1. REVIEW P R OT E I N F O L D I N G
An expanding arsenal of experimental methods
yields an explosion of insights into protein folding
mechanisms
Alice I Bartlett & Sheena E Radford
© 2009 Nature America, Inc. All rights reserved.
In recent years, improvements in experimental techniques and enhancements in computing power have revolutionized our
understanding of the mechanisms of protein folding. By combining insights gained from theory, experiment and simulation we are
moving toward an atomistic view of folding landscapes. Future challenges involve exploiting the knowledge gained and methods
developed to enable us to elucidate a molecular description of folding dynamics in the complex environment of the cell.
Most proteins are required to adopt a specific three-dimensional mutually supportive, weak interactions that cannot all be satisfied
structure to be biologically active. How they achieve this has been simultaneously during folding. As a result, energetic minimization of
the subject of immense scientific interest spanning the decades since individual interactions can be conflicting, leading to ‘frustration’ in
the first structures of proteins were elucidated1. The conformational the energy landscape5,7. This ruggedness may be attributed to oppos-
space accessible to the polypeptide chain is astronomically large, yet ing evolutionary pressures on protein sequences to enable them to
proteins fold on a biologically relevant timescale, with some obtaining fold reliably but also to avoid aggregation and to carry out specific
their native structure in vitro in just microseconds2. How rapid folding biological functions5,7.
is achieved has been rationalized by a number of concepts: the presence Landscape theory predicts an additional folding scenario in which
of nonrandom interactions in the initial denatured state that limit the the native structure is attained without encountering any substantial
conformational space available at the start of the folding reaction3; energy barriers, so-called ‘downhill folding’6. In principle, at least,
folding via intermediates that mark the way to the native structure4; proteins that fold in a downhill manner open the door to character-
and the realization that proteins fold on funnelled energy landscapes5 ization of the folding landscape in immense detail via the myriad of
that describe folding as the inevitable consequence of the requirement non-native conformations that are accessible experimentally for such a
to lower the free energy (increase stability) as more native contacts folding mechanism. Barrierless folding is difficult to demonstrate
form6 (Fig. 1). In this landscape view of folding, the denatured state of unequivocally by experiment, as proteins that fold in this manner
the protein populates a large ensemble of structures. The polypeptide are expected to obtain their native structure with rates close to the
chain may then fold by numerous pathways, potentially adopting folding ‘speed limit’2. In addition, the experimental hallmarks of this
multiple partially folded ensembles en route to the native state6. type of folding are difficult to define8–10. Nonetheless, downhill
For a protein that folds via a two-state transition (with a mechan- folding has been suggested (with much debate11,12) for a number of
ism in which only the denatured and native states are populated) the model proteins13–15. Single-molecule experiments have been proposed
energy landscape is relatively smooth. Such a landscape lacks deep as a means to differentiate downhill and two-state folding10,16,
valleys and high barriers and effectively funnels the polypeptide chain although this is not straightforward even with such a powerful
to its native state (Fig. 1a). Such an ideal folding scenario is rare4, and approach17. A more detailed discussion of fast protein folding and
many proteins fold on rough, rugged landscapes. Using new methods downhill folding scenarios can be found in ref. 18.
that can detect sparsely populated and/or transient non-native species Characterization of all the non-native species (unfolded states,
(Table 1), even small, simple proteins have now been shown to fold transition states and partially folded intermediates) encountered by
through one or more partially folded states4,5. In general, folding proteins that fold in a barrier-limited manner is essential if we are to
energy landscapes are rugged entities that are suboptimal for folding realize our quest to understand how proteins fold in all-atom detail.
(Fig. 1b) through which the polypeptide chain has to navigate to the Substantial advances toward this goal have been realized for a handful
native state5. Landscape ruggedness arises as the consequence of the of small proteins19–25. This has been enabled by the development of
simple fact that native protein structures are stabilized by thousands of experimental approaches with faster timescales of measurement26 and
enhanced sensitivity (Table 1 and references therein), together with
Astbury Centre for Structural Molecular Biology and Institute of Molecular and improvements in computing power and new theoretical tools6,27.
Cellular Biology, University of Leeds, Leeds, UK. Correspondence should be Today, the arsenal of biophysical methods available to the experimen-
addressed to S.E.R. (s.e.radford@leeds.ac.uk). talist allows transitions from picosecond to second (or longer)
Published online 3 June 2009; doi:10.1038/nsmb.1592 timescales to be monitored and species populated to as little as
582 VOLUME 16 NUMBER 6 JUNE 2009 NATURE STRUCTURAL & MOLECULAR BIOLOGY

2. REVIEW
Entropy nature of non-native species. Defining the structural properties of
a
non-native states, and determining how they interconvert so as to
align them in the context of a folding pathway, has remained a central
issue since this field began4. Today, the use of experimental methods
with enhanced time resolution and sensitivity, in combination with
molecular dynamics simulations, are beginning to reveal all-atom
models of non-native ensembles20–24,31. Although there remains
Energy
further room for optimization of this approach32,33, it has been
particularly useful in allowing visualization and interrogation of
ensembles of structures that represent the experimental data (rather
Native than a unique solution to the experimental observables). These models
state
can then inspire new experiments to test and refine the structural
ensembles produced22,34.
Entropy
b The denatured protein ensemble is of particular interest in the
context of the landscape view of folding, because this is the state
from which folding initiates. As the denatured protein ensemble is
rarely populated at equilibrium, obtaining structural information
about this species is a challenging task. This has been achieved
using naturally unstable proteins in which the native and denatured
Energy
© 2009 Nature America, Inc. All rights reserved.
Intermediate states are in equilibrium under ambient conditions35 or by creating
proteins by mutation19 or chemical modification36 that are denatured
under conditions that typically favor folding in their wild-type
Intermediate
counterparts. Alternative strategies involve denaturing the protein
Native under acidic conditions or adding chaotropes, with the caveat that
state
the structural properties of these ensembles may differ from those
under less harsh conditions37,38.
Figure 1 Schematic representation of folding funnels. Example of a smooth Studies of the denatured ensemble of the helical protein Im7,
energy landscape, through which the polypeptide chain is effectively formed in 6 M urea, using chemical shift analysis and NOE measure-
funneled to the native structure (a), and a more rugged landscape, through ments, revealed that this species lacks regular elements of secondary
which the polypeptide chain has to navigate, possibly via one or more structure. Despite this, the polypeptide chain contains clusters of
populated intermediates, to the native state (b). In both examples, the interacting hydrophobic side chains in those regions that ultimately
denatured state occupies a broad ensemble of structures containing
form helices in the native state, potentially priming the protein for
elements of both native and non-native interactions.
subsequent folding events37. The existence of structure in the dena-
tured state of Im7 in the presence of chaotrope builds on an increasing
0.5% to be identified and structurally assessed28. In principle, single- body of data that indicates conformational restriction in denatured
molecule techniques offer the potential to map folding events one chains35,36,38. Folding of the helical l repressor protein also com-
molecule at a time. Using this approach, rare species can be detected mences from a highly nonrandom state36. When denatured under
and characterized that may be hidden by the averaging inherent within ambient conditions by oxidation of methionine residues, this protein
ensemble experiments17. This approach also enables the measurement has been shown to possess nascent a-helical structure in the
of intramolecular diffusion coefficients in denatured and partially N-terminal region, whereas the C-terminal region remains nonhelical
folded states, providing detailed insights into the nature of the yet conformationally restrained36. All-atom images of the denatured
polypeptide chain at different stages of folding29,30. Perhaps most ensembles of the all-helical acyl coenzyme A binding protein (ACBP)
importantly, these experiments can link models based on chemical and the all b-sheet drkN SH3 domain have been obtained using
kinetics commonly used in protein folding with the more physical NMR paramagnetic relaxation enhancement to provide restraints for
description of folding in terms of quantitative free-energy surfaces17. simulations35,38. These experiments revealed denatured ensembles
A detailed review of the insights that have revolutionized our containing species, ranging from expanded to highly compact, that
understanding of protein-folding mechanisms and their impact on are stabilized by both native and non-native interactions. Together,
biology is beyond the scope of this short article. Here we focus on these studies indicate that the denatured states of proteins are highly
three areas that have seen major advances in recent years: (i) the heterogeneous, containing polypeptide chains that vary widely in their
structural diversity and properties of non-native states, (ii) current individual conformational properties.
knowledge about folding pathways and the extent to which protein By placing donor and acceptor chromophores at different positions,
sequences are optimized for folding efficiency, and (iii) new Schuler and co-workers have exploited the power of single-molecule
approaches that are beginning to allow us to take the knowledge ¨
Forster resonance energy transfer (FRET) and fluorescence correlation
gained from in vitro studies toward a molecular description of folding spectroscopy (FCS) to determine distance distributions in the
in the cell. In each area we highlight a selection of recent studies unfolded ensemble of the cold-shock protein CspTm and the rate of
showing how different experimental approaches have been used to intramolecular diffusion of the polypeptide chain in the denatured
elucidate new details of protein-folding mechanisms. state at different denaturant concentrations29. These results suggest
that the polypeptide behaves as a Gaussian chain even at low
The structural properties and diversity of non-native species denaturant concentrations where the protein is collapsed and contains
A major challenge in the structural, kinetic and thermodynamic charac- B20% of its native b-sheet structure30. This study reconciles the
terization of folding landscapes is the transient and heterogeneous seemingly contradictory results from small-angle X-ray scattering
NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 16 NUMBER 6 JUNE 2009 583

3. REVIEW
Table 1 Experimental techniques that have been applied to the study of protein folding
Technique Timescale Information content Comments Refs.
Intrinsic tryptophan Znsa Environment of tryptophan (through measurement Tryptophan can be introduced (or removed to create a 97
fluorescence of intensity and lmax) single-tryptophan protein) by protein engineering
Far UV CD Zmsa Secondary-structure content Synchrotron CD may allow a more accurate interpretation 98
of structure content. Can be complicated by aromatic
contributions to the spectrum
Near UV CD Zmsa Packing of aromatic residues Only fixed interactions give a near UV CD signal 98
Raman spectroscopy Zmsa Solvent accessibility, conformation of aromatic Information content depends on the frequency used. 99
residues Not widely applied to folding studies
Infrared spectroscopy Znsa Secondary-structure content Combined with solvent-exchange, information about 100
hydrogen-exchange protection can be obtained
ANS (1-anilino-8- Zmsa Exposure of aromatic surface area Care needs to be taken to ensure that ANS itself does not 97
napthalene sulfonic perturb folding
acid) binding
FRET Zpsa Molecular ruler, dependent on the distance between Information about rapid fluctuations is possible. Careful 17
two fluorophores (r–6 dependence assuming free design needed to incorporate dyes without perturbing
rotation of the dyes) folding
© 2009 Nature America, Inc. All rights reserved.
FCS Zps Diffusion time (and hence size and shape) A powerful method capable of resolving co-populated 101
conformers and their rates of interconversion over ps–ms
timescales
Anisotropy Zmsa Correlation time measurements provide information Can provide useful complementary information to FRET 97
about shape and size of molecule distance distributions
Small-angle X-ray Zmsa Radius of gyration With modeling, information about three-dimensional 102
scattering structure can be obtained
Absorbance Znsa Environment of chromophore Peptide bond, aromatic residue or extrinsic moiety may be 26
used
Real-time NMR 4min Structural information via chemical shifts and Powerful method for analysis of denatured states and 103
measurement of NOEs intermediates in slowly folding proteins
Native-state hydrogen h Global stability, detection of metastable states Rare species in equilibrium with the native state that are 104
exchange difficult to detect using other methods can be revealed
Pulsed H/D exchange by Zms Hydrogen exchange protection of folding Multiple exponential hydrogen-exchange behaviour indicates 104
NMR intermediates on a per-residue basis parallel pathways
Pulsed H/D exchange by Zms Hydrogen exchange protection of folding populations Quantification of the population of species within 105
ESI-MS heterogeneous ensembles with different hydrogen-exchange
properties
NMR relaxation Bms Nonrandom structure in denatured states and If exchange between species occurs at a suitable rate (ms), 28,103
methods conformational exchange between different species structural, kinetic and thermodynamic information about
rare species can be obtained
Protein engineering Depends Role of an individual residue in determining the rate Double-mutant cycles provide pairwise information. F-values 42
on probe of folding and stability of a species of interest provide indirect structural information via free-energy
used changes. C-values use metal chelation to bihistidine motifs
to identify specific side chain contacts
For a review of theories and simulation methods of folding, see refs. 6,27.
aThe timescale depends on the method used to initiate folding: temperature jump (ns), pressure jump (ms), ultra-rapid mixing (ms), stopped flow (ms) or manual mixing (s).
experiments that demonstrate that the overall dimensions of unfolded transition state ensembles of two homologous PDZ domains using
proteins are consistent with random coil models39 and increasing this approach demonstrated that the early transition states of the two
evidence of residual structure in unfolded ensembles from NMR domains are less similar in structure than the subsequent rate-limiting
studies35–38. Even within such a structured denatured state, the global transition state ensembles. This is consistent with the landscape view
reconfiguration time is rapid (B50 ns)29. A study of loop formation that conformational space is less restricted earlier in folding23. The late
in unfolded polypeptides using triplet-triplet energy transfer also transition state of both proteins adopts a narrow ensemble of
observed large-scale motions involving chain diffusion occurring on structures with native-like topology. This demonstrates that confor-
a timescale of 10–100 ns, whereas faster kinetics were observed on the mational sampling is highly restricted by this stage of folding, as has
50–500 ps timescale corresponding to local fluctuations40. been found previously in several other proteins21,31.
For the characterization of more highly structured non-native Although the interpretation of F-values (energetic parameters) in
species, such as partially folded intermediates and transition states, structural terms requires caution43–45, for populated intermediates
the use of protein engineering (F-value analysis) is well estab- independent analysis of F-values, chemical shifts and hydrogen
lished41,42. In recent years F-values have been used as restraints for exchange protection factors allows assessment of the quality of the
molecular dynamics simulations to generate atomic-level structural ensembles that result from molecular dynamics simulations using
models of these ensembles20–23,31. Recent analysis of the early and late different observable parameters as restraints20. The characterization
584 VOLUME 16 NUMBER 6 JUNE 2009 NATURE STRUCTURAL & MOLECULAR BIOLOGY

4. REVIEW
a b both proteins substantial numbers of non-native (as well as native)
contacts are formed in the intermediate ensembles20,21,25, indicative of
frustration in folding landscapes of even these small, simple proteins.
Direct observation of (un)folding trajectories in real time using
single-molecule fluorescence techniques offers further opportunities
to monitor folding reactions and to reveal rare events or species
hidden by the averaging of ensemble experiments17. Using immobi-
lization techniques on surfaces or encapsulation within liposomes to
increase observation times, the first trajectories of folding reactions
of individual proteins in real time are emerging48,49. Although
single-molecule fluorescence studies such as these should be able to
expose multiple species on the reaction coordinate, significant chal-
lenges lie ahead in developing experiments to allow the properties of
rapidly interconverting species to be discerned. Mechanical manipula-
tion using optical tweezers or the atomic force microscope has already
revealed the presence of intermediates when individual proteins are
Figure 2 Models of the conformational properties of the ensembles
representing the folding intermediates of the bacterial immunity protein Im7
unfolded under force50,51.
(a; ref. 21) and the rare folding intermediate of the G48V variant of the
Fyn SH3 domain24 (b). The native structure of each protein is shown Evolution of folding pathways
© 2009 Nature America, Inc. All rights reserved.
below (PDB 1AYI for Im7 (ref. 94); PDB 1SHF for Fyn SH3 domain95). The landscape view presents a powerful picture of protein folding, in
Comparison with the native structure demonstrates that the native topology that it allows a clear portrayal of the heterogeneity of species on the
is well defined in the folding intermediates of both these small proteins. folding surface. It also highlights the importance of native contacts in
Images of the intermediate ensembles reproduced from Nat. Struct. Mol.
Biol. (ref. 21) and Nature (ref. 24).
funneling the folding chain toward the native state6. As a consequence,
the native topology determines the sequence of folding events,
rationalizing why the structural mechanism of folding is conserved
of folding intermediates that are stably populated is particularly in protein families52 (even if the kinetic mechanism (for example,
important because, when on-pathway, they represent stepping stones two- or three-state) varies53). It also explains the observed correlation
en route to the native state4. Such species have also been implicated in between folding rate and the complexity of the native fold
misfolding diseases46, and their structural characterization offers (contact order)54.
prospects for therapeutic intervention. By the careful manipulation Important questions result from viewing folding as a multidimen-
of experimental conditions to modulate the population of intermedi- sional search process. These include how many routes to the native
ate species and their rates of interconversion, it is possible to state are taken by a folding polypeptide chain and the sensitivity of the
characterize these species using the range of approaches listed in pathways taken to the experimental conditions and protein sequence.
Table 1. Even ‘hidden’ intermediates that are kinetically invisible Some proteins seem to fold via a single route through the energy
(because they form after the rate-limiting transition state) can be landscape, as famously portrayed by chymotrypsin inhibitor 2
detected and structurally characterized using native-state hydrogen (ref. 55). For other proteins, the route map is more diverse56,57. For
exchange47. Rare intermediates can also be detected and structurally multidomain proteins, the possibility of folding via parallel routes is
analyzed using relaxation dispersion NMR28 (Table 1). Structural an obvious, and real56, possibility. Other long-established causes of
ensembles representing the folding intermediate of the Fyn SH3 parallel routes and alternative conformations involve cis-trans proline
domain have been calculated using chemical shifts determined from isomerization or disulfide oxidation58–60. In more recent work,
relaxation dispersion NMR experiments as restraints24 (Fig. 2). A Oliveberg and co-workers suggested that the number of pathways
similar strategy was used to determine an ensemble of intermediate accessible to a polypeptide chain may be linked to the number of
structures for Im7 using a combination of F-values, hydrogen
exchange protection factors and chemical shifts as restraints20,21
(Fig. 2). The finding that both of these small, single-domain proteins a b S6wt
fold via intermediates underlines the generic importance of partially
folded species in protein-folding reactions. The conformational prop-
erties of these species show that, for these proteins, the native topology
is well defined by this point in folding. Perhaps more surprisingly, for
N
C
N
P13-14
Figure 3 Overlapping nucleation motifs in the ribosomal protein S6. C
(a) Above, structure of wild-type S6 (S6wt; PDB 1RIS96) colored to show
the possibility for the protein to fold via different folding nuclei: a1 (red)
and a2 (blue). Both nuclei share the central b1 strand (purple). Below,
N C
schematic of the secondary structure of S6, demonstrating overlap of the
two folding nuclei. (b) Schematics demonstrating how local loop entropy
influences which of the two nuclei dominates folding and, hence, the α1 α2 P81-82 N
nucleus nucleus
structural folding mechanism of the protein. P13-14 and P81-82 are
C
circular permutants in which the N and C termini of the wild-type protein
are linked and new termini created between positions 13 and 14, and β1
81 and 82, respectively. Figure redrawn from ref. 63.
NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 16 NUMBER 6 JUNE 2009 585

5. REVIEW
nucleation motifs contained within its sequence61. These authors course of evolution, the cell has devised cunning schemes to enable
identified the minimal nucleation motif for folding of the ribosomal proteins (which are generally larger and more complex than those that
protein S6 and showed it to comprise an a-helix docked against two have been studied in detail biophysically to date) to fold correctly.
b-strands, a motif similar in size to the smallest cooperatively Other challenges include the requirements for post-translational modi-
stabilized proteins62 (Fig. 3). By creating circular permutants of S6, fication, cofactor binding, complex formation and compartmental-
the authors showed that the nucleation motif is conserved in all ization, all of which must be intricately controlled to ensure cellular
sequences and always includes the central b-strand 1 but is completed homeostasis. Now that many of the proteins involved in chaperoning,
by different structural elements that are dependent on the local loop targeting, modifying and degrading proteins have been identified, the
entropies of the individual permutants62,63 (Fig. 3). Recent studies challenge is to determine the shape of the folding landscape in the
using ankyrin repeat proteins have also revealed the presence of cellular context and to understand how it might be altered by changes in
multiple nucleation sites64–66. This suggests that the opportunity the cellular environment. In addition to describing the initial path to the
to fold via different routes may be a general feature of evolved native state that commences with protein synthesis, delineation of the
folding landscapes. shape of the folding landscape in vivo will allow the probability and
Using the range of methods now available, the folding mechanisms molecular nature of excursions from the native state to be identified.
of more than 20 small, water-soluble proteins have been interrogated This would allow the global and subglobal unfolding events that are
in detail. For most of these proteins, folding is remarkably efficient crucial for function, molecular recognition and degradation to be
in vitro. Thus, gross misfolding and aggregation are rare, folding is understood in atomistic detail.
rapid (usually occurring in less than 1 second), and intermediate Although detailed description of the folding landscape in vivo will
states, if formed, are transient4. By contrast with the behavior of these require further advances in methodology and increased computational
© 2009 Nature America, Inc. All rights reserved.
proteins, the 93-residue de novo designed protein Top7 folds with a power, significant steps have been made toward unraveling the
mechanism too complex to be solved kinetically, involving numerous mechanisms of folding in the cell. Theoretical and experimental studies
highly populated non-native states67. This suggests that the process of have shown that molecular crowding increases the stability of compact
evolution has yielded sequences that are relatively well designed for states (native and non-native) over their expanded counterparts75–77.
folding. Imperfections in the landscape presumably reflect additional Crowding can also enhance folding rates78,79 or result in conforma-
constraints that have limited the evolution of the sequence, such as the tional changes postulated to have important consequences for func-
requirement to avoid aggregation, to remain sufficiently soluble or to tion80. Confinement (either within chaperones or within the ribosome
be functional7,68. exit tunnel) may also have important consequences for folding in the
The conflict between folding and function is an emerging theme cellular environment77. Other Reviews in this issue81,82 deal with these
in current studies of protein folding. Several reports have docu- topics in detail. Exciting advances in the power of biophysical studies,
mented the effects of these opposing evolutionary pressures on the particularly in NMR83 and fluorescence techniques17,84,85, are begin-
folding landscape7,21,69–71. Examples include a statistical survey of ning to reveal insights into folding events in ribosome-bound nascent
natural proteins, which demonstrated that highly frustrated inter- chains83,86, within chaperones87,88 and even within living cells89–92.
actions colocalize with ligand binding sites in protein structures, These studies have shown the propensity for folding subsequent to the
underlining the opposing requirements of folding and function7. emergence of the polypeptide chain from the ribosomal exit tunnel83
Recent temperature-jump studies of the human PIN1 WW domain or, in some circumstances, even within the ribosomal exit tunnel86.
also revealed how the evolutionary requirement to endow function They have also demonstrated the consequences of conformational
is achieved at the expense of rapid folding and native-state stabi- restriction on the folding of polypeptide chains when they are confined
lity69. For Im7, the transient formation of non-native interactions within the GroEL folding cage87,88,93.
early in folding, which involves solvent-exposed residues that are Following folding in real time in intact cells is an immensely
vital for function, provides a further example of frustration in the challenging goal, and there is some way to go before we will be able
folding landscape21. Finally a recent molecular dynamics study of to depict realistic models of the folding landscape therein. A number of
interleukin-1b (IL-1b) revealed the phenomenon of ‘backtracking’, recent, innovative studies have taken the first steps in this direction.
in which subsets of native contacts form, break and then reform Exploitation of a tetracysteine motif that specifically binds a biarsenical
later during folding70. Such real-time editing of prematurely fluorescein dye (FlAsH) has enabled measurement of the unfolding free
formed native interactions contributes to the slow folding of this energy of a small protein, cellular retinoic acid binding protein
protein and is caused by residues within a functionally important (CRABP), in the Escherichia coli cytoplasm91. This approach has also
b-bulge70,71. It seems that today’s sequences are not perfected for been used to monitor protein aggregation in vivo92. The continuing
folding but represent a compromise of the different forces encoun- development of NMR techniques offers the potential to study protein
tered during their evolutionary history. As well as providing exciting structure and dynamics in whole cells89,90. Using 15N-labeled protein
opportunities for the experimentalist to improve upon nature’s (the B1 domain of protein G) injected into Xenopus laevis oocytes,
designs, this also furnishes the threat that minor changes to the Selenko et al. were able to record high-resolution HSQC spectra in the
sequence and/or environmental conditions may increase landscape eukaryotic cytosol. Analysis of line widths and chemical shifts allowed
ruggedness to such an extent that it has deleterious effects on the quantitative comparison of the structure and dynamics of this small
maintenance of a healthy living cell72. protein in buffer, in crude X. laevis oocyte extracts, in solutions
containing macromolecular crowding agents and in the intact cell90.
Toward a molecular description of folding in the cell The ultimate goal of monitoring folding kinetics in real time at the
Translating how the insights gained from biophysical studies of folding level of a single protein molecule in vivo awaits enhancements in dye
in vitro relate to the physiological process of folding in the cell is a technology, labeling strategies and instrument development. Impressive
further major challenge. In the cellular setting, folding proceeds in a achievements in this area have allowed the kinetics of binding of
crowded environment73 that is packed with molecular chaperones, individual repressor molecules to DNA in E. coli to be monitored in
which assist the folding process in this hostile environment74. Over the real time at the single-molecule level85. This technical tour de force
586 VOLUME 16 NUMBER 6 JUNE 2009 NATURE STRUCTURAL & MOLECULAR BIOLOGY

6. REVIEW
bodes well for innovations in this area and for the application of 11. Ferguson, N., Sharpe, T.D., Johnson, C.M., Schartau, P.J. & Fersht, A.R. Structural
single-molecule approaches to the study of protein folding and biology—analysis of ’downhill’ protein folding. Nature 445, E14–E15 (2007).
12. Zhou, Z. & Bai, Y.W. Structural biology—analysis of protein-folding cooperativity.
dynamics in the living cell. Nature 445, E16–E17 (2007).
13. Ma, H. & Gruebele, M. Kinetics are probe-dependent during downhill folding of an
engineered lambda(6-85) protein. Proc. Natl. Acad. Sci. USA 102, 2283–2287
Summary and outlook (2005).
The wide range of techniques developed over recent years has led to a 14. Sadqi, M., Fushman, D. & Munoz, V. Atom-by-atom analysis of global downhill protein
near-atomistic view of the folding landscape of small, model proteins folding. Nature 442, 317–321 (2006).
15. Fung, A., Li, P., Godoy-Ruiz, R., Sanchez-Ruiz, J.M. & Munoz, V. Expanding the realm
in vitro. Approaches combining both experiment and simulation have of ultrafast protein folding: gpW, a midsize natural single-domain with a+b topology
been crucial in achieving this goal, and the synergy of these that folds downhill. J. Am. Chem. Soc. 130, 7489–7495 (2008).
approaches will become even more important as the field develops 16. Huang, F., Sato, S., Sharpe, T.D., Ying, L.M. & Fersht, A.R. Distinguishing between
cooperative and unimodal downhill protein folding. Proc. Natl. Acad. Sci. USA 104,
in the future. At a fundamental level, much remains to be learned 123–127 (2007).
about the biophysics of protein folding: the prediction of folds and 17. Schuler, B. & Eaton, W.A. Protein folding studied by single-molecule FRET. Curr.
Opin. Struct. Biol. 18, 16–26 (2008).
folding mechanisms is far from routine27 and comparison of results 18. Dyer, R.B. Ultrafast and downhill protein folding. Curr. Opin. Struct. Biol. 17, 38–47
from simulations and experimental approaches remains challenging33. (2007).
Further enhancement of experimental techniques and theoretical 19. Religa, T.L., Markson, J.S., Mayor, U., Freund, S.M.V. & Fersht, A.R. Solution
structure of a protein denatured state and folding intermediate. Nature 437,
approaches are needed so that each can be used to better model, 1053–1056 (2005).
refine and test the outputs of the other. In addition, the folding 20. Gsponer, J. et al. Determination of an ensemble of structures representing the
mechanisms of only the simplest of proteins have been studied intermediate state of the bacterial immunity protein Im7. Proc. Natl. Acad. Sci.
USA 103, 99–104 (2006).
biophysically in detail so far. Larger proteins, chemically modified 21. Friel, C.T., Smith, D.A., Vendruscolo, M., Gsponer, J. & Radford, S.E. The mechanism
© 2009 Nature America, Inc. All rights reserved.
proteins, protein complexes and membrane proteins, which together of formation of a folding intermediate reveals the competition between functional and
kinetic evolutionary constraints. Nat. Struct. Mol. Biol. 16, 318–324 (2009).
comprise most of nature’s proteins, remain largely uncharacterized. 22. Salvatella, X., Dobson, C.M., Fersht, A.R. & Vendruscolo, M. Determination of the
Other future challenges include understanding folding in the context folding transition states of barnase by using phi(I)-value-restrained simulations
of interconverting ensembles and landscape theory in the living cell, validated by double mutant phi(IJ)-values. Proc. Natl. Acad. Sci. USA 102,
12389–12394 (2005).
where folding may be assisted by chaperones, challenged by stochastic 23. Calosci, N. et al. Comparison of successive transition states for folding reveals
events such as aggregation and modulated by the cellular status at any alternative early folding pathways of two homologous proteins. Proc. Natl. Acad.
moment in time. Armed with the arsenal of methods and concepts Sci. USA 105, 19241–19246 (2008).
24. Korzhnev, D.M. et al. Low-populated folding intermediates of Fyn SH3 characterized
about folding landscapes derived from studies in vitro over recent by relaxation dispersion NMR. Nature 430, 586–590 (2004).
years and fueled by the increasing awareness of the importance 25. Neudecker, P., Zarrine-Afsar, A., Davidson, A.R. & Kay, L.E. Phi-value analysis of a
three-state protein folding pathway by NMR relaxation dispersion spectroscopy. Proc.
of protein folding in the context of homeostasis and disease, Natl. Acad. Sci. USA 104, 15717–15722 (2007).
major advances toward this goal are sure to be realized in the 26. Roder, H., Maki, K. & Cheng, H. Early events in protein folding explored by rapid
forthcoming years. mixing methods. Chem. Rev. 106, 1836–1861 (2006).
27. Dill, K.A., Ozkan, S.B., Shell, M.S. & Weikl, T.R. The protein folding problem. Annu.
Rev. Biophys. 37, 289–316 (2008).
Note added in proof: Two recent articles106,107 have described innova- 28. Korzhnev, D.M. & Kay, L.E. Probing invisible, low-populated states of protein
tive NMR approaches to determine protein structure and dynamics in molecules by relaxation dispersion NMR spectroscopy: an application to protein
folding. Acc. Chem. Res. 41, 442–451 (2008).
living cells. 29. Nettels, D., Hoffmann, A. & Schuler, B. Unfolded protein and peptide dynamics
investigated with single-molecule FRET and correlation spectroscopy from pico-
ACKNOWLEDGMENTS seconds to seconds. J. Phys. Chem. B 112, 6137–6146 (2008).
We thank D. Brockwell for critical comments and many helpful insights. We also 30. Hoffmann, A. et al. Mapping protein collapse with single-molecule fluorescence and
acknowledge, with grateful thanks, members of our group, our collaborators past kinetic synchrotron radiation circular dichroism spectroscopy. Proc. Natl. Acad. Sci.
and present, and M. Oliveberg and J. Gsponer for their helpful discussions. A.I.B. USA 104, 105–110 (2007).
31. Va´rnai, P., Dobson, C.M. & Vendruscolo, M. Determination of the transition state
is supported by the UK Biotechnology and Biological Sciences Research Council
ensemble for the folding of ubiquitin from a combination of Phi and Psi analyses.
(BB/526502/1). J. Mol. Biol. 377, 575–588 (2008).
32. Allen, L.R. & Paci, E. Transition states for protein folding using molecular
Published online at http://www.nature.com/nsmb/ dynamics and experimental restraints. J. Phys. Condens. Matter 19, 285211
Reprints and permissions information is available online at http://npg.nature.com/ (2007).
reprintsandpermissions/ 33. van Gunsteren, W.F., Dolenc, J. & Mark, A.E. Molecular simulation as an aid to
experimentalists. Curr. Opin. Struct. Biol. 18, 149–153 (2008).
34. Oroguchi, T. et al. Atomically detailed description of the unfolding of a-lactalbumin by
1. Fersht, A.R. From the first protein structures to our current knowledge of protein the combined use of experiments and simulations. J. Mol. Biol. 354, 164–172
folding: delights and scepticisms. Nat. Rev. Mol. Cell Biol. 9, 650–654 (2008). (2005).
2. Kubelka, J., Hofrichter, J. & Eaton, W.A. The protein folding ‘speed limit’. Curr. Opin. 35. Marsh, J.A. et al. Improved structural characterizations of the drkN SH3 domain
Struct. Biol. 14, 76–88 (2004). unfolded state suggest a compact ensemble with native-like and non-native structure.
3. Karplus, M. The Levinthal paradox: yesterday and today. Fold. Des. 2, S69–S75 (1997). J. Mol. Biol. 367, 1494–1510 (2007).
4. Brockwell, D.J. & Radford, S.E. Intermediates: ubiquitous species on folding energy 36. Chugha, P. & Oas, T.G. Backbone dynamics of the monomeric l repressor denatured
landscapes? Curr. Opin. Struct. Biol. 17, 30–37 (2007). state ensemble under nondenaturing conditions. Biochemistry 46, 1141–1151
5. Bryngelson, J.D., Onuchic, J.N., Socci, N.D. & Wolynes, P.G. Funnels, pathways, and (2007).
the energy landscape of protein-folding—a synthesis. Proteins 21, 167–195 (1995). 37. Le Duff, C.S., Whittaker, S.B.M., Radford, S.E. & Moore, G.R. Characterisation of the
6. Onuchic, J.N. & Wolynes, P.G. Theory of protein folding. Curr. Opin. Struct. Biol. 14, conformational properties of urea-unfolded Im7: implications for the early stages of
70–75 (2004). protein folding. J. Mol. Biol. 364, 824–835 (2006).
7. Ferreiro, D.U., Hegler, J.A., Komives, E.A. & Wolynes, P.G. Localizing frustration in 38. Kristjansdottir, S. et al. Formation of native and non-native interactions in ensembles
native proteins and protein assemblies. Proc. Natl. Acad. Sci. USA 104, 19819– of denatured ACBP molecules from paramagnetic relaxation enhancement studies.
19824 (2007). J. Mol. Biol. 347, 1053–1062 (2005).
8. Hagen, S.J. Probe-dependent and nonexponential relaxation kinetics: unreliable 39. Kohn, J.E. et al. Random-coil behavior and the dimensions of chemically unfolded
signatures of downhill protein folding. Proteins 68, 205–217 (2007). proteins. Proc. Natl. Acad. Sci. USA 101, 12491–12496 (2004).
9. Gruebele, M. Comment on probe-dependent and nonexponential relaxation kinetics: 40. Fierz, B. et al. Loop formation in unfolded polypeptide chains on the picoseconds
unreliable signatures of ‘downhill’ protein folding. Proteins 70, 1099–1102 (2008). to microseconds time scale. Proc. Natl. Acad. Sci. USA 104, 2163–2168 (2007).
10. Knott, M. & Chan, H.S. Criteria for downhill protein folding: calorimetry, chevron plot, 41. Fersht, A.R., Matouschek, A. & Serrano, L. The folding of an enzyme: I. Theory of
kinetic relaxation, and single-molecule radius of gyration in chain models with protein engineering analysis of stability and pathway of protein folding. J. Mol. Biol.
subdued degrees of cooperativity. Proteins 65, 373–391 (2006). 224, 771–782 (1992).
NATURE STRUCTURAL & MOLECULAR BIOLOGY VOLUME 16 NUMBER 6 JUNE 2009 587

7. REVIEW
42. Zarrine-Afsar, A. & Davidson, A.R. The analysis of protein folding kinetic data 76. Engel, R. et al. Macromolecular crowding compacts unfolded apoflavodoxin and
produced in protein engineering experiments. Methods 34, 41–50 (2004). causes severe aggregation of the off-pathway intermediate during apoflavodoxin
43. Sa´nchez, I.E. & Kiefhaber, T. Origin of unusual j-values in protein folding: evidence folding. J. Biol. Chem. 283, 27383–27394 (2008).
against specific nucleation sites. J. Mol. Biol. 334, 1077–1085 (2003). 77. Zhou, H.-X., Rivas, G. & Minton, A.P. Macromolecular crowding and confinement:
44. Fersht, A.R. & Sato, S. F-value analysis and the nature of protein-folding transition biochemical, biophysical, and potential physiological consequences. Annu. Rev.
states. Proc. Natl. Acad. Sci. USA 101, 7976–7981 (2004). Biophys. 37, 375–397 (2008).
45. Cho, J.-H. & Raleigh, D.P. Denatured state effects and the origin of nonclassical 78. Mittal, J. & Best, R.B. Thermodynamics and kinetics of protein folding under
values in protein folding. J. Am. Chem. Soc. 128, 16492–16493 (2006). confinement. Proc. Natl. Acad. Sci. USA 105, 20233–20238 (2008).
46. Dobson, C.M. Protein folding and misfolding. Nature 426, 884–890 (2003). 79. Cheung, M.S. & Thirumalai, D. Effects of crowding and confinement on the structures
47. Zhou, Z. & Bai, Y. Detection of a hidden folding intermediate in the focal adhesion of the transition state ensemble in proteins. J. Phys. Chem. B 111, 8250–8257
target domain: implications for its function and folding. Proteins 65, 259–265 (2007).
(2006). 80. Homouz, D., Perham, M., Samiotakis, A., Cheung, M.S. & Wittung-Stafshede, P.
48. Rhoades, E., Cohen, M., Schuler, B. & Haran, G. Two-state folding observed in Crowded, cell-like environment induces shape changes in aspherical protein. Proc.
individual protein molecules. J. Am. Chem. Soc. 126, 14686–14687 (2004). Natl. Acad. Sci. USA 105, 11754–11759 (2008).
49. Kuzmenkina, E.V., Heyes, C.D. & Nienhaus, G.U. Single-molecule Fo ¨rster resonance 81. Hartl, F.U. & Hayer-Hartl, M. Converging concepts of protein folding in vitro and
energy transfer study of protein dynamics under denaturing conditions. Proc. Natl. in vivo. Nat. Struct. Mol. Biol. 16, 574–581 (2009).
Acad. Sci. USA 102, 15471–15476 (2005). 82. Kramer, G., Boehringer, D., Ban, N. & Bukau, B. The ribosome as a platform for
50. Cecconi, C., Shank, E.A., Bustamante, C. & Marqusee, S. Direct observation of the cotranslational processing, folding and targeting of newly synthesized proteins.
three-state folding of a single protein molecule. Science 309, 2057–2060 (2005). Nat. Struct. Mol. Biol. 16, 589–597 (2009).
51. Mickler, M. et al. Revealing the bifurcation in the unfolding pathways of GFP by using 83. Hsu, S.-T.D. et al. Structure and dynamics of a ribosome-bound nascent chain by
single-molecule experiments and simulations. Proc. Natl. Acad. Sci. USA 104, NMR spectroscopy. Proc. Natl. Acad. Sci. USA 104, 16516–16521 (2007).
20268–20273 (2007). 84. James, J.R. et al. Single-molecule level analysis of the subunit composition of the
52. Zarrine-Afsar, A., Larson, S.M. & Davidson, A.R. The family feud: do proteins with T cell receptor on live T cells. Proc. Natl. Acad. Sci. USA 104, 17662–17667
similar structures fold via the same pathway? Curr. Opin. Struct. Biol. 15, 42–49 (2007).
(2005). 85. Elf, J., Li, G.-W. & Xie, X.S. Probing transcription factor dynamics at the single-
53. Spudich, G.M., Miller, E.J. & Marqusee, S. Destabilization of the Escherichia coli molecule level in a living cell. Science 316, 1191–1194 (2007).
© 2009 Nature America, Inc. All rights reserved.
RNase H kinetic intermediate: switching between a two-state and three-state folding 86. Woolhead, C.A., McCormick, P.J. & Johnson, A.E. Nascent membrane and secretory
mechanism. J. Mol. Biol. 335, 609–618 (2004). proteins differ in FRET-detected folding far inside the ribosome and in their exposure
54. Plaxco, K.W., Simons, K.T. & Baker, D. Contact order, transition state placement and to ribosomal proteins. Cell 116, 725–736 (2004).
the refolding rates of single domain proteins. J. Mol. Biol. 277, 985–994 (1998). 87. Hillger, F. et al. Probing protein-chaperone interactions with single-molecule fluores-
55. Jackson, S.E. & Fersht, A.R. Folding of chymotrypsin inhibitor 2. 1. Evidence for a cence spectroscopy. Angew. Chem. Int. Ed. Engl. 47, 6184–6188 (2008).
two-state transition. Biochemistry 30, 10428–10435 (1991). 88. Sharma, S. et al. Monitoring protein conformation along the pathway of chaperonin-
56. Radford, S.E., Dobson, C.M. & Evans, P.A. The folding of hen lysozyme involves assisted folding. Cell 133, 142–153 (2008).
partially structured intermediates and multiple pathways. Nature 358, 302–307 89. McNulty, B.C., Young, G.B. & Pielak, G.J. Macromolecular crowding in the Escher-
(1992). ichia coli periplasm maintains a-synuclein disorder. J. Mol. Biol. 355, 893–897
57. Wright, C.F., Lindorff-Larsen, K., Randles, L.G. & Clarke, J. Parallel protein-unfolding (2006).
pathways revealed and mapped. Nat. Struct. Mol. Biol. 10, 658–662 (2003). 90. Selenko, P., Serber, Z., Gadea, B., Ruderman, J. & Wagner, G. Quantitative NMR
58. Eckert, B., Martin, A., Balbach, J. & Schmid, F.X. Prolyl isomerization as a molecular analysis of the protein G B1 domain in Xenopus laevis egg extracts and intact oocytes.
timer in phage infection. Nat. Struct. Mol. Biol. 12, 619–623 (2005). Proc. Natl. Acad. Sci. USA 103, 11904–11909 (2006).
59. Crespo, M.D., Simpson, E.R. & Searle, M.S. Population of on-pathway intermediates 91. Ignatova, Z. & Gierasch, L.M. Monitoring protein stability and aggregation in vivo
in the folding of ubiquitin. J. Mol. Biol. 360, 1053–1066 (2006). by real-time fluorescent labeling. Proc. Natl. Acad. Sci. USA 101, 523–528 (2004).
60. Weissman, J.S. & Kim, P.S. A kinetic explanation for the rearrangement pathway of 92. Ignatova, Z. & Gierasch, L.M. Inhibition of protein aggregation in vitro and in vivo
BPTI folding. Nat. Struct. Biol. 2, 1123–1130 (1995). by a natural osmoprotectant. Proc. Natl. Acad. Sci. USA 103, 13357–13361
61. Lindberg, M.O. & Oliveberg, M. Malleability of protein folding pathways: a simple (2006).
reason for complex behaviour. Curr. Opin. Struct. Biol. 17, 21–29 (2007). ¨
93. Horst, R., Fenton, W.A., Englander, S.W., Wuthrich, K. & Horwich, A.L. Folding
62. Lindberg, M.O., Haglund, E., Hubner, I.A., Shakhnovich, E.I. & Oliveberg, M. trajectories of human dihydrofolate reductase inside the GroEL-GroES chaperonin
Identification of the minimal protein-folding nucleus through loop-entropy perturba- cavity and free in solution. Proc. Natl. Acad. Sci. USA 104, 20788–20792 (2007).
tions. Proc. Natl. Acad. Sci. USA 103, 4083–4088 (2006). 94. Dennis, C.A. et al. A structural comparison of the colicin immunity proteins Im7 and
63. Haglund, E., Lindberg, M.O. & Oliveberg, M. Changes of protein folding pathways by Im9 gives new insights into the molecular determinants of immunity-protein speci-
circular permutation—overlapping nuclei promote global cooperativity. J. Biol. Chem. ficity. Biochem. J. 333, 183–191 (1998).
283, 27904–27915 (2008). 95. Noble, M.E.M., Musacchio, A., Saraste, M., Courtneidge, S.A. & Wierenga, R.K.
64. Ferreiro, D.U., Walczak, A.M., Komives, E.A. & Wolynes, P.G. The energy landscapes Crystal structure of the SH3 domain in human Fyn—comparison of the three-
of repeat-containing proteins: topology, cooperativity, and the folding funnels of one- dimensional structures of SH3 domains in tyrosine kinases and spectrin. EMBO J.
dimensional architectures. PLOS Comput. Biol. 4, e1000070 (2008). 12, 2617–2624 (1993).
65. Lowe, A.R. & Itzhaki, L.S. Rational redesign of the folding pathway of a modular 96. Lindahl, M. et al. Crystal structure of the ribosomal protein S6 from Thermus
protein. Proc. Natl. Acad. Sci. USA 104, 2679–2684 (2007). thermophilus. EMBO J. 13, 1249–1254 (1994).
66. Tripp, K.W. & Barrick, D. Rerouting the folding pathway of the notch ankyrin 97. Royer, C.A. Probing protein folding and conformational transitions with fluorescence.
domain by reshaping the energy landscape. J. Am. Chem. Soc. 130, 5681–5688 Chem. Rev. 106, 1769–1784 (2006).
(2008). 98. Kelly, S.M., Jess, T.J. & Price, N.C. How to study proteins by circular dichroism.
67. Watters, A.L. et al. The highly cooperative folding of small naturally occurring proteins Biochim. Biophys. Acta 1751, 119–139 (2005).
is likely the result of natural selection. Cell 128, 613–624 (2007). 99. Balakrishnan, G., Weeks, C.L., Ibrahim, M., Soldatova, A.V. & Spiro, T.G. Protein
68. Tartaglia, G.G., Pechmann, S., Dobson, C.M. & Vendruscolo, M. Life on the edge: dynamics from time resolved UV Raman spectroscopy. Curr. Opin. Struct. Biol. 18,
a link between gene expression levels and aggregation rates of human proteins. 623–629 (2008).
Trends Biochem. Sci. 32, 204–206 (2007). 100. Fabian, H. & Naumann, D. Methods to study protein folding by stopped-flow FT-IR.
69. Ja¨ger, M. et al. Structure-function-folding relationship in a WW domain. Proc. Natl. Methods 34, 28–40 (2004).
Acad. Sci. USA 103, 10648–10653 (2006). 101. Haustein, E. & Schwille, P. Fluorescence correlation spectroscopy: novel variations of
70. Gosavi, S., Whitford, P.C., Jennings, P.A. & Onuchic, J.N. Extracting function from a an established technique. Annu. Rev. Biophys. Biomol. Struct. 36, 151–169 (2007).
b-trefoil folding motif. Proc. Natl. Acad. Sci. USA 105, 10384–10389 (2008). 102. Lipfert, J. & Doniach, S. Small-angle X-ray scattering from RNA, proteins, and protein
71. Capraro, D.T., Roy, M., Onuchic, J.N. & Jennings, P.A. Backtracking on the folding complexes. Annu. Rev. Biophys. Biomol. Struct. 36, 307–327 (2007).
landscape of the b-trefoil protein interleukin-1b? Proc. Natl. Acad. Sci. USA 105, 103. Dyson, H.J. & Wright, P.E. Elucidation of the protein folding landscape by NMR.
14844–14848 (2008). Methods Enzymol. 394, 299–321 (2005).
72. Balch, W.E., Morimoto, R.I., Dillin, A. & Kelly, J.W. Adapting proteostasis for disease 104. Krishna, M.M.G., Hoang, L., Lin, Y. & Englander, S.W. Hydrogen exchange methods to
intervention. Science 319, 916–919 (2008). study protein folding. Methods 34, 51–64 (2004).
73. Ellis, R.J. Macromolecular crowding: obvious but underappreciated. Trends Biochem. 105. Maier, C.S. & Deinzer, M.L. Protein conformations, interactions, and H/D exchange.
Sci. 26, 597–604 (2001). Methods Enzymol. 402, 312–360 (2005).
74. Saibil, H.R. Chaperone machines in action. Curr. Opin. Struct. Biol. 18, 35–42 106. Sakakibara, D. et al. Protein structure determination in living cells by in-cell NMR
(2008). spectroscopy. Nature 458, 102–105 (2009).
75. Charlton, L.M. et al. Residue-level interrogation of macromolecular crowding effects 107. Inomata, K. et al. High-resolution multi-dimensional NMR spectroscopy of proteins in
on protein stability. J. Am. Chem. Soc. 130, 6826–6830 (2008). human cells. Nature 458, 106–109 (2009).
588 VOLUME 16 NUMBER 6 JUNE 2009 NATURE STRUCTURAL & MOLECULAR BIOLOGY