Synthesis and antibacterial properties of a hybrid of silver – potato starch nanocapsules by miniemulsion / polyaddition polymerization

Controlled synthesis of hollow polymeric nanocapsules has attracted significant attention in a wide range of applications. This paper reports a facile method for the synthesis of hybrid starch nanocapsules decorated with silver nanoparticles using the inverse miniemulsion polyaddition technique. Silver nanoparticles are formed and embedded in the shell of the nanocapsules during the polyaddition process without using any additional reducing agents. We found that silver also acts as a lipophobe that builds up osmotic pressure in the droplets facilitating the formation of stable round shaped nanocapsules. The nanocapsules' shell thickness could be tuned from 13 to 29 nm by varying the amount of cross-linker. We investigated the minimum inhibitory concentration (MIC) of the nanocapsules against Staphylococcus epidermidis ATCC 35984 and Escherichia coli ATCC 25922 which are two bacteria of medical relevance. The silver nanoparticle decorated nanocapsules showed antibacterial properties against both bacteria at the MIC of 2.315 mg mL 1 while control nanocapsules without silver had no antibacterial activity.


Introduction
Infections are the second leading cause of death worldwide aer heart diseases.The importance of the problem has led to extensive prevention and treatment efforts.However, despite great examples of success, infections are still part of our society causing suffering, death and substantial burden to the healthcare systems. 1,2Today, the main weapons to ght infections are antibiotics.This therapy saved millions of lives over the last 60 years; however, it has some disadvantages. 3ne of the major problems is the phenomenon associated with antimicrobial resistance. 4,5Since the 1970s, a trend to returning to silver and silver based compounds aer the development of antibiotic resistance has emerged. 6,7Over the last two decades, there has been increased interest in the use of silver, silver compounds and silver nanoparticles in medical applications.Examples are coatings for applications on wound dressing 8,9 and medical devices such as catheters. 10,11A number of commercial silver and silver nanoparticle containing products are currently on the market.A vast number of research papers on the subject have also been published.3][14][15][16][17][18][19] The reasons for this trend are that silver is universally active against all bacteria and not limited to certain bacterial types as antibiotics.Importantly, there is no convincing evidence that medically relevant bacteria are capable of developing resistance to silver. 20,21he goal of this work is to generate advanced biocompatible nanocapsules which are decorated with silver nanoparticles.Nanocapsules have been shown to be promising carries of drugs and other therapeutic molecules. 22The underlying motivation is to add antibacterial properties to nanocapsules' transport capability.We selected miniemulsion polymerization to achieve our goal.The technique is facile and provides molecular control over nanocapsule dimensions, structure and chemical composition. 22Narrowly distributed small droplets resulting from this process can act as nano-carriers or nano-reactors which makes them attractive in many applications such as targeted drug delivery systems, 23,24 self-healing coatings, [25][26][27] and optical molecular imaging. 28The technique has already been demonstrated for the successful encapsulation of a variety of materials such as nanoparticles, pigments, polymers, fragrances, drugs, and photoinitiators. 22,29We chose potato starch as the organic component for generating the nanocapsule wall structure because of its biocompatible and biodegradable nature. 30Starch nanocapsules have already been demonstrated as drug carriers with potential in targeted drug delivery and therapies. 31,32his study further aims to demonstrate the capacity to control the capsule wall thickness.In addition, we aim to provide functional chemical groups of the outer wall of the capsules which can be used for attachment of ligands for different applications (i.e.antibodies for cell targeting) or immobilization to surfaces.

Instrumental
A PSS Nicomp Particle Sizer 380 (Nicomp Particle Sizing Systems, USA) was used to measure the average size and size distribution of the nanocapsules.High resolution scanning electron microscopy (SEM) (Hitachi S-5200) and Transmission Electron Microscopy (TEM) (Philips CM200) were carried out to determine the particles' morphology.Samples were prepared by drying a droplet of diluted samples on silicon wafer for SEM and carbon coated copper grids for TEM.FTIR measurements were carried out using a Perkin Elmer Spectrum BX in the spectral range between 4000 and 400 cm À1 to determine the chemical composition of the nanocapsules.The equipment used for measuring the particle charge of nanocapsules was PCD 02 (Mütek Company) equipped with a Mettler DL21 Titrator.The amount of silver present in nanocapsules was analyzed by inductively coupled plasma (ICP) spectroscopy using a Perkin Elmer ICP-OES Optima 7300 DV to analyze Ag (mg L À1 ).The reading wavelength for Ag was 328.068 nm and the collected data were processed with MSF (Multicomponent Spectral Fitting).

Solid content measurement
The solid content of each sample was measured gravimetrically.An empty scintillation vial was weighed and the weight was recorded to the nearest thousandth of a gram (A). 100 mL of sample was then added to the vial and the weight was recorded (B).The specimen was then placed in an oven with temperature set at 105 C for 3 hours.The dried specimen was then removed from the oven and the weight was recorded to the nearest thousandth of a gram (C).The following equation calculates the percent of solids First, an aqueous solution consisting of 0.057 g AgNO 3 , 1.300 g sterilized water and 0.100 g soluble potato starch was prepared.To this phase, 7.400 g of cyclohexane pre-mixed with 0.100 g PGPR was added and stirred at 1000 rpm for one hour at 25 C using a magnetic stirrer.Then the mixture was ultrasonicated for 180 s at 90% amplitude in a pulse regime (20 s sonication, 10 s pause) using a Branson Sonier W-450-Digital and a 1/2 00 tip under ice cooling conditions in order to prevent evaporation of the solvent.Different amounts (0.200, 0.150 and 0.100 mg) of TDI dissolved in 4.970 g of cyclohexane and pre-mixed with 0.030 g PGPR were added dropwise over 5 min to the previously prepared emulsion under stirring conditions while maintaining the temperature at 25 C.The mixture was kept under stirring overnight at 25 C.The miniemulsion samples were then dispersed in 0.3% Lutensol AT50 at a ratio of 1 : 5. Another set of samples was prepared according to the procedure described above but without AgNO 3 .The designation and formulation compositions used in this study are denoted in Table 1.

Carboxymethylation of starch-silver nanocapsules
The carboxymethylation of starch-AgNP nanocapsules was performed using a method published previously. 24Briey, 0.4 mL of NaOH solution (0.1 M) was added to 4.0 g of starch-AgNP re-dispersion nanocapsules (solid content 1.0 wt%) and stirred at ambient temperature for 24 h to neutralize the non-reacted hydroxyl groups of the starch molecules on the nanocapsule surface.Next, 40 mL of MCA (20.0 wt%) was mixed with the above mentioned dispersion and stirred for 24 h at 40 C. Then 0.2 mL of NaOH solution (1.0 M) was added to the solution and stirred again for 24 h at ambient temperature.Finally, the nanocapsule dispersion was centrifuged at 4000 rpm for 20 min (Sigma 3k-30, RCF 1467) and aer removing the supernatant starch-AgNP nanocapsules were redispersed in demineralized water.The amount of carboxylic groups was determined by the polyelectrolyte titration method described below.Carboxyl acid groups were titrated against poly-DADMAC using 10 mL of the dispersion with a solid content of 0.01 g L À1 at a pH of 10.Each sample was titrated in triplicate, the number of carboxyl acid groups per nm 2 was determined from the average consumed volume of the titrant using the following equations: 33 V ¼ volume of the consumed polyelectrolyte in liters, M ¼ molar concentration of the polyelectrolyte in moles per liter, F ¼ density of polystyrene (1.045 Â 10 6 g L À1 ), N A ¼ Avogadro's constant, SC ¼ solid content of the sample in grams, D n ¼ average number diameter of the particle.

Determination of antibacterial activity in vitro
Staphylococcus epidermidis (ATCC 35984) and Escherichia coli (ATCC 25922) were used because these are common microorganisms involved in hospital-acquired infections.The nanocapsules were assayed for activity against these two strains in vitro. 34Bacteria were plated on Luria-Bertani agar plates from frozen stocks and incubated at 37 C. Individual bacterial colonies were picked and incubated overnight at 37 C in the oven.Nanocapsules were washed and serially diluted in quadruplicate in a 96 well microtiter plate with a nal volume of 50 mL per well.ABS 600 was used to adjust to approximately 1 Â 10 8 CFU mL À1 and further diluted to 1.33 Â 10 6 CFU mL À1 in tryptic soya broth (TSB).A total of 150 mL of bacterial suspension was added to each well to achieve a nal concentration of approximately 1 Â 10 6 CFU mL À1 .Bacterial suspensions without nanoparticles and wells containing particles and TSB without bacteria were used as controls.Using a microtiter plate reader (Biotek, ELx800), the minimal inhibitory concentration (MIC, the lowest concentration of an antimicrobial that will inhibit the visible growth of a microorganism aer overnight incubation) values were determined by measuring the turbidity of the bacterial suspension aer 24 h incubation at 37 C.The effect of light scattering induced by the nanocapsules was taken into account when reading ABS 600 .The results were conrmed visually.All assays were performed in triplicate.

Results and discussion
The general approach to synthesize potato starch nanocapsules with AgNPs embedded within their walls is shown in Fig. 1a.A macroemulsion is rst prepared by mixing potato starch and silver nitrate in aqueous medium with an amphiphilic surfactant polyglycerin-polyricinoleate (PGPR) in cyclohexane.A miniemulsion is then generated by sonication of this mixture in order to achieve homogenous aqueous droplets in the cyclohexane phase.In a next step 2,4-toluene diisocyanate (TDI) is added in order to facilitate the formation of the core-shell structure of starch-AgNP nanocapsules via polyaddition.The nanocapsule shell is formed by cross-linking of the starch OH groups with the -NCO groups of TDI.The mixture is then transferred from cyclohexane to an aqueous medium using the non-ionic surfactant Lutensol AT50, and this surfactant facilitates the redispersion of oil soluble nanocapsules into the water phase.In this synthetic procedure the starch acts as a reducing agent for silver nitrate.Reduction of silver nitrate into metallic silver occurs during the polyaddition and cross-linking steps facilitated by the release of free radicals 35 and D-glucose from the amylose component of starch. 36,37The nanocapsules were stable over several months in both the hydrophobic continuous phase (cyclohexane) and when redispersed in water (data not shown).The characteristic surface plasmon resonance (SPR) bands of AgNPs with the absorption maxima at around 412 nm are evidence for the presence of AgNPs in the nanocapsule solution (Fig. 2), and the inset in Fig. 2 shows the resultant solution of starch silver nanocapsules (grayish brown color) versus the starch only nanocapsules.The purple color of starch only nanocapsules is as a result of addition of sulforhodamine 101 (SR101) uorescent dye (data not shown).
The proportion of non-volatile material contained in the miniemulsions le aer the volatile solvent (water or cyclohexane) vaporized in the oven was calculated and expressed in the form of solid content (%) (Table 2).The loss in solid content of the redispersed sample is due to ltration of nanocapsules.The average sizes of the nanocapsules were determined by dynamic light scattering and are shown in Table 2. Formulations without addition of AgNO 3 (S046, S047 and S048) resulted in capsules in the size range of 700-800 nm.
SEM imaging further revealed that these nanocapsules are smaller in size but they were agglomerated and substantially deformed.A representative image of this type of nanocapsule is shown in Fig. 3d-f The SEM images of nanocapsules prepared with different amounts of TDI are shown in Fig. 3a-c.The images depict stable hollow spherical entities.Due to drying and vacuum required for SEM imaging the nanocapsules appear like hollow half balls.
The images also show bright entities within the walls of the nanocapsules which are assigned to silver nanoparticles.The presence of metallic silver in the walls of the nanocapsules was also conrmed by energy-dispersive X-ray (EDX) spectroscopy which is discussed later in the paper (Fig. 5).It is interesting to note that the silver nanoparticles do not appear clear in the images of the nanocapsules presented in Fig. 3a.The reason for this is the greater shell wall thickness.These nanocapsules were synthesized using the highest amount of TDI.The dependence of the shell wall thickness on the concentration of TDI is discussed later in the paper.The SEM observations are an indication of the important role of AgNO 3 as an osmotic pressure agent. 38Silver nitrate as an inorganic salt is insoluble in the oil phase and can act as a lipophobe builds up an osmotic pressure in the droplets. 39Thus, silver ions may be capable of suppressing Ostwald ripening and in this manner facilitate the formation of a well-dened core-shell structure.Analysis of the SEM images in Fig. 3a-c using the ImageJ soware package   suggested round shaped silver nanoparticles of 14 nm diameter and a narrow size distribution.Furthermore, the size of the silver nanoparticles does not seem to be affected by the shell thickness.We calculated an average of 5.5 silver nanoparticles per each 100 nm 2 of the shell surface area.We hypothesize that the growth of nanoparticles is controlled during the formation of the nanocapsules' shell structure.
An SEM backscattered electron image of the silver nanoparticle containing nanocapsules is shown in Fig. 4a.Silver nanoparticles appear as bright points because of their capacity to scatter a larger number of electrons.EDX spectra recorded at positions 25 and 26 of the image presented in Fig. 4a are shown in Fig. 4b.Several peaks can be identied.These peaks are representative of the chemical composition of the nanocapsules.The peak at 3 keV is characteristic of the silver K and L lines and evidence for the presence of silver nanoparticles.The peak at 0.2 keV represents carbon.The peaks located at 0.38 and 0.5 keV are associated with the nitrogen and oxygen characteristic lines, respectively.The sodium line comes from the surfactants used in the formulation.There is a small Si line which is from the silicon wafer substrate used as a support.The distribution map of Ag is shown in Fig. 4c and shows a discrete and uniform distribution of silver in the nanocapsules.
The nanocapsule shell wall thickness was determined from the SEM images.Fig. 5 shows shell wall thicknesses as a function of the TDI amount used in the synthesis for formulations involving AgNO 3 .It appears that the amount of added TDI signicantly inuences the nanocapsule shell thickness.As a general trend, the nanocapsule shell wall had a greater thickness with increasing TDI amount.For example, formulations S043, S044 and S045 where the TDI amount was 200, 150 and 100 mg had the shell thickness of 29 nm, 20 and 13 nm, respectively (Fig. 5).
Further insight into the distribution and size of the silver nanoparticles was revealed by TEM imaging.The TEM images in Fig. 6b demonstrate that a number of small 3-6 nm silver nanoparticles, which are beyond the resolution of SEM, are also available.On rare occasions we also observed the formation of some larger silver nanoparticles on a few tenths of nanometers as shown in the TEM image in Fig. 6a.The larger particles appear to have formed out of the nanocapsules and thus their growth has not been restricted by the walls.FTIR spectroscopy was applied to study the chemical composition of the samples (Fig. 7).The peak at 2276 cm À1 could be assigned to NCO groups; however, its low intensity suggests that in the nal redispersed product most of these groups were consumed by the hydrolysis reaction with water to form amines. 40 A close examination of FTIR spectra suggests that the incorporation of silver nanoparticles does not lead to changes in the chemical composition of nanocapsules as in both compositions peaks identied with chemical bonds appear at the same frequencies.The presence of hydrogen bonded -OH stretching and NH valence vibration (which is overlapped by O-H vibration) is evident in the form of a broad band in the region of 3600-3200 cm À1 .A strong band at 2899 cm À1 is characteristic of the presence of CH (sp 3 ) stretching in the aromatic structure.The weak broad band at 1968 cm À1 is attributed to N-H/C-N groups.Free C]O vibration and polyurethane C]O vibration are observed at 1734 and 1700 cm À1 , respectively.The weak broad band at 1640 cm À1 is due to the formation of polyurea.The bands observed in the region of 1650-1450 cm À1 also indicate C]C vibration in the aromatic ring.C-O-C anti-symmetric bridge stretching is observed at 1150 cm À1 in the form of a medium intensity band (cm À1 ).The bands at 1467 and 1343 cm À1 are attributed to CH 2 and CH bending respectively.The strong bands at 1300 and 1100 cm À1 are assigned to the C-(C]O)-C bending and C-O valence vibration.A strong band at 1060 cm À1 mainly originates from the tertiary alcohol (R-C(CH 3 )-OH).The bands at 963, 842, 717 and 529 cm À1 are also attributed to the skeletal vibrations of the polymeric shell.
4][45] The amount of carboxylic groups on the nanocapsule surface was determined by polyelectrolyte titration and the results are presented in Table 2.The average COOH functionalities in the samples containing AgNPs are in the range of 0.1-0.13 group per nm 2 and in pure starch nanocapsules they are between 0.28 and 0.45 group per nm 2 .Such functional group densities are believed to be sufficient for immobilization purposes and also provide a sufficient number of binding sites for attachment of ligands or proteins.
The concentration of silver based on ICP analysis was 37 mg mL À1 .The decrease in the silver concentration compared to the initial amount used in the synthesis can be attributed to removal of silver nanoparticles which were not embedded in nanocapsule walls during the washing and centrifuge steps.
Finally, we examined the MIC of the nanocapsules against Staphylococcus epidermidis (S. epidermidis) and Escherichia coli (E.coli).These two bacteria pose a signicant medical problem and cause a substantial part of the infections.The concentration of bacteria used for determination of the MIC values was 1 Â 10 6 CFU mL À1 .The minimum concentration of starch-silver nanocapsules required for inhibition of bacteria growth was studied by serial dilution.We found that a very low concentration of starch-silver nanocapsules was able to    This journal is © The Royal Society of Chemistry 2014 inhibit the bacteria growth.These concentrations are shown in Table 3 and were 2.315 mg mL À1 for both S. epidermidis and E. coli.The growth was completely inhibited at 2.315 mg mL À1 in starch-silver nanocapsules while starch nanocapsules without silver nanoparticles had no inhibitory effect on bacterial growth.7][48] In this work, we were able to employ the miniemulsion technique in combination with polyaddition polymerization to prepare in a controllable manner silver nanoparticles that are incorporated in the walls of potato starch nanocapsules.These nanocapsules have strong antibacterial properties with MICs in the range of a few micrograms per milliliter for S. epidermidis (Gram-positive) and E. coli (Gram-negative).This type of antibacterial nanocapsule could be of interest in biomedical applications because they might be capable of reducing the rates of infections when such carriers are used in drug delivery.The material is also interesting because starch is biodegradable, which means that it would be broken down by natural physiological processes if delivered into the body.Current work in our lab is focused on immobilization of these starch-silver nanocapsules to solid surfaces for applications as antibacterial coatings that can be applied to wound dressing and implantable biomedical devices.We are also exploring loading of various therapeutic agents in these hybrid nanocapsules that can be delivered either via solid carriers or in a suspension form.

Conclusion
In summary, we report a facile method for synthesis of hollow starch nanocapsules decorated with silver nanoparticles.The addition of AgNO 3 seems to play an important role facilitating the formation of stable spherical nanocapsules.The nanocapsule wall thickness can be controlled by the amount of TDI used in the synthesis.We were able to conduct further surface modication of the nanocapsule walls in order to provide carboxyl functional groups which can be potentially used for the surface immobilization or binding of ligands of interest for a particular application.The silver nanoparticles were formed in capsule walls and had diameters of about 14 nm.However, TEM images revealed the presence of some quantity of smaller particles.Occasionally, large nanoparticles of a few tenths of nanometers were also observed but they appear to have grown outside of capsule walls.MIC studies showed high antibacterial efficacy against Staphylococcus epidermidis ATCC 35984 and Escherichia coli ATCC 25922.The biocompatible nanostructures presented in this paper are of potential interest for various biomedical applications such as smart drug carriers and antibacterial coatings for medical devices and wound dressings.

Fig. 2
Fig. 2 UV-visible spectra of starch-silver nanocapsules versus starch only nanocapsules; the inset shows the appearance of the miniemulsion solutions.

Fig. 7
Fig. 7 FTIR spectra of the starch nanocapsules synthesized with (a) and without silver nitrate (b).

Fig. 5
Fig. 5 Shell thickness versus the TDI amount used in the synthesis.

Fig. 6
Fig. 6 TEM micrographs of a starch-silver nanocapsule (a) and a magnified image showing the capsule wall and smaller silver nanoparticles (b).

Table 1
Formulations used in the synthesis of nanocapsules This journal is © The Royal Society of Chemistry 2014 . When the synthesis was conducted in the presence of AgNO 3 , a capsule size of 290-370 nm was achieved.

Table 2
Characteristics of nanocapsules: solid content, DLS size determination and carboxyl acid functionalities

Table 3
MIC of starch-silver nanocapsules (S043) against Staphylococcus epidermidis ATCC 35984 and Escherichia coli ATCC 25922 a The mean values are shown with STD: <0.5.