Since genistein has been reported to inhibit tumor growth [12
], we first examined the effect of genistein on the proliferation of LM8 cells. The results of BrdU incorporation into DNA and DNA measurement revealed that genistein inhibited DNA replication and cell proliferation in cultures of LM8 cells (Figure A and 1C). Of course, there is concern that this inhibition could be due to a cytotoxic effect of genistein on LM8 cells. However, this does not appear to be the case because cells did not detach from the bottom of the plates during the 3-day treatment period. Moreover, the results of the trypan blue exclusion test showed that genistein-treated cells that attached to the bottom of the plates were viable cells (Figure D). Genistein at a higher concentration (100 μM) also inhibited the proliferation of LM8 cells without affecting cell viability (data not shown). These findings indicate that genistein-induced inhibition of cell proliferation was not due to a cytotoxic effect. There are reports on the effect of genistein on the proliferation of normal cells. In cultures of 3T3-L1 fibroblasts, genistein at 50 and 100 μM inhibits cell proliferation [20
]. In cultures of endothelial cells, genistein at lower concentrations (0.001-10 nM) stimulates cell proliferation, while at a higher concentration (1 μM) this chemical inhibits it [21
When LM8 cells are implanted s.c. into the backs of mice, a 100% pulmonary metastatic rate is observed within 4 weeks [6
]. The initial step of metastasis is cell detachment from the primary tumor. Since β-catenin is essential for cadherin-mediated cell-cell adhesion, reduced expression of β-catenin and cadherins at the cell surface is associated with tumor metastasis [22
]. Loss of membranous β-catenin occurs commonly in primary colorectal cancers with metastatic potential and in the corresponding colorectal liver metastases [23
]. Our results of immunofluorescence showed loss of membranous β-catenin in LM8 cells (Figure A, left panel), indicating that cell-cell adhesiveness may be reduced in LM8 cells.
Genistein affects the subcellular localization and expression level of β-catenin in normal cells. For example, treatment of mouse HC11 mammary epithelial cells with genistein increases membranous β-catenin [24
]. In mammary glands of young female rats fed a genistein-containing diet, β-catenin is predominantly localized to ductal epithelial cell membranes [25
] and the level of this protein is higher compared with those fed a control diet [24
]. Therefore, we examined whether genistein affects the subcellular localization and cellular level of β-catenin in LM8 cells. In genistein-treated cells, immunofluorescence staining of β-catenin was observed near the plasma membrane (Figure A, right panel). The results of Western blot showed that genistein-treated cells contained higher levels of β-catenin than untreated cells (Figure B). These findings suggest that genistein may promote intracellular adhesion of LM8 cells. This is the first report to focus on the subcellular localization and cellular level of β-catenin following genistein treatment in osteosarcoma cells.
The next step of metastasis is invasion of tumor cells into basement membranes, which involves distinct events, such as cell motility and MMP expression. Invasive tumors exhibit active migration and high levels of MMPs [9
]. LM8 cells have been reported to exhibit higher invasive potential, higher MMP-2 activity, and higher ALP activity than Dunn cells with no metastatic potential [6
]. In patients with osteosarcoma, high levels of ALP activity in serum are associated with a poor clinical outcome [3
]. In the present study, we found that genistein-treated cells were less invasive and less motile (Figure A and 3B), exhibited lower expression and secretion of MMP-2 (Figure C and 3D), and exhibited lower ALP activity (Figure E) compared with untreated cells. Taken together, our findings suggest that genistein-treated cells might lose metastatic potential.
When β-catenin is translocated to the nucleus, it activates target genes, such as cyclin D1 and c-myc
, and promotes the growth of tumor cells [15
]. In intestinal-type gastric cancer, β-catenin in the nucleus not only plays a role in early tumor growth, but it also functions in initiation of invasive processes [27
]. In osteosarcoma that has metastasized to the lung, β-catenin is detected in the cytoplasm and/or nucleus, whereas in primary osteosarcoma that has not metastasized for more than five years, it is not detected in the cytoplasm and/or nucleus [16
], indicating that the presence of β-catenin in the cytoplasm and/or nucleus is a marker of metastatic potential of osteosarcoma to the lung. Our results of immunofluorescence showed that β-catenin was present in the cytoplasm and/or nucleus of LM8 cells (Figure A, left panel) and that after treatment of LM8 cells with genistein, β-catenin was translocated near to the plasma membrane and was not found in the nucleus (Figure A, right panel). This translocation of β-catenin may result in decreases in proliferative rate and invasive potential of genistein-treated cells. A similar translocation of β-catenin from the nucleus to the plasma membrane has been reported in human HT-29 colon cancer cells treated with thiazolidinedione, a PPARγ ligand [17
Treatment of LM8 cells with genistein induced morphological changes (Figure A) and markedly increased the level of osteocalcin mRNA (Figure B). Moreover, this treatment markedly decreased the formation of multilayer masses (Figure A), suggesting that genistein-treated cells grew with contact inhibition. Contact inhibition is a characteristic of normal cells grown on plastic culture plates, which is lost in cancerous cells [28
]. On the basis of these findings, we concluded that genistein induced the differentiation of LM8 cells. There are a few reports on the association of genistein with cell differentiation. Treatment of mouse B16 melanoma cells with genistein induces morphological changes characteristic of differentiation, such as enlargement of the soma and nuclei together with the formation of dendritic-like structures [12
]. Treatment of human MG 63 osteosarcoma osteoblasts with genistein increases the level of collagen and stimulates the formation of calcification foci [13
Drg-1 is expressed at lower levels in tumors than in normal tissue [19
]. Overexpression of Drg-1
gene in human SW620 metastatic colon cancer cells induces morphological changes that are similar to differentiation-specific changes induced by known differentiation reagents such as tributyrin and a ligand of retinoid X receptor LG268 [30
], thus indicating that Drg-1 plays a role in differentiation of tumor cells. However, there is a contradictory report on the involvement of Drg-1 in cell differentiation. Human hepatocellular carcinomas express a higher level of Drg-1 than non-tumor liver or cirrhotic and benign liver lesions [31
]. Moreover, hepatocellular carcinomas that are moderately and poorly differentiated express a higher level of Drg-1 than those that are well differentiated [31
]. The positive or negative association of Drg-1 with cell differentiation depends on tumor types. Therefore, we examined the effect of genistein on the expression of Drg-1 mRNA by RT-PCR. LM8 cells significantly expressed Drg-1 mRNA. Genistein did not increase the expression of Drg-1 mRNA in LM8 cells (Figure B). Thus, the expression of Drg-1 in LM8 cells appears not to be associated with the induction of cell differentiation.
Drg-1 is located in the cytoplasm of human EJ bladder carcinoma cells [29
], human hepatocellular carcinoma cells [31
], and human prostate cancer cells [32
]. Our results of immunohistochemical staining also reveal that Drg-1 in LM8 cells was located in the cytoplasm (Figure C, left panel). After treatment of LM8 cells with genistein, Drg-1 mostly shifted in localization from the cytoplasm to the nucleus (Figure C, right panel). A study with human U2OS osteosarcoma cells expressing Drg-1 small interfering RNA showed that Drg-1 is associated with the production of osteocalcin [33
]. Taken together, our findings suggest that Drg-1 in the nucleus of genistein-treated cells may play a role in the expression of osteocalcin mRNA.
In postmenopausal Japanese women [11
] and Asian-American women [34
], the risk of breast cancer is inversely associated with isoflavone intake during adult life. Asian-American women who are high consumers of isoflavone (>12.68 mg/1,000 kcal) show a significantly reduced risk of breast cancer compared with those who are low consumers (≤1.79 mg/1,000 kcal) [34
]. However, isoflavones at the levels commonly consumed by non-Asian US women (an average intake equivalent to less than one serving of tofu per week) have little effect on breast cancer risk [35
]. In mice implanted with human 253J B-V bladder cancer cells, dietary genistin (0.14% of the diet), the natural form of genistein in soy, inhibits tumor growth at the primary site and pulmonary metastasis [14
]. In mice implanted with human PC3-M prostate cancer cells, dietary genistein (100 and 250 mg/kg diet) inhibits pulmonary metastasis without affecting tumor growth at the primary site [36