Blood

We investigated the proliferation-inducing effects of human recombinant interleukin-7 (IL-7) on acute lymphoblastic leukemia (ALL) cells. It is shown that IL-7 stimulates DNA synthesis in ALL cells of 6-cell precursor (n = 5) as well as immature T-cell origin (n = 2). Cytogenetic analysis of the cells of four patients proliferating in IL7supplemented cultures established the leukemic descendence of the IL-7-responsive cells. ’251-lL-7 binding

I murine growth factor derived from bone marrow stroma cells and is capable of supporting the growth of precursor B cells in vitro.' After the molecular cloning of the murine IL-7, a human IL-7 c D N A homologue has been isolated from a hepatoma cell line c D N A library, which has allowed the large scale production of pure recombinant human IL-7.2,3 It has become clear that murine IL-7 not only stimulates the proliferation of mouse precursor B cells, but that it also promotes the proliferation and development of murine thymocytes and their fetal Recently, it has been shown that IL-7 stimulates the proliferation of human peripheral blood T lymphocytes. 8 The effects of IL-7 on human Band T-cell precursors (BCP, TCP) have as yet not been established. In this study we investigate the action of IL-7 on B-cell precursor-acute lymphoblastic leukemia (BCP-ALL) and immature T-ALL cells. It is shown that IL-7 is able to induce a proliferative response in these cell types. I n addition, some of the binding characteristics of IL-7 membrane receptors present on A L L cells are presented.

MATERIALS AND METHODS
Bone marrow or peripheral blood samples of seven untreated ALL patients (five BCP-ALL, two T-ALL) were obtained after informed consent. Immunophenotypic characteristics of these cases are listed in Table 1. Leukemic cells were recovered by Ficoll-Isopaque (Nygaard, Oslo, Norway) separation.' E-rosette forming T lymphocytes were then removed from the ALL samples by rosetting with 2-aminoethylisothiouronium bromide-treated sheep red blood cells (AET-SRBC) and Ficoll separation.' This also was feasible in the two cases of T-ALL, as the leukemic cells did not form rosettes with the AET-SRBC. These cell samples contained less than 0.5% E-rosette-forming T cells. Subsequently, cells ( 107/mL culture medium) were incubated in 6-cm Petri dishes (Greiner, Alphen aan den Rijn, The Netherlands) for 1 hour at 37OC to remove plastic

Leukemic cells.
experiments with the cells of one patient and with two ALL cell lines showed the presence of two types of IL-7 receptors: one with a high affinity (kd 29 to 51 pmol/L) and one with a low affinity (kd 2.3 to 76 nmol/L) for the ligand. We conclude that IL-7 is one of the cytokines involved in the complex regulation of ALL cell proliferation. 0 1990 by The American Society of Hematology. adherent monocytes from the cells samples. ALL cell samples were either used fresh or after cryopreservation using a controlled freezing apparatus (Planer Biomed, Sunbury-on-Thames, UK).' BCP-ALL cell lines NALM-6 and ALL 202 were maintained in suspension culture as described.''," Immunofluorescence and purification of ALL cells byPuorescenceactivated cell sorting (FACSJ. Immunofluorescence procedures, flow cytometric analysis, and monoclonal antibodies (MoAbs) have been described in detail elsewhere.' In a majority of the experiments ALL cells were further purified by cell sorting (FACS 440, Becton Dickinson, Mountain View, CA) before incubations in culture. To this end, BCP-ALL were stained with a mixture of CDIO/CD19 MoAbs and goat-anti-mouse immunoglobulin (Ig) coupled to fluorescein isothiocyanate. Residual CD3 + normal T lymphocytes that could potentially contaminate the cultures were removed from the T-ALL samples by cell sorting as described." DNA synthesis in culture was assessed by uptake of 'H-thymidine ('H-TdR, 2 Ci/mmol/L, Amersham, UK) essentially as described.' However, a major modification is that in the present experiments a serum-free culture medium was used that consists of Iscove's modified Dulbecco's medium supplemented with human iron-saturated transferrin (7.7 x mol/L), insulin (1  Equilibrium binding of the '251-IL-7 occurred under these conditions. To determine nonspecific binding, incubations in the presence of excess (SO nmol/L) nonlabeled IL-7 were performed in parallel. Experimental procedures and calculations were identical to those described previously." Two-affinity receptor analysis was performed using the ENZ FITTER data analysis program (Sigma Chemical Co, St Louis, MO).

Radioiodination of IL-7 and binding experiments.
The pure IL-7 was labeled with '*'I using the Bolton-Hunter reagent (Amermacrophage colony-stimulating factor (GM-CSF).I3 Sodium do&-cy1 sulfate polyacrylamide (1 5%) gel electrophoresis under reducing conditions and subsequent autoradiography did not indicate the presence of carrier proteins in the IL-7 preparation. Self-displace-

IL-7-induced proliferative response in ALL cells. First,
Patients nos. 4 (Pre-B ALL) and 6 (T-ALL) to assess the optimal C o~~~~t r a t i o n for stimulation of DNA synthesis (Fig  1). In both pre-B ALL and T-ALL cells 30 U/mL of IL-7, which is equivalent to approximately 170 pmol/L, induced a sham), exactly as has been described for IL-3 and granulocyte  The data summarized in Table 2 demonstrate that only the  leukemic

cells entered metaphasis under the applied in vitro
Binding of titrated concentrations of radiolabeled IL-7 to the ALL cells of patient no. 3 are shown in Fig 2 (left panel). Scatchard conversion of these data produced a biphasic plot, indicative of the existence of high and low affinity binding sites (Fig 2, right panel). Similar IL-7 binding characterisitcs were observed with the ALL cell lines NALM-6 and ALL 202. Estimations of IL-7 receptor numbers and affinities are listed in Table 3.  Table 1. A significant proliferative response to IL-7 was evident in all cases with the exception of patient no. 3. The cells of the latter patient proliferated spontaneously, and addition of IL-7 to the Insights into the mechanisms controlling the proliferation and maturation of normal and neoplastic human B-and T-cell precursors have remained limited. This has been due to a lack of purified growth stimuli that specifically act on early human lymphoid cells. Although evidence has been obtained for a stimulatory role of IL-3 in inducing DNA synthesis in certain cases of BCP-ALL," in another group of patients no consistent proliferation inducing effects could be attributed to a panel of recombinant growth factors that included the ILs 1, 2, 3, 4, and 6, as well as GM-, G-, and M-CSF.9 A similar deficiency in growth factor responses has been noted in T-ALL, although variable proliferative responses to IL-2 have been rep~rted.'~.''.~~ culture only slightly elevated the rate of DNA synthesis. Because IL-3 has been reported to provoke a proliferative response in BCP-ALL," the effects of IL-7 on BCP-ALL were also tested in combination with IL-3. No additional or synergistic effects of IL-3 on the action of IL-7 were noted.
Synergistic effects of IL-2 or IL-4 on the IL-7-induced proliferative response in the T-ALL cells also were not apparent.

Cytogenetic analysis of cells proliferating in IL-7-
Based on the initial reports on the activity of murine IL-7,Is2 we speculated that the human homologue could be a candidate regulator of ALL cell pr~liferation.~ In the present study we have confirmed this suggestion and showed that IL-7 is capable of inducing a significant, although generally quite moderate, proliferative response in BCP-ALL as well as T-ALL cells. This relatively low responsiveness probably illustrates that IL-7 is only one of the factors among others controlling the growth of normal and neoplastic lymphoid containing cultures. To rule out the possibility that the cells which responded to IL-7 were contaminant normal cells, cytogenetic analysis was performed on cells of four patients that had been cultured for 2 to 3 days in the presence of IL-7.
precursors. The observations that IL-3 stimulated DNA synthesis in BCP-ALL patient nos. 2 and 4, and that some stimulatory effects of IL-2 and IL-4 in T-ALL case 6 were noted, support this idea. In view of this concept, the possibil-

1251-lL7 bound (pM)
For personal use only. on August 23, 2017. by guest www.bloodjournal.org From ity that the ILs 1, 2, 4, 5, 6, and tumor necrosis factor cooperate with IL-7 in inducing a proliferative response in ALL cells also has been addressed. We have not yet obtained evidence for such synergistic actions beteen IL-7 and other factors in ALL ( Table 1 and data not shown). It has also been examined in this study as to whether IL-7 induced maturation of the ALL cells. In three cases of BCP-ALL, ie, patient nos. 1, 2, and 4, IL-7 appeared to weakly induce/enhance the expression of CD20 on the cell membrane (data not shown), which is of interest given the evidence that the CD20 antigen is involved in the activation of B cells.'9 No other immunophenotypic alterations indicative of progressive maturation during culture, such as the acquisition of cytoplasmic or surface Ig, CD21, or the loss of CDlO were noted. Likewise, in the T-ALL patient nos. 6 and 7, no evidence for the inducibility of differentiation antigens CD1, surface CD3, CD4, and CD8 was obtained (data not shown). Thus, we consider the maturation-inducing effects of IL-7 on ALL cells very limited.
In a previous study' in which a serum containing culture assay had been applied, we encountered the outgrowth of contaminant normal cells as a serious problem for evaluating the proliferative response of BCP-ALL cells. Typically, the effects of a crude source of B-cell growth factors (BCGF) could only be measured when the ALL cells had been purified by cell sorting before culture. Without this purification step, only nonleukemic cells proliferated in response to BCGF. To circumvent this potential pitfall, we performed a major portion of our present experiments with FACS purified cell samples. However, it should be noted that under the currently applied serum-free conditions, exclusively the leukemic cells entered metaphase in IL-7 supplemented cultures, even when FACS purification had been omitted (Table   The data of radiolabeled IL-7 binding experiments (Fig 2,  Table 2) point toward the presence of both high and low affinity type IL-7 receptors on ALL cells. Presently we have no experimental data to explain the dual affinity characteristics of IL-7 receptors. One possibility that emerges is that this phenomenon reflects the involvement of di-or oligomerization of receptor chains in the formation of high affinity binding sites, and that low affinity binding occurs to single receptor chains, as has been proposed for a number of other growth factor receptors, eg, IL-2 receptors.20 Over the years, the application of in vitro cultures for human leukemic lymphoid precursor cells has remained particularly troublesome due to the lack of well-defined assays. Our finding that IL-7 is one of the factors stimulating ALL cell proliferation illustrates that the current interest in the molecular cloning and large-scale production of new growth factors is of vital importance for the development of reliable culture assays for ALL.