In the course of this

In the course of this study, K+PFOS did not appear to have any effect on the thyroid parameters evaluated (H&E histology, follicular epithelial cell proliferation, and follicular epithelial apoptosis). This is an interesting observation considering the fact that treatment with K+PFOS resulted in activation of PPARα and CAR/PXR in this study as well as in the 28-day study also reported in this issue (Elcombe et al., 2012). As discussed in detail in the just-cited article, other compounds that increase the expression of PPARα, CAR, and PXR have been shown to be capable of stimulating thyroid follicular epithelial proliferation in rats through a combination of increased phase II conjugation, increased expression of liver transport systems responsible for hepatocellular uptake of circulating thyroid hormones, as well as increased biliary efflux of conjugated and unconjugated hormones (Klaassen and Hood, 2001, Qatanani et al., 2005, Vansell and Klaassen, 2001, Vansell and Klaassen, 2002a, Vansell and Klaassen, 2002b, Wieneke et al., 2009, Wong et al., 2005). In fact, in the 28-day study (Elcombe et al., 2012), both Wy 14,643 and phenobarbital, model activators of PPARα and CAR/PXR, respectively, increased the thyroid follicular proliferation index as measured by BrdU incorporation; whereas, as in the present study, K+PFOS did not. PFOS treatment in rats has been shown to reduce circulating thyroid hormone via displacement of thyroxine (T4) from serum TG4-155 and increased turnover in liver (Chang et al., 2008, Yu et al., 2009, Yu et al., 2011); however, in those studies, serum TSH and release of TSH from the pituitary were unaffected, suggesting maintenance of thyroid hormone homeostasis. It is possible that PFOS either does not induce or only weakly induces the specific enzymes and/or transporters responsible for the PB- and Wy 14,643-mediated increase in thyroid follicular hyperplasia. It has been shown in rats that inducing compounds such as pregnenalone-16α-carbonitrile (PCN, a specific PXR agonist) that increase serum TSH concentrations leading to stimulation of proliferation in the thyroid follicular epithelium do so by increasing the glucuronidation of triiodothyronine (T3); whereas, inducing compounds that only increase the glucuronidation of T4 do not produce an associated TSH increase (Klaassen and Hood, 2001). It has been demonstrated that the specific PPARα agonist, nafenopin, like PFOS, is a weak competitor for T4 binding in serum which increases plasma free T4 as well as plasma and fecal clearance of T4 even in the presence of decreased 5′-deiodination (Kaiser et al., 1988); however, at the same time, plasma clearance of T3 decreased and TSH levels were maintained at euthyroid levels. Kaiser et al. suggested that increased T4 glucoronidation and biliary elimination explained the decrease in total T4. Therefore, it is possible that differences exist between PFOS and the inducers Wy 14,643 and phenobarbital relative to competition for serum T4 binding and/or induction in specific glucuronidation pathways for T4 and T3. In conclusion, the principal objective of this study was to evaluate the reversibility of the inductive effects of K+PFOS treatment in rats over an 84-day period following seven days of dietary treatment at dose levels used in the 28-day study reported also in this issue (Elcombe et al., 2012). As in the 28-day study, after seven days of feeding K+PFOS to male rats in the study reported herein, K+PFOS exhibited the combined hepatic effects of PPARα and CAR/PXR induction. Thus, the results obtained from this study offer further evidence supporting the role of K+PFOS-mediated induction of PPARα and CAR/PXR in the etiology of hepatocellular adenoma observed in the 104-week bioassay with K+PFOS in Sprague Dawley rats (Butenhoff et al., 2012). Many of the K+PFOS-induced effects present one day after the 7-day treatment period (recovery Day 1) resolved toward or returned to control levels during the 84-day recovery period. These included: hepatocellular proliferative index; activities associated with ACOX, CYP4A, and CYP2B; and liver weight. To an extent, these changes followed the elimination of PFOS anion from serum and liver over the 84-day period. Mean serum PFOS concentrations on recovery Day 1 were 39 and 140μg/mL (20ppm and 100ppm K+PFOS, respectively), decreasing to 4 and 26μg/mL by recovery Day 84. Liver PFOS concentrations were generally 3–5 times those of the corresponding serum concentrations. However, several K+PFOS-induced effects did not appear to subside significantly during the 84 days of recovery. These included: hepatocellular apoptotic index; centrilobular hepatocellular hypertrophy; liver microsomal total CYP450; and lowered cholesterol. Thus, hepatic effects in male rats resulting from K+PFOS-induced activation of PPARα and CAR/PXR resolved slowly or were still present after 84-days following a 7-day dietary treatment, consistent with the slow elimination rate of PFOS.