Photodynamic inactivation of Botrytis cinerea by an anionic porphyrin: an alternative pest management of grapevine

Botrytis cinerea is a necrotic plant fungus that causes gray mold disease in over 200 crops, including grapevine. Due to its genetic plasticity, this fungus presents strong resistance to many fungicides. Thus, new strategies against B. cinerea are urgently needed. In this context, antimicrobial photodynamic treatment (APDT) was considered. APDT involves the use of a photosensitizer that generates reactive oxygen species upon illumination with white light. Tetra-4-sulfonatophenyl porphyrin tetra-ammonium (TPPS) was tested on B. cinerea using light. 1.5 µM TPPS completely inhibited mycelial growth. TPPS (12.5 µM) was tested on three grapevine clones from Chardonnay, Merlot and Sauvignon, grown in vitro for 2 months. Treated root apparatus of the three backgrounds increased thiol production as a molecular protection against photoactivated TPPS, leading to a normal phenotype as compared with control plantlets. Finally, 2-month-old grapevine leaves were infected with 4-day-old mycelium of B. cinerea pre-incubated or not with TPPS. The pre-treated mycelium was unable to infect the detached leaves of any of the three grapevine varieties after 72 h growth when subjected to a 16 h photoperiod, contrary to untreated mycelium. These results suggest a strong potential of photo-treatment against B. cinerea mycelium for future agricultural practices in vineyard or other cultures.


Scientific Reports
| (2020) 10:17438 | https://doi.org/10.1038/s41598-020-74427-9 www.nature.com/scientificreports/ on both 14-day-old Arabidopsis and tomato plantlets. Nevertheless, while Arabidopsis plantlets were killed, tomato plantlets could be rescued after a 14 day-treatment 28 . Surprisingly, the anionic porphyrin tetra-4-sulfonatophenylporphyrin tetra-ammonium (TPPS) did not provoke any harmful effect on both plantlets even at concentrations as high as 50 µM 27,28 . With the aim to develop APDT for agriculture applications, TPPS could be a good candidate because of its low toxicity for plants. Moreover, TPPS remains negatively charged in many chemical environments even under acidic pH and does not aggregate in solution, allowing it to easily permeate cells through cell walls and membranes 29,30 . Therefore, we hypothesized that TPPS could be a good PS candidate to kill the plant pathogen B. cinerea and has potential to be a safe option for grapevine (Vitis vinifera L.) explants. B. cinerea is frequently responsible for drastic reductions in crop yields at harvest and for reducing wine quality 31,32 . This fungus displays very strong resistance to many fungicides, due to its genetic plasticity which confers its diversity in morphology, mycelial growth, sporulation and virulence [33][34][35][36][37] . For the Integrated Pest management, the grapevine's susceptibility to B. cinerea can be considered an essential management indicator [38][39][40][41][42][43] . According to the classification proposed by Dubos 42 and Fermaud et al. 43 , Chardonnay and Sauvignon are highly susceptible to B. cinerea infection whereas the Merlot variety is more resistant. Furthermore, these three varieties are listed in the top 10 most cultivated and famous grapevines for wine production in the world 44 .
Therefore, as a first and necessary step, TPPS was tested separately on the three grapevine backgrounds and on B. cinerea mycelium with the aim of killing the plant pathogen without affecting plantlet growth and development. As a second and final step, grapevine detached leaves infected with B. cinerea mycelium were tested with and without TPPS pre-treatment.

Results
Mycelium growth. The effect of three increasing concentrations of TPPS was monitored on B. cinerea mycelium growth under dark and light conditions as described in the material and methods section. As the four curves obtained with or without TPPS in the dark coincided within the standard deviation, we confirmed that, by itself, TPPS did not exhibit any cyto-and/or genotoxic activity against the fungus (Fig. 1a). Under light and at a low TPPS concentration (1.5 µM), mycelium growth was completely inhibited (Fig. 1b). Furthermore, 1 µM TPPS significantly slowed down mycelium growth under light; however, after a 7 day-culture, the colony reached a size similar to that of the control (Fig. 1b). Therefore, 1.5 µM of photoactivated TPPS has been chosen as the minimum fungicidal concentration (MFC). As shown by the growth curves in the dark and in light conditions,  www.nature.com/scientificreports/ the fourth day of culture corresponded to the end of the exponential growth phase (Fig. 1). Thus, all further experiments were conducted with 4-day old mycelium.
Effect of photoactivated TPPS on hyphae morphology. As the mycelium growth was affected by photoactivated TPPS, it was decided to carefully look at the hyphae structure, using ESEM. As expected, photoactivated TPPS induced important phenotypic changes of the hyphae, compared with the control that showed very regular hyphae with a well-organized structure (Fig. 2a). In presence of 1.5 µM TPPS, hyphae were notably less organized, exhibited irregular shapes and produced some spores (Fig. 2a). Moreover, the TPPS-treated hyphae presented a reduced width (1.36 µm) compared with the control (4.7 µm) (data not shown).  www.nature.com/scientificreports/

TPPS localization inside mycelial cells.
According to the phenotypical effects observed in TPPS-treated mycelium under light, it was of interest to localize TPPS inside the cells. TPPS was found inside several cells of 4-day-old mycelium ( Fig. 2b). This intracellular localization could explain the very strong inhibitory effect of photoactivated TPPS on mycelial growth. As TPPS was located inside cells, this showed that it could cross the cell wall and accumulate in the cell cytoplasm.
Biochemical activities of TPPS-treated B. cinerea mycelium. To gain insight into the effect of photoactivated TPPS on the mycelium, biochemical assays linked to oxidative stress and cell metabolism activity, were conducted. As expected, increases in both H 2 O 2 and MDA contents were observed in treated mycelium compared to the control, demonstrating at the molecular level that the fungus was stressed (Fig. 3). Metabolic activity, linked to mitochondrial respiration, was monitored with the MTT assay. Indeed, under light, formazan production decreased in the treated hyphae compared with the control, thus explaining the fungal growth inhibition (Fig. 3). While MDA indicated an increase in lipid peroxidation in the presence of photoactivated TPPS, the thiol content provided additional evidence that the fungus was not dead but only deeply stressed (Fig. 3).

TPPS and grapevine plantlets grown in vitro.
Internodal explants from each variety: Merlot, Sauvignon and Chardonnay, were placed on 12.5 µM of TPPS for two months in growth chamber as described in the material and methods section. Each culture was examined during a period of 3-months and the explant subculture was performed from 2-month-old plantlets. There was no phenotypical difference between the control and the treated plantlets after a 1-month culture (Fig. 4). To confirm the absence of the phenotypical effect of photoactivated TPPS, we measured the thiol content in aerial and root organs of treated and control plantlets from the three varieties (Table 1). Roots from the three varieties, that were in contact with photoactivated TPPS, showed a significant increase in the total thiol content compared to the control (Table 1). At the aerial level, while no difference in thiol content was observed in Sauvignon and Merlot, there was a significant difference in the total thiol content measured in Chardonnay (Table 1). This could suggest that the Chardonnay variety is more sensitive to photoactivated TPPS than the two other backgrounds although no visible outcome could be seen at the phenotypic level (Fig. 4).

TPPS effect on B. cinerea infected leaves.
The final interest of this work was to put together grapevine, B. cinerea and TPPS with the expectation to kill the pathogen without disturbing plants. As a preliminary assay, before the development of a complete plant pathosystem, our antifungal photodynamic treatment was tested on detached grapevine leaves from two-month-old plantlets. Experiments were conducted as described in the material and methods section, in the growth chamber. The infection was monitored for 72 h. Before 48 h, no change was observed in leaves from the three clones (data not shown). After 48 h, the untreated B. cinerea started to invade the leaf surface of the three varieties. The growth of the mycelium, pre-treated with 12.5 µM of TPPS for 8 h, was not completely inhibited (Fig. 5). However, after a 50 µM TPPS pre-treatment, the growth of mycelium was totally inhibited on the leaf surface suggesting that the strategy described in this study worked efficiently. ESEM analysis was conducted on infected leaves. In uninfected leaves, the structure was well-defined and organized (Fig. 6). However, in the infected leaves when B. cinerea was not pre-treated with TPPS, the fungus completely invaded the leaves, making the structure unrecognizable (Fig. 6). By contrast, when B. cinerea was pre-treated with 50 µM of TPPS, ESEM pictures showed a leaf structure almost identical to that of the control www.nature.com/scientificreports/ confirming the macroscopic phenotype ( Fig. 6). It was noticed for the Sauvignon variety, that spores were present in the pre-treated B. cinerea suggesting that Sauvignon was more susceptible than the other two varieties. For further investigation, biochemical assays were performed. H 2 O 2 quantification, linked to oxidative stress, was conducted for the three leaf conditions and the three varieties after infection or not with TPPS. After a 72 h treatment, a basal content of H 2 O 2 was detected in the three leaf varieties, very similar for Chardonnay and Sauvignon leaves and slightly lower for Merlot (Table 2). When the leaves were infected by a 4-day-old B. cinerea mycelium, the H 2 O 2 content significantly increased for the three varieties especially for Sauvignon (more than 3.5-fold higher than the control leaf) explaining its strong susceptibility to B. cinerea (Table 2). An increase in H 2 O 2 production in all leaf types in contact with B. cinerea was expected. Concerning the leaves infected by the fungus pre-treated with 50 µM TPPS for 8 h, no significant difference was observed between Chardonnay and Merlot leaves compared to leaves infected by the fungus. No signs of infection were observed after 72 h culture (Fig. 5, Table 2). B. cinerea, pre-treated with TPPS, was no longer able to induce a high production of H 2 O 2 in Sauvignon leaves ( Table 2). The H 2 O 2 content nearly reached the basal level detected in the control Sauvignon leaves ( Table 2). As a response to this H 2 O 2 production detected in the leaves induced by B. cinerea, and to gain insight into the leaf 's proper response, we measured the total thiol content in leaves that were subjected to the three different treatments. For Chardonnay and Merlot detached leaves, no significant difference was observed between the control, infected leaves and infected leaves pre-treated with TPPS (Table 2). A significant increase in thiol content was only observed for Sauvignon leaves between the control and the infected leaves. This result suggested that Sauvignon leaves were able to fight against the fungus infection with a thiol induced response (Table 2). Nevertheless, we also confirm that our pretreatment with TPPS inhibited B. cinerea growth on leaves for each variety. These results are promising for the development of APDT treatments in agriculture (Figs. 5, 6).

Discussion
Botrytis cinerea is a very serious problem in a large variety of plants. It is a necrophytic fungus that induces ROS production that contributes to plant cell destruction during its infection 45 . This fungus is able to infect leaves, stems, flowers and fruit, causing severe damages and commercial losses in agriculture. In vineyards, the fungus  Table 1. Total thiol content of 2-month old grapevine plantlets. Roots and aerial organs from each variety: Sauvignon, Chardonnay and Merlot, were separately frozen and analyzed. TPPS was tested at 12.5 µM. FW fresh weight. Results are the mean of three independent experiments ± sd. Statistical significance is determined by a one way ANOVA test (*P < 0.05, **P < 0.01). Statistical analysis was always performed against the control. When no indication, not significant.

Roots
Aerial organs www.nature.com/scientificreports/ induces several deleterious effects on both quality and quantity of vine production. Despite all the damage it can cause, under specific weather conditions, its growth on grapes induces noble rot that gives rise to sweet wine. However, the fight against this pathogen remains a daily struggle especially for fruit production in summer or fall. For more than 50 years, the use of specific fungicides has largely been envisaged and in that time, B. cinerea found coping strategies. As a consequence, fungicide treatments gradually became inefficient, even the famous CuSO 4 solution also known as 'Bordeaux mixture' lost its effectiveness against fungal pathogens 46,47 . Moreover, the copper divalent ion, also toxic for plants, contributes to soil pollution 48,49 .  Results are the mean of three independent experiments. Statistical significance is determined by a one way ANOVA test (*P < 0.05, **P < 0.01; ***P < 0.001). Statistical analysis was performed for each assay and was represented as follows: C = *** in IL column meant C compared to IL with a P < 0.001. No indication means not significant. www.nature.com/scientificreports/ Therefore, new strategies against fungi are urgently needed. Wang et al. 50 used naturally occurring eugenol (EC50 value of 235 µM for B. cinerea). This was found to mainly affect fungal mycelium growth rather than the germination of spores as shown in previous reports on fungicides, such as carbendazim and N-phenyl carbamates 51,52 .

Control (C) Infected leaves (IL) Pretreated B. cinerea (PBc) Control (C) Infected leaves (IL) Pretreated B. cinerea (PBc)
In a similar way, Fleurat-Lessard et al. 53 discovered a strategy whereby they investigated the potential of FeSO 4 and found that the sulfate anion determines the inhibition of mycelium growth in pathogenic fungi that is similar to B. cinerea at high concentrations (range of 0.5-20 mM). They also tested different iron salts and found that those with bromide, chloride and sulfate anions showed the best antifungal activity. In particular, the addition of an ammonium counterion to the sulfate moiety contributed to the inhibition of mycelium growth in the pathogenic fungus Eutypa lata, an ascomycete like B. cinerea.
Therefore, taking these results from previous works into account 27,28,30,53 , we decided to test TPPS, a molecule that presents four external sulfonate groups linked by a tetrapyrrole ring. Indeed, TPPS with an ammonium counterion could be an excellent antifungal candidate. Moreover, it was demonstrated that TPPS remains negatively charged in a large array of chemical environments, even under acidic pH and does not aggregate in media, allowing it to diffuse through cell walls and membranes more easily 29,30 . Therefore, this PS was thought to be an excellent candidate for APDT.
Under white light, TPPS at a very low concentration (MFC = 1.5 µM) induced a severe inhibition of B. cinerea mycelium growth which led to death. We did not succeed to rescue the mycelium after this treatment. To our knowledge, there is little to no information available on the fungistatic or fungicidal effect of light-activated photosensitizers on B. cinerea 54 . The first step was to investigate whether the anionic porphyrin was able to induce any changes in the mycelium structure. In the previous study, it was shown that the structure of the B. cinerea hyphae changed after treatment with antibiotics, eugenol, FeSO 4 and tea tree oil 50,53,55,56 . Moreover, it has been shown that stressed mycelium often produces spores and could show altered cell elongation 57 . According to our data, TPPS also induced a structural change on the fungus. More specifically, the structure of the PS-treated fungus was thinner than the structure of the control. The treated mycelium produced spores implying that the fungus was under stress. Therefore, it was of interest to localize TPPS inside the cells. In tobacco plant cells, TPPS was also tested at 3.5 µM and it was the most effective porphyrin PS to induce cell death under a short light period of 5 h and it was proven to mainly localize in the cell wall 30,58 . Thus, a similar localization of TPPS was expected in the fungal cells. The multi-layer fungal cell wall is enriched in neutral sugars and proteins and poor in chitin and uronic acids, suggesting a neutral global charge of the cell wall 59 . Nevertheless, the B. cinerea cell wall composition and it's global charge remain controversial. It was hypothesized that this neutrality could allow TPPS to cross the fungal cell wall which is completely different from the tobacco cell wall 30,59,60 . Further analyses, such as the H 2 O 2 content, MDA, MTT, total thiols and microscopy analysis, confirmed fungus stress to the point of dying. Our findings proved that TPPS, due to its characteristics, could be a valid alternative to classic fungicides 29 .
From our previous studies, 50 µM TPPS was our reference for plantlets grown in vitro from seeds: Arabidopsis and tomato 27,28 . In this new study, we changed our plant model: grapevines that are obtained from clones, to establish a pathosystem with B. cinerea. Comparing with the other plant species, the grapevine varieties did not grow optimally at root apparatus level in the presence of 50 µM TPPS. Thus, the concentration was reduced to 12.5 µM TPPS for all the chosen varieties: Chardonnay, Merlot and Sauvignon. The choice of these three varieties was due to their different susceptibility to B. cinerea 42,43 . Furthermore, 12.5 µM TPPS did not induce any phenotypical nor biochemical modification of the three grapevine plantlets. In addition, this TPPS concentration was approximately ten times higher than the minimal concentration inhibiting the mycelium growth (1.5 µM); thus, the strategy presented in this article could work against pathogens without altering plant growth and development.
The final aim was to demonstrate TPPS efficiency against B. cinerea in 2-month-old infected grapevine leaves to validate the hypothesis of mimicking a pathosystem. The in vitro cultures and artificial system demonstrated the potential of our strategy. ESEM images from infected leaves, of the three grapevine varieties, after 72 h treatment confirmed that these leaves infected with B. cinerea pre-treated with 50 µM TPPS were similar to that of the control. For the Sauvignon variety, the presence of spores on infected leaves confirmed that it is more sensitive to the fungus infection than the other two varieties 42,43 . This is also confirmed by the measure of the total thiol content, considered as a primary and strong defense to infection 61 .
In conclusion, TPPS was able to kill the pathogen B. cinerea without harming the grapevine leaves in vitro. Moreover, this molecule does not produce any biochemical nor phenotypical changes on the plantlets grown in vitro. These preliminary experiments carried out are indeed promising and, in the future, experiments could be done in a greenhouse and in fields to determine the real potential and efficacy of TPPS against plant pathogens. The results and findings presented herein are also very encouraging because the photodynamic treatment has been developed using a low concentration of PS. Therefore, we show that APDT can be used for the struggle against phytopathogens in the agronomic practices as the PS is effective against plant pathogens and exhibits non-toxic side effects toward plants.

Material and methods
Photosensitizer. 5 www.nature.com/scientificreports/ plug (0.6 cm diameter) of 2-week old B. cinerea mycelium was placed in the middle of plates containing PDA medium supplemented with or without TPPS in the concentration range 0.5-3.5 µM. TPPS was added to the PDA medium just after autoclaving (120 °C, 20 min). Plates were incubated at 22 °C, either in the dark or subjected to a photoperiod of 16 h (Osram large spectrum white lamp: photon flux density of 120 μmol m −2 s −1 ). Daily measurements of the diameter of the fungus were performed and reported to draw the growth curves.
Grapevine clone culture. Chardonnay (clone 7535) was provided by Pr. Clément (Université de Reims, Champagne-Ardenne, France). Sauvignon (clone 379) and Merlot (clone 373) were provided by the Institut Français de la Vigne et du Vin (Bordeaux, France). Intermodal explants of grapevine were dissected and placed in glass tubes or jars containing half Chée and Pool medium, and 2% (w/v) sucrose-solidified medium (pH 5.9) for 2 months. TPPS was added to the medium after autoclaving. Chée and Pool medium was purchased from Duchefa Biochemistry (Haarlem, Holland). The cultures were then exposed to white light (Osram large spectrum white lamp: photon flux density of 120 μmol m −2 s −1 ) for 16 h and the temperature was maintained at 24 °C.
Infection of young leaves with B. cinerea. Two-month-old grapevine leaves and four-day-old mycelium plug (0.6 cm diameter) were used to perform the experiment. Mycelium discs were firstly incubated in 12 Total thiol assay. Approximately 100 mg of fine powder has been obtained from the samples (fungus, plantlets or in vitro detached leaves). After grinding in liquid nitrogen, 1 mL of 0.2 N HCl was added to the powder. A centrifugation at 13,000×g for 20 min was performed. Afterward, 500 μL of the supernatant was neutralized with 400 μL NaOH (0.2 M) and 50 μL NaH 2 PO 4 (0.2 M). 700 μL of 0.12 M NaH 2 PO 4 , 6 mM EDTA and 0.1 mL of 6 mM dithiobis-2-nitro-benzoic acid (DTNB) was added to 200 μL of extract. A standard calibration curve was prepared by replacing the extract with 0, 5, 10, 25 and 50 μg/mL glutathione solutions (total volume 1 mL). Absorbance at 412 nm was read 5 min after the addition of glutathione or extract. Confocal microscopy analysis. Mycelium was cultivated for 4 days on media containing 3.5 μM TPPS in the dark. Data acquisition with a LSM510META Zeiss confocal microscope (Carl Zeiss France, Marly-le-Roi, France) was performed under the spectral acquisition mode for TPPS localization inside the mycelium (excitation at 405 nm, emission detected at 640 nm) and under the channel mode for examination of the mycelium.

Scientific Reports
| (2020) 10:17438 | https://doi.org/10.1038/s41598-020-74427-9 www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.