An environmentally benign antimicrobial nanoparticle based on a silver-infused lignin core

Silver nanoparticles have antibacterial properties, but their use has been a cause for concern because they persist in the environment. Here, we show that lignin nanoparticles infused with silver ions and coated with a cationic polyelectrolyte layer form a biodegradable and green alternative to silver nanoparticles. The polyelectrolyte layer promotes the adhesion of the particles to bacterial cell membranes and, together with silver ions, can kill a broad spectrum of bacteria, including Escherichia coli, Pseudomonas aeruginosa and quaternary-amine-resistant Ralstonia sp. Ion depletion studies have shown that the bioactivity of these nanoparticles is time-limited because of the desorption of silver ions. High-throughput bioactivity screening did not reveal increased toxicity of the particles when compared to an equivalent mass of metallic silver nanoparticles or silver nitrate solution. Our results demonstrate that the application of green chemistry principles may allow the synthesis of nanoparticles with biodegradable cores that have higher antimicrobial activity and smaller environmental impact than metallic silver nanoparticles.

polyelectrolyte. The results demonstrate that such engineered nanoparticles could be as efficient as their AgNPs analogs, but have time-limited antimicrobial activity because of depletion of the bioactive Ag + ions.

Fabrication and characterization of EbNPs
We choose lignin, one of the most abundant and underutilized biopolymers in nature 33 , as a benign particle core material. The used lignin was Indulin AT, which is extracted from biomass as by-product of the Kraft pulping processes 34 . It is soluble in alkali solutions, but insoluble in neutral and acid conditions and has high absorbance capacity for heavy metal ions in environmental remediation processes 35,36 . We showed earlier how biopolymer micro-and nanoparticles can be synthesized via an entirely water-based, solvent free, pH-drop flash precipitation method 37,38 . A similar pH-drop flash precipitation in ethylene glycol solution can be applied to synthesize negatively-charged pH-stable Indulin AT lignin nanoparticles of controlled sizes 38 . The nanoparticles could be synthesized in a broad size range (Fig. 2a); the ones used as EbNP precursors after dialysis had a mean hydrodynamic diameter of 84 ± 5 nm as measured by Dynamic Light Scattering and a zeta potential of -33 ± 1 mV (see supplementary information (SI) for measurement details).
In the next step, the lignin nanoparticle cores were infused with Ag + ions at pH 5.0 by using an aqueous solution of AgNO3 as silver source. The IAT lignin has several functional groups, e.g. carboxylic, thiol, phenolic and aliphatic hydroxyl groups (see Table S1 in SI) that can serve as binding sites for silver ions 35,36 .
Previous studies indicated that the particles are porous 38,39 , which would increase the total available surface area for silver ion loading and could potentially increase their degradation rate in the environment after disposal 38 . The interaction of Ag + ions with the particles was investigated by characterizing the adsorption equilibria ( Fig. 2b, see also Fig. S1 and Table S2 in SI) and the Ag + desorption kinetics (Fig. 2c). These data taken together indicate the presence of high and low energy binding sites, as not all adsorbed Ag + ions are released after 48h (Fig. 2c). The net binding of Ag + ions onto the EbNPs could be modeled on the basis of a modified Langmuir adsorption isotherm Γ( ) = Γ max /(1 + ) + Γ 1 , where Γ and c are the respective surface and bulk equilibrium concentrations of silver ions; Γ max is the maximum surface loading for weakly bound silver ions, K is the adsorption constant, which relates to the Ag + binding energy to the particles 40 , and Γ 1 is the adsorption of strongly bound silver ions in the high affinity region in mg/g of EbNPs. The least square fitting of the experimental data in Fig. 2b at value of Γ 1 = 1.4 Ag + per gram obtained from desorption data (see below) resulted in Γ max = 9.17 Ag + mg/g and K = 0.11 L/mg Ag + (see Table S3 in SI for details).
The Ag + release profile of nanoparticles infused with 4.9 Ag + mg/g in pure water is plotted in Figure   2c (see Table S4 in SI for Ag + ion adsorption measurements). The silver ions adsorbed in the high affinity region ( 35% of the initial concentration) may not be released readily. However the weakly bound Ag + ions adsorbed in the low affinity region are released during application. The release data can be fitted with exponential decay curve, offset by the strongly bound silver ions in the high affinity region (Fig. 2c), Γ( ) = Γ 1 + Γ 0 e − /τ , where Γ 0 is the weakly bound silver load, t is the time and τ the characteristic decay time.
The least square fit resulted in values of Γ 1 = 1.4 Ag + mg/g and τ = 6.2 h (see Table S3 in SI for fitting details). In summary, the data indicate that more than 95 % of the silver ions in the low affinity region are released within the first 24 h, which is consistent with their anticipated application time.
After silver ion infusion, EbNPs-Ag + are electrostatically stabilized and have a negative zeta-potential of -32 mV ± 1 mV (see SI for measurement details). In the last step, the surface charge of the particles was reversed by the adsorption of a cationic polyelectrolyte, polydiallyldimethylammonium chloride (PDAC), which may also delay slightly the release of silver ions from the particles (Fig. S2 in SI) due to formation of electrostatic barrier for their release. Stable EbNPs 0.05 wt% with positive surface potential could be fabricated at weight ratio of PDAC/Indulin AT of 0.20 or higher as shown in Fig. 2d. The TEM image in Fig. 1c displays nanosized non-spherical clusters with sizes below 70 nm. The accumulation of these cationic EbNPs on negatively charged E.coli was visualized by confocal microscopy (Fig. 1d). The ionic silver in these PDAC-coated, Ag + -loaded EbNPs (EbNPs-Ag + -PDAC) could be readily released via desorption near the cell walls, which could explain their strong antimicrobial efficacy reported below.

Antimicrobial activity of EbNPs
The antimicrobial activity of EbNPs-Ag + -PDAC nanoparticles was evaluated within 6 h after their preparation. Quantitative antimicrobial tests were performed on gram-negative E. coli BL21(DE3), a common testing bacteria related to pathogenic E.coli such as O157:H7, which has been genetically modified not to secrete toxin, and P. aeruginosa (ATCC 15442), a highly adaptive pathogen used in standardized disinfectant testing in the food industry 41 . Antimicrobial testing on a quaternary amine-resistant  Table S5 in SI). The EbNPs-Ag + -PDAC exhibited strong antimicrobial activity after 1 min exposure, while the corresponding nanoparticle-free supernatant showed no observable effect. The supernatant showed some CFU reduction after 30 minutes, which can be explained by residual amounts of unbound PDAC or silver. The conventional BPEI-AgNPs and AgNO3 solutions were less effective after 1 min, but after 30 min contact time reached 100% microbicidal efficiency at 50 mg/L Ag + equivalent. EbNPs-Ag + without PDAC coating did not result in significant CFU reduction (5%), which may be attributed to their negative surface charge reducing the particle adhesion to the bacteria.
However, EbNPs-PDAC without silver also showed strong CFU growth suppression, which may be attributed to the antimicrobial effect of the quaternary amine coating (note also that PDAC control reached 100% bacteria reduction after 30 min).
The antimicrobial efficiencies data on P. aeruginosa after 30 min are compared in Fig. 3c. The EbNPs-Ag + -PDAC sample resulted in strong CFU reduction at 5 mg/L Ag + equivalent (this sample reached 100% reduction after 24 h, see Fig. S6 in SI). Interestingly, BPEI-AgNPs did not exhibit observable antimicrobial activity even at 100 mg/L Ag + equivalent. AgNO3 solutions were inactive up to 40 mg/L Ag + equivalent and exhibited low CFU reduction at 60 mg/L Ag + . The control PDAC solutions at 0.02 wt% did not exhibit observable CFU reduction after 30 min contact time, and the PDAC solutions at 0.04 wt%, double the amount used in EbNPs-Ag + -PDAC sample, exhibited strong CFU reduction. However, again the EbNPs-Ag + -PDAC performed best in terms of antimicrobial efficiency, demonstrating higher efficiency than AgNO3 solution containing 12× more Ag + , and BPEI-AgNPs containing 20× higher Ag equivalent. These EbNPs were also 100% effective against the gram-positive bacteria S. epidermidis (see Table S7 in SI).
The CFU reduction data on PDAC-resistant Ralstonia sp. after 30 min contact time are compared in Fig. 3d. The EbNPs-Ag + -PDAC, AgNO3 solutions, and BPEI-AgNPs all resulted in 99%+ CFU reduction at  5 mg/L Ag + equivalent. The EbNPs-Ag + without PDAC coating reached moderate CFU reduction at 10 mg/L Ag + after 30 minutes. PDAC solution at 0.04 wt% did not show any CFU reduction even after 24 h of contact (see Fig. S7 and Table S6 in SI). Similarly, lignin-only EbNP controls did not result in any observable CFU reduction in all tests; on the contrary, they promoted CFU growth to some extent (see Fig.   S8 in SI).
In summary, comparing all active agents tested in terms of broad-spectrum antimicrobial efficiency, we establish that EbNPs-Ag + -PDAC proved most effective against four types of gram-positive and gramnegative bacteria. These EbNP particles also outperformed the common AgNPs and AgNO3 solutions on E. coli and P. aeruginosa while using at least 10× lower Ag + equivalent concentrations. The EbNPs-Ag + -PDAC also outperformed PDAC on Ralstonia sp. The high activity of EbNPs-Ag + -PDAC likely stems from the combination of the antimicrobial properties of Ag + ions and the PDAC coating, especially on cells that are sensitive to quaternary amines. However, for bacteria resistant to charged amines, such as the Ralstonia sp., the effect of EbNPs-Ag + -PDAC is a direct result of the bactericidal activity of silver ions.
Finally, the post-utilization activity of EbNPs-Ag + -PDAC depleted from silver ions was evaluated on PDAC-resistant Ralstonia sp. We simulated the post-application depletion process of silver from EbNPs-Ag + -PDAC, which were originally infused with 5 mg/L Ag + ions, by dialysis in DI water for 24 h. This led to silver ion desorption similar to the one that would occur if the particles were disposed in environmental water. The residual silver ion content in these dialyzed samples was determined to be 0.89 mg/L Ag + ions using Inductively-Coupled Plasma Optical Emission Spectroscopy (ICP-OES), which translates to 18 % of Ag + remaining bound on the particles and is in good correspondence with the release kinetics studies described above. The silver-depleted EbNPs-Ag + -PDAC sample exhibited moderate CFU reduction of multiplication at 96 h, while the non-silver-depleted EbNPs-Ag + -PDAC control had 100% CFU reduction at 24 h (see Fig. S8 in SI). This result indicates that EbNPs-Ag + -PDAC nanoparticles would lose their antimicrobial activity due to Ag + ion depletion when exposed to the environment. The lignin cores depleted from silver ions would in turn be further degraded by the microbial flora present in the environment.

High-throughput screening of EbNPs bioactivity
The biological activity and potential toxicity hazards of the particles were investigated in a comprehensive testing by the U.S. EPA's ToxCast program 44 protocols as described in the Methods. The 156 biological endpoints are described in the Methods. The aggregate assay targets are divided into ten biological function groups (see Fig. S9 in SI), namely zebrafish development toxicity, acute inflammation, cell cytotoxicity, cell proliferation, cellular stress modulation, immune response, nuclear receptor activation, tissue remodeling modulation, and vascular biology modulation. The testing results are summarized in the bioactivity "heat map" in Fig. 4. The data indicate that EbNPs and EbNPs-Ag + -PDAC had response profiles more similar to each other than to AgNP or AgNO3(aq). When evaluated on a total mass basis at Lowest Effective Concentration (LEC) > 0.1 μg/mL, they also affected fewer biological endpoints in all biological functional groups than AgNPs or AgNO3 (aq) (see Fig. S9 in SI). At LEC < 0.1 μg/mL the results show the same trend, although some endpoints (e.g. in immune response) were affected at lower Ag + ion concentration by EbNPs-Ag + -PDAC than by AgNPs. When evaluated by equivalent silver ion concentration, EbNPs-Ag + -PDAC appeared to affect zebrafish embryos at lower silver content than AgNPs (see Fig. S10 in SI).
Overall, the bioactivity screening results show that the lignin-only EbNPs without Ag + loading and the EbNPs-Ag + -PDAC interfered with fewer biological pathways than the AgNO3(aq) and AgNPs. Thus, it could be expected that the EbNPs-Ag + -PDAC would also have a much smaller environmental impact in terms of amount of silver released, as they would contain much less silver by mass than the common AgNPs in antimicrobial application scenarios relating to higher silver "atom economy" 19 . Their impact is likely to be minimized further as the EbNPs, unlike the persistent AgNPs, can rapidly lose their activity due to Ag + ion depletion in the environment or wastewater, followed by natural pathways for the neutralization of the minimal amount of ionic silver 45 . The EbNPs surface modifier, PDAC, commonly used as a primary organic coagulant in waste water treatment 46 , is also likely to be readily neutralized once it interacts with naturally occurring biopolymers 47 . The lignin NPs depleted of Ag + and PDAC would biodegrade similarly to lignin from plant biomass 48,49 . The EbNPs are also likely be more readily degraded and removed than common AgNPs in waste water treatment systems or in waste incineration plants.

Conclusions
We report a class of antimicrobial nanoparticles with biodegradable cores (from Indulin AT lignin), loaded with Ag + ions and coated with cationic polyelectrolyte PDAC. The EbNPs-Ag + -PDAC exhibit broad spectrum biocide action and are capable of neutralizing common gram-negative and gram-positive human pathogens as well as quaternary amine-resistant bacteria, while using 10× less silver when compared with conventional BPEI-AgNPs and AgNO3 aqueous solution. The PDAC coating, which boosts their adhesion to microbial cell membranes, may have some antibacterial activity on its own; however, both EbNP controls without PDAC or without Ag + ion loading demonstrated much lower antimicrobial efficiencies, indicating that both agents act synergistically. The array of high-throughput screening tests on mammalian cells and zebrafish embryos indicate that the EbNPs have decreased impact on the majority of biological endpoints, when compared with AgNPs and AgNO3 by total mass. In addition, the EbNPs-Ag + -PDAC showed timelimited antimicrobial action since they can lose their residual silver ions by post-utilization dilution in water.
We expect that the overall environmental impact of EbNPs-Ag + -PDAC is likely to be significantly lower when compared to AgNPs, however, more research and long-term testing in realistic real-world conditions (such as in water containing salts, surfactants and pollutants) would be needed to fully validate these claims.
In more general plan, these environmentally-benign nanoengineering results illustrate how green chemistry principles including atom economy, use of renewable feedstocks, and design for degradation can be applied to design more sustainable nanomaterials with increased activity and decreased environmental footprint.

Methods
Synthesis of Ag + -infused Indulin AT nanoparticles with PDAC coating. Native Indulin AT lignin nanoparticles were synthesized in ethylene glycol similarly to the method described earlier 38  approximately 1000 to 4400 CFU/mL in nutrient broth (Ralstonia sp. in PDAC 0.02 wt% solution), was added. The samples were continuously vortexed. After the bacteria were exposed to the active agent for a given time, e.g. 30 minutes, the survival rate of the bacteria was determined by plating 100 µL of each sample evenly distributed on Luria-Bertani agar plates. The procedure was repeated after various exposure time. After the plating procedure, the Petri dishes were sealed and incubated upside down for 48 h at 37 °C.
Depletion of silver-infused EbNPs. The EbNPs-Ag + samples for the depletion study were depleted from Ag + in 0. Biological activity and toxicity hazards tests. EbNPs-Ag + -PDAC, EbNPs, AgNO3, and AgNPs of 15 nm diameter (ENPRA NM300) 50 were screened for 156 endpoints in five cell-based platforms used in ToxCast, the U.S. EPA's chemical testing and prioritization program 44 . The bioactivity assays used human primary cell (co)culture and cell lines, rat primary hepatocytes, and developing zebrafish embryos with a diverse range of toxicity early responses and phenotype endpoints. Gold (Au) nanoparticles, included as relatively inert nanomaterial controls, and Ag microparticles affected very few assays ( Fig. S9f and S9g in SI), indicating that particles having the feature of being nanosized (AuNP), or containing Ag alone, do not exhibit broad or high bioactivity.