Dual-wavelength photo-killing of methicillin resistant Staphylococcus aureus

24 With the effectiveness of antimicrobials declining as antimicrobial resistance continues to 25 threaten public health, we must look to alternative strategies for the treatment of infections. In 26 this study, we investigated an innovative ‘drug-free’ dual-wavelength irradiation approach that 27 combines two wavelengths of light, 460 nm and 405 nm, against methicillin resistant 28 Staphylococcus aureus (MRSA). MRSA was initially irradiated with 460 nm light (90-360 J/cm) 29 and subsequently irradiated with aliquots of 405 nm light (54-324 J/cm). For in vivo studies, 30 mouse skin was abraded and infected with approximately 10 CFU of MRSA and incubated for 3 31 hours before irradiating with 460 nm (360 J/cm) and 405 nm (342 J/cm). Naïve mouse skin 32 was also irradiated to investigate apoptosis. We found that staphyloxanthin, the carotenoid 33 pigment in MRSA cells, promoted resistance to the antimicrobial effects of 405 nm light. In 34 addition, we found that the photolytic effect of 460 nm light on staphyloxanthin attenuated 35 resistance of MRSA to 405 nm light killing. Irradiation of 460 nm alone did not elicit any 36 antimicrobial effect on MRSA. In a proof-of-principle mouse skin abrasion infection model, we 37 observed significant killing of MRSA by the dual-wavelength irradiation approach. However, 38 when either wavelength of light was administered alone, no significant decrease in bacterial 39 viability was observed. Moreover, exposure of the dual-wavelength irradiation to naïve mouse 40 skin did not result in any visible apoptosis. In conclusion, dual-wavelength irradiation strategy 41 may offer an innovative, effective and safe approach for the treatment of skin infections caused 42 by MRSA. 43

In recent years, the threat of antimicrobial resistance has become one of the most important 47 concerns to public health. Infection outbreaks that result from multidrug resistant organisms that 48 have emerged remains a significant problem 1,2 . Staphylococcus aureus infections are amongst 49 the most important threats that can result in skin and soft-tissue infections, with methicillin 50 resistant Staphylococcus aureus (MRSA) being particularly important 3,4 . Therefore, novel, non-51 traditional approaches must be explored to quell these negative effects. Over the years, 52 antimicrobial blue light (aBL) at 405 nm wavelength has been emerging as a potential 53 alternative treatment for localized infections 5 . The accepted mechanism responsible for the 54 antimicrobial effects of aBL (405 nm) is through excitation of endogenous photosensitizing 55 porphyrins and the subsequent generation of singlet oxygen resulting in lipid peroxidation, DNA 56 damage, cell wall damage and cellular apoptosis of microbial cells 5 . However, previous studies 57 and the preliminary results in our laboratory showed that MRSA is much more tolerant of aBL at 58 405 nm than most of other species 6 . Recent findings by other groups have demonstrated the 59 antioxidant properties of staphyloxanthin (STX), which is a membrane-bound carotenoid 60 pigment of MRSA, responsible for its characteristic golden colonial phenotype 7,8 . In addition, it 61 has been demonstrated that STX is subject to photolysis through 460 nm light exposure, 62 rendering it more susceptible to H 2 0 2 mediated killing 9 . Therefore, we theorized that the limited 63 antimicrobial efficacy of aBL (405 nm) we have observed with MRSA was a direct result of STX, 64 as this is too a known singlet oxygen scavenger. As a result, we hypothesized that treatment of 65

STX photolysis alone mediated sensitization of MRSA to 405 nm aBL 94
To determine whether the enhancement in MRSA CFU reduction by 460 nm light was solely 95 based on STX photolysis or whether there is another underlying mechanism, Pig1:: ΔcrtM and 96 its parental WT strain were exposed to 405 nm light or the 460 nm + 460 nm combination. No 97 significant difference in CFU reduction was identified when the S. aureus Pig1:: ΔcrtM mutant 98 was exposed to the dual-wavelength irradiation (460 nm light at 180 J/cm 2 and 405 nm at 54 or 99 108 J/cm 2 ), compared with 405 nm light exposure (54,108 J/cm 2 ) alone ( Figure 3A;P = 0.24), 100 suggesting that the role of 460 nm light is solely to photolyze STX and subsequently render 101 MRSA more susceptible to ROS generated by 405 nm light. With respect to the WT strain, pre-102 exposure to 460 nm light (180 J/cm 2 ) significantly enhanced 405 nm aBL killing, further 103 confirming the role of 460 nm light in sensitizing MRSA to 405 nm aBL (P = 0.001; Figure 3B). 104

Enhancement of 405 nm light killing by 460 nm light was dose-dependent 105
In this study, we determined the effect of increasing radiant exposures of 460 nm light on the 106 effectiveness of 405 nm light. The representative clinical strain of MRSA (AF0003) was used. 107 MRSA was exposed to different radiant exposures of 460 nm light (90 J/cm 2 , 180 J/cm 2 , or 360 108 J/cm 2 ; reflecting a 15, 30 and 60 mins pre-exposure duration, respectively), prior to exposing 109 bacteria to 54 J/cm 2 or 108 J/cm 2 405 nm light. Pre-exposure to 360 J/cm 2 460 nm light resulted 110 in the most significant killing of MRSA following 54 J/cm 2 405 nm light, with a killing of 2.14 log 10 111 Figure 4). Conversely, exposing MRSA to 54 J/cm 2 405 nm light following exposure to 90 J/cm2 113 or 180 J/cm2 did not result in any significant improvement (P = 0.9). This suggests that the 114 enhancement of the antimicrobial efficacy of 405 nm light by 460 nm light is dose-dependent, 115 which is not surprising as increase photolysis of STX by 360 J/cm 2 may have increased sensitivity of MRSA to lower 405 nm doses.. When exposure to 405 nm aBL was increased to 117 108 J/cm 2 , pre-exposure to 180 J/cm 2 significantly enhanced killing of MRSA relative to 405 nm 118 light alone (P = 0.02). At a radiant exposure of 90 J/cm 2 460 nm light, however, there was no 119 significant improvement in the killing when 108 J/cm 2 405 nm light was exposed (P= 0.40). 120 Findings suggest that enhancement of 405 nm light mediated killing by pre-exposure to 460 nm 121 light, is contingent on the delivered radiant exposure of 460 nm light; as increasing radiant 122 exposures of 460 nm light resulted in increased susceptibility of MRSA to 405 nm light. We 123 additionally found there to be some increase in intracellular ROS (1.56-fold) when 405 nm light 124 was administered following photolysis by 460 nm light (360 J/cm 2 ), when compared to 405 nm 125 alone; however, this was not found to be statistically significant (P=0.24; Figure S2) 126 127 Dual-wavelength 460 nm + 405 nm light exposure effectively for reduced the viability of 128 MRSA biofilms. 129 In this study, we investigated whether pre-exposure to 460 nm light improved the efficacy of 405 130 nm aBL against 48-hour old MRSA biofilms. The MRSA AF0003 strain was used as the 131 representative strain for all biofilm experiments. The biofilms were initially exposed to 180 J/cm 2 132 460 nm light, immediately prior to exposing to 405 nm light, at 108, 216 or 324 J/cm 2 . We found 133 that irradiation of 108 J/cm 2 aBL at 405 nm did not result in any antimicrobial effects in the dual-134 wavelength irradiation exposed group, or in the 405 nm alone treated group ( Figure 5). When 135 the exposure of 405 nm light was increased to 216 J/cm 2 , however, the dual-wavelength 136 irradiation treated group showed a CFU reduction of 1.85 log 10 , compared with 0.73 log 10 CFU 137 reduction in 405 nm light alone treated group (P<0.05). When the exposure of 405 nm light 138 reached 324 J/cm 2 , the dual-wavelength irradiation resulted in 2.72 log 10 CFU reduction in 139 MRSA biofilms, compared with 405 nm light alone which inactivated 1.49 log 10 CFU (P =0.002; 140  nm light/t 360 J/cm 2 ) was administered to naive mouse skin tissue prior to assessing cellular 160 apoptosis using the TUNEL assay. Results showed no presence of apoptotic cells in the treated 161 group immediately following the dual-wavelength irradiation treatment,24 hours post-treatment, 162 or 48 hours post-treatment ( Figure 7A, B, C). The untreated group did not show any evidence of

DISCUSSION 165
In this study, we investigated the role of STX in promoting resistance to 405 nm light killing in 166 MRSA. In addition, we explored the STX photolytic effect of 460 nm light exposure as a means 167 of sensitizing MRSA to 405 nm light killing. With the use the STX deficient. Pig1:: ΔcrtM we 168 confirmed, for the first time, that the presence of STX within MRSA, is directly responsible for 169 the relative resistance phenotype to 405 nm light that has been observed in our and other 170 laboratories 6 . There have, however, been studies that have illustrated sensitivity of 405 nm light 171 killing, which conflict somewhat to ours 10 . It is possible that variabilities in the relative 172 abundance of endogenous STX may explain these findings. Furthermore, as STX is present in 173 90% of strains 11 it is feasible that improved efficacy of 405 nm light may be achieved using 174 these STX deficient strains, although further work is required to corroborate this hypothesis. 175 Given that STX has a carotenoid structure, and is thus a singlet oxygen scavenger, it is 176 unsurprising that its presence would interfere with damage elicited by singlet oxygen 12 . Liu et 177 al., fully characterized the role of STX and found it to be a powerful singlet oxygen scavenger. 178 They found the STX deficient isogenic mutant ΔcrtM, was >100-fold more susceptible to singlet 179 oxygen generated through methylene blue mediated photodynamic therapy (MB-PDT), relative 180 to its parental WT 7 . These findings, therefore, strongly support our findings illustrating the role of 181 STX in promoting resistance to 405 nm mediated killing. In a previous study, it was 182 demonstrated that photolysis using 460 nm light sensitized MRSA to killing by hydrogen 183 peroxide ,9 , further supporting the role of STX in eliciting resistance to ROS resulting from 405 184 nm light illumination. 185 We next sought to explore whether pre-exposure to 460 nm light would be adequate to 186 overcome the relative resistance observed when MRSA is exposed to 405 nm light. It was 187 demonstrated previously, that 460 nm light illumination was capable of STX lysis in MRSA. We found that the degree of STX photolysis achieved by 460 nm light illumination significantly 189 potentiated the killing of MRSA when subsequently exposed to 405 nm, suggesting that the 190 extent of STX lysis was enough to overcome any innate resistance to 405 nm light killing. 191 Exposure to 460 nm light alone, however, was insufficient to elicit any antimicrobial effects. Our We additionally found no evidence of apoptosis (immediately following treatment,24 hours later, 274 or 48 hours later) resulting from combining 460 nm +405 nm light against naïve mouse skin, at a 275 dose required to inactivate approximately 99% of MRSA within a mouse skin abrasion wound; 276 suggesting the therapy may be safely administered. It is important to appreciate, however, that 277 the doses required to significantly inactivate MRSA, within more established wound infections, 278 may necessitate higher therapeutic doses. In addition, the results present only present a 279 qualitative assessment safety denoted by the presence (or absence) of apoptotic cells, thus 280 suggesting that further studies are warranted to quantitatively evaluate the safety of using 460 281 nm +405 nm light against skin.. In conclusion, 460 nm + 405 nm combination therapy may offer 282 an effective and safe approach for the treatment of MRSA wound infections.

Blue Light Sources 285
Irradiations of 460 nm and 405 nm light were delivered using two light emitting diodes (LED; 286 M405L2 and M470L1; Thorlabs, Newton, New Jersey) with peak emissions of 460 nm and 405 287 nm, respectively, and a full width at half-maximum (FWHM) of 25 nm. The irradiance was 288 regulated by altering the distance of the light source aperture and the target with the use of a 289 PM100D power/energy meter (Thorlabs, Newton, New Jersey). 290

Bacterial Strains and Growth Conditions 291
The bacterial strains used in this study include a clinicals strains of methicillin resistant 292 Staphylococcus aureus (MRSA) AF0003 strain, that was isolated from an infected soldier 293 deployed in Afghanistan, and IQ00064 that was isolated from an infected soldier deployed in 294 Iraq. In addition, a bioluminescent MRSA USA 300 strain 23 , an MRSA Pig1 strain 7 . In addition, 295 the isogenic ΔcrtM mutant from the S. aueus Pig1 strain 7 was used for mechanistic studies. 296 With the exception of the S.aureus Pig1:: ΔcrtM strain, all are STX producers ( Figure S1). The 297 bacteria were cultured in brain heart infusion (BHI) medium (agar or broth) at 37°C, or in an 298 orbital incubator (37°C; 180 rpm), respectively. 299 300 Extraction of STX from S. aureus strains. All strains of S. aureus used in the study were initially 301 grown overnight on BHI agar from 25% glycerol freezer stocks. Subsequently, bacterial colonies 302 were collected with the use of a 10 μL loop and suspended in 10 mL Phosphate buffered saline 303 (PBS). The CFU was then adjusted to approximately 10 9 CFU/mL within PBS prior to extraction 304 of STX. The suspended bacteria were then centrifuged at 4000 x g for 4 minutes to pellet the 305 cells. The supernatant was then discarded and the pellet was suspended in 2 mL of 100% 306 methanol and incubated at 55 o C for 3 hours to ensure complete extraction of the pigment.
Following the incubation, the cells were pelleted and the supernatant containing STX was 308 transferred to a 1 mL cuvette. The absorption was then measured using am Evolution 300 UV-309 Vis Spectrophotometer (ThermoFisher Scientific, USA) in accordance with the manufacuter's 310 instructions. producing the skin abrasion wounds in mice, mice were injected intraperitoneally with the use of 355 a ketamine/xylazine cocktail (20 mg/kg -100 mg/kg). The mice were then shaved, and the tissue 356 was carefully abraded within a defined 1.0-cm x 1.0-cm area using a #15 sterile scalpel blade. 357 The scraped area either did not produce any blood or bled very little. Immediately following the 358 abrasion, a 100 µL bacterial suspension containing approximately 10 7 CFU of MRSA USA300 in 359 PBS was inoculated onto the wound and applied gently and uniformly using the side of a pipette 360 tip and left to incubate for 3 hours prior to treatment. To ensure that the inoculum was left intact 361 within the mouse abrasion wound, mice were kept anaesthetized until the inoculum dried within 362 the wound. In addition, to limit physical removal of bacteria from the wound, bacteria were 363 isolated within cages, with no more than 2 mice/cage. Experiments were performed in 364 septuplicate (7 independent replicates spanning 7 different days). Instrument (MP Biomedicals, USA). Samples were homogenized for 5 minutes, with 60 s 386 intervals, and samples were placed on ice for 5 minutes in between homogenization cycles, to 387 limit heat generation. Following homogenization, the CFU was determined by serial dilution (10 -388 2 -10 -7 dilution factors) on BHI agar plates as described previously 35. 389

TUNEL assay to detect apoptotic cells in mouse skin treated with 460 nm + 405 nm light 390
The presence of apoptotic cells that resulted from the dual-wavelength irradiation therapy was 391 determined in healthy mouse skin as described previously 28 . In brief, skin from the mouse was 392 isolated at 0 and 24 hours, following the treatment with 460 nm light (342 J/cm 2 ) and 405 nm 393 (360 J/cm 2 ). An untreated skin sample was also included as the control and was immediately 394