The Feline Thyroid Gland: Model for Endocrine Disruption by Polybrominated Diphenyl Ethers?

Polybrominated diphenyl ethers (PBDEs) are ubiquitous flame retardants and known endocrine disruptors used in many household products since the 1970s. They are lipophilic and bioaccumulate, with tissue residues in humans and wildlife exponentially increasing since their inception. Feline hyperthyroidism (FH) is a common disease of geriatric domestic cats first recognized only after PBDE production began. FH has been associated with a variety of risk factors and the associated thyroid abnormalities are morphologically and functionally analogous to human toxic (multi)nodular goiter (TNG), but its etiology remains unknown. Thyroid hyperplasia, a prelude of TNG, was linked with experimental exposure of rats to PBDEs in 1975. Structurally similar to thyroxine (T4) and triiodothyronine (T3), hydroxylated metabolites of PBDEs have been shown to bind to transthyretin, the dominant high-affinity serum binding protein in cats, and to thyroid hormone receptors (TRs), and also to inhibit 5???-deiodinases. Experimental exposures to PBDEs have resulted in decreased T4 levels via uridinediphosphoglucuronosyl transferase (UDPGT) induction. To test the hypothesis that PBDE exposure is associated with endocrine disruption of the feline thyroid gland and hyperthyroidism, we analyzed serum from 62 client-owned domestic and 9 feral cats (2 of which were pooled) for 15 PBDE congeners with gas chromatography/high resolution mass spectroscopy (GC/HRMS). To evaluate potential sources of exposure, dust samples from 19 homes and 10 commercial canned cat food samples were extracted using a modified EPA Method 3545 and analyzed via GC/HRMS. Total lipid-adjusted serum PBDE burdens ranged from 373 ng/g lipid to 51,063 ng/g lipid (mean: 5,865 ng/g lipid) for client-owned cats and 457-3712 ng/g lipid (mean: 1,203 ng/g lipid) for feral cats. This difference was statistically significant and may be attributable to environment, diet, and/or age. Among the client-owned cats, total lipid-adjusted PBDEs are extremely high compared to published values for human beings and represent the highest documented PBDE burdens in any species on the planet to date. Total lipid-adjusted PBDEs did not correlate with age or thyroid status. BDE-153 had a higher percent composition in serum of euthyroid cats than in hyperthyroid cats. The percent composition of BDE-183 was significantly higher in feral cats versus client-owned cats of either thyroid status. Elevated TSH was not detected any subject. In fact, all of the hyperthyroid cats and 79% of the euthyroid cats, including all of the euthyroid feral cats, had TSH concentrations below the detection limit of 0.03 ng/ml on the canine DPC IMMULITE?? assay standardized with recombinant feline TSH. Concentrations of PBDEs in dust ranged from 510 ng/g to 95,448 ng/g (mean: 8,098 ng/g) and were significantly higher in dust samples from homes of hyperthyroid cats as compared to those of euthyroid cats. PBDEs in canned foods ranged from 0.42-3.09 ng/g (mean: 1.79 ng/g). We estimated that total daily ingestion of PBDEs for domestic cats ranges from 32 to 3,906 ng/kg BW (mean: 375 ng/kg BW). Backward calculation from the mean feline serum body burden suggested daily exposure to 1,038 ng/kg/day (range: 66-9,038 ng/kg/day). At the high end, forward and backward calculation estimates equated to 6.5 and 15%, respectively, of the lowest single acutely toxic dose known to disrupt neurodevelopment in laboratory animals. These estimates also equated to 1.3 and 3.0%, respectively, of the lowest single acutely toxic dose known to disrupt thyroid function. Body burdens of PBDEs in cats chronically exposed to them may reach endocrine disruptive concentrations over time, a possible explanation as to why FH is predominantly a geriatric disease. We concluded that domestic cats are highly exposed to PBDEs, largely through ingestion of household dust during grooming. If PBDEs play a role in hyperplasia leading to thyroid autonomy, they apparently do not do so by markedly increasing TSH. It seems more plausible that PBDEs may act directly at the thyroid nuclear receptors, at pituitary TRs, and/or at an earlier, potentially developmental, time point.published or submitted for publicationis peer reviewe