Phytochemistry 72 (2011) 337–343

Contents lists available at ScienceDirect

Phytochemistry

journal homepage: www.elsevier .com/locate /phytochem

Isolation and characterization of a gene encoding a S-adenosyl-L-methionine-dependent halide/thiol methyltransferase (HTMT) from the marine diatomPhaeodactylum tricornutum: Biogenic mechanism of CH3I emissions in oceans

Hiroshi Toda, Nobuya Itoh ⇑Department of Biotechnology, Faculty of Engineering, Biotechnology Research Center, Toyama Prefectural University, 5180 Kurokawa, Imizu, Toyama 939-0398, Japan

a r t i c l e i n f o

Article history:Received 9 August 2010Received in revised form 19 November 2010Available online 10 January 2011

Keywords:Phaeodactylum tricornutumMethyl halidesMethyl iodideS-Adenosyl-L-methionine (SAM): halide/thiol methyltransferase

0031-9422/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.phytochem.2010.12.003

Abbreviations: HMT, halide ion methyltransferasease; HTMT, halide/thiol methyltransferase; SAM, S-adadenosyl-L-homocysteine; DMS, dimethyl sulfide.⇑ Corresponding author. Tel.: +81 766 56 7500x560

E-mail address: nbito@pu-toyama.ac.jp (N. Itoh).

a b s t r a c t

Several marine algae including diatoms exhibit S-adenosyl-L-methionine (SAM) halide/thiol methyltrans-ferase (HTMT) activity, which is involved in the emission of methyl halides. In this study, the in vivo bio-genic emission of methyl iodide from the diatom Phaeodactylum tricornutum was found to be clearlycorrelated with iodide concentration in the incubation media. The gene encoding HTMT (Pthtmt) was iso-lated from P. tricornutum CCAP 1055/1, and expressed in Escherichia coli. The molecular weight of theenzyme was 29.7 kDa including a histidine tag, and the optimal pH was around pH 7.0. The kinetic prop-erties of recombinant PtHTMT towards Cl�, Br�, I�, [SH]�, [SCN]�, and SAM were 637.88 mM, 72.83 mM,8.60 mM, 9.92 mM, 7.9 mM, and 0.016 mM, respectively, and were similar to those of higher-plantHTMTs, except that the activity towards thiocyanate was lower. The biogenic emission of methyl halidesfrom the cultured cells and the enzymatic properties of HTMT suggest that the HMT/HTMT reaction is keyto understanding the biogenesis of methyl halides in oceanic environments as well as terrestrial ones.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Several methyl halides (CH3Cl, CH3Br, CH3I, CH2Br2, and CHBr3)are released from oceanic sources, terrestrial plants, and fungi, aswell as by the burning of biomass (Harper et al., 1988; Itoh andShinya, 1994; Klick and Abbrahamsson, 1992; Lovelock, 1975;Manley et al., 1992, 2006; Manö and Andreae, 1994; Moore andTokarczyk, 1993; Yokouchi et al., 2002). Monohalomethanes(CH3Cl, CH3Br, and CH3I) are the simplest halogenated compoundsand exist in the atmosphere in large quantities (CH3Cl, 550; CH3Br,9; and CH3I, 5–10 pptv (parts per trillion by volume) in oceanic air)(Alicke et al., 1999; O’Dowd et al., 2002; World MeteorologicalOrganization, 2006). These compounds are the primary carriersof natural halogens from ocean to land. Since CH3Cl and CH3Brhave long half-lives in the atmosphere (1.0 and 0.7 years, respec-tively), they catalyze ozone destruction in the stratosphere byreleasing halogen radicals. Therefore, the Montreal Protocol regu-lates CH3Br use as a soil fumigant. In contrast, the half-life ofCH3I (7–10 days) is shorter than CH3Cl and CH3Br (Chameidesand Davis, 1980; World Meteorological Organization, 2006), and

ll rights reserved.

; TMT, thiol methyltransfer-enosyl-L-methionine; SAH, S-

; fax: +81 766 56 2498.

CH3I has been used as a substitute for CH3Br in agriculture. Mostmethyl halides (CH3X) are generated from oceanic and terrestrialenvironments as well as anthropogenic sources. However, thereis limited information about the origins, quantities, and physiolog-ical functions of methyl halides, especially CH3I from the oceans.

Chloride methyltransferase (S-adenosyl-L-methionine: halideion methyltransferase, HMT) that catalyzes the formation ofmethyl halides using S-adenosyl-L-methionine (SAM) as a methyldonor, was isolated from red marine alga Endocladia muricata,and the methylation mechanism is as follows (Wuosmaa andHager, 1990):

X� þ SAM! CH3Xþ S-adenosyl-l-methionine ðSAHÞ:

Homologues of this enzyme have been found in several organ-

isms, such as higher plants (Attieh et al., 1995, 2000a,b; Ni andHager, 1998), algae (Ohsawa et al., 2001), fungi (Saxena et al.,1998), and soil bacteria (Amachi et al., 2001), and some of thesehomologues have been characterized in detail. The data show thatthey also catalyze the methylation of thiol substrates, such as thebisulfide ion ([SH]�) or the thiocyanate ion ([SCN]�) to give CH3SHor CH3SCN, respectively (Attieh et al., 2000a,b; Drotar et al., 1987).

Genes encoding homologues of HTMT have been cloned fromseveral higher plants belonging to the family Brassicaceae and hal-ophytic plants, expressed in Escherichia coli, and the properties ofthe recombinant enzymes were characterized (Attieh et al., 2002;Itoh et al., 2009; Nagatoshi and Nakamura, 2007; Ni and Hager,

338 H. Toda, N. Itoh / Phytochemistry 72 (2011) 337–343

1998, 1999; Rhew et al., 2003). It is thought that these enzymes areinvolved in cellular homeostasis, such as in the detoxification ofsulfur compounds (Attieh et al., 2000a) and in salt tolerance viathe emission of methyl halides (Attieh et al., 1995; Ni and Hager,1998). Moreover, the completion of a number of whole-genome sequences has shown that HTMT homologues exist inseveral organisms. It was previously reported that several marinemacro- and microalgae, such as E. muricata, Papenfusiella kuromo,Sargassum horneri (macroalgae) (Itoh et al., 1997; Wuosmaa andHager, 1990), and Pavlova sp. (microalgae) (Ohsawa et al., 2001),exhibit HTMT activity and emit CH3I. Therefore, these organismsprobably have genes that are homologous to HTMT. In fact, wehave found a sequence for an HTMT homologue in the genome ofthe marine diatom Phaeodactylum tricornutum (Bowler et al.,2008). Moreover, expression sequence tag analysis confirmedexpression of the gene in P. tricornutum (Scala et al., 2002). How-ever, its precise function and roles in vivo are unclear.

In this study, the biogenic emission of methyl halide and theHTMT activity of several marine algae including P. tricornutumwere measured. Moreover, a gene encoding HTMT (Pthtmt) wasisolated from P. tricornutum, verified, expressed in E. coli, and char-acterized in detail. The results clarify the mechanism of biogenicemission of methyl halides from oceans.

Fig. 2. Schematic model of biosynthesis and emission of methyl halides frommarine microalgae.

2. Results and discussion

2.1. Profile of methyl halide emissions from marine algae and diatoms

To examine the relationships between HTMT activity and theemission strength of methyl halide, five marine microalgae thatare reported to have HTMT activity (Itoh et al., 1997) and a marinediatom were cultured, incubated with seawater containing 5 mMpotassium iodide in a headspace vial, and the amounts of CH3Iand the HTMT activity of crude cell extracts were assayed byGC–MS (Fig. 1). The marine diatom P. tricornutum showed the high-est rate of CH3I emission (2432 pmol/g cell/day), although theHTMT activity thereof was weaker than the activity of other micro-algae including P. pinguis CCAP 940/2 and P. gyrans CCAP 940/1,which showed lower rates of CH3I emission (538–604 pmol/gcell/day). Isochrysis galbana showed the same level of HTMT activ-ity as P. tricornutum; however, no CH3I emission was detected. On

HTM

T ac

tivity

(U/g

cel

l)

Emis

sion

rate

of C

H3I

(pm

ol/g

cel

l/day

)

0

20

40

60

80

100

120

Pavlov

a ping

uis C

CAP940/2

Pavlov

a gyra

ns C

CAP940/1

B

Isoch

rysis

galba

na N

BRC1028

13

Pavlov

a ping

uis N

BRC1028

07

Nanno

chlor

is ato

mus C

CAP251/4

A

Phaeo

dacty

lum tri

cornu

tum C

CAP1055

/10

500

1000

1500

2000

2500

3000

HTMT activityEmission rate of CH3I

Fig. 1. HTMT activity and rate of CH3I emission from various marine microalgae.The values represent the mean of three replicate analyses, and the error bars showvariance from the mean. The absence of bars indicates that the errors were toosmall to detect.

the other hand, P. pinguis NBRC 102807 emitted the same level ofCH3I as P. pinguis CCAP 940/2 and P. gyrans CCAP 940/1, althougha much lower level of HTMT activity was found in the crude extractof the cells. A trace amount of CH3I was emitted by Nannochlorisatomus that corresponded to a low level of HTMT activity.

Itoh et al. (1997) reported that P. gyrans CS-213 produced alarge amount of methyl halides, and N. atomus emitted a traceamount of them. Thus, the emission level of CH3I corresponds toHTMT activity in crude cell extracts. In this study, however, we ob-served some exceptions between CH3I emission level and HTMTactivity in the crude cell extracts. Such a discrepancy might becaused by several factors related to differences in enzymaticproperties between the variants of HTMT, such as Km values for

0

0.2

0.4

0.6

0.8

1

0 5 10 15

natural sea water

5 mM KI

10 mM KI

Cultivation time (days)

Con

cent

ratio

n of

CH

3Cl i

n vi

al (p

pm)

0

1

2

3

4

5

0 5 10 15

natural sea water

5 mM KI10 mM KI

Cultivation time (days)

Con

cent

ratio

n of

CH

I 3 in

vi

al (p

pm)

A

B

Fig. 3. Profiles of CH3I emissions from P. tricornutum cultivated in seawater with0–10 mM potassium iodide. Data points represent the mean values of threereplicate analyses, and the error bars show variance from the mean. The absence ofbars indicates that the errors were too small to detect.

Table 1Substrate specificity of recombinant PtHTMT compared with other HMT/HTMTs. Kinetic parameters for each methyl acceptor were measured at a fixed concentration of SAM(500 lM), and those for SAM were at constant concentration of iodide (20 mM).

Substrate Km (mM) kcat (s�1) kcat/Km (M�1 s�1) Km (mM) (E. muricata)a Km (mM) (R. sativus)b kcat (s�1) (R. sativus)b

SAM 0.016 – – 0.016 0.19 –Cl� 637.9 0.31 � 10�3 0.48 � 10�3 5 1657.4 1.63 � 10�3

Br� 72.8 1.04 � 10�3 1.43 � 10�2 40 177.3 16.9 � 10�3

I� 8.6 51.6 � 10�3 6.00 – 4.5 67.3 � 10�3

[SH]� 9.9 38.1 � 10�3 3.85 – 12.2 76.7 � 10�3

[SCN]� 7.9 10.2 � 10�3 1.29 – 0.04 89.5 � 10�3

KCN ND ND ND – ND NDIO3� ND ND ND – – –

ND, not detected.a Values from Wuosmaa and Hager (1990).b Values from Itoh et al. (2009).

H. Toda, N. Itoh / Phytochemistry 72 (2011) 337–343 339

I�, or the concentration and form of iodide ions (I�, IO3� or I2) in

the algal cells, since the concentration of free I� in seawater is quitelow (0.012 ppm). Leblanc et al. (2006) reported that haloperox-idase in brown kelp Laminaria digitata is involved in the accumula-tion of iodine from seawater by the following mechanism: I� isoxidized to HIO by haloperoxidase located in the cell wall; then,HIO and free I� spontaneously form the iodine molecule. Thehydrophobic property of the iodine molecule enables it to passthrough the cell membrane. Likewise, Amachi et al. (2007) re-ported that marine flavobacteria have a similar iodine uptakemechanism to that of L. digitata. Therefore, if marine microalgaehave the same iodine uptake system as L. digitata and marine flavo-bacteria, these properties would influence the concentration of I�

in the cells and affect the profile of methyl halide emission fromeach organism (Fig. 2).

To investigate the CH3I emission profile of P. tricornutum, cellswere incubated in natural seawater and natural seawater supple-mented with KI, and the emission of CH3I was measured. NoCH3I emissions were detected with natural seawater; however,the emission of CH3I increased with increasing concentration ofKI (Fig. 3A). This indicated that free I� in seawater was crucialand affected the level of CH3I emission. It was also confirmed thatiodate (IO3

�) was inert, not only as an iodine source in the incuba-tion media of P. tricornutum (data not shown), but also as a sub-strate for HTMT (Table 1). When 10 mM KBr was added toseawater instead of KI, CH3Br was not detected, in contrast toCH3I (data not shown). It has been reported that the HTMT en-zymes of other organisms show high levels of activity for I� as amethyl acceptor, and have much lower levels of activity for Br�

and Cl� in this order (Attieh et al., 1995; Itoh et al., 2009; Ni andHager, 1999; Rhew et al., 2003; Saxena et al., 1998). We found thatP. tricornutum HTMT showed high specificity towards I� asdescribed below. Therefore, the above-mentioned findings are rea-sonable. Interestingly, CH3Cl (at approximately 10% of the concen-tration of CH3I) was detected with CH3I after 3 days incubation inthe presence of KI, although no CH3Cl formation was detected inthe absence of KI (Fig. 3B). We confirmed that HTMT activity wasnot induced when P. tricornutum was cultivated in the presenceof KI (data not shown). These findings suggest that some CH3Iwas converted to CH3Cl via a chemical reaction, namely, a bimolec-ular nucleophilic substitution (SN2) reaction of CH3I with Cl�

because of the high concentration of Cl� in seawater (Fig. 2).Zafiriou (1975) proposed that atmospheric CH3Cl and CH3Br areproduced through intermediate biosynthesis from CH3I, and Elliotand Rowland (1993) and Jones and Carpenter (2007) reported thatCH3I with Cl� takes on a significant role in the production of CH3Clin seawater. The results of our experiments strongly support thesereports that CH3Cl is possibly produced not only by direct biosyn-thesis, as reported for specific tropical plants, including ferns,members of the family Dipterocarpaceae (Yokouchi et al., 2002),

and salt-marsh plants (Manley et al., 2006), but also by conversionfrom CH3I via a chemical reaction with Cl� in seawater, as illus-trated in Fig. 2.

In addition, Moore and Zafiriou (1994) proposed that a radicalreaction might catalyze the production of CH3I from iodide radicalsand methyl radicals which were photochemically formed fromhumic-like dissolved organic matters (DOM) derived frommicroorganisms, and Ooki et al. (2010) suggested that these photo-chemical reactions might be significant processes in theproduction of methyl halides in surface seawater. Thus, CH3Iformation in oceanic sources seems to be a complex mixture ofphotochemical and biochemical reactions.

2.2. Cloning and expression of gene encoding HTMT

The full length of the gene encoding HTMT (Pthtmt) was isolatedby PCR, and recombinant PtHTMT was expressed in E. coli for char-acterization (DDBJ/GenBank/EMBL BR000875). Total RNA andgenomic DNA prepared from cultured cells for 7 days were usedas templates for PCR respectively, and both amplified productsgave a single fragment of 760 bp. These PCR products were clonedinto pGEM-T Easy vector and their nucleotide sequences weredetermined. The resulting sequences agreed with those held in adatabase (GenBank Accession No.: EEC49246). The results indi-cated that the gene encoding HTMT from P. tricornutum has no in-trons in the open reading frame (ORF). Pthtmt was found to encodea protein consisting of 252 amino acid residues, which showedsimilarity with the higher-plant HTMT/TMT genes. The deducedamino acid sequence of Pthtmt contained several motifs that areknown to be involved in the SAM binding pocket in several meth-yltransferases (Fig. 4A) (Labahn et al., 1994; Vidgren et al., 1994),and the inferred secondary structure of PtHTMT was similar tothe S-adenosyl-L-methionine-dependent small-molecule methyl-transferases (Kagan and Clarke, 1994; Martin and McMillan,2002). Moreover, several homologous genes that show similaritywith Pthtmt were found in the genome sequences of several organ-isms including higher plants, fungi, bacteria, and unicellular algae.This fact suggests that the gene encoding HTMT has the same ori-gin, and is distributed widely throughout many organisms. Phylo-genetic analyses using amino acid sequences showed that PtHTMTexhibits similarity to higher-plant HTMT/TMTs rather than those ofunicellular algae such as Ostreococcus tauri (Fig. 4B).

2.3. Enzymatic properties of recombinant PtHTMT

To examine the enzymatic properties of PtHTMT, the Pthtmtgene was expressed in E. coli BL21(DE3) using the expressionvector pET-21b. Significant HTMT activity was observed in the cellswith pG-KJE8, which possessed chaperone-coding genes. Incontrast, no HTMT activity was found in the transformants lacking

A

B

Fig. 4. Properties of PtHTMT gene. (A) Alignment of amino acid sequences of PtHTMT and higher-plant HTMTs/TMTs. Amino acid sequences were aligned using ClustalW.Black boxes denote amino acids that are conserved in many HTMT/TMTs. (B) The phylogenetic tree of PtHTMT homologous proteins. Deduced amino acid sequences ofPtHTMT homologues were selected using BLAST search and aligned using ClustalW. The tree analysis was performed using the NJplot program. Aspergillus fumigatus (AfTMT;GenBank Accession No.: EAL89436), A. niger (AnTMT; CAK38144), Arabidopsis thaliana (AtHOL1; BAF01137), Batis maritima (BmMCT; AAD26120), Brassica oleracea (BoTMT1;AAK69760), Brassica rapa (BrTMT; ABL86248), Cryptococcus neoformance (CnTMT; EAL23163), Monosiga brevicollis (MbHMT; EDQ86113), Ostreococcus lucimarinus (OlMTR;ABO94880), Oryza sativa (OsHOL1; AAS07345), Ostreococcus tauri (OtTMT; CAL52768), Populus trichocarpa (PtHOL; EEF03847), Phaeodactylum tricornutum (PtHTMT;EEC49246), Raphanus sativus (RsHTMT; BAH84870), Sorghum bicolor (SbMTR; EER89221), Thalassiosira pseudonana (TpMTR; XP_002290733), Vitis vinifera (VvMTR;CAO46360), Zea mays (ZmHOL; ACG28672).

340 H. Toda, N. Itoh / Phytochemistry 72 (2011) 337–343

pG-KJE8, suggesting that a set of chaperones is necessary to nor-mally express Pthtmt in E. coli. Recombinant PtHTMT was obtainedas soluble protein with a histidine tag at the C-terminus and suc-cessfully purified by Ni-Sepharose resin column chromatography.In the final purified fraction, a single protein band of 34 kDa wasdetected by SDS–PAGE analysis (Fig. 5). The molecular weight ofthe recombinant PtHTMT estimated from HPLC analysis was30 kDa, suggesting that the native enzyme is a monomer. However,the molecular weight estimated by SDS–PAGE was slightly largerthan that (29.7 kDa) calculated from the deduced amino acid se-quence, perhaps as a result of the charge properties of the histidinetag attached to the expressed protein.

Substrate specificity and enzymatic properties of recombinantPtHTMT were characterized in detail. As shown in Table 1, theKm values for each substrate of recombinant PtHTMT revealed asimilar trend to those of previously reported HTMTs and exhibited

high specificity for I�, [SH]�, and [SCN]�, and low specificity for Cl�

and Br�. No HTMT activity was observed toward CN� or IO3�.

Recombinant PtHTMT exhibited similar specificities for these sub-strates, except that the specificity for [SCN]� was much lower thanthose of RsHTMT (Itoh et al., 2009). Considering the Km and kcat/Km

values of PtHTMT, it was concluded that the enzyme is specific forI� and adapted to the production of CH3I. The kinetic parameters ofPtHTMT supported the emission profile of methyl halides in vivo(Fig. 3).

The influence of several anionic thiol compounds on the HTMTactivity of recombinant PtHTMT was analyzed, and the results areshown in Table 2. Bisulfide ([SH]�) and thiocyanate ([SCN]�) ion,which are the most readily methylated substrates of PtHTMT, sig-nificantly inhibited the CH3I production of HTMT, and 2-mercap-tethanol showed an intermediate level of inhibition. Moreover,Cl� and Br� ions, which are probably natural substrates of HTMT

1 2 3 M

198.8

96.7

115.7

37.1

6.9

29.1

19.5

53.5

kDa

1 2 3 M

198.8

96.7

115.7

37.1

6.9

29.1

19.5

53.5

kDa

Fig. 5. SDS–PAGE analysis of recombinant PtHTMT expressed in E. coli cells. Lane 1,cell lysate of host strain (E. coli BL21(DE3) possessing pG-KJE8; lane 2, cell lysate ofE. coli transformant harboring pET-Pthtmt and pG-KJE8; lane 3, purified PtHTMT byNi-Sepharose column chromatography; M, molecular marker.

Table 2Effect of various compounds on HTMT/TMT activity. Enzyme activities are expressedas percentages of CH3I production under standard assay conditions in the presence ofeach compound (10 mM).a

Compound Relative activity (%)

P. tricornutum Pavlova pinguisb Brassica oleraceac

Control 100 100 100Cl� 94.2 121.0 –Br� 86.9 116.5 –[SH]� 0.5 15.7 53[SCN]� 13.9 40.4 0[CN]� 107.0 85.7 822-Mercaptoethanol 56.4 84.0 –

a The activity of each enzyme under the standard assay conditions was as fol-lows: P. tricornutum, 104,266 pmol min�1 mg protein�1; P. pinguis,1789 pmol min�1 mg protein�1; B. oleracea, 60 pmol min�1 mg protein�1.

b Values from Ohsawa et al. (2001).c Values from Attieh et al. (2000a).

0

50000

100000

150000

200000

2 4 6 8 10 12pH

Spec

ific

activ

ity (p

mol

/min

/mg)

Fig. 6. Effect of pH on HTMT activity. The enzyme activities were measured in thefollowing 50 mM buffers: citrate (�), MES (�), MOPS (N), Tris–HCl (j), and glycine-NaOH (d). HTMT activity is expressed as the amount of CH3I produced.

H. Toda, N. Itoh / Phytochemistry 72 (2011) 337–343 341

in the oceanic environment, showed a slight level of inhibition. Incontrast, [CN]� ion, which was not used as a substrate, did not in-hibit PtHTMT. These findings indicate that PtHTMT is competi-tively inhibited by effective methyl acceptors. It was reportedthat the HMT enzyme of A. thaliana and P. pinguis were stronglyinhibited by [SH]� or [SCN]�, as well as by PtHTMT (Ohsawa etal., 2001; Rhew et al., 2003). However, HMT from P. pinguis wasslightly inhibited by [CN]�, and it was not inhibited by Cl� andBr�. These differences between HTMTs and TMTs indicated that

these enzymes have different specificities for these methylacceptors.

Recombinant PtHTMT exhibited a sharp pH optimum at pH 7.0with the activity falling below 50% at pH 6.0 or 8.5 (Fig. 6), suggest-ing that it adapts to the cellular pH of the alga. This property issimilar to that of other HTMTs from Brassica oleracea (Attiehet al., 1995).

Many plants including marine algae have adapted to their livingenvironments, and different variants of HTMT might have led tothe development of different adaptations such as the ability togrow in a salt marsh via the emission of methyl halides or thedetoxification of glucosinolates for members of the family Brassic-aceae that release CH3SH or CH3SCN. The precise functions ofHTMT in the diatom P. tricornutum remain unknown. However,our data show that PtHTMT was suitable for producing CH3I; there-fore, it is speculated that its physiological function involves theproduction of CH3I. In addition, CH3I directly produced by marinealgae was partly converted to CH3Cl in seawater, as shown inFig. 2. Therefore, the HTMT activity of marine microalgae plays asignificant role in the biogenic emission of methyl halides fromoceans.

3. Conclusions

It was found that various marine microalgae including marinediatoms emit methyl halides and that these strains exhibit HTMTactivity. In addition, the level of methyl iodide emitted by the mar-ine diatom P. tricornutum increases in response to increased iodideion concentration in the incubation media. A gene encoding HTMT(Pthtmt) was isolated from P. tricornutum, verified, expressed inE. coli, and characterized in detail. Phylogenetic analysis shows thatgenes homologous to Pthtmt are distributed widely in variousorganisms and PtHTMT is similar to plant HTMTs. The kinetic prop-erties of recombinant PtHTMT are similar to those of higher-plantHTMT except for a lower specificity toward thiocyanate. These re-sults clarify the mechanism of the biogenic emission of methyl ha-lides from oceans.

4. Experimental

4.1. Strains and vectors

Phaeodactylum tricornutum (CCAP 1055/1), Pavlova pinguis(CCAP 940/2), Pavlova gyrans (CCAP 940/1B), and N. atomus (CCAP254/4A) were supplied by the CCAP (Culture Collection of Algaeand Protozoa), UK. P. pinguis (NBRC 102807) and I. galbana (NBRC102813) were supplied by NITE (National Institute of Technologyand Evaluation), Tokyo, Japan. To obtain the gene encoding Pthtmt,

342 H. Toda, N. Itoh / Phytochemistry 72 (2011) 337–343

chromosomal DNA and total RNA were isolated from P. tricornu-tum. pGEM-T Easy cloning vector and expression vector pET-21bwere used to clone the Pthtmt gene. E. coli DH5a and BL21(DE3)were used as cloning or expression hosts. The plasmid pG-KJE8,which carries genes encoding chaperones (i.e., DnaK–DnaJ–GrpEand GroES–GroEL), was coexpressed in order to successfully ex-press the recombinant protein.

4.2. Cultivation of marine algae and diatoms

Each marine alga was cultured in sterile natural seawater(20 mL) containing 0.252 g/L Daigo’s IMK medium (Nihon Pharma-ceutical Co., Ltd) at 20 �C and 200 lE/m2/s (16 h light and 8 h dark)for 1 week. Because of the silicate requirement of P. tricornutum,30 mg/L sodium metasilicate was added to the culture medium.Cultured cells (20 mL) were transferred into the new medium(2 L) mentioned above with sterile air bubbling (0.5 L air/min)using a filter (0.2 lm) and cultivated under the same conditionsas the previous culture for 10 days.

4.3. In vivo emission of methyl halides

Cultured cells of each marine alga were washed once and re-suspended with autoclaved seawater supplemented with anappropriate concentration of KI, KBr, or KIO3. Suspended cells (total5 mL; OD600 = 1.0) were transferred into an autoclaved 22-mLheadspace vial and sealed using a silicon septum and incubatedat 20 �C and 200 lE/m2/s (16 h light and 8 h dark) for a givenperiod of time (0.5–14 days). Thereafter, the headspace gas wasanalyzed on a Shimadzu QP-2010 gas chromatograph–mass spec-trometer (GC–MS).

4.4. Analytical conditions for gas chromatography and GC–massspectrometry

The measurement of methyl halides and sulfur-containingcompounds was performed in accordance with a previous study(Itoh et al., 2009). Methyl halides, CH3SH, and DMS were assayedon a Shimadzu QP-2010 GC–MS (quadrupole type) spectrometerequipped with a TurboMatrix HS40 (Perkin-Elmer) headspace sam-pler. Each sample vial was held at 70 �C for 20 min on the head-space sampler, and the gaseous phase was drawn for 0.2 minafter pressuring the tube for 3 min. The sample gas was injectedinto a DB-VRX capillary column (J&W Scientific; 60 m � 0.25 mmi.d., 1.4 lm film thickness) for analysis. The column temperatureprogram was as follows: 40 �C for 5 min, 50 �C at 2 �C/min, and180 �C at 10 �C/min. The retention times of CH3Cl, CH3Br, CH3I,and CH3SH were 5.05, 6.20, 9.00, and 5.85 min, respectively.

CH3CN and CH3SCN were measured using a GC-14A gas chro-matograph (Shimadzu) equipped with a flame ionization detector(FID). The packed column of a Thermon1000/ShimaliteW(Shimadzu GLC Inc.; 2.1 m � 3.2 mm) was used for analysis. A5-lL aliquot of reaction mixture was injected directly into thecolumn and kept at 80 �C. Nitrogen gas was used as a carrier andflow rate was 40 mL/min. The retention times of CH3CN andCH3SCN were 1.48 and 4.48 min, respectively.

Calibration for each compound was performed in accordancewith a previous study (Itoh et al., 2009) by using standard gas mix-tures and a standard chemical solution of CH3CN or CH3SCN.

4.5. Enzyme assay

The measurement of HTMT activity was performed in accordancewith a previous study (Itoh et al., 2009). A reaction mixture consist-ing of 0.5 mM SAM, 20 mM methyl donor (halides (KX), bisulfide(NaSH)), 20 mM MES (pH 7.0) was mixed with an appropriate

amount of enzyme solution to make a total volume of 5.0 mL in a22-mL vial sealed with a silicone septum. Reactions were startedby adding the enzyme solution, and the reaction mixture was incu-bated by shaking at 170 rpm at 30 �C for 60 min. Reactions werestopped by heating the reaction mixture at 70 �C for 5 min and theproducts were measured on the GC–MS QP-2010 spectrometer.

The CH3SCN or CH3CN content in the reaction solution was di-rectly measured on the GC-14A chromatograph, because of theirhigh solubility in water.

Enzyme activity was defined as 1 pmol of methyl halides,CH3SH, or CH3SCN produced in 1 min under the above-mentionedconditions. All measurements of enzyme activity were performedin triplicate.

4.6. Preparation of genomic DNA and first strand cDNA fromP. tricornutum

To clone the gene encoding Pthtmt, genomic DNA and total RNAwere extracted from P. tricornutum CCAP 1055/1 cells cultivated inDaigo’s IMK medium supplemented with metasilicate for 10 days.Genomic DNA was extracted by the cetyltrimethylammonium bro-mide (CTAB) extraction method described by Dellaporta et al.(1983). Total RNA was prepared using a TRI reagent (Cosmo BioCo., Ltd., Tokyo, Japan) and used for the synthesis of first strandcDNA as a template. The first strand cDNA was synthesized byusing a PrimeScript High Fidelity RT-PCR kit (TaKaRa Bio Inc.,Shiga, Japan). All manipulations for the synthesis of first strandcDNA were performed according to the manufacturer’s protocol.

4.7. Construction of expression plasmid for PtHTMT

To obtain the full-length genome and cDNA fragments of Pthtmt,PCR was performed using genomic DNA and the first strand cDNAas templates. Two oligonucleotide primers used for PCR were de-signed based on the nucleotide sequence encoding the Pthtmt heldin a data base (GenBank Accession No.: EEC49246). The nucleotidesequences of the primers were as follows: sense primer, 50-AGATCTAATGGCCCATCGTGTTCG-30, BglII site is underlined; antisenseprimer, 50-GTCGACTTAGTCGACAGTATTCCACCAGCACAC-30, SalI siteis underlined. Amplified fragments were cloned into pGEM-TEasy vector and the nucleotide sequences were determined usinga capillary DNA sequencer 310 (Applied Biosystems, Tokyo,Japan).

The full length of the Pthtmt cDNA fragment corresponding tothe mature PtHTMT was digested with BglII and SalI. The Pthtmtfragment was separated by agarose gel electrophoresis, then ex-tracted and purified. The purified DNA fragment was cloned intoexpression vector pET-21b treated with BglII and SalI. The resultingplasmid was named pET-Pthtmt and introduced into E. coliBL21(DE3) possessing pG-KJE8.

4.8. Expression and purification of recombinant enzyme

The expression of recombinant PtHTMT was induced by grow-ing the transformed E. coli cells in LB medium containing ampicillin(50 lg/mL), chloramphenicol (20 lg/mL), arabinose (0.5 mg/mL),and tetracycline (5 ng/mL) while shaking at 37 �C. When theOD600 of the culture reached 0.5, isopropyl-b-D-thiogalactoside(IPTG) was added to the culture to give a final concentration of0.1 mM. The culture was further incubated at 18 �C for 24 h, andthen cells were collected by centrifugation at 10,000g for 20 minat 4 �C. The recombinant PtHTMT was purified by the same manneras described previously (Itoh et al., 2009). The cells were washedtwice with cell lysis buffer (20 mM MES, 0.5 M NaCl, 5 mM DTT,and 10 mM imidazole, pH 7.0), and recovered by centrifugation.Cell pellets were re-suspended in this buffer, and disrupted by

H. Toda, N. Itoh / Phytochemistry 72 (2011) 337–343 343

sonication (five times for 30 s each). Clear lysate was recovered bycentrifugation at 20,000 g for 5 min, and then loaded onto a Ni-Sepharose high-performance column (1-mL bed volume). Afterwashing the column with cell lysis buffer (10 mL), PtHTMT witha histidine tag fused at the C-terminus was eluted with the samebuffer containing 500 mM imidazole. Fractions having HTMT activ-ity were collected, and passed through an Econo-pac column fordesalting with 20 mM MES buffer (pH 7.0) containing 5 mM DTT.The sample solution thus obtained was used to characterize theenzyme.

4.9. Chemicals

S-Adenosyl-L-methionine (SAM) was purchased from Sigma.Standard gaseous mixtures of methyl halides (CH3Cl, CH3Br, andCH3I, 1 or 5 ppm in N2) were obtained from Sumitomo Seika Co.,Osaka, Japan. Standard gaseous mixtures containing CH3SH andDMS (1 and 5 ppm in N2) were supplied by Takachiho ChemicalIndustrial Co., Tokyo, Japan. Other chemical compounds were ob-tained from Wako Pure Chemicals Industries, Ltd., Osaka, Japan.

Acknowledgements

This work was supported by a Grant-in-Aid for ScientificResearch on Priority Areas and the Western Pacific Air–Sea Interac-tion Study (W-PASS), provided by The Ministry of Education,Culture, Sports, Science and Technology (MEXT) of Japan.

References

Alicke, B., Hebestreit, K., Stutz, J., et al., 1999. Iodine oxide in the marine boundarylayer. Nature 397, 572–573.

Amachi, S., Kamagata, Y., Kanagawa, T., et al., 2001. Bacteria mediate methylation ofiodine in marine and terrestrial environments. Appl. Environ. Microbiol. 67,2718–2722.

Amachi, S., Kimura, K., Muramatsu, Y., et al., 2007. Hydrogene peroxide-dependentuptake of iodine by marine Flavobacteriaceae bacterium strain C-21. Appl.Environ. Microbiol. 73, 7536–7541.

Attieh, J., Djiana, R., Koonjul, P., et al., 2002. Cloning and functional expression oftwo plant thiol methyltransferases: a new class of enzymes involved in thebiosynthesis of sulfur volatiles. Plant Mol. Biol. 50, 511–521.

Attieh, J.M., Hanson, A.D., Saini, H.S., 1995. Purification and characterization of anovel methyltransferase responsible for biosynthesis of halomethanes andmethanethiol in Brassica oleracea. J. Biol. Chem. 270, 9250–9257.

Attieh, J., Kleppinger-Sparace, K.F., Nunes, C., et al., 2000a. Evidence implicating anovel thiol methyltransferase in the detoxification of glucosinolate hydrolysisproducts in Brassica oleracea L.. Plant Cell Environ. 23, 165–174.

Attieh, J., Sparace, S.A., Saini, H.S., 2000b. Purification and properties of multipleisoforms of a novel thiol methyltransferase involved in the production ofvolatile sulfur compounds from Brassica oleracea. Arch. Biochem. Biophys. 380,257–266.

Bowler, C., Allen, A.E., Badger, J.H., et al., 2008. The Phaeodactylum genome revealsthe evolutionary history of diatom genomes. Nature 456, 239–244.

Chameides, W.L., Davis, D.D., 1980. Iodine: its possible role in troposphericphotochemistry. J. Geophys. Res. 85, 7385–7398.

Dellaporta, S.L., Wood, J., Hicks, J.B., 1983. A plant DNA minipreparation: version II.Plant Mol. Biol. Rep. 1, 19–21.

Drotar, A., Burton Jr., G.A., Tavernier, J.E., et al., 1987. Widespread occurrence ofbacterial thiol methyltransferases and the biogenic emission of methylatedsulfur gases. Appl. Environ. Microbiol. 53, 1626–1631.

Elliot, S., Rowland, F.S., 1993. Nucleophilic-substitution rates and solubilities formethyl halides in seawater. Geophys. Res. Lett. 20, 1043–1046.

Harper, D.B., Kennedy, J.T., Hamilton, J.T.G., 1988. Chloromethane biosynthesis inporoid fungi. Phytochemistry 27, 3147–3153.

Itoh, N., Shinya, M., 1994. Seasonal evolution of bromomethanes from corallinealgae (Corallinaceae) and its effect on atmospheric ozone. Mar. Chem. 45, 95–103.

Itoh, N., Toda, H., Matsuda, M., et al., 2009. Involvement of S-adenosylmethionine-dependent halide/thiol methyltransferase (HTMT) in methyl halide emissionsfrom agricultural plants: isolation and characterization of an HTMT-coding genefrom Raphanus sativus (daikon radish). BMC Plant Biol. 9, 116.

Itoh, N., Tsujita, M., Ando, T., et al., 1997. Formation and emission ofmonohalomethanes from marine algae. Phytochemistry 45, 67–73.

Jones, C.E., Carpenter, L.J., 2007. Chemical destruction of CH3I, CH2H5I, 1-C3H7I, and2-C3H7I in saltwater. Geophys. Res. Lett. 34, L13804.

Kagan, R.M., Clarke, S., 1994. Widespread occurrence of three sequence motifs indiverse S-adenosylmethionine-dependent methyltransferase suggests acommon structure for these enzymes. Arch. Biochem. Biophys. 310, 417–427.

Klick, S., Abbrahamsson, K., 1992. Biogenic volatile iodated hydrocarbons in theocean. J. Geophys. Res. 97, 12683–12687.

Labahn, J., Granzin, J., Schluckebier, G., et al., 1994. Three-dimensional structure ofthe adenine-specific DNA methyltransferase M.Taq I in complex with thecofactor S-adenosylmethionine. Proc. Natl. Acad. Sci. USA 91, 10957–10961.

Leblanc, C., Colin, C., Cosse, A., et al., 2006. Iodine transfers in the coastalenvironment: the key role of brown algae and of their vanadium-dependenthaloperoxidases. Biochimie 88, 1773–1785.

Lovelock, J.E., 1975. Natural halocarbons in the air and in the sea. Nature 256, 193–194.

Manley, S.L., Goodwin, K., North, W.J., et al., 1992. Laboratory production ofbromoform, methylene bromide, and methyl iodide by macroalgae anddistribution in nearshore southern California waters. Limnol. Oceanogr. 37,1652–1659.

Manley, S.L., Wang, N.Y., Walser, M.L., et al., 2006. Coastal salt marshes as globalmethyl halide sources from determination of intrinsic production by marshplants. Global Biogeochem. Cycles 20, GB3015.

Manö, S., Andreae, M.O., 1994. Emission of methyl bromide from biomass burning.Science 263, 1255–1257.

Martin, J.L., McMillan, F.M., 2002. SAM (dependent) I AM: the S-adenosylmethionine-dependent methyltransferase fold. Curr. Opin. Struct.Biol. 12, 783–793.

Moore, R.M., Tokarczyk, R., 1993. Volatile biogenic halocarbons in the northwestAtlantic. Global Biogeochem. Cycles 7, 195–210.

Moore, R.M., Zafiriou, O.C., 1994. Photochemical production of methyl iodide inseawater. J. Geophys. Res. 99, 16415–16420.

Nagatoshi, Y., Nakamura, T., 2007. Characterization of three halidemethyltransferases in Arabidopsis thaliana. Plant Biotechnol. 24, 503–506.

Ni, X., Hager, L.P., 1998. CDNA cloning of Batis maritima methyl chloride transferaseand purification of the enzyme. Proc. Natl. Acad. Sci. USA 95, 12866–12871.

Ni, X., Hager, L.P., 1999. Expression of Batis maritima methyl chloride transferase inEscherichia coli. Proc. Natl. Acad. Sci. USA 96, 3611–3615.

O’Dowd, C.D., Jimenez, J.L., Bahreini, R., et al., 2002. Marine aerosol formation frombiogenic iodine emissions. Nature 417, 632–636.

Ohsawa, N., Tsujita, M., Morikawa, S., et al., 2001. Purification and characterizationof a monohalomethane-producing enzyme S-adenosyl-L-methionine: halide ionmethyltransferase from a marine microalga, Pavlova pinguis. Biosci. Biotechnol.Biochem. 65, 2397–2404.

Ooki, A., Tsuda, A., Kameyama, S., et al., 2010. Methyl halides in surface seawaterand marine boundary layer of the Northwest Pacific. J. Geophys. Res. 115,C10013.

Rhew, R.C., Østergaard, L., Saltzman, E.S., et al., 2003. Genetic control of methylhalide production in Arabidopsis. Curr. Biol. 13, 1809–1813.

Saxena, D., Aouad, S., Attieh, J., et al., 1998. Biochemical characterization ofchloromethane emission from the wood-rotting fungus Phellinus pomaceus.Appl. Environ. Microbiol. 64, 2831–2835.

Scala, S., Carels, N., Falciatore, A., et al., 2002. Genome properties of the diatomPhaeodactylum tricornutum. Plant Phys. 129, 993–1002.

Vidgren, J., Svensson, L.A., Liljas, A., 1994. Crystal structure of catechol O-methyltransferase. Nature 368, 354–358.

World Meteorological Organization. Available from: <http://www.wmo.int/pages/prog/arep/gaw/ozone_2006/ozone_asst_report.html>, 2006. Scientific Assessmentof Ozone Depletion: 2006.

Wuosmaa, A.M., Hager, L.P., 1990. Methyl chloride transferase: a carbocation routefor biosynthesis of halometabolites. Science 249, 160–162.

Yokouchi, Y., Ikeda, M., Inuzuka, Y., et al., 2002. Strong emission of methyl chloridefrom tropical plants. Nature 416, 163–165.

Zafiriou, O.C., 1975. Reactions of methyl halides with seawater and marine aerosols.J. Mar. Res. 33, 75–81.