Fig 1.
The Functional Cycle of Hsp70 Chaperones.
The functional cycle of Hsp70 chaperones. The main steps of the allosteric cycle include the nucleotide exchange in the closed, ADP-bound form (pdb id 2KHO); the formation of partially undocked ATP/substrate-bound intermediate; substrate release and formation of the domain-docked ATP-bound form (pdb id 4B9Q, 4JNE); substrate binding and partial domain undocking in the ATP/substrate-bound state; ATP hydrolysis and stabilization of the domain-undocked, ADP-bound form. The structures are shown in a surface representation and the main structural elements are annotated. The NBD subdomains are colored as follows: IA (in blue), IB (in red), IIA (in green), IIB (in cyan), the inter-domain linker (in black), SBD-α (in magenta), and SBD-β (in orange). The inter-domain interfaces NBD/SBD-β, NBD/SBD-α and SBD-β/SBD-α form allosteric hotspots of the Hsp70 functional cycle that are modulated through binding of nucleotides and substrates. The Pymol program was used for rendering protein structures (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, and LLC).
Fig 2.
The Structures and Domain Organization of Functional States in DnaK and Sse1p Chaperones.
A solution structure of an ADP-bound DnaK (pdb id 2KHO) (A) and the crystal structure of an ATP-bound DnaK (pdb id 4B9Q) (B). The structures are shown in a ribbon representation and main structural elements are annotated. The NBD subdomains are colored as follows: IA (in blue), IB (in red), IIA (in green), IIB (in cyan), the inter-domain linker (in black), SBD-α (in magenta), and SBD-β (in orange). (C) The crystal structure of the yeast Hsp110 (Sse1p) in a complex with ATP (pdb id 2QXL). (D) The crystal structure of Sse1p in a complex with the NBD of hHsp70 (pdb id 3D2F). The structures are shown in a ribbon representation and main structural elements are annotated and colored as in DnaK. The Sse1p insert in Sse1p-ATP structure is shown in yellow (C) and the NBD of hHsp70 in the Sse1p-Hsp70 complex is shown in pink (D).
Fig 3.
Conformational Dynamics of DnaK and Sse1p Functional States.
The computed B-factors obtained from 500 ns MD simulations of the solution structure of an ADP-bound DnaK (pdb id 2KHO) (A); the crystal structure of an ATP-bound DnaK (pdb id 4B9Q) (B); the crystal structure of a Sse1p-ATP (pdb id 2QXL) (C); and the crystal structure of Sse1p in a complex with the NBD of hHsp70 (pdb id 3D2F) (D). The thermal fluctuations are shown only for DnaK and Sse1p residues (B-factors of the hHsp70-NBD counterpart of Sse1p are omitted for clarity and uniformity of presentation). Equilibrium residue fluctuations are annotated and colored according to the adopted coloring scheme of the chaperone subdomains: IA (in blue), IB (in red), IIA (in green), IIB (in cyan), the inter-domain linker (in black), SBD-α (in magenta), and SBD-β (in orange).
Fig 4.
Analysis of Essential Motions in the Closed and Open DnaK Forms.
Functional dynamics maps and cross-correlation matrices of residue fluctuations for the ADP-bound DnaK structure (A, B) and ATP-bound DnaK form (C, D). Conformational dynamics profiles were computed by averaging protein motions in the space of three lowest frequency modes. The color gradient from blue to red indicates the decreasing structural rigidity of the protein residues. PCA computations are based on the Cα atoms. The axes denote Cα atoms of the protein residues in sequential order. Cross-correlations of residue-based fluctuations vary between +1 (fully correlated motion; fluctuation vectors in the same direction, colored in red) and -1 (fully anti-correlated motions; fluctuation vectors in the same direction, colored in blue). The residue ranges corresponding to the NBD, SBD-α, and SBD-β regions are highlighted.
Fig 5.
Analysis of Essential Motions in the Sse1p Structures.
Functional dynamics maps and cross-correlation matrices of residue fluctuations for the Sse1p-ATP structure (A, B) and Sse1p-hHsp70 structure (C, D). Conformational dynamics profiles of Sse1p were computed by averaging protein motions in the space of three lowest frequency modes. In (C, D) the plots depict the dynamics profiles for Sse1p structure only. The corresponding profile of the hHsp70-NBD counterpart of Sse1p in the complex is omitted for clarity of presentation. The color gradient from blue to red indicates the decreasing structural rigidity of the protein residues. PCA computations are based on the Cα atoms.
Fig 6.
The Protein Stability Analysis of DnaK Structures.
Protein stability changes ΔΔG are computed using a systematic alanine scanning. The protocol involved modification of the protein residues to alanine and computing the effect of each mutation on protein stability using FoldX (A,B) and DUET (C,D) methods respectively. The profiles are annotated using residue numbering in the solution structure of the ADP-bound DnaK (pdb id 2KHO) for (A, C) and the crystal structure of an ATP-bound DnaK (pdb id 4B9Q) for (B, D). The profiles are shown as bars colored according to the adopted scheme: IA (in blue), IB (in red), IIA (in green), IIB (in cyan), the inter-domain linker (in black), SBD-α (in magenta), and SBD-β (in orange). The inset shows protein stability changes of the SBD-β residues (orange-colored lines with marron-colored filled squares).
Fig 7.
The Protein Stability Analysis of Sse1p Structures.
Protein stability changes ΔΔG are computed using a systematic alanine scanning. The protocol involved modification of the protein residues to alanine and computing the effect of each mutation on protein stability using FoldX (A,B) and DUET (C,D) methods respectively. The profiles are annotated using residue numbering in the crystal structure of ATP-Sse1p (pdb id 2QXL) for (A,C) and the crystal structure of ATP-Sse1p in a complex with the NBD of hHsp70 (pdb id 3D2F) for (B,D). The profiles are shown as bars colored according to the adopted scheme as in Fig 6.
Fig 8.
Free Energy Calculations of the Substrate Binding with the DnaK Structures.
Binding free energies and alanine scanning of the NRLLLTG substrate-interacting residues with the ADP-DnaK (A) and ATP-DnaK (C). Computational alanine scanning employed MM-GBSA calculations to evaluate the effect of alanine mutations for the substrate binding site residues on binding affinity using MD trajectories of the nucleotide/substrate-bound WT structures. The protocol involved a systematic modification of the inhibitor-interacting residues to alanine by eliminating side-chain atoms beyond Cβ, and measuring the effect of each mutation on binding affinity. The close-up of the NRLLLTG substrate binding mode and interacting residues is shown for ADP-DnaK (B) and ATP-DnaK (D). The peptide is shown in atom-colored sticks and annotated. The substrate binding site residues from the SBD-β subdomain are shown in orange sticks and annotated.
Fig 9.
The Residue Depth Profiling of DnaK Structures.
The RD profiles are shown for ADP-DnaK (A), ATP-DnaK (B), and ATP-Sse1p structures (C, D). The RD values were obtained by averaging computations over the respective equilibrium ensembles. The profiles are annotated using residue numbering in the solution structure of the ADP-bound DnaK, (pdb id 2KHO) (A) the crystal structure of an ATP-bound DnaK, (pdb id 4B9Q) (B), the crystal structure of ATP-Sse1p (pdb id 2QXL) (C) and the crystal structure of ATP-Sse1p in a complex with the NBD of hHsp70 (pdb id 3D2F) (D). The profiles are shown as bars colored according to the adopted scheme: IA (in blue), IB (in red), IIA (in green), IIB (in cyan), the inter-domain linker (in black), SBD-α (in magenta), and SBD-β (in orange). The inset in each panel shows protein stability changes of the SBD-β residues (orange-colored lines with marron-colored filled squares).
Fig 10.
The Residue Depth Profiles in the DnaK States: Structural Mapping and Comparison with the HDX Experiments.
The computed RD profiles are compared with the HDX experiments, showing nucleotide-induced protection of functional regions. The effect of nucleotide binding is evaluated using differential plots of the RD profiles between the ATP-DnaK and ADP-DnaK (A). The profiles are shown as bars and colored according to the adopted scheme as in Fig 8. (B) The differential RD values are shown only for the SBD-β subdomain (orange bars) and RD values for residues with the experimentally known HDX peak intensity ratios between ATP-DnaK and ADP- DnaK are highlighted in marron bars. (C) The experimental values of the HDX peak intensity ratios between ATP-DnaK and ADP-DnaK [27]. Residues that are more protected in the ATP-DnaK (HDX peak ratio > 1.1) are shown in blue bars, and residues that are more protected in the ADP-DnaK form (HDX peak intensity ratio < 1.0) are shown in red bars. This annotation is based on the HDX peak assignments as prescribed in [27]. (D). Structural mapping of the ATP-DnaK protected residues depicted in blue spheres and annotated. (E) Structural mapping of the ADP-DnaK protected residues shown in red spheres and annotated. The SBD-β subdomain is shown in orange ribbons. The substrate binding loops L1,2 loop (residues 404–406), L3,4 loop (residues 428–434), and L5,6 loop (residues 458–473) are indicated.
Fig 11.
Force Constant and Network Centrality Profiles of DnaK Forms.
Residue-based force constant profiles and network centrality distributions for an ADP-bound DnaK form (A, B) and ATP-bound Dnak state (C, D). The profiles are annotated and colored according to the adopted scheme: IA (in blue), IB (in red), IIA (in green), IIB (in cyan), the inter-domain linker (in black), SBD-α (in magenta), and SBD-β (in orange). The residue-based dynamic profiles are annotated using the residue numbering in the solution structure of an ADP-bound DnaK, (pdb id 2KHO) and the crystal structure of an ATP-bound DnaK, (pdb id 4B9Q). The peaks of the force constant and centrality profiles corresponding to functionally important residues are indicated by arrows and annotated.
Fig 12.
Analysis of High Centrality Residues in DnaK States.
Structural mapping of common peaks in the force constant and network centrality distributions onto DnaK conformations (A, B). The structures are shown in a ribbon representation and main structural elements, including subdomains and functional residues are annotated and colored according to the adopted scheme. The functional residues of high centrality are shown in spheres and colored according to their respective subdomains. Structural positions of high centrality functional residues are indicated by arrows. The probability distributions of residue centrality in the ADP-bound (C) and ATP-bound DnaK forms (D). These profiles were obtained from MD trajectories by averaging computations of residue centrality over all protein residues in the conformational ensembles.
Fig 13.
Force Constant and Network Centrality Profiles of Sse1p Structures.
Residue-based force constant profiles and network centrality distributions for Sse1p-ATP (A, B), and Sse1p in a complex with the NBD of hHsp70 (C, D). The profiles are shown only for Sse1p residues. The respective profiles for the hHsp70-NBD counterpart of Sse1p in the complex are omitted for clarity of presentation. The profiles are annotated and colored according to the adopted scheme. The residue-based dynamic profiles are annotated using the residue numbering in the crystal structure of Sse1p-ATP conformation (pdb id 2QXL) and the crystal structure of Sse1p-Hsp70 complex (pdb id 3D2F). The profile peaks corresponding to functionally important residues are indicated by arrows and annotated.
Fig 14.
Analysis of High Centrality Residues in Sse1p Structures.
Structural mapping of common peaks in the force constant and network centrality distributions onto Sse1p conformations (A, B). The structures are shown in a ribbon representation and main structural elements, including subdomains and functional residues are annotated and colored according to the adopted scheme. The functional residues of high centrality are shown in spheres and colored according to their respective subdomains. Structural positions of high centrality functional residues are indicated by arrows. The probability distributions of residue centrality in the Sse1p-ATP (C) and Sse1p in a complex with the NBD of hHsp70 (D). These profiles were obtained from MD trajectories by averaging computations of residue centrality over all protein residues in the conformational ensembles.
Fig 15.
Network Analysis of Functional Effects in Sse1p Mutants.
Correlations between residue centrality and different functional effects caused by clusters of mutations in Sse1p. (A, B) The relationship between residue centrality in the Sse1p complex with Hsp70 (pdb id 3D2F) and rates of the nucleotide exchange induced by Sse1p mutants. (C, D) The relationship between residue centrality in the Sse1p complex and binding affinities of Sse1p mutants measured in [53]. The clusters of mutations are annotated as in the original experimental study [53]: Sse1-1 (K69M); Sse1-2 (N572Y,E575A); Sse1-3 (A280T,N281A); Sse1-4 (T365V, N367S); Sse1-5 (F392A, F394A); Sse1-6 (D396A); Sse1-7 (L489A, H490A); Sse1-8 (E554A, M557S, L558S); Sse1-9 (L433A, N434P); Sse1-10 (F439L, M441A). To account for clusters of mutations, used in the experiments, we computed the average betweenness value over all residues in a given cluster.