Analysis of dimerization and DNA binding functions in Fos and Jun by domain-swapping: involvement of residues outside the leucine zipper/basic region

The products of two cellular proto-oncogenes c-fos and c-jun form a heterodimeric complex that contribute to the DNA-binding activity referred to as AP-1 (activator protein-1). Two domains have been proposed to be required for heterodimer formation and protein-DNA complex formation. The leucine zipper domain mediated the interaction between the two proteins and a highly basic region immediately N-terminal to the leucine zipper forms a DNA binding domain. To assess the role of these two domains in dimerization and DNA binding and to determine what contribution, if any, is made by residues outside of these regions, we carried out an extensive domain swap analysis. Restriction sites created in the fos and jun cDNAs flanking the basic region and leucine zipper allowed these domains to be swapped between the two proteins either singly or in various combinations with adjacent domains. The chimeric proteins were assayed for their ability to dimerize with each other and to interact with the AP-1 consensus sequence. It was found that two Jun leucine zipper regions could mediate protein dimerization, whereas two Fos leucine zipper regions could not. The dimers formed between two Jun leucine repeats were less stable than those formed between a Fos and a Jun leucine zipper. A conserved His residue seven amino acids C-terminal of the last leucine of the zipper region contributed to the stability of protein-protein interactions. The basic region of both Fos and Jun was found to interact with DNA without the presence of the other, i.e. the combination of two Fos- or two Jun-DNA binding domains could bind to the AP-1 site. However, replacement of the Jun N-terminus with that of Fos resulted in a decrease in DNA binding, indicating that residues outside of the Jun basic region contribute to DNA binding. The results also suggest that the dimerization and DNA binding functions of each protein are not completely independent properties, but that each exerts an influence on the other.     

Both Jun and Fos contribute to transcription activation by the heterodimer

Comparison of the amino acid sequence of the three members of the mouse jun proto-oncogene family, c-jun, jun B and jun D, reveals several homologous segments. The most C-terminal of them including a leucine zipper motif and a cluster of basic amino acids was previously identified as the DNA binding domain. By deletion analysis, we show that three conserved domains in the N-terminal region are crucial for transactivation by Jun homodimers. Only one of these is predicted to form an acidic amphipathic alpha-helix. The addition of Fos and the formation of Jun-Fos heterodimers strongly increases the transactivation level. Jun mutants that are inactive alone gain partial or full activity in the presence of Fos. This increase strongly depends on the presence of the C-terminal domain of Fos. These results show that in Jun-Fos heterodimers both the N-terminal part of Jun and the C-terminal part of Fos contribute to transactivation with a more pronounced role for the latter.     

Organization of mouse DNA polymerase beta gene silencer elements and identification of the silencer-binding factor(s)

Different portions of the 5'-upstream region of the mouse DNA polymerase beta gene were combined with bacterial chloramphenicol acetyltransferase (CAT) gene of the CAT vector. Transfection of these recombinant plasmids into mouse NIH/3T3 cells has revealed that each of the previously identified two negatively acting regions (silencers I and II) of this gene consists of multiple sub-domains. The distal silencer (silencer I) at around -1.5 kb consists of four sub-domains (-1852 to -1667, -1663 to -1616, -1564 to -1525 and -1355 to -1257). The promoter-proximal silencer (silencer II) at around -0.5 kb consists of two functional domains (-681 to -523 and -490 to -447) separated by a neutral region of 33 base pairs. Silencer II functioned efficiently when silencer I was deleted. Conversely, the distal silencer I functioned efficiently when silencer II was deleted. Thus, these silencers functioned redundantly to each other in NIH/3T3 cells. Nucleotide sequence analysis revealed no extensive sequence similarity between these two silencers. Significant sequence similarity is present between a distal portion of silencer II and the c-myc gene silencer, and also between a proximal portion of silencer II and the mouse F9 cell-specific silencer. A protein factor(s) that specifically bound to the silencer elements was detected in nuclear extracts of NIH/3T3 cells and mouse liver in which DNA polymerase beta was expressed at a rather low level. The same binding factor(s) can bind to both silencer I and II regions, although its affinity for silencer II is much higher than that for silencer I.     

Organization of Mouse DNA Polymerase β Gene Silencer Elements and Identification of the Silencer-binding Factor(s)

Different portions of the 5-upstream region of the mouse DNA polymerase β gene were combined with bacterial chloramphenicol acetyltransferase (CAT) gene of the CAT vector. Transfection of these recombinant plasmids into mouse NIH/3T3 cells has revealed that each of the previously identified two negatively acting regions (silencers I and II) of this gene consists of multiple sub-domains. The distal silencer (silencer I) at around -1.5 kb consists of four sub-domains (-1852 to -1667, -1663 to -1616, -1564 to -1525 and -1355 to -1257). The promoter-proximal silencer (silencer II) at around – 0.5 kb consists of two functional domains (- 681 to – 523 and – 490 to – 447) separated by a neutral region of 33 base pairs. Silencer II functioned efficiently when silencer I was deleted. Conversely, the distal silencer I functioned efficiently when silencer II was deleted. Thus, these silencers functioned redundantly to each other in NIH/3T3 cells. Nucleotide sequence analysis revealed no extensive sequence similarity between these two silencers. Significant sequence similarity is present between a distal portion of silencer II and the c-myc gene silencer, and also between a proximal portion of silencer II and the mouse F9 cell-specific silencer. A protein factor(s) that specifically bound to the silencer elements was detected in nuclear extracts of NIH/3T3 cells and mouse liver in which DNA polymerase β was expressed at a rather low level. The same binding factor(s) can bind to both silencer I and II regions, although its affinity for silencer II is much higher than that for silencer I.     

In vitro DNA binding activity of Fos/Jun and BZLF1 but not C/EBP is affected by redox changes

The leucine zipper family of proteins have a DNA binding domain composed of a leucine zipper dimerisation interface and a basic DNA binding structure. We show here that redox changes affect the in vitro DNA binding ability of a select subset of leucine zipper proteins. The bacterially expressed DNA binding domains of Fos/Jun and BZLF1 are unable to bind DNA under non-reducing conditions whereas binding of the C/EBP DNA binding domain is unaffected. Sensitivity to redox state is due to the presence of a conserved cysteine residue in the basic DNA binding motif of Fos, Jun and BZLF1 but not C/EBP. Under non-reducing conditions an intermolecular disulphide bridge is formed between the cysteine residues of each basic motif within a dimer, which prevents DNA binding. We show that oxidation of these C residues can be achieved enzymatically, using glutathione peroxidase, and that DNA binding protects them from oxidation. These data raise the possibility that intracellular changes in the redox state may differentially regulate the activity of leucine zipper family members. In addition the loss of DNA binding activity under non-reducing conditions has implications for the purification methods used to isolate proteins of the leucine zipper family for structural analysis.     

Evolution of alpha-fetoprotein: sequence comparisons among AFP species and with albumin species

Alpha-fetoprotein (AFP) and albumin are related proteins, which contain about 585 amino acid residues that are organized in a characteristic structure of 3 homologous domains of about 195 amino acid residues. We have compared the domains of mouse, rat, and human AFP and rat, bovine, and human albumin using computer programs designed to quantify relationships between proteins. The comparisons of corresponding domains of the AFPs (e.g. domain I of rat and human AFP) reveal that each domain is well conserved. Similar results were found for the comparisons of corresponding domains in the albumins. In contrast, there was much less similarity between corresponding domains of albumin and AFP. These comparisons between AFP and albumin revealed that: (1) the amino acid sequences in domain III, which is at their carboxy terminus, are most conserved, and (2) the amino acid sequences in domain I, which is at their amino terminus, are least conserved. This suggests that there are differences in the constraints on amino acid substitutions among the domains of AFP and albumin during the approximately 400 million years since they diverged from a common ancestor. Also, computer studies revealed that there are substantial differences between domains I, II, and III in each AFP and albumin species, which indicates that these domains are significantly different from their approximately 195 residue ancestral domain. Finally, we find that mouse and rat AFP are more dissimilar to the albumins than is human AFP. Overall, our computer analyses indicate that AFP and albumin can be considered to be composed of distinct, but related, approximately 195 residue proteins, each of which could differ in some of their properties; for example, the binding of fatty acids.     

Comparative genomics on Fzd8 orthologs

WNT signaling pathway networks with Hedgehog, Notch and FGF signaling pathways during carcinogenesis and embryogenesis. FZD8 is up-regulated in HeLa S3 and A549 cells. Here we identified and characterized rat Fzd8 gene by using bioinformatics. Rat Fzd8 gene was identified within AC131883.2 genome sequence. Rat Fzd8 (684 aa) showed 99.1, 96.8, 71.8 and 71.6% total amino-acid identity with mouse Fzd8, human FZD8, zebrafish fzd8 and Xenopus fzd8, respectively. Vertebrate Fzd8 orthologs were seven-transmembrane receptors with Frizzled (Fz) domain within the N-terminal extracellular region, leucine zipper motif around the fifth transmembrane domain, and Dishevelled (Dvl)-binding motif within the C-terminal cytoplasmic region. Two Asn-linked glycosylation sites within the N-terminal extracellular region were conserved among vertebrate Fzd8 orhologs. One Asn-linked glycosylation site within the second extracellular loop was conserved among mammalian Fzd8 orthologs, but not in Xenopus and zebrafish fzd8 orthologs. These facts indicate the molecular evolution of Fzd8 orthologs. The 5'-flanking region and exonic region were well conserved among mammalian Fzd8 orthologs. Nucleotide position 92950-94221 of AC131883.2 genome sequence was identified as the evolutionarily conserved promoter region of rat Fzd8 gene, and nucleotide position 133421-132134 of AL121749.14 genome sequence as the evolutionarily conserved promoter region of human FZD8 gene. Match program revealed that ELK1- and PAX4-binding sites were conserved between rat Fzd8 and human FZD8 promoters. This is the first report on the rat Fzd8 gene as well as on comparative genomics for Fzd8 orthologs.     

Notch1 is a frequent mutational target in chemically induced lymphoma in mouse

Activating Notch1 mutations have been reported in human T-lineage acute lymphoblastic leukemia (T-ALL) and lymphomas from genetically modified mice. We report that Notch1 is a prevalent and major mutational target in chemically induced mouse lymphoma. The regions of the gene that are frequently mutated are the heterodimerization domain and the N-terminal ligand-binding region, important for protein stability, and the polypeptide rich in proline, glutamate, serine and threonine (PEST) domains, which is critical for protein degradation. Another gene, CDC4, is also involved in Notch1 degradation and shows frequent mutations. Mutations in the heterodimerization and the ligand-binding regions may cause ligand-independent signaling, whereas mutations preventing protein degradation result in accumulation of intracellular Notch1. We analyzed 103 chemical-induced mouse lymphomas for mutations in the Notch1 gene using single strand conformation analysis (SSCA) and DNA sequencing. Genetic alterations resulting in premature truncation of Notch1 were identified in 28 tumors, whereas 8 revealed alterations in the heterodimerization and 16 harbored deletions in the ligand-binding region. Dideoxycytidine-induced lymphomas displayed the highest frequency of Notch1 mutations (49%), whereas in butadiene- and phenolphthalein-induced tumors showed lower frequencies (26 and 10%, respectively). In total, 26 novel and 3 previously reported mutations were detected. This report shows that Notch1 is a prevalent and major mutational target for 2',3'-dideoxycytidine and butadiene-induced lymphoma.     

Molecular characterization of P2B/LAMP-1, a major protein target of a metastasis-associated oligosaccharide structure

Sialylated and GlcNAc beta 1-6Man alpha 1-6Man beta 1 (beta 1-6 branched) complex-type oligosaccharides linked to asparagine residues of membrane glycoproteins in metastatic murine tumor cells have been associated with efficient tumor cell metastasis. A large proportion of these oligosaccharide structures, in several unrelated malignant cell lines, have been shown to be associated with a glycoprotein termed P2B, with a molecular weight of 130,000. This glycoprotein has recently been purified from the metastatic MDAY-D2 cell line and shown to be biochemically similar to a lysosomal associated membrane glycoprotein (LAMP-1). We report here the details of a 2147 nucleotide complementary DNA encoding the entire murine P2B polypeptide which was immunoselected from a lambda gt11 expression library and sequenced. The sequence is similar to a complementary DNA coding for mouse LAMP-1 with the exception of a 5' untranslated region, a leader signal-sequence, and various insertions, deletions, and substitutions in the 3' untranslated domain. An open reading frame of 405 amino acids encodes a mature polypeptide of 382 residues with a predicted molecular weight of 42,000. P2B/LAMP-1 possesses 20 asparagine-linked glycosylation sites separated into equal halves by a central, putative hinge region and is anchored by a carboxy, membrane-spanning, domain. Topological considerations dictate that cell surface expression of P2B/LAMP-1 exposes the bulk of the glycoprotein into the extracellular compartment. Immunofluorescent staining of fibroblast cells indicated that P2B/LAMP-1 was associated with lysosomal membranes and, to a lesser degree, select surfaces of plasma membrane. An amino acid comparison of the murine sequence with its recently cloned rat, human, and chicken counterparts shows a conservation of 17 of 20 asparagine-linked glycosylation consensus sites, eight of eight cysteine residues, and other selected protein domains. The interspecies conservation of these domains suggests that they are important for the structure and function of the P2B/LAMP-1 glycoprotein. Northern analysis revealed that P2B/LAMP-1 is widely expressed in normal murine tissues and tumor cell lines. However, in two experimental models of metastasis, where changes in branching of oligosaccharides on P2B/LAMP-1 have been shown to occur, comparable levels of P2B/LAMP-1 mRNA were found in both metastatic and nonmetastatic cell lines.