Environ. Sci. Technol. 2007, 41, 5980-5985
Body Burdens of Persistent Halogenated Compounds during Different Development Stages of Anadromous Brown Trout (Salmo trutta) T O R E C . S V E N D S E N , * ,† KATRIN VORKAMP,‡ MARIE FREDERIKSEN,‡ BENT RØNSHOLDT,† AND JENS-OLE FRIER† Department of Biotechnology, Chemistry and Environmental Engineering, Sohngårdsholmsvej 57, Aalborg University, 9000 Aalborg, Denmark, and Department of Environmental Chemistry and Microbiology, National Environmental Research Institute, Frederiksborgvej 399, 4000 Roskilde, Denmark
Polychlorinated biphenyls (PCBs), DDTs, and polybrominated diphenyl ethers (PBDEs) were followed through the five life stages of a wild population of anadromous brown trout and related to variations in lipid content and exposure situations. Anadromous brown trout exhibits great variations in lipid content during its life cycle in the freshwater and marine environments. The results indicated substantial differences in PBDE and organochlorine exposure, with apparently more recent sources of PBDEs in the freshwater environment relative to the marine environment. Lipid and contaminant transfer were not always identical: The concentrations of PCBs, DDTs, and PBDEs (ng/g lipid weight) were about 15 times lower in the eggs compared to the muscle of their mother (e.g., 823 ng PCB/g lw vs 12 565 ng PCB/g lw, respectively). During the starving period from maiden to spawning trout the contaminant load increased by a higher factor than the lipid use. The data suggest a decoupling between lipid content and organohalogen concentrations for anadromous brown trout, which may contribute positively to reduce any potential negative effects of the transferred contaminants on eggs and fry.
Introduction During the past decades persistent halogenated compounds (PHCs) have been emitted to the environment and have accumulated in, e.g., lipid-rich tissues of marine organisms. Two compound groups regularly found in marine biota are organochlorines (PCBs and DDTs) and polybrominated diphenyl ethers (PBDEs). PCBs were used in a variety of products such as cooling and isolation fluids until they were banned in the mid 1970s, while DDT was used as an insecticide. PBDEs are a class of additive flame retardants that among others are used in paints, textiles, and electronics * Corresponding author phone: + 45 22117226; e-mail: ts@ bio.aau.dk. † Aalborg University. ‡ National Environmental Research Institute. 5980
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 17, 2007
(1). Both compound groups may have adverse effects on fish, as PDBEs have reduced spawning success of three-spined stickleback (Gasterosteus aculeatus) (2) and PCBs have caused early life stage mortality in rainbow trout (Oncorhynchus mykiss) (3). Many fish species exhibit the greatest toxicological sensitivity to contaminants during the early life stages (4). This indicates that knowledge about exposure during the early life stages is crucial. Owing to their semelparous life cycle, field studies on pacific salmon have suggested that the fish may leave considerable amounts of persistent pollutants in the rivers after spawning (5-7). Laboratory studies have also measured the accumulation of PHCs in salmonids, nevertheless only limited knowledge is currently available on transport of PHCs between the different life stages (3, 8). To investigate this, a population of anadromous brown trout from Lerkenfeldt Stream in Denmark has been studied. Hypotheses on how PHCs behave in the life cycle of anadromous brown trout were examined, in particular with regard to changes in lipid content. In agreement with previous studies, we assumed that the degradation or excretion of PHCs in the fish was slow and negligible in the development stages of the fish (9, 10). Furthermore, intercorrelation between the three compound groups has previously been shown (11). From egg to fry there is a net decrease in the lipid content of up to 50%, because the fry obtains nutrients and energy solely from the yolk sac (12). It is expected that this will lead to an increase in the concentration of PHCs on both wet weight (ww) and lipid weight (lw). As the egg lacks the enzyme system to metabolize PHCs (5) and thereby alter the composition, the PHCs pattern is expected to be similar between the two life stages. During the freshwater period the smolt feeds mostly on invertebrates (13). This results in a 100-200-fold weight gain over a period of approximately 2 years. Therefore the concentration in the smolt will not reflect the initial concentration in the fry, but the exposure to PHCs in the freshwater environment. Different PHC patterns are expected between fry and smolt, due to differences in PHC sources between the inherited marine pattern in the fry and freshwater pattern in the smolts (14). After smoltification, the trout migrates to the marine environment for 1-4 years before it returns to its natal stream for spawning (15). In this phase growth continues, leading to a 10-100 fold weight increase. A change in the PHC concentration, along with a change in the PHC pattern is expected due to the exposure to PHCs in the marine environment. During spawning migration and the subsequent resting period in the streams the trout stops feeding and loses 3050% of its total energy content (16). Part of the energy is invested in ovaries and roe which may comprise 15-20% of the fresh weight prior to spawning (12). The hypothesis is an increase in PHC concentration, but an unchanged PHC pattern since the period spent in the river is too short to permit substantial elimination or metabolic transformation of the PHCs (5, 9, 10). Only limited knowledge is available on how the eggs evolve inside the spawner. If there is free movement and equilibrium has occurred, equal concentrations of PHCs on a lipid basis are expected in the mother fish and the eggs. It is, however, anticipated that the concentration of PHCs in the eggs is lower than that in the mother fish (ng/g lw) since the production of eggs starts while the concentration in the 10.1021/es070746y CCC: $37.00
2007 American Chemical Society Published on Web 07/26/2007
FIGURE 1. Mean concentrations of ΣPDBEs, ΣPCB, and ΣDDT in ng/g wet weight and ng/g lipid weight in anadromous brown trout from Lerkenfeldt Stream. The error bars represent the standard deviation. mother is still low and ends before the mother is ready to spawn (17, 18).
Materials and Methods Sample Collection. The fish originated from Lerkenfeldt Stream, a medium-sized stream discharging into Limfjorden, Denmark. It does not run through any major cities and only three wastewater treatment plants with less than 1000 person equivalents and a fish farm discharge their water into the stream. PCBs, PBDEs, and DDTs were analyzed in trout from 5 different life stages. In July 2005, returning maiden trout were caught upon return to the stream. The trouts were killed and frozen (-20 °C) immediately after capture. Spawners were caught in November 2005 and kept at a fish farm until ready to spawn in the beginning of January 2006. The spawners were stripped, killed immediately after, and frozen. Following stripping, half of the eggs were frozen for analysis. Mixtures of pooled milt from five male individuals chosen at random were added to the other half of the eggs and the fertilized eggs were placed in hatching trays. The fry were kept until they had used most of their yolk sac, and then killed and frozen. Smolts were captured in the stream by electrofishing in April 2006. Length and weight of the individual fish (N ) 7) were as follows: smolt 14.7 ( 1.4 cm, 0.10 ( 0.05 kg; maiden: 44.5 ( 1.03 cm, 1.76 ( 0.16 kg; spawners (without eggs) 62.7 ( 9.5 cm, 2.37 ( 1.07 kg. For the chemical analysis, skinned fillet subsamples from the dorsal anterior part of the spawners and maiden trout were taken and whole egg and fry and the full skinned fillet of the smolt were used. Chemical Analysis. The lipid content of the muscle and liver samples was determined according to Smedes (19), and the dry matter content was calculated based on mass loss after drying at 105 °C until constant weight. The chemical analysis was described previously (20) and the protocol is briefly outlined here. The samples were homogenized, dried with diatomaceous earth, spiked with recovery standards
(CB-3, CB-40, CB-198, and BDE-77, all purity 98.9%, Promochem, Wesel, Germany) and Soxhlet extracted using 350 mL glass-distilled acetone and n-hexane (1:4) (Rathburn, Walkerburn, Scotland). The extracts were concentrated to approximately 1 mL using rotary evaporation and cleaned on a multilayer glass column consisting of anhydrous sodium sulfate, 5 g of silica impregnated with sulfuric acid (95-97% purity), 1 g of activated silica (24 h at 160 °C), and 5 g of aluminum oxide with 10% water. The column was eluted with 300 mL n-hexane and the eluates were concentrated by rotary evaporation and under nitrogen. Following addition of internal quantification standards (CB-53, CB-155, and BDE-71), the volume of the extracts was adjusted to 1 mL. Dual-column gas chromatography with electron capture detection (GC-ECD model HP 5890) was used for analysis of PCBs (CB-28, -32, -44, -49, -52, -101, -105, -110, -118, -128, -138, -149, -153, -156, -170, -180, -187, -194, -209) and DDTcompounds (o,p′-DDT, o,p′-DDE, p,p′-DDT, p,p′-DDE, p,p′DDD). The PBDEs (BDE-17, -28, -47, -49, -66, -85, -99, -100, -153, -154, and -183) were analyzed by GC-mass spectrometry in the electron capture negative ionisation (GC, Agilent 6890 series; MS, Agilent 5973). More detailed information is available from ref 20. The samples were processed in batches of 16 containing duplicate analysis of 1 or 2 samples, 2 blanks, and 2 sand eel oil (Ammodytes tobianus) samples used as internal reference material (Triple Nine Fish Protein, Esbjerg, Denmark), as described by ref 20. Compounds and congeners with concentrations deviating with more than 3× standard deviation from the mean concentration of the reference material were excluded from the data set. Only samples with >80% recovery for CB-40 and CB-198 were accepted. DLs were e0.25 ng/g ww for PCB congeners and DDT and ranged from 0.025 to 0.063 ng/g ww for PBDEs. BDE-183 had higher DLs of 0.25 and 0.125 ng/g ww. If the blanks showed concentrations >0.3 ng/mL the original detection limit was doubled. VOL. 41, NO. 17, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5981
TABLE 1. Ratios between the Mean Concentrations of the Individual Compound Groups (ΣPBDE, ΣPCB, and ΣDDT) on Both Wet Weight and Lipid Base in Successive Life Stages
PBDE wet weight significant PBDE lipid weight significant DDT wet weight significant DDT lipid weight significant PCB wet weight significant PCB lipid weight significant
spawner/ egg
egg/ fry
fry/ smolt
smolt/ maiden
maiden/ spawner
0.88 no 15.4 yes 0.86 no 15.1 yes 0.82 no 14.5 yes
2.7 no 0.77 no 0.86 no 0.25 no 0.95 no 0.27 yes
0.41 no 0.21 yes 7.4 yes 3.8 no 7.2 yes 3.7 yes
2.5 no 8.7 yes 0.67 no 2.3 no 0.35 no 1.2 no
2.5 no 21.8 yes 3.7 no 32.7 yes 2.0 no 17.4 yes
Data Analysis. A compound was removed from the data set if more than 20% of the samples from one life stage were below DL. This was the case for CB-209, BDE-153, and BDE183. If less than 20% of the individual samples had concentrations
