Effects of Mast Cell Growth Factor on Ara-C Mediated Acute Myeloid Leukemia Cell Killing A. Tafuri, L. De Felice, M. G. Mascolo, T. Valentini, M. T. Petrucci, M. C. Petti University La Sapienza of Rome, Hematology, Department of Biopathology, Rome, taly Key Words. AML Ara-C MGF Growth factors Abstract. Cell kinetic studies of acute myeloid leukemia (AML) have provided evidence for the presence of nonproliferating cells. Hemopoietic growth factors (GF) can regulate proliferation of leukemic cells, furnishing new possibilities for recruiting quiescent cells into the cycle and overcoming cytokinetic resistance in AML. To assess the role of the novel identified cytokine, mast cell growth factor (MGF), in enhancing cytosine arabinoside (Ara-C) cytotoxicity, we have primed AML blasts with MGF and then exposed these cells to the S phase specific agent Ara-C. Other growth factors such as PXY, interleukin 3 (L-3), granulocyte-macrophage colony stimulating factor (GM-CSF) and granulocyte CSF (G-CSF) and the combination of MGF plus PXY were also tested. Cytokinetic changes and clonogenic growth of leukemic colony forming unit (CFU-L) cells in methylcellulose were used to detect proliferative and cytotoxic effects on AML blasts. Expression of MGF receptor, the c-kit protein, was also measured by flow cytometry. We report in this preliminary study that MGF is able to increase proliferation in 75% of the samples studied and enhance Ara-C cytotoxicity in some of these cases. When MGF proliferative activity was compared with other GFs, individual cases showed heterogeneity in response, although the combination of MGF plus PXY was always the most effective. ntroduction Cell kinetic analysis of acute myeloid leukemia (AML) has identified a substantial Correspondence: Dr. Agostino Tafuri, Laboratory of Cell Kinetics, Department of Biopathology, University La Sapienza of Rome, Via Benevento 6, 00161 Rome, taly. Received March 9, 1993; accepted for publication March 9, 1993. OAlphaMed Press 1066-5099/93/$5.00/0 number of nonproliferating or very slowly proliferating blast cells [ 13. Their presence may play an important role in kinetic resistance to chemotherapy, which is usually aimed against cycling cells. Laboratory studies have suggested that hemopoietic growth factors (GFs), when combined with cytosine arabinoside (Ara-C), can enhance cytotoxic effects of this agent against AML cells [2]. Granulocyte-macrophage colony stimulating factor (GM-CSF) and interleukin 3 (L-3) can increase the percentage of cells in S phase recruiting Go cells and enhance Ara-C cytotoxic effects against AML blasts [3-51. We demonstrated that a combination of GM-CSF plus a -3 had a greater in vitro effect in decreasing quiescent cells versus treatment with the individual GF alone [5]. The GM-CSF/L-3 fusion protein has been recently referred to as the most effective in induction of proliferation of AML blasts and normal progenitors [6]. Mast cell growth factor (MGF) is a recently identified cytokine which acts on primitive hemopoietic precursors [7] through a specific receptor, the protein encoded by the proto-oncogene c-kit [8]. Expression of this receptor in AML blast cells [9] and preliminary reports of mitogenic activity of MGF on myeloid leukemia cells [ 101 indicate that MGF may play a role in the growth and proliferation of myeloid leukemic cells. Therefore, we investigated in this study the cell kinetic changes and the clonogenic blast cell proliferation induced by MGF in myeloid leukemia in order to detect effects on Ara-C cytotoxicity using this combined MGF/Ara-C strategy. Comparison of proliferative activity between MGF and other growth factors was also studied along with pattern of c-kit receptor expression. STEM CELLS 1993;11(~Uppl2):88-92
Tafuri et al. 89 Materials and Methods Samples Peripheral blood and bone marrow samples were obtained with informed consent, according to institutional policy, from four AML patients. Patient characteristics are listed in Table. Mononuclear cells were separated by Ficoll- Hypaque density gradient. Biological Reagents Recombinant human MGF and PXY 32 1 (GM-CSFLL-3 fusion protein) were kindly provided by mmunex, Seattle, WA, and were used respectively at a concentration of 50 ng/ml and 10 ng/ml. L-3 (Sandoz) was used at 20 ng/ml, GM-CSF and granulocyte CSF (G-CSF) at 500 U/ml (respectively from Schering and Roche, Basel, Switzerland). Suspension Culture Blast cells were resuspended in RPM supplemented by 10% fetal calf serum (FCS), 1% L-glutamine and 1 % penicillin-streptomycin (all from GBCO, Grand sland, NY). The number was adjusted to a starting concentration of 1.O x lo6 cells/ml. Cells were cultured for 48 h at 37 C in 5% C02 with and without (control culture) the following growth factors: MGF, PXY, G-CSF, GM-CSF, L-3, and MGF + PXY. Afterwards, cell number/viability was assessed by trypan blue exclusion. At this time kinetic studies and clonogenic assay were performed (see below). Cytotoxic Drug Effects After 48 h of incubation with cytokine, Ara-C (10 pm, 1 pm, 0.01 pm) was added to suspension cultures for 24 h. Cytotoxicity was determined by clonogenic assay in methylcellulose, and the results were expressed as the percentage of clonogenic cells killed at different Ara-C concentrations. Table. AML patient characteristics FAB Sample WBC/mrn3 Blasts(%) #1 M4E PB 19,000 83 #2 M5a BM 1,050 74 #3 M3 PB 24,400 80 #4 MO PB 23,800 80 Clonogenic Assay To measure cytokine-induced proliferative changes in clonogenic cells and changes in Ara-C cytotoxic effects, cultures were performed using the leukemic colony forming unit (CFU-L) assay, previously described [ 111. Cells were seeded at a density of lo5 in 1 ml MDM, 30% FCS, 0.8% methylcellulose and 10% 5637-culture medium (CM). Plates were scored for clusters (>8 cells) and colonies (>20 cells) after 7 days of incubation. Results are expressed as total number of clusters and colonies. Kinetic Studies To evaluate cell cycle distribution, flow cytometric acridine-orange (A.O.) technique was performed, as previously described [12]. A.O. was used to measure cellular DNA and RNA content (percentage of cells in Go, G, S, G2M, mean RNA content). The RNA content of G cells was expressed as RNA-index (RNA- G) determined as the ratio of mean RNA content of G cells of the samples times ten, divided by the median RNA content of control lymphocytes. Receptor Studies ndirect immunofluorescence for staining with the anti-kit antibody, clone 17F11 (mmunotech, S.A., Marseille, France) was performed as previously described [ 101 and then measured by flow cytometry. Flow Cytometric Measurement and Analysis A modified fluorescence activated cell sort (FACScan; Becton Dickinson, San Jose, CA) was used to measure fluorescence upon excitation at 488 nm. For each analysis 5-10 x lo3 cells were measured at separate wavelengths for green (F530-DNA [A.O.] FTC) and red (F > 620-RNA [A.O.]). Samples were analyzed using Becton Dickinson software, including CellFit and Lysis 11. Results AML blast proliferation induced by MGF was studied measuring cell cycle kinetic changes and clonogenic cell growth. Cell Kinetic Effects and c-kit Expression Table 1 shows the cell cycle distribution of the four samples studied after 48 h of liquid
~ 90 Effects of MGF on Ara- C Cytotoxicity in AML Table 11. Cell cycle effects # GO% G% S% RNA- G c-kit% 1 -Control 26.9 MGF 8.8 2-Control 53.0 MGF 27.2 3-Control 35.1 MGF 37.0 4-Control 93.0 MGF 73.8 55.2 70.0 42.8 51.6 56.4 57.3 6.6 19.4 15.9 17.8 3.9 18.0 8.1 5.4 1.o 5.2 17.9 n.d. 23.9 14.4 n.d. 19.9 14.5 n.d. 15.2 18.0 33 21.4 culture with and without cytokines. MGF induces recruitment into the cycle in 3/4 samples (#l, #2, #4) (see Fig. 1) with significant decrease (>25%) of Go cells and increase in S phase, only in sample #1 the changes in S phase were minor although proliferation was already high in the control culture. A different degree of response was seen in each case: Go cells decreased from 1.27- to 3.05-fold, and S phase cells exhibited an increase from 1.1 1- to 5-fold. Sample #3 failed to show any changes in the MOF Rna cell cycle distribution. Receptor studies for MGF demonstrated c-kit protein expression only in 113 of responder cases (#4). When proliferative activity induced by MGF was compared with other growth factors, the greatest decrease in Go cells and increase in S phase was induced by the combination of MGF plus PXY 321, and the last one was the more active among the single cytokines studied. Effect of MGF on Clonogenic Cell Growth Three out of four samples studied showed clonogenic growth (#l, #2, #4) (see Fig. 2). A significant (>30%) proliferative response to MGF priming was detected in 213 cases (#2, #4) as shown in Table 111. When the effect of MGF was compared with other cytokines, heterogeneity in response was seen in individual cases. Low proliferative effects were induced by both MGF and other GFs in sample #1 (MGF: 1.3-fold, MGF + PXY: 1.4-fold). Higher growth response to GFs was seen for case #2 (MGF: DDWJ MQF DXY Fig. 1. Cell kinetic effects of MGF, L-3, PXY and MGF + PXY in aneuploid AML sample as measured by A.O. flow cytometric technique. Significant recruitment of leukemic quiescent cells is apparent for all conditions, especially for MGF and MGF + PXY. Fig. 2. Clonogenic cell growth of three AML samples (#1. #2, #4). Comparison between control cultures (black bar) and MGF exposed cultures.
Tafuri et al. 91 Table 111. Effect of MGF on CFU-L growth #1 #2 #4 Control 1,901 5,854 85 MGF 2,391 26,888 226 clusters + colonies/los plated cells 4.6-fold, MGF + PXY 4.5-fold). n the third case (#4), G-CSF was the most effective single growth factor (4.7-fold), and MGF combined with PXY showed a synergistic proliferative effect (1 1.4-fold). Effects of MGF on Ara-C Cytotoxicity A correlation between MGF proliferative effects and Ara-C sensitization of leukemic progenitors was found in all cases. As shown in Table V, in samples with increased clonogenic growth after exposure to MGF (#2 and #4), a significant enhancement of CFU-L killing was detected at 1 pm Ara-C (58% versus 1% and 100% versus 29.6%). A similar correlation was also found at 0.01 pm Ara-C (23% and 56% versus 0). Sample #1 failed to show a significant increase in clonogenic cell growth, and the percentage of clonogenic cell killing appeared reduced after MGF exposure. We should notice that in this sample sensitivity to Ara-C in the control culture was already maximal (99.6% and 92.5% respectively at 10 ph4 and 1 N). Discussion n this study we investigated in vitro proliferative effects induced by MGF on AML blast Table V. MGF induced changes in Ara-C cytotoxicity Ara-C la.01 pm 1-Control 99.6 92.5 0 MGF 21 12 4 2-Control 98 1 0 MGF 98 58 23 4-Control 100 29.6 0 MGF 100 100 56 9% of killed CFU-L cells in order to determine changes in Ara-C cytotoxicity through a kinetic mechanism. Our preliminary data suggest that MGF is able to stimulate leukemia cell proliferation in the majority (75%) of the samples, although heterogeneity is apparent in individual cases. Recent reports [ 101 confirm the importance of MGF as a growth factor on myeloid leukemia cells, either as a direct stimulus or as a synergistic factor. Under the experimental conditions tested in this study, the combination of MGF and PXY induced the highest proliferation with synergistic effects detected by flow cytometric andfor clonogenic assay. The absence of detectable levels of flow cytometric c-kit protein in some of the responder samples showed a reduced sensitivity of this technique, confirming that more sensitive assays (i.e., Scatchard analysis) are required to detect SCF receptors in AML cases, as recently published [13]. Our data further suggests that MGF is effective in terms of Ara-C sensitization in some of the AML samples. A correlation was found between proliferative response to MGF and enhancement of Ara-C cytotoxic effects since in all cases with a significant increase in clonogenic cell growth, the highest percentage of CFU-L inhibition was detected. The only sample in which Ara-C sensitization after MGF exposure was not detected, was characterized by an already high value of S phase in the control culture; therefore, as reported for other cytokines [5], no further increase in proliferation could be detected. Several mechanisms have been reported to explain the effectiveness of the strategy of combined cytokine chemotherapy. A kinetic effect on quiescent cells has been described for other cytokines [5], and the present study suggests that the same mechanism also can be involved for MGF. Further studies are necessary to evaluate this preliminary in vitro observation on MGFlAra-C sensitization with careful evaluation of 1) beneficial net gain in cell killing (balance between expansion of leukemic cells and enhanced cell killing); and 2) biological features of leukemic cells at diagnosis that may be harmful if the leukemic population is increased by GFs (drug resistance). The role of MGF in stimulating residual normal progenitor proliferation in acute leukemia should also be established. MGF alone has been reported to be a poor stimulator of committed normal progenitors where a strong synergistic
92 Effects of MGF on Ara- C Cytotoxicity in AML activity in combination with other cytokines was found [ 141. t is therefore mandatory to evaluate the kinetic effects on normal progenitors when MGF is combined with other cytokines and chemotherapeutic drugs in acute leukemia. Acknowledgments The authors wish to thank Dr. Steven Gillis from mmunex, Seattle, WA, for kindly providing MGF and PXY 321. References 1 Clarkson BD, Fried J, Sakai Y, Ohkita T. Studies of cellular proliferation in human leukemia. 111. Behavior of leukemic cells in three adults with acute leukemia given continuous infusion of,h-thymidine for 8 or 10 days. Cancer 1970;25: 1327-1360. 2 Andreeff M, Welte K. Hemopoietic colony stimulating factors. Semin Oncol 1989;16:211-216. 3 Lista P, Porcu P, Avanzi GC, Pegoraro L. nterleukin-3 enhances the cytotoxic activity of 1 -beta-d arabinofuranosylcytosine (Ara-C) on acute myeloblastic leukemia (AML) cells. Br J Hematology 1988;69: 121-123. 4 Cannistra SA, Groshek P, Griffin JD. Granulocyte-macrophage colony stimulating factor enhances the cytotoxic effects of cytosine arabinoside in acute myeloblastic leukemia and in the myeloid blast crisis phase of chronic myeloid leukemia. Leukemia 1989;3:328-334. 5 Tafuri A, Andreeff M. Kinetic rationale for cytokine-induced recruitment of myeloblastic leukemia followed by cycle-specific chemotherapy in vitro. Leukemia 1990;12:826-834. 6 Williams DE, Park LS. Hemopoietic effects of a granulocyte macrophage colony stimulating factorlinterleukin-3 fusion protein. Cancer 1990;67:2705-2711. 7 Zsebo K, Wypych J, McNiece, Lu H, Smith K, Karkare S, Sachdev R, Yuschenkoff V, Birkett N, Williams L, Satyagal V, Tung W, Bosselman B, Mendaiaz E, Langley K. dentification, purification, and biological characterization of hematopoietic stem cell factor from Buffalo rat liver conditioned medium. Cell 1990;63:213-218. 8 Chabot B, Stephenson DA, Chapman VM, Besner P, Bernstein A. The proto-oncogene c-kit encoding a transmembrane tyrosine kinase receptor maps to the mouse Wlocus. Nature 1988;144:844-846. 9 Wang C, Koistinen P, Yang GS, Williams DE, Lyman SD, Minden MD, McCulloch EA. Mast cell growth factor, a ligand for the receptor encoded by c-kit, affects the growth in culture of the blast cells of acute myeloblastic leukemia. Leukemia 1991;5:493-499. 10 Pietsch T, Kyas U, Steffens U, Yakisan E, Hadam MR, Ludwig WD, Zsebo K, Welte K. Effects of human stem cell factor (c-kit ligand) on proliferation of myeloid leukemia cells: heterogeneity in response and synergy with other hemopoietic growth factors. Blood 1992;80:1199-1206. 11 Buick RN, Till JE, McCulloch EA. Colony assay for proliferative blast cells circulating in myeloblastic leukemia. Lancet 1977;i:862-863. 12 Andreeff M, Darzynkiewicz Z, Sharpless TK, Clarkson BD, Melamed MR. Discrimination of human leukemia subtypes by flow cytometric analysis of cellular DNA and RNA. Blood 1980;55:282-293. 13 Broudy VC, Smith FO, Lin N, Zsebo K, Egrie J, Bernstein D. Recombinant human stem cell factor (rhuscf) stimulates the growth of hemopoietic colonies from patient with acute nonlymphocytic leukemia (ANL) by binding to a specific receptor. Blood 1990;76:134a. 14McNiece K, Langley KE, Zsebo KM. Recombinant human stem cell factor synergizes with GM-CSF, G-CSF, L-3 and Epo to stimulate human progenitor cells of the myeloid and erythroid lineages. Exp Hematol 1991;19:226-230.