Occurrence and concentrations of halogenated natural products derived from seven years of passive water sampling (2007–2013) at Normanby Island, Great Barrier Reef, Australia

Marine Pollution Bulletin 137 (2018) 81–90

Contents lists available at ScienceDirect

Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul

Occurrence and concentrations of halogenated natural products derived from seven years of passive water sampling (2007–2013) at Normanby Island, Great Barrier Reef, Australia

T

Walter Vettera,b, , Sarit Kaserzonb, Christie Gallenb, Sarah Knolla, Michael Gallenb, Carolin Haulera, Jochen F. Muellerb ⁎

a b

University of Hohenheim, Institute of Food Chemistry, Garbenstrasse 28, D-70599 Stuttgart, Germany Queensland Alliance for Environmental Health Sciences (QAEHS), The University of Queensland, 20 Cornwall St. Woolloongabba, QLD 4102, Australia

ARTICLE INFO

ABSTRACT

Keywords: Halogenated natural products Great Barrier Reef Normanby Island Water concentrations Gas chromatography with high resolution mass spectrometry

Polydimethylsiloxane (PDMS) based passive water samplers deployed at Normanby Island, Great Barrier Reef (Australia) from 2007 to 2013 were analyzed for halogenated natural products (HNPs). Altogether, 38 samples, typically deployed for 30 days, were studied. Five HNPs (Q1, 2′‑MeO-BDE 68, BC-10, 2,4‑dibromoanisole and 2,4,6‑tribromoanisole) were detected in all samples. Most samples (> 90%) featured 2,2′‑diMeO-BB 80, 6‑MeOBDE 47, 2′,6‑diMeO-BDE 68 and 2,4‑dibromophenol. In addition, tetrabromo‑N‑methylpyrrole (TBMP) was detected in ~80% and Cl6-DBP in ~30% of the samples. Estimated time weighted maximum water concentrations were > 150 pg Q1 and 60 pg 2′‑MeO-BDE 68 per L seawater. Typically, the concentrations were varying from year to year. Moreover, time weighted average water concentration estimates did not reveal consistent maximum trend levels within a given year. Additional screening analysis via GC/MS indicated the presence of several polyhalogenated 1′‑methyl‑1,2′‑bipyrroles (PMBPs), 1,1′‑dimethyl‑2,2′‑bipyrroles (PDBPs), and 1‑methylpyrroles (PMPs) along with four brominated N‑methylindoles and several other polyhalogenated compounds at Normanby Island.

1. Introduction Halogenated natural products (HNPs) is a summarizing term for organohalogen compounds which are naturally produced by algae, sponges and other marine organisms. More than 5,000 structurally different HNPs have been discovered so far, mainly in marine environments (Gribble, 2010, 2012). Several of these HNPs have the potential to bioaccumulate in higher organisms, similarly to anthropogenic persistent organic pollutants (POPs) (Vetter, 2006). Occasionally, concentrations of such HNPs in marine organisms were comparable to or even higher than those of POPs (Vetter et al., 2001; Haraguchi et al., 2006; Stapleton et al., 2006; Alonso et al., 2014). The risks associated with exposure to HNPs are currently not elucidated, though studies suggest that some HNPs have toxic properties similar to polybrominated diphenyl ethers (PBDEs) and other POPs (Tittlemier et al., 2003; Wiseman et al., 2011). Likewise, sources and routes of entrance into the marine food web are sparsely known. For instance, HNPs have been repeatedly detected in marine mammals but it is unlikely that the mammals are directly consuming the natural producers

⁎

of HNPs (Vetter et al., 2002). Hence, release of HNPs into the water phase and “conventional” food chain enrichment similarly to POPs seems to be the most plausible route of exposure. However, both substance classes enter the environment in a different way. While POPs are mainly released into ocean water via atmospheric deposition and discharge of contaminated river water, HNPs are released into water wherever the natural producers are found. Hence, annual air concentration profiles are different from that of anthropogenic POPs (Melcher et al., 2008). However, the global and even local distribution of HNPs is currently poorly understood. First analyses of passive water samplers deployed at the Great Barrier Reef (Australia) showed that many of the relatively hydrophobic HNPs accumulate in semi-permeable membrane devices (SPMDs) (Vetter et al., 2009; Gaul et al., 2011). Deployed for a given period, passive water samplers not only accumulate (and concentrate) hydrophobic compounds; they also provide time weighted information on their concentration in the water phase during the deployment time of the samplers (Huckins et al., 1999; Allan et al., 2009). Yet, transfer of analyte concentration into concentrations in water requires an understanding of the sampling kinetics which is

Corresponding author at: University of Hohenheim, Institute of Food Chemistry, Garbenstrasse 28, D-70599 Stuttgart, Germany. E-mail address: [email protected] (W. Vetter).

https://doi.org/10.1016/j.marpolbul.2018.09.032 Received 10 July 2018; Received in revised form 7 September 2018; Accepted 18 September 2018 0025-326X/ © 2018 Elsevier Ltd. All rights reserved.

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

usually obtained in controlled calibration experiments assessing uptake and/or clearance of the chemicals (Rusina et al., 2010). Due to the lack of calibration data for HNPs with passive samplers at the time, however, the partly high mass accumulation of selected HNPs observed in samplers at 15 time points in the Great Barrier Reef, could not be translated into water concentration estimates (Vetter et al., 2009). Since then, such calibration of HNPs in passive water samplers based on polydimethylsiloxane (PDMS) and SPMD samplers (these samplers are among the most frequently used for organic contaminants with log Kow > 3) has been achieved (Kaserzon et al., 2014). In this study we aimed to investigate the temporal concentration profiles of several HNPs in archived PDMS samplers from a unique set of 38 samples deployed at Normanby Island on the Great Barrier Reef, over a period of seven years (2007–2013). The Great Barrier Reef in Australia stretches for about 2500 km along the northern coastline of eastern Australia and is home to a rich and diverse ecosystem and a known source of HNP production (Kennedy et al., 2012). Normanby Island was chosen since HNPs were previously identified in marine biota samples collected from around the Island which was recognized as a prime area for potential HNP biosynthesis (Vetter et al., 2009). Moreover, the sample number was sufficiently high for long term evaluation. Purified sample extracts were analyzed by gas chromatography coupled with high resolution electron ionization mass spectrometry (GC/EI-HRMS) in a targeted approach on several HNPs previously detected in samples from the Great Barrier Reef. Concentrations in the passive water samplers were converted into water concentrations and studied for time trends for the individual HNPs. In addition, qualitative data was collected for several homologs of major compound classes of HNPs.

Bremen, Germany). Helium (5.0 quality, BOC Gases, Sydney, Australia) was used as the carrier gas at a flow rate of 1.0 mL/min. The transfer line and ion source temperatures were set to 280 °C and 300 °C. Injections (1 μL) were conducted in splitless mode at an injection temperature of 250 °C. Separations were performed with a 30 m × 0.25 mm i.d., 0.25 μm film thickness DB-multi residue column (Zebron, ZB1-MS, Phenomenex, Torrance, Ca, USA). The GC oven temperature program started 1 min isothermal at 50 °C. Then, the oven was heated at 10 °C/min to 300 °C which was held for 5 min. Samples were analyzed at a resolution of R = 10,000, in selected ion monitoring (SIM) mode. For each compound two intense and high molecular mass fragments were selected for analysis. Generally, the most intense isotope peak served as a quantitation ion and the second most intense isotope peak as a verification ion. A compound was considered identified, if (i) the retention time did not deviate from the standard by > 0.02 s, (ii) both ions were detected and (iii) the ratio of both ions did not differ from the theoretical ratio by > 20%. Each time window included a lock mass and a calibration mass from perfluorotributylamine (PFTBA). SIM method 1: 5.0–12.3 min (2,4‑/2,6‑DBP, 2,4‑DBA): m/z 242.98508 (lock mass), m/z 249.86290, m/z 251.86090, m/z 263.87850, m/z 265.87650, m/z 268.98189 (calibration mass); 12.3–21.9 min (2,4,6‑TBP, 2,4,6‑TBA): m/z 230.98508 (lock mass), m/z 329.77140, m/z 331.76930, m/z 342.97869 (calibration mass), m/z 343.78700, m/z 345.78490; 21.9–25.0 min (TriBHD, 2′‑MeO-BDE 68/ 6-MeO-BDE 47, 2,2′‑diMeO-BB 80, 2′,6‑diMeO-BDE 68): m/z 454.97231 (lock mass), m/z 465.89650, m/z 467.89450, m/z 513.72380, m/z 515.72170, m/z 527.73940, m/z 529.73730, m/z 543.73430, m/z 545.73220, m/z 554.96592 (calibration mass); 25.0–31.0 min (TetraBHD): m/z 530.96592 (lock mass), m/z 543.80710, m/z 545.80500, m/z 554.96592 (calibration mass). SIM method 2 (α-PDHCH): 5.0–14.0 min: m/z 218.98508 (lock mass), m/z 221.94590, m/z 223.94300, m/z 230.98508 (calibration mass); 14.0–16.5 min (DPTE, TBMP): m/z 354.97869 (lock mass), m/z 369.80260, m/z 371.80060, m/z 394.69690, m/z 396.69400, m/z 404.97550 (calibration mass) 16.5–21.0 min (Q1, ATE/BATE, Cl6-DBP): m/z 304.98189 (lock mass), m/z 315.87090, m/z 317.86800, m/z 329.77140, m/z 331.76930, m/z 365.86330, m/z 367.86030, m/z 380.97550 (calibration mass); 21.0−31.0 min (BCIS, BC-10, Br5Cl-DBP, Br6-DBP): m/z 492.96911 (lock mass), m/z 513.72380, m/z 515.72170, m/z 543.65820, m/z 545.65530, m/z 587.60770, m/z 589.60480, m/z 631.55900, m/z 633.55700, m/z 642.95953 (calibration mass). Using these conditions 2′‑MeO-BDE 68 and 2,2′‑diMeO-BB 80 were coeluting but they were resolved by monitoring different SIM masses (2′‑MeOBDE 68: m/z 513.72374/515.72169; 2,2′‑diMeO-BB 80: m/z 527.73939/529.73734).

2. Materials and methods 2.1. HNP standards 2,4‑Dibromophenol (2,4‑DBP), 2,4,6‑tribromoanisole (2,4,6‑TBA) and 2,4,6‑tribromophenol (2,4,6‑TBP) were obtained from Sigma Aldrich (Steinheim/Taufkirchen, Germany), 2,6‑dibromophenol (2,6‑DBP) was from Lancaster Synthesis (Frankfurt, Germany), and 2,4‑dibromoanisole (2,4‑DBA) was from Alfa Aesar (Karlsruhe, Germany). 2,3,4,5‑Tetrabromomethylpyrrole (TBMP) was synthesized by Gaul et al. (2011). 2,3,3′,4,4′,5,5′‑heptachloro‑1′‑methyl‑1,2′‑bipyrrole (Q1) was synthesized according to Wu et al. (2002), 1,1′‑dimethyl‑3,3′,4,4′‑5,5′‑hexachloro‑2,2′‑bipyrrole (Cl6-DBP) was synthesized by Martin et al. (2011), 2,2′‑dimethoxy‑3,3′,5,5′‑tetrabromobiphenyl (2,2′‑diMeO-BB 80 or BC1) and 3,5‑dibromo‑2‑(3′,5′‑dibromo‑2′‑methoxy)phenoxyanisole, (2′,6‑diMeO-BDE 68 or BC-11) were synthesized according to Marsh et al. (2005), 4,6‑dibromo‑2‑(2′,4′‑dibromo)phenoxyanisole (2′‑MeOBDE 68 or BC-2) was synthesized by Vetter and Wu (2003), 3,5‑dibromo‑2‑(2′,4′‑dibromo)phenoxyanisole (6‑MeO-BDE 47 or BC-3) was synthesized according to Marsh et al. (2003), 5,5′‑dichloro‑1, 1′‑dimethyl‑3,3′,4,4′‑tetrabromo‑2,2′‑bipyrrole (BC-10) was synthesized according to Gribble et al. (1999). 2,7‑Dibromo‑4a‑ bromomethyl‑1,1‑dimethyl‑2,3,4,4a,9,9a‑hexahydro‑1H‑xanthene (TriBHD) and 2,5,7‑tribromo‑4a‑bromomethyl‑1,1‑dimethyl‑2,3,4,4a,9, 9a‑hexahydro‑1H‑xanthene (TetraBHD) were isolated and identified by Garson et al. (1989) and a quantitative solution was prepared by Melcher et al. (2007). Perdeuterated α-HCH (α-PDHCH) was synthesized by Vetter and Luckas (1995) and 6′‑methoxy‑2,3′,4,4′‑tetrabromodiphenylether (BCIS) was synthesized by Vetter et al. (2011).

2.3. GC/ECNI-MS analysis Homolog patterns of three classes of polyhalogenated alkaloids (polyhalogenated 1′‑methyl‑1,2′‑bipyrroles – PMBPs, polyhalogenated 1,1′‑dimethyl‑2,2′‑bipyrroles – PDBPs, and polyhalogenated 1-methylpyrroles - PMPs) were studied by means of a 6890/5975 GC/MS system operated in the electron capture negative ion (ECNI) mode using the setup of Hauler et al. (2013). Heptahalogenated PMBPs exist in a theoretical variety of 78 congeners which differ in number and positions of chlorine and/or bromine substituents (Vetter, 2012). Likewise, hexahalogenated PDBPs and tetrahalogenated PMPs exist in a variety of 36 congeners, respectively (Vetter, 2012). Since the discovery of these substance classes in environmental samples (Tittlemier et al., 1999; Vetter et al., 2001; Gaul et al., 2011), different PMBPs, PDBPs and PMPs were detected in the environment (Tittlemier et al., 1999; Vetter et al., 2007; Pangallo et al., 2008; Hauler et al., 2013, Hauler and Vetter, 2017; Alonso et al., 2017) These HNPs were not available as standards, but previous protocols indicated the GC elution range (Hauler et al., 2013; Hauler and Vetter, 2017), and the two most abundant isotope

2.2. GC/EI-HRMS analysis Quantitation of HNPs was performed with a TRACE 1310 gas chromatograph interfaced to a DFS high resolution magnetic sector mass spectrometer equipped with a Tri Plus auto sampler (Thermo, 82

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

peaks of the molecular ion were calculated and measured by GC/ECNIMS in the selected ion monitoring (SIM) mode in three time windows: (i) 8–18 min: m/z 79, 81, 216.9, 218.9, 260.8, 262.8, 264.8, 306.8, 308.8, 350.7, 352.7, 392.7, 394.7, 396.7; (ii) 18–20.8 min: m/z 79, 81, 365.9, 367.9, 385.8, 387.8, 409.8, 411.8, 429.8, 431.8, 473.7, 475.7, 519.7, 521.7; (iii) 20.8–50 min: m/z 79, 81, 543.7, 545.7, 587.6, 589.6, 609.6, 611.6, 631.6, 633.6, 653.5, 655.5, 697.5, 699.5. In addition, selected sample extracts were analyzed in the full scan mode (m/z 50–750) for possible detection of further compounds.

governs the chemical uptake process [Ako / Vs (Cw − (Cs / Ksw))], where A = surface area of the sampler, ko = the chemical mass transfer coefficient, Vs = volume of the sampler, Cs = concentration of chemical accumulated in the sampler and Ksw = sampler water partition coefficient (Kaserzon et al., 2014; Huckins et al., 1999). For logistical reasons, the deployment time of the samplers was varied and samples were not available from all months (Table S1). Hence, individual data points were not equidistant. In graphics, data points were shown with their regular (inconstant) distance in days. The median day of the corresponding deployment period was chosen as basis for bars in graphics. For instance, the first sample was deployed from April 22–May 24 in 2007, the median day was determined (May 8) which was selected to represent the value of the water concentration in the form of a bar. Concentration trend curves were calculated by multiplying the area of the actual data point (DP) by three, adding the double value of the DP left and right, adding the values of the next DP in both direction, followed by dividing the result by 9 according to Eq. (1), because this weighted smoothing mode moderates extreme values:

2.4. Great Barrier Reef temporal study Samples were available from a routine multi-year sampling campaign which has been carried out with the aim of monitoring the health of the Great Barrier Reef ecosystem and the inputs of anthropogenic organic contaminants (Kennedy et al., 2010). From May 2007 to February 2013, PDMS replicate samplers (two per cage) along with passive flow monitors (PFMs) were deployed for 21–100 days off the coast of Normanby Island on the Great Barrier Reef (wet tropics region, GPS: −17.20476, 146.07434). All PDMS samplers (92 cm × 2.5 cm × 0.5 cm) were pre-cleaned with acetone (LiChrosolv, Merck, Australia) and n-hexane (for gas chromatography and residue analysis, Merck, Australia) before loading into stainless steel cages. Blank PDMS samplers (n = 6; 1 for every ~6 samples) were prepared alongside samples and treated the same way during clean-up, processing extraction and analysis. No HNPs were detected in any blank samples. Before extraction each PDMS strip was cleaned by scrubbing with water and dried. PDMS strips were extracted in 2 × 200 mL of n-hexane on a shaker at room temperature (21 °C) for two 24 h periods. The combined extracts from each sampler were then reduced to about 1 mL using rotary evaporators. Extracts were then passed through a column with about 2 g of sodium sulfate to remove moisture. The extracts were reduced in volume and filtered (0.45 μm PTFE) into 10 mL of dichloromethane (DCM) and subjected to clean up using size exclusion gel permeation chromatography (GPC). The collected fraction was reduced to a final volume of 100 μL (PDMS extracts) in DCM. PDMS extracts were archived (stored at −20 °C) until analysis.

([DP

2] + 2 [DP

1] + 3 [DP] + 2 [DP + 1] + [DP + 2])/9

(1)

The least number of samples were available in 2007 and 2011. In these cases the concentration trend curves could not be considered as highly valid, because they are strongly dependent on the adjacent values of branching years. Specifically, 2011 was interpreted with caution because only three data points were available. 2.7. Recovery rates of the internal standards For the sample preparation an added internal surrogate mix consisting of four components was used as recovery standard for quality control. For the determination of the recovery, two of the components, i.e. 4,4′‑dibromobiphenyl (log KOW 5.54) and triphenyl phosphate (log KOW 4.7) were used, because these two substances have log KOW values in the range of the HNPs. Since not all samples contained this internal standard mix, recovery values from other passive sampler samples had to be used. The recovery rate of both compounds was somewhat variable and partly low (4,4′‑dibromobiphenyl: 24–57%; triphenyl phosphate: 30–76%). Variations were very likely due to different reasons (different persons, different training level). In particular, loss of 4,4′‑dibromobiphenyl was attributed to its high volatility, and evaporation procedures were considered the single most likely cause for its loss. Noteworthy, however, is that several HNPs are less volatile than 4,4′‑dibromobiphenyl, and loss was predominantly expected for volatile HNPs in the samples (especally 2,4‑DBA and 2,4,6‑TBA). We did not recovery correct the results and hence concentrations of these chemicals may be an underestimation and thus trends were not evaluated in detail. Last but not least, the sample cleanup protocol was not validated for phenolic compounds and it is likely that an unknown and probably varying share was lost during sample preparation (the virtual absence of bromophenols in samples collected and purified before 2010 indicated that changes in the sample cleanup may have occurred). Therefore, data of bromophenols (2,4‑DBP, 2,6‑DBP, 2,4,6‑TBP) is only indicative and will be only (briefly) mentioned in order to make readers aware of their potential relevance in the Great Barrier Reef.

2.5. Estimation of HNP concentrations in the water from the mass of HNP accumulated in the passive samplers In order to convert the accumulation of HNPs in passive samplers to concentrations of HNPs in the water we had to carry out an in-field calibration study using the PDMS samplers (Kaserzon et al., 2014). Using the field calibration, we could establish sampling rates describing the extraction of HNPs from the water as a function of time (L/day) for six HNPs where we could determine an uptake curve in the sampler and quantify the concentration in the water at the site reproducibly. This applied to chemicals including 2,6‑DBP, 2,4‑DBA, 2,4,6‑TBA, 2,4,6‑TBP, Q1 and 6‑MeO-BDE 47. Equilibration between some analytes and the PDMS was observed for chemicals with log Kow < 4 (i.e. bromophenols and -anisoles) and linear uptake in PDMS was observed for the chemicals with higher Kow (i.e. Q1 and 6‑MeO-BDE 47). Using the available calibration data we extrapolated sampling kinetics for other HNPs that were detected in the deployed PDMS samplers from the respective chemicals´ KOW (Kaserzon et al., 2014).

3. Results and discussion

2.6. Deployment and HNP concentrations in the samplers

3.1. General observations

Time-weighted average water concentrations for HNPs that were in linear accumulation in PDMS (i.e. HNPs with Kow > 5) were estimated from the linear approximation model [Cw = Ns/Rs t] where Cw = water concentration, Ns = amount of chemical accumulated in the sampler, Rs = the sampling rate of the respective HNP, and t = deployment time. Water concentrations for HNPs that had reached water-PDMS equilibrium were calculated based on the differential equation that

Five HNPs were detected in all 38 processed passive water samples, i.e. Q1, 2′‑MeO-BDE 68, 5,5′‑Cl2‑3,3′,4,4′‑Br4‑DBP (BC-10), 2,4‑DBA and 2,4,6‑TBA (Table 1). High detection frequencies (> 90%) were also observed for 2,2′‑diMeO-BB 80, 6‑MeO-BDE 47, 2′,6‑diMeO-BDE 68 and 2,4‑DBP (Table 1). In addition, TBMP was detected in ~80% and Cl6DBP in ~30% of the samples (Table 1). TriBHD and TetraBHD were not detected in any samples although they were initially identified by 83

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

Table 1 Time-weigthed minimum, maximum, and average water concentrations (pg/L) of different halogenated natural products derived from passive water samplers deployed at Normanby Island, Great Barrier Reef (2007–2013). Compound

Minimum

Maximum

Mean

Detection frequency [%] (n = 38)

2′‑MeO-BDE 68 2,2′‑diMeO-BB 80 6‑MeO-BDE 47 2′,6‑diMeO-BDE 68 Q1 Cl6‑DBP BC-10 2,4‑DBP 2,6‑DBP 2,4‑DBA 2,4,6‑TBA 2,4,6‑TBP TBMP

4.4 n.d. n.d. n.d. 9 n.d. 2,8 23 n.d. 21 6 n.d. n.d.

60 22 110 42 230 12 69 28,900 2370 1370 3280 320 2460

15 4.0 30 10 66 0.7 19 3840 75 450 170 21 220

100 98 98 98 100 45 100 100 30 100 100 50 90

a)

200

Q1 150 100 50

2,6‘-diMeO-BDE 68

14

18

TBMP

Irel

Jan-13

Sep-12

c)

10

CI6-DBP

8

10

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

Jan-10

May-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

BC-10

pg/L

2,4,6-TBP

6

Jan-13

b)

May-07

a)

Sep-12

Jan-12

pg/L 90 80 70 60 50 40 30 20 10 0

May-12

Sep-11

Jan-11

May-11

Sep-10

Jan-10

May-10

Sep-09

Jan-09

May-09

Sep-08

Jan-08

May-08

Sep-07

May-07

0

2,4-DBP 2,2‘-diMeO-BB 80/2‘-MeO-BDE 68 2,4-DBA 2,4,6-TBA 6-MeO-BDE 47 (IS)

Irel

pg/L

22

26

[min]

6

b)

Q1

4 2

BC-10

Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

(IS) Cl6-DBP

Sep-07

May-07

0

pg/L

6

10

14

18

22

26

[min]

d)

2000

TBMP

Fig. 1. GC/EI-HRMS chromatograms of HNPs detected in the sample from January 2013 with total ion current of (a) SIM method 1 and (b) SIM method 2.

1500

1000

Garson et al. (1989) in Wollongong (New South Wales, Australia), which is, however, ~2000 km south of Normanby Island. GC/EI-HRMS allowed for a good detection of the targeted analytes (Fig. 1a,b).

500

Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

3.2. Concentrations of heptachloro‑1′‑methyl‑1,2′‑bipyrrole (Q1) and presence of further polyhalogenated 1′‑methyl‑1,2′‑bipyrroles (PMBPs)

Sep-07

May-07

0

Fig. 2. Time weighted water concentration of (a) Q1, (b) BC-10, (c) Cl6-DBP and (d) TBMP derived from 38 passive water samplers.

Q1 was detected in all samples and individual values varied by about one order of magnitude (Table 1). The three top levels (> 100 pg/L) were determined in May (2007), November (2010) and January 2013 (Fig. 2a). When values were available, concentrations were higher in May than in February (2008–2010, 2012). However, no consistent trend with regard to year or season could be observed. Based on the concentration trend curve, Q1 concentrations were low in 2009, medium in 2008 and 2012, and high in 2010 (Fig. 2a). From January 2010 on, with two exceptions, the concentrations of Ql were > 50 pg/L (mean concentration in all samples: pg/L, Table 1). Vetter et al. (2009) determined Q1 for the first time in passive samplers of sea water from

the Great Barrier Reef with highest concentrations in the Outer Whitsundays, High Island and Normanby Island (average levels 40 ng/ sampler). Previous estimates without calibration indicated a time weighted maximum concentration of 92 pg/L in the Great Barrier Reef (Vetter et al., 2009). This previous estimate is only slightly lower than the data of the present study after calibration. The natural producer of Q1 is still unknown but it was assumed to be a marine bacterium (Vetter, 2006).

84

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

BrCl6-MBPs Irel

Irel

dolphin

BrCl6-MBPs

3.3. Concentrations of 5,5′‑Cl2‑3,3′,4,4′‑Br4‑DBP (BC-10) and presence of further polyhalogenated 1,1′‑dimethyl‑2,2′‑bipyrroles (PDBPs)

a)

BC-10 was detected in all 38 samples with a mean concentration of 19 pg/L (Table 1, Fig. 2b). Between 2007 and 2008, the concentration trend curves of BC-10 and Q1 were similar but BC-10 was lower in abundance by about a factor of 2 (Fig. 2b). Top concentrations of > 60 pg/L BC-10 were reached in May 2007, November 2008 and January 2013. In the last two samples of 2008 the estimates suggest that time weighted BC-10 concentrations exceeded for the only time those of Q1. Yet, in 2009 and throughout 2012, concentrations of BC-10 were low, but slightly increasing in early 2013. Within a given year, BC-10 concentrations were typically lowest from February to August and highest from November to January. As for Q1, the natural producer of BC-10 is still unknown (Vetter, 2012). BC-10 has been repeatedly determined in marine samples, not only from Australia (Tittlemier et al., 2002). Yet, BC-10 could not be determined in our previous study, because of its co-elution with 2,2′‑diMeO-BB 80 (and the use of GC/ECD) (Vetter et al., 2009). Hence, the present study based on GC/EI-HRMS denotes its first detection in water from the Great Barrier Reef (and most likely water samples worldwide). Actually, BC-10 was more concentrated than 2,2′‑diMeOBB 80, with the time weighted maximum level of BC-10 being four-fold higher than that of 2,2′‑diMeO-BB 80 (94 vs. 23 pg/L, see below). Initially, BC-10 and four further PDBPs were detected by Tittlemier et al. (1999, 2002) in various samples. Only recently, Hauler et al. (2013) described a range of novel PDBP homologs in marine biota (Holothuria sp.) and marine mammals (Sousa chinensis) from Australia. Noteworthy, however, is that BC-10 (5,5′‑Cl2‑3,3′,4,4′,‑Br4-DBP) was the only Br4Cl2-DBP isomer detected which is in contrast to mussels from the North Sea and other regions (Hauler et al., 2013). In addition, Cl6-DBP was detected in a few samples from 2007, 2008 and 2009 (Fig. 2c). Yet, from November 2012 on, Cl6-DBP concentrations were highest with up to 12 pg/L in January 2013 (Table 1, Fig. 2c). Furthermore, GC/ECNI-MS-SIM measurements enabled the detection of two Br5Cl-DBPs and Br6-DBP in one passive water sample extract from Normanby Island (Fig. 3b).

Br2Cl5-MBPs 17

18

19

20 [min]

Br2Cl5-MBPs Br3Cl4-MBPs 19.4

20.2

Irel

21.0 [min]

Br6-DBP

b)

Br5Cl-DBP

23.4

23.8

24.2

Irel

TBMP

[min]

c)

2,3,4-Br3-5-Cl-MP 2,3,5-Br34-Cl-MP

Br2Cl2-MP 13.2

14.0

14.8

3.4. Concentrations of tetrabromo 1‑methylpyrrole (TMBP) and presence of further polyhalogenated 1‑methylpyrroles (PMPs)

[min]

TBMP (Fig. 2d) was detected in all but four samples (May 07, March and May 2010, July 2011). Highest concentrations were measured from May 2012 on (top level ~2400 pg/L in January 2013, Table 1). Before mid-2012, the concentration trend line was low but still, several samples contained > 100 pg/L TBMP (time weighted maximum concentration ~400 pg/L in August 2007), which is more than the time weighted maximum concentrations of Q1 and BC-10 (see above). Throughout the study, the concentration range differed by two orders of magnitude, which is uncommon for HNPs (and also for anthropogenic POPs). Hence, a drastic change in the natural production of TBMP and/ or release of TBMP from the producing organism into the water phase must have occurred near the point of deployment of the passive water sampler. Transferred to HNPs this means that not only the annual cycle has to be taken into account but also extreme year-to-year varations may occur, dependent on environmental conditions/climate. No clear monthly/seasonal dependence of the annual progression of TBMP could be observed. For instance, in 2012, TBMP was highest in November/ December (2400 pg/L) and lowest in July (49 pg/L) while TBMP concentrations in 2009 were highest in July (256 pg/L) and low in November/December (5 and 30 pg/L). The high abundance in water in comparison with PMBPs and PDBPs was not matched in marine mammals. For instance, TBMP was not detected so far in dolphins from the Great Barrier Reef. This indicates that TBMP may not be persistent or not bioaccumulating in marine mammals. TBMP clearly dominated the PMP pattern but ~50% of the samples

Fig. 3. GC/ECNI-MS pattern of (a) BrCl6-, Br2Cl5-, and Br3Cl4-MBPs with an insert of BrCl6- and Br2Cl5-MBPs in a dolphin sample, (b) PDBPs and (c) PMPs in passive water sample extracts from Normanby Island.

Next to Q1, the samples featured also (at least) three much lower abundant BrCl6-MBPs, three even less abundant Br2Cl5-MBPs, and two least abundant Br3Cl4-MBPs in the typical ratio of 100 (Q1):13 (BrCl6MBPs):2.6 (Br2Cl5-MBPs):1 (Br3Cl4-MBPs) (Fig. 3a). Mixed brominated/ chlorinated PMBPs were not monitored in our previous study with passive samplers (Vetter et al., 2009), but they had already been detected in cetaceans from the Great Barrier Reef (Vetter et al., 2007). Hence, their occurrence in the samples was not surprising. In addition, the number of isomers (according to GC/ECNI-MS) was similar to the number determined in marine mammals from Australia, considering that different GC columns were used (Vetter et al., 2007). Also the intensity ratio in the mammals was similar as the ratio described above for the passive samplers: Q1:BrCl6-MBPs:Br2Cl5-MBPs:Br3Cl4MBPs = 100:20:2.5:0.5 in marine mammals (one example chosen). Given the two different sample matrices (passive sampler vs. marine mammal blubber) and sample origins, the observed small differences indicated that accumulation of isomers in passive samplers and marine mammals appears to be similar.

85

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

pg/L

3.5. 2′‑MeO-BDE 68, 6‑MeO-BDE 47, 2′,6‑diMeO-BDE 68, and 2,2′‑diMeO-BB 80

a)

50

2’-MeO-BDE 68

2′‑MeO-BDE 68 was detected in all samples from 2007 to 2013 (Table 1, Fig. 4a). Typically, 2′‑MeO-BDE 68 concentrations were lowest from February to September (~10 pg/L 2′‑MeO-BDE 68) and highest between November and January (Fig. 4a). An exception was observed in 2007 when the 2′‑MeO-BDE 68 concentration was high in May (~38 pg/L) and exceeded the level in November of the same year (Fig. 4a). Also, concentrations in May and August 2011 were higher than in December 2011/January 2012. In 2012, this was mainly due to the low concentration at the end of the year. Highest concentrations of 2′‑MeO-BDE 68 were measured in November 2008 and 2010 (Fig. 4a). Water concentrations of 2′‑MeO-BDE 68 in the Great Barrier Reef were not reported before, but concentrations in passive samplers from Normanby Island were lower than those from other locations of the Great Barrier Reef (mean 2′‑MeO-BDE 68 concentrations of 20 ng/sampler versus 12 ng/sampler this study) (Vetter et al., 2009). These comparisons indicate varying concentrations of 2′‑MeO-BDE 68 along the Great Barrier Reef with Normanby Island ranking below average. Recently Bidleman et al. reported 2′‑MeO-BDE 68 concentrations of up to 17 pg/ L in Baltic surface water samples from 2012 (Bidleman et al., 2016). In this study, two grab samples were partly taken on the same day and the 2′‑MeO-BDE 68 concentrations (combined share of dissolved and particulate fraction) were 17.7 pg/L on May 8 and 16.6 pg/L on May 9 (Bidleman et al., 2016). These concentrations were in the same order of magnitude as in our samples from Normanby Island. 6‑MeO-BDE 47 differed from its isomer 2′‑MeO-BDE 68 in that concentrations in May 2008 and 2010 were among the highest, together with November 2010 (Fig. 4b). Typically, concentrations of 6‑MeO-BDE 47 were about twice as high as those of 2′‑MeO-BDE 68, which was also found for the mean value (Table 1). This ratio is in agreement with our previous data from passive samplers deployed on the Great Barrier Reef (average 270 ng/sampler 6‑MeO-BDE 47 vs. 97 ng/sampler 2′‑MeOBDE 68) (Vetter et al., 2009). 6‑MeO-BDE 47 concentrations in 2008 and 2010 were higher than those in 2009, 2011 and 2012. 6‑MeO-BDE 47 and 2′‑MeO-BDE 68 are frequently detected in dolphins worldwide (Alonso et al., 2014), mostly being higher contaminated with 6-MeOBDE 47 than 2′‑MeO-BDE 68, except Australia. Here, the concentration ratio was reversed in passive water samples and mammals. Since structures and physico-chemical properties of 2′‑MeO-BDE 68 and 6‑MeO-BDE 47 (Kow 6.91 and 7.17, respectively (Yu et al., 2008)) are very similar, differences in the accumulation of both compounds in passive samplers compared to marine mammals is less likely. Hence, different distribution of both compounds along the Great Barrier Reef with higher abundance of 6‑MeO-BDE 47 in the habitat of the marine mammals investigated so far were the most plausible reason for this observation. 2′‑MeO-BDE 68 and 6‑MeO-BDE 47 are structurally related to PBDEs from which they differ only in the methoxy group. Yet, this structural feature apparently has a strong impact on the polarity with 2′‑MeO-BDE 68 and 6‑MeO-BDE 47 being more polar than tetrabromodiphenyl ethers (Vetter et al., 2011). 6‑MeO-BDE 47 was also measured in (air and) water samples from the Baltic Sea with up to 52 pg/L (Bidleman et al., 2016). This time weighted maximum concentration was ~50% of that reported in the present samples, which highlights the relevance of HNPs in the Great Barrier Reef. 2′,6‑diMeO-BDE 68 was discovered together with 2′‑MeO-BDE 68 in sponges (Cameron et al., 2000; Vetter et al., 2002). With few exceptions, the concentration trend curves of 2′,6‑diMeO-BDE 68 and 2′‑MeO-BDE 68 were similar (Fig. 4c). The concentration ratio of 2′‑MeO-BDE 68 to 2′,6‑diMeO-BDE 68 ranged from 0.9 to 5.1 (with 2′,6‑diMeO-BDE 68 being more abundant in one sample). 2′,6‑diMeOBDE 68 concentrations were highest in autumn (May) 2007 with > 40 pg/L, and the levels were lower after 2010 (Fig. 4c). Typically, the concentrations were higher in summer than in the Australian winter. Within the group of BCs, concentrations of 2,2′‑diMeO-BB 80 were

40 30 20 10

Jan-13

Sep-12

Jan-12

May-12

Sep-11

Jan-11

May-11

Sep-10

Jan-10

May-10

Sep-09

Jan-09

May-09

Sep-08

Jan-08

May-08

Sep-07

May-07

0

pg/L

b)

100

6-MeO-BDE 47 80 60 40 20

pg/L 40 35 30 25 20 15 10

Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

May-07

0

c) 2‘,6-diMeO-BDE 47

Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

May-07

5 0

pg/L

d)

20

2,2‘-diMeB-BB 80 15 10 5

Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

May-07

0

Fig. 4. Time weighted water concentration of (a) 2′‑MeO-BDE 68, (b) 6‑MeOBDE 47, (c) 2′,6‑diMeO-BDE 68 and (d) 2,2′‑diMeO-BB 80.

additionally contained lower amounts of two Br3Cl- and one Br2Cl2isomer (Fig. 3c). In all samples, the Br3Cl-PMP isomer eluting first from the GC/ECNI-MS system was more abundant than the second one (Fig. 3c). Recently, it was shown that 2,3,4‑Br3‑5‑Cl‑MP eluted first from DB-5-like columns (Hauler and Vetter, 2017). Hence, 2,3,4‑Br3‑5‑Cl‑MP was the more prominent isomer in the passive water samples. 2,3,4‑Br3‑5‑Cl‑MP was also found to dominate over 2,3,5‑Br3‑4‑Cl‑MP in mussels from different locations in Europe and one sample from Chile (Hauler and Vetter, 2017).

86

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

pg/L

up. This may explain relatively low detection frequencies in this study for 2,4,6‑TBP and 2,6-DBP (~20% and ~10% respectively in the samples). However, in the positive samples, the concentrations of 2,4‑DBP were particularly high in several samples in 2010 and 2012 (estimated Cw > 10,000 pg/L water) (Fig. 5). Due to the relatively low KOW of bromophenols observed in this study we previous estimated that samplers equilibrate relatively fast (Vetter et al., 2009). Hence, the estimated Cw is representative mainly for the latter part of sampler deployment where on occasions we estimated concentrations of 2,4‑DBP in the water 2 orders of magnitude higher than any other HNP measured in this study. Different to that, 2,4,6‑TBP was only found at high concentrations in 2007 and 2012 and at much lower concentrations in 2009 which is very different to the temporal profile of the DBPs (Fig. 5d). High abundance of bromophenols in water was not surprising since these chemicals have a higher water solubility and associated lower KOW compared to all other HNPs investigated. Bromophenols are produced by various algae and other marine organisms (Gribble, 2010). Flodin et al. already reported the annual and seasonal variation in the bromophenol concentration of the green marine alga Ulva lactuca which is known for its high bromophenol content (Flodin et al., 1999). Highest concentrations in alga were measured at the end of the summer and lowest in the remaining time of the year (Flodin et al., 1999). By contrast, the detection of bromoanisoles was not really expected because natural producers of bromoanisoles are not known. Bromoanisoles are thought to be microbial transformation products of bromophenols (Watanabe et al., 1985; Allard et al., 1987; Führer and Ballschmiter, 1998). Their detection in the passive samplers could be due to the presence of hitherto unknown producers or the (microbial) formation from bromophenols after their release into the water phase or even on the PDMS sampler. Yet, 2,4,6‑TBA (Fig. 5a) was already detected in the Great Barrier Reef (Vetter et al., 2009). Moreover, 2,4‑DBA and 2,4,6‑TBA concentrations of up to 200 pg/L were already reported in grab samples from the Atlantic Ocean (n = 4) and the Baltic Sea (n = 16) (Pfeifer and Ballschmiter, 2002; Bidleman et al., 2016). Though not verified due to sample preparation issues it was found from 2010 on, that the concentration trend curve of 2,4‑DBA was similar to the one of 2,4‑DBP but at ~10% of the DBP level (Fig. 5b,c). More data should be collected in this context. For instance, samplers may be deployed with isotopically labelled performance reference compounds (PRCs) in order to test for microbial methylation. Noteworthy, the time weighted maximum concentration of bromoanisoles at Normanby Island was about one order of magnitude higher than in water from the Atlantic Ocean and the Baltic Sea (Pfeifer and Ballschmiter, 2002; Bidleman et al., 2016).

a)

3000

2,4,6 TBA

2500 2000 1500 1000 500

Jan-13

Sep-12

Jan-12

May-12

Sep-11

Jan-11

May-11

Sep-10

Jan-10

May-10

Sep-09

Jan-09

May-09

Sep-08

Jan-08

May-08

Sep-07

May-07

0

pg/L

b)

2500

2,4 DBA 2000 1500 1000 500 Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

May-07

0

pg/L

c)

25000

2,4 DBP 20000 15000 10000 5000 Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

May-07

0

pg/L

d)

150

2,4,6,TBP 100

3.7. GC/ECNI-MS screening for further compounds

50

Selected samples were also analyzed by GC/ECNI-MS in full scan and low-mass SIM mode (Vetter, 2001). The low mass SIM mode indicated the presence of 50–100 polybrominated compounds due to the equally high abundance of the two Br− isotope peaks (Fig. S1). GC/ ECNI-MS chromatograms from different samples looked quite different (Fig. S1). The sample also featured several unknown polyhalogenated compounds forming m/z 114/116 (mixed aliphatic chlorinated-brominated compounds) and compounds with (substituted) diphenyl ether backbone ([HBr2]−, m/z 159/161/163). Full scan analysis confirmed the complexity indicated in the low mass SIM-run but none of the compounds indicated above could be identified. In most cases ions at higher mass were not present in the GC/ECNI-MS spectra or the fragmentation pattern could not be studied due to the bad quality of the mass spectra when analyte concentrations were low (see Fig. S2 for an example). However, these measurements verified the presence of two tribrominated and one tetrabrominated N‑methyl bromoindoles, and they produced evidence for one dibrominated N‑methylindole in the samples (Fig. 6). The GC/ECNI-MS spectra of the brominated N‑methylindoles featured the characteristic odd-mass molecular ion along

Jan-13

Sep-12

May-12

Jan-12

Sep-11

May-11

Jan-11

Sep-10

May-10

Jan-10

Sep-09

May-09

Jan-09

Sep-08

May-08

Jan-08

Sep-07

May-07

0

Fig. 5. Time weighted water concentration of (a) 2,4,6‑TBA, (b) 2,4-DBA, (c) 2,4‑DBP and (d) 2,4,6‑TBP.

lowest and below 10 pg/L except for March 2010 (11 pg/L) and April 2012 (23 pg/L). Concentrations in 2007, 2010 and 2012 were slightly higher than in 2008, 2009 and 2011 (Fig. 4d). Yet, no clear seasonal distribution was observed for 2,2′‑diMeO-BB 80. 2,2′‑diMeO-BB 80 was one of the most prominent HNPs in marine mammals from Northeastern Australia (Vetter et al., 2001). 3.6. Occurrence of bromophenols and concentrations of bromoanisoles As already mentioned, concentrations of bromophenols in our samples were equivocal due to potential loss during the sample clean 87

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

Irel

447

[M-Br+H]-

79

M-

367/9

Br4-MBI

X

a)

443

365

A

100 Irel

79

200

300

400 [m/z]

b)

X

Br3-MBI M367/9 365 100

200

300

X

Irel

M-

Br2-MBI

c)

289

287

79/81

60

[m/z]

160

260

[m/z]

Fig. 6. GC/ECNI-MS mass spectra of a (a) tetrabrominated, (b) tribrominated and (c) dibrominated N‑methylindole (MBI). “A” denotes the [M-CH3 + H]− fragment ion. The signal labelled “X” does not belong to the mass spectrum.

The screened samples also featured 2′‑OH-BDE 68 (Fig. S3). Brominated o‑hydroxylated diphenyl ethers (phenoxy phenols) are the precursors of brominated o‑methoxylated diphenyl ethers (phenoxy anisoles) with the latter being more lipophilic. Brominated phenoxyphenols were found to be widespread in water and biota from the Baltic Sea (Malmvärn et al., 2008; Dahlgren et al., 2015; Dahlberg et al., 2016; Choo et al., 2018). Previous attemps to detect this compound in passive water samples from the Great Barrier Reef failed although they were presumed to occur (Vetter et al., 2009). Similarly to bromophenols, these phenolic compounds could be partly dissociated at seawater pH (pKa values of ~7 were estimated for 2′‑OH-BDE 68 and 6‑OH-BDE

with the bromide isotope ions (m/z 79/81). The tetrabromo N‑methylindole additionally featured the [M-CH3 + H]− (labelled “A”) and [MBr + H]− fragment ions (Fig. 6a). Interestingly, one dibromo‑N‑methylindole, two tribromo‑N‑methylindoles and one tetrabromo‑N‑methylindole were reported in the marine environment (Hoh et al., 2009; Rosenfelder et al., 2010). Most likely these are the four compounds present in passive water samples from Normanby Island (Australia). Potential structures are 2,3,5‑ and 2,3,6‑tribromo‑ and 2,3,5,6‑tetrabromo‑N‑methylindole which were identified in algae and brittle star and red seaweed (Laurentia Brongniartii) (Tanaka et al., 1989; Liu and Gribble, 2002). 88

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al.

47, Bastos et al., 2009; Rayne and Forest, 2010), and passive sampling may lead to an underrating of the relevance. Furthermore, four isomers with the proposed M− at m/z 376 with dibromo pattern were detected whose structures could not be determined (Fig. S4).

from Rio de Janeiro, Brazil. Environ. Sci. Technol. 51, 1176–1185. Bastos, P.M., Eriksson, J., Bergman, Å., 2009. Photochemical decomposition of dissolved hydroxylated polybrominated diphenyl ethers under various aqueous conditions. Chemosphere 77, 791–797. Bidleman, T.F., Agosta, K., Andersson, A., Haglund, P., Liljelind, P., Hegmans, A., Jantunen, L.M., Nygren, O., Poole, J., Ripszam, M., Tysklind, M., 2016. Sea-air exchange of bromoanisoles and methoxylated bromodiphenyl ethers in the Northern Baltic. Mar. Pollut. Bull. 112, 58–64. Cameron, G.M., Stapleton, B.L., Simonsen, S.M., Brecknell, D.J., Garson, M.J., 2000. New sesquiterpene and brominated metabolites from the tropical marine sponge Dysidea sp. Tetrahedron 56, 5247–5252. Choo, G., Kim, D.-H., Kim, U.-J., Lee, I.-S., Oh, J.-E., 2018. PBDEs and their structural analogues in marine environments: fate and expected formation mechanisms compared with diverse environments. J. Hazard. Mater. 343, 116–124. Dahlberg, A.-K., Lindberg Chen, V., Larsson, K., Bergman, Å., Asplund, L., 2016. Hydroxylated and methoxylated polybrominated diphenyl ethers in long-tailed ducks (Clangula hyemalis) and their main food, Baltic blue mussels (Mytilus trossulus × Mytilus edulis). Chemosphere 144, 1475–1483. Dahlgren, E., Enhus, C., Lindqvist, D., Eklund, B., Asplund, L., 2015. Induced production of brominated aromatic compounds in the alga Ceramium tenuicorne. Environ. Sci. Pollut. Res. 22, 18107–18114. Flodin, C., Helidoniotis, F., Whitfield, F.B., 1999. Seasonal variation in bromophenol content and bromoperoxidase activity in Ulva lactuca. Phytochemistry 51, 135–138. Führer, U., Ballschmiter, K., 1998. Bromochloromethoxybenzenes in the marine troposphere of the Atlantic Ocean: a group of organohalogens with mixed biogenic and anthropogenic origin. Environ. Sci. Technol. 32, 2208–2215. Garson, M.J., Manker, D.C., Maxwell, K.E., Skelton, B.W., White, A.H., 1989. Novel bromo metabolites from a dictyoceratid sponge of the Cacospongia genus. Aust. J. Chem. 42, 611–622. Gaul, S., Bendig, P., Olbrich, D., Rosenfelder, N., Ruff, P., Gaus, C., Mueller, J.F., Vetter, W., 2011. Identification of the natural product 2,3,4,5‑tetrabromo‑1‑methylpyrrole in Pacific biota, passive samplers and seagrass from Queensland, Australia. Mar. Pollut. Bull. 62, 2463–2468. Gribble, G.W., 2010. Naturally occurring organohalogen compounds - a comprehensive update. Prog. Chem. Org. Nat. Prod. 91, 1–613. Gribble, G.W., 2012. Occurence of halogenated alkaloids. Alkaloids 71, 1–165. Gribble, G.W., Blank, D.H., Jasinski, J.P., 1999. Synthesis and identification of two halogenated bipyrroles present in seabird eggs. Chem. Commun. 1999, 2195–2196. Haraguchi, K., Hisamichi, Y., Endo, T., 2006. Bioaccumulation of naturally occurring mixed halogenated dimethylbipyrroles in whale and dolphin products on the Japanese market. Arch. Environ. Contam. Toxicol. 51, 135–141. Hauler, C., Vetter, W., 2017. Synthesis, structure elucidation and determination of polyhalogenated N‑methylpyrroles (PMPs) in blue mussels. Environ. Sci. Pollut. Res. 24, 26029–26039. Hauler, C., Martin, R., Knölker, H.-J., Gaus, C., Mueller, J.F., Vetter, W., 2013. Discovery and widespread occurrence of polyhalogenated 1,1′‑dimethyl‑2,2′‑bipyrroles (PDBPs) in marine biota. Environ. Pollut. 178, 329–335. Hoh, E., Lehotay, S.J., Mastovska, K., Ngo, H., Vetter, W., Pangallo, K., Reddy, C.M., 2009. Environ. Sci. Technol. 43, 3240–3247. Huckins, J.N., Petty, J.D., Orazio, C.E., Lebo, J.A., Clark, R.C., Gibson, V.L., Gala, W.R., Echols, K.R., 1999. Determination of uptake kinetics (sampling rates) by lipid-containing semipermeable membrane devices (SPMDs) for polycyclic aromatic hydrocarbons (PAHs) in water. Environ. Sci. Technol. 33, 3918–3923. Kaserzon, S., Gallen, C., Knoll, S., Gallen, M., Hauler, C., Vetter, W., Mueller, J., 2014. Levels of halogenated natural products on the great barrier reef, Australia from 20072013. Organohalogen Compd. 76, 918–921. Kennedy, K., Bentley, C., Paxman, C., Dunn, A., Heffernan, A., Kaserzon, S., Mueller, J., 2010. Final Report - Monitoring of Organic Chemicals in the Great Barrier Reef Marine Park Using Time Integrated Monitoring Tools (2009–2010). Coopers plains: The University of Queensland, The National Research Centre for Environmental Toxicology. Kennedy, K., Devlin, M., Bentley, C., Lee-Chue, K., Paxman, C., Carter, S., Lewis, S.E., Brodie, J., Guy, E., Vardy, S., Martin, K.C., Jones, A., Packett, R., Mueller, J.F., 2012. The influence of a season of extreme wet weather events on exposure of the World Heritage Area Great Barrier Reef to pesticides. Mar. Pollut. Bull. 64, 1495–1507. Liu, Y., Gribble, G.W., 2002. Syntheses of polybrominated indoles from the red alga Laurencia brongniartii and the brittle star Ophiocoma erinaceus. J. Nat. Prod. 65, 748–749. Malmvärn, A., Zebühr, Y., Kautsky, L., Bergman, Å., Asplund, L., 2008. Hydroxylated and methoxylated polybrominated diphenyl ethers and polybrominated dibenzo‑p‑dioxins in red alga and cyanobacteria living in the Baltic Sea. Chemosphere 72, 910–916. Marsh, G., Stenutz, R., Bergman, Å., 2003. Synthesis of hydroxylated and methoxylated polybrominated diphenyl ethers- natural products and potential polybrominated diphenyl ether metabolites. Eur. J. Org. Chem. 2003, 2566–2576. Marsh, G.A., Athanassiadis, M.I., Bergman, Å., Endo, T., Haraguchi, K., 2005. Identification, quantification, and synthesis of a novel dimethoxylated polybrominated biphenyl in marine mammals caught off the coast of Japan. Environ. Sci. Technol. 39, 8684–8690. Martin, R., Jäger, A., Knölker, H.-J., 2011. Transition metals in organic synthesis, part 97: silver-catalyzed synthesis of hexahalogenated 2,2′‑bipyrroles. Synlett 2795–2798. Melcher, J., Janussen, D., Garson, M., Hiebl, J., Vetter, W., 2007. Polybrominated hexahydroxanthene derivatives (PBHDs) and other halogenated natural products from the Mediterranean sponge Scalarispongia scalaris in marine biota. Arch. Environ. Contam. Toxicol. 52, 512–518. Melcher, J., Schlabach, M., Strand Andersen, M., Vetter, W., 2008. Contrasting the

3.8. Overall assessment None of the quantified HNPs showed distinct seasonal variations in the concentrations. No universal trends could be observed and the course of the concentrations appeared relatively random. Highest time weighted HNP concentrations of > 10,000 pg/L were determined for 2,4‑DBP in several samples (Fig. 5c). High time weighted maximum concentrations (> 1000 pg/L) were also occasionally determined for 2,4‑DBA and TBMP. Despite potential analytical issues, the detection of bromophenols at high concentrations in the positive samples illustrates their occurrence in the Great Barrier Reef, and their relevance needs to be studied more in detail in future. Aside from bromophenols and bromoanisoles, Q1 was the only HNP which was detected in several samples at > 100 pg/L. Comparison of the annual and seasonal progression resulted in individual patterns. However, no correlation was observed between HNP concentrations and rainfall during the dry (May–November) and wet (December–April) seasons. Although water concentration estimates suggest that aqueous levels of HNP are mostly in the subnanogram per litre levels and thus relatively low, their constant presence and persistence coupled with the high biomagnification potential of some compounds is worth noting. For example, high levels of Q1, 2,2′‑diMeO-BB 80, 2′‑MeO-BDE 68, and 6‑MeO-BDE 47 have previously been reported in dolphin blubber from Northeast Queensland, Australia (Vetter et al., 2001). This research focused on the identification of known HNPs, though additional yet undiscovered chemicals are likely present in the marine ecosystem. Events such as rising or fluctuating water temperatures, changing environmental conditions and nutrient inputs could result in changes to the biosynthesis of HNPs. Therefore, additional research that examines specific species and triggers for HNP production as well as the continued monitoring of these chemicals in marine and freshwater systems are warranted. Passive sampling techniques such as the PDMS are promising tools in this context. Acknowledgements This work was possible through scholarship and travel support from GO8-DAAD (German Academic Exchange Service). We thank the Water, Air Quality and Biomonitoring group at the Queensland Alliance for Environmental Health Sciences, University of Queensland and Heron Island Research Station (HIRS) staff for assistance with field deployments. We are also grateful to one anonymous reviewer of this manuscripts who's comments helped us to improve the quality of the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.marpolbul.2018.09.032. References Allan, I.J., Booij, K., Paschke, A., Vrana, B., Mills, G.A., Greenwood, R., 2009. Field performance of seven passive sampling devices for monitoring of hydrophobic substances. Environ. Sci. Technol. 43, 5383–5390. Allard, A.-S., Reemberger, M., Neilson, A.H., 1987. Bacterial O-methylation of halogensubstituted phenols. Appl. Environ. Microbiol. 53, 839–845. Alonso, M.B., Azevedo, A., Torres, J.P.M., Dorneles, P.R., Eljarrat, E., Barceló, D., LailsonBrito Jr., J., Malm, O., 2014. Anthropogenic (PBDE) and naturally-produced (MeOPBDE) brominated compounds in cetaceans - a review. Sci. Total Environ. 481, 619–634. Alonso, M.B., Maruya, K.A., Dodder, N.G., Lailson-Brito Jr., J., Azevedo, A., Santos-Neto, E., Torres, J.P.M., Malm, O., Hoh, E., 2017. Bottlenose dolphins (Tursiops truncatus)

89

Marine Pollution Bulletin 137 (2018) 81–90

W. Vetter et al. seasonal variability of halogenated natural products and anthropogenic hexachlorocyclohexanes in the Southern Norwegian atmosphere. Arch. Environ. Contam. Toxicol. 55, 547–557. Pangallo, K., Nelson, R.K., Teuten, E.L., Pedler, B.E., Reddy, C.M., 2008. Expanding the range of dehalogenated 1′‑methyl‑1,2′‑bipyrroles (MBPs) using GC/ECNI-MS and GC × GC/TOF-MS. Chemosphere 71, 1557–1565. Pfeifer, O., Ballschmiter, K., 2002. Halogenated methyl-phenyl ethers (HMPE; halogenated anisoles) in the marine troposphere and in the surface water of the Atlantic Ocean: an indicator of the global load of anthropogenic and biogenic halophenols. Organohalogen Compd. 57, 483–486. Rayne, S., Forest, K., 2010. pKa values of the monohydroxylated polychlorinated biphenyls (OH-PCBs), polybrominated biphenyls (OH-PBBs), polychlorinated diphenyl ethers (OH-PCDEs), and polybrominated diphenyl ethers (OH-PBDEs). J. Environ. Sci. Health A 45, 1322–1346. Rosenfelder, N., van Zee, N.J., Mueller, J.F., Gaus, C., Vetter, W., 2010. Gas chromatography/electron ionization-mass spectrometry-selected ion monitoring screening method for a thorough investigation of polyhalogenated compounds in passive sampler extracts with quadrupole systems. Anal. Chem. 82, 9835–9842. Rusina, T.P., Smedes, F., Koblizkova, M., Klanova, J., 2010. Calibration of silicone rubber passive samplers: experimental and modeled relations between sampling rate and compound properties. Environ. Sci. Technol. 44, 362–367. Stapleton, H.M., Dodder, N.G., Kucklick, J.R., Reddy, C.M., Schantz, M.M., Becker, P.R., Gulland, F., Porter, B.J., Wise, S.A., 2006. Determination of HBCD, PBDEs and MeOBDEs in California sea lions (Zalophus californianus) stranded between 1993 and 2003. Mar. Pollut. Bull. 52, 522–531. Tanaka, J., Higa, T., Bemardinelli, G., Jefford, C.W., 1989. Sulfur-containing polybromoindoles from the red alga Laurencia brongniartii. Tetrahedron 45, 7301–7310. Tittlemier, S.A., Simon, M., Jarman, W.M., Elliott, J.E., Norstrom, R.J., 1999. Identification of a novel C10H6N2Br4Cl2 heterocyclic compound in seabird eggs. A bioaccumulating marine natural product? Environ. Sci. Technol. 33, 26–33. Tittlemier, S., Borrell, A., Duffe, J., Duignan, P.J., Fair, P., Hall, A., Hoekstra, P., Kovacs, K.M., Krahn, M.M., Lebeuf, M., Lydersen, C., Muir, D., O'Hara, T., Olsson, M., Pranschke, J., Ross, P., Siebert, U., Stern, G., Tanabe, S., Norstrom, R., 2002. Global distribution of halogenated dimethyl bipyrroles in marine mammal blubber. Arch. Environ. Contam. Toxicol. 43, 244–255. Tittlemier, S.A., Kennedy, S.W., Hahn, M.E., Reddy, C.M., Norstrom, R.J., 2003. Naturally produced halogenated dimethyl bipyrroles bind to the aryl hydrocarbon receptor and induce cytochrome P4501A and porphyrin accumulation in chicken embryo hepatocytes. Environ. Toxicol. Chem. 22, 1622–1631.

Vetter, W., 2001. A GC/ECNI-MS method for the identification of lipophilic anthropogenic and natural brominated compounds in marine samples. Anal. Chem. 73, 4951–4957. Vetter, W., 2006. Marine halogenated natural products of environmental relevance. Rev. Environ. Contam. Toxicol. 188, 1–57. Vetter, W., 2012. Polyhalogenated alkaloids in environmental and food samples. Alkaloids 71, 211–276. Vetter, W., Wu, J., 2003. Non-polar halogenated natural products bioaccumulated in marine samples. II. Brominated and mixed halogenated compounds. Chemosphere 52, 423–431. Vetter, W., Luckas, B., 1995. Synthesis, isolation of perdeuterated α‑1,2,3,4,5,6‑hexachlorocyclohexane. J. High Resolut. Chromatogr. 18, 643–646. Vetter, W., Scholz, E., Gaus, C., Mueller, J.F., Haynes, D., 2001. Anthropogenic and natural organohalogen compounds in blubber of dolphins and dugongs (Dugong dugon) from northeastern Australia. Arch. Environ. Contam. Toxicol. 41, 221–231. Vetter, W., Stoll, E., Garson, M.J., Fahey, S.J., Gaus, C., Mueller, J.F., 2002. Sponge halogenated natural products found at parts-per-million levels in marine mammals. Environ. Toxicol. Chem. 21, 2014–2019. Vetter, W., Gaul, S., Olbrich, D., Gaus, C., 2007. Monobromo and higher brominated congeners of the marine halogenated natural product 2,3,3′,4,4′,5,5′‑heptachloro‑1′‑methyl‑1,2′‑bipyrrole (Q1). Chemosphere 66, 2011–2018. Vetter, W., Haase-Aschoff, P., Rosenfelder, N., Komarova, T., Mueller, J.F., 2009. Determination of halogenated natural products in passive samplers deployed along the Great Barrier Reef, Queensland/Australia. Environ. Sci. Technol. 43, 6131–6137. Vetter, W., Kirres, J., Bendig, P., 2011. Bromination of 2‑methoxydiphenyl ether to an average of tetrabrominated 2‑methoxydiphenyl ethers. Chemosphere 84, 1117–1124. Watanabe, I., Kashimoto, T., Tatsukawa, R., 1985. Brominated phenols and anisoles in river and marine sediments in Japan. Bull. Environ. Contam. Toxicol. 35, 272–278. Wiseman, S., Wan, Y., Chang, H., Zhang, X., Hecker, M., Jones, P.D., Giesy, J.P., 2011. Polybrominated diphenyl ethers and their hydroxylated/methoxylated analogs: environmental sources, metabolic relationships, and relative toxicities. Mar. Pollut. Bull. 63, 179–188. Wu, J., Vetter, W., Gribble, G.W., Schneekloth Jr., J.S., Blank, D.H., Görls, H., 2002. Structure and synthesis of the new natural heptachloro‑1′‑methyl‑1,2′‑bipyrrole Q1. Angew. Chem. Int. Ed. 41, 1740–1743. Yu, Y., Yang, W., Gao, Z., Lam, M.H.W., Liu, X., Wang, L., Yu, H., 2008. RP-HPLC measurement and quantitative structure–property relationship analysis of the n-octanol–water partitioning coefficients of selected metabolites of polybrominated diphenyl ethers. Environ. Chem. 5, 332–339.

90