Sorafenib, a dual Raf kinase/vascular endothelial growth factor receptor inhibitor has significant...

Sorafenib, a dual Raf kinase/vascular endothelial growth factorreceptor inhibitor has significant anti-myeloma activity andsynergizes with common anti-myeloma drugs

V Ramakrishnan1, M Timm1, JL Haug1, TK Kimlinger1, LE Wellik1, TE Witzig1, SVRajkumar1, AA Adjei2, and S Kumar11Division of Hematology, Mayo Clinic, Rochester, MN, USA2Department of Medicine, Roswell Park Cancer Institute, Buffalo, NY, USA

AbstractMultiple myeloma is characterized by increased bone marrow neovascularization driven in part byvascular endothelial growth factor (VEGF). In addition, the Ras/Raf/MEK/ERK pathway iscritical for the proliferation of myeloma cells and is often upregulated. Sorafenib (Nexavar) is anovel multi-kinase inhibitor that acts predominantly through inhibition of Raf-kinase and VEGFreceptor 2, offering the potential for targeting two important aspects of disease biology. In in vitrostudies, sorafenib-induced cytotoxicity in MM cell lines as well as freshly isolated patientmyeloma cells. It retained its activity against MM cells in co-culture with stromal cells or withinterleukin-6, VEGF or IGF; conditions mimicking tumor microenvironment. Examination ofcellular signaling pathways showed downregulation of Mcl1 as well as decreased phosphorylationof the STAT3 and MEK/ERK, as potential mechanisms of its anti-tumor effect. Sorafenib inducesreciprocal upregulation of Akt phosphorylation; and simultaneous inhibition of downstreammTOR with rapamycin leads to synergistic effects. Sorafenib also synergizes with drugs such asproteasome inhibitors and steroids. In a human in vitro angiogenesis assay, sorafenib showedpotent anti-angiogenic activity. Sorafenib, through multiple mechanisms exerts potent anti-myeloma activity and these results favor further clinical evaluation and development of novelsorafenib combinations.

Keywordsvascular endothelial growth factor; myeloma; angiogenesis; proliferation; apoptosis;microenvironment

IntroductionThe tumor microenvironment has an important role in myeloma and new treatments need totarget the tumor as well as the microenvironment to be effective. Demonstration of increasedbone marrow (BM) angiogenesis and studies highlighting the relevance of endothelial cell–myeloma cell interactions provides a compelling rationale for use of anti-angiogenic agentsin multiple myeloma (MM) (Vacca et al., 1994; Rajkumar et al., 2002; Kumar et al., 2004a).

© 2010 Macmillan Publishers Limited All rights reservedCorrespondence: Dr S Kumar, Division of Hematology, Mayo Clinic and Foundation, 200 First Street SW, Rochester, MN 55905,USA. [email protected] of interest The authors declare no conflict of interest

NIH Public AccessAuthor ManuscriptOncogene. Author manuscript; available in PMC 2010 July 21.

Published in final edited form as:Oncogene. 2010 February 25; 29(8): 1190–1202. doi:10.1038/onc.2009.403.

NIH

-PA Author Manuscript

NIH


NIH


Although several cytokines are implicated in the angiogenesis in multiple myeloma (MM),vascular endothelial growth factor (VEGF) is important and interruption of VEGF signalingmay have therapeutic potential. The interaction between the tumor cells and themicroenvironment is mediated through various mechanisms including cytokines such asVEGF, IL-6, IGF-1 and HGF among others. The Ras/Raf/MEK/ERK pathway liesdownstream of the receptors for these cytokines and has an important role in this disease(Uchiyama et al., 1993; Vacca et al., 1994; Ferlin et al., 2000; Podar et al., 2001; Rajkumaret al., 2002; Rowley and Van Ness, 2002; Kumar et al., 2004a). It enables activated cellsurface receptor–tyrosine–kinases to convey growth signals to the cell nucleus and thusinfluence transcriptional activity leading to cell cycle progression, downregulation of pro-apoptotic pathways and enhanced cell motility. The blockade of Ras/Raf/MEK/ERKpathway can induce apoptosis of MM cells even in the presence of stroma, which typicallyprotects them from conventional drugs such as dexamethasone (Chatterjee et al., 2002,2004). This pathway can also be upregulated by oncogenic activation of Ras, an event foundwith increasing frequency in the late stages of myeloma (Neri et al., 1989; Paquette et al.,1990; Portier et al., 1992; Liu et al., 1996; Bezieau et al., 2001). In newly diagnosed MM,Ras mutations can be observed in one third of the patients and appeared to correlate withshorter survival regardless of the response to treatment (Liu et al., 1996) and its acquisitionappears to correlate with disease progression. (Corradini et al., 1993; Brown et al., 1994;Pope et al., 1997) Given the important role of the Raf pathway in tumor progression in MM,it is only logical that it should be examined as a potential therapeutic target in MM.

Sorafenib is a bisaryl urea designed to specifically target Raf kinase by binding to theadenosine triphosphate binding site of Raf kinase. (Strumberg, 2005; Strumberg and Seeber,2005; Strumberg et al., 2005) Sorafenib has shown in vitro and in vivo efficacy in a broadrange of cancers including renal cell, hepatocellular, colon, breast, pancreas and ovariancancer and is currently approved for treatment of renal cell carcinoma. Given the importanceof Raf/MEK/ERK pathway and VEGF in myeloma biology, we examined the in vitroactivity of sorafenib as well as its potential mechanisms of action with the eventual goal ofdeveloping a rationale for its evaluation in clinical trials.

ResultsSorafenib inhibits the growth of multiple myeloma cell lines

Treatment of myeloma cell lines (RPMI 8226, ANBL-6, KAS-6/1, MM1.S, OPM-2, LR5,Dox40 and MM1R) with sorafenib for 48 h resulted in a dose-dependent growth inhibition(Figure 1a, not all cell lines shown). The median growth inhibitory concentration ofsorafenib was around 5 μM at 48 h with a range from 1 to 10 μM observed between cell lines.Maximum inhibition was observed at 48 h of incubation after a single treatment, with littleadditional effect observed at 72 h (data not shown). A similar degree of growth inhibitionwas also observed with two interleukin (IL)-6-dependent cell lines, ANBL-6 and KAS-6/1.More importantly, dose-dependent growth inhibition was observed with drug-resistantmyeloma cell lines MM1.R, LR5 and Dox-40, albeit at higher doses compared with therespective parental cell line (MM1.S, RPMI 8226).

Sorafenib overcomes the protective effect of BM microenvironment on MM cellsGiven that tumor microenvironment protects myeloma cells against cytotoxic effects ofvarious drugs, we examined if sorafenib can overcome this resistance. The tumormicroenvironment was simulated in vitro either by co-culture of myeloma cells (MM1.Scells) with BMSC or human umbilical vein endothelial cells or by growing myeloma celllines in the presence of different cytokines such as IL-6, VEGF and IGF-1. Although theBMSC (Figure 1b) and the human umbilical vein endothelial cells (Figure 1c) can stimulate

Ramakrishnan et al.

Oncogene. Author manuscript; available in PMC 2010 July 21.

NIH


NIH


NIH


the growth of the myeloma cells as measured by thymidine uptake, treatment with sorafenibcan overcome their protective effect on MM1S cells. In addition, sorafenib can inhibitcytokine (IL6 or VEGF or IGF)-induced increase in proliferation as observed by thymidineuptake (Figure 1d).

Sorafenib induces apoptosis of myeloma cell lines and primary myeloma cellsWe next examined if the cytotoxic effects of sorafenib were mediated through the inductionof apoptotic cell death. Sorafenib-induced apoptosis in MM1.S myeloma cell lines in a time-dependent manner as measured by flow cytometry using Annexin/PI staining. At 6-h post-treatment with sorafenib there was a minimal increase in apoptosis. At 24-h post-treatmentwith sorafenib there was a significant increase in apoptotic cells as indicated (Figure 2a).Immunoblotting of cellular lysates after sorafenib treatment showed a time-dependentcleavage of PARP, confirming induction of apoptosis. In addition, by performing bothwestern blotting and flow cytometry we can observe a time-dependent cleavage of caspases3, 8 and 9 in MM1.S cells confirming involvement of the intrinsic and extrinsic apoptoticpathways (Figure 2b). Sorafenib can induce cytotoxicity in ZVADfmk pretreated and non-ZVADfmk treated myeloma cells at similar levels indicating that although sorafenibtreatment leads to increase in caspase cleavage, it can induce apoptosis by caspase-independent mechanisms as well (data not shown).

We then treated primary myeloma cells with increasing doses of sorafenib and the degree ofapoptosis induction was determined by flow cytometric evaluation of Apo 2.7 staining as amarker of apoptosis. (Figure 2c) Comparison of treated cells and the untreated controlsrevealed varying degree of apoptosis that appeared to be dose dependent in some of thepatients. However, significant heterogeneity was observed between patient samples withrespect to their sensitivity to sorafenib. We also studied unsorted marrow cells from patientswith myeloma to evaluate the differential effects, if any, of the drug on the CD45 positiveand negative plasma cell populations given the biological differences between these two setsof plasma cells (Kumar et al., 2005b). When plasma cells were identified by their CD38expression, both the CD45 positive and negative cells were affected by treatment (Figure2d). To validate the cytotoxic effects of sorafenib on patient cells, we performed an MTTassay on two patients. Sorafenib can induce cytotoxicity on both the patients although atdifferent doses (Figure 2e). Sorafenib can induce cytotoxicity on patient 1 primary cells onlyat 20 μM whereas it can kill patient 2 primary cells at concentrations as low as 5 μM onceagain confirming the heterogeneity among patient samples.

Mechanisms of anti-myeloma activity of sorafenibWe then examined the intracellular events leading to induction of apoptosis to identifypotential mechanisms of action of sorafenib in myeloma cells. The changes were examinedboth at a protein level by immunoblotting as well as at a gene expression level using microarrays.

First, we examined pathways known to be important for myeloma cell proliferation andsurvival (Figure 3a). Treatment of myeloma cells lines (MM1.S and OPM-2) resulted intime-dependent downregulation of STAT3 phosphorylation. Consistent with sorafenib’seffect on the Raf/MEK/ERK pathway, we saw a time-dependent downregulation of ERKphosphorylation. However, we observed a transient upregulation of Akt phosphorylation,which returned to baseline by 6 h. As reported with sorafenib earlier, we observed adownregulation of Mcl1 after sorafenib treatment. Repeating the experiment in the presenceof the pan-caspase inhibitor ZVADfmk did not significantly affect the Mcl1 downregulation(data not shown).

Ramakrishnan et al.


NIH


NIH


NIH


We then specifically examined the effect of IL-6 and VEGF-mediated signaling and theeffect of drug treatment on these pathways. Pretreatment of myeloma cells (MM1S) withsorafenib resulted in abrogation of STAT3 phosphorylation induced by both VEGF and IL-6(Figure 3b). Similarly, the Akt phosphorylation induced by IL-6 was also abrogated by thepretreatment with sorafenib. This also led to abrogation of the Bcl-xL upregulation typicallyobserved with IL-6 and is responsible for some of the anti-apoptotic effects of IL-6. We thenexamined the effect on Mcl-1, given the ability of IL-6 and VEGF to upregulate Mcl-1 inmyeloma cell lines. Treatment with sorafenib resulted in suppression of the upregulation inMcl-1 levels in MM1.S cells observed after treatment with IL-6 or VEGF. These resultsconfirm the ability of sorafenib to target signal transduction pathways in the presence ofcytokines indicating the potential of this drug in vivo.

Gene expression profiles of myeloma cells were determined at three time points afterexposure to sorafenib (4, 8 and 24 h). A total of 261 genes were at least 10-folddifferentially expressed at any time point compared with the untreated sample, 77 weredownregulated and 139 genes were upregulated. We specifically identified genes thatincreased or decreased in a time-dependent manner, because these were most likely to bemediating some of the effects we see with treatment and can provide clues to mechanisms(Figure 3c). The genes included those involved in glucocorticoid receptor signaling (hsp70,c-fos), oxidative stress (hsp70, hsp90), small GTP-mediated signal transduction (HistoneH2), ECM remodeling and cell adhesion (serpine 2, MMP 10, HB-EGF, fibronectin andcollagen 1), hypoxia-induced HIF activation (hsp70, p53), ubiquitin pathway (hsp70),apoptosis (p53, Bcl2, Bim, GADD45, hsp70) and VEGF signaling (Neuropilin 1, c-fos,fibronectin).

Sorafenib synergizes with common myeloma drugs as well as the mTOR inhibitorrapamycin

We first examined the effect of combining sorafenib with commonly used myeloma drugssuch as dexamethasone and the proteasome inhibitor bortezomib. The combination caninduce synergistic killing of myeloma cells at different dose combinations (Figures 4a andb). The synergistic nature of the combination was confirmed by examination of thecombination index (CI) values using Chou–Talalay method in Calcusyn software (Biosoft,Ferguson, MO, USA). Synergy was not observed when concentrations of either of the drugwere reduced below the lowest dose indicated in the respective figures (Figures 4a and b).

Given that the transient upregulation of Akt phosphorylation after treatment of myelomacells could have potential effect of cell survival, as well as the importance of the PI3K/Aktpathway in the biology of myeloma, we examined the effect of an mTOR inhibitor on thesorafenib-induced cytotoxicity. An mTOR inhibitor was chosen because this wasdownstream of pAkt and because this class of drugs is currently available in the clinic andwould be amenable to incorporate in clinical trials. We examined the functional effect ofmTOR inhibition by combining sorafenib with rapamycin (an mTORC1 inhibitor) andobserved a synergistic effect on the cytotoxicity across a spectrum of doses (Figure 4c) aswell as in multiple cell lines (Figure 4d). The synergy was confirmed using an isobologramanalyses, which showed CI values of <1. Fa and CI values were calculated using Chou–Talalay method in Calcusyn software.

Sorafenib treatment effects the microenvironment and its interaction with myeloma cellsGiven the potent anti-VEGF receptor 2 antagonist activity of sorafenib, we examined itsability to inhibit angiogenesis in the context of myeloma. We have shown earlier the abilityof marrow plasma from patients with myeloma to stimulate angiogenesis in an in vitrohuman angiogenesis kit (Angiokit, TCS Cellworks, Buckinghamshire, UK) (Kumar et al.,

Ramakrishnan et al.


NIH


NIH


NIH


2004b). In this assay system, sorafenib treatment resulted in a dose-dependent inhibition ofthe tubule formation induced by myeloma marrow plasma with the effects obvious at verylow concentrations of the drug (Figure 5a). In the presence of the positive control (VEGF)and the negative control (suramin) there is increased and decreased tubule formationrespectively compared with the control (100%). Addition of myeloma marrow plasmaresulted in significant increase in the tubule formation, which was abrogated by sorafenib.

Previous studies have shown the enhanced secretion of VEGF by myeloma cells and themarrow stromal cells when they are grown in co-culture and the ability of VEGF to induceIL-6 secretion by the stromal cells. In addition, IL-6 has been shown to augment VEGFsecretion in co-culture. Given the ability of sorafenib to inhibit VEGF receptor and Rafkinase activity, we examined the ability of sorafenib to inhibit this upregulation of VEGFand IL-6 secretion. In a co-culture of myeloma cells (MM1.S) and marrow stromal cells,treatment with sorafenib at a dose (1 μM) that significantly lower than the median cytotoxicdoses, resulted in a dose-dependent decrease in the VEGF (Figure 5b) and IL-6 secretion(Figure 5c). These data confirm the ability of sorafenib to modulate the marrowmicroenvironment in myeloma.

DiscussionNew approaches to myeloma treatment must take advantage of the recent improvements inour understanding of the disease biology, especially the mechanisms of myeloma cellsurvival and its potential interactions with the microenvironment. This study represents suchan effort, to adapt a novel targeted agent with known safety profile for treatment ofmyeloma. We have presented evidence that would form the rationale for evaluation ofsorafenib, a Raf kinase/VEGF receptor 2 inhibitor, either alone or in combination with otherdrugs, for treatment of myeloma.

We show potent activity of sorafenib on a wide spectrum of myeloma cell lines and primarypatient cells. Interestingly, sorafenib had similar effect on the CD45− and + myelomapatient cells, which is important from a disease biology perspective. The CD45 positive PCsare believed to be the proliferative compartment and more dependent on themicroenvironment and cytokines, showing higher density of cytokine receptors such asVEGF, and hence potentially more sensitive to the action of this class of drugs (Kumar etal., 2005b; Kimlinger et al., 2006). In contrast, the CD45–PCs are likely less dependent onthe microenvironment signals depending more on constitutive activation of the signalingpathways and may be more susceptible to Raf kinase inhibition. The activity of this Rafkinase inhibitor is consistent with those described with other inhibitors of the pathway.Farnesyl transferase inhibitors, which inhibit farnesylation of the Ras and its membraneassociation, have been shown to be cytotoxic to myeloma cells that are resistant todexamethasone and other chemotherapeutic agents (Frassanito et al., 2002). MEK inhibitorshave been shown to induce apoptosis in myeloma cells (Dai et al., 2002). However, ERKinhibition in the myeloma cell line RPMI 8226, which harbors an activated K-Ras allele, didnot result in cell death showing the presence of other signaling pathways and highlightingthe importance of targeting an upstream mediator (Zhang and Fenton, 2002). Ability ofsorafenib to downregulate this pathway is confirmed by the downregulation of ERKobserved in the myeloma cell lines after treatment. Examination of the cellular signalingpathways identifies effects of sorafenib on multiple survival and proliferative signals,including those mediated by the MEK/ERK pathway as discussed as well as the JAK/STATand the PI3K/ Akt pathways. We can see an effective downregulation of the STAT3phosphorylation by sorafenib that can overcome the stimulatory effect of IL-6, a survivalcytokine for MM cells (Lentzsch et al., 2003). STAT3 has been shown to be constitutivelyupregulated in tumors and this upregulation leads to the aberrant activation of anti-apoptotic

Ramakrishnan et al.


NIH


NIH


NIH


proteins including BclXl and Mcl1 and cyclins. Upregulation of phospho-STAT3 levelshave been reported in BM plasma cells of myeloma patients and in the myeloma line U266.Inhibiting JAK/STAT pathway leads to downregulation of anti-apoptotic proteins leading toincreased apoptosis in myeloma cell lines (Catlett-Falcone et al., 1999a, b). Clearly thesimultaneous downregulation of MEK/ERK and JAK/STAT pathways can contribute to theanti-myeloma activity of sorafenib.

Given the importance of Mcl-1 in the survival of myeloma cells and previous reports ofMcl-1 regulation by sorafenib in other tumors, we specifically examined the effect on Mcl-1expression in myeloma cells (Le Gouill et al., 2004; Rahmani et al., 2005; Yu et al., 2005).We observed a time-dependent downregulation of Mcl-1 after treatment with sorafenib inmyeloma cell lines. Sorafenib can completely abrogate the stimulation of Mcl-1 expressiontypically induced by IL-6 and VEGF in myeloma cells. Pretreatment of the myeloma cellswith ZVAD-fmk, a pan caspase inhibitor resulted in only a minimal effect on the Mcl-1downregulation after exposure to the drug (data not shown) ruling out the possibility ofcaspase-mediated degradation of Mcl-1. Puthier et al. had shown earlier that the JAK/STATpathway and not the Ras/Raf/MEK/ERK pathway is involved in IL-6-induced Mcl-1expression suggesting that the effect of sorafenib on Mcl-1 expression may not be related toits ability to downregulate the Ras/Raf/MEK/ERK pathway (Puthier et al., 1999). Thesefindings are consistent with those reported in leukemia cell lines, in which the effect wasmediated in most part through a rapid decrease in Mcl-1 translation (Rahmani et al., 2005).Other studies have suggested an inhibitory effect of sorafenib on Mcl1 transcription in lungcancer cell lines (Yu et al., 2005).

Interestingly, we saw a paradoxical upregulation of Akt phosphorylation after treatment withsorafenib, confirming the presence of cross-talk between the PI3K/Akt and the Ras/Raf/MEK/ERK pathway observed in other studies. Although IL-6 can induce Ras/Raf/MEK/ERK pathway activation, this appears to be partly mediated through cross-talk from the PI3-K pathway because PI3-K inhibitor LY294002 partially blocked IL-6-triggered MEK/ERKactivation and proliferation in MM (Hideshima et al., 2001). In contrast, IL-6-mediatedactivation of PI3-K in MM tumor cells is at least partly mediated by signaling through Ras-dependent pathways (Hsu et al., 2004) and in this setting inhibition of Raf kinase may leadto increased Ras-mediated PI3K activation and explain the upregulation of Aktphosphorylation observed here. Conversely, treatment of myeloma cells with a selectivePI3-K inhibitor lead to MEK activation showing the presence of cross-talk between thepathways (Hideshima et al., 2006).

Given the role of PI3K/Akt pathway in survival and drug resistance of myeloma cells, wedecided to test the functional effect of this upregulation by targeting one of the downstreammediators in this pathway. mTOR inhibitors rapamycin and CCI-779 can inhibit IL-6-induced plasma cell proliferation by preventing p70 activation and 4E-BP1 phosphorylation(Shi et al., 2002). Given the importance of mTOR and the availability of clinically testeddrugs that can inhibit it, we tested the effect of adding rapamycin to sorafenib. There was aclear cut synergy, confirmed by isobologram analysis, between sorafenib and rapamycinconfirming the functional consequence of the pAkt upregulation. However, this does notexclude the possibility of active mTORC2 leading to an increase in rictormediated increasein pAkt levels, which in turn might have reduced the degree of synergy we saw. In addition,we also examined and confirmed synergistic combination of sorafenib with bortezomib anddexamethasone, combinations that should be evaluated through clinical trials.

The gene expression profiling of the myeloma cells after exposure to sorafenib, whereaslimited by the fact that it is representative of only one cell line, raises interesting findingsand hypotheses. We specifically focused on genes that were consistently modulated by

Ramakrishnan et al.


NIH


NIH


NIH


sorafenib in a time-dependent manner. One of the genes differentially regulated was the heatshock protein hsp70, the gene being nearly five log overexpressed by 24 h. This is likely astress response because one of the major roles for this heat shock protein is protection ofcells from apoptosis and previous studies have also shown a role for hsp70 in mediatingdrug resistance. The PI3K/Akt pathway has been shown to be capable of upregulating hsp70transcription and the upregulation of pAkt observed in our experiments may be having a role(Liu et al., 2006). Ongoing studies are examining the combinations of heat shock proteininhibitors with sorafenib. The GADD45 (growth arrest and DNA damage-inducible) familyof genes is involved in the regulation of cell cycle progression and apoptosis and is oftenupregulated by cellular stress. Other mediators of the apoptotic pathway were up to 13-foldupregulated and these include p53, bcl2 and Bim.

In this study, we observed potent effect of sorafenib on the microenvironment in terms of itsanti-angiogenic effect as well as its ability to modulate the interaction between myelomacells and stromal cells. Sorafenib can abrogate the angiogenic ability of myeloma marrowplasma, which could not overcome by VEGF. Importantly, this effect was obvious atconcentrations nearly one log lower than the median inhibitory doses for myeloma cells. Theanti-angiogenic effect observed here is similar to that observed with sorafenib in other tumorsystems (Wilhelm et al., 2004). VEGF is an important mediator of myeloma cell–stromalcell interaction and when myeloma cells come in contact with stromal cells there increasedVEGF secretion by myeloma cells, stromal cells and the endothelial cells, which in turnleads to stromal cell IL-6 secretion (Gupta et al., 2001). The Ras/Raf/MEK/ERK pathwayhas an important role in VEGF secretion and studies have shown that downregulation ofERK can inhibit VEGF secretion by myeloma cells (Giuliani et al., 2004). Consistent withthe role of this pathway in VEGF secretion, we observed decreased VEGF secretion in a co-culture system containing myeloma cells and BMSC in presence of non-cytotoxic doses ofsorafenib. We also noticed decreased secretion of IL-6 consistent with the role of VEGF instimulating IL-6 secretion by the BMSC. These effects of sorafenib will potentially have arole in indirectly decreasing the myeloma cell proliferation, abrogating some of the drugresistance phenotype and enhancing the activity of other drugs.

Materials and methodsMM cell lines and BM stromal cells

Multiple myeloma cell lines included dexamethasone sensitive (MM1.S) and resistant(MM1.R) cell lines; RPMI 8226, OPM-2, NCI-H929, U266 as well as doxorubicin-resistant(Dox 40), and melphalan-resistant (LR5) RPMI 8226 cell lines. All the cell lines werecultured in RPMI 1640 media (Sigma Chemical, St Louis, MO, USA) that contained 10%fetal bovine serum, 2mM L-glutamine (GIBCO, Grand Island, NY, USA), 100U/ml penicillinand 100 μg/ml streptomycin. Freshly obtained BM aspirates were subjected to Ficoll–Paquegradient separation, and the mononuclear cells were placed in 25mm2 culture flasks inRPMI 1640 media containing 20% fetal bovine serum, 2mM L-glutamine, 100 U/ml penicillinand 100 μg/ml streptomycin. Once the adherent stromal cells (BMSC) were confluent, theywere trypsinized and passaged as needed.

Sorafenib (Nexavar)Sorafenib was provided by Bayer Pharmaceuticals under an agreement with Cancer TherapyEvaluation Program and Bayer. Stock solutions were made in dimethylsulphoxide at aconcentration of 20mM, aliquoted and stored at −70°C to avoid repeated freeze thaw. Forindividual experiments, the drug aliquot was thawed and diluted taking care to avoid finaldimethylsulphoxide concentrations of over 0.01%.

Ramakrishnan et al.


NIH


NIH


NIH


Cell proliferation and cytotoxicity assaysMyeloma cells were incubated in 96-well culture plates in media alone, or with varyingconcentrations of sorafenib for 48 h at 37°C. In the experiments carried out to evaluate theeffect of growth factors, recombinant IL-6 (25 ng/ml), or VEGF (50 ng/ml) were added tothe MM cells and then incubated with or with out sorafenib. To confirm cytotoxicity ofsorafenib, colorimetric assays were performed using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrasodium bromide (MTT; Chemicon International Inc., Temecula, CA, USA) asdescribed earlier (Raje et al., 2004; Kumar et al., 2005a). All experiments were performed intriplicate.

In the experiments using co-culture of myeloma cells with BMSC, cell proliferation wasmeasured using tritiated thymidine uptake (3H-TdR; Perkin Elmer, Boston, MA, USA) asdescribed earlier (Raje et al., 2004; Kumar et al., 2005a). BMSC were plated onto flat-bottomed 96-well plates at a concentration of 5000 cells per well and allowed to adhereovernight. Myeloma cells were added to the wells in medium alone or with sorafenib andincubated for 48 h at 37°C. Cells were pulsed with 3H-TdR during the last 8 h of 48-hcultures, harvested on glass fiber filters using a Combi cell harvester (Molecular Devices,Sunnyvale, CA, USA), and incorporated radioactivity determined using a BeckmanLS6000SC scintillation counter. All experiments were performed in triplicate.

Detection of apoptosis in cell lines and patient tumor cellsApoptosis of tumor cell lines were detected by staining with Annexin fluoresceinisothiocyanate and PI as described earlier (Raje et al., 2004; Kumar et al., 2005a). Briefly,cells were cultured in media alone, or with varying concentrations of sorafenib, andharvested every 6 h for 30 h. Cells were washed twice with ice-cold phosphate-bufferedsaline, resuspended in Annexin binding buffer and incubated with Annexin V-fluoresceinisothiocyanate for 15 min at room temperature. Cells were then washed and resuspended inannexin binding buffer with 10 μg/ml propidium iodide added. The apoptotic fraction wasidentified as Annexin V-positive and PI-negative cells when analysed using a FACSCanto(BD Biosciences, San Jose, CA, USA). Levels of caspase 3, 8 and 9 were indirectlydetermined by the production of FL1 fluorescence by cleaved substrate using reagents fromOncoImmune (Gaithersburg, MD, USA). PhiPhiLux G1D2 was used for the detection ofcaspase 3, awhereas CaspaLux 8 L1D2 and CaspaLux 9 M1D2 were used for the detectionof caspases 8 and 9. All experiments were performed in triplicate.

For evaluation of patient myeloma cells, BM aspirates were subjected to ACK lysis andmyeloma cells separated by positive selection using CD138 coated magnetic beads in aRobosep system. The tumor cells were suspended in RPMI 1640 media containing 10% fetalbovine serum, placed in 24-well plates and cultured for 48 h with or with out sorafenib. Thecells were harvested, washed twice with phosphate-buffered saline and stained with PEconjugated Apo 2.7 antibody for identification of apoptotic cells. Additional samples wereanalysed using unsorted marrow mononuclear cells with the myeloma cells identified usinga CD38+/CD45+ gating strategy to analyse differential effects on the CD45 positive andnegative cells.

Western blottingMultiple myeloma cells were cultured with 5 μM sorafenib and lysed in buffer (50mM HEPES(pH 7.4), 150mM NaCl, 1% Triton X-100, 30mM sodium pyrophosphate, 5mM EDTA, 2mM

Na3VO4, 5mM NaF, 1mM phenylmethyl-sulfonyl-fluoride, 5 μg/ml leupeptin, and 5 μg/mlaprotinin) (Raje et al., 2004; Kumar et al., 2005a). Cell lysates were subjected to sodiumdodecyl sulfate–polyacrylamide gel electrophoresis, transferred to nitrocellulose membraneand immunoblotted with the relevant antibody. To characterize growth signaling,

Ramakrishnan et al.


NIH


NIH


NIH


immunoblotting was carried out using antibodies against pErk1/2, pStat3, pAkt, Akt, Stat3,Erk1/2, Mcl-1 and β-actin (Cell Signaling Technology, Danvers, MA, USA). Tocharacterize proteins involved in apoptosis immunoblotting was carried out using antibodiesagainst PARP, caspases 3, 8 and 9 (Cell Signaling Technology). Antigen-antibodycomplexes were detected using enhanced chemiluminescence (Amersham, ArlingtonHeights, IL, USA). Blots were stripped and re-probed with anti-actin antibody (Santa CruzBiotechnology, Santa Cruz, CA, USA) to ensure equivalent protein loading. All experimentswere repeated on three separate occasions. For cytokine experiments, MM1S cells weretreated with sorafenib (5 μM) for 6 h. In the last 30 min of incubation with the drug, cellswere treated with either IL-6 (25 ng/ml) or VEGF (50 ng/ml) for 0, 10, 20 or 30 min.Lysates were made as above and immunoblotting was carried out using antibodies againstpStat3, pAkt, Mcl-1, Bcl-xl, Stat3, Akt and β-actin (Cell Signaling Technology).

Isobologram analysisThe interaction between sorafenib and other drugs was analysed using the CalcuSynsoftware program. This program is based upon the Chou–Talalay method, which calculates aCI, and analysis is performed based on the following equation:

where (D)1 and (D)2 are the doses of drug 1 and drug 2 that have x effect when used incombination, and (Dx)1 and (Dx)2 are the doses of drug 1 and drug 2 that have the same xeffect when used alone. Data from the MTT viability assay were expressed as the fraction ofcells killed by the individual drug or the combination in drug-treated cells compared withuntreated cells. A CI of 1.0 indicates an additive effect, whereas CI values below 1.0indicate synergism.

Gene expression analysisThe effect of treatment with sorafenib on the gene expression profile of myeloma cells wasstudied using the Affymetrix U133 Plus 2.0 platform (Affymetrix, Santa Clara, CA, USA).RPMI cells were treated with sorafenib (10 μM) and cells were harvested at baseline, and at4, 8 and 24 h of treatment. Three separate experiments were carried out and samples fromcorresponding time points combined to reduce inter experiment variability. Total RNA wasextracted according to manufacturer’s instructions for the Qiagen RNeasy Mini Kit(Valencia, CA, USA). Double-stranded complementary DNA was synthesized from eightmicrograms of total RNA and the purified product was used to create biotin-labeledcomplementary RNA by in vitro transcription (Enzo, New York, NY, USA). Labeledproduct was hybridized onto Affymetrix U133 Plus 2.0 GeneChips. The hybridized arrayswere scanned using a GeneChip 3000 scanner and Affymetrix GeneChip Operating Systemsoftware v.1.3 (Santa Clara, CA, USA) were used to quantitatively analyse the scannedimage. The CEL files were GCRMA normalized on Genespring 7.2 software and thedifferential expression of genes across time points were examined.

Angiogenesis assayWe used the in vitro human angiogenesis kit Angiokit for evaluating the anti-angiogenicactivity of sorafenib (Kumar et al., 2004b). In the test wells, 10 μl of BM plasma, made cellfree by centrifugation was used. Assays were incubated at 37°C with 5% CO2 humidifiedatmosphere. On day 11, residual medium was aspirated, cultures fixed and stained withantibodies to CD31 to detect vessel formation. The degree of tubule formation wasevaluated by light microscopy and quantitated using computerized image analysis(Angiosys, TCS Cellworks). The total tubule length in each test well was expressed as a

Ramakrishnan et al.


NIH


NIH


NIH


percent of the NT control wells. Stimulation of angiogenesis in the test wells was defined astotal tubule length >125% of NT control and inhibition as <75% of NT control.

Enzyme-linked immunosorbent assayThe effect of sorafenib on cytokine secretion by human BMSCs, co-cultured with MM cells,was studied using enzyme-linked immunosorbent assay for IL-6 and VEGF. BMSCs wereharvested and cultured in 96-well plates with varying concentrations of sorafenib with orwithout MM1.S cells. After 24-h incubation, the supernatants were harvested and stored at−70°C until measurement. Cytokines were measured using Duoset ELISA DevelopmentKits (R&D Systems, Minneapolis, MN, USA). All measurements were carried out induplicate.

AcknowledgmentsWe acknowledge Roberta DeGoey and Christy Finke for their assistance with processing of tumor cells and all ofthe patients who provided us with the tumor samples. This work was supported by Mayo Clinic HematologicMalignancies Program, CR20 Award from Mayo Foundation and Career Development Award from LymphomaSPORE. Sorafenib was provided through CTEP drug evaluation program by Bayer Pharmaceuticals.

ReferencesBezieau S, Devilder MC, Avet-Loiseau H, Mellerin MP, Puthier D, Pennarun E, et al. High incidence

of N and K-Ras activating mutations in multiple myeloma and primary plasma cell leukemia atdiagnosis. Hum Mutat 2001;18:212–224. [PubMed: 11524732]

Brown RD, Pope B, Luo XF, Gibson J, Joshua D. The oncoprotein phenotype of plasma cells frompatients with multiple myeloma. Leuk Lymphoma 1994;16:147–156. [PubMed: 7696921]

Catlett-Falcone R, Dalton WS, Jove R. STAT proteins as novel targets for cancer therapy. Signaltransducer an activator of transcription. Curr Opin Oncol 1999a;11:490–496. [PubMed: 10550013]

Catlett-Falcone R, Landowski TH, Oshiro MM, Turkson J, Levitzki A, Savino R, et al. Constitutiveactivation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells.Immunity 1999b;10:105–115. [PubMed: 10023775]

Chatterjee M, Honemann D, Lentzsch S, Bommert K, Sers C, Herrmann P, et al. In the presence ofbone marrow stromal cells human multiple myeloma cells become independent of the IL-6/gp130/STAT3 pathway. Blood 2002;100:3311–3318. [PubMed: 12384432]

Chatterjee M, Stuhmer T, Herrmann P, Bommert K, Dorken B, Bargou RC. Combined disruption ofboth the MEK/ERK and the IL-6R/STAT3 pathways is required to induce apoptosis of multiplemyeloma cells in the presence of bone marrow stromal cells. Blood 2004;104:3712–3721.[PubMed: 15297310]

Corradini P, Ferrero D, Voena C, Ladetto M, Boccadoro M, Pileri A. The mutation of N-ras oncogenedoes not involve myeloid and erythroid lineages in a case of multiple myeloma. Br J Haematol1993;83:672–673. [PubMed: 7686038]

Dai Y, Landowski TH, Rosen ST, Dent P, Grant S. Combined treatment with the checkpoint abrogatorUCN-01 and MEK1/2 inhibitors potently induces apoptosis in drug-sensitive and -resistantmyeloma cells through an IL-6-independent mechanism. Blood 2002;100:3333–3343. [PubMed:12384435]

Ferlin M, Noraz N, Hertogh C, Brochier J, Taylor N, Klein B. Insulin-like growth factor induces thesurvival and proliferation of myeloma cells through an interleukin-6-independent transductionpathway. Br J Haematol 2000;111:626–634. [PubMed: 11122111]

Frassanito MA, Cusmai A, Piccoli C, Dammacco F. Manumycin inhibits farnesyltransferase andinduces apoptosis of drugresistant interleukin 6-producing myeloma cells. Br J Haematol2002;118:157–165. [PubMed: 12100143]

Giuliani N, Lunghi P, Morandi F, Colla S, Bonomini S, Hojden M, et al. Downmodulation of ERKprotein kinase activity inhibits VEGF secretion by human myeloma cells and myeloma-inducedangiogenesis. Leukemia 2004;18:628–635. [PubMed: 14737074]

Ramakrishnan et al.


NIH


NIH


NIH


Gupta D, Treon SP, Shima Y, Hideshima T, Podar K, Tai YT, et al. Adherence of multiple myelomacells to bone marrow stromal cells upregulates vascular endothelial growth factor secretion:therapeutic applications. Leukemia 2001;15:1950–1961. [PubMed: 11753617]

Hideshima T, Catley L, Yasui H, Ishitsuka K, Raje N, Mitsiades C, et al. Perifosine, an oral bioactivenovel alkylphospholipid, inhibits Akt and induces in vitro andin vivo cytotoxicity in humanmultiple myeloma cells. Blood 2006;107:4053–4062. [PubMed: 16418332]

Hideshima T, Nakamura N, Chauhan D, Anderson KC. Biologic sequelae of interleukin-6 inducedPI3-K/Akt signaling in multiple myeloma. Oncogene 2001;20:5991–6000. [PubMed: 11593406]

Hsu JH, Shi Y, Frost P, Yan H, Hoang B, Sharma S, et al. Interleukin-6 activates phosphoinositol-3′kinase in multiple myeloma tumor cells by signaling through RAS-dependent and, separately,through p85-dependent pathways. Oncogene 2004;23:3368–3375. [PubMed: 15021914]

Kimlinger T, Kline M, Kumar S, Lust J, Witzig T, Rajkumar SV. Differential expression of vascularendothelial growth factors and their receptors in multiple myeloma. Haematologica 2006;91:1033–1040. [PubMed: 16870555]

Kumar S, Gertz MA, Dispenzieri A, Lacy MQ, Wellik LA, Fonseca R, et al. Prognostic value of bonemarrow angiogenesis in patients with multiple myeloma undergoing high-dose therapy. BoneMarrow Transplant 2004a;34:235–239. [PubMed: 15170170]

Kumar S, Raje N, Hideshima T, Ishitsuka K, Roccaro A, Shiraishi N, et al. Antimyeloma activity oftwo novel N-substituted and tetraflourinated thalidomide analogs. Leukemia 2005a;19:1253–1261.[PubMed: 15858615]

Kumar S, Rajkumar SV, Kimlinger T, Greipp PR, Witzig TE. CD45 expression by bone marrowplasma cells in multiple myeloma: clinical and biological correlations. Leukemia 2005b;19:1466–1470. [PubMed: 15959533]

Kumar S, Witzig TE, Timm M, Haug J, Wellik L, Kimlinger TK, et al. Bone marrow angiogenicability and expression of angiogenic cytokines in myeloma: evidence favoring loss of marrowangiogenesis inhibitory activity with disease progression. Blood 2004b;104:1159–1165. [PubMed:15130943]

Le Gouill S, Podar K, Amiot M, Hideshima T, Chauhan D, Ishitsuka K, et al. VEGF induces Mcl-1up-regulation and protects multiple myeloma cells against apoptosis. Blood 2004;104:2886–2892.[PubMed: 15217829]

Lentzsch S, Gries M, Janz M, Bargou R, Dorken B, Mapara MY. Macrophage inflammatory protein 1-alpha (MIP-1 alpha) triggers migration and signaling cascades mediating survival and proliferationin multiple myeloma (MM) cells. Blood 2003;101:3568–3573. [PubMed: 12506012]

Liu M, Aneja R, Liu C, Sun L, Gao J, Wang H, et al. Inhibition of the mitotic kinesin Eg5 up-regulatesHsp70 through the phosphatidylinositol 3-Kinase/Akt pathway in multiple myeloma cells. J BiolChem 2006;281:18090–18097. [PubMed: 16627469]

Liu P, Leong T, Quam L, Billadeau D, Kay NE, Greipp P, et al. Activating mutations of N- and K-rasin multiple myeloma show different clinical associations: analysis of the Eastern CooperativeOncology Group Phase III Trial. Blood 1996;88:2699–2706. [PubMed: 8839865]

Neri A, Murphy JP, Cro L, Ferrero D, Tarella C, Baldini L, et al. Ras oncogene mutation in multiplemyeloma. J Exp Med 1989;170:1715–1725. [PubMed: 2681517]

Paquette RL, Berenson J, Lichtenstein A, McCormick F, Koeffler HP. Oncogenes in multiplemyeloma: point mutation of N-ras. Oncogene 1990;5:1659–1663. [PubMed: 2267133]

Podar K, Tai YT, Davies FE, Lentzsch S, Sattler M, Hideshima T, et al. Vascular endothelial growthfactor triggers signaling cascades mediating multiple myeloma cell growth and migration. Blood2001;98:428–435. [PubMed: 11435313]

Pope B, Brown R, Luo XF, Gibson J, Joshua D. Disease progression in patients with multiplemyeloma is associated with a concurrent alteration in the expression of both oncogenes andtumour suppressor genes and can be monitored by the oncoprotein phenotype. Leuk Lymphoma1997;25:545–554. [PubMed: 9250826]

Portier M, Moles JP, Mazars GR, Jeanteur P, Bataille R, Klein B, et al. p53 and RAS gene mutationsin multiple myeloma. Oncogene 1992;7:2539–2543. [PubMed: 1461658]

Ramakrishnan et al.


NIH


NIH


NIH


Puthier D, Bataille R, Amiot M. IL-6 up-regulates mcl-1 in human myeloma cells through JAK /STAT rather than ras/MAP kinase pathway. Eur J Immunol 1999;29:3945–3950. [PubMed:10602002]

Rahmani M, Davis EM, Bauer C, Dent P, Grant S. Apoptosis induced by the kinase inhibitor BAY43-9006 in human leukemia cells involves down-regulation of Mcl-1 through inhibition oftranslation. J Biol Chem 2005;280:35217–35227. [PubMed: 16109713]

Raje N, Kumar S, Hideshima T, Ishitsuka K, Chauhan D, Mitsiades C, et al. Combination of themTOR inhibitor rapamycin and CC-5013 has synergistic activity in multiple myeloma. Blood2004;104:4188–4193. [PubMed: 15319277]

Rajkumar SV, Mesa RA, Fonseca R, Schroeder G, Plevak MF, Dispenzieri A, et al. Bone marrowangiogenesis in 400 patients with monoclonal gammopathy of undetermined significance, multiplemyeloma, and primary amyloidosis. Clin Cancer Res 2002;8:2210–2216. [PubMed: 12114422]

Rowley M, Van Ness B. Activation of N-ras and K-ras induced by interleukin-6 in a myeloma cellline: implications for disease progression and therapeutic response. Oncogene 2002;21:8769–8775. [PubMed: 12483530]

Shi Y, Hsu JH, Hu L, Gera J, Lichtenstein A. Signal pathways involved in activation of p70S6K andphosphorylation of 4E-BP1 following exposure of multiple myeloma tumor cells to interleukin-6.J Biol Chem 2002;277:15712–15720. [PubMed: 11872747]

Strumberg D. Preclinical and clinical development of the oral multikinase inhibitor sorafenib in cancertreatment. Drugs Today (Barc) 2005;41:773–784. [PubMed: 16474853]

Strumberg D, Richly H, Hilger RA, Schleucher N, Korfee S, Tewes M, et al. Phase I clinical andpharmacokinetic study of the novel Raf kinase and vascular endothelial growth factor receptorinhibitor BAY 43-9006 in patients with advanced refractory solid tumors. J Clin Oncol2005;23:965–972. [PubMed: 15613696]

Strumberg D, Seeber S. Raf kinase inhibitors in oncology. Onkologie 2005;28:101–107. [PubMed:15665559]

Uchiyama H, Barut BA, Mohrbacher AF, Chauhan D, Anderson KC. Adhesion of human myeloma-derived cell lines to bone marrow stromal cells stimulates interleukin-6 secretion. Blood1993;82:3712–3720. [PubMed: 8260708]

Vacca A, Ribatti D, Roncali L, Ranieri G, Serio G, Silvestris F, et al. Bone marrow angiogenesis andprogression in multiple myeloma. Br J Haematol 1994;87:503–508. [PubMed: 7527645]

Wilhelm SM, Carter C, Tang L, Wilkie D, McNabola A, Rong H, et al. BAY 43-9006 exhibits broadspectrum oral antitumor activity and targets the RAF/MEK/ERK pathway and receptor tyrosinekinases involved in tumor progression and angiogenesis. Cancer Res 2004;64:7099–7109.[PubMed: 15466206]

Yu C, Bruzek LM, Meng XW, Gores GJ, Carter CA, Kaufmann SH, et al. The role of Mcl-1downregulation in the proapoptotic activity of the multikinase inhibitor BAY 43-9006. Oncogene2005;24:6861–6869. [PubMed: 16007148]

Zhang B, Fenton RG. Proliferation of IL-6-independent multiple myeloma does not require the activityof extracellular signal-regulated kinases (ERK1/2). J Cell Physiol 2002;193:42–54. [PubMed:12209879]

Ramakrishnan et al.


NIH


NIH


NIH


Figure 1.Sorafenib is cytotoxic to multiple myeloma (MM) cell lines including those resistant toconventional drugs and overcomes proliferative effect of BMSCs and human umbilical veinendothelial cells (HUVECs). When MM cell lines were incubated with sorafenib for 48 h,dose-dependent cytotoxic effects were observed as measured using the MTT cell viabilityassay (a). The IC50 value was around 5 μM for most of the cell lines tested. Sorafenibconcentration (μM) is indicated on the X axis and viability (as a percentage of the control) isindicated on the Y axis. Error bars represent one s.d. When MM1.S cells were grown incontact with stromal cells (b) or HUVECs (c), an enhanced proliferation of the MM cellswere observed (measured by thymidine uptake), which was completely reversed whenincubated with sorafenib. The Y axis represents the thymidine uptake (counts per minute)and the X axis represents the drug concentrations. The concentrations of sorafenib used areindicated in the figure and the duration of incubation with sorafenib was 48 h. When MM1Scells were grown with cytokines (25 ng/ml IL-6, 50 ng/ml VEGF,or 50 ng/ml IGF-1)sorafenib can inhibit the increase in cytokine-induced proliferation as measured bythymidine uptake (d). The concentrations of sorafenib used are indicated in the figure andthe time of incubation with sorafenib was 48 h. The Y axis represents the thymidine uptake(counts per minute) and the X axis represents the drug concentrations.

Ramakrishnan et al.


NIH


NIH


NIH


Ramakrishnan et al.


NIH


NIH


NIH


Figure 2.Sorafenib induces apoptosis in MM1.S myeloma cell line and myeloma patient cells. Time-dependent increase in the apoptotic cells was observed when MM1.S multiple myeloma(MM) cells were treated with increasing doses of sorafenib (a). Annexin V-fluoresceinisothiocyanate (FITC) staining is represented on the X axis and propidium iodide (PI)staining is represented on the Y axis. The proportion of cells in each quadrant is as indicatedin the figure. There is a time-dependent decrease in viable cells (lower left quadrant) withconcomitant increase in apoptotic (lower right quadrant) and necrotic (upper right quadrant)cells. When MM1.S MM cells were treated with sorafenib (5 μM), induction of apoptosis inMM cells was accompanied by a time-dependent (0, 3 and 6 h) cleavage of caspase 3,caspase 8 and caspase 9 as shown by flow cytometry (b). Induction of apoptosis is alsoconfirmed by the timedependent (0, 1, 2, 4 and 6) cleavage of caspases 3, 8 and 9 as well asPARP as shown by immunoblotting (b). Sorafenib induces apoptosis of freshly isolatedpatient MM cells when cultured with the indicated drug concentrations for 36–48 h asmeasured using Apo 2.7-PE staining and flow cytometry. Sorafenib concentrations (μM) areindicated and percentage of apoptotic cells (percentage of cells expressing membrane Apo2.7) is indicated on the Y axis (c). Results from 10 different patients are shown; proportionof cells positive for Apo 2.7 is shown on the Y axis. In addition, whole marrow mononuclear

Ramakrishnan et al.


NIH


NIH


NIH


cells were treated with 5 or 10 μM sorafenib for 48 h and the CD38/CD45 plasma cellcompartment was examined using flow cytometry (d). CD38 expression is denoted on the Yaxis and CD45 expression on the X axis. Plasma cells were identified as those cells stainingbrightly for CD38. Both the CD45 positive and negative MM cells appear to be sensitive tothe cytotoxic effects of sorafenib. Primary patient cells were treated with indicatedconcentrations of sorafenib for 48 h and cytotoxic effects of sorafenib on patient cellsmeasured using MTT assays (e). In the above experiments it must be noted that all groups(control cells and sorafenib treated cells) were analysed at the same time. Control groupswere left untreated for the duration of the experiment and analysed along with the sorafenib-treated cells. In the case of drug-treated cells, sorafenib was added at different time points tocells in culture that were set up at the same time. The duration of incubation with sorafenibis indicated in respective figures.

Ramakrishnan et al.


NIH


NIH


NIH


Ramakrishnan et al.


NIH


NIH


NIH


Figure 3.Sorafenib treatment induces changes in proliferation and survival signals in multiplemyeloma (MM) cells. Immunoblots of extracts from MM1.S and OPM2 cells treated withsorafenib (5 μM) for the indicated time periods show consistent downregulation of phospho-ERK, phospho-STAT3 and Mcl-1 (a). In contrast a time-dependent increase in Aktphosphorylation was observed that returned to near baseline levels by 6 h of treatment. Nochange was observed in total Akt, Erk or Stat3. Equal protein loading is confirmed byblotting for actin (a). MM1.S cells were then incubated with Interleukin-6 (IL-6) (25 ng/ml)or VEGF (50 ng/ml) for the indicated times with (+) or without a 6 h pre-incubation withsorafenib (5 μM) (b). IL-6 and VEGF treatments were carried out during the last 30, 20 or 10

Ramakrishnan et al.


NIH


NIH


NIH


min of the 6-h incubation with the drug as shown in the figure. Immunoblots wereperformed for pSTAT3, pAKT, MCl-1, BCl-xl, total STAT3, AKT and β-actin (control).Preincubation with sorafenib partially abrogates the vascular endothelial growth factor(VEGF) and IL-6-induced upregulation of pSTAT3 as well as Mcl1. MM1.S cells weretreated with sorafenib (5 μM) for 4, 8 or 24 h, cells harvested, and subjected to geneexpression profiling using Affymetrix U 133 Plus 2.0 platform (c). The experiment wasrepeated three times and the corresponding samples were combined to reduce experiment toexperiment variation. The output files were imported into Genespring 7.2, GCRMAnormalized and expression levels analysed at each time point. The graph and theaccompanying table show genes that show a consistent time-dependent differentialregulation.

Ramakrishnan et al.


NIH


NIH


NIH


Figure 4.Sorafenib synergizes with the proteasome inhibitor bortezomib and dexamethasone as wellas the mTOR inhibitor rapamycin. Synergistic killing of MM1.S cells were observed whensorafenib was combined with the anti-myeloma drug bortezomib (a) or with dexamethasone(b). MTT assays were performed after 48 h of drug treatment. The sorafenib concentrations(μM) and bortezomib or dexamethasone concentrations (nM) in the combination are as shownon the X axis. The combination indices (CI) are as shown below the X axis. When sorafenibwas combined with the mTOR inhibitor rapamycin, a synergistic cytotoxicity was observedin the MM1.S cells at different dose combinations after 48 h of treatment with the drugs asmeasured using the MTT assay (c). The sorafenib concentrations (μM) and rapamycinconcentrations (nM) in the combination are as shown on the X axis. Similar effects wereobserved across several myeloma cell lines and the figure represents myeloma cells treatedwith 5 μM of sorafenib and 5 nM of rapamycin in combination after 48 h of drug treatment(d).

Ramakrishnan et al.


NIH


NIH


NIH


Figure 5.Sorafenib has anti-angiogenic properties in a myeloma setting and modulates the interactionbetween myeloma cells and stromal cells: the angiogenic potential of the marrow plasmafrom a patient with myeloma was estimated by using the Angiokit Assay system (a). Thepositive control represents the effect of vascular endothelial growth factor (VEGF) treatmentand the negative control was treated with suramin. The initial set of bars shows the effect ofsorafenib alone on the tubule formation in the wells highlighting a dose-dependentinhibition. The subsequent bar (no drug) show near doubling of the tubule formation withmarrow plasma, which then is progressively inhibited by increasing concentration of thedrug with complete blockade observed at 1 μM concentration. MM1.S cells were then grownin co-culture with marrow stromal cells and VEGF (b) or interleukin-6 (IL6) (c) levels in theculture supernatants were measured using enzyme-linked immunosorbent assay (ELISA). Atime-dependent increase in the VEGF and IL-6 is observed in the absence of sorafenib witha significant decrease in the secretion of these cytokines in the presence of sorafenib (1 μM).

Ramakrishnan et al.


NIH


NIH


NIH


Sorafenib, a dual Raf kinase/vascular endothelial growth factor receptor inhibitor has significant...

Documents

Transcript of Sorafenib, a dual Raf kinase/vascular endothelial growth factor receptor inhibitor has significant...