Heterogeneity of tumorigenicity phenotype in murine tumors. I. Characterization of regressor and progressor clones isolated from a nonmutagenized ultraviolet regressor tumor

Mouse Hepa-1 tumor is rejected by H-2Db-restricted CTL despite decreased MHC class I antigen expression

It has recently been hypothesized that tumor cells with reduced levels of MHC class I antigens are more susceptible to NK-mediated lysis and are rejected by NK cells, whereas tumor cells with normal levels of class I are rejected by tumor-specific CTL. We have tested this hypothesis using a mouse hepatoma system. The Hepa-1 tumor is a spontaneous H-2Kb loss variant that arose from the BW7756 tumor, when BW7756 was adapted to growth in culture. Our studies have shown that despite the loss of H-2Kb antigen, Hepa-1 is not more susceptible to NK lysis than its H-2Kb-transfected variants. These studies also suggested that NK cells were not responsible for rejection of the Hepa-1 tumor. The Hepa-1 tumor, therefore, appears to contradict the hypothesized linkage of MHC levels and NK susceptibility. Because NK cells are not involved in immunity to this tumor, we have sought to identify the effector cell responsible for Hepa-1 rejection. Cytotoxic T lymphocyte assays demonstrate that in vitro, Hepa-1 cells are lysed by Hepa-1-specific H-2Db-restricted CD4-CD8+ T lymphocytes. Footpad assays demonstrate that in vivo, Hepa-1 rejection requires CD4+CD8- and CD4-CD8+ Hepa-1-primed splenocytes. These results indicate that immunity to Hepa-1 is T cell mediated. Hepa-1 is therefore an example of an unusual tumor in that down-regulation of MHC class I antigen expression is associated with increased CTL susceptibility.

Interaction between two cellular subpopulations of a rat colonic carcinoma when inoculated to the syngeneic host

From a single, chemically-induced rat colonic carcinoma, two subpopulations of tumor cells have been isolated. When injected into syngeneic rats, TR cells give rise to progressive tumors in most of the animals, whereas TS cells give rise to no tumors or to tumors which regress in less than 30 days. TS cells inhibit the growth of TR tumors, whether they are injected before TR cells or simultaneously, at a different site or mixed with the TR cells. Lymph nodes and spleen lymphocytes from animals having rejected TS tumors also inhibit TR-cell growth. On the other hand, TS cells are able to give rise to progressive tumors when they are injected into rats bearing established TR tumors. Lymph nodes and spleen cells of TR tumor-bearing rats are able to enhance TS cell growth. These results suggest that subpopulations of cancer cells contained in the same tumor interact with each other through their effect on the host immune system. The growth of a whole tumor probably depends not only on the growth potential of each constituent subpopulation, but also on the interaction between the subpopulations themselves.

Different cellular requirements for inducing contact sensitivity and non-specific unresponsiveness with hapten-conjugated lymphoid cells

Primary contact sensitivity (CS) to Ox and non-specific unresponsiveness to other unrelated delayed-type hypersensitivity (DTH) containing CS were induced simultaneously by i.v. administration of oxazolone(Ox)-conjugated spleen cells. The subpopulation of spleen cells which serves as carriers for the induction of primary CS consisted of plastic-adherent, Thy 1-negative macrophage-like cells, and the carriers for the induction of non-specific unresponsiveness were non-adherent, Thy 1-positive cells. Further, allogeneic spleen cells were not capable of acting as carriers for the induction of primary CS, though H-2 restriction was not observed in the induction of non-specific unresponsiveness by Ox-conjugated cells. Thus, these two phenomena were shown to be induced by distinct pathways or distinct mechanisms. Adult thymectomy or treatment with cyclophosphamide (CY) prevents the induction of non-specific unresponsiveness. Moreover, an interval of 7 days was required for almost complete suppression of CS between treatment with Ox-conjugated spleen cells and sensitization with picryl chloride (PCl). These results suggest that relatively short-lived, CY-sensitive T cells participate in the induction of the non-specific unresponsiveness induced with Ox-conjugated cells.

Enhanced tumorigenicity of cloned UV-regressor tumor lines following selected in vivo and in vitro manipulations

Passage of cloned ultraviolet (UV) radiation-induced fibrosarcomas with regressor phenotype through 500-rad-irradiated syngeneic mice resulted in their conversion to transplantable progressor tumors. A similar conversion in tumorigenic phenotype (regressor leads to progressor) was found to be inducible in vitro by coculturing a cloned regressor tumor with normal splenocytes, but not with splenocytes from tumor-immune or UV-irradiated animals. Recloning of regressor and converted progressor tumor lines yielded regressor and progressor phenotype subclones, respectively, suggesting a degree of stability in their growth phenotype. Although all of the cloned progressor tumors tested were found to be cross-reactive with related regressor tumor lines, suggesting that related clones share a similar tumor-specific transplantation antigen, the progressor clones appeared to be less immunogenic than the regressor clones. Potential mechanisms that influence this conversion in tumorigenic phenotype are discussed.

Cellular heterogeneity in tumours

Malignant tumours contain normal cells, descendants of transformed cells, and conceivably also initiated cells which have taken some but not all of the steps toward malignancy, and hybrid cells. Tumours are propagated by multiplication of clonogenic cells, which are a subclass of the descendants of transformed cells. The clonogenic cells of a tumour may differ in respect of morphology, karyotype, metastatic capacity, sensitivity to cytotoxic drugs, expression of cell surface antigens and hormone receptors, immunogenicity, sensitivity to the immune reaction of the host, and other properties. Evidence (disputed by some) suggests that selection of particular subpopulations plays a role in tumour metastasis and recurrence. Heterogeneity may be due to pleoclonal origin, generation of phenotypic diversity within a clone, or spontaneous hybridization and chromosome loss. The possibility of interaction between different subpopulations must be taken into account in discussing tumour cell population kinetics. Heterogeneity also has important therapeutic implications and may help to explain the failure of some therapeutic regimes and the success of others.

The selective nature of metastasis

The issue of whether metastases result from the random survival of cells released from a primary tumor or from the selective growth of specialized tumor subpopulations endowed with metastatic properties is important to our understanding of the metastatic process and to the development of therapeutic modalities against metastatic disease. We have found that the tumor cells populating spontaneous metastases are more metastatic than the cells populating the parent neoplasm, clearly indicating that metastasis is selective and not random. The selective nature of metastasis is a consistent observation, however, only when tumor cells are obtained from spontaneous metastases from mice bearing heterogenous, poorly metastatic tumors. Tumor cells from spontaneous metastases from mice bearing tumors that have been selected for metastatic potential or that are homogeneous (cloned) do not differ significantly in metastatic potential from tumor cells populating the parent tumor. Thus, under some conditions the process of metastasis can appear random. Although tumor cells from different individual metastases may be homogeneous with regard to a metastatic phenotype, they may be heterogeneous with regard to their sensitivity to chemotherapeutic agents. Thus, although metastasis selects for metastatic variants, resulting in the population of metastatic foci with tumor cells endowed with metastatic properties, it does not appear to select for phenotypes irrelevant to the process of metastasis such as sensitivity to therapeutic agents.

Immunoselection of tumor cell variants by mice suppressed with ultraviolet radiation

It has previously been shown that mice exposed to ultraviolet radiation (UV) fail to reject highly immunogenic UV-induced tumors, which are regularly rejected by normal mice. The present study shows, however, that this immunosuppresion is incomplete, as UV-treated mice can still mount certain tumor-specific immune responses and reject smaller inocula of tumor cells that regularly grow progressively in athymic nude mice. Furthermore, all tumor cell lines that were reisolated from the tumor mass resulting from one tumor passage through UV-treated recipients heritably lost a tumor-specific determinant present on the parental tumor cells used for transplantation, and a large percentage of these reisolated variant tumors had changed to progressively growing tumors, in that they were no longer rejected by normal mice. In contrast, none of the tumors reisolated from passage through athymic nude mice or anti-idiotypically suppressed mice showed this change in antigenicity and progressive growth behavior. Thus, it appears that the phenotypic change in tumors reisolated from UV-treated mice was caused by immunoselection, and that the tumor-specific immunity in these mice apparently restrained the outgrowth of the parental tumor cells despite the partial immunosuppression. Because of the regularity at which tumor variants arose in the UV-treated mice after tumor transplantation, it appears that the partial immunosuppression caused by UV-treatment may have favored the outgrowth of antigenic variants from the parental tumor cell population, possibly by allowing more time for the generation of tumor variants. A similar immunoselection process might be part of tumor progression during tumor development and preferentially occur in cancer-bearing individuals showing concomitant tumor immunity.

Frequency of UV-induced neoplastic transformation of diploid human fibroblasts is higher in xeroderma pigmentosum cells than in normal cells

If neoplastic transformation of diploid human cells results from carcinogen-induced mutations, cells deficient in excision repair of UV-induced DNA damage should be significantly more sensitive to transformation by UV light than normal cells. We tested this hypothesis by irradiating fibroblasts from a xeroderma pigmentosum patient (XP7BE, complementation group D) with low doses of Uv light (254 nm) and cells from a normal person with much higher doses and comparing the frequency of transformation to anchorage independence. Both sets of cells exhibited a dose-dependent increase in transformation which corresponded to a dose-dependent decrease in survival. At doses that caused equal cell killing, the frequency of anchorage-independent cells was approximately equal. Colonies of XP7BE and normal cells isolated from agar, propagated, and injected into X-irradiated athymic mice produced fibrosarcomas in 100% of the animals. Normal cells irradiated shortly before the onset of DNA synthesis exhibited a high frequency of anchorage-independent cells; cells irradiated in early G1 showed no increase over background. These results agree with those we observed for UV induction of 6-thioguanine-resistant mutants in these cells and support the hypothesis that anchorage independence results from mutations induced by DNA replication on a damaged template.

Mechanisms of syngeneic tumor rejection. Susceptibility of host-selected progressor variants to various immunological effector cells

The ultraviolet radiation-induced fibrosarcoma 1591 generally is rejected by normal syngeneic mice and only rarely exhibits progressive growth. We isolated five of these rare progressor tumors from normal animals to determine the selective pressures that had been exerted upon the parental tumor by normal immunocompetent hosts. We found that the variant tumor cell lines could neither induce nor be killed by tumor-specific lymphocytes, suggesting that selection had been exerted against tumor cells expressing the tumor-specific antigen. In contrast, no selection against natural killer cell activity or against nonspecific T cell-mediated immunity seems to have occurred because progressor tumor cells were highly sensitive to these types of effector cells and in fact induced these effector cells more effectively than did the parental tumor. Nude mice were found to be as capable as normal mice in generating natural killer activity in response to a challenge with progressor tumor cells, but they were unable to mount a nonspecific T lymphocyte response. This may account for the fact that the progressor tumors grew at a significantly faster rate in nude animals than in normal mice. Thus, our study shows that in this tumor system nonspecific T cell-mediated immunity may play a role in retarding tumor growth, but the absolute resistance of normal animals to progressive tumor growth critically depends upon the presence of T cell-mediated tumor-specific immunity. Furthermore, neither NK cells nor nonspecific cytotoxic T lymphocytes appear to play a role in immunoselection against this tumor in normal immunocompetent hosts.