The authors have declared that no competing interests exist.
Conceived and designed the experiments: LD SZ. Performed the experiments: CP HGH. Analyzed the data: JJH LD. Contributed reagents/materials/analysis tools: CP HGH. Wrote the paper: LD HGH.
Liver metastasis is a major cause of mortality from colorectal cancer (CRC). However, mechanisms underlying this process are largely unknown. Osteopontin (OPN) is a secreted phosphorylated glycoprotein that is involved in tumor migration and metastasis. The role of OPN in cancer is currently unclear. In this study, OPN mRNA was examined in tissues from CRC, adjacent normal mucosa, and liver metastatic lesions using quantitative real-time PCR analysis. The protein expression of OPN and its receptors (integrin αv and CD44 v6) was detected by using an immunohistochemical (
Colorectal cancer (CRC) continues to be the second leading cause of cancer-related deaths in the United States
OPN is a phosphoglycoprotein which is originally isolated from mineralized bone matrix
Here, we investigated the role of OPN in colon metastasis to liver and the mechanisms by which OPN promotes colon cancer metastatic activity.
Tumor specimens from 44 cases of CRC and 20 patients with liver metastasis at the Second Affiliated Hospital of Zhejiang University were obtained fresh from surgical resections with prior consent. The median age of the patients was 57 years at operation, ranging from 25 to 84. Pathological characteristics including tumor location, stage, and differentiation was recorded. This study has obtained human research ethics approval from the Ethics Committee of the Second Affiliated Hospital, School of Medicine, Zhejiang University.
Two colorectal adenocarcinoma cell lines Colo-205 and SW480 cells were obtained from the Cancer Institute of Zhejiang University and maintained in RMPI-1640 with 10% calf serum at 37°C with 5% CO2.
Tissue samples were snap-frozen immediately after surgical resection. The total RNA was isolated using Trizol (Invitrogen) according to the manufacture’s instructions. RNA integrity was examined by gel-electrophoresis with 1% formaldehyde gel. 2.0 µg total RNA was reversely transcribed with SUPER SCRIPT II RTPCR kit (Invitrogen). Real-time PCR was performed in a final volume of 50 µl containing 1 µl RT transcript, 5 µM of each primer, 0.5 unit of AmpliTaq DNA polymerase (Invitrogen). As an internal positive control, real-time PCR analysis was performed on the GAPDH gene in parallel. The following primers were used: OPN forward primer
PCR products were electrophoresed on agarose gels and stained with ethidium bromide and photographed. PCR products of the expected size were cloned into a pGEM-T-Easy vector (Promega) and the sequences of the resulting plasmids were confirmed.
To obtainan anti-sense OPN expression plasmid, we generated a 201 bp cDNA fragment (from 210 to 368) using PCR with OPN primers as mentioned above. cDNA fragment was then cloned into the
The open reading frame (ORF) of OPN was obtained by RT-PCR. OPN primer F: 5′-
The plasmids of OPN-antisense, OPN-sense and pcDNA3.1(+)were linearlized with
Specific probes complementary to the OPN mRNA were designed based on GenBank J04765 (Human osteopontin mRNA, complete cds). These oligonucleotide sequences have 100% homology with the OPN gene and minimal homology with other mammalian gene sequences as searched in GenBank by blast. DIG RNA Label Kit (SP6/T7) (Roche) was utilized to generate both sense and antisense. Paraffin-embedded tissues were sectioned at 4 μ m. The slides were deparaffinized and treated with 0.2% M. HCl for 20 min. After digested with proteinase K, slides were incubated with pre-hybridization solution (containing 500 µg/ml poly (A)) at 42°C for 30 min in a moisture chamber. The solution was replaced by hybridization solution (containing 0.3 µg/ml sense or anti-sense RNA single strand probe respectively, 50% formamide, 10% dextransulfate, and 2×SSC), and hybridized at 42°C overnight. After washed with 2×SSC and 1×SSC for 15 min succession, the slides were blocked with PBS containing 5% bovine serum albumin for 10 min, incubated with biotin-labeled mouse-anti-digoxin antibody for 30 min and then with alkaline phosphoesterase-affinitin for 30 min. NBT-BCIP was used to stain slides in pH 10 substrate buffer. After dehydration, slides were mounted with aqueous mounting media (Sigma).
Rat anti-human osteopontin monoclonal antibody (Chemicon, Billerica, MA) was used for
Monolayer Colo-205 cells and SW-480 cells were maintained in Ø 35 mm dishes and normal medium until 70%–80% confluence. Attached cells were then overlaid on the monolayer cells, mixed well and incubated for different periods of time.
Floating cells were collected after incubation for 10 min, 20 min, 30 min and 40 min, and the number was counted using a hemocytometer. Cell adhesion ratio was then calculated: adhesion ratio = (sum of cells added- suspended cells)/sum of sample cells. Homotypic adhesion assay was performed with same cell lines, each of them between navel vessel endothelium cells ecv 304 cells was used respectively in heterotypic adhesion assay.
1×106/ml cell samples were suspended thoroughly in PBS, mixed with 20 µl CD44-FITc antibody, controlled with 20 µl IgG1/2a antibody, cell incubated with Ab at RT for 20 min. Rinsed with PBS, mixed with 1% paraformal-dehyde, then detected with FAC Scan (BD company, Cell Quest TM Version 3.3 software).
Cells of 80% confluence were examined for fluorescence redistribution after photobleaching. Cells rinsed with Hanks (Ca2+, Mg2+), incubated 15 min with 10 µg/ml 5(6)-carboxyfluorescein diacetate (CFDA) in 37°C, 5% CO2. After rinsed by PBS, a cell adjacent to other cells was selected and its fluorescence was photobleached under LEICA TCS-SP laser co-focus microscope (Argon ion laser beam, 518 nM, Time/lampse program for photobleaching and scan, Physiology software for image and data). Digital images of the fluorescent emission excited by weak laser pulses were recorded at regular intervals for 12 min (scanning period 1 minute before and after photobleaching) and stored for subsequent analysis. In each experiment, one labeled, isolated cell was left unbleached as a reference for the loss of fluorescence due to repeated scanning and dye leakage, and an isolated, bleached cell served as a control.
Data are expressed as mean±SEM. Paired-samples
Based on quantitative real-time PCR, CT values are in inverse proportion to the log value of the original copy numbers of target sequence. We found that OPN mRNA was expressed in 44 cases of CRC tissues but at different levels; the maximum CT value was 1.47 and the minimum value was 0.85,
In
In 20 of examined liver metastatic tissues from CRC, the maximum CT value was 1.30 and the minimum value was 0.92,
Based on the Dukes stage, seven of total 44 CRC patients were at stage A and the OPN mRNA mean CT value was 1.49±0.38, 1.21±0.28 for stage B(17 patients), 1.11±0.29 for stage C(18 patients ),and 0.92±0.32 for stage D(2 patients). The expression level of OPN mRNA in CRC tissues at early stage was significantly higher than which at advanced stage (one-way ANOVA,
We then used
Using an immunohistochemical method, we further confirmed that OPN protein was expressed in the cytoplasm of CRC cells in both the primary lesion and metastatic tissues. The brown positive stains were also found in normal hepatocytes around the metastatic tissue. Adjacent normal colorectal tissues exhibited no signal. As control, OPN proteins stain in normal liver tissues without CRC metastasis
In
OPN has been shown interacting with a number of different integrins via the RGD sequence, including αvβ3, αvβ1 and αvβ5. CD44v6 is a splice variant of CD44 (CD44v) could also bind to OPN and likely promotes cancer cell adherence to the vascular endothelium and the base membranes, then enhances the invasion and metastasis of colonic carcinomas. To verify these receptors in normal liver, we assayed for integrin αvβ1 and CD44v6 by IHC. We found that both integrin αvβ1 and CD44v6 were stained positively in hepatocytes in normal liver tissues
In
CD44 plays a major role in cell-cell adhesion, cell-substrate interaction, lymphocyte homing, and tumor metastasis. Recent studies have shown that high expression of CD44 in certain types of tumors is associated with the hematogenic spread of cancer cells.
We detected CD44 expression in transferred cell lines using flow cytometry. We found that after down-regulation of OPN expression, CD44 expression on Colo-205-OPN-antisense cell membranes was decreased from 54.28±0.11 to 35.35±0.24 in Colo-205,
Cell adhesion is an important factor during cancer metastasis. The detachment of cancer cells from their parent tumors is an initial event in metastasis and is related to homotypic adhesion decreasing. Once tumor cells escape from the primary tumor, they interacts with the preexisting host basement membranes at a diverse stages during the metastatic cascade. Heterotypic adhesion helps cancer cells complete this whole process.
We detected the homotypic and heterotypic adhesion ability in Colo-205-OPN-antisense and SW-480-OPN-sense separately. Navel vessel endothelium cells ecv 304 was used as target cells in heterotypic adhesion. Homotypic adhesion ability was enhanced after OPN-antisense transfection in Colo-205, but was weakened in SW-480-OPN-sense. In contrast, heterotypic adhesion ability was weakened in Colo-205-OPN-antisense, but was enhanced in SW-480-OPN-sense.
GJIC consists of intercellular exchange of low molecular weight molecules, and is the only means for direct contact between cytoplasms of adjacent animal cells. Disturbances of GJIC have been associated with many pathological conditions, such as carcinogenesis or hereditary illness
We used gap-FRAP to measure GJIC in OPN transfected cells (SW-480-pcDNA3.1(+)-OPN) while Colo-205 were semi-suspension cells which were not appropriated for this method. Cell signal exchange was impaired and GJIC function was inhibited. Compared with control cells (vector transfected SW-480-OPN cells, SW-480-pcDNA3.1(+)), fluorescence was not redistributed well in SW-480-OPN-sense cells after both 5 minutes and 10 minutes of photobleaching. After 5 minutes of photobleaching, the percentage of fluorescence redistribution after photobleaching (FRAP%) was 24.65±4.08% in SW-480-pcDNA3.1(+)-OPN compared 44.74±6.23% in SW-480-pcDNA3.1(+)(
uniform fluorescence intensity before photobleaching in SW-480-pcDNA3.1(+).
Human colon cancer affects nearly 150,000 patients and 60,000 deaths in the United States per year. The liver is the primary extra-colonic site for CRC metastasis and represents the most common location and clinical presentation for recurrent disease in patients who fail locoregioal therapy
Using quantitative real-time, we verified that OPN was highly expressed in metastatic hepatic lesions from CRC compared to primary CRC tissue and adjacent normal mucosa. In addition, the expression of OPN mRNA in tumor tissues was significantly related with the CRC stages. These results implicate that OPN is a potential marker for tumor invasion and hepatic metastasis of CRC.
Notably, OPN mRNA was found in the cytoplasm of CRC cells by
We examined two different cell lines to investigate the role of OPN in regulating metastasis. The Colo-205 cell line was transfected with an OPN-antisense eukaryotic plasmid to inhibit OPN protein expression in this cell line. SW-480 cells were transfected with an OPN-sense eukaryotic plasmid, so that the OPN protein is expressed in this cell line.
After cell transfection, Colo-205-OPN-antisense transfected cells clustered (data not shown), homotypic adhesion was enhanced in these cells in comparison to the Colo-205-pcDNA3.1(+). However the adhesion ability was weakened in SW-480-OPN-sense cells, and the mobility was enhanced in these cells expressing OPN (data not shown). At the same time, we investigated the heterotypic adhesion with ecv304 as target cells, heterotypic adhesion ability was weakened in Colo-205-OPN-antisense cells and was enhanced in SW480-OPN-sense cells. These results suggest that OPN decreases the homotypic adhesion among CRC cells and enhances the heterotypic adhesion ability between CRC cells and endothelium cells.
Invasion and metastasis are important characteristics of malignant tumors. The whole process of metastasis includes angiogenesis adhesion, degradation, movement, and reattachment
Because of OPN, homotypic adhesion ability is weakened in CRC cells and the cells to depart from primary site more easily, and enter the blood circulation. The first step of metastasis is onset. However, the CRC cells are more easily to invade ECM (the extra cellular matrix) when heterotypic adhesion is enhanced. These two steps are important to metastasis in malignant cancer.
GJIC has been speculated to be a necessary, if not sufficient, biological function of metazoan cells in regulation of growth control, differentiation and apoptosis of normal progenitor cells
In our study, fluorescence redistribution after photobleaching was weakened in SW-480- OPN-sense cells. GJIC function was inhibited and signal exchange were impaired, So these tumor cells are more easily detached from primary lesions to initiate the first step of metastasis incidently.
Based on studies of the correlated mechanisms between expression of OPN and metastasis in combination with the pertinent literatures, we concluded that a potential mechanism of OPN promoting CRC liver metastasis is as follows: CRC cells express OPN, the homogeneity adherence ability are decreased, the function of GJIC is inhibited, the capability of invasion and movement is increased, which makes it easy for the CRC cells to depart from primary lesion, get into circulation and initiate the procedure of metastasis. At the same time, due to the OPN, the expression of CD44, a metastasis-related factor, is also increased. The heterotypic adhesion between CRC cells, ECM and vascular endothelial cell is enhanced. In the course of portal vein refluence, OPN allows cancer cells to invade peripheral vessels, settling down and planting in the liver. Furthermore, the receptors of OPN, CD44v6 and integrin αv were found in liver. The interaction between ligands and receptors could explain the liver-cling of CRC cells.
These results suggest that OPN is one of the important factors in tumor infiltration and metastasis. The effect on homotypic and heterotypic adhesion in CRC cells of OPN is essential in the process of liver metastasis.
We are grateful to Dr. Jiaping Peng and Dr. Xiaoming Fang for their invaluable scientific advice during the study.