Microbiology (2014), 160, 458–466

DOI 10.1099/mic.0.072470-0

Copper chloride induces antioxidant gene expression but reduces ability to mediate H2O2 toxicity in Xanthomonas campestris Phornphan Sornchuer,13 Poommaree Namchaiw,23 Jarunee Kerdwong,3 Nisanart Charoenlap,2 Skorn Mongkolsuk2,3,4 and Paiboon Vattanaviboon1,2,4 Correspondence Paiboon Vattanaviboon [email protected]

1

Program in Applied Biological Sciences: Environmental Health, Chulabhorn Graduate Institute, Bangkok 10210, Thailand

2

Laboratory of Biotechnology, Chulabhorn Research Institute, Bangkok 10210, Thailand

3

Department of Biotechnology, Faculty of Science, Mahidol University, Bangkok 10400, Thailand

4

Center of Excellence on Environmental Health and Toxicology, Bangkok, Thailand

Received 15 August 2013 Accepted 2 December 2013

Copper (Cu)-based biocides are currently used as control measures for both fungal and bacterial diseases in agricultural fields. In this communication, we show that exposure of the bacterial plant pathogen Xanthomonas campestris to nonlethal concentrations of Cu2+ ions (75 mM) enhanced expression of genes in OxyR, OhrR and IscR regulons. High levels of catalase, Ohr peroxidase and superoxide dismutase diminished Cu2+-induced gene expression, suggesting that the production of hydrogen peroxide (H2O2) and organic hydroperoxides is responsible for Cu2+induced gene expression. Despite high expression of antioxidant genes, the CuCl2-treated cells were more susceptible to H2O2 killing treatment than the uninduced cells. This phenotype arose from lowered catalase activity in the CuCl2-pretreated cells. Thus, exposure to a nonlethal dose of Cu2+ renders X. campestris vulnerable to H2O2, even when various genes for peroxidemetabolizing enzymes are highly expressed. Moreover, CuCl2-pretreated cells are sensitive to treatment with the redox cycling drug, menadione. No physiological cross-protection response was observed in CuCl2-treated cells in a subsequent challenge with killing concentrations of an organic hydroperoxide. As H2O2 production is an important initial plant immune response, defects in H2O2 protection are likely to reduce bacterial survival in plant hosts and enhance the usefulness of copper biocides in controlling bacterial pathogens.

INTRODUCTION Xanthomonas campestris is an important pathogenic bacterium that causes destructive diseases in economic crops worldwide. The increased generation and accumulation of reactive oxygen species (ROS) elicited by invading microbes has been recognized as one of the initial plant defence responses during plant–microbe interactions (Levine et al., 1994). The use of biocides containing copper compounds, i.e. copper hydroxide, copper sulfate, copper oxychloride, copper hydrazine sulfate and cuprous oxide, to control the spread of phytopathogenic fungi and bacteria has been a feature of most traditional agricultural 3These authors contributed equally to this work. Abbreviations: Ohr, organic hydroperoxide reductase; ROOH, organic hydroperoxide; ROS, reactive oxygen species; SOD, superoxide dismutase.

458

practices (Hopkins, 2004). Copper bactericides are also effective control measures of plant diseases caused by X. campestris (McGuire, 1988). Although copper is an essential trace element required as a cofactor for a variety of enzymes involved in many biological processes, an excess of copper can be highly toxic, especially under aerobic conditions. As redox-active metal ions, copper toxicity is thought to be the consequence of Fenton and Haber-Weiss reactions, where copper catalyses the generation of hydroxyl radicals from H2O2 and superoxide anions (Kehrer, 2000; Letelier et al., 2010). Micro-organisms such as Escherichia coli have evolved multiple systems to maintain intracellular copper homeostasis (Rensing & Grass, 2003). Little is known about copper resistance determinants in X. campestris. A previous study in X. campestris pv. juglandis identified copper resistance genes sharing high homology to Pseudomonas syringe copABCD and to the plasmidlocated pcoABCD genes in E. coli (Lee et al., 1994). X. 072470 G 2014 SGM Printed in Great Britain

Effects of copper chloride in Xanthomonas campestrisof copper chloride in Xanthomonas campestris

euvesticatoria possesses copABCD located divergently to copRS; these genes encode a two-component copper sensor and a regulator that controls the expression of the copperinducible copABCD operon (Basim et al., 2005). However, recent studies in E. coli have revealed that copper ions are unlikely to participate in Fenton chemistry, and the iron–sulfur clusters of dehydratase enzymes are the primary intracellular targets of copper toxicity (Macomber et al., 2007; Macomber & Imlay, 2009). The lethal effect of copper in X. campestris is currently unclear. Challenging bacterial cultures with sublethal concentrations of copper synergistically increased the killing effects of H2O2 and organic hydroperoxide (ROOH) through increased hydroxyl radical production and stimulation of lipid peroxidation, respectively (Patikarnmonthon et al., 2010). X. campestris has evolved an array of mechanisms to overcome oxidative stresses and to survive under such harsh conditions. Several antioxidant enzymes are inducibly produced upon exposure to increased levels of ROS. For example, the expression of KatA (a monofunctional catalase), KatG (a catalase–peroxidase) and AhpCF (an alkyl hydroperoxide reductase), which is regulated by the global peroxide sensor and transcriptional regulator OxyR, increase in cells subjected to either H2O2 or ROOH (Charoenlap et al., 2005; Chauvatcharin et al., 2005; Jittawuttipoka et al., 2009). Ohr (organic hydroperoxide reductase), whose expression is controlled by OhrR, is upregulated in cells challenged with ROOH and sodium hypochlorite (Chi et al., 2011; Panmanee et al., 2006). Some of these genes, including katA, katG, oxyR and soxR, are required for full virulence of this phytopathogen (Charoenlap et al., 2011; Jittawuttipoka et al., 2009; Mahavihakanont et al., 2012). OxyR and OhrR are thiolbased redox sensors and transcriptional regulators; peroxide activation of OxyR and deactivation of OhrR occur via the oxidation of a thiol group of the sensing cysteine residue and subsequent disulfide bond formation. OxyR is primarily recognized as an H2O2 sensor, but in X. campestris, OxyR is also activated by organic hydroperoxides, redox cycling drugs and disulfide stress (treatments with diamide and N-ethylmaleimide) (Vattanaviboon et al., 1999). Moreover, in vitro experiments have revealed that the reactive nitrogen species-generating agents S-nitrosoglutathione, S-nitrosocysteine and glutathione disulfide oxidized and activated E. coli OxyR (Kim et al., 2002). OhrR primarily senses organic hydroperoxides (Panmanee et al., 2006). However, sodium hypochlorite-mediated inactivation of OhrR has been reported (Chi et al., 2011). Although X. campestris possesses SoxR, it appears that this transcriptional regulator senses redox-active drugs rather than superoxide anions (Mahavihakanont et al., 2012). IscR is an iron–sulfur (Fe–S) cluster-containing transcriptional regulator involved in the oxidative stress response (Somprasong et al., 2012). IscR-regulated genes are predominantly induced by superoxide generators such as plumbagin (Somprasong et al., 2012). Superoxide anions reacted with the Fe–S clusters causing their depletion. IscR http://mic.sgmjournals.org

senses changes in the levels of Fe–S clusters and responds by either activating or repressing transcription of target genes (Ayala-Castro et al., 2008). Furthermore, treatments with H2O2 and organic hydroperoxides are able to induce the expression of IscR-regulated genes (Somprasong et al., 2012). We report here that treatment with nonlethal doses of copper induced high expression of antioxidant genes in the OxyR and OhrR regulon. In contrast, copper-pretreated cells showed enhanced sensitivity to H2O2.

METHODS Bacterial strains and growth conditions. X. campestris pv. campestris ATCC 33913 (Xcc) wild-type and mutant strains were grown aerobically at 28 uC in Silva-Buddenhagen (SB) medium (Charoenlap et al., 2011) and plant-like medium XVM2 (Wengelnik & Bonas, 1996) as indicated. Exponential-phase cells (OD600 of 0.5) were used in all experiments. Real-time reverse transcription PCR (qRT-PCR). The exponential-phase cultures were challenged with 75 mM CuCl2 for 15 min.

Total RNA preparation and qRT-PCR were performed as previously described (Buranajitpakorn et al., 2011). Real-time PCR was executed using 20 ng cDNA, SYBR green PCR Master Mix (Applied Biosystems) and a specific primer pair for each gene. The 16S rRNA gene was used for normalization. The specific primer pairs were primers BT2684 (59CGCAGCGTCTCGGTGACG39) and BT2685 (59AGTGGAAGACGCCGCTGA39) for ahpC, BT2239 (59TCTGCTTGCCACCGGACT39) and BT2240 (59TGTGGCAGGACCCGATCC39) for katG, BT2237 (59GGCCAGGTCGTCCGGCTT39) and BT2238 (59GAATCCACCCGCACGCTG39) for katA, BT3373 (59ACGTCAAGCTCTCCACCC39) and BT3374 (59GCTGGCATTGGAGTACG39) for ohr, BT1798 (59GGATGCAGTGTTCGATTC39) and BT1806 (59CATGCAGCTGGTTTTCGT39) for sufB, BT4244 (59GTCGATGCACATGATCATGT39) and BT4245 (59AGATCGCGACGCTGGAAGT39) for katE, and BT2781 (59GCCCGCACAAGCGGTGGAG39) and BT2782 (59ACGTCATCCCCACCTTCCT39) for the 16S rRNA gene. PCR amplifications were performed under the following conditions: denaturation at 95 uC for 30 s, annealing at 60 uC for 45 s and extension at 72 uC for 30 s, for 40 cycles. Relative expression was expressed as fold of expression relative to the uninduced level. Oxidant killing experiments. Oxidant killing experiments were

performed as previously described (Banjerdkij et al., 2005). Bacterial cultures were grown to the exponential phase before being challenged with 75 mM CuCl2 for 15 min. After treatment, bathocuproine sulfonate (copper chelator) was added to the final concentration of 150 mM. Aliquots of cells were treated with killing concentrations of H2O2, menadione or tert-butyl hydroperoxide (tBOOH) for 30 min. After treatment, cell survival was enumerated using viable cell count. The percentage survival was defined as the number of c.f.u. recovered after the treatment divided by the number of c.f.u. prior to treatment multiplied by 100. Peroxide degradation assay. The degradation of peroxides was

monitored using a ferrous oxidation-xylenol orange (FOX) assay as previously described (Chuchue et al., 2006) with some modifications. Exponential-phase cultures of Xcc strains grown under uninduced and 75 mM CuCl2-induced conditions were adjusted to an OD600 of 0.5 with fresh medium prior to addition of 200 mM H2O2. The residual peroxide in the supernatant was determined at indicated time points using a xylenol orange-iron reaction (Chuchue et al., 2006). 459

P. Sornchuer and others

Catalase and superoxide dismutase assays. Cell lysate prepara-

tion was performed as previously described (Chauvatcharin et al., 2005). Exponential-phase cells in XVM2 medium were induced with 75 mM CuCl2 for 15 min. Bacterial cells were harvested and lysed using brief sonication followed by centrifugation. The supernatants were used for enzymic activity assays. The catalase activity assay was monitored as previously described (Chauvatcharin et al., 2005). One unit of catalase is defined as the amount of enzyme required to decompose 1.0 mmol of H2O2 at 25 uC at pH 7.0. Total superoxide dismutase (SOD) activity was determined using the xanthine–xanthine oxidase-coupled reduction of cytochrome c as previously described (McCord & Fridovich, 1969). One unit of SOD activity is defined as the amount of enzyme required to inhibit the rate of reduction of cytochrome c by 50 %. Purification of X. campestris pv. campestris KatA. The 66His-

tagged X. campestris pv. campestris KatA was purified using E. coli and the pETBlue-2 system (Novagen). The full-length katA gene was PCR amplified from pKatA with forward primer BT4052 containing an NcoI restriction site (59-ACGCCATGGGCCCTGGATCCCTCA-39) and reverse primer BT3997 (59-TCAATGGTGGTGGTGGTGATGGTCCTGCAGGGTCGACGC-39) containing 66His and stop codons. The PCR product was cloned into pETBlue-2 cut with NcoI and NotI (filled in with a Klenow fragment), yielding pET-KatA. E. coli BL21(DE3) harbouring pET-KatA was grown in Terrific broth (Life Technology) supplemented with 0.5 M glycylglycine and 50 mg ml21 ampicillin to an OD600 of 1 before the culture was induced with 0.5 mM IPTG for 4 h at 28 uC. After harvesting, bacterial cells were lysed using intermittent sonication. An Ni-NTA agarose column (Qiagen) was used to purify 66His-tagged KatA from the crude lysate as per the manufacturer’s recommendations. The bound 66Histagged KatA protein was eluted with elution buffer (50 mM NaH2PO4, 200 mM NaCl, 250 mM imidazole, pH 8.0). Fractions that contain KatA protein were dialysed in PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4). The purity of KatA was 90 % as judged by SDS-PAGE analysis. Preparation of polyclonal anti-KatA antibody. Polyclonal anti-

KatA antibody was prepared by immunizing a mouse with purified 66His-tagged KatA (performed by Biomedical Technology Research unit, Faculty of Associated Medical Sciences, Chiang Mai University, Thailand). Western blot analysis. Bacterial cell lysate was separated using 12.5 % SDS-PAGE and electrotransferred to a PVDF membrane (Hybond-ECL; Amersham Biosciences) at 22 V for 45 min in blotting buffer (191 mM glycine, 24 mM Tris-base, 20 % methanol). The blotted membrane was blocked with 5 % skimmed milk in Trisbuffered saline Tween-20 (TBST) at 4 uC overnight prior to probing with a 1 : 4000 dilution of mouse anti-KatA antibody as a primary antibody. After washing with TBST (36 10 min), sheep anti-mouse IgG conjugated with HRP (GE Healthcare Life Sciences) was used as the secondary antibody. Detection was performed using chemiluminescent HRP substrate (Roche Diagnosis) and Hyperfilm (GE Healthcare Life Sciences). MIC determination. The broth dilution method was used to

determine the MIC of copper as previously described (Teitzel & Parsek, 2003) with some modifications. An exponential-phase culture of Xcc in XVM2 was adjusted to an OD600 of 0.3 and a 100 ml aliquot of this culture was inoculated into tubes containing 1 ml of XVM2 medium plus CuCl2 at concentrations ranging from 50 mM to 1 mM. The MIC was defined as the lowest concentration that showed no visible growth after 24 h at 28 uC. 460

RESULTS AND DISCUSSION The level of copper resistance in X. campestris pv. campestris Experiments were performed to determine the copper resistance levels of Xcc grown in XVM2, an hrp-inducing medium that provides an environment similar to the plant extracellular space (Wengelnik & Bonas, 1996). The MIC of CuCl2 in XVM2 medium for Xcc was 150 mM. The intracellular copper level in bacteria is tightly controlled by well-orchestrated mechanisms that are induced in response to increased copper levels (Rademacher & Masepohl, 2012). Analysis of the genome sequence of strain Xcc (da Silva et al., 2002; Qian et al., 2005; Vorho¨lter et al., 2008) revealed that this bacterium only contains homologues of copL (xcc0579), copA (xcc0577) and copB (xcc0576), which encode a copper resistance protein, multi-copper oxidase and outer-membrane protein involved in copper resistance, respectively. This copLAB operon is a copper resistance trait in Xcc (Hsiao et al., 2011). No open reading frames with high identity to CopRS (or PcoRS in E. coli) were identified, suggesting that copLAB might be regulated differently in Xcc than in other bacteria. In addition to copLAB, we have identified only two genes, cutA (xcc0520, putative copper binding protein) and cutC (xcc2913, putative copper transporter), that might have a role in copper homeostasis. These findings suggest that Xcc is poorly equipped to handle excess levels of copper. Nevertheless, it is possible that Xcc might possess other, as yet undiscovered, copper detoxification genes. Copper ions induce antioxidant gene expression We have proposed that one of the principal pathways of copper toxicity in Xcc involves induction of oxidative stress by the metal (Patikarnmonthon et al., 2010). Here, we determined the effects of exposure to sublethal concentrations of copper on the expression of genes involved in the oxidative stress response. Xcc cultures were challenged with either 75 or 100 mM of CuCl2 for 15 min. These treatment conditions did not affect bacterial viability (data not shown). The transcription levels of representative genes of each of the four regulons involved in oxidative stress protection were measured. The levels of ahpC, katA and katG (OxyR regulon), ohr (OhrR regulon), xcc0300 (SoxR regulon) and sufB (IscR regulon) expression were monitored using qRT-PCR. As illustrated in Fig. 1, 75 mM CuCl2 treatment enhanced expression of ahpC, katA, katG, ohr and sufB by 20.1±1.5, 33.2±2.5, 23.9±2.4, 33.1±1.9 and 11.6±1.2-fold, respectively, compared with the untreated levels (data expressed as means±SD of three independent experiments). No induction of gene expression in response to the treatment was observed for xcc0300 (data not shown). Treatment of Xcc with a higher concentration of CuCl2 (100 mM) induced ahpC, katA, katG, ohr and sufB expression by 10.8±1.2, 17.1±1.7, 13.4±2.5, 21.4±1.5 and 2.6±1.0-fold, respectively. These levels, Microbiology 160

Effects of copper chloride in Xanthomonas campestrisof copper chloride in Xanthomonas campestris

ahpC katA katG ohr sufB

35

Relative expression (fold)

30 25 20 15 10 5 0

75 mM

100 mM CuCl2

Fig. 1. The effects of CuCl2 on antioxidant gene expression. The expression levels of ahpC, katA, katG, ohr and sufB in X. campestris pv. campestris wild-type grown in XVM2 medium treated with 75 or 100 mM CuCl2 for 15 min were determined by qRT-PCR using specific primer pairs. Relative expression was calculated using the 16S rRNA gene for normalization and expressed as fold change relative to the untreated control. Data shown are means±SD of three independent experiments.

albeit elevated, were lower than the induction of gene expression attained by treatment of the bacteria with 75 mM CuCl2. This is probably due to partial growth inhibition at 100 mM CuCl2. Moreover, experiments were extended to measure the expression levels of katE, which encodes a growth-phase- and starvation-regulated catalase (Vattanaviboon & Mongkolsuk, 2000). The expression of katE was increased 7.0±2.4fold upon exposure to 75 mM CuCl2. Copper-induced katE expression was OxyR-independent as the induction was still observed in an oxyR mutant (data not shown). The expression of X. campestris katE is proposed to be regulated by the sigma factor, RpoS (Vattanaviboon & Mongkolsuk, 2000). This observation is in harmony with a previous report in E. coli indicating that copper stress induces the RpoS regulon (Macomber et al., 2007). Characterization of ROS generated upon CuCl2 exposure The question was then raised as to how exposure to elevated levels of copper triggered the expression of genes in the oxidative stress response. These transcriptional regulators sense various oxidants, thus CuCl2 treatment likely generated intracellular ROS. To determine whether the ROS participated in CuCl2-induced gene expression, CuCl2 treatments of strain Xcc harbouring plasmids that http://mic.sgmjournals.org

produced high levels of an antioxidant enzyme that preferentially degraded each type of ROS were performed and the resulting induction of gene expression was monitored by real-time PCR. We hypothesized that if superoxide radicals, H2O2, and ROOH are generated by CuCl2 treatment, high levels of oxidant-scavenging enzymes SOD (a superoxide scavenger), catalase (an H2O2 scavenger), and alkyl hydroperoxide reductase (an ROOH scavenger) should lower the magnitude of induction of sufB (an IscR-regulated gene), ahpC (an OxyR-regulated gene) and ohr (an OhrRregulated gene), respectively, in CuCl2-treated cells. Similar rationales have been used to differentiate the major ROS responsible for activation of transcriptional regulators (Gu & Imlay, 2011; Mahavihakanont et al., 2012). Hence, copper induction experiments were repeated in Xcc wild-type strain harbouring pSodBI (Saenkham et al., 2007), pKatG (Jittawuttipoka et al., 2009) and pAhpCF (Charoenlap et al., 2005). High levels of KatG catalase dramatically diminished CuCl2-induced expression of ahpC (from 20.1±1.5 to 5.0±0.9-fold), while high levels of AhpCF caused a major reduction in ohr expression (from 33.1±1.9 to 1.7±0.5fold; Fig. 2). High levels of SOD considerably reduced CuCl2-induced sufB expression from 11.6±1.2-fold in wildtype to 4.4±0.9-fold in the strain harbouring plasmid pSodBI but had no significant effect on CuCl2-mediated induction of ahpC and ohr (Fig. 2). Moreover, high levels of KatG and AhpCF also lowered CuCl2-induced expression of sufB to 6.8±1.3 and 5.9±1.1-fold (Fig. 2). These findings suggest that in CuCl2-treated cells, H2O2 and ROOH are responsible for oxidation activation and deactivation of OxyR and OhrR, respectively. In addition, superoxides are likely to be produced, as shown by high levels of SOD reducing sufB expression levels in CuCl2-induced cultures. This observation implies the direct contribution of superoxide anions in CuCl2 toxicity. The biochemical basis of how CuCl2 treatment resulted in generation and accumulation of H2O2, ROOH and superoxide anions is as yet unclear. A recent finding has shown that iron–sulfur clusters are primary targets for copper toxicity (Chillappagari et al., 2010; Macomber & Imlay, 2009). Increased intracellular Cu(I) ions destabilize iron– sulfur clusters, leading to functionally inactive proteins. Genome-wide transcriptome analysis of Bacillus subtilis has demonstrated that numerous genes for iron–sulfur cluster-containing enzymes, including those involved in the electron transport chain, are upregulated during copper stress conditions (Chillappagari et al., 2010). It is likely that excess copper ions can cause ineffective electron transport, resulting in increased generation of superoxide anions. This notion is supported by our evidence that superoxide anions are one of the radicals accumulated after CuCl2 treatment and by a previous finding that excess copper sulfate (CuSO4) acted as a redox cycling agent able to generate superoxide anions in the presence of molecular oxygen (Kimura & Nishioka, 1997); mutants lacking functional SOD displayed a CuSO4-sensitive phenotype. Dismutation of superoxide anions catalysed by SOD 461

P. Sornchuer and others

40

ahpC ohr sufB

35

Relative expression (fold)

30 25 20 15 10 5 0 Xcc

Xcc/pKatG Xcc/pAhpCF Xcc/pSodBI

Fig. 2. Characterization of ROS generated during CuCl2 treatment. Xcc wild-type harbouring pBBR1MCS- vector (Xcc), pKatG plasmid (Xcc/pKatG), pAhpCF plasmid (Xcc/pAhpCF) or pSodBI plasmid (Xcc/pSodBI) were cultivated in XVM2 medium and induced with 75 mM CuCl2 for 15 min. The expression level of antioxidant genes was monitored using qRT-PCR. Error bars represent means±SD from three independent experiments.

produces H2O2. H2O2 can further activate genes in the OxyR regulon via oxidation of OxyR. Furthermore, Xcc contains five putative open reading frames homologous to SOD, none of which are regulated by SoxR, the superoxide anion/redox cycling drug sensor and transcriptional regulator (Mahavihakanont et al., 2012). As a redox metal ion, excess copper can generate highly toxic hydroxyl radicals during redox cycling between Cu(I) and Cu(II) in the presence of H2O2 via a Fenton-like reaction (Feldberg et al., 1985). The hydroxyl radical is a highly reactive compound capable of initiating the lipid peroxidation chain reaction. In addition, copper ions can bind to reduced glutathione (GSH), lowering the level of GSH, an important antioxidant molecule (Freedman et al., 1989). Taken together, these biochemical changes caused by high levels of copper could be responsible for the accumulation of ROS. Copper-pretreated cells are vulnerable to subsequent H2O2 killing treatments Elevated expression of antioxidant enzymes is typically associated with enhanced resistance to oxidants. Therefore, we tested whether copper pretreatment could induce crossprotection to subsequent treatment with lethal concentrations of peroxides. CuCl2-pretreated and untreated cultures were subjected to 1, 2 and 5 mM H2O2 for 30 min. The 462

presence of copper ions (at a concentration of 100 mM) potentiates H2O2 killing effects through increased generation of hydroxyl radicals (Patikarnmonthon et al., 2010). The enhanced killing effect of H2O2 by copper was abolished by the addition of the copper-specific chelator bathocuproine sulfonate (Patikarnmonthon et al., 2010). Treatment with bathocuproine sulfonate can cause depletion of intracellular copper ions (Rae et al., 1999). In our experiments, after induction with 75 mM for 15 min, 150 mM bathocuproine sulfonate was added to the induced cultures prior to treatment with H2O2. The CuCl2-pretreated cells were hypersensitive to H2O2 treatment relative to the untreated cells: these cells were more than 104-fold more sensitive to treatment with 5 mM H2O2 (Fig. 3a). During plant–microbe interactions, oxidative burst is an early plant defence response against invading pathogens (Lamb & Dixon, 1997; Levine et al., 1994). A rapid burst of ROS, including superoxide anion and H2O2, is observed in both compatible and incompatible plant–microbe interactions (Lamb & Dixon, 1997). The concentration of H2O2 in the oxidative burst has been shown to reach low millimolar levels (Bargabus et al., 2003). The generated ROS not only protects plants against the invading pathogens, but is also involved in the activation of hypersensitivity responses and the strengthening of host cell walls (Bolwell & Daudi, 2009). The observation here that CuCl2-pretreated Xcc was extremely sensitive to low millimolar concentrations of H2O2 suggests that treatment with nonlethal doses of copper could affect plant–Xcc interactions. H2O2 hypersensitivity in the copper-pretreated cells is accompanied by lowered H2O2 degradation rates and total catalase activity The H2O2 hypersensitivity of the copper-pretreated cells raised the question of why copper-treated cells with high expression of katA, katG and katE are vulnerable to H2O2. A modified FOX assay was performed to determine bacterial ability to degrade peroxide. CuCl2 pretreated cells displayed lowered rates of H2O2 degradation relative to the untreated control (Fig. 3b). Thus, reduction of the ability to degrade H2O2 is, at least in part, responsible for the observed increase in H2O2 sensitivity of the CuCl2-pretreated cells. As catalases are the principal antioxidant enzymes that degrade H2O2 to water and molecular oxygen, total catalase activity was monitored in CuCl2pretreated and untreated cells. Total catalase activity in the CuCl2-pretreated cells was 6.5±1.2 U (mg protein)21, or approximately twofold lower than the level attained in untreated cells (14.6±2.5 U (mg protein)21). Hence, CuCl2 treatment enhanced transcription of katA while concomitantly decreasing catalase enzymic activity. As KatA is the principal catalase produced during exponential growth in Xcc (Charoenlap et al., 2011), we tested the effects of copper pretreatment on catalase activity in the Xcc wild-type strain harbouring pKatA plasmid for high expression of katA. Treating bacterial cultures with 75 mM Microbiology 160

Effects of copper chloride in Xanthomonas campestrisof copper chloride in Xanthomonas campestris

the untreated control, respectively. Thus, exposure to excess CuCl2 reduces enzymic catalase activity.

(a) 103 102

An additional experiment was performed to test whether the CuCl2-treated cells reduced catalase activity and high expression of katA was due to post-transcriptional modification of KatA, instability of KatA or other modifications of enzyme activity. Western blot analysis was conducted to measure the level of KatA protein in uninduced Xcc cultures and those induced with 75 mM CuCl2. KatA protein levels in the CuCl2-treated culture were 2.1-fold higher than in the uninduced culture (Fig. 4). Experiments were expanded to include Xcc harbouring pKatA (Xcc/pKatA). Treatment with either 75 or 100 mM CuCl2 caused no reduction in the levels of KatA protein (Fig. 4). Thus, reduced catalase activity in the CuCl2treated culture was not due to instability of KatA catalase upon exposure to CuCl2. This result supports the notion that CuCl2 treatment has no significant effects on KatA protein stability.

Survival (%)

101 100 10–1 10–2 10–3 10–4

0

1

2 3 H2O2 (mM)

4

5

(b)

Residual H2O2 (%)

100 80 60 40 20 0 0

1

2

3

4

Time (min)

Fig. 3. H2O2 resistance levels and degradation rates in Xcc induced with CuCl2. (a) Exponential cultures of Xcc grown in XVM2 medium were induced with 75 mM CuCl2 for 15 min before 150 mM bathocuproine sulfonate (copper chelator) was added. The uninduced (#) and CuCl2-induced ($) cultures were treated with 1, 2 or 5 mM H2O2 for 30 min. Cells that survived the treatment were enumerated using viable plate counts. Percent survival was defined as the ratio of the c.f.u. after treatment to the c.f.u. before treatment multiplied by 100. m represents CuCl2induced cultures without bathocuproine sulfonate treatment. (b) The rates of H2O2 degradation in culture medium containing 200 mM H2O2 by Xcc are shown. The levels of H2O2 remaining in the culture medium of uninduced (#) and 75 mM CuCl2-induced ($) cells at the various time points are determined, along with those of an uninoculated control (m). Error bars represent means±SD from three independent experiments.

We found here that excess CuCl2 reduced catalase activity. Three isozymes of catalases, i.e. KatA, KatG and KatE, are produced in X. campestris (Chauvatcharin et al., 2005; Jittawuttipoka et al., 2009; Vattanaviboon & Mongkolsuk, 2000). KatA and KatE are monofunctional catalases, while KatG is a catalase–peroxidase. All three catalases contain haem prosthetic groups as cofactors, and the iron ion (Fe3+) in haem is a key factor in the oxidation of H2O2 to molecular oxygen and water. Excess copper competes with and replaces other metals, including iron, in proteins (Kershaw et al., 2005). We therefore speculate that haemcontaining catalases are among the targets of copper toxicity though the mechanism for the effect of copper on catalase activity is not currently known. Copper-pretreated cells are vulnerable to menadione killing Menadione is a redox cycling drug that exerts its toxicity intracellularly through the generation of superoxide anions

Xcc/pKatA Xcc UN

Cu

UN

Cu (mM) 75 100 KatA

CuCl2 for 15 min resulted in 641±26 U (mg protein)21 total catalase activity, compared with 1056±55 U (mg protein)21 in the untreated cells. When the CuCl2 concentration was increased to 100 mM, total catalase activity decreased to 474±25 U (mg protein)21. This observation is consistent with the results of in vitro experiments using purified KatA protein. Treatment with either CuCl2 or CuCl (75 mM) for 15 min lowered KatA catalase activity to levels that were 78±9 % and 35±9 % of those observed in http://mic.sgmjournals.org

Fig. 4. Determination of KatA levels in CuCl2-induced cells. The levels of KatA catalase in Xcc and Xcc harbouring the katA expression plasmid pKat (Xcc/pKatA) cultivated under uninduced (UN), 75 mM CuCl2-induced (Cu) or 100 mM CuCl2-induced conditions were determined using Western blot analysis probed with anti-KatA antibody. Crude lysates (30 mg for Xcc and 5 mg for Xcc/pKatA) were separated on 12.5 % SDS-PAGE. 463

P. Sornchuer and others

in the presence of oxygen (Yamashoji et al., 2001). The effect of CuCl2 pretreatment on the protection of Xcc against menadione killing treatment was investigated. Xcc cells pretreated with 75 mM CuCl2 for 15 min were 20-fold more vulnerable to menadione killing treatments (50 mM) than the uninduced cells (Fig. 5). SODs are enzymes responsible for detoxifying superoxide anions. Total SOD activity in the CuCl2-pretreated cultures was determined. The level of total SOD activity in the CuCl2pretreated cells (6.2±0.1 U (mg protein)21) was comparable to that of the uninduced cells (6.0±0.8 U (mg protein)21). Thus, enhanced sensitivity to menadione killing of the CuCl2-pretreated cells was not a result of reduction of total SOD levels. Dismutation of superoxide anions generates H2O2. Inefficient removal of H2O2 in the CuCl2-pretreated cells would lead to increased generation of the most deleterious hydroxyl radical via a Haber-Weiss reaction, which is mediated by iron ions and other heavy metal ions including Cu ions (Kehrer, 2000; Letelier et al., 2010). Bathocuproine sulfonate was added to the CuCl2pretreated cultures prior to menadione killing treatments. Addition of bathocuproine sulfonate fully compensated for the menadione-sensitive phenotype of the CuCl2pretreated cells (Fig. 5). This observation supports the assumption that the metal ion-mediated Haber-Weiss reaction accounts for the menadione-sensitive phenotype of the CuCl2-pretreated cells.

103

Survival (%)

102 101 100

We tested whether elevated expression of antioxidant genes induced by copper pretreatment conferred resistance to ROOH killing treatments. CuCl2-induced cells were treated with lethal concentrations of tBOOH. No cross-protection was achieved (data not shown). This observation was consistent with our previous observation in Xanthomonas where pretreatment with ROOH, which triggers the expression of genes in both OxyR and OhrR regulons, did not induce adaptive protection to a subsequent killing treatment with organic hydroperoxide (Mongkolsuk et al., 1997).

CONCLUSION We present here the effect of sublethal concentrations of CuCl2 on antioxidant gene expression and the consequences of CuCl2 induction on bacterial survival of peroxide killing treatments. In an environment mimicking plant extracellular space, treatment with nonlethal doses of CuCl2 that caused no effect on bacterial viability resulted in the accumulation of peroxides, both H2O2 and ROOH. Enhanced levels of peroxides induce high expression of antioxidant genes ahpC, katA and katG that belong to the OxyR regulon, and ohr that is regulated by OhrR. Though nonlethal doses of CuCl2 induced antioxidant gene expression, the induced cells were more vulnerable to H2O2 than the uninduced cells due to inactivation of haemcontaining catalases by CuCl2. Thus, nonlethal concentrations of CuCl2 not only potentiate oxidative stress killing effects (Patikarnmonthon et al., 2010) but also reduce bacterial ability to cope with H2O2 that is commonly generated during plant–microbe interactions.

ACKNOWLEDGEMENTS

10–1 10–2 10–3

Copper pretreatment is unable to induce crossprotection against organic hydroperoxides

0

10

20

30

40

50

Menadione (mM)

The research was supported by grants from Chulabhorn Research Institute and the Center of Excellence on Environmental Health and Toxicology, Science & Technology Postgraduate Education and Research Development Office (PERDO), Ministry of Education, Thailand. P. S. was supported by the Chulabhorn Graduate Institute. The authors thank Dr James M. Dubbs for a critical reading of the revised manuscript.

REFERENCES Fig. 5. Menadione resistance levels in Xcc pretreated with CuCl2. Exponential-phase cultures of Xcc in XVM2 medium were induced with 75 mM CuCl2 for 15 min prior to treatment with killing concentrations (5–50 mM) of menadione for 30 min in the presence (g) and absence (m) of 150 mM bathocuproine sulfonate. Cells that survived the treatment were enumerated and the results were expressed as percent survival. # and $ represent uninduced cultures treated with menadione with and without bathocuproine sulfonate addition, respectively. Error bars represent means±SD from three independent experiments. 464

Ayala-Castro, C., Saini, A. & Outten, F. W. (2008). Fe-S cluster

assembly pathways in bacteria. Microbiol Mol Biol Rev 72, 110–125. Banjerdkij, P., Vattanaviboon, P. & Mongkolsuk, S. (2005). Exposure

to cadmium elevates expression of genes in the OxyR and OhrR regulons and induces cross-resistance to peroxide killing treatment in Xanthomonas campestris. Appl Environ Microbiol 71, 1843–1849. Bargabus, R. L., Zidack, N. K., Sherwood, J. E. & Jacobsen, B. J. (2003). Oxidative burst elicited by Bacillus mycoides isolate Bac J, a

biological control agent, occurs independently of hypersensitive cell death in sugar beet. Mol Plant Microbe Interact 16, 1145–1153. Microbiology 160

Effects of copper chloride in Xanthomonas campestrisof copper chloride in Xanthomonas campestris

Basim, H., Minsavage, G. V., Stall, R. E., Wang, J. F., Shanker, S. & Jones, J. B. (2005). Characterization of a unique chromosomal

by providing protection against low levels of H2O2. J Bacteriol 191, 7372–7377.

copper resistance gene cluster from Xanthomonas campestris pv. vesicatoria. Appl Environ Microbiol 71, 8284–8291.

Kehrer, J. P. (2000). The Haber-Weiss reaction and mechanisms of

Bolwell, G. P. & Daudi, A. (2009). Reactive oxygen species in plant-

Kershaw, C. J., Brown, N. L., Constantinidou, C., Patel, M. D. & Hobman, J. L. (2005). The expression profile of Escherichia coli K-12

pathogen interactions. In Reactive Oxygen Species in Plant Signalling, pp. 113–133. Edited by L. A. del Rio & A. Puppo. Berlin, Germany: Springer. Buranajitpakorn, S., Piwkam, A., Charoenlap, N., Vattanaviboon, P. & Mongkolsuk, S. (2011). Genes for hydrogen peroxide detoxifica-

tion and adaptation contribute to protection against heat shock in Xanthomonas campestris pv. campestris. FEMS Microbiol Lett 317, 60– 66.

toxicity. Toxicology 149, 43–50.

in response to minimal, optimal and excess copper concentrations. Microbiology 151, 1187–1198. Kim, S. O., Merchant, K., Nudelman, R., Beyer, W. F., Jr, Keng, T., DeAngelo, J., Hausladen, A. & Stamler, J. S. (2002). OxyR: a

molecular code for redox-related signaling. Cell 109, 383–396. Kimura, T. & Nishioka, H. (1997). Intracellular generation of super-

oxide by copper sulphate in Escherichia coli. Mutat Res 389, 237–242.

Charoenlap, N., Eiamphungporn, W., Chauvatcharin, N., Utamapongchai, S., Vattanaviboon, P. & Mongkolsuk, S. (2005). OxyR

Lamb, C. & Dixon, R. A. (1997). The oxidative burst in plant disease

mediated compensatory expression between ahpC and katA and the significance of ahpC in protection from hydrogen peroxide in Xanthomonas campestris. FEMS Microbiol Lett 249, 73–78.

Lee, Y. A., Hendson, M., Panopoulos, N. J. & Schroth, M. N. (1994).

Charoenlap, N., Buranajitpakorn, S., Duang-Nkern, J., Namchaiw, P., Vattanaviboon, P. & Mongkolsuk, S. (2011). Evaluation of the

virulence of Xanthomonas campestris pv. campestris mutant strains lacking functional genes in the OxyR regulon. Curr Microbiol 63, 232– 237.

resistance. Annu Rev Plant Physiol Plant Mol Biol 48, 251–275. Molecular cloning, chromosomal mapping, and sequence analysis of copper resistance genes from Xanthomonas campestris pv. juglandis: homology with small blue copper proteins and multicopper oxidase. J Bacteriol 176, 173–188. Letelier, M. E., Sa´nchez-Jofre´, S., Peredo-Silva, L., Corte´s-Troncoso, J. & Aracena-Parks, P. (2010). Mechanisms underlying iron and

Chauvatcharin, N., Atichartpongkul, S., Utamapongchai, S., Whangsuk, W., Vattanaviboon, P. & Mongkolsuk, S. (2005). Genetic

copper ions toxicity in biological systems: Pro-oxidant activity and protein-binding effects. Chem Biol Interact 188, 220–227.

and physiological analysis of the major OxyR-regulated katA from Xanthomonas campestris pv. phaseoli. Microbiology 151, 597–605.

Levine, A., Tenhaken, R., Dixon, R. & Lamb, C. (1994). H2O2 from the

oxidative burst orchestrates the plant hypersensitive disease resistance response. Cell 79, 583–593.

Chi, B. K., Gronau, K., Ma¨der, U., Hessling, B., Becher, D. & Antelmann, H. (2011). S-bacillithiolation protects against hypochlor-

Macomber, L. & Imlay, J. A. (2009). The iron-sulfur clusters of

ite stress in Bacillus subtilis as revealed by transcriptomics and redox proteomics. Mol Cell Proteomics 10, M111.009506.

dehydratases are primary intracellular targets of copper toxicity. Proc Natl Acad Sci U S A 106, 8344–8349.

Chillappagari, S., Seubert, A., Trip, H., Kuipers, O. P., Marahiel, M. A. & Miethke, M. (2010). Copper stress affects iron homeostasis

Macomber, L., Rensing, C. & Imlay, J. A. (2007). Intracellular copper

by destabilizing iron-sulfur cluster formation in Bacillus subtilis. J Bacteriol 192, 2512–2524. Chuchue, T., Tanboon, W., Prapagdee, B., Dubbs, J. M., Vattanaviboon, P. & Mongkolsuk, S. (2006). ohrR and ohr are the

primary sensor/regulator and protective genes against organic hydroperoxide stress in Agrobacterium tumefaciens. J Bacteriol 188, 842–851. da Silva, A. C., Ferro, J. A., Reinach, F. C., Farah, C. S., Furlan, L. R., Quaggio, R. B., Monteiro-Vitorello, C. B., Van Sluys, M. A., Almeida, N. F. & other authors (2002). Comparison of the genomes of two

Xanthomonas pathogens with differing host specificities. Nature 417, 459–463. Feldberg, R. S., Carew, J. A. & Paradise, R. (1985). Probing Cu(II)/

H2O2 damage in DNA with a damage-specific DNA binding protein. J Free Radic Biol Med 1, 459–466. Freedman, J. H., Ciriolo, M. R. & Peisach, J. (1989). The role of

glutathione in copper metabolism and toxicity. J Biol Chem 264, 5598–5605. Gu, M. & Imlay, J. A. (2011). The SoxRS response of Escherichia coli is

does not catalyze the formation of oxidative DNA damage in Escherichia coli. J Bacteriol 189, 1616–1626. Mahavihakanont, A., Charoenlap, N., Namchaiw, P., Eiamphungporn, W., Chattrakarn, S., Vattanaviboon, P. & Mongkolsuk, S. (2012).

Novel roles of SoxR, a transcriptional regulator from Xanthomonas campestris, in sensing redox-cycling drugs and regulating a protective gene that have overall implications for bacterial stress physiology and virulence on a host plant. J Bacteriol 194, 209–217. McCord, J. M. & Fridovich, I. (1969). Superoxide dismutase. An

enzymic function for erythrocuprein (hemocuprein). J Biol Chem 244, 6049–6055. McGuire, R. G. (1988). Evaluation of bactericidal chemicals for

control of Xanthomonas on citrus. Plant Dis 72, 1016–1020. Mongkolsuk, S., Vattanaviboon, P. & Praitaun, W. (1997). Induced

adaptive and cross-protection responses against oxidative stress killing in a bacterial phytopathogen, Xanthomonas oryzae pv. oryzae. FEMS Microbiol Lett 146, 217–222. Panmanee, W., Vattanaviboon, P., Poole, L. B. & Mongkolsuk, S. (2006). Novel organic hydroperoxide-sensing and responding mechan-

directly activated by redox-cycling drugs rather than by superoxide. Mol Microbiol 79, 1136–1150.

isms for OhrR, a major bacterial sensor and regulator of organic hydroperoxide stress. J Bacteriol 188, 1389–1395.

Hopkins, D. L. (2004). Management of Bacterial Diseases: Chemical

Patikarnmonthon, N., Nawapan, S., Buranajitpakorn, S., Charoenlap, N., Mongkolsuk, S. & Vattanaviboon, P. (2010). Copper ions potentiate

Methods. London, UK: Taylor & Francis. Hsiao, Y. M., Liu, Y. F., Lee, P. Y., Hsu, P. C., Tseng, S. Y. & Pan, Y. C. (2011). Functional characterization of copA gene encoding multi-

copper oxidase in Xanthomonas campestris pv. campestris. J Agric Food Chem 59, 9290–9302. Jittawuttipoka, T., Buranajitpakorn, S., Vattanaviboon, P. & Mongkolsuk, S. (2009). The catalase-peroxidase KatG is required

for virulence of Xanthomonas campestris pv. campestris in a host plant http://mic.sgmjournals.org

organic hydroperoxide and hydrogen peroxide toxicity through different mechanisms in Xanthomonas campestris pv. campestris. FEMS Microbiol Lett 313, 75–80. Qian, W., Jia, Y., Ren, S. X., He, Y. Q., Feng, J. X., Lu, L. F., Sun, Q., Ying, G., Tang, D. J. & other authors (2005). Comparative and

functional genomic analyses of the pathogenicity of phytopathogen Xanthomonas campestris pv. campestris. Genome Res 15, 757–767. 465

P. Sornchuer and others Rademacher, C. & Masepohl, B. (2012). Copper-responsive gene regulation in bacteria. Microbiology 158, 2451–2464. Rae, T. D., Schmidt, P. J., Pufahl, R. A., Culotta, V. C. & O’Halloran, T. V. (1999). Undetectable intracellular free copper: the requirement

of a copper chaperone for superoxide dismutase. Science 284, 805– 808. Rensing, C. & Grass, G. (2003). Escherichia coli mechanisms of

copper homeostasis in a changing environment. FEMS Microbiol Rev 27, 197–213. Saenkham, P., Eiamphungporn, W., Farrand, S. K., Vattanaviboon, P. & Mongkolsuk, S. (2007). Multiple superoxide dismutases in

Agrobacterium tumefaciens: functional analysis, gene regulation, and influence on tumorigenesis. J Bacteriol 189, 8807–8817. Somprasong, N., Jittawuttipoka, T., Duang-Nkern, J., Romsang, A., Chaiyen, P., Schweizer, H. P., Vattanaviboon, P. & Mongkolsuk, S. (2012). Pseudomonas aeruginosa thiol peroxidase protects against

hydrogen peroxide toxicity and displays atypical patterns of gene regulation. J Bacteriol 194, 3904–3912.

Vattanaviboon, P. & Mongkolsuk, S. (2000). Expression analysis and

characterization of the mutant of a growth-phase- and starvationregulated monofunctional catalase gene from Xanthomonas campestris pv. phaseoli. Gene 241, 259–265. Vattanaviboon, P., Varaluksit, T. & Mongkolsuk, S. (1999).

Modulation of peroxide stress response by thiol reagents and the role of redox sensor - transcription regulator, OxyR, in mediating the response in Xanthomonas. FEMS Microbiol Lett 176, 471–476. Vorho¨lter, F. J., Schneiker, S., Goesmann, A., Krause, L., Bekel, T., Kaiser, O., Linke, B., Patschkowski, T., Ru¨ckert, C. & other authors (2008). The genome of Xanthomonas campestris pv. campestris B100

and its use for the reconstruction of metabolic pathways involved in xanthan biosynthesis. J Biotechnol 134, 33–45. Wengelnik, K. & Bonas, U. (1996). HrpXv, an AraC-type regulator,

activates expression of five of the six loci in the hrp cluster of Xanthomonas campestris pv. vesicatoria. J Bacteriol 178, 3462–3469. Yamashoji, S., Manome, I. & Ikedo, M. (2001). Menadione-catalyzed

Teitzel, G. M. & Parsek, M. R. (2003). Heavy metal resistance of

O2- production by Escherichia coli cells: application of rapid chemiluminescent assay to antimicrobial susceptibility testing. Microbiol Immunol 45, 333–340.

biofilm and planktonic Pseudomonas aeruginosa. Appl Environ Microbiol 69, 2313–2320.

Edited by: N. L. Brun and others

466

Microbiology 160