A genetic model for interaction of the homeodomain recognition helix with DNA

Altering cell fates in sea urchin embryos by overexpressing SpOtx, an orthodenticle-related protein

While many general features of cell fate specification in the sea urchin embryo are understood, specific factors associated with these events remain unidentified. SpOtx, an orthodenticle-related protein, has been implicated as a transcriptional activator of the aboral ectoderm-specific Spec2a gene. Here, we present evidence that SpOtx has the potential to alter cell fates. SpOtx was found in the cytoplasm of early cleavage stage embryos and was translocated into nuclei between the 60- and 120-cell stage, coincident with Spec gene activation. Eggs injected with SpOtx mRNA developed into epithelial balls of aboral ectoderm suggesting that SpOtx redirected nonaboral ectoderm cells to an aboral ectoderm fate. At least three distinct domains on SpOtx, the homeobox and regions in the N-terminal and C-terminal halves of the protein, were required for the morphological alterations. These same N-terminal and C-terminal regions were shown to be transactivation domains in a yeast transactivation assay, indicating that the biological effects of overexpressing SpOtx were due to its action as a transcription factor. Our results suggest that SpOtx is involved in aboral ectoderm differentiation by activating aboral ectoderm-specific genes and that modulating its expression can lead to changes in cell fate.

Target gene identification: target specific transcriptional activation by three murine homeodomain/VP16 hybrid proteins in Saccharomyces cerevisiae

The mammalian homeodomain proteins encoded by Hox genes play an important role in embryonic development by providing positional queues which define developmental identities along the anteroposterior axis of developing organisms. These proteins bind DNA specifically through their homeodomain to sequences containing ATTA cores, and thereby are thought to exert their effect regulating downstream genes. Little is known about the specificity of binding of homeodomain proteins to their sequences and the identity of their target genes. We have developed a transcriptional activation assay in yeast which employs a homeobox/VP16 fusion gene as a transcriptional activator and a target construct in which test fragments of DNA are inserted upstream to a reporter gene. Using this assay, we compared transcriptional activation by three chimeric proteins containing the homeodomains of the mouse homeobox genes, Hoxa-5, Hoxb-6, and Hoxc-8. When tested on previously defined target sequences, strong differential specificities of activation were observed. In an effort to identify enhancers that normally respond to homeodomain transcriptional activators, random fragments of mouse genomic DNA were cloned upstream of the reporter gene. Genomic DNA fragments with distinct activation profiles were obtained and were found to share matches beyond the ATTA core with previously described enhancers. These results demonstrate that the transcriptional activation system in yeast can be used as a convenient system to detect DNA motifs which bind homeodomain proteins, and subsequently, to identify authentic target genes responsive to Hox gene proteins.

Alternative splicing of the Drosophila melanogaster rotundRacGAP gene

The rotund (rn) gene in Drosophila melanogaster codes for a RacGTPase-activating protein, RnRacGAP. Cellular studies have shown that RacGAP proteins function as negative regulators of substrate Rac proteins which, in turn, control the localization and polymerization state of actin within the cell. Previous sequence analysis of rn genomic DNA and incomplete cDNA clones suggested that at least two differentially spliced forms of the transcript exist, rnRacGAP(1) and rnRacGAP(2). Using nested reverse transcription-polymerase chain reaction (RT-PCR) methods, we have cloned missing exon and intron sequences, and detected differences between rnRacGAP(1) and rnRacGAP(2) involving 24 nucleotides (nt) of coding sequences and 119 nt of 3'UTR. This translates to a difference of seven amino acids at the C-termini of the polypeptide products. Utilization, in RT-PCR analysis, of form-specific primers provided a simple assay for the tissue specificity of expression of the two forms. rnRacGAP(1) is the predominant species in the testes and is expressed at a low level in the ovary and somatic tissues. rnRacGAP(2) is only very weakly expressed and is detectable solely in the testes.

rHox: a homeobox gene expressed in osteoblastic cells

Homeodomain proteins are characterized by a conserved domain with a helix-turn-helix motif. These proteins act as regulatory factors in tissue differentiation and proliferation. However, their role in the regulation of osteoblast differentiation is unknown. In this study we have identified and characterized a homeobox gene in osteoblast-like cells. This gene, termed rHox, was isolated from a cDNA library derived from rat osteoblast-like cells. The nucleotide sequence of the 1,375 base pair (bp) cDNA contains a noncoding leader sequence of 329 bp, a 735 bp open reading frame, and 312 bp of 3' noncoding sequence. Sequence comparison demonstrates that rHox is identical to the mouse Pmx gene (also called MHox) at the amino acid level and 90% homologous at the nucleotide level. Both Southwestern blotting and gel shift analyses indicate that rHox has potential to bind both the collagen I alpha 1 and the osteocalcin promoters. Transfection experiments using an rHox expression vector showed a strong repression of target promoter activity, regardless of whether the target promoters contained homeodomain binding response elements. These data suggest that rHox is a potent negative regulator of gene expression, although the specific role of rHox in bone gene regulation remains to be determined.

Crystal structure of the MATa1/MAT alpha 2 homeodomain heterodimer bound to DNA

The Saccharomyces cerevisiae MATa1 and MAT alpha 2 homeodomain proteins, which play a role in determining yeast cell type, form a heterodimer that binds DNA and represses transcription in a cell type-specific manner. Whereas the alpha 2 and a1 proteins on their own have only modest affinity for DNA, the a1/alpha 2 heterodimer binds DNA with high specificity and affinity. The three-dimensional crystal structure of the a1/alpha 2 homeodomain heterodimer bound to DNA was determined at a resolution of 2.5 A. The a1 and alpha 2 homeodomains bind in a head-to-tail orientation, with heterodimer contacts mediated by a 16-residue tail located carboxyl-terminal to the alpha 2 homeodomain. This tail becomes ordered in the presence of a1, part of it forming a short amphipathic helix that packs against the a1 homeodomain between helices 1 and 2. A pronounced 60 degree bend is induced in the DNA, which makes possible protein-protein and protein-DNA contacts that could not take place in a straight DNA fragment. Complex formation mediated by flexible protein-recognition peptides attached to stably folded DNA binding domains may prove to be a general feature of the architecture of other classes of eukaryotic transcriptional regulators.

Genetic identification of the site of DNA contact in the yeast heat shock transcription factor

The heat shock transcription factor (HSF), a trimeric transcription factor, activates the expression of heat shock genes in eukaryotes. We have isolated mutations in the HSF1 gene from Saccharomyces cerevisiae that severely compromise the ability of HSF to bind to its normal binding site, repeats of the module nGAAn. One of these mutations, Q229R, shows a "new specificity" phenotype, in which the protein prefers the mutant sequence nGACn. These results identify the region of HSF that contacts DNA, in complete agreement with the crystal structure of HSF of Kluyveromyces lactis and the nuclear magnetic resonance data from HSF of Drosophila melanogaster. To determine the orientation of the DNA-binding domain on the nGAAn motif, we performed site-specific cross-linking between cysteine residues of single-cysteine substitutions. Cysteines placed at the N terminus of the DNA contact helix formed cross-links readily, while cysteines placed at the C terminus of the helix did not.

Isolation and characterization of the chicken homeodomain protein AKR

Expression of the avian apoVLDLII gene is liver specific and completely dependent on estrogen. Previous analyses of protein binding sites in the apoVLDLII promoter revealed interactions between liver-enriched and ubiquitous factors at a location, site F', between nucleotides -229 and -260 relative to the major transcriptional start site. Site-directed mutagenesis of G residues contacted by these factors decreased expression from the promoter approximately 5-fold in the avian hepatoma cell line LMH2A. We have used this site to screen a cDNA expression library constructed from day 9 embryonic liver RNA. One of the two DNA binding factors isolated is a novel homeodomain protein. With the exception of the homeodomain itself, which is atypically located close to the protein N-terminus, the factor displays little similarity to any known DNA binding protein. Its homeodomain is most similar to that of the maize protein Knotted-1, while the most closely related vertebrate domain is that of the human proto-oncoprotein Pbx1. We demonstrate that the DNA binding specificity of the factor is consistent with its involvement in the ubiquitous complex formed with site F' and that it is capable of suppressing expression from the apoVLDLII promoter in short-term transfection experiments. These studies, combined with its DNA binding specificity, the tissue distribution of its mRNA and its developmental regulation, suggest a role as a negative regulator of gene expression in non-hepatic tissues and in the liver early during embryogenesis.

The optimal binding sequence of the Hox11 protein contains a predicted recognition core motif

HOX11 is a homeobox-containing oncogene of specific T-cell leukemias. We determined the DNA binding specificity of the Hox11 protein by using a novel technique of random oligonucleotide selection developed in this study. The optimal Hox11 binding sequence, GGCGGTAAGTGG, contained a core TAAGTG motif that is consistent with a prediction based on the residues at specific positions that potentially make DNA base contacts and models of homeodomain-DNA interaction proposed from studies with other homeodomains. The specific interaction between Hox11 and the selected optimal binding sequence was further confirmed by band-shift and DNA competition assays. Given that the Hox11 homeodomain shares low homology with other well studied homeodomains, the presence of a predictable recognition core motif in its optimal binding sequence supports the notion that different homeodomains interact with DNA in a similar manner, through highly conserved residues at specific positions that allow contact with DNA.

Sequence-specific targeting of nuclear signal transduction pathways by homeodomain proteins

Cells translate extracellular signals into specific programs of gene expression that reflect their developmental history or identity. We present evidence that one way this interpretation may be performed is by cooperative interactions between serum response factor (SRF) and certain homeodomain proteins. We show that human and Drosophila homeodomain proteins of the paired class have the ability to recruit SRF to DNA sequences not efficiently recognized by SRF on its own, thereby imparting to a linked reporter gene the potential to respond to polypeptide growth factors. This activity requires both the DNA-binding activity of the homeodomain and putative protein-protein contact residues on the exposed surfaces of homeodomain helices 1 and 2. The ability of the homeodomain to impart signal responsiveness is DNA sequence specific, and this specificity differs from the simple DNA-binding specificity of the homeodomain in vitro. The homeodomain imparts response to a spectrum of signals characteristic of the natural SRF-binding site in the c-fos gene. Response to some of these signals is dependent on the secondary recruitment of SRF-dependent ternary complex factors, and we show directly that a homeodomain can promote the recruitment of one such factor, Elk1. We infer that SRF and homeodomains interact cooperatively on DNA and that formation of SRF-homeodomain complexes permits the recruitment of signal-responsive SRF accessory proteins. The ability to route extracellular signals to specific target genes is a novel activity of the homeodomain, which may contribute to the identity function displayed by many homeodomain genes.

PCR-mediated recombination and mutagenesis. SOEing together tailor-made genes

Gene Splicing by Overlap Extension (gene SOEing) is a sequence-independent method for site-directed mutagenesis and/or recombination of DNA molecules. It is based on the idea that a PCR product can be engineered by adding or changing sequences at its ends so that the product can itself be used to prime DNA synthesis in a subsequent overlap-extension reaction to create mutant or recombinant molecules. As the engineered genes are created in vitro without reliance on host organisms or restriction sites, gene SOEing provides a powerful and versatile tool for genetic investigation and engineering.

Transcription factor GATA-1 regulates human HOXB2 gene expression in erythroid cells

The human HOXB2 gene is a member of the vertebrate Hox gene family that contains genes coding for specific developmental stage DNA-binding proteins. Remarkably, within the hematopoietic compartment, genes of the HOXB complex are expressed specifically in erythromegakaryocytic cell lines and, for some of them, in hematopoietic progenitors. Here, we report the study of HOXB2 gene transcriptional regulation in hematopoietic cells, an initial step in understanding the lineage-specific expression of the whole HOXB complex in these cells. We have isolated the HOXB2 5'-flanking sequence and have characterized a promoter fragment extending 323 base pairs upstream from the transcriptional start site, which, in transfection experiments, was sufficient to direct the tissue-specific expression of HOXB2 in the erythroid cell line K562. In this fragment, we have identified a potential GATA-binding site that is essential to the promoter activity as demonstrated by point mutation experiments. Gel shift analysis revealed the formation of a specific complex in both erythroleukemic lines K562 and HEL that could be prevented by the addition of a specific antiserum raised against GATA-1 protein. These findings suggest a regulatory hierarchy in which GATA-1 is upstream of the HOXB2 gene in erythroid cells.

Analysis of the conformation and stability of rat TTF-1 homeodomain by circular dichroism

The conformational stability of TTF-1HD has been determined by CD-monitored thermal denaturation and isothermal urea unfolding studies. The Gibbs free energy of stabilization found are 1.44 and 1.26 kcal.mol-1, respectively. TTF-1HD exhibits a Tm of 42 degrees C and a delta Cp of 80 cal.mol-1.K-1 indicating that TTF-1HD, when free in solution, is a mobile flexible segment folded into loose helices. Such a flexibility would be relevant for the DNA-binding function of this homeodomain. In fact, a small reduction of the alpha-helical content of TTF-1HD significantly modifies its DNA-binding activity.

Promoter analysis meets pattern formation: transcriptional regulatory genes in sea urchin embryogenesis

Analyses of spatial and temporal gene control mechanisms in the sea urchin embryo have identified several important trans-regulatory factors, including some that are related to known developmental control genes of the fly and mouse. Recent advances in gene perturbation technologies, including the use of antisense oligonucleotides to target mRNAs in early-stage embryos, as well as the injection of mRNAs into zygotes to express genes ectopically, have made it possible to test the functions of such factors directly.

Homeodomain determinants of major groove recognition

The homeodomain is a highly conserved structural module that binds DNA and participates in protein-protein interactions. Most homeodomains contain residues at positions 47 and 51 which mediate recognition of a TAAT core binding sequence in the major groove. The constraints imposed on the identity of these residues by homeodomain structure and DNA docking have been examined in the context of the POU domain of the Oct-1 transcription factor. A bacterial library, in which POU homeodomain residues 47 and 51 have been randomized, was probed on nitrocellulose filters for the binding of DNA fragments containing the consensus octamer sequence. The residues which provide for the highest affinity interaction with the octamer consensus sequence, and the greatest specificity, are the highly conserved wild-type residues valine 47 and asparagine 51. Interestingly, a class of variants containing arginine at position 51 was also detected in the screen and found to have moderate affinity for the consensus sequence but reduced specificity compared to the wild-type protein. A single variant containing arginine at both positions 47 and 51 was detected when the library was probed with fragments containing nucleotide substitutions at positions expected to be contacted by residues 47 and 51. This variant was used to alter the DNA-binding specificity of a transcriptional regulatory complex which depends upon Oct-1 for DNA recognition. These findings suggest that homeodomain structure and DNA docking constrain in the versatility of the domain in that only a limited set of amino acid determinants can endow the domain with specific, high-affinity DNA binding.

Functional differences between HOX proteins conferred by two residues in the homeodomain N-terminal arm

Hox genes encode homeodomain-containing transcriptional regulators that function during development to specify positional identity along embryonic axes. The homeodomain is composed of a flexible N-terminal arm and three alpha helices, and it differentially binds DNA. A number of homeodomains recognize sites containing a TAAT core motif. The product of the murine Hoxd-4 (Hox-4.2) gene functions in a positive autoregulatory fashion in P19 cells that is dependent on two TAAT motifs in the Hoxd-4 promoter. This effect is specific in that murine HOXA-1 (HOX-1.6) is unable to activate transcription through the Hoxd-4 autoregulatory element. Here we show that this is due to an inability of the HOXA-1 homeodomain to bind a HOXD-4 recognition site effectively. We have produced chimeras between HOXD-4 and HOXA-1 to map specific residues responsible for this functional difference. When positions 2 and 3 in the N-terminal arm of HOXA-1 were converted to HOXD-4 identity, both strong DNA binding and transcriptional activation were rescued. This substitution appears to confer an increased DNA-binding ability on the HOXA-1 homeodomain, since we were unable to detect a high-affinity recognition sequence for HOXA-1 in a randomized pool of DNA probes. The contribution of position 3 to DNA binding has been implicated by structural studies, but this is the first report of the importance of position 2 in regulating homeodomain-DNA interactions. Additionally, specific homeodomain residues that confer major differences in DNA binding and transcriptional activation between Hox gene products have not been previously determined. Identity at these two positions is generally conserved among paralogs but varies between Hox gene subfamilies. As a result, these residues may be important for the regulation of target gene expression by specific Hox products.

Sequence-specific DNA recognition by the thyroid transcription factor-1 homeodomain

The molecular basis for the DNA binding specificity of the thyroid transcription factor 1 homeodomain (TTF-1HD) has been investigated. Methylation and ethylation interference experiments show that the TTF-1HD alone recapitulates the DNA binding properties of the entire protein. Studies carried out with mutant derivatives of TTF-1HD indicate a precise correspondence of some of its amino acid residues with specific bases in its binding site, allowing a crude orientation of the TTF-1HD within the protein-DNA complex. TTF-1HD shows an overall geometry of interaction with DNA similar to that previously observed for Antennapedia class HDs, even though the binding specificities of these two types of HDs are distinct. We demonstrate that the crucial difference between the binding sites of Antennapedia class and TTF-1 HDs is in the motifs 5'-TAAT-3', recognized by Antennapedia, and 5'-CAAG-3', preferentially bound by TTF-1. Furthermore, the binding of wild type and mutants TTF-1 HD to oligonucleotides containing either 5'-TAAT-3' or 5'-CAAG-3' indicate that only in the presence of the latter motif the Gln50 in TTF-1 HD is utilized for DNA recognition. Since the Gln at position 50 is an essential determinant for DNA binding specificity for several other HDs that bind to 5'-TAAT-3' containing sequences, we suggest that utilization by different HDs of key residues may depend on the sequence context and probably follows a precise hierarchy of contacts.

Differential DNA-binding specificity of the engrailed homeodomain: the role of residue 50

To assess the importance of residue 50 in determining the binding specificity of the homeodomain from the engrailed transcription factor of Drosophila, the DNA-binding properties of isolated homeodomains containing glutamine (wild type), alanine, and lysine at this position have been studied. In binding site selection experiments using the wild-type engrailed homeodomain, TAATTA was identified as a high-affinity, consensus binding site. When the glutamine at position 50 was replaced by a lysine (QK50), the binding site preference changed to TAATCC. The half-life and affinity of the complex between the QK50 protein and a DNA site containing TAATCC were increased significantly compared to the half-life and affinity of the complex between the wild-type protein and a TAATTA site. This suggests that Lys50 forms a more favorable interaction with the TAATCC DNA than Gln50 does with the TAATTA site. In fact, the wild-type Gln50 side chain (which forms a hydrophobic interaction with the last A:T base pair of the TAATTA site in the cocrystal structure [Kissinger, C. R., Liu, B., Martin-Blanco, E., Kornberg, T. B., & Pabo, C. O. (1990) Cell 63, 579-590]) appears to play only a small role in determining binding affinity and specificity for the TAATTA site, as the QA50 mutant has only a 2-fold reduced affinity for the TAATTA site and discriminates between the TAATTA and TAATCC sites as well as the wild-type protein. As a result, determinants in addition to Gln50 must be involved in establishing the differential binding specificity of the engrailed homeodomain.

Functional differences between HOX proteins conferred by two residues in the homeodomain N-terminal arm

Hox genes encode homeodomain-containing transcriptional regulators that function during development to specify positional identity along embryonic axes. The homeodomain is composed of a flexible N-terminal arm and three alpha helices, and it differentially binds DNA. A number of homeodomains recognize sites containing a TAAT core motif. The product of the murine Hoxd-4 (Hox-4.2) gene functions in a positive autoregulatory fashion in P19 cells that is dependent on two TAAT motifs in the Hoxd-4 promoter. This effect is specific in that murine HOXA-1 (HOX-1.6) is unable to activate transcription through the Hoxd-4 autoregulatory element. Here we show that this is due to an inability of the HOXA-1 homeodomain to bind a HOXD-4 recognition site effectively. We have produced chimeras between HOXD-4 and HOXA-1 to map specific residues responsible for this functional difference. When positions 2 and 3 in the N-terminal arm of HOXA-1 were converted to HOXD-4 identity, both strong DNA binding and transcriptional activation were rescued. This substitution appears to confer an increased DNA-binding ability on the HOXA-1 homeodomain, since we were unable to detect a high-affinity recognition sequence for HOXA-1 in a randomized pool of DNA probes. The contribution of position 3 to DNA binding has been implicated by structural studies, but this is the first report of the importance of position 2 in regulating homeodomain-DNA interactions. Additionally, specific homeodomain residues that confer major differences in DNA binding and transcriptional activation between Hox gene products have not been previously determined. Identity at these two positions is generally conserved among paralogs but varies between Hox gene subfamilies. As a result, these residues may be important for the regulation of target gene expression by specific Hox products.

A homology-based molecular model of the proline-rich homeodomain protein Prh, from haematopoietic cells

A molecular structural model for the homeodomain of the haematopoietic protein Prh together with its DNA recognition sequence, has been built using the known crystal structure of the MAT alpha 2 homeodomain as a starting-point. The modelling procedure used main and side-chain optimisations by means of molecular mechanics/simulated annealing procedures to obtain stereochemically plausible geometries. The resulting structure has a number of specific interactions in both major and minor grooves of the DNA that serve to define the consensus binding sequence for Prh. In particular, the side-chain of glutamine 50 is postulated to be involved in hydrogen bonds to adjacent adenine and cytosine bases within the consensus sequence.

Genetic identification of residues involved in association of alpha and beta G-protein subunits

The GPA1, STE4, and STE18 genes of Saccharomyces cerevisiae encode the alpha, beta, and gamma subunits, respectively, of a G protein involved in the mating response pathway. We have found that mutations G124D, W136G, W136R, and delta L138 and double mutations W136R L138F and W136G S151C of the Ste4 protein cause constitutive activation of the signaling pathway. The W136R L138F and W136G S151C mutant Ste4 proteins were tested in the two-hybrid protein association assay and found to be defective in association with the Gpa1 protein. A mutation at position E307 of the Gpa1 protein both suppresses the constitutive signaling phenotype of some mutant Ste4 proteins and allows the mutant alpha subunit to physically associate with a specific mutant G beta subunit. The mutation in the Gpa1 protein is adjacent to the hinge, or switch, region that is required for the conformational change which triggers subunit dissociation, but the mutation does not affect the interaction of the alpha subunit with the wild-type beta subunit. Yeast cells constructed to contain only the mutant alpha and beta subunits mate and respond to pheromones, although they exhibit partial induction of the pheromone response pathway. Because the ability of the modified G alpha subunit to suppress the Ste4 mutations is allele specific, it is likely that the residues defined by this analysis play a direct role in G-protein subunit association.

Specific DNA recognition and intersite spacing are critical for action of the bicoid morphogen

We examined DNA site recognition by Bicoid and its importance for pattern formation in developing Drosophila embryos. Using altered DNA specificity Bicoid mutants and appropriate reporter genes, we show that Bicoid distinguishes among related DNA-binding sites in vivo by a specific contact between amino acid 9 of its recognition alpha-helix (lysine 50 of the homeodomain) and bp 7 of the site. This result is consistent with our earlier results using Saccharomyces cerevisiae but differs from that predicted by crystallographic analysis of another homeodomain-DNA interaction. Our results also demonstrate that Bicoid binds directly to those genes whose transcription it regulates and that the amino acid 9 contact is necessary for Bicoid to direct anterior pattern formation. In both Drosophila embryos and yeast cells, Bicoid requires multiple binding sites to activate transcription of target genes. We find that the distance between binding sites is critical for Bicoid activation but that, unexpectedly, this critical distance differs between Drosophila and S. cerevisiae. This result suggests that Bicoid activation in Drosophila might require an ancillary protein(s) not present in S. cerevisiae.

Mapping critical residues in eukaryotic DNA-binding proteins: a plasmid-based genetic selection strategy with application to the Oct-2 POU motif

Discrimination between allowed and disallowed amino acid substitutions provides a powerful method for analysis of protein structure and function. Site-directed mutagenesis allows specific hypotheses to be tested, but its systematic application to entire structural motifs is inefficient. This limitation may be overcome by genetic selection, which allows rapid scoring of thousands of randomly (or pseudorandomly) generated mutants. To facilitate structural dissection of DNA-binding proteins, we have designed two generally applicable bacterial selection systems: pPLUS selects for the ability of a protein to bind to a user-defined DNA sequence, whereas pMINUS selects against such binding. Complementary positive and negative selections allow rapid mapping of critical residues. To test and calibrate the systems, we have investigated the bipartite POU domain of the human B-cell-specific transcription factor Oct-2. (i) An invariant residue (Asn347) in the DNA-recognition helix of the POU-specific homeodomain (POUHD) was substituted by each of the 19 other possible amino acids. The mutant proteins each exhibited decreased specific DNA binding as defined in vivo by genetic selection and in vitro by gel retardation; relative affinities correlate with phenotypes in the positive and negative selection systems. An essential role for Asn347 in wild-type POUHD-DNA recognition is consistent with homologous Asn-adenine interactions in cocrystal structures of canonical homeodomains. (ii) Extension of pPLUS/pMINUS selection to the POU-specific subdomain (POUs) is demonstrated by analysis of mutations in its putative helix-turn-helix (HTH) element.(ABSTRACT TRUNCATED AT 250 WORDS)

Specific DNA recognition and intersite spacing are critical for action of the bicoid morphogen

We examined DNA site recognition by Bicoid and its importance for pattern formation in developing Drosophila embryos. Using altered DNA specificity Bicoid mutants and appropriate reporter genes, we show that Bicoid distinguishes among related DNA-binding sites in vivo by a specific contact between amino acid 9 of its recognition alpha-helix (lysine 50 of the homeodomain) and bp 7 of the site. This result is consistent with our earlier results using Saccharomyces cerevisiae but differs from that predicted by crystallographic analysis of another homeodomain-DNA interaction. Our results also demonstrate that Bicoid binds directly to those genes whose transcription it regulates and that the amino acid 9 contact is necessary for Bicoid to direct anterior pattern formation. In both Drosophila embryos and yeast cells, Bicoid requires multiple binding sites to activate transcription of target genes. We find that the distance between binding sites is critical for Bicoid activation but that, unexpectedly, this critical distance differs between Drosophila and S. cerevisiae. This result suggests that Bicoid activation in Drosophila might require an ancillary protein(s) not present in S. cerevisiae.

Genetic identification of residues involved in association of alpha and beta G-protein subunits

The GPA1, STE4, and STE18 genes of Saccharomyces cerevisiae encode the alpha, beta, and gamma subunits, respectively, of a G protein involved in the mating response pathway. We have found that mutations G124D, W136G, W136R, and delta L138 and double mutations W136R L138F and W136G S151C of the Ste4 protein cause constitutive activation of the signaling pathway. The W136R L138F and W136G S151C mutant Ste4 proteins were tested in the two-hybrid protein association assay and found to be defective in association with the Gpa1 protein. A mutation at position E307 of the Gpa1 protein both suppresses the constitutive signaling phenotype of some mutant Ste4 proteins and allows the mutant alpha subunit to physically associate with a specific mutant G beta subunit. The mutation in the Gpa1 protein is adjacent to the hinge, or switch, region that is required for the conformational change which triggers subunit dissociation, but the mutation does not affect the interaction of the alpha subunit with the wild-type beta subunit. Yeast cells constructed to contain only the mutant alpha and beta subunits mate and respond to pheromones, although they exhibit partial induction of the pheromone response pathway. Because the ability of the modified G alpha subunit to suppress the Ste4 mutations is allele specific, it is likely that the residues defined by this analysis play a direct role in G-protein subunit association.

Ingrowth by photoreceptor axons induces transcription of a retrotransposon in the developing Drosophila brain

The development of the lamina, the first optic ganglion of the fly visual system, depends on inductive cues from the innervating photoreceptor axons. lacZ expression from a P-element insertion, A72, occurs in the anlage of the lamina coincident with axon ingrowth from the eye imaginal disc. In eyeless mutants lacking photoreceptor axons, lacZ expression did not occur. The P-element was found to have inserted within the 3′ long terminal repeat (LTR) of a ‘17.6′ type retrotransposon. The expression pattern of 17.6 transcripts in the brain in wild-type and eyeless mutants paralleled the expression of the lacZ reporter. Analysis of 17.6 cis-regulatory sequences indicates that the lamina-specific expression is due to the combined action of an enhancer element in the LTR and a repressor element within the internal body of the retrotransposon. The regulation of the 17.6 retrotransposon provides a model for the study of innervation-dependent gene expression in postsynaptic cells during neurogenesis.

A framework for the DNA-protein recognition code of the probe helix in transcription factors: the chemical and stereochemical rules

BACKGROUND

Understanding the general mechanisms of sequence specific DNA recognition by proteins is a major challenge in structural biology. The existence of a 'DNA recognition code' for proteins, by which certain amino acid residues on a protein surface confer specificity for certain DNA bases, has been the subject of much discussion. However, no simple code has yet been established.

RESULTS

The principles of DNA recognition can be described at two levels. The 'chemical' rules describe the partnerships between amino acid side chains and DNA bases making favourable interactions in the major groove of DNA. Here I analyze the occurrence of nucleotide-amino acid contacts in previously determined crystal structures of DNA-protein complexes and find that simple rules pertain. I also describe 'stereochemical' rules for the probe helix type of DNA-binding motif found in certain transcription factors including leucine zipper and homeodomain proteins. These are a consequence of the binding geometry, and specify the amino acid and base positions used for the contacts, and the sizes of residues in the contact interface.

CONCLUSIONS

The chemical rules can be generalized for any DNA-binding motif, while the stereochemical rules are specific to a particular DNA-binding motif. The recognition code for a particular binding motif can be described by combining the two sets of rules.

Direct regulation of decapentaplegic by Ultrabithorax and its role in Drosophila midgut morphogenesis

Drosophila homeotic genes encode transcription factors thought to control segmental identity by regulating expression of largely unknown target genes. The formation of the second midgut constriction requires the Ultrabithorax (Ubx) and abdominal-A (abd-A) homeotic genes and decapentaplegic (dpp), a gene encoding a member of the TGF beta family of proteins. We identified a 674 bp enhancer of dpp controlling its expression in the second constriction domain of the visceral mesoderm (parasegment 7). Normal enhancer function requires positive regulation by Ubx and negative regulation by abd-A. This enhancer contains UBX- and ABD-A-binding sites defined in vitro. By generating complementary alterations of the binding sites and the binding specificity of UBX, we show that Ubx directly regulates dpp expression. These regulatory interactions are relevant to normal development, because a transgene made with this enhancer driving a dpp transcription unit rescues the second midgut constriction and larval lethality phenotypes of dpps mutations.

Mutation of a single lysine residue severely impairs the DNA recognition and regulatory functions of the VZV gene 62 transactivator protein

The product of varicella-zoster virus gene 62 (VZV 140k) is a potent transactivator protein. We have identified a region within the DNA binding domain of VZV 140k that shows a striking similarity to the DNA recognition helix of the homeodomain, with an especially highly conserved quartet of residues, WLQN. The 140k protein has functional counterparts within the other alphaherpesviruses, which include the major transcriptional regulatory protein of HSV-1, (ICP4), and the WLQN region is highly conserved among the members of this family of viral transactivators. Substitution of VZV 140k residue lysine 548, just adjacent to the WLQN region, drastically reduces the DNA binding activity of the 140k DNA binding domain and the intact 140k mutant protein fails to activate gene expression. Substitutions of two other VZV 140k residues in this conserved WLQN region result in alterations to the DNA binding interaction and reduced transactivation activities. All three mutations act at the level of DNA recognition, as they have no apparent effect on the dimerization state, solubility or efficiency of expression of the mutant peptides.

A proline-rich transcriptional activation domain in murine HOXD-4 (HOX-4.2)

The product of the murine Hoxd-4 (Hox-4.2) gene is a transcription factor that acts upon an autoregulatory element in Hoxd-4 upstream sequences (1). Using this activity as an assay in transient transfections of P19 embryonal carcinoma (EC) cells, we performed a mutational analysis to map functional domains in the HOXD-4 protein. The importance of the homeodomain was shown by a single amino acid change in this region that abolished activity. Deletion analysis revealed that many evolutionarily conserved regions outside of the homeodomain were dispensable for activation in our assay. Fusions to the GAL4 DNA-binding domain mapped a transcriptional activation function to the HOXD-4 proline-rich N-terminus. The proline-rich transcription factor AP2 squelched activation by HOXD-4 and by GAL4/HOXD-4 N-terminus fusion proteins. Together, these results suggest that HOXD-4 harbors a transcriptional activation domain of the proline-rich type.

Functional analysis of mouse Hoxa-7 in Saccharomyces cerevisiae: sequences outside the homeodomain base contact zone influence binding and activation

The murine developmental control gene product, Hoxa-7, was shown to function as a DNA-binding transactivator in Saccharomyces cerevisiae. The importance of the ATTA core, the preference for antp class flanking nucleotides, the importance of Asn-51 of the homeodomain (HD), and the synergism of multiple binding sites all reflect properties that have previously been described for HOM or Hox proteins in tissue culture systems. A comparison of contact positions among genes of paralog groups and classes of mammalian HDs points to a lack of diversity in positions that make base contact, suggesting that besides the combination of HD amino acid-base pair contacts, another means of recognizing differences between targets must exist if Hox genes select different targets. The HD of antennapedia is identical to the Hoxa-7 HD. The interaction of Hoxa-7 with the exact sequence used in the nuclear magnetic resonance three-dimensional structural analysis on the antennapedia HD was studied. Hoxa-7 binding and transactivation was influenced by sequences outside of the known base contact zone of this site. We conclude that Hoxa-7 protein has a second means to interact with DNA or/and that the sequences flanking the base contact zone influence HD interactions by distorting DNA within the contact zone (base or backbone). This result is discussed in terms of DNA flexure and two modes of transcription used in S. cerevisiae.

Homeodomain proteins in development and therapy

Homeobox genes encode transcriptional regulators found in all organisms ranging from yeast to humans. In Drosophila, a specific class of homeobox genes, the homeotic genes, specifies the identity of certain spatial units of development. Their genomic organization, in Drosophila, as well as in vertebrates, is uniquely connected with their expression which follows a 5'-posterior-3'-anterior rule along the longitudinal body axis. The 180-bp homeobox is part of the coding sequence of these genes, and the sequence of 60 amino acids it encodes is referred to as the homeodomain. Structural analyses have shown that homeodomains consist of a helix-turn-helix motif that binds the DNA by inserting the recognition helix into the major groove of the DNA and its amino-terminal arm into the adjacent minor groove. Developmental as well as gene regulatory functions of homeobox genes are discussed, with special emphasis on one group, the Antennapedia (Antp) class homeobox genes and a representative 60-amino acid Antennapedia peptide (pAntp). In cultured neuronal cells, pAntp translocates through the membrane specifically and efficiently and accumulates in the nucleus. The internalization process is followed by a strong induction of neuronal morphological differentiation, which raises the possibility that motoneuron growth is controlled by homeodomain proteins. It has been demonstrated that chimeric peptide molecules encompassing pAntp are also captured by cultured neurons and conveyed to their nuclei. This may be of enormous interest for the internalization of drugs.

Functional analysis of mouse Hoxa-7 in Saccharomyces cerevisiae: sequences outside the homeodomain base contact zone influence binding and activation

The murine developmental control gene product, Hoxa-7, was shown to function as a DNA-binding transactivator in Saccharomyces cerevisiae. The importance of the ATTA core, the preference for antp class flanking nucleotides, the importance of Asn-51 of the homeodomain (HD), and the synergism of multiple binding sites all reflect properties that have previously been described for HOM or Hox proteins in tissue culture systems. A comparison of contact positions among genes of paralog groups and classes of mammalian HDs points to a lack of diversity in positions that make base contact, suggesting that besides the combination of HD amino acid-base pair contacts, another means of recognizing differences between targets must exist if Hox genes select different targets. The HD of antennapedia is identical to the Hoxa-7 HD. The interaction of Hoxa-7 with the exact sequence used in the nuclear magnetic resonance three-dimensional structural analysis on the antennapedia HD was studied. Hoxa-7 binding and transactivation was influenced by sequences outside of the known base contact zone of this site. We conclude that Hoxa-7 protein has a second means to interact with DNA or/and that the sequences flanking the base contact zone influence HD interactions by distorting DNA within the contact zone (base or backbone). This result is discussed in terms of DNA flexure and two modes of transcription used in S. cerevisiae.

The Drosophila copia retrotransposon contains binding sites for transcriptional regulation by homeoproteins

We have identified in the 5' untranslated region of the Drosophila copia retrotransposon, 3' to the left LTR, a sequence for transcriptional regulation by homeoproteins. Co-transfection assays using expression vectors for homeoproteins and reporter vectors containing the lacZ gene under the control of either the entire copia LTR with 5' untranslated sequence, or a minimal heterologous promoter flanked with a 130 bp fragment containing the copia untranslated region, disclosed both positive and negative modulations of promoter activity in Drosophila cells in culture: a 5-10 fold decrease with engrailed, even-skipped and zerknüllt in DH33 cells, and a 10-30 fold increase with fushi tarazu and zerknüllt in Schneider II cells. In all cases, the regulatory effects were abolished with reporter plasmids deleted for a 58 bp fragment encompassing the putative homeoprotein binding sites. Mobility shift assays with a purified homeodomain-containing peptide demonstrated direct interaction with the 58 bp fragment, with an affinity in the 1-10 nM range as reported with the same peptide for other well characterized homeodomain binding regulatory sites. Foot-printing experiments with the extended LTR demonstrated protection of 'consensus' sequences, located within the 58 bp fragment. These homeodomain binding sites could be involved in the developmental regulation of the copia retrotransposon.

Cooperative dimerization of paired class homeo domains on DNA

Homeo domain-containing proteins mediate many transcriptional processes in eukaryotes. Because nearly all animal homeo proteins are believed to bind to short, highly related DNA sequences, the basis for their high specificity of action is not understood. We show that cooperative dimerization on palindromic DNA sequences can provide increased specificity to one of the three major classes of homeo domains, the Paired/Pax class. The 60-amino-acid homeo domains from this class contain sufficient information to bind cooperatively as homo- and heterodimers to palindromic DNA sequences; that is, the binding of one homeo domain molecule can increase the affinity of a second molecule by up to 300-fold. Different members of the Paired (Prd) class of homeo domains prefer different spacings between half-sites, as determined by the ninth amino acid residue of the recognition helix. In addition, this residue determines the identity of the base pairs at the center of the palindromic sites, as well as the magnitude of the cooperative interaction. The cooperative dimerization of homeo domains in the Prd class distinguishes them from other classes, whereas binding-site configuration and sequence specificity allow for distinctions within this class.

Isolation and characterization of target sequences of the chicken CdxA homeobox gene

The DNA binding specificity of the chicken homeodomain protein CDXA was studied. Using a CDXA-glutathione-S-transferase fusion protein, DNA fragments containing the binding site for this protein were isolated. The sources of DNA were oligonucleotides with random sequence and chicken genomic DNA. The DNA fragments isolated were sequenced and tested in DNA binding assays. Sequencing revealed that most DNA fragments are AT rich which is a common feature of homeodomain binding sites. By electrophoretic mobility shift assays it was shown that the different target sequences isolated bind to the CDXA protein with different affinities. The specific sequences bound by the CDXA protein in the genomic fragments isolated, were determined by DNase I footprinting. From the footprinted sequences, the CDXA consensus binding site was determined. The CDXA protein binds the consensus sequence A, A/T, T, A/T, A, T, A/G. The CAUDAL binding site in the ftz promoter is also included in this consensus sequence. When tested, some of the genomic target sequences were capable of enhancing the transcriptional activity of reporter plasmids when introduced into CDXA expressing cells. This study determined the DNA sequence specificity of the CDXA protein and it also shows that this protein can further activate transcription in cells in culture.

DNA-binding specificity of the S8 homeodomain

The murine S8 homeobox gene is expressed in a mesenchyme-specific pattern in embryos, as well as in mesodermal cell lines. The S8 homeodomain is overall similar to paired type homeodomains, but at position 50, which is crucial for specific DNA recognition, it contains a Gln, as is found in Antennapedia (Antp)-type homeodomains. We determined the DNA-binding specificity of the purified S8 homeodomain by in vitro selection of random oligonucleotides. The resulting 11-bp consensus binding site, ANC/TC/TAATTAA/GC resembles, but subtly differs from, the recognition sequences of Antp-type homeodomains. Equilibrium binding constants of down to 6.0 x 10(-10) M were found for binding of the S8 homeodomain to selected oligonucleotides. Using specific antibodies and an oligonucleotide containing an S8-site, we detected by band-shift two abundant DNA binding activities in mesodermal cell lines that correspond to S8 and two more that correspond to its close relative MHox. These S8 protein forms are differentially expressed in retinoic acid-treated P19 EC cells.

A universal target sequence is bound in vitro by diverse homeodomains

To determine the number of DNA binding proteins capable of binding a consensus Engrailed binding site, this consensus sequence was used to screen a library of Drosophila cDNA clones in a bacteriophage expression vector. We retrieved clones encoding 20 distinct DNA binding domains, 17 of which are homeodomains. Binding to a variety of oligonucleotides confirms the related sequence specificity of the retrieved binding domains. Nonetheless, the homeodomains have remarkably diverse amino acid sequences. We conclude that during the evolutionary divergence of homeodomains, the specificity of DNA binding has been much more highly conserved than the amino acid sequence.

Protein product of the somatic-type transcript of the Hoxa-4 (Hox-1.4) gene binds to homeobox consensus binding sites in its promoter and intron

The murine Hoxa-4 gene encodes a protein with a homeodomain closely related to those produced by the Antennapedia-like class of Drosophila genes. Drosophila homeodomain proteins can function as transcription factors, binding to several specific DNA sequences. One sequence that is frequently encountered contains a core ATTA motif within a larger consensus sequence, such as CAATTAA. The in vitro synthesized protein product of Hoxa-4 was shown to bind to a subset of restriction fragments of the Hoxa-4 gene itself as determined by gel retardation experiments. Direct examination of the sequences of the fragments bound by Hoxa-4 protein revealed the presence of four regions containing the core ATTA motif. Two regions contained sequences of the CAATTAA class and were located approximately 1 kb upstream from the putative somatic Hoxa-4 promoter and within the intron. Two additional binding sites containing the consensus target sequence involved in autoregulation of Drosophila Deformed gene were identified: one immediately downstream of the putative embryonic transcription start site and one within the intron, respectively. Specific binding of the in vitro produced Hoxa-4 protein to oligonucleotides corresponding to these sequences was observed in gel retardation assays. The same results were obtained with Hoxa-4 protein produced in a Baculovirus expression system. Experiments using oligonucleotides containing base substitutions in positions 1, 3, 4, and 5 in the sequence CAATTAA showed severely reduced binding. The use of truncated mutant Hoxa-4 proteins in gel retardation assays and in transient co-transfection experiments revealed that the intact homeodomain was required for the binding. These results also suggested that the Hoxa-4 gene has the potential to auto-regulate its expression by interacting with the homeodomain binding sites present in the promoter as well as in the intron.

Nucleotides flanking a conserved TAAT core dictate the DNA binding specificity of three murine homeodomain proteins

Murine homeobox genes play a fundamental role in directing embryogenesis by controlling gene expression during development. The homeobox encodes a DNA binding domain (the homeodomain) which presumably mediates interactions of homeodomain proteins with specific DNA sites in the control regions of target genes. However, the bases for these selective DNA-protein interactions are not well defined. In this report, we have characterized the DNA binding specificities of three murine homeodomain proteins, Hox 7.1, Hox 1.5, and En-1. We have identified optimal DNA binding sites for each of these proteins by using a random oligonucleotide selection strategy. Comparison of the sequences of the selected binding sites predicted a common consensus site that contained the motif (C/G)TAATTG. The TAAT core was essential for DNA binding activity, and the nucleotides flanking this core directed binding specificity. Whereas variations in the nucleotides flanking the 5' side of the TAAT core produced modest alterations in binding activity for all three proteins, perturbations of the nucleotides directly 3' of the core distinguished the binding specificity of Hox 1.5 from those of Hox 7.1 and En-1. These differences in binding activity reflected differences in the dissociation rates rather than the equilibrium constants of the protein-DNA complexes. Differences in DNA binding specificities observed in vitro may contribute to selective interactions of homeodomain proteins with potential binding sites in the control regions of target genes.

Nucleotides flanking a conserved TAAT core dictate the DNA binding specificity of three murine homeodomain proteins

Murine homeobox genes play a fundamental role in directing embryogenesis by controlling gene expression during development. The homeobox encodes a DNA binding domain (the homeodomain) which presumably mediates interactions of homeodomain proteins with specific DNA sites in the control regions of target genes. However, the bases for these selective DNA-protein interactions are not well defined. In this report, we have characterized the DNA binding specificities of three murine homeodomain proteins, Hox 7.1, Hox 1.5, and En-1. We have identified optimal DNA binding sites for each of these proteins by using a random oligonucleotide selection strategy. Comparison of the sequences of the selected binding sites predicted a common consensus site that contained the motif (C/G)TAATTG. The TAAT core was essential for DNA binding activity, and the nucleotides flanking this core directed binding specificity. Whereas variations in the nucleotides flanking the 5' side of the TAAT core produced modest alterations in binding activity for all three proteins, perturbations of the nucleotides directly 3' of the core distinguished the binding specificity of Hox 1.5 from those of Hox 7.1 and En-1. These differences in binding activity reflected differences in the dissociation rates rather than the equilibrium constants of the protein-DNA complexes. Differences in DNA binding specificities observed in vitro may contribute to selective interactions of homeodomain proteins with potential binding sites in the control regions of target genes.

Multitude of inverted repeats characterizes a class of anchorage sites of chromatin loops to the nuclear matrix

In order to understand the nature of DNA sequences that organize chromatin into domains or loops, we have cloned the nuclear matrix DNA (1.7% of the total DNA) from human myelogenous leukemia cells in culture. Nuclear matrix is formed by interactions between specific stretches of DNA of about 0.1 to 5.0 kb with protein transcription factors, nuclear enzymes, and structural proteins. Nuclear matrix is believed to be the exclusive nuclear microenvironment in which initiation of DNA replication, transcription, and repair take place. The matrix attachment regions (MARs) of DNA have transcriptional enhancer activity, harbor the origins of replication of the human genome, and define the borders between neighboring chromatin loops. In this study we report the sequence of the human MAR fragment 19.2 of a size of 542 bp. Hum. MAR 19.2 is composed of TG-, CA-, CT-, and GA-rich blocks and shows 8 perfect and imperfect inverted repeats. Thus, we have identified a novel class of MARs with sequence characteristics divergent from the AT-rich class of MARs. The inverted repeats of the 19.2 sequence might be stabilized into their cruciform configuration by torsional strain and by specific transcription/replication protein factors. This MAR might function in the initiation of replication of the flanking chromatin domain and in the regulation of the transcriptional activity of the gene(s) that reside in this domain.

Functional specificity of the homeodomain protein fushi tarazu: the role of DNA-binding specificity in vivo

The mechanisms determining the functional specificity of Drosophila homeodomain proteins are largely unknown. Here, the role of DNA-binding specificity for the in vivo function of the homeodomain protein fushi tarazu (ftz) is analyzed. We find that specific DNA binding is an important but not sufficient determinant of the functional specificity of ftz in vivo: The ftz DNA-binding specificity mutant ftzQ50K retains partial ftz wild-type activity in gene activation and phenotypic rescue assays. Furthermore, specificity mutations in a ftz-in vivo binding site only partially reduce enhancer activity as compared to null mutations of this site. Despite bicoid-like DNA-binding specificity ftzQ50K does not activate natural or artificial bcd target genes in the realms of ftz. These results are discussed in the light of recent observations on the mechanism of action of the yeast homeodomain protein alpha 2.

Molecular genetics of aristaless, a prd-type homeo box gene involved in the morphogenesis of proximal and distal pattern elements in a subset of appendages in Drosophila

Viable aristaless (al) mutations of Drosophila affect pattern elements at both ends of the proximodistal axis in a subset of adult appendages. The al gene has been cloned and identified by P-element-mediated germ-line transformation with a genomic DNA fragment, which rescues a lethal mutation of al as well as aspects of the adult al phenotype. The al gene contains a prd-type homeo domain and a Pro/Gln-rich domain and, hence, probably encodes a transcription factor. Its transcript distribution in third-instar imaginal discs closely corresponds to the anlagen of the tissues that later become visibly affected in adult al mutants. The striking similarity of a bimodal al expression in different imaginal discs indicates that al is under the control of a "prepattern," which is shared at least among antennal, leg, and wing discs. The al gene is also transcribed during embryogenesis. Apart from a function in the ontogeny of specific larval head and tail organs, its embryonic transcript pattern suggests a possible role in early imaginal disc development.

A human putative lymphocyte G0/G1 switch gene homologous to a rodent gene encoding a zinc-binding potential transcription factor

G0S24 is a member of a set of genes (putative G0/G1 switch regulatory genes) that are expressed transiently within 1-2 hr of the addition of lectin or cycloheximide to human blood mononuclear cells. Comparison of a full-length cDNA sequence with the corresponding genomic sequence reveals an open reading frame of 326 amino acids, distributed across two exons. Potential phosphorylation sites include the sequence PSPTSPT, which resembles an RNA polymerase II repeat reported to be a target of the cell cycle control kinase cdc2. Comparison of the derived protein sequence with those of rodent homologs allows classification into three groups. Group 1 contains G0S24 and the rat and mouse TIS11 genes (also known as TTP, Nup475, and Zfp36). Members of this group have three tetraproline repeats. Groups 1 and 2 have a serine-rich region and an "arginine element" (RRLPIF) at the carboxyl terminus. All groups contain cysteine- and histidine-rich putative zinc finger domains and a serine-phenylalanine "SFS" domain similar to part of the large subunit of eukaryotic RNA polymerase II. Comparison of group 1 human and mouse genomic sequences shows high conservation in the 5' flank and exons. A CpG island suggests expression in the germ line. G0S24 has potential sites for transcription factors in the 5' flank and intron; these include a serum response element. Protein and genomic sequences show similarities with those of a variety of proteins involved in transcription, suggesting that the G0S24 product has a similar role.

A novel murine homeobox gene isolated by a tissue specific PCR cloning strategy

We have identified a novel homeobox gene, designated K-2, using a reverse transcription PCR cloning strategy. Sequence analysis reveals that the homeobox of K-2 is 77.6% homologous at the nucleotide level and 97% identical at the amino acid sequence level to another murine gene, S8. Homeodomain sequence comparisons indicate that K-2 and S8 represent a distinct subclass of paired type homeobox genes. Northern blot analysis of RNA from murine embryos and adult tissues identified multiple transcripts that are expressed in a developmentally specific and tissue restricted manner. Alternate splicing of K-2 at the 3-coding region leads to the inclusion of a chain terminating sequence. In addition, the developmental expression pattern of this gene at day 12 of gestation was determined by in situ hybridization. Expression was observed in diverse mesenchymal cells in craniofacial, pericardial, primitive dermal, prevertebral, and genital structures.

Homeotic protein binding sites, origins of replication, and nuclear matrix anchorage sites share the ATTA and ATTTA motifs

Nuclear matrix organizes the mammalian chromatin into loops. This is achieved by binding of nuclear matrix proteins to characteristic DNA landmarks in introns as well as proximal and distal sites flanking the 5' and 3' ends of genes. Matrix anchorage sites (MARs), origins of replication (ORIs), and homeotic protein binding sites share common DNA sequence motifs. In particular, the ATTA and ATTTA motifs, which constitute the core elements recognized by the homeobox domain from species as divergent as flies and humans, are frequently occurring in the matrix attachment sites of several genes. The human apolipoprotein B 3' MAR and a stretch of the Chinese hamster DHFR gene intron and human HPRT gene intron shown to anchor these genes to the nuclear matrix are mosaics of ATTA and ATTTA motifs. Several origins of replication also share these elements. This observation suggests that homeotic proteins which control the expression level of many genes and pattern formation during development are components of the nuclear matrix. Thus, the nuclear matrix, known as the site of DNA replication, might sculpture the crossroads of the differential activation of origins during development and S-phase and the control of gene expression and pattern formation in embryogenesis.