Search

Fras1 is an extracellular protein of the basement membranes that surround embryonic epithelia, choroid plexuses and meninges in foetal mouse brain. Depletion of Fras1 in knockout mice results in sub-epidermal blistering and fusion of eyelids and digits as well as malformation of lungs and kidneys. Mutations in the human counterpart FRAS1 are responsible for the Fraser Syndrome with clinical manifestations similar to the murine phenotype. In addition, brain deformities or mental impairments have occasionally been reported in patients with Fraser Syndrome. In the present study, we explored the possible involvement of Fras1 in brain function, analysing its expression pattern in mouse brain and investigating aspects of Fras1-/- mice behaviour, related to the function of brain regions expressing Fras1. Transcripts were detected in choroid plexuses and in certain brain regions including cortical, hippocampal and amygdalar areas in juvenile mice. Behavioural tests revealed that Fras1-/- mice exhibit impaired egocentric spatial memory, aberrant olfactory learning and memory, markedly reduced fear memory in an auditory fear conditioning task, as well as reduced anxiety expression in open field and elevated plus maze tests. Moreover, the extracellular matrix organization has been severely affected in cortical and subcortical areas as demonstrated by Wisteria floribunda agglutinin immunolabelling. The widespread detection of Fras1 transcripts in the brain of both pre- and postnatal mice, as well as the behavioural and cellular disturbances exhibited by Fras1-/- adult mice provide evidence for the involvement of Fras1 in brain organization and function.

Fraser syndrome (OMIM 219000) is a multisystem malformation usually comprising cryptophthalmos, syndactyly and renal defects(1). Here we report autozygosity mapping and show that the locus FS1 at chromosome 4q21 is associated with Fraser syndrome, although the condition is genetically heterogeneous. Mutation analysis identified five frameshift mutations in FRAS1, which encodes one member of a family of novel proteins related to an extracellular matrix (ECM) blastocoelar protein found in sea urchin. The FRAS1 protein contains a series of N-terminal cysteine-rich repeat motifs previously implicated in BMP metabolism, suggesting that it has a role in both structure and signal propagation in the ECM. It has been speculated that Fraser syndrome is a human equivalent of the blebbed phenotype in the mouse(2), which has been associated with mutations in at least five loci including bl(3). As mapping data were consistent with homology of FRAS1 and bl, we screened DNA from bl/bl mice and identified a premature termination of mouse Fras1. Thus, the bl mouse is a model for Fraser syndrome in humans, a disorder caused by disrupted epithelial integrity in utero.

Basement membranes constitute architecturally complex extracellular matrix (ECM) protein networks of great structural and regulatory importance. Recently, a novel group of basement membrane proteins, Fras1 (Fraser syndrome protein (1) and the Fras1-related extracellular matrix proteins Frem1, Frem2 and Frem3, has emerged. They comprise components of the sublamina densa region and contribute to embryonic epithelial-mesenchymal integrity. Fras1/Frem share common polypeptide repetitive motifs with possible interactive and organizing functions. Mutations in genes encoding Fras1, Frem1 and Frem2 are causative for dermal-epidermal detachment in the plane of sublamina densa and have been identified in different classes of mouse bleb mutants, the murine model of human Fraser syndrome, the hallmark phenotypic characteristics of which are embryonic skin blistering, cryptophthalmos and renal agenesis. Indeed, defects in FRAS1 and FREM2 have been identified in Fraser syndrome patients. The phenotypic similarity of mouse bleb mutant strains can be attributed to the fact that Fras1, Frem1 and Frem2 have been experimentally shown to interact, forming a mutually stabilized protein complex, while Frem3, which has not yet been associated with any of the existing known mutations, operates in a more independent fashion. Fras1/Frem have been recently proposed to compensate for the activity of collagen VII, a major anchoring component of the sublamina densa, the levels of which rise only during late embryonic life. By focusing on the aforementioned data, in this review we will summarize the current knowledge about Fraser syndrome proteins and describe their contribution to basement membrane biology.

Fras1/Frem family of basement membrane proteins has been associated with the "bleb" phenotype in mouse and the Fraser syndrome in man. Fras1 and Frem2 proteins are known to be colocalized in all epithelial basement membranes during embryonic development. The functional significance of their colocalization has been demonstrated in the corresponding mouse mutants, where the absence of Fras1 results in complete depletion of Frem2 from epithelial basement membranes and vice versa. Nevertheless, under standard immunohistochemical procedures, we were able to detect Fras1, but not Frem2, in the basement membrane of adult mouse tail skin. After reevaluation of our protocol, we established 15-minute acidic buffer treatment to be of critical value upon Frem2 immunodetection, essentially operating as an antigen retrieval process. Testing more polyclonal antibodies revealed no negative effects, but rather reinforced the positive signal, rendering this technique suitable for incorporation to any standard immunohistochemical procedure.

Fras1 and the structurally related proteins Frem1, Frem2, and Frem3, comprise a novel family of extracellular matrix proteins, which localize in a similar fashion underneath the lamina densa of epithelial basement membranes. They are involved in the structural adhesion of the skin epithelium to its underlying mesenchyme. Deficiency in the individual murine Fras1/Frem genes gives rise to the bleb phenotype, which is equivalent to the human hereditary disorder Fraser syndrome, characterized by cryptophthalmos (hidden eyes), embryonic skin blistering, renal agenesis, and syndactyly. Recent studies revealed a functional cooperation between the Fras1/Frem gene products, in which Fras1, Frem1 and Frem2 are simultaneously stabilized at the lowermost region of the basement membrane by forming a macromolecular ternary complex. Loss of any of these proteins results in the collapse of the protein assembly, thus providing a molecular explanation for the highly similar phenotypic defects displayed by the respective mutant mice. Here, we summarize the current knowledge regarding the structure, function, and interplay between the proteins of the Fras1/Frem family and further propose a possible scenario for the evolution of the corresponding genes.

The Fraser syndrome protein Fras1 and the structurally related proteins Frem1, Frem2 and Frem3 comprise a novel family of extracellular matrix proteins implicated in the structural adhesion of the embryonic epidermis to the underlying mesenchyme. Fras1, Frem1 and Frem2 have been shown to be simultaneously and interdependently stabilized in the basement membrane by forming a ternary complex located underneath the lamina densa. However, the functional relationships between Frem3 and the other Fras1/Frem proteins remain unknown. Here we show that in the absence of Fras1 the basement membrane localization of Frem3 remains unaffected in contrast to Frem1 and Frem2 which are completely abolished from the basement membrane. This indicates that although Frem3 is localized in the sublamina densa similar to Fras1, Frem1 and Frem2 yet it is anchored in the basement membrane independently. We further demonstrate that loss of Fras1 results in the accumulation of Frem2 within epithelial cells. This finding reveals that Fras1 is not only essential as a component of a macromolecular complex for the extracellular stabilization of Frem2 but it is also required for its proper intracellular trafficking and export from embryonic epithelial cells.

Fras1 is a putative extracellular matrix protein that has been implicated in the structural adhesion of embryonic epidermis to dermis. Moreover, mutations in Fras1/FRAS1 have been associated with the mouse blebbed phenotype and the human rare genetic disorder Fraser syndrome, respectively. Here we report the mapping of Fras1 within the extracellular space and evaluate the effects of Fras1 deficiency on lung development in the mouse. Expression of Fras1 was detected in the mesothelial cells of the visceral pleura and in the conducting airway epithelia. Immunogold histochemistry identified Fras1 as a component of the extracellular matrix localized below the lamina densa of epithelial basement membranes in the embryonic lung. Embryos homozygous for a targeted mutation of Fras1 exhibited fused pulmonary lobes resulting from incomplete separation during development as well as a profound disarrangement of blood capillaries in the terminal air sacs. We demonstrate that loss of Fras1 causes alterations in the molecular composition of basement membranes, concomitant with local disruptions of epithelial-endothelial contacts and extravasation of erythrocytes into the embryonic respiratory lumen. Thus, our findings identify Fras1 as an important structural component of the sub-lamina densa of basement membranes required for lobar septation and the organization of blood capillaries in the peripheral lung.

The Fras1/Frem family of extracellular matrix proteins consists of Fras1 and its structurally related proteins, Frem1 (Fras1-related extracellular matrix protein 1), Frem2 and Frem3. These are co-localized in embryonic epithelial basement membranes (BMs), where they contribute to epithelial–mesenchymal adhesion. Although Fras1 localization pattern in epithelial BMs has been well defined, it has not yet been comprehensively studied in the central nervous system. Here, we demonstrate the immunohistochemical profile of Fras1 in the developing mouse brain and reveal an exclusively meningeal BM protein deposition. Interestingly, Fras1 displays a segmental localization pattern, which is restricted to certain regions of the meningeal BM. Frem2 protein displays a similar localization pattern, while Frem3 is rather uniformly distributed throughout the meningeal BM. Fras1 and Frem2 proteins are detected in regions of the BM that underlie organizing centers, such as the roof plate (RP) of diencephalon, midbrain and hindbrain, and the RP-derived structures of telencephalon (choroid plexus and hem). Organizing centers exert their activity via the production of bioactive molecules, which are potential Fras1 ligands. The restricted pattern of Fras1 and Frem2 proteins indicates a molecular compartmentalization of the meningeal BM that could reflect, yet unspecified, functional and structural differences.

The Pax gene family, encoding transcription factors, has been classified into four subfamilies according to their genomic organization, the sequences of the paired domains (PD) and the expression pattern. Pax1 and Pax9 constitute one subfamily, Pax2, Pax5 and Pax8 another, Pax3 and Pax7 another one and Pax4 and Pax6 the fourth subfamily. The PD exhibits DNA-binding activity, and is the most conserved functional motif in all Pax proteins. A high-resolution analysis of a PD structure has been performed [Xu et al., Cell 80 (1995) 639-650] and the DNA-binding characteristics of members of two Pax subfamilies (Pax2, Pax5 and Pax6) have been determined. Here, we have utilized a PCR-based selection approach to identify the DNA-binding sequences of the Pax3/PD, a member of a subfamily which has not yet been characterized. Comparison of the Pax3/PD-binding sequences with those of other PD proteins revealed both similarities and differences in the DNA-recognition sequence. This suggests that different Pax proteins can regulate the expression of the same target gene, but they can also regulate the expression of completely unrelated genes by binding to their DNA regulatory regions.

In recent years, the discovery of Pax genes in mouse has played an invaluable role in furthering our understanding in mouse developmental processes and disorders. To date, eight murine paired box-containing genes have been cloned. Seven of these exhibit a distinct spatiotemporal expression pattern in the developing nervous system implying a role in the regional specification of the developing spinal cord and brain. The Pax genes encode for sequence-specific DNA binding transcription factors that play a key role in embryonic development. Three of these developmental control genes are altered in mutant mice and two are associated with human diseases. Disruption of these Pax genes leads to abnormalities in neural crest derivatives, neuroectoderm, sclerotome or myotome-derived tissues. Disruption of the Pax-3 gene causes the Splotch phenotype in mice and Waardenburg syndrome in humans. Pax-6 mutations result in Small eye mice and the human genetic disorder aniridia. The Pax-1 gene is mutated in undulated mice. Pax proteins can transform cells in culture which then form tumours following injection in nude mice. Consistent with this activity, PAX3 has been recently implicated in the generation of the tumour alveolar rhabdomyosarcoma.

The mouse mammary tumor virus (MMTV) promoter, that responds to glucocorticoids and progestins, contains a complex hormone response element (HRE) in the long terminal repeat (LTR) region covered by a phased nucleosome. Hormone treatment leads to alterations in chromatin structure that make the HRE region more accessible to digestion by Dnase I and permit binding of transcription factors, including nuclear factor I (NFI), immediately downstream of the HRE. NFI acts as a basal transcription factor on the MMTV promoter in vitro but competes with the hormone receptors in terms of binding to free DNA. In uninduced chromatin, the precise positioning of the DNA double helix on the surface of the histone octamer precludes binding of NFI to its cognate sequence while still allowing recognition of the HRE by the hormone receptors. We postulate that receptor binding to the nucleosomally organized MMTV promoter disrupts the chromatin structure enabling NFI binding and subsequent formation of a stable transcription complex. Whether the receptor remains bound to DNA during induction or is displaced by NFI is not conclusively known, but our evidence supports a “hit and run” mechanism. NFI is not the only factor involved in hormonally induced transcription of the MMTV promoter. Two degenerated octamer motifs located immediately upstream of the TATA box are recognized by the ubiquitous transcription factor OTF-1 (Oct-1, NFIII), and are also important. In vitro , mutations in these motifs do not influence basal transcription, but completely abolish the stimulatory effect of purified progesterone receptor. Progesterone receptor bound to the HRE facilitates binding of OTF-1 to the two octamer motifs. Thus, OTF-1 is a natural mediator of progesterone induction of the MMTV promoter and acts through cooperation with the hormone receptor for binding to DNA.

Using the interaction of steroid hormone receptors with their palindromic response elements as an example, we show here that cloning in dam+ bacterial strains can lead to artifactual results due to methylation of adenine residues at the N-6 position. Substitution of the T by an A in the third position of the half palindromes of the hormone responsive element TGTTCT(1) yields a functional element only when amplification is made in dam+ bacteria. Mutant palindromes methylated at the N-6 position of this adenine exhibit the same affinity for progesterone and glucocorticoid receptors as the consensus response element, whereas their unmethylated counterpart binds with negligible affinity. These observations underline the significance of hydrophobic interactions between receptors and the major groove of the DNA for discrimination among various responsive elements, and point to the importance of using dam- bacterial strains for the correct identification of the nucleotide sequence of cis-acting elements.

Summary The paired domain is a conserved DNA binding motif which was first found in Drosophila segmentation gene products. This paired domain is encoded by a well conserved, paired box DNA sequence, also detected in the genomes of other species. The mouse paired box-containing genes are referred to as Pax genes and are expressed in a distinct spatiotemporal pattern during embryogenesis. Pax proteins are able to bind to specific DNA sequences and modulate transcriptional activity. Interestingly, three different Pax genes have already been shown to correspond to some mouse and human mutants, emphasizing their role as developmental control genes.

Steroid hormone receptors can be divided into two subfamilies according to the structure of their DNA binding domains and the nucleotide sequences which they recognize. The glucocorticoid receptor and the progesterone receptor (PR) recognize an imperfect palindrome (glucocorticoid responsive element/progesterone responsive element [GRE/PRE]) with the conserved half-sequence TGTYCY, whereas the estrogen receptor (ER) recognizes a palindrome (estrogen responsive element) with the half-sequence TGACC. A series of symmetric and asymmetric variants of these hormone responsive elements (HREs) have been tested for receptor binding and for the ability to mediate induction in vivo. High-resolution analysis demonstrates that the overall number and distribution of contacts with the N-7 position of guanines and with the phosphate backbone of various HREs are quite similar for PR and ER. However, PR and glucocorticoid receptor, but not ER, are able to contact the 5'-methyl group of thymines found in position 3 of HREs, as shown by potassium permanganate interference. The ER mutant HE84, which contains a single amino acid exchange, Glu-203 to Gly, in the knuckle of ER, creates a promiscuous ER that is able to bind to GRE/PREs by contacting this thymine. Elements with the sequence GGTCAcagTGTYCT that represent hybrids between an estrogen response element and a GRE/PRE respond to estrogens, glucocorticoids, and progestins in vivo and bind all three wild-type receptors in vitro. These hybrid HREs could serve to confer promiscuous gene regulation.

The Fras1/Frem gene family encodes for structurally similar proteins of the extracellular matrix, functionally correlated with embryonic dermal–epidermal adhesion as deduced from the appearance of sub-epidermal blisters in mouse mutants compromising the function of Fras1, Frem1 and Frem2 proteins. Mutations in the human counterparts FRAS1 and FREM2 have been detected in patients suffering from Fraser syndrome. So far, Fras1/Frem proteins have been shown to be strictly colocalized in the sublamina densa of mouse epithelial basement membranes during development. Here, we focused on the characterization of the localization pattern of the aforementioned proteins, in various parts of the adult mouse skin as well as a range of organs and tissues. Frem3 was present in a broad range of epithelial basement membranes where Fras1, Frem1 and Frem2 were missing. The localization profile of Frem3 coincided with that of collagen VII in all skin basement membranes but differed in that Frem3 was additionally found in the basement membrane of several internal epithelia, where collagen VII was absent. Fras1 and Frem2 were colocalized with Frem3 in the basement membrane of certain skin parts, underlying the thin-layer, of rapidly proliferating keratinocytes, whereas Frem1 was detected only in the basement membrane of the tail. The localization pattern of Fras1 and Frem2 was indistinguishable, while both proteins along with Frem3 could be detected even in the absence of Frem1.

A combination of the gel retardation assay and interference by hydroxyl radical modification (missing nucleoside technique) was used to analyze the interaction of the glucocorticoid receptor (GR) with various glucocorticoid responsive elements (GRE). Short oligonucleotides containing the 15-bp GRE and 1 to 3 flanking base pairs on each side, are bound with very low affinity. The same GREs, when positioned in the center of a large DNA fragment (40-50 bp), show high affinity for the receptor. However, when the GRE is positioned at the border of a 54-bp fragment, the affinity of the GR for the GRE decreases markedly. The DNA binding affinity increases linearly with each added flanking base pair and optimal binding is observed with 8-10 flanking bp. Thus, the nonconserved DNA sequences flanking the GRE contribute significantly to the free energy of receptor binding to DNA. Using larger DNA fragments (greater than 100 bp) and a smaller form of the receptor (40 kD), two retarded complexes are found that correspond to monomeric and homodimeric receptor DNA complexes. The DNA-binding domain of the GR (20 kD), expressed in bacteria, binds to the GRE as a monomer as well as a dimer and can form heterodimers with the native 94-kD GR. Insertion or deletion of one single base pair between the two halves of the GRE reduces the affinity for the homodimeric form of the native GR, and inhibits the function of the GRE in gene transfer experiments, suggesting that a dimer of the GR is the functional entity that binds to the GRE.

Fras1 is the first identified member of a protein family comprising Fras1 and the related extracellular matrix proteins Frem1, Frem2 and Frem3. Mutations in Fras1, Frem1 and Frem2 have been associated with the bleb phenotype in mouse, whereas mutations in the human orthologs FRAS1 and FREM2 have been implicated in the pathogenesis of the human Fraser syndrome. Bleb mutant mice are characterized by embryonic sub-epidermal blistering, unilateral or bilateral renal agenesis or dysgenesis, cutaneous syndactyly and fused eyelids. As revealed by immunofluorescence, Fras1 co-localizes with the markers of epithelial basement membranes and is ultrastructurally detected underneath the lamina densa of embryonic mouse epithelia. Since the loss of Fras1 mainly affects the cohesiveness of the embryonic skin basement membrane with its underlying mesenchyme, we compared here the ultrastructural localization of Fras1 in the dermal-epidermal junction and in the basement membrane of other embryonic epithelia that do not show any overt phenotype using preembedding immunocytochemistry. Fras1 immunoreactivity was detected in all epithelia examined, within the sublamina densa adjacent to stromal tissue, as clustered gold/silver enhanced depositions, usually attached to anchoring fibrils. Interestingly, clusters corresponding to Fras1 were frequently detected in close proximity to mesenchymal cells, indicating that Fras1 could serve as a direct link between the sublamina densa and mesenchyme. The localization of Fras1 is consistent with previous results indicating that Fras1 exerts its function below the lamina densa and that Fras1 displays the same localization pattern in all epithelial basement membranes.

The Fras1/Frem gene family encodes for structurally similar, developmentally regulated extracellular matrix proteins. Mutations in Fras1, Frem1 and Frem2 have been identified in different classes of mouse bleb mutants, while defects in the human orthologs FRAS1 and FREM2 are causative for Fraser syndrome. The hallmark phenotypic feature of bleb mice is embryonic skin blistering due to dermal-epidermal detachment. The similarity of the phenotypic characteristics among the bleb mouse mutants, together with the fact that Fras1/Frem proteins are co-localized in embryonic epithelial basement membranes, suggest that they operate in a common pathway. Here, we report for the first time the immunofluorescence pattern of Frem3 and provide a comparative analysis of the spatiotemporal localization of all Fras1/Frem proteins during mouse embryonic development. We demonstrate their overall co-localization in embryonic epithelial basement membranes, with emphasis on areas of phenotypic interest such as eyelids, limbs, kidneys, lungs and organs of the gastrointestinal tract and the central nervous system. We further studied collagen VII, impairment of which produces dystrophic epidermolysis bullosa, a postnatal skin blistering disorder. We show that basement membrane levels of collagen VII rise at late embryonic life, concomitant with descending Fras1/Frem immunolabeling.

Frem1 belongs to a family of structurally related extracellular matrix proteins of which Fras1 is the founding member. Mutations in Fras1 and Frem1 have been identified in mouse models for Fraser syndrome, which display a strikingly similar embryonic skin blistering phenotype due to impaired dermal-epidermal adhesion. Here we show that Frem1 originates from both epithelial and mesenchymal cells, in contrast to Fras1 that is exclusively derived from epithelia. However, both proteins are localized in an absolutely overlapping fashion in diverse epithelial basement membranes. At the ultrastructural level, Frem1 exhibits a clustered arrangement in the sublamina densa coinciding with fibrillar structures reminiscent of anchoring fibrils. Furthermore, in addition to its extracellular deposition, around E16, Frem1 displays an intracellular distribution in distinct epidermal cell types such as the periderm layer and basal keratinocytes. Since periderm cells are known to participate in temporary epithelial fusions like embryonic eyelid closure, defective function of Frem1 in these cells could provide a molecular explanation for the "eyes open at birth" phenotype, a feature unique for Frem1 deficient mouse mutants. Finally, we demonstrate loss of Frem1 localization in the basement membrane but not in periderm cells in the skin of Fras1(-/-) embryos. Taken together, our findings indicate that besides a cooperative function with Fras1 in embryonic basement membranes, Frem1 can also act independently in processes related to epidermal differentiation.

Fraser syndrome is a recessive, multisystem disorder presenting with cryptophthalmos, syndactyly and renal defects and associated with loss-of-function mutations of the extracellular matrix protein FRAS1. Fras1 mutant mice have a blebbed phenotype characterized by intrauterine epithelial fragility generating serous and, later, hemorrhagic blisters. The myelencephalic blebs (my) strain has a similar phenotype. We mapped my to Frem2, a gene related to Fras1 and Frem1, and showed that a Frem2 gene-trap mutation was allelic to my. Expression of Frem2 in adult kidneys correlated with cyst formation in my homozygotes, indicating that the gene is required for maintaining the differentiated state of renal epithelia. Two individuals with Fraser syndrome were homozygous with respect to the same missense mutation of FREM2, confirming genetic heterogeneity. This is the only missense mutation reported in any blebbing mutant or individual with Fraser syndrome, suggesting that calcium binding in the CALXbeta-cadherin motif is important for normal functioning of FREM2.

Cell adhesion to extracellular matrix (ECM) proteins is crucial for the structural integrity of tissues and epithelial-mesenchymal interactions mediating organ morphogenesis1,2. Here we describe how the loss of a cytoplasmic multi-PDZ scaffolding protein, glutamate receptor interacting protein 1 (GRIP1), leads to the formation of subepidermal hemorrhagic blisters, renal agenesis, syndactyly or polydactyly and permanent fusion of eyelids (cryptophthalmos). Similar malformations are characteristic of individuals with Fraser syndrome and animal models of this human genetic disorder, such as mice carrying the blebbed mutation (bl) in the gene encoding the Fras1 ECM protein3,4. GRIP1 can physically interact with Fras1 and is required for the localization of Fras1 to the basal side of cells. In one animal model of Fraser syndrome, the eye-blebs (eb) mouse, Grip1 is disrupted by a deletion of two coding exons. Our data indicate that GRIP1 is required for normal cell-matrix interactions during early embryonic development and that inactivation of Grip1 causes Fraser syndrome–like defects in mice.

Loss of tight association between epidermis and dermis underlies several blistering disorders and is frequently caused by impaired function of extracellular matrix (ECM) proteins1,2. Here we describe a new protein in mouse, Fras1, that is specifically detected in a linear fashion underlying the epidermis and the basal surface of other epithelia in embryos. Loss of Fras1 function results in the formation of subepidermal hemorrhagic blisters as well as unilateral or bilateral renal agenesis during mouse embryogenesis. Postnatally, homozygous Fras1 mutants have fusion of the eyelids and digits and unilateral renal agenesis or dysplasia. The defects observed in Fras1−/− mice phenocopy those of the existing bl (blebbed) mouse mutants3,4, which have been considered a model for the human genetic disorder Fraser syndrome5,6. We show that bl/bl homozygous embryos are devoid of Fras1 protein, consistent with the finding that Fras1 is mutated in these mice6. In sum, our data suggest that perturbations in the composition of the extracellular space underlying epithelia could account for the onset of the blebbed phenotype in mouse and Fraser syndrome manifestation in human.

The temporally and spatially restricted expression of the mouse Engrailed (En) genes is essential for development of the midbrain and cerebellum. The regulation of En-2 expression was studied using in vitro protein-DNA binding assays and in vivo expression analysis in transgenic mice to gain insight into the genetic events that lead to regionalization of the developing brain. A minimum En-2 1.0 kb enhancer fragment was defined and found to contain multiple positive and negative regulatory elements that function in concert to establish the early embryonic midhindbrain expression. Furthermore, the mid-hindbrain regulatory sequences were shown to be structurally and functionally conserved in humans. The mouse paired-box-containing genes Pax-2, Pax-5 and Pax-8 show overlapping expression with the En genes in the developing brain. Significantly, two DNA-binding sites for Pax-2, Pax-5 and Pax-8 proteins were identified in the 1.0 kb En-2 regulatory sequences, and mutation of the binding sites disrupted initiation and maintenance of expression in transgenic mice. These results present strong molecular evidence that the Pax genes are direct upstream regulators of En-2 in the genetic cascade controlling midhindbrain development. These mouse studies, taken together with others in Drosophila and zebrafish on the role of Pax genes in controlling expression of En family members, indicate that a Pax-En genetic pathway has been conserved during evolution.

The Pax gene family consists of nine genes identified in the mouse genome by their homology to the Drosophila segmentation gene, paired. The Pax gene products have the biochemical properties of transcription factors and share the conserved paired domain that exhibits DNA binding activity. Pax genes are expressed in temporally and spatially restricted patterns during embryogenesis and control key aspects of embryonic development. Pax gene mutations are semidominantly inherited, and homozygous mouse mutants are usually lethal.

Pax-3 is a member of the Pax family of transcription factors involved in transcriptional control events during embryonic development. Here we report a functional dissection of the Pax-3 protein and describe the protein domains which are responsible for different activities. A transcription inhibition activity is located in the first 90 N-terminal amino acids and includes part of the paired domain. Furthermore, the C terminus of Pax-3 is able to confer transcriptional activation of basal promoters. Pax-3 can utilize both transcription modulating functions and activates transcription over a narrow range of protein concentration in the presence of promoter elements containing functional binding sites.

The neural cell adhesion molecule L1, a member of the immunoglobulin superfamily, mediates cell interactions in the developing and regenerating nervous system of mammals and is also detectable in the immune system and in the epithelia of intestine, skin, lung, and kidney. This diverse pattern of expression begs the question as to the regulatory mechanisms underlying transcription of the L1 gene. We demonstrate here that the paired domain and homeodomain containing Pax-6 protein binds to three different sites in the promoter region of the L1 gene. The promoter proximal binding site is also recognized by Hoxa-1 and lies approximately 60 bp upstream from the transcription start site only few base pairs upstream of a putative binding site for the TFII-I transcription initiation factor. On the basis of this sequence, we have characterized the binding of Pax-6 and explored two modes of its DNA binding activities.

The mouse Pax-3 gene encodes a protein that is a member of the Pax family of DNA binding proteins. Pax-3 contains two DNA binding domains: a paired domain (PD) and a paired type homeodomain (HD). Both domains are separated by 53 amino acids and interact synergistically with a sequence harboring an ATTA motif (binding to the HD) and a GTTCC site (binding to the PD) separated by 5 base pairs. Here we show that the interaction of Pax-3 with these two binding sites is independent of their angular orientation. In addition, the protein spacer region between the HD and the PD can be shortened without changing the spatial flexibility of the two DNA binding domains which interact with DNA. Furthermore, by using circular permutation analysis we determined that binding of Pax-3 to a DNA fragment containing a specific binding site causes conformational changes in the DNA, as indicated by the different mobilities of the Pax-3-DNA complexes. The ability to change the conformation of the DNA was found to be an intrinsic property of the Pax-3 PD and of all Pax proteins that we tested so far. These in vitro studies suggest that interaction of Pax proteins with their specific sequences in vivo may result in an altered DNA conformation.

Pax genes control certain aspects of development, as mutations result in (semi)dominant defects apparent during embryogenesis. Pax-3 has been associated with the mouse mutant splotch (Sp) and the human Waardenburg syndrome type 1 (WS1). We have examined the molecular basis of splotch and WS1 by studying the effect of mutations on DNA binding, using a defined target sequence. Pax-3 contains two different types of functional DNA-binding domains, a paired domain and a homeodomain. Mutational analysis of Pax-3 reveals different modes of DNA binding depending on the presence of these domains. A segment of Pax-3 located between the two DNA-binding domains, including a conserved octapeptide, participates in protein homodimerization. Pax-3 mutations found in splotch alleles and WS1 individuals change DNA binding and, in the case of a protein product of the Sp allele, dimerization. These findings were taken as a basis to define the molecular nature of the mutants.

The murine paired box containing gene Pax-2 has been proposed to be involved in kidney and central nervous system (CNS) development. In this report, we show that expression cloning of Pax-2 cDNA allowed in vitro identification of specifically bound DNA sequences. When fused to the thymidine kinase (TK) promoter in front of reporter genes, these target sequences were able to mediate trans-activation by Pax-2 protein, thus demonstrating their in vivo function. Expression studies from adult mouse tissues revealed high levels of Pax-2 transcripts in male and female genital tracts, suggesting a second phase of Pax-2 activity. Sequence-specific DNA binding and subsequent modulation of promoter activities may constitute the molecular mechanism of Pax-2 action in specific adult tissues and during development.

Cytotactin is a morphoregulatory molecule of the extracellular matrix affecting cell shape, division, and migration that appears in a characteristic and complex site-restricted pattern during embryogenesis. The promoter region of the gene that encodes chicken cytotactin contains a variety of potential regulatory sequences. These include putative binding sites for homeodomain proteins and a phorbol 12-O-tetradecanoate 13-acetate response element (TRE)/AP-1 element, a potential target for transcription factors thought to be involved in growth-factor signal transduction. To determine the effects of homeobox-containing genes on cytotactin promoter activity, we conducted a series of cotransfection experiments on NIH 3T3 cells using cytotactin promoter-chloramphenicol acetyltransferase (CAT) reporter gene constructs and plasmids driving the expression of mouse homeobox genes Evx-1 and Hox-1.3. cotransfection with Evx-1 stimulated cytotactin promoter activity whereas cotransfection in control experiments with Hox-1.3 had no effect. To localize the sequences required for Evx-1 activation, we tested a series of deletions in the cytotactin promoter. An 89-base-pair region containing a consensus TRE/AP-1 element was found to be required for activation. An oligonucleotide segment containing this TRE/AP-1 site was found to confer Evx-1 inducibility on a simian virus 40 minimal promoter; mutation of the TRE/AP-1 site abolished this activity. To explore the potential role of growth factors in cytotactin promoter activation, chicken embryo fibroblasts, which are known to synthesize cytotactin, were first transfected with cytotactin promoter constructs and cultured under minimal conditions in 1% fetal bovine serum. Although the cells exhibited only low levels of CAT activity under these conditions, cells exposed for 12 h to 10% (vol/vol) fetal bovine serum showed a marked increase in CAT activity. Cotransfection with Evx-1 and cytotactin promoter constructs of cells cultured in 1% fetal bovine serum was sufficient, however, to produce high levels of CAT activity. These findings are consistent with the hypothesis that Evx-1, a homeobox-containing gene, may activate the cytotactin promoter by a mechanism involving a growth-factor signal transduction pathway. More generally, the results support the hypothesis that the place-dependent expression of morphoregulatory molecules may depend upon local cues provided by homeobox genes and their encoded proteins.

The murine paired box gene Pax-1 has been associated with the mouse developmental mutant undulated (un), which exhibits malformations in the vertebral column. In un mice, a point mutation leading to a Gly-Ser exchange in a conserved part of the paired domain of Pax-1 is present. Here we show that Pax-1 encodes a DNA-binding protein with transcriptional activating properties. The DNA-binding specificity of the Pax-1 protein has been extensively analyzed in gel shift assays, and in conjunction with binding interference experiments, a DNA-binding core motif was defined. Comparison of the DNA-binding properties of wild-type and un Pax-1 proteins demonstrates that the Gly-Ser replacement at position 15 within the paired domain dramatically decreases the DNA-binding affinity of the un Pax-1 protein and alters its DNA-binding specificity. These results decipher the molecular basis of the un mutation.

Understanding the mechanisms by which regulatory proteins recognize genetic information stored in DNA relies on the availability of methods to analyze their interaction with individual nucleotides and their reactive groups. Here we describe the use of KMnO4 to analyze the contacts between steroid hormone receptors and thymines within a hormone responsive element. Although several pyrimidine residues are highly conserved among different receptor binding sites their participation in sequence recognition has not been directly studied. Using an interference procedure based on selective modification of the thymine ring by KMnO4, we detect intimate contacts between the glucocorticoid or progesterone receptors and three thymine residues within the promoter distal receptor binding site of the mouse mammary tumor virus (-190/-160). A comparison of binding data obtained with oligonucleotides containing desoxyuridine, bromodeoxyuridine, cytosine, or 5'-methylcytosine instead of thymines demonstrates that the methyl group of those three thymines contributes to the free energy of binding. This simple method could be of general utility for the study of sequence-specific protein-DNA interactions.

The location, orientation, and structure of the hormone regulatory elements (HRE) in nine hormonally modulated genes is described. Based on analysis of the contact points between the glucocorticoid receptor (GR) and the DNA double helix within the HREs, a model for the interaction is proposed in which a dimer of the receptor in head-to-head orientation binds to the inverted symmetry element of the HRE. The relationship between the regulatory elements for glucocorticoids and progesterone in the long terminal repeat region (LTR) of mouse mammary tumor virus (MMTV), and in the promoter region of the chicken lysozyme gene, indicates that the recognition mechanism for both receptors is similar but not identical. Curiously, the hormone ligand is not an absolute requirement for the GR to bind its HRE, though it influences the kinetics of the interaction. Other possible functions of the hormone in vivo are discussed, as well as the molecular mechanism responsible for transcriptional regulation after receptor binding to the HRE.

The mouse mammary tumour virus long terminal repeat region contains regulatory sequences able to mediate transcriptional induction by different steroid hormones. Two clusters of binding sites for the glucocorticoid and the progesterone receptors have been identified in the region between -70 and -190, the so called hormone responsive or regulatory element. To understand the molecular details of the interaction between the receptors and the DNA we have used the high resolution technique of hydroxyl radical footprinting. Both in the promoter distal site and in the promoter proximal cluster additional contacts between the proteins and the double helix are detected by this technique, outside of the region identified by methylation protection. The pattern of contacts in the promoter distal region is compatible with a model involving the interaction of a receptor dimer with the major grooves of four subsequent turns of the double helix, each turn being contacted by a separate zinc finger. This model is illustrated by computer graphical methods and discussed in terms of sequence homologies with other hormone regulatory elements.

Gene regulation is generally mediated by DNA-binding proteins able to recognize regulatory information stored in the DNA double helix. In pro-karyotes there are several well-documented examples of regulatory proteins, such as the lambda repressor. The intimate molecular mechanism of the interaction between these prokaryotic regulatory proteins and the DNA double helix is known in great detail. However, it is not yet clear how this interaction modulates the efficiency of promoter utilization. The prevalent model assumes a direct interaction between the DNA-bound regulatory proteins and the RNA polymerase or some other component of the transcriptional machinery. Though this idea is appealing for cases in which the DNA regulatory sequences are close to the promoter, it cannot be extrapolated easily to situations in which the regulatory elements are far from the transcription start point. This is exactly the situation found in most of the regulatory elements described so far in mammalian cells. Indeed, a characteristic form of gene regulation in eukaryotes appears to be mediated through cis-acting elements located far upstream or even downstream of the promoter, the so-called enhancer elements. In most cases studied transacting factors that bind to the various enhancer elements are required for their function. Therefore, in order to accomodate the protein-protein interaction model to the eukaryotic situation, the DNA between the enhancer and the promoter will have to loop out. Of course, alternative mechanisms of regulation at a distance are conceivable and have been proposed in the past.

Gene regulation by steroid hormones is mediated through an interaction of the hormone receptors with DNA regulatory sequences called hormone regulatory or responsive elements (HRE). An analysis of the HRE's in the DNA of mouse mammary tumour provirus, human metallothionein IIA gene, chicken lysozyme gene, chicken and Xenopus vitellogenin genes, growth hormones genes, Moloney murine sarcoma provirus, rabbit uteroglobin gene, rat tyrosine aminotransferase gene, rat tryptophan oxygenase gene and rat acidic glycoprotein gene, yields the following consensus for positively modulated glucocorticoid responsive elements (GRE): 5'-GGTACAnnnTGTTCT-3'. This element can also mediate induction by progesterone and probably by androgens, but not by estrogens. Detailed analysis of the DNA protection pattern suggests that a dimer of the hormone receptor interacts with this palindromic 15-mer. In genes that are negatively regulated by glucocorticoids an imperfect copy of the GRE is found, and repression is probably due to competition between hormone receptor and other transcription factors or enhancer binding proteins for binding to overlapping DNA sequences. The receptors without bound hormone are able to interact specifically with DNA in vitro, but binding of hormone is needed for transcriptional activation in vivo. This could be due, at least in part, to changes in the rate parameters of the receptor-DNA interaction induced by binding of the hormone to the receptor. The possible role of precise chromatin organization in glucocorticoid induction is discussed on the basis of the nucleosome phasing found in the LTR region of mouse mammary tumour virus.

Binding to hormone responsive elements (HRE) is an essential step in gene regulation by steroid hormones. Using a combination of in vivo and in vitro studies we have analyzed the interactions of the estrogen receptor with genes from three different systems: the long terminal repeat (LTR) of the mouse mammary tumor virus (MMTV), the chicken and xenopus vitellogenin genes and the rabbit uteroglobin gene. The estrogen receptor binds to all four genes in vitro, but the MMTV LTR does not respond to estrogen in gene transfer experiments. Similarly, the xenopus vitellogenin gene binds the progesterone and glucocorticoid receptors in vitro, but only estrogen induces the xenopus vitellogenin gene in gene transfer. These results suggest that events distal to DNA binding are essential for transcriptional activation, and that the function of the HRE is not simply to position the hormone receptor in the vicinity of the regulated promoter.

Steroid hormone receptors induce genes by virtue of their interaction with DNA regulatory sequences. The hormonal response of a particular gene in vivo correlates with binding of the hormone to the receptor and supposedly reflects the degree of occupancy of the corresponding DNA regulatory sequences. However, in vitro the steroid-free glucocorticoid and progesterone receptors bind specifically to the regulatory sequences of mouse mammary tumor virus, thus raising questions on the role of the hormone in DNA binding in vivo. By using monoclonal antibodies, gel retardation assays, and filter binding techniques we show here that binding of a functional steroid to either the glucocorticoid or the progesterone receptors influences the kinetics of the protein-DNA interaction in vitro. In the presence of hormone the on rate of receptor binding to DNA fragments with or without regulatory sequences is accelerated 2- to 5-fold, and the off rate is accelerated 10- to 20-fold. The receptors complexed to an antihormone bind to DNA with kinetics intermediate between those of the steroid-free and the hormone-bound protein. Thus, ligand binding accelerates the kinetics of receptor binding to DNA and this could partly account for the behavior of the hormone receptor observed in vivo.

Modulation of gene expression frequently takes place at the level of transcription, and is mediated by the interaction of regulatory proteins with specific DNA sequences near the regulated promoters. Among the best-characterized examples of such regulatory proteins are the steroid hormone receptors. These proteins are members of a large family of nuclear proteins called the steroid/thyroid hormone receptor family or the nuclear receptor family. Most members of this gene family share a double specificity; on the one hand, they interact with small ligands, such as steroid hormones, thyroid hormones, vitamin D, and retinoic acid; on the other hand, they bind to regulatory elements in the DNA that mediate modulation of gene activity in cis. During the last decade the DNA complementary to the mRNAs encoding most of the members of the hormone receptor gene family have been cloned. A comparison of the primary structure of the various members of the family permits one to derive a general scheme for the modular architecture of these regulatory proteins (Fig. 1). The proteins are composed of three separate domains, one of which interacts with the hormone-ligand, the second binds to DNA, and the third domain modulates transcriptional activity. The most conserved region of the protein is located in the central domain, which consists of a short stretch of about 70 amino acids containing the DNA binding activity. This region of the protein exhibits an array of cysteine residues compatible with an organization of the domain into two so-called zinc fingers, in which two zinc ions are each tetrahedrally coordinated with four cysteines (for a review, see Evans 1988).

Gene regulation by steroid hormones leads to induction or repression of particular sets of genes. These effects are mediated by intracellular hormone receptors that, in the unliganded state, are maintained in an inactive form by unknown mechanisms possibly involving association with other cellular proteins. Induction of the mouse mammary tumor virus (MMTV) requires binding of the hormone receptor to a complex hormone-responsive element (HRE) located between 75 and 190 bp upstream from the start of transcription. The interaction of several receptor molecules with the four receptor binding sites in the HRE is highly cooperative on circular DNA molecules and each individual site is needed for optimal induction. In chromatin the HRE is precisely organized in phased nucleosomes. Following hormone treatment and receptor binding, changes in chromatin structure are detected that correlate with binding of transcription factors, including nuclear factor I, to the MMTV promoter. However, though nuclear factor I acts as a basal transcription factor on the MMTV promoter it does not cooperate with the hormone receptors in terms of binding to free DNA, and mutation of the nuclear factor I binding site does not eliminate hormonal stimulation. This residual induction is mediated by octamer motifs, upstream of the TATA box, that bind the ubiquitous transcription factor OTF-1. Mutation of these octamer motifs does not influence basal transcription in vitro, but completely abolishes the stimulatory effect of progesterone receptor.

Plasmids containing the hormone regulatory element of mouse mammary tumor virus linked to the thymidine kinase promoter of herpes simplex virus and the reporter gene chloramphenicol acetyltransferase of Escherichia coli respond to glucocorticoids and progestins when transfected into appropriate cells. In the human mammary tumor cell line T47D, the response to progestins, but not to glucocorticoids, is highly dependent on the topology of the transfected DNA. Although negatively supercoiled plasmids respond optimally to the synthetic progestin R5020, their linearized counterparts exhibit markedly reduced progestin inducibility. This is not due to changes in the efficiency of DNA transfection, since the amount of DNA incorporated into the cell nucleus is not significantly dependent on the initial topology of the plasmids. In contrast, cotransfection experiments with glucocorticoid receptor cDNA in the same cell line show no significant influence of DNA topology on induction by dexamethasone. A similar result was obtained with fibroblasts that contain endogenous glucocorticoid receptors. When the distance between receptor-binding sites or between the binding sites and the promoter was increased, the dependence of progestin induction on DNA topology was more pronounced. In contrast to the original plasmid, these constructs also revealed a similar topological dependence for induction by glucocorticoids. The differential influence of DNA topology is not due to differences in the affinity of the two hormone receptors for DNA of various topologies, but probably reflects an influence of DNA topology on the interaction between different DNA-bound receptor molecules and between receptors and other transcription factors.

We describe the isolation and characterization of Pax-3, a novel murine paired box gene expressed exclusively during embryogenesis. Pax-3 encodes a 479 amino acid protein with an Mr of 56 kd containing both a paired domain and a paired-type homeodomain. The Pax-3 protein is a DNA binding protein that specifically recognizes the e5 sequence present upstream of the Drosophila even-skipped gene. Pax-3 transcripts are first detected in 8.5 day mouse embryos where they are restricted to the dorsal part of the neuroepithelium and to the adjacent segmented dermomyotome. During early neurogenesis, Pax-3 expression is limited to mitotic cells in the ventricular zone of the developing spinal cord and to distinct regions in the hindbrain, midbrain and diencephalon. In 10-12 day embryos, expression of Pax-3 is also seen in neural crest cells of the developing spinal ganglia, the craniofacial mesectoderm and in limb mesenchyme of 10 and 11 day embryos.