Cold Atmospheric Plasma-Activated Composite Hydrogel for an Enhanced and On-Demand Delivery of Antimicrobials

We present the concept of a versatile drug-loaded composite hydrogel that can be activated using an argon-based cold atmospheric plasma (CAP) jet to deliver both a drug and CAP-generated molecules, concomitantly, in a tissue target. To demonstrate this concept, we utilized the antibiotic gentamicin that is encapsulated in sodium polyacrylate (PAA) particles, which are dispersed within a poly(vinyl alcohol) (PVA) hydrogel matrix. The final product is a gentamicin-PAA-PVA composite hydrogel suitable for an on-demand triggered release using CAP. We show that by activating using CAP, we can effectively release gentamicin from the hydrogel and also eradicate the bacteria effectively, both in the planktonic state and within a biofilm. Besides gentamicin, we also successfully demonstrate the applicability of the CAP-activated composite hydrogel loaded with other antimicrobial agents such as cetrimide and silver. This concept of a composite hydrogel is potentially adaptable to a range of therapeutics (such as antimicrobials, anticancer agents, and nanoparticles) and activatable using any dielectric barrier discharge CAP device.

2 S1: Media, buffers and stock solutions used in this study:

S3: Kirby-Bauer (KB) test assay
Single colonies were inoculated into 10 mL of MHB and incubated for 18 h at 37C with agitation at 200 rpm. Cultures were spun down at 10,000 rpm and pellet resuspended in PBS, subcultures were made in 10 mL of MHB and grown to OD 600 0.5. 100 L of bacterial culture was added to MHA and spread to create to create a lawn. Gel discs of 10 mm in diameter were placed in the centre of the lawn using sterile forceps. Gels were then activated or no activated as specified. After treatments plates were incubated statically for 24 h at 37C. The zone of inhibition (ZOI) was measured using a ruler and blank corrected to the diameter of the individual disc to account for any variation.

S4: Minimum Inhibitory Concentration (MIC)
Single colonies were inoculated into 10 mL of MHB and incubated for 18 h at 37C with agitation at 200 rpm. Cultures were spun down at 10,000 rpm and pellet resuspended in PBS, subcultures were made in 10 mL of MHB and grown to OD 600 0.01. 200 L of gentamicin antibiotic solution was added into the wells and serially diluted into 100 L, two-fold across a 96-well plate into MHB. 100 L of bacteria was then added to each well of these wells. As a negative control and blank, 200 L of bacteria solution and broth was added to separate wells, respectively. Plates were incubated for 18 h at 37C and MIC was read by eye and defined by the lowest concentration that inhibits bacterial growth denoted by lack of bacterial turbidity.

S5: Bacterial biofilm formation
Sterile discs (diameter 19 mm) were placed onto BHIA and UV sterilized for 10 minutes. 20 L of artificial wound fluid (AWF) (1:1 peptone water to fetal calf serum) was aliquoted onto the discs and spread and left to dry. Bacterial overnights were spun and resuspended in PBS and corrected to 0.01 OD 600 , 30 L of bacterial subculture was then added onto the disc. These were then incubated at 37C for 8 h and then removed. Discs were cut to a diameter of 20 mm to completely cover the biofilm and treated accordingly. Biofilms were then incubated for a further one hour to allow for killing effect of released antimicrobial. Biofilms were then removed from the BHIA using sterile forceps and placed into 5 mL of PBS in a 15 mL falcon tube. Tubes were then vortexed for 1 minute and sonicated for 15 minutes. This is repeated once. This is then serially diluted in PBS (0 -10 -7 ) and plated out on TSA or LB plate to enumerate viable cells.

S6: Ninhydrin assay for gentamicin release
Gentamicin sulphate (Sigma) was prepared to a range of concentrations 100-1000 g/mL in pH7.4 buffer solution. 500 L of antibiotic into an Eppendorf to which add 500 L of 1% ninhydrin solution was added. These are then incubated in 95C heat block for 45 minutes.

5
After cooling it down to room temperature, 200 L was dispensed into wells of 96-well plate and absorbance was measured at 540 nm.
S7: Gentamicin calibration curve Figure S1. Standard curve for the quantification of gentamicin using ninhydrin. Error bars represent the standard deviation (N=3).

S8: Loading of silver (Ag + ) in composite hydrogel system
Ag + loading in PAA particles were characterized by using scanning electron microscopyenergy dispersive X-ray (SEM-EDX). It was demonstrated by loading 1mg ml -1 silver nitrate in 10 mg of PAA (the protocol was same as the gentamicin loading described in the manuscript). The silver nitrate loaded dry particles were cut into half and analyzed using JEOL 6 JSM-7900F field emission scanning electron microscope ( Figure S2). An EDX spectra was generated at a specific location (box-marked in Figure S2) as shown in Figure S3. A peak at 3 keV corresponding to Ag confirmed the presence of Ag in the PAA particle. Figure S2. SEM image of the cross-sectioned PAA particle showing the specific area (marked as spectrum 3) on which the EDX analysis was performed. Figure S3. EDX spectrum from the cross-sectional surface of a silver nitrate-loaded PAA 7 particle. The spectra was obtained at an accelerating voltage of 10 kV, magnification of 320 times and a working distance of 10.1 mm.

S9: pH and Conductance change post CAP treatment
As shown in Figure S4A, a significant increase in conductance from ca. 15 to 119 µS cm -1 was observed as CAP treatment time increased. No change in conductance was observed n untreated water (control). On the other hand, pH value decreased from 7 to ca. 4 in a short treatment time of 4 minutes and further to ca. 3 after 20 mins of treatment ( Figure S4B).

S10: ZOI of CAP-activated polymyxin-B and cetrimide-loaded PAA-PVA particles
Cetrimide or polymyxin-B solution at 10% or 0.1% w/v, respectively, was prepared in DI water before adding to 1% w/v PAA in a round bottom flask. This followed the same drug loading procedure described in the manuscript. KB test assay was performed as described in section 8 S3. ZOI of drug-loaded composite hydrogels (polymyxin-B or cetrimide gels) with and without CAP-activation was compared to unloaded composite gels (PVA/PAA Gel) with CAPactivation. As shown in Figure S5A, no P. aeruginosa killing was observed with PVA/PAA gel even after CAP treatment. However, addition of cetrimide or to the gels in case of both untreated and CAP-treated cetrimide gels leads to bacterial killing. However, with CAP action a significantly higher ZOI (15 mm) was observed, cf. cetrimide only gel (ZOI = 5mm). This highlights the role of CAP in driving the cetrimide through the gel. A similar effect was observed in S. aureus wherein cetrimide-loaded gels when activated with CAP exhibited maximum ZOI ( Figure S5B). Polymyxin-B gels ( Figure S6) result in a very low passive release, potentially because of stronger (additive) interaction with the PAA, as it has multiple cationic groups but larger ZOIs following CAP application.

S11: Release of dendrimers from CAP-activated PAA-PVA particles
Generation 1 PAMAM dendrimer with ethylenediamine core (Sigma Aldrich) at a concentration of 0.213 mM in DI water was added to PAA particles (1% w/v) and left at room temperature for 1 hr. The particles were then treated with CAP jet (direct) and plasma-activated water (indirect). Blank and 0.1 mg/ml PAMAM solution was used as a negative and positive control, respectively. Ninhydrin assay was performed as described earlier. As shown in Figure   S7, direct CAP treatment resulted in purple color formation indicating the release of dendrimers into the solution and interaction with ninhydrin reagent to form a purple color. Absorbance measurements indicate a high absorbance with direct CAP treatment c.f. untreated and indirect CAP treatment.

S12: Encapsulation of anionic dye in PAMAM-loaded PAA particles
To validate whether PAMAM-loaded PAA can be used as a delivery vehicle for anionic agents (otherwise difficult to load in PAA particles), an anionic dye tartrazine (Sigma-Aldrich) was investigated. PAMAM:dye concentration was 1:10 in 0.1 g of PAA particles. The mixture was left at room temperature for 1 hr. The particles were then washed with copious amount of water over Buchner's funnel under vacuum. As shown in Figure S8, dye loaded in PAA particles (left) washed out completely as the particles fail to encapsulate the anionic agent. However, when trapped in the dendrimers (right), the dye remained trapped in the PAA particles. Figure S8. Proof of principle demonstrating the ability of PAA particles to encapsulate anionic dye -tartrazine in presence of dendrimers.