**3. Discussion**

PANI and P3ABA are promising additives to materials to create contamination resistance surfaces. Factors that may influence the antibacterial efficacy of the surface were explored over short treatment times (up to 4 h). Disrupting transmission pathways through surface decontamination can be best achieved with an antimicrobial agen<sup>t</sup> that kills over a short period of time [71]. The longer a bacterium persists on a surface, the greater the opportunity to be spread [4]. Therefore, rapid decontamination times will decrease the chance that bacteria may be transferred to a new surface before sterilisation is achieved, and will decrease the likelihood of resistance developing [72].

The activity of PANI and P3ABA was determined against *E. coli* and *S. aureus*, representing important pathogens that are found in settings requiring antimicrobial surfaces, such as hospitals and food processing plants. Overall, *E. coli* had greater susceptibility to both PANI and P3ABA in suspension when compared to *S. aureus* (Figure 2). A similar trend was observed for PANI or P3ABA in agar (Figures 4 and 5) and in films (Figure 6). These results demonstrate that PANI and P3ABA are active against the model Gram-negative and Gram-positive bacteria, *E. coli* and *S. aureus*, respectively, in suspension and in different types of surfaces. The differing levels of activity tha were observed against *E. coli* and *S. aureus* highlight how a broad spectrum antimicrobial agen<sup>t</sup> may be more or less effective against a range of bacteria and demonstrates why testing should be done against all the potential target organisms.

The effect of the presence of complex nutrients on the susceptibility of *E. coli* to PANI and P3ABA in suspension was examined. *E. coli* was more susceptible to the antimicrobial action of PANI in suspension when incubated in minimal media when compared to LB broth (Figure 3A). The more efficacious activity of PANI in a low nutrient environment supports the incorporation of this antimicrobial agen<sup>t</sup> in surfaces for applications that are associated with only minor contamination with organic matter. In contrast to this, P3ABA was more active against *E. coli* in rich media relative to minimal media (Figure 3B). The metabolic state of the cell may influence how it responds to bactericidal treatment [73]. The bactericidal action of antimicrobial agents is associated with increased respiration, while bacteriostatic action is characterised by suppressed cellular respiration [73]. The bacteriostatic effect reduces ATP demand and is often the dominant effect blocking bactericidal action [73]. Following from this, if cellular energy output is readily inhibited, such as in cells growing in energy poor conditions, antimicrobial action may result in the inhibition of growth rather than bactericidal killing [73]. Bacterial cells that are highly active may therefore be more susceptible to antimicrobial exposure because of accelerated respiration. The reduced sensitivity of *E. coli* cells to P3ABA in low nutrient conditions could be reflective of a predisposition to the bacteriostatic effect. The greater activity of P3ABA in the presence of nutrients that facilitate bacterial cell growth supports the use of P3ABA in surfaces in settings that are associated with contamination of organic matter, such as surfaces in the vicinity of patients with gastrointestinal infections, which are commonly contaminated with faecal matter containing the bacteria.

The feasibility of using PANI and P3ABA as additives to create antimicrobial surfaces was examined by determining the activity in suspension. Following confirmation of activity against *E. coli* and *S. aureus* (Figure 2), the activity of PANI and P3ABA as agents that are added to absorbent and non-absorbent surfaces was investigated. Overall, both PANI and P3ABA are most active in suspension, followed by in agar and then in films. *E. coli* treated with 0.5% PANI in suspension for 4 h was reduced in numbers by 2 log (Figure 3A), while 1% PANI in agar did not reduce the viable cell count, even after 8 h of treatment (Figure 4A). For surface incorporated PANI to achieve comparable activity to PANI in suspension, a higher concentration is required. This is demonstrated by total knockdown of *E. coli* after a 4 h exposure to 2% PANI in agar (Figure 4A); a result that was achieved by a concentration of 0.5% in suspension (Figure 3A). The reduction in activity of surface incorporated PANI and P3ABA is reflective of how immobilisation in a surface can affect bactericidal activity and how different surface matrixes may influence this in different ways [28].

In this study, the antibacterial activity of PANI and an fPANI were determined by the quantification of the viable cells remaining after a period of challenge, using either classical culture-based techniques, or measuring bioluminescence of genetically modified bacteria as a surrogate measure of viability. Future studies may be enhanced by coupling this type of analysis with scanning electron microscopy (SEM) of bacteria on surfaces and fluorescence microscopy after live/dead staining. SEM has previously allowed for visualisation of bacterial killing by fPANIs to the conclusion that the antimicrobial mode of action eventually leads to a loss of cell integrity [74,75]. Fluorescence microscopy of live/dead stained biofilms has allowed for the activity of another fPANI, polysulfanilic acid, to be followed, with the killing of bacteria being established in biofilms and the release of biomass from the surface, imaged [54]. In the study of bacterial attachment to surfaces real time imaging, e.g., using differential interference contrast microscopy [76] may allow for a better understanding of the interaction of bacteria with surfaces and the factors that influence resistance to colonisation.

It is believed that PANI and P3ABA exert antimicrobial action following contact with a bacterial cell [45,77]. Thus, the reduced contact that occurs between a bacterial cell and surface incorporated PANI and P3ABA (relative to in suspension) would mediate the decrease in antimicrobial efficacy. The least amount of contact between the antimicrobial agen<sup>t</sup> and a bacterial cell would occur for non-absorbent surfaces, which mirrors the decreased activity that was observed for PANI and P3ABA in films. 2% PANI and 2% P3ABA in agar (Figure 5) were able to mediate knockdown of *S. aureus* in 8 h and 15 min, respectively, while 5% PANI and 3% P3ABA in films were unable to reduce bacterial cell numbers after a 24 h treatment (Figure 6C). In this example, higher concentration and treatment time did not ameliorate the reduction of activity for polymers that were incorporated into a non-absorbent surface. The results of this work demonstrate why it is important to test antimicrobial agents, first in suspension (associated with quick and reproducible results) before testing as part of a surface, which should reflect the final application [71].

In real world settings, antimicrobial surfaces may be challenged with a range of inocula. It is well known that the size of the inoculum that is used can influence the magnitude of antimicrobial activity in susceptibility testing [56]. In general, higher inocula need a higher concentration of antimicrobial agen<sup>t</sup> and/or a longer treatment time to achieve knockdown [78]. The surface activity of PANI in film and P3ABA in film was affected by *S. aureus* inoculum size with activity demonstrated only for lower inocula (Figure 7B). The decreased surface activity in the presence of high numbers of bacteria may be mediated by the piling of bacterial cells on top of each other, thereby reducing direct contact with the antimicrobial agen<sup>t</sup> for a portion of the population [79]. The results that are presented demonstrate the necessity to perform antimicrobial surface testing with appropriate inocula to simulate the potential challenges that would occur in the real world application. Surfaces in hospitals are considered to be contaminated when aerobic colony counts exceed 2.5 CFU/cm2; however, sampling of objects in patient hospital rooms has demonstrated contamination with a range of bacterial loads (up to 10<sup>4</sup> CFU/m2, equivalent to 10<sup>2</sup> CFU/cm2), including 10<sup>3</sup> CFU/m<sup>2</sup> (equivalent to 10 CFU/cm2) of MRSA on door handles [80–83]. Therefore, antimicrobial surfaces in hospitals would need to be active against up to 10<sup>4</sup> CFU/m<sup>2</sup> of contaminants to prevent bacterial spread.

Organic soiling of antimicrobial surfaces is a known cause of loss of activity and thus was investigated for surfaces containing PANI and P3ABA [25,71,84]. Surface activity of both PANI in film and P3ABA in film was decreased in the presence of organic matter (Figure 8). Organic matter can interfere with contact between the bacterial cell and the antimicrobial agent—particularly for charged proteins and polysaccharides that can disrupt charge based interactions—thus providing protection from antimicrobial action [28,85,86]. Additionally, contaminating organic matter may inactivate antimicrobial agents [86]. Typical organic contaminants on hospital surfaces include blood and faecal matter [25,71]. It is important that antimicrobial surfaces are tested in conditions, including contamination with organic matter, relevant to the application to verify that the surfaces will be sufficiently active in these settings [25].

While it is not ideal that a reduction in surface activity was observed, the loss of activity upon soiling is common and the effect of organic soiling can be reduced by regular cleaning. Therefore, antimicrobial surfaces need to be able to withstand any adverse environmental conditions that are associated with cleaning [27]. PANI and P3ABA have thermal stability up to 300 ◦C and environmental stability in the conducting form [45–48]. An fPANI containing surface was demonstrated to retain activity against *E. coli* and *S. aureus* after 10 repeated challenges if hydrogen peroxide, but not bleach, was the cleaning agen<sup>t</sup> [87]. Future work will include examining the influence of current cleaning procedures on the activity of surface incorporated PANI and P3ABA.

P3ABA containing surfaces demonstrated potential as contamination resistant surfaces for applications. P3ABA as part of a non-absorbent surface reduced *E. coli* by 2 log after a 24 h incubation (Figure 6B), while an absorbent surface containing 2% P3ABA cleared the bacterial load after 15 min (Figure 4B). The P3ABA containing surfaces in this work indicate a superior performance than has been reported for triclosan, a popular additive claiming antimicrobial activity, which had no effect on the viable cell count of *E. coli* following a 24 h exposure [88]. Similarly, triclosan-incorporated plastic only inhibited *E. coli* O157:H7 after a 24 h incubation [89] and triclosan melt-mixed with 4.5% polystyrene inhibited *E. coli* Y 1090 for 5 h, after which the viable cell number increased [90]. Materials containing P3ABA may therefore have a future as a cost-effective antimicrobial surface to prevent or at least reduce the undesirable spread of micro-organisms.

### **4. Materials and Methods**

### *4.1. Bacterial Strains and Growth Conditions*

*E. coli* ATCC 25922 (referred to as *E. coli* 25922) and *S. aureus subsp*. *aureus* ATCC 6538 (referred to as *S. aureus* 6538) were used in this work because they are routinely used as control organisms to verify that antibiotic susceptibility results are accurate [56,91]. *E. coli* 25922 was tagged with an integrating plasmid (p16S*lux*) containing the bacterial luciferase (*lux*) operon (designated *E. coli* 25922 *lux*) [92,93]. *E. coli* 25922 *lux* was used for testing of surfaces containing PANI and P3ABA [28]. All strains were grown at 37 ◦C, with 200 rpm agitation where appropriate. The University of Auckland Institutional Biological Safety Committee approved the construction and use of genetically modified Enterobacteriaceae (GMO04-UA0027).

### *4.2. Media and Chemicals*

PANI and P3ABA were synthesised via chemical oxidation of aniline and 3-aminobenzoic acid monomers, respectively [45]. Cell biology reagents were purchased from Sigma-Aldrich (New South Wales, Australia). Bacteria were cultured in LB broth (BD) or in minimal media. Minimal A medium was used to support growth in a minimal environment, providing only essential nutrients. A 5× minimal A solution was made according to the following: 5 g (NH4)2SO4, 22.5 g KH2PO4, 52.5 g K2HPO4, 2.5 g sodium citrate·2H2O. After autoclaving, this solution was diluted to 1× with sterile water and the following sterile solutions, per litre: 1 mL 1 M MgSO4·7H2O, 0.1 mL 0.5% thiamine plus the carbon source (10 mL of 40% succinate solution per litre).

### *4.3. Preparation of PANI and P3ABA Suspensions*

PANI was finely ground using a mortar and pestle. This insoluble powder requires shaking at 200 rpm to stay in suspension. Reflecting the improved solubility of P3ABA, this polymer was suspended in broth by sonication (QSonica Q700 Sonicator, Newtown, CT, USA) at the following settings: amplitude 30, elapsed time 10 s, repeat 4×. Suspensions of PANI and P3ABA were prepared at 1% (*w*/*v*) for a final concentration of 0.5%.

### *4.4. Activity of PANI and P3ABA Suspensions against E. coli and S. aureus*

Turbid overnight cultures of test bacteria were diluted to 10<sup>6</sup> CFU/mL in LB broth (*E. coli* 25922 *lux* and *S. aureus* 6538) or minimal A salts with 0.4% succinate (*E. coli* 25922 *lux*) [94]. The inocula were retrospectively enumerated on LB agar plates [64]. 500 μL of PANI suspension, P3ABA suspension, and growth media (untreated cells) were inoculated with 500 μL of diluted culture. At 0.5 h, 1 h, 2 h, and 4 h time points, each experimental sample was enumerated on LB agar plates. Following incubation, colonies were counted and CFU/mL was calculated. At least three biological replicates were obtained.

Linear regression analysis was used to compare the sensitivity of test strains to PANI or P3ABA suspensions. Specifically, the sensitivity of *E. coli* and *S. aureus* in LB broth to each suspension was compared and the sensitivity of *E. coli* in LB broth and in minimal media to each suspension was compared. Statistical analysis by linear regression was performed using GraphPad Prism software version 6 (GraphPad Software, Inc., La Jolla, CA, USA). Data was graphed in a scatter plot that was generated with viable cell counts post-treatment (CFU/mL) represented on the *y*-axis and time (h) represented on the *x*-axis. Linear regression was used to fit a straight line (regression line) through the data for the categorical factor (strain type or media type) generating the best-fit value of the slope and intercept. An analysis of covariance (ANCOVA) was used to compare the regression lines from the categorical factors to determine if there was a statistically significant difference in sensitivity.

### *4.5. Activity of Absorbent Surfaces Containing PANI and P3ABA against E. coli and S. aureus*

Absorbent surfaces containing PANI or P3ABA can be modelled using agar, as drops of liquid containing bacteria will absorb into the agar surface [64]. Molten agar was mixed with varying amounts of PANI or P3ABA, which when left to set created absorbent surfaces containing the antimicrobial agents. PANI or P3ABA were established in agar at 1% and 2%; PANI was also established in agar at 8%. PANI and P3ABA containing absorbent surfaces were set up in triplicate in a 96 well plate by aliquoting 200 μL of each test agar and 200 μL of LB agar (for the untreated control) into individual wells. A turbid culture of test bacteria was diluted to 10<sup>6</sup> CFU/mL in broth and retrospectively enumerated. All of the test surfaces were inoculated with 10 μL diluted culture, resulting in 10<sup>4</sup> CFU in each well [55,56]. Agar samples for background readings received 10 μL LB broth.

At specified time points, bacterial cells were rescued in 200 μL fresh media in a 96 well plate [95]. Each type of absorbent surface was tested for the necessary time to achieve knockdown, therefore, highly active surfaces were tested only for the shorter treatment times. *E. coli* 25922 *lux* and *S. aureus* 6538 were challenged with PANI in agar for the following treatment times: 15 min, 30 min, 1 h, 2 h, 4 h, and 8 h. *E. coli* 25922 *lux* was exposed to P3ABA in agar for the following treatment times: 15 min, 30 min, 1 h, and 2 h. *S. aureus* 6538 was exposed to P3ABA in agar for the following treatment times: 15 min, 30 min, 1 h, 2 h, and 4 h. The viability of rescued *E. coli* 25922 *lux* was assessed after 16 h incubation by measuring bioluminescence using the VICTOR X Multilabel Plate Reader (Perkin Elmer, Foster City, CA, USA). The viability of rescued *S. aureus* 6538 was determined by measuring OD600 using the μQuant ™ Microplate Spectrophotometer (BioTek Instruments, Winooski, VT, USA). Three biological replicates were obtained for each experiment.

The Friedman test was used to analyse the differences between untreated cells and those that were treated with PANI or P3ABA in agar. When a significant difference was identified (*p* value less than 0.05), specific groups were compared to each other using Dunn's multiple comparison test. Dunn's multiple comparison test was used to compare the treated and untreated cells at each time point, with a *p* value of less than 0.05 indicating a significant difference. Thus, comparisons were made between each treatment time for every concentration tested to identify a concentration-contact time combination that is associated with significant surface activity.

### *4.6. Activity of Non-Absorbent Surfaces Containing PANI and P3ABA against E. coli and S. aureus*

Non-absorbent surface samples were prepared using SEBS films containing 5% PANI or 3% PANI or no additive (control film). The films were hole punched to generate ~5 mm diameter circles that fit into the wells of a 96 well plate. The film samples were disinfected by immersion in 70% ethanol for 10 min and dried in the Herasafe ™ KS (NSF) Class II, Type A2 Biological Safety Cabinet (Thermo Scientific, Auckland, New Zealand) [25].

A turbid overnight culture of test bacteria was diluted to 10<sup>6</sup> CFU/mL in broth and enumerated. The activity of the PANI and P3ABA containing film samples was determined using the MSTA adapted from Japanese Industry Standard (JIS Z-2801) method [28,70]. A piece of film was placed in an empty well, inoculated with 10 μL of diluted culture, and a second piece of the same type of film was placed on top of the inoculum [28,70]. Film samples for background readings received 10 μL of LB broth. The film treatments were established in triplicate. At the specified time point(s) bacterial cells were rescued in 190 μL LB broth in a fresh 96 well plate [95]. The rescued cells were incubated at 37 ◦C in a sealed container with moist tissue for 16 h and the viability of rescued cells was determined [95]. Three biological replicates were obtained for each experiment.

For *E. coli* 25922 *lux*, cells were rescued after 2 h and 24 h treatments. The viability of cells post-treatment was assessed by using plate counts and measuring bioluminescence. To this end, a 100 μL aliquot of rescued cells was used to enumerate by drop counts and the remaining 100 μL of rescued cells was added to a dark OptiPlate-96 well microtitre plate containing 100 μL of LB broth for the measurement of bioluminescence using the VICTOR X Multilabel Plate Reader. For *S. aureus* 6538, cells were exposed to film treatments for only 24 h and the viability of cells post-treatment was

assessed by incubating 200 μL of rescued cells in a 96 well plate for 16 h and measuring OD600 using the μQuant™ Microplate Spectrophotometer.

The activity of PANI and P3ABA in films against *E. coli* 25922 *lux* was analysed using a two-way repeated measures analysis of variation (2-way RM ANOVA). For both the plate counts and the bioluminescence data, the 2-way RM ANOVA determined how *E. coli* 25922 *lux* cell number was affected by two factors, treatment time (2 h and 24 h) and film type (PANI in film, P3ABA in film, no additive). A *p* value of less than 0.05 indicates that the cell number was significantly affected by at least one of the factors. When a significant difference was identified, treated cells were compared to the untreated control for each time point using Dunnett's multiple comparison test with a *p* value of less than 0.05, indicating a significant difference.

The Friedman test was used to analyse the differences between untreated *S. aureus* 6538 cells and those that were treated with PANI or P3ABA in film. The Friedman test is a nonparametric test that compares three or more matched groups—cells treated with 5% PANI in film, 3% P3ABA in film, and control film. A *p* value of less than 0.05 indicates that at least one of the groups differs from the rest. When a significant difference was identified, specific groups were compared to each other using Dunn's multiple comparison test. Dunn's multiple comparison test was used to compare the treated and untreated cells at each time point, with a *p* value of less than 0.05 indicating a significant difference.

### *4.7. Characterisation of the Action of PANI and P3ABA Films against E. coli and S. aureus*

4.7.1. Challenge of PANI and P3ABA Films with a Range of CFU Doses of *E. coli* 25922 *lux* and *S. aureus* 6538 in Saline

Film punches were prepared and decontaminated, as described above. The MSTA was performed with bacterial challenges (10<sup>4</sup> CFU, 10<sup>3</sup> CFU, 10<sup>2</sup> CFU, and 10 CFU) prepared in 10 μL saline. The 10<sup>6</sup> CFU/mL culture was enumerated. Following a 2 h treatment, cells were rescued in 190 μL LB broth and incubated in a fresh 96 well plate for 16 h. Viability of bacteria was assessed by measuring bioluminescence for *E. coli* 25922 *lux* and by measuring OD600 for *S. aureus* 6538. The activity of PANI and P3ABA in films against a range of CFU doses of *E. coli* 25922 *lux* and *S. aureus* 6538 was analysed using a 2-way RM ANOVA.

4.7.2. Assay to Evaluate the Influence of the Presence of Organic Matter on the Activity of PANI and P3ABA Films against *E. coli* 25922 *lux*

Film punches were prepared and decontaminated, as described above. MSTA was performed with bacterial challenges (10<sup>4</sup> CFU) in 10 μL saline or 10 μL LB broth. The inocula were enumerated. Following a 2 h treatment, cells were rescued in 190 μL LB broth and incubated in a fresh 96 well plate for 16 h. The viability of rescued cells was determined by measuring the bioluminescence.

### *4.8. Appendix A Methods*

### 4.8.1. Validation of Utilisation of *E. coli* 25922 *lux*

To examine if *E. coli* 25922 *lux* can be used for the testing of surfaces containing PANI and P3ABA, in place of the non-tagged version, the MIC and MBC of both strains were determined [55]. A range of concentrations of PANI and P3ABA in suspension were tested (0.03125–4%). The suspensions were established at 2× the final desired concentration in 500 μL. The insolubility of PANI required each suspension to be set up separately by weighing the powder into 5 mL tubes and adding 500 μL LB broth. P3ABA suspensions were established from a stock solution using a doubling dilution series. 500 μL LB broth was aliquoted to set up an untreated control.

The PANI and P3ABA suspensions were inoculated with 500 μL of 10<sup>6</sup> CFU/mL of *E. coli* 25922 and *E. coli* 25922 *lux*. The MIC was defined as the lowest concentration of PANI or P3ABA that was able to inhibit the visible growth of test bacteria following a 24 h treatment [55,56]. Tubes that were observed by eye to have no visible growth were selected for MBC testing. For this, 20 μL of the

experimental sample was spread onto six LB agar plates [55,56]. The spread plates were incubated at 37 ◦C for 16 h and the growth on these plates was determined. When countable colonies were present, the CFU/mL of the sample was calculated. The MBC was defined as the lowest concentration of PANI or P3ABA that either totally prevents growth or results in a ≥99.9% decrease in the initial inoculum following subculture on LB agar plates [55,96]. At least three biological replicates were obtained.

### 4.8.2. Determination of the Limit of Detection for *E. coli* 25922 *lux* and *S. aureus* 6538 Growing in a 96 Well Plate

The limit of detection of *E. coli* 25922 *lux* and *S. aureus* 6538 growing in a 96 well plate was examined by determining the lowest number of cells added to LB broth in a 96 well plate that can grow to detectable levels [95]. This was achieved by serially diluting an overnight culture in triplicate in a 96 well plate by transferring 20 μL of culture into wells containing 180 μL of LB broth. A range of inocula were established from ~10<sup>9</sup> CFU/mL to ~1 CFU/mL. The overnight culture was enumerated to confirm the cell numbers that were tested. The 96 well plate was incubated at 37 ◦C for 16 h in a sealed container with a moist tissue [95]. Growth of bacteria was assessed by measuring bioluminescence using the VICTOR X Multilabel Plate Reader for *E. coli* 25922 *lux* and by measuring OD600 using the μQuant™ Microplate Spectrophotometer for *S. aureus* 6538.
