Sequence-specific DNA-binding proteins which interact with (G + C)-rich sequences flanking the chicken c-myc gene

The interaction of nuclear sequence-specific DNA-binding proteins from definitive chicken erythrocytes, thymus and proliferating transformed erythroid precursor (HD3) cells with the 700-base-pair (700-bp) DNA 5'-flanking region of the chicken c-myc gene was investigated by in vitro footprint analysis. The major HD3 protein-binding activity binds to a site (site V) 200 bp upstream from the 'cap' site but, after further fractionation, a second distinct binding activity is detected to a site (site VIII) which contains both the 'CAAT' and 'SP1-binding' consensus sequences. Protein from thymus and erythrocyte cells which express c-myc at lower levels, bind to seven and eight sites respectively. In common with HD3 cell protein, they both bind to site VIII and, although binding to the sequence at site V is also detected, the footprint protection pattern is sufficiently different (site V') to suggest the involvement of different proteins in terminally differentiated and proliferating cells. The DNA-binding activities were partially fractionated by high-performance liquid chromatography gel filtration and include an erythrocyte-specific protein which binds to a c-myc gene poly(dG) homopolymer sequence similar to that found upstream of the chicken beta A-globin gene.     

Multiple sequence-specific DNA binding activities are eluted from chicken nuclei at low ionic strengths.

DNA sequence-specific binding proteins eluted from chicken erythrocyte and thymus nuclei, and fractionated as described by Emerson and Felsenfeld (19), have been investigated by filter binding and footprint analyses. The erythrocyte nuclear protein fraction specifically binds to at least two sites within the 5' flanking chromatin hypersensitive site of the chicken beta A-globin gene, and to a site 5' to the human beta-globin gene. The major chicken beta A globin gene binding site [G)18CGGGTGG) and the human beta-globin gene binding site [TA)6(T)8C(T)4) occur at or near sequences which are hypersensitive to S1 nuclease cleavage in supercoiled plasmids. Downstream, the second chicken beta A-globin gene binding site includes the beta-globin gene CACCC consensus sequence. Filter binding studies also show other sequence specific binding activities to human N-ras and human (but not chicken) c-myc gene sequences.     

Purification of an oligo(dG).oligo(dC)-binding sea urchin nuclear protein, suGF1: a family of G-string factors involved in gene regulation during development.

Contiguous deoxyguanosine residues (G strings) have been implicated in regulation of gene expression in several organisms via the binding of G-string factors. Regulation of expression of the chicken adult beta-globin gene may involve the interplay between binding of an erythrocyte-specific G-string factor, BGP1, and the stability of a positioned nucleosome (C. D. Lewis, S. P. Clark, G. Felsenfeld, and H. Gould, Genes Dev. 2:863-873, 1988). We have purified a 59.5-kDa nuclear protein (suGF1) from sea urchin embryos by DNA affinity chromatography. suGF1 has high binding affinity and specificity for oligo(dG).oligo(dC). The identity of the purified protein was confirmed by renaturation of sequence-specific DNA-binding activity from a sodium dodecyl sulfate-polyacrylamide gel slice and by Southwestern (DNA-protein) blotting. suGF1 binds in vitro to a G11 string present in the H1-H4 intergenic region of a sea urchin early histone gene battery. This suGF1 DNA recognition site occurs within a homopurine-homopyrimidine stretch previously shown to be incorporated into a positioned nucleosome core in vitro. DNase I footprinting shows that suGF1 protects the same base pairs on the promoter of the chicken beta A-globin gene as does BGP1. We show that a G-string cis-regulatory element of a sea urchin cell lineage-specific gene LpS1 (M. Xiang, S.-Y. Lu, M. Musso, G. Karsenty, and W. H. Klein, Development 113:1345-1355, 1991) also represents a high-affinity recognition site for suGF1. suGF1 may be a member of a family of G-string factors involved in the regulation of expression of unrelated genes during development of a number of different organisms.     

Characterisation of chicken erythroid nuclear proteins which bind to the nuclease hypersensitive regions upstream of the beta A- and beta H-globin genes.

Chicken erythrocyte sequence-specific nuclear DNA-binding proteins, which bind to the 5'-flanking DNAseI hypersensitive sites of the erythrocyte chromosomal beta A- and beta H-globin genes, have been fractionated by HPLC gel filtration. Three beta A-globin gene DNA binding activities (to sites A, B and B' (10-12)) were separated. The erythroid precursor cell line HD3 has beta A-globin gene sites B and B' binding activities, but binding to site A is detected only after the HD3 cells are induced to differentiate. The fractionated protein binds to a redefined site B', which contains at its center the globin CACCC consensus sequence. The chromosomal beta H-globin gene has two 5'-flanking DNAseI hypersensitive sites which bracket two sequences (H and H') bound by erythrocyte and HD3 nuclear protein in vitro. The beta H- and beta A-globin gene binding sites (H and B) contain variants of the sequences bound by Nuclear Factor 1 and the TGGCA-binding protein, and their protein binding activity(ies) co-purify after HPLC gel filtration.     

TGGCA protein is present in erythroid nuclei and binds within the nuclease-hypersensitive sites 5' of the chicken beta H- and beta A-globin genes

The developmentally regulated 5'-flanking DNase-I-hypersensitive site of the chicken beta H-globin gene in nuclei contains a subregion which is resistant to DNase I and which disappears when nuclei are extracted with 0.3 M NaCl, suggesting that there are salt-extractable proteins bound to sequences within this region. The 0.3 M NaCl extract contains two proteins which bind in vitro to these sequences. One of the binding sequences has an inverted repeat very similar to that bound by TGGCA protein. Partially purified TGGCA protein from chicken liver binds to this sequence in vitro giving exactly the same footprint as that obtained with erythroid nuclear proteins. Similarly TGGCA protein binds to an inverted repeat with the beta A-globin 5'-hypersensitive site giving a footprint identical to that obtained with erythroid nuclear protein extracts. From competition footprinting experiments and the electrophoretic mobility of the protein-DNA complex, it is concluded that the erythroid proteins previously described as binding to the beta H- and beta A-globin inverted repeats within the 5'-flanking hypersensitive sites both belong to the TGGCA protein family.     

The purification of an erythroid protein which binds to enhancer and promoter elements of haemoglobin genes.

An erythroid nuclear protein (EF1), originally detected as a protein binding within the nuclease hypersensitive site upstream of the chicken beta H-globin gene, has been purified. This protein of 37,000-39,000 molecular weight binds to three sites within the hypersensitive region: one between the CCAAT and TATA boxes, the second (further upstream) next to a NF1 binding site, and the third adjacent to a regulatory element found in a number of beta-globin genes. The EF1 protein also binds to an erythroid-specific promoter element of the mouse alpha-globin gene and to two sites within the chicken beta A-globin enhancer. These six EF1-binding sites are related by the consensus sequence A/TGATAA/GG/C. A minor protein of molecular weight 72,000 which co-purifies with EF1 also binds to the same sequences.     

Cationic metal-specific structures adopted by the poly(dG) region and the direct repeats in the chicken adult beta A globin gene promoter.

Naturally occurring contiguous deoxyguanine residues and their surrounding sequences in the chicken adult beta A globin gene promoter were analyzed for their inherent potential to adopt non-B DNA structures in supercoiled plasmid DNA. In particular, cationic effects on structure were studied by treating the supercoiled plasmid DNA harboring the chicken adult beta A globin 5' flanking sequence with an unpaired DNA base-specific probe, chloroacetaldehyde in the presence of either Mg++, Cu++, Zn++, Ca++ or Co++ ions. The chloroacetaldehyde-reactive bases were mapped at a single base resolution by a chemical cleavage method that specifically cleaves DNA at the chloroacetaldehyde modified sites. These experiments revealed that while Mg++ and Ca++ ions induce a dG.dG.dC triple helix structure at the contiguous dG residues, Zn++, Cu++ and Co++ ions induce yet another structure at the direct repeats immediately 5' of the dG residues. When Mg++ and Zn++ ions are both present, Zn++ inhibits the dG.dG.dC triplex at the contiguous dG residues and induces a particular non-B DNA structure at the adjacent direct repeats. The specific induction of non-B DNA structures by metal ions at the two adjacent sequences within the promoter region may be of biological significance.     

Different features of the MHC class I heterodimer have evolved at different rates. Chicken B-F and beta 2-microglobulin sequences reveal invariant surface residues.

Chicken beta 2-microglobulin (beta 2m) and class I (B-F19 alpha chain) cDNA clones were isolated and the sequences compared to those of B-F Ag isolated from chicken E. These clones represent the major expressed class I molecules on E, with B-F alpha size variants evidently due to alternative use of small exons in the cytoplasmic region. The cDNA sequences were compared to turkey beta 2m, the apparent allele B-F12 alpha and other vertebrate homologs, using the 2.6 A structure of the human HLA-A2 molecule as a model. Both chicken alpha 1 and alpha 2 domains resemble mammalian classical class I molecules and the MHC-encoded nonclassical molecules more than CD1 or the class I-like FcR. In contrast, the chicken alpha 3 domain is equally homologous to all alpha 3 domains, to beta 2m and to class II beta 2 domains. For each pair of extracellular domains (alpha 1 vs alpha 2, alpha 3 vs beta 2m), the level of sequence homology between mammalian and avian molecules is quite different. This suggests that the structurally homologous domains have been under different selective pressures during evolution. There is a very strong G + C bias in alpha 3 and beta 2m, leading to an overall change in amino acid composition in B-F compared to class I molecules from other taxa. Many of the surface residues are quite diverged, particularly in alpha 3 and beta 2m. There are fewer changes in intra- and interdomain contact sites. Some residues with important functions are invariant, including seven residues that bind the ends of the peptide, two residues that bind CD8, and three residues that are phosphorylated. The positions of the allelic residues are conserved. There are other patches of invariant residues on alpha 1, alpha 2, and beta 2m; these might bind TCR or other molecules involved in class I function.     

Occurrence of DNA structures which differ from the canonic B-form in sites of highly specific interaction with chromatin proteins

According to the three-dimensional structure of DNase I and the mechanism of its action on linear double-stranded DNA, helix regions in conformations considerably different from the canonical B-form should be resistant to endonucleolysis. A number of DNA sequences specifically bound by nonhistone factors within 5'-flanking regions of the chicken beta A-globin, beta H-globin and c-myc genes are shown to contain short DNase I-resistant DNA domains. Several examples of the occurrence of such DNase I-resistant domains within the sites for high-specific recognition by different proteins are given. The role of the DNA structural polymorphism in site-specific interaction with protein factors is discussed.     

A short sequence within domain C of duck carboxypeptidase D is critical for duck hepatitis B virus binding and determines host specificity

Virus-cell surface receptor interactions are of major interest. Hepadnaviruses are a family of partially double-stranded DNA viruses with liver tropism and a narrow host range of susceptibility to infection. At least in the case of duck hepatitis B virus (DHBV), host specificity seems controlled partly at the receptor level. The middle portion in the pre-S region of the viral large envelope protein binds specifically to duck carboxypeptidase D (DCPD) but not to its human or chicken homologue. Although domain C of DCPD is implicated in ligand binding, the exact pre-S contact site remains to be determined. We prepared and tested a panel of chimeric constructs consisting of DCPD and human carboxypeptidase D (HCPD). Our results indicate that a short region at the N terminus of domain C (residues 920 to 949) is critical to DHBV binding and is a major determinant for the host specificity of DHBV infection. Replacing this region of the DCPD molecule with its human homologue abolished the DHBV interaction, whereas introducing this DCPD sequence into HCPD conferred efficient DHBV binding. Extensive analysis of site-directed mutants revealed that both conserved and nonconserved residues were important for the pre-S interaction. There were primary sequence variations and secondary structural differences that contributed to the inability of HCPD to bind the DHBV pre-S domain.     

The telobox, a Myb-related telomeric DNA binding motif found in proteins from yeast, plants and human.

The yeast TTAGGG binding factor 1 (Tbf1) was identified and cloned through its ability to interact with vertebrate telomeric repeats in vitro. We show here that a sequence of 60 amino acids located in its C-terminus is critical for DNA binding. This sequence exhibits homologies with Myb repeats and is conserved among five proteins from plants, two of which are known to bind telomeric-related sequences, and two proteins from human, including the telomeric repeat binding factor (TRF) and the predicted C-terminal polypeptide, called orf2, from a yet unknown protein. We demonstrate that the 111 C-terminal residues of TRF and the 64 orf2 residues are able to bind the human telomeric repeats specifically. We propose to call the particular Myb-related motif found in these proteins the 'telobox'. Antibodies directed against the Tbf1 telobox detect two proteins in nuclear and mitotic chromosome extracts from human cell lines. Moreover, both proteins bind specifically to telomeric repeats in vitro. TRF is likely to correspond to one of them. Based on their high affinity for the telomeric repeat, we predict that TRF and orf2 play an important role at human telomeres.     

Fibroblast growth factor (FGF) homologous factors share structural but not functional homology with FGFs

Fibroblast growth factors (FGFs) interact with heparan sulfate glycosaminoglycans and the extracellular domains of FGF cell surface receptors (FGFRs) to trigger receptor activation and biological responses. FGF homologous factors (FHF1-FHF4; also known as FGF11-FGF14) are related to FGFs by substantial sequence homology, yet their only documented interactions are with an intracellular kinase scaffold protein, islet brain-2 (IB2) and with voltage-gated sodium channels. In this report, we show that recombinant FHFs can bind heparin with high affinity like classical FGFs yet fail to activate any of the seven principal FGFRs. Instead, we demonstrate that FHFs bind IB2 directly, furthering the contention that FHFs and FGFs elicit their biological effects by binding to different protein partners. To understand the molecular basis for this differential target binding specificity, we elucidated the crystal structure of FHF1b to 1.7-A resolution. The FHF1b core domain assumes a beta-trefoil fold consisting of 12 antiparallel beta strands (beta 1 through beta 12). The FHF1b beta-trefoil core is remarkably similar to that of classical FGFs and exhibits an FGF-characteristic heparin-binding surface as attested to by the number of bound sulfate ions. Using molecular modeling and structure-based mutational analysis, we identified two surface residues, Arg52 in the beta 4-beta 5 loop and Val95 in the beta 9 strand of FHF1b that are required for the interaction of FHF1b with IB2. These two residues are unique to FHFs, and mutations of the corresponding residues of FGF1 to Arg and Val diminish the capacity of FGF1 to activate FGFRs, suggesting that these two FHF residues contribute to the inability of FHFs to activate FGFRs. Hence, FHFs and FGFs bear striking structural similarity but have diverged to direct related surfaces toward interaction with distinct protein targets.