Research Halogenated Natural Products in Five Species of Antarctic Sponges: Compounds with POP-like Properties? W A L T E R V E T T E R * ,† A N D DORTE JANUSSEN‡ Institute of Food Chemistry, University of Hohenheim, D-70593 Stuttgart, Germany, and Section Marine Invertebrates I, Senckenberg Research Institute and Natural Museum, D-60325 Frankfurt/Main, Germany
Purified extracts of five species of Antarctic sponges (Demospongiae: Kirkpatrickia variolosa, Artemisina apollinis, Phorbas glaberrima, and Halichondria sp. and Calcarea: Leucetta antarctica) from King George Island were analyzed by GC/MS for the presence of persistent and lipophilic halogenated compounds to identify bioaccumulative halogenated natural products. Sample extracts were prepared using methods identical for the determination of POPs, namely, microwave-assisted extraction with organic solvents, gel permeation chromatography, and column chromatography on deactivated silica. In addition, samples were treated with sulfuric acid to remove aciddestructible compounds. PCBs were not detectable and only traces of lindane, p,p′-DDE, and R-HCH were detected in these samples in decreasing order of abundance, underscoring their uncontaminated state. In contrast, 146 brominated compounds were identified by correct isotopic ratios m/z 79 and 81, 50% of which eluted prior to lindane including the most abundant peaks. Each sponge sample contained g35 brominated compounds of natural origin, 14 of which were detected in all species. Estimated concentrations ranged from the high ng/kg to mg/kg (air-dried weights) and relative distributions of the same compounds in different sponges were highly variable. The high abundance of these compounds relative to known anthropogenic pollutants strongly suggests a natural origin. Multiple mode (EI-, ECNI-, and PCI-) GC/MS enabled identification of an aliphatic ketone tentatively identified as 1,1,2tribromo-oct-1-en-3-one, present in all species but highest in Phorbas glaberrima. Several halogenated phenols including 2,4,6-tribromophenol were also abundant in Phorbas glaberrima as were halogenated anisoles in lower relative abundances. The halogenated phenols were analyzed without derivatization. The sample of Halichondria sp. contained the dibromotrichloro monoterpene MHC1, a recently described environmental contaminant in fish and seals. Retrospective analysis of other marine samples confirmed that 2,4,6-tribromophenol was present in seal blubber from both the Arctic and the Antarctic. The presence * Corresponding author phone: +49 711 459 4016; fax: +49 711 459 4377; e-mail:
[email protected]. † University of Hohenheim. ‡ Senckenberg Research Institute and Natural Museum. 10.1021/es0484597 CCC: $30.25 Published on Web 04/21/2005
2005 American Chemical Society
of naturally occurring organohalogens such as 2,4,6tribromophenol and MHC-1 in Antarctic marine invertebrates thus provides a link to their occurrence in marine mammals.
Introduction Persistent organic pollutants (POPs) continue to be a serious threat to diverse environmental compartments. This group of chemicals comprises several halogenated compounds used as pesticides (e.g., DDT, lindane, chlordane, toxaphene, and others) as well as industrial chemicals (e.g., PCBs, PCNs, CPs, and brominated flame retardants). A wide range of compounds with diverse chemical structures have been identified in this context. Furthermore, chromatographic peaks of unknown structure detected in extracts containing POPs continue to surface in trace pollutant studies, since they display the same properties in standard cleanup procedures. Only recently, it was found that many of these “unknowns” originate from halogenated natural products (HNPs) which possess POP-like properties. Sometimes, HNPs have been found at remarkably high concentrations in marine birds, mammals, and fish (1-6). Because of their structural similarities with anthropogenic POPs, these substances may possess a potential risk for wildlife and man. Although the natural producers of HNPs remain largely unknown, marine plants and invertebrates have been reported as likely producers (7-12). For instance, a sponge from Australia was identified as the natural source of two HNPs previously identified in marine mammals (4). In this study, we collected and analyzed different species of Antarctic sponges for the occurrence of lipophilic and potentially persistent HNPs using protocols standardized for nonpolar POPs (13-14). Because our primary goal was to detect HNPs that persist in the environment, we treated all samples with concentrated sulfuric acid to eliminate labile halogenated compounds.
Materials and Methods Sample Collection. Sponges were collected by divers at 6-15 m depth in Maxwell Bay of King George Island (62°14′ S, 58°40′ W) on January 1 and 4, 2001. Samples were stored frozen until analysis. Portions of all samples were washed with water and treated with a potassium hypochlorite solution to dissolve the sponge soft body in order to obtain pure skeletal preparations. Furthermore, histological sections of paraffin preparations were made to investigate the skeletal architecture. The following species were investigated: Demospongiae: Kirkpatrickia variolosa (internal lab code sponge1), Halichondria sp. (sponge-3), Artemisina apollinis (sponge5), Phorbas glaberrima (sponge-7 and sponge-10), and Leucetta antarctica (sponge-12). In addition, we analyzed a sample extract of hooded seal (Cystophora cristata) from Jan Mayen (Northern Sea, Arctic, 1990), a sample extract of Weddell seal (Leptonychotes weddelli) from the Weddell Sea (Antarctic, 1990), and mussels from La Paz (Mexico, 2001) and New Zealand (origin not specified, 2003). Standards. Pesticide standards and standards of further anthropogenic compounds were from LGC Promochem (Wesel, Germany) or from Dr. Ehrenstorfer (Augsburg, Germany). 2,4-Dibromophenol, 2,6-dibromophenol, 2,4,6tribromophenol, and 2,4,6-tribromoanisole were from SigmaAldrich (Taufkirchen, Germany) or Lancaster (Frankfurt, Germany). VOL. 39, NO. 11, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3889
Sample Preparation. In the laboratory, individual species were sorted and washed with water. Subsamples were airdried for 48 h. Four to seven grams air-dried sample material was extracted by focused, open vessel microwave-assisted extraction (FOV-MAE) using ethyl acetate/cyclohexane (v/v, 1:1) as the solvent (13). The extraction program was 90 W for 7 min, 135 W for 8 min, and 150 W for 20 min. After solvent adjustment to 10 mL, the lipid phase was separated by gel permeation chromatography (GPC on BioBeads SX-3) using the extraction solvent mixture. After lipid removal, extracts were filtered through sodium sulfate, exchanged to isooctane, and concentrated to ∼1 mL using a gentle stream of nitrogen. Extracts were then brought up with 10 mL n-hexane and 5 mL sulfuric acid was added. The samples were stopped, shaken for ∼1 min, and stored for 1 week (H2SO4 was renewed three times). The concentrated n-hexane-phase (1-2 mL) was purified using a 1-cm i.d. glass column filled with 3 g of silica deactivated with 30% water (w/w) and eluted with 60 mL n-hexane. Final extracts were concentrated to 2.0 mL prior to GC analyses. The possibility that some of the detected compounds were produced during the acid treatment of the sponge extracts was investigated with three sponge samples. The samples were cleaned as described but were not treated with sulfuric acid. Comparison of treated/untreated samples confirmed that the bromophenols were actually in this form present in the samples. Only minor changes were observed but the H2SO4 treated samples were cleaner. Additional samples of marine organisms (e.g., seal blubber, fish, and mussels) were purified using the same methods excluding the lyophilization step. GC/MS Analyses. GC/MS analyses were performed with a Hewlett-Packard 5890/5989 GC/MS system. The ion source and quadrupole temperatures were set at 150 °C and 200 °C, respectively. He (quality 5.0, i.e., 99.9990% purity) was used as carrier gas at a constant flow rate of 1 mL/min. The injector temperature was 250 °C, and samples were introduced in the splitless mode (split opened after 1.5 min). An Ultra-2 column (50 m × 0.25 mm i.d. × 0.18 µm df) was installed in the oven. The GC oven was programmed as follows: 80 °C, hold time 1 min, ramp 1 at 20 °C/min to 150 °C, ramp 2 at 2 °C/min to 210 °C, hold time 5 min, ramp 3 at 20 °C/min to 280 °C, hold time 15 min. The total run time was 58 min. Additional measurements were performed with an HP-5 column (30 m, 0.25 mm i.d., 0.18 µm df). The GC oven on this column was the following: injection at 80 °C (hold time, 1 min), then at 20 °C/min to 150 °C (hold time, 2 min), at 2 °C/min to 200 °C (hold time, 5 min), and at 20 °C/min to 280 °C (hold time, 15 min). In electron capture negative ionization mass spectrometry using selected ion monitoring (GC/ECNI-MS-SIM) mode, m/z 35, 37, 79, 81, 114, 116, 158-161, 386, and 388 were recorded throughout the run (6-58 min) (15, 16). The CI gas was methane (4.5 quality, 99.995% purity). In the GC/ECNIMS, GC/PCI-MS, and GC/EI-MS, full scan modes m/z 33650, m/z 33-550, and m/z 33-450 were recorded after solvent delays of 6-8 min using otherwise the same conditions as shown above. Confirmatory analyses were carried out with a Varian 1200 GC/MSMS system operated in EI and ECNIMS modes.
Results and Discussion Brominated Natural Products versus Chlorinated POPs. Initial identification of halogenated compounds was based on GC/ECNI-MS-SIM measurements in which m/z 79 and m/z 81 (selective for organobromines) and m/z 35 and m/z 37 (selective for organochlorines) were monitored (see Materials and Methods). Of the anthropogenic chlorinated pollutants, lindane, R-HCH, and p,p′-DDE in decreasing order of abundance were detected in all samples at very low concentrations (Figure 1). PCBs were not identified in any 3890
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 11, 2005
FIGURE 1. GC-ECNI-MS-SIM (m/z 35) chromatogram of sponge extract (Leucetta antarctica) showing anthropogenic chlorinated POPs. sample. This is consistent with recent findings that marine mammals from the vicinity of King George Island contained the lowest levels of synthetic organohalogens reported to date (14). Since it is doubtful that sponges are able to metabolize POPs, our analyses indicate that lindane is more relevant at King George Island than R-HCH. In contrast, a large number of brominated compounds were detected using the bromide ion isotopes (m/z 79 and 81) as measured in the ECNI-MS-SIM mode. Compounds were considered brominated when (a) the ratio of m/z 79 to 81 was 1.00 ( 0.10, (b) the integration start and end times of both ion traces were identical or nearly so ((0.5 s), and (c) the signal-to-noise ratio was >6. On the basis of these criteria, our samples contained a total of 146 brominated compounds. In this paper, however, we report on only those 88 compounds that reached an S/N ) 30 (10 fold detection limit) in at least one sample (see footnote, Table 1), omitting discussion of 58 compounds that were found in very low concentrations. On the basis of the relative responses of Br- and Cl- using ECNI-MS, ∼90% of the brominated compounds were more abundant in the sponges than lindane, the most abundant chlorinated compound (Figure 1). Although formation of Brmay be preferable to loss of Cl-, additional full scan analysis confirmed the dominance of organobromines in the samples. Lindane subsequently served as the reference standard for the calculation of relative retention indices (RRIs) used to verify the presence of brominated compounds across samples. Approximately 50% of the brominated compounds eluted before lindane (Table 1). The highest RRI values were in the range of BDE 47 (absolute tR ) 36.6 min) and BDE 99 (absolute tR ) 41.1 min), components of anthropogenic flame retardants that were, as in the case of PCBs, not detected in any sample. The peak areas associated with m/z 79 ion using ECNIMS varied ∼4 orders of magnitude across samples. The highest concentrations were estimated at several ppm on the basis of air-dried sample weight. Since reference standards were not available and GC/ECNI-MS responses are known to vary, more precise details cannot be given. The two samples of P. glaberrima collected in near proximity to each other were similar in terms of total number and identity of brominated compounds. Sponge-7 and -10 contained 60 and 68 brominated compounds, respectively, 50 of which were common to both (Table 1). However, ECNIMS-SIM showed a remarkable difference in HNP pattern between these samples of the same species (Figure 2). Accordingly, the relative abundances in Table 1 should not be considered as absolute. Each sponge sample contained g35 brominated compounds of natural origin, and 14 brominated compounds were detected in all five species (n ) 6) but at very different concentrations (Table 1). No compound resembled Q1 (m/z 386 and 388) nor did any form [HBr2]- fragment ions which are typical of aromatic compounds such as most brominated phenoxyanisoles (15).
67 15 8 241 3 149 4 187 162 3584 1486 1389 1226 49 3821 7 70 15 41 4
166 13
889 11 7 78 16 34 162
a Eighty-eight out of 146 peaks were labeled and discussed in the manuscipt (on the basis of the prerequisite that one value was >10 in one sample). b Number with increasing retention time of the 146 brominated compounds detected c Retention time of lindane was 17.49 min. d Species: Kirk. ) Kirkpatrickia variolosa, Arte. ) Artemisina apollinis, Phor. ) Phorbas glaberrima, Hali. ) Halichondria sp., and Leuc. ) Leucetta antarctica. Numbers refer to the internal lab code (see Experimental). e Relative signal intensity, numbers represent the factor the signals exceeded a signal-to-noise ratio of three in the samples (3 S/N * number in table). f Tentative structure, Br,Cl in combination with A or P) type and number of halogen on backbone; P ) phenol, A ) anisole; TBP, TBA ) tribromophenol or -anisole; Br3-OEO ) tribromooctenone (see text), MHC-1 ) mixed halogenated compound 1 (see text). g Mass of the molecular ion (refers to the monoisotopic peak).
206 u 250 u 250 u 220 u 240 u 240 u 264 u 284 u 284 u 342 u 360 u 328 u 396 u BrCl-P 2,4-DBP 2,6-DBP BrCl-A BrCl2-P BrCl2-P Br2-A Br2Cl-P Br2Cl-P 2,4,6-TBA Br3-OEO 2,4,6-TBP MHC-1 22 4 11 31 110 165
20 20 21 3 50 205 281 151 467 104 281 2233 10 5 3 4 13 16 43
15 13 2 23 40 514 326 155 920 58 1323 33 8.18 9.30 9.70 9.98 10.53 10.61 11.31 12.12 12.23 13.81 14.05 14.29 30.49 4 7 10 12 13 14 16 21 22 28 29 30 70 1 2 3 4 5 6 7 8 9 10 11 12 13
7 13 16 18 23 24 29 36 37 45 46 47 109
mass of mol. iong tentative structuref Leuc. 12d Irele Phor. 7d Irele Phor. 10d Irele Arte. 5d Irele Hali. 3d Irele Kirk. 8d Irele ret. timec [min] brominated compound detectedb peaks labeleda no. no. in table
TABLE 1. Tentatively Identified Halogenated Natural Products in Sponges
FIGURE 2. Detection of brominated compounds using GC-ECNIMS-SIM (m/z 79) in Phorbas glaberrima (a) sponge-7 and (b) sponge10, as well as (c) Halichondria sp. Peak labeling according to Table 1. Peak #70 was identified as MHC-1. The diversity, commonality, and species-specific relative abundances of HNPs in these sponges do not help distinguish whether these compounds are produced internally as primary or secondary metabolites or are merely accumulated from the surrounding waters such as the anthropogenic POPs R-HCH, lindane, and p,p′-DDE (Figure 1). Mixed Halogenated Compounds. Mass spectra from only three of the dozens of halogenated compounds detected (#25, #27, #70) exhibited evidence of [Br2]- fragment ions, previously thought as an indicator of aliphatic/alicyclic compounds with at least two bromine substituents (15). Of these, only compound #70 formed [Cl]- fragment ions (m/z 35 and m/z 37) in Halichondria sp. On the basis of retention time and SIM analysis of m/z 79, 116, and 160, compound #70 was found to be MHC-1, a dibromotrichloro-monoterpene previously detected in different seal species and fish (3, 15) (Figure 3a,b). MHC-1 was detected in other sponge species (see Figure 3c and Table 1) but at much lower concentrations than in Halichondria sp, suggesting uptake/accumulation via the water phase. The relatively high concentration of MHC-1 in Halichondria sp. is unexpected since sponges are rarely known as producers or accumulators of mixedhalogenated compounds (17). It appears in this case, however, that MHC-1 is produced by the sponge itself or by a closely associated organism. What is clear is that at least one natural producer of MHC-1 is found in the waters of King George Island. Mixed halogenated phenols and anisoles are further discussed below. Tentative Assignment of a Tribromoketone. Extracts of Phorbas glaberrima (sponge-7) contained the highest concentrations of the majority of HNPs (Table 1, Figure 2). In VOL. 39, NO. 11, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3891
FIGURE 3. Confirmation of the mixed halogenated monoterpene (MHC-1) using GC-ECNI-MS-SIM in (a) standard, (b) Halichondria sp., (c) Phorbas glaberrima. P. glaberrima as well as in other species (Table 1), peak #29 was an intense peak that warranted further investigation. Although the GC/ECNI-MS spectrum of peak #29 (Figure 4a) showed highest mass at m/z 202 (suggesting only one Br), the second most abundant fragment at m/z 158 indicated at least two Br and a nonaromatic compound. GC/EI-MS revealed that m/z 281 contained at least one Br less than the molecular ion, ruling out m/z 304 as such and thus indicating a molecular weight of 360 u (Figure 4c). This was confirmed by GC/PCI-MS where the spectrum for peak #29 was dominated by the quasi molecular ion [M+H]+ at m/z 361 (Figure 4b). Successive elimination of Br- was indicated by fragment ions at m/z 281 (361-80 u) and 203 (361-158 u). The three fragment ions indicating tribromo isotope patterns (m/z 261, m/z 289, and m/z 304) in the GC/EI-MS spectrum of peak #29 are of particular interest (Figure 4c). The fragment ion m/z 261 (261-237 u (Br3) ) 24 u) corresponds with C2Br3. The second tribromo pattern at m/z 289 was 28 u from m/z 261 suggesting a C3Br3O+ (carbonyl) structure. R-Cleavage on either side of the carbonyl moiety resulting in the abundant fragment ions at m/z 261 and m/z 289 readily explains their presence. It also follows that the nonbrominated fragments resulting from the carbonyl cleavage, that is, (360-261 u) ) 99 u and (360-289 u) ) 71 u, would also be present, as was indeed observed (Figure 4c). The m/z 99 ion was also present using PCI-MS (Figure 4b). The most abundant “tribromo” fragment ion in the EIMS spectrum of peak #29 (Figure 4c) had even mass (M-56 u) at m/z 304; this was formed after movement of the γ-hydrogen to the carbonyl oxygen and C-C cleavage at the 3892
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 11, 2005
FIGURE 4. Mass spectra and structure of 1,1,2-tribromooct-1-en3-one, an important brominated compound tentatively identified in sponges. (a) GC/ECNI-MS (b) GC/PCI-MS, (c) GC/EI-MS, and (d) structure. R- and β-methylene group, a fragmentation typical of higher ketones (McLafferty rearrangement). Br elimination from the McLafferty ion at m/z 304 leads to m/z 225; elimination of two Br from m/z 289 leads to m/z 131 and elimination of Br from m/z 261 leads to m/z 182. Taken together, the fragmentation patterns described above indicate the structure of peak #29 to be 1,1,2-tribromooct-1-en-3-one (Figure 4d), previously described by Cueto et al. (20). These authors found the compound at high concentration (450 mg/kg dry weight) in an algae collected in the same geographic area as the sponges investigated in the present work. Thus, it may be possible that this compound originated from the alga and has been taken up via the water by the sponges. Brominated Phenols and Anisoles. Brominated phenols and anisoles are known primary products or secondary metabolites of marine origin. In this study, GC/MS confirmed the presence of 2,4,6-tribromoanisole (TBA, #28), a dibromoanisole (#16), and traces of a bromochloroanisole (#12) in extracts of P. glaberrima. TBA and other halogenated anisoles have been previously identified in Antarctic air and marine fish (3, 16, 21-22). 2,4-Dibromo- and 2,6-dibromoanisole are known biogenic compounds (23). 2,4,6-Tribromoanisole (#28) and the dibromoanisole congener (#16) were about 10-fold more concentrated in sponge-7 than in sponge-10 of the same species, collected on the same day in the same local environment. While 2,4,6tribromoanisole dominated in sponge-7, the broad peak (#30) in both samples which was more abundant in sponge-10
than TBA was identified by an authentic standard as 2,4,6tribromophenol (Figure 2a,b). Similarly, peak #7 and peak #10 were identified as 2,4- and 2,6-dibromophenol. While the mass spectrometric data for bromophenols were unequivocal, it is widely believed that bromophenols are not GC-amenable without derivatization. However, Ro¨hrig et al. (24) demonstrated that free 2,4,6-tribromophenol can be chromatographed as a sharp peak from 50-m Ultra 2-like columns. As with 2,4,6-TBA, both anthropogenic and natural sources are known for 2,4,6-tribromophenol. For instance, 2,4,6-TBP is used as a flame retardant in epoxy, polyurethane, plastics, paper, and textiles and is an important intermediate for the production of other commercial high-molecular weight flame retardants and fire extinguishing media (2527). Furthermore, 2,4,6-TBP is used as an impregnating agent by the wood industry (28). 2,4,6-TBP and dibromophenol isomers are also secondary metabolites found in marine macroalgae and are naturally occurring flavor compounds in seafood (29-31). P. glaberrima extracts contained additional halogenated phenols, including bromochloro (#4), dibromo (#7, #10), bromodichloro (#13, #14), and dibromochloro (#21, #22) homologues (Table 1). 2,4- and 2,6-dibromophenol are the major congeners of natural origin found in nature (30). Like MHC-1, the presence of mixed halogenated phenols in our samples was surprising, since these compounds have been reported mainly in marine macroalgae. Because the concentrations of 2,4,6-tribromophenol (#28) and dibromochlorophenols #21 und #22 were extremely high, biosynthesis of these compounds by P. glaberrima cannot be ruled out. Our samples were collected during the Antarctic summer (January), that is, the season with the highest bioproduction. Other investigators have shown that appearance of halogenated phenols in seafood coincided with the seasonal growth circle of seaweed or the presence of different amounts of symbionts associated with the production of natural products (32, 33). That halophenols and haloanisoles (#4, #7, #10, #12-14, #16, #21-22, #28, and #30) were elevated in P. glaberrima but were detected in other sponge species as well complicates the assignment of HNP producers among sponges. As an example, high concentrations of 2,4,6-TBA and the 2,4,6TBP were also determined in Kirkpatrickia variolosa whereas high concentrations of dibromoanisole were detected in Halichondria sp. (Table 1). In contrast, 2,4,6-TBP in Halichondria sp. and Artemisina apollinis was roughly 100 fold lower than in P. glaberrima. Interestingly, the molecular ions in GC/ECNI-MS (for instrument and conditions see Experimental section) of the halogenated phenols were between 3 and 15 times more abundant than the bromide ion, which is atypical for brominated compounds (see also below). This is consistent with differences observed between GC/ECNI-MS full scan and SIM modes. For example, the 2,4,6-TBP/2,4,6-TBA ratio (#28/#30) in sponge-7 of Phorbas glaberrima was ∼3 in the SIM mode compared with ∼90 in the full scan mode (see below). Therefore, the bromide ion based relative amounts of bromophenols in Table 1 are likely underestimated. Ecological Relevance of HNPs in Antarctic Sponges. The detection of high concentrations of bromophenols in the “POP-fraction” of purified sponges raised the question whether these compounds are also occurring in higher organisms. Faulkner predicted that phenolic compounds are considered as the most stable group of natural organohalogens (8). However, phenolic organohalogens such as OHPCBs have not been detected in adipose tissue of higher organisms (34). This suggests that sponges and their symbionts are likely producers of halogenated phenols but, obviously, only the simple molecules (bromophenols) may reach adipose tissue of higher organisms. Interestingly,
FIGURE 5. GC/ECNI-MS investigation of sponge brominated natural compound in marine mammals. Note that different retention times were obtained owing to different GC methods than those used for analysis of the sponge samples. (a) Detection of 2,4,6-TBA and 2,4,6-TBP in a mussel sample from Mexico. (b) Bromine-selective m/z 79 extracted from the full scan chromatogram of the blubber extract of a hooded seal from the Arctic (2,4,6-TBP and MHC-1 are labeled). Different GC conditions were used for the analysis of the seal sample. (c) Mass spectrum of tribromophenol in the blubber of an Antarctic Weddell seal. recently Smeds et al. reported on pentabromophenol in human adipose tissue, however, no 2,4,6-TBP was found (35). By contrast, the majority of phenolic organohalogen compounds (e.g., OH-BDEs (36)) are not bioaccumulative. The latter compound class is frequently detected in serum or fish tissue and may be bound to proteins and thus retained in the organisms as well (34, 36). In addition, simple TBP and other bromophenols have also been detected in human blood (37). The large number of brominated compounds in sponges with short retention times, treated with the same cleanup procedure established for POPs, raises the question of biomagnification in higher organisms such as marine mammals. Tribromoanisole is frequently found in fish (3, 22). Both 2,4,6-TBA and 2,4,6-TBP have been identified in mussels from New Zealand and Mexico (Figure 5a). Only 2,4,6-TBA has been previously decribed in mussels from Mexico (38). To our knowledge, no data for tribromophenol isomers exist in marine mammals. Since the bromide ion in tribromophenols was very low in abundance using ECNI-MS, we screened previously extracted marine mammal tissues for TBP by extraction of m/z 330 in GC/ECNI-MS full scan analyses. We found 2,4,6-TBP in both blubber of Arctic hooded seal (Cystophora cristata) and Antarctic Weddell seals (LeptonyVOL. 39, NO. 11, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3893
chotes weddelli) (Figure 5b,c). In addition, we also identified a dibromophenol isomer in the hooded seal. Higher concentrations of bromophenols are reported to exhibit biological activity (39). Moreover, a dibromochlorophenol was recently shown to be active against methicillin-resistant Staphylococcus aureus and Burkholderia cepacia (12, 40). Although the concentrations of bromophenols were low in seal blubber, their detection underscores the global relevance of the compound. Both natural and anthropogenic sources are known for TBP (see above), but at least in the Antarctic Weddell seal, the natural source for TBP is much more likely. Under this premise, it is however unlikely that uptake of bromophenols and other HNPs arose from consumption of sponges by marine mammals. It is much more likely that HNPs are released from their producers into seawater becoming thereby available to the entire food web; likewise, it is known to happen with anthropogenic contaminants. This study has documented a large number of naturally produced organohalogens in the Antarctic sponge community. That some of these compounds were also detected in marine mammals suggest that higher biota have been exposed to these nonpolar halogenated compounds during their evolution.
Acknowledgments W.V. is very grateful to H.-U. Peters (Institute of Ecology, University of Jena, Germany), the staff at the Bellingshausen Research Station, and the students who participated in the Antarctic research excursion in 2000/2001. We also thank researchers and divers at the Chilean Antarctic Research Station on King George Island for sample collection and E. Stoll (Institute of Nutrition, University of Jena) for sample cleanup. Roland von der Recke, Katja Lehnert, and Daniel Olbrich at the Institute of Food Chemistry, University of Hohenheim, are acknowledged for confirmatory measurements and the students of food chemistry, Melanie Mongili and Markus Persike, are acknowledged for data analysis/ preparation. Thanks to Ron van Soest (ETI, The Netherlands) for valuable discussions on taxonomic questions. We are grateful to three anonymous reviewers for their constructive comments on a previous version of this manuscript.
Supporting Information Available Tables showing the most prominent halogenated natural products in sponges. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Tittlemier, S. A.; Simon, M.; Jarman, W. M.; Elliott, J. E.; Norstrom, R. J. Identification of a novel C10H6N2Br4Cl2 heterocyclic compound in seabird eggs. A bioaccumulating marine natural product? Environ. Sci. Technol. 1999, 33, 26-33. (2) Wu, J.; Vetter, W.; Gribble, G. W.; Schneekloth, J. S., Jr.; Blank, D. H.; Go¨rls, H. Structure and synthesis of the new natural heptachloro-1′-methyl-1,2′-bipyrrole Q1. Angew. Chem., Int. Ed. 2002, 41, 1740-1743. (3) Vetter, W.; Hiebl, J.; Oldham, N. Determination and mass spectrometric investigation of a new mixed halogenated persistent component (MHC-1) in fish and seal. Environ. Sci. Technol. 2001, 35, 4157-4162. (4) Vetter, W.; Stoll, E.; Garson, M. J.; Fahey, S. J.; Gaus, C.; Mu ¨ ller, J. F. Sponge halogenated natural products found at parts-permillion levels in marine mammals. Environ. Toxicol. Chem. 2002, 21, 2014-2019. (5) Asplund, L.; Athanasiadou, M.; Sjo¨din, A.; Bergman, Å.; Bo¨rjeson, H. Organohalogen substances in muscle, egg and blood from healthy Baltic salmon (Salmo salar) and Baltic salmon that produced offspring with the M47 syndrome. Ambio 1999, 28, 67-76. (6) Vetter, W.; Alder, L.; Kallenborn, R.; Schlabach, M. Determination of Q1, an unknown organochlorine contaminant, in human milk, Antarctic air and further environmental samples. Environ. Pollut. 2000, 110, 401-409. 3894
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 11, 2005
(7) Gribble, G. W. Naturally occurring organohalogen compounds. Acc. Chem. Res. 1998, 31, 141-152. (8) Faulkner, D. J. Natural organohalogen compounds. In The Handbook of Environmental Chemistry; Hutzinger, O., Ed.; Springer-Verlag: Berlin, Heidelberg, New York, 1980; Volume 1, part A, pp 229-254. (9) Naumann, K.; Chlorchemie der Natur. Chem. uns. Zeit 1993, 27, 33-41. (10) Flodin, C.; Whitfield, F. B. Biosynthesis of bromophenols in marine algae. Water Sci. Technol. 1999, 40, 53-58. (11) Gribble, G. W. The diversity of naturally produced organohalogens. Chemosphere 2003, 52, 289-287. (12) Faulkner, D. J. Marine natural Products. Nat. Prod. Rep. 2002, 19, 1-48. (13) Weichbrodt, M.; Vetter, W.; Scholz, E.; Luckas, B.; Reinhardt, K. Determination of organochlorine levels in Antarctic skua and penguin eggs by application of combined focused open-vessel microwave-assisted extraction, gel-permeation chromatography, adsorption chromatography, and GC/ECD. Int. J. Environ. Anal. Chem. 1999, 73, 309-328. (14) Vetter, W.; Weichbrodt, M.; Stoll, E. Indication of geographic variations of organohalogen concentrations in the Blubber of Antarctic Weddell Seals (Leptonychotes Weddelli). Environ. Sci. Technol. 2003, 37, 840-844. (15) Vetter, W. A GC/ECNI-MS method for the identification of lipophilic anthropogenic and natural brominated compounds in marine samples. Anal. Chem. 2001, 73, 4951-4957. (16) Vetter, W.; Schlabach, M.; Kallenborn, R. Evidence for the presence of natural halogenated hydrocarbons in southern norwegian and polar air. Fresenius Environ. Bull. 2002, 11, 170-175. (17) Capon, R.; Ghisalberti, E. L.; Jefferies, P. R.; Skelton, B. W.; White, A. H. Structural studies of halogenated diphenyl ethers from marine sponge. J. Chem. Soc., Perkin Trans. 1 1981, 2464-2467. (18) D’Ambrosio, M.; Guerriero, A.; Pietra, F. Novel, racemic or nearlyracemic antibacterial bromoquinols and chloroquinols and gamma-lactams of the verongiaquinol and the cavernicolin type from the marine sponge Aplysina ()verongia) cavernicola. Helv. Chim. Acta 1984, 67, 1484-1492. (19) Clark, W. D.; Corbett, T.; Valeriote, F.; Crews, P. Cyclocinamide A. An unusual cytotoxic halogenated hexapeptide from the marine sponge Psammocinia. J. Am. Chem. Soc. 1997, 119, 92859286. (20) Cueto, M.; Darias, J.; San-Martı´n, A.; Rovirosa, J. New Acetyl Derivatives from Antarctic Delisea fimbriata. J. Nat. Prod. 1997, 60, 279-281. (21) Fu ¨ hrer, U.; Deiβler, A.; Ballschmiter, K. Determination of biogenic halogenated methyl-phenyl-ethers (halogenated anisols) in the picogram/m3 range in air. Fresenius J. Anal. Chem. 1996, 354, 333-343. (22) Watanabe, I.; Kashimoto, T.; Tatsukawa, R. Polybrominated anisoles in marine fish, shellfish and sediments in Japan. Arch. Environ. Contam. Toxicol. 1983, 12, 615-619. (23) Ballschmiter, K. Pattern and sources of naturally produced organohalogens in the marine environment: biogenic formation of organohalogens. Chemosphere 2003, 52, 313-324. (24) Ro¨hrig, L.; Pu ¨ ttmann, M.; Meisch, H.-U. Determination of persistent organochlorine compounds in blood by solid phase micro extraction and GC-ECD. Fresenius J. Anal. Chem. 1998, 361, 192-196. (25) Eriksson, J.; Rahm, S.; Green, N.; Bergman, Å.; Jakobsson, E. Photochemical transformations of tetrabromobisphenol A and related phenols in water. Chemosphere 2004, 54, 117-126. (26) Hakk, H.; Larsen, G.; Bowers, J. Metabolism, tissue disposition, and excretion of 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE) in male Sprague-Dawley rats. Chemosphere 2004, 54, 1367-1374. (27) Barontini, F.; Marsanich, K.; Petarca, L.; Cozzani, V.; The thermal degradation process of tetrabromobisphenol A. Ind. Eng. Chem. Res. 2004, 43, 1952-1961. (28) Mardones, C.; Palma, J.; Seplveda, C.; Berg, A.; von Baer D. Determination of tribromophenol and pentachlorophenol and its metabolite pentachloroanisole in Asparagus officinalis by gas chromatography/mass spectrometry. J. Sep. Sci. 2003, 26, 923-926. (29) Whitfield, F. B.; Helidoniotis, F.; Shaw, K. J.; Svoronos, D. Distribution of bromophenols in species of marine algae from Eastern Australia. J. Agric. Food Chem. 1998, 47, 2367-2373. (30) Whitfield, F. B.; Helidoniotis, F.; Shaw, K. J.; Svoronos, D. Distribution of bromophenols in species of ocean fish from Eastern Australia. J. Agric. Food Chem. 1998, 46, 3750-3757.
(31) Whitfield, F. B.; Drew, M.; Helidoniotis, F.; Svoronos, D. Distribution of bromophenols in species of marine polychaetes and bryozoans from Eastern Australia and the role of such animals in the flavor of edible ocean fish and prawns (shrimp). J. Agric. Food Chem. 1999, 47, 4756-4762. (32) Chung, H. Y.; Joyce Ma, W. C.; Kim, J. S. Seasonal distribution of bromophenols in selected Hong Kong seafood. J. Agric. Food Chem. 2003, 51, 6752-6760. (33) Unson, M. D.; Holland, N. D.; Faulkner, D. J. A brominated secondary metabolite synthesized by the cyanobacterial symbiont of a matine sponge and accumulation of the metabolite in the sponge tissue. Mar. Biol. 1994, 119, 1-11. (34) Letcher, R. J.; Klasson-Wehler, E.; Bergman, Å. Methyl sulfone and hydroxylated metabolites of polychlorinated biphenyls. In The Handbook of Environmental Chemistry; Paasivirta, J., Ed.; Springer: Berlin, 2000; Vol. 3, Part K, pp 315-359. (35) Smeds, A.; Saukko, P. Brominated flame retardants and phenolic endocrine disrupters in Finnish human adipose tissue. Chemosphere 2003, 53, 1123-1130. (36) Marsh, G.; Athanasiadou, M.; Bergman, Å.; Asplund, L. Identification of hydroxylated and methoxylated polybrominated diphenyl ethers in Baltic Sea salmon (Salmo salar). Environ. Sci. Technol. 2004, 54, 10-18.
(37) Hovander, L.; Malmberg, T.; Athanasiadou, M.; Athanassiadis, I.; Rahm, S.; Bergman, Å.; Klasson Wehler, E. Identification of hydroxylated PCB metabolites and other phenolic halogenated pollutants in human blood plasma. Arch. Environ. Contam. Toxicol. 2002, 42, 105-117. (38) Vetter, W.; Wu, J. Nonpolar halogenated natural products bioaccumulated in marine samples. II. Brominated and mixed halogenated compounds. Chemosphere 2003, 52, 423-431. (39) Olsen, C. M.; Meussen-Elholm, E. T.; Holme, J. A.; Hongslo, J. K. Brominated phenols: characterization of estrogen-like activity in the human breast cancer cell-line MCF-7. Toxicol. Lett. 2002, 129, 55-63. (40) Jiang, Z.; Boyd, K. G.; Mearns-Spragg, A.; Adams, D. R.; Wright, P. C.; Burgess J. G. Diketopiperazines and One Halogenated Phenol from Cultures of the Marine Bacterium, Pseudoalteromonas luteoviolacea. Nat. Prod. Lett. 2000, 14, 435-440.
Received for review September 30, 2004. Revised manuscript received March 11, 2005. Accepted March 16, 2005. ES0484597
VOL. 39, NO. 11, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3895
