|Home | About | Journals | Submit | Contact Us | Français|
Correspondence to: Da-Tian Bau, MD, Associate Professor, Chairman, Terry Fox Cancer Research Laboratory, China Medical University Hospital, 2 Yuh-Der Road, Taichung 404, Taiwan, China. wt.gro.humc.liam@naitad
Telephone: +886-422052121 Fax: +886-422053366
AIM: To investigate the association of Caveolin-1 (Cav-1) polymorphisms with colorectal cancer (CRC) risk in a central Taiwanese population.
METHODS: Three hundred and sixty-two patients with colorectal cancer and the same number of recruited age- and gender-matched healthy controls were genotyped. And only those matches with all single nucleotide polymorphisms data (case/control = 362/362) were selected for final analyzing.
RESULTS: There were significant differences between CRC and control groups in the distributions of their genotypes (P = 1.6 × 10-12 and 3.0 × 10-4) and allelic frequencies (P = 2.3 × 10-13 and 4.0 × 10-5) in the Cav-1 G14713A (rs3807987) and T29107A (rs7804372) polymorphisms respectively. As for the haplotype analysis, those who had GG/AT or GG/AA at Cav-1 G14713A/T29107A showed a 0.68-fold (95% CI: 0.48-0.98) decreased risk of CRC compared to those with GG/TT, while those of any other combinations were of increased risk. There were joint effects of Cav-1 G14713A and T29107A genotype with smoking status on individual CRC susceptibility.
CONCLUSION: This is the first report providing evidence of Cav-1 being involved in CRC and it may be novel useful genomic markers for early detection of CRC.
Colorectal cancer (CRC) is one of the most grave public health problems. There are nearly one million cases of CRC diagnosed worldwide each year. The prevalent incidence and age-adjusted mortality of CRC has kept on increasing in recent years in Taiwan. In 2008, the incidence and mortality of CRC has occupied third place among the common cancers. Etiological studies have attributed more than 85% of CRC to several environmental factors[1,2], in particular meat consumption, cigarette smoking and exposure to carcinogenic aromatic amines such as arylamines and heterocyclic amines[3-5].
In recent years, investigators have become interested in caveolae to define how these lipid domains participate in the pathogenesis of human cancers and what their possible utility may be for detection and treatment. Caveolae are vesicular invaginations of the plasma membrane, thought to play a critical role in transcytosis, communication between cell surface membrane receptors and intracellular signaling protein cascades such as apoptosis and tumorigenesis[7,8]. Caveolins are the major structural proteins of caveolae and this family contains three members in mammals, Caveolin-1 (Cav-1), Cav-2 and Cav-3[7,9], in which Cav-1 is the principal structural protein. It has been demonstrated that Cav-1 is down-regulated in sarcoma, lung carcinoma and ovarian carcinoma[10-12]. However, elevated expression of Cav-1 has been associated with the metastasis of esophageal squamous cell carcinoma and prostate cancer and negatively correlated with patient survival[13,14]. These findings indicate that the role of Cav-1 may vary considerably depending on the tissue involved.
Previous reports have found a differential display of Cav-1 in CRC cell lines and experimental colon adenocarcinomas when compared to normal tissue[15,16]. However, the role of Cav-1 in aberrant cellular physiology is not fully understood. Moreover, the functional role of Cav-1 in CRC is not precisely identified in vivo as of now. Therefore, the emerging evidence pointing to the role of Cav-1 gene in carcinogenesis led us to study whether different alleles of this gene are associated with CRC. Thus, the aims of the current study were to determine the genotypic frequency of six polymorphisms of the Cav-1 gene at C239A (rs1997623), G14713A (rs3807987), G21985A (12672038), T28608A (rs3757733), T29107A (rs7804372) and G32124A (rs3807992) and their association with CRC susceptibility. To the best of our knowledge, this is the largest study carried out to evaluate the contribution of Cav-1 polymorphisms in colorectal oncology.
The study population consisted of 362 CRC patients and 362 cancer-free control volunteers. Patients diagnosed with CRC were recruited at the outpatient clinics of general surgery during 2002-2008 at the China Medical University Hospital, Taichung, Taiwan. The clinical characteristics of patients, including histological details, were all graded and defined by expert surgeons (Dr. Yang’s team). All patients voluntarily participated, completed a self-administered questionnaire and provided peripheral blood samples. An equal number of non-cancer healthy volunteers were selected as controls by matching for age, gender and some indulgences after initial random sampling from the Health Examination Cohort of the hospital. The exclusion criteria of the control group included previous malignancy, metastasized cancer from other or unknown origin and any familial or genetic diseases. This study was approved by the Institutional Review Board of the China Medical University Hospital and written-informed consent was obtained from all participants.
Genomic DNA was prepared from peripheral blood leucocytes using a QIAamp Blood Mini Kit (Blossom, Taipei, Taiwan) and further processed according to our previous papers[17-25]. Briefly, the following primers were used for Cav-1 C239A (rs1997623): 5’-GTGTCCGCTTCTGCTATCTG-3’ and 5’-GCCAAGATGCAGAAGGAGTT-3’; for Cav-1 G14713A (rs3807987): 5’-CCTTCCAGTAAGCAAGCTGT-3’ and 5’-CCTCTCAATCTTGCCATAGT-3’; for Cav-1 G21985A (12672038): 5’-GGTGTCAGCAAGGCTATGCT-3’ and 5’-CCAGACACTCAGAATGTGAC-3’; for Cav-1 T28608A (rs3757733): 5’-GCTCAACCTCATCTGAGGCA-3’ and 5’-GGCCTATTGTTGAGTGGATG-3’; for Cav-1 T29107A (rs7804372): 5’-GCCTGAATTGCAATCCTGTG-3’ and 5’-ACGGTGTGAACACGGACATT-3’; and for Cav-1 G32124A (rs3807992): 5’-GGTGTCTTGCAGTTGAATG-3’ and 5’-ACGGAGCTACTCAGTGCCAA-3’. The following cycling conditions were performed: one cycle at 94°C for 5 min; 35 cycles of 94°C for 30 s, 55°C for 30 s and 72°C for 30 s; and a final extension at 72°C for 10 min. The PCR products were studied after digestion with Avr II, Bfa I, Hae III, Tsp509 I, Sau3AI and Nla III, restriction enzymes for Cav-1 C239A (cut from 485 bp C type into 170 + 315 bp T type), Cav-1 G14713A (cut from 268 bp A type into 66 + 202 bp G type), Cav-1 G21985A (cut from 251 + 43 bp A type into 153 + 98 + 43 bp G type), Cav-1 T28608A (cut from 298 bp T type into 100 + 198 bp A type), Cav-1 T29107A (cut from 336 bp A type into 172 + 164 bp T type) and Cav-1 G32124A (cut from 213 + 142+ 67 bp A type into 142 + 118 + 95 + 67 bp T type) respectively.
Only those matches with all single nucleotide polymorphisms (SNPs) data (case/control = 362/362) were selected for final analyzing. To ensure that the controls used were representative of the general population and to exclude the possibility of genotyping error, the deviation of the genotype frequencies of Cav-1 SNP in the control subjects from those expected under the Hardy-Weinberg equilibrium was assessed using the goodness-of-fit test. Pearson’s χ2 test or Fisher’s exact test (when the expected number in any cell was less than five) was used to compare the distribution of the Cav-1 genotypes between cases and controls. Cancer risk associated with the genotypes was estimated as odds ratio and 95% confidence intervals using unconditional logistic regression. Data was recognized as significant when the statistical P-value was less than 0.05. To evaluate effect modification by smoking, stratified analyses were conducted for chosen SNPs to compare the association across exposure categories of smoking status (never-smokers and smokers). All statistical tests were performed using SAS, Version 9.1.3 (SAS Institute Inc., Cary, NC, USA) on two sided probabilities.
The frequency distributions of selected characteristics of CRC patients and controls are shown in Table Table1.1. These characteristics of patients and controls are all well matched. None of these differences between groups were statistically significant (P > 0.05) (Table (Table1).1). The frequencies of the genotypes for the Cav-1 C239A, G14713A, G21985A, T28608A, T29107A and G32124A between controls and CRC patients are shown in Table Table2.2. Genotype distribution of various genetic polymorphisms of Cav-1 G14713A and T29107A were significantly different between CRC and control groups (P = 1.6 × 10-12 and 3.0 × 10-4 respectively), while those for Cav-1 C239A, G21985A, T28608A and G32124A were not significant (P > 0.05) (Table (Table2).2). To sum up, the polymorphism of Cav-1 G14713A and T29107A are associated with CRC risk and may be a biomarker for CRC early detection. The representative PCR-based restriction analyses for the Cav-1 G14713A and T29107A polymorphisms are shown in Figure Figure11.
The frequencies of the alleles for the Cav-1 C239A, G14713A, G21985A, T28608A, T29107A and G32124A between controls and CRC patients are shown in Table Table3.3. The two SNPs of Cav-1 found to be associated with CRC in Table Table2,2, G14713A and T29107A, are also found to be associated with higher CRC susceptibility in their allele frequency analysis here. As for the other four SNPs, the distributions of their allele frequencies are not significantly different in controls and CRC patients (Table (Table33).
Considering potential interactions between the two significant SNPs of Cav-1 gene and CRC susceptibility, the risk of CRC related to haplotype distributions of Cav-1 G14713A and T29107A were further analyzed (Table (Table4).4). Compared with GG/TT haplotype of Cav-1 G14713A and T29107A, the GG/AT or GG/AA group has a 0.68-fold lower risk of CRC (95% CI: 0.48-0.98). Other combinations of AG/TT, AG/AT or AG/AA, AA/TT and AA/AT or AA/AA conferred 2.78-fold (95% CI: 2.04-4.22), 2.02-fold (95% CI: 1.28-2.94), 3.48-fold (95% CI: 1.86-5.59) and 2.29-fold (95% CI: 1.49-3.06) increased risks compared to the GG/TT haplotype respectively (Table (Table44).
Since smoking is the predominant risk factor for CRC, the interaction between Cav-1 genotype and individual smoking habits was also analyzed by stratified individual smoking status (Table (Table5).5). We noticed that subjects with the hetero- or homozygous AA for Cav-1 G14713A had higher risks of CRC in both smoker and non-smoker groups, irrespective of before or after adjusting their age, gender and smoking pack-years. In the case of Cav-1 T29107A, the homozygous AA had lower risks of CRC in both smoker and non-smoker groups. The heterozygous AT of Cav-1 T29107A also had protective effects in the smoker group. To sum up, there was an obvious interaction between smoking status and Cav-1 genotypes in the CRC susceptibility.
Although several investigations have shown that Cav-1 plays a critical role in many tumors[10-14], few data are available which consider Cav-1 for genetic predisposition to cancers[26,27]. In 2004, the inactivation of Cav-1 by mutation models or via reducing its expression was found to involve in the pathogenesis of oral cancer. In that study, the exon 1 and 3 sequences of Cav-1 were investigated in 74 oral squamous cell carcinomas and 15 oral cancer cell lines and the expression of Cav-1 was examined. It was reported that only five mutations (1 missense and 4 silent mutations) of Cav-1 were identified in many cases and they were all found in exon 3. Since sequencing of exonic and promoter regions had not revealed variants in Cav-1 that might have been directly involved in any cancer risk, it is reasonable for us to select intronic SNPs from the NCBI database and to evaluate the role of Cav-1 polymorphisms which have never been reported to be associated with CRC risk.
The main finding of this study is that Cav-1 G14713A (rs3807987) and T29107A (rs7804372) polymorphisms are associated with the susceptibility to CRC (Table (Table22 and 3) while the other four polymorphisms were not. The combinative analysis of Cav-1 G14713A (rs3807987) and T29107A (rs7804372) showed that, when taking G14713A/T29107A GG/TT haplotype as a reference, those with GG/AT or GG/AA were of lower CRC risk, while those with other haplotypes including AG/TT, AG/AT or AG/AA, AA/TT, AA/AT or AA/AA were of 1.93- to 3.22-fold higher risk. The data also supported that A allele of G14713A was risky and A allele of T29107A was protective. Although these genetic variations do not directly result in amino acid coding change, it is plausible to suspect that the alternative splicing, intervention, modification, determination or involvement of these SNPs influence the expression level or stability of the Cav-1 protein. In our immunohistochemistry detection of tumor tissue from oral cancer patients, taking the distant parts from the same subjects as internal control, we have found that Cav-1 was down-regulated in the tumor sites (unpublished data). We have also checked for the possibility that the various genotypes of Cav-1 may have differential effects on the clinical outcomes. However, after performing all the analysis for the effects of Cav-1 genotypes (both for G14713A rs3807987 and T29107A rs7804372) on age, gender, habits, primary tumor site, histological differentiation, invasion and lymph node involvement of the patients, no positive correlation could be found.
Environmental factors such as cigarette smoking were reported to be closely related to CRC carcinogenesis. In this study, the joint effects of Cav-1 gene and individual smoking behaviors were analyzed and both significant genetic-environmental interactions were observed in Cav-1 G14713A (rs3807987) and T29107A (rs7804372) (Table (Table5).5). The sample size and similar trends of significant data after age- and behavior-adjustments strengthen the accuracy and reliability of our findings and the frequencies of Cav-1 polymorphisms variant alleles were similar to those reported in the NCBI website in other Asian population studies. For instance, the minor A allele frequencies of Cav-1 G14713A are 22.1% in our control group, close to those of 16.7% for Beijing and 22.2% for Tokyo populations in NCBI, which strongly suggest no selection bias for the subject’s enrolments in terms of genotypes. The smoking population in our patient group is rather low so the data itself and that of matched control group are disadvantageous for us to do the stratified analysis of smoking status (Table (Table5).5). We agree that it is important to verify our findings in further larger studies and clarify the role of Cav-1 with more phenotypic and functional evidence in CRC and other cancer. In conclusion, this is the first report to provide evidence that Cav-1 G14713A and T29107A but not C239A, G21985A, T28608A or G32124A, were associated with higher susceptibility to CRC. They both have joint effects with smoking status on CRC susceptibility. The G allele of Cav-1 G14713A and the A allele of Cav-1 T29107A might become potential biomarkers for the CRC early detection, prediction and targets for integrative cancer therapy.
Colorectal cancer (CRC) is one of the most grave public health problems. There are nearly one million cases of CRC diagnosis worldwide each year. Caveolin-1 (Cav-1) has been associated with the metastasis of esophageal squamous cell carcinoma and prostate cancer and negatively correlated with patient survival.
Caveolins are the major structural proteins of caveolae and this family contains three members in mammals, Cav-1, Cav-2 and Cav-3, in which Cav-1 is the principal structural protein. It has been demonstrated that Cav-1 is down-regulated in sarcoma, lung carcinoma and ovarian carcinoma. In this study, the authors demonstrate that Cav-1 is involved in CRC and may be novel useful genomic markers for early detection of CRC.
Recent reports indicate that the role of Cav-1 may vary considerably, depending on the tissue involved. This is the first report providing evidence of Cav-1 being involved in CRC and it may be novel useful genomic markers for early detection of CRC.
The emerging evidence pointing to the role of Cav-1 in carcinogenesis led us to study whether different alleles of this gene are associated with CRC. Thus, the current study was to determine the genotypic frequency of six polymorphisms of the Cav-1 gene and their association with CRC susceptibility. To the best of our knowledge, this is the largest study carried out to evaluate the contribution of Cav-1 polymorphisms in colorectal oncology.
The authors have done a careful evaluation of Cav-1 in a large cohort of CRCs and have found novel molecular alterations in their population sample. This is a well conducted study and the readership will find the results interesting.
We are grateful to Wen-Shin Chang, Hsiu-Min Hsieh, Judy Wang and the Tissue Bank in China Medical University Hospital for their technical assistance.
Supported by Research Grants from the China Medical University and Hospital (DMR-99-041 and CMU-99-NTU-10), the Terry Fox Cancer Research Foundation and the National Science Council (NSC 98-2320-B-039-010-MY3).
Peer reviewers: Runjan Chetty, Professor, Department of Pathology and Gene Regulation, University of Glasgow, Western Infirmary (Pathology), Dumbarton Road, Glasgow, G11 6NT, Scotland, United Kingdom; Seung Joon Baek, PhD, Associate Professor, Department of Pathobiology, College of Veterinary Medicine, The University of Tennessee, 2407 River Drive, Rm A228, Knoxville, TN 37996, United States; Xavier Sagaert, MD, PhD, Department of Pathology, University Hospital Leuven, Minderbroederstraat 12, Leuven 3000, Belgium
S- Editor Wang JL L- Editor Roemmele A E- Editor Yang C