Search

Polyamine synthesis inhibitors, such as a-difluoromethylornithine (DFMO), inhibit tumor cell growth in vitro and in vivo. However, upon cessation of treatment, tumor growth resumes. We hypothesized that incorporation of radioactive polyamines might kill the growth-arrested cells. This hypothesis was previously tested in rat 9L brain tumor cells in which DFMO increased both the uptake and the retention of [3H] putrescine. In these rat cells, DFMO-induced retention of high-specific-activity [3H] putrescine for 20 days resulted in several logs killing. In the present studies all of the 5 different human glioma cell lines tested with DFMO treatment also showed enhanced uptake of exogenous [3H] putrescine, reduced cell counts and enhanced killing of colony forming cells (CSF). Extending the time of DFMO treatment of cells that had taken up high-specific-activity (80 Ci/mmol) [3H] putrescine further increased the killing. A 10-day extension resulted in a 10,000-fold reduction in cumulative cell growth. A 5-day extension resulted in a 2-3 log decrease in numbers of surviving CFC. These data further support the hypothesis and suggest that DFMO-induced cell cycle arrest enhances cellular retention of [3H] putrescine, increasing the effective internal radiation dose enough to cause proliferative death. In a clinical setting, the short (approximately 1 microm) path-length of the tritium beta particle should limit effects to the tumor cells and spare adjacent normal cells. These results support the concept that treatment with the combination of polyamine inhibitors and radioactive polyamines might be a useful adjunct to current therapies for glioblastoma multiforme.

Nonstimulated fetal liver (FL) from 14.5-day gestation mice had no natural killer (NK) cell activity and <3% expressed NK1.1. Even after short-term (3-4 day) culture of FL with the late-acting cytokines, interleukin (IL)-15 or IL-2, little or no NK activity was detected. However, longer-term (13 day) culture with IL-2 plus stroma derived from bone marrow (BM) of adult mice, resulted in extensive proliferation and differentiation to mature NK cells. Cell numbers began to increase after 4 days, and by day 13, they had increased 40-fold and 69% of the cells were NK1.1+ with high NK activity and 5%-10% were NK1.1- B220+. With stroma, but no IL-2, equivalent proliferation occurred, but differentiated cells were predominantly NK1.1- B220+, not NK cells. Culture for 13 days without stroma, but with either IL-2, IL-15, FLTK3-ligand (L) or stroma-conditioned medium, resulted in less than fivefold expansion, and minimal NK activity. Culture with combinations of FLTK3-L or ckit-L plus IL-15 or IL-2 increased both cell number and NK activity, but the increase in cell number was less than that seen with stroma plus IL-2. By limiting dilution assay on stroma plus IL-2, the precursor frequency was 1/(2660+/-292) whole FL cells and the absolute number, but not the frequency, increased during culture on stroma without IL-2. The NK cell progenitors were found in sorted NK1.1- and Sca-1+ c-kit+ lineage- subpopulations at a frequency of 1/(156+/-52.5). Together, these data suggest that the NK lineage cells in FL are primarily in early stages of development. They are highly proliferative, respond to early acting cytokines and express stem cell markers.

Bone marrow stem cells develop into hematopoietic and mesenchymal lineages but have not been known to participate in production of hepatocytes, biliary cells, or oval cells during liver regeneration. Cross-sex or cross-strain bone marrow and whole liver transplantation were used to trace the origin of the repopulating liver cells. Transplanted rats were treated with 2-acetylaminofluorene, to block hepatocyte proliferation, and then hepatic injury, to induce oval cell proliferation. Markers for Y chromosome, dipeptidyl peptidase IV enzyme, and L21-6 antigen were used to identify liver cells of bone marrow origin. From these cells, a proportion of the regenerated hepatic cells were shown to be donor-derived. Thus, a stem cell associated with the bone marrow has epithelial cell lineage capability.

We have been exploring the feasibility of delivering ionizing radiation to brain tumor cells by using tritium labeled polyamines. Polyamines are taken up preferentially by dividing cells and form noncovalent bonds with DNA. Their uptake can be enhanced by drugs which deplete endogenous polyamines. To test this in vivo, 9L cells were implanted in the striatal region of the brain in male Fisher 344 rats. Osmotic pumps containing trace amounts of [3H] spermidine or [3H] putrescine with either difluoromethylornithine or combinations of 3 inhibitors of enzymes of the polyamine biosynthetic pathway were implanted subcutaneously and were connected to intratumoral cannulas. After 14-16 days the brains were removed and sliced in the coronal plane. The diameters of the tumors were measured and tumor tissue was dissected from each slice, weighed and lysed for scintillation counting. It was found that difluoromethylornithine enhanced the uptake of [3H] putrescine while a combination of inhibitors of enzymes of the polyamine biosynthetic pathway enhanced the uptake of [3H] putrescine and [3H] spermidine producing a localized region of radioactivity in the 9L tumor. It is estimated that if the [3H] polyamines were at higher specific activity (commercially available), instead of the trace dose given here, the [3H] polyamine uptake would be sufficient to kill 9L tumor cells within a 2 to 3 week period.

Although 14.5-day murine fetal liver (FL) has few, if any, mature natural killer (NK) cells, culture of FL with recombinant human IL-2 (rhIL-2) and stroma from irradiated NK longterm bone marrow cultures (NK-LTBMC) allows proliferation and differentiation of NK cell progenitors. Using this system, NK cell progenitors were found in both CD34+ and CD34- sorted subpopulations of FL. The CD34 antigen was expressed by 14+/-1.3% of whole FL cells, while mature NK cells cultured from NK cell precursors in FL did not express the CD34 antigen. Anti-TER-119 mAb reacted with 84%+/-10.3% of the FL cells, and NK cell progenitors were enriched in the TER-119- subpopulation. After coculture with rhIL-2 and stroma, neither TER-119- nor TER-119+ cells expressed antigens associated with T cells (CD3, CD4, and CD8) or myeloid cells (Gr-1 and Mac-1). Only the TER-119 subpopulation generated NK1.1+ (77%) and B220+ (87%) cells. Within the TER-119 subpopulation, both CD34+ and CD34- cells generated cytolytic and NK1.1+ cells after culture. By a limiting dilution assay (LDA) of the Lin (i.e., negative for NK1.1, CD3, CD4, CD8, B220, Gr-1, and TER-119) CD34 positive or negative subpopulations, the calculated mean frequency of NK cell progenitors was about 1/100 for the CD34+Lin- subpopulation and about 1/(200-300) for the CD34-Lin- subpopulation. In kinetic studies, we found that NK1.1 antigen expression continued to increase with time in culture for both the CD34+Lin- and CD34-Lin- fractions. In contrast, the percentage of CD34+ cells decreased rapidly and produced CD34- cells, and the CD34- population remained CD34-. These data suggest that both CD34+ and CD34- subpopulations of FL can differentiate into NK cells when cocultured for 13 days with irradiated NK-LTBMC stroma and rhIL-2, and that CD34+ progenitors differentiate to CD34- precursors, which in turn differentiate to CD34- mature NK cells.

Although there is no formalized area of study called phylogeny or ontogeny of the hematopoietic microenvironment, new models and molecular tools are now available for such studies. The concept of a hematopoietic microenvironment has developed from the need to answer basic questions about migration, control of proliferation and differentiation of lymphohematopoietic cells; e.g. how are cells with the same genes induced to express different sets of these genes which lead to differentiation. These questions were first approached when cells could only be identified morphologically. The ontogeny of hematopoiesis was traced from the blood islands of the embryonic yolk sac, to the fetal liver, spleen, and bone marrow. Cells with reticular morphology were associated with areas of hematopoiesis and, in the embryo, they were thought to give rise to both hematopoietic and supportive cells. In the 1960's the classic work of McCulloch, Till and Siminovitch led the study of hematopoietic precursors which have no distinctive morphological identity and are too infrequent to study microscopically. These cells were identified by their functions; e.g. colony formation in culture in the presence of certain factors, production of spleen colonies or rescue of lethally irradiated mice. Cells with these functions were also found sequentially in the yolk sac blood islands, in the aorta/mesonephros, fetal liver, spleen, and bone marrow during development. The question remained, what regulates the proliferation and differentiation of these cells and why do they home to different sites in different stages of development? Among the laboratories studying spleen colonies, a controversy arose as to whether differentiation decisions were stochastic or induced by extra cellular factors. Dexter and Greenberger developed the long-term bone marrow culture system which has aided in studying the roles of factors such as cell-cell contact and extracellular matrix in hematopoietic differentiation. The molecular identification of ligand/receptor pairs such as ckit and KL as well as transactivating factors that control whole sets of lineage related genes such as the GATAs and Ikaros, may lead to clarification of the stochastic versus induced differentiation issue. Chimeric bird and frog embryos and analysis of mutations effecting hematopoiesis in frogs and zebrafish have helped to trace the earliest hematopoietic development in the embryo and to determine what influences it. The identification of genes that alter development of hematopoiesis opens the possibility of comparing microenvironmental control mechanisms in various present day organisms and relating these to evolutionary events. Many basic questions relevant to the interaction between hematopoietic cells and their microenvironment can be addressed by studying "simple" organisms in which the answers may be more easily determined than in mice or humans. Examples of possibly useful organisms, range from the teliosts such as zebrafish to algae such as Volvox, a two cell organism, to Dictyostelium which change from 1 to many cell types and in the process, migrate, adhere and differentiate.

Treatment of mouse bone marrow (BM) with rabbit anti-mouse brain serum (RAMBS) plus complement (C') depletes several cell types, including T cells and facilitating cells (FCs), that is, cells that facilitate engraftment of sorted allogeneic stem cells (SCs) in vivo. In the present study, treatment of BM with RAMBS+C' resulted in the depletion of approximately half of the late cobblestone area (CA)-forming stem cells as assayed on irradiated long-term bone marrow culture (LTBMC) stroma. In addition, LTBMC of RAMBS+C'-treated BM produced functionally impaired stroma with reduced ability to support CA formation by nontreated exogenous SCs. This stromal impairment was not due to depletion of TCRalphabeta T cells in the BM, because BM cultures from TCR alpha-chain knockout mice supported normal numbers of exogenous CAs. Because CD8+/TCR- cells are enriched for FCs, we tested the effect of adding these cells back to the treated BM prior to culture. The sorted FCs alone did not produce CAs, but did improve the ability of the impaired stroma to support late CA formation by sorted SCs. These studies provide a new model for dissecting the roles of different cellular components of BM in producing functional stroma that supports CA formation by SCs, and show that the number of CAs formed depends on the "quality" of the stroma as well as the number of SCs seeded. These findings further suggest that CD8+/TCR- BM cells may be important for the establishment of functional stroma.

The role of Ikaros in early stages of natural killer (NK) cell differentiation was investigated using an in vitro system that promotes proliferation and differentiation of NK cell precursors into mature NK1.1+ cells. Day 14.5 fetal liver cells from mice, either homozygous for Ikaros Null or dominant negative (DN) mutations, had severe 55- to 79-fold reductions in functional NK cell precursors. Although there was no statistically significant difference between values for +/+ and +/- Null mice, the mean precursor frequency for DN mutant (+/-) mice was significantly above that for DN -/- mice and below that for DN +/+ mice. The NK activity values for cells generated from the NK cell precursors followed the same respective relationships found for NK cell precursor frequencies. These data suggest that the deficiency of mature NK cells in Ikaros mutant mice is related to lack of functional precursors.

The goal of the present work was to study the regulatory role of T lymphocytes on natural killer (NK) cell generation in NK long-term bone marrow cultures (LTBMCs), an established mouse long-term bone marrow (BM) culture system used for the study of NK cell differentiation from precursors. Activation of the few T cells present in NK-LTBMCs by addition of anti-CD3 monoclonal antibody (mAb) together with interleukin (IL)-2 inhibited the generation of NK cells. Coculture with NK-LTBMCs of a pure population of preactivated BM T cells completely inhibited NK cell development even when the T cells were separated from the NK-LTBMCs by transwells. Depletion of IL-2 by activated T cells was not the mechanism of the negative regulation because anti-CD3 mAb added to the cultures inhibited the generation of NK cells even in the presence of 10-fold higher concentrations of exogenous IL-2 than that used in controls. Medium from cultures in which suppression had occurred was also suppressive, suggesting that one or more soluble factors released in the medium was responsible. That this effect was exerted on NK cell development from precursors was indicated by the finding that T cell-conditioned medium stimulated proliferation of mature NK cells. In our experimental conditions, monoclonal antibodies to IL-10, IL-13, transforming growth factor-beta, and tumor necrosis factor receptor failed to reverse the inhibitory effect.

Difluoromethylornithine (DFMO) depletes endogenous putrescine and enhances the uptake of and retention of [3H] putrescine in vitro. To determine if DFMO also enhances uptake of [3H] putrescine in vivo, DFMO and trace doses of [3H] putrescine, dissolved in artificial CSF, were infused into growing (6-9 day) 9L brain tumors by means of osmotic pumps. When 7-day osmotic pumps were loaded with 1 microCi [3H] putrescine, with or without 10 or 100 mM DFMO, pumped at 1 microl/h, the mean uptake after 3 days was 168 +/- 62 cpm/mg tumor (17 rats) without DFMO, 300 +/- 197 cpm/mg tumor (11 rats) with 10 mM DFMO and 1088 +/- 421 cpm/mg tumor (11 rats) with 100 mM DFMO (p < or = 0.05 vs. control). Significantly less radioactivity was detected in the contralateral brain and in nonbrain tissues (0.5 +/- 0.1 to 14 +/- 5 cpm/mg). To measure the extent of [3H] putrescine distribution in the tumor, the same dose of drugs was delivered for a longer period of time, using 14-day pumps to allow tumors to become large enough to be divided into 1.4 mm thick transections. The mean radioactivity in the sections from eight control rats receiving [3H] putrescine without DFMO were not significantly different between the sections (174 +/- 61 cpm/mg tumor for sections containing the cannulas, 273 +/- 61 and 259 +/- 91 cpm/mg for adjacent sections). In the six rats given 100 mM DFMO there was a significant increase in mean radioactivity in the cannula containing section (2251 +/- 919 cpm/mg tumor). Mean counts from adjacent sections in these rats were 97 +/- 44 and 33 +/- 13 cpm/mg. Values for contralateral corpus striatum and nonbrain tissues ranged from 0.7 +/- 0.3 to 4.3 +/- 1.5 cpm/mg tissue. When DFMO was delivered directly to the tumors while [3H] putrescine was infused intraperitoneally, the uptake in the tumor slices was low (5-10 cpm/mg in different slices). These results demonstrate that infusion of DFMO directly into growing 9L brain tumors can selectively enhance the uptake of exogenous [3H] putrescine by rapidly dividing cells which are within a 1.4 mm diameter area at the cannula tip. Although these studies used [3H] putrescine at trace doses, it is estimated that infusion of higher doses of [3H] putrescine plus DFMO will selectively kill tumor cells.

Several laboratories have introduced the lacZ gene into 9L cells to improve the usefulness of this already popular rat brain tumor model. However, these laboratories were not concerned about possible changes in the phenotypic characteristics of the 9L cell line which can be induced by the selection of lacZ-expressing clones. Here, we describe a method for introducing the lacZ gene into 9L cells without selective cloning. The 9L parent cells (passaged the same number of times) and 9L/lacZ cells were compared in a number of tests and found to have the same phenotype. Specifically, they had the same sensitivity to radiation from external gamma or internal beta radiation, the same growth rates with or without frequent media changes and the same patterns of growth in rat brain. We demonstrated that the 9L/lacZ cells could be sorted from dissociated tumors by flow cytometry and the percentage of nonmalignant versus malignant cells determined. These percentages were variable from rat to rat. The colony-forming efficiency could be determined on the basis of whole tumor or, by using the percent of lacZ-positive cells, on the basis of malignant cells in a tumor. These novel approaches should render the 9L tumor model even more useful.

The role of recombinant hematopoietic growth factors in radiation repair has become a subject of increasing interest in both clinical and basic radiobiology. Combinations of cytokines such as hepatocyte growth factor, interleukin (IL)-3, IL-11, kit ligand, GM-CSF and erythropoietin were used to study the in vitro radiation dose response of human cord blood CD34+ hematopoietic progenitor cells using clonogenic survival assays. CD34+ cells were isolated by immunomagnetic selection and irradiated at 8 cGy/min. Irradiated cells were plated in methylcellulose with or without added cytokines, and hematopoietic colonies including CFU-GM, BFU-E and CFU-GEMM were scored on day 14. The radiation response characteristics of BFU-E and CFU-GEMM were similar for all culture conditions tested. The D0 values for BFU-E ranged between 1.29 and 2.40 Gy and n between 1.0 and 1.4. The D0 values for CFU-GEMM ranged from 86 cGy to 2.02 Gy and n between 1.0 and 1.5. The D0 for CFU-GM grown without added factors was 1.03 Gy. With single cytokine stimulation (IL-3, IL-11 or varying concentrations of HGF), D0 values ranged from 1.11 to 1.44 Gy. With the combination of IL-3, GM-CSF, kit ligand and HGF, D0 values were not significantly altered and ranged between 1.61 and 2.60 Gy. In contrast, the combination of IL-11 and HGF produced an increase in the shoulder of the radiation survival curve (n = 3.35). No increase in the shoulder was detected for any of the other conditions tested (n = 1.0-1.7). Thus the combination of HGF and IL-11 increased the radiation survival of hematopoietic progenitor cells forming CFU-GM. Understanding the mechanism by which combinations of early-acting growth factors support postirradiation recovery of primitive clonogenic hematopoietic cells may be relevant to the design of clinical protocols for improving hematopoietic recovery after total-body irradiation.

For over 40 years, the association between hemopoietic chimerism and donor-specific tolerance for allografts has been recognized. However, toxicity associated with lethal conditioning has prevented the clinical application of bone marrow (BM) chimerism to induce tolerance. We previously demonstrated that engraftment could be achieved with less than total recipient myeloablation (700 cGy) and that the incidence of engraftment correlated with the dose of total body irradiation (TBI). Administration of cyclophosphamide (CyP) on Day +2 reduced the minimum TBI dose sufficient to permit engraftment to 500 cGy. In the current study, addition of antilymphocyte globulin (ALG) to the TBI/CyP-based conditioning approach reduced the radiation required for engraftment to < or = 300 cGy. B10 (H-2b) mice conditioned with ALG on day -3, 300 cGy of TBI with transplantation of B10.BR (H-2k) or BALB/c (H-2d) BM on day 0, and CyP on day +2 exhibited evidence of donor chimerism (49.6 +/- 3.7% and 38.2 +/- 2.4%, respectively) in 97% of recipients. ALG eliminated CD4+ and CD8+ cells and decreased NK1.1+ cells in the peripheral circulation at the time of transplantation. Moreover, T and NK cells in the host BM were significantly decreased compared with cells of recipients conditioned with TBI alone. CyP delayed repopulation of host thymocytes, providing time for the establishment of donor chimerism before production of mature T cells. Chimeric animals exhibited stable multilineage chimerism and donor-specific tolerance to skin grafts and in in vitro assays. This model may provide a clinically acceptable approach for the induction of donor-specific transplantation tolerance.

To better understand immune responses to brain tumors and to develop possible approaches for immunotherapy, we have investigated the leukocyte populations infiltrating the rat 9L gliosarcoma. By immunocytochem-ical analyses of the cells infiltrating the tumor, we observed a substantial number of cells expressing natural killer cell receptor protein 1 (NKR-P1), a marker expressed only on rat lymphocytes capable of non-MHC-restricted cytotoxicity. Previous investigations have determined the existence of three populations of NKR-P1+ lymphocytes in normal rats, including NKR-P1bright/T-cell receptor (TCR)-/CD3-/CD5- (approximately 5-15%), NKR-P1dim-/TCRalphabeta+/CD3+/CD5+ (approximately 1-5%), and NKR-P1dim/TCRgammadelta+/CD3+/CD5+ (approximately 0.5-2%). By one-parameter flow cytometry, it was determined that NKR-P1+ cells constituted 30-60% of the lymphocytes in 9L tumors. Among splenic lymphocytes or peripheral blood leukocytes, NKR-P1bright cells are 1.5-4.5 times more numerous than NKR-P1dim cells. In striking contrast, NKR-p1dim cells were 4-5 times more numerous than NKR-P1bright cells among lymphocytes isolated from 9L tumors. Using quantitative analyses of laser confocal microscopic scans, we determined that NKR-P1dim cells were approximately 4 times as numerous as NKR-P1bright cells in situ, confirming flow cytometric findings. By two-color now cytometric analyses, it was observed that approximately 5-10% of the cells were NKR-p1bright/CD5-/TCR-, a phenotype representative of NK cells. Also, approximately 11-25% of the cells were NKR-P1dim/CD5+/TCR+ cells, corresponding to the T-cell subset with non-MHC-restricted lytic function. In addition, we observed a cell population among 9L-derived lymphocytes with a NKR- p1dim/CD5-/TCR- phenotype (approximately 15-25%). Cells of this phenotype have not been reported previously, and most likely represent NK cells down-modulated for expression of NKR-P1. Alternatively, they might represent cells of unknown origin or cells down-modulated for expression of T-cell markers in the microenvironment of 9L tumors. We also compared the lytic capacity of NKR-P1+ populations derived from normal animals and from 9L gliosarcomas. In these experiments, it was determined that, although cells isolated from 9L tumors had some capacity to lyse tumor target cells, they were clearly less efficient than cells isolated from normal splenocytes. Cumulatively, these data suggest that there is selective localization of cells capable of mediating antitumor responses in 9L, but that tumor-associated factors may down-regulate their function and expression of NKR-P1.

Transduction of mouse hematopoietic stem cells and their progeny was studied using a recombinant retroviral vector (MFG-ASA) which incorporates the human arylsulfatase A gene (ASA; EC 3.1.6.8). Successful transduction was demonstrated in spleen colonies of mice that received bone marrow transplantation, cultured bone marrow-derived macrophages, visceral tissues and brain of long-term reconstituted mice, and also the spleen colonies of secondarily transplanted mice. The efficiency of transduction in primary spleen colonies was 90%. Expression of the ASA transgene exceeded endogenous levels in spleen colonies and in cultured macrophages by 50-100%. Enzyme activity in the visceral tissues of long-term reconstituted mice consistently showed elevated ASA activity, greater than three-fold in the spleen and lung of one animal. Increased activity of ASA also could be detected in secondary spleen colonies. These data demonstrate the usefulness of the MFG-ASA vector for efficient gene transfer and expression in mouse hematopoietic stem cells and their differentiated progeny. The presence of vector DNA in the brain 4 months after transplantation suggests a role for gene transfer and stem cell transplantation in the treatment strategies for metachromatic leukodystrophy.

Following 200 cGy total body irradiation, 20-25% of CBA/Ca mice and their CBA/B and CBA/H sublines develop myeloid leukemia. To determine whether hematologic changes in vitro were detectable, long-term marrow cultures (LTBMCs) were established from the right and left hind limbs of 11 individual control and 11 CBA/B mice 100-114 days after 200 cGy total body irradiation. Individual cultures were studied weekly for cumulative production of nonadherent cells and colony-forming, hematopoietic progenitor cells. Control cultures produced significantly more nonadherent cells over 25 weeks in long-term marrow culture compared to those from irradiated (treated) mice. Permanent stromal cell lines were established from control and irradiated CBA/B mouse LTBMCs and clonal sublines were established. The stromal cell lines from LTBMCs of in vivo irradiated CBA/B mice had uniformly lower plating efficiencies, and only one formed a permanent clonal subline at 100-fold lower frequency compared to stromal cell lines from control mouse LTBMCs. The irradiation sensitivity of both uncloned and clonal sublines was similar by single-hit, multi-hit or by linear quadratic formula. Cocultivation of an IL-3 dependent hematopoietic progenitor cell line established from a control CBA/B, LTBMC with control of irradiated stromal cell lines derived from either a control (CC3) or the one successfully cloned in vivo irradiated (CT4) LTBMC, produced few cobblestone islands in the presence of IL-3. In contrast, formation of cobblestone islands in the presence of L cell-condition medium as a source of M-CSF was significantly greater, and these persisted for 21 days on both CC3 and CT4 stromal lines. The data provide evidence for irradiation induced changes in the bone marrow stromal cell compartment of CBA/B mice which persist and are detectable in vitro 6 months after explant of the cells to culture. These marrow stromal cell lines may provide valuable resources for analyzing the molecular biologic changes in the hematopoietic microenvironment during irradiation leukemogenesis.

Evidence in experimental models suggests that many autoimmune diseases can be prevented by transplantation of bone marrow from disease-resistant donors. For potential clinical application, it would be important to avoid the morbidity and mortality associated with lethal conditioning and achieve mixed chimerism using less than complete recipient ablation. We report here for the first time that stable chimerism achieved in NOD mice using a sublethal radiation-based conditioning approach is sufficient to prevent beta-cell destruction and abrogate insulitis in prediabetic NOD mice. The percentage of NOD mouse recipients (8 wk of age) that engrafted with donor bone marrow correlated with the dose of irradiation and number of bone marrow cells transplanted. Engraftment of B10.BR bone marrow occurred in > or = 94% of animals receiving > or = 750 cGy of total body irradiation before bone marrow transplantation and > or = 30 x 10(6) bone marrow cells, while reproducible engraftment did not occur at radiation doses of less than 700 cGy and cellular doses of less than 30 x 10(6) bone marrow cells. All chimeric animals remained free of diabetes (n = 38) for 10 mo following bone marrow transplantation. Moreover, in all animals examined, no insulitis was present from 12 to 36 wk following reconstitution. In striking contrast, 61% (22 of 36) of NOD recipients that were conditioned but did not receive bone marrow developed acute diabetes by 12 mo. Insulitis was present in all remaining animals. These results suggest that allogeneic chimerism achieved using a sublethal conditioning approach can prevent the onset of diabetes and even reverse preexisting insulitis in NOD mice.

This study was designed to test the feasibility and safety of long-term expression of high levels of secreted human interleukin-1 receptor antagonist (hIL-1ra) protein in mice by retroviral transduction of hematopoietic stem cells. The retroviral vector, CRIP-MFG-hIL-1ra (MFG-IRAP), carrying the hIL-1ra gene was used to infect mouse bone marrow (BM) which was subsequently injected into lethally irradiated mice. All of the mice survived and greater than 98% of the white blood cells (WBC) of these mice were of donor type from 2-13 months after transplantation. All of the mice had hIL-1ra protein in their sera (40-1200 ng of hIL-1ra/ml) at all assay periods for at least 15 months after transplantation. Bone marrow from seven of seven primary recipients produced at least one secondary recipient with sustained, high serum levels of hIL-1ra, indicating that hematopoietic stem cells had been successfully transduced. Although the hIL-1ra was biologically active when assayed in vitro, the mice appeared to be well and their WBC counts and hematocrit (HCT) were not significantly different from those of lethally-irradiated mice given BM cells infected with the same vector carrying the lacZ gene. There was also no evidence of alterations of white cell subpopulations. These results demonstrate that systemic production of biologically active hIL-1ra can be obtained by retrovirus-mediated gene transfer to hematopoietic stem cells and that this level of expression and secretion into the serum is compatible with normal BM engraftment, hematopoietic recovery and survival of the lethally irradiated recipient mice. These hIL-1ra-expressing mice represent a model to examine the functions of IL-1 and hIL-1ra and to determine the ability of hIL-1ra to reduce susceptibility to chronic diseases such as rheumatoid arthritis as well as effects of aging such as bone degeneration. The data further suggest that transduction and transplantation of hematopoietic stem cells is a potential method for delivery of hIL-1ra and other secreted therapeutic gene products for systemic diseases.

Reconstitution of lethally irradiated mice with a mixture of syngeneic and allogeneic (A+B-->A) bone marrow results in multilineage mixed allogeneic chimerism, donor-specific transplantation tolerance, superior immunocompetence and resistance to graft-vs-host disease. However, the morbidity and mortality associated with lethal irradiation would be a major limitation to the clinical application of chimerism to induce tolerance for solid organ grafts or treat other nonmalignant hematologic diseases. We report here that durable multilineage mixed allogeneic chimerism and donor-specific transplantation tolerance for skin and primarily vascularized allografts can be achieved across multiple histocompatibility barriers using a nonmyeloablative radiation-based approach. The percentage of B10 mouse recipients that engrafted directly correlated with the degree of disparity between donor and recipient and the dose of total body irradiation administered. Although the occurrence of engraftment following conditioning with doses of total body irradiation of > or = 600 cGy was similar for animals receiving bone marrow disparate at MHC or MHC, minor and hematopoietic (Hh-1) loci (67% vs 78%), the level of donor chimerism was significantly less when multiple histocompatibility barriers were present (94.6 +/- 3.8% vs 37.5 +/- 12.5%). Treatment of the recipient with cyclophosphamide 2 days following allogeneic bone marrow transplantation reduced the dose of radiation sufficient for reliable engraftment to only 500 cGy of total body irradiation, regardless of MHC and Hh-1 disparity. Donor chimerism was stable and present in all lineages, with production of lymphoid (T and B cell), NK, and myeloid (erythrocyte, platelet, granulocyte, and macrophage) cells. Mixed chimeras exhibited donor-specific tolerance in vitro, as assessed by mixed lymphocyte culture (MLR) and cytotoxicity (CML) assays, and in vivo to skin and primarily vascularized cardiac allografts. The observation that engraftment and tolerance can be achieved across multiple histocompatibility barriers using nonmyeloablative recipient conditioning may allow allogeneic bone marrow transplantation to be applied to nonmalignant disease states in which lethal conditioning cannot be justified, including the induction of donor-specific tolerance for solid organ transplantation and the treatment of hemoglobinopathies and enzyme deficiency states.

Transplantation of untreated rat bone marrow into lethally irradiated (950 cGy) mouse recipients results in durable xenogeneic (rat-->mouse) chimerism and confers donor-specific transplantation tolerance for subsequent xenografts. The purpose of the present study was to characterize the minimal dose of total body irradiation (TBI) which would allow engraftment of rat bone marrow in mouse recipients. We report here that durable and stable lymphohaematopoietic cross-species chimerism can be achieved using a less than totally ablative radiation-based conditioning approach. The percentage of B10 mouse recipients which engrafted with rat bone marrow cells correlated with the dose of TBI. Engraftment of rat bone marrow stem cells occurred in all animals receiving 750 cGy prior to bone marrow transplantation, while no engraftment was detected at doses less than 650 cGy. Although most of the recipients were repopulated with mixed mouse and rat multilineage chimerism, some exhibited a predominance of rat cells. Although mixed xenogeneic rat/mouse chimeras prepared by lethal TBI produced only mouse derived RBC (red blood cells), chimeras prepared by sublethal conditioning produced both rat and mouse RBC. Only animals with detectable chimerism exhibited specific functional transplantation tolerance to donor xenoantigens, as assessed in vitro by mixed lymphocyte reaction assay. This model may offer an in vivo approach to study the role of species-specific growth factors in stem cell biology as well as the mechanisms for the induction of tolerance across species barriers.

Studies were undertaken to determine whether the doses of FK506 that are effective for acute GVHD prophylaxis following lethal irradiation and bone marrow transplantation (BMT) would also suppress myeloid cell reconstitution. FK506 (3 mg/kg/day) abrogated acute lethal graft versus host disease (GVHD) in lethally irradiated C57BL/10SnJ (H-2b) recipient mice given histoincompatible BM plus spleen cells from B10.BR (H-2k) donors and this dose was used in all of the studies. Endogenous and exogenous myeloid repopulation was studied in mice given daily injections of either FK506, an equivalent amount of carrier solvent or no treatment throughout the interval between total body irradiation (TBI) and the day of assay. Repopulation was studied after 400 or 500 cGy TBI (endogenous) and after 950 cGy TBI plus injection with syngeneic BM (exogenous). No consistent adverse effects of FK506 were seen during either exogenous or endogenous recovery. Parameters studied included hematocrit (Hct), WBC count, cells per humerus, spleen weight, splenic colony-forming units, % spleen or BM 59Fe uptake and colony forming cells per humerus. Similarly, when lethally irradiated secondary recipients were reconstituted with BM from FK506 treated primary recipients (lethal irradiation plus exogenous BM), no consistent effects were observed. These data suggest that FK506 given to prevent GVHD would not compromise the myeloid recovery that is critical for survival in the interval of time following shortly after BMT.

The role of the adhesion molecule CD44 in the development of NK cells was analyzed in a mouse long-term bone marrow culture system. After 4 wk of culture (day 0), recombinant human IL-2 was added and 13 days later the cells generated were shown to have substantial cytotoxic activity against YAC-1 and to be enriched for NK cells, as assessed for NK-1.1 phenotype by flow cytometric analysis. Physical separation between stroma and precursors partially inhibited proliferation and, consequently, a lower number of cytotoxic cells were produced. Similar results were obtained when an anti-CD44 mAb was added together with IL-2 at day 0. The disruption of hyaluronic acid (HA), one of the ligands of CD44, by hyaluronidase or the competition for the binding of CD44 by soluble HA added with IL-2 on day 0 inhibited both proliferation and development of cytotoxicity to a greater degree than did anti-CD44. These results indicate that interaction of CD44 with HA plays an important role in the development of pre-NK cells into cytotoxic effector cells.

Gaucher disease (GD), the most common human lysosomal storage disorder, results from a genetic deficiency of the enzyme glucocerebrosidase (GC). The cloning of human GC cDNA, the benefits of allogeneic bone marrow transplantation and the success of enzyme replacement therapy support the feasibility of gene therapy as an approach to a cure for GD. We report the transfer of the GC gene to mobilized human peripheral blood (PB) CD34+ cells obtained from patients primed with granulocyte colony-stimulating factor and/or chemotherapy. A tenfold enrichment of CD34+ cells was achieved using an avidin-biotin immunoadsorption technique. Prestimulation of the CD34+ cells with cytokines, followed by infection for 5 days with a supernatant containing the MFG-GC retroviral vector, resulted in enzyme activity up to 2.5-times greater than non-infected and lac-Z infected controls. Southern blot hybridization of DNA from these cells demonstrated a transduction efficiency of 10-30%. These studies show that the GC gene is transferred efficiently to mobilized PB CD34+ cells by the MFG-GC retroviral vector and results in expression of enzyme activity in the population of cells capable of bone marrow reconstitution. These results advance the development of gene therapy for GD.

One promising strategy for gene therapy of Gaucher disease involves ex vivo retroviral transduction of autologous hematopoietic stem cells. Studies in small animals have demonstrated that this approach provides a life-long supply of the glucocerebrosidase (GC) enzyme. Human application has developed to the stage of a clinical trial. In this study, we describe development of a high titer amphotropic producer line for the vector, MFG-GC, and explore transduction of CD34+ cells from various human sources. Higher than three times the normal levels of glucocerebrosidase activity in non-transduced cells were achieved following transduction of CD34+ cells obtained from bone marrow or cord blood from normal donors. The improvement in enzyme activity in Gaucher marrow was about 40-fold above deficient levels. We examined the timing and stepwise effect of multiple rounds of infection and evaluated post-infection expansion of cells in two different cytokine mixtures. Transduction efficiency was determined using immunocytochemistry and Southern blot hybridization.

Purpose: This study explored the possible use of D,L- alpha-difluoromethylornithine (DFMO) to enhance the uptake of [3H] putrescine in order to selectively kill brain tumor cells., Methods and Materials: Gliosarcoma cells (9L) were grown for 4 or 20 day periods in monolayer cultures with or without [3H] putrescine and/or DFMO. Cells in culture incubated for 20 days were replated at 4-day intervals. Cells were counted on a Coulter Electronic Particle Counter and percent viability was determined by eosin dye exclusion. Survival of cells with proliferative capacity was assayed by their colony. Forming ability and surviving fraction was calculated. The radioactive counts due to [3H] putrescine were measured in 9L cells and in medium and expressed as cpm/100 cells or cpm/ml, respectively., Results: As previously reported (15), DFMO treatment resulted in termination of cell proliferation that was reversible by the addition of exogenous putrescine. Specifically, after 4 days in culture, cell counts in groups exposed to 10 mM DFMO were 55% of those in control groups and addition of 3 mM putrescine reversed the DFMO effects. Uptake of [3H] putrescine into untreated cells increased in proportion to the amount of exogenous putrescine present during 4 days of culture (range 0.01 nmol to 100 nmol) and the presence of DFMO in the medium enhanced the uptake 9 fold throughout these ranges. At activities greater than 100 cpm/100 cells the cell count was reduced to 23 to 48% of control after 4 days in culture. Extending the treatment to 20 days of incubation increased the killing of 9L cells. During the 20-day incubation, control cells increased from 5 x 10(5) to 13 x 10(12) of which 90% were colony forming cells. Treatment with either 25 microCi [3H] putrescine or 1 mM DFMO for 4 days followed by removal of these agents and incubation for an additional 16 days for a total of 20 days resulted in 31 x 10(8) or 18 x 10(7) colony forming cells, respectively. Combining [3H] putrescine and DFMO treatments during the first 4 days of the 20 day incubation reduced the colony forming cells to 21 x 10(5) (surviving fraction to 67%). When the DFMO treatment was present during the entire 20 days, it became cytotoxic since the colony forming cells were reduced to 35 x 10(3) (surviving fraction was 17%). The combination of the 4-day [3H putrescine and the 20 day DFMO treatments resulted in only 1200 surviving colony forming cells (surviving fraction was only 2%)., Conclusion: DFMO treatment of 9L cells for 20 days resulted in increased uptake of [3H] putrescine, a 10(10) fold inhibition of colony forming cells and extensive 9L cell killing relative to untreated controls.

The features of a mouse long-term bone marrow culture (LTBMC) system that produces natural killer (NK) cell activity are described. Over a 4-week period in the NK-LTBMC, cellularity dropped from approximately 2.5 x 10(7) to 8 x 10(5) cells/25-cm2 flask. About 3 x 10(5) of these cells were loosely adherent. The cultures at this time contained about one-third the spleen colony forming units, one-tenth the granulocyte macrophage colony forming units and about one-third the transplantable NK progenitor activity of fresh bone marrow (BM), and no detectable NK cell activity. In the 4-week NK-LTBMC, IL-2-responsive precursor cells appeared to be selectively maintained and gave an 8-fold higher activity after culture with human recombinant IL-2 (rIL-2) than did fresh BM. The addition of 50-5,000 IU/ml of rIL-2 resulted, after a minimal 3-day lag, in progressively increased cellularity for as long as 13 days. The percentage and staining intensity of NK-1.1+ cells increased with time after addition of rIL-2. CD3 epsilon + cells were occasionally seen and B220+ cells were present in low numbers at day 7 and slowly increased through day 13. The stroma was necessary for IL-2-dependent development of NK activity.

In an effort to increase the long-term production of hematopoietic cells in vitro, Origen hybridoma cloning factor (HCF) was added at the initiation of Dexter type cultures, in which whole bone marrow (BM) was seeded into tissue culture flasks and formed an adherent stromal layer that supported the proliferation and differentiation of primitive cells. After about six weeks, all the cultures were fully established, and continuous production of nonadherent cells was maintained for at least 27 weeks. In the groups with 20% HCF, there was a significant (three- to fourfold) increase in the steady-state cell production of 106 +/- 17 x 10(4) cells/ml compared to 26 +/- 10 x 10(4) in controls. In some cases the ability of HCF to increase productivity was limited by the nutrients and metabolic products in the culture medium. Cell number varied inversely with glucose and pH. HCF increased the concentration and absolute number of myeloid progenitors (granulocyte-macrophage colony forming units and spleen colony forming units) in the nonadherent layer and shifted the differentiation of granulocyte-macrophage colony forming units toward the production of cells of the monocyte/macrophage lineage. Spleen colonies produced from 10(5) cells from cultures with HCF were more numerous (8 +/- 2 versus 4 +/- 2) and larger than those from control cultures (2.6 versus 0.2 mg/colony), but they contained the usual cell lineages (erythrocytic, granulocytic and megakaryocytic).

We report the induction of stable and reliably detectable mixed xenogeneic chimerism through the coadministration of a mixture of untreated rat bone marrow plus T cell-depleted mouse bone marrow into B10 recipients conditioned with total body irradiation (TCD B10 mouse + untreated F344 rat----B10 mouse). Recipients repopulated as true mixed lymphopoietic chimeras, with from 1-21.6% rat-derived lymphoid cells in peripheral blood and splenic lymphoid tissue. Production of rat platelets was also demonstrated. Rat platelet and lymphoid chimerism was reliably detectable in chimeras from 1 to 7 months following reconstitution, suggesting engraftment of the rat bone marrow stem cell. Production of each stem cell-derived lineage appeared to be under independent regulation since a significantly greater proportion of platelets were rat-derived (24-81%) than were lymphocytes (1-21.6% rat), while erythrocytes were preferentially syngeneic (less than 2% rat). The tolerance induced by this model was highly donor strain-specific: donor-specific rat and mouse skin grafts were accepted while MHC-disparate third-party mouse (C3H; H-2k) and rat (Wistar Furth; Rt1Au) skin grafts were promptly rejected. Although specifically prolonged xenogeneic donor rat skin grafts underwent a slow chronic rejection, and some totally disappeared. In spite of this, chimeras retained their lymphoid chimerism, suggesting the presence of skin-specific antigens. This model for mixed xenogeneic chimerism with reliably detectable rat lymphoid cells may provide a model to study the existence of tissue-specific antigens across a species barrier, as well as mechanisms responsible for the induction and maintenance of this strain-specific transplantation tolerance.

The macrocytic anemia of W/Wv mice can be cured by injection of +/+ bone marrow cells (BMC) from WBB6F1 mice. However, it has been observed that some W/Wv recipients appear to "lose" their cure with time, an effect that does not appear to be related to the age of the BMC donor. The present study was undertaken to determine the effect of recipient age on W/Wv responses to BMC injection. The effect of aging on erythroid parameters was similar in untreated W/Wv mice and +/+ controls. In both genotypes, hematocrit (HCT) and red blood cell count (RBC) decreased, and the modal red blood cell size (peak) increased between 13 and 150 weeks of age. As anticipated, mean HCT and RBC values were lower and peak values higher in W/Wv mice compared to +/+ controls at every age. However, the rate of decrease in HCT and RBC with age was the same for both genotypes, suggesting that the age effect and W gene effect were independent. Peak values increased slightly more with age for W/Wv than for +/+ controls. When female W/Wv mice in three age groups (23.5, 70, and 91.5 weeks old) were injected with 5 x 10(5) BMC from 20-week-old +/+ female donors and HCT, RBC, and peak were determined monthly, improvement was seen in most W/Wv recipients. However, in the older mice this improvement was slower and often was not sustained; 100% of the youngest recipients, 80% of the middle-aged, and only 30% of the older groups were cured after 3 months. Taken together, these data suggest a latent deficiency of the aging hematopoietic microenvironment that is revealed in W/Wv mice by the stress of continuing erythroid demand on the limited number of normal donor BMC.

Transplantation of untreated F344 rat bone marrow into irradiated B10 mouse recipients (non-TCD F344----B10) to produce fully xenogeneic chimeras resulted in stable xenogeneic lymphoid chimerism, ranging from 82% to 97% rat. Survival of animals was excellent, without evidence for GVH disease. The specificity of tolerance which resulted was highly donor-specific; MHC disparate third party mouse and rat skin grafts were promptly rejected while donor-specific F344 grafts were significantly prolonged (MST greater than 130 days). Multi-lineage rat stem cell-derived progeny including lymphoid cells (T- and B-lymphocytes), myeloid cells, erythrocytes, platelets, and natural killer (NK) cells were present in the fully xenogenic chimeras up to 7 months after bone marrow transplantation. Immature rat T-lymphocytes matured and acquired the alpha/beta T-cell receptor in the thymus of chimeras in a pattern similar to normal rat controls, suggesting that immature T-lymphocytes of rat origin could interact with the murine xenogeneic thymic stroma to undergo normal maturation and differentiation. This model may be useful to study the mechanisms responsible for the induction and maintenance of donor-specific transplantation tolerance across a species barrier.

A request by the Institutional Animal Care and Use Committee for an alternative to death as an end point in a cancer research project using a rat brain 9L tumor cell model led to a search for reliable criteria for predicting time of death in this type of experiment. These experiments evaluated the therapeutic effectiveness of radiation alone, continuous intracerebral infusions of 5-iodo-2-deoxyuridine (IUDR) alone, and a combination of both therapies. We found that a characteristic pattern of body weight changes occurs after injection of 9L tumor cells into the brain ventricles or parenchyma. The initial phase was characterized by a loss of body weight which appeared to be related to surgery and, in the irradiated groups, to the subsequent doses of radiation under anesthesia on days 4, 6, and 7. After this initial phase (phase 1), a second period of weight change (phase 2) which was characterized by an overall gain of body weight interrupted temporarily in 76 out of the 149 rats by reversible episodes of weight loss of 1 to 5 days duration. The length of this phase 2 weight gain period was significantly extended by XRT-IUDR treatment in the rats with intraparenchymal tumors. The third and final phase consisted of a period of irreversible weight loss which may be related to cachexia. The third phase was similar in duration for control, XRT, IUDR and XRT-IUDR groups of rats and had a mean length of 9.8 +/- 0.27 days.(ABSTRACT TRUNCATED AT 250 WORDS)

Optimal electroporation efficiency of many cell types is associated with poor survival. We show that serum rapidly reseals the membranes of electroporated cells and that timely addition of serum following electroporation can improve cell survival and transfection efficiency.

The efficacy of 5-iodo-2-deoxyuridine (IUDR) as a radiosensitizer when administered by continuous infusion into the cerebral spinal fluid (CSF) of the lateral cerebral ventricle was evaluated in a 9L gliosarcoma rat brain tumor model. Stereotactic implantation of a 5 x 10(4) tumor cell suspension into the left caudate nucleus was carried out in four groups of 10 rats each. Control animals had a median survival of 16.9 days (range 16-21 days). IUDR, 8.4 mg over 7 days administered by continuous infusion into the left lateral ventricle produced a slight survival advantage (median survival 21.5 days, range 12-56). Irradiation of the entire brain, 8 Gy on days 4, 6 and 7 after tumor cell implantation also produced a slight improvement in survival (median 19.5 days, range 17-34). The combination of radiation and IUDR infusion into the CSF produced a marked survival advantage (median 30.5, range 22-54) compared to the control and single modality treatment groups. This is the first demonstration of the effectiveness of IUDR as a radiosensitizer when administered into the lateral cerebral ventricle in the treatment of an intraparenchymal brain tumor.

In this report, we describe a novel long-term bone marrow culture (LTBMC) system to study the origin and generation of natural killer (NK) cells from NK precursors. Rat bone marrow was cultured for 4 wk in RPMI 1640 with 5% fetal calf serum and 2-mercaptoethanol to allow the formation of an adherent stromal cell layer containing NK precursor cells. After addition of interleukin 2 (IL-2), the LTBMC generated high numbers (up to 100-fold expansion in 7 d) of pure 3.2.3+ large granular lymphocytes with lytic activity against NK-sensitive and -resistant tumor targets, as well as antibody-dependent cellular cytotoxicity. NK activity in LTBMC could be detected 3 d after addition of as little as 1 U/ml rIL-2, whereas lymphokine-activated killer activity was found 5 d after addition of at least 10 U/ml rIL-2. In vivo depletion and in vitro complement lysis studies showed that the NK precursor cells in LTBMC did not express the NK-associated surface markers asialo GM1 or 3.2.3. We also found that LTBMC cells did not exhibit colony growth in granulocyte/macrophage or spleen colony-forming unit assays. The generation of NK cells from NK precursors required, in addition to IL-2, a second growth/maturation factor(s), which was present in the conditioned medium of the LTBMC. This LTBMC system provides a unique in vitro model to study the development of NK cells from precursor cells, the role of the bone marrow stromal microenvironment in this development, and the lineage relationship of NK cells to other hematopoietic cells.

We evaluated the effects of 16,16-dimethyl prostaglandin E2 (dm-PGE2), with and without syngeneic bone marrow transplantation (BMT) on the survival and hematopoietic recovery of mice given 14-20 Gy total body irradiation (TBI). Survival of mice given combined dm-PGE2 and BMT was improved significantly over that of mice given either treatment alone. The 30-day survival after 14, 15, 16 or 18 Gy TBI for combined treatment was 97, 90, 20 or 10 percent, respectively. The corresponding 30-day survival rates for mice given BMT alone were 69, 60, 7 or 0 percent, respectively. For dm-PGE2 alone, 30-day survival was 63, 20, 10 or 0 percent, respectively. Deaths in both dm-PGE2 treated groups generally occurred after day 10 whereas deaths in the BMT group occurred before day 10. All irradiated controls were dead on or before day 10; after larger doses, deaths clustered around day 5. After 20 Gy TBI, all mice in all groups were dead by day 7. Studies of white blood cell recovery 1-9 days after 14 Gy TBI showed improvement with BMT, whereas dm-PGE2 did not enhance recovery. Nucleated cells per humerus, spleen weight, and spleen iron uptake (erythropoiesis) were also improved by BMT but not dm-PGE2.

A rat brain tumor model has been developed with the clinical and pathological features of dissemination via the cerebral spinal fluid (CSF) pathways. A precise number of 9L gliosarcoma cells (5 x 10(2) to 5 x 10(5)) is stereotactically injected into the CSF of the lateral ventricle. The interval until the onset of neurological symptoms and then death is reproducible and dependent upon the number of cells injected. The median survival of three groups of rats receiving 5 x 10(5) cells in three different experiments was 17, 18 and 19 days respectively. For three groups receiving 5 x 10(4) cells, the median survival was 23, 24 and 25.5 days respectively and for two groups receiving 5 x 10(3) cells the median survival was 28 and 30 days respectively. The animals developed multiple tumor implants along the CSF pathways usually resulting in hydrocephalus. This tumor model was developed to simulate dissemination via CSF pathways as seen with medulloblastoma and other primitive neuroectodermal tumors of the central nervous system. It will be used to evaluate the therapeutic efficacy of intraventricularly administered anti-neoplastic drugs against small implants and malignant cells in the CSF pathways.