Identification of differentially expressed genes in isogenic highly metastatic and poorly metastatic cell lines of R3230AC rat mammary adenocarcinoma.

Tumour metastasis occurs as a result of a cascade of events including alterations in the expression of various genes. The identification of such genes is essential to understanding formation of metastasis. In a previous study, highly metastatic (LN4.D6) and poorly metastatic (CAb.D5) cell lines were obtained from the rat mammary adenocarcinoma cell line R3230AC. Subtractive hybridization was used to identify differentially expressed genes between these two cell lines. We identified eight cDNA clones in CAb.D5 and six cDNA clones in LN4.D6 that were differentially expressed. One of the cDNA clones in each cell line had no homology with known sequences. Expression patterns of these differentially expressed genes were examined in a pair of rat mammary and prostate adenocarcinoma cell lines. Compared with cell lines examined, cDNA FF-10 was only expressed in CAb.D5; however, cDNA RB-8, RE-1, RF-5 were only expressed in the highly metastatic LN4.D6. No correlation was observed between expression patterns of the differentially expressed genes and metastatic potential of these cells. However, differential expression of genes, especially cytokeratins (CK8 and CK5) and collagens (III and IV) between highly metastatic and low metastatic rat mammary adenocarcinoma cell lines might initiate further investigation of these genes in metastatic process.

Phosphorylation of human N-myristoyltransferase by N-myristoylated SRC family tyrosine kinase members.

N-Myristoyltransferase (NMT) is an essential eukaryotic enzyme that catalyzes the cotranslational and/or posttranslational transfer of myristate to the amino terminal glycine residue of a number of important proteins especially the non-receptor tyrosine kinases whose activity is important for tumorigenesis. Human NMT was found to be phosphorylated by non-receptor tyrosine kinase family members of Lyn, Fyn and Lck and dephosphorylated by the Ca(2+)/calmodulin-dependent protein phosphatase, calcineurin. Deletion of 149 amino acids from the N-terminal end resulted in the absence of phosphorylation suggesting that the phosphorylation sites are located in the N-terminal end of NMT. Furthermore, a site-directed mutagenesis study indicated that substitution of tyrosine 100 with phenylalanine served NMT as a poor substrate for the Lyn kinase. A synthetic peptide corresponding to the amino-terminal region encompassing tyrosine 100 of NMT served as a good substrate for the Lyn and Fyn kinases. Our studies also indicated that NMT was found to interact with Lyn through its N-terminal end in a phosphorylation-dependent manner. This is the first study demonstrating the cross-talk between NMT and their myristoylated protein substrates in signaling pathways.

Isogloboside biosynthesis in metastatic R3230AC cells results from a decreased GM3 synthase activity.

We have determined that the production of a metastasis-associated neutral glycosphingolipid, isogloboside (iGb(4)Cer, GalNAcbeta1-3Galalpha1-3Galbeta1-4Glcbeta1-O-ceramide) is associated with the loss of G(M3) synthase activity. Assays for neutral glycosphingolipid-forming glycosyltransferases in cells producing various levels of iGb(4)Cer revealed no consistent differences that could account for the difference in iGb(4)Cer biosynthesis. However, comparison of the activity of G(M3) synthase in homogenates of these two cell types revealed that cells that did not synthesize iGb(4)Cer had activity significantly greater than that of cells possessing this antigen. Furthermore, somatic cell hybrids generated using clones of the iGb(4)Cer -producing and nonproducing cell lines lacked iGb(4)Cer while possessing high levels of G(M3) synthase activity. When iGb(4)Cer-producing cells were transfected with a G(M3) synthase expression vector, all of the resultant clones were negative for iGb(4)Cer production. The results of these studies clearly show that the presence of G(M3) synthase prevents the formation of iGb(4)Cer in these cells.

N-myristoyltransferase.

Myristoylation refers to the co-translational addition of a myristoyl group to an amino-terminal glycine residue of a protein by an ubiquitously distributed enzyme myristoyl-CoA:protein N-myristoyltransferase (NMT, EC 2.3.1.97). This review describes the basic enzymology, molecular cloning and regulation of NMT activity in various pathophysiological processes such as colon cancer and diabetes.

Isoglobotetraosylceramide is a marker for highly metastatic cells in rat mammary adenocarcinomas.

We have previously identified a neutral glycolipid antigen which appears to be a surface antigenic marker for the metastatic subpopulation in the R3230AC rat mammary adenocarcinoma (S.A. Carlsen, M. Barry, and K. Newton, Clin. Exp. Metastasis, 8: 141-151, 1990). In this article we describe the structural characterization of this glycolipid antigen. The sequence of the sugars in the saccharide portion of the molecule was determined by specific glycosidase cleavage and further confirmed by mass spectroscopic analysis. The nature of the linkages between the monosaccharide units was determined by methylation analysis. The final structure was confirmed by NMR analysis and found to be isoglobotetraosylceramide (GalNAc beta 1-3Gal alpha 1-3Gal beta 1-4Gle beta 1-O-ceramide). We also present evidence that the cells marked by this antigen have a higher metastatic potential than the cells lacking this glycolipid as measured by the formation of lung colonies after i.v. injection of the cells into the tail vein of the rat. Furthermore, isoglobotetraosylceramide seems to play a direct role in the metastatic process since the blocking of exposed antigen with monoclonal antibodies, or their Fab fragments, results in a highly significant decrease in lung colony formation.

The identification of a neutral glycosphingolipid antigenic marker for metastatic cells in the R3230AC rat mammary adenocarcinoma.

The process of metastasis is a complex process involving numerous steps, and it is thought that cells able to complete all of these steps and form metastatic foci form a unique subpopulation of cells within the tumor. To study this metastatic subpopulation, a marker for the metastatic cells is required. We have previously described the enrichment of soybean agglutinin binding cells in tumor populations enriched for lymphatic metastasis [1]. In this study we provide evidence that the cell-surface structure binding soybean agglutinin is a neutral glycosphingolipid. Using monoclonal antibodies generated against this glycolipid, highly metastatic tumor populations were depleted of cells containing this glycolipid. These depleted cell populations were found to be equally tumorigenic to that of the untreated population but were much less metastatic. These results suggest that this glycolipid may be a useful marker for metastatic cells.

Involvement of soybean agglutinin binding cells in the lymphatic metastasis of the R3230AC rat mammary adenocarcinoma.

Many human tumors, such as those of the breast, metastasize initially via the lymphatics. The tumor cell surface is believed to play a critical role in this process. To study the cell surface properties involved in dissemination, the poorly metastasizing R3230AC rat mammary adenocarcinoma was enriched for metastasizing cells by excising rare lymph node metastases arising after the s.c. injection of 10(6) cells and reinjecting these cells into another series of rats. By repeated enrichment cycles, the frequency of lymphatic metastasis was increased from 10 to 60-100% of the animals given injections. Fluorescein-conjugated lectins were used to probe the tumor cell surface. It was found that the percentage of cells in the population able to bind high levels of the lectin, soybean agglutinin (SBA), increased from 11 to almost 80% in the highly metastatic, enriched cell populations. A linear correlation (r = 0.92; P less than 0.001) was found between the percentage of cells in the population which bound high levels of SBA and the frequency of lymphatic metastasis in a series of enriched cell lines. Clones which bound high levels of SBA metastasized to lymph nodes at a high frequency, while clones which bound only low amounts of SBA exhibited a low frequency of lymphatic metastasis regardless of the metastatic potential of the cell line from which the clones were isolated. The binding of SBA to the cell was reduced by preincubation of the lectin with galactose, completely blocked by incubation with N-acetylgalactosamine, and unaffected by incubation with glucose or mannose, demonstrating that SBA was recognizing a N-acetylgalactosamine-containing component of the cell surface. Cells enriched for lymphatic metastasis were not similarly enriched for hematogenous metastasis. While cell lines enriched for lymphatic metastasis have been previously described, this is the first report of a specific cell surface property, SBA-binding, associated with lymphatic metastasis.

Stimulation of plasminogen activator production by dimethyl sulfoxide in Chinese hamster ovary cells.

The production of plasminogen activator activity in an auxotrophic mutant of the Chinese hamster ovary cell line was found be greatly stimulated by low concentrations of dimethyl sulfoxide. The production of both cell-associated and excreted plasminogen activator activities was stimulated maximally by dimethyl sulfoxide at a concentration of 2.5%. The stimulation of plasminogen activator activity production was found to be completely inhibited by actinomycin D and cycloheximide but not by mitomycin C, implying that new protein and RNA syntheses were required for this process. Using specific antibodies against plasminogen activator, the presence of a tissue-type plasminogen activator could only be detected in dimethyl sulfoxide treated cells. The dimethyl sulfoxide induced plasminogen activator production was observed only in a mutant auxotrophic for adenosine, glycine, and thymidine but not in wild-type cells. The ability of dimethyl sulfoxide to induce the synthesis of plasminogen activator was lost when the cells were hybridized with another complementary auxotrophic mutant. This implies that the ability of dimethyl sulfoxide to stimulate the production of plasminogen activator may be related to the auxotrophic mutation in this cell.

Involvement of plasminogen activator production with tumor metastasis in a rat model.

The process of metastasis involves numerous steps, many of which are thought to require the action of hydrolases, such as collagenase and other proteases. In this study, we investigate the role of the protease plasminogen activator in the metastasis of the rat mammary adenocarcinoma 13762. We observed that this tumor cell line is heterogeneous with respect to plasminogen activator (PAA) production. Clonal tumor cell populations were isolated which produced various levels of PAA. This phenotypic property of these clones remained stable for long periods of in vitro culture and did not affect their tumorigenicity. When the metastatic potential of these clones was determined using the lung colony assay, a strong correlation between PAA and metastatic potential was found. Furthermore, a threshold level of PAA production was observed, above which the lung colony-forming ability of the cells increased dramatically. These studies suggest that PAA production may play an important role in tumor metastasis.

A 6-thioguanine-resistant variant of the 13762 cell line which is no longer tumorigenic or metastatic.

A 6-thioguanine resistant (TGR) variant of the highly tumorigenic and metastatic mammary adenocarcinoma cell line 13762 was obtained. This variant was no longer tumorigenic or metastatic in normal syngeneic rats but did grow as a primary tumor in irradiated animals. Our results suggest that the TGR cell line was rejected by an irradiation-sensitive immunological mechanism. Although the TGR cells produced primary tumors in irradiated animals, there was no evidence of the extensive metastasis seen with the 13762 cells. This apparent inability to metastasize was confirmed by injecting the TGR cells intravenously. Whereas the 13762 cells produced large numbers of metastatic lung foci, there was no evidence of lung metastasis with the TGR cells, even in irradiated animals. Revertant cells for the 6-thioguanine-resistant phenotype were still non-tumorigenic and non-metastatic in normal rats, suggesting that 6-thioguanine resistance is not associated with the altered tumorigenic phenotype. From the TGR variant, cell lines were selected with an increased ability to produce tumors in normal rats. Although some of these revertants were capable of producing limited lung metastases in normal animals, extensive metastases were always seen when the cells were injected into irradiated animals. Differences between the 13762 and the TGR variants were also found in their ability to produce plasminogen activator. The TGR cells released far less plasminogen activator in culture than the 13762 cells. This could be a contributing factor in their different metastatic potentials.

Altered drug permeability in mammalian cell mutants.

The properties of colchicine uptake into Chinese hamster ovary cells were examined and found to be consistent with an unmediated diffusion mode. This uptake was stimulated several fold by metabolic inhibitors. The activation energy of colchicine uptake was found to be 19 kcal per mole; a similar value was obtained in cells stimulated by cyanide. Drug resistant mutants with greatly reduced colchicine permeability have been isolated. They displayed a pleiotropic phenotype, being cross-resistant to a variety of unrelated compounds. The basis of this pleiotropy was due also to reduced drug permeability. Examination of the lipids and fatty acids of parental and mutant cell membranes revealed no major differences. However, a 170,000 dalton surface glycoprotein was observed to be associated with colchicine resistance. This glycoprotein was postulated to be a modulator of drug permeability. All these data are consistent with the concept that mammalian cells are able to regulate the permeation of drugs entering by an unmediated diffusion process.

Modulation of membrane drug permeability in Chinese hamster ovary cells.

The kinetics of colchicine uptake into Chinese hamster ovary cells have been investigated and found to be consistent with an unmediated diffusion mode. A variety of compounds such as local anesthetics and non-ionic detergents as well as drugs such as vinblastine, vincristine, daunomycin and actinomycin D potentiate the rate of colchicine uptake into these cells and into colchicine resistant mutants. In all cases, higher concentrations of these compounds were required to stimulate colchicine uptake in the colchicine resistant mutants than in the cells of the parental line. This stimulation was observed also in the uptake of puromycin, a structurally and functionally different drug. These stimulatory agents did not, however, cause the cells to become nonspecifically leaky since the uptake of 2-deoxy-D-glucose was unaffected. In addition, the activation energy of colchicine uptake was unaltered in the presence of stimulating agents, implying that they were not causing colchicine to enter the cells via a different mechanism. The results are compatible with the view that these compounds are membrane-active, and are able to stimulate an increased rate of unmediated diffusion of colchicine into the cells. It appears that a mechanism for the regulation of passive permeability is modified in the resistant mutants.

Colchicine permeation is required for inhibition of concanavalin A capping in Chinese hamster ovary cells.

Several antimitotic, tubulin-binding agents, such as colchicine, colcemid, and podophyllotoxin, inhibit the capping of fluorescent-labeled concanavalin A in Chinese hamster ovary cells. By comparing the effects of these agents on parental cell lines on several independently selected colchicine-resistant mutants with decreased drug permeability, we have demonstrated that permeation of these drugs in required for inhibition of capping. These data support the hypothesis that these antimitotic agents interact with an intracellular component, probably microtubules, to prevent the directional movement of concanavalin A receptors on the surface membranes of Chinese hamster ovary cells.

An improved method for the isolation of surface membranes from cells in monolayer culture.

A modification of the procedure reported by Scher and Barland (Biochim. Biophys. Acta 255 (1972) 580-588) for the isolation of the surface membranes of cells grown in monolayer culture is described. The modification involves addition of a purification step utilizing the two-phase polymer system described by Brunette and Till (J. Membr. Biol. 5 (1971) 215-224). The modified method permits the isolation of surface membranes from cells not yet at confluence, when they may easily become detached from the substratum.