Protein Folding Energetics, Kinetics and Models

1
Protein FoldingEnergetics, Kinetics and Models

• Oznur Tastan
• oznur_at_cs.cmu.edu
• Graduate Student
• Carnegie Mellon University

2
Lecture Outline

• Introduction What is protein folding and why it
  is a problem?
• Globular Protein Folding Models
• Detection and characterization of denatured
  states, intermediates, such as the molten globule
  and comparison to folded states
• Kinetics and pathways
• Membrane Protein Folding Models
• 2-stage and 3-stage hypothesis
• New long range interaction hypothesis
• Summary

3
Lecture Outline

• Introduction What is protein folding and why it
  is a problem?
• Globular Protein Folding Models
• Detection and characterization of denatured
  states, intermediates, such as the molten globule
  and comparison to folded states
• Kinetics and pathways
• Membrane Protein Folding Models
• 2-stage and 3-stage hypothesis
• New long range interaction hypothesis
• Summary

4
Our focus in this lecture
http//www-nmr.cabm.rutgers.edu/academics/biochem6
94/2006BioChem412/Biochem.412_2-24-2006lecture.pdf

5
Anfinsens Experiment
Addition of mercaptoethanol and urea
Removal of mercaptoethanol and urea
Native, catalytically active state. Refolded
correctly!
Native, catalytically active state.
Unfolded catalytically inactive. Reduced
disulfide bonds.
1/105 random chance
Folding is encoded in the amino acid sequence.
Native state is the minimum energy state.
Anfinsen, 1973.
6
The Protein Folding ProblemWriting the book of
Protein Origami
Now collapse down hydrophobic core, and fold over
Helix A to the dotted line. Bring charged
residues of A into close proximity of B ..?
membrane
http//www.idi.ntnu.no/grupper/KS-grp/microarray/s
lides/drablos/Fold_recognition/sld006.htm
7
How does a protein fold?Levinthals Paradox

• Assume a chain of 100 amino acids.
• Allow only 3 conformations.
• - Possible conformations 3100 1048
• Assume bond rotation rate 1014 sec.
• - Reaching the native state would take
• 1026 years ! Longer than the age of
• the universe!

Simplest case random-walk
Energy
Entropy
Protein folding cannot be random-walk.
Dill Chan, 1997
Levinthal, 1968
8
Why is protein folding problem difficult?

• Folding can be very fast, millisecond to second
  (slow folding is easier)
• Small energy changes between the denatured state
  to the native state ( 1-15 kcal/mol)
• - equivalent to the strength of a few hydrogen
  bonds
• The states populated along pathway are ensembles
  of structures

Comparison from multiple complementary
techniques are required.
9
The Three Protein Folding Models
Framework model
Hydrophobic collapse model
Nucleation condensation model
http//www.makro.ch.tum.de/users/BFHZ/Scheibel/Sch
eibel20200320Bordeaux-1.pdf
10
Lecture Outline

• Introduction What is protein folding and why it
  is a problem?
• Globular Protein Folding Models
• Detection and characterization of denatured
  states, intermediates, such as the molten globule
  and comparison to folded states
• Kinetics and pathways
• Membrane Protein Folding Models
• 2-stage and 3-stage hypothesis
• New long range interaction hypothesis
• Summary

11
The Native State

• A complex balance between
• 1) Short-range local interactions
• -intrinsic conformational preferences of the
  amino acids
• 2) Medium-range interactions
• -stabilizing regions of secondary structure
• 3) Long-range interactions
• - tertiary interactions determining the
  global fold
• Generally single conformation (with small
  fluctuations around the mean torsion angles).

12
Random Coil and Denatured State
Florys isolated pair hypothesis
Rg values determined by SAXS
F,? angles of each residue is sterically
independent There should not exist any non-local
interactions.
Rg values of 28 denatured proteins obeys the
Florys power law.
Rg RgNv N Length (Residues)
v Solvent viscosity parameter
Sosnick, T.R., et al. 2004
Flory, 1969.
13
Testing the random coil statistics
For a protein 8 of the residues are varied the
remaining 92 of the residues remained fixed in
their native conformation.
33 proteins
Number of residues
Simulated Rg follows the power law.
Despite 92 of the native structure kept, random
coil statistics are obtained.
Fitzkee, N.C. and Rose, G.D. 2004
14
The Denatured StateDoes Florys hypothesis hold?
Conformations of polyalanine chains are
enumerated to test the hypothesis.
A,G,M,R,L,F,E,K,Q J,P,O,I,o
Florys hypothesis is not valid for polypeptide
chains. Backbone conformations are limited by
additional steric clashes.
Pappu et.al 2003.
15
Can we get a structure of the denatured state?

• When the folded state breaks down
• The dispersion of all resonances decreases
• - Extensive overlap of peaks.
• 2. Greater dynamic motions between residues
• - Weak or eliminated NOEs between protons.
• 3. Ensemble of conformations
• - Each NMR parameter reflects an average
  over a dynamic ensemble of conformations.

Attainment of a high-resolution structure is not
possible in the non-native state.
16
NMR as a tool to study denatured states
Unfolded lysozyme
Folded lysozyme
For small proteins, all backbone resonances can
be resolved even in the denatured state.
17
Which and how do we use NMR parameters?
1. Measurement of NMR parameters in 15N-labeled
unfolded protein 2. Comparison of NMR
parameters - unfolded with random coil
parameters (sources - statistical
analysis from unfolded peptides - random coil
models (e.g. polymer model, Model-free
analysis, Flory etc.) - unfolded with folded
state parameters - different degrees of unfolded
states

• Chemical shifts
• Relaxation rates
• Heteronuclear NOE

• Dipolar couplings
• Scalar couplings

18
Which and how do we use NMR parameters?
1. Measurement of NMR parameters in 15N-labeled
unfolded protein 2. Comparison of NMR
parameters - unfolded with random coil
parameters (sources - statistical
analysis from unfolded peptides - random coil
models (e.g. polymer model, Model-free
analysis, Flory etc.) - unfolded with folded
state parameters - different degrees of unfolded
states

• Chemical shifts
• Relaxation rates
• Heteronuclear NOE

• Dipolar couplings
• Scalar couplings

19
Persistence of native-like topology in the
denatured states
SNase N-H Dipolar couplings
Unfolded
Folded

• Denatured proteins can preserve long range
  ordering, in conflict with the random-coil models.

Shortie. et. al. 2001
20
Which and how do we use NMR parameters?
1. Measurement of NMR parameters in 15N-labeled
unfolded protein 2. Comparison of NMR
parameters - unfolded with random coil
parameters (sources - statistical
analysis from unfolded peptides - random coil
models (e.g. polymer model, Model-free
analysis, Flory etc.) - unfolded with folded
state parameters - different degrees of unfolded
states

• Chemical shifts
• Relaxation rates
• Heteronuclear NOE

• Dipolar couplings
• Scalar couplings

21
Residual structure in lysozyme
WL-SME in urea
WL-SME in water
There are six clusters of residual structure in
WL-SME.
Klein-Seetharaman, 2002.
22
Experiment Mutation of W62
A single point mutation, W62G in cluster 3,
disrupts all clusters in reduced and methylated
lysozyme.
Klein-Seetharaman, 2002.
23
The Molten Globule(MG) State

• Molten globule is characterized by
• Absence of specific tertiary contacts
• presence of some secondary structure
• Native-like compactness
• Presence of hydrophobic core

Example a-lactalbumin Molten globule observed
in low pH
24
The Molten Globule(MG) State

• Absence of specific tertiary contacts
• Presence of some secondary structure
• Native-like compactness
• Presence of hydrophobic core

native pH5.4
MG-state pH2
unfolded state (in 9M urea,pH2)
Kuwajima, K. 1989.
25
The Molten Globule(MG) State
Rg
Native 15.70.2
MG 17.20.3
Unfolded 30.00.7

1. Absence of specific tertiary contacts
2. Presence of some secondary structure
3. Native-like compactness
4. Presence of loosely packed hydrophobic core.

Katoka, 1997.
26
The Molten GlobuleSignificance for Protein
Folding Mechanism
Disordered polypeptide collapse into the molten
globule. According to one view,
http//www.bmb.psu.edu/courses/bmb401H/Chapter7and
8.pdf
27
How do small single-domain proteins fold?

• 20 small proteins(lt 100 aa) are showed to fold
• simple two-state folding kinetics
• show variation in their folding
  rates(microseconds to seconds)
• all structures has to pass the transition state
  in order to reach the native state

Dobson, 2003
28
Kinetics of Two-State Folding
Chevron Plot of CI2
ln ?
kfold
kunfold
GdnHCl
?GT-D GT - GD - RTlnkfold
?GN-D GN GT - RTlnkunfold ?
kfold kunfold
An indicator of 2-state kinetics.

29
?-value AnalysisCharacterization of the
Transition State(TS)
TS cannot be isolated or studied directly.
Systematically introduce mutations in the native
protein. Infer structure of TS from the
energetics of the folded state (mutant versus
wild-type).
?1 site of mutation is native-like in
TS. ?0 site of mutation is unfolded in
TS. Fractional ? value partial structure in TS.
? ??GT-D / ??GN-D
Reproduced form
30
Transition State AnalysisCase Study I
Chymotrypsin inhibitor(CI2)
In the transition state of CI2 three residues
with ?-values gt0.5 come together A16,L49,I57.
A hydrophobic core supporting the
nucleation-condensation mechanism
31
Complex pathwaysCase study II hen lysozyme
Most proteins (gt100 aa) fold with observable
intermediates.
Thus cannot be approximated with simple 2-state
kinetics.
lysozyme
Dill Chan et al.1997
32
Understanding how lysozyme folds
HX NMR
ß
a
a
a
ß
a
a
ß
a
a
ß
Alpha and beta domains are two distinct folding
units.
Radford,et.al,1992
33
The details?
a domain is structured independently of the
ß domain in the early stages of folding.
HX labeling EMS
Near-UV CD
unprotected
a
ß
Far-UV CD
Large secondary structure is formed within the
milliseconds of folding.
Single exponential, the a domain forms before the
near UV develops. The tertiary contacts are not
fixed yet.
Dobson,et.al. 1994
34
Intrinsic Trp flouresence
A change in the some or all of the Trp occurs in
early collapse in later intermediates and on
formation of the native structure.
Quenching of flouresence by iodine
Exclusion of water in the early stages of
folding
Binding of ANS
Maximal emission in the early stages.
A relatively loosely packed, condensed state
exists in the early stages of reaction.
Dobson,et.al. 1994
35
Folding pathway of lysozyme
Major gt30 Minor 10
Very rapidly alpha domain forms. Hydrophobic
interior develops. Few tertiary interactions.
Protective structure evolves. Dynamic and
fluctuating beta domain.
Alpha and beta domains are stabilized
Dobson,et.al. 1994
36
Complementary approaches are essential!
Dobson, 1998.
37
Summary
No clear unifying view of protein folding has yet
emerged.
38
Membrane Protein FoldingModel systems
The lipid environment
a-helical bundles
ß-barrels
OmpA
Bacteriorhodopsin
Mammalian Rhodopsin
39
Denaturation of bacteriorhodopsin

• Effects of Urea and Guanidinium Hydrochloride
• almost none, not even on tertiary structure.

40
Denaturation of Bacteriorhodopsin
Formic acid
Native
SDS
Secondary structure remains even in SDS.
41
Why is it so difficult to disrupt secondary
structure in membrane proteins?
42
Thermodynamic considerations
The engaging of polar backbone in H-bonds is
favorable.
White Wimley,1999.
43
Refolding of Bacteriorhodopsin in the lipid
bilayers
Refolded fragments have near-identical helix
content.
C-1
C-2
1. C-1 and C-2 in SDS 2. C-1 and C-2 lipid
3. Retinal omitted.
Retinal reconstituted 90 of the activity
regenerated.
BR can assemble in to the native structure when
helices are inserted into the membrane
independently.
Popot, 1987.
44
The two-stage hypothesis Independent helix
intermediate
1st Stage Helix folds Independently
2nd Stage Final packing and interactions between
the helices are formed.
New 3rd Stage Biding of prosthetic groups,
folding of loops, oligomerization
Two stage hypothesis may not hold in all cases.
Engelman Popot, 1990. Engelman Popot, 2003.
Figures from Klein-Seetharaman,2005.
45
HR unfolded with AFM
Helix G unfolded in two steps
Short cytoplasmic segment
Part of helix E
HelixF
Helix E unfolds with helix D
Helix C
Helix A unfolds in two steps
B-C loop
Helix B
Cooperative unfolding barriers are observed, in
conflict with the 2-stage hypothesis.
Cisneros, 2005.
46
Folding Core Prediction of Rhodopsin
Gaussian Network Model (GNM)
FIRST
Agreement with the mutational data (gt90 ).
Folding core lies in the EC-TM domain interface.
Evidence for the significance of long-range
interactions in the folding of rhodopsin.
Rader. et.al 2004
47
New Model Long Range Interaction intermediate
Klein-Seetharaman,2005.
48
References

• Anfinsen, C.B. (1973) "Principles that govern the
  folding of protein chains." Science 181 223-230.
• Cisneros, D.A., D. Oesterhelt, and D.J. Muller,
  Probing origins of molecular interactions
  stabilizing the membrane proteins halorhodopsin
  and bacteriorhodopsin. Structure, 2005. 13(2) p.
  235-42.
• Dobson, C.M.Sali A., and Karplus, M., "Protein
  Folding A Perspective from Theory and
  Experiment", Angew. Chem. Int. Ed. Eng. 37,
  868-893 ( 1998).
• Dobson, C.M. "Protein Folding and Misfolding",
  Nature 426, 884-890 ( 2003).
• Dobson, C.M., P.A. Evans, and S.E. Radford,
  Understanding how proteins fold the lysozyme
  story so far. Trends Biochem Sci, 1994. 19(1) p.
  31-7.
• Engelman, D. M., Chen, Y., Chin, C. N., Curran,
  R., Dixon, A. M., Dupuy, A, Lee, A., Lehnert, U.,
  Matthews, E., Reshetnyak, Y., Senes, A., Popot,
  J-L. Membrane Protein Folding Beyond the Two
  Stage Model FEBS Lett. (2003) 555122-5.
• Flory, P. J. (1969) Statistical Mechanics of
  Chain Molecules (Wiley, New York).
• Fitzkee, N.C. and Rose, G.D. (2004). Reassessing
  random-coil statistics in unfolded proteins.
  Proc. Natl. Acad. Sci. 101 1249712502.
• Klein-Seetharaman, J., Oikawa, M.,Wirmer, J.,
  Duchardt, E., Ueda, T., Imoto, T., Smith, L.J.,
  Dobson, C. and Schwalbe, H. (2002) Long-Range
  Interactions within a Non-Native Protein. Science
  295, 1719-1722.
• Klein-Seetharaman, J. (2005) Dual role of
  interactions between membranous and soluble
  portions of helical membrane receptors for
  folding and signaling. Trends in Pharmacological
  Science 26(4), 183-189
• Kuwajima, K. (1989). The molten globule state as
  a clue for understanding the folding and
  cooperativity of globular-protein structure.
  Proteins Struct. Funct. Genet. 6 87103.
• Kataoka, M., Y. Hagihara, K. Mihara, and Y. Goto.
  (1993). Molten globule of cytochrome c studied by
  the small angle X-ray scattering. J. Mol. Biol.
  229591-596.
• Radford, S. E., Dobson, C. M. Evans, P. A. The
  folding of hen lysozyme involves partially
  structured
• intermediates and multiple pathways. Nature 358,
  302-307 (1992).
• Pappu, R. V. , Srinivasan, R. Rose, G. D.
  (2000) Proc. Natl. Acad. Sci. USA 97,
  12565-12570.
• Popot, J-L and Engelman D.M."Membrane Protein
  Folding and Oligomerization The Two-Stage Model
  Biochemistry (1990), 29 (17), 4031-7.
• Popot J.L., Gerchman S.E., Engelman D.M. (1987)
  Refolding of bacteriorhodopsin in lipid bilayers.
  A thermodynamically controlled two-stage process.
  J. Mol. Biol. 198655-76
• Shortle D, Ackerman MS. (2001) Persistence of
  native-like topology in a denatured protein in 8
  M urea. Science. Jul 20293(5529)487-9.
• White S. H. and Wimley, W. C. (1999).  Membrane
  protein folding and stability Physical
  principles.  Annu. Rev. Biophys. Biomol. Struct.
  28319-365. 

49
Acknowledgements

• Dr. Judith Klein-Seetharaman
• Dr. Sanford Leuba
• The class of MB3 (Spring 2006)

50

• QUESTIONS?

http//www.otago.ac.nz/humannutrition/dietetics/gf
x/philosophy.jpg