International Journal of Cell Cloning (1991)

Transcription

1 Concise Review International Journal of Cell Cloning (1991) Hybrid Cytokines as Hematopoietic Growth Factors Douglas E. Williamsa, Linda S. Park", Hal E. Broxmeyerb, Li Lub 'Immunex Research and Development Corporation, Seattle, Washington, USA; bthe Walther Oncology Center, Indiana University School of Medicine, Indianapolis, Indiana, USA Key Words. Granulocyte-macrophage colony-stimulating factor * Interleukin 3 * Fusion protein Cytokines - Hematopoiesis Myelosuppression Abstract. A large body of in vitm and in vivo data suggests that combinations of cytokines provide the most effective mechamm ' for stimulating multilineage acceleration ofhematopoiesis. Creation of a granulocyte-macrophage colony-stimulating factor (GM-CSF)/ interleukin 3 (IL,-3) fusion protein has yielded a single therapeutic which has enhanced biological activity in comparison to the individual cytokines from which it is composed. In vivo studies with this fusion protein (PIXY321) suggest that it may provide a means to accelerate both neutrophil and platelet recovery in clinical settings in which hematopoiesis is suppressed. The biology of PIXY321 and the potential for other fusion proteins is discussed. Regulation of Hematopoiesis The daily demands for the production of mature, functional hematopoietic cells is met by the orderly and tightly regulated difkrentiation of immature precursor cells in the hematopoietic tissues. These processes are thought to be governed by the balance between stimulatory and inhibitory cytokines [l, 21. In the last several years, a large number of cytokines have been identified which either directly enhance the proliferation and/or differentiation of hematopoietic progenitor cells or act in concert with such direct acting factors to amplify the response. The direct acting cytokines include the granulocyte-macrophage colony-stimulating factors (GM-CSFs), whereas those which have demonstrable activity only in the presence of other factors would include interleukin (IL) 1. It is apparent from both in vitro and in vivo studies that maximal stimulation of progenitor cell growth is achieved with multiple growth factors [l-31. IL-1 has been shown to synergize with GM-CSF and granulocyte (G)-CSF [4,5] and the ligand for the c-kit proto-oncogene has been shown to potentiate the action of vir- Correspondence: Douglas E. Williams, Ph.D., Experimental Hematology, Immunex Corporation, 51 University Street, Seattle, WA 98101, USA. Received August 8, 1991; accepted for publication August 8, /91/$2.00/0 0AlphaMed Press

2 Hybrid Cytokines as Hematopoietic Growth Factors 543 tually all of the cytokines which influence myelopoiesis or lymphopoiesis [ It is therefore likely that optimal clinical utilization of hematopoietic growth factors will involve combination therapy with a cocktail of two or more cytokines. The Rationale for IL-3/GM-CSF Fusion Proteins GM-CSF and IL-3 are two CSFs with a very similar spectrum of in vitro activities on hematopoietic progenitor cells [I, 21. IL-3 appears, however, to act upon a slightly earlier phenotype of cell, and its actions in the granulocyte differentiation pathway do not seem to extend to the terminal stages of neutrophil maturation, in contrast to GM-CSF which is an activator of mature neutrophils [l The in vivo responses seen clinically with these two factors are quite distinct with respect to the kinetics of the response and the multilineage effects, and are consistent with a more primitive target cell for IL-3 than GM-CSF [l5-17]. IL-3 has been shown to stimulate tri-lineage responses in some patients, particularly those with secondary bone marrow failure following intensive chemotherapy, and has an effect upon platelet levels in a large percentage of patients [16]. IL-3 responses have also been shown to be maintained for several days after the cessation of cytokine treatment [16]. GM-CSF has been shown to promote granulo-monopoiesis and to functionally activate mature cells along this developmental pathway [l2, 131. Thus the clinical motivation for the creation of a GM-CSF/IL-3 fusion protein (referred to as PIXY321) was to determine whether this protein would stimulate multiple hematopoietic lineages. Receptor binding studies with GM-CSF and IL-3 have shown that a subclass of high-affinity receptors exists with the capacity to bind either growth factor as judged by the ability of these two ligands to cross-compete for receptor binding [18, 191. This cross-competible receptor or receptor complex appears to be expressed preferentially on more immature cells [20], but has also been observed on mononuclear phagocytes [21]. Our notion of creating a fusion protein between GM-CSF and IL-3 was based in part upon examination of these receptor systems and the hypothesis that such a double ended cytokine might preferentially target the crosscompetible receptors. Recent studies have shown that the GM-CSF receptor consists of a low-affinity binding subunit which is converted to a high-affinity binding complex by association with an additional nonbinding subunit [22]. This nonbinding subunit may also be shared with an IL-3 binding complex which could form the basis for the observed cross-competition and also account for the high degree of overlap in the biological activities of these cytokines [22]. The Biology of PIXY321 Initial characterization of recombinant PIXY321 involved the demonstration that both of the ligand domains of the fusion protein were accessible to their respec-

3 WilliamslParWBroxmeyerILu 544 Molecule8 BoundlCall Fig. 1. Scatchard analysis of equilibrium-binding data using radiolabelled GM-CSF or PIXY321 on purified human neutrophils. Both growth factors demonstrate a single class of high-affiity binding sites. The Ka and site number estimated from these analyses were not significantly different for GM-CSF or PIXY321. tive receptors [23]. This analysis was carried out on cell types expressing the pure, non-competible forms of the high- affinity GM-CSF and IL-3 receptors on HL-60 and JM-1 cells, respectively. The affinity constants (KJ and estimated number of binding sites determined by equilibrium binding experiments indicated that both the IL-3 and GM-CSF domains of PIXY321 interacted with their receptors in a similar way to the monomeric ligands [23]. Figure 1 illustrates the comparative results of PIXY321 and GM-CSF binding to human neutrophils. Both GM-CSF and PIXY321 show the presence ofa single class of high-affinity GM-CSF receptor with a K, in the nanomolar range and a similar number of sites per cell (Figure 1). PIXY321 binding was also assessed on cells expressing a subpopulation of cross-competible receptors such as KG-1 and AML-193 cells. PIXY321 was shown to possess an approximate tenfold enhanced binding capacity on these cells compared to IL-3 and an equivalent capacity to GM-CSF [23]. The biological activity of PIXY321 has been assessed on human factordependent cell lines, and nonnal human hematopoietic progenitor cells. An unanticipated finding of these studies has been the observed enhanced specific activity of PIXY321 compared to GM-CSF andor IL-3 [23]. The calculated specific activity of PIXY321 was 10- to 20-fold greater than GM-CSF or IL-3 alone, or a mixture of the two growth factors approximating what was present in the fusion protein [23]. This enhanced activity has been seen on unfractionated, T cell and adherent cell depleted, and highly purified human hematopoietic progenitors from bone marrow (Fig. 2) [23]. Figure 2 shows a comparative titration of PIXY321

4 Hybrid Cytokines as Hematopoietic Growth Factors 545 r O'o:, o ; Cytokine Concentration (nglmi) Fig. 2. BFU-e dose response curves generated in the presence of 1 unit/ml of recombinant erythropoietin alone (U), or in the presence of erythropoietin plus: PIXY321 (O...~),GM-CSF(O...O),IL-3(A..A),orGM-CSF + (O...O). Thetarget cell population was highly puritied human progenitor cells fmm bone marrow expressing high levels of CD34 and low levels of HLA-DR. Cumulative cloning efficiency for granulocyte-macrophage colony-forming units (CFU-gm), BFU-e, and mixed colonyforming units (CFU-gemm) for this population was 50%. versus GM-CSF andor IL-3 on bone marrow erythroid burst-forming units (BFU-e) expressing high levels of CD34 and low levels of HLA-DR isolated by fluorescence-activated cell sorting. These cells had a 50% cloning efficiency for granulocyte-macrophage and erythroid colonies and these data suggest a direct action of the fusion protein on these cells. It is apparent that PIXY321 stimulates BFU-e formation in the presence of erythropoietin at concentrations well below those required for GM-CSF, IL-3 or GM-CSF plus IL-3 (Fig. 2). Preliminary studies of the in vivo actions of PIXY321 in normal or sublethally irradiated rhesus monkeys have shown that the fusion protein has the demonstrable activities of both of the ligand domains [24,25]. In particular, sublethally irradiated animals respond to GM-CSF with accelerated recovery of the neutrophil compartment but no acceleration of platelet recovery. The IL-3 response is the opposite, with only a modest effect on neutrophils and an enhanced rate of platelet regeneration [24]. PIXY321 accelerated the recovery of both circulating compartments and in addition, has been shown to prime neutrophils in vivo, thus enhancing their functional capacity 124,251. These data suggest that the use of PIXY321 may have the significant clinical benefit of enhancing the recovery of both neutrophils and platelets in a number of myelosuppression states.

5 Williams/Park/Broxmeyer/Lu 546 Future Directions The use of GM-CSF and G-CSF to shorten the duration of neutropenia after chemotherapy or bone marrow transplantation [l5,26] is now well established. There is still a need for a drug which can consistently accelerate platelet regeneration in these clinical settings, and our preliminary preclinical data with PIXY321 suggests that this fusion protein may have that potential. In vitro data with PIXY321 also suggest that lower doses of this cytokine may prove efficacious based upon the observed enhanced specific activity as assessed on normal human progenitor cells [23]. The use of fusion proteins such as PIXY321 as a general approach to combination cytokine therapy will require additional study. The unusual biological properties of this GM-CSF/IL-3 combination may be related to the fact that these growth factors share a common receptor subunit which may be important for transducing the signal inside the cell. One could envision coupling two cytokines with a documented capacity to synergize however, as our experience with PIXY321 has shown, one cannot always predict the full range. of biological activities of a fusion protein based upon the biology of the individual monomeric proteins. Studies with additional fusion partners are currently ongoing and should provide some insight into the possibility of using this strategy to maximize the effectiveness of recombinant cytokines. References Broxmeyer, HE. Biomolecule-cell interactions and the regulation of myelopoiesis. Int J Cell Cloning 1986;4: Broxmeyer HE, Williams DE. The production of myeloid blood cells and their regulation during health and disease. CRC Crit Rev Hematol/Oncol 1988;8: Quesenberry PJ. Synergistic hematopoietic growth factors. Int J Cell Cloning 1986;4:3-15. Mochizuki DY, Eisenman JR, Conlon PJ, Larsen AD, Tushinski RT. Interleukin-1 regulates hematopoietic activity: a role previously ascribed to hemopoietin-1. Proc Natl Acad Sci USA 1987;84: Moore MAS, Warren D. Synergy of interleukin-1 and granulocyte colony-stimulating factor: in vivo stimulation of stem cell recovery and hematopoietic regeneration following 5-fluorouracil treatment of mice. Proc Natl Acad Sci USA 1987;84: Anderson DM, Lyman SD, Baird A, et al. Molecular cloning of murine mast cell growth factor: a hematopoietin that is active in both the membrane bound and soluble form. Cell 1990;63: Broxmeyer HE, Cooper S, Lu L, et al. Effect of murine mast cell growth factor (c-kit ligand) on colony formation by human marrow hematopoietic progenitor cells. Blood 1991;77: Zsebo KM, Wypych J, McNiece XK, et al. Identification, purification, and biological characterization of hematopietic stem cell factor from Buffalo rat liver-conditioned medium. Cell 1990;63:

6 Hybrid Cytokines as Hematopoietic Growth Factors de Vries P, Brasel KA, Eisenman JR, Alpert AR, Williams DE. The effects of recombinant mast cell growth factor on purified murine hematopoietic stem cells. J Exp Med l991;173: ll. 10 -McNiece IK, Langley KE, Zsebo KM. The role of recombinant stem cell factor in early B cell development: synergistic interaction with IL-7. J Immunol 1991;146: Namen AE, Widmer MB, Voice R, Christensen S, Braddy S, Lyman SD, Williams DE. A ligand for the c-kit proto-oncogene stimulates lymphoid progenitor cells in vitro. Exp Hematol 1991;19:497a. 12 Bot FJ, van Eijk L, Schipper P, Lowenburg B. Effects of human interleukin-3 on granulocytic colony-forming cells in human bone marrow. Blood 1989;73:1157-U Fletcher MP, Gasson JC. Enhancement of neutmphil function by granulocytemacrophage colony-stimulating factor involves recruitment of a less responsive subpopulation. Blood 1988;7l: Kaplan SS, Basford RE, Wing EJ, Shadduck RK. The effect of recombinant human granulocyte.-macrophage colony-stimulating factor on neutmphil activation in patients with refractory carcinoma. Blood l989;73: Glaspy JA, Golde DW. Clinical trials with myeloid growth factors. Exp Hematol 1990;18 :ll37-ll Ganser A, Siepelt G, Lindermann A, Effects of recombinant human interleukin-3 in patients with myelodysplastic syndromes. Blood 1990;76: Williams DE, Park LS. Hematopoietic effects of a granulocyte-macrophage colony stimulating factorlinterleukin-3 fusion protein. Cancer 1991; Gesner TA, Mufson RA, Norton CK, Turner KJ, Yang YC, Clark SC. Specific binding, internalization and degradation of human recombinant interleukin-3 by cells of the acute myelogenous leukemia line, KG-1. J Cell Physiol 1988;136: Park LS, Friend D, Price V, Anderson D, Singer J, Prickett KS, Urdal DL. Heterogeneity in human interleukin-3 receptors. J Biol Chem l989;264: Cannistra SD. Koenglesman M, DiCarlo J, Grosbek P, Griffii JD. Differentiationassociated expression of tm functionally distinct classes of granulocyte-macrophage colony-stimulating factor receptors by human myeloid cells. J Biol Chem 1990;265: Elliott MJ, Vadas M, Egllngton J, et al. Recombinant human interleukin-3 and granulocyte macrophage colony-stimulating factor shaw common biological effects and bdmg characteristics on human monocytes. Blood 1989;74: Hayashida K, Kitamura T, Gonnan DM, Arai K-I, Yokota T, Mipjima A. Molecular cloning of a second subunit of the receptor for human gmulocyte-macrophage colony-stimulating factor (GM-CSF): reconstitution of a high affinity GM-CSF mptor. Proc Natl Acad Sci USA 1990;87: Curtis BM, Williams DE, Broxmeyer HE, et al. Enhanced activity of a human granulocyte-macrophage colony-stimulating factor-interleukin-3 fusion protein. Proc Natl Acad Sci USA 1991;88: Williams DE, Farese A, Dunn J, Freiden E, Park LS, MacVittie TJ. In vivo effects of a GM-CSF/IL-3 fusion protein (PIXY321) in sublethally irradiated monkeys. Exp Hematol 1991;19:479a. 25 MacVittie TJ, Farese AM, Patchen ML, Williams DE. Hematologic effects of in viw administration of recombinant GM-CSFIIL-3 fusion protein (pixy321) in normal primates. Exp Hematol 1991;19:177a. 26 Peters WP. The myeloid colony-stimulating factors: Introduction and overview. Semin Hematol 1991;28:1-5.