DNA Sequence Determinants Controlling Affinity, Stability and Shape of DNA Complexes Bound by the Nucleoid Protein Fis - PubMed (original) (raw)

DNA Sequence Determinants Controlling Affinity, Stability and Shape of DNA Complexes Bound by the Nucleoid Protein Fis

Stephen P Hancock et al. PLoS One. 2016.

Erratum in

Abstract

The abundant Fis nucleoid protein selectively binds poorly related DNA sequences with high affinities to regulate diverse DNA reactions. Fis binds DNA primarily through DNA backbone contacts and selects target sites by reading conformational properties of DNA sequences, most prominently intrinsic minor groove widths. High-affinity binding requires Fis-stabilized DNA conformational changes that vary depending on DNA sequence. In order to better understand the molecular basis for high affinity site recognition, we analyzed the effects of DNA sequence within and flanking the core Fis binding site on binding affinity and DNA structure. X-ray crystal structures of Fis-DNA complexes containing variable sequences in the noncontacted center of the binding site or variations within the major groove interfaces show that the DNA can adapt to the Fis dimer surface asymmetrically. We show that the presence and position of pyrimidine-purine base steps within the major groove interfaces affect both local DNA bending and minor groove compression to modulate affinities and lifetimes of Fis-DNA complexes. Sequences flanking the core binding site also modulate complex affinities, lifetimes, and the degree of local and global Fis-induced DNA bending. In particular, a G immediately upstream of the 15 bp core sequence inhibits binding and bending, and A-tracts within the flanking base pairs increase both complex lifetimes and global DNA curvatures. Taken together, our observations support a revised DNA motif specifying high-affinity Fis binding and highlight the range of conformations that Fis-bound DNA can adopt. The affinities and DNA conformations of individual Fis-DNA complexes are likely to be tailored to their context-specific biological functions.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1

Fig 1. Fis binding site logo and contacts made to the Fis-bound DNA.

(A) Qualitative sequence logo generated from a compilation of well-defined Fis binding sites, many of which have been shown to function directly as regulatory sites in transcription or recombination reactions, mutagenesis studies [29], and X-ray crystal structures (this paper and [–30]). The core motif is defined here as sequences between -7 and +7, and flanking sequences extend beyond these limits. Bases depicted below the numbering inhibit binding. This symmetric motif differs from recent genome-wide chromatin immunoprecipitation studies where motif finders return binding logos that exhibit asymmetrically positioned A-tracts within the core [1,2]. (B) Ladder diagram of Fis-DNA contacts in the reference Fis-F1 crystal structure (PDB ID: 3IV5). DNA phosphates are shown as circles, ribose sugars as pentagons, and bases as rectangles. Phosphates that are contacted are filled blue. Contacts made by chain A side chains (light blue) and those made by chain B (green) are shown. Asterisks represent contacts made by protein backbone atoms. Arrows represent contacts made to the phosphate backbone, whereas lines represent contacts to the bases. (C) Crystal structure of the high affinity Fis-F1 DNA complex. Secondary structural elements are labeled and shown as a cartoon. Fis side chains that contact the flanking DNA backbone are shown as sticks. Sequence elements that are hallmarks of high affinity Fis binding sites including the conserved ±7(G/C) (red), the ±(3–4) Y-R step (orange), and the A/T rich center (blue). Lines through the helical axis have been drawn to highlight the points of helix axis deflection. (D) Zoomed-in representation of the residues that contact the flanking DNA backbone. The Arg71 side chain, for which experimental electron density is weak, has been modeled in purple to interact with the ±(12–13) phosphate here and in panel C (see also [22].

Fig 2

Fig 2. Effects of the F35 base substitutions on Fis-DNA binding and structure.

(A) Crystal lattice differences between the F35 (left) and F1 (right) complexes. The contents of one asymmetric unit are bound by a trapezoid. Protein chains A and B are represented as blue and green cartoons, respectively, and DNA chains C and D are shown as magenta and orange cartoons. The relationship between symmetry mates is reversed in F35 as compared to F1. (B) Binding affinities and lifetimes for the F36 and F35 sequences relative to F1. The 15 bp core sequences are highlighted in grey. (C) Plot of minor groove width (van der Waal’s radii subtracted) for the bound DNAs in the F35, F36, and F1 complexes.) The color of the plots corresponds to the color of the DNA sites below. Bases that are different than those in F1 are bold and underlined. (D) Roll angle plots of the bound DNAs in the F35 relative to F1 complexes. (E) Electrostatic potential calculations (±3 kT/e) mapped on to the surface of the F35 Fis molecule. Blue and red colors represent electropositive and electronegative surfaces, respectively. Note the DNA writhe as it follows the basic track extending along the sides of the Fis dimer. The surface potential map and DNA track of the F35 complex is indistinguishable from the F1 complex.

Fig 3

Fig 3. Effects of the flexible ±(3–4) Y-R step on Fis binding, stability, and DNA geometry.

(A) Zoomed in view of the contacts made to the phosphate backbone in the crystal structure of the reference high affinity Fis-F1 complex. Protein subunits (blue and green cartoon) and DNA chains (grey, except for base pairs 4–6 which are colored by atom) are shown and labeled. The Arg71 side chain has been modeled to contact the phosphate backbone (purple). The structure of Fis-F1, which contains a TG step at +(3–4), is nearly identical to that of Fis-F2 containing a TA step at +(3–4) [28]. (B) Sequences, equilibrium binding affinities, and lifetimes for DNA sites with displaced (F18 and F31) or removed (F32) ±(3–4) Y-R steps. Bases that differ from those in the F1 sequence are bold and underlined. (C-E) Roll angle plots for each of the major groove Y-R variants relative to F1. Bases that differ from those in the F1 sequence are underlined. (F) Minor groove plot for each of the major groove Y-R variants relative to F1. Bases that differ from those in the F1 sequence are underlined. (G & H) Position of the Arg89 side chain in the F1 (G) and F32 (H) structures. The 2Fo-Fc electron density map for the Arg89 side chain is shown for the F1 (pink) and F32 (blue) structures at 1.0σ. Dotted lines represent contacts that are within 3.4 Å. (I-K) Asn84 contacts in the F1, F31, and F32 crystal structures, respectively. Relevant protein and DNA elements are labeled.

Fig 4

Fig 4. Effects of ±8 substitutions on Fis-DNA binding, stability, and structure.

(A) Equilibrium binding affinities and lifetimes for DNA sites with symmetric ±8 substitutions. (B) Fis-F1 crystal structure highlighting contacts between nucleotides -7G and -8T and Fis residues Thr75 and Arg85. Note van der Waals surface contacts between the -8T C5 methyl and Thr75:Oγ1 and Arg85:CZ. (C and D) Same region as highlighted in panel B for the Fis-F1±A (C) and Fis-F1±G (D) crystal structures. Van der Waals contacts to the -8 base are absent in these as well as the F1±8C structure (not shown). (E) DNA backbone positional variations in the Fis-F1 (blue), Fis-8A (red), Fis-8C (green) and Fis-8G (orange) structures. Structures were aligned over the Fis proteins. Only Fis-F1 side chains that contact the flanking DNA backbone are shown. Contacts within 3.4 Å are represented by dotted lines. (F) Plot of minor groove widths for each of the ±8 variants. Bases that differ from those in the F1 sequence are underlined.

Fig 5

Fig 5. Effects of DNA base substitutions on DNA bending by Fis.

(A) Electrophoretic mobility shift assays showing migrations of free and Fis-bound DNA fragments. DNA fragments containing two to eight in-phase A-tracts were used as standards for estimating bend angles ([41]; left panel). Faster migrating fragments contain A-tracts at the end of the fragment (red squares), whereas slower migrating fragments contain A-tracts in the middle (red circles). Right panel: Fis-DNA complexes formed on 422 bp DNA fragments in which the Fis binding sites were located near the end (E) or the middle (M) of the fragment. Bands corresponding to unbound fragments (blue squares) and Fis-bound fragments (blue circles) are marked. (B) Representative FRET gel for Fis binding to F1 DNA. The intensity of the donor fluorophore emission in the absence (left titration) or presence (right titration) of the acceptor fluorophore (Alexa Fluor 555) was quantified from a scan of the gel where λex = 488 nm and fluorescence emission was collected through a 520 nm emission filter. The same gels were scanned by phosphorimaging as described in the Methods (not shown). Cartoons describe the species that correspond to the bands on the gel. The left titration is that for a donor-only labeled DNA (“D”; blue rectangle with the Fis site shown in red). The donor fluorophore (Alexa Fluor 488 –green hexagon) is conjugated to the 3′ end of the DNA oligonucleotide by a 6 carbon linker (black line). The free DNA is shifted to the upper band when bound by the Fis dimer (light blue ovals). The right half of the gel represents the titration of the donor and acceptor labeled DNA (“DA”; Acceptor is Alexa Fluor 555 –red hexagon). (C) A plot showing the correlation between the fluorescence intensities of the donor-acceptor labeled DNA (IDA) and donor-only labeled F1 DNA (ID) in the presence (blue diamonds) and absence (red squares) of Fis. The plots were fit to a line with the equation IDA[(complexD)/(complexDA)] = ID(1 –E), where E is FRET efficiency and the slope is (1-E) as described in the Methods. Axis units are x1000.

References

    1. Cho BK, Knight EM, Barrett CL, Palsson BO (2008) Genome-wide analysis of Fis binding in Escherichia coli indicates a causative role for A-/AT-tracts. Genome Res 18: 900–910. 10.1101/gr.070276.107 -DOI -PMC -PubMed
    1. Kahramanoglou C, Seshasayee AS, Prieto AI, Ibberson D, Schmidt S, Zimmermann J, et al. (2011) Direct and indirect effects of H-NS and Fis on global gene expression control in Escherichia coli. Nucleic Acids Res 39: 2073–2091. 10.1093/nar/gkq934 -DOI -PMC -PubMed
    1. Wang H, Liu B, Wang Q, Wang L (2013) Genome-wide analysis of the Salmonella Fis regulon and its regulatory mechanism on pathogenicity islands. PLoS One 8: e64688 10.1371/journal.pone.0064688 -DOI -PMC -PubMed
    1. Johnson RC, Johnson LM, Schmidt JW, Gardner JF (2005) Major nucleoid proteins in the structure and function of the Escherichia coli chromosome In: Higgins NP, editor. The Bacterial Chromosome. Washington, D.C.: ASM Press; pp. 65–132.
    1. Browning DF, Grainger DC, Busby SJ (2010) Effects of nucleoid-associated proteins on bacterial chromosome structure and gene expression. Curr Opin Microbiol 13: 773–780. 10.1016/j.mib.2010.09.013 -DOI -PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources