Archives of Environmental Contamination and Toxicology

, Volume 53, Issue 2, pp 293–302

A Comparative Analysis of Polybrominated Diphenyl Ethers and Polychlorinated Biphenyls in Southern Sea Otters that Died of Infectious Diseases and Noninfectious Causes

Authors

    • Wadsworth Center, New York State Department of Health and Department of Environmental Health Sciences, School of Public HealthState University of New York at Albany
  • Emily Perrotta
    • Wadsworth Center, New York State Department of Health and Department of Environmental Health Sciences, School of Public HealthState University of New York at Albany
  • Nancy J. Thomas
    • US Geological Survey-Biological Resources DivisionNational Wildlife Health Center
  • Kenneth M. Aldous
    • Wadsworth Center, New York State Department of Health and Department of Environmental Health Sciences, School of Public HealthState University of New York at Albany
Article

DOI: 10.1007/s00244-006-0251-8

Cite this article as:
Kannan, K., Perrotta, E., Thomas, N.J. et al. Arch Environ Contam Toxicol (2007) 53: 293. doi:10.1007/s00244-006-0251-8

Abstract

Southern sea otters (Enhydra lutris nereis) from the California coast continue to exhibit a slower population regrowth rate than the population in Alaska. Infectious diseases have been identified as a frequent cause of death. Infectious diseases caused by varied pathogens including bacteria, fungi, and parasites were suggestive of compromised immunological health of mature animals in this population. To test the hypothesis that elevated exposure to immunotoxic contaminants such as polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) contribute to disease susceptibility via immunosuppression, we determined concentrations of PBDEs and PCBs in livers of 80 adult female sea otters that died of infectious diseases, noninfectious causes, or emaciation. Concentrations of PBDEs and PCBs in sea otter livers varied widely (10–26,800 ng/g and 81–210,000 ng/g, lipid weight, respectively). Concentrations of PBDEs in sea otters were some of the highest values reported for marine mammals so far. Although PCB concentrations in sea otters have declined during 1992–2002, the mean concentration was at the threshold at which adverse health effects are elicited. Concentrations of PBDEs and PCBs were significantly correlated, suggesting co-exposure of these contaminants in sea otters. No significant association was found between the concentrations of PBDEs and the health status of sea otters. Concentrations of PCBs were significantly higher in otters in the infectious disease category than in the noninfectious category, suggesting an association between elevated PCB concentrations and infectious diseases in Southern sea otters.

The sea otter, once an abundant species with a range spanning from California in the east across the Pacific Rim to the Kuril Islands north of Japan in the west, was hunted to near-extinction in the 18th and 19th centuries for its fur (Estes 1990; Bacon et al. 1999). It was not until 1911 that the surviving population was legally protected by international law. The result was several remnant sea otter populations. Long-term studies exist for the sea otter subspecies of Attu Island of the Aleutian Archipelago, southeast Alaska, British Columbia, Washington, and California (Estes 1990; Bacon et al. 1999). Best estimates placed the California sea otter populations at 50 individuals in 1914. Prior to their exploitation, it was estimated that 16,000–18,000 sea otters inhabited the California coastline (DeMaster et al.1996; CDFG 2001). The California subspecies has exhibited a sluggish ∼5% per year population increase, whereas other sea otter populations have exhibited growth rates near species expansion limits, at 17–20% per year (Estes 1990). Additionally, the California sea otter has had two periods of population decline: one occurred during 1976–1984 and was attributed to mortalities from entanglement in near-shore fishing nets; the second took place from 1995 to 1999 and the cause remains unexplained (DeMaster et al.1996; CDFG 2001).

The California sea otter population has exhibited atypical mortality patterns. Approximately 40% of beached otters have died of infectious diseases, and mortality occurs throughout the year, with a seasonal peak in the spring through summer months (Thomas and Cole 1996; Estes et al. 2003). Additionally, the highest rate of mortality by age class was found among prime-aged adult otters (3–10 years), rather than among aged adults or pups (Otter Project 2005). It is believed that a high rate of infectious disease-related mortality, rather than low fecundity, was responsible for the stagnating population growth. Diseases caused by parasites, bacteria, or fungi, and diseases without a specified etiology, have been identified as the primary causes of death (Thomas and Cole 1996; Kreuder et al.2003). These findings suggest that the immunological health of mature animals in this population is compromised. In attempts to better understand the factors that contribute to high mortality and reduced immunocompetence in Southern sea otters, studies have examined residue levels of immunosuppressive chemical contaminants (Kannan et al. 1998; Nakata et al. 1998; Kannan et al. 2004a; Kannan et al. 2006a, 2006b), immunological parameters (Schwartz et al. 2005), and serology and clinical pathology (Hanni et al. 2003). However, because the contaminants exist in complex mixtures, and because interactions can potentially occur among contaminants, nutritional status, and other environmental factors, establishment of a link between contaminant exposure and disease mortality in marine mammals such as sea otters is a challenging task. Nevertheless, population-based epidemiologic studies to determine the risk of infection after exposure to potential contributing agents, such as polychlorinated biphenyls (PCBs) may help to discentangle this problem. (Hall et al.2006). Such case–control studies have been widely used in human and veterinary medicine (Shapiro 1989) but have received little attention in wildlife research, because of the logistical difficulties in obtaining disease and exposure data, and also because of the lack of necessary tools to investigate and understand the mechanisms of host disease resistance. A few earlier studies have compared residue levels of PCBs and butyltins between healthy and diseased sea otters (Kannan et al.1998; Nakata et al.1998) and other marine mammal populations (Kannan et al. 1993; Hall et al.2006), to examine whether high exposure levels were associated with the disease or mortality. Although the case–control studies still do not provide causal linkages, they can provide baseline information that allows us to develop a framework for future systematic investigations (Hall et al.2006). In this study, concentrations of polybrominated diphenyl ethers (PBDEs) and PCBs were measured in livers of sea otters, to compare exposure levels between individuals that died of infectious diseases and those that died of noninfectious causes.

Since the early 1990s, the putative link between exposure to PCBs and immunosuppression in marine mammals has received considerable attention (Lahvis et al.1995; De Swart et al. 1996; Jepson et al. 2005; Beineke et al. 2005). In addition to PCBs, several other halogenated contaminants such as DDTs, PBDEs, and dioxins can contribute to immunosuppression in marine mammals (Lahvis et al.1995; De Swart et al. 1996; Jepson et al. 2005; Beineke et al. 2005). Thymic atrophy and splenic depletion were significantly correlated with increased PCB and PBDE levels in harbor porpoises from the North Sea and the Baltic Sea (Beineke et al. 2005). PBDEs are flame retardants and are used in many household consumer products ranging from television sets and computers to sofas, blankets, and building materials (Hites 2004). Studies have shown that the concentrations of PBDEs in marine mammals have increased significantly over the last decade (She et al. 2002; Ikonomou et al. 2002; Kajiwara et al. 2004; Johnson-Restrepo et al. 2005a).

Although there are many known factors such as boat strike, entrapmentin fishing nets, human disturbances, and malnutrition that can contribute to the slow population growth rate of California sea otters, it remains unknown whether environmental contaminants contribute to the slowness of the recovery (Estes et al. 2003). In this study, we examined the association between exposure to PCBs and PBDEs and the disease status of the California sea otter. This was accomplished by determining and comparing hepatic concentrations of PCBs and PBDEs in sea otters that died of infectious diseases and those that died of noninfectious causes. The availability of large numbers of archived tissues of sea otters, which had been systematically examined for evidence of disease during postmortem investigations and necropsy, provided an opportunity to evaluate the association between PCB/PBDE concentrations and cause of mortality. Sea otters are unique and valuable sentinels that allow us to evaluate the health of coastal aquatic environments. The narrow home range, top-level predator status, and voracious appetite make the sea otter a suitable surrogate for human exposure to aquatic sources of environmental contaminants (Kajiwara et al. 2001; Kannan et al. 2004a).

Materials and Methods

Samples

A subset of samples from adult female sea otters (n = 80) were selected from an archived sample set of more than 300 beached Southern sea otters found freshly dead or dying between 1992 and 2002, along the central California coast (Figure 1). The otters were found terminally ill or dead and carcasses were in good postmortem condition. They were shipped refrigerated by overnight carrier, and necropsies were performed within a day of arrival. Sample selection was based on gender and age, to eliminate these variables as confounding factors. Additionally, female sea otters were chosen, because their more localized movement patterns make them more suitable indicators of local sources of pollution. Postmortem examinations were performed at the USGS National Wildlife Health Center (NWHC) in Madison, Wisconsin for the determination of cause of death (COD). A variety of tissues from sea otters were fixed in 10% buffered formalin for histopathology, then paraffin embedded, stained by hematoxylin and eosin, and examined by light microscopy. Selection of other diagnostic clinical laboratory tests was based on the history and gross lesions and included microbiologic, virologic, and parasitologic procedures. The COD was classified, based on necropsy findings, into one of four categories: emaciation, infectious disease, other, and trauma. Each class was further divided into more specific subclasses. In this study, we grouped animals that died of infectious diseases into an “infectious” group (n = 26), and animals that died of trauma and other causes into a “noninfectious” group (n = 27). Otters that died in emaciated condition and had no evidence of other causes of death were grouped into a separate “emaciation” category (n = 27). These animals may have died of emaciation due to inadequate food intake, but, alternatively, other debilitating conditions not apparent by postmortem may have been responsible for emaciation. Some subclasses of animals in this category had conditions that might have contributed to starvation, such as high demands of reproduction and dental disease, but these conditions alone were not expected to induce fatal debility. The nutritive/body condition was assessed by the amount of adipose tissue and muscle, and was scored from 1 to 5:1 for an animal in severely emaciated condition (NC1), 2 for poorly nourished (NC2), 3 for adequately nourished (NC3), 4 for well nourished (NC4), and 5 for an animal in excellent condition (NC5). Otters in the “infectious disease” category include those animals that died of acanthocephalan peritonitis (n = 3); protozoal encephalitis (n = 2); and fatal infections by bacteria (n = 13), fungi (n = 3), or parasites (n = 1). Also included in this category were other fatal cardiovascular (n = 2) and neurological (n = 2) infections. The category “other” comprised animals that died of gastrointestinal disorders (n = 5), miscellaneous individual problems (n = 4; blindness, urinary obstruction, twin fetuses causing a problem during birth), neoplasia (n = 3), and undetermined causes (n = 9). The category “trauma” included otters that died of shark bite (n = 3) and blunt traumatic injuries of uncertain cause (n = 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f1.jpg
Fig. 1.

Map of California showing sampling locations of sea otters

Liver tissue, rather than adipose tissue, was chosen for analysis for the following reasons. First, this organ plays a significant role in toxicant processing. Second, the sea otter lacks blubber. Although PCBs and PBDEs are lipophilic by nature, high concentrations in fat tissue may not indicate direct toxicity associated with exposure. It is thought that the toxic effects are pronounced when contaminants are mobilized from the lipid pool and distributed to vital organs such as liver, heart, and kidney.

Analysis of PCBs and PBDEs

Liver samples were collected from the carcasses at the time of necropsy, wrapped in aluminum foil, enclosed in sterile sampling bags (TWIRL’EM, Fisher Scientific Inc., Hampton, NH), and stored at –20°C until analysis. Liver samples (∼10 g) were homogenized with anhydrous sodium sulfate (120 g), spiked with the internal standards PCB-30 and PCB-204 (25 ng), and extracted with mixed solvents dichloromethane (300 mL) and hexane (100 mL) using a Soxhlet apparatus for 20 h (Kannan et al. 2004a). The solvent was concentrated (11 mL), and an aliquot (1 mL) was taken for gravimetric determination of fat content. A second aliquot (5 mL) was spiked with 13C-labeled PCBs (10 ng each congener; MBP-MXF and MBP-CG; Wellington Laboratories, Guelph, Ontario, Canada) and 13C-labeled PBDEs (10 ng each congener; EO5100; Cambridge Isotope Laboratories, Andover, MA) and subjected to column chromatography for removal of lipids. Columns (10 mm i.d.) were prepared by layering silica gel (2 g, 100–200 mesh), 40% acidic silica gel (1 g), silica gel (1 g), 40% acidic silica gel (1 g), and silica gel (1 g). Sample extracts were eluted with 15% dichloromethane in hexane (150 mL). Samples were further subjected to lipid removal by treatment with concentrated sulfuric acid, if and when needed. Sample extracts (2 μL) were injected into a Hewlett-Packard 6890 gas chromatograph interfaced with a Hewlett-Packard 5973 mass spectrometer (GC-MS). Injections were made in the splitless mode, and samples were separated on a 30 m DB-5 (5% diphenyl/dimethylpolysiloxane) analytical capillary column with a 250 μm i.d. and a 0.25-μm film thickness. The oven temperature program was set to 100°C for 1 min, 10°C/min to 160°C, hold 3 min, and 2.5°C/min to 260°C, hold for 10 min. The inlet and interface temperatures were set to 270°C and 300°C, respectively. The MS was operated in an electron impact mode (70 eV) and selected ion monitoring mode (SIM). PBDE congeners were identified and quantified by SIM at m/z 406, 408; 486, 484; 564, 566; and 642, 644 for tri-; tetra-; penta-; and hexa-BDEs, respectively. Quantification was based on external calibration standard. PCB congeners were monitored at the two most intense ions of the molecular ion cluster. An equivalent mixture of Kanechlor (KC300, 400, 500, and 600), with known PCB composition, was used in the identification of PCB congeners. Quantification of PCB congeners was based on external calibration standards containing known concentration of di- through deca-CB congeners. Concentrations from individually resolved peaks of PCB isomers were summed, to obtain total PCB concentrations.

Quality Assurance and Quality Control

In order to assure the quality of our measurements, we have taken several steps. Extracts used for PBDE analysis were also analyzed for PCBs. Therefore, samples were spiked with internal standards PCB-30, PCB-204, and 13C-labeled PCBs and 13C-PBDEs, one from each homologue group. The recoveries of PCB-30 and PCB-204, spiked into samples prior to extraction, were 86 ± 15% and 91 ± 19%, respectively. Recovery of the 13C-labeled PCBs and 13C-PBDEs, spiked prior to lipid removal, ranged from 86% to 99%. The measured concentrations were not corrected for the recoveries of internal or surrogate standards. The limit of quantification (LOQ) was determined by several measures. The LOQ was considered to be two times greater than the signal determined in blank samples. If no signal was observed in the blank, the lowest point on the calibration curve that was within 30% of the back-calculated 1/x weighted calibration curve was used. Dilution factors, sample weight, and final extract volume were taken into account for calculating LOQ. A LOQ of 1 ng/g, wet weight, for total PBDEs and total PCBs was determined. Procedural blanks were analyzed simultaneously with every batch of five samples as a check for interferences or contamination arising from solvents and glassware. Total PBDEs represent the sum of BDE congeners 47, 99, 100, 153, and 154. Total PCBs represent the sum of all of the di- through decachlorobiphenyl congeners. Concentrations are reported on a lipid weight (wt) basis, unless reported otherwise.

Statistical Analysis

All statistical analyses were performed using Statgraphics version 5 (Manugistics, Inc., Rockville, MD). The distribution of PBDE and PCB concentrations was skewed. Therefore, nonparametric tests were used when comparisons were made between mortality groups. The effects of location and biological parameters (e.g., nutritive status) on the concentrations of PBDEs and PCBs were tested using one-way ANOVA and the Kruskall-Wallis test or Fisher’s least significant difference test to determine whether median concentrations varied among sea otters by location or nutritive condition or through time. The level of significance used for all statistical tests was α ≤ 0.05.

Results and Discussion

PBDEs

Concentrations of PBDEs in sea otter livers ranged from 10 to 26,800 ng/g, lipid wt (mean ± SD: 2200 ± 3700; median: 974) (Table 1). The variation of three orders of magnitude in PBDE concentrations suggests remarkable differences in exposure levels among individual sea otters. The highest concentration of 26,800 ng/g, lipid wt, was found in an otter that died of bacterial infection in 1995. This value qualified as an outlier (Grubbs’ test, p < 0.05), but the repeated analysis of the sample yielded similar results.
Table 1.

Concentrations (ng/g) of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) in livers of adult female sea otters from the California coast (1992–2002) stratified by mortality categories

 

Fat (%)

PBDEs (lipid wt)

PBDEs (wet wt)

PCBs (lipid wt)

PCBs (wet wt)

Noninfectious causes (n = 27)

  Mean ± SD

4.7 ± 2.7

2120 ± 2630

88.5 ± 107

25,600 ± 48200

970 ± 1520

  Min-max

1.1–11

20–10400

0.7–449

81–210000

3.0–5960

  Median

4.0

1070

24.6

3090

114

Infectious disease (n = 26)

  Mean ± SD

4.0 ± 1.6

2980 ± 5500

113 ± 191

15500 ± 32200

558 ± 1040

  Min-max

1.4–7.8

12–26800

0.2–862

310–143000

4.4–4440

  Median

3.7

1020

40.1

4390

195

Emaciation (n = 27)

  Mean ± SD

3.4 ± 1.5

1430 ± 1860

48.4 ± 79.2

10300 ± 23600

340 ± 940

  Min-max

1.2–8.3

10–8810

0.5–363

271–123000

12.2–5060

  Median

3.3

831

23.7

2763

69.4

Overall (n = 80)

  Mean ± SD

4.0 ± 2.1

2170 ± 3710

82.8 ± 136

17100 ± 36700

624 ± 1220

  Min-max

1.1–11

10–26800

0.2–862

81–210000

3.0–5960

  Median

3.7

974

27.2

3510

131

The mean concentration of PBDEs in the livers of sea otters in this study was higher than the mean values reported for blubber from harbor seals from central California (She et al. 2002) and in blubber from bottlenose dolphins from the Florida coast (Johnson-Restrepo et al. 2005a) (Figure 2). However, it was similar to the concentration reported for California sea lions stranded between 1993 and 2003 (570–24,240 ng/g, lipid wt; Stapleton et al. 2006). Overall, the concentrations of PBDEs found in sea otter were comparable to those reported for marine mammals from the United Kingdom and the Mediterranean Sea, but they were two to three orders of magnitude higher than those reported for marine mammals from Asian coastal areas and the Arctic Ocean (Figure 2) (Ikonomou et al. 2002; Lebeuf et al. 2004; Kajiwara et al. 2004; Pattersson et al. 2004; Kannan et al. 2005a, 2005b; Law et al. 2002, 2005; Ramu et al. 2005; Stapleton et al. 2006; Kajiwara et al. 2006; Muir et al.2006).
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f2.jpg
Fig. 2.

Mean (±SD) concentrations of polybrominated diphenyl ethers in livers of sea otters compared with those in the blubber of marine mammals from other selected locations. Sea otter, harbor seal, and California sea lion are from California (CA) coast (this study, She et al. 2002; Stapleton et al. 2006). Bottlenose dolphin from Florida (FL) coast (Johnson-Restrepo et al. 2005a). Data for the UK marine mammals are from Law et al. (2002, 2005). Data for the Mediterranean Sea (MES) are from Pattersson et al. (2004); St. Lawrence Estuary (SLE) beluga whales from Lebeuf et al. (2004); Alaska (AK) polar bears from Kannan et al. (2005a); Svalbard (SVAL) polar bears from Muir et al. (2006); Canadian Arctic (ARC CAN) ringed seals from Ikonomou et al. (2002); Japanese (JPN) northern fur seal from Kajiwara et al. (2004); India’s humpback (HB) and spinner dolphin and Philippines (PHI) spinner dolphin and Japanese marine mammals from Kajiwara et al. (2006); India’s Irrawaddy dolphin from Kannan et al. (2005b); Hong Kong (HK) finless porpoise and humpback dolphin from Ramu et al. (2005)

High concentrations of PBDEs are known to occur in both inland and marine coastal regions of the United States. A study of the San Francisco Estuary, at the northern end of the California sea otters’ range, showed PBDE concentrations in water ranging from 3 to 513 pg/L; concentrations in sediment ranging from below detection limit to 212 ng/g, dry wt; and concentrations in clams (Corbicula fluminea) at 106 ng/g, dry wt (Oros et al. 2005). Water and sediment contained high proportions of octa- and deca- BDEs, whereas clams contained BDE-47 as the dominant congener. We did not analyze octa- or deca-BDE in sea otter tissues. Nevertheless, most studies have reported BDE-47, BDE-99, and BDE-100 as the most abundant congeners in biological tissues (Stapleton et al.2006). Sewage treatment plant effluents were suggested as a source of PBDEs for the San Francisco Estuary; an estimated 22.7 kg of PBDEs was discharged into this estuary every year (Oros et al. 2005). Mean concentration of PBDEs in edible portion of fish from San Francisco Bay in 2000 was 302 ng/g, lipid wt (Brown et al. 2006), approximately sevenfold lower than the level that we found in the sea otters.

BDE-47 was the dominant congener in livers of all but one sea otter analyzed in our study, and it accounted for 37% of the total PBDEs, followed by BDE-99 (28%) and BDE-100 (16%) (Figure 3). This pattern is similar to that reported for other marine mammals, except that sea otter livers contained a relatively higher proportion of BDE-99 than BDE-100 (Hites 2004). The distribution of PBDE congeners among the categories of emaciated, noninfectious COD, and infectious COD was similar (BDE-47 > BDE-99 > BDE-100 > BDE-153 > BDE-154), except that otters classified as infectious COD contained higher proportions of BDE-99 (32%) than did the noninfectious-COD otters (25%). Furthermore, in the noninfectious-COD category, BDE-47 was slightly higher (40%) than that in the infectious-COD (33%) and emaciated categories (38%) (Figure 3). Indeed, the otter that had the highest concentration of total PBDEs contained BDE-99 as the most abundant congener, followed by BDE-47. In clams (a prey item of sea otters) collected from the San Francisco Estuary, concentrations of BDE-47 were higher than those of BDE-99 (Oros et al. 2005). The elevated proportion of BDE-99 in the otter that had the highest concentration of total PBDEs suggests the inability of this diseased otter to metabolize BDE-99. In general, BDE-99 is the predominant congener in the penta-BDE commercial mixture, followed by BDE-47 and BDE-100 (Stapleton et al.2006).
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f3.jpg
Fig. 3.

Percent contributions of various polybrominated diphenyl ether (PBDE) congeners to total PBDE concentrations in the livers of sea otters that died of noninfectious causes, infectious diseases, and emaciation

PCBs

Concentrations of PCBs in livers of sea otters ranged from 81 to 210,000 ng/g, lipid wt (mean ± SD: 17,100 ± 36,700; median: 3510) (Table 1). The highest concentration was found in an otter that died of undetermined causes in 1995. The next highest concentration, 143,000 ng/g, lipid wt, was found in an otter beached in 1994, and the COD of this otter was found to be an infectious neurological disease. An earlier study reported PCB concentrations as high as 266,000 ng/g, lipid wt, in livers of adult sea otters (Kannan et al.2004a), similar to that found in the present study.

The mean concentration of PCBs in livers of our sea otters was 2.5-fold lower than the concentration reported for California sea otters collected during 1992–1996 (Nakata et al. 1998). This suggests a reduction in PCB concentrations in sea otters in recent years. Similarly, concentrations of PCBs in California sea lions showed a trend of decline in the 1990s (Kannan et al.2004b). Despite the declining trend, concentrations of PCBs in sea otters and California sea lions were some of the highest concentrations (410,000 ng/g, lipid wt) found for marine mammals in recent years (Kajiwara et al. 2001; Kannan et al. 2004b). The mean concentration of PCBs in sea otters (17,100 ng/g, lipid wt) was at the threshold concentration (17,000 ng/g, lipid wt) that has been reported to elicit adverse physiological effects in marine mammals (Kannan et al.2000). A threshold PCB concentration of 45,000 ng/g, lipid wt, was reported for an increased risk of infection in harbor porpoises (Hall et al. 2006). A PCB concentration of 10,000–22,000 ng/g, lipid wt, was associated with suspected immunotoxic effects in harbor porpoises (Beineke et al. 2005). More than 25% of the sea otters analyzed in the present study contained PCB concentrations greater than 10,000 ng/g, lipid wt. These results suggest that individual sea otters may experience adverse effects due to exposure to high levels of PCBs.

The distribution of PCB homologues in sea otter livers followed the order of hexa-CB > hepta-CB > penta-CB > octa-CB > nona-CB > tetra-CB > tri-CB/di-CB (Figure 4). The relative proportion of each homologue group for sea otters classified as infectious COD, emaciated, and noninfectious COD was similar. Hexa-CB congeners CB-153 and CB-138 were the most abundant in the sea otter livers, accounting for 20% and 14%, respectively, of the total PCB concentrations (Figure 4). Other major CB congeners include CB-187, CB-180, and CB-118, respectively accounting for 8.3%, 7.2%, and 5% of the total PCB concentrations. The toxic mono-ortho congeners, CB-105, CB-118, and CB-156, were present at 1.8%, 5.0%, and 2.7%, respectively, of the total PCB concentrations.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f4.jpg
Fig. 4.

Relative distribution of polychlorinated biphenyls homologues (top) and congeners (bottom) in livers of sea otters. Concentrations of individual congeners were normalized to that of CB-153, which is set to 100. Congeners comprising <1% of the CB-153 were not included

Comparison of Concentrations by Location, Nutritive Condition, and Through Time

The effect of stranding location on the concentrations of PBDEs and PCBs was tested with one-way ANOVA. For this analysis, the sea otters’ present-day range was divided into three segments: north of Seaside, Seaside-Cayucos, and south of Cayucos (Figure 5). The three segments were chosen based on findings of past studies (Estes et al.2003), natural divisions of the developed coastline, and availability of data for each segment. Because of the clustering of female sea otters in the center of the range and the dominance of males in the outer regions, lower numbers of female samples exist for the north of Seaside and south of Cayucos segments than for the central segment. No significant difference was found for PBDE concentrations among the three segments. Nevertheless, PCB concentrations were significantly higher (p < 0.05; ANOVA and Mann-Whitney W test) in the central segment (Seaside-Cayucos) than in the south-of-Cayucos segment. PCB concentrations were not significantly different between the north of Seaside and the central segments (p > 0.05) (Figure 5). These results suggest widespread contamination by PBDEs and PCBs along the central California coast.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f5.jpg
Fig. 5.

Box-and-whisker plots of polybrominated diphenyl ethers (PBDE) and polychlorinated biphenyl (PCB) concentrations in sea otters stratified by location. White line is the median and white circle is the mean; lower and upper limits of the box represent 25th and 75th percentiles; the whiskers extend to the last observation within 1.5 times the interquartile range. The patterned circles are far outside points (outliers). Categories represented by the two different letters (a,b) suggest significant difference (p < 0.05)

The influence of nutritive condition of sea otters at the time of death on the concentrations of PBDEs and PCBs was examined. Only one sea otter was in excellent nutritive condition (NC5), and therefore this individual was excluded from the subsequent comparison. Concentrations of PBDEs differed significantly among the remaining four groups. Concentrations of PBDEs were significantly lower in sea otters that were well-nourished than in poorly nourished or emaciated individuals (Figure 6). Similarly, concentrations of PCBs in severely emaciated individuals (NC1 and NC2) were significantly higher than those in well-nourished individuals (NC4) (Mann-Whitney W test; p < 0.05). The differences in contaminant status among various nutritional groups suggest the influence of lipid mobilization as a consequence of starvation and/or disease. In particular, elevated concentrations of PCBs and PBDEs in emaciated sea otters suggest sequelae of lipolysis. Pre-existing disease in harbor porpoises was suggested to increase the mobilization of fat from body tissues, resulting in emaciation (Beineke et al. 2005). Emaciation can also induce immunosuppression due to increased secretion of adrenal glucocorticoids and reduced thyroid hormone levels (Wyllie 1980). Because emaciation appears to confound the comparison of PBDE and PCB concentrations with health status, we placed emaciated otters into a separate category for the comparative analysis.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f6.jpg
Fig. 6.

Box-and-whisker plots of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyl (PCBs) concentrations in sea otters stratified by nutritive condition. NC1 emaciated, NC2 poor, NC3 fair, NC4 good. White line is the median and white circle is the mean; lower and upper limits of the box represent 25th and 75th percentiles; the whiskers extend to the last observation within 1.5 times the interquartile range. The patterned circles are far outside points (outliers). Categories represented by the two different letters (a,b) suggest significant difference (p < 0.05)

Because the adult female sea otter samples analyzed in this study had been collected from 1992 to 2002, temporal trends in the concentrations of PBDEs and PCBs could be examined for that decade (Figure 7). Concentrations of PCBs showed a trend of marginal decline (p = 0.07; ANOVA) between 1992 and 2002. Concentrations of PCBs in sea otters collected in 1993 and 1995 were significantly higher than in those collected in 2002 (Mann-Whitney W test; p < 0.05). However, no significant difference was found for concentrations measured for other sampling years. Concentrations of PBDEs did not exhibit a trend of increase during 1992–2002 (Figure 7). Concentrations of PBDEs determined in 1995 were higher than those in 2002 (Mann-Whitney W test; p < 0.05). Despite the existence of several studies describing an exponential increase in PBDE concentrations in marine mammals (Ikonomou et al.2002; She et al. 2002; Kajiwara et al. 2004; Johnson-Restrepo et al. 2005a), sea otters collected during 1992–2002 did not show a temporal increase (i.e., an increase over that decade) in PBDE concentrations. This lack of increase is similar to the pattern found for California sea lions (Stapleton et al. 2006). It is probable that PBDE concentrations reached their peak in the late 1990s. It should be noted that our findings have limitations, because of the lack of uniformity in the number of samples analyzed per year (n = varied from 2 to 19 per year).
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f7.jpg
Fig. 7.

Temporal trends in polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyl (PCBs) concentrations in livers of sea otters from the California coast from 1992 to 2002

Despite the lack of increase in total PBDE concentrations during 1992–2002, the ratios of concentrations of PBDEs to PCBs increased significantly during this decade (p < 0.05) (Figure 8A). This suggests that the rate of increase of PBDE concentrations exceeded that for PCBs in sea otters. Recent studies have reported the predominance of PBDEs over PCBs in human tissues (Johnson-Restrepo et al. 2005b). Although restrictions have been placed on the production and use of several commercial PBDE mixtures, some of these mixtures are still in use in various household and commercial products. Because the marine ecosystem is the ultimate sink of persistent halogenated contaminants (Loganathan and Kannan 1994), there may be a lag period before the ultimate concentrations are detected in coastal and pelagic organisms. Therefore, further studies are needed to monitor the trends of PBDEs in marine mammals, including sea otters. Concentrations of PCBs were significantly correlated with concentrations of PBDEs in sea otters (Figure 8B). This suggests co-exposure to the two groups of contaminants and has implications for the health of otters, because both groups of contaminants are immunotoxic, either by direct or indirect mechanisms of toxic action (Fowles et al.1994; Law et al.2005).
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f8.jpg
Fig. 8.

Relationship between (A) polybrominated diphenyl ether/polychlorinated biphenyl (PBDE/PCB) concentration ratio in livers of sea otters and sampling year, and (B) relationship between PBDE and PCB concentrations (μg/g, lipid wt) in livers of sea otters

Comparison of Concentrations in the Infectious and Noninfectious Mortality Categories

Mean concentrations of PBDEs in the infectious COD, emaciation, and noninfectious COD categories were 2980, 1430, and 2120 ng/g, lipid wt, respectively. PBDE concentrations did not differ significantly among the three categories (Kruskall-Wallis test; p < 0.05; (Figure 9). Mean concentrations of total PCBs in the infectious COD, emaciation, and noninfectious COD categories were 15,500, 10,300, and 25,600 ng/g, lipid wt, respectively. Similar to the findings for PBDEs, concentrations of PCBs did not vary significantly among the three categories (Figure 9). An earlier study reported the occurrence of fourfold higher concentrations of PCBs in sea otters that died of infectious disease than in those that died of trauma (Nakata et al. 1998). Also, significant correlation was found between PCB concentrations and infectious disease mortality in harbor porpoises from the North Sea (Jepson et al.2005). A significant correlation was found between thymic atrophy/splenic depletion and total PCB and PBDE concentrations in harbor porpoises (Beineke et al. 2005). Based on a case–control study comprising diseased and nondiseased harbor porpoises, Hall and coworkers (2006) quantified the risk of infectious disease in relation to exposure to PCBs. Like PCBs, PBDEs can augment the risk of infection, because certain PBDE congeners have been shown to cause lymphoid depletion and decreased thyroid hormone levels, which are associated with bacterial infections in mice (Fowles et al. 1994). The significant correlation between PBDE and PCB concentrations in sea otters suggests that these two contaminant groups potentially could be acting concurrently to produce adverse health effects.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f9.jpg
Fig. 9.

Box and whisker plots of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyl (PCBs) concentrations in livers of sea otters categorized in infectious (n = 26), noninfectious (n = 27), and emaciation (n = 27) mortality categories. White line is the median and white circle is the mean; lower and upper limits of the box represent 25th and 75th percentiles; the whiskers extend to the last observation within 1.5 times the interquartile range. The patterned circles are far outside points (outliers). The inset refers to the comparisons after removing “miscellaneous” and “undetermined” subclasses from noninfectious category

As mentioned above, the noninfectious-COD category comprised otters that died of gastrointestinal disorders (n = 5), miscellaneous problems (n = 4), undetermined (n = 9) causes, neoplasia (n = 3), gun shot (n = 3), or shark bite (n = 3). The subclasses “undetermined causes” contained sea otters for which no definitive COD or etiology could be discerned at the time of necropsy. The data were next analyzed based on concentrations found in the individual subclasses within each of the three COD categories (Figure 10). Of the 17 subclasses identified, only one individual otter was available for each of the subclasses “infectious-parasites” and “other-cardiovascular”; these therefore were eliminated from further analysis. When the data were analyzed for the 15 remaining subclasses within the three broad COD categories, the highest mean PBDE concentrations were found for “other-undetermined” (n = 9; 2520 ng/g), “infectious-bacterial” (n = 13; 4180 ng/g), “infectious-neurological” (n = 2; 5180 ng/g), and “trauma-miscellaneous” (n = 3; 5310 ng/g). However, PBDE concentrations were not significantly different (p > 0.05) among the subclasses. PCB concentrations were significantly higher (Mann-Whitney W test; p < 0.05) in the subclass “infectious-neurological” (n = 2; 81,500 ng/g) than in other subclasses. Furthermore, concentrations of PCBs in the subclasses “other-neoplasia” (n = 3; 35,600 ng/g), “other-undetermined” (n = 9; 42,500 ng/g), and “trauma-miscellaneous” (48,800 ng/g) were significantly higher than those in other groups.
https://static-content.springer.com/image/art%3A10.1007%2Fs00244-006-0251-8/MediaObjects/244_2006_251_f10.jpg
Fig. 10.

Mean (±SD) concentrations of polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) in livers of sea otters in various subclasses of the emaciation, infectious, and noninfectious mortality categories. EM emaciation-mating trauma, ES emaciation-starvation, EU emaciation-undetermined, IA infectious-acanthocephalan peritonitis, IB infectious-bacterial, IC infectious-cardiovascular, IE infectious-protozoal encephalitis, IF infectious-fungal, IN infectious-neurological, IP infectious-parasitic, OC other-cardiovascular, OG other-gastrointestinal, OM other-miscellaneous, ON other-neoplasia, OU other-undetermined, TM trauma-miscellaneous (gunshot), TS trauma-shark bite

Nine of the 20 individuals that contained PCB concentrations above 10,000 ng/g, lipid wt, were in either the subclass “undetermined causes” or “miscellaneous problems,” which were categorized as noninfectious COD in the present study (Figure 10). When the PCB concentrations for “undetermined” subclasses were removed (because of the lack of specific etiology) from the noninfectious COD category, concentrations of PCBs in the infectious COD otters were significantly greater than those in noninfectious-COD and emaciated otters (Mann-Whitney W test, p < 0.05) (Figure 9). Nevertheless, concentrations of PBDE did not differ significantly even after the undetermined and miscellaneous subclasses were removed from the noninfectious COD category.

In summary, these results suggest that elevated PCB concentrations in Southern sea otters are associated with infectious diseases. The sea otters analyzed in this study were also examined for the presence of a wide variety of contaminants including trace metals, butyltins, polycyclic aromatic hydrocarbons, pesticides, perfluorochemicals, and synthetic musks (Perrotta 2005; Kannan et al. 2005c). A recent study has reported the association between perfluorochemicals and infectious diseases (Kannan et al. 2006a) and between certain trace elements and disease (Kannan et al.2006b) in sea otters. Based on these results, PCBs and perfluorochemicals were found to be strongly associated with the disease status. Co-occurrence and exposure to multiple contaminants in sea otters may also contribute to immunosuppression. Further research is needed to establish the association between immunological parameters and contaminant levels as well as controlled-exposure studies to establish cause–effect linkages.

Acknowledgments

This study was funded by a grant from the Monterey Bay Sanctuary Foundation.

Copyright information

© Springer Science+Business Media, LLC 2007