IL-15 has been found to activate NF-κB in various types of cells. However, the role of this transcription factor in IL-15- and IL-21-stimulated murine bone marrow (BM) cells is unclear. In this study, we demonstrated that both IL-15 and IL-21 are capable of delaying BM cell factor deprivation-induced apoptosis, but only IL-15 induced their proliferation. Following separation of BM cells into myeloid (CD11b+) and lymphoid (CD11b−) cell populations, we found that IL-15, but not IL-21, significantly induced proliferation in both cell populations. Both cytokines significantly delayed apoptosis, but only in CD11b− BM cells. IL-15Rα, CD122 (IL-2/15Rβ), and common γ-chains (CD132) were expressed in both populations, whereas IL-21Rα was expressed only in CD11b− BM cells. In addition, we demonstrated that IL-15-induced BM cell proliferation was significantly inhibited in NF-κBp50−/− mice when compared with littermate controls. The ability of IL-15 and IL-21 to delay BM cell apoptosis was slightly inhibited in NF-κBp50−/− mice, whereas the antiapoptotic effect of LPS was markedly reversed. We conclude that IL-15, but not IL-21, induces BM cell proliferation and that both cytokines delay BM cell apoptosis. These biological activities were preferentially observed in CD11b− BM cells. Using NF-κBp50−/− mice, we demonstrated for the first time that NF-κB plays a greater role in IL-15-induced cell proliferation than in IL-15- and IL-21-induced suppression of apoptosis.
Cytokines IL-15 and IL-21 are members of the CD132 common γ-chain (γc)3-dependent cytokine family, which also includes IL-2, IL-4, IL-7, and IL-9 (1, 2, 3, 4). Structurally, IL-21 is most closely related to IL-2, IL-4, and IL-15. Although IL-21 is not itself a potent mitogenic factor, it has been found to promote the expansion and maturation of NK cells from bone marrow (BM) progenitor cells in vitro, in synergy with Flt-3 ligand and IL-15 (5, 6). However, it was reported, in one study, that IL-21 induced proliferation of three different IL-6-dependent human myeloma cell lines, OH-2, ANBL-6, and 1H1 (7). In contrast to IL-21, the role of IL-15 as a mitogenic factor is better established. IL-15 is known to induce the proliferation of a variety of cells and is known as a potent growth factor, particularly for T and B lymphocytes as well as for NK cells (8, 9, 10). Recently, a synergism between IL-21 and IL-15 in regulating CD8+ T cell expansion and function was reported (11). IL-21 was also shown to be a growth and survival factor for human myeloma cells and to display antitumor activity (12), like IL-15 (13, 14). A critical role for IL-21 in regulating Ig production has been recently reported (15). This attests to the importance of IL-21 in general immunology.
IL-15 and IL-21 are known to mediate their biological activities through their respective receptors, IL-15R and IL-21R. These receptors, as well as IL-2R, IL-4R, IL-7R, and IL-9R, share the CD132 component (or γc chain) (1). In addition to this latter component, the IL-15R is composed of a specific IL-15Rα-chain and the IL-2/15Rβ subunit. To date, the only other known component of the IL-21R is IL-21Rα (1). Both cytokines are known to mediate their effects through activation of Jak/STAT pathway, but they also activate other pathways, including the MAPK (1, 2, 3, 4). Although IL-15 is known to activate NF-κB in different experimental conditions (16, 17, 18, 19), the role of this transcription factor in IL-21-induced biological activity is not clear. For example, it was reported in one study that IL-21 did not activate NF-κB in the myeloma OH-2 cell line (7). In another study, it was reported that IL-21 did not increase nuclear accumulation of NF-κB in splenic B cells (26). The importance of NF-κB activation by IL-21 in vivo has never been clearly established.
Although some previous observations had suggested that IL-21 was a proinflammatory cytokine (12, 21, 22), it is only recently that we have established that IL-21 is a proinflammatory cytokine in vivo, based on recruitment of neutrophil and monocyte populations in the murine air pouch model (23). In contrast to LPS, administration of IL-21 into the air pouch did not significantly increase the concentration of IL-6, CCL3 (MIP-1α), CCL5 (RANTES), and CXCL2 (MIP-2). In vitro, IL-21 induced the release of CXCL8 (IL-8) by human monocyte-derived macrophages. Interestingly, we demonstrated that although IL-21 is a not a direct human neutrophil agonist, this cytokine activated human promyelocyte HL-60 cells by inducing phosphorylation of ERK1/2 (23). This result correlates with the presence of IL-21Rα in HL-60 cells and its absence in human neutrophils. Recently, we demonstrated that IL-15 also attracted neutrophils in the same in vivo model (24).
IL-15 is known as a general inhibitor of apoptosis because it suppresses or delays apoptosis in virtually all cells tested. A good example of IL-15’s potent ability to inhibit apoptosis is the study conducted by Bulfone-Paus et al. (25) in which anti-Fas-induced lethal multisystem apoptosis in mice was suppressed by an IL-15-IgG2b fusion protein. In contrast to IL-15, IL-21 possesses both pro- and antiapoptotic properties. Kasaian et al. (26) studied the effect of IL-21 on IL-15 expanded murine splenic cultures restimulated for 2 days with IL-15 or IL-21. IL-21 was found to delay apoptosis caused by removal of IL-15 after 1 day, whereas the majority of NK cells in the culture were apoptotic after 2 days. Others reported that IL-21 induced apoptosis in murine B cells (27).
Although IL-15 is known to activate immature cells isolated from BM or cord blood (16, 28, 29), the role of IL-21 in such cells is not clear. Based on the described differential effects between IL-15 and IL-21, we conducted this study to establish the role of IL-15 and IL-21 on murine BM cell proliferation and apoptosis, as well as to elucidate the role of NF-κB in IL-15 or IL-21-stimulated BM cells. In addition, we investigated the role of each cytokine on two major murine BM leukocyte cell populations, namely the myeloid (CD11b+) and lymphoid (CD11b−) cells.
Materials and Methods
Chemicals, reagents, and Abs
Recombinant murine IL (mIL)-15 was purchased from PeproTech and recombinant mIL-21 was purchased from R&D Systems. The goat anti-mouse IL-15Rα Ab, the goat anti-mouse IL-21R Ab, and the isotypic normal goat IgG were from R&D Systems. The monoclonal anti-actin (clone AC-40) from mouse ascite fluid was from Sigma-Aldrich. The polyclonal rabbit anti-human NF-κBp50 (C20) was from Delta Biolabs and the polyclonal rabbit anti-mouse IκBβ (C-20) was from Santa Cruz Biotechnology. Peroxidase-conjugated rabbit anti-goat IgG and FITC-conjugated rabbit anti-goat IgG were from Jackson ImmunoResearch Laboratories. The PE anti-mouse CD122 (IL-2/15Rβ, clone Tm-b1) Ab was purchased from eBioscience, the corresponding PE-conjugated isotypic control (rat IgG2b), the PE-conjugated rat anti-mouse CD132 (γc chain) mAb, and the corresponding PE-conjugated isotypic control, were from BD Biosciences. Annexin V-FITC conjugate was purchased from BioSource International. [Methyl-3H]thymidine was obtained from ICN Pharmaceuticals. RNase A, LPS, human IgG, propidium iodide, and MTT were purchased from Sigma-Aldrich.
C57BL/6 mice were obtained from Charles River Breeding Laboratories. Mice homozygous for the p50NF-κB1 mutation (B6;129P2-Nfkb1tm1Bal/J) and their littermate controls (B6;129PF2/J) were obtained from The Jackson Laboratory.
BM cell preparation
Mice were killed by CO2 asphyxiation and BM was extracted from both femurs with 5 ml of RPMI 1640 using a 26-gauge needle. BM cells were washed (1200 rpm, 10 min, 4°C) and RBC were removed using a lysis buffer (150 mM NH4Cl, 1 mM KHCO3, and 0.1 mM Na2EDTA; pH 7.2) for 10 min at room temperature. Remaining cells were washed before passage through a pasteur pipette containing nylon wool. Cells were washed twice and viability (>95%) was systematically evaluated before and after each experiment using the trypan blue exclusion assay. Cells (0.5 × 106 cells/ml RPMI 1640-HEPES-penicillin/streptomycin-10% FCS) were incubated with IL-15 (0.1, 1.0, 10, and 100 ng/ml), IL-21 (0.1, 1.0, 10, and 100 ng/ml), or LPS (100 ng/ml, positive control) for 72 h.
Magnetic separation of myeloid (CD11b+) and lymphoid (CD11b−) cells from BM
Positive selection of myeloid cells from mouse BM was performed using CD11b microbeads from Miltenyi Biotec on a MACS LS column according to manufacturer’s instructions. In mouse, the CD11b Ag is expressed on monocytes/macrophages, and to a lesser extent on granulocytes, NK cells, CD5+ B1 cells, and a subset of dendritic cells. Briefly, prepared BM cells were suspended in 90 μl of buffer (degassed PBS (pH 7.2), 0.5% BSA, and 2 mM EDTA) and 10 μl of CD11b microbeads per 107 cells. Cells were mixed and incubated for 15 min at 4°C. Cells were then washed and suspended in 500 μl of buffer before applying cell suspension onto the prepared LS column. Unlabeled fraction (CD11b− lymphoid cells) was collected by washing the column three times with 3 ml of buffer. The column was then removed from the separator, and the magnetically labeled fraction (CD11b+ myeloid cells) was flushed out with 5 ml of buffer by applying the plunger supplied with the column. Cells were washed and viability (>95%) was also evaluated using trypan blue. Purity of each fraction was evaluated by flow cytometry analysis and by microscopy (cytospins stained with Hema-Stain) (23). Cells (0.5 × 106 cells/ml RPMI 1640-HEPES-penicillin/streptomycin-10% FCS) were then incubated with IL-15 (0.1, 1.0, 10, and 100 ng/ml), IL-21 (0.1, 1.0, 10, and 100 ng/ml), or LPS (100 ng/ml, positive control) for 72 h.
Mitochondrial activity evaluation using MTT
MTT (final concentration 0.5 mg/ml) was added into each well for the last 3 h at 37°C. Plates were spun and supernatants were removed. DMSO (200 μl) was added to each well and absorbance was evaluated at 570 nm (reference wavelength at 690 nm) using a plate reader.
Proliferation assay using [3H]thymidine
[3H]Thymidine (1 μCi) was added 24 h before cells were collected onto borosilicate glass fiber paper with a multiple-cell culture harvester (Skatron Instruments). Sections of the filter corresponding to each microwell were then punched out and placed into scintillation counting vials with 4 ml of ScintiSafe Econo 1 (Fisher Scientific) and placed in a beta counter. Results are expressed as stimulation indices (cpm from tested cells/cpm from cells treated with buffer alone).
Assessment of apoptosis by Annexin V
After 72 h of incubation, cells were collected, washed twice with PBS and suspended in 100 μl of binding buffer containing 3 μl of Annexin V-FITC (23). Cells were incubated 15 min at room temperature (light protected) before FACS analysis. Flow cytometry analysis (10,000 events) was performed using a FACScan (BD Biosciences).
Assessment of apoptosis by propidium iodide staining
Cells were harvested and washed twice with PBS. Cells were then fixed in cold 70% ethanol for at least 30 min at 4°C. Cells were washed twice with PBS-2% FCS and treated with 100 μg/ml RNase A. Propidium iodide (50 μg/ml) was added and cells were analyzed by flow cytometry (23). Apoptosis was determined as the percentage of subdiploid (sub-G0/G1) cells from a cell cycle profile.
Cells were lysed in 2× Laemmli’s sample buffer and aliquots corresponding to 0.5 × 106 cells were loaded onto 10% SDS-PAGE and transferred from gel to polyvinylidene difluoride membranes. Nonspecific sites were blocked with 5% nonfat dry milk (Carnation) in TBS-Tween (25 mM Tris-HCl (pH 7.8), 190 mM NaCl, 0.15% Tween 20) for 1 h at room temperature. Membranes were then washed with TBS-Tween 30 and incubated overnight at 4°C with goat anti-mIL-21Rα at 1 μg/ml in TBS-Tween 20. Membranes were then washed with TBS-Tween 20 and incubated for 1 h at room temperature with a rabbit anti-goat HRP secondary Ab (Jackson ImmunoResearch Laboratories) at 1/20,000 in TBS-Tween plus 5% nonfat dry milk, followed by washes. Membranes were revealed with ECL. Membranes were stripped with agitation for 30 min at 65°C with stripping buffer (100 mM 2-ME, 2% SDS, 62.5 mM Tris; pH 6.7) and washed extensively with TBS-Tween 20. Membranes were blocked with either TBS-Tween 20 plus 2% BSA for 1 h at room temperature (anti-actin) or TBS-Tween 20 plus 1% BSA (anti-NF-κBp50 and anti-IκBβ) overnight at 4°C. Membranes were then probed overnight at 4°C with mouse anti-actin (3 μg/ml) in TBS-Tween 20 plus 1% BSA or 1 h at room temperature with rabbit anti-NF-κBp50 (C-20) 1/250 or rabbit anti-IκBβp45 1/500 in TBS-Tween 20 plus 1% BSA. Membranes were then washed with TBS-Tween 20 and incubated for 1 h at room temperature with a goat anti-mouse HRP secondary Ab at 1/20,000 in TBS-Tween 20 plus 2% BSA (anti-actin) or with goat anti-rabbit HRP secondary Ab at 1/20,000 in TBS-Tween 20 plus 5% nonfat dry milk (anti-NF-κBp50 and anti-IκBβ) followed by washes. Membranes were revealed with ECL.
Evaluation of mIL-15Rα, CD122, and CD132 expression by flow cytometry
Cell surface expression of mIL-15Rα, CD122, and CD132 was monitored by indirect and direct fluorescence, respectively. Briefly, cells (5 × 106 cells/ml) were suspended in PBS containing 5 μg/ml human IgG for 30 min at 4°C to block Fc receptors and then washed with PBS. For IL-15Rα, cells were stained in 100 μl of PBS for 30 min at 4°C with 3 μg/ml goat anti-mouse IL-15Rα Ab or 3 μg/ml of the corresponding isotypic control. Cells were washed twice and incubated in 50 μl of PBS containing 10 μg/ml FITC rabbit anti-goat Ab for 30 min at 4°C, washed twice and analysis was performed with a FACScan (BD Biosciences) (23). For CD122 and CD132, cells were treated as described but were incubated with 10 μg/ml PE rabbit anti-rat CD122, CD132, or the corresponding isotypic control.
Statistical analysis was performed with SigmaStat for Windows version 3.0 with either a one-way ANOVA or a two-way ANOVA. Statistical significance was established at p < 0.05.
IL-15, but not IL-21, induced BM cells to proliferate
IL-15 is known to activate proliferation in a variety of cells (1, 8, 10, 16), whereas IL-21 is not well recognized as a direct mitogenic cytokine (5). In this model, we were interested in answering whether these cytokines can induce murine BM cell proliferation. As illustrated in Fig. 1,A, IL-15, but not IL-21, induced cell proliferation, as assessed by [3H]thymidine incorporation, in a concentration-dependent manner. We have also verified the effect of IL-15 and IL-21 at higher concentrations (up to 1000 ng/ml) and observed that IL-15 further increased cell proliferation, whereas IL-21 did not (data not shown). In parallel, we investigated potential effects of IL-15 and IL-21 on mitochondrial activity using the MTT assay. We expected an effect only for IL-15 because this assay is frequently used for measuring cell proliferation. However, as illustrated in Fig. 1 B, both IL-15 and IL-21 significantly increased the response in the MTT assay, suggesting that BM cells expressed functional IL-15R and IL-21R. Also, in two separate experiments, we have treated BM cells with a mixture of 10 ng/ml IL-15 plus 10 ng/ml IL-21 and observed a significant increase of the MTT activity by a factor of 1.5 (data not shown). This observation is in agreement with the synergy effect between IL-15 and IL-21 reported previously (11). Note that LPS, a broad spectrum agonist used as a positive control, also increased both [3H]thymidine incorporation and mitochondrial activity.
The cytokines IL-15 and IL-21 delayed BM cell apoptosis
Because IL-15 is known as a potent antiapoptotic cytokine, and IL-21 increased mitochondrial activity, we investigated the effects of IL-15 and IL-21 on BM cell factor deprivation-induced apoptosis. BM cells were incubated for a period of 3 days without addition of any factors, except FCS and IL-15 or IL-21. As illustrated in Fig. 2, both IL-15 and IL-21 significantly suppressed BM cell apoptosis, as assessed by two different methods, FITC-Annexin V binding assay (Fig. 2,A) and DNA staining with propidium iodide (Fig. 2 B).
IL-15, but not IL-21, induced cell proliferation in myeloid (CD11b+) and lymphoid (CD11b−) BM cell populations
Because we demonstrated that IL-15 induced BM cell proliferation, we next investigated the effect of IL-15 (and IL-21 in parallel as a negative control) on the proliferation of BM cells separated into myeloid (CD11b+) and lymphoid (CD11b−) cell populations (Fig. 3, A and B). As illustrated in Fig. 3, IL-15, but not IL-21, induced the proliferation of both myeloid (Fig. 3,C) and lymphoid (Fig. 3,D) cell populations. Of note, IL-15 was more effective on cells of lymphoid origin than cells of myeloid origin because cell proliferation was increased by ∼11-fold vs ∼4-fold, respectively. Both cytokines increased the MTT assay response in a concentration-dependent manner (Fig. 3, E and F), but only IL-15 significantly induced mitochondrial activity in BM cells of lymphoid origin.
IL-15 and IL-21 delayed apoptosis, especially in BM cells of lymphoid origin
Because we previously demonstrated that IL-15 and IL-21 modulated apoptosis of BM cells, we next investigated the role of these two cytokines in delaying apoptosis of CD11b− and CD11b+ BM cells. As illustrated in Fig. 4, both IL-15 and IL-21 significantly delayed apoptosis in BM lymphoid cells, but neither cytokine significantly delayed apoptosis in BM myeloid cells.
Expression of IL-15 and IL-21 receptors in BM cells
We next verified the expression of IL-15Rα, CD122, CD132, and IL-21Rα in total CD11b− and CD11b+ BM cells to confirm whether a lack of biological activity in response to these cytokines was related to absence of cytokine receptor expression. As illustrated in Fig. 5, BM cells expressed IL-15Rα, CD122, and CD132 in both CD11b+ and CD11b− cells, as assessed by flow cytometry. Because the anti-IL-21Rα Ab was inefficient in flow cytometry (data not shown), we performed Western blot experiments and found CD11b− cells, but not CD11b+, expressed IL-21Rα (Fig. 5 D).
Involvement of NF-κBp50 in IL-15-induced BM cell proliferation
Because IL-15 is known to activate NF-κB for mediation of cell signaling (16, 17, 18, 19), we investigated the potential role of this transcription factor in IL-15-stimulated BM cells isolated from mice deficient in NF-κBp50, using LPS as a positive control (30, 31, 32). We also investigated the role of this transcription factor in IL-21-stimulated BM cells because NF-κB activation by IL-21 is poorly documented. As illustrated in Fig. 6, A–C, mitochondrial activity was increased by IL-15 and IL-21 in BM cells, whether cells were isolated from NF-κBp50−/− knockout (KO) mice or NF-κBp50+/+ wild-type (WT) mice. The response was not significantly decreased in KO vs WT mice. Mitochondrial activity was increased by LPS in both WT and KO mice, but this increase was significantly reduced only in KO mice. The ability of IL-15 and LPS to induce BM cell proliferation was significantly decreased in KO vs WT mice (Fig. 6, D and E), but not totally reversed. As expected, IL-21 did not modulate BM cell proliferation in both WT and KO mice.
Role of NF-κBp50 in IL-15- or IL-21-induced suppression of BM cell apoptosis
We next investigated the role of NF-κB in IL-15-induced suppression of BM cell apoptosis. Interestingly, the basal level of BM cell factor deprivation-induced apoptosis was decreased in KO mice vs WT, suggesting that BM cells from mice lacking NF-κBp50 are more refractory, and therefore, less likely to undergo apoptosis (Fig. 7, A and B). The antiapoptotic effect of IL-15 and IL-21 was slightly reversed in KO mice, but markedly reversed in LPS-induced BM cells. The results differ slightly, depending on the assay used for evaluating BM cell apoptosis. IL-15 and IL-21 significantly delayed BM cell apoptosis in KO mice as assessed by propidium iodide staining, whereas the slight decrease observed when using FITC-Annexin V was not significant. In contrast, the antiapoptotic effect of LPS was markedly reversed in KO mice (Fig. 7, A and B). The differences observed between WT and KO mice cannot be attributed to the expression of receptors because no differences were observed between both groups (Fig. 7 C).
This is the first study investigating the biological activity of IL-15 and IL-21 on murine BM cells by focusing on 1) proliferation and apoptosis in unfractionated cells, as well as in myeloid (CD11b+) and lymphoid (CD11b−) cell populations and 2) the role of NF-κB in BM cell activation. We found that IL-15, but not IL-21, induced total BM cell proliferation. This biological activity of IL-15 is in agreement with previous studies, which have reported that IL-15 is itself mitogenic (8, 9, 10, 33). Although IL-21 was recently found to induce proliferation in IL-6-dependent human myeloma cell lines (7) and in adult-T cell leukemia cells (34), we did not observe any mitogenic effect of IL-21 on murine BM cells. In humans, IL-21 was also reported as a nonmitogenic cytokine in isolated B cells (5). However, although IL-21 is no mitogenic by itself, this cytokine is known to exert critical roles in regulating Ig production (15). A synergism between IL-21 and IL-15 has been recently reported in regulating CD8+ T cell expansion and functions (11). We have also observed a synergy between IL-15 and IL-21 when BM cells were treated with a mixture of IL-15 and IL-21, suggesting that BM cells respond like mature cells.
Because IL-21 increased mitochondrial activity, this suggests that some BM cells express functional IL-21R. The presence of IL-21R was previously detected in C57BL/6 BM cells (34). A recent study from Jin et al. (36) reported that IL-21R was expressed on a variety of murine cell lines, including some B and T cell lines. IL-21Rα was not expressed in three different pre-B cell lines, in macrophages (tested in the P338D1 cell line), and in fibroblasts. In BM, it was demonstrated that IL-21R was negligibly expressed on CD43highIgM−B220low pro-B cells, weakly expressed on CD43intIgM−B220low pre-B cells, but highly expressed on most immature B220highIgMlow and all mature B220highIgMhigh B cells (36). Interestingly, when we fractionated total BM cells into myeloid and lymphoid cell populations, we found that IL-21Rα was expressed in total and in lymphoid, but not in myeloid BM cells. This concurs with the inability of IL-21 to stimulate myeloid cells. Unlike IL-21Rα, IL-15Rα, CD122, and CD132 are expressed in both lymphoid and myeloid BM cells. These observations are in agreement with our previous findings, which demonstrated that IL-15 (24, 34), but not IL-21 (23) was a neutrophil (CD11b+) agonist. Of note, in the present study, the majority of BM CD11b+ cells exhibit a neutrophil phenotype (Fig. 3).
Our results demonstrate that IL-21 exerts its biological activity, namely the capacity to increase mitochondrial activity and to delay apoptosis, in the lymphoid but not myeloid BM cell populations. This finding is also true for IL-15. However, the myeloid cell population was also induced to proliferate in response to IL-15, albeit with less potency than the lymphoid cell population. Even if at first sight our results suggest that the effect of IL-15 is marginal on CD11b− cells, the fact that cells expressing IL-15Rα can trans-present IL-15 to other cells expressing only CD122 and CD132 components (37) suggests that IL-15 can mediate potent biological activities in different BM cells. Although we did not separate the BM lymphoid cells into different subset populations, our results are in agreement with previous data, demonstrating that IL-21 preferentially acts on lymphoid B cells (15, 36). Interestingly, in their study, Jin et al. (36) demonstrated that very few B220− cells in the BM expressed IL-21R. Knowing that, it is tempting to speculate that IL-21 acts principally on lymphoid B cell subsets; however, this needs further investigation.
To the best of our knowledge, this is the first study to report that IL-21 can itself delay murine BM cell apoptosis. Several studies have reported that IL-21 acts as a coactivator. IL-21 was shown to costimulate human B cell proliferation induced by anti-CD40 (5). In contrast, IL-21 was found to suppress B cell proliferation induced by anti-IgM and IL-4 (36). Recently, it was reported in mice that IL-21 did not costimulate anti-CD40-induced B cell proliferation, but rather induced apoptosis (27). At contrast in humans, IL-21 inhibited rather than enhanced the IL-15-induced expansion of resting and activated NK cells (26, 27). In this model, we found that, on its own, IL-21 did not induce BM cell proliferation, but acted as an antiapoptotic molecule in lymphoid CD11b− cells expressing IL-21Rα.
Both IL-15 and IL-21, like many other class I cytokines, are known to activate the Jak-STAT pathway for mediation of their biological actions (1, 2, 3, 4). In particular, IL-21 activates Jak1, Jak3, STAT1, STAT3, and STAT5 in different human T cell and B cell lines and in NK cells (1, 2, 3, 4). In addition to the Jak-STAT pathway, IL-21 was found to activate the MAPK pathway because it induced phosphorylation of Erk1/2 (p44/42 MAPK), at least in myeloma cells (7). Recently, we reported that IL-21 also induced phosphorylation of ERK1/2 in human promyelocytic HL-60 cells expressing IL-21Rα, but not in IL-21Rα negative neutrophil cells (23). IL-15 is particularly known to activate Jak1/3-STAT3/5 in a variety of cells, including B and T cells and NK cells (1, 16, 38). In human neutrophils, IL-15 induced phosphorylation of Jak2, p38, and ERK1/2, but not STAT5 (39). Although IL-15 is known to activate NF-κB under several experimental conditions, involvement of this transcription factor in IL-21-activated cells was not clear before the present study. We decided to investigate the role of NF-κB in IL-15- and IL-21-induced BM cells by using mice deficient in NF-κBp50. NF-κB is composed of homo- and heterodimers of Rel family proteins including p65, RelB, c-Rel, p52, and p50, and targeted disruption of the p50 subunit in mice caused multifocal defects in the immune response (40). These mice showed no developmental abnormalities, but interestingly, their B cells did not proliferate in response to LPS. According to Sha et al. (40), there was no evidence that p50-deficient B cells expressed larger amounts of Rel, RelB, p65, and p52. These results imply that compensation by other members does not occur in B cells (40). Recently, using EMSA, an absence of NF-κB activation was demonstrated in myocardium of p50 KO mice (41). Our results show that the p50 component of the NF-κB transcription factor is implicated in IL-15-induced BM cell proliferation. However, we do not exclude the possibility that other components (p65, RelB, c-Rel, and p52) may also be involved. In future it would be interesting to verify the state of activation of all the components.
In this study, although we focused our study on the two proinflammatory cytokines IL-15 and IL-21, we used LPS throughout our experiments because of the known importance of NF-κB in LPS-induced biological actions (30, 31, 32). We confirm the importance of this transcription factor for LPS-induced signaling in BM cells because the ability of LPS to induce mitochondrial activity and proliferation was significantly diminished in NF-κBp50-deficient mice. Although several studies have been performed with NF-κBp50 KO mice (42, 43, 44), this is the first study to investigate the role of cytokines in BM cell proliferation and apoptosis. As expected, and unlike IL-15, NF-κBp50 is not involved in IL-21-induced BM cell proliferation, correlating with the fact that IL-21 did not stimulate BM cell proliferation in normal C57BL/6 cells. This finding is in agreement with previous studies reporting that IL-21 did not activate NF-κB (7, 20). However, we have observed that IL-21 was able to increase moderately mitochondrial activity in NF-κBp50-deficient BM cells. This is probably due to an additive effect, obtained from both myeloid and lymphoid cell populations because IL-21 slightly, but not significantly, increased the MTT assay in both BM cell populations (Fig. 3). The MTT assay is known as a good indicator of cell proliferation, based on the ability of a mitochondrial dehydrogenase enzyme to cleave the tetrazolium rings, leading to the formation of dark blue formazan crystals that are measured by colorimetry (45). This assay is also recognized as a cell viability assay because of the enzyme activity observed in viable cells. Knowing that apoptotic cells exclude trypan blue, we explain the increase in MTT activity by IL-21 as a consequence of delayed BM cell apoptosis rather than its ability to induce cell proliferation, which indirectly, increases cell viability.
In contrast to IL-21, the role of NF-κB in IL-15-induced biological activities has been studied previously in a variety of in vitro conditions. For example, using a gel EMSA, NF-κB was found to be one of the major signal molecules to mediate IL-15-induced cyclooxygenase-2 up-regulation in rheumatoid synoviocytes (46). IL-15 was also found to inhibit spontaneous apoptosis in human eosinophils via autocrine production of GM-CSF and NF-κB activation (17). In addition, IL-15 has been found to induce NF-κB activation and IL-8 production in human neutrophils (19). Interestingly, NF-κB was reported to be activated in human BM and cord blood progenitor cells via constitutive endogenous secreted IL-15 from CD34+ cells (16). Despite these observations, there has been no previous study investigating the role of NF-κB in IL-15-stimulated murine BM cells. The use of these mice permitted the definitive demonstration that NF-κB is at least partially involved in IL-15-induced BM cells because IL-15 can still mediate its biological activity in these cells even in the absence of NF-κBp50.
Collectively, our results are the first to demonstrate that the two proinflammatory cytokines IL-15 and IL-21 can alter murine BM cell physiology. These cytokines preferentially act on BM lymphoid (CD11b−) rather than myeloid (CD11b+) cell populations. Both cytokines can mediate some biological activities in BM cells isolated from NF-κBp50 mice. Although BM cells can accumulate in the presence of IL-15 or IL-21, the mechanisms involved are distinct between both cytokines. IL-15 can simultaneously induce cell proliferation and delay apoptosis, whereas IL-21 cannot induce BM cell proliferation, but acts by suppressing BM cell apoptosis. This study provides the first evidence that presence of IL-15 and/or IL-21 in BM can participate in increasing the number of total BM cells, especially of the CD11b− lymphoid phenotype.
We thank Mary Gregory for reading this manuscript.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This study was supported in part by Canadian Institutes of Health Research (CIHR) Grants MOP-89534 and MOP-144416. M.P. holds a PhD CIHR Award, and D.G. is a Scholar from Fonds de la Recherche en Santé, Québec.
Abbreviations used in this paper: γc, common γ-chain; BM, bone marrow; mIL, murine IL; KO, knockout; WT, wild type.