Limits...
Balanced Protein-Water Interactions Improve Properties of Disordered Proteins and Non-Specific Protein Association.

Best RB, Zheng W, Mittal J - J Chem Theory Comput (2014)

Bottom Line: The modification also results in more realistic protein-protein affinities, and average solvation free energies of model compounds which are more consistent with experiment.Most importantly, we show that this scaling is small enough not to affect adversely the stability of the folded state, with only a modest effect on the stability of model peptides forming α-helix and β-sheet structures.The proposed adjustment opens the way to more accurate atomistic simulations of proteins, particularly for intrinsically disordered proteins, protein-protein association, and crowded cellular environments.

View Article: PubMed Central - PubMed

Affiliation: Laboratory of Chemical Physics, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health , Bethesda, Maryland 20892, United States.

ABSTRACT
Some frequently encountered deficiencies in all-atom molecular simulations, such as nonspecific protein-protein interactions being too strong, and unfolded or disordered states being too collapsed, suggest that proteins are insufficiently well solvated in simulations using current state-of-the-art force fields. To address these issues, we make the simplest possible change, by modifying the short-range protein-water pair interactions, and leaving all the water-water and protein-protein parameters unchanged. We find that a modest strengthening of protein-water interactions is sufficient to recover the correct dimensions of intrinsically disordered or unfolded proteins, as determined by direct comparison with small-angle X-ray scattering (SAXS) and Förster resonance energy transfer (FRET) data. The modification also results in more realistic protein-protein affinities, and average solvation free energies of model compounds which are more consistent with experiment. Most importantly, we show that this scaling is small enough not to affect adversely the stability of the folded state, with only a modest effect on the stability of model peptides forming α-helix and β-sheet structures. The proposed adjustment opens the way to more accurate atomistic simulations of proteins, particularly for intrinsically disordered proteins, protein-protein association, and crowded cellular environments.

No MeSH data available.


Related in: MedlinePlus

Stability of folded proteins.We show native distance matrix RMS(dRMS) calculated for all heavy atom pairsin contact in the native crystal structure, for (A) ubiquitin, (B)CspTm, (C) human lysozyme, and (D) spectrin R15 andresidue-averaged root-mean-square fluctuations (RMSF) for each ofthe proteins (E) ubiquitin, (F) CspTm, (G) humanlysozyme, and (H) spectrin R15. Black and red curves correspond respectivelyto results obtained with the Amber ff03w and Amber ff03ws force fieldsin 200 ns simulations at 300 K. The experimental structures are presentedas insets in E–H.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4230380&req=5

fig6: Stability of folded proteins.We show native distance matrix RMS(dRMS) calculated for all heavy atom pairsin contact in the native crystal structure, for (A) ubiquitin, (B)CspTm, (C) human lysozyme, and (D) spectrin R15 andresidue-averaged root-mean-square fluctuations (RMSF) for each ofthe proteins (E) ubiquitin, (F) CspTm, (G) humanlysozyme, and (H) spectrin R15. Black and red curves correspond respectivelyto results obtained with the Amber ff03w and Amber ff03ws force fieldsin 200 ns simulations at 300 K. The experimental structures are presentedas insets in E–H.

Mentions: Lastly,we consider thestability of native proteins: it might be anticipated that increasingthe protein-water interaction strength may destabilize folded states.However, this is a difficult issue to address computationally as computingthe stability of native proteins is rather challenging. Instead, weshow that a more limited condition is satisfied, i.e. that foldedproteins stay close to the native state in simulations starting fromthe experimental structures. We have chosen a set of four proteinsrepresentative of different structural classes so as to obtain a goodcoverage of protein structure space. These are the extensively characterizedubiquitin (α/β), spectrin R15 (all-α), the coldshock protein from Thermotoga maritima, CspTm (all-β), and human lysozyme (α/β) –the last being an example of a slightly larger protein with two distinctα and β domains. For each protein, we have run 200 nssimulations starting from the experimentally determined, folded structure.In all cases, using the Amber03ws model results in all-atom dRMS deviations of ∼0.2 nm from the X-raystructures. These deviations are comparable to, or in some cases (CspTm) slightly better than those obtained with the originalAmber03w. (Figure 6 A-D). To further quantifythe fluctuations on a per-residue basis, we have computed root-mean-squarefluctuations (RMSF) averaged over each residue, after aligning thebackbone to the experimental structure. The RMSF (Figure 6E–H) obtained with Amber03w and Amber03wsis also comparable. Lastly, we have calculated all-atom contact mapswith a 0.6 nm cutoff (i.e., for each pair of residues, we calculatethe fraction of the time for which at least one pair of heavy atomsfrom the two residues is within 0.6 nm). These contact maps, shownin Figure 7, show that almost identical contactsare formed with the original and modified force field; that is veryfew native contacts are broken and few non-native contacts are formed.We have also plotted the residue–residue contacts on the α-carbontrace of each protein, highlighting the contacts formed in the Amberff03w simulations and not in the Amber ff03ws simulations, and viceversa. As can be seen there are generally very few differences. Thelargest changes are for CspTm, as already suggestedby the differences in the dRMS plots;in this case, the ff03ws is in fact closer to the experimental structure.Finally, we have directly calculated backbone amide order parametersfor comparison with those measured by NMR, for the proteins wheredata is available (ubiquitin, human lysozyme).82 The results, shown in Figure 8 indicatethat the Amber ff03w and ff03ws force fields yield similar valuesfor the order parameters in regions of secondary structure, and arein both cases in good agreement with experiment. For ubiquitin, theloops are clearly more flexible in Amber ff03ws than in either Amberff03w or in experiment; in lysozyme, both ff03w and ff03ws exhibitsome excess flexibility in the loops over experiment. One caveat tothis interpretation is that the method we have used to compute orderparameters averages the internal dynamics over the entire simulation,82 whereas the fit to experiment is sensitive onlyto dynamics shorter than the reorientational correlation time of themolecule. The calculated order parameters may therefore be an underestimate,but much longer simulations would be needed to calculate them in theanalogous way to experiment.83


Balanced Protein-Water Interactions Improve Properties of Disordered Proteins and Non-Specific Protein Association.

Best RB, Zheng W, Mittal J - J Chem Theory Comput (2014)

Stability of folded proteins.We show native distance matrix RMS(dRMS) calculated for all heavy atom pairsin contact in the native crystal structure, for (A) ubiquitin, (B)CspTm, (C) human lysozyme, and (D) spectrin R15 andresidue-averaged root-mean-square fluctuations (RMSF) for each ofthe proteins (E) ubiquitin, (F) CspTm, (G) humanlysozyme, and (H) spectrin R15. Black and red curves correspond respectivelyto results obtained with the Amber ff03w and Amber ff03ws force fieldsin 200 ns simulations at 300 K. The experimental structures are presentedas insets in E–H.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4230380&req=5

fig6: Stability of folded proteins.We show native distance matrix RMS(dRMS) calculated for all heavy atom pairsin contact in the native crystal structure, for (A) ubiquitin, (B)CspTm, (C) human lysozyme, and (D) spectrin R15 andresidue-averaged root-mean-square fluctuations (RMSF) for each ofthe proteins (E) ubiquitin, (F) CspTm, (G) humanlysozyme, and (H) spectrin R15. Black and red curves correspond respectivelyto results obtained with the Amber ff03w and Amber ff03ws force fieldsin 200 ns simulations at 300 K. The experimental structures are presentedas insets in E–H.
Mentions: Lastly,we consider thestability of native proteins: it might be anticipated that increasingthe protein-water interaction strength may destabilize folded states.However, this is a difficult issue to address computationally as computingthe stability of native proteins is rather challenging. Instead, weshow that a more limited condition is satisfied, i.e. that foldedproteins stay close to the native state in simulations starting fromthe experimental structures. We have chosen a set of four proteinsrepresentative of different structural classes so as to obtain a goodcoverage of protein structure space. These are the extensively characterizedubiquitin (α/β), spectrin R15 (all-α), the coldshock protein from Thermotoga maritima, CspTm (all-β), and human lysozyme (α/β) –the last being an example of a slightly larger protein with two distinctα and β domains. For each protein, we have run 200 nssimulations starting from the experimentally determined, folded structure.In all cases, using the Amber03ws model results in all-atom dRMS deviations of ∼0.2 nm from the X-raystructures. These deviations are comparable to, or in some cases (CspTm) slightly better than those obtained with the originalAmber03w. (Figure 6 A-D). To further quantifythe fluctuations on a per-residue basis, we have computed root-mean-squarefluctuations (RMSF) averaged over each residue, after aligning thebackbone to the experimental structure. The RMSF (Figure 6E–H) obtained with Amber03w and Amber03wsis also comparable. Lastly, we have calculated all-atom contact mapswith a 0.6 nm cutoff (i.e., for each pair of residues, we calculatethe fraction of the time for which at least one pair of heavy atomsfrom the two residues is within 0.6 nm). These contact maps, shownin Figure 7, show that almost identical contactsare formed with the original and modified force field; that is veryfew native contacts are broken and few non-native contacts are formed.We have also plotted the residue–residue contacts on the α-carbontrace of each protein, highlighting the contacts formed in the Amberff03w simulations and not in the Amber ff03ws simulations, and viceversa. As can be seen there are generally very few differences. Thelargest changes are for CspTm, as already suggestedby the differences in the dRMS plots;in this case, the ff03ws is in fact closer to the experimental structure.Finally, we have directly calculated backbone amide order parametersfor comparison with those measured by NMR, for the proteins wheredata is available (ubiquitin, human lysozyme).82 The results, shown in Figure 8 indicatethat the Amber ff03w and ff03ws force fields yield similar valuesfor the order parameters in regions of secondary structure, and arein both cases in good agreement with experiment. For ubiquitin, theloops are clearly more flexible in Amber ff03ws than in either Amberff03w or in experiment; in lysozyme, both ff03w and ff03ws exhibitsome excess flexibility in the loops over experiment. One caveat tothis interpretation is that the method we have used to compute orderparameters averages the internal dynamics over the entire simulation,82 whereas the fit to experiment is sensitive onlyto dynamics shorter than the reorientational correlation time of themolecule. The calculated order parameters may therefore be an underestimate,but much longer simulations would be needed to calculate them in theanalogous way to experiment.83

Bottom Line: The modification also results in more realistic protein-protein affinities, and average solvation free energies of model compounds which are more consistent with experiment.Most importantly, we show that this scaling is small enough not to affect adversely the stability of the folded state, with only a modest effect on the stability of model peptides forming α-helix and β-sheet structures.The proposed adjustment opens the way to more accurate atomistic simulations of proteins, particularly for intrinsically disordered proteins, protein-protein association, and crowded cellular environments.

View Article: PubMed Central - PubMed

Affiliation: Laboratory of Chemical Physics, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health , Bethesda, Maryland 20892, United States.

ABSTRACT
Some frequently encountered deficiencies in all-atom molecular simulations, such as nonspecific protein-protein interactions being too strong, and unfolded or disordered states being too collapsed, suggest that proteins are insufficiently well solvated in simulations using current state-of-the-art force fields. To address these issues, we make the simplest possible change, by modifying the short-range protein-water pair interactions, and leaving all the water-water and protein-protein parameters unchanged. We find that a modest strengthening of protein-water interactions is sufficient to recover the correct dimensions of intrinsically disordered or unfolded proteins, as determined by direct comparison with small-angle X-ray scattering (SAXS) and Förster resonance energy transfer (FRET) data. The modification also results in more realistic protein-protein affinities, and average solvation free energies of model compounds which are more consistent with experiment. Most importantly, we show that this scaling is small enough not to affect adversely the stability of the folded state, with only a modest effect on the stability of model peptides forming α-helix and β-sheet structures. The proposed adjustment opens the way to more accurate atomistic simulations of proteins, particularly for intrinsically disordered proteins, protein-protein association, and crowded cellular environments.

No MeSH data available.


Related in: MedlinePlus