Using single-walled carbon nanotubes nonwoven films as scaffolds to enhance long-term cell proliferation in vitro

Using single-walled carbon nanotubes nonwoven films asscaffolds to enhance long-term cell proliferation in vitro

Jie Meng,1 Li Song,2 Jie Meng,3 Hua Kong,3 Guangjin Zhu,1 Chaoying Wang,2 Lianghua Xu,4

Sishen Xie,2 Haiyan Xu3

1Department of Pathophysiology, Institute of Basic Medicine, Chinese Academy of Medical Sciencesand Peking Union Medical College, Beijing 1000052Institute of Physics, Chinese Academy of Sciences, Beijing 100080, People’s Republic of China3Department of Biomedical Engineering, Institute of Basic Medicine, Chinese Academy of Medical Sciencesand Peking Union Medical College, Beijing 100005, People’s Republic of China4Department of Carbon and Composite, Beijing University of Chemical Technology, Beijing 100029,People’s Republic of China

Received 25 November 2005; accepted 28 February 2006Published online 30 June 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.30787

Abstract: Carbon nanotubes have attracted intensive inter-ests in biomedical research in recent years. In this study, anovel type of carbon nanotubes material so called nonwo-ven single-walled carbon nanotubes (SWNTs) with nanoto-pographic structure and macroscopic volume was used ascell growing scaffold. The morphology and surface chemis-try of nonwoven SWNTs were observed and characterizedthrough scanning electron microscopy and X-ray photoelec-tron spectroscopy, respectively. The cells were cultivated innonwoven SWNTs and in other types of substrate as con-trol. The cells growth behaviors including adhesion, prolifer-ation, and cytoskeletal development was investigated byusing cell viability assay and confocal observation. The ex-perimental results indicated that nonwoven SWNTs exhib-ited significant enhancement to the cells adhesion and pro-

liferation in at least 3 weeks. Numerous and highly organ-ized cytoskeletal structures were observed when the cellswere cultured in nonwoven SWNTs. Furthermore, anobvious promotional influence of the cells cultivated in non-woven SWNTs scaffold upon the proliferation of thosegrowing in the other kind of substrate through cell–cellcommunication had been found. The results obtained in thiswork are of significance to in vitro cell amplification in largescale, tissue regeneration, or guided repair, as well as bio-medical device application. � 2006 Wiley Periodicals, Inc.J Biomed Mater Res 79A: 298–306, 2006

Key words: single-walled carbon nanotubes; scaffold; cellgrowth; cell–cell communication; tissue regeneration

INTRODUCTION

Extracellular matrix (ECM) plays a crucial role incell–substrate adhesion, proliferation, and migration.Engineering scaffolds, providing artificial ECM forcells growing in vitro, are of strategic significance toregenerative therapeutic, novel drug development,and cell biology research. Materials for artificial 3Dscaffolds used in most previous work have been bio-degradable synthetic polymers such as poly(L-lacticacid), poly(glycolic acid), or copolymers of lactic

acid and glycolic acid, as well as biopolymers suchas collagen, fibroin, and peptide. Scaffolds composedof polymeric materials are typically fabricated po-rous objects, fabrics, or films. As several micro/nanofabrication techniques recently developed canproduce a wide range of nanostructures, includingnanoimprint lithography,1,2 electrospinning,3,4 anddimix microphase separation,5,6 latest studies oncell–scaffold interactions give evidences that demon-strated that some types of nanotopography of scaf-folds appear to be an important factor that has strongeffects on cell behavior depending on the topographicfeatures.

In the previous studies,7 we reported a novel ma-terial with unique nanotopographic feature and purecarbon atom composition, single-walled carbonnanotubes (SWNTs) nonwoven film, which is a kindof flexible fabric composed of thousands of highly

Correspondence to: H. Xu; e-mail: [email protected] grant sponsor: National Natural Science Founda-

tion of China; contract grant numbers: 30270394 and 90306004Contract grant sponsor: National Center for Nanoscience

and Technology of China

' 2006 Wiley Periodicals, Inc.

entangled SWNT bundles. Compared with conven-tional carbon nanotube particulates, nonwovenSWNTs has many attractive characteristics of bothstructure and chemistry, which may have promisingpotentials of therapeutic strategies and biomedicaldevice applications, including macroscopic nanoto-pographic surface up to tens of square centimeters,pure carbon atom composition, high conductivity,extraordinary strength, and stiffness. It was at thispoint, we took up the investigation of using nonwo-ven SWNTs as a scaffold for cell culture, and themorphology of nonwoven SWNTs was observedthrough scanning electron microscopy (SEM); theinteractions of SWNTs to the protein molecules con-tained in the cell culture were investigated with X-ray photoelectron spectroscopy (XPS) and SEM aswell; the cell growth behavior on nonwoven SWNTsscaffold was studied by cytoskeletal observationsand cell viability assay, respectively, for a long termof 3 weeks, and compared with those on other bio-compatible materials. More importantly, the promo-tional influence of the cells growing in nonwovenSWNTs scaffolds upon the cells cultivated on theother substrates through cell–cell communicationwas found by using a modified transwell device. Theexperimental results indicated that nonwoven SWNTsexhibited excellent functions of enhancing long-termcell proliferation in vitro, which is of significance to cellamplification in large scale and in long term; addition-ally, the interactions of nonwoven SWNTs to the cellsliving on it could enhance the proliferation of the cellsgrowing on the other materials in a manner of cell–cellcommunication that had implications to tissue regen-eration and repairing.

MATERIALS AND METHODS

Preparation of nonwoven SWNTs

Nonwoven SWNTs was prepared directly by the float-ing chemical vapor deposition method, then oxidized inair at 400–5008C for 48 h, and subsequently treated withconcentrated HCl to remove impurities.7

SEM observation and XPS analysis for proteinadsorption on nonwoven SWNTs

A purified large piece of nonwoven SWNTs was cutinto small pieces with an average area of 5 mm � 5 mm.The pieces of nonwoven SWNTs were incubated at 378Cin Iscove’s modified Dulbecco’s medium (IMDM, Invitro-gen) supplemented with 10% of newborn bovine serum(TBD, Tianjin China), 1% of penicillin, and streptomycin at378C in a humidified incubator with 5% CO2 for 6 h,respectively. The samples were then rinsed thoroughly 5times by phosphate-buffered saline (PBS, pH ¼ 7.4) to

remove unadsorbed proteins. After being freezing dried at–258C for 12 h, the resulting samples were spread on cup-per grid to observe with SEM (Hitachi S-5200), and spreadon a cupper plate for XPS (Thermo ESCALAB 250, UK)analysis.

Preparation of substrates for cell culture

In this work, we used four kinds of materials as cell cul-ture substrates that are as follows: (1) blank cell cultureplate without any adhesion cue, (2) polyether polyur-ethane films, (3) carbon fibers (CF), and (4) nonwovenSWNTs. The CF and nonwoven SWNTs were spread andadhered on the polyurethane substrate films respectivelyprior to cell culture.

Medical grade segmented polyether polyurethane Jm80(PU) was purchased from the Manufacturer of Polyur-ethane Products (Tianjin, China) and used without furtherpurification. PU films were prepared by casting the solu-tion of 5% PU in tetrahydrofuran, followed drying in airand in vacuum oven at 508C for 48 h, respectively, toremove the left solvent completely. The film thickness is�200 lm. CF with diameter of 7 lm were prepared by Bei-jing University of Chemical Technology. One strand of CFcontained 1000 fibers. In the experiments, the strand wasscrubbed with hand to make the fibers separate, followedadhering them to PU substrate films. The nonwovenSWNTs were also spread and adhered to PU substratefilm. All the films were trimmed carefully to fit the size ofwell in cell culture plate.

Cell culture and viability evaluation

Cell culture

3T3-L1 mouse fibroblasts were obtained from the Centerfor Cell Culture of Institute of Basic Medicine, ChineseAcademy of Medical Sciences, and maintained in IMDMsupplemented with 10% of newborn bovine serum (TBD,Tianjin China), 1% of penicillin, and streptomycin at 378Cin a humidified incubator with 5% CO2. The various filmsprepared above were set in the wells of cell culture platesand were irradiated with ultraviolet light for 1 h prior tothe cells incubation. Freshly confluent flasks of 3T3-L1mouse fibroblasts were incubated with Trypsin/EDTA(0.125% trypsin in 0.01% EDTA in Ca2þ/Mg2þ-free PBS)for 1 min and then resuspended in the culture medium.

Cell adhesion assay

The measurement of 3T3-L1 cells adhesion was per-formed 1, 3, 5, and 6 h after the cells were seeded on tothe various substrates at a density of 3 � 104 cells/cm2.All the samples were placed in a humidified incubatorwith 5% CO2 at 378C. The cells cultivated to the desig-nated time were rinsed with PBS (pH ¼ 7.4) twice toremove unadhered ones, and the numbers of rest cellswere determined using CellTiter 96 AQueous One Solu-

NONWOVEN SWNT FILMS AS SCAFFOLDS TO ENHANCE LONG-TERM CELL PROLIFERATION 299

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

tion Cell Proliferation Assay (MTS assay, Promega). TheMTS assay is a colorimetric method for determining thenumber of viable cell. The MTS reagent contains a tetra-zolium compound that is bioreduced presumably byNADPH or NADH produced by dehydrogenase enzymesin metabolicallyactive cells. When measuring, the adhered3T3-L1 cells were incubated for 1.5 h with aliquots ofmedia with MTS reagent and transferred to a 96-wellplate. The absorbance was measured at 490 nm (UV-2450,Shimadzu) and the values obtained were transferred tocell numbers using a calibration curve established accord-ing to the instruction of MTS assay.

Cell proliferation assay

3T3-L1 cells were seeded at a density of 12,000 cells/cm2 onto the various substrates set in the cell culture plateof 96-well and placed in a humidified incubator with 5%CO2 at 378C. The medium was exchanged once per twodays. The cells were allowed to proliferate for 21 days,and at each designated time the number of attached cellswas again determined by MTS assay as described earlier.

Cell morphology and actin observation

To learn the cell morphology changes on nonwovenSWNTs during the growth period, 3T3-L1 cells wereseeded at a density of 4000 cells/cm2 onto nonwovenSWNTs and placed in a humidified incubator with 5%CO2 at 378C. The cells growing on the substrates wereobserved timely with microscope (Nikon TE2000-U) andtook pictures during 3 weeks.

For the actin observation, 3T3-L1 cells were seeded at adensity of 10,000 cells/cm2 onto nonwoven SWNTs in 6-well culture plate and placed in a humidified incubatorwith 5% CO2 at 378C. After 6 h, 24 h, and 3 days of cul-ture, the films were rinsed with PBS (pH ¼ 7.4) twice andfixed in 4% formaldehyde/PBS, with 1% sucrose at 378Cfor 15 min. The samples were washed with PBS (pH ¼7.4), and a permeabilizing buffer (10.3 g sucrose, 0.292 gNaCl, 0.06 g MgCl2, 0.476 g HEPES buffer, 0.5 mL TritonX, in 100 mL water, pH ¼ 7.2) was added at 48C for5 min. The samples were then incubated at 378C for 5 minin 1% BSA/PBS, followed by the addition of 10 lg/mLFITC-conjugated phalloidin (Sigma) for 40 min at 378C,and given a final wash. The Samples were then viewedwith fluorescence microscope (Nikon TE2000-U) and laserscanning confocal microscopy (Confocal Vokogawa CSU10),respectively.

Modified transwell culture assay

A modified transwell device was used to examine theinfluence of the cells growing on nonwoven SWNTs uponthe cells growing on polyurethane control. Transwell is adevice for cells culture (Costar). It has two wells separate,bottom well and upper well. The area of upper well isabout 1/6 of that of bottom well. There are pores on thesidewall of the upper well so that the cells growing in the

upper and bottom well can exchange substance throughthe culture medium, but the two cell populations are notable to contact directly. In our experiments, some trans-well devices were designed to have PU films both in thebottom wells and in the upper wells, the other transwelldevices were designed to have PU films in the bottomwells and nonwoven SWNTs in the upper wells. 3T3-L1cells were seeded at a density of 18,000 cells/cm2 eachwell. The amount of culture medium in each transwell de-vice was given enough to make sure the culture solutionsin upper and in bottom wells were connected. After culti-vated for 8 days, the viable cells number of attached ineach well were determined by MTS assay, respectively, asdescribed earlier.

Statistical analysis

All attachment and proliferation measurements were col-lected in triplicate and expressed as mean 6 standard de-viation (SD). Single-factor analysis of variance (ANOVA)was employed to assess statistical significance of the results.Difference was considered statistically significant at p < 0.05.

RESULTS

Characterization of nonwoven SWNTs beforeand after treated with cell culture medium

Figure 1(a) showed the optical graph of nonwovenSWNTs, and it could be seen that nonwoven SWNTswas a black membrane with macroscopic area. Thearea of the nonwoven SWNTs could reach to severaltens of square centimeters. Observed via SEM [Fig.(1b,c)], the nonwoven SWNTs with compact struc-ture was composed of thousands of highly en-tangled SWNTs bundles with diameters of 20–30 nmand several hundred micrometers in length. The av-erage size of pores formed by the entangled SWNTbundles ranged from 50 to 200 nm. The thickness ofthe nonwoven SWNTs observed was about severalmicrometers.

Figure 2 gave the morphology of nonwovenSWNTs after immersion in the cell culture mediumfor 6 h. It could be seen from the image that theedges of carbon nanotubes became a little obscurecompared with those of bare SWNTs shown in Fig-ure 1. It was implied that protein adsorptionoccurred when nonwoven SWNTs was exposed tothe culture medium.

The surface chemistry of nonwoven SWNTs afterimmersion in the medium for 6 h was analyzed byXPS. Figure 3(a,b) gave the survey scan of bare non-woven SWNTs and the one of nonwoven SWNTsafter immersion in the medium for 6 h, respectively.In Figure 3(a), there was one major strong peak cor-responding to C1s (binding energy 284.72 eV), and a

300 MENG ET AL.


very weak peak corresponding to O1s (bindingenergy 533.48 eV). The scan spectrum in Figure 3(b)exhibited the peaks corresponding to C1s (bindingenergy 286.95 eV), N1s (binding energy 400.07 eV),and O1s (binding energy 532.15 eV), respectively. ItsC1s spectrum was resolved into six characteristicpeaks in Figure 3(d), in which the binding energiesof 284.56, 285.35, 286.25, 287.03, 288.35, and 289.55eV were attributed by C��C, C��O, C¼¼O, O��C��N, O¼¼C��O, and N��COO, respectively. Figure3(c) showed the C1s spectrum of nonwoven SWNTsnot treated with the medium, exhibiting a singlecharacteristic peak. Comparing the XPS spectra ofmedium-treated and untreated nonwoven SWNTs,we could infer that the nitrogen content detected inthe surface of nonwoven SWNTs treated with cul-ture medium came from the serum proteins thatwere adsorbed onto the nonwoven SWNTs by non-specific bindings. The results from XPS gave a spec-

troscopic evidence of the protein adsorption ontononwoven SWNTs.

Cells growth behavior in various materials

Observation of cell morphology

Figure 4 gave representative micrographs showing3T3 L1 fibroblasts spreading and proliferating onnonwoven SWNTs. It could be seen that the cellssurvived well in the nonwoven SWNTs scaffoldsand persisted for at least 3 weeks. In the first twoweeks there were not significant differences of celldensity and growth state between the cells growingin nonwoven SWNTs scaffolds and the cells growingon PU films. However, in the third week, the stateof the cells growing in PU substrate went down(graphs were not shown here), and the cells growing

Figure 1. Optical graph and SEM image of nonwoven SWNTs, in which (a) optical graph; (b) SEM image; and (c) SEM imagewith higher magnification. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 2. Morphology of nonwoven SWNTs incubated in the complete IMDM medium for 6 h.



in nonwoven SWNTs scaffold still kept good state ofproliferation with high cell density.

Cell adhesion assay

Results from cell adhesion experiments (Fig. 5)were shown that nonwoven SWNTs scaffold withnanostructure enhanced 3T3-L1 cell adhesion signifi-cantly over time compared to the conventional poly-mers (blank plate and PU) and micronstructuredcontrol (CF), although CF exhibited higher capabilityof cell adhesion compared to the blank plate andPU. Through statistical analysis, it was indicated

that there were no significant differences of cell ad-hesion between the blank plate and PU (p > 0.05 ateach testing time point), while there was significantdifference of cell adhesion at the point of 6 hbetween CF and the two controls (PU and blankplate); among the four kinds of materials, nonwovenSWNTs exhibited distinguished advantage to celladhesion compared with CF, PU, and blank plate,and it presented p < 0.05 at each testing time point.

Cell proliferation assay

Figure 6 presented the proliferation of the cellsgrowing in various materials during 3 weeks. It

Figure 3. XPS spectra of nonwoven SWNTs after immersion in the culture medium for 6 h. [Color figure can be viewedin the online issue, which is available at www.interscience.wiley.com.]

Figure 4. Optical micrographs showing the morphology of 3T3-L1 fibroblasts cultured in the nonwoven SWNTs scaf-folds for 1, 2, and 3 weeks, respectively.

302 MENG ET AL.


could be seen from the graph that the proliferationextent of the cells growing on PU, CF, and nonwo-ven SWNTs was higher respectively than that on theblank plate at each testing time. For the cells grow-ing on PU and CF, there was no significant differ-ence of proliferation extent observed in the firstweek (p > 0.05); however, the number of viable cellsgrowing on CF was significantly more than thosegrowing on PU from the second week. Among thefour kinds of materials, nonwoven SWNTs exhibitedthe strongest function of promoting cells prolifera-tion. The enhanced cell numbers in nonwovenSWNTs scaffolds were detectable only after 1 dayand situation persisted for 3 weeks. In particular, inthe third week, the number of viable cells growingin nonwoven SWNTs scaffold was much higher thanthose in the other materials respectively (p < 0.05).

Actin observation

Figure 7 displayed the actin observation of the cellsgrowing in nonwoven SWNTs and PU, respectively.It could be seen that at the beginning of cultivation (6h), the FITC-labeled actin (F-actin) in cells growing onnonwoven SWNTs was concentrated in some areasalong the cell margin, where cells would extend pseu-dopodia [Fig. 7(a,a-1)]; however, no obvious concen-trated F-actin was observed in the cells on the PU film[Fig. 7(b)]. When the cells were cultivated for 24 h,staining of actin showed that cells cultured both onthe nonwoven SWNTs and on the PU had developedstress fibers and that actin was located homogene-ously in the cytoplasma [Fig. 7(c,d)]. After growing 3days, the cells on the PU film exhibited normal devel-oped stress fibers [Fig. 7(f)], while the cells growingon nonwoven SWNTs displayed high intensity of fluo-rescent, and the F-actin was highly organized, withstress fibers and microtubules observed clearly through

the cytoplasm of the fibroblasts [Fig. 7(e,e-1)]. Theobservations suggest that the cells sensed their growingenvironments and produced stronger responses to thenanotopographic structure of nonwoven SWNTs thanto the plain PU film.

Cell–cell communication

Figure 8 shows the proliferation of the cells grow-ing in different wells in the modified transwell devi-ces. Comparing the cell numbers in the upper wellswith nonwoven SWNTs and with PU film, respec-tively, it could be seen that the number of the cellsgrowing in nonwoven SWNTs was significant higherthan that growing on PU film, which was in agree-ment with the result obtained from proliferationassay. Moreover, it is interesting that the growthbehavior of the cells cultivated in the bottom wellcould be affected or induced by the cells cultivated inthe upper wells with the substrate of nonwovenSWNTs. Comparing the number of the cells growingin the bottom wells, we could find that the viable cellsin the bottom well with nonwoven SWNTs in theupper well were obviously more than those in thebottom well with PU film in the upper well after cul-tivated for 8 days, although the starting numbers ofthe cells in both bottom wells were the same. Becausethe culture medium for the upper well and the bot-tom well was connected in a transwell device, it wasimplied that the cells growing in nonwoven SWNTsscaffolds could promote the proliferation of the cellsgrowing on the PU film through cell–cell communica-tion by substance diffusion and exchange.

DISCUSSION

When a tissue cell comes into contact with bioma-terials, it will perceive the chemistry of a surface

Figure 5. Cell numbers attached on the various materialsat different culture time.

Figure 6. Viability of the cells growing on various materialsduring 3 weeks.



using integrin transmembrane proteins to find suita-ble site for adhesion, growth, and maturation. It hasbeen realized that in vitro cells sense not only thechemistry but also the topography of the scaffoldsthat are living environment for them. Topographical

cues, independent of biochemistry, generated by theECM may have significant effects upon cellularbehavior. Clearly, substratum topography has directeffects on the abilities of cells to orient, migrate, andproduce organized cytoskeletal arrangements. Hence,

Figure 7. FITC-labeled actin observation of cytoskeleton, in which (a, c, e) exhibited the cells growing in nonwovenSWNTs scaffolds for 6 h, 24 h, and 3 days, respectively, viewed by fluorescence microscope; (b, d, f) presented the cellsgrowing on PU for 6 h, 24 h, and 3 days, respectively, viewed by fluorescence microscope; (a-1) and (e-1) gave the graphsviewed with laser scanning confocal microscope corresponding to the (a) and (e), respectively. [Color figure can beviewed in the online issue, which is available at www.interscience.wiley.com.]

304 MENG ET AL.


it is very significant and attractive for damage repair-ing and tissue regeneration to establish artificial scaf-folds for cells preferential adhesion, and further pro-moted proliferation and induced specification.

Based on the characterizations of nonwoven SWNTs,it could be known that nonwoven SWNTs possesseda unique nanotopogaphic structure with average di-ameter of 20–30 nm for the SWNTs and average di-ameter of 200–300 nm for the pores in the mem-brane. These size features are similar to the ones ofnatural ECM. For instance, basement membrane is aubiquitous component of ECM, possessing a com-plex mixture of pore, ridges, and fibers with sizes inthe nanometer range.8 An example is the cornealepithelial basement membrane of macaque monkey;it consists of a porous membrane with a network oftightly cross-linked fibers, with the pores averaging72 nm and the fibers 77 nm in diameter, respec-tively. In another example of artificial basementmembrane, the commercially available Matrigel (BDBioscience, San Jose, Calif), similar nanoscale fea-tures can be seen. Matrigel is one of the most popu-lar coating materials for cell culture, which polymer-izes at room temperature to produce biological activematrix material resembling the mammalian cellularbasement membrane. Its features when viewed bySEM also reveal topography of fibers and pores at thenanoscale. The aforementioned detail highlights thefact that cells interact with nanoscale protein fibrils inthe ECM all the time.

It has been reported by some research groups thatSWNTs have strong interactions with most kinds ofproteins.9,10 The XPS analysis and SEM observationin this work indicated that the serum proteins in theculture medium would adsorb to the sidewalls ofcarbon nanotubes in the nonwoven SWNTs; thus, ascaffold with topography pores and tubes at thenanoscale, and pure carbon atom composition insideand serum protein layer outside was obtained whennonwoven SWNTs was immersed in the culture me-

dium. Hence, we would suggest that the nanostruc-tured scaffold of nonwoven SWNTs provided thecells an ideal environment compared with the othercontrol substrates in this work, which explained thesignificant long-term promotion of cell proliferationof nonwoven SWNTs.

Some in vitro studies with nanophase biomaterialshave also shown that cells respond differently tomaterials with nanoscale than to micron-sized rough-ness. For example, Dubly group11–14 had reportedthat the surfaces with nanoislands induced cellsstrong responses depending on the island height.Scaffolds with fibers in nanoscale, made of biode-gradable polymers through electrical spinning, hadshown improved serum proteins adsorption and pro-moted cell proliferation.15 Stupp group16 used murineneural progenitor cells to study in vitro the use of aself assembling artificial scaffold to direct cell differ-entiation, and they found that the gel scaffold com-posed of nanofibers with IKVAV epitope presentedto cells a nanostructured scaffold with high densityof available epitopes, which promoted their differen-tiation.

The advantages of nonwoven SWNTs differentfrom the synthetic or biological polymer materialslie in the high strength/weight ratio of SWNTs thatprovides stronger support, pure carbon atom com-position, and aromatic graphite structure that notonly being beneficial for protein adsorption but alsoprovides excellent biocompatibility, and high con-ductivities that provide more efficient transportationof energies and substances. It has been reported17

that high conductivity is a promising property aselectrical stimulation shown to be beneficial for tis-sue regeneration. Additionally, it has been shownthat increasing the conductivity of a material corre-lates directly to a decreasing foreign body response.

In this work, we had also observed that the cellsgrowing in nonwoven SWNTs scaffold exhibited sig-nificant proliferation and well-developed skeleton,and more importantly, their growth behavior hadobvious influence upon that of the cells cultivated inthe PU film when the two populations were culti-vated in a common culture system (modified trans-well device) in which the culture medium for theupper well and bottom well was connected. It hasbeen known that cells in higher animals communi-cate by means of hundreds of kinds of signal mole-cules. Cells can send signals to other cells of thesame type, as well as to themselves. Thus, it wasrational to consider that the cells cultivated in non-woven SWNTs scaffold (in the upper well) secretedsome kinds of signal substance that were deliveredinto the common medium through diffusion and tar-geted to either the cells growing in PU (in the bot-tom well) to encourage them make proliferation andto themselves (in the upper well) as well.

Figure 8. Proliferation of the cells cultivated in the trans-wells.



CONCLUSION

In this work, we have demonstrated that nonwo-ven SWNTs with nanotopographic structures in sim-ilarity to natural ECM provided more desirablegrowth environment for the cells compared with theother substrates including CF with microscopicstructure features, and the polymer films (polyur-ethane and blank plate). The cells cultivated in non-woven SWNTs scaffolds obtained promoted prolifer-ation in 3 weeks and developed highly organizedskeletal system. Additionally, we would suggest thatthe enhancement role of nonwoven SWNTs to prolif-eration of the cells cultivated in nonwoven SWNTsscaffold could affect the cells growing on the otherkinds of substrate through cell–cell communicationand encourage their proliferation. The results are ofsignificance not only to in vitro cell amplification inlarge scale, tissue regeneration, and repair but alsoto biomedical device applications.

Authors thank Prof. Boqin Qiang, Academician of Chi-nese Academy of Sciences, Institute of Basic Medicine,Chinese Academy of Medical Sciences, for his helpfulinstructions in the research. Authors thank Prof. XingyuJiang, National Center for Nanoscience and Technology ofChina for offering Confocal analysis, and Prof. Fei Lu,University of Science and Technology of China, for pro-viding XPS analysis.

References

1. Gallagher JO, McGhee KF, Wilkinson CDW, Riehle MO.Interaction of animal cells with ordered nano-topography.IEEE Trans Nanobioscience 2002;3:24–28.

2. Curtis ASG, Gadegaard N, Dalby MJ, Riehle MO, WilkinsonCDW, Aitchison G. Cells react to nanoscale order and sym-metry in their surroundings. IEEE Trans Nanobioscience 2004;3:61–65.

3. Shin M, Yoshimoto H, Vacanti JP. In vivo bone tissue engi-neering using mesenchymal stem cells on a novel electro-spun nanofibrous scaffold. Tissue Eng 2004;10:33–41.

4. Yang F, Murugan R, Wang S, Ramakrishna S. Electrospin-ning of nano/micro scale poly(L-lactic acid) aligned fibers

and their potential in neural tissue engineering. Biomaterials2005;26:2603–2610.

5. Dalby MJ, Riehle MO, Johnstone H, Affrossman S, CurtisASG. In vitro reaction of endothelial cells to polymer dem-ixed nanotopography. Biomaterials 2002;23:2945–2954.

6. Affrossman S, Henn G, O’Neill SA, Pethrick RA, Stamm M.Surface topography and composition of deuterated polysty-rene–poly(bromostyrene) blends. Macromolecules 1996;29:5010–5016.

7. Song L, Ci LJ, Li LV, Zhou ZP, Yan XQ, Liu DF, Yuan HJ,Gao Y, Wang JX, Liu LF, Zhao XW, Zhang ZX, Dou XY,Zhou WY, Wang G, Wang CY, Xie SS. Direct synthesis ofmacroscale single-walled carbon nanotubes nonwoven. AdvMater (Weinheim, Ger) 2004;16:1529.

8. Yim EKF, Leong KW. Significance of synthetic nanostructuresin dictating cellular response. Nanomed: Nanotechnol BiolMed 2005;1:10–21.

9. Chen RJ, Zhang YG, Wang DW, Dai HJ. Noncovalent side-wall functionalization of single-walled carbon nanotubes forprotein immobilization. J Am Chem Soc 2001;123:3838,3839.

10. Chen RJ, Bangsaruntip S, Drouvalakis KA, Kam NWS, ShimM, Li YN, Kim W, Utz PJ, Dai HJ. Noncovalent functionali-zation of carbon nanotubes for highly specific electronic bio-sensors. Proc Natl Acad Sci USA 2003;100:4984–4989.

11. Dalby MJ, Yarwood SJ, Riehle MO, Johnstone HJH, Affross-man S, Curtis ASG. Increasing fibroblast response to materi-als using namotopography: Morphological and genetic meas-urements of cell response to 13-nm-high polymer demixedislands. Exp Cell Res 2002;276:1–9.

12. Dalby MJ, Childs S, Riehle MO, Johnstone HJH, Affrossman S,Curtis ASG. Fibroblast rection to island topography: Changesin cytoskeleton and morphology with time. Biomaterials 2003;24:927–935.

13. Dalby MJ, Giannaras D, Riehle MO, Gadegaard N, Affross-man S, Curtis ASG. Rapid fibroblast adhesion to 27 nm highpolymer demixed nano-topography. Biomaterials 2004;25:77–83.

14. Dalby MJ, Riehle MO, Johnstone HJH, Affrossman S, CurtisASG. Polymer-demixed nanotopography: Control of fibro-blast spreading and proliferation. Tissue Eng 2002;8:1099–1108.

15. Woo KM, Chen VJ, Ma PX. Nano-fibrous scaffolding archi-tecture selectively enhances protein adsorption contributingto cell attachment. J Biomed Mater Res 2003;67A:531–537.

16. Silva GA, Czeisler C, Niece KL, Beniash E, Harrington DA,Kessler JA, Stupp SI. Selective differentiation of neural progen-itor cells by high-epitope density nanofibers. Science 2004;303:1352–1355.

17. Mckenzie JL, Waid MC, Shi R, Webster TJ. Decreased func-tions of astrocytes on carbon nanofiber materials. Biomateri-als 2004;25:1309–1317.

306 MENG ET AL.


Documents

Using single-walled carbon nanotubes nonwoven films as scaffolds to enhance long-term cell proliferation in vitro