Characteristics of the chemotactic factor-mediated cell swelling response of tumor cells.

The interaction between Walker carcinosarcoma cells, maintained in ascites form in noninbred Sprague-Dawley rats and in cell culture, and appropriate factors induced directed migration (chemotaxis) of these cells as measured in the Boyden chamber assay. The response of the tumor cells to these factors was very similar to that of leukocytes to their chemotactic factors. In addition to inducing chemotactic responses, the interaction between Walker carcinosarcoma cells and appropriate chemotactic factors also led to a swelling of the cells, which could be measured with the use of Coulter Channelyzer C-1000. Associated with the cell-swelling response was a temporary drop in the cell count of a suspension of cells as measured with the use of a Coulter counter, model ZBI. These responses were also similar to what occurs when leukocytes interact with their chemotactic factors. When the suspension of tumor cells was pretreated with 2-deoxyglucose, the responses measured with the Coulter counters were almost completely abrogated. In contrast to this, treatment of the cells with various chemical agents, including cytochalasin B, colchicine, cycloheximide, chlorpromazine, and the Ca2+ ionophore A23187 failed to significantly reduce the cell-swelling response. Cytochalasin B slightly potentiated the response. These similarities to the responses of leukocytes suggested a common underlying mechanism.

Induction of K562 lysis in autologous LGL and dense T cell cocultures.

Human large granular lymphocytes (LGL) were cultured with autologous dense T lymphocytes for 0-8 days. The mixed LGL dense T cell cultures had enhanced lytic activity against K562 on day 4, as compared to LGL cultured alone. Dense T cells cultured alone had no K562 lytic activity. The mixed LGL dense T cell cultures were reseparated on day 4 into low dense SRBC- and SRBC+ and into high dense subsets. Both low dense subsets did lyse K562 whereas the high dense subset did not lyse K562. Since the LGL population was the low dense SRBC- cells, the results suggested that the day 4 low dense SRBC+ cytotoxic population in the mixed LGL dense T cell culture originated from noncytotoxic dense T cells. Dense T cells were also separated into OKT8- and OKT8+ subsets and cultured together with LGL for 0-8 days. The LGL dense OKT8- T cell coculture had enhanced K562 lytic activity while the LGL dense OKT8+ T cell coculture had no enhanced lytic activity, as compared to LGL cultured alone. These results indicated that LGL interacted with the dense OKT8- T cells to produce the enhanced K562 lysis.

NK-patterned binding expressed by non-NK mouse leukocytes.

Previous studies demonstrated that NK-cell-enriched spleen cells bind to sensitive but not to resistant mouse lymphoma targets, measured by a single-cell target binding assay. In the present report we further analyzed this "NK-patterned" binding using effector cells with high, low or no NK activity. In agreement with previous results, nylon-wool-passed spleen cells as well as a cloned cytotoxic cell line with NK activity bound tumor targets largely according to their NK sensitivity; NK-sensitive tumors had a higher frequency of binders than the more resistant ones. The correlation coefficient relating the percentage target binding cells (TBC) for each tumor with the ability of the same tumor to inhibit lysis in cold target competition assays was consistently higher than that relating percentage TBC with direct NK lysis. A non-lytic variant of the cloned cytotoxic cell line had a binding pattern identical to that of the lytic line, demonstrating that "NK-patterned" binding occurred in the absence of lysis. Thymocytes, which are totally devoid of NK activity, also bound preferentially NK-sensitive targets, and their binding was decreased after trypsin treatment or in the presence of EDTA. Peritoneal macrophages also showed "NK-patterned" binding, thus demonstrating that this was not restricted to cells in the T-cell lineage. Variants from a mouse lymphoma selected for decreased NK sensitivity bound a lower frequency of NK-active (nylon-wool-passed spleen cells) and inactive cells (thymocytes, peritoneal macrophages). This study therefore showed that the membrane property conferring NK-binding selectivity exists on several types of non-NK leukocytes.

Responses of normal and malignant cells to collagen, collagen-derived peptides and the C5-related tumor cell chemotactic peptide.

Two variant subpopulations of murine fibrosarcoma cells that differ significantly in their malignant potential and normal mouse fibroblasts were compared with regard to ability to respond chemotactically to collagen, collagen-derived peptides and the C5-derived tumor cell chemotactic peptide. Two distinct patterns of responsiveness were observed. The normal fibroblasts and non-metastasizing fibrosarcoma cells responded to the collagen products but not the C5 peptide. The metastasizing fibrosarcoma cells responded to the C5 peptide but not to the collagen products. These findings emphasize the similarities between the normal fibroblasts and the non-metastasizing fibrosarcoma cells.

YAC-1 MHC class I variants reveal an association between decreased NK sensitivity and increased H-2 expression after interferon treatment or in vivo passage.

Two H-2 negative variants of the YAC-1 lymphoma were selected by mutagenization and sequential in vitro selections and compared with wild-type cells for changes in NK sensitivity and H-2 expression after interferon treatment or in vivo passage. The H-2 negative variants and the low H-2 expressor YAC-1 wild-type cells had similar NK sensitivity. However, IFN-beta or recombinant IFN-gamma pretreatments increased the H-2 expression of YAC-1 and protected them from NK lysis, whereas the H-2 variants, which remained H-2 negative, were not protected and often more sensitive to NK lysis. The H-2 variants were similarly susceptible as wild-type cells to three other cellular effects of interferon: protection from virus infection, modulation of Con A capping, and inhibition of cell proliferation. Thus, the only interferon-mediated effect that distinguished the H-2 negative variants from wild-type cells was the inability of the former to increase their H-2 expression and decrease their NK sensitivity. The wild-type YAC-1 line showed increased H-2 expression and decreased NK sensitivity after in vivo passage. In contrast, in vivo passaged H-2 variants showed no reexpression of H-2, and remained NK sensitive. The altered responses to interferon and in vivo passage were specific for loss or down-regulation of H-2, because Thy-1 loss (H-2 positive) YAC-1 variants behaved as the wild-type cells in all respects. This study supports the hypothesis that NK cells may function in vivo to eliminate host cells that fail to express H-2 after interferon stimulation during an immune response; such cells are a potential threat because they may escape recognition by T lymphocytes despite the expression of viral or tumor-associated antigens.