Comparative Heat Sensitivity of Murine and Human Hemopoietic Progenitors and Clonogenic Leukemia Cells

Transcription

1 Comparative Heat Sensitivity of Murine and Human Hemopoietic Progenitors and Clonogenic Leukemia Cells Julia Giddi: Imre Fehe'qa Pe'ter Kova'csb "National Institute of Haematology, Blood Transfusion and Immunology, Budapest, Hungary; binstitute of Pharmacology, Medical School, Debrecen, Hungary Key Words. Thermal sensitivity Human CFU-GM Murine CFU-GM Clonogenic leukemic cells HL-60 WEHI 3-B Abstract. The aim of this study was to compare the thermal sensitivity of normal murine and human hemopoietic progenitors to that of leukemic murine and human clonogenic cells in order to assess the clinical relevance of experimental data. Colony forming units-granulocyte-macrophage (CFU-GM) from normal human bone marrow and from bone marrow of patients with acute lymphocytic leukemia (ALL), acute myelogenous leukemia (AML), and Hodgkin's disease in complete remission proved to be less sensitive to 42.5"C in vitro hyperthermia (Do: 93.9 min) than murine bone marrow CFU-GM (DO: 49.6 min). Leukemic colony forming cells (CFU-L) from HL-60 suspension culture-when compared to human CFU-GM-showed significantly increased thermal sensitivity (Do: 22.8 min). While the thermal sensitivity of CFU-L from a murine leukemia cell line (WEHI 3-B) was not statistically significant when compared to that of CFU- L from HL-60 (Do values 17.0 versus 22.8 min), the vertical difference between the parallel regression lines suggested an approximately three-fold greater survival for human CFU-L. Although carefully controlled hyperthermia is an easy purging technique, the relevance of murine data to human clinical practice must be considered critically. Introduction Autologous bone marrow transplantation (ABMT) is a promising therapy for patients for Correspondence: Dr. J. Gidfili, National Institute of Haematology, Blood Transfusion and Immunology, Budapest, P.O. Box 44, H-1502, Hungary. Received December 20, 1993; provisionally accepted February 16, 1994; accepted for publication April 18, OAlphaMed Press /94/$5.00/0 whom no matched donor is available for allogeneic transplantation. Although the role of contaminating leukemia cells infused with the graft in the high relapse rate following ABMT is controversial, many attempts have been made to eliminate malignant cells from the transplanted bone marrow or peripheral mononuclear cells [l-41. An ideal purging method would kill a substantial fraction of tumor cells while sparing enough hemopoietic stem cells for successful engraftment. One candidate technique is hyperthermia. Using various end points, malignant hemopoietic cells have been proven to be more sensitive to hyperthermia than their normal counterparts [ Long-term engraftment could be achieved by transplanting those hemopoietic stem cells that survive heat exposure [7]. This result appears to indicate that primitive stem cells are heat resistant [12]. The fact that neither engraftment nor leukemia incidence can be satisfactorily assayed in humans prompted the development of murine experimental systems. Even the protocols applied to treatment of human tumors are based on rodent systems, assuming that the thermal sensitivity of human and rodent tumor cells does not differ significantly. When the thermal response of human and rodent tumor cells was compared, however, human cells proved to be more heat resistant and acquired thermotolerance readily, while rodent cells were unable to do so [13]. While a recent case report supported the applicability of hyperthermia for bone marrow purging [14], a careful comparison must be performed prior to extrapolating murine experimental data to human therapy. In the present experiments, therefore, the thermal sensitivity of STEM CELLS 1994;12:

2 GidBlilFehCrlKovks 534 normal murine and human hemopoietic progenitors and that of murine and human leukemic cell lines were compared. Materials and Methods Bone Marrow Samples Murine bone marrow cells were obtained from the femora of week old BDFl mice. After informed consent, human bone marrow samples were obtained from the iliac crest for ABMT (acute myelogenous leukemia [AML] and acute lymphocytic leukemia [ALL] patients) or for allogeneic transplantation (normal bone marrow). Assay of Clonogerzic Cells Murine granulocyte-macrophage colony forming cells (CFU-GM) were assayed by a modified method based on that of Bradley and Metcalf[l5]. Briefly, the cells were plated in McCoy's 5A medium supplemented with 20% horse serum (Flow Laboratories, UK) in 0.3% agar in the presence of 10% WEHI 3-B conditioned medium as colony stimulating factor (CSF). Leukemic colony forming cells (CFU-L) from the WEHI 3-B cell line were assayed in a similar system without added CSF. After a seven-day incubation, aggregates of more than 50 cells were scored as colonies. Human CFU-GM were assayed by a modified technique based on that of Robinson and Pike [ 161. Briefly, unseparated cells were plated in McCoy's 5A medium supplemented with 30% fetal calf serum ([FCS], Sebak, Germany) in 0.3% agar in the presence of 10% supernatant of 5,637 human bladder carcinoma cell line cultures as CSF source. On day 9, aggregates of more than 40 cells were scored as colonies. HL-60 CFU-L were assayed in a similar system without added CSF. Incubation time was 14 days. Leukemic Cells WEHI 3-B cells were obtained from Dr. E. Spooncer (Paterson Institute of Cancer Research, Manchester, U.K.) through the generosity of the originator of the cell line (Dr. D. Metcalj Walter & Eliza Hall Institute, Melbourne, Australia). Cells were maintained in suspension culture in McCoy's 5A medium supplemented with 15% FCS by serial passage every third day, with occasional intervals of storage in liquid nitrogen. Hyperthermia Cells were heated by immersion in a precision thermostatically-controlled water bath. Cell suspensions were prepared in McCoy's 5A medium supplemented with 10% FCS and were heated in a cell concentration of 5-8 x lo6 per ml in a final volume of 2 ml in sterile, airtight polystyrene centrifuge tubes. Before hyperthermia, the cell suspensions were kept at 37 C for 15 min then transferred to the precision temperature-controlled water bath. After exposure to 42.5"C, appropriate dilution of cells was prepared with cold McCoy's medium and plated immediately. Control samples were kept at 37 C for 120 min. Statistical Analysis Linear regression of the natural logarithm of the surviving fraction was computed and a statistical comparison of the regression curves and Do values was carried out [17]. Results First, survival curves of normal human and murine CFU-GM at 42.5"C were compared. Both curves were characterized by a shoulder region followed by an exponential decline. A statistically significant difference in the slope of the curves and Do values (93.89 * for human versus * 8.2 for murine CFU-GM) suggested that human CFU-GM are less sensitive to 42.5"C hyperthermia (Fig. 1). Since cells for ABMT are usually collected after repeated courses of cytotoxic chemotherapy (i.e., an increased cycling rate can be postulated) before comparing thermal sensitivity of CFU-GM and CFU-L, the survival curves of normal human CFU-GM, and those from bone marrow collected from chronic myelogenous leukemia (CML) patients or from patients for ABMT (three ALL, three AML and one Hodgkin's lymphoma in complete remission) were compared. The Do value of the CFU-GM curves did not differ significantly (Fig. 2). Consequently, in the following experiments the sensitivity of CFU-L could be compared to that of the normal CFU-GM. CFU-L from HL-60 suspension cultures showed significantly increased sensitivity to hyperthermia (Do value: 22.8 * 3.2 min). The Do value differed significantly from the normal (Fig. 3).

3 535 survival (K) suwlval (%) 0.1 Heat Sensitivity of CFU-GM and CFU-L Fig. 1. Heat sensitivity of normal human and murine CFU-GM as a function of time at 42.5'C. Controls: cells incubated at 37'C for 120 min. Curves are calculated from five (human) to 10 (murine) experiments. Murine CFU-GM: In y = x + In 171.4; r = Human CFU-GM: In y = x + In ; r = Fig. 3. Heat sensitivity of normal human CFU-GM and HL-60 leukemic clonogenic cells (CFU-L) as a function of time at 42.5'C. Curves are calculated from six experiments. Normal CFU-GM: see legend to Figure 2. CFU-L: In y = x + In ; r = When the thermal sensitivity of the CFU-L of murine (WEHI 3-B) and human (HL-60) cell lines was compared, Do values did not show a significant difference (17.0 ~t 2.2 versus 22.8 i3.2, p > 0.1). If, however, due to the non-significant survival (K) difference in the slopes, the curves were hypothetically regarded as parallel, the vertical distance between the parallel regression lines was 3.02, suggesting that the same exposure time allowed 3.02 times more survival of HL-60 (human) CFU-L than WEHI 3-B (murine) CFU-L (Fig. 4). survival (K) 11 I I40 Fig. 2. Heat sensitivity of CFU-GM from bone marrow of normal donors, or from bone marrow collected for ABMT as a function of time at 42.5'C. Controls: see legend to Figure 1. Curves are calculated from five to six experiments. Normal bone marrow: In y = x + In ; r = ABMT bone marrow: In y = x + In ; r = I Fig. 4. Heat sensitivity of human (HL-60) and murine (WEHI 3-B) CFU-L as a function of time at 42.5'C. Curves are calculated from six experiments. HL-60 CFU-L: see legend to Figure 3. WHI 3-B CFU-L In y = -0,05867 x + In ; r = a

4 GidBlilFehbr/Kovhcs 536 Table I. Comparison of Do value and extrapolation number for murine and human bone marrow CFU-GM, and for CFU-L from murine (WEHI 3-B) and human (HL-60) leukemia cell lines Type of cells Do value (min) P' Pb n (mean f SE) Murine CFU-GM f WEHI 3-B CFU-L * 2.20 < Human CFU-GM * < HL-60 CFU-L * 3.20 < > Human CFU-GM (ABMT)' f > Do value and n (extrapolation number) were calculated from the curves. p: referred to the normal counterpart (murine or human CFU-GM, respectively). bp: referred to murine counterparts: CFU-GM or CFU-L, respectively. 'ABMT: bone marrow collected for ABMT from AML or ALL patients in complete remission. For better comparison, Do values, extrapolation numbers calculated from Figures 1-4 and p values are summarized in Table I. Discussion Our present knowledge on the differential sensitivity to hyperthermia of malignant versus normal hemopoietic cells has been based either on data obtained on human or rodent clonogenic cells in vitro [8, 10-12, 18, 191 or from murine experiments in vivo [5-9, 121. Since murine experiments offer the advantage of assaying repopulating ability [7, 121 parallel to leukemogenic effectiveness of purged bone marrow [7], animal models might play a special role in planning optimal purging systems to be applied to human ABMT. Before considering any extrapolation from murine experiments to clinical purging, however, possible species differences have to be ruled out. Some data have already suggested different heat sensitivities of human fibroblasts and Chinese hamster cells, as well as human and rodent tumor cells [ 131. Heat sensitivity of human and murine normal clonogenic hemopoietic progenitors or that of human and murine clonogenic leukemic cells has not yet been compared. The present experiments suggested that normal human CFU-GM are more resistant to 42.5"C in vitro hyperthermia than their murine counterparts. Similar to other data [ 191, the sensitivity of human bone marrow CFU-GM was not affected by the cycling rate: similar Do values were obtained for CFU-GM from normal bone marrow or from AML or ALL patients in complete remission after repeated courses of cytotoxic chemotherapy. The same phenomenon was observed for day-9 CFU-S from steady state or regenerating bone marrow [7], suggesting that different cycling properties cannot be responsible for the phenomenon observed. Division independent hyperthermic death has already been proven in Chinese hamster ovary cells [20]. Although high survival levels were found in our experiments for normal progenitors (25-35% survival for human, 15% for murine CFU-GM at 120 min exposure to 42.5"C), division associated death predominated. One cannot categorically exclude the possibility of division independent hyperthermic death. The effect of hyperthermia, however, is not simply related to energy deposition. The relationship between required time and temperature to produce a certain biological effect suggested that there is an inflection point around 42.5"C [21]. 42.5"C is a relatively mild hyperthermia and 120 min is a rather long exposure time. Our approach was a pragmatic one: this dose and time produced a more than two-log difference between surviving fractions of clonogenic malignant murine cells and CFU-S or repopulating ability in our earlier studies [7, 221. Based on these experiments at this purging temperature, human clonogenic HL- 60 cells showed significantly increased sensitivity to hyperthermia than human CFU-GM, suggesting that this is a promising purging temperature.

5 537 Heat Sensitivity of CFU-GM and CFU-L However, when the surviving fractions of HL-60 (human) and WEHI 3-B (murine) CFU-L were compared, the difference in their heat sensitivity was similar to that observed for normal human and murine clonogenic CFU-GM (Table I). If all conditions (the rate at which hyperthermic temperature is achieved, a strictly controlled temperature and serum concentration during hyperthermia, and a quick return to normal temperature) are carefully controlled, hyperthermia is technically an easy purging method, and isoeffect curves can be easily precalculated. Extrapolation from murine experiments to clinical purging, however, needs careful consideration. Acknowledgments The authors wish to express their thanks to Dr. E. Spooncer for the WEHI 3-B cells, and Ms. E. Kovdcs for skillful technical assistance. This work was partly supported by the National Research Foundation of the Hungarian Academy of Sciences (No. 194) and the Scientific Research Council of the Ministry of Welfare, Hungary (T-53/1990 and T-476/1990). References 1 Gale RP. Bone marrow purging: current status, future directions. Bone Marrow Transplant 1987;2(~~ppl2): Dicke KA, Spinolo JA. High dose chemotherapy and autologous bone marrow transplant (ABMT) in acute leukemia: is purging necessary? Bone Marrow Transplant 1989;4(suppl 1): Gale RP, Butturini A, Reizenstein P. Autotransplants in leukemia: current state, future progress. Leuk Res 1991;15: Gorin NC, Aegerter P, Auvert B, Meloni G, Goldstone AH, Burnett A, Carella A, Korbling M, Herve P, Maraninchi D, Lowenberg R, Verdonck LF, de Planque M, Hermans J, Helbig W, Porcellini A, Rizzoli V, Alesandrino EP, Franklin IM, Reiffers J, Colleselli P, Goldman JM. Autologous bone marrow transplantation for acute myeloid leukemia in first remission: a European survey of the role of marrow purging. Blood 1990;75: Robins HI, Steeves RA, Clark AW, Martin PA, Miller K, Dennis WH. Differential sensitivity of AKR murine leukemia and normal bone marrow cells to hyperthermia. Cancer Res 1983;43: Flentje M, Flentje D, Sapareto SA. Differential effect of hyperthermia on murine bone marrow colony forming units and AKR and L1210 leukemia stem cells. Cancer Res 1984;44: Gidhli J, Szamosvolgyi S, FehtSr I, Kovhcs P. Survival and characteristics of murine leukaemic and normal stem cells after hyperthermia: a murine model for human bone marrow purging. Leuk Res 1990;14: Wang SB, Hendry JH, Testa NG. Thermal sensitivity of haemopoietic and stromal progenitor cells in different proliferative states. Br J Cancer 1985;52: Wierenga PK, Konings AWT. Effect of a hyperthermic treatment on the pluripotent haemopoietic stem cell in normal and anaemic mice. Int J Hyperthermia 1990;3: Moriyama Y, Narita M, Sat0 K, Urushiyama M, Koyama S, Hirosawa H, Kishi K, Takahashi M, Takai K, Shibata A. Application of hyperthermia to the treatment of acute leukemia: purging human leukemic progenitors by heat. Blood 1986;67: Blackbum MJ, Wheldon TE, Field SB, Goldman JM. The sensitivity of hyperthermia of human granulocyte/macrophage progenitor cells (CFU- GM) derived from blood or bone marrow of normal subjects and patients with chronic granulocytic leukaemia. Br J Cancer 1984;50: Wierenga PK, Konings AWT. Studies on the hyperthermic sensitivity of the murine hematopoietic stem cell compartment. I. Heat effects on clonogenic stem cells and progenitors. Exp Hematol 1993;21 : Hahn GM, Ning SC, Elizaga M, Kapp DS, Anderson RL. A comparison of thermal responses of human and rodent cells. Int J Radiat Biol 1989;56: Herrmann RP, O Reilly J, Meyer BF, Lazzaro G. Prompt haemopoietic reconstitution following hyperthermia purged autologous marrow and peripheral blood stem cell transplantation in acute myeloid leukaemia. Bone Marrow Transplant 1992;10: Bradley TR, Metcalf D. The growth of mouse bone marrow cells in vitro. Austr J Exp Biol Med Sci 1966;44: Pike BL, Robinson WA. Human bone marrow colony growth in agar-gel. J Cell Physiol 1970;76: Diem K, Lentner G, eds. Scientific Tables 7th Edition. Basel, Switzerland: Ciba-Geigy Ltd, 1970: Moriyama K, Nikkuni H, Saito H, Aoki A, Furukawa T, Imanari A, Narita M, Kishi K, Takahashi M, Shibata A. In vitro sensitivity of

6 GidAli/Fehtr/KovBcs Moriyama Y, Nikkuni K, Saito H, Aoki A, Kishi K, Takahashi M, Shibata A. Effect of hyperthermia on both primary proliferation and self renewal of human leukemic progenitor cells in vitro: Its application to in vitro purging. Leukemia 1991;5: human hematopoietic progenitor cells to hyperthermia: critical temperature for cells to survive and its application to in vitro purging. Bone Marrow Transplant 1990;6: Vidair CA, Dewey WC. Division associated and division independent hyperthermic cell death: comparison with other cytotoxic agents. Int J Hyperthermia 1991;7: Dewey WC, Hopwood LE, Sapareto SA, Derweck LE. Cellular responses to combinations of hyperthermia and radiation. Radiology 1976; 123 : GidAli J, Fehtr I. The effect of combined purging with mafosfamide and hyperthermia on murine haemopoietic stem cells and leukemogenic cells. Bone Marrow Transplant 1992;10: