General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: [email protected]
SID 5 Research Project Final Report
SID 5 (2/05) Page 1 of 24
NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.
A SID 5A form must be completed where a project is paid on a monthly basis or against quarterly invoices. No SID 5A is required where payments are made at milestone points. When a SID 5A is required, no SID 5 form will be accepted without the accompanying SID 5A.
This form is in Word format and the boxes may be expanded or reduced, as appropriate.
ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.
Project identification
1. Defra Project code AE1142
2. Project title
Population impacts of endocrine disrupters on UK seals - effects on juvenile survival and interactions with disease
3. Contractororganisation(s)
Sea Mammal Research UnitUniversity of St AndrewsSt AndrewsFifeScotlandKY16 8LB
54. Total Defra project costs £ 53409
5. Project: start date................ 01 February 2004
end date................. 01 July 2005
SID 5 (2/05) Page 2 of 24
6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They
should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.
(b) If you have answered NO, please explain why the Final report should not be released into public domain
Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the
intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.
SID 5 (2/05) Page 3 of 24
Background and ObjectivesThis study comprised two distinct sections relating to the two different species of seal (the grey seal Halichoerus grypus and the harbour or common seal Phoca vitulina) that inhabit UK waters. The scientific questions addressed in this study were
(a) Does exposure to endocrine disrupters (particularly polybrominated diphenyl ethers (PDBEs, the main components of brominated flame retardant chemicals) and polychlorinated biphenyls (PCBs) affect the first year survival of grey seal pups? Previous research found a significant relationship between thyroid hormone levels in the blood and PBDEs in the blubber of these animals. This study specifically investigated if this relationship consequently affected the animals’ first year survival probability.
(b) Harbour seals inhabit coastal waters. Are there regional differences in blubber contaminant concentrations, particularly among compounds that are putative endocrine disrupters? This question is particularly pertinent following the two outbreaks of phocine distemper (PDV) that have had a major impact on the abundance of the harbour seal in Europe. The population in the Wash (southeast England) was particularly affected, suffering between 20 and 50% mortality in the two epidemics. By contrast in Scotland mortality rates were often below 10%. Studies funded by DEFRA (then DoE) in 1989 indicated a correlation between varying mortality rates in different UK harbour seal populations and contaminant levels and higher levels of contaminants in seals which died of PDV compared to those that survived. In addition to finding out the current levels of exposure to potential endocrine disrupters in harbour seals and whether levels were again higher in victims than survivors of PDV, we also investigated the relationship between blubber levels of PCBs, PBDEs and other OC pesticides and thyroid and steroid hormone levels in the blood of free-living animals from different populations with different contaminant exposure.
MethodsIn order to investigate whether blubber contaminants in post-weaned grey seal pups (transferred to them from their mothers in the milk during lactation) were important covariates of first year survival, we used a novel mark-recapture study. Post-weaned pups were fitted with mobile phone tags, which regularly automatically attempted to send text messages to a central phone number. When animals were at sea and out of mast coverage messages were not received but when animals returned to a haulout site in the coastal zone text messages stored in the tags’ buffer would be sent and received. This method of ‘resighting’ animals produced an encounter history for each of 60 pups from the Isle of May in the Firth of Forth and this data was used in a live-resighting, Cormack Jolly Seber model using the program MARK to estimate the first year survival of the pups. Individual covariates of survival (including mass at weaning, sex, immunoglobulin levels – all factors we know from previous work to be important in predicting survival – and in addition blubber contaminant levels and serum thyroid hormones) were embedded in the model in a fashion similar to logistic regression. Contaminants were measured in blubber biopsy samples obtained when animals were tagged using GC-MS. Hormones in blood samples obtained at the same time were measured using ELISA assays. A set of models was then fitted to the all the encounter history and covariate data and the best models (using Akaike’s Information selection criteria) were used to make inferences about the important predictors of first year survival.
Harbour seals were sampled from five populations around the UK coast; the Wash, southeast England; the Tay estuary, southeast Scotland; the Moray Firth, northeast Scotland; Orkney northern Scotland and Islay/Jura, south western Scotland. Populations were chosen following differential mortality during the 2002 phocine distemper epidemic. Sampling trips were carried out between March and October 2003 and blood and blubber samples were collected from 10 animals at each site except the Wash where 20 were sampled. Additional morphometric and mass and sex data were also obtained. Wherever possible adult animals only were used in the study to ensure comparability among sites. All procedures were carried out under Home Office Animal (Scientific Procedures) Act licences. Dead seals collected during the 2002 PDV epidemic were sampled by the Institute of Zoology, London and where cause of death was established as PDV blubber samples were collected. Only fresh carcasses in good condition were used in this study. An additional three samples were available from Scotland but the sample size was small because only very few animals had died of PDV. Blubber samples were analysed for contaminants as above and hormones were also assayed using ELISA methods.
Findings of the Research(a) Grey seal pup survival - Although we again found a significant relationship between blubber PCBs
and PBDEs and total serum thyroxine and triiodothyronine levels (which themselves are positively correlated), after taking account of condition and age effects, this did not appear to affect the survivorship of the pups to age one. The most important predictors of survival were condition (total fat) at weaning and sex (males have a lower survival probability than females). These results confirmed our earlier findings suggesting the study was robust. However, within the dataset there was no substantial evidence to support the hypothesis that blubber endocrine disrupters (transferred from the mother in the milk during lactation) significantly affects first year
SID 5 (2/05) Page 4 of 24
survival or that these effects are mediated through impacts on the levels of the thyroid hormones thyroxine and triiodothyronine.
(b) Harbour seals - We found distinct regional differences among contaminant levels in harbour seals from around the UK with seals hauling out on Islay and Jura (southwest Scotland) having the highest levels of PCBs and seals in the Wash having the highest levels of PBDEs. Animals from the north of Scotland (Moray Firth and Orkney) had the lowest levels of all contaminant groups studied. Significant relationships between blubber contaminants and total triiodothyronine concentrations were found in the harbour seals, after controlling for the effects of confounding factors. PBDEs and pesticides were the most significant predictors and in all cases the relationship was again a positive one, with higher levels of triiodothyronine being seen with increasing concentrations of contaminants. This interesting finding is in line with our previous studies on grey seal pups where hyperthyroid effects were related to blubber levels of PBDEs. This was not due to sampling pregnant females who have increased triiodothyronine levels in their blood, as the relationship was more significant when females and breeding season males were removed from the dataset. The increased levels in animals with higher contaminant exposure could be due to preferential secretion of triiodothyronine but information on levels of thyroid stimulating hormone (TSH) in the blood would be required to further understand this phenomenon. We currently do not have a method for measuring TSH in seals, as the reagents available for humans and laboratory species do not cross-react in seal serum.
In addition amongst the adult male harbour seals there was a significant negative relationship between blubber PBDEs levels and blubber hexachlorocyclohexane levels and circulating testosterone concentrations (after controlling for the effects of age and differences among regions). A similar negative relationship between organochlorine contaminants and testosterone levels has been reported for other species of marine mammal and indeed for polar bears but not for harbour seals, in which studies of the potential effects of exposure PBDEs are currently limited. Finally we investigated the difference between potential ED contaminant concentrations in harbour seals that died of PDV during the 2002 epidemic with concentrations in the survivors. This was restricted to seals from the Wash in southeast England as only 3 carcasses sampled in good condition were available from Scotland. We found significantly higher concentrations of PCBs (but not other contaminants) in the seals that died of PDV than the live animals. Again we controlled for other factors (such as sex, age and size) that might have been driving this finding. PCB levels in the dead animals were also above the estimated threshold for effects on the immune system of marine mammals, as were the levels in live animals from Islay/Jura where no exposure to PDV (animals did not have any antibodies) was seen in 2002.
Although the production and use of PCBs and many of the OC pesticides has been banned both in Europe and worldwide and the production and use of PBDEs is similarly now controlled, many compounds are still entering the marine environment through various waste streams. Comparing the concentrations of specific PCB congeners over the period between the PDV epidemics in 1988 and 2002 we found little evidence of a reduction in blubber levels in harbour seals. In fact geometric mean levels in the blubber of animals that died were an order of magnitude higher in 2002 than in the animals that died in 1988. Some of this difference is likely to be due to differences in the age structure of the two datasets (9/10 animals sampled in 1988 were <2 years old whereas all the 2002 animals were adults) and in the analytical methods used in the two studies, however it does indicate that these species are long-term integrators of the environment that are slow to respond to improvements in the quality of the marine ecosystem.
The brominated flame-retardants such as PBDEs may be EDs in harbour seals and concerns regarding the continued use of deca-bromodiphenyl ether that may degrade to lower brominated compounds remain. These findings are clearly also important to our predictions of the impact of another PDV outbreak among UK seals. Harbour seals in the Wash seem most vulnerable but perhaps of similar concern is the population on the southeast coast of Scotland as they have the highest levels of PCBs, again above the threshold for effects. PDV did not find its way into this population in 2002 but this may have been a chance event and future introductions of the virus may have a different pattern of spread.
In both grey and harbour seals relationships between blubber contaminants and serum hormones suggest these organohalogenated compounds are potential endocrine disrupters in these species. However, for post-weaned grey seal pups the relationship does not appear to affect their first year survival probability. Further research, particularly using in vitro approaches, would elucidate the mechanisms underlying the relationships reported here.
SID 5 (2/05) Page 5 of 24
Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with
details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).
1.0 Introduction
The effect of various persistent organic contaminants on immune function in marine mammals was the focus of much research in the 1990s (De Swart et al., 1996; Ross et al., 1996). However, more recently attention has turned to their effect on the endocrine system (Hall et al., 1998). The close relationship between the immune and endocrine systems suggests that endocrine disrupters (EDs) may have synergistic effects on immunity. Thus compounds identified as immunotoxic may also be EDs. Many studies on laboratory animals have shown polychlorinated biphenyls (PCBs) and polybrominated diphenyl ethers (PBDEs) to be thyrotoxic (Byrne et al., 1987). These compounds preferentially bind to the thyroid hormone transport proteins (Brouwer et al., 1989). We recently completed a study that investigated the interaction between blubber PBDEs in grey seal (Halichoerus grypus) pups and circulating thyroid hormone levels. The pups are fed milk with a very high fat content from their mothers. The organic compounds, which are bound to the lipid, are also transferred, often giving the pups very high doses of contaminants at critical times during their development. We found significant positive correlations between thyroid hormones and blubber PBDEs, after taking age and nutritive condition into account (Hall et al., 2003). An obvious follow up to these findings is to investigate the population consequences of this relationship, specifically how does this affect survival? Previous studies used a mark-recapture model to investigate if weaning mass, sex, location and immunoglobulin (IgG) levels were important determinants of survivorship (Hall et al., 2002). Here we determine if these putative EDs are also significant covariates of survival probability to age one. Studies on harbour seals (Phoca vitulina) following the 1988 outbreak of phocine distemper (PDV) concluded that their immune system was affected by exposure to a mixture of organochlorine contaminants (DeSwart et al., 1994). However, effects on the endocrine system were not studied. The 2002 outbreak of PDV had a similar major impact on the population of harbour seals in England (c.22% of the Wash population died) whereas mortality rates in Scotland were very much lower (often <10%) (Härkönen et al., Submitted). Research funded by DEFRA in 1989 found that blubber levels of PCBs and DDTs were high in areas of high mortality and were higher among animals which died compared to the survivors (Hall et al., 1992). Here we investigate whether there is also a link between EDs and contaminant burden in harbour seals from different populations around the UK. In addition to determining how exposure to various groups of potential EDs varies in harbour seals around the UK (from the Wash in southeast England to Orkney in the north of Scotland) we investigate relationships between exposure and levels of hormones in the blood. Findings that immune function in harbour seals is affected by exposure to persistent organic compounds may suggest additional effects on the endocrine system. For example, there are well-established effects of thyroid hormones on immune function. Hypothyroid mice have a reduced thymic and splenic mass and decreased numbers of T-cells (Byrne et al., 1987). The effects of thyroid hormones on immune function are also modulated by the thymic peptide thymulin. Several human disorders characterized by reductions in thyroid hormones are also associated with reduced thymulin concentrations concomitant with immune deficiency. It may be that some compounds targeting thyroid function produce a synergistic effect on immunity. After the 1988 epidemic Kendall et al. (1992) found a significant relationship between thymulin and PCB levels when time since exposure to PDV was accounted for. Harbour seals are also susceptible to the reproductive effects of organochlorine contaminants. In the early 1980s the significant reduction in the harbour seal population in the Wadden Sea was attributed to the effects of PCB exposure (Reijnders, 1986). In addition other studies have found negative relationships between testosterone levels and blubber DDTs and PCBs in marine mammals (Subramanian et al., 1987). Here we investigate how circulating thyroid hormones (thyroxine and triiodothyronine) and testosterone or progesterone levels relate to ED blubber burdens in harbour seals.
SID 5 (2/05) Page 6 of 24
1.1 Scientific objectives as set out in the proposalObjective 1. To measure the concentration of PBDEs, PCBs and chlorinated pesticides in the blubber of 60 post-weaned grey seal pups from the Isle of May. Objective 2. To measure levels of thyroid hormones in blood samples from the same pupsObjective 3. To model the survival probability of grey seal pups, using the mark-recapture framework currently being developed for the mobile phone tags, with blubber contaminants and thyroid hormone levels as individual covariates.Objective 4. To measure the concentration of PBDEs, PCBs and chlorinated pesticides in the blubber of harbour seals from 5 UK populations that were victims and survivors of the 2002 PDV outbreak.Objective 5. To measure levels of thyroxine, tri-iodothyronine, progesterone and testosterone in 50 harbour seal survivors of the 2002 PDV outbreakObjective 6. To model the relationship between blubber contaminants, hormone levels and regional mortality levels in five UK harbour seal populations (the Wash, Tay, Moray Firth, Orkney and the West coast of Scotland).
All the above objectives of this study were met and the findings are detailed below.
2.0 Materials and methods
2.1 Endocrine disrupting chemicals and probability of first-year survival in grey seal pups
2.1.2 Study population and survival covariate measuresA random sample of 60 pups born at the Isle of May in the Firth of Forth in 2002 were weighed, measured and sampled for this study. Pups were tagged with flipper tags (Dalton Animal Identification Systems) and weighed twice during lactation. Marked animals were recaptured as post-weaned pups and re-weighed. Pre-weaned mass gain and post-weaned mass loss rates were used to determine estimated mass at weaning (see below). Blood samples for thyroid hormone and total IgG assays and blubber biopsies were collected when animals were first post-weaned. Pups were mildly sedated with a 10-20 mg does of Zoletil 100 (Virbac, France). Length measurements were taken and blood samples were collected from the extradural vein using the Vacutainer™ system (Becton Dickinson, Oxford). This system consists of a double-pointed needle with one end shorter than the other. The long end of the needle is inserted into the vein and the shorter end is used to pierce the rubber stopper of a vacuum tube. The glass vacuum blood collection (Vacuainer™) tubes are sealed with a partial vacuum inside by rubber stoppers. The air pressure inside the tube is negative and after inserting the longer needle into the vein, the tube is pushed into a holder retaining the glass tubes so that the shorter needle pierces the stopper. The difference in pressure between the inside of the tube and the vein causes blood to fill the tube. After clotting, the tubes were spun at 1000g for 10 min to separate the serum from the clot. The serum was then removed using a Pasteur pipette into 1.5ml storage tubes for freezing until analysis. Blubber samples were collected from the lateral pelvic region using a 6mm biopsy punch (Acuderm, Florida, USA). Serum and blubber were stored at –20oC until processing. Mobile phone tags were also fitted (glued to the fur with two part quick setting epoxy resin) and animals were individually colour marked for identification and estimation of tag loss during re-sighting trips. All procedures were carried out under Home Office Animal (Scientific Procedures) Act licences.
Serum samples were analysed for total IgG using a protein A enzyme linked immunosorbent assay (ELISA) as previously reported (Hall et al., 2002). Protein A strongly binds mammalian IgG including seal (Ross et al., 1993) and all samples were analysed in duplicate. In-house grey seal standards were used from purified IgG. Inter and intra-assay coefficients of variation (CVs) were 19.4 and 8.7% respectively. Total thyroxine concentrations were also measured using a commercially available ELISA kit (Fortress Diagnostics, Belfast), which has been validated for grey seals in previous studies. Again all samples were analysed in duplicate and the mean of the two results were used in the final analysis. Controls and standards were included in all assay runs and the inter assay CV was 22.3% with an intra-assay CV of 10.4%. Total triiodothyronine concentrations were also measured using an ELISA kit (Fortress Diagnostics, Belfast). The inter-assay CV was 18.7% and the intra-assay CV was 7.8%. These are all comparable to those reported in other studies where inter-assay CVs for hormones in seals using ELISAs and radioimmunoassay varied between 25.5 and 30% (Gardiner et al., 1999). Intra-assay CVs in these studies were also similar to our CV at between 5.7 and 11.6%. It should be noted that many studies have reported lower inter and intra-assay CVs for hormone ELISAs in seals (e.g. Sørmo et al. 2005, <7.8%) but these were calculated using the kit standards rather than duplicating seal serum samples within and between plates, which results in lower but unrealistic CVs.
2.1.3 Estimating effect of endocrine disrupters on first year survival probabilityIn this study we extended the models of first year survival in grey seal pups (Hall et al., 2001; Hall et al., 2002) to include a number of blubber contaminants that are putative endocrine disrupters. Total serum thyroxine concentrations were also included to establish if any effect of contaminants were due to perturbations in circulating concentrations of these hormones. We used a mark-recapture framework with a live resighting model (Cormack Jolly Seber, or CJS model) and a new method of “resighting” animals using mobile phone tags (McConnell et al., 2004). Live recaptures are the basis of the standard Cormack Jolly Seber (CJS) model.
SID 5 (2/05) Page 7 of 24
Marked animals are released into the population, in our study these were pups fitted with mobile phone tags. Then, marked animals are “encountered”. In many studies this is by catching them alive and re-releasing them. In our study the phone tags sent an SMS text message every 6 hours. The successful receipt of a text message ashore therefore constituted a resighting event or encounter within the coastal zone of phone coverage. If marked animals are released into the population on occasion 1, then each succeeding encounter (successful receipt of a text message) occasion is one encounter occasion, e.g.
Release ----S(1)-----> Encounter 1 -------S(2)------> Encounter 2
Animals survive from initial release to the first re-encounter with probability S(1), and from the first encounter occasion to the second encounter occasion with probability S(2). The recapture probability at encounter occasion 1 is p(2), and p(3) is the recapture probability at encounter occasion 2. At least 2 encounter occasions are required to estimate the survival rate between the first release occasion and the first encounter occasion, i.e., S(1). The survival rate between the last two encounter occasions is not estimable because only the product of survival and recapture probability for this occasion is identifiable.
Generally, the survival rates of the CJS model are labelled as Phi(1), Phi(2), etc., because the quantity estimated is the probability of remaining available for recapture. Thus, animals that emigrate from the study area are not available for recapture, so appear to have died in this model. Thus, Phi(i) = S(i)(1 - E(i)), where E(i) is the probability of emigrating from the study area. However, our phone tags had ‘roaming’ agreements with European mobile phone service providers thus allowing the study area to extend to all the European coastal zones where grey seal pups can disperse.
For each individual seal fitted with a mobile phone tag (n=60) the receipt of a text message to a central phone at SMRU constituted a resighting event. Each tag’s phone number identified it individually. Tags were programmed to attempt to send messages every 6 hours. However, on many of these occasions animals were away from the coastal zone and therefore outside the phone mast coverage area. When animals returned to the haulout site (grey seals return to haulout after approximately 5-6 day foraging trips and will therefore return to within phone coverage if they survive) tags were then able to send a message, which would be coded as a ‘resighting’ event (see McConnell et al., 2004 for more details). Text messages received were concatenated into monthly encounter histories from which first year survivorship could be estimated. Tags lasted 6 months due to battery life so survival estimates were calculated for the period December to June and annual estimates were adjusted accordingly.
Information from double tagging (animals were given highly coloured individual glue marks on their fur in addition to phone tags) and resighting surveys to the major haulout sites on the east coast of the UK allowed us to estimate tag loss rates. Monthly tag retention rates were then used to adjust the final survival estimates. Of the 60 deployed data from 55 were used in the final analysis due to the failure of a small number of tags. Model fitting was carried out using the Program MARK (White and Burnham, 1999). As mentioned above, in the CJS model structure Phi (Φ) is used to denote apparent survival since emigration is often confounded with survival when animals leave the study area. However, we had roaming mobile phone agreements with other European countries. Thus for pups that moved away from UK coverage we were still able to receive messages (messages were received from animals that went to Germany and Norway) and Φ is equivalent to survival without emigration confounding.
Mass/length, as a condition index that reflects the pup’s stored blubber reserves was calculated for each animal. This has been shown by other studies to provide a useful index of total body fat (Hall et al., 2001; Thorson and Le Boeuf, 1994). Because animals could not be sampled exactly on the day they weaned, marked pups were weighed repeatedly before and after weaning and linear regression models were fitted to the pre-weaned growth data and the mass loss data for each animal. Estimated mass at weaning was then calculated from the intersection between the two regression lines (see Hall et al., 2001). The standard length of the pups was assumed to be the same at weaning as when animals were measured and tagged during the postweaning fast.
Blood and blubber biopsy samples were collected when animals were first recaptured after weaning, and this varied by individual. Because contaminant concentrations increase with days after weaning as blubber reserves are utilized for fuel, and recalcitrant contaminants are concentrated in less blubber, the concentrations were also adjusted to give an estimated contaminant concentration at weaning. We found the serum IgG and total thyroid hormone levels in these individuals did not change with days postweaning so the level measured postweaning was assumed to be the same as at weaning (Hall et al., 2001).
Individual covariates of survival (condition, serum IgG, total thyroxine, total PCBS, mono-ortho PCBs and total PBDEs, total DDTs and OC pesticides), are embedded in the survival model in an approach similar to logistic regression. Details of the maximum likelihood mark-recapture method to estimate survival are given in Annex 1.
SID 5 (2/05) Page 8 of 24
2.2 Harbour seals
2.2.1 Study populationsFive populations of harbour seals were studied; the Wash, southeast England; the Tay estuary, southeast Scotland; the Moray Firth, northeast Scotland; Orkney northern Scotland and Islay/Jura, south western Scotland (Fig. 1). Populations were chosen following differential mortality during the 2002 phocine distemper epidemic. Sampling trips were carried out between March and October 2003. Animals were captured on land whilst hauled out using hand held hoop nets or at sea in tangle nets deployed from boats. After capture the seals were weighed, sedated with Zoletil when length and axilliary girth measurements were taken. Animals were categorised as adult or sub-adult based on mass and body length measurements (Thompson and Rothery, 1987). Animals >110cm standard body length (nose-tail) were considered to be adult or sub-adult and those >60kg were categorised as sub-adult with those >70kg as adult (Härkönen and Heide-Jørgensen, 1990). Pups and juveniles were excluded from the study. Blood and blubber samples were collected as for the grey seal pups above and stored at –20oC before analysis in the laboratory. All procedures were carried out under Home Office Animal (Scientific Procedures) Act licences.
2.2.2 Hormone analysesBlood samples were analysed for total thyroxine, total triiodothyronine and progesterone in the females and testosterone in the males. All assays used were commercially available ELISA kits (Fortress Diagnostics, Belfast for thyroid hormones and Serozyme kits, BioStat, Stafford for steroid hormones). These have been validated for use in harbour seals (Gardiner et al., 1996) and all samples were assayed in duplicate with controls and standards included in all assay runs. Total thyroxine inter assay CV was 22.3% with an intra-assay CV of 10.4%. Total triiodothyronine interassay CV was 20% with intra-assay CV of 7.6%. Progesterone inter-assay CV was 24.9% and intra-assay was 5.7%. Testosterone inter-assay CV was 25.2% and intra-assay CV was 6.6%. These are all comparable to those reported in other studies where inter-assay CVs for hormones in seals using ELISAs and radioimmunoassay varied between 25.5 and 30% (Gardiner et al., 1999). Intra-assay CVs in these studies were also similar to our CV at between 5.7 and 11.6%. It should be noted that many studies have reported lower inter and intra-assay CVs for hormone ELISAs in seals (e.g. Sørmo et al. 2005, <7.8%) but these were calculated using the kit standards rather than duplicating seal samples within and between plates which results in lower but unrealistic CVs.
A total of 60 live animals were sampled, 10 from each population except the Wash where 20 samples were analysed. Twenty samples from harbour seals that died of PDV in the Wash in 2002 were included and 3 from the east coast of Scotland. All those chosen from the dead animals were from fresh adult carcasses with no signs of decomposition (assessed by the veterinarians at the Institute of Zoology Marine Mammal Stranding Group using standard condition codes (i.e. code 2) and criteria established under the DEFRA funded UK Marine Mammal Strandings Project) to ensure comparability with samples from live animals. Very few were available from Scotland because of the very low PDV mortality seen there. Of the 37 animals retrieved for post-mortem examination in Scotland during the outbreak only 3 fitted the criteria of condition code 2, adult or sub-adult and died of PDV on the east coast of Scotland.
2.3 Contaminant concentrations in blubber biopsy samplesAll samples were still frozen upon receipt at Lancaster University, where they were kept frozen at –20 °C until analysis. Sample analysis was performed at Lancaster University using methods based on those reported elsewhere (e.g. Thomas et al., 2004). Samples were mixed with anhydrous sodium sulphate, extracted with dichloromethane (DCM) using Soxhlet apparatus, and an aliquot taken for gravimetric lipid determination before the rest of the sample was transferred to hexane. Samples were then cleaned by chromatography using silica gel treated with concentrated sulphuric acid, followed by gel permeation chromatography. Sample extracts were then analysed by GC-MS.
Samples were analysed for 45 PCBs (PCBs 18, 22, 28, 31, 41/64, 44, 49, 52, 54, 60/56, 70, 74, 87, 90/101, 95, 99, 104, 105, 110, 114, 118, 123, 138, 141, 149, 151, 153, 155, 156, 157, 158, 167, 170, 174, 180, 183, 187, 188, 189, 194, 199 and 203) using a GC-MS system (Finnigan TRACE) in EI mode. The concentration of 12 organochlorine pesticides in the blubber samples was also determined. Nine of these (namely -chlordane, -chlordane, HCB, o,p’-DDD, p,p’-DDD, o,p’-DDE, p,p’-DDE, o,p’-DDT and p,p’-DDT,) were also analysed using the Finnigan TRACE GC-MS in EI mode. Samples were injected in splitless mode, and separation was achieved on a 50m CPSil8 chromatography column (Chrompak) with a 2m retention gap. The mobile phase was helium, with a flow rate of 1 ml/min. The mass spectrometer was operated in SIM mode, with 2 ion masses monitored for each chemical of interest. 21 PBDEs (BDEs 17, 28, 32, 35, 37, 47, 49, 71, 75, 77, 85, 99, 100, 119, 138, 153, 154, 166, 181, 183 and 190) and the 3 remaining pesticides, -HCH, -HCH and -HCH, were analysed using a GC-quadrupole MS system (Fisons MD800 or Finnigan TRACE) in NCI mode, using ammonia as the reagent gas. Samples were injected in splitless mode, and separation was achieved on a 30m DB5MS chromatography column (J&W) with a 2m retention gap. The mobile phase was helium, with a flow rate of 1 ml/min. The mass spectrometer was operated in SIM mode, with 2 ion masses monitored for each chemical of interest.
SID 5 (2/05) Page 9 of 24
Calibration and quantification was achieved in the same way on each instrument. A set of seven calibration standards, containing all of the internal standards, recovery standards, and analytes, was run on the instrument before and after a batch of up to 24 samples. Within the batch, after every 6 samples, a specially prepared ‘QC’ standard was run to check the continuing validity of the calibration. Samples were quantified using the Thermo ‘Xcaliber’ instrument software, and calibration and quantification was achieved using an internal standards method. To be accepted as the analyte of interest a peak in a sample had to be eluted from the GC column at the correct retention time, and show the correct ratio (compared to the standards) between the two ions measured. After quantification, data processing was completed using Excel spreadsheets. Quality control information is given in Annex 1.
3.0 Results
3.1 Endocrine disrupters as covariates of survival in juvenile grey seals
3.1.1 Individual chemical congenersPolychlorinated biphenyls (PCBs)Forty-one individual PCB congeners were identified and quantified in the blubber biopsy samples obtained from 60 post-weaned pups from the Isle of May. The concentrations of all congeners measured, expressed on a lipid weight basis were lognormally distributed. Due to this non-normality parametric statistical analyses were only carried out on the log10 transformed concentrations. Details of the average concentrations by congener are given in Annex 1. To examine the congener profile patterns in the blubber samples, median and interquartile ranges for each congener were calculated and are plotted in Fig. 2. The profiles are dominated by the hexachlorobiphenyl congeners (particularly IUPAC No.s CB138 and CB153) followed by the hepta- and pentachlorobiphenyls. These higher chlorinated compounds are more recalcitrant in marine mammals, are not substantially metabolised and excreted and therefore bioaccumulate in the lipid rich blubber tissues.
Of the 41 congeners measured in the blubber biopsy samples, 15 were the mono-ortho PCBs (congeners that have chlorine substitute at one and only one of the 2,2’ or 6,6’ positions) that are potential thyroid hormone disrupters (Ness et al., 1993). Fig. 3 shows the congener profiles for these compounds, which are dominated by the tri and tetrachlorobiphenyls with CBs 118 and 156 (penta and hexachlorobiphenyls) also at relatively higher concentrations. However, the levels of these congeners were an order of magnitude lower than the other hexachlorobiphenyls CBs138 and 153 that dominate the overall pattern. As has been found elsewhere, concentrations of many of the individual congeners were highly correlated with each other. Pearson correlation coefficients among the 41 individual PCB congeners were calculated and the majority (71% ) showed significant correlations. As expected the lower chlorinated congeners were more correlated with each other than they were to the higher chlorinated congeners, with a similar trend being seen among the higher chlorinated congeners.
DDT and OC PesticidesThe geometric mean and 95% geometric confidence limits for the DDT and OC pesticides measured in grey seal blubber samples are given in Annex 1. The DDT congener profile is dominated by p,p’-DDE and p,p’-DDT with the other congeners being at the detection limit. Of the other OC pesticides measured , gamma-hexachlorocyclohexane (γ-HCH or the insecticide Lindane which is 99% pure γ-HCH) was the dominant compound in this group followed by β-HCH, used as an insecticide and by α-HCH, which is also a component of Lindane. Smaller quantities of hexachlorobenzene (HCB, used as a fungicide and as a waste product of other chlorinated pesticides) were also found. Concentrations of chlordane were at or just above the detection limit.
Polybrominated Diphenyl Ethers (PBDEs)Within the PBDE group of compounds, 21 individual congeners were measured in the grey seal pup blubber samples and the profiles (median and interquartile range) are shown in Fig. 4. As with the OC compounds, concentrations were lognormally distributed. The geometric mean and 95% geometric confidence limits for each congener are given in Annex 1. The profile is totally dominated by BDE47, the tetra-bromodiphenyl ether. All the other congeners make only a small contribution to the total PBDE concentration in the blubber. Unlike the PCBs the levels of the higher brominated compounds are not significantly higher than the lower brominated congeners. The Pearson correlation coefficients were also calculated for all the PBDE congeners. There was a very high degree of correlation among all the congener concentrations (76% of them were significantly correlated with each other), except for the most abundant congeners BDE47 and BDE100. Deca-BDE (BDE209) was not found in the seal blubber samples analysed.
3.1.2 Total Blubber Contaminant ConcentrationsThe congener sums for the different groups of contaminants were calculated to give an overall estimate of blubber concentrations for each chemical group. Because of the high degree of correlation between the congeners within each group these totals were used in the survivorship models, with the exception of the PCBs where the relationship between first-year survival and the sub-set of mono-ortho PCBs was also investigated. The profiles for the different contaminant groups (median concentrations, ± interquartile ranges) are shown in Fig.
SID 5 (2/05) Page 10 of 24
5. These are dominated by the PCBs with the DDTs, PBDEs and HCHs making up the remainder at approximately the same concentrations.
3.1.3 Total serum thyroxine, triiodothyronine and IgG concentrationsThe mean total thyroxine (TT4) concentration was 34.36 nmol/l ± SD 13.24 (26.70 ± SD 10.29 ng/ml). A Kolmogorov-Smirnov (KS) test indicated the data were normally distributed (p=0.298). The range of levels was between 13.1 and 70.6 nmol/l (10.18 – 54.86 ng/ml). The concentrations of total triiodothyronine (TT3) were not normally distributed (KS test, p<0.0001) thus levels were log-transformed for analysis. The geometric mean concentration was 1.805 nmol/l (1.17 ng/ml), with 95% geometric confidence interval of 1.58 – 2.07 nmol/l 1.03 – 1.34 ng/ml). The mean total IgG concentrations were not normally distributed (KS test p=0.002). The geometric mean was 23.4 g/l with 95% geometric confidence interval of 19.5 – 27.5 g/l.
3.1.4 Relationship between serum thyroxine, triiodothyronine and blubber contaminantsA linear least squares model was used to investigate the relationship between serum thyroxine and blubber contaminants in the post-weaned pups. As we have found in a previous study total PCBs, DDTs and PBDEs (with interaction between the contaminants) and after controlling for girth and age (days post-weaning) as confounders, were significant predictors of serum thyroxine (F8,46=2.89, p=0.011). The coefficients from this model were positive suggesting animals with higher contaminant levels had increased concentrations of thyroxine in the blood. Similarly for serum triiodothyronine total PCBs and PBDEs (but not DDTs) were significant predictors of circulating concentrations of triiodothyronine, after controlling for girth (F4,50=2.71, p=0.04). Again the coefficients were positive and there was a significant positive linear relationship between thyroxine and triiodothyronine concentrations (F1,53=15.04, p=0.0003); data not shown.
3.1.5 Mass, size and condition of pups and sex differencesThere was a significant difference between the sizes of sexes, with males being slightly larger than females (two-sample t-test, unequal variances, p=0.03, females mean=38.85 kg ± SD 5.0, n=28; males 42.3, ± SD 6.46, n=27). All morphometric measurements were highly linearly and significantly positively related (p<0.0001; mass vs girth adjusted R2=0.850; mass vs length adjusted R2=0.592; length vs girth adjusted R2=0.491). There were no significant differences (two-sample t-test, unequal variances) between the sexes among any of the contaminant concentrations or in the total thyroxine or triiodothyronine concentrations. The mean or geometric mean for the lognormally distributed parameters (contaminants and triiodothyronine) and standard deviations are shown in Table 1.
3.1.6 Relationship between blubber contaminants in post-weaned grey seal pups and first year survival probabilityGoodness of fitIn the first stage of modelling a goodness of fit test of the model to the data without covariates was carried out. This is because the program MARK is currently unable to assess the goodness of fit for models with covariates. Males and females were considered separately as group covariates and survival and recapture probabilities were allowed to vary with time. All combinations of sex and time varying survival and recapture were included in the first candidate set of models. The reduced m-array for the data by sex is given in Annex 1. The goodness of fit tests found there were no significant recapture or survival problems in this dataset. To account for the effect of lack of fit on the model selection, a parametric bootstrap goodness of fit procedure incorporated in the program MARK was used (Burnham and Anderson, 1998) to estimate an overdispersion factor, which was then used to adjust the AICc statistics. The adjusted AICc is known as QAICc. The over-dispersion factor c is estimated using the ratio of the observed deviance statistic to the average value obtained from the bootstrap replicates. This indicated some mild overdispersion (č = 1.157).
Survivorship modelsModels constructed in the first stage of model fitting are shown in Table 2. These are the models without the individual covariates. The best model from this stage was found to be survival varying by sex (males and females having different survival probabilities) and recapture varying with time. Although the model with time dependent survival has ΔAIC of <2 and weight of 0.386, a likelihood ratio test comparing the two models found no evidence that it is important (p=0.0997), so the second stage of modelling incorporated sex-dependent survival and time-dependent recapture. Variations in the recapture probabilities with time are shown in Fig. 6. Lower recaptures in February to April probably reflect an exploratory behavioural phase where the pups disperse to sea before establishing foraging patterns similar to adults and returning to haul out sites on a more regular basis.
The results of the model selection process for the second stage are shown in Table 3. The best models all had condition and covariate effects on survival, survival was different between males and females and recapture probabilities varied with time. The top three models with a combined QAICc weight of 0.6 indicated that condition (i.e. total body fat) at weaning was again found to be an important predictor of survival and the second model found total serum IgG to be important but the ΔAICc value = 2.25 and model likelihood = 0.324, indicating support for this effect in the dataset is not very strong. The next model included condition and total thyroxine levels but, again the ΔAICc > 2 and the models that include both PCBs and PBDEs after controlling for the effects of
SID 5 (2/05) Page 11 of 24
condition do not have substantial support. This was also seen when the models were run with the mono-ortho PCBs, the DDTs and the OC pesticide concentrations (data not shown). Despite a significant relationship between serum thyroxine and blubber PCBs and PBDEs being found in this sample of grey seal pups we conclude that this is unlikely to reduce the probability of first year survival.
A third stage of model building was carried out to investigate the relationship between triiodothyronine levels, blubber contaminants and first year survival. In addition models including concentrations of both thyroid hormones were investigated. The results are shown in Table 4, where again the best model included condition dependent survival by sex and time dependent recapture. The second best model given the data, with a ΔAICc value of 1.03 (and therefore with some evidence for support), includes triiodothyronine in addition to condition. However, again the models with PCB and PBDE dependent survival both alone and in conjunction with triiodothyronine levels were not ranked highly. Although there was some limited evidence that triiodothyronine levels in addition to condition and sex are important predictors of first year survival, this is not mediated through endocrine disrupting effects of blubber PCBs and PBDEs.
From the best model that included condition dependent survival, the coefficients for the sex and condition effects were males 1.61 ± SE 0.55; male condition 2.42 ± SE 1.08; females 2.04 ± SE 0.48; female condition 0.69 ± 0.42 thus the odds of survival was estimated to be 1.5 times higher for females that males, regardless of condition. Males in poor condition had the lowest probability of survival. This gives us an estimated annual survival probability for males of 47.5% (after accounting for the effect of tag loss). For females the survival was found to significantly higher at 63.9%.
3.2 Endocrine disrupters in UK harbour seals and relationships with serum thyroid and reproductive hormone levels.
3.2.1 Regional and seasonal variations in blubber contaminants of live sealsAs for the grey seal samples concentrations of individual congeners within each chemical group were highly correlated with each other and frequency distributions found them to be lognormally distributed. Thus for this phase of the study relationships with total contaminant levels within chemical groups based on summation of individual congeners (as described above) were used in the final analysis, log transformed on a lipid weight basis.
Table 5 shows the number of harbour seals sampled by sex and age category at each location. For each site we were able to sample an even number of males and females (5) except in the Wash where the sample comprised significantly more females (15) than males (5). Most samples were obtained from adults except 3. There were significant differences in size (mass, length and girth) of harbour seals between the sites (Fig. 7). Animals from Orkney were heavier and larger than those captured elsewhere and two animals from the Wash were significantly smaller than the others. In a general linear model after controlling for differences by sex and age, length and mass were significant predictors of location (analysis of deviance Pr χ2 <0.0001). Seals in Orkney were captured in June just prior to the breeding season when all females in the sample were pregnant. Since time of year is therefore confounded with location we were unable to explicitly control for seasonal variation. However, when Orkney was excluded from the analysis mass, length and girth remained significantly different among the sites.
Blubber concentrations of total PCBs, DDTs, OC pesticides, HCHs and PBDEs are given in Table 6. Geometric mean and 95% confidence intervals are shown. The highest concentrations of PCBs were found in the animals from Islay/Jura and the lowest in Orkney. Harbour seals from the Wash and Tay estuaries had comparable levels. DDTs were however highest in the Tay estuary animals and in the Wash whereas PBDEs were highest in the Wash samples (over 2.5 times those seen elsewhere). However, absolute concentrations were an order of magnitude lower than the PCB levels.
Linear least squares models were used to determine if the above differences between contaminant concentrations by location or season were significant. It was possible to control for differences by sex, age and size (mass, length and girth) by including these as predictors of contaminant concentrations in the model. Condition indices were not used for harbour seals, as we have no independent data to indicate how useful this is for predicting total body fat for this species. Stepwise regression was used to determine the most important predictors and the best fitting models given the data for each contaminant group. The analysis of variance table results for each model are shown in Table 7. Differences by location were seen among all contaminant groups except DDTs where location was not a significant variable. For all contaminant groups except HCHs sex and age were important after controlling for location differences and size (mass, length or girth) was important for the most abundant contaminants, PCBs, DDTs and PBDEs. The adjusted R2 values for each model were: PCBs 0.73, DDTs 0.57, PBDEs 0.77, HCHs 0.18, and pesticides 0.32. Thus for the abundant contaminants the model accounted for between 57 and 77% of the variance in blubber concentrations whereas the models were less predictive for the less abundant contaminants.
SID 5 (2/05) Page 12 of 24
3.2.2 Serum thyroid hormone and reproductive hormone concentrations.Mean thyroid hormone concentrations in harbour seals were: total thyroxine 90.1 nmol/l (95% CI 77.3, 103.0; mean 70 ng/ml, 95% CI 60.1 – 80.0 ng/ml) total triiodothyronine 1.78 nmol/l (95% CI 1.6, 2.0; mean 1.16 ng/ml 95% CI 1.04 – 1.30 ng/ml). Mean testosterone concentrations measured in the mature males were 12.1 ng/ml (95% CI 5.5, 18.7) and 0.85 ng/ml (95% CI 0.6, 1.1) in the immature males. There was a significant positive linear relationship between total triiodothyronine and total thyroxine serum concentrations (p=0.02, Fig. 8). Progesterone concentrations were lognormally distributed as some pregnant females had very high levels. Geometric mean progesterone concentrations were 99.1 ng/ml (95% CI 60.25, 165.9). Analyses were carried out using the transformed values. Significant differences among locations were only found for the progesterone concentrations and two females in Orkney who had extremely high levels associated with parturition largely drove this.
3.2.3 Relationships between serum thyroid hormone and blubber contaminant concentrations.In the next stage of the analysis relationships between thyroid hormone levels in the blood of harbour seals and different categories of contaminants were explored. Linear least squares models were again used. Table 8 shows the analysis of variance table results from the linear models for the best predictors of thyroid hormones controlling for variables such as age, sex and size which from the previous stage of analysis were found to be related to contaminant concentrations. Significant relationships between blubber PCBs, PBDEs and pesticides and total triiodothyronine concentrations were found after controlling for the effect of the confounding variables identified previously. PBDEs and pesticides were the most important of the contaminant groups. In all cases the relationships were positive where higher levels of triiodothyronine were found with increasing concentrations of contaminants in the blubber. Total triiodothyronine levels are increased during pregnancy but when all females were excluded from the analysis this increased the significance of the relationship between triiodothyronine and contaminants (Table 8). Total thyroxine, despite the correlation between these two hormones, showed only marginally significant relationships with the pesticide group of contaminants (Table 9).
3.2.4 Relationships between serum reproductive hormone and blubber contaminant concentrations.There was no significant relationship between progesterone concentrations and any of the blubber contaminant groups in the female harbour seals after controlling for differences due to location or season, age and size. For the testosterone concentrations however, PBDEs and HCHs were significant predictors of testosterone after controlling for the effect of location or season (which are confounded) and age (Table 10). Age was also significant as expected (adjusted R2 PBDEs = 0.64, HCHs = 0.61). This relationship was negative thus lower concentrations were found in animals with higher PBDE and HCH blubber levels. PBDEs accounted for 23% of the variance in testosterone and HCHs for 19%.
3.2.5 Differences between victims and survivors of the phocine distemper epidemicFig. 9a shows the blubber concentration of total PCBs and total PBDEs (log scale) in harbour seals from the Wash and east coast of Scotland that were victims (Wash n=20, east coast n=3) and survivors (Wash n=20, Tay estuary n=10) of the 2002 PDV epidemic. The concentrations of total PCBs were significantly higher in the animals that died compared to the survivors in the Wash but not the Scottish animals (although only 3 samples were available from dead animals). In a binomial generalized linear model with state (live or dead) as the dependent variable (data for the Wash only) and accounting for differences in size and condition (girth), total PCBs were significantly related to state (analysis of deviance Pr χ2 = 0.005). There was however no difference in the other contaminant groups (total PBDEs are shown in Fig. 9b) between victims and survivors.
4.0 Discussion
Endocrine disrupters in post-weaned grey seal pups and effects on first year survival
We used a mark-recapture survival study to investigate if blubber EDs (various groups of persistent organic contaminants) in post-weaned grey seal pups were important covariates of first-year survival probability. Although we again found a significant relationship between blubber PCBs and PBDEs and total serum thyroxine levels, after taking account of condition and age effects, this did not appear to affect the survivorship of the pups to age one. The most important predictors of survival were condition (total fat) at weaning and sex (males have a lower survival probability than females). These results confirmed our earlier findings but within the dataset there was no substantial evidence to support the hypothesis that blubber EDs (transferred from the mother in the milk during lactation) significantly affect survival or that these effects are mediated through impacts on the levels of the thyroid hormone thyroxine. It may be that we did not have sufficient detection power in our study but the model fit to the data was good and we were able to adjust the models for the small sample size by using variance inflation. In addition the standard errors around the resulting survival estimates were not large and the first year survival estimates of between ~40-60% again fit with our earlier findings (Hall et al., 2001) and with those produced from the grey seal population model (Duck, personal communication). By using the new mobile phone tag technology developed at SMRU to replace passive tags to mark seals we were able to conduct a study with similar power using only half the number of study animals.
SID 5 (2/05) Page 13 of 24
Other studies into the immunosuppressive effects of these contaminants on grey seals have found them to be relatively resistant, particularly when compared to harbour seals (Hammond et al., In Press). Indeed a study of the effects of lactational exposure on immunity in grey seal pups from the same population in the 1990s found a similar negative effect (Hall et al., 1997). Further research is needed to establish whether the relationship between thyroid hormones and contaminants we have found in this and previous studies (Hall et al., 2003) is causing any longer term health effects in this species. We have only been able to investigate effects on survival during the first year of life but it is possible, since thyroid hormones are important in growth and metabolism that there are sub-lethal effects we have not yet observed. Other studies in rodents found perinatal exposure to PCB congener CB 118 led to a significant increase in thyroxine in the offspring as a ‘thyroid resistance syndrome’ (Kuriyama et al., 2003). They suggest that exposure of the offspring to the PCB during gestation or lactation may have altered expression of the TR-β gene leading to reduced responsiveness of target tissues to thyroid hormone characterised by elevated serum levels of thyroxine that appears to be invariably associated with mutations in the TR-β gene. We did not find a relationship between CB118 alone and thyroxine levels in the pups but the studies on rodents were single CB congeners rather than the mixtures (both of PCB congeners and other contaminants) that may produce somewhat different responses in these wild species. Further studies at the molecular level would help to resolve whether our finding is also related to changes in TR-β.
Endocrine disrupters in harbour seals and interactions with disease
Harbour seals are more susceptible to both the immunosuppressive effects of contaminants and to infection with PDV than the sympatric grey seals (De Swart et al., 1996; Hammond et al., In Press; Pomeroy et al., 2005). We found distinct regional differences among contaminant levels in harbour seals from around the UK with seals on Islay and Jura having the highest levels of PCBs and seals in the Wash having the highest levels of PBDEs. Animals from the north of Scotland had the lowest levels of all contaminant groups studied. However, for logistical reasons and restrictions caused by the weather, it was not possible for us to sample all the populations simultaneously so seasonal differences are confounded with regional differences in this study. However, we were able to control for morphometric variations in the harbour seals sampled (particularly using mass, length and girth measurements) and investigate residual variation by location. If seasonal changes are mostly manifest as changes in size and sex effects are additionally controlled for, then we can conclude that the differences we have found are largely driven by variation in contaminant uptake from the environment where the seals forage. A further caveat is that for the pregnant females we sampled in Orkney hormonal changes during the late stages of pregnancy might affect the way contaminants are metabolised since the cytochrome P450 enzymes that are induced by exposure to certain organochlorine contaminants are also involved in the metabolism of steroid hormones. Indeed the PCB, DDT and PBDE levels found in the seals from this region were the lowest compared to elsewhere. However, since the congener profiles are dominated by the recalcitrant congeners CB153 and CB138 this is unlikely to have been a large confounder biasing the regional differences observed.
Significant relationships between blubber contaminants and total triiodothyronine concentrations were found in the harbour seals, after controlling for the effects of confounding factors. PBDEs and pesticides were the most significant predictors and in all cases the relationship was again a positive one, with higher levels of triiodothyronine being seen with increasing concentrations of contaminants. This interesting finding is in line with our previous studies on grey seal pups where hyperthyroid effects were related to blubber levels of PBDEs. This was not due to sampling pregnant females who have increased triiodothyronine levels in their blood, as the relationship was more significant when females and breeding season males were removed from the dataset. The increased levels in animals with higher contaminant exposure could be due to preferential secretion of triiodothyronine but information on levels of thyroid stimulating hormone (TSH) in the blood would be required to further understand this phenomenon. We currently do not have a method for measuring TSH in seals as the reagents available for humans and laboratory species do not cross-react in seal serum.
Significant negative relationships between contaminant levels in the blubber and testosterone levels in males (again after controlling for confounders) were also found. This has been noted for other marine mammal species (Subramanian et al., 1987) but not previously in harbour seals. The mechanism for this effect has not been determined but similar relationships were more recently observed for PCBs and OC pesticides in polar bears from Svalbard (Oskam et al., 2003). Further studies at the cellular and molecular level would allow us to determine the nature of this observation and find out how important anti-androgen effects, particularly of PBDEs would be for harbour seals.
Finally we investigated the difference between potential ED contaminant concentrations in harbour seals that died of PDV during the 2002 epidemic with concentrations in the survivors. This was restricted to seals from the Wash in southeast England as only 3 carcasses sampled in good condition were available from Scotland. We found significantly higher concentrations of PCBs (but not other contaminants) in the seals that died of PDV than the live animals. Again we controlled for other factors (such as sex, age and size) that might be driving this finding. This is in line with the results we obtained following the 1988 epidemic and is suggestive of an interaction between the contaminants and the virus. Studies have shown PCBs particularly to be immunosuppressive and Kannan et al. (2000) collated all the data on marine mammals (much of which is on harbour seals) to determine a threshold for
SID 5 (2/05) Page 14 of 24
immunosuppressive effects. The geometric mean concentration of total PCBs in the animals that died was above this threshold. This finding combined with our observations that contaminants are related to thyroid hormone and testosterone levels suggests that further work is needed to study the potential synergistic or additive effects of the compounds that are EDs and cause immunosuppression. Given the close relationship between immunity and the endocrine system (and indeed specifically the links between the thyroid gland and immune function) it is not inconceivable that both systems are being impacted.
Although the production and use of PCBs and many of the OC pesticides has been banned both in Europe and worldwide and the production and use of PBDEs is similarly now controlled, many compounds are still entering the marine environment through various waste streams. Comparing the concentrations of specific PCB congeners over the period between the PDV epidemics in 1988 and 2002 we found little evidence of a reduction in blubber levels in harbour seals. In fact geometric mean levels in the blubber of animals that died were an order of magnitude higher in 2002 than in the animals that died in 1988. Whilst the method for measuring PCBs in blubber samples has changed between 1988 and 2002 (previous methods used high resolution capillary glass chromatography with electron capture detection, with determinants being confirmed by gas chromatography-mass spectrometry, Hall et al, 1992) allowing more accurate determination particularly of lower chlorinated congeners, the log10 concentrations of CB28 and CB101 were not significantly different between the two years (two sample t-test, unequal variances, p>0.05). However, significant differences were observed for the other 5 congeners measured in the two sets of samples and in the log10 sum of the 7 congeners (p=0.011). Levene’s test for equality of the variances indicated the variances were unequal. This indicates the high degree of individual variability in the blubber concentrations, with a few individuals contributing to the very high levels. In addition it was found that 9/10 animals sampled in 1988 were less than 2 years old whereas all the animals sampled in 2002 were adults and 12 were adult males. Although a legacy of contaminants are passed from mother to pup in the milk during lactation and young animals can have relatively high blubber PCB concentrations, animals continue to store PCBs in their blubber throughout their life. Whereas females can then offload some of their burden during annual reproduction and lactation, adult males continue to accumulate blubber PCBs. This sampling variability could therefore account for some of the difference between the two datasets. However, after the 1988 epidemic live harbour seals from Strangford Lough in Northern Ireland sampled in 1989 had levels comparable to those seen in the harbour seals that died of PDV in the Wash in 2002 (Hall et al., 1992). Long term datasets and monitoring by CEFAS and DEFRA on other marine mammals in the North Sea, particularly harbour porpoise (Phocoena phocoena) also found blubber PCB concentrations to be stable between 1989 and 2001 (Sabin et al., 2004) and levels in guillemot eggs have not declined since the mid 1980s (European Environmental Health Strategy, 2005: http://europa.eu.int/comm/environment/health/pdf/annex5_ktl.pdf) indicating that although the sources and inputs of PCBs to the environment have been controlled and are declining, top predators as environmental integrators with long life spans and generation times are slow to benefit from these improvements in quality.
The brominated flame-retardants such as PBDEs appear to be EDs in harbour seals and concerns regarding the continued use of deca-bromodiphenyl ether that may degrade to lower brominated compounds remain (Keum and Li, 2005). These findings are clearly also important to our predictions of the impact of another PDV outbreak among UK seals. Harbour seals in the Wash seem most vulnerable but perhaps of similar concern is the population on the southeast coast of Scotland as they have the highest levels of PCBs, again above the threshold for effects. PDV did not find its way into this population in 2002 but this may have been a chance event and future introductions of the virus may have a different pattern of spread.
In both grey and harbour seals relationships between blubber contaminants and serum hormones suggest these organohalogenated compounds are potential endocrine disrupters in these species. However, for post-weaned grey seal pups the relationship does not appear to affect their first year survival probability. Further research, particularly using in vitro approaches would elucidate the mechanisms underlying the relationships reported here.
SID 5 (2/05) Page 15 of 24
Table 1. Mean and geometric mean (and 95% confidence intervals) of total PCBs, total PBDEs, total DDTs, total HCHs and total pesticides, total thyroxine and total triiodothyronine concentrations in post-weaned grey seals from the Isle of May, 2002 (n=55).
Total PCBs (ng/g lipid)1
Total PBDEs (ng/g lipid)1
Total DDTs (ng/g lipid)1
Total HCHs (ng/g lipid)1
Total Pesticides (ng/g lipid)1
Total thyroxine (ng/ml)
Total triiodothyronine (ng/ml)1
Mean 1096 141 229 132 14 26.7 1.1795% CI 933 - 1288 129 - 158 195 - 269 120 - 144 12 - 17 23.9 – 29.4 1.02 – 1.341 Geometric means for lognormally distributed data
Table 2. QAICc, ΔQAICc, QAICc weights, model likelihoods, number of parameters and model deviances from the best 5 candidate models from the first stage of the model selection process. Models are ordered by their weights (t=time, phi=survival, p= recapture).
Model AICcDeltaAICc
AICcWeight
Model Likelihood No. Parameters Deviance
{phi(sex),p(t)} 355.775 0 0.58728 1 8 84.859{phi(t),p(t)} 356.613 0.84 0.38637 0.6579 11 78.601{phi(sex),p(sex)} 362.21 6.44 0.02352 0.04 4 100.202{phi(sex),p(sex*t)} 366.455 10.68 0.00282 0.0048 14 80.962{phi(sex*t),p(sex*t)} 377.554 21.78 0.00001 0 22 69.958
Table 3. QAICc, ΔQAICc, QAICc weights, model likelihoods, number of parameters and model deviances from the best 12 candidate models from the second stage of the model selection process. Models are ordered by their weights (t=time, phi=survival, p= recapture).
Delta QAICc Model
Model QAICc QAICc Weight LikelihoodNo. Parameters QDeviance
1{phi(sex*condition),p(t)} 295.673 0 0.37079 1 10 273.7432{phi(sex*condition + IgG),p(t)} 297.927 2.25 0.12018 0.3241 11 273.593{phi(sex*condition + total thyroxine),p(t)} 298.055 2.38 0.11268 0.3039 11 273.7194{phi(sex*condition + PBDEs),p(t)} 298.061 2.39 0.11235 0.303 11 273.7255{phi(sex*condition + PCBs),p(t)} 298.073 2.4 0.11169 0.3012 11 273.7376{phi(sex*mass + PBDEs),p(t)} 299.544 3.87 0.05352 0.1443 11 275.2087{phi(sex*condition* PBDEs),p(t)} 300.34 4.67 0.03595 0.097 12 273.5558{phi(sex*condition + IgG+ total thyroxine),p(t)} 300.365 4.69 0.03551 0.0958 12 273.5799{phi(sex*condition + PBDE+thyroxine),p(t)} 300.472 4.8 0.03366 0.0908 12 273.68610{phi(sex*condition + PBDE+thyroxine+IgG),p(t)} 302.839 7.17 0.01031 0.0278 13 273.55911{phi(sex*condition*PBDE*thyroxine),p(t)} 305.278 9.6 0.00304 0.0082 14 273.4612{phi(sex),p(t)} 309.838 14.16 0.00031 0.0008 8 292.596
Table 4. QAICc, ΔQAICc, QAICc weights, model likelihoods, number of parameters and model deviances from the best 10 candidate models from the third stage of the model selection process. Models are ordered by their weights (t=time, phi=survival, p= recapture).
Delta QAICc Model
Model QAICc QAICc Weight LikelihoodNo. Parameters QDeviance
1{phi(sex*condition),p(t)} 295.673 0 0.28830 1 10 273.7432{phi(sex*condition + triiodothyronine),p(t)} 296.703 1.03 0.17233 0.5978 11 272.3663{phi(sex*condition + IgG),p(t)} 297.927 2.25 0.09344 0.3241 11 273.5904{phi(sex*condition + thyroxine),p(t)} 298.055 2.38 0.08762 0.3039 11 273.7195{phi(sex*condition + PBDEs),p(t)} 298.061 2.39 0.08735 0.3030 11 273.725
SID 5 (2/05) Page 16 of 24
6{phi(sex*condition + PCBs),p(t)} 298.073 2.40 0.08684 0.3012 11 273.7377{phi(sex*condition* PBDEs + triiodothyronine),p(t)} 298.372 2.7 0.07479 0.2594 12 271.5868{phi(sex*condition + thyroxine + triiodothyronine),p(t)} 298.995 3.32 0.05476 0.1899 12 272.2109{phi(sex*condition + PCBs + triiodothyronine),p(t)} 299.011 3.34 0.05434 0.1885 12 272.22512{phi(sex),p(t)} 309.838 14.16 0.00024 0.0008 8 292.596
Table 5. Numbers of harbour seals sampled by sex, age and locationIslay/Jura Moray Firth Orkney Tay Estuary Wash
Adult Males 5 5 5 5 3Sub-adult Males 0 0 0 0 2Total Males 5 5 5 5 5Adult Females 4 5 5 5 14Sub-adult Females 1 0 0 0 1Total Females 5 5 5 5 15Total 10 10 10 10 20
Table 6. Geometric mean and geometric 95% confidence limits for blubber contaminants (ng/g lipid) in harbour seals by location and month
Location Sum PCBs Sum DDTs Sum PBDEsSum Pesticides Sum HCHs
Tay Estuary (Mar) LCL 2003.85 588.87 188.77 4.48 85.89mean 3947.82 830.45 269.93 5.15 97.13UCL 7777.68 1171.15 385.99 5.93 109.84
Moray Firth (Apr) LCL 1035.99 277.59 101.49 3.50 97.79mean 1746.50 391.87 118.87 5.49 120.05UCL 2944.29 553.19 139.21 8.61 147.37
Orkney (June) LCL 556.84 105.01 68.75 4.13 91.35mean 1264.55 220.65 96.54 5.18 108.60UCL 2871.70 463.63 135.56 6.49 129.10
Islay/Jura (Sep) LCL 2532.49 235.79 154.67 5.88 110.69mean 8177.08 544.20 245.03 7.29 142.28UCL 26402.68 1256.02 388.17 9.05 182.89
Wash (Oct) LCL 3394.46 500.28 476.73 4.49 109.73mean 4975.30 650.16 630.03 5.01 123.39UCL 7292.34 844.95 832.64 5.58 138.75
Table 7. Analysis of variance tables from linear least squares best fitting models showing the significant independent variables for each contaminant group.Sum PCBs d.f SS MS F Value p-valueLocation 4 4.718 1.179 14.297 <0.0001Sex 1 6.399 6.399 77.566 <0.0001Age 1 0.530 0.530 6.426 0.014Mass 1 0.392 0.392 4.750 0.034Sum DDTsSex 1 1.841 1.841 28.768 <0.0001Age 1 0.013 0.013 0.204 0.653Mass 1 2.432 2.432 38.010 <0.0001Girth 1 0.386 0.386 6.039 0.017Sum PBDEsLocation 4 5.971 1.493 37.591 <0.0001Sex 1 0.638 0.638 16.076 0.000Age 1 0.028 0.028 0.696 0.408Length 1 0.132 0.132 3.336 0.074Sum HCHsLocation 4 0.159 0.040 3.014 0.026Sum Pesticides
SID 5 (2/05) Page 17 of 24
Location 4 0.197 0.049 2.643 0.044Sex 1 0.217 0.217 11.620 0.001Age 1 0.049 0.049 2.616 0.112
Table 8. Analysis of variance tables from linear least squares models for the best predictors of total triiodothyronine concentrations in the serum for each group of contaminants in the blubber showing significant relationships.Sum PCBs df SS MS F Value Pr(F)Location 4 20.441 5.110 15.130 <0.0001Sex 1 3.997 3.997 11.833 0.001Age 1 0.537 0.537 1.589 0.213Mass 1 0.010 0.010 0.029 0.865logPCBs 1 1.560 1.560 4.618 0.037Sum PBDEsLocation 4 20.441 5.110 16.139 <0.0001Sex 1 3.997 3.997 12.622 0.001Age 1 0.537 0.537 1.695 0.199Mass 1 0.010 0.010 0.031 0.860Girth 1 0.443 0.443 1.398 0.243logPBDEs 1 2.490 2.490 7.862 0.007Sum PesticidesLocation 4 20.441 5.110 16.260 <0.0001Sex 1 3.997 3.997 12.716 0.001Age 1 0.537 0.537 1.708 0.197logPesticides 1 2.428 2.428 7.726 0.008Males Only (excluding breeding season) Sum PCBs SS MS F-value Pr(F)Location 3 19.374 6.458 18.438 <0.0001Age 1 0.258 0.258 0.738 0.395Mass 1 0.073 0.073 0.209 0.650logPCBs 1 4.809 4.809 13.726 0.0006Sum PBDEsLocation 3 19.37 6.458 18.711 <0.0001Age 1 0.258 0.258 0.748 0.392Mass 1 0.073 0.073 0.212 0.647logPBDEs 1 5.028 5.028 14.568 0.0004Sum PesticidesLocation 3 19.374 6.458 18.381 <0.0001Age 1 0.258 0.258 0.735 0.396Mass 1 0.073 0.073 0.208 0.650logPesticides 1 4.761 4.761 13.552 0.0006
SID 5 (2/05) Page 18 of 24
Table 9. Analysis of variance tables from linear least squares models for the best predictors of total thyroxine concentrations in the serum for Sum Pesticides in blubber.Sum Pesticides df SS MS F Value Pr(F)Location 4 29435.500 7358.868 3.558 0.012Sex 1 1127.500 1127.511 0.545 0.464Age 1 82.600 82.635 0.040 0.842logPesticides 1 8049.200 8049.197 3.892 0.054
Table 10. Analysis of variance tables from linear least squares models for the best predictors of serum testosterone in adult males.Sum PBDEs df SS MS F Value Pr(F)Location/Season 4 363.514 90.879 1.157 0.377Age 1 731.734 731.734 9.317 0.010logPBDE 1 600.863 600.863 7.651 0.017
Sum HCHsLocation/Season 4 363.514 90.879 1.056 0.420Age 1 731.734 731.734 8.501 0.013logHCH 1 510.368 510.368 5.929 0.031
Fig. 1
Locations of harbour seal study populations
SID 5 (2/05) Page 19 of 24
0
50
100
150
200
250
300
350
400
450
500
Congener
Con
cent
ratio
n (n
g/g
lipid
)
tri tetra penta
hexa
heptaocta
0
2
4
6
8
10
12
14
16
18
20
Congener
Con
cent
ratio
n (n
g/g
lipid
)
tri
tetra
penta
hexa
hepta
0
20
40
60
80
100
120
Congener
Con
cent
ratio
n (n
g/g
lipid
)
tri
tetra pentahexa
hepta
Fig. 2 Congener profile of PCBs in grey seal blubber samples
Fig. 3. Congener profile of mono-ortho PCBs in grey seal blubber samplesFig. 4. Congener profiles of PBDEs in grey seal blubber samples.
SID 5 (2/05) Page 20 of 24
0
200
400
600
800
1000
1200
1400
1600
SumPCBs SumPest SumDDTs SumHCH SumPBDEs
Contaminant Group
Con
cent
ratio
n (n
g/g
lipid
)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 1 2 3 4 5 6 7
Month
Pr(
Rec
aptu
re)
Fig. 5. Median and interquartile ranges of concentrations of contaminants in grey seal pup blubber samples.
Fig. 6. Recapture probabilities and standard errors, by time (1=January) from Model 1 {phi(sex*condition),p(t)}
Fig. 7. Mass, length and girth relationships among harbour seals by location.
SID 5 (2/05) Page 21 of 24
Mass Girth Length
Leng
thG
irth
Mas
s
LocationIslay/JuraMoray FirthOrkneyTay EstuaryWash
3
3.5
4
4.5
5
5.5
Live Dead Live Dead
Tay Estuary East Coast Wash Wash
Log(
10) S
um P
CB
s ng
/g li
pid
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Live Dead Live Dead
Tay Estuary East Coast Wash Wash
Log(
10) S
um P
BD
Es
ng/g
lipi
dFig. 8. Relationship between total triiodothyronine and total thyroxine concentrations in the serum of harbour seals.
Fig. 9. Log10 (Sum PCB) and Log10 (Sum PBDE) concentrations in victims and survivors of the 2002 PDV epidemic. Horizontal line shows threshold level for immunosuppressive effects of PCBs in marine mammals as estimated by Kannan et al. (2000)
SID 5 (2/05) Page 22 of 24
0 1 2 3 4
Total triiodothyronine (nmol/l)
0
50
100
150
200
250
Tota
l thy
roxi
ne (n
mol
/l)
References
Brouwer, A., P. J. H. Reijnders, J. H. Koeman. 1989. Polychlorinated biphenyl (PCB)-contaminated fish induces vitamin A and thyroid hormone deficiency in the common seal (Phoca vitulina). Aquatic Toxicology 15, 99-106.
Byrne, J. J., J. P. Carbone, E. A. Hanson. 1987. Hypothyroidism and Abnormalities in the Kinetics of Thyroid- Hormone Metabolism in Rats Treated Chronically with Polychlorinated Biphenyl and Polybrominated Biphenyl. Endocrinology 121, 520-527.
De Swart, R. L., P. S. Ross, J. G. Vos, A. D. M. E. Osterhaus. 1996. Impaired immunity in harbour seals (Phoca vitulina) exposed to bioaccumulated environmental contaminants: Review of a long-term feeding study. Environmental Health Perspectives 104, 823-828.
DeSwart, R. L., P. S. Ross, L. J. Vedder, H. H. Timmerman, S. Heisterkamp, H. Van Loveren, J. G. Vos, P. J. H. Reijnders, A. D. M. E. Osterhaus. 1994. Impairment of immune function in harbor seals (Phoca vitulina) feeding on fish from polluted waters. Ambio 23, 155-159.
Gardiner, K. J., I. L. Boyed, B. K. Follett, P. A. Racey, P. J. H. Reijnders. 1999. Changes in pituitary, ovarian and testicular activity in harbour seals (Phoca vitulina) in relation to season and sexual maturity. Canadian Journal of Zoology 77, 211-221.
Gardiner, K. U., I. L. Boyd, P. A. Racey, P. J. H. Reijnders, P. M. Thompson. 1996. Plasma progesterone concentrations measured using an enzyme-linked immunosorbent assay useful for diagnosing pregnancy in harbor seals (Phoca vitulina). Marine Mammal Science 12, 265-273.
Hall, A., O. Kalantzi, G. Thomas. 2003. Polybrominated diphenyl ethers (PBDEs) in grey seal pups during their first year of life - are they thyroid hormone endocrine disrupters? Environmental Pollution 126, 29-37.
Hall, A., B. McConnell, R. Barker. 2001. Factors affecting first-year survival in grey seals and their implications for life history strategy. Journal of Animal Ecology 70, 138-149.
Hall, A., P. Pomeroy, N. Green, K. Jones, J. Harwood. 1997. Infection, haematology and biochemistry in grey seal pups exposed to chlorinated biphenyls. Marine Environmental Research 43, 81-98.
Hall, A. J., N. Green, K. Jones, P. Pomeroy. 1998. Thyroid hormones as biomarkers in grey seals. Marine Pollution Bulletin 36, 424-428.
Hall, A. J., R. J. Law, D. E. Wells, J. Harwood, H. M. Ross, S. Kennedy, C. R. Allchin, L. A. Campbell, P. P. Pomeroy. 1992. Organochlorine levels in common seals (Phoca vitulina) that were victims and survivors of the 1988 phocine distemper epizootic. Science of the Total Environment 115, 145-162.
Hall, A. J., B. J. McConnell, R. J. Barker. 2002. The effect of total immunoglobulin levels, mass and condition on the first-year survival of grey seal pups. Functional Ecology 16, 462-474.
Hammond, J. A., A. J. Hall, L. Dyrynda. In Press. Polychlorinated biphenyls (PCBs) suppress innate immune function of harbour but not grey seals in vitro. Aquatic Toxicology.
Härkönen, T., R. Dietz, P. J. H. Reijnders, J. Teilmann, P. M. Thompson, K. Harding, A. J. Hall, S. M. J. M. Brasseur, U. Siebert, S. Goodman, P. Jepson, T. Dau Rasmussen. Submitted. Review of the seal epizootics in Europe. Diseases in Aquatic Organisms.
Härkönen, T. and M.P. Heide-Jørgensen, 1990. Comparative life histories of East Atlantic and other harbor seal populations. Ophelia 32, 211-235.
Kannan, K., A. L. Blankenship, P. D. Jones, J. P. Giesy. 2000. Toxicity reference values for the toxic effects of polychlorinated biphenyls to aquatic mammals. Human and Ecological Risk Assessment 6, 181-201.
Kendall, M. D., B. Safieh, J. Harwood, P. Pomeroy. 1992. Plasma thymulin concentrations, the thymus and organochlorine contaminant levels in seals infected with phocine distemper virus. Science of the Total Environment 115, 133-144.
Keum, Y., Q. Li. 2005. Reductive debromination of polybrominated diphenyl ethers by zerovalent iron. Environmental Science and Technology 39, 2280-2286.
Kuriyama, S., A. Fidalgo-Neto, W. Mathar, R. Palavinskas, K. Friedrich, I. Chahoud. 2003. Effect of low dose mono-ortho 2,3’,4,4’,5 pentachlorobiphenyl on thyroid hormone status and EROD activity in rat offspring: consequences for risk assessment. Toxicology 186, 11-20.
McConnell, B. J., R. Beaton, E. Bryant, C. Hunter, P. Lovell, A. J. Hall. 2004. Phoning home - a new GSM mobile phone telemetry system to collect mark-recapture data. Marine Mammal Science 20, 274-283.
Ness, D., S. Schantz, J. Moshtaghian, L. Hansen. 1993. Effects of perinatal exposure to specific PCB congeners on thyroid hormone concentrations and thyroid histology. Toxicological Letters 68, 311-323.
Oskam, I., E. Ropstad, l. E. Dah, E. Lie, A. Derocher, O. Wiig, S. Larsen, R. Wiger, J. Skaare. 2003. Organochlorines affect the major androgenic hormone, testosterone, in male polar bears (Ursus maritimus) at Svalbard. J Toxicol Eviron Health A 28, 2119-2139.
Pomeroy, P. P., J. A. Hammond, A. J. Hall, M. Lonergan, C. D. Duck, V. J. Smith, H. Thompson. 2005. Morbillivirus neutralizing antibodies in Scottish grey seals: assessing the effects of the 1988 and 2002 PDV epizootics. Marine Ecology Progress Series 287, 241-250.
Reijnders, P. J. H. 1986. Reproductive failure in common seals feeding on fish from polluted coastal waters. Nature, London 324, 456-457.
Ross, P. S., R. L. De Swart, H. H. Timmerman, P. J. H. Reijnders, J. G. Vos, H. Van Loveren, A. D. M. E. Osterhaus. 1996. Suppression of natural killer cell activity in harbour seals (Phoca vitulina) fed Baltic Sea herring. Aquatic Toxicology 34, 71-84.
SID 5 (2/05) Page 23 of 24
Ross, P. S., B. Pohajdak, W. D. Bowen, R. F. Addison. 1993. Immune function in free-ranging hrbour seal (Phoca vitulina) mothers and their pups during lactation. Journal of Wildlife Diseases 29, 21-29.
Subramanian, A., S. Tanabe, R. Tatsukawa, S. Saito, N. Miyazaki. 1987. Reduction in the testosterone levels by PCBs and DDE in Dall's porpoises of northwestern North Pacific. Marine Pollution Bulletin 18, 643-646.
Sabin, R. C., D. J. Chimonides, C.J.H. Spurrier, P.D. Jepson, R. Deaville, R.J. Reid, A. P. Patterson, R. Penrose and R. Law. 2004. Trends in cetacean strandings around the UK coastline and cetacean and marine turtle post-mortem investigations for the year 2003. Report to DEFRA, No. ECM 516F/04, pp57.
Sørmo, E.G., I. Jüssi, M. Jüssi, M. Braathen, J. U. Skaare, and B. M. Jenssen. 2005. Thyroid hormone status in grey seal (Halichoerus grypus) pups from the Baltic sea and the Atlantic ocean in relation to organochlorine pollutants. Environ. Toxicol. Chem. 24, 610-616.
Thomas, G.O., Moss, S.E.W., Asplund, L. and Hall, A.J. 2004. Absorption of decabromodiphenyl ether and other organohalogen chemicals by grey seals Environmental Pollution 133, 581-586
Thompson, P., P. Rothery. 1987. Age and sex differences in the timing of moult in the common seal. Journal of Zoology, London 212, 597-603.
Thorson, P., B. Le Boeuf. 1994. Developmental aspects of diving in Northern elephant seal pups. Pages 271-289 in B. Le Boeuf,R. Laws, editors. Elephant Seals. University of California Press, Berkeley and Los Angeles.
White, G. C., K. P. Burnham. 1999. Program MARK: survival estimation from populations of marked animals. Bird Study 46, 120-139.
References to published material9. This section should be used to record links (hypertext links where possible) or references to other
published material generated by, or relating to this project.
SID 5 (2/05) Page 24 of 24
