Effects of 2-iodohexadecanal in the physiology of thyroid cells

Iodide has direct effects on thyroid function. Several iodinated lipids are biosynthesized by the thyroid and they were postulated as intermediaries in the action of iodide. Among them, 2-iodohexadecanal (2-IHDA) has been identified and proposed to play a role in thyroid autoregulation. The aim of this study was to compare the effect of iodide and 2-IHDA on thyroid cell physiology. For this purpose, FRTL-5 thyroid cells were incubated with the two compounds during 24 or 48 h and several thyroid parameters were evaluated such as: iodide uptake, intracellular calcium and H₂O₂ levels. To further explore the molecular mechanism involved in 2-IHDA action, transcript and protein levels of genes involved in thyroid hormone biosynthesis, as well as the transcriptional expression of these genes were evaluated in the presence of iodide and 2-IHDA. The results obtained indicate that 2-IHDA reproduces the action of excess iodide on the "Wolff-Chaikoff" effect as well as on thyroid specific genes transcription supporting its role in thyroid autoregulation.

Functional inactivation of thyroid transcription factor-1 in PCCl3 thyroid cells

Thyroid transcription factor-1 (TTF-1) is a key regulator of thyroid development and function. In order to identify the genes whose expression depends on TTF-1 transcriptional activity within the thyrocyte we analyzed the consequence of the functional inactivation of this factor in PCCl3 cells. The expression of a fusion protein composed of the DNA binding domain of TTF-1 and of the strong repressive domain of the engrailed protein resulted in a dramatic loss of epithelial cell morphology and in proliferation arrest. These changes were reversed when the inhibition of endogenous TTF-1 was relieved. No change was observed when a similar fusion protein containing point mutations abolishing DNA binding activity was produced in the cells. Besides the expected down-regulation of expression of the main genes linked to the differentiated thyroid function, we observed a decreased expression of the transcription factors Hhex, Pax 8 and TTF-2 and of E-cadherin. By contrast, both ThOX-1 and DUOXA-1 genes were up-regulated, as well as the ones encoding vimentin and several proteins involved in cell cycle arrest. Our data thus extend the known roles of TTF-1 in thyroid development and in the expression of differentiated function in the adult organ to the control of epithelial morphology and of cell division in mature thyrocytes.

Delimitation and functional characterization of the bidirectional THOX-DUOXA promoter regions in thyrocytes

The THOX and DUOXA genes encode components of the oxidative machinery involved in thyroid hormone biosynthesis. Both of these genes are duplicated in mammalian genomes and are positioned in a head-to-head configuration, THOX1 facing DUOXA1 and THOX2 facing DUOXA2, respectively. The intergenic regions in both couples of genes exhibit dissimilar compositions, being highly GC-rich in the case of THOX1-DUOXA1 but not in the other case. In this study we localized precisely the transcription starts of all four genes using the RLM-RACE technique. It revealed that the distance between THOX1 and DUOXA1 transcription units is of about 70bp only, whereas THOX2 and DUOXA2 transcription starts are separated by 170bp. Analysis of these putative promoter regions revealed the presence of several potential binding sites for transcription factor Sp1 within the THOX1-DUOXA1 intergenic space, and of a TATA box and an Inr element in front of DUOXA2 and THOX2 genes, respectively. The putative promoter regions were inserted into a specifically designed vector harbouring two distinct reporter genes facing each other and their activity was investigated in transient transfection experiments in rat thyroid PCCl3 cells. Both regions exhibited bidirectional promoter activity in the assay. Gel shift experiments using extracts obtained from PCCl3 cells demonstrated the existence of at least one functional Sp1 binding site within the THOX1-DUOXA1 promoter. When Sp1 binding was abolished by mutation of the DNA sequence, a clear reduction in promoter activity in both THOX1 and DUOXA1 directions was observed in the functional assay. As these promoter sequences are well conserved in mammalian genomes, it appears very likely that the results we obtained here in the rat may be extended to the other species.

Identification of a short basic peptide motif able to drive copy-number dependent nuclear accumulation of a linked protein

The repetitive [RTRG](6) peptide was fortuitously identified as a potent nuclear localization signal when linked to the green fluorescent reporter protein. Replacing the arginines by lysines, or the threonines by glycines, both resulted in a decreased nuclear targeting ability of the peptide within this context. By contrast, the sequence [RT](12) proved able to drive nuclear accumulation of the linked protein as efficiently as the starting peptide. Remarkably, [RTRG](n) peptides where n=2 to 6 showed a gradual, copy-number dependent, increase in their ability to target the green fluorescent protein to the cell nucleus. As a consequence, the nuclear to cytoplasmic concentration ratio of the linked protein within the cell could be adjusted to different values depending on the number of repeats used in the fusion. Our observation may open the way to the use of [RTRG](n) repeats of given lengths (n=2 to 6) for fixing the nuclear-cytoplasmic partition of shuttling protein domains in the course of their functional study.

Human Thyroid Oxidases genes promoter activity in thyrocytes does not appear to be functionally dependent on Thyroid Transcription Factor-1 or Pax8

Thyroid Oxidases (ThOX/DUOX) genes encode proteins that are thought to play a crucial role in the biosynthesis of thyroid hormone by providing the oxidizing agent required to allow the organification of iodine. The expression of these genes is not restricted to the thyroid, but the corresponding mRNAs are found in the thyrocyte more abundantly than in several other cell types. It raises the question whether the same transcription factors, namely Thyroid Transcription Factor-1 (TTF-1) and Pax8, that control the expression of other genes involved in the differentiated thyroid function, also regulate ThOX/DUOX gene transcription in the thyrocyte. We set up a functional co-transfection assay in which fusion proteins composed of the DNA-binding domain of either TTF-1 or Pax8 fused to the repressive domain of the drosophila engrailed protein were used to competitively counteract the activity of endogenous TTF-1 or Pax8 factor in the differentiated thyroid cell line PCCl3. Contrary to the Thyroglobulin or Thyroid Peroxidase promoter, the known regulatory elements of the human ThOX/DUOX genes displayed no reduction in transcriptional activity when either TTF-1 or Pax8 competitor was produced in the cell, indicating that the presently characterized control elements of human ThOX/DUOX genes are not responsive to these thyroid-specific transcription factors.

Sequences downstream of the bHLH domain of the Xenopus hairy-related transcription factor-1 act as an extended dimerization domain that contributes to the selection of the partners

XHRT1 is a member of the HRT/Hey protein subfamily that are known as Notch effectors. XHRT1 is expressed in the developing floor plate and encodes a basic helix-loop-helix (bHLH) transcription repressor. Here, we show that XHRT1 misexpression in the neural plate inhibits differentiation of neural precursor cells and thus may be important for floor plate cells to prevent them from adopting a neuronal fate. Deletion analysis indicated that inhibition of differentiation by XHRT1 requires the DNA-binding bHLH motif and either the Orange domain or the C-terminal region. XHRT1 could efficiently homodimerize and heterodimerize with hairy proteins. Among those hairy genes, Xhairy2b shows extensive overlap of expression with XHRT1 in floor plate precursors and may be a biologically relevant XHRT1 partner. Dimerization is mediated through both the bHLH and downstream sequences, the Orange domain being particularly important for the efficiency of the interaction. Using chimeric constructs between XHRT1 and the ESR9 bHLH-O protein that does not interact with Xhairy1 and Xhairy2b, we found that both the bHLH domain and downstream sequences of XHRT1 were required for heterodimerization with Xhairy2b, while only the XHRT1 sequences downstream of the Orange domain are required for the interaction with Xhairy1. Together, these results suggest that XHRT1 plays a role in floor plate cell development and highlight the importance of the Orange and downstream sequences in dimerization and in the selection of the bHLH partners.

Transcriptional repression by the bHLH-Orange factor XHRT1 does not involve the C-terminal YRPW motif

Hairy-related transcription factors (HRTs) constitute a recently identified subfamily of basic-helix-loop-helix transcription factors containing an Orange domain (bHLH-O factors). As compared to the related HES proteins, HRTs exhibit distinct DNA-binding activities in vitro and the molecular mechanisms underlying their transcriptional activity remain poorly understood. We have identified here the sequence "ggCACGTGcc" as predominant binding site for Xenopus HRT1 (XHRT1). In transiently transfected 3T3 cells, XHRT1 represses the expression of a luciferase reporter gene under the control of multimerized XHRT1 binding sites. Deletion analysis indicated that repression by XHRT1 requires the presence of the DNA-binding bHLH motif and the Orange domain. However, the presence of the sequence motif YRPWGTEIGAF located at the very C-terminus of XHRT1 is dispensable. Accordingly, the groucho co-repressor, which is known to mediate transcriptional repression by HES factors through binding their C-terminal WRPW sequence, does not recognize the related YRPW motif present in the C-terminal part of XHRT1 significantly in vitro. As the C-terminus of HRTs is well conserved, our observation indicates that this part of HRTs, unlike the corresponding part of HES proteins, does not recruit the groucho co-repressor efficiently.

TM4SF10 gene sequencing in XLMR patients identifies common polymorphisms but no disease-associated mutation

Background

The TM4SF10 gene encodes a putative four-transmembrane domains protein of unknown function termed Brain Cell Membrane Protein 1 (BCMP1), and is abundantly expressed in the brain. This gene is located on the short arm of human chromosome X at p21.1. The hypothesis that mutations in the TM4SF10 gene are associated with impaired brain function was investigated by sequencing the gene in individuals with hereditary X-linked mental retardation (XLMR).

Methods

The coding region (543 bp) of TM4SF10, including intronic junctions, and the long 3' untranslated region (3 233 bp), that has been conserved during evolution, were sequenced in 16 male XLMR patients from 14 unrelated families with definite, or suggestive, linkage to the TM4SF10 gene locus, and in 5 normal males.

Results

Five sequence changes were identified but none was found to be associated with the disease. Two of these changes correspond to previously known SNPs, while three other were novel SNPs in the TM4SF10 gene.

Conclusion

We have investigated the majority of the known MRX families linked to the TM4SF10 gene region. In the absence of mutations detected, our study indicates that alterations of TM4SF10 are not a frequent cause of XLMR.

Characterization of a novel loss of function mutation of PAX8 in a familial case of congenital hypothyroidism with in-place, normal-sized thyroid

Thyroid dysgenesis is the most common cause of congenital hypothyroidism, a relatively frequent disease affecting 1 in 3000-4000 newborns. Whereas most cases are sporadic, mutations in transcription factors implicated in thyroid development have been shown to cause a minority of cases transmitted as monogenic Mendelian diseases. PAX8 is one of these transcription factors, and so far, five mutations have been identified in its paired domain in patients with thyroid dysgenesis. We have identified a novel mutation of PAX8, in the heterozygous state, in a father and his two children both presenting with congenital hypothyroidism associated with an in-place thyroid of normal size at birth. In addition, one of the affected siblings displayed unilateral kidney agenesis. The mutation substitutes a highly conserved serine in position 54 of the DNA-binding domain of the protein (S54G mutation) by a glycine. Functional analyses of the mutant protein (PAX8-S54G) demonstrated that it is unable to bind a specific cis-element of the thyroperoxidase gene promoter in EMSAs and that it has almost completely lost the ability to act in synergy with Titf1 to transactivate transcription from the thyroglobulin promoter/enhancer. These results indicate that loss of function mutations of the PAX8 gene may cause congenital hypothyroidism in the absence of thyroid hypoplasia.

The control of thyroid-specific gene expression: what exactly have we learned as yet?

The identification of transcription factors TTF 1 and Pax 8 and the demonstration of their pivotal role in thyroid development and in the control of thyroid-specific gene expression, although representing remarkable openings in our understanding of cell-specific transcription in the thyroid, still leave a lot of open questions. Recent work investigating the development of thyroid-specific gene expression in transgenic mouse models, now reveal that some basic assumptions have to be reconsidered also. Altogether, currently available data indicate that the regulatory machinery undergoes significant changes during thyroid organogenesis and confirm the existence of still unknown factors whose roles appear at least as critical as the ones played by TTF 1 and Pax 8 in the control of specific gene expression.

Identification of BOIP, a novel cDNA highly expressed during spermatogenesis that encodes a protein interacting with the orange domain of the hairy-related transcription factor HRT1/Hey1 in Xenopus and mouse

Hairy-related transcription factor (HRT/Hey) genes encode a novel subfamily of basic helix-loop-helix (bHLH) transcription factors related to the Drosophila hairy and Enhancer-of-split (E(spl)) and the mammalian HES proteins that function as downstream mediators of Notch signaling. Using the yeast two-hybrid approach, a previously uncharacterized protein was identified in Xenopus that interacts with XHRT1 (originally referred to as bc8), one member of the HRT/Hey subclass. This protein is evolutionarily conserved in chordates. It binds to sequences adjacent to the bHLH domain of XHRT1 known as the Orange domain and has been named bc8 Orange interacting protein (BOIP). BOIP shows a rather uniform subcellular localization and is recruited to the nucleus upon binding to XHRT1. In Xenopus, XBOIP mRNA is detected by RNase protection analysis throughout embryogenesis. In the adult, the strongest expression is detected in testis. In the mouse, high levels of BOIP mRNA are also found in adult testis. No expression is detected in the embryo and in any of the other adult organs tested. In situ hybridization revealed that BOIP transcripts were detected almost exclusively in round spermatids and that this expression overlaps with that of Hey1 (HRT1), which is expressed throughout spermatogenesis. In view of the importance of the Orange domain for HRT/Hey function, the newly identified BOIP proteins may serve as regulators specifically of HRT1/Hey1 activity.

Structural and functional characterization of the two human ThOX/Duox genes and their 5'-flanking regions

A crucial step in thyroid hormone synthesis is the oxidative coupling of iodide to thyroglobulin that is catalyzed by thyroperoxidase. The limiting factor of this reaction is the supply of hydrogen peroxide. The generation of hydrogen peroxide has been linked to an enzymatic system located at the apical pole of thyrocytes. This enzymatic activity is assumed to be exerted by NADPH oxidases encoded by two recently cloned genes hThOX1 and hThOX2. Both genes are expressed at high levels in thyrocytes. In this study we report the chromosomal organization of these two genes and the functional characterization of their respective promoter regions. The two human ThOX genes are arranged in a head to head configuration and are separated by a 16 kb-long region. Human ThOX1 and ThOX2 genes span 75 kb and are composed of 35 and 34 exons, respectively. The promoters of both genes do not resemble each other and differ from promoters of other known thyroid-specific genes. No TATA box is present in either ThOX gene promoter. Functional studies confirm that both promoters display significant transcriptional activities after transfection in differentiated thyroid cell lines. However, in contrast to that of thyroglobulin or Na(+)/I(-) symporter gene promoter, hThOX promoter activity is not restricted to thyroid cells. Additionally, functional studies show that both hThOX promoters are not positively controlled by cAMP.

XHRT-1, a hairy and Enhancer of split related gene with expression in floor plate and hypochord during early Xenopus embryogenesis

We have isolated a Xenopus homologue of the mammalian hairy and Enhancer of split related gene HRT1. XHRT1 expression in late gastrula and early neurula embryos is restricted to two stripes of cells in the medial neural plate and in dorsal endodermal cells. At later stages, XHRT1 is expressed in the floor plate, in hypochord cells and in the somitogenic and anterior presomitic mesoderm. By tailbud stage, XHRT1 is also highly expressed in the dorsal hindbrain, telencephalon and eye vesicles, olfactory placodes, pronephros, branchial arches and tail fin. We also show that XHRT1 expression in medial neural cells is induced by Notch signaling and that there are differences in the way XHRT1 and other H/E(spl) genes are regulated.

Suboptimal enhancer sequences are required for efficient bovine leukemia virus propagation in vivo: implications for viral latency

Repression of viral expression is a major strategy developed by retroviruses to escape from the host immune response. The absence of viral proteins (or derived peptides) at the surface of an infected cell does not permit the establishment of an efficient immune attack. Such a strategy appears to have been adopted by animal oncoviruses such as bovine leukemia virus (BLV) and human T-cell leukemia virus (HTLV). In BLV-infected animals, only a small fraction of the infected lymphocytes (between 1 in 5,000 and 1 in 50,000) express large amounts of viral proteins; the vast majority of the proviruses are repressed at the transcriptional level. Induction of BLV transcription involves the interaction of the virus-encoded Tax protein with the CREB/ATF factors; the resulting complex is able to interact with three 21-bp Tax-responsive elements (TxRE) located in the 5' long terminal repeat (5' LTR). These TxRE contain cyclic AMP-responsive elements (CRE), but, remarkably, the "TGACGTCA" consensus is never strictly conserved in any viral strain (e.g.,AGACGTCA, TGACGGCA, TGACCTCA). To assess the role of these suboptimal CREs, we introduced a perfect consensus sequence within the TxRE and showed by gel retardation assays that the binding efficiency of the CREB/ATF proteins was increased. However, trans-activation of a luciferase-based reporter by Tax was not affected in transient transfection assays. Still, in the absence of Tax, the basal promoter activity of the mutated LTR was increased as much as 20-fold. In contrast, mutation of other regulatory elements within the LTR (the E box, NF-kappa B, and glucocorticoid- or interferon-responsive sites [GRE or IRF]) did not induce a similar alteration of the basal transcription levels. To evaluate the biological relevance of these observations made in vitro, the mutations were introduced into an infectious BLV molecular clone. After injection into sheep, it appeared that all the recombinants were infectious in vivo and did not revert into a wild-type virus. All of them, except one, propagated at wild-type levels, indicating that viral spread was not affected by the mutation. The sole exception was the CRE mutant; proviral loads were drastically reduced in sheep infected with this type of virus. We conclude that a series of sites (NF-kappa B, IRF, GRE, and the E box) are not required for efficient viral spread in the sheep model, although mutation of some of these motifs might induce a minor phenotype during transient transfection assays in vitro. Remarkably, a provirus (pBLV-Delta 21-bp) harboring only two TxRE was infectious and propagated at wild-type levels. And, most importantly, reconstitution of a consensus CRE, within the 21-bp enhancers increases binding of CREB/ATF proteins but abrogates basal repression of LTR-directed transcription in vitro. Suboptimal CREs are, however, essential for efficient viral spread within infected sheep, although these sites are dispensable for infectivity. These results suggest an evolutionary selection of suboptimal CREs that repress viral expression with escape from the host immune response. These observations, which were obtained in an animal model for HTLV-1, are of interest for oncovirus-induced pathogenesis in humans.

Identification of the gene encoding Brain Cell Membrane Protein 1 (BCMP1), a putative four-transmembrane protein distantly related to the Peripheral Myelin Protein 22 / Epithelial Membrane Proteins and the Claudins

BACKGROUND

A partial cDNA clone from dog thyroid presenting a very significant similarity with an uncharacterized mouse EST sequence was isolated fortuitously. We report here the identification of the complete mRNA and of the gene, the product of which was termed "brain cell membrane protein 1" (BCMP1).

RESULTS

The 4 kb-long mRNA sequence exhibited an open-reading frame of only 543 b followed by a 3.2 kb-long 3' untranslated region containing several AUUUA instability motifs. Analysis of the encoded protein sequence identified the presence of four putative transmembrane domains. Similarity searches in protein domain databases identified partial sequence conservations with peripheral myelin protein 22 (PMP22)/ epithelial membrane proteins (EMPs) and Claudins, defining the encoded protein as representative of the existence of a novel subclass in this protein family.Northern-blot analysis of the expression of the corresponding mRNA in adult dog tissues revealed the presence of a huge amount of the 4 kb transcript in the brain. An EGFP-BCMP1 fusion protein expressed in transfected COS-7 cells exhibited a membranous localization as expected. The sequences encoding BCMP1 were assigned to chromosome X in dog, man and rat using radiation hybrid panels and were partly localized in the currently available human genome sequence.

CONCLUSIONS

We have identified the existence in several mammalian species of a gene encoding a putative four-transmembrane protein, BCMP1, wich defines a novel subclass in this family of proteins. In dog at least, the corresponding mRNA is highly present in brain cells. The chromosomal localization of the gene in man makes of it a likely candidate gene for X-linked mental retardation.

A method for the large-scale cloning of nuclear proteins and nuclear targeting sequences on a functional basis

We describe here a selection strategy allowing the cloning of sequences that contain a functional nuclear targeting signal. Our method relies on the use of green fluorescent protein fusion proteins to identify nuclear targeting sequences. Transfected cells expressing nuclear protein fusions were isolated on the basis of their nuclear fluorescence using flow cytometry and the transfected DNAs were recovered after bacterial transformation with total DNA from pools of sorted cells. Starting from a cDNA expression library, in which only 1% of the expressed proteins were nuclear, we obtained a 70-fold enrichment in nuclear protein-encoding clones after a single round of selection. Among the 63 clones that have been partially sequenced to date, 25 (40%) corresponded to known nuclear proteins and 13 (20%) to previously uncharacterized sequences. Despite their ability to target the green fluorescent protein marker to the cell nucleus, about half of the cloned sequences did not encode canonical basic or bipartite nuclear localization signals. The method can thus be applied to the large-scale cloning of functional nuclear targeting sequences, which opens the way to a wide investigation of nuclear import mechanisms and to the identification of previously unknown nuclear proteins.

Nuclear targeting of proteins: how many different signals?

The nuclear import of proteins into the cell nucleus involves the recognition of a nuclear localization signal sequence, borne by the protein to be transported, by complex molecules called importins, that will subsequently mediate the crossing over of the nuclear envelope. The most frequently encountered signal sequence is made up of short stretches of basic amino acid residues and is recognized by importins alpha and/or beta. Other signal sequences have been described, and some have been shown to mediate the association with importins other than importin alpha or beta. Recently, approaches have been developed that allow the cloning, on a functional basis, of sequences able to specify the nuclear localization of proteins. A variety of peptidic motifs of limited size which do not contain previously described signal sequences were isolated in such assays. It reveals that the spectrum of sequences that are able to target a protein to the cell nucleus may be wider than currently expected. It will probably also lead to the identification of novel target sequences for importins and will demonstrate the implication of additional members of this family of proteins in nuclear transport.

Critical residues of the homeodomain involved in contacting DNA bases also specify the nuclear accumulation of thyroid transcription factor-1

The N-terminal end of thyroid transcription factor-1 (TTF-1) homeodomain is composed of a stretch of five basic amino-acids that is conserved in both POU- and NK2-class homeodomains and constitutes a functional nuclear localization signal. By analyzing the cellular distribution of fusion proteins, composed of a jellyfish green fluorescent variant and different parts of TTF-1, we show here that the presence of this basic sequence is not sufficient by itself to confer complete nuclear accumulation. By mutagenesis, we identified a second region located in the center of the DNA recognition helix of the homeodomain that is also able to specify a predominantly nuclear localization of the chimeric proteins, independently of the presence of the basic NLS. The destruction, by mutagenesis, of both the basic stretch and the motif in the DNA recognition helix led to the total loss of nuclear accumulation, indicating that complete nuclear accumulation of TTF-1 results from the concerted action of these two proteic signals. Both of the regions of the homeodomain that are involved in nuclear targeting also encompass critical amino-acids responsible for DNA binding site recognition, as evidenced by the loss of DNA binding activity in vitro upon mutagenesis. Specifically, residues in the central part of the DNA recognition helix are involved in contacting bases in the major groove of DNA and are the most conserved in homeodomain proteins, suggesting that this part of the homeodomain could play a general role in the nuclear localization of members of this family of proteins.

A canonical nerve growth factor-induced gene-B response element appears not to be involved in the cyclic adenosine monophosphate-dependent expression of differentiation in thyrocytes

The expression of transcriptionally active nerve growth factor-induced gene-B (NGFI-B) is rapidly induced in thyroid follicular cells in response to cAMP stimulation. As the transcription of thyrocyte-specific genes is controlled by the cAMP cascade, we have investigated a possible involvement of NGFI-B in this control. Recombinant adenoviruses driving the expression of either the intact NGFI-B protein or a truncated form of it that lacks the capacity to transactivate a NBRE-dependent promoter, were used to infect dog thyrocytes maintained in primary culture. Northern blot analysis of total RNA from infected cells revealed that the expression of NGFI-B was not sufficient to induce a significant accumulation of specific transcripts (thyroglobulin, thyroperoxidase, sodium-iodide symporter) in unstimulated thyrocytes. The overproduction of the transcriptionally inactive form of NGFI-B in thyrocytes maintained in the presence of forskolin after infection did not impair the accumulation of the thyroid-specific transcripts. These data show that NGFI-B does not control the expression of differentiation in thyrocytes by acting through a canonical NBRE. As a consequence, we must consider that either the expression of NGFI-B in cAMP-stimulated thyrocytes is not critically linked to the expression of differentiation or that NGFI-B is implicated in a regulatory mechanism which differs from its known action at the level of a NBRE.

Two binding sites for thyroid transcription factor 1 (TTF-1) determine the activity of the bovine thyroglobulin gene upstream enhancer element

A thyroid-specific enhancer element located upstream from the bovine thyroglobulin gene had been shown to contain three contiguous regions that are protected by thyroid transcription factor 1 (TTF-1) in footprinting experiments in vitro. The functional relevance of the individual TTF-1 binding sites was investigated in a transient assay in primary cultured thyrocytes. Using reporter constructs containing synthetic oligonucleotides overlapping the protected sequences we were able to show that only two out of the three TTF-1 binding sites exhibit transcription enhancing activity. Within the context of the complete enhancer sequence, the central 'inactive' TTF-1 site could be deleted whithout any consequence on the activity of the enhancer in the assay, whereas the presence of both terminal 'active' TTF-1 sites had previously been shown to be strictly required for enhancer function. Our results thus show that the activity of the bovine thyroglobulin upstream enhancer relies on the presence of a pair of TTF-1 binding sites separated by about 30 bp. These results also emphasize the need to assess experimentally the functional relevance of TTF-1 binding sites identified in footprinting experiments.

TTF-2 does not appear to be a key mediator of the effect of cyclic AMP on thyroglobulin gene transcription in primary cultured dog thyrocytes

TTF-2 is a thyroid-specific winged-helix transcription factor which has been proposed to play a key role in the hormonal control of thyroglobulin and thyroperoxidase genes transcription in FRTL-5 cells. We have analyzed TTF-2 DNA-binding activity in primary cultures of dog thyrocytes maintained in control condition or in the presence of the cAMP agonist forskolin. Binding of 35S-labelled nuclear proteins to the TTF-2 recognition sequence identified the presence of two molecular species of 41.5 and 42.5 kDa. TTF-2 DNA-binding activity was clearly detectable in nuclear extracts from unstimulated cells and appeared increased in forskolin-treated cells. Thus, the presence of TTF-2 DNA-binding activity does not correlate with the cAMP-dependent activity of thyroglobulin and thyroperoxidase genes in this cell system. In addition, the mutation of the TTF-2 binding site in the thyroglobulin promoter resulted in a very reduced but still clearly cAMP-dependent promoter activity when assayed by transient expression in the same cells. These results do not support a dominant role for TTF-2 in the cAMP-dependent control of thyroglobulin gene transcription in primary cultured thyrocytes.

Production of dog calcyphosine in bacteria and lack of phosphorylation by the catalytic subunit of protein kinase A in vitro

Calcyphosine is a calcium-binding protein containing four EF-hand domains that is found in several epithelia and in some cells of the central nervous system. In thyroid follicular cells, calcyphosine is synthesized and phosphorylated in response to stimulation by thyrotropin and cAMP agonists. The cDNA coding for dog calcyphosine has been expressed in bacteria under the control of the T7 promoter. Recombinant calcyphosine was purified from crude bacterial lysates by a combination of anion-exchange and hydrophobic interaction chromatography. Phosphorylation assays using the purified catalytic subunit of protein kinase A and the recombinant or the native calcyphosine revealed that, contrary to a previous report, calcyphosine is not significantly phosphorylated by this enzyme in vitro.

Cloning and sequence analysis of human calcyphosine complementary DNA

Calcyphosine, initially identified as thyroid protein p24, is a calcium-binding protein containing four EF-hand domains. It was first cloned and characterized in the dog and corresponds to R2D5 antigen in rabbit. Using the canine calcyphosine cDNA sequence as a probe, we have isolated its human counterpart from a thyroid cDNA library. The two sequences display a high degree of conservation, both at nucleotide and deduced amino acid levels. Sequence comparison with other proteins showed that the closest homologue of calcyphosine is the crustacean CCBP-23 protein. Northern blot analysis revealed that calcyphosine messenger RNA is much less abundant in human than in canine thyrocytes. Western blot experiments indicated that the amount of protein is also dramatically reduced in man compared to dog.

A zinc-dependent DNA-binding activity co-operates with cAMP-responsive-element-binding protein to activate the human thyroglobulin enhancer

Footprinting experiments involving the human thyroglobulin gene enhancer and thyroid nuclear extracts revealed a protected region called X2, containing an incomplete cAMP-responsive element (CRE). Band-shift experiments identified two binding activities recognizing the X2 element: a CRE-binding protein (CREB)/activating transcription factor (ATF) relative that binds the half CRE motif and a second factor that interacts with a G-rich motif located just upstream from the CRE. The first factor appears to be CREB itself, as indicated by the supershifting when using an antibody directed against CREB, and the second DNA-binding activity involved was shown to be zinc-dependent and exhibited an apparent molecular mass of 42-44 kDa in South-Western blotting experiments. This factor may represent a novel entity, which we named CAF, for 'CREB Associated Factor'. Three copies of X2 sequence conferred a strong cAMP-dependent transcriptional activation to a heterologous promoter in transient transfection assay in cAMP-stimulated primary thyrocytes and HeLa cells. Transfection experiments of constructs containing the X2 element mutated in either the CRE or the G-rich site showed that both motifs were required for this transcription activating function. Moreover, the combination of several individual X2 elements mutated in either the CRE or the G-rich motif did not exhibit full transcriptional activity. This suggests that, in the context of the X2 element, CREB requires a close interaction with CAF to achieve both basal and cAMP-dependent transcriptional activation.

Expression of a transactivation-deficient form of thyroid transcription factor I decreases the activity of co-transfected thyroglobulin and thyroperoxidase promoters

Thyroid transcription factor I (TTF-1) plays a critical role in thyroid organogenesis and in the control of expression of several thyroid-specific genes, like those coding for thyroglobulin and thyroperoxidase. We have expressed the isolated DNA-binding homeodomain of TTF-1 in cultured thyroid cells by transient transfection. A specific reduction in the activity of co-transfected thyroglobulin and thyroperoxidase promoters was observed in the presence of the isolated TTF-1 homeodomain, as compared to their activity measured in the presence of a mutated homeodomain unable to bind DNA. The activity of the SV40 early promoter, used as a control, was only marginally affected in these experiments. The transactivation-deficient form of TTF-1 described here may thus be used for investigating other cellular processes that are dependent on TTF-1 transcriptional activity.

Induction of nerve growth factor-induced gene-B (NGFI-B) as an early event in the cyclic adenosine monophosphate response of dog thyrocytes in primary culture

We investigated the induction of nerve growth factor-induced gene-B (NGFI-B) in dog thyrocytes in primary culture stimulated by different agents. The dog NGFI-B complementary DNA (cDNA) was cloned from a cDNA library of dog thyrocytes and used to study, by Northern blotting, the level of NGFI-B messenger RNA (mRNA) in those cells. We have shown that TSH and forskolin, which both induce proliferation and differentiation of the thyroid cells by activation of the protein kinase A pathway, lead to a strong and transient expression of two NGFI-B mRNA species, which differ in the length of the poly(A) tail. In contrast, 12-O-tetradecanoyl-13-phorbol-acetate (TPA) and epidermal growth factor, which induce proliferation and dedifferentiation of those cells by activation of the protein kinase C and the protein tyrosine kinase cascade, respectively, lead to a weaker expression of NGFI-B mRNA. In parallel, we studied the transactivation capacity of NGFI-B in the same cell system by transient transfection of a chloramphenicol acetyl transferase reporter construction containing a NGFI-B-dependent synthetic promoter. The highest transactivation was observed after forskolin stimulation, whereas transactivation after TPA stimulation was weak and no significant transactivation was observed after epidermal growth factor stimulation. Taken together, these results show that NGFI-B is an immediate early gene product that is mainly induced by the cAMP-dependent pathway in dog thyrocytes. Moreover they suggest that NGFI-B expression could be one of the early transcriptional changes induced specifically by this cascade and leading to differentiation and/or proliferation of these cells.

Identification of a thyroid-specific and cAMP-responsive enhancer in the upstream sequences of the human thyroglobulin promoter

Functional analysis of remote 5'-flanking sequences from the human thyroglobulin gene in primary cultured dog thyrocytes led to the identification of a partly cAMP-responsive enhancer, located between -3.6 to -2.2 kb from the transcriptional start site. Deletion analysis of the 1.4 kb-long region localised the enhancer activity in a 0.5 kb-long fragment (located between -3.2 and -2.7 kb relative to transcription start), which could be divided into two functional sub-fragments of 0.2 and 0.3 kb. A potential binding site for the CREB/ATF transcription factors was found in the 0.3 kb element. The complete enhancer region had no detectable activity when assayed in Hela cells, suggesting that it constituted a thyroid-specific regulatory element. Accordingly, footprinting experiments revealed the presence of several binding sites for Thyroid Transcription Factor-1 (TTF-1) in both the 0.2 and 0.3 kb elements.

Pax 8 expression in primary cultured dog thyrocyte is increased by cyclic AMP

Pax 8 proteins are paired domain-containing transcription factors expressed in thyroid, kidney, ovary, placenta and developing brain. Thyroglobulin (Tg) and thyroperoxidase (TPO) genes, which are specifically expressed in thyroid follicular cells, both harbor a Pax 8 binding site in their proximal promoter region. The transcription of these genes is, as is the expression of most of the other differentiated functions of the thyrocyte, positively regulated by thyrotropin (TSH) via a cyclic-AMP (cAMP)-dependent mechanism. However, no typical cAMP-responsive element has been detected in the promoter region of Tg and TPO genes. We therefore investigated whether Pax 8 activity itself could be regulated by cAMP, which would support a role for these factors in the cAMP-dependent expression of differentiation in thyroid cells. In this paper we show that the expression of Pax 8 mRNA and proteins are increased by treatment of the thyrocyte with forskolin. This suggests that Pax 8 could indeed participate in the mediation of the transcriptional activation of thyroid specific genes by cAMP. We also show that Pax 8 are nuclear phosphoproteins, although neither their phosphorylation, nor their nuclear translocation seem to be highly regulated by cAMP. During the course of this study, a new splicing variant of dog Pax 8, termed Pax 8g, has been isolated.

Study of TTF-1 gene expression in dog thyrocytes in primary culture

TTF-1 is a homeodomain-containing transcription factor mainly expressed in the thyroid where it controls the tissue-specific expression of the thyroglobulin, thyroperoxidase and TSH receptor genes. It is therefore potentially implicated in the hormonal control exerted by thyrotropin via the second messenger cyclic AMP on the transcription of these genes in thyrocytes. In order to investigate whether there exists a relationship between the stimulation of the cAMP pathway and TTF-1 gene expression in these cells, we have compared the amounts of TTF-1 protein, its state of phosphorylation and its subcellular distribution in control and cAMP-stimulated dog thyrocytes in primary culture. Dog TTF-1 was expressed in bacteria as a fusion protein and antibodies were raised against the dog TTF-1 moiety. Stimulation of the thyrocytes by cyclic AMP agonist only marginally increased TTF-1 gene expression as shown for the mRNA by RNase protection assay and for the protein by immunoblotting and immunoprecipitation of extracts from 35S-methionine labelled cells. The phosphorylation state of TTF-1 was investigated by immunoprecipitation of extracts from 32P-labelled thyrocytes. Phosphorylation level appeared to be essentially unaffected by forskolin treatment of the cells. We also looked for differences in the use of phosphorylation sites by partial proteolytic digestion of immunoprecipitated 32P-labelled TTF-1 with Glu-C and Asp-N endoproteases. Comparison of radioactivity distribution amongst the generated fragments did not reveal any difference in the pattern of TTF-1 phosphorylation in control and forskolin conditions. Lastly, in situ detection of TTF-1 by immunofluorescence demonstrated that the protein was localized in the nucleus of the cells, irrespective of the culture conditions. No major change in TTF-1 gene expression upon stimulation of the thyrocyte with a cAMP agonist could thus be detected in this study. The absence of an obvious modification of the TTF-1 protein itself in response to cAMP stimulation may indicate that other transcription factor(s) or co-factor(s) are involved in the control exerted by cAMP on the expression of thyroid-specific genes.

Unmethylated thyroglobulin promoter may be repressed by methylation of flanking DNA sequences

The thyroglobulin gene, like many other tissue-specific genes, appears to be specifically less methylated in the differentiated cell type where it is transcribed. The thyroglobulin gene promoter elements themselves are highly CG-deficient and do not contain any HpaII/MspI sites. In this study, using DNA constructs that were methylated in vitro with HpaII or MspI methylases, we show that DNA methylation of vector sequences is sufficient to repress the activity of the thyroglobulin gene promoter in transient transfection experiments. Reporter-gene expression from a plasmid containing only the proximal thyroglobulin gene promoter is sensitive to DNA methylation even in fully differentiated thyrocytes. Transcription from methylated plasmids containing the thyroglobulin gene enhancer and proximal promoter is also clearly reduced when the transfected cells are maintained under less-differentiated conditions. These results indicate that DNA methylation can influence, from a distance, the activity of an unmodified promoter. Our results also agree with the view that loss of DNA methylation does not constitute a prerequisite for thyroglobulin gene expression in differentiated thyrocytes, where the thyroglobulin gene enhancer and promoter are activated. However, the production of thyroglobulin transcripts could be severely impaired when this activation is not maximal, as is the case in less-differentiated cells or when the enhancer element is lacking. We suggest that DNA methylation helps to maintain the thyroglobulin gene in an inactive state unless all of the conditions required for its expression are fulfilled, and that the thyroid-specific demethylation events are a consequence of the activation state of the gene.

Human, bovine, canine and rat thyroglobulin promoter sequences display species-specific differences in an in vitro study

The proximal promoter regions of the thyroglobulin gene from man, beef, dog and rat were compared by transient expression in primary cultured dog thyrocytes. All four promoter regions were able to control properly the expression of a reporter gene in response to cyclic AMP stimulation. Surprisingly, despite extensive sequence conservation, the transcriptional activities of these four mammalian thyroglobulin promoters were differently affected by equivalent mutations. Homologous sequence elements from these promoter regions also exhibited distinct binding characteristics in mobility-shift experiments conducted in the presence of nuclear proteins from bovine thyroids. Our observations show that the highly conserved thyroglobulin promoters may exhibit unexpected functional differences in a specific assay and indicate that some of the molecular mechanisms involved in the control of thyroglobulin gene expression have evolved differently within mammals.

Identification of a transcriptional enhancer upstream from the bovine thyroglobulin gene

The DNA sequences corresponding to a DNaseI-hypersensitive region identified previously in bovine thyroglobulin gene chromatin (Hansen et al. (1988) Eur. J. Biochem. 178, 387-393) exhibited the properties of a transcriptional enhancer in a transient assay in primary cultured dog thyrocytes, but did not so in transfected HeLa cells. By contrast to the thyroglobulin proximal promoter, the enhancer element did not require cyclic AMP stimulation of the thyrocytes to be active. Using a bi-directional deletion approach, the minimal region displaying enhancer activity has been localized between positions -1906 and -1744 relative to the thyroglobulin gene transcription start. DNA-footprinting experiments revealed the presence of several binding sites for the thyroid-specific transcription factor TTF-1 within the enhancer sequence.

Functional role of TTF-1 binding sites in bovine thyroglobulin promoter

We have studied the binding of purified TTF-1 on the bovine thyroglobulin gene promoter. DNase I footprinting experiments revealed three binding sites which corresponded in location to the A, B and C sites found in the rat thyroglobulin promoter. Mutants in the A and C regions showing reduced binding of TTF-1, also exhibited largely decreased promoter activity in transient expression experiments in primary-cultured dog thyrocytes. Two mutants in the B site that exhibited a reduced capacity to bind TTF-1 also displayed a drastically affected transcriptional activity in transient assays. As in the rat, sites A and C only are critical for promoter activity, these results suggest that full occupancy of the B site is required for thyroglobulin promoter activity in the cow only.

Functional study of the human thyroid peroxidase gene promoter

Structure/function relationships in the human thyroid peroxidase gene promoter have been studied by deletion and mutation analyses and confronted with footprint patterns obtained with thyroid nuclear extracts and the purified thyroid transcription factor TTF-1. Crude nuclear extracts from dog thyroid primary cultures were shown to contain a binding activity recognizing the -119 to -105 segment of the promoter (coordinates relative to the transcriptional start site). Deletion, or site-directed mutagenesis of this segment dramatically reduced transcriptional activity in transient expression experiments on gene fusions of the thyroid peroxidase promoter and the growth hormone reporter. This binding activity was increased in nuclear extracts from thyrocytes cultured in the presence of the cAMP-agonist forskolin. A mutation that decreased the promoter function in forskolin-stimulated thyrocytes resulted in weakening of the corresponding footprint. The binding site displays no significant sequence similarities with known cAMP-responsive elements. Mutagenesis of another region of the promoter (-99 to -94) induced the binding of an additional factor, resulting in a dramatically enhanced promoter activity. We show that the thyroid-specific transcriptional factor TTF-1 is not directly involved in the above-mentioned interactions and provide evidence suggesting that, in spite of displaying a similar binding pattern to thyroperoxidase and thyroglobulin promoters in vitro, TTF-1 plays a less important role in the former. Altogether, our data delineate the minimal thyroid peroxidase gene promoter in the human and identify the binding sites of two trans-activating factors, one of them being potentially the mediator of a non-conventional cAMP control, independent of the cAMP-responsive element and factor AP-2.

The 5' region of the human thyroglobulin gene contains members of the Alu family

Repeated sequences were identified in the 5' region of the human Tg gene in introns 4, 5, 10, and 11. Another repeated cluster was located in the 5' flanking sequences, approximately 6 Kb upstream from the first exon. The nucleotide sequence analysis indicated that these repeated sequences are members of the Alu family. The homology between the sequences of the intron 4 and the Alu consensus was 86%. The Alu member studied was oriented in the direction of transcription of the Tg.

Isolation and characterization of the canine thyroglobulin gene promoter region

The 5' flanking sequences from the canine thyroglobulin gene were isolated by homology screening with the evolutionary conserved sequence from the bovine thyroglobulin promoter and sequenced. Transient expression in primary cultured dog thyrocytes demonstrated that the canine clone contains a functional promoter inducible by cAMP. DNAse I footprinting assays showed that the thyroid-specific transcription factor TTF-1, purified from bovine thyroid, also recognizes the canine thyroglobulin promoter. Similar footprints were obtained with crude nuclear extracts from primary cultured dog thyrocytes.

Cloning and sequence analysis of TFE, a helix-loop-helix transcription factor able to recognize the thyroglobulin gene promoter in vitro

A cDNA that encodes a transcription factor able to recognize the thyroglobulin gene promoter in vitro was isolated from a dog thyroid cDNA expression library in lambda gt11. The library was screened with a multimerized 20 bp-oligonucleotide probe corresponding to the -126 to -107 bp region of the bovine thyroglobulin gene promoter. The specificity of DNA sequence recognition was demonstrated by DNA binding experiments realized with beta-galactosidase-fusion protein immobilized on nitrocellulose filters and various unlabelled multimerized competing DNA fragments. The encoded protein, TFE, appears to be the canine counterpart of a recently cloned human transcription factor, ITF-2, that binds to the mu E5 kappa E2 motif found in both immunoglobulin heavy and light chains genes enhancers and belongs to the basic-Helix-Loop-Helix family of transcription factors. When TFE protein was produced in a rabbit reticulocyte lysate, it displayed the same specificity of DNA sequence recognition as the beta-galactosidase fusion protein and immobilization of the translation product on nitrocellulose still appeared to be essential for detecting in vitro DNA binding activity. Functional data failed to assign a role for TFE in the control of thyroglobulin gene transcription in vitro, suggesting that the selection of TFE clone resulted from the fortuitous presence of a high affinity binding site in the probe used for screening the expression library.

The thyroglobulin gene: evolutionary and regulatory issues

The coding information for thyroglobulin synthesis is contained in a large transcription unit, which is made from the juxtaposition of short repetitive gene modules and of a copy of an ancient acetylcholinesterase homologue. Minor alternatively spliced transcripts with conservation of the reading frame seem to be common. Whether they have a role or represent noise in the splicing phenomena is unknown. Transcription of the thyroglobulin gene is controlled by cAMP through a pathway involving sequence motifs and trans-acting factors that differ from those identified so far in other systems.

Differentiation expression during proliferative activity induced through different pathways: in situ hybridization study of thyroglobulin gene expression in thyroid epithelial cells

In canine thyrocytes in primary culture, our previous studies have identified three mitogenic agents and pathways: thyrotropin (TSH) acting through cyclic AMP (cAMP), EGF and its receptor tyrosine protein kinase, and the phorbol esters that stimulate protein kinase C. TSH enhances, while EGF and phorbol esters inhibit, the expression of differentiation. Given that growth and differentiation expression are often considered as mutually exclusive activities of the cells, it was conceivable that the differentiating action of TSH was restricted to noncycling (Go) cells, while the inhibition of the differentiation expression by EGF and phorbol esters only concerned proliferating cells. Therefore, the capacity to express the thyroglobulin (Tg) gene, the most prominent marker of differentiation in thyrocytes, was studied in proliferative cells (with insulin) and in quiescent cells (without insulin). Using cRNA in situ hybridization, we observed that TSH (and, to a lesser extent, insulin and insulin-like growth factor I) restored or maintained the expression of the Tg gene. Without these hormones, the Tg mRNA content became undetectable in most of the cells. EGF and 12-0-tetradecanoyl phorbol-13-acetate (TPA) inhibited the Tg mRNA accumulation induced by TSH (and/or insulin). Most of the cells (up to 90%) responded to both TSH and EGF. Nevertheless, the range of individual response was quite variable. The effects of TSH and EGF on differentiation expression were not dependent on insulin and can therefore be dissociated from their mitogenic effects. Cell cycling did not affect the induction of Tg gene. Indeed, the same cell distribution of Tg mRNA content was observed in quiescent cells stimulated by TSH alone, or in cells approximately 50% of which had performed one mitotic cycle in response to TSH + insulin. Moreover, after proliferation in "dedifferentiating" conditions (EGF + serum + insulin), thyrocytes had acquired a fusiform fibroblast-like morphology, and responded to TSH by regaining a characteristic epithelial shape and high Tg mRNA content. 32 h after the replacement of EGF by TSH, cells in mitosis presented the same distribution of the Tg mRNA content as the rest of the cell population. This implies that cell cycling (at least 27 h, as previously shown) did not affect the induction of the Tg gene which is clearly detectable after a time lag of at least 24 h. The data unequivocally show that the reexpression of differentiation and proliferative activity are separate but fully compatible processes when induced by cAMP in thyrocytes.(ABSTRACT TRUNCATED AT 400 WORDS)

The poly (purine) poly (pyrimidine) sequence in the 5' end of the thyroglobulin gene used as a probe, identifies a DNA fingerprint in man

Images

Thyroid peroxidase gene promoter confers TSH responsiveness to heterologous reporter genes in transfection experiments

The cyclic AMP-mediated transcriptional regulation of the enzyme thyroperoxidase by thyrotropin (TSH) in thyroid follicular cells was examined at the molecular level. The 5' end of the human thyroperoxidase gene was isolated and sequenced and the transcription start site was mapped by S1 nuclease analysis. A 0.9 kilobase pair DNA fragment of the promoter was shown to confer responsiveness to thyrotropin, and cyclic AMP, in transient expression assays using two different reporter genes. Several potential sites for specific interaction with nuclear transcription factors which could be involved in the regulation of thyroperoxidase gene transcription were identified.

Control of thyroperoxidase and thyroglobulin transcription by cAMP: evidence for distinct regulatory mechanisms

The expression of the genes coding for thyroglobulin (TG), and thyroperoxidase (TPO), are regulated by TSH. These effects are mediated by cAMP as they are reproduced by forskolin. In vitro run-on transcription assays performed on nuclei isolated from dog thyrocytes in culture or from dog thyroid slices, indicate that the forskolin-induced transcriptional stimulation of TG and TPO genes are very different. For the TG gene, the kinetics of transcriptional activation vary according to the experimental model: it is rapid (1 h) in thyroid slices and slow (8 h) in primary cultures. In contrast, TPO induction is rapid in both cases. In primary cultures, insulin is responsible for the basal level and for a part of forskolin-induced TG transcription, whereas TPO transcription is not affected by insulin. The forskolin-induced increase of TG transcription requires ongoing protein synthesis, as it is blocked by cycloheximide. TPO gene transcription is unaffected by cycloheximide. Taken together with previous data on the two genes, our results suggest that while TPO regulation corresponds to the classical model of genes in which the promoter is regulated directly via cAMP regulatory elements, TG gene regulation involves the synthesis of an intermediary, rapid turnover trans-acting protein.

Identification of a cAMP-responsive region in thyroglobulin gene promoter

The DNA sequences involved in transcription control by a cAMP-dependent mechanism have been localized in the thyroglobulin gene promoter region by a functional assay. The proximal 5'-flanking sequences from the bovine thyroglobulin gene were linked to the bacterial chloramphenicol acetyl-transferase gene. Transient expression of this reporter gene was studied in dog thyrocytes in primary culture in the presence, or absence, of cAMP stimulation. Deletion analysis showed that the cAMP-responsive region is contained within the first 250 base-pairs of the promoter, and suggests that it could correspond to a sequence conserved between species. These DNA sequences do not bear significant homology with cAMP-responsive elements (CRE) described previously. By contrast, some similarities were found with the fat-specific element (FSE2) of genes under cAMP control in adipocytes and with DNA elements mediating cAMP-dependent regulation of expression of two different genes in the lower eukaryote Dictyostelium discoideum. This suggests that control of Tg gene transcription by cAMP could involve a mechanism different from the one mediated by a classical CRE.