preclinicalactivityoftheoralproteasomeinhibitormln9708...

Cancer Therapy: Preclinical

Preclinical Activity of theOral Proteasome InhibitorMLN9708in Myeloma Bone Disease

AntonioGarcia-Gomez1,2,3, Dalia Quwaider1,MiriamCanavese5, EnriqueM.Ocio1,3, Ze Tian6, Juan F. Blanco3,Allison J. Berger7, Carlos Ortiz-de-Solorzano4, Teresa Hern�andez-Iglesias1, Anton C.M. Martens8,9,Richard W.J. Groen8, Joaquín Mateo-Urdiales4, Susana Fraile1, Miguel Galarraga4, Dharminder Chauhan6,Jes�us F. San Miguel1,2,3, Noopur Raje5, and Mercedes Garayoa1,2,3

AbstractPurpose: MLN9708 (ixazomib citrate), which hydrolyzes to pharmacologically active MLN2238

(ixazomib), is a next-generation proteasome inhibitor with demonstrated preclinical and clinical anti-

myeloma activity, but yet with an unknown effect onmyeloma bone disease. Here, we investigated its bone

anabolic and antiresorptive effects in themyeloma setting and in comparisonwithbortezomib inpreclinical

models.

Experimental Design: The in vitro effect ofMLN2238was tested on osteoclasts and osteoclast precursors

from healthy donors and patients with myeloma, and on osteoprogenitors derived from bone marrow

mesenchymal stem cells also from both origins. We used an in vivo model of bone marrow–disseminated

human myeloma to evaluate MLN2238 antimyeloma and bone activities.

Results:Clinically achievable concentrations ofMLN2238markedly inhibited in vitro osteoclastogenesis

and osteoclast resorption; these effects involved blockade of RANKL (receptor activator of NF-kB ligand)-

induced NF-kB activation, F-actin ring disruption, and diminished expression of aVb3 integrin. A similar

range of MLN2238 concentrations promoted in vitro osteoblastogenesis and osteoblast activity (even in

osteoprogenitors from patients with myeloma), partly mediated by activation of TCF/b-catenin signaling

and upregulation of the IRE1 component of the unfolded protein response. In a mouse model of bone

marrow–disseminated human multiple myeloma, orally administered MLN2238 was equally effective as

bortezomib to control tumor burden and also provided a marked benefit in associated bone disease

(sustained by both bone anabolic and anticatabolic activities).

Conclusion: Given favorable data on pharmacologic properties and emerging clinical safety profile of

MLN9708, it is conceivable that this proteasome inhibitormay achieve bone beneficial effects in addition to

its antimyeloma activity in patients with myeloma. Clin Cancer Res; 20(6); 1542–54. �2014 AACR.

IntroductionMultiplemyeloma is a hematologic malignancy resulting

from the growth of plasma cells within the bone marrow(1). The presence of osteolytic lesions is a hallmark of thedisease, characterized by an increase in bone-resorptiveactivity and number of osteoclasts and impairment of

bone-forming activity and differentiation of osteoblasts(2, 3). Osteoclast hyperactivation and enhanced differen-tiation from precursors is primarily mediated throughincreased secretion of osteoclast-activating factors both bymultiple myeloma cells and other cells from the bonemarrow microenvironment [e.g., receptor activator ofNF-kB ligand (RANKL), chemokine C-C motif ligand 3(CCL3), interleukin (IL)-7, IL-3, activin A], whereas levelsof osteoprotegerin, a soluble decoy receptor for RANKL, arereduced. At the same time, osteoclasts promote multiplemyeloma cell survival and growth by the release of IL-6,CCL3, B-cell–activating factor and a proliferation-inducingligand, creating a vicious cycle between bone lesions andtumor expansion (4, 5). On the other hand, reduced oste-oblast formation and activity is in part sustained throughaugmented expression of Wnt and bone morphogeneticprotein (BMP) signaling antagonists and other osteoblastsuppressive factors (such as DKK-1, secreted frizzled-relatedprotein 2, sclerostin, transforming growth factor-b, hepa-tocyte growth factor, activin A, IL-7, CCL3, and tumornecrosis factor-a; reviewed in refs. 2, 3, 5), and by blockade

Authors' Affiliations: 1Centro de Investigaci�on del C�ancer, IBMCC (Uni-versidaddeSalamanca-CSIC); 2Centro enReddeMedicinaRegenerativa yTerapia Celular de Castilla y Le�on; 3Hospital Universitario de Salamanca-IBSAL, Salamanca; 4Laboratorio de Imagen del C�ancer, Centro deInvestigaci�onM�edica Aplicada, Universidad de Navarra, Pamplona, Spain;5MGH Cancer Center, Massachusetts General Hospital; 6Dana-FarberCancer Institute, Harvard Medical School, Boston; 7Millennium Pharma-ceuticals, Inc., Cambridge, Massachusetts, USA; and Departments of8Cell Biology and 9Immunology, University Medical Center Utrecht,Utrecht, The Netherlands

Corresponding Author: Antonio García, Centro de Investigaci�on delC�ancer, IBMCC (Universidad de Salamanca-CSIC); Campus Miguel deUnamuno, Avda. Universidad de Coimbra s/n. 37007 Salamanca, Spain.Phone: 34-923-294-812; Fax: 34-923-294-743; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-13-1657

�2014 American Association for Cancer Research.

ClinicalCancer

Research

Clin Cancer Res; 20(6) March 15, 20141542

on June 25, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 31, 2014; DOI: 10.1158/1078-0432.CCR-13-1657

http://clincancerres.aacrjournals.org/

of activity of the critical osteoblast transcription factorRunx2/Cbfa1 via cell–cell contact interactions of myelomaand osteoprogenitor cells (6).The incorporation of the first in class proteasome inhib-

itor, bortezomib, to the antimyeloma armamentarium canbe considered one of the major milestones in the treatmentof patients with myeloma, greatly improving the responserates and overall survival in both front-line and relapsed/refractory settings (7). Bortezomib is a dipeptide boronicacid, primarily inhibiting the chymotrypsin-like activity (b5subunit) of the 20S proteasome with slowly reversiblekinetics (110 minutes in an in vitro enzyme assay; ref. 8).Myeloma plasma cells seem to be especially sensitive tobortezomib, and multiple effects have been reported afterbortezomib treatment (e.g., blockade ofNF-kBactivity, cell-cycle arrest, induction of intrinsic and extrinsic apoptosis,inhibition of DNA repair enzymes, and inhibition of adhe-sion to bonemarrow stromal cells), which likely contributeto the clinical success of this agent in multiple myeloma(reviewed in ref. 9).Moreover, bortezomib also restrains theprogression of myeloma bone disease (MBD), as evidencedby changes in bone turnover markers and radiologic datafavoring bone healing (10–12). This beneficial impact ofbortezomibonbonemetabolism is notmerely secondary toits antimyeloma activity, but rather this agent acts directlyboth interfering with osteoclastogenesis and resorption andpromoting osteoblast formation and function, as shown innumerous in vitro studies (13–15), and inmyelomatous andnonmyelomatous in vivo models (16, 17).However, some patients are not sensitive to bortezomib,

andmain limitations associatedwith its use are the eventualdevelopment of resistance and the emergence of peripheralneuropathy (8). This has prompted the development of

structurally different second-generation proteasome inhi-bitors, with the aim of retaining or improving bortezomibefficacy, overcoming bortezomib resistance, and having asafer toxicity profile. In fact, several next-generation protea-some inhibitors are currently under various stages of clinicalevaluation (7).

In this context, MLN9708 was selected from a pool ofboron-containing proteasome inhibitors because of its oralbioavailability and its distinct physicochemical profilecompared with that of bortezomib (8, 18). MLN9708(ixazomib citrate) hydrolyzes to the biologically activecompound MLN2238 (ixazomib), which preferentiallyand reversibly inhibits the b5 chymotryptic-like subunitof the proteasome (IC50 ¼ 3.4 nmol/L), albeit with asubstantially shorter dissociation half-life than bortezo-mib (18 minutes in an in vitro enzyme assay; ref. 8). Thisdifferential binding kinetics is thought to play an impor-tant role in the tissue distribution of MLN2238, with astronger and more sustained proteasome inhibitionin xenograft tumors, and a weaker and less sustained effectin whole blood, as compared with bortezomib (18).MLN2238 has shown in vitro cytotoxicity on multiplemyeloma cells, even on those resistant to conventionaltherapies (including bortezomib), and also inhibitedtumor growth and reduced recurrence in a human plas-macytoma model (19). In fact, clinical trials of orallyadministered MLN9708 (now entering phase III) haveevidenced its tolerability with infrequent peripheral neu-ropathy (20, 21) and clinical antimyeloma activity eitheralone or in combination with other antimyeloma drugs(22). Signs of additional beneficial bone effects ofMLN2238 have been observed in a model of disseminatedmouse plasma cell malignancy (23), suggesting that thisagent, similar to bortezomib, may also display direct reg-ulatory effects on bone remodeling. We have, therefore,conducted in vitro studies to comparatively evaluate theability of MLN2238 (relative to that of bortezomib) topromote the osteogenic differentiation frommesenchymalprogenitors from both healthy donors and patients withmyeloma, and also to inhibit osteoclast formation andfunction. In addition, we present evidence of antimyelomaand global bone anabolic effects of orally administeredMLN2238 in a bone marrow–disseminated human mye-loma mouse model, demonstrating both antiresorptiveand bone anabolic activities of this agent.

Materials and MethodsReagents and immunochemicals

Bortezomib (Velcade) and MLN2238 were provided byMillenniumPharmaceuticals, Inc.Melphalanwaspurchasedfrom Sigma-Aldrich. Macrophage colony-stimulating factor(M-CSF) and RANKL were obtained from Peprotech. All cellculturemedia and reagents were supplied byGibco. Primaryantibodies were directed against: p65, ERK (extracellularsignal–regulated kinase), b-catenin, dephospho-b-catenin,phospho-GSK3b (glycogen synthase kinase 3b), GAPDH(glyceraldehyde-3-phosphate dehydrogenase), and IRE1a

Translational RelevanceSecond-generation proteasome inhibitors are now

being evaluated as putative alternatives to bortezomibfor the treatment of patients with multiple myeloma. Inthis context, MLN9708 (ixazomib citrate), hydrolyzingto pharmacologically active MLN2238 (ixazomib), is anorally bioavailable proteasome inhibitor with demon-strated preclinical and clinical activity against multiplemyeloma, but yet with unknown beneficial effect onmyeloma bone disease (MBD).Our in vitro data demonstrate that clinically relevant

concentrations of MLN2238 promote osteoblast differ-entiation and function and inhibit osteoclast formationand resorption. In a mouse model of disseminatedhuman myeloma, oral administration of MLN2238 wasfound to be at least equally effective as intraperitoneallyadministered bortezomib in the control of myelomagrowth and in prevention of bone loss, sustained byboth bone-forming and anticatabolic activities. Thesedata provide the rationale for clinical assessment of thebone effects of MLN9708 in patients with MBD.

Preclinical Activity of MLN9708 in Myeloma Bone Disease

www.aacrjournals.org Clin Cancer Res; 20(6) March 15, 2014 1543




(inositol-requiring protein 1a) (Cell Signaling Technology),phospho-ERK (Santa Cruz Biotechnology), a-tubulin (Cal-biochem), T-cell factor 4 (TCF4; TCF7L2; Millipore), andCD51/61 (R&D Systems). Rhodamine-conjugated phalloi-din was purchased from Invitrogen.

Cell culturesHuman multiple myeloma cell lines MM.1S and MM.1R

were provided by Dr. Steven Rosen (Northwestern Univer-sity, Chicago, IL). TheMM.1S-luc cell linewas a gift fromDr.C.S. Mitsiades (Dana-Farber Cancer Institute, Boston, MA),whereas RPMI8226 cells purchased from theAmerican TypeCulture Collection were lentivirally transduced to stablyexpress firefly luciferase as in Groen and colleagues (24).Cell viability ofmultiplemyeloma cells inmonoculturewasassessed by MTT assay as previously described (19).

The humanmesenchymal stem cell line immortalized byexpression of the telomerase reverse transcriptase gene(hMSC–TERT) was obtained from Dr. D. Campana (St.Jude Children’s Research Hospital, Memphis, TN). PrimaryMSCsderived from thebonemarrowof healthydonors (n¼5), and patients with multiple myeloma with (n ¼ 5) orwithout (n ¼ 3) osteolytic lesions were generated asdescribed by Garayoa and colleagues (25). These studieswere performed with the approval of the Hospital Univer-sitario de Salamanca Review Board, and samples wereobtained after written informed consent of subjects.

In vitro osteoclast formation, resorption pits, F-actinring formation, and CD51/61 expression

Peripheral blood mononuclear cells (PBMCs) fromhealthy donors were differentiated by culture in osteo-clastogenic medium (containing 25 ng/mL M-CSF and 50ng/mL RANKL) as described in Garcia-Gomez and col-leagues (26). Alternatively, PBMCs from patients withmyeloma were also used. Assays related to osteoclastformation and function included: RANKL-induced NF-kB activation, F-actin ring formation, and CD51/61(aVb3 integrin) expression (preosteoclasts, 14 days dif-ferentiation); resorption capacity (17 days differentia-tion), and osteoclast formation (21 days differentiation),and were performed as in Hurchla and colleagues (27).

In vitro osteoblast differentiation, alkalinephosphatase activity, and mineralization assays

Osteoblasts were generated from mesenchymal osteo-progenitors by culture in osteogenic medium (containing5 mmol/L b-glycerophosphate, 50 mg/mL ascorbic acid,and 80 nmol/L dexamethasone) and assayed as in Garcia-Gomez and colleagues (26). Briefly, primary MSCs (pas-sage 2–3) or the hMSC–TERT cell line were exposed todifferent MLN2238 or bortezomib concentrations andcultured in osteogenic medium for 11 or 21 days foralkaline phosphatase (ALP) activity and matrix mineral-ization assays, respectively. ALP activity was quantified byhydrolysis of p-nitrophenylphosphate into p-nitrophenol(Sigma-Aldrich), whereas mineralization was assessed byquantitative measurement of Alizarin Red staining.

Reverse transcription real-time PCR analysesThe hMSC–TERT cell line was cultured for 14 days in

osteogenic medium (preosteoblasts) and total RNA wasisolated using TRIzol reagent (Invitrogen). TaqMan GeneExpression Assays (Applied Biosystems) were performedaccording to the manufacturer’s instructions: Runx2,Hs01047976_m1; Osterix, Hs00541729_m1; Osteopontin,Hs00959010_m1; Osteocalcin, Hs01587814_g1 and DKK-1, Hs00183740_m1. The endogenous expression ofGAPDHwas used for normalization and relative quantification oftarget gene expression was calculated by the comparativethreshold cycle method.

TCF luciferase reporter assayMSCs (3� 105)were transiently cotransfectedwith either

400 ng/mL TOPflash or FOPflash firefly luciferase reporters(Addgene plasmids #12456-7; ref. 28) and 40 ng/mL pRL-SV40 Renilla luciferase reporter (Promega), using Nucleo-fector II as per the manufacturer’s instructions (Lonza).After overnight recovery, cells were treated with or withoutproteasome inhibitors for 24 hours and luciferase activitywas evaluated using a Dual-Luciferase Reporter System(Promega). The firefly luciferase/Renilla luciferase ratio wascalculated to normalize for transfection efficiency.

Gene silencinghMSC–TERT cells were transfected (three times/week for

3 weeks) using SAFEctin-STEM (Deliverics) as per the sup-plier’s instructions, either with ON-TARGETplus SMART-pool siRNA targeting human IRE1 or ON-TARGETplusnontargeting pool as negative control (Dharmacon).

ImmunoblottingGeneration of protein lysates and Western blotting were

performed following standard procedures (26). A luminoldetection system with p-iodophenol enhancement forchemiluminescence was used for protein detection.

MM.1S-luc coculturesThe hMSC–TERT cell line was plated overnight in 96-

well plates (1.25� 104 cells/well) and MM.1S-luc cells (2�104/well) were then added and cocultured for 48 hours inmedium supplemented with 0.5% fetal bovine serum (FBS).Similarly, MM.1S-luc cells (4 � 103/well) were added topreviously generated preosteoclasts and maintained in thesame low-serummedium for 5 days. MM.1S-luc growth wasassessed by measurement of luciferase activity.

Mouse model of bone marrow–disseminated humanmultiple myeloma

Animal experiments were conducted according to institu-tional guidelines for the use of laboratory animals and afteracquired permission from the local Ethical Committee foranimal experimentation. Bortezomib and melphalan weresolubilized in 0.9% saline, whereas MLN2238 was preparedin a 5% solution of 2-hydroxypropyl-b-cyclodextrin (Sigma-Aldrich). RPMI8226-luc cells (5 � 106) were injectedintravenously into 6-week-oldNOD-SCID-IL-2Rg�/� (NSG)

Garcia-Gomez et al.

Clin Cancer Res; 20(6) March 15, 2014 Clinical Cancer Research1544




mice (Charles River Laboratories) and tumor developmentwas monitored by noninvasive bioluminescence imaging(BLI) with a Xenogen IVIS 50 system (Caliper Life Sciences).After 3 weeks, animals were randomized into four groups(n ¼ 6/group) receiving: vehicle (5% solution of 2-hydro-xypropyl-b-cyclodextrin, 2 times/week by oral gavage),melphalan (2.5 mg/kg, three times/week by oral gavage),bortezomib (0.8 mg/kg, three times/week by intraperito-neal injection), and MLN2238 (7.5 mg/kg, 2 times/

week by oral gavage). Serum levels of human Igl (secretedby RPMI8226-luc cells) were also used as a measureof tumor burden and determined by ELISA (BethylLaboratories).

Microcomputed tomography analysisOne femur of each animal was fixed in 10% formalin and

scanned using a microcomputed tomography (microCT)system(MicroCATII; Siemens) as describedpreviously (26).Analysis of trabecular microarchitecture in the distal femurwas performed using BoneJ (29), a bone morphometryimage analysis ImageJ plugin.

Bone turnover markersCarboxy-terminal telopeptide collagen cross-links (CTX;

indicating bone resorption) and N-terminal propeptide oftype I procollagen (P1NP; indicating bone formation) weremeasured in mice sera using ELISA systems (Immunodiag-nostic Systems).

Histologic and immunohistochemical analysesThe other femur from each animal was fixed in 10%

formalin, decalcified, and paraffin embedded. Bone histo-logic evaluation on paraffin sections was performed afterhematoxylin–eosin or immunohistochemical staining forTCF4 as previously described (26).

Statistical analysesEach in vitro assay was performed at least three times

using MSCs or PBMCs from different individuals, andduplicates (real-time PCR) or triplicates were measuredfor each of the tested conditions. Quantitative data areexpressed as mean � SD or SEM, as specified. The non-parametric Mann–Whitney U test was used for compar-ison of vehicle control and specific treatments, and dif-ferences considered statistically significant for values of P< 0.05 (SPSS Statistics 15.0).

ResultsMLN2238 inhibits in vitromyeloma cell growth even inthe presence of microenvironment support

The antimyeloma activity of MLN2238 and bortezomibon various myeloma cell lines (dexamethasone-sensitiveMM.1S, dexamethasone-resistant MM.1R and RPMI8226)was evaluated after a 48-hour culture by MTT assay (Fig. 1Aand B). In agreement with previous reports, IC50 forMLN2238 ranged from ffi10 to 17 nmol/L (19), whereasbortezomib presented a more potent cytotoxic effect (IC50

A

C

2.5 5 10 20 400

25

50

75

100 MM.1S

MM.1R

% o

f M

TT

Meta

bo

lizati

on

MLN2238 (nmol/L)

0Veh 1 2.5 5 10 50

Bortez (nmol/L) MLN2238 (nmol/L)

1 2.5 5 10 25

100

200

300

400

Veh 1 2.5 5 10 251 2.5 5 100

% o

f R

LU

100

200

300

400

0 1.25 2.5 5 10 20

Bortezomib (nmol/L)

MM cell line viability

MM.1S

MM.1R

RPMI8226

10

MM cell line viabilityB

0

25

50

75

100

DMM.1S growth MM.1S growth

% o

f M

TT

Meta

bo

lizati

on

RPMI8226

% o

f R

LU


MM.1S-luc

MM.1S-luc + MSCsMM.1S-luc

MM.1S-luc + MSCs

Figure 1. MLN2238 is cytotoxic tomultiple myeloma (MM) cells andeffectively suppresses the bonemarrow microenvironmentproliferative advantage to MMcells. A and B, MM.1S, MM.1R,and RPMI8226 cell lines werecultured in the presence ofindicated concentrations ofbortezomib or MLN2238 for 48hours and cell viability wasassessed byMTT assay. C andD,MM.1S-luc cells were coculturedwith thehMSC–TERTcell line for 2days (C) or preosteoclastsderived from healthy donors for 5days (D) inmediumsupplementedwith 0.5% FBS. After thecoculture period, MM.1S-lucgrowth was assessed byluciferase bioluminescencesignal. In all graphs, data, mean�SD of three independentexperiments.






ranging from ffi1.7–3 nmol/L) for the same myeloma celllines (Fig. 1A and B).

MSCs and osteoclasts within the bone marrow microen-vironment have been reported to support growth andsurvival of myeloma cells and to protect multiple myelomacells from chemotherapy-induced apoptosis (4, 30). Wetherefore tested the ability of MLN2238 to decrease thegrowth of MM.1S-luc cells when grown in coculture withMSCs (Fig. 1C) or osteoclasts (Fig. 1D). As expected, higherdoses of both bortezomib and MLN2238 were necessary toinhibit MM.1S-luc cell growth in the coculture as comparedwithmonoculture conditions. Overall, althoughMLN2238clearly overcame the proliferative advantage conferredby microenvironmental cells (Fig. 1C and D), higherMLN2238 concentrations were required to attain similarefficacy as bortezomib on myeloma cells.

MLN2238 inhibits in vitro osteoclastogenesis andresorption

To compare the relative potency of MLN2238 versusbortezomib on osteoclast formation, osteoclastogenic cul-tureswere exposed to awide range of concentrations of eachagent along thedifferentiationprocess. As previously shown(13), bortezomib markedly inhibited the formation ofTRAPþ multinucleated cells (IC50 ffi 1.2 nmol/L) derivedfrom PBMCs of healthy donors. MLN2238 was able tosuppress osteoclastogenesis with similar efficacy but whenadded at higher concentrations (IC50 ffi 10.9 nmol/L; Fig.2A). A higher sensitivity to MLN2238 inhibitory effect onosteoclast formationwas observedwhenusing PBMCs frompatients with myeloma (IC50 ffi 4.8 nmol/L; data notshown). Noteworthy, as observed in representative micro-graphs of Fig. 2A, cell densities in our cultures at 10 nmol/LMLN2238 (a concentration similar to the IC50 for this agenton osteoclast formation) were not significantly reduced,thus, suggesting a selective effect ofMLN2238 on inhibitionof osteoclast formation rather than an effect on cell viability.

We next examined the effect of MLN2238 on osteoclasto-genic cultures establishedoncalciumsubstrate-coated slides.A marked and dose-dependent reduction of the area ofresorbed pits was observed with MLN2238 treatment (Fig.2B), similarly to that already reported for bortezomib(13, 31). Specifically forMLN2238, doses required to inhibitosteoclast resorption were well below the ones required toinhibit osteoclast formation, thus suggesting that this pro-teasome inhibitor not only inhibits osteoclastogenesis butalso has a particular effect on osteoclast resorption.

It is known that early osteoclast differentiation mediatedby RANKL–RANK stimulation is primarily mediatedthrough the adapter protein TRAF6 (TNF receptor-associ-ated factor 6), which then leads to activation of mitogen-activated protein kinases (MAPK; i.e., c-Jun N-terminalkinase, p38, and ERK), NF-kB, and activator protein-1(32). To gain some insight into the mechanistics mediatingMLN2238 effect in inhibition of osteoclast formation andfunction, we first tested its effect on MAPK pathway activa-tion. As shown in Fig. 2C, both MLN2238 and bortezomibinhibited RANKL-stimulated phosphorylation of ERK.

Downstream signaling after RANKL stimulation inosteoclast precursors activates NF-kB transcription factor(32), which is not only a key regulator of osteoclastformation but also of osteoclast function and survival.Both MLN2238 and bortezomib targeted NF-kB signalingon preosteoclasts, preventing RANKL-induced transloca-tion of NF-kB into the nucleus (Fig. 2D). This effect isthought to be due to impaired degradation of I-kB (thenatural NF-kB inhibitor) by the proteasome, which thenremains bound to NF-kB retaining it in an inactive formin the cytoplasm (13).

Relative to othermechanismsmediating the inhibition ofosteoclast resorption, we assessed that preosteoclasts dif-ferentiated under 2.5 nmol/L bortezomib and 10 nmol/LMLN2238 (concentrations still allowing osteoclastogen-esis) were associated with a partial or complete disruptionof the F-actin ring (Fig. 2E), as has alreadybeendescribed forother proteasome inhibitors (33). Furthermore, and relatedto this latter effect, MLN2238 at the same concentrationwascapable of reducing the expression of the aVb3 integrin(CD51/61) on viable 7-AAD� osteoclast precursors (Fig.2F). The reduced levels of aVb3 integrin and the disruptionof the F-actin ring would impair maintenance of the sealingzone and prevent an effective bone resorption (34).

MLN2238 promotes in vitro osteogenic differentiationand matrix mineralization of osteoprogenitor cellsfrom patients with myeloma

To check whether MLN2238 was capable of promotingosteoblast differentiation and function in vitro, primaryMSCs from patients with myeloma (5/8 with bone lesions)were allowed to differentiate in osteogenic medium alongwith different concentrations of the agent, and ALP activitywas determined as a surrogate marker of early osteoblasticactivation. As observed in Fig. 3A, MLN2238 dose depen-dently increased ALP activity in early-stage osteoblasts tolevels similar to that attained by bortezomib. Akin to itseffect on ALP activity, MLN2238 also augmented matrixmineralization and calcium deposition when present alongthe complete osteogenic differentiation process (Fig. 3B).MLN2238 and bortezomib were capable of increasing ALPactivity and/ormineralization ofMSCs both frommyelomapatients with osteolytic lesions (see representative micro-graphs in Fig. 3B) and without osteolytic lesions, althoughbasal levels of these parameters without treatment weregenerally higher in osteoblasts derived from the latterMSCs.Likewise, increased ALP activity and matrix mineralizationwere observed withmesenchymal progenitors derived fromhealthy donors (data not shown).

The relative expression of several late bone formationmarkers (namely, the downstream Osterix transcriptionfactor, osteopontin, and osteocalcin) was also determinedafter proteasome inhibitor treatment ofMSCs frompatientswith myeloma at concentrations yielding maximal induc-tion of ALP activity and bone mineral formation (Fig. 3C).Both proteasome inhibitors significantly augmented theexpression of the mentioned osteoblast markers. Becausebortezomib has been found to suppress DKK-1 expression

Garcia-Gomez et al.





in mouse calvarial cultures and bone marrow stromal cells(14), we testedwhether a similar effect could be obtained inour experimental setting. MLN2238 was also capable ofachieving a modest yet significant reduction of DKK-1mRNA levels in preosteoblasts differentiated from MSCsfrom patients with myeloma (Fig. 3C).

MLN2238 activates TCF4/b-catenin signaling and theUPR response in mesenchymal precursors

Numerous studies have demonstrated a pivotal regula-tory role for the canonical Wnt signaling pathway in com-mitment of mesenchymal progenitors to the osteoblastlineage, in osteoblast function, and in skeletal development

(er

lati

ev

toc

no

trl

o)

0 1 2.5 5 10 25 0 1 2.5 5 10 250

50

100

150

% o

f L

ac

un

ae

are

a(r

ela

tiv

e t

o c

on

tro

l)

0 1 2.5 5 0 2.5 10 250

50

100

150

TRAP+ Osteoclast formation Bone resorption

A B

*

*

* * *

*

*

* *

*

*

*

*

*

Bortezomib (nmol/L) MLN2238 (nmol/L) Bortezomib (nmol/L) MLN2238 (nmol/L)

MLN2238

Vehicle 2.5 nmol/L 10 nmol/L 25 nmol/L

MLN2238

Vehicle 2.5 nmol/L 10 nmol/L 25 nmol/L

Vehicle Bortez 2.5 nmol/L

C D

E

MLN2238 10 nmol/L

DA

PI

p65

Merg

e

1

2.5

2.5

10

ααVβ3 Integrin (CD51/61)

Proteasome

inhibitors

Vehicle

Bortezomib

MLN2238

% of MFI ± SD

1000

nmol/L

81.28 ± 21.77

64.51 ± 15.96

76.05 ± 18.62

70.10 ± 15.87

FVehicle Bortez 2.5 nmol/L MLN2238 10 nmol/L

MLN2238

Bort

ez

VehicleDay 5

p-Erk1/2 42/44 kDa

42/44 kDaErk1/2

GAPDH 37 kDa

*

*

% o

f O

ste

oc

las

ts/w

ell

Figure 2. MLN2238 inhibits in vitro osteoclast formation and resorption. A, PBMCs from healthy donors were cultured in osteoclastogenic medium undercontinuousproteasome inhibitor treatment for 21days, andosteoclastswere identifiedasmultinucleated (�3nuclei) TRAPþcells.Graphs,mean�SDof threeindependent experiments. Representative micrographs of TRAPþ osteoclasts are shown. B, to test the effect of MLN2238 on osteoclast resorption,PBMCs were cultured on calcium-coated slices in the presence of osteoclastogenic medium and continuous proteasome inhibitor treatment as specified.After 17 days, osteoclasts were removed and the area of resorption lacunae analyzed. Data, mean � SD. Representative micrographs of resorptive pitsafterMLN2238 treatment are reported. A andB: �,P < 0.05 versus vehicle control. Bar, 50 mm.C, PBMCswere differentiated in osteoclastogenicmedium for 5days and then exposed to 2.5 nmol/L bortezomib or 10 nmol/L MLN2238 for 24 hours. Both proteasome inhibitors inhibit RANKL-induced ERK1/2phosphorylation as shown by Western blot analysis. D, preosteoclasts (14 days in osteoclastogenic medium) were cultured overnight in the absence ofRANKL, treated for 3 hours with proteasome inhibitors and stimulated with RANKL for 30 minutes before immunostaining for the p65 subunit of NF-kB. Inabsence of proteasome inhibitors (vehicle), RANKL induces p65 translocation to the nucleus, whereas after proteasome inhibitor treatment p65subunit is retained in the cytoplasm. The p65 subunit is visualized in red and the nuclei by DAPI (40,6-diamidino-2-phenylindole) staining. Representativemicrographs of each condition are shown. Bar, 12.5 mm. E, preosteoclasts differentiated under drug exposure were stained with rhodamine-conjugatedphalloidin andexaminedby fluorescencemicroscopy to visualize theF-actin ring. Representativemicrographsare shown.Bar, 50mm.F, in the samesetting asin (E), flow cytometry analysis revealed a decreased expression of CD51/61 (aVb3 integrin) in 7-AAD� cells. Data are expressed as percentage of meanfluorescence intensity (% MFI) � SD. �, P < 0.05 versus vehicle control.






(35, 36). Upon binding of canonical Wnt ligands to appro-priate receptors in osteoprogenitors, b-catenin is stabilizedand translocated to the nucleus where it interacts with amember of the T-cell factor/lymphoid enhancer factor(TCF/LEF) transcription factor family, to synergisticallyregulate the expression of osteoblast differentiating genes(i.e., Runx2; ref. 35). Because TCF4 is the most abundantTCF family transcription factor expressed in osteoblast celllines and primary hMSCs (37), we explored whetherMLN2238 treatment was able to enhance TCF4 transcrip-tional activity using a TOPflash reporter system. WhenMSCs from patients with myeloma were proteasome inhib-itor treated for 24 hours, a dose-dependent increase inluciferase activity of the reporter plasmidwas observed (Fig.4A). Therefore, activation of TCF4 transcriptional activityseems to mediate, at least in part, MLN2238 promotion ofosteoblastogenesis and osteoblast function.

Stabilization and nuclear translocation of b-catenin is alsoused as a metric of enhanced canonical Wnt signaling, andthus, we also tested whether MLN2238 could modulateb-catenin levels. As seen in Fig. 4B, MLN2238 treatment ofMSCs from patients with myeloma slightly upregulated

levels of dephosphorylated b-catenin, which is the stabilizedactive form of b-catenin able to interact with TCF/LEF tran-scription factors. These increased stabilized b-catenin levelsmaybeascribed to amodest inhibitionofGSK3b activity dueto increasedphosphorylation (Fig. 4B; ref. 38). Taken togeth-er, our data show thatMLN2238 treatment leads to increasedTCF4 transcriptional activity along with modest augmentedlevels of active b-catenin, thus, leading to activation ofTCF4/b-catenin signaling on osteoprogenitor cells.

Osteoblasts produce large amounts of extracellularmatrix proteins and must activate the unfolded proteinresponse (UPR) to handle the accumulation of misfoldedproteins. In fact, the IRE1a–XBP1 (X-boxbindingprotein 1)pathway of the UPR has been recently shown to be essentialfor osteoblast differentiation by driving the transcription ofOsterix (39). Increased levels of IRE1a protein were readilyobserved after bortezomib and MLN2238 treatment ofhMSC–TERT cells (Fig. 4C). Conversely, silencing of IRE1amarkedly abrogated proteasome inhibitor–enhanced min-eralization (Fig. 4D and E), thus underscoring the essentialrole of the IRE1a component of the UPR in the promotionof osteoblast function by proteasome inhibitors.

AL

P A

cti

vit

y (

nm

ol/L

/min

/μg

pro

tein

)

Alizari

n R

ed

(n

mo

l/L

)

Mineralization

0 1 2.5 5 0 2.5 5 10


A B

C

Osterix OsteocalcinOsteopontin DKK-10

1

2

3

Rela

tive e

xp

ressio

n

Vehicle

Bortezomib 5 nmol/L

MLN2238 10 nmol/L

*

* **

*

**

*

*

* **

*

*

*

* *

*

**

OB markers

10 nmol/L

5 nmol/L

Vehicle

MLN2238

0

10

20

30

40

50ALP Activity

0.0

0.5

1.0

1.5

2.0

0 1 2.5 5 0 2.5 5 10


Figure 3. MLN2238 increases ALP activity, stimulates matrix mineralization, and augments the expression of bone formation markers in in vitrodifferentiating osteoblasts (OB). A, primaryMSCs frompatients withmyelomawere cultured in osteogenicmedium in the presence or absence of proteasomeinhibitors at indicated concentrations and ALP activity was measured at day 11. Results, mean � SEM. B, matrix mineralization was assessed byAlizarin Red staining and subsequent dye quantification in OBs differentiated from MSCs from patients with MM at day 21. Graphs, mean values � SD.Representative images derived from MSCs of a patient with osteolytic lesions are shown. C, total RNA was isolated from primary MSCs from patientswith MM differentiated toward the osteogenic phenotype for 14 days in the presence or absence of bortezomib or MLN2238. Real-time RT-PCR wasperformed to determine the expression of osteoblastogenicmarkers (Osterix, osteopontin, and osteocalcin), togetherwith the expression of theWnt signalinginhibitor DKK-1. Expression levels for each gene were normalized to GAPDH expression and referred to those in the absence of proteasome inhibitors.Data, mean � SD. �, P < 0.05 versus vehicle control.

Garcia-Gomez et al.





MLN2238 prevents tumor-associated bone loss besidesreducing multiple myeloma tumor burdenWe next explored whether proosteogenic, antiosteoclas-

togenic, and antimyeloma in vitro effects ofMLN2238 couldbe translated into a disseminated murine model of humanmyeloma (27). RPMI8226-luc cells were intravenouslyinjected to NSG mice and myeloma engraftment (as mon-itored by bioluminescence) occurred after 3 weeks. At thispoint mice were randomized to receive vehicle, melphalan,bortezomib, or MLN2238 treatments as indicated (seeMaterials and Methods; Fig. 5A). Melphalan was used asa control drug with expected antimyeloma activity but not abone effect (17). Compared with the vehicle control group,both orally administered MLN2238 and intraperitoneallydelivered bortezomib very efficiently controlled tumor pro-gression, as measured by bioluminescence (Fig. 5A) or by

serum levels of hIgl secreted by RPMI8226-luc cells (Fig.5B). Melphalan treatment, at doses used in our study,reduced myeloma burden at only about 50% of levelsobserved with MLN2238 or bortezomib (Fig. 5A and B).Representative microCT images at the methaphyses of

distal femurs showed evident tumor-associated bone loss in

vehicle and melphalan-treated mice, in contrast with tra-becular structures observed in both MLN2238 and borte-zomib-treated animals (Fig. 6A). This was also reflected bybonemorphometric parameterswhich resulted in increasedtrabecular bone volume and connectivity and reducedtrabecular separation only in proteasome inhibitor–treatedanimals, as compared with vehicle control (Fig. 6B–D). Inagreement with these data, histologic sections of MLN2238and bortezomib-treated mice presented higher number oftrabeculae at the growth plate of the distal femur (Fig. 6E),together with immunoreactive TCF4 osteoblasts lining thetrabeculae (Fig. 6F). Of interest, TCF4-positive osteoblastsin proteasome inhibitor–treatedmice presented a square orrectangular section indicative of active bone matrix secre-tion, whereas flat osteoblasts in vehicle and melphalangroups reflected a resting secretory state (Fig. 6G). Finally,these findings correlated with levels of bone turnover mar-kers analyzed in serum after drug treatment. The boneresorption marker CTX was only significantly diminishedin proteasome inhibitor–treated animals (Fig. 6H). P1NP, abone formation marker, augmented significantly in theproteasome inhibitor–treated groups versus the control

TCF4 Activity

Bortez MLN2238

6 12 24 6 12 24Veh

icle

IRE1α

α-Tubulin 60 kDa

110 kDa

C

A

D

0

500

1,000

1,500

2,000

2,500

Rela

tive l

ucif

era

se u

nit

s

0 5 10 25 NC 0 50 100 250 NC

*

**

*

*

*

0

50

100

siRNA NC IRE1

NC IRE1

IRE1α 110 kDa

α-Tubulin 60 kDa

siRNA

IRE

1α

Rela

tive e

xp

ressio

n

Vehicle Bortez MLN2238

0.0

0.5

1.0

1.5

2.0

NC IRE1 NC IRE1 NC IRE1

Ali

za

rin

re

d (

mm

ol/

L)

siRNA

E

MLN2238

Bort

ez

Vehicle

deP-β-Catenin 92 kDa

92 kDaβ-Catenin

p-GSK3β 46 kDa

37 kDaGAPDH

B

Bortez nmol/L MLN2238 nmol/L

*

* * *

Figure 4. MLN2238 promotes OB differentiation and function through activation of TCF/b-catenin signaling and the UPR. A, primary MSCs from patientswith myeloma were transiently transfected with the TOPflash (containing 7 TCF/LEF–binding sites) or FOPflash (with 6 mutant TCF/LEF–binding sites)luciferase reporters. One day after transfection, MSCs in osteogenic medium were treated for 24 hours with indicated concentrations of bortezomib orMLN2238 and luciferase activity wasmeasured in cell lysates. Results,mean�SD.Negative control (NC) refers to luciferase activity after transfectionwith theFOPflash reporter at 10 nmol/L bortezomib and 100 nmol/L MLN2238. B, MSCs from patients with MM were differentiated for 7 days in osteogenicmedium and treated with 2.5 nmol/L bortezomib or 10 nmol/L MLN2238 for 72 hours. Levels of active (dephosphorylated) b-catenin and inhibited(phosphorylated) GSK3b were evaluated by Western blot analysis. C, proteasome inhibitor treatment (bortezomib 25 nmol/L and MLN2238 100 nmol/L)of hMSC–TERT cells over a 24-hour period in osteogenic medium augmented IRE1a levels as determined by Western blot analysis. D, transfection ofhMSC–TERT cells with IRE1a targeting siRNAs (100 nmol/L, 48 hours) effectively diminished IRE1a expression at the mRNA and protein level. NC refersto IRE1a expression after transfection with nontargeting siRNAs. E, the hMSC–TERT cell line was maintained in osteogenic medium for 21 days andtransfected three times per week with IRE1a or NC siRNAs. When IRE1a was silenced, matrix mineralization was abolished even in the presence ofproteasome inhibitors (5 nmol/L bortezomib; 10 nmol/L MLN2238). Data, mean � SD of at least three independent transfections for each condition, each ofthem run in triplicate. For A, D, and E: �, P < 0.05 versus vehicle or NC.






group (Fig. 6I), whereasmelphalan-treated animals showedno change.

DiscussionOsteolytic lesions are the most common complication of

myeloma, being developed in more than 80% of patientssuffering the disease. These lesions often lead to skeletal-related events (SRE), including focal lytic lesions, patho-logic fractures, hypercalcemia, severe bone pain, and ver-tebral compression fractures (2, 40). Importantly, MBD notonly negatively affects the quality of life of patients withmyeloma, but has been found to be associated withdecreased overall survival (41). This highlights the needfor supportive bone-targeting treatments besides antimye-loma chemotherapies to reduce the risk of bone complica-tions in multiple myeloma. Moreover, because many of thefactors dysregulated in MBD are also important contribu-tors to myeloma progression and chemoresistance, it isexpected that restoring bone homeostasis would indirectlyand additionally lead to tumor inhibition (42).

Bisphosphonates are anticatabolic agents representingthe current standard supportive treatment for patients withmyeloma with MBD. Being pyrophosphate analogs,bisphosphonates become adsorbed to mineralized matrixand cause apoptosis when osteoclasts incorporate them byendocytosis (43). Second-generation nitrogen-containingbisphosphonates now in use, such as pamidronate andzoledronic acid, have dramatically reduced the incidenceof SREs and significantly improved the quality of life inpatients withmultiplemyeloma (40).Nevertheless, adverseside effects may result from bisphosphonate therapy,including osteonecrosis of the jaw, impaired renal function,and accumulation of bone microfractures (2, 44); in addi-tion, MBDmay progress during bisphosphonate treatment,

presumably because these compounds inhibit bone resorp-tion but do not restore bone formation (3). Recent progressin the understanding of themolecular interplayers involvedin MBD has stimulated the development alternative bonetherapeutic compounds either focusing on osteoclast inhi-bition, or importantly, on osteoblastogenesis promotion(reviewed in refs. 2, 40, 42). It is likely that these agents, nowat different stages of clinical development, will contribute toa more efficacious and improved management of MBD inthe near future.

Within current agents used in antimyeloma chemother-apy, bortezomib is unique because of its concurrent anti-myeloma, anticatabolic, and bone anabolic activities, asevidenced in multiple preclinical studies and in bortezo-mib-treated patients (10, 13, 15, 17). This represents anunprecedented advantage for the treatment of both multi-ple myeloma and MBD either as single agent (12) or incertain combinations with other antimyeloma drugs (11).Preclinical studies with second-generation proteasomeinhibitors carfilzomib and ONX0912, have shown that inaddition to their antimyeloma properties (45, 46) andsimilar to bortezomib, they effectively target osteoblast andosteoclast populations to shift the bonemicroenvironmentfrom a catabolic to an anabolic state (27). Akin to theseproteasome inhibitors, herewith, we show that MLN2238also combines antimyelomaactivitywith antiresorptive andbone-forming effects on bone.

As observed in our in vitro experiments, MLN2238markedly inhibited osteoclast formation from humanPBMC progenitors as well as osteoclast resorption. Mech-anistically, these effects on osteoclast function were at leastpartially mediated through inhibition of RANKL-inducedNF-kB signaling, F-actin ring disruption, and diminishedexpression of the aVb3 integrin, molecular sequelae alsoshared by bortezomib and other proteasome inhibitor

0 4 7 10 14 18 22 25 28

0.0

0.5

1.0

1.5

2.0

2.5

Vehicle

Melphalan 2.5 mg/kg p.o. (day 1, 3, and 5) 4 wk

Bortezomib 0.8 mg/kg i.p. (day 1, 3, and 5) 4 wk

MLN2238 7.5 mg/kg p.o. (day 1,4) 4 wk

Vehicle Bortez MLN2238

1.0 2.0 3.0 4.0

Radiance

(photons/sec/cm2/sr)

× 106

hIg

Gλ

(ng

/mL

)

Melphalan

To

tal fl

ux

( ×1

01

2 p

ho

ton

s/s

ec)

Days of treatment

0

500

1,000

1,500

Serum human Igλ

MLN

2238

Bor

tez

Veh

icle

**

*

Mel

phal

an

BLI ImagingA B

** ** **********## ## ## ## ## ## ##

Figure 5. Oral administration of MLN2238 decreased tumor burden in a mouse model of disseminated MM. RPMI8226-luc cells (5 � 106) wereintravenously injected into NSGmice. After 3 weeks, mice were randomized into 4 groups [receiving vehicle (MLN2238 vehicle), melphalan, bortezomib, andMLN2238; n ¼ 6/group], and treated for additional 4 weeks with dosing and regimen schedules as specified. Both intraperitoneal administeredbortezomib and orally delivered MLN2238 effectively decreased tumor burden as measured by BLI (A) and serum levels of human Igl secreted byRPMI8226-luc cells (B). At doses and administration regimens used in this study, melphalan was less effective than proteasome inhibitors in controllingtumor progression. Data (A and B), mean � SD. In A, ��, P < 0.01 bortezomib versus vehicle control at each time point; ##, P < 0.01 MLN2238 versus vehiclecontrol at each time point; in B,�, P < 0.05 versus the vehicle control group.

Garcia-Gomez et al.





treatments (13, 31, 33). Proteasome inhibition has alsobeen linked to induced in vitro differentiation of MSCs intoosteoblasts and enhanced osteoblast function (14, 15, 37).We present here that, in a similar range of concentrationsaffecting osteoclasts, MLN2238 was also able to stimulatethe in vitro osteogenic differentiation of MSCs and to pro-mote osteoblast function; of interest, MLN2238 effects wereobserved on MSCs irrespective of being derived fromhealthy donors or patients with myeloma (including thosefrom patients with osteolytic lesions). As already reportedfor bortezomib and carfilzomib (37, 47), MLN2238 dose

dependently increased TCF transcriptional activity inreporter assays on osteoprogenitors. In relation with TCFtransactivation,we also found thatMLN2238 treatmentwaslinked to modest increased levels of active dephosphory-lated b-catenin (35), all of which would enhance TCF/b-catenin signaling and promote osteoblast function. Inter-estingly, we have also shown that MLN2238 treatmentmarkedly augments levels of the IRE1a component of theUPR, and as evidenced by siRNA strategies, reduced levels ofthis protein impaired matrix mineralization even in thepresence of proteasome inhibitors. This is in agreement

0

5

10

15

20

B

MicroCT imaging MicroCT parameters

H Serum CTX

(bone resorption)

I Serum P1NP

(bone formation)

CT

X (

ng

/mL

)

*

*

**

MLN

2238

Bor

tez

Veh

icle

Mel

phal

an

MLN

2238

Bor

tez

Veh

icle

Mel

phal

an

0

50

100

150

200

P1

NP

(n

g/m

L) *

*

MLN

2238

Bor

tez

Veh

icle

Mel

phal

an

BV

/TV

0.0

0.2

0.4

0.6

Trabecular bone

volume

*

*

0.0

0.1

0.2

0.3

0.4

0.5

Trabecular

separationVehicle

Melphalan

Bortez

MLN2238

A

E

Hem

ato

xylin

an

d

eo

sin

sta

inin

g T

CF

4

sta

inin

g

F

*

*

Connectivity

MLN2238

Tb

.Sp

(m

m)

0

50

100

150

200

250*

*

C D

Co

nn

ec

tiv

ity

BortezomibVehicle MLN2238Melphalan

G

MLN2238

Vehicle

TCF4 staining

MLN

2238

Bor

tez

Veh

icle

Mel

phal

an

MLN

2238

Bor

tez

Veh

icle

Mel

phal

an

BortezomibVehicle Melphalan

Figure 6. MLN2238 preventedmyeloma-associated bone loss.A, representative microCT cross-sections at the metaphyses ofdistal femurs show loss oftrabecular architecture in vehicleand melphalan-treated mice, butnot in mice treated withbortezomiborMLN2238. B toD, ascompared with vehicle controlanimals, MLN2238 significantlyincreased bone parameters oftrabecular bone volume andconnectivity, whereas trabecularbone separation was decreased.Results, mean � SD. �, P < 0.05versus vehicle control. E and F,representative femur sections afterdrug treatment stained withhematoxylin and eosin (E), or forTCF4 immunohistochemistry (F).Bar, 25 mm. Dotted line delimitsinfiltrating RPMI8226-luc cells andhost tissue. OB-like cellsimmunostained for TCF4 can beobserved lining the trabeculae(arrowheads). G, representativeimages at higher magnificationshowing flat OBs lining thetrabeculae of vehicle-treatedanimals (arrows), whereas OBswere cuboidal in shape andpresented TCF4 immunostainingin MLN2238-treated mice(arrowheads). Bar, 12.5 mm. H andI, MLN2238 significantly changedbone turnover markers in serum,increasing levels of P1NP, a boneformation marker, whiledecreasing levels of CTX, a boneresorption marker, as comparedwith vehicle-treated animals. Hand I, graphs,mean values�SEM.�, P < 0.05; ��, P < 0.01 versusvehicle or between indicatedgroups.






with recent reports highlighting the critical role of the UPR,and specifically the IRE1a–XBP1 pathway, in osteoblastdifferentiation through promotion of Osterix transcription(39). Although both myeloma cells and MSCs/osteoblastsare protein secretory cells dependent on the UPR to elim-inate misfolded proteins and to cope with endoplasmicreticulum(ER) stress, proteasome inhibitor treatment exertsapoptosis on the former and promotes osteogenic differ-entiation on the latter. The reason for these differentialeffects may reside on the lower threshold of multiplemyeloma cells for proteasome inhibitor–induced terminalUPR and ER stress–induced apoptosis, because these cellsalready highly express prosurvival UPR components tofunction as secretory cells (48).

Our studies on osteoclasts and osteoblasts, as well asthose on myeloma cell lines alone or in coculture withmicroenvironment cells, clearly showed that MLN2238,although less potent, reached a similar effect as bortezomib.This difference in potency may relate to the shorter protea-some dissociation half-life of MLN2238 compared to bor-tezomib (8), thus, higher concentrations of MLN2238being needed to attain an equivalent effect. Nevertheless,clinical pharmacokinetic studies from once-weekly orallyadministeredMLN9708 estimate plasmaCmax� 150 nmol/L (49), well over the required in vitro doses for the effects ofMLN2238 on osteoblasts and osteoclasts in our studies. It istherefore conceivable that therapeutically administeredMLN9708will achieve antiresorptive andproanabolic boneeffects (besides its antimyeloma activity) in patients withmyeloma.

In fact, as observed in our in vivo murine model,MLN2238 and bortezomib treatments were very efficientin controlling the growth of RPMI8226-luc tumor cells inthe bone marrow, as assessed by decreased BLI and

diminished serum levels of human IgGl. Increased bonetrabecular architecture in MLN2238 and bortezomib-trea-ted animals was evident on microCT images at meta-physes of distal femurs as compared with vehicle control,correlating with significantly augmented bone parametersof trabecular bone volume and connectivity and reducedtrabecular separation. This global bone anabolic effect ofMLN2238 and bortezomib was also associated with (i)increased trabeculae at the growth plate as observed onhistologic sections, (ii) immunostaining for TCF4 andactive morphologic appearance of osteoblast cells liningthe trabeculae, and (iii) reduced CTX (bone resorptionmarker) and increased P1NP (bone formation marker)serum levels. All this evidence clearly supports, in addi-tion to our in vitro data, that these proteasome inhibitorshave a dual direct effect on bone, both promoting boneformation and inhibiting bone resorption. To exclude thepossibility that the effects of MLN2238 on bone were asecondary consequence of its antimyeloma activity, wecompared the in vivo effect of MLN2238 with that ofmelphalan, an agent with accepted antimyeloma but nodirect effects on bone (17). Although with our specific invivo melphalan dosing and schedule this compound wasonly 50% as effective as bortezomib or MLN2238 in

controlling tumor burden, its antimyeloma activity wasnot linked to an indirect effect on bone, with no signif-icant changes observed with respect to vehicle control onbone turnover serum markers, trabecular architecture,osteoblast histologic appearance, or TCF4 immunostain-ing. These data reinforce our previous interpretation ofbone effects of MLN2238 being primarily due to directtargeting of osteoblast and osteoclast populations in thebone marrow, albeit indirect bone beneficial conse-quences of its antimyeloma activity cannot be excluded.

Taken together, we have shown that clinically relevantconcentrations of the next-generation proteasome inhibitorMLN2238, promote osteoblast differentiation and function(even in osteoprogenitor cells from multiple myelomapatients bearing osteolytic lesions) and inhibit osteoclastformation and resorption in ex vivo experiments. In amousemodel of disseminated human myeloma, oral administra-tion of MLN2238 was found to be at least as effective asintraperitoneally administered bortezomib in the control ofmyeloma growth and in prevention of bone loss, sustainedby both bone-forming and anticatabolic activities. Ourstudies, therefore, provide rationale for the clinical evalu-ation ofMLN9708 in the treatment ofMBD and other bonepathologies. Looking at the future, it is likely that the nextyears will show the real value of MLN9708 and othersecond-generation proteasome inhibitors, such as carfilzo-mib and oprozomib, as both anti–multiple myeloma andbone-directed agents.

Disclosure of Potential Conflicts of InterestJ.F. San Miguel is a consultant/advisory board member for BMS, Celgene,

Janssen, Millennium, MSD, Onyx, and Novartis. N. Raje is a consultant/advisory board member of Millennium. No potential conflicts of interestwere disclosed by the other authors.

Authors' ContributionsConception anddesign:A.Garcia-Gomez,M.Canavese, A.J. Berger, J.F. SanMiguel, M. GarayoaDevelopment of methodology: C. Ortiz-de-Solorzano, T. Hern�andez-Iglesias, A.C.M. Martens, S. Fraile, M. GarayoaAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): Z. Tian, J.F. Blanco, J. Mateo-UrdialesAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): A. Garcia-Gomez, M. Canavese, C. Ortiz-de-Solorzano, J. Mateo-Urdiales, M. Galarraga, N. Raje, M. GarayoaWriting, review, and/or revisionof themanuscript: A.Garcia-Gomez,M.Canavese, E.M. Ocio, A.J. Berger, C. Ortiz-de-Solorzano, J.F. San Miguel,N. Raje, M. GarayoaAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): A.C.M. Martens, R.W.J. Groen,J. Mateo-Urdiales, D. ChauhanStudy supervision: E.M. Ocio, J.F. San Miguel, M. Garayoa

AcknowledgmentsThe authors thank Dr. Randall Moon (Howard Huges Medical Institute,

Chevy Chase, MD) for making M50 Super 8� TOPflash and M51 Super 8�FOPflash plasmids available at Addgene nonprofit plasmid repository. Theauthors are also indebted to Montserrat Mart��n, Susana Hern�andez-Garc��a,Laura San Segundo, Isabel Isidro, Teresa Prieto, and Almudena Mart��n(Centro de Investigaci�on del Cancer and Hospital Universitario de Sala-manca) for their help and excellent technical work.

Grant SupportThis work was supported by grants from the Spanish MICINN—ISCIII

(PI081825; PI1202591), Centro en Red of Regenerative Medicine andCellular Therapy from Castilla and Le�on, RTICC Hematology Program

Garcia-Gomez et al.





(RD12/0036/0058), Spanish FIS (PS09/01897), MINECODPI2012-38090-C03-02, and CIMA (The "UTE project CIMA"). A. Garcia-Gomez wassupported by the "Spanish Society of Hematology and Hemotherapy."

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received June 27, 2013; revised December 8, 2013; accepted January 4,2014; published OnlineFirst January 31, 2014.

References1. Kyle RA, Rajkumar SV. Multiple myeloma. Blood 2008;111:2962–72.2. Raje N, Roodman GD. Advances in the biology and treatment of bone

disease in multiple myeloma. Clin Cancer Res 2011;17:1278–86.3. Vallet S, Raje N. Bone anabolic agents for the treatment of multiple

myeloma. Cancer Microenviron 2011;4:339–49.4. Basak GW, Srivastava AS, Malhotra R, Carrier E. Multiple myeloma

bone marrow niche. Curr Pharm Biotechnol 2009;10:345–6.5. Yaccoby S. Advances in the understanding of myeloma bone disease

and tumour growth. Br J Haematol 2010;149:311–21.6. Giuliani N, Colla S, Morandi F, Lazzaretti M, Sala R, Bonomini S, et al.

Myeloma cells block RUNX2/CBFA1 activity in human bone marrowosteoblast progenitors and inhibit osteoblast formation and differen-tiation. Blood 2005;106:2472–83.

7. Ocio EM, Mateos MV, San-Miguel JF. Novel agents derived from thecurrently approved treatments for MM: novel proteasome inhibitorsand novel IMIDs. Expert Opin Investig Drugs 2012;21:1075–87.

8. Dick LR, Fleming PE. Building on bortezomib: second-generationproteasome inhibitors as anti-cancer therapy. Drug Discov Today2010;15:243–9.

9. Hideshima T, Anderson KC. Preclinical studies of novel targetedtherapies. Hematol Oncol Clin North Am 2007;21:1071–91.

10. Terpos E,DimopoulosMA, SezerO, RoodmanD, AbildgaardN, VescioR, et al. The use of biochemical markers of bone remodeling inmultiplemyeloma: a report of the International Myeloma Working Group.Leukemia 2010;24:1700–12.

11. Delforge M, Terpos E, Richardson PG, Shpilberg O, Khuageva NK,Schlag R, et al. Fewer bone disease events, improvement inbone remodeling, and evidence of bone healing with bortezomibplus melphalan-prednisone vs. melphalan-prednisone in thephase III VISTA trial in multiple myeloma. Eur J Haematol 2011;86:372–84.

12. Zangari M, Yaccoby S, Pappas L, Cavallo F, Kumar NS, RanganathanS, et al. A prospective evaluation of the biochemical, metabolic,hormonal and structural bone changes associated with bortezomibresponse in multiple myeloma patients. Haematologica 2011;96:333–6.

13. vonMetzler I, KrebbelH,HechtM,ManzRA, FleissnerC,MiethM, et al.Bortezomib inhibits human osteoclastogenesis. Leukemia 2007;21:2025–34.

14. Oyajobi BO, Garrett IR, Gupta A, Flores A, Esparza J, Munoz S, et al.Stimulation of new bone formation by the proteasome inhibitor, bor-tezomib: implications for myeloma bone disease. Br J Haematol2007;139:434–8.

15. Giuliani N,Morandi F, Tagliaferri S, Lazzaretti M, Bonomini S, CrugnolaM, et al. The proteasome inhibitor bortezomib affects osteoblastdifferentiation in vitro and in vivo in multiple myeloma patients. Blood2007;110:334–8.

16. MukherjeeS, RajeN, Schoonmaker JA, Liu JC,Hideshima T,WeinMN,et al. Pharmacologic targeting of a stem/progenitor population in vivois associated with enhanced bone regeneration in mice. J Clin Invest2008;118:491–504.

17. Pennisi A, Li X, LingW,KhanS, ZangariM,YaccobyS. Theproteasomeinhibitor, bortezomib suppresses primary myeloma and stimulatesbone formation in myelomatous and nonmyelomatous bones in vivo.Am J Hematol 2009;84:6–14.

18. Kupperman E, Lee EC, Cao Y, Bannerman B, Fitzgerald M, Berger A,et al. Evaluation of the proteasome inhibitor MLN9708 in preclinicalmodels of human cancer. Cancer Res 2010;70:1970–80.

19. Chauhan D, Tian Z, Zhou B, Kuhn D, Orlowski R, Raje N, et al. In vitroand in vivo selective antitumor activity of a novel orally bioavailableproteasome inhibitor MLN9708 against multiple myeloma cells. ClinCancer Res 2011;17:5311–21.

20. Kumar S, Bensinger WI, Reeder CB, Zimmerman TM, Berenson JR,Berg D, et al. Weekly dosing of the investigational oral proteasomeinhibitor MLN9708 in patients with relapsed and/or refractory multiplemyeloma: results from a phase 1 dose-escalation study. Blood (ASHAnnual Meeting Abstracts) 2011;118:371–2. Abstract 816.

21. Richardson PG, Baz R, Wang L, Jakubowiak AJ, Berg D, Liu G, et al.Investigational agent MLN9708, an oral proteasome inhibitor, inpatients (Pts) with relapsed and/or refractory multiple myeloma (MM):results from theexpansion cohorts of a phase 1dose-escalation study.Blood (ASH Annual Meeting Abstracts) 2011;118:140. Abstract 301.

22. Kumar SK, Berdeja JG, Niesvizky R, Lonial S, Hamadani M, StewartAK, et al. A phase 1/2 study of weeklyMLN9708, an investigational oralproteasome inhibitor, in combination with lenalidomide and dexa-methasone in patients with previously untreated multiple myeloma(MM). Blood (ASH Annual Meeting Abstracts) 2012;120. Abstract 332.

23. Lee EC, Fitzgerald M, Bannerman B, Donelan J, Bano K, Terkelsen J,et al. Antitumor activity of the investigational proteasome inhibitorMLN9708 in mouse models of B-cell and plasma cell malignancies.Clin Cancer Res 2011;17:7313–23.

24. GroenRW,NoortWA,RaymakersRA, PrinsHJ, Aalders L, Hofhuis FM,et al. Reconstructing the human hematopoietic niche in immunode-ficient mice: opportunities for studying primary multiple myeloma.Blood 2012;120:e9–16.

25. Garayoa M, Garcia JL, Santamaria C, Garcia-Gomez A, Blanco JF,Pandiella A, et al. Mesenchymal stem cells from multiple myelomapatients display distinct genomic profile as compared with those fromnormal donors. Leukemia 2009;23:1515–27.

26. Garcia-Gomez A, Ocio EM, Crusoe E, Santamaria C, Hernandez-Campo P, Blanco JF, et al. Dasatinib as a bone-modifying agent:anabolic and anti-resorptive effects. PLoS ONE 2012;7:e34914.

27. Hurchla MA, Garcia-Gomez A, Hornick MC, Ocio EM, Li A, Blanco JF,et al. The epoxyketone-based proteasome inhibitors carfilzomib andorally bioavailable oprozomib have anti-resorptive and bone-anabolicactivity in addition to anti-myeloma effects. Leukemia 2013;27:430–40.

28. Veeman MT, Slusarski DC, Kaykas A, Louie SH, Moon RT. Zebrafishprickle, a modulator of noncanonical Wnt/Fz signaling, regulatesgastrulation movements. Curr Biol 2003;13:680–5.

29. DoubeM, KlosowskiMM, Arganda-Carreras I, Cordelieres FP, Dough-erty RP, Jackson JS, et al. BoneJ: free and extensible bone imageanalysis in ImageJ. Bone 2010;47:1076–9.

30. Hideshima T, Mitsiades C, Tonon G, Richardson PG, Anderson KC.Understanding multiple myeloma pathogenesis in the bone marrow toidentify new therapeutic targets. Nat Rev 2007;7:585–98.

31. Breitkreutz I, RaabMS,Vallet S,HideshimaT,RajeN,MitsiadesC, et al.Lenalidomide inhibits osteoclastogenesis, survival factors, and bone-remodelingmarkers inmultiplemyeloma. Leukemia 2008;22:1925–32.

32. Kuroda Y, Matsuo K. Molecular mechanisms of triggering, amplifying,and targeting RANK signaling in osteoclasts. World J Orthop 2012;3:167–74.

33. Ang E, Pavlos NJ, Rea SL, Qi M, Chai T, Walsh JP, et al. Proteasomeinhibitors impair RANKL-inducedNF-kappaBactivity in osteoclast-likecells via disruption of p62, TRAF6, CYLD, and IkappaBalpha signalingcascades. J Cell Physiol 2009;220:450–9.

34. Vaananen HK, Laitala-Leinonen T. Osteoclast lineage and function.Arch Biochem Biophys 2008;473:132–8.

35. Lin GL, Hankenson KD. Integration of BMP, Wnt, and notch signalingpathways in osteoblast differentiation. J Cell Biochem 2011;112:3491–501.

36. Monroe DG, McGee-Lawrence ME, Oursler MJ, Westendorf JJ.Update on Wnt signaling in bone cell biology and bone disease. Gene2012;492:1–18.






37. Qiang YW, Hu B, Chen Y, Zhong Y, Shi B, Barlogie B, et al. Bortezomibinduces osteoblast differentiation via Wnt-independent activation ofbeta-catenin/TCF signaling. Blood 2009;113:4319–30.

38. Sutherland C, Leighton IA, Cohen P. Inactivation of glycogen synthasekinase-3 beta by phosphorylation: new kinase connections in insulinand growth-factor signalling. Biochem J 1993;296:15–9.

39. Tohmonda T, Miyauchi Y, Ghosh R, Yoda M, Uchikawa S, Takito J,et al. The IRE1alpha-XBP1 pathway is essential for osteoblast differ-entiation through promoting transcription of Osterix. EMBO Rep2011;12:451–7.

40. Longo V, Brunetti O, D'Oronzo S, Dammacco F, Silvestris F. Thera-peutic approaches to myeloma bone disease: an evolving story.Cancer Treat Rev 2012;38:787–97.

41. SonmezM, Akagun T, Topbas M, Cobanoglu U, Sonmez B, Yilmaz M,et al. Effect of pathologic fractures on survival in multiple myelomapatients: a case control study. J Exp Clin Cancer Res 2008;27:11.

42. Wu P, Morgan GJ. Targeting bone as a therapy for myeloma. CancerMicroenviron 2011;4:299–311.

43. Drake MT, Clarke BL, Khosla S. Bisphosphonates: mechanism ofaction and role in clinical practice. Mayo Clin Proc 2008;83:1032–45.

44. Terpos E, Sezer O, Croucher PI, Garcia-Sanz R, Boccadoro M, SanMiguel J, et al. The use of bisphosphonates in multiple myeloma:

recommendations of an expert panel on behalf of the EuropeanMyeloma Network. Ann Oncol 2009;20:1303–17.

45. Kuhn DJ, Chen Q, Voorhees PM, Strader JS, Shenk KD, Sun CM, et al.Potent activity of carfilzomib, a novel, irreversible inhibitor of theubiquitin-proteasome pathway, against preclinical models of multiplemyeloma. Blood 2007;110:3281–90.

46. Chauhan D, Singh AV, Aujay M, Kirk CJ, Bandi M, Ciccarelli B, et al.A novel orally active proteasome inhibitor ONX 0912 triggers invitro and in vivo cytotoxicity in multiple myeloma. Blood 2011;116:4906–15.

47. Hu B, Chen Y, Usmani SZ, Ye S, Qiang W, Papanikolaou X, et al.Characterization of the molecular mechanism of the bone-anabolicactivity of carfilzomib in multiple myeloma. PLoS ONE 2013;8:e74191.

48. Obeng EA, Carlson LM, Gutman DM, Harrington WJ Jr, Lee KP, BoiseLH.Proteasome inhibitors induce a terminal unfoldedprotein responsein multiple myeloma cells. Blood 2006;107:4907–16.

49. Gupta N, Noe D, Liu G, Berg D, Kalebic T, Shou Y, et al. Clinicalpharmacokinetics of intravenous andoralMLN9708, an investigationalproteasome inhibitor: pooled analysis from monotherapy and combi-nations studies across various indications. Clin Pharmacol Ther(ASCPT Annual Meeting Abstracts) 2013;93:S32. Abstract PI-51.

Garcia-Gomez et al.





2014;20:1542-1554. Published OnlineFirst January 31, 2014.Clin Cancer Res Antonio Garcia-Gomez, Dalia Quwaider, Miriam Canavese, et al. Myeloma Bone DiseasePreclinical Activity of the Oral Proteasome Inhibitor MLN9708 in

Updated version

10.1158/1078-0432.CCR-13-1657doi:

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/20/6/1542.full#ref-list-1

This article cites 48 articles, 14 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/20/6/1542.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/20/6/1542To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-1657

http://clincancerres.aacrjournals.org/content/20/6/1542.full#ref-list-1

http://clincancerres.aacrjournals.org/content/20/6/1542.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/20/6/1542


preclinicalactivityoftheoralproteasomeinhibitormln9708...

Documents