Kinome Profiling of Osteoblasts onHydroxyapatite Opens New Avenues onBiomaterial Cell Signaling

Sara Gemini-Piperni,1 Renato Milani,2 Sergio Bertazzo,3 Maikel Peppelenbosch,4

Esther R. Takamori,5 Jose Mauro Granjeiro,6 Carmen V. Ferreira,2 Anna Teti,1

Willian Zambuzzi5

1Department of Biotechnological and Applied Clinical Sciences, University of L’Aquila,

L’Aquila, Italy2Departamento de Bioqu�ımica, Instituto de Biologia, Universidade Estadual de Campinas

(UNICAMP), Campinas, SP, Brazil3Department of Materials, Imperial College London, London, United Kingdom4Department of Gastroenterology and Hepatology, Erasmus MC, University Medical

Center Rotterdam, Rotterdam, The Netherlands5Laborat�orio de Bioensaios e Dinamica Celular, Department of Chemistry and

Biochemistry, Biosciences Institute, UNESP, Campus Botucatu, Botucatu, Sao Paulo,

Brazil; telephone: þ55-14-3880-0599; fax: þ55 14 3815 3744; e-mail: wzambuzzi@ibb.

unesp.br6Instituto Nacional de Metrologia, NormalizaSc~ao e Qualidade Industrial (INMETRO),

Diretoria de Programas (DIPRO)/Bioengenharia, Xerem, RJ, Brazil

ABSTRACT: In degenerative diseases or lesions, bone tissuereplacement and regeneration are important clinical goals.The most used bone substitutes today are hydroxyapatite(HA) scaffolds. These scaffolds, developed over the last fewdecades, present high porosity and good osteointegration, buthaven’t completely solved issues related to bone defects.Moreover, the exact intracellular mechanisms involved in theresponse to HA have yet to be addressed. This prompted us toinvestigate the protein networks responsible for signaltransduction during early osteoblast adhesion on syntheticHA scaffolds. By performing a global kinase activity assay, weshowed that there is a specific molecular machineryresponding to HA contact, immediately triggering pathwaysleading to cytoskeleton rearrangement due to activation ofAdducin 1 (ADD1), protein kinase A (PKA), protein kinase C(PKC), and vascular endothelial growth factor (VEGF).

Moreover, we found a significantly increased phosphoryla-tion of the activating site Ser-421 in histone deacetylase 1(HDAC1), a substrate of Cyclin-Dependent Kinase 5(CDK5). These phosphorylation events are hallmarks ofosteoblast differentiation, pointing to HA surfaces ability topromote differentiation. We also found that AKT was keptactive, suggesting the maintenance of survival pathways.Interestingly, though, the substrate sequence of CDK5 alsopresented higher phosphorylation levels when compared tocontrol conditions. To our knowledge, this kinase has neverbefore been related to osteoblast biology, opening a newavenue of investigation for novel pathways involved in thismatter. These results suggest that HA triggers a specificintracellular signal transduction cascade during early osteo-blast adhesion, activating proteins involved with cytoskeletonrearrangement, and induction of osteoblast differentiation.

Biotechnol. Bioeng. 2014;111: 1900–1905.

� 2014 Wiley Periodicals, Inc.

KEYWORDS: bioengineering; hydroxyapatite; osteoblast;signal transduction; kinome profile; peptide array;biotechnology

Bone is a specialized type of dense connective tissue. It isconsidered a dynamic tissue since it is continually beingformed and resorbed through the processes involved in boneremodeling. These processes are known to be influencedby several factors, like hormonal changes, physical activity(or lack thereof), and drugs.

Conflict of interest: none.

Correspondence to: W.F. Zambuzzi

Contract grant sponsor: Conselho Nacional de Desenvolvimento Cientıfico e

Tecnol�ogico (CNPq)

Contract grant sponsor: FundaSc~ao de Pesquisa do Estado do Rio de Janeiro (Faperj)

Contract grant sponsor: Marie Curie International Research Staff Exchange Scheme

Fellowship (PIRSES-GA-2011-295181)

Contract grant sponsor: Rosetrees Trust and the Junior Research Fellowship scheme

at Imperial College London

Received 27 January 2014; Revision received 7 March 2014; Accepted 19 March 2014

Accepted manuscript online 25 March 2014;

Article first published online 11 June 2014 in Wiley Online Library

(http://onlinelibrary.wiley.com/doi/10.1002/bit.25246/abstract).

DOI 10.1002/bit.25246

COMMUNICATION TO THE EDITOR

1900 Biotechnology and Bioengineering, Vol. 111, No. 9, September, 2014 � 2014 Wiley Periodicals, Inc.

Because of this natural mechanism of remodeling, bonetissue is able to regenerate when affected by small lesions,but not so when lesions are larger. In these circumstances,current surgical practices for bone reestablishment are to useautologous bone grafts, bone allografts, or synthetic graftmaterials (Matassi et al., 2011).Apatite, a group of phosphate minerals, are widely spread

in nature (Bose and Tarafder, 2012; Dorozhkin, 2012). Inparticular, the inorganic phase of hard vertebrate tissues isvery similar to synthetic hydroxyapatite (HA) Ca10(PO4)6(OH)2, (Bertazzo et al., 2010a,b; Bose and Tarafder, 2012;Dorozhkin, 2012; Zambuzzi et al., 2012). The similarity withbiological apatites accounts for the high biocompatibility ofsynthetic HA, making it a potential osteo-substitute material(Bertazzo et al., 2010b). However, the exact intracellularmechanisms triggered in response to synthetic HAs are notcompletely known. The bioactivity of any material, heredefined as the ability of osteoblasts to adhere and spread uponit, depends on these signaling mechanisms (Kokubo, 1991;Zambuzzi et al., 2011).Therefore, a better understanding of the intracellular

mechanisms by which cells adhere to the surface of HA isparamount to improve proliferation and differentiation ofosteoblasts over synthetic surfaces, increasing the effective-ness of bone grafts. Cell migration, adhesion, and prolife-ration on implant surfaces are important prerequisites toinitiate the process of tissue regeneration, while modifica-tions of the implant surface by incorporation of biologicmediators of cell growth and differentiation may bepotentially beneficial in enhancing wound healing followingimplant placement.At the intracellular level, reversible protein phosphoryla-

tion is a kind of post-translational modification capable ofmediating most of the signal transduction processes in livingcells (Zambuzzi et al., 2010, 2011). Since these processes areof the utmost importance to adequate cell development,differentiation, intercellular communication, homeostasisand survival, they are usually under tight regulation, whereasabnormal phosphorylation events lead to unwanted substrateactivity and are the cause of many diseases, such as cancer andneurological disorders. Massive studies of this set of enzymeshave been performed over the last years (Krüger et al., 2008;Milani et al., 2010; Rigbolt et al., 2011; Zambuzzi et al., 2011).In this context, advances in bioinformatics and molecularbiology have been developed to allow a broader and moreefficient analysis of cellular metabolism and signalingpathways.In order to understand the signaling mechanisms behind

osteoblast adhesion and how this may impact the develop-ment of better biomaterials, we investigated the globalnetwork of signaling molecules responsible for signaltransduction during early MC3T3-E1 cells adhesion onsynthetic HA substrates. It is a well-known fact that cellsrespond differently to diverse material surfaces (Anselme,2000). As suggested by Caplan and Shah (2009), the centralparadigm for cell interpretation and response to a givenmaterial surface is that surface receptors are involved in the

transduction of information through signaling cascades thatalter gene expression and, eventually, cell behavior. There arealso consistent evidences, through micromanufacturing andnanotechnology, that such intracellular responses aretriggered even by nanoscaled surface patterns, indicating avery complex topographical perception by adhering cells(Alves et al., 2010). Over the last few years, we haveinvestigated kinase activity as a parameter to screenbiomaterials regarding their biocompatibility (Zambuzziet al., 2011). In turn, kinases are pivotal regulators of cellularmetabolism and physiology through their ability to phos-phorylate selected substrates at specific sites. In this context,we decided to investigate the network of signal transductioninvolved with early osteoblast behavior adhering onHAdiscs.In osteoblast cells subjected to adhesion on HA compared totissue culture plastic (TCP) surfaces, our data showed aremarkably different kinome profile. This reinforces ourprevious statement (Milani et al., 2010) regarding theimportance of surface type and molecular structure forosteoblast adhesion.The kinome profiling of osteoblast cells adhering on HA

and TCP showed a number of differentially phosphorylatedtarget substrates. All differentially phosphorylated substratesfor HA versus TCP surfaces can be seen in Figure 1. Most ofthe significant differentially phosphorylated substrates wereup-phosphorylated when adhering onHA (in Fig. 1, spots are

Figure 1. Median phosphorylation intensity for all 1,024 PepChip proteins and

substrates. HA refers to a hydroxyapatite surface. Control refers to a TCP surface.

Colored spots identify spots that have presented consistent replicates across PepChip

experiments. Red means less phosphorylated and green means more phosphorylated

than the control sample. Graph lines represent Fold Change¼ 0.75; 1; 1.5, top to bottom.

p before a protein name indicate it is a substrate for phosphorylation, instead of a

kinase. Comments regarding activity or inhibition refer to the phosphorylated substrate

state on HA surfaces versus TCP surfaces when this information is available. FC

indicates the individual fold change for each colored spot.

Gemini-Piperni et al.: Hydroxyapatite-Induced Osteoblast Signaling 1901

Biotechnology and Bioengineering

colored red if down-phosphorylated and green if up-phosphorylated when compared to the control sample).Among up-phosphorylated substrates, some proteins deserveattention because of their association to osteoblast differen-tiation: HDAC1 (activated by phosphorylation at S421, foldchange: 3.2, P< 0.01), EP300 (inactivated by phosphoryla-tion at S89, fold change: 1.2, P< 0.05) and ADD1 (activatedby phosphorylation at T445, fold change: 1.9, P< 0.05).HDAC1 deacetylates lysine residues on the N-terminal regionof histones, repressing transcriptional activity. HDACproteins are critical in cellular events such as cell cyclecontrol and differentiation (Lee et al., 2006; Pflum et al.,2001). Accordingly, histone acetyltransferase p300 (EP300),a protein that activates transcription through histoneacetylation, was found up-phosphorylated at different sites.Phosphorylation at S89 (fold change: 2.33, P< 0.01) byAMPK (50-AMP-activated protein kinase) reduces inter-action of p300 with nuclear receptors, impairing its activity(Yang et al., 2001). Moreover, phosphorylation at S1834 (foldchange: 1.2, P< 0.05) by AKT disrupts the interactionbetween p300 and transcription factors, indicating additionalregulation of transcriptional activity that may be related todifferentiation signaling.

We also found increased phosphorylation at S229 ofConnexin 32 (fold change: 1.94, P< 0.01), a protein involvedin the formation of gap junctions. Donahue et al. (2000)reported a correlation between Connexin 43 mRNA andprotein expression and differentiation of a human fetalosteoblastic cell line (hFOB 1.19). While there are noavailable data for Connexin 32, Cooper and Lampe (2002)have determined that gap junction assembly may be regulatedby direct phosphorylation of Connexin 43. Moreover, AKTalso was up-phosphorylated in osteoblasts responding to HA,suggesting survival pathways remained active, anotherhallmark of osteoblast differentiation (Ruggiu et al., 2013).

Another protein surprisingly activated was ADD1 (Addu-cin 1, activated by phosphorylation at T445, fold change: 1.9).Adducins control growth of filamentous actins (F-actins)via their interaction with the fast-growing barbed ends ofF-actins, spectrin, protein kinase C (PKC), protein kinase A(PKA), and calmodulin kinase, thereby affecting actinremodeling at the cortex underneath the plasma membrane(Barkalow et al., 2003; Kuhlman et al., 1996; Matsuokaet al., 2000). The C terminus of adducins containsmyristoylated alanine-rich C-kinase substrate (MARCKS)domain that interacts with those proteins involved inremodeling of F-actins (Matsuoka et al., 2000). Because ofthis, we suggest ADD1 as a important signaling proteininvolved with osteoblast adaptation on HA surfaces,governing an adequate actin rearrangement during the firstmoments of cell-HA interaction. Furthermore, there isevidence to suggest ADD1 interacts with PKA and PKCduring these events. We have previously reported that, duringthe adhesion of osteoblasts on TCP, there is also an intenserearrangement of signaling pathways involved in cytoskeletalremodeling that affects proteins such as cofilin, heat shockprotein-27 (HSP-27), mitogen-activated protein kinases,

PKA, PKC, and protein phosphatase 2A (Zambuzzi et al.,2009).

Interestingly, the substrate sequence of CDK5, a kinasepreviously associated with neuronal development (Pagliniand Cáceres, 2001), presented higher phosphorylation levels(fold change: 2.02, P< 0.01, activating residue) (Sasaki et al.,2002) when compared to control conditions (Figs. 1 and 2).To our knowledge, this kinase has never been related toosteoblast biology, opening a new avenue of investigation forpathways involved in this process. Furthermore, CDK5 is ableto phosphorylate and activate HDAC1, which is consistentwith our previously mentioned results (Pflum et al., 2001).

In addition, we have identified an increase on VEGFactivation in osteoblasts adhering on HA. It is worthy toremark that Liu et al. (2012) showed that there is an intracrinemechanism involving VEGF related to osteoblast/adipocytedifferentiation. They also demonstrate that VEGF secreted bybone marrow stromal cells has paracrine effects on osteoclast

Figure 2. Differentially phosphorylated substrates in MC3T3-E1 cells adhering on

HA surfaces compared to TCP surfaces. Green bars indicate higher phosphorylation on

HA surfaces while red bars indicate the opposite. Some of these proteins are directly

related to adhesion and differentiation processes: HDAC1, a cell cycle and

differentiation regulator, was found phosphorylated more than three times higher

on HA on an activating residue (S421). Likewise, ADD1, a protein involved in the control

of actin filament growth, was found almost two times more phosphorylated on an

activating residue (T445), suggesting a specific signaling network for HA adhesion.

1902 Biotechnology and Bioengineering, Vol. 111, No. 9, September, 2014

differentiation in assays using total bonemarrow cell cultures.Furthermore, it is known that osteoblast-derived VEGF canaffect bone formation in a paracrine manner via stimulationof angiogenesis (Wang et al., 2007). Altogether, the end resultof this complex cascade is the promotion of a wound healingprocess that includes angiogenesis.In order to determine if there would be any bias towards a

specific location for up-phosphorylated substrates duringadhesion, we have created Venn diagrams (Fig. 3) bringingsubcellular localization for all differentially activated kinases.Highly phosphorylated substrates are mostly located in themembrane in both conditions, suggesting similar mecha-nisms for adhesion on HA and on polystyrene, but throughdifferent effectors.Through the Pathway Interaction Database clustering

algorithm we found that 2 major pathways associatedwith proliferation and differentiation are relevant for thedifferentially phosphorylated protein subset: E2F transcrip-tion factor network, and glucocorticoid receptor regulatorynetwork. Another pathway, related to cytoskeleton rear-rangement, was also found to be relevant: Regulation ofnuclear beta catenin signaling and target gene transcription.A protein interaction network showing activation andinhibition relationships among differentially phosphorylatedproteins in these pathways can be seen in Figure 4.Given the complex cell-biomaterial surface interaction,

understanding the signal transduction sequence of eventsis fundamental in tailoring cell adhesion to differentbiomaterial surfaces. Should the ability to properly modulateor influence the activities of these enzymes come togetherwith a proper understanding of the underlying signaltransduction mechanisms, considerable improvement in

the next generation of dental and orthopedic devices/biomaterials may be achieved. The ability of biomaterialsto support and promote osteoblast adhesion (Bertazzoet al., 2009, 2010b) and subsequent integration betweensurgical implants and the surrounding tissue (Coelhoet al., 2009) is one of such improvements. In spite of the

Figure 3. Subcellular location for all differentially phosphorylated substrates, classified into Membrane, Cytoplasm, Nucleus, and Mitochondria. A: Substrates presenting

higher phosphorylation on hydroxyapatite surfaces. B: Substrates presenting higher phosphorylation on TCP surfaces. Highly phosphorylated substrates are mostly located in the

membrane, while no mitochondrial substrates were found to be differentially phosphorylated in either condition, suggesting adhesion on HA and on polystyrene might happen by

similar mechanisms but through different effectors.

Figure 4. Protein network depicting interactions among more phosphorylated

substrates in HA and their interaction partners in associated pathways. Blue lines

indicate binding while green arrows and perpendicular red lines indicate activation or

inhibition, respectively. Yellow lines indicate expression, purple lines indicate catalysis

and pink lines indicate post-translational modification. Black lines indicate a generic

reaction while ball-ends point to a directional interaction of unknown nature.

Gemini-Piperni et al.: Hydroxyapatite-Induced Osteoblast Signaling 1903

Biotechnology and Bioengineering

fact that HA is already used in surgical procedures, this is thefirst work within current literature reporting global osteo-blast kinase activity during adhesion, outlining a molecularprofile for adhesion on HA.

Our results show that HA requires a specific intracellularsignal transduction pathway during early osteoblast adhesionby promoting phosphorylation events that activate proteinsinvolved with cytoskeleton rearrangement and prepareosteoblasts to differentiation.

Materials and Methods

Hydroxyapatite

HA was synthesized according to the method described byLopez-Macipe et al. (1998). A thorough characterization ofthe material can be found in previous papers published byour group (Bertazzo and Bertran, 2008).

Cell Line and Culture Conditions

MC3T3-E1 (subclone 4), a preosteoblast cell line frommousecalvaria, was obtained from American Type Culture Collec-tion (Manassas, VA) and grown at 37�C in a-MEM mediumsupplemented with 10% fetal bovine serum (FBS), 100U/mLof penicillin, and 100 g/mL of streptomycin under ahumidified 5% CO2 atmosphere.

Experimental Design

50� 103 MC3T3-E1 cells were plated on 24 HA discs (0.8 cmof diameter) fitted into 24 wells dish plates and after 3 h thesamples were collected. Tissue culture plastic (TCP) was usedas an internal control.

Kinome Arrays

Kinome arrays were performed essentially as described before(Milani et al., 2010). In short, cells were washed in PBS andlysed in a non-denaturing complete lysis buffer. The peptidearrays (Pepscan, Lelystad, TheNetherlands), containing up to1,024 different kinase substrates in triplicate, were incubatedwith the cell lysates for 2 h in a humidified incubator at 37�C.Subsequently, the arrays were washed in 2M NaCl, 1%Triton-X-100, PBS, 0.1% Tween and water; thereafter slideswere exposed to a phospho-imaging screen for 24–72 h andscanned on a phospho-imager (Fuji Storm 860, Stanford,CA). The level of incorporated radioactivity, which reflectsthe extent of phosphorylation, was quantified with specificarray software (EisenLab ScanAlyze, version 2.50).

Statistics

Datasets from chips were then analyzed statistically usingPepMatrix, as described by Milani et al. (2010). In short,spot replications were scrutinized for consistency using twoindexes: one being the standard deviation/average (SD/A)

ratio and the other being the ratio between the average andthe median (A/M) of all three replications for each chip.Parameters applied to the indexes were an SD/A< 50% and80% <A/M< 120%. The fold change in phosphorylationbetween control and treated cells was assessed using Student’st-test, with P< 0.05 indicating significance.

Pathway Analysis

The Pathway Interaction Database clustering algorithm wasused to return relevant pathways containing one or moreproteins from the differentially phosphorylated subset. ThePathway Interaction Database is a structured collectionof signaling pathways assembled from known molecularinteractions. All information on the database is curated andreviewed by invited specialists.

The authors would like to thank FAPESP, FAPERJ, CNPq. Thisresearch was also supported by a Marie Curie International ResearchStaff Exchange Scheme Fellowship to S.G.P. within the 7th EuropeanCommunity Framework Programme. S.B. was supported by theRosetrees Trust and the Junior Research Fellowship scheme atImperial College London. W.F.Z. is supported by a fellowship atCNPq (PQ-2). Sara Gemini Piperni, Anna Teti, Willian Zambuzzi arethe members of INTERBONE Corsortium, financed by the EuropeanCommission, Grant nr. PIRSES-GA-2011-295181.

References

Alves NM, Pashkuleva I, Reis RL, Mano JF. 2010. Controlling cell behaviorthrough the design of polymer surfaces. Small 6:2208–2220.

Anselme K. 2000. Osteoblast adhesion on biomaterials. Biomaterials 21:667–681.

Barkalow KL, Italiano JE, Jr. Chou DE, Matsuoka Y, Bennett V, Hartwig JH.2003. Alpha-adducin dissociates from F-actin and spectrin duringplatelet activation. J Cell Biol 161:557–570.

Bertazzo S, Bertran CA. 2008. Effect of hydrazine deproteination on bonemineral phase: A critical view. J Inorg Biochem 102:137–145.

Bertazzo S, Zambuzzi WF, Campos DD, Ferreira CV, Bertran CA. 2010a. Asimple method for enhancing cell adhesion to hydroxyapatite surface.Clin Oral Implants Res 21:1411–1413.

Bertazzo S, Zambuzzi WF, Campos DD, Ogeda TL, Ferreira CV, Bertran CA.2010b. Hydroxyapatite surface solubility and effect on cell adhesion.Colloids Surf B Biointerfaces 78:177–184.

Bertazzo S, Zambuzzi WF, da Silva HA, Ferreira CV, Bertran CA. 2009.Bioactivation of alumina by surface modification: A possibility forimproving the applicability of alumina in bone and oral repair. Clin OralImplants Res 20:288–293.

Bose S, Tarafder S. 2012. Calcium phosphate ceramic systems in growthfactor and drug delivery for bone tissue engineering: A review. ActaBiomater 8:1401–1421.

Caplan MR, Shah MM. 2009. Translating biomaterial properties tointracellular signaling. Cell Biochem Biophys 54:1–10.

Coelho PG, Granjeiro JM, Romanos GE, Suzuki M, Silva NR, Cardaropoli G,Thompson VP, Lemons JE. 2009. Basic research methods and currenttrends of dental implant surfaces. J Biomed Mater Res B Appl Biomater88:579–596.

Cooper CD, Lampe PD. 2002. Casein kinase 1 regulates connexin-43 gapjunction assembly. J Biol Chem 277:44962–44968.

Donahue HJ, Li Z, Zhou Z, Yellowley CE. 2000. Differentiation of humanfetal osteoblastic cells and gap junctional intercellular communication.Am J Physiol Cell Physiol 278:C315–C322.

Dorozhkin S. 2012. Calcium orthophosphates and human beings: Ahistorical perspective from the 1770s until 1940. Biomaterials 2:53–70.

1904 Biotechnology and Bioengineering, Vol. 111, No. 9, September, 2014

Kokubo T. 1991. Bioactive glass ceramics: Properties and applications.Biomaterials 12:155–163.

Krüger M, Kratchmarova I, Blagoev B, Tseng YH, Kahn CR, Mann M.2008. Dissection of the insulin signaling pathway via quantitativephosphoproteomics. Proc Natl Acad Sci USA 105:2451–2456.

Kuhlman PA, Hughes CA, Bennett V, Fowler VM. 1996. A new function foradducin. Calcium/calmodulin-regulated capping of the barbed ends ofactin filaments. J Biol Chem 27:7986–7991.

Lee HW, Suh JH, Kim AY, Lee YS, Park SY, Kim JB. 2006. Histone deacetylase1-mediated histone modification regulates osteoblast differentiation.Mol Endocrinol 20:2432–2443.

Liu Y, Berendsen AD, Jia S, Lotinun S, Baron R, Ferrara N, Olsen BR.2012. Intracellular VEGF regulates the balance between osteoblast andadipocyte differentiation. J Clin Invest 122:3101–3136.

Lopez-Macipe A, Rodrigues-Clemente R, Hidalgo-L�opez A, Arita I, Garc�ıa-Gadu~no MV, Rivera E, Casta~no VM. 1998. Wet chemical synthesisof hydroxyapatite particles from nonstoichiometric solutions. J MaterSynth Process 6:21–26.

Matassi F, Nistri L, Chicon Paez D, Innocenti M. 2011. New biomaterials forbone regeneration. Clin Cases Miner Metab 8(1):21–24.

Matsuoka Y, Li X, Bennett V. 2000. Adducin: Structure, function andregulation. Cell Mol Life Sci 57:884–895.

Milani R, Ferreira CV, Granjeiro JM, Paredes-Gamero EJ, Silva RA,Justo GZ, Nader HB, Galembeck E, Peppelenbosch MP, Aoyama H,Zambuzzi WF. 2010. Phosphoproteome reveals an atlas of proteinsignaling networks during osteoblast adhesion. J Cell Biochem 109:957–966.

Paglini G, Cáceres A. 2001. The role of the Cdk5–p35 kinase in neuronaldevelopment. Eur J Biochem 268:1528–1533.

Pflum MK, Tong JK, Lane WS, Schreiber SL. 2001. Histone deacetylase 1phosphorylation promotes enzymatic activity and complex formation.J Biol Chem 276:47733–47741.

Rigbolt KT, Prokhorova TA, Akimov V, Henningsen J, Johansen PT,Kratchmarova I, Kassem M, Mann M, Olsen JV, Blagoev B. 2011.

System-wide temporal characterization of the proteome and phospho-proteome of human embryonic stem cell differentiation. Sci Signal 4:rs3.

Ruggiu A, Ulivi V, Sanguineti F, Cancedda R, Descalzi F. 2013. The effect ofPlatelet Lysate on osteoblast proliferation associated with a transientincrease of the inflammatory response in bone regeneration. Bio-materials 34:9318–9330.

Sasaki Y, Cheng C, Uchida Y, Nakajima O, Ohshima T, Yagi T, Taniguchi M,Nakayama T, Kishida R, Kudo Y, Ohno S, Nakamura F, Goshima Y. 2002.Fyn and Cdk5 mediate semaphorin-3A signaling, which is involved inregulation of dendrite orientation in cerebral cortex. Neuron 35:907–920.

Wang Y, Wan C, Deng L, Liu X, Cao X, Gilbert SR, Bouxsein ML, FaugereMC, Guldberg RE, Gerstenfeld LC, Haase VH, Johnson RS, Schipani E,Clemens TL. 2007. The hypoxia-inducible factor alpha pathway couplesangiogenesis to osteogenesis during skeletal development. J Clin Invest117:1616–1626.

Yang W, Hong YH, Shen XQ, Frankowski C, Camp HS, Leff T. 2001.Regulation of transcription by AMP-activated protein kinase: Phos-phorylation of p300 blocks its interaction with nuclear receptors. J BiolChem 276:38341–38344.

Zambuzzi WF, Bruni-Cardoso A, Granjeiro JM, Peppelenbosch MP, deCarvalho HF, Aoyama H, Ferreira CV. 2009. On the road to under-standing of the osteoblast adhesion: Cytoskeleton organization isrearranged by distinct signaling pathways. J Cell Biochem 108:134–144.

ZambuzziWF, Coelho PG, Alves GG, Granjeiro JM. 2011. Intracellular signaltransduction as a factor in the development of “smart” biomaterials forbone tissue engineering. Biotechnol Bioeng 108:1246–1250.

Zambuzzi WF, Fernandes GV, Iano FG, Fernandes Mda S, Granjeiro JM,Oliveira RC. 2012. Exploring anorganic bovine bone granules asosteoblast carriers for bone bioengineering: A study in rat critical-sizecalvarial defects. Braz Dent J 23:315–321.

Zambuzzi WF, Milani R, Teti A. 2010. Expanding the role of Src and protein-tyrosine phosphatases balance in modulating osteoblast metabolism:Lessons from mice. Biochimie 92:327–332.

Gemini-Piperni et al.: Hydroxyapatite-Induced Osteoblast Signaling 1905

Biotechnology and Bioengineering