1 Introduction

Phytopathogens cause worldwide crop yield losses of up to 40%, thereby leading to severe economic losses for the agricultural industry. These shortfalls lead to waste of valuable resources, such as water, energy, fertilizers, and labor. In addition, the fall in crop yields is anticipated to cause increasing food prices, posing a threat to adequate access to nutritionally beneficial food and endangering worldwide food supply [1, 2]. Global warming will further endanger crop yields, especially in the Northern Hemisphere, due to shifts in pathogen distributions and microbial species compositions. Moreover, changes in physiologic, biochemical, and evolutionary processes of the host and pathogen interaction will be affected by climate change, leading to the rise of new pathogens [3, 4].

Phytopathogenic fungi are a serious hazard in agriculture. Botrytis cinerea, a necrotrophic plant pathogen, especially concerns farmers as it causes substantial damage in horticultural crops. Therefore, it gained attention as a critical target for the development of novel management strategies in agricultural research [5]. Following Magnaporthe oryzae, B. cinerea was ranked the second most important fungal plant pathogen in 2012 [6]. And still, a decade later, B. cinerea is an issue of concern, due to its many different routes of attack and multi-layered and multifunctional infection pathways [7]. As a generalist pathogen it is capable of causing gray-mold disease in over 200 different crop species, making it a universal challenge to most sectors of agricultural industry [8]. Pathogenicity begins with its ability to persist as mycelia, conidia or sclerotia in crop remnants. Upon entering the fresh host plants during early development stages, it remains quiescent until more favourable growth conditions appear. Consequently, damages caused by B. cinerea are frequently detected post-harvest [9]. Since the gray mold disease affects such a wide range of crop plants and pathogenicity occurs in all parts of the plant, the costs of damage control are difficult to assess. However, chemical control measures against Botrytis species are estimated to cost up to 40 € per hectare, accounting for 10% of the global fungicide market [6].

Chemical control agents in form of synthetic fungicides are the main measure for management of B. cinerea [10]. Yet, the overuse of synthetic fungicides in agriculture since the 1950s lead to an increasing number of resistant strains also within the Botrytis genus. Managing resistance in B. cinerea is particularly challenging because, in addition to single target site resistance, certain strains have developed multi-drug resistance (MDR) [11]. Multi-drug resistant B. cinerea strains exhibit enhanced efflux activity of fungicides and overexpression of membrane transporter genes. The incidence of MDR strains is rapidly increasing in French and German vineyards. In contrast, the development of fungicides is lagging behind due to common obstacles, such as off-target toxicity or food safety issues, emphasizing the need for alternative management strategies against resistant B. cinerea strains [12]. Alternating several fungicides, varying doses, and using drug mixtures can decelerate, but not prevent the development of resistances [13]. Particularly for organic agriculture, the number of approved fungicide compounds is limited. Consequently, copper-based formulations were introduced against Botrytis infections [14]. Yet, the extensive use of copper may lead to accumulation in soils, compromising the groundwater, soil microbiomes and plants [15]. Moreover, besides residual pesticides, copper accumulates in the fruit and hence in the processed product, as recently demonstrated for European organic wines [16].

As an alternative method to combat fungal phytopathogens in agriculture, Photodynamic Inactivation (PDI) was recently introduced [17]. PDI utilizes photosensitizers (PS). These photoactive molecules generate reactive oxygen species upon illumination which are then capable of oxidizing various cellular components, such as cell wall and DNA, leading to death of pathogenic cells. Through PDI a wide spectrum of pathogens can be targeted, without any known development of resistance [18]. Sodium–magnesium–chlorophyllin (Chl) is a natural, water-soluble PS, approved as a food additive in the European Union (E140), thereby proving its safety in humans. Chl was already proven to show a phototoxic effect against fungal and bacterial phytopathogens [17, 19]. Recently, Chl has achieved an antibacterial effect against a Streptomycin resistant strain of the Gram-negative bacteria Erwinia amylovora [20].

The aim of this study is to investigate Chl for its potential as a PS in PDI of fungicide-susceptible versus -resistant strains of B. cinerea. The formulations SUN-D-05, -06 and -07, containing the naturally derived Chl in different concentrations, are used in addition with the chelator EDTA disodium salt dihydrate (NA2EDTA) and a surfactant (Alkylsulfosuccinate). As we successfully demonstrated the effectiveness of SUN-D-06 in vitro, this formulation was further examined within a greenhouse trial using the pepper ‘California Wonder’.

2 Experimental

2.1 Fungal culture

Botrytis cinerea strains for the in vitro study were purchased from the DSMZ (Braunschweig, Germany [21]) or were kindly provided and resistance typed by Anne-Sophie Walker (Palaiseau, France) (Table 1). Cultivation and PDI were preformed according to the protocol of Hamminger et al. [17]. The fungal strains were grown on malt extract peptone (MEP) agar plates (30 g L−1 malt extract (Carl Roth GmbH & Co. KG), 3 g L−1 peptone (Peptone ex casein, Roth) and 15 g L−1 Agar–Agar (Kobe I, Roth)) at 26 °C. The imides resistant strain Bc-80 was cultivated on MEP agar plates supplemented with 5 mg L−1 cycloheximide (Sigma-Aldrich Chemie GmbH), while the benzimidazole-resistant strain (Bc-842) and the MDR strains (Bc-VA101 and Bc-VA664) were cultured on MEP agar supplemented with 1 mg L−1 carbendazim (Sigma-Aldrich Chemie GmbH). The fungicides carbendazim and cycloheximide were diluted in 2.5 mL sterile ddH2O and added after autoclaving at 75 °C. For growing mycelial spheres, fungal material was transferred from agar plates into 20 mL liquid MEP medium at 26 °C and 200 rpm in a shaking incubator (MaxQ 4000, Thermo Scientific) for at least 48 h until mycelial spheres of 2–3 mm formed.

Table 1 PDI-treated Botrytis cinerea strains including isolate name, known resistances, date and origin of collection
Full size table

2.2 In vitro Photodynamic Inactivation

PS formulations were received as powders from Suncor AgroScience (Mississauga, ON, Canada). Aqueous solutions of the dry material were prepared (for concentrations refer Table 2) and stored at − 20 °C until use. The absorption signal of the PS formulation with medium Chl concentration (SUN-D-06) was measured at 0.23% concentration using an Infinite M200 microplate reader (Tecan Trading AG). The mycelial spheres were transferred into 24-well plates (one sphere per well) using a pipette. To each well 500 µL of treatment were added. Samples were incubated for 100 min in the dark at room temperature under constant agitation. Negative controls were kept in Dulbecco’s Phosphate Buffered Saline (DPBS, Sigma-Aldrich Chemie GmbH) and positive controls in hydrogen peroxide (H2O2, Roth). After incubation, each sphere was washed once in 500 µL DPBS. Samples were illuminated in fresh 500 µL DPBS. The illumination was performed using an LED array (22 mW cm−2, 106 J cm−2, diode type L-7113UVC, dominant wavelength 395 nm, Kingbright Electronic Europe GmbH). Dark controls of each treatment were kept in the dark for the same time period. Subsequently, each sphere was transferred to 6 cm MEP agar plates (without fungicide) and incubated at 26 °C for 7 days. The photokilling effect was evaluated by counting the samples for which no growth was observed. The percentage of dead samples was calculated as follows:

Table 2 Formulation composition for in vitro Photodynamic Inactivation of B. cinerea spheres including concentrations of active photosensitizer compound sodium–magnesium–chlorophyllin (Na–Mg–Chl), concentrations of the chelator EDTA disodium salt dihydrate (Na2EDTA), surfactant Alkylsulfosuccinate and dilution percentages
Full size table
$$\text{Percentage of dead samples} \; \left[\%\right]= \frac{\text{treated samples without growth}}{\text{amount treated samples}}\times 100$$

2.3 Greenhouse trial

A greenhouse trial was conducted on pepper plants (Capsicum annum L., California Wonder, Houwelings Nursery Ltd.) in a research greenhouse located at Frost Environmental, Abbotsford, BC, Canada. The greenhouse contained 24 LED lights (each 0.9 m long, 120 W per light (~ 10,800 lm per light, Monios-L)) on a 12 h timer as well as supplemental heat. Temperature and humidity logs are given in the Supplementary Materials. The 7-week-old plants at the 6–8 leaf stage were transplanted into coco fiber bags (Richgrow™, Terralink) in the greenhouse. Each plot of a one m2 area contained two bags with a total of four plants laid out in a randomized complete block (RCB) design with four replicates. Each plant was fertigated with a single emitter (0.5 gph low flow rate, Netafim Emitter). Foliar sprays of PS were prepared in a solution volume of 1000 L ha−1, accounting for 100 mL per plot. Controls were sprayed with water. Treatments are summarized in Table 3, water control was used as negative control, while RHAPSODY ASO (Plant Products) was chosen as a positive control. Besides SUN-D-06 in different rates also a combination of SUN-D-06 with BRIJ L4 was tested. This ethylene oxide-based surfactant is a spray additive which enables the foliar uptake of compounds into leaves and is already used in emulsifier systems in agriculture [23]. Foliar applications were made with a CO2 backpack sprayer at 276 kPa, equipped with a single nozzle (Tip 8002VS, TeeJet Technologies). Minor foaming was observed when mixing the PS products, approximately 0.5 cm per 100 mL. Foaming dissipated within a minute of mixing and posed no problem to the application.

Table 3 Treatments, product rates and solution volumes for the bell pepper greenhouse trial
Full size table

Treatment was applied weekly for a total of 4 weeks (Fig. 1). Forty-eight h after the first treatment, wetted leaves were inoculated under cool, cloudy conditions in the late evening. Sporulating diseased leaves from extra plants were brushed against the trial plants to dislodge the conidia from the diseased plants onto the healthy plants. Conidia from the diseased plants were confirmed under the microscope as B. cinerea. A high level of natural inoculum was present in the greenhouse as well.

Fig. 1
figure 1

Experimental schedule of the greenhouse trial. Four applications of treatments were done in 7-day intervals. The inoculation of Botrytis cinerea was done 48 h after the first treatment application

Full size image

Gray-mold symptoms were assessed weekly, prior to each application and at 7 days after the last application for a total of 5 assessments. Foliar symptoms, such as leaf lesions, wilting, stem lesions, and leaf drop, were counted and recorded. The number of lesions were recorded at each date as well as the number of leaves affected and total leaves lost. Visual assessments of diseased leaf area (leaf area covered by B. cinerea lesions) and plant vigor were also recorded at each assessment. The percentage of leaf area diseased was rated weekly on an Horsfall-Barratt (HB) scale. The HB ratings were transformed to percentages following the standard grade formula by Redman, King and Brown [24] (Table 4).

Table 4 Horsfall–Barratt (HB) scale for the determination of diseased leaf area including transformed HB scale
Full size table

Vigor was rated on a 1–10 scale where 10 = healthy, green and vigorous and 1 = dead. The area under the disease progress curve (AUDPC) [25] was calculated at the end of the trial using the percent leaf area data. Due to the high level of disease pressure in combination with cool winter conditions, the plants did not mature to fruiting stage, so final yield was assessed on the plant’s marketability and quality. Each plant was determined if it was marketable or non-marketable, one week after the final treatment, based on the plant vigor, number of Botrytis spots and percent leaf area covered by the spots. Plants with vigor ratings of 7 or higher and with a percentage of leaf area covered of 1% or less were considered marketable.

The mean disease severity for all evaluated plant characteristics was calculated using following equation:

$$\text{Mean disease severity} \; \left[\%\right]=\frac{\text{mean value }\left({\text{treatment}}\right)}{\text{mean value }\left(\text{water control}\right)}\times 100$$

3 Results

3.1 In vitro Photodynamic Inactivation

The absorption spectrum of the PS formulation SUN-D-06 and the structure of the photoactive molecule Chl are shown in Fig. 2. The percentage of dead samples was measured after a 7-day incubation period by evaluating if growth of mycelial spheres occurred on agar plates. Sample images of mycelial patches on agar plates are provided in Fig. 3A. In all experiments, untreated controls (co-/-) of all strains showed clear growth, resulting in a complete coverage of the dishes with mycelia after 7 days. Illumination with 395 nm LED light without PS had no effect on all samples (light only control).

Fig. 2
figure 2

Absorption spectrum of the photosensitizer formulation SUN-D-06 at a 0.23% dilution and the chemical structure of the photoactive molecule sodium–magnesium chlorophyllin

Full size image
Fig. 3
figure 3

Results of in vitro tests for five tested B. cinerea strains after SUN-D-05, -06 and -07 PDI treatment. A Sample photographs of agar plates of the wild-type B. cinerea 7 days post PDI treatment using the PS SUN-D-06. Mean percentages of dead samples [%]: B wild-type strain (Bc-WT). C Strain Bc-842, carbendazim-resistant (D) a cycloheximide resistant strain Bc-80 and two multidrug resistant strains BC-VA101 (E) and Bc-VA664 (F). PDI treatment was done using an LED array (395 nm, 22 mW cm−2, 106 J cm−2). Co-/-: double negative control, DPBS, no illumination; light only: DPBS with illumination; 3% H2O2 only: positive control, no illumination; 3% H2O2: positive control with illumination. The numbers above bars represent the number of tested fungal spheres per treatment. Summarized numerical values can be found in the Supplementary Materials

Full size image

A complete eradication (100%, no growth detected on agar plates) was given for the wild-type strain (Bc-WT, Fig. 3B) after PDI with a radiant exposure of 106 J cm−2 and the PS formulations SUN-D-06 and SUN-D-07. Even though SUN-D-05 contains the highest concentration of the photoactive compound Chl, it eradicated only 55.5% of samples. The positive control 3% hydrogen peroxide induced killing of all mycelial spheres. As for dark toxicity of the PS formulations, incubation of Bc-WT spheres with SUN-D-06 without illumination resulted in growth inhibition of 22.2% of spheres. SUN-D-07 was less toxic and SUN-D-05 did not show dark toxicity at all.

As for the benzimidazole resistant strain Bc-842, treatment with all three tested-PS formulations resulted in complete photokilling (100%, Fig. 3C). Dark toxicity of SUN-D-06 without illumination was 33.3%, and again, SUN-D-05 was not dark-toxic. SUN-D-07 showed less toxicity than SUN-D-06. The positive control (H2O2, illuminated and in the dark) achieved a somewhat lower toxicity than PDI (87.5%).

Likewise, phototreatment of the imides resistant strain Bc-80 was successful with all tested-PS formulations (100% killing of mycelial spheres, Fig. 3D). Equally, the positive control H2O2 eradicated all tested mycelial spheres. The highest dark toxicity was detected for SUN-D-07 (37.5% killing of spheres). SUN-D-05 (14.3%) and SUN-D-06 (25.0%) showed lower toxicity in the dark.

PDI with all tested PS formulations effectively photokilled the MDR strain Bc-VA101 (100% killing of spheres, Fig. 3E), as did the positive control H2O2. The dark toxicity of the tested formulations did not exceed the values achieved for strain Bc-80 (< 25%).

In like manner, all samples of the second MDR strain Bc-VA664 were entirely eradicated by SUN-D-05, SUN-D-06 and SUN-D-07 (Fig. 3F). Likewise, the positive control H2O2 showed an antifungal effect (100% killing). Toxicity of SUN-D-05 alone without illumination was 33.3%, not exceeded by the other formulations (14.3% for SUN-D-06 and SUN-D-07).

3.2 Greenhouse trial

The number of Botrytis leaf spots (Fig. 4A) after 3 weeks showed significant differences in Tukey’s HSD at P = 0.05 compared to the water control. All tested products and the standard RHAPSODY ASO were significantly different from the water-treated control. However, SUN-D-06 at the 2× rate, and the 1× rate in combination with 0.05% BRIJ L4, performed the best with zero spots (100% reduction from the water control). At the final assessment, 7 days after the 4th application, all test products were significantly different from the water-treated control, but not significantly different from each other or the standard RHAPSODY ASO. Plants treated with RHAPSODY ASO had the lowest average number of spots (86.2% percent of inhibition).

Fig. 4
figure 4

A Mean disease severity percentage of Botrytis leaf spots per plant in respect to the water control. B Mean disease severity percentage of number of diseased leaves in respect to the water control. C Mean disease severity percentage of diseased leaf area in respect to the water control. Each characteristic was evaluated in 7-day intervals before the next treatment application. All numerical values in tables can be found in the Supplementary Materials

Full size image

At each assessment, the number of affected leaves and total leaves was counted per plant to calculate the percentage of diseased leaves (Fig. 4B). Significant differences in Tukey’s HSD at P = 0.05 were first observed in the third week. All test products and the standard RHAPSODY ASO were significantly different from the water control. However, the 2× rate of SUN-D-06 and SUN-D-06 in combination with 0.05% BRIJ L4 performed the best with 100% of inhibition compared to the water-only treated control. After the final assessment, all test products were significantly different from the water control but not significantly different from each other or the standard product RHAPSODY ASO. Plants treated with RHAPSODY ASO had the lowest average percentage of leaves affected (5%), with SUN-D-06 in combination with BRIJ L4 the next best-performing product at an average of 9.5% leaves affected, lower than all the other test products.

Significant differences in diseased leaf area were first observed in week 4 (Tukey’s HSD, P = 0.05) (Fig. 4C). The 1× rate of SUN-D-06 in combination with BRIJ L4 and the 2× rate of SUN-D-06 were the highest performing products with an average 1.3–1.4% leaf area diseased (87.4–88.3% less than the water control), both treatments were significantly different from the water-treated control, but not significantly different from the other test products or the standard RHAPSODY ASO. For the final assessment, all test products and the standard RHAPSODY ASO were significantly different from the water control in Tukey’s HSD at P = 0.05, but not significantly different from each other. The standard RHAPSODY ASO performed the best with an average percent of inhibition of 90.1%, while SUN-D-06 in combination with BRIJ L4 was the next best treatment with an average 88.7% percent of inhibition compared to the water control.

SUN-D-06 in combination with BRIJ L4 performed the best for AUDPC with the lowest AUDPC of 37.6 but it was not statistically different from the other product treatments or the standard RHAPSODY ASO (Table 5). All product treatments including the standard RHAPSODY ASO were significantly different from the water control in Tukey’s HSD at P = 0.05.

Table 5 Integrated disease progress curve (AUDPC), visual rating of plant vigor and the mean inhibition percentage of dropped leaves compared to the water control
Full size table

Each plant was rated for vigor on a 1–10 scale where 10 = healthy, vigorous, and green in color and 1 = dead (Table 5). Significant differences from the water control in Tukey’s HSD were first observed after the first treatment. Plants treated with SUN-D-06 in combination with BRIJ L4 had the highest average plant vigor rating of 8.7, a 10.1% increase in vigor from the control. After the final assessment, SUN-D-06 in combination with BRIJ L4 was the highest performing treatment with an average vigor rating of 7.9. SUN-D-06 in combination with BRIJ L4 was significantly different from the water control but not significantly different from the standard RHAPSODY ASO or the 2× rate of SUN-D-06.

Leaf drop is a common symptom of Botrytis gray mold on greenhouse pepper. Each week the number of leaves per plant dropped was counted (Table 5) Significant differences were first observed in Tukey’s HSD at P = 0.05 after the final assessment. At least one leaf had dropped in all treatments however, the 1× rate of SUN-D-06 had the highest average leaf drop per plant (0.4), 50.0% less than the water control but not significantly different from the control or the other test products.

The percentage of marketable plants was evaluated after the final assessment (Fig. 5). SUN-D-06 in combination with BRIJ L4 and the standard RHAPSODY performed the best for total marketable yield (100%) with the highest percentages of marketable plants but were not significantly different from the 1× or 2× rate of SUN-D-06.

Fig. 5
figure 5

Total marketable plants. Numbers were evaluated after the final assessment. Plants with vigor ratings of 7 or higher and with a percentage of leaf area covered of 1% or less were considered marketable. Mean and standard deviation of four plants per plot, 4 replicates per treatment, RCB design. Letters indicate Tukey’s HSD at P = 0.05 results, bars labeled with same letters are not significantly different. All numerical values in tables can be found in the Supplementary Materials

Full size image

4 Discussion

The need for environmentally friendly Botrytis management agents is expected to rise over the next few decades [26]. Classic treatments using regular fungicides bear the risk of novel resistant strains and as the world population grows, crop protection must be prioritized to ensure food safety globally [27]. We successfully demonstrated that PDI using the natural photosensitizer Na-Mg-Chlorophyllin (Chl, food additive E140), combined with a surfactant and Na2EDTA, represents a promising approach to combat B. cinerea (Fig. 3). This is especially meaningful, as Chl-based PDI proved highly effective also against strains with challenging fungicide resistances. When illuminated, photoactive substances will undergo a change in their energy state, resulting in the production of toxic species of either oxygen radicals (type-I photochemical reaction) or singlet oxygen molecules (type II photochemical reaction). Chlorin-based molecules, such as Chl, favour to produce singlet oxygen molecules rather than radicals, therefore are considered type II PS [28,29,30]. In vitro experiments could show a 100% photokilling effect against all tested strains, independent on resistance to conventional treatment. Our observations indicate that SUN-D-06 and -07 are the most promising formulations with little to no dark toxicity. The SUN-D-05 formulation could achieve a killing effect of 100% for the resistant strains, however only a 55.5% killing rate for the wild-type. As working hypothesis, we suggest that due to the darker color of the solution resulting from the higher Chl concentration in SUN-D-05, a filter effect is likely in the treated mycelial spheres. The stained outside hinders the light penetration into in the inner sphere, most likely leading to a potential decreased photokilling effect. This indicates that Chl-based PDI, preferably in low concentrations, provides one universal management strategy regardless of the resistance mechanisms of B. cinerea.

Furthermore, under high disease pressure in a research greenhouse in British Columbia, Canada, SUN-D-06 in combination with BRIJ L4 was the best-performing treatment against B. cinerea. Reduction of Botrytis lesions in number and in diameter in planta was best achieved in our study by addition of BRIJ L4. BRIJ L4 is an ethylene oxide-based surfactant enabling the foliar uptake of compounds into leaves and is presently used in agricultural settings [23]. Compared to the water-treated control, plants treated with SUN-D-06 in combination with 0.05% BRIJ L4 had 75.9% fewer Botrytis lesions per plant, 73.8% fewer leaves affected, and 88.7% lower leaf area diseased (Fig. 4), an AUDPC that was 80.7% less, and the highest plant vigor (Table 5) and percentage of marketable plants at the end of the trial (Fig. 5).

The need for a cell permeabilizing agents in antifungal PDI-based approaches results from fungal cell-wall peculiarities [31]. The negatively charged Chl alone cannot permit the wall and therefore cannot introduce reactive oxygen species to the pathogen [17]. Other studies overcome this problem by using positively charged synthetic PS. For example, the cationic PS Tetra-4-sulfonatophenyl porphyrin tetra-ammonium, in concentrations as low as 1.5 µM inhibited the growth of B. cinerea cultures completely [32]. However also another naturally occurring photoactive compound, curcumin, can kill B. cinerea effectively (4 log10 steps of B. cinerea spores and reduced cell viability). Nevertheless, the highly hydrophobic molecule must laboriously be dissolved in ethanol [33]. In our study, Na2EDTA was used to achieve cell-wall penetration. The effect is supported by Hamminger et al., who achieved an almost complete Botrytis eradication (91.7%) using 100 µM Chl in combination with 5 mM Na2EDTA and the same light conditions as in this study, indicating that Chl is a suitable antifungal agent, if used in combination with the cell permeabilizing agent. Moreover, the same study revealed a 94.1% photokilling effect against Alternaria solani, the inducing pathogen of early blight, under the same experimental conditions. Furthermore, it could be shown that Na2EDTA does not influence Chl absorption, nor its ability to generate reactive oxygen species. Hence, the addition of cell permeabilizing agents is not altering the photoactive effects of Chl, but rather enhance the toxicity against microbes [17].

The antimicrobial effect of Chl for agricultural applications has been shown against other pathogens already, however mainly against bacteria. Chl at a concentration of 100 µM showed a 7 log10 step reduction against a streptomycin resistant strain of Erwinia amylovora [20]. The application of Chl-based PDI for human-relevant pathogens mainly focuses on food disinfections, as Chl is approved as a food additive in the European Union (E140). PDI using Chl was introduced as promising method to ensure food safety on sprouts and cherry tomatoes [34, 35]. Using 100 µM Chl, a 4 log10 step reduction of Listeria innocua could be achieved in mung bean seeds, whereby no alteration of the further germination of the seed was reported [34]. The Gram-positive food pathogens Bacillus cereus and Listeria monocytogenes were reduced by 1.5 log10 steps in 150 µM Chl on tomato fruits with no harm to the tomato plants [35]. The copper salt of Na-chlorophyllin was found as an effective antimicrobial on food package surfaces against Listeria monocytogenes and Bacillus cereus [36, 37]. Additionally, Chl has an antibacterial effect against Staphylococcus aureus on lettuce [38]. Chl concentrations of as little as 15 µM were also already tested against Gram-negative food pathogens, in order to establish edible coatings for strawberry fruits. While Chl alone was not able to eradicate Salmonella enterica, a chlorophyllin-chitosan complex could reduce the bacteria by 7 log10 steps [39]. Recently, the application range of Chl used in antibacterial materials was further expanded to hospital settings. Fabrics based on Chl achieved an 99.998% and 99.994% inactivation against clinically relevant and antibiotic resistant strains of Enterococcus faecium and Staphylococcus aureus, respectively [40].

To our understanding, until now little is known to the effect of PDI against resistant phytopathogenic fungi. To date, one single other study has introduced PDI against a fungicide-resistant Penicillium digitatum strain using a water-soluble vitamin K3 analog as a PS illuminated by natural sunlight [41]. The results presented in this study and available results from previous studies conclude that PDI using Chl-based PS is a working tool against phytopathogenic B. cinerea. The in vitro experiments of this study could show that SUN-D-06 and SUN-D-07, PS formulations based on the naturally derived Chl can achieve a 100% photokilling effect in several B. cinerea strains, independent on their resistance to conventional fungicides. A total eradication was achieved with PS formulations containing Chl at a concentration as little as 470 µM.

Moreover, because of the limited number of in planta studies of PDI applications, knowledge about the effect in actual agricultural use is limited. So far, studies implicate that Chl-based PS do not harm plant growth or integrity and do not alter the quality of edible plants, e.g., tomatoes [17, 34, 35, 42]. The data we present here show a good compatibility of the used PS and plant growth. After four applications of the PS formulations on greenhouse pepper plants through spraying, the marketability of the plants was enhanced compared to the control and was equally high as for the competitor product RHAPSODY ASO. Due to the growth conditions in the greenhouse and the reduced temperature of the area of experiment, the growth of fruits could not be assessed and might be a question for further research. RHAPSODY ASO uses the strain QST 713 of Bacillus subtilis as an active ingredient and was shown to be active against B. cinerea [43]. However, pesticides based on replacement microbes possess several downsides. Proper storage environments and culture conditions appropriate for the used active microorganism must be kept in mind. Also, the production of microorganism-based products serves several challenges for example the missing standardization [27]. Likewise, the production of ready-to-use PS for fungal control comes with challenges [44]. Photostability of the product is welcome, however a degradation of the product through light into products that plants can easily metabolize is adding to the eco-friendliness of the product. Chl degrades into natural products, which do not tend to be accumulated in the soil [45]. This study, including in vitro and in planta results give a promising outlook to future field applications of Ch-based PDI. The spectrum of SUN-D-06, given in Fig. 2, indicates by the absorption peaks in the visible light range that an excitation via sunlight is a possible option for a cost-effective use of the tested-PS formulations. Larger field trials on different model plants, such as grapes, and against different phytopathogens are however needed for a better insight into Chl as a PS in organic agriculture.

To conclude, PDI using a natural, water-soluble Chl-based PS provides a reliable antifungal effect in vitro against B. cinerea irrespective of resistance to conventional antifungals. Chl-based PS are effective for photoinactivation of fungal spheres. In addition, a good control of gray mold of greenhouse peppers was granted using the same PS, comparable to the standard RHAPSODY ASO, with no visible crop injury. The efficacy of the product was enhanced by the addition of the surfactant BRIJ L4 at 0.05%. Our findings strengthen the idea of using tailored formulations of Chl as photofungicides, allowing for economic and eco-friendly disease control.