This article appeared in a journal published by Elsevier. The attached

copy is furnished to the author for internal non-commercial research

and education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Effects of single wall carbon nanotubes and its functionalization with sodiumhyaluronate on bone repair

Renato M. Mendes a, Gerluza A.B. Silva a, Marcelo V. Caliari b, Edelma E. Silva c,Luiz Orlando Ladeira c, Anderson J. Ferreira a,!a Department of Morphology, Federal University of Minas Gerais, Belo Horizonte MG, 31.270-901, Brazilb Department of General Pathology, Federal University of Minas Gerais, Belo Horizonte MG, 31.270-901, Brazilc Department of Physics, Federal University of Minas Gerais, Belo Horizonte MG, 31.270-901, Brazil

a b s t r a c ta r t i c l e i n f o

Article history:Received 25 February 2010Accepted 11 June 2010

Keywords:Carbon nanotubesHyaluronanHealing processSocketsHyaluronic acidTooth

Aims: Sodium hyaluronate (HY) accelerates the repair of bone defects. However, the weak stability of HYformulations in aqueous environments has hindered its wide utilization. The functionalization of carbonnanotubes (SWCNT) with HY (HY–SWCNT) results in a reinforced hydrogel with an increased stability.Nevertheless, the biological effects of HY–SWCNT have not been explored. Thus, our objective was toevaluate whether this biomaterial preserves the bioactivity of the HY.Main methods: Wistar rats were subjected to molar extraction and the sockets were treated with SWCNT(50–400 !g/mL), 1% HY, HY–SWCNT (50–400 !g/mL) or carbopol (vehicle). After seven days of surgery,histological and morphometric analyses were performed to evaluate the trabecular bone formation and thenumber of cell nuclei in the sockets. Expression of collagen types I and III was determined byimmunohistochemistry.Key !ndings: Treatment with SWCNT did not alter the bone deposition, as well as the cell nuclei counting.Additionally, no signi!cant evidence of toxicity was observed in SWCNT-treated sockets. Contrastingly, bothHY and HY–SWCNT induced a marked increase in the bone formation (HY: 10.10±1.99%; HY–SWCNT100 !g/mL: 10.90±1.13%; control: 3.69±1.17%) and decreased the cell nuclei amount in the sockets.Moreover, collagen type I expression was more pronounced in HY- and HY–SWCNT-treated sockets. Nosigni!cant differences were viewed in the expression of collagen type III.Signi!cance: Our results indicate that SWCNT is a feasible material to deliver HY to bone defects. Importantly,the functionalization of SWCNT with HY preserved the bene!cial biological properties of HY in the healingprocess, thereby suggesting that HY–SWCNT scaffolds are potentially useful biomaterials for the restorationof bone defects.

© 2010 Elsevier Inc. All rights reserved.

Introduction

The development of scaffolds for structural support and sustainingand controlling cell growth is of great interest inmedicine and in bonetissue engineering (Abarrategi et al. 2008; Bhattacharyya et al. 2008;Sitharaman et al. 2008; Silva et al. 2009; Tutak et al. 2009). The idealscaffold should possess suitable stability, biocompatibility, bioperme-ability, and three-dimensional structure and promote bone tissuegrowth (Ch!opek et al. 2006; Sitharaman et al. 2008). In this aspect,carbon nanotubes (CNT)-based networks allow extensive cellulargrowth, spreading and adhesion of cells (Correa-Duarte et al. 2004;

Ch!opek et al. 2006; Galvan-Garcia et al. 2007; Matsumoto et al. 2007)and possess excellent biocompatibility (Ch!opek et al. 2006). It hasbeen demonstrated that CNT support the growth of osteoblastic cellsby stimulating the production of extracellular matrix (ECM), a centralstep during the formation of bone tissue (Tutak et al. 2009).Additionally, CNT improve the stability of synthetic and naturalpolymeric scaffold materials (Abarrategi et al. 2008; Bhattacharyyaet al. 2008; Sitharaman et al. 2008; Silva et al. 2009) and theinteraction of polymers with CNT enhances its capacity of inducingnucleation and growth of hydroxyapatite crystals (Silva et al. 2009),as well as osteogenesis (Sitharaman et al. 2008).

Hyaluronan (HA) or sodium hyaluronate (hyaluronic acid–HY) is apolysaccharide with high molecular weight composed by repeateddisaccharide units of D-glucoronic acid and N-acetyl-D-glucosamine(Moseley et al. 2002; Deschrevel et al. 2008). It is widely distributed inthe ECM of mammalian (!gren et al. 1997; Fraser et al. 1997; Dechertet al. 2006; Deschrevel et al. 2008; Kappler et al. 2009; Rügheimeret al. 2009), playing crucial roles in tissue repair duringwound healing

Life Sciences 87 (2010) 215–222

! Corresponding author. Department of Morphology, Av. Antônio Carlos, 6627, ICB,UFMG, 31.270-901, Belo Horizonte, MG, Brazil. Tel.: +55 31 3409 2811; fax: +55 3134092810.

E-mail address: anderson@icb.ufmg.br (A.J. Ferreira).

0024-3205/$ – see front matter © 2010 Elsevier Inc. All rights reserved.doi:10.1016/j.lfs.2010.06.010

Contents lists available at ScienceDirect

Life Sciences

j ourna l homepage: www.e lsev ie r.com/ locate / l i fesc ie

Author's personal copy

and in"ammation by stimulating migration, adhesion, proliferationand cell differentiation leading to bone formation (Grigolo et al. 2001;Lisignoli et al. 2002; Toole et al. 2002; Arosarena and Collins 2005;David-Raoudi et al. 2008; Pasquinelli et al. 2008). Thus, HA-basedscaffold might be potentially useful for bone tissue defect repairaccelerating the bone matrix formation and deposition (Pasquinelliet al. 2008). However, the weak stability of HA formulations inaqueous environments has hindered its wide use in bone rehabilita-tion of patients. This might be circumventing by the functionalizationof CNT with HA (HY–SWCNT). In fact, as a result of rheological tests,Bhattacharyya et al. (2008) found a considerable increase in theviscosity of the biomaterial due to the incorporation of SWCNT. Theviscoelastic property analysis revealed the formation of strong andrigid gels and an increase in the storage modulus for the hybridhydrogels compared to native HY, indicating that the functionaliza-tion of SWCNT with HY gel results in a reinforced hydrogel with anincreased stability. However, a missing point of this studywas that theauthors did not evaluate the biological properties of the HY–SWCNT.Thus, our objective in the current study was to evaluate whether thisbiomaterial preserves the bioactivity of the HY. To achieve this aim,we used an animal model of bone repair (tooth sockets) which wehave used before to demonstrate the ability of the HY alone toaccelerate the bone formation (Mendes et al. 2008).

Materials and methods

Carbon nanotube synthesis

SWCNT were prepared according to the arc discharge methodusing cobalt and nickel as catalysts in a helium atmosphere at totalpressure of 500 Torr (Trigueiro et al. 2007). The material was puri!edby a sequence of thermal oxidation and acid treatments. The puri!edSWCNT were re"uxed in 3 mol/L nitric acid in a domestic microwaveoven for 15 min in order to add carboxylic groups to the SWCNTstructure. Afterward, the material was centrifuged at 7000 rpm andrepeatedly washed with deionized water until complete removal ofthe nitric acid. The !nal solution, a carboxylated SWCNT, was dried for12 h at 60 °C.

Carbon nanotube characterization

Transmission electron microscopy (TEM) was performed to !rstcharacterize the SWCNT, as described previously (Ajayan 1999; Zhaoet al. 2005). TEM provides a rapid analysis of the CNT and allowsverifying the number and thickness of CNT walls. Additionally, Ramanspectra were obtained by using the 514-nm line of an argon laser anda triple-grating spectrometer (Jobin-Ivon, France). The energy regionof the so-called disorder (D) and graphite (G) bands were analyzed. Itis well known that the D band centered at approximately 1350 cm!1

is related to the presence of amorphous carbon or structural defectson the CNT (Zhou et al. 2001; Jorio et al. 2003; Arepalli et al. 2004),whereas the G band is originated of the E2g tangential mode ofgraphite which is also active in SWCNT and multi wall carbonnanotubes (MWCNT). Thus, the relative intensity between the D andG bands can be used as an indicative of the presence of amorphouscarbon and/or defects on CNT materials.

Sodium hyaluronate-functionalized carbon nanotubes

A solution containing carboxylated SWCNT (0.5 mg/mL) and HY(0.5 mg/mL) was prepared using an ultrasonic bath. This solution waswashed several times with deionized water and !ltered with aMillipore !lter. This process eliminated the HY and SWCNT excessesfrom the samples. The material remained over the !lter was dried andutilized for the experimental protocols. The success of the functiona-lization of SWCNT with HY (HY–SWCNT) was con!rmed by Fourier

transform infrared (FTIR) spectroscopy (Bhattacharyya et al. 2008).FTIR spectra were performedwith a Nexusmicroscope (magni!cation10!, probed region 150!150 !m2) connected to a Nicolet spectrom-eter in the mid-infrared region (650–4000 cm!1). CarboxylatedSWCNT and HY–SWCNT powder were mixed with carbopol, an inertgel, to prepare gels with different concentrations of SWCNT and HY–SWCNT, respectively.

Animals

Male Wistar rats (3 months old) weighting 260–300 g werehoused in a climate-controlled environment under a 12-h light/darkcycle with access ad libitum to rat chow and water. All experimentalprotocols were performed in accordance with the guidelines for thehumane use of laboratory animals established at our Institution.

Tooth extraction surgery

Rats were anaesthetized with a mixture of 10% ketamine and 2%xylazine (1:1, 0.1 mL/100 g body weight, i.m.) and subjected to theupper !rst molar (right and left) extraction. Sockets were randomlydivided and immediately treated with SWCNT [50 (n=8), 100(n=11), and 400 (n=6) !g/mL], 1% HY (NIKKOL, Galena, Brazil;n=9), HY–SWCNT [50 (n=6), 100 (n=7), and 400 (n=6) !g/mL]or carbopol (vehicle) ("0.1 mL; n=11). The variability of the numberof animals (samples) was due to the exclusion of some sockets whichpresented unde!ned limits or fragments of roots. The gel of HYcontained less than 2 ppm of arsenic and heavy metal, 0% of protein,5.2 mg/mL of glucuronic acid and 1% of hyaluronic acid (pH 6.1). HY,at a concentration of 1%, presented as a gel due to its high hygroscopiccapacity. Carbopol gel was used as a control (vehicle) because SWCNTand HY–SWCNT were dispersed in this gel. It was necessary becausetheywere supplied as a powder. Our previous results have shown thatcarbopol did not interfere on the socket healing after tooth extraction(Mendes et al. 2008). The gels were injected directly into the socketsusing a small needle (26 G). To maintain the gels in the extractionsites after the surgery, the rats were kept in the supine position forapproximately 4 h, period of duration of the anaesthesia. In addition,the animals received pasty chow for 2 days after the surgeries.

Histological and morphometric evaluation

After 7 days of tooth extraction, the animals were sacri!ced bydecapitation under anaesthesia with 10% ketamine and 2% xylazine(1:1, 0.1 mL/100 g body weight, i.m.). The sockets (upper maxillaries)were dissected and !xed in neutral 10% buffered formalin for 48 h atroom temperature. After !xation, the upper maxillaries weredemineralized in 10% ethylenediamine tetraacetic acid (EDTA) pH7.2–7.4, dehydrated through graded ethanol solutions, embedded inparaf!n and serially sectioned at 6 !m in the sagittal plane. Thesections were stained with Masson's trichrome for histological andmorphometric analyses. The sockets were anatomically divided inapical, medium and cervical thirds. Twelve images (40!) wereobtained from each third (apical and medium) of the distal root ofsockets with visible limits (total area of 6.4!105 !m2 for each third).The percentage of trabecular bone area in each third of the socket wascalculated dividing the trabecular bone area by the total area using theKS300 software (Carl Zeiss). This software was also used to count thenumber of cell nuclei in the total area of the apical and medium third,as described previously (Mendes et al. 2008).

Immunohistochemical analysis

Histological sections of sockets treated with SWCNT (100 !g/mL), 1%HY,HY–SWCNT(100 !g/mL)or carbopol (n=4per group)wereused forimmunohistochemical evaluation of collagen type I and III expression.

216 R.M. Mendes et al. / Life Sciences 87 (2010) 215–222

Author's personal copy

The sections were deparaf!nized with xylene and rehydrated through agraded series of ethanol. Then, they were incubated with 3% H2O2 toneutralize the endogenous peroxidase activity, washed with PBS andincubatedwith2%BSA inPBS for 1 h fornonspeci!c binding site blocking.Afterward, the sections were incubated overnight at 4 °Cwith one of thefollowing primary antibodies: rabbit polyclonal anti-collagen type I(1:50; Fitzgerald, 70R-CR007X) or rabbit polyclonal anti-collagen type III(1:50; Fitzgerald, 70R-CR014). After rinses in PBS, the sections wereincubated with biotinylated anti-rabbit antibody for 30 min at roomtemperature (Dako, Glostrup, Denmark). Following PBS rinses, thesections were incubated with streptavidin–horseradish peroxidaseconjugate for 30 min at room temperature, stained brown with asolution containing 350 !M3,3"-diaminobenzidine (Sigma Chemical Co.)and 1% H2O2 in PBS, and counterstained with hematoxylin. Negativecontrols were obtained by omission of primary antibodies. All sampleswere submitted to the immunoreactions at the same time to avoiddifferences among the assays. Eight images (40!) were obtained fromeach socket and collagen type I and III immunostaining were quanti!edusing the KS300 software.

Statistical analysis

All data were reported as mean±SEM. Statistical analysis wasperformed using one-way ANOVA followed by the Newman–Keulsmultiple comparison test. Pb0.05was considered statistically signi!cant.

Results

Carbon nanotube synthesis and characterization

TEM analysis revealed that CNT presented as bundles composed bydifferent numbers of CNT units (Fig. 1A). Further, it showed that eachCNT had approximately 1.25 nm thickness and was formed by singlewalls (Fig. 1B). The low relative intensity of the D band comparedwithG band of the Raman spectra indicated that there was a very lowamount of amorphous carbon and/or few structural defects on theSWCNT (Fig. 2A). In addition, the radial breathing mode (RBM)showed the presence of peaks with low energy spectra which are!ngerprints of SWCNT (Fig. 2B). The presence of bands at low energyspectra allowed estimating the diameter of the nanotubes inapproximately 1–1.3 nm. No signi!cant differences were observedbetween the SWCNT and HY–SWCNT. These results indicated that theCNT obtained by the arc discharge method presented single walls.

As expected for carboxylated SWCNT (COOH-SWCNT), the peak at1741 cm!1 of theFTIR spectrum is attributed to theC=Ostretchingof thecarboxylic group (COOH), the peak at 1595 cm!1 is due to themolecularC–C bond and the peak at 1234 cm!1 is related to the C–O stretching(Fig. 3). The HY spectrum showed peaks at 950 to 1200 cm!1

corresponding to the stretching of the C–OH; at 1160 cm!1 attributedto the asymmetric C–O–C stretching in glycoside group; at 1400 cm!1

related to the symmetric COO-stretching of the carboxylic group; at1500 to 1700 cm!1 attributed to the amide I and amide II; at 2500 to3600 cm!1 corresponding to the N–H stretching and to the O–Hstretching in hydrogen bond; and at 3100 cm!1 related to the N–Hstretching of the amide II (Fig. 3). Furthermore, the FTIR spectrumof theHY-SWCNT presented the following peaks: at 950 to 1200 cm!1

representing the stretching of the C–O; at 1160 cm!1 attributed tothe asymmetric C–O–C stretching in glycoside group; a contribution at1720 cm!1 disappeared; and at 3100 cm!1 related to the hydroxylgroup. Thus, the FTIR spectroscopy showed that SWCNT was functio-nalized with HY (Fig. 3).

Histology

After 7 days of tooth extraction, control sockets (carbopol-treatedsockets) were !lled with a dense conjunctive tissue. Newly formed

blood vessels and bone trabeculae were also observed in controlsockets (Fig. 4A and B). Sockets treated with SWCNT at differentconcentrations (50–400 !g/mL) did not show any signi!cant histo-logical difference when compared with control group (Fig. 4C and D—SWCNT 100 !g/mL). However, the deposition of bone matrix (bonetrabeculae) was markedly increased and the cell density wasdecreased in HY- and HY–SWCNT-treated sockets (HY: Fig. 4E andF; HY–SWCNT 100 !g/mL: 4G and 4H). No signi!cant differences wereobserved between HY- and HY–SWCNT-treated sockets, indicatingthat the functionalization of SWCNT with HY did not interfere withthe stimulatory effect of HY on bone healing. In sockets treated withHY or HY–SWCNT, the trabecular bone presented as an immaturetissue with large medullary spaces; however, it was more organizedthan control sockets.

Morphometry and immunohistochemistry

The treatment of socketswith SWCNTat different concentrations (50–400 !g/mL) did not change the amount of bone trabeculae deposition(SWCNT 100 !g/mL: 3.66±1.15% vs. control sockets: 3.69±1.17%,Fig. 5A), as well as the number of cell nuclei (SWCNT 100 !g/mL:22070±607 nuclei vs. control sockets: 22010±861 nuclei, Fig. 5B)when compared with control group. On the other hand, SWCNT

Fig. 1. Representative transmission electron micrographs (TEM) of SWCNT. (A) SWCNTwere organized in bundle. (B) One single CNT presented approximately 1.25 nmthickness. The arrow represents the opening of one single CNT.

217R.M. Mendes et al. / Life Sciences 87 (2010) 215–222

Author's personal copy

functionalized with HY markedly increased the percentage of bonetrabeculae (HY–SWCNT 100 !g/mL: 10.90±1.13% vs. control sock-ets: 3.69±1.17%, Fig. 5C) without any signi!cant effect on cell nuclei

counting (Fig. 5D). The effect of HY–SWCNT on bone deposition wascomparablewith that effect observedwhensocketswere treatedwithHYalone (HY–SWCNT 100 !g/mL: 10.90±1.13% vs. HY: 10.10±1.99%,Fig. 5E). Importantly, in contrast to HY, HY–SWCNT treatment causedonly a tendency to decrease the number of cell nuclei (Fig. 5F). However,because the timecourse of healing is different considering the anatomicalregion of the sockets (Teó!lo et al. 2004), we analyzed the cell nucleicountingand formationof bone trabeculae ineach third separately. Itwasobserved that both HY and HY–SWCNT decreased the number of cellnuclei only in the apical third and increased the percentage of bonetrabeculae in theapical andmediumthirds (Table 1). Since thenumber ofcell nuclei decreases with the progression of the healing process ofsockets (Mendes et al. 2008), these !ndings indicate that both HY andHY–SWCNT accelerate the bone repair.

Seven days after tooth extraction, both collagen types I and III wereexpressed in control and treated sockets. The quantitative analysisrevealed that the expression of collagen type I was 122% and 46%increased in HY and HY–SWCNT-treated sockets, respectively, whencompared with control animals (Fig. 6A). No signi!cant differences inthe expression of collagen type III were observed among any of thegroups (Fig. 6B). Additionally, SWCNT alone did not cause anysigni!cant change in the expression of collagen types I and III in ratsockets.

Discussion

In this study we synthesized SWCNT and functionalized thisnanomaterial with HY. Notably, this association preserved thebene!cial biological properties of HY on the healing process. In fact,treatment of sockets with HY–SWCNT signi!cantly increased theformation of bone trabeculae, decreased the amount of cells andincreased the expression of collagen type I in sockets after 7 days oftooth extraction.

The !rst aim of this study was to synthesize SWCNT andfunctionalize them with HY. The production of SWCNT was validatedby TEM and Raman spectroscopy and the presence of carboxylicgroups in the SWCNT structure was veri!ed by FTIR spectroscopy. Thefunctionalization of CNT with HY and its viability were reportedpreviously by Bhattacharyya et al. They showed that the interactionbetween CNT and HY resulted in a biomaterial with improveddynamic mechanical properties when compared with HY alone(Bhattacharyya et al. 2008).

If CNT are going to turn out into new biological applications, theirsafety must be assessed (Flahaut et al. 2006). Indeed, many studieshave been designed to evaluate the CNT toxicity. SWCNT did not causeany cytotoxicity when added to cultures of !broblasts (Galvan-Garciaet al. 2007; Shi et al. 2007), osteoblasts (Tutak et al. 2009) andendothelial cells (Flahaut et al. 2006). The absence of toxicity was alsoobserved in pharmacokinetics studies using SWCNT in mammalian(Cherukuri et al. 2006). In keeping with these data, our !ndings didnot reveal any signi!cant evidence of toxicity in sockets treated withSWCNT. In addition, this nanomaterial did not in"uence the healingprocess at 7 days of tooth extraction, indicating that SWCNT is an inertmaterial. However, it has been demonstrated that CNT are able toinduce nucleation and growth of hydroxyapatite crystals underphysiological concentrations of calcium and phosphates in vitro(Zhao et al. 2005).

In a previous study, we found that HY accelerates the healing processin tooth sockets of rats by stimulating the early bone trabeculaedeposition and expression of osteogenic proteins, i.e. osteopontin andbone morphogenetic protein 2 (Mendes et al. 2008). Now, we observedthat HY–SWCNT induced similar effects on sockets healing processwithout any further increase in the trabeculae bone deposition whencompared with HY alone. Thus, our present results showed that HY–SWCNT preserve the bene!cial biological properties of HY in the healingprocess. This is of extreme importance because this association (HY–

Fig. 2. Representative Raman spectra of SWCNT (A) at D and G bands and (B) at RBMmode. It showed that the carbon nanotubes presented single walls. COOH:carboxylated.

Fig. 3. Representative Fourier transform infrared (FTIR) spectra of COOH-SWCNT, HYand HY–SWCNT. The different pro!les observed indicated that SWCNT werefunctionalized with HY, thereby forming a distinct biomaterial. COOH: carboxylated.

218 R.M. Mendes et al. / Life Sciences 87 (2010) 215–222

Author's personal copy

Fig. 4. Representative micrographs of histological sections of sockets 7 days after tooth extraction. (A) Carbopol-treated sockets (control— lowmagni!cation); (B) carbopol-treatedsockets (control — high magni!cation); (C) SWCNT 100 !g/mL-treated sockets (low magni!cation); (D) SWCNT 100 !g/mL-treated sockets (high magni!cation); (E) HY-treatedsockets (low magni!cation); (F) HY-treated sockets (high magni!cation); (G) HY–SWCNT 100 !g/mL-treated sockets (low magni!cation); and (H) HY–SWCNT 100 !g/mL-treatedsockets (high magni!cation). Sockets treated with HY or HY–SWCNT presented an increased trabecular bone formation when compared with control sockets (arrows). S: boneseptum.

219R.M. Mendes et al. / Life Sciences 87 (2010) 215–222

Author's personal copy

SWCNT) can allow the administration of HY using different pharmaco-logical formulations. Consequently, this increases the stability of the HY,especially in aqueous environments such as themouth, and allows its usein large bone defects. Surprisingly, HY–SWCNT at 400 !g/mL failed toaccelerate the bone formation in ourmodel. One possible explanation for

this is that high concentrations of HY–SWCNT act as a physical barrierimpairing the healing process. Nevertheless, further experiments areneeded to con!rm or not this possibility.

During the bone healing process, the collagen !bril formation isinitiated when collagen !laments released by osteoblasts assemble

Fig. 5. Effects of SWCNT,HY–SWCNTandHYonhealingprocess of sockets 7 daysafter tooth extraction. (A, C, E)Quanti!cationof thepercentage of bone trabeculae and (B,D, F) cellular nucleicounting in sockets of rats. HY–SWCNT at 100 !g/mL and HY induced an increase in the trabeculae bone formation when compared with control sockets. However, only the HY treatmentdecreased the number of cell nuclei. Data are shown as mean±S.E.M. *pb0.05 vs. carbopol.

Table 1Percentage of bone trabeculae and cell nuclei counting in the apical and medium thirds.

Apical third Medium third

Trabeculae (%) Cell nuclei (!103) Trabeculae (%) Cell nuclei (!103)

Carbopol (n=11) 5.92±1.90 11570±467 1.70±0.6 10440±558SWCNT (n=11) 6.62±2.18 11410±314 0.51±0.32 10660±361HY (n=9) 17.07±3.48a 9406±453a 3.26±0.74a 9186±627HY–SWCNT (n=7) 18.16±1.92a 9404±723a 3.71±0.44a 10610±828

Data are shown asmean±SEM. apb0.05 vs. carbopol. Carbopol (control); SWCNT (single wall carbon nanotubes)— 100 !g/mL; HY (sodium hyaluronate)— 1%; HY–SWCNT (SWCNTfunctionalized with HY) — 100 !g/mL.

220 R.M. Mendes et al. / Life Sciences 87 (2010) 215–222

Author's personal copy

Fig. 6. Representative photomicrographs of immunostaining for (A) collagen type I and(B) collagen type III 7 daysafter tooth extraction. (a) Carbopol-treated sockets; (b)HY-treated sockets;(c) SWCNT100 !g/mL-treated sockets; (d)HY–SWCNT100 !g/mL-treated sockets; (e) negative control. Collagen type I expressionwas increased in sockets treatedwithHY andHY–SWCNT.SWCNT sockets presented similar expression of collagen type I when compared with control group. No signi!cant changes in collagen type III expression were observed among any of thegroups.

221R.M. Mendes et al. / Life Sciences 87 (2010) 215–222

Author's personal copy

extracellularly into striated !brils to form the osteoide (Sodek andMcKee 2000). This immature tissue allows the deposition of crystalsof carbonated hydroxyapatite responsible by mineralization of thebone matrix (Sodek and McKee 2000; Bouletreau et al. 2002). Theamount of collagen type I increases in the osteoide while theexpression of collagen type III diminishes progressively. Thus, anaugmented expression of collagen type I in HY and HY–SWCNT-treated sockets further suggests that the healing process is acceleratedin these groups.

Conclusions

Our results demonstrated that the functionalization of SWCNTwith HY is a feasible strategy to improve the physical properties of theHY with preservation of its biological effects. Thus, the observedactions of HY–SWCNT are probably due to the HY effects. Overall,these !ndings suggest that HY–SWCNT scaffolds are potentially usefulbiomaterials for the restoration of bone defects.

Con!ict of interest statement

The authors declare that there are no con"icts of interest.

Acknowledgements

This study was supported in part by CNPq (Conselho Nacional deDesenvolvimento Cientí!co e Tecnológico), CAPES (Coordenação deAperfeiçoamento de Pessoal de Nível Superior) and FAPEMIG(Fundação de Amparo à Pesquisa do Estado de Minas Gerais). Theauthors RMM, GABS, LOL and AJF declare that they have deposited apatent covering the use of HY–SWCNT as a biomaterial for bonerepair. The data presented is part of the Ph.D. thesis of RMM.

References

Abarrategi A, Gutiérrez MC, Moreno-Vicente C, Hortigüela MJ, Ramos V, López-LacombaJL, Ferrer ML, del Monte F. Multiwall carbon nanotube scaffolds for tissueengineering purposes. Biomaterials 29, 94–102, 2008.

!gren UM, Tammi RH, Tammi MI. Reactive oxygen species contribute to epidermalhyaluronan catabolism in human skin organ culture. Free Radical Biology &Medicine 23, 996–1001, 1997.

Ajayan PM. Nanotubes from carbon. Chemical Reviews 99, 1787–1799, 1999.Arepalli S, Nikolaev P, Gorelik O, Hadjiev VG. Protocol for the characterization of single-

wall carbon nanotube material quality. Carbon 42, 1783–1791, 2004.ArosarenaOA, CollinsWL. Bone regeneration in the ratmandible with bonemorphogenetic

protein-2: a comparison of two carriers. Journal of Otolaryngology — Head & NeckSurgery 132, 592–597, 2005.

Bhattacharyya S, Guillot S, Dabboue H, Tranchant J-F, Salvetat J-P. Carbon nanotubes asstructural nano!bers for hyaluronic acid hydrogel scaffolds. Biomacromolecules 9,505–509, 2008.

Bouletreau PJ, Warren SM, Longaker MT. The molecular biology of distractionosteogenesis. Journal of Craniomaxillofacial Surgery 30, 1–11, 2002.

Cherukuri P, Gannon CJ, Leeuw TK, Schmidt HK, Smalley RE, Curley SA, Weisman RB.Mammalian pharmacokinetics of carbon nanotubes using intrinsic near-infrared"uorescence. Proceedings of the National Academy of Sciences of the United Statesof America 103, 18882–18886, 2006.

Ch!opek J, Czajkowska B, Szaraniec B, Frackowiak E, Szostak K, Béguin F. In vitro studiesof carbon nanotubes biocompatibility. Carbon 44, 1106–1111, 2006.

Correa-Duarte MA, Wagner N, Rojas-Chapana J, Morsczeck C, Thie M, Giersig M.Fabrication and biocompatibility of carbon nanotube-based 3D networks asscaffolds for cell seeding and growth. Nano Letters 4, 2233–2236, 2004.

David-Raoudi M, Tranchepain F, Deschrevel B, Vincent J-C, Bogdanowicz P, Boumediene K,Pujol J-P. Differential effects of hyaluronan and its fragments on !broblasts: relation towound healing. Wound Repair and Regeneration 16, 274–287, 2008.

Dechert TA, Ducale AE, Ward SI, Yager DR. Hyaluronan in human acute and chronicdermal wounds. Wound Repair and Regeneration 14, 252–258, 2006.

Deschrevel B, Tranchepain F, Vincent J-C. Chain-length dependence of the kinetics ofthe hyaluronan hydrolysis catalyzed by bovine testicular hyaluronidase. Matrix 27,475–486, 2008.

Flahaut E, Durrieu MC, Remy-Zolghadri M, Bareille R, Baquey Ch. Investigation of thecytotoxicity of CCVD carbon nanotubes towards human umbilical vein endothelialcells. Carbon 44, 1093–1099, 2006.

Fraser JRE, Laurent TC, Laurent UBG. Hyaluronan: its nature, distribution, functions andturnover. Journal of Internal Medicine 242, 27–33, 1997.

Galvan-Garcia P, Keefer EW, Yang F, Zhang M, Fang S, Zakhidov AA, Baughman RH,Romero MI. Robust cell migration and neuronal growth on pristine carbonnanotube sheets and yarns. Journal of Biomaterials Science, Polymer Edition 18,1245–1261, 2007.

Grigolo B, Roseti L, Fiorini M, Fini M, Giavaresi G, Aldini NN, Giardino R, Facchini A.Transplantation of chondrocytes seeded on a hyaluronan derivative (Hyaff®-11)into cartilage defects in rabbits. Biomaterials 22, 2417–2424, 2001.

Jorio A, Pimenta MA, Souza AG, Saito R, Dresselhaus G, Dresselhaus MS. Characterizingcarbon nanotube samples with resonance Raman scattering. New Journal of Physics5, 139, 2003.

Kappler J, Hegener O, Baader SL, Franken S, Gieselmann V, Häberlein H, Rauch U.Transport of a hyaluronan-binding protein in brain tissue. Matrix 28, 396–405,2009.

Lisignoli G, Fini M, Giavaresi G, Aldini NN, Toneguzzi S, Facchini A. Osteogenesis of largesegmental radius defects enhanced by basic !broblast growth factor activated bonemarrow stromal cells grown on non-woven hyaluronic acid-based polymerscaffold. Biomaterials 23, 1043–1051, 2002.

Matsumoto K, Sato C, Naka Y, Kitazawa A, Whitby RLD, Shimizu N. Neurite outgrowthsof neurons with neurotrophin-coated carbon nanotubes. Journal of Bioscience andBioengineering 103, 216–220, 2007.

Mendes RM, Silva GA, Lima MF, Caliari MV, Almeida AP, Alves JB, Ferreira AJ. Sodiumhyaluronate accelerates the healing process in tooth sockets of rats. Archives of OralBiology 53, 1155–1162, 2008.

Moseley R, Leaver M, Walker M, Waddington RJ, Parsons D, Chen WYJ, Embery G.Comparison of the antioxidant properties of HYAFF®-11p75, AQUACEL® andhyaluronan towards reactive oxygen species in vitro. Biomaterials 23, 2255–2264,2002.

Pasquinelli G, Orrico C, Foroni L, Bonafè F, Carboni M, Guarnieri C, Raimondo S, Penna C,Geuna S, Pagliaro P, Freyrie A, Stella A, Caldarera CM,Muscari C. Mesenchymal stemcell interaction with a non-woven hyaluronan-based scaffold suitable for tissuerepair. Journal of Anatomy 213, 520–530, 2008.

Rügheimer L, Olerud J, Johnsson C, Takahashi T, Shimizu K, Hansell P. Hyaluronansynthases and hyaluronidases in the kidney during changes in hydration status.Matrix 28, 390–395, 2009.

Shi X, Sitharaman B, Pham QP, Spicer PP, Hudson JL, Wilson LJ, Tour JM, Raphael RM,Mikos AG. In vitro cytotoxicity of single-walled carbon nanotube/biodegradablepolymer nanocomposites. Journal of Biomedical Materials Research 86, 813–823,2007.

Silva EE, Colleta HHMD, Ferlauto AS, Moreira RL, Resende RR, Oliveira S, Kitten GT,Lacerda RG, Ladeira LO. Nanostructured 3-d collagen/nanotube biocomposites forfuture bone regeneration scaffolds. Nano Research 2, 462–473, 2009.

Sitharaman B, Shi X, Walboomers XF, Liao H, Cuijpers V, Wilson LJ, Mikos AG, Jansen JA.In vivo biocompatibility of ultra-short single-walled carbon nanotube/biodegrad-able polymer nanocomposites for bone tissue engineering. Bone 43, 362–370, 2008.

Sodek J, McKee MD. Molecular and cellular biology of alveolar bone. Periodontology 24,99–126, 2000.

Teó!lo JM, Brentegani LG, Lamano Carvalho TL. Bone healing in osteoporotic female ratsfollowing intra-alveolar grafting of bioactive glass. Archives of Oral Biology 49,755–762, 2004.

Toole BP, Wight TN, Tammi MI. Hyaluronan–cell interactions in cancer and vasculardisease — minireview. The Journal of Biological Chemistry 277, 4593–4596, 2002.

Trigueiro JPC, Silva GG, Lavall RL, Furtado CA, Oliveira S, Ferlauto AS, Lacerda RG, LadeiraLO, Liu JW, Frost RL, George GA. Purity evaluation of carbon nanotube materials bythermogravimetric, TEM, and SEM methods. Journal of Nanoscience and Nano-technology 7, 3477–3486, 2007.

Tutak W, Park KH, Vasilov A, Starovoytov V, Fanchini G, Cai S-Q, Partridge NC, Sesti F,Chhowalla M. Toxicity induced enhanced extracellular matrix production inosteoblastic cells cultured on single-walled carbon nanotube networks. Nanotech-nology 20, 255101, 2009.

Zhao B, Hu H, Mandal SK, Haddon RC. A bone mimic based on the self-assembly ofhydroxyapatite on chemically functionalized single-walled carbon nanotubes.Chemistry of Materials 17, 3235–3241, 2005.

Zhou W, Ooi YH, Russo R, Fischer JE, Papanek P. Structural characterization anddiameter-dependent oxidative stability of single wall carbon nanotubes synthe-sized by the catalytic decomposition of CO. Chemical Physics Letters 350, 6–14,2001.

222 R.M. Mendes et al. / Life Sciences 87 (2010) 215–222