Arsenic trioxide (ATO) induces apoptosis of malignant plasma cells through multiple mechanisms, including inhibition of DNA binding by nuclear factor kappa-B, a key player in the development of chemoresistance in multiple myeloma (MM). This activity suggests that ATO may be synergistic when combined with other active antimyeloma drugs. To evaluate this, we examined the antimyeloma effects of ATO alone and in combination with bortezomib, melphalan and ascorbic acid (AA) both in vitro and in vivo using a severe combined immunodeficient (SCID)-hu murine myeloma model. Marked synergistic antimyeloma effects were demonstrated when human MM Los Angeles xenograft IgG lambda light chain (LAG lambda-1) cells were treated in vitro with ATO and any one of these agents. SCID mice bearing human MM LAG lambda-1 tumours were treated with single-agent ATO, bortezomib, melphalan, or AA, or combinations of ATO with either bortezomib or melphalan and AA. Animals treated with any of these drugs alone showed tumour growth and increases in paraprotein levels similar to control mice, whereas animals treated with ATO-containing combinations showed markedly suppressed tumour growth and significantly reduced serum paraprotein levels. These in vitro and in vivo results suggest that addition of ATO to other antimyeloma agents may result in improved outcomes for patients with relapsed or refractory MM.

Pleiotrophin (PTN) is an important developmental cytokine that is highly expressed during embryogenesis but shows very limited expression in adult tissues, where it is largely restricted to the brain. High PTN serum levels are associated with a variety of solid tumors. We recently showed that patients with multiple myeloma (MM) also have elevated serum levels of this protein and the amount of PTN correlated with the patients' disease status and response to treatment. In this study, we demonstrate that MM cell lines and the malignant cells from MM patients' bone marrow produced PTN and secreted PTN protein into the supernatants during short-term culture. Moreover, Ptn gene expression correlated with the patients' disease status. Inhibition of PTN with a polyclonal anti-PTN antibody reduced growth and enhanced apoptosis of MM cell lines and freshly isolated bone marrow tumor cells from MM patients in vitro. Importantly, this antibody also markedly suppressed the growth of MM in vivo using a severe combined immunodeficiency (SCID)-hu murine model. This represents the first study showing the importance of PTN in the growth of any hematological disorder. Because the expression of this protein is very limited in normal adult tissues, PTN may represent a new target for the treatment of MM.

Inorganic arsenics like arsenic trioxide (ATO) are novel anti-cancer drugs active in acute promyelocytic leukemia (APL) and multiple myeloma (MM). ATO induces apoptosis of plasma cells by several mechanisms including down-regulation of BCL-2 expression and inhibition of DNA-binding by NF-κB. The amount of ATO that can be safely given is low because of QTc-prolongation. ZIO-101, a new organic arsenic, is in phase-1/-2 trials. ZIO-101 can be safely given at much higher doses than ATO. We evaluated the in vivo anti-myeloma activity of ZIO-101 in a SCID-hu mouse model of human myeloma. LAGλ-1 was developed from a person with melphalan-resistant and LAGλ-1B from a person with bortezomib-resistant myeloma. Each severe combined immunodeficient (SCID) mouse was implanted with a 2–4 mm3 fragment of LAGλ-1B or LAGλ-1 into the left superficial gluteal muscle. Fragments were allowed to grow for 14 d when human IgG was first detectable. ZIO-101 was given IV 1 or 2 times/d using three different schedules: one day/w, 3 days/w or 5 days/w at doses of 50– 200 mg/kg/d. Tumor volume and human IgG were assessed weekly. Doses up to 200 mg/kg were well-tolerated. Anti-myeloma effects were observed in both models at doses of 100 – 200 mg/kg on all 3 schedules. Mice receiving 100 mg/kg twice daily thrice weekly and those receiving 200 mg/kg once weekly showed marked anti-myeloma activity (100 mg/kg, P = 0.03; 200 mg/kg, P = 0.001) and reduced human IgG levels (100 mg/kg, P < 0.001; 200 mg/kg, P = 0.01) compared to controls. We are evaluating ZIO-101 in other SCID-hu models of human myeloma and exploring different doses and schedules of ZIO-101 alone or combined with other anti-myeloma agents. In summary, these data show activity of ZIO-101 in human myeloma in vivo. These studies provide the bases for future clinical trials.

Histone deacetylase (HDAC) inhibitors represent a new mechanistic class of anti-cancer therapeutics that inhibit HDAC enzymes and have been shown to have anti-proliferative effects in cancer cells (including drug resistance subtypes), induce apoptosis, inhibit angiogenesis, and sensitize cancer cells when combined with other available anti-cancer therapies. PXD101 is a novel investigational small molecule drug that selectively inhibits HDAC enzymes. In recent preclinical studies, PXD101 has been shown to have the potential to treat a wide range of solid and hematological malignancies either as a monotherapy or in combination with other active agents. In this study, we evaluated the activity of PXD101 on multiple myeloma samples when used as monotherapy or in combination with the proteasome inhibitor bortezomib. In vitro experiments indicated that PXD101 pretreatment (20 mM; 3h) sensitized RPMI-8226 human multiple myeloma cells to subsequent bortezomib exposure (5 nM; 72h). To examine PXD101 and bortezomib in vivo, two mouse models of human multiple myeloma were utilized (LAGλ-1 and LAGκ-1B). LAGλ-1 was generated from a patient resistant to melphalan therapy and LAGκ-1B from a patient who progressed on bortezomib treatment (Campbell et al, International Journal of Oncology 2006). SCID mice were implanted with LAGλ-1 or LAGκ-1B tumor fragments into the left superficial gluteal muscle. Tumors were allowed to grow for 14 days at which time human IgG levels were detectable in the mouse serum, and mice were randomly assigned into treatment groups. Groups consisted of Vehicle only, PXD101 alone (40 mg/kg), bortezomib alone (0.5 mg/kg), or PXD101 (40 mg/kg) + bortezomib (0.5 mg/kg). In one cohort, PXD101 and bortezomib were administered twice weekly (M, Th) and in another cohort PXD101 was administered 5 days a week (M-F) and bortezomib twice weekly (M, Th). When administered, PXD101 was given i.p twice daily and bortezomib once daily intravenously. The results of these animal experiments will provide preclinical information on the activity of PXD101 monotherapy and PXD101/bortezomib combination therapy on drug-resistant myeloma samples, and may help to define the optimal schedule for potential clinical evaluation of this drug combination.

We have recently reported that rituximab treatment of B-NHL cell lines, like Ramos, inhibited the PI3K/AKT signaling pathway and downregulated Bcl-xL expression. The role of the AKT pathway in chemosensitization was corroborated by treating Ramos cells with the AKT inhibitor, LY294002, and resulted in sensitization of the cells to drug-induced apoptosis. We have investigated a potential underlying mechanism responsible for the rituximab-mediated inhibition of the AKT pathway. PTEN (phosphatase and tensin homologue deleted on chromosome 10) is a tumor suppressor phosphatase and functions as a negative regulator of the PI3K pathway through its phosphatase activity. We hypothesized that rituximab may upregulate PTEN expression and thus, inhibiting the AKT pathway. Treatment of Ramos cells with rituximab (20μg/ml for 20h) resulted in significant upregulation of PTEN expression (as assessed by both Western and RT-PCR). Time kinetic analysis showed that PTEN is upregulated as early as 6–9h post rituximab treatment. In addition, since rituximab inhibits cell proliferation and cell growth, we have examined the effect of rituximab on the expression of the growth factor pleitrophin (PTN). PTN is a heparin-binding and secreted growth differentiation factor that mediates various functions such as cell motility and migration, survival, growth and differentiation. Treatment of Ramos cells with rituximab inhibited PTN expression as early as 12h post treatment. The present findings suggested that rituximab-mediated induction of PTEN expression and inhibition of both the AKT pathway and PTN epxression may be interrelated and play an important role in rituximab-mediated cell growth inhibition and chemosensitization. A recent report by Li et al., (JBC, 281:10663,2006) demonstrated that PTEN null cells exhibited upregulation of PTN expression and activation of the PI3K/AKT pathway. Further, inhibition of PTN resulted in inhibition of the AKT pathway, thus establishing a feedback mechanism. Our findings with rituximab are consistent with the Li's findings'. The mechanism by which rituximab inhibits PTN expression is not clear. Reported studies have implicated the role of AP-1 in PTN transcription and in agreement, our findings have also demonstrated that rituximab inhibits AP-1. Overall, the present studies suggest novel targets modified by rituximab namely, PTN and PTEN, which can be considered for therapeutic intervention in the treatment of both rituximab-sensitive and rituximab-resistant tumors.

We have discovered a novel mechanism leading to blood vessel formation involving transdifferentiation of monocytes into endothelial cells by tumor cell production of pleiotrophin (PTN), a protein highly produced by myeloma (H. Chen et al, Blood, 2005; Yeh et al BJH, 2006). Arsenic trioxide (ATO) induces apoptosis of cancer cells directly through a number of mechanisms, and this drug has also been shown to inhibit angiogenesis. However, it remains unknown whether ATO affects the earliest stages of angiogenesis and vasculogenesis important in tumor development. We purified human monocytes (CD14+) and cultured these cells on collagen I-coated dishes. mCSF was added to the cells after 1 hour of culture. PTN was added twice to the culture, once after 24 hours and again after 5 days with or without ATO or bortezomib. FLK-1 expression (VEGFR-2) showed that the cells incubated on collagen I without drugs formed tube-like structures in the presence of PTN and mCSF. However, the tube-like structures disappeared after adding either the IC50 (5x10−6M) dose or low (5x10−7M) dose of ATO. FLK-1 staining remains in the tube-like structures with low doses (3x10−12M) of bortezomib. In order to examine whether ATO or bortezomib affects endothelial gene expression when monocytes are induced to transdifferentiate in the presence of these cytokines, we also examined expression using RT-PCR on endothelial cell genes (vascular endothelial growth factor receptor-2 (Flk-1), Tie-2 and von Willebrand factor (vWF)) and Western blot analysis for protein expression. The results of both RT-PCR and Western blot analysis showed that the expression of endothelial markers was blocked at both the higher (5x10−6M) and lower (5x10−7M) doses of ATO. In contrast, the expression of endothelial markers was not reduced by adding low dose bortezomib (3x10−12M). We further examined the effects of ATO and bortezomib on early stage angiogenesis in vivo using the chorioallantoic membrane (CAM) assay. Fertilized chick eggs were incubated horizontally at 38°C in a humidified incubator, windowed by day 3 of incubation and processed by day 8. The tested micro-sponge with ATO (5x10−6M) or bortezomib (3x10−11M) or control reagents was implanted on the CAM. The eggs were sealed with adhesive tape and returned to the incubator for 48 hours. The assay scored positive when two independent observers reported a significant reduction of vessels in the treated area. The results of the CAM assay showed that compared to saline, ATO significantly reduced new macroscopic and microscopic vessel formation. In contrast, bortezomib did not affect angiogenesis in the CAM assay. These experiments define a previously unrecognized novel mechanism by which ATO may have anti-angiogenetic effects in cancer patients-preventing the transdifferentiation of monocytes into endothelial cells by PTN. They also suggest ATO as a potential new specific agent to inhibit angiogenesis resulting from transdifferentiation of monocytes into vascular endothelial cells driven by pleiotrophin and mCSF. These results suggest a novel way by which anti-cancer agents may impact angiogenesis.

The peripheral benzodiazepine receptor (mPBR) appears to be a potential target to induce apoptosis in tumor cells. The expression of this receptor has been linked to a poor prognosis in cancer patients. PK11195 may represent a new, well-tolerated potent chemosensitizing agent that affects multiple resistance mechanisms within malignant cells. We have evaluated whether PK11195 inhibits multiple myeloma (MM) cell growth in vitro; and, furthermore, whether this drug can chemosensitize a melphalan resistant human MM tumor, LAGλ-1 (Campbell et al, International Journal of Oncology 2006), to arsenic trioxide (ATO) and melphalan using an in vivo SCID-hu model. The MM cell lines RPMI8226 and U266 were treated with varying concentrations of PK11195 (1 – 100 mM). After incubating with PK11195 for 24 hours, cell growth was measured by MTT assay. Those cells treated with PK11195 showed decreased proliferation at concentrations as low as 1 mM compared to the untreated cells. Next, we investigated the chemosensitizing effects of PK11195 using an in vivo model of human MM. To accomplish this, each immunodeficient (SCID) mouse was implanted with a 2.0 – 4.0 mm3 LAGλ-1 tumor fragment into the left superficial gluteal muscle. The tumors were allowed to grow for 14 days at which time human IgG levels were detectable in the mouse serum or when tumors became palpable (21 days) and mice were blindly assigned into treatment groups. PK11195 (10, 50 and 100 mg/kg) was administered via oral gavage once weekly when combined with melphalan and once daily five times per week when combined with ATO. Melphalan (3 mg/kg) was administered once weekly via intraperitoneal (i.p.) injection. ATO (1.25 mg/kg) was administered once daily five times per week via i.p. injection. Mice receiving the combination of PK11195 and melphalan (3 mg/kg) showed marked inhibition of tumor growth (PK11195 10 mg/kg, P = 0.03; PK11195 50 mg/kg, P = 0.02; PK11195 200 mg/kg, P < 0.01) compared to mice receiving no therapy. Animals treated with melphalan, as a single agent, did show minimal tumor growth inhibition and reduced paraprotein levels whereas mice treated with single agent PK11195 showed tumor growth similar to the control mice. Mice receiving the combination of PK11195 and low dose ATO (1.25 mg/kg) also showed inhibition of tumor growth (PK11195 200 mg/kg, P < 0.01) whereas treatment with either single agent PK11195 or ATO demonstrated growth similar to the control groups. Treatment with the highest dose of PK11195 (200 mg/kg) was not associated with any observed toxicity suggesting that high doses can be safely administered and are well tolerated. In this study, we showed PK11195 inhibits MM cell growth in vitro at very low concentrations and can chemosensitize drug resistant tumor cells in vivo at doses that have no observable toxicity. We are further evaluating PK11195 as a single agent and in combination therapy both in vitro and in vivo..

We have previously shown that multiple myeloma (MM) patients express pleiotrophin (PTN) and it is found at high levels in MM serum as well as PTN is a key factor in the transdifferentiation of monocytes into endothelial cells. We determined the level of PTN expression in myeloma and breast cancer and determined whether PTN produced by these tumor cells could induce endothelial cell expression in human monocytes. Both myeloma and breast cancer cells produced high levels of PTN and secreted this growth factor into the culture medium whereas normal bone marrow showed no expression of this protein. Next, MM cell lines, human bone marrow (BM) from MM patients or control subjects or breast cancer cells were cultured with CD14 + PBMCs using transwell culture plates coated with collagen I. CD14 + monocytes exposed to cells from MM cell lines or fresh BM or breast cancer cells showed expression of endothelial genes (Flk-1, Tie-2, CD144, and vWF) and lost expression of monocyte genes (c-fms). Induction of endothelial gene expression was blocked with an anti-PTN antibody. In contrast, CD14 + cells exposed to normal bone marrow as well as cell lines lacking PTN expression did not show endothelial gene expression. We determined whether human monocytes could be incorporated in vivo as vascular endothelium within human tumors that express PTN. Human myeloma LAGλ-1 cells which highly express and secrete PTN were mixed with THP1 monocytes transduced with the green fluorescent protein (GFP) gene and injected subcutaneously into SCID mice. Mice were sacrificed 6 weeks later and tumor was fixed and frozen sections. MM cells or THP1 monocytes alone did not demonstrate the presence of GFP + blood vessels. Notably, GFP + THP1 cells were found in blood vessels within the PTN-expressing LAGλ-1 tumor in animals injected with both cells together. When GFP + h2Kd - blood vessels were stained for anti-human and anti-mouse CD31, 60% of the endothelial cells stained positive for human CD31 and the remaining cells stained positive for mouse CD31 whereas none of these cells stained positive for both mouse and human markers. These results show that the blood vessels containing GFP + cells do not result from fused cells. In addition, an anti-PTN antibody but not control IgG antibody blocks the incorporation of GFP + cells into the vasculature of the LAGλ-1 tumors. Staining of serial sections with anti-Tie-2 and CD31 antibodies showed a similar distribution pattern. We further examined endothelial gene expression in these in vivo -generated samples using RT-PCR. The results showed that the THP1 monocytes or LAGλ-1 tumor cells alone did not express endothelial genes whereas THP1 monocytes mixed with PTN-expressing LAGλ-1 showed endothelial gene expression. This endothelial gene expression was blocked by anti-PTN antibody. These data show that hematologic and solid tumors through expression of PTN support new blood vessel formation by the transdifferentiation of monocytes into endothelial cells and provide a new potential target for inhibiting blood vessel formation in solid and liquid tumors.

Dominant negative inhibition is most commonly seen when a mutant subunit of a multi-subunit protein is co-expressed with the wild-type protein so that assembly of a functional oligomer is impaired. Studies have shown that TRAF6 plays a key role in the regulation of NF-κB through the IL-1R/TLR-TRAF6-TAK1-TAB1-TAB2-IkB-NF-κB pathway. We previously demonstrated that TRAF6 is an important factor for the activation of nuclear factor (NF)-κB signaling in multiple myeloma cell proliferation through the c-Jun N-terminal kinase (JNK) pathway and the pathway can be silenced by TRAF6 siRNA. (H. Chen et al. Oncogene, 2006). We targeted the TRAF6 function domain by designing primers targeting positive 1115 to 1818 (Forward: ggctagcatgtcagaggtccggaatttggag (Nhe1) Reverse: cgaagtactgatgcaggggtatagctcgagc (Xho1)) for hTRAF6dn according to GeneBank (NCBI) nucleotide sequence of human TRAF6 (#U78798). We cloned TRAF6 negative domain cDNA into PCRII-TOPO vector and subsequently re-cloned into the pLenti6.2 expression vector (pLenti6.2-hTRAF6dn). All constructs were confirmed by sequencing. Viral titers for all transfections were determined to be 107 plaque-forming units/ml. Expression levels as determined by flow cytometric analysis were >95% for all lentivirally encoded GFP gene products. The pLenti6.2-hTRAF6dn vector continually expressed the peptide for TRAF6dn during tumor cell proliferation. We found that TRAF6dn began to inhibit MM cell proliferation in the U266 myeloma cell line after 72 hours of culture and most prominently on day 6. However, the inhibition of RPMI8226 cell proliferation by TRAFdn started after 24 hours of culture whereas effects on inducing MM cell apoptosis were most prominent at 72 hours. The decrease in cell proliferation and increase in cell apoptosis occurred in a dose-dependent fashion. We also examined the effects of TRAF6dn on the NF-κB and JNK pathway since this signaling pathway is associated with cell cycle effects in myeloma. Phosphorylated NF-κB protein levels were reduced using the TRAF6dn expression vector. We also determined the phosphorylation of JUN kinase kinase (JNKK), which activates the MAP kinase homologues SAPK and JNK in response to IL-1 receptor stimulation. The results showed that the phosphorylation of JNKK is clearly reduced following blocking the TRAF6 function domain with the TRAF6dn. Furthermore, we examined c-Jun, a component of the transcription factor complex AP-1, which binds and activates transcription at TRE/AP-1 elements. The transcriptional activity of c-Jun is regulated by SAPK/JNK binding to c-Jun and phosphorylation of c-Jun at Ser63/73. We found that total endogenous c-Jun is reduced after blocking the TRAF6 function domain with TRAF6dn in the RPMI8226 and U266 MM cell lines. Comparing TRAF6dn with TRAF6 siRNA, only the TRAF6dn inhibited the TRAF6 function domain. These studies suggest that the TRAF6dn peptide may impede myeloma cell signaling pathways resulting in inhibition of tumor cell growth and may represent a new approach to treating patients with MM.

Proc Amer Assoc Cancer Res, Volume 47, 2006 4859 ATN-224 is an orally available small molecule inhibitor of the cytoplasmic enzyme superoxide dismutase 1 (SOD1) that is currently entering Phase II clinical trials in multiple cancer indications. ATN-224 has pleiotropic effects on both endothelial cells and tumor cells, leading to the inhibition of angiogenesis and the induction of tumor cell apoptosis in vitro and in vivo. ATN-224 interferes with a number of signaling pathways in parallel, including those mediated by VEGF, FGF-2 and IGF-1. These pathways have been implicated in myeloma progression, leading us to evaluate the effects of ATN-224 on myeloma cells in vitro as well as in murine models of myeloma progression in vivo. ATN-224 inhibited IGF-1 mediated activation of ERK1/2 in MM.1S cells and induced apoptosis in several myeloma cell lines including MM.1S, MM.1R (a dexamethasone resistant subclone of MM.1S) and RPMI-8226 in a dose and time-dependent manner. ATN-224 was also evaluated in SCID-hu models of myeloma. LAGκ-1B tumors, originally isolated from a patient refractory to lenalidomide, melphalan and bortezomib, were transplanted intramuscularly into SCID mice. Fourteen days post-implantation, mice were randomized based on human IgG levels and treated with ATN-224 at 5 mg/kg, 15 mg/kg, and 50 mg/kg per day via oral gavage on a Monday-Friday schedule. Mice receiving the intermediate (15 mg/kg) and high doses of ATN-224 (50 mg/kg) showed marked inhibition of tumor growth and reduction of human IgG levels while there was little effect observed in the low dose treatment group (5 mg/kg). Additionally, when a late dose of bortezomib (0.125 mg/kg) was administered with low dose ATN-224 (5 mg/kg), the combination had significantly better antitumor activity than ATN-224 or bortezomib alone, indicating that ATN-224 could re-sensitize tumors that were previously resistant to bortezomib. ATN-224 was also found to inhibit the expression of a novel myeloma growth and pro-angiogenic factor produced by myeloma cells, pleiotrophin, in a dose-dependent fashion. We have previously demonstrated that pleiotrophin expression declines in myeloma patients that respond to anti-myeloma therapy, further substantiating the anti-myeloma activity of ATN-224. These data suggest that ATN-224 may be useful in treating patients with highly refractory myeloma and provide a rationale for the clinical development of ATN-224 in this indication.

We set out to generate new human myeloma tumors that grow in immunodeficient mice and can be used for pathophysiological studies and rapid evaluation of new therapies. Fresh whole core bone marrow (BM) biopsies taken from 33 myeloma patients were engrafted into the hind limb muscle of severe combined immunodeficient (SCID) mice. Human Ig was detected in 28/33 mice and three grew palpable tumors displaying many features of human myeloma including morphology, immunophenotype and BM plasmacytosis. Following intramuscular passage, we generated large numbers of mice with predictable increases in tumor growth and human paraprotein levels. We further characterized the model generated from an IgGlambda-producing tumor known as LAGlambda-1 and determined the effects of the proteasome inhibitor bortezomib, the alkylating agent melphalan, and the DNA damaging agent liposomal doxorubicin, on the growth of this tumor. LAGlambda-1-bearing mice receiving higher doses of bortezomib showed reduced tumor growth whereas a lower dose had no effect. In contrast, melphalan did not significantly alter tumor growth, except minimally at high doses, reflecting the resistance of this patient's tumor to this drug. We also used our intramuscular (i.m.) LAGlambda-1 model to optimize the dosing schedule of liposomal doxorubicin. Low doses administered once daily three days per week decreased tumor growth and human paraprotein levels whereas much higher doses given once weekly had no anti-myeloma effects. Furthermore, LAGlambda-1 cells produce local tumors when injected subcutaneously and lytic lesions when injected intravenously allowing for multiple methods of evaluating the anti-myeloma effects of a variety of agents. Our new clinically relevant SCID models of human myeloma should greatly facilitate drug development and enable novel therapies to quickly move from the laboratory to the clinic.

Vascular endothelial growth factor (VEGF) is an important signaling protein that plays a critical role in vasculogenesis and angiogenesis, and serves as one of the contributors to physiological or pathological conditions that can stimulate the formation of new blood vessels. The uncontrolled growth of new blood vessels is an important contributor to a number of pathological conditions, including multiple myeloma (MM). In support of this, bone marrow angiogenesis has been shown to correlate with disease status and poor prognosis in MM. VEGF also directly induces myeloma cell proliferation. Based on these findings, we evaluated a new mouse/human anti-VEGF antibody using our SCID-hu mouse models of human MM. Each immunodeficient (SCID) mouse was implanted with a 2.0 – 4.0 mm3 LAGκ-1A tumor fragment into the left superficial gluteal muscle. The tumors were allowed to grow for 14 days at which time human IgG levels were detectable in the mouse serum, and mice were blindly assigned into one of two treatment groups. In one group, anti-VEGF antibody was administered via intraperitoneal injection twice per week at a dose of 5 mg/kg. In the other cohort, control mice were given a control IgG antibody (5 mg/kg) on the same schedule. Mice receiving the anti-VEGF antibody showed marked inhibition of tumor growth ( P = 0.0005) and reduction of paraprotein levels ( P = 0.0002) compared to mice receiving control antibody. On day 42, LAGκ-1A-bearing mice receiving the anti-VEGF antibody showed a 70% reduction in human paraprotein levels and an 80% decrease in tumor volume compared the control antibody treated animals. Treatment with the anti-VEGF antibody was not associated with any observed toxicity. We are currently evaluating this anti-VEGF antibody in several of our mouse models of human MM and plasma cell leukemia. Based on these data with anti-VEGF monotherapy, we are currently investigating the anti-tumor activity of anti-VEGF antibody plus bortezomib as well as other available anti-MM agents using our in vivo SCID-hu myeloma murine models. Preliminary results are encouraging with single agent anti-VEGF antibody and additional studies may be used to direct the clinical development of anti-VEGF antibody treatment alone and in combination regimens for patients with relapsing or refractory MM.

Pleiotrophin (PTN), a tightly regulated angiogenic and mitogenic heparin-binding protein, is markedly elevated in a variety of aggressive solid tumours. The role of PTN in haematological malignancies, however, has not been previously evaluated. This study demonstrated that PTN serum levels were elevated in multiple myeloma (MM) patients when compared with healthy subjects (P < 0.0001). Serum levels of this protein significantly increased during progression of disease, and decreased during response to anti-MM therapy (P < 0.001). These results suggest that serum PTN may be a new biomarker for monitoring the disease status and therapeutic response of MM patients.

Abstract CD40 is a TNF receptor found on the cell surface of mature B cells (B lymphocytes) and most B-cell malignancies including multiple myeloma (MM). SGN-40 is a high-affinity, humanized monoclonal antibody that targets the CD40 antigen. Recently, it has been shown that SGN-40 decreases the proliferation of malignant B cells by partial agonistic signaling and effector functions in vitro. In this study, we examined the anti-MM effects of SGN-40 in vivo using a CD40+ SCID-hu murine model of human myeloma, LAGκ-1A. Each immunodeficient (SCID) mouse was implanted with a 2.0 – 4.0 mm3 LAGκ-1A tumor fragment into the left hind limb muscle. The tumor was allowed to grow for 14 days at which time human IgG levels were detectable in the mouse serum. Mice were then randomly assigned to one of four SGN-40 treatment groups (6 mice per treatment group). SGN-40 was administered via intraperitoneal injection twice per week at doses of 0.1, 0.3, 1, and 3 mg/kg. Control mice were given a control IgG antibody (3 mg/kg) using the same schedule. Mice receiving the higher doses of SGN-40 showed marked inhibition of tumor growth (0.3 mg/kg, P < 0.02; 1 mg/kg, P < 0.03; and 3 mg/kg, P < 0.04) and reduction of paraprotein levels (1 mg/kg, P < 0.05; and 3 mg/kg, P < 0.03) compared to mice receiving control antibody. At the lowest dose of SGN-40 evaluated (0.1 mg/kg) a slight inhibition of tumor growth was observable, but there was no effect on human paraprotein. Treatment with SGN-40 was not associated with any observed toxicity. Based on these data with SGN-40 monotherapy, we are currently investigating the antitumor activity of SGN-40 plus bortezomib as well as other available anti-MM agents using our in vivo SCID-hu myeloma murine model. These data for single-agent SGN-40 are encouraging and support testing SGN-40 both alone and in combination regimens to treat MM patients.

The tumor necrosis factor receptor (TNFR)-associated factor (TRAF) family of six adaptor proteins (TRAF1-6) links the TNFR superfamily to the nuclear factor kappa B (NF-kappaB) and activator protein-1 (AP-1) transcriptional activators. Unlike other TRAFs, TRAF6 is also involved in Toll-like/interleukin (IL)-1 receptor (TIR) signal transduction. Thus, inhibition of TRAF6 function could interrupt both CD40 (TNFR family) and IL-1 growth signals, pathways critical to myeloma proliferation. To block TRAF6-mediated IL-1 signaling, we constructed small interfering RNA (siRNA) against TRAF6. We found that siRNA targeting the TRAF6 C-terminal (siTRAF6C) receptor interaction domain specifically reduced only TRAF6 protein expression, without affecting TRAF2 or 5 levels, and substantially interfered with IL-1-induced NF-kappaB and c-Jun/AP-1 activation. Inhibition by siTRAF6C was concentration-dependent. SiTRAF6C also significantly reduced myeloma proliferation and enhanced apoptosis in a similar dose-dependent fashion in vitro. More importantly, marked siTRAF6C growth inhibition was detected in vivo when these cells were implanted into the bone marrow of irradiated normal mice. In contrast, introduction of siRNA derived from the TRAF6 Zn-finger domain or an irrelevant siRNA construct failed to alter cell growth or cell death. These studies suggest that TRAF6 may be a new molecular target to block cell signal transduction important for the survival and proliferation of multiple myeloma cells.

Myeloma survival and proliferation in the bone marrow (BM) depends on the expression of a variety of autocrine and paracrine growth factors, including IL-6, insulin-like growth factor I (IGF-I), vascular endothelial growth factor (VEGF), and hepatocyte growth factor (HGF). We recently discovered an additional new autocrine myeloma growth factor, pleiotrophin (PTN). PTN is an 18kD heparin-binding protein normally expressed during early development and downregulated in adults, but aberrant PTN re-expression has been associated with a variety of aggressive solid tumors including neuroblastoma, glioma, melanoma and lung, breast and prostate cancers. We found that interference with PTN using a polyclonal anti-PTN antibody inhibited the proliferation of myeloma cells in vitro and in vivo by inducing cell cycle arrest but not apoptosis. To determine the mechanism by which PTN stimulates myeloma proliferation we analyzed the expression of the known PTN receptors, syndecans 1 (CD138) and 3, the anaplastic lymphoma kinase (ALK) and the receptor tyrosine phosphatase beta/zeta (RPTPβ/ζ) on myeloma cells. The expression of ALK and RPTPβ/ζ has not previously been investigated in any hematologic malignancy. In addition to syndecan 1 we found that a subset of myeloma cell lines and BM mononuclear cells (BMMCs) from myeloma patients are RPTPβ/ζ+ by RT-PCR, Western blot and flow cytometry. Myeloma cells, however, do not express ALK. RPTPβ/ζ inhibition by PTN binding leads to the accumulation of β-catenin, and downstream activation of Wnt, NF-κB, MAPK and Akt-mediated cell signals in cells from solid tumors, signaling pathways known to contribute to myeloma cell survival and proliferation. The myeloma cell line MM-1S and the SCID-hu myeloma model LAGλ-1 are both CD138+ but RPTPβ/ζ−. Nonetheless, the growth of these cells is inhibited by anti-PTN antibody. CD138 lacks cytoplasmic signaling motifs suggesting that additional novel receptors are required for PTN-stimulated cell growth in these cells. To continue to define PTN-mediated signaling in myeloma cells we compared the induction of PTN-regulated signaling pathways in RPTPβ/ζ+ RPMI 8226 cells compared to RPTPβ/ζ − MM-1s cells by phospho-protein Western blot. We found differences in PTN-stimulated tyrosine phosphorylation between RPTPβ/ζ+ and RPTPβ/ζ −cells. Specifically, we also discovered that RPTPβ/ζ+ RPMI 8226 cells activate the MAPK Erk1/2. In contrast, we found Akt, but not Erk1/2, to be activated by PTN in RPTPβ/ζ − MM-1S cells. We are now continuing our analysis of PTN signaling in these cells. We are using also identifying new PTN receptors on the surface of myeloma cells by affinity chromatography using biotinylated-PTN. These studies will provide potential new targets that should lead to the development of novel targeted anti-myeloma therapies.

Hodgkin and Reed–Sternberg (HRS) cells of classical Hodgkin lymphoma (cHL) and primary effusion lymphoma (PEL) are derived from germinal center (GC) and post-GC B cells, respectively. Neither express many of the B cell genes or surface markers typically expressed by other GC-derived B cell lymphomas or normal B cells. This loss of B cell gene expression is not due to a lack of essential transcription factors, as studies have shown that the ectopic expression of missing transcription factors failed to reactivate endogenous target genes. These results implicate epigenetic mechanisms extinguishing B cell gene expression. Silenced endogenous B cell genes representing a surface receptor, B29 (Igβ, CD79b), a signaling molecule, TCL1, and a transcription factor, Bob1 (OCA-B, OBF-1), were reactivated by 5-aza-2′-deoxycytidine, indicating that gene silencing in HRS and PEL cells is due to DNA methylation. Genomic bisulfite sequencing corroborated this prediction and revealed three distinct patterns of methylation for the silenced B29 and TCL1 promoters. These distinct patterns consisted of 5′ promoter CpG methylation alone, 5′ and 3′ promoter CpG methylation sparing sites in the central cores, and complete CpG methylation throughout the promoter regions. The silenced Bob1 promoter showed one pattern of dense CpG methylation at essentially all sites. These consistent patterns predict that, although gene silencing in many HRS and PEL cells mimics appropriate gene silencing, in some cases of complete CpG methylation throughout entire promoters both the activation and targeting of methylation is abnormal.

Pleiotrophin (PTN) is a secreted angiogenic protein. We have recently shown that malignant plasma cells express high amounts of PTN and that it is elevated in MM patient serum. In this study, we show that this protein leads to angiogenesis through a novel mechanism - transdifferentiation of monocytes into endothelial cells. First, we isolated human monocytes (CD14+ cells) from peripheral blood using immunomagnetic bead isolation. We excluded the presence of endothelial cells in these CD14-expressing cells using RT-PCR on 106 monocytes with primers to genes expressed by endothelial cells including Flk-1, Tie2, CD144, and von Willebrand factor (vWF). We cultured these purified monocytes on collagen I for one week in the presence of PTN, mCSF and VEGF. Cells cultured with the combination of mCSF and PTN developed Flk-1-expressing tube-like structures, and the addition of VEGF increased tube formation. Next, we performed RT-PCR analysis with primers to these endothelial genes on monocytes cultured with PTN, mCSF and VEGF following serial dilution in cells (T or B lymphocytes) that lack monocyte or endothelial cell gene expression. Many of the cells expressed Tie2 RNA (>10%), and a smaller proportion (0.1–1%) also expressed Flk-1, CD144 and vWF RNA. In contrast, purified monocytes incubated with mCSF, VEGF, or PTN alone or the combinations of mCSF and VEGF or PTN and VEGF lacked Flk-1 staining, did not form tubes and failed to express endothelial cell RNA. We also assessed these monocytes in three dimensional matrices using Matrigel. Cells treated with mCSF and PTN invaded the matrix and began to form tube-like structures in the three dimensional gels as early as 7 days following culture whereas monocytes treated with mCSF, VEGF, or PTN alone did not form these structures. Next, we transduced human monocytic THP-1 cells that lack PTN expression with PTN-sense or -anti-sense constructs. Using RT-PCR, THP-1 cells transduced with PTN expressed endothelial cell genes and lost expression of the monocyte genes c-fms and CD68. In contrast, endothelial cell RNA was not detected in either THP-1 cells infected with anti-sense or the GFP control vectors. We used Transwell plates to co-culture THP-1 monocytes with human MM RPMI8226 or U266 cells or cell lines lacking PTN expression. We also added serum from MM patients with high levels of PTN or normal controls lacking PTN to THP-1 cells. THP-1 cells cultured with the MM cell lines or MM serum expressed endothelial genes and lost expression of monocyte RNA. Endothelial gene expression was blocked by the addition of an anti-PTN antibody but not by a control antibody. Control serum and cell lines lacking PTN did not induce endothelial gene expression or changes in monocyte RNA expression in the THP-1 cells. These experiments define a previously unrecognized novel mechanism leading to angiogenesis in cancer patients - the transdifferentiation of monocytes into endothelial cells by a factor highly produced by the malignant cells in MM. These findings also suggest a potential new specific target, PTN, to inhibit angiogenesis in cancer patients and should have profound clinical implications.

Tie2, an endothelial cell-specific receptor kinase, plays an important role in tumor angiogenesis. This protein is essential to the development of embryonic vasculature as well as vascular growth and maintenance in adult tissues. Because of the increasing importance that angiogenesis has been shown to play in multiple myeloma (MM), we determined the number of Tie2-expressing cells in the peripheral blood (PB) of MM patients and its relationship to the serum levels and gene expression of a recently identified angiogenic factor, pleiotrophin (PTN). We have recently demonstrated that PTN is expressed and secreted by MM tumor cells, and serum levels of this protein are highly elevated in MM patients. We quantified the number of Tie2-positive cells in MM patients (n=15) and age-matched control subjects (n=10) using an immunohistochemical technique. Tie2-expressing cells were significantly elevated in the PB mononuclear cells (MCs) from MM patients compared to the normal controls (p<0.05). We also analyzed gene expression for Tie2 in these same samples using RT-PCR. The results showed that Tie2 mRNA was strongly expressed in the PBMCs from MM patients whereas control samples showed no or low expression of this gene. Serum levels of PTN were tested with ELISA, and PTN mRNA concentrations were quantified by RT-PCR in PBMCs from these same patients and control subjects. The results showed that serum levels of PTN correlated with the number of Tie2-expressing PBMCs in MM patients (R2=0.5778). PTN mRNA levels also correlated with Tie2 gene expression in PBMC samples. We further examined whether monocyte colony stimulating factor (mCSF), PTN and vascular endothelial growth factor (VEGF) may be capable of inducing Tie2 expression in highly purified human monocytes that lack Tie2 expression. Normal PB monocytes were purified using density centrifugation followed by anti-CD14 micro-bead affinity column selection. Although none of these three proteins alone or the combinations of either VEGF and mCSF or VEGF and PTN induced Tie2 gene expression in the monocytes following one week of incubation, the combination of PTN (100 nM) and mCSF (20 nM) led to expression of Tie2 in these cells. We quantified the proportion of cells expressing Tie2 in these samples with RT-PCR using serial dilutional analysis with B or T cells that lack Tie2 expression, and showed that approximately 0.1–1.0% of the monocytes expressed this gene following incubation with PTN and mCSF. Moreover, the addition of VEGF (20 ng/ml) to PTN and mCSF increased the proportion of cells expressing Tie2 (to >10%). Anti-PTN antibody blocked the induction of Tie2 gene expression in these monocytes by this cytokine combination. These results show that Tie2-expressing cells are elevated in the peripheral blood of MM patients, and correlate with PTN serum and PTN mRNA expression. PTN in combination with VEGF and mCSF induces Tie2 gene expression in a large proportion of circulating human monocytes. These results suggest that MM patients show increased numbers of vasculogenic progenitors in their circulation that may result from the presence of elevated levels of circulating angiogenic factors including PTN and VEGF.