Fashion & Beauty

Bisphosphonates Inhibit the Growth of Mesothelioma Cells In vitro and In vivo

Bisphosphonates Inhibit the Growth of Mesothelioma Cells In vitro and In vivo
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  BisphosphonatesInhibit theGrowthofMesotheliomaCells Invitro  and Invivo Savita Wakchoure, 1 MelindaA.Merrell, 1 WayneAldrich, 1 TelishaMillender-Swain, 1 Kevin W.Harris, 1,2 Pierre Triozzi, 1 andKatriS.Selander 1 Abstract  Purpose:  Bisphosphonates (such as risedronate and zoledronate) are widely used inhibitorsofboneresorption.Despitetheir invitro  antiproliferativeeffectsinvariouscancercells,bisphosph-onateshavenotexhibitedsignificantantitumorefficacyinanimalmodelsofvisceralcancer,whichmay be due to their poor bioavailability.The diagnostic use of radioactive bisphosphonates hasrevealed the accumulation of bisphosphonates in mesothelioma, which prompted us to test theantitumorefficacyofbisphosphonatesinthisdisease. Experimental Design and Results:  Treatment with either risedronate or zoledronate(2  10  4 to2  10  6 mol/L)inhibitedthegrowthofAB12andAC29mousemesotheliomacellsand induced the accumulation of unprenylated Rap1A in these cells. Both these  in vitro  effectswere reversed by geranygeraniol, an end product of the mevalonate pathway that thesebisphosphonates inhibit. Both bisphosphonates also induced the phosphorylation of the p38mitogen-activated protein kinase in AB12 and AC29 cells.The inhibition of p38 augmentedbisphosphonate-induced growth inhibition in these cells. Bisphosphonate-induced p38 phos-phorylation was not reversible by geranylgeraniol. Risedronate (15 mg/kg) and zoledronate(0.5 mg/kg) inhibited the growth of s.c. tumors andincreased the median survivalof mice withi.p.mesotheliomatumors invivo . Discussion:  In conclusion, risedronate and zoledronate inhibit the mevalonate pathway andinduce p38 activation in mesothelioma cells  in vitro .The effects on the mevalonate pathwaydominatebecausethenetresultisgrowthinhibition.Bothbisphosphonatesalsoinhibitmesothe-liomatumorgrowth invivo  andprolongthesurvivalofmesothelioma-bearingmice.Theseresultssupportfurtherstudyofbisphosphonatesinthemanagementofmesothelioma. M esothelioma, an asbestos-related neoplasm of the pleuraland peritoneal space, occurs in  f 10,000 patients worldwide(1). Due to the long latency period for tumor development andthe widespread use of asbestos for many years, the incidence isexpected to rise until the year 2020 (2). The biological behavior is distinct from other solid tumors in that mesothelioma tendsto grow in a sheet-like fashion, covering the surface of thepleura or peritoneum. It shows little tendency to invade,especially early in the course of the disease (3). Mesotheliomatypically recurs even after the most aggressive attempts at surgical resection and is poorly responsive to radiotherapy andchemotherapy. Multimodality approaches have had a relatively small effect on the majority of the patients and have beenassociated with toxicity. The survival of patients with mesothe-lioma ranges between 4 and 12 months (4, 5). Clearly, new treatment modalities are needed.Bisphosphonates are synthetic analogues of the naturally occurring pyrophosphate. Depending on their molecular structure, these drugs can be divided into pyrophosphate-resembling (p-bisphosphonates, such as clodronate) and nitro-gen-containing bisphosphonates (n-bisphosphonates, such asrisedronate and zoledronate; ref. 6). At the cellular level, thedifferent bisphosphonates have different mechanisms of action;n-bisphosphonates inhibit the mevalonate pathway, whereasthe effects of p-bisphosphonates are mediated via intracellular  ATP-like analogues. The main effect of all bisphosphonates istheir ability to inhibit osteoclast-mediated bone resorption. These drugs are therefore widely clinically used in the treatment of metabolic bone diseases that are due to increased boneresorption, such as osteoporosis (7). Bisphosphonates alsoinhibit the osteolytic complications of bone metastases of solidtumors and multiple myeloma (8). Data from animal modelssuggest that in addition to osteoclast inhibition at the site of bone metastasis, these drugs may also inhibit cancer cell pro-liferation in bone (9, 10). In particular, the newer n-bisphosph-onates have also been suggested to actually inhibit the cancer  Cancer Therapy: Preclinical  Authors’Affiliations:  1 Department of Medicine, Division of Hematology-Oncology, University of Alabama at Birmingham and  2 Birmingham Veteran’sAdministrationMedicalCenter,Birmingham,AlabamaReceived12/19/05;revised2/6/06;accepted2/23/06. Grantsupport: UniversityofAlabamaComprehensiveCancerCenterMesotheliomaCenterCoregrant(5P30CA13148).Thecostsofpublicationofthisarticleweredefrayedinpartby thepaymentofpagecharges.This article must therefore behereby marked  advertisement   inaccordancewith18U.S.C.Section1734solely toindicatethisfact. Note: P.TriozziandK.S.Selandercontributedequally tothiswork. Requests for reprints:  Katri Selander, Department of Medicine, Division ofHematology-Oncology, University of Alabama at Birmingham,WTIT558,1824 6thAvenue South, Birmingham, AL 35294-3300. Phone: 205-975-5973; Fax: 205-975-5650;E-mail:Katri.Selander @  F  2006AmericanAssociationforCancerResearch.doi:10.1158/1078-0432.CCR-05-2766 www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2862 Cancer Research. on February 12, 2016. © 2006 American Association Downloaded from   spread to bones in animal models (11). Although these drugssignificantly inhibit the growth of various cancer cells  in vitro ,they have not, however, proven effective as single agents in pre- venting tumor growth at visceral sites in various animal modelsofcancer(9–16).Thismaybeduetotheirpoorbioavailabilitytothe tumors; bisphosphonates are poorly absorbed from thegastrointestinaltractandwhengiveni.v.,theyarequicklyclearedfrom the circulation and adsorbed to bone matrix hydroxyap-atite, where they are retained for prolonged periods (17). It isindeed thought that this bone-seeking propensity of theotherwise very hydrophilic drugs makes them available for thecancer cells residing in the bone microenvironment. The uptake of the bone scan agent   99  Tc m diphosphonate, which is structurally similar to the actual bisphosphonatedrugs, in malignant pleural effusions and, rarely, in nonmalig-nant pleural effusions is well established. Although the exact mechanism of uptake in these conditions remains unclear,passive transudation has been implicated (18). There are alsoseveral reports of uptake of   99  Tc m diphosphonate by cancerslocalized at the soft tissue sites (19–21). Because mesotheliomahas been associated with such an accumulation of   99  Tc m diphosphonate, we hypothesized that bisphosphonates may also exhibit direct antitumor activity against this tumor type in vivo  (22, 23). We show here that the new nitrogen-containing bisphosphonates, risedronate and zoledronate, effectively inhibit the proliferation of mesothelioma cells  in vitro  andthe growth of mesothelioma tumors  in vivo . Furthermore, weshow that administration of these drugs after tumor formationcan significantly extend the survival of tumor-bearing mice inexperimental models of mesothelioma. Materials and Methods Bisphosphonates.  Risedronate (a gift from Leiras OY, Turku, Finland) was dissolved in PBS and the stock solution was set to pH 7.4 withNaOH. Zoledronate (Novartis, Geneva,Switzerland) was obtained fromthe pharmacy and diluted into the cell culture medium. For animalstudies, both bisphosphonates were diluted into sterile 0.9% saline. Cell culture.  The mouse mesothelioma cell lines AB12 and AC29 were provided by Dr. Steven Albelda (University of Pennsylvania,Philadelphia, PA). These cell lines were srcinally generated by Dr.Bruce Robinson at the Queen Elizabeth II Medical Center (Nedlands,Perth, Western Australia) by i.p. implantation of asbestos fibers inBALB/c and CBA/J mice, respectively, and have been well characterized(24). AB12 and AC29 cells were cultured and maintained in completemedium consisting of high-glucose DMEM (Mediatech, Washington,DC) supplemented with 10% heat-inactivated FCS, 100 units/mL penicillin, 100  A g/mL streptomycin, and 2 mmol/L glutamine (Sigma,St. Louis, MO). All cell cultures were done in incubators in a 37 j Catmosphere of 5% CO 2  /95% air. In vitro  growth assay.  Mesothelioma cells were plated in 96-wellplates in normal culture medium, and treated for the indicated periodsof time with various concentrations of zoledronate, risedronate, or PBS with or without the p38 inhibitor SB202190 or the inactive controlcompound SB207420 (both at the final concentration of 10  5 mol/L;Calbiochem, San Diego, CA), 25  A mol/L geranylgeraniol (cold, alltrans, American Radiolabeled Chemicals, St. Louis, MO), or the same volume of ethanol as a vehicle control. DNA synthesis was measured asan indication of cell proliferation, using nonisotopic bromodeoxyur-idine (BrdU) incorporation immunoassays (Exalpha Biologicals, Watertown, MA), according to the manufacturer’s instructions. Briefly,10 3 cells were plated onto 96-well plates in 100  A L of normal culturemedium. The cells were then treated with the indicated agents for  various times. BrdU was added to the wells for the final 24 hoursand incorporated BrdU was detected with sequential additions of monoclonal mouse anti-BrdU antibody and horseradish peroxidase–conjugated anti-mouse antibody. After addition of the substrate for horseradish peroxidase, the intensity of the colored reaction product, which is proportional to the amount of BrdU incorporated into thecells, was read with a spectrophotometer at 450 nmol/L. Western blotting.  AB12 and AC29 cells were plated on six-well platesin normal culture medium until near confluency. The cells were thenrinsed with sterile PBS and cultured for a further 24 hours in serum-freeculture medium, in the presence or absence of 2  10  4 to 10  5 mol/L risedronate, zoledronate, or PBS control, with or without 25  A mol/L of geranylgeraniol, or the same volume of ethanol as a vehicle control.Culture medium was discarded and the cells were harvested in lysisbuffer [20 mmol/L Tris (pH 7.4), 150 mmol/L NaCl, 1 mmol/L EDTA,1 mmol/L EGTA, 1% Triton, 2.5 mmol/L sodium PPi, 1 mmol/L  h -glycerolphosphate, 1 mmol/L Na 3  VO 4 , 1  A g/mL leupeptin; Cell Sig-naling, Beverly, MA] and clarified by centrifugation. After boiling thesupernatants in reducing SDS sample buffer, equal amounts of protein( f 50  A g) were loaded per lane and the samples were electrophoresedon 10% polyacrylamide SDS gel and transferred to a nitrocellulosemembrane. Unprenylated Rap1A was detected with the antibody SC-1482 and total Rap1 (both prenylated and unprenylated forms of bothRap1A and Rap1B) was detected with the antibody SC-65 (Santa CruzBiotechnology, Santa Cruz, CA) according to the manufacturer’srecommendations (25, 26). The phosphorylation status of p38 wasstudied with anti-phospho-p38 and anti-total p38 antibodies (CellSignaling), as recommended by the manufacturer (27). The proteinbands were visualized by chemiluminescence using SuperSignal West Pico enhanced chemiluminescence kit (Pierce, Rockford, IL). In vivo  mesothelioma models.  Female BALB/c mice, 4 to 8 weeks of age, were obtained from the National Cancer Institute-Frederick Cancer Research Facility (Frederick, MD) and were housed in the Pathogen-FreeRodent Shared Facility (Comprehensive Cancer Center, University of  Alabama at Birmingham). All animal procedures were done inaccordance with recommendations for the proper care and use of laboratory animals and were approved by the local IACUC. S.c. and i.p.mouse mesothelioma models were evaluated. In the s.c. model, 3  10 6  AB12 cells were first injected s.c. into cohorts of BALB/c mice. Treatments with i.p. bisphosphonates or vehicle were started whentumors became palpable on day 10 and continued every 6 days for atotal of four treatments. Tumor size was measured bidimensionally  with calipers every 2 to 3 days, and tumor volume was calculated by theformula (length    width 2 ) / 2. Mice were euthanized before tumorsreached the size of 2,000 mm 3 . In the i.p. model, AC29 or AB12 cells(5    10 5  / 0.5 mL) were injected i.p. into cohorts of 10 to 12 BALB/cmice using a 26-gauge needle. Treatment was initiated 6 days after tumor inoculation and the mice were followed for survival. In the i.p.model, risedronate (15 mg/kg), zoledronate (0.5 mg/kg), or PBS weregiven i.p. thrice a week for 2 weeks. Statistical analysis.  Kaplan-Meier survival curves were analyzed withthe Mantel-Cox Log-rank test. Fisher exact test was used to examinedifferences in the proportion of tumors responding and proportion of mice surviving. Student’s  t   test (two-tailed) was used to examinedifferences in growth assays and for the time to death/sacrifice. Resultsare expressed as mean F SD.  P   < 0.05 was considered to be statistically significant. Results Risedronate and zoledronate effects on unprenylated Rap1Aaccumulation and growth inhibition can be partially reversed by  geranylgeraniol in mesothelioma cells.  Nitrogen-containing bisphosphonates have previously been shown to inhibit thegrowth of various epithelial cancer cells  in vitro  via inhibit-ing the mevalonate pathway (28, 29). This inhibition resultsin the depletion of intracellular prenyl-groups, such as BisphosphonatesInhibitMesotheliomaGrowth Clin Cancer Res 2006;12(9) May1, 20062863 Cancer Research. on February 12, 2016. © 2006 American Association Downloaded from   geranylgeraniol, which are needed for the posttranslationalmodification and activation of small GTP-binding proteins,such as Ras, Rho, Rac, and Rap (7). For example, treatment withn-bisphosphonates has been shown to result in the accumula-tion of unprenylated Rap1A in CaCo-2 and leukemia cells(25, 30). To investigate whether risedronate and zoledronatesimilarly inhibit the mevalonate pathway in mesothelioma cells, AB12 and AC29 cells were treated for 24 hours with PBS or with2    10  4 to 2    10  6 mol/L risedronate or zoledronate, with25  A mol/L geranylgeraniol, or the same volume of ethanol as a vehicle control. The cells were then lysed and prepared for  Western blot analysis. Accumulation of unprenylated Rap1A wasused as a surrogate marker for the inhibition of the mevalonatepathway (26). Zoledronate and risedronate induced a dose-dependent accumulation of unprenylated Rap1A in both celllines. Risedronate-induced accumulation of unprenylated Rap1A  was almost completely reversed by 25  A mol/L geranylgeraniol inboth cells. Zoledronate-induced accumulation of unprenylatedRap1A was partially reversed in both cells. Stripping and re-blotting the membranes with the anti-total Rap1 antibodyclearly indicated that the findings were not due to unevenloading of thegels (Fig. 1). Higher concentrations of geranylgeraniol were alsotested and found effective, but because they compromised cell viability, they were not routinely used. Geranylgeraniol also re- versed the bisphosphonate-induced inhibition of DNA syn-thesis in both cells, but the extent of this reversal was dependent on the cell line and the bisphosphonates used (Fig. 2). Inhibition of p38 augments n-bisphosphonate induced growthinhibition.  We have previously shown that in addition to theinhibitory effects on the mevalonate pathway, in particular,the n-bisphosphonates also activate the p38 mitogen-activatedprotein kinase in breast cancer cells. This activation signals for resistance against bisphosphonate-induced growth inhibitionbecause blocking of the p38 mitogen-activated protein kinasepathway augments the growth-inhibitory effects of bisphosph-onates (27). To study whether a similar mechanism operates inmesothelioma cells, we first investigated the effects of risedronate or zoledronate on p38 phosphorylation in thesecells. Using phospho-p38-specific and total p38 antibodies in Western blotting, we show here that both risedronate andzoledronate induce a dose-dependent increase of p38 phos-phorylation in AB12 and AC29 cells. Unlike the accumulationof unprenylated Rap1A, this effect was, however, not reversibleby excess geranylgeraniol (25  A mol/L). Increasing the geranyl-geraniol dose did not affect the bisphosphonate-induced,increased phosphorylation status of p38 either (data not shown for higher geranylgeraniol concentrations; Fig. 3). AB12 and AC29 cells were then cultured with risedronate or zoledronate, with or without the specific p38 inhibitor SB202190 (10  5 mol/L), or with the same concentration of an inactive control compound SB202474. Inhibition of p38augmented both risedronate- and zoledronate-induced growthinhibition in both cell lines, even though there were cell-specific differences between the bisphosphonate concentrationsat which these effects were seen. In general, AC29 cells weremore sensitive to the effects of p38 inhibition (Fig. 4). Risedronate and zoledronate inhibit mesothelioma growth in vivo.  The antitumor activity of n-bisphosphonates wastested  in vivo  in a s.c. tumor model using AB12 cells, which aresyngeneic in BALB/C mice, because AB12 tumors are more Fig.1.  n-BisphosphonatesinducetheaccumulationofunprenylatedRap1Ainmesotheliomacells.AccumulationofunprenylatedRap1AwasdetectedinAB12(  A )andinAC29( B )cellsafter treatment for24hourswithPBSorwiththeindicatedconcentrationsofrisedronateorzoledronate,with25 A mol/Lgeranylgeraniol( GG ),orthesamevolumeofethanolasavehiclecontrol.ThecellswerethenlysedandpreparedforWesternblotanalysis.ThelevelsofunprenylatedRap1A( top )wereusedasasurrogatemarker todetecttheinhibitionofthemevalonatepathway,usingtheantibodySC-1482.Theblots,whichrepresentreplicateexperiments,werestrippedandtotalRap1wasdetectedwiththeantibodySC-65,toshow that theeffectswerenotduetoaloadingerror. Cancer Therapy: Preclinical  www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2864 Cancer Research. on February 12, 2016. © 2006 American Association Downloaded from   aggressive than the AC29 cells and have been resistant to most cancer chemotherapeutics  in vivo  (31). Groups of 10 BALB/cmice were inoculated s.c. with AB12 cells. Ten days later, whenthe tumors were palpable and between 100 and 175 mm 3 insize, the mice were treated with risedronate or zoledronate,using higher doses and more infrequent dosing schedules thanpreviously applied in mouse tumor models. Inoculations withPBS served as a vehicle control (9, 32). Tumor volume wasmeasured over time. Mice were sacrificed when tumors reached2,000 mm 3 . Both risedronate ( P   < 0.02) and zoledronate( P   < 0.003) inhibited s.c. tumor growth (Fig. 5A). However,neither of these bisphosphonates-treated tumors completely regressed, and after initial suppression, most tumors grew at arate comparable to control tumors (data not shown). Theeffects of risedronate and zoledronate on survival wereexamined  in vivo  in an i.p. tumor model. In particular, AB12cells form diffuse tumors throughout the peritoneal cavity following i.p. injection, a pattern similar to the presentation of human peritoneal mesothelioma (31). Six days after i.p.inoculation of AB12 or AC29 cells, groups of 10 to 12 mice were treated by i.p. injection of either risedronate, zoledronate,or PBS. Administration of zoledronate led to a significant increase in median survival (43 days for zoledronate versus 26days for PBS;  P   < 0.001). Median survival in the risedronategroup was 30 days, which was not statistically significantly different from the PBS group. All PBS-treated mice died by day 35. In contrast, there were three long-term (>60 days) and twolong-term (>85 days) survivors in the risedronate andzoledronate-treatment groups, respectively (Fig. 5B). A similar survival experiment was also done with mice bearing AC29cells. Although the differences were not statistically significant ( P   = 0.08) after a total of six inoculations with the drug, themedian survival of mice in the zoledronate-treatment group was 39 days, whereas in the control group, the median survival was 26.5 days (Fig. 5C). Discussion  The  in vivo  anticancer effects of bisphosphonates have beenpreviously detected in models of bone metastases. In thissetting, the growth-inhibitory effects of these drugs have beenattributed to their ability to inhibit osteoclasts and thereby, tothe inhibition of growth factor liberation from the bone matrix during bone resorption (9, 10, 16). We show here for the first time that both risedronate and zoledronate inhibit mesotheli-oma tumor growth  in vitro  and  in vivo . Treatment of mesothelioma-bearing mice with these agents also promotedtheir survival. Thus, bisphosphonates might also have bone-independent, direct anticancer effects against certain tumors.Our   in vivo  results with mesothelioma tumors establishedat soft tissue sites seem to be superior to those achieved withbisphosphonates in other tumor models. For example,bisphosphonates have not been shown to inhibit tumor growthin a mouse breast cancer model using s.c. injection of humanMDA-MB-231 breast cancer cells (9, 10). Furthermore, in a Fig. 2.  Geranylgeraniolpartiallyreversesthen-bisphosphonate-inducedgrowthinhibition.AB12(  A )andAC29( B )wereculturedinthepresenceofindicatedconcentrationsofrisedronateorzoledronate,with25 A mol/Lgeranylgeraniol( GG )or thesamevolumeofethanolasavehiclecontrol.DNA synthesisasanindicatorofcellproliferationratewasmeasuredwithBrdUincorporationafter5daysoftreatment. Columns,  meanpercentageofPBScontrol; bars, F SD;**. P  < 0.01;***, P  < 0.001versus thecorrespondinginactivecontrolcompound( n  =5). Fig. 3.  n-Bisphosphonatesinducep38phosphorylation.AB12andAC29cellswereculturedfor24hoursinthepresenceof theindicatedconcentrationsofrisedronate,zoledronate,orPBS,with25 A mol/Lgeranylgeraniol( GG )orethanolasavehiclecontrol.Phosphorylationofp38wasdetectedinAB12(  A )andinAC29( B )celllysatesinWesternblots,usingphospho-p38( top )andafterstrippingofthesameblot,totalp38( bottom )specificantibodies. BisphosphonatesInhibitMesotheliomaGrowth Clin Cancer Res 2006;12(9) May1, 20062865 Cancer Research. on February 12, 2016. © 2006 American Association Downloaded from   mouse model of breast cancer metastasis, zoledronate increasedthe overall survival of the tumor-bearing mice by, at most,4 days, whereas the median survival in our studies was increasedby 17 days with zoledronate (32). Although given lessfrequently, the bisphosphonate doses used in our study weremuch higher than those used in the earlier studies. Therefore,despite the less frequent dosing, which mimics the use of i.v.bisphosphonates in oncology, the cumulative doses are highand possibly not directly translatable into clinical use (8).Our findings are surprising in light of the poor bioavailability of bisphosphonates to visceral tumors. For example, serum Fig.4.  Inhibitionofp38augmentsn-bisphosphonate-inducedgrowthinhibition.AB12andAC29cellswereculturedwiththeindicatedconcentrationsofrisedronate(  A )orzoledronate( B ),withthespecificp38inhibitorSB202190(10  5 mol/L)orthesamevolumeofaninactivecontrolcompoundSB202474.DNA synthesisasanindicatorofcellproliferationratewasmeasuredwithBrdUincorporationafter5daysoftreatment. Columns,  meanpercentageofPBScontrol; bars, F SD;***, P  < 0.001versus thecorrespondinginactivecontrolcompound( n  =5). Fig.5.  Risedronateandzoledronatemediateantitumoractivity invivo.A,  AB12cellswereinoculateds.c.intotheflanksofmice.Tendayslater,groupsof10miceweretreateds.c.withPBS,zoledronate(0.5mg/kg),orrisedronate(15mg/kg)every6daysfor fourinjections. Points,  meantumor volume; bars, F SE. B,  AB12cellswereinoculatedintotheperitonealcavitiesofmice.Sixdayslater,groupsof12miceweretreatedbyi.p.injectionofzoledronate(0.5mg/kg),risedronate(15mg/kg),oranequalvolumeofPBSthriceaweekfor2weeks.Dataisexpressedasthepercentageofsurvival. C,  AC29cellswereinoculatedintotheperitonealcavitiesofmice( n  =10).Sixdayslater,themiceweretreatedbyi.p.injectionofzoledronate(0.5mg/kg)orwithanequalvolumeofPBSthriceaweekfor2weeks.Dataisexpressedasthepercentageofsurvival. Cancer Therapy: Preclinical  www.aacrjournals.orgClin Cancer Res 2006;12(9) May1, 2006 2866 Cancer Research. on February 12, 2016. © 2006 American Association Downloaded from 
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks