Peptide Analogs of a Trichoderma Peptaibol Effectively Control Downy Mildew in the Vineyard

Downy mildew causes substantial yield losses and quality deterioration in several crops and ornamental plants worldwide (Chang et al. 2013; Cohen et al. 2017; Keinath and de Figueiredo Silva 2022; Salgado-Salazar et al. 2018a, b; Spring et al. 2018). The causal agents of downy mildew are obligate biotrophic pathogens and rely entirely on the host to complete their life cycle (Spring et al. 2018). They belong to the kingdom Chromista, subphylum Oomycota, family Peronosporaceae (Thines 2014). Downy mildew is mainly caused by oomycetes of the genus Peronospora, counting about 400 species, but other pathogens are members of additional 18 genera, including Pseudoperonospora, Bremia, Plasmopara, Hyaloperonospora, and Sclerospora (Thines and Choi 2016). Among phytopathogenic oomycetes, Plasmopara viticola (Berk. & M. A. Curtis) Berl. & De Toni affects grape production in all viticultural regions around the world and, without protection, causes over 50% losses under disease-favorable conditions (Agrios 2005; Gessler et al. 2011; Leroy et al. 2013).

P. viticola as the causal agent of downy mildew in grapevines was first reported in 1876. Downy mildew is endemic in North America, and it arrived in Europe in 1878 (Gessler et al. 2011) and then started spreading worldwide (Fontaine et al. 2021). Vitis vinifera varieties are highly susceptible to downy mildew. Although some resistant varieties have been released (Pedneault and Provost 2016), the most traditional wine varieties remain highly susceptible (Pertot et al. 2017).

The primary infections occur in the spring and result from zoospores released by zoosporangia differentiated after germination of resting oospores, whereas the secondary infections are caused by zoospores released from sporangia formed on infected tissues (Gobbin et al. 2005; Kennelly et al. 2007). Stomatal openings are essential for zoospore germination, infection (Gessler et al. 2011; Müller-Thurgau 1911), and mycelium sporulation; thus, all vegetative tissues are susceptible even if ontogenic resistance occurs in berries (Kennelly et al. 2005).

Disease management relies on chemical control with repetitive use of traditional copper-based products and/or synthetic fungicides (Agreste 2021; ISTAT 2011). Although these chemical substances effectively reduce disease pressure, they are a threat to the health of vine growers and populations living near the vineyards and contaminate the environment (EUROSTAT 2007; Merz et al. 2015). In addition, the pathogen may become resistant, especially to more recent selective fungicides (Gessler et al. 2011).

European Union (EU) regulations provide for the sustainable use of pesticides by promoting integrated pest management (Directive 2009/128/EC) (European Parliament and Council of the European Union 2009a). The EU adopted stricter criteria for the authorization of plant protection products (Regulation EC 1107/2009) (European Parliament and Council of the European Union 2009b), discontinuing many active substances or including them in the list of candidates for substitution (ec.europa.eu/food/plant/pesticides/approval_active_substances_en). Copper compounds, the main ingredients of fungicides allowed in organic viticulture, are also candidates for substitution, and their application must not exceed 28 kg of copper per hectare over 7 years (Regulation EC 2018/1981) (European Parliament and Council of the European Union 2009b). More recently, environmental and safety concerns have been incorporated into the EU Farm to Fork strategy (European Commission 2020) through the revision of legislation of the Sustainable Use of Pesticides action, which contemplates a 50% reduction in using pesticides by 2030. In this context, identifying and developing alternative molecules, mainly of biological origin, for the control of downy mildew is a research challenge.

Natural compounds derived from plants, animals, or microorganisms may be a source of antifungal molecules for crop protection (Copping and Duke 2007). As an example, Trichoderma spp. produce an array of secondary metabolites that can be used for controlling phytopathogens (Mayo-Prieto et al. 2019; Zeilinger et al. 2016). As natural secondary metabolites, antimicrobial peptides (AMPs) have attracted great attention as candidates for plant protection products and inspired the design of new semisynthetic analogs (Montesinos 2007). Among AMPs, peptaibols produced by Trichoderma spp. have gained interest from the scientific community for their bioactivity (Marik et al. 2019).

Peptaibols are peptides of 8 to 20 residues with nonproteinogenic amino acids that can aggregate, affect cell membrane integrity (Afanasyeva et al. 2019; Milov et al. 2016; Szekeres et al. 2005), and trigger programmed cell death in phytopathogenic fungi, such as Fusarium oxysporum and Botrytis cinerea (Shi et al. 2012; Zhao et al. 2018). Moreover, the well-characterized peptaibols alamethicin and trichokonin are capable of inducing pathogenic resistance in plants (Kredics et al. 2013; Leitgeb et al. 2007; Li et al. 2014).

The peptaibol trichogin GA IV has bactericidal activity and remarkable resistance to proteolysis, but poor water solubility (De Zotti et al. 2009). In our previous work, we produced water-soluble analogs of the short-length peptaibol trichogin GA IV from Trichoderma longibrachiatum by solid-phase synthesis (De Zotti et al. 2020). Some of these peptides have significant fungicidal activity at a concentration lower than 50 μM. Peptides with higher in vitro antifungal activity reduce disease symptoms produced by B. cinerea on grapevine, tomato leaves, and grape berries (Baccelli et al. 2022; De Zotti et al. 2020). Selected peptides protect barley and rice from Pyricularia oryzae infection (Sella et al. 2021). However, few studies addressed in vitro and in vivo biocidal activity of peptaibols against oomycetes of plant pathogens (Lederer et al. 1992; Otto et al. 2016). In this study, we investigated the efficacy of trichogin GA IV analogs in preventing P. viticola infection in grapevine leaf disks and their biocidal activity against sporangia and zoospores. Finally, we present the results of a 2-year field trial to assess the efficacy of one of the most promising peptides in protecting vineyards from downy mildew.

Materials and Methods

Water-soluble analogs of trichogin

Trichogin and its water-soluble analogs were synthesized as previously reported (De Zotti et al. 2020). The peptide sequences used in this study are given in Table 1, and apart from peptide K9r, the other peptides are described in Baccelli et al. (2022) and De Zotti et al. (2020). They are classified into full-length peptides, C-terminal-modified analogs (carrying a C-terminal amide), and shorter analogs.

Table 1. Sequences of the trichogin analogs and efficacy in preventing Plasmopara viticola infection and activity against Botrytis cinerea conidia^u

Peptide ID	Sequencesv	Efficacyw
		P. viticolax	B. cinereay
Full-length peptides
Trichogin GA IV	nOct–Aib–Gly–Leu–Aib–Gly–Gly–Leu–Aib–Gly–Ile–Lol	–	–
1	nOct–Aib–Lys–Leu–Aib–Lys–Gly–Leu–Aib–Gly–Ile–Lol	+	–
2	nOct–Aib–Lys–Leu–Aib–Gly–Gly–Leu–Aib–Gly–Ile–Lol	–	–
3	nOct–Aib–Gly–Leu–Aib–Lys–Gly–Leu–Aib–Gly–Ile–Lol	–	–
4	nOct–Aib–Gly–Leu–Aib–Lys–Lys–Leu–Aib–Gly–Ile–Lol	+	+
5	nOct–Aib–Gly–Leu–Aib–Gly–Lys–Leu–Aib–Gly–Ile–Lol	–	+
6	nOct–Aib–Gly–Leu–Aib–Lys–Aib–Leu–Aib–Gly–Ile–Lol	+	–
7	nOct–Aib–Lys–Leu–Aib–Gly–Lys–Leu–Aib–Gly–Ile–Lol	–	–
K9	nOct–Aib–Gly–Leu–Aib–Gly–Gly–Leu–Aib–Lys–Ile–Lol	+	+
C-terminal-modified analogs
4r	nOct–Aib–Gly–Leu–Aib–Lys–Lys–Leu–Aib–Gly–Ile–Leu–NH2	+	+
6r	nOct–Aib–Gly–Leu–Aib–Lys–Aib–Leu–Aib–Gly–Ile–Leu–NH2	–	nd
K9r	nOct–Aib–Gly–Leu–Aib–Gly–Gly–Leu–Aib–Lys–Ile–Leu–NH2	+	nd
Shorter analogs
4c	nOct–Aib–Lys–Lys–Leu–Aib–Gly–Ile–Lol	+	+
4c1z	nOct–Aib–Gly–Leu–Aib–Lys–Lys–Leu–NH2	–	+
4c2z	nOct–Aib–Lys–Lys–Leu–Aib–Gly–Ile–Leu–NH2	+	+

^u Modified residues are highlighted in bold.

^v nOct, n-octanoyl; Aib, α-aminoisobutyric acid; Lol, leucinol.

^w +, high antimicrobial activity; –, partial or no antimicrobial activity; nd, not determined.

^x Data obtained in this work.

^y Data obtained by in vitro assays on B. cinerea (De Zotti et al. 2020), and the in vitro activity of the K9 peptide has not previously been published.

^z Shorter analogs carrying a C-terminal amide.

View as image HTML

Plant material, P. viticola strain, propagation, and inoculum production

One- to four-year-old potted plants of a white grapevine variety (Vitis vinifera ‘Glera’) grafted onto the Kober 5BB rootstock were used for inoculum production and leaf disk assays (discussed below in Inoculation experiments and peptide treatments). One-year-old shoots of the dormant plants were pruned to three to four buds, and at different times from January to April of the years 2017 to 2021, the plants were transferred into a climatic chamber (20 to 22°C, 16 h photoperiod, and about 75% RH). From about 1 month after bud opening, leaves were harvested as needed. Once the necessary leaves were collected, the plants were placed outdoor. About 20 plants were managed in total.

Sporangia of P. viticola were collected from infected grapevine leaves harvested at the beginning of June 2017 in a vineyard in the municipality of Nervesa della Battaglia located in the Venetian region of Italy (45°49′23″N, 12°12′21″E). The pathogen was maintained by spraying a sporangia suspension on the abaxial surface of detached fresh grapevine leaves on a weekly basis. The leaves were arranged on moist towels in plastic trays and maintained at 22 to 25°C in the dark under humid conditions to allow infection and sporulation. After 6 to 9 days from leaf inoculation, sporangia were collected by washing with sterilized water. In absence of fresh leaves, the vitality of sporangia was preserved by storing air-dried sporulating leaves at −20°C.

Inoculation experiments and peptide treatments

Grapevine leaves, from the fourth to the sixth from the shoot tip, were collected from different plants. Leaf disks of 1.7 cm in diameter were excised by a cork borer and randomly distributed with the adaxial surface facing down on moistened sterile filter papers placed in 15-cm-diameter Petri dishes (20 disks/plate).

Trichogin GA IV 1 mM stock solution was solubilized in 5% (v/v) ethanol. Trichogin and its derivatives (Table 1) were dissolved in water at 50 μM, and about 0.1 ml of each peptide solution was sprayed on the surface of each of the 20 leaf disks with a 20-ml pump atomizer vial amber (Arco Scientifica, Limena, Italy). After drying for 10 min in a laminar flow hood, the leaf disks were sprayed with a sporangia suspension (about 0.75 ml of suspension per 20 disks) containing 4 × 10⁵ sporangia/ml (counted using a hemocytometer), and the plates were incubated in the dark at room temperature (22 to 23°C).

Different treatment sessions were carried out. Each session comprised one or more peptide treatments (one plate per treatment) and spraying a control plate with water only. Twelve days postinoculation (dpi), the disease incidence was calculated in each plate with the following formula: (the number of sporulating leaf disks/the total number of inoculated disks) × 100. Then, the effectiveness of each treatment was calculated with the following formula: ([disease incidence of the control plate − disease incidence of the treated plate]/disease incidence of the control plate) × 100. Overall, two or three plates (replicates) were performed for the less active peptides and at least three or more replicates for the most active peptides.

To compare the effectiveness of peptides 4 and 4r in controlling downy mildew, we also assayed them at a concentration of 15, 20, and 30 μM. To establish the effectiveness of the peptides in comparison with a commercial fungicide, a tribasic copper sulfate fungicide (Tricopperland, Isagro s.p.a., Milan, Italy) was assayed against the pathogen. In a preliminary test, the fungicide administered at a field dosage (7.3 mM) of Cu metal provided complete sporulation inhibition (data not shown). To identify the Cu concentrations capable of giving effectiveness comparable with that of the peptides, the fungicide was assayed at a concentration of 50, 200, and 730 μM of Cu metal. Data were statistically analyzed by applying the one-way ANOVA and Bonferroni–Holm tests.

Microscopic observations

Aliquots of 200 μl of a water suspension containing 5 × 10⁵ sporangia were treated with peptide 4 at 50 μM, and after 15, 30, and 60 min, sporangia were examined using an optical microscope (Laborlux 12, Leitz). At least four optical fields with 50 to 100 sporangia each were analyzed, and sporangia with shape alterations were counted. At the same time points, untreated sporangia (negative control) were also examined and counted for morphological alterations.

A suspension containing 3 × 10⁶ sporangia/ml was incubated for 6 h in sterile deionized water to allow the release of zoospores (Islam et al. 2016). Then, peptide 4 was added, and the zoospores were examined after 15 and 30 min. At least four optical fields containing 40 to 80 zoospores were analyzed, and the lyzed zoospores were counted. Untreated zoospores (negative control) were also analyzed. Each experiment on sporangia and zoospores was repeated three times. Data were statistically analyzed by applying the one-way ANOVA Tukey–Kramer test, considering the treatment as a fixed effect and the experiment as a random effect.

Field trials

Field trials were conducted in 2020 and 2021 in V. vinifera vineyards of Pinot noir and Moscato Bianco in Costigliole d’Asti (AT) (44°45′26.5″N, 8°12′59.3″E) and Vesime (AT) (44°37′4.7″N, 8°12′38.8″E) municipalities of the Piedmont region, Italy, respectively. The plants, grafted onto the Kober 5BB rootstock, were grown in a Guyot training system. The plants were spaced 0.8 m along the row and 4 m (Pinot noir) or 2.5 m (Moscato Bianco) between the rows. A randomized complete block design with four replicates was adopted, each parcel comprising seven plants along the row. The treatments were performed in 2020 with a standard atomizer equipped with a conical nozzle and in 2021 with a motorized backpack sprayer equipped with a five-flat fan nozzle boom, approximately at a weekly time interval, from the BBCH-stages 53 to 75 and from BBCH 15 to 78 in the 2020 and 2021 seasons, respectively.

In both years, the first and second treatments (Table 2) were carried out on the entire experimental fields with a commercial formulation of copper oxychloride (Zetaram Plus, Sipcam Italia s.p.a., Milan, Italy). Then, the successive treatments differed according to the experiment design: the control plots (C1) were treated with water, the peptide plots were treated with peptide 4r, and the fungicide plots were treated with the copper oxychloride fungicide. Peptide 4r was administered at 129.3 g/ha (equivalent to 10³ liter of application volume with the peptide at the concentration of 100 μM; MW 1293) in 2020 and 129.3, 64.7, 32.3, 16.2, and 1.62 g/ha in 2021. The copper oxychloride fungicide was sprayed with 760 and 494 g/ha of pure metal in 2020 and 2021, respectively. The reduced copper dosage (494 g/ha) was aimed at reducing the metal pollution according to EU disposition (Commission Implementing Regulation [EU] 2018/1981) without losing efficacy (Cabús et al. 2017). In 2021, the commercial adjuvant Silwet L-77 AG (Momentive Performance Materials Inc., Waterford, NY) at 0.01% v/v was added as a surfactant to the peptide mixture to improve the uniformity of distribution and, therefore, additional control plots (C2) were included and treated with the adjuvant only. In both years, plots without any treatment were included in the experimental field (C0).

Table 2. Treatment schedule based on the plant development stage

2020		2021
Datez	Plant developmental stage (BBCH)	Datez	Plant developmental stage (BBCH)
12 May	53	8 May	15
20 May	55	15 May	19
27 May	57	22 May	53
1 June	60	29 May	55
10 June	63	6 June	57
15 June	68	13 June	63
22 June	71	20 June	69
30 June	75	25 June	71
		30 June	73
		5 July	75
		11 July	77
		17 July	78

^z In both years, on the first two dates, the plots were sprayed only with the copper oxychloride fungicide; on the remaining dates, each plot was treated according to the experiment design.

View as image HTML

To control powdery mildew (Erysiphe necator), sulfur and tetraconazole were sprayed in 2020, and fluxapyroxad, tetraconazole, metrafenone, cyflufenamid, and boscalid were sprayed in 2021. In both years, one treatment with acetamiprid was performed to control the leafhopper Scaphoideus titanus, the vector of the Flavescence dorée phytoplasma.

The grapevine downy mildew infection in each plot was evaluated on 100 leaves and 50 bunches. In 2020, leaf symptoms were recorded on five dates (May 26; June 11, 19, and 30; and July 13, corresponding to the BBCH-scales 57, 65, 69, 75, and 79, respectively), while symptoms on bunches were scored on July 7 (BBCH 79). In 2021, leaf symptoms were recorded on six dates (June 21 and 30 and July 5, 10, and 20, corresponding to the BBCH-scales 69, 73, 75, 77, and 79, respectively), and bunches were scored on August 5 (BBCH 81). Disease incidence was recorded as a percentage of infected leaves or bunches per total number of leaves or bunches. Disease severity was determined by visual inspection and classification of leaves and bunches according to the following percentage values of symptomatic area: 0% = healthy; 1.25% = 0 to 2.5%; 3.75% = 2.6 to 5%; 7.5% = 5.1 to 10%; 17.5% = 10.1 to 25%; 37.5% = 25.1 to 50%; 62.5% = 50.1 to 75%; 82.5% = 75.1 to 90%; and 95% = 90.1 to 100%.

Disease incidence and severity data recorded in 2021 were also used to calculate areas under the disease progress curve (AUDPC). Data were statistically analyzed by applying a one-way ANOVA technique followed by the Tukey HSD test (P < 0.05).

Results

The in vitro activity of the Trichoderma-derived peptides against P. viticola

When trichogin GA IV and peptides 2, 3, 5, and 7 were assayed on leaf disks at 50 μM, they did not significantly reduce the number of sporulating disks compared with the untreated control. By contrast, peptides 1, 4, 6, and K9 were highly effective as they reduced downy mildew incidence by 80 to 90% (Fig. 1A). At this point, further analyses with the expensive peptide 1 were not carried out while peptides 4, 6, and K9 were also assayed in their Rink version (r), in which the relatively expensive C-terminal Lol moiety was replaced on Rink Amide resin by a Leu-NH2 (leucine amide) residue. Results showed that peptides 4r and K9r were as effective as their parental molecules in preventing P. viticola infection, while peptide 6r was less effective (Fig. 1B). Previously, peptides 4c and 4c2, shorter and cheaper versions of peptides 4 and 4r, respectively, were demonstrated active against the fungus B. cinerea (De Zotti et al. 2020). By comparison, their activity was verified also against the oomycetes of P. viticola, and these shorter peptides appeared slightly less active than their parental peptides. The 4c1 peptide, i.e., a shorter version of peptide 4r, was relatively inactive (Fig. 1B). No phytotoxic effects were detected on leaves after treatment with any peptides (data not shown).

Download as PowerPoint

The effective peptide 4 and its cheaper analog 4r were also compared for their effectiveness in protecting the leaves from infection at doses lower than 50 μM (Fig. 2). At 30 μM, peptides 4 and 4r prevented sporangia production to an extent not significantly different from the 50-μM doses (P < 0.05). Their efficacy decreased by about 52% (peptide 4) or 77% (peptide 4r) at 20 μM. Thus, for both peptides, the half maximal inhibitory concentration (IC₅₀) value was comprised between 20 and 30 μM. To identify a copper concentration capable of giving an effectiveness comparable with that of the peptides, a copper fungicide was assayed. Compared with the untreated control, the copper fungicide used at 730, 200, and 50 μM of copper reduced the disease incidence by 83.2, 40.9, and 14.3%, respectively (Fig. 2). Thus, a similar level of protection (approximately 80% of disease incidence reduction) was obtained with 30 μM of peptides 4 and 4r and 730 μM of copper.

Download as PowerPoint

Microscopic observations

Microscopic observations were carried out on zoosporangia after treatment with peptide 4 at 50 μM. Compared with the untreated sporangia (Fig. 3A), the treated sporangia lost their integrity and released cytoplasmic material after 15 min (Fig. 3B). This effect was more evident after 30 min (Fig. 3C and Table 3), and all sporangia exhibited abnormal morphology after 1 h (not shown).

Download as PowerPoint

Table 3. Percentage of collapsed zoosporangia and zoospores of Plasmopara viticola after the treatment with peptide 4 at 50 μM^z

	Collapsed zoosporangia (%) after 30 min	Collapsed zoospores (%) after 15 min
Control	6.2 ± 3.2 a	7.2 ± 1.96 a
Treated	82.4 ± 19.9 b	87.3 ± 5.15 b

^z Data are the mean ± SD of values obtained from three separated experiments. Each experiment consisted of the counting of four microscopic fields, each containing at least 40 zoosporangia or zoospores. Different letters indicate statistically significant differences according to the Tukey–Kramer test results (P < 0.05).

View as image HTML

The same treatment performed on zoospores confirmed the detrimental effect of peptide 4. Compared with the untreated zoospores (Fig. 3D), the membrane of most of the treated zoospores appeared completely disrupted after 15 min (Fig. 3E and Table 3).

Field trial experiments

Because of the possible industrial production as plant protection products, peptide 4r was chosen for experimental trials conducted in the field in 2020 and 2021 aimed at protecting vineyards from downy mildew. Peptide 4r was administered at 100 μM (129.3 g/ha), considering that the efficacy in the field is usually lower than that detected in the laboratory experiments.

In 2020, mild and frequent rain events throughout the experimental period and maximum temperatures below 30°C until the second 10 days of June (Fig. 4A) were the conditions suitable for downy mildew infection (Fig. 5A and B). Disease incidence increased from the first assessment date (May 26th) to the end of June. (Fig. 5A). Both peptide and copper treatments decreased the disease incidence and severity. However, the disease incidence rates were not statistically different at every assessment date, and at the last evaluation date, only copper was effective in reducing the incidence compared with the C1 control (P < 0.05). The protective effect was remarkable considering the disease severity, which, at the last date, decreased by 53.5 and 57.5% in the treatments with the peptide and copper, respectively (Fig. 5B). Disease incidence and severity values on grape bunches mirrored those detected on leaves. The effect of the peptide treatment on bunches was remarkable, with a 72.8% reduction in severity (Fig. 6).

Download as PowerPoint

Download as PowerPoint

Download as PowerPoint

In 2021, the infection started at the beginning of June (not shown), but the scarcity of rain (Fig. 4B) delayed the restart of the infection until the end of the month (Fig. 5C and D). The effect of the peptide and copper treatments in protecting plants from downy mildew was remarkable, as highlighted by the significant reduction in both disease incidence and severity (Fig. 5C and D). At the last assessment date, compared with their corresponding controls (C1 and C2), copper and peptide treatments significantly reduced the disease incidence by 63.6 and 52.5%, respectively (Fig. 5C). Similarly, the two treatments decreased the disease severity by 84.6 and 73.8%, respectively (Fig. 5D). The treatment with the adjuvant alone (C2) displayed disease incidence and severity values not significantly different from those of the C1 control sprayed with water (Fig. 5C and D). On bunches, both incidence and severity were markedly reduced (78.0 and 94.7%, respectively). These values were comparable with those detected with the copper treatment (Fig. 6A and B). In 2021, treatments with concentrations of peptide 4r lower than 129.3 g/ha were also included in the trial. To summarize the effect of the peptide dosage on disease incidence and severity, AUDPC values were calculated (Table 4). Although a reduction in the protective effect was observed by decreasing the peptide concentration, the protection levels on both leaves and bunches obtained with the peptide at 64.7 g/ha were similar to those obtained with 129.3 g/ha of peptide, or with the copper fungicide. At 32.3 g/ha, the protection against the disease was comparable with that of the higher peptide doses, except for a significant increase in disease incidence on bunches. At 16.2 g/ha, the protection significantly decreased compared with the higher doses of the peptide, but moderate protection was still evident as the AUDPC values were significantly lower than those recorded on the C2 control plants (Table 4).

Table 4. Downy mildew area under the disease progress curve (AUDPC) values determined in the 2021 field trial^w

Treatment	Leaves		Bunches
	Incidence AUDPC (% days)x	Severity AUDPC (% days)x	Incidence AUDPC (% days)y	Severity AUDPC (% days)y
Untreated control (C0)	1,275 ± 106 a	355 ± 79 a	1,491 ± 113 a	466 ± 156 a
Partially untreated control (C1)z	934 ± 110 ab	271 ± 55 ab	1,225 ± 14 ab	413 ± 47 a
Adjuvant 0.01% (C2)	870 ± 153 bc	200 ± 26 bc	1,148 ± 46 abc	379 ± 70 ab
Copper (494 g/ha)	152 ± 58 f	18 ± 7 e	175 ± 27 f	10 ± 5 d
Peptide 4r 129.3 g/ha	319 ± 96 def	63 ± 29 de	231 ± 35 f	20 ± 11 d
Peptide 4r 64.7 g/ha	377 ± 97 def	80 ± 22 de	315 ± 42 ef	27 ± 17 d
Peptide 4r 32.3 g/ha	450 ± 62 def	119 ± 28 cde	441 ± 58 de	96 ± 47 cd
Peptide 4r 16.2 g/ha	577 ± 165 cde	135 ± 4 cd	511 ± 35 cde	156 ± 25 cd
Peptide 4r 1.62 g/ha	670 ± 193 bcd	161 ± 33 bcd	609.0 ± 48 bcd	219 ± 38 bc

^w Different amounts of peptide 4r were compared with the cupric fungicide treatment and three different controls (C0, C1, and C2).

^x Calculated on 100 leaves per plot sampled for 29 days (from June 21 to July 20). Means followed by the same letters in the same column are not significantly different by the Tukey HSD test (P < 0.01).

^y Calculated on 50 bunches per plot sampled for 28 days (from July 8 to August 5). Means followed by the same letters in the same column are not significantly different by the Tukey HSD test (P < 0.01).

^z Only the two initial treatments with the cupric fungicide were performed (see Table 2).

View as image HTML

It is important to point out that no treatments gave toxicity symptoms on leaves or bunches.

Discussion

To identify new bio-inspired molecules capable of preventing downy mildew infection, we explored the capacity of trichogin-derived peptides to control the grapevine pathogen P. viticola. Trichogin GA IV is a poorly amphipathic 11-residue peptaibol produced by T. longibrachiatum with a helical structure capable of interacting with microbial membranes and has a remarkable resistance to proteolytic degradation conferred by three α-aminoisobutyric acid (Aib) residues (Yamaguchi et al. 2003). However, trichogin has poor solubility in water and is inactive against fungi (De Zotti et al. 2009, 2020; Sella et al. 2021). Similarly, here we report that trichogin is inactive against the oomycete species P. viticola since it does not prevent sporulation on grapevine leaf disks. Conversely, some Gly-to-Lys substitutions in the trichogin sequence provide cationic properties, higher solubility in water, and the ability to prevent P. viticola infections.

Four of the eight full-length peptides (1, 4, 6, and K9) administered at 50 μM effectively prevented the grapevine leaf infection, while the remaining four peptides (2, 3, 5, and 7) did not significantly reduce the infection rate. The number of Lys substitutions alone does not explain the difference in activity, as a single Lys substitution is present in the effective peptides 6 and K9 and in the ineffective peptides 2, 3, and 5, and double Lys substitutions are present in the effective peptides 1 and 4 and also in the ineffective peptide 7 (Table 1).

The C-terminal amide modification of the active peptides 4, 6, and K9 and of the shorter version of peptide 4 reduces the cost of synthesis. The observation that the full-length 4r and K9r peptides, as well as the shorter versions 4c and 4c2, are active against P. viticola while the 6r peptide is only partially active and the shorter version 4c1 is not active, demonstrates that modification or shortening of the N-terminal part of the sequence may also influence the activity of the peptides. Thus, the cheaper peptides 4r, 4c, and 4c2, previously demonstrated active against the fungus B. cinerea (De Zotti et al. 2020), are interesting candidates for their excellent activity against P. viticola as well.

A comparison of peptide activity against B. cinerea (De Zotti et al. 2020) and P. viticola (this work) as a function of the Lys number and position reveals some similarities and differences between the results on these two pathogens. Among the peptides with a single Lys substitution, peptides 2 (Lys²) and 3 (Lys⁵) are poorly or not effective against both P. viticola and B. cinerea (Table 1), while peptide K9 (Lys⁹) is active against both pathogens. Moreover, peptides 5 (Lys⁶) and 6 (Lys⁵ substitution with an additional Aib in position 6) were effective against B. cinerea or P. viticola, respectively (Table 1). Among the peptides with double Lys substitutions, peptides 4 and 4r (Lys⁵ and Lys⁶) and peptide 7 (Lys² and Lys⁶) behave similarly against both pathogens, being active and inactive, respectively, while peptide 1 (Lys² and Lys⁵) is active only against P. viticola (Table 1). Thus, in some cases, the position of the Lys residues seems to trigger the selectivity of the peptides toward the two pathogens.

The relative position of the two Lys residues in the 3D structure of peptides 1, 4, and 7 does not help explain their activity against P. viticola. Previously, the spatial vicinity between Lys residues was argued as being responsible for diminished activity against the necrotrophic fungus B. cinerea (De Zotti et al. 2020), and this explanation helped justify the ineffectiveness of peptide 1 and the poor activity of peptide 7 against B. cinerea. However, this behavior does not fit with P. viticola, as peptide 1 affects zoospore infection. Capability of self-assembling, forming pores on the pathogen cell membrane, and the membrane composition may play an essential role in the effectiveness of the peptides. For instance, in model liposomes the presence of cholesterol stabilizes the self-assembly of the transmembrane peptide. In fact, it was previously reported that trichogin insertion in cholesterol-containing membranes is accompanied by self-aggregation of parallelly aligned transmembrane peptide molecules, while in cholesterol-lacking membranes, the peptides are monomolecularly distributed (Syryamina et al. 2012). The membrane of the oomycetes of P. viticola differs from that of fungi. Indeed, sterols are absent in the membrane of oomycetes, while ergosterol, the sterol found in fungal membranes, is responsible for fungal membrane fluidity and resistance to stress (Bloch 1983; Wise et al. 2014). Thus, we assume that the difference in membrane lipid composition between fungi and oomycetes may affect the activity of some peptides. Microscopic observations also point to a different interaction of peptides with the plasma membrane of fungi or oomycetes. After treatment with peptide 4, the protoplast of B. cinerea conidia shrinks and the membrane detaches from the cell wall (De Zotti et al. 2020), while sporangia and zoospores of P. viticola lyze. Finally, since we assayed the peptide activity on B. cinerea spore germination in vitro and against P. viticola on leaf disks, possible interference with the grapevine leaf surface can affect the availability of some peptides to interact with P. viticola zoospores. Further investigation is needed to address these topics.

The 4 and 4r peptides have previously been shown to be equally effective in protecting plant leaves from the fungi B. cinerea and P. oryzae (De Zotti et al. 2020; Sella et al. 2021), with a negligible impact on the leaf metabolism (Baccelli et al. 2022). Particularly, peptide 4r did not induce reactive oxygen species (ROS) production in tomato or Arabidopsis leaves, whereas ROS production induced by peptide 4 did not result in a significant modulation of plant defense genes (Baccelli et al. 2022), corroborating the observation that trichogin analogs are not phytotoxic to plants.

In the grapevine leaf disk assay, peptide 4r has been confirmed as effective as peptide 4 also against P. viticola, and both peptides provide a 100% reduction in the disease incidence when used at 50 μM. The cost of synthesis of peptide 4r is lower than that of the parental peptide 4; for this reason, peptide 4r was selected in the field experiments.

To establish a suitable amount of peptide 4r to be used in the field trial, we compared different dosages of the peptide with those of a cupric fungicide in the leaf disk assay. A similar level of protection was obtained with about 39 mg/liter of peptide 4r (30 μM) and 46 mg/liter of copper (730 μM).

In field practice, the recommended copper concentration for vineyard protection is between 0.5 and 1 g/liter (Cabús et al. 2017), which is higher than the concentration that proved effective in our leaf disk assay. However, these high values are recommended considering the low solubility and gradual release of copper ions in the water film wetting the leaf surface and the need to ensure the persistence of the active ingredient in case of rain wash-off events.

Considering that peptide 4r is soluble in water and may also be subjected to rain wash-off, a dose higher than that effective in the leaf disk assay (i.e., 129.3 mg/liter against 64.7 mg/liter) was used in field experiments. In this sense, it should be pointed out that the dose of fungicides recommended for plant treatments in the field is higher than the minimum inhibitory concentration determined in vitro (Andrieu et al. 2001).

The field trials conducted in 2020 and 2021 underwent different climatic conditions influencing the onset and course of infections. In 2020, the climatic conditions were more favorable for P. viticola infections, as highlighted by both disease incidence and severity values in the untreated control plots (C0). As expected, in both years on the first survey dates, the disease level detected on leaves of the C1 and C2 (C2, present only in 2021) control plants was lower than that recorded in the untreated plants (C0) because of the two copper treatments carried out at the beginning of the season. Those treatments delayed the rise of downy mildew symptoms. Later, the differences diminished, and at the end of the experiment, the disease levels in the C0, C1, and C2 plants were similar (Fig. 5).

In both field trials, peptide 4r significantly reduced the disease severity on leaves and bunches compared with the corresponding controls (C1 or C2). The effectiveness and duration of protection were remarkable and comparable with those obtained with a cupric fungicide with the same administration frequency. An uneven distribution of the peptide on the vine canopy, possibly determined by the absence of adjuvants, may explain the slightly higher disease incidence observed in 2020 by the peptide treatment compared with the cupric treatment.

In the second-year field trial, the results showed that even lower doses of the peptide could effectively contain the disease. This encouraging result highlights that there is room for reducing the cost of the treatment while achieving an excellent protection level by reducing the dosage or improving the peptide formulation.

The trichogin analogs increase the availability of effective bio-inspired linear AMPs. So far, AMPs have been mainly tested to contain some widespread bacterial plant diseases (Baró et al. 2020; Cabrefiga and Montesinos 2017; Mariz-Ponte et al. 2021; Mendes et al. 2021). Increasing the efficacy of the trichogin analogs against some important plant bacterial diseases is the goal of one of our current studies.

In conclusion, water solubility, persistence, absence of phytotoxicity, and excellent fungicidal activity make peptide 4r an interesting new molecule for controlling filamentous pathogens in viticulture. If this peptide meets the approval criteria of Regulation (EC) No. 1107/2009, it will be necessary to consider the issues related to the production and development costs and effective formulations.