A shared susceptibility locus in p53 for both gastric cancer and esophageal cancer in a northwestern Chinese population

Background Upper gastrointestinal cancers are the leading causes of cancer-related deaths in Northwest China and share many similarities in terms of histological type, risk factors and genetic variants. We hypothesized that shared common genetic SNPs among eight SNPs in the p53 pathway existed among Ningxia gastric cancer (GC) and esophageal cancer (EC) patients. Methods A total of 180 GC cases, 113 EC cases and 358 cancer-free control subjects from a high-incidence area for upper gastrointestinal cancers in Ningxia, China, were enrolled in this study. The genotyping of 8 SNPs was performed using PCR direct sequencing. P53 expression in GC and EC tissues was examined using the S-P immunohistochemical method. Multiple logistic regression analyses were used to estimate the association between genotypes and GC or EC risks. Kaplan-Meier and multivariate Cox regression analyses were carried out to evaluate the associations between genetic variants and overall survival. Result rs1042522 was a common genetic locus shared by both Ningxia GC and EC patients. Compared with the rs1042522 Pro allele, the rs1042522 Arg allele increased the GC risk by 1.810 times and the EC risk by 2.285 times. Additionally, patients who carried the rs1042522 Arg allele and who also smoked or consumed alcohol had an increased risk for GC and EC. Cox survival analysis showed that neither p53 nor rs1042522 had an effect on the prognosis of GC and EC patients. Conclusion rs1042522 was a common genetic locus responsible for susceptibility shared by both northwestern GC and EC Chinese patients. Tobacco smoking and alcohol drinking further enhanced the cancer risk in our study.


Abstract
Background Upper gastrointestinal cancers are the leading causes of cancer-related deaths in Northwest China and share many similarities in terms of histological type, risk factors and genetic variants. We hypothesized that shared common genetic SNPs among eight SNPs in the p53 pathway existed among Ningxia gastric cancer (GC) and esophageal cancer (EC) patients. Methods A total of 180 GC cases, 113 EC cases and 358 cancer-free control subjects from a high-incidence area for upper gastrointestinal cancers in Ningxia, China, were enrolled in this study. The genotyping of 8 SNPs was performed using PCR direct sequencing. P53 expression in GC and EC tissues was examined using the S-P immunohistochemical method.
Multiple logistic regression analyses were used to estimate the association between genotypes and GC or EC risks. Kaplan-Meier and multivariate Cox regression analyses were carried out to evaluate the associations between genetic variants and overall survival. Result rs1042522 was a common genetic locus shared by both Ningxia GC and EC patients. Compared with the rs1042522 Pro allele, the rs1042522 Arg allele increased the GC risk by 1.810 times and the EC risk by 2.285 times.
Additionally, patients who carried the rs1042522 Arg allele and who also smoked or consumed alcohol had an increased risk for GC and EC. Cox survival analysis showed that neither p53 nor rs1042522 had an effect on the prognosis of GC and EC patients. Conclusion rs1042522 was a common genetic locus responsible for susceptibility shared by both northwestern GC and EC Chinese patients. Tobacco smoking and alcohol drinking further enhanced the cancer risk in our study.

Background
Gastric cancer (GC) and esophageal cancer (EC) are two deadly malignancies worldwide. Both GC and EC have been prevalent in China for decades with approximately fifty percent of the world's GC-related or EC-related morbidity and mortality [1]. In  China. The Ningxia region, located in Northwest China, has had high incidence of GC and EC for decades, ranking first and sixth among all cancers in this region, respectively [2 , 3].
Due to the lack of highly effective biomarkers for early diagnosis and prognosis prediction, advanced stages of GC and EC are commonly present at the time of initial diagnosis. The age-standardized 5-year relative survival rates were only 27.4% and 21% in Chinese GC and EC patients, respectively [4]. Although many research works have been undertaken to elucidate the genetic factors for the etiology of GC and EC, the established genes or loci could only explain only a small part of cancer risk. In the remaining large portion of GC and EC patients, there has been a limited success in determining the genetic architecture of GC and EC. Some recent studies have focused on the investigating about whether GC and EC share common genetic SNPs. For example, Yao et al. [5] explored whether known 21 genetic susceptibility loci for EC were also important in the development of GC.
They finally demonstrated that three genetic variants at 10q23. 33 Demographic data, including the age at diagnosis, gender, alcohol drinking and cigarette smoking, were obtained from medical records. Subjects with previous cancer and previous chemotherapy or radiotherapy were excluded.
The 358 cancer-free control subjects matching the cases for gender (1:1) and age (±5 years) were randomly selected from a pool of healthy volunteers who came to the general health check-up center at the General Hospital during the same period.
The recruited subjects were confirmed to be free of contraindications regarding endoscopic and cytology examinations. Demographic and risk factor data were collected by brief interviews. The study was approved by the Ningxia Medical University Ethical Review Committee, and all the participants provided written informed consent before entry into this study according to the regulations.
DNA preparation and genotyping.
Genomic DNA was extracted from peripheral blood leukocytes of the participants by using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions. SNP genotyping was performed using an improved multiplex ligase detection reaction method (iMLDR, Genesky Bio-Tech Cod., Ltd., Shanghai, China). Eight SNPs, including rs1042522, rs2395655, rs3176320, rs3829963, rs3829964, rs4135234, rs730506 and rs762624, in p53 or CDKN1A were tested. The polymerase chain reaction (PCR) primer pairs used to amplify the eight SNPs were the same as previously described

Follow-up
All patients were followed up to confirm their living status or loss of follow-up by telephone or text messages. The GC and EC patients' data were followed up to January 24, 2019. Patients with one-call failure were contacted at least three times afterwards. Patients who could not be contacted after three attempts or who had wrong numbers were regarded as lost visits. Death was considered an event. Those who were alive, lost to follow-up or died from other diseases at the closing date were considered censored data. The survival data were available for 103 of 180 GC cases (57.2%) and 85 of 113 EC cases (75.2%). Overall survival was calculated from the operation time to the date of death or the last follow-up.

Immunohistochemical analysis
For further immunohistochemical analysis, 103 GC cases and 85 EC cases were retrieved from the 180 GC and 113 EC surgical patients mentioned above. All the specimens had been routinely formalin-fixed, paraffin-embedded and serially sectioned at 4 μm in thickness. p53 protein expression in GC and EC tissues was examined using the S-P immunohistochemical method. Briefly, after routine deparaffinization and hydration, slides were treated with 1% hydrogen dioxide and then heated in EDTA (pH 8.0) for antigen retrieval. Following blocking in 10% goat serum blocking, tissue sections were then incubated with mouse anti-Human p53 protein (Biogenex, San Raman, CA) at 4°C overnight. After rinsing, sections were subsequently incubated with goat anti-mouse biotin-conjugated IgG for 15 min and then with streptavidin-peroxidase conjugate for 15 min. The signal was developed with diaminobenzidine, and slides were counterstained with 5% hematoxylin. The brown signals located in the nucleus represent positive staining for P53 protein.
Sections without monoclonal antibody treatment were used as negative controls.
The staining was scored on a scale from 0 to III as follows: 0, less than 10% cells were stained; I, 10-25% cells were stained; II, 26-50% cells were stained; and III, >50% cells were stained. Scores I-III were classified as positive, while score 0 was negative.

Statistical analysis
Descriptive statistical analysis and unconditional logistic regression were performed by using the Statistical Package for the Social Sciences 17.0 (SPSS. Inc. Chicago IL, USA, Microsoft) and Statistical Analysis System software (version 8.01; SAS Institute, Cary, NC). Multiple logistic regression analyses were used to estimate the association between genotypes and GC or EC risks. Odds ratios (ORs) were adjusted for demographic variables, including age, gender, drinking and smoking. Crossover analyses were used to investigate the association between rs1042522 and GC or EC by cigarette smoking or alcohol drinking status.
Cumulative survival rates were calculated from the date of initial diagnosis of GC or EC to the date of death or last follow-up. The Kaplan-Meier estimator was used to plot the survival curves, and multivariate analysis was performed using a Cox regression. Statistical significance was set at P<0.05, and all statistical tests were two-sided.

Part1
Demographic data of the participants A total of 651 Han people, including 180 GC cases, 115 EC cases and 358 cancerfree controls, were recruited in this study. Demographic data, including age, gender, cigarette smoking and alcohol drinking, for the subjects recruited in this study are summarized in Additional file: Table S1. The percentages of males in GC (71.7%) and EC ( 75.2%) in the case group were higher than those in the control group (57.0%), which was the same in cigarette smoking and alcohol drinking, and there were significant differences. All of the indexes in the GC or EC group were higher than those in the control group (P<0.01).
Genotype and allele frequencies of the SNPs in the case-control study The genotype and allele frequencies in the case and control groups are shown in Additional file: Table S2. Single SNP analyses showed that both p53 rs1042522 in 17p13.1 and CDKN1A rs730506 in 6p2 were significantly associated with EC susceptibility. The rs1042522 Arg allele was associated with an increased risk of EC

Associations between SNPs and GC or EC
A multiple logistic regression model was used to estimate the association between the genotypes and the risk of GC or EC. The results are listed in Table 1 and Table   2, the results were adjusted by age, gender, cigarette smoking and alcohol drinking.
Only one of the eight SNPs, rs1042522 in 17p13. 1 The results are illustrated in Table 3 and Table 4. The results indicated that there was a significantly increased risk of GC among participants with the rs1042522 Arg allele who were also cigarette smokers or alcohol drinkers when compared to those non-cigarette smokers or non-alcohol drinkers who carried the rs1042522 Pro  Table S5. The multivariate Cox regression showed that both p53 and rs1042522 had no effect on GC and EC patients' prognosis, and smoking and N stage were associated with the prognosis in EC patients (Table 5, 6 and Figure 1I-L).

Discussion
In the present study, we focused on investigating eight SNPs, including rs1042522 in the p53 gene and seven polymorphisms of p21 W af1/Cip1 among Chinese Ningxia GC and EC patients and cancer-free controls.
First, it was showed that rs1042522 was the only common genetic locus shared by both Chinese Ningxia GC and EC patients among eight SNPs. Compared with the rs1042522 Pro allele, the rs1042522 Arg allele increased the GC risk by 1.810 times, and the EC risk by 2.285 times as well. A group of studies supported that rs1042522 Arg was a potential cancer genetic risk factor, such as in breast cancer in Turkish patients and cutaneous melanoma in Caucasians [22]. Cheng et al. [25] reported that the rs1042522 Arg allele increased the GC risk by 1.17-fold compared with the rs1042522 Pro allele in a meta-analysis with 7,444 GC cases and 9,984 controls among Eastern Asians. Our previous study also found that the rs1042522 Arg allele contributed to an elevated esophageal squamous cell carcinoma risk by 6.48 times among a northern Chinese population [26]. These two studies were consistent with our present study. The functional mechanism behind this may result from the changing functions of the two alleles of rs1042522 in different cell types. Pim et al.
[26] used an inducible switch system for expressing both rs1042522 Arg and rs1042522 Pro to investigate how the 2 forms of rs1042522 brought about a cessation of cell growth. They found that within the primary Saos-2 cells, the rs1042522 Arg allele increased the ability of p53 to locate to mitochondria, which induced apoptosis more efficiently than the rs1042522 Pro allele. However, Schneider-Stock, et al. [28] reported that in squamous cell carcinomas of the head and neck, the rs1042522 Arg allele was associated with insufficient or absent apoptosis, because it lacked the apoptosis-related protein activities, such as the coexpression of Fas and FasL or high expression of Bcl2 protein. Furthermore, Garima et al. [29] found that compared to the rs1042522 Pro allele, the rs1042522 Arg allele led to reduced expression of the cell proliferation inhibitor p21 and increased angiogenesis-mediating VEGF expression, promoting carcinoma growth through increased cell proliferation and vascularity in tumors arising in patients. Therefore, these changes in rs1042522 may affect the function of the p53 protein, attenuate the loss or alteration of p53 binding capacity to the targets and may induce aberrant cell amplification accompanying cellular transformation, which leads to the occurrence of different cancers [29].
Some studies have also shown that rs1042522 Arg allele represents a significant risk factor in the development of human papillomavirus (HPV) associated cancers.
Steven et al. [31]evaluated the role of rs1042522 polymorphism and HPV status on the initiation, progression, and development of cervical cancer. They found that there was a significantly higher odds of progression from squamous intraepithelial lesions (SIL) to cervical cancer with the rs1042522 Arg allele in HPV-positive white and East Asian individuals. Storey et al. [32] reported there was a marked overrepresentation of homozygous rs1042522 Arg when compared with heterozygous or homozygous rs1042522 Pro, making individuals homozygous for rs1042522 Arg 7 times more likely to develop HPV-associated cervical cancer than individuals having one or more rs1042522 Pro alleles. This could be explained by the studies from Storey et al. and Thomas et al. [32 , 33], who showed that rs1042522 Arg was significantly more susceptible to the degradation induced by HPV E6 protein than rs1042522 Pro. Therefore, the rs1042522 Arg allele represents a significant risk factor in the development of HPV-associated cancers.
However, several other studies reported that the rs1042522 Pro allele was associated with the risk of cancer. For instance, Venkat R et al. [34] reported that African Americans, but not Caucasians, with the rs1042522 Pro homozygote had significantly higher mortality and 2.15-fold colorectal adenocarcinoma risk than those with the rs1042522 Arg allele. A group of studies also reported that the Additionally, different environmental exposures or inadequate sample sizes may also be attributable to the inconsistent results.
GC and EC are multifactorial disorders. Some common risk factors for GC include tobacco smoking, alcohol consumption, foods preserved by salting, older age, male gender, race, family history, low physical activity, low fiber intake and radiation [37]. Several studies demonstrated that cigarette smoking and alcohol consumption were risk factors for GC and EC [38 , 39] and may also lead to death from other smoking-related complications. In the present study, the crossover analysis by cigarette smoking or alcohol drinking status indicated that rs1042522 Arg allele carriers together with cigarette smoking or alcohol drinking further increased GC and EC risk.
Second, we also estimated the association of P53 expression and rs1042522 with the prognosis of cancer patients. Although the differences were not statistically significant, the overall survival rates of both GC and EC patients with p53 positive expression were higher than those with p53 negative expression, which was consistent with Wang's and Li's [41 , 42] investigations among 2,013 renal cell carcinoma patients and 150 GC patients, respectively. However, although some groups also reported that rs1042522 had an effect on cancer survival [34 , 43], we did not find any relationship between this SNP and survival of GC or EC, which was consistent with the results from Hedayatizadeh-Omran et al. [44].

Conclusion
In conclusion, we comprehensively analyzed the associations between eight SNPs and their association with both GC and EC patients from Ningxia, China and found that rs1042522 was a common genetic locus shared by GC and EC patients.
Additionally, tobacco smoking and alcohol drinking could further enhance the risk of cancer susceptibility in our study. However, some limitations need to be addressed.      Figure 1 Immunohistochemistry staining of the P53 protein in GC and EC tissues and the relationship

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download. Supplementary_CJ_June19.doc