A monomeric mycobacteriophage immunity repressor utilizes two domains to recognize an asymmetric DNA sequence

Abstract

Regulation of bacteriophage gene expression involves repressor proteins that bind and downregulate early lytic promoters. A large group of mycobacteriophages code for repressors that are unusual in also terminating transcription elongation at numerous binding sites (stoperators) distributed across the phage genome. Here we provide the X-ray crystal structure of a mycobacteriophage immunity repressor bound to DNA, which reveals the binding of a monomer to an asymmetric DNA sequence using two independent DNA binding domains. The structure is supported by small-angle X-ray scattering, DNA binding, molecular dynamics, and in vivo immunity assays. We propose a model for how dual DNA binding domains facilitate regulation of both transcription initiation and elongation, while enabling evolution of other superinfection immune specificities.

Fig. 1. Overall structure of the repressor:DNA complex.

a Cartoon representation of the repressor bound to DNA. The protein monomer is colored by domain and includes the N-terminal HTH (residues 15–55, slate blue), the helix bridge (residues 56–74, magenta), and the Stoperator (residues 75–181, orange). The protein secondary structural elements are labeled as either alpha helices (α), beta strands (β), or 3₁₀ helices (η), and the two DNA strands are colored in green and white. 2F_o – F_c density, contoured at 1 sigma, is shown in gray for the DNA helix. The two DNA-binding domains engage the DNA via insertion into adjacent openings of the major groove, while the helix bridge serves as a linker between the two DNA-binding domains and lies above the minor groove. b The repressor is shown in surface view and is color-coded as in a. The arrows shown in a, b indicate the direction of transcription. c Surface view of the lambda cI dimer bound to DNA. This image was generated from coordinates 3BDN and shows the two monomers of the cI dimer colored in cyan and brown, with the two DNA strands colored in green and white. In cI, the HTH domain from each monomer of the dimer binds adjacent openings of the major groove. d Superposition of the repressor HTH (slate blue) and the DNA-binding motif of the Stoperator domain (orange). The Stoperator lacks the α2 helix of the HTH domain and also contains a small 3₁₀-helix (η1) at its N-terminus. e The region of the Stoperator domain that does not bind the DNA substrate is emphasized with a black circle. An electrostatic surface rendering (red: negative potential, blue: positive potential, white: neutral) reveals that this portion of the protein is acidic in nature.

Fig. 2. SAXS analysis of the repressor on and off DNA.

a SEC elution profiles for the repressor only (blue), repressor mixed with a 13-bp DNA (orange), and repressor mixed with a 24-bp DNA (red), along with masses calculated from MALS. The masses confirm that the repressor is a monomer in solution both on and off DNA. b P(r) functions calculated from the experimental data that is shown in panel c, with the traces color-coded as in a. The distance r, in Angstroms (Å), on the x-axis where the P(r) function approaches zero intensity represents the maximal dimension (D_max) for each sample. c Experimental SAXS curves are shown in gray/black along with the theoretical scattering profiles, fit-residuals, and χ² values for the atomistic models of the protein:24-bp DNA (red trace), protein:13-bp DNA (orange trace), and protein only (blue trace) samples shown in panel d. The Guinier plots (inset) were used to calculate the radius of gyration values (R_g) for each sample. d Atomistic models derived from SAXS-fitting, with protein colored in gray and DNA in yellow. The percentages define how much each model contributes to the theoretical scattering profiles shown in panel c. Top panel: the protein:24-mer complex provides an excellent fit to the experimental data (χ² = 1.6), confirming that the structure observed in the crystal matches the conformation of the complex in solution. The long loop at the N-terminus represents the first 14 residues of the protein as well as the His-tag that was disordered in the crystal structure. Middle panel: the protein:13-bp sample is best described by a mixture of free protein and free DNA in solution, in agreement with the SEC trace in panel a. Bottom panel: The free protein matches the SAXS data when it contains solvent-exposed N- and C-termini.

Fig. 3. Repressor:DNA interactions.

a Double-stranded DNA sequence present in the crystal structure is shown, with the consensus sequence underlined. The numbers are present to identify each nucleotide of the consensus. Residues that contact the DNA are listed, with dashed lines indicating interactions with the DNA bases, while solid lines designate residues that contact the DNA backbone. All polar contacts shown are 3.2 Å or less, and an asterisk indicates contacts that are only observed in the higher resolution selenomethionine structure. Residues are color-coded as in Fig. 1a, and the arrow indictates the direction of transcription. b Interactions between R45, Q46, and W50 of the α3 helix in the HTH domain and DNA bases are shown. c Interactions between the Stoperator domain and DNA bases. At the beginning of this domain, the η1 3₁₀ helix properly positions K75, D78, and K79 to contact bases of the DNA. K81 sits at the base of the α5 helix. R108 in the α6 helix is properly positioned to bind DNA via an interaction (colored red) with D104. In both panels b, c, the protein and DNA are colored coded as in Fig. 1a.

Fig. 4. Residues critical for repressor function.

a TipsytheTRex virus was serially diluted and spotted onto top agar containing cells integrated with a single copy of the pMH94 empty vector, vector plus wild-type (WT) repressor and its endogenous promoter, or the indicated mutants. Spot titers and the efficiency of virus plating (EOP) were calculated as compared to vector-only containing cells. Mutants tested are color-coded to indicate whether they are from the HTH (blue) or Stoperator (orange) domains. Results shown are representative of three independent experiments. b Mutants underlined in panel a were tested for their ability to bind DNA using electrophoretic mobility shift assays. For each gel, fluorescein-labeled DNA was mixed with 0, 0.02, 0.04, 0.08, 0.16, 0.31, 0.63, 1.25, 2.5, or 5 μM protein. The mass of free DNA in the absence of protein (19.0 kDa) is indicated for each gel. A titration of the wild-type repressor (WT Rep) shows a shift in band size indicative of protein:DNA complex formation. The smearing present in the last two lanes represent non-specific complex formation present at high protein concentrations. While the D104A mutant retained the ability to bind DNA, the R108A mutant has lost the ability to form a specific complex with the DNA substrate. All mutants with an asterisk in panel a showed DNA-binding behavior similar to R108A (see Supplementary Fig. 6). This experiment was performed in three independent experiments, with similar results. c, d Both the wild-type (panel c, blue trace) and D104A (panel d, red trace) repressors bind the DNA ligand, with the D104A mutant displaying an ∼3.5-fold weaker DNA-binding affinity as compared to wild-type. Plotted are the mean and standard deviation values calculated from three independent experiments. Source data are provided in the Source Data file.

Fig. 5. DNA binding promotes altered conformational dynamics.

a, b Cross-correlation analysis indicates a combination of positively (blue rods) and negatively (red rods) correlated conformational motions for the apo (a, slate blue) versus DNA-bound (b, yellow) TipsytheTRex repressor structure. All structural representations in a, b were prepared using Pymol 2.4.1 in ribbon view. c A Scree plot illustrates the contribution of individual principal components to overall system variance as a percentage based on MD trajectories generated using the apo repressor structure. y-axis labeling highlights the contribution of major principal components to system variance. Internal plot labeling presents total variance captured with each additional principal component. d, e Structural representations depict the conformational motions captured in Principal Components 1 (PC1, d, pink/cyan) and 2 (PC2, e, green/yellow). f A Scree plot illustrates the contribution of individual principal components to overall DNA-bound system variance as a percentage. g, h Structural representations depict the conformational motions captured in Principal Components 1 (PC1, g, pink/cyan) and 2 (PC2, h, green/yellow). All principal component structural representations are shown in surface view with the direction of motion indicated by an arrow.

Fig. 6. DNA-binding dissociation energies are different for the HTH and Stoperator domains.

Steered molecular dynamics techniques were employed alongside umbrella sampling methods to simulate DNA dissociation from full-length repressor. a Frames extracted along the COM pulling trajectory are superimposed to highlight dissociation path. All structural representations were prepared using Pymol 2.4.1 in ribbon view with frames corresponding to t = 0, 100, 200, 300, 400, 500, and 600 ps colored as red, green, blue, yellow, magenta, cyan, and orange, respectively. The TipsytheTRex repressor protein structure is position restrained, while DNA is sequentially pulled along a defined path by application of a static force vector. The resulting SMD trajectories were then subjected to umbrella sampling techniques to calculate Potential of Mean Force (PMF). The amplitude of a resulting plot of PMF versus distance between protein and DNA centers-of-mass yields an estimate of the dissociation free energy (b). Adequate sampling was confirmed by weighted histogram analysis (c) and error estimation was obtained by bootstrap methods (n = 200). Figure labeling is present to indicate binding dissociation energy value, ΔG.

Fig. 7. Model for transcriptional silencing in cluster A mycobacteriophages.

a The Mycobacterium smegmatis RNA polymerase (PDB 5VI5) is colored as follows: α subunits: pink, β subunit: gray, β′ subunit: cyan, ω subunit: dark purple, σ^A subunit: red, and RNA polymerase-binding protein RBPA: black. The TipsytheTRex repressor is shown in surface view and colored as in Fig. 1a, and it has been rotated forward by ∼90° relative to the orientation observed in Fig. 1. All nucleic acid in the figure is colored green, and the direction of transcription is indicated with an arrow. The repressor may inhibit transcription elongation by either serving as a steric block or by halting the polymerase via protein:protein interactions. b Zoomed in view showing the position of the RNA polymerase β′ insert (colored cyan) relative to the C-terminal region of the Stoperator domain of the repressor (colored orange). The remaining regions of the RNA polymerase are colored in gray for clarity. c Same view as in panel b but with electrostatic potentials shown (red: negative potential, blue: positive potential, white: neutral) for the RNA polymerase β′ insert and C-terminal region of the Stoperator domain of the repressor. Circles are drawn around areas of positive and negative charge for the β′ insert and C-terminal region of the repressor, respectively.