Conceived and designed the experiments: SL BJG AAG JAM. Performed the experiments: SL BJG. Analyzed the data: SL BJG. Contributed reagents/materials/analysis tools: BJG AAG GHG. Wrote the paper: SL BJG AAG JAM.
The authors have declared that no competing interests exist.
Ras proteins regulate signaling cascades crucial for cell proliferation and differentiation by switching between GTP- and GDP-bound conformations. Distinct Ras isoforms have unique physiological functions with individual isoforms associated with different cancers and developmental diseases. Given the small structural differences among isoforms and mutants, it is currently unclear how these functional differences and aberrant properties arise. Here we investigate whether the subtle differences among isoforms and mutants are associated with detectable dynamical differences. Extensive molecular dynamics simulations reveal that wild-type K-Ras and mutant H-Ras A59G are intrinsically more dynamic than wild-type H-Ras. The crucial switch 1 and switch 2 regions along with loop 3, helix 3, and loop 7 contribute to this enhanced flexibility. Removing the gamma-phosphate of the bound GTP from the structure of A59G led to a spontaneous GTP-to-GDP conformational transition in a 20-ns unbiased simulation. The switch 1 and 2 regions exhibit enhanced flexibility and correlated motion when compared to non-transitioning wild-type H-Ras over a similar timeframe. Correlated motions between loop 3 and helix 5 of wild-type H-Ras are absent in the mutant A59G reflecting the enhanced dynamics of the loop 3 region. Taken together with earlier findings, these results suggest the existence of a lower energetic barrier between GTP and GDP states of the mutant. Molecular dynamics simulations combined with principal component analysis of available Ras crystallographic structures can be used to discriminate ligand- and sequence-based dynamic perturbations with potential functional implications. Furthermore, the identification of specific conformations associated with distinct Ras isoforms and mutants provides useful information for efforts that attempt to selectively interfere with the aberrant functions of these species.
The proto-oncogene Ras mediates signaling pathways controlling cell proliferation and development by cycling between active and inactive conformational states. Mutations that affect the ability to switch between states are associated with over 25% of human tumors. However, despite much effort, details of how these mutations affect the fidelity of activating conformational transitions remain unclear. Here we employ extensive molecular dynamics simulations combined with principal component analysis to investigate whether the subtle differences among functionally distinct isoforms and oncogenic mutants are associated with detectable dynamical differences. Our results reveal that wild-type K-Ras, the most prevalent isoform in a number of cancers, and mutant H-Ras A59G are intrinsically more dynamic than wild-type H-Ras. Furthermore, we have observed the first spontaneous GTP-to-GDP transition of H-Ras A59G during unbiased molecular dynamics simulation. These results indicate that key changes in sequence can lead to different dynamic properties that may be relevant for the unique physiological and aberrant functions of Ras isoforms and mutants. Furthermore, the current results shed further light on the conformational transition mechanism of this important molecular switch.
Ras proteins couple cell-surface receptors to intracellular signaling cascades involved in cell proliferation, differentiation and development. Signal propagation through Ras is mediated by a regulated GTPase cycle that leads to active and inactive conformations with distinct affinity for downstream effectors. Regulatory proteins including guanine nucleotide exchange factors (GEFs) and GTPase-activating proteins (GAPs) stimulate the intrinsically slow GTPase cycle promoting proper signal flow. Ras mutants with an impaired GTPase activity that are insensitive to the action of GAPs and GEFs result in prolonged downstream signaling associated with oncogenic cell growth in diverse human cancers and leukemia
Ras genes encode multiple isoforms of which H-, N-, and K-Ras are the most abundant. K-Ras can be found as two splice variants termed K-Ras4A and K-Ras4B. Although these isoforms share a high degree of similarity (over 90% sequence identity), their physiological functions are not necessarily equivalent
The unique functions of Ras isoforms are mediated by their preferences for different binding partners
(A) The sequence of H-Ras and K-Ras (amino acid differences between K-Ras and H-Ras are indicated in cyan). Residues 12, 13, and 61, whose substitutions are associated with a large number of cancers, are colored magenta. Other residues where mutations have been reported associated with various cancers and developmental diseases, are colored in brown. Secondary structure content is indicated on top of amino acid positions with α-helices (black) and β-sheets (gray) (B) The Ras catalytic domain is composed of a six-stranded central β-sheet surrounded by five α-helices. Nucleotide binds to a conserved phosphate-binding loop (P-loop comprising residues 10–17, green) and two switch loop regions (switch 1, residues 25–40, blue, and switch 2, residues 57–75, red). Oncogenic mutations often occur at these regions, particularly residues 12, 13 and 61
Small changes in sequence can lead to different dynamic properties, which are manifested as only subtle changes in the average structure observed by x-ray crystallography. Recently we reported the observation of spontaneous nucleotide-dependent transition during unbiased molecular dynamics (MD) simulation of the oncogenically active H-Ras G12V variant
Previous classical and accelerated MD simulations of H-Ras have successfully characterized its dynamic features and proposed a reaction path for the ligand-associated conformational changes
Here we employ multi-copy MD simulations to investigate whether the subtle differences between H- and K-Ras isoforms are associated with detectable dynamical differences that might have potential functional consequences. We first conducted an expanded bioinformatic analysis of available Ras crystallographic structures and found that K-Ras has similar conformational features to the H-Ras A59G mutant. We further performed MD simulations on active, GTP-bound wild-type K-Ras, H-Ras, and H-Ras A59G. Indeed, we observed that both the wild-type K-Ras and H-Ras A59G variant are similarly more dynamic than wild-type H-Ras. However, wild-type K-Ras also has dynamic features that are similar to those of wild-type H-Ras, hence wild-type functions are preserved. The different dynamic features between wild-type K- and H-Ras may also provide clues on distinct preferences for binding proteins and hence subsequent downstream signaling functions. Another major contribution of this study is the first report on the spontaneous GTP-to-GDP transition of the H-Ras A59G variant. The H-Ras A59G variant crystallized with a GTP-analog, when simulated with a GDP, is capable of spontaneously adopting GDP-bound conformations within 20 ns. The atomic details of this transition are important for understanding Ras signaling and function.
We first examined the structure of K-Ras in relation to other available Ras experimental structures. This analysis revealed a similarity to a A59G H-Ras variant. We then performed multiple MD simulations on wild-type K-, H-Ras, and H-Ras A59G. The Ras isoforms exhibited differences in their active, GTP-bound conformational dynamics. H-Ras A59G variant simulated with GDP was able to achieve a spontaneous GTP-to-GDP transition.
To investigate the relationship of K-Ras to other available Ras structures, we compiled a crystallographic ensemble comprising 51 chains from the 47 unique Ras structures in the RCSB PDB
Structures are projected onto the first three principal components (PC) with the largest variances and colored based on the dominant grouping obtained from hierarchical clustering of the projected structures. Representative structures corresponding to distinct bound-nucleotides, wild-type or mutant states, and different isoforms are labeled with their PDB IDs.
While the first principal component revealed distinct clusters of Ras conformations corresponding to different bound-nucleotides (
To further probe the conformational dynamics of the different Ras isoforms we performed multi-copy molecular dynamics simulations on the following systems: wild-type K-Ras, wild-type H-Ras, and H-Ras A59G. Although the starting wild-type K-Ras and the wild-type H-Ras structures are similar in overall conformation (Cα RMSD is 1.03 Å), the active site of wild-type K-Ras is more similar to that of H-Ras A59G (Cα RMSD is 1.16 Å) than wild-type H-Ras (Cα RMSD is 1.74 Å) (
Comparison of the wild-type K-Ras and H-Ras trajectories in the GTP-bound form revealed variations in their dynamic behaviors, consistent with previous MD simulations of homology-built K-Ras and wild-type H-Ras in the nucleotide-free state
The grey and green points represent crystallographic and MD conformers respectively.
The standard deviations about the average are shown in gray.
To identify regions undergoing correlated motions we analyzed dynamic cross-correlation maps for the three sets of trajectories. The correlated motion between loop 3 and α5 helix is absent in H-Ras A59G but present in the wild-type K- and H-Ras (
Only correlation coefficients with an absolute value greater than 0.25 are displayed. Motion occurring along the same direction is represented by positive correlation (blue), whereas anti-correlated motion occurring along the opposite direction is represented by negative correlation (red).
Residue Q61 of switch 2 is essential for GTP hydrolysis with its side chain involved in orienting and activating a nucleophilic water molecule
Y64 is more distant to the γ-phosphate in both wild-type K-Ras and H-Ras A59G than in wild-type H-Ras (
The switch 2 of 15-ns K-Ras conformation resembles the switch 2 in the wild-type H-Ras crystallographic conformation.
The crystallographic structure of H-Ras A59G, on which the current simulations are based, was proposed to represent the conformation of wild-type H-Ras following β/γ-phosphate bond breakage but before γ-phosphate dissociation
We note that only one of our three multi-copy simulations managed to reach the GDP-cluster of the Ras crystallographic ensemble within 20-ns. Comparing the residue-wise flexibility from this simulation to the other two sets, indicates that the GTP-to-GDP transition involves higher RMSF at the loop 2, loop 4, and α2 regions, as well as lower RMSF at the loop 3 region (
(A) set 1, (B) 2, and (C) 3. Within 20-ns, only the set 2 simulation managed to reach the GDP-bound cluster of the Ras crystallographic ensemble.
When the MD-derived conformers are projected onto the first two principal components (PC) obtained from analyzing the Ras crystallographic ensemble, the GDP-bound H-Ras A59G MD conformers sample the GDP cluster of crystallographic conformers (PC 1: 15 to 20, PC 2: −5 to 0) (
(A) The projection of GDP-bound H-Ras A59G MD conformers onto the first two principal components of the crystallographic ensemble. (B) The RMSDs of H-Ras A59G MD conformers with respect to the GTP-bound H-Ras A59G crystal structure (PDB 1LF0, black) and the GDP-bound H-Ras A59G crystal structure (PDB 1LF5, gray). (C) The RMSD of switch 1 (blue) and switch 2 (red) during the MD simulation.
(A) The dynamics of secondary structures of the GDP-bound H-Ras A59G MD conformers. (B) The distances between Y32-Y40 (black), Y32-β phosphate (gray). (C) The distances between G60 and β-phosphate (black), E37-R68 (blue), E37-Y64 (yellow), E37-Y71 (green), E37-Q61 (brown) between 17.5 and 18.5 ns of MD simulation.
The side chain of Y32 lies across the nucleotide binding pocket in GTP-bound conformations but is displaced away from the binding pocket in GDP-bound H-Ras A59G and wild-type H-Ras crystal structures
The MD conformers at different time frames also indicated interesting differences in their torsional angles with respect to the GTP-analog-bound versus the GDP-bound H-Ras A59G crystal conformers (
Combining the current results with previous reports
Oncogenic mutations may interfere with the intrinsic ability of Ras to hydrolyze GTP or lead to insensitivity to the action of GAPs. The loss or deregulation of intrinsic GTP hydrolysis seems to be sufficient for causing diseases
We speculate that the observed differences in dynamics may be related to the measured differences in the intrinsic GTPase activity of the three proteins. The intrinsic GTP hydrolysis rate of K-Ras (1.2×10−4 s−1
We have performed multiple unbiased MD simulation on wild-type K-Ras, H-Ras A59G and wild-type H-Ras. Results from these simulations support the observation that the active, GTP-bound wild-type K-Ras and H-Ras A59G variant are more dynamic than wild-type H-Ras. We observed spontaneous GTP-to-GDP transition during an unbiased MD simulation of H-Ras A59G. The approach of multivariate clustering of crystal structure conformations to reveal differences due to ligands and sequences highlights intermediate conformers, such as the H-Ras G12V
The Bio3D package
Systems for molecular dynamics simulations were prepared from high-resolution crystal structures of wild-type K-Ras, wild-type H-Ras, and H-Ras A59G variant (PDB: 2PMX, 1QRA, and 1LF0 respectively). Each system was simulated with Mg2+GDP and Mg2+GTP. All simulations were performed with the AMBER 10 package
Active site similarities of H-Ras (blue), K-Ras (black) and H-Ras A59G (red). The GTP nucleotide (yellow), switch 1 (residues 25–40) and switch 2 (residues 57–75) regions are displayed from the PDB entries 1QRA, 2PMX, and 1LF0 corresponding to H-, K- and H-Ras A59G respectively.
(0.39 MB TIF)
RMSD values of conformers obtained from three independent MD simulations with respect to Ras crystal structures. (A) The RMSD of wild-type K-Ras MD conformers with respect to the wild-type H-Ras (black) and H-Ras A59G (gray) crystal conformer. (B) The RMSD of H-Ras A59G MD conformers with respect to the wild-type H-Ras (black) and the wild-type K-Ras (gray) crystal conformer.
(0.34 MB TIF)
Root mean square deviations (RMSD) of switch 1 (residues 25–40, blue), switch 2 (residues 57–75, red), switch 3 (residues 47–49, 161–165, green) of H-Ras, K-Ras, and H-Ras A59G in three sets of MD trajectories of (A) wild-type H-Ras, (B) wild-type K-Ras, (C) H-Ras A59G.
(0.73 MB TIF)
Pseudo alpha Carbon torsion angle differences in MD conformers at different time point with respect to crystal structure of (A) GTP-analog-bound, (B) GDP-bound H-Ras A59G crystal conformers. Absolute values of the difference are plotted with secondary structures schematically depicted in black for helices and grey for strands.
(0.12 MB TIF)
The DCCM of GDP-bound H-Ras A59G MD simulation that achieved a spontaneous GTP-to-GDP transition.
(0.11 MB TIF)
The GTP-to-GDP transition observed in the MD trajectory of H-Ras A59G. The N-terminal (residue 65–67) of α 2 helix unwinds. Important residues 37 (pink), 60 (black), 61 (red-brown), 64 (orange), 66 (magenta, to aid unwinding visualization), 68 (blue), and 71 (green), are highlighted in CPK style.URL:
(1.72 MB AVI)
We thank the National Science Foundation (NSF) Supercomputer Centers, the Center for Theoretical Biological Physics (CTBP), the Cambridge High Performance Computing Service (UK) and National Biomedical Computation Resource for computational resources.