Dual energy landscape: The functional state of the β-barrel outer membrane protein G molds its unfolding energy landscape

RESEARCH ARTICLE

Dual energy landscape: The functional state of the

b-barrel outer membrane protein G molds its unfolding

energy landscape

Mehdi Damaghi1,2�, K. Tanuj Sapra2,3�, Stefan Koster4, Ozkan Yildiz4, Werner K .uhlbrandt4

and Daniel J. Muller1,2

1 ETH Z .urich, Department of Biosystems Science and Engineering, Basel, Switzerland2 Biotechnology Center, University of Technology, Tatzberg, Dresden, Germany3 Chemistry Research Laboratory, University of Oxford, Oxford, UK4 Max-Planck-Institute of Biophysics, Department of Structural Biology, Frankfurt am Main, Germany

Received: April 13, 2010

Revised: May 16, 2010

Accepted: May 17, 2010

We applied dynamic single-molecule force spectroscopy to quantify the parameters (free energy

of activation and distance of the transition state from the folded state) characterizing the energy

barriers in the unfolding energy landscape of the outer membrane protein G (OmpG) from

Escherichia coli. The pH-dependent functional switching of OmpG directs the protein along

different regions on the unfolding energy landscape. The two functional states of OmpG take

the same unfolding pathway during the sequential unfolding of b-hairpins I–IV. After the

initial unfolding events, the unfolding pathways diverge. In the open state, the unfolding of

b-hairpin V in one step precedes the unfolding of b-hairpin VI. In the closed state, b-hairpin V

and b-strand S11 with a part of extracellular loop L6 unfold cooperatively, and subsequently

b-strand S12 unfolds with the remaining loop L6. These two unfolding pathways in the open

and closed states join again in the last unfolding step of b-hairpin VII. Also, the conformational

change from the open to the closed state witnesses a rigidified extracellular gating loop L6.

Thus, a change in the conformational state of OmpG not only bifurcates its unfolding pathways

but also tunes its mechanical properties for optimum function.

Keywords:

Atomic force microscopy / Interactions / Mechanical properties / Nanoproteomics / pH

gating / Single-molecule force spectroscopy

1 Introduction

Outer membrane proteins (Omps) are found in the outer

membranes of Gram-negative bacteria, mitochondria, and

chloroplasts. These b-barrel-forming transmembrane proteins

are imperative to a cell’s survival owing to their function in

controlling the transport of solutes in and out of a cell.

Whereas some Omps, like OmpC, OmpF, and OmpG, are

non-selective in what they transport through their pores, others

like LamB, FhuA, BtuB, are solute specific [1–6]. The gating

mechanisms of the b-barrel forming Omps attract continuous

interest and remain to be investigated in detail [1]. Several of

the transmembrane pores formed by Omps of Escherichia coliare pH gated. Low pH induces the closing of the pores of, for

example, OmpC, OmpF, OmpG, LamB, and PhoE [2–6]. What

conformational changes drive pore closure has long been

debated [1, 7]. In 1999, high-resolution atomic force micro-

scopy (AFM) imaging, for the first time, showed that at low pH

the large extracellular loops of the OmpF collapsed onto theAbbreviations: aa, amino acids; AFM, atomic force microscopy;

DFS, dynamic single-molecule force spectroscopy; F-D curve,

single-molecule force-distance curve; Omp, outer membrane

protein; SMFS, single-molecule force spectroscopy; WLC, worm-

like-chain

�These authors contributed equally to this work.

Colour Online: See the article online to view fig. 3 in colour.

Correspondence: Professor Daniel J. Muller, ETH Z .urich,

Department of Biosystems Science and Engineering, Matten-

strasse 26, 4058 Basel, Switzerland

E-mail: [email protected]

Fax: 141-61-387-39-94

& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com

Proteomics 2010, 10, 4151–4162 4151DOI 10.1002/pmic.201000241

pore entrance [8]. This supported the hypothesis that the Omp

pores are gated by the conformational changes of the flexible

extracellular loops. Experiments on the maltoporin LamB from

E. coli, which is specific for malto-oligosaccharides, corrobo-

rated the gating model. When lacking the major extracellular

loops L4 and L6, LamB failed to close at lower pH [9].

Among the pH-gated Omps, the structure and function

relationship of the OmpG from E. coli represents possibly the

best-studied example. The OmpG structure solved by X-ray

crystallography [10, 11] and NMR [12] comprises 14 b-strands

(S1–S14) that form a transmembrane b-barrel. On the peri-

plasmic side the b-strands are connected by six short poly-

peptide turns (T1–T6). On the extracellular side the b-strands

are connected by seven longer loops (L1–L7) that exhibit

enhanced intrinsic flexibility [11, 12]. A pH-dependent gating

controls the flux of small molecules through the OmpG pore

[6]. X-ray structures obtained from three-dimensional OmpG

crystals grown at neutral (pH 7.5) and acidic (pH 5.6) pH

provide insight into the conformational changes that may

guide the gating mechanism [10]. At low pH, the largest

extracellular loop L6 folds into the pore, thereby constricting

its entrance. However, the three-dimensional crystals of

solubilized OmpG grown at different pHs showed different

packing arrangements, and some extracellular loops formed

crystal contacts with adjacent OmpG molecules. Thus, it may

be assumed that the conformations observed may not

represent those that naturally occur in the gating mechanism

of OmpG. To test this hypothesis, OmpG was reconstituted

in native E. coli lipids and imaged by high-resolution AFM in

buffer solution at room temperature [13]. The AFM topo-

graphs confirmed that the pH-dependent gating mechanism

suggested from the X-ray structures indeed occurred in

physiological conditions.

Besides the gating mechanisms of the Omps, the

mechanisms that guide their folding and unfolding are of

pertinent interest [14–20]. So far most experiments investi-

gating the folding of Omps have first denatured Omps in

detergent and/or urea and then characterized the refolding

into a lipid bilayer or in a detergent [14, 21–24]. Such bulk

unfolding experiments suggest that OmpG unfolds and

refolds reversibly [22]. The folding process of Omps is

described as being coupled with membrane insertion [17]. A

folding and insertion process has been recently described for

the b-barrel transmembrane protein PagP from E. coli [25].

PagP solubilized and denatured in 10 M urea is found to

adsorb to the lipid headgroups of the bilayer, where it forms a

transition state that tilts and inserts into the lipid membrane

to complete the folding process. Apparently, these models

contrast with the results from single-molecule force spectro-

scopy (SMFS) in which single OmpG molecules have been

mechanically stressed to induce their unfolding from the

native lipid membrane in a buffer solution [26]. These single-

molecule experiments clearly show that OmpG molecules

unfold via many sequential unfolding intermediates

describing a detailed unfolding pathway. The unfolding step

of a single b-hairpin characterizes the transition from one

unfolding intermediate to the next one. However, the differ-

ences between chemical denaturation and refolding and

mechanical unfolding experiments may have different

origins. First, one may assume that the refolding mechanism

in the absence of any external force does not reflect unfolding

under an applied force. Second, it may be that the experi-

mental conditions alter the unfolding and folding pathways

chosen by b-barrel membrane proteins. In case of a-helical

transmembrane proteins it has been shown that alterations in

the temperature and buffer solution within the physiological

relevant range can considerably modify their unfolding

pathways [27, 28]. Therefore, it is not surprising that the

exposure of membrane proteins to urea, detergent, and

mechanical stress may force them along very different

unfolding and folding pathways.

SMFS has been particularly successful in characterizing

the unfolding pathways of membrane proteins and to

quantify the interactions and energies of the intermediates

in the unfolding pathways [28, 29]. SMFS provides detailed

insights into the nature of molecular interactions and most

importantly allows to locate and quantify these interactions

structurally with an accuracy of E2–6 amino acids (aa). In

this work we have performed dynamic SMFS (DFS) to probe

the strength of the interactions that stabilize the unfolding

intermediates of OmpG at different loading rates (applied

force over time). The dependence of these interaction

strengths on the loading rate allows quantifying the

unfolding energy barriers of the intermediates [30, 31].

These measurements provide the position of the transition

state, the transition rate of the intermediate from the folded

to the unfolded state, and the energy of activation to cross

the transition barrier. Because the sensitivity of SMFS

permits to directly determine the sequence at which the

unfolding barriers are located along an unfolding pathway,

we can chart the unfolding energy landscape of OmpG in

the two pH-dependent conformational and functional states.

The energy landscapes reveal detailed insights into how

interactions can change the unfolding pathways and the

gating mechanism of OmpG. We show that the molecular

interactions associated with a change in pH not only drive a

conformational change but also influence the mechanical

properties of the region responsible for the conformational

change. We propose that the functional properties of the

protein are related to its mechanical properties.

2 Materials and methods

2.1 SMFS and DFS

OmpG was purified from inclusion bodies, refolded in

detergent, and reconstituted into native E. coli lipids [10].

Membranes showed the OmpG molecules being densely

packed and assembled into two-dimensional crystals. These

OmpG membranes were adsorbed onto freshly cleaved mica

(E30 min) in buffer solution (pH 7.0, 25 mM Tris-HCl,

4152 M. Damaghi et al. Proteomics 2010, 10, 4151–4162


25 mM MgCl2, 300 mM NaCl or pH 5.0, 25 mM Na-acetate,

25 mM MgCl2, 300 mM NaCl). The adsorbed membranes

were localized by AFM in the same buffer solution at room

temperature [32]. For SMFS, the AFM cantilever tip (60 mm

long Biolever, Olympus) was pushed onto the OmpG

membrane applying forces E500–750 pN for E500 ms. In

E0.1% of all cases the OmpG terminus attached to the

AFM tip. Then the AFM tip was retracted at a specific

pulling velocity to induce unfolding. A force-distance (F-D)

curve recorded the forces required to overcome the interac-

tion strengths that stabilized the unfolding intermediates of

the membrane protein. F-D spectra recorded from OmpG,

which was either densely packed or crystallized two-

dimensionally showed no difference in the force pattern.

DFS experiments were performed at seven pulling velocities

(100, 300, 600, 900, 1200, 2500, and 5000 nm/s). Before

and after each experiment the spring constant of each

cantilever (E0.03 N/m) was estimated from its thermal

noise using the equipartition theorem [33]. To minimize

errors that may occur due to uncertainties in the

cantilever spring constant calibration, OmpG was unfolded

using at least three different cantilevers for each pulling

velocity.

2.2 Data selection and analysis

We analyzed only F-D curves that corresponded to the

length (475 nm) of a fully stretched and unfolded OmpG

polypeptide (281 aa). This selection criterion ensured that

OmpG was mechanically unfolded by stretching one of its

termini [28]. Previously we have shown that OmpG predo-

minantly attaches with its N-terminal end to the AFM tip

[26]. All F-D curves having a length of 475 nm showed an F-

D pattern similar to those published previously [26, 34].

Thus, we could conclude that the F-D curves were recorded

upon mechanically unfolding of OmpG from the N-terminal

end. For analysis each force peak of each F-D curve was

fitted using the worm-like-chain (WLC) model to reveal the

contour lengths of the unfolded polypeptides [28]. Deter-

mining the unfolded polypeptide stretches allowed assign-

ing the structural regions that form stable unfolding

intermediates [26, 28].

2.3 Calculating xu and k0 from DFS data

According to the Bell–Evans theory [35, 36], the

most probable unfolding force F� plotted versus ln ðr�f Þdescribes the most prominent unfolding energy barriers

that have been crossed along the force-driven reaction

coordinate [30]. The relation between F� and r�f can be

described by:

F� ¼kBT

xuln

rur�fkBTk0

� �ð1Þ

where kB is the Boltzmann constant, T the absolute

temperature, r�f the loading rate, xu the distance between the

free energy minimum of the folded intermediate state and

transition state barrier, and k0 the unfolding rate of the

intermediate at zero force. r�f is the product of pulling

velocity, v, and the slope of the WLC fit at each force peak.

Experimental loading rate and force histograms (Supporting

Information Figs. S1 and S2) were fitted with Gaussian

distributions. The resulting F� was semi-logarithmically

plotted versus r�f . xu and k0 were obtained by fitting Eq. (1)

using a non-linear least-squares algorithm. Only unfolding

forces and loading rates corresponding to the main force

peaks were considered for analysis.

2.4 Calculating transition barrier height and rigidity

The height of the free energy barrier, DGz, separating the

folded and the unfolded states was assessed using an

Arrhenius-like expression:

DGz ¼ �kBT lnðtDk0Þ ð2Þ

where tD denotes the diffusive relaxation time. Typical

values for tD found for proteins are in the order of

10�7–10�9 s [37, 38]. Therefore, assuming tD 5 10�9 s seems

to be reasonable for determining the free energy barrier

heights. This value has also been used for molecular

dynamics simulations of protein folding [39, 40]. We have

used tD 5 10�9 s in all our calculations. Varying tD within

the above-mentioned range changes the free energy of

activation by o15%. Moreover, even if tD was wrong by

orders of magnitude, the influence on the error of tD would

be the same for all conditions and values and, hence, would

not affect the qualitative results. Errors in DGz were esti-

mated by propagation of the errors of k0.

Without any information on the energy potential shape,

we assumed the shape of the energy potential to be a simple

parabola. Hence, the spring constant k of the folded struc-

ture was calculated using DGz and xu [41, 42]:

k ¼2DGz

x2u

ð3Þ

Errors in DGz and xu were propagated for estimation of

errors in k.

3 Results and discussion

3.1 The functional state of OmpG directs its

unfolding route

In previous experiments we have established SMFS-based

unfolding of single OmpG molecules that were recon-

stituted into membranes of native E. coli lipids [26]. In those

experiments we first localized OmpG membranes by AFM

imaging [13, 32]. Then the AFM tip was pressed onto OmpG

Proteomics 2010, 10, 4151–4162 4153


to facilitate the non-specific attachment of the terminal end

(Fig. 1A). The non-specific attachment of OmpG occurred

predominantly via its N-terminal end to the AFM tip [26].

Withdrawal of the AFM tip stretched the polypeptide and

induced the unfolding of OmpG. During withdrawal we

recorded F-D curves that showed a series of force peaks

occurring at similar positions. The superimposition of all

F-D curves showed a clear pattern with the predominant

occurrence of force peaks (Supporting Information Fig. S1).

Every force peak of an F-D curve reflects the strength of an

interaction that has been established by an OmpG unfolding

intermediate [26, 28]. Fitting a force peak using the WLC

model allowed estimating the length of the unfolded poly-

peptide and locating the structural region at which this

Figure 1. Unfolding intermediates of OmpG

detected at pH 7.0 and 5.0 using SMFS. (A)

Illustration of the SMFS experiment. OmpG

reconstituted into native E. coli lipids was

adsorbed to mica in buffer solution at room

temperature. Single OmpGs were non-

specifically attached by their N-terminal end

to the tip of the AFM cantilever [26]. Retrac-

tion of the AFM tip stretched polypeptide

and induced the unfolding of OmpG. F-D

curves recorded showed a saw-tooth-like

pattern of force peaks (Supporting Informa-

tion Fig. S1). Each of these force peaks

recorded an unfolding event of OmpG. In

previous studies [26, 34] we have used these

force peaks to assign the unfolding inter-

mediates and to reconstruct the unfolding

pathways of OmpG exposed to (B) pH 7.0

(25 mM Tris-HCl, 25 mM MgCl2, 300 mM

NaCl) and (C) pH 5.0 (25 mM Na-acetate,

25 mM MgCl2, 300 mM NaCl). Yellow-circled

aa (positions labeled) indicate structural

regions at which interactions stabilizing the

unfolding intermediates were detected in the

F-D curves. The interaction located at 224 aa

(pH 5.0) is detected by the force peak at a

contour length of 213 aa (Supporting Infor-

mation Fig. S1). Because the contour length

of 213 aa suggests that the interaction lies at

the membrane surface opposite to the pull-

ing AFM tip, a membrane compensation of

E11 aa had to be added to locate the inter-

action [34, 60]. A mechanical force applied to

the N-terminal end initiated unfolding of the

first b-hairpin (red colored), with the next

unfolding step being established by the

second b-hairpin (green colored) and so

forth. (B) At pH 7.0 single b-hairpins (equally

colored) formed the predominant unfolding

steps (equally colored b-strands) guiding

one unfolding intermediate (remaining

folded structures) to the next. (C) In contrast

to unfolding at pH 7.0, at pH 5.0 the b-strands

S9, S10, and S11 formed one unfolding step

(purple colored in (C)) and b-strand S12 by

itself (bright blue colored in (C)) formed one.



interaction was established (Fig. 1B). The sequence of force

peaks, or unfolding intermediates, was used to assign the

unfolding pathways of single OmpG molecules [26].

At neutral pH OmpG predominantly unfolds one

b-hairpin after the other until the entire membrane

protein has been unfolded [26], i.e. the unfolding inter-

mediates of OmpG were formed by unfolding subsequent

b-hairpins in single steps (equally colored b-strands in

Fig. 1B). Similarly, at pH 5.0, which induces pore closure

[10, 13], most b-strands unfolded with an adjacent b-strand

as b-hairpin. A noticeable exception was observed for the

intermediate composed of b-hairpins V (b-strands S9

and S10) and VI (b-strands S11 and S12) in the closed

state of OmpG; b-strands S9, S10, and S11 unfolded in one

step followed by the unfolding of b-strand S12 in

a single step (Fig. 1C). This structural region of OmpG is

also known to undergo a pH-dependent conformational

change. Thus, the occurrence of new intermediates in the

closed state exhibiting a bifurcation of the unfolding path-

ways may have its origin in the conformational states of the

protein.

3.2 Quantifying the unfolding energy barriers of

OmpG in the closed and open states

In our previous work we have used SMFS to characterize the

unfolding pathways and interactions of OmpG in the open

and closed states [26, 34]. Here, we probe the strength of the

interactions that stabilize the unfolding intermediates of

OmpG at different loading rates by using DFS. This data is

used to quantify the parameters that define the energy

barriers (Fig. 2) of the unfolding intermediates of OmpG in

the open (at pH 7.0) and closed (at pH 5.0) conformations.

We unfolded single OmpGs at pulling speeds of 100, 300,

600, 900, 1200, 2500, and 5000 nm/s, and superimposed the

F-D curves recorded at each speed (Supporting Information

Fig. S1). The F-D curves of OmpG unfolded in the open

state showed seven main force peaks at contour lengths of 8,

43, 83, 126, 166, 204, and 246 aa, and the F-D spectra of

OmpG unfolded in the closed state showed seven main

force peaks at contour lengths of 8, 43, 83, 126, 166, 213, and

246 aa. For both the open and the closed states, the positions

of these force peaks did not change with the pulling speed at

which OmpG was unfolded. Thus, we could conclude that

the main unfolding pathways of OmpG did not depend on

the loading rate.

In agreement with theoretical considerations [30, 36] and

previous force spectroscopy studies on soluble [43, 44] and

membrane proteins [45–48], the unfolding forces increased

with increased pulling velocities. This dependency of the

unfolding force became clear on plotting the most probable

unfolding force of every main unfolding peak (Supporting

Information Fig. S2), F�, versus the logarithm of the loading

rate, r�f (Fig. 3). The dynamic force spectra of every structural

segment of OmpG showed one linear regime (Fig. 3).

According to the Bell–Evans model [30, 35, 49], the existence

of one linear regime indicates that a single energy barrier

separates the folded from the unfolded state. Using

Eqs. (1)–(3) (see Section 2) we extrapolated the parameters

characterizing the energy barriers stabilizing the unfolding

intermediates (Table 1).

3.3 Transition state distances of unfolding energy

barriers

The distance, xu, separating the folded structural segments

of OmpG from the transition state (Fig. 2) ranged from

0.23 to 0.85 nm (Table 1). These values agree well

with those measured for a-helical membrane proteins

such as bacteriorhodopsin (0.32–0.77 nm) [45], bovine

rhodopsin (0.21–0.47 nm) [48], the sodium-proton antiporter

NhaA (0.28–0.7 nm) [46], and the aa transporter SteT

(0.38–1.34 nm) [50] by DFS.

At pH 7.0 and 5.0, xu increased with an increasing

number of b-strands unfolded. An exception was the

unfolding intermediate of b-hairpin V, which at pH 7.0

Figure 2. Conceptual unfolding energy barrier tilted by an

externally applied force according to the Bell–Evans theory

[30, 35, 49]. (A) A simple two-state model that describes the

mechanical unfolding experiments. The energy potential exhi-

bits one energy barrier separating the folded low-energy state,

N, from the unfolded state, U. The activation energy for

unfolding is given by DGu, xu describes the distance from the

folded to the transition state, z, and ku gives the transition rate

for crossing the energy barrier. In absence of an externally

applied force ku equals the thermal transition rate k0. (B) Appli-

cation of an external force, F, changes the thermal likelihood of

reaching the top of the energy barrier. It is assumed that

although for a sharp barrier the distance, xu, of the folded state

relative to the energy barrier is not changed, the thermally

averaged projection of the energy profile along the pulling

direction is tilted by the mechanical energy (-Fcosy)x (short-

dashed line). y gives the angle of the externally applied force

relative to the molecular coordinate x. This tilt decreases the

energy barrier (black energy potential). At low loading rates, the

thermal contribution to overcome the energy barrier is higher,

and therefore, the mechanical energy required to overcome the

barrier is smaller. With increasing loading rates, the mechanical

force increases owing to a reduced lifetime of the folded state.

Proteomics 2010, 10, 4151–4162 4155


showed the largest xu of 0.85 nm. Recently, we compared

the unfolding energy landscapes of bacteriorhodopsin and

bovine rhodopsin [48]. The comparison showed that the xu

values increased as the unfolding of bacteriorhodopsin

progressed whereas no specific trend was observed during

the unfolding of rhodopsin. Also, rhodopsin showed the

presence of rigid structural cores suggesting the presence of

long-range interactions, which make the folding and

unfolding of multiple structural segments a cooperative

process [51, 52]. Such rigid structural cores were absent in

bacteriorhodopsin suggesting that the a-helices of bacter-

iorhodopsin, which unfold independent of each other, are

stabilized mainly by short-range interactions. Similar to

bacteriorhodopsin, OmpG shows a clear trend of increasing

xu values as the unfolding progresses and an absence of

rigid structural cores.

An increase in xu with unfolding may indicate that

OmpG is destabilized at an advanced unfolding stage.

The physical origin in the destabilization of the protein

structure with an increase in xu may lie in the breaking

Figure 3. DFS plots of OmpG

unfolded in the open (pH 7.0)

and closed (pH 5.0) state.

Loading rate dependence of

interactions that stabilize the

individual unfolding inter-

mediates of OmpG at pH 7.0

(red) and pH 5.0 (black). Each

data point represents the most

probable unfolding force of a

structural segment at the

given loading rate. Structural

segments unfolded are shown

in Fig. 1B (pH 7.0) and C

(pH 5.0). Fitting the loading

rate-dependent force (lines)

using Eq. (1) provides the

parameters of the energy

barriers that stabilize the

unfolding intermediates of

OmpG (Table 1). Fits were

weighted using the standard

error (S.E.) of the most prob-

able force. Error bars represent

the S.E. of force and loading

rate. The data was extracted

from the F-D spectra shown in

Supporting Information Fig. S1.

Experiments were performed at

pH 7.0 (25 mM Tris-HCl, 25 mM

MgCl2, 300 mM NaCl) and pH

5.0 (25 mM Na-acetate, 25 mM

MgCl2, 300 mM NaCl).



of the hydrogen bonds as the unfolding proceeds. This

interpretation may be supported by a decreasing trend

in the most probable (Supporting Information Fig. S2)

and average (Supporting Information Fig. S3) unfolding

force of the intermediates along the unfolding pathway. A

larger xu value implies an increase in the conformational

entropy of the unfolding intermediates and consequently

that they are located in broad energy wells (Fig. 2). Being

situated in a broad energy well, the structural segment has

to be stretched further to reach the transition state

than if it is sitting in a narrow energy well. The accom-

panying decrease in the spring constant, k, (Table 1)

signifies the highly frustrated nature of the energy land-

scape. A greater flexibility of the structure denotes more

conformations being trapped by many local minima

of the rough energy landscape. In contrast, a more rigid

structure is indicative of a minimally frustrated, smooth

energy landscape. The rigidity of the structural segments

decreases as observed in the case of bacteriorhodopsin.

Thus, it can be suggested that the folding of OmpG is

guided by short-range interactions and not by long-range

interactions.

3.4 Activation free energy of b-strands and

b-hairpins

At pH 7.0 and 5.0, the height of the free energy barriers,

DGz, separating the folded from the unfolded state of the

b-strands and -hairpins ranged from 20 to 23 kBT. DGz of

transmembrane a-helices of bacteriorhodopsin range

from 20 to 26 kBT [47], of bovine rhodopsin from 20

to 26 kBT [48], and of SteT from 20 to 36 kBT [50]. Thus, the

height of the free energy barriers stabilizing transmembrane

b-strands and b-hairpins is similar to the activation free

energies of transmembrane a-helices. Although empirical

evidence is needed, the similarity in the stabilizing

energies of b-hairpins and a-helices could be explained

based on the fact that most of the energy input during the

unfolding process is expended in reaching the transition

state of the intermediate. In other words, irrespective of

whether it is a b-strand or an a-helix, stretching a structural

segment by, in most cases, 1–3 aa brings the intermediate to

its transition state. Since the nature of physical interactions

(hydrogen bonds and hydrophobic interactions) broken to

reach the transition state and finally the intermediate

state is the same for an a-helix as well a b-strand or a

b-sheet, the energy required to reach the transition

state falls within a defined range. Thus, we assume that the

height of the free energy barrier to be overcome to initiate

the unfolding of a b-hairpin and a b-strand is independent

of the number of aa in the structural segment. Assuming

that a transmembrane a-helix has in average E20 aa

and a b-strand E10 aa, the average energy per residue

stabilizing a b-strand is approximately twofold higher than

that stabilizing an a-helix.Tab

le1.

Para

mete

rsch

ara

cteri

zin

gth

een

erg

yb

arr

iers

an

dsp

rin

gco

nst

an

tso

fth

eu

nfo

ldin

gin

term

ed

iate

so

fO

mp

Gin

the

op

en

(pH

7.0

)an

dcl

ose

d(p

H5.0

)st

ate

s

xu

(nm

)k u

(s�

1)

DGz

(kBT

)k(

N/m

)

Str

uct

ura

lse

gm

en

t(f

orc

ep

eak

po

siti

on

aa)

pH

7.0

pH

5.0

pH

7.0

pH

5.0

pH

7.0

pH

5.0

pH

7.0

pH

5.0

b-H

air

pin

I(8

aa)

0.2

47

0.0

20.2

57

0.0

23.37

4.0

3.47

1.0

19.57

0.9

19.57

0.9

2.67

1.0

2.67

0.7

b-H

air

pin

II(4

2aa)

0.2

87

0.0

40.2

67

0.0

20.97

1.0

1.27

0.7

20.77

1.8

20.57

0.6

2.17

1.7

2.47

0.6

b-H

air

pin

III

(83aa)

0.3

17

0.0

50.3

07

0.0

30.57

0.6

0.77

0.4

21.57

1.0

21.07

0.5

1.87

0.7

1.87

0.4

b-H

air

pin

IV(1

24aa)

0.3

57

0.0

90.3

77

0.0

30.47

0.6

0.27

0.8

21.77

1.0

22.27

1.6

1.47

0.2

1.37

0.6

b-H

air

pin

V(1

65aa)

0.8

57

0.1

90.0

017

0.7

–27.27

2.7

–0.37

0.2

–b-

Str

an

ds

S9–S

11

an

dp

art

of

loo

pL6

(165aa)

–0.4

67

0.0

4–

0.27

0.2

–22.47

1.0

–0.97

0.2

b-H

air

pin

VI

(204aa)

0.5

27

0.0

6–

0.17

0.1

–23.07

0.7

–0.77

0.2

–b-

Str

an

dS

12

an

dp

art

of

loo

pL6

(213aa)

–0.6

57

0.0

4–

0.17

0.1

–23.07

0.8

–0.47

0.1

b-H

air

pin

VII

(246aa)

0.6

17

0.1

10.6

27

0.1

00.17

0.2

0.37

0.2

22.87

0.5

22.97

0.8

0.57

0.2

0.57

0.1

Para

mete

rsw

ere

deri

ved

fro

mfi

ttin

gth

eD

FS

data

sho

wn

inFig

.3.

Err

ors

rep

rese

nt

stan

dard

devia

tio

ns,

xu

measu

res

the

dis

tan

cefr

om

the

en

erg

yw

ell

of

the

nati

ve

state

toth

etr

an

siti

on

state

,an

dk 0

desc

rib

es

the

kin

eti

ctr

an

siti

on

rate

at

wh

ich

the

stru

ctu

ral

seg

men

tu

nfo

lds

at

zero

forc

e.

Barr

ier

heig

hts

,D

Gz,

an

dsp

rin

gco

nst

an

ts,k,

were

calc

ula

ted

as

desc

rib

ed

un

der

Sect

ion

2.

Proteomics 2010, 10, 4151–4162 4157


3.5 Mechanical properties of OmpG

Structural rigidity defines the resistance of a material to

structurally deform in response to a mechanical force. The

structural rigidity of a protein depends on the curvature of

the potential well of the energy profile, the height of the

energy barrier, DGz, and the distance xu separating its

folded state from the transition state (Fig. 2). An energy

landscape describes the energy of a protein structure as a

function of its conformational entropy [53–55]. Accordingly,

the width of an energy valley defines the conformational

entropy of a protein structure. With increasing (decreasing)

width of an energy valley a protein structure can adopt more

(less) conformational substates owing to an increase

(decrease) in its flexibility [39, 48, 50, 56]. Because different

functional states have different conformational states,

structural rigidity is intrinsically related to the function of a

protein.

We have assumed a parabolic energy potential and a

sharp, static transition barrier at all loading rates for all

unfolding intermediates of OmpG [42, 48]. To approximate

the rigidity of the individual structural segments (b-hairpins

and b-strands) of the unfolding intermediates we calculated

their spring constants using Eq. (3). The spring constants of

the b-hairpins and b-strands of OmpG ranged from 0.3 to

2.6 N/m (Table 1). The spring constants of transmembrane

a-helices of bacteriorhodopsin ranged from 0.5 to 7.3 N/m

[48], of bovine rhodopsin from 0.9 to 3.8 N/m [48], and of

SteT from 0.2 to 2.8 N/m [50]. Thus, we conclude that

b-hairpins and b-strands of OmpG show a similar structural

flexibility as transmembrane a-helices. Changing OmpG

from the open to the closed state did not significantly affect

the mechanical properties of most of the b-hairpins and

b-strands.

As discussed in Section 3.3, the average rigidity of the

unfolding intermediates decreased with the increasing

number of secondary structures unfolded. A possible reason

for this decrease in the rigidity could be the breaking of the

interstrand hydrogen bond network as unfolding proceeds.

The first b-hairpin is the most rigid, and as hydrogen bonds

are broken the remaining b-strands get flexible. The

contribution of the inherent rigidity of each structural

segment, i.e. the rigidity of structural segments

in the absence of hydrogen bonds, is a daunting experi-

mental task. However, it may be assumed that the rigidity of

the last b-hairpin VII is the approximate inherent rigidity of

a b-hairpin in OmpG, and that the rigidity increases with

increasing complexity of the hydrogen bond network. In line

with such an interpretation, it has been shown that the

individual interactions of a hydrogen bond network can

contribute to folding in a cooperative manner [57]. A greater

flexibility of the structure denotes more conformations

being trapped by many local minima in a rough energy

landscape. The accompanying decrease in the spring

constant, k, (Table 1) also signifies the highly frustrated

nature of the energy landscape. In contrast, a more rigid

structure is indicative of a minimally frustrated, smooth

energy landscape.

The rigidity or the spring constant of the structural

segment constituted of b-strands S9–S11 and part of

loop L6 at pH 5 was calculated to be three times more

than that of b-hairpin V (b-strands S9–S10) at pH 7. It may

be assumed that the extra mechanical rigidity of the

gating region in the closed state reduces its structural fluc-

tuations to ensure that the gated pore remains closed.

In the open conformation, however, this may not be a

necessity and a flexible loop L6 can be tolerated by the

organism. Thus, a change in conformation of OmpG

changes its mechanical properties to ensure an efficient

functional state.

3.6 Mapping the unfolding energy landscapes of

OmpG in the open and closed states

It is worth noting that a change in the mechanical properties

during the pH-dependent conformational change of OmpG

also led to a bifurcation in the unfolding pathways, i.e.specifically the structural region gating the pore unfolded

via different intermediates at pH 7.0 and 5.0 (Figs. 1a

and B). Based on the F-D spectra it was possible to construct

the sequence of unfolding events that occurred until single

OmpG molecules were entirely unfolded to a fully stretched

conformation (Fig. 4). The DFS spectra enabled to quantify

the parameters associated with the energy barriers of the

main unfolding intermediates of OmpG (Table 1). Based on

these data the unfolding energy landscape was mapped for

OmpG in the two functional conformations corresponding

to the open and the closed states (Fig. 4). In both the states

OmpG shows similar unfolding events for the first four

b-hairpins I, II, III, and IV (Fig. 1), i.e. the same inter-

mediates populated the unfolding pathway from b-hairpin

I–IV. All parameters characterizing the energy barriers of

these first four unfolding events did not show significant

differences (Table 1). Thus, the first four b-hairpins of

OmpG unfolded via the same pathway independent of the

channel’s functional state. Unfolding of b-hairpins V and VI

depended on whether OmpG resided in the open or closed

state. b-hairpins V and VI unfolded in individual events

when OmpG was in the open state (Fig. 1). When OmpG

was in the closed conformation, b-hairpin V unfolded

together with b-strand S11. In this case, b-strand S12

unfolded in an individual event. Thus, the unfolding path-

ways of OmpG bifurcated upon reaching the unfolding of b-

hairpins V and VI, and depended on whether OmpG was in

the closed or in the open conformation. Accordingly, the

parameters, xu and ku, characterizing the unfolding

energy barriers of both the pathways also differed.

The last unfolding event of OmpG in the open and in the

closed states was characterized by the unfolding

of the b-hairpin VII. Within experimental error the

parameters characterizing the energy barrier of this



unfolding event did not depend on the conformational state

of OmpG (Table 1). Because we did not observe any

significant difference in xu and ku at pH 5 and pH 7, we

assumed that the last unfolding event of OmpG is very

similar or even the same whether OmpG is in the open state

or the closed state.

3.7 How stable is a membrane protein in the cell?

Bulk unfolding experiments denaturing b-barrel membrane

proteins in the presence of urea or detergent suggest a

surprisingly low stability (o10 kcal/mol) [17], similar to that

of the water-soluble b-barrel protein GFP (410–25 kcal/

Figure 4. Unfolding energy landscape of OmpG being set in the open (pH 7.0, red) and closed (pH 5.0, black) conformation. Upon applying

a sufficiently high mechanical force to the N-terminal end, the structural segments of OmpG start unfolding sequentially. Individual

unfolding intermediates of OmpG are trapped in energy valleys (rainbow colored). Structures forming the unfolding intermediates along

the unfolding pathway are shown for each energy valley. For simplicity we have assumed the energy valleys stabilizing the folded native

state of OmpG in the open (pH 7.0) and closed (pH 5.0) state to be the same (valley indicated by black number 1). Overcoming each energy

barrier that separates two energy valleys from each other induces the unfolding of a b-hairpin. The first four unfolding steps (stepwise

unfolding of b-hairpins I, II, III, and IV) that guide one unfolding intermediate to the other are the same for OmpG in the open and closed

states. After this, the unfolding pathway differs for OmpG being set in the open (top, red pathway at pH 7.0) and the closed (bottom, black

pathway at pH 5.0) state. In the open conformation, unfolding of b-hairpin IV is followed by the unfolding of b-hairpins V (transition from

energy valley numbered with black 5 to energy valley numbered red 6), then VI (transition from energy valley numbered with red 6 to

energy valley numbered black 7), and finally VII (transition from energy valley numbered with black 7 towards completely unfolded

OmpG). In the closed conformation, unfolding of b-hairpin IV is followed by the unfolding of b-hairpin V together with b-strand S11

(transition from energy valley numbered with black 5 to energy valley numbered black 6), then of b-strand S12 (transition from energy

valley numbered with black 6 to energy valley numbered black 7), and finally of b-hairpin VII (transition from energy valley numbered with

black 7 towards completely unfolded OmpG). Because the parameters characterizing the unfolding of b-hairpin VII do not differ between

OmpG being set in the open or in the closed state we assume that the last unfolding step is the same for both unfolding pathways.

The schematic representation of the unfolding energy landscape was reconstructed from parameters revealed by SMFS and DFS (Fig. 1,

Table 1). Shown is only the predominant (main) unfolding pathway for each functional conformation of OmpG at pH 7.0 and pH 5.0.

Proteomics 2010, 10, 4151–4162 4159


mol) [58]. However, the experimental conditions used to

chemically or thermally denature solubilized OmpG (or

GFP) are very different from those used here to mechani-

cally unfold OmpG. Most importantly in our experiments

OmpG was embedded in the native lipid bilayer and exposed

to physiological relevant buffer solution. This lipid bilayer is

important to maintain the physiological structural and

functional integrity, and establishes naturally occurring

interactions of OmpG. In addition, a force externally

applied by SMFS guides a membrane protein along

different unfolding pathways [27, 28, 30, 49]. Our results

show that the mechanical unfolding of single b-hairpins

and b-strands requires overcoming energy differences

between 20 and 23 kBT, i.e., E50–57 kcal/mol at room

temperature. Assuming that we have to unfold seven

b-hairpins (at pH 7.0) the energy required to unfold OmpG

entirely sums up to E140–160 kcal/mol. Thus, the high

unfolding free energy determined in our experiments

show that the b-barrel membrane protein OmpG in the

native lipid membrane may be much more stable than

thought by conventional unfolding experiments. However,

to be able to answer how stable a membrane protein is in the

living cell and via which pathways membrane proteins

indeed unfold, future experimental setups may be specifi-

cally designed. This not trivial task may be addressed in

the future.

4 Concluding remarks

In this work we have quantified the parameters that char-

acterize the unfolding energy barriers of OmpG in its open

and closed conformations. The unfolding intermediates of

OmpG depend on the functional state of the transmem-

brane pore. Deciphering the unfolding pathways showed

that the initial unfolding intermediates are the same for

both the open and closed conformational states of OmpG.

Till the unfolding of b-hairpin IV, OmpG takes the same

pathway in both the states. After the unfolding of b-hairpin

IV the unfolding pathways bifurcate on the energy land-

scape. Crucially, only the structural region involving extra-

cellular loop L6, which shows pH-dependent gating, unfolds

via different intermediates. The parameters characterizing

the last unfolding intermediate, i.e. the unfolding of

b-hairpin VII, do not differ in both the functional states of

OmpG. Besides the unfolding pathways, the spring

constants of most structural segments of OmpG remain the

same. However, a rigid extracellular loop L6 in the closed

state becomes more flexible in the open state suggesting a

crucial role in maintaining the functional integrity of the

protein. Finally, we propose that the two conformational

states of OmpG have different mechanical properties, which

are intricately related to the protein’s unfolding pattern and

function.

We have determined for the first time how the functional

state guides a native b-barrel forming membrane protein

along a different pathway of its unfolding energy landscape.

A localized interaction that changes the conformation

of an extracellular loop gates the transmembrane pore

of OmpG [13, 34] and at the same time changes the

(un-)folding pathway of the membrane protein. In the

crowdedness of a living cell [59], membrane proteins are

exposed to many specific and unspecific interactions. It

remains to be shown which of these interactions are

sufficient to influence the unfolding and the folding of

membrane proteins.

We thank C. Bippes for helpful discussions. This work wassupported by DFG (YI 96/3-1 and MU 1791) and EU.

The authors have declared no conflict of interest.

5 References

[1] Schulz, G. E., The structure of bacterial outer membrane

proteins. Biochim. Biophys. Acta 2002, 1565, 308–317.

[2] Xu, G. Z., Shi, B., McGroarty, E. J., Tien, H. T., Channel-

closing activity of porins from Escherichia coli in

bilayer lipid membranes. Biochim. Biophys. Acta 1986, 862,

57–64.

[3] Heyde, M., Portalier, R., Regulation of major outer

membrane porin proteins of Escherichia coli K 12 by pH.

Mol. Gen. Genet. 1987, 208, 511–517.

[4] Todt, J. C., Rocque, W. J., McGroarty, E. J., Effects of pH on

bacterial porin function. Biochemistry 1992, 31,

10471–10478.

[5] Todt, J. C., McGroarty, E. J., Acid pH decreases OmpF and

OmpC channel size in vivo. Biochem. Biophys. Res.

Commun. 1992, 189, 1498–1502.

[6] Conlan, S., Zhang, Y., Cheley, S., Bayley, H., Biochemical

and biophysical characterization of OmpG: A monomeric

porin. Biochemistry 2000, 39, 11845–11854.

[7] Schulz, G., Bacterial porins: structure and function. Curr.

Opin. Cell Biol. 1993, 5, 701–707.

[8] Muller, D. J., Engel, A., Voltage and pH-induced channel

closure of porin OmpF visualized by atomic force micro-

scopy. J. Mol. Biol. 1999, 285, 1347–1351.

[9] Andersen, C., Schiffler, B., Charbit, A., Benz, R., PH-induced

collapse of the extracellular loops closes Escherichia coli

maltoporin and allows the study of asymmetric sugar

binding. J. Biol. Chem. 2002, 277, 41318–41325.

[10] Yildiz, O., Vinothkumar, K. R., Goswami, P., Kuhlbrandt, W.,

Structure of the monomeric outer-membrane porin OmpG

in the open and closed conformation. EMBO J. 2006, 25,

3702–3713.

[11] Subbarao, G. V., van den Berg, B., Crystal structure of the

monomeric porin OmpG. J. Mol. Biol. 2006, 360, 750–759.

[12] Liang, B., Tamm, L. K., Structure of outer membrane protein

G by solution NMR spectroscopy. Proc. Natl. Acad. Sci. USA

2007, 104, 16140–16145.



[13] Mari, S. A., Koster, S., Bippes, C., Yildiz, O. et al., pH-

induced conformational change of the beta-barrel forming

protein OmpG reconstituted into native E. coli lipids. J. Mol.

Biol. 2010, 396, 610–616.

[14] Kleinschmidt, J. H., Tamm, L. K., Folding intermediates of a

b-barrel membrane protein. Kinetic evidence for a multi-

step membrane insertion mechanism. Biochemistry 1996,

35, 12993–13000.

[15] White, S. H., Wimley, W. C., Membrane protein folding and

stability: Physical principles. Annu. Rev. Biophys. Biomol.

Struct. 1999, 28, 319–365.

[16] Kleinschmidt, J. H., Tamm, L. K., Secondary and tertiary

structure formation of the beta-barrel membrane protein

OmpA is synchronized and depends on membrane thick-

ness. J. Mol. Biol. 2002, 324, 319–330.

[17] Tamm, L. K., Hong, H., Liang, B., Folding and assembly of

beta-barrel membrane proteins. Biochim. Biophys. Acta

2004, 1666, 250–263.

[18] Bond, P. J., Sansom, M. S., The simulation approach to

bacterial outer membrane proteins. Mol. Membr. Biol. 2004,

21, 151–161.

[19] Wimley, W. C., White, S. H., Reversible unfolding of

b-sheets in membranes: A calorimetric study. J. Mol. Biol.

2004, 342, 703–711.

[20] Booth, P. J., Unravelling the folding of bacteriorhodopsin.

Biochim. Biophys. Acta 2000, 1460, 4–14.

[21] Surrey, T., Schmid, A., Jahnig, F., Folding and membrane

insertion of the trimeric b-barrel protein OmpF. Biochem-

istry 1996, 35, 2283–2288.

[22] Conlan, S., Bayley, H., Folding of a monomeric porin,

OmpG, in detergent solution. Biochemistry 2003, 42,

9453–9465.

[23] Hong, H., Park, S., Jimenez, R. H. F., Rinehart, D., Tamm,

L. K., Role of aromatic side chains in the folding and ther-

modynamic stability of integral membrane proteins. J. Am.

Chem. Soc. 2007, 129, 8320–8327.

[24] Buchanan, S. K., Beta-barrel proteins from bacterial outer

membranes: structure, function and refolding. Curr. Opin.

Struct. Biol. 1999, 9, 455–461.

[25] Huysmans, G. H. M., Baldwin, S. A., Brockwell, D. J.,

Radford, S. E., The transition state for folding of an outer

membrane protein. Proc. Natl. Acad. Sci. USA 2010, 107,

4099–4104.

[26] Sapra, K. T., Damaghi, M., Koester, S., Yildiz, O. et al., One b

hairpin after the other: Exploring mechanical unfolding

pathways of the transmembrane b-barrel protein OmpG.

Angew. Chem. Int. Ed. 2009, 48, 8306–8308.

[27] Janovjak, H., Kessler, M., Gaub, H. E., Oesterhelt, D., Muller,

D. J., Unfolding pathways of native bacteriorhodopsin

depend on temperature. EMBO J. 2003, 22, 5220–5229.

[28] Kedrov, A., Janovjak, H., Sapra, K. T., Muller, D. J., Deci-

phering molecular interactions of native membrane

proteins by single-molecule force spectroscopy. Annu. Rev.

Biophys. Biomol. Struct. 2007, 36, 233–260.

[29] Engel, A., Gaub, H. E., Structure and mechanics of membrane

proteins. Annu. Rev. Biochem. 2008, 77, 127–148.

[30] Evans, E., Ritchie, K., Dynamic strength of molecular

adhesion bonds. Biophys. J. 1997, 72, 1541–1555.

[31] Evans, E., Kinoshita, K., Using force to probe single-

molecule receptor-cytoskeletal anchoring beneath the

surface of a living cell. Methods Cell Biol. 2007, 83C,

373–396.

[32] Muller, D. J., Engel, A., Atomic force microscopy and

spectroscopy of native membrane proteins. Nat. Protoc.

2007, 2, 2191–2197.

[33] Butt, H.-J., Jaschke,M., Calculation of thermal noise in

atomic force microscopy. Nanotechnology 1995, 6, 1–7.

[34] Damaghi, M., Bippes, C., Koester, S., Mari, S. A. et al., pH-

dependent interactions guide the folding and gate the

transmembrane pore of the beta-barrel membrane protein

OmpG. J. Mol. Biol. 2010, 397, 878–882.

[35] Bell, G. I., Models for the specific adhesion of cells to cells.

Science 1978, 200, 618–627.

[36] Evans, E., Energy landscapes of biomolecular adhesion and

receptor anchoring at interfaces explored with dynamic

force spectroscopy. Faraday Discuss. 1998, 111, 1–16.

[37] Bieri, O., Wirz, J., Hellrung, B., Schutkowski, M. et al., The

speed limit for protein folding measured by triplet-triplet

energy transfer. Proc. Natl. Acad. Sci. USA 1999, 96, 9597–9601.

[38] Krieger, F., Fierz, B., Bieri, O., Drewello, M., Kiefhaber, T.,

Dynamics of unfolded polypeptide chains as model for the

earliest steps in protein folding. J. Mol. Biol. 2003, 332,

265–274.

[39] Gr .ater, F., Grubm .uller, H., Fluctuations of primary ubiquitin

folding intermediates in a force clamp. J. Struct. Biol. 2007,

157, 557–569.

[40] Nunes, J., Hensen, U., Ge, L., Lipinsiky, M. et al., A ‘‘force

buffer’’ protecting immunoglobulin titin. Angew. Chem. Int.

Ed. 2010, 49, 3528–3531.

[41] Dietz, H., Berkemeier, F., Bertz, M., Rief, M., Anisotropic

deformation response of single protein molecules. Proc.

Natl. Acad. Sci. USA 2006, 103, 12724–12728.

[42] Howard, J., Mechanics of Motor Proteins and the Cytoske-

leton, Sinauer Associates Inc., Sunderland, Massachusetts

2001.

[43] Carrion-Vazquez, M., Marszalek, P. E., Oberhauser, A. F.,

Fernandez, J. M., Atomic force microscopy captures length

phenotypes in single proteins. Proc. Natl. Acad. Sci. USA

1999, 96, 11288–11292.

[44] Williams, P. M., Fowler, S. B., Best, R. B., Toca-Herrera, J. L.

et al., Hidden complexity in the mechanical properties of

titin. Nature 2003, 422, 446–449.

[45] Janovjak, H., Struckmeier, J., Hubain, M., Kedrov, A. et al.,

Probing the energy landscape of the membrane protein

bacteriorhodopsin. Structure 2004, 12, 871–879.

[46] Kedrov, A., Appel, M., Baumann, H., Ziegler, C., Muller,

D. J., Examining the dynamic energy landscape of an anti-

porter upon inhibitor binding. J. Mol. Biol. 2008, 375,

1258–1266.

[47] Sapra, K. T., Balasubramanian, G. P., Labudde, D., Bowie,

J. U., Muller, D. J., Point mutations in membrane proteins

Proteomics 2010, 10, 4151–4162 4161


reshape energy landscape and populate different unfolding

pathways. J. Mol. Biol. 2008, 376, 1076–1090.

[48] Sapra, K. T., Park, P. S., Palczewski, K., Muller, D. J.,

Mechanical properties of bovine rhodopsin and bacter-

iorhodopsin: possible roles in folding and function. Lang-

muir 2008, 24, 1330–1337.

[49] Evans, E., Probing the relation between force–lifetime–and

chemistry in single molecular bonds. Annu. Rev. Biophys.

Biomol. Struct. 2001, 30, 105–128.

[50] Bippes, C., Zeltina, A., Casagrande, F., Ratera, M. et al.,

Substrate binding tunes conformational flexibility and

kinetic stability of an amino acid antiporter. J. Biol. Chem.

2009, 28, 18651–18663.

[51] Sapra, K. T., Park, P. S. H., Filipek, S., Engel, A. et al.,

Detecting molecular interactions that stabilize bovine

rhodopsin. J. Mol. Biol. 2006, 358, 255–269.

[52] Park, P. S., Sapra, K. T., Jastrzebska, B., Maeda, T. et al.,

Modulation of molecular interactions and function

by rhodopsin palmitylation. Biochemistry 2009, 48,

4294–4304.

[53] Frauenfelder, H., Sligar, S. G., Wolynes, P. G., The energy

landscapes and motions of proteins. Science 1991, 254,

1598–1603.

[54] Wolynes, P. G., Onuchic, J. N., Thirumalai, D., Navigating

the folding routes. Science 1995, 267, 1619–1620.

[55] Dill, K. A., Chan, H. S., From Levinthal to pathways to

funnels. Nat. Struct. Biol. 1997, 4, 10–19.

[56] Sanders, C. R., Nagy, J. K., Misfolding of membrane

proteins in health and disease: the lady or the tiger? Curr.

Opin. Struct. Biol. 2000, 10, 438–442.

[57] Horovitz, A., Serrano, L., Avron, B., Bycroft, M., Fersht, A. R.,

Strength and co-operativity of contributions of surface salt

bridges to protein stability. J. Mol. Biol. 1990, 216,

1031–1044.

[58] Ishii, M., Kunimura, J. S., Jeng, H. T., Penna, T. C., Cholewa,

O., Evaluation of the pH- and thermal stability of the

recombinant green fluorescent protein (GFP) in the

presence of sodium chloride. Appl. Biochem. Biotechnol.

2007, 137– 140, 555–571.

[59] Ellis, R. J., Macromolecular crowding: an important but

neglected aspect of the intracellular environment. Curr.

Opin. Struct. Biol. 2001, 11, 114–119.

[60] Muller, D. J., Kessler, M., Oesterhelt, F., Moeller, C. et al.,

Stability of bacteriorhodopsin alpha-helices and loops

analyzed by single-molecule force spectroscopy. Biophys.

J. 2002, 83, 3578–3588.



Documents

Dual energy landscape: The functional state of the β-barrel outer membrane protein G molds its unfolding energy landscape