Combined TRAF6 Targeting and Proteasome Blockade Has Anti-myeloma and Anti-Bone Resorptive Effects

This article is featured in Highlights of This Issue, p. 499

Multiple myeloma is characterized by the aberrant accumulation of malignant plasma cells in the bone marrow. Osteolytic bone lesions and its associated bone pain and fractures are hallmarks of the disease, causing significant morbidity (1). Until recently, most patients were treated with chemotherapy and glucocorticosteroids as initial therapy. Novel anti-myeloma agents such as the proteasome inhibitors (PI) bortezomib (2, 3)and carfilzomib (4), histone deacetylase inhibitors (5–7), the immunomodulatory agents thalidomide, lenalidomide (8), and pomalidomide (9), pegylated liposomal doxorubicin (1, 10), and arsenic trioxide (11) have proven to be potent inhibitors of myeloma tumor growth in laboratory and clinical studies. Initially, these agents were used alone, but their efficacy was shown to be much improved when combined with each other or with chemotherapy or glucocorticosteroids (12, 13). Despite these improvements in anti–multiple myeloma therapy, the disease remains incurable, and nearly all patients eventually develop resistance to these therapies. Thus, additional tumor targets and therapeutic options that are capable of overcoming drug resistance are critical to help improve the outcome for these patients. To find a new treatment target that inhibits multiple myeloma growth, improves the efficacy of other anti–multiple myeloma agents, and prevents bone loss should both improve the survival and quality of life for patients with this B-cell malignancy. Previously, we showed that inhibition of IκBα degradation with the PI bortezomib decreased NF-κB activity and greatly increased the sensitivity of chemoresistant myeloma cells to chemotherapy without damaging normal bone marrow cells (14, 15). We have also demonstrated that silencing of the TNF receptor–associated factor (TRAF) 6 with siRNA significantly down regulated NF-κB phosphorylation and inhibited multiple myeloma tumor cell proliferation (16).

TRAF6 has been implicated in regulating the transcription factor NF-κB and members of the MAPK family, including the MAPK and the c Jun N-terminal kinase (JNK) signaling transduction pathways (17). TRAF6 is an ubiquitin E3 ligase that catalyzes lysine-63 (K63) polyubiquitination and activates the canonical IκB kinase (IKK) through a proteasome-independent mechanism. TAK1 kinase complex is activated through TRAF6-catalyzed K63 polyubiquitination and phosphorylates IKKβ at 2 serine residues within the activation loop, leading to IKK activation (18, 19). The latter kinase specifically phosphorylates the inhibitor of kappa B (IκB) α protein leading to its degradation which results in NF-κB activation. The TRAF6 ubiquitin–activated TAK1 also phosphorylates MAP kinase kinases (MKK), leading to the activation of the JNK and p38 kinase pathways (20). TRAF6 forms a family of cytoplasmic adaptor proteins that mediate the intracellular signal transduction of TLR/IL-1 superfamilies (21). The TLR/TRAF6/NF-κB pathway plays several major roles in multiple myeloma (17). NF-κB activation results in the production of cytokines, leading to the expression of genes associated with abnormal growth and prevention of apoptosis of tumor cells and their resistance to chemotherapy. TRAF6 was recently found to be a direct E3 ligase for protein kinase B (AKT) ubiquitination, membrane recruitment, and phosphorylation upon growth factor stimulation (22). Furthermore, TRAF6 is also an essential signaling component of the receptor for activation of NF-κB ligand (RANKL)/RANK signaling which stimulates development of osteoclasts and their precursors (OCP). After OCPs are stimulated with RANKL, IL1, or lipopolysaccharide, TRAF6 promotes K63 polyubiquitination on target proteins involved in NF-κB signaling (23). This polyubiquitination through the action of TRAF6 can be reversed with deubiquitinating enzymes (DUB; refs. 24, 25). TRAF6 activity is crucial for osteoclast formation, which is decreased through its deubiquitination (26). Silencing of TRAF6, TRAF6 dominant negative (TRAF6dn) overexpression, or a cell-permeable TRAF6 decoy peptide decreases IL1-induced NF-κB and c-Jun/AP-1 signaling pathway activity, reduces tumor proliferation, enhances their apoptosis, and reduces RANKL-mediated osteoclastogenesis and bone resorption (27, 28). In this study, we investigated TRAF6 expression in multiple myeloma bone marrow samples. Next, we constructed TRAF6dn peptides and determined their effects alone and in combination with a PI, either bortezomib or carfilzomib, on NF-κB phosphorylation, cell signaling, and tumor cell proliferation. We also assessed the effects of these peptides on osteoclast formation and bone resorption.

Peripheral blood and bone marrow aspirate and biopsy specimens were collected from patients with multiple myeloma, individuals with monoclonal gammopathy of undetermined significance (MGUS), and healthy subjects after obtaining institutional review board approval (Western Institutional Review Board) and an informed consent in accordance with the Declaration of Helsinki. Patients were defined as showing progressive disease (PD) or in complete remission (CR) according to the International Myeloma Working Group Uniform Response Criteria guidelines. The human multiple myeloma cell lines RPMI-8226, U266, and MM1S were obtained from ATCC. Cells were cultured with RPMI-1640 supplemented with 10% FBS. Bone marrow mononuclear cells (BMMC) were isolated with density-gradient centrifugation using Histopaque-1077 (Sigma-Aldrich).

Fresh bone marrow biopsies were fixed in 4% paraformaldehyde overnight at 4°C and decalcified with Decalcifying Solution-Lite (Sigma-Aldrich). Samples were imbedded into a paraffin frame. Five-micrometer sections were blocked with 0.05% Tween-20 (TBST) and 3% BSA for 1 hour at room temperature. The samples were exposed to the following antibodies: anti-CD138 (BD Bioscience), anti-TRAF6 (BD Bioscience), and control IgG. For double staining, the slides were washed 3 times with PBS for 15 minutes at room temperature and incubated with FITC-conjugated swine anti-goat or anti-mouse antibodies conjugated to phycoerythrin (PE; BD Bioscience). DAPI (4′,6-diamino-2-phenylindole) was added to slides as a nuclear marker. The slides were washed as before and mounted with aqueous mounting media (Biomeda). Proteins markers were identified under the microscope (Olympus BX51), and merged cells were analyzed using the Microsuite Biological Suite program (Olympus BX51).

Total RNA was isolated from BMMCs. RNA was resuspended in 0.1% diethyl pyrocarbonate-treated water, digested with DNase I (Sigma-Aldrich) to remove contaminating DNA, and extracted with phenol/chloroform followed by ethanol precipitation. Total RNA (1 μg) was reverse-transcribed to cDNA and amplified using the Thermo-Script RT-PCR System (Invitrogen). PCR was performed again using the same system and a GeneAmp PCR System 9700 (Applied Biosystems) for 1 cycle at 94°C for 2 minutes, followed by 35 cycles at 94°C for 30 seconds, 58°C for 30 seconds, 72°C for 1 minute, and 1 cycle at 72°C for 5 minutes. The following primers were used: TRAF6 (L) tgccatgaaaagatgcagag, (R) cgtggttttgccttacaggt; GAPDH, (L) actgccacccagaagactgt, (R) ccagtagaggcagggatgat.

The human TRAF6 probe was fluorescently labeled with 6FAM-ATGCAGAGGAATCAC-MGBNFQ and amplified with the following PCR primers: TRAF6 qPF1, TTCAGTACTTTTGGTTGCCATGA and TRAF6 qPR2, TGTGACTGGGTGTTCTCTTGTAGGT. The human hypoxanthine phosphoribosyl transferase (HPRT) gene was used as the housekeeping control gene. The HPRT probe was labeled with a different fluorescent probe (VIC CATGTTTGTGTCATTAGTG-MGBNFQ) and amplified with PCR primers HPRT qPF3, TGTAGGATATGCCCTTGACTATAATGA and HPRT qPR6, AGGCTTTGTATTTTGCTTTTCCA. Probes and primers were designed, standardized, and tested using Primer Expression 3 software (Life Technologies). This design allowed for the simultaneous use of 2 sets of probes and amplification primers with the same PCR conditions. All amplicons were designed to span introns to eliminate possible false-positive signals due to DNA contamination. The relative abundance of TRAF6 mRNA was represented as the fold change of TRAF6 mRNA compared with HPRT mRNA. The values represent the average of triplicates.

We used the full-length human TRAF6 open reading frame to clone the TRAF6dn cDNA into the PCRII TOPO vector and synthesized a 167–amino acid peptide to produce a TRAF6dn from the NIH GenBank (U78798) and subsequently re-cloned this into an expression vector (pLenti6.2-hTRAF6dn). The TRAF6dn cDNA was generated using reverse transcription in human monocytes and amplification via PCR with specific primers for TRAF6dn (5′-GGTTAGCATGTCAGAGGTCCGGAATTTCCAG and 5′-GCTCGAGCTATACC CCTGCATCAGTACTTGG). The PCR product was cloned into a pCRII-TOPO vector (Invitrogen), and 5 clones were selected for further analysis. Nucleotide sequence analysis of the clones revealed that they contained full-length TRAF6dn (1,115–1,818) that matched the TRAF6 nucleotide and amino acid sequences found in the GenBank database. We subcloned the TRAF6dn gene into an expression pLenti6.2 vector (pLenti6.2-hTRAF6dn) to transduce monocytes and RPMI-8226 cells.

Forty micrograms of protein lysates from BMMCs or multiple myeloma cell lines was electrophoresed on a 4% to 15% SDS-PAGE, and then proteins were transferred to Immobilon-NC nitrocellulose membranes (Millipore) overnight at 50 mA at 4°C. For analysis of signaling molecules, multiple myeloma tumor cells were exposed to TRAF6dn and cultured in reduced serum media containing 5% FBS for 24 hours and stimulated with 20 ng/mL IL1β (Sigma-Aldrich) or IGF (10 ng/mL; Sigma-Aldrich) for 30 minutes. Primary multiple myeloma tumor cells were treated withTRAF6dn, bortezomib, or carfilzomib alone or a combination of the peptide with one of these PIs for 24 hours. The cells were lysed and the proteins were transferred to membranes (Millipore), and the membranes were blocked with 10% non-fat dry milk. Membranes were probed with anti-TRAF6 (Sigma-Aldrich) antibodies, anti-phospho-NF-κB p65 (S536), anti-phospho-IκBα (S32), anti-phospho-Jun (S63/67), anti-IκBα, anti-NF-κB p65, anti-c-Jun (Cell Signaling Technology), anti-AKT, anti-phospho-AKT (Cell Signaling Technology), or anti-β-actin-horseradish peroxidase (HRP; Sigma) antibodies, and the protein blots were incubated overnight at 4°C. Membranes were washed with TBST 3 times followed by incubation with the HRP-conjugated secondary antibody (KPL). Proteins were detected using ECL detection reagent (Amersham) and analyzed using a FluorChem imaging system.

The MTS assay was used to determine cell proliferation. RPMI-8226 multiple myeloma cells (1 × 10⁴) were stably transfected with 0, 1, 10, or 100 nmol/L pLenti6.2 hTRAF6dn vector. These cells were incubated for 72 hours, and cell viability was determined using the MTS assay as per the manufacturer's protocol (Sigma). Data represented the mean of experiments performed 4 times. Statistical significance was determined using the Student t test. We also examined the efficacy of the TRAF6dn peptide alone and in combination with bortezomib or carfilzomib on RPMI-8226 or primary multiple myeloma tumor cells. The multiple myeloma tumor cells were cultured with the TRAF6dn alone for 48 hours at concentrations ranging from 1 to 200 μmol/L or a combination of the TRAF6 inhibitory peptide 25 μmol/L with 6 nmol/L bortezomib or 10 nmol/L carfilzomib. Anti–multiple myeloma effects of this combination were assessed using the MTS assay.

RPMI-8226 cells were transfected with pLenti6.2-hTRAF6dn or pLenti6.2-h vectors as above. Transfected cells were incubated for 48 hours, and apoptosis was measured using the Annexin V assay per the manufacturer's protocol (Boehringer Mannheim) followed with flow cytometric analysis (FC-500 cytometer using Cytomics CXP software, both Beckman Coulter). Statistical significance was determined using the Student t test.

CD14⁺ monocytes were purified from normal human peripheral blood mononuclear cells (PBMC) or multiple myeloma BMMCs using magnetic bead selection. Briefly, 1 × 10⁸ PBMCs or BMMCs were incubated with 200 μL anti-CD14 microbeads (Miltenyi Biotec) in ice for 30 minutes. The cells were washed with cold 1× PBS with 2% FCS and centrifuged at 300 × g for 10 minutes and then resuspended in 1× PBS with 2% FCS. The cell suspension was applied to the magnetic column, and unbound cells were passed through after washing 3 times with 1× PBS with 2% FCS. The CD14⁺ monocytes were treated with 50 ng/mL RANKL and 20 ng/mL monocyte colony-stimulating factor (mCSF) at the beginning of the culture and during a medium change every 3 days. The cells were transfected with pLenti6.2 hTRAF6dn or control vector on the second day of culture. The cells were fixed and then tartrate-resistant acid phosphatase (TRAP) staining was performed with microscopic examination on day 21. The experiment was performed at least 3 times which is represented by the error bars. For assessing bone resorption, the cells were cultured on the dentin slides for 28 days and examined using toluidine blue staining for bone resorption. Resorption pits were examined using light microscopy and the percentage surface area of lacunar resorption on each dentin slice provided by Dr. David Roodman (University of Indiana, Indianapolis, IN) was measured using an image analysis system.

First, we examined TRAF6 and CD138 protein expression using immunofluorescence in bone marrow biopsy samples from 52 patients with plasma cell dyscrasia, including those specifically with multiple myeloma with PD (n = 37) or in CR (n = 11) or MGUS (n = 4). The merged fluorescent image showed that TRAF6 was expressed at high levels in CD138⁺ BMMCs from patients with multiple myeloma especially among bone marrow samples obtained from patients with multiple myeloma with PD (Fig. 1A), whereas the expression of this factor was much lower among patients in CR (Fig. 1B) and individuals with MGUS (Fig. 1C). TRAF6 protein expression was significantly higher in patients with multiple myeloma with PD than in patients with multiple myeloma with CR or individuals with MGUS (P < 0.03; Fig. 1D).

To further assess the levels of TRAF6 transcripts, we determined gene expression from total RNA that was purified from fresh multiple myeloma BMMCs, and RT-PCR was performed using TRAF6 primers. Patients with multiple myeloma with PD expressed high levels of TRAF6 transcripts, whereas levels were much lower among patients in CR (Fig. 2A). This result was further validated using qPCR which demonstrated that TRAF6 gene expression levels were significantly higher in patients with multiple myeloma PD than CR patients (Fig. 2B). Next, Western blot analysis was used to examine the relative TRAF6 protein levels in patients with multiple myeloma. Notably, TRAF6 protein expression was higher in BMMCs from patients with PD than those with CR and healthy controls (Fig. 2C). TRAF6 protein levels were much higher at the time of PD compared with when the same patients were in CR following treatment (Fig. 2D). CD138-expressing BMMCs from patients with multiple myeloma with PD showed higher levels of TRAF6 expression compared with those in CR (Fig. 2E).

We previously reported that silencing TRAF6 gene expression with TRAF6 C-terminal siRNA inhibited multiple myeloma cell proliferation and enhanced apoptosis through interference with NF-κB and c-Jun NH2-terminal kinase signaling in multiple myeloma cell lines (17). In this study, we extended our investigation to determine whether TRAF6 contributes to the cell proliferation of primary multiple myeloma cells using our TRAF6dn peptides. We transfected RPMI-8226 cells with the pLenti6.2-hTRAF6dn expression vector (Supplementary Fig. S1). Introduction of the pLenti6.2-hTRAF6dn vector markedly reduced multiple myeloma cell proliferation in a concentration-dependent fashion. Our data showed that there was a strong negative correlation between the concentration of peptide added and percentage of viable cells remaining after following treatment (Spearman ρ = −0.9383; P = 0.0025; Fig. 3A). In contrast, the pLenti6.2 control vector had no effect on the cells. To investigate whether growth inhibition occurred through an increase in apoptosis, RPMI-8226 cells were transfected with the pLenti6.2 hTRAF6dn vector, stained with Annexin V, and analyzed using flow cytometric analysis. The multiple myeloma cells transfected with pLenti6.2-hTRAF6dn vector demonstrated increased apoptosis in a concentration-dependent fashion. The proportion of apoptotic cells following transfection with the pLenti6.2-hTRAF6dn vector was much higher than among cells containing the pLenti6.2 control vector (P < 0.005; Fig. 3B and C).

To determine whether the TRAF6dn mutants would block NF-κB and/or AP-1 activation in response to IL1 stimulation, RPMI-8226 cells were transfected with TRAF6dn expression vector and then stimulated with IL1β, and cells were assessed using Western blot analysis for the total protein and phosphorylated IκBα (S32/36), NF-κB (S536, NF-κB pathways), and c-Jun (S63, AP-1 pathway) levels. The cells showed maximal phosphorylation after 30 minutes of IL1β exposure (data not shown). Following transient transfection with the TRAF6dn expression vector, the cells were stimulated with IL1β, and NF-κB and AP-1 activation were determined. TRAF6dn reduced serine phosphorylation of IκBα and NF-κB as well as c-Jun in multiple myeloma tumor cells (Fig. 3D). Thus, TRAF6dn efficiently inhibited the phosphorylation cascade that transduces signals to the nucleus from the binding of IL1 to its surface receptor.

To facilitate the clinical applicability of blocking TRAF6, we synthesized a human-specific TRAF6-inhibitory peptide that included a cell membrane permeabilizing–specific peptide sequence (AAVALLPAVLLALLAP RQMPTEDEY). This small molecular TRAF6 inhibitory peptide contained a sequence from the site of TRAF6 that binds to the TRAF6-binding motif (T6DP), thereby preventing binding of RANK to TRAF6. When this peptide was incubated with cells from the RPMI-8226, U266, and MM1S multiple myeloma cell lines, the tumor cells took up 48% (RPMI-8226), 32% (U266), 23% (MM1S) of these modified peptides after exposure to them for 72 hours (Supplementary Fig. S2). We further determined the effects of the TRAF6 inhibitory peptides on primary multiple myeloma tumor cells alone and in combination with PIs. Multiple myeloma cells treated with TRAF6dn, bortezomib, or carfilzomib alone showed a concentration-dependent decrease in cell viability. When the TRAF6dn peptide (25 μmol/L) was combined with bortezomib or carfilzomib, this resulted in a marked decrease in cell viability in RPMI-8226 cells (Fig. 4A) and primary multiple myeloma cells (Fig. 4B) compared with the agents alone. To investigate whether TRAF6dn peptide had an effect on tumor cell apoptosis, primary multiple myeloma cells were treated with increasing concentrations of TRAF6dn peptide followed by Annexin V staining and flow cytometric analysis. A concentration-dependent increase in apoptosis was observed in TRAF6dn-treated multiple myeloma cells. In addition, the proportion of apoptotic cells was significantly higher in TRAF6dn peptide-treated cells than in control peptide–treated cells (P < 0.01; Fig. 4C).

To determine whether TRAF6dn peptide blocked IL1-induced NF-κB activation through the specific inhibition of TRAF6-related pathways in multiple myeloma, primary myeloma cells were stimulated with IL1β for 30 minutes. Next, the stimulated cells were evaluated for induction of the NF-κB and AP-1 pathways with Western blot analysis to determine their total protein levels as well as amounts of phosphorylated IκBα and NF-κB. Phosphorylation of NF-κB was slightly reduced when cells were exposed to the TRAF6dn peptide alone but when this peptide was combined with bortezomib or carfilzomib, the amount of its phosphorylation was markedly decreased as compared with the drugs alone (Fig. 4D).

TRAF6 also has been reported to catalyze Lys63-linked ubiquitination of other signaling molecules such as AKT. Yang and colleagues have recently identified AKT as a target of TRAF6 (29). It is an ubiquitin ligase which promotes AKT translocation to the plasma membrane and increases its activation in the membrane through phosphorylation of AKT at T-308 (29). Thus, inhibition of TRAF6 could be used to augment the efficacy of mTOR inhibitors and increase the efficacy of these drugs for the treatment of patients with cancer (30). AKT phosphorylation decreased when multiple myeloma tumor cells were treated with TRAF6dn peptides compared with primary multiple myeloma tumor cells treated with control peptides (Fig. 4E; Supplementary Fig. S3A). Similarly, when TRAF6dn peptide was combined with the PI3K inhibitor idelalisib (1 μmol/L), the amount of NF-κB phosphorylation was markedly decreased as compared with cells treated with the agents alone (Fig. 4E; Supplementary Fig. S3B). Moreover, gene expression of cyclin D1 which is downstream of AKT was reduced in multiple myeloma cells treated with TRAF6dn (Supplementary Fig. S3C). However, the gene expression of mTOR was slightly increased in myeloma cells (data not shown). Therefore, it is possible that increased mTOR expression in myeloma cells may result from a negative feedback induced through blockage of TRAF6 ubiquitination of AKT.

We evaluated the effect of the TRAF6dn on osteoclast formation using human monocytes isolated with an anti-CD14 microbead affinity column from multiple myeloma patients' PBMCs or BMMCs. Cells containing the pLenti6.2 hTRAF6dn or the control vector pLenti6.2/GW/EmGFP were treated with RANKL and mCSF at the beginning of the culture and during a medium change after 3 days with the same growth factors added. The cells were fixed for a TRAP staining assay on day 21. We found that TRAF6dn markedly inhibited osteoclast cell formation from CD14⁺ cells from PBMCs (Fig. 5A) and BMMCs (Fig. 5B) induced with RANKL and mCSF in a concentration-dependent fashion. The experiment was performed 3 times which is represented by the error bars (Fig. 5C; P < 0.03). TRAF6dn also inhibited bone resorption as determined via the dentin pit assay (Fig. 5D).

We report that TRAF6, a TAK1 K63-specific ubiquitin E3 ligase that leads to the activation of TAK1 and thus subsequent activation of IKK and NF-κB (31), is over expressed in tumor cells from patients with multiple myeloma. Inhibition of its activity decreases growth and increases apoptosis of multiple myeloma cells. It also reduces osteoclast formation and bone resorption in monocytes derived from PBMCs and BMMCs from patients with multiple myeloma. Furthermore, overexpression of TRAF6 in cells has been shown to activate NF-κB (32), which has been shown to have an important role in myeloma, specifically in the regulation of genes involved in tumor cell proliferation, survival, and chemoresistance (33, 34). We and others have previously reported that NF-κB activity is increased in both multiple myeloma cell lines (34) and fresh BMMCs from patients with multiple myeloma with a high proportion of tumor cells (>90%) when compared with normal BMMCs (35). TRAF6 is critical to the activation of NF-κB and JNK in response to IL1 stimulation in multiple myeloma (17). In this study, patients with multiple myeloma with PD had markedly higher levels of TRAF6 gene and protein expression in their BMMCs than patients with multiple myeloma in CR, MGUS individuals, or healthy subjects. Notably, TRAF6 protein expression was higher in patients with multiple myeloma at the time of PD when compared with the same patients evaluated when they were in CR. Thus, changes in TRAF6 gene and protein expression in multiple myeloma bone marrow correlate with changes in disease status among myeloma patients.

We further investigated whether TRAF6 is a potential target molecule for the treatment of multiple myeloma. Several members of the TRAF family, including TRAF2, TRAF3, TRAF5, and TRAF6, have been implicated in regulating signal transduction from various TRAF family members (36). However, the TRAF-C domain largely determines the unique biologic function of TRAF6, whereas it does not interact with peptide motifs that are recognized by TRAF1, 2, 3, or 5. TRAF6 plays a crucial role in the signaling activation which occurs when CD40, another member of the TNF receptor superfamily, is ligand bound or through IL1 which leads to the stimulation of multiple myeloma growth. TRAF6 binds to the amino terminal region of the CD40 cytoplasmic tail, which is distinct from the binding domain for TRAF2, TRAF3, and TRAF5. On the basis of the sequences and crystal structures of TRAF6-, CD40-, and the RANK-binding site (37), we targeted the TRAF6 C-terminal domain residues from 420 to 440 which are part of CD40- or RANK-binding domain (residues 333 to 508) of TRAF6. In this study, we found that a TRAF6dn mutant, expressed using a pLenti6.2 hTRAF6dn expression vector, inhibited both the NF-κB and AP-1 signal transduction pathways resulting in inhibition of the growth of multiple myeloma cells. It has been reported that TRAF6 through its ubiquitination properties plays an important role in the activation of other kinases such as AKT (23). We also examined AKT phosphorylation downstream of AKT and cyclin D gene expression when multiple myeloma tumor cells were treated with the TRAF6 inhibitory peptide and compared with multiple myeloma cells treated with control peptides in both RPMI-8226 and primary multiple myeloma tumor cells. AKT phosphorylation and cyclin D gene expression were both reduced following treatment with TRAF6dn. Our study confirms the involvement of TRAF6 in the IL1R/TLR/NF-κB and JNK signaling pathways in multiple myeloma. Using dominant-negative expression vectors, we provide evidence that blocking this adapter protein may produce anti–multiple myeloma effects.

Stimulation of osteoclasts via the RANK/RANKL interaction increases bone resorption (38). Bone destruction is a hallmark of myeloma (1). Malignant plasma cells exert major osteoclastogenic effects through enhancement of RANK signaling leading to increases in the recruitment, differentiation, and activation of osteoclasts and their progenitor cells within the bone marrow. TRAF6 is an essential signaling component of RANKL/RANK signaling in osteoclasts and OCPs. Osteoclasts are multinucleated cells derived from monocytes that play critical roles in bone resorption (39). TRAF6 enhances bone resorption among patients with multiple myeloma and other cancers metastatic to bone. This occurs through IL1 stimulation of PGE2 production in osteoblasts and stromal cells. This stimulation leads to an increased production of RANKL and enhanced RANK activation, resulting in an increased differentiation of monocytes into osteoclasts (40). Monocytes showing TRAF6 gene mutations do not differentiate into functional bone resorbing osteoclasts despite RANKL stimulation; and, thus, this emphasizes the critical importance of this adapter protein in mediating RANKL-induced osteoclast formation and its bone resorptive activities (41, 42). In this study, we demonstrated that the blockage of TRAF6 signaling with an inhibitory peptide reduces osteoclast formation and bone resorption.

Multiple myeloma is characterized by an aberrant accumulation of malignant plasma cells in the bone marrow. Novel anti-myeloma agents such as the immunomodulatory agents thalidomide, lenalidomide, and pomalidomide and the PIs bortezomib and carfilzomib have shown clinical benefits for patients with multiple myeloma (42, 43). Despite these improvements in anti–multiple myeloma therapy, the disease remains incurable, and nearly all patients eventually develop drug resistance. Thus, additional therapeutic options that are capable of overcoming drug resistance are critical to help improve the outcome for these patients (44, 45). For example, the recent finding by Shah and colleagues demonstrated that eliminating the cytoprotective heat shock response by targeting heat shock factor 1 (HSF1), a protein that protects myeloma cells from proteasome inhibition, may serve as one potential mechanism for overcoming PI-based therapies resistance (46). In addition, bone disease is a major cause of morbidity and mortality in patients with multiple myeloma (47–49). Thus, finding treatments that both overcome drug resistance and reduce bone resorption should improve overall survival and quality of life for patients with this common hematologic malignancy. Mechanistically, the PIs bortezomib and carfilzomib inhibit the chymotryptic activity of the 20S subunit of the proteasome (50–52). TRAF6 is an IKK upstream activator that mainly activates the canonical NF-κB signal transduction pathway. Blockade of TRAF-6 reduces this factor's induction of polyubiquitin synthesis and also prevents proteasome activity (53).

In this study, we provide evidence that combining TRAF6 inhibitors with PIs may provide a novel approach to treat patients with multiple myeloma, as blocking TRAF6 activity not only reduces growth and increases apoptosis of multiple myeloma cells, enhances the anti–multiple myeloma activity of PIs but also decreases osteoclast formation and bone loss. Specifically, we have demonstrated that the combination of the TRAF6-inhibitory peptide with either bortezomib or carfilzomib has markedly decreased NF-κB phosphorylation and induced enhanced anti–multiple myeloma activities on primary multiple myeloma cells. Furthermore, inhibition of TRAF6 results in reduced osteoclast formation and bone resorption in multiple myeloma bone marrow monocytes induced with RANKL and mCSF.

No potential conflicts of interest were disclosed.

Conception and design: H. Chen, J.R. Berenson

Development of methodology: H. Chen, M. Li, C.S. Wang, T. Lee, G.Y. Tang, J.R. Berenson

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): E. Sanchez, T. Lee, C.M. Soof, C.E. Casas, K.A. Udd, K. DeCorso

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): H. Chen, M. Li, T. Lee, K. DeCorso, G.Y. Tang, J.R. Berenson

Writing, review, and/or revision of the manuscript: H. Chen, M. Li, J. Cao, K.A. Udd, G.Y. Tang, T.M. Spektor, J.R. Berenson

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): H. Chen, E. Sanchez, C.S. Wang, C.M. Soof, C.E. Casas, J. Cao, C. Xie, K. DeCorso, J.R. Berenson

Study supervision: H. Chen, J.R. Berenson

This research was supported by the Skirball Foundation, Annenberg Foundation (Los Angeles, CA), Kramer Foundation (San Francisco, CA), and Myeloma Research Fund (San Jose, CA).

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.