The surface sterilized seeds of Solanum lycopersicum L. cv. Micro-Tom wildtype (WT) and transgenic lines and empty vector (VC) were grown in green house under following conditions: 16 h/8 h light/dark cycle, 25°C/18°C day/night temperature, light intensity 250 μ mol m⁻²s⁻¹, and 80% relative humidity. For gene expression analysis, plant parts, such as root, leaves, stem, flowers (in bud and fully opened), and flower parts (sepal, petal, carpel, and stamens) were harvested from 4-week-old plants (Waseem et al., 2018). For each sample, each plant part was collected from 10 plants were mixed and frozen in liquid nitrogen.

For overexpression, the K303 expression vector (Gateway technology) under CaMV 35S promoter was constructed as described in Waseem et al. (2019). For RNAi a 217 bp long SlbHLH22 fragment for sense and antisense silencing in pCAMBIA 2301 vector. The specific primers used for RNAi are listed in Supplementary Table S1. The transgenic line plants were generated by agrobacterium-mediated (Agrobacterium tumefaciens strain, GV310) transformation in WT tomato plants as described by Xian et al. (2014). The transgenic plants were screened on MS media supplemented with kanamycin (100 mg L⁻¹). The generated kanamycin resistant seedlings were transferred to green house for further growth under control conditions and then verified with successful qRT-PCR. The homozygous T₃ lines were used for further analysis.

The SlbHLH22 ORF without the stop codon was amplified and cloned into the pGreen0029 vector. The recombinant plasmid containing the SlbHLH22-GFP fusion gene and the control plasmid with GFP alone were transformed into onion epidermal cells using Agrobacterium-mediated transformation as described by Sun et al. (2007). For trans-activation assays, to produce pBD- SlbHLH22 the coding sequence of SlbHLH22 was amplified and ligated into the yeast expression vector pGBKT7 (Clontech, United States). pBD-SlbHLH22, pGBKT7 (plasmid for negative control), and pGBKT7-53+pGADT7-T (plasmid combination for positive control) were transformed separately into the yeast strain AH109 according to the manufacturer’s protocol. Transformants were selected on SD/-Trp or SD/-Ade/-His/-Trp drop-out medium (Clontech, United States). After colony formation, the trans-activation activity of each protein was examined by comparing growth on permissive and selective medium and the activity of X-gal (40 μg mL⁻¹, 5-bromo-4-chloro-3-indoxyl-α-D-galactopyranoside).

Total flavonoid contents were determined by the AlCl₃ method as described by Koolen et al. (2013) with slight modifications. 2.5 g of tomato leaves (WT, 35S:SlbHLH22 and RNAi lines) were ground in liquid nitrogen and dissolved in 70% (by vol.) ethanol solution and incubated at room temperature for 24 h. 1 mL of ethanol extract was diluted with 1 mL of AlCl₃ (5%, w/v) and incubated for 1 h at room temperature. The mixture was centrifuged at 10,000 rpm for 10 min and supernatant was collected in a new tube. One volume of chloroform was added to remove chlorophylls. The mixture was centrifuged at 8000 rpm for 5 min and supernatant was used to measure absorbance at 430 nm. Total flavonoids were expressed in mg quercetin equivalent/g dry weight (Koolen et al., 2013).

For salt, osmotic, and oxidative stresses, the leaves from 5-week-old WT seedling were sprayed with 200 mM NaCl (Waseem and Ahmad, 2019), 100 mM mannitol, and 100 mM hydrogen peroxide, respectively. Leaves were harvested at 0, 3, 6, 9, 12, and 24 h. Leaves harvested at 0 h were used as control. For each sample, leaves were collected from 10 plants, mixed and all the experiments were performed in triplicate.

For salinity and drought stresses, each 15 plants of WT and transgenic lines were placed in big pot, watered twice in week to make sure water was uniform in all pots and were grown under same light and temperature conditions. For salinity stress, 6-week-old plants were irrigated with 200 mM (200 mL per pot, 9 cm) NaCl for every 48 h in the following 18 days. However, for drought stress, withhold water for up to 30 days followed by rehydrated for 10 days. The control plants were watered normally. During treatment, the relative water content (RWC, %) and total chlorophyll content was assessed (Pan et al., 2012). The leaves at same development stage were harvested and store immediately at −80°C till further analysis.

During salinity and drought treatment, leaves at same development stages were harvested from plants for antioxidant enzyme activity such as catalase (CAT. EC 1.11.1.6) superoxide dismutase (SOD, EC 1.15.1.1), peroxidase (POD, EC 1.11.1.7), H₂O₂ content, Malondialdehyde (MDA), soluble sugar content, and proline content assessment. MDA following method by Heath and Packer (1968). Briefly, about 0.5 g of tomato leaves were ground in 2 mL of the chilled reagent [0.25% (w/v) thio-barbituric acid in 10% (w/v) trichloroacetic acid]. The extracts were incubated at 100°C for 30 min, cooled to room temperature. The extracts were centrifuged at 12,000 rpm for 15 min and absorbance of the supernatant was measured at 450, 532, and 600 nm. The MDA content was calculated based on the following equation: 6.45 × (OD₅₃₂–OD₆₀₀) −0.559 × OD₄₅₀.

Soluble sugar content was measured according to method described by Fukao et al. (2006). Proline contents were determined following Bates et al. (1973). About 0.5 g of tomato leaves were ground into powder with liquid nitrogen and extracted in 3% sulfosalicylic acid. After centrifuging at 12,000 rpm for 10 min, the supernatant (2 mL) was mixed with equal volume of ninhydrin reagent [2.5% (w/v) ninhydrin, 60% (v/v) glacial acetic acid, and 40% 6 M phosphoric acid] and of glacial acetic acid, incubated at 100°C for 40 min. The reaction was terminated in an ice bath. Then, the reaction mixture was extracted with 4 mL of toluene. The absorbance was measured at 520 nm with a UV-5200 spectrophotometer.

For SOD activity, 1 g of frozen leaves tomato leaves were homogenized in 5 ml of cold 20 mM HEPES buffer (pH 7.2, 1 mM EGTA, 210 mM mannitol, 70 mM sucrose) then centrifuged at 2,500 rpm for 5 min at 4°C. The enzyme activity SOD were measured following Mittova et al. (2000). Total protein from tomato leaves was extracted with 0.05 M potassium phosphate buffer (pH 7.0). After centrifuging at 12,000 rpm for 15 min at 4°C, the supernatant was used for the measurement of POD and CAT activities. POD activity was determined using the previously described method by Morohashi (2002) and Morohashi et al. (2003). The 5 mL reaction mixture contained 0.1 mL of the supernatant, 1 mL of 0.5% (v/v) H₂O₂, 2.9 mL of 0.05 M potassium phosphate buffer (pH 5.5), and 1 mL of 0.05 M guaiacol as substrates. The oxidation of guaiacol was monitored by the absorbance measured at 470 nm every 10 s. CAT activity was confirmed using a Catalase Assay Kit (Jiancheng Bioengineering Company, Nanjing, China) according to the manufacturer’s instructions. H₂O₂ content was determined according to instruction available in commercial kit from Jiancheng Bioengineering Company (Nanjing, China). For ABA quantification, ABA extracted from 1 g of leaves (WT, transgenic lines under stress, and mock) as described by Jia et al. (2011).

Total RNA was extracted from all harvested samples using Invitrogen^TM TRIzol^® reagent (Thermo Fisher Scientific, New York, NY, United States) according to the manufacturer’s instruction. The RNA concentration was determined using NanoDrop Lite UV-Vis spectrophotometer (Thermo Fisher Scientific^TM). The cDNA was synthesized with 2 μg of total RNA using PrimeScriptTM RT reagent Kit with gDNA Eraser (TaKaRA, Japan). All the primers (Supplementary Table S1) used in this study were designed in primer premier 5 (PREMIER Biosoft International, Palo Alto CA, United States). The real-time PCR was performed using SYBR^® Premix Ex Taq^TM II (TliRNaseH Plus) (Clontech, TaKaRa, Shiga, Japan) in 96 well plate, Bio-Rad CFX system (Bio-Rad, United States). The relative changes in gene expression was calculated by adopting 2^−Δ(ΔCT) method (Livak and Schmittgen, 2001) using SlUBI3 (Solyc01g056940) as an internal control. All the experiments were performed in triplicate.

All the experiments were performed in triplicate, reproducible and were presented as means ± standard error (SE). Statistical analysis of data was performed using Sigmaplot 12.1. (SYSTAT and MYSTAT Products, United States, and Canada) and two-tailed Student’s t-tests for salinity and drought comparison or Dunnett’s tests were used to compare between WT, empty vector plants, and each overexpression line to determine significant differences. The significance values of p ≤ 0.01^∗∗/0.05^∗ were considered.

To gain insight into the roles of SlbHLH22 in plant growth and development, we analyzed the expression profile of SlbHLH22 in various plant parts using qRT-PCR. The results suggested that SlbHLH22 expressed in all the tested tissues, but high expressions were found in leaves and flowers compared to root (Figure 1A). We further examined the expression of SlbHLH22 response to osmotic and oxidative stresses. WT plants were treated with NaCl, H₂O₂, and D-mannitol. The results indicated that the transcript of SlbHLH22 was upregulated after salt treatment across all time points (Figure 1B). For D-mannitol, SlbHLH22 was upregulated after exposure, but was peaked at 24 h time point (Figure 1C). For H₂O₂ was only peaked at 3 h interval and was downregulated in the remaining time points (Figure 1D). The observation of expression profile for SlbHLH22 indicated that SlbHLH22 gene might be very important for plant resistance against abiotic stress.

To determine the subcellular localisation of SlbHLH22 protein, the vector 35S-SlbHLH22-GFP was transiently expressed in living onion epidermal cells. Confocal imaging of protein fluorescence showed that the cells transformed with the vector containing GFP alone displayed fluorescence throughout the cells, whereas the green fluorescence signal of 35S-SlbHLH22-GFP was exclusively detected in the nucleus (Figure 2A). A Y2H experiment was used to examine the transcriptional activity of SlbHLH22. A GAL4 DNA-binding domain SlbHLH22 fusion protein was expressed in yeast cells, which were then assayed for their ability to activate transcription from the GAL4 sequence. SlbHLH22 promoted yeast growth in the absence of histidine and adenine, and showed X-α-gal activity, whereas the control vector pGBKT7 did not (Figure 2B). Moreover, string database was used to predict interaction network of SlbHLH22 with other proteins. It was found that tomato SlbHLH22 may interact with genes involved in flavonoid and ABA biosynthesis pathway include; CHS, CHI, PAL, PSY1, FLS, AAO, NCED, and ZEP (Supplementary Figure S1). Taken together, our results suggested that SlbHLH22 has transcriptional activity and is targeted to the nucleus in plant cells.

To further study the function of SlbHLH22 gene, the transgenic plant lines were generated by overexpressing the ORF and RNAi silencing by agrobacterium-mediated transformation. Three independent transgenic lines for overexpression (L18, L20, L23) and two RNAi lines (L14i and L19i) were detected exhibiting significant changes in expression fold (Figure 3A). Two transgenic plant lines L18 and L23 for overexpression and two lines for RNAi L14i and L19i were selected for further characterization. The transgenic plants showed pleiotropic phenotypes, such as plant height and leaf size (Figure 3B,C). Our results displayed that the tomato SlbHLH22 have remarkable effects on development of tomato plant.

As it was observed, SlbHLH22 expressions was activated by salinity. Thus, we hypothesized that SlbHLH22 might increase the resistance of transgenic plants to salinity and drought. To prove it, we explore the role the performance of transgenic and WT plants treated with salt solution for 18 days and deprived of water for 30 days. As shown in Figure 4, significant morphological changes were observed between transgenic plant lines with SlbHLH22 and WT plants after stresses. However, after stresses, the overexpression plant lines showed slight changes in their physical appearance, but RNAi lines and WT plants showed typical severe desiccation symptoms (Figure 4). In comparison, WT and RANi lines plants under drought showed severe damages than under salt treatment. However, upon exposure to normal conditions for 10 days, the overexpression plants recover very fast, but RNAi and WT plants under drought stress was unable to recover (Figure 4). The relative water content (RWC %) and total chlorophyll content was decreased in WT and transgenic lines during treatment, but relatively higher in the overexpression plant lines (Supplementary Figures S2A,B).

The total flavonoid contents were assessed in WT and transgenic lines under normal conditions and under salinity and drought stresses. It was found that the total flavonoid contents in overexpression plant lines were induced and suppressed significantly in RNAi lines compared to WT (Supplementary Figure S3). The expression levels of genes in the flavonoid biosynthesis pathway were further analyzed at molecular level in the overexpression and RNAi lines. The results indicated that the transcript levels of the flavonoid biosynthesis genes, such as SlCHS, SlCHI, SlF3’H, SlF3H, SlFLS, and SlPAL were peaked in overexpression lines than in WT. In SlbHLH22-RNAi lines the expression of all genes significantly downregulated than in WT plants (Supplementary Figure S3). These results demonstrating that SlbHLH22 affect flavonoid accumulation by modulating flavonoid biosynthesis pathway.

Salinity and drought lead to production of reactive oxygen species (ROS), that cause damages to membrane structure (Zhai et al., 2016). We investigated the changes in the accumulation of H₂O₂ in transgenic lines and WT plant under salinity and drought stresses. It was found that more H₂O₂ accumulated in WT plant than in transgenic lines (Figure 5). To explore the possible physiological mechanism responsible for the increased drought and salt tolerance, we compared the changes in contents of proline, MDA concentration, and total soluble sugars in the leaves from transgenic and WT plants grown under normal and stress conditions. MDA content was significantly peaked in RNAi-lines and WT plants under drought and salt stresses compared to overexpression lines (Figure 5). It was examined that the soluble sugars were accumulated more in overexpression lines than in RNAi-lines and WT plant, after salt and drought stresses (Figure 5). Furthermore, we examined the enzyme activities of antioxidants enzymes such as SOD, POD, and CAT of the leaves from transgenic plant lines and WT plant under stresses and normal growth. The enzymatic activities of SOD, POD, and CAT in transgenic lines and WT were almost same under normal conditions. For drought stress, the activities of CAT and SOD were significantly upregulated in overexpression lines as compared with WT and SlbHLH22-RNAi lines. However, POD activity was 10 points more in SlbHLH22 overexpression lines than in the WT and RNAi (Figure 5). The activities of SOD, POD, and CAT were significantly upregulated in overexpression lines under salt stress in compassion with the mocked corresponding transgenic lines, RNAi lines, and WT plants. The proline contents were increased in overexpression plant lines under drought and salt stress in mocked transgenic lines but peaked in transgenic lines under salt stress. However, in RNAi lines proline content was significantly downregulated than WT plants. It was found that the overexpression plant accumulated more prolines contents under salinity stress than under drought (Figure 5). Collectively, our results show that SlbHLH22 in tomato can help improve the resistance of transgenic plant to salinity and drought stresses.

The overexpression of SlbHLH22 improved plant lines resistance against salinity and drought leads us to examine whether the overexpression of SlbHLH22 affects the endogenous level of ABA and its biosynthesis genes in transgenic tomato. We measured the ABA levels in leaves of transgenic lines overexpressing SlbHLH22 and WT plants under normal growth conditions and under stresses. The results showed a significant difference in ABA levels in WT and transgenic lines under normal conditions. However, the endogenous ABA contents was significantly higher in transgenic lines than that in the WT under salt and drought stresses (Figure 5). To further insight into the role of SlbHLH22 in ABA biosynthesis, we examined ABA levels by generating RNAi lines. It was observed that ABA level in RNAi is downregulated than WT under mock and salinity and drought stress (Figure 5). To ascertain the molecular mechanism involved in ABA biosynthesis, we analyzed the expression level of ABA biosynthesis genes between transgenic lines and WT plants. The results indicated that the expression of genes involved in ABA biosynthesis, such as SlAAO, SlABA2, SlNCED, and SlZEP were upregulated in transgenic lines under salt and drought than in WT and RNAi lines (Figure 6). Meanwhile, we investigated the expression profiles of genes involved in ROS scavenging system, including SlCAT, SlPOD, SlSOD (Supplementary Figure S4), and proline biosynthesis such as SlP5CS and SlP5CR (Supplementary Figure S4) in WT and transgenic lines under normal and stress conditions. The results suggested minor changes in the expression levels of SlCAT, SlPOD, and SlSOD in WT plants under normal and stress conditions but significantly downregulated in SlbHLH22-RNAi lines. Moreover, the transcripts of SlCAT, SlPOD, and SlSOD accumulated more in overexpression lines under stresses than WT and mocked corresponding overexpression plant lines grown under normal conditions (Supplementary Figure S4). SlP5CS and SlP5CR are two key genes in proline biosynthesis. The expression levels of SlP5CS and SlP5CR were upregulated under stress in WT plants, but more high transcript levels were detected in overexpressing plant lines grown under stress treatments (Supplementary Figure S4). These observations suggested that tomato SlbHLH22 may enhance tomato resistance to salt and drought stresses through ABA and/or other pathways.