the osteogenic activity of strontium loaded titania ... · the osteogenic activity of strontium...

The osteogenic activity of strontium loaded titania nanotube arrays on titaniumsubstrates

Lingzhou Zhao a,***,1, Hairong Wang b,1, Kaifu Huo b,c,d,**, Xuming Zhang c, Wei Wang e, Yumei Zhang e,Zhifen Wu a, Paul K. Chu c,*

aDepartment of Periodontology and Oral Medicine, School of Stomatology, The Fourth Military Medical University, Xi’an 710032, Chinab School of Materials and Metallurgy, Wuhan University of Science and Technology, Wuhan 430081, ChinacDepartment of Physics and Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, Chinad State Key Laboratory of High Performance Ceramics and Superfine Microstructure, Shanghai Institute of Ceramics, CAS, Shanghai, 200050, ChinaeDepartment of Prosthetic Dentistry, School of Stomatology, The Fourth Military Medical University, Xi’an 710032, China

a r t i c l e i n f o

Article history:Received 31 July 2012Accepted 20 September 2012Available online 6 October 2012

Keywords:Titania nanotubesStrontiumProtein adsorptionMesenchymal stem cellsOsteogenic differentiation

a b s t r a c t

Development of biomedical titanium implants with high osteogenic ability for fast and good osseoin-tegration under normal as well as osteoporotic conditions is hotly pursued. Strontium (Sr) loadednanotubular structures (NT-Sr) that allow controlled and long-term Sr release are expected to yieldfavorable osteogenic effects and properties. NT-Sr structures with different tube diameters are fabricatedby hydrothermal treatment of titania nanotubes formed at 10 and 40 V (NT10 and NT40). The loaded Sramounts are regulated by the hydrothermal treatment time of 1 and 3 h (samples NT10-Sr1, NT10-Sr3,NT40-Sr1 and NT40-Sr3) in the Sr(OH)2 solution. Long lasting and controllable Sr release is observedfrom the NT-Sr samples with no cytotoxicity. The samples NT10 and NT10-Sr have multiple nanocues,comprising bundles of nanotubes of less than or equal to 30 nmwith bundle diameters between 100 and400 nm separated by about 80 nm. Sr incorporation enhances proliferation of rat mesenchymal stemcells (MSCs) on the NT structure, especially NT10-Sr which promotes the spread of the MSCs intoa polygonal osteoblastic shape. Both the NT and NT-Sr samples promote osteogenesis to varying degreesas indicated by gene expression and among the various samples, samples NT10-Sr3 and NT40-Srsignificantly up-regulate the expressions of the osteogenesis related genes in the absence of an extraosteogenic agent. Samples NT10 and NT10-Sr generate big nodular alkaline phosphatase (ALP) productsand induce extracellular matrix (ECM) mineralization, and the effects on NT10-Sr3 are most obvious dueto the multiple scaled nanostructure and proper amount of incorporated Sr. In comparison, less ALPproducts and failure to induce ECM mineralization are observed from sample NT40-Sr, possibly due tocell function impairment by the uneven protein distribution. NT10-Sr3 which shows excellent osteogenicproperties is very attractive and has large clinical potential.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Mesenchymal stem cells (MSCs) play a crucial role in boneregeneration and bony fixation of implanted biomaterials. Most ofthe osteoblastic cells that colonize the implant surface to induce

bone growth originate from MSCs [1], and hence, in order toaccomplish good osseointegration, it is critical to induce thedifferentiation of MSCs preferentially toward osteogenitor cells andthen into osteoblasts in lieu of other cell lineages. Titanium (Ti)implants are commonly used in orthopedics and dentistry, but themetallic Ti surface fails to preferentially induce MSC commitmenttoward osteoblasts compared to other cell lineages to generatebetter osseointegration, and improvement is thus needed to meetthe demand for better clinical performance. Surface nano-topography such as TiO2 nanotube (NT) coatings have been shownto alter cell behaviors such as adhesion, orientation, cell differen-tiation, and migration significantly [2e4] due to their similardimensional scale with bone collagen fibrils and elasticity

* Corresponding author. Tel.: þ852 27887724; fax: þ852 27889549.** Corresponding author. School of Materials and Metallurgy, Wuhan University ofScience and Technology, Wuhan 430081, China.*** Corresponding author.

E-mail addresses: [email protected] (L. Zhao), [email protected], [email protected] (K. Huo), [email protected] (P.K. Chu).

1 These authors contributed equally to this work

Contents lists available at SciVerse ScienceDirect

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

0142-9612/$ e see front matter � 2012 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biomaterials.2012.09.041

Biomaterials 34 (2013) 19e29

resembling that of bones [5,6]. In addition, the NT also constitutesan excellent drug delivery platform, particularly for inorganicbioactive elements such as silver and trace elements. They arestable and function at low doses and can potentially generatinglong-lasting delivery effects [7e9].

Strontium (Sr) has aroused tremendous clinical interests espe-cially after the development of the anti-osteoporosis drug stron-tium ranelate (SR) which exhibits pronounced effects to decreasethe bone fracture risk in osteoporotic patients [10,11]. The favorableeffects which have been observed from osteoporotic and normalanimals [12,13] are closely related to the modulation of boneturnover toward osteogenesis [14e16]. Sr increases osteoblastreplication, differentiation, and bone matrix mineralization prob-ably via a calcium sensing receptor (CaR) dependent mechanism[14,16e18]. Sr has also been recently reported to direct the MSCcommitment to the bone lineage via the Wnt/b-catenin and MAPKpathways as well as induction of cyclooxygenase (COX)-2 andprostaglandin E2 expressions while simultaneously repressingtheir commitment to other lineages such as adipocytes [19e23]. Onthe other hand, Sr can inhibit bone resorption by reducing osteo-blast matrix metalloproteinase production, reduce osteoclastdifferentiation and resorbing activity, and induce osteoclastapoptosis through a CaR dependent mechanism [14,16,24]. Oraladministration of SR has been reported to effectively enhance boneimplant fixation in osteoporotic and intact rats [25,26], butconstant in situ Sr release of a proper dose directly to implantetissue interface gives rise to better results while avoiding thepotential deleterious effects associated with oral administration athigh concentrations [8]. Sr has been experimentally incorporatedinto biomaterials such as calcium phosphates and bioactive glassesto enhance the bone formation [18,27e30]. Nonetheless, becausethe solubility of calcium phosphates and bioglasses changesappreciably with the amounts of Sr [31], it is difficult to achievelong-lasting Sr release at a reasonably constant rate. In this respect,a more efficient delivery platform with controlled Sr release is ofclinical interest.

We have fabricated a Sr loaded nanotubular structure (NT-Sr) onTi implants by transforming the NT into SrTiO3 while retaining thenanotubular structure [8]. This Sr releasing NT-Sr coating is ex-pected to expedite osteointegration especially in osteoporoticpatients. Although a low dose of Sr benefits bone formation, anexcessively high dose may induce toxicity thereby posing delete-rious effects on bone mineralization [16,31,32]. Hence, an NT-Srcoating with the optimal dose and proper Sr release is requiredfor osteointegration. Recent investigations suggest that the NTdiameter could affect the activity of bone cells and an optimaldiameter is thus required for cell functionality. In this work, a seriesof NT and NT-Sr samples with large and small tube sizes as well ashigh and low Sr release amounts are prepared and the influence ofthe NT diameter and Sr contents on rat MSC functionalities in theabsence of exogenic osteogenic factors (OS) are investigatedsystematically.

2. Materials and methods

2.1. Sample synthesis and characterization

Ti foils (99.7% pure, Aldrich) with the size of 10� 10� 1mm3 were first polishedby SiC sandpaper and then ultrasonically cleaned with acetone, ethanol, anddeionized water sequentially. NT10 and NT40 were fabricated by electrochemicalanodization in an ethylene glycol solution with 0.5 wt% NH4F, 5 vol% CH3OH, and 5vol% H2O at 10 V for 1 h and 40 V for 40min, respectively. Afterward, NT10 and NT40were placed in 40 ml of 0.02 M Sr(OH)2 solution in a 60 ml Teflon-lined autoclaveand heated at 200 �C for 1 and 3 h to obtain the NT-Sr samples. Then the specimenswere ultrasonically washed with 1 M HCl for 5 min to remove residual Sr(OH)2,rinsed with distilled water, and dried in air. The samples were characterized by field-emission scanning electron microscopy (FE-SEM, HITACHI S-4800), X-ray diffraction

(XRD, Philips X0 Pert Pro) with Cu Ka radiation, micro Raman spectroscopy(Renishaw 2000), and X-ray photoelectron spectroscopy (XPS, ESCALB MK-II). Watercontact anglemeasurements were carried out by the sessile-dropmethod (EasyDropStandard, KRUSS).

2.2. Sr release determination

The NT-Sr samples were immersed in 5ml of phosphate buffered saline (PBS) for1 day, taken out, and then immersed again in 5 ml of fresh PBS. This process wasrepeated for a total of 1 month to generate solutions at different time points in orderto determine the Sr release time profile. The PBS solution containing released Sr wasanalyzed by inductively-coupled plasma atomic emission spectrometry (ICP-AES,IRIS Advantage ER/S). The NT-Sr layers on the Ti surfaces were completely dissolvedin the HF and HNO3 mixture when ultrasonically treated at room temperature for5 min. The total amounts of Sr were determined by ICP-AES.

2.3. Protein adsorption assay

The protein adsorption assay was conducted in a-MEM containing 10% fetal calfserum (FCS). After incubation in the medium for 2 h at 37 �C, the proteins adsorbedonto the samples were detached by 1% sodium dodecyl sulfate (Solarbio) anddetermined using a MicroBCA protein assay kit (Pierce). The protein adsorptionpattern after incubating for 2 days in a-MEM containing 10% FCS was inspected byFE-SEM.

2.4. Cell cultures, cytotoxicity, cell adhesion and proliferation

MSCs and green fluorescenceprotein (GFP) labeled ones were obtained fromSpragueeDawley rats (Lab Animal Center, the Fourth Military Medical University)and GFP transgenic rats (Rat Resource & Research Center, the University of Missouri)respectively and the detailed cell isolation procedures are described in Ref. [4]. Thecells were cultured in a-MEM containing 10% FCS at 37 �C with the medium changedevery three days. The cells of passage 2e4 were used in the experiments. Thesamples were placed in 24 well plates (Costar) and the MSCs were seeded ata density of 4 � 104/well for the cell adhesion assay and 2 � 104/well for the otherassays unless other mentioned. For the cell adhesion assay, after incubation of 30, 60and 120 min, the attached cells were stained by 40 ,6-diamidino-2-phenylindole(DAPI, Sigma). Images were captured from five random fields by a fluorescencemicroscope (Leica) and the cell number in each field was counted. The activity oflactate dehydrogenase (LDH) in the culture media released by the cells was used asan index of cytotoxicity. After culturing for 3 days, the culture mediumwas collected(among this period the medium was not changed) and centrifuged, and the LDHactivity in the supernatant was determined spectrophotometrically according to themanufacturer’s instruction (Sigma). In order to assess cell proliferation, the cellswere cultured on samples for 1, 3 and 7 days and the cell numbers were assessedusing the Cell Counting Kit-8 assay (CCK-8, Beyotime).

2.5. Cell morphology

The cell morphology and details of cell/biomaterial interaction were studied byFE-SEM and fluorescence image of the GFP labeled MSCs. After incubation for 2 daysin a-MEM containing 10% FCS, the samples with attached cells were fixed in 3%glutaraldehyde, dehydrated in a graded ethanol series, freeze-dried, sputter coatedwith gold, and observed by high-resolution FE-SEM. The GFP labeled MSCs wereinoculated at a density of 1 �104/well, and after 2 h, 3 and 8 days, the images werecaptured by fluorescence microscopy.

2.6. Cell migration

In thewound healing assay experiments, the GFP labeledMSCs were cultured onsamples at a density of 1.2 � 105/well for 18 h to reach confluence. The monolayerswere wounded with a plastic pipette. After washing with PBS, the cells were incu-bated for another 2 days and then observed as aforementioned.

2.7. Quantitative reverse transcription polymerase chain reaction (qRT-PCR),alkaline phosphatase (ALP) and mineralized nodule staining

MSC osteogenic differentiation was assessed. The expression levels of osteo-genesis related genes were measured using the qRT-PCR. The cells were seeded ata density of 2 � 104 cells/well, cultured for 2 weeks and harvested using TRIzol(Invitrogen) to extract the RNA. The obtained RNA was reverse transcribed intocomplementary DNA (cDNA) using PrimeScript RT reagent Kit (Takara) and the qRT-PCR analysis was performed on the Bio-Rad C1000 using SYBR Premix Ex Taq II(Takara). The primers for the target genes are listed in Table 1. The expression levelsof the target genes were normalized to that of the housekeeping gene GAPDH. ALPstaining was performed with the BCIP/NBT alkaline phosphatase color developmentkit (Beyotime) after culturing for 5 and 10 days. After culturing for 2 weeks, extra-cellular matrix (ECM) mineralization by the cells on the samples was assessed byAlizarin Red staining [33,34].

L. Zhao et al. / Biomaterials 34 (2013) 19e2920

2.8. Statistical analysis

The data were collected from three separate experiments and expressed asmeans � standard deviation. The one way ANOVA and Student-Newman-Keuls posthoc tests were used to determine the level of significance and p values less than 0.05and 0.01 were considered to be significant and highly significant, respectively.

3. Results and discussion

3.1. Synthesis and structure characterization

TheNTsamples producedbyanodization at10 and40V (NT10andNT40) are subjected to ahydrothermal treatment for 1 and3h to formdifferent Sr-containing NT samples (denoted as NT10-Sr1, NT10-Sr3,NT40-Sr1, and NT40-Sr3). SEM images indicate that NT10 and NT40have diameters of 30 and 80 nm (Fig.1B). In particular, NT10 containsmultiple nanocues comprising bundles of 30 nm nanotubes withbundle diameters between 100 and 400 nmseparated by 80nmwidecracks. The NT10 structure is similiar to the highly ordered hierar-chical structure of the natural bone ECM and good biological perfor-mance is therefore expected. The top ends of 80nmnanotubewall onNT40 are not completely smooth and have flakes. After the hydro-thermal treatment, the nanotubular architecture is preserved but thetube diameters decrease with treatment time due to volume expan-sion from the transformation of TiO2 into SrTiO3. The flakes on top ofthe nanotube wall of NT40 are blunt on NT40-Sr1 but smoother onNT40-Sr3. During the hydrothermal process, the surface TiO2dissolves and react to form SrTiO3 and the dissolution and re-recipitation processes make the sample surface smoother.

The XRD spectra in Fig. 2A disclose only Ti and anatase TiO2 inthe annealed NT while the peaks from crystalline SrTiO3 areobserved for NT-Sr. The SrTiO3 peaks become more intense whilethe TiO2 peaks weaken as the hydrothermal reaction timeincreases, indicating gradual transformation of TiO2 into SrTiO3during the hydrothermal process. The Raman spectra of therepresentative samples NT40 and NT40-Sr3 in Fig. 2B furthercorroborate the XRD results. The XPS spectra in Fig. 2C acquiredfrom the NT-Sr reveal peaks corresponding to Ti, O, and Sr. The XPSdepth profiles of Ti, O, and Sr acquired from NT-Sr (Fig. S1) suggestthat Sr is distributed along the entire length of the nanotubes. Sincethe nanotubes on NT40 are much longer than those on NT10, Srpenetrates a larger depth in NT40-Sr than NT10-Sr. The Sr contentsalong the nanotube depth of NT10-Sr1, NT10-Sr3, NT40-Sr1, andNT40-Sr3 are shown in Fig. 2D, indicating that the Sr content islarger on the surface and decreases gradually with depth. Thisphenomenon is attributed to the insufficient hydrothermal reactionat the deeper sites since it is more difficult for the reactants todiffuse to here. In addition, the Sr contents increase with thehydrothermal treatment time. The water contact angle measure-ments (Fig. S2) reveal that the NT samples are hydrophilic and Srincorporation does not alter the water contact angle significantly.

3.2. Sr release

The Sr release kinetics is assessed by immersing the NT-Sr in5 mL of PBS for as long as 1 month, and as shown in Fig. 3A, the

released Sr amounts at various time slots follow the order of NT40-Sr3 > NT40-Sr1 > NT10-Sr3 > NT10-Sr1. There are two possiblemechanisms by which Sr is released from the NT-Sr samples,namely dissolution of strontium titanate and ion exchangebetween Sr ions in strontium titanate and hydrogen ions in thesurrounding solution. Generally, NT-Sr shows an initial Sr releaseburst but four days later, the released Sr amounts are relativelyconstant and in fact exhibit a slight decline. The average Sr amountsreleased from NT10-Sr1, NT10-Sr3, NT40-Sr1, and NT40-Sr3 dailyare 0.025, 0.039, 0.042, and 0.053 ppm, respectively, with theexception of the burst release at day 1. The total Sr contents leachedfrom the four samples are 45.48, 54.95, 88.20, and 118.00 mg basedon the 1 cm2 coatings, respectively (Fig. 3B). Theoretically, Sr maybe delivered from the NT-Sr for a period longer than 1 year.

3.3. Protein adsorption

The proteins adsorbed on the NT and NT-Sr samples are inves-tigated because they play a crucial role in conveying the biologicaleffects of the topographical cue [35]. In comparison with the pris-tine Ti control, the NT samples adsorb more proteins and NT40absorbs evenmore (Fig. 4A). The adsorbed protein amount dependsmainly on the nanotopography irrespective of Sr incorporation,even though Sr incorporation alters the nanotopography to somedegree. Besides the amount, the species, conformation, and orien-tation of the adsorbed proteins have been reported to influence thecell/biomaterials interaction [35]. The details of protein adsorptionare examined by FE-SEM and the results are displayed in Fig. 5. OnNT10 and NT10-Sr, there are abundant proteins on the top ends ofthe nanotube bundles along the nanotube walls or protein aggre-gates covering the tubes. The proteins are distributed relativelyevenly along the nanotopography. NT40 and NT40-Sr induce moreprotein deposition and this is in good agreement with the assayresults in Fig. 4A. Unexpectedly, the adsorbed proteins are notdistributed evenly but rather form pillars, especially on NT40 onwhich tall protein pillars with relatively small top dimensionsmostly<50 nm are distributed at certain distances from each other.The uneven protein distribution on NT40 and NT40-Sr can beascribed to the nonsmooth top ends of the nanotube wall withflakes, which provide local nucleation sites for protein aggregation.The proteins on NT40-Sr are more even because the top ends of thenanotube wall are smoother, but they are still uneven compared toNT10 and NT10-Sr. The nanotubes formed in an inorganic electro-lyte have a flat top consequently inducing a uniform distribution ofproteins [4]. Hence, even subtle changes in the nanotopography canlead to dramatic alteration in the protein deposition pattern.Because the immediate substrate to that cells interact is theadsorbed proteins, the notably uneven protein deposition willmake the ultimate topographical cues exposed to cells quitedifferently from the primitive nanotopographies thereby alteringthe biological performance. This issue will be discussed in moredetails later in this paper.

3.4. Cytotoxicity, cell adhesion, and proliferation

Initial cell adhesion is the key step for the ensuing cell prolif-eration and differentiation on biomaterials [36]. As shown in Fig. 4Band Fig. S3, the big or small NT samples do not induce obviousdifference in the initial adherent cell number and this is consistentwith our previous observation from NT formed in an inorganicelectrolyte [4,37]. The available evidence demonstrates that the NTsamples have a negligible effect on the initial adherent cell number.After Sr incorporation, NT10-Sr1, NT10-Sr3, and NT40-Sr1 stillshow no obvious difference in the initial adherent cell number incontrast with the Sr free counterparts, whereas NT40-Sr3 induces

Table 1Primers used for qRT-PCR.

Gene Forward primer sequence (50-30) Reverse primer sequence (50-30)

RUNX2 CCATAACGGTCTTCACAAATCCT TCTGTCTGTGCCTTCTTGGTTCALP AACGTGGCCAAGAACATCATCA TGTCCATCTCCAGCCGTGTCOCN GGTGCAGACCTAGCAGACACCA AGGTAGCGCCGGAGTCTATTCACol-1 GCCTCCCAGAACATCACCTA GCAGGGACTTCTTGAGGTTGBMP-2 CAACACCGTGCTCAGCTTCC TTCCCACTCATTTCTGAAAGTTCCGAPDH GGCACAGTCAAGGCTGAGAATG ATGGTGGTGAAGACGCCAGTA

L. Zhao et al. / Biomaterials 34 (2013) 19e29 21

a larger adherent cell number at 1 h. Panzavolta et al. show that Srincorporation has a slight positive effect on the early adherent cellnumber [38], while Park et al. find that Sr incorporated Tie6Ale4Vfabricated by a hydrothermal process lead to no difference in theinitial attached cell numbers [30]. Hence, Sr incorporation has noobvious influence on the initial adherent cell number. The variation

shown in different reports may be explained by the different Sramounts and release kinetics.

Since Sr overdosemay lead to cytotoxicity [31], the LDH releasedby cells when co-cultured with the samples is evaluated as anindication of cytotoxicity. As shown in Fig. 4C, all the NT and NT-Srsamples exhibit no cytotoxicity compared to the Ti control and Sr

Fig. 1. (A) Schematic showing that the NT-Sr coating, combining the effect of Sr and the nanomorphology of NT, dramatically promotes MSC spreading and induces MSC selectivedifferentiation toward osteoblasts. (B) FE-SEM images of NT and NT-Sr.


amounts released from all the NT-Sr samples are safe. The LDHrelease from NT10-Sr is slightly smaller than that from the Ticontrol, indicating even better cytocompatibility on NT10-Sr. Thegood cytocompatibility is further verified by the good attachment,spread and proliferation of cells after incubation for 8 days asshown by the FE-SEM and fluorescence pictures in Figs. 6 and 7A.

Cell proliferation is assessed by CCK-8 and as shown in Fig. 4D,the cells on all the samples proliferate with time. Cell proliferationon NT is obviously lower than that on the Ti control, and NT10 and

NT40 induce similar levels of cell proliferation. NT10-Sr and NT40-Sr show better cell proliferation than NT10 and NT40, except at day7, NT10-Sr, NT10 and NT40 generate similar cell proliferation. Cellproliferation on NT40-Sr is similar to that on the control, whereasthat on NT10-Sr is still lower than the control. The reduced cellproliferation on NT10, NT40, and NT10-Sr cannot originate fromcytotoxicity which can be ruled out based on the LDH assay and socan be partly attributed to the extraordinary extension and intimateattachment of cells on them, especially on NT10-Sr, because

Fig. 3. (A) Non-cumulative Sr release time profiles from NT-Sr into PBS and (B) total Sr content of NT-Sr.

Fig. 2. (A) XRD patterns, (B) Raman spectra, (C) XPS survey spectra and (D) Sr depth profiles of the samples: (a) NT10, (b) NT10-Sr1, (c) NT10-Sr3, (d) NT40, (e) NT40-Sr1 and (f)NT40-Sr3.


detachment from the substrate is needed for cells to undergodivision and excessively strong adhesion will hinder this processand cell proliferation. Another potential reason is the osteogenicdifferentiation tendency of MSCs for the reciprocal relationshipbetween cell proliferation and differentiation [39]. On NT40, thecompromised cell proliferation may be ascribed to the proteinnanopillars which compromise cell focal adhesion and prolifera-tion. Previous reports have demonstrated the dose dependenteffects of Sr incorporation in cell proliferation [29,40] and our

results are consistent. The enhanced cell proliferation by NT10-Sr incontrast to NT10 should be mainly related to Sr, but the higher cellproliferation on NT40-Sr than NT40 is the combined effects of Srand smoothing of the nanotube wall. Even though the cells onNT40-Sr also show the osteogenic differentiation tendency, thelarger amount of released Sr still within the safe dosage range,relatively loose cell attachment to the substrate, and smaller cellspread area give rise to relatively high cell proliferation comparedto NT10-Sr.

Fig. 4. (A) Protein adsorption to the samples after 2 h of immersion in a-MEM containing 10% FCS, (B) Initial adherent MSC number at 0.5, 1 and 2 h measured by counting cellsdisplayed with DAPI, (C) LDH amount released by cells during the first 3 days of incubation and (D) cell proliferation measured by CCK-8 after 1, 3 and 7 days of culture. *, **p < 0.05or 0.01 vs Ti, #, ##p < 0.05 or 0.01 vs NT10, and %, %% p < 0.05 or 0.01 vs NT40.

Fig. 5. Protein adsorption pattern after 2 days of incubation in a-MEM containing 10% FCS.


3.5. Cell morphology

The cell morphology is closely related to the cell fate [41]. At 2 h,the cells do not spread completely and there is no discernibledifference in the cell morphology (Fig. S4). However, after completespreading, dramatic differences in the cell shape can be observed(Fig. 6). The MSCs on the flat Ti spread poorly into a spindle shapeindicative of undifferentiated quiescent cells. On the other hand,those on NT10 are more extended exhibiting a typical polygonalosteoblastic shape, but cell spreading is not promoted appreciably

on NT40. Longer incubation reveals that the effects on the cellmorphology are long lasting, as shown in Fig. 7A and Fig. S4. At firstglance, the results appear to contradict our previous observationthat both NT samples (30 and 80 nm) fabricated in an inorganicelectrolyte show notable promoting effects on MSC adhesion andspread [4]. Regarding the distribution of the adsorbed proteins, thepresent results actually are intrinsically consistent with ourprevious observation and they can be explained by cell/nano-topography interactions. Stable cell adhesion relies on the forma-tion of focal adhesion involving clustering of ligated integrins and

Fig. 6. SEM views of cells after 2 days of culture on different samples: (a) Ti, (b) NT10, (c) NT10-Sr1, (d) NT10-Sr3, (e) NT-40, (f) NT40-Sr1 and (g) NT40-Sr3.


requiring a distance smaller than 50e70 nm between two neigh-boring integrins. A distance beyond that will impede integrinclustering and focal adhesion assembly [4]. With regard to NT10,although focal adhesion will be interrupted by the 80 nm cracksand constrained to the top area of the nanotube bundles, theabundant and relatively evenly distributed nanoscale proteins onthe bundles allow the formation of high quality focal adhesion tofoster good cell adhesion, as clearly indicated in Fig. 6b3. However,for NT40, the relatively sparsely distributed nanoscale proteinpillars, to some extent like the quasi-aligned nanowire arrayspreviously reported by us [42], do not favor cell adhesion, becausethe focal adhesion will mostly and maybe even totally be con-strained to the small top end of the protein pillars, thus resulting incompromised cell adhesion and spreading. The cell mobility assayresults in Fig. 7B further demonstrate the impairing effect of NT40on cell functions. The cell mobility on NT40 is severely impairedand it becomes better on NT40-Sr possibly due to the more evenprotein distribution. Our study demonstrates that the proteindeposition pattern can be influenced by the nanotopography whichin turn can affect the biological performance of biomaterials.

Incorporation of Sr into the NT promotes cell spreading asshown in Figs. 6, 7A, and Fig. S4. Further improvement in cellspreading on NT10-Sr compared to NT10 is attributed to Sr incor-poration. With regard to the better cell spreading on NT40-Srrelative to NT40, besides Sr incorporation, smoothing of thenanotube wall is also a factor. As shown in Fig. 6f3 and g3, the moreevenly distributed proteins on NT40-Sr, especially on NT40-Sr3,leads to closer cell adhesion. Nonetheless, the cell spreading onNT40-Sr is still worse than that on NT10-Sr. In addition, an inter-esting phenomenon is that the cells on NT40-Sr have a thin cellbody but that on NT10-Sr is thicker, as shown in Fig. 6. The

abnormally thin cell body on NT40-Sr is due to two factors, the Srincorporation and the uneven protein distribution. Sr incorporationenables a larger spread area, but the unevenly deposited proteinsgive rise to smaller and fewer focal adhesions allowing fewerskeleton fibers to be linked to them. As a result, a thin cell bodyoccurs. On the contrary, NT10-Sr allows formation of more andbigger focal adhesions that translate into more skeleton fibersanchored to them and eventually a thicker cell body. Sr incorpo-ration into biomaterials has been previously observed to promotespreading of osteoprecursor cells and osteoblasts [18,31,43]. Ourstudy definitively confirms the robust effect of Sr incorporation onMSC adhesion and spreading, while complete understanding of theunderlying mechanism still needs further studies.

3.6. Osteogenic differentiation

It has been reported that well spread MSCs are inclined toundergoing osteogenesis [41], and thus the effects of NT-Sr oninducing fast and good spreading of MSCs bode well for theosteogenic ability. The surrogate makers for osteogenic differen-tiation of MSCs cultured on the samples are studied in theabsence of OS. The gene expression level analysis results in Fig. 8show that NT10-Sr3 and NT40-Sr significantly promote theexpressions of osteogenesis related genes including the runt-related transcription factor 2 (RUNX2, a key transcript factor forbone formation), ALP (an early marker for osteogenic differ-entiaton), osteocalcin (OCN, a late marker for osteogenic differ-entiaton), type 1 collagen (Col-1, main content of bone ECM), andbone morphogenic protein 2 (BMP-2), demonstrating excellentostogenic activity. In contrast, NT10-Sr1 promotes the expressionof ALP, Col-1, and BMP-2 to a much lesser degree. The

Fig. 7. Fluorescence images showing: (A) morphology of GFP labeled MSCs after 8 days of culture and (B) cell migration on different samples.


topographical cue modulates the cell shape and focal adhesionand alters the indirect mechanotransduction, namely the integrindependent signal pathways such as MAPK and the direct one inwhich gene expressions originate from cell nuclei distorted bycytoskeletal tension [4,44,45]. Sr has been found to modulate theWnt/b-catenin and MAPK pathways in MSCs to initiate osteo-genic differentiation [19,20]. Although NT10-Sr3 and NT10-Sr1have similar topography and cell spreading area, there isa dramatic difference in the gene expression of the MSCs culturedon these two samples. NT10-Sr3 shows striking effects but NT10-

Sr1 displays a weak effect. The observation is in line with the Srrelease trend shown in Fig. 3. In addition, the NT40-Sr samplesthat exhibit relatively smaller cell spreading induce similar highgene expression than NT10-Sr3 as a result of the large Sr release.The present results indicate that the released Sr has a strongereffect on initiating MSC osteogenic differentiation than thetopography and cell area. Actually, the strong effect of Sr toinduce MSC osteogenic differentiation has been reported byChoudhary et al. [21] and previous studies indicate that theosteogenesis inducing effect of topography may be limited [4,46].

Fig. 8. Relative expressions of RUNX2, ALP, OCN, Col-1 and BMP-2 by MSCs cultured on different substrates for 2 weeks, all values normalized to GAPDH. *, **p< 0.05 or 0.01 vs Ti, #,##p < 0.05 or 0.01 vs NT10, and %, %% p < 0.05 or 0.01 vs NT40.


To further verify the Sr-induced enhanced osteogenic activity,the production of ALP and ECMmineralized nodules is investigated(Fig. 9). As early as 5 days during incubation, the cells alreadyproduce a certain amount of ALP, as shown in Fig. S5. After culturingfor 10 days, there are large nodular ALP positive areas on NT10 andNT10-Sr. The NT10-Sr samples induce more ALP than NT10 and Ticontrol, especially NT10-Sr3 although the amounts of ALP on NT40and NT40-Sr are much less. After incubation for 2 weeks, themineralized nodules are clearly observed from NT10 and biggerones appear on the NT10-Sr samples, but no discernible mineral-ized nodule is found on NT40, NT40-Sr, or flat Ti. It is quite unex-pected that the NT40-Sr samples do not show high ALP productionand fail to induce mineralized nodules. These results are differentfrom the gene expression analysis results and may be attributed tothe uneven protein distribution. Our results indicate that thenanotopography also plays an important role in the biologicalperformance of NT-Sr, even though Sr induces stronger osteogenicsignaling. The structure of NT10 allows high-quality focal adhesionformation to support cell spreading and induces big nodular ALPproducts and mineralized nodules on account of the suitablemultiple nanoscale structure. It can be further enhanced by Srrelease, especially NT10-Sr3. Kaur [47] has also demonstrated thesynergistic effects of nanotopography provided by tobacco mosaicvirus and phosphate and on osteogenic differentiation of MSCs. Onthe contrary, NT40, due to the protein nanopillar topography,triggers unfavorable signaling in cell functions, shows small ALPproduction, and fails to induce mineralized nodules. Further Srloading and release (NT40-Sr) cannot retrieve the cell functions. Onaccount of the uneven protein distribution, the cells on NT40-Srhave an abnormally thin body and the protein protrusions seemto break into the cell body thus impairing cell functions andpotentially inducing apoptosis. In brief, on NT40-Sr1 the proteinnanopillars generate unfavorable signaling and thus bad results,but on NT10-Sr3, the multiple nanoscale structure that facilitatesgood cell spreading and proper Sr release synergistically give rise togood osteogenic activity.

4. Conclusion

NT-Sr coatings with different tube diameters and Sr release rateshave been fabricated. Long-lasting and controllable Sr release isobserved from the NT-Sr samples with no cytotoxicity. In particular,sample NT10-Sr3 shows the best effects in inducing MSC osteo-genic differentiation due to the proper Sr release amount andsuitable multiple nanoscale structure with even protein adsorptionfavoring cell functions. The NT-Sr gives rise to good osteogenicactivity without the need to apply foreign complex biomolecules.The materials are easy to fabricate and have good stability thatfacilitating large-scale industrial production, storage, transport,sterilization and clinical use. They are very attractive to boneimplants and clinical applications.

Acknowledgments

This work was supported by National Natural Science Founda-tion of China Nos. 50902104 and 31200716, Hong Kong ResearchGrants Council (RGC) General Research Funds (GRF) No.112510, andthe Opening Project of State Key Laboratory of High PerformanceCeramics and Superfine Microstructure (SKL201109SIC). LingzhouZhao also thanks the grants from The Fourth Military MedicalUniversity and School of Stomatology, The Fourth Military MedicalUniversity.

Appendix A. Supplementary data

Supplementary data related to this article can be found athttp://dx.doi.org/10.1016/j.biomaterials.2012.09.041.

References

[1] Tasso R, Fais F, Reverberi D, Tortelli F, Cancedda R. The recruitment of twoconsecutive and different waves of host stem/progenitor cells during the

Fig. 9. (A) ALP product and (B) Extracellular matrix mineralized nodules generated by MSCs cultured on different samples.


development of tissue-engineered bone in a murine model. Biomaterials2010;31:2121e9.

[2] Oh S, Brammer KS, Li YSJ, Teng D, Engler AJ, Chien S, et al. Stem cell fatedictated solely by altered nanotube dimension. Proc Natl Acad Sci USA 2009;106:2130e5.

[3] Wang N, Li H, Lu W, Li J, Wang J, Zhang Z, et al. Effects of TiO2 nanotubes withdifferent diameters on gene expression and osseointegration of implants inminipigs. Biomaterials 2011;32:6900e11.

[4] Zhao L, Liu L, Wu Z, Zhang Y, Chu PK. Effects of micropitted/nanotubulartitania topographies on bone mesenchymal stem cell osteogenic differentia-tion. Biomaterials 2012;33:2629e41.

[5] Crawford GA, Chawla N, Das K, Bose S, Bandyopadhyay A. Microstructure anddeformation behavior of biocompatible TiO2 nanotubes on titanium substrate.Acta Biomater 2007;3:359e67.

[6] Tzaphlidou M. The role of collagen in bone structure: an image processingapproach. Micron 2005;36:593e601.

[7] Popat KC, Eltgroth M, La Tempa TJ, Grimes CA, Desai TA. Titania nanotubes:a novel platform for drug-eluting coatings for medical implants? Small 2007;3:1878e81.

[8] Xin Y, Jiang J, Huo K, Hu T, Chu PK. Bioactive SrTiO3 nanotube arrays: stron-tium delivery platform on Ti-based osteoporotic bone implants. ACS Nano2009;3:3228e34.

[9] Zhao L, Wang H, Huo K, Cui L, Zhang W, Ni H, et al. Antibacterial nano-structured titania coating incorporated with silver nanoparticles. Biomate-rials 2011;32:5706e16.

[10] Meunier PJ, Roux C, Seeman E, Ortolani S, Badurski JE, Spector TD, et al. Theeffects of strontium ranelate on the risk of vertebral fracture in women withpostmenopausal osteoporosis. N Engl J Med 2004;350:459e68.

[11] Seeman E, Vellas B, Benhamou C, Aquino JP, Semler J, Kaufman JM, et al.Strontium ranelate reduces the risk of vertebral and nonvertebral frac-tures in women eighty years of age and older. J Bone Miner Res 2006;21:1113e20.

[12] Ammann P, Shen V, Robin B, Mauras Y, Bonjour JP, Rizzoli R. Strontiumranelate improves bone resistance by increasing bone mass and improvingarchitecture in intact female rats. J Bone Miner Res 2004;19:2012e20.

[13] Marie PJ, Hott M, Modrowski D, De Pollak C, Guillemain J, Deloffre P, et al. Anuncoupling agent containing strontium prevents bone loss by depressingbone resorption and maintaining bone formation in estrogen-deficient rats.J Bone Miner Res 1993;8:607e15.

[14] Bonnelye E, Chabadel A, Saltel F, Jurdic P. Dual effect of strontium ranelate:stimulation of osteoblast differentiation and inhibition of osteoclast formationand resorption in vitro. Bone 2008;42:129e38.

[15] Marie PJ. Strontium ranelate: a dual mode of action rebalancing bone turnoverin favour of bone formation. Curr Opin Rheumatol 2006;18(Suppl. 1):S11e5.

[16] Braux J, Velard F, Guillaume C, Bouthors S, Jallot E, Nedelec JM, et al. A newinsight into the dissociating effect of strontium on bone resorption andformation. Acta Biomater 2011;7:2593e603.

[17] Coulombe J, Faure H, Robin B, Ruat M. In vitro effects of strontium ranelate onthe extracellular calcium-sensing receptor. Biochem Biophys Res Commun2004;323:1184e90.

[18] Capuccini C, Torricelli P, Sima F, Boanini E, Ristoscu C, Bracci B, et al. Stron-tium-substituted hydroxyapatite coatings synthesized by pulsed-laser depo-sition: in vitro osteoblast and osteoclast response. Acta Biomater 2008;4:1885e93.

[19] Yang F, Yang D, Tu J, Zheng Q, Cai L, Wang L. Strontium enhances osteogenicdifferentiation of mesenchymal stem cells and in vivo bone formation byactivating Wnt/catenin signaling. Stem Cells 2011;29:981e91.

[20] Peng S, Zhou G, Luk KD, Cheung KM, Li Z, Lam WM, et al. Strontium promotesosteogenic differentiation of mesenchymal stem cells through the Ras/MAPKsignaling pathway. Cell Physiol Biochem 2009;23:165e74.

[21] Choudhary S, Halbout P, Alander C, Raisz L, Pilbeam C. Strontium ranelatepromotes osteoblastic differentiation and mineralization of murine bonemarrow stromal cells: involvement of prostaglandins. J Bone Miner Res 2007;22:1002e10.

[22] Fournier C, Perrier A, Thomas M, Laroche N, Dumas V, Rattner A, et al.Reduction by strontium of the bone marrow adiposity in mice and repressionof the adipogenic commitment of multipotent C3H10T1/2 cells. Bone 2012;50:499e509.

[23] Peng S, Liu XS, Wang T, Li Z, Zhou G, Luk KD, et al. In vivo anabolic effect ofstrontium on trabecular bone was associated with increased osteoblasto-genesis of bone marrow stromal cells. J Orthop Res 2010;28:1208e14.

[24] Hurtel-Lemaire AS, Mentaverri R, Caudrillier A, Cournarie F, Wattel A, Kamel S,et al. The calcium-sensing receptor is involved in strontium ranelate-inducedosteoclast apoptosis. New insights into the associated signaling pathways.J Biol Chem 2009;284:575e84.

[25] Li Y, Feng G, Gao Y, Luo E, Liu X, Hu J. Strontium ranelate treatment enhanceshydroxyapatite-coated titanium screws fixation in osteoporotic rats. J OrthopRes 2010;28:578e82.

[26] Maimoun L, Brennan TC, Badoud I, Dubois-Ferriere V, Rizzoli R, Ammann P.Strontium ranelate improves implant osseointegration. Bone 2010;46:1436e41.

[27] Li Y, Li Q, Zhu S, Luo E, Li J, Feng G, et al. The effect of strontium-substitutedhydroxyapatite coating on implant fixation in ovariectomized rats. Biomate-rials 2010;31:9006e14.

[28] Zreiqat H, Ramaswamy Y, Wu C, Paschalidis A, Lu Z, James B, et al. Theincorporation of strontium and zinc into a calcium-silicon ceramic for bonetissue engineering. Biomaterials 2010;31:3175e84.

[29] Gentleman E, Fredholm YC, Jell G, Lotfibakhshaiesh N, O’Donnell MD, Hill RG,et al. The effects of strontium-substituted bioactive glasses on osteoblasts andosteoclasts in vitro. Biomaterials 2010;31:3949e56.

[30] Park JW, Kim HK, Kim YJ, Jang JH, Song H, Hanawa T. Osteoblast response andosseointegration of a Ti-6Al-4V alloy implant incorporating strontium. ActaBiomater 2010;6:2843e51.

[31] Zhang W, Shen Y, Pan H, Lin K, Liu X, Darvell BW, et al. Effects of strontium inmodified biomaterials. Acta Biomater 2011;7:800e8.

[32] Qiu K, Zhao XJ, Wan CX, Zhao CS, Chen YW. Effect of strontium ions on thegrowth of ROS17/2.8 cells on porous calcium polyphosphate scaffolds.Biomaterials 2006;27:1277e86.

[33] Zhao L, Mei S, Chu PK, Zhang Y, Wu Z. The influence of hierarchical hybridmicro/nano-textured titanium surface with titania nanotubes on osteoblastfunctions. Biomaterials 2010;31:5072e82.

[34] Zhao L, Mei S, Wang W, Chu PK, Wu Z, Zhang Y. The role of sterilization in thecytocompatibility of titania nanotubes. Biomaterials 2010;31:2055e63.

[35] Lord MS, Foss M, Besenbacher F. Influence of nanoscale surface topography onprotein adsorption and cellular response. Nano Today 2010;5:66e78.

[36] Anselme K. Osteoblast adhesion on biomaterials. Biomaterials 2000;21:667e81.[37] Zhao L, Mei S, WangW, Chu PK, Zhang Y, Wu Z. Suppressed primary osteoblast

functions on nanoporous titania surface. J BiomedMater Res A 2011;96:100e7.[38] Panzavolta S, Torricelli P, Sturba L, Bracci B, Giardino R, Bigi A. Setting prop-

erties and in vitro bioactivity of strontium-enriched gelatin-calcium phos-phate bone cements. J Biomed Mater Res A 2008;84:965e72.

[39] Stein GS, Lian JB, Owen TA. Relationship of cell growth to the regulation oftissue-specific gene expression during osteoblast differentiation. FASEB J1990;4:3111e23.

[40] Chung CJ, Long HY. Systematic strontium substitution in hydroxyapatitecoatings on titanium via micro-arc treatment and their osteoblast/osteoclastresponses. Acta Biomater 2011;7:4081e7.

[41] McBeath R, Pirone DM, Nelson CM, Bhadriraju K, Chen CS. Cell shape, cyto-skeletal tension, and RhoA regulate stem cell lineage commitment. Dev Cell2004;6:483e95.

[42] Zhao L, Hu L, Huo K, Zhang Y, Wu Z, Chu PK. Mechanism of cell repellence onquasi-aligned nanowire arrays on Ti alloy. Biomaterials 2010;31:8341e9.

[43] Xue W, Moore JL, Hosick HL, Bose S, Bandyopadhyay A, Lu WW, et al.Osteoprecursor cell response to strontium-containing hydroxyapatiteceramics. J Biomed Mater Res A 2006;79:804e14.

[44] Biggs MJ, Richards RG, Dalby MJ. Nanotopographical modification: a regulatorof cellular function through focal adhesions. Nanomedicine 2010;6:619e33.

[45] Biggs MJ, Richards RG, Gadegaard N, Wilkinson CD, Oreffo RO, Dalby MJ. Theuse of nanoscale topography to modulate the dynamics of adhesion formationin primary osteoblasts and ERK/MAPK signalling in STRO-1þ enriched skeletalstem cells. Biomaterials 2009;30:5094e103.

[46] Kilian KA, Bugarija B, Lahn BT, Mrksich M. Geometric cues for directing thedifferentiation of mesenchymal stem cells. Proc Natl Acad Sci USA 2010;107:4872e7.

[47] Kaur G, Wang C, Sun J, Wang Q. The synergistic effects of multivalent liganddisplay and nanotopography on osteogenic differentiation of rat bone marrowstem cells. Biomaterials 2010;31:5813e24.


the osteogenic activity of strontium loaded titania ... · the osteogenic activity of strontium...

Documents