Various environmental pollutants and harmful pathogenic microorganisms can be neutralized through advanced oxidation processes (AOPs), which are physicochemical treatments that generate potent reactive oxygen radicals. Currently, dozens of chemicals and treatments are applied in AOPs. Among these, peracetic acid (PAA) has emerged as a versatile disinfectant for microbes in water and on surfaces (Biswal et al. 2014; Cai et al. 2017; De Souza et al. 2015; Gehr et al. 2003; Kibbee and Örmeci 2020; Sun et al. 2018; Weng et al. 2018). PAA offers several advantages over classic chlorination owing to the reduced formation of harmful disinfection by-products (DBPs), mutagens of organisms, and persistent residues in the environment. On a mechanistic basis, PAA generates strong reactive oxygen species, such as hydroxyl (∙OH), hydroperoxyl (∙HO₂), superoxide (∙O^2-), methyl peroxyl (∙OOCH₃), and acetyloxyl (∙CH₂C(O)O^-) radicals and H₂O₂ (Zhang and Huang 2020). PAA can penetrate the cell wall at a broad pH range and is not affected by catalase activity. These radicals are formed through the co-treatment of PAA and UV radiation and/or other oxidants, which increases the disinfection efficiency (Cai et al. 2017; Zhang et al. 2020, 2022). Specifically, when used concurrently, UV and PAA exert the effects by counterbalancing their limitations. Moreover, requirement of a high PAA dosage can be compromised by UV irradiation, and a highly particulate matrix with low UV transmittance can be efficiently exposed to PAA. The combination of UV and PAA induces hemolytic breakage of the O-O bond in the PAA molecule to form a hydroxyl radical (Caretti and Lubello 2003). Thus, the combination of PAA and UV irradiation is more effective than the combination of H₂O₂ and UV irradiation.

In terms of microbial disinfection activity and the underlying mode of action, concurrent or sequential treatment with UV and PAA exhibited synergistic and/or additive effects against various gram-negative (Escherichia coli and Pseudomonas aeruginosa) and gram-positive (Enterococcus durans and Staphylococcus epidermidis) bacteria (Caretti and Lubello 2003; De Souza et al. 2015; Hassaballah et al. 2019; Sun et al. 2018; Zhang et al., 2020, 2022). De Souza et al. (2015) showed that combined treatment with PAA and UV irradiation provided superior efficacy in disinfecting Escherichia coli (4 ppm PAA + 60/90 s UV for 10 min). Moreover, Sun et al. (2018) showed that UV-activated PAA (9 ppm) led to greater inactivation of Escherichia coli than either PAA or UV alone. In a recent study, combined PAA and UV treatment achieved greater microbial inactivation even with a pilot-scale wastewater treatment for a long time (Hassaballah et al. 2019).

Furthermore, as irrigation is essential to cultivate crops and vegetables, water disinfection is important to ensure food safety in the agricultural industry. Based on its proven efficacy and safety, PAA is a potential AOP agent for disinfecting microbes during plant cultivation processes. However, the effects of PAA on plant growth warrant investigation. To date, several studies have explored the effects of PAA on plant growth. PAA is an effective chemical biocide and commercially used for direct foliar application (5.6%; Peragreen^®, Enviro Tech) against powdery mildew. When applied to tomatoes for a short period (< 2 h) under hydroponic conditions, PAA inhibited vegetative growth, transient wilting, and retardation in roots, which was accompanied by oxidative stress due to H₂O₂ dissociation from PAA (Vines et al. 2003). However, with prolonged exposure, tomatoes developed PAA tolerance. Moreover, treatment a PAA mixture (< 40 ppm) for a long time improved the growth and yield of watercress under hydroponic conditions (Carrasco et al. 2011). Notably, 1% PAA treatment via soil-drenching significantly reduced bacterial wilt in tomato seedlings (Hong et al. 2018). Meanwhile, treatment with moderate PAA concentrations (40 ppm) induced oxygenation in the roots but improved the vegetative growth of watercress (Carrasco et al. 2011). Nevertheless, the effects of PAA on plant growth remain controversial and warrant comprehensive examination. Therefore, research into the effects of UV-pre-activated PAA on plant growth is meaningful.

Plants defend themselves against various pathogenic attacks through the innate immune system, largely known as the pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) or effector-triggered immunity (ETI) (Jones and Dangl 2006; Ngou et al. 2022). PTI may be accompanied by the “priming” process-a form of plant immunological memory in response to external stimuli (pests, pathogens, or chemicals, among others) (Mauch-Mani et al. 2017). Priming pre - activates defense mechanisms that lead to a faster and stronger defense response upon subsequent attacks by a pest or pathogen. From the genetic perspective, the most reliable hallmark of priming is the expression level of PR1, which shows the fastest induction. In addition, priming is correlated with the hyper-induction of PTI marker genes, such as WRKY53, FLG22-INDUCED RECEPTOR KINASE1 (FRK1), and NDR1/HIN1-LIKE10 (NHL10) (Boudsocq et al. 2010; Xiao et al. 2007).

In the present study, we tested the disinfection efficiency of a PAA solution pre-irradiated with UV (UV/PAA) against Escherichia coli - a representative species of bacteria in water. Further, we investigated the impact of UV/PAA on plant growth by monitoring the physiological parameters of Arabidopsis plants treated with UV/PAA as the major water source. In addition, we examined the defense of Arabidopsis treated with UV/PAA solution against Pseudomonas syringae pv. tomato (Pst) DC3000 as a potential pathogenic bacterial challenge during cultivation.

Experimental materials

PAA was purchased in commercial (PROXITANE® 15:10, Solvay, Belgium). The stock solution (50 mg·L^-1) was prepared based on PAA concentration in the original bottle determined by iodometric titration, and its pH was adjusted to 7.0 (Dominguez-Henao et al. 2018). Just prior to use, the working PAA solution (i.e., 5 and 10 ppm) was diluted from the stock solution using 10 mM phosphate- buffered saline (PBS, pH 7.0).

Arabidopsis thaliana ecotype Col-0 was used in all experiments. Seeds were sterilized by rocking in 2.5% NaOCl containing 0.02% Triton X-100 for 7 min, followed by rinsing for five times with sterilized water. The seeds were inoculated on Murashige and Skoog (MS) medium [2.15 g·L^-1 MS salt, 0.5 g·L^-1 MES [2-(N-morpholino) ethanesulfonic acid], 5 g·L^-1 sucrose, and 0.8% phyto agar; pH 5.7]. After vernalization for 2 days at 4°C in the dark, the seeds germinated and were grown under 8 h of light (120 µmol·m^-2·s^-1) and 16 h of dark conditions for 10 days. Then, the seedlings were transferred to pots containing artificial peat pellets (Jiffy, Netherlands) or soil and placed on homemade hydroponic growth platforms under the same light regime as that during the germination stage.

Disinfection of Escherichia coli with the UV/PAA solution

After overnight culture in LB broth medium at 37°C and 220 rpm, the density of Escherichia coli was adjusted to 1.2 × 10⁸ CFU·mL^-1 using 10 mM PBS (pH 7.0) and subjected to stirring in 30 mL of PAA working solution under UV irradiation (0.2 mW·cm^-2). Then, 1 mL of the reaction mixture was removed at different time points, and the dissociated radicals were quenched with sodium thiosulfate to achieve a concentration of 30 mM. The Escherichia coli pellet was harvested via centrifugation, suspended in 200 µL of water, and spread on LB agar plates. After 1 day of incubation at 37°C, the colonies were counted manually.

Monitoring of the vegetative growth of UV/PAA-treated Arabidopsis

PAA working solutions (5 and 10 ppm) were exposed to UV-C light (0.2 mW·cm^-2) for 1 min. Nano-pure water exposed to the same intensity of UV-C light was used as the non-treated (NT) control (0 ppm PAA). To monitor the effect of UV/PAA on the vegetative growth of Arabidopsis, seedlings were transferred from MS plates to hydroponic pots. Seedling roots were submerged in a tank filled with 1× Hoagland solution (van Delden et al. 2020) with or without the UV/PAA solution, which was refreshed every second day. The growth parameters were scored, and photographs were obtained on the 21^th day.

Measurement of Arabidopsis defense against Pst DC3000

To examine the effect of UV/PAA on Arabidopsis defense against Pst DC3000 infecting leaves, the seedlings were transferred from MS plates to peat pellets (44 mm in diameter and 42 mm in height) and then watered with 5 mL of distilled water or UV/PAA solution daily for 21 days. Then, rosette leaves were infected via bacterial infiltration using a piston syringe (1 mL) filled with 500 µL of Pst DC3000 suspension (1 × 10⁷ CFU·mL^-1 in 10 mM MgCl₂), which was prepared by culturing the bacteria in LB broth supplemented with 50 µg·mL^-1 of rifampicin. After 3 days of infection, colonies of residual Pst DC3000 in leaves (leaf disc with a diameter of 6 mm) were counted on LB agar plates containing 50 µg·mL^-1 of rifampicin. For the infection of whole plants, seedlings grown in meshed soil pots (50 mm in diameter and 70 mm in height) watered with 5 mL of 1× Hoagland solution with or without UV/PAA every day for 1 month were immersed upside down in Pst DC3000 suspension (1 × 10⁷ CFU·mL^-1 in 10 mM MgCl₂ containing 0.02% Silwet 77) for 30 s. The whole plants were drained and air-dried in the dark for 3 h and then placed under normal growth conditions. After 10 days, fresh weight (FW) was measured, and the plants were photographed.

H₂O₂ measurement

For qualitative measurement of H₂O₂, whole leaves were excised and immersed in 3,3’-diaminobenzidine (DAB) staining solution (Daudi and O’Brien 2012) for 9 h. For destaining, the leaves were immersed in a solution containing glycerol, ethanol, and acetic acid (4:4:2). For quantitative measurement, leaf discs (6 mm in diameter) were excised from leaves infected with Pst DC3000 and homogenized with 0.05 M PBS (pH 7.4). H₂O₂ content was measured using an EZ-hydrogen peroxide/peroxidase assay kit (DoGenBio Co., Ltd, Korea) according to the manufacturer’s instructions.

Callose deposition measurement

To monitor callose (β-1,3-d-glucan polymer) accumulation in leaves infected with Pst DC3000, the leaves were detached and sequentially immersed in 95% and 50% ethanol for 1 h, in that order. The leaves were transferred to 67 mM K₂HPO₄ for 1 h, followed by staining with 0.01% aniline blue solution for 1 h. After washing with 67 mM K₂HPO₄ for 1 h, the leaves were observed under UV or visible light with a microscope. Whole leaves were homogenized using a mortar and pestle and subjected to fluorescence measurements using a fluorometer (FP-8300; Jasco, Japan).

Gene expression profiling

Twenty-one-day-old seedlings grown under hydroponic conditions with or without UV/PAA solution were removed and submerged in a Pst DC3000 suspension (1 × 10⁹ CFU·mL^-1 in 10 mM MgCl₂). Seedlings were harvested at indicated time points (0-9 h) and snap-frozen in liquid nitrogen. Total RNA was isolated from the seedlings using TRIzol™ (Invitrogen, CA, USA). First-strand cDNA was synthesized using oligo-dT primers and reverse transcription reagents provided in the PrimeScript^TM 1^st Strand cDNA Synthesis Kit (Takara, Japan). Designated genes were real-time amplified using specific primers (Table 1) and PCR reagents provided in TOPreal^TM qPCR 2X PreMIX (SYBR Green with low ROX) (Enzynomics, Korea). PCR (denaturation at 95°C for 30 s, annealing at 53°C for 30 s, and extension at 72°C for 30 s) and C_t value calculation were performed using the CFX Connect^TM Real-Time PCR system (Bio-Rad Laboratories, CA, USA).

Table 1 . Oligonucleotide sequences of primers used for qRT-PCR in the present study.

Gene	Nucleotide sequences	Genome locus ID/annotated function
ACT2	5′-GGTAACATTGTGCTCAGTGGTGG-3′	AT3g18780/cytoskeletal actin
	5′-AACGACCTTAATCTTCATGCTGC-3′
GST1	5′-AACCGTTGTTGAAGAAGA-3′	AT1g02930/glutathione S-transferase
	5′-GTCAGCAACCCAAGCACTCACAT-3′
PR1	5′-ATAATCAGTTGCAACTATGATCCTC-3′	AT2g14610/pathogenesis-related gene
	5′-AAATAGATTCTCGTAATCTCAGCTC-3′
NHL10	5′-TTCCTGTCCGTAACCCAAAC-3′	AT2g35980/Arabidopsis NDR1/HIN1-like 10
	5′-CCCTCGTAGTAGGCATGAGC-3′
FRK1	5′-GCCAACGGAGACATTAGAG-3′	AT2g19190/FLG22-induced receptor-like kinase 1
	5′-TCTAGACCCGGCACATACAA-3′

Data on gene locus IDs in the Arabidopsis genome and the annotated functions of the encoded proteins were obtained from http://www.arabidopsis.org..

Disinfection efficiency of UV/PAA against Escherichia coli in water

The efficacy of combined treatment with UV irradiation and PAA for the disinfection of various bacteria has been extensively documented. In the present study, we recapitulated the disinfection activity of UV/PAA against Escherichia coli in water (Fig. 1). At the dose of 12 mJ·cm^-2 (equivalent to irradiation with 0.2 mW·cm^-2 UV for 60 s), UV treatment, per se, inactivated Escherichia coli at a 4-log scale. Under UV/PAA treatment, Escherichia coli inactivation was facilitated as a function of the UV dose and PAA concentration. The averaged reduction scale of Escherichia coli population following treatment with 5 and 10 ppm UV/PAA was 5.1 and 6.7 log, respectively. Meanwhile, PAA treatment without UV irradiation showed a much lower disinfection efficiency than UV/PAA treatment (Fig. 1. inset). As such, 5 and 10 ppm of PAA without UV irradiation achieved respectively 0.5 and 1.5 log reduction in Escherichia coli. Therefore, UV/PAA treatment efficiently disinfected Escherichia coli in water.

Figure 1. Disinfection kinetics of UV-activated PAA against Escherichia coli. Escherichia coli cells (1 × 10⁸ CFU·mL^-1) were suspended in different PAA concentrations with or without UV irradiation (0.2 mW·cm^-2) and then stirred. At the indicated time points, 1 mL of the reaction suspension was removed, and the residual radicals were quenched using sodium thiosulfate. The UV intensity at the given time points was expressed as the UV dose (mJ·cm^-2). The inset shows the disinfection efficiency of PAA alone. Error bars indicate standard deviation (n = 3)

Effect of UV/PAA on Arabidopsis growth under hydroponic conditions

To investigate the response of Arabidopsis to UV/PAA treatment, we monitored the growth of its seedlings under hydroponic conditions. Arabidopsis seedlings continuously exposed to UV/PAA solution exhibited reduced size and greenish phenotype of leaves compared with NT seedlings (Fig. 2A and 2B). The average FW of seedlings under NT, 5 ppm UV/PAA, and 10 ppm UV/PAA was 0.044, 0.038, and 0.036 g, respectively (Fig. 2C). Moreover, primary root elongation was inhibited in seedlings exposed to the UV/PAA solution. Specifically, the average length of primary roots under 0, 5, and 10 ppm of UV/PAA was 5.0, 3.1, and 2.0 cm, respectively (Fig. 2D). Since leaf bleaching and short roots are well-known phenotypes expressed under oxidative stress, we measured the content of H₂O₂ - a reactive oxygen species (ROS) - in whole seedlings. Compared with NT controls (2.98 nmole·g^-1), seedlings exposed to 5 (4.06 nmole·g^-1) and 10 (4.30 nmole·g^-1) ppm UV/PAA showed elevated H₂O₂ levels (Fig. 2E). Therefore, following continuous exposure of roots to UV/PAA solutions, Arabidopsis may experience moderate oxidative stress.

Figure 2. Physiological responses of Arabidopsis to UV/PAA treatment under hydroponic conditions. (A) Morphology of Arabidopsis seedlings grown in hydroponic media supplemented with UVtreated PAA solution. (B) Morphology of whole seedlings grown in UV/PAA-supplemented hydroponic media. (C) Average fresh weight (FW) of seedlings. (D) Average length of primary roots. (E) Average H₂O₂ content in whole seedlings. Scale bar = 1 cm. Error bars indicate standard deviation (n = 9). In C, D, and E, statistical significance between UV/PAA and NT treatments was determined using Student’s t-test (*p < 0.05; **p < 0.01)

Effect of UV/PAA on Arabidopsis defense against Pst DC3000

To test whether pre-exposure to UV/PAA altered the defense response of Arabidopsis to pathogenic attacks, we infected plants with Pst DC3000. According to leaf infiltration with the pathogen, Arabidopsis leaves exhibited bleaching marks of infection, which were similarly observed under all conditions at 1 day post-infection (DPI) (Fig. 3A). However, at 3 DPI, the leaves of Arabidopsis grown under NT conditions exhibited severe curling, bleaching, and wilting (Fig. 3A). Meanwhile, although the leaves of Arabidopsis treated with UV/PAA were bleached, their turgor was maintained. Moreover, the density of bacteria was higher in the infected leaves of NT controls (6 × 10⁵ CFU·mL^-1) than in those of seedlings treated with 5 (5.5 × 10⁴ CFU·mL^-1) and 10 (1.3 × 10⁴ CFU·mL^-1) ppm UV/PAA (Fig. 3B). Therefore, based on the morphological phenotype and residual bacterial population in infected leaves, UV/PAA treatment enhanced the defense of Arabidopsis against Pst DC3000.

Figure 3. Defensive responses of UV/PAA-treated Arabidopsis against Pst DC3000 infection. (A) Leaf morphology of UV/PAAtreated Arabidopsis samples according to Pst DC3000 infection. (B) Average residual bacterial density in Arabidopsis leaves at 3 days post-infection (DPI). Scale bar = 1 cm. NT, non-treated control. Error bars indicate standard deviation (n = 3). Statistical significance between UV/PAA and NT treatments was determined using Student’s t-test (*p < 0.05; **p < 0.01)

H₂O₂ and callose accumulation in UV/PAA-treated Arabidopsis

H₂O₂ accumulation is a typical response of plants challenged with pathogenic attacks, and it is often correlated with hyper-response (HR) or programmed cell death (PCD). Therefore, in addition to scoring bacteria density (CFU·mL^-1), we measured H₂O₂ content in the infected leaves of Arabidopsis treated with the UV/PAA solution. In DAB staining, Arabidopsis leaves exhibited a faint yellow color on the day of infection, indicating trace H₂O₂ accumulation. This was similarly observed under all conditions on the day of infection (Fig. 4A). However, at 3 DPI, leaves of UV/PAA-treated plants exhibited more intense staining than those of controls (Fig. 4A). In quantitative measurements, H₂O₂ levels in plants that were not infected with Pst DC3000 were below 2 nmol·g^-1 under all conditions (Fig. 4B). However, with Pst DC3000 infection, H₂O₂ levels increased under all UV/PAA treatments. Specifically, H₂O₂ levels under NT, 5 ppm UV/PAA, and 10 ppm UV/PAA treatments were 6.8, 9.0, and 12.1 nmole·g^-1, respectively. Therefore, the UV/PAA treatment facilitated H₂O₂ accumulation in Arabidopsis leaves challenged with Pst DC3000.

Figure 4. H₂O₂ and callose accumulation in the leaves of UV/PAAtreated Arabidopsis according to Pst DC3000 infection. (A) DAB staining patterns of Arabidopsis leaves. (B) Average H₂O₂ accumulated in Arabidopsis leaf tissues. (C) Fluorescence pattern resulting from callose accumulation in Arabidopsis leaves. (D) Fluorescence signal intensity in Arabidopsis leaf extracts. Scale bar = 1 cm (A) and 0.5 cm (C). DPI, days post-infection; FL, images of fluorescence emitted by UV irradiation; BR, brightfield images. Error bars indicate standard deviation (n = 3). Statistical significance between UV/PAA and NT treatments was determined using Student’s t-test (**p < 0.01)

Callose is a β-1,3-d-glucan polymer. In Arabidopsis, its synthesis increases under pathogenic attack, as this polymer rigidifies the cell wall structure, acting as a physical barrier against pathogens. To examine the effects of UV/PAA on Arabidopsis defense, we assessed callose deposition in the leaves of UV/PAA-treated plants according to Pst DC3000 infection (Fig. 4C). Callose accumulation was inferred based on fluorescence using aniline blue staining. At 3 DPI, callose deposition was increased in the leaves of Arabidopsis treated with UV/PAA compared with that in the leaves of NT controls. In semi-quantitative measurements of fluorescence levels in the whole leaf, signal intensity in the leaves of plants treated with 5 and 10 ppm UV/PAA was respectively 1.7-fold and 2.4-fold higher than that in the leaves of NT controls (Fig. 4D). Together, these data indicate that the UV/PAA treatment promoted callose synthesis and deposition in leaves infected by Pst DC3000.

Post-infection growth of UV/PAA-treated Arabidopsis infected with Pst DC3000

To investigate whether plants treated with UV/PAA maintained their growth during the post-infection period, we monitored the phenotype and biomass of Arabidopsis for an extended period (10 days) after Pst DC3000 infection. Whole Arabidopsis plants grown in soil pots were submerged upside down in a Pst DC3000 suspension for bacterial infiltration and then recovered under normal growth conditions. As shown in Fig. 2, in the absence of bacterial infection, Arabidopsis plants grown under NT conditions were slightly larger and showed greater biomass accumulation than those exposed to UV/PAA (Fig. 5A). Average FW ranged between ~0.7 and 0.8 g per plant across all conditions (Fig. 5B). However, in the presence of Pst DC3000 infection, Arabidopsis plants grown under NT conditions exhibited severe growth retardation with pathogenic symptoms (e.g., wilting and bleaching). In contrast, in Arabidopsis plants treated with UV/PAA, most leaves remained greenish and healthy. Moreover, the average FW (0.35 g per plant) of plants grown in NT conditions significantly decreased post-infection, whereas that of plants treated with 5 and 10 ppm UV/PAA was respectively 0.67 and 0.64 g per plant (Fig. 5B). Therefore, UV/PAA likely mitigated the biotic stress induced by Pst DC3000 infection and helped maintain Arabidopsis growth.

Figure 5. Long-term growth behavior of UV/PAA-treated Arabidopsis after Pst DC3000 infection. (A) Morphology of Arabidopsis plants after Pst DC3000 infection. (B) Average fresh weight (FW) of Arabidopsis plants after Pst DC3000 infection. UI, Pst DC3000- uninfected; INF, Pst DC3000-infected. Error bars indicate standard deviation (n = 3). Statistical significance between UV/PAA and NT treatments was determined using Student’s t-test (**p < 0.01)

Expression of defense marker genes in UV/PAA-treated Arabidopsis

To explore the effects of UV/PAA on the stress response and defense potential of Arabidopsis in greater detail, we assessed the expression profiles of several well-known defense marker genes following Pst DC3000 infection (Fig. 6). As shown in Fig. 2, UV/PAA treatment increased H₂O₂ levels in Arabidopsis even without pathogen challenge. Consistently, the expression level of GST1, which encodes glutathione S-transferase that detoxifies oxides at the cellular level, was higher in UV/PAA-treated plants than in NT controls. Moreover, following Pst DC3000 infection, the transcript levels of GST1 were higher in UV/PAA- treated plants than in NT controls. Meanwhile, PR1 - the most reliable marker for defense potential - was upregulated after 9 h of Pst DC3000 infection. Remarkably, PR1 expression was rapidly and markedly induced in Arabidopsis treated with UV/PAA according to Pst DC3000 infection. Interestingly, after peaking at 6 h post-infection, the transcript levels of PR1 in Arabidopsis treated with UV/PAA reduced to the base level at 9 h. These results indicate that UV/PAA treatment expedited PR1 expression faster than NT treatment. Other defense markers, such as PR2 and PR5, exhibited similar expression patterns to PR1 (data not shown). Previously, NHL10 and FRK1 have been implicated in the regulation of PTI and “priming” processes in Arabidopsis. After 3 h of Pst DC3000 infection, both NHL10 and FRK1 were rapidly upregulated, and their transcript levels were significantly higher in UV/PAA-treated plants than in NT controls. In Arabidopsis exposed to UV/PAA, NHL10 and FRK1 were downregulated, and their expression dropped to the base level, whereas these genes were gradually upregulated in Arabidopsis grown under NT conditions. Taken together, UV/PAA treatment stimulated the expression of defense genes against oxidative stress and bacterial infection in Arabidopsis.

Figure 6. Transcript profiles of defense marker genes in UV/PAAtreated Arabidopsis according to Pst DC3000 infection. According to Pst DC3000 infection, the expression profiles of representative oxidative (GST1) and defense marker (PR1, NHL10, and FRK1) genes in Arabidopsis samples with or without UV/PAA were quantitatively assessed at different time courses. Relative expression was presented as the averaged 2(ΔCt) values of each gene and ACT2, an internal control. Error bars indicate standard deviation (n = 3). Statistical significance between UV/PAA and NT treatments after 3 to 9 h was determined using Student’s t-test (*p < 0.05; **p < 0.01)

With increasing demand for irrigation and fresh green products, water disinfection is recognized as the first-line method to ensure food safety. Based on accumulated evidence, combined treatment with UV and PAA is a reasonable AOP for efficient disinfection of a broad spectrum of bacteria in water. In addition, this technique is versatile in reducing pathogenic contamination of food materials in the agricultural industry. However, the effects of UV and PAA co-treatment on plant growth and development remain largely unknown. Therefore, in the present study, we confirmed that a UV-activated PAA solution moderately inhibited vegetative growth and altered developmental processes (Fig. 2) but improved the defense of Arabidopsis against pathogenic Pst DC3000 (Figs. 3 and 5). This increase in defense potential was accompanied by the accumulation of H₂O₂ and deposition of callose (Fig. 4). In addition, based on the gene expression profiles of several defense markers, the innate immunity in Arabidopsis exposed to UV/PAA was enhanced through rapid induction of defense mechanisms (Fig. 6).

In the present study, the UV/PAA treatment induced oxidative stress in Arabidopsis seedlings (Fig. 2). UV/ PAA generates various ROS, including ∙OH, ∙HO₂, ∙O^2-, ∙OOCH₃, and ∙CH₂C(O)O^- radicals and H₂O₂ (Zhang and Huang 2020). Therefore, exogenous UV/PAA treatment likely formed such radicals in the solution as well as inside the plant cells. Typically, ROS overaccumulation interrupts functions in many compartments, such as chloroplasts, mitochondria, peroxisomes, and plasma membranes, and induces oxidative stress within cells, producing numerous detrimental effects on the growth and developmental processes of plants (Dmitrieva et al. 2020; Hasanuzzaman et al. 2020). Thus, quantitative assessment of ROS in Arabidopsis tissues provides accurate insights into stress levels. However, accurate measurement of ROS levels within plant cells is challenging because of their rapid turnover rates. Nonetheless, despite limited accuracy of stoichiometric assessment, ROS generated via UV/PAA treatment may induce oxidative stress in Arabidopsis. Furthermore, PAA treatment caused transient wilting, reduced transpiration, and promoted CO₂ assimilation in tomatoes (Vines et al. 2003). In particular, PAA-treated tomato and watercress exhibited oxygenation and root damage (Carrasco et al. 2011; Vines et al. 2003). Interestingly, however, long-term treatment induced PAA tolerance and improved the growth and yield of plants treated with < 40 ppm PAA. Consistently, in the present study, Arabidopsis treated with UV/PAA exhibited marginal growth retardation but maintained normal growth (Fig. 5). Therefore, plants can acclimate and adapt to PAA-driven oxidative stress if treated at a moderate level.

PAA is manufactured as a mixed solution in which PAA, acetic acid, H₂O₂, and water are in equilibrium. Therefore, acetic acid and H₂O₂ may be the other stressors affecting plant growth. However, we adjusted the pH of the PAA solution to 7.0, at which most of the acetic acid (pKa = 4.8) is protonated. According to Vines et al. (2003), oxidative rather than acidic mechanisms are primarily responsible for the phytotoxicity of PAA solutions. Nonetheless, the effect of H₂O₂ on the stress level of Arabidopsis must be taken into account because of its abundance and similar roles of ROS in PAA solutions. In the present study, the original PAA solution contained 15% PAA and 10% H₂O₂. Accordingly, the concentration of H₂O₂ in 10 ppm PAA solution was approximately 0.2 mM. However, the dissociation of PAA into H₂O₂ following UV irradiation increased H₂O₂ concentration in the PAA solution. Therefore, dissociated H₂O₂ from the PAA solution may be attributed to the increased H₂O₂ levels in Arabidopsis tissues. Based on the direct measurements on Arabidopsis grown under hydroponic conditions (Fig. 2), H₂O₂ levels in plants treated with UV/PAA were higher than those in untreated controls. In addition, the expressions of GST1 encoding a ROS scavenging enzyme was highly induced in Arabidopsis treated with UV/PAA solution even in the absence of pathogen infection (Fig. 6). Nonetheless, whether the increase in H₂O₂ levels in UV/PAA-treated plants was due to exogenous input or endogenous de novo generation induced as a secondary response triggered by PAA remains unclear.

In the present study, the most prominent effect of UV/PAA treatment on Arabidopsis was the enhancement of defense against pathogenic challenge. The effects of PAA on plant protection against pathogens have been previously reported (Carrasco et al. 2011; Hong et al. 2018; Vines et al. 2003). Nevertheless, whether these outcomes were driven by the direct action of PAA on pathogens or increase in the tolerance of plants to (a)biotic stress remains unclear. In the present study, we designed an experiment to exclude the possibility of direct contact between UV/PAA and Pst DC3000. Therefore, the enhancement of defense potential acquired by the UV/PAA treatment may be attributed to the augmentation of plant innate immunity mediated via pre-acclimation or pre-exposure to UV/PAA. This phenomenon is called “priming” (pretreatment of plants through exposure to stressors or chemicals, improving tolerance to subsequent or secondary stress events). Priming is an important mechanism of induced resistance to biotic stresses (Beckers and Conrath 2007; Borges et al. 2014; Tanou et al. 2012). The increased priming potential of Arabidopsis treated with UV/PAA is consistent with the rapid induction of the PR1 marker gene (Fig. 6) as well as the hyper-induction of PTI marker genes, such as WRKY53, FRK1, and NHL10 (Singh et al. 2014). PTI is the first innate immune layer activated upon pathogen perception. Among PTI genes, WRKY53, FRK1, and NHL10 (Boudsocq et al. 2010; Xiao et al. 2007) also serve as the markers for priming. In the present study, the rapid upregulation of NHL10 and FRK1 in Arabidopsis treated with UV/PAA indicates that the priming processes expedited PTI against Pst DC3000.

Interestingly, in response to Pst DC3000 infection, Arabidopsis leaves accumulated higher levels of H₂O₂. This de novo generation of H₂O₂ differs from exogenous H₂O₂ present in PAA. In general, plants undergo an oxidative burst against pathogen attacks, which results in a drastic increase in H₂O₂ levels and a concurrent hypersensitive response (HR). In the present study, under Pst DC3000 challenge, more endogenous H₂O₂ was accumulated in UV/PAA-treated plants than in untreated controls (Fig. 4). Moreover, the expression of GST1, which encodes glutathione S-transferase - a radical scavenging enzyme - was highly and rapidly induced upon Pst DC3000 challenge in Arabidopsis treated with UV/PAA (Fig. 6). These results confirm that UV/PAA mediates signaling pathways that regulate endogenous H₂O₂ production and defense mechanisms. Taken together, although the precise roles of PAA and/or H₂O₂ within tissues remain obscure, endogenous UV/PAA-mediated increase in H₂O₂ levels must be an important factor underlying the enhancement of Arabidopsis defense potential.

Endogenous H₂O₂, generated through various environmental stimuli, acts as a stressor or signaling molecule regulating stress adaptation and PCD (Apel and Hirt 2004). Specifically, H₂O₂ plays a dual role in plants. At low concentrations, it acts as a signaling molecule triggering stress tolerance (Fukao and Bailey-Serres 2004; Mittler et al. 2004; Quan et al. 2008), whereas at high concentrations, it orchestrates PCD (Dat et al. 2000) and regulates various developmental and physiological processes, such as senescence (Liao et al. 2012), flowering (Liu et al. 2013), and root system architecture (Liao et al. 2009; Ma et al. 2014) via crosstalk with other signaling components. Interestingly, pretreatment with a low concentration of H₂O₂ can induce protection against subsequent severe oxidative or abiotic stress. Such cross-priming (priming by one type of stressor enhancing plant response to other types of stressors) likely occurs in Arabidopsis treated with UV/PAA. In the present study, upon Pst DC3000 challenge, UV/PAA-treated Arabidopsis exhibited increased H₂O₂ and callose accumulation at the onset of infection. Therefore, H₂O₂ acted as a protective signal triggering stress adaptation and innate immunity (Petrov and Van Breusegem 2012), and priming processes were initiated by the pre-exposure of Arabidopsis to H₂O₂ (Hossain et al. 2015). Overall, at micromolar concentrations, H₂O₂ induced a stress response without causing severe damage to plants in our experiment.

In conclusion, combined treatment with UV irradiation and PAA might be a highly efficient AOP for inactivating microbes in water and on various surfaces in agricultural industry. Our findings suggest that prolonged treatment with a low concentration of UV/PAA enhances the innate immunity of plants against various secondary pathogen attacks through cultivation platforms, potentially through the stimulation of priming processes. Considering the broad effects of priming on stress adaptation, UV/PAA treatment may improve plant growth under various abiotic and biotic stresses. Furthermore, investigation of the effects of UV/PAA on plant growth in presence of various environmental pollutants would be interesting.

This work was supported by National Research Foundation of Korea (NRF-2020R1I1A1A01069595) and the Cooperative Research Program for Agricultural Science and Technology Development (project numbers RS-2023-00230820) and the Rural Development Administration, the Republic of Korea. The authors declare that they have no conflicts of interest.