Introduction

It is well established that pathogen infection triggers modifications in photosynthesis processes and consequently carbohydrate metabolism. On one side, the pathogen redirects sugars for its own benefit and on the other side, the plant shut down its primary metabolism to develop its defense strategy. Plants have to make trade-offs between stress defense responses and primary metabolism (Demmig-Adams et al. 2017; Foyer and Noctor 2005; Karpinski et al. 2003). The chloroplast, in which photosynthesis takes place, is now considered as a key defense organelle as it is able to perceive external signals occurring during biotic or abiotic stresses (Serrano et al. 2016). Some pathogen effectors are able to interfere with some chloroplastic functions to facilitate their multiplication (de Torres Zabala et al. 2015). It was recently demonstrated that Sclerotinia suppress host defense via antagonizing ABA biosynthesis by manipulating the xanthophyll cycle in early pathogenesis (Zhou et al. 2015), suggesting that photoprotective metabolites could be integrated into the defense responses. Oxylipin (OPDA and/or JA) levels are increased in photoprotection mutants either deficient in components involved in thermal dissipation and/or detoxification (Demmig-Adams et al. 2013). By using Arabidopsis over-expressing or lacking PsbS (a chlorophyll-binding protein of photosystem II), it was shown that differences in herbivore preferences towards plants were due to differences in the primary metabolism of these plants rather than in their contents of typical defence compounds (Johansson Jänkänpää et al. 2013). Similarly, PsbS-deficient rice plants were also more resistant to fungal and bacterial pathogens (Zulfugarov et al. 2016). Moreover, redox metabolism and related signaling are key players in tolerance to biotic stresses in plants (Munné-Bosch et al. 2013). The foliar hemibiotrophic bacterium Pseudomonas syringae pv. tomato strain DC3000 (Pst) is able to cause disease symptoms in a wide range of plant species, including Arabidopsis (Xin and He 2013). Arabidopsis recognizes type III virulence effectors of Pst (Shan et al. 2007; Xin and He 2013) and led to activation of several defense mechanisms, including stomatal closure (Boureau et al. 2002; Melotto et al. 2008), programmed cell death at the infection site (Lenz et al. 2011; Dong and Chen 2013) and activation of salicylic acid (SA) signaling pathway (Glazebrook 2005), which could lead to systemic resistance (Mishina and Zeier 2007). Moreover, two Pst effectors are imported into chloroplasts, meaning that this bacterium targets the chloroplast during its infection process (de Torres Zabala et al. 2015). Thilmony et al. (2006) and Cartieaux et al. (2008) showed an induction of defense-related genes correlated with a downregulation of photosynthetic genes. Infection of Arabidopsis with Pst also induces modifications on photosynthesis, carbon metabolism and carbohydrate distribution in tomato plants (Berger et al. 2004) and Arabidopsis (Bonfig et al. 2006; de Torres Zabala et al. 2015). However, repression of photosynthetic parameters was restricted at the site of infection (Berger et al. 2004; Bonfig et al. 2006). Infections with the bacterium inhibit photosynthetic CO2 assimilation through disruption of photosystem II. Li et al. (2015) showed that elevated CO2 concentration-induced stomatal closures not only reduce entry of Pst by controlling stomatal apertures, but also involve a stomata-independent pathway to resist against Pst.

Plants are sessile organisms and do not possess an adaptive immune system. To restrict pathogen infection, plants have evolved an array of defense mechanisms. The success of plant resistance firstly relies on the capacity of the plant to recognize its invader. Plants detect the presence of microorganisms by microbe- or pathogen-associated molecular patterns, such as bacterial lipopolysaccharides, flagellin, and fungal chitin (Boller and Felix 2009; Henry et al. 2012; Walters 2015). Pattern recognition by specific receptors located on the plant plasma membrane is required to trigger plant innate immune mechanisms (pattern-triggered immunity, PTI) (Bittel and Robatzek 2007; Walters 2015). Some pathogens are able to bypass this first line of plant defense by preventing host plant detection or by deleting PTI signals (Dodds and Rathjen 2010; Walters 2015). However, many plants could recognize specific effectors of pathogens, resulting in more powerful defense responses (effector–triggered immunity, ETI) (Dodds and Rathjen 2010). Activation of the PTI or the ETI limits the development of pathogens at the site of infection, and could trigger an induced resistance in intact tissues with one or more long-distance signals (Newton et al. 2014; Walters 2015). The plant hormones SA, jasmonic acid (JA), ethylene (ET), and abscisic acid (ABA) are key players in defense signaling network. In Arabidopsis, SA is involved in plant resistance against (hemi)biotrophic pathogens, whereas JA and ET are thought to be involved in resistance against necrotrophic pathogens (Thomma et al. 2001; Glazebrook 2005). Pathogen perception then initiates a large array of immune responses including modification of cell walls as well as the production of anti-microbial proteins and metabolites like pathogenesis-related (PR) proteins (Schwessinger and Ronald 2012).

Several rhizobacteria, the so-called plant-growth-promoting rhizobacteria (PGPR), are capable to promote plant growth through uptake facilitation of some soil nutrients (Vessey 2003; Yadav et al. 2015), modification of phytohormones homeostasis (Castillo et al. 2015; Kumar et al. 2015), and/or improvement of photosynthesis efficiency (Stefan et al. 2013; Cohen et al. 2015). Some PGPR could also restrict pathogen developing in plants (Compant et al. 2005; Wang et al. 2015). PGPR could inhibit directly pathogen infection and propagation by their antagonistic activity, including production of antibiotics, siderophores, and hydrolytic enzymes (Vessey 2003; Somers et al. 2004; Pathma et al. 2011). Indirectly, PGPR application could activate plant defense mechanisms to lead plant in a systemically resistant state (induced systemic resistance, ISR) (van Loon et al. 1998; Kloepper et al. 2004; Chandler et al. 2008). The perception of flagellin from Bp by Arabidopsis cells induced defense responses, like production of reactive oxygen species and expression of defense-related genes (Trdá et al. 2014). Bp has been shown to protect plants against abiotic stresses such as cold (Su et al. 2015; Theocharis et al. 2012), drought (Naveed et al. 2014), or salt (Pinedo et al. 2015), or biotic stresses such as the necrotrophic pathogen Botrytis cinerea (Miotto-Vilanova et al. 2016). However, little is known about the impact of Bp inoculation on an attack by a (hemi)biotrophic pathogen (Sharma and Novak 1998). Furthermore, mechanisms underlying the induced resistance are still unclear. The present study thus aims to investigate if Bp could be efficient to protect plants against a hemibiotrophic pathogen. We first evaluated the eventual direct antibacterial effect of Bp on Pst growth. Since biotic stress and photosynthesis are intimately linked, photosynthetic parameters (PSI and PSII activity, RuBisCO levels) and defense-related gene expression were quantified when Arabidopsis plants were co-inoculated with the two bacteria. To decipher how the presence of Bp in the plant could modify tolerance to Pst attack, Bp was locally (leaf) or systemically (root) inoculated.

Materials and methods

Plant material and growth conditions

Arabidopsis thaliana ecotype Col-0 plants were grown in soil condition in a controlled chamber at 20/15 °C (day/night), with 60% of relative humidity and a 12-h photoperiod (photosynthetically active radiation, PAR = 120 µmol m−2 s−1). For all of the experiments, measurements were performed on mature leaves of 6-week-old plants. Three plants were used per treatment with three biological replicates.

Bacterial strains and inoculation

PGPR Burkholderia phytofirmans strain PsJN (Bp) tagged with GFP and pathogen Pseudomonas syringae pv. tomato strain DC3000 (Pst) were grown for 24 h at 28 °C at 180 rpm in King’s B liquid medium supplemented with 50 µg mL−1 antibiotics (kanamycin and cycloheximide or rifampicin for Bp or Pst, respectively). Bacteria were collected after centrifugation at 4500×g for 10 min and suspended in 10 mM MgCl2. The concentration of bacterial inocula was adjusted by spectrophotometry at 600 nm (Pillay and Nowak 1997).

For direct antibacterial effect on growth assays, Pst (103 colony forming units per mL, cfu mL−1) or Bp (103 cfu mL−1) were inoculated alone or side by side on the King’s B medium supplemented with kanamycin (50 µg mL−1). Plates were then placed in a growth chamber at 28 °C. Colony growth of the two bacteria was observed under UV 365 nm to assess possible direct antagonism between these two bacteria.

Bp inoculum (105 cfu mL−1) was infiltrated into Arabidopsis mature leaves using a needleless syringe (Bp). Control plants (Mock) were infiltrated with 10 mM MgCl2. Arabidopsis seeds were immersed in bacterial inoculum of 5.108 cfu mL−1 (SBp) or PBS for 3 h at 4 °C (Mock). Pst infection was performed 8-h post Bp application by dipping Bp and Mock plants in Pst suspension (108 cfu mL−1) supplemented with Silwet L77 0.02% (Bp + Pst and Pst, respectively), or 10 mM MgCl2 Silwet L77 0.02% as Mock. 3 days after Bp and Pst inoculation, Pst symptoms were photographed on Pst and Bp + Pst inoculated plants.

Protein extraction and western blotting analysis

Total proteins were extracted from 0.2 g of leaf with 500 μL cold extraction buffer (250 mM sorbitol, 50 mM Tris–HCl, pH 8.0, 2 mM EDTA, 7 g L−1 PVPP, 5 mM DTT, 1 mM PMSF and 1/100 Halt Protease Inhibitor Cocktail-Thermo Scientific) and centrifuged at 10,000×g for 10 min at 4 °C. The supernatant was then collected and proteins were quantified by the Bradford method using bovine serum albumin as the standard (Bradford 1976). Protein samples (2 μg) were solubilized for 3 min at 95 °C in Laemmli buffer (Laemmli 1970) and separated by SDS-PAGE in 12% (w/v) polyacrylamide gels, using Mini-protean three Cell electrophoresis equipment (Bio-Rad). Proteins were electro-transferred onto a polyvinylidene difluoride (PVDF) membrane using iBlot system (Invitrogen). Western blotting was performed according to standard procedures using rabbit anti-RbcL or -RbcS antibodies (Agrisera; 1:10,000) and peroxidase-coupled anti-rabbit IgG antibodies (Cell signaling; 1:5000). Actin (Agrisera; 1:1000) was used as internal quantification control.

RNA extraction and real-time RT-PCR

RNA extraction and real-time RT-PCR analysis were performed as described by Magnin-Robert et al. (2015). For each sample, 100 mg of leaves was ground in liquid nitrogen. Total RNA was isolated using Extract’ All (Eurobio) and followed reverse transcription by using the Verso cDNA Synthesis Kit (Thermo Scientific) according to the manufacturer’s instructions. The transcript levels were determined by qPCR using the CFX 96TM Real Time System (Bio-Rad, France) and the SYBR Green Master Mix PCR kit as recommended by the manufacturer (Applied Biosystems). PCR conditions were 95 °C for 15 s (denaturation) and 60 °C for 1 min (annealing/extension) for 40 cycles on CFX 96TM Real Time System (Bio-Rad, France). Traditional reference genes were evaluated with Bio-Rad CFX MANAGER software v.3.0 (Actin2, UBQ5, UBQ10, EF1α, and Tubulin2) to select a reference gene with a stable expression in all tested conditions (Hong et al. 2010). The expression stability geNorm M value of UBQ5 was below the critical value of 0.5 in Arabidopsis samples. Transcript level and Pst relative quantity were calculated using the standard curve method and normalized against UBQ5 gene as an internal control. The specific primers used in this study were listed in Table S1 in Supplementary Material.

PSI and PSII photochemistry

Chlorophyll fluorescence parameters and the redox change of P700 were assessed simultaneously with a Dual-PAM-100 measuring system (dual-wavelength pulse-amplitude-modulated fluorescence monitoring system, Heinz Walz, Germany) on Arabidopsis leaves.

PSII photochemistry

Leaves were dark adapted for 30 min to determine the minimal level of fluorescence (F0) and the maximal fluorescence (F m) after a saturating flash (1 s, 13,000 μmol m−2 s−1). The ratio of variable to maximal fluorescence [F v/F m = (F m − F 0)/F m] was calculated. Actinic illumination (216 μmol m−2 s−1) was applied and after fluorescence stabilization, a second saturating flash (1 s) was imposed to determine the maximal fluorescence (F m′) of a light-adapted inflorescence. Removal of the actinic light and exposure to a short period of far-red light allowed measurement of the zero level of fluorescence (F 0′). In both dark- and light-adapted states, the fluorescence parameters were calculated according to Genty et al. (1989) and Schreiber et al. (1994). Quantum yields, designated by YII [YII = (F m′ − F)/F m′] for photochemical energy utilization in PSII, YNPQ for regulated energy dissipation in PSII and YNO for non-regulated energy dissipation in PSII, were calculated according to Kramer et al. (2004). The electron flow through PSII (ETRII) was calculated according to Miyake et al. (2005) according to ETRII = YII × PPFD × αII [αII = fraction of the incident light absorbed by organ (p) × fraction of the absorbed light distributed to PSII (dII)].

PSI photochemistry

Together with fluorescence measurement, the saturation pulse method was used to determine P700 parameters following the method of Klughammer and Schreiber (1994, 2008). The P700+ signals may vary between a minimal (P700 fully reduced, closed) and a maximal level (P700 fully oxidized). P700 fully oxidized (P m) was determined by application of a saturation pulse after far-red pre-illumination. YNA, the quantum yield of non-photochemical energy dissipation due to acceptor side limitation, was calculated based on a P m′ determination at 216 μmol m−2 s−1 actinic light according to the following: YNA = (P m − P m′)/P m. YND, the non-photochemical quantum yield of PSI due to donor side limitation, was calculated according to the following: YND = 1 − P700 red. YI, the photochemical quantum yield of PSI, was defined by the fraction of overall P700, which is reduced and not limited by acceptor side. YI was calculated from the complementary PSI quantum yields of non-photochemical energy dissipation, YND, and YNA according to the following: YI = 1 − YND − YNA. The electron transport rate of PSI, ETRI, was calculated by Dual-PAM software.

Statistical analyses

All experiments were repeated independently at least three times. Photosynthesis parameters considered to vary significantly between treatments were those with P < 0.05 using Student’s t-tests.

Results

No direct antagonistic relationship between Bp and Pst in vitro

The direct antifungal effect of Bp could partially explain the protection of grapevine against B. cinerea (Miotto-Villanova et al. 2016). We thus analyzed the antibacterial effect of this PGPR on the hemibiotrophic bacterium Pst. First, Pst growth on King’s B medium supplemented with either kanamycin or kanamycin and rifampicin was quantified. Pst growth was similar between the two media (supplemental figure S1), suggesting that removal of rifampicin from culture medium is not a stressful condition for Pst growth. Then, growth of Pst colonies with or without Bp was thus followed 2, 3, and 4 days after inoculation of the plate to assess in vitro a direct interaction between the two strains (Fig. 1). The strain Bp is labeled with GFP (green) and Pst produces a blue fluorescent pigment (Fackrell and Sinha 1983), allowing to distinguish between the two bacterium growth under UV light. In vitro, the presence of a strain does not inhibit the development of the other even after 4 days of co-culture. Bp has thus no direct antagonism against Pst or inversely.

Fig. 1
figure 1

Co-culture of Bp and Pst in vitro. Pst (103 cfu mL−1) was co-cultured with Bp (103 cfu mL−1) in King’s B medium supplemented with kanamycin (50 μg mL−1) at 28 °C. Observations of two bacterial colonies were carried out under UV. Photographs display the results of one representative experiment among three independent repetitions

Full size image

Only local presence of Bp alleviates Pst growth in planta during the early stage of infection

As previously demonstrated, Bp promotes Arabidopsis aerial plant growth when inoculated on seed or by root irrigation. Similarly to most of rhizobacteria, Bp is strictly localized to the root system independently of inoculation method (Su et al. 2015). In order to assess a potential ISR against Pst infection, Bp was seed inoculated. Moreover, in order to determine if Bp is able to protect locally Arabidopsis plants against Pst, Bp was also infiltrated in plant leaves (Su et al. 2016) prior to Pst infection. Bp-seed inoculated or -leaf infiltrated plants do not display symptoms (Fig. 2a, b). 3 days after infection, no difference between symptoms developed by Pst-infected plants subjected to leaf infiltration or seed inoculation with Bp was observed (Fig. 2a, b). Quantification of Pst was thus performed by qPCR. In seed bacterized plants, Pst growth was enhanced compared to non seed bacterized plants at 1 dpi (Fig. 2c; 26-fold). However, this growth stimulation was less induced at 3 dpi (Fig. 2c; 2.5-fold). Moreover, Pst only slightly grew between 1 and 3 dpi (×1.3-fold) in seed bacterized plants compared to non-bacterized plants (×14-fold). This suggested that Bp inoculation slightly slowed Pst spread when it is systemically present. Interestingly, Pst growth was drastically reduced in leaf bacterized plants at 1 dpi (Fig. 2d; 12-fold less). However, the presence of Bp at the same place than Pst has no influence on Pst growth 3 dpi.

Fig. 2
figure 2

Pst development in Arabidopsis leaves. The plants bacterized by Bp on seeds (S-; a, c) or leaves (L-; b, d) were dipped in a suspension of Pst (108 cfu mL−1) at the age of 6 weeks. Leaf symptoms caused by Pst (a, b) observed 3 days post infection. Photographs display the results of one representative experiment among three independent repetitions. Quantification of Pst (c, d) estimated by the abundance of Pst mRNA (PsyrOPRF) relative to the respective Pst sample on the first day post infection, referred to as the ×1 expression level. Values shown are mean ± SE of duplicate data from one representative experiment among three independent repetitions

Full size image

Priming of both SA- and JA-related genes by Bp

To determine whether the altered Pst growth in Bp-inoculated plants is linked to changes in the regulation of defense genes, the expression pattern of defense-related genes (PR1 and PDF1.2) was monitored before and after challenge with Pst (Fig. 3). PR1 is well-known SA-dependent defense marker gene and PDF1.2 expression is regulated by JA and ET (Glazebrook 2005). Bp inoculation, either in roots or in leaves, did not significantly trigger PR1 or PDF1.2 expression, except when performed in leaves at 3 dpi for PR1 (Fig. 3). As already described, colonization by Pst induced expression of PR1 in control plants (Fig. 3a, b). PDF1.2 expression is stimulated at the early stage of infection but this induction fell progressively down until the end of the time-course (Fig. 3c, d). In seed bacterized plants, induction of PR1 is earlier and stronger than in plants infected only with Pst. However, PR1 expression was similar in S-Pst and S-Bp/Pst plants at 3 dpi (Fig. 3a). In contrast, expression of PDF1.2 is continuously enhanced in seed Bp-treated plants compared to S-Pst plants (Fig. 3c). Infiltration of Bp in leaves did not alter PR1 or PDF1.2 expression after challenge with Pst (Fig. 3b, d).

Fig. 3
figure 3

Expression levels of the defense-related genes PR1 (a, b) and PDF1.2 (c, d) in Arabidopsis leaves after seed inoculation (S-; a, c) or leaf infiltration (L-; b, d) with Bp, infection by Pst, or co-inoculation with the two bacterial strains. Data represent mean fold increases in mRNA levels relative to those of control plants (Mock on the first day), referred to as the ×1 expression level. Values shown are mean ± SE of duplicate data from one representative experiment among three independent repetitions

Full size image

Photosystem I and II activities

Infection of plants with pathogens results in photosynthetic parameter disturbance. To evaluate the impacts on photosynthetic activity triggered by a PGPR, a virulent bacterium, or a co-inoculation with both bacteria, the regulation of the PSI and PSII activities was monitored.

PSI photochemistry

Excitation energy transferred to the PSI centers will result in photochemical charge separation with quantum yield ΦPSI or in non-photochemically conversion to heat (Nelson and Yocum 2006). The quantum yield of non-photochemical energy dissipation can be due to limitation of the acceptor-site Y(NA) or the donor-site Y(ND).

Seed inoculation of Bp altered neither the regulation of PSI (Y(NA), ΦPSI) nor the transport rate of electrons (ETRI) (Fig. 4a). A slight but significant decrease in non-photochemical efficiency due to the limitation of electron donor side, Y(ND), was measured in SBp at 3 dpi. At 3 dpi, infection with Pst caused an increase of Y(NA), which represents the limitation of the electron acceptor side, while ΦPSI and ETRI decreased (Fig. 4a). Infection with Pst in SBp increased Y(ND) at 2 dpi comparatively to the plant without Bp.

Fig. 4
figure 4

PSI photochemistry in Arabidopsis leaves after seed inoculation (S-; a) or leaf infiltration (L-; b) with Bp, infection by Pst, or co-inoculation with the two bacterial strains. PSI acceptor side limitation [Y(NA)], PSI donor side limitation [Y(ND)], efficient quantum yield of PSI (ΦPSI), and PSI electron transport rate (ETRI) are shown. Data are mean ± SE of three experimental replicates, each with three plants per treatment (n = 9). Same letters indicate non-significant difference among all conditions on the same day (Student’s t-test; P < 0.05)

Full size image

Leaf infiltration of Bp triggered a decrease in Y(ND) at 1 dpi, and an increase in ΦPSI and ETRI values at 3 dpi (Fig. 4b). Pst infection decreased ΦPSI and ETRI (2 dpi) (Fig. 4b). The presence of Bp did not modify the impact of Pst on PSI activity.

PSII photochemistry

The excitation energies absorbed by PSII centers are contributed in photochemical utilization (ΦPSII) or heat dissipation (Nelson and Yocum 2006). In PSII, the heat dissipation includes regulated and non-regulated energy dissipation [Y(NPQ) and Y(NO), respectively].

Seed inoculation by Bp did not provoke modifications of PSII parameters at all 3 days monitored (Fig. 5a). However, Pst invasion damaged PSII activity, presented by decreased Fv/Fm and Y(NPQ) values at 2 dpi, and ΦPSII and ETRII values at 3 dpi. Along with the decrease in ΦPSII, Pst-infected plants showed an increase in Y(NO) compared to mock-treated plants. A greater reduction of ΦPSII, Fv/Fm, and ETRII and a greater increase in Y(NO) at 2 dpi were observed in plants inoculated with Bp and infected by Pst compared to plants inoculated only with Pst.

Fig. 5
figure 5

PSII photochemistry in Arabidopsis leaves after seed inoculation (S-; a) or leaf infiltration (L-; b) with Bp, infection by Pst, or co-inoculation with the two bacterial strains. Quantum yield of non-regulated energy dissipation in PSII [Y(NO)] and regulated energy dissipation in PSII [Y(NPQ)], efficient quantum yield of PSII (ΦPSI), maximum photochemical efficiency of PSII (F v/F m), and PSII electron transport rate (ETRII) are shown. Data are mean ± SE of three experimental replicates, each with three plants per treatment (n = 9). Same letters indicate non-significant difference among all conditions on the same day (Student’s t-test; P < 0.05)

Full size image

Leaf infiltration of Bp did not induce modifications of PSII activity (Fig. 5b). Pst infection led to an increase of Y(NO) at 2 dpi and a decrease of ΦPSII, Y(NPQ), F v/F m, and ETRII. However, no change in PSII activity could be detected when plants were inoculated with Bp and infected by Pst compared to plants inoculated only with Pst.

RuBisCO accumulation

To complete photosynthetic activities, protein accumulation of two subunits of the RuBisCO (the nuclear-encoded RbcS and the chloroplast-encoded RbcL) was followed (Fig. 6). As previously shown (Su et al. 2015), Bp seed-inoculated plants did not modify RbcS or RbcL accumulation. However, no change in these protein accumulations was detected in bacterized or non-bacterized plants after challenge with Pst.

Fig. 6
figure 6

RbcL and RbcS accumulation in Arabidopsis leaves after seed inoculation (S-) or leaf infiltration (L-) with Bp, infection by Pst, or co-inoculation with the two bacterial strains. Normalization was carried out with Actin. Numbers on the right indicate molecular mass in kilodaltons. Data display the results of one representative experiment among three independent repetitions

Full size image

Discussion

Previous studies demonstrated that Bp protected grapevine against the necrotrophic pathogen Botrytis cinerea (Ait Barka et al. 2006; Miotto-Vilanova et al. 2016). However, only few information is available on the capability of Bp to protect plants against an attack by a (hemi)biotrophic pathogen (Sharma and Novak 1998). Here, we described the mechanisms underlying the interaction between Bp and the hemibiotrophic bacterium Pst. We focused on photosynthetic parameters and on some defense-related processes underlying the link between photosynthesis and biotic stress.

Defense mechanism regulation during infection with Bp and/or Pst

It is well known that PGPR could protect plants against pathogens by anti-microbial activities (Lugtenberg and Kamilova 2009). Recently, it was shown that Bp possesses antifungal activity against B. cinerea since it reduced its growth development (Miotto-Vilanova et al. 2016). This inhibition by Bp partially protect grapevine against B. cinerea. Here, co-culture of Bp and Pst did not display any antibiosis impact between these two bacteria.

Sur-expression of defense-related gene PR1 normally relates to SA-mediated signaling pathway, and responses to (hemi)biotrophic pathogens infection in Arabidopsis (Reddy 2013; Vidhyasekaran 2015), whereas induction of PDF1.2 expression is under control of JA and ET accumulation. Here, inoculation with Bp either on seeds or in leaves did not induce the accumulation of mRNA of PR1. In contrast to Poupin et al. (2013), inoculation of Bp on Arabidopsis plants did not induce PDF1.2 expression as well. This discrepancy might be due probably to the different methods of bacterial inoculation and plant growth conditions. Plants can establish the so-called “primed” state (Ton et al. 2005; Conrath 2009) following root colonization by beneficial microbes or after chemical treatment by activating their defense responses faster or more strongly when subsequently challenged by microbial pathogens (Pozo et al. 2008; Conrath 2009), resulting in a better tolerance towards this stress. The priming of PR gene expressions by Bp was described in grapevine in order to confer tolerance towards low non-freezing temperatures or B. cinerea (Theocharis et al. 2012; Miotto-Vilanova et al. 2016). Interestingly, Bp inoculation on Arabidopsis seeds conditioned the plant to produce PR1 and PDF1.2 mRNA more rapidly, suggesting that Bp potentiated the accumulation of PR1 and PDF1.2 mRNA during the first stage of virulent bacterium attack. Such a simultaneous activation of expressions of both SA- and JA-related genes has already been observed after inoculation with diverse PGPRs, including Bp (Conn et al. 2008; Niu et al. 2011; Brock et al. 2013; Miotto-Vilanova et al. 2016). All these data suggested that priming of PR1 expression could significantly slowed Pst growth in S-Bp/Pst plants compared to S-Pst plants at 3 dpi. Whereas PR1 potentiation is only transitory, PDF1.2 potentiation is maintained at 3 dpi. It is well admitted in Arabidopsis that SA pathway has antagonistic effect on JA signaling and reciprocally (Thaler et al. 2002; Bostock 2005; Glazebrook 2005; Spoel et al. 2007; Derksen et al. 2013). This hypothesized that activation of JA signaling via the extended induction of PDF1.2 expression could inhibit effect of PR1 potentiation, resulting in an absence of tolerance. This also hypothesized that Bp protect plants against necrotrophic pathogen but not against (hemi)biotrophic invaders.

Photosynthesis regulation during infection with Bp and/or Pst

In order to better understand the link existing between photosynthesis and biotic stress, different photosynthetic parameters were analyzed in response to BP and/or Pst infection. Our results showed small and transient modifications of PSI activity after Bp inoculation. A decrease in Y(ND) at 3 dpi (PSI) was recorded after seed bacterization. Leaf infiltration of Bp triggered a decrease in Y(ND) at 1 dpi, and an increase in ΦPSI and ETRI values at 3 dpi (PSI), without PSII modification. These data are in agreement with our previous results (Su et al. 2015). However, effects of PGPR are highly dependent on the plant–bacteria interaction (van Loon 2007). Ait Barka et al. (2006) reported that Bp-inoculated grapevine plantlets exhibit a higher photosynthetic activity compared to non-bacterized plantlets. Many PGPR strains, such as Azospirillum brasilense (Ruíz-Sánchez et al. 2011; Cohen et al. 2015), Bacillus subtilis GB03 (Zhang et al. 2008), Pseudomonas fluorescens (Rincón et al. 2008), and Bp (in maize) (Naveed et al. 2014), were reported to enhance PSII activities. A. brasilense (Cohen 2008, 2015) or B. subtilis GB03 (Zhang et al. 2008) also promote the accumulation of photosynthetic pigment contents. Zhang et al. (2008) found that PGPR strain B. subtilis GB03 increased chlorophyll content and photosynthetic efficiency of Arabidopsis by modulation of endogenous glucose concentration and abscisic acid signaling. Miotto-Vilanova et al. (2016) showed in grapevine no modification in Y(II), Y(NO), Y(NPQ) 3–6 days after root-bacterization with Bp. For the first time, we showed that there is also no variation in the regulation of energy at the PSI level.

The photosynthetic activity is due to the RuBisCO activity, enzyme able to fix carbon in the chloroplast. The protein is composed of two subunits: the nuclear-encoded RbcS gene and the chloroplast-encoded RbcL gene. Local or systemic presence of Bp in planta and of Pst in leaf did not modify the RuBisCO protein accumulation pattern. In Arabidopsis and tomato, many genes encoding chloroplast-localized proteins involved in photosynthesis and the Calvin cycle are repressed by Pst (Bonfig et al. 2006; Thilmony et al. 2006), but no data on protein accumulation were available.

Pathogen infection not only affects defense reactions but also leads to changes in carbohydrate metabolism (Ehness et al. 1997; Roitsch 1999). Several studies reported a decrease in photosynthesis after infection, which might be due to a downregulation of photosynthesis or damage of the photosynthetic apparatus (Chou et al. 2000; Berger et al. 2004). The damaged in PSII led a decreased of F v/F m, ΦPSII, ETRII, Y(NPQ), and an increase of Y(NO). The results on PSII were in accordance with previous studies showing a decrease in F v/F m, ΦPSII after infiltration of Pst in Arabidopsis or in bean (Bonfig et al. 2006; Pérez-Bueno et al. 2015). The partitioning of absorbed excitation energy in PSII takes place into three fundamental pathways, (i) photochemical utilization, (ii) regulated heat dissipation (a loss process serving for protection) and (iii) non-regulated heat dissipation (a loss process due to PSII inactivity). A higher Y(NPQ) value than Y(NO) value indicates that excess excitation energy is safely dissipated at the antenna level, meaning that photosynthetic energy fluxes are well regulated. In variance, high values of Y(NO) signify that excess excitation energy is reaching the reaction centers, resulting in strong reduction of PSII acceptors and photodamage, e.g., via formation of reactive oxygen species. The impact on NPQ seems to be dependent on the studied pathosystem: an increase in Arabidopsis inoculated with Pst (Berger et al. 2007) but a decrease in bean plants inoculated with Pst or P. syringae pv. phaseolicola (Rodríguez-Moreno et al. 2008) was observed. A decrease in NPQ was also detected following the infection of tobacco with tobacco mosaic virus (Balachandran et al. 1994), Abutilon striatum with Abutilon mosaic virus (Lohaus et al. 2000), and oat leaves with Puccinia (Scholes and Rolfe 1996). In contrast, an increase was observed in response to the biotrophic fungus Albugo candida in Arabidopsis (Chou et al. 2000) and to the necrotrophic fungus B. cinerea in tomato (Berger et al. 2004). Indeed, stress and particularly disease can promote excess excitation energy even at light intensities that would not pose a problem under control conditions (Mullineaux and Karpinski 2002).

Photosystem II is considered to be more vulnerable than PSI when plants encounter stresses (Barth et al. 2001). Several researches have been conducted on PSII regulation after challenge with Pst. However, only few information are available on PSI activity after pathogen infection. Infection with Pst caused a modification in PSI and PSI activity and regulation. In PSI, we recorded an increase in the limitation of the electron acceptor side Y(NA) and a decrease of ΦPSI and ETRI values. Cheng et al. (2016) measured also a decrease of PSI activity after P. syringae pv. tabaci infection and indicated a degradation of PsbO, D1, and PsaA proteins in tobacco.

Overall, the presence of Bp did not modify the impact of Pst on PSI and PSII regulation and on RuBisCO contents, except an increased Y(ND) at 3 dpi in Pst/SBp infected plants.

Conclusions

PGPR may prevent attack from pathogenic microorganisms (van Loon et al. 1998). The ability of the PGPR Burkholderia phytofirmans strain PsJN to induce a local or a systemic resistance on Arabidopsis against an infection with the hemibiotrophic bacterium Pseudomonas syringae pv. tomato DC3000 was evaluated. Since photosynthesis mechanisms are also involved in plant resistance to biotic stress, photosynthetic parameters and defense response were both quantified. No antibiosis between these two microorganisms was detected. Pre-inoculation of plant roots with Bp primed both PR1 and PDF1.2 expressions that led to a slower spread of Pst during the first stages after pathogen inoculation. However, such priming effect was not displayed when Bp was infiltrated in leaves. Infection with Pst decreased PSI and PSII activities. More precisely, the pathogen inoculation triggered a decrease in ΦPSI and ETRI and an increase in Y(NA). Additionally, photosynthetic parameters were not modified by a pre-inoculation with Bp. Barriuso et al. (2008) showed that infection with Pst of Arabidopsis plants pre-inoculated with different PGPR (Staphylococcus, Bacillus, Curtobacterium, or Arthrobacter oxidanspathogen) led to significantly higher values of F v/F m than the pathogen control. The bacterium Bp induced plant tolerance against abiotic and biotic stresses. In B. cinerea-infected leaves of root-inoculated grapevine with Bp, an increase of Y(II) was observed in bacterized plantlets, suggesting the protective role of bacteria on grapevine photosynthetic apparatus during the first step of infection (without modification of Y(NO), Y(NPQ), Fv/Fm, or ETR) (Miotto-Vilanova et al. 2016). In response to abiotic stress, the results are also reversed. Beneficial effects triggered by PGPR colonization against cold are associated to photosynthesis, carbohydrates, and related metabolites (Ait Barka et al. 2006; Fernandez et al. 2012). But Su et al. (2015) showed that Bp inoculation did not affect PSII activity and gas exchange during and after one cold night at 0 or −1 °C but increased net photosynthesis after one night at −3 °C.

In conclusion, the role of Bp here is not multifaceted like in grapevine tolerance towards B. cinerea, and relies only on priming of defense mechanisms but not on improving photosynthetic activity.