Tabulation of results on the heterogeneity of cellular characteristics among cells from B16 mouse melanoma cell lines with different colonization potentials. A summary of sixty reports.

The analysis of the complex, multi-step process of metastasis has been facilitated by the use of variant cell lines with differing metastatic or colonization potentials. In order to assess the full extent of the heterogeneity that can be expressed within a single cell line, the results of experiments describing individual ultrastructural, biochemical and behavioral aspects of the metastatic phenotype of B16 mouse melanoma cells have been collected. Of the 84 observations from 60 reports presented, there are 53 positive and 16 negative correlations with colonization after intravenous injection and/or metastatic potential, and 15 cases of no correlation. The tabulated results indicate that colony formation and metastasis can be associated with one or more of many different cellular traits so that genes responsible for each of these many traits could contribute to the metastatic phenotype.

Relationship of in vitro cell motility and colonization potential in a mouse colon adenocarcinoma (MCA-38) cell line.

Cell motility is an important factor in the metastatic process that can be affected by environmental conditions. A quantitative study was made of the relationship between cell motility and the colonization potential of a mouse colon adenocarcinoma cell line (MCA-38). MCA-38 cells grown in culture did not produce hepatic or pulmonary colonies following ileocolic or tail vein injection, respectively. In contrast, MCA-38 cells adapted to grow in the mouse produced colonies in both organs. The motility of the MCA-38 cells that did not produce colonies, as determined by the depth of penetration into cellulose nitrate filters (8 micron pore size), was significantly less than that of MCA-38 cells with colony-forming potential. Return to in vitro growth resulted in both a loss of colonization potential and a reduction in motility. In this system, secondary organ colonization and in vitro cell motility are positively correlated, suggesting an association between cell motility and metastatic potential.

Collagenase effects on cancer cell invasiveness and motility.

The invasion by malignant cells through extracellular matrix is an important part of the metastatic process, providing access to points of dissemination. Cell migration in tissues, however, depends not only on the destruction of extracellular matrices, but also on the locomotory behavior of the cells themselves. A quantitative study of aspects of cell behavior related to invasiveness was made using cellulose nitrate filters, both unimpregnated and filled with collagen, as models for some aspects of basement membrane. The relative penetration of mouse malignant cells into filters correlated with their spontaneous metastatic potential. Penetration of collagen-impregnated filters was greater than in unfilled filters. Pretreatment with collagenase enhanced the penetration of some cells into both collagen-impregnated and unfilled filters, and enhanced their motility on a plastic substrate; other cells showed enhanced penetration when incubated on collagenase-pretreated filters and no change in motility on the plastic substrate when incubated in collagenase-containing medium. These results emphasize the variability in response of different malignant cell types to factors present in the tumor environment and suggest that the effect of collagenase during invasion may be to enhance cell motility as well as to degrade the extracellular matrix.

Some effects of the tumor promoter 12-0-tetradecanoylphorbol 13-acetate on cell interactions in vitro.

Studies were made of the effects of the potent tumor promoter 12-O-tetradecanoylphorbol 13-acetate (TPA), 4-O-methyl TPA (4-O-MeTPA), and 4 alpha phorbol 12,13-didecanoate (4 alpha PDD) on the aggregation of embryonic chick cells in gyratory shaker culture, a model system useful for the study of cell adhesion and cell interactions. TPA and, to a lesser extent, 4-O-MeTPA significantly reduced the neural retina aggregate size at concentrations as low as 10(-9) M and 10(-8) M, respectively. An inactive isomer, 4 alpha PDD, had no effect up to 10(-6) M. The reduction in aggregate size appeared related to promoter activity since dexamethasone, a steroid that inhibits tumor promotion by TPA, significantly reversed the inhibitory effect of TPA. None of the agents tested affected the sorting pattern in mixed neural retina and heart cultures. The results indicate that intercellular adhesion, as determined by extent of aggregation, is reduced in the presence of TPA. This inhibition is considered to be related to the tumor-promoting activity of TPA.

General inhibition of embryonic cell aggregation by neoplastic cells.

The presence of small numbers of cancer cells cause an inhibition in the aggregation of embryonic chick neural retina cells. Experiments were made to investigate the inhibitory action of malignant mouse and virus-transformed chick neural retina cells on the aggregation and adhesion of several embryonic cell types. Malignant mouse cells, and their conditioned medium, inhibited the aggregation and adhesion of embryonic skeletal muscle heart and liver cells. The transformed retinal cells inhibited the aggregation of neural retina cells. These results suggest that the effect of cancer cells on the behavior of embryonic cells is a general one not related to the origin of the cancer cells.

The effect of neuraminidase- and ribonuclease-susceptible surface anionic groups on the aggregation of embryonic chick neural retina cells.

The role of cell surface charge in cellular interactions has been the subject of conflicting reports. The major contribution to the net cell surface negativity of all mammalian cells studied is made by the sialic acid moieties of the surface glycoproteins, while ribonuclease-susceptible sites have been shown to contribute to the lesser extent on some cell types. Experiments were done to determine whether these anionic groups at the cell periphery affect the aggregation and sorting behaviour of embryonic chick neural retina cells when cultured alone or in combination with embryonic heart cells. The net negative surface charge density, as determined by cell electrophoretic mobility, of neuraminidase- or ribonuclease-treated cells was significantly decreased immediately after incubation with the enzymes, and the treatment with neuraminidase resulted in a reduction in the binding of colloidal iron hydroxide particles at the cell surface. Both enzymes caused reduced aggregate size in gyratory shaker cultures of neural retina and mixed cell suspensions, and fewer neural retina cells adherent to microtest plate surfaces, but no differences were seen in their histological appearance or sorting pattern in mixed shaker culture. The results indicate that the neuraminidase- and ribonuclease-susceptible groups at the periphery of embryonic neural retina cells play a role in some aspects of cell contact behaviour in ways other than reduction in net negative surface charge.

Cells from normal and malignant human colon mucosa differentially inhibit embryonic cell aggregation.

A model system for studying some aspects of the interaction of cancer cells in tumors and their surrounding nonmalignant tissue is the co-culture of cancer cells and embryonic chick neural retinal cells in gyratory shakers. Neural retina cell aggregation, under these conditions, has been shown to be differentially inhibited by small numbers of cultured mouse and human cancer cells. We extend here these observations to co-cultures of retinal cells with small numbers (60:1 ratio) of human cells isolated from normal colon mucosa or colonic adenocarcinoma tissue. The cells from the malignant tissue samples inhibited aggregation to a significantly greater extent than the cells from normal mucosa, even when both were from the same individual. Cells derived from nonmalignant tumors were more inhibitory than those from normal individuals, which is consistent with described differences in this 'transitional' region. Thus, the aggregation inhibition assay appears applicable to freshly isolated human tissues.

Differential inhibition of in vitro malignant cell assays by B16 mouse melanoma variants.

Neural retina aggregation and adhesion assays, shown to be sensitive to the presence of small number of malignant cells, were differentially inhibited by cells or medium from B16 mouse melanoma sublines with higher metastatic characterizations, which suggests a correlation between inhibition of adhesion and aggregation and metastatic potential.

Contribution of macrophages to interactions between tumor cells and other tissues assessed in an in vitro system.

The modification of the tissues surrounding solid tumors is usually attributed exclusively to interactions between the normal tissues and the cancer cells in the tumor, in spite of the fact that tumors contain many different types of non-cancer cells including macrophages. As an experimental model for some facets of the cellular interactions between tumor and normal tissues, we have assessed the individual contributions of macrophages and cancer cells to the differential inhibition of neural retina (NR) cell aggregation by co-culturing NR cells with small numbers of macrophages or cancer cells alone, as well as with both natural (ascites tumors) and artificial mixtures. Cells from human colonic adenocarcinoma inhibited NR aggregation to a greater extent than normal colon mucosa. Aggregation of NR cells was inhibited by macrophages from mice and rats, and to a greater extent by cancer cell suspensions of mouse Ehrlich and rat Walker 256 lines from spinner culture or in the ascites form. Combinations of macrophages and cancer cells indicated that their inhibitory effects were neither additive nor synergistic. Cell-free media from macrophages and cancer cell cultures were equally effective inhibitors of aggregation. The results suggest that the interaction between a malignant tumor and the surrounding normal tissue can be modified by cancer cells, tumor macrophages and their products.

The effects of tumour cells on embryonic cell aggregation, adhesion and detachment.

The presence of small numbers of tumour cells inhibits the aggregation of embryonic chicken neural retina cells grown in gyratory shaker culture. The aggregation of neural retina cells was also inhibited by ascites cell medium. We investigated whether the inhibitory effect of the tumour cells on aggregate size is effected by inhibition of the initial adhesion or by enhancement of their separation. The number of neural retina cells adherent to microtest plate surfaces was significantly reduced after incubation with either Ehrlich ascites cells or cell-free, conditioned medium, while the percentage of cells removed from glass by shearing was unchanged under those conditions. These results suggest that the reduction in neural retina cell aggregate size produced by Ehrlich ascites cells and their products is due to partial inhibition of neural retina cell adhesion processes, as distinct from enhancement of separation.

Differential inhibition of embryonic cell aggregation by cultured human cells with "malignant" of "normal" characteristics.

The effect that cultured human cells have on chick embryonic neural retina cell aggregation was examined. Different types of human cultured cells inhibited aggregation of chick neural retinal cells to differetn degrees when mixed at a human cell:retina cell ration of 1:60. It appeared from the eleven cell lines studied that cells with "malignant" characteristics inhibited retinal cell aggregation to a greater extent than those with more "normal" characteristics. The assay could be used as a further test for abnormality of cell types and also as a method for studying the interactions of malignant cells with cultured cells.

Some effects of positively charged surface groups on cell aggregation.

A study was made of the effects of 2,3-dimethylmaleic anhydride (DMA), a reagent removing positive charges, on the aggregation and surface charge of embryonic chick neural retina cells. Neural retina cells, recovered from the dissociation procedure, were cultured on a gyratory shaker and the aggregate dimaeters formed in the presence of DMA or DMA-serum dialysate, following DMA-pretreatment, or in appropriate control cultures measured. The electrophoretic mobilities of similarly treated cells were also determined. In addition, cellulose acetate electrophoresis was carried out on samples of serum containing DMA, and the incorporation of 14C-amino acids into DMA-treated cells studied. Aggregates formed in the presence of DMA, or following DMA-pretreatment, were significantly smaller than aggregates from control cultures. The electrophoretic mobility of DMA-treated cells was significantly increased in serum-containing medium, but not serum-free Hanks' solution. At 24 h after removal of DMA-containing medium, the mobilities of pretreated cells were similar to those of controls. The electrophoretic pattern of DMA-treated serum was changed only with concentrations of DMA many times that affecting cell aggregation or mobility. DMA-serum dialysate did not significanlty reduce aggregate size. The incorporation of 14C-amino acids in DMA-treated cells and the structure of aggregates were unchanged from controls. It is concluded that positively charged consituents of the cell periphery play a demonstrable, but not limiting, role in cell aggregation, while a minor role for positive charges on serum protein cannot be totally excluded.

Inhibition of embryonic cell aggregation by neoplastic cells.

The effects of normal and malignant cells on the aggregation of embryonic cells in gyratory shaker cultures were compared. The addition of 1 times 10-5 simian virus 40 (SV40)-transformed BALB/3T3 (SV40-3T3) cells to 6 times 10-6 embryonic neural retina cells caused a highly significant greater reduct on (22.7 percent) in aggregate diameter than the addition of untransformed BALB/3T3 (3T3) cells. The ratio of the number of single cells to the number of aggregates was significantly higher for cultures containing SV40-3T3 cells than for the cultures containing 3T3 cells. This effect was concentration dependent in the presence of cultured Ehrlich-Lettre hyperdiploid (ELD) ascites cells; however, media from ELD cell cultures or ELD cell sonicates resulted in aggregates of greater diameter and lower ratios of single cells to aggregates. This approach may provide a sensitive assay system for the interactions of tumor and other cells in vitro.

Cytochalasin B prevents specific sorting of reaggregating embryonic cells.

The effect of cytochalasin B on specific sorting during reaggregation of embryonic chick heart and neural retina cells was studied. At a dose that did not measurably affect uptake of precursors of protein and RNA synthesis, ratios of potassium to sodium ions, and nonspecific aggregation, cytochalasin B disrupted the formation of the characteristic pattern of islands of heart cells within a retinal continuum.