Nanoparticles and Surfaces Presenting Antifungal, Antibacterial and

Apr 30, 2012 - Nanomedicine: Nanotechnology, Biology and Medicine 2017 13 (2), 353- .... Self-Defensive Biomaterial Coating Against Bacteria and Yeast...
19 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Nanoparticles and Surfaces Presenting Antifungal, Antibacterial and Antiviral Properties D. Botequim,† J. Maia,† M. M. F. Lino,†,* L. M. F. Lopes,‡ P. N. Simões,§ L. M. Ilharco,‡ and L. Ferreira∥,⊥ †

Matera, Núcleo 4, Lote 2, Parque tecnológico de Cantanhede, 3060-197 Cantanhede, Portugal Centro de Química-Física Molecular and IN-Institute of Nanoscience and Nanotechnology, Instituto Superior Técnico, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal § Chemical Engineering Department, University of Coimbra, Pinhal de Marrocos, 3030-290, Coimbra, Portugal ∥ CNC- Centre for Neuroscience and Cell Biology, University of Coimbra, Coimbra, Portugal ⊥ Biocant, Centro de Inovaçaõ em Biotecnologia, Cantanhede, Portugal ‡

S Supporting Information *

ABSTRACT: Here, we present new antimicrobial nanoparticles based on silica nanoparticles (SNPs) coated with a quaternary ammonium cationic surfactant, didodecyldimethylammonium bromide (DDAB). Depending on the initial concentration of DDAB, SNPs immobilize between 45 and 275 μg of DDAB per milligram of nanoparticle. For high concentrations of DDAB adsorbed to SNP, a bilayer is formed as confirmed by zeta potential measurements, thermogravimetry, and diffuse reflectance infrared Fourier transform (DRIFT) analyses. Interestingly, these nanoparticles have lower minimal inhibitory concentrations (MIC) against bacteria and fungi than soluble surfactant. The electrostatic interaction of the DDAB with the SNP is strong, since no measurable loss of antimicrobial activity was observed after suspension in aqueous solution for 60 days. We further show that the antimicrobial activity of the nanoparticle does not require the leaching of the surfactant from the surface of the NPs. The SNPs may be immobilized onto surfaces with different chemistry while maintaining their antimicrobial activity, in this case extended to a virucidal activity. The versatility, relative facility in preparation, low cost, and large antimicrobial activity of our platform makes it attractive as a coating for large surfaces.

1. INTRODUCTION The development of microbicidal coatings has received increasing attention in recent years to prevent the propagation of pathogenic microbes. Microbicidal coatings can be prepared by impregnating, adsorbing, or covalently attaching microbicidal agents to the object. Considerable research efforts have been performed in the development of nonleaching surfaces capable of killing microorganisms on contact.1−3 This is because the antimicrobial coatings can produce longer-lasting antimicrobial effectiveness, and ensure that microbes encountering the antimicrobial agent are only exposed to high surface concentrations as opposed to low ones created by slow release surfaces.3 Several methodologies have been proposed to prepare nonleaching antimicrobial coatings, but in most cases, they require a multistep procedure and are only specific for certain surface chemistries.1,4 Alternative approaches include the use of coatings that are noncovalently linked to the surface, acting as a “painting”.5,6 Hydrophobic polycations such as N-alkyl-poly(ethyleneimines) can be applied to surfaces as a “painting”, being very efficient in killing microorganisms by contact and relatively nonleaching in aqueous solution.6,7 Unfortunately, hydrophobic coatings might facilitate protein adsorption with the concomitant loss of the long-term antimicrobial activity. An © 2012 American Chemical Society

alternative approach has been recently proposed which uses amphiphilic polycations containing a catechol derivative that is known to adhere to substrates of variable chemistry.8 The polymeric coatings were effective against Escherichia coli but less effective against Staphylococcus aureus, and no data were presented about their antifungal properties. Therefore, the development of new amphiphilic polycationic coatings with improved bioactivity and a large spectrum of antimicrobial activity is desirable. To that end, we investigated the use of amphiphilic polycations adsorbed to silica nanoparticles (SNPs) as a general approach to coat surfaces. SNPs were chosen due to their noncytotoxicity, low price, high stability and durability, and ease of modification by organosilane chemistry, allowing the incorporation of an array of different functional groups.9 Although the adsorption of quaternary ammonium cationic surfactants to large silica nanoparticles (300 nm) and polystyrene nanoparticles has been described,10,11 there is no systematic work on their antimicrobial activity and their potential use as a surface coating. Received: March 5, 2012 Revised: April 20, 2012 Published: April 30, 2012 7646

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

2.4. DRIFT Analysis. Infrared analysis of lyophilized DDAB vesicles and of the bare and DDAB coated SNPs by DRIFT spectroscopy was performed using a Mattson RS1 FTIR spectrometer with a Specac Selector, in the 400−4000 cm−1 range (wide band MCT detector), at 4 cm−1 resolution. The spectra were the result of 500 coadded scans for each sample, ratioed against the same number of scans for the background (ground KBr, FTIR grade from Aldrich). The samples were previously ground and mixed with KBr in appropriate proportions to obtain spectral absorbance in the range of applicability of the Kubelka−Munk transformation.12 2.5. Determination of the Critical Micelle Concentration (CMC). The critical micelle concentration was determined for each surfactant used, following the method described elsewhere.13 The surfactants tested were as follows: dodecyltrimethylammonium bromide (DTAB, TCI), didodecyldimethylammonium bromide (DDAB), tridodecylmethylammonium chloride (TMAC, Sigma), ditetradecyldimethylammonium bromide (DTDAB, TCI), dihexadecyldimethylammonium bromide (DHDAB, TCI), and dioctadecyldimethylammonium bromide (DODAB, Sigma). Briefly, the surfactants were dissolved in PBS (pH 7.4), serially diluted, and mixed with Nphenyl-1-naphthylamine (NPN; TCI). The fluorescence of each sample was quantified (excitation at 350 nm; emission at 420 nm) and plotted against the concentration of each surfactant. The CMC was found at the intersection of two straight lines traced on the emission intensity plot, defining the aqueous and micellar environments. 2.6. Coating of Glass Coverslips with SNPs-DDAB. The coating was performed as described elsewhere.3 Basically, round glass coverslips (Ø = 12 mm) were initially cleaned by ultrasonication in successive 10 min steps in acetone, methanol/H2O (1:1), and chloroform. The coverslips were then placed in 24-well plates, one per well, immersed in 0.5 mL dopamine hydrochloride solution (2 mg mL−1 in 10 mM Tris buffer pH 8.5), shaken overnight (150 rpm orbital shaking), and finally rinsed with distilled water. The coverslips were dried and covered with a suspension of SNP8 or SNP8-DDAB250 (20 μL each, 5 mg mL−1 in 10 mM Tris buffer pH 8.5), dried, and immersed in 0.5 mL Tris buffer overnight. This process was repeated three times. The washing water was analyzed by DLS via Zeta PALS Zeta Potential Analyzer in order to assess the number of particles removed from the surface (given by the counts per second). 2.7. Antimicrobial Activity. 2.7.1. Quaternary Ammonium Surfactants. Growth kinetics of Candida albicans ATCC 10231 (C. albicans), Staphylococcus aureus ATCC 6538 (S. aureus), and Escherichia coli ATCC 25922 (E. coli) (1 × 105 cells mL−1) exposed to variable concentrations (from 62.5 to 500 μg mL−1 for fungi and from 15.6 to 500 μg mL−1 for S. aureus and E. coli) of the surfactants was evaluated by absorbance at 600 nm. The surfactants tested were as follows: DTAB, DDAB, TMAC, DTDAB, DHDAB, and DODAB. Inoculum corresponds to prokaryotic cells incubated without surfactants. The samples were prepared in a 96-well plate and incubated for 18 h at 30 °C (fungi) or 37 °C (bacteria) with constant shaking in a Biotek Synergy Mx spectrophotometer. 2.7.2. SNPs in Suspension. SNP8 or SNP80 coated with DDAB (SNP8-DDAB250 or SNP80-DDAB250) were tested in suspension against C. albicans, Aspergillus oryzae ATCC 46244 (A. oryzae; mold), Penicillium ochrochloron ATCC 9112 (P. ochrochloron; mold), S. aureus (bacteria gram-positive), and E. coli (bacteria gram-negative). Yeast Peptone Dextrose (YPD) and Tryptone Soy Yeast (TSY) media were used to culture yeast and bacteria, respectively. SNP8-DDAB250 or SNP80-DDAB250 suspended in culture medium (500 μg mL−1) were incubated with a suspension of yeast (1 mL YPD containing 1 × 105 yeast cells) or bacteria (1 mL TSY containing 1 × 106 bacteria cells) for 6 h at 30 °C (fungi) or 37 °C (bacteria) with orbital shaking (150 rpm). Then, an aliquot of the medium was serially diluted in sterile water and plated on YPD agar plates (1% yeast extract, 2% peptone, 2% dextrose, 2% agar) or TSY agar plates (3.3% tryptic soy broth, 0.3% yeast extract, 1.7% agar). Finally, the plates were incubated at 30 and 37 °C for 18 h, and the number of colony forming units (CFU) was counted and compared with the controls (SNP8 or SNP80) and the inoculum (prokaryotic cells incubated without NPs). Minimum inhibitory concentrations (MICs) were determined by the broth

Here, we show that SNPs coated with a quaternary ammonium cationic surfactant, didodecyldimethylammonium bromide (DDAB), have high antimicrobial activity against bacteria, yeast, molds, and viruses. We demonstrate that the coated SNPs can be reused without substantially losing their antimicrobial activity. Importantly, the antimicrobial activity is not due to the leaching of DDAB from the surface of the SNPs, since media that have been in contact with the DDAB-coated SNPs have no significant antimicrobial activity. We further show that these nanoparticles (NPs) can be immobilized onto flat surfaces forming an antimicrobial and antiviral coating.

2. MATERIALS AND METHODS 2.1. Physical Immobilization of Surfactants on SNPs. SNPs of 8 nm (SNP8) in diameter were kindly offered by EKA (Sweden), while SNPs of 80 nm (SNP80) in diameter were purchased from PlasmaChem GmbH (Germany). An aqueous suspension of SNP8 or SNP80 (5% w/v, in 0.1 M citrate/sodium citrate buffer pH 3.0, or distilled water or 0.1 M borate/NaOH buffer pH 9.0) was added to an aqueous solution of DDAB (50, 125, 250, and 500 μg per mg of NPs; Sigma), for variable time (in distilled water; 30 min, 1 h, or 3 h), under vigorous magnetic stirring (800 rpm) at room temperature. The suspension of NPs was then centrifuged (30 min, 40 000 rpm for SNP8; 20 min, 10 000 rpm for SNP80), washed two times with distilled water and finally freeze−dried. 2.2. Characterization of the NPs. SNPs suspended in distilled water (5 μL, 0.5 mg mL−1) were deposited on 0.5 cm2 glass slides. The solvent was allowed to evaporate and the slides mounted on a SEM sample stub using conductive carbon cement. The samples were then carbon-coated by plasma vapor deposition and analyzed by a Hitachi SU-70, with a STEM detector at 4 kV. For TEM analysis, the suspension of SNPs (10 μL, 5 mg mL−1 in PBS) was spray-coated on a TEM 400 mesh grid. The SNPs were then observed by TEM on a FEI microscope (model TECNAI G2 20 S-TWIN) at 200 kV. NP size was determined using dynamic light scattering (DLS) via Zeta PALS zeta potential analyzer and ZetaPlus Particle Sizing Software, v 2.27 (Brookhaven Instruments Corporation). An aliquot (20 μL) of NPs suspended in water (2 mg mL−1) was added to 2 mL of PBS pH 7.4, vortexed, and sonicated. All sizing measurements were performed at ca. 25 °C, and all data were recorded at 90°, with an equilibration time of 5 min and individual run times of 60 s (5 runs per measurement). The average diameters described in this work are number-weighted, and were collected from 3 independent measurements. The zeta potential of the previous NP suspensions was recorded in at least 6 runs with a relative residual value (measure of data fit quality) of 0.03. The specific surface area was assessed on an ASAP 2000 instrument, using the BET isotherm model with a relative pressure range of 0.05− 0.15. 2.3. Thermogravimetric Analysis. The thermal behavior of the samples was first evaluated by simultaneous thermal analysis (STA; heat-flux DSC and TGA), by using a TA Instruments SDT Q600 equipment (thermobalance sensitivity 0.1 μg), which was previously calibrated in the range 25−1000 °C by running tin and lead as melting standards, at a heating rate of 10 °C min−1, using open alumina crucibles and a dry nitrogen purge flow of 100 mL min−1. The mass loss measured in the temperature range from 200 to 350 °C was used to calculate the percentage of surfactant adsorbed to a certain mass of NPs. The mass loss process was further studied in detail by highresolution modulated thermogravimetric analysis (HiRes-M-TGA) in a TA Instruments Q500 thermogravimetric apparatus (thermobalance sensitivity: 0.1 μg). The temperature calibration was performed in the range 25−1000 °C by measuring the Curie point of nickel standard. Open platinum crucibles and a dry nitrogen purge flow of 100 mL min−1 were used. The experiments were performed under a dynamic rate mode with a (maximum) heating rate of 2 °C min−1, a modulation period of 200 s, and a temperature amplitude of ±5 °C. 7647

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

parafilm to spread the drop. After 30 min of incubation at room temperature, the samples were washed with 1.98 mL PBS and 2-fold serial dilutions were made. Then, confluent MDCK monolayers were washed twice with PBS and infected with 200 μL of each dilution for 1 h at room temperature. After incubation, the virus solution was aspirated and the cells were covered with agar medium (1:1 DME/F12 modified supplemented with 0.01% DEAE dextran, 0.1% NaHCO3, 4 μg mL −1 trypsin, 100 units mL −1 penicillin, 100 μg mL −1 streptomycin, and 0.6% agar) and incubated for 4 days at 37 °C in a humidified-air atmosphere containing 5% CO2. Finally, the agar overlay was removed and the cells were fixed with paraformaldehyde and stained with crystal violet (0.1% in 20% v/v aqueous methanol). The antiviral activity of the surfaces due to the leaching of NPs was evaluated by washing the coverslips with PBS (1 mL) for 6 h under orbital shaking and incubating the washing with 20 μL of virus solution for 1 h at room temperature. The number of viral colonies was determined as before.

microdilution method according to the National Committee for Clinical Laboratory Standard (for yeasts; NCCLS; M27A2E) and the Clinical and Laboratory Standards Institute (for bacteria; CLSI; M07A8) guidelines. To evaluate the antimold activity of SNPs, suspensions of spores from Aspergillus or Penicillium were prepared from grown cultures on Potato Dextrose Agar (PDA) plates at 30 °C. SNP80-DDAB250 suspended in sterile distilled water (500 μg mL−1) was incubated with a suspension of spores (1 × 105 spores mL−1, in sterile distilled water) for 6 h, at 150 rpm and 30 °C. At the end of the incubation, an aliquot of the medium was serially diluted in sterile distilled water and plated on PDA plates (3.3% potato dextrose broth, 1.1% agar). The plates were then incubated at 30 °C and CFU counted after 3 days and compared with the controls (SNP8 or SNP80) and the inoculum (spores without NPs). To evaluate the biological activity of SNP8-DDAB250 or SNP80DDAB250 against yeast, bacteria, and molds in multiple challenges, the particle suspension (3 mg mL−1) was centrifuged (20 min, 14 000 rpm) after the previous antimicrobial activity assay and the SNPs resuspended in YPD medium containing C. albicans (1 × 105 cells mL−1), TSY medium containing S. aureus or E. coli (1 × 106 cells mL−1), or distilled water containing spores (1 × 106 spores of A. oryzae or P. ochrochloron). The suspension was incubated in an orbital shaker for 6 h at 30 °C (fungi and molds) or 37 °C (bacteria) and the number of CFU was determined as described above. Growth kinetics of C. albicans, S. aureus, and E. coli (1 × 105 cells mL−1) exposed to variable concentrations of SNP8-DDAB250 or SNP80-DDAB250 (from 3.1 to 50 μg mL−1 for fungi and S. aureus and from 150 to 1000 μg mL−1 for E. coli) was evaluated by absorbance at 600 nm. Inoculum corresponds to prokaryotic cells incubated without NPs. The samples were prepared as described above. 2.7.3. Aging of NPs. SNP80-DDAB250 (1 mg mL−1) samples were maintained in 0.1 M citrate/sodium citrate buffer pH 3.0, PBS pH 7.4, 0.1 M borate/NaOH buffer pH 9.0, and YPD medium during 60 days. The buffers and the YPD medium were changed every 3 days. For that purpose, NPs were centrifuged at 14 000 rpm for 20 min, and the buffer/medium replaced by a new one. After 60 days, the NPs were centrifuged (same conditions as before), washed with distilled water, and freeze−dried before testing antifungal activity with C. albicans (1 × 105 cells mL−1). 2.7.4. SNPs Immobilized on Glass Surface. The antimicrobial activity due to the leaching of the SNPs was evaluated by washing the glass coverslips for 6 h (sterilized for 30 min under UV light) coated with SNP8 and SNP8-DDAB250 with 1 mL of YPD (at 30 °C) or TSY (at 37 °C), under orbital shaking (150 rpm). The washing medium was tested against 1 × 105 C. albicans cells, 1 × 106 E. coli or S. aureus cells. After confirming the absence of antimicrobial activity in the washing solutions, the antimicrobial activity of the surfaces was tested by adding 1 mL of YPD with 1 × 105 C. albicans cells or 1 mL of TSY with 1 × 103 E. coli or 1 × 106 S. aureus cells to each coverslip and incubating at 30 °C (fungi) or 37 °C (bacteria) for 6 h, under orbital shaking (150 rpm). An aliquot of the medium was serially diluted in sterile distilled water and plated on YPD or TSY agar plates. The number of CFU was counted after incubation of the plates at 30 °C or at 37 °C for 24 h. The remaining medium was removed and the coverslips were rinsed twice with 1 mL of sterile water. After drying rapidly, the coverslips were plated with coated side down on YPD or TSY agar plates and incubated at 30 °C or at 37 °C. 2.8. Antiviral Activity. MDCK cells obtained from the European Collection of Cell Cultures agency (ECACC), were grown in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine serum (FBS), 100 units mL−1 penicillin, and 100 μg mL−1 streptomycin, at 37 °C in a humidified-air atmosphere containing 5% CO2. Influenza A/PR/8/34 (H1N1) was obtained from Advanced Biotechnologies. The antiviral activity of the coated glass coverslips was assessed by a methodology previously reported.14 Briefly, glass coverslips (glass, glass dopamine coated with SNP8, and glass dopamine coated with SNP8-DDAB250) were placed in 24-well plates and 20 μL of virus solution was deposited on the top of each sample and covered with

3. RESULTS 3.1. Antimicrobial Activity of Quaternary Ammonium Cationic Surfactants. To evaluate the effect of the molecular structure of quaternary ammonium cationic surfactants in their antimicrobial activity (against fungi and bacteria), we used single-chained (DTAB), dichained (DDAB), and trichained (TMAC) surfactants (Figure 1A). The selected surfactants have

Figure 1. Antimicrobial activity of quaternary ammonium cationic surfactants. (A) Schematic representation of surfactants with variable number of hydrocarbon chains (DTAB, DDAB, and TMAC) (A.1) and dialkyldimethylammonium bromide surfactants with variable hydrocarbon chain length (DTDAB, DHDAB, and DODAB) (A.2). (B) Growth kinetics of yeast and bacteria exposed to surfactants. A suspension of C. albicans (B.1), S. aureus (B.2), or E. coli (B.3) (1 × 105 cells mL−1) was incubated with a certain concentration of surfactant and the absorbance monitored at 600 nm overtime. Inoculum is in black. The following concentrations (μg mL−1) have been used: C. albicans, 62.5; S. aureus, 15.6; E. coli, 62.5. Results are mean ± standard deviation (n = 4).

a hydrocarbonated chain formed by 12 carbons. We also used dialkyldimethylammonium bromide surfactants with variable chain length including 12 (DDAB), 14 (DTDAB), 16 (DHDAB), and 18 (DODAB) carbons (Figure 1B). Their effect on the growth curves of C. albicans (fungi), S. aureus (gram-positive bacteria), and E. coli (gram-negative bacteria) was assessed at variable concentrations (from 62.5 to 500 μg mL−1 for fungi and from 15.6 to 500 μg mL−1 for S. aureus and E. coli) (Figure 1C). Table 1 presents the minimal inhibitory concentrations (MICs), defined as the lowest drug concen7648

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

Table 1. Critical Micelle Concentrations (CMCs) and Minimal Inhibitory Concentrations (MICs) for the Surfactants Testeda MIC (μg mL−1) surfactant

# chains

# carbons

CMC (μg mL−1)

C. albicans

S. aureus

E. coli

DTAB DDAB TMAC DTDAB DHDAB DODAB

1 2 3 2 2 2

12 12 12 14 16 18

3515.1 32.4 246.2 7.8 5.2 -

250 125 n.d. n.d. n.d. n.d.

31.25 31.25 125 n.d. n.d. n.d.

62.5 125 n.d. n.d. n.d. n.d.

a

n.d. − not determined under the studied concentration range.

tration that results in complete inhibition of visible growth, and the critical micelle concentrations (CMCs) of the surfactants tested, defined as the concentration of surfactant at which micelles form. With regard to E. coli, the toxicity ranking of all surfactants studied in this work was DTAB > DDAB > DHDAB > TMAC ≈ DTDAB > DODAB; for C. albicans, it was DDAB > DHDAB > DTDAB > TMAC > DTAB > DODAB; and finally, for S. aureus it was DDAB > DTAB > DHDAB > DODAB ≈ DTDAB > TMAC. DDAB is the most effective in the inhibition of C. albicans (125 μg mL−1) and S. aureus (31.3 μg mL−1), while DTAB is the most effective in the inhibition of E. coli (125 μg mL−1). Our results also indicate that the antimicrobial activity is higher for the surfactant having hydrocarbon dichains formed by 12 carbons than with 14, 16, or 18 chains. On the basis of these results, DDAB was used for subsequent tests. 3.2. Characterization of SNPs Coated with DDAB. Two types of SNPs were used in this work: SNPs with a diameter of 78.9 ± 12.6 nm by DLS (Table 2), corresponding to a diameter

Figure 2. Characterization of SNPs. (A) TEM and SEM micrographs of SNP8 and SNP80, respectively. Scale bars correspond to 20 nm for SNP8 and 100 nm for SNP80. (B) DRIFT analysis of SNP8 and SNP80.

of four Gaussian functions by a nonlinear least-squares fitting method, assuming that the elementary SiO4 units are arranged mostly in (SiO)4 and (SiO)6 siloxane rings15 and that in each type of ring this mode is split into a pair of optic components (longitudinal, LO, and transverse, TO). In both SNPs, the 4rings are predominant (their relative area is 64% in SNP8 and 66% in SNP80), as is usual for silica gels. However, the LO− TO splitting is smaller for both types of rings in SNP80, which is associated with a higher porosity16 and confirms the density measurements. The deconvolution results are shown in Table S2, Supporting Information. Furthermore, the SNP8 are considerably more hydrophilic, since their spectrum evidence strong νSi−O (dangling from broken siloxane bridges or in silanol groups) and νO−H bands, at 960 and centered at 3300 cm−1, respectively, and also some adsorbed water (δHOH band, at 1630 cm−1). These bands are extremely weak (νO−H and δHOH) or even absent (νSi−Od) in the spectrum of SNP80. Therefore, the SNP80 have a considerably lower content in surface hydroxyl groups. Finally, the broad νO−H band for SNP8 also indicates that the hydroxyl and/or silanol groups are involved in a variety of hydrogen bonds, suggesting that some aggregation between particles may occur. These features are unlikely to occur in SNP80. Both samples have a small proportion of free OH groups, which are more significant in SNP80. To identify the best conditions to physically adsorb DDAB onto SNPs, the NPs were mixed with aqueous solutions of DDAB at different concentrations and pHs, over different time frames. Concentrations of DDAB above its CMC (32.4 μg mL−1) (Table 1) were tested (i.e., 200, 500, 1000, and 2000 μg mL−1 for SNPs-DDAB50, SNPs-DDAB125, SNPs-DDAB250, and SNPs-DDAB500, respectively). The concentrations used were 50, 125, 250, and 500 μg of DDAB per mg of SNPs (suspension of 5%, w/v) and per 0.293 m2 or 0.218 m2 in the case of SNP8 and SNP80, respectively (taking into account the surface areas of 293.0 ± 3.4 m2 g−1 and 218.5 ± 68 m2 g−1 for SNP8 and SNP80, respectively). SNPs with an average diameter of 80 nm (SNP80) were used in most of the tests,

Table 2. Sizes and Antimicrobial Activity of SNPs Coated with DDAB MIC (μg mL−1)b SNPs SNP8 SNP8-DDAB250 SNP80 SNP80DDAB250

size (nm)a

C. albicans

S. aureus

E. coli

± ± ± ±

25 (4.5) 50 (9.7)

12.5 (2.2) 25 (4.9)

500 (89.1) >1000 (194.4)

7.5 34.6 78.9 98.5

1.7 13.5 12.6 6.0

Data corresponds to the average ± standard deviation of three independent measurements by DLS. bThe value in parentheses is the amount of DDAB (μg) in the SNPs. a

of 58.3 ± 3.8 by SEM (Figure 2A), henceforth named as SNP80; and SNPs with a diameter of 7.5 ± 1.7 nm by DLS (Table 2), corresponding to a diameter of 5.0 ± 2.0 by TEM (Figure 2A), designated as SNP8. SNP80 have a surface area of 218.5 ± 68 m2g−1 (as measured by BET analysis) and a density of 1.20 g cm−3 (as measured by helium pycnometry), while SNP8 have a surface area of 293.0 ± 3.4 m2g−1 and a density of 2.18 g cm−3. DRIFT results show that SNP80 and SNP8 have different structures (Figure 2B and Table S1 in Supporting Information): the main silica band (νasSi−O−Si, at ∼1100 cm−1) is shifted ∼12 cm−1 to higher wavenumbers for SNP80 and is much narrower (FWHH ∼173 cm−1, versus ∼207 cm−1 for SNP8). The νasSi−O−Si band was decomposed into a sum 7649

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

and in the interactions of DDAB chains among them and with the silica nanoparticles. According to Figure 4A, as the concentration of DDAB to SNP80 increased, the zeta potential of NPs became positive,

since they were relatively easier to process than SNP8, which tend to aggregate, as suggested by the DRIFT spectra (Figure 2B). After the adsorption period, the suspension of the NPs was centrifuged. Importantly, aqueous solutions of DDAB (therefore without SNPs, forming vesicles with an average of 57 nm and a zeta potential of 17.11 ± 2.15 mV), centrifuged at the same velocities, did not form a pellet. The adsorption of DDAB to SNPs was monitored by conventional and high-resolution modulated TGA (Figure 3).

Figure 4. Influence of concentration, pH, and time in the physical immobilization of DDAB onto SNPs. (A) Influence of pH in the physical immobilization of DDAB onto SNP80. Size and zeta potentials of the NPs were evaluated. (B) Influence of NP size in the physical immobilization of DDAB. (C) Influence of time in the physical immobilization of DDAB onto SNPs. (D) Antifungal activity of SNP80 coated with DDAB (initial coating conditions: 250 μg of DDAB per mg of NPs). SNP80-DDAB250 (500 μg mL−1) was incubated with C. albicans (1 × 105 cells mL−1) for 6 h at 30 °C, and then an aliquot of the medium serially diluted in water and plated in YPD agar plates to assess their survival. Inoculum corresponds to cells suspended in YPD medium without NPs. Results are mean ± standard deviation (size: n = 3; ζ potential: n = 6; antifungal activity: n = 4).

Figure 3. Quantification of DDAB physically immobilized onto SNPs. (A) The amount of DDAB physically immobilized onto SNP80 and SNP8 was assessed by HiRes-M-TGA analysis. TGA curves are as follows: (1) SNP8 or SNP80 without DDAB; (2−5) SNP8 or SNP80 coated with an initial concentration of DDAB of (2) 50 μg, (3) 125 μg, (4) 250 μg, or (5) 500 μg per mg of NP; (6) DDAB. (B) Derivatives of the TGA plots of SNP80-DDAB and SNP8-DDAB samples. The numbers 1, 2, and 3 indicate the three main peaks.

Depending on the initial concentration of DDAB, SNP8 immobilizes between 57 and 210 μg of DDAB per mg of SNP, while SNP80 immobilizes between 45 and 275 μg of the conjugate per mg of SNP (Figure 3A). The derivatives of the TGA plots of SNP80-DDAB samples indicate the presence of three major peaks at 160−170 °C, 190−200 °C, and 210−230 °C (Figure 3B). The first peak at 160−170 °C is attributed to desorption of monomeric DDAB molecules, whose hydrocarbon chains are partially inserted in the bilayer; the second peak at 190−200 °C is ascribed to desorption of the DDAB molecules in the outer layer of the DDAB bilayer coated on SNPs; finally, the third peak at 210−230 °C is related to desorption of the DDAB molecules bonded to the SNPs.17,18 Monomeric DDAB adsorbed to the bilayer (first peak) is only present in SNP80-DDAB250 and SNP80-DDAB500 samples. For the initial DDAB concentrations of 50 and 125 μg mg−1 silica, the amount of surfactant physically adsorbed onto SNP8 is higher than on SNP80, while the opposite was observed when an initial concentration of DDAB of 500 μg was used per mg of silica. The derivatives of the TGA plots in SNP8-DDAB indicate the presence of two major peaks: the first one between 150 and 195 °C, and the second one between 218 and 255 °C (Figure 3B). The first peak is likely attributed to desorption of DDAB molecules in the outer layer of the bilayer coated on SNPs, while the second peak is due to the desorption of DDAB bonded to the SNPs. The differences between the temperatures of the peaks in SNP8-DDAB and SNP80-DDAB samples are likely due to differences in the curvature of the nanoparticles

peaking for concentration of 250 μg of DDAB per mg of NPs. Similar zeta potential results were obtained for SNP8 (Figure 4B). In the case of SNP80, the increase in the zeta potential was followed by an increase in the average diameter of the NPs due to the adsorption of the surfactant (Figure 4A). Next, we evaluated the effect of time and pH in the adsorption of DDAB to SNP80 (Figure 4A,C). No significant differences in terms of zeta potential and size were found for all NP formulations obtained from adsorption times above 30 min and at different pHs. Importantly, NP formulations obtained at different pHs and adsorption times were fungicidal (Figure 4D). Therefore, for subsequent experiments, an adsorption time of 3 h and distilled water (pH 7.6) were used. To further characterize the adsorption of DDAB to SNPs we performed DRIFT analysis (Figure 5, Figures S1 and S2 and Table S1 in Supporting Information). The maximum of the νO−H band shifts considerably to lower wavenumbers upon DDAB adsorption and as DDAB concentration increases up to 250 μg mg−1 silica, indicating that the fraction of the more interacting hydroxyl groups increases with DDAB adsorption. This effect is clearer for SNP80 and suggests the involvement of OH groups as anchor sites for DDAB. However, the shift becomes very small when increasing the DDAB load to 500 μg mg−1 silica, which shows that further modifications at the SNP surface are negligible (Figure 5A). The two deformation modes assigned to methyl and methylene groups bonded to N+ (at 7650

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

larger particles keep their silica structure, in the small SNP8 the νasSi−O−Si becomes narrower. Its quantitative analysis (Table S2 in Supporting Information) showed that the populations of fourfold and sixfold rings invert, i.e., part of the tensioned siloxane bonds in (SiO)4 rings are broken due to the electrostatic interactions with DDAB, converting into (SiO)6 units. When the DDAB load is very high (500 μg mg−1 silica), this tendency reverts, suggesting that part of the vesicles assemble around aggregated silica particles. In Figure 5D, we observe a significant shift of the νasCH2 to a higher wavenumber for SNP8-DDAB50 and SNP80-DDAB50 when compared to DDAB vesicles (from 2920 to 2927 and 2928 cm−1, respectively) and a more subtle shift for the νsCH2 band (from 2852 to 2856 and 2858 cm−1, respectively). These shifts decrease when increasing the DDAB load, and become within the spectral resolution for samples SNP8-DDAB500 and SNP80-DDAB500. It is known that the symmetric and antisymmetric modes of the methylene groups of DDAB are used as indicators of the ordering of the alkyl chains, higher frequencies corresponding to a higher density of gauche defects.20 The observed spectral changes show that the alkyl chains of adsorbed DDAB have more gauche conformations (i.e., form a more disordered layer) for lower loads, becoming more ordered (all-trans) as coverage increases, probably stabilized by interchain dispersion interactions. The relative intensity of the νsCH2 band (2854 cm−1) is smaller for those DDAB chains with more gauche defects adsorbed on SNPs. The relative intensity of the νasCH3 band (∼2953 cm−1) is higher for all the coated SNPs than for DDAB vesicles, but it decreases as the surfactant concentration increases, indicating that the terminal methyl groups lose some freedom as the first layer becomes denser and the bilayer is formed.21 There is one exception for sample SNP8-DDAB500, which is consistent with some vesicles forming at the surface of aggregated silica particles, resulting in less confined terminal methyl groups. Taking into account that the SNP8 are much richer in OH groups, a higher adsorption yield would be expected for low DDAB concentrations, while the first layer is being formed (up to 125 μg mg−1 silica). This was confirmed by the TGA results and also by DRIFT, comparing the relative intensities of the CH2 and CH3 deformation bands versus the main silica band. On the basis of the concentrations of DDAB adsorbed to SNPs and on the efficiency of the adsorption process, the ratio 250 μg of DDAB per mg of NPs was selected (SNP-DDAB250) for further experiments. 3.3. SNPs Coated with DDAB Have Antimicrobial Activity and Can Be Reused. To evaluate the antimicrobial activity spectrum, SNPs coated with DDAB (SNP80-DDAB250) (3 mg mL−1) were added to (i) YPD media containing 1 × 105 cells of C. albicans, (ii) TSY medium containing 1 × 106 cells of E. coli or S. aureus, or (iii) water containing 1 × 105 cells mL−1 of A. oryzae or P. ochrochloron spores. After 6 h of exposure, an aliquot of the cell suspension was plated on agar plates for 18 h and the number of yeast, bacteria, or mold colonies was counted (Figure 6A). SNP80 without physically adsorbed DDAB showed no significant antifungal activity. In contrast, SNPs incorporating DDAB killed 100% of the microorganisms. In addition, SNP80 modified with DDAB can be reused for 5 times (maximum number of rounds tested) without losing their activity against C. albicans, S. aureus, A. oryzae, and P. ochrochloron. SNP80-DDAB250 lost their activity against E. coli at the third round (Figure 6A).

Figure 5. DRIFT spectra of bare and DDAB coated SNPs: SNPs (black), DDAB vesicles (pink), SNPs-DDAB50 (red), SNPs-DDAB125 (green), SNPs-DDAB250 (blue), and SNPs-DDAB500 (gray). (A) O−H stretching region, normalized to the main silica band; (B) CH deformation region, normalized to the δscCH2 mode; (C) silica fingerprint region; (D) C−H stretching region, normalized to the νasCH2 band.

1408 and 1398 cm−1 in DDAB vesicles) appear shifted to higher and lower frequencies, respectively (Figure 5B). This perturbation must be induced by the interaction of the polar head with the silica surface,19 confirmed by the changes observed in the C−N+ stretching modes in samples SNP80 (the four bands observed at 974, 966, 941, and 920 cm−1 in DDAB appear as just one shoulder at 980 to 966 cm−1 when it adsorbs on SNP80) (Figure 5C). The relative intensities of the δscCH2(N+) and δasCH3(N+) bands with respect to the δscCH2(chain), at 1468 cm−1, decrease with surface coverage as a consequence of some inhibition due to the increasing density of neighbor heads. Figure 5C shows an interesting difference between SNP8 and SNP80 upon coating: while the 7651

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

antimicrobial activity of the NPs is mainly mediated by contact (Figure 6B). 3.4. SNPs Coated with DDAB Can Maintain Their Antimicrobial Activity after Aging in Aqueous Solutions and Culture Medium. The antimicrobial activity of SNP80DDAB250 was evaluated after aging in 0.1 M citrate/sodium citrate buffer pH 3.0, PBS pH 7.4, 0.1 M borate/NaOH buffer pH 9.0, and YPD medium for 60 days at 30 °C (Figure 6C). During this time, the buffer or medium was replaced every 3 days. After the 60 days, the NPs were washed and resuspended in medium containing 1 × 105 cells/mL of C. albicans. After an incubation of 6 h, an aliquot of the medium was serially diluted in distilled water and plated in agar to assess the number of microorganisms. SNP80-DDAB250 killed 100% of C. albicans, while SNP80 without DDAB showed no antifungal activity (Figure 6C). 3.5. Microorganisms Are Rapidly Killed by SNPs Coated with DDAB. Yeast (C. albicans), bacteria (E. coli and S. aureus), and molds (A. oryzae or P. ochrochloron) were rapidly killed by SNPs coated with DDAB. The antimicrobial activity was evaluated as before. No viable microorganism was found in the medium after 2 h (Figure 7A). S. aureus and molds

Figure 6. Effect of reuse and aging in the antimicrobial activity of SNPs coated with DDAB. (A) Antimicrobial activity of SNP80DDAB250 after several cycles of reuse. SNP80-DDAB250 (3 mg mL−1) was incubated with C. albicans (1 × 105 cells mL−1), E. coli (1 × 106 cells mL−1), S. aureus (1 × 106 cells mL−1), P. ochrochloron (1 × 105 spores mL−1), or A. oryzae (1 × 105 spores mL−1) for 6 h at 30 °C (C. albicans, A. oryzae, and P. ochrochloron) or 37 °C (E. coli and S. aureus). The NPs were then centrifuged, and an aliquot of the medium serially diluted in water and plated in agar plates to assess microorganism survival. Subsequently, NPs were ressuspended in medium containing an inoculum of a specific microorganism and a new run initiated. (B) SNPs coated with DDAB kill microorganisms mainly by contact. After each activity assay, the supernatant was collected and tested against microorganisms. (C) Effect of aging in the antimicrobial activity of SNP80-DDAB250. After 60 days of aging, SNP80 and SNP80-DDAB250 (1 mg mL−1) were incubated with a suspension of C. albicans (1 × 105 cells mL−1) for 6 h at 30 °C, and the survival of the microorganisms evaluated in agar plates. In all graphs, black and white columns correspond to survival of microorganisms incubated with SNP80 and SNP80-DDAB250, respectively. Results are mean ± standard deviation (n = 4).

Figure 7. Time course of microbial killing by SNP80-DDAB250 and SNP8-DDAB250. C. albicans (●, 1 × 105 cells mL−1), E. coli (▲, 1 × 106 cells mL−1), S. aureus (▼, 1 × 106 cells mL−1), P. ochrochloron (▽, 1 × 105 spores mL−1), or A. oryzae (△, 1 × 105 spores mL−1) were incubated with SNP80-DDAB250 (A and B, in black) or SNP8DDAB250 (B, in red) (500 μg mL−1) at 30 °C (C. albicans, A. oryzae, and P. ochrochloron) or 37 °C (E. coli and S. aureus). Inoculum (■) corresponds to a cell suspension without NPs. Data corresponds to mean ± standard deviation (n = 6).

The MIC for SNP8-DDAB250 against C. albicans, S. aureus, and E. coli was 25, 12.5, and 500 μg mL−1, respectively, while for SNP80-DDAB250, it was 50, 25, and above 1000 μg mL−1 (Table 2). At these concentrations, coated SNP8 formulations have 4.5, 2.2, and 89.1 μg of DDAB, while SNP80 have 9.7, 4.9, and 194.4 μg of DDAB, respectively (Table 2). Importantly, with the exception of MIC values obtained against E. coli, the MIC values of SNPs coated with DDAB are much below the MIC of DDAB solutions (C. albicans: 125 μg mL−1; S. aureus: 31.25 μg mL−1; E. coli: 125 μg mL−1). To assess the relative contributions from surfactant release and NP killing, SNP80-DDAB250 were incubated in YPD or TSY mediums or distilled water for 6 h, centrifuged, and the supernatant collected and tested against microorganisms. Limited killing was observed (below 25%) suggesting that

were the most susceptible to the effect of SNP80-DDAB250 followed by C. albicans and E. coli, respectively. Importantly, SNP8-DDAB250 kills E. coli more rapidly than SNP80-DDAB250 (Figure 7B). 3.6. SNPs Coated with DDAB Can Be Immobilized onto Surfaces. Encouraged by the antimicrobial activity of the SNPs coated with DDAB, we immobilized them on top of glass 7652

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

using polydopamine.3,22 Dopamine was polymerized on top of a coverslip (1.13 cm2) for ∼18 h at pH 8.5 followed by the coating with SNP8-DDAB250. We were able to immobilize 222.8 and 243.9 μg cm−2 of SNP8 and SNP8-DDAB250, respectively. Because no amine or sulfhydryl groups exist in SNP8-DDAB250, van der Waals interactions likely mediate the immobilization of NPs to the polydopamine-coated surface.22,23 To confirm that the NPs were permanently bound to the surface, the coverslip was washed in YPD or TSY medium for 6 h under agitation (150 rpm), and the washing medium was tested against fungi and bacteria. No measurable antimicrobial activity was detected (Figure 8A). Then, the coverslips were

activity of glass coverslips coated with SNP8-DDAB250 against inf luenza A/PR/8/34 (H1N1) virus. Initially, the antiviral activity of the surfaces due to the leaching of NPs was evaluated by washing the coverslips with PBS for 6 h, and the washing solution tested against a virus suspension for 30 min at room temperature. No measurable antiviral activity was observed (Figure 9). Then, a virus suspension was added to the surface

Figure 9. Antiviral activity of glass coverslips coated with SNPs coated with DDAB. Coverslips were washed in PBS for 6 h under agitation and the washing medium incubated for 30 min with inf luenza A virus. Then, the coverslips were also incubated for 30 min with inf luenza A virus to assess their antiviral activity. Results correspond to mean ± standard deviation (n = 6).

of the coverslips followed by incubation at room temperature for 30 min. Virus survival was evaluated by their ability to infect MDCK cells. According to Figure 9, no survival was observed after virus exposure to coverslips coated with SNP8-DDAB250 relative to the control (coverslips coated with SNP8).

4. DISCUSSION This work reports a methodology to create nonleaching antimicrobial and antiviral surfaces. Our method comprises three steps:3 (i) adsorption of DDAB to SNPs, (ii) treatment of the substrate surface with an adhesive layer, and (iii) coating of the substrate surface with the SNPs coated with DDAB. This method allows creating durable antibacterial, antifungal, and antiviral nanocoatings in any material surface, independent of the nature of the bulk material. From all the surfactants tested in this work, DDAB was the surfactant with high antimicrobial activity against C. albicans and S. aureus. Previous studies have compared the antimicrobial activity of anionic, cationic, and zwitterionic surfactants and showed that cationic surfactants had higher activity against Gram-negative bacteria (E. coli, Pseudomonas aeruginosa), Gram-positive bacteria (Streptocococus agalactiae), and fungi (C. albicans) than the other ones.24 Furthermore, some studies have shown that dichained cationic surfactants have higher ability to penetrate cell model membranes than monochained or trichained cationic surfactants.11 Therefore, our results confirm these findings. Interestingly, few studies have reported the use of DDAB as an antimicrobial agent.25 Adsorption of surfactants at the solid/aqueous solution interface has been extensively studied to understand the interactions between the surfactant and the solid surface, including SNPs. For example, studies from adsorption of quaternary ammonium cationic surfactants with one, two, and three alkyl chains on SNPs showed that the adsorbed amounts at saturation decrease with increasing chain number of the surfactants.10 The adsorbed amount of surfactant is also dependent on the surface area of the NP. For example, the

Figure 8. Antimicrobial activity of glass coverslips coated with SNP8DDAB250. (A) Coverslips were washed in YPD or TSY medium for 6 h under agitation and the washing medium was tested against C. albicans (1 × 105 cells mL−1), S. aureus (1 × 106 cells mL−1), and E. coli (1 × 103 cells mL−1). Then, the coverslips were incubated for 6 h with fungi and bacteria and cell survival in the medium assessed by plating in YPD agar. Counts are normalized relatively to the control and expressed as mean ± standard deviation (n = 4). (B) Photographs of coverslips coated with SNP8 and SNP8-DDAB250 plated on YPD or TSY agar after incubation in medium with C. albicans (1 × 105 cells mL−1), S. aureus (1 × 106 cells mL−1), or E. coli (1 × 103 cells mL−1) for 6 h.

incubated for 6 h with C. albicans (1 × 105), E. coli (1 × 103), or S. aureus cells (1 × 106) to assess their antimicrobial activity (Figure 8A). A 100% reduction in C. albicans, E. coli, and S. aureus was observed in the media containing the coverslips coated with SNP8-DDAB250 relative to the control (coverslips coated with SNP8). Finally, the remaining medium was removed and the coverslips were rinsed with sterile water to remove nonadherent cells and plated upside down on YPD or TSY agar plates. After 24 h, no C. albicans, E. coli, or S. aureus colonies were observed on the coverslips coated with SNP8DDAB250, whereas fungi and bacteria colonized the control coverslips coated with SNP8 (Figure 8B). Overall, our results show that glass coverslips coated with SNP8-DDAB250 have antimicrobial activity against C. albicans, E. coli, and S. aureus. 3.7. SNPs Coated with DDAB and Immobilized onto Substrate Surfaces Are Antiviral. Next, we evaluated the 7653

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

temperatures observed for desorption of the bilayer in SNP8DDAB compared to SNP80-DDAB. Furthermore, our results indicate that, for an initial concentration of 500 μg of DDAB per mg of silica, SNP80 adsorbed more DDAB than SNP8. This high adsorption capacity of larger SNP80 is linked to their ability to retain additional monomeric DDAB molecules by partial insertion on the outer layer of DDAB bilayer. Our FTIR results indicate that the packing of DDAB at the surface of the SNPs depends on the initial DDAB ratio to SNPs. The progressive shift observed in the asymmetric and symmetric stretching modes of methylene groups when increasing the DDAB load on SNP80 indicate that a higher coverage of DDAB restricts the freedom of movement of the methylene chains that acquire more ordered conformations as coverage increases and as the bilayer is formed. Although NPs coated with surfactants have been used for the delivery of DNA, proteins, and small molecules, their use as antimicrobial agents has not been explored.30 Our results show that SNPs with low diameter (∼34.6 nm; Table 2) have lower MIC than SNPs with high diameter (∼98.5 nm; Table 2). At the moment, it is unclear whether the differences observed are due to differences in the concentration of adsorbed surfactant on top of NPs, differences in the charge density (Figure 4B), or differences in the interaction of the coated NPs with biological membranes. An important finding of our work is that SNP-DDAB formulations have a rapid and broad antibacterial, antifungal, and antiviral activity. SNP-DDAB formulations were effective against C. albicans (fungi), S. aureus (Gram-positive bacteria), E. coli (Gram-negative bacteria), A. oryzae (mold), P. ochrochloron (mold), and inf luenza A/PR/8/34 (H1N1; virus). These microorganisms were selected as representative examples for each category. Another important finding of our work is related to the demonstration that SNP-DDAB has a lower MIC than the one obtained for DDAB in solution. The MIC of SNP8DDAB250 against C. albicans, S. aureus, and E. coli was 4.5, 2.2, and 89.1 μg mL−1 (these are amounts on top of the SNPs), while the MIC for DDAB in solution was 250, 31.5, and 125 μg mL−1. To the best of our knowledge, this effect has not been previously reported. This may be due to differences in the size, charge density, and stiffness of both formulations, which may have an impact on their interaction with biological systems. Work is in progress to clarify this issue. Importantly, we demonstrate that SNPs coated with DDAB and immobilized on top of a substrate by a chemical adhesive maintains their antibacterial, antifungal, and antiviral activities without leaching. Our results suggest that the antimicrobial activity of the NP formulation is mainly mediated by contact. Our technology has several advantages: first, it will produce longer-lasting antibacterial, antifungal, and antiviral effectiveness; and second, it will ensure that microbes encountering the surfactant are exposed to only high surface concentrations. Very few studies have reported materials or surfaces with antiviral properties.14,31,32 Recent advances indicate that hydrophobic polycations render surfaces permanently antiviral. The mechanism of inactivation of influenza virus by polycations has been shown recently.32 The viruses adhere to hydrophobic polycationic surfaces followed by structural damage and inactivation. Subsequently, viral RNA is released into solution while proteins remain adsorbed. In the same work, the authors showed that a surfactant similar to DDAB, dodecyltrimethylammonium bromide (DTAB), had antiviral activity because of its hydrophobic quaternary ammonium monomeric unity. A

adsorption of DDAB in SNPs with a diameter of 300 nm has been previously reported.10,18 The maximum concentration of DDAB adsorbed to the SNPs was 60 μmol of DDAB per gram of silica (i.e., 28 μg per mg of SNP). This is clearly lower than the values described in this work. Our results show that SNP8 could immobilize up to 210 μg of DDAB per milligram of NPs. This is likely due to the high surface area (∼300 m2 g−1 for SNP8 and ∼220 m2 g−1 for SNP80) of our SNPs compared to the previous ones (16.7 m2 g−1) or to a higher concentration of surface anchor sites. The range of times (0.5, 1, and 3 h) tested in this work had little effect in the adsorption level of DDAB. This agrees with other studies using cationic surfactants. For example, dioctadecyldimethylammonium bromide surfactant is rapidly adsorbed (below 1 h incubation period) by SNPs with an average diameter of 50 nm (surface area of 26 m2 g−1).26 In addition, the range of pH values (3.0, 7.6, and 9.0) tested in this work had little effect on the adsorption level of DDAB. This is not surprising since the range of pH values tested are above the isoelectric point of silica (∼2.5),27 and consequently, the silica surface is negatively charged, which favors the formation of ion pairs between ammonium ions from DDAB and silanol groups from SNPs. Previous studies are not conclusive about the effect of pH in the adsorption of cationic surfactants. For example, the adsorption of dioctadecyldimethylammonium bromide to SNPs showed that the nature of the buffer was more important than its pH for the adsorption of the surfactant.26,28 Previous results have shown that DDAB formed vesicle-like structures in NPs and the layers were highly compacted, as confirmed by TGA, FTIR, fluorescence, and electron spin resonance (ESR) studies.10,18 Our zeta potential results show that SNP-DDAB250 and SNP-DDAB500 have a strong positive charge (between 20 and 30 mV), which indicates that the surfactant polar heads are at the top of the NP, which in turn suggests the formation of a bilayer of DDAB. The zeta potentials for the formulations SNP-DDAB50 (−15 mV) and SNP-DDAB125 (0 mV) indicate that the majority of the SNPs do not have a bilayer of DDAB. TGA results also indicate that most of the SNPs in the formulations SNP-DDAB250 and SNPDDAB500 had a bilayer of DDAB. This is confirmed by the TGA curves showing similar values of temperature for the second and third peaks, corresponding to desorption of the outer and inner layers of the bilayer, respectively, and similar enthalpy (areas of the peaks). Our TGA results also indicate that most of the SNPs in the formulations SNP-DDAB125 and SNP-DDAB50 do not have a bilayer of DDAB (corroborating the zeta potential results). It should be noted that the TGA profiles in this work are similar to those previously reported for gold nanoparticles coated with DDAB although with differences in the temperature ranges.18 DDAB-coated gold nanoparticles (4−6 nm in diameter) presented three major weight losses with peaks at 222, 259, and 291 °C. The differences in temperatures are likely due to differences in packing and energy of adsorption of the surfactant to the NPs. Recent studies on the packing of lipids on SNPs indicate an increased interdigitation and increased lateral packing order between the chains with decreasing NP size, which improves hydrophobic association and decreases the voids that occur in normal bilayers.29 Given the small dimension and high content of silanol groups of SNP8, a stronger adsorption of DDAB to the silica surface is achieved, as well as more ordered and denser layers of surfactant, which enhances the lateral interactions between chains. This may explain the increased 7654

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

pathogenic bacteria. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 17667− 71. (8) Han, H.; Wu, J.; Avery, C. W.; Mizutani, M.; Jiang, X.; Kamigaito, M.; Chen, Z.; Xi, C.; Kuroda, K. Immobilization of amphiphilic polycations by catechol functionality for antimicrobial coatings. Langmuir 2011, 27, 4010−9. (9) Slowing, I. I.; Trewyn, B. G.; Giri, S.; Lin, V. S. Y. Mesoporous silica nanoparticles for drug delivery and biosensing applications. Adv. Funct. Mater. 2007, 17, 1225−1236. (10) Esumi, K.; Matoba, M.; Yamanaka, Y. Characterization of Adsorption of Quaternary Ammonium Cationic Surfactants and Their Adsolubilization Behaviors on Silica. Langmuir 1996, 12, 2130−2135. (11) Peetla, C.; Labhasetwar, V. Effect of molecular structure of cationic surfactants on biophysical interactions of surfactant-modified nanoparticles with a model membrane and cellular uptake. Langmuir 2009, 25, 2369−77. (12) Kubelka, P. a. F. M. Ein Beitrag zur Optik der Farbanstriche. Z. Technol. Phys. 1931, 10, 593−601. (13) Brito, R. M.; Vaz, W. L. Determination of the critical micelle concentration of surfactants using the fluorescent probe N-phenyl-1naphthylamine. Anal. Biochem. 1986, 152, 250−5. (14) Haldar, J.; Weight, A. K.; Klibanov, A. M. Preparation, application and testing of permanent antibacterial and antiviral coatings. Nat. Protoc. 2007, 2, 2412−7. (15) Fidalgo, A. a. L. M. I. Chemical Tailoring of Porous Silica Xerogels: Local Structure by Vibrational Spectroscopy. Chem.Eur. J. 2004, 10, 392−398. (16) Fidalgo, A. R., M. E.; Ilharco, L. M. Chemical Control of Highly Porous Silica Xerogels: Physical Properties and Morphology. Chem. Mater. 2003, 15, 2186−2192. (17) Nikoobakht, B.; El-Sayed, M. A. Evidence for bilayer assembly of cationic surfactants on the surface of gold nanorods. Langmuir 2001, 17, 6368−6374. (18) Zhang, L.; Sun, X.; Song, Y.; Jiang, X.; Dong, S.; Wang, E. Didodecyldimethylammonium bromide lipid bilayer-protected gold nanoparticles: synthesis, characterization, and self-assembly. Langmuir 2006, 22, 2838−43. (19) Hostetler, M. J.; Stokes, J. J.; Murray, R. W. Infrared Spectroscopy of Three-Dimensional Self-Assembled Monolayers: NAlkanethiolate Monolayers on Gold Cluster Compounds. Langmuir 1996, 12, 3604−3612. (20) Kung, K. H. S.; Hayes, K. F. Fourier-Transform Infrared Spectroscopic Study of the Adsorption of Cetyltrimethylammonium Bromide and Cetylpyridinium Chloride on Silica. Langmuir 1993, 9, 263−267. (21) Cheng, W.; Dong, S.; Wang, E. Synthesis and Self-Assembly of Cetyltrimethylammonium Bromide-Capped Gold Nanoparticles. Langmuir 2003, 19, 9434−9439. (22) Lee, H.; Rho, J.; Messersmith, P. B. Facile Conjugation of Biomolecules onto Surfaces via Mussel Adhesive Protein Inspired Coatings. Adv. Mater. 2009, 21, 431−434. (23) Lee, H.; Scherer, N. F.; Messersmith, P. B. Single-molecule mechanics of mussel adhesion. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 12999−3003. (24) Vieira, O. V.; Hartmann, D. O.; Cardoso, C. M. P.; Oberdoerfer, D.; Baptista, M.; Santos, M. A. S.; Almeida, L.; Ramalho-Santos, J.; Vaz, W. L. C. Surfactants as Microbicides and Contraceptive Agents: A Systematic In Vitro Study. PLOS One 2008, 3. (25) Mechin, L.; Dubois-Brissonnet, F.; Heyd, B.; Leveau, J. Y. Adaptation of Pseudomonas aeruginosa ATCC 15442 to didecyldimethylammonium bromide induces changes in membrane fatty acid composition and in resistance of cells. J. Appl. Microbiol. 1999, 86, 859−66. (26) Rapuano, R.; Carmona-Ribeiro, A. M. Supported bilayers on silica. J. Colloid Interface Sci. 2000, 226, 299−307. (27) Wu, Z. Surface properties of submicrometer silica spheres modified with aminopropyltriethoxysilane and phenyltriethoxysilane. J. Colloid Interface Sci. 2006, 304, 119−124.

similar antiviral mechanism is anticipated for SNP-DDAB immobilized onto glass surfaces.

5. CONCLUSIONS Our results show that SNPs coated with a quaternary ammonium cationic surfactant are very effective against bacteria, molds, yeast, and inf luenza A − H1N1 virus, and can be reused several times without significantly losing their antimicrobial activity. Importantly, either in suspension or as a coating, the antimicrobial and antiviral activities are not due to the leaching of surfactant from the surface of the SNPs. The nanoparticles can be potentially attached to any substrate, by a chemical adhesive formed by polydopamine, forming antimicrobial and antiviral coatings. The platform described herein may offer an alternative to current technologies to render objects with antimicrobial properties.

■

ASSOCIATED CONTENT

S Supporting Information *

Table S1: Proposed assignments for the DRIFT spectra of DDAB vesicles, bare and DDAB coated SNPs. Table S2: Results of the νasSi−O−Si band deconvolution in Gaussian components for bare SNPs and coated with DDAB. Figure S1: DRIFT spectra of bare and DDAB-coated SNP8. Figure S2: DRIFT spectra of bare and DDAB-coated SNP80. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare the following competing financial interest: D.B., J.M., and M.M.F.L. are employees of Matera. L.F. is a coowner of Matera, a company that has an option to license IP generated from L.F. The remaining authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the financial support of Biocant Ventures, QREN (Project No. 6627), and FCT (PTDC/Qui-Qui/105000/ 2008).

■

REFERENCES

(1) Ferreira, L.; Zumbuehl, A. Non-leaching surfaces capable of killing microorganisms on contact. J. Mater. Chem. 2009, 19, 7796− 7806. (2) Zumbuehl, A.; Ferreira, L.; Kuhn, D.; Astashkina, A.; Long, L.; Yeo, Y.; Iaconis, T.; Ghannoum, M.; Fink, G. R.; Langer, R.; Kohane, D. S. Antifungal hydrogels. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 12994−8. (3) Paulo, C. S.; Vidal, M.; Ferreira, L. S. Antifungal nanoparticles and surfaces. Biomacromolecules 2010, 11, 2810−7. (4) Tiller, J. C.; Liao, C. J.; Lewis, K.; Klibanov, A. M. Designing surfaces that kill bacteria on contact. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 5981−5. (5) Klibanov, A. M. Permanently microbicidal materials coatings. J. Mater. Chem. 2007, 17, 2479−2482. (6) Park, D.; Wang, J.; Klibanov, A. M. One-step, painting-like coating procedures to make surfaces highly and permanently bactericidal. Biotechnol. Prog. 2006, 22, 584−9. (7) Haldar, J.; An, D.; Alvarez de Cienfuegos, L.; Chen, J.; Klibanov, A. M. Polymeric coatings that inactivate both influenza virus and 7655

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656

Langmuir

Article

(28) Rapuano, R.; Carmona Ribeiro, A. M. Physical adsorption of bilayer membranes on silica. J. Colloid Interface Sci. 1997, 193, 104− 111. (29) Ahmed, S.; Nikolov, Z.; Wunder, S. L. Effect of Curvature on Nanoparticle Supported Lipid Bilayers Investigated by Raman Spectroscopy. J. Phys. Chem. B 2011, 115, 13181−13190. (30) Li, P. C.; Zhang, L. X.; Ai, K. L.; Li, D.; Liu, X. H.; Wang, E. K. Coating didodecyldimethylammonium bromide onto Au nanoparticles increases the stability of its complex with DNA. J. Controlled Release 2008, 129, 128−134. (31) Haldar, J.; Chen, J.; Tumpey, T. M.; Gubareva, L. V.; Klibanov, A. M. Hydrophobic polycationic coatings inactivate wild-type and zanamivir- and/or oseltamivir-resistant human and avian influenza viruses. Biotechnol. Lett. 2008, 30, 475−479. (32) Hsu, B. B.; Wong, S. Y.; Hammond, P. T.; Chen, J. Z.; Klibanov, A. M. Mechanism of inactivation of influenza viruses by immobilized hydrophobic polycations. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 61− 66.

7656

dx.doi.org/10.1021/la300948n | Langmuir 2012, 28, 7646−7656