Quantitative assessment of pituitary resistance to thyroid hormone from plots of the logarithm of thyrotropin versus serum free thyroxine index.

Previous studies have shown that, in patients with primary alterations in thyroid hormone secretion, the level of the natural logarithm of serum TSH (lnTSH) is negatively related to the level of free T4. Because such patients can generally be assumed to exhibit normal tissue responsivity to thyroid hormone, we were interested in determining whether the lnTSH/free T4 index (FTI) relationship in patients with established thyroid hormone resistance (THR) exhibit a lower slope than patients with normal tissue sensitivity to thyroid hormone. We have therefore analyzed the relationship between the lnTSH and the FTI in members of three families with documented THR. In these patients, a given dose of T4 was maintained for a 1- to 2-month period, to achieve hormonal equilibration. Two of the families, though not related, exhibited the same mutation, E460K. The third was identified as A317T. As anticipated, the slope of the lnTSH/FTI ratio was significantly lower in the patients with THR than in T4-treated patients who were presumed to have normal sensitivity to thyroid hormone. The slope of the lnTSH/FTI relationship seemed to be characteristic of the specific mutation involved in the three genotypes (wild-type and two mutations) examined. Further, the in vivo slope of the lnTSH/FTI relationship seemed to be linearly related to the T3 association constant of the in vitro translated receptor. These findings support the potential usefulness of measuring the slope of lnTSH, as a function of the FTI, in quantitating pituitary THR.

Evolving concepts of thyroid hormone action.

The past 25 years have witnessed dramatic changes in our concepts of thyroid hormone action. Progress in this area was made possible by the recognition of the central role of triiodothyronine in mediating thyroid hormone action and the recognition of specific nuclear receptors in target tissues as demonstrated by displacement studies. The cloning of the receptors and receptor variants has enabled investigators to undertake detailed analyses of the biochemical events which underlie the physiological and pathological action of thyroid hormone.

Chicken ovalbumin upstream promoter-transcription factor (COUP-TF) modulates expression of the Purkinje cell protein-2 gene. A potential role for COUP-TF in repressing premature thyroid hormone action in the developing brain.

The cerebellar Purkinje cell-specific PCP-2 gene is transcriptionally activated by thyroid hormone during the 2nd and 3rd weeks of postnatal life in the rat. In contrast, thyroid hormone has no detectable effects on PCP-2 expression in the fetal rat. We now present data that suggest that the orphan nuclear receptor chicken ovalbumin upstream promoter-transcription factor (COUP-TF) represses triiodothyronine (T3)-dependent transcriptional activation of PCP-2 in the immature Purkinje cell. Gel shift assays show that the PCP-2 A1TRE and adjoining sequences (-295/-199 region) bind to rat and mouse brain nucleoproteins in a developmentally regulated fashion and that one of these nucleoproteins could be the orphan nucleoprotein COUP-TF. In support of this hypothesis, in vitro translated COUP-TF binds to the -295/-199 region and COUP-TF represses T3-dependent activation of the PCP-2 promoter in transient transfection analyses. Finally, immunohistochemical studies reveal that COUP-TF is specifically expressed in the immature fetal and early neonatal Purkinje cell and that this expression diminishes coincident with thyroid hormone induction of PCP-2 expression. Our findings are consistent with the hypothesis that the presence or absence of inhibitory proteins bound to the thyroid hormone response element of T3-responsive genes governs the responsivity of these genes to thyroid hormone during brain development.

Beta receptor isoforms are not essential for thyroid hormone-dependent acceleration of PCP-2 and myelin basic protein gene expression in the developing brains of neonatal mice.

In rat pups, thyroid hormone dependent brain development coincides with the appearance of the thyroid hormone receptor (TR)beta1 isoform. This finding led to the suggestion that TRbeta1 plays an essential role in brain development. The recent availability of a mouse TRbeta knockout strain allowed us to test this possibility by determining whether TRbeta is essential for the normal developmental pattern of expression of two thyroid hormone regulated brain genes, myelin basic protein (MBP), and Purkinje cell protein 2 (Pcp-2). Northern analysis of total mRNA from the brains of wild-type mice established that, as in the rat pup, the initial rate of rise of the MBP and Pcp-2 mRNA is slowed in the hypothyroid state. Supporting the effectiveness of TRbeta gene deletion was the finding that the thiiodothyronine (T3) nuclear binding capacity in the livers and brains of knockout animals was consistent with the fractional contribution of TRbeta1 to total binding capacity in the wild-type tissues. Further, no TRbeta1 could be detected by isoform-specific immunoprecipitation of nuclear receptor extracts. However, deletion of the functional TRbeta in the TRbeta knockout mice did not affect the normal ontogeny of expression of the Pcp-2 and MBP genes in the postnatal pup. We conclude that TRbeta is not essential for the normal developmental expression of these T3 dependent brain genes.

Lack of effect of thyroid hormone on late fetal rat brain development.

Studies were undertaken to test whether alterations in fetal brain thyroid hormone levels during the final week of gestation can prematurely induce gene expression in brain or affect cerebellar morphogenesis. Pregnant dams were treated either by administration of 0.025% methimazole (MMI) in the drinking water from day 14 post conception (PC14) or administration of 2.5 mg T4/100 g BW on PC15. On PC21, treatment with MMI resulted in a 53% fall in fetal brain T3 levels and excess T4 resulted in a 2- to 3-fold increase to concentrations observed in adult brains. Neither excess nor reduced levels of T3 caused alterations in the expression of the myelin basic protein, Pcp-2 or calmodulin kinase IV genes. Cerebella of control brains showed early evidence of foliation and the presence of a several cell thick Purkinje cell layer and an external granule layer. No treatment induced effects were evident. Thus, at the late fetal stage in the rat, the developing brain appears to be unresponsive to thyroid hormone despite the presence of thyroid hormone receptors. We infer the presence of as yet unidentified factors that suppress precocious response to thyroid hormone or the absence of cofactors essential for such a response.

Purkinje cell protein-2 cis-elements mediate repression of T3-dependent transcriptional activation.

Previous studies in our laboratory show that triiodothyronine upregulates expression of the cerebellar Purkinje cell-specific gene Pcp-2 during the first 2 weeks of rat neonatal life. A specific thyroid hormone response element, the A1 TRE, mediates this regulation. The finding that the contiguous 68 bases (-267/ -199) of the Pcp-2 promoter 3' to the A1 TRE repressed T3 response in transactivation studies suggested that this sequence could play a role in preventing premature T3-dependent activation of Pcp-2 in the fetus. We now show that deletion of this region resulted in enhanced T3-dependent activation of the native Pcp-2 promoter. The sequence is not a generalized silencer since it does not alter basal activity of mouse mammary tumor virus (MMTV) or thymidine kinase (TK) promoters. Deletion and linker scanning studies indicate that the 5' 30 bases of the -267/ -199 region mediate most of the response silencing activity. The -267/ -199 region also attenuates T3-induced transactivation mediated by other TREs. Gel shift analysis reveals that nuclear proteins from fetal but not adult brains complex with the -267/ -199 region, supporting the hypothesis that this region binds proteins that suppress Pcp-2 expression early in brain development.

Transient stimulation of myelin basic protein gene expression in differentiating cultured oligodendrocytes: a model for 3,5,3'-triiodothyronine-induced brain development.

We compared the regulation of myelin basic protein (MBP) gene expression by T3 in differentiating oligodendrocytes in culture with that previously observed by us in the neonatal rat brain. As in intact brain, expression of the T3R alpha gene preceded that of the T3R beta gene. Although the absence of T3 retarded the rate of accumulation of MBP messenger RNA, the level ultimately attained was similar to that reached in the presence of T3. This relationship mirrored the pattern observed in the neonatal brain. Transient transfection experiments showed that T3 regulates MBP expression at the transcriptional level, but only for a limited period during differentiation. These observations imply that the early rise of MBP messenger RNA is T3 dependent, whereas the terminal levels are maintained independently of T3. Both the T3-dependent and, surprisingly, the T3-independent expression of MBP require the presence of an intact T3 response element. T3 receptor may regulate MBP expression in a ligand-independent manner, or a nuclear factor other than T3 receptor may bind to the T3 response element of MBP to regulate terminal gene expression. These findings support the use of differentiating oligodendrocytes as a model of T3-induced brain development.

Isoform-specific 3,5,3'-triiodothyronine receptor binding capacity and messenger ribonucleic acid content in rat adenohypophysis: effect of thyroidal state and comparison with extrapituitary tissues.

Although the role of the three functional thyroid hormone receptor isoforms (TR beta 1, TR beta 2, and TR alpha 1) remains unclear, studies by Hodin and Lazar et al. have suggested that restriction of TR beta 2 messenger RNA (mRNA) to rat pituitary could reflect a specific regulatory role in the pituitary. Supporting their hypothesis was a significant fall in pituitary TR beta 2 mRNA after T3 administration. These observations prompted us to assess the effect of thyroidal state on the level of TR beta 2 protein, as inferred by immunoprecipitation of TR beta 2 nuclear binding activity. In contrast to the behavior of the mRNA, we noted surprising stability in the levels of total nuclear TR binding capacity and TR isoform distribution in the transition from hypo- to hyperthyroid states. Calculations based on these and previous data from this laboratory (7) show that the average cellular content of TR beta 2 mRNA in pituitary is 0.6 molecules, whereas the content of TR beta 2 mRNA molecules in extrapituitary tissues is less than 0.007 molecule/cell. A high TR beta 2 protein/mRNA ratio in extrapituitary tissues thus could reflect a rapid turnover of TR beta 2 mRNA compared to TR beta 2 protein. This would explain the widespread distribution of TR beta 2 protein and the scarcity of mRNA in extrapituitary tissues.

Tissue-specific regulation of malic enzyme by thyroid hormone in the neonatal rat.

Two recent studies have claimed that thyroid hormone administration accelerates malic enzyme gene expression in the neonatal brain in contrast to the well-documented lack of effect of triiodothyronine on malic enzyme gene expression in the adult brain. Since these observations conflict with earlier observations in our laboratory, we reinvestigated the effect of thyroid hormone status on the ontogeny of malic enzyme gene expression in the neonatal rat. Neither hypothyroidism nor hyperthyroidism influenced the ontogenesis of malic enzyme activity in neonatal brain whereas the patterns of gene expression and enzyme activity in liver were markedly affected. Our results suggest that tissue-specific factors in brain block thyroid hormone-induced gene expression by thyroid hormone.

Thyroid hormone receptor isoform content in cultured type 1 and type 2 astrocytes.

Immunohistochemical studies previously reported from this laboratory showed that astrocytes in adult rat brain appear devoid of all thyroid hormone receptor (TR) isoforms. These findings, however, contrast with reports of measurable nuclear T3 binding in astrocytes in cell culture. To address this discrepancy, TR protein and messenger RNA (mRNA) content of type 1 and type 2 astrocytes in culture were assayed. Type 1 cells represent astrocytes present in brain in vivo. Type 2 astrocytes differentiate in culture from bipotential progenitor O-2A cells in the presence of serum. Under serum-free conditions, these progenitor cells differentiate into oligodendroglia. Total nuclear T3 binding capacity in both type 1 and type 2 astrocytes was approximately 3000 sites/cell. Northern blots showed the presence of mRNA for TRbeta1, TRalpha1, and TRalpha2 in type 2 cells but failed to reveal the presence of these mRNAs in type 1 astrocytes. Moreover, Northern blots also failed to reveal TRbeta2 mRNA in both type 1 and type 2 astrocytes. These findings, therefore, raised a question as to which receptor isoform was responsible for the nuclear binding capacity observed in type 1 astrocytes. As anticipated, immunocytochemical analysis demonstrated prominent nuclear signals for TRbeta1, TRalpha1, and TRalpha2 mRNA in type 2 astrocytes but failed to demonstrate TRbeta1, TRalpha1, or TRalpha2 in type 2 astrocytes. Application of RT-PCR, however, revealed the presence of low levels of TRbeta2 mRNA in type 1 astrocytes. When stained with a specific anti-TRbeta2 antiserum, both type 1 and type 2 astrocytes showed a strong fluorescent signal concentrated in the nucleus. These data indicate that under the special conditions of cell culture, expression of the TRbeta2 isoform in type 1 accounts for the measured nuclear T3 binding capacity.

A Purkinje cell protein-2 intronic thyroid hormone response element binds developmentally regulated thyroid hormone receptor-nuclear protein complexes.

Two thyroid hormone response elements (TREs), designated A1 TRE (-295/-268) and B1 TRE (+207/+227), have been identified within the Purkinje cell-expressed Pcp-2 gene. Previous studies have characterized the A1 TRE (Zou et al., 1994). This article analyzes the structural and functional characteristics of the intronic B1 TRE. The B1 sequence contains four overlapping TRE half-sites. The 3' DR4 motif, consisting of the second and forth half-sites, is responsible for the T3 induction observed with the B1 sequence. Gel-shift analysis reveals developmentally regulated complexes that are abundant in the fetus and at birth and then fall precipitously in the neonate bind to B1. The observed time-course of these complexes varies inversely with the rise in Pcp-2 expression, thus raising the possibility that the complexes may represent inhibitory factors. Supershift analysis indicates that endogenous TR alpha 1 is present in the fetal nuclear protein complexes that bind to B1. Competition analysis also indicates the second B1 TRE half-site is important in binding the TR alpha 1-TRAP complexes. These studies suggest that the B1 sequence may bind potential TR alpha 1-TRAP repressor complexes in the fetus, whereas in the neonate, these TRE sites may be involved in the activation of Pcp-2 by binding other TR-TRAP-activating complexes.

Thermogenesis and thyroid function.

The past 10 years have seen tremendous progress in the definition of the nuclear mechanism of action of thyroid hormones. Although the way in which these nuclear mechanisms underlie the 3,5,3'-triiodo-L-thyronine (T3)-dependent stimulation of metabolic rate remains to be clarified, evidence favoring non-nuclear pathways is limited. Clearly, T3 stimulates both the production and consumption of energy within cells. It also exerts a number of parallel effects that result in increased oxygen consumption, e.g. on mitochondrial structure and composition; on the metabolism of lipids, carbohydrates, and proteins, and on cardiac function. Additionally, T3 may increase the proton permeability of the inner mitochondrial membrane, which implies that it may decrease the efficiency of energy production. These metabolic effects of T3 appear to be restricted to homeothermic-animals, representing a coordinated response to the challenge of maintaining body temperature.

Immunofluorescent localization of thyroid hormone receptor isoforms in glial cells of rat brain.

The three currently recognized T3 binding thyroid hormone receptor (TR) isoforms, TR alpha 1, TR beta 1, and TR beta 2, arise from two distinct genes (alpha and beta), whereas two closely related non-T3-binding receptor variants, collectively designated TR alpha 2, arise from alternate splicing of the alpha gene transcript. Using a panel of specific antisera to these isoforms we have assessed the presence or absence of TRs in oligodendrocytes and astrocytes of rat cerebrum and cerebellum. Inferences as to colocalization of the receptor isoforms and cell-specific marker proteins were based on immunohistochemical analysis of the differential emissions of paired immunofluorescent probes. Antisera against myelin basic protein (MBP) identified oligodendroglia, and glial fibrillary acidic protein identified astrocytes. MBP-positive oligodendrocytes displayed positive fluorescent signals with each of the three TR isoform-specific antisera and the antiserum to the receptor variants. These findings are consistent with the concept that the MBP gene is a direct target for thyroid hormone action. TR immunoreactivity appeared to localize primarily to the nuclei of these cells. In contrast, we observed no immunofluorescent signals for any of the TR isoforms in glial fibrillary acidic protein-positive astrocytes. These findings raise the possibility that any effect of thyroid hormone on astrocyte function and structure is mediated indirectly as a result of interaction of thyroid hormone with receptors situated in nonastrocyte cells or as a result of nonnuclear mechanisms.

Widespread distribution of immunoreactive thyroid hormone beta 2 receptor (TR beta 2) in the nuclei of extrapituitary rat tissues.

Messenger RNA for thyroid hormone receptor (TR) isoforms alpha 1 and beta 1 are widely distributed in rat tissues. Until recently, TR beta 2 mRNA was believed to be limited to the pituitary and the assumption was made that TR beta 2 protein was similarly restricted. We determined the distribution of TR beta 2 protein in selected adult and fetal rat tissues using three anti-TR beta 2 antisera directed to different amino acid sequences of the distinctive A/B domain of TR beta 2. The proportion of total nuclear binding capacity cleared by each antiserum was determined by saturation analysis. 10-20% of total binding capacity in adult brain, liver, kidney, and heart was immunoprecipitated by each antiserum. Use of specific antibodies to TR beta 1 and TR alpha 1 showed these isoforms accounted for the remainder of total T3 binding. Fetal liver and brain, however, contained only TR alpha 1. Immunohistochemical analysis of the adult tissues showed TR beta 2 present in nuclei. Reverse transcription polymerase chain reaction detected low levels of TR beta 2 mRNA in the adult tissues. We infer that TR beta 2 accounts for a significant fraction of TR in adult rat tissues despite the low levels of its mRNA.

Kinetic model of the response of precursor and mature rat hepatic mRNA-S14 to thyroid hormone.

We found in preliminary experiments that multiple daily injections of triiodothyronine (T3) resulted in an apparent prolongation in the half time (t1/2) of mRNA-S14 decay. To appropriately interpret these observations, we developed a mathematical model of the fluctuations of mRNA-S14 and its nuclear precursor after a single injection or multiple daily injections of T3. The model parameters include 1) the effect of plasma protein binding and metabolic clearance rates on receptor-bound nuclear T3, 2) the threefold circadian variation in mRNA-S14, 3) a 12-min t1/2 for the nuclear precursor and a 1.5-h t1/2 for the mature mRNA-S14, 4) previously derived relationships between the level of plasma T3 and nuclear occupancy, and 5) direct proportionality between nuclear transcription of the S14 gene and T3 nuclear occupancy. The model faithfully predicted the excursions of the mature mRNA-S14 and its nuclear precursor. Nuclear retention of T3 and the effects of circadian variation on S14 gene transcription explain the apparent prolongation in the t1/2 of decay of mature mRNA. Our findings illustrate the feasibility of incorporating parameters at the molecular level into a comprehensive kinetic analysis of hormone action.

Identification of thyroid hormone response elements in rodent Pcp-2, a developmentally regulated gene of cerebellar Purkinje cells.

In a previous study, we have shown that in vivo expression of the cerebellar Purkinje cell-specific gene Pcp-2 is regulated by thyroid hormone (T3) during neonatal development. In addition, transient cotransfection studies using thyroid hormone receptors (TRs) and a Pcp-2-lacZ construct pointed to direct regulation of Pcp-2 gene expression by T3. Therefore, we have initiated the following series of studies to define more precisely the location of the thyroid hormone regulatory elements in the Pcp-2 gene. By transfection and in vitro receptor binding analyses, we have identified two thyroid hormone response elements, A1 (-295/-268) and B1 (+207/+227). A1 contains a central half-site flanked by two similar half-sites. B1 contains two pairs of alternate half-sites. When these elements were ligated to the modified mouse mammary tumor virus promoter (delta MMTV), both induced a 8-14-fold expression of the reporter gene, but only in the presence of T3. Gel mobility assays demonstrated that both A1 and B1 bind TRs in the presence of thyroid hormone receptor auxiliary proteins or the retinoid X beta receptor. Mutations of the G residues to T within the individual half-site sequences of A1 caused a variable decrease in the transactivation of the MMTV-CAT construct and a corresponding reduction in TR binding in vitro. Thus, mutational analysis of A1 pointed to the interaction of the flanking half-site motifs with the central AGGTCA half-site. Interestingly, lengthening of the A1 sequence at its 3'-end caused a progressive dampening of the T3 response. The results suggest that the neighboring sequence may function as a silencer of the A1 element. Since thyroid hormone regulation of Pcp-2 is manifest only during the first 2 weeks after birth, we hypothesize that A1 and B1 act as T3-dependent response elements operative only during early neonatal Purkinje cell development and that their function is suppressed by a neighboring silencer element operative when expression of Pcp-2 becomes hormone-independent.