affected by AURKA gene SNPs. Existing research demonstrated that Gli2/miR-124/AURKA axis might be the key to influencing AURKA's effect on glioma. Gli2 is a glioma-associated oncogene transcription factor. The target gene of Gli2 can mediate the Hedgehog signaling pathway to facilitate the occurrence of glioma [44]. Briefly, miR-124 is suppressed because the Gli2 combines with the upstream region of miR-124 transcriptional initiation locus excessively, whereas miR-124 existence is relevant to AURKA expression, giving rise to the anomaly of AURKA expression in glioma cells, which ultimately promotes cell proliferation and tumorigenesis [45,46]. The Gli2/miR-124/AURKA occupies an essential position during the growth and development of human glioma cells.
In this study, we have shed light on that AURKA rs8173 was associated with reduced CNS tumor risk. At the same time, there were no apparent correlations between other AURKA gene SNPs (rs1047972 and rs2273535) and CNS tumor susceptibility. Moreover, the combined effect of protective genotypes and stratification analysis suggested that only subjects with 3 protective genotypes reduced CNS tumor risk significantly. Besides, stratified analysis manifested carriers with AURKA gene rs8173 GC/CC were linked to decreased CNS tumor risk in the subgroup of male and clinical stages I + II diseases, with no findings in age and subtypes. In our study, although, AURKA rs1047972 revealed negative results for altering CNS tumor risk, AURKA rs1047972 variants were identified to change lung adenocarcinoma (LADC) development probably [47]. Moreover, the rs1047972 variants were associated with elevated gastric cancer (GC) risk and basal-like breast cancer [48,49]. With no association of AURKA rs2273535 with CNS tumor susceptibility, AURKA rs2273535 polymorphisms were associated with overall enhanced cancer susceptibility, particularly breast cancer [50]. Furthermore, the SNP was a favorable prognostic factor for urothelial carcinoma patients treated with alisertib (an AURKA inhibitor) [51]. In a previous study, the correlation between AURKA rs8173 and breast cancer risk was reported by Hong Shi et al. [52]. In addition, the rs8173 G > C was associated with reduced Wilms tumor risk [40]. This study also indicated the critical role of AURKA rs8173 in cancer modification.
However, serval deficiencies in our study deserve to point out. In the first place, what we analyzed was not a large sample size but a relatively small sample size, only containing 191 patients and 248 controls, which was insufficient to produce a convincing statistic. For the sake of reducing fortuitous events to the utmost, more samples are needed for verification. In the second place, we only tested three AURKA SNPs in the research. More potentially functional CNS tumor risk-associated SNPs in the AURKA gene and other genes should be investigated later. Of note, our previous study indicated that FTO gene SNPs are unlikely to have significant effects on CNS tumor risk [53]. In that study, we also included the samples of Guangzhou and Wenzhou. More SNPs in diverse genes are needed to genotype. Finally, the case-control study conducted in this study might have been biased because the genotype distribution of the general population might not be represented by the genotype distribution studied in the hospital.
In this research, AURKA gene rs8173 G > C polymorphism was identified to confer decreased CNS tumor susceptibility significantly in Chinese children. Continuous studies with a larger sample size are warranted to elucidate the effect of AURKA gene SNPs on CNS tumor risk.