Light tolerance of extended spectrum β-lactamase producing Escherichia coli strains after repetitive exposure to far-UVC and blue LED light

Abstract

Aims

The aim of this study was to investigate dual far-UVC (Ultraviolet-C) (222 nm) and blue LED (Light Emitting Diode) (405 nm) light on the inactivation of extended spectrum β-lactamase-producing Escherichia coli (ESBL-Ec) and to determine if repetitive exposure to long pulses of light resulted in changes to light tolerance, and antibiotic susceptibility.

Methods and results

Antimicrobial efficiency of dual and individual light wavelengths and development of light tolerance in E. coli was evaluated through a spread plate method after exposure to light at 25 cm. Dual light exposure for 30 min resulted in a 5–6 log₁₀ CFU mL⁻¹ reduction in two ESBL-Ec and two antibiotic-sensitive control E. coli strains. The overall inhibition achieved by dual light treatment was always greater than the combined reductions (log₁₀ CFU) observed from exposure to individual light wavelengths (combined 222–405 nm), indicating a synergistic relationship between blue LED and far-UVC light when used together. Repetitive long pulses of dual and individual far-UVC light exposure resulted in light tolerance in two ESBL-Ec strains but not the antibiotic-sensitive E. coli strains. Subsequent passages of repetitive light-treated ESBL-Ec strains continued to exhibit light tolerance. Antibiotic susceptibility was determined through a standard disk diffusion method. No changes were observed in the antibiotic susceptibility profiles for any of the four strains after exposure to either dual or individual wavelengths.

Conclusions

Dual light exposure was effective in the disinfection of ESBL-Ec in solution; however, antibiotic-resistant E. coli were able to develop light tolerance after repetitive exposure to light.

405 nm blue light, 222 nm far-UVC light, light tolerance, antibiotic resistance, light disinfection, extended spectrum β-lactamase producing E. coli

Significance and impact of the study

This study demonstrates the novel antimicrobial efficiency of dual far-UVC and blue LED light against ESBL-Ec and that the correct application of these light technologies is important to minimize the potential for development of light tolerance.

Introduction

In 2019, the World Health Organization declared antimicrobial resistance (AMR) to be one of the top ten health concerns globally, with the latest Global Antimicrobial Resistance and Use Surveillance System (GLASS) report confirming >3.1 million laboratory-confirmed infections related to AMR pathogens (WHO 2021b). Since 2019, the number of countries reporting AMR-related infections and mortalities has increased significantly, with Escherichia coli, responsible for blood stream infections (BSIs) and urinary tract infections (UTIs), the most commonly reported pathogen ( WHO 2021b, Ikuta 2022). Extended spectrum β-lactamase producing E. coli (ESBL-Ec) are commonly associated with UTIs (Hapuarachchi IU 2021), can be associated with severe clinical outcomes (Foxman 2002), and are listed as Priority 1: ‘Critical pathogens’ on the World Health Organization priority pathogens list (WHO 2021a). The overuse and misuse of antibiotics to treat these and other infections has led to increased levels of antimicrobial resistance and an extra burden across healthcare systems (Cosgrove 2006). This highlights the need to identify effective non-chemical antimicrobial technologies for disinfection of bacteria and environmental decontamination that does not result in the increased prevalence of AMR bacteria and the emergence of new AMR variants.

The use of ultra-violet (UV) light (200–400 nm) for the inactivation of microorganisms is a technology that has been used for several decades, most commonly in safety biological cabinets in laboratories in the form of mercury lamps (254 nm) (Rea 2000). In recent years, its use has expanded into different industries, including food production, health care, wastewater treatment, and hospitality, for surface, water, air, and PPE disinfection (Hadi et al. 2020). UV light is grouped into three wavelength ranges, where each group has distinct biological effects: UVA (Ultraviolet A) (320–400 nm), UVB (Ultraviolet B) (290–320 nm), and UVC (200–290 mm) (Soehnge et al. 1997). UV radiation causes DNA damage through the formation of thymine dimers and photoproducts, which lead to cytotoxic effects and cell death (Sinha et al. 2002, Friedberg et al. 2006), with the most efficient antimicrobial activity being exhibited by UVC radiation. There are many mechanisms, however, that allow bacteria to adapt to UV exposure and become tolerant to light; for example, DNA repair pathways assisted by DNA polymerase, Pol V, and production of proteins such as nucleotide excision repair (NER) proteins that counteract the effects of DNA damaging agents such as UV light (Krishna et al. 2007) and the selection of mutations that are advantageous to survival (Alcántara-Díaz et al. 2004). Further studies have investigated the effect of UV radiation on antibiotic susceptibility, where repetitive exposure to UV light resulted in light tolerance and an increase in antibiotic resistance in various bacteria, including E. coli and Salmonella species (Álvarez-Molina et al. 2020).

Due to the health risks of UVC radiation, alternative technologies such as antimicrobial blue LEDs (400–450 nm) and far-UVC (222 nm) light have been implemented in recent years. This is due to the non-detrimental effects on human tissue while exhibiting antimicrobial activity against gram-positive and gram-negative bacteria, microbial biofilms, viruses, yeasts, and endospores (Maclean et al. 2009, 2013, Kleinpenning et al. 2010, Buonanno et al. 2020, Narita et al. 2020, Rathnasinghe et al. 2021, Eadie et al. 2022). The antimicrobial mechanism of blue light causes cellular damage and cell death through the generation of reactive oxygen species (ROS), such as hydrogen peroxide (H₂O₂) and reactive singlet oxygen (¹O₂), as light is absorbed by endogenous porphyrin molecules acting as photosensitizers (Amin et al. 2016, Angarano et al. 2020). Far-UVC krypton chloride lamps are believed to exhibit a similar mechanism as germicidal mercury UV lamps (254 nm); however, far-UVC light has limited penetrability, reducing the harmful effects on human tissues that result from exposure to traditional UV (Buonanno et al. 2017). Blue LED and far-UVC wavelengths have other notable advantages to traditional UVC in that they can be operated in a wider range of temperatures and humidity, and are safer to use and handle (Buonanno et al. 2016).

The use of dual blue LEDs and far-UVC light in combination is novel, and its effect on bacterial disinfection has not yet been investigated. Our previous work showed that dual blue LED and far-UVC wavelengths were virucidal against feline infectious peritonitis virus (FIPV) in solution and on different surfaces, indicating the potential of dual light wavelengths to be effective against bacteria (Gardner et al. 2021). In this study, we aimed to investigate the effect of dual blue LED (405 nm) and far-UVC (222 nm) light on the inactivation of clinically relevant ESBL-Ec and to determine whether repetitive exposure to long pulse doses of dual or individual light wavelengths results in changes to light tolerance, and antibiotic susceptibility.

Materials and methods

Bacterial strains

Two extended spectrum β-lactamase producing E. coli (ESBL-Ec) strains (AGR5151 and AGR6128) isolated from freshwater (Dannevirke, New Zealand) and two antibiotic-sensitive E. coli strains from environmental sources (AGR4060, from avian faeces; AGR4318, from mammal faeces) isolated in previous studies were used in this work (Table 1) (Burgess et al. 2022, Cookson et al. 2022). Single colonies were incubated aerobically in Luria Bertani (LB) broth (Fort Richard, New Zealand) at 37°C for 18 h to obtain overnight bacterial culture (∼9.5 log₁₀ CFU mL⁻¹).

Table 1.

Summary of the extended spectrum β-lactamase producing E. coli and antibiotic-sensitive E. coli control strains.

Isolate	Phylogroup	Sequence type	Serotype	Source	AMR phenotype*
AGR4060	D	ST-69	O17/O77:H18 O17/O44:H18	Avian faeces	Sensitive
AGR4318	B2	ST-131	O25:H4	Feline faeces	Sensitive
AGR5151	B2	ST-131	O16:H5	Fresh water	ESBL, StrR, TetR, and CipR
AGR6128	B2	ST-131	O16:H5	Fresh water	ESBL, StrR, TetR, and CipR

*ESBL: extended-spectrum β-lactamase; Str: streptomycin; Tet: tetracycline; Cip: ciprofloxacin.

Light equipment

A custom-made light emitting unit was used in this study, Violeta 2.0 (EnergyLine Ltd., Christchurch, New Zealand), which contained three blue LED 405 nm bars (total radiant flux = 36 W) and four 12 W USHIO far-UVC 222 nm mercury-free excimer lamps (total radiant flux = 21.5 mW cm⁻²). The arrays were equipped with heat sinks and fans to minimize heat transfer to the bacterial samples. The light source was placed 25 cm above the surface of the polystyrene 24-well culture plates (Nunc, ThermoFisher Scientific, Waltham, MA, USA), giving an average irradiance of 16,820 μW cm⁻² for 405 nm bulbs and 323 μW cm⁻² for 222 nm lamps. Spatial variation across the light unit was previously determined, and variation was minimized as much as possible across the 24 wells by placing the plate in the middle of the light unit. The average irradiance was then calculated based on this position and spatial data available (Gardner et al. 2021). All irradiance measurements were performed with a SpectriLight ILT950 spectroradiometer (International Light Technologies, Peabody, MA, USA).

Light exposure to far-UVC and blue LED light

Effects of light exposure on bacterial viability in suspension

A test suspension was obtained by centrifuging overnight bacterial culture in LB broth at 5000 g (Heraeus Multifuge X3R, ThermoFisher Scientific, Waltham, MA, USA) for 10 min, followed by the removal of supernatant and resuspension of the cell pellet in 10 mL of 0.01 M phosphate buffer solution (PBS) (pH 7.4). A 500 μL aliquot of test suspension (∼8.8 log₁₀ CFU mL⁻¹) was pipetted into each of the six wells (A, B, and C in columns one and five) of the 24-well plate. Half of the plate (column one) was exposed to light (treated wells), and the second half (column five) was covered with a strip of black paper to prevent exposure to the light (untreated controls). Replicate plates were prepared for each time point, and each bacterial strain analysed. The plates were placed inside the light box and exposed to dual or individual lights for 30 min, equating to a dose of 30.85 J cm⁻² for dual, 0.58 J cm⁻² for far-UVC only, and 30.28 J cm⁻² for blue LED only. After exposure to light, the supernatant in each well was removed and pipetted into separate Eppendorf tubes (1.5 mL). The tubes were centrifuged to pellet cells and then resuspended in 1 mL of PBS after the removal of supernatant. The bacterial suspension was serially diluted 10-fold in 0.1% peptone diluent (Fort Richard, New Zealand) and plated on to Columbia Sheep Blood Agar (SBA) (Fort Richard, New Zealand). The plates were incubated at 37°C for 24 h to determine the viable cell count (CFU mL⁻¹).

Synergistic study of far-UVC and blue LED light inactivation

To determine whether the relationship between far-UVC and blue LED light was synergistic or additive, the bacterial strains were exposed to dual (405 nm + 222 nm) light treatment, as well as individual 405 nm and 222 nm light wavelengths for 30 min. Inhibition (log reduction) was calculated by subtracting the log₁₀ CFU mL⁻¹ of the treated samples from the log₁₀ CFU mL⁻¹ of the corresponding untreated controls. The combined total of both bacterial log₁₀ CFU reductions per mL for individual light treatments was compared to the bacterial log₁₀ CFU reductions per mL achieved with dual light exposure using a one-way analysis of variance (ANOVA) test.

Tolerance study of far-UVC and blue light inactivation

To determine the possible development of tolerance to dual far-UVC and blue LED light in ESBL-Ec and antibiotic-sensitive E. coli, eight 5-min long pulses of light dose (5.14 J cm⁻²) on bacteria in suspension was carried out, interspersed with 5-min periods in darkness. After repeated exposure to dual lights, samples were serially diluted 10-fold and plated onto SBA, followed by incubation overnight at 37°C for 24 h. The following day, separate individual colonies were picked and re-inoculated into LB broth to produce overnight cultures that were subjected to further light treatment as outlined in 2.3.1. The procedure was repeated independently three times for each of the four strains, and resultant bacterial counts and log₁₀ CFU reductions were compared against controls that had not been exposed to repeated long pulses using a one-way analysis of variance test. The experiment was repeated with far-UVC light only with a dose of 0.15 J cm⁻² to determine if any light tolerance observed was due to dual exposure to far-UVC and blue light or only far-UVC light.

Cell passage number and light tolerance

To determine if light tolerance was maintained after subsequent cell passages, E. coli AGR5151 colonies were picked from SBA plates after eight repeated long pulses of dual light. Colonies were passaged a total of five times onto SBA plates and incubated for 24 h at 37°C after each passage. Colonies from passage number four (P4) and five (P5) were picked and inoculated into LB broth to produce overnight cultures that were then subjected to dual light treatment as outlined in 2.3.1.

Antibiotic susceptibility testing

After treatment with either far-UVC only or dual far-UVC and blue LED light (Section 2.3.3), the E. coli cultures in PBS were stored overnight at 4°C for antibiotic susceptibility testing according to EUCAST guidelines, using a Kirby-Bauer disk diffusion assay (Toombs-Ruane et al. 2020). The effect of repetitive long doses of light exposure was evaluated by comparison of the treated and untreated E. coli cultures against antibiotics, namely, cefotaxime (30 μg), cefpodoxime (10 μg), cefoxitin (30 μg), tetracycline (30 μg), streptomycin (10 μg), and ciprofloxacin (10 μg). The antibiotic susceptibility profiles of these E. coli strains had been elucidated in previous work (Burgess et al. 2022, Cookson et al. 2022).

Data and statistical analysis

All light treatments were repeated independently three times in triplicate for each of the four E. coli strains, and bacterial survival rates were compared against controls that were not exposed to light using a one-way analysis of variance test. Light-tolerant strains were compared to control strains by an unpaired t-test. For all light experiments, averages and standard deviation for graphical representation were performed using GraphPad Prism version 9.1.2 for Windows, GraphPad Software, San Diego, CA, USA, www.graphpad.com. Antibiotic susceptibility graphs were plotted on an Excel (Microsoft) spreadsheet.

Results

Synergistic effect of far-UVC and blue LED light on the inactivation of E. coli

Exposure of the two ESBL-Ec strains (AGR5151 and AGR6128) and two antibiotic-sensitive control E. coli strains (AGR4060 and AGR4318) to dual blue LED (405 nm) and far-UVC (222 nm) lights resulted in a 5–6 log₁₀ CFU mL⁻¹ reduction within 30 min (30.86 J cm⁻²) (Fig. 1). In comparison, when exposed to blue LED light only, 0.18 log₁₀ CFU mL⁻¹ reduction was achieved for E. coli strain AGR4318, with no inhibition observed for AGR5151, AGR6128, or AGR4060. Compared to blue LED light, there was an increased level of inhibition observed for individual light exposure to far-UVC, with a 4–5 log₁₀ CFU mL⁻¹ reduction at 30 min (0.58 J cm⁻²). The inhibition achieved by dual lights was always greater than the sum of the log₁₀ CFU reductions from exposure to individual wavelengths (combined 405 nm + 222 nm) (Fig. 1), signifying a synergistic relationship between far-UVC and blue LED light when used dually.

Figure 1.

Log reduction of two extended spectrum β-lactamase-producing E. coli strains (AGR5151 and AGR6128) and two antibiotic-sensitive control E. coli strains (AGR4060 and AGR4318) in 0.01 M phosphate buffered solution (PBS) for far-UVC (222 nm) (0.58 J cm⁻²), blue LED (405 nm) (30.28 J cm⁻²), dual (405/222 nm) (30.85 J cm⁻²), and combined sum (405 nm + 222 nm) lights. All exposures were performed at 25-cm distance. Data is displayed as the reduction in mean log₁₀ CFU mL⁻¹ ± standard deviation. Statistically significant differences are indicated by the *(P < 0.05).

Effect of repeated sublethal doses of blue LED and far-UVC on bacterial inactivation

Exposure to eight repetitive long pulses (5.14 J cm⁻²) of dual light (405 nm + 222 nm) prior to 30 min of continuous dual light resulted in less inhibition of the ESBL-Ec strains (AGR5151 and AGR6128) by 2–3 logs as compared to the same strains (T30) that were not exposed to repetitive long pulses prior to a continuous dose of dual light for 30 min (Fig. 2A and B). Antibiotic-sensitive E. coli strains (AGR4060 and AGR4318) did not show any significant difference between the controls (T30) and long pulse-treated (T8 × 5) strains when exposed to a continuous dose of dual light for 30 min (Fig. 2C and D).

Figure 2.

The effect of repeated long pulse doses (5.14 J cm⁻²) of dual blue LED 405 nm and far-UVC 222 nm light on two extended spectrum β-lactamase-producing E. coli strains AGR5151 and AGR6128 (Fig. 2A and B) and two antibiotic-sensitive control E. coli strains AGR4060 and AGR4318 (Fig. 2C and D) in 0.01 M phosphate buffered solution (PBS). All strains were treated with 30 min (30.85 J cm⁻²) of dual light exposure with or without prior exposure to repeated long pulses of dual light (8 × 5 min). Data is displayed as average log₁₀ CFU mL⁻¹ ± SD. All exposures were performed at 25-cm distance. Statistically significant differences are indicated by the ^****(P < 0.0001). ns denotes not significant.

Effect of repeated sublethal doses of far-UVC on bacterial inactivation

Exposure to eight repetitive long pulses (0.15 J cm⁻²) of far-UVC (222 nm) light only prior to a 30-min continuous far-UVC dose resulted in a significant (P < 0.0001 and P < 0.01) difference in bacterial inhibition (log₁₀ CFU mL⁻¹) between the control bacteria (T30) and the long pulse treated (T8 × 5) ESBL-Ec (AGR5151 and AGR6128) (∼1–1.5 log₁₀ CFU difference) but not for antibiotic-sensitive control E. coli strains (AGR4060 and AGR4318) (Fig. 3).

Figure 3.

The effect of repeated long pulse doses (0.15 J cm⁻²) of far-UVC 222 nm light on extended spectrum β-lactamase-producing E. coli AGR5151 and AGR6128 (Fig. 3A and B) and two antibiotic-sensitive control E. coli strains AGR4060 and AGR4318 (Fig. 3C and D) in 0.01 M phosphate buffered solution (PBS). All strains were treated with 30 min (0.58 J cm⁻²) of far-UVC light exposure with or without prior exposure to repeated long pulses of far-UVC light (T8 × 5 min). All exposures were performed at 25-cm distance. Data is displayed as average log₁₀ CFU mL⁻¹ ± SD. Statistically significant differences are indicated by the **(P < 0.01) and ^**** (P < 0.0001) after unpaired t-test. ns denotes not significant.

Effect of cell passage number on light tolerance

Multi-drug-resistant ESBL-Ec AGR5151, that was previously exposed to repetitive long pulses (T8 × 5 min) of dual far-UVC and blue LED light, consecutively passaged four (P4) and five (P5) times, continued to exhibit light tolerance to continuous dual light at a dose of 30.85 J cm⁻² as compared to controls only treated with 30 min of continuous light (Fig. 4).

Figure 4.

Inherited light tolerance exhibited by multidrug-resistant extended spectrum β-lactamase-producing E. coli AGR5151 after repetitive dual long pulse (5.14 J cm⁻²) exposure (T8 × 5 min) followed by 30 min (30.85 J cm⁻²) of continuous dual blue LED and far-UVC light. Colonies were passaged one (P1), four (P4), or five (P5) times before 30 min of continuous dual light treatment. All exposures were performed at 25-cm distance. Data is displayed as average log₁₀ CFU mL⁻¹ ± SD. Statistically significant differences are indicated by the ^****(P < 0.0001).

Effect of light treatments on the antibiotic susceptibility of E. coli strains

The antibiotic susceptibility for both ESBL-Ec and antibiotic-sensitive E. coli strains that had been previously exposed to repetitive long pulse doses of dual light showed no change in their antibiotic profiles. More specifically, the resistant strains remained resistant (Fig. 5), and the sensitive strains remained sensitive to antibiotic tested after treatment (Fig. 6).

Figure 5.

Antibiotic susceptibility of extended spectrum β-lactamase-producing E. coli AGR5151 (Fig. 5A) and AGR6128 (Fig. 5B) after repeated long pulsed (8 × 5 min) (5.14 J cm⁻²) dual far-UVC 222 nm and blue LED 405 nm light treatment in comparison to control strains (C8 × 5) not exposed to light. Each bar in the graph represents the average diameter of the inhibitory zone (in mm, including the 6 mm diameter of the disc) for different antibiotics tested, where grey bars represent the treated samples and black bars represent the controls (untreated). R = resistant, S = sensitive. The results were compiled after three independent assays performed in duplicates; error bars indicate + SD.

Figure 6.

Antibiotic susceptibility of antibiotic-sensitive E. coli strains AGR4060 (Fig. 6A) and AGR4138 (Fig. 6B) after repeated long pulsed (T8 × 5 min) (5.14 J cm⁻²) dual far-UVC 222 nm and blue LED 405 nm light treatment in comparison to control strains (C8 × 5) not exposed to light. Each bar in the graph represents the average diameter of the inhibitory zone (in mm, including the 6 mm diameter of the disc) for different antibiotics tested, where grey bars represent the treated samples and black bars represent the controls (untreated). S = sensitive. The results were compiled after three independent assays performed in duplicates; error bars indicate + SD.

Similarly, no change in antibiotic profile was observed in all strains tested after a continuous dose of dual far-UVC and blue LED exposure for 30 min following the repeated long pulse light treatment (Supplementary Figure 1). The applied light treatments showed insignificant influence on antibiotic susceptibility of all four E. coli strains tested.

Discussion

The growing global incidence of antimicrobial resistance has highlighted the need for safe, non-chemical technologies that can be used for the disinfection of bacteria efficiently without producing further antimicrobial resistance (WHO 2021b). In this study, we investigated the potential of dual blue LED (405 nm) and far-UVC (222 nm) light in the inactivation of clinically relevant extended spectrum β-lactamase-producing E. coli (ESBL-Ec) and determine whether repetitive long pulses of individual and dual light exposure resulted in light tolerance or changes in antibiotic susceptibility. Clinically relevant ESBL-Ec were chosen as the target antibiotic-resistant bacteria in this study as they are listed as ‘Priority 1: Critical pathogens’ on the WHO priority pathogens list for research and development of new antibiotics (WHO 2021a).

We demonstrated that dual exposure to far-UVC and blue LED light had bactericidal effects on all E. coli strains tested in PBS solution after 30 min (30 J cm⁻²) of exposure, with a 5–6 log₁₀ CFU mL⁻¹ reduction in all strains tested. This is the first time that the application of far-UVC (222 nm) and blue LED light (405 nm) in combination has been demonstrated to effectively disinfect bacteria in solution. The use of individual far-UVC and blue LED lights in whole-room disinfection scenarios has shown to be effective in several studies (Amodeo et al. 2022, Eadie et al. 2022, Xie et al. 2022). In one study, blue LED (405 nm) light was applied at 2–3 m distances resulting in a 2–4 log₁₀ reduction in E. coli over a 12-h period with an irradiance of 20–40 J cm⁻² (Amodeo et al. 2022). Blue LED on its own was not shown to be effective in the inhibition of E. coli in PBS in this current work as compared to the study by Amodeo et al.(2022). This is possibly due to the differences in the environment of the bacteria when exposed to the light, e.g. dried on surfaces or in air, compared to a suspension in PBS in a 24-well plate. Furthermore, the length of exposure time and distance of the blue LED light treatment has a significant impact on the efficacy of inhibition. The American Conference of Governmental Industrial Hygienist (Slonczewski et al. 2020) recently published guidelines for exposure to far-UVC light to ensure reduced risk to human health with doses <161–479 mJ cm⁻² suggested as safe (Slonczewski et al. 2020). The use of dual light in whole room disinfection could reduce the time and therefore the dose required to complete disinfection safely for surfaces, air, or water as lower far-UVC doses would be expected to achieve the same outcome. However, the safety of dual far-UVC and blue LED light applications in public places would need to be investigated, considering the spaces are occupied during disinfection.

The log₁₀ CFU mL⁻¹ reduction of E. coli when exposed to dual far-UVC and blue LED light was always greater than the sum of the individual light wavelength log₁₀ CFU mL⁻¹ reductions, indicating a synergistic relationship between the two wavelengths of light studied. Blue LED light exposure on its own resulted in very little reduction in bacterial count; however, when combined with far-UVC light, it facilitated in a higher reduction. Combining UVC and UVA wavelengths of light that exhibit different mechanisms of inactivation have previously been shown to be synergistic in behaviour (Chevremont et al. 2012). The sequential exposure of UVA (365 nm) followed by UVC (265 nm) wavelengths showed an increase in bacterial log reduction as compared to individual wavelengths (Xiao et al. 2018). The reactive oxygen species (ROS) generated by UVA light facilitated the effect of UVC capability while inhibiting repair pathways. Blue LED 405 nm light also produces ROS (Wang et al. 2017), and it is likely this has facilitated the efficacy of far-UVC light in the same way. In another study, it was reported that a synergistic relationship between UVA and UVC was dependent on bacterial strain and that using small doses of UVA prior to UVC exposure facilitated E. coli recovery by aiding the repair of DNA damage (Song et al. 2019). Combining blue LED and far-UVC together with a higher irradiance may reduce the likelihood of DNA repair mechanisms from being activated.

Our results show that there was no difference between the inhibition of the ESBL-Ec and the antibiotic-sensitive E. coli control strains after 30 min of dual UVC and blue light; however, there was a difference in the inhibition of ESBL-Ec after exposure to repetitive long pulses of dual and individual far-UVC light as compared to the antibiotic-sensitive control strains. Repetitive exposure to UVC radiation has been previously shown to cause resistance in E. coli because of the generation of mutations in genes responsible for repair and replication of DNA (Alcántara-Díaz et al. 2004). Far-UVC is thought to elicit the same mechanism of DNA damage; therefore, it is likely that repetitive exposure to low doses of far-UVC would also result in light tolerance. Light tolerance from repetitive long pulses of blue LED light alone was not investigated in this study as only minimal bacterial cell reduction was observed within the 30-min experimental duration, and therefore comparisons between treated and untreated would have been difficult. Until now, light tolerance to blue LED light has not been noted, but the potential for light tolerance development through multiple cellular pathways has been postulated (Hadi et al. 2021).

One possible explanation for the difference in light tolerance between ESBL-Ec strains and antibiotic-sensitive E. coli control strains is their isolation from freshwater streams may have allowed exposure to higher UV radiation in the aquatic environment as compared to the antibiotic-sensitive strains isolated from faeces, and therefore may have developed potential DNA repair systems that were initiated upon exposure to low repetitive doses of dual blue LED and far-UVC light. There are several studies that have investigated the response of bacteria in aquatic environments to UV radiation, especially those in high-altitude environments such as alpine lakes that are exposed to high UV radiation. These studies have reported that bacteria living in these conditions have high tolerance to UV and use several strategies to minimize the damage caused by UV radiation, as compared to bacteria from less irradiated environments (Sommaruga 2001). Another possible explanation for the difference in light tolerance exhibited by the ESBL-Ec strains compared to the antibiotic-sensitive E. coli controls is that the antibiotic-resistant mechanisms the ESBL-EC strains have may provide cross-protection against sublethal doses of light. These mechanisms could provide protection in a similar manner to bacterial stress responses exhibited (Davies et al. 1998). A number of cross-protection mechanisms have been proposed, including general bacterial stress response with sigma factors and two component systems (TCS), SOS response pathway, efflux pumps, and ribosomal mutations (Liao et al. 2020). In one study, it was found that Listeria monocytogenes cells resistant to sulphanilamide was more resistant to UV radiation than cells that were not antibiotic-resistant to sulphanilamide (McKinney et al. 2009). Further work could be investigated to examine the effect of exposure to repetitive dual or far-UVC light wavelengths on clinical AMR isolates or antibiotic-sensitive isolates from water to prove which of these explanations may explain the differences in light tolerance.

Subsequent passaging of ESBL-Ec AGR5151 colonies that had been exposed to repetitive long pulses of dual blue LED and far-UVC light showed daughter cells continued to exhibit light tolerance at the same rate as the ancestor. This observation could be due to mutations in the bacterial DNA generated by light exposure, resulting in genetic changes that are inheritable (Shibai et al. 2017). Another possible explanation for the light tolerance in the daughter cells could be due to epigenetic changes such as DNA methylation and changes in gene expression, which have been shown to be inheritable and play a role in the development of resistance of bacteria to antibiotics and other environmental factors (Casadesus et al. 2006, Adam et al. 2008). Further analysis is required to elucidate whether the changes are genetic or epigenetic after repetitive exposure to long pulses of dual and individual light wavelengths.

We demonstrated there were no changes in antibiotic susceptibility after repetitive exposure to long pulses of dual blue LEDs and far-UVC light or far-UVC light alone for any of the E. coli strains tested. A further continuous 30 min of dual or individual light wavelengths also showed no changes in antibiotic susceptibility. This indicates that far-UVC and blue light treatment of either low or high doses does not change the antibiotic susceptibility of E. coli strains tested and therefore could be used as an efficient antimicrobial technology for the disinfection of antibiotic-resistance bacteria.

There are some limitations of the use of blue LED and far-UVC in real-world applications that were not investigated in this study. The first limitation is the inherent shadowing effect of light-based technology. It is therefore important to consider this when designing the light units and the placement of these to ensure effective disinfection. One possible option is the use of hand-held units that have been developed recently for far-UVC that allow finer control of the application of the light (Ma et al. 2023). The generation of ozone by UV light technologies is another limitation and one that should be considered when using in a real-world application. Far-UVC KrCl excimer lamps do produce ozone, but the levels are significantly lower than traditional UVC lamps; however, it would be a requirement to measure the ozone generation rates with a standardized method to ensure the concentrations are below the regulated levels when these light units are installed for use (Blatchley et al. 2021). Finally, the effect of light exposure on the PBS solution that the bacteria were treated in was not assessed further for the production of any reactive oxygen species (ROS) and should also be investigated in the future to determine any effects prior to use in real-world applications.

Conclusion

The use of dual far-UVC and blue LED light in the inactivation of antibiotic-resistant E. coli has not been reported before. This is the first study to report the potential for this technology to be used for the disinfection of bacteria while being safe to use and without the potential to create further antimicrobial resistance. However, low doses of repetitive far-UVC light have shown the potential to create light tolerance in antibiotic-resistant E. coli, and therefore its application should be considered carefully before being implemented.

Acknowledgement

Authors wish to acknowledge Rose Collis and Lynn Rogers for their technical inputs.

Conflict of interest

All authors declare that there is no conflict of interest in this work. The funders had no role in the design of the study, in the collection or analysis of data, in the writing of the manuscript, or decision to publish the findings.

Funding

Financial support was provided by the New Zealand Ministry of Business, Innovation, and Employment through the AgResearch Ltd. Strategic Science Investment Fund (SSIF).

Author contributions

Amanda Gardner (Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing), Aswathi Soni (Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing), Adrian Cookson (Methodology, Resources, Validation, Writing – review & editing), and Gale Brightwell (Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing)

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.