The Agriculture Research Center in Egypt supplied a virus-free tomato (Solanum lycopersicum) cultivar, GS-12, showing susceptibility to PVY infection. The PVY strain DA55 employed in this research has been previously documented (Aseel et al., 2023b). The purified PVY served as the viral inoculum for all treatments challenged with PVY. The concentration of PVY inoculum was 20 µg/mL, prepared in a 10 mM phosphate buffer at pH 7.2, with the addition of 0.1% sodium sulfite.

AMF was kindly obtained from the Plant Pathology Research Institute, Agricultural Research Centre, Egypt. The inoculum used in this study was made up of equal amounts of spores from Rhizophagus irregularis (Blaszk., Wubet, Renker, and Buscot), Rhizoglomus clarum (Nicolson and Schenck), and Funneliformis mosseae (Nicolson and Gerd.). We cultivated the AMF inoculum in sterile soil beneath Sudan grass (Sorghum bicolor L.) as a potential host. The AMF inoculum, exhibiting 83.1% colonization, consists of root fragments, mycelia, spores, and rhizospheric soil. The T. viride isolate Tvd44 (Ac# OQ991378) utilized in this study was previously isolated and characterized (Aseel et al., 2023b). The Tvd44 culture filtrate (spraying solution) was prepared by inoculating 1 mL containing 1 × 10^9 conidia into 100 mL of potato dextrose broth and culturing it at 28°C for 6 days on a rotary shaker at 150 rpm. Next, we filtered the culture by using Whatman filter paper No. 1. The filtrate was subjected to a 0.2 μm pore biological membrane filter before application to the plant leaves.

Before planting, tomato seeds were surface-sterilized with a 70% ethyl alcohol solution and a 0.5% sodium hypochlorite (NaOCl) solution. At 20 days after seeds were planted in sterilized peat moss soil, tomato seedlings were transferred to 25-cm-diameter pots filled with a mixture of equal parts sterilized soil, sand, and clay, with five seedlings in each pot. Each treatment had five replicates. Fertilization was not applied, and all pots received regular irrigation. The plants were grown in a controlled greenhouse environment maintained at 26-28°C during the day and night, with a humidity level of 65%. Seven treatments were administered. The first treatment (control) consisted of tomato plants that were foliarly treated with sterilized medium-free Tvd44 and mechanically inoculated with viral inoculation buffer. The second treatment (AMF) consisted of tomato plants colonized by AMF, supplemented with a foliar application of sterilized medium-free Tvd44, and mechanically inoculated using a viral inoculation buffer. The third treatment (Tvd44) consisted of tomato plants that were foliarly treated with T. viride culture filtrate and mechanically inoculated with viral inoculation buffer. The fourth treatment (PVY) included tomato plants that were foliarly treated with sterilized medium-free Tvd44 and mechanically rubbed with PVY. The mechanical inoculation involved dusting the upper two leaves of each tomato seedling with carborundum and gently rubbing with 1 mL of PVY using the forefinger technique (Omar et al., 2022). The fifth treatment (AMF+PVY) consisted of tomato plants colonized with AMF, foliarly treated with sterilized medium-free Tvd44, and mechanically inoculated with PVY. The sixth treatment (Tvd44+PVY) included tomato plants that were foliarly treated with Tvd44 culture filtrate and mechanically inoculated with PVY. The seventh treatment (dual+PVY) involved tomato plants colonized by AMF, foliarly treated with T. viride culture filtrate, and mechanically inoculated with PVY. In all AMF-treated plants, AMF was applied during the transfer of tomato seedlings into new pots (at 20 days post-germination) by incorporating 10 g of the AMF inoculum into each seedling bed. The viral inoculation was conducted on all PVY-inoculated plants fifteen days after seedling transplantation, corresponding to 35 days post-germination. In every instance of Tvd44-treated plants, the culture filtrate was administered to the tomato leaves 24 h before the inoculation with PVY, corresponding to 34 days post-germination (at two fully expanded true leaves). A handheld pressure sprayer was employed to apply culture filtrate or sterilized medium-free Tvd44 to the entire plant until runoff occurred. All plants were maintained in insect-proof greenhouses for more than three weeks, and the manifestation of viral symptoms was monitored daily (Abdelkhalek et al., 2022a).

Tomato plants were collected 25 days post-PVY inoculation (dpi) and underwent various physiological and biomolecular assessments. The plants collected from each group were subjected to multiple washes with running water. For further analysis, three upper leaves from each plant were collected from every pot (15 leaves per pot) and categorized as a biological sample within the same treatment group, resulting in 5 biological samples. Every biological sample underwent three technical replicates. The disease incidence and severity were evaluated by observing visual viral symptoms, utilizing a rating scale from 0 to 5, and assessing the extent of leaf damage (Imran et al., 2012; Aseel et al., 2023b).

Five tomato roots from each applied treatment were estimated for mycorrhizal colonization with PVY at 25 dpi. Small segments (1 cm) of each root were prepared and stained with trypan blue (Sigma-Aldrich, St. Louis, USA), as previously described (Phillips and Hayman, 1970). Mycorrhizal colonization was assessed in root segments of each treatment using a light microscope, adhering to the previously described approach (Trouvelot, 1986).

At 25 dpi, all plants from each treatment were meticulously uprooted, washed under running water, and assessed for the number of leaves, root and shoot lengths (in cm), and shoot and root dry weights (in g). The dry weights were measured following a 72-hour drying period of the plant samples in an oven set to 40°C.

The nitrogen (N) content in the tomato leaves was determined through the Kjeldahl method (Horneck and Miller, 2019). As previously described, total sulfur (S) was determined by precipitating the sulfate with barium chloride and measuring turbidity using a spectrophotometer at 420 nm (Olsen and Sommers, 1982).

The evaluation of photosynthetic chlorophyll a and b in tomato leaves was conducted as described previously (Harborne, 1998). The techniques for extracting and assessing the enzymes peroxidase (POX), polyphenol oxidase (PPO), and catalase (CAT) were established by Maxwell and Bateman (1967); Galeazzi et al. (1981), and Aeby (1984), respectively. The extraction and methodology for quantifying total protein content were conducted following Bradford (1976).

Total RNA was extracted from 100 mg of fresh tomato leaves collected at 25 days post-inoculation (dpi) using the RNeasy Plant Mini Kit according to the manufacturer’s instructions. Following the RNA concentration and purity assessment, a reverse transcription procedure was conducted to convert one µg of DNase-treated RNA into cDNA utilizing M-MuLV reverse transcriptase, as previously outlined (Aseel et al., 2023b). The cDNA was stored at -20°C until it was used. The evaluation of AMF colonization and T. viride treatments on the expression of defense-related genes and the accumulation level of the PVY-CP gene in tomato plants was conducted using real-time quantitative PCR (RT-qPCR). The expression ratio of the PVY-CP gene to the housekeeping gene in tomato control plants was utilized to determine the level of viral accumulation. The sequences of specific primers for tomato defense pathway genes WRKY1, WRKY19, JERF3, GST1, CHS, C3H, CHI2, FLS, F3H, F3'H, and HQT, as well as PVY-CP, are listed in Supplementary Table S1 . The β-actin gene was selected as a housekeeping gene due to its stable expression in tomatoes, as reported in prior studies (Aseel et al., 2023a; Rabie et al., 2024). The RT-qPCR experiment utilized a CFX Connect TM Real-Time System (BIO-RAD, USA). Each reaction contained 10 µL of 2xSYBR Green RT Mix (Bioline, Germany), 7.4 µL of RNase-free water, 0.8 µL of each primer (10 pmol/µL), and 1 µL of cDNA. The RT-qPCR protocol consisted of an initial cycle at 95°C for 5 min, followed by 45 cycles of 95°C for 5 seconds, 60°C for 10 seconds, and 72°C for 15 seconds. The relative gene expression was determined using the comparative C_T approach (2^−ΔΔCT) as described previously (Schmittgen and Livak, 2008).

Volcano plots were created using the ggplot2 package, and a t-test was employed to determine the statistical significance of gene expression compared to the control treatment. The significance was expressed as -log_10, while gene expression was expressed as log₂ fold-change. Heatmap and cluster analysis figures of gene expression were generated using pheatmap package. The data were first transformed using log₂; then, Euclidean distance was applied, followed by the Ward method of hard clustering. Thirty internal validation indexes were employed to determine the optimal number of clusters in the data (Charrad et al., 2014). Principal component analysis was conducted using function prcomp in stats package. The figures of contribution and biplot were built using the factoextra package. Gene expression and physiological trait networks were generated using the qgraph package, based on the Spearman rank correlation coefficient, to connect gene expression results with physiological data. All statistical analyses were performed using R software version 4.3.2 (2023).

Under controlled greenhouse conditions, the tomato plants treated with PVY exhibited characteristic viral symptoms at 19 dpi. The pronounced leaf mosaic, distinct veins, yellowing, and necrosis were distinctly observable at 22 dpi ( Figure 1 ). Interestingly, the AMF+PVY, Tvd44+PVY, and dual+PVY treatments resulted in a delay of approximately 5 days in the appearance of symptoms, with mild symptoms observed at 25 dpi ( Figure 1 ). No symptoms were observed in the control, AMF, and Tvd44 treatments. The findings demonstrated that the PVY treatment resulted in a 100% incidence of disease and a severity level reaching 96.8% ( Table 1 ). The findings correspond to the accumulation level of the PVY-CP gene, indicating an approximately 27.8-fold increase in PVY-treated plants relative to control plants. Compared to PVV treatment, the AMF+PVY, Tvd44+PVY, and dual+PVY treatments exhibited significantly reduced PVY accumulation expressions, with relative expression levels of 4.82-, 5.84-, and 3.29-fold, respectively. The dual+PVY treatment resulted in a decrease in disease incidence and severity to 60% and 46.62%, respectively. The AMF+PVY and Tvd44+PVY exhibited disease severity levels of 51.96% and 54.64%, respectively, as shown in Table 1 . Nonetheless, the data presented in Figure 2 indicated that the roots of infected plants with PVY were shorter than those treated with AMF+PVY, Tvd44+PVY, and/or the dual+PVY treatments, which exhibited a significantly greater length.

Concerning mycorrhizal colonization in tomato roots infected with PVY and/or treated with Tvd44 at 25 dpi, the data revealed that no mycorrhizal colonization was detected in tomato plants that did not receive the AMF inoculum. On the contrary, the other treatments with AMF inoculum showed varying degrees of mycorrhizal colonization. Different typical mycorrhizal structures were observed in the tomato roots through microscopic examination ( Figure 3 ). Successful colonization was confirmed by light microscopy investigation, which revealed the presence of both arbuscules and intraradical mycelium in the root cortex of the colonized plants ( Figures 3b–d ). Finally, as indicated by our results, PVY-uninfected tomato plants treated only with AMF inoculum exhibited the highest degree of colonization frequency (83.1%).

Figure 4 displays the mean growth parameters of 25 dpi tomato plants in response to the various treatments. More precisely, treating tomato plants with AMF and foliar Tvd44 improved their shoot height, root length, fresh weight (both root and shoot), dry weight (both root and shoot), and leaf number compared to control tomato plants. Meanwhile, AMF root colonization promoted all assessed growth parameters. In contrast, PVY treatment significantly decreases all assessed growth parameters compared to control tomato plants. The colonization with AMF, foliar Tvd44, and dual treatments reduces the adverse impacts of PVY infection on the estimated parameters compared with untreated infected plants. Compared to PVY treatment, the dual+PVY treatment was more efficient than the AMF colonization and foliar Tvd44 alone. The dual+PVY treatment resulted in a significant increase of 180% and 254% in shoot and root fresh weight, respectively; additionally, leaf numbers increased by 185% ( Figure 4 ).

The total nitrogen (TN) content in tomato leaves varied among the different treatments ( Table 2 ). All treatments resulted in increased TN levels compared to the control treatment. The most significant increases in TN content were noted in the dual+PVY treatment at 3.84%, followed by AMF at 3.45% and Tvd44 at 3.39%. AMF colonization reached its peak in total carbon content at 34.95%, closely followed by Tvd44+PVY at 33.95%. Moreover, the majority of the treatments implemented resulted in elevated levels of total hydrogen (TH) and total sulfur (TS). The observed C/N ratios decreased across all treatments compared to the control ( Table 2 ). Figure 5 presents the findings from the Tukey multiple comparison test concerning the effects of AMF and Tvd44 on biochemical traits. The dual+PVY treatment resulted in a significant increase in chlorophyll (a), (b), CAT, and TP compared to the PVY treatment, with enhancements of 191%, 91%, 147%, and 34%, respectively.

The present study assessed the relative expression levels of eleven genes involved in regulating the five key components of the defense pathways, using RT-qPCR at 25 dpi ( Figure 6 ). The analysis revealed that the infection of tomato plants with PVY resulted in a 7.2-fold increase in the expression level of WRKY1 compared to the control group. Furthermore, the treatments AMF+PVY and Tvd44+PVY applications exhibited the highest expression levels of WRKY1, 66.3- and 71.9-fold increase, respectively ( Figure 6 ). The expression levels of WRKY19 exhibited varying degrees of increase in response to PVY, AMF, and Tvd44 treatments compared to the control group. JERF3 exhibited stimulation following infection with PVY, AMF, and Tvd44 treatments compared to control plants. Concurrently, the expression levels of AMF+PVY, Tvd44+PVY, and the dual+PVY treatment increased by 2.9-, 2.9-, and 20.2-fold, respectively ( Figure 6 ). Concerning GST1, the infected plants underwent treatment with AMV and Tvd44, resulting in a significant increase in the transcript level of GST1. Notably, the dual+PVY treatment elicited a more pronounced response than the treatments involving infection alone. The findings from Figure 6 indicate that all the treatments applied led to an increase in the expression levels of three genes: F3H and CHI2 at 25 dpi, with notable enhancement in the AMF+PVY, Tvd44+PVY, and dual+PVY treatment groups. Conversely, the PVY treatment showed no impact on the expression of FLS gene transcripts. In contrast, treatments involving AMF+PVY, Tvd44+PVY, and dual+PVY induction resulted in transcript gene expression levels of 59.9-, 45.7-, and 134.9-fold, respectively. The data presented in Figure 6 indicate no effect on the F3'H expression for the tomato plant treatments utilizing AMF or Tvd44 alone. However, an impact was noted for all applied treatments on the F3H gene expression, particularly in tomato plants infected with PVY, which exhibited a 3.54-fold increase. All applied treatments induce the expression of HQT gene transcripts in the chlorogenic pathway. In C3H, a downregulating effect was noted when treating AMV or Tvd44 individually, in contrast to all treatments infected with PVY, which increased gene transcript expression.

Results from the volcano plots in Figure 6a showed that infected plants with the virus led to significant upregulation of all the studied genes except for GST1, C3H, and FLS. When plants were treated with AMF, as shown in Figure 6b , seven out of eleven genes were significantly upregulated (JERF3, WRKY1, WRKY19, CHS, CHI2, F3H, and FLS), while four genes (GST1, HQT, C3H, and F3’H) were unchanged. In Figure 6c , when treatment Tvd44 was applied, six genes were significantly upregulated (WRKY1, WRKY19, F3H, CHI2, JERF, and GST1), while five genes were unchanged (C3H, FLS, CHS, HQT, and F3’H). Figure 6d revealed that in plants treated with AMF+PVY, eight out of eleven genes were significantly upregulated (JERF3, WRKY1, WRKY19, CHS, CHI2, F3H, GST1, and FLS), while three genes (HQT, C3H, and F3’H) were unchanged. When the treatment Tvd44 was applied before virus infection (Tvd44+PVY), as shown in Figure 6e , six out of eleven genes were significantly upregulated (JERF3, WRKY1, WRKY19, CHI2, F3H, and FLS), while five genes (HQT, C3H, CHS, GST1, and F3’H) were unchanged. The application of dual+PVY treatments resulted in a significant upregulation of all eleven studied genes ( Figure 6f ).

The heatmap in Figure 7 was constructed based on two treatment groups. The first group consists of treatments with PVY, AMF, and Tvd44 against the control, while the second group comprises treatments with AMF+PVY, Tvd44+PVY, and dual+PVY against PVY. The heatmap demonstrates that the eleven studied genes were grouped into four clusters based on their expression. Four genes in the second cluster (C3H, GST1, F3’H, and HQT) exhibited similar expression patterns across all treatments, meaning they responded similarly to each treatment. In contrast, two genes (WRKY1 and CHI2) in the third cluster and three genes (WRKY19, FLS, and F3H in the fourth cluster exhibited different expression patterns at each treatment. For example, in the dual+PVY~PVY treatment, the genes of the third and fourth clusters exhibited different expression levels (represented by different colors), with WRKY1 and F3H being highly expressed, followed by CHI2, FLS, and WRKY19. For the same treatment, the expression of the first cluster genes (JERF and CHS) was similar in their expression (same color). These results highlight the significant role of the third and fourth clusters of genes in response to PVY, compared to the first and second clusters.

Figures 8a, b show that the first two principal components (PCs) account for 87.7% of the total variance, with 69.6% and 18.1%, respectively. The first principal component was represented by leaf number, chlorophyll (a), shoot length, shoot weight (both fresh and dry), and root dry weight, with higher loadings (longer arrows) located in the left area of the biplot. At the same time, the second principal component was represented by CAT, POX, and TP, which had higher loadings and were located in the upper center of the biplot ( Figures 8c, d ). These results illustrated that all traits were suitable for distinguishing between control and PVY treatments, as well as the rest.

Figures 9 , 10 represent the network of eleven gene expressions and eleven physiological traits in response to the seven treatments. The network revealed that all eleven studied genes were significantly correlated, except for gene CH3 with JERF3, CHI2, HQT, C3H, F3’H, and CHI2 with F3’H. The least correlated physiological traits to both genes and other traits were root dry weight, POX, and chlorophyll (b). The most correlated were CAT, TP, and the number of leaves. On the other hand, the gene least correlated with both other genes and physiological traits was F3’H, while the most correlated ones were F3H, FLS, WRKY1, WRKY19, GST1, JERF3, and CHS. The most highly correlated gene with the physiological traits was CHS, which correlated with nine traits, followed by FLS, which correlated with eight traits, and CHI2, WRKY19, and F3H, which correlated with six traits each. GST1 and WRKY1 correlated with five traits, followed by JERF3, which correlated with four traits. Finally, HQT, F3’H, and C3H were the least correlated with the physiological traits, as they were correlated with only three traits. The network identified two local communities; the first community (nodes with blue borders) comprised 10 genes, excluding CHS, and three physiological traits (CAT, POX, and TP). The second community (nodes with green border) contained eight physiological traits, except for CAT, POX, TP, and one gene (CHS).