FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2079500084> ?p ?o ?g. }

• W2079500084 endingPage "16497" @default.
• W2079500084 startingPage "16490" @default.
• W2079500084 abstract "A 1.6-kilobase pair full-length cDNA encoding a transcription factor homologous to the Maf family of proteins was isolated by screening a K562 cDNA library with the NFE2 tandem repeat probe derived from the globin locus control region. The protein, which was designated hMAF, contains a basic DNA binding domain and an extended leucine zipper but lacks any recognizable activation domain. Expressed in vitro, the hMAF protein is able to homodimerize in solution and band-shift the NFE2 tandem repeat probe. In addition to homodimers, hMAF can also form high affinity heterodimers with two members of the NFE2/CNC-bZip family (Nrf1 and Nrf2) but not with a third family member, p45-NFE2. Although hMAF/hMAF homodimers and hMAF/Nrf1 and hMAF/Nrf2 heterodimers bind to the same NFE2 site, they exert functionally opposite effects on the activity of a linked γ-globin gene. In fact, whereas all hMAF/CNC-bZip heterodimers stimulate the activity of a γ-promoter reporter construct in K562 cells, the association into homodimers that is induced by overexpressing hMAF inhibits the activity of the same construct. Thus variations in the expression of hMAF may account for the modulation in the activity of the genes that bear NFE2 recognition sites. A 1.6-kilobase pair full-length cDNA encoding a transcription factor homologous to the Maf family of proteins was isolated by screening a K562 cDNA library with the NFE2 tandem repeat probe derived from the globin locus control region. The protein, which was designated hMAF, contains a basic DNA binding domain and an extended leucine zipper but lacks any recognizable activation domain. Expressed in vitro, the hMAF protein is able to homodimerize in solution and band-shift the NFE2 tandem repeat probe. In addition to homodimers, hMAF can also form high affinity heterodimers with two members of the NFE2/CNC-bZip family (Nrf1 and Nrf2) but not with a third family member, p45-NFE2. Although hMAF/hMAF homodimers and hMAF/Nrf1 and hMAF/Nrf2 heterodimers bind to the same NFE2 site, they exert functionally opposite effects on the activity of a linked γ-globin gene. In fact, whereas all hMAF/CNC-bZip heterodimers stimulate the activity of a γ-promoter reporter construct in K562 cells, the association into homodimers that is induced by overexpressing hMAF inhibits the activity of the same construct. Thus variations in the expression of hMAF may account for the modulation in the activity of the genes that bear NFE2 recognition sites. In the last decade, meticulous searches along the β-globin gene cluster have led to the identification of numerous regulatory DNA sequences located either in close proximity to the genes or at a distance in regions that were originally identified for their DNase I hypersensitivity (1Tuan D. Solomon W. Li Q. London I.M. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 6384-6388Crossref PubMed Scopus (434) Google Scholar, 2Forrester W.C. Takegawa S. Papayannopoulou T. Stamatoyannopoulos G. Groudine M. Nucleic Acids Res. 1987; 15: 10159-10177Crossref PubMed Scopus (291) Google Scholar, 3Forrester W.C. Thompson C. Elder J.T. Groudine M. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1359-1363Crossref PubMed Scopus (269) Google Scholar). The latter regions, which are referred to individually as hypersensitive sites (from 5′-HS1 to HS4) and collectively as the locus control region of the β-globin gene cluster (4Orkin S.H. Cell. 1990; 63: 665-672Abstract Full Text PDF PubMed Scopus (320) Google Scholar), contain elements with different functions such as enhancers (5Moi P. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 9000-9004Crossref PubMed Scopus (77) Google Scholar, 6Ney P.A. Sorrentino B.P. McDonagh K.T. Nienhuis A.W. Genes Dev. 1990; 4: 993-1006Crossref PubMed Scopus (209) Google Scholar, 7Sorrentino B. Ney P. Bodine D. Nienhuis A.W. Nucleic Acids Res. 1990; 18: 2721-2731Crossref PubMed Scopus (55) Google Scholar, 8Tuan D.Y. Solomon W.B. London I.M. Lee D.P. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2554-2558Crossref PubMed Scopus (205) Google Scholar), silencers (9Raich N. Papayannopoulou T. Stamatoyannopoulos G. Enver T. Blood. 1992; 79: 861-864Crossref PubMed Google Scholar, 10Wada-Kiyama Y. Peters B. Noguchi C.T. J. Biol. Chem. 1992; 267: 11532-11538Abstract Full Text PDF PubMed Google Scholar), origins of replication (11Aladjem M.I. Groudine M. Brody L.L. Dieken E.S. Fournier R.E. Wahl G.M. Epner E.M. Science. 1995; 270: 815-819Crossref PubMed Scopus (196) Google Scholar, 12Fiering S. Epner E. Robinson K. Zhuang Y. Telling A. Hu M. Martin D.I. Enver T. Ley T.J. Groudine M. Genes Dev. 1995; 9: 2203-2213Crossref PubMed Scopus (295) Google Scholar, 13Kitsberg D. Selig S. Keshet I. Cedar H. Nature. 1993; 366: 588-590Crossref PubMed Scopus (247) Google Scholar), and putative insulators (14Li Q. Stamatoyannopoulos G. Blood. 1994; 84: 1399-1401Crossref PubMed Google Scholar). By a series of structural and functional experiments based on DNA-protein interactions (15Caterina J.J. Ryan T.M. Pawlik K.M. Palmiter R.D. Brinster R.L. Behringer R.R. Townes T.M. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 1626-1630Crossref PubMed Scopus (95) Google Scholar, 16Caterina J.J. Ciavatta D.J. Donze D. Behringer R.R. Townes T.M. Nucleic Acids Res. 1994; 22: 1006-1011Crossref PubMed Scopus (80) Google Scholar, 17Curtin P.T. Liu D.P. Liu W. Chang J.C. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 7082-7086Crossref PubMed Scopus (78) Google Scholar, 18Fraser P. Hurst J. Collis P. Grosveld F. Nucleic Acids Res. 1990; 18: 3503-3508Crossref PubMed Scopus (140) Google Scholar, 19Fraser P. Pruzina S. Antoniou M. Grosveld F. Genes Dev. 1993; 7: 106-113Crossref PubMed Scopus (197) Google Scholar, 20Ikuta T. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10188-10192Crossref PubMed Scopus (70) Google Scholar, 21Pruzina S. Antoniou M. Hurst J. Grosveld F. Philipsen S. Biochim. Biophys. Acta. 1994; 1219: 351-360Crossref PubMed Scopus (18) Google Scholar, 22Talbot D. Philipsen S. Fraser P. Grosveld F. EMBO J. 1990; 9: 2169-2177Crossref PubMed Scopus (234) Google Scholar, 23Ryan T.M. Behringer R.R. Martin N.C. Townes T.M. Palmiter R.D. Brinster R.L. Genes Dev. 1989; 3: 314-323Crossref PubMed Scopus (179) Google Scholar) as well as selective mutagenesis and expression studies in cell lines and transgenic mice (17Curtin P.T. Liu D.P. Liu W. Chang J.C. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 7082-7086Crossref PubMed Scopus (78) Google Scholar, 18Fraser P. Hurst J. Collis P. Grosveld F. Nucleic Acids Res. 1990; 18: 3503-3508Crossref PubMed Scopus (140) Google Scholar, 22Talbot D. Philipsen S. Fraser P. Grosveld F. EMBO J. 1990; 9: 2169-2177Crossref PubMed Scopus (234) Google Scholar, 23Ryan T.M. Behringer R.R. Martin N.C. Townes T.M. Palmiter R.D. Brinster R.L. Genes Dev. 1989; 3: 314-323Crossref PubMed Scopus (179) Google Scholar, 24Grosveld F., van Assendelft G. Greaves D.R. Kollias G. Cell. 1987; 51: 975-985Abstract Full Text PDF PubMed Scopus (1430) Google Scholar, 25Philipsen S. Talbot D. Fraser P. Grosveld F. EMBO J. 1990; 9: 2159-2167Crossref PubMed Scopus (201) Google Scholar, 26Philipsen S. Pruzina S. Grosveld F. EMBO J. 1993; 12: 1077-1085Crossref PubMed Scopus (81) Google Scholar, 27Pruzina S. Hanscombe O. Whyatt D. Grosveld F. Philipsen S. Nucleic Acids Res. 1991; 19: 1413-1419Crossref PubMed Scopus (112) Google Scholar), several short DNA consensus sequences have been identified to bind regulatory proteins that represent the effectors of the activities ascribed to the locus control region. One such sequence (TGAGTCA) that is repeated twice in the core of the HS2 enhancer is recognized by proteins of the AP1 (28Lee W. Mitchell P. Tjian R. Cell. 1987; 49: 741-752Abstract Full Text PDF PubMed Scopus (1360) Google Scholar), cAMP-responsive element-binding protein, and NFE2/CNC-bZip families of transcription factors (29Mignotte V. Eleouet J.F. Raich N. Romeo P.H. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 6548-6552Crossref PubMed Scopus (181) Google Scholar, 30Mignotte V. Wall L. de Boer E. Grosveld F. Romeo P.H. Nucleic Acids Res. 1989; 17: 37-54Crossref PubMed Scopus (216) Google Scholar) and is known as the NFE2/AP1 consensus sequence. As the latter motif is frequently found along the globin clusters in DNA elements with enhancer activity, cloning p45-NFE2 (31Andrews N.C. Erdjument-Bromage H. Davidson M.B. Tempst P. Orkin S.H. Nature. 1993; 362: 722-728Crossref PubMed Scopus (565) Google Scholar, 32Chan J.Y. Han X.L. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11366-11370Crossref PubMed Scopus (109) Google Scholar, 33Ney P.A. Andrews N.C. Jane S.M. Safer B. Purucker M.E. Weremowicz S. Morton C.C. Goff S.C. Orkin S.H. Nienhuis A.W. Mol. Cell. Biol. 1993; 13: 5604-5612Crossref PubMed Scopus (162) Google Scholar) has raised much interest as it could provide useful insights on the transcription factor enhancement of the globin gene expression, which in turn might lead to novel therapeutic approaches for inherited hemoglobin diseases such as sickle cell and thalassemia syndromes.Soon after the cloning of NFE2, we and others have cloned two more genes, NRF1 (also known as LCR-F1 andTCF11) (34Chan J.Y. Han X.L. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11371-11375Crossref PubMed Scopus (293) Google Scholar, 35Caterina J.J. Donze D. Sun C.W. Ciavatta D.J. Townes T.M. Nucleic Acids Res. 1994; 22: 2383-2391Crossref PubMed Scopus (123) Google Scholar, 36Luna L. Johnsen O. Skartlien A.H. Pedeutour F. Turc-Carel C. Prydz H. Kolsto A.B. Genomics. 1994; 22: 553-562Crossref PubMed Scopus (84) Google Scholar) and NRF2 (37Moi P. Chan K. Asunis I. Cao A. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 9926-9930Crossref PubMed Scopus (1186) Google Scholar), which is highly related to NFE2, thus extending the NFE2 family to three members and predicting the existence of a fourth member on the basis of linkage with other large gene families (COL,INT, and HOX) on specific chromosomes (38Chan J.Y. Cheung M.C. Moi P. Chan K. Kan Y.W. Hum. Genet. 1995; 95: 265-269Crossref PubMed Scopus (58) Google Scholar). The three genes are highly homologous in the DNA binding domain and leucine zipper but as is usually the case among related transcription factors, they are completely different in the activation domain. The genes are also differently regulated as NFE2 is restricted to hematopoietic tissues, whereas NRF1 and NRF2 are ubiquitously expressed. Despite the dramatic decrease in the HS2 enhancer activity produced by the disruption of the NFE2 consensus enhancer sequence (5Moi P. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 9000-9004Crossref PubMed Scopus (77) Google Scholar, 6Ney P.A. Sorrentino B.P. McDonagh K.T. Nienhuis A.W. Genes Dev. 1990; 4: 993-1006Crossref PubMed Scopus (209) Google Scholar), knockout of the p45-NFE2 gene in mice resulted in a disorder of megakaryocyte maturation but only minimally decreased the expression of the globin genes (39Shivdasani R.A. Orkin S.H. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 8690-8694Crossref PubMed Scopus (182) Google Scholar), suggesting that other factors are able to compensate for the lack of p45-NFE2 activity. Although Nrf1 and Nrf2 are ubiquitous factors, both are highly expressed in erythroid tissues and are able to transactivate globin gene promoters. Thus Nrf1 and Nrf2 could potentially compensate for p45-NFE2 function in the p45-NFE2 knockout mouse. Although this would suggest an evolutionary redundancy to protect and maintain the crucial body function of oxygen delivery, the observation that none of these factors bind DNA by itself and the identification of the p45-NFE2-associated protein p18 (40Andrews N.C. Kotkow K.J. Ney P.A. Erdjument-Bromage H. Tempst P. Orkin S.H. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11488-11492Crossref PubMed Scopus (239) Google Scholar) raise the possibility that p45-NFE2 and the related factors might indeed play distinct functions by dimerization with their respective partners. Interestingly, the p45-NFE2 partner p18 belongs to another family of bZip transcription factors (the Maf family (41Fujiwara K.T. Kataoka K. Nishizawa M. Oncogene. 1993; 8: 2371-2380PubMed Google Scholar, 42Kataoka K. Nishizawa M. Kawai S. J. Virol. 1993; 67: 2133-2141Crossref PubMed Google Scholar, 43Kawai S. Goto N. Kataoka K. Saegusa T. Shinno K.H. Nishizawa M. Virology. 1992; 188: 778-784Crossref PubMed Scopus (36) Google Scholar)) whose members in chickens have different levels of tissue expression and could therefore drive the functional specificity of the proteins with which they associate. Even though the small Maf proteins do not seem to have an activation domain, a recent report suggests that they are able to modulate the activity of p45-NFE2 according to their preferential association into homodimers or heterodimers, resulting in a negative or positive regulatory activity on the target genes, respectively (44Igarashi K. Kataoka K. Itoh K. Hayashi N. Nishizawa M. Yamamoto M. Nature. 1994; 367: 568-572Crossref PubMed Scopus (396) Google Scholar).These studies emphasize how our knowledge on the globin regulation would benefit from isolation of the proteins that associate with Nrf1 and Nrf2. It was also predictable on the basis of similarities in the dimerization domains among members of the NFE2 family that the partners for Nrf1 and Nrf2 could also be found within the family of the Maf oncogenes. Here we describe the cloning of a small human MAFcDNA (hMAF) through recognition site screening of a K562 cDNA library with a probe derived from the NFE2 tandem repeat motif of HS2. Even though hMAF shares strong structural homology with the other small Maf proteins in its leucine zipper, it heterodimerizes specifically with Nrf1 and Nrf2 but not with p45-NFE2. As a consequence of heterodimerization, Nrf1 and Nrf2 acquire the ability to bind and stimulate the activity of the target promoters, whereas hMAF homodimers (lacking any activation domain) apparently repress transcription by keeping the heterodimers from binding to their recognition sites.RESULTSThe recognition site probe screening yielded a dozen clones, 10 of which bound with clear specificity to the wild-type but not to the mutated NFE2 probe. Some of these clones have already been reported (37Moi P. Chan K. Asunis I. Cao A. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 9926-9930Crossref PubMed Scopus (1186) Google Scholar), and here we present the cloning and characterization of one of the remaining clones.cDNA CloningThe primary screening yielded four overlapping cDNA clones of 300, 550, 600, and 1631 base pairs. As expected from an expression cloning procedure, all cDNAs contained a functional DNA binding domain composed of at least four heptads of the leucine zipper motif (51Landschulz W.H. Johnson P.F. McKnight S.L. Science. 1988; 240: 1759-1764Crossref PubMed Scopus (2521) Google Scholar). The longest of the cDNAs contained a relatively short open reading frame as well as 5′- and 3′-untranslated regions (Fig. 1). Since the 3′-end of the clone terminated with a putative poly(A) tail (a stretch of 23 adenines that was 14 nucleotides downstream from a poly(A) addition signal (TATAAA)), the 3′-untranslated region is likely to be complete. On the other hand, we found an in-frame stop codon within the sequences upstream of the initiation codon that will prevent the upstream extension of the internal open reading frame, confirming that the upstream sequences are truly untranslated and that the cDNA clone encodes a full protein product. This conclusion is further supported by the fact that the sequences surrounding the first ATG conform well to the Kozak rule (52Kozak M. Nucleic Acids Res. 1987; 15: 8125-8148Crossref PubMed Scopus (4151) Google Scholar) for optimal initiation of translation.Based on these observations the coding region for hMAF begins at nucleotide 190 and terminates at nucleotide 676. Other noteworthy features of the cDNA are the AT richness of the 3′-untranslated region and the presence of several putative destabilization signals (ATTTA). The latter feature suggests that the mRNA may undergo rapid turnover in vivo (53Domen J. Von Lindern M. Hermans A. Breuer M. Grosveld G. Berns A. Oncogene Res. 1987; 1: 103-112PubMed Google Scholar, 54Luthman H. Soderling-Barros J. Persson B. Engberg C. Stern I. Lake M. Franzen S.A. Israelsson M. Raden B. Lindgren B. Hjelmqvist L. Enerbäck S. Carlsson P. Bjursell G. Povoa G. Hall K. Jörnvall H. Eur. J. Biochem. 1989; 180: 259-265Crossref PubMed Scopus (45) Google Scholar, 55Uchida K. Morita T. Sato T. Ogura T. Yamashita R. Noguchi S. Suzuki H. Nyunoya H. Miwa M. Sugimura T. Biochem. Biophys. Res. Commun. 1987; 148: 617-622Crossref PubMed Scopus (110) Google Scholar).Predicted Protein StructureTranslation of the open reading frame embedded in our cDNA predicts a short protein of 162 amino acids and a molecular mass of 17.9 kDa. Since the amino acid sequence comparison in the protein data base revealed homology with the oncogene v-Maf and especially with the small Maf proteins, the clone was designated hMAF. Similar to the other small Maf proteins, hMAF has a classical bZip domain that takes up most of the protein structure. The leucine zipper is comprised of seven heptad repeats with the D position of the α-helix occupied by five leucines and by two leucine zipper-compatible highly conserved residues, a methionine in the middle and a valine in the terminal heptad. Comparison with the published small Maf protein sequences (Fig. 2) shows the highest degree of homology (93.2%) with the recently cloned chicken MafG (cMafG) (56Kataoka K. Igarashi K. Itoh K. Fujiwara K.T. Noda M. Yamamoto M. Nishizawa M. Mol. Cell. Biol. 1995; 15: 2180-2190Crossref PubMed Scopus (199) Google Scholar) and the lowest homology with the chicken MafF (64%) (41Fujiwara K.T. Kataoka K. Nishizawa M. Oncogene. 1993; 8: 2371-2380PubMed Google Scholar). Thus even though our clone is referred to as hMAF throughout the text and in the figures, it should be considered the human homologue of cMafG. The evolution of hMAF and cMafG from the other small Maf proteins appears to be driven by an insertion of 14 amino acids C-terminal to the leucine zipper and by a premature termination in the polypeptide chain resulting in the truncation of the last eight residues.Figure 2Alignment of the known members of the small Maf family. The alignment covers the full protein sequences. Identical residues are located in the gray background. Gaps are represented by dashes. The alignment was generated with the Mac DNAsis software (Hitachi) by the Higgins-Sharp algorithm in automatic mode and at the following settings: gap penalty, 20; top diagonals, 10; fixed gap penalty, 4; K-tuple, 2; window size, 5; floating gap penalty, 10. The homology cMafG/hMAF is 93.2%.View Large Image Figure ViewerDownload Hi-res image Download (PPT)hMAF Binding in Solution to the NFE2 Repeat MotifThe full-length and partial clones of hMAF were subcloned into prokaryotic (pET3a, Novagen) and eukaryotic (pcDNAI, Invitrogen) expression verctors flanked by phage RNA polymerase promoters. Proteins were prepared by in vitro transcription and translation and assayed in band-shift experiments for their ability to bind probes derived from the core HS2 enhancer containing either the full NFE2/AP1 tandem repeat or the isolated left and right repeats. In a previous paper (5Moi P. Kan Y.W. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 9000-9004Crossref PubMed Scopus (77) Google Scholar) we presented evidence that the most 5′ (left) repeat had a more extended consensus sequence and a higher binding affinity for NFE2 than the 3′ (right) repeat. Thus we postulated that in vivoNFE2 might be mainly recognizing the left repeat, whereas members of the AP1 family were probably binding to the right repeat. Both the full-length hMAF and a shorter protein truncated at amino acid position 110 (hMAFΔ) were able to bind strongly to the NFE2 tandem repeat probe (Fig. 3 A, lane 1, band c and Fig. 3 C, lane 15, band c), whereas binding to a single NFE2 site required the complete hMAF protein (Fig. 3 B, lane 3 and Fig. 3 C, compare lanes 5 and 9 to lane 15, where the truncated hMAFΔ produced band c).Figure 3Band-shift analysis. In panels A, B, C, and E the transcription factors assayed were the products of the in vitrotranslation reactions from reticulocyte lysates. In panel Dnuclear extracts from K562 were compared with in vitrotranslated protein products. fNFE2 and hNFE2 are the protein products of the two splicing variants of the NFE2 gene. The Δ symbol is used to indicate truncated proteins. All probes were derived from the NFE2/AP1 tandem repeat of HS2, which is schematically indicated on thetop of each panel by two linked open rectangles (a broken line rectangle is used to indicate the absence of a single repeat). Unlabeled oligonucleotides used in competition are preceded by opposing arrows (><) and derive from the left repeat (NFE2), the right repeat (AP1), and a mutant left repeat (NFM) carrying a nucleotide substitution that selectively abolishes NFE2 binding. In thelanes marked F the band-shift reactions did not contain any protein extract, whereas in the lanes L orLys they contained extracts of reticulocyte lysates incubated in the absence of DNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)As the bZIP proteins bind only after dimerization, these results were taken as evidence that hMAF can form stable homodimers and bind DNA in solution. Overall, the binding and competition experiments with the full tandem repeat and with the isolated left and right motifs indicate that hMAF binds with the highest affinity to the full NFE2 tandem repeat and, when tested individually, to the left NFE2 repeat.hMAF Heterodimerizes with Nrf1 and Nrf2Band-shift assays were also performed to assess the ability of hMAF to form heterodimers with the known members of the NFE2/CNC-bZip family. As p45-NFE2, Nrf1, and Nrf2 do not bind the DNA as homodimers and as they all have greater molecular masses than hMAF, the formation of heterodimers was expected to band-shift a second complex with slower mobility than the hMAF homodimer. Fig. 3 A (lanes 2, 5, and7) shows that a second slower mobility complex is indeed formed only when hMAF is mixed and preincubated with Nrf1 and Nrf2 and is never observed when combined with NFE2 (either the fetal (fNFE2) or the adult (hNFE2) splicing isoform (lanes 9 and11)). An even clearer result is obtained when the same experiments are repeated using truncated versions of the hMAF and Nrf1 proteins. An NRF1 cDNA spanning nucleotides 2184–3016 (Nrf1Δ) also cloned through recognition site screening with the tandem repeat probe encodes a shorter polypeptide that acquires the ability to form homodimers in band shifts of the tandem repeat probe (Fig. 3 C, lane 14 and Fig. 3 E,lane 3, band a). Thus Nrf1Δ in combination with the other partial clone hMAFΔ produces any possible species of homo- and heterodimers, greatly simplifying the assignment of the observed bands. The heterodimer hMAFΔ/Nrf1Δ (Fig. 3 C, lane 13 and Fig. 3 E, lane 4, band b) now shows a mobility intermediate between one of the two homodimers Nrf1Δ/Nrf1Δ and hMAFΔ/hMAFΔ (bands a andc, respectively, in Fig. 3 C, lanes 13–15 and Fig. 3 E, lanes 3–5). The appearance of the intermediate mobility complex (Fig. 3 C,lane 13 and Fig. 3 E, lane 4,band b) coincides with the attenuation of the signal corresponding to the hMAFΔ homodimer complex (Fig. 3 C,lane 13 and Fig. 3 E, lane 4,band c), confirming what was previously observed in the interaction between hMAF and Nrf2 (Fig. 3 A, lane 7, band a versus band c). Furthermore, splitting the NFE2 tandem repeat probe in the middle and assaying each NFE2 site separately in band shift shows that the heterodimer binds almost exclusively to the left site (Fig. 3 C, compare band b in lanes 3 and 7), whereas hMAFΔ and Nrf1Δ homodimers bind neither the left nor the right site (Fig.3 C, lanes 4, 5, 8, and9).It is also interesting to note that the complexes formed in band-shift analysis with the in vitro translated Nrf1/hMAF proteins show electrophoretic mobilities comparable to the most abundant complexes obtained with crude nuclear extracts from induced K562 cells (Fig. 3 D, lanes 3 and 4, bands a and b), indicating that Nrf1/hMAF heterodimers may participate in the formation of these complexes, which appear more abundant and migrate slower than the p18/NFE2 complexes (barely seen in these extracts).When the hMAF homodimers and hMAF/Nrf1 heterodimers bound to the tandem repeat are assayed for competition, the left (NFE2) and right (AP1) motifs competed efficiently, whereas the mutant left repeat NFM, which carries a T to G mutation in the NFE2 binding site thought to discriminate specifically among transcription factors NFE2 and AP1 (30Mignotte V. Wall L. de Boer E. Grosveld F. Romeo P.H. Nucleic Acids Res. 1989; 17: 37-54Crossref PubMed Scopus (216) Google Scholar), was unable to compete (Fig. 3 B, lanes 4–6and 8–10). These results indicate that both hMAF homodimers and hMAF/Nrf1 heterodimers have DNA binding affinities identical to the heterodimer p18-NFE2.Glutathione S-Transferase ExperimentsWe wanted to further confirm the selectivity of the hMAF/Nrf1 and hMAF/Nrf2 interactions with an independent assay based on the glutathioneS-transferase fusion analysis. The complete hMAF cDNA fused in-frame with the glutathione S-transferase gene in the vector pGEX2T (Pharmacia) and expressed in bacterial cells was anchored to glutathione-Sepharose beads and assayed for the ability to retain [35S]methionine-labeled Nrf1, Nrf2, NFE2, and hMAF proteins. After stringent washings the only proteins retained in the beads were Nrf1, Nrf2, and hMAF (Fig. 4). Thus the preferential interactions as defined by band-shift assays were confirmed.Figure 4Glutathione S-transferase assay. Denaturing SDS-PAGE of the [35S]methionine-labeled proteins specifically and not specifically retained by the beads of glutathione-Sepharose primed with the GST/hMAF fusion or with the GST protein, respectively, is shown. The same amount of labeled protein assayed in the binding reactions was loaded in the input lanes as a size reference marker.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Methylation Interference AnalysisA methylation interference assay was performed only for Nrf1Δ and hMAFΔ as they interacted to form any possible combination of homodimers and heterodimers, whereas for Nrf2 we were unable to find any truncation variant that allowed homodimer formation. The bands corresponding to the hMAFΔ/Nrf1Δ heterodimer (Fig. 3 C, lane 13 and Fig.3 E, lane 4, band b) and to the hMAFΔ and Nrf1Δ homodimers (Fig. 3 C, lanes 14 and15 and Fig. 3 E, lanes 3 and4, bands a and c, the latter separated in top and bottom sub-bands) were cut from a preparative gel and subjected to methylation-interference analysis. In the bottom sub-band formed by the Nrf1Δ homodimer (N/NB) the protection of G residues on the tandem NFE2 repeat probe was more pronounced on the left motif, whereas the top sub-band (N/NT) showed protection on both repeats (Fig.5 B, lanes 2 and 3). The Nrf1Δ/hMAFΔ heterodimer, on the other hand, produced a unique band with protection restricted to the left motif (Fig. 5 B,lane 4), confirming that it preferentially binds to the 5′ NFE2 site of the tandem repeat. The lack of interference in the right motif may indicate that heterodimers binding to the left motif probably sterically hinder the binding of a second complex to the right motif. The hMAF homodimer (M/M) has an interference pattern that closely resembles that of the hMAF/Nrf1 heterodimers (Fig. 5 A,lanes 1 and 2), confirming a previous observation in which the small subunit of the p18-NFE2 heterodimer drives binding site specificity (40Andrews N.C. Kotkow K.J. Ney P.A. Erdjument-Bromage H. Tempst P. Orkin S.H. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11488-11492Crossref PubMed Scopus (239) Google Scholar)).Figure 5Methylation-interference analysis. The pattern of binding interference with the methylated G nucleotides in the sense strand of the HS2 core enhancer is shown. Lane labels correspond to the bands produced by the homodimers hMAFΔ/hMAFΔ (M/M), Nrf1Δ/Nrf1Δ (top band,N/NT), Nrf1Δ/Nrf1Δ (bottom band,N/NB), heterodimers Nrf1Δ/hMAFΔ (N/M), and free DNA (F). On the right side of eachpanel the G nucleotides that interfered with protein binding are represented with open circles for homodimers and withopen squares for heterodimers. Gray andblack shading indicates partial and absent interference, respectively. In the sequence on the left, the NFE2/AP1 repeats are boxed. Open linked rectangles on thetop represent the NFE2 tandem repeat probe used in the preparative band-shift assay.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Tissue Expression of hMAFNorthern blot analysis showed a complex pattern of hybridization with multiple bands ranging in size from 1.7 to 9 kilobases (Fig. 6). The intensity of the bands appeared to vary widely among the different human tissues but also in the same tissues among the differently sized mRNAs. Although we could observe a mRNA band of 1.7 kilobases corresponding to" @default.
• W2079500084 created "2016-06-24" @default.
• W2079500084 creator A5002234465 @default.
• W2079500084 creator A5004204696 @default.
• W2079500084 creator A5030084867 @default.
• W2079500084 creator A5038994974 @default.
• W2079500084 creator A5046272153 @default.
• W2079500084 creator A5079507122 @default.
• W2079500084 date "1997-06-01" @default.
• W2079500084 modified "2023-10-11" @default.
• W2079500084 title "hMAF, a Small Human Transcription Factor That Heterodimerizes Specifically with Nrf1 and Nrf2" @default.
• W2079500084 cites W139144773 @default.
• W2079500084 cites W1493457372 @default.
• W2079500084 cites W1565316587 @default.
• W2079500084 cites W1591462951 @default.
• W2079500084 cites W1714683344 @default.
• W2079500084 cites W1966844004 @default.
• W2079500084 cites W1967868770 @default.
• W2079500084 cites W1968092959 @default.
• W2079500084 cites W1970500252 @default.
• W2079500084 cites W1973595140 @default.
• W2079500084 cites W1974525592 @default.
• W2079500084 cites W1974858651 @default.
• W2079500084 cites W1978370818 @default.
• W2079500084 cites W1978667215 @default.
• W2079500084 cites W1978713358 @default.
• W2079500084 cites W1988690249 @default.
• W2079500084 cites W1991265255 @default.
• W2079500084 cites W1995985230 @default.
• W2079500084 cites W2004563262 @default.
• W2079500084 cites W2004995908 @default.
• W2079500084 cites W2005393972 @default.
• W2079500084 cites W2006918190 @default.
• W2079500084 cites W2010057451 @default.
• W2079500084 cites W2011509827 @default.
• W2079500084 cites W2014912471 @default.
• W2079500084 cites W2014925424 @default.
• W2079500084 cites W2015413754 @default.
• W2079500084 cites W2020506179 @default.
• W2079500084 cites W2023881776 @default.
• W2079500084 cites W2028553942 @default.
• W2079500084 cites W2038210364 @default.
• W2079500084 cites W2041798627 @default.
• W2079500084 cites W2045348184 @default.
• W2079500084 cites W2052042646 @default.
• W2079500084 cites W2055643269 @default.
• W2079500084 cites W2057321580 @default.
• W2079500084 cites W2068446924 @default.
• W2079500084 cites W2071919914 @default.
• W2079500084 cites W2077643007 @default.
• W2079500084 cites W2081135713 @default.
• W2079500084 cites W2081875871 @default.
• W2079500084 cites W2083295035 @default.
• W2079500084 cites W2083517254 @default.
• W2079500084 cites W2087652779 @default.
• W2079500084 cites W2091046288 @default.
• W2079500084 cites W2094921802 @default.
• W2079500084 cites W2096998693 @default.
• W2079500084 cites W2101141723 @default.
• W2079500084 cites W2112436307 @default.
• W2079500084 cites W2118167494 @default.
• W2079500084 cites W2118299921 @default.
• W2079500084 cites W2121858619 @default.
• W2079500084 cites W2122297685 @default.
• W2079500084 cites W2142342476 @default.
• W2079500084 cites W2150498320 @default.
• W2079500084 cites W2169578753 @default.
• W2079500084 cites W4239493272 @default.
• W2079500084 doi "https://doi.org/10.1074/jbc.272.26.16490" @default.
• W2079500084 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9195958" @default.
• W2079500084 hasPublicationYear "1997" @default.
• W2079500084 type Work @default.
• W2079500084 sameAs 2079500084 @default.
• W2079500084 citedByCount "119" @default.
• W2079500084 countsByYear W20795000842012 @default.
• W2079500084 countsByYear W20795000842013 @default.
• W2079500084 countsByYear W20795000842014 @default.
• W2079500084 countsByYear W20795000842015 @default.
• W2079500084 countsByYear W20795000842016 @default.
• W2079500084 countsByYear W20795000842017 @default.
• W2079500084 countsByYear W20795000842018 @default.
• W2079500084 countsByYear W20795000842019 @default.
• W2079500084 countsByYear W20795000842020 @default.
• W2079500084 countsByYear W20795000842021 @default.
• W2079500084 countsByYear W20795000842022 @default.
• W2079500084 countsByYear W20795000842023 @default.
• W2079500084 crossrefType "journal-article" @default.
• W2079500084 hasAuthorship W2079500084A5002234465 @default.
• W2079500084 hasAuthorship W2079500084A5004204696 @default.
• W2079500084 hasAuthorship W2079500084A5030084867 @default.
• W2079500084 hasAuthorship W2079500084A5038994974 @default.
• W2079500084 hasAuthorship W2079500084A5046272153 @default.
• W2079500084 hasAuthorship W2079500084A5079507122 @default.
• W2079500084 hasBestOaLocation W20795000841 @default.
• W2079500084 hasConcept C104317684 @default.
• W2079500084 hasConcept C54355233 @default.
• W2079500084 hasConcept C74438686 @default.
• W2079500084 hasConcept C86339819 @default.

• next