Limits...
High temperature unfolding simulations of the TRPZ1 peptide.

Settanni G, Fersht AR - Biophys. J. (2008)

Bottom Line: But, the speed of the folding process is mainly determined by the transition from the completely unfolded state to the intermediate and specifically by the closure of the hairpin loop driven by formation of two native backbone hydrogen bonds and hydrophobic contacts between tryptophan residues.The temperature dependence of the unfolding time provides an estimate of the unfolding activation enthalpy that is in agreement with experiments.The unfolding time extrapolated to room temperature is in agreement with the experimental data as well, thus providing a further validation to the analysis reported here.

View Article: PubMed Central - PubMed

Affiliation: Centre for Protein Engineering, Cambridge, United Kingdom. gs@mrc-lmb.cam.ac.uk

ABSTRACT
We report high temperature molecular dynamics simulations of the unfolding of the TRPZ1 peptide using an explicit model for the solvent. The system has been simulated for a total of 6 mus with 100-ns minimal continuous stretches of trajectory. The populated states along the simulations are identified by monitoring multiple observables, probing both the structure and the flexibility of the conformations. Several unfolding and refolding transition pathways are sampled and analyzed. The unfolding process of the peptide occurs in two steps because of the accumulation of a metastable on-pathway intermediate state stabilized by two native backbone hydrogen bonds assisted by nonnative hydrophobic interactions between the tryptophan side chains. Analysis of the un/folding kinetics and classical commitment probability calculations on the conformations extracted from the transition pathways show that the rate-limiting step for unfolding is the disruption of the ordered native hydrophobic packing (Trp-zip motif) leading from the native to the intermediate state. But, the speed of the folding process is mainly determined by the transition from the completely unfolded state to the intermediate and specifically by the closure of the hairpin loop driven by formation of two native backbone hydrogen bonds and hydrophobic contacts between tryptophan residues. The temperature dependence of the unfolding time provides an estimate of the unfolding activation enthalpy that is in agreement with experiments. The unfolding time extrapolated to room temperature is in agreement with the experimental data as well, thus providing a further validation to the analysis reported here.

Show MeSH

Related in: MedlinePlus

Kinetic network representation of the sampled conformational space of TRPZ1 at 450 K. Each circle represents a cluster obtained using the leader algorithm based on a DRMS metric between conformations. The size of the circle is proportional to the number of structures in the cluster. Pair of clusters are connected by lines when a transition has been sampled along the simulations. Clusters with similar connectivity pattern are placed close together in the plot. The large cluster on the top of the figure and the connected ancillary clusters within the blue boundary corresponds to the native state N. The large set of small clusters evenly distributed on the left of the figure within the green boundary corresponds to the D state. The couple of sets of clusters below the Native state and within the red boundary corresponds to the I state. The I state provides the more frequent connection between the N and D states, as revealed by the density of connecting lines. For graphical reason, small clusters were lumped together according to the dRMSD between cluster centers, so that, at the end of the procedure, all the clusters contain more than 20 conformations. Please note that the VIR was computed using unlumped clusters. The figure was prepared using VISONE (http://visone.info), similarly to what was done in Rao and Caflisch (48).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2480669&req=5

fig2: Kinetic network representation of the sampled conformational space of TRPZ1 at 450 K. Each circle represents a cluster obtained using the leader algorithm based on a DRMS metric between conformations. The size of the circle is proportional to the number of structures in the cluster. Pair of clusters are connected by lines when a transition has been sampled along the simulations. Clusters with similar connectivity pattern are placed close together in the plot. The large cluster on the top of the figure and the connected ancillary clusters within the blue boundary corresponds to the native state N. The large set of small clusters evenly distributed on the left of the figure within the green boundary corresponds to the D state. The couple of sets of clusters below the Native state and within the red boundary corresponds to the I state. The I state provides the more frequent connection between the N and D states, as revealed by the density of connecting lines. For graphical reason, small clusters were lumped together according to the dRMSD between cluster centers, so that, at the end of the procedure, all the clusters contain more than 20 conformations. Please note that the VIR was computed using unlumped clusters. The figure was prepared using VISONE (http://visone.info), similarly to what was done in Rao and Caflisch (48).

Mentions: The NMR conformation (1) represents a free energy minimum for the force field used in this work both in the control run at room temperature and in the unfolding simulations at high temperature. As anticipated in Methods, three density peaks were observed in the projection of the trajectories on the VIR-Q-RMSD space (Fig. 1, right). This picture is confirmed by a dRMSD-based cluster analysis of the trajectories (Fig. 2). The dwelling times in each state has been reported in Table 2. As expected, the larger the temperature the smaller the amount of time the peptide spends in the N state, both because of the decreasing stability of N and because of faster kinetics. Interestingly, apart from the results obtained at 373 K (where D is not reached), the population of I is always smaller than D. A detailed analysis of the I state shows that the residual structure consists of the backbone hydrogen bonds between the carboxy oxygen of Thr3 and the amide hydrogen of Thr10 and between the amide hydrogen of Glu5 and carboxy oxygen of Lys8. Further, the ordered Trp-zip motif is lost in the I state. The distances between the Trp residues on the same hairpin strand decrease, while those between Trp residues on opposite strands increase, with the exception of the pair Trp4 and Trp9, indicating the formation of an extended hydrophobic cluster that is less ordered than the zip motif (see Fig. S13 in Data S1). Several populated rotamers for the Trp side chains are present in the I and N states (see Fig. S13 in Data S1). However, the exchange rate between these rotamers is considerably faster than the rate of exchange among the states I, N, and D (see Fig. S14 in Data S1).


High temperature unfolding simulations of the TRPZ1 peptide.

Settanni G, Fersht AR - Biophys. J. (2008)

Kinetic network representation of the sampled conformational space of TRPZ1 at 450 K. Each circle represents a cluster obtained using the leader algorithm based on a DRMS metric between conformations. The size of the circle is proportional to the number of structures in the cluster. Pair of clusters are connected by lines when a transition has been sampled along the simulations. Clusters with similar connectivity pattern are placed close together in the plot. The large cluster on the top of the figure and the connected ancillary clusters within the blue boundary corresponds to the native state N. The large set of small clusters evenly distributed on the left of the figure within the green boundary corresponds to the D state. The couple of sets of clusters below the Native state and within the red boundary corresponds to the I state. The I state provides the more frequent connection between the N and D states, as revealed by the density of connecting lines. For graphical reason, small clusters were lumped together according to the dRMSD between cluster centers, so that, at the end of the procedure, all the clusters contain more than 20 conformations. Please note that the VIR was computed using unlumped clusters. The figure was prepared using VISONE (http://visone.info), similarly to what was done in Rao and Caflisch (48).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2480669&req=5

fig2: Kinetic network representation of the sampled conformational space of TRPZ1 at 450 K. Each circle represents a cluster obtained using the leader algorithm based on a DRMS metric between conformations. The size of the circle is proportional to the number of structures in the cluster. Pair of clusters are connected by lines when a transition has been sampled along the simulations. Clusters with similar connectivity pattern are placed close together in the plot. The large cluster on the top of the figure and the connected ancillary clusters within the blue boundary corresponds to the native state N. The large set of small clusters evenly distributed on the left of the figure within the green boundary corresponds to the D state. The couple of sets of clusters below the Native state and within the red boundary corresponds to the I state. The I state provides the more frequent connection between the N and D states, as revealed by the density of connecting lines. For graphical reason, small clusters were lumped together according to the dRMSD between cluster centers, so that, at the end of the procedure, all the clusters contain more than 20 conformations. Please note that the VIR was computed using unlumped clusters. The figure was prepared using VISONE (http://visone.info), similarly to what was done in Rao and Caflisch (48).
Mentions: The NMR conformation (1) represents a free energy minimum for the force field used in this work both in the control run at room temperature and in the unfolding simulations at high temperature. As anticipated in Methods, three density peaks were observed in the projection of the trajectories on the VIR-Q-RMSD space (Fig. 1, right). This picture is confirmed by a dRMSD-based cluster analysis of the trajectories (Fig. 2). The dwelling times in each state has been reported in Table 2. As expected, the larger the temperature the smaller the amount of time the peptide spends in the N state, both because of the decreasing stability of N and because of faster kinetics. Interestingly, apart from the results obtained at 373 K (where D is not reached), the population of I is always smaller than D. A detailed analysis of the I state shows that the residual structure consists of the backbone hydrogen bonds between the carboxy oxygen of Thr3 and the amide hydrogen of Thr10 and between the amide hydrogen of Glu5 and carboxy oxygen of Lys8. Further, the ordered Trp-zip motif is lost in the I state. The distances between the Trp residues on the same hairpin strand decrease, while those between Trp residues on opposite strands increase, with the exception of the pair Trp4 and Trp9, indicating the formation of an extended hydrophobic cluster that is less ordered than the zip motif (see Fig. S13 in Data S1). Several populated rotamers for the Trp side chains are present in the I and N states (see Fig. S13 in Data S1). However, the exchange rate between these rotamers is considerably faster than the rate of exchange among the states I, N, and D (see Fig. S14 in Data S1).

Bottom Line: But, the speed of the folding process is mainly determined by the transition from the completely unfolded state to the intermediate and specifically by the closure of the hairpin loop driven by formation of two native backbone hydrogen bonds and hydrophobic contacts between tryptophan residues.The temperature dependence of the unfolding time provides an estimate of the unfolding activation enthalpy that is in agreement with experiments.The unfolding time extrapolated to room temperature is in agreement with the experimental data as well, thus providing a further validation to the analysis reported here.

View Article: PubMed Central - PubMed

Affiliation: Centre for Protein Engineering, Cambridge, United Kingdom. gs@mrc-lmb.cam.ac.uk

ABSTRACT
We report high temperature molecular dynamics simulations of the unfolding of the TRPZ1 peptide using an explicit model for the solvent. The system has been simulated for a total of 6 mus with 100-ns minimal continuous stretches of trajectory. The populated states along the simulations are identified by monitoring multiple observables, probing both the structure and the flexibility of the conformations. Several unfolding and refolding transition pathways are sampled and analyzed. The unfolding process of the peptide occurs in two steps because of the accumulation of a metastable on-pathway intermediate state stabilized by two native backbone hydrogen bonds assisted by nonnative hydrophobic interactions between the tryptophan side chains. Analysis of the un/folding kinetics and classical commitment probability calculations on the conformations extracted from the transition pathways show that the rate-limiting step for unfolding is the disruption of the ordered native hydrophobic packing (Trp-zip motif) leading from the native to the intermediate state. But, the speed of the folding process is mainly determined by the transition from the completely unfolded state to the intermediate and specifically by the closure of the hairpin loop driven by formation of two native backbone hydrogen bonds and hydrophobic contacts between tryptophan residues. The temperature dependence of the unfolding time provides an estimate of the unfolding activation enthalpy that is in agreement with experiments. The unfolding time extrapolated to room temperature is in agreement with the experimental data as well, thus providing a further validation to the analysis reported here.

Show MeSH
Related in: MedlinePlus