Induced Systemic Resistance in the Bacillus spp.—Capsicum chinense Jacq.—PepGMV Interaction, Elicited by Defense-Related Gene Expression

Induced systemic resistance (ISR) is a mechanism involved in the plant defense response against pathogens. Certain members of the Bacillus genus are able to promote the ISR by maintaining a healthy photosynthetic apparatus, which prepares the plant for future stress situations. The goal of the present study was to analyze the effect of the inoculation of Bacillus on the expression of genes involved in plant responses to pathogens, as a part of the ISR, during the interaction of Capsicum chinense infected with PepGMV. The effects of the inoculation of the Bacillus strains in pepper plants infected with PepGMV were evaluated by observing the accumulation of viral DNA and the visible symptoms of pepper plants during a time-course experiment in greenhouse and in in vitro experiments. The relative expression of the defense genes CcNPR1, CcPR10, and CcCOI1 were also evaluated. The results showed that the plants inoculated with Bacillus subtilis K47, Bacillus cereus K46, and Bacillus sp. M9 had a reduction in the PepGMV viral titer, and the symptoms in these plants were less severe compared to the plants infected with PepGMV and non-inoculated with Bacillus. Additionally, an increase in the transcript levels of CcNPR1, CcPR10, and CcCOI1 was observed in plants inoculated with Bacillus strains. Our results suggest that the inoculation of Bacillus strains interferes with the viral replication, through the increase in the transcription of pathogenesis-related genes, which is reflected in a lowered plant symptomatology and an improved yield in the greenhouse, regardless of PepGMV infection status.


Introduction
The disease management caused by viruses represents high costs and generates serious economic losses in agricultural productions worldwide [1]. The begomovirus genera, family Geminiviridae, is the most diverse and widespread member of the family worldwide, and it includes more than 400 reported species, including Pepper golden mosaic virus

Induced Systemic Resistance by Bacillus spp. in Capsicum chinense Jacq. against PepGMV
Symptoms in plants infected with PepGMV were first observed at 7 days post-inoculation (dpi) ( Figure 2B). The first symptoms were downward leaf curling and yellow spots. In plants treated with B. subtilis K47, B. cereus K46, and Bacillus spp. M9, the symptoms of yellow spots, leaf curl, and dwarfism, characteristic of a Level 1 PepGMV severity, were observed until 14 dpi ( Figure 2H-J). Symptoms were attenuated in the treatment with the B. subtilis K47-PepGMV ( Figure 2H). PepGMV-infected plants showed severe golden mosaics, leaf curling, and dwarfing symptoms ( Figure 2G), which corresponded to Level 4 on the severity scale. Mock-inoculated plants did not present any symptoms

Induced Systemic Resistance by Bacillus spp. in Capsicum chinense Jacq. against PepGMV
Symptoms in plants infected with PepGMV were first observed at 7 days postinoculation (dpi) ( Figure 2B). The first symptoms were downward leaf curling and yellow spots. In plants treated with B. subtilis K47, B. cereus K46, and Bacillus spp. M9, the symptoms of yellow spots, leaf curl, and dwarfism, characteristic of a Level 1 PepGMV severity, were observed until 14 dpi ( Figure 2H-J). Symptoms were attenuated in the treatment with the B. subtilis K47-PepGMV ( Figure 2H). PepGMV-infected plants showed severe golden mosaics, leaf curling, and dwarfing symptoms ( Figure 2G), which corresponded to Level 4 on the severity scale. Mock-inoculated plants did not present any symptoms (Figure 2A,F,K).
Viral symptoms were less severe in plants infected with PepGMV at 21 dpi (Level 2 severity, Figure 2L) compared to 14 dpi, with Level 4 severity ( Figure 2G). The symptoms in plants inoculated with the Bacillus strains were similar at 14 dpi with Level 1 severity ( Figure 2M,O), except for the plants inoculated with the M9 strain, where yellow mosaics and leaf curling were observed, which corresponded to Level 2 severity ( Figure 2O). PepGMV was detected in all the plants inoculated with the virus, and was determined by PCR amplification. Taken together, these results indicate that Bacillus spp. induces resistance to PepGMV in pepper plants.

Accumulation of PepGMV in Capsicum chinense
The relative accumulation of viral DNA was quantified using the PepGMV-AC2 gene as a qPCR target. Viral DNA was detected at 7 dpi in all treatments, with mean differences among them. In PepGMV-infected plants, the viral DNA titer was up to 25 times lower than the viral DNA detected in plants inoculated with the K47 strain, and 9 and 17 times lower than in those treated with the K46 and M9 strains, respectively ( Figure 3). In contrast, a significant reduction in the viral accumulation was observed in plants inoculated with the Bacillus spp. K47, K46, and M9 at 14 and 21 dpi. The viral accumulation of PepGMV in plants without the inoculation of Bacillus spp. K47, K46, and M9 increased over time; the highest values were observed at 7 dpi and then decreased at 14 and 21 dpi, with mean differences among treatments ( Figure 3). These results strongly suggest that the strains of Bacillus spp. are involved in the reduction in the viral titer in PepGMV-infected plants.

Effect of Plan Inoculation with Bacillus in Gene Expression of Genes Related with the ISR
After infection with PepGMV, the expression levels of the CcNPR1 gene increased in pepper plants treated with Bacillus. A 3-fold increase was observed in leaves from inoculated plants with B. cereus K46 when compared to the mock treatment at 2 hours post inoculaction (hpi). The expression levels of the CcNPR1 gene were significantly higher in plants inoculated with B. subtilis K47 at 24 hpi ( Figure 4A), with a 19-fold increase when compared to mock-treated plants. The expression levels of CcNPR1 decreased as the inoculaction (hpi). The expression levels of the CcNPR1 gene were significantly higher in plants inoculated with B. subtilis K47 at 24 hpi ( Figure 4A), with a 19-fold increase when compared to mock-treated plants. The expression levels of CcNPR1 decreased as the PepGMV infection progressed in Bacillus-treated plants, although in PepGMV-inoculated plants without Bacillus, a 9.5-fold increase in CcNPR1 was observed at 7 dpi when compared to the mock treatment ( Figure 4B). The aforementioned results suggest that treatments with B. cereus K46 and B. subtilis K47 sharply increased CcNPR1 gene expression in pepper plants infected with PepGMV, and the increase will depend on the strain used.  Gene expression levels of CcPR10 gradually increased in plants treated with B. subtilis K47, which showed a 10-fold increase and 190-fold increase at 8 and 12 hpi, respectively, when compared to the mock treatments ( Figure 4C). The CcPR10 gene expression levels showed a 30-fold increase in plants treated with Bacillus spp. M9 at 21 dpi, when compared to mock treatments ( Figure 4D). These results indicate an involvement of the CcPR10 gene during the early stages of the PepGMV infection in plants inoculated with B. subtilis K47.
CcCOI1 gene expression increased significantly at 24 hpi in leaves treated with B. subtilis K47 and Bacillus spp. M9 when compared to mock-treated plants, with a 16and 57-fold increase ( Figure 4E), respectively. CcCOI1 gene expression levels remained significantly higher in pepper plants treated with Bacillus spp. M9, when compared to mock treatments with a 16-and 27-fold increase at 24 hpi and 21 dpi, respectively ( Figure 4F,G). The above results indicate that treatment with Bacillus spp. markedly increased the expression of the CcCOI1 gene in pepper plants infected with PepGMV progressively at more advanced stages of the disease.

Greenhouse Yield of Plants during the Bacillus spp.-C. chinensee-PepGMV Interaction
The agronomic traits evaluated at the greenhouse level were total yield, fruit number, and fruit weight, which were quantified at 200 dpi. Plants without Bacillus inoculation (H 2 O) and plants treated with the K47 strain had the highest yield across all analyzed treatments (Tukey, α = 0.05) ( Table 1). The results showed that the yields obtained in plants treated with three strains of Bacillus and inoculated with the PepGMV had mean differences compared to the plants treated with water and the plants inoculated with the virus and without Bacillus spp.  (Table 1). Taken together, these results showed that seeds of C. chinense treated with different strains of Bacillus spp. increased the habanero pepper greenhouse production in plants infected with PepGMV (Table 1).

Discussion
In this study, our results indicate that the accumulation of PepGMV DNA increased over time in plants without Bacillus. These results have been reported previously in symptomatic pepper leaves infected with PepGMV, which showed a high accumulation of viral DNA and RNA, due to high rates of replication and transcription [7]. In contrast, in this study, we observed that in plants treated with Bacillus spp., the PepGMV viral titer and the symptoms decreased over time. Different reports demonstrate that inoculation with Bacillus spp. can reduce viral replication in infected plants [26,37]. Viral replication and movement are fundamental processes in the cycle of the disease; if both are affected, the result is a low viral concentration, and as a consequence, a decrease in symptoms [38]. Given that the PepGMV infection is associated with mechanical wounding, a mock treatment was prepared; during the early phases of the experiments, increases in the CcNPR1, CcPR10, and CcCOI1 was observed when compared to plants without mechanical damage. The observed up-regulations of these genes suggest that they are responsive, to some extent, to mechanical damage; this type of damage is caused by insect vectors (B. tabaci) during viral transmission [10,15].
Some studies indicate that NPR1 is a plant gene, and it has been observed that its expression is at low levels in healthy plants [32]. Moreover, the evidence demonstrates that NPR1 is a fundamental component of the pathway of the SA-mediated signal transduction pathway, inducing defense genes [16,39]. We observed that in plants inoculated with B. cereus K46, CcNPR1 levels sharply increase at 2 hpi, and these increases are statistically higher than in plants infected with PepGMV and not inoculated with Bacillus strains. It was suggested that during ISR by Bacillus spp. in C. chinense, CcNPR1 is involved in the defense response to PepGMV immediately after infection, and their behavior is specific to each Bacillus strain. Although most strains of B. cereus are recognized as pathogenic microbes for humans, there are some strains used in the bio-fertilization and biological control of plant viruses [40][41][42]. In contrast, in plants infected with PepGMV and without Bacillus inoculation, the highest expression of CcNPR1 was observed at 7 dpi. Similar studies report that in C. annuum plants infected with Euphorbia mosaic virus-Yucatan Peninsula (EuMV-YP), the expression of NPR1 increased at 7 dpi [32]. After the accumulation of SA in response to a pathogen attack, NPR1 oligomer disassociates in the cytoplasm, and after this, the monomer is translocated into the nucleus, and together with TGA transcription factors, they induce the pathogenesis-related gene (PR's) expression [38,39].
The PR protein family has a complex pattern of expression, and many members of this family of genes are differentially expressed under conditions of environmental stress and in response to pathogen attacks within the signaling pathway systemic acquired resistance (SAR) [24,39]. In this study, the transcript levels of CcPR10 were highest at 8 hpi in plants inoculated with B. subtilis K47, and its expression was statistically higher than in plants without rhizobacteria. In this way, several studies have shown that the Tobacco mosaic virus (TMV), Cucumber mosaic virus (CMV), Tobacco etch virus (TEV), and Tobacco vein mottling virus (TVMV) trigger the activation of PR10, and this protein functions as a ribonuclease [43,44]. Furthermore, it was demonstrated that the inoculation of leaves in C. annuum L. 'Bukwang' with B. amyloliquefaciens 5B6 caused an increased in the expression of PR10 during CMV infection [45]. Therefore, PR10 could not only be activated during the SAR, but it also participates in the ISR in viral diseases.
It has been known that the COI1 gene is the central regulatory component of the signaling pathway of SA/JA, and it is required for the defense responses of plants [28,39]. Recently, studies have reported that the C2 protein of geminivirus suppresses the defense response signaling pathway mediated by jasmonates because this protein affects the functioning of complex SCFCOI1 [46,47]. However, our data indicate that the CcCOI1 levels of C. chinense plants treated with Bacillus spp. increased consistently for 7 dpi during the disease caused by PepGMV. Similarly, in tobacco plants infected with TMV and inoculated with Bacillus spp., the activation of the genes NtPR1, NtCOI1, and NtNPR1 has been observed, generating a modulated ISR by Bacillus spp. [28]. These results suggest that despite the disease caused by PepGMV, Bacillus spp. promotes ISR through increased levels of transcripts of CcCOI1.
Multiple investigations on plant-rhizobacteria-pathogen interactions have demonstrated the benefits in disease resistance and increased crop yield [48,49]. In this study, we consistently observed that plants treated with B. subtilis K47 had less severity of symptoms and better fruit quality (larger fruits) despite viral infection than healthy plants. In this sense, in tomato plants treated with the B. amyloliquefaciens strain MBI600, resistance to Tomato spotted wilt virus and Potato virus Y was increased through the salicylic acid pathway [1]. Similarly, the application of B. velezensis CE 100 on strawberry plants controls fungal diseases and improves yield [50].
Previously, our studies showed that B. subtilis K47 increased the photosynthetic parameters of the plant and prepared it for stress situations, such as viral diseases. Furthermore, this study reveals that inoculation with B. subtilis K47 increases the defense gene expression of CcNPR1, CcPR10, and CcCOI1, decreases the viral titer and severity of symptoms, and increases yield. It is evident that Bacillus activates ISR through a complex network of mechanisms, not only defending the plant from pathogen attacks, but also allow it to continue producing fruit. These results could be interesting in evaluating their effect as complex bacteria on the response to biotic and abiotic stresses during agricultural sustainable production.

Selection of Plant Material, Germination, and Growth Conditions
Seeds of C. chinense, accession H-224, were used [51]. The disinfection procedure consisted of the immersion of C. chinense seeds in a solution of commercial bleach (sodium hypoclorite, 2% v/v) for 15 min; the seeds were then rinsed three times with sterile distilled water and air-dried in absorbent paper. The seed germination was performed in trays with 200 cavities filled with sterile substrate (peat moss). The substrate was moistened up to field capacity with distilled sterile water. The trays were placed in a growth room (25 ± 2 • C, photoperiod of 16/8 light/dark), and they were watered every two days and leaf-fertilized weekly at a dose of 1 gL −1 (UltraFol, Biochem systems, Querétaro, México. After 18 days of germination, the seedlings were transferred into 500 mL Styrofoam cups, filled with sterile substrate, and maintained in the same controlled conditions [36].

Severity Scale in C. chinense Plants under Controlled Conditions
The symptoms caused in C. chinense seedlings bioballistically infected with PepGMV were evaluated at 9 and 15 dpi in a growth room under controlled conditions, the scale of the severity of the symptoms was modified from Samaniego-Gámez et al. [36], with the following values: 1. Golden mosaics; 2. Golden mosaics and leave distortion; 3. Golden mosaics, leave distortion, and chlorosis; 4. Golden mosaics, leave distortion, chlorosis, and leaf curling. The severity analyses were assessed from ten observations per treatment [54].

Viral Titer Determination in C. chinense
Leaves of C. chinense from systemically infected plants with PepGMV were sampled (1 g), and the total DNA was isolated with CTAB, as described by Doyle and Doyle [55] without modifications. The detection reactions for the virus were performed within a thermalcycler (TC-412, Techne, Bibby Scientific Ltd., Chicago, IL, USA), using a 100 ng of total DNA and primers that targeted the AC2 gene ( Table 2)  A gene normalization analysis was performed in order to determine the most stable housekeeping genes, such as Actin, 18S, Gliceraldehyde-3-phosphate-dehydrogensae, Ubiquitin, and β-Tubulin, with the geNorm software (v.3.0, qBASE, Beavercreek, OH, USA) The normalization showed that β-Tubulin was the most stable gene among the samples, which agrees with previous reports in Capsicum [56,57]. The viral titter was determined with the 2 −∆∆Ct method [56,58], normalized with β-tubulin (Table 2), and expressed as RDA.

Relative Expression of CcNPR1, CcPR10, and CcCOI1 by Real-Time PCR
To determine the transcript levels of CcNPR1, CcPR10, and CcCOI1 genes, leaf tissue was collected during the time course of the disease, 0 h before infecting with the virus, 2, 4, 8, 12, and 24 h after infection with PepGMV, and 7, 14, and 21 days after infection with PepGMV. Total RNA was isolated from leaves with the TRIZOL reagent, as described by Chomczynski and Sacchi [44]. The isolated RNA was treated with TURBO DNase (2238, Ambion, Life Technologies, Sunnyvale, CA, USA), and 2.5 µg was used for the cDNA synthesis with the SuperScript III Reverse Transcriptase (18080-044, Invitrogen, Carlsbad, CA, USA) and oligodT 18 (Table 2) and expressed as relative transcript levels with the 2 −∆∆Ct method [56][57][58].

Evaluation of Agronomic Parameters
Plants at 28 dpi were transferred into black polyethylene bags (400 gauge) of 5 kg capacity (35 cm diameter, 40 cm height) filled with sterile substrate, and maintained in a greenhouse under controlled conditions (30 ± 2 • C, 65 ± 3% HR and 1100 mmol luminous intensity). The agronomic parameters were performed as previously described [36].

Experimental Design
A complete randomized design was used with 30 plants per treatment. For qPCR experiments, three biological replicates per treatment were analyzed. PCR products were cloned, sequenced, and compared via nucleotide BLAST in the NCBI. The detection of mean differences in each treatment was examined by an ANOVA with the Tukey's HSD test at p ≤ 0.05 (Statistica, v.7.0.0, StatSoft Hamburg, Germany).

Conclusions
The seed inoculation with Bacillus promotes the ISR in C. chinense plants, which is reflected in the reduction in the viral accumulation and the symptom severity, which suggests that Bacillus could participate in the ISR through various mechanisms, such as the inhibition of the viral replication and an increase in the transcription rate of defense genes. Therefore, the ISR is a mechanism that could be implemented in disease management programs in sustainable agriculture. Future research should be focused on the study of the viral movement in Bacillus-treated plants and the transcriptional profile of the Bacillusplant-geminivirus interaction in order to elucidate the mechanisms involved at transcript and the protein profile of this type of interaction.