Search

Cancer cells show characteristic gene expression profiles. Recent studies support the potential importance of microRNA (miRNA) expression signatures as biomarkers and therapeutic targets. The membrane-anchored protease regulator RECK is downregulated in many cancers, and forced expression of RECK in tumor cells results in decreased malignancy in animal models. RECK is also essential for mammalian development. In this study, we found that RECK is a target of at least three groups of miRNAs (miR-15b/16, miR-21 and miR-372/373); that RECK mutants lacking the target sites for these miRNA show augmented tumor/metastasis-suppressor activities; and that miR-372/373 are upregulated in response to hypoxia through HIF1alpha and TWIST1, whereas miR-21 is upregulated by RAS/ERK signaling. These data indicate that the hypoxia- and RAS-signaling pathways converge on RECK through miRNAs, cooperatively downregulating this tumor suppressor and thereby promoting malignant cell behavior.

Accumulating evidence indicates that Reversion-inducing cysteine-rich protein with Kazal motifs (RECK), a membrane-anchored matrix metalloproteinase regulator, plays crucial roles in mammalian development and tumor suppression. Its mechanisms of action at the single cell level, however, remain largely unknown. In mouse fibroblasts, RECK is abundant around the perinuclear region, membrane ruffles and cell surface. Cells lacking Reck show decreased spreading, ambiguous anterior-posterior (AP) polarity, and increased speed and decreased directional persistence in migration; these characteristics are also found in transformed fibroblasts and fibrosarcoma cells with low RECK expression. RECK-deficient cells fail to form discrete focal adhesions, have increased levels of GTP-bound Rac1 and Cdc42, and a marked decrease in the level of detyrosinated tubulin, a hallmark of stabilized microtubules. RECK-deficient cells also show elevated gelatinolytic activity and decreased fibronectin fibrils. The phenotype of RECK-deficient cells is largely suppressed when the cells are plated on fibronectin-coated substrates. These findings suggest that RECK regulates pericellular extracellular matrix degradation, thereby allowing the cells to form proper cell-substrate adhesions and to maintain AP polarity during migration; this mechanism is compromised in malignant cells.

Histone deacetylase (HDAC) inhibitors are expected to be effective for refractory cancer because their mechanism of action differs from that of conventional antineoplastic agents. In this study, we examined the effect of the HDAC inhibitor FK228 on malignant melanoma, as well as its molecular mechanisms. FK228 was highly effective against melanoma compared with other commonly used drugs. By comparing the gene expression profiles of melanoma cells and normal melanocytes, we defined a subset of genes specifically upregulated in melanoma cells by FK228, which included Rap1, a small GTP-binding protein of the Ras family. The expression of Rap1 mRNA and protein increased in FK228-treated melanoma cells in both a dose- and a time-dependent manner. A decrease in the phosphorylation of c-Raf, MEK1/2, and ERK1/2 was accompanied by an increase in Rap1 expression in both FK228-treated and Rap1-overexpressing cells. Inhibition of Rap1 upregulation by small interfering RNA (siRNA) abrogated the induction of apoptosis and suppression of ERK1/2 phosphorylation in FK228-treated melanoma cells. These results indicate that the cytotoxic effects of FK228 are mediated via the upregulation of Rap1. Furthermore, we found that Rap1 was overexpressed and formed a complex with B-Raf in melanoma cell lines with a V599E mutation of B-Raf. The siRNA-mediated abrogation of Rap1 overexpression increased the viability of these cells, suggesting that Rap1 is also an endogenous regulator of Ras-MAP kinase signaling in melanomas.

The membrane-anchored MMP-regulator RECK is down regulated in many solid tumors; the extent of RECK down regulation correlates with poor prognosis. Forced expression of RECK in tumor cells results in suppression of angiogenesis, invasion, and metastasis. Studies on the roles and the mechanisms of regulation of the RECK gene during normal development may therefore yield important insights into how the malignant behaviors of tumor cells arise and how they can be controlled. Our previous studies indicate that mice lacking RECK die around E10.5 with reduced tissue integrity. In the present study, we have found that in later stage wild-type embryos, RECK is abundantly expressed in skeletal muscles, especially in the areas where the myoblast differentiation factor MRF4 is expressed. Consistent with this finding, the RECK-promoter is activated by MRF4 in cultured cells. In contrast, a myoblast determination factor MyoD suppresses the RECK-promoter. Myoblastic cells lacking RECK expression give rise to myotubes at higher efficiency than the cells expressing RECK, indicating that RECK suppresses myotube formation. These findings suggest that MyoD down regulates RECK to facilitate myotube formation, whereas MRF4 up regulates RECK to promote other aspects of myogenesis that require extracellular matrix integrity.

The mutant of Ras protein with serine to asparagine mutation at residue 17 (Ras-17N) is known to interfere with the signaling function of the wild-type Ras protein by sequestering its guanine-nucleotide exchange factors (GEFs). The similar mutant of another Ras family protein Rap1 (Rap1-17N) fails to effectively interfere with the interaction between the wild-type Rap1 and one of its GEFs, C3G, in vitro. In the present study, we have attempted to isolate Rap1 mutants with increased affinity for C3G using random mutagenesis and yeast two-hybrid screening. Based on the pattern of mutations found among these mutants, we could design a potent C3G-binder, named Rap1-AGE, harboring mutations in three sites (17A, 29G, and 117E). The association of Rap1-AGE with C3G in the cells was confirmed by co-immunoprecipitation experiments. The ability of Rap1-AGE to inhibit C3G-mediated Rap1-activation and cell spreading was also demonstrated. On the other hand, Rap1 activation mediated by two other GEFs, Epac and smgGDS, was not inhibited by Rap1-AGE. These results suggest that Rap1-AGE acts as a dominant interfering factor against C3G and serves as a useful tool in analyzing the roles of C3G-Rap1 signaling pathway in various biological processes.

In COS cells, Ral GDP dissociation stimulator (RalGDS)-induced Ral activation was stimulated by RasG12V or a Rap1/Ras chimera in which the N-terminal region of Rap1 was ligated to the C-terminal region of Ras but not by Rap1G12V or a Ras/Rap1 chimera in which the N-terminal region of Ras was ligated to the C-terminal region of Rap1, although RalGDS interacted with these small GTP-binding proteins. When RasG12V, Ral and the Rap1/Ras chimera were individually expressed in NIH3T3 cells, they localized to the plasma membrane. Rap1Q63E and the Ras/Rap1 chimera were detected in the perinuclear region. When RalGDS was expressed alone, it was abundant in the cytoplasm. When coexpressed with RasG12V or the Rap1/Ras chimera, RalGDS was detected at the plasma membrane, whereas when coexpressed with Rap1Q63E or the Ras/Rap1 chimera, RalGDS was observed in the perinuclear region. RalGDS which was targeted to the plasma membrane by the addition of Ras farnesylation site (RalGDS-CAAX) activated Ral in the absence of RasG12V. Although RalGDS did not stimulate the dissociation of GDP from Ral in the absence of the GTP-bound form of Ras in a reconstitution assay using the liposomes, RalGDS-CAAX could stimulate it without Ras. RasG12V activated Raf-1 when they were coexpressed in Sf9 cells, whereas RasG12V did not affect the RalGDS activity. These results indicate that Ras recruits RalGDS to the plasma membrane and that the translocated RalGDS induces the activation of Ral, but that Rap1 does not activate Ral due to distinct subcellular localization.

A cDNA (termed CT124) encoding a carboxyl-terminal fragment of the human homeobox protein MSX-2 was found to induce flat reversion when expressed in v-Ki-ras-transformed NIH3T3 cells. Although the expression of endogenous MSX-2 gene is low in most of the normal adult tissues examined, it is frequently activated in carcinoma-derived cell lines. Likewise, the gene is inactive in NIH3T3 cells but is transcriptionally activated after transformation by v-Ki-ras oncogene, suggesting that the intact MSX-2 may play a positive, rather than suppressive, role in cell transformation. To test this possibility, we isolated a near full-length human MSX-2 cDNA and tested its activities in two cell systems, i.e. fibroblast and myoblast. In NIH3T3 fibroblasts, although the gene by itself failed to confer a transformed phenotype, antisense MSX-2 cDNA as well as truncated CT124 cDNA interfered with the transforming activities of v-Ki-ras oncogene. In C2C12 myoblasts, MSX-2 was found to suppress MyoD gene expression, as do activated ras oncogenes, under certain culture conditions, and CT124 was found to inhibit the activities of both MSX-2 and ras in this system as well. Our findings not only suggest that CT124 may act as a dominant suppressor of MSX-2 but also raise the possibility that MSX-2 gene may be an important downstream target for the Ras signaling pathways.