Figure 1.
TRF-DNA complex structure and comparison of the TRF1/TRF2 DBD domains.
(A) A schematic representation of the system used for the calculation of the binding free energy profiles. The DNA dodecamer (5′-GGTTAGGGTTAG-3′) is aligned with the z axis, and the xy-distance between the centers of mass of TRF and DNA serves as a convenient reaction coordinate describing the binding process (for a precise definition of the reaction coordinate, see Methods and Fig. S1). The C-terminal recognition helix interacting with the DNA major groove is shown in green and the N-terminal linker binding within the minor groove is in purple. (B) Distribution of different types of amino acid residues in the TRF2 structure: hydrophobic residues are depicted in white, hydrophilic in green, basic in blue and acidic in red. (C) Differences in amino acid sequence between TRF1 and TRF2 mapped onto the TRF2 structure in a color coded manner: identical amino acids are marked in blue, green denotes conservative substitutions (little change) and red non-conservative substitutions (significant change). (D) Alignment of TRF1 and TRF2 sequences. Color-coding is consistent with panel C. The purple and green highlights denote the N-terminal linker and the C-terminal helix, respectively.
Figure 2.
Free energy profiles for binding of TRF1 and TRF2 to telomeric DNA.
Dashed green lines show the boundaries between the three defined bound states and the unbound state (for the corresponding binding free energies, see Table 1).
Table 1.
Standard binding free energies of the TRF DBD domains to telomeric DNA (kcal/mol).
Figure 3.
Direct and water-mediated hydrogen bonds between the TRF proteins and DNA bases.
The pattern of direct and water-mediated hydrogen bonds between the TRF proteins and DNA bases. Only interactions between amino acid residues and nucleic bases are considered, as these base-specific contacts are potentially critical for sequence recognition. Filled circles at individual bases are scaled to reflect the probability that two residues are connected through a direct (orange) or water-mediated (cyan) hydrogen bond (for numeric probability values and hydrogen bond lifetimes, see Table S1 in File S1 and Table S5 in File S1). All contacts that are made with probability 0.05 are included. Grey percentage bars show the probability for a given base to be involved, at any given time, in either direct or water-mediated hydrogen bonds with the protein.
Figure 4.
Detailed view of the interface between TRFs and DNA in the sequence-specific complex.
Four amino acid residues important for recognition of telomeric dsDNA are shown explicitly. See legend to Fig. 1A for base color-coding.
Figure 5.
Protein entropy changes upon DNA binding.
Conformational entropy per atom for individual protein residues in the tightly-bound complex with telomeric DNA (narrow bars) and in the isolated state (wide bars, colored according to amino acid type), estimated using the quasi-harmonic approximation. Only the regions interacting directly with DNA in the tightly-bound complex are presented: the N-terminal linker (purple) and the C-terminal helix (green).
Figure 6.
Spatial probability distributions for the critical residues of the TRF1 C-terminal helix.
High-probability isosurfaces indicate conformations available to the recognition residues in the unbound (A) and tightly-bound (B) state. Distributions for Lys-421, Asp-422 and Arg-425 are plotted in yellow, blue and orange, respectively.
Figure 7.
Hydrogen bond profiles for individual TRF residues.
The probability of hydrogen bond formation between individual protein residues and DNA as a function of the distance from the DNA axis. Only the residues for which the probability exceeds 0.2 are presented. The protein structure is subdivided into three separate regions: the N-terminal linker (top), the middle region (middle) and the C-terminal helix (bottom).
Figure 8.
Interaction energy profiles for individual protein residues.
Interaction energy (computed as a sum of electrostatic and van der Waals contributions) between the charged amino acid residues and their surroundings (DNA and solvent combined) as a function of the distance between the protein and the DNA axis. The protein structure is subdivided into three separate regions: the N-terminal linker (top), the middle region (middle) and the C-terminal helix (bottom). For a full set of amino acid interaction energies, see Fig. S7.
Figure 9.
Solvent accessible surface of TRFs as a function of their distance from the DNA axis.
Hydrophobic (top), hydrophilic (middle) and overall (bottom) solvent accessible surface area of TRF1 and TRF2 as a function of the xy-distance between the protein and the DNA dodecamer.
Figure 10.
Orientations of TRF1/TRF2 DBDs with respect to DNA as a function of xy-distance.
Two-dimensional free energy surfaces for the xy-distance and the three rigid-body rotation angles defining the orientation of TRF1 and TRF2 DBD domains with respect to the DNA dodecamer (aligned with the z axis). To calculate the angles, we first obtained a rotation matrix which gives the best-fit of the instantaneous protein structure to the initial one in the tightly-bound complex with DNA. The Rotation matrix was then expressed as a product of three matrices representing rotations about the x, y and z axis and the rotation angles were derived from these matrices.