Structure of the DNA-binding domain of the response regulator PhoP from Mycobacterium tuberculosis

Abstract

The PhoP-PhoR two-component signaling system from Mycobacterium tuberculosis is essential for the virulence of the tubercle bacillus. The response regulator, PhoP, regulates expression of over 110 genes. In order to elucidate the regulatory mechanism of PhoP, we determined the crystal structure of its DNA-binding domain (PhoPC). PhoPC exhibits a typical fold of the winged helix-turn-helix subfamily of response regulators. The structure starts with a four-stranded antiparallel beta-sheet, followed by a three-helical bundle of alpha-helices, and then a C-terminal beta-hairpin, which together with a short beta-strand between the first and second helices forms a three-stranded antiparallel beta-sheet. Structural elements are packed through a hydrophobic core, with the first helix providing a scaffold for the rest of the domain to pack. The second and third helices and the long, flexible loop between them form the helix-turn-helix motif, with the third helix being the recognition helix. The C-terminal beta-hairpin turn forms the wing motif. The molecular surfaces around the recognition helix and the wing residues show strong positive electrostatic potential, consistent with their roles in DNA binding and nucleotide sequence recognition. The crystal packing of PhoPC gives a hexamer ring, with neighboring molecules interacting in a head-to-tail fashion. This packing interface suggests that PhoPC could bind DNA in a tandem association. However, this mode of DNA binding is likely to be nonspecific because the recognition helix is partially blocked and would be prevented from inserting into the major groove of DNA. Detailed structural analysis and implications with respect to DNA binding are discussed.

Figure 1

Electrophoretic mobility shift assay of the binding of PhoPC with the promoter DNA of the phoP gene. Binding of PhoPC retards the electrophoretic mobility of DNA and thus causes a shift of the DNA band.

Figure 2

Ribbon diagram of the structure of the C-terminal domain of MTB PhoP. Secondary structural elements are labeled. The β-strands and α-helices are numbered starting from β6 and α6, respectively, for consistency with the N-terminal regulatory domain structure. Side chains of residues on the recognition helix (α8) that are exposed and are likely to be involved in DNA binding and recognition are shown as sticks. The side chain of residue Arg237 at the wing is also shown. Amino acids are referred to as single letter codes in the figure for clarity. The figure was generated with the program PYMOL (http://www.pymol.org).

Figure 3

Sequence alignment of the DNA-binding domain of response regulators of the OmpR/PhoB subfamily. Sequences were aligned with the program CLUSTALW (46), and the figure was generated with the program ALSCRIPT (47). The secondary structural elements shown are of the PhoPC structure. Residues labeled with a “ * “ are those found interacting with DNA in the structure of the PhoB-DNA complex (PDB code 1GXP) (16). Identical residues are highlighted in black boxes, while residues with high similarity as defined by CLUSTALW are highlighted in gray boxes. Among the sequences compared, OmpR of E. coli is the most distant from others. Several residues, including Val192, Trp203, and Arg237 (residue numbers refer to those of MTB PhoP sequence), are conserved among PhoP of MTB, DrrD of T. maritima, PhoP of B. subtilis, and PhoB of E. coli, but not in OmpR. Of these residues, Trp203 and Arg237 interact with DNA in the PhoB-DNA complex. Residues in the MTB PhoP sequence highlighted in rectangle boxes are involved in hydrophobic interactions between the N-terminal β-sheet and helix α6, and those in circles are involved in interactions between helices α6 and α7. These two hydrophobic clusters are connected by a cluster of side chains of residues Phe153 and Ala154 from the N-terminal β-sheet and residues Leu193, Leu232, Leu233, Tyr241, and Leu243 from the C-terminal β-sheet. Side chains of Val214, Val218, Leu221, and Ile225 from helix α8, as well as those of Leu174 and Trp203, also contribute to the hydrophobic core.

Figure 4

Ribbon diagrams of hexamer and dodecamer in the crystal structure of PhoPC. (a) The six PhoPC molecules in the asymmetric unit form a hexamer with a loose six-fold non-crystallographic symmetry. The subunits are designated as A to F as described in text. Each subunit is colored differently. The secondary structure elements of subunit F are labeled. The recognition helix α8 is in the front while the N-terminus is in the back. The subunits interact with each other in a head-to-tail fashion. The molecular interface involves the N-terminal four-stranded β-sheet of one molecule and one side of the recognition helix α8, the N-terminal end of helix α6, and the transactivation loop of the other molecule. (b) The two-fold crystallographic symmetry relates two hexamers to give a double ring dodecamer, shown as a side-view of (a). Individual subunits in each ring are colored differently as in (a). Each subunit of one ring is related by a two-fold symmetry to the neighboring subunit of the other ring to give a symmetric dimer, as shown by the two magenta subunits (molecule B) in the front (recognition helix α8 is labeled). Major dimeric interface involves the loop following the recognition helix α8. The amino termini of all subunits point away from the dodecamer.

Figure 5

Electrostatic potential surface diagrams of PhoPC. Positive charge is shown in blue, and negative charge is in red. The diagram in (a) shows the electrostatic potential surface in the same orientation as shown in the ribbon diagram in Figure 2. Shown in (b) is the backside of (a), i.e. approximately 180 ° rotation of (a) along a vertical axis. Panel (c) shows the surface potential of the hexamer in the same orientation as in Figure 4(a). Electrostatic potential was calculated and the figures were generated with the program Swiss-PdbViewer (44).

Figure 6

Stereo views of structural superposition of molecules A, B and C in the PhoPC structure (a) and of PhoPC with structures of the C-terminal domain of OmpR and PhoB (b). The structures are shown as Cα traces. Molecules A, B, and C in (a) are shown in magenta, orange, and green, respectively. Helices and the two termini are labeled in (a). In (b), the molecule B of MTB PhoPC is colored in orange, OmpR of E. coli (PDB code 1OPC (39)) is colored in black, and PhoB of E. coli, (1GXQ (16)) is colored in green. PhoPC is more similar to PhoB in structure, while OmpR has the largest deviations, especially in the transactivation loop and the loop following the recognition helix. The figures were prepared with the program MOLSCRIPT (45).

Figure 7

Structural superposition of the molecule B of MTB PhoPC and the PhoB-DNA complex (PDB code 1GXP). The two structures were aligned based on the recognition helix. The PhoB structure is shown in magenta as a Cα trace, and the PhoPC structure is shown in green as a Cα trace and a ribbon diagram. The side chains shown are some of those found in the PhoB-DNA complex to interact with DNA and the corresponding residues in PhoP. Other side chains at the N-terminal end of helix α6 and at strands β11 and β12 that also interact with DNA in the PhoB-DNA complex are not shown for clarity. Also shown is the side chain of Arg223, which occupies the position of Arg200 in PhoB even though these two residues are not aligned in sequence. The residue labels are for those of the PhoP sequence with single letter codes for amino acids. Based on the structural superposition, PhoP is likely to be able to bind DNA in a similar way with most of the interactions with phosphate and sugar groups of DNA conserved. The figure was generated with the programs MOLSCRIPT (45) and RASTER3D (48).