Protective specific immunity induced by doxorubicin plus TNF-alpha combination treatment of EL4 lymphoma-bearing C57BL/6 mice.

The therapeutic efficacy of a single (day 8), moderate dose (4 mg/kg, i.v.) of doxorubicin (DOX, Adriamycin) combined with recombinant human TNF-alpha (3 different doses and 5 different schedules, i.v.) was evaluated in C57BL/6 mice bearing an implant (s.c.) of the DOX-sensitive, TNF-alpha-resistant EL4 lymphoma. In parallel to monitoring survival, the levels of several host anti-tumor cytolytic effector functions of splenocytes and thymocytes were evaluated throughout the treatment period and in long-term survivors (LTS). DOX treatment alone resulted in a moderate (approx. 20%) increase in life span but no cures. TNF-alpha alone, at any tested dose or schedule, had little or no positive effect on survival. The combinations of DOX and TNF-alpha were only slightly better than DOX alone with respect to the time to death of mice that died (approx. 29% increase); however, each of the combinations involving 1,000 U TNF-alpha/injection produced a fraction (20% to 80%) of LTS. The host defense activities examined included those of splenic and thymic cytolytic T lymphocytes (CTL) and lymphokine-activated killer cells as well as splenic tumoricidal macrophages. Although most activities were modulated by tumor growth and/or treatment, only CTL responsiveness appeared to correlate with survival. CTL activity in the treated groups with LTS was significantly higher than in control groups late in the treatment period. Finally, ex vivo analyses of splenocytes and thymocytes together with the rejection of implanted tumor at 17 months established that LTS displayed specific long-term immune memory.

Thymic anti-tumor effectors in mice cured of lymphoma by cyclophosphamide plus TNF-alpha therapy: phenotypic and functional characterization up to 20 months after initial tumor inoculation.

As reported previously, cyclophosphamide plus tumor necrosis factor-alpha treatment of C57BL/6 mice bearing advanced EL4 lymphoma induced approx. 60% long-term (i.e., >60 days) survivors. These mice developed protective immunity, as evidenced by 1) rejection (100% survival) of EL4 tumor re-implanted on day 60 (day 0 = initial tumor implantation); and 2) development of significant levels of specific EL4 tumor cell killing activity by both splenocytes and thymocytes. Using this model, age-related changes in functionally and phenotypically definable thymocyte subsets were assessed. In thymocytes from 90 to 308 day survivors, specific immune memory was long term; both CD4+ and CD8+ cells were required for the ex vivo stimulation of lytic activity, but the specific anti-EL4 cytotoxic effector was CD4-CD8+. On day 520, the surviving mice were randomized into 2 groups. One group received a second re-challenge with EL4 tumor cells and all survived. The other group was sacrificed on day 520. Their thymocytes, exposed to X-irradiated EL4, developed anti-EL4 lytic activity and, in comparison with thymocytes of young and age-matched control mice, were markedly enriched in CD4-CD8+CD44+ cells. On day 625, thymocytes from the survivors of the day 520 re-challenge were evaluated and were found to have developed specific anti-EL4 lytic activity. Phenotypically, they had returned toward the pattern seen in age-matched control mice although CD4-CD8+CD44+ cells remained increased. These mice were > or = 2 years old, the median life span of C57BL/6 mice. Thus, mice cured of tumor by an immuno-modulating regimen rejected re-implanted primary tumor and maintained specific thymic anti-tumor immune memory for life.

Expression of B cell markers on SR-4987 cells derived from murine bone marrow stroma.

The murine cell line SR-4987 was originated in our laboratory from adherent cells of a long term bone marrow culture. SR-4987 cells do not express p21-ras and c-fms products on membrane whereas secrete M-CSF, evidence a fibroblast-like morphology and are vimentine positive. This line shows a very poor "in vitro" agar clonogenicity which is not modulated by the addition of different cytokines and growth factors (M-CSF, GM-CSF, G-CSF, IL-3, IL-7, alpha-TNF, PDGF, and EGF). On the contrary, a dramatic increase in clonogenicity is observed in the presence of bFGF. The RT-PCR investigation evidences the mRNA encoding for bFGF, IL-7, GM-CSF, and SCF (c-kit ligand). The analysis of CD antigen expression on SR-4987 cell membrane indicates a phenotype (CD5+, CD44+, 45R(B220)+, sIg+, 5'-nucleotidase+) that is consistent with a B cell feature. Our observations suggest that exogenous bFGF might represent an appropriate stimulus for inducing the SR-4987 cells proliferation also in the absence of cell-substrate anchorage. Further, they indicate that SR-4987 cells could represent a particular differentiation stage in which characters of "stromal cell" and "B cell" are coexpressed in agreement with the hypothesis of a common stromal-hematopoietic differentiation.

Specific anti-EL4-lymphoma immunity in mice cured 2 years earlier with doxorubicin and interleukin-2.

This laboratory has reported the conditions for an effective, non-toxic, chemoimmunotherapy utilizing doxorubicin in combination with prolonged administration of interleukin-2 and the identification of the critical role of activated CD8+ T cells in the therapeutic effect. Mice (C57BL/6) cured in those studies have been followed for the remainder of their life spans. These mice, approximately 2 months of age when initially inoculated with syngeneic EL4 lymphoma, survived for more than 2 years, the normal life span of C57BL/6 mice. Mice 4 months old reinoculated with the EL4 cells all survived. At about 1 year of age mice were sacrificed and the ability of their thymocytes and splenocytes to develop specific CD8+ anti-EL4 activity was as high as it had been at the time of tumor rejection. At about 2 years of age EL4 was reimplanted into mice; all of them survived. These surviving mice, at 2 years 2 months of age, as well as a group of 2-year-old mice not rechallenged, were killed and functional antitumor activity and phenotype characteristics of various lymphocyte populations were determined in comparison to those of young and age-matched control mice. The phenotyping of the lymphocytes from the cured mice indicated very notable differences in subset distribution and increased CD44 expression. Functionally they developed high levels of anti-EL4 activity, which was ablated by combined treatment with monoclonal antibodies against CD8 and CD44, indicating the role of memory cells. Consistent with cells from aged mice, these same cell populations had a very reduced allogeneic responsiveness. It appears that cured mice have developed an immune memory specific for EL4.

Cyclophosphamide plus tumor necrosis factor-alpha chemoimmunotherapy cured mice: life-long immunity and rejection of re-implanted primary lymphoma.

Changes in functionally and phenotypically definable splenocyte subsets in aging mice which had been rendered tumor-free in early life by immunochemotherapy (cyclophosphamide plus tumor necrosis factor-alpha) were studied in the syngeneic EL4 lymphoma-C57BL/6 murine model. Treatment-induced long-term survivors (LTS) surviving rechallenge are termed "immune-LTS". On day 120 (day 0, initial tumor inoculation), splenocytes from day 60 rechallenged immune-LTS developed significantly greater specific anti-EL4 cytolytic activity in an ex vitro assay than those from non-rechallenged LTS. Splenocytes from combination-treated groups developed significantly higher activity than those from cyclophosphamide-induced immune-LTS. The splenic effector precursor was a CD8+ T cell. The specific anti-EL4 effector cell from the cyclophosphamide-induced immune-LTS was CD4- CD8+; however, approximately 50% of those from combination-treated immune-LTS appeared to be CD4+CD8+. On day 520 immune-LTS were randomized into 2 groups. One group was re-implanted with EL4 tumor; all mice survived. The other group was killed and, even though their splenocytes developed considerable anti-EL4 activity, their allogeneic responsiveness was as reduced as that of age-matched controls. Phenotypic analysis, compared with splenocytes from young and age-matched controls, revealed changes in the makeup of each T-cell subset, except the CD4+CD8+, and all subsets, except the CD4-CD8-, had increases in CD44 positivity. On day 625, the age of these mice was equivalent to the median life-span of C57BL/6 mice; nevertheless, their splenocytes developed high anti-EL4 activity. Phenotypic analysis indicated that, compared to day 520, there was a major decrease in CD4-CD8+ splenocytes; we suggest that these cells had migrated to the site of tumor eradication.

Doxorubicin and cyclosporin A affect murine lymphoid cells expressing different antigenic determinants.

Cyclosporin A (CsA) enhances the antitumor activity of doxorubicin (Dox) as well as that of some other cytotoxic drugs against drug-resistant tumor variants. In some cases, however, such combination treatments result in severe unexplained toxicities. In this study the possibility was explored that the effects of CsA and Dox on lymphoid cells may be instrumental, at least in part, in determining their toxicological effects. Subsets of spleen and thymus lymphoid cells, from mice with or without Dox or CsA treatment, were identified by flow cytometry based upon their plasma membrane antigenic determinants. The results indicate that there is essentially no cross-sensitivity/resistance between the two agents. The most Dox-susceptible cells were immature (non-proliferating) CD4+CD8+ thymocytes and CD3-B220- as well as CD3-B220+ splenocytes. These populations were intact following CsA treatment, but the numbers of mature CD4+CD8- and CD4-CD8+ cells were substantially reduced. Similar "mirror image" differences were found for other populations examined. When considered together, these findings suggest that in combination Dox and CsA would affect nearly all subsets of lymphoid cells, providing one possible explanation of why increased leukopenia, toxicity and immunosuppression are found after their combined administration. Since leukemias, lymphomas and, to a more limited extent, certain solid tumors express these same phenotypic markers, similar analyses should be considered for monitoring and perhaps even predicting neoplastic cell sensitivity to treatment with such agents.

Lipopolysaccharide and splenic tumoricidal macrophage activation.

Splenic macrophage tumoricidal activity was examined and a splenic macrophage tumoricidal assay was established. Initially, mixtures of lipopolysaccharide (LPS) and spleen single cell suspensions (SSCS) were cultured for 1-4 days. Adherent macrophages, washed free of nonadherent cells and LPS, were then examined and were found to lack tumoricidal activity in a standard 18-h 51Cr release assay. However, tumoricidal activity was generated if LPS was added to the SSCS cultures at later time points during the 4-day incubation period; maximal activity was seen when LPS was added on day 3. In parallel, significant changes in macrophage autofluorescence and morphology, but not phenotype, were observed. Next, SSCS were cultured for 1-4 days without stimulating agents. Adherent macrophages were then washed free of nonadherent cells and LPS was added. Significant tumoricidal activity developed in time- and LPS concentration-dependent fashions. The presence of nonadherent spleen cells in physical contact with the macrophages during the SSCS culture was essential for the macrophages in the resultant monolayer to be responsive to LPS. Activated splenic macrophage-mediated lysis of tumor cells was shown to depend on the contact between the two cells.

Doxorubicin-induced DNA degradation in murine thymocytes.

Exposure of murine thymocytes to doxorubicin (Dox) (0.5-1.0 microM, 24 hr) triggered rapid DNA degradation, as indicated by the appearance of a major subdiploid population demonstrated by DNA flow cytometry. Electron microscopic comparison of samples with large subdiploid populations versus those with little or no such subset revealed significantly more cells with the characteristic features of apoptosis, the morphologically definable stage of programmed cell death. These features include unipolar condensed chromatin, zeiosis, and electron-dense cytoplasm. Dox-induced apoptosis occurred without prior S or G2/M phase arrest or cell size increase. The subset most susceptible to Dox-induced apoptosis in vitro and in vivo was CD3-CD4+CD8+. The same subset is affected by dexamethasone (Dex); as reported for Dex-induced apoptosis, actinomycin D and cycloheximide also blocked Dox-induced apoptosis. Thymocytes exposed to higher Dox concentrations (2-10 microM) did not have a subdiploid population. Although at 2-10 microM Dox significantly reduced cell numbers (probably as a result of necrosis), at least 5-10% of the population was viable at 24 hr. Thymocytes exposed to low concentrations of Dox (0.001-1.0 microM) plus Dex (0.1 microM) exhibited additive induction of apoptosis, whereas those exposed to high concentrations of Dox (2-10 microM) plus Dex were completely devoid of any evidence of apoptosis. These results indicate that the Dox-induced killing in thymocytes (mostly noncycling cells) occurs via different mechanisms depending upon the Dox concentration.

Effect of perioperative chemoimmunotherapy with cyclophosphamide and autologous tumor vaccine in murine MBT-2 bladder cancer.

The in vitro cytotoxic activity of splenocytes from C3H/He mice implanted subcutaneously with 10(6) syngeneic MBT-2 tumor cells on day 0 was significantly enhanced after cyclophosphamide (100 mg./kg., intraperitoneally) given 2 days before tumor resection on day 17, with or without active specific immunization with BCG plus autologous irradiated tumor cells (vaccine) 1 week after tumor resection. Furthermore, a significantly lower tumor incidence was seen in mice challenged with 10(5), but not 10(6), tumor cells per mouse 24 hours after tumor resection on day 17 and treated with cyclophosphamide on day 15 and postoperatively with vaccine than was found in nontreated tumor resected mice. Phenotypic analysis of cells from spleen showed that cyclophosphamide pretreatment and postoperative vaccine, either singly or in combination, induced a significant increase of both CD44+ memory T cells and CD11b+ myeloid/macrophage cells. Thus, in addition to a specific antitumor immune response, a nonspecific cytolytic mechanism may also play a role in the observed antitumor effect.

Intracellular doxorubicin kinetics in lymphoma cells and lymphocytes infiltrating the tumor area in vivo: a flow cytometric study.

Recently we have reported the development of a safe and effective chemoimmunotherapy protocol involving doxorubicin (Dox) in combination with interleukin 2 which, in C57BL/6 mice, boosts local T cell responses, and, in 50 to 80% of the cases, this resulted in the complete eradication of established syngeneic EL4 lymphoma or its Dox-resistant variant, EL4/A. Accumulation of host-derived leukocytes in the peritoneal cavity was increased up to 8-fold after tumor inoculation, but, in absolute numbers, did not increase further following Dox administration. The cellular pharmacokinetic studies undertaken to clarify the role of Dox following a single IV injection indicate that 4 h later, lymphocytes found in the peritoneal cavity have detectable levels of Dox; but the lymphoma cells (both EL4 and EL4/A) have, in proportion to their larger size, taken up more drug as judged by flow cytometry. The estimated drug "concentration" (i.e., intracellular amount divided by estimated cell size) at the 4-h time point, however, was found to be essentially equivalent in both the lymphoma cells and the lymphocytes. Thereafter, the drug content and intracellular "concentration" in the EL4/A cells rapidly declined while their numbers progressively increased. In contrast, the EL4 lymphoma cells and the lymphocytes found in the peritoneal cavity in the presence of either lymphoma consistently exhibited higher levels of drug 24-48 h than at 4 h. Splenic and tumor-infiltrating mature T (CD3+) cells were completely insensitive to Dox cytotoxicity and actually showed increased CTL activity when examined ex vivo. Although EL4 cells had identical Dox uptake patterns to those of CD3+ cells, they were sensitive to the drug and their numbers decreased, resulting in increased host/tumor cell ratios in these mice. The pharmacokinetic parameters of the drug and the insensitivity of the mature T cells to the drug determined in this study can explain, in part, the efficacy of a chemoimmunotherapy protocol boosting local T-cell responses.

Immunological responses critical to the therapeutic effects of adriamycin plus interleukin 2 in C57BL/6 mice bearing syngeneic EL4 lymphoma.

A safe and effective therapeutic combination of moderate doses of Adriamycin (doxorubicin, 4 mg/kg, IV, Days 1 and 8 or only Day 8) plus prolonged administration of moderate doses of interleukin 2 (2 micrograms, b.i.d., Days 9-40) was developed in the syngeneic EL4 (5 x 10(4) cells, IP, Day 0) lymphoma--C57B1/6 mouse model and has been reported in the companion paper. The studies described herein demonstrate that the effectiveness of this combination treatment against EL4 lymphoma growing intraperitoneally in C57B1/6 mice was dependent upon the presence of CD8+ cells. Thus, the induction of long-term survivors (60-80%) by Adriamycin plus interleukin 2 was completely ablated by pretreatment of mice with anti-CD8 monoclonal antibody (MAb), whereas pretreatment with anti-CD4 MAb only partially inhibited the therapeutic effects and anti-NK1.1 MAb had no effect. A close association between survival, an increase in phenotypically identified CD8+ cells, and an increase in specific anti-EL4 cytolytic activity was demonstrated with cells from the tumor site (peritoneum) but not consistently with cells from the spleen. No association was observed between survival and modulations in natural killer (NK), lymphokine-activated killer (LAK), or tumoricidal macrophage activity of spleen or peritoneal cells. Taken together the results indicate that, in this model, the most relevant correlate of a therapeutically effective host antitumor response is the level of specific EL4 tumor killing by cells present at the tumor site. Based on the findings reported herein, it can be predicted that weakly immunogenic tumors may be eradicated by immunologic mechanisms elicited in conjunction with properly designed therapeutic regimens.

Development of a safe and effective adriamycin plus interleukin 2 therapy against both adriamycin-sensitive and -resistant lymphomas.

This laboratory has extensively studied Adriamycin (doxorubicin)-induced immunomodulation. Despite demonstration of favorable effects, little therapeutic advantage was seen, and it was decided to test Adriamycin in combination with interleukin 2 (IL2). Considerable toxicity was seen with either high-dose IL2 or high-dose Adriamycin alone, using the syngeneic C57B1/6-EL4 T cell lymphoma model. When the doses of either agent were reduced to decrease toxicity, little therapeutic effect was seen. In contrast, an effective protocol without apparent toxicity was developed by combining a moderate dose of Adriamycin (4 mg/kg, IV, Days 1 and 8 or only Day 8) with prolonged administration of a moderate dose of IL2 (2 micrograms, b.i.d., i.p., Days 9 to 40). This protocol resulted in up to 80% long-term survivors among mice inoculated on Day 0 with EL4 lymphoma (5 x 10(4) cells). It should be noted that under these conditions, neither agent, when administered singly, induced long term survivors, and that following the inoculation of only 10-100 EL4 tumor cells all animals died in the absence of treatment. The survivors developed protective immunity as demonstrated by their ability to resist reimplantation with EL4 tumor. Furthermore, this resistance to tumor reimplantation could be transferred into naive hosts with spleen cells from tumor-bearing mice receiving the combination protocol; exposure of mice to sublethal whole body irradiation prior to tumor implantation completely abrogated the efficacy of this combination treatment. Finally, it was shown that this combination protocol was equally effective against an Adriamycin-resistant subline of EL4 that expresses the multidrug resistance phenotype.

Establishment and characterization of a new murine cell line (SR-4987) derived from marrow stromal cells.

A new murine cell line designated as SR-4987 was established by treating a long-term bone marrow culture with the supernatant from Y-1 cells which actively produce viral C-particles (MuLV). The line showed a fibroblast-like morphology and its mesodermal origin was confirmed by immunocytochemical staining. Flow cytometric analysis of DNA index evidenced a tetraploid number of chromosomes whereas cell cycle analysis showed 34.8% of cells in S phase and 60.7% in G1. In vitro growth studies demonstrated a population doubling time of 14.7 h, a good plating efficiency (52.3%) and a very poor agar clonogenic capacity (0.6%). SR-4987 was tumorigenic only in syngeneic mice in which sarcomas were induced. The line produced M-CSF in the culture supernatant whereas G-CSF, IL-3 and GM-CSF were not detected. Studies are in progress to assess the production of other cytokines and to verify if same autocrine growth factor is involved in the control of SR-4987 proliferation. Our line provides a further model of stromal cells for studying the interaction between hemopoietic progenitors and their microenvironment, as well as to study factors produced by stromal cells acting as modulators of proliferation and differentiation of related cell populations.

[Granulocyte functional characteristics of mice with transplanted tumors]. / Funktsional'nye osobennosti granulotsitov u myshei s perevivnymi opukholiami.

The Ehrlich ascitic carcinoma (EAC) growing subcutaneously in (CBA X C57B1)F1 mice was found to induce leukocytosis with the elevated number of mature granulocytes in the peripheral blood. At the same time depletion of the cell amount in the thymus and an increase of cells, especially granulocytic ones, in the spleen was observed. Granulocyte contamination of the ficoll-verografin gradient interphase was detected in mononuclear cells separated from tumour host blood. Simultaneously, expression of Fc G-receptors of granulocytes decreased in tumour-bearing mice. These data are discussed in terms of inhibitory effect of the EAC on granulocyte functions in the presence of paraneoplastic leukocytosis.