The study included 162 patients with HCC at the Cancer Center of Sun Yat-sen University (Guangzhou, China) from 2001 to 2005. All patients underwent hepatectomy as initial therapy, and did not receive chemotherapy or radiotherapy as follow-up treatment before recurrence. All samples were histologically confirmed. After surgical resection, the tissue samples were immediately frozen in liquid nitrogen and then stored at -80°C until use.

Clinicopathological details and follow-up information were obtained from hospital records. The patients enrolled in the study were residents of Guangdong Province. Infection with hepatitis B virus (HBV) or hepatitis C virus (HCV) was diagnosed when HBV surface antigen or HCV antibody was detected by enzyme linked immunosorbent assay in the serum isolated from peripheral blood. The TNM criteria and the Edmondson and Steiner grading system were used to classify tumor stages and differentiation grades, respectively. Informed consent was obtained from each patient. This study was approved by the Clinical Research Ethics Committee of Sun Yat-sen University Cancer Center.

Total genomic DNA was isolated with a standard protocol that included proteinase digestion, phenol-chloroform extraction, and ethanol precipitation. Polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis was used to detect the genotype. A 160-bp fragment that covered the G-1666A polymorphism was generated using sense primer 5'-GCAACAGCATAACTGGCATT-3' and reverse primer 5'-GAGACCTCCCCCCAATCA-3'. PCR was performed as follows: 20 ng genomic DNA was amplified in a 20-μL reaction mixture that contained 2 mmol/L MgCl₂, 0.4 mmol/L dNTPs, 0.2 μmol/L each primer, and 0.5 U Taq polymerase (Promega, Madison, WI, USA). After a total of 36 cycles of amplification at an annealing temperature of 58°C, 3 μL PCR products was then incubated overnight at 37°C with 15 U HpaII (MBI Fermentas, Hanover, MD, USA). Digested products were analyzed by 2% agarose gel. PCR fragments that demonstrated altered electrophoretic patterns were purified and characterized by direct DNA sequencing. Results represent two independent experiments.

Liver cancer cell lines Huh7 and Hep3B were obtained from the American Type Culture Collection (Manassas, VA, USA) and grown in Dulbecco’s Modified Eagle’s Medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum in a humidified environment of 37°C that contained 50 mL/L CO₂. Nuclear protein extracts from Hep3B and Huh7 cells were prepared according to the manufacturer’s protocol (NucBuster Protein Extraction Kit; Novagen, Darmstadt, Germany).

Electrophoretic mobility shift assay (EMSA) was performed with the Gel Shift Assay System (Promega), according to the manufacturer’s instructions. The following oligonucleotides that corresponded to the promoter region of MRP1 and covered the G-1666A polymorphism were synthesized (underline letters indicate polymorphism): -1666A allele, 5'-GGGGGACCCGGGCCAATAAAAAAATCA-3'; -1666G allele, 5'-GGGGGACCCAGGCCAATAAAAAAATCA-3'; nonspecific (scrambled) oligonucleotide, 5'-GAAGCGGTGACACGGAACATCACGAAA-3'. Oligonucleotides were annealed and end-labeled with [γ-³²P]-ATP. Five micrograms of Hep3B or Huh7 nuclear extracts were added in each binding reaction. For the competition assay, a 10-, 50- or 100-fold molar excess of unlabeled oligonucleotide was added to the binding reaction mixture as a competitor. The products were separated on pre-electrophoresed 5% polyacrylamide gels at 4°C. The gels were then dried at 80°C for 4 h and exposed to a Storage Phosphor Screen (Amersham Bioscience, Sunnyvale, CA, USA), which was subsequently read with a Typhoon Phosphor Imager (Amersham Bioscience). The putative transcription factors that recognized the sequences that overlapped the G-1666A site were predicted with Alibaba2.1 (http://www.gene-regulation.com/pub/programs/alibaba2/index.html) and the transcription element search software (TESS, http://www.cbil.upenn.edu/cgi-bin/tess/tess).

The χ² and Fisher’s exact tests were used for the analysis of the relationship between the genotypes and clinicopathological characteristics. Disease-free survival (DFS) was calculated from the day of surgery to either relapse or death without relapse, and it was censored only for patients who were alive and recurrence-free at the last follow-up. Overall survival (OS) was measured from the date of hepatectomy to the time of death or the last follow-up. Survival curves were obtained by the Kaplan-Meier method, and the statistical significance of the differences in survival among subgroups was evaluated with the log-rank test. The Cox proportional hazards model was employed to assess the independent prognostic values of the polymorphisms. Statistical analyses were all performed with SPSS software package (version 13.0; SPSS, Inc., Chicago, IL, USA). All statistical tests were two-sided, and P < 0.05 was considered to be statistically significant.

Demographic and clinicopathological characteristics of the 162 patients with HCC are summarized in Table 1. The mean age at first diagnosis of HCC was 48 years. Consistent with our previous study[15], most patients showed excessive γ-glutamyl transpeptidase and α-fetoprotein, along with liver cirrhosis, and > 80% of the enrolled patients were infected with HBV (140/161, 87.0%), which implicated HBV infection as a leading cause of HCC in Southeast China. In contrast, only a small number of patients were infected with HCV.

Genotyping was performed by PCR-RFLP. A 160-bp MRP1 promoter region that covered the G-1666A variant was digested with HpaII. After full digestion of the amplified PCR products, those from AA homozygotes still existed as a single 160-bp fragment, whereas those from the GG homozygotes had been divided into two fragments of 71 bp and 89 bp, respectively. The allele frequency of patients with HCC was 0.61 for MRP1 -1666A and 0.39 for -1666G. However, no significant correlations were found between the nucleotide variants and clinical variables (data not shown).

Growing evidence suggests that SNPs are closely related to the risk and outcome of cancer[14,16,17]. To investigate the impact of the G-1666A polymorphism on the prognosis of patients with HCC, we next analyzed the 4-year DFS and OS of patients with different genotypes. A significant correlation between the -1666 polymorphism and postoperative survival was found. The mean survival times of patients with the AA, AG and GG genotypes were 30.4 ± 18.2, 30.7 ± 17.4 and 24.8 ± 17.3 mo, respectively. The survival curves showed that the 4-year rate of DFS among patients who carried GG decreased significantly compared with those who carried the AA or AG allele (P = 0.031, Figure 1A). Moreover, if the patients with AA and AG genotypes were combined, the discrepancy became more obvious (P = 0.011, Figure 1B). Further analysis revealed a similar, albeit non-significant, trend between the -1666 polymorphism and 4-year OS (Table 2). Multivariate Cox proportional hazard analysis was then performed, and the variables that showed significance by univariate analysis were adopted as covariates (Table 2). The results revealed that the MRP1 G-1666A polymorphism was an independent prognostic factor for 4-year DFS [hazard ratio (HR) = 3.067, 95% confidence interval (CI): 1.587-5.952, P = 0.001, Table 3].

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Kaplan-Meier disease-free survival curves for hepatocellular carcinoma patients who carried different multidrug resistance related protein-1 -1666 genotypes. A: Comparison between three genotypes; B: GG genotype compared with the other two genotypes. Log-rank P values are indicated. Tick marks represent censored data. DFS: Disease-free survival.

One of the key features of HCC is the much higher incidence in men than in women[1]. In our study cohort, the male/female ratio was 9.8:1. Further stratification of the patients by sex revealed an even more pronounced association of the -1666GG genotype with poorer survival (P < 0.05, Figure 2). Multivariate analysis suggested that the -1666GG genotype was an especially powerful independent prognostic factor of 4-year DFS (HR = 3.154, 95% CI: 1.604-6.201, P = 0.001) and OS (HR = 3.342, 95% CI: 1.474-7.576, P = 0.004) in the men with HCC (Table 3).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Kaplan-Meier curves for disease-free survival (A and B) and overall survival (C and D) for male patients with hepatocellular carcinoma and different multidrug resistance related protein-1 -1666 genotypes. A: Comparison of disease-free survival (DFS) between three genotypes; B: AA and AG grouped together and compared to GG genotype; C: Comparison of overall survival (OS) between three genotypes; D: OS of AA and AG genotypes compared with GG genotype. Log-rank P values are indicated. Tick marks represent censored data.

We performed EMSA to evaluate the influence of the G-1666A polymorphism on the binding affinity of the MRP1 promoter with putative transcription factors. The radiolabeled -1666G probe showed strong DNA-protein binding ability in the presence of nuclear proteins extracted from the Hep3B cell line, whereas the -1666A probe barely showed any interaction (Figure 3A, lane 2 and lane 10, respectively). In order to assess the binding specificity and the differences in binding affinity between the G and A alleles, competition assays were performed with unlabeled -1666A and -1666G oligonucleotides. A 50-fold excess of unlabeled -1666A oligonucleotides only partially disrupted the binding of the radiolabeled -1666G probe with nuclear extracts (Figure 3A, lane 7 and Figure 3B, lane 6), whereas this amount of unlabeled -1666G oligonucleotides almost completely abolished the binding (Figure 3A and B, lane 4). In contrast, a non-specific competitor had no effect (Figure 3B, lane 8). Similar results were obtained when using the nuclear extracts from Huh7 cells (data not shown). These data suggested that the G-1666A polymorphism could affect the binding affinity of the MRP1 promoter with transcription factors, and that the G allele had a stronger binding affinity than the A allele.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Electrophoretic mobility shift assay of the multidrug resistance related protein-1 promoter region that contained the G-1666A site. A: Analysis was performed in the presence (+) or absence (-) of Hep3B nuclear extract. Each binding reaction contained γ-³²P-labeled -1666G (lanes 2-8) or -1666A (lanes 10-14) probes. A 10-, 50-, or 100-fold (as indicated) excess of unlabeled (cold) -1666A or G oligonucleotides (lanes 6-8, 11, and 12 or 3-5, 13, and 14) were included in the binding reactions as specific competitors. Labeled oligonucleotides incubated without the nuclear extracts were included as negative controls (lanes 1 and 9); B: In the present of Hep3B nuclear extract, 10- or 50-fold more excess of unlabeled -1666G oligonucleotides (lanes 3 and 4) or -1666A oligonucleotides (lanes 5 and 6) or non-specific oligonucleotides (lanes 7 and 8) were used as competitors. Lane 1 was the negative control. Lane 2 indicated the labeled -1666G oligonucleotides incubated with the nuclear extracts only. The asterisks indicated the DNA-protein complex.

Hepatocellular carcinoma (HCC) is the fifth most common malignancy worldwide and the third leading cause of cancer death, and long-term survival remains poor because of high rates of tumor recurrence or progression. Therefore, markers that can be used for improving therapeutic regimens and survival prediction are urgently needed.

The finding that human multidrug resistance protein-1 (MRP1) is expressed in unusually large amounts in HCC suggests it has a role in the growth and progression of this cancer. Expression of MRP1 is affected by the genetic sequence in its promoter region, therefore, the authors of this study examined the potential of different sequences (polymorphisms) in the MRP1 promoter to serve as indicators of prognosis and outcome in patients with HCC.

Recent studies have demonstrated that mutations within the MRP1 gene have value in predicting the response to chemotherapy in different cancers, but the clinical significance of such mutations in the MRP1 promoter for patients with HCC is unknown. This is believed to be the first study to identify a polymorphism in the promoter of MRP1 that is an independent prognostic factor for 4-year overall survival in men with HCC. The correlation between the MRP1 polymorphism and the survival of women with HCC requires further investigation. The authors also demonstrated that the polymorphism altered the affinity of nuclear proteins for the DNA in the HCC cells, which might explain the mechanism by which the expression of MRP1 was reducing.

The genetic sequence identified in this study can be used to test tissue samples from patients with HCC to help predict their outcome after therapy. This information can be used to guide treatment decisions and improve therapeutic regimens for individual patients.

MRP1 is one of a family of proteins found on the surface of cells. These proteins transport a wide variety of substances and can contribute to resistance to chemotherapy by transporting anticancer drugs out of cancer cells. The promoter region of a gene is a sequence of nucleotides that regulates whether and how much of a protein is synthesized from that gene.

This is a study of an important area of cancer genomics. The authors found that a single nucleotide polymorphism of MRP1 promoter region is an independent prognostic factor for HCC patients. The study design was well-organized and they reached a conclusion by making full use of the data.