Antibacterial Effect of Chitosan–Gold Nanoparticles and Computational Modeling of the Interaction between Chitosan and a Lipid Bilayer Model

Pathogenic bacteria have the ability to develop antibiotic resistance mechanisms. Their action consists mainly in the production of bacterial enzymes that inactivate antibiotics or the appearance of modifications that prevent the arrival of the drug at the target point or the alteration of the target point itself, becoming a growing problem for health systems. Chitosan–gold nanoparticles (Cs-AuNPs) have been shown as effective bactericidal materials avoiding damage to human cells. In this work, Cs-AuNPs were synthesized using chitosan as the reducing agent, and a systematic analysis of the influence of the synthesis parameters on the size and zeta potential of the Cs-AuNPs and their UV-vis spectra was carried out. We used a simulation model to characterize the interaction of chitosan with bacterial membranes, using a symmetric charged bilayer and two different chitosan models with different degrees of the chitosan amine protonation as a function of pH, with the aim to elucidate the antibacterial mechanism involving the cell wall disruption. The Cs-AuNP antibacterial activity was evaluated to check the simulation model.


Surface charge density calculation
The surface charge density ( ) of a spherical colloidal particle can be obtained following the method proposed by Makino & Ohshima [1] on the basis of the zeta potential and the ionic strength of the medium. For a nanoparticle with constant surface charge density, the surface charge density is a more characteristic quantity than the zeta potential, 0 , because for such particles the zeta potential is not a constant and depends on the electrolyte concentration. The method is based on Equation (1): where k is the Boltzmann constant, T is the absolute temperature, Ɛr is relative permittivity, Ɛ0 is the permittivity of the vacuum, Ɛ is the permittivity, NA is the Avogadro's number, e is the elementary electric charge, I is the Ionic strength; a is the radius of the spherical nanoparticles, Ψ0 is the zeta potential, z is the electrolyte valence and where is the Debye-Hückel parameter defined by Equation (2) The results of the calculated surface charge density for each experiment have been included in Table S1 and all the parameter values used for the calculation are shown in Table S2. Table S1. Surface charge density (C/m 2 ).

UV-vis Spectroscopy
UV-vis absorbance spectra of the synthesized Cs-AuNPs were recorded on a Thermo Spectronic Heλios α spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) in the wavelength range of 750-450 nm. The spectra were obtained for preliminary confirmation of AuNP formation from the surface plasmon resonance band around 530 nm [2]. One milliliter of sample and a quartz cell were used for the measurements and double deionized water as blank. The diameter, d, of the AuNPs obtained in each experiment was estimated from UV-vis parameters using Equation (3), as in Haiss et al. [3]: where is the absorbance at the surface plasmon resonance peak, 450 is the absorbance at λ = 450 nm and 1 and 2 are empirical parameters with values of 3.00 and 2.20, respectively.

Influence of chitosan molecular weight on the physical features of Cs-AuNPs
The color of the resulting dispersion after Cs-AuNPs formation was observed to depend on the chitosan molecular weight ( Figure S1). The color of AuNP solutions can be red, pink, violet, purple, blue or yellow with increasing particle size and the formation of aggregates [4]. The formation and resulting stability of AuNPs can be followed by UV-vis spectroscopy. AuNPs usually absorb light in the visible range depending on their size, local refractive index and aggregation leading to plasmon coupling. For bigger AuNPs, the plasmon absorption band is shifted to longer wavelengths [5]. The intensity of the absorption band is also sensitive to the size and shape of the AuNPs, the increasing intensity of the absorption maximum corresponding to greater AuNP formation growth [6]. Figure S2 shows the UV-vis spectra of the Cs-AuNPs synthesized from chitosan of different molecular weights. The spectra have been normalized with respect to the absorbance at the lowest wavelength. The absorption bands of the Cs-AuNPs were usually observed at around 530 nm, which is indicative of AuNP formation [7]. Only the peak of the spectra of VLMw at 0.1% and of HMw at 0.4% chitosan concentrations were displaced to longer wavelengths of 564 nm and 538 nm, respectively. The shift to 538 nm was attributed to a local change in the refractive index provoked by the chitosan coating, while the shoulder at ca. 570 nm might correspond to small aggregates of AuNPs within a shell of chitosan. It was also seen that the intensity of the absorption increased as the molecular weight decreased (when the concentration of CS was 0.4%), indicating a higher degree of reduction of the metal ions. This trend was not so evident for the lower chitosan concentration.

Influence of the chitosan concentration on the biosynthesis of Cs-AuNPs
As can be seen in Figure S1, the color of the colloid after Cs-AuNP formation follows two different trends as a function of the concentration of chitosan. In the case of 0.1 w/v % chitosan, VLMW and LMW led to purple colloidal dispersions, indicating aggregation of the AuNPs. This aggregation is further confirmed by the UV-vis data (see Figure S2). With greater molecular weights of chitosan, the extent of aggregation was reduced and AuNPs appeared as individual nanoparticles rather than aggregates. Therefore, for a 0.1 w/v % of chitosan, there is a threshold molecular weight below which the polymer chains do not provide sufficient steric stabilization and a modest aggregation occurs. By contrast, the aggregation of the AuNPs was much more difficult for 0.4 w/v % chitosan and individual unaggregated AuNPs were found for the entire range of molecular weight (see Figure S2). Only for the highest molecular weight was a small shift observed, probably corresponding to a local variation in the refractive index, as might be expected from the rather bulky polymer shell. Figure S3 shows the UV-vis spectra of the Cs-AuNPs synthesized from different chitosan concentrations. As can be seen, for LMw, MMw and HMw chitosans, the higher concentration (0.4 w/v %), i.e. higher Cs:Au molar ratio, led to a lower rate of AuNP synthesis. This may have been due to the increased viscosity, which controls the nucleation and growth of AuNPs [8]. These results agree with those of other authors who claimed that, as the gold precursor concentration increases, i.e. lower Cs:Au molar ratio, the plasmon peak shows greater absorbance, indicating greater progress in the gold reduction process [9].

Influence of the gold concentration on the biosynthesis of Cs-AuNPs
As regards the color of the resultant colloids ( Figure S4), an increase in the concentration of gold led to a more intense color, as was to be expected from the higher amount of AuNPs formed. The same conclusion was reached by Regiel-Futyra et al. [9].  Figure S5 shows the UV-vis absorption spectra of the AuNPs obtained as a function of the gold concentration. As can be seen, an increase in the gold concentration clearly contributed to a higher rate of reduction and hence a higher AuNP synthesis; see the increase in peak absorbance in the plasmon absorption band (around 530 nm).

Influence of the reaction solvent on the biosynthesis of Cs-AuNPs
The pairs of experiments 3-10 and 12-13 are useful for determining the effect of methanol and experiments 9-11 for determining the effect of isopropanol. It is important to remark that in all cases the alcohol was incorporated in the reaction mixture in a proportion of 50% volume, the rest being acetic acid. Figure S6 shows the differences in the color of the colloids obtained. At a first glance, it seems that the presence of alcohols in the reaction mixture contribute to making the solution cloudier, suggesting an enhanced interaction with the chitosan coating and leading to the formation of micron-size aggregates.  Figure S7 shows the UV-vis absorption spectra for AuNP formation with MMw chitosan as a function of the reaction solvent system (acetic acid, acetic acid/methanol 50 v/v % or acetic acid/isopropanol 50 v/v %). Based on the synthesis parameters used for the experiments of Figures S7a and S7c, we can assume that the alcohols, especially methanol, do not contribute to improving the Cs-AuNP yield that can be obtained using acetic acid alone because their spectra show a lower degree of intensity at maximum absorbance. However, in the case of Figure S7b, methanol seems to slightly improve Cs-AuNP synthesis. This discrepancy is probably due to the different Cs:Au molar ratios used. In the experiments of Figures S7a and S7c, the molar ratios were lower, 40:1 and 20:1, respectively, but in the experiments of Figure S7b a molar ratio of 80:1 was used.

Estimation of size and concentration of AuNPs from UV-vis spectra
The size of AuNPs is an important variable for any proposed medical or industrial application. In this case, the size of Cs-AuNPs, i.e. the whole entity formed by the core and the polymer shell, should be differentiated from the size of the AuNP itself (the core). The location of the absorption band in the UV-vis spectra provides approximate information on the core size as does TEM, but the total size of the core-shell particle is equally important. The latter can be assessed by DLS and hydrodynamic diameter measurements. Table S4 shows the estimated values of the diameter based on the UV-vis spectra using Equation (3), and the values obtained from DLS measurements and from TEM. Note that DLS gives the hydrodynamic diameter, which includes the AuNP, the chitosan coating and the water of the hydration sphere, while TEM only provides the size of the AuNP, meaning that these values should be significantly lower than those obtained by DLS.