Photochemical regulation of an artificial hydrolase by a backbone incorporated tertiary structure switch

Article abstract of DOI:10.1002/chem.200801808

A stilbene chromophore has been incorporated into the turn region of a 42 amino acid peptide, linking two helical peptide sections. Spatial proximity between these sections, as well as aggregation into dimers, is required to facilitate the catalytic function of this artificial hydrolase. Photomodulation of the hydrolase activity results in an increase of the activity of 42% upon switching from the trans to the cis isomer of the chromophore. This is rationalized by a change in the aggregation state of the peptidomimetic, which is supported by diffusion coefficients obtained from PFG-NMR experiments. The results show that incorporation of a small, relatively flexible chromophore into a large peptide is capable of inducing a considerable change in tertiary structure and thus, functionality.

Full text of DOI:10.1002/chem.200801808

FULL PAPER
DOI: 10.1002/chem.200801808
Photochemical Regulation of an Artificial Hydrolase by a Backbone
Incorporated Tertiary Structure Switch
N. Johan V. Lindgren, Miranda Varedian, and Adolf Gogoll*[a]
Abstract: A stilbene chromophore has
been incorporated into the turn region
of a 42 amino acid peptide, linking two
helical peptide sections. Spatial prox-
imity between these sections, as well as
aggregation into dimers, is required to
facilitate the catalytic function of this
artificial hydrolase. Photomodulation
of the hydrolase activity results in an
increase of the activity of 42% upon
switching from the trans to the cis
isomer of the chromophore. This is ra-
tionalized by a change in the aggrega-
tion state of the peptidomimetic, which
is supported by diffusion coefficients
obtained from PFG-NMR experiments.
The results show that incorporation of
a
small, relatively flexible chromo-
phore into a large peptide is capable of
inducing a considerable change in terti-
ary structure and thus, functionality.
Keywords: aggregation
phores enzyme catalysis
peptidomimetics
· chromo-
·
·
Introduction
modulated properties of the resulting peptidomimetics in-
clude change of secondary structures such as b-hairpins and
a-helical motifs,[6,7] and modulation of gramicidin channel
permeability for ionic currents.[8] A few examples address
the control of catalytic activity. Thus, a photoinduced
change in the rate of hydrolysis of para-nitrophenol acetate
by a cyclodextrin-peptide hybrid tagged with azobenzene
has been reported.[9] Woolley et al. showed that juxtaposing
a phenylazophenylalanine adjacent to the catalytic residue
in RNase S led to small changes of enzyme activity between
the isomers of the azobenzene derivative for the hydrolysis
of RNA.[10] Both of these examples use the strategy of at-
taching azobenzene derivatives to amino acid side chains, a
concept that previously has been shown to affect the confor-
mation of the peptide to a much smaller extent than the
backbone approach.[11]
Nature relies on photoswitchable molecules to trigger bio-
logical functions essential for life. One of the most familiar
examples is the cis–trans isomerization of retinal, a small
molecule covalently attached to opsin. The protein changes
its conformation when retinal is isomerized, and this confor-
mational change activates an associated G-protein and trig-
gers a second messenger cascade, ultimately resulting in the
perception of vision.[1] Another example is provided by the
kindling fluorescent proteins (GFP, green fluorescent pro-
teins) of certain marine invertebrates, in which chromo-
phore isomerization triggers bioluminescence.[2] Recently,
the enzyme-catalyzed cis–trans isomerization of backbone
proline residues has attracted attention as a natural mecha-
nism to modulate protein function.[3]
Incorporation of photoswitchable chromophores into pep-
tides and proteins opens new possibilities to modulate the
structure and hence the function of these compounds.[4]
Azobenzenes constitute the dominating chromophores in
these studies,[4,5] probably due to their well understood pho-
tochemistry and ease of synthetic accessibility. The photo-
In this work we report a profiling on the incorporation of
the stilbene photoswitch into the backbone of a peptide,
thus triggering a change of tertiary rather than secondary
peptide structure. The target peptide of choice is a previous-
ly reported artificial hydrolase (HN1), comprising two paral-
lel helical segments linked by a Gly-Pro-Val-Asp turn
(Figure 1). Spatial proximity of the two helical segments has
been established to be crucial to the enzymatic function of
this artificial hydrolase.[12,13] The prime objective was wheth-
er it would be possible to modulate this enzymatic activity
with a rather flexible photoswitch[14] that was to be incorpo-
rated into the peptide backbone. Examples of this so called
optical nanotweezer approach are very rare,[4a] as opposed
to side chain incorporation. Since photomodulation of large
[a] N. J. V. Lindgren, M. Varedian, Prof. A. Gogoll
Department of Biochemistry and Organic Chemistry
Uppsala University, Box 576, 751 23 Uppsala (Sweden)
Fax : (+46)18-471-3818
E-mail: adolf.gogoll@biorg.uu.se
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801808.
Chem. Eur. J. 2009, 15, 501 – 505
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
501

hypothesized that the exchange
of the entire GPVD sequence
with
a meta-substituted stil-
bene derivative, might mimic
the loop upon isomerization to
its cis isomer. Conversely, the
trans isomer might generate a
turn sequence less favorable
for the hydrophobic interaction
between the helical segments
(Figure 2).
Thioaurones are another
type of photoswitchable com-
pounds and have recently been
used to regulate the helical
conformation of a peptide by
linkage of two amino acid side
chains
in
the
peptide
Figure 1. Sequences of the HN1 and JM1 peptides (Nle=norleucin). The C-terminal was amidated and the N-
strand.[8,21]
Conformational
terminal was acetylated.
studies have indicated that thi-
oaurone derivatives are more
rigid than stilbenes, and thus
molecular entities, for example, dendrimers (Mꢀ6500), by
incorporation of stilbene chromophores has been report-
ed,[15] this task appeared to be viable.
potentially impose larger conformational restraint differen-
ces between their E- and Z-isomers onto peptides upon iso-
merization.[15] However, the use of thioaurones in photo-
The hydrolytic activity of the target peptide depends on
the proximity of histidines, which then promote a general
acid–general base catalysis.[16,17] The sequence of the HN1
peptide (Figure 1) is based on the template SA42 and has
earlier been reported to catalyze the hydrolysis of an RNA
mimic.[18] The role of each residue has not been unequivo-
cally determined but the proximity of several histidines to
each other is of utmost importance. The two helical seg-
ments are amphiphilic and designed to interact through the
hydrophobic effect. They are connected head to tail by the
tetrapeptide sequence Gly-Pro-Val-Asp, which acts as helix
breaker and forms a loop region. The design can be de-
scribed in terms of the heptad repeat pattern[19] (a-b-c-d-e-f-
g)n, in which the residues in the a and d position form the
hydrophobic core, the residues in the c and g position form
the exposed surface of the dimer, and the residues in the b
and e position are at the dimer interface and control the di-
merization. The design of the polypeptide in this manner
drives it to fold into antiparallel dimers, constituting four-
helix-bundles. Here, it has to be noted that the accurate sec-
ondary and tertiary structures of these peptides have not
been established yet.
The stilbene switch substituted for the Gly-Pro-Val-Asp
sequence (Figure 2), has previously been used to photo-
chemically regulate the secondary structure of a b-hairpin.[14]
It provides access to two photoisomers of different geometry
(i.e., cis and trans isomers) but has at the same time an in-
herent conformational flexibility. Hence, it is likely to have
the potential to adopt to the secondary structure of a pep-
tide, if other interactions, for example, such as those be-
tween parallel peptide strands, overpower the conformation-
al preferences of the chromophore itself.[14,20] Therefore, we
Figure 2. Photoisomerization of the stilbene chromophore and of the
photoswitchable peptidomimetic JM1.
502
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 501 – 505

Photochemical Regulation of an Artificial Hydrolase
FULL PAPER
switchable peptidomimetics appears to be limited, due to
their chemical instability during solid-phase peptide synthe-
sis conditions.[22,23]
To test the performance of this chromophore, we synthe-
sized another peptide sequence, JM2, in which the GPVD
sequence was replaced by the thioaurone switch[24] in the
loop region of the polypeptide (Scheme 1).
likely, it indicates a change of peptide aggregation state, that
is, from monomers (trans-JM1) into dimers (cis-JM1). Inter-
estingly, although the above-mentioned relationship be-
tween peptide molecular weight (Mr) and diffusion coeffi-
cient Dt was devised for spherical peptides, calculated diffu-
sion coefficients based on the molecular mass of JM1 are
surprisingly similar to the observed values for trans-JM1 and
cis-JM1. This suggests that trans-JM1 exists predominantly
as monomers, whereas cis-JM1 forms dimers. The validity of
this interpretation receives further support from investiga-
tions on the HN1 peptide. Analogues of HN1 have previ-
ously been shown, by ultracentrifugation, to exist in dimeric
form in aqueous solution, and HN1 has been reported to
behave similarly.[18] Again, the diffusion coefficient, as deter-
mined by PFG-NMR experiments, fits well with this as-
sumption (Table 1). Furthermore, when HN1 is dissolved in
D2O/CD3CN (65:35 vol%), which is known to suppress pep-
tide aggregation,[27] the diffusion coefficient increases to the
value that would be expected for the monomeric peptide.
Here, the different viscosity of the solvent mixture has been
accounted for.
Scheme 1. The thioaurone chromophore incorporated into the peptide se-
quence JM2.
Results and Discussion
Previous studies of the stilbene photoswitch (Figure 2a)
have shown that both cis and trans isomers, due to their in-
trinsic flexibility, promote a loop motif in peptidomimetics,
albeit of different geometry.[14,25] Photoisomerization of the
peptide JM1 was followed by analysis with analytical HPLC
(see Figure S1 in the Supporting Information). This indicat-
ed a 1:9 cis:trans ratio of the stilbene moiety after synthesis,
which is supported by the 1H NMR data. Photoisomeriza-
tion for 90 min at 300 nm changed this to a 4:1 cis:trans
ratio, comparable to other peptidomimetics with the stilbene
chromophore.[14,20,25]
CD spectroscopy of the peptidomimetics indicates a high
helical content for a broad range of peptide concentrations
(see Supporting Information for data). The JM1 peptide has
a higher helical content compared to HN1. No significant
differences were detected between the cis-JM1 and the
trans-JM1 structures. Estimates of helical contents have pre-
viously been used to obtain indirect information about the
aggregation state of this type of peptides. However, in the
absence of detailed structural information, it is difficult to
draw clear conclusions from the current CD data.
Measurements of translational diffusion coefficients by
PFG-NMR experiments provide comparably straightforward
information on the molecular weight of peptides and their
aggregates in solution, as has been elaborated by Grꢁslund
et al.[26] These parameters are presented in Table 1. The con-
siderable divergence between the diffusion coefficients for
trans-JM1 and cis-JM1, is unlikely to originate only in an al-
teration in peptide tertiary structure such as a different ori-
entation of the two helical segments (Figure 2b). More
The hydrolysis of para-nitrophenylacetate was used to
monitor the enzymatic performance of the peptidomimetics.
Second-order rate constants k2 for the peptides HN1, and
the photostationary states of cis-JM1 and trans-JM1 are pre-
sented in Table 2. A reference peptide HN1-ref in which the
histidines are replaced with alanines is also included. As ex-
Table 2. Second-order rate constants of the peptides HN1, JM1 and
HN1-ref. The third column shows the rate constants normalized to the
rate constant of the HN1 peptide.
Peptide
k2 [mꢁ1 sꢁ1
]
k2rel =k2 (pept)/k2 (HN1)
HN1
0.049
0.043
0.061
0.004
1.00
0.88
1.25
0.07
trans-JM1
cis-JM1
HN1-ref
pected, the HN1-ref peptide shows almost no effect of gen-
eral acid–general base catalysis (k2rel =0.07). The small resid-
rel
ual effect (i.e., k2 ¼0) can be attributed to secondary ef-
fects, most likely transition-state stabilization by the posi-
tively charged arginines.[17] The lower value of k2rel =0.88 for
trans-JM1 as compared to k2rel =1.25 for cis-JM1 shows that
the histidines are situated in a less favorable position in the
trans isomer. The higher k2rel-value for cis-JM1 compared to
that for the reference peptide HN1 (k2rel =1.00), is likely
due to different geometries of the loop regions.
Thus, it appears that the
loop provided by the cis-stil-
bene promotes arrangement of
the helical peptide segments
into a structure even more
Table 1. Observed (obs) and calculated (calcd) diffusion coefficients for HN1 and JM1 peptides.
Peptide
Solvent
Mr
[gmolꢁ1
Dobsd 1010
[m2 sꢁ1
Dcalcd 1010
[26]
A
]
A
]
G
]
suitable for catalytic activity
than what the Gly-Pro-Val-Asp
sequence does in the original
HN1 peptide.[28] In addition,
HN1 (dimer)
D2O
9018
4509
4416
8832
1.14
1.48
1.58
1.03
1.10
1.50
1.51
1.11
HN1 (monomer)
trans-JM1 (monomer)
cis-JM1 (dimer)
D2O:CD3CN (65:35)
D2O
D2O
Chem. Eur. J. 2009, 15, 501 – 505
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
503

A. Gogoll et al.
terminal was amidated upon cleavage from the resin. The Fmoc-protect-
ed stilbene derivative was, together with the two following amino acids,
introduced manually in the sequence (to minimize the excess) using (ben-
and most likely even more important, the dimerization of
the peptidomimetic, which promotes its catalytic activity, is
augmented for the cis isomer. It should also be noted that
trans-JM1 appears to be monomeric while retaining its high
helix content, in contrast to previously reported congeners
without the stilbene motif.[12]
To test the hypothesis that a more rigid switch might gen-
erate a larger structural and thus catalytic difference be-
tween the photoisomers, the thioaurone switch was incorpo-
rated in the novel sequence JM2. The synthesis of JM2 was
successful according to MALDI-TOF MS and HPLC data
(see Supporting Information for data). Following cleavage
from the resin, subsequent purification steps, which were
monitored by MALDI-TOF MS, indicated successive de-
composition, which also was detected by changes in reten-
tion times and the UV spectrum. A likely explanation is an
irreversible chemical conversion of the thioaurone moiety,
probably due to intramolecular attack by nucleophiles pres-
ent in the peptide sequence.
zotriazol-1-yloxy)tripyrrolidinophosphonium
hexafluorophosphate
(PyBop) as activating agent (0.6m), with a threefold amino acid excess al-
lowing a coupling time of 120 min. The global cleavage of the peptide
was done in a mixture containing trifluoroacetic acid (TFA)/triisopropyl-
silane/H2O (95:2.5:2.5 v/v) for four hours. The solution was concentrated
and the peptide was precipitated with cooled diethyl ether. The crude
material was lyophilized prior to purification by reversed-phase HPLC
(semi-preparative hypersil C-18 column 250ꢂ20 mm, 5 mm particle size)
using a gradient of 40–50% acetonitrile containing 0.1% TFA (v/v) for
60 min. The purified peptide was obtained upon lyophilization of the col-
lected HPLC-fractions. Identification of the peptide was done by
MALDI-TOF MS (Applied Biosystems, Voyager-DE PRO). The sample
was mixed with a saturated solution of a-cyano-4-hydroxy cinnamic acid
in acetonitrile/H2O (1:1) containing 0.1% TFA. Purity was confirmed by
amino acid analysis (Uppsala University Amino Acid Analysis Center)
and analytical HPLC (Hypersil C-18 250ꢂ4.6 mm, 5 mm particle size)
using a gradient of 40–60% acetonitrile containing 0.1% TFA for 40 mi-
nutes. No impurities were detected (diode-array detector 229–400 nm).
The peptide JM2 was synthesized in a similar way, but due to stability
problems of the thioaurone switch during solid-phase peptide synthesis,
cleavage conditions were altered by replacing triisopropylsilane by ani-
sole.
CD spectroscopy: CD spectroscopy (JASCO J-810) was done at 298 K
using peptide concentrations of 600, 60, and 6 mm in 10 mm phosphate
buffer. The pH was set to 7.0 prior to measuring.
Conclusion
Herein we have shown that when replacing the loop region
in a catalytically active helix-loop-helix peptide with a pho-
toswitchable stilbene derivative, the rate of hydrolysis of an
activated substrate was not only retained, but slightly im-
proved. Photoisomerization resulted in a modulation of cat-
alytic activity, with a 42% higher rate constant for the cis
isomer. Most likely, in the photoswitchable peptidomimetics
the inter-helix distances are more (cis isomer) or less (trans
isomer) optimal for the catalytic activity than in the refer-
ence peptide HN1. Thus, incorporation of a comparatively
small and even somewhat flexible photoswitchable unit into
a large peptidomimetic appears to be a viable concept for
photomodulation of functionality in a biocatalyst. Optimiza-
tion of the switch structure to render larger conformational
differences between the photoisomers, for example, by
choosing a more rigid switch, holds the promise to generate
more drastic effects. However, such a switch must also show
sufficient chemical stability.
Photoisomerization: The photochemical reactions were carried out for
acetonitrile/water under N2 gas flow using an Oriel 1000 W Xe ARC
light source and a 300 or 280 nm Oriel UV filter. The emitted light inten-
sity was determined by using a UV enhanced Si photodiode (5.8 mm2) at-
tached to a current meter.
Kinetics: The second-order rate constant k2 of JM1 for the hydrolysis of
para-nitrophenylacetate was determined by linear regression of the plot
of the observed velocity (vobs) at three different concentrations of the cat-
alyst against the catalyst concentration. The vobs was determined by dilut-
ing a stock solution of peptide of known concentration determined by
amino acid analysis to a new stock solution, making it 2 mm of peptide in
50 mm sodium acetate as buffer. The pH was fixed to 5.0. From the 2 mm
peptide solution, three concentrations of peptide (0.4, 0.6, and 0.8 mm)
were made by dilution with buffer in separate Eppendorf tubes. The sub-
strate was dissolved in acetonitrile to give a stock solution of 20 mm.
Measurement of the product formation started as soon as possible after
addition of 4 mL of substrate stock to 796 mL of peptide solution, thor-
ough mixing, and division of each peptide into two 0.1 cm cuvettes for
dual samples. The evolution of the hydrolysis was studied at 320 nm for
detection of the para-nitrophenol on a CARY 100 BIO equipped with a
temperature controller. The reaction proceeded at 298 K. The concentra-
tion of product formed was determined from the absorption of the para-
nitrophenol species using an absorption coefficient of 10000 cmꢁ1 mꢁ1
.
The reactions were monitored to complete hydrolysis of the substrate.
The vobs was determined from the initial rate for the first 60 min of the
measurement.
Experimental Section
Diffusion measurements: For the LED-PGSE diffusion experiments,[29]
z
Synthesis: The synthesis of the photoswitchable meta-substituted stilbene
derivative was conducted according to the literature.[14] Characterization
and detailed synthetic procedures can be found in the Supporting Infor-
mation. The peptide JM1 was synthesized on an automated peptide syn-
thesizer (Pioneer, Applied Biosystems) by standard Fmoc (9-fluorenyl-
methoxycarbonyl) chemistry on a 0.2 mmol scale. The solid phase used
gradients were employed and 64 scans were acquired. A relaxation delay
of 0 s, 9 ms gradient pulse duration, 40 ms diffusion delay, 5 ms storage
delay was used and the gradient pulse strength was arrayed between 0
and 20 Gausscmꢁ1 (20 steps). The diffusion coefficients were calculated
based on the diffusion coefficient of HDO (1.902ꢂ10ꢁ9 m2 sꢁ1),[30] with an
accuracy of approximately 3%.
was Fmoc-PAL-PEG-PS (Applied Biosystems) with
a loading of
0.20 mmolgꢁ1. The Fmoc-protected amino acids were coupled in a four-
fold excess by benzotriazol-1-yl-N-tetramethyluronium hexafluorophos-
phate (HBTU; 0.8m in DMF) and diisopropylethylamine (DIPEA; 1m in
DMF) as activators. The Fmoc protecting group was deprotected by
using 20% piperidine in DMF (v/v). Standard coupling times for the
amino acids were 60 min except for His and Gln (90 min), Arg and Asn
(120 min). The N-terminal was capped with acetic anhydride and the C-
Acknowledgement
The Swedish Research Council is gratefully acknowledged for financial
support. We also thank Prof. Lars Baltzer for fruitful discussions.
504
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 501 – 505

Photochemical Regulation of an Artificial Hydrolase
FULL PAPER
[16] K. S. Broo, L. Brive, P. Ahlberg, L. Baltzer, J. Am. Chem. Soc. 1997,
119, 11362–11372.
[17] K. S. Broo, H. Nilsson, J. Nilsson, A. Flodberg, L. Baltzer, J. Am.
Chem. Soc. 1998, 120, 4063–4068.
[18] J. Razkin, H. Nilsson, L. Baltzer, J. Am. Chem. Soc. 2007, 129,
14752–14758.
[19] J. W. Bryson, S. F. Betz, H. S. Lu, D. J. Suich, H. X. Zhou, K. T.
OꢅNeil, W. F. DeGrado, Science 1995, 270, 935–941.
[20] M. Varedian, M. Erdꢄlyi, ꢆ. Persson, A. Gogoll, J. Pept. Sci., in
press.
[21] M. Fꢃllbeck, E. Michalsky, I. S. Jaeger, P. Henklein, H. Kuhn, K.
Rꢃck-Braun, R. Preissner, Genome Informatics 2006, 17, 141–151.
[22] M. Varedian, V. Langer, J. Bergquist, A. Gogoll, Tetrahedron Lett.
2008, 49, 6033–6035.
[23] W. Steinle, PhD thesis, Technische Universitꢁt Berlin (Germany),
2004.
[24] W. Steinle, K. Rꢃck-Braun, Org. Lett. 2003, 5, 141–144.
[25] M. Erdꢄlyi, M. Varedian, C. Skçld, I. B. Niklasson, J. Nurbo, ꢆ.
Persson, A. Gogoll, Org. Biomol. Chem.2008, 6, 4356–4373.
[26] J. Danielsson, J. Jarvet, P. Damberg, A. Grꢁslund, Magn. Reson.
Chem. 2002, 40, S89–S97.
[1] C. K. Mathews, K. E. van Holde, K. G. Ahern in Biochemistry, Ad-
dison Wesley Longman, San Francisco, 1999, pp. 697–852.
[2] a) O. Shimomura, Biol. Bull. 1995, 189, 1–5; b) D. C. Prasher, V. K.
Eckenrode, W. W. Ward, F. G. Prendergast, M. J. Cormier, Gene
1992, 111, 229–233; c) J. N. Henderson, S. J. Remington, Physiology
2006, 21, 162–170.
[3] a) M. Schutkowski, A. Bernhardt, X. Z. Zhou, M. Shen, U. Reimer,
J.-U. Rahfeld, K. P. Lu, G. Fischer, Biochemistry 1998, 37, 5566–
5575; b) K. P. Lu, G. Finn, T. H. Lee, L. K. Nicholson, Nat. Chem.
Biol. 2007, 3, 619–629.
[4] a) P. Gorostiza, E. Isacoff, Mol. Biosyst. 2007, 3, 686–704; b) D. D.
Young, A. Deiters, Org. Biomol. Chem. 2007, 5, 999–1005.
[5] C. Renner, L. Moroder, ChemBioChem 2006, 7, 868–878.
[6] T. Schrader, W. J. Schreier, T. Cordes, F. O. Koller, G. Babitzki, R.
Denschlag, C. Renner, M. Lçweneck, S.-L. Dong, L. Moroder, P.
Tavan, W. Zinth, Proc. Natl. Acad. Sci. USA 2007, 104, 15729–
15734.
[7] G. A. Woolley, Acc. Chem. Res. 2005, 38, 486–493.
[8] T. Lougheed, V. Borisenko, T. Hennig, K. Rꢃck-Braun, G. A. Wool-
ley, Org. Biomol. Chem. 2004, 2, 2798–2801.
[9] A. Ueno, T. Shimizu, H. Mihara, K. Hamasaki, K. Pitchumani, J. In-
clusion Phenom. Macrocyclic Chem. 2003, 44, 49–52.
[10] D. Liu, J. Karanicolas, C. Yu, Z. Zhang, G. A. Woolley, Bioorg.
Med. Chem. Lett. 1997, 7, 2677–2680.
[27] W. Bocian, J. Sitkowski, E. Bednarek, A. Tarnowska, R. KawÞcki, L.
Kozerski, J. Biomol. NMR 2008, 40, 55–64.
[28] B.-C. Lee, T. K. Chu, K. A. Dill, R. N. Zuckermann, J. Am. Chem.
Soc. 2008, 130, 8847–8855.
[11] P. Hamm, J. Helbing, J. Bredenbeck, Ann. Rev. Phys. Chem. 2008,
59, 291–317.
[12] S. Olofsson, G. Johansson, L. Baltzer, J. Chem. Soc. Perkin Trans. 2
1995, 2047–2056.
[29] a) S. J. Gibbs, C. S. Johnson, Jr., J. Magn. Reson. 1991, 93, 395–402;
b) G. A. Morris, H. Barjat in Methods for Structure Elucidation by
High Resolution NMR (Eds.: G. Batta, K. E. Kçvꢄr, C. Szꢇntay),
Elsevier, 1997, pp. 209.
[13] S. Olofsson, L. Baltzer, Folding Des. 1996, 1, 347–356.
[14] M. Erdꢄlyi, A. Karlꢄn, A. Gogoll, Chem. Eur. J. 2006, 12, 403–412.
[15] a) A. Momotake, T. Arai, Polymer 2004, 45, 5369–5390; b) A. Mo-
motake, T. Arai, J. Photochem. Photobiol. C 2004, 5, 1–25.
[30] R. Mills, J. Phys. Chem. 1973, 77, 685–688.
Received: September 2, 2008
Published online: December 8, 2008
Chem. Eur. J. 2009, 15, 501 – 505
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
505