Dynamic Article LinksC<Journal ofEnvironmentalMonitoringCite this: J. Environ. Monit., 2012, 14, 845

www.rsc.org/jem PAPER

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online / Journal Homepage / Table of Contents for this issue

Stable carbon isotope composition (d13C values) of the halogenatedmonoterpene MHC-1 as found in fish and seaweed from differentmarine regions

Natalie Rosenfelder and Walter Vetter*

Received 17th October 2011, Accepted 5th December 2011

DOI: 10.1039/c2em10838k

Six samples of the red seaweed Plocamium cartilagineum (two both from Helgoland/Germany, Ireland,

and the Antarctic) were analyzed by stable isotope analysis. The bulk d13C values (�31 to �38&)

confirmed that this seaweed was exceptionally highly depleted in 13C. The d15N values were in the

reported range for algae, and a moderate correlation between 13C and 15N content was observed. These

measurements indicated that the season had a higher impact on the d13C values than the location.

Compound specific carbon isotope analysis by gas chromatography coupled with isotope ratio mass

spectrometry (GC-IRMS) was used to verify an even stronger depletion in 13C in the halogenated

natural product (HNP) MHC-1 isolated from the seaweed samples. The d13C values of MHC-1 in the

range �42 to �45& are the lightest determined to date for polyhalogenated compounds in the

molecular mass range between 250 and 800 Da. MHC-1 was also isolated from two fish samples. The

much higher 13C content in MHC-1 from the fish samples (�35.6 to �39.2&) indicated carbon isotope

enrichment due to partial transformation. In processes such as the reductive debromination, the partial

transformation is associated with an increase in the 13C content in the remaining non-metabolized share

of a compound. Stable isotope analysis may thus provide insights into the degree of transformation of

HNPs in environmental samples.

Introduction

Since the 1970s, natural product chemists have determined the

structures of more than 4500 halogenated natural products

(HNPs), predominantly after extraction from samples collected

in the marine environment.1 The major HNP producers are

sponges and algae that occur in virtually every marine environ-

ment.1 Like anthropogenic persistent organic pollutants, many

HNPs have been found to be bioaccumulative, persistent, and

toxic. Thus, HNPs are of interest because they have been

Institute of Food Chemistry (170b), University of Hohenheim,Garbenstr.28, D-70599 Stuttgart, Germany. E-mail: walter.vetter@uni-hohenheim.de; Fax: +49 711 459 24377; Tel: +49 711 459 24016

Environmental impact

In this study we isolated the halogenated natural product MHC-1

gineum as well as from two fish samples which had accumulated th

MHC-1 isolated from the six seaweed samples from Ireland, Germ

�42&). Seasonal factors had a higher impact on the d13C values th

heavier in carbon (�39 to�36&). This indicated that MHC-1 has b

and the accumulation in the fish. Thus, stable isotope analysis pr

compounds in the environment.

This journal is ª The Royal Society of Chemistry 2012

detected in food and environmental samples at elevated

concentrations.2 A commonly occurring HNP is the so-called

mixed halogenated compound 1 (MHC-1) initially identified in

fish.3 In 2008, the structure of MHC-1 was shown to

be (1R,2S,4R,5R,10E)-2-bromo-1-bromomethyl-1,4-dichloro-5-

(20-chloroethenyl)-5-methylcyclohexane, a compound previously

discovered in the red seaweed P. cartilagineum.4–6 Consequently,

P. cartilagineum was identified as a natural producer of MHC-1.

P. cartilagineum is widely distributed and abundant in diverse

climate zones such as the North Sea and Antarctic waters.

Likewise, MHC-1 was detected in all habitats of this red

seaweed.3,7 The wide distribution in different regions raised the

question of whether P. cartilagineum samples from different

from its natural producer, the red seaweed Plocamium cartila-

e natural product in their tissues.

any, and the Antarctic was depleted in 13C (d13C values �45 to

an the location. We then found that MHC-1 in fish was much

een partly degraded on its way from its release from the seaweed

oved to be a valuable tool for fate studies of polyhalogenated

J. Environ. Monit., 2012, 14, 845–851 | 845

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online

habitats would provide the same or different stable carbon

isotope signatures of MHC-1.

Stable carbon isotope analysis is a valuable tool that has been

repeatedly used in biogeochemistry, food chemistry, and envi-

ronmental chemistry. Using elemental analyzer-isotope ratio

mass spectrometry (EA-IRMS), a wide diversity of bulk d13C

values have been reported for different seaweed genera.8,9 Of all

samples analyzed by Maberly et al., P. cartilagineum proved to

be most depleted in 13C.8 This was attributed to indications that

Plocamium sp. was the only reported genus which assimilated

only CO2 and not HCO3� under natural conditions.8 Raven

et al.10 confirmed this strong depletion in 13C for P. cartilagineum.

However, they also found variations in the dependence of the

origin of the seaweed.10 For instance, d13C values of P. cartila-

gineum samples ranged from �29 to �34&.10 In contrast to bulk

d13C values of seaweed samples, isotopic signatures of individual

HNPs have been scarcely reported. When HNPs isolated from

algae and sponge extracts are analyzed by EA-IRMS, the risk of

artefacts by co-extracted compounds is high.11 For this reason,

gas chromatography coupled to IRMS is the method of choice

for compound specific isotope analysis (CSIA).

The goal of this study was to isolate MHC-1 from different

samples of P. cartilagineum and to compare the d13C values of the

HNP with those of the bulk values of the seaweed from different

origins. Finally, we attempted to determine the d13C values of

MHC-1 isolated from fish samples highly contaminated with

MHC-1.

Material and methods

Samples

Two samples from each site were analyzed and the first sample

mentioned in the following has already been described by Vetter

et al.7 The Helgoland (North Sea, Germany) samples (1.06 g and

0.6 g air-dried material) were collected by scuba diving in May

2007 and March 2008. The samples from Ireland (3.24 g lyoph-

ilized material and 5.37 g air-dried) were collected in June and

October 2007 by scuba diving (�3 m depth at low tide from

a rocky shore close to Galway). The Antarctic samples originated

from King George Island. The first sample (0.22 g air-dried

material) was collected at low tide at a beach close to Ardley in

January 2001. The second one (5.18 g air-dried material) was

collected in late January 2009 in Elephant Bay. The distance

between both sites is about 10 km.

Two fish samples from official German food inspections in

2001 and 2008 (Bavaria, Southern Germany) were selected

because of the high MHC-1 levels detected in these specifically

under previous analysis. A processed sample extract of a pollack

(Pollachius pollachius) fillet from Denmark was available from

a previous study.3 The salmon (Salmo salar) fillet sample origi-

nated from Norway and was collected in summer 2008.

Sample preparation

Cold extracts of P. cartilagineum were prepared at ambient

temperature according to Vetter et al.7 The seaweed samples were

placed in Erlenmeyer flasks and covered with n-hexane (for

residue analysis, ChemSolute, ThGeyer, Renningen, Germany).

The extraction solvent was renewed three times for each sample

846 | J. Environ. Monit., 2012, 14, 845–851

after 6 h, 24 h and 6 h, respectively. Combined extracts were

condensed to �2 mL and treated with �50% by volume

concentrated H2SO4. The organic phase was separated and

subjected to adsorption chromatography with 3 g silica gel (silica

60, for column chromatography, Fluka, Steinheim, Germany)

previously deactivated with 30% water (w/w). The elution with

60 mL n-hexane was performed according to Weichbrodt et al.12

Extracts of one P. cartilagineum sample from Helgoland (March

2008) and one from Antarctica (January 2009) were prepared by

means of accelerated solvent extraction (ASE, Dionex, Sunni-

vale, USA) with n-hexane using the instrumental setup of

Weichbrodt et al.12 One part of the P. cartilagineum sample from

October 2007 from Ireland was extracted by ASE in parallel to

the cold-extraction. Further sample treatment was the same as

shown above for cold extracts. Extraction residues were re-

extracted using the same method again. All but one re-extract

were free ofMHC-1. However, theMHC-1 residue in this sample

could not be analyzed without interferences by GC-IRMS. A

co-eluting non-halogenated compound could not be separated by

GC and would have falsified the d13C value of MHC-1 in the

residue.

The sample preparation of the pollack sample from Denmark

was described elsewhere.3 Lipophilic compounds from the

Norwegian salmon sample were gathered by ASE extraction

using a pre-distilled, azeotropic mixture (54 : 46, w/w) of ethyl

acetate (purity >99.5%) and cyclohexane (purity >99.0%), both

obtained from Sigma-Aldrich, Steinheim. Further purification

steps included gel permeation chromatography (GPC) and

adsorption chromatography on modified silica followed by

a modified group separation on 8 g activated silica.12 Eight

replicate samples of the salmon sample from Norway were

pooled for analysis. In either case, the solutions were evaporated

under a gentle stream of nitrogen and re-dissolved in iso-octane.

Gas-chromatography in combination with mass spectrometry

(GC-MS)

GC-MS analyses were performed on a Varian CP-3800/1200

triple quadrupole system (Varian, Darmstadt, Germany). Three

GC columns coated with DB-5, CP-Sil 2 and CP-Sil 19 and

different GC oven programs were tested in order to facilitate

a non-interfered elution of MHC-1 (data not shown). The best

results were obtained on the very nonpolar CP-Sil 2 column

(30 m, 0.25 mm i.d., 0.25 mm film thickness) from Chrompack

(Middelburg, The Netherlands). Helium (purity 99.9990%;

Westfalen, M€unster, Germany) was used as the carrier gas with

a constant flow rate of 1.2 mL min�1. Injections (1 mL) were

performed in splitless mode (split opened after 2 min) at 250 �C.The GC oven temperature program started at 70 �C. After 2 min,

the temperature was increased by 10 �C min�1 to 230 �C,followed by 5 �C min�1 to 270 �C and finally at 10 �C min�1 to

285 �C (hold time 10.5 min). Additionally, the following slow

oven program was used for the measurement of fish samples:

70 �C (hold time 2 min), then at 10 �Cmin�1 to 240 �C and at 1 �Cmin�1 to 265 �C and finally at 15 �C min�1 to 270 �C (hold time

5.7 min). GC-MS full scan (m/z 50–m/z 700) measurements in the

electron ionization (EI) mode were carried out at an ion source

temperature of 220 �C. In the electron capture negative ion

(ECNI) mode, nitrogen (purity 99.9990%, Westfalen, M€unster,

This journal is ª The Royal Society of Chemistry 2012

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online

Germany) was used as the reagent gas with an ion source pres-

sure of 7.0 Torr.13 The ion source temperature was held at

200 �C. GC/ECNI-MS full scan analyses covered the mass range

m/z 50–m/z 650.

Gas chromatography in combination with flame ionization

detection (GC/FID)

Quantification of MHC-1 in individual sample extracts was

performed using GC/FID. In addition, the abundances of

MHC-1 in the individual samples were adjusted to equal

concentrations by means of GC/FID analyses performed with

the system previously described by Vetter et al.7 The chromato-

graphic conditions were the same as those used in GC-MS except

that a constant column head pressure of 1.25 bar was used. Final

sample adjustment was done by dilutions in iso-octane (Pestanal,

Fluka, Darmstadt, Germany) or evaporation under a gentle

stream of nitrogen to a specific volume. By this measure, the

sample solutions were adjusted by GC/FID to give concentra-

tions of 65 � 5 ng mL�1 MHC-1. Exceptions formed the first

seaweed sample from the Antarctic (�15 ng mL�1 MHC-1) and

the pollack sample from Denmark (�25 ng mL�1 MHC-1) due to

limited amounts that were available. Once these measurements

were finished, the CP-Sil 2 column was installed into the

GC-IRMS system.

Isotope ratio mass spectrometry (IRMS)

Isotope ratio mass spectrometry (IRMS) analyses were per-

formed on a Delta plus XP instrument (Thermo Finnigan MAT,

Bremen, Germany). Stable nitrogen and carbon isotope values

were expressed in per mil (&) relative to the reference scale of

N2-Air for nitrogen isotope values and VPDB for carbon isotope

values (International Atomic Energy Agency (IAEA), Vienna,

Austria). They are defined as shown in eqn (1):

d13C or d15N ¼ ((Rsample/Rreference) � 1) � 1000 (1)

where R is the isotopic ratio of the heavy (13C or 15N) to the light

isotope (12C or 14N).

The IRMS instrument was equipped with a Euro EA 3000

elemental analyzer (Hekatech, Wegberg, Germany) as well as

with a GC.14–16 The two reference gases, N2 for the determination

of d15N values and CO2 for the determination of d13C values, were

calibrated by elemental analysis. USGS-40 (L-glutamic acid)

(IAEA, Vienna, Austria), certified reference material17,18 with

known 13C and 15N isotopic composition, was repeatedly

measured (a minimum of six runs in a row). The average of the

resulting d13C value and d15N value of the certified standard was

compared to the known isotopic composition (d13C ¼ �26.24%

relative to VPDB and d15N ¼ �4.52% relative to N2-Air).17 The

resulting difference between the measured and theoretical values

was used to adjust the d13C and d15N values of the reference gases,

regularly. Subsequent measurements of USGS-40 with devia-

tions <0.1& from theoretical to measured delta values confirmed

the accuracy of the calibration of the IRMS system. In a second

step, the in-house standard sodium L-glutamate was repeatedly

measured against the calibrated reference gas (d13C ¼ �28.07%

relative to VPDB and d15N ¼ �2.72% relative to N2-Air).

This journal is ª The Royal Society of Chemistry 2012

EA-IRMS measurements were performed relative to this

calibrated standard (see below).

For bulk d13C and d15N value analysis, approximately 1 mg of

air-dried samples (except one sample from Ireland which was

lyophilized) was loaded into 4 � 6 mm tin capsules (IVA Ana-

lysentechnik, Meerbusch, Germany) and combusted in the EA

using the parameters previously described.14,15 Overall measure-

ment uncertainty (including instrumental performance, sample

combustion by the elemental analyzer and sample homogeneity)

was determined by running all samples in triplicates (see below).

The in-house standard sodium L-glutamate was run at the

beginning and the end of each sample sequence to verify constant

instrumental conditions as well as to monitor the correct refer-

encing of the d values to the international reference scale (see

above). The same d13C (deviation <0.2&) and d15N (deviation

<0.2&) values were determined at the beginning and the end of

the measured sequence and no samples had to be discarded.

Following this protocol, the standard deviations (n ¼ 3) for

determined d13C and d15N values were # � 0.2& for all samples

in a sequence.

Compound specific isotope analysis (CSIA) measurements by

GC-IRMS were based on the method described in detail else-

where.16 Chromatographic separation on CP-Sil 2 was done as

described for GC-MS. Constant instrumental performance was

verified by running a BDE 47 standard (d13C value �30.0&)

(synthesized in our lab according to published procedures19) and

a sample of isolated MHC-1 (d13C �43.3&, see below) at the

beginning and end of each sequence. All samples were run in

minimum of triplicates. The volumes of the sample extracts were

adjusted to deliver similar peak areas of MHC-1 in the

GC-IRMS chromatograms because high variations of the peak

areas (> 50%) resulted in varying d13C values of up to 1&.

Standard deviations (n $ 3) for determined d13C values of

MHC-1 ranged from �0.2& to �0.5& in the different sample

extracts. In addition, BDE 47 was measured at least once per

sequence and its d13C value was compared to previous

measurements.16 The resulting d13C values of BDE 47 were

generally within the range of �30.0 � 0.5& previously

reported.16

Results and discussion

Bulk stable carbon (d13C values) and nitrogen (d15N values)

isotope ratios of seaweed (P. cartilagineum) samples

The bulk d13C values of the dried seaweed samples ranged

from �31& to �38& (Table 1). In accordance with reported

bulk d13C values in the literature (�29.1& to �34.5&8,10), the

samples of P. cartilagineum were heavily depleted in 13C. Our

measurements confirmed that P. cartilagineum belongs to the

seaweed samples lightest in carbon.10

The strongest depletion in 13C was observed in the sample

collected in March at Helgoland (Table 1). This sample was more

depleted in carbon than any other sample of P. cartilagineum

analyzed so far. Moreover, the origins of the seaweed samples

seemed to have less of an impact on the d13C values compared to the

season during which the samples were collected. For instance, the

samples collected in May/June in Helgoland/Ireland showed the

same d13C value of �34.2&. By comparison, the d13C value of the

J. Environ. Monit., 2012, 14, 845–851 | 847

Table 1 Origins, collection date and stable isotope data of six samples of the red seaweed Plocamium cartilagineum as well as its halogenated naturalproduct MHC-1 from the seaweed and from fish

Sample (sampling site) DateConcentration ofMHC-1 [mg g�1]a d13C value of MHC-1 [&] d13C bulk [&] d15N bulk [&]

P. cartilagineum ASE-extract (Helgoland) March 2008 177 �45.0 (�0.4) �38.1 (�0.1) 10.2 (�0.1)P. cartilagineum Cold-extract (Helgoland) May 2007 1900 �43.3 (�0.3) �34.2 (�0.2) 9.8 (�0.1)P. cartilagineum Cold-extract (Ireland) June 2007 9 �43.1(�0.2) �34.2 (�0.1) 8.4 (�0.2)P. cartilagineum Cold-extract (Ireland) October 2007 40 �42.3 (�0.5)b �33.1 (�0.2) 7.2 (�0.1)P. cartilagineum Cold-extract (Antarctica) January/July 2001c 2 �42.9 (� 0.4) �32.3 (�0.2) 5.7 (�0.1)P. cartilagineum ASE-extract(Antarctica) January/July 2009 c <1 Not detected �31.1 (�0.01) 5.2 (�0.1)Pollack ASE-extract (Denmark) 2001 2.3 (l.w.) �35.6 (�0.4) Not determined Not determinedSalmon ASE-extract (Norway) Summer 2008 0.6 (w.w) �39.2 (�0.5) �25.4 (�0.2) 10.8 (�0.1)

a Dry weight except where noted; l.w.¼ lipid weight, w.w.¼ wet weight. b The d13C value in an ASE-extract was�42.0� 0.4& (concentration 37 mg g�1

dry weight). c January in the Southern Hemisphere corresponds with July in the Northern Hemisphere.

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online

Helgoland sample, collected in March was depleted by 3.9& while

the d13C value of the October sample collected in Ireland

was enriched by 1.1& (Table 1). These differences in the carbon

isotopic signatures in the different seasons were about one order of

magnitude larger than the variations in the method. The

samples collected in Antarctica in January were the least depleted

in 13C (Table 1).

The bulk d15N values of the seaweed samples ranged from

5.2& to 10.2& (Table 1). Dunton reported a slightly less

positive d15N value for P. cartilagineum of 3.6&.20 Vizzini and

Mazzola measured d15N values in the range of our samples (6.4–

8.2&) for two green algae from the Mediterranean Sea.21 These

measurements indicated that the bulk d15N values of P. carti-

lagineum were in the typical range of other seaweed species

(in contrast to the d13C values). The Plocamium samples from

the Antarctic were the most depleted in 15N, which was attrib-

uted to the distinctly different environmental conditions in this

polar region (Table 1). Interestingly, we observed a negative

correlation between stable carbon and stable nitrogen isotope

ratios. Samples most depleted in carbon were most enriched in

nitrogen (Fig. 1). The coefficient of determination (R2 ¼ 0.7827,

corresponding correlation coefficient r ¼ 0.8847) with a slope of

��1 between these isotopic signatures indicated that 78.27% of

the total variability can be explained by this linear relationship.

One factor contributing 78.27% to the total variability is

considered high for biological systems but it could not be

explained.

Fig. 1 Correlation of the bulk d13C values [&] over bulk d15N values [&]

of six Plocamium cartilagineum samples collected at different time at three

different locations.

848 | J. Environ. Monit., 2012, 14, 845–851

Halogenated natural products in six samples of P. cartilagineum

One sample of P. cartilagineum from each of the three sites had

been analyzed before.7 These three samples showed different

HNP patterns and MHC-1 was the only HNP detected in all the

samples.7 Differences in the HNP patterns were verified in the

new set of samples from Helgoland, Ireland and the Antarctic.

While MHC-1 was dominant in the samples from Helgoland/

Germany, its concentration in the Antarctic was low in one

sample and it was not detected in the second sample (Table 1).

Differences were also observed in the samples from Ireland where

MHC-1 was more than four fold higher concentrated in one

sample. Variations in the pattern of halogenated monoterpenes

in P. cartilagineum from different marine regions including the

Antarctic and the North Sea were also reported and summarized

by San-Mart�ın and Rovirosa.22Twelve HNPs in all were detected

in the six seaweed samples (Fig. 2). This rich palette is in agree-

ment with the range of alicyclic and aliphatic halogenated

monoterpenes identified in P. cartilagineum samples worldwide

since the 1970s.4,5,23–25

GC/EI-MS spectra of alicyclic compounds showed more

abundant fragment ions in the high mass range, and partially

even the molecular ion was detected (with up to 10% of the

intensity of the base peak). By contrast, the GC/EI-MS spectra

of aliphatic monoterpenes were characterized by strong frag-

mentation, and the C4 fragment ions m/z 167 ([C4H5BrCl]+) and

m/z 123 ([C4H5Cl2]+) (see Fig. 3) were frequently found to be the

base peak in accordance with Stierle and Sims.23 Compound 1

and compound 2 were tentatively identified as the alicyclic

monoterpenes shown in Fig. 3 due to the excellent match with

MS data provided by Stierle and Sims.4 The corresponding

molecular ion was found at m/z 238 (C10H13Cl3). Compound 3

showed the base peak at m/z 165, which corresponds with

[C10H10Cl]+. The low abundant m/z 244 was tentatively identi-

fied as [C10H10BrCl]+. Although this mass agreed with the

molecular mass of other compounds eluting in this retention

time range, this molecular formula would require five double

and/or ring bonds, which is difficult to construct. Therefore, the

nature of this compound could not be unravelled. The charac-

teristic C4 fragment ion of compound 4 showed at m/z 123 (see

above) pointed to an aliphatic monoterpene backbone. This

compound also featured two very low abundant fragment ions

at m/z 226 ([C6H7BrCl2]+) and m/z 202 (corresponding with the

This journal is ª The Royal Society of Chemistry 2012

Fig. 2 GC/EI-MS full scan analysis of the halogenated natural products

extracted from Plocamium cartilagineum samples. (a) Re-extracted

sample from Helgoland (Germany) from May 2007; (b) ASE extract of

the sample fromHelgoland (Germany) fromMarch 2008; (c) cold extract

of the Ireland sample from June 2007; (d) ASE extract of the Ireland

sample from October 2007; (e) cold extract of the Ireland sample from

October 2007; (f) cold extract of the Antarctic sample from January 2001

and (g) ASE extract of the Antarctic sample from January 2009. Peaks

labelled with an * represent non-halogenated compounds.

Fig. 3 Structure of characteristic fragment ions of aliphatic mono-

terpenes frequently detected by GC/EI-MS m/z 167 ([C4H5BrCl]+) and

m/z 123 ([C4H5Cl2]+), tentative structures of halogenated metabolites

(compounds 1,2 and 5) and structure of MHC-1 from Plocamium

cartilagineum.

This journal is ª The Royal Society of Chemistry 2012

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online

elimination of 26 Da or a neutral loss of C2H2 from the

former one).

By contrast, compound 5 was alicyclic and the highest

mass was found at m/z 247 ([C10H13BrCl]+). The dichloro

pattern of m/z 203 indicated that the molecular ion could be m/z

282 (a low abundant cluster of isotope peaks was observed at

this mass), which agrees with structure 5 (Fig. 3) previously

described by Stierle and Sims.4 The highest mass for compound

6 was at m/z 239, corresponding with [C10H14Cl3]+. The

molecular ion of the alicyclic compound could be either m/z 274

([C10H14Cl4]+) or m/z 318 ([C10H14BrCl3]

+). Congeners with this

feature were isolated by K€onig et al. from Plocamium hamatum.6

A small fragment ion detected at m/z 282 indicated the presence

of the [M-Cl]+ fragment ion of the BrCl3 variant. According to

the GC retention time, the C10H14BrCl3 variant was more likely.

This molecule differs from MHC-1 by the lack of one bromine

substituent.

Compounds 7 and 8 were most likely isomers. Both formed

the base peak at m/z 114 and also shared m/z 167 for aliphatic

Br/Cl containing monoterpenes. Compound 9 was identified by

the molecular ion at m/z 362 as an alicyclic monoterpene with

the composition C10H14Br2Cl2 (which lacks one chlorine

substituent compared to compound 12, i.e. MHC-1).

Compound 10 and 11 were aliphatic and featured m/z 123

(compound 10) or m/z 167 (compound 11). Apparently,

both compounds were structurally related and may only have

differed in the substitution of the Br/Cl in the C4 fragment ions

shown in Fig. 3.

Compound specific d13C values of MHC-1 isolated from P.

cartilagineum

Initial GC/EI-MS measurements demonstrated that MHC-1

could be determined in all samples without interference on the

CP-Sil 2 column using the GC oven program shown in the

section Material and Methods. Using GC-IRMS we found

that the strong depletion observed in the bulk d13C values of

the seaweed was even more pronounced in the compound-

specific analysis of MHC-1. The d13C values of MHC-1

ranged from �42& to �45& (Table 1). These d13C values

indicated that MHC-1 was more depleted in 13C than any

anthropogenic organohalogen compound with m.w. >200 Da

analyzed so far.14,26 Only very few small and volatile haloge-

nated compounds such as CHCl3 and CCl4 were similarly

depleted in 13C.26,27 Furthermore, the impact of seasonal

trends on the d13C values tentatively observed in the bulk d13C

values of the seaweed samples was also observed for MHC-1.

Considering that the Antarctic is in the Southern Hemisphere,

the samples collected there in January (Antarctic summer)

corresponded with July in the Northern Hemisphere (Fig. 4).

If the Antarctic samples were assigned to the season, there

was a trend visible with increasing d13C values of MHC-1

from spring (March) to autumn (October). In this period of

time, MHC-1 became steadily heavier in carbon (Fig. 4a). For

all except the Antarctic sample, the progression of the d13C

value was similar for MHC-1 and the bulk value of the

seaweed (Fig. 4b).

J. Environ. Monit., 2012, 14, 845–851 | 849

Fig. 4 Correlation of a) CSIA d13C values [&] of isolated MHC-1 from

Plocamium cartilagineum and b) bulk d13C values [&] of Plocamium

cartilagineum with different collection times at the three different loca-

tions (H: Helgoland (Germany), I: Ireland, A: Antarctica). The month of

the Southern and Northern Hemispheres (SH and NH) are assigned on

the x-axis.

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online

d13C values of other halogenated monoterpenes from P.

cartilagineum

Due to limited sample amounts only a few further halogenated

monoterpenes detected in the seaweed samples could be studied

by GC-IRMS (Table 2). These measurements showed that

MHC-1 was generally the halogenated compound most depleted

in 13C. As shown above, MHC-1 was the highest halogenated

monoterpene in this series (Table 2). Such a correlation of

depletion in 13C with increasing number of halogen atoms was

also reported for industrial halogenation reactions.26,28 In

chemical reactions, this is because covalent bonds between heavy

isotopes are less easily formed. Thus, halogens preferably add to

the lighter 12C isotope. Carbon isotope discrimination in the

formation of carbon-halogen-bonds could also take place during

the enzymatic biohalogenation of HNPs. As shown above,

compound 6 and compound 9 differed from MHC-1 in that they

Table 2 GC retention times, tentative structure information and d13C valuesPlocamium cartilagineum

No. tR [min] Back-bonea Formula Detectedb Helgolan

1 16.4 C C10H13Cl3 I —2 16.7 C C10H13Cl3 H, I �38.0 (�3 17.1 — — H, A �42.9 (�4 17.4 A — A —5 17.9 C C10H13BrCl2 H, I —6 19.3 C C10H14BrCl3 H, I —7 19.4 — — I —8 19.8 — — I, A —9 20.3 C C10H14Br2Cl2 I —10 21.0 A — I —11 22.4 A — I —12 23.1 C C10H13Br2Cl3 H, I, A —

a C¼ alicyclic, A¼ aliphatic monoterpene. b Detected in samples fromH¼Hextract.

850 | J. Environ. Monit., 2012, 14, 845–851

have one halogen less (Table 2). Compound 6 with one less Br

than MHC-1 was more depleted in 13C compared to compound 9

with one Cl less (Table 2).

d13C values of MHC-1 from fish

MHC-1 was also isolated from the two fish samples that had

bioaccumulated this HNP similarly to anthropogenic persistent

organic pollutants (POPs). In the fish samples, MHC-1 (d13C

values of �35.6 and �39.2&) was more enriched in 13C than in

any of the seaweed samples (Table 1). Moreover, the difference in

the d13C values of�7& ofMHC-1 in the pollack compared to the

heaviest seaweed sample was larger than the variations associ-

ated with seasonal effects in the seaweed. Hence, MHC-1 in the

fish must have become heavier in carbon due to its partial

degradation. Enrichment in 13C due to food chain enrichment

seemed to be less likely because isotopic fractionation in animals

during assimilation and respiration is small.29 By contrast,

reductive dehalogenation of anthropogenic POPs was accom-

panied with an increase in 13C in the portion of the compound

that remained unmetabolized.16,26 Therefore, it is very likely that

the high d13C values in fish were, at least in part, the result of the

partial degradation of MHC-1 (most likely in the fish itself or by

other organisms in its food chain). Enrichment in 13C would be

highest if MHC-1 was degraded without concurrent uptake by

the fish (either from its feed or from water). Consequently, the

enrichment of 13C during the degradation of MHC-1 will be the

more reflected in the d13C value, the less it is superimposed by the

input of isotopically lighter MHC-1. Due to seasonal variations

in the production and availability of HNPs,30 this scenario

pertains more to HNPs than anthropogenic POPs. Little data is

available on the effect of transformation of anthropogenic POPs

on the d13C values. For the UV transformation of PBDEs,

transformation of about 3/4 of a highly-brominated congener

was accompanied with an increase in the 13C content by �3&when the loss of the compound was not compensated by new

input.16 Due to the rather constant presence of anthropogenic

POPs in the marine environment, transformation without

parallel uptake is unlikely. Thus, smaller variations and more

stable d13C values would be more expected for anthropogenic

POPs.

occurrence of twelve halogenated natural products in different samples of

d, March 2008 Ireland, June 2007 Ireland, October 2007

�40.4 (�0.9) —0.3) �38.8 (�0.7) �41.1 (�0.03)d

0.3) — —— —— �37.9 (�0.5)d

�42.4 (�1.3) �42.1 (�0.4)d

— —— —�41.8 (�0.8) —— —— —�43.1 (�0.2) �42.0 (�0.4) c �42.3 (�0.5) d

elgoland (Germany), I¼ Ireland, A¼Antarctica. c ASE extracted. d Cold

This journal is ª The Royal Society of Chemistry 2012

Publ

ishe

d on

17

Janu

ary

2012

. Dow

nloa

ded

by U

nive

rsity

of

Cal

ifor

nia

- Sa

nta

Cru

z on

25/

10/2

014

15:3

2:16

. View Article Online

Conclusions

Stable isotope analysis was successfully used to show that tissue

material of P. cartilagineum and MHC-1 isolated from the

seaweed was isotopically depleted in 13C. By contrast, the higher13C content ofMHC-1 isolated from fish indicated that a relevant

part of the compound had been transformed since its release

from the seaweed. Compared to effects observed during light-

induced hydrodebromination of anthropogenic PBDEs (see

above), the degree of transformation could be >75%. More

studies are required to verify the comparability of the dimension

of these effects. In the long term, stable carbon isotope analysis

may represent a valuable tool to add a time scale and fate

dimension to the analysis of HNPs and POPs.

Acknowledgements

We are grateful to Andreas Wagner (Alfred-Wegener-Institut f€ur

Polar- und Meeresforschung, Germany) for the collection and

organisation of the seaweed samples from Helgoland, Stefan

Kraan (Irish Seaweed Centre, Galway, Ireland) for the collection

and organisation of the seaweed samples from Ireland, and

Christina Buesser (Institute of Ecology, Friedrich-Schiller-

University of Jena) for the collection and transport of the 2nd

Antarctic seaweed sample. Hans-Ulrich Peter (Institute of

Ecology, Friedrich-Schiller-University of Jena) is acknowledged

for the organisation and guidance of the Antarctic expedition

2000/2001 during which the first Antarctic seaweed sample was

collected by one of the authors. We are grateful to Dr J. Hiebl

(LGL Oberschleissheim) for the donation of the fish samples. We

also thank Axel Lang (Institute of Food Chemistry of the

University of Hohenheim) for valuable input during the cali-

bration of the IRMS instrument.

References

1 G. W. Gribble, Prog. Chem. Org. Nat. Prod., 2010, 91, 1–505.2 W. Vetter, Rev. Environ. Contam. Toxicol., 2006, 188, 1–57.3 W. Vetter, J. Hiebl and N. J. Oldham, Environ. Sci. Technol., 2001, 35,4157–4162.

4 D. B. Stierle and J. J. Sims, Tetrahedron, 1979, 35, 1261–1265.

This journal is ª The Royal Society of Chemistry 2012

5 M. D. Higgs, D. J. Vanderah and D. J. Faulkner, Tetrahedron, 1977,33, 2775–2780.

6 G. M. K€onig, A. D. Wright and A. Linden, Phytochemistry, 1999, 52,1047–1053.

7 W. Vetter, N. Rosenfelder, S. Kraan and J. Hiebl,Chemosphere, 2008,73, 7–13.

8 S. C. Maberly, J. A. Raven and A. M. Johnston, Oecologia, 1992, 91,481–492.

9 J. S. Albertus, A. Brearley, G. A. Hyndes, P. S. Lavery andD. I. Walker, Aquat. Bot., 2006, 84, 277–282.

10 J. A. Raven, A. M. Johnston, J. E. K€ubler, R. Korb, S. G. McInroy,L. L. Handley, C. M. Scrimgeour, D. I. Walker, J. Beardall,M. Vanderklift, S. Fredriksen and K. H. Dunton, Funct. PlantBiol., 2002, 29, 355–378.

11 E. L. Teuten, L. Xu and C. M. Reddy, Science, 2005, 307, 917–920.12 M. Weichbrodt, W. Vetter and B. Luckas, J. Assoc. Off. Anal. Chem.

Int., 2000, 83, 1334–1343.13 N. Rosenfelder andW. Vetter,Rapid Commun.Mass Spectrom., 2009,

23, 3807–3812.14 W. Vetter, W. Armbruster, T. R. Betson, J. Schleucher, T. Kapp and

K. Lehnert, Anal. Chim. Acta, 2006, 577, 250–255.15 N. Rosenfelder, K. Lehnert, S. Kaffarnik, J. P. M. Torres, M. Vianna

and W. Vetter, Environ. Sci. Pollut. Res., 2011, 19, 379–389.16 N. Rosenfelder, P. Bendig and W. Vetter, Environ. Pollut., 2011, 159,

2706–2712.17 H. Qi, T. B. Coplen, H. Geilmann, W. A. Brand and J. K. Boehlke,

Rapid Commun. Mass Spectrom., 2003, 17, 2483–2487.18 T. B. Coplen, W. A. Brand, M. Gehre, M. Groening, H. A. J. Meijer,

B. Toman and R. M. Verkouteren, Anal. Chem., 2006, 78, 2439–2441.19 W. Vetter and W. Jun, Chemosphere, 2003, 52, 423–431.20 K. H. Dunton, Am. Zool., 2001, 41, 99–112.21 S. Vizzini and A. Mazzola, Mar. Biol., 2003, 142, 1009–1018.22 A. San-Mart�ın and J. Rovirosa,Biochem. Syst. Ecol., 1986, 14, 459–461.23 D. B. Stierle, R. M. Wing and J. J. Sims, Tetrahedron, 1979, 35, 2855–

2859.24 P. Crews, J. Org. Chem., 1977, 42, 2634–2636.25 R. S. Norton, R. G. Warren and R. J. Wells, Tetrahedron Lett., 1977,

18, 3905–3908.26 W. Vetter in Handbook of Environmental Isotope Geochemistry,

Advances in Isotopic Geochemistry, ed. M. Baskaran, SpringerBerlin Heidelberg, 2012, ch. 32, pp. 679–705.

27 L. Zwank, M. Berg, T. C. Schmidt and S. B. Haderlein, Anal. Chem.,2003, 75, 5575–5583.

28 W. M. Jarman, A. Hilkert, C. E. Bacon, J. W. Collister,K. Ballschmiter and R. W. Risebrough, Environ. Sci. Technol.,2001, 32, 833–836.

29 M. J. DeNiro and S. Epstein, Geochim. Cosmochim. Acta, 1978, 42,495–506.

30 J. Melcher, M. Schlabach, M. S. Andersen and W. Vetter, Arch.Environ. Contam. Toxicol., 2008, 55, 547–557.

J. Environ. Monit., 2012, 14, 845–851 | 851