Antimicrobial biopolymer formation from sodium alginate and algae extract using aminoglycosides

Antimicrobial biopolymers provide a biodegradable, sustainable, safe, and cheap approach to drug delivery and wound dressing to control bacterial infection and improve wound healing respectively. Here, we report a one-step method of making antimicrobial alginate polymer from sodium alginate and aqueous extract of Wakame using antibiotic aminoglycosides. Thin layer chromatography of commercially available sodium alginate and Wakame extract showed similar oligosaccharide profiles. Screening of six aminoglycosides showed that kanamycin disulfate and neomycin sulfate produces the highest amount of biopolymer; however, kanamycin disulfate produces the most malleable and form fitting biopolymer. Image texture analysis of biopolymers showed similar quantification parameters for all the six aminoglycosides. Weight of alginate polymer as a function of aminoglycoside concentration follows a growth model of prion protein, consistent with the aggregating nature of both processes. Slow release of antibiotics and the resulting zone of inhibition against E. coli DH5α were observed by agar well diffusion assay. Inexpensive method of production and slow release of antibiotics will enable diverse applications of antimicrobial alginate biopolymer reported in this paper.


Introduction
According to the World Health Organization (WHO), the emergence of multidrug resistance among bacterial pathogens is a global public-health challenge [1]. Recently, WHO has listed Acinetobacter baumannii [2,3], carbapenem-resistant Pseudomonas aeruginosa [4,5], and carbapenem-resistant Enterobacteriaceae [6,7] as critical priority pathogens. These pathogens are responsible for infections of burn, wound, blood stream, nervous system, urinary and respiratory tracts. In this context, antimicrobial biopolymer synthesis is one of the current areas of antimicrobial drug delivery research to control bacterial infection in biomedical devices [8], wound healing [9], food packaging [10], textiles [11], cosmetic products [11], and water treatment systems [12]. Furthermore, antimicrobial biopolymers are safe, less toxic and more efficacious as compared to the low molecular weight quaternary ammonium compounds [13]. While some polymers have intrinsic antimicrobial activity [14], others have antimicrobial compounds PLOS

Oligosaccharides detection using TLC, quantification of polymerization efficiency, and quality of alginate polymers
We used TLC to check the oligosaccharide profiles [37] of commercial SA and aqueous Wakame extract (Fig 1a). Similar TLC profiles confirm the similarities of oligosaccharide compositions of SA and Wakame extract. For quantifying polymerization efficiency, we measured OD at 600 nm using a plate reader to quantify turbidity due to alginate polymerization to screen six aminoglycosides (Fig 1b and 1d). Optical density (OD), affected by both scattering and absorption, is a common technique to quantify cell growth, fluorophore concentration, and turbidity due to suspended particles [38]. OD of the polymerization reactions at varying concentrations of aminoglycosides showed initial growth followed by saturation. We fitted the OD measurements to y = ax / (k + x), a general equation describing ligand binding to substrate where k is the binding affinity [39]. Fit parameters for aqueous solution of sodium alginate (Fig 1b)  . We used k as the measure of efficiency of aminoglycosides in polymerizing alginate. Neomycin and kanamycin disulfate resulted in the best polymerization. We selected KDS for further testing because it is relatively inexpensive and made alginate biopolymer from SA and Wakame extract, both of which resulted in good quality polymer (Fig 1f and 1g); however, SA solution provided more malleable (Fig 1f, left panel) and form fitting (Fig 1g, right panel) polymer.

Melting temperature of polymer
We measured the melting temperature by leveraging temperature-dependent dissolution of alginate polymer. Polymer was incubated in 1 M NaOH solution to break ionic bonds in polymer and dissolve. Reaction was centrifuged and absorption spectrum of the supernatant was measured. Experiment was repeated at different temperatures (Fig 2a). Integrated absorption, i.e., area under the absorption spectrum, as a function of temperature (Fig 2b) was fitted to a sigmoidal function, y = a / (1 + exp(T m − T) / σ), where T m is the melting temperature and σ is the width. Fit parameters are: a = 130.8 ± 4.3(std), T m = 34.7 ± 2.5(std), and σ = 29.4 ± 2.1(std).

Microscopy to visualize and quantify polymer texture
To visualize the texture of alginate polymers, we performed reactions on microscopes slides and imaged using an optical microscope at 10X magnification as shown in Fig 3. Since biopolymer texture can affect surface reactions and absorption of ligands [40], quantitative analysis of polymer is useful and has been reported for alginate and chitosan films [41]. To avoid subjective bias in visual observation, we quantified the polymer texture using four quantitative parameters (energy, E; contrast, C; homogeneity, H; and entropy, S) of Gray Level Co-Occurrence Matrix (GLCM) algorithm [42] and one quantitative parameter (fractal dimension, F) of Shifting Differential Box Counting (SDBC) algorithm [43,44] (see Methods for definitions and detailed procedures). These five parameters have been previously used for texture analysis of alginate gel images [41]. We calculated E, C, H, S, and F for nine 640 pixel × 480 pixel images for each condition. Table 1 shows the quantitative parameters for different aminoglycosides. To determine polymer texture at higher resolutions, we imaged KDS-based alginate polymer using SEM ( Fig  4). Wet polymer is porous (Fig 4a), whereas dry polymer is not porous (Fig 4b).

Dry weight of pellet after centrifugation of polymerization reaction to quantify the total amount of polymer
Quantitative weight-based assay approach was used to quantify the optimum amount to antibiotic and sodium alginate for polymer synthesis. Different concentrations were tested with a fixed concentration of sodium alginate and algae extract. To quantify the total amount of alginate polymer created after adding aminoglycosides to alginate solutions, the reactions were centrifuged. Supernatants were decanted and pellets were dried. Dried polymer weights were measured for different concentrations of aminoglycoside ( Fig 5). We fitted the experimental polymer weights against concentrations ( Fig 5) to a model similar to prion protein growth [45]. Best fit parameters (Table 2) were used to calculate the dry weight of alginate polymer per mg of antibiotics.

Measurement of zone of bacterial inhibition to quantify antimicrobial property of alginate-aminoglycoside polymer against E. coli DH5α
Aminoglycoside biopolymers were prepared in microcentrifuge tubes and after centrifugation, pellet was washed three times with sterile deionized water to remove any free antibiotic. After washing, the polymer pellet was carefully removed using a spatula and placed at the center of     Table 3.

Biocompatibility assay
We grew COS-1 cells on alginate polymer. As shown in Fig 7, COS-1 cells attached, grew, and formed a nest of polymer. Presence of viable cells clearly indicated that aminoglycoside-based alginate polymer is not toxic to COS-1 cells.

Sodium alginate polymerizes due to acid-base reaction mechanism
Algae extract and commercially available sodium alginate showed similar polymerization behaviors with aminoglycosides. The most likely mechanism of polymerization is interactions between sodium ions in alginate polymer and sulfate ions in aminoglycoside antibiotics.  Immediately after addition of aminoglycoside in sodium alginate, the sulfate ions bind the sodium ions to form sodium sulfate and aminoglycoside binds to alginate via amine-carbonyl interactions. The addition of polymer in sodium hydroxide solution reversed the interaction, which was proved by adding sodium hydroxide in polymer pellet followed by heating for 30 min (Fig 2).

Polymerization efficiency depends on the type of aminoglycoside
We considered the cost of aminoglycosides and the amount of alginate polymer produced using 1 mg of aminoglycoside to determine the efficiency of polymerization, which follows the sequence GS>NS>SS>TS>KS>KDS in decreasing order of efficiency. However, KDS results in the most malleable and form fitting polymer. According to the mechanism of polymerization described before, both the number of amines and sulfates should affect polymerization. While there is a general trend to support the importance of amines and sulfate, the order does not follow exactly, which suggests other factors in polymerization that we do not know yet.

Growth of alginate polymer is similar to a model of prion protein growth
In general, alginate undergoes complex hierarchical crosslinking and aggregation [46] typical of polysaccharides [47]. To simplify, we modeled alginate polymerization [46] similar to the pathogenic form of prion protein [45] because both are unbranched linear polymers and form stable aggregates. Experimentally, we waited 30 min to account for the initial time-dependent polymerization so that the total amount of time-independent alginate polymer can be described by [45]: where y is the total alginate polymer, y(0) is the initial size of alginate polymers, n represents the minimum size of stable alginate polymers, r1 and r2 are the growth and dissociation rates of alginate polymers, and b is the breakage rate of a alginate polymeric chain. As shown in Fig  5, this model fits polymer weights well as a function of concentrations for all the aminoglycosides.

Heat-induced dissociation results in slow release of aminoglycoside from alginate biopolymers
Analogous to hydrogen bonds in double-stranded DNA, we posit that aminoglycoside-based sodium alginate polymer is held together by ionic bonds as a result of acid-base reaction. Therefore, bonds in alginate polymer are expected to break in NaOH solution in a temperature-dependent manner as shown in Fig 2. Indeed, the measured melting temperature T m = 34.7 ± 2.5(std) of alginate polymer (Fig 2b) is of the same order as that of DNA. Since the alginate polymerization occurs due to electrostatic interactions in water (Fig 8), the bond strength is similar to hydrogen bonds [46]. As such, thermal breathing of bonds between aminoglycoside and alginate is possible similar to the thermal breathing of hydrogen bonds in DNA [48]. Thermal breathing along with any base in the media that can create competing ions provides a plausible mechanism of the slow release of aminoglycoside from alginate biopolymers. Additionally, when a piece of aminoglycoside-based alginate polymer is placed at the center, aminoglycosides stochastically may detach from alginate due to thermal breathing and quickly diffuses away due to the concentration gradient. Since streptomycin sulfate has only two amines, it detaches easily from alginate polymer and leads to highest zone of inhibition (Fig 6). In contrast, neomycin sulfate has six amines, which leads to more chance of attachment after detachment due to thermal breathing. As a result, neomycin sulfate provides more stable alginate polymer and slower release leading to smaller zone of inhibition (Fig 6). Release time can possibly be controlled by the number of sulfate ions in aminoglycosides because an argument similar to the number of amines can be made regarding more reversible detachment with more sulfate ions. However, further studies are needed to confirm these hypotheses.

Aminoglycoside-based alginate polymer as potential niche for tissue engineering
Alginate is an well-established naturally derived alternative to synthetic hydrogels such as polyethylene glycol (PEG) and acts as an excellent extracellular mimics to provide the biological cues to cells and surrounding tissue [49]. Both alginate and alginate functionalized with an RGD peptide (RGD-Alginate) with enhanced cell adhesion have been used as extracellular matrix analogs for tissue engineering [50][51][52][53][54][55][56][57][58][59][60]. Flexible (Fig 1f and 1g), porous (Fig 4a), and biocompatible (Fig 7) aminoglycoside-based alginate polymer developed in this paper provides an easy alternative niche for tissue engineering without added chemicals.

Conclusion
In conclusion, we have described a method of preparing biocompatible antimicrobial alginate polymer from aqueous solution of commercial sodium alginate and aqueous extract of Wakame using aminoglycoside antibiotics. The underlying acid-base mechanism involves interactions between negatively charged oxygen due to dissociated sodium ions in alginate and protonated amine in aminoglycosides. Polymerization efficiency seems to loosely correlate with the number of amines and sulfate ions in aminoglycosides. Slow release of aminoglycosides from alginate polymers is evident from the microbial zone of inhibition. Antimicrobial alginate polymers from Wakame, one of the most invasive species in the world that grows in diverse conditions of vast oceans, provides a sustainable and biodegradable alternative for wound dressing with slow release of antibiotics.

Alginate extraction from Undaria pinnatifida (Wakame) using mechanical grinding and orbital shaking
Commercially available dry Wakame algae leaves (Amazon.com) were converted into powder using a coffee grinder. 20 g of dry powder was added to 500 ml water in an Erlenmeyer flask and incubated at 37˚C for 15 hr with 250 rpm orbital shaking. Following incubation, aqueous sodium alginate supernatant was collected after centrifugation at 10000 rpm for 10 min. All steps were performed under sterile conditions to avoid microbial contamination. The supernatant was used for making alginate polymers using aminoglycoside antibiotics.

Thin layer chromatography (TLC)
Sodium alginate and algae extract were spotted on TLC plates (Millipore, Cat# Hx71642853, TLC silica gel, 60 aluminum sheets, 20×20 cm) at several places using 5, 10, 15, 20, 25, and 30 μl of solutions. TLC plates were developed using a mixture of 1-butanol, formic acid, and water in 4:6:1 (v:v:v) ratio. The developed TLC plates were heated at 110˚C for 5 min after spraying with 10% (v/v) sulfuric acid in ethanol to test for the presence of alginate oligosaccharides [37].

Measurement of melting temperature of alginate polymer
Sodium alginate (20 mg/ml) was mixed with kanamycin disulfate (50 mg/ml) in a volume ratio of 1:1 ratio (10 ml each). Resultant solution with alginate polymer was centrifuged at 10,000 rpm for 10 min. The supernatant was discarded and the pellet was washed three times with 5 ml of sterile 0.1 M (pH 7.4) phosphate-buffered saline (PBS) buffer. Six pellets were made and incubated with 2 ml of 1 M NaOH at 10˚, 20˚, 40˚, 60˚, 80˚and 100˚C for 30 min. After incubation, tubes were centrifuged at 10000 rpm for 10 min and absorption spectra of supernatants (Fig 2) were measured using a spectrophotometer (Thermo Scientific Evolution 260 Bio Spectrophotometer).
Imaging and analysis of alginate polymer texture using light microscope 100 μl of 10 mg/ml aqueous solutions of antibiotics were mixed drop-wise with either 100 μl of algae extract or 100 μl of 10 mg/ml sodium alginate on a clean glass slide. The immediate polymerization reaction was observed and imaged using a light microscope at 10X magnification (Fig 3). To describe the polymer texture quantitatively, we converted the images into 8-bit gray scale images with 256 gray levels. Using the gray scale values, i, we determined four quantitative parameters of the Gray Level Co-Occurrence Matrix (GLCM) using the ImageJ Texture Analyzer plugin [61]: the energy, E = ∑p(i, j) 2 ; the contrast, C = ∑|i − j| 2 p(i, j); the homogeneity, H = ∑p(i, j) / (1 + |i − j|); the entropy, S = −∑p(i, j)log[p(i, j)]; where the sums are for all distinct gray scale values and p(i, j) is the (i, j)th element of the normalized gray scale spatial dependence matrix. Three images were chosen for each biopolymer, converted to 8-bit images, and analyzed with step size of 1 pixel at 0˚angle. From the same 8-bit images, the fractal dimension was measured using the FracLac plugin for ImageJ by using the default settings of the FracLac program [62]. All five parameters with standard deviations are given in Table 1.

Scanning Electron Microscopy (SEM) of alginate polymer
Biopolymer was made by mixing sodium alginate and kanamycin disulfate in a volume ratio of 1:1. The polymer suspension was spread on a clean glass slide and imaged using Phenom Pro-Scanning Electron Microscope (Fig 4). It should be noted that drying can lead to a loss of polymer flexibility and therefore, should be imaged quickly.

Quantitative weight-based assay of polymer formation
To quantify polymer formation, 500 μl of either 10 mg/ml aqueous sodium alginate solution or aqueous algae extract was added to a microcentrifuge tube and antibiotics were added at varying final concentrations. The reactions were vortexed and incubated for 10 min at room temperature. After incubation, tubes were centrifuged at 10000 rpm for 15 min and polymer pellets were weighed using an analytical balance. Weight of polymer pellets in mg as a function of aminoglycoside concentrations in mg/ml were plotted (Fig 5), which were fitted to a model that also described the growth of pathogen prion protein growth [45]. Fit parameters are given in Table 2.

Antimicrobial activity of alginate polymer
Antimicrobial activity of polymers was checked against E. coli DH5α using agar diffusion assay [63]. Single colony of E. coli was inoculated in 10 ml of Luria Broth (LB) and incubated till OD600 reached 0.5 (log-phase culture). 100 μl of bacterial culture was spread over LB agar plates. To check the antimicrobial activity of the antibiotics, wells were made by punching a hole in LB agar plates. For reference plates, 50 μl of 10 mg/ml aqueous stock solution of each antibiotic was added in the wells (Fig 6, top row). For experiments with alginate polymers, polymer pellets were washed 3 times with sterile DI water to remove any unbound free antibiotics. Each washing step included the addition of 500 μl of sterile DI water and centrifugation at 10000 rpm for 15 min. After washing, polymer pellets were placed at the center of the LB plates with E. coli DH5α. Plates were incubated at 37˚C for 18 hr. Zone of inhibition, i.e., the area with no growth of bacteria around well or polymer pellet was measured using a ruler ( Fig  6, middle and bottom rows; Table 3).

Biocompatibility assay
We used COS-1 cell lines to test biocompatibility of antimicrobial alginate polymer. Sterile stock solutions of sodium alginate (10 mg/ml) and kanamycin disulfate (100 mg/ml) were mixed in a volume ratio of 1:1 in microtiter plate wells under sterile conditions. After 30 min incubation at 20˚C, thin biopolymer layers were visible at the bottom of microtiter wells. We washed wells three times with 2 ml of sterile 0.1 M (pH 7.4) PBS buffer to remove unbound polymer. 1.5 ml of culture stock of COS-1 cells were added to the wells under sterile condition and incubated for 3 days at 37˚C at 5% CO 2 environment. After incubation, plates were imaged before and after staining with methylene blue using a light microscope and analyzed using ImageJ (Fig 7).