Benzoate, Gallate, And Salicylate Regulate Redox-Enzymes, Ultrastructure, And Accumulation of Boron To Counteract Boron Toxicity On Tomato Callus Cells

The participation of benzoate (BA), gallate (GA), and salicylate (SA) in various biochemical and physiological processes in plants under conditions of excessive boron (EB) is largely unknown to date. Here, the relationships between phenolic acids (PAs) and the regulation of redox-enzymes, the ultrastructure of cells, and boron forms in the mitigation of EB-induced oxidative stress within tomato calli were studied. Tomato calli were exposed to 2 mM boron (B) in the presence or absence of three concentrations of benzoate, gallate, and salicylate. The data showed that different concentrations of PA counteracted the inhibition of growth and oxidative stress of EB stress by reducing hydrogen peroxide (H 2 O 2 ) production, lipoxygenase (LOX) activity, boron accumulation forms, cell wall thickening, and moderate concentrations were the most effective. Applications of PAs reduced the catalytic impacts of EB on superoxide dismutase (SOD) and catalase (CAT) activity. Likewise, benzoate and gallate increased the inuences of EB stimulation on peroxidase (POD) and ascorbate peroxidase (APX) activities; whereas, SA reduced these effects on both enzymes. PA treatments enhanced the insignicant catalytic effect of EB on the activity of phenylalanine ammonia-lyase (PAL), as well as the stimulation of the negative inuence of EB on polyphenol oxidase (PPO) activity. The ndings highlight that PAs play an important role in alleviating EB stress in tomato plants by regulating redox enzymes, B-accumulation forms, and cell wall thickening. This study provides new perspectives for strategies related to excess boron tolerance in tomato plants and thus can be used as plant growth promoters.


Introduction
Boron is known as a micronutrient and is essential in relation to vascular plants, which  Plants have a complex defensive response strategy, which can appear either primarily or after a stress challenge. The plant's constitutive defense mechanism, after stress recognition, triggers a complex defense signaling chain that varies from one stress to another (Rejeb et al. 2014). Plants constitutively cope with oxidative stress through internal defense strategies consisting of various enzymatic antioxidants such as superoxide dismutase (SOD), catalase (CAT), ascorbate peroxidase (APX), peroxidase (POD), and non-enzymatic antioxidants (Hasanuzzaman et al. 2020). Under EB, Kaya et al. (2020) reported an increase in CAT, SOD, and POD activity within tomato plants. However, other studies have demonstrated that APX activity in lettuce (Eraslan et al. 2007) and CAT in citrus leaves (Han et al. 2009) decreased or was insensitive in tomatoes to some EB doses (Cervilla et al. 2007). In plants, phenolics are also important secondary metabolites that serve as effective antioxidants (Hossain et al. 2009). Phenylalanine ammonia-lyase (PAL) is the main enzyme in phenolate synthesis, and EB has been shown to enhance PAL activity in grape cultivars (Sarabandi et al. 2019). Furthermore, Sofo et al. (2005) concluded that under abiotic stress, a decrease in polyphenol oxidase (PPO, having catechol oxidase activity) was related to the enhancement of the antioxidant ability and phenolics accumulation. Tomatoes are among the most cultivated vegetable crops worldwide (Srividya et al. 2014) and a rich source of vitamins, minerals (USDA 2016) and antioxidants (Di Masico et al. 1989). They are considered one of the most prominent vegetable crops in Egypt, they are cultivated throughout the year. The incidence of B toxicity in tomato cultivation elds has been reported in many countries from dry areas of the world, including Egypt (Landi et al. 2012). The effects of benzoate and gallate applications on redox enzymes, B-accumulation forms, and cell wall thickening were not reported in tomatoes under excess boron conditions, while SA is an active hormone for mitigating boron excess stress on antioxidant enzymes (Radi et al. 2014). This investigation provides new insights into the coordinating effects of redox enzymes induced by BA, GA, or SA and reducing forms of B accumulation, cell wall thickening to improve resistance against oxidative stress induced by EB in tomato calli.

Materials And Methods
Plant tissue culture The in vitro experiment was performed with the cultivar "Castle Rock" of tomato (Solanum lycopersicum L.). Seeds were disinfected with 5% NaClO prior to germination with half strength of MS (Murashige and Skoog 1962). Seedlings were grown (10-12 days) under growth chamber conditions (16/8 light period, temperature 25 ± 1 ° C). The explants (1.0 cm of the hypocotyl) were grown in a pre-prepared MS medium and incubated in a growth chamber for a month (Farghaly et al. 2021).
The medium included MS (4.4 g L -1 ), sucrose (30 g L -1 ), a-naphthalene-acetic acid (0.1 mg L -1 ), and 6benzyl-amino-purine (1 mg L -1 ), 4 various treatments were as follows: (1) The control set was MS-nutrient medium alone (control) and 2 mM boric acid on the MS-nutrient medium Calli were harvested after a month, rinsed with sterile distilled water (dw), sucked with lter paper, for fresh weight (FW) assessment (FW), oven-dried for dry weight (DW) assessment, and the other calli were frozen at -80°C.

Hydrogen peroxide
The H 2 O 2 content in the callus was assessed by the Velikova et al. (2000) method. After the callus sample was homogenized in 0.1% (w/v) trichloroacetic acid (TCA), the mixture was properly centrifuged, the ltrate (0.5 mL) was treated with 0.5 mL of 0.1 M K phosphate buffer (PPB; pH 7.0) and 1 mL of 1 M KI then after 20 min, the absorbance was assessed at 390 nm and expressed in mg g -1 fresh weight (FW).

Enzyme extraction
Frozen samples (0.5 g of callus tissue) were ground in 5 mL of 100 mM PPB (pH 7.8) comprising 0.1 mM ethylenediaminetetraacetic acid (EDTA) and 0.1 g polyvinylpyrrolidone and then properly centrifuged. Lowry et al. (1951) technique was used to quantify the soluble protein in the supernatants. The activities of the examined enzymes were calculated based on the difference in absorption wavelength (nm) per mg protein per min.
Lipoxygenase (EC 1.13.11.12) The Minguez-Mosquera et al. (1993) technique was followed to assess LOX activity. To dw (5 mL), Tween-20 (5 μL), and linoleic acid (35 μL) were mixed and NaOH (0.2 M) was added until the mixture was clear and the pH was adjusted to 9.0, then the mixture was completed to 100 mL with PPB (0.1 M) and the pH was adjusted to 6.5. To the previous mixture (2.95 mL) enzyme aliquots (50 μL) were added and then the change in absorbance was measured at 234 nm.

Boron forms
The different B forms were extracted as reported by Du et al. (2002) and Li et al. (2017). To dry callus powder, dw was added, left on a shaker at 25°C for 24 h, ltered, and free B was measured in the ltration, and NaCl (1 M) was put on the residue and left on a shake at 25°C for 24 h, ltered. Semi-bound B was measured in the ltrate, then HCl (1 M) was put on the residue and left on a shaker at 25°C for 24 h, ltered, and the bound B was measured in the ltrate. The curcumin-acetate method was applied to measure the B forms concentration (Mohan and Jones 2018).

Transmission electron microscope imaging
The plant samples were prepared using an adaptation of the method used by Bozzola and Russell (1991). Samples were viewed using eld-emission TEM (JEOL transmission electron microscope JEM-100CX II).

Statistical analysis
The results obtained were a mean (± standard deviation) of 4 replicates, with all three technical measurements, in most cases, and statistical tests were calculated by SPSS software. ANOVA (One-way analysis of variance) was employed and followed by Multiple Range Tukey Test for all PA treatment without or with EB. Pearson correlation test was performed to perceive the relationship between the mean rate of different parameters of tomato under BA, GA, or SA without or with EB and asterisks show signi cant correlation (* and ** at 5 and 1%, respectively). Correspondence analysis was performed to investigate the relationships between redox enzymes and different levels of each treatment with PA with or without EB. Excel's t-test was used to investigate the signi cant difference between treatment with PA with or without EB (ns = non-signi cant; * P< 0.05; ** P< 0.01; *** P< 0.001).

Results
Callus growth EB stress has been declared to inhibit cell and plant growth factors. ANOVA analysis of FW and DW data of tomato calli showed that EB stress resulted in a signi cant decrease in the mass gain of the studied tomato calli ( Fig. 1A

Hydrogen peroxide
The H 2 O 2 contents, a product of redox metabolites, were evaluated within tomato calli that underwent various treatments to assess the degree of oxidative regulation resulting from EB stress and the in uences of benzoate, gallate, and salicylate in reducing this negative damage ( Fig. 2A, Tables S1-6). As revealed in Fig. 2A

Lipoxygenase activity
To test whether the positive effects of PA treatments on EB-stressed calli were related to their ability to reduce membrane damage, LOX activity was assessed in calli exposed to EB stress with or without PAs application (Fig. 2B, Tables S1-6). LOX activity in calli was increased with an increase of B in the MS nutrient medium and the increase was 65.96% compared to the control calli. Treating EB-stressed calli with the three tested PAs attenuated the negative impact of EB stress on LOX activity. LOX activity decreased signi cantly, in most cases, and recorded the highest drops in moderate levels of benzoate, gallate, and salicylate; whereas, compared to EB-stressed calli the decrease was 36.43%, 25.64%, and 15.65%, respectively.
Under MS-boron conditions, signi cant increments in LOX activity were noticed with the application of PAs, and only BA recorded non-signi cant changes. Moreover, the application of benzoate, gallate, and salicylate induced strong positive relations between LOX activity and H 2 O 2 content in calli-treated with or without EB.
Superoxide dismutase activity SOD activity was estimated as a remarkable scavenger for ROS (Fig. 3A, Tables S1-6). SOD activity in callus cells signi cantly increased when the callus was exposed to EB, compared to the control calli, and the recorded increase was 23.63%. The use of gallate and salicylate did not considerably alter SOD activity in EB-stressed calli; whereas benzoate statistically reduced its activity.
Under MS-boron conditions, benzoate and gallate treatments did not alter SOD activity, however, its activity gradually increased with increasing salicylate level in the nutrient medium. Furthermore, the relations between SOD activity and H 2 O 2 content were minimal in calli-treated with PAs and stressed or unstressed with EB, only the correlations in calli-treated with SA (0.895**) without EB and BA (0.579*) with EB were signi cant.
Catalase activity CAT activity was estimated in tomato calli subjected to different levels of PAs with or without EB because it aids in the decomposition of H 2 O 2 into H 2 O and O 2 (Fig. 3B, Tables S1-6). As appeared in Fig. 3B, it can be observed that the EB had a considerable catalytic effect on CAT activity in the callus (about 2-fold above the control calli). However, applications of the three concentrations of benzoate, gallate, or salicylate to EB-stressed calli reduced the stimulating effect of EB on CAT activity.
Treatment of EB-stressed calli with a moderate concentration of GA showed a higher decrease in CAT activity (60.51%) compared to moderate concentrations of benzoate and SA, which recorded a decrease of 39.46% and 34.44%, respectively, of B-stressed calli.
Under MS-B cases, the three phenolic acids applied, in the most cases, signi cantly increased CAT activity in callus cells compared to MS-B-unstressed calli. Interestingly, PA applications caused strong positive relations between CAT activity and H 2 O 2 content in calli stressed or unstressed with EB, only GA without EB stress caused a non-signi cant relationship.
Peroxidase activity POD activity was determined because it induces oxidation by H 2 O 2 for a wide range of organic materials ( Fig. 3A, Tables S1-6). POD activity was stimulated with increased B in the MS-nutrient medium which showed a 35.11% increase in POD activity compared to MS-B-unstressed calli. Supplementation of benzoate and gallate at different concentrations with excess boron increased POD activity, while salicylate treatments decreased its activity signi cantly. Compared to EB-stressed calli, moderate levels of BA and GA showed increases in POD activity of 26.16% and 83.47%, respectively, while SA at moderate concentration reduced its activity by 43.63%.
Under MS-B conditions, benzoate and gallate treatments progressively improved POD activity in the corresponding absolute controls. In contrast, POD activity decreased signi cantly with the application of SA levels in the MS medium. Furthermore, in the callus treated with excess boron and benzoate, gallate, and salicylate the correlation between POD activity and H 2 O 2 content was signi cant (-0.603*, -0.789**, and +0.865**, respectively), while it was insigni cant in the PAs-treated-callus only.
Ascorbate peroxidase activity APX activity was determined because it induces the hydrogen peroxide dependent oxidation of AsA in plants (Fig. 3B, Tables S1-6). Treating tomato callus with 2 mM B positively affected APX activity. Under EB states, the increase in APX activity, compared to MS controls, was 34.99%. Treatment with different levels of benzoate and gallate plus excess boron enhanced APX activity (10.46%, 22.34%, 16.23% for benzoate and 10.25%, 9.16%, 20.32 for gallate, respectively, over B-stressed calli), while salicylate treatments did not signi cantly alter its activity.
Under MS-B states, BA treatments reduced APX activity in calli, only the higher level did not alter its activity. GA treatments catalyzed APX activity, in most cases, under MS conditions. However, salicylate applications did not alter APX activity, only the higher level enhanced its activity, under MS conditions. Additionally, our results showed that APX activity in calli treated with only SA (0.655*) and BA with EB (-0.744**) showed a strong association with H 2 O 2 content; whereas, in other PA treatments with or without excess boron, the correlations were insigni cant.
Phenylalanine ammonia-lyase activity PAL activity was examined to mark whether co-applications of PAs with or without EB treatments affected phenolic biosynthesis in tomato calli (Fig. 4A, Tables S1-6). The data showed that EB signi cantly failed to enhance PAL activity in tomato calli. Nonetheless, the moderate and high concentrations of the tested PAs signi cantly enhanced PAL activity compared to EB-stressed calli. Combined application of BA, GA, SA, and EB at the highest level resulted in increased PAL activity, which showed a higher increase, in most cases, of 16.83%, 42.03%, and 18.95%, respectively, than that found in EB-stressed calli. Also, our results manifested that correlations between PAL activity and H 2 O 2 content were not signi cant in the existence of PAs with EB, only there was a signi cant relationship in the BA case with EB (-0.684*).
Correspondingly, under MS-B conditions, PAL activity decreased by 21.74% and 29.82%, respectively, at moderate and high concentrations of BA compared to MS-calli. However, gallate and salicylate treatments did not affect PAL activity in calli exposed to MS-B conditions; only the highest level of SA boosted its activity (46.39%). Moreover, under BA without EB treatments, PAL activity showed a negatively strong association with H 2 O 2 content (-0.673*); whereas, the same relationship was positively strong in cases of GA (0.714**) and SA (0.820**) without EB.
Polyphenol oxidase activity PPO activity in calli was examined to look at the degree of phenolic oxidation induced by PAs with or without EB treatments (Fig. 4B, Tables S1-6). EB negatively affected PPO activity in calli, with PPO activity reduced by 31.86%, compared to EB-unstressed calli. The PA treatments had, in most cases, signi cant catalytic effects on PPO activity in calli under EB conditions. GA in high concentration was the most active PA in stimulating PPO activity in callus tissues; the stimulation rate was 179.19%, compared to EB-stressed calli. Moreover, the results manifested that the PAs (BA, GA, and SA) together with the EB treatments caused signi cant relationships between PPO activity and H 2 O 2 content in calli (-0.777**, -0.918**, and -0.763**, respectively).
PA treatments without EB led to altered effects on PPO activity in callus tissues. The applications of different levels of benzoate signi cantly reduced PPO activity; however, gallate greatly enhanced its activity. In contrast, SA did not considerably alter PPO activity under MS-B states. Additionally, the results revealed that the PAs without EB treatments induced non-signi cant relationships between PPO activity and H 2 O 2 content in calli.

Boron forms
Under MS-B cases, B was found mainly in free form in the tomato callus, with free B present at 68.4%, semi-bound B at 23.4%, and bound B at 8.2% of total B. However, in cases of EB, free and semi-bound B decreased by 6.90% and 4.38% to increase bound B by 11.28%. It can also be seen that EB caused signi cant increases in free, semi-bound, and bound B contents in B-stressed calli higher than in nonstressed calli ( Fig. 5A Transmission electron microscopy TEM enables an estimate of the cell's microstructure, particularly in the internal features of cells and organelles. Hence, this technique was harnessed to look at the differences in ultrastructural resulting from excessive B stress and whether applications of phenolic acid enhanced callus cell resistance ( Fig. 6 and 7). Imaging obtained from TEM showed that EB-untreated (control) cells have a normal shape, large nuclei, large vacuoles, and thin cell walls (ranging from 333 to 559 nm). The EB caused the cell walls to be thicker between 1167 and 1347 nm, which means 2.4 to 3.5 times more than the control cell wall. However, the moderate concentration of benzoate, gallate, and salicylate treatments reduced cell wall thickness in EB-stressed calli. Compared to B-stressed calli, the highest decrease in cell wall thickness was found at 78.17-85.00%, 42.54-64.52%, 83.59-83.72%, respectively after exposure to benzoate, gallate, and salicylate. Moreover, the addition of benzoate, gallate, and salicylate alone without increasing B to the nutrient medium reduced the thickness of the cell walls.

Discussion
In plants, PAs are a subclass of phenols that have a single COOH group with or without one or more OH groups, causing the H atom to be donated to produce antioxidant properties. However, the performance of the action of PAs in mitigating the inhibition of EB stress on plant growth, cell ultrastructure, accumulation of B forms, and antioxidant enzymes have not been quite proven. Therefore, this research was conducted to test the mechanisms underlying PAs stimulating plant growth, to nd out the response of antioxidant enzymes, the association of boron forms, and modi cation of the cell ultrastructure in alleviating EB stress.
Under our investigational conditions, we found that the use of PAs mitigated the negative effect of EB stress on tomato calli, including their growth. Consistent with our previous studies (Metwally et al. 2018;Farghaly et al. 2021), the increase in boron in the present study remarkably impeded the growth of callus and the gain of fresh and dry biomass. In this research, EB stress (2 mM) induced oxidative stress (Fig. 2) that could be linked to a defect in antioxidant enzymes. Moreover, inhibition of callus growth could be due to the accumulation of boron in various forms in the callus cells that were in direct contact with EB and the thickening of the cell walls. Inhibition of callus growth due to excessive boron stress may result from inhibition of cell division and rate of cell expansion, which is mainly caused by cell wall thickening. Accordingly, the increase in cell wall-bound B (Fig. 6) and increased cell wall thickening (Fig. 8) in this study may refer to decreased tomato callus growth. Podgórska et al. (2017) reported that the thickness of cell walls depends on B ion availability.
However, when PAs were co-applied with EB, a signi cant increase in FW and DW of calli was observed in boron-stressed calli (Fig. 1). Furthermore, our results indicated that in the presence or absence of EB, BA was the most active PAs in improving calli growth, and for all PAs used, moderate concentration was most active in increasing growth. The strong negative relations between callus growth, H 2 O 2 content, and the concentration of B forms under the treatments of PAs and EB con rmed that the higher the callus growth, the less B accumulation and the decreased oxidative stress in the plants. These results indicate that phenolic acids play signi cant roles in mitigating the damage to tomato calli that occurs under conditions of excess boron stress, as they regulate antioxidant enzymes involved in detoxifying ROS, and reduce the accumulation of boron forms in cell walls and cytoplasm. Likewise, the increase in callus growth was recorded in tomatoes using PAs under EB (Farghaly et al. 2021 (Dat et al. 2000). In this investigation, the H 2 O 2 content in calli was assessed to screen for oxidative stress triggered by EB stress.
The results (Fig. 1A)  In contrast, considerable increases in LOX activity in GA and SA treated calli without EB indicate that the oxidative metabolites of fatty acids resulting from LOX activity are predominantly involved in growth rather than aging. Consistent with our results, Siedow (1991) stated that oxygenated fatty acid products of LOX activity have a clear role in growth, aging, and reaction to external stress.
Most of the enzymatic antioxidants were affected under EB stress with or without PAs, indicating that those enzymes enable calli to withstand the EB stress.
The SOD enzyme is metallic in nature, which forms the primary frontier of the defense against ROS (Jackson et al. 1978 CAT plays a critical function in withstanding plant stress, and increasing CAT activity under stressor conditions is a signal for oxidative stress tolerance, as it degrades H 2 O 2 the SOD product (Mittler 2002).
According to our data, CAT activity in calli-stressed with EB signi cantly increased; indicating that overexposure to EB increased the content of ROS in tissues; however, it also improved defense systems.
CAT activity has likewise been reported with EB use in tomato seedlings (Kaya et al. 2020 (2020) found that GA alone and GA + Cd did not induce CAT activity in wheat seedlings. In addition, Amist and Singh (2018) observed a decrease in CAT activity of wheat seedlings-treated with hydrotic stress and BA. Parallel to our nding, decreased CAT activity was detected during EB-toxicity and SA application in canola plants (Radi et al. 2014).
Both CAT and POD, in concert with SOD, play a remarkable defensive role in removing ROS (Jaleel et al. 2009). The current results demonstrated that with increased B in the MS-nutrient medium, POD activity in calli increased; indicating that tomato calli adapt to EB through effective defense systems. Moreover, this increase in activity was bound with high phenolic content, as observed in the current study. Likewise, Kaya et al. (2020) found that excess boron increased POD activity in SC 2121 tomato cultivars. Under MS or EB conditions, treatments with benzoate and gallate were further active in stimulating POD activity in calli; whereas, SA treatments reduced this activity. Our results also revealed that the association between POD activity and H 2 O 2 content was signi cantly negative in calli-treated with EB + BA or EB + GA, while it was positively signi cant in calli-treated with EB + SA. Nonetheless, the same association was minimal in the callus treated with PAs without increasing boron. These results suggest that, in the existence of EB, this enzyme was responsible for H 2 O 2 elimination in calli-treated with benzoic or gallic acid; however, it was not responsible for this elimination in calli-treated with salicylate. Yadav and Singh (2013) concluded that the elevated activity of POD by application of BA to Cd-stressed-wheat seedling is dose-dependent.
Recently, Yetişsin and Kurt (2020) found that GA stimulated POD activity in Cu-stressed maize seedlings.
The decrease in POD activity in EB-stressed calli treated with SA con rmed by a previous study on canola who found that protein accumulation, decreased CAT activity, and increased POD, and PPO activities in the apoplastic region could be the best strategy to tolerate cold temperature in wheat leaves. In calli treated with BA without EB, the decrease in PPO activity was mostly related to the increase in cysteine content (Farghaly et al. 2021), as Mishra and Gautam (2016) reported that cysteine inhibited PPO activity.
Under excess boron, an increase in PPO activity and decreased cysteine content in calli could con rm the previous suggestion (Farghaly et al. 2021). These results require additional investigation.
Boron possesses a pKa of 9.2 and is thus present in the MS-nutrient medium in an uncharged form that can be absorbed by passive diffusion across the plasma membranes. Boron is normally present in the cell in three major forms, free, semi-bound, and bound, the free is soluble B that is directly available for potential physiological roles, semi-bound is mainly involved in the synthesis of the cell wall, and bound is a component of cell walls (Du et al. 2002). The results illustrated that the increased supply of B into the MS-nutrient medium led to increased uptake of B forms into calli (Fig. 6). The results illustrated that free B (61.5%) was most prevalent in tomato calli, which may be related to symptoms of EB stress.
Interestingly, the bound B was increased by about 11% under EB stress than the control calli, this may be due to the increased binding sites on cell walls by the rhamnogalacturonan RG-II. Thickening of callus cell walls under the EB con rmed the increase of bound B thus decreasing callus expansion and growth.
Accordingly, Reid (2007) stated that additional boron binding on the cell wall may further disrupt the cell expansion and growth processes.
However, when PAs were co-applied with EB, a signi cant decrease in the accumulated B forms was observed in B-stressed tomato calli (Fig. 6), resulting in a considerable increase in the growth criteria of the stressed calli (Fig. 1). Our results clearly demonstrated that the applications of PAs reduced free, semi-bound, and bound B accumulation (Fig. 6), which resulted in the overall enhancement in the callus growth under excess boron stress (Fig. 1) (Fig. 9A-B).
The strategy of the three PAs applied was to promote plant growth under EB by reducing H 2 O 2 production through the activation of CAT activity and reducing cell wall thicknesses by reducing the accumulation of various B forms, especially the bound form. These results suggested that the most effective phenolic acids were BA > GA > SA. Moreover, benzoate was more active at the moderate concentration used (1 µM), which was the lowest concentration of the total concentrations of the three phenolic acids used. These results could contribute to improving the management strategy to alleviate excess boron stress in tomatoes, but more studies are needed.  (2)   indicate signi cance (one-way ANOVA; Tukey HSD-post-hoc), and asterisks indicate signi cant differences between treatments with or without boron (t-test; * P < 0.05; ** P < 0.01; *** P < 0.001; ns = not signi cant).