Non-resorbing osteoclasts induce migration and osteogenic differentiation of mesenchymal stem cells.

Osteoclast activity has traditionally been regarded as restricted to bone resorption but there is some evidence that also non-resorbing osteoclasts might influence osteoblast activity. The aim of the present study was to further investigate the hypothesis of an anabolic function of non-resorbing osteoclasts by investigating their capability to recruit mesenchymal stem cells (MSC) and to provoke their differentiation toward the osteogenic lineage. Bone-marrow-derived human MSC were exposed to conditioned media (CM) derived from non-resorbing osteoclast cultures, which were generated from human peripheral blood monocytes. Osteogenic marker genes (transcription factor Runx2, bone sialoprotein, alkaline phosphatase (AP), and osteopontin) were significantly increased. Osteogenic differentiation (OD) was also proved by von Kossa and AP staining occurred in the same range as in MSC cultures stimulated with osteogenic supplements. Chemotactic responses of MSC were measured with a modified Boyden chamber assay. CM from osteoclast cultures induced a strong migratory response in MSC, which was greatly reduced in the presence of an anti-human platelet-derived growth factor (PDGF) receptor beta antibody. Correspondingly, significantly increased PDGF-BB concentrations were measured in the CM using a PDGF-BB immunoassay. CM derived from mononuclear cell cultures did not provoke MSC differentiation and had a significantly lower migratory effect on MSC suggesting that the effects were specifically mediated by osteoclasts. In conclusion, it can be suggested that human non-resorbing osteoclasts induce migration and OD of MSC. While effects on MSC migration might be mainly due to PDGF-BB, the factors inducing OD remain to be elucidated.

In vitro studies on the radiosensitivity of multipotent hemopoietic progenitors in canine bone marrow.

The in vitro radiation response to 280-kV x-rays (does rate 72 cGy/min) of multipotent hemopoietic progenitor cells, mixed colony-forming units (CFU-mix), from canine bone marrow was assayed and compared to the radiation response characteristics of early erythroid progenitors, erythroid burst-forming units (BFU-E). To improve the colony-forming efficiency, the effect of various bone marrow cell separation techniques on colony formation of both progenitors was examined. The separation of bone marrow aspirates by discontinuous buoyant gradient centrifugation using the lymphocyte separation medium Lymphoprep with a density of 1.070 g/ml allowed the establishment of reproducible survival curves. The survival curves for both progenitors were strictly exponential, and CFU-mix were found to be more radiosensitive (D0 = 12 +/- 2 cGy) than BFU-E (D0 = 16 +/- 2 cGy).

Growth of erythroid burst-forming units (BFU-E) in cultures of canine bone marrow and peripheral blood cells: effect of serum from irradiated dogs.

Erythroid burst-forming units (BFU-E) from canine bone marrow and peripheral blood could be grown in methylcellulose in the presence of an appropriate batch of fetal calf serum (FCS), transferrin, and erythropoietin (Epo). However, improved colony formation (size and number of bursts) was obtained when serum from total body irradiated dogs was present in the culture. This serum, obtained from dogs at day 9 after total body irradiation with a dose of 3.9 Gy, reduced markedly the Epo requirement of BFU-E. Furthermore, it allowed the omission of FCS from the culture medium if cholesterol and bovine serum albumin (BSA) were used as FCS substitutes. BFU-E concentrations were found to be rather different in the peripheral blood and in bone marrow samples from different sites (i.e., iliac crest, sternum, and humerus) of normal beagles. The studies further show that canine bone marrow BFU-E can be cryopreserved in liquid nitrogen.

In vivo studies on hemopoietic stem cells and target cells for Friend virus infection in vitro.

Friend virus replicating target cells have been characterized by in vitro infection of bone marrow cells from DBA/2 mice after multiple injections of hydroxyurea (HU), after bleeding and/or hypertransfusion and after busulfan treatment. The concentration of CFUS, BFUE and CFUE in the cell suspension was correlated with the number of infectious centers (IC) induced after infection. HU treated mice were also infected in vivo. The results demonstrate that there is almost no target cell 4 h after the last of 4 HU injections. The number of IC induced could correlate with the numbers of CFUS in the HU experiment, but this possibility can be rejected based on the busulfan experiments. All results, including those after bleeding and/or hypertransfusion are compatible with a target cell between BFUE and CFUE, as suggested by other in vitro infection methods and also in vivo experiments.

Hematopoietic stem cells in Friend murine leukemia virus infected mice undergoing chemotherapy: failure to eradicate Ep-independent erythropoiesis by BCNU.

DBA/2 mice, 12 days after infection with the polycythemia-inducing Friend virus (F-MulV-P), were treated with a single dose of BCNU. The effect of 35 mg/kg and lower doses on the different stem cell pools (CFUS, CFUC, CFUE) and the F-MulV-P specific ICPC (infectious centers producing cells) and Ep-independent CFUE was studied in detail. The depression of CFUS and CFUC was comparable in normal and F-MuLV-P infected mice. The CFUE population of virus infected mice, which had been completely Ep independent before treatment, showed only some normal Ep dependent colony growth thereafter. This finding was in contrast to the effect of multiple doses of Hydroxyurea, studied previously, which had given the same quantitative depression of CFUS and CFUC. Possible reasons for this failure are discussed; they may be the more proliferation independent mode of action of BCNU and the delayed hemopoietic regeneration.

Erythroid stem cell regeneration in normal and plethoric mice treated by cytosinarabinoside.

Cytosinarabinoside (6 x 40 mg/kg in 4 h intervals) was used to study hemopoietic regeneration in normal and in hypertransfused mice. About 70-80% of pluripotent stem cells (CFUs), more than 80% of granuloid committed precursors (CFUc) and more than 95% of erythroid committed precursors (CFUE) were eliminated after treatment, and also more than 99% of the erythroblasts in the bone marrow smears. There was no regular difference between normal and hypertransfused groups. Regeneration started at day 2, with control levels of CFUs at day 3 and then overshooting, control levels of CFUc at days 4-5 and a single increase of CFUE at day 3. This rise in CFUE, with values well above control levels, was also present in hypertransfused mice, although less pronounced. In hypertransfused mice the CFUE peak was not allowed by an increase in erythroblast numbers in the femora or 6 h 59Fe uptake into femora or spleen-in contrast to non-plethoric controls. The results indicate the partial Ep independence of regenerating CFUE, as found after treatment with hydroxyurea.

On the role of the spleen in Friend virus (F-MuLV-P) erythroleukemia.

About 30% of Friend virus (F-MuLV-P)-infected C57BL/6 mice became leukemic more than ten weeks after virus infection. This late leukemia development could not be essentially influenced by drug treatment, such as injection of hydroxyurea (2 X 500 mg/kg), actinomycin D (3 X 120 micrograms/kg), phenylhydrazine (3 X 60 mg/kg), or 30 micrograms endotoxin or by bone marrow transplantation following lethal irradiation. Endotoxin (30 micrograms) given prior to virus caused the loss of resistance to F-MuLV-P, but it had only a slight effect if applied one or three months after virus infection. Leukemia development has never been observed in C57BL/6 mice after splenectomy. In DBA/2 mice, highly susceptible to F-MuLV-P, leukemia development was markedly impaired if the mice were splenectomized. These results clearly indicate that the spleen plays a crucial role in the mechanism of susceptibility or resistance to the Friend virus.

Characterization of a Friend virus-replicating target cell.

The FV-replicating target cell characterized in this study is an erythroid precursor cell, as shown by its response to bleeding and hypertransfusion. The target cell is not identical with the BFU-E compartment as demonstrated by different model velocities and different cell cycle characteristics as compared to BFU-E. Different sensitivity of FV-target cells and CFU-E to AMD, differences in the growth kinetics of both cell populations in bone marrow and spleen of mice after bleeding, and large quantitative differences of both cell populations in hemopoietic organs, suggest only partial identity or non-identity of these two cell types. Finally it is shown that spleen colonies originating from FV-infected target cells are a primary erythroleukemic lesion, since they contain more Ep-independent CFU-E than do intercolony areas of the spleen.

Natural killer (NK) cells and interferon in Friend-virus-infected susceptible and resistant mice.

Natural killer (NK) cells, as determined by the lysis of YAC-1 and Eveline target cells, were determined in F-MuLV-P susceptible DBA/2 and resistant C57/Bl/6 and in beige mice. No difference was found between DBA/2 and C57/Bl/6 mice. Spleens from beige mice showed, as expected, a much lower lysis. The NK cell activity and interferon levels were also studied in these mice after F-MuLV-P infection. Peak NK activities of spleen cells against both target cell types were found at day 3; peak interferon levels were detected in the serum 24 h after infection in all three strains. At day 12, the splenic NK cell activity remained above control levels only in C57/Bl/6 mice. Beige mice were as resistant to Friend leukemia development as C57/Bl/6 mice. The injection of endotoxin, which makes C57/Bl/6 mice virus-susceptible, was followed by increased NK cell activity. The significance of the NK cell system for the resistance of C57/Bl/6 mice against the development of Friend leukemia is discussed.

Hematologic effects of recombinant human interleukin-6 in dogs exposed to a total-body radiation dose of 2.4 Gy.

The hematologic effects of recombinant human interleukin-6 (rhIL-6) were studied in dogs exposed to a total-body irradiation (TBI) of 2.4 Gy. IL-6 was administered over a period of 14 days at a daily dose of 18 micrograms/kg by single subcutaneous injection. Treatment was started 1 day after TBI. The data obtained for the different hematologic parameters of the irradiated IL-6-treated dogs were compared with the data obtained from dogs who received TBI of 2.4 Gy and were treated with the carrier (control). No clear influence of IL-6 treatment on the pattern of recovery of lymphocytes could be detected in comparison to the irradiated control animals. The thrombocyte counts in the period from day 1 to 16 after TBI were similar for both groups of dogs, showing a sharp decrease in counts between days 6 and 12 with a stabilization thereafter at approximately 30 x 10(3)/microL. In three of the four IL-6-treated dogs, however, thrombocyte counts increased at day 18 after the beginning of treatment. This increase occurred 7 days earlier than in the controls. In two of the three dogs showing an accelerated recovery of platelet counts, however, treatment with IL-6 caused a strong decrease in the erythrocyte counts associated with a prolonged depression in reticulocyte concentration. There was no influence on the recovery of blood granulocytes. In one of the animals responding with an accelerated thrombocyte recovery, IL-6 had no adverse effect on erythropoiesis. However, IL-6 forced the recovery of blood granulocytes in the period beyond day 10 after TBI. Another animal showed no influence of IL-6 on thrombocyte recovery but a strong depressive effect on erythrocyte and reticulocyte counts. The results show that for standardized conditions of radiation-induced bone marrow damage, the pattern of response to IL-6 in different hematopoietic lineages may show considerable variations between individuals, in contrast to what has been observed in irradiated animals treated with granulocyte-macrophage or granulocyte colony-stimulating factor (GM- or G-CSF).

The effect of recombinant human stem cell factor and basic fibroblast growth factor on the in vitro radiosensitivity of CD34+ hematopoietic progenitors from human umbilical cord blood.

Human umbilical cord blood (CB) cells selected by immunomagnetic beads for expression of the CD34 antigen were irradiated with increasing doses of x-rays (72 cGy/min). Clonogenic survival of the hematopoietic progenitors, including mixed colony-forming cells (Mix-CFC), erythroid burst-forming units (BFU-E), and granulocyte-macrophage colony-forming cells (GM-CFC), was determined in methylcellulose cultures containing placenta conditioned medium (PCM) and erythropoietin (Epo). Exponential survival curves were fitted to the data of all the colonies, resulting in D0 = 95 cGy for Mix-CFC, 136 cGy for BFU-E, and 136 cGy for GM-CFC. Additionally, the radiosensitivity of CD34+ cells was studied employing cultures containing either recombinant human stem cell factor (rhSCF) or basic fibroblast growth factor (b-FGF) in combination with PCM and Epo. It was found that the colony-forming efficiency (CFE) of non-irradiated CD34+ cells of 5.5% (range 1.4 to 14.4%) did not increase after the addition of SCF or b-FGF to the culture. The radiation response characteristics showed, however, that in the presence of SCF, the D0 value and the extrapolation number n increased significantly. This suggests the stimulation of what operationally is termed "recovery from potentially lethal damage." In contrast, no response modifying effect could be seen for b-FGF.

Acceleration of hemopoietic recovery in dogs after extended-field partial-body irradiation by treatment with colony-stimulating factors: rhG-CSF and rhGM-CSF.

PURPOSE: The influence of treatment with the two colony-stimulating factors, rhG-CSF and rhGM-CSF, on the hemopoietic recovery in aplastic bone marrow sites after extended-field irradiation was studied in a canine model. METHODS AND MATERIALS: The dogs received irradiation of the cranial part of their body with a single dose of 11.7 Gy, comprising approximately 72% of the total bone marrow mass. Anatomically this type of exposure corresponds to upper body irradiation (UBI) as employed under clinical conditions. Treatment with both the CSFs was employed for 7 days by daily injections of 30 microg/kg, starting 24 hr after irradiation. RESULTS: Treatment with rhGM-CSF did not completely prevent the initial decrease of the granulocyte counts, but caused an accelerated, though incomplete, recovery in the period from day 5 to day 15. In contrast, treatment with rhG-CSF caused two phases of granulocytosis and an early recovery to normal levels at day 11 after irradiation. Treatment with rhG-CSF, but not with rhGM-CSF, was associated with a strong supra-normal increase of progenitor cells in the blood within the first 8 days and an accelerated hemopoietic recovery in the irradiated sites particularly within the first 7 days after the exposure. CONCLUSIONS: These results indicate that under conditions of partial-body irradiation short term treatment with G-CSF is superior to GM-CSF in initiating the hemopoietic recovery on the basis of endogenous stem cell seeding.

Donor cell leukemia after bone marrow transplantation in the Friend leukemia in mice.

After lethal irradiation (800 R) Friend virus (FV-P)-infected leukemic DBA/2 mice were transplanted with normal bone marrow cells. Isogeneic transplantation led to an immediate relapse of leukemia. Therefore, allogeneic bone marrow cells were taken from almost FV-P resistant C57BL/6 mice. A measure of leukemia development was given by the number of erythropoietin-independent erythroid colonies (CFU-EI) in bone marrow and spleen, characteristic for the Friend leukemia. Even after allogeneic transplantation leukemia recurred after 5 to 19 days. By an electrophoretic analysis of the hemoglobin, it could be shown that the transformed erythropoiesis was donor derived. Thus, marrow of C57BL/6 origin loses its FV-P resistance in allogeneic leukemic lethally irradiated recipients and is transformed by the surviving virus.

T-cell markers in the thymus during methylnitrosourea-induced leukemogenesis.

T-cell leukemias were induced in adult BDF1 mice by a single i.v. injection of 50 mg/kg of methylnitrosourea (MNU). During the latency period and in leukemic mice the expression of peanut agglutinin (PNA) receptors, terminal deoxynucleotidyl transferase (TdT) and thymus leukemia antigen (TLm4) in the thymus was studied. First, TLm4-positive thymocytes were found 6 wk after MNU. Further, there was a continuous reduction of the percentage of PNA-positive thymocytes. The expression of TdT remained unaltered late in the latency period and in leukemic mice. Four phenotype patterns could be defined for the expression of TdT and PNA in thymocytes: PNA+ TdT+, PNA+ TdT-, PNA- TdT+ and PNA- TdT-. Thymuses negative for TdT were rare, the majority was PNA+ TdT+. However, thymuses late after MNU treatment and leukemic thymus were PNA- TdT+. The discrepancy in the expression of PNA and TdT as cellular markers in leukemic mice as compared to normal cell populations is discussed.

The sensitivity to methylnitrosourea-induced T-cell leukemogenesis in mice is modified by the injection of hydrocortisone.

T-cell leukemias were induced in adult BDF1 mice by a single i.v. injection of methylnitrosourea (MNU). Leukemogenesis was delayed by a single or repeated injections of hydrocortisone (HC) after MNU and also when HC was given one day before MNU. Enhancement of leukemogenesis was seen in experiments with 10 and 14 days' intervals between HC and MNU. The T-cell subset composition of the thymus after HC treatment was studied at these time intervals, but a specific target cell for the action of MNU, reduced one day after HC and increased in number during the thymic regeneration at 10 and 14 days could not be defined. HC did not prohibit the toxic action of MNU as measured by hemopoietic stem cell numbers in the femur.

T-cell phenotypes in chemically induced leukemia in mice.

T-cell leukemias were induced in adult BDF mice by a single i.v. injection of 50 mg kg-1 of methylnitrosourea (MNU). Leukemic cells from the thymus and other organ sites showed the theta antigen and were peanut-negative, but were heterogeneous with respect to Lyt-1 and Lyt-2. The acute cytotoxic effect of the MNU showed no preferential toxicity to a special T-cell subset in the thymus. During the latency period of leukemogenesis a continuous fall in peanut-positive cells in the thymus was found.

Natural killer cells in leukemogenesis.

In order to relate a reduced natural killer (NK) cell function to leukemogenesis, NK cells in the spleen and peritoneal exudate cells, with and without stimulation by Corynebacterium parvum, were tested in mice of various strains after split dose irradiation and after leukemogenic treatment with butyl- and methylnitrosourea. The investigations included also mice submitted to non-leukemogenic irradiation (1 X 1.5 and 1 X 4.5 Gy) and mice submitted to an additional treatment with hydrocortisone, which delays leukemia development after methylnitrosourea. There was, indeed, a NK-cell depression, but no major differences were seen between mice prone to leukemia development and those after cytotoxic, but nonleukemogenic, treatment.

Leukemia induction by methylnitrosourea (MNU) in selected mouse strains. Effects of MNU on hemopoietic stem cells, the immune system and natural killer cells.

T cell leukemias were induced by a single dose of methylnitrosourea (MNU) in DBA/2, C57/Bl/6, NMRI, BDF1, and CBA mice. The latency period in the CBA strain was much longer than in the others. Studies on the pluripotent stem cells in the bone marrow and T cell reactions of thymus and spleen cells showed a toxicity of MNU for these parameters but no significant differences between the strains. The activity of the natural killer (NK) cells in the spleen and peritoneal exudate cells, studied also after additional stimulation by injection of Corynebacterium parvum, was influenced by MNU, but again no relation to leukemogenesis could be established. The first leukemic (transplantable) cells were found in the thymus. The presence of leukemic cells could be responsible for low NK cell activities found in BDF1 and DBA/2 mice late after MNU.

The seeding of a transplanted murine leukemia and its influence on hemopoietic stem cells and immune function of the host.

A butylnitrosourea-induced murine T-cell leukemia, L40, was transplanted in BDF1 mice; 1 X 10(3) cells killed all recipients after conditioning with 400 rad, whereas 1 X 10(5) were needed with normal recipients. No leukemic cells could be detected by transplantation or cytogenetic analysis in the femur or the spleen at day 6 after L40 inoculation and, at day 11, leukemic cells were found in one out of two experiments, more if the host had been irradiated. Up to day 17, when leukemic cells were present, the CFU-S and CFU-C content of the femur was normal, but later a loss was observed with an increase in the enlarging spleen. Lymphocyte-stimulation assays with spleen cells gave normal results up to day 17, but later the 3H-thymidine uptake of stimulated T and B cells was reduced. The NK-cell activity with and without stimulation by Corynebacterium parvum in the spleen began to fall at day 17 and was absent later; this loss was also observed with peritoneal exudate cells. In vitro mixing experiments of L40 cells with normal spleen cells showed "cold target inhibition" by L40 cells in the NK-cell assay as well as interference with the lymphocyte stimulation.

Effects of total-body irradiation on bone marrow erythroid burst-forming units (BFU-E) and hemopoietic regeneration in dogs.

The acute and long-term effects of total-body X irradiation (TBI) on early erythroid progenitors, burst-forming units (BFU-E), in the bone marrow of beagles were studied for midline tissue doses of 1.6 and 2.4 Gy. After both radiation doses the initial reduction in the concentration of BFU-E was greater than that found for the granulocyte-macrophage progenitor cells (GM-CFC) and thus was in general agreement with the higher in vitro radiosensitivity of BFU-E compared to GM-CFC. After TBI with 1.6 Gy the GM-CFC and BFU-E returned to their normal levels within 2-4 weeks without showing long-term radiation effects. In contrast, after TBI with 2.4 Gy the concentrations of GM-CFC and BFU-E remained below the pretreatment levels up to 1 year after exposure. For a given midline tissue dose, the extent of the long-term effect of radiation on the BFU-E in a certain bone marrow site seems to be dependent on the local radiation dose in the respective bone marrow space. The minor radiation effects observed in the erythrocyte concentration in the peripheral blood, the hematocrit, and the hemoglobin concentration point to the enormous compensatory capacity of the more mature erythropoietic transit population to increase the proliferative capacity upon demand.