Background

The incidence of colorectal cancer (CRC) is higher in the United States than in China, where it is the third leading cause of cancer-related death in both sexes. However, its rate in China has increased steeply in recent years. Most Chinese patients with CRC have metastatic disease at diagnosis; earlier detection of their disease would greatly improve their odds of survival [1]. Although recent advances in chemotherapy have prolonged survival of patients with advanced disease, these treatments are handicapped by the lack of early-presenting biomarkers for CRC metastasis. Exploration of candidate genes to establish potent biomarkers for earlier detection of lymph node metastasis (LNM) would permit adoption of more suitable chemotherapeutic regimens, although prognoses of patients with CRC are also affected by such factors as tumor localization, quality of surgical procedures, gender, age, and patient’s overall performance status. Monitoring of high-risk individuals increases their 5-year survival rate and decreases chances of cumulative recurrence.

The excision repair cross-complementing group 1 gene (ERCC1) is an essential member of the nucleotide excision repair (NER) pathway, which accounts for most platinum–DNA adduct repairs. ERCC1 has been established as a useful molecular marker for NER activity. Early studies have shown that higher ERCC1 mRNA levels are associated with more active DNA repair processes in various tissues [2]. Interestingly, ERCC1 expression is also associated with cellular and clinical resistance to platinum compounds and to platinum-based chemotherapy, including those for lung and gastric malignancies [3, 4].

Breast cancer susceptibility gene 1 (BRCA1) is an essential component of several DNA-repair pathways that affect homologous recombination repair, non-homologous repair and NER. BRCA1 is considered to be a differential modulator of tumor response to cisplatin and taxanes [5–7], and BRCA1 levels are reportedly associated with chemosensitivity to cisplatin [8] and taxanes [9, 10]. Although the aforementioned studies suggest that both ERCC1 and BRCA1 are effective biomarkers for chemosensitivity in primary tumors, information on their expression in metastases is limited. Therefore, we explored the applicability of these biomarkers as predictive factors in CRC metastasis.

The current study is thus designed to investigate the possibility of using ERCC1 and BRCA1 as biomarkers in CRC metastatic specimens from Chinese patients. We examined mRNA levels of ERCC1 and BRCA1 in CRC with LNM (LNM CRC) or without LNM (non-LNM CRC), using real-time quantitative polymerase chain reaction (RT-PCR). We also verified the relationship of ERCC1 and BRCA1 levels on prognosis in CRC patients.

Methods

Patient population and characteristics of tissue samples

Samples from a total of 120 patients with colorectal carcinoma were collected from surgical resections performed in our hospital (Fudan University Shanghai Cancer Center, Shanghai, China), after obtaining informed consent. None of the patients received chemotherapy or radiotherapy before surgery. Resected specimens were reviewed by two senior pathologists according to the criteria described in the American Joint Committee on Cancer’s Cancer Staging Manual (7th edition, 2010) [11]. At least 12 lymph nodes each were retrieved from patients with non-LNM CRC, none of whom had distant metastasis. The fresh colorectal tumor tissues were obtained immediately after surgery, washed twice with chilled phosphate-buffered saline (PBS), immediately stored in liquid nitrogen and at –80°C in our tissue bank until further use. Ethical approval was obtained from the Cancer Center Research Ethics Committee of Fudan University.

Gene expression analysis by real-time quantitative PCR

ERCC1 and BRCA1 gene expression was assessed in SYBR Green Supermix (Promega). Samples were treated using a laser capture microdissection technique (Palm Microlaser, Oberlensheim, Germany) to ensure a minimum of 80% of tumor tissue. RNA was then extracted with phenol-chloroform-isoamyl alcohol, followed by precipitation with isopropanol in the presence of glycogen and sodium acetate, resuspension in diethyl pyrocarbonate water (Ambion Inc., Austin, TX), and treatment with DNAse I (Ambion Inc., Austin, TX) to avoid DNA contamination. Complementary DNA was synthesized using Maloney Murine Leukemia Virus retrotranscriptase enzyme. Template cDNA was added to Taqman Universal Master Mix (AB, Applied Biosystems, Foster City, CA) in a 12.5-μl reaction with specific primers and probe for each gene. Primer and probe sets were designed using Primer Express 2.0 Software (AB) and RefSeq sequences (http://www.ncbi.clm.cih.gob/gene). Quantification of gene expression was carried out using the ABI Prism 7900HT Sequence Detection System (AB).

Relative gene expression quantification was calculated according to the comparative cycle threshold (Ct) method [12] using β-actin as an endogenous control and commercial RNA controls (Stratagene, La Jolla, CA) as calibrators. Final results were determined as follows: 2– (ΔCt sample – ΔCt calibrator), where ΔCt values of the calibrator and sample are determined by subtracting the Ct value of the target gene from the value of the β-actin gene. In all experiments, only triplicates with a standard deviation of the Ct value < 0.20 were accepted. In addition, for each sample analyzed, a retrotranscriptase minus control was run in the same plate to assure lack of genomic DNA contamination.

Western blotting

Briefly, 30-μg protein samples from each case were separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and subsequently transferred to poly (vinylidene fluoride) membranes. The membranes were incubated with rabbit polyclonal antibody against ERCC1 or BRCA1 (1:1000 dilution; Abcam, Cambridge, UK) and then incubated with a horseradish-peroxidase-conjugated secondary antibody (1:100 dilution; Proteintech, Chicago, IL, USA). β-Actin was detected simultaneously as a loading control (anti-β-actin, 1:1000 dilution; Kangchen, Beijing, China). All blots were visualized using an ECL detection system (Amersham, Arlington Heights, IL, USA) and quantitated by densitometry using an LAS-3000 imager.

Immunohistochemistry

Both ERCC1 and BRCA1 expression were examined immunohistochemically using paraffin-embedded tissues. In brief, 4-μm-thick tissue sections were heated in 6.5 mmol/L citrate buffer (pH 6.0) at 100°C for 28 min, and incubated with antibodies against ERCC1 or BRCA1 (1:200 dilution). Immunostaining was performed using the DAKO En-Vision System (Dako Diagnostics, Zug, Switzerland). In the negative control group, PBS was used instead of primary antibody. Expression was scored by two independent experienced pathologists. Each sample was graded according to intensity and extent of staining. The intensity of staining was scored as 0 (no staining), 1 (weak staining), and 2 (strong staining). The extent of staining was based on the percentage of positive tumor cells: 0 (no staining), 1 (1–25%), 2 (26–50%), 3 (51–75%), and 4 (76–100%). These two scores were added together for a final score. The case was considered negative if the final score was 0 or 1 (−) or 2 or 3 (±), and positive if the score was 4 or 5 (+) or 6 or 7 (++). In most cases, the two reviewers provided consistent results. Any inconsistencies were resolved by discussion to achieve a consensus score.

Statistical analysis

The PCR analysis results were expressed as ratios between two absolute measurements (gene of interest/ internal reference gene). Student’s t test was used to evaluate differences in ERCC1 and BRCA1 expression between LNM CRC and non-LNM CRC. The χ2 test was used to assess relationships between ERCC1 and BRCA1 expression and clinicopathological factors. The cumulative recurrence and survival probability were estimated using the Kaplan–Meier method; differences were calculated by log-rank test. Prognostic factors were determined using Cox regression analysis. Recurrence-free and overall survival times were calculated from the first resection of the primary tumor to first evidence of recurrence or to death from any cause, respectively. The diagnosis of recurrence was based on the typical features presented on computed tomography/magnetic resonance imaging and elevated serum carcinoembryonic antigen. All P values were two-sided; P < 0.05 was considered to be significant. Statistical analyses used SPSS 13.0 software.

Results

Confirmation of ERCC1 and BRCA1 expression in non-LNM and LNM CRC specimens

Real-time quantitative PCR was used to analyze ERCC1 and BRCA1 expression in different groups of CRC. Relative gene expression quantifications were calculated according to the comparative Ct method using β-actin as an endogenous control. Median ERCC1 mRNA expression was 6.6 in non-LNM CRC (range: 2.8–8.52; n = 60) and 3.4 in LNM CRC (range: 2.2–8.16; n = 60). Median BRCA1 mRNA expression was 4.5 in non-LNM CRC (range: 3.21–10.52; n = 60) and 2.6 in LNM CRC (range 1.3–10.16; n = 60). Both ERCC1 and BRCA1 mRNA expression were down-regulated in LNM CRC compared with non-LNM CRC. Significant correlations between ERCC1 and BRCA1 expression levels (Spearman r = 0.516; P < 0.001) are shown in Figure 1.

Figure 1
figure 1

Confirmation of the overexpression of BRCA1 and ERCC1 in colorectal cancer.

Full size image

To compare the RT-PCR results with ERCC1 and BRCA1 protein levels, we extended the experiments in the same samples described above with western blot. Thirty micrograms of total proteins from LNM CRC and non-LNM CRC were analyzed using western blotting. Expression of ERCC1 and BRCA1 was dramatically higher in non-LNM CRC compared with LNM CRC (P < 0.001). A representative western blotting result is presented in Figure 1A.

Association of ERCC1 and BRCA1 expression with clinicopathological features and postoperative prognosis of patients with CRC

To study the relationships between ERCC1 or BRCA1 expression and clinicopathological features, and to assess whether ERCC1 or BRCA1 levels could predict clinical outcomes for patients with CRC, an immunohistochemistry study was used to confirm the PCR results, using the same samples.

Statistical analysis showed positive expressions of ERCC1 and BRCA1 were significantly associated with LNM, and advanced TNM stage (P < 0.001). However, no significant correlations were observed between ERCC1 or BRCA1 expression and other clinicopathological parameters of sex, age, tumor size, tumor differentiation and tumor location (Tables 1 and 2).

Table 1 BRCA1 expression and relationship with clinicopathological factors in CRC
Full size table
Table 2 ERCC1 expression and relationship with clinicopathological factors in CRC
Full size table

Furthermore, we have found that patients whose CRC specimens were negative for ERCC1 or BRCA1 had significantly poorer prognoses than those with ERCC1+/BRCA1+ CRC (Figure 2). The 5-year cumulative recurrence rate was significantly higher for patients in the ERCC1+/BRCA1+ group (P < 0.05). The 5-year estimated probability cumulative survival rate was also different in both group patients with BRCA1– CRC or ERCC1– CRC than in the ERCC1+/BRCA1+ group (P < 0.05). Univariate analyses revealed that LNM, TNM stage, ERCC1 expression and BRCA1 expression were related to recurrence and overall survival. In multivariate analysis, LNM, TNM stage, ERCC1 expression and BRCA1 expression were also independent prognostic factors for recurrence and overall survival (P < 0.05, Table 3).

Figure 2
figure 2

Expression of ERCC1 and BRCA1 correlated with poor prognosis in colorectal cancer patients.

Full size image
Table 3 Univariate and multivariate analyses of recurrence and survival (Cox regression)
Full size table

Discussion

CRC is an aggressive cancer, with 300,000 newly diagnosed cases and 200,000 CRC-caused deaths each year in Europe and the United States [13]. Biomarkers that could help diagnose CRCs before metastases occur can lead to earlier, more successful treatments. Moreover, accurate biomarkers for metastases could aid clinicians in identifying the most appropriate chemotherapies for patients with CRC who have had resections; up to 50% of patients who undergo potentially curative surgeries ultimately suffer recurrence and die of metastatic disease [14].

More than 85% of CRCs have been attributed to environmental factors, which can produce adducts, damage and strand breaks in DNA. Damaged DNA can be removed and recovered by DNA-repairing enzymes, which are critical for the genome protection and cancer prevention [15, 16]; the relationship between DNA repair genes and various cancers has been widely studied [17]. Adjuvant therapy is important to the clinical outcomes of the stage II–III cases that represent approximately 70% of CRC patients; adjuvant treatment following primary surgery could be improved by prognostic biomarkers. In this study, we investigated whether ERCC1 and BRCA1 could be such biomarkers and found that their down-expression is associated with poor prognosis in CRC.

ERCC1 is an endonuclease that helps perform NER of DNA [18]. We found negative expression of ERCC1 to correlate with LNM and advanced TNM stage, implying that ERCC1 decreases CRC metastasis, and by extension, that its reduced expression might be an early event in colorectal carcinogenesis. Patients with CRC that expresses negative ERCC1 have been shown to have high cumulative recurrence and low cumulative survival. On the other hand, the NER pathway is thought to repair DNA damage cause by platinum agents; several studies demonstrated an inverse relationship between impaired DNA-repair capacity and increased response rates to platinum drugs. Patients with CRC whose tumors show low levels of ERCC1 gene amplification reportedly have superior overall survival if treated with fluorouracil/oxaliplatin. However, many factors affect metastasis. ERCC1 could be restrained or promoted by other genes, which could influence LNM somehow; this area is already the subject of advanced research.

BRCA1 also has a role in DNA repair similar to that of ERCC1. Variants of BRCA1 are markers for breast and ovarian cancers, but it is unclear whether mutations in this gene increase the risk of CRC. Our data show that BRCA1 might be involved in CRC metastasis, as with ERCC1, and therefore it is a potential biomarker for CRC. However, changes in BRCA1 expression seem to be less sensitive to early-stage disease, compared with ERCC1, although changes in expressions of both proteins appear to correlate with LNM. This correlation between ERCC1 or BRCA1 expression and CRC metastasis has been suggested recently [19, 20], but no statistical association could be established between reduced expression of ERCC1 or BRCA1 and tumor stage and lymph node involvement. Our data show relationships between negative ERCC1 or BRCA1 expression and clinical CRC LNM, implying that both ERCC1 and BRCA1 are involved in CRC metastasis, and that reduced expression of these proteins are early events in colorectal carcinogenesis.

Arguably, the most important relationship we found was that between negative ERCC1 or BRCA1 expression in CRC and poor patient survival. We found negative ERCC1 or BRCA1 expression to correlate with LNM and advanced TNM stage, which suggests that ERCC1 and/or BRCA1 affect CRC progression from localized to LNM disease. In addition, patients with negative ERCC1 or BRCA1 expression in CRC have increased risk of recurrence and significantly reduced overall survival rates. Univariate and multivariate analyses indicate that ERCC1 or BRCA1 expression could serve as independent prognostic factors for recurrence and overall survival in patients with CRC.

Conclusions

Our quantitative analysis of mRNA and protein expression showed ERCC1 and BRCA1 to be significantly negatively expressed in LNM CRC. Further evaluation using the same sample set suggests that ERCC1 and BRCA1 are biomarkers for LNM and predictors of prognosis in CRC.