bs_bs_banner
Environmental Microbiology Reports (2015) 7(5), 795–802
doi:10.1111/1758-2229.12320
Visualizing the relevance of bacterial blue- and red-light receptors during plant–pathogen interaction Ada Ricci,1 Lucia Dramis,1 Rashmi Shah,2 Wolfgang Gärtner2 and Aba Losi3* Departments of 1Life Sciences and 3 Physics and Earth Sciences, University of Parma, 43124 Parma, Italy. 2 Max-Planck-Institute for Chemical Energy Conversion, 45470 Mülheim, Germany. Summary The foliar pathogen Pseudomonas syringae pv. tomato DC3000 (Pst) leads to consistent losses in tomato crops, urging to multiply investigations on the physiological bases for its infectiveness. As other P. syringae pathovars, Pst is equipped with photoreceptors for blue and red light, mimicking the photosensing ability of host plants. In this work we have investigated Pst strains lacking the genes for a blue-light sensing protein (PstLOV), for a bacteriophytochrome (PstBph1) or for hemeoxygenase-1. When grown in culturing medium, all deletion mutants presented a larger growth than wildtype (WT) Pst under all other light conditions, with the exception of blue light which, under our experimental conditions (photon fluence rate = 40 μmol m−2 s−1), completely suppressed the growth of the deletion mutants. Each of the knockout mutants shows stronger virulence towards Arabidopsis thaliana than PstWT, as evidenced by macroscopic damages in the host tissues of infected leaves. Mutated bacteria were also identified in districts distant from the infection site using scanning electron microscopy. These results underscore the importance of Pst photoreceptors in responding to environmental light inputs and the partial protective role that they exert towards host plants during infection, diminishing virulence and invasiveness. Introduction Recent research has suggested that visible light (400– 700 nm), particularly in the blue (BL) and red-/far red light
Received 2 February, 2015; revised 26 June, 2015; accepted 29 June, 2015. *For correspondence. E-mail [email protected]; Tel. +390521905293; Fax +390521905223.
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd
(RL/FRL) spectral regions, may influence growth patterns, virulence and infectivity of pathogenic bacteria via the activity of microbial photoreceptor proteins (Gomelsky and Hoff, 2011; Losi et al., 2014). In this work we focus on the blue- and red-light (BL and RL) photoreceptors of Pseudomonas syringae pv. tomato DC3000 (Pst), an agronomically important pathogen of tomato plants, and their role during infection. Pst is in fact often studied as a model system for plant–pathogen interactions, given its ability to infect Arabidopsis thaliana (Xin and He, 2013). As other P. syringae pathovars, Pst enter plant tissues via wounds or natural surface openings such as stomata, but they are weak epiphytes and die quickly if they fail to enter plant tissues. P. syringae bacteria are hemibiotrophs, i.e. microorganisms that parasite living tissues for a certain time period and kill host tissues in late stages of infection. Besides the modulation of stomatal closure (Lim et al., 2014), infected plants respond with multiple strategies that prevent aggressive multiplication of P. syringae strains, such as callose deposition and enhanced formation of reactive oxygen species (ROS) (Xin and He, 2013; Lim et al., 2014). Recent research suggested that also visible light, particularly the BL, RL and FRL spectral regions, may influence growth patterns, virulence and infectivity of P. syringae (Moriconi et al., 2013; Río-Álvarez et al., 2013; Wu et al., 2013). From a plant perspective, it is becoming evident that visible light has direct effects on immune responses (Karpinski et al., 2003; Roden and Ingle, 2009), but could also act as a determinant for pathogens virulence (Elías-Arnanz et al., 2011; Losi et al., 2014). Indeed, photoreceptors present in plant-pathogenic and plant-symbiotic bacteria strikingly resemble the light-sensing ability of the host. In particular bacteriophytochromes (BphP) and LOV (light, oxygen and voltage) proteins are related to plant phytochromes (phy) and phototropins (phot) respectively (Mandalari et al., 2013). Similar to plant phy, in BphP photoperception relies on a photoisomerizable open-chain tetrapyrrole chromophore, bound within a photosensory module fused to a kinase function (Müller et al., 2008; Dasgupta et al., 2009; Shah et al., 2012). They are photochromic proteins, typically switchable between a red (Pr) and a far red (Pfr) absorbing form (Auldridge and Forest, 2011). In plant phy, the chromophore is phytochromobilin, whereas BphP bind biliverdin IXα (Rockwell and Lagarias, 2010).
796
A. Ricci et al.
The LOV domains are small α/β folds (c. 110 aa) that respond to light through a bound flavin chromophore and are largely spread among plants, bacteria, archaea and fungi, being fused to a variety of effector domains (Herrou and Crosson, 2011; Zoltowski and Gardner, 2011; Mandalari et al., 2013). In the dark-adapted state, the flavin is non-covalently bound and absorbs maximally around 450 nm, exhibiting a typical bright green fluorescence. The photocycle involves the reversible formation of a covalent adduct between the isoalloxazine ring of flavin and a nearby cysteine, with complete loss of fluorescence (Losi, 2007). In bacteria and archaea, LOV proteins are extremely variegate: most of them are His-kinases (HK) of the two-component systems or regulators for the turnover of cyclic diguanylate, but also DNA binding proteins, phosphatases and other effector/regulative domains are represented (Losi et al., 2014). All P. syringae pathovars sequenced so far bear genes coding for a hybrid LOV-HK with a fused response regulator (RR), and for two BphP proteins (BphP1 and BphP2) (Mandalari et al., 2013); however, BphP2 carries a compromised HK and second HK domain following in C-terminal direction (Shah et al., 2012). The gene encoding BphP2 is part of an operon that also contains a gene for a canonical RR of CheY type, whereas bphP1 is associated with a heme oxygenase-1 (here HO) encoding gene. The occurrence of similar photoreceptors in plants and in plant-pathogenic bacteria prompts the question whether this shared light-sensing capability is important for plant–bacteria interactions, chiefly for colonization, infectivity or virulence. Available data for BL flavinphotoreceptors pointing in this direction have recently been compiled (Losi et al., 2014). Considering selectively examples for P. syringae and related species, the results obtained so far can be summarized as follows: (i) quorum sensing in P. aeruginosa is affected by a BphP protein; however, the role of the light quality was not inspected (Barkovits et al., 2011). (ii) P. syringae pv. phaseolicola causes diseases in bean plants only after the loss of a genomic island (GI) that contains the gene for a BphP. Again, the role of light had not been investigated (Lovell et al., 2011). (iii) In P. syringae pv. syringae (Pss), BphP1 negatively regulates swarming motility in response to RL/FRL, whereas a LOV-HK suppresses negative regulation by BphP1 under BL (Wu et al., 2013). (iv) In Xanthomonas citri subsp. citri (Xcc, formerly X. axonopodis pv. citri), responsible for citrus canker, LOV-HK protein (XccLOV) not only modulates physiological attributes of the bacterium that are relevant for plant colonization and promotes adhesion to leaf surface, but also influences the development of disease symptoms on citrus in a light-dependent manner: in the light, XccLOV exerts a protective effect against necrosis, but not against canker (Kraiselburd et al., 2012). These effects seem to
be raised via the participation of XccLOV in several plant responses induced by infection, e.g. alteration of photosynthesis and defence response (Kraiselburd et al., 2013). BL absorption from XccLOV might be involved in a bacterial mechanism aiming to partially reduce the plant defence response and maintain tissue integrity, which is vital for hemibiotrophic pathogens. (v) Finally, in Pst, a LOV-HK protein (PstLOV) inhibits bacterial growth and expression of genes controlled by sigma factors (e.g. regulation of general stress response), and mediates BL-driven reduction of virulence in infected plants. In parallel, PstLOV enhances bacterial adhesion to leaves (Moriconi et al., 2013; Río-Álvarez et al., 2013). The reduction of virulence may be related to a reduced invasiveness and proliferation within leaves mediated by this photoreceptor (Moriconi et al., 2013). Different from Pss, BL has also an inhibiting effect on Pst virulence towards plant leaves, while the opposite has been reported for RL (Río-Álvarez et al., 2013), suggesting a still unidentified role for BphP proteins. Even less experimental results are available for the role of RL in plant–pathogen interactions. We have thus further investigated the in vivo relevance of BL sensing and have extended our work to the effect of RL sensing in Pst. The role of photoreceptors was studied with respect to bacterial growth under selected light conditions, bacterial proliferation/invasiveness within plant tissues and plant hypersensitive response (HR). For these investigations, we employed Arabidopsis plants as model hosts, Pst wild-type (PstWT) and deletion mutants in PstLOV, PstBphP1 and PstHO (HO catalysing formation of bilins). Our results indicate that the presence of both BL and RL photoreceptors renders Pst less virulent and less invasive towards host plants, and that these photoreceptors regulate bacterial growth under different light conditions.
Results and discussion Bacterial growth under different light conditions and influence of photoreceptors Knockout protocol for the PstLOV encoding gene (PSPTO2896) is described in Moriconi and colleagues (2013): genome positions 3258886 until 3260929 (PstLOV coding region) were cloned into the pJET1.2 vector (Life Technologies, Darmstadt, Germany). The BstZ17I restriction site located within the gene was used to insert a kanamycin cassette, isolated from the pBSL15 vector (Creative Biogene, Shirley, NY, USA) by SmaI digestion. This method yielded the inserted DNA in both orientations. Generation of the other knockout strains was carried out according to the following protocols: for the BphP1 encoding gene (PSPTO_1902) chromosomal positions from
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Bacterial photoreceptors in plant–pathogen interaction 2077242 to 2080311 was amplified and cloned; here a spectinomycin resistant omega (Ω-Spr) cassette isolated from pHP45-Ω by SmaI digestion was bluntend ligated into the EcoRV site of the gene. Heme oxygenase-1: chromosomal positions 2077011 to 2077625 was cloned into the pET14 vector (Novagene/Merck, Germany); a kanamycin cassette was inserted into a StuI site located within the gene. All gene constructs carried 5′and 3′-flanking regions of at least 250 bp in length. Sequencing of isolated genomic DNA with appropriate primers clearly showed the different size for the WTand the antibiotics cassette enlarged gene. Disrupted genes were then cloned into a suicide pSUP plasmid (Simon et al., 1983) (kindly donated by Thomas Drepper, Forschungszentrum Jülich, Germany), carrying a spectinomycin resistance cassette and the vector was linearized. Linear DNA (50–100 ng) was added to 100 μl of Pst cells (washed three times with ice-cold distilled water and kept in a 10% glycerol aqueous solution). Transformation (0.1 cm cuvette) was done with a Bio-Rad gene pulser [25 μF, 2.4 kV and 200 Ohm (5 ms)], immediately followed by addition of ice-cold KB (King’s B) medium. Further details will be described elsewhere (Shah, R., Pathak, G.P., Drepper, T., and Gärtner, W., in preparation). Cells that grew on the selective media after electroporation were re-streaked on plates containing higher amount of antibiotics (50 μg ml−1 of kanamycin or 100 μg ml−1 of spectinomycin) for several rounds. Finally, individual colonies from the antibiotic plate were withdrawn, and mutations of the chromosomal genes were verified by polymerase chain reaction (PCR), using gene-specific primers. Bacterial cultures were grown overnight in Luria– Bertani (LB) medium (Sambrook et al., 1989) at 27 ± 1°C with antibiotics added as outlined above. The PstWT and the various knockout mutant strains (ΔLOV-A, ΔLOV-B, ΔHO-A, ΔHO-B and ΔBphP1) were kept under different continuous light qualities or in darkness in order to determine the roles of different photoreceptors for their growth behaviour (Fig. 1). Cells were grown under these conditions for 48 h on solid medium, after which time the amount of colonies was counted as an estimate for the number of viable cells (cfu = colony-forming units) (Fig. 1). The growth of the six strains (WT and deletion mutants) was statistically different in all light conditions tested (oneway analysis of variance), with the exception of BL. This latter light quality completely repressed the growth of the mutants, under the conditions used. With reference to darkness conditions, RL caused an increase in growth capacity for ΔLOV-A, ΔLOV-B, and to a lesser extent for ΔHO-B (significantly different, Student’s t-test). None of the other applied light qualities yielded a significantly different growth behaviour in comparison to darkness. We can thus infer that both PstLOV and PstBphP1
797
Fig. 1. Effects of different light conditions on the growth of investigated bacterial strains. Error bars indicate standard deviation. cfu, colony-forming units. RL (λmax = 660 nm; 6 μmol m−2 s−1), FRL (λmax = 770 nm; 17 μmol m−2 s−1) or BL (λmax = 462 nm; 40 μmol m−2 s−1) were provided by light-emitting diodes (LEDs) from Roithner Lasertechnik, Vienna, Austria (diameter 5 mm each). The different photon fluence rates were due to the necessity to position the plates at a given distance from the light sources, as described in Wu and colleagues (2013). To determine the cfu, an exponential-phase culture (OD600 = 0.6) of each bacterial strain was used to prepare serial 1:10 dilutions necessary to get countable colonies. Of each dilution, 100 μl was spread on the surface of 2% (wt vol−1) LB agar plates. The plates were incubated at 28°C in separate chambers under the different light conditions. Plates kept in darkness were used as control. Three plates per treatment were handled in parallel and experiments were repeated twice. The colony number was determined from each plate after 48 h, and the significance of differences among multiple groups was evaluated using the analysis of variance (one-way ANOVA) test, while comparisons between two groups were done by Student’s t-test. Results were considered to be statistically significant at p ≤ 0.01.
downregulate bacterial growth under WL, RL, FRL and in darkness. Deletion of the HO, required for bilin production, has the same effect. In fact, each mutant strain shows a greater growth rate than the PstWT when exposed to these light conditions (continuous illumination). On the contrary, under BL only PstWT strain retains a growth capacity similar to that shown in darkness, while the mutant strains did not grow at all under our experimental conditions. Apparently, the simultaneous presence of PstLOV, PstHO and PstBphP1 photoreceptor proteins is an indispensable requirement to guarantee growth under BL in the described culturing conditions. One might assume that photoreceptors trigger a still unknown protective response towards BL that is potentially phototoxic. The results also suggest an interplay between LOV and BphP proteins and an integrated regulatory network of these photoreceptor proteins in Pst, as recently demonstrated for the regulation of swarming motility in P. syringae B728a (Wu et al., 2013). We note here that the growth of PstWT is significantly reduced under FRL, possibly accounting for the slightly reduced growth under WL. These findings suggest a sort of balancing among the effects elicited by the different wavelengths, as PstWT growth under WL is only slightly reduced.
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
798 A. Ricci et al. Fig. 2. Representative plates displaying bacterial growth resulting from homogenate smear of Arabidopsis leaves infiltrated (I) and non-infiltrated (NI) with PstWT (A, B) or ΔLOV-A (C, D) bacterial strains. For plant inoculation, the different bacterial strains were suspended at a density of approximately 1 × 108 or 2.5 × 108 cfu ml−1 (OD600 = 0.2 or 0.5 respectively). Pictures were taken after incubation at 25 ± 1°C for 48 h in darkness. For the remaining strains, see Fig. S1.
Our results on light-regulated growth appear to be in contrast to previously published data by Moriconi and colleagues reporting that only WL was able to promote growth of a ΔLOV Pst strain (Moriconi et al., 2013), and more importantly these authors did not observe any difference between PstWT and ΔLOV-A under BL. Due to the different experimental conditions, the two sets of data are difficult to compare, because in the work by Moriconi and colleagues the cell density in liquid medium was measured (OD600), whereas here the number of viable cells was used as parameter. Yet these authors note that PstWT showed a longer initial growth lag than ΔLOV mutant strains under WL (Moriconi et al., 2013). Further work, also taking into account e.g. the diverse media composition or the illumination regime, is required to fully clarify this aspect and the discrepancy among different datasets. Infectivity and bacterial growth within infected leaves Arabidopsis thaliana plants, ecotype Col-0, were grown from seed on a sterile mixture of soil and vermiculite (1:1) in a growth chamber at WL intensity of approximately 30 μmol m−2 s−1 at 21 ± 1°C under 16/8 h day/night photoperiod. For plant inoculation, overnight-cultured bacteria were suspended in 10 mM MgCl2 at a density of approximately 1 × 108 or 2.5 × 108 cfu ml−1 (OD600 = 0.2 or 0.5, respectively), and leaves of 5- to 6-week-old plants were infiltrated into their abaxial surface using a 1 ml syringe without needle, following a previously described protocol with minor modifications (Katagiri et al., 2002; Griebel and
Zeier, 2008). Infiltration with 10 mM MgCl2 was used as control. Infiltrated plants were placed in plastic boxes filled with wet, expanded clay covered with a transparent polythene cover to ensure high humidity. The boxes were kept at WL intensity of approximately 30 μmol m−2 s−1 at 19 ± 1°C under a 16/8 h day/night photoperiod. Bacterial growth in planta was determined 72 h after inoculation. Infiltrated and non-infiltrated opposite leaves were sampled, surface-sterilized with 70% ethanol for 1 min, and then washed twice with sterile, distilled water to remove bacteria from the surface. Leaves were ground in 1 ml LB medium, the homogenate was shaken and plated on solid LB medium (2% agar wt vol−1) supplemented with the appropriate antibiotics. Following incubation at 25 ± 1°C for 48 h in darkness, bacterial growth, and accordingly distribution inside the plants, was assessed. For each experiment, repeated three times, six to ten plants were infiltrated with the different bacterial strains. Bacteria were detected from the homogenate of all infiltrated and non-infiltrated leaves, independently of the bacterial density of the inoculum. In both cases, knockout mutants showed a larger number of colonies (Fig. 2 and Fig. S1). These results agree with the growth assays of bacteria illustrated above. A more straightforward investigation of the infectivity of PstWT and mutated strains was obtained by means of scanning electron microscopy (SEM) on cross-sections of infiltrated and non-infiltrated leaves, according to the method of Misas-Villamil and colleagues (2011) with minor modifications. Only A. thaliana plants infiltrated with bacteria at a density of approximately 2.5 × 108 cfu ml−1
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Bacterial photoreceptors in plant–pathogen interaction
799
Fig. 3. SEM representative pictures of infiltrated leaves taken 24 h after infiltration with bacterial strains [(A) PstWT, (B) ΔLOV-A, (C) ΔLOV-B, (D) ΔHO-A, (E) ΔHO-B and (F) Δbphp1, respectively] at a density of approximately 2.5 × 108 cfu ml−1 (OD600 = 0.5). No clearly recognizable bacteria were visible in the cross-sections of leaves infiltrated with the wild-type bacterial strain, but only tissue alteration. For the control condition (infiltration with 10 mm MgCl2) and the other times of investigation, see Fig. S2.
(OD600 = 0.5) or with 10 mM MgCl2 solution as control were used in this assay. Infiltrated and non-infiltrated opposite leaf samples were scored at 24, 48 and 72 h after infiltration, fixed in 3% glutaraldehyde in phosphate buffer 0.1 M, pH 7.0, dehydrated using acetone and dried in a critical point drier (Balzers Union CPD 020) under a CO2 atmosphere. Specimens were gold-coated using a Balzers Union Sputter Coater and examined using SEM (Philips SEM 501, provided by Dipartimento di Scienze Biomediche, Biotecnologiche e Traslazionali, University of Parma). This approach is able to reveal the presence of bacteria in planta, and possibly their distant location from the infiltration spot. For the mutated strains, a huge amount of bacteria was clearly visible in the apoplast regions near the xylem vessels, and also trapped in granular and/or fibrillar structures of the infiltrated leaves (Fig. 3 and Fig. S2). On the other hand, only the presence of granular and fibrillar structures altering the internal leaf architecture without clearly recognizable bacteria was achieved by the SEM analysis of the cross-sections of leaves infiltrated with PstWT (Fig. 3 and Fig. S2). Neither bacteria nor fibrillar structures were ever detected under control conditions (leaves infiltrated with 10 mM MgCl2 solution; Fig. S2). An example of complete collapse of host cells could be documented 48 h after infiltration: upper and lower epidermis surround an empty space, while ‘buds’ of unknown matrix embedding bacteria are extruded from a lot of stomata of the upper epidermis (Fig. S3). In order to investigate the distant location of bacteria, non-infected leaves were sampled and analysed as well.
Only the mutant strains rapidly moved inside the plants, allowing the detection of colonization of non-infiltrated leaves (Fig. S4). In contrast, no distant location of WT bacteria was achieved at 24, 48 or 72 h after infiltration, or under control conditions. Plant HR assay Arabidopsis thaliana plants infiltrated with the different bacterial strains at a density of approximately 2.5 × 108 cfu ml−1 (OD600 = 0.5) or with 10 mM MgCl2 solution, as control, were used in this assay. Infiltrated leaves were scored for tissue collapse (denoting HR) at 24, 48 and 72 h after infiltration (Fig. 4). The evolution of macroscopic symptoms was registered with a Canon Ixus 200 IS digital camera as previously described (Kabisch et al., 2005). For each experiment, repeated three times, six to ten plants were infiltrated with the different bacterial strains. The first symptoms of HR in plants infiltrated with the mutated strains appeared 48 h after inoculation in the form of diffuse chlorosis (Fig. 4H, K, N, Q) and random necrotic spots (Fig. 4E). Tissue collapse became more severe at 72 h after inoculation, with yellowing occurring on the entire leaf (Fig. 4F, L) and necrotic zones being either more frequent or wider in size (Fig. 4I, O, R). Leaves of plants inoculated with PstWT did not develop comparable symptoms during the period of investigation (Fig. 4A–C). In these plants, yellowing appeared 8–10 days after inoculation (data not shown). Control plants that were injected with a 10 mM
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
800 A. Ricci et al. MgCl2 solution without bacteria did not develop symptoms (Fig. 4S, T, U). From our data we can infer that all bacterial photoreceptors analysed in this study downregulate infectivity versus Arabidopsis plants and growth of bacteria within infected leaves. As for LOV proteins, this is consistent with previous work either on Pst or on X. citri (Kraiselburd et al., 2012; 2013; Moriconi et al., 2013). The present data, obtained with strains unable to synthesize functional BphP proteins, suggest a similar role also for bacteriophytochromes during Pst–plant interaction. Conclusions
Fig. 4. Representative phenotypes of leaves displaying a hypersensitive reaction (HR) after inoculation with Pst DC3000 (A–C), ΔLOV-A (D–F), ΔLOV-B (G–I), ΔHO-A (J–L), ΔHO-B (M–O), Δbphp1 (P–R) and 10 mM MgCl2 as control (S–U). Leaves were scored at 24, 48 and 72 h after infiltration. *: evident reaction; **: more severe collapse.
The data presented here underscore the striking observation that bacterial photoreceptors can downregulate infectivity towards host plants, with which they share the spectral range sensed (BL and RL/FRL). Still, this phenomenon is only sparsely investigated, as is evident from the low number of reports published so far. This feature, first brought forth for LOV proteins, apparently holds also for bacteriophytochromes, even if the available data are still limited to very few systems. It is clear that during plant–pathogen interactions different interests and survival strategies meet: pathogens need to grow inside plant tissues from which they obtain nutritional compounds; plants must rapidly react by activating efficient defence responses. This is clearly observable when A. thaliana plants are infected with PstWT. Bacterial pathogenicity depends on effectors, either proteins delivered by the type III secretion system (T3SS) directly into host cells or phytotoxins as coronatine. Conversely, plants display immune signalling pathways resulting in callose deposition, antibacterial phytoalexins, ROS, HR and accumulation of salicylic acid (Kabisch et al., 2005; Návarová et al., 2012; Xin and He, 2013). Many T3SS effectors of plant-pathogenic bacteria suppress plant innate immunity by different mechanisms altering plant behaviour and development, up to death. Could bacterial photoreceptors play a role in plant–pathogen interactions being rapidly recognized by plant cells as non-host proteins and eliciting the different plant immune signalling pathways? In order to confirm or discard this hypothesis, deeper investigations about plant defence responses are essential. With respect to this, in a recently published paper the growth, swarming and biofilm formation capacity of Pseudomonas cichorii JBC1 was tested under different light conditions, and no significant difference to dark-grown cells was detected for red, green and WL illumination (Nagendran and Lee, 2015). Nevertheless, tomato seedlings infected with P. cichorii JBC1 showed significant reduction in disease incidence and internal bacterial growth under green and red light, concomitant with significant upregulation of
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Bacterial photoreceptors in plant–pathogen interaction plant defence-related gene. The role of bacterial photoreceptors was nevertheless not tested in that work (P. cichorii JBC1 has an LOV-HK and two BphPs, similar to Pst). The data presented here and elsewhere (Kraiselburd et al., 2012; 2013; Moriconi et al., 2013; Río-Álvarez et al., 2013; Wu et al., 2013) suggest that bacterial photoreceptors could indeed have a stimulating function as protective factors during plant infection, although the integration of bacterial light-sensing with plant photosensory transduction pathways is an open research field. The knowledge that genes for LOV and BphP proteins in P. syringae pathovars have been localized on GI that bears other genes relevant for virulence suggests that they might have been acquired via horizontal gene transfer (Lovell et al., 2011; Moriconi et al., 2013). Although the cross-talk with virulence genes and the actual significance of photoreceptors in such GI is still to be clarified, we consider this a key point for understanding the role of LOV and BphP proteins during infection, at least for this species. Complementation and interpretation of experiments with double mutants, already performed in some cases (Moriconi et al., 2013; Wu et al., 2013), also have to take into account the localization of these genes on GI. The importance of the topic dealt in this manuscript, i.e. the impact of bacterial photoreceptors on plant–pathogen interactions, is likely to grow in the near future, given the increasing importance of bacterial photosensors for bacterial lifestyles and metabolism (Mandalari et al., 2013; Losi et al., 2014). Future work will explore the role of Pst photoreceptors during infection of tomato plants, the main agronomically relevant host of this species (Xin and He, 2013). Acknowledgements The authors are deeply grateful to Davide Dallatana (Dipartimento di Scienze Biomediche, Biotecnologiche e Traslazionali, University of Parma) for his excellent technical assistance in the SEM analyses. The authors also acknowledge Dr. Alessio Sardella for his partial contribution to the experimental procedures. PstWT was kindly provided by Robin Buell (Institute for Genomic Research, Rockville, MD, USA). The authors declare no conflict of interest for this manuscript.
References Auldridge, M.E., and Forest, K.T. (2011) Bacterial phytochromes: more than meets the light. Crit Rev Biochem Mol 46: 67–88. Barkovits, K., Schubert, B., Heine, S., Scheer, M., and Frankenberg-Dinkel, N. (2011) Function of the bacteriophytochrome BphP in the RpoS/Las quorumsensing network of Pseudomonas aeruginosa. Microbiology 157: 1651–1664.
801
Dasgupta, J., Frontiera, R.R., Taylor, K.C., Lagarias, J.C., and Mathies, R.A. (2009) Ultrafast excited-state isomerization in phytochrome revealed by femtosecond stimulated Raman spectroscopy. Proc Natl Acad Sci USA 106: 1784–1789. Elías-Arnanz, M., Padmanabhan, S., and Murillo, F.J. (2011) Light-dependent gene regulation in nonphototrophic bacteria. Curr Opin Microbiol 14: 128–135. Gomelsky, M., and Hoff, W.D. (2011) Light helps bacteria make important lifestyle decisions. Trends Microbiol 19: 441–448. Griebel, T., and Zeier, J. (2008) Light regulation and daytime dependency of inducible plant defenses in Arabidopsis: phytochrome signaling controls systemic acquired resistance rather than local defense. Plant Physiol 147: 790– 801. Herrou, J., and Crosson, S. (2011) Function, structure and mechanism of bacterial photosensory LOV proteins. Nat Rev Microbiol 9: 713–723. Kabisch, U., Landgraf, A., Krause, J., Bonas, U., and Boch, J. (2005) Type III secretion chaperones ShcS1 and ShcO1 from Pseudomonas syringae pv. tomato DC3000 bind more than one effector. Microbiology 151: 269– 280. Karpinski, S., Gabrys, H., Mateo, A., Karpinska, B., and Mullineaux, P.M. (2003) Light perception in plant disease defence signalling. Curr Opin Plant Biol 6: 390–396. Katagiri, F., Thilmony, R., and He, S.Y. (2002) The Arabidopsis thaliana-Pseudomonas syringae interaction. Arabidopsis Book 1: e0039. doi:10.1199/tab.0039. Kraiselburd, I., Alet, A.I., Tondo, M.L., Petrocelli, S., Daurelio, L.D., Monzón, J., et al. (2012) A LOV protein modulates the physiological attributes of Xanthomonas axonopodis pv. citri relevant for host plant colonization. PLoS ONE 7: e38226. Kraiselburd, I., Daurelio, L.D., Tondo, M.L., Merelo, P., Cortadi, A., Talón, M., et al. (2013) The LOV protein of Xanthomonas citri subsp. citri plays a significant role in the counteraction of plant immune responses during citrus canker. PLoS ONE 8: e80930. Lim, C.W., Luan, S., and Lee, S.C. (2014) A prominent role for RCAR3-mediated ABA signaling in response to Pseudomonas syringae pv. tomato DC3000 infection in Arabidopsis. Plant Cell Physiol 55: 1691–1703. Losi, A. (2007) Flavin-based blue-light photosensors: a photobiophysics update. Photochem Photobiol 83: 1283– 1300. Losi, A., Mandalari, C., and Gärtner, W. (2014) From plant infectivity to growth patterns: the role of blue-light sensing in the prokaryotic world. Plants 3: 70–94. Lovell, H.C., Jackson, R.W., Mansfield, J.W., Godfrey, S.A.C., Hancock, J.T., Desikan, R.A.D.H., and Arnold, D.L. (2011) In planta conditions induce genomic changes in Pseudomonas syringae pv. phaseolicola. Mol Plant Pathol 12: 167–176. Mandalari, C., Losi, A., and Gärtner, W. (2013) Distancetree analysis, distribution and co-presence of bilin- and flavin-binding prokaryotic photoreceptors for visible light. Photochem Photobiol Sci 12: 1144–1157. Misas-Villamil, J., Kolodziejek, I., and van der Hoorn, R. (2011) Pseudomonas syringae colonizes distant tissues in
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
802 A. Ricci et al. Nicotiana benthamiana through xylem vessels. Plant J 67: 774–782. Moriconi, V., Sellaro, R., Ayub, N., Soto, G., Rugnone, M.A., Shah, R., et al. (2013) LOV-domain photoreceptor, encoded in a genomic island, attenuates the virulence of Pseudomonas syringae in light-exposed Arabidopsis leaves. Plant J 76: 322–331. Müller, M., Lindner, I., Martin, I., Gärtner, W., and Holzwarth, A.R. (2008) Femtosecond kinetics of photoconversion of the higher plant photoreceptor phytochrome carrying native and modified chromophores. Biophys J 94: 4370– 4382. Nagendran, R., and Lee, Y.H. (2015) Green and red light reduces the disease severity by Pseudomonas cichorii JBC1 in tomato plants via upregulation of defense-related gene expression. Phytopathology 150: 412–418. Návarová, H., Bernsdorff, F., Döring, A.C., and Zeier, J. (2012) Pipecolic acid, an endogenous mediator of defense amplification and priming, is a critical regulator of inducible plant immunity. Plant Cell 24: 5123–5141. Río-Álvarez, I., Rodríguez-Herva, J.J., Martínez, P.M., González-Melendi, P., García-Casado, G., Rodríguez-Palenzuela, P., and López-Solanilla, E. (2013) Light regulates motility, attachment and virulence in the plant pathogen Pseudomonas syringae pv. tomato DC3000. Environ Microbiol 16: 2072–2085. Rockwell, N.C., and Lagarias, J.C. (2010) A brief history of phytochromes. Chem Phys Chem 11: 1172–1180. Roden, L.C., and Ingle, R.A. (2009) Lights, rhythms, infection: the role of light and the circadian clock in determining the outcome of plant-pathogen interactions. Plant Cell 21: 2546–2552. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor, NY, USA: Cold Spring Harbor Laboratory Press. Shah, R., Schwach, J., Frankenberg-Dinkel, N., and Gärtner, W. (2012) Complex formation between heme oxygenase and phytochrome during biosynthesis in Pseudomonas syringae pv. tomato. Photochem Photobiol Sci 11: 1026– 1031. Simon, R., Priefer, U., and Pühler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Biotechnology 1: 784–791. Wu, L., McGrane, R.S., and Beattie, G.A. (2013) Light regulation of swarming motility in Pseudomonas syringae integrates signaling pathways mediated by a bacteriophytochrome and a LOV protein. mBio 4: e00334–13.
Xin, X.F., and He, S.Y. (2013) Pseudomonas syringae pv. tomato DC3000: a model pathogen for probing disease susceptibility and hormone signaling in plants. Annu Rev Phytopathol 51: 473–498. Zoltowski, B.D., and Gardner, K.H. (2011) Tripping the light fantastic: blue-light photoreceptors as examples of environmentally modulated protein-protein interactions. Biochemistry 50: 4–16.
Supporting information Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Fig. S1. Representative plates displaying bacterial growth resulting from homogenate smear of Arabidopsis leaves infiltrated (I) and non-infiltrated (NI) with all the bacterial strains. For plant inoculation, the different bacterial strains were suspended at a density of approximately 1 × 108 or 2.5 × 108 cfu ml−1 (OD600 = 0.2 or 0.5 respectively). Infiltration with 10 mM MgCl2 solution was used as control. Pictures were taken after incubation at 25 ± 1°C for 48 h in darkness. Fig. S2. SEM representative pictures of infiltrated leaves taken 48 and 72 h after infiltration with bacterial strains [(A) PstWT, (B) ΔLOV-A, (C) ΔLOV-B, (D) ΔHO-A, (E) ΔHO-B and (F) ΔBphP1, respectively], at a density of approximately 2.5 × 108 cfu ml−1 (OD600 = 0.5). No clearly recognizable bacteria were visible in the cross-sections of leaves infiltrated with the wild-type bacterial strain, only tissue alteration could be detected. SEM representative pictures of leaves infiltrated with 10 mM MgCl2 control solution taken 24, 48 and 72 h after infiltration are also shown. Fig. S3. Example of host cell collapse 48 h after the infiltration of cell suspension of the mutant strain ΔBphP1 had been performed. (A) Upper and lower epidermis surrounding an empty space; arrows indicate ‘buds’ on upper edge of the empty space; (B) upper epidermis full of ‘buds’; (C) ‘buds’ emerging from the stomata; and (D) bacteria embedded in an undefined matrix extruded from the stomata. Fig. S4. (A) Example of colonization of non-infiltrated leaf at 24 h after infiltration with the ΔHO-B mutant strain; (B) magnification of (A) showing bacteria embedded in a compact substance (arrow). (C) Example of colonization of a noninfiltrated leaf at 72 h after infiltration with ΔHO-B mutant strain; (D) magnification of (C) showing bacteria in the apoplast region. (E) Example of colonization of non-infiltrated leaf 72 h after infiltration with ΔLOV-B mutant strain; (F) magnification of (E).
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Environmental Microbiology Reports (2015) 7(5), 795–802
doi:10.1111/1758-2229.12320
Visualizing the relevance of bacterial blue- and red-light receptors during plant–pathogen interaction Ada Ricci,1 Lucia Dramis,1 Rashmi Shah,2 Wolfgang Gärtner2 and Aba Losi3* Departments of 1Life Sciences and 3 Physics and Earth Sciences, University of Parma, 43124 Parma, Italy. 2 Max-Planck-Institute for Chemical Energy Conversion, 45470 Mülheim, Germany. Summary The foliar pathogen Pseudomonas syringae pv. tomato DC3000 (Pst) leads to consistent losses in tomato crops, urging to multiply investigations on the physiological bases for its infectiveness. As other P. syringae pathovars, Pst is equipped with photoreceptors for blue and red light, mimicking the photosensing ability of host plants. In this work we have investigated Pst strains lacking the genes for a blue-light sensing protein (PstLOV), for a bacteriophytochrome (PstBph1) or for hemeoxygenase-1. When grown in culturing medium, all deletion mutants presented a larger growth than wildtype (WT) Pst under all other light conditions, with the exception of blue light which, under our experimental conditions (photon fluence rate = 40 μmol m−2 s−1), completely suppressed the growth of the deletion mutants. Each of the knockout mutants shows stronger virulence towards Arabidopsis thaliana than PstWT, as evidenced by macroscopic damages in the host tissues of infected leaves. Mutated bacteria were also identified in districts distant from the infection site using scanning electron microscopy. These results underscore the importance of Pst photoreceptors in responding to environmental light inputs and the partial protective role that they exert towards host plants during infection, diminishing virulence and invasiveness. Introduction Recent research has suggested that visible light (400– 700 nm), particularly in the blue (BL) and red-/far red light
Received 2 February, 2015; revised 26 June, 2015; accepted 29 June, 2015. *For correspondence. E-mail [email protected]; Tel. +390521905293; Fax +390521905223.
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd
(RL/FRL) spectral regions, may influence growth patterns, virulence and infectivity of pathogenic bacteria via the activity of microbial photoreceptor proteins (Gomelsky and Hoff, 2011; Losi et al., 2014). In this work we focus on the blue- and red-light (BL and RL) photoreceptors of Pseudomonas syringae pv. tomato DC3000 (Pst), an agronomically important pathogen of tomato plants, and their role during infection. Pst is in fact often studied as a model system for plant–pathogen interactions, given its ability to infect Arabidopsis thaliana (Xin and He, 2013). As other P. syringae pathovars, Pst enter plant tissues via wounds or natural surface openings such as stomata, but they are weak epiphytes and die quickly if they fail to enter plant tissues. P. syringae bacteria are hemibiotrophs, i.e. microorganisms that parasite living tissues for a certain time period and kill host tissues in late stages of infection. Besides the modulation of stomatal closure (Lim et al., 2014), infected plants respond with multiple strategies that prevent aggressive multiplication of P. syringae strains, such as callose deposition and enhanced formation of reactive oxygen species (ROS) (Xin and He, 2013; Lim et al., 2014). Recent research suggested that also visible light, particularly the BL, RL and FRL spectral regions, may influence growth patterns, virulence and infectivity of P. syringae (Moriconi et al., 2013; Río-Álvarez et al., 2013; Wu et al., 2013). From a plant perspective, it is becoming evident that visible light has direct effects on immune responses (Karpinski et al., 2003; Roden and Ingle, 2009), but could also act as a determinant for pathogens virulence (Elías-Arnanz et al., 2011; Losi et al., 2014). Indeed, photoreceptors present in plant-pathogenic and plant-symbiotic bacteria strikingly resemble the light-sensing ability of the host. In particular bacteriophytochromes (BphP) and LOV (light, oxygen and voltage) proteins are related to plant phytochromes (phy) and phototropins (phot) respectively (Mandalari et al., 2013). Similar to plant phy, in BphP photoperception relies on a photoisomerizable open-chain tetrapyrrole chromophore, bound within a photosensory module fused to a kinase function (Müller et al., 2008; Dasgupta et al., 2009; Shah et al., 2012). They are photochromic proteins, typically switchable between a red (Pr) and a far red (Pfr) absorbing form (Auldridge and Forest, 2011). In plant phy, the chromophore is phytochromobilin, whereas BphP bind biliverdin IXα (Rockwell and Lagarias, 2010).
796
A. Ricci et al.
The LOV domains are small α/β folds (c. 110 aa) that respond to light through a bound flavin chromophore and are largely spread among plants, bacteria, archaea and fungi, being fused to a variety of effector domains (Herrou and Crosson, 2011; Zoltowski and Gardner, 2011; Mandalari et al., 2013). In the dark-adapted state, the flavin is non-covalently bound and absorbs maximally around 450 nm, exhibiting a typical bright green fluorescence. The photocycle involves the reversible formation of a covalent adduct between the isoalloxazine ring of flavin and a nearby cysteine, with complete loss of fluorescence (Losi, 2007). In bacteria and archaea, LOV proteins are extremely variegate: most of them are His-kinases (HK) of the two-component systems or regulators for the turnover of cyclic diguanylate, but also DNA binding proteins, phosphatases and other effector/regulative domains are represented (Losi et al., 2014). All P. syringae pathovars sequenced so far bear genes coding for a hybrid LOV-HK with a fused response regulator (RR), and for two BphP proteins (BphP1 and BphP2) (Mandalari et al., 2013); however, BphP2 carries a compromised HK and second HK domain following in C-terminal direction (Shah et al., 2012). The gene encoding BphP2 is part of an operon that also contains a gene for a canonical RR of CheY type, whereas bphP1 is associated with a heme oxygenase-1 (here HO) encoding gene. The occurrence of similar photoreceptors in plants and in plant-pathogenic bacteria prompts the question whether this shared light-sensing capability is important for plant–bacteria interactions, chiefly for colonization, infectivity or virulence. Available data for BL flavinphotoreceptors pointing in this direction have recently been compiled (Losi et al., 2014). Considering selectively examples for P. syringae and related species, the results obtained so far can be summarized as follows: (i) quorum sensing in P. aeruginosa is affected by a BphP protein; however, the role of the light quality was not inspected (Barkovits et al., 2011). (ii) P. syringae pv. phaseolicola causes diseases in bean plants only after the loss of a genomic island (GI) that contains the gene for a BphP. Again, the role of light had not been investigated (Lovell et al., 2011). (iii) In P. syringae pv. syringae (Pss), BphP1 negatively regulates swarming motility in response to RL/FRL, whereas a LOV-HK suppresses negative regulation by BphP1 under BL (Wu et al., 2013). (iv) In Xanthomonas citri subsp. citri (Xcc, formerly X. axonopodis pv. citri), responsible for citrus canker, LOV-HK protein (XccLOV) not only modulates physiological attributes of the bacterium that are relevant for plant colonization and promotes adhesion to leaf surface, but also influences the development of disease symptoms on citrus in a light-dependent manner: in the light, XccLOV exerts a protective effect against necrosis, but not against canker (Kraiselburd et al., 2012). These effects seem to
be raised via the participation of XccLOV in several plant responses induced by infection, e.g. alteration of photosynthesis and defence response (Kraiselburd et al., 2013). BL absorption from XccLOV might be involved in a bacterial mechanism aiming to partially reduce the plant defence response and maintain tissue integrity, which is vital for hemibiotrophic pathogens. (v) Finally, in Pst, a LOV-HK protein (PstLOV) inhibits bacterial growth and expression of genes controlled by sigma factors (e.g. regulation of general stress response), and mediates BL-driven reduction of virulence in infected plants. In parallel, PstLOV enhances bacterial adhesion to leaves (Moriconi et al., 2013; Río-Álvarez et al., 2013). The reduction of virulence may be related to a reduced invasiveness and proliferation within leaves mediated by this photoreceptor (Moriconi et al., 2013). Different from Pss, BL has also an inhibiting effect on Pst virulence towards plant leaves, while the opposite has been reported for RL (Río-Álvarez et al., 2013), suggesting a still unidentified role for BphP proteins. Even less experimental results are available for the role of RL in plant–pathogen interactions. We have thus further investigated the in vivo relevance of BL sensing and have extended our work to the effect of RL sensing in Pst. The role of photoreceptors was studied with respect to bacterial growth under selected light conditions, bacterial proliferation/invasiveness within plant tissues and plant hypersensitive response (HR). For these investigations, we employed Arabidopsis plants as model hosts, Pst wild-type (PstWT) and deletion mutants in PstLOV, PstBphP1 and PstHO (HO catalysing formation of bilins). Our results indicate that the presence of both BL and RL photoreceptors renders Pst less virulent and less invasive towards host plants, and that these photoreceptors regulate bacterial growth under different light conditions.
Results and discussion Bacterial growth under different light conditions and influence of photoreceptors Knockout protocol for the PstLOV encoding gene (PSPTO2896) is described in Moriconi and colleagues (2013): genome positions 3258886 until 3260929 (PstLOV coding region) were cloned into the pJET1.2 vector (Life Technologies, Darmstadt, Germany). The BstZ17I restriction site located within the gene was used to insert a kanamycin cassette, isolated from the pBSL15 vector (Creative Biogene, Shirley, NY, USA) by SmaI digestion. This method yielded the inserted DNA in both orientations. Generation of the other knockout strains was carried out according to the following protocols: for the BphP1 encoding gene (PSPTO_1902) chromosomal positions from
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Bacterial photoreceptors in plant–pathogen interaction 2077242 to 2080311 was amplified and cloned; here a spectinomycin resistant omega (Ω-Spr) cassette isolated from pHP45-Ω by SmaI digestion was bluntend ligated into the EcoRV site of the gene. Heme oxygenase-1: chromosomal positions 2077011 to 2077625 was cloned into the pET14 vector (Novagene/Merck, Germany); a kanamycin cassette was inserted into a StuI site located within the gene. All gene constructs carried 5′and 3′-flanking regions of at least 250 bp in length. Sequencing of isolated genomic DNA with appropriate primers clearly showed the different size for the WTand the antibiotics cassette enlarged gene. Disrupted genes were then cloned into a suicide pSUP plasmid (Simon et al., 1983) (kindly donated by Thomas Drepper, Forschungszentrum Jülich, Germany), carrying a spectinomycin resistance cassette and the vector was linearized. Linear DNA (50–100 ng) was added to 100 μl of Pst cells (washed three times with ice-cold distilled water and kept in a 10% glycerol aqueous solution). Transformation (0.1 cm cuvette) was done with a Bio-Rad gene pulser [25 μF, 2.4 kV and 200 Ohm (5 ms)], immediately followed by addition of ice-cold KB (King’s B) medium. Further details will be described elsewhere (Shah, R., Pathak, G.P., Drepper, T., and Gärtner, W., in preparation). Cells that grew on the selective media after electroporation were re-streaked on plates containing higher amount of antibiotics (50 μg ml−1 of kanamycin or 100 μg ml−1 of spectinomycin) for several rounds. Finally, individual colonies from the antibiotic plate were withdrawn, and mutations of the chromosomal genes were verified by polymerase chain reaction (PCR), using gene-specific primers. Bacterial cultures were grown overnight in Luria– Bertani (LB) medium (Sambrook et al., 1989) at 27 ± 1°C with antibiotics added as outlined above. The PstWT and the various knockout mutant strains (ΔLOV-A, ΔLOV-B, ΔHO-A, ΔHO-B and ΔBphP1) were kept under different continuous light qualities or in darkness in order to determine the roles of different photoreceptors for their growth behaviour (Fig. 1). Cells were grown under these conditions for 48 h on solid medium, after which time the amount of colonies was counted as an estimate for the number of viable cells (cfu = colony-forming units) (Fig. 1). The growth of the six strains (WT and deletion mutants) was statistically different in all light conditions tested (oneway analysis of variance), with the exception of BL. This latter light quality completely repressed the growth of the mutants, under the conditions used. With reference to darkness conditions, RL caused an increase in growth capacity for ΔLOV-A, ΔLOV-B, and to a lesser extent for ΔHO-B (significantly different, Student’s t-test). None of the other applied light qualities yielded a significantly different growth behaviour in comparison to darkness. We can thus infer that both PstLOV and PstBphP1
797
Fig. 1. Effects of different light conditions on the growth of investigated bacterial strains. Error bars indicate standard deviation. cfu, colony-forming units. RL (λmax = 660 nm; 6 μmol m−2 s−1), FRL (λmax = 770 nm; 17 μmol m−2 s−1) or BL (λmax = 462 nm; 40 μmol m−2 s−1) were provided by light-emitting diodes (LEDs) from Roithner Lasertechnik, Vienna, Austria (diameter 5 mm each). The different photon fluence rates were due to the necessity to position the plates at a given distance from the light sources, as described in Wu and colleagues (2013). To determine the cfu, an exponential-phase culture (OD600 = 0.6) of each bacterial strain was used to prepare serial 1:10 dilutions necessary to get countable colonies. Of each dilution, 100 μl was spread on the surface of 2% (wt vol−1) LB agar plates. The plates were incubated at 28°C in separate chambers under the different light conditions. Plates kept in darkness were used as control. Three plates per treatment were handled in parallel and experiments were repeated twice. The colony number was determined from each plate after 48 h, and the significance of differences among multiple groups was evaluated using the analysis of variance (one-way ANOVA) test, while comparisons between two groups were done by Student’s t-test. Results were considered to be statistically significant at p ≤ 0.01.
downregulate bacterial growth under WL, RL, FRL and in darkness. Deletion of the HO, required for bilin production, has the same effect. In fact, each mutant strain shows a greater growth rate than the PstWT when exposed to these light conditions (continuous illumination). On the contrary, under BL only PstWT strain retains a growth capacity similar to that shown in darkness, while the mutant strains did not grow at all under our experimental conditions. Apparently, the simultaneous presence of PstLOV, PstHO and PstBphP1 photoreceptor proteins is an indispensable requirement to guarantee growth under BL in the described culturing conditions. One might assume that photoreceptors trigger a still unknown protective response towards BL that is potentially phototoxic. The results also suggest an interplay between LOV and BphP proteins and an integrated regulatory network of these photoreceptor proteins in Pst, as recently demonstrated for the regulation of swarming motility in P. syringae B728a (Wu et al., 2013). We note here that the growth of PstWT is significantly reduced under FRL, possibly accounting for the slightly reduced growth under WL. These findings suggest a sort of balancing among the effects elicited by the different wavelengths, as PstWT growth under WL is only slightly reduced.
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
798 A. Ricci et al. Fig. 2. Representative plates displaying bacterial growth resulting from homogenate smear of Arabidopsis leaves infiltrated (I) and non-infiltrated (NI) with PstWT (A, B) or ΔLOV-A (C, D) bacterial strains. For plant inoculation, the different bacterial strains were suspended at a density of approximately 1 × 108 or 2.5 × 108 cfu ml−1 (OD600 = 0.2 or 0.5 respectively). Pictures were taken after incubation at 25 ± 1°C for 48 h in darkness. For the remaining strains, see Fig. S1.
Our results on light-regulated growth appear to be in contrast to previously published data by Moriconi and colleagues reporting that only WL was able to promote growth of a ΔLOV Pst strain (Moriconi et al., 2013), and more importantly these authors did not observe any difference between PstWT and ΔLOV-A under BL. Due to the different experimental conditions, the two sets of data are difficult to compare, because in the work by Moriconi and colleagues the cell density in liquid medium was measured (OD600), whereas here the number of viable cells was used as parameter. Yet these authors note that PstWT showed a longer initial growth lag than ΔLOV mutant strains under WL (Moriconi et al., 2013). Further work, also taking into account e.g. the diverse media composition or the illumination regime, is required to fully clarify this aspect and the discrepancy among different datasets. Infectivity and bacterial growth within infected leaves Arabidopsis thaliana plants, ecotype Col-0, were grown from seed on a sterile mixture of soil and vermiculite (1:1) in a growth chamber at WL intensity of approximately 30 μmol m−2 s−1 at 21 ± 1°C under 16/8 h day/night photoperiod. For plant inoculation, overnight-cultured bacteria were suspended in 10 mM MgCl2 at a density of approximately 1 × 108 or 2.5 × 108 cfu ml−1 (OD600 = 0.2 or 0.5, respectively), and leaves of 5- to 6-week-old plants were infiltrated into their abaxial surface using a 1 ml syringe without needle, following a previously described protocol with minor modifications (Katagiri et al., 2002; Griebel and
Zeier, 2008). Infiltration with 10 mM MgCl2 was used as control. Infiltrated plants were placed in plastic boxes filled with wet, expanded clay covered with a transparent polythene cover to ensure high humidity. The boxes were kept at WL intensity of approximately 30 μmol m−2 s−1 at 19 ± 1°C under a 16/8 h day/night photoperiod. Bacterial growth in planta was determined 72 h after inoculation. Infiltrated and non-infiltrated opposite leaves were sampled, surface-sterilized with 70% ethanol for 1 min, and then washed twice with sterile, distilled water to remove bacteria from the surface. Leaves were ground in 1 ml LB medium, the homogenate was shaken and plated on solid LB medium (2% agar wt vol−1) supplemented with the appropriate antibiotics. Following incubation at 25 ± 1°C for 48 h in darkness, bacterial growth, and accordingly distribution inside the plants, was assessed. For each experiment, repeated three times, six to ten plants were infiltrated with the different bacterial strains. Bacteria were detected from the homogenate of all infiltrated and non-infiltrated leaves, independently of the bacterial density of the inoculum. In both cases, knockout mutants showed a larger number of colonies (Fig. 2 and Fig. S1). These results agree with the growth assays of bacteria illustrated above. A more straightforward investigation of the infectivity of PstWT and mutated strains was obtained by means of scanning electron microscopy (SEM) on cross-sections of infiltrated and non-infiltrated leaves, according to the method of Misas-Villamil and colleagues (2011) with minor modifications. Only A. thaliana plants infiltrated with bacteria at a density of approximately 2.5 × 108 cfu ml−1
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Bacterial photoreceptors in plant–pathogen interaction
799
Fig. 3. SEM representative pictures of infiltrated leaves taken 24 h after infiltration with bacterial strains [(A) PstWT, (B) ΔLOV-A, (C) ΔLOV-B, (D) ΔHO-A, (E) ΔHO-B and (F) Δbphp1, respectively] at a density of approximately 2.5 × 108 cfu ml−1 (OD600 = 0.5). No clearly recognizable bacteria were visible in the cross-sections of leaves infiltrated with the wild-type bacterial strain, but only tissue alteration. For the control condition (infiltration with 10 mm MgCl2) and the other times of investigation, see Fig. S2.
(OD600 = 0.5) or with 10 mM MgCl2 solution as control were used in this assay. Infiltrated and non-infiltrated opposite leaf samples were scored at 24, 48 and 72 h after infiltration, fixed in 3% glutaraldehyde in phosphate buffer 0.1 M, pH 7.0, dehydrated using acetone and dried in a critical point drier (Balzers Union CPD 020) under a CO2 atmosphere. Specimens were gold-coated using a Balzers Union Sputter Coater and examined using SEM (Philips SEM 501, provided by Dipartimento di Scienze Biomediche, Biotecnologiche e Traslazionali, University of Parma). This approach is able to reveal the presence of bacteria in planta, and possibly their distant location from the infiltration spot. For the mutated strains, a huge amount of bacteria was clearly visible in the apoplast regions near the xylem vessels, and also trapped in granular and/or fibrillar structures of the infiltrated leaves (Fig. 3 and Fig. S2). On the other hand, only the presence of granular and fibrillar structures altering the internal leaf architecture without clearly recognizable bacteria was achieved by the SEM analysis of the cross-sections of leaves infiltrated with PstWT (Fig. 3 and Fig. S2). Neither bacteria nor fibrillar structures were ever detected under control conditions (leaves infiltrated with 10 mM MgCl2 solution; Fig. S2). An example of complete collapse of host cells could be documented 48 h after infiltration: upper and lower epidermis surround an empty space, while ‘buds’ of unknown matrix embedding bacteria are extruded from a lot of stomata of the upper epidermis (Fig. S3). In order to investigate the distant location of bacteria, non-infected leaves were sampled and analysed as well.
Only the mutant strains rapidly moved inside the plants, allowing the detection of colonization of non-infiltrated leaves (Fig. S4). In contrast, no distant location of WT bacteria was achieved at 24, 48 or 72 h after infiltration, or under control conditions. Plant HR assay Arabidopsis thaliana plants infiltrated with the different bacterial strains at a density of approximately 2.5 × 108 cfu ml−1 (OD600 = 0.5) or with 10 mM MgCl2 solution, as control, were used in this assay. Infiltrated leaves were scored for tissue collapse (denoting HR) at 24, 48 and 72 h after infiltration (Fig. 4). The evolution of macroscopic symptoms was registered with a Canon Ixus 200 IS digital camera as previously described (Kabisch et al., 2005). For each experiment, repeated three times, six to ten plants were infiltrated with the different bacterial strains. The first symptoms of HR in plants infiltrated with the mutated strains appeared 48 h after inoculation in the form of diffuse chlorosis (Fig. 4H, K, N, Q) and random necrotic spots (Fig. 4E). Tissue collapse became more severe at 72 h after inoculation, with yellowing occurring on the entire leaf (Fig. 4F, L) and necrotic zones being either more frequent or wider in size (Fig. 4I, O, R). Leaves of plants inoculated with PstWT did not develop comparable symptoms during the period of investigation (Fig. 4A–C). In these plants, yellowing appeared 8–10 days after inoculation (data not shown). Control plants that were injected with a 10 mM
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
800 A. Ricci et al. MgCl2 solution without bacteria did not develop symptoms (Fig. 4S, T, U). From our data we can infer that all bacterial photoreceptors analysed in this study downregulate infectivity versus Arabidopsis plants and growth of bacteria within infected leaves. As for LOV proteins, this is consistent with previous work either on Pst or on X. citri (Kraiselburd et al., 2012; 2013; Moriconi et al., 2013). The present data, obtained with strains unable to synthesize functional BphP proteins, suggest a similar role also for bacteriophytochromes during Pst–plant interaction. Conclusions
Fig. 4. Representative phenotypes of leaves displaying a hypersensitive reaction (HR) after inoculation with Pst DC3000 (A–C), ΔLOV-A (D–F), ΔLOV-B (G–I), ΔHO-A (J–L), ΔHO-B (M–O), Δbphp1 (P–R) and 10 mM MgCl2 as control (S–U). Leaves were scored at 24, 48 and 72 h after infiltration. *: evident reaction; **: more severe collapse.
The data presented here underscore the striking observation that bacterial photoreceptors can downregulate infectivity towards host plants, with which they share the spectral range sensed (BL and RL/FRL). Still, this phenomenon is only sparsely investigated, as is evident from the low number of reports published so far. This feature, first brought forth for LOV proteins, apparently holds also for bacteriophytochromes, even if the available data are still limited to very few systems. It is clear that during plant–pathogen interactions different interests and survival strategies meet: pathogens need to grow inside plant tissues from which they obtain nutritional compounds; plants must rapidly react by activating efficient defence responses. This is clearly observable when A. thaliana plants are infected with PstWT. Bacterial pathogenicity depends on effectors, either proteins delivered by the type III secretion system (T3SS) directly into host cells or phytotoxins as coronatine. Conversely, plants display immune signalling pathways resulting in callose deposition, antibacterial phytoalexins, ROS, HR and accumulation of salicylic acid (Kabisch et al., 2005; Návarová et al., 2012; Xin and He, 2013). Many T3SS effectors of plant-pathogenic bacteria suppress plant innate immunity by different mechanisms altering plant behaviour and development, up to death. Could bacterial photoreceptors play a role in plant–pathogen interactions being rapidly recognized by plant cells as non-host proteins and eliciting the different plant immune signalling pathways? In order to confirm or discard this hypothesis, deeper investigations about plant defence responses are essential. With respect to this, in a recently published paper the growth, swarming and biofilm formation capacity of Pseudomonas cichorii JBC1 was tested under different light conditions, and no significant difference to dark-grown cells was detected for red, green and WL illumination (Nagendran and Lee, 2015). Nevertheless, tomato seedlings infected with P. cichorii JBC1 showed significant reduction in disease incidence and internal bacterial growth under green and red light, concomitant with significant upregulation of
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
Bacterial photoreceptors in plant–pathogen interaction plant defence-related gene. The role of bacterial photoreceptors was nevertheless not tested in that work (P. cichorii JBC1 has an LOV-HK and two BphPs, similar to Pst). The data presented here and elsewhere (Kraiselburd et al., 2012; 2013; Moriconi et al., 2013; Río-Álvarez et al., 2013; Wu et al., 2013) suggest that bacterial photoreceptors could indeed have a stimulating function as protective factors during plant infection, although the integration of bacterial light-sensing with plant photosensory transduction pathways is an open research field. The knowledge that genes for LOV and BphP proteins in P. syringae pathovars have been localized on GI that bears other genes relevant for virulence suggests that they might have been acquired via horizontal gene transfer (Lovell et al., 2011; Moriconi et al., 2013). Although the cross-talk with virulence genes and the actual significance of photoreceptors in such GI is still to be clarified, we consider this a key point for understanding the role of LOV and BphP proteins during infection, at least for this species. Complementation and interpretation of experiments with double mutants, already performed in some cases (Moriconi et al., 2013; Wu et al., 2013), also have to take into account the localization of these genes on GI. The importance of the topic dealt in this manuscript, i.e. the impact of bacterial photoreceptors on plant–pathogen interactions, is likely to grow in the near future, given the increasing importance of bacterial photosensors for bacterial lifestyles and metabolism (Mandalari et al., 2013; Losi et al., 2014). Future work will explore the role of Pst photoreceptors during infection of tomato plants, the main agronomically relevant host of this species (Xin and He, 2013). Acknowledgements The authors are deeply grateful to Davide Dallatana (Dipartimento di Scienze Biomediche, Biotecnologiche e Traslazionali, University of Parma) for his excellent technical assistance in the SEM analyses. The authors also acknowledge Dr. Alessio Sardella for his partial contribution to the experimental procedures. PstWT was kindly provided by Robin Buell (Institute for Genomic Research, Rockville, MD, USA). The authors declare no conflict of interest for this manuscript.
References Auldridge, M.E., and Forest, K.T. (2011) Bacterial phytochromes: more than meets the light. Crit Rev Biochem Mol 46: 67–88. Barkovits, K., Schubert, B., Heine, S., Scheer, M., and Frankenberg-Dinkel, N. (2011) Function of the bacteriophytochrome BphP in the RpoS/Las quorumsensing network of Pseudomonas aeruginosa. Microbiology 157: 1651–1664.
801
Dasgupta, J., Frontiera, R.R., Taylor, K.C., Lagarias, J.C., and Mathies, R.A. (2009) Ultrafast excited-state isomerization in phytochrome revealed by femtosecond stimulated Raman spectroscopy. Proc Natl Acad Sci USA 106: 1784–1789. Elías-Arnanz, M., Padmanabhan, S., and Murillo, F.J. (2011) Light-dependent gene regulation in nonphototrophic bacteria. Curr Opin Microbiol 14: 128–135. Gomelsky, M., and Hoff, W.D. (2011) Light helps bacteria make important lifestyle decisions. Trends Microbiol 19: 441–448. Griebel, T., and Zeier, J. (2008) Light regulation and daytime dependency of inducible plant defenses in Arabidopsis: phytochrome signaling controls systemic acquired resistance rather than local defense. Plant Physiol 147: 790– 801. Herrou, J., and Crosson, S. (2011) Function, structure and mechanism of bacterial photosensory LOV proteins. Nat Rev Microbiol 9: 713–723. Kabisch, U., Landgraf, A., Krause, J., Bonas, U., and Boch, J. (2005) Type III secretion chaperones ShcS1 and ShcO1 from Pseudomonas syringae pv. tomato DC3000 bind more than one effector. Microbiology 151: 269– 280. Karpinski, S., Gabrys, H., Mateo, A., Karpinska, B., and Mullineaux, P.M. (2003) Light perception in plant disease defence signalling. Curr Opin Plant Biol 6: 390–396. Katagiri, F., Thilmony, R., and He, S.Y. (2002) The Arabidopsis thaliana-Pseudomonas syringae interaction. Arabidopsis Book 1: e0039. doi:10.1199/tab.0039. Kraiselburd, I., Alet, A.I., Tondo, M.L., Petrocelli, S., Daurelio, L.D., Monzón, J., et al. (2012) A LOV protein modulates the physiological attributes of Xanthomonas axonopodis pv. citri relevant for host plant colonization. PLoS ONE 7: e38226. Kraiselburd, I., Daurelio, L.D., Tondo, M.L., Merelo, P., Cortadi, A., Talón, M., et al. (2013) The LOV protein of Xanthomonas citri subsp. citri plays a significant role in the counteraction of plant immune responses during citrus canker. PLoS ONE 8: e80930. Lim, C.W., Luan, S., and Lee, S.C. (2014) A prominent role for RCAR3-mediated ABA signaling in response to Pseudomonas syringae pv. tomato DC3000 infection in Arabidopsis. Plant Cell Physiol 55: 1691–1703. Losi, A. (2007) Flavin-based blue-light photosensors: a photobiophysics update. Photochem Photobiol 83: 1283– 1300. Losi, A., Mandalari, C., and Gärtner, W. (2014) From plant infectivity to growth patterns: the role of blue-light sensing in the prokaryotic world. Plants 3: 70–94. Lovell, H.C., Jackson, R.W., Mansfield, J.W., Godfrey, S.A.C., Hancock, J.T., Desikan, R.A.D.H., and Arnold, D.L. (2011) In planta conditions induce genomic changes in Pseudomonas syringae pv. phaseolicola. Mol Plant Pathol 12: 167–176. Mandalari, C., Losi, A., and Gärtner, W. (2013) Distancetree analysis, distribution and co-presence of bilin- and flavin-binding prokaryotic photoreceptors for visible light. Photochem Photobiol Sci 12: 1144–1157. Misas-Villamil, J., Kolodziejek, I., and van der Hoorn, R. (2011) Pseudomonas syringae colonizes distant tissues in
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
802 A. Ricci et al. Nicotiana benthamiana through xylem vessels. Plant J 67: 774–782. Moriconi, V., Sellaro, R., Ayub, N., Soto, G., Rugnone, M.A., Shah, R., et al. (2013) LOV-domain photoreceptor, encoded in a genomic island, attenuates the virulence of Pseudomonas syringae in light-exposed Arabidopsis leaves. Plant J 76: 322–331. Müller, M., Lindner, I., Martin, I., Gärtner, W., and Holzwarth, A.R. (2008) Femtosecond kinetics of photoconversion of the higher plant photoreceptor phytochrome carrying native and modified chromophores. Biophys J 94: 4370– 4382. Nagendran, R., and Lee, Y.H. (2015) Green and red light reduces the disease severity by Pseudomonas cichorii JBC1 in tomato plants via upregulation of defense-related gene expression. Phytopathology 150: 412–418. Návarová, H., Bernsdorff, F., Döring, A.C., and Zeier, J. (2012) Pipecolic acid, an endogenous mediator of defense amplification and priming, is a critical regulator of inducible plant immunity. Plant Cell 24: 5123–5141. Río-Álvarez, I., Rodríguez-Herva, J.J., Martínez, P.M., González-Melendi, P., García-Casado, G., Rodríguez-Palenzuela, P., and López-Solanilla, E. (2013) Light regulates motility, attachment and virulence in the plant pathogen Pseudomonas syringae pv. tomato DC3000. Environ Microbiol 16: 2072–2085. Rockwell, N.C., and Lagarias, J.C. (2010) A brief history of phytochromes. Chem Phys Chem 11: 1172–1180. Roden, L.C., and Ingle, R.A. (2009) Lights, rhythms, infection: the role of light and the circadian clock in determining the outcome of plant-pathogen interactions. Plant Cell 21: 2546–2552. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual. Cold Spring Harbor, NY, USA: Cold Spring Harbor Laboratory Press. Shah, R., Schwach, J., Frankenberg-Dinkel, N., and Gärtner, W. (2012) Complex formation between heme oxygenase and phytochrome during biosynthesis in Pseudomonas syringae pv. tomato. Photochem Photobiol Sci 11: 1026– 1031. Simon, R., Priefer, U., and Pühler, A. (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteria. Biotechnology 1: 784–791. Wu, L., McGrane, R.S., and Beattie, G.A. (2013) Light regulation of swarming motility in Pseudomonas syringae integrates signaling pathways mediated by a bacteriophytochrome and a LOV protein. mBio 4: e00334–13.
Xin, X.F., and He, S.Y. (2013) Pseudomonas syringae pv. tomato DC3000: a model pathogen for probing disease susceptibility and hormone signaling in plants. Annu Rev Phytopathol 51: 473–498. Zoltowski, B.D., and Gardner, K.H. (2011) Tripping the light fantastic: blue-light photoreceptors as examples of environmentally modulated protein-protein interactions. Biochemistry 50: 4–16.
Supporting information Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Fig. S1. Representative plates displaying bacterial growth resulting from homogenate smear of Arabidopsis leaves infiltrated (I) and non-infiltrated (NI) with all the bacterial strains. For plant inoculation, the different bacterial strains were suspended at a density of approximately 1 × 108 or 2.5 × 108 cfu ml−1 (OD600 = 0.2 or 0.5 respectively). Infiltration with 10 mM MgCl2 solution was used as control. Pictures were taken after incubation at 25 ± 1°C for 48 h in darkness. Fig. S2. SEM representative pictures of infiltrated leaves taken 48 and 72 h after infiltration with bacterial strains [(A) PstWT, (B) ΔLOV-A, (C) ΔLOV-B, (D) ΔHO-A, (E) ΔHO-B and (F) ΔBphP1, respectively], at a density of approximately 2.5 × 108 cfu ml−1 (OD600 = 0.5). No clearly recognizable bacteria were visible in the cross-sections of leaves infiltrated with the wild-type bacterial strain, only tissue alteration could be detected. SEM representative pictures of leaves infiltrated with 10 mM MgCl2 control solution taken 24, 48 and 72 h after infiltration are also shown. Fig. S3. Example of host cell collapse 48 h after the infiltration of cell suspension of the mutant strain ΔBphP1 had been performed. (A) Upper and lower epidermis surrounding an empty space; arrows indicate ‘buds’ on upper edge of the empty space; (B) upper epidermis full of ‘buds’; (C) ‘buds’ emerging from the stomata; and (D) bacteria embedded in an undefined matrix extruded from the stomata. Fig. S4. (A) Example of colonization of non-infiltrated leaf at 24 h after infiltration with the ΔHO-B mutant strain; (B) magnification of (A) showing bacteria embedded in a compact substance (arrow). (C) Example of colonization of a noninfiltrated leaf at 72 h after infiltration with ΔHO-B mutant strain; (D) magnification of (C) showing bacteria in the apoplast region. (E) Example of colonization of non-infiltrated leaf 72 h after infiltration with ΔLOV-B mutant strain; (F) magnification of (E).
© 2015 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology Reports, 7, 795–802
