Hormone-Independent In Vitro Erythroid Colony Formation by Bone Marrow Cells From Rauscher Virus-Infected Mice Kees
1,2
Nooter,s and Riccardo Ghio 4,6
SUMMARY-Bone marrow cells from BALB/c mice Infected with Rauscher erythroblastosis virus produced five to twenty· five times more erythroid colonies in vitro in the absence of erythropoietin (EP) as compared to normal cells. A good eerrelatlon existed between the state of the disease and the number of hormone-independent erythroid colony-forming cells (CFU· E). A significant number of hormone·independent CFU·E was found as early as 3 days after Infection. A linear relationship existed between the number of cells plated and the number of erythroid colonies formed in vitro. Addition of EP did not enhance colony formation, even at low cell concentrations. Feeder layer experiments demonstrated that EP-independent colony formation was not due to the production of endoge. nous EP. Repeated injections of phenylhydrazine Into normal mice did not lead to the loss of EP responsiveness in vitro; this indicated that the hormone independency Induced by the virus was not due to continuous erythropoietic stimulation in vivo. Besides hormone Independency, the CFU-E from Infected mice required less serum in the culture medium. Normal ery· throld colonies regressed after 4 days of culture, but EP·inde· pendent colonies from infected mice persisted for more than 2 weeks. These three phenomena may be regarded as indicative for a physiologic transformation.-J Natl Cancer Inst 55: 59-64, 1975.
COLONY FORMATION by bone marrow cells from normal BALBJc mice may differ from that in Rauscher virus-infected mice. Murine leukemia viruses replicate well in fibroblasts, but in contrast to sarcoma viruses (1, 2) transformation by this group of viruses is rare (3-6). Several members of the group induce various kinds of hematopoietic neoplasms in vivo within a relatively short period. It is obvious that, for the assessment of the transforming capacity of leukemia viruses in vitro, hematopoietic cells and not fibroblasts are the appropriate target cells. To develop a suitable transformation assay, possible discrimination between normal and leukemic cells on the basis of differences in growth patterns in vitro is required. Techniques have been developed to produce myeloid (7-9) and erythroid colonies (10, 11) in vitro from mouse bone marrow cells. Our study represents a search for differences in colony formation between bone marrow cells from normal BALB Jc mice and BALB Jc mice infected with Rauscher erythroblastosis virus (12-14). MATERIALS AND METHODS
Virus isolation.-Rauscher leukemia virus (RLV) was isolated from plasma of leukemic BALB Jc mice. Plasma (100 ml) was centrifuged for 20 minutes at 1,200xg. The supernatant was spun for 20 minutes at 10,000 rpm (± 10,000Xg). An equal volume of buffer (1.5 mM Tris-HCI+0.5 M sucrose, pH 7.2) was added to the supernatant and centrifuged for 30 minutes at 228,000Xg (Beckman 50 Ti rotor; 50,000 rpm). The virus pellet was resuspended in I ml buffer (1.5 mM TrisHCl+0.025 M sucrose, pH 7.2) and spun for 25 minutes at 23,000 rpm (Beckman SW 27, I rotor; 85,000xg) on
a linear gradient (5-20% sucrose in 1.5 mM Tris-HCl, pH 7.2). The obtained opalescent band contained purified RLV. The virus was stored in liquid nitrogen until used. The virus preparation was assayed with the reverse XC plaque test of Niwa et al. (15) and contained 3 X 107 plaque-forming unitsJml. The purified virus stock was not contaminated with lactic dehydrogenase virus. It contained 8 mg proteinJml as measured by the method of Lowry et al. (16). Mice.-Six-week-old female BALBJc mice were given ip injections of approximately 0.01 mg purified virus and thereafter examined twice a week for malignancy. A continuing increase in spleen weight began at day 5 post infection; a maximum weight of 3 g was reached in the terminal stage (4-5 wk post inoculation). Six- to eight-week-old female BALBJc mice became anemic as a result of sc injections of 0.5 mg phenylhydrazine for 3 consecutive days. We determined the hematocrit 2 days later by puncturing the orbital venous plexus. Hypertransfusion experiments were done with normal and RLV-infected BALBJc mice. Blood for transfusions was collected aseptically in heparinized phosphate-buffered saline and washed three times. Each recipient was inoculated ip with 0.5 ml packed red cells on 3 successive days. The hematocrit was determined before and 3 days after transfusion when the animals were killed for bone marrow collection. Bone marrow cwltures.-The technique was based on that of Stephenson et al. (10). Bone marrow cells were aspirated from the femurs. The bone marrow cells were washed in Hanks' balanced salt solution by centrifugation at 250 X g at 4 0 C. Thereafter, 2 X 105 nucleated cells were plated in 35-mm plastic petri dishes (Greiner, Niirtingen, Germany) along with I ml of a mixture of Dulbecco's modified Eagle's minimum essential medium (DMEM) (Flow Laboratories, Irvine, Scotland), 0.8% methyl cellulose (Methocel, 4,000 cps; Dow Chemical Co., Bogota, Colombia), 30% fetal bovine serum (FBS) (Flow Laboratories), and varying concentrations of sheep erythropoietin (EP) (step III, Connaught Medical Research Laboratories, Toronto, Canada) . All experiments were done in duplicate. The cultures were incubated for 3 days at 37 0 C in a humidified atmosphere of 10% CO2 in air. To each plate were then added a few Received October 9, 1974; accepted March 7, 1975. Supported in part by a grant from The Netherlands Organization for Fundamental Medical Research, The Hague. 3 Radiobiological Institute TNO, Lange Kleiweg 151, Rijswijk (Z-H), The Netherlands. 4 On leave of absence from the Instituto Scientifico di Medicina Interna, Universita Cattedra di Ematologia, Genoa, Italy; recipient of a fellowship from The Netherlands Organization for the Advancement of Science, The Hague, under the auspices of a treaty between this organization and the Centro Nazionale di Richerche, Italy. 5 We thank Drs. K. A. Dicke and P. Bentvelzen for their interest in this investigation, Mr. J. Brinkhof for purification of the virus. and Dr. A. C. Ford for aid in the preparation of the manuscript. 1
2
JOURNAL OF THE NATIONAL CANCER INSTITUTE, VOL. 55, NO. I, JULY 1975
59
60
NOOTER AND GHIO
drops of 0.2% dimethoxybenzidine (Baker, Deventer, The Netherlands) in 0.5 M acetic acid with 0.5% hydrogen peroxide. Only colonies staining blue due to the presence of hemoglobin and containing more than eight cells as revealed by an inverted microscope were scored. Mixed colonies of erythroid and myeloid cells were not included. We stained individual colonies by transferring them with a pipette and putting them on a glass slide. Slides were air dried, fixed in methanol, and washed in water to remove the methyl cellulose. Staining for hemoglobin was done by the Lepehne procedure (17) and counterstained for 8 minutes with Giemsa (1 :20). In some experiments we prepared feeder layers by plating 2 X 105 irradiated (2,000 rad) bone marrow cells suspended in an agar medium consisting of 1% agar in DMEM with 30% FBS. The bone marrow cells in the feeder layers were obtained from either normal or erythroblastotic mice and were overlaid with 1 ml of the 0.8% methyl cellulose mixture containing 2 X lOs bone matrow cells from either normal or erythroblastotic mice. RESULTS
The production of in vitro erythroid colonies at several concentrations of EP by normal, bone marrow cells and cells taken from mice several days after infection with RLV is presented in text-figure 1. The controls cor-
I
1,2
Nooter,s and Riccardo Ghio 4,6
SUMMARY-Bone marrow cells from BALB/c mice Infected with Rauscher erythroblastosis virus produced five to twenty· five times more erythroid colonies in vitro in the absence of erythropoietin (EP) as compared to normal cells. A good eerrelatlon existed between the state of the disease and the number of hormone-independent erythroid colony-forming cells (CFU· E). A significant number of hormone·independent CFU·E was found as early as 3 days after Infection. A linear relationship existed between the number of cells plated and the number of erythroid colonies formed in vitro. Addition of EP did not enhance colony formation, even at low cell concentrations. Feeder layer experiments demonstrated that EP-independent colony formation was not due to the production of endoge. nous EP. Repeated injections of phenylhydrazine Into normal mice did not lead to the loss of EP responsiveness in vitro; this indicated that the hormone independency Induced by the virus was not due to continuous erythropoietic stimulation in vivo. Besides hormone Independency, the CFU-E from Infected mice required less serum in the culture medium. Normal ery· throld colonies regressed after 4 days of culture, but EP·inde· pendent colonies from infected mice persisted for more than 2 weeks. These three phenomena may be regarded as indicative for a physiologic transformation.-J Natl Cancer Inst 55: 59-64, 1975.
COLONY FORMATION by bone marrow cells from normal BALBJc mice may differ from that in Rauscher virus-infected mice. Murine leukemia viruses replicate well in fibroblasts, but in contrast to sarcoma viruses (1, 2) transformation by this group of viruses is rare (3-6). Several members of the group induce various kinds of hematopoietic neoplasms in vivo within a relatively short period. It is obvious that, for the assessment of the transforming capacity of leukemia viruses in vitro, hematopoietic cells and not fibroblasts are the appropriate target cells. To develop a suitable transformation assay, possible discrimination between normal and leukemic cells on the basis of differences in growth patterns in vitro is required. Techniques have been developed to produce myeloid (7-9) and erythroid colonies (10, 11) in vitro from mouse bone marrow cells. Our study represents a search for differences in colony formation between bone marrow cells from normal BALB Jc mice and BALB Jc mice infected with Rauscher erythroblastosis virus (12-14). MATERIALS AND METHODS
Virus isolation.-Rauscher leukemia virus (RLV) was isolated from plasma of leukemic BALB Jc mice. Plasma (100 ml) was centrifuged for 20 minutes at 1,200xg. The supernatant was spun for 20 minutes at 10,000 rpm (± 10,000Xg). An equal volume of buffer (1.5 mM Tris-HCI+0.5 M sucrose, pH 7.2) was added to the supernatant and centrifuged for 30 minutes at 228,000Xg (Beckman 50 Ti rotor; 50,000 rpm). The virus pellet was resuspended in I ml buffer (1.5 mM TrisHCl+0.025 M sucrose, pH 7.2) and spun for 25 minutes at 23,000 rpm (Beckman SW 27, I rotor; 85,000xg) on
a linear gradient (5-20% sucrose in 1.5 mM Tris-HCl, pH 7.2). The obtained opalescent band contained purified RLV. The virus was stored in liquid nitrogen until used. The virus preparation was assayed with the reverse XC plaque test of Niwa et al. (15) and contained 3 X 107 plaque-forming unitsJml. The purified virus stock was not contaminated with lactic dehydrogenase virus. It contained 8 mg proteinJml as measured by the method of Lowry et al. (16). Mice.-Six-week-old female BALBJc mice were given ip injections of approximately 0.01 mg purified virus and thereafter examined twice a week for malignancy. A continuing increase in spleen weight began at day 5 post infection; a maximum weight of 3 g was reached in the terminal stage (4-5 wk post inoculation). Six- to eight-week-old female BALBJc mice became anemic as a result of sc injections of 0.5 mg phenylhydrazine for 3 consecutive days. We determined the hematocrit 2 days later by puncturing the orbital venous plexus. Hypertransfusion experiments were done with normal and RLV-infected BALBJc mice. Blood for transfusions was collected aseptically in heparinized phosphate-buffered saline and washed three times. Each recipient was inoculated ip with 0.5 ml packed red cells on 3 successive days. The hematocrit was determined before and 3 days after transfusion when the animals were killed for bone marrow collection. Bone marrow cwltures.-The technique was based on that of Stephenson et al. (10). Bone marrow cells were aspirated from the femurs. The bone marrow cells were washed in Hanks' balanced salt solution by centrifugation at 250 X g at 4 0 C. Thereafter, 2 X 105 nucleated cells were plated in 35-mm plastic petri dishes (Greiner, Niirtingen, Germany) along with I ml of a mixture of Dulbecco's modified Eagle's minimum essential medium (DMEM) (Flow Laboratories, Irvine, Scotland), 0.8% methyl cellulose (Methocel, 4,000 cps; Dow Chemical Co., Bogota, Colombia), 30% fetal bovine serum (FBS) (Flow Laboratories), and varying concentrations of sheep erythropoietin (EP) (step III, Connaught Medical Research Laboratories, Toronto, Canada) . All experiments were done in duplicate. The cultures were incubated for 3 days at 37 0 C in a humidified atmosphere of 10% CO2 in air. To each plate were then added a few Received October 9, 1974; accepted March 7, 1975. Supported in part by a grant from The Netherlands Organization for Fundamental Medical Research, The Hague. 3 Radiobiological Institute TNO, Lange Kleiweg 151, Rijswijk (Z-H), The Netherlands. 4 On leave of absence from the Instituto Scientifico di Medicina Interna, Universita Cattedra di Ematologia, Genoa, Italy; recipient of a fellowship from The Netherlands Organization for the Advancement of Science, The Hague, under the auspices of a treaty between this organization and the Centro Nazionale di Richerche, Italy. 5 We thank Drs. K. A. Dicke and P. Bentvelzen for their interest in this investigation, Mr. J. Brinkhof for purification of the virus. and Dr. A. C. Ford for aid in the preparation of the manuscript. 1
2
JOURNAL OF THE NATIONAL CANCER INSTITUTE, VOL. 55, NO. I, JULY 1975
59
60
NOOTER AND GHIO
drops of 0.2% dimethoxybenzidine (Baker, Deventer, The Netherlands) in 0.5 M acetic acid with 0.5% hydrogen peroxide. Only colonies staining blue due to the presence of hemoglobin and containing more than eight cells as revealed by an inverted microscope were scored. Mixed colonies of erythroid and myeloid cells were not included. We stained individual colonies by transferring them with a pipette and putting them on a glass slide. Slides were air dried, fixed in methanol, and washed in water to remove the methyl cellulose. Staining for hemoglobin was done by the Lepehne procedure (17) and counterstained for 8 minutes with Giemsa (1 :20). In some experiments we prepared feeder layers by plating 2 X 105 irradiated (2,000 rad) bone marrow cells suspended in an agar medium consisting of 1% agar in DMEM with 30% FBS. The bone marrow cells in the feeder layers were obtained from either normal or erythroblastotic mice and were overlaid with 1 ml of the 0.8% methyl cellulose mixture containing 2 X lOs bone matrow cells from either normal or erythroblastotic mice. RESULTS
The production of in vitro erythroid colonies at several concentrations of EP by normal, bone marrow cells and cells taken from mice several days after infection with RLV is presented in text-figure 1. The controls cor-
I
