Possible roles of reactive chlorine II: assessing biotic chlorination as a way for organisms to handle oxygen stress

Authors

  • Per Bengtson,

    1. Department of Biology – Microbial Ecology, Lund University, The Ecology Building, Lund SE-223 62, Sweden
    Search for more papers by this author
  • David Bastviken,

    1. Water and Environmental Studies, Linköping University, Linköping SE-581 83, Sweden
    Search for more papers by this author
  • Gunilla Öberg

    Corresponding author
    1. Institute for Resources, Environment and Sustainability, University of British Columbia, 429-2202 Main Mall, Vancouver, British Columbia, Canada V6T 1Z4
      E-mail goberg@ires.ubc.ca; Tel. (+1) 604 306 9900; Fax (+1) 604 822 9250.
    Search for more papers by this author

E-mail goberg@ires.ubc.ca; Tel. (+1) 604 306 9900; Fax (+1) 604 822 9250.

Summary

Natural formation of organically bound chlorine is extensive in many environments. The enzymes associated with the formation of chlorinated organic matter are produced by a large variety of organisms. Little is known about the ecological role of the process, the key question being: why do microorganisms promote chlorination of organic matter? In a recent paper we discuss whether organic matter chlorination may be a result of antagonistic interactions among microorganisms. In the present paper we evaluate whether extracellular microbial formation of reactive chlorine may be used as a defence against oxygen stress, and we discuss whether this process is likely to contribute to the formation of chlorinated organic matter. Our analysis suggests that periodic exposure to elevated concentrations of reactive oxygen species is a common denominator among the multitude of organisms that are able to enzymatically catalyse formation of reactive chlorine. There is also some evidence suggesting that the production of such enzymes in algae and bacteria is induced by oxygen stress. The relative contribution from this process to the extensive formation of chlorinated organic matter in natural environments remains to be empirically assessed.

Introduction

Chlorine mainly occurs in the environment as chloride (Cl-), which has traditionally been viewed as a non-reactive species. During the last few decades, this view has gradually been revised, and it is now generally accepted that chlorine takes part in a complex biogeochemical cycle, although this is not yet textbook knowledge. In fact, evidence suggests that formation of chlorinated organic matter is quite extensive and ubiquitous both in soil and on decaying plant material (Öberg and Sanden, 2005; Ortiz-Bermudez et al., 2007; Bastviken et al., 2009; Clarke et al., 2009), and an immense diversity of natural chlorinated organic compounds has been described (Gribble, 1992; 2004; Gribble, 2003; Vaillancourt et al., 2006). In boreal forest soils, and possibly other locations as well, the amount of organically bound chlorine may exceed the amount of inorganic chlorine (chloride; Cl-) as a consequence of the extensive organic matter chlorination (Johansson et al. 2003).

Enzymes believed to be responsible for the formation of chlorinated organic matter are produced by a large variety of organisms (Clutterbuck et al., 1940; Bengtson et al., 2009; Hofrichter et al., 2010). Chloroperoxidases is one group of commonly occurring enzymes that is capable of catalysing organic matter chlorination. Three different classes of chloroperoxidases have been identified: haem-dependent, no haem vanadium-dependent and a few bacterial chloroperoxidases with no prosthetic group (van Pee, 2001; van Pee and Unversucht, 2003; Winter and Moore, 2009; Hofrichter et al., 2010). Chloroperoxidase presumably produce reactive chlorine at the active site of the enzyme (Reaction 1), which then reacts with surrounding organic substrates to produce chlorinated organic compounds according to Reaction 2 (Murphy, 2003; van Pee and Unversucht, 2003; Manoj, 2006):

image((Reaction 1))
image((Reaction 2))

where A is an organic nucleophilic acceptor. HOCl can also react spontaneously with H2O2 to produce water and singlet oxygen according to Reaction 3 (Renirie et al., 2003):

image((Reaction 3))

Little is known about the ecological role of the extensive chlorine cycling. That is, why do microorganisms promote chlorination of organic matter? Several explanations have been proposed, one such is that reactive chlorine is used to depolymerize lignocellulose and other complex compounds in order to access carbon (Öberg et al., 1996; 1997; Ortiz-Bermudez et al., 2003; 2007). This line of reasoning is based on the assumption that carbon is a limited resource in soil (Demoling et al., 2007) and production of reactive chlorine species would thus increase its availability. Another line of reasoning suggests that the chlorination of organic compounds serves as a chemical defence against, e.g. antibiotics, while Barnett and colleagues (1997) has proposed that HOCl is produced by plant pathogens/symbionts to dissolve the cell wall.

In a recent paper we assess the hypothesis that that the ability to produce reactive chlorine has evolved as a strategy to compete with other microorganisms for limited resources by means of antagonistic interactions among microorganisms (Bengtson et al., 2009). In the present paper we evaluate another proposed hypothesis: that extracellular microbial formation of reactive chlorine is used as a defence against O2 stress, and we discuss whether this process is likely to contribute to the formation of chlorinated organic matter.

Method

The study was conducted as a critical literature review of relevant articles published in peer-reviewed journals. Microorganisms producing chloroperoxidases were identified in Peroxibase (Passardi et al., 2007), where all sequences have been manually annotated and checked. The chloroperoxidases belong to a group of enzymes called ‘haloperoxidases’, which are divided into chloroperoxidases, bromoperoxidases and iodoperoxidases according to the most electronegative halide ion they can oxidize. In our literature searches, we searched for chloroperoxidases, bromoperoxidases and iodoperoxidases, as well as haloperoxidases. It was, however, not always clear if a study dealing with ‘haloperoxidases’ dealt with chloroperoxidases or bromoperoxidases, and it seems as if chloroperoxidases at times has been reported as bromoperoxidase. When this was the case, we use the term ‘haloperoxidase’ and ‘halogenation’ to denote these organisms. The focus of the present paper is chlorination, but due to mentioned limitations in the literature, we at times extend the discussion to halogenation.

We also made a search in UniProtKB, and identified both reviewed and non-reviewed sequences, as well as the organisms in which the sequences occurred (2011-05-23, query: (name:chloroperoxidase) OR (name:haloperoxidase) OR (name:bromoperoxidase) OR (name:iodoperoxidase) OR (name:chloride peroxidase) OR (name:bromide peroxidase). The findings were analysed in the light of the following three questions:

  • i. Are chloroperoxidases primarily produced by organisms living in environments where they are (at least periodically) exposed to high concentrations of reactive oxygen species (ROS)?
  • ii. Do high ROS levels induce the expression of genes encoding chloroperoxidase?
  • iii. Can the mechanism be extended to defend against extracellular ROS, hence contributing to organic matter chlorination that occurs outside of cells?

Background

Oxidative stress is an inevitable consequence of aerobic metabolism and arises from exposure to high concentrations of ROS. High levels of ROS are toxic against a diversity of cellular compounds necessary for cell functioning and division, such as DNA, RNA, proteins, lipids, etc. (Halliwell and Gutteridge, 1984; Schrader and Fahimi, 2006; Imlay, 2008). The intracellular response to ROS and oxidative stress is well orchestrated (reviewed by, e.g. Imlay, 2008; Latifi et al., 2009). For example, in bacteria the regulatory protein SoxRS generally controls the defence against O2- by inducing production of, e.g. superoxide dismutase (SOD), which catalyses the dismutation of O2- to O2 and H2O2 (Imlay, 2008; Latifi et al., 2009). The intracellular response to H2O2 is regulated by OxyR, and in some bacteria by PerR (Latifi et al., 2009). At low concentrations, peroxidases such as peroxiredoxin AhpCF are the most common intracellular H2O2 scavenger (Imlay, 2008). When concentrations increase, catalase is strongly induced and becomes the main scavenger.

In addition to the intracellular production of ROS, microorganisms in certain environments are also exposed to high extracellular concentrations of ROS. Since H2O2, unlike O2-, is an uncharged molecule that diffuses across cellular membranes, high extracellular concentrations of H2O2 will inevitably lead to intracellular H2O2 stress (Tavares et al., 2007; Imlay, 2008). Microorganisms inhabiting environments where extracellular H2O2 concentrations are periodically high would thus have a competitive advantage if they could detoxify H2O2 extracellularly as well as intracellularly. One such way would be to produce and release extracellular peroxidases, but most peroxidases use specific organic reductants to reduce H2O2 to H2O. Consequently, low concentrations or a lack of these reductants would result in inefficient removal and protection of organisms from H2O2 by peroxidases depending on specific organic substrates.

Proposed mechanism

Haloperoxidases catalyse the reaction between H2O2 and halide ions, and are thus different from peroxidases catalysing reactions with specific organic substrates. The discussion below applies to all haloperoxidases but we focus primarily on chloroperoxidases, which convert chloride, one of the earth's most common ions, to HOCl and H2O. HOCl reacts almost instantly with any oxidizable group, and thus, in contrast to H2O2, does not diffuse over longer distances (Ortiz-Bermudez et al., 2007).

In the case of extracellular chlorination, intracellular and extracellular HOCl concentrations should therefore not be proportional, since the extracellular formation of HOCl would transfer oxidizing power from H2O2 to reactive chlorine. The reactive chloride would in turn react with any organic compound present, which in soil is predominantly organic matter, thus forming chlorinated organic matter, which is considerably less toxic to cells than H2O2. Taken together, the highly reactive nature of HOCl, in combination with the abundance of a reductant (Cl-), makes the haloperoxidase system a plausible candidate for extracellular detoxification of H2O2 to prevent oxidative stress. Below follows an analysis and discussion of each of the research questions posed above.

i. Are chloroperoxidases primarily produced by organisms living in environments where they are (at least periodically) exposed to high concentrations of ROS?

As further developed below, the search in Peroxibase and Uniprot revealed that most identified haloperoxidases are found in organisms living in environments where they are likely to be exposed to high concentrations of ROS (Tables 1–3 and Table S1). Haem-dependent haloperoxidases are mainly found in ascomycetes, but also in a few basidiomycetes and in the oomycete Phytophthora infestans (Table 1). A common denominator for these organisms seems to be that they are either plant pathogens, symbionts or moulds growing in the phyllosphere. Haem-dependent haloperoxidases are also found in Metarhizium species, a genus of entomopathogenic ascomycetes.

Table 1. Haem-dependent haloperoxidases reported in Peroxibase.
NameTaxonomic groupOrganismEcology
AfumHalPrx01 Ascomycota Aspergillus fumigatus An opportunistic pathogen found in almost all oxygen-rich environments.
AniHalPrx01 Ascomycota Aspergillus nidulans See above.
AorHalPrx01 Ascomycota Aspergillus oryzae See above.
AteHalPrx02 Ascomycota Aspergillus terreus See above.
CfuHalPrx Ascomycota Caldariomyces fumago Sooty mould colonizing living plant surface habitats.
CgHalPrx01 Ascomycota Chaetomium globosum Yeast normally found in soil, air and plant debris. Also live endophytically in plants and shows antagonistic effects against fungal plant pathogens.
GzHalPrx01 Ascomycota Gibberella zeae A plant pathogen causing the wheat headblight disease.
HjHalPrx01 Ascomycota Trichoderma reesei Common in soil and root environments. Opportunistic, avirulent plant symbionts. Trichoderma spp. are know to produce a wide variety of antibiotics and can parasitize other fungi.
MpiHalPrx01 Ascomycota Mycosphaerella pini Parasitic, causing needle blight on coniferous trees.
NcHalPrx01 Ascomycota Neurospora crassa Red bread mould. Occur naturally mainly on dead plant matter after fires in tropical and sub-tropical regions.
PnoHalPrx01 Ascomycota Phaeosphaeria nodorum Plant pathogen that causes Septoria diseases.
PanHalPrx01 Ascomycota Podospora anserina Saprophytic fungi on herbivore dung.
AbHalPrx01 Basidiomycota Agaricus bisporus Saprophytic fungi growing on e.g. litter and compost.
CcinHalPrx01 Basidiomycota Coprinopsis cinerea (Coprinus cinereus) Saprophytic fungi growing on dung.
LbiHalPrx01 Basidiomycota Laccaria bicolor Forms ectomycorrhizal associations with a wide variety of tree species.
PinvHalPrx Basidiomycota Paxillus involutus Forms ectomycorrhizal associations with a wide variety of tree species.
PcHalPrx01 Basidiomycota Phanerochaete chrysosporium A lignin degrading white rot fungi.
PplHalPrx01 Basidiomycota Postia placenta Brown rot fungi commonly found in forest ecosystems. A major cause of wood decay.
UmHalPrx01 Basidiomycota Ustilago maydis Pathogenic fungi causing corn smut on maize.
PiHalPrx01Other Stramenopiles Phytophthora infestans Pathogenic oomycete causing potato blight.
Table 2. No haem, no metal haloperoxidases reported in Peroxibase.
NameTaxonomic groupOrganismEcology
MvaHalNPrx Actinobacteria Mycobacterium vanbaalenii Commonly found in contaminated soils and sediments.
RerHalNPrx Actinobacteria Rhodococcus erythropolis An opportunistic pathogen that also has the ability to tolerate and degrade organic pollutants.
STaHalNPrx01 Actinobacteria Streptomyces aureofaciens Root associated bacteria producing anti fungal compounds.
ScoHalNPrx Actinobacteria Streptomyces coelicolor Rhizosphere bacteria that also colonize some aquatic plants.
SliHalNPrx Actinobacteria Streptomyces lividans Root associated bacteria.
GfoHalNPrx Bacteroidetes Gramella forsetii Found on marine snow. Degrade high molecular weight compounds in both the dissolved and particulate fraction of marine organic matter.
MmarHalNPrx Bacteroidetes Microscilla marina Found on marine snow. Degrade high molecular weight compounds in both the dissolved and particulate fraction of marine organic matter.
PtorHalNPrx Bacteroidetes Psychroflexus torquis A psychrophilic bacterium found in sea ice.
SYspHalNPrx Cyanobacteria Synechocystis sp. Cyanobacteria capable of phototrophic growth as well as heterotrophy.
LacHalNPrx Firmicutes Lactobacillus acidophilus A lactic acid bacterium that occurs naturally in the human and animal mouth, vagina and gastrointestinal tract.
LplHalNPrx01 Firmicutes Lactobacillus plantarum Lactic acid bacterium found in vegetables and saliva.
PpenHalNPrx Firmicutes Pediococcus pentosaceus Lactic acid bacterium found in vegetables.
WpaHalNPrx Firmicutes Weissella paramesenteroides Lactic acid bacterium found in vegetables.
BjaHalNPrx01 Alphaproteobacteria Bradyrhizobium japonicum Nitrogen fixing root nodule forming bacteria.
BRspHalNPrx Alphaproteobacteria Bradyrhizobium sp.Nitrogen fixing root nodule forming bacteria.
BcHalNPrx Betaproteobacteria Burkholderia cepacia A soil bacterium responsible for bulbiferous Aliaceae root rot diseases. Also a human pathogen.
PpyrHalNPrx Betaproteobacteria Pseudomonas pyrrocinia A rhizoplane associated bacteria possibly belonging to the genus Burkholderia.
PaerHalNPrx_PAO1 Gammaproteobacteria Pseudomonas aeruginosa A rhizosphere bacteria producing a wide range of anti fungal compounds.
PfHalNPrx01_Pf5 Gammaproteobacteria Pseudomonas fluorescens Root colonizing bacteria producing a wide range of anti fungal compounds.
BdeHalNPrx Chytridiomycota Batrachochytrium dendrobatidis A pathogenic fungus causing chytridiomycosis in amphibians.
Table 3. No haem vanadium-dependent haloperoxidases reported in peroxibase.
NameTaxonomic groupOrganismEcology
StrVBPo Actinobacteria Salinispora tropica A marine bacterium sediment dwelling bacteria producing a wide range of structurally unique secondary metabolites.
BRspVCPo Alphaproteobacteria Bradyrhizobium sp.Nitrogen fixing root nodule forming bacteria.
RpVCPo Alphaproteobacteria Rhodopseudomonas palustris Phototrophic bacteria with a wide very range of metabolic capabilities.
CinaVCPo Ascomycota Curvularia inaequalis A plant pathogenic fungi that causes Curvularia leaf spot disease in maize.
EdVCPo Ascomycota Embellisia didymospora  
MagVCPo Ascomycota Magnaporthe grisea (Pyricularia grisea) Plant pathogenic fungi that causes blight disease in e.g. rice and grains.
PnoVCPo Ascomycota Phaeosphaeria nodorum Plant pathogenic fungi that causes Septoria glume blotch in wheat.
PtritVCPo Ascomycota Pyrenophora tritici-repentis Plant pathogenic fungi that causes tan spot disease in wheat.
CoVBPo Corallina Corallina officinalis A red seaweed inhabiting the lower and mid-littoral zones.
CpiVBPo01 Corallina Corallina pilulifera A red seaweed found in the sub-littoral zone. Allelopathic against red tide microalgae.
CcriVBPo Florideophyceae Chondrus crispus A widely distributed marine red algae found in the inter-tidal to sub-littoral zones.
GcVBPo Florideophyceae Gracilaria changii An agar producing red algae found in mangrove swamps.
AnoVBPo Phaeophyceae Ascophyllum nodosum A brown algae found in the intertidal and mid-littoral zones.
FdVBPo Phaeophyceae Fucus distichus A widely distributed brown algae found in the intertidal zone.
LdVBPo01 Phaeophyceae Laminaria digitata A brown algae found in the lower littoral zone.
AvaVBPo Cyanobacteria Anabaena variabilis An N-fixing phototrophic cyanobacteria also capable of heterotrophy. Forms symbiotic associations with plants and fungi.
CwaVBPo Cyanobacteria Crocosphaera watsonii An N-fixing phototrophic cyanobacteria found in warm marine waters.
NspuVBPo Cyanobacteria Nodularia spumigena An N-fixing, UV-tolerant phototrophic cyanobacteria found in brackish and marine surface waters where it can form intense blooms.
NOspVCPo Cyanobacteria Nostoc sp.A genus of cyanobacteria found in a wide range of habitats. Also forms symbiotic relationships with plants and fungi.
SspVBPo02_CC9311 Cyanobacteria Synechococcus sp.Marine photoautotrophic bacteria.
CboVBPo_TypeA Firmicutes Clostridium botulinum Anaerobic soil bacteria.
RbaVCPoOther Bacteria Rhodopirellula baltica A widely distributed marine bacterium found associated with phytoplankton blooms and on macroscopic organic particles in the photic zone.
SuVBPoOther Bacteria Solibacter usitatus An aerobic Acidobacteria found in soil.

The no haem no vanadium haloperoxidases seem to be widespread among plant-associated microorganisms, such as Streptomyces and Pseudomonas spp. (commonly found in the rhizosphere/rhizoplane), and endosymbiotic or parasitic Bradyrhizobium, Burkholderia, Cupriavidus, Frankia, Rhizobium spp. (Table 2). No haem no vanadium haloperoxidases are also produced by several Agrobacterium, Clostridium, Mycobacterium and Xanthomonas spp., all of which are known to cause disease in plants and mammals. Another bacterial phylum producing no haem no vanadium haloperoxidase, Bacteriodetes, are commonly found as free living assemblages in nutrient-rich microenvironments associated with phytoplankton blooms, and on macroscopic organic particles (marine snow) formed in the photic zone (Bauer et al., 2006). Bacteria that thrive in polluted environments and are able to tolerate high concentrations of organic pollutants, metals and oxidative stress, e.g. Arthrobacter, Deinococcus, Ralstonia and Rhodoccocus spp., also seem to have the ability to produce no haem no vanadium haloperoxidase. Finally, several Lactobacillus spp., whose metabolism results in self-induced creation of a high ROS environment, have the ability to produce no haem no vanadium haloperoxidase.

Vanadium-dependent haloperoxidases are mainly produced by red and brown algae and plant pathogenic fungi, but also by cyanobacteria and a few other phototrophic bacteria such as Citreicella and Rhodopseudomonas palustris (Table 3). Other bacteria-producing vanadium-dependent haloperoxidases include Deinococcus radiodurans, which has an unrivalled ability to overcome oxidative stress (Slade and Radman, 2011), and Bacillus mycoides, a soil bacterium that also can induce an oxidative burst, leading to systemic resistance against pest in sugar beets, without causing plant cell death or tissue necrosis (Bargabus et al., 2002; 2003). Vanadium-dependent haloperoxidases are also produced by opportunistic pathogens (e.g. Flavobacteria), plant endophytes (e.g. Bradyrhizobium and Dyadobacter fermentans) and the bacteria Rhodopirellula baltica and Gramella forsetii, which similarly to many Bacteriodetes are found associated with phytoplankton blooms and on macroscopic organic particles formed in the photic zone (Gade et al., 2005).

Based on the observations above and the information in Tables 1–3 and Table S1, we conclude that many of the haloperoxidase producing organisms can be placed into one of the following five groups/niches:

  • (i) microorganisms associated with plants and animals (e.g. rhizosphere/rhizoplane associated microorganisms, endophytes/endosymbionts and pathogens/parasites);
  • (ii) photoautotrophic organisms such as algae, cyanobacteria and brown seaweed;
  • (iii) bacteria associated with marine snow;
  • (iv) microorganisms that thrive in polluted environments; and
  • (v) microorganisms that create high concentrations of ROS in their own environment.

Organisms inhabiting these environments are at least periodically exposed to high concentrations of ROS. For example, plants and animals react to microbial invasion with a respiratory burst that produces H2O2 (Doke et al., 1991; Jacks and Davidonis, 1996; Nauseef, 2007). In the first group we find opportunistic pathogens as well as endosymbiotic and endopathogenic plant associated microorganisms that are severely exposed to ROS, especially H2O2, during colonization of their host (Tavares et al., 2007). In fact, there is increasing evidence that ROS are needed for successful colonization and establishment of the legumes–rhizobia symbiosis (Pauly et al., 2006; Kopcinska, 2009). Furthermore, leaf-associated microorganisms are exposed to high levels of UV radiation and ROS (Liu et al., 2000; Lindow and Leveau, 2002; Cohen and Yamasaki, 2003; Lindow and Brandl, 2003), which might explain why early studies on microbial chlorination revealed that microorganisms growing on plants, such as Caldariomyces fumago, produce chlorinated organic compounds at especially high rates (Clutterbuck et al., 1940). In the rhizosphere, microbial reduction of ferric iron to Fe(II) induces ROS production via the Fenton reaction (Halliwell and Gutteridge, 1984; Laturnus et al., 2005). The rhizosphere is also an environment where antagonistic interactions and the ability to produce antibiotics are common (Hassett and Imlay, 2007).

The second group, photoautotrophic microorganisms, are heavily exposed to ROS as a result of high intracellular production of ROS during photosynthesis (Ashur et al., 2009; Latifi et al., 2009). The same is true for the third group, bacteria associated with phytoplankton blooms and marine snow in the photic zone of oceans, which are exposed to high concentrations of ROS due to photochemical generation of ROS from dissolved organic matter (Zepp et al., 1977; Cooper and Zika, 1983; Cory et al., 2010).

The fourth group is tied to contaminated environments, which contain a complex mixture of chemicals that induce the formation of ROS. For example, the genotoxicity of polycyclic aromatic hydrocarbons seems to be the end-result of incomplete degradation resulting in the production of ROS (Park et al., 2008). A second cause of oxidative stress in contaminated environments is metal-mediated formation of ROS (Hrimpeng et al., 2006; Park et al., 2008; Rico et al., 2009).

The fifth group, finally, is haloperoxidase producing organisms, such as the lactic acid bacteria, which are also exposed to high levels of ROS because of their lack of haem, which leads to a self-induced creation of a high ROS environment: The lack of haem means that they use flavoproteins rather than the haem-dependent cytochrome system for terminal oxidations (Klebanoff, 2005). Oxygen is thus converted to hydrogen peroxide rather than to water as in most organisms. The lack of haem also means that these organisms do not possess the catalase system that is normally used for scavenging accumulating hydrogen peroxide (Barloy-Hubler et al., 2004; Hansen et al., 2004). Haloperoxidases are also produced by radiation-resistant and halophilic bacteria and archaea (Table S1), which might explain their effective scavenging of ROS produced by radiolysis of water (Roh et al., 2007; Kish et al., 2009).

The ability to produce chloroperoxidases (and thus the potential capability of inducing the formation of chlorinated organic matter) seems to occur mainly in organisms being exposed to high concentrations of ROS either intracellularly or extracellularly. Oxidative stress might therefore be a trigger for evolving chlorinating enzymes. For at least some of the environments mentioned above there is a correspondence between high abundance of ROS and chlorinating activity. For example, high rates of production of volatile chlorinated organic compounds have been observed in the marine environments (Palmer et al., 2005; Leblanc et al., 2006), and as mentioned above microorganisms growing on plants also produce chlorinated organic compounds at especially high rates (Clutterbuck et al., 1940). We have not been able to identify any studies addressing if chlorinating activity are enhanced in polluted environments, or if invasion of plants and animals by pathogens and endosymbionts result in a chloroperoxidase mediated production of chlorinated organic compounds. There are, however, indications that formation of HOCl by myeloperoxidase, a functional analogue of chloroperoxidase, accounts for a high proportion of the reactive oxygen consumed in the oxidative burst in mammal defence systems (Klebanoff, 2005).

ii. Do high ROS levels induce the expression of genes encoding chloroperoxidase?

Evidence regarding the expression of genes encoding chloroperoxidases has recently been made available. A microarray analysis of genes expressed in response to hydrogen peroxide exposure demonstrates that the Sinorhizobium meliloti gene smc01944, which encodes a no haem chloroperoxidase, is strongly induced under oxidative stress (Barloy-Hubler et al., 2004). Sinorhizobium meliloti is a soil bacterium that forms a symbiotic relationship with host plants by infecting their roots and forming nitrogen fixing root nodules. During the infection stage the plant respond with a strong oxidative burst which produces ROS, mainly hydrogen peroxide. It seems likely that the induction of smc01944 is a means to defend against the high levels of this ROS. The amount of smc01944 mRNA in the study increased 50-fold in response to hydrogen peroxide, while that of katA (encoding a catalase) only increased 10-fold.

In contrast to catalase, the chloroperoxidase Smc01944 is excreted by the cells (Barloy-Hubler et al., 2004) and could, thus, be effective in detoxifying ROS both within cells and in the external medium, which might explain the high levels of smc01944 mRNA compared with katA mRNA. The smc01944 gene encodes a non-haem chloroperoxidase that has also been found in, e.g. Pseudomonas flourescens.

Furthermore, the widespread ability to produce brominated and iodinated organic compounds among brown macrophytic algae is believed to play a central role in oxidative detoxification (La Barre et al., 2006). Expressed sequence tag (EST) analysis of genes expressed in the brown algae Laminaria digitata has revealed that a family of haloperoxidase genes encoding for vanadium bromoperoxidase (vBPO) is highly expressed under oxidative stress (Roeder et al., 2005). Interestingly, the previously unidentified gene family vBPO-II was specifically induced in protoplast cells, which must cope with high levels of toxic ROS during experimental isolation, suggesting that vBPO-II proteins play an active role as ROS scavenging enzymes. EST analysis of the unrelated red algae, Chondrus crispus, produced similar induction of vBPO genes. vBPO production is also induced by copper stress in the brown alga Ectocarpus siliculosus (Ritter et al., 2010). Copper induces the production of ROS, and it seems as if vBPO plays an important role in the detoxification of ROS (Kupper et al., 2008; Ritter et al., 2010). A second marine red algae, Gracilaria changii, is known to produce both vanadium chloroperoxidase (vCPO) and vBPO as a response to stress (Teo et al., 2007), while kelps produce iodoperoxidas (Palmer et al., 2005; Leblanc et al., 2006). Kelps also seem to produce vBPO (e.g. Saccharina japonica) and the production is increased during summer, possibly as a defence against ROS (Yotsukura et al., 2010).

Taken together, these observations support the hypothesis that high ROS levels induce the expression of genes encoding haloperoxidases, and the mechanisms seem to occur in several phyla. We have not been able to find any studies that explicitly address the relative importance of the proposed mechanism to detoxify ROS compared with known pathways. However, as mentioned above the formation of HOCl by myeloperoxidase accounts for a high proportion of the reactive oxygen consumed in the oxidative burst in mammal defence systems (Klebanoff, 2005). Other and more direct evidence includes observations that haloperoxidase activity is synchronized with Superoxide dismutase activity, but not with catalase activity, in the brown seaweed Corallina pilufiera. While superoxide dismutase produces H2O2, BPO functions to eliminate H2O2 and thus compensates for the lack of catalase (Ohsawa et al., 2001). Further evidence consists of observations that katA mutants of Sinorhizobium meliloti are still successful in forming root nodules, despite lacking the ability to produce catalase to protect themselves against exposure to ROS during formation of the root nodules. A possible reason is that haloperoxidase production is strongly enhanced in response to the elevated ROS concentrations (Barloy-Hubler et al., 2004).

iii. Can the mechanism be extended to defence against extracellular ROS?

Most studies on chloroperoxidases are based on isolated enzymes from intact cells, and literature on studies of extra cellular levels is limited. However, organisms from at least three different taxa are known to produce haloperoxidases that are excreted and remain active in the extracellular medium. Already in 1940, Clutterbuck and colleagues (1940) demonstrated that plant moulds such as Caldariomyces fumago excrete high levels of a chlorinating enzyme. It has later been confirmed that the enzyme is a chloroperoxidase (Conesa et al., 2001). The marine fungi Curvularia inaequalis excrete vanadium chloroperoxidase into the extra cellular medium during the iodophase of growth, and the phenomenon has also been demonstrated for several plant pathogenic ascomycetes (Barnett et al., 1997; Ortiz-Bermudez et al., 2007). The haloperoxidase smc01944, which is produced in response to H2O2 stress by Sinorhizobium meliloti, is also excreted (Barloy-Hubler et al., 2004). Marine algae are the third taxa known to produce and release haloperoxidases to the external medium (Roeder et al., 2005). The ability to exude haloperoxidases might therefore be a rather common feature, even though this remains to be investigated.

Is it reasonable to think that the proposed mechanism is relevant compared with known pathways to detoxify ROS?

We have not been able to find any studies that explicitly address this hypothesis. However, although direct evidence regarding the importance of the proposed mechanism to detoxify ROS is lacking, there are several interesting indications. For example, a high proportion of the reactive oxygen consumed in the oxidative burst in mammal defence systems can be assigned to myeloperoxidase activity (Klebanoff, 2005). Further, superoxide dismutase activity in the brown seaweed Corallina pilufiera is synchronized with haloperoxidase activity, but not with catalase activity, suggesting that H2O2 produced by superoxide dismutase is removed by haloperoxidases (in this case BPO) rather than by catalase (Ohsawa et al., 2001). The hypothesis also finds support in the above-mentioned observations that katA mutants of Sinorhizobium meliloti are able to cope with ROS and successfully form root nodules. This co-occurs with a strongly enhanced haloperoxidase production, although the causal connection remains to be established (Barloy-Hubler et al., 2004).

Conclusion

In the present paper we have evaluated the hypothesis that chloroperoxidases are produced by some organisms to defend against O2 stress, and we have discussed if this may contribute to organic matter chlorination that occurs outside of cells. Our analysis suggests that it is by no means unlikely that chloroperoxidases play a role in the detoxification of ROS, causing subsequent formation of chlorinated organic matter. A multitude of organisms belonging to several different taxa have the ability to produce haloperoxidases. Our analysis suggests that periodic exposure to high concentrations of ROS is a common denominator among the multitude of organisms that are able to enzymatically catalyse formation of reactive chlorine, and there is evidence suggesting that the production of haloperoxidase is induced by oxygen stress in both algae and bacteria. Our study also demonstrates that chloroperoxidases produced by these organisms are often exuded, and are therefore likely to contribute to the extensive formation of chlorinated organic matter that occurs in terrestrial as well as marine environments. However, it remains to be tested as to whether these conclusions can be extended to include the majority of the organisms producing this type of enzymes. Calculations in our previous study (Bengtson et al., 2009) suggest that soil concentrations of chlorinating enzymes (e.g. chloroperoxidase) are high enough to fully account for the chlorination of organic matter in soil. Also, our previous study suggests that soil organic matter chlorination is more likely to be limited by the supply of H2O2, rather than by the presence of chlorinating enzymes. The relative contribution from this process to the extensive formation of chlorinated organic matter in natural environments remains, however, to be empirically assessed.

Acknowledgements

This work was sponsored by the Swedish Research Council, the Natural Sciences and Engineering Research Council of Canada and the Canada Foundation for Innovation.

Ancillary