Original Manuscript | Published:

Adherence of multiple myeloma cells to bone marrow stromal cells upregulates vascular endothelial growth factor secretion: therapeutic applications

Leukemia volume 15, pages 19501961 (2001) | Download Citation

Subjects

Abstract

Increased angiogenesis has recently been recognized in active multiple myeloma (MM). Since vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF) are two key mediators of angiogenesis, we characterized the production of VEGF, b-FGF and interleukin-6 (IL-6) (a MM growth and survival factor) in MM cell lines and Epstein–Barr virus (EBV) transformed B cell lines from MM patients, patient MM cells, as well as bone marrow stromal cells (BMSCs) from normal healthy donors and MM patients. We detected secretion of VEGF, but no bFGF and IL-6, in MM cell lines (MM.1S, RPMI 8226 and U266); EBV transformed B cell lines from MM patients (IM-9, HS-Sultan and ARH77); MM cell lines resistant to doxorubicin (RPMI-DOX40), mitoxantrone (RPMI-MR20), melphalan (RPMI-LR5) and dexamethasone (MM.1R); and patient MM cells (MM1 and MM2). BMSCs from MM patients and normal donors secreted VEGF, b-FGF and IL-6. Importantly, when MM cells were adhered to BMSCs, there was a significant increase in VEGF (1.5- to 3.1-fold) and IL-6 (1.9- to 56-fold) secretion. In contrast, the bFGF decreased in co-cultures of BMSCs and MM cells. Paraformaldehyde fixation of BMSCs or MM cells prior to adhesion revealed that VEGF was produced both from BMSCs and MM cells, though it may come primarily from BMSCs in some cultures. IL-6 was produced exclusively in BMSCs, rather than MM cells. Moreover, when MM cells were placed in Transwell insert chambers to allow their juxtaposition to BMSCs without cell to cell contact, induction of VEGF and IL-6 secretion persisted, suggesting the importance of humoral factors. Addition of exogenous IL-6 (10 ng/ml) increased VEGF secretion by BMSCs. Conversely, VEGF (100 ng/ml) significantly increased IL-6 secretion by BMSCs. Moreover, anti-human VEGF (1 μg/ml) and anti-human IL-6 (10 μg/ml) neutralizing antibodies reduced IL-6 and VEGF secretion, respectively, in cultures of BMSCs alone and co-cultures of BMSCs and MM cells. Finally, thalidomide (100 μM) and its immunomodulatory analog IMiD1-CC4047 (1 μM) decreased the upregulation of IL-6 and VEGF secretion in cultures of BMSCs, MM cells and co-cultures of BMSCs with MM cells. These data demonstrate the importance of stromal–MM cell interactions in regulating VEGF and IL-6 secretion, and suggest additional mechanisms whereby thalidomide and IMiD1-CC4047 act against MM cells in the BM millieu.

Introduction

MM is a B cell neoplasm associated with clonal proliferation of malignant plasma cells. It is characterized by infiltration of BM with tumor cells, cortical and medullary bone destruction, and suppression of hematopoiesis. In the year 2001, MM will be diagnosed in 13700 patients in the United States and account for 20% deaths from hematological malignancies.1 The disease course is variable but unfortunately remains fatal in the majority of cases despite advances in high-dose chemotherapy and transplantation.

We2 and others3,4 have demonstrated that IL-6 is a major growth and survival factor for MM cells. IL-6 is secreted by BMSCs and some MM cells and is markedly upregulated by adhesion of MM cells to BMSCs.5 IL-6 mediates both MM cell proliferation6 and inhibits Fas and dexamethasone induced apoptosis.7 The observation that IL-6 levels are usually high in patients with active MM,8 coupled with the decrease in MM cells after administration of anti-IL-6 monoclonal antibodies,9 suggests its in vivo significance. Thus IL-6 appears to be an important cytokine involved in the maintenance and progression of MM.10

Many malignant neoplasms are dependent on neovascularization to sustain growth.11 Various studies have demonstrated that the number and density of microvessels in solid tumors including breast,12 prostate13 and non small cell lung cancers,14 as well as neuroblastomas,15 directly correlates with growth, invasion, and dissemination. Increased angiogenesis often predicts poor prognosis. Angiogenic cytokines such as bFGF and VEGF are important mediators of angiogenesis.16 Overexpression of VEGF is associated with increased angiogenesis, growth, and metastasis in solid tumors. Various human cancers, including lung,17 gastrointestinal18 and prostate19 cancers, secrete VEGF, and in solid cancers, VEGF mRNA and protein expression has been correlated with advancing disease.17 Elevated levels of bFGF have been reported in serum20 and urine21 of patients with cancer.

Although the importance of angiogenesis in solid tumors is well known, its role in growth and survival of hematological malignancies has only recently been realized. Ribatti et al22 reported an increase in microvessel density in lymph nodes in B cell non-Hodgkin's lymphoma (NHL), which correlated with the severity of disease.22 Fielder et al23 reported transcription of VEGF in a high proportion of patients with acute myeloid leukemia (AML). Increased angiogenesis and higher levels of VEGF, bFGF, and hepatocyte growth factor (HGF) have been reported in AML, chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML) and myelodysplastic syndromes (MDS).24 In contrast, low intracellular VEGF protein in AML cells correlated with higher complete response rates and longer survival.25

In MM, increased angiogenesis26,27,28 and elevated mast cell density26 is highly correlated with plasma cell labelling index (LI).26,27 This increased angiogenesis persists even after achievement of complete response following stem cell transplantation, suggesting the utility of combined chemotherapeutic and antiangiogenic drug regimens.29 Elevated serum levels of angiogenic cytokines such as VEGF, bFGF, and HGF have been reported both in patients with MM30 and polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy and skin changes (POEMS) syndrome.31 Recently, a remarkable anti-tumor effect of thalidomide has been noted by multiple groups in patients with refractory and relapsed MM.32,33,34,35,36,37 Although this marked anti-tumor activity of thalidomide is attributed to its anti-angiogenic properties,38 its ability to modulate cytokines, alter adhesion molecule expression, and enhance host cell anti-tumor immunity may also mediate anti-MM activity.39

In the present study, we characterized the regulation of angiogenic cytokines (VEGF and bFGF) and IL-6 secretion in the BM millieu. Importantly we demonstrate that both VEGF and IL-6 secretion are upregulated by MM to BMSC binding, and that thalidomide and its analog, IMiD1 (CC4047), abrogate this response.

Materials and methods

Reagents

Cytokines and antibodies:

Recombinant human IL-6, recombinant human VEGF165, neutralizing anti-human IL-6 monoclonal antibody (MoAb), and neutralizing goat anti-human VEGF polyclonal antibody were purchased from R&D Systems (Minneapolis, MN, USA). These reagents were reconstituted with sterile PBS and stored at 4°C. Recombinant IL-6 was used at 10 ng/ml (ED50 on an IL-6 dependent plasmacytoma cell line = 0.2–0.8 ng/ml) and recombinant VEGF165 at 100 ng/ml (ED50 on HUVEC cells = 2–6 ng/ml). Anti-human IL-6 MoAb was used at 10 μg/ml (ND50 on an IL-6 dependent plasmacytoma cell line in the presence of 2.5 ng/ml of rhIL-6 = 0.05–0.15 μg/ml). Goat anti-human VEGF165 polyclonal antibody was used at 1 μg/ml (ND50 on VEGF-dependent human umbilical vein endothelial cells (HUVEC) in the presence of 10 ng/ml of rhVEGF = 0.01–0.03 μg/ml).

Thalidomide and its analog IMiD 1:

Thalidomide and its potent analog immunomodulatory drug (IMiD1-CC4047) (Celgene Corporation, Warren, NJ, USA) were dissolved in dimethyl sulfoxide (DMSO) (Sigma, St Louis, MO, USA) and stored at −20°C. These drugs were diluted in culture medium (0.0001 to 100 μM) immediately before use. The thalidomide analog (IMiD1-CC4047) stimulates T cell proliferation, as well as interleukin-2 and interferon-γ secretion, and is not a phosphodiesterase 4 inhibitor. This drug also inhibits tumor necrosis factor-α, interleukin-1β and IL-6 secretion, but greatly increases interleukin-10 production by lipopolysaccharide-stimulated peripheral blood mononuclear cells (PBMCs).40 Thalidomide and its analog-IMiD 1 (CC4047) were used at a final concentration of 100 μM and 1 μM, respectively.

Human MM cells

The human MM cell lines, RPMI 8226 and U266, and EBV-transfected B cell lines from MM patients, HS-Sultan, IM-9 and ARH 77 were obtained from the American Type Culture Collection (Rockville, MD, USA). Dexamethasone (Dex) sensitive (MM.1S) and Dex resistant (MM.1R) cells were a kind gift from Dr Steven Rosen (Northwestern University, Chicago, IL, USA). Doxorubicin (DOX40), mitoxantrone (MR20) and melphalan (LR5) resistant RPMI 8226 cell lines were kindly provided by Dr William Dalton (Moffitt Cancer Center, Tampa, FL, USA). Malignant plasma cells (MM1 and MM2) were obtained by BM aspiration from two patients with newly diagnosed MM and were 95% CD38+CD45RA. All cell lines and patients’ MM cells were cultured in modified RPMI 1640 (Mediatech 800, Cellgro, VA, USA) containing 10% fetal bovine serum (FBS) (Harlan, Indianapolis, IN, USA), 100 U/ml penicillin (Pen), 100 μg/ml streptomycin (Strep) and 2 mM L-glutamine (Mediatech). The drug-resistant cell lines were cultured with doxorubicin or mitoxantrone or melphalan to confirm their lack of drug sensitivity.

BMSCs

BM specimens were obtained from patients with MM and from normal healthy donors. PBMCs were separated by Ficoll–Hypaque density sedimentation and were used to establish long-term BM cultures according to the method described by Gartner and Kaplan, with slight modifications.41 PBMCs were suspended in 8 ml growth medium containing Iscove's modified Dulbecco's medium (IMDM) (Sigma), 20% FBS, Pen (100 U/ml), Strep (100 μg/ml), and 8 mM L-glutamine. The non-adherent cells were removed after 2 days, and thereafter the media was changed once or twice weekly. In some cultures, human recombinant β-epidermal growth factor (4 ng/ml) was added every 3–4 days to augment the growth of BMSCs. When an adherent monolayer had developed, the cells were harvested in Hank's buffered saline solution (HBSS) containing 0.05% trypsin and 0.53 mM EDTA (Mediatech). The trypsin was inactivated using RPMI-1640 with 10% FBS, and the cells were washed and collected by centrifugation. The confluent adherent BMSCs showed predominantly fibroblast morphology and were suspended in IMDM, 10% FBS, Pen (100 U/ml), Strep (100 μg/ml), and 8 mM L-glutamine.

Preparation of conditioned medium from BMSC cultures

Trypsinized BMSCs from normal donors and MM patients were layered on 24-well polystyrene plates (Falcon; Becton Dickinson, Franklin Lakes, NJ, USA) in a concentration of 1 × 105 cells per well. After the BMSCs formed a confluent adherent monolayer, the wells were washed twice with IMDM containing 10% FBS. The MM/EBV transformed B cell lines (RPMI, MM.1S, HS-Sultan) and MM patient cells (MM.1) (2.5 × 105 cells/well) were either cultured with BMSCs, or placed in Transwell inserts (Costar, Cambridge, MA, USA), and then cultured with BMSCs. In some experiments, either BMSCs or MM cells were fixed with 1% paraformaldehyde before addition to the cultures in order to determine the source of cytokines. In some cultures, IL-6, VEGF, neutralizing anti-IL-6 antibody, neutralizing anti-VEGF antibody (with appropriate isotype antibody controls), thalidomide, and IMiD1 (CC4047) were used in appropriate final concentrations to discern their effects on cytokine modulation. After incubation at 37°C for 48 h in a 5% CO2 incubator with 100% humidity (Steri-cult 200 incubator, Forma Scientific, OH, USA), the supernatants were collected after centrifugation (Sorvall RT6000B Refrigerated Centrifuge; DuPont, CT, USA) at 1600 r.p.m. for 5 min. The supernatants were filtered through 0.22 μm syringe filters (Costar, Corning, NY, USA), aliquotted in three to four vials, and were frozen at −70°C. All experiments were done in triplicate and the levels depicted are a mean of three experiments for each cell line or patient MM cells.

Measurements of cytokine levels

VEGF165 levels in culture supernatants were measured using Enzyme Linked Immunosorbent Assay (ELISA) development reagents (R&D Systems); the levels of IL-6 were measured by DuoSet ELISA development systems (R&D Systems), according to the manufacturer's recommendations. We used quantikine human FGF basic colorimetric sandwich ELISA (R&D Systems) for bFGF quantitation. Briefly, these assays use a sandwich enzyme immunoassay with MoAbs specific for each cytokine precoated on 96-well microtiter plates. Standard control and sample (100 μl) are added into wells in triplicate. After an incubation, the wells were washed and an enzyme linked polyclonal antibody specific for each cytokine was added. After washing thoroughly, a substrate solution was added, and color developed in proportion to the amount of cytokine bound. The optical density of each well was determined by a microplate reader at 450 nm. The value of blank was subtracted from the standard control and samples. A standard curve was generated by plotting the mean absorbance of each standard vs the known concentration of cytokine on a log-log scale. Concentrations of various cytokines were reported in picograms/ml. The limit of detection for IL-6 and b-FGF was 10 pg/ml, and for VEGF was 30 pg/ml. Statistical analysis was done using the Microsoft Excel program for mean and standard deviation, and Student's t-test was performed using unpaired t-test in DA stat program. All P values were two-tailed; values 0.05 were considered as to be statistically significant.

Results

MM cell lines secrete VEGF

We first measured secretion of VEGF. IL-6 and bFGF in culture supernatants of MM cells/EBV-transformed B cells from MM patients using specific ELISAs. VEGF was secreted by IM-9, HS-Sultan, ARH 77 (EBV transfected B cell lines from MM patients), U266 (Figure 1a), RPMI 8226 (Figure 1b) and MM1.S (Figure 1c) MM cell lines. We next compared VEGF secretion in drug-sensitive vs drug-resistant MM cell lines. RPMI 8226 cells produced more VEGF (mean ± s.d. = 10740 ± 57 pg/ml) than RPMI-derived MM cells resistant to mitoxantrone (RPMI/MR20) (mean ± s.d. = 6710 ± 10 pg/ml), melphalan (RPMI/LR5) (mean ± s.d. = 1286 ± 144 pg/ml), or doxorubicin (RPMI/DOX40) (mean ± s.d. = 287 ± 0 pg/ml) (RPMI > RPMI/MR20 > RPMI/LR5 > RPMI/DOX40) (Figure 1b) at 48 h, but VEGF secretion by these cell lines (except RPMI/DOX 40) was comparable at 72 h. The VEGF secretion was higher in MM1.S cells (Dex sensitive) (mean ± s.d. = 1135 ± 4 pg/ml) compared to MM1.R cells (Dex resistant) (mean ± s.d. = 454 ± 15 pg/ml) (Figure 1c) at 48 and 72 h. Finally, MM cells from two patients (MM1 and MM2) also secreted VEGF (Figure 1d). No IL-6 or basic-FGF secretion was detectable in cultures of these cell lines and patient MM cells (data not shown).

Figure 1
Figure 1

MM cell lines secrete VEGF. VEGF levels (pg/ml) from 1 × 106 MM cells at 24, 48 and 72 h were measured by ELISA: (a) U266, HS-Sultan, ARH 77 and IM-9 MM cell lines; (b) RPMI 8226 as well as RPMI/DOX40, RPMI/MR20 and RPMI/LR5 (drug resistant) MM cell lines; (c) MM1.S (Dex-sensitive) and MM1.R (Dex-resistant) MM cell lines; and (d) 2 MM patient cells: MM1 and MM2.

MM patient and normal donor BMSCs secrete VEGF, IL-6 and bFGF

To assay for paracrine production of these cytokines in the BM, long-term BMSCs were grown as in previous studies.5,42 MM-BMSCs and N-BMSCs secreted variable amounts of VEGF, IL-6 and b-FGF (Figure 2a, b and c). The secretion of VEGF in MM-BMSCs was 760 ± 129 (range, 552–917) pg/ml and in N-BMSCs was 150 ± 102, (range, 65–243) pg/ml (Figure 2a). The secretion of IL-6 in N-BMSCs was 107 ± 90 (range, 9–200) pg/ml and in MM-BMSCs was 187 ± 91 (range, 46–300) pg/ml (Figure 2b). bFGF secretion was 243 ± 130 (range, 128–373) pg/ml in N-BMSCs and in MM-BMSCs was 36 ± 20 (range, 17–71) pg/ml (Figure 2c).

Figure 2
Figure 2

MM patient and normal donor BMSCs secrete VEGF, IL-6 and bFGF. Secretion of (a) VEGF, (b) IL-6 and (c) bFGF (pg/ml) from BMSCs of two normal donors and BMSCs from four MM patients (BMSC 1, 2, 3 and 4) were measured by ELISA.

Adhesion of MM cell lines and MM patient cells to N-BMSC and MM-BMSCs induces VEGF and IL-6 secretion

Given our prior studies that adhesion of MM cells to BMSCs can induce NF-κB-dependent upregulation in IL-6 secretion,43 we assayed whether VEGF and bFGF were similarly affected. We observed increased secretion of VEGF and IL-6 when either MM cell lines or patient MM cells were adhered to either N-BMSC or MM-BMSCs (Figure 3a, b and c). Specifically, VEGF secretion increased by 1.5- to 3.1-fold (P  0.003), when compared to the sum of VEGF levels from BMSCs and MM cells alone (Figure 3a). Similarly IL-6 levels increased by 1.9- to 56-fold (P  0.002) in co-cultures of BMSCs and MM cells, relative to cultures of these cells alone (Figure 3b). In contrast, bFGF levels decreased (P  0.09) in cultures of BMSCs and MM cells, compared to either MM cells or BMSCs alone (Figure 3c). As depicted in Figure 3, we found this phenomenon in all three cell lines (HSS, RPMI, MM.1S) and MM patient cells (MM1).

Figure 3
Figure 3

Adhesion of MM cells to MM patient and normal donor BMSCs induces VEGF and IL-6 secretion. Levels of (a) VEGF, (b) IL-6, and (c) bFGF (pg/ml) from N-BMSC or MM-BMSCs, MM cell lines and co-cultures of BMSCs and MM cell lines were measured by ELISA.

Source of VEGF and IL-6 secretion triggered by MM to BMSC binding

In order to determine whether the increased secretion of VEGF and IL-6 triggered by MM to BMSC binding was from MM cells and/or BMSCs, either MM cells or BMSCs were fixed with paraformaldehyde prior to adhesion, as in prior studies.5 As seen in Figure 4a, fixation of BMSCs before HS-Sultan MM cell binding blocked the upregulation of VEGF secretion (P < 0.001), whereas fixation of HS-Sultan MM cells before adhesion did not alter upregulation of VEGF secretion (P = 0.18), suggesting that BMSCs are the major source for VEGF secretion in this experiment. In the case of IL-6, fixation of tumor cells partially (P < 0.001), and fixation of BMSCs to a greater extent (P < 0.001), inhibited adhesion-induced IL-6 secretion (Figure 4g). None of our cell lines produced IL-6, and it can therefore be concluded that the major source of IL-6 in cocultures is BMSCs (Figure 4h, i, j, k and l). However, both BMSCs and MM cells are responsible for increased VEGF secretion in co-cultures (Figure 4b, c, d, e and f).

Figure 4
Figure 4

Source of VEGF and IL-6 secretion triggered by MM cell to BMSC binding. Levels of VEGF and IL-6 from MM BMSCs; MM cell lines; co-cultures of MM cells and BMSC; co-cultures of BMSCs and MM cells (fixed by 1% paraformaldehyde); and co-cultures of MM cells and BMSCs (fixed by 1% paraformaldehyde; were measured by ELISA. Panels a, b, c, d, e and f show VEGF secretion and panels g, h, i, j, k and l show IL-6 secretion in experiments using HSS, MM.1S, RPMI, U266, MM1 and MM2 cells. IL-6 was secreted exclusively by BMSCs in these co-cultures whereas VEGF levels were contributed both by BMSCs and MM cells.

In order to determine whether MM cell to BMSC contact is required for upregulation of VEGF and IL-6 secretion, MM cells were placed in Transwell inserts to allow diffusion of humoral factors, but preclude MM cell to BMSC contact. We have previously utilized this system to demonstrate that MM cells secrete TGF-β which upregulates IL-6 secretion from BMSCs.42 When MM.1S cells in Transwell insert chambers were placed in proximity to BMSCs, the induction of VEGF was similar to that noted when MM.1S cells were adhered to BMSCs (P = 0.177) (Figure 5a), suggesting the importance of a humoral factor/s in these co-cultures. IL-6 secretion was also comparable (P = 0.44) in cultures where MM.1S cells were placed in Transwell insert chambers compared to cultures where MM.1S cells were allowed to adhere to BMSCs (Figure 5f), suggesting that humoral factor/s may be involved leading to enhanced secretion. These data are consistent with our previous study delineating the role of TGFβ secreted by MM cells in triggering IL-6 secretion in BMSCs.42 Similar trends in VEGF (Figure 5b, c, d and e) and IL-6 (Figure 5g, h, i and j) secretion were noted in co-cultures of BMSCs with and without adhesion of RPMI8226, U266, MM1 and MM2 cells.

Figure 5
Figure 5

Source of VEGF and IL-6 secretion triggered by MM cell to BMSC binding. Levels of VEGF and IL-6 (pg/ml) from MM-BMSCs; MM cells; co-cultures of MM cells and BMSCs; as well as co-cultures of MM cells in Transwell insert chambers and BMSCs; were measured by ELISA. Panels a, b, c, d and e show VEGF secretion, and panels f, g, h, i and j show IL-6 secretion in experiments using MM.1S, RPMI, U266, MM1 and MM2 cells. High levels of VEGF and IL-6 persisted in cultures were MM cells were placed in Transwell insert chambers indicating the role of humoral factors in increased secretion of these cytokines in co-cultures of BMSCs and MM cells.

Exogenous recombinant VEGF165 increases IL-6 secretion by BMSCs

Having shown increased VEGF and IL-6 secretion triggered by MM cell binding to BMSCs, we next explored the interplay between VEGF and IL-6 in the BM millieu. BMSCs and MM cells were first stimulated by recombinant VEGF165 (100 ng/ml), and the secretion of IL-6 was measured by ELISA. As can be seen in Figure 6a, IL-6 secretion in BMSCs increased markedly (158-fold, P < 0.001) in response to recombinant VEGF165 stimulation. In contrast, no IL-6 secretion was detected from MM cell lines or patient MM cells after recombinant VEGF165 stimulation (data not shown). Importantly, goat anti-human VEGF165 neutralizing polyclonal antibody (1 μg/ml) reduced IL-6 secretion by 27% in BMSC cultures (P = 0.002) (Figure 6c) and by 50% in cultures of MM cells adherent to BMSCs (P < 0.001) (Figure 6e). These data demonstrate the role of VEGF secreted by BMSCs and MM cells in triggering IL-6 secretion from BMSCs. Figure 6g also shows a significant decrease in IL-6 secretion (P = 0.001) in co-culture of MM2 cells and BMSCs after addition of anti-human VEGF165 polyclonal antibody.

Figure 6
Figure 6

(a) Exogenous recombinant VEGF165 increases IL-6 secretion by BMSCs. Levels of IL-6 (pg/ml) were measured by ELISA in supernatants from BMSCs before and after stimulation with recombinant VEGF (100 ng/ml). (b) Exogenous recombinant IL-6 increases VEGF secretion by BMSCs. Levels of VEGF secretion from BMSCs before and after stimulation with 10 ng/ml of recombinant human IL-6 were measured by ELISA. (c) Anti-VEGF neutralizing antibody (1 μg/ml) reduces IL-6 secretion from BMSCs. Levels of IL-6 from BMSCs cultures without anti VEGF neutralizing antibody and with anti-VEGF neutralizing antibody were measured by ELISA. (d) Anti-IL-6 neutralizing antibody (10 μg/ml) does not reduce VEGF secretion from BMSCs. Levels of VEGF from BMSCs cultures without anti-IL-6 neutralizing antibody and with anti-IL-6 neutralising antibody were measured by ELISA. (e) Goat anti-human VEGF165 neutralizing antibody reduces IL-6 secretion in co-cultures of BMSCs and MM cells. Levesl of IL-6 were measured by ELISA in culture supernatants of BMSCs; HS-Sultan cells; BMSCs + HS-Sultan cells as well as BMSC + HS-Sultan cells + anti-VEGF antibody. Panel g shows similar reductions in IL-6 levels in cocultures of MM2 + BMSCs. (f) Anti-human IL-6 neutralizing MoAb reduced VEGF secretion in co-cultures of BMSCs and MM cells. Levels of VEGF were measured by ELISA in culture supernatants of BMSCs; HS-Sultan cells; BMSCs + HS-Sultan cells; as well as BMSC + HS-Sultan cells + anti-IL-6 antibody. Panel h shows similar reductions in VEGF levels in cocultures of MM2 + BMSCs.

Exogenous recombinant IL-6 increases VEGF secretion by BMSCs and MM cells

BMSCs and MM cells were next stimulated by recombinant IL-6 (10 ng/ml), and the secretion of VEGF was measured by ELISA. As can be seen in Figure 6b, recombinant IL-6 increased VEGF secretion from BMSCs (2.5-fold, P < 0.001). In addition, HS-Sultan cells stimulated with recombinant IL-6 also upregulated VEGF secretion (1.5-fold, data not shown). However, stimulation of RPMI8226, MM.1S, U266, MM1 and MM2 cells by IL-6 did not increase VEGF secretion. Figure 6d shows no significant change in VEGF secretion by BMSCs alone upon addition of anti-human IL-6 (10 μg/ml) MoAb (P = 0.06). Importantly, anti-human IL-6 MoAb reduced VEGF levels by 16.8% (P = 0.02) in cultures of HS Sultan cells adhered to BMSCs (Figure 6f) and by 19% (P = 0.004) in cultures of MM2 cells adhered to BMSCs (Figure 6h). These data demonstrate the role of IL-6 secreted from BMSCs in triggering VEGF secretion in both MM cells and BMSCs.

Thalidomide and immunomodulatory drug (IMiD1-CC4047) reduce IL-6, and VEGF secretion in co-cultures of MM cells and BMSCs

Given the known ability of thalidomide and immunomodulatory drug (IMiD1-CC4047) to modulate cytokines and inhibit angiogenesis,40 we next examined the effect of thalidomide on the secretion of VEGF and IL-6 in MM-BMSC cultures. Thalidomide (100 μM) reduced VEGF and IL-6 secretion triggered by HS Sultan cells to BMSC binding by 87% (P < 0.001) and 94% (P < 0.001), respectively (Figure 7a, g). Reductions in secretion of VEGF after addition of thalidomide was also noted in cocultures of BMSCs and RPMI8226 (Figure 7c), U266 (Figure 7d), MM1 (Figure 7e) and MM2 (Figure 7f) cells. IL-6 secretion also decreased in co-cultures of BMSCs and MM.1S (Figure 7h), RPMI 8226 (Figure 7i), U266 (Figure 7j), MM1 (Figure 7k) and MM2 (Figure 7l) cells in the presence of thalidomide.

Figure 7
Figure 7

Thalidomide (100 μM) and immunomodulatory drug (IMiD1-CC4047) (1 μM) reduce IL-6 and VEGF secretion in cocultures of MM cells and BMSCs. Levels of VEGF (a) and IL-6 (pg/ml) (g) were measured by ELISA in cultures of BMSCs; HS-Sultan cells; BMSC + HS-Sultan cells; BMSC and HS-Sultan cells with thalidomide (THD); as well as BMSC and HS-Sultan MM cells with IMiD1. Panels b, c, d, e, f show VEGF levels in cultures using MM.1S, RPMI, U266, MM1 and MM2 cells and panels h, i, j, k and l show IL-6 levels in cultures using MM.1S, RPMI, U266, MM1 and MM2 cells.

Immunomodulatory drug (IMiD1-CC4047) (1 μM) reduced VEGF (P < 0.001) and IL-6 (P < 0.001) secretion in co-cultures of HS Sultan cells and BMSCs (Figure 7a, g). We detected significant decrease in the secretion of VEGF in cultures of BMSCs and MM.1S (Figure 7b), RPMI 8266 (Figure 7c), U266 (Figure 7d), MM1 (Figure 7e) and MM2 (Figure 7f) after addition of IMiD1. IL-6 secretion also reduced significantly in co-cultures of BMSCs and RPMI8226 (Figure 7i), MM1 (Figure 7k) and MM2 (Figure 7l) with IMiD1.

Thalidomide and immunomodulatory drug (IMiD1-CC4047) reduce VEGF secretion in cultures of MM cells

We also directly examined the effect of these drugs on VEGF secretion in MM cells. Thalidomide reduced VEGF secretion significantly in cultures of HS-Sultan (P = 0.02), MM1.S cells (P = 0.03), MM2 cells (P = 0.078) (Figure 8a, b). Similarly, IMiD1-CC4047 lowered VEGF secretion significantly in HS-Sultan (P = 0.007), RPMI (P = 0.01), U266 (P = 0.001) and MM2 cells (P < 0.001) (Figure 8a, b).

Figure 8
Figure 8

Thalidomide and immunomodulatory drug (IMiD1-CC4047) reduce VEGF secretion in cultures of MM cells. Levels of VEGF were measured by ELISA in cultures of HS-Sultan cells, patient MM1 cells, and MM.1S cells before and after treatment with THD or ImiD1-CC4047 (a). Panel b shows levels of VEGF in cultures of RPMI, U266 and MM2 cells with thalidomide and IMiD1.

Thalidomide and immunomodulatory drug (IMiD1-CC4047) reduced VEGF and IL-6 secretion in cultures of BMSCs

We also examined the effect of thalidomide and IMID1-CC4047 on secretion of VEGF and IL-6 by myeloma BMSCs alone. Significant reductions in VEGF levels were noted in BMSCs cultures with thalidomide (P = 0.007) and IMID1 (P = 0.004) in experiment 1 (Figure 9a), and significant reduction in IL-6 secretion was also observed in cultures of BMSCs with IMID1 in experiment 1 (P = 0.05) and in experiment 2 (P = 0.008) (Figure 9b).

Figure 9
Figure 9

Thalidomide and immunomodulatory drug (IMiD1-CC4047) reduce VEGF and IL-6 secretion in cultures of BMSCs. Panels and b show the levels of VEGF and IL-6 in BMSCs cultures before and after treatment with THD or IMiD1-CC4047.

Discussion

Many studies have associated increased angiogenesis with growth and progression of MM.26,27,28 Elevated serum levels of angiogenic cytokines, such as VEGF44 and bFGF,45 have been correlated with progression in MM. This prompted us to study these angiogenic cytokines in the context of MM–BMSC interactions, given their pivotal role in the paracrine IL-6-mediated growth and survival of MM cells.5 We first detected VEGF secretion by all MM cell lines, EBV-transfected B cell lines from MM patients, and patient MM cells, consistent with previous reports.46,47 The levels of VEGF in our cultures of MM cell lines, EBV-transfected B cell lines from MM patients, and patient MM cells were well within the range of its biological activity (150–300 pg/ml).48

We also detected VEGF in supernatants from BMSCs derived from normal donors and from MM patients. It is difficult to say whether BMSCs from MM patients secreted more VEGF than BMSCs from normal donors; the numbers in both groups are small and this observation needs confirmation in studies of large numbers of closely matched normal donors and MM patients. There was no difference in IL-6 secretion between normal and MM patient BMSCs, consistent with our previous report.5 Importantly, we found up to three-fold increases in VEGF secretion in co-cultures of BMSCs and MM cells, as well as increased IL-6 secretion, as in our previous studies.5,43,49 The source of increased secretion of IL-6 was primarily BMSCs, since fixation of BMSCs prior to MM cell binding abrogated this response. However, increased VEGF secretion was contributed to both by BMSCs and MM cells. The increased VEGF secretion persisted even when MM cells in Transwell insert chambers were juxtaposed to BMSCs, suggesting that VEGF secretion in co-cultures is dependent not only on direct cell to cell contact and adhesion, but also triggered by a diffusible factor. This is analogous to IL-6 secretion, which is induced both by adhesion of MM cells to BMSCs and TGFβ secreted by MM cells acting on BMSCs.5,42 This diffusible factor could be a single cytokine or more likely multiple cytokines – such as IL-6, TNF-α, TGF-β, IL-1β, PDGF, HGF, bFGF, M-CSF, GM-CSF or G-CSF – produced by BMSCs and/or MM cells, which increase the secretion of VEGF and IL-6 and thereby enhances growth and survival of MM cells in the BM microenvironment. In this study, we have shown that exogenous IL-6 stimulates VEGF secretion from BMSCs, and conversely, that exogenous VEGF increases IL-6 secretion from BMSCs, consistent with a prior report.47 Neutralizing antibody to IL-6 or VEGF in co-cultures of MM cells (HSS and MM2) and BMSCs also reduced the secretion of VEGF or IL-6, respectively, further confirming the interplay of these cytokines in MM.

We have shown that VEGF causes significant proliferation of MM.1S cell line and patient MM cells (n = 5) at 100 ng/ml (P < 0.001) and 1000 ng/ml (P < 0.001).50 The biological functions of VEGF are mediated by two VEGF receptors – VEGFR-1 (flt1)51 and VEGFR-2 (flk-1/KDR).52 In recent studies, we were able to identify VEGFR-1 in MM.1S and patient MM1 cells by RT-PCR and immunoprecipitation. We also confirmed the biological activity of VEGF in MM.1S cells, as evidenced by specific phosphorylation of tyrosine kinase (Flt-1) receptor after stimulation with VEGF.50 Our results are consistent with one previous report demonstrating moderate expression of Flt-1 in ARH77 MM cells.46 These observations indicate that VEGF may play a predominant role in paracrine, and a minor role in autocrine, MM cell growth and are consistent with prior reports.46,47 We did not find any increase in proliferation of MM.1S cells in vitro when increasing concentrations of IL-6 (1, 10, 30, 50 ng/ml) were added to VEGF (500 ng/ml) as compared to cultures of MM.1S cells with IL-6 and VEGF separately.50

In this study, the anti-angiogenic drugs, thalidomide and IMiD1 (CC 4047), reduced VEGF and IL-6 secretion in co-cultures of BMSCs and MM cells, highlighting the importance of studying MM cells within the BM millieu. These drugs also caused reduction of VEGF secretion in MM cell lines, and both VEGF and IL-6 secretion in BMSCs alone. Thalidomide was more effective in reducing VEGF and IL-6 levels in co-cultures of BMSCs and MM cells, whereas IMID1 (CC 4047) caused greater reductions in VEGF secretion from MM cells and VEGF and IL-6 secretion from BMSCs. Our prior studies show that thalidomide, even at a concentration of 100 μM, does not cause apoptosis or G1 growth arrest in MM cell lines,53 suggesting that this reduction in VEGF secretion triggered by MM-BMSC binding may represent an important mechanism for its anti-MM activity. The more potent IMiD1-CC4047 (1 μM) does act directly to induce G1 growth arrest and apoptosis in HS-Sultan, RPMI, U266 and MM2 cells.53 Therefore, the observed downregulation in VEGF secretion could be due both to direct MM cell toxicity and to inhibition of VEGF secretion in MM–BMSC cultures.

In co-cultures of BMSCs and MM cells, bFGF levels were low. One explanation of this phenomenon could be binding of bFGF to the heparan sulfate proteoglycan syndecan-1, which is abundantly expressed on MM cell lines.54 Syndecan-1, which is expressed on MM cells, mediates adhesion of tumor cells to extracellular matrix proteins and inhibits tumor cell invasion in collagen gels. Moreover, the shed form of syndecan-1 has been found to inhibit osteoclast formation, modulate bone disease, and induce apoptosis of MM cells.55 There is evidence that syndecan-1 acts as a positive regulator of the activity of bFGF.56 Basic FGF subsequently binds to a high affinity tyrosine kinase bFGF receptor, thereby triggering signal transduction. In our study, bFGF levels were higher in cultures of MM cells in Transwell insert chambers and thereby separated from BMSCs (data not shown). This suggests that the decreased bFGF in cultures of MM cells adherent to BMSCs may be due to binding to syndecan-1 on MM cells and possible subsequent downregulation of secretion of bFGF by BMSCs.

We conclude that angiogenic cytokines such as VEGF and bFGF are secreted by BMSCs, and that VEGF is also secreted by most MM cell lines. VEGF and IL-6 secretion increases significantly in cultures of MM cells adherent to BMSCs, due to enhanced secretion of these cytokines from BMSCs in a predominant paracrine fashion. Importantly, anti-angiogenic drugs such as thalidomide and IMiDs abrogate this adhesion-related upregulation of both VEGF and IL-6, besides having a direct effect on MM and BMSCs. These studies therefore delineate important novel mechanisms of activity of these drugs against MM and further suggest the clinical utility of novel treatment paradigms targeting not only the tumor cell directly, but also cellular interactions and cytokine secretion in the BM millieu.

References

  1. 1.

    , , , . Cancer statistics, 2000 CA Cancer J Clin 2000 501: 7–33

  2. 2.

    , , , , . Response pattern of purified myeloma cells to hematopoietic growth factors Blood 1989 73: 1915–1924

  3. 3.

    , , , , , , , . Paracrine rather than autocrine regulation of myeloma-cell growth and differentiation by interleukin-6 Blood 1989 73: 517–526

  4. 4.

    , , , , , , , , , . Autocrine generation and essential requirement of BSF-2/IL-6 for human multiple myeloma Nature 1988 332: 83–85

  5. 5.

    , , , , . Adhesion of human myeloma derived cell lines to bone marrow stromal cells stimulates interleukin-6 secretion Blood 1993 82: 3712- 3720

  6. 6.

    , , , , , , . Interleukin-6 triggers cell growth via the ras-dependent mitogen-activated protein kinase cascade J Immunol 1997 159: 2212–2221

  7. 7.

    , , , , , , , . Interleukin-6 inhibits Fas-induced apoptosis and stress-activated protein kinase activation in multiple myeloma cells Blood 1997 89: 227–234

  8. 8.

    , , , . Serum levels of interleukin-6, a potent myeloma cell growth factor, as a reflect of disease severity in plasma cell dyscrasias J Clin Invest 1989 84: 2008–2011

  9. 9.

    , , , . Interleukin-6 in human multiple myeloma Blood 1995 85: 863–872

  10. 10.

    , , . Multiple myeloma: increasing evidence for a multistep transformation process Blood 1998 91: 3–21

  11. 11.

    . Clinical applications of research on angiogenesis N Engl J Med 1995 333: 1757–1763

  12. 12.

    , , , . Tumor angiogenesis and metastasis – correlation in invasive breast carcinoma N Engl J Med 1991 324: 1–8

  13. 13.

    , , , , . Tumor angiogenesis correlates with metastasis in invasive prostate carcinoma Am J Pathol 1993 143: 401–409

  14. 14.

    , , , , , , , . Prognostic value of angiogenesis in operable non-small cell lung cancer J Pathol 1996 179: 80–88

  15. 15.

    , , , . Tumor angiogenesis correlates with metastatic disease, n-myc amplification, and poor outcome in human neuroblastoma J Clin Oncol 1996 14: 405–414

  16. 16.

    , . Angiogenic factors Science 1987 235: 442–447

  17. 17.

    , , , , , , , . Significance of vascular endothelial growth factor messenger RNA expression in primary lung cancer Clin Cancer Res 1996 2: 1411–1416

  18. 18.

    , , , , , , . Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in adenocarcinomas of the gastrointestinal tract Cancer Res 1993 3: 4727–4735

  19. 19.

    , , , . Androgens regulate vascular endothelial growth factor content in normal and malignant prostatic tissue Clin Cancer Res 1997 3: 2507–2511

  20. 20.

    , , , , , , , . Elevated levels of the angiogenic cytokines basic fibroblast growth factor and vascular endothelial growth factor in sera of cancer patients Br J Cancer 1997 76: 238–243

  21. 21.

    , , , , , . Elevated levels of an angiogenic peptide, basic fibroblast growth factor, in the urine of patients with a wide spectrum of cancer J Natl Cancer Inst 1994 86: 356–361

  22. 22.

    , , , , , . Angiogenesis spectrum in the stroma of B-cell non-Hodgkin's lymphoma. An immunohistochemical and ultrastructural study Eur J Haematol 1996 56: 45–53

  23. 23.

    , , , , , , . Vascular endothelial growth factor, a possible paracrine growth factor in human acute myeloid leukemia Blood 1997 89: 1870–1875

  24. 24.

    , , , , , , , , , , , . Angiogenesis in acute and chronic leukemias and myelodysplastic syndromes Blood 2000 96: 2240–2245

  25. 25.

    , , , , , , , , . Cellular vascular endothelial growth factor is a predictor of outcome in patients with acute myeloid leukemia Blood 1999 94: 3717–3721

  26. 26.

    , , , , , , , , , . Bone marrow angiogenesis and mast cell density increase simultaneously with progression of human multiple myeloma Br J Cancer 1999 79: 451–455

  27. 27.

    , , , , , , , , , , . Prognostic value of bone marrow angiogenesis in multiple myeloma Clin Cancer Res 2000 6: 3111–3116

  28. 28.

    , , , , , , , , , , . Angiogenesis in newly diagnosed multiple myeloma: poor prognosis with increased microvessel density (MVD) in bone marrow biopsies Blood 1998 92: 98a

  29. 29.

    , , , , . Bone marrow angiogenesis in patients achieving complete response after stem cell transplantation for multiple myeloma Leukemia 1999 13: 469–472

  30. 30.

    , , , , , , , . Angiogenic factors in multiple myeloma: higher levels in bone marrow than in peripheral blood Haematologica 2000 85: 800–805

  31. 31.

    , , , , , , , , , . Growth factors in POEMS syndrome: evidence for a marked increase in circulating vascular endothelial growth factor Arthr Rheum 1997 40: 786–787

  32. 32.

    , , , , , , , , , , , . Antitumor activity of thalidomide in refractory multiple myeloma N Engl J Med 1999 341: 1565–1571

  33. 33.

    , , , , , , , , . Thalidomide in the treatment of relapsed multiple myeloma Mayo Clin Proc 2000 75: 897–901

  34. 34.

    , , , , , . Frequent good partial remissions from thalidomide including best response ever in patients with advanced refractory and relapsed myeloma Br J Haematol 2000 109: 89–96

  35. 35.

    , , , , , , . Therapy with thalidomide in refractory multiple myeloma patients – the revival of an old drug Br J Haematol 2000 108: 391–393

  36. 36.

    , , . Successful treatment of multiple myeloma relapsing after high-dose therapy and autologous transplantation with thalidomide as a single agent Bone Marrow Transplant 2000 25: 1319–1320

  37. 37.

    , , , , , , . Low-dose thalidomide in the treatment of refractory myeloma Haematologica 2000 85: 1111–1112

  38. 38.

    , , , . Thalidomide is an inhibitor of angiogenesis Proc Natl Acad Sci USA 1994 91: 4082–4085

  39. 39.

    , . Thalidomide – a revival story N Engl J Med 1999 341: 1606–1608

  40. 40.

    , , , , , , , , . Differential cytokine modulation and T cell activation by two distinct classes of thalidomide analogs that are potent inhibitors of TNF-α J Immunol 1999 163: 380–386

  41. 41.

    , . S. Long term culture of human bone marrow cells Proc Natl Acad Sci USA 1980 77: 4756–4759

  42. 42.

    , , , , , , , . Transforming growth factor β1: differential effects on multiple myeloma versus normal B cells Blood 1996 87: 1928–1938

  43. 43.

    , , , , , , . Multiple myeloma cell adhesion-induced interleukin-6 expression in bone marrow stromal cells involves activation of NF-kappa B Blood 1996 87: 1104–1112

  44. 44.

    , , , , , . Vascular endothelial growth factor and angiogenesis in multiple myeloma Blood 1999 94 (Suppl. 1): 305b

  45. 45.

    , , , , , . Basic fibroblast growth factor serum levels in multiple myeloma patients correlate with stage of disease and lapse under therapy Blood 1999 94 (Suppl. 1): 305b

  46. 46.

    , , , . Expression of vascular endothelial growth factor and its receptors in hematopoietic malignancies Cancer Res 1999 59: 728–733

  47. 47.

    , , , , , , , , . Vascular endothelial growth factor and interleukin-6 in paracrine tumor–stromal cell interactions in multiple myeloma Blood 2000 95: 2630–2636

  48. 48.

    , , , , , , , , . Human vascular permeability factor: isolation from U937 cells J Biol Chem 1989 264: 20017–20024

  49. 49.

    , , , , . Regulation of interleukin-6 in multiple myeloma and bone marrow stromal cells Stem Cells 1995 13 (Suppl. 2): 35–39

  50. 50.

    , , , , , , , , , , , , . Vascular endothelial growth factor (VEGF)-triggers signaling cascades mediating multiple myeloma cell growth and migration Blood 2001 98: 428–435

  51. 51.

    , , , , , . The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor Science 1992 255: 989–991

  52. 52.

    , , , , , , . Identification of the KDR tyrosine kinase as a receptor for vascular endothelial growth factor Biochem Biophys Res Commun 1992 187: 1579–1586

  53. 53.

    , , , , , , , , , , , , . Thalidomide and its analogs overcome drug resistance of human multiple myeloma cells to conventional therapy Blood 2000 96: 2943–2950

  54. 54.

    , , . Basic fibroblast growth factor–syndecan complex at cell surface or immobilized to matrix promotes cell growth J Biol Chem 1992 267: 17606–17610

  55. 55.

    , . Syndecan-1 (CD 138) in myeloma and lymphoid malignancies: a multifunctional regulator of cell behavior within the tumor microenvironment Leuk Lymphoma 1999 34: 35–43

  56. 56.

    , , . The cell surface proteoglycan syndecan-1 mediates fibroblast growth factor-2 binding and activity J Cell Physiol 1998 174: 310–321

Download references

Acknowledgements

This work was supported by National Institute of Health Grant PO-1 78378 and the Doris Duke Distinguished Clinical Research Scientist Award (KCA), Multiple Myeloma Research Foundation and the National Institute of Health Career Development Award (SPT).

Author information

Affiliations

  1. Department of Adult Oncology, Dana Farber Cancer Institute, Boston, MA, USA

    • D Gupta
    • , SP Treon
    • , Y Shima
    • , T Hideshima
    • , K Podar
    • , YT Tai
    • , B Lin
    • , S Lentzsch
    • , FE Davies
    • , D Chauhan
    • , RL Schlossman
    • , P Richardson
    •  & KC Anderson
  2. Department of Medicine, Harvard Medical School, Boston, MA, USA

    • D Gupta
    • , SP Treon
    • , Y Shima
    • , T Hideshima
    • , K Podar
    • , YT Tai
    • , B Lin
    • , S Lentzsch
    • , FE Davies
    • , D Chauhan
    • , RL Schlossman
    • , P Richardson
    •  & KC Anderson
  3. Genentech Inc., South San Francisco, CA, USA

    • P Ralph
  4. Celgene Corporation, Warren, NJ, USA

    • L Wu
    • , F Payvandi
    • , G Muller
    •  & DI Stirling

Authors

  1. Search for D Gupta in:

  2. Search for SP Treon in:

  3. Search for Y Shima in:

  4. Search for T Hideshima in:

  5. Search for K Podar in:

  6. Search for YT Tai in:

  7. Search for B Lin in:

  8. Search for S Lentzsch in:

  9. Search for FE Davies in:

  10. Search for D Chauhan in:

  11. Search for RL Schlossman in:

  12. Search for P Richardson in:

  13. Search for P Ralph in:

  14. Search for L Wu in:

  15. Search for F Payvandi in:

  16. Search for G Muller in:

  17. Search for DI Stirling in:

  18. Search for KC Anderson in:

Corresponding author

Correspondence to KC Anderson.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/sj.leu.2402295

Further reading