An unusual dehalogenating peroxidase from the marine terebellid polychaete Amphitrite ornata.

The terebellid polychaete Amphitrite ornata produces no detectable volatile halogenated secondary metabolites, but frequently inhabits coastal marine sediments heavily contaminated with anthropogenic or biogenic haloaromatic compounds. This animal contains high levels of two very unusual enzymes, dehalogenating peroxidases. We have purified and partially characterized one of these dehaloperoxidases, DHP I. DHP I is a heme enzyme (Mr = 30,790) composed of two identical subunits (Mr = 15,529) and is very rich in the amino acids aspartic acid (+ asparagine) and glutamic acid (+ glutamine). The enzyme converts trihalogenated phenols, such as 2,4,6-tribromophenol, into dihalogenated quinones. The optimum pH for this reaction is 5.0. DHP I is also active against di- and monohalogenated phenols and will oxidize bromo-, chloro-, and fluorophenols. We have identified similar dehaloperoxidase activities in other infaunal polychaetes, including halometabolite-producing species.

Contamination of coastal marine sediments by anthropogenic haloaromatic compounds found in agricultural, industrial, and urban runoff is well known (1,2). Less recognized, but common and very widespread sources of haloaromatics are biotic (3)(4)(5). Among these are species of sediment-dwelling marine polychaetes and hemichordates which produce high levels of volatile brominated secondary metabolites, such as bromophenols, bromopyrroles, bromoindoles, bromohydroquinones, and bromobenzylalcohols (5)(6)(7)(8), through the action of haloperoxidases (9,10). Such worms occupy immense areas of coastal sediments (11)(12)(13)(14)(15), contaminating them with volatile, malodorous, and toxic bromometabolites. Bromophenols and related compounds are respiratory inhibitors and presumably express their toxicity through inhibition of mitochondrial function. The toxicity of these compounds to vertebrates is well established (16 -19), but has not been as thoroughly examined in invertebrates. It is clear from field and laboratory studies that sediment contamination with bromoaromatics inhibits recruitment of non-bromometabolite-producing invertebrate species (15) and appears to select for specific organisms which may be resistant to the toxic effects of haloaromatic compounds. Both these resistant species and the bromometabolite producers themselves, which face bromometabolite autotoxicity, must have some means of detoxifying these compounds. Dehalogenating enzymes provide one such detoxification mechanism.
Dehalogenating enzymes are uncommon in higher organisms. The cytochrome P-450 enzymes are exceptions, being broadly distributed among animals, plants, fungi, and bacteria and capable of reductive dehalogenation of alkyl halocarbons to the corresponding alkanes under anaerobic conditions (20). Other oxygenases may also participate in similar reactions, but haloaromatic compounds present a unique problem due to their relative stability and toxicity. The terebellid polychaete Amphitrite ornata produces no detectable volatile halometabolites of its own 1 and lacks detectable halogenase activity. However, this animal is often found in beds also inhabited by Notomastus lobatus (Polychaeta) and Saccoglossus kowalewskyi (Hemichordata), which produce and contaminate sediments with bromophenols and bromopyrroles (8,9,13,21). We examined A. ornata to determine the basis for its tolerance of haloaromatic compounds and found that, instead of a typical oxygenase, it produces high levels of two dehalogenating peroxidases. We have purified and partially characterized dehaloperoxidase I (DHP I), 2 which has a number of unusual properties.

EXPERIMENTAL PROCEDURES
Materials-A. ornata was collected at low tide from mixed oyster rubble and sandy mud at Debidue flats, in the North Inlet estuary (Georgetown, SC, 33°20ЈN, 79°10ЈW) and immediately frozen on dry ice. Worms could be stored frozen at Ϫ70°C for at least 6 months without substantial loss of dehaloperoxidase activity, but recently collected material was used in most purifications.
Dehaloperoxidase Activity, Products, and Protein Assays-A. ornata dehaloperoxidase I activity was assayed on the basis of disappearance of substrate, typically 2,4,6-tribromophenol. Activity was assayed using a reaction mixture containing 50 mM KH 2 PO 4 , pH 5.0, 5 mM H 2 O 2 , from 1 to 20 M halophenol substrate, and 0.15 g of pure DHP I in a 1-ml reaction volume, for 5 min at room temperature. An internal standard (2,6-dichlorophenol) was then added, and the reaction was stopped immediately by extraction with high performance liquid chromatography grade pentane. Substrate remaining after incubation was resolved via gas chromatography and quantified using standard curves derived from known concentrations of reagent grade substrate halophenols and of the internal standard (9,10). Product identities were determined via mass spectroscopy (8,9). Halophenol disappearance was linear with time and with enzyme quantity. The DHP I reaction achieved maximum velocity at 1.5 mM H 2 O 2 and 20 M 2,4,6-tribromophenol. Saturating concentrations of chlorophenols and fluorophenols were lower. Optimum pH for enzyme activity was determined using reaction mixtures buffered to different pH values with 100 mM sodium acetate buffer (pH 3.8 -6.0) or 50 mM potassium phosphate buffer (pH 5.0 -7.0). One unit of dehaloperoxidase activity was defined as 1 nmol of 2,4,6-tribromophenol debrominated per min. Specific activities are reported as units/mg of protein. Protein was determined using the method of Lowry et al. (22) with bovine serum albumin as a standard.
Preparation of A. ornata Extract and Purification of Dehaloperoxidase-Frozen A. ornata tissue was thawed at room temperature and homogenized by hand grinding with washed sea sand using an ice-cold mortar and pestle. The homogenate was diluted with 50 mM sodium phosphate buffer, pH 5.0, and filtered through cheesecloth. All subse-* This work was supported by National Science Foundation Grant OCE-9201857 and the Environmental Protection Agency EPSCoR Program Grant R821838-01-0. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ To whom correspondence should be addressed. Tel.: 803-777-7036; Fax: 803-777-4002; E-mail: lovell@biol.scarolina.edu. quent steps were performed either on ice or at 4°C. The crude extract was centrifuged to remove debris and loaded directly onto a DE52 DEAE-cellulose (Whatman Labsales, Hillsboro, OR) column pre-equilibrated with 50 mM sodium phosphate buffer, pH 5.0 (PB). Proteins were eluted with a linear gradient of NaCl from 0.0 to 0.5 M in 300 ml of PB. Fractions containing the Peak 1 dehaloperoxidase (DHP I) were pooled and concentrated by ultrafiltration, and the proteins were separated using a Bio-Gel A-0.5m (Bio-Rad) column with PB as the mobile phase. Fractions containing peak DHP I activity were again pooled and concentrated, loaded onto a Sephacryl S-300 (Sigma) column, and separated using PB as the mobile phase. The DHP I recovered was homogeneous by native and SDS-polyacrylamide gel electrophoresis (Ref. 23 and Fig. 1).
Gel Electrophoresis and Molecular Weight Determination-The molecular weight of the native DHP I was determined from gel permeation chromatography using a Sephacryl S-300 column (1.2 ϫ 57 cm) calibrated with proteins of known molecular weight. Electrospray mass spectroscopy was used to determine subunit molecular weight. SDSpolyacrylamide gel electrophoresis with proteins of known molecular weight (Bio-Rad Low Range Molecular Weight Marker Set) was also used in estimating subunit molecular weight. A typical SDS-polyacrylamide gel is shown in Fig. 1.
Determination of Metal Content, Amino Acid Composition, and Absorption Spectra-Metal content of DHP I was determined by atomic absorption spectroscopy (Model 965 Plasma Atomcomp; Jarrell-Ash Co., Waltham, MA). Amino acid composition was determined from pure fractions by hydrolysis in 6 N HCl at 110°C for 24 h under N 2 , followed by analysis with a Beckman System 6300 amino acid analyzer (Beckman Instruments). Absorption spectra of native and pyridine-modified DHP I were recorded in PB.

RESULTS AND DISCUSSION
Dehaloperoxidases are abundant in A. ornata, accounting for roughly 3% of soluble protein recovered in crude extracts (Table  I). Initial chromatography steps provided a 10.1-fold purification of dehaloperoxidase I with loss of 38% of total DHP activity. This major activity loss is accounted for by separation of the two dehaloperoxidases present in crude extract. The two redcolored DHP peaks were always visible in the ion exchange chromatography fractions. Peak 1 eluted at about 25 ml (approximately 0.1 M NaCl), while Peak 2 eluted at about 76 ml (approximately 0.25 M NaCl). Peak 1 (DHP I) was selected for further purification. The purification procedure resulted in 31.1-fold purification and a homogeneous preparation of DHP I with recovery of about 25% of the starting enzyme activity ( Fig.  1 and Table I).
The pure DHP I native molecular weight was 30,790, with a subunit molecular weight of 15,529. These data support a holoenzyme composed of two identical subunits. The dehaloperoxidase is a heme-containing protein, as indicated by the Soret absorbance peak at 420 nm and strong secondary peaks at 540 nm and 575 nm, and by modification of this spectrum by reaction with pyridine (data not shown). DHP I also contains 1 mol of iron per mol of dimer, as determined by atomic absorption spectroscopy. This is consistent with a content of one heme group per dimeric holoenzyme molecule. No other metals were found in significant quantities. The amino acid composition of the enzyme is given in Table II with the composition of the N. lobatus chloroperoxidase provided for comparison. As is typical of many peroxidases (see Ref. 9), both enzymes are rich in aspartic acid (ϩ asparagine) and glutamic acid (ϩ glutamine). Interestingly, the sizes of the DHP I subunit (15,529) and the chloroperoxidase ␤ subunit (15,500) are very similar and much smaller than subunits of the many well characterized haloperoxidases from plants, fungi, and bacteria (9). The optimum pH for DHP I activity is 5.0, which is also optimal for the phenol bromination reaction catalyzed by the N. lobatus chloroperoxidase (9). The trihalophenols 2,4,6-tribromophenol and 2,4,6trichlorophenol are initially oxidized to dihalogenated quinone intermediates (Fig. 2). Degradation of these compounds continues to as yet unidentified products.
The A. ornata DHP I is very unusual in that it oxidizes a wide variety of halophenols, including mono-, di-, and trisubstituted halophenols with bromine, chlorine, or fluorine substituent groups (Table III). This is particularly interesting given the stability of carbon-chlorine and carbon-fluorine bonds in haloaromatic compounds. DHP I is the first enzyme from a higher organism shown to oxidize these bonds at high specific activities and the first peroxidase capable of oxidizing carbonfluorine bonds. As would be expected, bromophenols are oxidized at the highest rates of the halophenols tested. This is noteworthy since the great majority of known volatile halometabolites of infaunal worms are brominated aromatic compounds, and these are the compounds most abundant at sites inhabited by A. ornata (7). The enzyme also shows little preference for either degree or position of substitution and can oxidize monohalogenated phenols at rates similar to those for di-and trihalogenated compounds.
DHP I represents approximately 3% of the soluble protein in crude extracts of A. ornata, an animal which lacks detectable haloperoxidase activity. DHP II is present at a roughly equivalent level. The other two polychaetes in which we have detected dehalogenase activity, Thelepus crispus and N. lobatus (Table IV), both produce halometabolites which are released into surrounding sediments (9,13,15). In those species, dehalogenase activity is consistent with a role in prevention of  autotoxicity. Since A. ornata produces no detectable volatile halometabolites and lacks detectable halogenase activity, autotoxicity seems unlikely. It lives, however, in close proximity to N. lobatus and S. kowalewskyi which contaminate the sediments with bromophenols and bromopyrroles (8,13,21). A. ornata feeds on surface deposits with its tentacles and lives in a mudlined tube; its lifestyle therefore involves intimate contact with the sediments and associated contaminants (24). The range of toxic bromoaromatic compounds encountered by A. ornata requires a detoxification mechanism which can neutralize high levels of several haloaromatic compounds having dif-   ferent structures and degrees and positions of bromine substitution. Sediments contaminated by anthropogenic sources, such as sediments downstream from sulfate process pulp mills, can also contain high levels of a diversity of haloaromatic compounds including chlorophenols, chloroguaiacols, and chlorocatechols, all potentially toxic to sediment-dwelling invertebrates (2,25). The high rates of A. ornata DHP I activity and its broad substrate specificity are consistent with its proposed function in neutralization of environmental haloaromatic toxins. We hypothesize that production of dehaloperoxidases allows A. ornata to survive in locations contaminated with these toxic compounds and that similar animals inhabiting sediments contaminated with haloaromatic compounds from both anthropogenic and biogenic sources will also produce these unusual dehaloperoxidases. Further structural and catalytic investigations of the A. ornata DHP I are ongoing.