Tumor necrosis factor-alpha (TNF; cachectin), a peptide secreted from stimulated macrophages, mediates some of the metabolic derangements in inflammatory and neoplastic disorders. To determine whether TNF is responsible for the changes in hypothalamic-pituitary-thyroid (HPT) function in nonthyroid illnesses, we administered synthetic human TNF to male Sprague-Dawley rats. The rats were given TNF or saline (control; both pair fed and nonpair fed) iv (six to eight per group). HPT function was tested 8 h after administration of 200 micrograms TNF/kg BW, 8 h after 5 days of 150 micrograms TNF/kg BW, and 8 h after a 3-day series of 50, 200, and 800 micrograms TNF/kg BW. The single injection of 200 micrograms TNF/kg significantly reduced (all P less than 0.05) serum TSH, T4, free T4, T3, and hypothalamic TRH compared to the corresponding hormone levels in saline-injected control rats. Serum TSH and hypothalamic TRH recovered to normal levels after 5 days of 150 micrograms/kg TNF treatment. With the increasing daily doses of TNF, serum TSH and hypothalamic TRH fell significantly. Hepatic 5'-deiodinase activity was reduced after 1 day of TNF treatment, but increased after the 3-day series of injections. TNF treatment reduced pituitary TSH beta mRNA, but did not affect alpha-subunit mRNA. TNF treatment also reduced thyroid 125I uptake and reduced thyroidal release of T4 and T3 in response to bovine TSH, but did not change the TSH response to TRH. TNF treatment reduced the binding of pituitary TSH to Concanavalin-A, indicating that it alters the glycosylation of TSH. The TSH with reduced affinity for this lectin had reduced biological activity when tested in cultured FRTL-5 rat thyroid cells. In vitro, TNF inhibited 125I uptake by cultured FRTL-5 rat thyroid cells and blocked the stimulation of [3H]thymidine uptake by these cells. The data indicate that TNF acts on the HPT axis at multiple levels and suggest that TNF is one of the mediators responsible for alterations in thyroid function tests in patients with nonthyroidal illnesses.

To clarify the role of various thyroid stimulators in normal human pregnancy, we measured serum TSH, chorionic TSH (hCT), hCG, bioassayable thyroid-stimulating activity, T4, T3, T3 uptake, free T4 and free T3 indexes, free T4, and free T3 by dialysis in 339 serum samples from pregnant women at various intervals of pregnancy and in 40 normal female controls. Serum T4 and T3 and free T4 and free T3 indexes were significantly elevated throughout pregnancy in comparison with controls. Free T4 concentration was elevated after 10 weeks of pregnancy and free T3 concentration was elevated at 13--20 weeks. Bioassayable thyroid-stimulating activity was elevated from 9--16 weeks when serum hCG concentrations were highest. Serum TSH levels were significantly lower at 9--12 weeks compared with the rest of pregnancy. hCT was detected in only 35% of sera tested; the mean detectable value was 0.60 +/- 0.04 (SE) microU/ml; only 15% of the detectable values exceeded 1 microU/ml. The level of hCG correlated with bioassayable thyroid-stimulating activity (P less than 0.01). The data indicate that hCT is not a significant thyroid stimulator. We propose that hCG, as a weak thyroid stimulator, causes a modest rise in free thyroid hormone levels early in pregnancy which in turn causes a modest reduction in pituitary TSH secretion.

Iodide uptake, which is necessary for thyroid hormone synthesis, can be inhibited by aging, withdrawal of TSH, or increased tumor necrosis factor (TNF) and transforming growth factor (TGF)-beta1 levels resulting from the nonthyroid illness syndrome. TNF induces receptor-mediated activation of sphingomyelinase, which converts sphingomyelin to ceramide, a mediator of TNF actions. Thyroid follicular cells transport iodide from blood into the follicular lumen against an iodide gradient by means of coupled transport of Na+ ions and I- ions via the Na+/I- symporter (NIS). An inward Na+ gradient is maintained by Na+/K+-ATPase. The recent cloning and sequencing of the rat NIS complementary DNA has made possible studies on the mechanism by which TSH, aging, and cytokines regulate I- uptake by thyroid cells. Young (<20 passages) and aged (>40 passages) FRTL-5 cells grown with or without TSH were treated with various concentrations of TNF, TGF-beta1, sphingomyelinase, or ceramide. NIS messenger RNA (mRNA) levels in aged cells were only 2% of those in young cells. Withdrawal of TSH from young cells reduced NIS mRNA levels by more than 90%. TNF reduced NIS mRNA levels in young cells grown with TSH at t1/2 = 0.62 days, a cycloheximide inhibitable effect. Similar treatments with TGF-beta1, sphingomyelinase, or ceramide reduced NIS mRNA by 70-90%. Ceramide reduced 125I(-)-uptake by 50%. The addition of TNF increased both the sphingomyelin and ceramide levels 3- to 5-fold in young and old cells. We conclude that 1) the decline in iodide uptake due to aging, a fall in serum TSH or an increase in TNF or TGF-beta1 is mediated primarily by a reduction in thyroid NIS expression; and 2) that receptor-mediated activation of sphingomyelinase is an important, protein synthesis-dependent, intracellular pathway for inhibition of NIS expression by TNF.

No abstract