Recently, increasing attention has been paid to antibiotic resistance in stormwater runoff. However, there is no available literature about the control of antibiotic resistant bacteria (ARB) through 365 nm ultraviolet light-emitting diode (UVA/LED). In this study, batch experiments were conducted to investigate ARB inactivation kinetics, effects of light intensity and water matrix (including suspended solid (SS) concentration, initial pH and bacteria concentration), and potential transmission risks after UVA/LED irradiation. Results showed that ARB inactivation efficiencies reached 6.31 log reduction at 8 mW/cm2 (86 J/cm2) of UVA/LED for 180 min. ARB inactivation efficiencies increased with the increase of light intensity, and showed a linear relationship. ARB inactivation decreased with increasing SS levels, and the largest inactivation efficiencies was 3.56 log reduction at 50 mg/L of SS. Initial pH had slight effect on ARB inactivation through UVA/LED irradiation. A low initial bacteria concentration (105 CFU/mL) was not necessarily associated with good ARB inactivation (3.59 log reduction). After UVA/LED irradiation, ARB was hardly detected during 12 hr of dark repair, and the transfer frequency of kanamycin resistance gene was increased to 5.43 × 10−4. These suggested that the application of UVA/LED to inactivate ARB in stormwater runoff was feasible and desirable in this study.

  • ARB inactivation reached 6.31 log reduction at 8 mW/cm2 of UVA/LED for 180 min.

  • UVA/LED inactivation kinetics could be well described by the first-order model.

  • Initial pH and bacteria levels had weak effects for ARB inactivation.

  • Potential risks of ARB could be blocked after 60 min of UVA/LED disinfection.

Graphical Abstract

Graphical Abstract
Graphical Abstract

In recent years, environmental problems of antibiotic resistance have attracted attention. Since the late 20th century, a large number of different kinds of antibiotics have been found and widely used in the prevention and control of disease (Davies & Davies 2010; Larsson & Flach 2022). The pollution of antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs) caused by long-term abuse of antibiotics had become increasingly prominent, which seriously endangered environments and human health, animals, and plants. Most antibiotics remaining in the environment were difficult to degrade, and microorganisms became adaptive, leading to antibiotic failure (Hendriksen et al. 2019). Meanwhile, the ARGs carried by ARB had the biological characteristics of transmission and diffusion, which presented a substantial threat to the control of infectious diseases (such as scarlet fever and tuberculosis) (Hatfull et al. 2022). Particularly, as an important medium for the migration and transformation of non-point source pollution, stormwater runoff was one of the important paths for antibiotic resistance spread in the environment (Zhang et al. 2016; Almakki et al. 2019), where both ARB and ARGs were found significantly (Lee et al. 2020; Zuo et al. 2022a). To mitigate this potential public health risks, it is highly desired to develop effective methods to inactivate ARB in stormwater runoff.

It is well known that the conventional disinfection technologies are mainly comprised of chlorination, ozonation, and UV irradiation. The chlorination process led to the generation of potentially carcinogenic by-products such as trihalomethanes (THM) (David et al. 2018; Moreno-Andrés & Peperzak 2019). For ozonation, cost and safety concerns limited its practical application (Wang et al. 2019). By contrast, UV irradiation had received extensive attention in the field of disinfection. From the perspective of current disinfection application, UVC disinfection was favored because of its excellent germicidal power (Cai et al. 2021). However, this damage was easily repaired, leading to microbial reactivation (Grob & Pollet 2016; Hess-Erga et al. 2019). In addition, since the main UVC light source was low-pressure or medium-pressure UV mercury lamp, the problem with its application was that it contained highly toxic mercury, which might cause fatal damage to the environment (Song et al. 2016; Xiao et al. 2018). Mercury lamps also suffered from long start-up times, high maintenance costs, high energy consumption, and short lifespans (Lui et al. 2016; Sholtes et al. 2016).

UV light-emitting diodes (UV/LEDs) have been considered as one of the alternative and environmentally friendly disinfection technologies (Kebbi et al. 2020; Wang et al. 2022). The high-power LEDs showed longer life and better resistance to mechanical shock than conventional UV lamps, and they were more energy efficient, compact in size, and space-saving (Jo & Tayade 2014; Song et al. 2019). UVA/LED had been proved to be effective for sterilization in water (Ferreira et al. 2016; Cai & Hu 2017). The inactivation efficiency of E. coli could reach 5.7 log reduction at 315 J/cm2 of UVA/LED irradiation (Hamamoto et al. 2007). Moreover, bacteria might be more inactivated in the presence of photocatalysts (such as TiO2) (Xiong & Hu 2013). After UVA irradiation, the absorbed energy could negatively affect the DNA by forming pyrimidine dimers, thereby inactivating bacteria (Senavirathne et al. 2014). More important was that there was the indirect inactivation mechanism through producing reactive oxygen species (ROS) after absorbing UVA by specific molecules (Mcguigan et al. 2012; Cadet et al. 2014), which was irreversible and provided more possibilities for inhibition of microbial reactivation. However, there is still a lack of information about the inactivation of ARB in stormwater runoff by UVA/LED, which could be a new challenge for stormwater reuse, although the removal of ARB in stormwater runoff through stormwater bioretention cell and biofilter (Rugh et al. 2022; Zuo et al. 2022b) has been reported.

UV inactivation kinetic model was closely related to irradiation time and dose, which could more intuitively evaluate the effect of experimental factors on UV sterilization, and provided more relevant theoretical basis, and the microbial inactivation often followed a first-order model (Zhou et al. 2017). Light intensity was one of the important factors for UV disinfection (Said & Otaki 2013; Kim et al. 2016). Due to the different sensitivity of various microorganisms to UV radiation, the irradiation doses with the best inactivation efficiencies were different. Furthermore, the water matrix was also important for UV disinfection (Wetzel & McBride 2020). For example, pH could indirectly affect the disinfection effect by affecting other substrates in solutions (Ng et al. 2016; Figueredo et al. 2021). Turbidity could affect light transmittance, which was manifested in SS concentration and initial bacteria concentration (Li et al. 2017; Farrell et al. 2018; Wang et al. 2020a). On the other hand, Song et al. (2019) reported that E. coli dark repair reached 12% in the first hour after UVA pretreatment. Similarly, Xiao et al. (2018) indicated that the repair of E. coli 15597 327 after two hours of dark repair decreased from 16% to 3% when the bacterial suspension was treated with UVA pre-radiation. The antibiotic resistance can be disseminated by sharing ARGs among microorganisms through horizontal gene transfer (including conjugation transfer and transformation). Conjugative transfer is mainly a process of genetic material transfer, and its vectors were mainly autonomously transferable plasmids and conjugative transposons (Lu et al. 2018). The transformation process was a process in which free extracellular ARGs were taken up by competent recipient bacteria and integrated and expressed in the recipient bacteria to acquire resistance (Wang et al. 2020b). Pokhum et al. (2017) claimed that UV disinfection could destroy the integrity of cell structure and release ARGs, suggesting the possibility of conjugation transfer and transformation after UVA/LED treatment. This indicated that it was necessary to explore the transmission risk after UVA/LED inactivation due to the limited damage potential of UV to ARGs in water (Martín-Sómer et al. 2017; Pei et al. 2019). However, the data on ARB repair and antibiotic resistance transmission after UVA/LED disinfection in stormwater runoff were limited until now.

Therefore, in this study, UVA/LED was used as radiation source for the batch experiments about ARB inactivation efficiencies, the effects of light intensity and water matrix, and potential transmission risks after UVA/LED irradiation treatment. It provides a promising reference to control antibiotic resistance diffusion in stormwater runoff through UVA/LED disinfection.

Bacterial strains and chemicals

E. coli K-12 with plasmid RP4 that carries blaTEM, tetR and aphA was selected as target ARB in this study, which was the same as the one used in our previous literature (Zuo et al. 2022b). It is a multi-antibiotic resistant bacterium with little interference in detection. The strains were incubated in Luria broth medium (LB, 25 g/L, pH of 7.4), shaken overnight (18–24 h) at 30 °C and 140 rpm, and then centrifuged at 5,000 × g for 5 min. The supernatants were removed after the centrifugation, and the pellet was resuspended in 10 mL sterilized normal saline (0.9% NaCl) solution (Zuo et al. 2022b).

Tetracyclines, kanamycin, ampicillin, and streptomycin were purchased from Aladdin, China. Analytically pure NaCl, NaOH, HCl, and kaolin were purchased from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). Ultrapure water (Aquapro Pure Water System) was used for all solutions and suspensions. The main chemicals of simulated stormwater were obtained from Nanjing Jiaodeng Science Equipment Co. Ltd (China), including starch (purity > 99%), ammonium chloride (NH4Cl, purity > 99%), and potassium dihydrogen phosphate (purity 99%). The stock solutions of chemicals were prepared by dissolving the corresponding compound into sterile DI-water (autoclaving for 15 min at 121 °C) to yield the desired concentrations for COD, ammonia nitrogen and total phosphorus (simulated stormwater quality), which was in line with the one in our previous literature (Zuo et al. 2022b).

The disinfection experiments

The UVA/LED system is standard, including light source, heat sink, DC power and magnetic stirrer. The beaker (500 mL) with 100 mL simulated stormwater was placed below the center of the light source (20 W 365 nm UVA/LED, L × W ∼ 178 × 139 mm, Zhonglian UV optical factory, Shenzhen, China). This UVA/LED device consists of 20 individual lamp arrangements. The control experiment took place in the dark and other conditions were consistent with the disinfection experiment. In the disinfection experiments, initial bacteria concentration was set to 107 CFU/mL, light intensities of 5, 6.5, 8, 9.5, 11, 12.5 mW/cm2 were used for the disinfection experiment by adjusting the distance from the UV lamp to the center of water surface. NaOH and HCl solutions were both added to adjust the pH of the simulated stormwater and the initial pH was 7. The total duration of irradiation was 60 min (29 J/cm2 when light intensity was 8 mW/cm2), and then the samples were diluted to different gradients using 0.9% NaCl solutions through the multiple dilution method, then 100 μL of diluted suspension was uniformly spread onto the prepared LB solid medium plates (carrying different antibiotics, 50 mg/L of kanamycin sulfate, 10 mg/L of tetracycline hydrochloride, 100 mg/L of ampicillin), and incubated at 37 °C for 24 h to obtain ARB counts before and after the experiments. The enumerative method used colony forming units, selecting the appropriate plate for the total number of colonies (30–300) to count by colony counter, and the actual number of colonies in sample was obtained according to different dilution concentrations.

Dark repair and transmission experiments

After the inactivation, the dark repair was conducted for a set time (2, 4, 6, 8, 12, and 24 h) in dark conditions. The system construction and experimental conditions of the conjugation transfer and transformation system were based on the studies of Lu et al. (2018) and Wang et al. (2020b), respectively. In this study, E. coli (K-12) carrying RP4 plasmid was used as the donor bacterium, and E. coli (HB101) carrying streptomycin resistance gene was used as the recipient bacterium to build the binding conjugation transfer system. Based on counting the colonies on the selection plates for both transconjugants and the recipient bacteria, transfer frequency was calculated by dividing the number of transconjugants by the number of recipient bacteria. In the transformation system, E. coli (pWH1266) carrying tetracycline and ampicillin resistance was used as donor and Acinetobacter A. baylyi ADP1 as the recipient. Based on counting the colonies on the transformant selection plate and the recipient bacteria selection plate, the transformation frequency was calculated by dividing the number of transformants by the number of recipient bacteria. The donor and recipient bacteria were incubated according to the method in the "Bacterial strains and chemicals" section, and mixed at a ratio of 1:1 in a simulated water sample treated with UVA/LED. Both were left to react for 6 h at room temperature. The water sample (1 mL) was extracted at pre-determined reaction times. After that, the collected sample was cultured and counted on the corresponding antibiotic screening plate to measure bacteria concentrations.

Analysis

Initial pH was detected by a pH meter (PSH-3C, YUEPING). The light intensity of UVA/LED was determined by an optical power meter (CEL-NP2000, CEAULIGHT). The number of colonies to count by a colony counter (OMJ-2, SHANGHAI OUMENG).

Plates between 30 colonies and 300 colonies were included in the analysis. Inactivation rate was reported in log reduction form, as follows:
formula
where N0 is ARB count before the disinfection, Nt is ARB count after the disinfection at time t.

All experiments were done in triplicate, and then the average of three replicate experiments was calculated and used for data interpretation. Data analysis graphs were obtained by ORIGIN 2017.

ARB inactivation efficiency

Results from Figure 1 showed that ARB inactivation efficiencies increased over time during UVA/LED irradiation. However, at the first 30 min, the inactivation efficiencies of ARB were not obvious with less than 1.0 log reduction, indicating that it was very insensitive to UVA for short irradiation times, which was consistent with the maximum inactivation of four E. coli strains being lower than 0.05 log reduction after 30 min of irradiation at 365 nm (Xiao et al. 2018). This could be attributed to cells having a self-protective effect when stimulated by external stimulus, and the microorganism being in the growth adaptation period (Liang et al. 2019). It would produce enough ROS to inactivate ARB with the persistent irradiation of UVA (Pezzoni et al. 2014). On the other hand, the maximum inactivation efficiencies (5.54 log reduction) could be reached after 120 min of UVA/LED exposure (Figure 1), which was higher than the one (4 log reduction) after 120 min of UVA irradiation observed by Robertson et al. (2005). Moreover, Pezzoni et al. (2014) reported that Pseudomonas aeruginosa in the form of planktonic bacteria and biofilm achieved more than 3 log reduction within 180 min at 20 W/m2 (216 kJ/m2) applying UVA. These indicated that UVA/LED could efficiently inactivate ARB in stormwater runoff.
Figure 1

ARB inactivation under UVA/LED irradiation or in the dark (Control), where light intensity was 8 mW/cm2, pH 7, and initial bacteria concentration 107 CFU/mL.

Figure 1

ARB inactivation under UVA/LED irradiation or in the dark (Control), where light intensity was 8 mW/cm2, pH 7, and initial bacteria concentration 107 CFU/mL.

Close modal

The role of light intensity

As presented in Figure 2, ARB inactivation efficiencies improved significantly with the increase of UVA/LED light intensity. According to the fitting results, the inactivation efficiencies of ARB had a good linear relationship with light intensity under different conditions, and the correlation coefficient R2 > 0.98. These clarified that the increase in inactivation efficiencies was likely due to the accelerated kinetics, which was customarily expected in photochemical reactions. The number of photons increased with the increase of light intensity, and thus the number of photo-generated electron holes was enhanced, resulting in improving the photocatalytic activity. Similarly, Pousty et al. (2021) found that at the same irradiation dose, the combination of high UV intensity and short time could cause high inactivation efficiencies, which might be due to the bacteria self-repair ability being destroyed at high light intensity (Wang et al. 2011). High levels of UVA damage intracellular antioxidant enzymes, causing bacteria to fail to repair themselves (Pezzoni et al. 2014). Meanwhile, ROS (such as hydroxyl radical (·OH)) levels in cells increased due to UVA irradiation, which caused oxidative damage to cell components (Pezzoni et al. 2014). For example, Chevremont et al. (2013) showed that the inactivation of E. coli ATCC 1,130.03 reached 2.7 log reduction after UVA/LED exposure for 20 s at 350 mW (7 J), which was much higher than the one (less than 0.5 log reduction when light intensity was 8 mW/cm2 for 15 min) in this study. This could be explained by the higher light intensity used in the mentioned literature.
Figure 2

Effect of light intensity on ARB inactivation by UVA/LED under irradiation time 60 min, pH 7, and initial bacteria concentration 107 CFU/mL.

Figure 2

Effect of light intensity on ARB inactivation by UVA/LED under irradiation time 60 min, pH 7, and initial bacteria concentration 107 CFU/mL.

Close modal

Effects of water matrix on ARB inactivation

SS concentrations

Under the condition without UVA/LED irradiation (only SS), ARB inactivation efficiencies hardly changed with the varying SS levels (Figure 3). This indicated that SS had little effect on ARB inactivation. However, the inactivation efficiencies of ARB decreased with increase of SS concentrations under UVA/LED irradiation. The lowest ARB inactivation efficiency was 2.52 ± 0.23 log reduction when the initial SS concentration was 150 mg/L in this study. SS in solutions would hinder the transmission of light, such as reflecting and absorbing ultraviolet rays (Li et al. 2017), and SS also provided a shielding effect for the bacteria, resulting in no ultraviolet radiation and thus making it difficult for them to be killed (Cantwell & Hofmann 2011; Yu et al. 2014). Interestingly, ARB inactivation efficiency was 3.56 ± 0.31 log reduction at 50 mg/L of SS level, which was slightly higher than the one at 0 mg/L SS concentration (3.47 ± 0.32 log reduction). This was in line with the results reported by Friedler et al. (2021) that UV disinfection might not be subject to the interference of particles in the water with low SS concentrations (less than 50 mg/L), such as absorbing or scattering the UV light. Another reason might be due to SS dispersing the original aggregated bacteria, allowing UV to kill more bacteria (Kollu & Ormeci 2015).
Figure 3

Effect of SS concentration on ARB inactivation by UVA/LED under irradiation time 60 min, light intensity 8 mW/cm2, pH 7, and initial bacteria concentration 107 CFU/mL.

Figure 3

Effect of SS concentration on ARB inactivation by UVA/LED under irradiation time 60 min, light intensity 8 mW/cm2, pH 7, and initial bacteria concentration 107 CFU/mL.

Close modal

Initial pH

Results from Figure 4 showed that the inactivation of ARB was the lowest with 3.47 ± 0.18 log reduction at neutral conditions (pH = 7), and the highest with 3.66 ± 0.12 log reduction at alkaline conditions (pH = 9). ARB inactivation efficiencies had only a slight amplitude of variation –5.5%. It was consistent with the finding by Chevremont et al. (2012) that alkaline condition was more conducive to bacterial inactivation. It might be because the growth and reproduction of strains had different sensitivities to different solution pH (Peng et al. 2019). For example, the extreme pH environment was very unfavorable for the survival of most microorganisms (Akkermans & Van Impe 2018; Tovar et al. 2020). However, Ng et al. (2016) reported that adjusting initial pH into alkaline could improve the disinfection performance five times. The discrepancy between the results in this study and other studies might be associated with the role of photocatalyst. For example, TiO2 was excited to generate hole and electron pairs under UVA irradiation, and then hydroxyl ions and H2O could react with the holes to generate ·OH, increasing the bacterial inactivation efficiencies (Zuo et al. 2015). In this study, alkaline conditions had a weak effect on improving ARB inactivation. The reason might be that alkaline conditions were conducive to the production of ROS (such as ·OH), but its output was very low and its role was limited (Chevremont et al. 2012; Huang et al. 2020). In general, initial pH had little effect on the ARB inactivation by UVA/LED irradiation.
Figure 4

Effect of initial pH on ARB inactivation by UVA/LED under irradiation time 60 min, light intensity 8 mW/cm2, and initial bacteria concentration 107 CFU/mL.

Figure 4

Effect of initial pH on ARB inactivation by UVA/LED under irradiation time 60 min, light intensity 8 mW/cm2, and initial bacteria concentration 107 CFU/mL.

Close modal

Initial bacteria concentrations

ARB inactivation efficiencies increased firstly and then decreased with the increase of initial bacteria concentration (Table 1). Appropriate initial bacteria concentration (106 CFU/mL) was beneficial to the largest ARB inactivation with 3.59 ± 0.14 log reduction in this study. On the contrary, ARB inactivation efficiencies with low initial bacteria concentration were only 3.09 ± 0.14 log reduction. This might be because at higher initial bacterial concentrations, the more bacteria per unit volume of simulated stormwater were exposed to UV light, and the greater the probability that bacteria would be inactivated. But, the inactivation of ARB decreased when the initial bacteria concentration continued to increase to 107 CFU/mL. Excessive bacteria counts could affect the turbidity, thereby impeding UV transmittance (Farrell et al. 2018). Moreover, it was proved that UV dose determined the yield of ROS (such as ·OH) (Song et al. 2019), which implied that ROS decreased with the increase of initial bacteria levels, and the redundant ARB would be not attacked.

Table 1

Effect of initial bacteria concentration on ARB inactivation by UVA/LED

Initial bacteria concentrations (CFU/mL)wavelength λ (nm)irradiation time (min)light intensity (mW/cm2)pHLog reduction
105 365 60 3.09 ± 0.14 
106 3.59 ± 0.14 
107 3.47 ± 0.18 
Initial bacteria concentrations (CFU/mL)wavelength λ (nm)irradiation time (min)light intensity (mW/cm2)pHLog reduction
105 365 60 3.09 ± 0.14 
106 3.59 ± 0.14 
107 3.47 ± 0.18 

Potential risks after the UVA/LED inactivation

Dark repair of ARB

ARB counts decreased rapidly within the beginning 2 h of the repair, and then tended to decrease slightly, which might be because UV light at 365 nm could destroy tRNA (Pezzoni et al. 2018). It increased sharply between 6 and 8 h of the repair, although there was no obvious dark repair of ARB after UVA/LED irradiation (Figure 5). Metral et al. (2018) reported that a small number of bacteria could use undamaged nucleic acids to provide a model for cell replication for repair, which might be the reason for the curve increase in the number of bacteria. And then it decreased greatly, which could be attributed to limited nutrients in water and residual ROS (such as H2O2, formed when exposed to UVA) (Sayed & Mitani 2017). Meanwhile, there were irreversible processes, that ROS destroyed repair enzymes and thus inhibited bacteria reactivation (Xiao et al. 2018; Cai et al. 2021), which could be proved by the fact that ARB was hardly detected after 12 h of the repair. These implied that the resurrection of ARB in the stormwater after UVA/LED irradiation was inhibited, ensuring the safety of effluent.
Figure 5

ARB dark repair in the following 24 h after 60 min of UVA/LED disinfection under light intensity 8 mW/cm2, pH 7, and initial bacteria concentration 106 CFU/mL.

Figure 5

ARB dark repair in the following 24 h after 60 min of UVA/LED disinfection under light intensity 8 mW/cm2, pH 7, and initial bacteria concentration 106 CFU/mL.

Close modal

Transmission of ARG

The conjugation transfer frequency of most ARGs was reduced, especially the transfer of ampicillin-resistant genes was significantly inhibited, and there was no obvious transfer of double and triple ARGs (Figure 6(a)). The possible reason was that the generated ROS penetrated the cell membrane, destroyed genomic DNA and inhibited the transfer function (Zhang et al. 2019). However, for single ARGs, there was no conjugation of kanamycin resistance gene in the control, occurring after UVA irradiation (transfer frequency was 5.43 × 10−4), implying that there was still a risk of antibiotic resistance transmission after UVA irradiation. It was in line with the finding that UV could destroy ARB cells and release ARGs to increase their concentration (Pokhum et al. 2017). Similarly, Dunlop et al. (2015) found that the sublethal stress induced by TiO2 under UVA irradiation increased the conjugate transfer of ARGs between E. coli. The possible reason was that UV caused the production of intracellular ROS and promoted the expression of bacterial junction-related genes (Cai et al. 2021).
Figure 6

Transmission risks of antibiotic resistance after 60 min of UVA/LED irradiation (a reflects conjugation transfer frequency of antibiotic resistant genes; and b reflects transformation frequency of antibiotic resistant genes, where light intensity 8 mW/cm2, pH 7, and initial bacteria concentration 106 CFU/mL; Control represented no UVA/LED irradiation).

Figure 6

Transmission risks of antibiotic resistance after 60 min of UVA/LED irradiation (a reflects conjugation transfer frequency of antibiotic resistant genes; and b reflects transformation frequency of antibiotic resistant genes, where light intensity 8 mW/cm2, pH 7, and initial bacteria concentration 106 CFU/mL; Control represented no UVA/LED irradiation).

Close modal

On the other hand, results from Figure 6(b) indicated that ARGs transformation phenomenon was not observed after UVA/LED irradiation compared with the one before UVA/LED disinfection. The significant decrease in ARB counts after UVA/LED disinfection resulted in a decrease in the contacting chance between ARB and the recipient bacteria, and thus reduced the transformation frequencies of ARGs. Meanwhile, this might be related to the increase of intracellular ROS levels caused by UVA irradiation (Song et al. 2019), which destroyed genetic elements and prevented transformation (Pousty et al. 2021). This indicated that 60 min of UVA/LED irradiation process could eliminate the risk of ARGs transformation.

In this study, ARB inactivation increased with the increase of light intensity and showed a good linear relationship, where the largest ARB inactivation reached 6.31 log reduction at 8 mW/cm2 of UVA/LED irradiation for 180 min. At the lower SS concentration (50 mg/L), ARB inactivation was the largest with 3.56 log reduction. There were low amplitudes of variation (less than 14%) for ARB inactivation under different initial pH and bacteria concentration, indicating that initial pH and bacteria levels had weak effects for ARB inactivation during UVA/LED irradiation. After the optimal UVA/LED irradiation, ARB was hardly detected during 12 h of dark repair, and little conjugation transfer and no transformation was found, implying that potential risks could be blocked after UVA/LED irradiation. These suggested that the application of UVA/LED to inactivate ARB in stormwater runoff was feasible and desirable in this study.

This work was supported by a grant from ‘the project of Jiangsu Specially-Appointed Professor’ (R2018T24) and ‘National Natural Science Foundation of China’ (52170099).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

Almakki
A.
,
Jumas-Bilak
E.
,
Marchandin
H.
&
Licznar-Fajardo
P.
2019
Antibiotic resistance in urban runoff
.
Science of the Total Environment
667
,
64
76
.
Cadet
J.
,
Grand
A.
&
Douki
T.
2014
Solar UV radiation-induced DNA bipyrimidine photoproducts: formation and mechanistic insights
. In: Barbatti, M., Borin, A. & Ullrich, S. (eds)
Photoinduced Phenomena in Nucleic Acids II
. Topics in Current Chemistry, vol. 356. Springer, Cham.
249
275
. https://doi.org/10.1007/128_2014_553.
Chevremont
A. C.
,
Farnet
A. M.
,
Sergent
M.
,
Coulomb
B.
&
Boudenne
J. L.
2012
Multivariate optimization of fecal bioindicator inactivation by coupling UV-A and UV-C LEDs
.
Desalination
285
,
219
225
.
Chevremont
A. C.
,
Boudenne
J. L.
,
Coulomb
B.
&
Farnet
A. M.
2013
Fate of carbamazepine and anthracene in soils watered with UV-LED treated wastewaters
.
Water Research
47
,
6574
6584
.
Davies
J.
&
Davies
D.
2010
Origins and evolution of antibiotic resistance
.
Microbiology and Molecular Biology Reviews
74
,
417
433
.
Dunlop
P. S. M.
,
Ciavola
M.
,
Rizzo
L.
,
McDowell
D. A.
&
Byrne
J. A.
2015
Effect of photocatalysis on the transfer of antibiotic resistance genes in urban wastewater
.
Catalysis Today
240
,
55
60
.
Farrell
C.
,
Hassard
F.
,
Jefferson
B.
,
Leziart
T.
,
Nocker
A.
&
Jarvis
P.
2018
Turbidity composition and the relationship with microbial attachment and UV inactivation efficacy
.
Science of the Total Environment
624
,
638
647
.
Ferreira
L. C.
,
Lucas
M. S.
,
Fernandes
J. R.
&
Tavares
P. B.
2016
Photocatalytic oxidation of reactive black 5 with UV-A LEDs
.
Journal of Environmental Chemical Engineering
4
,
109
114
.
Figueredo
M.
,
Rodríguez
E. M.
,
Rivas
J.
&
Beltrán
F. J.
2021
Photocatalytic ozonation in water treatment: is there really a synergy between systems?
Water Research
206
,
117727
.
Grob
C.
&
Pollet
B. G.
2016
Regrowth in ship's ballast water tanks: think again!
.
Marine Pollution Bulletin
109
,
46
48
.
Hamamoto
A.
,
Mori
M.
,
Takahashi
A.
,
Nakano
M.
,
Wakikawa
N.
,
Akutagawa
M.
, Ikehara, T., Nakaya, Y. & Kinouchi, Y.
2007
New water disinfection system using UVA light-emitting diodes
.
Journal of Applied Microbiology
103
,
2291
2298
.
Hatfull
G. F.
,
Dedrick
R. M.
&
Schooley
R. T.
2022
Phage therapy for antibiotic-resistant bacterial infections
.
Annual Review of Medicine
73
,
197
211
.
Hendriksen
R. S.
,
Munk
P.
,
Njage
P.
,
Van Bunnik
B.
,
McNally
L.
,
Lukjancenko
O.
, Röder, T., Nieuwenhuijse, D., Pedersen, S. K., Kjeldgaard, J., Kaas, R. S., Clausen, P. T. L. C., Vogt, J. K., Leekitcharoenphon, P., van de Schans, M. G. M., Zuidema, T., Husman, A. M. de R., Rasmussen, S., Petersen, B., The Global Sewage Surveillance project consortium, Amid, C., Cochrane, G., Sicheritz-Ponten, T., Schmitt, H., Alvarez, J. R. M., Aidara-Kane, A., Pamp, S. J., Lund, O., Hald, T., Woolhouse, M., Marion P. Koopmans, M. P., Håkan Vigre, H., Petersen, T. N. & Aarestrup, F. M.
2019
Global monitoring of antimicrobial resistance based on metagenomics analyses of urban sewage
.
Nature Communications
10
,
1
12
.
Hess-Erga
O. K.
,
Moreno-Andrés
J.
,
Enger
Ø.
&
Vadstein
O.
2019
Microorganisms in ballast water: disinfection, community dynamics, and implications for management
.
Science of the Total Environment
657
,
704
716
.
Huang
Y.
,
Su
L.
,
Zhang
S.
,
Zhao
Q.
,
Zhang
X.
,
Li
X.
, Li, H., Liu, L., Chen, J. & Wei, X.
2020
Opposite pH-dependent roles of hydroxyl radicals in ozonation and UV photolysis of genistein
.
Science of the Total Environment
709
,
136243
.
Jo
W. K.
&
Tayade
R. J.
2014
New generation energy-efficient light source for photocatalysis: LEDs for environmental applications
.
Industrial & Engineering Chemistry Research
53
,
2073
2084
.
Kebbi
Y.
,
Muhammad
A. I.
,
Sant'Ana
A. S.
,
do Prado-Silva
L.
,
Liu
D.
&
Ding
T.
2020
Recent advances on the application of UV-LED technology for microbial inactivation: progress and mechanism
.
Comprehensive Reviews in Food Science and Food Safety
19
,
3501
3527
.
Kollu
K.
&
Ormeci
B.
2015
UV-induced self-aggregation of E. coli after low and medium pressure ultraviolet irradiation
.
Journal of Photochemistry & Photobiology B Biology
148
,
310
321
.
Larsson
D. G. J.
&
Flach
C. F.
2022
Antibiotic resistance in the environment
.
Nature Reviews Microbiology
20
,
257
269
.
Lee
S.
,
Suits
M.
,
Wituszynski
D.
,
Winston
R.
,
Martin
J.
&
Lee
J.
2020
Residential urban stormwater runoff: a comprehensive profile of microbiome and antibiotic resistance
.
Science of the Total Environment
723
,
138033
.
Li
M.
,
Li
W.
,
Wen
D.
,
Qiang
Z.
&
Blatchley
E. R.
III
2017
Experimental evaluation of turbidity impact on the fluence rate distribution in a UV reactor using a microfluorescent silica detector
.
Environmental Science & Technology
51
,
13241
13247
.
Lu
J.
,
Wang
Y.
,
Li
J.
,
Mao
L.
,
Nguyen
S. H.
,
Duarte
T.
, Coin, L., Bond, P., Yuan, Z. & Guo, J.
2018
Triclosan at environmentally relevant concentrations promotes horizontal transfer of multidrug resistance genes within and across bacterial genera
.
Environment International
121
,
1217
1226
.
Lui
G. Y.
,
Roser
D.
,
Corkish
R.
,
Ashbolt
N. J.
&
Stuetz
R.
2016
Point-of-use water disinfection using ultraviolet and visible light-emitting diodes
.
Science of the Total Environment
553
,
626
635
.
Martín-Sómer
M.
,
Pablos
C.
,
van Grieken
R.
&
Marugán
J.
2017
Influence of light distribution on the performance of photocatalytic reactors: LED vs mercury lamps
.
Applied Catalysis B: Environmental
215
,
1
7
.
McGuigan
K. G.
,
Conroy
R. M.
,
Mosler
H. J.
,
du Preez
M.
,
Ubomba-Jaswa
E.
&
Fernandez-Ibanez
P.
2012
Solar water disinfection (SODIS): a review from bench-top to roof-top
.
Journal of Hazardous Materials
235
,
29
46
.
Metral
E.
,
Rachidi
W.
,
Damour
O.
,
Demarne
F.
&
Bechetoille
N.
2018
Long-term genoprotection effect of Sechium edule fruit extract against UVA irradiation in keratinocytes
.
Photochemistry and Photobiology
94
,
343
350
.
Ng
T. W.
,
Li
B.
,
Chow
A.
&
Wong
P. K.
2016
Effects of bromide on inactivation efficacy and disinfection byproduct formation in photocatalytic inactivation
.
Journal of Photochemistry and Photobiology A: Chemistry
324
,
145
151
.
Pei
M.
,
Zhang
B.
,
He
Y.
,
Su
J.
,
Gin
K.
,
Lev
O.
, Shen, G. & Hu, S.
2019
State of the art of tertiary treatment technologies for controlling antibiotic resistance in wastewater treatment plants
.
Environment International
131
,
105026
.
Pezzoni
M.
,
Pizarro
R. A.
&
Costa
C. S.
2014
Protective role of extracellular catalase (KatA) against UVA radiation in Pseudomonas aeruginosa biofilms
.
Journal of Photochemistry and Photobiology B: Biology
131
,
53
64
.
Pokhum
C.
,
Viboonratanasri
D.
&
Chawengkijwanich
C.
2017
New insight into the disinfection mechanism of Fusarium monoliforme and Aspergillus niger by TiO2 photocatalyst under low intensity UVA light
.
Journal of Photochemistry and Photobiology B: Biology
176
,
17
24
.
Pousty
D.
,
Hofmann
R.
,
Gerchman
Y.
&
Mamane
H.
2021
Wavelength-dependent time–dose reciprocity and stress mechanism for UV-LED disinfection of Escherichia coli
.
Journal of Photochemistry and Photobiology B: Biology
217
,
112129
.
Robertson
J. M.
,
Robertson
P. K.
&
Lawton
L. A.
2005
A comparison of the effectiveness of TiO2 photocatalysis and UVA photolysis for the destruction of three pathogenic micro-organisms
.
Journal of Photochemistry and Photobiology A: Chemistry
175
,
51
56
.
Rugh
M. B.
,
Grant
S. B.
,
Hung
W. C.
,
Jay
J. A.
,
Parker
E. A.
,
Feraud
M.
, Li, D., Avasarala, S., Holden, P. A., Liu, H., Rippy, M. A., Van DeWerfhorst, L. C., Kefela, T., Peng, J., Shao, S., Graham, K. E., Boehm, A. B., Choi, S., Mohanty, S. K. & Cao, Y.
2022
Highly variable removal of pathogens, antibiotic resistance genes, conventional fecal indicators and human-associated fecal source markers in a pilot-scale stormwater biofilter operated under realistic stormflow conditions
.
Water Research
219
,
118525
.
Sayed
A. E. D. H.
&
Mitani
H.
2017
Immunostaining of UVA-induced DNA damage in erythrocytes of medaka (Oryzias latipes)
.
Journal of Photochemistry and Photobiology B: Biology
171
,
90
95
.
Senavirathne
G.
,
Bertram
J.
,
Jaszczur
M.
,
Pham
P.
,
Mak
C.
,
Goodman
M. F.
& Rueda, D.
2014
Watching aid scanning single stranded and transcribed DNA with single molecule resolution
.
Biophysical Journal
106
,
692a
.
Sholtes
K. A.
,
Lowe
K.
,
Walters
G. W.
,
Sobsey
M. D.
,
Linden
K. G.
&
Casanova
L. M.
2016
Comparison of ultraviolet light-emitting diodes and low-pressure mercury-arc lamps for disinfection of water
.
Environmental Technology
37
,
2183
2188
.
Tovar
M.
,
Mahler
L.
,
Buchheim
S.
,
Roth
M.
&
Rosenbaum
M. A.
2020
Monitoring and external control of pH in microfluidic droplets during microbial culturing
.
Microbial Cell Factories
19
,
1
9
.
Wang
X.
,
Hu
X.
,
Hu
C.
&
Wei
D.
2011
Sequential use of ultraviolet light and chlorine for reclaimed water disinfection
.
Journal of Environmental Sciences
23
,
1605
1610
.
Wang
J.
,
Liu
H.
,
Wang
Y.
,
Ma
D.
,
Yao
G.
,
Yue
Q.
, Gao, B. & Xu, X.
2022
A new UV source activates ozone for water treatment: wavelength-dependent ultraviolet light-emitting diode (UV-LED)
.
Separation and Purification Technology
280
,
119934
.
Wetzel
D.
&
McBride
S. M.
2020
The impact of pH on Clostridioides difficile sporulation and physiology
.
Applied and Environmental Microbiology
86
,
e02706
19
.
Zhang
S.
,
Pang
S.
,
Wang
P.
,
Wang
C.
,
Han
N.
,
Liu
B.
, Han, B., Li, Y. & Anim-Larbi, K.
2016
Antibiotic concentration and antibiotic-resistant bacteria in two shallow urban lakes after stormwater event
.
Environmental Science and Pollution Research
23
,
9984
9992
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).