Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009 Dec;29(23):6232-44.
doi: 10.1128/MCB.00708-09. Epub 2009 Sep 21.

Structural basis of alternative DNA recognition by Maf transcription factors

Hirofumi Kurokawa  1 , Hozumi Motohashi, Shinji Sueno, Momoko Kimura, Hiroaki Takagawa, Yousuke Kanno, Masayuki Yamamoto, Toshiyuki Tanaka
Affiliations

Structural basis of alternative DNA recognition by Maf transcription factors

Hirofumi Kurokawa et al. Mol Cell Biol. 2009 Dec.

Abstract

Maf transcription factors constitute a family of the basic region-leucine zipper (bZip) factors and recognize unusually long DNA motifs (13 or 14 bp), termed the Maf recognition element (MARE). The MARE harbors extended GC sequences on each side of its core motif, which is similar to TRE or CRE (7 or 8 bp) recognized by the AP1 and CREB/ATF families, respectively. To ascertain the structural basis governing the acquirement of such unique DNA recognition, we determined the crystal structure of the MafG-DNA complex. Each MafG monomer consists of three helices in which the carboxyl-terminal long helix organizes one DNA-contacting element and one coiled-coil dimer formation element. To our surprise, two well-conserved residues, Arg57 and Asn61 in the basic region, play critical roles in Maf-specific DNA recognition. These two residues show unique side-chain orientations and interact directly with the extended GC bases. Maf-specific residues in the amino-terminal and basic regions appear to indirectly stabilize MARE recognition through DNA backbone phosphate interactions. This study revealed an alternative DNA recognition mechanism of the bZip factors that bestows specific target gene profiles upon Maf homodimers or Maf-containing heterodimers.

PubMed Disclaimer

Figures

FIG. 1.
FIG. 1.
Maf recognition elements and sequence conservation in Maf-type, CNC-type, and AP1-type bZip factors. (A) Consensus DNA sequences recognized by bZip transcription factors. The TRE and the CRE lie within the T-MARE and the C-MARE, respectively, and act as the established core sequences of these MAREs. Three bases on each side of the core form the flanking sequence, which is critical for the recognition of the MARE by the Maf protein family. The NF-E2 binding site and ARE are asymmetrical sequences composed of half of the TRE site and half of the MARE site with one flanking sequence. The core and flanking regions are in red and blue, respectively. cAMP, cyclic AMP. (B) Sequence alignment of various subfamilies of dimerizing bZip transcription factors and their consensus sequences. The conserved arginine (Arg57 of MafG) and asparagine (Asn61 of MafG) are highlighted in blue and orange, respectively. Maf-specific tyrosine and the corresponding alanine (serine for BBF-2) are in pink and yellow, respectively. (C) EMSA of MafG mutant molecules with a probe containing the consensus MARE. MafG(1-123), MafG(21-123), and MafG(45-123) were incubated with the MARE probe. The concentrations of each protein are 250 nM (lanes 2, 6, and 10), 50 nM (lanes 3, 7, and 11), 10 nM (lanes 4, 8, and 12), and 2 nM (lanes 5, 9, and 13). (D) Sequence alignment of small Maf, large Maf, CNC, Bach, and AP1-type transcription factors. The EHR of Maf proteins, the basic regions, and the partial CNC domains of the CNC family are boxed. H1, H2, and H3 along the top indicate the α-helical regions. The amino acid sequences of the EHR and CNC domains are aligned to each other according to structural correspondence. The amino acid positions in the leucine heptad repeats are indicated at the bottom, and amino acids at position “d” are highlighted in green.
FIG. 2.
FIG. 2.
Main-chain structure of MafG in complex with MARE DNA. (A and B) Orthogonal views of MafG in complex with its cognate DNA. MafG consists of three helices: H1 and H2 in the EHR and H3 in both the basic region that recognizes DNA and the leucine zipper region that forms a dimer with the other subunit. The sequence of the duplex oligonucleotide MARE25 that was used for cocrystallization is shown (A). Subunit A (in green) is in contact with the GA flanking region, and subunit B (in cyan) is in contact with the GC flanking region. (C) Comparison with Fos (in red)-Jun (in purple) in complex with DNA. The AP1 cores are superimposed. Structural alignment yields root mean square deviations for 106 superimposed Cα atoms (rmsd106Cα) of 2.5 Å when the AP1 cores are superimposed and 1.7 Å when the protein regions are superimposed. (D) Comparison of MafG with Skn1 (in orange). The protein regions are used for superimposition. Structural alignment yields rmsd57Cα of 2.4 Å when the AP1 half-cores are superimposed and 1.8 Å when only the protein regions are superimposed.
FIG. 3.
FIG. 3.
Interhelical interactions. (A) Detailed view of the leucine zipper region. Amino acids constituting the heptad repeat region (subunit A, in green) and those forming interactions as a counterpart (subunit B, in blue) are indicated. Dotted lines indicate hydrogen bonds. Intersubunit interactions between each Gln82 and Lys83 and between each Gln75 and Lys76 via a water molecule are indicated. (B and C) Helical-wheel diagram of the MafG homodimer (B) and the Fos-Jun heterodimer (C). The coiled-coil sequence is read from the N terminus to the C terminus outwards from the wheel. Arrows with solid lines and broken lines indicate direct and indirect (via a water molecule) electrostatic interactions, respectively.
FIG. 4.
FIG. 4.
Conserved Asn61 and Arg57 in the basic region of MafG contact with the MARE sequence. (A and B) Orthogonal views of the basic region of subunit A. Some of the amino acids making critical contacts are indicated. The AP1 core and extended sequence elements (Fig. 1A) are in pink and blue, respectively. (C and D) The simulated-annealing Fo-Fc omit map is contoured at 5 σ. Arg57, Asn61, and Tyr64 were omitted from the calculation. Note that the electron density of these three residues is unambiguously visible. (C) In subunit A, side chains of Tyr64 and Arg57 contact via a water molecule. An indirect interaction between Arg57 and Asn61 was also observed. Hydrogen bonds are shown as broken lines. (D) In subunit B, these interactions via a water molecule were not observed. (E and F) Comparison of the side chains of conserved residues within MafG (green), c-Fos (blue), and Skn1 (yellow). The image in E was rotated by 90° and is shown in F.
FIG. 5.
FIG. 5.
MARE recognition by the basic region of MafG and comparison with AP1-type bZip factors. (A) The phosphate groups of MARE25 engaged by the MafG protein are shown. Chain C and chain D are the bottom and top strands of MARE25, respectively (Fig. 2A). The figure was drawn by NUCPLOT (23). (B to F) The conserved amino acid side chains of the motifs that make direct contacts with the DNA bases of the consensus sequence are shown for MafG subunit A (B), MafG subunit B (C), Skn1 (D), Pap1 (E), and C/EBP (F). Hydrogen bonds and van der Waals contacts are shown as solid and broken lines, respectively. PDB accession numbers 1SKN for Skn1 (D), 1GD2 for Pap1 (E), and 1GU4 for C/EBP (F) were used to estimate the DNA base recognition mode.
FIG. 6.
FIG. 6.
A missing-phosphate assay demonstrates the critical contribution of the backbone phosphate in the flanking region of MARE. (A) Three probe types were used for EMSAs. BglHS4-MARE is a parent probe of intact double-stranded oligonucleotides containing the conserved core region and one side of the flanking region, which allows the oriented binding of the p45-MafG heterodimer but not of the MafG homodimer. The nicked MARE-1 and MARE-2 probes lack the 5′ phosphate of the flanking “GC” bases and the corresponding phosphate on the other strand, respectively. (B) EMSA of p45ΔN-MafG(1-123) with intact double-stranded parent probe (lanes 1 to 7) and the nicked probes (lanes 8 to 21). p45ΔN is an N-terminally truncated p45 that contains CNC and bZip motifs. The concentrations of p45ΔN are 0.1 μM (lanes 3, 10, and 17), 0.3 μM (lanes 4, 11, and 18), 0.9 μM (lanes 5, 12, and 19), and 3 μM (lanes 6, 7, 13, 14, 20, and 21), with 0.3 μM of MafG added in lanes 2 to 6, 9 to 13, and 16 to 20.
FIG. 7.
FIG. 7.
Interactions of the DNA binding domain of MafG with DNA. (A) van der Waals forces between Maf-specific Thr58 and the DNA backbone phosphate. The hydrogen bond between Gln54 and Thr58 fixes the Thr58 side chain in an optimal orientation against the backbone phosphate. (B) The DNA backbone phosphate referred to above (A) interacts with Arg507 of Skn1. Arg507 corresponds to Arg57 of MafG and forms indirect contacts with the invariant Asn511 via a water molecule, thereby stabilizing the orientation of the Asn511 side chain for recognition of the core TRE half-region.
FIG. 8.
FIG. 8.
Comparison of the interactions between basic regions and juxtaposed ancillary domains. (A) Schematic representation of the interactions between the EHR and the basic region of MafG (subunit A). (B) Graphic depiction of the contacts made between the CNC domain and the basic region of Skn1.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Bessant, D. A., A. M. Payne, K. P. Mitton, Q. L. Wang, P. K. Swain, C. Plant, A. C. Bird, D. J. Zack, A. Swaroop, and S. S. Bhattacharya. 1999. A mutation in NRL is associated with autosomal dominant retinitis pigmentosa. Nat. Genet. 21:355-356. - PubMed
    1. Brunger, A. T., P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W. Grosse-Kunstleve, J. S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu, R. J. Read, L. M. Rice, T. Simonson, and G. L. Warren. 1998. Crystallography & NMR system: a new software suite for macromolecular structure determination. Acta Crystallogr. D Biol. Crystallogr. 54:905-921. - PubMed
    1. DeLano, W. L. 2002. The PyMOL molecular graphics system. DeLano Scientific, San Carlos, CA.
    1. Dlakic, M., A. V. Grinberg, D. A. Leonard, and T. K. Kerppola. 2001. DNA sequence-dependent folding determines the divergence in binding specificities between Maf and other bZIP proteins. EMBO J. 20:828-840. - PMC - PubMed
    1. Ellenberger, T. E., C. J. Brandl, K. Struhl, and S. C. Harrison. 1992. The GCN4 basic region leucine zipper binds DNA as a dimer of uninterrupted alpha helices: crystal structure of the protein-DNA complex. Cell 71:1223-1237. - PubMed

Publication types

MeSH terms

LinkOut - more resources

-