The Catalase-Peroxidase KatG Is Required for Virulence of Xanthomonas campestris pv. campestris in a Host Plant by Providing Protection against Low Levels of H₂O₂

Role of katG in protection of X. campestris pv. campestris against H₂O₂.

To evaluate the role of the putative katG gene annotated as Xcc1205 in the X. campestris pv. campestris genome (7), a katG knockout mutant was constructed by insertional inactivation using pKNOCK (1). The 300-bp EcoRI fragment of katG was cloned into pKNOCK-Km digested with EcoRI. The recombinant plasmid was subsequently transferred into X. campestris pv. campestris by electroporation, and recipient cells were selected by kanamycin resistance. The X. campestris pv. campestris katG mutant was verified by PCR and Southern blot analysis (data not shown). Next, the physiological role of the gene in oxidative stress protection was evaluated. The extent of resistance to H₂O₂ in the katG mutant and the X. campestris pv. campestris wild type was validated primarily using the plate sensitivity assay (23), in which bacterial cultures were serially diluted and plated on Silva Buddenhagen (SB) medium (17) containing 200 μM H₂O₂. Bacterial colonies capable of growing on H₂O₂-containing medium were counted after a 48-h incubation. The katG mutant was 10⁴-fold less resistant to H₂O₂ than the isogenic wild-type strain (Fig. 1A).

• Open in new tab
• Download powerpoint

FIG. 1.

Phenotypes of X. campestris pv. campestris strains. (A) H₂O₂ resistance levels were determined using the plate sensitivity assay (23). Serial dilutions of exponential-phase cultures of the X. campestris pv. campestris wild type (Xcc), the katG mutant and its complemented strain (katG/pKatG), the katA mutant, and a katA katG double mutant harboring the control plasmid pBBR1MCS4 (pBBR), pKatG, or pKatA were spotted on SB agar plates containing 200 μM H₂O₂, 150 μM tert-butyl hydroperoxide (tBOOH), or 200 μM menadione. Bacterial colonies surviving the treatment were counted after 48 h of incubation at 28°C and expressed as surviving fractions, which were calculated by dividing the number of CFU from plates containing H₂O₂ by the number of CFU from plates without an oxidant. NG indicates no surviving cell growth. (B) H₂O₂ resistance levels were determined using the H₂O₂ killing assay (5). One-milliliter aliquots of exponential-phase cultures of X. campestris pv. campestris (⋄), the katG mutant (○), the katA mutant (▵), and a katA katG double mutant harboring the vector control (▴), pKatA (□), or pKatG (▪) were treated with 0, 5, 10, and 20 mM H₂O₂ for 30 min. Cells surviving the treatment were determined using viable cell counts. The surviving fraction was calculated by dividing the number of CFU from treated samples by the number of CFU from nontreated samples. (C) The virulence of X. campestris pv. campestris was determined on Chinese radish (Raphanus sativus), a compatible host plant, using the leaf-clipping method (9) with some modifications. Overnight cultures of X. campestris pv. campestris and the katG mutant and its complemented strain (katG/pKatG) in AB medium (a) supplemented with 0.1% (wt/vol) Casamino Acids were diluted to an optical density at 600 nm of 1.0 in fresh AB medium. The lesion lengths were measured at 14 days postinoculation. Experiments were repeated three times.

We have previously observed that an X. campestris pv. phaseoli katA mutant is also less resistant to H₂O₂ (5). Nonetheless, the roles of each of these two catalase genes in protecting the bacteria from H₂O₂ have not been evaluated and compared. Therefore, an X. campestris pv. campestris katA knockout mutant was constructed using the pKNOCK system. A 200-bp katA (Xcc3949) DNA fragment was PCR amplified from X. campestris pv. campestris genomic DNA with BT2237 and BT2238 primers (Table 1). The resulting PCR products were cloned into pGEM-T Easy (Promega) before an EcoRI fragment was subcloned into pKNOCK-Gm. The resultant recombinant plasmid (pKNOCK-katA) was electroporated into X. campestris pv. campestris. The gentamicin-resistant transformants were screened for katA inactivation. The katA mutant was verified using both PCR and Southern analysis. A katA katG double mutant was also constructed by electroporating pKNOCK-katA into a katG mutant. Next, the levels of resistance to a low dose (200 μM) of H₂O₂ in katG and katA mutants and the parental strain were determined by the plate sensitivity assay. The results show that the katG mutant was roughly 500-fold less resistant to 200 μM of H₂O₂ than the katA mutant. The katA mutant was 50-fold less resistant to H₂O₂ than a wild-type strain (Fig. 1A). This suggests that katG has a principal role in protecting the X. campestris pv. campestris cells from a low dose of H₂O₂.

Next, we determined the levels of catalase in various kat mutants. The katG mutant had reduced total catalase activity (4.7 ± 0.5 U mg⁻¹ protein) compared to that in an isogenic wild-type strain (6.0 ± 0.4 U mg⁻¹ protein). The katA mutant had a drastically lower catalase activity (1.2 ± 0.3 U mg⁻¹ protein). There are no direct correlations between resistance to 200 μM of H₂O₂ and total catalase activities in various strains; KatA is a major source of catalase activity in exponentially growing X. campestris pv. campestris cells. The katG mutant has 4-fold-higher levels of catalase and yet 500-fold less resistance to 200 μM H₂O₂ than the katA mutant, supporting the idea that KatG catalase-peroxidase has a primary role for bacterial survival under micromolar concentrations of H₂O₂. Generally, degradations of H₂O₂ by KatG catalase-peroxidase and KatA monofunctional catalase are two-step reactions. The first step involves heme oxidation by H₂O₂ to form an oxyferryl species, compound I. The second reaction is a reduction of compound I by H₂O₂ to regenerate the resting state enzyme, water, and oxygen. While the apparent K_m values for H₂O₂ in the heme oxidation of both KatG catalase-peroxidases and KatA monofunctional catalases are somewhat similar (∼ 300 μM), the apparent K_m values of KatGs for the reduction of compound I (2.4 to 4.5 mM) are much lower than those of KatA (38 to 200 mM) (26, 27). This reflects a higher affinity of KatG for H₂O₂ degradation.

Additionally, the katA katG double mutant was hypersensitive to H₂O₂. This mutant was unable to grow on medium containing 200 μM H₂O₂ (Fig. 1A). Nevertheless, the katA katG double mutant had no aerobic growth defects when grown in either liquid or solid enriched medium (SB). This is unlike a Bradyrhizobium japonicum katG mutant that has defects in aerobic growth, suggesting that the gene has a primary role in the detoxification of H₂O₂ generated from aerobic life (22). We have shown in X. campestris pv. phaseoli that an ahpC (alkyl hydroperoxide reductase) mutant has reduced ability to form colonies on an agar plate, but this ability could be rescued by the addition of a peroxide scavenger, 0.1% pyruvate, to the medium (19). In Xanthomonas, AhpC has a primary role in scavenging physiologically generated H₂O₂ analogous to that of Escherichia coli AhpC (25).

If catalatic activity is crucial for detoxification of high concentrations of H₂O₂, one would anticipate that a katA mutant is less resistant to treatment with a high concentration of H₂O₂ than a katG mutant. We performed additional experiments using the H₂O₂ killing assay (5). In this procedure, bacterial cultures are subjected to lethal doses of H₂O₂ (5, 10, and 20 mM) for 30 min. Cells surviving the treatments were determined by viable cell count. As illustrated in Fig. 1B, compared to the X. campestris pv. campestris wild type, the katA mutant was 10³-fold less resistant to 20 mM H₂O₂ killing. The katG mutant was 500-fold more resistant to the same concentration of H₂O₂ than the katA mutant and only 5-fold less resistant than a wild-type strain. This supports the role of KatA as a major catalase responsible for protecting X. campestris pv. campestris against high levels of H₂O₂, whereas katG has only a minor role in the process (Fig. 1B). Additionally, the levels of resistance to high concentrations of H₂O₂ show a correlation with total catalase activities of the cells. As expected, the double mutant, which produced no detectable level of catalase activity, was the most sensitive to killing by a high concentration of H₂O₂ (Fig. 1B). These data support the hypothesis that the catalatic activity of KatA as well as that of KatG is responsible for the protection of Xanthomonas from killing by millimolar concentrations of H₂O₂.

We extended the investigation by assessing the ability of high levels of expression of either katA or katG from an expression vector to complement the H₂O₂-hypersensitive phenotype of the katA katG double mutant. The double mutant harboring pKatA or pKatG (pBBR1MCS4 expression vector [11] containing X. campestris pv. campestris katA or katG [full-length]) produced catalase activities of 195 ± 23 and 73 ± 8 U mg⁻¹ protein, respectively. Although the katA katG double mutant harboring pKatA had more than twofold-higher total catalase activity, it failed to fully complement the H₂O₂-hypersensitive phenotype, as determined by the plate sensitivity assay using 200 μM H₂O₂ (Fig. 1A), but the strain was highly resistant to the high-concentration H₂O₂ killing treatment (Fig. 1B). On the other hand, a high-level expression of katG fully restored the H₂O₂-hypersensitive phenotype of the katA katG double mutant determined using both the 200 μM H₂O₂ plate sensitivity assay (Fig. 1A) and millimolar H₂O₂ killing treatments (Fig. 1B). The results further support the role of KatG in protection against micromolar concentrations of H₂O₂. In addition, the resistance to millimolar concentrations of H₂O₂ shows correlation with total catalase activity.

Taken together, the results of H₂O₂ resistance tests and the catalase levels indicate a departure from a current paradigm for the roles of different catalases. In X. campestris, both katG and katA have primary roles in protecting the bacteria against different levels of H₂O₂. The basis of the mechanism is the enzymes’ different biochemical properties (bifunctional peroxidase/catalase versus monofunctional catalase) and not the activities of each enzyme.

The katG mutant has altered levels of resistance to other oxidants.

The katG mutant showed a 10-fold increase in resistance to tert-butyl hydroperoxide compared with the wild-type strain, and the phenotype could be complemented by pKatG (Fig. 1A). Inactivation of X. campestris pv. phaseoli katA has been shown to produce a compensatory increase in ahpC expression along with enhanced bacterial tolerance to organic hydroperoxides (4). To test whether the level of ahpC expression is altered, an end-point reverse transcription (RT)-PCR was performed with RNA prepared from the X. campestris pv. campestris wild type, the katG mutant, and a katG-complemented strain by using ahpC-specific primers (BT2684 and BT2685 [Table 1]). The level of ahpC transcripts in the katG mutant was higher than the level in either the X. campestris pv. campestris wild type or the katG-complemented strain (Fig. 2A). Thus, this compensatory increase in ahpC expression accounted for, at least in part, the enhanced resistance to organic hydroperoxide observed in the katG mutant. However, the precise mechanism responsible for this resistance is not yet clear. Likewise, inactivation of ahpC leads to compensatory upregulation of katG in X. campestris pv. phaseoli (19).

• Open in new tab
• Download powerpoint

FIG. 2.

Expression analysis of ahpC and katG. (A) The levels of ahpC transcripts in X. campestris pv. campestris (Xcc) and the katG mutant and its complemented strain (katG/pKatG) were measured using end-point RT-PCR with BT2684 and BT2685 primers and total RNA (5 μg) as the template. The RT reaction mixture was incubated at 25°C for 10 min and then at 42°C for 1 h. The cDNA products were PCR amplified under the following conditions: 95°C (for 30 s), 62°C (for 30 s), and 72°C (for 30 s) for 25 cycles. The amplified DNA products were separated on a 1.8% agarose gel. (B) katG expression in the X. campestris pv. campestris wild type and an oxyR mutant was analyzed by end-point RT-PCR with katG-specific primers (BT2239 and BT2240). Total RNAs (5 μg) extracted from exponentially growing cells either uninduced (UN) or induced with 100 μM H₂O₂, menadione (MD), or cumene hydroperoxide (CHP) were added to the RT reaction as described for panel A. The PCR was performed under the following conditions: 95°C (for 30 s), 58°C (for 30 s), and 72°C (for 30 s) for 30 cycles.

We also determined the resistance levels of a superoxide generator, menadione, in the X. campestris pv. campestris wild type and the katG mutant by using the plate sensitivity assay with 200 μM menadione. The results in Fig. 1A illustrate that the katG mutant was 100-fold less resistant to menadione than the wild-type strain. Furthermore, the phenotype could be complemented by pKatG. The superoxide anions formed are dismutated to yield H₂O₂ and other reactive oxygen species. Efficient degradation of H₂O₂ in part contributes to protection against menadione toxicity. Thus, impaired H₂O₂ detoxification in the katG mutant is likely responsible for its reduced menadione resistance phenotype.

The X. campestris pv. campestris katG mutant is avirulent.

X. campestris pv. campestris is a causative agent of black rot disease in cruciferous crops. Oxidative stress is an important component of the plant defense response against microbes (12). Hence, we determined the effects of katG inactivation on the virulence of the bacteria on a host plant. The X. campestris pv. campestris wild type, the katG mutant, and the katG-complemented strain (katG/pKatG) were inoculated into Chinese radish (Raphanus sativus) leaves by using the leaf-clipping method (9). The results in Fig. 1C reveal that the katG mutant is avirulent on the radish, as shown by the lack of detectable lesions in all tests, while the X. campestris pv. campestris wild type and the katG-complemented strain (katG/pKatG) produced similarly sized lesions (10.7 ± 2.7 mm for X. campestris pv. campestris and 9.8 ± 2.4 mm for the katG/pKatG strain), and no significant differences in the length of lesions (P > 0.05 by paired t test) could be detected. The results suggest that katG is required for full virulence of X. campestris pv. campestris on the Chinese radish host plant. A study of the sugar beet plant defense response reveals that the levels of H₂O₂ in the oxidative burst of the phase I response peaked at 2 mM in elicited plant leaves (3). The phase II burst occurs at approximately 2 h postinoculation and peaks at 4 mM H₂O₂. These levels of H₂O₂ would have lethal effects on bacterial pathogens, including Xanthomonas. We speculate that the avirulent phenotype of the katG mutant is due in part to a reduction in the ability of the bacteria to cope with plant-generated H₂O₂. Our speculation is supported by the in vitro experiments that show hypersensitivity of the katG mutant to H₂O₂ (Fig. 1A and B). In addition, we found that the X. campestris pv. campestris katA mutant showed an avirulent phenotype on the Chinese radish host, indicating that monofunctional catalase, which contributes to the protection of millimolar levels of H₂O₂, also plays a crucial role in Xanthomonas pathogenicity (S. Mongkolsuk, S. Buranajitpakorn, and P. Vattanaviboon, unpublished data).

katG expression is regulated by OxyR.

The expression profile of katG in X. campestris pv. campestris was investigated using end-point RT-PCR. The experiments were performed using total RNAs extracted from untreated X. campestris pv. campestris cultures and cultures treated with various oxidants and katG-specific primers (BT2239 and BT2240 [Table 1]). Representative results demonstrate that the levels of katG transcripts were markedly increased in cells treated with a superoxide generator, menadione (100 μM), and to a lesser extent in those treated with organic hydroperoxides (100 μM tert-butyl or cumene hydroperoxides) or H₂O₂ (100 μM) (Fig. 2B). The katG expression profiles in response to oxidants resemble typical OxyR-dependent gene expression in X. campestris (5, 14, 17). The RT-PCR experiments were repeated in an X. campestris pv. campestris oxyR knockout mutant that was constructed using pKNOCK-Gm containing a 210-bp oxyR (Xcc0832) fragment amplified with BT1413 and BT1414 primers (Table 1). The oxidant's induction of katG expression was abolished in the oxyR mutant (Fig. 2B). This indicates that OxyR regulates the peroxide-inducible expression of katG and that the gene is a member of the OxyR regulon, as in several bacteria (6, 16, 20).

OxyR commonly regulates transcription of a target gene through binding to specific motif sequences located in close proximity to the −35 promoter region. Primer extension was performed to map the 5′ end of katG mRNA and to localize the promoter by using a labeled BT2652 primer (Table 1) and total RNA extracted from uninduced and menadione-induced X. campestris pv. campestris cultures. The transcription start site (+1) based on the primer extension product was mapped to the A located 21 nucleotides upstream of the ATG codon (Fig. 3A). The putative −35 and −10 elements were identified as TTCCAC and GCAGCT, respectively, and were separated by a suboptimal distance of 20 nucleotides. Both motifs share low sequence identity to the proposed consensus σ⁷⁰ binding sequences for X. campestris promoters that consist of the −35 element, TTGTNN, and the −10 element, T/AATNAA/T (10). The poor promoter sequence and structure of katG are a likely reason for a relatively low expression level of the gene. A putative OxyR binding site could be identified immediately upstream of the −35 element (Fig. 3A). This proposed binding site poorly matches (6 of 16) the consensus sequence of the E. coli OxyR binding site (ATAGN₇CTATN₇ATAGN₇CTAT) (28). This is quite surprising because we have shown that the putative oxyR binding sites in the katA and ahpC promoters in X. campestris pv. phaseoli highly match (10 of 16) the E. coli consensus sequence (5, 14). Nonetheless, in several microorganisms, the OxyR binding sequences of the target promoters are diverse and differ from the consensus sequence (20, 30).

• Open in new tab
• Download powerpoint

FIG. 3.

Analysis of the katG promoter. (A) Primer extension was carried out as previously described (24). Reverse transcription was performed using the ³²P-labeled primer BT2652 and RNA extracted from the X. campestris pv. campestris wild type and the oxyR mutant either uninduced (UN) or induced with 100 μM menadione (MD). The extension products were separated on a 6% acrylamide-7 M urea sequencing gel. C, T, A, and G represent the DNA sequence ladder prepared using a PCR sequencing kit with labeled pUC/M13 forward primer and pGEM-3Zf (+) as the template (Applied Biosystems). Numbers to the left indicate fragment lengths. The arrowhead indicates the transcription start site (+1). Putative −35 and −10 elements are underlined, and the ribosome binding site is in italic font. The proposed OxyR binding site sequence is in boldface and correspondingly aligned with the E. coli consensus sequence for OxyR binding that is shown above the sequence line. The conserved bases are indicated by asterisks. The bold ATG represents the initiation codon. RBS indicates the putative ribosome binding site. (B) The gel mobility shift assay was done as described previously (8). Purified OxyR at the indicated concentrations was mixed with a ³²P-labeled 366-bp DNA fragment spanning the putative katG promoter that was prepared from PCR amplification using BT2637 and BT2652 primers in 25-μl binding reaction mixtures. The binding complexes were separated on a nondenaturing polyacrylamide gel. F and B indicate free and bound probes, respectively. CP represents the binding complexes that were treated with a cold probe. BSA indicates that bovine serum albumin was added to the binding reaction instead of purified OxyR.

We next tested the binding of X. campestris OxyR to the katG promoter fragment by using a gel mobility shift assay. Purified X. campestris pv. phaseoli OxyR (14), which shares 99% homology with X. campestris pv. campestris OxyR, was mixed with radioactively labeled katG promoter DNA fragments in the binding buffer. The results clearly illustrate that purified OxyR bound to the katG promoter fragment (Fig. 3B). The binding showed high specificity, as shown by competition experiments; only cold katG promoter DNA fragments could compete with the labeled probe in the binding reaction. The results support a notion that OxyR binds in the vicinity of the katG promoter and regulates its expression.

Conclusions.

X. campestris pv. campestris has katA, katG, katE, and ahpC genes that are involved in the degradation of H₂O₂. A major puzzle in the overall bacterial peroxide stress response is whether these peroxide-scavenging genes are redundant or whether they serve disparate roles in the process. katE has no protective roles during the exponential phase of growth but is involved in protecting the bacteria from H₂O₂ in the stationary phase (29). The remaining three genes are regulated by OxyR and are presumably activated by similar concentrations of H₂O₂. Hence, the physiological roles of these genes could not be distinguished at the transcription activation level. Induction of three distinct types of peroxide-degrading enzymes enables the bacteria to detoxify H₂O₂ at a wide range of concentrations as well as other organic peroxides. AhpC is associated with the detoxification of low-concentration H₂O₂ generated from normal aerobic metabolism and organic hydroperoxides (25). At intermediate levels, KatG is better suited for detoxification of H₂O₂ at high micromolar levels. KatA, being the most abundant σ⁷⁰ and also having a high apparent K_m , is responsible for protection against millimolar concentrations of H₂O₂.