Abstract
In healthy plants, the innate immune system contributes to maintenance of microbiota homoeostasis, while disease can be associated with microbiome perturbation or dysbiosis, and enrichment of opportunistic plant pathogens like Xanthomonas. It is currently unclear whether the microbiota change occurs independently of the opportunistic pathogens or is caused by the latter. Here we tested if protein export through the type-2 secretion system (T2SS) by Xanthomonas causes microbiome dysbiosis in Arabidopsis thaliana in immunocompromised plants. We found that Xanthomonas strains secrete a cocktail of plant cell wall-degrading enzymes that promote Xanthomonas growth during infection. Disease severity and leaf tissue degradation were increased in A. thaliana mutants lacking the NADPH oxidase RBOHD. Experiments with gnotobiotic plants, synthetic bacterial communities and wild-type or T2SS-mutant Xanthomonas revealed that virulence and leaf microbiome composition are controlled by the T2SS. Overall, a compromised immune system in plants can enrich opportunistic pathogens, which damage leaf tissues and ultimately cause microbiome dysbiosis by facilitating growth of specific commensal bacteria.
Similar content being viewed by others
Main
Host-associated microbial communities, collectively referred to as microbiota, promote development, growth and adaptation to abiotic and biotic stress in healthy host organisms. Bacteria are abundant members in the microbiota and assemble into taxonomically structured communities in animals and plants1,2,3.
Under certain circumstances, the relationship between the host and its microbiota can become unbalanced, resulting in an alternative state of the microbial community termed dysbiosis, which is commonly associated with disease and with an alteration in the composition or function of the microbiome4,5. The host immune system plays a central role in maintaining and controlling microbiota homoeostasis to prevent dysbiosis4,6. In addition, opportunistic pathogens are particularly relevant in dysbiosis as they are normally harmless for the host but are equipped with potential virulence functions and, under conducive conditions, eventually cause context-dependent diseases. In mammals, opportunistic pathogens present in the gut or on the skin have been associated with disease in hosts that have a compromised immune system and have a reduced microbiota diversity4,7,8. Therefore, dysbiosis has underlying contributions both from individual species with pathogenic potential and from the microbiota.
Dysbiosis can also occur in plant leaf microbiota9,10. A reverse genetic screen in Arabidopsis thaliana mutants with defects in the immune system revealed that rbohD knockout plants, among others, harbour an altered phyllosphere microbiota and develop disease9. In this case, two Xanthomonas strains were identified as opportunistic pathogens in rbohD plants and as the driver of plant disease after inoculation with a bacterial synthetic community (SynCom) of more than 200 strains that contained these opportunistic pathogens9. The two Xanthomonas strains, Leaf131 and Leaf148, are part of the representative At-LSPHERE strain collection1 and were recently placed into distinct phylogenetic clades, that is Xanthomonas hortorum and Xanthomonas dyei, respectively11. Both strains lack a type-3 secretion system, a typical virulence factor of bona fide pathogens, which might render them non-virulent on A. thaliana Col-0 wild type. Opportunistic Xanthomonas in plants have been reported previously to cause soft rot in wounded plant tissue due to their pectolytic activity12,13.
In plants, the NADPH oxidase RBOHD produces apoplastic reactive oxygen species (ROS) and is involved in several pathways related to growth, development and stress response14,15,16. Moreover, RBOHD is an important component of the plant immune system17. Plants recognize microorganisms due to microbe- or danger-associated molecular patterns or microbial effector proteins that lead to activation of RBOHD, which is a convergence point of pattern-triggered immunity and effector-triggered immunity signalling pathways18. RBOHD-produced ROS also function in cell wall polymer crosslinking during pathogen-induced lignification19,20. Apart from plants, other multi-cellular organisms possess NADPH oxidases, including fungi, where they serve both defence and differentiation signalling21, and in mammals14,22, they are involved in gut epithelial immune responses and prevent intestinal dysbiosis23,24.
In this study, we dissect the contribution of opportunistic Xanthomonas strains, their context-dependent virulence and host genotype to the bacterial community composition in the phyllosphere of A. thaliana mutants defective in RBOHD. We used a SynCom approach, which has emerged as a decisive tool to study the processes and interactions shaping the microbiota and affecting the host9,25,26,27,28,29, and both targeted and random bacterial mutagenesis. Our results link plant immunity to dysbiosis by establishing a causal relationship between a plant protein (RBOHD) and a bacterial trait (enzyme secretion via T2SS) within a rather complex microbiome.
Dysbiosis caused by opportunistic Xanthomonas in rbohD plants
A. thaliana plants with defective RBOHD, but not wild-type plants, show impaired growth and disease when inoculated with a synthetic community and exhibited a dysbiotic microbiota. The rbohD phenotype can be remediated by removing the Xanthomonas Leaf131 strain from a 137-member microbiota community9. To determine whether the opportunistic pathogen not only drives plant disease but also alters the microbiota composition in rbohD plants, we inoculated microbiota-free A. thaliana seedlings with a SynCom of 137 strains that did or did not include Xanthomonas Leaf131 and analysed the community composition on Col-0 wild type, rbohD knockout and the complementation line rbohD/RBOHD by 16S ribosomal RNA (rRNA) amplicon sequencing.
As an indicator for monitoring overall community changes, we used effect size to quantify how much of the total variance in the microbiota is explained by the plant genotype. As expected, we observed that the microbiota composition in rbohD plants when compared with Col-0 was significantly altered when Xanthomonas Leaf131 was included in the microbiota, that is, SynCom-137+Leaf131 with an effect size of 12.5% (P = 0.0001). In contrast, the community composition did not significantly change when Leaf131 was omitted from the SynCom-137 (effect size 2.8%, P = 0.71) (Fig. 1a). Consistent with this, the difference in community composition of SynCom-137 in rbohD plants was observed when Xanthomonas Leaf131 was included, but not in the absence of the opportunistic pathogen, as indicated by a principal component analysis (PCA) (Extended Data Fig. 1a). Analysis of the effect of addition of Xanthomonas Leaf131 to the SynCom on the overall community composition for each genotype confirmed the rbohD-specific impact (Fig. 1 and Source Data). By analysing the changes in relative abundance of each strain in the SynCom-137, we found that specific strains were enriched in rbohD compared with Col-0, resulting in the characteristic microbiota shift in diseased rbohD plants as observed previously9. In addition to Xanthomonas Leaf131, we found that the Gammaproteobacteria Pseudomonas Leaf58, Leaf127 and Leaf434, the Alphaproteobacteria Sphingobium Leaf26 and Brevundimonas Leaf168, the Bacteroides Pedobacter Leaf41, as well as the Actinobacterium Sanguibacter Leaf3 were enriched in their relative abundance (Fig. 1b and Extended Data Fig. 1c). None of the changes in the relative abundance of these strains could be observed in rbohD plants in the absence of Xanthomonas Leaf131, which is also a Gammaproteobacterium.
a, Composition of synthetic bacterial communities SynCom-137 + Xanthomonas Leaf131 or SynCom-137 in rbohD or rbohD/RBOHD plants was compared with Col-0 wild-type plants. Effect size represents percentage of total variance explained by genotype (shown by dot size and absolute value) and statistical significance is expressed with P values determined by PERMANOVA (Benjamini–Hochberg adjusted, n = 16). Number of differentially abundant strains (as shown in b) is shown by dot colour. b, Heatmap shows subset of strains in SynCom-137 with significant log2 fold changes (log2FC, P < 0.05) in rbohD or rbohD/RBOHD compared with Col-0 wild-type plants in the presence (+) or absence (−) of Xanthomonas Leaf131. Black rectangles show significant changes, P < 0.05 (n = 16, two-sided Wald test, Benjamini–Hochberg adjusted). Complete heatmap of all strains in SynCom-137 is shown in Extended Data Fig. 1c. c, Fresh weight of aboveground plant tissue of Col-0, rbohD and rbohD/RBOHD mock inoculated, with SynCom-137 or SynCom-137 + Xanthomonas Leaf131. Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range (n = 16, two-sided Mann–Whitney U test, P values indicated above box plots). Corresponding plant phenotypes are shown in Extended Data Fig. 1d. d, CFU counts of Pseudomonas Leaf434 per gram plant fresh weight after inoculation of germ-free Col-0, rbohD and rbohD/RBOHD plants with Pseudomonas Leaf434 as single inoculation or in binary inoculation with Xanthomonas Leaf131 Tn7::Gm-lux. Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range (n = 12, two-sided Mann–Whitney U test, P values indicated above box plots).
As the SynCom-137 did not show significant differences in community composition in rbohD compared with the control Col-0 without Xanthomonas Leaf131 (Fig. 1a,b and Extended Data Fig. 1a,b), we conclude that rbohD does not affect the microbiota per se, but rather indirectly via Xanthomonas Leaf131. Consistently, only rbohD knockout plants showed disease symptoms and a reduced average plant fresh weight after inoculation with SynCom137+Leaf131, but not Col-0 or rbohD/RBOHD (Fig. 1c and Extended Data Fig. 1d).
To exemplarily validate that certain members of the microbiota benefit from the presence of Xanthomonas Leaf131 on rbohD plants but not on Col-0 wild-type plants, we selected a commensal strain, Pseudomonas Leaf434, that was enriched on rbohD plants on the basis of our data from the SynCom experiment (Fig. 1b), and assessed its absolute abundance in a binary inoculation experiment together with Xanthomonas Leaf131. Substantiating the results of the SynCom experiment, Pseudomonas Leaf434 showed higher plant colonization levels only in rbohD plants when inoculated together with Xanthomonas Leaf131 compared with single inoculation or in control Col-0 and rbohD/RBOHD plants (Fig. 1d).
Overall, our data show that the presence of the opportunistic pathogen Xanthomonas Leaf131 leads to dysbiosis and an enrichment, possibly through the promotion of growth, of specific microbiota members in rbohD plants.
Plant tissue degradation by opportunistic Xanthomonas
When examining possible virulence mechanisms, we found that Xanthomonas Leaf131 and also Leaf148, previously identified as opportunistic pathogens9, degrade leaf tissue. We therefore set up a quantitative A. thaliana assay to assess tissue degradation using leaf discs (Fig. 2). Both Xanthomonas strains degraded the tissue, which we quantified using pixel brightness. We observed that tissue degradation was markedly more severe in leaf discs of rbohD plants compared with Col-0 plants (Fig. 2a,b), corroborating the stronger virulence phenotype of these Xanthomonas strains in rbohD plants9.
a, Leaf discs of Col-0 and rbohD plants (6 weeks old) were mock inoculated (10 mM MgCl2) or inoculated with Xanthomonas Leaf131 or Leaf148 (OD of 0.02) and incubated for 20 h. b, Time-course measurement and quantification of leaf disc brightness (arbitrary unit, AU) from experiment described in a. Statistical differences between Col-0 and rbohD at varying timepoints are indicated above box plots, with P value on the right or left of horizontal line indicating comparison (two-sided Mann–Whitney U test, n = 8). Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range. c, Leaf disc of 5-week-old rbohD plants mock (10 mM MgCl2) inoculated or with Xanthomonas Leaf131 or Leaf148 (OD of 0.02) and incubated for 48 h. Scale bar, 1 mm.
Leaf tissue degradation progressed gradually over time starting at the edges of the leaf discs (Fig. 2b and Extended Data Fig. 2a). After complete degradation of leaf tissue, the leaf disc was translucent and eventually lost its cellular cohesion and fragmented after mechanical impact (Fig. 2c and Extended Data Fig. 2b). In contrast to the effective degradation of leaf discs from rbohD plants, those from Col-0 plants showed reduced and patchy degradation even after 48 h (Extended Data Fig. 2c). We tested other plant genotypes impaired in pattern-triggered immunity signalling upstream of RBOHD, such as hyper-susceptible mutants lacking cell surface localized receptors (for example, Flagellin Sensitive 2 (FLS2)) or mutants of co-receptors (for example, BRI1-Associated Receptor Kinase 1 (BAK1))30,31. We found that the triple co-receptor mutant bak1/bkk1/cerk1 (bbc) but not the triple receptor mutant fls2/efr/cerk1 was susceptible to leaf disc degradation similar to rbohD (Extended Data Fig. 3a, b). Consistently, bbc plants showed disease symptoms and reduced growth after inoculation with Xanthomonas Leaf131 and Leaf148 (Extended Data Fig. 3c). Besides plant genotype, plant age influenced leaf disc degradation, with 5-week-old Col-0 plants being more susceptible compared with 6-week-old plants (Supplementary Fig. 1), suggesting that multiple plant factors affect the phenotype. While we found that intact leaves remained visually unaffected upon exposure to Xanthomonas Leaf131 and Leaf148 within 2 days of observation, wounded leaves showed signs of degradation over the same time period, as expected due to more accessible tissue (Extended Data Fig. 2d). Xanthomonas Leaf131 caused disease and stunted plant growth in germ-free rbohD not only upon inoculation of 10-day-old seedlings9, but also resulted in disease symptoms and reduced growth in older rbohD plants after spray inoculation (Supplementary Fig. 2a,b). Despite rbohD-dependent disease symptoms, bacterial colonization was not significantly different between genotypes Col-0 and rbohD (Supplementary Fig. 2c,d). Spray inoculation of 5-week-old microbiota-free rbohD plants with Xanthomonas Leaf131 Tn7::Gm-lux led to disease symptoms 2 days after infection and co-localized with bacterial colonization based on luminescence (Supplementary Fig. 2e).
In general, our data indicate that opportunistic Xanthomonas spp. act as commensals in Col-0 plants and reveal their pathogenic potential in immunocompromised mutant plants where they elicit strong disease symptoms, in particular in the absence of a microbiota.
Secretion of cell wall-degrading enzymes via T2SS Xps
Leaf tissue degradation by Xanthomonas as a proxy for a virulence phenotype was observed by live bacteria but also by cell-free supernatants of liquid cultures (Fig. 3a), indicating that the phenotype is mediated by secreted factors. Consistent with this finding, the secretion of plant cell wall-degrading enzymes (CWDE) by the T2SS is a known virulence function of other Xanthomonas species32,33. Xanthomonas Leaf131 and Leaf148 each possess two T2SS gene clusters, designated xps and xcs by homology search. To test whether the degradation activity is dependent on the T2SS, we deleted the core genes of the two T2SS operons (Fig. 3b) and generated double mutants in Xanthomonas Leaf131 and Leaf148. In both strains, the xps mutant and the double knockout xpsxcs did not show tissue degradation, in contrast to the xcs mutants, which were still able to degrade leaf discs (Fig. 3c,d). This indicates that the T2SS Xps is required for leaf degradation by Xanthomonas, which is in line with studies of other Xanthomonas bacteria reporting the importance of xps for virulence34,35,36.
a, Leaf discs of Col-0 and rbohD plants (5 weeks old) were mock treated (0.5× LB) or treated with cell-free supernatant (0.22 µm filter sterilized) of Xanthomonas Leaf131 or Leaf148 liquid cultures and incubated for 48 h. b, Genomic region of the T2SS operons xps and xcs in Xanthomonas Leaf131 and Leaf148. Letters indicate gene names and black line shows region of gene deletion. c,d, Leaf disc brightness was measured 24 h after inoculation with mock solution or with Xanthomonas wild-type or mutant strains of Leaf131 (c) or Leaf148 (d). Leaf discs were generated from Col-0 or rbohD plants (6 weeks old). Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range. Significant differences were calculated with ANOVA and two-sided Tukey’s honest significant difference post hoc test (n = 8, letters indicate significance groups, α = 0.05). e, Agar plates containing either skimmed milk, PGA, CMC, azo-xyloglucan or RBB-Xylan. Drops of 4 µl Xanthomonas Leaf131 wild-type or mutant suspension were pipetted onto agar plate. Pictures were taken 24 h after incubation at 22 °C. Quantification of halo diameter is shown in Supplementary Fig. 4. f, Leaf discs were treated with 0.22 µm filter-sterilized supernatant of liquid cultures from Xanthomonas Leaf131 or Leaf148 wild-type and xpsxcs mutants or mock solution (0.5× LB). Leaf discs were incubated for 48 h at 22 °C. AU, arbitrary unit; SN, supernatant; WT, wild type.
In addition, we deleted the hrpX and hrpG genes in Xanthomonas Leaf131, which encode master regulators of virulence factors including T2SS-secreted enzymes in various Xanthomonas pathogens37,38,39,40,41. However, the hrpXhrpG double knockout mutant still showed leaf degradation activity (Supplementary Fig. 3) suggesting that the production or secretion of the degradative enzymes is not, or not exclusively, controlled by HrpX or HrpG in Xanthomonas Leaf131. In line with the absence of a phenotype for the hrpXhrpG knockout, transcriptomic studies in Xanthomonas campestris pv. campestris found that the T2SS genes and some (but not all) T2SS substrates were regulated by HrpG and that only a small subset of all genes regulated in planta were part of the HrpG regulon42.
To validate the finding that Xps is the primary T2SS involved in the secretion of plant polymer-degrading enzymes, we conducted agar plate assays using substrates for CWDE. We tested Xanthomonas Leaf131 and found that the strain was able to degrade milk powder, polygalacturonic acid (PGA), carboxymethyl cellulose (CMC), xyloglucan and xylan, suggesting secretion of proteases, pectate lyases, glucanases and xylanases, respectively, as shown by halos that formed around the bacterial colonies after incubation indicating substrate degradation (Fig. 3e). Notably, the T2SS mutants xps and xpsxcs showed reduced or delayed polymer degradation, unlike the xcs mutant strain (Fig. 3e). However, xps and xpsxcs mutants still resulted in a small halo indicating substrate degradation on xyloglucan plates after 24 h of incubation. At later timepoints, halos were observable on all plates (Supplementary Fig. 4), which might be due to cell lysis or alternative secretion mechanisms, such as outer membrane vesicles43.
Next, we tested the leaf degradation activity of supernatants from Xanthomonas grown in liquid culture. In contrast to the wild type, cell-free supernatant of Xanthomonas Leaf131 and Leaf148 T2SS mutant xpsxcs did not cause rbohD leaf disc degradation (Fig. 3f). Sodium dodecyl sulfate–polyacrylamide gel electrophoresis analysis of supernatants revealed the presence of protein bands (35–55 kDa) in the wild type that were absent in the xpsxcs mutant (Extended Data Fig. 4a). Identification of the corresponding protein fractions by liquid chromatography tandem mass spectometry (LC–MS/MS) showed T2SS-dependent secretion (Supplementary Table 1) and several candidate proteins predicted to harbour a secretion signal peptide and a function potentially involved in plant interaction (Extended Data Fig. 4d). This included genes annotated to encode an endoglucanase (ASF73_13775), a serine protease (ASF73_18370), two pectate lyases (ASF73_04230 and ASF73_20170) and a lysyl endopeptidase (ASF73_20190), which is in line with the activities observed in the agar plate assays. We generated in-frame deletion knockout strains in Xanthomonas Leaf131 and tested the mutant strains for their leaf tissue degradation activity. For ASF73_20170 and ASF73_20190, which are located in proximity in the genome, we deleted the whole cluster (Extended Data Fig. 4b). Degradation was not affected in these mutant strains compared with wild type in rbohD leaf discs (Extended Data Fig. 4c). The mutant strains lacking a serine protease (ASF73_18370), a pectate lyase (ASF73_04230) or the gene cluster mutant ASF73_20170-20190 showed a difference in degradation in Col-0 (Extended Data Fig. 4c); however, this difference was not observed consistently, as leaf degradation in Col-0 is, in general, less pronounced, slower and more variable compared with rbohD (Fig. 2b and Extended Data Fig. 2c).
Overall, our data suggest that Xanthomonas secretes a cocktail of potential CWDE responsible for leaf degradation via the T2SS Xps.
Involvement of T2SS Xps in virulence during plant infection
To test the importance of the T2SS for virulence in planta, we inoculated Col-0 and rbohD plants with Xanthomonas Leaf131 wild type and the T2SS mutants. Plant health was monitored by assessing disease symptoms and measuring plant fresh weight 3 weeks after inoculation using an established gnotobiotic growth system9. The infection experiment revealed that the virulence of Xanthomonas Leaf131 was dependent on the presence of the T2SS Xps, while Xcs did not contribute to virulence (Fig. 4a, b), corroborating the results of the leaf degradation assay (Fig. 3c). While the Xanthomonas Leaf131 xps mutant was non-virulent in Col-0 plants, as indicated by similar plant weight compared with mock inoculation, the xps mutant showed residual virulence in rbohD plants (Fig. 4b), which suggests the presence of additional T2SS-independent virulence factors or alternative secretion pathways of leaf-degrading enzymes. Moreover, the overall colonization level of these disease-attenuated T2SS mutants xps and xpsxcs was significantly reduced by up to two orders of magnitude compared with Xanthomonas Leaf131 wild type in Col-0 and rbohD plants (Fig. 4c and Source Data for statistical results), highlighting the importance of the T2SS Xps for bacterial growth during plant colonization.
a, Phenotype of 5-week-old Col-0 plants (blue arrow) and rbohD plants (green arrow) mock inoculated or with Xanthomonas Leaf131 wild type (WT) or T2SS mutants xps, xcs and xpsxcs. Scale bars, 1 cm. b, Measurement of fresh weight from plants shown in a. c, CFU counts of Xanthomonas Leaf131 per gram plant fresh weight from samples in b. Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range. Significant differences in b (n = 20) and c (n = 12) were calculated with ANOVA and two-sided Tukey’s honest significant difference post hoc test (letters indicate significance groups, α = 0.05). Log reduction of bacterial abundance shown in Source Data Fig. 4.
In addition to using a targeted approach by mutating the T2SS and genes for proteins that we found excreted under in vitro conditions, we used an untargeted approach by setting up a forward genetic screen in Xanthomonas Leaf131. The screening procedure was effective as we identified transposon (Tn) mutants that we had already confirmed as being important, that is the T2SS xps, and by identifying multiple independent transposon insertions in the same gene, suggesting high coverage (Supplementary Table 2). We identified 16 Tn mutant candidates with reduced or delayed leaf tissue degradation activity (Supplementary Note). To validate the results of the Tn screen, we generated and tested mutants in candidate genes by assessing leaf degradation phenotypes (Supplementary Fig. 5 and Supplementary Table 2) and virulence in planta (Supplementary Fig. 6). The selected targets were dsbB (ASF73_01480) encoding a thiol-disulfide interchange protein; gtf, encoding a predicted glycosyltransferase (ASF73_08425), located upstream of a flagellum gene cluster; and a gene (ASF73_19940) encoding for a hypothetical protein with glucanase/lectin domain. In addition, we deleted a gene cluster including the operon encoding the identified glucanase, a TonB-dependent receptor, a pectin methylesterase and a pectate lyase, as well as the TonB-dependent receptor, which is named iroN (ASF73_19920) and has been identified in the transposon screen (Supplementary Fig. 7).
We examined the ability of the gene deletion strains to degrade plant tissue and their impact on plant fresh weight during rbohD infection. With the exception of the gtf mutant strain, all other mutants showed phenotypes. Leaf degradation by the dsbB mutant was abolished in rbohD, similar to the xps mutant (Fig. 5a). In accordance with the impaired leaf degradation, the dsbB mutant was also reduced in virulence as indicated by higher fresh weight of dsbB colonized rbohD plants (Fig. 5b) and had a lower colonization level compared with the wild type, similar to xps and xpsxcs mutants (Fig. 5d). DsbB is involved in post-translational modification of secreted enzymes, including proteins of the T2SS, which therefore explains the similar phenotypes between the mutants44. The glucanase and iroN-glucanase mutants showed reduced or delayed leaf degradation in rbohD (Fig. 5a) and cell-free supernatant of liquid culture from the respective mutants revealed reduced degradation activities in rbohD leaf discs (Fig. 5c). This finding suggests that the glucanase might be directly involved in polymer degradation. All mutants with reduced degradation activity were also attenuated in overall virulence as indicated by higher fresh weight of rbohD plants (Fig. 5b), while glucanase and iroN-glucanase mutants maintained wild-type colonization levels (Fig. 5d). The gene encoding glucanase, which is absent in the glucanase and iroN-glucanase mutants (Supplementary Fig. 7), encodes a protein belonging to the glucanase superfamily (pfam13385) and contains a signal peptide for secretion. This glucanase contributed to leaf degradation and virulence in planta, which was notable given the functional redundancy common to tissue-degrading enzymes.
a, Leaf discs of 5-week-old rbohD plants were mock (10 mM MgCl2) inoculated or inoculated with Xanthomonas Leaf131 wild type (WT) or gene deletion mutants (OD of 0.02) and incubated for 24 h. b, Fresh weight of aboveground plant tissue of 5-week-old gnotobiotic rbohD plants, either mock inoculated or inoculated with Xanthomonas Leaf131 wild type or gene deletion mutants. c, Leaf discs of 5-week-old rbohD plants were mock treated (0.5× LB) or treated with cell-free supernatant (0.22 µm filter sterilized) of liquid cultures from Xanthomonas Leaf131 wild type and gene deletion mutants. Leaf discs were incubated for 24 h at 22 °C. Black circles, rectangles and squares indicate data from three bacterial cultures. d, CFU counts of Xanthomonas Leaf131 per gram plant fresh weight from samples in b. Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range. Significant difference in a (n = 8), b (n = 20), c (n = 24) and d (n = 12) of gene deletion mutants compared with Leaf131 wild type was determined by two-sided Mann–Whitney U test, and P values are indicated above box plots.
Xanthomonas T2SS drives community shifts in rbohD plants
The secretion of extracellular enzymes is a crucial virulence factor of opportunistic Xanthomonas bacteria for plant colonization (Figs. 4 and 5), and our SynCom experiments revealed that plant disease and the microbiota shift in rbohD depend on the presence of Xanthomonas Leaf131 (Fig. 1). To investigate whether both phenotypes are causally linked and dependent on T2SS-related virulence, we inoculated plants with the SynCom-137 and added either Xanthomonas Leaf131 wild-type or attenuated mutant strains. We determined the microbiota profiles by 16S rRNA amplicon sequencing and compared the community composition of the SynCom-137 containing Xanthomonas Leaf131 wild type or mutants with the SynCom-137 without Leaf131 as a control. Quantification of the impact of each Xanthomonas strain on the community composition revealed a significant effect only of Xanthomonas Leaf131 wild type (effect size 10.3%, P = 0.0002), as observed previously (Fig. 1a), but not the attenuated mutants in rbohD plants (Fig. 6a). Consistently, only the presence of virulent Xanthomonas Leaf131, but not mutants with defective T2SS or dsbB knockout, increased the relative abundance of other commensals (Fig. 6b). The addition of Xanthomonas Leaf131 wild type to the SynCom-137 showed the characteristic shift in specific strains (Fig. 6b and Extended Data Fig. 5a), as observed previously (Fig. 1b). In contrast, inoculation of rbohD plants with SynCom-137 containing the Xanthomonas Leaf131 mutants xps, xpsxcs or dsbB resulted in a similar overall community composition as the SynCom-137 alone in rbohD and in Col-0 plants, as indicated by few changes of individual strains in their relative abundance (Fig. 6b and Extended Data Fig. 5b) and by overlapping clusters of the different conditions in a PCA (Extended Data Fig. 5c). In addition, the T2SS mutants showed reduced relative abundance, and dsbB was hardly detected by 16S rRNA amplicon sequencing (Fig. 6c), which underlines the importance of these features for the competitiveness of Xanthomonas in the context of a bacterial community, similar to the plant inoculations with only Xanthomonas Leaf131 (Fig. 4c).
a, Composition of synthetic bacterial community SynCom-137 containing Xanthomonas Leaf131 wild type or mutants xps, xpsxcs and dsbB was compared with SynCom-137 alone in Col-0 and rbohD plants. Effect size represents percentage of total variance explained by genotype (shown by dot size and absolute value) and statistical significance is expressed with P values determined by PERMANOVA (Benjamini–Hochberg adjusted, n = 16). Number of differentially abundant strains (as shown in b) is represented by dot colour. b, Heatmap shows subset of strains of SynCom-137 with significant log2 fold changes (log2FC, P < 0.05) in rbohD plants inoculated either with only SynCom-137 or with SynCom-137 containing Xanthomonas Leaf131 wild type or the mutants xps, xpsxcs and dsbB. Black rectangles show significant changes, P < 0.05 (n = 16, two-sided Wald test, Benjamini–Hochberg adjusted). The heatmap of all strains in SynCom-137 is shown in Extended Data Fig. 5b. c, Relative abundance of Xanthomonas Leaf131 wild type or the mutants xps, xpsxcs and dsbB within SynCom-137 in Col-0 and rbohD plants. Ratios below violin plots represent frequency of samples where Xanthomonas Leaf131 was not detected. Violin plots show the median with upper and lower quartiles (n = 16, two-sided Mann–Whitney U test; P values are indicated above violin plots). d, CFU counts of Pseudomonas Leaf434 per gram plant fresh weight after inoculation of germ-free Col-0 and rbohD plants with Pseudomonas Leaf434 as single inoculation (−) or as binary inoculation with either Xanthomonas Leaf131 wild type or the mutants xps and xpsxcs. Box plots show the median with upper and lower quartiles and whiskers present 1.5× interquartile range. Significant differences were calculated with ANOVA and two-sided Tukey’s honest significant difference post hoc test (n = 12, letters indicate significance groups, α = 0.05).
Furthermore, we examined in a binary strain inoculation experiment the colonization level of the commensal Pseudomonas Leaf434 in response to attenuated Xanthomonas Leaf131. Strikingly, the increase in the commensal Pseudomonas Leaf434 observed during co-inoculation with virulent Xanthomonas Leaf131 was significantly reduced when the T2SS mutants xps and xpsxcs were paired with the Pseudomonas strain (Fig. 6d). This finding supports the conclusion that commensals are enriched in their abundance in plants due to the virulence of Xanthomonas Leaf131, which is particularly pronounced in immunocompromised rbohD plants.
In summary, our results indicate that specific microbiota members benefit indirectly from Xanthomonas Leaf131 due to T2SS-dependent virulence causing plant disease, rather than from the presence of Xanthomonas Leaf131 or the knockout of RBOHD per se.
Discussion
Dysbiosis is considered a condition with distorted microbiota with various compositional states, but associated to disease and often characterized by weakening of host control over microbial growth45,46. However, the concept of dysbiosis is controversial because the causal relationships are often unclear, that is whether the observed changes in the microbiota are caused by the host genotype or by infection with a pathogen, and whether a shift in the microbial composition is the consequence of host disease or promoting disease4,47. Several studies have reported dysbiosis in the phyllosphere of plants infected with a pathogen48,49,50,51,52; however, it remains to be shown whether the pathogen invaded the microbial community as external agent or was part of the microbiota that was initially kept under control. Indeed, environmental conditions and protective microbiota members determine the virulence of pathogens28,30,53,54,55. Recently, experimental studies described that a functional immune system is required to maintain microbiota homoeostasis and prevent dysbiosis9,10,56.
Our previous finding that A. thaliana rbohD mutants display a microbiota shift and the identification of Xanthomonas Leaf131 and Leaf148 as opportunistic pathogens9 gave us the opportunity to disentangle the causation of dysbiosis in a representative leaf microbiome context. We found that conditional pathogenicity of the microbiota member Xanthomonas Leaf131 in immunocompromised rbohD mutants is governed by the T2SS and results in a dysbiotic microbial community characterized by increased abundance of Xanthomonas Leaf131 and other strains (Figs. 1 and 6). The enriched commensals might benefit from nutrients released from the plant as ‘public good’ and depend on their metabolic capacity that shapes microbiota composition57,58,59.
We found that leaf tissue damage caused by secreted CWDE via the T2SS Xps is a major virulence strategy of Xanthomonas Leaf131 and Leaf148 during infection of A. thaliana. However, it is still unclear what underlies the context-dependent pathogenicity of Xanthomonas Leaf131 and Leaf148 in rbohD plants. Pathogenicity of these opportunistic Xanthomonas strains could be regulated by rbohD-specific cues (for example, nutrients, signalling molecules or absence of ROS) that trigger a behavioural switch in Xanthomonas towards a pathogenic lifestyle. It was recently shown in Xanthomonas citri pv. citri that degradation products from the plant cell wall polymer xyloglucan induce transcription of virulence factors60. In our leaf degradation experiments using supernatants, Xanthomonas produced CWDE during incubation in rich media, suggesting that CWDE production is a constitutive trait and might not be dependent on host signals. A complementary study on Xanthomonas Leaf148 confirmed the T2SS-dependent pathogenicity, and a bacterial transcriptomics experiment indicates that genes encoding the T2SS and CWDE are more highly expressed in rbohD knockout plants61. We identified several T2SS-dependent proteins to be secreted by Xanthomonas Leaf131 and Leaf148 in liquid culture. Notably, some of these have been described in the context of virulence of Xanthomonas pathogens34,36,43,62. We have not observed a reduction in leaf tissue degradation in the gene knockout strains for two of the identified enzymes, an endoglucanase or serine protease, which is, however, not surprising given that a deletion of a single T2SS substrate often does not show a phenotype presumably due to functional redundancy among the secreted proteins34,36.
Context-dependent pathogenicity of opportunistic Xanthomonas strains might rely on plant susceptibility due to altered immune signalling or physical barriers. The plant immune system detects microbial activity and monitors the cell wall integrity63,64. Loss of microbe/danger-associated molecular patterns-induced ROS production by RBOHD results in impaired immune signalling and increased susceptibility to bacterial and fungal pathogens17,65,66 as well as reduced cell wall remodelling and lignification19,20,67. To explain susceptibility to opportunistic Xanthomonas, rbohD plants could mount an insufficient defence response. Many pathogens secrete CWDE to degrade plant polymers at certain stages during the infection process64,68 and, in turn, defects in cell wall composition make plants more susceptible69. As such, rbohD plants might have cell wall defects due to altered polymer crosslinking, which is in accordance with our data showing that tissue degradation activity of cell-free supernatants is higher in rbohD compared with Col-0 leaf discs. In that case, opportunistic Xanthomonas would secrete CWDE that break down a vulnerable (pre-formed) cell wall of rbohD plants. Strikingly, we have identified a single gene (ASF73_19940), which is required for full leaf tissue degradation and virulence in rbohD plants, and encodes a protein annotated with a secretion signal and a glucanase/concanavalin A-like lectin domain, which is potentially involved in carbohydrate processing or adhesion. Importantly, in A. thaliana wild-type plants, both Xanthomonas Leaf131 and Leaf148 protect from the virulent pathogen Pseudomonas syringae pv. syringae DC300053 and the related Xanthomonas WCS2014-23 is enriched in A. thaliana plants and limits infections with Hyaloperonospora arabidopsidis70,71 highlighting that these Xanthomonas can also be advantageous for the host when their pathogenicity is constrained. Mammalian NADPH oxidases produce ROS as a cell-to-cell messenger regulating the intestinal barrier, which is required for microbiota homoeostasis72, and ROS also form a physical barrier, which is thought to keep certain bacteria at distance from the epithelial surface24,73. This draws attention to striking similarities in the molecular mechanisms for host control of microbiota homoeostasis across animal and plant kingdoms.
In conclusion, our study revealed the importance of the T2SS for opportunistic Xanthomonas strains both for their interaction with the plant and for their competitiveness within the microbiota. The conditional pathogenicity of this opportunistic microbiota member depends on the host genotype and impacts both plant health and the microbial community. Our findings establish a causal link between a single plant gene to a specific genus of bacteria that drives a microbiota shift and highlight the crucial role of opportunistic pathogens in dysbiosis.
Methods
Plant growth conditions in soil
A. thaliana wild-type Col-0, bbc30, fls2/efr/cerk131, rbohD knockout mutant17 and complementation line rbohD/pRBOHD::RBOHD-FLAG (rbohD/RBOHD)65 were used in this study.
A. thaliana plants for leaf degradation assays were grown in peat-based potting soil (substrate 1, Klasmann-Deilmann) in a growth chamber (CU-41L4, Percival) under controlled conditions (11 h light cycle, 22 °C, 65% relative humidity, light intensity (photosynthetic active radiation) 200 µmol s−1 cm−2). Seeds were treated with 70% ethanol for 2 min, sown on soil and stratified for 2 days at 4 °C in the dark.
Gnotobiotic plant growth and bacterial inoculation
Gnotobiotic plants were prepared and grown in sterile microboxes filled with calcined clay as described previously9.
For the SynCom, 138 strains were selected on the basis of the At-LSPHERE strain collection (Supplementary Table 3) to have maximal phylogenetic diversity and to distinguish all strains with 100% sequence identity representing amplicon sequence variants (ASVs)9. Xanthomonas Leaf131 was used as single inoculum or mixed into the SynCom-137.
Bacterial growth, mixing of the synthetic community and plant inoculation were done as described before9. Each strain was mixed in equal volume ratio for inoculum mix. Germ-free, 11-day-old seedlings were inoculated with 200 µl bacterial solution. Plants were harvested between 35 and 38 days after germination. Experiments with SynCom, single strain or binary strain inocula were done in the same procedure. Axenic plants in gnotobiotic system were inoculated with buffer only and used as control for contamination by plating plant homogenate to monitor bacterial growth and were included as negative control in 16S rRNA amplicon sequencing. To extract DNA for 16S rRNA amplicon sequencing, the phyllosphere was harvested, weighed and stored at −80 °C.
Spray inoculation was done with sterilized glass sprayer in 24-day-old or 38-day-old gnotobiotic plants with bacterial culture diluted in 10 mM MgCl2 to optical density (OD)600 of 0.2 or 0.001, as indicated in corresponding figure legend.
To determine bacterial colonization levels, the phyllosphere was harvested, weighed and homogenized in 10 mM MgCl2 and a dilution series plated on R2A and methanol (MeOH) agar plates to count colony forming units (CFUs) after 2 days incubation at 28 °C. We excluded completely necrotic or dead plants from CFU count analysis as this would introduce inaccuracies depending on the time passed between plant death and the sampling timepoint. In the binary plant colonization experiments, the dilution series was plated on R2A-MeOH agar plates containing 10 µg ml−1 gentamycin and 25 µg ml−1 chloramphenicol to select for Xanthomonas Leaf131 Tn7::Gm-lux and Pseudomonas Leaf434, respectively. In addition, Xanthomonas Leaf131 and Pseudomonas Leaf434 can be distinguished by yellow and white colony pigmentation, respectively.
Bacterial luminescence of Xanthomonas Leaf131 Tn7::Gm-lux was measured in planta using IVIS spectrum imaging system (Xenogen). Exposure was set to 50 s and emission filter to 500. Radiance values (p−1 s−1 cm−2 sr−1) were extracted and normalized to plant size by adjusting elliptic region of interest for each plant.
16S rRNA amplicon sequencing and analysis
DNA extraction and 16S rRNA amplicon sequencing was done as previously published9, but polymerase chain reaction (PCR) reactions for 16S rRNA amplification and barcoding were not done in technical triplicate here.
16S rRNA amplicon data processing was done as described previously9,26. The ASV table (Supplementary Table 3) of each experiment was processed in R v.3.6.3 as described previously9. To account for varying sequencing depths between samples, the ASV table was log normalized and variance stabilized by DESeq2 v.1.14.1. To examine the effect on individual strains between the test and control conditions, the output of DESeq2 provided log2 fold change values and strains were considered to be differentially abundant according to Wald test implemented in DESeq2. P values were adjusted for multiple testing using the Benjamini–Hochberg method implemented in DESeq2. The differential strain abundances between the test and control conditions were visualized as a heatmap. To assess the overall effect on communities, PCA was performed with the transformed ASV table using the prcomp command. The effect size represents the variance explained by the compared factor and was calculated on Euclidean distances followed by a permutational multivariate analysis of variance (PERMANOVA) to test for statistical significance using the adonis command of the package vegan v.2 v.5-4. To summarize the relative abundance of Xanthomonas Leaf131 in a sample, the relative abundance values were calculated by proportional normalization of each sample by its sequencing depth.
The following R packages were used during analysis and visualization: ape v.5.4 (ref. 74), ggplot2 v.3.3.0 (ref. 75), vegan v.2.5-4 (ref. 76), DESeq2 v.1.14.1 (ref. 77) and ggpubr v.0.3.0 (ref. 78).
Leaf disc degradation assay
Leaf discs of 5- or 6-week-old A. thaliana plants grown in soil were collected using a 4-mm-diameter biopsy puncher (BPP-40F, KAI MEDICAL) and placed with the adaxial side up in a clear flat-bottom 96-well plate (655101, Greiner Bio-One) filled with 90 µl Milli-Q purified water. Xanthomonas were grown on R2A-MeOH agar plates for 2 days at 22 °C; bacterial cells were scraped off, resuspended in 10 mM MgCl2 by vortexing for 2 min and the bacterial solution was adjusted to OD600 of 0.1. Leaf discs were inoculated with 10 µl of bacterial suspension and incubated at 22 °C for up to 48 h in the dark. Digital images were taken at regular intervals under standardized conditions using a black box and a light screen illuminating leaf discs from below to monitor leaf tissue degradation.
Quantification of leaf disc brightness
To quantify leaf tissue degradation, we developed a computational script MatlabR2022a (MathWorks), which recognizes leaf discs in a 96-well plate, measures surface area, brightness of the red channel (in RGB images) and computes a ‘roughness’ parameter.
In short: the script normalizes the brightness of the images using the ‘illumgray’ and ‘chromadapt’ functions implemented in MATLAB. Subsequently, a binary mask is created, separating the area occupied by leaf discs from the rest of the image. Discs that deviate in ‘roundness’ are discarded from the analysis since they are probably broken or folded. The roughness parameter is created using the ‘Sobel’ edge detection function on the isolated discs in the red channel and computing the total number of pixels recognized as edge within each disc. The area of each disc is computed by counting the number of pixels per disc times the pixel size retrieved from an image scaling step. The brightness value represents the mean brightness value of the red channel for each individual leaf disc.
The MatlabR2022a code and user manual is available79.
Transformation of electrocompetent Xanthomonas cells
Electrocompetent Xanthomonas cells were made by an established protocol80. Exponentially growing Xanthomonas cells in 200 ml lysogeny broth (LB) at 28 °C with an OD600 between 0.6 and 1 were cooled on ice for 20 min and kept on ice for the entire procedure. Cells were collected by centrifugation for 15 min at 4,000g at 4 °C and washed three times in chilled sterile 10% glycerol to remove growth medium. After the final washing step, cells were concentrated approximately 100-fold compared with initial volume in 10% glycerol and aliquots frozen at −80 °C.
Electrocompetent Xanthomonas cells were thawed on ice and 50 µl was mixed with 200 ng plasmid. Cells were transformed by electroporation in 1-mm electro-cuvettes applying 1.8 kV electric current. Cells were recovered in LB medium for 2–4 h shaking at 28 °C before plating 100 µl on LB agar plates containing selective antibiotics.
Xanthomonas Leaf131 cells were transformed with pUC18-mini-Tn7T-Gm-lux81 and helper plasmid pTNS3 (ref. 82) for site-specific Tn7 integration of luxCDABE. Transformed cells of Xanthomonas Leaf131 Tn7::Gm-lux were selected on LB agar plates containing 10 µg ml−1 gentamycin.
Bacterial gene knockout strains
Markerless gene deletion in Xanthomonas strains (Supplementary Table 4) were made according to a method based on double homologous recombination using the suicide plasmid pK18mobSacB as vector83. Gene deletion plasmids were designed to result in in-frame deletion of the gene of interest while leaving an open reading frame of three to four amino acid peptide. Briefly, 500 bp of flanking regions upstream and downstream of the gene of interest were amplified by PCR and cloned into pK18mobSacB plasmid. The plasmids were cloned using either classical restriction enzyme digest or Gibson Assembly (oligonucleotides are provided in Supplementary Table 4) in Escherichia coli DH5α. Gene deletion constructs were confirmed by Sanger sequencing.
Electrocompetent Xanthomonas cells were transformed, recovered in LB medium for 2–4 h and transformed cells were selected on LB agar plates containing 50 µg ml−1 kanamycin. Transformed cells were re-streaked on fresh selective LB agar plates and a single colony resuspended in LB medium for 2 h before plating on LB agar plates containing 5% sucrose to select for double cross-over events due to homologous recombination and chromosomal deletion of the gene of interest and the vector backbone. After sucrose selection, individual colonies were tested for sensitivity to kanamycin. Cells were re-streaked to obtain single colonies that were cultured and frozen in 25% glycerol at −80 °C. Genomic deletion was confirmed by PCR using primers outside of flanking regions and Sanger sequencing the PCR product and by the absence of PCR product using primers inside genomic deletion.
Bacterial supernatant of liquid culture
Xanthomonas were grown in triplicates in 100 ml liquid 0.5× LB medium until late exponential growth phase (approximately OD600 of 2) at 28 °C while shaking. Cells were harvested by centrifugation at 4,000g for 15 min and washed twice in 10 mM MgCl2. Bacterial cells were resuspended in 10–20 ml fresh 0.5× LB medium at an OD600 of 3 and incubated for 4 h in flasks at 28 °C while shaking. To obtain the cell-free supernatant, we centrifuged the samples at 4,000g for 15 min to remove bacteria and filter sterilized the supernatant using 0.22-µm filter units (no. 99505, ‘rapid’-Filtermax, TTP) and a vacuum pump. A total of 10 ml of cell-free supernatant was concentrated ten-fold by using Ultrafiltration Units Amicon-15 with a molecular weight cutoff 10 kDa (Merck) and centrifugation at 3,500g at 4 °C for 20–40 min. Cell-free supernatants were directly tested for leaf degradation activity and kept on ice until further processing for protein analysis.
Cell-free supernatant or concentrated supernatant was applied to leaf discs to test for tissue degradation activity. Leaf discs were collected from 5- or 6-week-old plants and floated in 40 µl Milli-Q purified water in a 96-well plate. To each leaf disc, 40 µl supernatant was added. Leaf discs were incubated at 22 °C and photographs taken at regular intervals.
Analysis of protein bands by LC–MS/MS
To test for the secretion of proteins, the concentrated cell-free supernatant of Xanthomonas liquid cultures was obtained as described above. Protein concentration of the concentrated supernatant was determined by the Pierce BCA assay kit (Thermo Fischer Scientific) according to the manufacturer’s instructions. Protein content of supernatant samples were normalized and analysed using sodium dodecyl sulfate–polyacrylamide gel electrophoresis (mPAGE Bis-Tris 8%, Merck) revealing specific protein bands in the supernatant while comparing wild type and the T2SS mutant (xpsxcs) of Xanthomonas Leaf131 and Leaf148. The protein bands of interest were cutout and identified by in-gel digestion and LC–MS/MS analysis as described previously84. Reference genomes of Xanthomonas Leaf131 and Xanthomonas Leaf148 accessed under NCBI:txid1736270 and NCBI:txid1736275, respectively.
Substrate degradation by secreted enzymes in agar plates
Agar plate assays to detect glucanase, xylanase, pectate lyase and polygalacturonase or protease activity were modified after refs. 85,86,87.
Xanthomonas strains were streaked on R2A-MeOH plates and grown at 22 °C for 2 days. Bacterial cells were scraped off and resuspended in 1 ml 10 mM MgCl2 by vortexing for 5 min to disperse cell aggregates. Cell density was adjusted to OD600 of 0.4 and 4 µl of the bacterial suspensions were spotted on R2A agar plates either containing 0.5% sodium CMC (Sigma-Aldrich, C5678), 0.05% Remazol Brilliant Blue-Xylan (RBB-Xylan; Sigma-Aldrich, M5019), 0.1% azo-xyloglucan (Megazyme, S-AZXG) or 0.1% PGA in 1 M sodium phosphate buffer pH 7.0 (Sigma-Aldrich, 81325) or on 1.5% agar plates containing 3% skimmed milk powder (Rapilait), 1% peptone, 0.025% MgSO4 and 0.05% K2HPO4, respectively. The plates were incubated at 22 °C, and photographs were taken at regular intervals.
Glucanase activity can be detected by yellow halos against the red background after staining with 0.1% Congo red (Sigma-Aldrich, C6767) dye solution (solved in 50% ethanol) for 30 min and destaining with 1 M NaCl for 15 min. Pectate lyase or polygalacturonase activity can be detected by light-pink halos against the darker pink background after staining with 0.05% ruthenium red (Sigma-Aldrich, R2751) dye solution (solved in water) for 30 min and destaining with water. Xylanase or protease activity can be detected by a light-blue or clear halo forming around the colonies, respectively.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
Raw data of 16S rRNA amplicon sequencing can be found at the European Nucleotide Archive under accession number PRJEB64618. Source data are provided with this paper.
Code availability
Customized code to analyse data and generate figures can be found at https://github.com/MicrobiologyETHZ/phylloR/releases/tag/v1.1. The code to quantify leaf disc brightness using MatlabR2022a and user manual is available at https://github.com/gaebeleinC/leaf-disc_quantification.
References
Bai, Y. et al. Functional overlap of the Arabidopsis leaf and root microbiota. Nature 528, 364–369 (2015).
Hacquard, S. et al. Microbiota and host nutrition across plant and animal kingdoms. Cell Host Microbe 17, 603–616 (2015).
Lloyd-Price, J. et al. Strains, functions and dynamics in the expanded Human Microbiome Project. Nature 550, 61–66 (2017).
Levy, M., Kolodziejczyk, A. A., Thaiss, C. A. & Elinav, E. Dysbiosis and the immune system. Nat. Rev. Immun. 17, 219–232 (2017).
Petersen, C. & Round, J. L. Defining dysbiosis and its influence on host immunity and disease. Cell. Microbiol. 16, 1024–1033 (2014).
Paasch, B. C. & He, S. Y. Toward understanding microbiota homeostasis in the plant kingdom. PLoS Pathog. 17, e1009472 (2021).
Byrd, A. L., Belkaid, Y. & Segre, J. A. The human skin microbiome. Nat. Rev. Microbiol. 16, 143–155 (2018).
Kamada, N., Seo, S.-U., Chen, G. Y. & Núñez, G. Role of the gut microbiota in immunity and inflammatory disease. Nat. Rev. Immun. 13, 321–335 (2013).
Pfeilmeier, S. et al. The plant NADPH oxidase RBOHD is required for microbiota homeostasis in leaves. Nat. Microbiol. 6, 852–864 (2021).
Chen, T. et al. A plant genetic network for preventing dysbiosis in the phyllosphere. Nature 580, 653–657 (2020).
Agarwal, V., Stubits, R., Nassrullah, Z. & Dillon, M. M. Pangenome insights into the diversification and disease specificity of worldwide Xanthomonas outbreaks. Front. Microbiol. 14, 1213261 (2023).
Liao, C. H. & Wells, J. M. Association of pectolytic strains of Xanthomonas campestris with soft rots of fruits and vegetables at retail markets. Phytopathology 77, 418–422 (1987).
Gitaitis, R. D., Sasser, M. J., Beaver, R. W., Mcinnes, T. B. & Stall, R. E. Pectolytic xanthomonads in mixed infections with Pseudomonas syringae pv syringae, Pseudomonas syringae pv tomato, and Xanthomonas campestris pv vesicatoria in tomato and pepper transplants. Phytopathology 77, 611–615 (1987).
Mittler, R., Zandalinas, S. I., Fichman, Y. & Van Breusegem, F. Reactive oxygen species signalling in plant stress responses. Nat. Rev. Mol. Cell Biol. 23, 663–679 (2022).
Planas-Riverola, A., Markaide, E. & Caño-Delgado, A. I. New role for LRR-receptor kinase in sensing of reactive oxygen species. Trends Plant Sci. 26, 102–104 (2021).
Hu, C.-H. et al. NADPH oxidases: the vital performers and center hubs during plant growth and signaling. Cells 9, 437 (2020).
Torres, M. A., Dangl, J. L. & Jones, J. D. G. Arabidopsis gp91phox homologues AtrbohD and AtrbohF are required for accumulation of reactive oxygen intermediates in the plant defense response. Proc. Natl Acad. Sci. USA 99, 517–522 (2002).
Ngou, B. P. M., Ahn, H.-K., Ding, P. & Jones, J. D. G. Mutual potentiation of plant immunity by cell-surface and intracellular receptors. Nature 592, 110–115 (2021).
Denness, L. et al. Cell wall damage-induced lignin biosynthesis is regulated by a reactive oxygen species- and jasmonic acid-dependent process in Arabidopsis. Plant Physiol. 156, 1364–1374 (2011).
Hamann, T., Bennett, M., Mansfield, J. & Somerville, C. Identification of cell-wall stress as a hexose-dependent and osmosensitive regulator of plant responses. Plant J. 57, 1015–1026 (2009).
Takemoto, D., Tanaka, A. & Scott, B. NADPH oxidases in fungi: diverse roles of reactive oxygen species in fungal cellular differentiation. Fungal Gen. Biol 44, 1065–1076 (2007).
Suzuki, N. et al. Respiratory burst oxidases: the engines of ROS signaling. Curr. Op. Plant Biol. 14, 691–699 (2011).
Grasberger, H. et al. Increased expression of DUOX2 is an epithelial response to mucosal dysbiosis required for immune homeostasis in mouse intestine. Gastroenterology 149, 1849–1859 (2015).
Miller, B. M. et al. Anaerobic respiration of NOX1-derived hydrogen peroxide licenses bacterial growth at the colonic surface. Cell Host Microbe 28, 789–797.e785 (2020).
Vorholt, J. A., Vogel, C., Carlstrom, C. I. & Mueller, D. B. Establishing causality: opportunities of synthetic communities for plant microbiome research. Cell Host Microbe 22, 142–155 (2017).
Carlström, C. I. et al. Synthetic microbiota reveal priority effects and keystone strains in the Arabidopsis phyllosphere. Nat. Ecol. Evo. 3, 1445–1454 (2019).
Schäfer, M., Vogel, C. M., Bortfeld-Miller, M., Mittelviefhaus, M. & Vorholt, J. A. Mapping phyllosphere microbiota interactions in planta to establish genotype–phenotype relationships. Nat. Microbiol. 7, 856–867 (2022).
Durán, P. et al. Microbial interkingdom interactions in roots promote Arabidopsis survival. Cell 175, 973–983.e914 (2018).
Finkel, O. M. et al. A single bacterial genus maintains root growth in a complex microbiome. Nature 587, 103–108 (2020).
Xin, X. F. et al. Bacteria establish an aqueous living space in plants crucial for virulence. Nature 539, 524–529 (2016).
Gimenez-Ibanez, S., Ntoukakis, V. & Rathjen, J. P. The LysM receptor kinase CERK1 mediates bacterial perception in Arabidopsis. Plant Signal. Behav. 4, 539–541 (2009).
Büttner, D. & Bonas, U. Regulation and secretion of Xanthomonas virulence factors. FEMS Microbiol. Rev. 34, 107–133 (2010).
Alvarez-Martinez, C. E. et al. Secrete or perish: the role of secretion systems in Xanthomonas biology. Comp. Struct. Biotechn. J. 19, 279–302 (2021).
Jha, G., Rajeshwari, R. & Sonti, R. V. Bacterial type two secretion system secreted proteins: double-edged swords for plant pathogens. Mol. Plant Microbe Interact. 18, 891–898 (2005).
Potnis, N. et al. Comparative genomics reveals diversity among xanthomonads infecting tomato and pepper. BMC Genom. 12, 146 (2011).
Szczesny, R. et al. Functional characterization of the Xcs and Xps type II secretion systems from the plant pathogenic bacterium Xanthomonas campestris pv vesicatoria. New Phytol. 187, 983–1002 (2010).
Guo, Y., Figueiredo, F., Jones, J. & Wang, N. HrpG and HrpX play global roles in coordinating different virulence traits of Xanthomonas axonopodis pv. citri. Mol. Plant Microbe Interact. 24, 649–661 (2011).
Wengelnik, K., Van den Ackerveken, G. & Bonas, U. HrpG, a key hrp regulatory protein of Xanthomonas campestris pv. vesicatoria is homologous to two-component response regulators. Mol. Plant Microbe Interact. 9, 704–712 (1996).
Noël, L., Thieme, F., Nennstiel, D. & Bonas, U. cDNA-AFLP analysis unravels a genome-wide hrpG-regulon in the plant pathogen Xanthomonas campestris pv. vesicatoria. Mol. Microbiol. 41, 1271–1281 (2001).
Roux, B. et al. Genomics and transcriptomics of Xanthomonas campestris species challenge the concept of core type III effectome. BMC Genom. 16, 975 (2015).
Yamazaki, A., Hirata, H. & Tsuyumu, S. HrpG regulates type II secretory proteins in Xanthomonas axonopodis pv. citri. J. Gen. Plant Pathol. 74, 138–150 (2008).
Luneau, J. S. et al. Xanthomonas transcriptome inside cauliflower hydathodes reveals bacterial virulence strategies and physiological adaptations at early infection stages. Mol. Plant Pathol. 23, 159–174 (2022).
Solé, M. et al. Xanthomonas campestris pv. vesicatoria secretes proteases and xylanases via the Xps Type II secretion system and outer membrane vesicles. J. Bacteriol. 197, 2879–2893 (2015).
Jiang, B.-L. et al. DsbB is required for the pathogenesis process of Xanthomonas campestris pv. campestris. Mol. Plant Microbe Interact. 21, 1036–1045 (2008).
Miller, B. M. & Bäumler, A. J. The habitat filters of microbiota-nourishing immunity. Annu. Rev. Immunol. 39, 1–18 (2021).
Lee, J. Y., Tsolis, R. M. & Baumler, A. J. The microbiome and gut homeostasis. Science 377, eabp9960 (2022).
Olesen, S. W. & Alm, E. J. Dysbiosis is not an answer. Nat. Microbiol. 1, 16228 (2016).
Duran, P. et al. A fungal powdery mildew pathogen induces extensive local and marginal systemic changes in the Arabidopsis thaliana microbiota. Environ. Microbiol. 23, 6292–6308 (2021).
Seybold, H. et al. A fungal pathogen induces systemic susceptibility and systemic shifts in wheat metabolome and microbiome composition. Nat. Commun. 11, 1910 (2020).
Yang, F. et al. Bacterial blight induced shifts in endophytic microbiome of rice leaves and the enrichment of specific bacterial strains with pathogen antagonism. Front. Plant Sci. 11, 963 (2020).
Luo, L. et al. Variations in phyllosphere microbial community along with the development of angular leaf-spot of cucumber. AMB Express 9, 76 (2019).
Runge, P., Ventura, F., Kemen, E. & Stam, R. Distinct phyllosphere microbiome of wild tomato species in central Peru upon dysbiosis. Microb. Ecol. https://doi.org/10.1007/s00248-021-01947-w (2022).
Vogel, C. M., Potthoff, D. B., Schäfer, M., Barandun, N. & Vorholt, J. A. Protective role of the Arabidopsis leaf microbiota against a bacterial pathogen. Nat. Microbiol. 6, 1537–1548 (2021).
Kwak, M.-J. et al. Rhizosphere microbiome structure alters to enable wilt resistance in tomato. Nat. Biotechnol. 36, 1100–1109 (2018).
Cheng, Y. T., Zhang, L. & He, S. Y. Plant-microbe interactions facing environmental challenge. Cell Host Microbe 26, 183–192 (2019).
Wolinska, K. W. et al. Tryptophan metabolism and bacterial commensals prevent fungal dysbiosis in Arabidopsis roots. Proc. Natl Acad. Sci. USA 118, e2111521118 (2021).
Hemmerle, L. et al. Dynamic character displacement among a pair of bacterial phyllosphere commensals in situ. Nat. Commun. 13, 2836 (2022).
Ryback, B., Bortfeld-Miller, M. & Vorholt, J. A. Metabolic adaptation to vitamin auxotrophy by leaf-associated bacteria. ISME J. 16, 2712–2724 (2022).
Schäfer, M. et al. Metabolic interaction models recapitulate leaf microbiota ecology. Science 381, eadf5121 (2023).
Vieira, P. S. et al. Xyloglucan processing machinery in Xanthomonas pathogens and its role in the transcriptional activation of virulence factors. Nat. Commun. 12, 4049 (2021).
Entila, F., Han, X., Mine, A., Schulze-Lefert, P. & Tsuda, K. Commensal lifestyle regulated by a negative feedback loop between Arabidopsis ROS and the bacterial T2SS. Preprint at bioRxiv https://doi.org/10.1101/2023.05.09.539802 (2023).
Laia, M. L. et al. New genes of Xanthomonas citri subsp. citri involved in pathogenesis and adaptation revealed by a transposon-based mutant library. BMC Microbiol. 9, 12 (2009).
Wolf, S. Cell wall signaling in plant development and defense. Annu. Rev. Plant Biol. 73, 323–353 (2022).
Dora, S., Terrett, O. M. & Sánchez-Rodríguez, C. Plant–microbe interactions in the apoplast: communication at the plant cell wall. Plant Cell 34, 1532–1550 (2022).
Kadota, Y. et al. Direct regulation of the NADPH oxidase RBOHD by the PRR-associated kinase BIK1 during plant immunity. Mol. Cell 54, 43–55 (2014).
Mélida, H. et al. Arabinoxylan-oligosaccharides act as damage associated molecular patterns in plants regulating disease resistance. Front. Plant Sci. 11, 1210 (2020).
Liu, C., Yu, H., Voxeur, A., Rao, X. & Dixon, R. A. FERONIA and wall-associated kinases coordinate defense induced by lignin modification in plant cell walls. Sci. Adv. 9, eadf7714 (2023).
Bellincampi, D., Cervone, F. & Lionetti, V. Plant cell wall dynamics and wall-related susceptibility in plant–pathogen interactions. Front. Plant Sci. 5, 228 (2014).
Molina, A. et al. Arabidopsis cell wall composition determines disease resistance specificity and fitness. Proc. Natl Acad. Sci. USA 118, e2010243118 (2021).
Berendsen, R. L. et al. Disease-induced assemblage of a plant-beneficial bacterial consortium. ISME J. 12, 1496–1507 (2018).
Goossens, P. et al. Obligate biotroph downy mildew consistently induces near-identical protective microbiomes in Arabidopsis thaliana. Nat. Microbiol. https://doi.org/10.1038/s41564-023-01502-y (2023).
Aviello, G. & Knaus, U. G. NADPH oxidases and ROS signaling in the gastrointestinal tract. Mucosal Immunol. 11, 1011–1023 (2018).
Chanin, R. B. et al. Epithelial-derived reactive oxygen species enable AppBCX-mediated aerobic respiration of Escherichia coli during intestinal inflammation. Cell Host Microbe 28, 780–788.e785 (2020).
Paradis, E., Claude, J. & Strimmer, K. APE: analyses of phylogenetics and evolution in R language. Bioinformatics 20, 289–290 (2004).
Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer, 2009).
Oksanen, F. et al. vegan: community ecology package. R Project Version 2.5-6 https://cran.r-project.org/web/packages/vegan/index.html (2019).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Kassambara, A. ggpubr: ‘ggplot2’ Based Publication Ready Plots. R Project Version 0.3.0 https://cran.r-project.org/web/packages/ggpubr/index.html (2020).
leaf-disc_quantification. GitHub https://github.com/gaebeleinC/leaf-disc_quantification (2023).
Sun, Q. et al. High-quality mutant libraries of Xanthomonas oryzae pv. oryzae and X. campestris pv. campestris generated by an efficient transposon mutagenesis system. FEMS Microbiol. Lett. 226, 145–150 (2003).
Choi, K.-H. et al. A Tn7-based broad-range bacterial cloning and expression system. Nat. Meth. 2, 443–448 (2005).
Choi, K.-H. et al. Genetic tools for select-agent-compliant manipulation of Burkholderia pseudomallei. Appl. Environ. Microbiol. 74, 1064–1075 (2008).
Marx, C. J. Development of a broad-host-range sacB-based vector for unmarked allelic exchange. BMC Res. Notes 1, 1 (2008).
Hemmerle, L., Ochsner, A. M., Vonderach, T., Hattendorf, B. & Vorholt, J. A. Mass spectrometry-based approaches to study lanthanides and lanthanide-dependent proteins in the phyllosphere. Methods Enzymol. 650, 215–236 (2021).
Sun, Q. H. et al. Type-II secretion pathway structural gene xpsE, xylanase- and cellulase secretion and virulence in Xanthomonas oryzae pv. oryzae. Plant Pathol. 54, 15–21 (2005).
Tayi, L., Maku, R., Patel, H. K. & Sonti, R. V. Action of multiple cell wall-degrading enzymes is required for elicitation of innate immune responses during Xanthomonas oryzae pv. oryzae infection in rice. Mol. Plant Microbe Interact. 29, 599–608 (2016).
Pailin, T., Kang, D. H., Schmidt, K. & Fung, D. Y. C. Detection of extracellular bound proteinase in EPS-producing lactic acid bacteria cultures on skim milk agar. Lett. Appl. Microbiol. 33, 45–49 (2001).
Acknowledgements
The authors thank C. Lefebvre and T. Steward for experimental help, C. Field for bioinformatic support, M. Schäfer for fruitful discussions, and S. Kobel from the Genetic Diversity Centre Zurich for assistance in library preparation and sequencing. The authors acknowledge funding from the German Research Foundation (DECRyPT, no. SPP2125), from the ETH Zurich Foundation (Career Seed Award), and from the NCCR Microbiomes funded by the Swiss National Science Foundation (51NF40_180575).
Funding
Open access funding provided by Swiss Federal Institute of Technology Zurich.
Author information
Authors and Affiliations
Contributions
S.P. and J.A.V. designed the study and supervised the work. A.W. designed and conducted in vitro and in planta experiments with bacterial mutants and assisted in the SynCom experiment with mutant strains, including library preparation for 16S rRNA amplicon sequencing. M.O. established and conducted the forward genetic screen with input from C.M.P. M.B.-M. and P.K. did the bacterial genetic engineering and assisted in experiments. A.K. designed and assisted in binary plant colonization experiment. L.H. did the proteomic experiment of culture supernatants. C.G.G. developed the script for leaf disc brightness measurements. G.C.P. discovered the leaf degradation phenotype and did the plant colonization experiment of immunity mutants. S.W. conducted spray infection experiments. S.P. performed all other experiments and did the final analysis of all the data. S.P. and J.A.V. wrote the paper with input from all co-authors. All authors approved the final version.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Microbiology thanks Gwyn Beattie, Blanca Landa, Neha Potnis and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 Microbiota composition of SynCom-137 and plant phenotype in presence and absence of Xanthomonas Leaf131.
a) Principal component analysis of SynCom-137+Xanthomonas Leaf131 and of b) SynCom-137 in Col-0 (blue), rbohD (green) and rbohD/RBOHD (light blue). Axes show principal components PC1 and PC2 with their explained variance (%). Statistical analysis (PERMANOVA) analysis is represented by effect size shown in Fig. 1Ac) Heatmap shows log2 fold changes (log2FC) of strains in SynCom-137 in rbohD or rbohD/RBOHD compared to Col-0 wild-type plants in the presence (+) or absence (-) of Xanthomonas Leaf131. Black rectangles show significant changes, p-value < 0.05 (n = 16, two-sided Wald test, Benjamini–Hochberg adjusted). Subset of same data is shown in Fig. 1b. d) Plant phenotype of Col-0 (blue arrow), rbohD (green arrow) and rbohD/RBOHD (light blue arrow) mock inoculated or with SynCom-137 or SynCom-137+Xanthomonas Leaf131. Two representative replica growth boxes are shown for each treatment.
Extended Data Fig. 2 Xanthomonas disrupts leaf tissue cohesion and requires plant wound.
a) Time-course of leaf discs from five-week-old rbohD plants inoculated with Xanthomonas Leaf131 (OD=0.02). Scale bar represents 1 mm. b) Leaf discs of rbohD plants in 96-well plate after 48 hours incubation with Xanthomonas Leaf131 were vortexed for two seconds. Scale bar represents 1 mm in left and middle panel, and 0.5 mm in right panel. c) Leaf discs of five-week-old Col-0 plants after 48 hours incubation with Xanthomonas Leaf131. Scale bar represents 1 mm in left and middle panel, and 0.5 mm in right panel. Experiment shown in panel A-C was repeated at least ten times. d) Leaves of five-week-old rbohD plants floating in water with undamaged leaf edge (left panels) or wounded leaf edge (right panels) were mock inoculated (10 mM MgCl2) or with Xanthomonas Leaf131 or Leaf148 (OD=0.02).
Extended Data Fig. 3 Plant genotype influences leaf disc degradation and susceptibility to Xanthomonas.
a) Time-course of leaf discs brightness from six-week-old Col-0, rbohD, fls2/efr/cerk1 (fec) and bak1/bkk1/cerk1 (bbc) plants inoculated with Xanthomonas Leaf131 or b) Leaf148. Statistical differences of leaf disc brightness between plant genotype at varying time points is indicated with p-value above box plots (two-sided Mann–Whitney U-test, n = 4). Box plots show the median with upper and lower quartiles and whiskers present 1.5x interquartile range. c) Plant phenotype of gnotobiotic Col-0, rbohD and bbc plants mock inoculated (blue arrow) or with Xanthomonas Leaf131 (yellow arrow) or Xanthomonas Leaf148 (orange arrow) at 24 days post inoculation.
Extended Data Fig. 4 Proteomic analysis of supernatant from Xanthomonas Leaf131 and Leaf148 liquid culture identified T2SS-specific proteins.
a) Coomassie stained SDS-PAGE of cell-free supernatant from Xanthomonas Leaf131 and Leaf148 wildtypes and xpsxcs mutants liquid cultures (n=3). Black arrows with numbers indicate protein bands excised from gel and analysed by LC-MS/MS with results of different fractions shown in Supplementary Table 2. b) Genomic region encoding T2SS-dependent secreted proteins pectate lyase (ASF73_20170) and lysyl endopeptidase (ASF73_20190). Orange line indicates in-frame deletion of gene cluster. c) Leaf discs of Col-0 or rbohD plants (six weeks old) were mock treated or with Xanthomonas Leaf131 wildtype or mutant strains with gene deletions of ASF73_13775, ASF73_18370, ASF73_04230, ASF73_20170-ASF73_20190. Box plots show the median with upper and lower quartiles and whiskers present 1.5x interquartile range. Sample size, n = 8. d) Table shows candidate genes of T2SS-dependent secreted proteins identified by LC-MS/MS (Supplementary Table 1) and gene identifiers of homologues in other Xanthomonas species as described in the literature.
Extended Data Fig. 5 Microbiota composition in presence of virulent or attenuated Xanthomonas Leaf131.
a) Heatmap shows log2 fold changes (log2FC) of strains in SynCom-137 in rbohD compared to Col-0 wild-type plants in the presence (+) or absence (−) of Xanthomonas Leaf131. b) Heatmap shows log2 fold changes (log2FC) of strains in the presence of either Xanthomonas Leaf131 wildtype or the mutants xps, xpsxcs, or dsbB compared to SynCom-137 without Leaf131. Black rectangles show significant changes, p-value < 0.05 (n = 16, two-sided Wald test, Benjamini–Hochberg adjusted). Subset of same data is shown in Fig. 6a. c) Principal component analysis of community in rbohD plants (upper panel) and Col-0 (lower panel) inoculated only with SynCom-137 or SynCom-137 containing either Xanthomonas Leaf131, xps, xpsxcs, or dsbB. Axes show principal components PC1 and PC2 with their explained variance (%).
Supplementary information
Supplementary Information
Supplementary information.
Supplementary Tables 1–5
Supplementary Table 1. Proteomics of Leaf131 culture supernatant. a, Identification of proteins in fractions of Xanthomonas Leaf131 supernatant from wild type (fractions 1 and 2) or xpsxcs mutant (fractions 4 and 5). b, Identification of proteins in fractions of Xanthomonas Leaf148 supernatant from wild type (fractions 3 and 4). Fractions are protein bands excised from SDS–PAGE (Extended Data Fig. 4a). Selection for gene knockout highlighted in orange. c, Table presents selected protein candidates for gene knockout in Xanthomonas Leaf131. Supplementary Table 2. Transposon mutagenesis screen in Xanthomonas Leaf131. a, Overview of transposon screen and identified candidate genes. b, Selection of validated candidate genes. Supplementary Table 3. SynCom strains and microbiota composition data. a, At-LSPHERE strains used in SynCom-137 and Xanthomonas Leaf131 and Leaf148. b,c, ASV count table for drop-out experiment (b) and corresponding metadata (c). d,e, ASV count table for drop-in experiment (d) and corresponding metadata (e). Supplementary Table 4. Knockout strains and oligonucleotides used in this study. Supplementary Table 5. Statistical analysis of data shown in Supplementary Fig. 1. Results of two-way ANOVA of data shown in Supplementary Fig. 1.
Source data
Source Data Fig. 1
Data tables for graphs and statistical analysis.
Source Data Fig. 2
Data tables for graphs and statistical analysis.
Source Data Fig. 3
Data tables for graphs and statistical analysis.
Source Data Fig. 4
Data tables for graphs and statistical analysis.
Source Data Fig. 5
Data tables for graphs and statistical analysis.
Source Data Fig. 6
Data tables for graphs and statistical analysis.
Source Data Extended Data Fig. 1
Data tables for graphs and statistical analysis.
Source Data Extended Data Fig. 3
Data tables for graphs and statistical analysis.
Source Data Extended Data Fig. 4
Data tables for graphs and statistical analysis.
Source Data Extended Data Fig. 5
Data tables for graphs and statistical analysis.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Pfeilmeier, S., Werz, A., Ote, M. et al. Leaf microbiome dysbiosis triggered by T2SS-dependent enzyme secretion from opportunistic Xanthomonas pathogens. Nat Microbiol 9, 136–149 (2024). https://doi.org/10.1038/s41564-023-01555-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41564-023-01555-z
