Hypoxia-inducible transcription factor-1 (HIF-1α) is overexpressed in multiple myeloma (MM) cells within the hypoxic microenvironment. Herein, we explored the effect of persistent HIF-1α inhibition by a lentivirus short hairpin RNA pool on MM cell growth either in vitro or in vivo and on the transcriptional and pro-angiogenic profiles of MM cells. HIF-1α suppression did not have a significant impact on MM cell proliferation and survival in vitro although, increased the antiproliferative effect of lenalidomide. On the other hand, we found that HIF-1α inhibition in MM cells downregulates the pro-angiogenic genes VEGF, IL8, IL10, CCL2, CCL5 and MMP9. Pro-osteoclastogenic cytokines were also inhibited, such as IL-7 and CCL3/MIP-1α. The effect of HIF-1α inhibition was assessed in vivo in nonobese diabetic/severe combined immunodeficiency mice both in a subcutaneous and an intratibial MM model. HIF-1α inhibition caused a dramatic reduction in the weight and volume of the tumor burden in both mouse models. Moreover, a significant reduction of the number of vessels and vascular endothelial growth factors (VEGFs) immunostaining was observed. Finally, in the intratibial experiments, HIF-1α inhibition significantly blocked bone destruction. Overall, our data indicate that HIF-1α suppression in MM cells significantly blocks MM-induced angiogenesis and reduces MM tumor burden and bone destruction in vivo, supporting HIF-1α as a potential therapeutic target in MM.
Multiple myeloma (MM) is characterized by the accumulation of malignant plasma cells in the bone marrow (BM) microenvironment that critically supports MM cell growth and survival.1 Particularly, an increase of BM angiogenesis and osteoclastogenesis occurs with MM cell growth and proliferation and plays a critical role in the pathophysiology and progression of MM.2, 3, 4, 5, 6, 7, 8
The angiogenic switch in MM is mainly sustained by the overproduction of vascular endothelial growth factors (VEGFs) by MM cells and the BM microenvironment.2, 9, 10 Other pro-angiogenic molecules, such as basic fibroblast growth factor,11, 12 interleukin-8 (IL-8),13 angiopoietin-1 14 and osteopontin15 are involved in the pro-angiogenic process. Hypoxia is a common feature of solid tumors and is associated with angiogenesis and malignant phenotype. It is critical to the regulation of the angiogenic switch in those cancers, by regulating the production of the pro-angiogenic molecules.16, 17, 18 Tumor adaptation to hypoxia is mainly due to the hypoxia-inducible factor (HIF)-1, a key transcription factor that regulates pro-angiogenic factors, angiogenesis and tumor progression in solid tumors.19, 20, 21, 22, 23 HIF-1 is a heterodimeric DNA binding complex highly inducible by hypoxia and is composed of two basic helix-loop-helix proteins, including the constitutive HIF-1β subunit and the hypoxia-inducible α-subunit.19, 20, 21 Under normoxic conditions HIF-1α has a very short half-life undergoing proteosomal degradation by oxygen-dependent hydroxylation. In contrast, under hypoxic condition, hydroxylation is suppressed and HIF-1α protein escapes proteasomal destruction and can accumulate and translocate to the nucleus.19, 20, 21 HIF-1α is a critical trigger and regulator of tumor associated angiogenesis.22, 23 Interestingly a critical role of HIF-1α in osteoclast formation and regulation has been shown24, 25 as well as its role in the development of osteolytic metastasis in breast cancer model.26
Recently it has been demonstrated that the BM microenvironment is hypoxic in MM patients27 and that HIF-1α is overexpressed by MM cells, and modulates their transcriptional and pro-angiogenic profiles.27, 28, 29 Moreover, the role of hypoxia in tumor progression and dissemination has been demonstrated in MM mouse models.30, 31 These findings suggest that hypoxia could be a target in MM. The role of HIF-1α as a therapeutic target in MM is under investigation. In this study, we determined the effect of stable HIF-1α inhibition in MM cells on cell proliferation and survival in vitro and growth, angiogenesis and osteolysis in vivo using plasmocytoma xenograft or intratibial mouse models previously used to evaluate the effect on myeloma cell growth and bone destruction.32, 33, 34
Materials and methods
Cells and cell culture conditions
Human myeloma cell lines (HMCLs) JJN3, Roswell Park Memorial Institute medium (RPMI)-8226 and OPM-2 were purchased from DSMZ (Braunschweig, Germany), while the U266 was obtained from the ATCC (LGC Standards S.r.l., Venezia, Italy). HMCLs were cultured in RPMI media at 10% fetal bovine serum with 2 nM of glutamine and antibiotics (Invitrogen Life Technologies, Milan, Italy).
Lentivirus short hairpin RNA (shRNA) pool anti-HIF-1α (Sigma-Aldrich, Milan, Italy) was used for HIF-1α stable knockdown in HMCLs, whereas the pLKO.1 lentiviral vector was the empty control. Recombinant lentivirus was produced by transient transfection of 293T cells following a standard protocol. HMCLs were infected as previously described32, and selected in culture by the presence of 4 ug per ml puromycin for 21 days. Selected clones of HMCLs were then screened for HIF-1α, HIF-1β, HIF-2α and HIF-3α mRNA and/or protein expression. Stably transfected HMCLs were maintained in RPMI medium containing 10% of fetal bovine serum with 4 ug per ml puromycin until use.
Hypoxic and drugs treatments
HMCLs stably transfected with pLKO.1 or anti-HIF-1α shRNA were incubated in the presence or absence of hypoxic conditions (1%O2, 5%CO2) or treated with the hypoxic mimetic drug cobalt chloride at 100 μM (Sigma-Aldrich, St Louis, MO, USA) or vehicle for 12–24 h. In selected experiments HMCLs stably transfected with pLKO.1 or anti-HIF-1α were treated either in normoxic or in hypoxic conditions in the presence of the absence of Bortezomib (supplied from Janssen-Cilag; Milan, Italy) at concentrations ranging from 1–50 nM of Lenalidomide (supplied by Celgene Italy srl, MiIan, Italy) (0.2–10 mM) or vehicle (dimethyl sulfoxide) for 24–72 h.
Cell proliferation and viability assays
HMCLs transfected with pLKO.1 or anti-HIF-1α were cultured in 96-well microtiter plates for 48–96 h in the presence of 3H-thymidine (3H-TdR) (Biocompare South San Francisco, CA, USA) and thymidine incorporation was detected by liquid scintillation spectroscopy (1205 Betaplate; Wallac; Markham, ON, Canada). Viability of HMCLs stably transfected with pLKO.1 or anti-HIF-1α was evaluated under both normoxic and hypoxic conditions after 24–72 h of culture by adapted 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test assay (Cell Counting Kit-8; Alexis, Vinci-Biochem srl, Italy). Cell apoptosis was determined by Apo 2.7 mAb staining (Immunotech, Fullerton, CA, USA) and verified by FACScan, (BD Biosciences Italy, Milan, Italy).
In vivo studies
Four to six-week-old males and females severe combined immunodeficiency/nonobese diabetic (SCID-NOD) mice (Harlan Laboratories, Udine, Italy) were housed under specific pathogen-free conditions. All procedures involving animals were performed in accordance with the National and International current regulations (D.l.vo 27/01/1992, n.116, European Economic Community Council Directive 86/609, OJL 358, Dec. 1, 1987). Three groups of six animals each were injected subcutaneously with 5 × 106 JJN3 cells stably transfected with anti-HIF1α containing plasmid vectors (JJN3-anti-HIF-1α), or with JJN3 stably transfected with empty vector (JJN3-pLKO.1), or JJN3 wild type (JJN3). Twenty days after tumor cell inoculation, mice were killed and autopsies were performed. Tumor mass was measured as previously described.32 Maximum length and width of the tumor masses were measured with a caliper, and tumor volume (mm3) was calculated according to the following formula: 0.523 × length × width2. Tumors were removed and subjected to immunohistochemical staining.
In a separate set of experiments SCID and NIH-III nude mice (4 weeks of age) were injected intratibially with 20 μl of 5 × 104 cells of JJN3-anti-HIF-1α or JJN3-pLKO.1 or JJN3 or saline alone. All research protocols were approved by the Pittsburgh VA Healthcare System Institutional Animal Care and Use Committee. Four weeks after injection, the animals were killed and the tibias were dissected out. Images of dissected tibias were acquired on a vivaCT 40 scanner (Scanco Medical, Brüttisellen, Switzerland) at resolution of 21 μm isotropic, reconstructed, and segmented for 3-dimensional display using the instrument's analysis algorithm software. Tissue samples and cell extracts were obtained for immunohistochemical staining.
Gene expression profiling and microarray analysis
The transcriptional profiles of JJN3-anti-HIF-1α and JJN3-pLKO.1 cells exposed to hypoxic or normoxic conditions were analyzed. To perform gene expression profiles, total RNA was purified using the RNeasy total RNA Isolation Kit (Qiagen, Valencia, CA, USA). Preparation of biotin-labeled complementary RNA, hybridization to GeneChip Human Genome U133 Plus2.0 Arrays and scanning (GeneChip Scanner 3000 7G, Affymetrix Inc., Santa Clara, CA, USA) were performed according to manufacturer’s protocols. The raw intensity signals were extracted from cell intensity (CELL) files and normalized using the robust multi-chip average (RMA) package for Bioconductor and custom GeneAnnot-based Chip Definition Files version 2.2.0 in R software. The most differentially expressed genes between two experimental conditions were defined as follows: for each gene, the ratio between the difference and the average value in the expression signals under the two conditions was calculated. Those genes with a minimum of 1.5 absolute fold change between hypoxic and normoxic conditions were selected for further analysis. Then, those genes with ratios exceeding two s.d. from the mean were considered differentially expressed. NetAffx (https://www.affymetrix.com/analysis/netaffx/) and Database for Annotation, Visualization and Integrated Discovery (DAVID) (http://david.abcc.ncifcrf.gov) tools were used for the functional annotation studies of the selected lists. The data have been deposited at NCBI Gene Expression Omnibus repository (http://www.ncbi.nlm.mih.gov/geo) and are accessible through Gene Expression Omnibus series accession number GSE40326.
RNA isolation and reverse-transcriptase polymerase chain reaction amplification
Total RNA was extracted from the cells using the RNeasy total RNA isolation kit (Qiagen, Valencia, CA, USA), and then quantified using a Nanodrop ND-100 (Celbio S.p.A., Milan, Italy). 1 μg of RNA was reverse- transcribed with 400 U Moloney murine leukemia reverse transcriptase (Applied Biosciences, Life Sciences, Carlsbad, CA, USA) in accordance with the manufacturer’s protocol. The complementary DNAs were amplified by means of PCR using the following specific primer pairs:
HIF1A: F: 5′-IndexTermCTCAAAGTCGGACAGCCTCA-3′; R: 5′-IndexTermCCCTGCAGTAGGTTTCTGCT-3′
VEGFA: F:5′-IndexTermCGAAGTGGTGAAGTTCATGGATG-3′; R: 5′-IndexTermTTCTGTTCAGTCTTTCCTGGTGAG-3′
IL8: F: 5′-IndexTermTACTCCAAACCTTTCCACCC-3′; R: 5′-IndexTermAACTTCTCCACAACCCTCTG-3′
MMP9: F: 5′-IndexTermCGCAGACATCGTCATCCAGT-3′; R: 5′-IndexTermGGATTGGCCTTGGAAGATGA-3′
CCL2: F: 5′-IndexTermACTGAAGCTCGTACTCTC-3′; R: 5′-IndexTermCTTGGGTTGTGGAGTGAG-3′
GAPDH: F 5'-IndexTermCAACGGATTTGGTCGTATTG-3'; R: 5'-IndexTermGGAAGATGGTGATGGGATTT-3'
Annealing temperature: HIF1A: 59 °C; VEGFA: 66 °C; IL8: 64 °C; CCL2: 60 °C MMP9: 58 °C; GAPDH: 58 °C. Product size: HIF1A: 454 bp; VEGFA: 375 bp; IL8: 158 bp; CCL2: 354 bp MMP9: 369 bp GAPDH: 209 bp. Pictures of the electrophoresed complementary DNAs were recorded with a digital DC 120 Kodak camera and quantified.
Real-time quantitative PCR and angiogenesis PCR array
Real-time PCR was performed using the TaqMan Gene Expression Assay (Applied Biosystems Applera, Milan, Italy) for the following genes:
HIF-1B (ARNT): HS00231048_M1
HIF2A (EPAS1): HS01026149_M1
HIF2B (ARNT2): HS00208298_M1
The PCR amplifications were performed in duplicate using the iCycler iQ Real-Time Detection System (Bio-Rad, Milan, Italy). The comparative Ct method was applied to normalize the differences in the quantity and quality of RNA, and mRNA was quantified using the comparative ΔCt method using the endogenous reference gene ABL (ΔCt=mean Ct gene–mean Ct ABL); ΔΔCt was evaluated as the difference between the ΔCt of a sample and the ΔCt of the control. The fold change in mRNA expression was calculated as 2−ΔΔCt.
The expression levels of the pro-angiogenic molecules were evaluated on mRNA extracted from tumors removed from SCID-NOD mice with the Human Angiogenesis RT2 Profiler PCR Array and RT2 Real-Timer SyBR green/ROX PCR Mix (PAHS-024, Superarray, SABiosciences, Frederick, MD, USA) that profiles the expression of 84 key genes involved in modulating the biological processes of angiogenesis.
Western blot analysis
Western blot analysis was performed as previously described.26 Nuclear extracts were prepared using the Nuclear extraction kit (Active Motif, Carlsbad, CA, USA) according to the manufacturer's protocol. A total of 40 μg of nuclear extracts were tested. A polyclonal goat anti-HIF-1α Ab (1:1000; R&D system, Minneapolis, MN, USA) or anti-HIF-1β or anti-HIF-2α or anti-HIF-3α (1:500; Novus Biologicals, Littleton, CO, USA) were used to detect HIF-1α, HIF-1β, HIF-2α and HIF-3α, respectively, whereas anti-Histone H1 monoclonal Ab was used as internal control (Upstate, Lake Placid, NY, USA). A monoclonal anti-p27 (BD Pharmingen, San Jose, CA, USA) was used as primary antibody to detect p27 on whole lysates. Appropriate secondary antibodies that were used: anti-goat Immunoglobulin G (Rockland Immunochemicals, Gilberrtsville, PA, USA) for HIF-1α, anti-mouse Immunoglobulin G (BD Biosciences) for HIF-1β, HIF-2α and p27 and anti-rabbit Immunoglobulin G (Chemicon International, Millipore, Billerica, MA, USA) for HIF-3α.
HIF-1α activation in nuclear extracts of HMCLs was evaluated using an Enzyme-linked immunosorbent assay based method (TransAM HIF-1 kit, Active Motif) according to the manufacturer’s procedures.
Enzyme-linked immunosorbent assays
Soluble VEGF, IL-8, IL-7 and CCL3 proteins were detected in the conditioned media of HMCLs using Enzyme-linked immunosorbent assays purchased from R&D system according to the manufacturer’s protocols. Cytokine levels in the conditioned media were normalized to the number of cells at the end of culture period.
Histological and Immunohistochemical analysis
Tissue samples obtained from tumors removed from mice injected either subcutaneously or intratibially with JJN3-pLKO.1, JJN3-anti-HIF-1α and JJN3 were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 3 μm, and stained with hematoxylin and eosin or with Toluidine blu and Gomori’s three-chromic methods. Tumors obtained from intratibial injections underwent EDTA decalcification before embedding in paraffin. On the basis of cell location and morphology, the number of all Osteoclasts (OCLs) and active prismatic osteoblasts was evaluated on the bone surface of each section (3 × 10 mm2). Osteocyte number and vitality were recorded on a total of 500 lacunae per histological section.
Sections were immuno-stained either with mouse monoclonal anti-HIF-1α Ab (NOVUS Biologicals Littleton, CO; Working dilution 1:100) or mouse anti-human VEGF (R&D; dilution 1:20) or with 1:100 diluted mouse anti-Ki67 primary antibody (Clone MIB-1, Dako, Carpinteria, CA, USA) for 30 min. HIF-1α and VEGF staining were revealed using the UltraVision LP Large Volume Detection System HRP polymer (Thermo Scientific, Erembodegem, Belgium) and quantified according to semiquantitative immunohistochemical score. Detection of Ki67 was performed using a high-sensitive detection system (Advance-HRP, Dako Carpinteria, CA, USA) and 3,3-diaminobenzidine was used as chromogen substrate. Angiogenesis was evaluated on frozen tissues samples obtained from both series of mice. Tissues were fixed in acetone and treated with rabbit anti-mouse CD34 (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, USA). After washing sections were incubated with a secondary antibody (1:250; Rat anti-Immunoglobulin G horseradish peroxidase; Millipore) and reaction revealed with a solution of 3-3′-diaminobenzidine tetrahydrochloride (liquid DAB substrate chromogen system, DAKO, Glostrup, DK).
Permanent HIF-1α silencing in HMCLs: effects on MM cell proliferation, survival and drug sensitivity
HMCLs were initially checked for HIF1A mRNA and HIF-1α protein expression under normoxic conditions (Figure 1a). All HMCLs tested expressed HIF1A mRNA, whereas HIF-1α protein was present at low levels in JJN3 and U266, but not in RPMI-8226 and OPM-2 (Figure 1a). All HMCLs overexpressed HIF-1α protein upon hypoxic treatment. The JJN3 and RPMI-8226 were used as cell lines for HIF-1α inhibition studies. The selected clones were screened both under normoxic and hypoxic conditions for HIF1A mRNA by qualitative PCR (Figure 1b), and for HIF1A, BNIP3, HIF1B, HIF2A, HIF2B and HIF3A mRNA levels by real-time PCR. These studies showed a selective inhibition of HIF1A and its target gene BNIP3 by the shRNA pool used (Figure 1c left panel). HIF1B, HIF2A, HIF2B and HIF3A mRNA were not inhibited by the shRNA pool (Figure 1c). Western blot for HIF-1α, HIF-1β and HIF-3α proteins showed specific inhibition of HIF-1α protein under both normoxia and hypoxia (Figure 1d). HIF-2α and HIF-2 β were not expressed either in normoxia or upon treatment with hypoxic mimetic (data not shown). Consistently HIF-1α activity was significantly inhibited in JJN3 anti-HIF-1α as compared with JJN3 pLKO.1 under hypoxic conditions (Figure 1e). Similar results were obtained with RPMI-8226 (data not shown).
The effect of HIF-1α suppression on HMCLs proliferation and survival was next investigated. Interestingly, there was no significant inhibition of cell proliferation either with JJN3 or RPMI-8226 (Figure 2a) at different time points. Cell viability was not significantly changed by HIF-1α suppression when checked at 12 and 24 h either under normoxia or in hypoxia (Figure 2b). A statistically significant effect on cell viability was observed in normoxia condition at 72 h for JJN3 cells (−23%; P=0.01) (Figure 2c upper panel) and maintained at the same entity for longer culture period, but not for RPMI-8226 cells (Figure 2c, lower panel). Treatment with lenalidomide at a wide range of concentrations induced a significantly higher inhibition of cell proliferation in JJN3 anti-HIF-1α as compared with JJN3 pLKO.1 (P=0.01) (Figure 2d) without changing their viability significantly. Consistently p27 expression was increased by lenalidomide in JJN3 anti-HIF-1α as compared with JJN3 pLKO.1 (Figure 2e). On the other hand, Bortezomib (4–10 nM) induced a similar rate of cell proliferation and cell death in JJN3 anti-HIF-1α and JJN3 pLKO.1 (data not shown).
HIF-1α suppression affects the transcriptional profile of HMCLs: inhibitory effect on the pro-angiogenic and pro-osteoclastogenic genes
The transcriptional profiles of JJN3 cells transduced with shRNA anti-HIF-1α (JJN3-anti-HIF-1α) were compared with those infected with the control vector pLKO.1 (JJN3-pLKO.1) either under hypoxic or normoxic conditions. Among the significantly modulated genes (326 and 361 genes under hypoxic and normoxic condition, respectively) (Supplementary Table S1), we found downregulation of the pro-angiogenic molecules VEGFA, IL8, CCL2, MMP9 in JJN3-anti-HIF-1α cells under both hypoxic and normoxic conditions. Microarray data were further validated by both qualitative PCR (Figure 3a) and real-time quantitative PCR, which showed that VEGFA and IL8 were induced by hypoxia and inhibited by HIF-1α suppression (Figure 3b), whereas CCL2 mRNA was not induced by hypoxia but was inhibited by HIF-1α suppression either under normoxia or hypoxia. Finally, other genes with pro-angiogenic properties, such as IL10 and CCL5, were inhibited by HIF-1α suppression in JJN3 under both normoxic and hypoxic conditions (Supplementary Table S1). Consistent with our previous observation on small interfering RNA-mediated HIF-1α inhibition,27 stable HIF-1α suppression by shRNA significantly blunted vessels formation induced by the conditioned media of HMCLs as determined by AngioKit (TCS Biologies, London, UK) (data not shown).
The effect of HIF-1α suppression on the production of pro-osteoclastogenic cytokines by HMCLs was then investigated. A significant inhibitory effect was observed for both MIP1A/CCL3 and IL7 mRNA levels either under normoxia or in hypoxia as shown for JJN3 (Figure 3c). This inhibitory effect was also confirmed at translational level under normoxia as shown for JJN3 (Figure 3d). In contrast, no significant effect was observed on the expression of the osteoblast inhibitor gene DKK1 by JJN3 with HIF-1α suppression (data not shown).
HIF-1α suppression in JJN3 cells blocks the growth of subcutaneous MM in SCID-NOD mice and inhibits angiogenesis
We next investigated whether inhibition of HIF1α in JJN3 may influence tumor growth in vivo. Therefore, we analyzed the tumorigenicity of JJN3 wild type, JJN3 transfected with a plasmid with silenced HIF1α (JJN3-anti-HIF-1α) and JJN3 with empty vector (JJN3-pLKO.1) cells injected subcutaneously into SCID-NOD animals. Twenty days after cell inoculation, mice were killed, tumors removed and measured. At this time point, all animals developed tumors that grew at the site of injection in the absence of metastases to distant sites. Mice injected with the JJN3-anti-HIF-1α cells developed significantly smaller tumors than mice inoculated with the JJN3-pLKO.1 (P=0.00018) or with JJN3 (P=0.032) (Figure 4a and b). Both weight and volume of the tumors in JJN3-anti-HIF-1α mice were significantly reduced as compared with JJN3-pLKO.1 (Figure 4c and d). The median weight of tumors formed by JJN3-anti-HIF-1α was 0.05 g vs 0.3 g for JJN3-pLKO (P=0.0007) (Figure 4c). The median volume of tumors formed by JJN3-anti-HIF-1α was 72.6 mm3 (range 3–221 mm3) and that of JJN3-pLKO.1 was 369.8 mm3 (range 221.8–671 mm3) (P=0.0012) (Figure 4d). Interestingly in the tumoral mass removed from JJN3-anti-HIF-1α mice the microvascular density (number of vessels positive for CD34/mm3) was significantly reduced as compared with that obtained from JJN3-pLKO.1 (JJN3-anti-HIF-1α vs. JJN3-pLKO.1: −76%; P=0.003). Similarly VEGF immunostaining was reduced in JJN3-anti-HIF-1α derived tumors as compared with JJN3 pLKO.1 injected mice (Figure 4e and f).
HIF-1α suppression in JJN3 cells blunts angiogenesis and bone destruction in an intratibial mouse model
The effect of HIF-1α suppression in MM cells was further investigated in vivo in intratibial mouse model. The SCID mice were injected intratibially with saline or JJN3-pLKO.1 or JJN3-anti-HIF-1α cells, and lytic lesions were allowed to develop for 2–4 weeks before the mice were analyzed. Histological analysis showed a significant reduction in the tumor mass with an increase in the bone thickness in JJN3-anti-HIF-1α as compared with JJN3 pLKO.1 injected animals (Figure 5a). However, immunostaining for Ki67 showed a reduction in the number of Ki67-positive cells in JJN3-anti-HIF-1α as compared with JJN3 pLKO.1 mice although the difference did not reach statistical significance (Figure 5b). However, immunostaining for CD34 showed a significant reduction in the number of CD34-positive vessels in JJN3-anti-HIF-1α as compared with JJN3 pLKO.1 mice (median number of vessels positive for CD34/mm3 22 vs 64, P=0.001) (Figure 5c) similarly to what observed in the subcutaneous model. Finally, micro-QCT analysis demonstrated that mice injected with JJN3-pLKO.1 cells started developing detectable lytic lesions at 2 weeks after cell injection with continued further bone deterioration through the 4 weeks, that ultimately involved the entire tibia leading to animal death from advanced disease. In contrast, the saline injected controls at 4 weeks were similar to the 0-week time point, demonstrating that the effects detected were not the result of the injection process. Interestingly, mice injected with JJN3-anti-HIF-1α cells showed a dramatic reduction of osteolytic lesions (Figure 5d). Finally, histological analysis showed a significant reduction in the number of OCLs in JJN3-anti-HIF-1α as compared with JJN3-pLKO.1 bone sections (P<0.05), whereas the number of active prismatic osteoblasts arranged in laminae did not significantly changed as well as that of viable osteocytes for 500 lacunae (P=0.4; NS, data not shown).
It is well established that intratumoral hypoxia and consequently HIF-1α activation in solid tumors critically trigger the angiogenic switch, and induce modifications of cancer cell metabolism leading to tumor progression and metastasis.16, 17, 18, 22 HIF-1α is overexpressed in many tumors22, 23 including some hematological malignancies such as MM.27, 28, 29, 35 We and others27, 35 have recently shown that BM microenvironment is hypoxic both in MM patients27 and in mice.35 In addition HIF-1α also may be overexpressed by MM cells under normoxic conditions27, 28 dependent at least in part on c-myc upregulation.27 Finally, it has been reported that hypoxia modulates the expression of pro-angiogenic genes such as VEGF and IL-827 by MM cells, and that HIF-1α is involved in MM-induced angiogenesis in vitro.27 In turn, angiogenesis is able to support MM cell growth and survival.2 All these results suggest that HIF-1α could be a potential therapeutic target for MM. To test this hypothesis, we performed HIF-1α knockdown in HMCLs using a pool of shRNA anti-HIF-1α and evaluated the effect permanent suppression of HIF-1α on MM cell proliferation, survival, the pro-angiogenic profile and tumor growth in vivo. Firstly, we showed that the inhibition of HIF-1α by shRNA was specific because no significant inhibitory effect was demonstrated on HIF1B, HIF2A, HIF2B and HIF3A expression. Interestingly, we did not find a significant inhibitory in vitro effect on MM cell proliferation by HIF-1α suppression only showing a significant inhibitory effect on cell survival after 72 h in JJN3, but not in RPMI-8226. These results suggest that HIF-1α inhibition did not directly induce a significant effect on MM cell proliferation and survival. Similarly, it has been previously reported that small interfering RNA anti-HIF-1α alone is not able to directly affect MM cell survival, but enhanced MM cell sensitivity to melphalan.36 Therefore, we further evaluated the potential effect of HIF-1α stable suppression on the sensitivity of HMCLs to IMiDs and Bortezomib in vitro. Interestingly, we found that shRNA anti-HIF-1α increased the inhibitory effect of IMiDs on HMCL proliferation. This result was also supported by the previous finding that the response to lenalidomide treatment involves oxidative stress response in MM cells.37 In contrast, we did not find any sensitization of MM cells to Bortezomib treatment. It has been reported that anti-MM effects of Bortezomib may be increased by hypoxia and HIF-1α upregulation38 rather than by HIF-1α inhibition.
Although, we did not find a significant effect on cell proliferation and survival by knockdown of HIF-1α, stable HIF-1α suppression by shRNA modified the pro-angiogenic and osteoclastogenic profiles of MM cells either under hypoxic or normoxic conditions. Microarray analysis and real-time PCR showed that HIF-1α suppression significantly inhibited some pro-angiogenic factor genes such as VEGFA, IL-8, CCL2, MMP9. The capacity of HIF-1α to regulate VEGFA and IL8 gene expression by MM cells confirmed our previous data,27, 39 and supports HIF-1α as a key regulator of angiogenesis also in MM cells because VEGF9, 10 and at least in part of IL-839, 40 are well-known factors involved in MM-induced angiogenesis. CCL2 and MMP9 could be also involved in the pro-angiogenic role of HIF-1α, because both cytokines are produced by MM cells and contribute to angiogenesis and BM spreading.41, 42, 43
Interestingly, our data suggest that other than angiogenic cytokines, HIF-1α suppression affects expression and production of MM-derived pro-osteoclastogenic factors that are known to be involved in osteoclast formation and bone destruction such as CCL3/MIP1α44 and IL-7.45 To the best of our knowledge, this is the first time that the capacity of HIF-1α has been reported to regulate pro-osteoclastogenic cytokine in MM cells, although the evidence that CCL3/MIP1α and IL-7 are potential target genes of HIF-1α has been shown in other cell types such as acute myeloid leukemia cells for CCL3/MIP1α46 and mesenchymal/osteoblastic cells for IL-7.47
To evaluate the role of HIF-1α as a potential target in MM we used two different in vivo models: the subcutaneously and the intratibial model. A dramatic effect on tumor burden was demonstrated in both models by HIF-1α suppression showing a significant reduction of tumor mass. Because of the lack of a significant effect of HIF-1α suppression on cell proliferation and survival of HMCLs in vitro, we guess that the anti-MM in vivo effect was mainly due to the inhibitory effect on angiogenesis. This was supported by the capacity of HIF-1α to regulate the pro-angiogenic profiles of HMCLs in vitro and by our previous results showing that HIF-1α suppression blocks MM-induced in vitro angiogenesis.27 The significant inhibition of angiogenesis (evidenced by reduction in the number and length of vessels in mice injected with JJN3-anti-HIF-1α) confirmed that the antitumor effect of HIF-1α suppression was mainly due to the inhibition of angiogenesis. In support of these data, the inhibitory effect on angiogenesis by HIF-1α suppression was accompanied by a reduction of VEGF immunostaining and of the proliferative index, Ki67, that could be related to angiogenesis in MM,6 although it did not reach a statistical significance. Finally, in line with our data, it has been previously reported that HIF-1α inhibition by RNA interference inhibits in vivo angiogenesis in mice in other tumor models.48
In addition to the inhibition of angiogenesis, we showed in the intratibial bone model that HIF-1α suppression also affects MM-induced in vivo formation of osteolytic bone lesions. This could be due in part to the inhibitory effect on MM growth via the inhibition of angiogenesis, but also to a direct effect on the production of pro-osteoclastogenic cytokines as shown in vitro. Accordingly, histological analysis showed a significant reduction of the number of OCLs in JJN3 anti-HIF-1α as compared with JJN3-pLKO.1.
The capacity of HIF-1α suppression to block the development of osteolytic lesions in the intratibial MM model may explain in turn its anti-MM effect, because it is well-known that osteoclast activation supports MM cells growth in vivo.7 Similarly, it has been previously reported that hypoxia and HIF-1α lead to the development of osteolytic lesions and tumor growth in breast cancer in vivo models.26
In conclusion, as in other tumoral models where it has been shown that HIF-1α inhibition exerts a potent in vivo antitumor effect, including the squamous cell carcinoma, liver and lung cancer,48, 49 our data indicate that selective HIF-1α inhibition results in a potent anti-MM effect by blocking angiogenesis and the development of osteolytic bone lesions. These results suggest that HIF-1α is a potential therapeutic target in MM. In this context, it has been recently reported that adaphostin inhibits HIF-1α expression and induces a significant antiangiogenic and anti-MM activity in a MM xenograft mouse model.28 Furthermore, a growing number of novel anticancer agents have been demonstrated to inhibit HIF-1α with different molecular mechanisms including histone deacetylases, molecules acting on RAS-RAK-MEK-ERK pathways and PI3K-AKT, and mTOR inhibitors, that are also currently in phase I-II clinical trials in MM.1, 29 Overall, our study provides the rationale for clinical trial with selective inhibitors of HIF-1α mRNA, such as EZN-296850 in MM patients.
Podar K, Richardson PG, Hideshima T, Chauhan D, Anderson KC . The malignant clone and the bone-marrow environment. Best Pract Res Clin Haematol 2007; 20: 597–612.
Vacca A, Ribatti D, Presta M, Minischetti M, Iurlaro M, Ria R et al. Bone marrow neovascularization, plasma cell angiogenic potential, and matrix metalloproteinase-2 secretion parallel progression of human multiple myeloma. Blood 1999; 93: 3064–3073.
Vacca A, Ribatti D . Bone marrow angiogenesis in multiple myeloma. Leukemia 2006; 20: 193–199.
Rajkumar SV, Leong T, Roche PC . Prognostic value of bone marrow angiogenesis in multiple myeloma. Clin Cancer Res 2000; 6: 3111–3116.
Munshi NC, Wilson C . Increased bone marrow microvessel density in newly diagnosed multiple myeloma carries a poor prognosis. Semin Oncol 2001; 28: 565–569.
Alexandrakis MG, Passam FH, Dambaki C, Pappa CA, Stathopoulos EN . The relation between bone marrow angiogenesis and the proliferation index Ki-67 in multiple myeloma. J Clin Pathol 2004; 57: 856–860.
Yaccoby S . Advances in the understanding of myeloma bone disease and tumour growth. Br J Haematol 2010; 149: 311–321.
Roodman GD . Targeting the bone microenvironment in multiple myeloma. J Bone Miner Metab 2010; 28: 244–250.
Dankbar B, Padro T, Leo R, Feldmann B, Kropff M, Mesters RM et al. Vascular endothelial growth factor and interleukin-6 in paracrine tumor-stromal cell interactions in multiple myeloma. Blood 2000; 95: 2630–2636.
Podar K, Anderson KC . The pathophisiologic role of VEGF in hematologic malignancies: therapeutic implications. Blood 2005; 105: 1383–1395.
Bisping G, Leo R, Wenning D, Dankbar B, Padró T, Kropff M et al. Paracrine interactions of basic fibroblast growth factor and interleukin-6 in multiple myeloma. Blood 2003; 101: 2775–2783.
Colla S, Morandi F, Lazzaretti M, Polistena P, Svaldi M, Coser P et al. Do human myeloma cells directly produce basic FGF? Blood 2003; 102: 3071–3072.
Shapiro VS, Mollenauer MN, Weiss A . Endogenous CD28 expressed on myeloma cells up-regulates interleukin-8 production: implications for multiple myeloma progression. Blood 2001; 98: 187–193.
Giuliani N, Colla S, Lazzaretti M, Sala R, Roti G, Mancini C et al. Pro-angiogenic properties of human myeloma cells: production of angiopoietin-1 and its potential relationship to myeloma-induced angiogenesis. Blood 2003; 102: 638–645.
Colla S, Morandi F, Lazzaretti M, Rizzato R, Lunghi P, Bonomini S et al. Human myeloma cells express the bone regulating gene Runx2/Cbfa1 and produce osteopontin that is involved in angiogenesis in multiple myeloma patients. Leukemia 2005; 19: 2166–2176.
Brahimi-Horn MC, Chiche J, Pouysségur J . Hypoxia and cancer. J Mol Med 2007; 85: 1301–1307.
Rankin EB, Giaccia AJ . The role of hypoxia-inducible factors in tumorigenesis. Cell Death Differ 2008; 15: 678–685.
Liao D, Johnson RS . Hypoxia: a key regulator of angiogenesis in cancer. Cancer Metastasis Rev 2007; 26: 281–290.
Weidemann A, Johnson RS . Biology of HIF-1α. Cell Death Differ 2008; 15: 621–627.
Lisy K, Peet DJ . Turn me on: regulating HIF transcriptional activity. Cell Death Differ 2008; 15: 642–649.
Yee Koh M, Spivak-Kroizman TR, Powis G . HIF-1 regulation: not so easy come, easy go. Trends Biochem Sci 2008; 33: 526–634.
Semenza GL . Regulation of cancer cell metabolism by hypoxia-inducible factor 1. Semin Cancer Biol 2009; 19: 12–16.
Hirota K, Semenza GL . ReguIation of angiogenesis by hypoxia-inducibIe factor 1. Crit Rev OncoI Hematol 2006; 59: 15–26.
Utting JC, Flanagan AM, Brandao-Burch A, Orriss IR, Arnett TR . Hypoxia stimulates osteoclast formation from human peripheral blood. Cell Biochem Funct 2010; 28: 374–380.
Bozec A, Bakiri L, Hoebertz A, Eferl R, Schilling AF, Komnenovic V et al. Osteoclast size is controlled by Fra-2 through LIF/LIF-receptor signalling and hypoxia. Nature 2008; 454: 221–225.
Hiraga T, Kizaka-Kondoh S, Hirota K, Hiraoka M, Yoneda T . Hypoxia and hypoxia-inducible factor-1 expression enhance osteolytic bone metastases of breast cancer. Cancer Res 2007; 67: 4157–4163.
Colla S, Storti P, Donofrio G, Todoerti K, Bolzoni M, Lazzaretti M et al. Low bone marrow oxygen tension and hypoxia-inducible factor-1α overexpression characterize patients with multiple myeloma: role on the transcriptional and proangiogenic profiles of CD138(+) cells. Leukemia 2010; 24: 1967–1970.
Zhang J, Sattler M, Tonon G, Grabher C, Lababidi S, Zimmerhackl A et al. Targeting angiogenesis via a c-Myc/hypoxia-inducible factor-1alpha-dependent pathway in multiple myeloma. Cancer Res 2009; 69: 5082–5090.
Martin SK, Diamond P, Gronthos S, Peet DJ, Zannettino AC . The emerging role of hypoxia, HIF-1 and HIF-2 in multiple myeloma. Leukemia 2011; 25: 1533–1542.
Azab AK, Hu J, Quang P, Azab F, Pitsillides C, Awwad R et al. Hypoxia promotes dissemination of multiple myeloma through acquisition of epithelial to mesenchymal transition-like features. Blood 2012; 119: 5782–5794.
Hu J, Handisides DR, Van Valckenborgh E, De Raeve H, Menu E, Vande Broek I et al. Targeting the multiple myeloma hypoxic niche with TH-302, a hypoxia-activated prodrug. Blood 2010; 116: 1524–1527.
Storti P, Donofrio G, Colla S, Airoldi I, Bolzoni M, Agnelli L et al. HOXB7 expression by myeloma cells regulates their pro-angiogenic properties in multiple myeloma patients. Leukemia 2011; 25: 527–537.
Campbell RA, Berenson JR . Animal models of multiple myeloma and their utility in drug discovery. Curr Protoc Pharmacol 2008; Chapter 14, Unit 14: 9.
D'Souza S, del Prete D, Jin S, Sun Q, Huston AJ, Kostov FE et al. Gfi1 expressed in bone marrow stromal cells is a novel osteoblast suppressor in patients with multiple myeloma bone disease. Blood 2011; 118: 6871–6880.
Asosingh K, De Raeve H, de Ridder M, Storme GA, Willems A, Van Riet I et al. Role of the hypoxic bone marrow microenvironment in 5T2MM murine myeloma tumor progression. Haematologica 2005; 90: 810–817.
Hu Y, Kirito K, Yoshida K, Mitsumori T, Nakajima K, Nozaki Y et al. Inhibition of hypoxia-inducible factor-1 function enhances the sensitivity of multiple myeloma cells to melphalan. Mol Cancer Ther 2009; 8: 2329–2338.
Colla S, Zhan F, Xiong W, Wu X, Xu H, Stephens O et al. The oxidative stress response regulates DKK1 expression through the JNK signaling cascade in multiple myeloma plasma cells. Blood 2007; 109: 4470–4477.
Veschini L, Belloni D, Foglieni C, Cangi MG, Ferrarini M, Caligaris-Cappio F et al. Hypoxia-inducible transcription factor-1 alpha determines sensitivity of endothelial cells to the proteosome inhibitor bortezomib. Blood 2007; 109: 2565–2570.
Colla S, Tagliaferri S, Morandi F, Lunghi P, Donofrio G, Martorana D et al. The new tumor-suppressor gene inhibitor of growth family member 4 (ING4) regulates the production of proangiogenic molecules by myeloma cells and suppresses hypoxia-inducible factor-1 alpha (HIF-1alpha) activity: involvement in myeloma-induced angiogenesis. Blood 2007; 110: 4464–4475.
Negaard HF, Iversen N, Bowitz-Lothe IM, Sandset PM, Steinsvik B, Ostenstad B et al. Increased bone marrow microvascular density in haematological malignancies is associated with differential regulation of angiogenic factors. Leukemia 2009; 23: 162–169.
Arendt BK, Velazquez-Dones A, Tschumper RC, Howell KG, Ansell SM, Witzig TE et al. Interleukin 6 induces monocyte chemoattractant protein-1 expression in myeloma cells. Leukemia 2002; 16: 2142–2147.
Van Valckenborgh E, Bakkus M, Munaut C, Noël A, Pierre Y, Asosingh K et al. Upregulation of matrix metalloproteinase-9 in murine 5T33 multiple myeloma cells by interaction with bone marrow endothelial cells. Int J Cancer 2002; 101: 512–518.
Purushothaman A, Chen L, Yang Y, Sanderson RD . Heparanase stimulation of protease expression implicates it as a master regulator of the aggressive tumor phenotype in myeloma. J Biol Chem 2008; 283: 32628–32636.
Choi SJ, Cruz JC, Craig F, Chung H, Devlin RD, Roodman GD et al. Macrophage inflammatory protein 1-alpha is a potential osteoclast stimulatory factor in multiple myeloma. Blood 2000; 96: 671–675.
Giuliani N, Colla S, Sala R, Moroni M, Lazzaretti M, La Monica S et al. Human myeloma cells stimulate the receptor activator of nuclear factor-kappa B ligand (RANKL) in T lymphocytes: a potential role in multiple myeloma bone disease. Blood 2002; 100: 4615–4621.
Hatfield KJ, Bedringsaas SL, Ryningen A, Gjertsen BT, Bruserud O . Hypoxia increases HIF-1α expression and constitutive cytokine release by primary human acute myeloid leukaemia cells. Eur Cytokine Netw 2010; 21: 154–164.
Nishida C, Kusubata K, Tashiro Y, Gritli I, Sato A, Ohki-Koizumi M et al. MT1-MMP plays a critical role in hematopoiesis by regulating HIF-mediated chemokine/cytokine gene transcription within niche cells. Blood 2012; 119: 5405–5416.
Krishnamachary B, Berg-Dixon S, Kelly B, Agani F, Feldser D, Ferreira G et al. Regulation of colon carcinoma cell invasion by hypoxia-inducible factor 1. Cancer Res 2003; 63: 1138–1143.
Li L, Lin X, Staver M, Shoemaker A, Semizarov D, Fesik SW et al. Evaluating hypoxia-inducible factor-1alpha as a cancer therapeutic target via inducible RNA interference in vivo. Cancer Res 2005; 65: 7249–7258.
Greenberger LM, Horak ID, Filpula D, Sapra P, Westergaard M, Frydenlund HF et al. A RNA antagonist of hypoxia-inducible factor-1alpha, EZN-2968, inhibits tumor cell growth. Mol Cancer Ther 2008; 7: 3598–3608.
This study was supported by grants from the ‘International Myeloma Foundation’; and ‘Associazione Italiana per la Ricerca sul Cancro’ A.I.R.C, IG2009 no. 8530 (N.G.), no. IG4569 (A.N.) and no. 13018 (I.A.); Special Program Molecular Clinical Oncology 5 per mille no. 9965 (N.G.) and no. 9980 (A.N.), the Multiple Myeloma Research Foundation and grant support R01 AR059679, R21 CA141426, R01AR057308, and Merit Review funds from the Veterans Administration. (G.D.R.). We thanks Ministero Italiano dell’ Istruzione, Università e Ricerca (MIUR) and Italian Association against Leukemia, Lymphoma and Myeloma (AIL). We thanks Carla Palumbo and Marzia Ferretti for their support in the histological analysis.
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Leukemia website
About this article
Expert Opinion on Investigational Drugs (2019)
Functional insights into nucleotide-metabolizing ectoenzymes expressed by bone marrow-resident cells in patients with multiple myeloma
Immunology Letters (2019)
Journal of Clinical Investigation (2019)
Hematological Malignancy-Derived Small Extracellular Vesicles and Tumor Microenvironment: The Art of Turning Foes into Friends
Sema4D expression and secretion are increased by HIF-1α and inhibit osteogenesis in bone metastases of lung cancer
Clinical & Experimental Metastasis (2019)