IL-21 is a member of the type I cytokine family related most closely to IL-2 and IL-15. IL-21 is a pleiotropic cytokine, produced by T, NKT, and dendritic cells, which modulates lymphoid and myeloid cell functions. Besides its activities on normal lymphoid cells, it has been shown that IL-21 is a growth factor for myeloma cells. In the present study, we demonstrate that IL-21 generated myeloma colonies from 9 of 24 human myeloma cell lines (HMCL) in a collagen-based assay. Of major interest, the capacity of IL-21 to stimulate clonogenicity was restricted to CD45− HMCL. We found that IL-21 induced tyrosine phosphorylation of STAT-3, STAT-1, and Erk1/2. Interestingly, an Akt activation was observed lately after 30 min to 1 h of IL-21 stimulation, indicating that this Akt phosphorylation could be due to an IGF-1 autocrine loop. This hypothesis was sustained both by the fact that IL-21 treatment induced an IGF-1 mRNA synthesis and that an antagonistic anti-IGF-1 receptor mAb (AVE1642) strongly inhibits the IL-21-induced clonogenicity. Thus, we demonstrated by quantitative PCR that IL-21 induced clonogenicity through an autocrine IGF-1 secretion in HMCL and primary myeloma cells. Because we have previously demonstrated that CD45 phosphatase inhibits the IGF-1 signaling, this inhibitory effect of CD45 explains why the IL-21-induced clonogenicity was restricted to CD45− HMCL. These results support that therapy against IGF-1R, which are presently under investigation in multiple myeloma, could be beneficial, not only to suppress IGF-1-mediated myeloma cell growth, but also IL-21-mediated myeloma cell growth.
Multiple myeloma (MM)2 is a fatal neoplasia characterized by the accumulation of malignant long-lived plasma cells. MM presents as a heterogeneous disease with very different clinical outcome. Differences in survival are linked to biological heterogeneity including morphology, phenotype, and genotype (1). Among the different prognostic markers, CD45 expression appears as a powerful one (2). Indeed, CD45− patients have a very poor survival compared with CD45+ patients. This phenotypic heterogeneity can be explained by the fact that CD45 is essential to discriminate signaling and proliferation of human myeloma cells in response to either IL-6 or IGF-1, the two major growth factors for MM cells. Thus, the CD45 phosphatase appears as a positive regulator of IL-6 response by controlling the lyn src activity and a negative regulator of IGF-1 response (3, 4, 5, 6). Consequently, CD45+ and CD45− human myeloma cell lines (HMCL) proliferate primarily in response to IL-6 and IGF-1, respectively. In a recent study, we demonstrated that IGF-1 stimulates the clonogenicity of CD45− HMCL only (7). Beside the cytokines of the IL-6 family and IGF-1, it was recently shown that IL-21 is a growth and survival factor for MM cells (8). IL-21 is produced by activated CD4+ T and NKT cells and its receptor is expressed on T, B, NK, and dendritic cells (9, 10, 11). IL-21R contains the common cytokine-receptor γ-chain, which is also shared by the receptors for IL-2, IL-4, IL-7, IL-9, and IL-15 (12). IL-21R has highest homology to the IL-2Rβ-chain, which is a component of IL-2 and IL-15 receptors (13). IL-21 has a wide range of effect and modulate the function of T, B, and NK cells (9). On B cells, the effects of IL-21 are different either proapoptotic on naive B cells or promoting their differentiation into plasma cells after costimulation with signal from T cells (14). Indeed, IL-21 induces the differentiation of B cells into plasma cells through up-regulation of BLIMP1 (15). Moreover, a recent study demonstrated the requirement of IL-21 in B cell activation expansion and plasma cell differentiation induced by direct interaction of B cells with activated T cells (16). The role of IL-21 in regulating Ig production is supported by knockout mouse models. IL-21R−/− mice have diminished IgG1, but elevated IgE level in response to Ag (17).
Because a recent report indicates that IL-21 is a growth and survival factor for MM, we currently evaluated the capacity of IL-21 to induce myeloma colonies in a collagen-based assay. We show that the capacity of IL-21 to generate myeloma colonies is restricted to CD45− HMCL. Furthermore, we demonstrate that IL-21 essentially acts through an autocrine IGF-1 secretion loop.
Materials and Methods
Human myeloma cell lines and culture conditions
LP-1, L363, NCI-H929, and OPM-2 HMCL were purchased from DSMZ and RPMI 8226 and U266 were purchased from the American Type Culture Collection. JIM-3 and JJN-3 were provided by Prof. L. Bergsagel (Mayo Clinic, Scottsdale, AZ) and Prof. B. Van Camp (Brussels, Belgium), respectively. MM.1S and Karpas 620 were a gift from Dr. S. T. Rosen (Northwestern University, Chicago, IL) and Dr. A. Karpas (Cambridge University, Cambridge, U.K.), respectively. AMO1 was provided by Dr. Shimizu (Kanazawa Medical University, Ishikawa). KMS-11, KMS-18 and KMS-12-PE were provided by Dr. T. Ohtsuki (Kawasaki Medical School, Okayama, Japan). The XG-1, XG-2, XG-5, XG-6, XG-7, NAN-1, NAN-3, NAN-6, SBN, and BCN HMCLs have been previously established in our laboratory from peripheral blood samples or pleural effusion of patients with MM (1) and are cultured in the presence of 3 ng/ml of r-IL-6 (Novartis). Cell lines were maintained in RPMI 1640 medium supplemented with 5% FCS, 2 mM glutamine, antibiotics, and 5 × 10−5M 2-βME.
Reagents and Abs
The antagonistic Ab anti-IGF-R (AVE1642) was provided by sanofi-aventis (6
Cells (0.5 × 106
Myeloma cell colony-forming assay
Myeloma cells (103 cells) were plated in 1 ml serum-free, cytokine-free, collagen-based semisolid medium (stemαIII, StemAlpha SA) in triplicates (330 μl/well) in 4-well plates and grown for 15 days. The gels were harvested on glass slides, dried and stained with May-Grunwald-Giemsa. Colonies were counted on triplicate gels. The number of colonies was expressed as average per 103 cells.
Quantitative real-time PCR
Cells (4 × 106) were resuspended in 0.5% NP40 lysis buffer (10 mM Tris-HCl (pH 7.6), 150 mM NaCl, 5 mM EDTA, and protease inhibitors). After 40 min on ice, lysates were cleared by centrifugation at 12000 × g for 30 min at 4°C. Protein concentration was measured using bicinchoninic acid protein assay (Pierce). One hundred micrograms of proteins were loaded for each lane. The proteins were separated by 10% SDS-PAGE and then electrotransfered to PVDF membranes. Western blot analysis was performed by standard techniques with ECL detection (Roche).
IL-21 is a clonogenic factor for 9 out of 24 HMCL being all CD45− HMCL
We evaluated the capacity of IL-21 to induce myeloma colonies in a collagen-based assay as previously described (7). We investigated the effect of IL-21 on 24 HMCL and we showed that IL-21 generated myeloma colonies in 9 out of 24 HMCL. The clonogenicity ranged from 1 to 26% (Fig. 1⇓). Of major interest, the capacity of IL-21 to stimulate clonogenicity was exclusively restricted to CD45− HMCL; indeed, IL-21 stimulated clonogenicity of 9 of 17 CD45− HMCL vs 0/7 CD45+ HMCL (Fisher exact test p < 0.022). We next examined whether the absence of clonogenicity could be related to an absence of IL-21R expression. The expression of IL-21R transcripts was examined by both quantitative real-time PCR analysis and confirmed by flow cytometry. Most of the HMCL including both CD45− and CD45+ HMCL expressed IL-21R (19 out of 24 HMCL) as shown in Fig. 2⇓A. It is worth noting that the level of expression of IL-21R is weak for most of the HMCL (as illustrated in Fig. 2⇓C for LP1), except for XG-5, which expressed very high level of IL-21R detected either by both methods (Fig. 2⇓, A and C). Thus, the level of IL-21R is not directly correlated to the IL-21 induced clonogenicity. To investigate the biological significance of the IL-21/IL21R system in MM patients we examined the expression of IL-21R transcripts by quantitative real-time PCR in primary myeloma cells. We showed that 9 out of 18 purified CD138+ myeloma cells express IL-21R transcripts (Fig. 2⇓B), demonstrating that IL-21/IL-21R could be potentially important in MM development.
IL-21 activates first STAT-3, STAT-1, and Erk then lately Akt in CD45− HMCL
It was previously shown that IL-21 mediated its effects through STAT-3 and Erk1/2 activation in myeloma cells (8). We found that IL-21 induced tyrosine phosphorylation of STAT-3, STAT-1, and Erk1/2 and thus only in HMCL able to clone in the presence of IL-21 as shown for LP-1, L363, and KMS-12-PE HMCL (Fig. 3⇓). Indeed, IL-21 neither induced STAT-3, STAT-1, or Erk1/2 phosphorylation in JIM-3, a HMCL that did not clone in the presence of IL-21. Of note, we have previously shown that IL-6 induced the clonogenicity of these cell lines (7) and, as expected, IL-6 induced STAT-3 and Erk 1/2 activation in all HMCL tested (Fig. 3⇓). We next examined the kinetics of activation of these proteins in the presence of IL-21. IL-21 induced a rapid (after a 5-min stimulation), strong, and sustained STAT-3 phosphorylation up to 3 h, whereas the rapid STAT-1 activation was more transient and started to decrease after 1-h stimulation. Surprisingly, an Akt activation was observed after 30 min to 1 h of IL-21 stimulation (Fig. 4⇓). Thus, we hypothesized that Akt phosphorylation could be due to an autocrine loop secondary to IL-21 stimulation in CD45− HMCL.
IL-21 response corresponds to an IGF-1 autocrine loop in CD45− HMCL
IGF-1 is a well-known growth factor for HMCL which induced a strong Akt phosphorylation, but the response to IGF-1 is related to the presence or not to the CD45 phosphatase (5). Thus, we hypothesized that IL-21 could induce an IGF-1 autocrine loop in CD45− HMCL leading to Akt phosphorylation. Also, we first analyzed the effects of the antagonistic mAb anti-IGF-1R (AVE1642) on the clonogenicity induced by IL-21. This Ab triggers a total inhibition of the IL-21-induced clonogenicity of LP-1, KMS-12-PE, Karpas 620, NCI-H929, KMS-11, and RPMI 8226 and a partial inhibition of that of L363 (44%) (Fig. 5⇓A). These results indicate that an autocrine IGF-1 loop was either totally or partially responsible for the IL-21-observed clonogenicity. According to this concept, we observed that AVE1642 mAb induced a total inhibition of Akt phosphorylation induced by IL-21 as shown for LP-1 and KMS-12-PE (Fig. 5⇓B) whereas it had no effect on Stat-1, 3, and Erk1/2 phosphorylation. These results indicate that the Akt phosphorylation is necessary to the clonogenicity induced by IL-21. We confirm this hypothesis showing that wortmaninn totally inhibits the IL-21-induced clonogenicity (Fig. 5⇓C). Furthermore, quantitative PCR analysis was performed showing that a 3-h IL-21 treatment triggered an IGF-1 mRNA level increase in all the cell lines tested (n = 6) and in a MM patient, confirming the hypothesis of IGF-1 autocrine loop (Fig. 5⇓D).
IL-21 and IGF-1 have an additive effect on clonogenicity
We have previously shown that IGF-1 induced clonogenicity of CD45− HMCL only (7). We show in this study that all the HMCL that cloned with IL-21 were also able to clone with IGF-1 (Fig. 6⇓A). Whereas IGF-1 is usually a more powerful clonogenic factor than IL-21, it induced fewer colonies than IL-21 in L363. We next analyzed the effects of IL-21 in combination with IGF-1. We found that this combination triggered an additive effect on clonogenicity in three of four HMCL tested (Fig. 6⇓A). When the signal transduction was analyzed (Fig. 6⇓B), we observed an increase of the phosphorylation of Akt and Erk with the combination of IL-21 and IGF-1 compared with either IL-21 or IGF-1 alone, which can explain this additive effect on clonogenicity.
Numerous studies have documented that IL-21 stimulates B cell differentiation and IgG production, and a recent study demonstrated that IL-21 is required in B cell activation expansion and plasma cell generation during CD4+ T cell-B cell interaction (15, 16, 17). Although, IL-21 has been shown to be a growth and survival factor in MM several years ago, no subsequent study has been devoted to the biological significance of the IL-21/IL-21R system in MM. In the present study, we have demonstrated that a majority of HMCL expressed IL-21R, and this expression is also found in 50% of primary myeloma cells. In contrast to the expression of IL-21R on malignant plasma cells, it was recently shown that normal plasma cells isolated from the spleen lacked expression of IL-21R (18). Thus, sustained expression of IL-21R on malignant plasma cells supports the notion that sustained IL-21 signaling may be involved in the development of MM either by increasing proliferation or giving a survival advantage to the malignant cells. To evaluate the biological significance of IL-21R expression, we examined the capacity of IL-21 to induce self renew of isolated myeloma cells in a serum-free collagen-based assay. In the absence of cytokines, most of the HMCL do not retain the capacity to self renew (with the exception of RPMI 8226, KMS-18, and AMO1, and to a less extend NCI-H929 and JJN-3). Using a large panel of HMCL, we show that IL-21 generates myeloma colonies in 37% of the cases, indicating that IL-21 is an important clonogenic factor for MM. Indeed, we have previously shown that besides IL-6 and IGF-1, which are very potent clonogenic factors for myeloma cells, the other documented growth factors for MM as fibroblast growth factor, hepatocyte growth factor, and heparin-binding-epidermal-like growth factor were very poor clonogenic factors (7). IL-21, as IGF-1, but in contrast to IL-6, generates myeloma colonies depending on their CD45 phenotype since only CD45− HMCL cloned with IL-21.
We searched for a correlation between clonogenicity and signaling pathway activated by IL-21, showing that only HMCL able to clone are also able to activate all the downstream molecules of IL-21 signaling pathways. Similar to IL-2, IL-4, IL-7, and IL-15 it has been shown that IL-21 activates JAK1 and JAK3 with JAK1 binding IL-21R (13) and JAK3 binding γc (19). According to JAK1 and JAK3 activation, we found that IL-21 induced activation of both STAT-1 and STAT-3 in HMCL but not of STAT-5 (result not shown). IL-21 is thus distinct from IL-2, IL-7, and IL-15, which mainly activate STAT-5 (19). Although, the activation of STAT-1 by IL-21 was not previously detected by Brenne et al. (8) in myeloma cells, we found it in the current study that is in agreement with numerous other studies on other cell types (12, 20). Kinetics study indicates that Stat-1 activation is more transient than Stat-3 activation showing that STAT-3 might play a greater role in IL-21 induced clonogenicity. This result is consistent with the notion that STAT-1 is a negative regulator of cell growth and survival (21) whereas Stat-3 is a positive regulator of survival and promotes oncogenesis (22). In addition to the JAK/STAT pathway, certain γc dependent cytokines can activate MAPK and PI3K/Akt pathways (20). In myeloma cells, we found that IL-21 triggers a rapid Erk 1/2 phosphorylation, a pathway known to contribute to proliferation. Because IL-21 activates strongly STAT-3 but also Erk1/2, we show an important similitude in the downstream signaling molecule activated by IL-21 and IL-6. Finally, we found that IL-21 induced a delayed Akt phosphorylation that starts 30 min to 1 h after IL-21 stimulation. Thus, in CD45− HMCL, IL-21-mediated clonogenicity required cooperative effects of the three pathways: JAK/STAT, Erk/MAPK, and Akt/PI3K pathways. We can assume that the JAK/STAT pathway is involved in survival effect whereas both Erk/MAPK and Akt/PI3K pathways are involved in the proliferation effect. Because in myeloma cells, IGF-1 is the major growth factor known to induce a strong Akt phosphorylation and since the effects of IGF-1 as those of IL-21, are restricted to CD45− HMCL, we have investigated the effect of an antagonistic anti-IGF-1R Ab (AVE1642) on the IL-21 induced clonogenicity. AVE1642 induced a strong inhibition of the IL-21-induced clonogenicity indicating that an autocrine IGF-1 loop was involved in the IL-21 clonogenicity sustained by the result of quantitative PCR. Thus, this result explains that the clonogenicity of IL-21 was restricted to CD45− HMCL as it was already demonstrated for IGF-1 induced clonogenicity (7). Indeed, we have previously demonstrated that the presence of CD45 phosphatase inhibits IGF-1 signaling by a direct interaction between CD45 and IGF-1R, which probably dephosphorylates the IGF-1R and results in an inhibition of IGF-1 signaling (5). Thus, in CD45+ HMCL, the clonogenic effect of autocrine IGF-1 loop induced by IL-21 was inhibited by the presence of the CD45 phosphatase. Although, the proliferative effects of IL-21 seem to be due to an autocrine IGF-1 loop, we still observed an additive effect of IL-21 and IGF-1 on clonogenicity. We can hypothesize that Stat-3 phosphorylation induced by IL-21 will give a survival advantage which finally result in increased clonogenicity after exposure to both IL-21 and IGF-1. Moreover, we observed that Akt phosphorylation induced by IL-21 can be increased by the combination of IL-21 and IGF-1 explaining in part the additive effect of the IL-21/IGF-1 combination on clonogenicity. In conclusion, our data confirm that IL-21 is a potent MM cell growth factor as previously published (8). However, our results demonstrate that IL-21 is not a really new independent growth factor for MM but that it is dependent on IGF-1 to act, reinforcing the major role of IGF-1 as a growth factor of MM. These results support the notion that therapy against IGF-1R, which is presently under investigation in MM, could be beneficial, not only to suppress IGF-1-mediated MM cell growth, but also IL-21-mediated MM cell growth.
This work was supported by The Ligue Nationale contre le Cancer (équipe labelisée 2008) and by Institut National du Cancer (PL06_070).
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.
↵1 Address correspondence and reprint requests to Dr. Martine Amiot, Unité Mixte de Recherche 892, Département de Recherche en Cancérologie, 9, quai Moncousu, Nantes, France. E-mail address:
↵2 Abbreviations used in this paper: MM, multiple myeloma; HMCL, human myeloma cell line.
- Received May 23, 2008.
- Accepted September 20, 2008.
- Copyright © 2008 by The American Association of Immunologists