Involvement of bacterial TonB-dependent signaling in the generation of an oligogalacturonide damage-associated molecular pattern from plant cell walls exposed to Xanthomonas campestris pv. campestris pectate lyases

Abstract

Background: Efficient perception of attacking pathogens is essential for plants. Plant defense is evoked by molecules termed elicitors. Endogenous elicitors or damage-associated molecular patterns (DAMPs) originate from plant materials upon injury or pathogen activity. While there are comparably well-characterized examples for DAMPs, often oligogalacturonides (OGAs), generated by the activity of fungal pathogens, endogenous elicitors evoked by bacterial pathogens have been rarely described. In particular, the signal perception and transduction processes involved in DAMP generation are poorly characterized.

Results: A mutant strain of the phytopathogenic bacterium Xanthomonas campestris pv. campestris deficient in exbD2, which encodes a component of its unusual elaborate TonB system, had impaired pectate lyase activity and caused no visible symptoms for defense on the non-host plant pepper (Capsicum annuum). A co-incubation of X. campestris pv. campestris with isolated cell wall material from C. annuum led to the release of compounds which induced an oxidative burst in cell suspension cultures of the non-host plant. Lipopolysaccharides and proteins were ruled out as elicitors by polymyxin B and heat treatment, respectively. After hydrolysis with trifluoroacetic acid and subsequent HPAE chromatography, the elicitor preparation contained galacturonic acid, the monosaccharide constituent of pectate. OGAs were isolated from this crude elicitor preparation by HPAEC and tested for their biological activity. While small OGAs were unable to induce an oxidative burst, the elicitor activity in cell suspension cultures of the non-host plants tobacco and pepper increased with the degree of polymerization (DP). Maximal elicitor activity was observed for DPs exceeding 8. In contrast to the X. campestris pv. campestris wild type B100, the exbD2 mutant was unable to generate elicitor activity from plant cell wall material or from pectin.

Conclusions: To our knowledge, this is the second report on a DAMP generated by bacterial features. The generation of the OGA elicitor is embedded in a complex exchange of signals within the framework of the plant-microbe interaction of C. annuum and X. campestris pv. campestris. The bacterial TonB-system is essential for the substrate-induced generation of extracellular pectate lyase activity. This is the first demonstration that a TonB-system is involved in bacterial trans-envelope signaling in the context of a pathogenic interaction with a plant.

Figure 1

Genomic organization of the TonB-related genes in X. campestris pv. campestris B100. (A) A circular genome plot indicates the locations of the TonB-related genes on the chromosome. The core of the TonB system is encoded by the genes tonB, exbB and exbD. In X. campestris pv. campestris B100 multiple isoforms of these genes were identified. Their genomic locations on the circular chromosome are indicated. So far, this multiplicity was only known for tonB genes in Pseudomonas[68] and for the exbD genes in Flavobacterium psychrophilum, where two paralogous genes were found in tandem in a cluster combined with tonB and exbB[64] close to the chromosomal origin of replication (B). Size and direction of transcription is illustrated by arrows for this gene cluster. Genes that were predicted with convincing evidence are symbolized by shaded arrows, while an open arrow indicates a putative protein-coding sequence (CDS) that was predicted with less confidence. Now a third copy of exbD was found downstream of exbD2, separated from exbD2 only by a hypothetical gene for which nor functionality neither expression could be indicated. Further copies of tonB and the genes exbB-exbD were found at different chromosomal positions. To facilitate discriminating the individual genes, unique numbers were added to their names.

Figure 2

Test for pectate lyase activity in TonB-related mutants of X. campestris pv. campestris.X. campestris pv. campestris wild-type strain B100 and mutants derived from it with disrupted genes coding for core components of the TonB system were grown for two days on M9 minimal medium supplemented with pectate and FeSO₄. The positions of the inocula are indicated by dashed circles. Staining with Ruthenium Red unveiled halos encircling the inocula of the wild-type and a control strain that indicate activity of extracellular pectate lyases [64], while no halos were visible when the genes tonB1, exbB1, exbD1, and exbD2 were disrupted. The mutant strain B100-6.01 [64], carrying an ΩKm(cat) insertion in the non-coding region between tonB and exbB, was tested as a positive contro.

Figure 3

Complementation of an X. campestris pv. campestris exbD2 mutant by a constitutively expressed pglI gene from X. campestris pv. campestris 8004. When compared to the X. campestris pv. campestris wild-type strain B100, it becomes obvious that the mutant strain defective in exbD2, B100-11.03, which had been demonstrated before to induce no symptoms like necrotic lesions [66], could be functionally complemented with a constitutively expressed pglI gene on plasmid pHGW267 that was integrated into the chromosome. (A) The complemented mutant strain regained its pectate lyase activity, although not to the full extent of the wild-type strain. (B) This correlates well with the reconstituted but attenuated hypersensitive response that this complemented mutant evoked on C. annuum.

Figure 4

Oxidative burst reactions in heterologous N. tabacum cell suspension cultures after elicitation with supernatants of X. campestris pv. campestris co-incubated with plant cell wall material. The production of hydrogen peroxide was quantified by means of an H₂O₂-dependent chemiluminescence reaction (A). For each measurement, 200 μl of the respective supernatants were added to the cell cultures. The hydrogen peroxide formation was monitored at different time intervals upon the addition of supernatants of C. annuum cell wall material (✶), supernatants of X. campestris pv. campestris cultures (▲), supernatants of X. campestris pv. campestris cultures co-incubated with C. annuum cell wall material (●), and for a negative control of 200 μl water (♦). There was a clear oxidative burst upon the addition of a supernatant of X. campestris pv. campestris co-incubated with cell wall material, but an almost similar explicit reaction when a supernatant of X. campestris pv. campestris was added that had not been co-incubated with cell wall material. (B) Supernatants of X. campestris pv. campestris cultures were treated with polymyxin B agarose to remove LPS. Then the effect of the purified supernatants on N. tabacum cell suspension cultures was analyzed. The formation of H₂O₂ was monitored upon the addition of supernatants of X. campestris pv. campestris cultures (▲), supernatants of X. campestris pv. campestris cultures co-incubated with C. annuum cell wall material (●), supernatants of X. campestris pv. campestris cultures purified from LPS (■), supernatants of X. campestris pv. campestris cultures co-incubated with C. annuum cell wall material and purified from LPS (✶), and after adding 200 μl water as a negative control (♦). Removing the LPS reduced the response to X. campestris pv. campestris supernatant to the level of the water control. In contrast to this, the removal of LPS reduced the amplitude of the cell culture response to X. campestris pv. campestris co-incubated with cell wall material, but this supernatant still evoked a clear oxidative burst reaction.

Figure 5

Hydrogen peroxide formation in C. annuum cell suspension cultures upon elicitation with supernatant of an X. campestris pv. campestris exbD2 mutant co-incubated with cell wall material. To assess the role of the exbD2 gene in provoking defense reactions in non-host plants, cultures of the X. campestris pv. campestris mutant strain B100-11.03 were co-incubated with cell wall material from C. annuum. Then the formation of H₂O₂ was monitored in cell suspension cultures of C. annuum upon the addition either supernatants of X. campestris pv. campestris wild-type cultures (●), supernatants of X. campestris pv. campestris cultures affected in exbD2 that were co-incubated with C. annuum cell wall material (♦), invertase as a positive control (■), or C. annuum cell wall material employed as negative control (✶). The mutated bacterial mutant strain deficient in exbD2 could not evoke an oxidative burst reaction.

Figure 6

Effect of the co-incubation of X. campestris pv. campestris with plant cell wall material on the composition of the dissolved monosaccharides. The identity and relative amounts of the monosaccharides in the supernatant of X. campestris pv. campestris co-incubated with cell wall material of C. annuum was determined by HPAEC. The sugars were separated and identified using an isocratic elution with10 mM sodium hydroxide and amperometric detection on a CarboPac® PA-100 column, a set up that allows detecting the dissolved neutral sugars. The results were compared to the supernatant of an X. campestris pv. campestris culture that had had no contact to plant cell wall material, and to analogously treated cell wall material that had not been incubated with bacteria. The supernatants of plant cell wall material (A) and the X. campestris pv. campestris culture (B), which were analyzed as controls, were both mainly composed of glucose (Glc), galactose (Gal), and rhamnose (Rha). When plant cell wall material and X. campestris pv. campestris culture were co-incubated (C), the amounts of rhamnose and galactose increased dramatically, reverting the original relative abundances. In addition, small amounts of mannose (Man) became detectable.

Figure 7

HPAEC characterization of the elicitor-active compound. A sodium acetate gradient ranging at 0.1 M NaOH from 0.01 M to 1 M sodium acetate with a plateau of 10 min. at a concentration of 0.7 M facilitated the identification of oligosaccharides on a CarboPac® PA-100 column with pulsed amperometric and UV-detection. Supernatants of C. annuum cell wall material (A) and an X. campestris pv. campestris culture (B) displayed no oligosaccharide signals. However, when C. annuum cell wall material was co-incubated with an X. campestris pv. campestris culture (C), characteristic peaks were detected that eluted between 10 min and 20 min. and that indicated the formation of oligosaccharides. A pectate standard of OGAs generated by digesting commercially available pectin with pectate lyase was analyzed as a control (D). The characteristic oligosaccharide peaks of both runs (C and D) were eluted at similar retention times. When the pectate standard was mixed with co-incubation supernatant, the HPAEC analysis indicated perfect overlapping of the congruent oligosaccharide peaks (E). Hence it was plausible to identify the oligosaccharides from the co-incubation of C. annuum cell wall material and X. campestris pv. campestris culture as OGAs.

Figure 8

MALDI-TOF MS of oligosaccharides released from C. annuum cell walls by co-incubation with X. campestris pv. campestris. Cell walls of C. annuum and bacteria were co-incubated over night and the cell-free supernatant was desalted and lyophilized. This material was applied to MALDI-TOF MS using the negative-ion mode. A characteristic ladder of negatively charged ions was obtained. Mass differences correspond to that of OGAs of different degrees of polymerization (DP). Ions that correspond to DP 7 to 12 are indicated.

Figure 9

Isolation of oligogalacturonides (OGAs) with varying degree of polymerization. As the chromatogram of C. annuum cell wall material co-incubated with an X. campestris pv. campestris culture was identical to OGAs derived from pectin digested with a pectate lyase, the products of the co-incubation were assumed to be OGAs, too. The activity of the X. campestris pv. campestris culture supernatant had obviously generated a diverse set of OGAs varying by their degree of polymerization (DP), with a minimal DP of 2, see main text. To allow a further characterization of the OGAs, eluted fractions representing the different individual OGAs were isolated by a sodium acetate gradient, ranging from 0.01 M to 1 M, 0.1 M NaOH with a plateau of 10 min. at a concentration of 0.7 M on a semi-preparative CarboPac® PA-1 column.

Figure 10

Oxidative burst reactions in heterologous N. tabacum cell suspension cultures after elicitation with isolated OGAs. To functionally characterize OGAs differing in their DPs they were checked for their capacity to evoke oxidative burst reactions in cell-suspension cultures of the non-host plant N. tabacum. Samples of the OGAs were added to the cell suspension cultures to a final concentration of 5 μg/ml. The amount of H₂O₂ produced upon the addition of the OGAs was monitored as described before. The addition of water used as negative control (♦) had no effect, and OGAs with a DP of 2 (■), a DP of 3 (●), a DP of 4 (▲), and a DP of 6 (◊) had only minimal effects on the N. tabacum suspension cells. The response to OGAs with DPs of 5 (✶), 7 (□), and 8 (∆) was slightly stronger but still small. But for OGAs whereof the DP exceeded 8 (○), a clear oxidative burst reaction was observed. This indicated the largest OGA fraction as elicitor of the non-host plant defense against X. campestris pv. campestris.

Figure 11

Oxidative burst reaction in homologous C. annuum suspension cell cultures after elicitation with OGAs of a DP exceeding 8. A fraction of isolated OGAs, which had a DP of at least 8, was able to elicit a strong oxidative burst reaction in heterologous N. tabacum suspension cell cultures (Figure 10). Now this OGA fraction was tested in homologous C. annuum suspension cell cultures. Samples were added to the C. annuum culture to a final concentration of 5 mg/ml (○). A negative control contained only water (♦). Once more this OGA fraction evoked a strong oxidative burst, similar to the reaction in N. tabacum. These observations show that OGAs with a DP of at least 8 that were generated by an X. campestris pv. campestris culture from co-incubated C. annuum cell wall material are a powerful endogenous elicitor.

Figure 12

Schematic overview on the interactions of X. campestris pv. campestris and C. annuum analyzed in this work. A major plant cell wall component is pectate, a polygalacturonide (PGA). Pectate is perceived by X. campestris pv. campestris by means of the TonB system. ExbD2, which is not required for ferric iron uptake, is essential for this process. This induces extracellular pectate lyase activity, resulting in the generation of OGAs. Extracellular OGAs consisting of at least 8 galacturonate residues are recognized by C. annuum as a DAMP, resulting in the initiation of defensive measures like an oxidative burst reaction. The presence of a PRR similar to WAK1 is supposed for C. annuum. WAK1 has been identified recently in A. thaliana as a receptor that specifically perceives OGAs [23].