Dr Kazuhiro Nishii, The Second Department of Internal Medicine, Mie University School of Medicine, 2-174 Edobashi, Tsu, Mie 514-8507, Japan. e-mail: firstname.lastname@example.org
We searched for cytokines with the potential to support the survival of human B-cell precursor acute lymphoblastic leukaemia (pre-B ALL) cells. 47 patients with pre-B ALL were classified into four stages: stage I, CD19+CD10−CD20−; stage II, CD19+CD10+CD20−; stage III, CD19+CD10+CD20+cytoplasmic μ-heavy chain (cμ)−; stage IV, CD19+CD10+CD20+cμ+. Interleukin (IL)-3 receptor α chain (IL-3Rα) was expressed in all stages, whereas the expressions of IL-7Rα and IL-2Rα were pronounced in stages IV and II, respectively. Neither IL-3, IL-7 nor IL-2 supported the survival of pre-B ALL cells. When pre-B ALL cells were layered on stromal, MS-10, cells, viability of the pre-B ALL cells increased. Addition of IL-3 to culture containing MS-10 cells enhanced the survival of pre-B ALL cells in all cases, whereas addition of IL-7 augmented the survival of pre-B ALL cells of some cases of stage III and all cases of stage IV. The survival of pre-B ALL cells was also supported by the conditioned media of MS-10 cells. Stromal-cell-derived factor 1 (SDF-1) supported the survival of pre-B ALL cells. Effects of the conditioned media of MS-10 cells were abrogated by an anti-SDF-1 neutralizing antibody. The extent of survival of pre-B ALL cells supported by stromal cells and IL-3 and IL-7, correlated with the expression level of bcl-2 protein. The effects of stromal cells may be in part related to SDF-1.
We searched for cytokines which support the survival of pre-B ALL cells and found that although interleukin (IL)-3 or IL-7 alone did not promote the survival of pre-B ALL cells, these cytokines were effective in supporting survival of pre-B ALL cells, in the presence of stromal cells. The enhanced survival of pre-B ALL cells was associated with the up-regulation of Bcl-2. Conditioned media of the stromal cells also supported the survival of pre-B ALL cells. Stromal cell-derived factor 1 (SDF-1) (Tashiro et al, 1993; Nagasawa et al, 1994) was found to support survival of pre-B ALL cells. Effects of the conditioned media were neutralized by an anti-SDF-1 monoclonal antibody.
MATERIALS AND METHODS
Patients and leukaemic cells
47 Japanese adult patients with untreated pre-B ALL were studied following informed consent. From leukaemic bone marrow and peripheral blood samples, mononuclear cells were separated by Ficoll-Hypaque density gradient centrifugation. Samples stored at −190°C in RPMI 1640 medium (Nissui Pharmaceutial Co. Ltd, Tokyo, Japan), with 20% heat-inactivated fetal calf serum (FCS) (Hyclone Laboratories, Logan, U.S.A.), and 10% dimethylsulphoxide, were also used as needed. In all cases, >90% of the isolated cells had leukaemic cell-like morphology. Before immunostaining, the cells were treated with 5% heat-inactivated human AB serum to block non-specific binding of the antibodies to the receptor for IgG Fc portion (FcγR). CD19 (B4), CD10 (J5), CD20 (Leu16) and CD22 (Leu14) were used as B-cell markers. CD2 (T11), CD3 (Leu4), CD4 (Leu3a), CD5 (Leu1), CD7 (TP40) and CD8 (Leu2a) were used as T-cell markers. CD11b (OKM1), CD13 (MCS2), CD33 (My9) and CD34 (My10) were used as myeloid cell markers. For patients with >80% of leukaemic blasts expressing both CD19 and HLA-DR and lacking surface immunoglobulins, pre-B ALL was diagnosed. Cytoplasmic μ-heavy chain (cμ) was detected on acetone-fixed cytospin smears by a direct immunofluorescence procedure using a fluorescein isothiocyanate (FITC)-conjugated F(ab′)2 fragment of goat anti-human immunoglobulin.
Flow cytometric analysis of cytokine receptors and apoptosis-associated molecules
Mouse monoclonal antibodies against human IL-3 receptor α chain (IL-3Rα), c-kit molecule (c-Kit) and granulocyte colony-stimulating factor (G-CSF) receptor (G-CSFR) were purchased from PharMingen (San Diego, U.S.A.). Mouse monoclonal antibody against human IL-7Rα was obtained from Immunotec (Marseille, France). Mouse monoclonal antibody against human IL-2Rα was obtained from Becton Dickinson Immunocytometry Systems (San Jose, Calif.). Mouse monoclonal antibody against human granulocyte/macrophage colony-stimulating factor (GM-CSF) receptor α chain (GM-CSFRα) was purchased from Genzyme (Cambridge, U.S.A.). The optimal concentration of the antibodies was determined in titration experiments. The second reagent used was an FITC-conjugated goat anti-mouse immunoglobulin (GAM-FITC) (Zymed Laboratories, San Francisco, U.S.A.). Purified mouse IgG1 (Zymed Laboratories) and IgG2a (Becton Dickinson Immunocytometry Systems) were used as isotype controls. Expression of cytokine receptors was analysed using a flow cytometer (Cytron Absolute, Ortho-Clinical Diagnostic, Tokyo, Japan). Cytokine receptors were quantitatively expressed as the antibody binding capacity (ABC), the number of antibody molecules bound per cell, using DAKO QIFIKIT software (Dakopatts, Glostrup, Denmark), according to the manufacturer's recommendations (Lanza et al, 1997). The ABC values referred to a gated leukaemic population set on scatter properties.
For cytoplasmic staining of apoptosis-associated molecules, including Bcl-2, bcl-x protein (Bcl-x) and bax protein (Bax), an FITC-conjugated mouse monoclonal antibody against Bcl-2 (DAKO, Glostrup, Denmark), a rabbit polyclonal antibody against Bcl-x (Santa Cruz Biotechnology, Santa Cruz, U.S.A.) and a rabbit polyclonal antibody against Bax (Santa Cruz Biotechnology) were used. Leukaemic cells were fixed with Ca2+, Mg2+-free phosphate-buffered saline (PBS) supplemented with 4% paraformaldehyde at 4°C for 30 min. To assess the expression of Bcl-2, the fixed cells were washed twice with the staining buffer (1% FCS and 0.1% sodium azide in PBS) and incubated with the FITC-conjugated anti-Bcl-2 antibody in the permeabilization buffer (1% FCS, 0.1% sodium azide, and 0.1% saponin in PBS). To detect intracytoplasmic Bcl-x and Bax, the cells were washed with PBS containing the staining buffer and incubated with the FITC-conjugated goat anti-rabbit IgG antibody (DAKO) in the permeabilization buffer. FITC-conjugated mouse IgG1 (DAKO) and rabbit IgG (Zymed Laboratories) served as isotype controls. Samples were analysed using a Cytron Absolute cytometer. The expression of Bcl-2, Bcl-x and Bax was quantitatively expressed as ABC, as described above. In some experiments, samples were analysed using a FACScan cytometer (Becton Dickinson Immunocytometry Systems).
Leukaemic cells were incubated at 37°C in a humidified atmosphere flushed with 5% CO2 in the 25 cm2 flask (Nunc, Roskilde, Denmark) at a final concentration of 5 × 105 cells/ml in RPMI 1640 with 5% FCS, 5 × 10−5m 2-mercaptoethanol (Sigma Chemical, St Louis, U.S.A.), and designated cytokines. Cell culture experiments were restricted to patients from whom leukaemic cells had been obtained in an adequate number. In some experiments leukaemic cells were cultured on the feeder layer of stromal cells. MS-10 cells served as stromal cells (Itoh et al, 1989; Jiang et al, 1998). MS-10 cells were maintained in RPMI 1640 with 5% FCS. When a 60% confluent monolayer of MS-10 cells had been prepared, the supernatant was removed from the culture flask. The monolayer in the flask was gently washed three times with fresh medium, and thereafter leukaemic cells were cultured onto the monolayer. The leukaemic cells were recovered by gently pipetting and non-adherent cells were assessed for subsequent experiments. Recombinant human IL-3, recombinant human IL-7 and recombinant human IL-2 were purchased from Genzyme. Recombinant human SDF-1 and mouse anti-human SDF-1 neutralizing monoclonal antibody were obtained from R & D Systems (Minneapolis, U.S.A.). Concentrations of the cytokines were as follows: IL-3, 10 ng/ml; IL-7, 10 ng/ml; IL-2, 100 U/ml; SDF-1, 250 ng/ml. The neutralizing antibody against SDF-1 was used at the concentration of 75 μg/ml.
Preparation of stromal cell-conditioned media
MS-10 cells were cultured in RPMI 1640 containing 5% FCS. When approximately 60% of the flask was covered with MS-10 cells, the supernatant was gently removed and fresh medium was added. After 7 d of incubation, the conditioned media were collected, centrifuged, and passed through a 0.22 μm filter (Millipore, Bedford, Mass.). The conditioned media were stored at −80°C until use. The conditioned media were added to cultures at a final concentration of 20% (vol/vol).
Determination of viable leukaemic cells
Leukaemic cells were incubated in RPMI 1640 with 5% FCS. At various intervals after the initiation of culture, viable cells were counted by trypan blue dye (Sigma Chemical) exclusion, using a Neubauer counting chamber. Leukaemic cells obtained from cultures containing the stromal cells were readily identified, as stromal cells could be characterized by their size and morphology.
Cell-cycle analysis was done as described, but with some modification (Nishii et al, 1996). Leukaemic cells were fixed with 70% ethanol PBS at 4°C for 30 min. After being washed twice with PBS, the cells were resuspended in 1 ml of PBS containing 40 μg of propidium iodide (PI) (Sigma Chemical) and 10 μg of RNase (Sigma Chemical) for 30 min in the dark and, subsequently, analysed using a FACScan cytometer. Apoptotic cells were detected in the red fluorescence channel (585/42 nm) as a sub-G1 peak, as viewed on a linear scale. Doublets were discriminated by gating on a plot of area against width on the red fluorescent signal. Cell debris was excluded by appropriately raising the forward-scatter threshold.
Western blotting was done using a modification of the technique described elsewhere (Nishii et al, 1996). Leukaemic cells (1 × 106) were homogenized in lysis buffer containing 10 mm phosphate, pH 7.0, 150 mm EDTA, 1% Triton-X-100, 0.05% SDS, 20 mm NaF, 1 mm NaVO3, 5 μm molybdic acid, 60 mg/ml soybean trypsin inhibitor, 60 mg/ml leupeptin, 60 mg/ml bestatin and 1 ml PNSF. Total protein was subjected to 15% SDS-PAGE and transferred onto IPVDF membrane (Nihon Millipore, Yonezawa, Japan). The membrane was probed with mouse anti-Bcl-2 monoclonal antibody (DAKO), rabbit anti-Bcl-x polyclonal antibody (Santa Cruz Biotechnology) or rabbit anti-Bax polyclonal antibody (Santa Cruz Biotechnology). The blots were washed four times with PBS containing 0.05% Tween 20 for 1 h at room temperature, and then incubated with horse-radish peroxidase-conjugated goat anti-mouse antibody or goat anti-rabbit antibody (Amersham International, Little Chalfont, U.K.) for 30 min at room temperature. The proteins were visualized using chemiluminescence (ECL Western blotting detection kit, Amersham International) followed by exposure to autoradiography film.
Expression of cytokine receptors
We first categorized pre-B ALL into four immunophenotypically defined groups (Nadler et al, 1984; Foon & Todd, 1986). The phenotype of each group was as follows: stage I, CD19+CD10−CD20−; stage II, CD19+CD10+CD20−; stage III, CD19+CD10+CD20+cμ−; stage IV, CD19+CD10+ CD20+cμ+. Stages I, II, III and IV comprised eight, 18, nine and 12 cases, respectively. The expression of IL-3Rα, IL-7Rα, IL-2Rα, c-Kit, G-CSFR and GM-CSFRα were screened using flow cytometry. Among these cytokine receptors, c-Kit, G-CSFR and GM-CSFRα were below the detectable level (data not shown). IL-3Rα, IL-7Rα and IL-2Rα were detected on pre-B ALL cells (Fig 1). The expression of IL-3Rα on pre-B ALL cells was observed in all stages, although IL-3Rα was highly expressed on pre-B ALL cells of stage III. Pre-B ALL cells of stage IV expressed IL-7Rα at a level higher than for pre-B ALL cells of other stages. IL-2Rα was detected in pre-B ALL cells of stage II alone.
Survival of pre-B ALL cells in the presence of IL-3, IL-7 or IL-2
Since IL-3Rα, IL-7Rα and IL-2Rα were expressed on pre-B ALL cells, pre-B ALL cells may respond to IL-3, IL-7 or IL-2. We examined the effects of IL-3, IL-7 and IL-2 on survival of pre-B ALL cells from 14 patients, using a suspension culture system. Leukaemic cells (5 × 105/ml) were incubated for 96 h in the presence of IL-3, IL-7 or IL-2. The survival of pre-B ALL cells was assessed by directly counting viable cell numbers. As shown in Fig 2, IL-3, IL-7 or IL-2 did not promote the survival of pre-B ALL cells, compared with cultures in the absence of each cytokine. As much as 10% pre-B ALL cells of stage IV (cases 12–14) were viable in the cultures, whereas >20% of pre-B ALL cells of stages I, II and III were viable. After 96 h of culture, some pre-B ALL cells showed morphologic evidence for apoptosis such as nuclear fragmentation and loss of cell volume (data not shown). Thus, DNA contents of pre-B ALL cells were analysed in a representative case for each stage (case 3, stage I; case 6, stage II; case 10, stage III; case 14, stage IV), using flow cytometry (Fig 3). The sub-G1 population was high in pre-B ALL cells of stage IV, relative to that of pre-B ALL cells of other stages. The addition of IL-3, IL-7 or IL-2 to cultures did not affect the DNA profile of pre-B ALL cells, in any case. These data suggest that IL-3, IL-7 or IL-2 did not stimulate the growth of pre-B ALL cells. After 10 d of culture, >95% of pre-B ALL cells for each stage were not viable (Fig 4).
Expression of apoptosis-associated molecules in pre-B ALL cells
We next analysed the expression of Bcl-2, Bcl-x and Bax in pre-B ALL cells by flow cytometry. Bcl-2 was detectable in pre-B ALL cells of all stages. The expression level of Bcl-2 was low in pre-B ALL cells of stage IV (cases 12–14), in comparison to the levels in pre-B ALL cells of other stages (Fig 5). Western blotting was also done and results were comparable to data obtained with flow cytometric analysis (Fig 6). Although Bcl-x and Bax were also detected in pre-B ALL cells, expression levels did not differ among stages (data not shown).
Effects of IL-3, IL-7 or IL-2 on survival of pre-B ALL cells in the presence of stromal cells
It has been reported that bone marrow stromal cells can promote the survival of normal (Rawlings et al, 1995; Nishihara et al, 1998) and leukaemic B-cell precursors in vitro (Manabe et al, 1992; Kumagai et al, 1996; Nishigaki et al, 1997). We first tested the effects of stromal cells on survival of pre-B ALL cells, using a murine stromal cell-supported culture system. MS-10 cells served as stromal cells (Itoh et al, 1989; Jiang et al, 1998). When viable cells were counted after 96 h of incubation, the majority of pre-B ALL cells of all samples were recovered in the presence of MS-10 cells (data not shown). Accordingly, we attempted to determine whether IL-3, IL-7 or IL-2 is virtually functional. We counted the viable cells after 10 d of incubation (Fig 7). IL-3 efficiently augmented the survival of pre-B ALL cells of all stages in cultures for 10 d in the presence of MS-10 cells. The addition of IL-7 to cultures with MS-10 cells also enhanced the survival of pre-B ALL cells in some cases of stage III (cases 7, 9 and 11) and all cases of stage IV (cases 12–14). The capability of IL-7 to promote the survival of pre-B ALL cells cultured with MS-10 cells was significant in the cases of stage IV. In the presence of MS-10 cells, IL-2 did not affect the survival of pre-B ALL cells. After culture, the immunophenotype of pre-B ALL cells did not differ from that seen before the culture.
Effects of conditioned media of MS-10 cells and SDF-1 on survival of pre-B ALL cells
To determine whether the effects of MS-10 cells are mediated by direct cell to cell interaction or through soluble secreted factors, we examined the effects of conditioned media of the MS-10 cells that had been cultured for 7 d on survival of pre-B ALL cells of cases 7, 13 and 14 (Fig 8). The conditioned media potently enhanced the survival of pre-B ALL cells cultured in the absence of MS-10 cells. A factor, termed SDF-1 or pre-B cell growth-stimulating factor, produced by murine stromal cells has been purified and its gene cloned (Tashiro et al, 1993; Nagasawa et al, 1994). We next studied the effects of SDF-1 on survival of pre-B ALL cells by estimating the number of viable cells after 96 h of incubation. SDF-1 increased the viability of pre-B ALL cells in the absence of MS-10 cells. We also tested the effects of anti-SDF-1 neutralizing monoclonal antibody on the survival of pre-B ALL cells supported by the conditioned media of MS-10 cells. The anti-SDF-1 neutralizing monoclonal antibody substantially inhibited the effects of conditioned media of MS-10 cells (Fig 8). This effect was not shown when control antibody was added in the conditioned media of MS-10 cells.
Up-regulation of Bcl-2 by stromal cells and cytokines
The mechanism by which MS-10 cells plus IL-3 or IL-7 enhanced the survival of pre-B ALL cells was also investigated. The low level of Bcl-2 expression in pre-B ALL cells of stage IV, the viability of which was extremely decreased (Figs 2–6) raises the possibility that Bcl-2 may be involved in survival of pre-B ALL cells. Using flow cytometry, we examined the expression of Bcl-2 in pre-B ALL cells cultured for 96 h in the presence of MS-10 cells plus IL-3 or IL-7. We selected case 5 of stage II as a representative of pre-B ALL cases which highly responded to the combination of MS-10 cells plus IL-3 but not of MS-10 cells plus IL-7. As shown in Fig 9, MS-10 cells alone up-regulated the expression of Bcl-2, compared to the absence of MS-10 cells. The additional up-regulation of Bcl-2 was apparent in the presence of MS-10 cells plus IL-3 but not of MS-10 cells plus IL-7. Case 14 of stage IV was also tested because the leukaemic cells were highly reactive with both IL-3 and IL-7 in the presence of MS-10 cells. IL-3 and IL-7 enhanced the up-regulation of Bcl-2 in pre-B ALL cells induced by MS-10 cells. We also confirmed changes in Bcl-2 expression in some samples by Western blot analysis (Fig 10). The results were comparable to those obtained using flow cytometry techniques.
We investigated factors with the potential to support the survival of pre-B ALL cells. Uckun et al (1989a) reported that although pre-B-ALL cells had IL-3 binding capacity, the response to IL-3 was observed in only 2/12 samples. Although other investigators have also reported that IL-3 or IL-7 stimulates pre-B ALL cells, the responsiveness varied considerably among patients (Wörmann et al, 1989; Masuda et al, 1990; Skjønsberg et al, 1991; Digel et al, 1991; Smiers et al, 1995). Our data show that pre-B ALL cells expressed IL-3Rα, IL-7Rα and IL-2Rα but did not respond to IL-3, IL-7 or IL-2. However, the finding that survival of pre-B ALL cells was promoted by IL-3 in the presence of stromal cells provides evidence that IL-3 does act on pre-B ALL cells. This is reminiscent of another report showing that IL-3, in combination with other factors, induces the generation of normal human B-cell precursors (Saeland et al, 1991). In agreement with the observation of Nishihara et al (1998) that IL-7 stimulated normal human CD19+CD10+CD20+ but not CD19+CD10+CD20− B-cell precursors, IL-7 enhanced the survival of pre-B ALL cells supported by MS-10 cells, preferentially pre-B ALL cells with mature phenotypes. The responses of leukaemic B-cell precursors to IL-3 and IL-7 appear to be similar to those of normal corresponding cell populations. Although it has been demonstrated that IL-2 stimulated the proliferation of pre-B ALL cells (Touw et al, 1985), our findings indicated that IL-2 was not active on pre-B ALL cells, regardless of stromal cells. Further studies will be necessary to clarify the action of IL-2 on pre-B ALL cells. Stem cell factor (SCF), the ligand for c-Kit, plays an important role in the development of B-cell precursors in mice (McNiece et al, 1991; Galli et al, 1994; Broudy, 1997). It has been reported that human B-cell precursors do not express c-Kit, although they are generated from CD34+ haemopoietic progenitors in response to SCF (Ryan et al, 1997; Nishihara et al, 1998). We did not detect c-Kit on the pre-B ALL cells we tested. Therefore it is conceivable that SCF may be not active on normal or leukaemic human B-cell precursors.
It is known that bone marrow stromal cells support the growth of normal and leukaemic B-cell precursors by direct cell contact (Manabe et al, 1992; Nishihara et al, 1998). We observed that MS-10 cells increased the number of viable pre-B ALL cells. Our data obtained using conditioned media from MS-10 cells indicated that the soluble factor(s) secreted was responsible for the potential of MS-10 cells to promote the survival of pre-B ALL cells. Interestingly, SDF-1 promoted the survival of pre-B ALL cells and the anti-SDF-1 neutralizing antibody significantly suppressed the survival of pre-B ALL cells supported by the conditioned media of MS-10 cells. It is therefore suggested that SDF-1 is one of the potent regulators of pre-B ALL cells. This could be important, because factors required for the survival of pre-B ALL cells have been difficult to identify.
Although overexpression of Bcl-2 has been noted in pre-B ALL (Campana et al, 1993; Coustan-Smith et al, 1996; Ucken et al, 1997), the relevance of Bcl-2 remains unclear. We observed that the level of Bcl-2 was low in cμ+ pre-B ALL cells, which are more susceptible to apoptosis. During murine pre-B cell development, Bcl-2 expression is low in cμ+ pre-B cells (Merino et al, 1994; Nishii et al, 1998). Therefore the expression level of Bcl-2 may differ among the distinctive developmental stages of pre-B cells. We next tested the Bcl-2 expression in pre-B ALL cells, the survival of which was enhanced by stromal cells and the cytokines. The enhancement of the survival of pre-B ALL cells was associated with the up-regulation of Bcl-2 expression in pre-B ALL cells, a finding which strongly suggests that Bcl-2 is involved in the survival of pre-B ALL cells. Our present study has shed some light on signals that regulate the survival of human pre-B ALL cells.
We thank Dr K. J. Mori, Niigata University, Niigata, Japan, for providing the cloned murine bone marrow stromal cells (MS-10 cells), Professor M. F. Greaves and Dr I. Titly, Leukaemia Research Fund Centre at the Institute of Cancer Research, London, for helpful discussion, and M. Ohara for critical commemts.
This work was supported by research grants from the Ministry of Education, Science, Sports, and Culture and from the Ministry of Health and Welfare, Japan, and a Grant-in-Aid (1998: K.N.) from the Mie Medical Research Foundation.