Structure of the tripeptide model Ac-Val-Tyr(Me)-NHMe and its cluster with water investigated by IR/UV double resonance spectroscopy

Article abstract of DOI:10.1039/b407174c

In this paper the structures of the isolated tripeptide model Ac-Val-Tyr(Me)-NHMe (Val = valine, Tyr = tyrosine) and its cluster with water are investigated by mass-, isomer- and state-selective IR/UV double resonance spectroscopy. From the IR spectra bot

Full text of DOI:10.1039/b407174c

View Article Online / Journal Homepage / Table of Contents for this issue
R E S E A R C H P A P E R
Structure of the tripeptide model Ac–Val–Tyr(Me)–NHMe and its
cluster with water investigated by IR/UV double resonance
spectroscopy
H. Fricke,a A. Gerlach,a C. Unterberg,a P. Rzepecki,b T. Schraderb and M. Gerhards*a
a Heinrich-Heine Universitat Dusseldorf, Institut fur Physikalische Chemie I, UniversitatsstraXe
¨
26.33.O2, 40225 Dusseldorf, Germany. E-mail: gerhards@uni-duesseldorf.de
¨
¨
¨
¨
b Fachbereich Chemie, Philipps-Universitat Marburg, Hans-Meerwein-Str., 35032 Marburg,
Germany
¨
Received 12th May 2004, Accepted 7th September 2004
First published as an Advance Article on the web 16th September 2004
In this paper the structures of the isolated tripeptide model Ac–Val–Tyr(Me)–NHMe (Val ¼ valine, Tyr ¼
tyrosine) and its cluster with water are investigated by mass-, isomer- and state-selective IR/UV double resonance
spectroscopy. From the IR spectra both in the region of the NH and C O stretching vibrations and in
Q
combination with force field and ab initio calculations it can unambiguously be derived that the peptide contains
a stretched, b-sheet related structure. Thus the peptide serves as an ideal candidate for b-sheet model systems.
By adding one water molecule to the peptide the b-sheet related structure seems to be preserved with a water
molecule being attached to the NHMe group.
but also the region of the C O stretching and NH bending
Q
I. Introduction
vibrations (about 1500–1800 cmꢀ1) has been investigated. In
The conformational preferences of isolated protein building
blocks, amino acids and peptides, have been subject of great
interest in recent years. Due to their flexibility peptides can
occur in many different conformers. In order to learn some-
thing about the driving forces to form a conformer, investiga-
tions can start by applying molecular beam experiments on
isolated amino acids and peptides. The experimental discrimi-
nation between different conformers requires the use of mass-,
isomer- and state-selective methods.
If the investigated species contains an aromatic chromo-
phore (like the common amino acids phenylalanine, tyrosine or
tryptophan) R2PI (resonant 2-photon ionization) spectroscopy
can be used. This was first done by Levy and coworkers1–3 who
recorded R2PI spectra of the free amino acids phenylalanine
(Phe), tyrosine (Tyr) and tryptophan (Trp) as well as some di-
and tripeptides. In order to get information on the structure of
the amino acids in the electronic grounds states (S0) vibrational
frequencies of the important functional groups have to be
recorded. IR/UV double resonance spectroscopy (IR/R2PI;
see e.g. ref. 4–13) turns out to be an excellent method to
investigate the IR spectra of different neutral species in their
S0 states. In combination with force field, ab initio, and density
functional theory (DFT) calculations structural assignments
can be derived.
most cases information on the C O stretching vibrations are
Q
necessary to yield an unambiguous assignment of a structural
arrangement. In order to open the new spectral window in the
region of 1700 cmꢀ1 for the IR/R2PI spectroscopy of C
O
Q
stretching vibrations a new laser system has been con-
structed.12,15
Zwier and coworkers investigated the protected amino acids
Ac–Trp–NH2 and Ac–Trp–NHMe in the region of the NH
stretching vibrations by applying IR/laser induced fluorescence
(LIF) spectroscopy.16 Three (two) isomers of Ac–Trp–NHMe
(Ac–Trp–NH2) have been observed. In order to discuss the
conformational dynamics of Ac–Trp–NHMe after vibrational
excitation the new hole-filling spectroscopy has been applied.17
The backbones of (protected) amino acids and peptides are
characterized by a set of torsional angles f and c for each
amino acid (Ramachandran plot).18,19 Furthermore, the side-
chain configurations are characterized by the dihedral angles w1
and w2, cf. Fig. 1. In the case of a b-sheet related structure (bL)
the angles f and c fall into the region of about ꢀ1801 o f o
ꢀ1201 or 1201 o f o 1801 and ꢀ1801 o c o ꢀ1201 or 1201 o
c o 1801, i.e. NH and C O groups of an amino acid are
Q
nearly parallel to each other. Structures forming an intramo-
lecular hydrogen bond between a NH and a C O group can
Q
form gL arrangements with angles f and c of the backbone
lying between: ꢀ1201 o f o 01; 01 o c o 1201. A detailed
description of the nomenclature with respect to the angles f, c,
w1 and w2 is given in refs. 11,19.
The first applications of the IR/R2PI technique to the amino
10
acids phenylalanine12 and tryptophan have been performed
by Simons and coworkers. The IR spectra have been recorded
in the region of the NH and OH stretching vibrations and
several isomers have been deduced from the IR spectra con-
taining intramolecular hydrogen bonds between the COOH
and NH2 group. In a recent publication Bakker et al.13
investigated the spectral fingerprint region of tryptophan from
330 to 1500 cmꢀ1 by using a free electron laser as IR light
source.
In contrast to the unprotected amino acids the protected
species (at N and C terminus) cannot form intramolecular
hydrogen bonds between the polar acid (COOH) and basic
(NH2) groups leading to a higher selectivity of the structures.
In the case of Ac–Phe–OMe11,12 only a b-sheet related (bL)
conformer has been observed as prominent isomer, whereas for
Ac–Phe–NHMe14 four isomers are obtained with a bL struc-
ture being the most stable one. These protected amino acids
serve both as most simple model systems to describe b-sheet
related structures as well as b-sheet models. b-sheet models
In our group the protected amino acids Ac–Phe–OMe (Ac ¼
acetyl, Me ¼ methyl)11,12 and Ac–Phe–NHMe14 have been
investigated. In these studies not only the NH stretch region
4636
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 4 6 3 6 – 4 6 4 1
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
structures similar to the one observed for isomers of Ac–Phe–
NHMe,14 Ac–Trp–NH2,16 and Ac–Trp–NHMe.16
In contrast to unprotected amino acids the polar NH and
CO groups of Ac–Val–Tyr(Me)–NHMe are located in the
backbone and not at the end of the molecule, i.e. water will
be attached to the backbone of the peptide. In this paper we
will report on the influence of the first water molecule attached
to the model tripeptide Ac–Val–Tyr(Me)–NHMe.
II. Experimental
(a) Synthesis of Ac–Val–Tyr(Me)–NHMe
295 mg (1.00 mmol, 1.00 eq) N-tert-butyloxycarbonyl-O-
methyl-(S)-tyrosine were dissolved in 5 ml of dry dichloro-
methane under argon and cooled to ꢀ10 1C. To this solution
162 mg (1.00 mmol, 1.00 eq) of N,N0-carbonyldiimidazole were
added and the reaction mixture was stirred for another 2 h at
ꢀ10 1C. Subsequently, 500 ml (1.10 mmol, 1.10 eq) of a 2 M
solution of methylamine in tetrahydrofuran were added. The
mixture was warmed to 0 1C and stirred for 15 min. The excess
of methylamine was removed by passing an argon gas stream
through the reaction vessel for 15 min. The organic phase was
then washed three times with 3% aqueous citric acid, three
times with sodium carbonate/sodium bicarbonate buffer (pH
10) and once with water, and finally dried over magnesium
sulfate. After evaporation of the solvent (membrane pump) 250
mg (810 mmol, 81%) of a colorless solid (N-tert-butyloxycar-
bonyl-O-methyl-(S)-tyrosine methyl amide, cf. Fig. 2a) were
obtained.
230 mg (746 mmol, 1.00 eq) N-tert-butyloxycarbonyl-O-
methyl-(S)-tyrosine methyl amide (cf. Fig. 2a) were stirred in
a mixture of 5 ml trifluoroacetic acid and 20 ml of dichlor-
omethane for 2.5 h with ice cooling. The solvent was removed
in vacuum with toluene as azeotropic mixture. After vacuum-
drying the product O-methyl-(S)-tyrosine methyl amide tri-
fluoracetate (cf. Fig. 2b) could be isolated in quantitative yield
as a ductile colorless oil.
190 mg (1.20 mmol, 1.20 eq) of N-acetyl-(S)-valine, 494 mg
(1.30 mmol, 1.30 eq) of the coupling reagent HATU, and 408
mg (3.00 mmol, 3.00 eq) of the coupling reagent HOAt, as well
as 350 ml (3.00 mmol, 3.00 eq) of lutidine were dissolved under
argon in a solvent mixture of dry dichloromethane/DMF (5:1)
and stirred for 10 min at 0 1C. Subsequently, a mixture of 322
mg (1.00 mmol, 1.00 eq) O-methyl-(S)-tyrosine methyl amide
trifluoroacetate (cf. Fig. 2b) and 107 ml (1.00 mmol, 1.00 eq) of
lutidine in little dichloromethane/DMF (5:1) were added. The
reaction mixture was initially stirred with ice-cooling for 2 h.
The organic layer was extracted three times with 1 M aqueous
KHSO4. The precipitating solid was filtered off and treated
separately. The organic layer was subsequently washed three
times with saturated sodiumbicarbonate and once with water
and with saturated sodium chloride solution, as well as the
solid. The organic layer was dried over sodium sulfate and the
solvent was completely removed at the rotatory evaporator.
Fig. 1 Structure of the tripeptide model Ac–Val–Tyr(Me)–NHMe.
The angles fV, cV and fT, cT define the backbone, whereas the angles
wV1 , wT1 and wT2 describe the side-chain orientation. The depicted con-
former is of the bLV(gꢀ)–bLT(a)1 type.
contain peptides or protected amino acids that undergo inter-
molecular hydrogen bonds like two amino acids in a b-sheet
secondary structure.11,12,14,20
The conformational properties of amino acids depend on
many different influences. The interaction with water plays an
important role in directing an amino acid or peptide into a
certain conformer. In order to investigate the process of
microsolvation the following questions have to be answered:
What will happen if water is added to an isolated peptide?
What is the preferred hydrogen bonding site? How many water
molecules have to be attached to cause a conformational change?
Clusters of amides and amino acids with water have been
investigated by different molecular beam experiments,21–28
but no hydrated peptides have been investigated up to now.
Snoek et al.25 have been the first to examine the hydrated
clusters of a free amino acid, tryptophan, with up to three
water molecules by applying IR/R2PI spectroscopy and UV/
UV hole-burning spectroscopy. In the case of the investigated
unprotected amino acid the water molecules are attached to the
polar end groups.
There are some experimental studies on larger neutral pep-
tides, like dipeptides or tripeptides. The first IR/R2PI spectrum
on a dipeptide has been published by our group choosing the
Ac–Val–Phe–OMe (Val ¼ valine) system.29 The IR spectra
have been recorded both in the region of the CH and NH as
well as the C O stretching modes. In very recent publications
Q
other groups also extended their applications of IR/R2PI
spectroscopy to peptides containing tryptophan and glycine30
or phenylalanine and proline.31 It should be mentioned that
UV/UV hole burning spectroscopy32 has also been applied to
some dipeptides.33,34 These method yields no direct informa-
tion on the structure of the electronic ground state, but the UV/
UV-hole burning spectroscopy is an ideal tool to distinguish
between different isomers especially in the case of congested
R2PI spectra. An overview on newest developments in the field
of biological molecules in the gas phase is given in ref. 35.
In this paper we extend our investigations on the model
tripeptide Ac–Val–Tyr(Me)–NHMe (cf. Fig. 1) and its cluster
with one water molecule. Tyr(Me) represents the amino acid
tyrosine with a methylated OH group in the side-chain. This
modification is important to avoid any hydrogen bonds via a
polar side-chain. Due to the use of a NHMe instead of an OMe
protecting group, Ac–Val–Tyr(Me)–NHMe contains three
amid groups and serves as model for a tripeptide. Since Val
and Tyr contain bulky side-chains both amino acids are ideal
candidates to form b-sheet related structures. The existence of
b-sheet related structures is important to build up an extended
b-sheet model. On the other hand the molecule gets more
flexible compared to Ac–Val–Phe–OMe, i.e. the additional
NH group may lead to intramolecular hydrogen-bonded (gL)
Fig.
2 (a) N-tert-butyloxycarbonyl-O-methyl-(S)-tyrosine methyl
amide and (b) O-methyl-(S)-tyrosine methyl amide trifluoracetate.
Both substances have been synthesized as precursor for the synthesis
of Ac–Val–Tyr(Me)–NHMe.
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 4 6 3 6 – 4 6 4 1
4637

View Article Online
The resulting solid was combined with the previously filtered
and thoroughly washed solid; the combined solids were va-
cuum dried, affording 158 mg (453 mmol, 45%) of the colorless
solid Ac–Val–Tyr(Me)–NHMe (cf. Fig. 1).
arrangements in the linear conformers of Ac–Val–Phe–OMe
and Ac–Val–Tyr(Me)–NHMe should be nearly identical. Ac-
cording to the results on Ac–Phe–NHMe14 and Ac–Trp–
NHMe16 it is known that the NH stretching vibration of the
NHMe protecting group is located around 3470 cmꢀ1. There-
fore the transition at 3477 cmꢀ1 is assigned to the N3–H3
(cf. Fig. 1) vibration of Ac–Val–Tyr(Me)–NHMe. Thus the
remaining transition at 3411 cmꢀ1 has to be correlated with
the N–H vibration of the tyrosine moiety. Unfortunately
the protected amino acid Ac–Tyr(Me)–NHMe has not been
investigated up to now. Thus a direct comparison of the NH
stretching frequency of the tyrosine moiety in Ac–Val–Tyr(Me)
–NHMe with the one in Ac–Tyr(Me)–NHMe is not possible
but the observations obtained from other protected amino
acids and peptides explain the relative low frequency of 3411
cmꢀ1 (compared to the value of 3451 cmꢀ1 of the Phe residue in
Ac–Val–Phe–OMe29): By comparing the IR spectra of the
protected amino acids Ac–Phe–OMe11 and Ac–Phe–NHMe14
as well as Ac–Trp–OMe37 and Ac–Trp–NHMe16 strong red-
shifts (20 and 25 cmꢀ1) of the NH stretching frequency of the
Phe and Trp residues are observed resulting from the change of
the OMe to NHMe protecting groups. By going from the
protected amino acid Ac–Phe–OMe to the protected dipeptide
Ac–Val–Phe–OMe an additional small red shift (7 cmꢀ1) of
the (Phe) NH stretching mode is observed, i.e. both the choice
of a NHMe protecting group and the extension of the
backbone (from a protected amino acid to a tripeptide model)
leads to a red-shift of the NH stretching frequency of the
tyrosine residue.
(b) Experimental set-up
The experimental set-up has been described elsewhere (cf. refs.
11,12,36). Thus only a short description is given: The R2PI and
IR/R2PI spectra were measured in a vacuum system consisting
of a differentially pumped linear time-of-flight mass spectro-
meter and a pulsed valve (General Valve Iota One, 500 mm
orifice) for skimmed jet expansion (X/D ¼ 130). A frequency-
doubled dye laser (Lumonics HD 300), pumped by a Nd:YAG
laser (Lumonics HY 400), was used for excitation to the S1
state and for ionization. The IR light in the region of 2.63–3.08
mm (3250–3800 cmꢀ1) was generated with a LiNbO3 crystal by
difference frequency mixing of the fundamental (1064 nm) of a
seeded Nd:YAG laser (Spectra-Physics PRO-230) and the
output of a dye laser (Sirah, Precision Scan) pumped by the
second harmonic (532 nm) of the same Nd:YAG laser. The IR
output is amplified by an optical parametric amplification (in a
LiNbO3 crystal) of the output of the IR laser (2.63–3.08 mm)
and the fundamental of the Nd:YAG laser.36
IR light in the region of 6 mm is generated by a third
nonlinear process12,15 The signal (1.79–1.82 mm, 5500–5600
cmꢀ1) and idler (2.56–2.63 mm, 3800–3900 cmꢀ1) of the
OPA process are the inputs to a difference frequency mixing
process in an AgGaSe2 crystal leading to the IR light of 1600–
1800 cmꢀ1
.
The assignment to a linear b-sheet related structure is further
Since the time delay chosen for the two lasers is not longer
than 100 ns, all lasers have been spatially overlapped. In order
to obtain IR/R2PI spectra the IR laser is fired 60 ns prior to the
UV laser. The substance (Ac–Val–Tyr(Me)–NHMe) and the
valve are heated to 150 1C. Helium was used as carrier gas
(2000 mbar).
supported by vibrational frequencies obtained for the C
O
Q
groups of Ac–Val–Tyr(Me)–NHMe. The IR/R2PI spectrum
exhibits three transitions at 1694, 1700, and 1715 cmꢀ1. The
transition at 1715 cmꢀ1 is in good agreement with frequencies
of the C O (acetyl) group (cf. Fig. 1) observed for Ac–Val–
Q
1
1
Phe–OMe (1711 cmꢀ1 29). Furthermore, the value of 1694 cmꢀ1
is nearly identical to the frequency observed for the corre-
sponding valine residue of Ac–Val–Phe–OMe (1696 cmꢀ1 29).
Thus the vibration at 1694 cmꢀ1 can easily be assigned to the
III. Results and discussion
(a) Structure of Ac–Val–Tyr(Me)–NHMe
stretching vibration of the C O group of Ac–Val–Tyr(Me)–
2
Q
2
NHMe. Finally the remaining transition at 1700 cmꢀ1 has to
be correlated with the C O stretching mode (cf. Fig. 1) of
the Tyr residue.
The IR/R2PI spectrum of Ac–Val–Tyr(Me)–NHMe in the
range of 1600–1800 and 3300–3600 cmꢀ1 is depicted in Fig. 3.
The spectrum has been recorded via the electronic origin of
the S1 ’ S0 transition at 35 434 cmꢀ1. No further prominent
transitions can be observed in the R2PI spectrum from 35 200
to 35 900 cmꢀ1. In the region of the NH stretching modes of
the IR/R2PI spectrum three vibrational transitions are ob-
served at 3411, 3440 and 3477 cmꢀ1. All vibrations are located
above 3400 cmꢀ1 which strongly indicates that a linear b-sheet
related structure and no intramolecular hydrogen bonds are
formed. The frequencies of hydrogen-bonded NH stretching
Q
3
3
In order to support the vibrational assignments force field,
ab initio (Hartree–Fock, HF) and DFT (B3LYP functional)
calculations have been performed using the Discover38 and
Gaussian 9839 programs. To generate starting geometries for
ab initio calculations, a preselection via molecular dynamics
using the CFF force field38,40–43 has been done. Molecular
dynamics have been performed at a temperature of 750 K and
with a time step of 1 fs. After 1000 steps the structure has been
minimized. This structure has been used to continue the
dynamics in the same manner. The following minimized struc-
ture has been compared to the first via its relative energy to
discriminate between different minima. With this technique
30,000 structures have been minimized and compared to each
other. The dynamics have been repeated several times to
explore the PES. The preselection using the CFF force field
shows that b-sheet related structures turn out to be the most
stable conformers. This is in agreement with the analysis of the
IR spectra indicating to b-sheet related structures, which can
differ in their side-chain arrangements. In a general overview
on several peptides and clusters44 we will show that the class II
force field CFF can be taken for the interpretation of ‘‘high
resolution’’ IR spectra of isolated species.
modes are typically located below 3400 cmꢀ1 11,14,16
The NH
.
stretching frequency at 3440 cmꢀ1 is very close to the value of
the NH stretching mode of Val at 3441 cmꢀ1 in the spectrum of
Ac–Val–Phe–OMe29 and can therefore be assigned to the NH
stretching mode of the valine moiety. The identity of the
observed frequencies is not surprising, since the structural
Due to the size of the system and in order to compare the
results with earlier studies, we have chosen the 3-21G(d) basis
set for the subsequent HF calculations. All structures with b-
sheet related backbone arrangements have been fully optimized
and the relative energies have been corrected by the zero point
Fig. 3 IR/R2PI spectrum of Ac–Val–Tyr(Me)–NHMe in the region
from 1600–1800 cmꢀ1 and from 3300–3600 cmꢀ1
.
4638
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 4 6 3 6 – 4 6 4 1
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
energies. The calculated frequencies have been scaled with
factors derived from our investigations on Ac–Phe–OMe
(0.9076 for the NH vibrations and 0.9140 for the C
O
Q
stretching modes12). The most stable isomer has a bV(gꢀ)–
bT(a) (superscripts: V ¼ valine, T ¼ tyrosine) arrangement
(cf. Fig. 1). The torsional angles characterizing the backbone
and side-chain conformations have the following values: fV(C1–
N1–CVa –C2) ¼ ꢀ1401, cV(N1–CaV–C2–N2) ¼ 1641, wV1 (N1–CaV–
CVb –CVg ) ¼ ꢀ681, fT(C2–N2–CTa–C3) ¼ ꢀ1661, cT(N2–CaT–C3–
N3) ¼ 1731, wT1 (N2–CaT–CbT–CTg ) ¼ ꢀ1471. It has to be men-
tioned that two structures of all bV–bT conformers have to be
discussed which only differ by the orientation of the OMe
group in the tyrosine side-chain. If the angle wT2 (cf. Fig. 1) has
a positive (negative) value the structure is characterized by
aþ(ꢀ) sign. All ꢁ structures of a given conformer have nearly
identical energies. The structure shown in Fig. 1 is a bV(gꢀ)–
bT(a)1 arrangement with a torsional angle of w2T (CaT–CbT–CTg –
CTd ) ¼ 681. In the bV(gꢀ)–bT(a)ꢀ structure the OMe group
would be rotated about the C–O axis by 1801.
Fig. 4 IR/R2PI spectrum of Ac–Val–Tyr(Me)–NHMe(H2O)1 in the
region from 3250–3800 cmꢀ1
.
interpretation of the spectrum the experimentally observed NH
stretching frequencies are compared with the one of the Ac–Val–
Tyr(Me)–NHMe monomer; in addition the OH stretching
frequencies of the water moiety have to be discussed.
The values f and c of the bV(gꢀ)–bT(a)ꢁ structures deviate
not more than 71 from the values obtained for the minimum
energy structure of Ac–Val–Phe–OMe.29 There are two dif-
ferent structural arrangements (bV(gþ)–bT(a)ꢁ; bV(a)–bT(a)ꢁ)
which are close in energy to the most stable structure of Ac–Val–
Tyr(Me)–NHMe, i.e. the energy difference between the struc-
tures is about 100 cmꢀ1 including zero point energy correc-
tions. The calculation of relative energies cannot be
satisfactory at the HF/3-21G(d) level of theory and more
sophisticated theoretical applications have to be performed in
order to define the global minimum energy structure of Ac–Val
–Tyr(Me)–NHMe. In contrast to the relative energies the
scaled frequencies obtained from calculations at the HF/3-
21G(d) level of theory are often in good agreement with the
experimental values which is a result of a cancellation of errors.
The vibrational frequencies obtained for the bV(gꢀ)–bT(a)1
structure are listed in Table 1 yielding a reasonable agreement
to the experimentally observed frequencies. Additionally HF
calculations using the 6-31G(d,p) basis and DFT calculations
(B3LYP functional, 6-31þG(d) basis set) have been performed
for the bV(gꢀ)–bT(a)1 structure. The following dihedral angles
of the backbone structure are obtained from the HF/
6-31G(d,p) (B3LYP/6-31þG(d)) calculations: fV ¼ ꢀ1291
(ꢀ1301), cV ¼ 1601 (1611), wV1 ¼ ꢀ631 (ꢀ621), fT ¼ ꢀ1571
(ꢀ1571), cT ¼ 1721 (1601), w1T ¼ ꢀ1691 (ꢀ1671), wT2 ¼ 741 (701).
The resulting vibrational frequencies are also listed in Table 1.
The frequencies obtained from the DFT calculations strongly
support our vibrational assignments.
A water molecule can be attached as hydrogen acceptor (by
forming an N–Hꢂ ꢂ ꢂO bond) or as hydrogen donor (by forming
an C Oꢂ ꢂ ꢂH bond). These arrangements lead to significantly
Q
different OH stretching frequencies. In case of a C Oꢂ ꢂ ꢂH
Q
hydrogen donor bond an OH donor stretching frequency in the
region of about 3450–3600 cmꢀ1 would be expected, e.g. in
23
trans-N-phenylformamide(H2O)1
mide22 the vibrations are located at 3536 and 3525 cmꢀ1
or trans-N-benzylforma-
,
respectively. Similar results are obtained in investigations on
other hydrated clusters with aromatic chromophores45 or in
our investigations on Ac–Phe–OMe(H2O)1, a hydrated cluster
of a protected amino acid with water.46 In Ac–Phe–OMe
(H2O)1 two isomers are discussed. The first one contains a C
Oꢂ ꢂ ꢂH donor bond and in the second one a water molecule
Q
forms a bridge between a N–H (hydrogen bond acceptor) and a
C O group (hydrogen bond donor). Also for the second
Q
isomer containing a combined donor/acceptor binding motif
the OH stretching frequency is significantly below 3600 cmꢀ1
(at 3575 cmꢀ1 46). If water acts as proton donor the remaining
free OH stretching mode would be expected around 3720 cmꢀ1
(cf. e.g. refs. 6,8,45,47). The experimental spectrum shows two
transitions in the OH stretching region at 3635 and 3718 cmꢀ1
(cf. Fig. 4). The value observed for the free OH stretching mode
is in very good agreement with the value expected for a donor
bonded H2O molecule but no transition can be observed
between 3450 and 3600 cmꢀ1. Thus no OH stretching donor
vibration can be assigned. On the other hand the transition at
3635 cmꢀ1 is only shifted by ꢀ22 cmꢀ1 with respect to the free
symmetric vibration of isolated water48,49 strongly indicating
that water cannot act as a donor. If the water molecules acts as
(b) Structure of Ac–Val–Tyr(Me)–NHMe(H2O)1
The IR/R2PI spectrum of Ac–Val–Tyr(Me)–NHMe(H2O)1
in the spectral region from 3250 to 3800 cmꢀ1 is shown in
Fig. 4. The spectrum has been recorded via the electronic origin
of the S1 ’ S0 transition at 35636 cmꢀ1. In order to get an
proton acceptor like in indole(H2O)1,50 phenol(H2O)1 or N-
6
phenylformamide(H2O)1 a frequency of about 3650 cmꢀ1 is
23
obtained for the symmetric OH stretching mode. In some cases
Table 1 Comparison of the experimentally observed vibrational frequencies with the one obtained at the Hartree–Fock and DFT (B3LYP
functional, 6-31þG(d) basis set) level of theory. The scaling factors for the HF calculations (3-21G(d) and 6-31G(d,p) basis sets) are derived from
earlier studies on Ac–Phe–OMe:11,12 0.9076 (0.8881) for NH stretching vibrations and 0.9140 (0.8771) for C O stretching vibrations. The NH
stretching vibrations obtained from DFT calculations are scaled with a factor taken from ref. 16 (0.9605). The C O stretching modes are scaled
Q
Q
with respect to the C O stretching vibration (0.9755)
Q
1
Calculated frequencies
Type of vibration
HF/3-21G(d)
HF/6-31G(d,p)
DFT/B3LYP 6-31þG(d)
Experimental frequencies
NH (Val, N1–H1)
NH (Tyr, N2–H2)
NH (NHMe, N3–H3)
3451
3433
3464
1726
1664
1707
3439
3440
3463
1718
1677
1701
3434
3415
3467
1715
1672
1698
3440
3411
3477
1715
1694
1700
CO (Acetyl, C O )
Q
1
CO (Val, C O )
Q
2
CO (Tyr, C O )
Q
3
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 4 6 3 6 – 4 6 4 1
4639

View Article Online
this value can also shift to frequencies of about 3635 cmꢀ1, e.g.
in the p-bonded fluorene(H2O)1 cluster (3632 cmꢀ1 51). It is
interesting to note that a p-interaction can also be discussed for
the Ac–Val–Tyr(Me)–NHMe(H2O)1 cluster (see below).
Concerning the asymmetric OH stretching vibration its
frequency is often located at about 3745 cmꢀ1 6,8,50
wavenumbers lower than the corresponding frequency of the
free water molecule at 3756 cmꢀ1 48,49
But there are also some
,
i.e. a few
.
acceptor bound species which show a much stronger shift down
to about 3710 cmꢀ1, cf. e.g. 2-naphthyl-1-ethanol(H2O)2 with
an asymmetric OH stretching vibration of the acceptor water
at 3708 cmꢀ1 52
This water molecule also undergoes a p-
.
interaction with the aromatic chromophore. Additionally the
51
p-bonded fluorene(H2O)1 is an example which shows a shift
of the asymmetric OH stretching mode to 3719 cmꢀ1
.
Furthermore, it is interesting that the splitting of symmetric
(n1) and asymmetric (n3) OH stretching mode in Ac–Val–
Tyr(Me)–NHMe(H2O)1 is only 83 cmꢀ1. This value is lower
than the corresponding difference of an isolated water molecule
(99 cmꢀ1 48,49) or other acceptor bound clusters6,8,45,50, but a
similar low splitting has also been observed for the water dimer
(about 85 cmꢀ1 53) and for the fluorene(H2O)1 cluster (87 cmꢀ1
51). It should be noted that our ab initio calculations performed
at the HF/6-31G(d,p) level of theory (see below) predict a
splitting of 87 cmꢀ1, if the difference is scaled to the corre-
sponding value obtained for the free water molecule calculated
at the same level of theory, i.e. the n1/n3 splitting is significantly
reduced in the complex of Ac–Val–Tyr(Me)–NHMe with
water. This is in agreement with the experimental result.
According to the discussion of the OH stretching vibrations
it can be assumed that the water molecule undergoes a hydro-
gen acceptor bond, i.e. water is attached to one of the three NH
groups and forms a N–Hꢂ ꢂ ꢂO hydrogen bond. In order to find
out which NH group is hydrogen-bonded the remaining transi-
tions at 3393, 3418 and 3435 cmꢀ1 (cf. Fig. 4) are compared
with the NH stretching frequencies of the Ac–Val–Tyr(Me)–
NHMe monomer. It is obvious that the bands at 3435 and 3418
cmꢀ1 show only small shifts (ꢀ5 and þ7 cmꢀ1) with respect to
the corresponding transitions of the monomer (3440 and 3411
cmꢀ1) and are therefore assigned to free NH stretching modes
of the valine and tyrosine residues. Since no transition is
observed in the region from 3435 to 3635 cmꢀ1, the free NH
stretching vibration of the N3–H3 group (located at 3477 cmꢀ1
in the monomer) can definitely not be observed. This leads to
the conclusion that the water molecule is hydrogen-bonded to
the N3–H3 group of the NHMe protecting group (cf. Fig. 1),
yielding a broader N–H stretching transition at 3393 cmꢀ1. The
experimentally observed shift (84 cmꢀ1) of the N3–H3 stretch-
ing vibration between monomer and cluster is in agreement
with the value obtained for the N-phenylformamide cluster.23
It can be concluded that the IR/R2PI spectrum of Ac–Val–
Tyr(Me)–NHMe(H2O)1 can be interpreted by predicting a
structure which contains a water molecule undergoing a N3–
H3ꢂ ꢂ ꢂO acceptor bond. As mentioned above the NH stretching
frequencies of the Val and Tyr residues are very close to their
corresponding values in the monomer. Thus it is very likely to
assume that the conformation of the backbone has not chan-
ged significantly. Ab initio calculations at the HF level of
theory are performed using the same basis sets (3-21G(d) and
6-31G(d,p)) as for the monomer. According to the experimen-
tal results a b-sheet related structure with water attached to the
N3–H3 group has been used as starting geometry. Subse-
quently, structures including all possible side-chain configura-
tions have been optimized at the HF/3-21G(d) level of theory.
As in the monomer a bV(gꢀ)–bT(a) conformer turns out to be
the most stable structure (cf. Fig. 5). Thus this bV(gꢀ)–bT(a)
structure has been further optimized at the HF/6-31G(d,p)
level yielding a complex with an almost linear intermolecular
hydrogen bond, i.e. the angle between N3, H3 (cf. Fig. 1)
and the oxygen atom of the water molecule is 1741, the
Fig.
5
(H2O)1 as obtained from HF/6-31G(d,p) calculations.
Structure of the binary cluster Ac–Val–Tyr(Me)–NHMe
N3–H3ꢂ ꢂ ꢂOH2 bond length is 2.04 A. One hydrogen atom of
the water molecule lies above the center of the aromatic ring
(Tyr residue, cf. Fig. 5). This p-interaction may lead to an
additional stabilization of this structure. The torsional angles
of backbone and side-chain of the Val moiety have identical
values with respect to the monomer. Due to the influence of the
attached water at the NHMe group, the torsional angles of the
Tyr residue show a small shift (21 for fT, 161 for cT and 01 for
wT1 ), but the bL-arrangement is still conserved.
The frequencies calculated for the NH stretching vibrations
of the N3–H3 group as well as the Tyr and Val moieties (3400,
3440 and 3441 cmꢀ1) are in agreement with the experimental
frequencies (3393, 3418 and 3435 cmꢀ1). The calculated fre-
quencies have been scaled with the same factor as the one
obtained for the monomer at the same level of theory.
Furthermore, a stabilization energy of 1288 cmꢀ1 (15.4 kJ
molꢀ1) is obtained for the cluster at the HF/6-31G(d,p) level
including BSSE corrections.54
IV. Conclusion
The tripeptide model Ac–Val–Tyr(Me)–NHMe is an extended
molecule that is flexible enough to form several structural
conformers, both stretched (b-sheet related) and intramolecu-
lar hydrogen-bonded arrangements. From our spectroscopic
investigations a b-sheet related arrangement can be derived.
This result indicates a high selectivity with respect to b-sheet
related structures if appropriate amino acids are chosen.
According to these results Ac–Val–Tyr(Me)–NHMe is a good
candidate for b-sheet model systems, i.e. intermolecular hydro-
gen-bonded cluster with other peptides can be formed leading
to a structural configuration similar to the one observed
between peptides in an extended b-sheet.
It is interesting to note that the monomer seems only slightly
influenced by the addition of one water molecule. Water forms
an almost linear hydrogen bond to the NH group of NHMe
which is the strongest donor of the molecule. It will be a further
question, whether the addition of more water molecules will
change the backbone configuration of the peptide.
Acknowledgements
The authors thank the Deutsche Forschungsgemeinschaft
(DFG) for financial support and the Rechenzentrum der
‘‘Heinrich-Heine-Universitat Dusseldorf’’ for the granted
computer time.
¨
¨
References
1
T. R. Rizzo, Y. D. Park, L. A. Peteanu and D. H. Levy, J. Chem.
Phys., 1986, 84, 2534.
4640
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 4 6 3 6 – 4 6 4 1
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
2
3
4
5
6
S. J. Martinez III, J. C. Alfano and D. H. Levy, J. Mol. Spectrosc.,
1992, 156, 421.
R. Cable, M. J. Tubergen and D. H. Levy, J. Am. Chem. Soc.,
1988, 110, 7349.
R. H. Page, Y. R. Shen and Y. T. Lee, J. Chem. Phys., 1988, 88,
4621.
C. Riehn, C. Lahmann, B. Wassermann and B. Brutschy, Chem.
Phys. Lett., 1992, 197, 3197.
S. Tanabe, T. Ebata, M. Fujii and N. Mikami, Chem. Phys. Lett.,
1993, 215, 347.
T. S. Zwier, Anal. Chem., 1996, 47, 205, and references therein.
M. Gerhards, C. Unterberg and K. Kleinermanns, Phys. Chem.
Chem. Phys., 2000, 2, 5538.
32 R. J. Lipert and S. D. Colson, J. Phys. Chem., 1989, 93, 3894.
33 R. Cohen, B. Brauer, E. Nir, L. Grace and M. S. de Vries, J. Phys.
Chem. A, 2000, 104, 6351.
34 I. Hunig, K. A. Seefeld and K. Kleinermanns, Chem. Phys. Lett.,
¨
2003, 369, 173.
35 J. P. Simons, Phys. Chem. Chem. Phys., 2004, 6((10)), E7, Bio-
active molecules in the gas phase special issue.
36 C. Unterberg, A. Jansen and M. Gerhards, J. Chem. Phys., 2000,
113, 7945.
37 A. Gerlach, C. Unterberg and M. Gerhards, in preparation.
38 CFF Consistent Force Field and DISCOVER Molecular Simulation
Program, Version 2000, Accelrys Inc., San Diego, CA, 2000.
39 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zarkrzewski, J. A. Montegomery
Jr, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A.
D. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M.
Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G. A. Peterrson, P. Y. Ayala, Q. Cui, K. Morokuma,
D. K. Malick, A. D. Rabuck, K. Raghhavachari, J. B. Foresman,
J. Ciosloski, J. V. Ortiz, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C.
Gonzales, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, J. L. Andres, C. Gonzales, M. Head-Gordon, E. S.
Replogle, J. A. Pople, GAUSSIAN 98, (Revision A.6), Gaussian
Inc., Pittsburgh, PA, 1998.
7
8
9
L. C. Snoek, E. G. Robertson, R. T. Kroemer and J. P. Simons,
Chem. Phys. Lett., 2000, 321, 49.
10 M. Mons, I. Dimicoli, B. Tardivel, F. Piuzzi, E. G. Robertson and
J. P. Simons, J. Phys. Chem. A, 2001, 105, 969.
11 M. Gerhards and C. Unterberg, Phys. Chem. Chem. Phys., 2002,
4, 1760.
12 M. Gerhards, C. Unterberg and A. Gerlach, Phys. Chem. Chem.
Phys., 2002, 4, 5563.
13 J. M. Bakker, L. M. Aleese, G. Meijer and G. v. Helden, Phys.
Rev. Lett., 2003, 91, 203003.
14 M. Gerhards, C. Unterberg, A. Gerlach and A. Jansen, Phys.
Chem. Chem. Phys., 2004, 6, 2682.
15 M. Gerhards, Opt. Commun., in press.
16 B. C. Dian, A. Longarte, S. Mercier, D. A. Evans, D. J. Wales and
T. S. Zwier, J. Chem. Phys., 2002, 117, 10688.
17 B. C. Dian, A. Longarte and T. S. Zwier, Science, 2002, 296, 2369.
18 G. N. Ramachandran, V. Sasisekharan and C. Ramakrishnan,
Biochim. Biophys. Acta, 1966, 112, 168.
40 J. R. Maple, M.-J. Hwang, T. P. Stockfisch, U. Dinur, M.
Waldmann, C. S. Ewig and A. T. Hagler, J. Comput. Chem.,
1994, 15, 162.
41 M. J. Hwang, T. P. Stockfisch and A. T. Hagler, J. Am. Chem.
Soc., 1994, 116, 2515.
19 A. Perczel, J. G. Angyan, M. Kajtar, W. Viviani, J. L. Rivail, J. F.
Marcoccia and I. G. Csizmadia, J. Am. Chem. Soc., 1991, 113, 6256.
20 C. Unterberg, A. Gerlach, T. Schrader and M. Gerhards, Eur.
Phys. J. D, 2002, 20, 543.
21 J. A. Dickinson, M. R. Hockridge, E. G. Robertson and J. P.
Simons, J. Phys. Chem. A, 1999, 103, 6938.
22 E. G. Robertson, M. R. Hockridge, P. D. Jelfs and J. P. Simons,
J. Phys. Chem. A, 2000, 104, 11714.
23 E. G. Robertson, Chem. Phys. Lett., 2000, 325, 299.
24 E. G. Robertson, M. R. Hockridge, P. D. Jelfs and J. P. Simons,
Phys. Chem. Chem. Phys., 2001, 3, 786.
42 J. R. Maple, M.-J. Hwang, K. J. Jalkanen, T. P. Stockfisch and
A. T. Hagler, J. Comput. Chem., 1998, 19, 430.
43 C. S. Ewig, R. Berry, U. Dinur, J. R. Hill, M. J. Hwang, H. Li, C.
Liang, J. Maple, Z. Peng, T. P. Stockfisch, T. S. Thacher, L. Yan,
X. Ni and A. T. Hagler, J. Comput. Chem., 2001, 22, 1782.
44 A. Jansen, H. Fricke and M. Gerhards, J. Am. Chem. Soc.,
submitted.
45 H.-D. Barth, K. Buchhold, S. Djafari, B. Reimann, U. Lom-
matzsch and B. Brutschy, Chem. Phys., 1998, 239, 49.
46 H. Fricke, C. Unterberg, A. Gerlach and M. Gerhards, in
preparation.
25 G. M. Florio and T. S. Zwier, J. Phys. Chem. A, 2003, 107, 974.
26 L. C. Snoek, R. T. Kroemer and J. P. Simons, Phys. Chem. Chem.
Phys., 2002, 4, 2130.
47 G. M. Florio, C. J. Gruenloh, R. C. Quimpo and T. S. Zwier,
J. Chem. Phys., 2000, 113(24), 11143.
48 P. E. Fraley and K. N. Rao, J. Mol. Spectrosc., 1969, 29, 348.
49 J. M. Flaud, C. Camy-Peyret and J. P. Maillard, Mol. Phys., 1976,
32, 449.
50 J. R. Carney, F. C. Hagemeister and T. S. Zwier, J. Chem. Phys.,
1998, 108(9), 3379.
51 N. Guchhait, T. Ebata and N. Mikami, J. Phys. Chem. A, 2000,
104, 11891.
27 F. Lecomte, B. Lucas, G. Gre
Desfrancois, Eur. Phys. J. D, 2002, 20, 449.
28 B. Lucas, F. Lecomte, B. Reimann, H.-D. Barth, G. Gre
Bouteiller, J.-P. Schermann and C. Desfrancois, Phys. Chem.
Chem. Phys., 2004, 6, 2600.
´
goire, J. P. Schermann and C.
¸
´
goire, Y.
¸
29 C. Unterberg, A. Gerlach, T. Schrader and M. Gerhards, J. Chem.
Phys., 2003, 118(18), 8296.
52 N. Seurre, J. Sepioz, F. Lahmani, A. Zehnacker-Rentien and K. Le
Barbu-Debus, Phys. Chem. Chem. Phys., 2004, 610.1039/
b405276e.
30 I. Hunig and K. Kleinermanns, Phys. Chem. Chem. Phys., 2004
¨
6, 2650.
31 W. Chin, M. Mons, J. P. Dognon, F. Piuzzi, B. Tardivel and
I. Dimicoli, Phys. Chem. Chem. Phys., 2004, 6, 2700.
53 U. Buck and F. Huisken, Chem. Rev., 2000, 100, 3863.
54 S. F. Boys and D. Bernardi, Mol. Phys., 1970, 19, 553.
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 4 6 3 6 – 4 6 4 1
4641