Bacterial Photosensory Proteins and Their Role in Plant–pathogen Interactions^†

Light regulation and photoreceptor proteins in bacteria of the Xanthomonas genus

Xanthomonas citri subsp. citri (Xcc) is the bacterium responsible for citrus canker, a disease that affects all commercial varieties of Citrus. Bacterial infections that occur early in fruit development result in serious production losses. Late infections only result in scattered necrotic corky lesions on leaves, stems and fruits; however, these fruits become unacceptable for international commercialization 17, 18. Xcc presents multiple virulence factors involved in host colonization, plant immune response evasion and disease development. These include type II and type III secretion systems, responsible for secreting enzymes for the degradation of plant cell components and for injecting of virulence effectors inside the host cells, respectively, and other surface structures important for plant colonization such as adhesins, lipopolysaccharides and exopolysaccharides 19. The Xcc genome sequence includes four genes encoding putative photoreceptors: a phytochrome, two BLUF proteins and a LOV protein 20.

The LOV protein of Xcc (Xcc-LOV) presents an N-terminal LOV domain associated with a C-terminal histidine kinase (HK) domain and a response regulator (RR) domain (hybrid HK-RR). The amino acid sequence of Xcc-LOV presents all the essential amino acids for LOV photochemistry, including the characteristic GXNCRFLQ motif containing the conserved cysteine residue involved in the covalent adduct formation upon blue light absorption (C76 in Xcc) 21, 22. In fact, Xcc-LOV is a legitimate photoreceptor capable of absorbing blue light and initiates a canonical LOV photocycle 20. In this protein, photochemistry does not take place in C76 mutations; confirming the involvement of this residue in the generation of the covalent photoadduct essential for photochemistry 22. The mechanism by which the Xcc-LOV photoreceptor is activated upon an incoming blue light photon to start a signal transduction cascade has been elucidated by a complete spectroscopic analysis. This study demonstrated that upon photoconversion, the Xcc-LOV protein does not undergo significant conformational changes, as indicated by unaffected fluorescence spectra of tryptophan and tyrosine residues and fluorescence anisotropy. Moreover, laser-induced optoacoustic spectroscopy (LIOAS) studies revealed that, unlike previously characterized LOV proteins, the photoconversion of Xcc-LOV does not involve a volume contraction within the observation time window. However, as the back-conversion into the dark state is, in fact, accompanied by a volume expansion, it can be assumed that the volume contraction associated with adduct formation in the wild-type falls at the limit of the time window for LIOAS or its absolute value is very small. On the other hand, the Xcc-LOV protein exhibits quantum yields for triplet (ΦT = 0.68) and photoadduct (Φ₃₉₀ = 0.46) formation as well as the lifetime for triplet decay (τ_T = 2.4–2.8 μs) similar to other described bacterial LOV proteins including those from Pseudomonas syringae pv. tomato and Bacillus subtilis 22.

The light state of the Xcc-LOV protein is the active (signaling) state, probably initiating a signal transduction cascade involved in a physiological response. This was demonstrated by a radioactivity-based enzyme function assay which revealed a light-induced upregulation of the kinase function of Xcc-LOV 22. Regarding the role of the Xcc-LOV in the bacterial physiology, this protein modulates bacterial features directly associated with the Xcc ability to colonize host plants, such as motility, adhesion, biofilm formation and oxidative stress resistance, by means of the synthesis of flagellum, exopolysaccharides and adhesins. Moreover, disease development during the bacterial infection of orange trees depends on the functionality of this protein, as considerably different symptoms developed in plants infected with a lov-mutant strain of Xcc than those developed upon Xcc wild-type infection, with a higher degree of tissue necrosis 20. An expression profiling of Citrus sinensis leaves revealed that genes directly involved in plant defense are upregulated upon infection with a lov-mutant strain. The principal biological processes differentially affected by treatments include photosynthesis, sucrose catabolism, secondary metabolism and defense response. Moreover, leaves inoculated with the lov-mutant strain present a more pronounced decrease in photosynthesis, a higher lignin accumulation and tissues integrity are more severely affected compared to wild-type Xcc-inoculated leaves 23. Furthermore, a qualitative and quantitative evaluation of reactive oxygen species in C. sinensis shows a more marked oxidative burst in plants inoculated with the lov-mutant strain of Xcc (I. Kraiselburd, L. Moyano, S. Petrocelli, E. G. Orellano, unpublished) (Fig. 1).

Based on its lifestyle, Xcc is classified as a hemibiotrophic pathogen, which relies on metabolites derived from plant cells to grow. Therefore, maintaining plant photosynthesis and the integrity of host plant tissues favors bacterial establishment and survival. In fact, in Xcc, bacterial modulation of host tissue damage is an important mechanism for surviving in the host long enough to develop the disease and to spread to neighboring tissues 24. Moreover, although the initiation of plant immune responses and the biosynthesis of defense compounds are energy-demanding processes, plant photosynthesis is negatively regulated upon pathogen attack to prioritize the metabolic processes involved in the production of defense compounds 25. The high representation of upregulated genes related to plant defense mechanisms against pathogen attack in tissues inoculated with the lov-mutant strain of Xcc, as well as with the enhanced biochemical and structural host tissue alterations, shows a stronger host response upon infection with this strain. These results suggest that Xcc-LOV participates in a bacterial strategy to counteract plant defense responses and to maintain plant energetic metabolism and integrity, thus favoring bacterial survival in host tissues and disease development (plants were grown in a greenhouse with a photoperiod of 16 h light = 150 μE m⁻² s⁻¹, and 8 h dark at a temperature of 25 °C and 80% humidity) 23.

In the case of Xcc BLUF proteins, preliminary data show that the protein encoded by the bluf3278 gene is involved in the modulation of different physiological features such as swimming and swarming motility, exopolysaccharide and biosurfactant production and adhesion to orange leaves. Furthermore, this gene participates in the regulation of the interaction between Xcc and the host plant (A. Carrau, J. Tano, L. Moyano, S. Petrocelli, L. M. Moreira, J. Setubal, E. G. Orellano, unpublished). The physiological role of bacterial BLUF proteins is still largely unexplored, and this information represents the first evidence of a functional BLUF photoreceptor in a bacterial plant pathogen.

The related bacterium Xanthomonas campestris pv. campestris (Xcca) is responsible for crucifer black rot, a disease affecting all cultivated varieties of brassicas worldwide. Infections by Xcca can occur at any stage of the plant life cycle and typical symptoms are v-shaped chlorotic necrotic lesions extending out from the leaf margins and blackening of vascular tissues 26. This bacterium presents several proteins potentially involved in blue, tricolor (blue, red and far-red) or red/far-red light signaling, belonging to three functional groups: HKs, GGDEF proteins and hybrid HKs. Mao et al. demonstrated that these proteins modulate growth, motility and virulence of Xcca and are regulated both by the light quality (wavelength) and quantity (intensity) 27. Bacterial growth is modulated by exposure to red (4.30 mW cm⁻²) and far-red light (3.36 mW cm⁻²), this effect being dependent on two HKs, two hybrid HKs and three GGDEF proteins. White (12000 lux) and blue light (763 μW cm⁻²) has a stronger influence on the behavior of Xcca, with three HKs, four GGDEF proteins and four hybrid HK putative involved in the signaling process in response to these radiations 27. Considering blue light in particular, a deletion mutant in an HK protein exhibits an impaired growth under this wavelength, in contrast to what was observed in Xcc, where light does not influence bacterial growth. The influence on bacterial virulence has been established for four light-sensing proteins, three of which present GGDEF domains, confirming that the virulence of Xcca is modulated in response to lighting conditions 27.

Xcca has a single bacteriophytochrome. The crystal structure bearing a full-length BphP with a canonical structure PAS–GAF–PHY photosensory module C-terminally linked to a PAS output domain was recently published 28. In addition, Otero et al. demonstrated that the phytochrome of Xcca acts as a bathy-like phytochromes 28, 29. These kinds of phytochromes have a longer wavelength Pfr form than the ground state 30, 31. The specific functions of the Xcca-phytochrome have been recently described. In this bacterium, the bacteriophytochrome downregulates exopolysaccharide production and attachment in response to light (15 μE m⁻² s⁻¹). Also, a phytochrome-mutant strain showed enhanced virulence. Moreover, the mutant strain showed an alteration in the stomata aperture regulation and callose deposition 32. Furthermore, Bonomi et al. demonstrated that the far-red light (733 nm; 0.7 W LEDs at a 15 cm distance) or the Xcca-phytochrome overexpression produces the inhibitions of several virulence factors in this bacterium 32.

Effects of light and photosensory proteins on the physiology and virulence of Pseudomonas plant pathogenic bacteria

Pseudomonas syringae pv. tomato DC 3000 (Psto) bacterium is the causal agent of bacterial speck in tomatoes and other plants including A. thaliana. In the first step of infection, Psto develops an epiphytic phase on the plant surface. Then, the bacterium enters into the apoplast through natural openings and wounds and produces the disease symptoms. Bacterial speck is characterized by the development of necrotic symptoms in leaves, stems and fruits 33, 34. The Psto genome was completely sequenced 35, and it presents a LOV-HK-RR protein (PSPTO_LOV) with a light-regulated HK activity 36 and two bacteriophytochromes (PSPTO_PHY1 and PSPTO_PHY2), whose activity is triggered by red light 7. The Psto-LOV protein presents a canonical LOV photochemistry, with an absorbance spectrum typical for oxidized flavin species and fluorescence emission in the dark, with the appearance of a species absorbing maximally at 390 nm and loss of flavin fluorescence upon blue light irradiation, indicating the formation of a covalent flavin-protein photoadduct 36. This adduct slowly reverts to the dark state with a lifetime of 5650 s. LIOAS evaluation of this protein revealed similar properties to those observed in other LOV proteins, such as Xcc-LOV. Transient species are formed within 20 ns corresponding to the FMN triplet state, with a small volume contraction. Then, triplet decays with a lifetime of 1.5 μs generating the covalent photoadduct. The quantum yields of triplet and adduct formation are consistent with the values reported for most LOV proteins; however, unlike other described LOV proteins, adduct formation does not result in a further contraction, but rather in a small expansion 36. Río-Álvarez et al. demonstrated that while light does not affect bacterial growth in liquid and solid media, Psto exhibits an inhibition of its typical swarming motility when grown under white (70 μE m⁻² s⁻¹) or blue light (20 μE m⁻² s⁻¹). This effect is a consequence of a downregulation of the expression of genes involved in the synthesis of bacterial flagella and in the regulation of flagellar functions. However, bacterial growth under red light (20 μE m⁻² s⁻¹) and darkness favors bacterial motility, showing the participation of the Psto-LOV protein in the inhibition of this process. In addition, bacterial attachment and biofilm formation in the leaves of host plants are favored under blue light. This is thought to be the result of an upregulation of genes involved in the alginate biosynthesis, an exopolysaccharide involved in adhesion and resistance mechanisms during the epiphytic stage. Based on these effects, the authors suggest that blue light provides a signal for the bacterial switch to a nonmotile attached lifestyle 37. On the other hand, light is involved in the regulation of bacterial virulence in Psto, because bacterial populations and disease symptoms developed in tomato and A. thaliana plants are significantly reduced when bacteria are grown under blue light conditions. Bacterial growth in red light causes the opposite effect, increasing the disease symptoms compared to the treatment with bacteria grown in the darkness 37. The reduced virulence of Psto under blue light conditions is thought to be associated with the Psto-LOV-dependent effect of this radiation on bacterial flagella, as these structures are essential for the initiation of bacterial infection 38. In a Psto-LOV-mutant strain, bacterial virulence is reduced in comparison with the wild-type strain despite the light conditions and it is not affected by blue or white light. Moreover, the opposite effect is observed under red light, suggesting that bacteriophytochromes also have a role in the virulence of this bacterium 37. LOV protein in Psto is encoded by a gene present in a pathogenicity island (LOV domain gene genomic island or LOV–GI), which also encodes genes involved in circadian timing (a gene homologous to labA from Synechococcus) 39, light responses 40 and putative virulence factors 41. Despite not affecting bacterial growth in rich media, Psto-LOV protein seems to be involved in the light-dependent inhibition of the growth of Psto in minimum media. Consistent with these facts, Psto-LOV protein has been shown to be a master negative regulator of gene expression in this plant pathogen. Psto-LOV is involved in the light-dependent downregulation of the expression of genes encoding the principal sigma factors highly expressed during bacterial exponential growth and alternative sigma factors involved in general stress response, secondary metabolite production, nutrient-scavenging stress response and production of type III secretion system and type III virulence effectors 42, 43. As established by Río-Álvarez et al. and Moriconi et al. repression of these genes by Psto-LOV protein can account for the reduced bacterial proliferation in A. thaliana plants inoculated with Psto grown under white and blue light conditions, not only by an impaired bacterial motility and thus inability of bacterial infection, as previously mentioned, but also as a result of a reduced bacterial replication once inside the plant apoplast 37, 41. Moreover, the Psto-LOV-directed repression of type III secretion system and effectors probably results in the reduced virulence reported for Psto in the light 37. In fact, Moriconi et al. demonstrated that this effect is the result of an impaired bacterial invasiveness in leaf tissues, that is the capacity of the pathogen to grow and spread within the host 41. Psto is a hemibiotrophic pathogen that grows epiphytically on plant surfaces without causing disease symptoms until it enters the plant leaf apoplast, where it multiplies by exploiting live host cells. This bacterium also survives as a saprophyte in the soil 44-47. Moriconi et al. also suggested that while the lack of effect of Psto-LOV in the darkness may favor bacterial survival in saprophytic conditions, reduced bacterial proliferation in the light can reduce the damage in the host tissue, thus optimizing the bacterial nutritional resources in order to allow the dispersal to new hosts 41. Phytochromes of Psto are biliverdin IXα (BV)-binding phytochromes. PstobphP1 is arranged in an operon with a heme oxygenase (PstoBphO)-encoding gene (PstobphO), whereas PstobphP2 is flanked downstream by a gene encoding a CheY-type response regulator 7. Concerning the role of the bacteriophytochromes in the physiology of Psto, these proteins modulate several features related to the epiphytic phase. The ability to adhere to biotic surfaces and biofilm development are regulated by phytochromes, as these proteins have been shown to be necessary for a successful bacterial attachment. The production of biosurfactants, which is another feature related to the epiphytic life of bacteria 48, is markedly reduced in ΔbphP1 and ΔbphP2 strains, indicating that phytochromes of Psto have a positive regulation of these features. In agreement with this result, epiphytic fitness assays showed that tomato leaves inoculated with mutant strains presented less typical bacterial speck symptoms. Swarming motility has an essential role in the colonization of natural environments, the flagellum being the most important requirement for this kind of motility 49. Migration zones of Psto strains exposed to darkness are higher than those observed when bacteria are exposed to white (20 μE m⁻² s⁻¹) and red light (40 μE m⁻² s⁻¹) conditions. In red light, swarming motility is decreases but is not completely repressed. Moreover, ΔbphP1 and ΔbphP2 mutants migrate further than the wild-type strain in every light condition. The examination of the flagella from swarming plates by transmission electron microscopy (TEM) showed that, in agreement with the bacterial swarming observation, the wild-type strain in darkness presented thicker flagella than in white and red light. On the other hand, the wild-type strain grown under red light presented a fewer number of flagella, while both mutant strains presented a greater number of flagella than the wild type in all lighting conditions. The biological significance of different light treatments and the phytochromes of Psto in the pathogenesis process were determined by studying the ability of this bacterium to develop disease symptoms in tomato plants under different light conditions. Leaves inoculated with wild-type and ΔbphP1 strains in tomato plants exposed to white light showed typical bacterial speck lesions, while ΔbphP2-inoculated leaves produced highly necrotic lesions. In red light, no lesions were observed after the inoculation with all the Psto strains. On the other hand, when inoculated leaves were maintained in dark conditions, the mutant strains showed a higher necrotic lesion than the wild-type strain. This pattern of infection, which depends on the different light conditions, is also observed in epiphytic fitness assays suggesting that the light is an important factor both in Psto pathogenicity and in the plant response (L. Moyano, A. Carrau, S. Petrocelli, I. Kraiselburd, W. Gärtner, E. G. Orellano, unpublished) (Fig. 2).

Ricci et al. studied some Psto strains lacking genes in LOV protein (ΔlovA and ΔlovB), in bacteriophytochrome 1 (ΔBphP1) and in heme oxygenase (ΔHO-A and ΔHO-B) and under different continuous light qualities or in the dark. The authors showed that PstLOV and PstBphP1 downregulate bacterial growth under white light, red (6 μE m⁻² s⁻¹) and far-red (17 μE m⁻² s⁻¹) and in darkness. In addition, under blue light (40 μE m⁻² s⁻¹), the mutant strains are not able to grow in liquid culture suggesting that the simultaneous presence of PstLOV, PstHO and PstBphP1 is essential to guarantee growth under blue light. On the other hand, Ricci et al. inferred that these bacterial photoreceptors downregulated the infectivity in A. thaliana host plants 50.

Another bacterium from the Pseudomonas genus is the bean pathogen Pseudomonas syringae pv. syringae (Pss). This bacterium also presents two bacteriophytochromes, BphP1 and BphP2, and a LOV-HK protein 51. In Pss, swarming motility is modulated by light like Psto and Xcc. Wu et al. demonstrated that in Pss, LOV-HK protein behaves as a positive regulator of swarming motility in response to blue light (5 μE m⁻² s⁻¹) and that BphP1 represses swarming motility in response to red (10 μE m⁻² s⁻¹) and far-red light (0.8 μE m⁻² s⁻¹), both being the red-light-absorbing (Pr) and a far-red-light-absorbing (Pfr) forms responsible for the inhibition phenotype. Interestingly, BphP1 is also involved in the response of this bacterium to blue light, acting as a repressor of swarming motility downstream of LOV-HK signal transduction pathway. On the other hand, BphP2 seems to have no role in bacterial motility through light modulation 51. In this way, in Pss, BphP1 and LOV-HK proteins appear to constitute an integrated signaling network where BphP1 responds to blue light as well as red and far-red light and LOV-HK suppresses BphP1-mediated blue light signaling. Wu et al. suggested that this plant pathogen, sensing distinct aspects of light quality, probably aims at favoring bacterial epiphytic survival and host colonization 51.

Nagrendran et al. studied the influence of light on the interaction between another Pseudomonas strain, Pseudomonas cichorii JBC1 and tomato plants. This bacterium is responsible for the leaf spots on soybean, mid-rib rot on lettuce and leaf spot and stem necrosis in tomato. Like Psto, P. cichorii does not present significant changes in the growth rates when it is grown under different lighting. Tomato seedlings inoculated with P. cichorii grown under white light (120 μE m⁻² s⁻¹) or in the dark show severe necrotic lesions. However, disease symptoms are significantly lower when seedlings are exposed to red (120 μE m⁻² s⁻¹) or green light (120 μE m⁻² s⁻¹). Assays performed with mature tomato plants showed the same phenotype. According to this result, the analysis of the expression of defense-related genes in P. cichorii-infected tomato seedlings grown under red and green light showed that these genes are upregulated in these conditions. Furthermore, lighting did not influence swarming motility and biofilm formation in this bacterium. Nagendran et al. showed that tomato plants grown under red and green light suppress disease development by regulating defense-related gene expression 52.

Light-regulated processes in the soil bacterium Agrobacterium tumefaciens

Finally, light effects on the physiology and virulence of the plant pathogen Agrobacterium tumefaciens have been reported. This bacterium is the causal agent of crown gall disease in over 140 plant species, including several economically important crops. Disease symptoms include the formation of tumor-like swellings (galls) at the crown of the plant, just above soil level as a consequence of the insertion of a small segment of DNA from a bacterial plasmid into the plant cell, which is incorporated at a semi-random location into the plant genome 53. Agrobacterium tumefaciens presents two phytochrome genes and a gene encoding a cryptochrome/photolyase-type protein 54, 55. In A. tumefaciens, white light (150 μE m⁻² s⁻¹) represses the expression of flagellum genes, producing a reduction in bacterial motility. Also, bacteria grown in darkness adhere better to the plant tissue and are more virulent 56. Oberpichler et al. suggested that, in the dark, A. tumefaciens exhibits an increased virulence to optimize the colonization and the infection of host plants when plant defenses are less efficient. Alternatively, the authors suggested that an increased motility in the dark could be part of a mechanism to address the bacteria to the base of the plant, where tumors are preferentially formed 56. The mechanism responsible for these light responses in A. tumefaciens remains unknown and it is still unclear which photoreceptor is responsible for the light regulation of bacterial virulence 56. Both phytochromes of A. tumefaciens, AtBphP1 and AtBphP2, have been extensively characterized. AtBphP1 behaves as a typical BphP; it assembles with biliverdin IXα to generate a Pr ground state that has high histidine autophosphorylation activity. After photoconversion to Pfr by red light absorption, this activity is immediately repressed but slowly restored as Pfr reverts back to Pr in the dark. In contrast, AtBphP2 assembles with biliverdin IXα and first generates a transient Pr-like intermediate, which then transforms nonphotochemically into a stable Pfr form with high histidine kinase activity. After photoconversion to Pr by far-red light absorption, this kinase activity drops but is quickly restored as the unstable Pr form rapidly reverts to Pfr 54, 57. Karniol and Vierstra suggest that A. tumefaciens uses these mutually opposing BphPs to detect its location within the soil strata sensing the red light/far-red light ratio 30.