Amphitrite ornata, a Marine Worm, Contains Two Dehaloperoxidase Genes

Mar. Biotechnol. 3, 287–292, 2001 DOI: 10.1007/s10126-001-0003-8

© 2001 Springer-Verlag New York Inc.

Amphitrite ornata, a Marine Worm, Contains Two Dehaloperoxidase Genes Kaiping Han, Sarah A. Woodin, David E. Lincoln, Kevin T. Fielman,* and Bert Ely† Department of Biological Sciences, University of South Carolina, Columbia, SC 29208, U.S.A.

Abstract: Amphitrite ornata, a terebellid polychaete, inhabits marine environments that are contaminated by biogenically produced halometabolites. These halogenated organic compounds are toxic and quite diverse. To survive in this environment, A. ornata produces a novel dehaloperoxidase (DHP I) that detoxifies haloaromatic compounds. In this study we identified and characterized two dehaloperoxidase genes, designated dhpA and dhpB, from an A. ornata complementary DNA library. The deduced amino acid sequences (DHP A and DHP B) of the two dhp genes both contain 137 amino acid residues, but they differ at 5 amino acid positions. Allelic variation was observed for both genes as well. Polymerase chain reaction–restriction fragment length polymorphism assays of genomic DNA from 19 in individuals showed that each individual contains both the dhpA and the dhpB genes. Therefore, the two types of DHP are encoded by separate genes and are not alleles of a single gene. Furthermore, DHP A and DHP B may have different substrate specificities since they have amino acid differences in the active site. Key words: Amphitrite ornata, dehaloperoxidase genes, haloaromatic compounds, heme protein, polychaete.

I NTRODUCTION A number of infaunal marine polychaete and hemichordate worms produce high levels of halogenated secondary metabolites (King, 1986; Woodin et al., 1987; Woodin, 1991; Gribble, 1999, 2000). Fielman et al. (1999) examined 40 taxa within several temperate benthic assemblages and found that 43% of the taxa contained halometabolites, indicating the widespread occurrence of natural halogenated organics among temperate marine infauna. These haloaromatics are toxic; for example, they cause liver neoplasms in Accepted January 16, 2001. *Present address: Department of Genetics, University of Georgia, Athens, GA 30609, U.S.A. †Corresponding author: telephone 803-777-2768; fax 803-777-4002; e-mail [email protected]

fish (Malins et al., 1987; McConnell, 1992) and interfere with respiration and membrane integrity (Escher et al., 1996). In addition to being found within worm tissues, the sediments surrounding the worm burrows contain halogenated compounds that significantly affect patterns of recruitment (Woodin et al., 1993, 1997; Hardege et al., 1998) and predation (S.A. Woodin et al., unpublished data). Thus these compounds can dramatically alter the composition of benthic assemblages (Fielman, 2000). Amphitrite ornata, a terebellid polychaete, does not produce halogenated organics, but co-occurs with the haloaromatic-producing worms Notomastus lobatus and Saccoglossus kowalewskyi. Chen et al. (1996) documented that A. ornata actively degrades toxic haloaromatics, with dehalogenating enzymes comprising up to 3% of A. ornata’s soluble protein. One dehalogenating peroxidase (DHP I)

288

Kaiping Han et al.

from A. ornata has been well characterized (Chen et al., 1996). This enzyme uses a novel mechanism to catalyze the H2O2-dependent conversion of mono-, di-, or trihalophenols into quinones. It is the only heme-containing dehaloperoxidase known to be capable of removing all halogens (including fluorine) from haloaromatics (Chen et al., 1996; Roach et al., 1997). The crystal structure of DHP I was recently determined and a catalytic mechanism was suggested (Roach et al., 1997; Lebioda et al., 1999; LaCount et al., 2000). In a population study DHP I enzyme levels were highest in field populations of A. ornata with the greatest body burden of bromophenols (Fielman, 2000). However, the worms’ overall level of dehalogenase activity did not change with bromophenol body burden. Thus A. ornata must have multiple dehalogenases, as suggested previously by Chen et al. (1996). In addition, the increased expression of DHP I in field populations must be matched by a decrease in the expression of other dehalogenases. This variation in the composition of dehalogenating enzymes may be important for tolerance to the variable halo-organic environments experienced by many benthic organisms. The evidence for multiple dehalogenase enzymes in A. ornata prompted us to clone the dhp genes. Using a probe designed from the amino-terminal amino acid sequence of DHP I (L. Lebioda and Y. Chen, unpublished data), we screened an A. ornata complementary DNA (cDNA) library. Two classes of dhp cDNAs were identified: one designated dhpA that encodes DHP I, and a second, designated dhpB, that encodes a second dehalogenase. Population analyses demonstrated that dhpA and dhpB are different genes and not alleles of the same gene.

M ATERIALS

AND

Construction and Screening of an Amphitrite ornata cDNA Library The RNA obtained from the 6 individuals was combined and sent to Stratagene (La Jolla, Calif.) for construction of a cDNA library in the Uni-ZAP XR vector. A 125-bp fragment containing part of the dhp gene was amplified from the library by polymerase chain reaction (PCR) using primers designed from the DHP I amino-terminal amino acid sequence (L. Lebioda and Y. Chen, unpublished data). The fragment was labeled with fluorescein-11-dUTP using random priming (Amersham Life Science, England) and used to screen the library. Positive plaques were replated and purified through 3 rounds of repeated screening. Plasmid DNA was excised using ExAssist helper phage with the SOLR strain (Stratagene). To improve the efficiency of screening for dhp genes in the cDNA library, a 688-bp fragment identified in the initial screening was labeled and used as a probe for subsequent screens. The 688-bp fragment was obtained by EcoRI digestion of a plasmid DNA containing most of dhp gene and 282 bp of the downstream noncoding region.

Nucleotide Sequence and Amino Acid Sequence Analysis The nucleotide sequence of the 13 presumptive dhp clones was determined using an automated DNA sequencer (LICOR, Lincoln, Neb.). DNA sequences were aligned and translated into amino acid sequence using the Eyeball SEquence Editor (ESEE version 3.0s). Nucleotide sequences of the dhp genes identified in this study are listed in GenBank under accession numbers AF284381 for dhpA and AF285090 for dhpB.

M ETHODS

Sample Collection, RNA Isolation, and DNA Isolation Six individuals were collected at North Inlet, Georgetown, S.C. Total RNA was extracted from the anteriormost segments including the prostomium and labial palps of all individuals, using a “one-step” total RNA preparation method (Chomczynski and Sacchi, 1987). Subsequently, 19 more individuals were collected from the same location, and genomic DNA was extracted from tissue with the genomic DNA isolation reagent DNAzol (Molecular Research Center, Inc., Cincinnati, Ohio).

PCR-RFLP Assay of dhp Genes from Genomic DNA A set of primers (BE440 5⬘-AAACAAGATATTGCCACC and DL15 5⬘-GCCATTGACTTGAGCTCTT) were developed based on the dhp cDNA sequences to amplify gene fragments from genomic DNA. PCR reactions contained 50 mM Tris-HCl, pH 8.3, 2 mM MgCl, 250 µ/ml bovine serum albumin, 0.2 mM each of dNTP, 0.5 µM of each primer, and 1.25 units of Taq DNA polymerase (Qiagen Inc., Valencia, Calif.) in a final volume of 50 µl. PCR reactions were performed under the following conditions: an initial denaturation at 94°C for 3 minutes, 31 amplification cycles (94°C for 40 seconds, 48°C

Two Dehaloperoxidase Genes from A. ornata

289

for 40 seconds, 72°C for 40 seconds), and a final extension at 72°C for 5 minutes. Since two RsaI restriction sites were present in the cDNA sequence of the dhpA gene, but only one was present in the dhpB gene (see below), we devised an RsaI restriction fragment length polymorphism (RFLP) assay to distinguish the two genes. Amplified dhp DNA was digested with RsaI and analyzed in 4% agarose high-resolution gels (FisherBiotech BP 1360-100) stained with ethidium bromide. Digestion of dhpA resulted in 3 fragments, while digestion of dhpB resulted in 2 fragments.

R ESULTS Isolation and Characterization of Amphitrite ornata dhp Genes Screening of approximately 100,000 recombinant bacteriophage clones with a DHP I probe resulted in 13 clones that showed a strong hybridization signal. These clones were purified, and up to 1016 bp of nucleotide sequence was determined for each cloned fragment. Analyses of the nucleotide sequence information showed that 11 of the clones encoded an amino acid sequence that corresponds to that determined for the DHP I protein (L. Lebioda and Y. Chen, unpublished data). The other 2 clones contained only the downstream noncoding region. In addition, 1 of the 11 clones lacked the first 270 bp of the coding region, so it was not included in the sequence comparisons. When the nucleotide sequences were compared, variation was observed at more than 50 positions (the nucleotide sequence alignment is available at www.biol.sc.edu/∼elygen). Nineteen of these variable sites were observed in the 411-bp coding region. The nucleotide sequences could be separated into two classes, designated dhpA and dhpB, that had fixed differences at 9 nucleotide positions within the 411 bp that encode DHP. No fixed differences were observed in the 600 bp of nucleotide sequence downstream from the coding region. When the 6 dhpA nucleotide sequences were compared, each sequence was unique. There were 5 variable sites in the coding region and 25 variable sites in the downstream noncoding region. Similarly, the 4 unique dhpB nucleotide sequences contained 6 variable sites in the coding region and 39 variable sites in the downstream noncoding region. Thus multiple alleles were observed among both classes of dhp genes.

Figure 1. Comparison of the deduced amino acid sequences of independent isolates of the two dehaloperoxidase genes, dhpA and dhpB. Dots represent identity to the reference sequence.

Deduced Amino Acid Sequences from dhpA and dhpB The deduced amino acid sequences of the dhp genes are shown in Figure 1. Both DHP A and DHP B are composed of 137 amino acid residues, but they differ at 5 amino acid positions (9, 32, 34, 81, and 91). Among the DHP A amino acid sequences, variation occurred only at amino acid position 134. Thus most of the nucleotide sequence variation occurs among synonymous codons. Similarly, amino acid variation was observed only at position 90 in the DHP B sequences. In both cases, the variable amino acid was either alanine or serine.

290 Kaiping Han et al.

Figure 3. Restriction map of the dhp genes. Primers BE440 and DL15 will amplify a 270-bp fragment of both genes. However, RsaI cleaves dhpA twice and dhpB once, so that the two genes can be resolved in a PCR-RFLP assay as shown in Figure 2, B.

Figure 2. A: PCR amplification of 270 bp of the dehaloperoxidase genes from genomic DNA of A. ornata. Lanes 1–8 are amplifications of DNA from 8 individual samples; lane 9 is a negative control containing no DNA; lane 10 is a 100-bp DNA ladder. Electrophoresis was performed in a 1% agarose gel. B: RsaI restriction endonuclease digestion of the PCR products shown in A. Lanes 1–8 correspond to samples in A. Lane 9 is a 100-bp DNA ladder. Lane 10 is an uncut PCR product. The 49-bp fragment is indicative of the presence of the dhpA gene, and the 67-bp fragment is indicative of the presence of the dhpB gene. Electrophoresis was performed in a 4% agarose high-resolution gel. The light bands present below the fragment migrating at 200 bp are not due to the RsaI digestion since they can be observed in the corresponding undigested samples shown in A after electrophoresis in 4% agarose gel.

PCR-RFLP Assays of dhp Genes from Genomic DNA The cDNA nucleotide sequence of the dhp genes derived from the original 6 A. ornata individuals was used to develop primer sets for amplifying the dhp genes from genomic DNA derived from 19 additional individuals. For the primer set BE440/DL15, the size predicted from the cDNA sequence is 143 bp. However, the fragment amplified from genomic DNA was approximately 270 bp (Figure 2, A), indicating that an intron of approximately 130 bp occurs in this fragment. The nucleotide sequence of the clones from the cDNA library revealed that the coding region of dhpA had 2 RsaI restriction sites, while dhpB had only 1. There-

fore, RsaI digestion of the 270-bp fragment amplified from the dhpA gene should result in 3 fragments (200, 49, and 18 bp), while digestion of dhpB should result in 2 fragments (200 and 67 bp) (Figure 3). When the genomic DNA from 19 individuals was analyzed, all 19 contained both the dhpA and the dhpB genes (Figure 2, B). This result confirms that the PCR-RFLP assay can be used to distinguish between the two loci and indicates that DHP A and DHP B are encoded by two different genes. If dhpA and dhpB were alleles of a single gene, then individuals homozygous for one or the other alleles should have been observed.

D ISCUSSION Organisms co-inhabiting sediments with producers of haloaromatic compounds are exposed to a diverse array of biogenic compounds (Fielman et al., 1999). Amphitrite ornata, for example, lives in habitats containing the large capitellid polychaete N. lobatus and several smaller capitellids that contaminate the sediments with mono-, di-, and tribromophenols and mono- and dibromovinylphenols. In addition, several bromopyroles are associated with the co-occurring cirratulid polychaetes and the hemichordate S. kowalewskyi. Thus the haloaromatic compounds challenging A. ornata are quite diverse. To survive in this environment, A. ornata may require a diverse set of dehalogenating enzymes to detoxify all the haloaromatic compounds it encounters. In the present study two types of dehaloperoxidase genes were identified from A. ornata cDNA clones. PCRRFLP assays of genomic DNA from 19 individuals showed that each individual contained both dhpA and dhpB (Figure

Two Dehaloperoxidase Genes from A. ornata

2). Thus dhpA and dhpB are separate genes and not alleles of a single gene. Their 98% nucleotide sequence identity suggests that the two genes may have arisen from a recent gene duplication. Since A. ornata are exposed to a variety of halogenated compounds, multiple dehalogenating enzymes with differing specificities may be needed for survival. The DHP A and DHP B proteins are predicted to differ at 5 amino acid positions (Figure 1). The variation between serine (DHP A) and glycine (DHP B) at position 91 is probably the most important difference because it occurs in the active site close to the histidine at position 89. This histidine is important for binding the heme, and motion in the loop containing His 89 is thought to be important for peroxidase activity (Roach et al., 1997; LaCount et al., 2000). The change between serine and glycine at position 91 could be very important because glycine is the most flexible amino acid residue and the two residues differ in hydrogenbonding capabilities. Thus differences in hydrogen bonding and flexibility in the active site could contribute to differences in substrate specificity that may be important for survival in variable halo-organic environments. Additional variation occurs among alleles of DHP B. Either serine or alanine was found at position 90 (Figure 1). As discussed above, this region encodes the active site. Therefore, variation at position 90 may affect substrate specificity. Allelic variation was also observed in DHP A at position 134. However, because this region is near the end of the peptide sequence and is not thought to be important for DHP structure, it is unlikely that variation at position 134 causes a functional difference. We observed far more nucleotide sequence variation at both the dhpA and the dhpB loci than one would expect. At the dhpA locus, the nucleotide sequences of 5 of the 6 clones analyzed could be grouped into two classes that differed at 11 of 900 positions (the sequence alignments can be observed at www.biol.sc.edu/∼elygen). Within each group the most similar sequences differed at a minimum of 3 additional positions. The sequence of the sixth clone differed from all the others at a minimum of 13 positions. At the dhpB locus, all 4 sequences were unique with the most similar sequences differing from each other at 9 of 820 positions. Why do we observe so much nucleotide sequence variation? One explanation could be that artifactual changes in the nucleotide sequence occurred during the cloning process. This is unlikely because many differences were observed. Nonetheless, if this explanation were true, we would expect to find that the changes occurred randomly through-

291

out the sequence. In fact, most of the changes occurred in the 3⬘ noncoding region, and those that did occur in the coding region were primarily third codon position changes that did not affect the DHP amino acid sequence. Thus the observed changes are clearly not due to a random process. In addition, since one of the changes at the dhpB locus results in an altered MspI restriction site, we amplified this region directly from the genomic DNA of 19 individuals and verified that this restriction site polymorphism is present in the natural population (data not shown). Therefore, we conclude that the observed nucleotide sequence variation is present in the natural population. One alternative explanation for the high level of genetic diversity could be that both dhp loci are mutational hotspots. Since the dhp genes probably arose from a duplication event, a mutational hotspot could have been part of the duplicated region. Another possibility is that there has been selection for genetic diversity at the two dhp loci, or at loci closely linked to them. Selection for genetic diversity could produce a set of alleles that differ at multiple positions, as we have observed at these two loci. Further experiments are needed to discriminate between these two hypotheses. Chen et al. (1996) indicated that there are two types of DHP enzymes. In this study we found that the amino acid sequence of DHP A is identical to the 60 amino acid aminoterminal sequence of DHP I (L. Lebioda and Y. Chen, unpublished). Therefore, dhpA encodes DHP 1. The aminoterminal amino acid sequence of DHP II has not been determined. Therefore, we do not know whether dhpB encodes DHP II.

A CKNOWLEDGMENTS We thank L. Lebioda, G. Leclerc, C.R. Lovell, Y. Chen, and J. Sun for valuable discussions and technical assistance. We also thank E. Thelen for performing the nucleotide sequence determinations. Funding for this work was provided by the College of Science and Mathematics at the University of South Carolina and National Science Foundation grant OCE-9811435 to S.A. Woodin. This is contribution number 1258 of the Belle W. Baruch Institute.

R EFERENCES Chen, Y.P., Woodin, S.A., Lincoln, D.E., and Lovell, C.R. (1996). An unusual dehalogenating peroxidase from the marine terebellid polychaete Amphitrite ornata. J Biol Chem 271:4609–4612.

292 Kaiping Han et al.

Chomczynski, P., and Sacchi, N. (1987). Single step method of RNA isolation by guanidinium thiocynate–phenol–chloroform extraction. Anal Biochem 162:156–159. Escher, B.I., Snozzi, M., and Schwarzenbach, R.P. (1996). Uptake, speciation, and uncoupling activity of substituted phenols in energy transducing membranes. Environ Sci Technol 300:3071–3079. Fielman, K.T. (2000). Biogenic halogenated organic compounds and their dehalogenation as determinants of marine benthic community structure. PhD dissertation. Columbia: University of South Carolina. Fielman, K.T., Woodin, S.A., Walla, M.D., and Lincoln, D.E. (1999). Widespread occurrence of natural halogenated organics among temperate marine infauna. Mar Ecol Prog Ser 181:1–12. Gribble, G.W. (1999). The diversity of naturally occurring organobromine compounds. Chem Soc Rev 28:335–346. Gribble, G.W. (2000). The natural production of organobromine compounds. Environ Sci Pollut R7:37–49. Hardege, J.D., Bentley, M.G., and Snape, L. (1998). Sediment selection by juvenile arenicola marina. Mar Ecol Prog Ser 166:187– 195. King, G.M. (1986). Inhibition of microbial activity in marine sediments by a bromophenol from a hemichordate. Nature 323:257– 259. LaCount, M.W., Zhang, E., Chen, Y.P., Han, K.P., Whitton, M.M., Lincoln, D.E., Woodin, S.A., and Lebioda, L. (2000). The crystal structure and amino acid sequence of dehaloperoxidase from Amphitrite ornata indicate common ancestry with globins. J Biol Chem 275:18712–18716.

Lebioda, L., LaCount, M.W., Zhang, E., Chen, Y.P., Han, K.P., Whitton, M.M., Lincoln, D.E., and Woodin, S.A. (1999). An enzymatic globin from a marine worm. Nature 401:445. Malins, D.C., McCain, B.B., Myers, M.S., Brown, D.W., Krahn, M.M., Roubal, W.T., Schiewe, M.H., Landahl, J.T., and Chan, S.-L. (1987). Field and laboratory studies of the etiology of liver neoplasms in marine fish from Puget Sound. Environ Health Perspect 71:5–16. McConnell, L.L. (1992). Air-water gas exchange of chlorinated organic compounds. PhD dissertation. Columbia: University of South Carolina. Roach, M.P., Chen, Y.P., Woodin, S.A., Lincoln, D.E., Lovell, C.R., and Dawson, J.H. (1997). Notomastus lobatus chloroperoxidase and Amphitrite ornata dehaloperoxidase both contain histidine as their proximal heme iron ligand. Biochemistry 36:2197–2202. Woodin, S.A. (1991). Recruitment of infauna: positive or negative cues? Am Zool 31:797–807. Woodin, S.A., Walla, M.D., and Lincoln, D.E. (1987). Occurrence of brominated compounds in soft-bottom benthic organism. J Exp Mar Biol Ecol 107:209–217. Woodin, S.A., Marinelli, R.L., and Lincoln, D.E. (1993). Allelochemical inhibition of recruitment in a sedimentary assemblage. J Chem Ecol 19:517:530. Woodin, S.A., Lindsay, S.M., and Lincoln, D.E. (1997). Biogenic bromophenols as negative recruitment cues. Mar Ecol Prog Ser 157:303–306.