Portal de Periódicos da CAPES

Identification of Metastasis-related Genes in a Mouse Model Using a Library of Randomized Ribozymes

2004; Elsevier BV; Volume: 279; Issue: 37 Linguagem: Inglês

10.1074/jbc.c400313200

1083-351X

Eigo Suyama, Renu Wadhwa, Kamaljit Kaur, Makoto Miyagishi, Sunil C. Kaul, Hiroaki Kawasaki, Kazunari Taira,

Chemical Synthesis and Analysis

Libraries of randomized ribozymes have considerable potential as tools for the identification of functional genes critically involved in a biological phenotype of interest in vitro. We have used a ribozyme library in an in vivo mouse model to identify genes related to metastasis. We injected weakly metastatic melanoma cells that had been treated with the library intravenously into mice. We then isolated ribozymes that accelerated metastasis from pulmonary tumors that had developed from metastasizing cells. As candidates for metastasis-related genes that were targets of the isolated ribozymes, we identified five unknown and three known genes: stromal interaction molecule 1 (STIM1), polymerase γ2 accessory subunit (Polg2), and cytochrome P450, family 2, subfamily d, polypeptide 22 (Cyp2d22). Repression of four of these by small interfering RNAs indeed resulted in the accelerated mobility of cells in in vitro scratch-wound assay. The further characterization of these candidate genes would provide clues to the complex mechanism(s) of metastasis. Libraries of randomized ribozymes have considerable potential as tools for the identification of functional genes critically involved in a biological phenotype of interest in vitro. We have used a ribozyme library in an in vivo mouse model to identify genes related to metastasis. We injected weakly metastatic melanoma cells that had been treated with the library intravenously into mice. We then isolated ribozymes that accelerated metastasis from pulmonary tumors that had developed from metastasizing cells. As candidates for metastasis-related genes that were targets of the isolated ribozymes, we identified five unknown and three known genes: stromal interaction molecule 1 (STIM1), polymerase γ2 accessory subunit (Polg2), and cytochrome P450, family 2, subfamily d, polypeptide 22 (Cyp2d22). Repression of four of these by small interfering RNAs indeed resulted in the accelerated mobility of cells in in vitro scratch-wound assay. The further characterization of these candidate genes would provide clues to the complex mechanism(s) of metastasis. Hammerhead ribozymes (referred to herein after as ribozymes) that specifically cleave the transcripts of particular genes can be produced when the nucleotide sequences of the substrate-recognition arms are designed such that they are complementary to the sequences of individual target mRNAs. Such ribozymes have been used successfully to interfere with the expression of specific genes via cleavage of the respective target mRNAs in mammalian cells (1Uhlenbeck O.C. Nature. 1987; 328: 596-600Crossref PubMed Scopus (873) Google Scholar, 2Haseloff J. Gerlach W.L. Nature. 1988; 334: 585-591Crossref PubMed Scopus (967) Google Scholar, 3Zhou D.M. Taira K. Chem. Rev. 1998; 98: 991-1026Crossref PubMed Scopus (215) Google Scholar, 4Kawasaki H. Eckner R. Yao T.P. Taira K. Chiu R. Livingston D.M. Yokoyama K.K. Nature. 1998; 393: 284-289Crossref PubMed Scopus (302) Google Scholar, 5Gasteland R.F. Cech T.R. Atkins J.F. The RNA World. 2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1999Google Scholar, 6Krupp G. Gaur R.K. RIBOZYME: Biochemistry and Biotechnology. Eaton Publishing, Natick, MA2000Google Scholar). The use of a library of ribozymes with randomized substrate-recognition arms allowed the identification of genes that are involved directly in certain phenomena, such as the transformation of NIH3T3 fibroblasts, anchorage-independent cell growth, tumor necrosis factor-α-induced apoptosis and tumor invasion (7Kruger M. Beger C. Li Q.X. Welch P.J. Tritz R. Leavitt M. Barber J.R. Wong-Staal F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 8566-8571Crossref PubMed Scopus (92) Google Scholar, 8Li Q.X. Robbins J.M. Welch P.J. Wong-Staal F. Barber J.R. Nucleic Acids Res. 2000; 28: 2605-2612Crossref PubMed Scopus (47) Google Scholar, 9Welch P.J. Marcusson E.G. Li Q.X. Beger C. Kruger M. Zhou C. Leavitt M. Wong-Staal F. Barber J.R. Genomics. 2000; 66: 274-283Crossref PubMed Scopus (55) Google Scholar, 10Beger C. Pierce L.N. Kruger M. Marcusson E.G. Robbins J.M. Welcsh P. Welch P.J. Welte K. King M.C. Barber J.R. Wong-Staal F. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 130-135Crossref PubMed Scopus (200) Google Scholar, 11Kawasaki H. Onuki R. Suyama E. Taira K. Nat. Biotechnol. 2002; 4: 376-380Crossref Scopus (76) Google Scholar, 12Kawasaki H. Taira K. EMBO Rep. 2002; 3: 443-450Crossref PubMed Scopus (55) Google Scholar, 13Suyama E. Kawasaki H. Kasaoka T. Taira K. Cancer Res. 2003; 63: 119-124PubMed Google Scholar, 14Suyama E. Kawasaki H. Nakajima M. Taira K. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5616-5621Crossref PubMed Scopus (51) Google Scholar, 15Nelson L.D. Suyama E. Kawasaki H. Taira K. Targets. 2003; 2: 191-200Crossref Google Scholar). The metastatic spread of tumor cells around the body is a key target for cancer therapy (16Tannock I.F. Hill R.P. The Basic Science of Oncology. 3rd Ed. The McGraw-Hill Companies, New York1998Google Scholar, 17Holland J.F. Frei E. Cancer Medicine. 5th Ed. B. C. Decker, Hamilton, Canada2000Google Scholar). An experimental assay of metastasis in vivo that involves intravenous injection of tumor cells into mice was developed by Fidler (18Fidler I.J. Nat. New Biol. 1973; 242: 148-149Crossref PubMed Scopus (1277) Google Scholar) for the analysis of the metastatic properties of melanoma cells (Fig. 1). In this assay, cells of interest are injected into the tail vein of mice and allowed to proliferate and/or metastasize. After several weeks, pulmonary metastases are observed in the case of strongly metastatic melanoma cells, while relatively few metastases are formed after injection of cells with limited metastatic potential. This assay allows the selection of strongly metastatic cells from a heterogeneous population of weakly metastatic tumor cells (18Fidler I.J. Nat. New Biol. 1973; 242: 148-149Crossref PubMed Scopus (1277) Google Scholar, 19Clark E.A. Golub T.R. Lander E.S. Hynes R.O. Nature. 2000; 406: 532-535Crossref PubMed Scopus (1309) Google Scholar). Thus, the assay allows not only the study of the metastatic characteristics of melanoma cells but also the isolation of strongly metastatic cells. In the present study, we used this mouse model and mouse B16F0 melanoma cells, a weakly metastatic melanoma cell line, to investigate the applicability of a library of ribozymes in vivo, to isolate ribozymes that converted weakly metastatic melanoma cells to strongly metastatic cells by the apparent disruption of expression of specific gene(s), and to identify genes related to metastasis. Construction of a Ribozyme Library—A library of ribozymes was constructed based on the retrovirus expression vector pMXpuro (20Morita S. Kojima T. Kitamura T. Gene Ther. 2000; 7: 1063-1066Crossref PubMed Scopus (1377) Google Scholar). First, fragments of DNA carrying randomized hammerhead ribozymes were obtained by PCR amplification using oligonucleotide DNAs as follows, a sense primer (5′-TCC CCG GTT CGA AAC CGG GCA-3′), an antisense primer (5′-GCT TGC ATG CCT GCA GGT CGA CGC GAT AGA AAA AAA GAT ATC CGG GGT-3′), and a template (5′-TCC CCG GTT CGA AAC CGG GCA CTA CAA AAA CCA ACT TTN NNN NNNN CTG ATG AGG CCG AAA GGC CGA AAN NNN NNNG GTA CCC CGG ATA TCT TTT TTT-3′, where N represents A, T, G, or C). The fragments were cloned into a plasmid vector, pGEM-T (Promega, Madison, WI), then were digested with Csp45I and KpnI, and were again cloned into pPUR-KE (21Koseki S. Tanabe T. Tani K. Asano S. Shioda T. Nagai Y. Shimada T. Ohkawa J. Taira K. J. Virol. 1999; 73: 1868-1877Crossref PubMed Google Scholar). Next, fragments encoding tRNAVal-fused randomized ribozymes were obtained by digestion of the library of ribozymes based on pPUR-KE with EcoRI and BamHI. Finally, the fragments were inserted into the EcoRI and BamHI sites in pMXpuro. Cell Culture and Retroviral Infection—B16F0 melanoma cells (number CRL-6322; ATCC, Manassas, VA) were cultured in Dulbecco's modified Eagle's medium (Sigma) supplemented with 10% fetal bovine serum (Invitrogen) and an antibiotics mixture (Invitrogen). Retroviral preparation and infection with the retroviral vector pMXpuro and the packaging cell line Plat-E (kindly provided by Professor Toshio Kitamura, Institute of Medical Science, University of Tokyo, Tokyo, Japan) were performed as described elsewhere (14Suyama E. Kawasaki H. Nakajima M. Taira K. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5616-5621Crossref PubMed Scopus (51) Google Scholar, 20Morita S. Kojima T. Kitamura T. Gene Ther. 2000; 7: 1063-1066Crossref PubMed Scopus (1377) Google Scholar). Infected cells were treated with medium that contained puromycin (1 μg/ml, Sigma) for 4 weeks. Experimental Metastasis Assay—5 × 105 of B16F0 cells, infected with retroviruses carrying a ribozyme library or not, were suspended in phosphate-buffered saline (PBS 1The abbreviations used are: PBS, phosphate-buffered saline; RT, reverse transcriptase; siRNA, small interfering RNA; STIM1, stromal interaction molecule 1 gene; Polg2, polymerase γ2 accessory subunit gene; Cyp2d22, cytochrome P450, family 2, subfamily d, polypeptide 22 gene.; Takara Bio, Shiga, Japan) and intravenously injected into the lateral tail vein of C57BL/6NCrj mice (Charles River Japan, Yokohama, Japan). Two weeks after injection, lungs were removed from the mice and were washed with ice-cold PBS. The pulmonary metastases on the surface of the lungs were observed and counted. Then the nodules were removed, minced, and cultured in dishes. RT-PCR Analysis and Ribozyme Rescue—Total RNA was prepared from cells using the Isogen reagent (Nippon Gene, Toyama, Japan) according to the manufacturer's protocol. cDNA was synthesized using 2 μg of total RNA and Moloney murine leukemia virus reverse transcriptase (Promega). Using the cDNA as templates, PCR amplification was performed with primers as follows: ribozyme sense, 5′-TCC CCG GTT CGA AAC CGG GCA-3′; ribozyme antisense, 5′-GCT TGC ATG CCT GCA GGT CGA CGC GAT AGA AAA AAA GAT ATC CGG GGT-3′; β-actin sense, 5′-GCA CGG CAT CGT CAC CAA CT-3′; and β-actin antisense, 5′-AAG GCT GGA AGA GTG CCT CA-3′. The amplified fragments of DNA were confirmed by agarose gel electrophoresis. Fragments carrying ribozymes that were derived from the metastatic nodules were cloned into the pGEM-T vector (Promega) and sequenced. Search of DNA Data Bases—By searching of mouse cDNA databases with the BLAST program, identification of target genes of the selected ribozymes was performed (22Altschul S.F. Madden T.L. Schaffer A.A. Zhang J. Zhang Z. Miller W. Lipman D.J. Nucleic Acids Res. 1997; 25: 3389-3402Crossref PubMed Scopus (60233) Google Scholar) (www3.ncbi.nlm.nih.gov/BLAST/). Parameters for the data base searches were set to optimize searches for short, nearly exact sequences. Construction and Transfection of siRNA Expression Plasmid—Plasmid vectors encoding siRNAs for four of the eight identified targets were constructed as described (23Wadhwa R. Kaul S.C. Miyagishi M. Taira K. Rev. Mutat. Res. 2004; (in press)PubMed Google Scholar). Two target sites for each of the genes were selected using an algorithm (www.igene-therapeutics.co.jp). Transfections were performed using LipofectAMINE Plus (Invitrogen). Typically, 3 μg of plasmid DNA was used per 80% confluent 6-cm dish culture. Transfected B16F0 cells were selected in medium containing puromycin (2.5 μg/ml) for 48 h and then subjected to RT-PCR analysis and in vitro scratch-wound assay. The expression of targeted genes was examined by gene-specific primers, STIM1 (sense: 5′-GGG AAG ACC TCA ATT ACC A-3′ and antisense: 5′-CAG CTG CAG CTT CTG CCT GT-3′), Polg-2 (sense: 5′-GGG AAG CAA ACT TTA CTA CAA CT-3′ and antisense: 5′-ATC CAA AGC GAC CTT AAT AG-3′), and Cyp2d22 (sense: 5′-AAG GAG GAA GCT GGA TTC CTA CC-3′ and antisense: 5′-CCC TAT GAC TTC ATC GAT TT-3′), on standard condition for RT-PCR (25 cycles of 94 °C for 30 s, –54 °C for 60 s, –72 °C for 60 s). Scratch-wound Assay—B16F0 cells that had been treated with siRNA expression plasmid were allowed to form a monolayer on a fibronectin (10 μg/ml)-coated dish surface. A wound was made in the monolayer of cells by completely scratching the cells in a line with a pipette tip. Cells were washed a few times with PBS to remove cell debris and fed with fresh medium as described (14Suyama E. Kawasaki H. Nakajima M. Taira K. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5616-5621Crossref PubMed Scopus (51) Google Scholar). The time of the scratching wound was designated as time 0. Cells were allowed to proliferate and migrate into the wound during the next 30 h. Migration of cells into the wound was recorded under a phase contrast microscope with a 10× phase objective. We constructed a library of ribozymes that was based on the retroviral expression vector pMXpuro (20Morita S. Kojima T. Kitamura T. Gene Ther. 2000; 7: 1063-1066Crossref PubMed Scopus (1377) Google Scholar). Retroviruses carrying randomized ribozymes were prepared in the packaging cell line Plat-E as described elsewhere (14Suyama E. Kawasaki H. Nakajima M. Taira K. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5616-5621Crossref PubMed Scopus (51) Google Scholar, 20Morita S. Kojima T. Kitamura T. Gene Ther. 2000; 7: 1063-1066Crossref PubMed Scopus (1377) Google Scholar). Then we infected weakly metastatic B16F0 cells with the retroviruses for the selection of ribozymes that enhanced the metastatic properties of the cells (Fig. 2, A and B). After antibiotic (puromycin) selection, we confirmed the expression of randomized ribozymes in cells by RT-PCR analysis with primers specific for the parent ribozyme (Fig. 2C). Then we injected 5 × 105 control or randomized ribozymes harboring B16F0 cells intravenously into the lateral tail vein of individual C57BL/6N mice. Cells with enhanced metastatic properties were selected by their isolation from lung as pulmonary tumors. Only a few pulmonary metastases were detected in the mice injected with parent B16F0 cells as reported earlier (18Fidler I.J. Nat. New Biol. 1973; 242: 148-149Crossref PubMed Scopus (1277) Google Scholar, 19Clark E.A. Golub T.R. Lander E.S. Hynes R.O. Nature. 2000; 406: 532-535Crossref PubMed Scopus (1309) Google Scholar). Fig. 3, A and B, show the difference in terms of the resultant pulmonary metastases between cells that had been treated with the ribozyme library and untreated cells, 2 weeks after injection. The number of metastatic melanoma nodules derived from library-treated cells was clearly higher than those derived from untreated cells. Thus, certain ribozymes in the library appeared to have promoted the pulmonary metastasis of B16F0 cells (Fig. 3, A and B).Fig. 3Selection in vivo of ribozymes that enhanced the metastatic properties of B16F0 melanoma cells, using a mouse model and cells treated with the library of randomized ribozymes. A, isolation of pulmonary metastases. B, number of pulmonary tumors isolated after the injection of B16F0 cells that had been treated with randomized ribozymes (RRz) or tRNAVal alone (RRz minus).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Next, we removed the pulmonary nodules of melanoma cells derived from ribozyme harboring cells (Fig. 3A), and we collected cells from these nodules by mincing and treating them with trypsin. The resultant cells were cultured, and RNA was isolated from these cells. Then ribozymes were amplified by RT-PCR strategy and cloned into plasmid vector (15Nelson L.D. Suyama E. Kawasaki H. Taira K. Targets. 2003; 2: 191-200Crossref Google Scholar) (Fig. 1), and the results of sequencing of the ribozymes indicated that the selected ribozymes represented eight types of ribozyme that contained specific substrate-recognition arms (Table I). By this way, we recovered ribozymes that appeared to have raised the metastatic potential of B16F0 cells.Table ISelected ribozymes and their target genes Open table in a new tab We then identified their target genes by searching data bases with the BLAST program, using settings that optimized searches for short, nearly exactly matching sequences (22Altschul S.F. Madden T.L. Schaffer A.A. Zhang J. Zhang Z. Miller W. Lipman D.J. Nucleic Acids Res. 1997; 25: 3389-3402Crossref PubMed Scopus (60233) Google Scholar) (www3.ncbi.nlm.nih.gov/BLAST/) (Table I). Among the selected ribozymes, our search indicated that ribozyme A targeted the transcript of the gene known as stromal interaction molecule 1 (STIM1). STIM1 is a gene for a transmembrane glycoprotein. This gene is located at human chromosome region 11p15.5, which is involved in tumorigenesis. Moreover, overexpression of STIM1 causes growth arrest and cell death in several lines of cells (24Sabbioni S. Barbanti-Brodano G. Croce C.M. Negrini M. Cancer Res. 1997; 57: 4493-4497PubMed Google Scholar, 25Manji S.S. Parker N.J. Williams R.T. van Stekelenburg L. Pearson R.B. Dziadek M. Smith P.J. Biochim. Biophys. Acta. 2000; 1481: 147-155Crossref PubMed Scopus (200) Google Scholar). These earlier observations indicate that STIM1 functions as a tumor suppressor. Therefore, our identification of a ribozyme that targets STIM1 in our assay of metastasis by cells treated with a ribozyme library seems eminently reasonable. Our data base search also identified genes whose functions have not yet been well characterized, such as the sequence AW551984, expressed in the mouse, which contains a VWA domain that seems to mediate adhesion of eukaryotic cells (a target for ribozyme B; GenBank™ accession number NM_178737.1; Table I). We next constructed siRNA expression vectors against four (STIM1, AW551984, Polg2, and Cyp2d22) out of the eight identified genes and examined their functional involvement in metastasis by carrying out a scratch-wound assay that is commonly used to study the ability of cells to migrate (26Magdalena J. Millard T.H. Machesky L.M. J. Cell Sci. 2003; 116: 743-756Crossref PubMed Scopus (99) Google Scholar, 27Ettenson D.S. Gotileb A.I. Arterioscler. Thromb. Vasc. Biol. 1995; 15: 515-521Crossref PubMed Scopus (29) Google Scholar), a reliable parameter for metastasis. We prepared two different siRNA expression vectors for each of target genes (Table II). Cells harboring the expression plasmid were selected in puromycin supplemented medium. We examined suppression of the expression of target genes and confirmed reduced levels of expression of three genes, STIM1, Polg2, and Cyp2d22 (unfortunately, AW551984 transcript could not be detected by RT-PCR analysis) (Fig. 4A). Then we subjected the cells to the scratch-wound assay (Fig. 4B). Targeting of all the four genes led to significant acceleration of migration, which is one of the properties of metastatic cells. AW551984- and Polg2-targeted B16F0 cells moved more than 90% of the width of the wound in 30 h, whereas control cells, which had been treated with empty vector, only initiated to migrate into the wound. Parallel observation on STIM1- and Cyp2d22-targeted cells revealed about 50% movement into the wound (Fig. 4B). Although siRNA-induced suppression of AW551984 could not be confirmed due to technical difficulty in RT-PCR analysis, the change in phenotype (increased migration) was obtained in two independent experiments. Taken together, these four genes appeared to have a role in control of the metastatic properties of B16F0 melanoma cells. Further investigation of genes identified in this study would provide important clues to the complex mechanism(s) of metastasis.Table IIsiRNA target sequenceGeneTarget siteSequenceSTIM1Site 1GGGAAGACCTCAATTACCASite 2GGCCAAGAAGACATTTATAAW551984Site 1GAACAAACTTCTATCTTTASite 2AGTCAAACTTGTAAATAGAPolg2Site 1GAAGCAAACTTTACTACAASite 2GGGAAAGGAGCCAATAGAACyp2d22Site 1GAAGGAAACTCTTAAGGAASite 2GCGCCGAGTACAACAGGAA Open table in a new tab To the best of our knowledge, this is the first demonstration of the applicability of a ribozyme library to the identification of genes using animal model. To date, such a potential has been demonstrated only in cultured cells (7Kruger M. Beger C. Li Q.X. Welch P.J. Tritz R. Leavitt M. Barber J.R. Wong-Staal F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 8566-8571Crossref PubMed Scopus (92) Google Scholar, 8Li Q.X. Robbins J.M. Welch P.J. Wong-Staal F. Barber J.R. Nucleic Acids Res. 2000; 28: 2605-2612Crossref PubMed Scopus (47) Google Scholar, 9Welch P.J. Marcusson E.G. Li Q.X. Beger C. Kruger M. Zhou C. Leavitt M. Wong-Staal F. Barber J.R. Genomics. 2000; 66: 274-283Crossref PubMed Scopus (55) Google Scholar, 10Beger C. Pierce L.N. Kruger M. Marcusson E.G. Robbins J.M. Welcsh P. Welch P.J. Welte K. King M.C. Barber J.R. Wong-Staal F. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 130-135Crossref PubMed Scopus (200) Google Scholar, 11Kawasaki H. Onuki R. Suyama E. Taira K. Nat. Biotechnol. 2002; 4: 376-380Crossref Scopus (76) Google Scholar, 12Kawasaki H. Taira K. EMBO Rep. 2002; 3: 443-450Crossref PubMed Scopus (55) Google Scholar, 13Suyama E. Kawasaki H. Kasaoka T. Taira K. Cancer Res. 2003; 63: 119-124PubMed Google Scholar, 14Suyama E. Kawasaki H. Nakajima M. Taira K. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5616-5621Crossref PubMed Scopus (51) Google Scholar, 15Nelson L.D. Suyama E. Kawasaki H. Taira K. Targets. 2003; 2: 191-200Crossref Google Scholar, 28Onuki R. Bando Y. Suyama E. Katayama T. Kawasaki H. Baba T. Tohyama M. Taira K. EMBO J. 2004; 23: 959-968Crossref PubMed Scopus (143) Google Scholar, 29Kuwabara T. Hsieh J. Nakashima K. Taira K. Gage F.H. Cell. 2004; 116: 779-793Abstract Full Text Full Text PDF PubMed Scopus (387) Google Scholar). It should now be possible to identify important functional genes in vivo using randomized ribozyme libraries. We thank Dr. Toshio Kitamura (Institute of Medical Science, University of Tokyo) for supplying the retroviral vector pMXpuro and the packaging cell line Plat-E, as well as Drs. Motowo Nakajima and Tatsuhiko Kasaoka (Tsukuba Research Institute, Novartis Pharma, Tsukuba Science City) for critical comments and helpful discussions.