Although obesity is a risk factor for colorectal cancer, the underlying mechanism is not clear. Adiponectin is an adipokine that binds to 2 types of receptors, AdipoR1 and AdipoR2. The plasma concentrations of adiponectin are reduced in obese individuals and adiponectin has been reported to have anticarcinogenic properties. Furthermore, AdipoR1 and AdipoR2 have been reported to be expressed in several malignancies. However, little is known about the expression of AdipoR1 and AdipoR2 in colorectal cancer and its clinicopathological implications. In addition, the relationship between adiponectin and colorectal cancer has not yet been determined. Here, we sought to investigate adiponectin and adiponectin receptors in relation to colorectal cancer. AdipoR1 and AdipoR2 immunostaining was detected in 72 and 68% of human colorectal cancer tissue, respectively. AdipoR1 and AdipoR2 expression levels were inversely related to T stage. The lowest AdipoR1 and AdipoR2 expression were detected in poorly differentiated adenocarcinoma. RT-PCR also showed the expression of AdipoR1 and AdipoR2 in HCT116 and SW620. MTT assay and TUNEL assay demonstrated the tendency of growth inhibition and apoptosis induction in both cell lines after full-length adiponectin treatment although statistically insignificant. Microarray analysis revealed several gene responses to full-length adiponectin, including upregulation of ENDOGL1 and MT1G. In conclusion, AdipoR1 and AdipoR2 may be intimately related to the progression of colorectal cancer. Further studies may be warranted to assess adiponectin and its receptors as a novel target for inhibition of colorectal cancer growth.
Colorectal cancer has long been prevalent in Western populations and was estimated to comprise 10.4% of new cancer cases and be the second leading cause of cancer death in the United States in 2008.1 Over the past few decades, the incidence of colorectal cancer has increased in Asia and it is now one of the most common malignancies worldwide.2
Obesity is one of the risk factors for colorectal cancer. Several case-control studies assessing body mass index and colorectal cancer incidence have reported an increased risk of colorectal cancer in obese individuals compared with normal weight individuals.3–5 In addition, prospective cohort investigations have reported a positive association between body mass index and colorectal cancer, with relative risks of 1.2–3.4.6–9 Indices such as waist-to-hip ratio or waist circumference, which indicate abdominal or visceral adiposity, have also been suggested to be independent risk factors for colorectal cancer.9–11 Together with other researchers, we have further demonstrated the importance of visceral adiposity to colorectal cancer risk by the direct measurement of visceral fat by computed tomography (CT) analysis.12, 13
Although the pathophysiological mechanism by which obesity is linked to colorectal cancer is not completely understood, several mechanisms have been proposed. Obesity-induced insulin resistance increases the level of plasma insulin, which has a mitogenic effect on the colonocyte. Insulin resistance also results in increased levels of plasma glucose and fatty acids, which may induce metabolic perturbations and alterations in cell signaling pathways and oxidative stress.14, 15 Insulin-like growth factor 1 and chronic inflammation associated with obesity have also been suggested to have a role in the development of colorectal cancer.15, 16 The importance of adipokines, which are adipocyte-secreted hormones, has recently been highlighted and may represent an additional mechanism by which obesity is associated with colorectal carcinogenesis.16, 17
Adiponectin, a 244 amino-acid protein, is one of the most abundant adipokines and circulates in human plasma as full-length adiponectin (f-adiponectin) or globular adiponectin produced by the proteolytic cleavage of f-adiponectin at amino acid 110. The plasma concentrations of adiponectin, which exerts antidiabetic and anti-inflammatory actions,18, 19 are significantly reduced in obese individuals.20 Recently, adiponectin has been reported to have anticarcinogenic properties. Epidemiologic studies have shown an inverse association between adiponectin plasma level and the risk of various malignancies including breast, endometrial and prostate cancer.21–23 Furthermore, in vitro studies have shown that adiponectin may control cell number by inhibiting cell proliferation and inducing apoptosis of some breast and prostate cancer cells although not all the cell lines tested were responsive to adiponectin.24, 25 Because 2 types of adiponectin receptor, AdipoR1 and AdipoR2, have been identified in various tissues and cell types,26 the action of adiponectin may be mediated by these receptors. Recently, the expression of AdipoR1 and AdipoR2 has been reported in human lung, breast, pancreas and gastric cancer tissue.27–30 Immunohistochemical studies have also evidenced the presence of AdipoR1 and AdipoR2 in human colorectal cancer tissue.31, 32 However, little is known about the clinicopathological implications of adiponectin receptor expression in colorectal cancer.
In this study, we sought to analyze the expression of adiponectin receptors in human colorectal cancer tissues and the clinicopathological implications. We further determined whether adiponectin can affect the growth and apoptosis of colorectal cancer cells. We also investigated genes that response to adiponectin treatment on colorectal cancer cells.
Material and Methods
Of 730 patients who had undergone surgical resection of primary colorectal cancer at the Asan Medical Center in 2001, 100 were randomly selected for the immunohistochemical study. The surgically resected colorectal cancer tissue specimens were fixed in 10% buffered formalin and embedded in paraffin. Formalin-fixed, paraffin-embedded tissue sections were stained for AdipoR1 and AdipoR2. The 5-μm paraffin tissue sections were deparaffinized with 100% xylene for 7 min twice, with 100% alcohol for 5 min twice, with 95% alcohol for 5 min, with 90% alcohol for 5 min, with 80% alcohol for 5 min, and finally rehydrated with tap water. For antigen retrieval, samples were treated under Trilogy™ buffer (#920P-06, Cell Marque, CA) for 15 min at 121°C using electronic pressure cooker (CMM977, Cell Marque), and were cooled at room temperature for 20 min. Subsequently, samples were incubated in 3% H2O2 for 10 min at room temperature for endogenous blocking. Slides were incubated with primary antibodies against AdipoR1 (1:700, Phoenix Pharmaceuticals, CA) and AdipoR2 (1:300, Phoenix Pharmaceuticals), incubated with primary antibody enhancer for 10 min, then with Polymer HRP for 15 min (TL-125-HL, Ultravision LP Large Volume Detection System HRP Polymer Kit, LabVision, CA). Color reaction was performed by incubation with Chromogen DAB (3,3′ diaminobenzidine; TA-125-HDX, LabVision) for 2 min. The slides were counterstained with Meyer's hematoxylin (HMM999, Scytek Laboratories, UT).
Interpretation of immunostaining
AdipoR1 and AdipoR2 expression levels were graded on a scale of 0 to 2 based on staining intensity and proportion of positive tumor cells by an expert pathologist (MJ Kim) who was blinded to patients' clinical records. AdipoR1 and AdipoR2 staining levels were scored as 0 if no cancer cells were reactive, 1 if staining was weakly positive in <2/3 of cancer cells or strongly positive in <1/3 of cancer cells, and 2 if staining was weakly positive in >2/3 of cancer cells or strongly positive in >1/3 of cancer cells. Staining of epithelial cells in normal crypts was used as an endogenous positive control.
Medical records of the 100 patients were reviewed; the location of colorectal cancer, gender, age and body mass index (BMI, kg/m2) at the diagnosis of colorectal cancer, postoperative pathological tumor node metastasis stage, differentiation and lymphovascular invasion status of resected cancer specimen were assessed. Recurrence during the follow-up period and survival status at the last follow-up were also recorded. These clinicopathological findings were analyzed according to the grade of AdipoR1 and AdipoR2 immunostaining.
Colorectal cancer cell lines, culture conditions and cytokines
Human colorectal cancer cell lines HCT116 and SW620 have routinely been cultured in our laboratory. In this study, all the cell lines were grown in RPMI-1640 medium. All media were supplemented with 10% fetal bovine serum plus penicillin (100 U/mL) and streptomycin (100 μg/mL). Recombinant human f-adiponectin was purchased from R&D Systems (Minneapolis, MN).
RT-PCR analysis of AdipoR1 and AdipoR2 expression
RT-PCR analyses were performed in both HCT116 and SW620. Total RNAs from cell lines were extracted with RNeasy Mini kit (QIAGEN GmbH, Germany) according to the manufacturer's protocol. cDNA was synthesized from 1 μg of total RNA extracted from each samples using high fidelity RT-PCR system. The primers for adiponectin receptor amplification were as follows; 5′-AATTCCTGAGCGCTTCTTTCCT-3′ (forward) and 5′-CATAGAAGTGGACAAAGGCTGC-3′ (reverse) for AdipoR1 and 5′-TGCAGCCATTATAGTCTCCCAG-3′ (forward) and 5′-GAATGATTCCACTCAGGCCTAG-3′ (reverse) for AdipoR2. GAPDH cDNA amplification was used as an internal control with 5′-TCGGAGTCAACGGATTTGGTCGTA-3′ (forward) and 5′-AGCCTTCTCCATGGTGGTGAAGA-3′ (reverse). Each of the 30 cycles of amplification was performed as follows; 94°C denature for 30 sec, 54°C annealing for 30 sec, 72°C extension for 1 min. Then, the PCR products were analyzed in 2% agarose gel. Additional RT-PCR analyses were performed in MCF-7, a human breast cancer cell line, as a positive control.
HCT116 (5 × 103/well) and SW620 (5 × 103/well) cells were seeded in 96-well plates 24 hr prior to serum starvation. Cells were starved of serum for 24 hr and then incubated in the serum-free medium without f-adiponectin for 96 hr. Cells were also incubated with increasing concentrations of f-adiponectin (0, 0.1, 1 and 10 μg/mL) for 96 hr.
MTT was added to each well at a concentration of 0.5 mg/mL. After incubation at 37°C for 3 hr, cells were lysed in 50% dimethylformamide and 20% SDS at 37°C. Optical densities (OD) at 550 and 670 nm were measured using a plate reader, and differential OD between 550 and 670 nm (OD 550-670 nm) were determined. Experiments were repeated 3 times in total.
After HCT116 (6 × 105/well) and SW620 (6 × 105/well) cells were grown in 6-well plates for 24 hr, the medium was replaced by serum-free medium. Cells were then cultured in the presence of 10 μg/mL f-adiponectin for 96 hr. Attached cells were harvested by trypsinization, combined with floating cells and suspended in phosphate-buffered saline (PBS). Cells were fixed in 4% paraformaldehyde at 4°C overnight and washed twice with PBS. Cells were labeled for DNA fragmentation by terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling (TUNEL, Roch Diagnostics, Mannheim, Germany) according to the manufacturer's instructions. Apoptotic data were analyzed by flow cytometry using a FACScan (Becton Dickinson GmbH, Heidelberg, Germany). Apoptotic assay was performed 3 times in total.
Total RNA was extracted using TRIzol (Invitrogen Life Technologies, Carlsbad, CA) and purified using RNeasy columns (Qiagen, Valencia, CA) according to the manufacturer's protocol. After processing with DNase digestion and clean-up procedures, RNA samples were quantified, aliquoted and stored at −80°C until use. For quality control, RNA purity and integrity were evaluated by denaturing gel electrophoresis, assessment of the OD 260/280 ratio, and analysis using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA).
Total RNA was amplified and purified using the Ambion Illumina RNA amplification kit (Ambion, Austin, TX) to yield biotinylated cRNA according to the manufacturer's instructions. Briefly, 550 ng of total RNA was reverse-transcribed to cDNA using a T7 oligo (dT) primer. Second-strand cDNA was synthesized, in vitro transcribed and labeled with biotin-NTP. After purification, the cRNA was quantified using a ND-1000 Spectrophotometer (NanoDrop, Wilmington, DE).
Labeled cRNA samples (750 ng) were hybridized to each expression bead array for 16–18 hr at 58°C, according to the manufacturer's instructions (Illumina, San Diego, CA). Array signals were detected using Amersham fluorolink streptavidin-Cy3 (GE Healthcare Bio-Sciences, Little Chalfont, UK) according to the manufacturer's instructions. Arrays were scanned with an Illumina bead array reader confocal scanner according to the manufacturer's instructions. Array data export processing and analysis were performed using an Illumina BeadStudio.
The quality of hybridization and overall chip performance were monitored by visual inspection of internal quality control checks and the raw scanned data. Raw data were extracted using the software BeadStudio v.3.3 provided by the manufacturer (Illumina). Array data were filtered using a detection p value < 0.05 (similar to signal to noise) in at least 50% of samples (a filtering criterion for data analysis was applied and a higher signal value was required to obtain a detection p value < 0.05). Selected gene signal values were transformed by logarithmic analyses and normalized using the quantile method. Comparative analyses were performed using (|fold| > or < 1.5). Hierarchical cluster analysis was performed using complete linkage and Euclidean distance as a measure of similarity. All data analyses and visualization of differentially expressed genes were conducted using ArrayAssist® (Stratagene, La Jolla, CA).
The chi-square test using linear-by-linear association was used to compare categorical variables. Kruskal–Wallis test was performed to analyze the results of MTT assay. Wilcoxon signed rank test was used to evaluate the results of TUNEL assay. Cumulative overall survival and recurrence-free survival were calculated using the Kaplan–Meier method and between-group comparisons were made using the log-rank test. A p value of less than 0.05 was considered statistically significant. Statistical evaluation was performed using the statistical software package SPSS 14.0 for Windows (SPSS, IL). The study was approved by the Institutional Review Board of the Asan Medical Center and was in compliance with the Helsinki Declaration.
Expression of adiponectin receptors in human colorectal cancer tissues
Descriptive characteristics of immunostaining study subjects are presented in Table 1. Immunostaining of AdipoR1 was positive in 72 of 100 cases (72%), whereas immunostaining of AdipoR2 was positive in 68 of 100 cases (68%; Fig. 1). Demographic characteristics associated with AdipoR1 and AdipoR2 expression are described in Table 2. No correlation was detected between BMI and the grade of AdipoR1 or AdipoR2 immunostaining. Clinicopathological factors associated with AdipoR1 and AdipoR2 expression are summarized in Table 3. As T stage progressed, the grade of AdipoR1 immunostaining decreased (p = 0.007). The proportion of negative AdipoR1 immunostaining was significantly higher in poorly differentiated adenocarcinoma than in well or moderately differentiated adenocarcinoma (p = 0.002). The grade of AdipoR2 immunostaining also showed a significant correlation with T stage and differentiation of cancer cell. In addition, the grade of AdipoR2 immunostaining was correlated with AJCC (American Joint Committee on Cancer) stage and lymphovascular invasion.
Table 1. Descriptive characteristics of patients with colorectal cancer
Table 2. Association between AdipoR1/AdipoR2 expression and demographic characteristics
Table 3. Association between AdipoR1/AdipoR2 expression and clinicopathological factors
We analyzed overall survival and recurrence-free survival according to the grade of AdipoR1 immunostaining (Fig. 2). The grade of AdipoR1 immunostaining was not associated with overall survival and recurrence-free survival of colorectal cancer patients. There were also no differences in overall or recurrence-free survival between grades of AdipoR2 immunostaining (Fig. 3).
Expression of adiponectin receptors in colorectal cancer cell lines
RT-PCR analysis showed that both colorectal cancer cell lines HCT116, SW620 and the positive control, MCF-7 all expressed AdipoR1 and AdipoR2 (Fig. 4).
Inhibition of growth of colorectal cancer cell lines by adiponectin
Treatment with f-adiponectin for 96 hr appeared to inhibit cell growth in a dose-dependent manner in HCT116 cells; however, it was not statistically significant (p = 0.055). Similar tendency was observed in SW620 (p = 0.067). Growth of HCT116 and SW620 was suppressed by 31 and 38%, respectively, after treatment with 10 μg/mL f-adiponectin for 96 hr.
Induction of colorectal cancer cell apoptosis by adiponectin
Exposure of HCT116 cells to 10 μg/mL f-adiponectin resulted in slightly increased apoptotic cell proportion compared to negative control, which was not statistically significant (13.3% ± 16.7% vs. 5.4% ± 5.7%, p = 0.109). The results were similar in SW620 (25.6% ± 3.1% vs. 8.5% ± 7.2%, p = 0.109).
Gene expression of colorectal cancer cell lines after stimulation with adiponectin
In microarray analyses, expression levels of several genes involved in ion transport, apoptosis and cell-cycle control were affected by treatment with f-adiponectin. Interestingly, the expression levels of many metallothionein (MT) genes were upregulated after treatment with f-adiponectin. Genes responding to f-adiponectin in both HCT116 and SW620 are listed in Table 4. Results of hierarchical clustering are shown in Figure 5.
Table 4. Genes regulated in both HCT116 and SW620 cells after stimulation with f-Adiponectin1
250 ng/mL of f-adiponectin was used in this study.
In the present study, we found that AdipoR1 and AdipoR2 are expressed in human colorectal cancer cells both by RT-PCR and by immunohistochemical analyses. In addition, we demonstrated that expression of AdipoR1 and AdipoR2 are associated with T-stage progression and differentiation of cancer cells. Although expression of adiponectin receptors in human colorectal cancer tissue has been reported,31, 32 few studies have investigated their clinicopathological implications in colorectal cancer. We found an inverse correlation between the T stage of colorectal cancer and the grade of AdipoR1 and AdipoR2 expression in immunostaining study. This finding may support the hypothesis that adiponectin can inhibit the growth of colorectal cancer, because abundant expression of adiponectin receptors in colorectal cancer tissue may facilitate the anticarcinogenic effect of adiponectin; by contrast, low expression levels of adiponectin receptors may promote progression of colorectal cancer by protecting against the effects of adiponectin. This assumption may be further applicable to AdipoR2, because the grade of AdipoR2 expression was inversely correlated to AJCC stage and the status of lymphovascular invasion. Based upon these results, we believe that AdipoR1 and AdipoR2 expression may be intimately related to the progression of human colorectal cancer. We also found that levels of AdipoR1 and AdipoR2 expression were lower in poorly differentiated adenocarcinomas compared with well or moderately differentiated adenocarcinomas.
We assessed whether grade of adiponectin receptor expression is associated with the survival of patients with colorectal cancer. The grade of AdipoR1 and AdipoR2 expression was not a prognostic factor of colorectal cancer. This finding is in contrast with a recent report by Barresi et al. in which they performed immunohistochemical assays for AdipoR1 and AdipoR2 in 49 surgically resected gastric cancers. They found that the expression of AdipoR1 and AdipoR2 is significantly associated with longer overall survival in patients with gastric cancer.30 The reason why the prognostic role of adiponectin receptor expression is different between colorectal cancer and gastric cancer is unclear. Larger prospective studies are necessary to clearly determine the prognostic role of adiponectin receptor expression in colorectal cancer and gastric cancer.
As obesity is a key factor in circulating adiponectin levels, obesity indices such as BMI can be related to adiponectin receptor levels in tissues. However, in our immunostaining analysis, we could not find any association between the grade of adiponectin receptor expression and BMI. This finding is in accord with a recent study demonstrating that body adiposity was not related to adiponectin receptor levels in rat colon.33 We suggest that adiponectin receptor expression may be controlled not only by fat mass and resultant circulating adiponectin, but also by other factors such as cancer cell characteristics including differentiation because our data showed significantly lower expression of AdipoR1 and AdipoR2 in poorly differentiated adenocarcinomas.
Several recent studies have demonstrated probable direct inhibitory effect of adiponectin on colorectal cancer growth.34–37 Fujisawa et al. showed a significant increase in cell proliferative activity in the colonic epithelial cells of the adiponectin-deficient mice compared with wild-type mice fed a high-fat diet.34 Sugiyama et al. demonstrated the inhibition of colorectal cancer cell growth by adiponectin using MTT assay.35 Although our MTT study results showed a statistically insignificant inhibition by f-adiponectin, we suggest that higher concentration of f-adiponectin may significantly inhibit colorectal cancer growth based upon 31–38% of growth inhibition by 10 μg/mL f-adiponectin in our data and the aforementioned previous reports. Because the physiological plasma concentration of adiponectin in healthy nonobese humans is ∼10 μg/mL,38 supra-physiological doses of adiponectin could be used for the treatment of colorectal cancer. However, further studies addressing the safety of supra-physiological dose of adiponectin should be performed before the efficacy trial.
Induction of apoptosis by f-adiponectin was statistically insignificant in the present study. Several studies have shown that adiponectin induces apoptosis in breast cancer and myelomonocytic leukemia cell lines; this was mediated by downregulation of antiapoptotic genes such as Bcl-2 and Bcl-xL and upregulation of proapoptotic genes such as Bax and p53.24, 39, 40 The reason for different apoptotic response between cancer cells is not clear, which necessitate further studies.
We performed microarray analysis of HCT116 and SW620 cells after treatment with f-adiponectin to identify possible candidate genes that mediate the direct effects of adiponectin on colorectal cancer cells; the results showed that several genes were upregulated and downregulated (Table 4). ENDOGL1 showed increased expression with f-adiponectin treatment. It is a member of the DNA/RNA endonuclease family including human endonuclease G (ENDOG),41 which is a mitochondria-specific nuclease that translocates to the nucleus during apoptosis. We suggest that the role of ENDOGL1 in the apoptotic pathway, especially in regard to colorectal cancer progression, should be investigated in further studies. Several MT genes were found to be upregulated in HCT116 and SW620 cells after f-adiponectin treatment. Genes of the MT family encode a class of metal-binding proteins involved in several cellular processes, including potent antioxidant function against various types of oxidative damage as well as regulation of zinc and copper homeostasis, and their expression is often dysregulated in human tumors. Microarray analysis of thyroid tumors showed downregulated MT1G, and restoration of MT1G by cDNA transfection led to a reduced growth rate and in vivo tumorigenicity of papillary thyroid carcinoma-derived K1 cells, indicating an oncosuppressive role for MT1G in thyroid papillary tumorigenesis.42 Downregulation of MT1G expression in renal cell carcinoma and MT1X expression in advanced prostate cancer have also been reported.43, 44 Based on these results and those of the present study, we suggest that MT genes may be involved in the mechanism by which adiponectin affects colorectal cancer risk and they may have an anticarcinogenic role in colorectal carcinogenesis. However, other studies have shown increased expression levels of MT genes in various human tumors of the breast, liver, lung and urinary bladder.45 Hence, the function and expression of MT genes may not be universal, but may depend on several factors such as type of tumor, differentiation status and other gene mutations; further studies are needed to determine these relationships.
Our present study has several limitations. First, f-adiponectin concentrations used for in vitro experiments may not have been sufficiently varied. Both HCT116 and SW620 cells showed 31–38% growth inhibition when treated with f-adiponectin at a concentration of 10 μg/mL. Experiments with various concentrations of f-adiponectin, including higher concentrations, may be necessary to confirm the inhibitory effect of f-adiponectin on colorectal cancer cells. Second, we evaluated the inhibition of growth, induction of apoptotic response, and gene responses only at 96 hr after treatment of f-adiponectin. Further studies should include several earlier time periods to fully determine the effects of f-adiponectin. Third, gene responses in microarray were weak, which leaves the necessity of further clarification through the application of various concentration of f-adiponectin at various time points. Despite these limitations, we believe that this study is meaningful because, to the best of our knowledge, the present study is the first to identify the clinicopathological implications of adiponectin receptor expression in human colorectal cancer such as an inverse correlation between adiponectin receptor expression and T stage.
In conclusion, adiponectin receptor expression may be intimately related to the progression of colorectal cancer. Further studies may be warranted to assess adiponectin and its receptors as a novel target for inhibition of colorectal cancer growth.