N,N,N-Trimethyl chitosan (TMC) is a cationic polysaccharide obtained from chitosan reduction ^{[1, 2, 3]}. TMC has properties such as,mucoadhesivity,biocompatibility,degradability and antimicrobial activity ^{[1, 2]}. Sodium alginate (ALG) is an anionic polysaccharide composed of a-(1-4)-L-guluronic (G) acid and β- (1-4)-D-mannuronic (M) acid units ^[4]. Polyelectrolyte complexes (PECs) are made by attractive coulombic interactions and by other secondary forces ^[5]. The most recent works published by our research group highlighted the preparation and characterization of PECs based on TMC/ALG ^{[4, 6]}. In these cases,TMC/ALG PECs were associated with gold nanoparticles (AuNPs) and curcumin (CUR) ^{[4, 6]}. TMC/ALG PECs promoted the stabilization and protection of AuNPs ^[6]. Furthermore,TMC/ALG loaded with AuNPs was lightly cytotoxic on Caco-2 and VERO cells,compared to neat TMC/ALG beads ^[6]. Such materials also acted as a vehicle for efficient CUR delivery at pH close to physiological conditions ^[4]. So,this work is aimed at associating TMC/ALG beads with AgNPs (beads/AgNPs). In this case,the antimicrobial activity of beads/AgNPs on Escherichia coli (E. coli) was evaluated at pH 7.4. Finally,Ag⁺ releasing study from AgNPs loaded into beads was performed only at pH 7.4,and the obtained results are discussed in the light of the bactericidal assay.

N,N,N-Trimethyl chitosan (TMC) with acetylation degree equal to 15%; quaternization degree (DQ) of ≈20%,and MV of 26×10³ g mol^-1 was synthesized from chitosan ^{[4, 6]}. Sodium alginate (ALG) was purchased from Across Organics (New Jersey, USA),and the ratio of mannuronic acid to guluronic acid (M/G) of the ALG was 1.56,as stated by the manufacturer. It has already been reported that the values of the average number (Mn) and average-weight (Mw) molecular weights for this alginate are 339,000 g mol^-1 and 1,073,000 g mol^-1,respectively ^{[4, 6]}. Silver nitrate,sodium citrate,potassium dihydrogen phosphate,and acetic acid were purchased from Sigma-Aldrich. All reagents were used as received,i.e.,without any further purification.

Silver nitrate solution (1.0 mmol L^-1) in a reflux system was preheated at 90 ℃ until boiling,and then 2.5 mL of sodium citrate solution (0.30 mol L^-1) were added to the system. So,after 4 min of heating at boiling,the system was taken off the heat. Finally,AgNPs suspension was poured into an ice bath and stored in an amber vial under refrigeration at 4 ℃. The average diameter size of AgNPs was obtained from Transmission Electron Microscopy (TEM) images, using the program Statistic version 8 ^[6].

The methodology used for TMC/ALG beads preparation was published in detail by Martins et al. ^{[4, 6]}.

TMC and ALG solutions were separately prepared from solubilization of each polymer in acetic acid solution. Then,AgNPs suspension (5.0 mL containing 0.54 mg of Ag⁰) were added to 10 mL of ALG-solution (0.5%,wt/v) and the mixture homogenized. So,ALG/AgNPs mixture was slowly dropped into 20 mL of TMCsolution (1.0%,wt/v) under stirring at 28 ℃. Each drop of the ALG/ AgNPs suspension added into TMC-solution was responsible for the formation of only one bead containing AgNPs. Beads/silver nanoparticles (beads/AgNPs) were separated by decantation as previously described by Martins et al. ^[6].

In vitro silver ion release assay was performed in buffer solution at pH 7.4. Then,an amount of dried beads/AgNPs was deposited in a sealed flask with 110 mL at pH 7.4 (37 ℃). At a desired time interval,aliquots were removed from the flask in order to quantify the amount of Ag⁺ released.

Antimicrobial activity tests of beads and beads/AgNPs against E. coli (ATCC 26922) were carried out according to the viable-cellcounting method ^[7]. Briefly,15 mg of beads or beads/AgNPs were transferred to 15 mL of 0.137 mol L^-1 PBS at pH 7.4 and then 150μL of calibrated inoculum of E. coli (10⁷ CFU mL^-1) were added. This solution was allowed to incubate for 6 h at 37 ℃ under constant shaking. Aliquots of 100 μL were collected from each supernatant and diluted with Tryptic Soy Broth (TSB) to a final concentration of 10⁵ CFU mL^-1. Then,30 μL of each supernatant were plated on Tryptic Soy Agar (TSA) plates,incubated for 24 h at 37 ℃,and then the CFU number was counted. Other experiments were also carried out using the same procedure but with 50 mg of samples and incubated for 24 h under constant shaking. In parallel, a test with a control group (TMC/ALG beads without AgNPs) was carried out using only 15 mL of sterilized PBS. All experiments were performed in triplicate.

Molecular absorption of silver nanoparticles (AgNPs) suspension was obtained from a Biochrom Libra S12 spectrometer,in the range of 300-900 nm. AgNPs morphology characterization was performed by Transmission Electron Microscopy (TEM) using a Shimadzu JOEL - JEM 1400 operated at 120 kV. Differential Scanning Calorimetry (DSC) analyses of neat beads and beads/ AgNPs were performed on a calorimeter (Netzsch,model STA 409 PG/4/G Luxx,USA) operating at heating rate of 10 ℃ min^-1, nitrogen flow rate of 50 mL min^-1 and temperature range from 40 ℃ to 210 ℃. Flame Atomic Absorption Spectroscopy (FAAS) analysis was used to evaluate the amount of Ag⁺ released from beads/AgNPs,using a spectrometer (Varian,model AA-175,USA). Silver hollow cathode lamp from the same equipment-supplier,operating at 5 mA,was used to measure the absorption signals. The analytical wavelength of 328.07 nm at silver resonance line was selected with a spectral band pass of 0.2 nm.

Maximum absorption of AgNPs colloidal suspensions occurred at 403 nm (Supporting information,Fig. S1). TEM images showed that the AgNPs size distribution was 2.3±0.5 nm (Supporting information,Fig. S2). DSC curves of beads and beads/AgNPs are shown in Fig. 1. DSC curves showed significant differences in the highlighted region (Fig. 1). Endothermic peaks at 174,185,188 and 192 ℃ were attributed to fusion of the samples and such peaks presented greater intensities on beads DSC curve concerning beads/AgNPs DSC curve. AgNPs surfaces have citrate passivation layer,which increases the negative charge density on beads/AgNPs matrix. This fact destabilized the beads,and the endothermic peak intensities decreased on beads/ AgNPs DSC curve (Fig. 1),confirming the AgNPs encapsulation process. Fig. 2 shows photos of neat beads and beads/AgNPs. Dry samples presented ca. 0.1-0.2 mm of diameter while neat swollen beads or swollen beads/AgNPs can reach 5 mm of diameter ^{[4, 6]}. The neat beads showed white coloration (Fig. 2,left side),while beads/AgNPs are yellowish (Fig. 2,right side),as expected,because the AgNPs are yellow.

Fig. 1

	Download: JPG larger image
Fig. 1.DSC curves of neat beads and beads/AgNPs.

Fig. 2

	Download: JPG larger image
Fig. 2.Photos of neat beads (left side) and beads/AgNPs (right side) processed immediately after the washing with acetone.

Release assays using beads/AgNPs showed that the Ag⁺ amount released at pH 7.4 reached 3.3 μg per each hydrogel milligram (Fig. 3). Equilibrium state was reached up to 5 h,being that 61% of loaded AgNPs were released in this time interval. UV-vis measurements indicated that the AgNPs releasing does not occur during the tests (not shown). The interactions between beads- AgNPs maintain the AgNPs stability in the hydrogel matrices, inhibiting their agregation. AgNPs suspensions are not stable and should be stored under refrigeration. Nevertheless,the AgNPs aggregation occurs after three to four months ^[8]. In situ AgNPs synthesis in the hydrogel matrices or metals evaporation processes in the heated polymer surfaces are methods used to stabilize metallic nanoparticles ^{[9, 10]}. However,in this work another more simple method for AgNPs stabilization was showed when compared to the methods previously described. In these cases,AgNPs were synthesized concomitantly to the beads/AgNPs obtention process. So,the AgNPs loading occurred at same time that the hydrogel matrices were prepared.

Fig. 3

	Download: JPG larger image
Fig. 3.Fraction released (μgmg-1) of Ag+ from beads/AgNPs at pH 7.4.

Swelling beads/AgNPs favors the water molecules diffusion and the Ag⁺ releasing process. Hydrogel/AgNPs matrices disintegrates after ca. 10 h,however the complete dissolution of beads/AgNPs no occurred up to 24 h (Fig. 3). This fact was associated with the permanent interactions among N-quaternized moieties of TMC and carboxylate groups on ALG chains,at pH 7.4 ^{[4, 6]}. The pK_a values of M- and G-residues are 3.38 and 3.65,respectively. Therefore,the beads/AgNPs swelling occurs because the -COOH groups on ALG backbone remain ionized. So,at pH 7.4 the interactions among carboxylate anions of ALG with water molecules by ion-dipole forces,promotes Ag⁺ releasing through diffusion mechanism and macromolecular relaxation of the polymer chains ^[4].

Bactericidal activities of neat beads and beads/AgNPs were evaluated at pH 7.4 using concentrations of 1.0 and 3.3 mg mL^-1. Fig. 4 demonstrates the antimicrobial activities of such systems against E. coli. When the bacteria were treated with the neat beads for 6 and 24 h,no bactericidal activity was observed. TMC has low quaternization degree (DQ = 20%),and at pH 7.4,only -⁺N(CH₃)₃sites are positively charged on the bead structures. However,these groups are complexedwith carboxylate anions on ALG,i.e. such sites are neutralized on the hydrogel matrices. Therefore,the neat beads were not bactericidal against E. coli because -⁺N(CH₃)₃ groups are not free to interact with the microbial cells ^[2]. On the other hand,when the beads/AgNPs concentration was increased to 3.3 mg mL^-1,the antibacterial effect increased to 91% inhibition (Fig. 4). The higher beads/AgNPs bactericidal activity regarding neat beads occurs due to the amount of Ag⁺ released. According to Hoang et al. ^[11] AgNPswith diameter lower than 12 nm show minimum inhibitory concentration (MIC) of 10 μg mL^-1. In this case,when the bacteria (E. coli) were treated with beads/AgNPs at a concentration of 1.0 mg mL^-1; 3.3 μg mg^-1 of Ag⁺ was released at pH 7.4 (Fig. 3). This concentration is lower than MIC. Accordingly,the beads/ AgNPs does not present bactericidal activity at a concentration of 1.0 mg mL^-1 (Fig. 4),since the Ag⁺ amount released was not above the MIC value. On the other hand,when the bacteria were treated with beads/AgNPs at a concentration of 3.3 mg mL^-1, 10.9 μg mL^-1 of Ag⁺ were released at pH 7.4 under incubation time of 24 h. This concentration is higher than MIC value and explains the 91% inhibition by beads/AgNPs,whichmay be due to the increase in the amount of Ag⁺ in solution.

Fig. 4

	Download: JPG larger image
Fig. 4.Cell counts treated with neat beads and beads/AgNPs against E. coli after incubation for 6 h (a) and 24 h (b) at pH 7.4.

Several studies also have indicated that AgNPs (＜10 nm) are able to penetrate cell membranes and possess bactericidal effects ^[11]. However,according to Lee et al. ^[12],the oxidation of Ag⁰ atoms present in the AgNPs suspension predominates only in the first 6 h of the release assay. Therefore,the results of Ag⁺ releasing (Fig. 3) are consistent with the data published by Lee et al. ^[12],since amount of Ag⁺ ion released achieved equilibrium after 5 h. This suggests that the bactericidal activity of the beads/AgNPs system against E. coli is due to Ag⁺ ion action ^{[3, 13]} on E. coli microbial cells,whereas the concentration of Ag⁺ released was constant after 5 h. Moreover,UV-vis measurements showed that the AgNPs were not present in aliquots in which the Ag⁺ ions were quantified (not shown). Therefore,at pH 7.4,the beads released 61% of Ag⁺ and 39% of AgNPs remained in their matrices. It was found that the concentration of Ag⁺ released from beads/AgNPs hydrogel (10.9 mg) was sufficient to kill E. coli cells.

Hydrogel beads based on TMC/ALG polyelectrolyte complexes were obtained,and concomitantly silver nanoparticles (AgNPs) were loaded into beads. TMC/ALG hydrogel stabilized the AgNPs and inhibited their aggregation. On the other hand,on buffer solution (at pH 7.4) the beads promoted Ag⁺ releasing. In this case, 61% of Ag⁰ on AgNPs-loaded were oxidized to Ag⁺ and 39% of AgNPs remained in the hydrogel matrices. So,Ag⁺ amount can be modulated to have cytotoxic effect on E. coli cells.

The authors thank CNPq/CAPES,Brazil for the financial support (Nos. 481424/2010-5 and 308337/2013-1).

Supplementary data associated with this article can be found,in the online version,at http://dx.doi.org/10.1016/j.cclet.2015.04.032.