Recombinant human granulocyte colony-stimulating factor accelerates hematopoietic recovery after DLA-identical littermate marrow transplants in dogs.

We studied whether treatment of dogs with recombinant human granulocyte colony-stimulating factor (rhG-CSF), after 920 cGy total body irradiation (TBI) and transplantation of 3.3 +/- 1.0 x 10(8) bone marrow cells per kilogram from a DLA-identical littermate, accelerated hematopoietic recovery and influenced the incidence of subsequent marrow graft failure or graft-versus-host disease (GVHD). Ten animals were treated with 100 micrograms rhG-CSF/kg/d from days 1 through 10 after TBI. Results were compared with those of historical control of 14 dogs not administered rhG-CSF. Neither group of dogs received GVHD prophylaxis. The median time to recovery of 1,000 neutrophils/mm3 was 8 days for dogs administered rhG-CSF compared with 14 days in controls (logrank test: P less than .03). The median time to reach 100 monocytes/mm3 was 17 days in G-CSF-treated dogs compared with 49 days in controls (P less than .002). The median time to attain 500 lymphocytes/mm3 was 15 days versus 31 days, respectively (P less than .01). The median time to reach 20,000 platelets/mm3 was 26 versus 20 days (P = .68). Graft failure occurred in 1 of 10 G-CSF-treated dogs versus 2 of 14 controls (two-tailed Fisher's exact test: P = 1.00). GVHD was seen in 4 of 9 rhG-CSF-treated dogs compared with 1 of 12 controls (P = .12). Two G-CSF-treated dogs died of GVHD versus none of the controls (P = .17). No unusual toxicities were seen in dogs receiving rhG-CSF. In summary, rhG-CSF significantly accelerated recovery of neutrophils, monocytes, and lymphocytes after DLA-identical littermate marrow transplantation without altering platelet recovery. Graft failure was not seen more often than in controls, but there was a trend toward an increased incidence of GVHD.     

Improved retroviral transfer of genes into canine hematopoietic progenitor cells kept in long-term marrow culture

Amphotropic helper-free retroviral vectors containing either the bacterial neomycin phosphotransferase gene (NEO) or a mutant dihydrofolate reductase gene (DHFR*) were used to infect canine hematopoietic progenitor cells. In previous experiments, successful transfer and expression of both genes in canine CFU-GM were achieved after 24-hour cocultivation with virus-producing cells. The average rate of gene expression was 10% (6% to 16%) as measured by the number of CFU-GM resistant to either the aminoglycoside G418 or methotrexate. In an attempt to increase the efficiency of gene transfer, marrow was cocultured for 24 hours with either NEO or DHFR* virus-producing packaging cells and then kept in long-term marrow culture fed three times with virus-containing supernatant (2 to 5 x 10(6) CFU/mL). After six days, cells were harvested and cultured in CFU-GM assay with and without a selective agent. The average rate of gene expression in CFU- GM in five independent experiments was 46% and ranged from 19% to 87%. In conclusion, the efficiency of gene transfer into canine hematopoietic progenitor cells has been increased fourfold by combining cocultivation with long-term marrow culture as compared with results obtained with cocultivation only.     

Prevention of transfusion-induced graft-versus-host disease in dogs by ultraviolet irradiation

Ten dogs were given 9.2 Gy of total body irradiation and autologous bone marrow infusion followed by ten daily transfusions of leukocytes for a total of 11.5 to 36.2 (median, 18.8) x 10(8)/kg obtained via leukapheresis from histoincompatible unrelated donors. Four dogs were given unirradiated leukocytes, and all developed graft-versus-host disease (GVHD). In contrast, only two of three dogs given leukocytes irradiated with 20 mJ/cm2 of ultraviolet (UV) light (200 to 300 nm), and none of three dogs given leukocytes irradiated with 1,000 mJ/cm2 developed GVHD. These data indicate that UV irradiation abrogates the alloreactive potential of transfused leukocytes, and suggest that UV irradiation can be used to prevent the development of transfusion- induced GVHD.     

Allogenic transplant of canine peripheral blood stem cells mobilized by recombinant canine hematopoietic growth factors

We have studied graft-versus-host disease (GVHD) after transplantation of allogeneic peripheral blood stem cells (PBSC) mobilized by either recombinant canine granulocyte colony-stimulating factor (rcG-CSF) alone or combined with stem cell factor (rcSCF). These studies were prompted by the observation of extremely rapid and sustained engraftment of growth factor-mobilized PBSC in the autologous setting using genetically marked cells and changes in function of T lymphocytes from donors that had undergone mobilization. Specifically, lymphocytes from growth factor-treated donors were hyporesponsive in mixed leukocyte culture and in response to Con A, raising hopes that GVHD in dogs given growth factor mobilized allogenic PBSC might be altered in a beneficial way. Eighteen dogs were given a median of 17.1 x 10(8) PBSC/kg from littermate donors after 920 cGy of total body irradiation without postgrafting immunosuppression. Donors were either genotypically DLA-identical (n = 9) or DLA-haploidentical (n = 9). The median number of colony-forming unit-granulocyte macrophage (CFU-GM) infused was 27 x 10(4)/kg, and the number of CD34+ cells in the transplant was on the order of 4.6 x 10(6)/kg. The dogs received a median of 52.8 x 10(7) CD4 cells/kg and 13.7 X 10(7) CD8 cells/kg. All 18 dogs had prompt hematopoietic engraftment of donor cells as assessed by chimerism studies using variable number tandem repeat, as well as cytogenetic markers. Three of the nine dogs given grafts from DLA- identical littermates had fatal GVHD, five had transient GVHD, and one had no GVHD. All nine DLA-haploidentical recipients of PBSC developed fatal hyperacute GVHD. In conclusion, the expectation about rapid engraftment was fulfilled. However, incidence and severity of acute GVHD after transplantation of mobilized PBSC were not different than previously reported for nonmobilized PBSC or marrow. This model will allow for further studies, including T-cell depletion to minimize GVHD without increasing graft rejection.     

Functional dendritic cells are required for transfusion-induced sensitization in canine marrow graft recipients

Untransfused dogs given 9.2 Gy total-body irradiation and hematopoietic grafts from DLA-identical littermate donors uniformly achieve sustained engraftment, whereas dogs given three transfusions (Tx) of whole blood from the intended marrow donor 24, 17 and 10 days pretransplant uniformly reject their graft. Sensitization appears to be mediated by peripheral blood mononuclear cells and can be prevented by prior irradiation of the Tx product with UV light, known to inactivate leukocytes, in particular, cells with accessory function. In the present study we investigated which leukocyte population was responsible for Tx-induced sensitization and subsequent marrow graft rejection. Surprisingly, neither monocytes nor macrophages or dendritic cells induced sensitization, and all dogs so treated achieved engraftment; however, all four evaluable dogs transfused with UV- exposed blood to which small numbers of normal dendritic cells (12.5 x 10(3)/kg) were added rejected their marrow graft. Among five dogs given UV-exposed blood and normal monocytes (12.5 x 10(3)/kg) only one rejected its graft, and four achieved sustained engraftment. We conclude that donor dendritic cells are necessary, albeit not sufficient for in vivo sensitization. Sensitization is prevented by elimination or inactivation of dendritic cell.