Science

Bone-derived soluble factors and laminin-511 cooperate to promote migration, invasion and survival of bone-metastatic breast tumor cells

Description
Tumor intrinsic and extrinsic factors are thought to contribute to bone metastasis but little is known about how they cooperate to promote breast cancer spread to bone. We used the bone-metastatic 4T1BM2 mammary carcinoma model to investigate the
Categories
Published
of 17
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
Share
Transcript
     2   0   1  4 http://informahealthcare.com/grf ISSN: 0897-7194 (print), 1029-2292 (electronic) Growth Factors, 2014; 32(2): 63–73 ! 2014 Informa UK Ltd. DOI: 10.3109/08977194.2014.894037 RESEARCH PAPER Bone-derived soluble factors and laminin-511 cooperate to promotemigration, invasion and survival of bone-metastatic breast tumor cells Delphine Denoyer 1 , Nicole Kusuma 1 , Allan Burrows 1 , Xiawei Ling 1 , Lara Jupp 1 , Robin L. Anderson 1,2,3 , andNormand Pouliot 1,3 1 Metastasis Research Laboratory, Peter MacCallum Cancer Centre, Melbourne, VIC, Australia,  2 Department of Pathology, Melbourne, VIC, Australia,and   3 Sir Peter MacCallum Department of Oncology, University of Melbourne, Melbourne, VIC, Australia Abstract Tumor intrinsic and extrinsic factors are thought to contribute to bone metastasis but littleis known about how they cooperate to promote breast cancer spread to bone. We usedthe bone-metastatic 4T1BM2 mammary carcinoma model to investigate the cooperativeinteractions between tumor LM-511 and bone-derived soluble factors  in vitro . We show thatbone conditioned medium cooperates with LM-511 to enhance 4T1BM2 cell migration andinvasion and is sufficient alone to promote survival in the absence of serum. These responseswere associated with increased secretion of MMP-9 and activation of ERK and AKT signalingpathways and were partially blocked by pharmacological inhibitors of MMP-9, AKT-1/2 or MEK.Importantly, pre-treatment of 4T1BM2 cells with an AKT-1/2 inhibitor significantly reducedexperimental metastasis to bone  in vivo . Promotion of survival and invasive responses bybone-derived soluble factors and tumor-derived LM-511 are likely to contribute to themetastatic spread of breast tumors to bone. Keywords AKT-1/2, breast cancer, bone metastasis,laminin-511, MMP-9, TGF- b 1 History Received 25 October 2013Revised 9 February 2014Accepted 10 February 2014Published online 5 March 2014 Introduction Metastatic breast cancer is the most common cause of cancer-related death in women worldwide (Desantis et al., 2011; Jemal et al., 2011). Involvement of bone occurs in approxi-mately 70% of advanced breast cancer patients, resulting indebilitating skeletal complications that invariably impact onthe quality of life (Coleman et al., 1998). Despite recentclinical advances, bone metastasis remains largely an incur-able disease for which patients receive primarily palliativetreatments that aim to extend life while minimizing cancer-related pain.Current evidence indicates that cross-talk between mul-tiple tumor-derived and bone-derived factors (chemokines,growth factors, and extracellular matrix proteins) dictatesorgan-specific tropism and promotes homing, survival, andgrowth of metastatic cells in bone (Casimiro et al., 2009; Daiet al., 2006; Dittmar et al., 2008; Joyce & Pollard, 2009; Kingsley et al., 2007; White & Muller, 2007; Nguyen et al., 2009; Theriault, 2012; Weilbaecher et al., 2011). These bi-directional interactions between tumor cells and the bonestroma, therefore, represent potential therapeutic targets forthe treatment of breast cancer bone metastasis (Clezardin,2011; Theriault, 2012). The basement membrane protein laminin (LM)-511 isexpressed at high levels in metastatic human and murinebreast tumors (Chia et al., 2007). It is a potent pro-migratory/ invasive substrate for metastatic breast tumor lines  in vitro ,a property essential for successful metastasis  in vivo (Chia et al., 2007). To further demonstrate the functionalrelevance of LM-511 to bone metastasis, we recentlyextended these observations to show that isolation of cellsubsets that migrate rapidly towards LM-511  in vitro  enrichesfor tumor variants that are highly metastatic to bone from themammary gland compared with the parental line (Kusumaet al., 2012). In particular, 4T1BM2 cells isolated from a bonemetastatic lesion derived from LM-511-selected variants gaverise to extensive bone metastasis in up to 85% of animalscompared with approximately 20% of mice inoculated withparental 4T1 cells (Kusuma et al., 2012). Together, theseobservations indicate that the ability to express and respondto LM-511 may contribute to metastatic dissemination of advanced breast tumors to bone. Evidence of a similarrelationship between high tumor expression of LM-511 andbone metastatic propensity in other cancer types (e.g. prostateand lung carcinomas) supports this contention (reviewedin Pouliot & Kusuma, 2013). However, it is unclear howfactors produced by bone cells and tumor-derived LM-511may cooperate to facilitate homing, survival, and colonizationof bone.To begin to address this, we used surrogate  in vitro  assaysof metastasis to mimic some of the functional interactionsamong tumor cells, LM-511, and soluble factors secreted by Correspondence: Dr. Normand Pouliot, Peter MacCallum Cancer Centre,Locked Bag #1, A’Beckett St., Melbourne, VIC 8006, Australia. Tel:+61 3 9656 1285. Fax: +61 3 9656 1411. E-mail: normand.pouliot@petermac.org    G  r  o  w   t   h   F  a  c   t  o  r  s   D  o  w  n   l  o  a   d  e   d   f  r  o  m   i  n   f  o  r  m  a   h  e  a   l   t   h  c  a  r  e .  c  o  m   b  y   P  e   t  e  r   M  a  c   C  a   l   l  u  m   C  a  n  c  e  r   C  e  n   t  r  e  o  n   0   7   /   2   4   /   1   4   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y .  bone cells. Our study provides supporting evidence thatsignals triggered by bone-derived soluble factors and attach-ment to LM-511 cooperate to promote breast cancercell migration, invasion, and survival through activation of multiple signaling pathways. These responses are likely tocontribute to bone metastasis. Methods Cell lines and cell culture 4T1BM2 bone-metastatic cells were derived from the 4T1mouse mammary tumor line (obtained from Dr. F. Miller,Karmanos Cancer Institute, Detroit, MI) in our laboratory(Kusuma et al., 2012). For routine culture, 4T1BM2 cellswere maintained in the  a -minimal essential medium( a -MEM) supplemented with 5% fetal calf serum (FCS) and1% penicillin–streptomycin (Invitrogen Australia Pty Ltd,Mount Waverley, Victoria, Australia) and limited to 4 weeksin culture. Recombinant human LM-511 used for functionalassays was produced from human embryonic kidney 293 cellstransfected with full-length LM- a 5,  b 1, and  g 1 cDNA(provided by BioStratum Inc., Durham, NC) and purifiedfrom serum-free culture supernatant as described previously(Doi et al., 2002). Preparation of serum-free bone-conditionedmedium (BCM) Matrix metalloproteinase (MMP)-9 null mice were kindlyprovided by Dr. S. D. Shapiro (University of California inSan Francisco, San Francisco, CA) and backcrossed onto aBalb/c background in our laboratory. Femurs and tibiasfrom either wild-type or MMP-9 null Balb/c mice wereharvested and excess connective tissue carefully removed.The bones were minced using a scalpel blade and digestedwith a solution of collagenase II (3mg/mL) in PBS for 1hat 37  C with agitation. Debris and undigested bone wereremoved by filtration through a 70- m m nylon filter andsingle cell suspensions diluted in PBS and pelleted twiceby centrifugation. The mixture of cells was used unfractio-nated to maintain the natural cellular heterogeneity of thebone environment as closely as possible. The cells wereseeded in  a -MEM supplemented with 20% FCS, glutamine(2mmol/L), sodium pyruvate (1mmol/L), 3% penicillin–streptomycin, and fluconazole (6 m g/mL). The medium waschanged after 3d and every 2d thereafter until sub-confluent.Sub-confluent cultures were serum-starved overnight, themedium replaced with fresh serum-free medium and the cellsincubated for a further 48h at 37  C. The BCM was filteredthrough a 0.2- m m filter and stored at   80  C until required.BCM remained active during   80  C storage for at least2months. Prior to their use in functional assays, eachbatch of BCM was titrated and their effect was reproducibleand optimal when used at 50% dilution or higherconcentration. Detection of cytokines in BCM A glass slide-based sandwich ELISA fluorescent detectionsystem was used for the identification of cytokines presentin the BCM and in the serum-free conditioned medium from4T1BM2 cells. BCM and 4T1BM2 conditioned medium wereprepared from sub-confluent cultures as described abovefor BCM. The RayBio  Mouse Cytokine Antibody Array GSeries 6 (RayBiotech, San Diego, CA) allowing semi-quantitative detection of up to 97 cytokines was used as permanufacturer’s instructions. The mouse TGF- b 1 quantikinecolorimetric sandwich ELISA kit was used for specificdetection of TGF- b 1 as per manufacturer’s instructions(R&D Systems Inc., Minneapolis, MN). Cell proliferation and survival assays  In vitro  cell proliferation and survival assays were completedusing the sulforhodamine B (SRB) colorimetric assay (Sloanet al., 2006). Proliferation of 4T1BM2 cells was measuredover 5d in the complete  a -MEM medium with an initial celldensity of 500/well and five replicate wells per time point.To determine the effect of LM-511, BCM and the combin-ation of both on the proliferation of 4T1BM2 cells, the cellswere seeded in serum-free medium alone or with 50% BCMin uncoated or LM-511-coated wells (2 m g/mL). For thesurvival assay, cells (500/well) were seeded in serum-freemedium or 50% BCM with or without an AKT inhibitor VIII(10 m M, Calbiochem, EMD Millipore Corporation, Billerica,MA) or the MEK inhibitor U0126 (10 m M, Calbiochem) andkept on ice for 30min before seeding into 96-well cultureplates coated with LM-511. After 24h incubation at 37  Cunder serum-free conditions, the medium, containing inhibi-tors and dead cells, was removed and replaced with freshmedium supplemented with 10% serum. Remaining livecells were allowed to proliferate for an additional 96h at37  C. For comparison, control cells were seeded directlyinto wells in the presence of 10% FCS and allowed toproliferate for 96h. After each time point, the culture plateswere processed for standard SRB staining and the absorbancemeasured by a spectrophotometry at 550nm. Proliferation andsurvival assays were completed three times and the dataare presented as the mean±SD of a representative experimentwith five replicate wells. Statistical significance was deter-mined using a one-way ANOVA and Tukey’s multiplecomparisons test (survival assay) or a two-way ANOVA(proliferation assay);  p 5 0.05 was considered statisticallysignificant. Migration and invasion assays Migration and invasion assays were run in triplicate Transwellmigration chambers (8- m m pore size, Corning Inc., Corning,NY) as described previously (Chia et al., 2007; Sloan et al.,2006). Where indicated, LM-511 coated on the undersideof the porous membranes and/or soluble BCM added to thelower well were used as chemoattractants. For inhibition of migration or invasion, the cells were pre-incubated for 30minon ice in serum-free medium containing a MMP-9 inhibitor-1(50nmol/L, Calbiochem) or the AKT inhibitor, AKT inhibitorVIII (10 m mol/L, Calbiochem) before seeding into the wells.After 4h (migration) or 18–24h (invasion) incubation at37  C, migrated/invaded cells on the underside werefixed with 10% buffered formalin, permeabilized in 0.1%Triton-X 100, and stained with 0.5 m g/mL 4 0 -6-diamodino-2-phenylindole (DAPI, Sigma, St. Louis, MO). The membranes 64  D. Denoyer et al.  Growth Factors, 2014; 32(2): 63–73    G  r  o  w   t   h   F  a  c   t  o  r  s   D  o  w  n   l  o  a   d  e   d   f  r  o  m   i  n   f  o  r  m  a   h  e  a   l   t   h  c  a  r  e .  c  o  m   b  y   P  e   t  e  r   M  a  c   C  a   l   l  u  m   C  a  n  c  e  r   C  e  n   t  r  e  o  n   0   7   /   2   4   /   1   4   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y .  were mounted on glass slides and three random fields permembrane photographed on a Olympus BX-51 fluorescencemicroscope for quantitative scoring of migrated/invaded cellsusing Metamorph software (Molecular Devices, SiliconValley, CA). Assays were completed three times. The resultsfrom a representative experiment ( n ¼ 3) are shown andexpressed as the mean number of migrated/invaded cells/ field±SD of nine fields of view per condition. Statisticalsignificance was determined using a one-way ANOVA andTukey’s multiple comparisons test;  p 5 0.05 was consideredsignificant. Immunoblotting The cells were serum-starved overnight and starved for anadditional 4h again in fresh serum-free medium beforeseeding in 6-well plates coated or not with LM-511 (2 m g/mL)at a density of 10 6 cells in 2mL of serum-free medium orBCM. Plates were spun for 2min at 1200rpm at 4  C andincubated at 37  C for 5 or 20min. An aliquot of 10 6 cells waskept on ice to determine the basal activation of AKT afterdetachment of the cells ( T  0 ). After the incubation, the cellswere washed twice with ice-cold PBS and lysed in RIPAbuffer containing 10mmol/L PO 4  (pH 7.4), 500mmol/LEDTA, 1% (v/v) Nonidet P-40, 0.5% (w/v) sodium deox-ycholate, 1mmol/L Na 4 P 2 O 7 , 1mmol/L Na 3 VO 4 , 50mmol/LNaF, 0.1mmol/L phenylmethylsulfonyl fluoride, 1 m g/mLleupeptin, 1 m g/mL aprotinin, 0.1% (w/v) SDS, and150mmol/L NaCl. Lysates were clarified by centrifugation(16,000 g , 15min at 4  C), and protein concentrationsmeasured by D C  protein assay (Bio-Rad, Hercules, CA).Cell lysates (30 m g) were separated by 12% SDS-PAGE andtransferred to a PVDF membrane. The membrane wasblocked with 5% (w/v) non-fat dried milk in PBS, 0.1%Tween 20, and incubated with rabbit polyclonal anti-AKT(1:1000, Cell Signaling, Rockford, IL) and rabbit polyclonalanti-pAKT Ser473 (1:1000, Cell Signaling) or a rabbit poly-clonal anti-ERK-1/2 (1/1000, Cell Signaling) and rabbitpolyclonal anti-pERK-1/2 T202/Y204 (1/1000, Cell Signaling)antibodies overnight at 4  C, followed by incubation withappropriate horseradish peroxidase-conjugated secondaryantibodies for 1h at room temperature. Blots were developedusing the enhanced chemoluminescence kit (ECL, AmershamBiosciences, Piscataway, NJ), according to the manufacturer’sinstructions. Flow cytometry Standard flow cytometry on a Canto flow cytometer (BectonDickinson, San Jose, CA) (Kusuma et al., 2012) was used todetect the expression of Axl, CCR1, and CXCR6. Briefly,4T1BM2 cells were detached in PBS containing 0.01%EDTA, blocked in  a -MEM supplemented with 2% FCS and2% BSA for 30min on ice and reacted with primaryantibodies for 1h on ice. Excess primary antibodies wereremoved in wash buffer (PBS, 2% FCS) and incubatedfor a further 45min in the presence of appropriate FITC-conjugated secondary antibodies. Primary antibodies wererat anti-mouse Axl (clone MAB8541), rat anti-mouseCCR1 (clone 643854), and rat anti-mouse CXCR6 (cloneMAB2145) from R&D Systems (Minneapolis, MN). Gelatin zymography Gelatinase assays were run in 12-well plates precoated or notwith LM-511 (2 m g/mL) or collagen-IV (20 m g/mL). 4T1BM2cells (10 6 in 1mL final) were seeded in the absence orpresence of BCM (66% v/v) derived from MMP-9 null mice.FCS (0.3% v/v) was added to all wells to maintain viabilityand the supernatants were collected after a 24h incubation at37  C. Total cell lysates were prepared from pooled adherentand non-adherent cells in each well by lysis with extractionbuffer (30mmol/L HEPES, 150mmol/L NaCl, 1% TritonX-100) supplemented with complete EDTA-free proteaseinhibitor cocktail (Roche, New York, NY). Volumes of culturesupernatant corresponding to 15 m g of cell lysate were loadedand separated on an 8% SDS-PAGE containing 2.5% bovinegelatin (Sigma, St. Louis, MO). Proteins were renatured byrocking in three changes of 2.5% Triton X-100 solution over2h followed by incubation in activation buffer (50mmol/LTris-HCl pH 7.5, 150mmol/L NaCl, 10mmol/L CaCl 2 , 1%Triton X-100, and 0.02% sodium azide) for 18h. MMP-2 andMMP-9 bands were revealed by staining of the gels with asolution of Coomassie blue for 1h and destaining (30%methanol and 10% acetic acid). Clear bands were scanned andquantitated using Image-J software (NIH, Bethesda, MD).Statistical significance was determined using Student’s  t   test;  p 5 0.05 was considered significant. Experimental bone metastasis andcolony-forming assays Female 6–8 week old Balb/c mice (Walter & Eliza HallInstitute, Melbourne, Australia) were maintained in a specificpathogen-free environment and fed ad libitum. All proceduresinvolving mice conformed to National Health and MedicalResearch Council animal ethics guidelines. 4T1BM2 cells(10 6  /mL in PBS) were pretreated with AKT inhibitor VIII(AKTi-1/2, Calbiochem/Merck Biosciences, Darmstadt,Germany, 10 m mol/L) or control DMSO (0.1%) for 30minon ice. The mice were anesthetized by isoflurane inhalationand the cells (50,000) inoculated into the left ventricle of theheart. The number of mice that developed experimentalmacrometastases in bone (ribs, femurs and spine) after 14dwas evaluated in each group semi-quantitatively by visualinspection and microscopic examination of H&E stainedtissue sections by two independent observers. For femurs andspines, at least three sections/bone (100 m m apart) wereexamined. The statistical significance was analyzed usingFisher’s exact test.Alternatively, real-time quantitative PCR using TaqmanChemistry (PE Biosystems, Foster City, CA) was used foraccurate quantitation of relative metastatic burden in bone(Eckhardt et al., 2005; Sloan et al., 2006). Briefly, genomic DNA was extracted from spines and femurs and a multiplexreaction was performed on a HT7900 cycler to determinethe ratio of signals between a marker gene (mCherry)present in tumor cells only and vimentin gene present inall cells. Primers and probes were designed using primerexpress (Applied Biosystems, Foster City, CA) and were asfollows: Vimentin: forward 5 0 -AGCTGCTAACTACCAGGACACTATTG-3 0 , reverse 5 0 -CGAAGGTGACGAGCCATCT-3 0 , probe VIC-CCTTCATGTTTTGGATCTCATCCTGCAG DOI: 10.3109/08977194.2014.894037  Laminin-511 and bone factors promote breast cancer bone metastasis  65    G  r  o  w   t   h   F  a  c   t  o  r  s   D  o  w  n   l  o  a   d  e   d   f  r  o  m   i  n   f  o  r  m  a   h  e  a   l   t   h  c  a  r  e .  c  o  m   b  y   P  e   t  e  r   M  a  c   C  a   l   l  u  m   C  a  n  c  e  r   C  e  n   t  r  e  o  n   0   7   /   2   4   /   1   4   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y .  GT-TAMRA; mCherry: forward 5 0 -GACCACCTACAAGGCCAAGAAG-3 0 , reverse 5 0 -TCAACATCAAGTTGGACATCACCT-3 0 , probe 6FAM-CAGCTGCCCGGCGC-TAMRA.Multiplex PCR reactions were performed as describedpreviously (Eckhardt et al., 2005; Sloan et al., 2006), and the  D C  T  is used to calculate the relative tumor burden (RTB)as RTB ¼ 10 ; 000 = 2 D C  T .For colony-forming assays, control (DMSO) and AKTinhibitor VIII-treated cells were seeded in 6cm dishes(five replicates per group) at a final density of 100cells/dishin 4mL of   a -MEM medium supplemented with 10% FCS.Colonies ( 4 50cells) formed after 10d at 37  C were fixedand stained with a solution of crystal violet and counted. Thestatistical differences between groups were analyzed usingStudent’s  t   test;  p 5 0.05 was considered significant. Results We reported previously that LM-511 is a potent pro-migratory/invasive factor for bone metastatic breast tumorcells (Chia et al., 2007). Since soluble factors produced in thebone microenvironment have also been implicated in bonetropism (Casimiro et al., 2009; Joyce & Pollard, 2009), we tested whether serum-free BCM derived from primarycultures of collagenase-digested whole crushed bones couldcooperate with LM-511 to promote migration and Matrigelinvasion of 4T1BM2 cells (Figure 1). As expected, nosignificant migration was observed in the absence of LM-511or BCM but migration was induced by LM-511 coated on theunderside of the porous membrane (Figure 1A). Interestingly,migration was negligible in response to BCM alone butincreased more than 3-fold above that seen in the presence of coated LM-511 alone when used in combination with LM-511(Figure 1A). Similarly, 4T1BM2 cells did not invade Matrigelsignificantly in the absence of BCM or coated LM-511. WhileBCM or LM-511 used singly promoted invasion, combinationof both LM-511 and BCM enhanced invasion above theadditive effect of LM-511 or BCM alone (Figure 1B). Theseresults indicate that LM-511 and serum-free bone-conditionedmedium have a cooperative effect on the migration andinvasion of 4T1BM2 cells.Bone metastatic cells often express high levels of MMP-9that contributes to their migration, invasion and metastasis(Testeretal.,2000).Wehaveshownpreviouslythatattachmentofanotherbonemetastaticclonallineofthe4T1model(4T1.2)to purified LM-511 induces the expression of MMP-9(Kusuma et al., 2011), whereas others have reported that apeptide derived from the LM-511  a 5 chain induces MMP-9expression in macrophages (Adair-Kirk et al., 2005). It was,therefore, of interest to determine if enhanced MMP-9expression in response to LM-511 could contribute to thesynergy between LM-511 and BCM in 4T1BM2 cells.For these experiments, the cells were seeded on plastic orcollagen-IV or LM-511 in the absence or presence of BCMderived from MMP-9 null mice to allow specific measurementof tumor-derived MMP-9. The assays were carried out in thepresence of 0.3% serum to maintain long-term viability.Secreted MMP-9 was measured after 24h by gelatinzymography (Figure 2A) and quantitated by densitometry(Figure 2B). MMP-9 was not detectable in medium containing0.3% serumor inBCM derived fromMMP-9nullmice butwasreadily detected in the supernatant of 4T1BM2 cells seeded onplastic (Control). Importantly, either LM-511 or BCM stimu-lation alone were sufficient to enhance MMP-9 expression in4T1BM2 cells by approximately 2-fold (  p 5 0.03) comparedwith cells seeded on plastic. Combining LM-511 and BCMdid not increase MMP-9 expression above the level seen in thepresence of LM-511 or BCM alone. In contrast to thestimulatory effect of LM-511, MMP-9 expression decreasedwhen 4T1BM2 cells were plated on collagen-IValone or in thepresence of BCM indicating that MMP-9 induction byattachment to extracellular matrix (ECM) proteins is morespecific for LM-511.Analysis of signaling pathways triggered by attachment toLM-511 and/or stimulation with BCM revealed an associationbetween the migratory and invasive responses and activationof the extracellular-regulated kinase (ERK)-1/2 and PI3-kinase/AKT pathways (Figure 3). While phospho-ERK-1 wasnegligible at 5min in 4T1BM2 cells seeded on plastic, it wasstrongly induced by seeding on LM-511 alone or in thepresence of BCM. Combination of LM-511 and BCM had anadditive effect on the level of ERK-1 phosphorylationthat was sustained for at least 20min. In contrast tophospho-ERK-1, phospho-ERK-2 was readily detectableafter 5min when 4T1BM2 cells were seeded on plastic andits level decreased after 20min. Unlike cells seeded onplastic, phosphorylation of ERK-2 in the presence of LM-511and BCM was sustained for at least 20min. Figure 1. LM-511 and bone-derived solublefactors cooperate to enhance 4T1BM2migration and invasion. (A) Migration and(B) invasion of 4T1BM2 cells were measuredin Transwell chambers in the absence of serum. BCM in the bottom well or LM511(5 m g/mL) coated on the underside of theporous membrane or both were used aschemoattractants as indicated. The number of migrated and invaded cells was measuredafter 4h (A) and 18h (B). The results areexpressed as the number of invaded cells/ field of view and represent the mean±SD of a representative experiment in triplicate wells( n ¼ 3). Statistical significance wasdetermined using a one-way ANOVA andTukey’s multiple comparisons test.***  p 5 0.001. 66  D. Denoyer et al.  Growth Factors, 2014; 32(2): 63–73    G  r  o  w   t   h   F  a  c   t  o  r  s   D  o  w  n   l  o  a   d  e   d   f  r  o  m   i  n   f  o  r  m  a   h  e  a   l   t   h  c  a  r  e .  c  o  m   b  y   P  e   t  e  r   M  a  c   C  a   l   l  u  m   C  a  n  c  e  r   C  e  n   t  r  e  o  n   0   7   /   2   4   /   1   4   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y .  Prior to stimulation ( T  0 ), the activating phosphorylationof AKT on serine 473 was not detectable. Low level of AKTphosphorylation was observed after 20min when 4T1BM2cells were seeded on plastic and this was not increased furtherby seeding on LM-511 alone. However, a robust activationof AKT was observed in 4T1BM2 cells in response toBCM used alone or in combination with LM-511. AKTphosphorylation was evident after 5min of stimulation withBCM and remained elevated for at least 20min. Quantitationof P-AKT/T-AKT ratio at 5 and 20min by densitometryconfirmed that AKT activation was induced primarilyby BCM rather than by LM-511 (Figure 3B). Quantitationof P-ERK-1/2 and P-AKT after 20min stimulation in anotherindependent experiment performed under the same conditionsshowed that the effect of LM-511 and BCM on ERK andAKT activation was reproducible between experiments(Supplemental Figure 1).To demonstrate a causal relationship among MMP-9levels, ERK-1/2, and AKT phosphorylation and inductionof invasion by LM-511 and BCM, we made use of specificinhibitors of these pathways. First, we confirmed that 20 m Mof the MEK inhibitor U0126 potently and specifically Figure 3. Time-course of ERK-1/2 and AKT activation in 4T1BM2cells. The cells were seeded in culture plates coated or not with LM-511(2 m g/mL) with or without 50% v/v BCM as indicated. (A) Cell lysateswere analyzed for the presence of phosphothreonine 202  /tyrosine 204 -ERK (P-ERK) and phosphoserine 473 -AKT (P-AKT) by immunoblotting at thetimes indicated. (B) Total amount of ERK (T-ERK) and AKT (T-AKT)in each sample was used for normalization and quantitation of P-ERK and P-AKT by densitometry. A representative immunoblot is shown( n ¼ 3) and the data are expressed as fold-activation relative to plastic T  5min  (given a value of 1).Figure 2. Zymographic analysis of MMP-9 levels in 4T1BM2 cells.4T1BM2 cells (10 6 ) were seeded in wells precoated or not with LM-511(2 m g/mL) or collagen-IV (20 m g/mL) in the absence or presence of BCM(66% v/v) derived from MMP-9 null mice. FCS (0.3% v/v) was addedto all wells to maintain viability and the supernatants were collectedafter a 24h incubation at 37  C. (A) MMP-9 expression in the culturesupernatants was detected by gelatin zymography as described inMethods section. (B) Clear bands were scanned and quantitated usingImage-J software (NIH). Statistical significance was determined usingStudent’s  t   test; *  p 5 0.05. DOI: 10.3109/08977194.2014.894037  Laminin-511 and bone factors promote breast cancer bone metastasis  67    G  r  o  w   t   h   F  a  c   t  o  r  s   D  o  w  n   l  o  a   d  e   d   f  r  o  m   i  n   f  o  r  m  a   h  e  a   l   t   h  c  a  r  e .  c  o  m   b  y   P  e   t  e  r   M  a  c   C  a   l   l  u  m   C  a  n  c  e  r   C  e  n   t  r  e  o  n   0   7   /   2   4   /   1   4   F  o  r  p  e  r  s  o  n  a   l  u  s  e  o  n   l  y .
Search
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
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x