Bisphosphonates regulate cell growth and gene expression in the UMR 106-01 clonal rat osteosarcoma cell line

Bisphosphonates regulate cell growth and gene expression in the UMR 106-01 clonal rat osteosarcoma cell line
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
  Despite significant improvements in patient survival and localdisease control, 25% of patients with osteosarcoma developmetastases and the surgery for many extremity lesions stillincludes amputation. Thus, modalities that inhibit tumour growthand the metastatic cascade will have a significant impact onpatient survival and potential for limb sparing surgery. Tumours induce bone destruction by cellular processes, such asosteoclast-mediated bone lysis. The importance of osteoclast-mediated lysis has been demonstrated in other malignancies,where the development of osseous metastases has been shown tobe mediated by soluble tumour-related osteoclast activatingfactors (Galasko, 1976). More recently, the induction of specificosteoclast recruiting factors in osteoblasts/stromal cells followingcontact with myeloma and breast cancer cells has been demon-strated (Chikatsu et al, 2000). Bisphosphonates have recognizedefficacy in reducing bone destruction, pain and pathological frac-ture in a variety of lytic primary and metastatic diseases of theskeleton (Thiébaud et al, 1991; Coleman and Purohit, 1993; Shipman et al, 1997; Bloomfield, 1998; Diel et al, 1998). They have more recently been shown to inhibit establishment andgrowth of prostate cancer metastases, which are generally consid-ered to be osteoblastic in their growth pattern (Adami, 1997). Weproposed that, as with the intraosseous growth of prostate cancer,the local spread of osteosarcoma involves proteolytic and osteo-clast-mediated bone destruction. The bisphosphonates may reducebone destruction by uncoupling close regulation of osteoclastactivity by osteoblast-like cells. The aims of our study were to examine the effects on the regula-tion of proliferation and apoptosis by bisphosphonate treatment of a clonal osteosarcoma cell line. Further, we assessed the expres-sion of an osteoclast differentiating factor, receptor activator of NF- κβ (RANKL) (Yasuda et al, 1998b) and an osteoclastogenesisinhibitory factor, osteoprotegerin (OPG) (Simonet et al, 1997;Yasuda et al, 1998a). Finally, we investigated the effects onosteoblast-related gene expression by an aminobisphosphonateand a non-aminobisphosphonate. The use of a bisphosphonatefrom each group allows comparison of their relative potencieswith the known effects on osteoclasts and bone resorption studies(Fleisch, 1993).  MATERIALS AND METHODS Cell culture UMR 106-01 cells derived from a 32 P-induced osteosarcoma inrats (Martin et al, 1976) were used. Cells were cultured in 75cm 2 tissue culture flasks (Greiner Cellstar) in α -modified MinimalEssential Medium ( α -MEM) containing hepes 4g l  1 , sodiumbicarbonate 1.95g 6  1 , gentamicin 80mg 6  1 , pH 7.4 and 10%fetal bovine serum (FBS), incubated at 37˚C and equilibrated in5% CO 2 in air. Subcultures were performed using 0.0125% trypsinin 0.5mM Na 2 EDTA in calcium and magnesium-free phosphatebuffer (1  versene) to harvest the cells. Bisphosphonates regulate cell growth and geneexpression in the UMR 106-01 clonal rat osteosarcomacell line PS Mackie 1 , JL Fisher 1 , H Zhou 2 and PFM Choong 1 1 Department of Orthopedics and 2 Department of Medicine, St. Vincent’s Hospital, Fitzroy, Melbourne, Victoria 3065, Australia Summary Local growth of osteosarcoma involves destruction of host bone by proteolytic mechanisms and/or host osteoclast activation.Osteoclast formation and activity are regulated by osteoblast-derived factors such as the osteoclast differentiating factor, receptor activator ofNF- κ  B ligand (RANKL) and the inhibitor osteoprotegerin (OPG). We have investigated the in vitro effects of bisphosphonates on a clonal ratosteosarcoma cell line. The aminobisphosphonate pamidronate was added to UMR 106-01 cell cultures (10  8 M to 10  4 M up to 5 days). Thenon-aminobisphosphonate clodronate was administered for the same time periods (10  6 M to 10  2 M). Cell proliferation, apoptosis and mRNAexpression was assessed. Both agents inhibited cell proliferation in a time- and dose-dependent manner. ELISA analysis demonstrated anincrease in DNA fragmentation although there was no significant dose-related difference between the doses studied. Bisphosphonate-treatedcultures had a greater subpopulation of cells exhibiting morphological changes of apoptosis. Expression of mRNA for osteopontin and RANKLwas down-regulated by both agents, while the expression of mRNA for alkaline phosphatase, pro- α 1(I) collagen and OPG was not altered.Out in vitro work suggests the bisphosphonates not only have direct effects on osteosarcoma cell growth and apoptosis, but also, by alteringthe relative expression of osteoclast-regulating factors, they may inhibit the activity of osteoclasts and their recruitment. ©2001 CancerResearch Campaign Keywords : osteosarcoma; bisphosphonates; apoptosis; osteopontin; RANKL; osteoprotegerin 951 Received 17 August 2000 Revised 3 January 2001 Accepted 4 January 2001 Correspondence to:  PFM Choong British Journal of Cancer  (2001) 84 (7), 951–958 ©2001 Cancer Research Campaigndoi: 10.1054/ bjoc.2001.1679, available online at on  Bisphosphonates The aminobisphosphonate pamidronate (Aredia, NovartisPharmaceuticals, Australia) and non-aminobisphosphonateclodronate (provided by Prof TJ Martin of St Vincent’s Institute of Medical Research) were dissolved in sterile water and storedfrozen in aliquots of 10  6 M to 10  2 M until use. Proliferation studies Cell count / proliferation studies were performed on UMR 106-01cells grown in monolayer in 6-well culture plates (10.06cm 2 NuncProducts), with a plating density of 1000–1500 cells per well in α -MEM medium containing 10% FBS. Trypan blue was used toconfirm that only cell populations with greater than 90% cellviability were used for seeding. Medium was replaced after 1 daywith α -MEM containing 2% FBS for a further 24 hours prior totreatment of bisphosphonates, at concentrations of pamidronate10  8 to 10  4 M and clodronate 10  6 to 10  2 M. Medium wasreplaced every 48 hours and bisphosphonates replenished with thenew media, to ensure continual exposure of cells to a relativelyuniform concentration of the bisphosphonates. Cells wereharvested at various time points with 0.0125% trypsin in 1  versene and were counted on a coulter counter (Model DN,Coulter Electronics, Bedfordshire, England). Experiments wererepeated at least 3 times with 6 wells per treatment group withineach experiment. Three separate coulter counts were performed oneach well sample and the results compared against controls usingStudent’s t  -test. Apoptosis assessment Apoptotic cell ratios in cell cultures were determined by ELISAquantification of DNA fragmentation (Boehringer Mannheim)according to the manufacturer’s specifications. Within microtitrewells, 1  10 4 cells were incubated for 4 hours with serial concen-trations of the two bisphosphonates. After centrifugation (toensure detached apoptotic cells were retained) the cells were lysedand an aliquot of the supernatant transferred to streptavidin-coatedELISA microtitre wells and exposed to anti-histone biotin andanti-DNA POD solutions for 2 hours. Unbound antibodies wereremoved by washing and the fragmented DNA-POD immunocom-plexes were detected photometrically, using ABTS (2,2 ′ -Azino-di[3-ethylbenzthiazolin-sulphonate]) substrate. A microtitre platereader (Titertek, Multiskan®Plus) was used to record the resultantcolor change (at 405nm against substrate solution as a blank,wavelength 490nm) and differences assessed using 1-wayANOVA and Fisher’s PLSD analyses. To confirm the ELISA findings, cultures were assessed for thecharacteristic morphological features of apoptosis. Cell cultureswere grown in monolayer on 8 well chamber slides (Lab-Tek, NuncProducts) and exposed after 24 hours to the two bisphosphonatesfor 24 and 48 hours. As with the ELISA studies, slides were brieflycentrifuged on microtitre plate carriers to return detached cells tothe monolayer and the slides were fixed in 3% (v/v) acetic acid in95% methanol for 1 minute before staining with haematoxylin andeosin. Three representative high-powered fields (400  ) wereanalysed for the percentage of cells exhibiting apoptotic character-istics of nuclear pyknosis, apoptotic bodies, membrane blebbingand cytoplasmic inclusions. Results were recorded as an apoptoticsubpopulation of less than 5%, 5 to 20%, or greater than 20%. Genotype studies UMR 106-01 cells were seeded at 1  10 4 cells per 56cm 2 plate(Falcon, Becton Dickinson) in α -MEM containing 10% FBS.Medium was replaced after 24 hours with α -MEM containing 2%FBS to reduce effects of endogenous cytokines on cell growth.Bisphosphonate treatment was not started until 24 hours after thischange and media and bisphosphonates ware replenished every48 hours. Bisphosphonate was added at concentrations of between1  10  8 and 1  10  4 M for pamidronate and 1  10  6 and 1  10  2 M for clodronate over time frames of between 6 days and4 hours, prior to confluence of control cultures. Previous experi-ments on the UMR 106-01 cells line allowed characterization of the culture growth pattern and accurate prediction of times tocontrol cell culture confluence. All times and concentrations wereassessed in triplicate. Total RNA was isolated from the UMR 106-01 cultures usingTRIZOL (Gibco PRL, Life Technologies) according to the manu-facturer’s instructions and redissolved in RNase-free TE Buffer,pH 8.0, to a concentration of 2  µ g µ l  1 and stored at – 80˚C. Astotal RNA from each culture was diluted to a standard 2  µ g µ l  1 ,prior to agarose gel electophoresis, the resultant Northernhybridization studies allow comparative analysis of mRNA as apercentage of total RNA in each treatment group. Synthesis of riboprobes Probes were labelled with digoxigenin (DIG) using an RNAlabeling kit (Boehringer Mannheim GmbH, Mannheim, Germany)according to the manufacturer’s instructions. A 2.4kb  Eco RI fragment of rat alkaline phosphatase (ALP)cDNA, Dr. G. Rodan, Merck, Sharp and Dohme ResearchLaboratories, West Point, PA, was further cut with  Bam H1 and the1.1kb  Eco R1-  Bam H1 fragment was subcloned into StratagenepBluescript plasmid (pBS). The plasmid was linearized with  Bam HI and transcribed with T7 RNA polymerase to generate a1.1-kb antisense riboprobe. A 500bp  Eco RI fragment of rat boneGLA protein/osteocalcin (BGP) cDNA (Dr J Wozney, GeneticsInstitute, Cambridge, MA) was subcloned into pBS plasmid vectorand then linearized with  Hind  III before being transcribed into theantisense strand with T3 RNA polymerase. Rat matrix GLAprotein (MGP) cDNA (Dr P Price, University of San Diego, LaJolla, CA) was subcloned into pBS. The plasmid was linearizedwith  Hind  III and transcribed with T3 to produce an antisense ribo-probe. A 1.6kb  Xba I-  Xho I insert of human osteopontin (OPN)cDNA in pBS (Dr L Fisher, National Institute of Dental Research,USA) was linearized with  Xba I and transcribed with T7 poly-merase to generate the antisense riboprobe. Pro- α 1(I) collagenRNA riboprobe was obtained by subcloning a 1.6kb Pst I frag-ment of rat pro- α 1(I) collagen cDNA (Dr J Bateman, RoyalChildren’s Hospital, Melbourne, Australia) into the pSPT 19. Thisplasmid vector was linearized using  Eco RI and was then tran-scribed with T7 RNA polymerase to generate a 1.6kb antisenseriboprobe. A 750bp RANKL cDNA fragment, (bp 381–1130)(Wong et al, 1997) and a 545bp OPG cDNA fragment (bp778–1323) (Simonet et al, 1997) were obtained from N Horwoodand R Thomas, St Vincent’s Institute of Medical Research,Melbourne, Australia. Each reverse transcription polymerasechain reaction product was cloned into pGEM-T, linearized with  Nde I and transcribed with T7 RNA polymerase to generate therespective antisense riboprobes (Kartsogiannis et al, 1999).  952 PS Mackie et al British Journal of Cancer (2001) 84  (7), 951–958 ©2001 Cancer Research Campaign   Northern blot analysis Total RNA was loaded at 20  µ g per lane and separated overnight ina 1.5% agarose-formaldehyde gel. The RNA was transferred byvacuum suction to nylon hybridization filters (Genescreen, NENLife Science Products) hybridized overnight in a buffer solution(formamide 50%, 5  SSC, block reagent 2% (BoehringerMannheim), SDS 0.02%) with DIG labeled riboprobes at aconcentration of 25ng ml  1 . Probes for OPN, ALP, Pro- α 1 (I)collagen, BGP, MGP, RANKL and OPG were used. Sequentialwashes (2  SSC / 0.1% SDS, 2 washes of 5 minutes at roomtemperature, 2 washes at 65˚C in 0.1  SSC / 0.1% SDS). Thehybridized probes were subsequently detected with anti-DIG anti-body coupled to alkaline phosphatase (Boehringer Mannheim) andvisualized by chemiluminescence autoradiographic detection withCDP-Star (Boehringer Mannheim). Filters were stripped of probe by washing in DEPC-treated H 2 Ofollowed by 15 minutes washing in 50% formamide, 50nM Tris-HCl (pH8.0) and 1% SDS at 100˚C, and stored in 2  SSC.Hybridized probes were normalized against a DIG-labelled 18 Soligonucleotide probe (amino acid (180–156) sequence: CGGCAT GTA TTA GCT CTA GAA TTA CCA CAG) (Chan et al,1984) and detected with Anti-DIG as described above. mRNA was quantitated by densitometry (ImageQuaNT soft-ware, Molecular Dynamics), normalizing each probe expressionagainst 18 S expression on the same filter, and repeated with atleast 3 filters for each probe. RESULTS Cell proliferation and cell death study There was a dose-related inhibition of cellular proliferation of UMR 106–01 cells in monolayer culture with both bisphosphon-ates used. At 1  10  4 M pamidronate exposure and > 1  10  2 Mclodronate exposure marked cytotoxicity was noted, with death of all cells within 24 to 48 hours (results not shown). Inhibition of growth occurred at concentrations of pamidronate between 1  10  6 M and 1  10  5 M and clodronate between 1  10  2 M and 1  10  2 M (Figures 1A, 1B). DNA internucleosomal fragmentation, detected by ELISAutilizing anti-DNA and anti-histone antibodies, increased morethan 3-fold in cell cultures exposed for 4 hours to 10  4 Mpamidronate (Figures 2A, 2B). Fragmentation at pamidronateconcentrations of 10  5 M to 10  9 M and with clodronate concen-trations of 10  3 M and 10  5 M was greater than in control cultures( P < 0.05). Using 1-way ANOVA analysis no statistical dose-dependent difference was noted between the concentrationsstudied within each treatment group. Examination of cells treated at 1  10  5 to 1  10  4 M of pamidronate by phase contrast microscopy demonstrated largenumbers of non-viable cells within the cultures. Large numbers of cells were seen exhibiting intracytoplasmic inclusions, loss of cellshape, loss of cell to cell and cell to surface adhesion, and the pres-ence of apoptotic bodies (Figures 3A, 3B). Visual quantification of the percentage of cells exhibiting these features demonstrated anincrease in the proportion of apoptotic cells in cultures exposed tohigher bisphosphonate concentrations (Table 1). Exposuretoclodronate at doses approximately 100-fold higher than those of pamidronate produced similar morphological features of apoptosiswithin the osteosarcoma cultures. These features of cytotoxicitywere not thought to be due to the calcium-chelating effect of bisphosphonates, as cultures exposed to equimolar concentrationsof EDTA did not produce the same effect (results not shown). Gene expression At 1  10  5 M pamidronate the expression of OPG was unalteredover treatment periods of 0 to 5 days when normalized against18 S expression on the same filter. RANKL mRNA expressionwas down regulated by prolonged exposure of the cells topamidronate, when normalized against 18 S and hybridizing equalamounts of total sample RNA compared with control RNA(Figure 4A, 4B). There was a significant reduction in OPN mRNA expression,again when normalized against 18 S on the same filters andcompared with equal control total RNA hybridization. We notedno change in the expression of ALP or pro- α 1 (I) collagenmRNA (Figures 4A, 4C). Similar effects on mRNA expressionfor the above proteins were observed in cells treated withclodronate, albeit at a concentration 100-fold that of pamidronate (results not shown). BGP and MGP are expressedin some osteoblast-like celllines however there is no inherentexpression of mRNA for thesematrix proteins by UMR106-01cells and exposure to bisphosphonate did not alter this. Bisphosphonate inhibition of osteosarcoma  953 British Journal of Cancer (2001) 84  (7), 951–958 ©2001 Cancer Research Campaign  108642ControlPamidronate 10  –8 Pamidronate 10  –7 Pamidronate 10  –6 Pamidronate 10  –5    C  e   l   l  c  o  u  n   t  p  e  r  w  e   l   l   (  x   1   0    5    ) Day 1Day 2Day 3Day 4108642ControlClodronate 10  –6 Clodronate 10  –4 Clodronate 10  –2    C  e   l   l  c  o  u  n   t  p  e  r  w  e   l   l   (  x   1   0    5    ) Day 1Day 2Day 3Day 4 BA Figure 1 The effect of bisphosphonate exposure on proliferation of UMR106–01 cells grown in monolayer. Cells were detached by trypsinization andcounted by coulter counter on the days indicated. Points are mean of threesamples. Experiments were performed more than three times with similarresults and low standard error of means (not discernible at this resolution). ( A ) – Pamidronate exposure. 10  8 M to 10  5 M. ( B ) – Clodronate exposure.10  6 M to 10  2 M. ( * ) = P  -value < 0.05 compared against control  Exposure of the cultures to concentrations ofpamidronate <10  5 M and clodronate <10  3 M did not significantly altermRNA expression, as detected by Northern hybridization(results not shown). DISCUSSION To study the effects of bisphosphonates on osteosarcoma cells invitro we have utilized the UMR 106-01 cell line. This cell line isvery well characterized and has been used, in our laboratory, toinduce reproducible and representative tibial osteosarcomas innude mice (unpublished). We have shown that pamidronate andclodronate have a dose- and time-dependent inhibitory effect onmonolayer growth of osteosarcoma cell. Studies on the effect of bisphosphonates on osteoblasts have differed in their findingsalthough several have demonstrated inhibitory effects at higherdoses (Evans and Braidman, 1994; Goziotis et al, 1995; García Moreno et al, 1998). Plotkin et al showed that bisphosphonatesprevent osteoblast apoptosis using several criteria, however noted 954 PS Mackie et al British Journal of Cancer (2001) 84  (7), 951–958 ©2001 Cancer Research Campaign    A  p  o  p   t  o  s   i  s  e  n  r   i  c   h  m  e  n   t   f  a  c   t  o  r Control10  –9 10  –7 10  –5 10  –4    A  p  o  p   t  o  s   i  s  e  n  r   i  c   h  m  e  n   t   f  a  c   t  o  r Control10  –7 10  –5 10  –3 BA Figure 2 Assessment of apoptosis after 4 hours of bisphosphonateexposure to 1  10 4 UMR 106–01 cells. The apoptotic subpopulation wasdetected by ELISA photometric assay (405nm against substrate solutionapprox 490nm) and the enrichment factor for apoptosis compared withuntreated cells graphically shown. Error bars represent SE mean over threeexperiments. 2( A ) – Pamidronate exposure 10  9 , 10  7 , 10  5 , 10  4 M. 2( B ) – Clodronate exposure 10  7 , 10  5 , 10  3 M. (*) = P  -value < 0.05 comparedagainst control. 25 microns25 micronsBA Figure 3 Representative microphotographs of UMR 106–01 cells grown inmonolayer, fixed with acetic acid 3% v.v. in methanol and stained withhematoxylin and cosin. ( A ) – Untreated cells with distinct cellular structureand frequent mitoses (arrow heads)  400 ( B ) – Following 24 hours ofpamidronate (10  5 M) apoptotic cells are more frequent, with many apoptoticbodies developing from nuclear fragmentation (arrows)  400 24 Hours<5%48 Hours<5% Control Pamidronate Clodronate 10 − 9 M10 − 7 M10 − 5 M10 − 7 M10 − 5 M10 − 3 M< 5%< 5%< 5%< 5%< 5%< 5%< 5%< 5%< 20%5     5%5     20%5     20% Table 1 Morphological assessment of apoptotic cells in monolayer on 8 well tissue culture microscopy slides. Three representative high-power fields wereexamined and the number of cells exhibiting characteristic features of apoptosis determined as a percentage of the total population. (<5% of cells apoptotic. 5–20 % of cells apoptotic. >20% of cells apoptotic  a biphasic response, and were unable to resolve an increase inmorphological signs of apoptosis seen in osteocytes exposed tobisphosphonates (Plotkin et al, 1999). An inhibitory effect wasnoted by Tan et al in 3 H-thymidine uptake studies in UMR 106osteosarcoma cells exposed to pamidronate (Tan et al, 1988),although the effect on cellular metabolic activity (rather than cellnumber) at 5  10  4 M was assessed. Time-or concentration-dependent differences were not reported. As observed in our study, pamidronate exposure of >10  4 Mwas cytotoxic to the cell monolayer whereas at 1  10  5 Mpamidronate we noted an inhibition of proliferation and anincrease in the proportion of cells in monolayer undergoingapoptosis. At this lower concentration we demonstrated nochange over time in the expression of ALP or pro- α 1 (I) collagenmRNA but there was progressive down regulation of OPN andRANKL mRNA expression. This selective alteration in mRNAexpression suggests that pamidronate, at a dose of 1  10  5 M,may regulate specific cell functions, such as proliferation andgene expression rather than inducing direct cytotoxicity. Thedose at which effects on the osteosarcoma cell line were noted ishigher than those used in many studies on osteoclasts, althoughsimilar concentrations have been demonstrated to cause apop-tosis in breast cancer cell lines (pamidronate >1  10  5 M)(Senaratne et al, 2000) and myeloma cell line cultures(pamidronate >1  10  4 M) (Shipman et al, 1997). These find-ings may still be significant if translated to the in vivo setting, asthe strong hydroxyapatite affinity of bisphosphonates is thoughtto drive their local concentration within the bone microenviron-ment significantly higher than the serum level. Novel studies bySato et al demonstrated localization of radiolabelled alendronatebeneath resorption lacuna at concentrations approaching 1  10  3 M (Sato et al, 1991) and further in vitro assays demon-strated an inhibitory IC 50 of 10  6 –10  5 M for pamidronate in anivory slice assay of osteoclast activity. We have studied the effects of bisphosphonates on the expres-sion of two factors by an osteosarcoma cell line which are knownto be physiological regulators of osteoclast activity. The osteoclastdifferentiating factor (RANKL) (Yasuda et al, 1998b) and M-CSF(Yoshida et al, 1990) are required for osteoclast differentiation.Regulation of these two factors is thought to be the common finalpathway by which the resorption-inducing agents such as parathy-roid hormone, 1,25 dihydroxyvitamin D 3 and interleukins promoteosteoclastogenesis (Horwood et al, 1998). The soluble TNF familymember (OPG) acts as a decoy receptor for RANKL, and expres-sion by osteoblasts under hormone regulation is thought to inhibit Bisphosphonate inhibition of osteosarcoma  955 British Journal of Cancer (2001) 84  (7), 951–958 ©2001 Cancer Research Campaign  Days123456ContOPNRANKLALPColOPG18 S Figure 4(A) Detection of mRNA expression for bone-related proteins andosteoclast-regulating factors. UMR 106–01 osteosarcoma cell monolayercultures were exposed to 10  5 M pamidronate for 1 to 6 days prior toconfluence of controls. Total RNA (20  µ g) was loaded in each lane, separatedby gel electrophoresis and mRNA visualized as described in the text. 18Sribosomal detection was performed to normalize the amount of RNA presentin each lane. Numbers above each lane represent number of days ofpamidronate exposure. Control = untreated cells. OPN = osteopontin:RANKL = receptor activator of NF- κ  B ligand: ALP = alkaline phosphatase:Col = pro- α 1(1) collagen: OPG = osteoprotegerin: 18S = Housekeeping gene (Days pamidronate treatment) pamidronate treatment)OPGRANKLALPColOPN Figure 4 (B and C) Densitometric analysis of Northern hybridization forbone-related factors and osteoclast regulating genes. Filters with bound andvisualized mRNA probes were analysed and normalized against thecorresponding 18S control bands for each filter. Y-axis represents the valueexpressed as a ratio of the normalized untreated control for the same timepoint and concentration. Average and range of three filters shown for eachprobe. ( B ) OPG = osteoprotegerin. RANKL = receptor activator of NF- κ  Bligand ( C ) ALP = alkaline phosphatase. Col = pro- α 1(1) collagen. OPN =osteopontin
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