Activation of sweet pepper defense responses by novel and known biocontrol agents of the genus Bacillus against Botrytis cinerea and Verticillium dahliae

The Fresno-Benavente Pepper (F-BP) Protected Geographical Indication (PGI) is a horticultural crop characterized by its great agronomic, economic and cultural importance in the region of Castilla y León (Spain). Field production is threatened by verticillium wilt caused by Verticillium dahliae and postharvest losses due to grey mould caused by Botrytis cinerea. Eight Bacillus spp. strains endophytically isolated from F-BP roots were used in the study. By conducting an in vitro antagonism study, we found that all Bacillus strains were effective against B. cinerea and five of them showed high antagonism against V. dahliae, with B. siamensis and B. proteolyticus strains being the most effective against both pathogens. Eight Bacillus strains were used for an infection test in F-BP fruits and plants to test their activity against both fungal pathogens. We report that Bacillus thuringiensis, B. siamensis and B. pumilus (SCFC 1–2) could control B. cinerea in pepper fruits through direct action and local activation of plant defences. In the case of V. dahliae root infection, plant roots inoculated with B. siamensis and B. proteolyticus were able to significantly decrease the occurrence of disease through direct action and local activation of jasmonic acid as a defence response. Therefore, we propose that B. siamensis could be used to control B. cinerea and V. dahliae in F-BP fruits and plants, respectively, through direct antagonism as well as the induction of local plant defence responses.


Introduction
Sweet pepper (Capsicum annuum L.) is an annual herbaceous plant belonging to the Solanaceae family and originating from South America (Fratianni et al., 2020;Guevara et al., 2021). It is part of the diet especially in European countries (Fratianni et al., 2020). Sweet peppers represent an important nutritional resource of antioxidant compounds, such as carotenoids, ascorbic acid and polyphenols (Fratianni et al., 2020). The world production of this crop stands at 36 million tons per year, on 2 million hectares (FAO, 2020).
According to the European Union Council Regulation (EC) No 925/2012, the Fresno-Benavente Pepper (F-BP) ("Pimiento de Fresno-Benavente, in Spanish) is produced in three specific regions in Castilla y León (Spain), namely Benavente and Los Valles (Zamora province) and Fresno (León province), all of them watered by the Esla and Tera rivers. F-BP was granted a Protected Geographical Indication (PGI) in 2012 (EC 925/2012). The F-BP describes a type of pepper that has been specifically adapted to its production area, including 55 municipalities in the north of the province of Zamora, 33 municipalities in the southeast of the province of León, and a municipality in the province of Valladolid. It is a "cow's nose" or "bell pepper" type, large (10 cm minimum width), red in colour, rectangular in shape and with thick and juicy flesh (EC 925/2012).
Wilt caused by Verticillium dahliae and grey mould caused by Botrytis cinerea are the two primary diseases impacting production. V. dahliae is a soilborne pathogen, causing serious economic losses in more than 150 different crops (Tubeileh & Stephenson, 2020). It is the most important root pathogen in pepper cultivation, causing a vascular wilt that leads to plant death, including symptoms such as a delay in plant growth, chlorosis, discoloration of the vascular bundles, and a whitish colour can be seen in the roots with serious damage that causes the loss of the outermost layers (Tyvaert et al., 2019). B. cinerea is a plant pathogenic fungus that causes serious diseases in more than 1400 different plant species. In peppers, B. cinerea can attack the leaves and fruits in the field, but the real damage occurs at postharvest when fruits develop the characteristic grey mould that spreads rapidly and causes fruit rot (Mekawi et al., 2019;Poveda et al., 2020).
Plant endophytes are microorganisms that develop their entire life or part of it within plant tissues without causing visible damage or disease . With regard to bacteria, they can be found inside different organs and plant tissues, such as seeds, tubers, roots, stems, leaves or fruits; the roots being the organ in which the greatest number and diversity of microorganisms can be found. The most commonly isolated bacterial genera are Pseudomonas, Bacillus, Enterobacter and Agrobacterium (Eljounaidi et al., 2016). Endophytic bacteria can be of great importance in the agricultural system, promoting plant growth, favouring tolerance to abiotic stresses or reducing the incidence of biotic stresses. Regarding the effects against phytopathogens, endophytic bacteria reduce the damage caused to plants due to the release of chemical compounds, such as antibiotics, a mechanism known as antibiosis, through direct parasitism, competition for space and/or nutrients, or the induction of specific plant defensive responses (Eljounaidi et al., 2016;Poveda & González-Andrés, 2021).
In a previous work, different endophytic bacterial strains were isolated locally from F-BP plants (in 2010, and in the municipalities of: San Cristóbal de Entreviñas, Micereces de Tera and Fresno de la Vega) and selected for their outstanding biocontrol capacity against the root pathogenic oomycete of pepper, Phytophthora capsici. Out of the 122 isolated strains, two strains of the Bacillus pumilus species were described as the ones that most reduced the P. capsici infection in pepper (Barquero et al., 2016). The bacterial strains with the greatest biocontrol potential were those used in the present work.
The objective of this work was to identify effective endophytic and locally isolated bacterial biological control agents (BCAs) against V. dahliae and B. cinerea in F-BP and to assess potential modes of action. We utilized a local bacterial strain collection (described in the following section), isolated from F-BP plants grown in their geographical demarcation, because they are the bacteria that are fully adapted to the host plant and the environmental conditions of the crop. Strains were assessed for biocontrol activity through antagonistic in vitro assays, plant inoculations and infections in vivo and in planta, and molecular and biochemical analysis to assess potential activation of plant defences.

In vitro antagonism assay
The evaluation of the antagonistic activity of Bacillus species against B. cinerea and V. dahliae, was assessed by the dual culture technique in vitro, based on the methodology described by Slama et al. (2019), with some modifications. Mycelium disks (6 mm diameter) of each pathogenic fungus, obtained from the edge of the colonies grown on PDA medium, were placed in the centre of the Petri plate with TSA medium (pathogenic fungi grow similarly on TSA medium and PDA medium: Fig. S1). From an overnight pure bacterial culture on TSA agar, a loopful of cells was suspended in 0.9% NaCl solution, applying two 5 μl drops of 10 8 CFU mL −1 at 30 mm distance from the fungus inoculum. Controls were performed with and without the application of 5 μl of 0.9% NaCl solution, and no difference was found between them. The inhibition rate (IR) was calculated according to the following formula (Lyousfi et al., 2021): where A C means the colony area of the pathogenic fungus in the control treatment (TSA medium without bacterial treatments), and A T means the colony area of the pathogenic fungus in bacterial treatments. Fungal areas were measured using ImageJ photographic analysis software, at 3, 5 and 7 days for B. cinerea, and at 7, 10 and 14 days for V. dahliae. The experimental unit is represented by each of the Petri dishes. Each treatment was performed on eight plates (replicates per treatment), and the experiment was repeated twice.

In vivo biocontrol of B. cinerea in fruits assay
The in vivo study against the pathogen B. cinerea with the different Bacillus species was carried out on pepper fruits of the F-BP. For this, the methodology previously described by Shi and Sun (2017) for tomatoes was carried out, with some modifications. The fruits were superficially sterilized by washing with 70% ethanol and 2% sodium hypochlorite. A 3 × 3 mm crossshaped wound was made in the epidermis of the fruits where each of the bacteria was inoculated with 40 μl at a concentration of 10 8 CFU ml −1 . Two hours after bacterial inoculation, the wounds were inoculated with 15 μl of B. cinerea spores at a concentration of 10 5 spores ml −1 . Subsequently, the fruits were placed in plastic bags to maintain constant-high humidity and kept at 23°C and in the dark for 7 days. Different controls were used: peppers without B. cinerea inoculation and without bacterial treatment, peppers without B. cinerea inoculation and with bacterial treatment, and peppers with B. cinerea inoculation and without bacterial treatment.
After one week of fungal infection, measurements were taken of the areas of lesions produced by B. cinerea in peppers, tissue viability and reactive oxygen species (ROS) accumulation. In addition, infected tissues from the fruits were collected and immediately frozen in liquid nitrogen to analyse levels of pathogen colonization and expression of defence-related genes.
The experimental unit is represented by each fruit. Each treatment was performed on eight fruits (replicates per treatment), and the experiment was repeated three times. Treatments and B. cinerea-inoculations on each of the peppers were performed randomly.
In planta biocontrol of V. dahliae assay F-BP seeds were surface-sterilized by vigorous sequential shaking in 70% ethanol and 5% sodium hypochlorite solutions for 10 min each and then washed thoroughly four times in sterile, distilled water. Pepper seeds were individually transferred to 1 L pots, containing a sterilized mixture of field soil/vermiculite (3:1) that was previously prepared by autoclaving a local field soil twice (24 h apart). Seeded pots were maintained in a greenhouse.
Bacillus strains were applied to three-week-old pepper seedlings as described by Mena-Violante and Olalde-Portugal (2007), with some modifications. Pepper seedlings were treated by applying a 1 ml suspension (10 7 CFU ml −1 ) of each bacterium to individual roots. Inoculation with the pathogenic fungus V. dahliae was carried out as described by Mercado-Blanco et al. (2004), with some modifications. Two weeks after applying bacterial root treatments, the pepper plants were root inoculated with 3 ml of a V. dahliae spores suspension (10 7 spores ml −1 ).
Four weeks after inoculation with the fungus, the plants were collected and measurements were taken of the dry weight of the aerial part of the plant (after drying at 65°C for 48 h), disease severity, root vitality and ROS accumulation in root tissues. In addition, roots were collected and immediately frozen in liquid nitrogen to analyse levels of pathogen colonization and defence gene expression. Disease severity was analysed by a 0-4 rating scale according to the percentage of affected leaves and twigs (0 = no symptoms, 1 = 1-33%, 2 = 34-66%, 3 = 67-100%, and 4 = dead plant) (Mercado-Blanco et al., 2004).
The experimental unit is represented by the individual plants. Each treatment was performed on eight plants (replicates per treatment), and the experiment was repeated three times. Treatments and V. dahliaeinoculations on each of the plants were performed randomly. In addition, plants were randomly arranged in the greenhouse.

Vitality test and indirect quantification of ROS in tissues
To analyse and quantify the damage produced by both pathogens on F-BP fruits and roots, the vitality test and ROS quantification were assayed, as reported by Poveda (2021a).
The vitality test is based on the reduction of triphenyltetrazolium chloride (TTC) by living tissues to redcoloured insoluble triphenylformazan (TF), due to the activity of the mitochondrial respiratory chain. Thus, only living tissues should reduce TTC to TF. Pools were formed from eight fruits (B. cinerea-infected plants) or roots (V. dahliae-infected) per condition. From these pools, 100 mg was transferred to 1 mL of 1% TTC in triplicate and incubated for 48 h at 37°C. After incubation, 200 mg of Ballotini Glass Balls (0.15-0.25 mm diameter) was added to each sample in 1.5 mL Eppendorf tubes and shaken vigorously. After centrifuging the samples for 15 min at 10,000 rpm, the supernatant was removed, and 1 mL of isopropanol was added to each tube. The samples were again agitated and centrifuged in the same way, and the supernatant was used to quantify the absorbance at 620 nm; this acted as an indirect measure of the vitality of the F-BP tissues.
The indirect quantification of reactive oxygen species (ROS) in F-BP tissues was carried out by measuring electrolyte leakage, which is a measure of cellular oxidative damage related to the production of ROS, a sign of stress in plant tissues. This methodology of indirect analysis of cell damage caused by a plant pathogen has been used in many other works, with a direct relationship between the quantified damage and the indirect ROS data obtained (Poveda, 2020(Poveda, , 2021a(Poveda, , b, 2022a, demonstrating its success. From each pool formed from fruits (B. cinerea-infected plants) or roots (V. dahliae-infected), 1 cm 2 of fresh tissue was briefly mixed with water and floated on 5 mL of double-distilled water at room temperature for 6 h. The conductivity of the water was measured using a Crison™ Conductimeter GLP31 (Crison, Barcelona, Spain). This represented the electrolyte leakage from the tissues (Reading 1). Then, samples were boiled for 20 min at 90°C. After the liquid cooled down, the conductivity of the water was measured again. This represented the total concentration of ions present in the tissues (Reading 2). Electrolyte leakage, an indirect measurement of ROS, was represented as the percentage of total ions released [(Reading 1/Reading 2) × 100].

Tissue-colonization analysis
In order to determine the differences in F-BP tissue colonization by B. cinerea and V. dahliae, they were quantified by qPCR. For the analysis of B. cinerea-fruit and V. dahliae-root colonization, pools formed from eight fruits (B. cinerea-infected) or roots (V. dahliaeinfected) per treatment were used. All material was immediately frozen with liquid nitrogen and pulverized with a mortar.
Following the methodology described by Poveda (2022a), DNA was extracted using the cetyltrimethylammonium bromide (CTAB) extraction method. A mix was prepared in a 10-μL volume using 5 μL of Brilliant SYBR Green qPCR Master Mix (Roche, Penzberg, Germany), 10 ng of DNA, the forward and reverse primers at a final concentration of 100 nM, and nuclease-free PCR-grade water to adjust the final volume. The endogenous genes of pathogens and pepper used for the quantification are represented in Table 1. Amplifications were performed in an ABI PRISM 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA), programmed for 40 cycles under the following conditions: denaturation, 95°C for 15 s; annealing, 60°C for 1 min; extension, 72°C for 1 min. Each PCR was performed in triplicate by using the DNA extracted from three tissue pools of eight plants each -one for each treatment. Cycle threshold values served to calculate the amount of pathogen DNA using standard curves. The values of pathogen DNA were referred to the amount of pepper DNA in every corresponding sample.

Defence gene expression analysis
To analyse the defence responses following F-BP infection with B. cinerea and V. dahliae, an expression analysis of defence genes was performed by RT-qPCR. For gene expression studies, the pools formed from eight fruits (B. cinerea-infected) or roots (V. dahliaeinfected) per treatment were used for RNA extraction with TRIzol reagent (Invitrogen, Waltham, Massachusetts, USA). RNA was cleaned up with the Direct-zol RNA Miniprep Kit (Zymo Research, Irvine, CA, USA) including a DNase enzymatic treatment. cDNA was synthesized from RNA with Reverse Transcription System (Promega, Madison, WI, USA). Gene expression was analysed by RT-qPCR using an Agilent Mx3005P Real-Time PCR System (Agilent Technologies, Santa Clara, CA, USA) with Luna ® Universal qPCR Mastermix (New England Biolabs, Ipswich, MA, USA). All PCR reactions were performed in triplicate in a total volume of 10 μL for 40 cycles under the following conditions: denaturation at 95°C for 15 s; and annealing and extension at 60°C for 30 s; and melt curve. Threshold cycles (CT) were calculated using the pepper Actin gene as an endogenous control. The primers used are shown in Table 1, as follows: phenylalanine ammonia lyase 1 (PAL1), pathogenesis-related protein 1 (PR-1), lipoxygenase 1 (LOX1), and defensin 1 (DEF1) genes. Data are expressed using the 2 −ΔΔCT method (Livak & Schmittgen, 2001).

Statistical analysis
Statistical analysis of the data was carried out with Statistix 8.0 software. One-way ANOVA using Tukey's multiple range test at P < 0.05 was used for pairwise comparisons; the different letters indicate significant differences.
In vivo biocontrol of B. cinerea in fruits The treatment of pepper wounds with the Bacillus species caused a significant reduction in the lesions caused by B. cinerea (reductions of between 41% and 98%) (Fig. 3). The most efficient strain as BCA was B. safensis MTA 1-2, with a significantly smaller area of average lesions produced (reduction of 98%) by the pathogen than the following, B. thuringiensis FVA 2-3, B. pumilus SCFC 1-2 and B. siamensis SCFB 2-2 and SCFB 3-4 (Fig. 3b). B. thuringiensis (FVA 2-3). Data are the mean of two biological replicates for each condition with eight plates in each one. Oneway analysis of variance (ANOVA) was performed, followed by the Tukey's test. Different letters represent significant differences (P < 0.05) Six out of eight Bacillus strains significantly improved tissue viability based on relative TTC staining, compared to the non-treated control. From the effective strains, B. safensis MTA 1-2 produced the highest values of viability (0.078), significantly higher than B. siamensis SCFB 3-4 (0.039) and B. pumilus The columns represent the average of three biological replicates with eight fruits in each one. One-way analysis of variance (ANOVA) was performed, followed by a Tukey test. The different letters signify significant differences (P < 0.05) FVA1-5 (0.037); moreover B. siamensis SCFB 2-2 and SCFB 3-4, B. pumilus SCFC 1-2 and B. thuringiensis FVA 2-3 produced intermediate values (Fig. 4a). Concomitantly, only three strains reduced ROS levels compared with the non-treated control (4.23), namely B . s a f e n s i s M T A 1 -2 ( 1 . 2 2 ) f o l l o w e d b y B. thuringiensis FVA 2-3 (2.34) and B. pumilus SCFC 1-2 (2.81) (Fig. 4b).
The expression analysis of several genes involved in the defensive responses of the plant was carried out in the tissues of pepper infected with B. cinerea (Fig. 6). Strains B. safensis MTA 1-2 and B. thuringiensis FVA 2-3 significantly reduced the expression of PAL1 and B. megaterium; FVA 1-5 : B. pumilus; FVA 2-3 : B. thuringiensis. The columns represent the average of three biological replicates with eight fruits in each one. One-way analysis of variance (ANOVA) was performed, followed by a Tukey test. The different letters signify significant differences (P < 0.05) PR-1 (88% reduction) and significantly increased the expression of LOX1 and DEF1 (>400% increase). Similarly, inoculation with B. siamensis SCFB 2-2 and SCFB 3-4 and B. pumilus SCFC 1-2 caused a significant increase in the expression of LOX1 and DEF1, although the decrease in the expression of PAL1 and PR-1 was significantly less (except PAL1 expression for B. siamensis SCFB2-2 and PR-1 expression for B. siamensis SCFB3-4), compared to the non-treated control fruit. Fruit treated with B. proteolyticus SCFC 2- lipoxygenase 1 (LOX1) and defensin 1 (DEF1). The measurements were referenced to the value obtained for the plants without inoculation with Bacillus (2 -ΔΔCt = 1). The pepper Actin gene was used as an internal reference gene. The columns represent the average of three biological replicates with 8 fruits in each one. One-way analysis of variance (ANOVA) was performed, followed by a Tukey test. The different letters signify significant differences (P < 0.05) 2, B. megaterium MMM 3-6 and B. pumilus FVA 1-5 showed a significant increase in the expression of PAL1 a n d P R -1 ( e x c e p t e x p r e s s i o n o f P R -1 f o r B. proteolyticus SCFC2-2 and B. pumilus FVA1-5) and a significant decrease in LOX1 and DEF1 compared to the non-treated control (Fig. 6).
The quantification of the root colonization of pepper plants by the pathogen V. dahliae showed that roots FVA 2-3: B. thuringiensis. The columns represent the average of three biological replicates with eight plants in each one. One-way analysis of variance (ANOVA) was performed, followed by a Tukey test. The different letters signify significant differences (P < 0.05) treated with B. proteolyticus SCFC 2-2, B. siamensis SCFB 2-2 and SCFB 3-4, and B. megaterium MMM 3-6, significantly reduced the root colonization by the pathogen, compared to the non-treated control (reductions of between 69% and 83%) (Fig. 9).

Discussion
From the species included in our work, B. proteolyticus inhibited in vitro the growth of B. cinerea and FVA 2-3: B. thuringiensis. The columns represent the average of three biological replicates with eight plants in each one. One-way analysis of variance (ANOVA) was performed, followed by a Tukey test. The different letters signify significant differences (P < 0.05) Fig. 10 Quantitative reverse transcription polymerase chain reaction (RT-qPCR) analysis of the expression of some defence genes in F-BP plants infected with V. dahliae. Control: non-treated, inoculated control; SCFB 2-2: B. siamensis; SCFB 3-4: B. siamensis; SCFC 1-2: B. pumilus; SCFC 2-2: B. proteolyticus; MTA 1-2: B. safensis; MMM 3-6: B. megaterium; FVA 1-5 : B. pumilus; FVA 2-3 : B. thuringiensis. The genes analysed were phenylalanine ammonium lyase 1 (PAL1), pathogenesis-related protein (PR-1), lipoxygenase 1 (LOX1) and defensin 1 (DEF1). The measurements were referenced to the value obtained for the plants without inoculation with Bacillus (2 -ΔΔCt = 1). The pepper Actin gene was used as an internal reference gene. The columns represent the average of three biological replicates with eight plants in each one. One-way analysis of variance (ANOVA) was performed, followed by a Tukey test. The different letters signify significant differences (P < 0.05) V. dahliae, being the first description for both pathogens, although it has recently been described as an in vitro antagonist of pathogens Nigrospora sphaerica, Pestalotiopsis theae, Curvularia eragrostidis, Glomerella cingulata, Rhizoctonia solani and Fusarium oxysporum (Dutta & Thakur, 2021). On the other hand, B. siamensis is a bacterial species widely described as a BCA against pathogens, such as Alternaria alternata (Xie et al., 2021), Colletotrichum fructicola (Park et al., 2021), or fungi producing postharvest diseases, including B. cinerea (You et al., 2021). The two isolates from this study belonging to B. siamensis, inhibited the growth of B. cinerea. Moreover, they also inhibited the growth of the root pathogen V. dahliae, previously described (Gao et al., 2021). Similarly, B. safensis has been widely described as a potent BCA, including pathogens such as Aspergillus flavus (Einloft et al., 2021), A. alternata (Prakash & Arora, 2021), Magnaporthe oryzae (Rong et al., 2020), Plenodomus destruens (Mateus et al., 2021) and B. cinerea (Berrada et al., 2012;Hassan et al., 2021); the B. safensis strain from this study controlled the growth of B. cinerea in vitro as previously reported, but it also stood out as a BCA of V. dahliae, never reported before, to the best of our knowledge.
On the other hand, the potential of B. pumilus as a BCA has been described for R. solani and F. oxysporum (Agarwal et al., 2017). In our work, we have demonstrated in vitro effect against B. cinerea and V. dahliae. Interestingly, the inhibitory effect was strain-dependent in the case of B. cinerea, with B. siamensis SCFC1-2 being more effective than B. pumilus FVA 1-5, whilst the two strains were very effective against V. dahliae.
Finally, B. megaterium has previously been identified as an efficient BCA in vitro against A. flavus (Kong et al., 2010), Septoria tritici (Kildea et al., 2008) or Alternaria japonica (Vásconez et al., 2020) and B. cinerea, (Donmez et al., 2011). The B. megaterium strain from this work inhibited in vitro the growth of B. cinerea, but unexpectedly, it promoted V. dahliae growth, contrary to what other authors have observed for other strains (Abada et al., 2018). However, it has been previously described that B. megaterium promotes the growth of other fungi such as mycorrhizal ones (Marulanda et al., 2009). In this sense, Sarathambal et al. (2022) suggested that the reason for the positive interaction could be related to the bacteria zinc solubilizing capacity.
The results obtained in the in vivo infection of F-BP fruits with the pathogen B. cinerea after the different bacterial inoculations with Bacillus showed that the strains that best controlled the pathogen were B. safensis MTA 1-2, B. thuringiensis FVA 2-3, B. siamensis SCFB 3-4 and SCFC 1-2, and B. pumilus SCFC 1-2. Furthermore, it has been found that the bacterial strains B. proteolyticus SCFC 2-2, B megaterium MMM 3-6 and B. pumilus FVA 1-5 of do not induce the activation of defence genes in colonized pepper tissues.
B. safensis MTA 1-2 was able to reduce the disease caused by B. cinerea in pepper fruits. However, its in vitro antagonism was very low. Therefore, its ability as a BCA on fruits could only be a consequence of an activation of plant defence responses. The ability of B. safensis to induce plant defences has been determined in rice plants root inoculated with the bacteria and infected in the aerial part with the pathogen Bipolaris sorokiniana (Sarkar et al., 2018). In the specific case of B. cinerea, the present work could be the first description of B. safensis as an inducer of plant defences against the pathogen, although further tests are required to confirm this, e.g., by physically separating bacterial strain and pathogen.
Based on the results obtained in the study, B. thuringiensis FVA 2-3 would be able to act against B. cinerea in pepper fruits due to a direct antagonism and the activation of plant defensive responses. The ability of B. thuringiensis to activate plant defence responses has been previously described with tomato root treatments, against the fungus F. oxysporum f. sp. lycopersici (Akram et al., 2013) or against the bacteria Ralstonia solanacearum (Takahashi et al., 2014). Against B. cinerea, the ability of B. thuringiensis to activate plant defences and reduce the disease produced by the pathogen in tomato leaves has been described due to the induction of the expression of different defence genes (Yoshida et al., 2019).
Similarly, both strains of B. siamensis used would be capable of reducing the ability of B. cinerea to infect pepper fruit tissues through direct antagonism and activation of genes related to plant defence. In this sense, it has been previously described how the root inoculation of tobacco plants with B. siamensis significantly reduced the foliar lesions caused by the pathogenic fungus A. alternata, due to the activation of plant defence genes (Xie et al., 2021). In the specific case of B. cinerea, our study represents the first description of activation of plant defences against the pathogen mediated by B. siamensis.
Finally, regarding the two strains of B. pumilus, we have verified how the strain SCFC 1-2 would be able to reduce the damage caused by B. cinerea in pepper fruits thanks to the direct antagonism and activation of the plant defences, while the FVA 1-5 strain did not show plant defences activation. The ability of B. pumilus to activate plant defences has been previously described in tobacco and pepper plants against foliar infection by the bacteria Xanthomonas axonopodis, through the induction of SA-related genes Yi et al., 2013).
The results obtained in the in planta study showed that the bacterial species that best reduces the disease caused by V. dahliae was B. siamensis. In addition, B. proteolyticus and B. megaterium also reduced disease, to a lesser degree than B. siamensis.
In the control of V. dahliae in F-BP roots by B. siamensis mechanisms of both direct antagonism and activation of plant defences would be involved. In this sense, the ability of the bacterium to activate plant defences systemically when inoculated into roots has been previously described (Xie et al., 2021), its ability to activate defence genes locally in roots has not been previously described. Therefore, our study represents the first description of activation of root defences by B. siamensis inoculation and the first description of its effectiveness against the pathogen V. dahliae.
In the case of the bacteria B. proteolyticus, we have described its ability to control V. dahliae in pepper roots also through direct antagonism and activation of plant defences. These results represent a highly relevant finding since B. proteolyticus had never been described as a BCA against V. dahliae and our study places it as one of the most effective bacterial species.
Finally, B. megaterium would be able to reduce the disease caused by V. dahliae in pepper roots mainly through the activation of plant defences mediated by JA (specific response against necrotrophic pathogens, such as V. dahliae). The ability of B. megaterium to induce systemic resistance in plants has been previously described in cucumber roots against the oomycete Pythium aphanidermatum (Liang et al., 2011) and in Camellia sinensis roots against the fungus Fomes lamaoensis (Chakraborty et al., 2006). Against V. dahliae, the present work would be the first description of the ability of B. megaterium to control the pathogen due to the activation of plant defence responses, since in the direct in vitro interaction we found that the bacteria promoted the growth of the pathogenic fungus.
Our results show how the best protection effect against B. cinerea and V. dahliae occurs when SArelated genes (PAL1 and PR1) are downregulated, and JA-related genes (LOX1 and DEF1) are upregulated. The SA pathway activates plant defences against biotrophic pathogens, whereas the JA pathway activates plant defences against necrotrophic pathogens (Li et al., 2019;Thaler et al., 2012). Therefore, SA pathway activation is known to compromise plant defence against necrotrophes through JA pathway suppression (Li et al., 2019). Our results are consistent with those studies, by showing a clear antagonism between SA-related genes and JA-related genes activations. Those bacterial strains that downregulate SA-related genes and upregulate JA-related genes (B. siamensis SCFB 2-2, B. siamensis SCFB 3-4, B. thuringiensis FVA 2-3 and B. safensis MTA 1-2 in the case of B. cinerea infection; and B. siamensis SCFB 2-2, B. siamensis SCFB 3-4, B. proteolyticus SCFC 2-2 and B. megaterium MMM 3-6 in the case of V. dahliae infection) are the ones that show the best results in fruit injury area, aerial biomass, ion leakage and viability. JA-related genes activation explains this protective effect as B. cinerea and V. dahliae are necrotrophic pathogens. However, the mechanism used by the strains to activate the JA pathway has not been uncovered by our investigation, thus requiring further research in order to know how it is done. One possibility is the production of molecules that mimic JA by the bacteria (Thaler et al., 2012), but they have not been detected in the present study.
In conclusion, the endophytic species B. siamensis is the best choice as a biological control agent for the main diseases suffered by the F-BP. This choice is because B. siamensis can reduce the disease caused by B. cinerea in fruits and V. dahliae in roots, through direct antagonism and activation of plant defences. Although these results have been obtained under laboratory and greenhouse conditions (with local soil), the application and efficiency of the treatment in field and storage conditions can give very different results, as there are other variables, such as microbiota already present in the soils and fruits, or adverse weather conditions. Therefore, tests with B. siamensis should be carried out in real situations and its effectiveness in the agricultural system of the area should be determined.
Author contribution J.P. is the Principal Researcher of the research project. J.P., M.B. and F.G.-A. thought and designed the work. J.P. and J.C. performed the assays and analyzed the results. M.B. provided the endophytic bacteria used in the study. J.P. wrote the first version of the manuscript. J.C., M.B. and F.G.-A. contributed to the manuscript correction and critical reading, as well as to the knowledge on the bacteria field. All authors have read and agreed to the published version of the manuscript.
Funding Open Access funding provided by Universidad Pública de Navarra. This work was supported by the Instituto de Estudios Zamoranos "Florián de Ocampo" (IEZFO) as part of the XXVIII Call for Research Grants in 2020.
Data availability Not applicable.
Code availability Not applicable.

Declarations
Conflict of interest The authors declare that they have no conflict of interest.
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://creativecommons.org/licenses/by/4.0/ .