Using%20two%20recently-developed%20molecular%20dynamics%20protocols%20for%20protein%20folding

1
Using two recently-developed molecular dynamics
protocols for protein folding

• Timothy H. Click
• Department of Chemistry and Biochemistry
• University of Oklahoma
• Norman, Oklahoma

2
Outline

• Introduction to MD protocols
• Previous work
• Simulations of tryptophan zipper 2
• Simulation of Streptococcal protein G B1 domain
  (residues 41-56)
• Conclusions
• Future directions
• Acknowledgements

3
Protein geometry optimization
Dill, K.A. Chan, H.S. Nat. Struct. Biol., 1997,
4, 10-19.
4
1st MD protocol DIVE

• Disrupted Velocity (DIVE) search protocol
• Velocity reassignment of coordinate histories
• Magnitude rescaling energy perturbation
• Direction changes
• Reassignment every n steps (defined by user)
• Heating and cooling cycles
• Conformations sampled near absolute zero
• Overall, protocol disrupts equilibrium
• Energy barriers overcome or circumvented
• Several potential energy minima sampled

5
How conformations are selected
ß
ßlow to DIP
ßlow
ßlow2 to DIP
6
2nd MD protocol DIP

• Divergent Path (DIP) search strategy
• Coordinate histories at same constant temperature
• Simulations involve multiple coordinate histories
• Individual coordinate histories randomly assigned
  initial velocities
• Velocities can be altered allowing for different
  conditions
• Constant temperatures maintained by rescaling
  velocity magnitudes
• Broader sampling of potential energy surface
  allowed

7
DIP simulation
8
DIP simulation (contd)
ltEgt -624.22 8.27 kcal/mol ltRMSDgt 1.6 0.5 Å
ltEgt -632.42 8.44 kcal/mol ltRMSDgt 1.8 0.2 Å
ltEgt -587.16 9.50 kcal/mol ltRMSDgt 13.2 0.5
Å
ltEgt -633.94 7.87 kcal/mol ltRMSDgt 1.5 0.2 Å
nmr
ltEgt -592.02 8.38 kcal/mol ltRMSDgt 12.5 0.5
Å
ltEgt -600.29 13.05 kcal/mol ltRMSDgt 11.3
0.7 Å
9
Protocol procedures

• Modified Amber force field (Okur,A. Strockbine,
  B. Hornak, V. Simmerling, C., J. Comput. Chem.,
  2003, 21)
• Constraints on atoms covalently bonded to
  hydrogen
• Implicit solvent
• 2 fs time step
• 4,000,000 steps
• Velocity disruption every 20,000 steps (DIVE)
• T 300 20 K (DIP)
• 6 independent coordinate histories/simulation

10
Previous work with a-helices

• Zunnan Huang
• 13-residue polyalanine
• Trp-cage (a-helix and 310-helix
• Huang and Zhanyong Guo
• Peptide F
• Timothy H. Click
• C-peptide of ribonuclease A (residues 1-13)

11
Tryptophan zipper 2 (trpzip2)

• De novo 12-residue polypeptide
• Sequence (S1WTWENGKWTWK12-NH2)
• PDB code 1LE1 (20 NMR models)
• Stable ß-sheet in aqueous solution by
  cross-stranded pairs of four tryptophans
• Simulations completed by other groups

1 Cochran, A.G. Skelton, N.J. Starovasnik, M.A.
P. Natl. Acad. Sci. USA, 2001, 98, 5578-5583.
12
Trpzip2 DIVE Results
extlow E -496.59 kcal/mol RMSD 6.9 ?
ßlow E -494.05 kcal/mol RMSD 5.0 ?
alow E -498.22 kcal/mol RMSD 6.1 ?
ß E -489.44 kcal/mol RMSD 0.9 ?
extlow2 E -499.23 kcal/mol RMSD 7.0 ?
ßlow2 E -497.06 kcal/mol RMSD 5.7 ?
alow2 E -498.17 kcal/mol RMSD 6.1 ?
13
Trpzip2 DIP Results
ext ltEgt -360.82 7.66 kcal/mol ltRMSDgt 6.5
0.8 ?
ß ltEgt -381.65 4.55 kcal/mol ltRMSDgt 0.9 0.1
?
a ltEgt -369.79 9.01 kcal/mol ltRMSDgt 6.7 0.4
?
extlow ltEgt -375.47 6.28 kcal/mol ltRMSDgt 7.5
0.1 ?
ßlow ltEgt -374.14 7.18 kcal/mol ltRMSDgt 7.0
0.3 ?
alow ltEgt -368.77 7.38 kcal/mol ltRMSDgt 6.1
0.1 ?
ß ltEgt -381.65 5.04 kcal/mol ltRMSDgt 0.8
0.1 ?
extlow2 ltEgt -381.44 5.74 kcal/mol ltRMSDgt 7.4
0.1 ?
ßlow2 ltEgt -378.75 6.04 kcal/mol ltRMSDgt 7.0
0.3 ?
alow2 ltEgt -368.93 7.54 kcal/mol ltRMSDgt 6.2
0.3 ?
14
Trpzip2 Summary

• PES rough at low temperatures
• ß-hairpin challenging secondary structure
• ß-hairpin as relative global PE conformation
• a-helices metastable conformation

15
B1 domain of Streptococcal protein G

• Natural ß-hairpin stable in aqueous solution.
• Sequence (G41EWTYDDATKTFTVTE56)
• PDB 2GB1 (x-ray crystal structure)
• Stabilization factors
• Hydrophobic core
• Terminal salt bridge
• Several simulations

2 Gronenborn, A. M. Filpula, D. R. Essig, N.
Z. Achari, A. Whitlow, M. Wingfield, P. T.
Clore, G. M. Science, 1991, 253, 657-661.
16
Protein G DIVE results
extlow E -784.40 kcal/mol RMSD 7.4 ?
ßlow E -774.96 kcal/mol RMSD 6.8 ?
alow E -783.53 kcal/mol RMSD 8.4 ?
ß E -770.54 kcal/mol RMSD 0.9 ?
extlow2 E -785.85 kcal/mol RMSD 7.3 ?
ßlow2 E -781.75 kcal/mol RMSD 6.4 ?
alow2 E -785.64 kcal/mol RMSD 8.4 ?
17
Protein G DIP results
ext ltEgt -612.63 9.31 kcal/mol ltRMSDgt 10.6
0.6 ?
ß ltEgt -638.42 5.84 kcal/mol ltRMSDgt 1.6 0.4
?
a ltEgt -646.18 6.90 kcal/mol ltRMSDgt 8.9 0.2
?
extlow ltEgt -640.60 7.13 kcal/mol ltRMSDgt 9.1
0.3 ?
ßlow ltEgt -651.05 6.34 kcal/mol ltRMSDgt 6.9
1.2 ?
alow ltEgt -646.14 6.01 kcal/mol ltRMSDgt 9.0
0.3 ?
ß ltEgt -644.28 6.08 kcal/mol ltRMSDgt 1.6
0.2 ?
extlow2 ltEgt -633.74 7.13 kcal/mol ltRMSDgt 9.0
0.7 ?
ßlow2 ltEgt -643.91 7.86 kcal/mol ltRMSDgt 9.0
0.3 ?
alow2 ltEgt -650.14 6.33 kcal/mol ltRMSDgt 9.0
0.2 ?
18
Protein G summary

• ß-hairpin stable at 300 K
• Helical conformation lower in energy
• Better energy compensation3
• Agreement with other simulation4
• Various factors may overstabilize helices (e.g.,
  implicit solvent, salt bridges)

3 Muñoz, V. Thompson, P. A. Hofrichter, J.
Eaton, W. A. Nature, 1997, 390, 196-199. 4
Krivov, S. V. Karplus, M. P. Natl. Acad. Sci.,
USA, 2004, 101, 14766-14770.
19
Conclusions

• DIVE and DIP locate several PE minima
• PES mapped by DIVE
• PES of conformations at desired temperature with
  DIP
• Conformations in good, if not excellent,
  agreement with experimental structures using DIP
  and DIVE

20
Future directions

• Continue validation of MD protocols with larger
  ß-sheet
• Further test MD protocols with tertiary structure
• Predict structure of small protein

21
Acknowledgements

• Ralph A. Wheeler
• Zunnan Huang and Adam Hixson
• National Research Service Award 5 F31 GM067560-03
  to THC from the NIH/NIGMS
• Oklahoma Center for the Advancement of Science
  and Technology (OCAST) HR01-148
• Oklahoma Supercomputing Center for Education and
  Research (OSCER)
• NSF/NRAC supercomputer time MCA96-N019