A Coating Based on Bioactive Compounds from Streptomyces spp. and Chitosan Oligomers to Control Botrytis cinerea Preserves the Quality and Improves the Shelf Life of Table Grapes

Botrytis cinerea is the most harmful postharvest disease of table grapes. Among the strategies that can be envisaged for its control, the use of coatings based on natural products is particularly promising. The study presented herein focuses on the assessment of the antagonistic capacity of two Streptomyces species and their culture filtrates against B. cinerea. Firstly, the secondary metabolites were characterized by gas chromatography–mass spectrometry, with N1-(4-hydroxybutyl)-N3-methylguanidine acetate and 2R,3S-9-[1,3,4-trihydroxy-2-butoxymethyl]guanine acetate as the main compounds produced by S. lavendofoliae DSM 40217; and cyclo(leucyloprolyl) and cyclo(phenylalanylprolyl) as the most abundant chemical species for S. rochei DSM 41729. Subsequently, the capacity of S. lavendofoliae DSM 40217 and S. rochei DSM 41729 to inhibit the growth of the pathogen was tested in dual culture plate assays, finding 85–90% inhibition. In agar dilution tests, their culture filtrates resulted in effective concentration values (EC90) in the 246–3013 μg·mL−1 range. Upon the formation of conjugate complexes with chitosan oligomers (COS) to improve solubility and bioavailability, a synergistic behavior was observed, resulting in lower EC90 values, ranging from 201 to 953 μg·mL−1. Ex situ tests carried out on ‘Timpson’ and ‘Red Globe’ table grapes using the conjugate complexes as coatings were found to maintain the turgor of the grapes and delay the appearance of the pathogen by 10−15 days at concentrations in the 750−1000 µg·mL−1 range. Hence, the conjugate complexes of COS and the selected Streptomyces spp. culture filtrates may be put forward as promising protection treatments for the sustainable control of gray mold.


Introduction
Table grapes are considered a product with functional characteristics since they offer a wealth of health benefits due to their high nutrient and antioxidant contents [1], which are essential for a healthy human diet [2]. Table grapes are one of the most consumed fruits globally and are cultivated in about 90 countries [3]. Like other fresh fruits, grapes are very delicate and the losses at harvest and during distribution are very high [4], which makes them a highly perishable food with relatively short postharvest storage lives [2].
Postharvest diseases of table grapes in favorable conditions-and particularly in developing countries-can cause losses in total production of up to 55% [5]. This situation is aggravated by the fact that only 5% of all agricultural research funding is allocated to postharvest challenges, with the majority of investments going toward expanding

In vitro Growth Inhibition Tests
The results of the mycelial growth inhibition tests are summarized in Figure 3. When tested alone, a higher efficacy of COS than of both secondary metabolites was observed: full inhibition was reached at 1500 µg·mL −1 for the former, whereas secondary metabolites reduced mycelial growth but did not reach complete inhibition at the highest assayed concentration. Upon conjugation of the secondary metabolites with COS, an enhancement in terms of efficacy was found in both cases: the conjugated COS-S. lavendofoliae DSM 40217 metabolites treatment reached full inhibition at 1000 µg·mL −1 and the COS-S. rochei DSM 41729 metabolites treatment did so at 750 µg·mL −1 ( Figure S2).  Figure S1). Such results are in agreement with those previously reported by Ghanem et al. [24] for S. rochei and two other Streptomyces spp.

In vitro Growth Inhibition Tests
The results of the mycelial growth inhibition tests are summarized in Figure 3. When tested alone, a higher efficacy of COS than of both secondary metabolites was observed: full inhibition was reached at 1500 μg·mL −1 for the former, whereas secondary metabolites reduced mycelial growth but did not reach complete inhibition at the highest assayed concentration. Upon conjugation of the secondary metabolites with COS, an enhancement in terms of efficacy was found in both cases: the conjugated COS-S. lavendofoliae DSM 40217 metabolites treatment reached full inhibition at 1000 μg·mL −1 and the COS-S. rochei DSM 41729 metabolites treatment did so at 750 μg·mL −1 ( Figure S2). Upon comparison of the effective concentrations (Table 1), differences in the efficacy of the treatments could be observed more clearly. The highest efficacy (i.e., the lowest EC50 and EC90 values) corresponded to the COS-S. rochei DSM 41729 metabolites conjugate complex, followed by the COS-S. lavendofoliae DSM 40217 metabolites conjugate complex, COS, S. rochei DSM 41729 metabolites, and S. lavendofoliae DSM 40217 metabolites. Regarding the synergy factors (Table 2), a synergistic behavior was found between COS and the metabolites (SF values ≥ 1). The highest synergy factor was obtained for the EC90 of COS-S. lavendofoliae DSM 40217 (SF = 2.33).  Upon comparison of the effective concentrations (Table 1), differences in the efficacy of the treatments could be observed more clearly. The highest efficacy (i.e., the lowest EC 50 and EC 90 values) corresponded to the COS-S. rochei DSM 41729 metabolites conjugate complex, followed by the COS-S. lavendofoliae DSM 40217 metabolites conjugate complex, COS, S. rochei DSM 41729 metabolites, and S. lavendofoliae DSM 40217 metabolites. Regarding the synergy factors (Table 2), a synergistic behavior was found between COS and the metabolites (SF values ≥ 1). The highest synergy factor was obtained for the EC 90 of COS-S. lavendofoliae DSM 40217 (SF = 2.33).

Ex Situ Growth Inhibition Tests
As shown in Figure 4, the two treatments were very effective, achieving a high inhibition of the fungus until the end of the trial. Disease incidences on both cultivars of table grapes were calculated on days 1, 5, 10, and 15 (Table S3). In the negative controls, the fungus did not proliferate (discarding contamination), while in the clusters that had been artificially inoculated with the fungus, but which were not treated, B. cinerea managed to invade 100% of the cv. 'Timpson' berries and ca. 80% of the cv. 'Red Globe' berries after 15 days. Regarding the cv. 'Timpson' clusters treated with COS-S. lavendofoliae DSM 40217 conjugate complex, no colonization by the pathogen was observed after 15 days, while in the cv. 'Red Globe' grapes some colonization was observed after 10 days, reaching an incidence of ca. 20% after 15 days. On the other hand, in the tests carried out with COS-S. rochei DSM 41729 conjugate complex coating, 100% effectiveness against B. cinerea after 15 days was observed for both cultivars (and at a lower concentration).

Ex Situ Growth Inhibition Tests
As shown in Figure 4, the two treatments were very effective, achieving a high inhibition of the fungus until the end of the trial. Disease incidences on both cultivars of table grapes were calculated on days 1, 5, 10, and 15 (Table S3). In the negative controls, the fungus did not proliferate (discarding contamination), while in the clusters that had been artificially inoculated with the fungus, but which were not treated, B. cinerea managed to invade 100% of the cv. 'Timpson' berries and ca. 80% of the cv. 'Red Globe' berries after 15 days. Regarding the cv. 'Timpson' clusters treated with COS-S. lavendofoliae DSM 40217 conjugate complex, no colonization by the pathogen was observed after 15 days, while in the cv. 'Red Globe' grapes some colonization was observed after 10 days, reaching an incidence of ca. 20% after 15 days. On the other hand, in the tests carried out with COS-S. rochei DSM 41729 conjugate complex coating, 100% effectiveness against B. cinerea after 15 days was observed for both cultivars (and at a lower concentration).  It is worth noting that in both cases, the treatments resulted in better preservation of the firmness and turgor of the berries. Regarding the berries' color, it was not affected in the case of the cv. 'Red Globe' clusters, but the COS-S. rochei DSM 41729 conjugate complex coating conferred a slightly yellowish hue to the cv. 'Timpson' berries after day 5 (an effect that was not observed in the clusters treated with COS-S. lavendofoliae DSM 40217 conjugate complex). This should be taken into consideration, as it may affect consumers' preferences.
Deoxyspergualin (also present in S. rochei DSM 41729) has been reported to promote resistance to Candida albicans (apart from being one of the best immunosuppressive, radiationprotective, antineoplastic, and hypoglycemic agents) [30]. Regarding the secondary metabolites identified in S. rochei DSM 41729, cyclo(leucyloprolyl) and cyclo(phenylalanyl-prolyl) regulate the growth of some fungi [31] and the latter one has been reported to exhibit activity against C. albicans. Oxazinoquinoxalines have also been reported to exhibit antibacterial activity [32].

Efficacy of the Treatments
In relation to the efficacy of the treatments, the review paper by De Simone et al. [10] provides a qualitative and quantitative comparison of different biological compounds-including chitosan-for the control B. cinerea in table grapes. However, specific inhibition rates and the associated concentrations/effective concentrations of Streptomyces spp.-based treatments were not covered. A survey of such values against the pathogen under study is summarized in Table 3 for comparison purposes.
Taking into consideration that both the B. cinerea isolates and the Streptomyces species/strains differ from one article to another, a word of caution seems necessary regarding the comparisons presented below.
Concerning the antibiosis assay, a comparison of the inhibition percentages obtained herein (80% for S. lavendofoliae DSM 40217 and 85% for S. rochei DSM 41729) with the other Streptomyces spp. in the table suggests that only Streptomyces sp. A3265, S. globisporus JK−1, and Streptomyces sp. S97 would have a higher antifungal potential against B. cinerea in dual plate assays.
Regarding the minimum inhibitory concentration when metabolites were tested, the EC 90 values of 1723 and 3013 µg·mL −1 obtained for S. rochei DSM 41729 and S. lavendofoliae DSM 40217, respectively, indicate that total inhibition of pathogen growth would be achieved at a dose approximately ten times lower than that reported for S. globisporus JK−1 (30,000 µg·mL −1 ), one of the species that outperformed the species tested herein in the dual plate assays. Therefore, the active compounds present in the secondary metabolites of S. rochei DSM 41729 and S. lavendofoliae DSM 40217 may have more promising antifungal activity. In relation to chitosan, it has been widely studied for the postharvest treatment of grapes. In some previous studies, chitosan failed to fully inhibit the radial growth of the fungus, reducing the mycelium growth to 3.84 cm, as compared to 8.4 cm for the control [40]. In the work by Ramos-Bell et al. [41], in which a concentration of 15,000 µg·mL −1 was tested (approximately ten times higher than the MIC obtained in our trials, 1500 µg·mL −1 ), a 93.38% inhibition was found. Muñoz et al. [42] obtained an EC 50 of 1770 µg·mL −1 , a concentration approximately equal to the EC 90 of the assay. These differences may be attributed to the use of chitosan oligomers instead of medium or high molecular-weight chitosan, given that the former ones have a higher antifungal activity [43].
With the ex situ experiment, it has been shown that cv. 'Red Globe' is more resistant to B. cinerea than other grape varieties, such as 'Italia' or 'Victoria' [9]. These results would be in agreement with our findings, provided that cv. 'Timpson' was more sensitive to B. cinerea than cv. 'Red Globe'. In terms of the appearance of the treated grapes vs. untreated control grapes, in the case of cv. 'Red Globe', the obtained results are similar to those reported by Xu et al. [40]. Although there was no fungal development in the grapes treated with the coatings, on day 10 the firmness and turgor began to decline. Given that the berries of this cultivar have thinner skin than those of other cultivars, it could be a possible explanation for the proliferation of the fungus [44]; microcracks make the skin more permeable, which speeds up transpiration and water absorption (depending on humidity and internal pressure in the berry), resulting in increased fruit deterioration and a decrease in fruit firmness [45].

Mechanism of Action
In relation to the antifungal activity of COS, it is well established in the literature and may be ascribed to several mechanisms of action on the fungal cell wall [46]. The presence of chitosan polysaccharides confers a positive molecular charge, which, upon interaction with the negatively charged fungal cell membrane (lipopolysaccharides, phospholipids, or lipoproteins) [47], leads to changes in the cell membrane, such as increased cell permeability [48]. In turn, this results in changes in its functionality, in addition to the loss of intracellular components such as proteins and Na + and Ka + ions, which affects the osmotic pressure and finally cause cell death [49]. It has also been hypothesized that COS can significantly alter cellular chitin levels and thereby disturb the equilibrium between degradation and synthesis, resulting in a weakening of the cell wall [50]. In addition, excessive ROS attacks intracellular biomolecules such as lipids, proteins, and DNA, causing irreversible oxidative damage and thereby triggering apoptosis and necrosis. Furthermore, COS can infiltrate into the nucleus of fungus and disrupt the synthesis of DNA, as well as RNA [51]. As for its effect on grapes, chitosan can significantly increase the enzymatic activities of grapefruit, including superoxide dismutase, peroxidase, catalase, and ascorbate peroxidase, reducing the damage by active oxygen [43]. Regarding its role in the conjugate complexes, the observed synergism may result from an improved additive fungicidal activity per se or by simultaneous action on various fungal metabolic sites [52], but it may also be due to the fact that chitosan oligomers can enhance the solubility and bioavailability of the bioactive compounds present in the secondary metabolite filtrate [53,54].
Regarding the mode of action of S. rochei DSM 41729 and S. lavendofoliae DSM 40217, volatile organic compounds (VOCs) produced by Streptomyces spp. have been previously shown to control postharvest diseases. For instance, the volatiles emitted from S. setonii WY228 effectively controlled the postharvest pathogen of sweet potato Ceratocystis fimbriata Ellis et Halsted, 1890 [55]. Volatiles mainly downregulated the ribosomal synthesis genes and activated the proteasome system of the fungus, while the genes for spore development, cell membrane synthesis, mitochondrial function, and hydrolase and toxin synthesis were also downregulated, suggesting that VOCs have a variety of mechanisms of action on fungal pathogens [56].
Concerning other grapevine diseases, secondary metabolites from S. rochei DSM 41729 and S. lavendofoliae DSM 40217 have also been demonstrated to have antifungal activity against downy mildew [57]. Such activity would arise from the presence of bioactive substances with antifungal capacity, as discussed in the previous section. However, further research would be needed to gain insight into the mechanisms of action of each of them and into the differences in terms of effectiveness depending on whether they act together or separately.

Bacterial and Fungal Isolates, Reagents, and Table Grapes
The two bacteria of the genus Streptomyces, viz. Streptomyces rochei (DSM 41729) and

Preparation of Secondary Metabolites of Streptomyces spp. and Preparation of B. cinerea Conidial Suspension
Lyophilized S. lavendofoliae (DSM 40217) and S. rochei (DSM 41729) were inoculated in TSB at 28 • C for 24 h and were seeded on SCA medium plates at 28 • C for 10 days. The plates were stored at 4 • C. Secondary metabolites were obtained following the method described by Sadigh-Eteghad et al. [58]. After fermentation, each final solution of the two cultures was treated with 50 mL of phosphate buffer and then sonicated for 5 min (pH 6.4). After centrifuging the filtrates, the bioactive compound-containing supernatant was extracted with 100 mL of ethyl acetate, concentrated under pressure, and then freeze-dried. In agreement with Pazhanimurugan et al. [59], the culture filtrates had a concentration of approx. 2000 µg·mL −1 (1958 µg·mL −1 for S. lavendofoliae DSM 40217 secondary metabolites and 1877 µg·mL −1 for S. rochei DSM 41729 secondary metabolites). For in vitro and ex situ antifungal activity tests, the freeze-dried samples were dissolved in Milli-Q ® water.
For the preparation of B. cinerea conidial suspension, lyophilized B. cinerea (CECT 20973) was inoculated in PSB at 26 • C for 24 h and was seeded on PDA medium plates at 26 • C for 7 days. Subsequently, 5 plugs (10 mm in diameter) were inoculated in 150 mL of PDB. Seven-day-old PDB cultures (150 mL cultures kept at 26 • C, 135 rpm in an orbital stirrer incubator) were harvested for B. cinerea conidia. The suspension was filtered twice to eliminate hyphal fragments before measuring the spore concentration using a hemocytometer (Weber Scientific International Ltd., Teddington, Middlesex, UK). It was then amended with 0.2% Tween 20 ® and adjusted to a final concentration of 1 × 10 6 spores (conidia)·mL −1 .

Synthesis of Chitosan Oligomers and COS-Secondary Metabolites Conjugate Complexes
The procedure reported by Santos-Moriano et al. [60] with some modifications was used to obtain chitosan oligomers. 20 g of chitosan was stirred with 20 g of citric acid in T1 L of Milli-Q ® water at 50 • C until a homogeneous solution was obtained. Then, the enzyme Neutrase ® 0.8 L (1.67 g·L −1 ) was added to degrade the polymer chains. Subsequently, the mixture was subjected to sonication in 3 min cycles with 1 min stops in between. Once the process was completed, oligomers of molecular weight < 2000 Da were obtained in a solution with a pH of 4-6. The molar mass of the COS samples was determined by measuring the viscosity, in agreement with Yang et al. [61] in a solvent of 0.20 mol·L −1 NaCl + 0.1 mol·L −1 CH 3 COOH at 25 • C using an Ubbelohde capillary viscometer. Molar mass was determined using the Mark-Houwink equation [η] = 1.81 × 10 −3 M 0.93 [62].
Conjugate complexes of secondary metabolites of S. rochei DSM 41729 or S. lavendofoliae DSM 40217 and COS were obtained according to the procedure previously described in [63], by mixing in a 1:1 ratio (w/w: 2000 µg·mL −1 /2000 µg·mL −1 ), followed by sonication. The resulting stock solution (at a concentration of 2000 µg·mL −1 ) was diluted in Milli-Q ® water to obtain the different concentrations assayed in the mycelial growth inhibition tests.

In Vitro Tests of Mycelial Growth Inhibition
For the antagonistic study of B. cinerea versus both Streptomyces spp., the methodology of Ghanem et al. [24] was used, with slight modifications. PDA was used as the culture medium, in which 4 drops of 30 µL from TSB tubes previously seeded at 26 • C for 18 h were added to reach a concentration of 1 × 10 8 CFU·mL −1 . The plates were then incubated at 26 • C for 3 days. Then, a 5 mm plug of B. cinerea was added in the center while another plug was inoculated on a control plate. Both were incubated for 7 days at 26 • C. The experiment was performed in triplicate with three replicates. The growth inhibitory effects on the pathogenic fungi were represented as a percentage of pathogen growth inhibition (PGI) and calculated according to Zambrano et al. [64], using the following equation: PGI = (R 1 − R 2 )/R 1 × 100, where R 1 = growth of pathogenic fungus in the control plate and R 2 = growth of pathogenic fungus interacting with the antagonist.
For the mycelial growth inhibition assays with the five treatments (viz. COS, S. lavendofoliae DSM 40217 secondary metabolites, S. rochei DSM 41729 secondary metabolites, COS-S. lavendofoliae DSM 40217 secondary metabolites, and COS-S. rochei DSM 41729 secondary metabolites), the methodology reported by Sánchez-Hernández et al. [65] was chosen. PDA plates with 10 concentrations of each treatment (ranging from 62.5 to 1500 µg·mL −1 ) were inoculated with 5 mm diameter plugs, and incubated at 26 • C for 7 days, together with the control plate, containing only culture medium. Growth inhibition was calculated according to the formula: ((d c − d t )/d c ) × 100, where d c is the average colony diameter in the control and d t is the average diameter of the treated colony. EC 50 and EC 90 (50% and 90% effective concentrations, respectively) were estimated using PROBIT analysis in IBM SPSS Statistics v.25 software (IBM; Armonk, NY, USA). The level of interaction, i.e., the synergy factor (SF), was estimated according to Wadley's method [66].

Ex Situ Tests of Mycelial Growth Inhibition
The effectiveness of the different conjugate complex-based coatings on gray mold incidence was also assayed ex situ on cv. 'Red Globe' and 'Timpson' table grapes. The MIC values obtained in the in vitro mycelial growth inhibition assays were used, viz. 750 and 1000 µg·mL −1 for COS-S. rochei DSM 41729 secondary metabolites and COS-S. lavendofoliae DSM 40217 secondary metabolites, respectively.
Two cultivars of organic table grapes were tested: green seedless cv. 'Timpson' grapes, which are more sensitive to B. cinerea, and cv. 'Red Globe' red grapes with seeds, which are more tolerant to B. cinerea. Once harvested, the grapes were sent from Murcia to the University of Valladolid facilities by an express courier service at 4 • C. The trial began 24 h after the grapes had been harvested.
Forty bunches of uniform size, color, and shape were selected to obtain homogeneous lots, all of which were checked to ensure that they were free of lesions. The trials were carried out in triplicate (120 bunches). At the start of the trials, grapes were superficially disinfected with a 3% NaOCl solution for 2 min, washed three times with sterile distilled water, and dried in a laminar flow hood using sterile absorbent paper. In this state, the negative control grapes (i.e., no fungus and no treatment) were reserved. Subsequently, the grapes were immersed in the respective treatments (viz. COS-S. lavendofoliae DSM 40217 and COS-S. rochei DSM 41729 secondary metabolites) for 5 min. Once dry, 2 mm cuts were made with a scalpel on the grapes, which were then inoculated with 10 µL of a conidia suspension (10 6 conidia mL −1 ) [67]. Some untreated grapes were reserved to make cuts and add the conidia, using them as positive controls (with fungus and without treatment). The grapes were then stored in a chamber at 22 • C with a relative humidity of 90%, i.e., under conditions suitable for the development of B. cinerea. The trial lasted 15 days, the time it took for the fungus to completely colonize the grapes of the positive control. During this time, the quality of the grapes was studied in terms of morphological, color, and textural changes, together with the time of appearance and colonization of the fungus.

Gas Chromatography-Mass Spectrometry Analysis of Streptomyces spp. Secondary Metabolites
Part of the culture filtrates was reserved to identify the active compounds produced by S. rochei DSM 41729 and S. lavendofoliae DSM 40217. For this purpose, the culture filtrates were centrifuged and the supernatant, which contained the bioactive compounds, was extracted with 100 mL of ethyl acetate, concentrated under pressure, and freeze-dried. Then, 25 mg of the obtained freeze-dried aqueous extract was dissolved in 5 mL of HPLCgrade methanol to obtain a 5 mg·mL −1 solution, which was filtered and used for GC-MS analysis. The GC-MS analysis was carried out at the Research Support Services (STI) at Universidad de Alicante (Alicante, Spain), using a gas chromatograph model 7890A coupled to a quadrupole mass spectrometer model 5975C (both from Agilent Technologies, Santa Clara, CA, USA). Chromatographic conditions: 1 µL injection volume; 280 • C injector temperature, in splitless mode; initial oven temperature = 60 • C, 2 min, followed by a ramp of 10 • C·min −1 up to a final temperature of 300 • C, 15 min. The chromatographic column used for the separation of the compounds was an Agilent Technologies HP-5MS UI of 30 m in length, 0.250 mm diameter, and 0.25 µm film. Mass spectrometer conditions: temperature of the electron impact source of the mass spectrometer = 230 • C and of the quadrupole = 150 • C; ionization energy = 70 eV. The identification of components was based on a comparison of their mass spectra and retention time with those of the authentic compounds and by computer matching with the database of the National Institute of Standards and Technology (NIST11).

Statistical Analysis
Data were subjected to analysis of variance (ANOVA) in IBM SPSS Statistics v.25 software. Tukey's HSD test at 0.05 probability level (p < 0.05) was used for the post hoc comparison of means.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.