Purpose: The first step of metastasis is the detachment of cancer cells from the surrounding matrix and neighboring cells; however, how cancer cells accomplish this process remains unclear. Thus, we aimed to investigate the underlying mechanism that controls the early event of metastasis.
Experimental Design: One hundred and thirty-seven paired colorectal carcinoma and normal colon tissues were examined by immunohistochemical staining and Western blot for the expression of CD151, a member of the tetraspanin family that plays important roles in cell adhesion and motility. The effect of CD151 on cancer cell adhesion was investigated under normoxia and hypoxia conditions.
Results: The level of CD151 was down-regulated in colon cancer compared with the paired normal counterparts. Expression of CD151 was negatively regulated by hypoxia inducible factor-1–dependent hypoxic stress. Suppression of CD151 by hypoxia caused the detachment of cancer cells from the surrounding matrix and neighboring cells whereas restoration of CD151 expression during reoxygenation facilitated the adhesion capacity. Clinical examination further showed that metastasized cancer cells expressed a greater level of CD151 compared with that of primary tumor.
Conclusion: Regulation of CD151 by oxygen tension may play an important role in cancer metastasis by regulating the detachment from the primary site and homing in the secondary site.
- Colon cancer
- cell adhesion
A necessary early event in cancer metastasis is the conversion of the stationary phenotype to the migratory phenotype. However, how cancer cells detach from the highly organized architecture of cell-cell and cell-matrix adhesions was largely unknown. We report here that the expression of CD151 in colorectal cancer is lower than in its normal counterpart and is inversely correlated with the level of hypoxia inducible factor 1α. Inhibition of CD151 expression by hypoxia reduces the strength of cell-cell and cell-matrix adhesions, which may play important roles in the early event of cancer metastasis. In contrast, restoration of CD151 under reoxygenation increases cell adhesion and homing in the remote site. These data provide useful information for developing new treatment regimens to improve cancer therapy by targeting disruption of hypoxia-mediated down-regulation of CD151 in primary site or blocking the restoration of CD151 expression to inhibit the adhesion to the secondary site.
Cancer metastasis is a complex, multistep process with mechanisms largely unknown despite intensive investigation in the past. The first step of metastasis is the detachment of cancer cells from the surrounding matrix and neighboring cells. Adhesions between normal epithelial cells and cell-matrix are strong so local cell invasion is not able to occur. Therefore, the necessary early event in cancer invasion and metastasis is the conversion of the stationary phenotype to the migratory phenotype. This can be achieved by down-regulation of cell adhesion molecules such as cadherins and integrins. Indeed, studies have revealed that loss or down-regulation of E-cadherin and one or more of the integrins is a common feature of several epithelial malignancies (1, 2). Nonetheless, the underlying mechanisms of causing the down-regulation of these molecules are not yet clear. In addition, whether there are other molecules that also control cancer cell adhesion needs to be elucidated.
Many factors have been identified as playing important roles in the regulation of cancer metastasis. Among these, hypoxia is the most common and critical one. Because oxygen can only diffuse for 150 to 200 μm, cells will face the hypoxic stress when distances of cells from a capillary exceed this range. Owing to the rapid growing nature, cancer cells of solid tumors frequently encounter reduced oxygen tension. The hypoxic stress will force cancer cells to develop necessary processes such as induction of angiogenesis, metabolic switch, and migration to avoid cell death. These responses to hypoxia are mainly mediated through genes regulated by the hypoxia inducible factor (HIF; ref. 3). HIF is a heterodimer that comprises an α subunit and a β subunit. HIF-1β is constitutively expressed whereas HIF-1α is dynamically regulated according to the concentrations of oxygen.
CD151 (PETA-3/SFA-1) is a membrane protein belonging to the tetraspanin superfamily that comprises at least 26 highly conserved members (4) and is widely expressed by a variety of cell types. As a member of the tetraspanin family, CD151 contains four transmembrane domains, two extracellular loops, and intracellular NH2- and COOH-terminal domains. This feature enables CD151 to act as an adaptor or organizer by assembling multimolecular complexes of cell surface proteins including integrins and cadherins (5). Hence, CD151 involves in numerous biological processes, including cell adhesion (6), motility (7), angiogenesis (8), and formation of hemidesmosomes (9). So far, the best-known function of CD151 is its engagement of integrins in basal lateral cell surface complexes, such as α3β1 and α6β4, to facilitate cell-cell and cell-matrix adhesions (5, 9, 10). Particularly, the α3β1-CD151 interaction is stoichiometric and resistant to detergents (5), which has been shown to be essential for the formation of highly ordered cell-cell and cell-matrix contacts (11).
Although CD151 plays an important role in integrin-mediated cell functions, unlike integrins, CD151 knockout mice are viable and fertile (12), indicating that CD151 is not essential for embryonic development or reproductive function. However, CD151-null mice developed focal glomerulosclerosis and tubular cystic dilation and exhibited defects in platelet aggregation and keratinocyte migration (12, 13). Furthermore, the wound healing process was impaired in CD151-null mice due to defective basement membrane formation and epithelial cell adhesion (14).
The expression of CD151 in a small number of cancers, such as non–small cell lung cancer, colon cancer, pancreatic cancer, prostate cancer, and breast cancer, has been examined (15–19). The expression of CD151 in these cancers ranged from 55% to 77%, and CD151-positive patients tended to have lower survival rates within 5 years (15–17). Studies using dispersed cancer cells with or without overexpressed CD151 revealed that CD151 is important in controlling cancer cell adhesion to tissues (20). These data suggested that CD151 plays an important role in cancer biology. Surprisingly, the levels of CD151 expression between paired normal and cancer tissues were never compared, and the mechanisms of regulation of CD151 expression in normal and/or cancer cells had not been investigated thus far.
Considering the importance of CD151 in cell-cell and cell-matrix adhesion, it is possible that the level of CD151 in cancer cells may play a pivotal role in the initiation of metastasis (i.e., regulating the attachment/detachment status of the cancer cells). Furthermore, as hypoxia is the driving force to induce cancer cell migration, we hypothesized that hypoxia may regulate the expression of CD151, which would result in decreasing cell-cell and/or cell-matrix adhesion, thus increasing the migratory ability of cancer cells. Therefore, this study was designed to elucidate the effect of hypoxia on CD151 expression and the functional role of CD151 in causing colon cancer metastasis.
Materials and Methods
Collection of cancer and normal colon tissues. Tumor specimens were obtained from 137 patients with colon cancer who underwent surgery at the Department of Surgery of the National Cheng Kung University Hospital. Specimens of tumor tissue and adjacent tissues, at 1 cm and 5 cm away from the tumor site were collected. The postsurgical stage of each tumor was classified and histologically confirmed by pathologists. Human ethics approval was obtained from the Clinical Research Ethics Committee at The National Cheng Kung University Medical Center, and informed consents were obtained from the patient.
Immunohistochemistry. The procedure used was described previously (21) with minor modification. In brief, sections were incubated with anti-CD151 monoclonal antibody (Vision BioSystems) at 1:200 dilution at 4°C overnight (1:200 dilution for HIF-1α; Norvus). After washing with PBS, sections were incubated with biotinylated goat antimouse IgG (Vector Laboratories) at 1:1,000 dilution for 1 h at room temperature, washed again, and then incubated with Vectastain (Vector Laboratories) for 30 min. Colors were developed by AEC kit (Zymed Laboratories) for 8 min and stopped by PBS washing. Hematoxylin was used for counter stain. Ten high-power fields were examined per section. Scores of 1 to 3 were assigned to samples with no membrane staining but with cytoplasm staining, intermediate membrane staining, and strong membrane staining, respectively. Samples without positive staining were assigned a score of 0. All sections were scored by two pathologists independently.
Cell culture. Human colon cancer cell lines SW480, SW620, and HCT116 were used in this study. Culture media and conditions for each cell type were selected according to suggestions made by the American Type Culture Collection. When cells reached 70% confluence, they were cultured under hypoxia (1% O2) or normoxia (21% O2) for various time points as indicated in the figure legends. For chemical-mimicked hypoxia, cells were incubated in media containing different doses of desferrioxamine (iron chelator) for 24 h.
Western blot. Equal amounts of proteins were loaded into each lane, separated on SDS-PAGE, and transferred onto a polyvinylidene difluoride membrane (Millipore) according to standard procedure. Membranes were incubated with primary antibody overnight at 4°C followed by incubation with secondary antibody and detected by enhanced chemiluminescence (Perkin-Elmer). Membranes were then stripped, reprobed with antibodies against housekeeping genes, and detected by enhanced chemiluminescence for internal controls.
Promoter activity assay. Two oligonucleotides corresponding to human cd151 hypoxia-response element (HRE) matrices located in the promoter (−2467/−2484; forward: 5′-ctagGTGATACGTGAGGCCCAG-3′, reverse: 5′-gatcCTGGGCCTCACGT-3′) and intron II (1694/1718; forward: 5′-ctagTGTCTGGAAACGTGGCCCAGGGCGC-3′, reverse: 5′-tcgaGCGCCCTGGGCCACGTTTCCAGACA-3′) were synthesized and cloned into SV40-driven pGL3 plasmid after annealing as previously described (22). Commercial plasmid containing a cytomegalovirus-driven β-galactosidase reporter system (Promega) were cotransfected into the cells for internal control. At 6 h after transfection, the medium was changed, and cells were subjected to hypoxic or other treatments for the indicated time. Luciferase assays were done using the Dual Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions. Each luciferase assay experiment was done in duplicate and repeated as indicated in the figure legends.
Short interference RNA and overexpression of HIF-1α. Short interference RNAs (siRNA) against human CD151 and guanine-cytosine content-matched scramble control were purchased from Invitrogen. The siRNA was used to transfect cells with a final concentration of 40 nmol/L. After transfection, cells were incubated in normoxia or hypoxia for 24 h and subjected to Western blot analysis or adhesion assay as described above. The specificity of siRNA was determined by quantification of the INF-γ-sensitive gene 2′5′-oligoadenylate synthetase (OAS1).
Human HIF-1α (CEP4/HIF-1α, ATCC Cat no.: MBA-2) plasmid was purchased from the American Type Culture Collection (The Johns Hopkins University). Cells were transfected with 200 ng plasmid using Lipofectamine 2000 (Invitrogen Life Technologies) as described previously (23, 24). Cells were harvested at 24 h after transfection and subjected to Western blot analysis.
Chromatin immunoprecipitation assay. The protocol used was as described before (23, 24) with modifications. In brief, HIF-1α and DNA in normoxia- or hypoxia-treated cells were cross-linked by incubation for 10 min at room temperature with 1% formaldehyde. After aspiration of the formaldehyde, cells were washed twice with ice-cold PBS containing protease inhibitors. Genomic DNA was sheared to lengths of 0.5 to 1 kb by sonicating the cell lysate. Two percent of the diluted lysate was kept for input control. Anti-HIF-1α antibody or rabbit IgG (for negative control) was added to the supernatant fraction and subjected to chromatin immunoprecipitation assay following standard procedures. After being separated from protein, genomic DNA was amplified by PCR using specific primers (5′-GAGCTTCTGTCCACCTGTCC-3′ and 5′-CAGTGGGCAAGCTGTGAG-3′ for intronic HRE; 5′-GAGAGCGAGCACGAGAGG-3′ and 5′-CCATGCAAAAGAGGTACACG-3′ for promoter HRE).
Shake-off assay. Cells were seeded in 3-cm Petri-dish and then subjected to normoxia or hypoxia treatment for 4, 12, 24, or 48 h after overnight incubation. At each time point, dishes were removed from the incubator and shaken for 10 min at 300 rpm. The attached cells were then trypsinized and counted under a microscope using trypan blue after washing off floating cells. At 4 and 48 h, media were collected for evaluation of floating cells.
Adhesion assay. Substrates for adhesion assays were 20 mg/mL laminin or 10 mg/mL heat-inactivated bovine serum albumin (negative control). After overnight coating, wells were rinsed and blocked with 10 mg/mL heat-inactivated bovine serum albumin. Cells were treated with hypoxia or normoxia for 48 h, harvested, resuspended in serum-free medium, and 2 × 105 live cells were seeded into single wells of a 24-well plate. Cells were incubated for 0.5, 2, 12, or 24 h at 37°C. After incubation, the detached cells were removed by serum-free medium and attached cells were fixed with 1% glutaraldehyde for 10 min and stained with 0.1% crystal violet for 25 min. The plate was reimmersed in fresh tap water to remove excess dye and then allowed to air-dry for 5 to 10 min. A total of 50 μL of 0.5% Triton X-100 was added to each well to solubilize the cells overnight at room temperature. The absorbance was measured at OD595 using a microplate reader.
Cell aggregation assay. Cells were trypsinized in the presence of calcium. A single cell suspension was obtained, and 5 × 104 live cells were placed in a 0.2-mL tube and incubated on a rotation apparatus for 0, 1, or 3 h at room temperature. At the end of the incubation, cells were diluted into single wells of a 24-well plate to prevent further aggregation. After allowing cells to settle for 10 min at 37°C, the number of single cells and cells in clusters were manually counted. Ten low-power fields were counted using an inverted tissue culture microscope. The percentage of cells in clusters was calculated as the number in clusters of five or more cells divided by the total number of cells (single cells plus cells in clusters).
Statistical analysis. Differences between groups were analyzed using one-way ANOVA using commercial statistical software (GraphPad Prism 4.02). Tukey's procedure was used for post test. Student's t test was used to compare differences between normoxic and hypoxic groups in cell adhesion and aggregation assays. Statistical significance was set at P < 0.05.
Decreased expression of CD151 in colon cancer tissues. Equal amounts of proteins obtained from 5 cm distal or proximal of cancer, 1 cm proximal of cancer, and cancer tissues obtained from colorectal carcinoma biopsies were subjected to Western blotting. Expression of CD151 was decreased in the cancer region compared with other regions (n = 69; Fig. 1A and B ). By breaking down into four different groups according to stage of cancer, the result showed that amounts of CD151 were greater in the normal tissues compared with cancer lesion in each individual stage (Fig. 1C).
Next, we used immunohistochemical staining to examine cell type–specific expression of CD151 in another 68 samples. Expression of CD151 was predominantly in epithelial cells in both normal and cancer tissues (Fig. 2A ) whereas stromal cells only showed weak immunoreactivity. Normal epithelial cells showed strong staining of CD151, especially in the basal and lateral sites (Fig. 2A). In contrast, cancer cells showed weak to intermediate staining and were primarily present in the cytoplasm (Fig. 2A). This was especially evident in sections that contained both normal and cancer cells (Fig. 2B). All normal cells were CD151 positive, with about 22% showing weak staining, 50% intermediate staining, and 28% strong staining (Fig. 2C). In samples classified as stage I, 18% showed weak staining whereas 82% were CD151 negative. In stage II, about 84% samples were CD151 positive, with 52.6% weak staining and 31.5% intermediate staining. In stage III samples, 82% showed weak staining whereas 9% showed intermediate staining. In stage IV, 75% of cancer samples showed weak staining and 18.7% intermediate staining (Fig. 2C). Further analysis of CD151 expression in cancer samples revealed that the levels of CD151 were positively correlated with stage of the disease (Supplemental Table S1).
Expression of CD151 is down-regulated by hypoxia. The expression of HIF-1α protein was examined by Western blot, and results revealed that HIF-1α protein was undetectable or very low in normal tissues but was elevated in cancer cells (Fig. 3A ). This was inversely correlated with level of CD151 (Fig. 1A). Thus, we sought to determine whether hypoxia could inhibit CD151 expression. Accumulation of HIF-1α protein was evident at 24 and 48 hours after hypoxia treatment (Fig. 3B). Concordantly, levels of CD 151 were suppressed when SW480 cells were cultured under hypoxia condition (Fig. 3B). Similar results were also observed when two other colon cancer cell lines, SW620 and HCT116, were cultured under hypoxia condition (Fig. 3B), indicating this is not a strain-specific phenomenon. Furthermore, the level of CD151 was also inhibited by desferrioxamine-induced accumulation of HIF-1α (Supplementary Fig. S1) or overexpression of HIF-1α complementary DNA under normoxia (Fig. 3C) suggesting it is a HIF-1α-dependent event.
Next, we aimed to determine whether decreased CD151 expression under hypoxia condition is regulated by reduced CD151 transcription. SW480 and HCT116 cells were cultured under normoxia or hypoxia conditions and levels of mRNA were determined by real time reverse transcription-PCR. Amounts of mRNA were reduced when cells were cultured under hypoxia condition (Supplementary Fig. S2). Concordantly, chemical hypoxia also resulted in decreased CD151 mRNA expression (Supplementary Fig. S2).
By using a bioinformatic analytic platform,4 we identified putative HRE in human cd151 promoter and intron II, respectively (Fig. 3D). Results from promoter activity assay revealed that both HREs were functional in hypoxia-mediated reporter gene down-regulation (Fig. 3D). Similarly, desferrioxamine treatment also suppressed reporter gene in plasmids containing promoter HRE or intron II HRE (Fig. 3D). Finally, chromatin immunoprecipitation data further showed the binding of HIF-1α to the HRE of the human cd151 gene (Fig. 3D).
Hypoxia-mediated CD151 deregulation attenuates cell adhesion to extracellular matrix. To delineate the biological significance of hypoxia-mediated CD151 down-regulation in cancer progression, we tested whether hypoxia-mediated deregulation of CD151 would reduce cellular adherence to the extracellular matrix. Results showed that the number of attached cells in normoxia condition was greater than that in hypoxia condition after 12 hours of treatment (Fig. 4A ). The difference was more evident at 24 and 48 hours after hypoxia treatment. In agreement with this result, the percentage of floating cells in the hypoxia-treated group was significantly increased compared with that in the normoxia group at 48 hours after culture (Fig. 4A).
The aforementioned data showed that the adhesive ability of cells was reduced when cultured under hypoxia condition. We next aimed to determine whether preexposure to hypoxia would reduce cell adhesion ability. Equal numbers of live colon cancer cells were plated in 24 wells after preincubation under normoxia or hypoxia condition for 48 hours. At 30 minutes after cell plating, the normoxia-pretreated group had more cells attached to the surface compared with the hypoxia-pretreated one (Fig. 4B). The difference was markedly enhanced at 2 hours after plating (Fig. 4B). A similar result was also observed when HCT116 cells were used to repeat this experiment (Fig. 4C).
Suppression of CD151 expression by hypoxia inhibits cell-cell adhesion. Because CD151 is an important molecule in cell-cell adhesion, we sought to test whether the loss of cell-cell adhesion could be contributed by hypoxia-mediated reduced CD151 expression. Hypoxia- and normoxia-pretreated cells were harvested and live cells were kept in suspension on a rotator. Cells that aggregated as clusters were counted under microscope. Results showed that hypoxia-pretreated cells had a significantly less number of cells adhered together at 1 hour compared with those precultured under normoxia condition. This phenomenon was even more evident when cells were allowed to aggregate for 3 hours (Fig. 4D).
Hypoxia-inhibited cell adhesion is mediated by CD151. To show that the effect of reduced cell-cell and cell-matrix adhesion capability by hypoxia is mediated via down-regulation of CD151, we directly tested whether the level of CD151 is essential for controlling cell adhesion. First, that the administration of neutralizing antibody significantly inhibited cell adhesion to laminin-coated surface (Fig. 5A ) supports the notion that CD151 is important for cell adhesion. Next, two sets of siRNAs against CD151 were transfected to colon cancer cells and the expression of CD151 protein was determined by Western blot. Transfection of siRNAs against CD151 reduced the expression CD151 protein by 70% to 80% (Fig. 5B). Concomitantly, adhesion ability was significantly inhibited in siCD151 but not guanine-cytosine content-matched scramble siRNA-transfected cells (Fig. 5C).
Reexpression of CD151 in normoxic condition and metastatic site. Four sets of samples containing normal colon, colon cancer, and metastasized cancer were used to examine the expression of CD151 protein. As expected, the level of CD151 in primary cancer was less than its normal counterpart (Fig. 6A ). Interestingly, the levels of CD151 in metastasized tissue were greater than that in the primary site, which was inversely correlated with levels of HIF-1α (Fig. 6A). Immunohistostaining also revealed a strong staining of CD151 in metastasized cancer cells but not surrounding hepatocytes (Supplementary Fig. S3). This result suggested that CD151 is reexpressed by cancer cells once they reach the remote site and supported the notion that CD151 is important for increased cancer metastasis.
Finally, we tested the hypothesis whether suppression of CD151 by HIF-1α is reversible. The result showed that prolonged incubation of hypoxia-pretreated cells in normoxia condition rescued hypoxia-inhibited CD151 expression (Fig. 6B). Concomitantly, the adhesion capacity of hypoxia-pretreated cells was restored when they were subsequently cultured under normoxia condition for 12 or 24 hours (Fig. 6C and D). Taken together, these data show that expression of CD151 in cancer cells is reversibly regulated by oxygen tension, which plays significant roles in controlling cancer cell metastasis.
Metastasis is a complex, multistep cascade of cellular events including detachment from matrix and neighboring epithelial cells, migration through the surrounding stroma, entry into the circulatory system, and finally arrest, extravasation, and growth at a secondary site (25). Previous studies in cancer metastasis primarily focused on the latter events of this process whereas the fundamental question of how cancer cells detach from the highly organized architecture such as tight cell-cell and cell-matrix adhesion was never investigated. In the current study, we showed, for the first time, that expression of CD151 is reduced in cancer cells compared with their normal counterparts. More importantly, we showed that hypoxia markedly reduced the expression of CD151 in cancer cells. Suppression of CD151 expression by HIF-1α inhibited cell-cell adhesion and adhesion to laminin-coated surface. The inhibition was reversible when cells were reexposed to normal oxygen level or grew at a distantly secondary site. These data provide novel information to advance our understanding of the long-standing puzzle of hypoxia-mediated cancer metastasis.
CD151 is one of the most ubiquitously expressed tetraspanins that function as multimolecule organizers in cell-cell and cell-matrix adhesion. Consistent with previous reports (15–17, 19), we showed that there are CD151-positive and CD151-negative cancer cells and the percentage of CD151-positive cells increases in more severe stages. In addition, our results showed that the level of CD151 is significantly lower in cancerous tissues compared with their normal counterparts irrespective of the pathologic status. Recently, Takeda et al. reported that CD151 was not detected in normal colon cancer cell using immunostaining in six tissue array sections (8). This result was not consistent with our current finding that 137 normal colon biopsies were all CD151 positive as detected by both Western blot and immunostaining. The discrepancy was not known. Nevertheless, our current results are consistent with previous findings that CD151 is predominantly expressed in epithelial cells (26, 27).
The expression of CD151 was primarily in the basal and lateral sites of the plasma membrane of epithelial cells in normal colon cells, which showed highly ordered glandular morphology. In contrast, cancer cells showed reduced intensity of staining and most of the CD151-positive staining appeared in the cytoplasm. These results reveal two distinct properties of CD151 in cancer progression. First, as CD151 is critical for cell-cell and cell-matrix adhesion, reduced expression and mislocalization of CD151 may result in poor organization of cancer cells and loosen contact with extracellular matrix and neighboring cells. This in term prepares cancer cells to be “ready” for migrating out of the primary, hypoxic site. Second, as CD151 also plays important roles in integrin-dependent cell migration, reexpression of CD151 to facilitate cell migration becomes critical in cancer cells. In more advanced-stage cancer cells, CD151 will be induced by factors other than HIF-1α to ensure the capability of cell movement. However, due to the inhibitory effect of HIF-1α, the overall level of CD151 in these cancer cells remains lower than that of normal cells (Figs. 1 and 2). Nevertheless, with such level of CD151, it might represent an ideal situation for metastasis because the level of CD151 is enough for migration but never too high for immobilizing cancer cells to extracellular matrix. At present, the factor that induces CD151 expression is not known but is likely to be present in advanced-stage but not earlier-stage cancer cells. Further investigation is warranted to investigate the factor that induces CD151 expression in cancer cells.
Elevated protein levels of HIF-1α in solid cancer have been shown to be associated with tumor malignancy and unfavorable prognosis (28), partly due to HIF-1α-induced increase in metastatic capability. Many genes that contribute to increase cancer metastasis are regulated by HIF-1α; however, almost all genes characterized thus far are involved in the later stages of metastasis (29). Recently, it has been shown that hypoxia-induced lysyl oxidase overexpression is critical for cancer metastasis (30). However, detachment of cancer cells from their surrounding matrix and other cells cannot be explained by overexpression of lysyl oxidase because it is an extracellular matrix protein. Overexpression of lysyl oxidase increases the cross-linking of collagens and elastin, thereby enhancing insoluble matrix deposition and tensile strength (31), which promotes cancer cell invasion and migration to remote sites but not for detaching from the primary site. Herein, we showed that hypoxia-induced CD151 down-regulation in colon cancer leads to a decrease in cell adhesion to laminin-coated matrix. The importance of this finding is to provide the first evidence to show that hypoxia-induced cancer cell metastasis may start from the very first step via reducing cell-matrix adhesion.
Our data also show that metastasized cancer cells expressed greater amount of CD151 than cancer cells in situ. The exact mechanism of this phenomenon is not known, but there are two possible explanations. First, CD151 might be induced during the time of migration in the circulatory system as shown by the reappearance of CD151 in cancer cells after reoxygenation. Reexpression of CD151 in circulating cancer cells may restore adhesive capacity, thus increasing the opportunity for growth in the secondary site. Alternatively, factors synthesized or secreted by the cells in the secondary site might induce the production of CD151 leading to an increase in adhesive ability.
In conclusion, we have shown for the first time that hypoxia down-regulates CD151 expression in cancer cells. Decreased expression of CD151 attenuates cell adhesion ability. In colon cancer tissue in which HIF-1α is elevated, expression of CD151 is inhibited, which may explain the dissociation of glandular morphology of cancer epithelium. More importantly, decreased expression of CD151 in cancer cells may facilitate the initiation of cancer metastasis, a hallmark of tumor malignancy and poor prognosis. Further investigations on disruption of hypoxia-mediated down-regulation of CD151 in the primary site or blocking the restoration of CD151 expression to inhibit the adhesion to the secondary site may provide useful information for improving cancer therapy.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant support: Grants NSC95-2627-B-006-001 and NSC95-3112-B-006-001 from the National Research Program for Genomic Medicine and the National Science Council of Taiwan.
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Received June 26, 2008.
- Revision received August 7, 2008.
- Accepted August 8, 2008.