Trichogin GA IV-derived peptides (Table 1) were produced by solid-phase peptide synthesis as described in De Zotti et al. (2020). The C-terminal Leucinol was obtained by using a 2-chlorotrytil resin preloaded with the 1,2-amino alcohol. A yield of about 70% on the purified peptides (purity > 95%) was achieved for all analogs. Pep K9 and Pep 6r were designed and synthesized in the present study. Details on the design and synthesis of Pep 4, Pep 4r, Pep 5, and Pep 6 are provided in De Zotti et al. (2020). C-terminal amide peptides were produced on a Rink-amide resin. Purity was checked by analytical HPLC on a C18 column (Phenomenex Jupiter 5 μm, 300 Å, 250 × 4.5 mm). The gradient used was 60%–100% B in 30 min; eluent A H₂O/CH₃CN 9:1 with 0.05% trifluoroacetic acid (TFA); eluent B CH₃CN/H₂O 9:1 with 0.05% TFA. HPLC retention times were used to evaluate hydrophilicity. Helix stability was evaluated by combining literature computational considerations (Chou and Fasman, 1978; Lyu et al., 1991), and circular dichroism analysis performed on Trichogin analogs in solvents of different polarity (De Zotti et al., 2012a).

Tomato (Solanum lycopersicum) seeds belonging to the cultivars Marmande and Micro-Tom were purchased from two specialty stores: Cooperativa Agricola Legnaia (Florence, Italy) and Gargini Sementi (Lucca, Italy), respectively. Plants were sown in Jiffy-7 peat pellets and transferred into commercial soil 2 weeks after germination. Plants were grown on a bench in a growth room under LED lights “Toplight Greenhouse Plus—Natural” (35% Blue, 17% Green; and 48% Red; C-LED srl, Imola, Italy) with 12 h of light (650 μmol m² s⁻¹), at 28°C (day)/21°C (night), until treatment.

For ROS detection, Arabidopsis thaliana ecotype Columbia-0 (Col-0) plants were grown in Jiffy-7 peat pellets in a dedicated growth chamber at 21°C (day)/18°C (night), 12 h of light (100 μmol m² s⁻¹), for 5 weeks.

Botrytis cinerea strain B05.10 (Staats and van Kan, 2012) was grown on potato dextrose agar (PDA for microbiology, VWR International srl, Italy) at 21°C (day)/18°C (night), with 12 h of light (100 μmol m² s⁻¹) to stimulate and standardize conidia production. These conditions were obtained in the growth chamber used for A. thaliana growth, where detached tomato leaves and whole plantlets were also incubated after peptide treatment and during infections. Botrytis cinerea cultures were maintained under these conditions until conidia collection generally obtained after 10–15 days.

Treatments with Trichogin GA IV-derived peptides were performed on either detached leaves or whole plantlets, as described below. Peptide treatments for ROS detection were performed as described in the appropriate paragraph.

To test the efficacy of the peptides in preventing B. cinerea infection, 10–12 mature and healthy leaves were cut from different Micro-Tom or Marmande plants and placed in 90 mm-Petri dishes containing a filter paper disc wet with 1 ml sterile water, to ensure high internal relative humidity (RH) conditions during incubation. Leaves were sprayed with a 10-ml pump atomizer on the lower (abaxial) leaf surface with 50 μM peptide solution (approximately 300 μl per leaf), or sterile distilled water as a control. Peptides at the concentration of 100 μM were tested only on cv. Marmande leaves, according to the same procedure. In most cases, peptides were sprayed as peptide solution in water, without any addition of adjuvant. However, the use of the commercial surfactant Silwet L-77 AG (Momentive Performance Materials Inc., NY, United States) was assessed in some tests, as specified below and in the result section, with the aim of enhancing the biological activity of the peptides. The plates were sealed with parafilm and incubated in the growth chamber at 21°C (day)/18°C (night), with 12 h of light (100 μmol m² s⁻¹) until pathogen inoculation. Inoculations were performed 48 or 120 h after treatment, as described in the next paragraph.

Tests for curative efficacy, i.e., the ability of the peptides to stop ongoing infections, were performed by spraying detached leaves with 50 μM peptides 24 h after B. cinerea inoculation (see next paragraph). In this case, Micro-Tom or Marmande infected leaves were sprayed with a peptide solution containing 0.01% v/v Silwet L-77 AG.

To demonstrate the efficacy of Pep 4 to prevent B. cinerea infection on whole tomato plants, six 2-week-old Micro-Tom plants were sprayed on the adaxial leaf surfaces with 50 μM Pep 4 (approximately 700 μl per plant) containing 0.01% v/v Silwet L-77 AG. Six control plants were sprayed with sterile distilled water containing 0.01% v/v Silwet L-77 AG. After treatment, plants were incubated for 48 h at 21°C (day)/18°C (night), 12 h of light (100 μmol m² s⁻¹) until infection.

For gene expression studies, 30 2-week-old Micro-Tom plants were used to investigate the ability of the peptide to induce or prime defense gene expression against B. cinerea. In this case, 15 plants were sprayed with 50 μM Pep 4 containing 0.01% v/v Silwet L-77 AG, and 15 plants were sprayed with water containing 0.01% v/v Silwet L-77 AG as a control. Plants were incubated for 48 h as described above, then three control and three Pep 4-treated plants were sampled for RNA extraction (see appropriate paragraph). The remaining plants were inoculated (see next paragraph) or mock treated with potato dextrose broth (PDB, Laboratorios Conda S.A., Spain) yielding the following samples for gene expression studies: control + mock at 6 and 24 h post infection (hpi; i.e., treated with water 48 h before inoculation and then inoculated with PDB and collected at 6 or 24 hpi); control + B. cinerea at 6 and 24 hpi (i.e., treated with water 48 h before inoculation and then inoculated with B. cinerea and collected at 6 or 24 hpi); Pep 4 + mock at 6 and 24 hpi (i.e., treated with Pep 48 h before inoculation and then inoculated with PDB and collected at 6 or 24 hpi); Pep 4 + B. cinerea at 6 and 24 hpi (i.e., treated with Pep 48 h before inoculation and then inoculated with B. cinerea and collected at 6 or 24 hpi). At the end of each sampling time, all true leaves from each plant were frozen in liquid nitrogen and stored at −80°C until RNA extraction.

All treatments and inoculations in this study were performed 2–3 h after lights in the growth chamber were on. Inoculations on detached leaves were performed by applying a single 10-μl droplet of B. cinerea B05.10 conidial suspension (1 × 10⁶ conidia/ml in PDB) on a side of the midrib. In a few cases, bigger leaves were inoculated with two droplets applied on each side of the midrib. Inoculations were performed 48 h after peptide treatment either on the previously treated leaf surface or on the opposite untreated surface, in order to highlight translaminar or resistance-inducing activity (see Results). Inoculations 120 h after treatment were also performed on Pep 4- and Pep 4r-treated leaves in order to assess the duration of the protection. For curative efficacy, untreated healthy leaves were inoculated on the abaxial leaf surface 24 h before peptide treatment. Infected leaves were sealed in Petri dishes containing 1 ml water as described above to maintain high RH conditions needed for infection development, and incubated at 21°C (day)/18°C (night), 12 h of light (100 μmol m² s⁻¹). Disease severity was determined 3 days after infection by measuring the necrotic lesion diameter.

Two-week-old Micro-Tom plants were inoculated by spraying adaxial leaf surfaces with a conidial suspension containing 1 × 10⁶ conidia/ml in PDB. Approximately, 800 μl of conidial suspension were sprayed on each plant. For gene expression studies, that amount was decreased to approximately 600 μl per plant, while mock plants were sprayed with the same amount of PDB. Plants were incubated at 21°C (day)/18°C (night), with 12 h of light (100 μmol m² s⁻¹), in closed transparent boxes containing free water on the bottom to ensure high RH conditions needed for infection development. Disease severity was determined after 3 days by classifying each plant on the basis of the visible leaf necrosis, as follows: Class I: no or point-limited necrosis; Class II: spreading lesions affecting <50% of the leaf surface; Class III: spreading lesions affecting >50% of the leaf surface; and Class IV: extensive necrosis and plant collapse.

Concerning the plants used for gene expression studies, disease severity was determined at the end of the experiment (24 hpi) by counting the number of necrotic spots visible on the leaf surface.

Total RNA was extracted from 80 to 100 mg of ground leaf material by using RNeasy Plant Mini Kit (Qiagen, Italy), with guanidine thiocyanate extraction buffer RLT. RNA was quantified in a Qubit fluorometer (Thermo Fisher Scientific, MA, United States) by using Qubit RNA BR Assay Kit (Life Technologies, Italy). RNA integrity was evaluated by agarose gel electrophoresis. RNA samples (400 ng each) were treated with Amplification Grade DNase I (Sigma-Aldrich) and reverse-transcribed into cDNA by using Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Fisher Scientific, MA, United States). qPCRs were performed in a StepOne Real-Time PCR System (Applied Biosystems). Reactions (20 μl) were carried out with 20 ng of cDNA, 250 nM of each primer, and Fast SYBR Green Master Mix (Applied Biosystems) by setting the recommended thermal-cycling conditions for the fast mode (Fast SYBR® Green Master Mix Protocol, Applied Biosystems). The following genes were analyzed: Proteinase inhibitor II (PIN2), Beta-1,3-glucanase A (GluA), Pathogenesis-related protein 1 (PR1), 1-aminocyclopropane-1-carboxylate oxidase 1 (ACO1), Pathogenesis-related genes transcriptional activator (PTI5), Lipoxygenase A (Lox1.1), and Polygalacturonase inhibitor protein (PGIP). Relative gene expression values were calculated by 2^-ΔΔCt method as described in Livak and Schmittgen (2001). Actin-7 was selected as the endogenous reference gene after comparing its transcriptional stability against three candidate reference genes: Ubiquitin, Tubulin, and Glyceraldehyde-3-phosphate dehydrogenase (GPDH; Supplementary Figure S1). Transcriptional stability was tested in 10 cDNA samples representative of all the experimental conditions analyzed. Before performing relative expression calculations, the performance of each amplification run was checked by melting curve analysis, and the amplification plots were compared to ensure that amplification efficiencies between target and reference genes were approximately equal (data not shown; Schmittgen and Livak, 2008). The list of genes analyzed in this study and their primer sequences are provided in Supplementary Table S1. The primer sequences were designed in the present study, with the exception of ACO1, Actin-7, Ubiquitin, and GPDH (Mannucci et al., 2022).

Trichogin GA IV-derived peptides at the concentration of 50 μM in water were applied as 10-μl droplets on the abaxial surface of detached Micro-Tom and A. thaliana leaves. Five droplets were applied on each leaf, and three leaves per treatment were analyzed. Incubations were performed in Petri dishes for 24 and 48 h as previously described. ROS production by leaves was visualized by staining with the cell-permeable probe 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA; Sigma-Aldrich, United States), as reported in Luti et al. (2020). For staining, leaves were soaked for 1 h in a 10 μM H₂DCFDA solution in 20 mM phosphate buffer, pH 6.8. After H₂DCFDA incubation, leaves were washed twice with phosphate buffer and mounted on glass slides for microscopy analysis, which was performed under a confocal Leica TCS SP5 scanning microscope (Leica, Germany; λex 460 and λem 512 nm).

The efficacy of different peptide treatments against B. cinerea infection on detached leaves was analyzed by one-way ANOVA and Tukey–Kramer multiple comparison post-test when p < 0.05. Efficacies between two peptide concentrations, or between control and treated leaves, were analyzed by unpaired t-test (p < 0.05). Gene expression data were analyzed by unpaired t-test by comparing to the appropriate calibrator sample (control 48 hpt or control + mock 6 and 24 hpi) as detailed in the figure legends (p < 0.05). Analyses were performed in GraphPad software (GraphPad Software Inc., CA, United States). Disease severity was evaluated by Fisher’s exact test (p < 0.05).

Following a published protocol (De Zotti et al., 2020), we synthesized six Trichogin GA IV-derived peptides with improved water solubility and/or conformational stability (Table 1). The analogs were characterized by one to three amino acid substitutions in the native Trichogin GA IV sequence. Gly to Lys substitutions were performed, with positional differences, in all peptide variants to confer water solubility (Table 1). The Gly-Gly dipeptide in the middle of the native sequence of Trichogin GA IV forms a kink in the 3D-structure, which lessens its helical content (Venanzi et al., 2009). Accordingly, being a better helix promoter than Gly (Chou and Fasman, 1978), Lys was also able to stabilize the peptide structure when placed in that kink. Similarly, the Gly-to-Aib substitution in that kink (Pep 6 and Pep 6r) further rigidified the helix, since Aib is a strong helix-inducer (Toniolo et al., 2001b). Finally, by replacing Gly with Lys at position 9 (Pep K9), i.e., far from that kink, we aimed to selectively evaluate the enhancement of hydrophilicity against helix stabilization effects shown by the other variants (position 9 is already involved in a strong helical segment, see Toniolo et al., 1994). The C-terminal leucine amide in the two most helical peptide analogs (Pep 4r and Pep 6r) introduced an additional hydrogen bond, further strengthening the helix (Table 1).

The efficacy of the six Trichogin-derived peptides (Pep 4, Pep 4r, Pep K9, Pep 5, Pep 6, and Pep 6r, Table 1) against B. cinerea infections on tomato was firstly tested on cv. Marmande leaves. Leaves were sprayed with a 50 μM peptide solution 48 h before point inoculating B. cinerea conidia. In a first test, both treatments and infections were performed on the abaxial surface of detached leaves. The peptides Pep 4, Pep 4r, and Pep K9 turned out to reduce significantly disease severity compared to control, as determined by necrotic lesion measurements (Figure 1A) and no efficacy increase occurred when leaves were treated with 100 μM peptide solutions (Supplementary Figure S2). Neither 50 nor 100 μM spray treatments caused visible phytotoxic effects on leaves (data not shown), and 50 μM was selected as a suitable concentration for the following assays.

To get clues on the ability of the peptides to act as penetrating fungicides or induce localized resistance, the peptides were sprayed on the abaxial leaf surface and the pathogen was inoculated 48 h later on the opposite (adaxial) untreated leaf surface. In this case, the treatments were not effective in restricting lesion development compared to control (Figure 1B). The surfactant Silwet L-77 AG, which was used as an adjuvant in this experiment, did not interfere with fungal infection compared to the untreated control (Figure 1B). The protective effect of peptide treatment was also tested on a different tomato cultivar (Micro-Tom; Figure 2). Again, while Pep 4, Pep 4r, and Pep K9 turned out to be effective as protectants against B. cinerea (Figure 2A), the peptides did not reduce significantly disease progression when the pathogen was being inoculated on the opposite, untreated leaf surface (Figure 2B). The duration of the protection was investigated by treating leaves with Pep 4 and Pep 4r and inoculating the fungus 5 days later. As shown in Figure 3, the peptides were still effective in reducing lesion diameters 5 days after the peptide treatments.

Finally, to ascertain whether the protective effect detected on detached leaves occurred on whole tomato plants, 2-week-old Micro-Tom plants were sprayed with 50 μM Pep 4 containing 0.01% Silwet L-77 AG to promote leaf adhesion, and inoculated 48 h later by spraying B. cinerea conidia. As visible in Figure 4, Pep 4-treated plants displayed reduced leaf necrotic symptoms in comparison to control plants.

Curative efficacy against B. cinerea infection was investigated on both tomato cultivars, Marmande and Micro-Tom. Leaves were treated with the peptides 24 h after point inoculation of B. cinerea conidia, i.e., when infection symptoms beneath the inoculation droplet were already visible (not shown). Pep 4, Pep 4r, and Pep K9 at 50 μM with 0.01% Silwet L-77 AG were ineffective in reducing disease severity compared to control on either cultivar (Figures 5A,B), indicating that Trichogin GA IV-derived peptides do not possess any curative efficacy against B. cinerea infections.

To understand more about their mode of action, peptides Pep 4, Pep 4r, and Pep K9 (all showing efficacy against B. cinerea), and Pep 6 and Pep 6r for comparison, were applied to healthy Micro-Tom and A. thaliana leaves. ROS were determined by fluorescence microscopy with the H₂DCFDA probe. As visible in Figure 6A, only Pep 4-treated leaves showed ROS production on tomato leaves after 24 and 48 h. Fluorescent-stained ROS were mostly confined beneath the point of droplet application with no apparent spread to the surrounding zones. A similar result was obtained on A. thaliana leaves, although in that case both Pep 4 and Pep K9 induced ROS production on leaves (Figure 6B).

The induction of ROS production prompted us to investigate the ability of Pep 4 to prime defenses. To do so, a gene expression analysis was designed. The following defense-related genes were selected: the ethylene (ET) biosynthesis gene 1-aminocyclopropane-1-carboxylate oxidase 1 (ACO1), the jasmonic acid (JA)-related gene Lipoxygenase A (Lox1.1), the salicylic acid (SA)-marker gene Pathogenesis-related protein 1 (PR1), and some defense genes known to be affected during B. cinerea infection in tomato, such as Pathogenesis-related genes transcriptional activator 5 (PTI5), Polygalacturonase-inhibiting protein (PGIP), Proteinase inhibitor II (PIN2), and Beta-1,3-glucanase A (GluA; Harel et al., 2014; Sun et al., 2017; Gupta et al., 2021). Gene expression was analyzed at three different time points in order to investigate whether Pep 4 could directly induce defenses ahead of infection (48 hpt), or rather prime them for earlier/stronger induction during infection (6 and 24 hpi; Mauch-Mani et al., 2017). At each time point, control and mock-inoculated plants were collected for RNA extraction. Before collecting 24 hpi samples, the effectiveness of the treatment was verified and confirmed (Supplementary Figure S3).

Pep 4 did not significantly upregulate defense genes at 48 h post treatment (Figure 7A). The same held true at 6 and 24 hpi in mock-treated plants (i.e., 54 and 72 h after initial Pep 4 treatment; Figures 7B,C). The only exception was represented by a moderate down-regulation detectable for GluA in Pep 4 + mock plants after 24 hpi (i.e., 72 h after initial Pep 4 treatment; Figure 7C). In contrast, B. cinerea infection markedly altered gene expression: at 6 hpi, mRNA levels for Pti5, ACO1, and PR1 genes were significantly higher (≥3-fold) in infected plants not treated with Pep 4 (Figure 7B). At 24 hpi, the same three genes turned out to be strongly up-regulated (>20-fold; Figure 7C), and all the genes analyzed were modulated: PIN2 was significantly upregulated only in control + infection plants; GluA, Pti5, PGIP, ACO1, and PR1 were upregulated in both control + infection and Pep 4 + infection plants, and their relative expression reached a similar fold-change value in control + infection and Pep 4 + infection; Lox1.1 was instead strongly down-regulated in both control + infection and Pep 4 + infection plants (>50-fold; Figure 7C). Overall, these results suggested that Pep 4 treatments did not significantly induce defense gene expression nor did they enhance the defense response of the plant during infection.