Caspase-3 (CASP3) is the main executioner of apoptosis, mediating both extrinsic and intrinsic cell death signaling pathways, and is involved in tumor behaviors. In this study, we investigated the association of two regulatory variants in CASP3 and the risk of esophageal squamous cell carcinoma (ESCC) in 1026 cases and 1270 healthy controls. Odds ratios (OR) and 95% confidence intervals (CI) were computed by logistic regression. The function of the CASP3 829 A>C polymorphism was examined by luciferase reporter assay and real-time PCR. A significant increased risk of ESCC was found for the CASP3 829 AC and CC genotypes with OR (95% CI), 1.53 (1.26–1.89) and 1.42 (1.11–1.82), respectively. When stratified by age and gender, the risk of ESCC was more significant in younger (≤57 years) and male individuals. No significantly changed risk of ESCC was related to 20541 C>T variant. Luciferase reporter assay showed 829 A>C variant dramatically reduced the transcriptional activity of luciferase reporter gene by over 95% in both KYSE30 and KYSE450 esophageal cancer cells. Remarkably, the transcriptional activity of the 829C-containing construct was much lower than the activity of the pGL3-basic construct, with over 85% reduction in both cell lines. Real-time PCR analyses showed that 829 AA genotype carriers had significantly higher RNA levels (0.015 ± 0.00216, n = 24) than the 829 AC genotype carriers (0.00969 ± 0.00136, n = 36), and 829 CC genotype carriers (0.00663 ± 0.00097, n = 20). These findings suggest that CASP3 829 A>C polymorphism may highly affect the function of caspase-3 and play an important role in the development of ESCC in Chinese populations. (Cancer Sci 2012; 103: 555–560)
Esophageal cancer has high incidence and mortality rates, with a variable geographic distribution.(1) Nearly 50% of esophageal carcinoma cases worldwide occur in China.(2) Histologically, esophageal cancer is classified into squamous cell cancer and adenocarcinoma. Esophageal squamous cell carcinoma (ESCC), which is the most common histological subtype, accounts for approximately 90% of all esophageal cancers diagnosed in China each year.(3) The development of ESCC is a multifactor-mediated progressive process. Several environmental factors such as tobacco smoking, alcohol consumption, exposure to nitrosamines, and nutritional deficiencies(4–8) have been reported to increase the risk of ESCC. However, individuals exposed to these risk factors have different susceptibilities to ESCC, indicating that genetic factors also play an important role in development of the disease.
Genetic variants in the MDM2, COX-2, GSTP1, survivin, FAS, XRCC1, c1orf10, and S100A14(9–15) genes have been shown to be associate with susceptibility to ESCC. Recently, three genome-wide association studies of esophageal cancer also identified several variants that strongly confer their risks to ESCC.(16–18) These studies highlighted the key role of genetic factors in the development of ESCC. Caspases play a crucial role in the initiation and execution of apoptosis, and the dysregulation of apoptosis contributes to tumor development and progression.(19) Caspases have been divided into initiator and effector caspases. The initiator caspases (caspase-2, -8, -9, and -10) activate the effector caspases (caspase-3, -6, and -7), which in turn cleave multiple substrates including the structural and regulatory proteins in the cell nucleus, cytoplasm, and cytoskeleton.(20) In apoptotic cells, caspase-3 is the main effector as it can be activated through both extrinsic and intrinsic cell death pathways.(21)
There are an increasing number of studies on the genetic variants of caspase genes and the risk of various cancers. It has been shown that the caspase-8 promoter region six-nucleotide deletion/insertion and a coding region D302H polymorphism could cause dysregulated apoptosis and thus carcinogenesis, and is associated with multiple cancer risk, including esophageal cancer.(22–26) Caspase-3 plays a critical role in the apoptosis machinery and has been reported to harbor somatic mutations in multiple cancers.(27) The polymorphisms of the caspase-3 (CASP3) gene have also been explored by direct sequencing in a Korean group of subjects, and the −928 A>G, 77 G>A, and 17532 A>C polymorphisms were associated with a significantly decreased risk of lung cancer.(28) In a Han Chinese population, Xie et al. selected 261 DNA samples from healthy subjects to search for genetic variants within the regulatory region and coding region of CASP3. They identified three single nucleotide polymorphisms (SNPs), 829 A>C, 17532 A>C, and 20541 C>T, in the 5′-regulatory region, intron 4, and 3′-regulatory region.(29) They also found that the genetic variants 17532 A>C and 20541 C>T were in complete linkage disequilibrium (LD).(29) To further illustrate the effect of CASP3 829 A>C (rs4647602) and 20541 C>T (rs1049216) variants on the susceptibility to esophageal cancer and gene expression, we carried out a case–control study in a Chinese group of subjects.
Materials and Methods
Study subjects. Recruitment of study subjects was described previously.(14) Briefly, this study included 1026 patients with histologically confirmed ESCC and 1270 cancer-free controls. Patients were consecutively recruited between July 1999 and July 2003 at the Cancer Hospital of the Chinese Academy of Medical Sciences (Beijing, China). The response rate for patients was 92%. All subjects were unrelated ethnic Han Chinese that had not received any radiotherapy or chemotherapy at the time of recruitment and blood donation. Controls were cancer-free individuals randomly selected from a community cancer-screening program for early detection of cancer carried out in the same regions during the same period the patients were recruited. The response rate for controls was 89%. The selection criteria for the controls included no individual history of cancer and frequency-matched to case patients based on age (±5 years), sex, and ethnicity. At recruitment, informed consent was obtained from each subject and personal demographic characteristics such as sex, age, and related risk factors including tobacco smoking were collected through a questionnaire. Subjects were considered smokers if they smoked up to 1 year prior to the date of diagnosis for cancer patients or the date of interview for controls, otherwise they were defined as non-smokers. Alcohol drinkers were defined as subjects who had participated in drinking any type of alcohol (wine, beer, herb liquors, other spirits) in the last 12 months. This study was approved by the institutional review board of the Chinese Academy of Medical Sciences Cancer Institute (Beijing, China).
Single nucleotide polymorphism genotyping. Genomic DNA was extracted from peripheral blood lymphocytes of all controls and most patients. Of all DNA samples from patients, 28% were isolated from surgically removed normal tissues adjacent to the tumors. Genotypes were determined by PCR-RFLP. The primers were designed using the SNP Cutter program.(30) Briefly, the PCR primer pairs used to amplify the CASP3 regulatory region containing the 829 A>C (rs4647602) site was 5′-TAG TTG CAG GGT TTA AAC TCC AAT GC-3′ and 5′-CTA ACT CCT CAC GGC CTG GGA T-3′. The underlined nucleotide is an introduced mismatch from T to G at the forward primer, so that the nearby 829 A>C polymorphism can be recognized by restriction enzyme BglI. The primer pairs used to amplify the CASP3 3′-regulatory region containing site 20541 C>T (rs1049216) was 5′-GTG AAA AAG TTA AAC ATT GAA TTA A-3′ and 5′-TTC TTC CAC ATC ATC ATT TCT A-3′. The underlined nucleotide is an introduced mismatch from G to T at the forward primer, so that the nearby 20541 C>T polymorphism can be recognized by restriction enzyme AseI. Polymerase chain reaction was carried out using a 25 μL reaction mixture containing 100 ng DNA, 0.1 μmol/L each primer, 0.2 mmol/L deoxynucleoside triphosphate, and 1.0 U Taq DNA polymerase (Promega, Madison, WI, USA). The PCR profile consisted of an initial melting step of 95°C for 4 min, followed by 35 cycles 94°C for 30 s, 60°C for 30 s, 72°C for 30 s, and a final extension step of 72°C for 7 min. The amplified PCR products for 829 A>C and 20541 C>T were 137 bp and 103 bp, respectively. Restriction enzymes BglI and AseI (New England Biolabs, Beverly, MA, USA) were used to delineate the SNPs, which resulted in 112 bp and 25 bp fragments and 82 bp and 21 bp fragments, respectively. Genotyping was carried out without knowledge of the case/control status of the subjects. The genotype methods were confirmed by sequencing. Ten percent of the samples were randomly selected for repeated assays, and the results were 100% consistent.
Plasmid construction. To explore whether the 829 A>C SNP influences the transcriptional activity of CASP3, we constructed a reporter plasmid encompassing the base pairs of human CASP3 promoter. To amplify this DNA fragment, the Prime 3 program was used to design the primers(31) and the sequences were 5′-ata cGC TAG CTA CCC AGT GAC CAG CAA GTG-3′ and 5′-gat aAA GCT TGG TGG CAA AAC AAA CAC TCC-3′, which contain NheI and HindIII cloning sites (underlined sequences), respectively. To ensure high-fidelity amplification, LA Taq polymerase (TaKaRa, Dalian, China) was used. The PCR product was digested with NheI and HindIII and ligated into an appropriately digested pGL3-basic vector (Promega) containing the firefly luciferase gene as a reporter. To produce the luciferase construct containing the 829A allele, a pair of primers 5′-GGT TTA AAC TCC AAT TCA TTT TCG GCC C-3′ and 5′-GAA TTG GAG TTT AAA CCC TGC AAC TAT CTC-3′ was used to make the single site mutagenesis (Invitrogen, Carlsbad, CA, USA). All constructs used in this study were sequenced to confirm their authenticity.
Luciferase assay. Promoter activity was measured using the Lucifierase Reporter Assay System (Promega).(14,32,33) KYSE30 and KYSE450 cells were grown in RPMI-1640 supplemented with 10% FBS (Invitrogen) in a humidified, 5% CO2 incubator at 37°C. Cells (1 × 104) were plated in a 96-well plate and grown to 80–90% confluence. The firefly luciferase constructs containing the 829A or 829C allele as well as pGL3-basic vector (100 ng) were transfected into KYSE30 and KYSE450 cells using Lipofectamine 2000 reagent (Invitrogen). To monitor transfection efficiency, cells were cotransfected with 1 ng pRL-SV40 plasmid, which encodes Renilla luciferase. Luminescence was measured 24 h after transfection using a dual-luciferase reporter assay system (Promega). All transfections were carried out in triplicate, and data were analyzed by normalizing firefly luciferase activity to Renilla luciferase activity for each sample.
Real-time analysis of CASP3 RNA. Total RNA was isolated from surgically removed normal esophageal tissues adjacent to the tumors of 80 individual patients then converted to cDNA using the SuperScript II First-Strand Synthesis System (Invitrogen). Caspase-3 RNA was measured by real-time quantitative PCR by the ABI 7900 HT Real-time System (Applied Biosystems, Carlsbad, CA, USA) based on the SYBR-Green method. Each assay was done in triplicate, and the expression of individual CASP3 RNA measurements was calculated relative to expression of β-actin using a modification of the method described by Lehmann and Kreipe.(34) The primers and probes used for detection of CASP3 RNA are available on request.
Statistical analysis. We used Student’s t-test to compare the means of age and the χ2-test to compare the frequency distributions between cases and controls. Odds ratios (OR) and 95% confidence intervals (CI) were used to evaluate the risk of ESCC using multivariate logistic regression analysis with adjustment for possible confounding factors (age, sex, and smoking). All statistical tests were two-sided, and a P-value of <0.05 was considered significant using Statistical Analysis System software (version 8.2; SAS Institute, Cary, NC, USA). Pairwise LD was measured with D′ and r2 in Haploview.(35)
Subject characteristics. The characteristics of the study group are shown in Table 1. The mean age (±SD) was 58.2 ± 9.5 years for the controls and 59.0 ± 9.9 years for the patients. No significant difference was found in the mean age between cases and controls (P = 0.057). The median age was 57 years (range, 26–84 years) for the controls and 59 years (range, 25–89 years) for the controls (P = 0.899). There was no significant difference in sex distribution between 1026 cases and 1270 controls (P = 0.526). However, the cases included more smokers than the controls (61.9%vs 54.2%, P < 0.0001). Among our study subjects, only 792 controls and 1025 cases had alcohol-drinking data. There were more drinkers in the cases than in the controls (46.3%vs 31.4%, P = 0.004). The patients with tumor grade I, II, III, and IV accounted for 10.0%, 34.1%, 41.8%, and 4.8% in the case group, respectively.
Table 1. Demographic characteristics of study groups including patients with esophageal squamous cell carcinoma (cases) and controls
Cases (n = 1026)
Controls (n = 1270)
†Two-sided χ2-test. NA, not applicable.
Caspase-3 variants and risk of developing ESCC. The genotype distributions of CASP3 829 A>C and 20541 C>T SNPs in the cases and controls are summarized in Table 2. The observed genotype frequencies of 829 A>C and 20541 C>T were in Hardy–Weinberg equilibrium in the controls (P = 0.997 and P = 0.809, respectively). The 829 C>A variant in the 5′-UTR of caspase-3 was separated from 3′-UTR 20541 C>T variant by >9.7 kb. The LD was not strong (D′ = 0.768, r2 = 0.16).
Table 2. Caspase-3 genotype frequencies and association with risk of esophageal squamous cell carcinoma
Controls (n = 1270)
Cases (n = 1026)
OR (95% CI)†
†Adjusted by age, sex, and smoking. CI, confidence interval; OR, odds ratio; Ref, reference.
Compared with the carriers with the 829 AA genotype, the carriers with 829 AC or 829 CC genotypes had significantly increased risk of ESCC with OR (95% CI) of 1.53 (1.26–1.89) and 1.42 (1.11–1.82), respectively. No significant changed risk of ESCC was found to relate to the 20541 C>T SNP. The OR (95% CI) was 1.15 (0.96–1.37) and 0.64 (0.39–1.05) for the carriers with 20541 CT and 20541 TT genotypes, respectively.
To identify any susceptible subgroup, we focused on stratified analysis of associations between CASP3 829 A>C and risk of ESCC by selected variables listed in Table 1. As shown in Table 3, the significantly increased risk associated with CASP3 829 AC or CC genotype was evident in the subgroups of younger subjects, with OR (95% CI) of 1.63 (1.23–2.15) and 1.67 (1.15–2.41), respectively, and also in males, with OR (95% CI) of 1.58 (1.28–1.96) and 1.37 (1.03–1.81), respectively, compared with the common homozygous 829 CC genotype (Table 3). However, we did not find an association between increased risk of ESCC and tobacco smoking, drinking (Table 3), or tumor grade (data not shown). Also, there was no evidence for possible interaction between the 20541 C>T variant and these selected variables (data not shown).
Table 3. Caspase-3 829A>C genotype frequencies in cases of esophageal squamous cell carcinoma and controls, stratified by age and sex
829 A>C Genotype
Controls (n = 1270)
Cases (n = 1026)
OR (95% CI)†
†Adjusted by other covariates (excluded drinking). CI, confidence interval; OR, odds ratio; Ref, reference.
Age ≤57 years
Age >57 years
Effects of 829 A>C SNP on caspase-3 promoter activity. In order to investigate the effect of promoter SNP 829 A>C on the transcriptional activity of CASP3, we created promoter luciferase constructs containing 829A or 829C alleles and dual luciferase assay was carried out in esophageal cancer cell lines, KYSE30 and KYSE450. As shown in Fig. 1, the CASP3 829 A>C variant dramatically reduced the transcriptional activity of luciferase reporter gene to 3.6% in KYSE30 cells and 3.9% in KYSE450 cells. Remarkably, the transcriptional activity of the 829C-containing construct was much lower than the activity of the pGL3-basic construct, with a reduction of more than 85% in both cell lines. These results suggest that the CASP3 829 A>C variant may dramatically enhance the binding of some transcriptional corepressors to the regulatory region of CASP3.
Effects of caspase-3 SNPs on caspase-3 RNA expression. We also evaluated the association between CASP3 829 A>C genotype and CASP3 RNA levels in surgically removed normal lung tissues adjacent to the tumors of 80 esophageal cancer patients. Real-time RT-PCR analyses show that 829 AA genotype carriers had significantly higher RNA levels (0.015 ± 0.00216, n = 24) than the 829 AC genotype carriers (0.00969 ± 0.00136, n = 36), and 829 CC genotype carriers (0.00663 ± 0.00097, n = 20) (Fig. 2). Using the R program, we further calculated the power of this sample size. When given at least 20 samples (the smallest group in our study) for each group, the detection power would be 0.867. These results further suggested that the 829 A>C change is associated with decreased caspase-3 transcription.
In this study, we investigated the association of CASP3 829 A>C and 20541 C>T SNPs located in the 5′- and 3′-regulatory regions of CASP3 with the risk of ESCC in a case–control study in Chinese subjects. We found that the 829 CC genotype highly reduced the transcriptional activity of caspase-3 and was associated with an increased risk of ESCC. Also, the risk was more evident in the subgroups of younger (≤57 years) and male subjects.
Apoptosis is a fundamental biological process, the dysfunction of which has been considered to be one of the key events in the development of cancer.(36) The process of apoptosis is controlled by activation of multiple caspases, which act as apoptosis initiators and effectors. Investigators have shown that caspases are at the crossroads of immune-cell life and death and the reduced expression of caspases is associated with cancer development.(37) Caspase-3 is the crucial executioner of apoptosis as it can be frequently activated through both intrinsic receptor-mediated and extrinsic mitochondrial pathways. By catalyzing the specific cleavage of a broad spectrum of cellular proteins, caspase-3 degrades the cellular target to trigger cell apoptosis.(21,38,39) Xie et al.(29) screened the regulatory region of the caspase-3 gene and identified one SNP (rs4647602 A>C) in the 5′-UTR and one SNP (rs1049216 C>T) in the 3′-UTR in a group of Chinese subjects. It is reasonable that these genetic variants in the caspase-3 regulatory region affect an individual’s susceptibility to certain cancers.
There are a few studies about the mutations of CASP3 and cancer risk. One case–control study investigated the role of two other CASP3 SNPs in the 3′-UTR region (rs6948:Ex8-280 C>A and rs1049216:Ex8-567 T>C) in non-Hodgkin’s lymphoma susceptibility, and their results showed that these two variants in CASP3 were significantly associated with a decreased risk for non-Hodgkin’s lymphoma.(40) Another study showed that CASP3 polymorphisms (−928 A>G, 77 G>A, and 17532 A>C) contribute to genetic susceptibility to lung cancer.(28) Our present study showed that functional CASP3 829 A>C polymorphism increased the risk of ESCC. This is consistent with our results of functional assay at the transcriptional level and RNA expression level. Chen et al.(41) carried out an association study on three SNPs in the caspase-3 gene, rs4647601 G>T, rs4647602 C>A, and rs4647603 G>A to examine the relation between the gene and head and neck squamous cell carcinoma in Caucasians. They found a significant association with rs4647601, but not with the two others. However, their haplotype-based analysis revealed that haplotype Trs4647601Crs4647602Grs4647603 and haplotype Grs4647601Crs4647602Ars4647603 had significant increased cancer risk, suggesting that rs4647602 C is a causal variation of the relation. This finding supported the possibility that SNP rs4647602 is functional.
Most of the mutations in caspases result in decreased caspase activity and thus the attenuation and deficiency of autoimmune lymphocyte apoptosis.(25,42,43) Sun et al.(25) have shown that CASP8 6N del variant in the promoter destroys a stimulatory protein 1 binding site and decreases caspase-8 transcription. This genetic variant has a protective effect on multiple cancers. They further showed that T lymphocytes with the deletion variant had lower activation-induced cell death after stimulation with cancer cell antigens. Activation-induced cell death of T lymphocytes may help malignant cells to escape from cytotoxic T lymphocyte killing and therefore confer susceptibility to cancer by influencing immune surveillance.(44–46) Opposite to the previous study(25) showing that individuals with lower caspase-8 activity were less susceptible to cancers, we found that individuals harboring the variant allele C of CASP3 promoter SNP 829 A>C had reduced CASP3 transcriptional activity and increased susceptibility to ESCC. This may be due to the variant allele C leading to decreased CASP3 transcriptional activity and gene expression, which might inhibit the apoptosis of tumor cells and thus be associated with an increased risk of ESCC. However, further studies are needed to elucidate the function of this polymorphism in precarcinogenic epithelial cells.
Evasion of apoptosis is the hallmark of carcinogenesis; the acquired ability to resist apoptotic stimuli is a common characteristic of tumor cells. Our luciferase assay showed that variant C allele of CASP3 promoter SNP 829 A>C significantly repressed caspase-3 transcriptional activity, which explains the aberrations in apoptotic pathways in cancer patients. Our epidemiological findings are consistent with the results of functional analyses, indicating a significant enhanced susceptibility of people with variant allele C to ESCC.
We found a significantly increased risk of ESCC in younger subjects, which suggested the promoter SNP 829 A>C may be a genetic marker for susceptibility to ESCC in Chinese people due to an early age of cancer onset. It has been found that gender plays a role in genetic susceptibility to lung cancer. Many researchers have reported that DNA repair capacity, mutations in tumor suppressor gene p53, and oncogenes K-ras and c-erbB-2 were responsible for the sex difference.(47) In this study, we found an increased risk of ESCC in males with the C allele, which suggests that sex hormones or other gene mutations could explain this difference. The findings show that breastfeeding and sex hormones are associated with a decreased risk of oesophageal and gastric junction adenocarcinoma.(48) Furthermore, the different expression of the cytochrome P4501A1 and GSTM1 genes(49) and sex hormones also provide supporting evidence.
In conclusion, CASP3 829 A>C polymorphism was significantly associated with the risk of ESCC, which might be a marker for genetic predisposition to ESCC in Chinese populations. However, our current study has limitations due to the lack of replication data. Our findings need to be validated in other groups. Additionally, future studies on how CASP3 polymorphisms affect gene function are needed and a larger case–control study will further clarify the association of SNP with ESCC in other ethnic groups.
We greatly thank Dr. Y. Shimada (Kyoto University, Kyoto, Japan) for providing esophageal cancer cell lines KYSE30 and KYSE450. This work was supported by the National Natural Science Foundation of China (Grant No. 81101483).