NANO LETTERS

Toxicological Impact Studies Based on Escherichia coli Bacteria in Ultrafine ZnO Nanoparticles Colloidal Medium
Roberta Brayner,*, Roselyne Ferrari-Iliou, Nicolas Brivois,, Shakib Djediat, Marc F. Benedetti, and Fernand Fievet
Interfaces, Traitements, Organisation et Dynamique des Systemes (ITODYS), ` UMR-CNRS 7086, UniVersite Paris 7-Denis Diderot, case 7090, 2 place Jussieu, 75251 Paris Cedex 05, France, Laboratoire de Geochimie des Eaux, UMR-CNRS 7047, UniVersite Paris 7 & IPGP, 2 place Jussieu, 75251 Paris Cedex 05, France, and Departement RDDM, USM 505, Museum National dHistoire Naturelle, case 39, 57 rue CuVier, 75005 Paris, France
Received November 24, 2005; Revised Manuscript Received February 23, 2006

2006 Vol. 6, No. 4 866-870

ABSTRACT
We report here preliminary studies of biocidal effects and cellular internalization of ZnO nanoparticles on Escherichia coli bacteria. ZnO nanoparticles were synthesized in di(ethylene glycol) (DEG) medium by forced hydrolysis of ionic Zn2+ salts. Particle size and shape were controlled by addition of small molecules and macromolecules such as tri-n-octylphosphine oxide, sodium dodecyl sulfate, polyoxyethylene stearyl ether, and bovine serum albumin. Transmission electron microscopy (TEM) and X-ray diffraction analyses were used to characterize particle structure, size, and morphology. Bactericidal tests were performed in LuriaBertani medium on solid agar plates and in liquid systems with different concentrations of small and macromolecules and also with ZnO nanoparticles. TEM analyses of bacteria thin sections were used to study biocidal action of ZnO materials. The results confirmed that E. coli cells after contact with DEG and ZnO were damaged showing a Gram-negative triple membrane disorganization. This behavior causes the increase of membrane permeability leading to accumulation of ZnO nanoparticles in the bacterial membrane and also cellular internalization of these nanoparticles.

Introduction. Toxicological studies carried out in the last 10 years showed that ultrafine particles (d < 100 nm) pose serious problems to the lungs.1-3 It was demonstrated that these nanoparticles cause more inflammation than larger respirable particles made from the same material when delivered at the same mass dose. This behavior has been observed for a range of different materials of generally low toxicity such as carbon black (CB) and TiO2.4 With the increased presence of nanomaterials in commercial products, a growing public debate is emerging on toxicological and environmental effects of direct and indirect exposure to these materials, and these effects are not completely elucidated. For example, micronized or ultrafine (20-50 nm) TiO2, socalled microreflectors (US Federal Register, 43FR38206, 25 August 1978), can be employed as a safe physical sunscreen because it reflects and scatters UVB and UVA in sunlight. Moreover, it was demonstrated that these samples catalyze sunlight-illuminated DNA damage in cells both in vitro and
* Corresponding author: brayner@ccr.jussieu.fr. Interfaces, Traitements, Organisation et Dynamique des Systemes ` (ITODYS), UMR-CNRS 7086, Universite Paris 7-Denis Diderot. Laboratoire de Geochimie des Eaux, UMR-CNRS 7047, Universite Paris 7 & IPGP. Departement RDDM, USM 505, Museum National dHistoire Naturelle. 10.1021/nl052326h CCC: $33.50 Published on Web 03/11/2006 2006 American Chemical Society

in vivo.5 These results may be relevant to the overall effects of sunscreens. The crystalline forms of TiO2, anatase and rutile, are semiconductors with gap energies of about 3.23 and 3.06 eV, respectively.5 Here, we used ZnO nanoparticles because it is also a sunscreen semiconductor with a band gap of 3.3 eV,6 very close to TiO2 anatase. ZnO semiconductor is a versatile material that has achievable applications in sensors, photocatalysis, solar cells, transparent electrodes, electroluminescent devices, and ultraviolet laser diodes.7-13 The results described in this work provide a range of concentrations that ZnO nanoparticles, in a polyol medium, were toxic to Escherichia coli cultured cells and also show the cellular internalization of these nanoparticles. Materials and Methods. ZnO nanoparticles were synthesized in di(ethylene glycol) (DEG) medium by forced hydrolysis of zinc acetate.14,15 The general procedure involves addition of zinc acetate to a given volume of polyol and H2O to reach a final concentration between 0.06 and 0.6 molL-1. nH2O The hydrolysis ratio H ) was varied from 10 to nZn2+ 80. To control particle size and shape, small molecules and macromolecules such as tri-n-octylphosphine oxide (TOPO),

Table 1. Characteristics of ZnO Samples

MET (nm) sample molecule molecule concn ZnO-1 ZnO-2 ZnO-3 ZnO-4 ZnO-5 ZnO-6 BSA TOPO TOPO SDS Brij-76 10-2 M 10-1 M 10-2 M 10-1 M 25 g L-1 H length diameter 10.8 ( 2.2 7.7 ( 0.9 14.0 ( 0.9 13.1 ( 1.5 11.7 ( 2.5 11.7 ( 1.4

10 70 10 30 32.5 ( 1.4 10 22.7 ( 5.7 10 83.8 ( 28.4

sodium dodecyl sulfate (SDS), polyoxyethylene stearyl ether (Brij-76), or bovine serum albumin (BSA) were added to zinc acetate and polyol solution with concentrations between 10-2 and 10-1 M. The mixture was then heated under vigorous stirring, to a temperature varying between 150 and 220 C. Escherichia coli strain MG1655 was used in this study. This strain was kindly supplied by Dr. Dukan (Bacterial Chemistry Laboratory, IBSM, CNRS, UMR 9043, Marseille, France). It corresponds to the wild type (FLambda-), and it has been used as the E. coli strain reference for genome sequencing. E. coli was grown aerobically at 37 C for 12-16 h with shaking (200 rpm) in Luria-Bertani medium (LB) containing 5 gL-1 of yeast extract, 10 gL-1 of tryptone, and 10 gL-1 of NaCl. The saturated cultures were diluted in fresh LB medium starting at OD600 ) 0.05 (optical density) and incubated at 37 C overnight in Erlenmeyer flasks with vigorous shaking at 200 rpm as well as on solid agar plates. Bacteriological tests were performed on solid agar plates with different concentrations of small molecules and macromolecules and also with different concentrations of ZnO nanoparticles (from 10-2 to 10-4 M). The inoculated cells were estimated to 200 cfu (colonyforming units) per plate, and all results were compared to a control without ZnO nanoparticles. E. coli and ZnO nanoparticles in polyol medium cultured on LB agar plates were incubated overnight at 37 C, and the number of colonies was counted. ZnO-free LB agar plates and DEG-LB agar plates were used as a control. Transmission electron microscopy (TEM) and X-ray diffraction (XRD) analyses were used to characterize particle structure, size, and morphology. TEM analyses of bacteria thin sections were used to study biocidal action of ZnO nanoparticles. Bacteria samples, grown in a liquid culture medium, at 37 C for 12 h were then diluted in fresh LB medium starting at OD600 ) 0.05. In the exponential phase, a 10-3 M concentration of ZnO-TOPO nanoparticles in a polyol medium was added to the culture medium. The final solution was then incubated overnight, a sample containing 5.8 106 bacteria fixed in a mixture containing 2.5% glutaraldehyde, 1.0% acrolein, and 0.1% ruthenium red in a phosphate Sorengen buffer (0.1 M, pH 7.2). Results and Discussion. Table 1 presents the characteristics of ZnO samples and Figure 1 shows TEM micrographs of ZnO nanoparticles obtained with different small molecules and macromolecules. There are two factors that influence size and shape of ZnO materials synthesized by forced hydrolysis in polyol medium: (i) the nature of molecular template added during ZnO formation and (ii) the hydrolysis
Nano Lett., Vol. 6, No. 4, 2006

Figure 1. TEM micrographs of ZnO nanoparticles synthesized in DEG medium: (a) ZnO without addition of macromolecules; (b) ZnO-BSA (top left schematic BSA conformation); (c) ZnO-TOPO (H ) 10); (d) ZnO-TOPO (H ) 30); (e) ZnO-SDS; and (f) ZnOBrij-76.

ratio (Table 1). For ZnO prepared without addition of small molecules or macromolecules (Figure 1a and Table 1), spherical nanoparticles with a broad size distribution were observed. On the other hand, a unique kind of onedimensionally organized ZnO nanoparticles was obtained by electrostatic stabilization with BSA as templating material (Figure 1b). BSA protein has been extensively studied in previous works as suitable for drug delivery.16-17 The advantages of BSA utilization as a template material are its biodegradability18 and nontoxic19 and antigenic20 properties. In addition, because of their defined primary structure, and high content of charged amino acids (i.e., lysine), BSA could allow electrostatic adsorption of positively or negatively charged molecules and/or nanoparticles. A very narrow size distribution of ZnO spherical nanoparticles was obtained with addition of TOPO as a template (Figure 1c and Table 1). In this case, TOPO acts as a template and also as a linker in the self-assembly of ZnO nanoparticles. For ZnO-TOPO synthesis, when hydrolysis ratio increased from H ) 10 to H ) 30 and TOPO concentration decreased from 10-1 to 10-2 M, ZnO nanorod formation was observed (Figure 1d). Finally, ZnO anisotropic morphologies were obtained by coalescence of spherical and cubic nanoparticles in the presence of SDS and Brij-76 macromolecules, respectively (parts e and f of Figure 1). In the case of Brij-76, used as a structure director, assemblies of ZnO nanocubes were observed. These nanocubes have a coalescence tendency that allows formation of ZnO nanobelts (Figure 1f). Figure 2 presents XRD patterns of ZnO nanoparticles after hydrolysis reaction. For all samples, the peaks can be indexed as pure
867

Figure 3. Bactericidal tests as a function of (a) different small molecules and macromolecules and (b) ZnO concentration after incubation at 37 C overnight.

Figure 2. XRD patterns of (a) ZnO-1 sample, (b) comparison between ZnO-1 and ZnO-4 samples, and (c) HRTEM of ZnO nanorod (growth orientation along c axis).

ZnO wurtzite structure (hcp structure; Figure 2a). Crystallite sizes inferred from X-ray line broadening compared to particle sizes obtained from TEM are quite similar. In addition, the increase of the (002) peak of ZnO samples presenting anisotropic forms (Figure 2b) shows that the preferred growth orientation of these samples is along the c-axis. An isolated nanorod detected by high-resolution TEM (HRTEM) analysis is single crystalline (Figure 2c). The analysis of HRTEM images revealed hcp structure with the same orientation confirming that the c-axis of this structure is parallel to nanorod revolution axis. The bactericidal effect of small molecules and macromolecules (used in this work)
868

on Gram-negative bacteria is presented in Figure 3a. For statistical studies, all tests were repeated three times after culture incubation at 37 C overnight. The number of bacterial colonies grown on control and DEG plates was quite similar (100%). Polyol is a sugar, and it can be metabolized to short chain fatty acids and gases by E. coli (Figure 3a) such as glucose presents in the culture medium. Bacterial inhibition growth (15%) was observed in the presence of SDS surfactant (Figure 3a). This behavior was likeky due to denaturalization of bacterial protein by SDS.22 For TOPO and Brij-76 a positive effect was observed (Figure 3a). After their metabolization, these molecules may promote bacterial growth. This effect was more important for E. coli supplemented with TOPO. The same test was made only with ZnO nanoparticles (ZnO-1 sample, Figure 1a) in DEG medium to avoid molecule interaction effects (Figure 3b). The concentration on ZnO nanoparticles was varied from 10-2 to 10-3 M, and free LB plates were also used as a control. The presence of these nanoparticles at a concentration between 10-2 and 3.0 10-3 M caused 100% inhibition of bacterial growth. Concentrations between 3.0 10-3 and 1.5 10-3 M inhibited bacterial growth by 85%. For concentrations between 1.5 10-3 and 10-3 M, an increase of E. coli colonies was observed. Maybe one could also predict through these results whether this increase is metabolism-dependent, because bacteria can metabolize Zn2+ as an oligoelement showing that between these concentrations, ZnO nanoparticles are not toxic for E. coli. To corroborate our results, it was demonstrated by Roselli et al.23a that ZnO nanoparticle concentration from 10-4 and 10-3 M did not affect cell permeability (measured by transepithelial electrical resistance) even after 31 h of treatment. In their study, cell permeability slightly increases at 5 10-3
Nano Lett., Vol. 6, No. 4, 2006

Scheme 1.

Cell Wall

Figure 4. TEM micrographs of E. coli thin sections: (a) and (b) E. coli grown in free LB medium; (c) and (d) E. coli in DEG medium; (e) and (f) E. coli cellular division; (g-j) ZnO nanoparticles (10-3 M) internalization. (NP ) ZnO nanoparticles.)

M compared with untreated cells. In addition, recently, efficient Zn uptake by growing cells of Aspergillus spp. isolated from industrial waste has been reported.23b In all cases, the dynamics of bacterial growth was not elucidated. TEM analysis of sample thin sections allows direct visualization of morphological changes resulting in the microorganisms upon contact with polyol medium and ZnO nanoparNano Lett., Vol. 6, No. 4, 2006

ticles (ZnO-3 sample) both at 10-3 M concentration in a liquid culture medium (Figure 4). This sample was chosen because its presents a narrow size distribution of spherical nanoparticles (Figure 1c). Moreover, it was demonstrated, in this work, that TOPO molecule does not prevent bacterial growth (Figure 3a). Figure 4a shows thin sections of Escherichia coli grown in free LB liquid medium. This bacterium is a moderately sized Gram-negative bacillus that presents a tubular form. The E. coli Gram-negative cell wall (Figure 4b) is composed of an organized triple membrane containing a thin inner layer of peptidoglycan between an outer membrane consisting of porins, phospholipids molecules, lipopolysaccharides (LPS), lipoproteins, surface proteins, and a cytoplasmic membrane consisting of phospholipids molecules and porins (Scheme 1). E. coli grown with DEG in LB liquid medium shows changes in the morphology of the cells ( parts c and d of Figure 4). These cells were larger and shorter than the one grown in free LB liquid medium (Figure 4c). Moreover, damage and disorganization in the cell wall were also observed (Figure 4d). In parts e and f of Figure 4, a cellular division was represented for E. coli grown in free LB liquid medium and also in DEG, respectively. In the last case, the new cells presented considerable damage with a much disorganized cell wall and affected morphology. Finally, TEM micrographs of E. coli grown in the presence of ZnO-3 sample showed preliminary results of cellular internalization of ZnO nanoparticles and cell wall disorganization (Figure 4g-j). The cell membrane in all E. coli is extensively damaged and, most probably, the intracellular content has leaked out. Figure 4g shows ZnO nanoparticles inside and outside the cell surrounded maybe by lipopolysaccharides released by the bacteria. The cellular internalization was also observed with MgO,24 CdSe@ZnS,25 and Ag26 nanoparticles. Conclusions. Our work presents the preliminary studies on toxicological impact of ZnO nanoparticles synthesized in a liquid polyol on Escherichia coli bacteria. These results showed that lower concentrations of ZnO nanoparticles did not induce any damage, as also reported by Roselli et al.23a
869

Moreover, E. coli cells after contact with ZnO nanoparticle concentrations higher than 1.3 10-3 M and with some small molecules and macromolecules in DEG medium were damaged and the cell contents may have leaked out. Cellular internalization of these nanoparticles was observed. Complementary studies are in progress including (i) the monitoring of bacterial growth in the presence of ZnO varying nanoparticle concentrations, sizes and shapes, (ii) the impact of other surfactants present in commercial products, (iii) the mechanism of cellular internalization of these nanoparticles, and (iv) conductmetric studies of cell content release. References
(1) Donaldson, K.; Stone, V.; Macnee, W. In Particulate Matter: Properties and effects upon health; Maynard, R. L., Howards, C. D., Eds.; BIOS Scientific Publishers: Oxford, 1999; p 115. (2) Donaldson, K.; Stone, V.; Clouter, A.; Renwick, L.; Macnee, W. Occup. EnViron. Med. 2001, 58, 211. (3) Donaldson, K.; Stone, V.; Gilmour, P. S.; Brown, D. M.; Macnee, W. Philos. Trans. R. Soc. London, Ser. A 2000, 358, 2741. (4) Wilson, M. R.; Lightbody, J. H.; Donaldson, K.; Sales, J.; Stone, V. Toxicol. Appl. Pharmacol. 2002, 184, 172. (5) Dunford, R.; Salinaro, A.; Cai, L.; Serpone, N.; Horikoshi, S.; Hidaka, H.; Knowland, J. FEBS Lett. 1997, 418, 87. (6) Strehlow, W. H.; Coo, E. L. J. Phys. Chem. Ref. Data 1973, 2, 163. (7) Lin, F. C.; Takao, Y.; Shimizu, Y.; Egashira, M. Sens. Actuators, B 1995, 24-25, 843. (8) Minne, S. C.; Manalis, S. R.; Quate, C. F. Appl. Phys. Lett. 1995, 67, 3918. (9) Baxter, J. B.; Aydil, E. S. Appl. Phys. Lett. 2005, 86, 53114.

(10) Hoffman, R. L.; Norris, B. J.; Wager, J. F. Appl. Phys. Lett. 2003, 82, 733. (11) Park, W. I.; Yi, G. C. AdV. Mater. 2003, 16, 1907. (12) Chen, S.; Liu, Y.; Shao, C.; Mu, R.; Lu, Y.; Zhang, J.; Shen, D.; Fan, X. AdV. Mater. 2005, 17 (5), 586. (13) Shen, G.; Cho, J. H.; Lee, C. J. Chem. Phys. Lett. 2005, 401, 414. (14) Jezequel, D.; Guenot, J.; Jouini, N.; Fievet, F. J. Mater. Res. 1995, 10 (1) 77. (15) Feldmann, C. AdV. Funct. Mater. 2003, 13 (2), 101. (16) Lin, W.; Coobes, A. G.; Davies, M. C.; Davies, S. S.; Illum, L. J. Drug Targeting 1993, 1, 237. (17) Merodio, M.; Arnedo, A.; Remedo, M. J.; Frache, J. M. Eur. J. Pharm. Sci. 2001, 12, 251. (18) Morimoto, Y.; Fujimoto, S. Crit. ReV. Ther. Drug Carrier Syst. 1985, 2, 19. (19) Rhodes, B. A.; Zolle, I.; Buchanan, J. W.; Wagner, H. N. J. Radiology 1969, 92, 1453. (20) Arshadi, R., J. Controlled Release 1990, 14, 11. (21) Zhao, X.; Hilliard, L. R.; Mechery, S. J.; Wang, Y.; Bagwe, R. P.; Jin, S.; Tan, W. Proc. Natl. Acad. Sci. U.S.A. 2004, 101 (42), 15027. (22) Ramnath, M.; Beukes, M.; Tamura, K.; Hastings, J. W. Appl. EnViron. Microbiol. 2000, 66 (7), 3098. (23) (a) Roselli, M.; Finamore, A.; Garaguso, I.; Britti, M. S.; Mengheri, E. Biochem. Molec. Actions Nutriments 2003, 4077. (b) Sharma, S.; Dastidar, M. G.; Sreekrishnan, T. R. ASCE: Pratice Period Toxic Hazard Radioact. Waste Management 2002, 6 (4), 256. (24) Stoimenov, P. K.; Klinger, R. L.; Marchin, G. L.; Klabunde, K. J. Langmuir 2002, 18, 6679. (25) Rozenzhak, S. M.; Kadakia, M. P.; Caserta, T. M.; Westbrook, T. R.; Stone, M. O.; Naik, R. Chem. Commun. 2005, 2217. (26) Sondi, I.; Salopek-Sondi, B. J. Colloid Interface Sci. 2004, 275, 177.

NL052326H

870

Nano Lett., Vol. 6, No. 4, 2006