ABSTRACT

Isolated porcine thyroid follicular cells were cultured on a collagen-coated Millipore filter to form a monolayer. The monolayer could translocate ¹²⁵I added in the medium beneath the filter (basal medium) into the medium above the monolayer (apical medium) and form an iodide concentration gradient of several-fold. Transcellular iodide pump activity was observed when the cells were cultured with TSH in the basal medium. In the absence of TSH, the translocation of iodide was very slow. The concentration of TSH required to activate the iodide pump was 0·1–0·3 mU/ml. Addition of ClO₄ ⁻ to the basal medium inhibited transcellular transport, whilst addition of ClO₄ ⁻ to the apical medium was much less effective.

Constituents labelled with ¹²⁵I in the apical medium were analysed. The amount of protein-bound ¹²⁵I measured by acid precipitation was 3–8% of the total radioactivity. The residual radioactivity was found to be iodide ion by paper chromatography. Further analysis by sodium dodecylsulphate–polyacrylamide gel electrophoresis revealed that most of the ¹²⁵I-labelled protein was at the position of bovine serum albumin which had been added to the culture medium.

The monolayer culture of cells on collagen-coated filter would be a useful experimental system for analysing thyroid cell functions for which the cell polarity is essential.

Journal of Endocrinology (1990) 126, 275–281

Abstract

The functional capacity of the recombinant human FSH (hFSH) receptor was tested on the basis of gonadotrophin stimulation of cyclic AMP (cAMP) production by transient transfections of 293 cells and stable transfections of Chinese hamster ovary (CHO) cells. A CHO cell line expressed with the hFSH receptor cDNA covering the entire amino acid coding region revealed the presence of FSH binding site (K _d 6·2 × 10⁻¹⁰ m) on the plasma membrane. Treatment of transfected cells with hFSH induced dose-dependent increases in intracellular cAMP production. These results indicate that the hFSH receptor functionally couples with endogenous adenylyl cyclase. Although rat FSH also induced dose-dependent increases in cAMP production, bovine FSH was effective only at high doses and human chorionic gonadotropin did not alter cAMP levels compared with control values.

Northern blot analysis with a cRNA probe derived from hFSH receptor cDNA indicated the presence of two common FSH receptor mRNA transcripts (2·4 and 4·1 kb) in RNA prepared from a human ovary and transfected cell lines.

Preincubation of CHO cells expressing a functional hFSH receptor (CHO-FSHR) with FSH for 16 h decreased the subsequent cAMP production resulting from a 30-min pulse of FSH stimulation. These results indicate that desensitization of the adenylyl cyclase response to FSH stimulation occurs in CHO-FSHR cells. This cell line therefore provides a tool with which to pursue detailed studies on the molecular basis of FSH-induced desensitization.

Journal of Endocrinology (1994) 141, 369–375

Abstract

The acquisition of FSH receptor during preantral folliculogenesis is believed to be a key step in the subsequent development of follicles. We examined the interaction between activin and cAMP in FSH receptor induction in rat granulosa cells by measuring ¹²⁵I-FSH binding and FSH receptor mRNA. In the ¹²⁵I-FSH binding study, 0·2 mm 8-Br-cAMP and 1 μm forskolin were maximally effective in FSH receptor induction (169 and 220% respectively of control), while higher concentrations gave attenuated responses. It appears that cAMP has ambivalent effects on FSH receptor induction depending on the concentration and length of exposure. Activin alone dramatically increased the number of FSH receptors (314% of control). Moreover, synergistic effects of activin and 8-Br-cAMP or forskolin were observed on FSH receptor induction: a combination of activin (80 ng/ml) and low doses of 8-Br-cAMP (0·2 mm) or forskolin (1 μm) was most effective (160 or 140% of that induced by activin alone) and receptor levels reached a maximum at 24 h. These levels then markedly decreased after 72 h of incubation. Northern blot analysis revealed that the combination of activin (80 ng/ml) and 8-Br-cAMP (0·2 mm) or forskolin (1 μm) increased FSH receptor mRNA to about 140% of that induced by activin alone. These results indicate that activin and cAMP induced FSH receptor synergistically. However, activin did not enhance the production of cAMP induced by forskolin. In addition, a protein kinase A inhibitor (H89) (2 μm), which inhibited the effects of forskolin, had no effect on the action of activin. Taken together, the present findings suggest that the action of activin is not via a cAMP pathway, and that activin works co-operatively with cAMP on folliculogenesis.

Journal of Endocrinology (1995) 147, 103–110

Hormonal responsiveness in peripheral tissues is variable in patients with resistance to thyroid hormone (RTH). One cause of this may be differential interaction of RTH mutants of thyroid hormone receptor beta (TR beta) with TR auxiliary proteins (TRAPs). We used gel shift mobility assays to examine the interaction of wild-type and mutant TR beta s with retinoid X receptors (RXRs) and endogenous TRAPs. Some mutants showed reduced homodimerization but retained heterodimerization with recombinant RXRs. Wild-type TR beta formed heterodimeric complexes with multiple TRAPs in nuclear extracts of rat tissues, but RTH mutants showed variably altered heterodimerization with each TRAP. With liver nuclear extract, all mutants with impaired homodimerization also showed impaired TR beta-TRAP heterodimerization. Thus heterodimerizations with RXRs and TRAPs are differently affected by RTH mutations. Our results suggest that multiple TRAPs are expressed in tissue-specific patterns. The variability of TR beta heterodimerization with TRAPs may account, in part, for the variable tissue responsiveness in RTH.

ABSTRACT

GH accelerates hepatic regeneration in the rat. Hepatocyte growth factor (HGF), a potent hepatocyte mitogen in vitro, is considered to be a major regulator of hepatic regeneration. In the present study, the effects of GH and insulin-like growth factor-I (IGF-I) on HGF gene expression in regenerating rat liver was investigated. In hypophysectomized rats treated with GH, hepatic HGF mRNA levels were increased 3 h after partial hepatectomy and reached peak levels after 5 h. In rats with intact pituitaries and in hypophysectomized rats not given GH treatment, HGF mRNA levels in liver were unchanged during the first 5 h following hepatectomy and reached peak levels after 10-18 h. DNA synthesis in the liver of GH-treated rats increased from low levels 10 h after hepatectomy to peak levels after 18 h. In rats without GH treatment the synthesis of DNA was still low 18 h after hepatectomy and was increased after 26 h. Treatment of hypophysectomized rats with IGF-I promoted increases in hepatic HGF mRNA levels and DNA synthesis 3·5 h and 15 h after hepatectomy respectively. HGF mRNA levels were constantly lower after sham-hepatectomy than after partial hepatectomy. In summary, in hypophysectomized rats the responses of hepatic HGF gene expression and DNA synthesis to partial hepatectomy were both accelerated by treatment with GH or IGF-I.

Journal of Endocrinology (1992) 135, 59–67

Abstract

The effect of FSH on the induction of FSH receptors in granulosa cells is believed to be mediated, at least in part, by the cAMP second messenger system. We examined the effect of activin and cAMP on FSH receptor expression in this culture system. Steady-state levels of FSH receptor mRNA, analyzed by Northern blot hybridization, increased 3·5-fold in response to 24-h incubation with activin and 1·7-fold with 12-h incubation with 8-bromoadenosine 3,5-cyclic monophosphate (8-Br-cAMP; 0·2 mm). We have investigated whether 8-Br-cAMP- and/or activin-induced increases in FSH receptor mRNA levels are the result of increased transcription and/or altered mRNA stability. The rates of FSH receptor mRNA gene transcription, assessed by nuclear run-on transcription assay, increased 3-fold in cells treated with activin and 1·5-fold in cells treated with 8-Br-cAMP for 2 h. To examine the degradation rates of FSH receptor mRNA transcripts, granulosa cells were preincubated with 8-Br-cAMP, activin, or medium alone for 6 h. After the preincubation period, 5 μm actinomycin-D or 200 μm 5,6-dichloro-1-β-ribofuranosyl benzimidazole were added to arrest new RNA synthesis. The decay curves for the 2·4 kb FSH receptor mRNA transcript in granulosa cells were not significantly different in the absence or presence of 8-Br-cAMP. Activin, on the other hand, significantly altered the slope of the FSH receptor mRNA decay curve and increased the half-life of the 2·4 kb FSH receptor mRNA transcript. These data provide evidence that cAMP induces FSH receptor mRNA levels by stimulating the transcription rate and that activin increases FSH receptor mRNA levels both by stimulating transcription rates and by stabilizing the FSH receptor mRNA transcripts.

Journal of Endocrinology (1997) 153, 465–473

Human thyroid hormone receptor (TR) is encoded by two distinct genes, TR alpha and TR beta. TR heterodimerizes with retinoid X receptor (RXR) and binds efficiently to the thyroid hormone (T(3)) response element (TRE) of target genes. In the absence of T(3), unliganded TR suppresses the basal promoter activity of positively regulated genes (silencing). Silencing mediator for retinoid and thyroid hormone receptors (SMRT) and nuclear receptor co-repressor (N-CoR) interact with unliganded TR and function as corepressor proteins. Previously, we found beta F451X with carboxyl (C)-terminal 11-amino acid deletion had stronger silencing potency than wild-type TR beta 1 and beta E449X with C-terminal 13-amino acid deletion on a subset of TREs. In the present study, to assess the isoform-specific effects of the C-terminal truncations on TR silencing, we constructed two mutant TR alpha 1s (alpha F397X and alpha E395X) with the same respective C-terminal truncations as beta F451X and beta E449X and analysed their silencing activities. Unlike beta F451X and beta E449X, alpha F397X and alpha E395X showed similarly stronger silencing potency than wild-type TR alpha 1. We further studied the abilities of wild-type and the mutant TR beta 1s and alpha 1s on RXR and co-repressor binding by a two-hybrid interference assay. beta F451X had significantly stronger abilities to bind to RXR and SMRT than did wild-type TR beta 1 and beta E449X. In contrast, wild-type TR alpha 1, alpha F397X and alpha E395X showed similar abilities to bind to RXR and SMRT. beta E449X and alpha E395X, which have identical C-terminal truncation, showed less ability to bind to N-CoR than did wild-type TR beta 1 and beta F451X and wild-type TR alpha 1 and alpha F397X respectively. These results indicate that an identical C-terminal truncation gives rise to different effects on TR beta 1 and alpha1 with respect to silencing potency, RXR binding and SMRT binding. The difference in the silencing potency among wild-type TR beta 1, beta F451X and beta E449X correlated well with the difference in the ability to bind co-repressor SMRT.

Abstract

Clinical resistance to thyroid hormone (RTH) has been classified into generalized resistance to thyroid hormone (GRTH) and pituitary resistance to thyroid hormone (PRTH) types. Since similar mutations have been identified in tri-iodothyronine (T₃) receptor (TR) β gene in GRTH and PRTH, and since considerable overlap has been seen in the clinical manifestations in patients with GRTH and PRTH, two subtypes of RTH are now considered to be a continuous spectrum with the same genetic defect. A point mutation at amino acid Arg 338 to Trp (R338W) which we identified in a patient with PRTH is very interesting, since R338W has been found in several other patients with PRTH, raising the possibility that this mutation may tend to associate with a phenotype of PRTH.

In our previous study, we found that R338W had relatively less impaired transcriptional potency, weaker dominant negative activity on various T₃ response elements and poor homodimer formation, as compared with another GRTH mutant. In this study, to investigate the functional properties of R338W further, especially in terms of the relation between transcriptional activity and dimer formations, we introduced the R338W mutation into the mutant receptors, K443E and F451X, constructing the double mutants, R338W/K443E and R338W/F451X. Both R338W/K443E and R338W/F451X showed negligible T₃ binding and transcriptional activities. The dominant negative activities of K443E and F451X were, however, significantly weakened by introducing the R338W mutation. As a control, a double mutant G345R/K443E was constructed by introducing a point mutation, G345R, located in the same exon 9 as R338W, into the K443E mutant. Dominant negative activity did not differ between G345R/K443E and K443E. Homodimer formation was significantly reduced in the double mutants containing R338W, but not G345R.

In summary, introducing the R338W mutation, but not G345R, into the mutant TR significantly weakened the dominant negative activity, despite further impairment of the T₃ binding and transcriptional activities.

Journal of Endocrinology (1996) 151, 293–300

We have previously reported that a decrease in hepatocyte growth factor (HGF), which has many protective functions against endothelial damage by high d-glucose, might be a trigger of endothelial injury. However, the regulation of vascular HGF in diabetes mellitus (DM) has not been clarified in vivo, although vascular disease is frequently observed in DM patients. In addition, our previous report revealed that a prostaglandin I(2) (PGI(2)) analogue prevented endothelial cell death through the induction of vascular HGF production in cultured human epithelial cells. Thus, in this study, we examined the effects of a PGI(2) analogue in the regulation of the local HGF system using DM rats. A PGI(2) analogue (beraprost sodium; 300 and 600 micro g/kg per day) or vehicle was administered to 16-week-old DM rats induced by administration of streptozotocin for 28 days. Endothelial function was evaluated by the vasodilator response to acetylcholine, and the expression of vascular HGF mRNA was measured by Northern blotting. Of importance, expression of HGF mRNA was significantly decreased in the blood vessels of DM rats as compared with non-DM (P<0.01). In addition, the in vitro vasodilator response of the abdominal aorta to acetylcholine was markedly impaired in DM rats. Importantly, the vasodilator response was restored by PGI(2) treatment in a dose-dependent manner (P<0.01), whereas N(omega)-nitro-l-arginine methyl ester inhibited the restoration of endothelial function. Of particular interest, vascular HGF mRNA and protein were significantly increased in the blood vessels of DM rats treated with PGI(2) as compared with vehicle. Similarly, an increase in HGF protein was also confirmed by immunohistochemical analysis. In addition, the specific HGF receptor, c-met, was also increased by PGI(2) treatment. Overall, this study demonstrated that treatment with a PGI(2) analogue restored endothelial dysfunction in DM rats, accompanied by the induction of vascular HGF and c-met expression. Increased local vascular HGF production by a PGI(2) analogue may prevent endothelial injury, potentially resulting in the improvement of endothelial dysfunction.