Key message

  • Cold plasma treatment alters water properties, increasing H2O2, NO2−, and NO3− concentrations.

  • Plasma activated water (PAW) irrigation boosts leaf composition, reflectance, plant size, and trichome densities.

  • PAW-irrigated plants reduce settling and oviposition avoidance of two-spotted spider mites.

  • PAW-irrigated plants reduce two-spotted spider mite populations.

  • PAW should be considered an innovative and sustainable component in 21st century pest management.

Introduction

Ensuring sustainable crop productivity and global food security is one of the defining challenges of the new millennium (Farooq et al. 2019). Moreover, it is crucial to examine and promote technologies that can boost food production without causing adverse risks to surrounding ecosystems and urban environments. In this context, there has been growing interest in use of “cold plasma technologies” in crop production (Xu et al. 2023; Guo et al. 2021; Yepez et al. 2022).

Cold plasma, also known as non-thermal plasma, is typically generated by applying a discharge (thermal, electromagnetic, or electric fields, microwave, and radio frequencies) to gases (air, Ar, O2, He, or N2) at low or atmospheric pressure (Guo et al. 2021). Cold plasma treatments produce a dynamic mixture of ions, new electrons, excited molecules, and free radicals (Guo et al. 2021; Thirumdas et al. 2015; Pankaj and Keener 2017). When applied to water, active compounds, like reactive oxygen and nitrogen species (RONS) (H2O2, NO2−, or NO3−), are generated (Gao et al. 2022; Han et al. 2023). Reactive species generated during the plasma activation process increase water salinity, electrical conductivity, and acidity (Chew et al. 2023). PAW has been shown to enhance seed germination, root and shoot growth, photosynthesis, nutrient availability, floral regulation, and yield (Fan et al. 2020; Vichiansan et al. 2023; Priatama et al. 2023; Sivachandiran and Khacef 2017; Škarpa et al. 2020; Guragain et al. 2021). It is also a disinfectant with antimicrobial properties against bacteria (Perez et al. 2019; Lee et al. 2023), fungi (Ambrico et al. 2020; Ju et al. 2023), and viruses (Filipić et al. 2019; Hanbal et al. 2018). Additionally, PAW has been demonstrated to possess significant insecticidal activity against citrus mealybug Planococcus citri (Risso) (Hemiptera: Pseudococcidae) (Ten Bosch et al. 2017).

The above-mentioned studies highlight the potential of PAW and its direct effects on pests (pathogens and insects). Less research attention has been given to the potential of PAW as a booster of plant defense mechanisms (Laurita et al. 2021; Zambon et al. 2020; Adhikari et al. 2019). This topic is here referred to as the “indirect effects” of PAW—how treatments (either foliar applications or via means of supplemental irrigation) may elicit significant reduction in host plant suitability to herbivorous pests. As part of resistance mechanisms against arthropod pests, plant trichomes and their chemical compounds play a significant role in crop protection (Simmons and Gurr 2005; Savi et al. 2022). Tomato (Solanum lycopersicum L.) plants, for instance, possess seven types of trichomes, which are categorized as non-glandular (types II, III, and V) and glandular (types I, IV, VI, and VII) (Simmons and Gurr 2005). Non-glandular trichomes can serve as physical barriers to micro-arthropods on the leaf surface while glandular trichomes act as physical and deterrent barriers to small herbivores, through their high viscosity of allelochemical secretions such as acyl sugars, methyl-ketones, and sesquiterpenes (Rakha et al. 2017; Savi 2019; Yang et al. 2023). Trichomes, when present in high densities, can suppress pest movement, or negatively impact pest survival, development, and reproduction (Savi 2019; Savi et al. 2022; Salazar-Mendoza et al. 2023). Furthermore, nutritional composition of plants may be positively correlated with suitability to herbivorous pests (Chaboussou 2004; Joern et al. 2012). If so, it may be possible to indirectly manage and suppress pest populations by manipulating element composition of crops (Altieri and Nicholls 2003; Huber et al. 2012). However, to our knowledge, it has not yet been explored to what extent PAW may be used to alter plant element composition or boost defensive traits, such as trichome density, in ways that would reduce suitability of crops to herbivorous pests.

Use of trichome density and element composition as response variables in studies of indirect plant responses to PAW, and other treatments, requires destructive sampling of crop leaves. Additionally, these response variables require processing of leaf materials and are hampered by their general “representativeness.” Meaning, trichome density can only feasibly be counted on small pieces of leaves from individual plants, so it is possible that within-leaf variation and vertical variation among leaves within canopies affect counts. Regarding element composition, costs of analyses may hamper inclusion of individual samples from each plant from all combinations of treatments and replications, so there may be statistical constraints due to few observations (degrees of freedom). Furthermore, selection of leaves to be included in each sample may also pose some concerns about subjectivity and representativeness of leaf materials being analyzed. As an attempt to address such challenges and to explore possible complementarity and highly innovative technologies, optical sensing of plants may be considered. As reviewed elsewhere (Li et al. 2017; Nansen 2016), when lighting conditions, projection angle, distance between camera and objects are quasi-constant over time and space, then reflectance (or transmission) features acquired from plant leaves may be used to quantify differences in qualitative traits among objected subjected in different categories (i.e., subjected to different treatments). Furthermore, suitability of a given host plant may be associated with its element composition, which is detectable based on features in leaf profiles (Nansen et al. 2021). Accordingly, leaf reflectance profiling may be proposed as an approach to, non-destructively, detect and diagnose effects of PAW treatments and provide valuable and quantitative insight into crop plant responses to PAW, other growth-promoting treatments.

As illustrated in Fig. 1, we examined the hypotheses that: (1) trichome density and element composition of tomato plants are significantly affected by supplementary PAW irrigation, (2) PAW irrigations elicit significant changes to leaf reflectance profiles, (3) based on preference bioassays, arthropod herbivores will show avoidance of PAW-irrigated plants, and (4) based on performance bioassays, arthropod pest populations will grow significantly slower on PAW-irrigated plants. To address these hypotheses, we used two-spotted spider mites, Tetranychus urticae Koch (Acari Tetranychidae) as a model herbivore. Two-spotted spider mite is an important herbivore that poses considerable damage to many plants including tomato due to their short life cycle and high reproductive capacity (Tiftikçi et al. 2020; Savi et al. 2021). Control of this important pest is widely based on application of synthetic miticides, which may lead to environmental contamination and evolution of resistance (Attia et al. 2013). Resistance development of this herbivore to 96 active ingredients has been reported (APRD, http://www.pesticideresistance.org/). Due to these issues, alternative control measures have been sought globally, including the use of predatory mites, such as Phytoseiulus persimilis Athias-Henriot (Acari: Phytoseiidae) (McMurtry 2013). However, implementation of biological control programs has not been widely adopted by commercial tomato growers due to inconsistent performance and need for extensive pest population monitoring (Weinblum et al. 2021). Therefore, it has been recommended that growers use a combination of soft miticides, and biological control methods, or mite-resistant plant genotypes (Tiftikçi et al. 2020, 2022; Lucini et al. 2015; Savi et al. 2022). Although these practices are somewhat effective, they may not always provide sufficient control over two-spotted spider mite populations, particularly under high-pressure conditions. Hence, additional management tactics are needed to complement these practices and enhance their effectiveness. Applications of supplementary irrigation with PAW are likely highly compatible with other IPM tactics and should be considered an innovative and sustainable component in twenty-first-century pest management.

Fig. 1
figure 1

Schematic representation of the experimental design to evaluate the effect of plasma-activated water

Full size image

Material and methods

Two-spotted spider mites

To establish two-spotted spider mite colony used in this study, some individuals were collected from soybean (Glycine max L.) at the University of California, UC Davis, USA. These mites were then reared on tomato plants (variety Micro-tom) for several generations in a greenhouse maintained at the following conditions: 25.2 ± 1 °C, 77 ± 10% RH, and a L:D 12:12 photoperiod.

Production and characterization of PAW

We evaluated two PAWs, in which UC Davis tap water (control) was treated for 6.0 min (PAW 6.0) and 9.4 min (PAW 9.4); PAW 6.0 and PAW 9.4 were specifically chosen for this study based on their observed higher responsiveness to plant growth in preliminary studies involving a broader range of PAWs. PAWs were generated using the Openair™ Plasma System from Plasmatreat USA, Inc., which included TS CD50 treatment station, FG5001 plasma generator equipped with 1KVA digital processor and 10,145 nozzle, CD50 plasma disinfection jet, EPU2 hydrogen peroxide evaporator system, DVE10 pressure supply control unit, HTR11 plasma high-voltage transformer, PTU2 high-voltage cabinet for two transformers and accessories, FLC air/O2 and flow control. We operated the system under the following parameters: voltage (302 V), air pressure (38–42 mBar), frequency (23 kHz), and current (12.6A). Each treatment consisted of 1 L of water in a 2.5-L Pyrex glass beaker (diameter 131 mm) with a stir bar maintained on a hot magnetic stirrer device operating at 600 rpm for continuous agitation during plasma treatment. To prevent water spillage during plasma treatment, a corrosion-resistant 316 stainless steel sheet with a central 10-cm-diameter opening, fitted for the plasma nozzle inlet, was placed atop the beaker. A consistent distance of 1 cm was maintained between the water surface and plasma nozzle inside the beaker. To ensure consistent performance throughout the treatment process, an air compressor was regularly monitored and maintained at 85 psi. 24 h after treatment, we measured the physicochemical properties of PAW 6.0, PAW 9.4, and control. Levels of pH and electrical conductivity (EC; indicative of active ions in water) were monitored using pH/EC meter (OHAUS-AB331M-F) while oxidation–reduction potential (ORP; ability of solution to gain or lose electrons) was estimated through ORP meter (Hanna HI2202-01). Concentrations of reactive species, such as nitrate (NO3−) and nitrite (NO2−) as well as hydrogen peroxide (H2O2), were obtained using a benchtop multiparameter photometer (HI83399-01). Physicochemical parameters measurements were collected based on a completely randomized design with eight replicates (different times that the PAW was prepared) per treatment. PAW samples were stored in glass containers in a refrigerator (5 °C) for a maximum of four days before application to tomato plants.

Plant growing conditions

Tomato plants variety BQ273 were grown in the greenhouse facilities at UC Davis. Initially, seeds were sown in tray cells filled with a homogeneous mixture of pumice, sphagnum peat moss, sand, redwood sawdust, and dolomite, at a ratio of 5.23 kg per m3 of soil mix, which had been sterilized in an autoclave at 121 °C for 1 h. Trays were then placed in a greenhouse, and automated sprinkling irrigation using a Mix Rite injector (Model 2502) provided water and fertilizer four times a day (at 7 am, 10 am, 2 pm, and 5 pm). The fertilizer was prepared using UC Davis modified Hoagland's solution with the following element composition (values in ppm): N = 150, P = 50, K = 200, Ca = 175, Mg = 55, S = 120, Fe = 2.5, Cu = 0.02, B = 0.5, Mn = 0.5, Mo = 0.01, and Zn = 0.05. Four weeks after sowing, seedlings were individually transplanted to 1.5-L pots 80% filled with a homogeneous mixture of soil, sand, and tanned bovine manure (1:1:1) autoclaved at 121 °C for 1 h. Thirty pots were randomly divided into three treatment groups (PAW 6.0, PAW 9.4, and control). Pots were maintained in a greenhouse under the following conditions: 25.2 ± 1 °C, 77 ± 10% RH, and a L:D 12:12 photoperiod. A 35 ml water and fertilizer solution (prepared at rates previously mentioned) was supplied twice a day (at 7 am and 6 pm) to the pots through automated drip irrigation. Additionally, 50 ml of PAW 6.0, PAW 9.4, or tap water was applied once a day, between 11 am and 12 noon, to individual plants until the end of the experiments. Thus, PAW treatments represented 42% of total water supplied to each tomato plant.

Element composition and trichome densities

After two weeks of treatments, leaf samples were collected from the median third of the plant canopy for analysis of elemental composition (N, P, K, Ca, Mg, S, Fe, Cu, B, Mn, Zn and Na). Each sample consisted of 0.5 g of dried leaflet material obtained from two plants, for a total of four samples per treatment. Element composition analyses were performed by University of California Davis analytical laboratory, and standard methods for analyses can be found at https://anlab.ucdavis.edu. Regarding trichome assessment (Fig. 1), five leaflets excised from the median third of the canopy of each plant amounting to 50 leaflets per treatment were examined. All glandular and non-glandular trichomes were counted in an area of 1 mm2 on the abaxial and adaxial sides close to the middle portion of the main vein. We used 10× magnification microscope (SZ61; Olympus SZ61 Trinocular Stereo Zoom Microscope, 8x-40x, SZ2-LGB Illuminator Camera), equipped with a digital camera (Olympus DP27) connected to a computer software, Cell Sens Entry.

Hyperspectral imaging

After two weeks of supplementary PAW irrigation, hyperspectral reflectance data were collected from all tomato plants (Fig. 1). We utilized a push-broom hyperspectral camera (PIKA L; www.resonon.com) mounted on a custom-built robotic rail system positioned approximately 1.5 m above the plant canopy, achieving a spatial resolution of approximately 9 pixels mm−2 (Nguyen et al. 2020; Nansen et al. 2021). The hyperspectral camera acquires data in 150 spectral bands ranging from 380 to 1015 nm, with a spectral resolution of 4.2 nm. Light source consisted of 12 and 15 W, 12 V halogen light bulbs placed on either side of the lens. White calibration was performed using a piece of white Teflon; both dark and white calibrations were carried out immediately before each imaging event to obtain relative reflectance. Hyperspectral imaging data were collected around 2 pm inside a dark room, maintained at 20–23 °C and 50–75% RH. Number of green pixels (after radiometric filtering to omit background pixels) per tomato plant was determined and used as an indicator of plant size.

Preference bioassays

Attractiveness and oviposition preferences of two-spotted spider mites were assessed through two-choice bioassays. The experimental unit involved placing two 30-mm leaf discs (each from a different treatment), with their abaxial surface facing up and made from leaves excised from the middle third of the canopy of 35-day-old tomato plants (irrigated with treatments for two weeks), inside a Petri dish (10 cm in diameter and 1.5 cm high). The Petri dish contained a 1-cm-thick foam mat lined with hydrophilic cotton soaked in deionized water (Fig. 1). Leaf discs were positioned in close proximity and connected by a plastic piece to allow the movement of mites from one to the other. Ten 6-day-old adult females were introduced into the center of the plastic piece. The experimental units were maintained under controlled conditions in a growth chamber with the following settings: 25.2 ± 1 °C, 77 ± 10% RH, and a L:D 12:12 photoperiod. The numbers of two-spotted spider mite females and their respective eggs on each leaf disc were recorded after 24 h. The experimental design was fully randomized, with 20 replicates for each of the three paired combinations (control–PAW 6.0, control–PAW 9.4, and PAW 6.0–PAW 9.4).

Performance bioassays

After 20 days of treatments, all tomato plants were infested with two-spotted spider mites (Fig. 1), placing 15 adult females on each of the second and fourth leaves (counted from the bottom). Population growth of two-spotted spider mites was monitored on all leaves on each plant at weeks 1, 2, and 3, by counting the numbers of mobile stages (larvae, nymphs–pooling protonymphs and deutonymphs–and adults) with a magnifying glass 10x. Upon completion of the last evaluation, all leaves of each plant were collected to count the number of eggs on each of them, under a stereomicroscope.

Data analyses

All data processing, analyses, and classifications were performed in R v3.6.1 (The R Foundation for Statistical Computing, Vienna, Austria). Leaf compositions of N, K, Ca, Mg, Zn and Mn followed Shapiro assumptions of normality (library ‘dplyr’) and were therefore subjected to Tukey HSD analysis of variance. Conversely, leaf compositions of P, S, B, Fe, and Cu and physicochemical properties did not meet homoscedasticity and normality assumptions, even after transformations; accordingly, these variables were analyzed based on nonparametric Kruskal–Wallis’ comparison of averages of treatments (library ‘AgroR’). Similarly, average trichome densities in response to treatments did not meet homoscedasticity and normality assumptions and were therefore analyzed using a Generalized Linear Model (GLM) with negative binomial distribution (library ‘AgroR’). In two-choice and performance bioassays, counts of adult female spider mites and eggs were analyzed using GLM with binomial error distribution. In case of significant differences between treatments, multiple comparisons for GLM models were performed using “emmeans” function in the “multcomp” library and Šidák-adjusted confidence intervals. Furthermore, principal component analysis (library ‘devtools’) was performed with element composition, trichome densities, and bioassay data as explanatory variables.

Tukey analysis of variance was used to examine average numbers of pixels per tomato plant among the three treatments. To compare leaf reflectance among treatments, random subsamples of 1000 green pixels were drawn from each tomato plant (total of 30,000 pixels). Leaf reflectance among treatments was classified based on support vector machine (svm) (library ‘e1071’) with radial kernel function and no specific hyperparameters (i.e., cost or gamma). Kappa values were generated for the assessment of classification performance and interpreted as follows: 0 = poor, 0.01–0.20 = slight, 0.21–0.40 = fair, 0.41–0.60 = moderate, 0.61–0.80 = substantial, and 0.81–1.00 = almost perfect (Landis and Koch 1977). In addition, we included tenfold cross-validation (Fauvel et al. 2015; Huang et al. 2019; Jin et al. 2009). In k-fold cross-validation, training data sets are divided into 'k' equal portions, which in this study was set at 10 [so 1500 observations (pixels) in each portion]. The classification model is trained on 'k − 1' of these portions, while the remaining portion is used for validation. This process is repeated 'k' times, with each fold serving as validation, and results from 'k' tests are averaged to produce a single estimation of model performance.

Results

Physicochemical properties of the water

pH of PAW 6.0 and PAW 9.4 elicited significant reductions (χ2 = 23.26, df = 2, p < 0.001) compared to the control (Fig. 2a). Electrical conductivity (EC) increased significantly (χ2 = 23.14, df = 2, p < 0.001) with significantly higher values for PAW 9.4 intermediate for that PAW 6.0 and lower in control. ORP followed the same trend (χ2 = 23.16, df = 2, p < 0.001), being also significantly higher for PAW 9.4 (563.2), intermediate for PAW 6.0, and lower for control (Fig. 2b). Concentration of H2O2 in both PAW 6.0 and PAW 9.4 increased significantly (χ2 = 12.3, df = 2, p = 0.002), compared to control (0.41). Concentrations of NO2−(χ2 = 23.67, df = 2, p < 0.001) and NO3−(χ2 = 23.15, df = 2, p < 0.001) increased substantially, with higher values for PAW 9.4, intermediate for that of PAW 6.0 and lower for control (Fig. 2c).

Fig. 2
figure 2

a pH, conductivity, b oxidation–reduction potential (ORP), hydrogen peroxide concentration (H2O2), c concentrations of nitrate (NO2−) and nitrite (NO3−) in untreated tap water (control) and 6.0 (PAW 6.0) and 9.4 min (PAW 9.4) plasma-activated water. Lines followed by different letters are significantly different (Kruskal–Wallis's test, p < 0.05)

Full size image

Element composition and trichome densities

Compared to control, both PAW 6.0 and PAW 9.4 irrigations significantly increased the accumulation of N, P, K S, Ca, and Mg in tomato leaves, but decreased Zn and Fe content (Table 1). While PAW 6.0 increased B accumulation, PAW 9.4 treatment reduced it compared to control. No significant differences were observed between control and both PAW 6.0 and PAW 9.4 irrigations for Mn and Cu.

Table 1 Element composition of leaf samples from untreated and irrigated PAW tomato plants
Full size table

PAW-irrigated plants showed a significant increase in the numbers of non-glandular and glandular trichomes compared to control plants. This increase was observed when considering the abaxial and adaxial leaf surfaces together or separately, except for non-glandular trichomes on the adaxial surface (Table 2).

Table 2 Means (± SE) density of non-glandular and glandular trichomes (no. trichomes/mm2) on both abaxial and adaxial surfaces of leaves from untreated and irrigated PAW tomato plants
Full size table

Hyperspectral imaging

Average size of tomato plants (based on green pixels) varied significantly among all three treatments, with PAW 6.0 plant being the largest (F = 42.36, df = 2, 27, p < 0.001) (Fig. 3a and b). Thus, it appeared that especially PAW 6.0 positively affected overall plant growth. Based on k = 10 validation, 30,000 pixels (10,000 from each treatment) were classified with 61% accuracy. In addition, the svm classification was associated with a kappa value = 0.57, which is considered “moderate” accuracy (Landis and Koch 1977). In the interpretation of the confusion matrix of classified pixels (Table 3), control pixels were classified with only 9% mis-classification, suggesting a marked difference between control plants and plants receiving supplementary PAW irrigation. The same analysis also showed that only 10–14% of pixels from PAW 6.0 and PAW 9.4 were mis-classified as control pixels. Thus, classification of hyperspectral imaging data provided strong support for the claim that this technology can be used to assess differences between PAW-irrigated plants. The difference between treatments was greater in spectral bands around 445 nm and between 700 and 1105 nm wavelength, as indicated by the separation of the three lines in Fig. 4a, and the height of the PAW lines in relation to the fixed control line (Fig. 4b).

Fig. 3
figure 3

Representative images of plants grown under different treatments a control (untreated tap water), PAW 6.0, and PAW 9.4. The two PAWs corresponded to plasma activation of water for 6.0 (PAW 6.0) and 9.4 min (PAW 9.4). Average (± SE) number of pixels per tomato plant from the three treatment, which is a relative indicator of canopy size

Full size image
Table 3 Confusion matrices of support vector machine (svm) classification of leaf reflectance from untreated and irrigated PAW tomato plants
Full size table
Fig. 4
figure 4

a Leaf reflectance data from plant grown under grown under different treatments control (untreated tap water), PAW 6.0, and PAW 9.4. The two PAWs corresponded to plasma activation of water for 6.0 (PAW 6.0) and 9.4 min (PAW 9.4). b Relative difference of PAW treatments compared to control (PAW/Control) with red line denoting no difference = 1

Full size image

Preference bioassay

Out of the 10 females tested on each leaf disc (n = 20) in each two-choice test, 83.5%, 84.5%, and 95.5% chose one of the two options: control vs. PAW 6.0, control vs. PAW 9.4, and PAW 6.0 vs. PAW 9.4, respectively (Fig. 5a). The number of settled mites was significantly lower on leaf discs from both PAW 6.0 (χ2 = 18.38, df = 1, p < 0.001)- and PAW 9.4 (χ2 = 14.602, df = 1, p < 0.001)-irrigated plants compared to those of the control. No significant difference was observed for the number of settled mites in test involving leaf discs from PAW 6.0- and PAW 9.4-irrigated plants (χ2 = 0.09, df = 1, p = 0.75). Similarly, there was no significant difference in the per capita daily oviposition rate between leaf discs in tests involving PAW 6.0 vs. PAW 9.4 (χ2 = 0.24, df = 1, p = 0.62), or control vs PAW 9.4 (χ2 = 1.64, df = 1, p = 0.19), but it was significantly lower on PAW 6.0 leaf discs in test involving PAW 6.0 vs control (χ2 = 10.32, df = 1, p = 0.001) (Fig. 5b).

Fig. 5
figure 5

Preference bioassays results. Average (± SE) percentages of females of two-spotted spider mites (a) and per capita daily oviposition rate (b) in leaf discs from both PAW-irrigated and untreated tomato plants through dual-choice tests 24 h after release in dual-choice leaf disc assay (initial number of females = 10; number of replicates = 20), contrasting control versus PAW 6.0, control versus PAW 9.4, and PAW 6.0 versus PAW 9.4. Asterisk (*) indicates significant preference for one of the options (GLM with two side binomial tests: p < 0.05; df = 1)

Full size image

Performance bioassay

Significant effects of PAW-irrigated plants or control (χ2 = 138.821, df = 2; p < 0.001), time (χ2 = 171.69, df = 3; p < 0.001) and their interaction (χ2 = 36.38, df = 6; p < 0.001) were observed on population growth of mobile two-spotted spider mites (Fig. 6a). One week after infestation, PAW 6.0- or PAW 9.4-irrigated plants showed a significant delay in population growth of mobile spider mites compared to control. Population growth on PAW 6.0, PAW 9.4, or control treatments 2 weeks after infestation followed the same trend reported for the previous week. After three weeks of the infestation, a similar pattern was observed for the number of mobile mites, also lower on tomato plants grown under PAW 6.0, and PAW 9.4 compared with control. In the destructive measurement evaluation, number of eggs was significantly (χ2 = 29.79, df = 2; p < 0.001) lower on PAW 6.0- and PAW 9.4-irrigated plants than in the control plants (Fig. 6b).

Fig. 6
figure 6

Performance results. Average number (± SE) of a mobile two-spotted spider mites (TSSM) at weeks 0, 1, 2, and 3 after mite infestation and b eggs recorded in destructive measurements (3-week after infestation) on plants growth under control or PAW treatments (initial females infested/ plant = 30, number of replicates = 10)

Full size image

Principal component analysis

Principal component analysis was deployed to element composition, trichome, and bioassay data to elucidate trends (Fig. 7). Except for B and PAW 6.0 (which were highly correlated with PCA 2), all variables were positively or negatively correlated with PCA 1. That is, Cu, Fe, Zn, Mn, and all two-spotted spider mite parameters from preference and performance bioassays (preference by adults, oviposition, and population density) were positioned along the positive side of PCA 1 and highest in control. Conversely, Mg, N, P, K, S, and trichome densities were positioned along the negative side of PCA 1 and highest in PAW 6.0 and PAW 9.4. As PCA 1 explained more than seven times (88.04%) the variance explained by PCA 2 (11.96%), this principal component analysis suggested that most of the variance was caused by the difference between control, PAW 6.0, and PAW 9.4.

Fig. 7
figure 7

Principal component analysis (PCA) of element composition (Cu, Fe, K, Mg, Mn, S, and Zn), trichome densities (Gland T—glandular trichomes, and Non-gland. T—non-glandular trichomes), and bioassays data (Set avoid—avoidance of settling by two-spotted spider mites, Ovip avoid—avoidance of oviposition by two-spotted spider mites, Ep. Grow—egg population growth, MMPop Grow—mobile spider mite population growth) from tomato grown under control, PAW 6.0, and PAW 9.4 treatments

Full size image

Discussion

In this study, we examined physicochemical properties of PAW 6.0 and PAW 9.4, and their indirect effects on the suitability of tomato plants as host of two-spotted spider mites. Our results showed significant differences in physicochemical parameters between PAW 6.0 and PAW 9.4 compared to control. We demonstrated that supplementary PAW irrigation elicited significant increase in N, P, K S, Ca, and Mg leaf retention, plant size (based on green pixels), leaf reflectance profiles, and trichome densities (except non-glandular trichomes on the adaxial surface). Additionally, preference bioassays revealed significant avoidance of two-spotted spider mites, as evidenced by reduced settling and oviposition on leaf discs from PAW-irrigated plants compared to control plants. Performance bioassays over three weeks showed significantly lower two-spotted spider mite populations on PAW-irrigated plants. Thus, results corroborated our study hypothesis that indirect effects of PAW might reduce host plant suitability against arthropod pest attack. These results provide novel insights into the versatile potential of PAW as a tool to manipulate plant resistance as part of IPM strategies.

PAW physicochemical properties

Most of physicochemical properties data from PAW 6.0 and PAW 9.4 in our study are comparable to those obtained by other studies using atmospheric plasma jet technology. For instance, as reported by Ma et al. (2015), Shen et al. (2016), Sivachandiran and Khacef (2017), Liao et al. (2018), and Guragain et al. (2023), pH of tap water decreased from 6.0–7.19 to 2.3–4.69 after 5–30 min of plasma treatment, as observed in the present study. Previous studies have shown that water acidification is the result of formation and accumulation of active ions in the solution, specifically nitrite (NO2–), nitrate (NO3–) ions, and H2O2 (Dunand et al. 2007; Ma et al. 2015). It was determined in those studies that NO2–, NO3–, and H2O2, initially absent in untreated water, increased, respectively, to the ranges of 0.49–0.86, 1.49–75.0, and 2.15–70.0 mg/L, after 5–20 min of plasma treatment (Sivachandiran and Khacef 2017; Liao et al. 2018; Stoleru et al. 2020; Than et al. 2022; Guragain et al. 2021), which also aligns with our findings. However, values of NO2–, NO3–, and H2O2 obtained in our studies after plasma treatment are much higher than found in above-mentioned studies. These differences may be attributed to the plasma reactor design, treatment time, voltage, and feed gas used for plasma generation (Ma et al. 2015; Zhou et al. 2018).

Properties of PAW-irrigated tomato plants

Physicochemical properties of PAW are recognized for their substantial impact on plant properties (Guo et al. 2021), and alterations in leaf reflectance profiles may indicate such changes. In our study, we observed a significant increase in relative leaf reflectance at 445 and 700–1015 nm, which correspond to the blue and near-infrared spectrum, respectively. These specific spectral bands are known for their strong absorption properties by chlorophyll and leaf pigmentation (Sims and Gamon 2002). This suggests that plants irrigated with water whose physicochemical properties have been altered by plasma discharge may improve the uptake and utilization of essential nutrients necessary for efficient photosynthetic processes in plants (Kučerová et al. 2021). Additionally, previous studies have indicated that changes in plant leaf reflectance within specific wavelength bands, such as 775–850 nm and 910–960 nm, can be indicative of nitrogen levels in plants (Sishodia et al. 2020; Rubo and Zinkernagel 2022; Rubio-Delgado et al. 2021). Given that PAW contains reactive nitrogen species (RNS), changes in leaf reflectance due to PAW treatments may be attributed to nitrogen assimilation. This finding is supported by our element composition analysis, which showed an enhancement in leaf nitrogen content in response to PAW treatments. Moreover, our results demonstrated an improvement in the levels of other essential elements such as P, S, Ca, and Mg. These positive effects of supplementary PAW irrigation on nutrient uptake and maintenance of a balanced nutrient profile in plants can be attributed to the beneficial role of hydrogen peroxide (H2O2), a reactive oxygen species found in PAW, which interacts with RNS. This aligns with the findings of Nurnaeimah et al. (2020), who observed that exogenous application of 8-mM H2O2 not only significantly improved nitrogen mineral uptake, but also the uptake of other elements including arsenic, iron, calcium, and potassium. Wang et al. (2023) also reported that PAW supplementation enhanced accumulation of both macro-elements (Ca, P, Na, K, S, and Mg) and microelements (Mn, Fe, and Zn) in fresh wheat (Triticum aestivum L.). A truly innovative and important aspect of this study is how we were able to associate leaf reflectance features with element composition, and our results were corroborated by the existing body of literature. Non-glandular and glandular trichomes densities have been extensively documented as the primary mechanism by which plants defend themselves against arthropod pests (Savi et al. 2022; de Lima Filho et al. 2022; Lucini et al. 2015; Koller et al. 2007). In our study, we observed an increase in the density of both non-glandular and glandular trichomes on PAW-irrigated plants. This is the first report showing that the plants irrigated with PAW resulted in a significant increase in both non-glandular and glandular trichomes density on tomato leaves. These findings are promising and could potentially enhance plant resistance, thus contributing to efforts to reduce reliance on synthetic pesticides and promote sustainable pest management practices.

Paw-irrigated plants responses to two-spotted spider mites

Leaf discs from both 6- and 9.4-min PAW-irrigated plants revealed a significant avoidance of settling by two-spotted spider mites indicating a potential deterrent effect of PAW supplementation. Additionally, significant avoidance of oviposition by two-spotted spider mites was observed on leaf discs from 6-min PAW-irrigated plants suggesting that plants irrigated with a specific PAW regime may provide more robust deterrent effect against arthropod pests. On the other hand, development and reproduction of two-spotted spider mites were substantially delayed on both 6- and 9.4-min PAW-irrigated plants as evidenced by their lower population growth on those plants. These findings support the evidence that PAW is potentially useful as a tool to enhance antixenosis and antibiosis resistance mechanisms in plants against arthropod pests (Savi et al. 2021, 2022), with PAW 6.0 appearing to provide better protection to tomato plants than PAW 9.4. The observed deterrent effects of PAW on two-spotted spider mites in our study may be attributed to the increased trichome densities, particularly glandular trichomes resulting from PAW irrigation. These trichomes produce a range of secondary metabolites including methyl-ketones (notably 2-tridecanone), sesquiterpene, zingiberene, and acyl sugars, which are known have toxic, repellent, and/or anti-nutritive effects on spider mites (Savi et al. 2021, 2022; Lucini et al. 2015; Oliveira et al. 2020; Rioja et al. 2017). However, our study did not assess the differences in secondary metabolites on PAW-irrigated tomato plants in the present study. Further research is required to identify the specific compounds associated with the trichomes found in PAW-irrigated plants that provide enhanced defense against two-spotted spider mite attacks.

Additionally, exogenous application of H2O2 has been shown to enhance common bean (Phaseolus vulgaris L.) plant resistance against two-spotted spider mites (Alakhdar and Shoala 2021). H2O2 acts as a messenger molecule, triggering the activation of OXI1 kinases, induction of callose production, triggering of MAPK cascades, and promotion of the transcription of defense-related genes (Neill et al. 2002; Rentel et al. 2004). Therefore, the presence of H2O2 in PAW supplemented to the plants may have also contributed to the significant deterrent effect on two-spotted spider mites.

Previous studies have also reported that H2O2 in PAW stimulates the production of phytohormones, such as salicylic acid and jasmonic acid, which are essential for activating plant immune responses (Laurita et al. 2021; Adhikari et al. 2019). Additionally, PAW has been reported to increase the activity of antioxidant enzymes such as superoxide dismutase, peroxidase, and catalase (Ramazzina et al. 2022; Kučerová et al. 2019, 2021), which might neutralize harmful reactive oxygen species produced during pest attacks. However, at excessive concentration, H2O2 can induce oxidative stress in plants, potentially leading to suboptimal responses to abiotic and biotic stresses (Alakhdar and Shoala 2021). This could explain the low deterrent effect on two-spotted spider mite oviposition observed on plants irrigated with PAW 9.4, which contains a high level of H2O2. Therefore, PAW regime must carefully consider H2O2 levels and other properties to avoid inducing stress and compromising plant resilience. Moreover, other factors, such as the nutritional balance resulting from PAW irrigation in our study, might have contributed to the resistance of tomato plants against two-spotted spider mites, although consensus on the relationship between plant nutritional state and defense mechanisms is still elusive (Singh and Sood 2017; Chaboussou 2004; Altieri and Nicholls 2003). It is essential to consider environmental factors such as temperature, humidity, soil conditions, and plant genotype which might affect PAW-irrigated plants and two-spotted spider mite interactions.

This study provides for the first time evidence that PAW could serve be a valuable tool for sustainable pest management. In addition to its established roles in seed germination, plant growth, yield enhancement, and plant pathogen control, PAW was found to improve plant nutrient uptake, increase leaf reflectance, and increase trichome densities (a key plant defense mechanism against arthropod pests). As a result, it reduced the preference and performance of two-spotted spider mites. These findings are encouraging and suggest the need for further investigation to determine the practical use of PAW in pest management. This would involve assessing its effectiveness on different plant crops, varieties, and pest species, as well as examining potential synergistic effects with other pest management strategies. It is also important to evaluate the benefits of repeated administration of PAW throughout the crop cycle under various environmental conditions and to consider the quality of the plant products and any possible side effects. This comprehensive assessment will provide valuable insights into the broader implications and limitations of PAW for sustainable pest management practices.