BC nanofibres In vitro study of genotoxicity and cell proliferation

BC nanofibres In vitro study of genotoxicity and cell proliferation

(Parte 1 de 3)

Toxicology Letters journal homepage: w.elsevier.com/locate/toxlet

BC nanofibres: In vitro study of genotoxicity and cell proliferation Susana Moreiraa,1, Naisandra Bezerra Silvab,1, Jailma Almeida-Limab, Hugo Alexandre Oliveira Rochab,

Silvia Regina Batistuzzo Medeirosc, Clodomiro Alves Jr.d, Francisco Miguel Gamaa,∗IBB, Institute for Biotechnology and Bioengineering, Universidade do Minho, Campus de Gualtar, 4710-057 Braga, PortugalDepartamento de Bioquímica, Universidade Federal do Rio Grande do Norte – UFRN, BrazilDepartamento de Biologia Celular e Genética, Universidade Federal do Rio Grande do Norte – UFRN, BrazilDepartamento de Física, Universidade Federal do Rio Grande do Norte – UFRN, Brazil article info

Keywords: Nanofibres Nonotoxicology Bacterial cellulose Genotoxicity abstract

Nanomaterialshaveunusualpropertiesnotfoundinthebulkmaterials,whichcanbeexploitedinnumerous applications such as biosensing, electronics, scaffolds for tissue engineering, diagnostics and drug delivery. However, research in the past few years has turned up a range of potential health hazards, which has given birth to the new discipline of nanotoxicology. Bacterial cellulose (BC) is a promising material for biomedical applications, namely due its biocompatibility. Although BC has been shown not to be cytotoxic or genotoxic, the properties of isolated BC nanofibres (NFs) on cells and tissues has never been analysed. Considering the toxicity associated to other fibre-shaped nanoparticles, it seems crucial to evaluate the toxicity associated to the BC-NFs.

Inthiswork,nanofibreswereproducedfrombacterialcellulosebyacombinationofacidandultrasonic treatment. The genotoxicity of nanofibres from bacterial cellulose was analysed in vitro, using techniques previously demonstrated to detect the genotoxicity of fibrous nanoparticles. The results from single cell gel electrophoresis (also known as comet assay) and the Salmonella reversion assays showed that NFs are not genotoxicity under the conditions tested. A proliferation assay using fibroblasts and CHO cells reveals a slight reduction in the proliferation rate, although no modification in the cell morphology is observed. © 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

The development of artificial materials with biomimetic behaviour is essential for tissue engineering purposes. Scaffolds based on nanofibres (NFs) mimic the natural extracellular matrix and its nanoscale fibrous structure. Several approaches have been described in order to achieve materials based on nanofibres from synthetic or natural polymers (Ma et al., 2005; Ashammakhi et al., 2007).

Bacterial cellulose (BC), secreted by Gluconacetobacter xylinus, has been presented as a biocompatible scaffold for the engineering ofcartilageandbloodvessels,wounddressing,guidedtissueregeneration, among other applications (Astley et al., 2003; Entcheva et al.,2004;Svenssonetal.,2005;TabuchiandBaba,2005;Czajaetal., 2007; Teeri et al., 2007; Andrade et al., 2008; Backdahl et al., 2008; Maneerung et al., 2008; Rambo et al., 2008). BC has unique characteristics including high purity, high crystallinity and remarkable mechanical properties, due to the uniform ultrafine-fibre network structure,thehighplanarorientationoftheribbon-likefibreswhen compressed into sheets, the good chemical stability, and the high water holding capacity (Svensson et al., 2005). Several materials based on bacterial cellulose, recognized as non-genotoxic and noncytotoxic, have been commercialized (Schmitt et al., 1991; Jonas and Farah, 1998).

Since nanomaterials have unusual properties, not found in the bulk material, such as high surface reactivity and ability to cross-cell membranes, concerns about their safety and toxicology emerged. The impact of nanostructural features in the interaction of a material with cells and tissues is dependent on the size, chemical composition, surface structure, solubility, shape, and on the supramolecular structural organization (Barnes et al., 2008). A major concern with fibres is their carcinogenic potential. There is sufficient evidence that all forms of asbestos (generic term for a group of six naturally occurring fibrous silicate minerals) are carcinogenic and co-carcinogen to man (Speit, 2002; Dopp et al., 2005). Moreover, recent studies described the toxicity of materials associated to size or shape; namely, the toxicity of carbon nanotubes (Donaldson et al., 2006; Poland et al., 2008) and the size-dependence toxicity of gold or ferric oxide nanoparticles was reported (Pan et al., 2007; Backdahl et al., 2008; Wang et al., 2009).

The toxicity associated with inhaled fibres such as asbestos has been described. Inhaled fibres may be toxic, particularly when they are “long, thin and durable” (Donaldson et al., 2006). Asbestos

fibres are dangerous because the fibres split lengthwise, producing thin fibres that can enter the lungs, being “moderately durable” once there (Speit, 2002). Although cellulose fibres, from wood pulp and textile fabric, are used without “significant concern”, cellulose fibres share similar features with asbestos, including the needle-like shape and biopersistence. Moreover, the inflammatory responses of respirable cellulose fibres (wood pulp) using animal models were already reported (Cullen et al., 2000). In light of these results,itseemscrucialtoevaluatethetoxicityoftheBCnanofibres. It must be remarked that, although BC cannot be enzymatically degraded in the human body, the inflammatory processes may actuallydegradecellulosetosomeextent.Giventhecurrentfocusof BCasapromisingbiomaterialwithavarietyofapplications,itisrelevanttoevaluatenotonlythetoxicityofBCmembranesorscaffolds, but also of its degradation products, including BC nanofibres.

Indeed, although in vivo studies demonstrate the BC biocompatibility (Helenius et al., 2006), and lack of mutagenicity (Schmitt et al., 1991), no reports are available on the BC nanofibres toxicity. Although BC is not expected to be degraded in vivo, safety concerns makes this study mandatory. It is well accepted that in vitro studies using cell systems are valuable tools to clarify the cellular mechanismsinvolvedingenototoxiceffects,includingDNAdamage(Speit, 2002; Dusinská et al., 2004). Therefore, the aim of this study is to evaluate the genotoxicity of cellulose nanofibres at cellular level using the single cell gel electrophoresis and the Salmonella reversion assays. The cell proliferation in the presence of nanofibres was also evaluated. These tests are useful as a screening tool for setting priorities because they are an inexpensive and a quick way to help single out substances that should be targeted for further testing. Furthermore, these assays were already used to demonstrate the genotoxic effect of asbestos fibres in mammalian cells in vitro (Speit, 2002; Dusinská et al., 2004).

2. Materials and methods 2.1. Bacterial strain, cells and culture medium

The cellulose was produced by G. xylinus (ATCC 53582), purchased from the

American Type Culture Collection, grown statically in Hestrin and Schramm (1954) medium, pH 5 at 30 C,5d ays.

In the Salmonella reversion assay, four strains of Salmonella tryphimurium (Dr.

The proliferation assays were performed using mouse embryo fibroblasts 3T3

(ATCCCCL-164),growninDulbecco’smodifiedEagle’smedia(DMEM)supplemented with 10% newborn calf serum (Invitrogen), and Chinese Hamster Ovary (CHO), grown in DMEM media supplemented with 10% fetal bovine serum (Invitrogen), both culture medium were supplemented with penicillin/streptomycin (1 g/ml) (Sigma–Aldrich, St. Louis, USA) and the incubation was at 37 C, in a fully modified air containing 5% CO . The same conditions were used to grow CHO cells in Comet assay. The cell viability was assessed using the MTT (3-[4,5-dimethylthiazol-2-yl]- 2,5-diphenyl-tetrazolium bromide) assay, obtained from Invitrogen.

2.2. Production of BC nanofibres

The production of bacterial cellulose was performed by growing G. xylinus in

Hestrin–Schramm medium, pH 5. After inoculation, the culture (100ml) was incubated, first with agitation during 8h, and then statically at 30 C, for 5–7 days. BC pellicles were purified in a 4% NaOH solution at 70 C, for 90min. BC was then neutralised by thoroughly washing with water. Finally, BC pellicles were lyophilised prior to use.

The nanofibres production, by acidic and/or ultrasonic treatment, was based on previousworks(RomanandWinter,2004;Zhaoetal.,2007).Theacidhydrolysiswas performed as follows: 20mg of dry BC was sliced in small pieces and 2ml of H SO (50%,v/v)wasadded.Themixturewaskeptat40 C,for2hwithvigorousstirring.To stop the hydrolysis, 10ml of cold water was added and the cellulose was recovered by filtration, using a membrane with a 0.45 m pore size. Then, the cellulose was washed out with 20ml of water and the recovered pellet was resuspended in 10ml of water. This suspension was treated by sonication at 40W (Branson Ultrasonic Disruptor, Sonifier I/W450) for 10min (samples were maintained on ice during sonication). Then, the NFs suspension was centrifuged (1h, 15,000rpm), and the pellet resuspended in water and sonicated again, in the same condition, for another 10min. The yield of the process was evaluated by quantifying the total sugar in the samples, using the phenol-sulphuric method (Dubois et al., 1956).

2.3. TEM analysis

TheNFsobtainedwerestainedwithuranylacetateandanalysedbytransmission electronic microscopy (TEM, Zeiss 902A Orius SC 1000; 50kV).

2.4. Evaluation of cellulose nanofibres mutagenicity by Salmonella reversion assay

Four S. tryphimurium strains were used to study the potential mutagenicity effect of the cellulosic NFs. The procedure was to some extent modified from the original description by Kado et al. (1986). This assay was performed in miscrosuspension with or without S9 mixture (Moltox , North Carolina, USA), using 0.1, 0.5 or 1.0mg/ml of NFs suspension. The negative control (NC) was distilled water, and the positive controls (PC) employed were: 0.1 g/plate 4NQO (4-nitroquinoline 1- oxide) for the TA97a and TA98 strains; 5.0 g/plate sodium azide for the TA100 strain;and0.5 g/platemytomicynCfortheTA102strain.Briefly,105 lofamixture containing the NFs suspension and cell suspension (10 cells/ml) were incubated at 37 C for 90min. Then, 2.5ml of molten Top agar (0.6% bacto-agar and 0.5% NaCl) was added, before plating in a Petri dish containing minimal agar (1.5% agar, Vogel- BonnerEmedium).TheHis revertantcolonieswerecountedafter72hofincubation at 37 C. All experiments were repeated at least three times with three replicas. The mutagenicity of cellulose NFs was evaluated according to the following parameters: the maximum number of revertants in the presence of the NFs should be 2-fold or more relative to the negative control; a dose-dependent increase in the number of revertants should be observed (Mortelmans and Zeiger, 2000).

2.5. Proliferation assays

The proliferation assays were performed in vitro as follows: 1ml of the CHO or mouse embryo fibroblast 3T3 cell suspension (10 cells/ml) was seeded in a 24-well polystyreneplate(TPP,Switzerland).Thecellswereallowedtoadherefor4h.Before the addition of cellulose NFs, the medium with non-adherent cells was removed and the NFs containing medium (to a final concentration of 1, 0.5 or 0.1mg/ml) was added.AcontrolwithoutNFswascarriedout.Thecellulargrowthat0,24,48and72h ofincubationwasevaluatedbyMTTassay,acolorimetrictestthatgivesameasureof the mitochondrial activity. The effect of NFs on the cell morphology was evaluated by microscopic observation using a Nikon Eclipse TE300 Inverted Microscope.

2.6. Evaluation of cellulose nanofibres genotoxicity by single cell gel assay (comet assay)

The DNA integrity was evaluated by alkaline single cell gel assay (also kwon as comet assay) using CHO cells grown in the presence of different NFs concentration.

In this assay, 2ml of CHO cell suspension (10 cells/ml) were seeded on a 6- well polystyrene plate (TPP, Switzerland). After 16h, the medium was refreshed with medium containing the NFs (0.1, 0.5 or 1mg/ml). Cells were incubated with NFs suspension during 48h. Hydrogen peroxide (100mM) and water were used as positive and negative controls, respectively. The alkaline comet assay was performedasdescribedbySinghetal.(1988).Briefly,cellsweretrypsinizedfrom6-well polystyrene plate, and resuspended in 50 l of medium. The cell viability was determined in a Neubauer counting chamber using the trypan blue exclusion test. A volume of 10 l of the cellular suspension were embedded in 0.5% low-meltingpoint agarose and plated on agarose-coated microscope slide. Then, the slides with cells were treated with lysis solution (2.5M NaHO, 0.1M EDTA, 0.010M Tris, 1% Triton X-100, 10% DMSO, adjusted to pH 10) for 12h at 4 C, rinsed with distilled water, and placed in the electrophoresis buffer (0.3M NaOH, pH 13 and 0.001M EDTA), for 20min to allow DNA unwinding. Following electrophoresis (30min, at 25V and 300mA), the slides were neutralised with 0.4M Tris buffer (pH 7.5) and stained with ethidium bromide (20mg/ml). The slides were analysed through fluorescence microscopy (Nikon Eclipse TE300 microscope equipped with a Nikon E600 camera, 0.488 m/pixel). At least 300 cells per condition tested were analysed.

The DNA damages were evaluated by image analysis using the “Comet Assay

IV version 4.2” image analysis system. Data collected from each cell included tail length (TL), tail migration (TMi), percent tail DNA (TI), and tail moment (TM), which correspondtheproductofthecometlengthandtheamountofDNAinthetail(Olive and Durand, 1992).

2.7. Statistic analysis

Theone-wayanalysisofvariance(ANOVA)wasappliedtostatisticsevaluationof thecometscoresandtotheproliferationassaysresults.Thepost-testTukey–Kramer Multiple Comparisons test was used to compare the scores of the samples and positive control, the analysis were performed using GraphPad Prisma v 3.05.

3. Results and discussion 3.1. Production of BC nanofibres

G. xylinum synthesizes cellulose nanofibres with 40–50nm width (the bacterial cellulose ribbons), which assemble in a static culture as a white gelatinous material (pellicle) on the surface of the culture liquid. The native cellulose consists of sets of parallel chains of -1,4-d-glucopyranose units interlinked by intermolecular hydrogen bonds (Czaja et al., 2007). Several works describe the production of nanofibres from different cellulosic sources, using acid hydrolysis (Araki et al., 1999; Roman and Winter, 2004)o r mechanictreatment(Zhaoetal.,2007).Thesetwoapproacheswere usedinordertoextractNFsfromBC.Theacidhydrolysiswastested using a range of acid concentrations, temperatures and treatment time. Concentrations of H2SO4 superior to 50% resulted in extensive hydrolysis, yielding less than 20% of the material used (data not shown). The acid concentration is in fact the critical parameter in the acid hydrolysis approach. The use 50% H2SO4, for 2h at 40◦C, yielded 50% of nanofibres. According to Zhao et al. (2007), sonica- tion can also be successfully used to extract NFs from natural materials,includingcellulosefromwood,cotton,bamboo.Thisapproach wasalsoappliedtoBC.Usingacidhydrolysis(50%H2SO4,2h,40◦C) and sonication (20min, 40W), needle shaped cellulose NFs with

3.2. Evaluation of cellulose nanofibres mutagenicity by Salmonella reversion assay

The purpose of the bacterial reverse mutation assay is to evaluate the mutagenicity of the cellulose NFs, by measuring its ability to induce reverse mutations at selected loci, in several bacte- rial strains. Having into account that mutations are essential for cancer formation, the reliable characterization of mutagenicity is mandatory, while characterizing the safety of a biomaterial. The Kado test (Kado et al., 1983) is a modification with improved sensitivity of the Ames test (Ames et al., 1972). This is a simple, quick and inexpensive mutagenicity test, required for safety testing of a variety of compounds, including drugs, medical devices, food additives, industrial chemicals and pesticides (McCann et al., 1975). Furthermore, the potential mutagenicity of BC and of some its derivates were already accessed using the Ames assay (Schmitt et al., 1991), therefore it was selected as a first approach in this work to investigate the possible mutagenicity of the cellulose NFs. The strains used were specially constructed to allow detection of mutagens acting via different mechanisms, namely frameshift mutations (TA97a and TA98 strains), base-pair substitution mutations (TA100 and TA102), detection of oxidative and alkylatingmutagensandactiveformsofoxygen(TA102)(Hakuraet al., 2005). Table 1 presents the results obtained with the different strains.

The reversion of the histidine phenotype in Salmonella strains is often adopted as a criteria for the classification of molecules as mutagenic. The results obtained in the presence of the cellulose NFs, without S9 mixture, correspond to the spontaneous reversion for each strain and are similar to those obtained to negative control (Table 1). In the presence of S9 mixture, an increase of revertant colonies per plate, for the TA98 and TA100 strains, is detected as compared with control; however, the increases was in each case <2-fold and does not appear to be dose-related. The results suggest that, under the conditions tested, the cellulose NFs does not present mutagenic behaviour, as described previously for BC and some fibrous BC-based materials (Schmitt et al., 1991).

Fig. 1. TEM image of cellulose nanofibres (50kV; Zeiss 902A Orius SC 1000).

Table 1 Results obtained in Salmonella reversion assay.

PC: positive control: 0.1 g/plate of 4NQO to TA97a and TA98, 5.0 g/plate sodium azide to TA100 and 0.5 g/plate mytomicyn C to TA102; NC: negative control: H O; SD:

standard deviation.Strain.

3.3. In vitro proliferation assay

The cellular morphology and proliferation may be affected by the presence of nanostructural patterns. Several studies analysed the proliferation of different cell lines on BC membranes, confirming its non-toxicity and applicability as scaffold for cell proliferation. However, depending on the cells used, the effect of the biomaterial on the proliferation rate and the cell morphology may be quite different (Sanchavanakit et al., 2006). Several studies showed that the cytotoxicity of a nanomaterial is many times cell-specific (Cullen et al., 2002). Recently, De Nicola et al. (2007) reported that, although carbon nanotubes do not present cytotoxic effect on human leukemic U937 cells, the proliferation rate is deeplyaltered.Moreover,Bottinietal.(2006)showedthatthesame nanotubes refereed above induce apoptosis in T lymphocytic cells, suggesting that cytotoxicity may be cell-specific. In addition, it has been reported that asbestos fibres inhibits the growth of CHO cells (Speit, 2002), and yet the same fibres stimulate the proliferation of different kinds of cells, in vitro, including fibroblasts (Bernstein et al., 2005). Taking in consideration the evidence of contradictory, cell-specific effects arising from the interaction cell-biomaterial, the evaluation of the NFs effect on proliferative rate was performed both with CHO cells and fibroblasts. In both cases, the proliferation was about 15–20% lower in the presence of NFs, after 72h of cell culture, irrespective of the concentration used (Fig. 2). The lower proliferation rate may stem from the insolubility of NFs and their slow deposition on the polystyrene plate. It is known that cell proliferation is dependent on characteristics of material surface, such as it roughness. In addition, it was also described that cell proliferation on BC membrane is slower than on the cell culture plate

Fig. 2. MTT results from proliferation assays using mouse embryonic fibroblast 3T3 and CHO (mean±SD; **P<0.05; ***P<0.005). Image obtained by optical microscopy of fibroblasts grown in the presence of cellulose NFs during 72h. Scale bar=20 m.

(Backdahl et al., 2006). However, the microscopic observations did not reveal differences in the cellular morphology.

3.4. Evaluation of cellulose nanofibres genotoxicity by comet assay

(Parte 1 de 3)