In vivo requirement for asialo GM1 and Thy1 positive leukocytes for antitumor effect of rIL-2 +/- rIFN-gamma.

Localized treatment with recombinant human interleukin-2 (rIL-2) +/- recombinant murine interferon-gamma (rIFN-gamma) results in regression of early B16 melanomas in normal C57BL/6 (B6) mice, but not in syngeneic beige mice, which have defective natural killer (NK) cells. Injection of antibodies to asialo GM1 (a-AGM1) or Thy1 abolishes the tymoricidal effects of rIL-1 +/- rIFN-gamma. Thus, cells activated by these cytokines must be either NK-like cells that are AGM1+ Thy1+, or NK-like cells (AGM1+) cooperating with T lymphocytes (Thy1+), since either antibody (a-AGM1 or a-Thy1) independently abrogates the in vivo antitumor effect.

Eradication of mouse melanoma by combined treatment with recombinant human interleukin 2 and recombinant murine interferon-gamma.

Successful immunotherapy of early s.c. or i.p. (B16) melanoma in syngeneic C57BL/6 (B6) mice was achieved with s.c. peri-lesional injections (for s.c. tumors) or i.p. injections (for i.p. tumors) of recombinant human interleukin 2 (rIL-2) and recombinant murine interferon-gamma (rIFN-gamma). Over a 28-day period, rIL-2 and rIFN-gamma were injected 14 times. Results with this combination were additive with s.c. tumors and synergistic with i.p. tumors. Treatment with 6,250 U-25,000 U of rIL-2 and 2 micrograms of rIFN-gamma began 1-3 days after s.c. inoculation of melanoma. On day 50, 87% (72/83) of mice thus treated were completely free of tumor. None of the 78 control mice (tumor + buffer) survived. Of mice receiving either rIL-2 or rIFN-gamma alone, 59% (47/79) and 53% (44/83), respectively, were tumor-free. I.p. tumors were also eradicated by i.p. injections of rIL-2 (50,000 U) with rIFN-gamma (5, 10, and 15 micrograms) as judged by absence of tumor in 81% (21/26) of mice autopsied between days 45 and 65. No control mice survived, and only 17% (2/12) and 20% (6/30) given either rIL-2 or rIFN-gamma separately (i.p.) were tumor-free. Doses of rIFN-gamma from 1-4 micrograms were more beneficial in eliminating 1-day s.c. melanomas than were higher doses, and local s.c. treatment was far superior to distant or systemic treatment. Non-adherent peritoneal or splenic cells from mice bearing 6-day-old i.p. melanomas and treated with one or both lymphokines on days 3 and 4 were used in cytotoxicity assays in vitro. Significant cytotoxicity against cultured melanoma cells was displayed by cells harvested from lymphokine-treated mice, but there was no evidence of the synergism observed with combination treatment of i.p. tumors in vivo. rIFN-gamma inhibited proliferation of melanoma cells in vitro, whereas rIL-2 stimulated proliferation at 1,000 U/ml. Plating efficiency was increased by at least 30% by culture with 100 U or 1,000 U of rIL-2/ml and both concentrations neutralized the inhibitory effect of 0.05 ng/ml of IFN-gamma, but not of 0.5 or 5.0 ng/ml.

Successful immunotherapy of mouse melanoma and sarcoma with recombinant interleukin-2 and cyclophosphamide.

Recombinant human interleukin-2 (rIL-2, courtesy of Cetus Corporation, Emeryville, CA, U.S.A.) is highly effective in eradicating syngeneic tumor when administered to C57BL/6 mice implanted with a nonimmunogenic, rapidly growing clone of B16 melanoma, or an immunogenic methylcholanthrene-induced sarcoma, 1-3 days, before beginning treatment at 5 X 10(4) U/injection daily for 5 days, and then on alternate days for 4 weeks. Low-dose cyclophosphamide (CY, 50 mg/kg), injected intraperitoneally (i.p.) 4 times at weekly intervals, greatly facilitated tumor eradication in rIL-2-treated mice. Most early subcutaneous (s.c.) tumors were cured by combining CY i.p. with repeated perilesional s.c. injections of rIL-2, i.e, 100% cure for 1-day and 87-91% for 3-day melanomas. When rIL-2 was administered without CY, the cure rate was 64% for 1-day and 67% for 3-day melanomas. This cure rate indicated a synergism between rIL-2 and CY, since CY alone did not affect tumor incidence. CY was therefore administered in all subsequent experiments. Treatment with rIL-2, localized to the site of the melanoma or sarcoma, was most effective, although systemic (i.p.) administration achieved results regardless of tumor site. When either tumor was implanted s.c. and rIL-2 treatment was also given s.c. locally, beginning 1-3 days later, 87-100% of the mice were cured, compared with 35-50% cured when rIL-2 was administered i.p., and 0% cured in excipient buffer-injected controls. Conversely, with i.p. treatment of i.p. tumors, 60-83% of the mice were tumor-free on day 50, as compared with only 17% tumor-free if treatment was s.c. These in vivo model systems should prove useful in helping establish protocols for human therapy.

Ecotropic murine leukemia virus-induced fusion of murine cells.

Extensive fusion occurs upon cocultivation of murine fibroblasts producing ecotropic murine leukemia viruses (MuLVs) with a large variety of murine cell lines in the presence of the polyene antibiotic amphotericin B, the active component of the antifungal agent Fungizone. The resulting polykaryocytes contain nuclei from both infected and uninfected cells, as evidenced by autoradiographic labeling experiments in which one or the other parent cell type was separately labeled with [3H]thymidine and fused with an unlabeled parent. This cell fusion specifically requires the presence of an ecotropic MuLV-producing parent and is not observed for cells producing xenotropic, amphotropic, or dualtropic viruses. Mouse cells infected with nonecotropic viruses retain their sensitivity toward fusion, whereas infection with ecotropic viruses abrogates the fusion of these cells upon cocultivation with other ecotropic MuLV-producing cells. Nonmurine cells lacking the ecotropic gp70 receptor are not fused under similar conditions. Fusion is effectively inhibited by monospecific antisera to gp70, but not by antisera to p15(E), and studies with monoclonal antibodies identify distinct amino- and carboxy-terminal gp70 regions which play a role in the fusion reaction. The enhanced fusion which occurs in the presence of amphotericin B provides a rapid and sensitive assay for the expression of ecotropic MuLVs and should facilitate further mechanistic studies of MuLV-induced fusion of murine cells.

DNA-mediated transfer of human melanoma cell surface glycoprotein gp130: identification of transfectants by erythrocyte rosetting.

DNA sequences encoding a human melanoma membrane-bound sialoglycoprotein of 130,000 molecular weight (gp130) were introduced into a clonal derivative of mouse B-16 melanoma cells with the selectable neomycin resistance gene (aminoglycoside phosphotransferase). Mouse transfectants were identified by a rapid and precise screening method with mouse monoclonal antibodies and erythrocyte rosetting. The frequency of gp130 transfectants was approximately 1 in 2,000 to 5,000 colonies with neo+ cells. Analysis of secondary mouse transfectants has revealed that the transfected gp130 has a molecular weight, isoelectric point, intracellular processing, peptide map, and spatial orientation of surface-exposed epitopes indistinguishable from those seen with gp130 from human melanoma cells. In contrast to primary transfectants, secondary transfectants expressing gp130 lack demonstrable human repetitive sequences.

Efficient DNA-mediated transfer of selectable genes and unselected sequences into differentiated and undifferentiated mouse melanoma clones.

We have found that three phenotypically dissimilar mouse B16 melanoma subclones are competent recipients for DNA-mediated gene transfer. Two of these approach and a third, amelanotic clone B78H1, surpasses mouse LTK cells in frequencies of transferent colony formation after treatment with either of two codominantly selectable plasmid vectors, pSV2gpt or pGCcos3neo. Melanoma transferents incorporate both selectable plasmid-homologous sequences and substantial amounts of unselected donor DNA into their cellular DNAs. In addition they retain the distinctive states of differentiation characteristic of the untreated clones. Frequencies of pGCcos3neo-mediated transfer of neo gene-encoded antibiotic resistance into B78H1 can reach 10(-2) in response to treatment with as little as 15 ng plasmid/ml coprecipitate/dish. B78H1 cells readily give rise to "secondary" transferents for the neo gene after treatment with DNA from a "primary" B78H1 neo transferent. This gene transfer system has potential applications for study of regulation of melanoma and neural crest differentiation and malignancy.

Leukocyte subpopulations elicited by a nontumorigenic variant of B16 melanoma: their role in direct rejection of the melanoma and in prevention of tumorigenesis in Winn assays.

The mechanisms by which various leukocyte subpopulations elicited by an immunogenic, nontumorigenic subclone (C3471) of B16 melanoma caused rejection of the tumorigenic parental melanoma (B559), were investigated. Leukocytes from C3471-immune mice were co-injected with B559 tumor cells in Winn assays into normal syngeneic recipients. Tumor formation by B559 cells was prevented when C3471-immune (a) unfractionated peritoneal leukocytes, or (b) glass-adherent peritoneal cells (90% macrophages), or (c) nylon wool purified nonadherent cells (95% Thy-1.2+) were used in the Winn assays. If the C3471-immunized mice were treated with antithymocyte serum before harvest of their peritoneal cells, none of these leukocyte populations were effective in the Winn assay. However, macrophages from these immunologically compromised donors regained their tumoricidal activity after incubation in vitro with T lymphocytes from untreated C3471-immune donors; similarly, C3471-immune lymphocytes rendered normal resident peritoneal macrophages tumoricidal in Winn assays. When C3471-immunized mice were irradiated or treated with antithymocyte serum before direct challenge with B559 cells, melanomas developed, thus providing additional evidence for the need for intact T cell function to establish immunity against the melanoma. Furthermore, when Winn assay recipients were treated with antithymocyte serum, neither C3471-immune macrophages nor T cells were able to prevent tumor formation. These findings indicate that antithymocyte serum-sensitive (Thy-1.2+) lymphocytes are necessary both for the generation of tumoricidal leukocytes in C3471-immunized mice, and for the rejection of B559 melanoma by demonstrably tumoricidal macrophages in Winn assay recipients. In addition, long-lasting immunity developed in 50% of the normal mice that had received both C3471-immune peritoneal cells and B559 tumor cells, as manifested by their capacity to reject a second challenge with B559 cells 40-60 d later.

Relationship between rejection of several syngeneic tumors and retrovirus production by 5-bromodeoxyuridine-grown melanoma cells: lack of protection in natural killer-deficient beige mice.

Bromodeoxyuridine-grown B16 melanoma cells (C3471) immunize mice against not only the parent melanoma but also two other C57BL/6 tumors: a mammary adenocarcinoma and a methylcholanthrene-induced sarcoma. We have shown that the endogenous retrovirus induced in C3471 cells by bromodeoxyuridine can persistently infect feral mouse (SC1) cells and that they then become as efficient as C3471 cells in preventing tumors. Uninfected SC1 cells cannot protect. Neither C3471 nor virus-infected SC1 can prevent B6MS5 or B6MS7, two other non-virus-producing sarcomas, from forming tumors. Meth 4, mammary adenocarcinoma, and L-cells produce retrovirus and prevent melanoma formation in half the mice challenged. Significantly, C57BL/6 mice homozygous for the beige mutation are unable to reject melanoma challenge after C3471 immunization, although their normal littermates do so efficiently. We conclude that production of retrovirus is in some way responsible for the cross-reactive immunizing capacity of C3471 cells and that cells in which the beige mouse is deficient play a role in the rejection process. Beige mice have been shown to be deficient in natural killer cells and abnormal in macrophage kinetics. In addition, C3471-induced protection against B559 melanoma appears to involve host cells sensitive to anti-thymocyte and anti-lymphocyte sera. We hypothesize that the retrovirus-producing cells may cause induction of interferon, which augments the cytocidal activity of natural killer cells and macrophages, killing sensitive tumor cells.

Macrophages elicited with heat-killed bacillus Calomette-Guérin protect C57BL/6J mice against a syngeneic melanoma.

We have demonstrated that a murine cytotoxic peritoneal cell can be elicited by intraperitoneal immunization with heat-killed Mycobacterium bovis, strain Bacillus Calmette-Guérin (BCG). When these cells are injected together with cells of clone B(5)59 of B16 melanoma in a Winn-type transfer assay into syngeneic C57BL/6J mice, the tumorigenic potential of the melanoma is completely abrogated. Similarly, mice immunized intraperitoneally with dead BCG are protected against intraperitoneal challenge with a number of B16 melanoma cells sufficient to cause tumors in 100% of control mice. However, mice immunized intraperitoneally with dead BCG are not protected against tumor formation when B16 melanoma cells are injected subcutaneously. Co-injection of BCG-elicited peritoneal cells with B16 melanoma cells into nude or sublethally irradiated (650 rad) mice inhibits tumor formation in > 85% of the mice, indicating that additional participation of host bone marrow- or thymus-derived leukocytes is not required to eradicate the tumor implant. The effector cell in the BCG-induced peritoneal exudate is adherent and phagocytic and is a mononuclear phagocyte. Nonadherent lymphoid cells from the same BCG-induced peritoneal exudate and from thioglycollate-broth-elicited granylocytes and macrophages neither prevent nor delay B16 tumor formation.

Recovery of three distinct biologically active type C viruses from cloned C57Bl/6 melanoma cells.

The genome of a single cell derived from the B16 melanoma contains information for the expression of three distinct biologically active viruses: the N- and B-tropic ecotropic viruses and the xenotropic virus. Their release results in reduction of melanin production. A possible relationship of virus replication to differentiation may be involved.

Co-cultivation of tumorigenic mouse melanoma cells with cells of a non-tumorigenic subclone inhibits plasminogen activator expression by the melanoma cells.

Clone B559 mouse melanoma cells are highly tumorigenic and produce plasminogen activator. Cells of clone C3471, a line obtained by continued growth of B559 cells in medium containing 5-bromodeoxyuridine (1 microgram/ml), have no plasminogen activator and are non-tumorigenic. When B559 cells are co-cultivated with C3471 cells, the ability of B559 cells to activate plasminogen is suppressed. Under these conditions cell fusion occurs. Lack of expression of plasminogen activators is not a consequence of cell fusion, inhibition of cell division or release of soluble inhibitors of either plasminogen activators or plasmin. No inhibitors of plasminogen activators could be demonstrated in association with sub cellular fractions of C3471 cells or with the C-type viral particles released from C3471 cells. Close contact between cells of the two lines is shown to be essential for suppression of plasminogen activation.

Malignant mouse melanoma cells do not form tumors when mixed with cells of a non-malignant subclone: relationships between plasminogen activator expression by the tumor cells and the host's immune response.

Mouse melanoma clones B559 and B78 are highly tumorigenic when injected into C57BL/6J mice. Tmor formation by these cells is suppressed when they are mixed with nonmalignant bromodeoxyuridine-grown clone C3471 before injection. C3471 cells suppress tumor formation only in immunocompetent hosts; mixtures of B559 and C3471 cells or C3471 cells alone form tumors in antithymocyte serum (ATS)-treated mice. Explants of C3471 tumors grown in ATS-treated mice form tumors in immunocompetent mice, most of which regress. Inability of C3471 or mixtures of C3471 with malignant cells to grow in normal mice, as contrasted with ability to grow in immunosuppressed mice, indicates that host response is involved. Both tumorigenic clones have high plasminogen activator activity, whereas nontumorigenic clone C3471 has none. Mixture of either tumorigenic clone with C3471 cells decreases plasminogen activator in vitro. C3471 tumor explants from ATS-treated mice initially express plasminogen activator, but lose the capacity to express this activity upon prolonged cultivation in vitro. Explants from B559 tumors retain plasminogen activator in long term culture. Close physical contact between C3471 and B559 cells appears essential both for inhibiton of plasminogen activator expression by B559 cells in vitro, and for tumor suppression in vivo. These findings suggest that production of plasminogen activators by tumor cells may play an important role in suppressing the host's immune response locally to an inoculum of syngeneic tumor cells.

Reversible suppression of malignancy and differentiation of melanoma cells.

Tumorigenicity is reversibly suppressed in mouse melanoma cells grown with 5-bromodeoxyuridine (BrdU). The nontumorigenic cells are immunogenic, and preinjection of these cells can protect mice against tumors inevitably formed when the parental, untreated melanoma cells are inoculated into inbred strain C57BL/6. A mixture of highly immunogenic clone, C(3)471, with malignant cells is also nontumorigenic. These effects are related to the host immune response since they occur only in immunocompetent mice. BrdU also reversibly suppresses functions related to pigment formation and plasminogen activation. These effects require incorporation of BrdU into DNA, emphasizing the value of the thymidine analog, BrdU as a tool to relate normal regulation of gene activity to perturbations of this regulation which produce malignant cells. This research can facilitate basic understanding of the malignant state and its relationship to host response as well as a method for immunizing melanoma patients after surgery to prevent tumor recurrence.

Suppression of melanoma cell tyrosinase activity and tumorigenicity after incorporation of bromouracil for one or two cell divisions.

We have studied the kinetics of suppression of tyrosinase activity and tumorigenicity in unsynchronized B16 mouse melanoma cells (clone B559) exposed to 5-bromodeoxyuridine (BrdU, 3 mug/ml) for one or two cell divisions, then cultured in BrdU-free medium (RM) for five or six days. Bromouracil replaced about 23% of thymine residues after 24 hours (1 cell division) and almost 40% after 48 hours (2 cell divisions) in the presence of BrdU. Upon subsequent growth in RM the extent of replacement declined in a manner consistent with dilution by new DNA synthesis, reaching 5-10% substitution by day 7 of these experiments. Tyrosinase activity was significantly reduced after treatment with BrdU for 24 or 48 hours but continued to decline after the cultures were changed to RM, approaching undetectable levels on day 7. The time course of reduction was similar to that previously determined in cells grown continuously for seven days in the presence of BrdU. Therefore, suppression of tyrosinase activity can result from incorporation of BrdU during a single cell cycle, but requires about seven days for full manifestation of the effect. Tumorigenicity decreased to 55% after 24 hours and to 15% after 48 hours with BrdU but rapidly reversed to approach that of untreated melanoma cells when subsequently grown in RM for 5-6 days. The effects of BrdU on total RNA or protein synthesis, or on plating efficiency appeared insufficient to account for the degree of suppression observed. Our results indicate that substitution by bromouracil into either strand of DNA loci controlling tyrosinase activity or tumorigenic potential may be sufficient for suppression. In addition, they demonstrate that such brief treatment with BrdU may be used to probe the regulation of differentiated function and tumorigenicity in these melanoma cells.

Correlated suppression by 5-bromodeoxyuridine of tumorigenicity and plasminogen activator in mouse melanoma cells.

The decrease of tumorigenicity by mouse melanoma clone B559 after growth in the presence of 5-bromodeoxyuridine (BrdU) has been correlated with a decrease in detectable cellular plasminogen activator. Reduction of both activities occurs after one to two cell divisions in the presence of this thymidine analog and is virtually complete within three to four cell cycles. These changes are fully reversible; four to five cell divisions in the absence of BrdU are sufficient to allow both tumorigenicity and plasminogen activator levels to return to normal. These results support the hypotheses that (a) the expression of a cellular plasminogen activator is closely associated with the transformation of normal to malignant cells and that (b) the suppression of tumorigenicity by BrdU reflects the capacity of this base analog to inhibit the expression of specialized functions which accompany the malignant state.