Starvation-Induced Decrease in the Maximal Binding Capacity for Triiodothyronine of the Thyroid Hormone Receptor Is Due to a Decrease in the Receptor Protein Tetsuya Tagami, Hirotoshi Nakamura, Sigekazu Sasaki, Yoji Miyoshi, and Kazuwa Nakao Biological responses to thyroid hormones are mediated by the nuclear thyroid hormone receptor (TR). Alterations in the maximal triiodothyronine (T3)-binding capacity (Cmax) of TR measured using a ligand binding assay have been reported under some pathophysiological conditions. Northern blot analysis has indicated that TR mRNA concentrations do not necessarily correlate with Cmax levels. For example, although the decrease in Cmax in rat liver induced by prolonged fasting is well established, TR mRNA concentrations have been reported to be constant. In the present study, we examined starvation- induced changes in TR by Western blot with anti-TR(al + [3) antiserum and by Scatchard plot analysis. Starvation of rats for 72 hours decreased Cmax in the liver to 72.5% of control levels. The 47- and 55-kd TR proteins detected in hepatic nuclear extract by Western blotting also decreased to 64% and 66% of control values, respectively. The starvation-induced changes in Cmax and TR protein levels paralleled the change in total hepatic nuclear protein concentration. These results suggest that the decrease in T3-binding activity of the TR is due to a reduction of the TR protein itself, Copyright © 1996 by W,B. Saunders Company T HYROID HORMONE RECEPTORS (TRs) are li- gand-dependent transcription factors. Tissue respon- siveness to triiodothyronine (T3) is related to TR content. Many studies have reported that the maximal T3-binding capacity (Cmax) of TR changes under physiological and pathological conditions. Tumor growth, ~ uremia, 2 and dia- betes mellitus3 have been reported to reduce TR Cmax. In particular, a starvation-induced decrease in Cmax in rat liver has been well established. This reduction in receptor number, in combination with reduced serum T3 concentra- tion, is assumed to protect cells from the catabolic effects of thyroid hormone during acute caloric and amino acid deficiency. 4-7 The mechanism responsible for the starvation- induced Cmax decrease is still unknown. There are three mRNAs encoding rat TRs with T3- binding activity: al, [31, and [32. 8-1° TRal and TR[31 mRNAs are expressed ubiquitously in rat tissues, u-~3 whereas TR[32 mRNA is mainly found in the pituitary. 1°a4 When the amounts of individual TR mRNAs in several rat tissues were determined by Northern blot analysis in conjunction with solution hybridization, TR[31 mRNA was the predominant receptor in the liver.15 Lane et a116 reported that the level of hepatic TR[31 mRNA did not change by prolonged fasting despite a decrease in Cmax. Discrepancies between TRod and TRI31 mRNA levels and Cmax have also been found in some tissues and at specific stages in development. 15,17 These results raise a question as From the Second Division, Department of Internal Medicine, Kyoto University School of Medicine, Kyoto; and the Second Division, Department of Internal Medicine, Hamamatsu Medical School, Hamamatsu, Japan. SubmittedAugust 29, 1995;accepted February 1, 1996. Supported in part by Grants-in-Aid for Scientific Research from Fellowships of the Japan Society for the Promotion of Science for Japanese Junior Scientists (no. 2037 in 1993 to ZT.) and from the Ministry of Education of Japan (no. 06671018in 1994 to H.N.). Address reprint requests to Tetsuya Tagami, MD, Endocrinology, Metabolism and Molecular Medicine, Tarry 15, Northwestern Univer- sity Medical School, 303 E Chicago Ave, Chicago, IL 60611. Copyright © 1996 by W.B. Saunders Company 0026-0495/96/4508-0010503.00/0 to whether starvation-induced alterations of Cmax corre- late with changes in TR protein levels. Previously, we have measured TR protein content in rat tissues using Western blot analysis with an anti-TR antise- rum that recognizes TRal and TR[3.18 The 47- and 55-kd proteins were identified as TR proteins. Although discrep- ancies between levels of functional TR mRNA and Cmax were known to exist in rat liver and brain, we demonstrated that the relative concentration of TR proteins correlated with Cmax in both tissues. In the present study, we measured Cmax and TR protein level by a T3-binding assay and Western blot analysis, respectively, and examined whether the decrease in Cmax induced by starvation was due to a reduction of TR protein content. MATERIALS AND METHODS Preparation of Nuclear Extracts Wistar male rats (150 g body weight) purchased from Shimizu (Kyoto, Japan) were fed a laboratory diet (Oriental Yeast Indus- trial, Chiba, Japan) for a few days. The manipulated group of rats were fasted for 72 hours before killing. All rats were allowed unlimited access to water. After the animals were killed by exsanguination through the abdominal aorta, the liver was immedi- ately removed and nuclear extracts were prepared as described previously. 19 The nuclear proteins were fractionated by high- performance liquid chromatography (HPLC) using a gel-filtration column (G3000SW; Tosoh, Tokyo, Japan) in 0.05 mol/L sodium phosphate buffer, pH 7.4, with 1 mmol/L MgClz. The TR fraction was obtained by measurement of T3-binding activity in each fractionJ 8 Protein concentration was measured using Coomassie brilliant blue G-250 (Bio-Rad Laboratories, Hercules, CA) with bovine serum albumin as the standard. DNA content was determined by Burton's method. 2° Western Blot Analysis The procedure for raising the anti-TR antiserum, 4BII, and its characteristics have been described previously. 19,21 4BII recognizes functional TRs (cd and [31) but not the variant form (a2) biochemicallyand immunohistochemicaUy. 19,21,22 TR proteins were analyzed by Western blotting as described previously. TM Briefly, proteins (20 to 60 Ixg per lane) in the nuclear 970 Metabolism, Vo145,No 8 (August),1996: pp 970-973