ADVANCED VIBRATIONAL
SPECTROSCOPIC STUDIES OF
BIOLOGICAL MOLECULES
A thesis submitted to the University of Manchester for the degree of
Doctor in Philosophy in the Faculty of Life Sciences
2012
Saeideh Ostovar pour
Table of Contents
List of Contents ............................................................................................................... 2
List of tables ..................................................................................................................... 6
List of figures ................................................................................................................... 7
Abstract .......................................................................................................................... 11
Declaration ..................................................................................................................... 12
Copyright statement...................................................................................................... 13
Abbreviations List ......................................................................................................... 14
Glossary .......................................................................................................................... 15
Acknowledgements ........................................................................................................ 17
Chapter 1
1.0 Introduction ............................................................................................................... 19
1.1 Conformational analysis of biological molecules ..................................................... 19
1.2 Infrared and Raman spectroscopies .......................................................................... 20
1.2.1 Principles of Raman spectroscopy .................................................................. 23
1.2.2 Resonance Raman spectroscopy ..................................................................... 24
1.2.3 Raman spectroscopy of biomolecules ............................................................. 25
1.3 Raman Optical Activity (ROA) ................................................................................ 28
1.3.1 ROA spectroscopy of biomolecules ................................................................ 32
1.4 Surface-Enhanced Raman Scattering ........................................................................ 34
1.4.1 The Electromagnetic Mechanism.................................................................... 35
1.4.2 The Chemical Mechanism (CE) ...................................................................... 36
1.4.3 Biological applications of SERS ..................................................................... 37
1.5 Surface enhanced Raman optical activity (SEROA) ................................................ 39
1.6 References ................................................................................................................. 45
Chapter 2
Use of a Hydrogel Polymer for Reproducible Surface Enhanced Raman Optical
Activity (SEROA)
2.0 Declaration ................................................................................................................ 50
2
Table of Contents
2.1 Abstract ..................................................................................................................... 51
2.2 Introduction ............................................................................................................... 51
2.3 Experimental ............................................................................................................. 53
2.4 Results and Discussion.............................................................................................. 54
2.5 Conclusion ................................................................................................................. 57
2.6 References ................................................................................................................. 59
2.7 Supplementary Information ...................................................................................... 63
2.7.1 Colloid Preparation ......................................................................................... 63
2.7.2 Sample Preparation for Raman and ROA Measurements ............................... 63
2.7.3 Atomic Force Microscopy .............................................................................. 64
2.7.4 SERS Time Dependence ................................................................................. 66
Chapter 3
Induced Chirality to Non-chiral Surfaces of Silver Silica Nanotags
3.0 Declaration ................................................................................................................ 71
3.1 Abstract ..................................................................................................................... 72
3.2 Introduction ............................................................................................................... 73
3.3 Experimental ............................................................................................................. 75
3.4 Results and Discussion.............................................................................................. 77
3.5 Conclusion ................................................................................................................. 82
3.6 References ................................................................................................................. 83
3.7 Supplementary Information ...................................................................................... 88
Chapter 4
Phosphorylation Detection and Characterization in
Ribonucleotides Using Raman and Raman Optical Activity (ROA)
Spectroscopies
4.0 Declaration ................................................................................................................ 90
4.1 Abstract ..................................................................................................................... 91
4.2 Introduction ............................................................................................................... 91
4.3 Experimental ............................................................................................................. 93
4.4 Results and Discussion.............................................................................................. 94
4.5 Conclusion ................................................................................................................. 99
3
Table of Contents
4.6 Acknowledgement................................................................................................... 100
4.7 References ............................................................................................................... 101
4.8 Supplementary Information .................................................................................... 107
4.8.1 Colloid Preparation ....................................................................................... 107
4.8.2 Surface enhanced Raman spectroscopy (SERS) ........................................... 107
4.8.3 References ..................................................................................................... 109
Chapter 5
Study of Experimental and Computational Raman and Raman Optical Activity
(ROA) Spectra of Cyclic and Linear L-Ala-L-Ala in Solution
5.0 Declaration .............................................................................................................. 113
5.1 Abstract ................................................................................................................... 114
5.2 Introduction ............................................................................................................. 115
5.3 Experimental ........................................................................................................... 117
5.4 Computational methods .......................................................................................... 118
5.5 Results and Discussion............................................................................................ 119
5.6 Conclusion .............................................................................................................. 125
5.7 References ............................................................................................................... 129
5.8 Supplementary Information .................................................................................... 135
Chapter 6
Combined Experimental and Computational Study of Raman and Raman
Optical Activity (ROA) Spectra of Linear and Cyclic L-Ser-L-Ser in Solution
6.0 Declaration .............................................................................................................. 140
6.1 Abstract ................................................................................................................... 141
6.2 Introduction ............................................................................................................. 142
6.3 Experimental ........................................................................................................... 143
6.4 Computational methods .......................................................................................... 144
6.5 Results and Discussion............................................................................................ 146
6.6 Conclusion .............................................................................................................. 150
6.7 References ............................................................................................................... 152
6.8 Supplementary Information .................................................................................... 161
4
Table of Contents
Chapter 7
7.0 Conclusion ............................................................................................................. 166
7.1 Future work ............................................................................................................ 170
7.2 References .............................................................................................................. 172
Chapter 8
Appendix
8.0 Declaration .............................................................................................................. 174
37,600 words
5
List of Figures
List of Tables
Table 1.1
Advantages and disadvantages of Raman spectroscopy use ...... 26
Table 1.2
Important spectral regions of protein vibrations ......................... 27
Table S2.1
Raman and ROA band assignments of L- and D-ribose in aqueous
solution........................................................................................ 68
Table 4.1
Raman band assignments of adenosine, AMP, ADP, ATP, A(2)MP,
A(2,3)MP, A(3)MP and A(3,5)MP ............................................. 105
Table 4.2
ROA band assignments of adenosine, AMP, ADP, ATP, A(3,5)MP,
A(2,3)MP, A(2)MP and A(3)MP ................................................ 106
Table S4.1
SERS band assignments of adenosine, AMP, ADP, ATP, A(2)MP,
A(2,3)MP, A(3)MP and A(3,5)MP ............................................. 111
Table 5.1
Calculated and experimental wavenumber band assignments for
Raman and ROA of cyclic and linear L-Ala-L-Ala in H2O........ 131
Table 5.2
Calculated and experimental wavenumber band assignments for
Raman and ROA of cyclic and linear L-Ala-L-Ala in D2O........ 133
Table S5.1
Calculated and experimental bond lengths (Å) for cyclic and linear
L-Ala-L-Ala ................................................................................ 135
Table S5.2
Calculated and experimental bond angles (o) for cyclic and linear
L-Ala-L-Ala ................................................................................ 136
Table S5.3
Calculated and experimental torsion angles (o) for cyclic and linear
L-Ala-L-Ala ................................................................................ 137
Table 6.1
Calculated and experimental Raman and ROA bands for cyclic and
linear L-Ser-L-Ser in H2O........................................................... 157
Table 6.2
Calculated and experimental Raman and ROA bands for cyclic and
linear L-Ser-L-Ser in D2O........................................................... 159
Table S6.1
Calculated and experimental bond lengths (Å) for cyclic and linear
L-Ser-L-Ser ................................................................................. 161
Table S6.2
Calculated bond angles (o) for cyclic and linear L-Ser-L-Ser .... 161
Table S6.3
Calculated torsion angles (o) for cyclic and linear L-Ser-L-Ser . 163
6
List of Figures
List of Figures
Figure 1.1
An energy level diagram showing the transitions involved in
Raman scattering......................................................................... 22
Figure 1.2
Schematic diagram of the basic ROA experiment which measures a
small difference in the intensity of Raman scattering in right (R)
and left (L) circularly polarized light from chiral molecules ...... 28
Figure 2.1
Raman (IR + IL, top) and SCP ROA (IR – IL, bottom) spectra of Dand L-ribose in aqueous. Sample concentrations were 2.66 M at pH
5.46 (D-) and 5.60 (L-ribose), data collection time was 237.02 min,
and laser power at the sample 0.625 W for each ........................ 60
Figure 2.2
Raman spectrum of polycarbopol in solution (A), SERS spectra
before (B) and after addition of L- and D-ribose (0.25 mg ml-1) in
the presence of silver citrate reduced colloid and K2SO4 at 0.020 M
concentration, data collection time: 20 min (C), SERS spectra of Land D-ribose (0.25 mg ml-1) in the presence of polycarbopol
polymer, data collection time: 20 min (D), ROA spectrum of
polycarbopol polymer in solution, sample concentration 40 mg ml1
, data collection time: 218 min (E), SEROA spectra of silver
citrate reduced colloids in presence of aggregating salt before (F)
and after addition of L- and D-ribose, data collection time: 35 min
(G), SEROA spectra of L- and D-ribose with addition of
polycarbopol, datacollection time of 35 min (H) ........................ 61
Figure S2.1
A schematic diagram of SEROA sample preparation with use of the
polycarbopol polymer ................................................................. 63
Figure S2.2
Dual views of AFM images of silver citrate reduced colloids only
(A), polycarbopol polymer only (B) and a mixture of silver citrate
reduced colloids and polycarbopol polymer (C) ......................... 65
Figure S2.3
Time dependence SERS study for D-ribose molecule with and
without addition of polymer to SERS solution ........................... 66
Figure S2.4
Example repeats of SEROA spectra for D- (top) and L-ribose
(bottom) with the same conditions for each spectrum, where
concentration of each sample was 0.25 mg/ml and K2SO4
7
List of Figures
concentration was 0.020 M, with addition of 20 mg/mL of
polycarbopol, data collection time of 35 mins ............................ 67
Figure S2.5
Raman (IR + IL, A and B) and SCP ROA (IR – IL, E and F) spectra
of D- and L-tryptophan in aqueous. Sample concentrations were 50
mg/ml at pH 1.90 (D-) and 1.58 (L-tryptophan), data collection
time was 4-8hrs, and laser power at the sample 0.625 W for each.
SERS spectra after addition of D- and L-tryptophan (C and D),
(0.0002 M) in the presence of silver hydroxyalamine reduced
colloid and MgSO4 at 0.050 M concentration, data collection time:
5 min, SEROA spectra of D- and L-tryptophan (G and H) with
addition of polycarbopol, data collection time of 35 min ........... 69
Figure 3.1
SERRS spectra of nanotag (tri-functional benzotriazole dye)
without silica coated silver colloids (A), with silica coated silver
colloids (B) and SERROA spectra of A (C) and B (D), data
collection time of 35 min and laser power at source 0.20 W ...... 85
Figure 3.2
SERRS spectra of D- and L-ribose that attached to silver silica
nanotag (A), SERROA of D- and L-ribose replicates 1 (B) and
batch 2 (C), data collection time of 35 min and laser power at
source 0.20 W ............................................................................. 86
Figure 3.3
SERRS spectra of D- and L-tryptophan that attached to silver silica
nanotag (A), SERROA spectra of D- and L-tryptophan (B), data
collection time of 35 min and laser power at source 0.20 W ...... 87
Figure S3.1
Structure of tri-functional benzotriazole dye .............................. 88
Figure 4.1
Raman spectra of adenosine (pH= 12.95), AMP (pH= 6.02), ADP
(pH= 5.18), ATP (pH= 4.20), A(2)MP (pH= 3.13), A(2,3)MP
(pH=5.54), A(3)MP (pH= 8.10) and A(3,5)MP (pH=6.67) in
solution. The concentration for each sample was 100 mg/ml and
laser power was 0.6 W at the sample .......................................... 103
Figure 4.2
ROA spectra of adenosine (pH= 12.95), AMP (pH= 6.02), ADP
(pH= 5.18), ATP (pH= 4.20), A(2)MP (pH= 3.13), A(2,3)MP
(pH=5.54), A(3)MP (pH= 8.10) and A(3,5)MP (pH=6.67) in
solution. The concentration for each sample was 100 mg/ml and
laser power at the sample was 0.6 W .......................................... 104
8
List of Figures
Figure S4.1
SERS spectra of adenosine ribonucleotides and adenosine in the
presence of silver citrate reduced colloid. Analyte concentrations
were 1x10-5 mg/ml, K2SO4 concentration was 0.020 M, data
collection time was 50 seconds, and laser power was 0.20 W at the
laser ............................................................................................. 110
Figure 5.1
The chemical structure with atom numbering scheme (left) and
calculate minimum energy conformation (right) of linear L-Ala-LAla ............................................................................................... 126
Figure 5.2
The chemical structure with atom numbering scheme (left) and
calculate minimum energy conformation (right) of cylic L-Ala-LAla ............................................................................................... 126
Figure 5.3
Experimental and computed Raman (top) and ROA (bottom)
spectra of linear (pH= 7.0) and cyclic L-Ala-L-Ala (pH= 7.0) in
aqueous solution. The concentration for each sample was 50 mg/ml
and laser power was 0.6 W at the sample. The marker bands that
are induced upon cyclization are highlighted by shading ........... 127
Figure 5.4
Experimental and computed Raman (top) and ROA (bottom)
spectra of linear (pH= 7.0) and cyclic L-Ala-L-Ala (pH= 7.0) in
D2O. The concentration for each sample was 50 mg/ml and laser
power was 0.6 W at the sample. The marker bands that are induced
upon cyclization are highlighted by shading .............................. 128
Figure 6.1
The chemical structure with atom numbering scheme (left) and
calculate minimum energy conformation (right) of linear L-Ser-LSer ............................................................................................... 154
Figure 6.2
The chemical structure with atom numbering scheme (left) and
calculate minimum energy conformation (right) of cyclic L-Ser-LSer ............................................................................................... 154
Figure 6.3
Experimental and computed Raman (top) and ROA (bottom)
spectra of linear (pH= 7.0) and cyclic L-Ser-L-Ser (pH= 7.0) in
aqueous solution. The concentration for each sample was 50 mg/ml
and laser power was 1.2 W at the laser. The marker bands that are
induced upon cyclization are highlighted by shading ................. 155
9
List of Figures
Figure 6.4
Experimental and computed Raman (top) and ROA (bottom)
spectra of linear (pH= 7.0) and cyclic L-Ser-L-Ser (pH= 7.0) in
D2O. The concentration for each sample was 50 mg/ml and the
laser power was 0.6 W at the sample. The marker bands that are
induced upon cyclization are highlighted by shading ................. 156
10
Abstract
Abstract
Raman optical activity (ROA) is a powerful probe of the structure and behaviour of
biomolecules in aqueous solution for a number of important problems in molecular
biology. Although ROA is a very sensitive technique for studying biological samples, it
is a very weak effect and the conditions of high concentration and long data collection
time required limit its application for a wide range of biological samples. These
limitations could possibly be overcome using the principle of surface enhanced Raman
scattering (SERS). The combination of ROA with SERS in the form of surface
enhanced ROA (SEROA) could be a solution for widening the application of ROA. In
the last few years, the generation of reliable SEROA spectra of biomolecules has been
problematic due to non-homogenous colloidal systems forming and low signal-to-noise
ratios which complicated detection of the true SEROA signal from the analyte. L- and
D-enantiomers give full or partially mirror image chiroptical spectra, this property of
enantiomers can be employed to prove the chiroptical activity of the SEROA technique.
In this thesis we employed a hydrophilic polycarbopol polymer as stabilising media
which has led to the first report of mirror image SEROA bands for enantiomeric
structures. This new technique of incorporating the hydrogel polymer as a means to
stabilise the colloidal system has proven to be reliable in obtaining high quality SEROA
spectra of D- and L-enantiomers of ribose and tryptophan.
In an extension of the hydrogel-stabilised SEROA work, we also demonstrate that
single nanoparticle plasmonic substrate such as silver silica nanotags can enhance the
weak ROA effect. These dye tagged silica coated silver nanoparticles have enabled a
chiral response to be transmitted from a chiral analyte to the plasmon resonance of an
achiral metallic nanostructure. The measurement of mirror image SERROA bands for
the two enantiomers of each of ribose and tryptophan was confirmed for this system.
The generation of SEROA for both systems was achieved and confirmed SEROA as a
new sensitive tool for analysis of biomolecular structure.
In a related project, Raman and ROA spectra were measured for adenosine and seven of
its derivative ribonucleotides. Both of these spectroscopic techniques are shown to be
sensitive to the site and degree of phosphorylation, with a considerable number of
marker bands being identified for these ribonucleotides. Moreover, the SERS studies of
these ribonucleotides were also performed. The obtained SERS spectra were shown
similar features that confirm these analytes interact with the surface in a similar manner,
hence limiting the structural sensitivity of this method towards phosphate position.
Short dipeptides such as diketopiperazine (DKP) have been investigated during the last
decades as both natural and synthetic DKPs have a wide variety of biological activities.
Raman and ROA spectra of linear and cyclic dialanine and diserine were measured to
charecterize their solution structures. Density functional theory (DFT) calculations were
carried out by a collaborator to assist in making vibrational band assignments.
Considerable differences were observed between the ROA bands for the cyclic and
linear forms of both dialanine and diserine that reflect large differences in the
vibrational modes of the polypeptide backbone upon cyclicization. In this study, the
ROA spectra of cyclic dialanine and diserine have been reported for the first time which
demonstrated that ROA spectroscopy when utilised in combination with computational
modelling clearly provides a potential tool for characterization of cyclic peptides.
11
Declaration
Declaration
No portion of the work referred to in the thesis has been submitted in support of an
application for another degree or qualification of this or any other university or other
institute of learning.
12
Copy right Statement
Copyright Statement
1. The author of this thesis (including any appendices and/or schedules to this
thesis) owns certain copyright or related rights in it (the “Copyright”) and he
has given The University of Manchester certain rights to use such Copyright,
including for administrative purposes.
2. Copies of this thesis, either in full or in extracts and whether in hard or
electronic copy, may be made only in accordance with the Copyright,
Designs and Patents Act 1988 (as amended) and regulations issued under it
or, where appropriate, in accordance with licensing agreements which the
University has from time to time. This page must form part of any such
copies made.
3. The ownership of certain Copyright, patents, designs, trade marks and other
intellectual property (the “Intellectual Property”) and any reproductions of
copyright
works
in
the
thesis,
for
example
graphs
and
tables
(“Reproductions”), which may be described in this thesis, may not be owned
by the author and may be owned by third parties. Such Intellectual Property
and Reproductions cannot and must not be made available for use without the
prior written permission of the owner(s) of the relevant Intellectual Property
and/or Reproductions.
4. Further information on the conditions under which disclosure, publication and
commercialisation of this thesis, the Copyright and any Intellectual Property
and/or Reproductions described in it may take place is available in the
University IP Policy (see http://www.campus.manchester.ac.uk/medialibrary/
policies/intellectualproperty.pdf),
in
any
relevant
Thesis
restriction
declarations deposited in the University Library, The University Library’s
regulations
(see
http://www.manchester.ac.uk/library/aboutus/regulations)
and in The University’s policy on presentation of These.
13
Abbreviations List
Abbreviations List
AMP
Adenosine 5’-monophosphate
ADP
Adenosine 5’-diphosphate
A(2)MP
Adenosine 2’-monophosphate
A(2,3)MP
Adenosine cyclic 2,3’- monophosphate
A(3)MP
Adenosine 3’-monophosphate
A(3,5)MP
Adenosine 3’,5’-cyclic monophosphate
ATP
Adenosine 5’-triphosphate
CID
Circularly intensity difference
CP
Circular polarization
Cyclic L-Ala-L-Ala
Cyclic L-Alanine-L-Alanine
Cyclic L-Ser-L-Ser
Cyclic L-Serine-L-Serine
DCP
Dual circularly polarised
DNA
Deoxyribonucleic acid
EM
Electromagnetic
IR
Infrared
Linear L-Ala-L-Ala
Linear form of L-Alanine-L-Alanine
Linear L-Ser-L-Ser
Linear form of L-Serine-L-Serine
NMR
Nuclear magnetic resonance
SCP
Scattered circular polarisation
SEROA
Surface-enhanced Raman optical activity
SERROA
Surface enhanced resonance Raman optical activity
SERRS
Surface enhanced resonance Raman scattering
SERS
Surface-enhanced Raman scattering
14
Glossary
Glossary
Anti-stokes scattering
Raman scattering event where the scattered photon has more
energy than incident radiation
Chirality
Property of molecules with a handedness to their chemical
Structure (i.e. molecules which cannot super impose on its
mirror image) and possessing optical activity
Colloids
Type of chemical mixture when one substance is dispersed
evenly throughout another
D-configuration
Dextrorotatory- configuration rotating the plane of vibration
of polarized light to the right
Elastic scattering
Specific form of scattering where the energy of incident
photon is equal to the energy of scattered photon.
Enantiomers
Molecules that are optical isomers, or mirror images, of one
another. Enantiomers can be distinguished by the direction in
which they rotate the plane of polarization of polarized light
Inelastic scattering
Raman scattering where the energy of scattered photon is not
equal to Incident photon
L-configuration
Laevorotatory-Rotating the plane of vibration of polarized
light to the left
Nanoparticles
Particles whose diameter is between 1 and 100 nm
Nucleoside
are glycosylamines consisting of a nucleobase (often referred
to simply base) bound to a ribose or deoxyribose sugar
Nucleic acids base, sugar and phosphate
Nucleotides
Purine
Heterocyclic aromatic organic compound which is consisting
of a pyrimidine ring fused to an imidazole ring
Pyrimidine
Nitrogenous organic base, containing two nitrogen atoms at
position 1 and 3 of the six-member ring
Raman scattering
The phenomenon change in wavelength of a photon as result
of inelastic scattering of light by molecule
Rayleigh scattering
Occurs as result of elastic collision between the photons and
molecules in the samples (no energy change in energy level of
the analyte)
15
Glossary
ROA
Raman optical activity one form of vibrational spectroscopies
which is reliant on the difference in intensity of Raman
scattered right and left circularly polarized light due to
molecular Chirality
Single plasmonic
nanoparticles substrate
Metallic nanoparticles which have localized surface plasmon
resonances that match the excitation wavelengths of lasers
used in
Stokes scattering
Raman scattering event where those photon scattered with
less energy than the incident radiation
Surface plasmons
The oscillations that occur as result of the interaction of light
beam with conduction electrons held in a lattice by the
presence of positive charge from the metal centre
16
Acknowledgements
Acknowledgements
I would like to thank to my supervisor Dr Ewan Blanch for his endless enthusiasm
for making this project feasible. It was a great experience working with him. To all
group mates, Clare, Myra, Ben, Christian, Grant, Nicola, Lorna, Heather, Kieaibi and
my friends Zahra and Soumya for making my last three years here a wonderful
experience, to Steve Prince, David Ellis and Elon Correa for giving me advice on
different subjects. I am truly fortunate to know you all.
To my parents, Jafar, Suosan and my grandmother for their endless support,
encouragement and love that made me believe in myself and pursue my dreams.
Without you all this would not be achievable.
My final and greatest appreciation is for my sister Soheyla, without her nothing was
possible.
17
Chapter 1
Chapter 1
Introduction
18
Chapter 1
1.0 Introduction
1.1 Conformational analysis of biological molecules
The characterisation of the three-dimensional (3D) structures of biological molecules
and their relationship to their functions has made a tremendous impact on all
subsequent biochemical investigations [1,2]. X-ray protein crystallography is
currently the primary methodology used for determining the 3D structure of
biological molecules at near-atomic or atomic resolution, the other notable atomic
resolution technique being nuclear magnetic resonance (NMR) [1-4]. However,
typically around 20–40% of all protein molecules, including many important
proteins, are difficult or impossible to crystallize, and hence their structures have not
been accessible by crystallography [1,2]. NMR has limitations on the size and type of
molecules that can be structurally characterized [3,4]. Overcoming these limitations
has therefore necessitated the development of novel approaches for structural
biology.
One of these novel approaches is the employment of vibrational spectroscopy which
is concerned with the interaction of electromagnetic radiation with matter.
Vibrational spectroscopic techniques, such as infrared (IR) [5-7] and Raman [8-10]
spectroscopies have been widely applied in different fields of science such as
biochemistry and biomolecular structure analysis. Infrared spectroscopy measures
the light intensity of light absorbed in the infrared region of the electromagnetic
spectrum. The electric dipole of the molecule must change during a molecular
vibration in order for the molecule to absorb infrared radiation [5]. In the mid- and
far-infrared spectral regions, the absorption occurs where the frequencies of light and
molecular vibrations are equal which causes promotion of the molecule to a
vibrationally excited state [5]. Raman spectroscopy is also based on vibrational
19
Chapter 1
transitions that give rise to narrow spectral features characteristic of the investigated
sample [11] and measures the intensity of light that is inelastically scattered from the
molecule.
Raman and infrared spectroscopies possess several advantages in contrast to other
analytical techniques since they are non-invasive, the requirement for sample
mass/volume is minimal and more importantly, there is no requirement for chemical
labelling/probes. Protein aggregation, stability, conformational changes induced by
different factors, the accurate prediction of structure and folding can all be assessed
by vibrational spectroscopy [5-11].
1.2 Infrared and Raman spectroscopies
Light is a form of electromagnetic (EM) radiation composed of electric and magnetic
waves that are oriented perpendicular to each other when they oscillate in single
planes. The interaction of electric waves with matter can either lead to the
absorbance or scattering of the incident light. If the energy of the photon matches
that of the energy difference between the ground and excited states of a molecule, the
photon may be absorbed and the molecule is promoted to a higher energy excited
state. The energy change resulting from this phenomenon is measured by absorbance
spectroscopy through the detection of the energy lost from the initial radiation.
Infrared spectroscopy is a technique derived from the vibrations of atoms of a
molecule. An infrared spectrum is commonly obtained by passing infrared radiation
through a sample and determining what fraction of the incident radiation is absorbed
at a particular energy. The energy at which any peak in an absorbance spectrum
appears corresponds to the frequency of a vibration of the sample molecule. This
20
Chapter 1
interaction of infrared radiation with matter can be explained by changes in the
molecular
dipole
associated
with
vibrations
and
rotations
[5-7,11].
Light scattering occurs when there is no necessity for a photon to possess an energy
matching the energy difference between the ground and excited vibrational levels.
This phenomenon of change in wavelength as a result of inelastic scattering of light
by matter was first observed by the Indian scientist C.V. Raman in 1928 [12]. When
the incident photon collides with a molecule, the electron distribution is perturbed
which results in the Raman effect. The majority of photons are scattered elastically
which is known as Rayleigh scattering, with no change in wavelength of the
scattered photon [11]. However, a transfer of energy can occur either from the
incident photon to the molecule or from the molecule to the incident photon, if
nuclear motion is induced. This results in inelastic scattering of the photon for which
the energy of the incident photon is different to that of the scattered photon, termed
Raman scattering [13].
In the energy level diagram shown in Figure 1.1, at room temperature most
molecules are present in the lowest energy vibrational level [11], and with a
monochromatic light beam of photon energy hv and wavelength λ the Raman
scattering process from the ground vibrational state m leads to the absorption of
energy to a higher excited vibrational state n. This difference in energy between the
incident and scattered photons is represented by Stokes and anti-Stokes lines. The
scattered photon for Stokes lines has a lower energy than the incident photon, and for
anti-Stokes lines the incident photon has less energy than the scattered photon [11].
Variation in energy from the excitation state is correlated to the vibrational energy
spacing in the ground state of the molecule, hence, the vibrational energy of the
molecule can be a probe of molecular chemistry of the sample via quantification of
21
Chapter 1
the frequencies of Stokes and anti-Stokes lines [14]. Vibrational motion is sensitive
to chemical modification and therefore the molecular chemistry of samples can be
studied. In Raman spectroscopy typically the Stokes scattering is used for analysis of
molecular structure since the anti-Stokes scattering is weak.
Virtual State
Stokes
Rayleigh
Anti Stokes
Vibrational
Levels
n
m
Ground Electronic State
Figure 1.1: An energy level diagram showing the transitions involved in Rayleigh
and Raman scattering, adapted from [11].
22
Chapter 1
1.2.1 Principles of Raman spectroscopy:
As an oscillating beam of light interacts with a molecule, the electron cloud of the
molecule is perturbed periodically with the same frequency as the electric field of the
incident wave. The perturbation of the electron cloud results in a periodic separation
of charge within the molecules that is termed an induced dipole moment. The
oscillating induced dipole moment is a source of EM radiation, thereby resulting in
scattered light [15].
As discussed above, the electric field associated with the laser radiation induces a
dipole moment in the molecule which is proportional to the electrical field strength E
and to the molecular polarizability α (the ability of the electron charge distribution to
be distorted by an electric field) that depends on the molecular structure and the
nature of the molecular bonds.
The strength of the induced dipole moment μ is given by [15],
(1)
Because of the vector nature of the dipole moment and electric field, α is not a
simple constant, but can be written as a tensor by taking account of the contributions
with respect to the three Cartesian axes x, y and z. All three components of E
contribute to each of the three components of μ shown in tensor form, as they refer to
the Cartesian axis directions [11].
(2)
23
Chapter 1
In the case of aqueous solution samples, there is no ordering of the axes of the
molecule to the polarization direction of the light but it is possible to measure these
from polarization measurements. In practical situations, the ratio of depolarization is
determined where the intensity of a given band is measured with respect to the plane
of polarization of the incident light being parallel or perpendicular to the scattered
light. The average polarizability can also be described in terms of isotropic (with the
analyzer parallel to the plane of the incident radiation) and anisotropic (with the
analyzer perpendicular to the plane) components of the tensors as represented in
equations (3) and (4), where
and
represent isotropic and anisotropic terms,
respectively [11,14],
(3)
(4)
1.2.2 Resonance Raman spectroscopy
Historically, coloured compounds were avoided by most Raman spectroscopists.
This was mainly due to decomposition of the sample by the powerful lasers used that
prevents Raman analysis as a result of strong fluorescence [16,17]. However, if the
frequency of the laser beam is close to the frequency of an electronic transition,
scattering enhancement of up to 106 can be observed [11,16-17]. This results in a
more sensitive technique in contrast to conventional Raman spectroscopy since a
chromophore provides more efficient scattering that is selective for the molecule
involving the chromophore [16,17]. The resonance scattering can provide both
electronic and vibrational information concerning the molecule of interest.
24
Chapter 1
Resonance Raman scattering can occur when an incident laser beam has an
excitation frequency close to that of an electric transition. A tuneable laser beam can
be used for excitation, and the frequency would correspond to the energy difference
between the ground vibrational state and the first or second vibronic state of the
excited state.
The resonance condition is met when the energy difference between the lowest
vibrational state of the ground electronic state and the resonant vibronic state is of
the same energy as the excitation resulting from the incident light. The obtained
enhancement of Raman scattering is mainly due to an increase in polarizability
[16,17]. Elucidation of structural information from deep within complex biological
samples was enabled through development of this technique [18-20].
1.2.3 Raman spectroscopy of biomolecules
The molecular information provided by Raman spectroscopy is the same as that from
infrared spectroscopy. However, the Raman effect has advantages over IR absorption
for aqueous environments since less interference occurs from the solvent [5-10]. This
advantage is beneficial for studying biological samples in solution since water in
most cases is a pre-requisite for functioning in the surrounding physiological
environment. Raman spectroscopy has a number of advantages and disadvantages
compared to other analytical techniques for studying biological samples, which are
summarised in Table 1.1.
Raman spectroscopy still remains a practical method for probing the interplay
between structure, dynamics and function of biomolecules [21-25]. Understanding
the precise structure of biomolecules in terms of their vibrational spectra can have a
large impact on discovery of exact physiological function in living systems. The
25
Chapter 1
vibrational modes of biomolecules that can be studied by vibrational spectroscopies
such as Raman are characteristic of their molecular structure. However, due to the
large number of vibrational modes in biomolecules, it is a complex task to elucidate
detailed information based on the measurements of vibrational spectra. Even so,
important information on secondary structure elements may frequently be derived
[21-25].
Table 1.1: Advantages and disadvantages of Raman spectroscopy [5-11].u Column2
Advantages
Disadvantages
Distinct characteristic vibrations that can be
used as finger prints for qualitative/
quantitative identification
Weak effect
Lack of interference with other vibrational
bands which results in narrower absorption
bands from the laser beam
Interference from fluorescence
No or minimal sample preparation required
Decomposition of coloured samples as a
result of heating
Minimal volume
Minimum absorption by water molecules
Non-invasive
Can be used for a wide range of conditions
e.g. aqueous, gas, solid, tissue
Table 1.2, displays the most notable spectral regions from protein vibrations, which
are called amide I, II and III [26-28]. The spectral information obtained in these
regions is a sensitive indicator of the presence of secondary structure within
biomolecules such as proteins and peptides where they have been used to estimate
the amount of α-helix and β-sheet content [26-28].
26
Chapter 1
Table 1.2: Important spectral regions for protein vibrations [26-28]
Band
Wavenumber (cm-1)
Vibrational Assignment
Amide I
Amide II
Amide III
1630-1700
1480-1575
1230-1330
C=O Stretching
N-H bend/ C-N stretching
N-H bend/ C-N stretching/
Cα-H deformation
Bands in the vicinity of 1655-1659, 1300 and 1340 cm-1 in the amide I and III
regions indicate α-helical conformations whereas bands in the vicinity of 1670, 1700
and 1229-1240 cm-1 usually indicate β-sheet conformations [26-28].
27
Chapter 1
1.3 Raman Optical Activity (ROA)
The vibrational optical activity of chiral molecules, exemplified by Raman optical
activity (ROA), was predicted by Atkins and Barron in 1969 [29]. They noticed a
new optical process involving interference between light waves scattered through the
polarizability and optical activity tensors of a chiral molecule that was first
experimentally measured by Barron, Bogaard and Buckingham, in 1973, who
observed a small difference in the intensity of Raman scattering in right- and leftcircularly polarized light from α-phenylethylamine and α-phenylethanol [30]. This
observation was independently verified by Hug in 1975 (Figure 1.2).
x
Raman
IR+ IL
R
L
ω
z
y
  
ROA
IR- IL
Figure 1.2: Schematic diagram of the basic SCP ROA experiment which measures
as a small circular component in the scattered light in right (R) and left (L) using
unpolarized incident light (adapted from [11]).
The ROA measurement can be represented in terms of the circular intensity
difference (CID) that is defined by;
(5)
28
Chapter 1
where I R and I L are scattered Raman intensities in right and left circularly
polarized light, respectively.
ROA measures the optical activity related to Raman scattering and the chirality
associated with molecular vibrational transitions [32], where a chiral molecule is one
that is not super-imposable on its mirror image. The two mirror image forms of a
chiral molecule are referred to as enantiomers [33]. Chiral molecules scatter left- and
right-circularly polarized light to different degrees which leads to the resultant ROA
spectrum. Unlike conventional Raman spectroscopy, in which only the electric
dipole interacts with the incident light, in ROA spectroscopy, contributions from
magnetic dipole and electric quadrupole optical activity tensors must also be
considered.
The oscillating electric dipole, magnetic dipole and electric quadrupole moments are
characteristic of the scattered radiation field induced in a molecule by the incident
light. The electric dipole, magnetic dipole m and electric quadrupole moments 
are described by,
(6)
(7)
(8)
where particle i at distance ri has charge ei , mass mi , linear momentum pi and the
Kronecker delta,   , is a function of two variables which is equal to 1 if they are
equal and 0 otherwise [31,32].
29
Chapter 1
The molecular multipole moments and quantum mechanical expressions for the
dynamic molecular property tensors can be defined by the fields and field gradients
that are assessed at the origin of the molecule. The field and field gradients are
derived from the time-dependent perturbation theory and are defined as,
Electric dipole-electric dipole tensor:
(9)
Electric dipole-magnetic dipole optical activity tensor:
(10)
Electric dipole-electric quadrupole tensor:
(11)
where n and j represent, respectively, the initial and virtual intermediate states of the
molecule,  jn   j  n is their angular frequency separation and
is Plank’s
constant [31]. Circularly polarized light scattering and diffraction are caused by the
electric dipole-electric dipole tensor  , optical rotation in aqueous solution is
generated by the electric dipole-magnetic dipole tensor G
 ; and A is the electric
dipole-electric quadrupole tensor which leads to additional contributions to optical
rotation in oriented samples [34,35].
By averaging the different polarizability-polarizability and polarizability-optical
activity tensors components for all possible orientations of a molecule, we can
30
Chapter 1
consider their tensorial components that are invariant to axis rotations as shown in
equations 12-15.
The isotopic invariants are defined as,
and
(12)
and the anisotropic invariants as [34, 35],
(13)
(14)
(15)
The scattering angle can be varied, for example we can have forward (0) or
backward (180) scattering. Right angle scattering can also be measured using a
linear polarization analyzer in the scattered beam to select either the perpendicular
(x) or parallel (z) transmission axis to the scattering plane (yz). CID expressions for
 and A ,
different scattering geometries can be written in terms of  , G
(16)
(17)
31
Chapter 1
(18)
(19)
When a molecule consists of idealized axially symmetric bonds, for which
 (G)2   ( A)2 and  G  0 , a simple bond polarizability theory explains that ROA
is generated entirely by anisotropic scattering in which case the CID expressions
reduce to :
(20)
and
(21)
These equations illustrate that ROA scattering intensities are maximized in the
backward direction and are zero in forward scattering. This is unlike the case of
conventional Raman spectroscopy, where the forward and backward scattering
intensities are the same [34,35].
1.3.1 ROA spectroscopy of biomolecules
By understanding the connection that exists between protein structure and function,
the behaviour of proteins can be studied. A range of techniques have been applied to
the clarification of 3D structures of proteins, ranging from prediction based on the
sequence and physico-chemical properties of the constituent amino acids to high
resolution methods for the detection of atoms and determination of their molecular
32
Chapter 1
coordinates. As a result, investigations of protein structure (at primary, secondary,
tertiary and quaternary levels) are important as probes of protein function in living
organisms [36]. For a number of important problems in molecular biology such as
protein folding, protein-protein interactions, and protein-nucleic acid interactions,
quantitative measurement of the secondary structure provides significant insight into
structural features critical to biological function [36,37].
Almost all biological compounds are chiral, so it is logical to investigate them not
only by Raman spectroscopy but also by ROA, where additional features can be
determined. Although ROA is a very weak effect with  -values typically being ~
103  105 , it provides more structural information than Raman spectroscopy as
ROA spectral details are more sensitive to stereochemistry [31].
ROA has been measured for a wide range of biological molecules including proteins,
carbohydrates [37], nucleic acids [38, 39] and viruses [40]. The ROA spectra of
proteins are dominated by bands originating in the peptide backbone which directly
reflect their solution conformations [37]. The bands from side chains are usually not
as significant in the ROA spectra of polypeptides and proteins as they are in the
conventional Raman spectra [41] since the largest ROA signals are often associated
with vibrational coordinates from the most rigid and chiral parts of the biomolecules
[31].
33
Chapter 1
1.4 Surface-Enhanced Raman Scattering
Surface-enhanced Raman scattering (SERS) is a powerful tool for determining
chemical information about molecule substrates. The enhanced Raman signals in
SERS are due to enhanced electromagnetic fields that result from adsorption of
molecules on nanotextured metallic surfaces [42]. The SERS observation was first
reported by Fleischman and co-workers, Hendra and McQuillan in 1974 [43]. In
1977 two separate papers by Jeanmaire and Van Duyne, and by Albrecht and
Creighton confirmed the observation of a surface Raman spectrum of pyridine
adsorbed on electrochemically roughened silver electrodes as a result of successive
oxidation-reduction cycles [44,45].
The SERS method can increase the intensity of the Raman signal with enhancement
factors of 102  106 in scattering efficiency over conventional Raman scattering [45].
It has been reported that the enhancement of Raman signals can be up to 1011 or
greater for some experiments which proposes the possibility of single molecule
detection levels [46,47]. Silver and gold are the typical substrates used in SERS
technique and in various metal forms, for instance colloids, roughened electrodes,
deposited layers and nanoshells [48]. While the type and preparation methods of
metal substrates have an effect on the outcome of SERS signals, other factors such as
temperature, pressure, nature of the analyte, aggregating agents and laser power can
also have significant influence on SERS signals [49-51].
Several theories have been suggested to explain the mechanisms involved in SERS
enhancement. The enhancement occurs due to increases in both the molecular
polarizability of adsorbed species and the local electric field in the vicinity of the
metallic surface [52]. However, the exact nature of SERS is still unknown, though it
34
Chapter 1
is now accepted that the electromagnetic and charge transfer (chemical)
enhancements are the two most important mechanisms [50].
1.4.1 The Electromagnetic Mechanism
Various electromagnetic (EM) theories have been developed over the past decades.
Complete electrodynamic calculations have been performed for simpler systems and
the effects of dielectric responses have been discussed by Moskovits [52]. EM
enhancement only depends on the characteristics and morphology of the metal
surface [53]; therefore the same enhancement factor of vibrational modes should be
obtained for the same surface morphology [54]. In order to explain the theory, the
morphologies of roughened metal surface need to be understood. Electrons circulate
on the metal surface which is held in a lattice by the presence of positive charges.
This electron density on the surface expands in a significant distance from the
surface which has the freedom of movement in the lateral direction [55,56]. As the
incident electromagnetic radiation interacts with the electron density that surrounds
the atomic lattice sites of the metal, vibrations in the molecule are initiated, resulting
in a collective oscillation which is known as a plasmon [50,57-58]. Surface plasmons
from small uniform particles or from single periodic roughness features of a surface
have a resonance frequency which results in scattering of electromagnetic radiation
[57]. The dielectric constant of the metal has a direct effect on the frequency of the
surface plasmon oscillation. The resonance frequency should match with the visible
frequencies of Raman scattering in order to generate the SERS enhancement [57]. To
facilitate enhancement it is necessary for the oscillation of the plasmon to be
perpendicular to the surface plane which is usually achieved by roughening of the
surface [58]. This results in an increase of the coupling concentration of the
electromagnetic field in certain regions on metallic surfaces [58]. An analyte
35
Chapter 1
molecule on or near the metal surface interacting with a surface plasmon experiences
a large electromagnetic field, resulting in enhancement of the vibrational modes in
the Raman spectrum [58].
EM theory cannot entirely explain the mechanisms involved in SERS; in particular it
predicts the uniform enhancement of all Raman active bands. However, this is not
the case in practice since some intense bands that can be observed in Raman spectra
weaken or disappear in their corresponding SERS spectra. Therefore, mechanisms
other than EM must be implicated in the SERS phenomenon in order to fully explain
SERS enhancement.
1.4.2 The Chemical Mechanism (CE)
Other studies suggest that there is a second enhancement mechanism for SERS
which operates independently from the EM mechanism [48,59]. Different molecules
with identical polarizability, adsorbed onto the same metallic substrate under the
same experimental conditions, demonstrate different enhancement factors [48]. This
would suggest that an EM mechanism is not the only mechanism involve in SERS
enhancement. Further evidence in support of the chemical mechanism comes from
potential-dependent electrochemical experiments. When an electrode potential is
scanned at a fixed laser frequency, or the laser frequency is scanned at fixed
potential, broad resonance is observed [48].
Enhancement resulting from the chemical mechanism occurs due to the formation of
a chemical bond between the atomic scale of metal roughness and the adsorbed
analyte [10]. This chemical contribution generates surface species which consist of
the analyte and surface metal atoms [10]. This in turn enables feasibility of charge
transfer from the metal surface into the analyte which causes an increase in the
36
Chapter 1
molecular polarizability of the analyte due to interaction with the metal’s electrons
[59]. Basically, these observations can be explained by resonant intermediates in
Raman scattering in which either the electronic states of the adsorbate are shifted and
broadened by their interaction with the surface, or the formation of a new electronic
state arising from chemisorption [50]. The highest occupied molecular orbital and
lowest unoccupied molecular orbital of the adsorbate are symmetrically disposed in
energy with respect to the Fermi level of the metal [50]. In this case, charge transfer
excitations can occur either from the metal substrate to the molecule or from the
molecule to the metal substrate at the about half the energy of the intrinsic
intramolecular excitation of the adsorbate [10,50]. Most charge transfer excitations in
SERS take place at visible wavelengths since the molecule’s lowest-lying excitation
energy is in the near ultraviolet region [59].
As discussed above, it is very difficult to separate the contribution effects resulting
from EM and chemical mechanisms on systems which support SERS enhancement.
Although the majority of evidence suggests that both mechanisms play a key role in
SERS enhancement, EM enhancement has the greater effect on enhancement as the
charge transfer enhancement of Raman signals drops off by 1 r 3 with distance ‘r’
from the surface [10].
1.4.3 Biological applications of SERS
Metal substrates can be applied to obtain more precise information for structural
determination of nucleic acids and peptides as they can give more enhanced signals
in short illumination periods (less than 1s), lower laser power and sample
concentrations. SERS spectra of different biomolecules such as amino acids,
nucleotides, peptides, enzymes, DNA and RNA have been reported using different
37
Chapter 1
metal substrates e.g. Ag and Au [60-77]. One of the main advantages of SERS for
analysis of biomolecules is the reduction of the luminescent background that often
obscures Raman scattering from biological molecules [78]. The compatibility of the
metal substrate with biomolecules and the morphology of the surface play an
important role in SERS activation. This compatibility facilitates a better coupling
between adsorbed sample molecules and the metallic surface for enhancement of
Raman scattering. The potential for application of SERS for analysis of biological
components is illustrated by its use in the diagnosis of tissue lesions [79,80], analysis
of blood components and study of tissues [79,80]. A highly sensitive and selective
SERS detection has been reported for DNA using plasmonic nanoparticle substrates,
highlighting the potential of this approach as a rapid genetic analysis tool for
understanding biological process, for unlocking the underlying molecular cause of
diseases and for development of biosensors [81].
38
Chapter 1
1.5 Surface enhanced Raman optical activity (SEROA)
Existing problems with ROA spectroscopy for studying biological molecules are
principally that the conditions required include high sample concentrations (10-100
mg/ml) and long acquisition periods in comparison to those required for Raman and
SERS [82]. Although ROA spectroscopy gives more incisive information about
stereochemistry of biomolecules, it suffers from being a weak effect hence
preventing the application of ROA to a wider range of target molecules at present.
Alternatively, SERS generates signals that can be ~ 106 larger than the conventional
Raman signals [45]. The combination of ROA with SERS in the form of surface
enhanced Raman optical activity (SEROA) is a promising solution for widening the
application of ROA to other biomolecules that are not easily accessible to other
structural methods, such as unfolded proteins and viruses.
In recent decades, various theoretical aspects of surface enhancement of ROA signals
have been investigated. Efrima proposed that measurements of ROA are possible for
a molecule adsorbed on the metal surface which contains information on the local
electric field, their gradients and, in general, local dielectric properties of the metalsolution interface [83,84]. The model relies upon large electric field gradients close
to the metal surface where a molecule is subjected to a larger electric field than in the
bulk solution. Enhancement of the ROA signal can be achieved by this model as the
electric dipole-electric quadrupole contribution is predicted to be large. According to
Efrima’s calculations, SEROA spectra can be influenced by several properties of
electromagnetic radiation once it interacts with metal surface. These include the
magnitude, direction, spatial dependence and polarization of the electromagnetic
radiation. In summary, Efrima proposed that SEROA can be obtained if certain
conditions are met [83,84]. These are the existence of an electromagnetic field
39
Chapter 1
gradient near the surface of the metal surface, a phase difference between the electric
field and its gradient and finally induction of resonance where the interaction
between the molecule and the metal surface occurs [83,84].
Hecht and Barron also considered enhancement of the ROA signal using a metal
substrate, through the approximation of a pure electric dipole surface ROA model
[85,86]. They also investigated the possibility of a SEROA spectrum being generated
by an achiral molecule. They postulated that when an achiral molecule adsorbs onto
the metal surface and is randomly oriented, no ROA signal can be achieved. This is
mainly due to cancellation of the signal as a result of different enantiomeric
projections. However, if they align in a manner to form an ordered surface, obtaining
a SEROA signal from an achiral molecule is feasible.
A decade later Janesko and Scuseria considered the effect of averaging over all
orientations of the metal surface, using three different models; a dipolar sphere,
quadrupole sphere and a dipolar nanorod [87]. In contrast with Efrima’s models, they
predicted significantly smaller CID values.
Etchegoin et al. [88] have proposed the generation of SEROA signals by modelling
of single SERS experiments. In their theoretical work, they have predicted that the
high enhancements associated with ‘hot spots’ for SERS single molecule detection
affect the behaviour of circularly polarized light in the vicinity of the surface
plasmons [88]. Given that ROA is a weaker effect than conventional Raman, they
predicted that SEROA signals may be small and difficult to distinguish from
background noise. Also, the detected SEROA signal comes from the electric field of
a number of different molecules that may have different polarization directions
which may result in cancelling of the SEROA signals and prevent reliable
measurement of a SEROA spectrum.
40
Chapter 1
Various theoretical studies modelling ROA responses in the vicinity of metal
substrates have facilitated a better understanding of the fundamental phenomenon of
SEROA [89]. The first simulation of SEROA using the time-dependent density
function theory has recently been proposed by Janesko and Scuseria where
interaction of adenine with a Ag cluster was investigated [90]. They predicted an
enhancement factor of 104 for both SERS and SEROA and concluded that the
observed enhancement due to charge transfer is larger than that of SERS. The
chemical effect of analyte-colloid binding on the metal surface was also calculated
[90]. The results suggest that observed SEROA bands, in terms of signs and
intensities, are very sensitive to analyte-metal orientation. As a result, they proposed
that future SEROA experiments may require utilising ordered monolayers of chiral
analytes to minimise this orientation effect [90]. Yang et al. [91] have studied the
interaction of electromagnetic fields with light and the phenomena of enhancement
of magnetic and electric field gradients. They have highlighted the potential
applicability of SEROA as a chiroptical analytical tool.
Halas et al. [92] also developed a SEROA model for molecules moving near
spherical metal nanoshells where the excitation profiles for a simple chiroptical
model was analyzed in detail and suggested a preferred excitation wavenumber. Very
recently, the matrix polarization theory was employed to model the SEROA spectra
of ribose and cysteine molecules and enabled comparison with experimental results
[93]. Findings showed a strong distance dependence of enhancement between
molecules and the metal surface along with dependence of the ROA ratio and Raman
intensities (CID) on distance and rotational averaging. Findings from this study
confirmed that maximum enhancement can be obtained by colloidal aggregates
rather than asymmetry of individual particles and validated the experimental
41
Chapter 1
observation of L- and D-ribose which was reported as part of this thesis [94]. The
importance of controlling the colloid-sugar distance was emphasised which was done
in the experimental set up for SEROA measurement of the ribose molecule.
Currently, few experimental studies have been reported to validate the technique
with a high degree of certainty. Kneipp et al. [95] have claimed to observe a SEROA
spectrum of adenine molecule adsorbed onto silver colloidal nanoparticles. They
correlated two SEROA bands to the most enhanced peak in the SERS spectrum of
adenine. Since adenine is an achiral molecule, they suggested that symmetry of the
molecule can be lost once it is adsorbed onto the metal surface which then becomes
chiral. They have also postulated that the adenine molecule aligns in the same
orientation on the metal surface that then result in prevention of signal cancelling
from various pro-chiral attachments. Abdali and Blanch [82] in their review article
dismissed these results and suggested that they are artefacts as Kneipp et al.
neglected the problems associated with changes in the polarization state modelled by
Etchegoin et al. [84]. These problems arise from reflection of electromagnetic light
from the metal surface that modify circularly polarized light and create elliptically
polarized light, which may result in the generation of birefringent artefacts [84].
Also, the generation of an ROA signal under these conditions would require a
significant proportion of circularly polarized light, since a significant proportion of
linearly polarized light would result in SEROA artefacts.
Abdali et al. reported SEROA spectra of two resonant molecules; cyctochrome c and
myoglobin as well as a nonresonant molecule; Met-enkephalin [96-98]. However, it
is difficult to verify these results since there may be artefacts arising from the parent
SERS bands and also there were no enantiomers of these compounds available to
42
Chapter 1
verify SEROA mirror image band responses which are very important for validating
of chiroptical techniques such as SEROA.
More recently, Osinska et el. have reported the measurement of mirror image
SEROA spectra of L- and D-cysteine using an electrochemically roughened solid
silver based system [99]. Although this was an interesting observation, this study did
not consider a number of details concerning the experimental procedure utilised.
These results cannot be deemed reliable since the authors reported that SERS spectra
of L- and D-cysteine for the same experiment could not be observed. Instead they
reported SERS spectra for a different colloid-based experiment with no
corresponding SEROA. Since any SEROA measurement must logically be weaker
than the corresponding SERS measurement, and therefore more difficult, this calls
into doubt the reliability of these spectra. It appears that they did not measure
SEROA but rather the solution-phase ROA, as sample concentrations were very high
in their study, complicated by the presence of large reflection-based artefacts from
the metal surface.
In 2008 [100] and 2009 [101], two PhD theses presented attempts to prove SEROA
as a technique, but which were unsuccessful in both cases due to the complexity of
the process. However, both illustrated the importance of controlling the experimental
SERS conditions such as colloidal type, analytes, aggregating agents, pH and
concentration since they have a direct effect on obtaining not only reliable SERS but
also any associated SEROA spectra. The time-dependent nature of the enhancement
process was shown to have a significant effect on the obtained results laying the
platform for this thesis.
SEROA can still be considered an unproven technique. The main objective of this
PhD thesis is to prove and develop reliable SEROA. Obtaining a more consistent
43
Chapter 1
colloidal system may assist to validate SEROA as a feasible technique and this could
be achieved by controlling the colloidal aggregation over long time periods using a
hydrogel polymer as a stabilising agent. The measurement of mirror image bands for
two enantiomers of each of ribose and tryptophan were undertaken in order to prove
the validity of SEROA measurements, and this work is presented in Chapter 2. Silver
silica nanotags, which have been proven to provide strong SERS enhancement, as
well as stable metal nanoparticles were both used to assess the potential application
of SERROA as a nanoprobe for biomolecule analysis. The chiroptical properties of
these nanotags were confirmed by the measurement of mirror image surface
enhanced resonance Raman optical activity (SERROA) spectra of the two
enantiomers of each of ribose and tryptophan which is demonstrated in Chapter 3. As
part of the PhD research undertaken a number of other studies were performed to
investigate outstanding problems in biomolecular structure using analytical
spectroscopies. Chapter 4 presents work in which adenosine and seven of its
derivative ribonucleotides were studied by Raman, ROA and SERS in order to
identify spectral markers of site-specific phosphorylation in nucleic acids. Raman
and ROA spectroscopies in combination with computational modelling were used to
study the structural changes in short linear dipeptides, specifically diserine and
dialanine, due to cyclization; are presented in Chapters 5 and 6. The conclusion
chapter highlights the importance of optimization of the correct experimental
protocols for obtaining reliable SERS and SEROA spectra. The conclusion also
discusses how that combination of four different spectroscopic techniques, Raman,
ROA, SERS and SEROA is more advantageous for studying of biological samples
since they can provide more structural information on the nature of biomolecules
along with their chirality.
44
Chapter 1
1.6 References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
T.Takahashi , S. Nakatani, Surface Science. 1995, 326, 347-360.
F. H. C. Crick, Methods in Enzymology. 1957, 4, 127-146.
A. Krushelnitsky, D. Reichert, Progress in Nuclear Magnetic Resonance
Spectroscopy. 2005, 47, 1-25.
M. W. F. Fischer, A. Majumdar, E. R. P. Zuiderweg, Progress in Nuclear
Magnetic Resonance Spectroscopy. 1998, 33, 207-272.
M. Subirade, D. Marion, M. Pézolet, Thin Solid Films. 1996, 284-285, 326329.
I. Martin, E. Goormaghtigh, J. M. Ruysschaert, Biochimica et Biophysica
Acta (BBA) Biomembranes. 2003, 1614, 97-103.
I. T. Arkin, Current Opinion in Chemical Biology. 2006, 10, 394-401.
V. A. Shashilov, M. Xu, V. V. Ermolenkov, I. K. Lednev, Journal of
Quantitative Spectroscopy and Radiative Transfe. 2006, 102, 46-61.
S. W. Ellepola, S. M. Choi, D. L. Phillips, C. Y. Ma, Journal of Cereal
Science. 2006, 43, 85-93.
Jr. V. Kopecký , V. Baumruk, Vibrational Spectroscopy. 2006, 42, 184-187.
E. Smith, G. Dent, Moder Raman Spectroscopy: A Practical Approach. 2005:
Wiley.
C. V. Raman, K. S. Krishnan, Nature. 1928, 121, 501-502.
P. R. Carey, Biochemical Application of Raman and Resonance Raman
Spectroscopies. 1982: Academic Press.
R. L. McCreery, Raman Spectroscopy for Chemical Analysis. 2000: Wiley.
H. T. Helmut, A. G. Urena, J. Robert, Laser Chemistry: Spectroscopy,
Dynamics and Aplication. 2007: Wiley.
B. Robert, Photosynthesis Research. 2009, 101, 147-155.
A. Albrecht, Journal of Chemical Physics. 1961, 34, 1476-1484.
A. V. Ruban, R. Berera, C. Ilioaia, Ivo. H. M. van Stokkum, J. T. M. Kennis,
A. A. Pascal , H. V. Amerongen, B. Robert, P. Horton, R. V. Grondelle,
Nature. 2007, 450, 575–578.
A. M. Dokter, M. C. Van Hemert, C. M. In’t Velt, K. van der Hoef, J.
Lugtenburg, H. A. Frank, E. J. J. Groenen, Journal of Physical Chemistry A.
2002, 106, 9463-9469.
M. Cecarelli, M. Lutz, M. Marchi, Journal of the American Chemical Society.
2009, 122, 3532-3533.
L. J. Lis, J. W. Kauffman, D. F. Shriver, Biochimica et Biophysica Acta
Biomembranes. 1975, 406, 453-464.
L. J. Lis, J. W. Kauffman, D. F. Shriver, Biochimica et Biophysica Acta
Biomembranes. 1976, 436, 513-522.
A. Schneemann, Annual Review of Microbiology. 2006, 60, 51-67.
G. Cerchiaro, A. C. Sant'Ana, M. L. A. Temperini, A. M. da Costa Ferreira,
Carbohydrate Research. 2005, 340, 2352-2359.
G. Zhang, S. Shuang, C. Dong, J. Pan, Spectrochimica Acta Part A:
Molecular and Biomolecular Spectroscopy. 2003, 59, 2935-2941.
T. M. Theophanides, Infrared and Raman Spectroscopy of Biological
Macromolecules. 1979, D. Rediel Pub: Boston.
B. G. Frushour, J. K. Koeing , Advances in IR and Raman Spectroscopy.
1975, Heyden & son: Berkshire.
45
Chapter 1
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
R. W. Williams, Enzymology. 1986, Academic Press. 311.
P. W. Atkins, L. D. Barron, Molecular Physics. 1969, 16, 453-466.
L. D. Barron, M. P. Bogaard, A. D. Buckingham, Journal of the American
Chemical Society. 1973, 95, 603-605.
L. D. Barron, L. Hecht. I. H. McColl, E. W. Blanch, Molecular Physics.
2004, 102, 731-734.
L. D. Barron, F. Zhu, L. Hecht, G. E. Tranter, N. W. Isaacs, Journal of
Molecular Structure. 2007, 834, 7-16.
L. D. Barron, L. Hecht, E. W. Blanch, A. F. Bell, Progress in Biophysics &
Molecular Biology. 2000, 73, 1-49.
L. D. Barron, E. W. Blanch, L. Hecht, Unfolded Proteins Studied by Raman
Optical Activity, in Advances in Protein Chemistry. 2002, Elsevier Science &
Technology Academic Press Inc.
E. W. Blanch, L. Hecht, L. D. Barron, Methods. 2003, 29, 196-202.
J. King, Protein and Nucleic Acid Structure and Dynamics. 1985,
Benjamin/Cummings Pub. Co. 587.
L. D. Barron, E. W. Blanch, H. I. McColl, C. D. Syme, L. Hecht, K. Nielsen,
Spectroscopy: An International Journal. 2003, 17, 101-126.
W. L. Peticolas, Raman spectroscopy of DNA and proteins, in Methods in
Enzymology, S. Kenneth, Editor. 1995, Academic Press. p. 389-416.
A. J. Hobro, M. Rouhi, E. W. Blanch, G. L. Conn. Nucleic Acids Research.
2007, 35, 1169-1177.
E. W. Blanch, L. Hecht, C. D. Syme, V. Volpetti, G. P. Lomonossoff, K.
Nielsen, L. D. Barron, Journal of General Virology. 2002, 83, 2593-2600.
L. D. Barron, Current Opinion in Structural Biology. 2006, 16, 638-643.
B. Pettinger, U. Wenning, Chemical Physics Letters. 1978, 56, 253-257.
M. Fleishmann, P. J. Hendra, A. McQuillan, Journal of Chemical Physics
Letter. 1974, 24, 163-166.
D. L. Jeanmaire, R. P. Van Duyne, Journal of Electroanalytical Chemistry
and Interfacial Electrochemistry. 1977, 84, 1-20.
M. G. Albrecht, J. A. Creighton, Journal of the American Chemical Society.
1977, 99, 5215-5217.
S. Nie, S. R. Emory, Science. 1997, 275, 1102-1106.
K. Kneipp, Y. Wang, H. Kneipp, L. T. Perelman, I. Itzkan, R. R. Dasari, M.
S. Feld, Physical Review Letters. 1997, 78, 1667-1670.
A. Campion, P. Kambhampati, Chemical Society Reviews. 1998, 27, 241-250.
S. H. Macomber, T. E. Furtak, Solid State Communications. 1983, 45, 267271.
A. G. Brolo, D. E. Irish, B. D. Smith, Journal of Molecular Structure. 1997,
405, 29-44.
R. K. Chang, B. L. Laube, Critical Reviews in Solid State and Materials
Sciences. 1984, 12, 1-73.
M. Moskovits, Reviews of Modern Physics. 1985, 57, 783-826.
W. E. Smith, Chemical Society Reviews. 2008, 37, 955-964.
H. Seki, Raman Spectra of Molecules Considered to be Surface Enhanced, in
Studies in Surface Science and Catalysis. D. A. King, N. V. Richardson and
S. Holloway, Editors. 1986, Elsevier. p. 289-310.
J. A. Creighton, Surface Science. 1983, 124, 209-219.
M. Kerker, Accounts of Chemical Research. 1984, 17, 271-277.
46
Chapter 1
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
E. J. Zeman, G. C. Schatz, The Journal of Physical Chemistry. 1987, 91, 634643.
J. C. Tsang, J. R. Kirtley, J. A. Bradley, Physical Review Letters. 1979, 43,
772-775.
A. Otto, Journal of Raman Spectroscopy. 2005, 36, 497-509.
Z. M. Pan, S. H. Ueda, A. Mu, R. Cui, Y. Guo, M. Burger, A. Yeh Yin,
Journal of Raman Spectroscopy. 2005, 36, 1082-1087.
E. Kai, S. Sawata, K. Ikebukuro, T. Iida, T. Honda, I. Karube, Analytical
Chemistry. 1999, 71, 796–800.
P. A. van der Merwe, A. N. Barclay, Current Opinion in Immunology. 1996,
8, 257–261.
M. Besenicar, P. Macek, J. H. Lakey, G. Anderluh, Chemistry and Physics of
Lipids. 2006, 141, 169–178.
H. Celia, E. Wilson-Kubalek, R. A. Milligan, L. Teyton, Proceedings of the
National Academic Sciences. U. S. A. 1999, 96, 5634–5639.
E. Hutter, M. P. Pileni, Journal of Physical Chemistry B. 2003, 107, 6497–
6499.
B. P. Nelson, T. E. Grimsrud, M. R. Liles, R. M. Goodman, R. M. Corn,
Analytical Chemistry. 2000, 73, 1–7.
T. T. Goodrich, H. J. Lee, R. M. Corn, Journal of the American Chemical
Society. 2004, 126, 4086–4087.
H. J. Lee, A. W. Wark, R. M. Corn, Langmuir. 2006, 22, 5241–5250.
C. S. Thaxton, D. G. Georganopoulou, C. A. Mirkin, Clinica Chimica Acta.
2006, 363, 120–126.
H. Li, L. J. Rothberg, Proceeding of the National Academic Sciences. U. S. A.
2004, 101, 14036–14039.
S. Stewart, P. M. Fredericks, Spectrochimica Acta Part A. 1999, 55, 1615–
1640.
E. Podstawka, Y. Ozaki, L. M. Proniewicz, Applied Spectroscopy. 2005, 59,
1516–1526.
E. Podstawka, Y. Ozaki, L. M. Proniewicz, Applied Spectroscopy. 2004, 58,
570–580.
A. E. Aliaga, I. Osorio-Roman, C. Garrido, P. Leyton, J. Carcamo, E. Clavijo,
J. S. Gomez-Jeria, G. Diaz, M. M. Campos-Vallette, Vibrational
Spectroscopy. 2009, 50, 131–135.
E. J. Bjerneld, P. Johansson, M. Kall, Single Molecules. 2000, 1, 239–248.
A. Kudelski, Vibrational Spectroscopy. 2008, 46, 34–38.
T. Deckert-Gaudig, E. Bailo, V. Deckert, Physical Chemistry Chemical
Physics. 2009, 11, 7360–7362.
V. P. Drachev, M. D. Thoreson, V. Nashine, E. N. Khaliullin, D. Ben-Amotz,
V. Jo Davisson, V. M. Shalaev, Journal of Raman Spectroscopy. 2005, 36,
648-656.
S. C. Pînzaru, L. M. Andronie, I. Domsa, O. Cozar, S. Astilean, Journal of
Raman Spectroscopy. 2008, 39, 331-334.
T. Vo-Dinh, L. R. Allain, D. L. Stokes, Journal of Raman Spectroscopy.
2002, 33, 511-516.
Y. C. Cao, R. Jin, C. A. Mirkin, Science. 2002, 297, 1536-1540.
S. Abdali, E. W. Blanch, Chemical Society Reviews. 2008, 37, 980-992.
S. Efrima, Chemical Physics Letters. 1983, 102, 79-82.
S. Efrima, The Journal of Chemical Physics. 1985, 83, 1356-1362.
47
Chapter 1
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
L. Hecht, L. D. Barron, Journal of Molecular Structure. 1995, 348, 217-220.
L. Hecht, L. D. Barron, Chemical Physics Letters. 1994, 225, 525-530.
B. G. Janesko, G. E. Scuseria, Journal of Chemical Physics. 2006, 125,
124704-124710.
P. G. Etchegoin, C. Galloway, E. C. Le Ru, Physical Chemistry Chemical
Physics. 2006, 8, 2624-2628.
P. Bour, The Journal of Chemical Physics. 2007, 126, 136101-136103.
B. G. Janesko, G. E. Scuseria, The Journal of Physical Chemistry C. 2009,
113, 9445-9449.
N. Yang, Y. Tang, A. E. Cohen, Nano Today. 2009, 4, 269-279.
R. Lombardini, R. Acevedo, N. J. Halas, B. R. Johnson, Journal of Physical
Chemistry C. 2010, 114, 7390-7400.
V. Novak, J. Sebestík, P. Bour, Journal of Chemical Theory and
Computation. 2012, 8, 1714-1720.
S. Ostovar Pour, S. E. J. Bell, E. W. Blanch, Chemical Communications.
2011, 47, 4754-4756.
H. Kneipp, J. Kneipp, K. Kneipp, Analytical Chemistry. 2006, 78, 1363-1366.
S. Abdali, C. Johannessen, J. Nygaard, T. Nørbygaard, Journal of Physics:
Condensed Matter. 2007, 19, 285205.
S. Abdali, Journal of Raman Spectroscopy. 2006, 37, 1341-1345.
C. Johannessen, P. C. White, S. Abdali, Journal of Physical Chemistry
A, 2007. 111, 7771–7776
K. Osińska, M. Pecul, A. Kudelski, Chemical Physics Letters. 2010, 496, 8690.
A. J. Hobro, PhD thesis, Structural Investigation of RNA through Application
of Raman, Raman Optical Activity and Surface Enhanced Spectroscopies.
2008, University of Manchester, UK.
N. R. Yaffe, PhD thesis, Raman Spectroscopic Studies of Biological
Molecules. 2009, University of Manchester, UK.
48
Chapter 2
Chapter 2
Use of a Hydrogel Polymer for Reproducible
Surface Enhanced Raman Optical Activity
(SEROA)
49
Chapter 2
2.0 Declaration
This chapter consists of one published full paper: S. Ostovar Pour, S. E. J. Bell, E.
W. Blanch, Chemical Communication. 2011, 47, 4754–4756.
The manuscripts have been incorporated in a format identical to that for journal
submission, except for minor adjustments to incorporate them into this thesis. As
first author on this publication I carried out all of the associated experimental and
spectroscopic analysis. The polymer was provided by Dr Steve Bell at the Queen’s
University of Belfast.
50
Chapter 2
Use of a Hydrogel Polymer for Reproducible Surface
Enhanced Raman Optical Activity (SEROA)
Saeideh Ostovar Pour,*a Steven E. J. Bellb and Ewan W. Blancha
Received 30th November 2010, Accepted 25th February 2011
a
Faculty of Life Sciences, Manchester Interdisciplinary Biocentre, The University of Manchester, 131
Princess Street, Manchester, UK M1 7DN.
E-mail: [email protected]; Fax: +44 (0)161 236 0409; Tel: +44 (0)161 306 5819
b
School of Chemistry and Chemical Engineering, The Queen’s University of Belfast, Belfast, UK
BT9 5AG.
E-mail: [email protected] ; Fax: +44(0) 2890 976524; Tel: +44(0)2890 974470
2.1 Abstract
We present surface enhanced Raman optical activity (SEROA), as well as Raman,
SERS and ROA, spectra of D- and L-ribose. By employing a gel forming polyacrylic
acid to control colloid aggregation and associated birefringent artefacts we observe
the first definitive proof of SEROA through measurement of mirror image bands for
the two enantiomers.
2.2 Introduction
As a result of its sensitivity to chirality, Raman optical activity (ROA), which
measures a small difference in the intensity of vibrational Raman scattering from
chiral molecules in right- and left-circularly polarized light [1,2] is a powerful probe
of the structure and behaviour of biomolecules in aqueous solution [3–7].
However, ROA is a very weak effect being ~3–5 orders of magnitude smaller than
the parent Raman scattering. The conditions of high concentration and long data
collection time required for ROA currently limit its application for a wide range of
51
Chapter 2
biological samples. These limitations could possibly be overcome using the principles
of surface enhanced Raman scattering (SERS) [8–10] in which a sample in the
presence of surface plasmons localized on a neighbouring nanostructured feature of a
metal surface can interact with the incident light leading to large enhancement of the
Raman scattering. However, the generation of reliable SEROA spectra of
biomolecules has been problematic due to difficulties in controlling spectral artefacts
and low signal-to-noise ratios which complicate detection of true SEROA signals.
Although several papers have presented possible SEROA spectra [11, 12] currently a
proof demonstrating mirror image SEROA spectra from opposite enantiomers has not
been reported. Recently, observation of SEROA spectra for the L- and D-enantiomers
of cysteine has been claimed [13]; however the authors stated that no corresponding
SERS spectra could be measured under the same conditions. Therefore, no surface
enhancement had occurred in the earlier study [13] either SERS or SEROA with the
observed spectral features probably being reflection associated birefringent artefacts.
Thus, SEROA has still not been confirmed as an experimental technique. SEROA
spectral features depend on SERS experimental conditions, since they reflect the
stability of colloids over time periods longer than those typically used in conventional
SERS [14] Contributing factors, including the concentration of analyte and
aggregating agents, pH, type of colloid and time dependence, have been studied in
order to determine the effects of these parameters on SERS spectra [15–17] as they
should also be optimal for measuring SEROA [18]. However, it has proven difficult
to stabilise the extent of aggregation in colloidal systems sufficiently to control the
fluctuation of bands in SEROA experiments, complicating validation of observed
spectral features. Etchegoin et al. modelled the effect of surface plasmons on
circularly polarized light [19]. Their calculations suggest that large artefacts in
52
Chapter 2
SEROA spectra would be highly sensitive to the nature of colloid–colloid
interactions, explaining the origin of the intense and fluctuating features often
observed in SEROA experiments. Slowing down changes in the aggregation state to
minimize changes in colloidal interactions should improve the reliability and
reproducibility of SEROA spectra.
In this study we employ a hydrophilic polyacrylic acid ‘‘polycarbopol’’ polymer as a
stabilising medium. This polymer has small Raman and surface-enhanced Raman
cross sections, minimising interference from background signals, does not
significantly change the UV-vis absorption spectra when added to silver colloids and
is known to stabilise even aggregated colloids for extended periods of time [20,21].
We report SERS and SEROA spectra, along with the Raman and ROA spectra of Dand L-ribose measured in the presence of citrate-reduced silver colloids and the
polycarbopol polymer, providing the first definitive observation of SEROA.
2.3 Experimental
Silver nitrate (99%), sodium borohydride (99%), sodium citrate (99%), sodium
hydroxide (99%), potassium sulphate (99%), D- and L-ribose (99%) were purchased
from Sigma- Aldrich UK and used without further purification.
Citrate reduced silver colloids were prepared by reduction of silver nitrate with citrate
ions [22] see supplementary information for details. The polycarbopol polymer was
purchased from B.F. Goodrich Ltd and used without further purification to form the
polymer-sol mixture.
The Raman (IR + IL), scattered circularly polarized (SCP) ROA (IR - IL), SERS (IR +
IL) and SCP SEROA (IR - IL) spectra were all measured using a ChiralRAMAN SCP
spectrometer (BioTools Inc., Jupiter FL) operating in the backscattering configuration
53
Chapter 2
at an excitation wavelength of 532 nm with spectral resolution of 7 cm-1. Raman and
ROA spectra were taken with laser power of 0.625 W at the sample with data
collection times of 4–6 h. The laser power for SERS and SEROA was 0.25 W at the
sample with data collection times of 35 min. The details of sample, aggregating agent
and colloid concentration are given in each figure legend.
All SERS samples were prepared to 1 ml, the sample was left to sit for 15 min in
order to obtain maximum SERS enhancement, which was determined from time
dependence measurements, and then 20 mg of polycarbophil polymer powder was
added and stirred vigorously for a few seconds, then left for 60 min in order to allow
full hydration and swelling of the polymer prior to data collection.
2.4 Results and Discussion
The Raman and ROA spectra in aqueous solution obtained for D- and L-ribose are
shown in Figure 2.1. All spectra (Raman, ROA, SERS and SEROA) presented in this
study are raw data without any smoothing, base lining, normalization or any other
data pretreatment. The Raman and ROA band assignments for both enantiomers of
ribose are summarized in Table S2.1 in supplementary information [23–25]. The
Raman and ROA spectra of D-ribose are in excellent agreement with those reported
by Wen et al. [23] and those measured recently by Dr C. Johannessen in Glasgow
(personal communication). We have repeated the Raman and ROA spectra for Lribose, but these have not been previously reported. Mirror image responses are
observed for most ROA bands, though it is not known why no ROA band appears
near 877 cm-1 for L-ribose, though this spectrum is reproducible.
Figure 2.2 A and E presents the Raman and ROA spectra, respectively, of
polycarbopol in solution, measured at the same concentration as used in the SEROA
54
Chapter 2
experiments, with SERS and SEROA spectra of D- and L-ribose shown in Figure 2.2
C, D, G and H, respectively, before and after addition of polycarbopol polymer. The
Raman spectrum of polycarbopol shows that the polymer does not generate any
significant Raman signal as this polymer has a very small Raman cross section [20],
and only the spectrum of water is evident. Although the polycarbopol subunit is
chiral, it has a low Raman cross-section [21] so helping to minimise its ROA
spectrum. Together, this leads to the ROA spectrum of polycarbopol being very
weak, barely above the noise level. Figure 2.2 B and F shows spectra for the
combination of silver colloids, aggregating salt and polycarbopol (no analyte). The
two stronger bands at ~1394 and 1452 cm-1 in the SERS spectrum are a fingerprint of
the sol with polycarbopol.
The corresponding SEROA spectrum has negative features which are very noisy, that
arise from both the polycarbopol and the interaction of plasmon resonances with
circularly polarized light. Figure 2.2 C and D presents the experimental SERS spectra
of D- and L-ribose before and after, respectively, the addition of the polymer. The
SERS spectra for D- and L-ribose shown in Figure 2.2 C and D were obtained using
the optimum type of aggregating agent (K2SO4), its concentration (20 mM) and pH
(8.7). The optimum concentration of the polycarbopol polymer was found to be 20
mg ml-1 which generated a viscous solution that was dilute enough to pipette but
thick enough to control the aggregation of colloids, and gave rise to strong SERS
signals for an extended period of time. The spectra demonstrate that the SERS signals
for the two enantiomers are similar both in the presence (Figure 2.2 D) and the
absence (Figure 2.2 C) of the polymer. All bands measured in the conventional SERS
experiments appear at the same position in the presence of the polymer with only
small differences in relative intensities of bands, confirming that the polymer does
55
Chapter 2
not interfere with signals from ribose molecules. Time dependence measurements,
see Figure S2.3 in supplementary information, show that SERS intensity is stable for
over 35 min with the polymer, but for only 10 min without polymer. We conclude,
therefore, that the addition of the polymer increases the stability of the aggregated
colloids, allowing measurement of reliable SERS signals from the analyte. The
SEROA spectra of D- and L-ribose measured in the absence of polymer are shown in
Figure 2.2 G. These spectra present a common problem that can occur in attempts to
measure SEROA spectra. The SEROA spectrum of D-ribose gives rise to a mix of
+ve and -ve bands, which are what may be expected in a chiroptical measurement,
but in the spectrum of L-ribose all of the bands are negative in sign, due to difficulties
in controlling the highly birefringent background signal. Therefore, we do not
observe a mirror image response in Figure 2.2 G for any of the purported SEROA
bands generated by the two enantiomers due to the large birefringence generated by
the surface plasmons from the aggregating colloids, making it difficult to have
confidence in the reliability of either of these two spectra. Furthermore, though the
SERS spectra presented in Figure 2.2 C, which are insensitive to this problem, could
be reproduced many times, the corresponding SEROA spectra demonstrated poor
reproducibility both from sample-to-sample and as a function of time.
Figure 2.2 H shows the SEROA spectra of D- and L-ribose with polycarbopol
polymer. Both D- and L-ribose give highly reproducible SEORA spectra (see Figure
S2.4 in supplementary information for replicate measurements) with positive and
negative bands. Critically, despite baseline variations, mirror image bands are now
observed for the two enantiomers. The SEROA spectrum of D-ribose displays a
number of bands that clearly show the opposite sign to their L-ribose counterparts.
The +ve SEROA bands for D-ribose at 1247, 1273 and 1315 cm-1 correspond to the -
56
Chapter 2
ve SEROA bands for L-ribose at 1242, 1270 and 1310 cm-1, respectively. A complex
-ve/+ve/-ve triplet exhibited by D-ribose from ~1100–1230 cm-1 is nicely replicated
as a +ve/-ve/+ve triplet by L-ribose with similar band shapes and intensities, as is the
+ve/-ve couplet for D-ribose from ~1430–1500 cm-1. A strong +ve SEROA band at
1571 cm-1 for D-ribose gives rise to an equivalent -ve feature for L-ribose. The
regions between 1300–1400 cm-1 and below 1000 cm-1 reveal a number of weak
features that appear to show opposite sign for the two enantiomers, though variations
in local baselines due to residual birefringent background signals complicate their
analysis. However, several features in the SEROA spectrum for D-ribose do not lead
to a mirror image for the L-enantiomer, most notably the +ve bands at ~1014 and
1539 cm-1 , while there are also no counterpart features to the -ve SEROA bands
displayed by L-ribose at ~1699 and 1739 cm-1 . The reasons for these differences are
not known, but they are reproducible (Figure S2.4, supplementary information) so do
not originate from variable birefringent artefacts or shot noise.
In order to verify the reliability of this method further, L- and D-tryptophan were also
measured, Figure S2.5 (spectra provided in supplementary information). In both case
L- and D-tryptophan provided mirror image response in SEROA Spectra.
2.5 Conclusion
We have demonstrated the first experimental proof of SEROA by recording SEROA
spectra for two enantiomers, D- and L-ribose, along with their corresponding SERS
and ROA spectra. Addition of the polycarbopol polymer provides a solution to the
problem of how to stabilize the aggregated colloids, and so reduce the effect of
plasmon resonance induced changes in circularly polarized light that typically plague
57
Chapter 2
SEROA experiments. This strategy will allow the potential of SEROA to be more
effectively explored.
58
Chapter 2
2.6 References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
P. W. Atkins, L. D. Barron, Molecular Physics. 1969, 16, 453–466.
L. D. Barron, M. P. Bogaard, A. D. Buckingham, Journal of the American
Chemical Society. 1973, 95, 603–605.
L. D. Barron, L. Hecht, E. W. Blanch Molecular Physics. 2004, 102, 731–
744.
L. D. Barron, Current Opinion in Structural Biology. 2006, 16, 638–643.
T. Uchiyama, M. Sonoyama, Y. Hamada, R. K. Dukor, L. A. Nafie, F.
Hayashi, K. Oosawa, Vibrational Spectroscopy. 2008, 48, 65–68.
L. D. Barron, E. W. Blanch, I. H. McColl, C. D. Syme, L. Hecht, K. Nielsen,
Spectroscopy. 2003, 17, 101–126.
E. W. Blanch, L. Hecht, L. D. Barron, Methods. 2003, 29, 196–202.
D. L. Jeanmaire, R. P. Van Duyne, Journal of Electroanalytical Chemistry.
1977, 84, 1–20.
K. Kneipp, Y. Wang, H. Kneipp, L. T. Perelman, I. Itzkan, R. Dasari, M. S.
Feld, Physical Review Letters. 1997, 78, 1667–1670.
E. Koglin, H. H. Lewinsky, J. M. Sequaris, Surface Science. 1985, 58, 370–
380.
C. Johannessen, P. C. White, S. Abdali, Journal of Physical Chemistry A.
2007, 111, 7771–7776.
N. A. Brazhe, A. R. Brazhe, O. V. Sosnovtseva, S. Abdali, Chirality. 2009,
21, E307–E312.
K. Osinska, M. Pecul, A. Kudelski, Chemical Physics Letters. 2010, 496, 86–
90.
S. Abdali, Journal of Raman Spectroscopy. 2006, 37, 1341–1345.
A. J. Hobro, S. Jabeen, B. Z. Chowdhry, E. W. Blanch, Journal of Physical
Chemistry C. 2010, 114, 7314–7323.
N. R. Yaffe, E. W. Blanch, Vibrational Spectroscopy. 2008, 48, 196–201.
N. R. Yaffe, A. Ingram, D. Graham, E. W. Blanch, Journal of Raman
Spectroscopy. 2009, 41, 618–623.
S. Abdali, E. W. Blanch, Chemical Society Reviews. 2008, 37, 980–992.
P. G. Etchegoin, C. Galloway, E. C. Le Ru, Physical Chemistry Chemical
Physics. 2006, 8, 2624–2628.
S. E. J. Bell, S. J. Spence, Analyst. 2001, 126, 1–3.
S. E. J. Bell, N. M. S. Sirimuthu, Analyst. 2004, 129, 1032–1036.
P. C. Lee, D. Meisel, Journal of Physical Chemistry. 1982, 86, 3991.
Z. Q. Wen, L. D. Barron, L. Hecht, Journal of the American Chemical
Society. 1993, 115, 285–292.
P. Carmona, M. Molina, Journal of Raman Spectroscopy. 1990, 21, 395–400.
M. Mathlouthi, A. M. Seuvre, J. L. Koenig, Carbohydrate Research. 1983,
122, 31–47.
59
Chapter 2
1270
1467
D - Ribose
L - Ribose
1640
D- Ribose
L- Ribose
1363
1260
ROA
1167
1048
1069
1105 1135
964
1009
510
IR - IL
455
7.3x10
652
550
6
598
877
970
1014
879
10
4.4x10
919
601
682
729
805
547
10
6.6x10
420
IR + IL
8.8x10
1127
10
1083
Raman
400
800
1000
1200
1467
1363
1262
1009
967
600
1048
1069
1105
1135
1167
874
652
510
544
598
6
-7.3x10
452
0.0
1400
1600
1800
-1
wavenumber (cm )
Figure 2.1: Raman (IR + IL, top) and SCP ROA (IR – IL, bottom) spectra of D- and Lribose in aqueous. Sample concentrations were 2.66 M at pH 5.46 (D-) and 5.60 (Lribose), data collection time was 237.02 min, and laser power at the sample 0.625 W
for each.
60
Chapter 2
A
8
6.9x10
8
4.6x10
8
1394
IR + IL
1.5x10
1452
B
7
7.5x10
1366
1624
D-ribose
L- ribose
9
2.4x10
C
9
1.2x10
0.0
8
7.0x10
D
8
3.5x10
4
7.8x10
E
0.0
0.0
F
4
IR - IL
-5.0x10
5
-1.0x10
0.0
G
5
-6.5x10
1539
1247
H
1273
5
2.5x10
1014
1315
0.0
1699 1739
1242
400
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure 2.2: Raman spectrum of polycarbopol in solution (A), SERS spectra before
(B) and after addition of L- and D-ribose (0.25 mg ml-1) in the presence of silver
61
Chapter 2
citrate reduced colloid and K2SO4 at 0.020 M concentration, data collection time: 20
min (C), SERS spectra of L- and D-ribose (0.25 mg ml-1) in the presence of
polycarbopol polymer, data collection time: 20 min (D), ROA spectrum of
polycarbopol polymer in solution, sample concentration 40 mg ml-1, data collection
time: 218 min (E), SEROA spectra of silver citrate reduced colloids in presence of
aggregating salt before (F) and after addition of L- and D-ribose, data collection
time: 35 min (G), SEROA spectra of L- and D-ribose with addition of polycarbopol,
data collection time of 35 min (H).
62
Chapter 2
2.7 Supplementary Information
*
n
Polycarbopol
Figure S2.1: A schematic diagram of SEROA sample preparation with use of the
polycarbopol polymer.
2.7.1 Colloid Preparation
Citrate-reduced silver colloids were prepared by reduction of silver nitrate with
citrate ions (Lee and Meisel method),21 where 0.094 g of AgNO 3 was dissolved in
500 ml of distilled H 2O and heated to boiling point, then 10 ml of 1% trisodium
citrate solution was added drop wise to the mixture. Heating was continued for
another hour with constant stirring and then the solution was allowed to cool to
room temperature. Approximately 300 ml of a green-grey solution was obtained
at ~0.5 M concentration. All glassware used to prepare the colloids was washed
prior to use with aqua regia followed by gentle scrubbing with a 2% Helmanex
solution and thorough rising with distilled water.
2.7.2 Sample Preparation for Raman and ROA Measurements
Samples of D- and L-ribose for Raman and ROA spectra were prepared by
dissolving into distilled water at 100 mg/ml, then were microcentrifuged for 5
minutes at 3000 rpm (1000 g) to minimize dust particles prior to loading into
quartz microflourescence cells.
63
Chapter 2
2.7.3 Atomic Force Microscopy
Micrographs were obtained using a Veeco Picoforce Multimode AFM with
standard extender module, Nanoscope IIIA controller and a Picoforce scanner.
Each AFM plate was prepared by adding 50 µl of sample to freshly cleaved mica
and left at room temperature for 30 minutes. The mica was then rinsed carefully
under distilled water for approximately 10 seconds and dried under a gentle
stream of nitrogen. AFM was carried out in air in tapping mode with a scan size
of 5 microns and a scan rate of 0.5 Hz using a Silicon ‘TAP300’ AFM cantilever
and tip (oscillated at approximately 260 kHz). Figure S2.2 presents AFM images
of silver citrate reduced colloids before and after addition of polycarbopol
polymer. The micrograph of polycarbopol without metal colloids (Figure S2.2 B)
shows a very smooth surface whereas the metal colloids give rise to a very rough
surface (Figure S2.2 A). Addition of polymer to metal colloids does not induce
significant change to morphologies of the metal particles that are observed in
Figure S2.2 C and confirms that aggregation is significantly reduced.
64
Chapter 2
Figure S2.2: Dual views of AFM images of silver citrate reduced colloids only (A),
polycarbopol polymer only (B) and a mixture of silver citrate reduced colloids and
polycarbopol polymer (C).
Random regions of different micrographs were selected and the diameters of
nanoparticles contained within were measured by using the measuring tool in the
Nanoscope 7.2 software. The average particle sizes of silver colloids with and
without polycarbopol polymer were ~67 and 70 nm, respectively, so the particle size
in the polymer gel was very similar to that observed for normal silver colloids.
However, the individual colloidal particles are much more distinct in the AFM
images upon addition of the polymer. This is not an issue as the maximum SERS
enhancement was obtained before controlling the aggregation process. The
micrographs in combination with the time dependent SERS data (Figure S2.3), which
are discussed below, confirm that the addition of polycarbopol to silver colloid
controls the aggregation process of the nanoparticles.
65
Chapter 2
2.7.4 SERS Time Dependence
The time dependence study was performed by measuring the intesities of bands at
1366 and 1409 cm-1. Figure S2.3 shows the intensity changes in both normal and
polymer-SERS solution over a 35 mins time interval. The intensity in normal SERS
experiments decreased significantly after 10 mins. As is shown in Figure S2.3, the
enhancement was kept constant in the polymer-SERS sample for a substantially
longer time.
D-ribose in polymer-SERS solution
D-ribose in normal SERS solution
9
1.8x10
9
1.6x10
9
1.4x10
SERS Intensity
9
1.2x10
9
1.0x10
8
8.0x10
8
6.0x10
8
4.0x10
8
2.0x10
0.0
5-10
10-15
15-20
20-25
25-30
30-35
Time Intervals
Figure S2.3: Time dependence SERS study for D-ribose molecule with and without
addition of polymer to SERS solution.
66
Chapter 2
5
2.3x10
D-Ribose
0.0
5
2.3x10
0.0
5
2.8x10
0.0
5
IR - IL
2.8x10
0.0
5
1.3x10
L-Ribose
0.0
5
1.4x10
0.0
5
1.4x10
0.0
5
1.8x10
0.0
400
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure S2.4: Example repeats of SEROA spectra for D- (top) and L-ribose (bottom)
with the same conditions for each spectrum, where concentration of each sample was
0.25 mg/ml and K2SO4 concentration was 0.020 M, with addition of 20 mg/mL of
polycarbopol, data collection time of 35 mins.
67
Chapter 2
Table S2.1: Raman and ROA band assignments of L- and D-ribose in aqueous
solution
Raman
D-ribose (cm-1)
420
467
547
601
652
682
729
805
879
919
970
1014
1083
1127
1270
1327
1467
1640
L-ribose (cm-1)
420
467
547
601
652
682
729
805
879
919
970
1014
1083
1127
1270
1327
1467
1640
ROA
D-ribose (cm-1)
420 (+ve)
452 (-ve)
550 (+ve)
598 (+ve)
652 (-ve)
682w (+ve)
726 (+ve)
802 (+ve)
877 (+ve)
914 (-ve)
964 (+ve)
1009 (-ve)
1048 (+ve)
1069 (-ve)
1105 (-ve)
1135 (+ve)
1260 (+ve)
1363 (-ve)
1467 (-ve)
-
L-ribose (cm-1)
408 (-ve)
455 (+ve)
544 (+ve)
595 (-ve)
652 (+ve)
682w (+ve)
732 (+ve)
802 (+ve)
874 (-ve)
914 (+ve)
967 (-ve)
1009 (+ve)
1048 (-ve)
1069 (+ve)
1105 (+ve)
1135 (-ve)
1262 (-ve)
1363 (+ve)
1467 (-ve)
-
*ν = stretching mode, δ = bending, τ = torsion, ω = wagging
68
Assignments
δ CCO +δ CCC23
δ CCO +δ CCC23
Sym, ring bend23
δ CCO +δ CCC23
δ OCO, β anomer22,23
δ OCO, α anomer22,24
ν CC, δ CCO +δ CCC23
ν CC+ ν CO+ δ OH24
ν CC+ ν CO+ δ OH24
ν CC+ ν CO24
ν CC+ ν CO + δ CH24
ν CC+ ν CO24
ν CC+ ν CO24
ν CC+ ν CO24
ν CC+ ν CO24
ν CO24
τ CH2 + δ OH23
δ CH+ ω CH223
δ CH223
δ CH223
Chapter 2
A
10
1.20x10
0.00
9
2.80x10
B
9
IR + IL
1.40x10
0.00
7
C
6.0x10
7
3.0x10
8
1.18x10
D
7
5.90x10
6
2.20x10
E
0.00
6
IR - IL
1.30x10
F
0.00
4
3.20x10
G
0.00
4
2.70x10
H
0.00
400
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure S2.5: Raman (IR + IL, A and B) and SCP ROA (IR – IL, E and F) spectra of Dand L-tryptophan in aqueous. Sample concentrations were 50 mg/ml at pH 1.90 (D-)
and 1.58 (L-tryptophan), data collection time was 4-8hrs, and laser power at the
sample 0.625 W for each. SERS spectra after addition of D- and L-tryptophan (C and
D), (0.0002 M) in the presence of silver hydroxyalamine reduced colloid and MgSO4
at 0.050 M concentration, data collection time: 5 min, SEROA spectra of D- and Ltryptophan (G and H) with addition of polycarbopol, data collection time of 35 min.
69
Chapter 3
Chapter 3
Induced Chirality to Non-chiral Surfaces of
Silver Silica Nanotags
70
Chapter 3
3.0 Declaration
This chapter consists of one draft paper awaiting submission to Journal of the
American Chemical Society: S. Ostovar pour 1*, L. Rocks 2, K. Faulds 2, D. Graham2
and E.W. Blanch1, Journal of the American Chemical Society. 2012
The manuscript has been incorporated in a format identical to that for journal
submission except for minor adjustments to incorporate them into this thesis. As the
first author on this publication I have carried out all associated spectroscopic
measurements and analysis. The experimental work involving the silica nanotag
preparation was carried out at the University of Strathclyde by Ms. Louise Rocks.
71
Chapter 3
Induced Chirality to Non-chiral Surfaces of Silver Silica
Nanotags
S. Ostovar pour 1*, L. Rocks 2, K. Faulds 2, D. Graham2 and E.W. Blanch1
1
Faculty of Life Sciences, Manchester Interdisciplinary Biocentre, The University of Manchester, 131
Princess Street, Manchester, UK, M1 7DN.
2
Centre for Molecular Nanometrology, WestCHEM, Department of Pure and Applied Chemistry,
University of Strathclyde, 295 Cathedral Street, Glasgow, UK, G1 1XL.
*
Corresponding author: [email protected]; Fax: +44 (0)161 236 0409; Tel: +44 (0)161 306
5819
3.1 Abstract
Nanoprobes can offer advantages for chemical analysis, since spectroscopic detection
of biomolecules is limited by the inherently weak Raman effect. In the present work,
we demonstrate that silver silica nanotags can provide an enhancement for the
chirally sensitive but even weaker Raman optical activity (ROA) effect enabling
highly sensitive detection of chiral biomolecules. These single nanoparticle
plasmonic substrates function as achiral plasmonic nanomaterials that produce
intense optical activity. This system enables a chiral response to be transmitted from a
chiral analyte to the plasmon resonance of an achiral metallic nanostructure. The
chiroptical properties of these nanotags were confirmed by the measurement of
mirror image surface enhanced resonance Raman optical activity (SERROA) spectra
of the two enantiomers of each of ribose and tryptophan. This highly sensitive probe
of chiral molecules can provide a new approach for studying biomolecules in
solution. This is the first report of colloidal metal nanoparticles in the form of single
plasmonic substrates displaying an intrinsic chiral sensitivity once attached to a chiral
72
Chapter 3
molecule. Therefore, the observed mechanism is a novel and remarkable fundamental
effect which can provide a new route for engineering chiral plasmonic nanomaterials.
3.2 Introduction
Single nanoparticle plasmonic substrates, such as hollow gold nanospheres [1], silver
triangles [2], gold nanorods [3] and gold/silver silica nanoshells [4], have localized
surface plasmon resonances that match the excitation wavelengths of lasers used in,
for example, Raman spectroscopy. They provide a strong electromagnetic field that
increases the Raman cross-section giving rise to the technique of surface-enhanced
Raman scattering (SERS), as opposed to relying upon high conjunction potentials or
‘hot spots’ that are characteristic of aggregated metal colloids. Additionally, they
provide large scattering cross-sections that allow colorimetric detection of analytes at
relatively low concentrations [5].
In previous studies, silver silica nanotag particles conjugated to dye molecules have
been used as an approach to enhance the sensitivity of SERS [6]. Silica coated silver
colloids that have been conjugated to dyes such as benzotriazole have an electronic
transition that can be tuned to the excitation frequency of the laser, leading to
surfaced-enhanced resonance Raman scattering (SERRS). Although the Raman
signal enhancement by single nanoparticle plasmonic substrates can detect and
provide useful information about the conformations of biomolecules, this technique
is blind to chirality. Determination of molecular stereochemistry and the consequent
impact of chirality on bioactivity are of particular significance owing to the fact that
99% of biomolecules are chiral. The aim of this paper is to determine the possibility
of identification of chiral molecules by nanotags, in the form of chirally sensitive
nanoprobes, using Raman optical activity (ROA) spectroscopy. ROA measures the
73
Chapter 3
intensity difference between Raman scattering of left- and right-circular polarized
light from chiral molecules) [7-10]. No ROA signals are detected if a molecule
possesses mirror-symmetry planes or centres of symmetry, and spectral bands of
opposing signs are detected for enantiomers [11]. Although ROA is a very sensitive
technique for studying biological samples it is also a very weak effect, being
approximately 3-5 orders of magnitude less than the parent Raman scattering. The
conditions of high concentration and long data acquisition time required for ROA
currently limit its application for a wide range of biological samples. These
limitations could possibly be circumvented using the application of SERRS [12-14]
due to the large increase in sensitivity which makes the technique worthy of
consideration for the enhancement of ROA spectra.
Recently, a class of nanomaterials, known as chiral plasmonic nanomaterials [15-17],
has attracted much attention due to their ability to be used as broad band circular
polarisers and to generate superchiral electromagnectic fields for ultrasensitive
conformational detection of biomolecules [18]. Plasmonic nanomaterials such as
metallic nanoparticles experience strong absorption in the visible wavelength range
but are achiral with no inherent chiroptical properties [15-17]. It has previously been
demonstrated that when biomolecules are assembled with nanomaterials, chirality
from the biomolecule can be imparted to the nanomaterial, resulting in the artificial
yield of a plasmon-induced CD signal in the visible spectral region [19]. This was
investigated for a limited number of chiral biomolecules/nanomaterial complexes,
examples of which include DNA and peptide nanotubes decorated with gold/silver
nanoparticles [21-23]. Optical chirality of nanostructured systems is a remarkable
area of research that is receiving increasing interest as a result of its potential
74
Chapter 3
applications in optically active component devices for biomedical science,
environmental sensing and bioterrorism detection.
In view of the fact that plasmonic substrates offer huge electromagnetic fields for
SERS measurements, in this report fluorescent dye-labelled silver colloids have been
investigated to ascertain whether they can provide the same level of enhancement for
chiroptical spectroscopy, in this particular case ROA. Here, we have demonstrated
chiroptical behaviour of silver silica nanotags that possess an achiral plasmonic
nanostructure. Furthermore, we show that chirality was induced into the achiral
plasmonic surface of the substrate by binding to L- and D-enantiomeric analytes.
3.3 Experimental
Synthesis of EDTA-reduced silver colloid (AgEDTA)
Silica encapsulated silver nanotags were synthesised following the method outlined
in Graham et al. [6]. Nanoparticles of approximately 40 nm diameter were
synthesised according to methods described by Fabrikano et al. [24]. Briefly, sodium
hydroxide (0.4 M, 10 mL) was injected into a boiling 1 L solution of EDTA (1.62 x
10-4 M). Silver nitrate (0.026 M, 10 mL) was added to the boiling solution in 2.5 mL
aliquots. Following 15 minutes of continued heating, the solution was allowed to
equilibrate with room temperature. Stirring was maintained throughout.
Conjugation of dye to silver nanoparticles
Three separate replicates of nanoparticle-dye conjugates were synthesised for
analysis. The tri-functional benzotriazole dye shown in Figure S3.1 (refer to
supplementary information) was added to AgEDTA (1 mL, 1.00 x 10-10 M,) to a final
75
Chapter 3
concentration of 10-7 M. Samples were agitated prior to centrifugation (6000 rpm
(3500 g), 20 mins) and resuspension in 500 μL dH2O.
Silica encapsulation of silver – precursor conjugates
The nanoparticle (NP) – dye conjugates were prepared to 1 mL with the slow
addition of ethanol. Silica growth was initiated by the addition of triethylamine (10
µL, 1% v/v in ethanol) and the sequential addition of tetraethyl orthosilicate (TEOS)
(10 µL, 4% v/v in ethanol) over a three hour period until the final concentration of
TEOS was 5.4 mM.
Functionalisation of silica encapsulated nanotags
Based on the original NP concentration, the nanotags were functionalised with
approximately 4000x molar excess of the selected enantiomer, L- or D-, of the
analyte, ribose or tryptophan. This was achieved by reacting 1.21 molar equivalents
of triethoxysilylpropyl isocyanate with the required enantiomer; L- or D-ribose, or Lor D-tryptophan, in NaHCO3 buffer (0.1 M at pH 9) at 4°C overnight. The
“silanised” molecules were added to unwashed nanotags and agitated prior to
centrifugation (7000 rpm (4700 g), 20 mins) and resuspension (500 μL dH2O). Due
to the formation of dimers, trimers and possibly small aggregates it was difficult to
determine the actual nanotag:analyte molar ratio, therefore a 1:500 ratio is quoted
based on the initial nanoparticle concentration that had been used to prepare each
sample.
All Raman and ROA spectra were measured using a ChiralRaman spectrometer
(BioTools Inc., Jupiter FL, USA) configured in the backscattering geometry and
operating at a wavelength of 532 nm and spectral resolution of 7 cm-1. The laser
76
Chapter 3
power was set to 0.20 W, with laser power at the sample being approximately 0.10
W, and data acquisition times ranged from 5 min to 2 hour.
3.4 Results and Discussion
Silver silica nanotags were designed to possess maximum absorption corresponding
to their plasmon resonance at ~514 nm. The silver silica nanoshells have been
previously linked with benzotriazole dye molecules to construct SERRS nanoprobes
[6]. Certain dyes such as benzotriazole can act as both a resonant reporter for SERRS
and a precursor for the formation of silica shells around silver nanoparticles. They
have also been used to stabilise silver nanoparticle cores. To determine whether the
silica shell interferes with the ROA signal, spectra of silver nanoparticles bound to
the benzotriazole dye with and without the silica shell were recorded and are
presented in Figure 3.1. This serves as a control experiment since obtaining a true
ROA signal from an analyte can be challenging when either it is a resonant molecule
or silver nanoparticles are present [25,26]. The SERRS spectra of silver nanotags
both with and without the silica nanoshell coating are shown in Figure 3.1 (A) and
(B). These SERRS spectra demonstrate an identical profile with strong SERRS bands
attributed to the resonance effect of the dye. It is clear that the silica shell does not
interfere with the SERRS signal generated. The corresponding SERROA spectra of
silver nanotags with and without a silica nanoshell, Figure 3.1 (C) and (D), also show
similar spectral features being dominated by positive bands. The positive SERROA
bands resemble the parent SERRS bands with a lower signal to noise ratio which
indicates in both cases that the observed bands principally originate from interaction
of linear contaminants in the scattered circularly polarized light with the SERRS
bands rather than being a true measure of optical activity, therefore indicating that
77
Chapter 3
they are artefacts. This is a known problem in attempts to measure the SEROA effect
[26,27] and signifies non-optical activity of the silver silica nanotags as expected due
to the achiral structures of the dye and silver nanotags.
Figure 3.2 displays the SERRS and SERROA spectra of L- and D-ribose that were
attached covalently to silver silica nanotags. As is clearly shown, the SERRS spectra
of the two enantiomers of ribose have identical profiles to each other as well as the
SERRS spectra of the silver silica nanotag (Figure 3.1), since the resonance effect of
the dye dominates the spectra. The SERS spectra of L- and D-ribose previously
reported [25] are significantly different from the SERRS spectra of the same analytes
shown here since direct interaction of ribose molecules with the metal nanoparticle
surfaces occurred in the previous study which is not the case here. The SERROA
spectra presented in Figure 3.2 (B) and (C) which represent different replicates of
these experiments for ribose, exhibit a mirror image response and very different
features to their corresponding SERRS bands, unlike the already discussed spectra
obtained from the achiral silver silica nanotags. The corresponding SERRS spectra
for Figure 3.2 (C) were identical to 3.2 (B), therefore are not shown here. SERROA
spectra from the two enantiomers of ribose have both positive and negative bands,
the positions of which correlate well with each other. Strong peaks that appear above
1200 cm-1 in both cases can clearly be resolved from the background noise verifying
the detection of the optical activity of L- and D-ribose, and can be used to confirm
the stereochemistry of each analyte. For example, the strong +ve/-ve/+ve SERROA
bands at 1391, 1431 and 1450 cm-1 for L-ribose have corresponding –ve/+ve/-ve
bands at 1389, 1431 and 1450 cm-1 for the D-enantiomer. In addition the band above
1600 cm-1 shows a strong mirror image response at 1616 and 1623 cm-1 for L- and
D-ribose, respectively. The observed SERROA bands are probably due to the effect
78
Chapter 3
of ribose chirality on the plasmon at the silver nanotag surface as it modifies the
surface state. This chiral response then appears to be imprinted on the SERRS
spectrum of the bezotriazole dye, leading to the enantiomeric-sensitivity observed for
the SERRS bands.
In order to further evaluate these results, the SERRS and SERROA spectra of L- and
D-tryptophan were also measured and are presented in Figure 3.3. Once again, the
SERRS profiles for both enantiomers are identical to the aforementioned SERRS
profile of the benzotriazole dye-tagged silver colloids. The corresponding SERROA
spectra for the two enantiomers of this amino acid display mirror image responses, in
particular for the bands at 1317, 1347 and 1390 cm-1. The SEROA band intensities
observed for the two enantiomers of both ribose and tryptophan are similar. This
indicates that both molecules provide around the same level of chiral response in the
SERRS bands from the dyes despite having different molecular weights, chemical
structures and number of chiral centres. More mirror image bands are observed for
L- and D-ribose in contrast to L- and D-tryptophan within the region of 1400-1600
cm-1, mainly the two +ve doublet bands at 1531 and 1564 cm-1 for D-ribose.
The observation of mirror image SERROA bands originating from the SERRS
spectrum of the benzotriazole dye can be explained on the basis that chiral
molecules, in this study L- and D-ribose or tryptophan, once attached to a silver
surface possess the enantioselectivity required to break the symmetric environment
in the achiral metallic cluster, which is reported here by the SERRS response from
the dye molecules which are in close proximity to the surface plasmons. Thus,
induced dissymmetry in the surface plasmon interaction with the benzotriazole
molecules is responsible for the observed mirror image bands in the SERROA
spectra of L- and D-ribose and tryptophan. The SERROA phenomenon measured
79
Chapter 3
here is fundamentally different from that responsible for our previously reported
SEROA spectra of L- and D-ribose [25] and the SERROA spectra of two resonant
proteins myoglobin and cyctochrome c [28,29], since the direct interaction between
the chiral molecules and the surface plasmons from metal nanoparticles in those
cases was responsible for the enhancement of ROA signals. The SEROA spectral
details in those previous studies also originate directly from the analyte investigated,
whereas in the SERROA spectra presented in this study we observe a chiral influence
on the SERRS spectrum of the benzotriazole dye. Therefore, the molecular
dissymmetry has a direct effect on the observed SEROA spectra mainly due to the
field gradient generated by the plasmon resonance. Chirality observed in such
systems mainly originates from dipolar interactions with chiral molecules [19]. The
current observation of chiroptical behaviour monitored by our silver silica nanotags
is principally due to radiative electromagnetic coupling between metallic particles
and nanotag plasmons and the surrounding chiral molecules interacting over a long
range [30]. The chiral electromagnetic currents generated by the perturbation existent
in the presence of chiral chromophores have induced optical activity to the metal
nanoparticles.
The mechanism proposed as being responsible for these SERROA results is
comparable to that recently reported for a class of hybrid plasmonic nanomaterials
[30] exhibiting nanolithographed achiral gold crosses on the surface utilised with
indirect adsorption of chiral molecules. Abdulrahman et al. [30] demonstrated that a
chiral response was induced into the plasmonic resonance of the achiral
nanostructure using circular dichroism, so via measurement of electronic excitation.
In this paper we have observed a similar response but through monitoring vibrational
excitation. As proposed by Abdulrahman et al., this far-field effect that we have
80
Chapter 3
observed occurs due to the radiative electromagnetic interaction between a nonabsorbing isotropic chiral medium and a strongly absorbing metallic plasmon
resonance. In our case the plasmon resonance from the silver surface induces the
SERRS signal from the tethered benzotriazole dye molecules, with the chiral
analytes then interacting with that SERRS signal to generate an enantiomericallysensitive response.
Although a mirror image response has been obtained in this work for at least the
main SERROA bands for L- and D-ribose and L- and D-tryptophan, it is not possible
to assign these bands to the vibrational modes of ribose and tryptophan molecules.
This is, as explained above, because we are monitoring the induction of a chiral
influence on the SERRS spectrum of benzotriazole which acts here as a reporter
molecule. It is as yet unknown whether the differences observed in the details of the
SERROA band structures shown in Figures 3.2 and 3.3 are due to the still less than
perfect control of birefringent artefacts arising from distortions induced in the
scattered circularly polarized light by the surface plasmons, or whether they are
indicative of a difference in the interactions between the different chiral analytes and
the surface plasmons. However, this will be pursued in future work and the aim of
this current study is to confirm measurement of the optical activity of chiral analytes
through monitoring vibrational excitation in sensitive achiral nanostructures. These
spectra clearly show that SERROA bands of opposing sign are obtained from the two
enantiomers of chiral molecules and that these are signatures of the chiroptical nature
of the interaction between these analytes and the surface plasmons of these dyetagged nanoprobes.
81
Chapter 3
3.5 Conclusion
Molecular adsorption onto a nanostructured surface is of fundamental importance to
many processes involving separation, bio-sensing, surface processing, lubrication
and heterogeneous catalysis [31]. Therefore, the conformations of surface molecules
may have a pronounced effect on the physiochemical properties of biomolecules.
However, gaining information about surface organization at a molecular level is
often rather difficult and the use of spectroscopic sensing probes such as silver silica
nanotags has proven indeed to be very useful.
The results obtained here have demonstrated that achiral plasmonic substrates can be
used to detect optical activity through a chiral signature present in a SERRS
spectrum and that this is due to the novel mechanism of induced dissymmetry in a
plasmonic resonance monitored through vibrational excitation. Although much
further work needs to be performed to investigate this new phenomenon, and to
optimize it, this study already raises interesting possibilities for the enantioselective
detection of chiral molecules, and in particular biomolecules, hence extending the
scope of nanoplasmonic devices.
82
Chapter 3
3.6 References:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
H. Chon, S. Lee, S. W. Son, C. Oh, H. Choo, Journal of Analytical
Chemistry. 2009, 81, 3029–3034.
J. P. Camden, J. A. Dieringer, J. Zhao, R. P. Van Duyne, Accounts of
Chemical Research. 2008, 41, 1653–1661.
E. Temur, I. Boyaci, U. Tamer, H. Unsal, N. Aydogan, Analytical and
Bioanalytical Chemistry. 2010, 397, 1595–1604.
J. B. Jackson, S. L. Westcott, L. R. Hirsch, J. L. West, N. Halas, Journal of
Applied Physics Letter. 2003, 82, 257–259.
A. W. H. Lin, N. A. Lewinski, M. H. Lee, R. A. Drezek, Journal of
Nanoparticles Research. 2006, 8, 681–692.
L. Rocks, K. Faulds, D. Graham, Chemical Communication. 2011, 47, 44154417.
L. D. Barron, L. Hecht, Bimolecular conformational studies with vibrational
Raman optical activity. In Biomolecular Spectroscopy; Clark, R. J. H.,
Hester, R. E., Editors. Wiley: Chichester, 1993; Part B, pp 235.
T. B. Freedman, L. A. Nafie, T. A. Keiderling, Biopolymers. 1995, 37, 265279.
L. A. Nafie, Applied Spectroscopy. 1996, 50, 14A-26A.
L. D. Barron, L. Hecht, Vibrational Raman optical activity: From
fundamentals to biochemical applications. In Circular Dichroism, Principles
and Applications; Nakanishi, K., Berova, N., Woody, R. W., Eds.; VCH
Publishers: New York, 1994; pp 179.
L. D. Barron, L.Hecht. I. H. McColl, E. W. Blanch, Molecular Physics. 2004,
102, 731-734.
D. Van Duyne, L. Jeanmaire, Journal of Elecroanalytical Chemistry. 1977,
84, 1-20.
K. Kneipp, Y. Wang, H. Kneipp, L. T. Perelman, I. Itzkan, R. R. Dasari, M.
S. Feld, Physical Review Letters. 1997, 78, 1667-1670.
E. Koglin, H. H. Lewinsky, J. M. Sequaris, Surface Science. 1985, 158, 370380.
J. K. Gansel, M. Thiel, M. S. Rill, M. Decker, K. Bade, V. Saile, G. von
Freymann, S. Linden, M. Wegener, Science. 2009, 325, 1513-1515.
A. S. Schwanecke, A. Krasavin, D. M. Bagnall, A. Potts, A. V. Zayats, N. I.
Zheludev, Physical Review Letters. 2003, 91, 247404-247409.
M. Kuwata-Gonokami, N. Saito, Y. Ino, M.Kauranen, K. Jefimovs, T.
Vallius, J. Turunen, Y. Svirko, Physical Review Letters. 2005, 95, 227401227404.
E. Hendry, T. Carpy, J. Johnston, M. Popland, R. Mikhaylovskiy, A. J.
Lapthorn, S. M. Kelly, L. D. Barron, N. Gadegaard, M. Kadodwala, Nature
Nanotechnology. 2010, 5, 783-787.
J. M. Slocik, A. O. Govorov, R. R. Naik, Nano Letters. 2011, 11, 701–705.
G. Shemer, O. Kruchevski,G. Markovich, T. Molotsky, I. Lubitz, A. B.
Kotlyar, Journal of the American Chemistry Society. 2006, 128, 11006–
11007.
J. George, G. Thomas, Journal of the American Chemistry Society. 2010,
132, 2502–2503.
83
Chapter 3
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
N. Shukla, M. A. Bartel, A. J. Gellman, Journal of the American Chemistry
Society. 2010, 132, 8580–8575.
P. Rezanka, A. Zaruba, V. Kral, Colloids Surface A: Physicochemical and
Engineering Aspects. 2011, 374, 77-83.
V. Fabrikanos, S. Athanassiou, K. Leiser, Zeitschrift für Naturforschung B: A
Journal of Chemical Sciences. 1963, 186, 612-615.
S.Ostovar Pour, S.E.J. Bell, E. W. Blanch, Chemical Communications. 2011,
37, 4754-4756.
S. Abdali, E. W. Blanch, Chemical Society Reviews. 2008, 19, 980-992.
P. G. Etchegoin, C. Galloway, E.C. Le Ru, Physical Chemistry Chemical
Physics. 2006, 8, 2624-2628.
S. Abdali, C. Johannessen, J. Nygaard, T. Nørbygaard, Journal of Physics:
Condensed Matter. 2007, 19, 285205-285212.
C. Johannessen, P. C. White, S. Abdali, Journal of Physical Chemistry
A, 2007. 111, 7771–7776
N. A. Abdulrahman, Z. Fan, T. Tonooka, S. M. Kelly, N. Gadegaard, E.
Hendry, A. O. Govorov, M. Kadodwala, Nano Letters. 2012, 12, 977-983.
S. K. Basiruddin, A. Saha, N. Pradhan, N. R. Jana, Journal of Physical
Chemistry C. 2010, 114, 11009-11017.
.
84
Chapter 3
A
8
IR+ IL
5x10
B
8
5x10
C
IR- IL
4
5x10
0.0
D
4
5x10
0.0
200
400
600
800
1000
1200
-1
1400
1600
1800
Wavenumber (cm )
Figure 3.1: SERRS spectra of nanotag (tri-functional benzotriazole dye) without
silica coated silver colloids (A), with silica coated silver colloids (B) and SERROA
spectra of A (C) and B (D), data collection time of 35 min and laser power at source
0.20 W.
85
Chapter 3
A
IR+ IL
D-ribose
L-ribose
8
7x10
B
1616
1391
5
2.5x10
IR- IL
1431
1433
C
1389
1623
5
2.5x10
200
400
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure 3.2: SERRS spectra of D- and L-ribose that are attached to silver silica
nanotags (A), SERROA of D- and L-ribose replicates 1 (B) and batch 2 (C), data
collection time of 35 min and laser power at source 0.20 W.
86
Chapter 3
A
D-tryptophan
L-tryptophan
9
IR+ IL
1x10
1317
B
1347
1393
1637
4
IR- IL
4x10
1320
1347
1390
200
400
600
800
1000
1200
1400
1630
1600
-1
Wavenumber (cm )
Figure 3.3: SERRS spectra of D- and L-tryptophan that are attached to silver silica
nanotags (A), SERROA spectra of D- and L-tryptophan (B), data collection time of
35 min and laser power at source 0.20 W.
87
1800
Chapter 3
3.7 Supplementary Information
Figure S3.1: Structure of tri-functional benzotriazole dye.
88
Chapter 4
Chapter 4
Phosphorylation Detection and
Characterization in
Ribonucleotides Using Raman and Raman
Optical Activity (ROA) Spectroscopies
89
Chapter 4
4.0 Declaration
This chapter consists of one published full paper: S. Ostovar Pour, E. W. Blanch, Applied
Spectroscopy. 2012, 289-293.
The manuscripts have been incorporated in a format identical to that for journal submission,
except for minor adjustments to incorporate them into this thesis. As first author on this
publication, I carried out all of the associated experimental and spectroscopic analysis.
90
Chapter 4
Phosphorylation Detection and Characterization in
Ribonucleotides Using Raman and Raman Optical Activity
(ROA) Spectroscopies
Seideh Ostovar Pour and Ewan W. Blanch*
Received 8 September 2011; accepted 23 November 2011
Manchester Interdisciplinary Biocentre and Faculty of Life Sciences, The University of Manchester,
131 Princess Street, Manchester M1 7DN, United Kingdom
* Author to whom correspondence should be sent. E-mail: [email protected].
4.1 Abstract
Raman and Raman optical activity (ROA) spectra are presented for adenosine and
seven of its derivative ribonucleotides. Both of these spectroscopic techniques are
shown to be sensitive to the site and degree of phosphorylation, with a considerable
number of marker bands being identified for these ribonucleotides. ROA spectra are
shown to provide the most sensitive diagnostic tool for phosphorylation
characterization and quantification.
Index
Headings:
Raman
spectroscopy;
Raman
optical
activity;
ROA;
Ribonucleotides; Phosphorylation detection.
4.2 Introduction
Protein phosphorylation is an important regulatory mechanism that plays a role in a
wide range of cellular processes, causing or preventing the mechanisms of diseases
such as cancer and diabetes [1–3]. Although phosphorylation pathways have been
studied widely, the detection and quantification of phosphorylated species in
complex mixtures has proven difficult due to the limited availability of suitable
91
Chapter 4
methods. The methods that are available for characterization of phosphorylated
biomolecules include specific antibody staining, two-dimensional (2D) gel
electrophoresis with specific fluorescent probe labeling, and, most recently, mass
spectrometry [3–12]. Although mass spectrometry is a very accurate technique
compared to other available methods, it is not intrinsically sensitive to the site of
phosphorylation. Fluorimetric methods have in part replaced these problems by
attaching fluorescent markers to specific locations of the active molecule. This
technique can provide high spatial and temporal resolution; however, the addition of
substantial structural elements such as fluorescent markers will affect the kinetic and
thermodynamic behavior of biomolecules and may even influence the outcome of a
complex enzymatic process [13,14]. Due to the limitations inherent to the above
techniques, there is a need to develop a structurally sensitive and label-free method
for characterizing the degree and location of phosphorylation.
In recent years, Raman spectroscopy has been widely applied in different fields of
science for recognition of chemical and biological samples as well as for the
classification of molecular structure [15–17]. Another promising technique that is
even more sensitive to structural changes than Raman spectroscopy is Raman optical
activity (ROA), which measures a small difference in the intensity of vibrational
Raman scattering from chiral molecules in right and left circularly polarized light
[18,19]. As a result of its sensitivity to chirality, ROA is a powerful probe of the
structure and behavior of biomolecules in aqueous solution [20]. Recently, an
investigation of protein phosphorylation using Raman and ROA spectroscopies has
highlighted the sensitivity of these methods when studying phosphorylation patterns
in both amino acids and proteins [21].
92
Chapter 4
We present here a detection platform based on Raman and ROA spectroscopies for
sensitive detection of phosphate groups in adenosine ribonucleotides with positional
specificity. A considerable benefit in using Raman and ROA lies in their ability to
provide information about molecular structure with label-free sensitivity, making
them promising candidates for detecting, both qualitatively and quantitatively,
phosphorylation. Ribonucleotides present a diverse range of phosphorylated variants
and are known to be amenable to spectroscopic study, and thus provide a suitable
model system for testing sensitivity to phosphorylation. To our knowledge, this is the
first study to present combined Raman and ROA spectra of adenosine
ribonucleotides containing one or more phosphate groups at different positions on the
ribose ring, obtained under the same experimental conditions.
4.3 Experimental
Samples of adenosine, adenosine 2’,3’-cyclic monophosphate sodium salt
[A(2,3)MP], adenosine 3’,5’-cyclic monophosphate sodium salt monohydrate
[A(3,5)MP], adenosine 2’-monophosphate [A(2)MP], adenosine 3’-monophosphate
[A(3)MP], adenosine 5’-monophosphate [AMP], adenosine 5’-diphosphate [ADP],
and adenosine 5’-triphosphate [ATP] were obtained from Sigma-Aldrich Ltd (Poole,
Dorset, UK) and used without further purification. For Raman and ROA
measurements, each sample was prepared by dissolving dry powder in distilled
deionized H2O at a concentration of ~100 mg/mL, which was then transferred to a
quartz microflourescence cell. Each cell was then microcentrifuged at 3000 rpm
(1000 g) for 5 min in order to minimize dust particles prior to spectral measurements.
The Raman and ROA measurements were performed using a ChiralRaman
spectrometer (BioTools Inc., Jupiter, FL) configured in the backscattering geometry,
93
Chapter 4
with an excitation wavelength of 532 nm and spectral resolution of 7 cm-1. The
Raman and ROA experimental conditions were as follows: power at the laser head
was 0.60 W, data collection times of 4–12 h. A baseline subtraction was performed
for each spectrum prior to analysis using MATLAB 7.6 software by employing the
inbuilt bioinformatics toolbox.
4.4 Results and Discussion
The Raman spectra of the seven adenosine ribonucleotides, along with that of
adenosine, are presented in Figure 4.1. In Table 4.1, the Raman assignments
associated with these molecules are summarized. The results are in agreement with
previous studies on adenosine, AMP, A(3,5)MP, ADP, and ATP molecules [22–24].
In general, the spectral features of all samples except ATP are similar in terms of
band intensity and profile, but several differences are observed. For example, the
band observed around 754 cm-1, which is assigned to a ring breathing mode in
adenosine molecules [22], shifts to a lower wavenumber with the addition of a
phosphate group. This band remains at 731 cm-1 for both AMP and ADP, whereas for
ATP the band shifts lower to 693 cm-1. As the position of a phosphate group changes
along the ribose ring from C2 to C5, the peak becomes more intense as well as
shifting to a lower wavenumber.
The Raman bands that are associated with phosphate group vibrations principally
occur in the region between 900 and 1200 cm-1 [23,24]. The peaks observed in this
region indicate that Raman spectra are responsive to phosphate group numbers and
position around the ribose ring. Therefore, the marker band for each analyte under
study can be found in this region. There are weak/no bands present in this region for
adenosine, so that any intense band within this region for any ribonucleotide studied
94
Chapter 4
here can be assigned to a phosphate group vibrational mode. Bands at 1083 and 1113
cm-1, assigned to the phosphate group vibration, are noted in the Raman spectra of
AMP and ADP, respectively, and so serve as the sole marker bands for
distinguishing between these two analytes as the rest of their Raman spectra are
virtually identical. The most intense band arising from this phosphate group vibration
can be observed for ATP at 1092 cm-1, presumably because it has the highest number
of phosphates [23,24]. Previous researchers [25–33] pointed out that the band near
~1100 cm-1 in the spectra for polyribonucleotides is due to the symmetric stretching
vibration of the PO32- moiety and shows similar relative intensities to our results. As
the phosphate group moves from C5 to C3 around the ribose ring, the spectral
appearance changes and the peak at 1083 cm-1 shifts to lower Wavenumber at 946
cm-1.
The signature band of the phosphate group vibration for A(2)MP molecule appears at
1048 cm-1 while this band is upshifted to 1081 cm-1 for A(2,3)MP. The most
significant difference for A(3)MP and A(3,5)MP appears in the bands at 946 and
1056 cm-1, respectively, which are assigned to phosphate vibrations, while other
bands for these two molecules are almost identical.
The Raman spectra of all analytes investigated in this work are dominated by the
adenine ring breathing C–H, C–N, and C=C stretching modes [34], which appear as a
triplet of bands from 1275 to 1392 cm-1, apart from the case of ATP, which shows a
doublet of peaks at 1295 and 1377 cm-1. The relative intensity of these triplet bands
varies for adenosine, ATP, A(2)MP, and A(2,3)MP but not for AMP, A(3)MP,
A(3,5)MP, and ADP, for which the positions and intensities of these bands are
identical. Raman bands observed at 1499 and 1524 cm-1 for adenosine are associated
with ribose CH2 bending and C–C and C–N stretching coordinates [33], while these
95
Chapter 4
bands are downshifted to 1487 and 1514 cm-1 for AMP and ADP. The same bands
shift to 1477 and 1529 cm-1 for ATP, which illustrates the significant changes in the
vibrational modes upon addition of the third phosphate group. The Raman spectra for
A(2)MP, A(2,3)MP, and A(3)MP show identical bands at 1455 and 1480 cm-1 for
these vibrational modes, which shift by 3 to 5 cm-1 for A(3,5)MP. There is a band
observed at 1584 cm-1 in the Raman spectra of AMP and ADP that is assigned to a
combined C=C and C=N stretching mode that appears at 1582 cm-1 for ATP and
1550 cm-1 for A(2)MP, A(2,3)MP, A(3)MP, and A(3,5)MP [23,35].
In addition to Raman spectra measurement, surface enhanced Raman spectroscopy
(SERS) was also performed (see Supplemental Material, available online). It was
found that SERS generated very similar spectra for A(2,3)MP, A(3)MP, and
A(3,5)MP, suggesting that all of these analytes interact with the surface in a similar
manner where the adenine moiety interacts most directly with the metal surface
while the phosphate group is located further away, hence limiting the structural
sensitivity of this method towards phosphate position.
Figure 4.2 presents the ROA spectra of the seven adenosine ribonucleotides and that
of adenosine in aqueous solution. The ROA band assignments for AMP, ADP, ATP,
A(2)MP, A(2,3)MP, A(3)MP, and A(3,5)MP are listed in Table 4.2. In general, all
spectra show very distinct profiles for each individual species and a number of
marker bands are obvious. Notable differences are apparent between the ROA bands
in the region of 600–800 cm-1, which involves the adenosine backbone vibration. The
ROA spectrum of adenosine gives a single band at 793 cm-1 that is due to the ribose
ring breathing mode [26–28].
The +ve/-ve couplet band at 761/782 cm-1 for AMP and ADP corresponds to the
couplet at 723/749 cm-1 for A(2)MP, presumably because the adenosine backbone
96
Chapter 4
appears to be in a more stereochemically restricted environment and has less freedom
to rotate, since the ROA spectra of ATP, A(2,3)MP, A(3)MP, and A(3,5)MP show
only a weaker single band at 738, 744, 735, or 732 cm-1, respectively. From the ROA
spectra of all seven ribonucleotides in this region it is noted that more bands are
detected in contrast to the case of adenosine; therefore, the addition of phosphate to
the ribose ring causes considerable change to its conformation and vibrational
dynamics. The region of 900–1200 cm-1 includes bands corresponding to the
vibrations of PO32-, the ribose moiety, C–O, and C–C in the ribose ring [26–28]. The
intense band at 925 cm-1 for adenosine that is mainly assigned to a combined C–O
and C–C stretching vibration in the ribose ring is found at the same position for AMP
and ADP but shifts to 888 cm-1 for ATP. The ROA spectra of AMP and ADP in this
region are very similar except for appearance of the weak bands at 1180 and 1050
cm-1 that are associated with the PO32- symmetric vibration mode [26–28].
The absence of bands at 1075 and 1127 cm-1 in the ROA spectrum of adenosine, but
their presence in the spectrum for ATP, indicates that these bands arise from
phosphate groups. As the phosphate group moves from C5 to C2 and C3 positions
within these nucleotides, more intense ROA bands are detected. The +ve/-ve/+ve
triplet at 879/905/931 cm-1 for A(2)MP, which appears at 882/914/936 cm-1 for
A(3)MP, evidently is sensitive to the location of the phosphate group on the ribose
ring. The -ve/+ve couplet at 1064/1110 cm-1 corresponding to the phosphate group
vibration in the ROA spectrum of A(2,3)MP undergoes an inversion of sign and
shifts to 1078/1113 cm-1 for A(3,5)MP. This striking change in spectral signature
from the phosphate group when attached to two different positions on the ribose ring,
so giving rise to a more rigid and chiral structure upon cyclization, indicates that the
signals coming from phosphate are stronger for A(2,3)MP and A(3,5)MP. We note
97
Chapter 4
the absence of this couplet in the ROA spectra of AMP, ADP, ATP, A(3)MP, and
A(2)MP for which the phosphate groups have a more flexible conformation.
The higher wavenumber bands are the most intense and mainly correspond to purine
ring stretching modes, for which adenosine gives an intense peak at 1361 cm-1. Upon
phosphorylation of the ribose ring of adenosine more changes are noted within this
region that distinctly indicate the effect of phosphorylation upon the purine ring. The
set of +ve/-ve/+ve bands at 1299/1348/1381 cm-1 for AMP shift slightly to
1299/1345/1376 cm-1 for ADP, with these bands arising from combined C–C, C=N,
and C–H stretching modes. The relative intensities of these triplet bands for AMP
vary significantly in comparison to those measured for ADP. However, the higher
peak in the triplet changes sign and shifts downward by 10 cm-1 for ATP, which
indicates that the addition of the third phosphate has a large and direct effect on the
purine ring’s vibrations. As the phosphate group transfers from C2,3 to C3,5
linkages, more intense bands become apparent. The +ve/-ve/+ve triplet peaks at
1312/1351/1384 cm-1 for A(2,3)MP correspond to the +ve/-ve/+ve triplet bands at
1309/1345/1386 cm-1 for A(3,5)MP. The +ve/-ve couplet at 1207/1265 cm-1 for
A(3,5)MP is absent for A(2,3)MP; however, this couplet appears for A(2)MP at
1210/ 1268 cm-1. The intense band at 1422 cm-1 in the ROA spectrum of AMP,
assigned to a combined C–C and C–N stretching mode, is downshifted to 1417 and
1419 cm-1 for ATP and ADP, respectively. No band of similar relative intensity can
be observed for the other analytes with fewer phosphate groups or where the
phosphate group is moved to a different position of the ribose ring, that is from the
C5 to C2 or C3 position.
The sharp +ve/-ve couplet at 1509/1560 cm-1, from a combined C–C, C–N, C=C, and
C=N stretching mode, for ATP is reduced in relative intensity and is accompanied by
98
Chapter 4
inversion of sign in the cases of AMP and ADP. This couplet also shifts to
1512/1580 and 1489/1585 cm-1 for A(2)MP and A(3)MP, respectively. A ROA
feature above 1600 cm-1 that is due to C=N, C–C, and C–N stretching is only
apparent as an intense band for AMP and ADP at 1626 and 1621 cm-1, respectively,
while only a very weak or no ROA band is observed in this region for ATP, A(2)MP,
A(2,3)MP, A(3)MP, and A(3,5)MP.
4.5 Conclusion
Raman and ROA spectra of adenosine ribonucleotides in aqueous solution have been
investigated to study the conformational changes of these analytes upon
phosphorylation. It was found that Raman spectra were sensitive to the number of
phosphate groups but not to their position around the ribose ring. The general
features of the ROA spectra vary dramatically among these samples and provide a
fingerprint sensitive to both the number and position of phosphate groups. Although
the ROA spectra of AMP and ADP are very similar, there are several bands that can
be used as markers for differentiating between them. This is notable as only an
achiral phosphate group is added and no direct structural change is made to a chiral
center. Therefore, several corresponding ROA bands for AMP and ADP show
apparent sensitivity to an indirect change upon a chiral center. Upon cyclization of
the phosphate group in A(2,3)MP and A(3,5)MP, which appears to lead to a more
rigid structure in each case, unique ROA signatures are observed, which also indicate
that signals for the phosphorus groups in these two cyclic ribonucleotides are
opposite in sign and reflect opposite stereochemistry in the environment of the
phosphorus atom in each case.
99
Chapter 4
The ROA technique is more insightful for investigating phosphorylation of
adenosine ribonucleotides because it provides more informative details on the
phosphate groups. ROA signals associated with phosphate groups evidently can
distinguish between achiral and chiral environments of these phosphate groups. By
contrast, Raman spectra are, of course, blind to stereochemistry. The ROA spectra
reported here have demonstrated that it is feasible to distinguish between adenosine
ribonucleotides that vary in the occurrence and position of phosphorylation even
when the ribonucleotides have very similar structure. An earlier study from our
group has reported Raman and ROA spectra for uridine and its mono and
triphosphates, which showed similar strong marker band differences as a function of
phosphorylation [36]. We therefore expect that all other nucleotides will possess
unique phosporylation-sensitive marker bands, with ROA spectra being particularly
informative.
In conclusion, we have shown that Raman and ROA spectra are very sensitive to
structural differences between these adenosine ribonucleotides and specifically to the
site and degree of their phosphorylation. Furthermore, of these two spectroscopic
techniques ROA provides the most sensitive diagnostic of phosphorylation.
4.6 Acknowledgement
The authors thank Dr. Christian Johannessen and Professor Laurence Barron for
helpful discussions. Additionally, the authors would like to thank the Royal Society
of Chemistry and Analytical Trust Fund for a studentship to S.O.
100
Chapter 4
4.7 References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
P. Cohen, Trends in Biochemical sciences. 2000, 25, 596-601.
T. H. Steinberg, K. R. Agnew, K. R. Gee, W. Leung, T. Goodman, B.
Schulenberg, J. Hendrickson, J. M. Beechem, R. P. Haugland, W. F. Patton,
Proteomics. 2003, 3, 1128-1144.
G. Guy, R. Philip, Y. Tan, Electrophoresis. 1994, 15, 417-440.
V. Sockic, M. Gorlach, S. Poznanovic, F. D. Boehmer, J. GodovacZimmermann, Biochemistry. 1999, 38, 1757-1764.
H. Kaufmann, J. E. Fussenegger, Proteomics. 2001, 1, 194-199.
M. Mann, O. N. Jensen, Nature. 2003, 21, 255-261.
R. R. E. Schweppe, C. E. Haydon, T. S. Lewis, K. A. Resing, N. G. Ahn,
Accounts of Chemical Research. 2003, 36, 453-461.
R. D. Aebersold, R. Goodlet, Chemical Reviews. 2001, 101, 269-295.
S. P. Gygi, B. Rist, S. A. Gerber, F. Turecek, M. H. Gelb, R. Aebersold,
Nature Biotechnology. 1999, 17, 994-999.
K. Zhang, H. Tang, L. Haung, J. W. Blankenship, P. R. Jones, F. Xiang, P.
M. Yau, A. L. Burlingame, Biochemistry. 2002, 306, 259-269.
A. G. Marshall, C. L. Hendrickson, G. S. Jackson, Mass Spectrometry
Reviews. 1998, 17, 1-35.
S. E. Martin, J. Shabanowits, D. F. Hunt, J. A. Marol, Analytical Chemistry.
2000, 72, 4266-4274.
R. J. Beynon, M. Pratt, Molecular & Cellular Proteomics. 2005, 4, 857-872.
S. E. Ong, M. Mann, Nature Chemical Biology. 2005, 1, 252-262.
E. B. Hanlon, R. Manoharan, T. W. Koo, K. E. Shafer, J. T. Motz, M.
Fitzmaurices, J. R. Kramer, I. Itzkan, M. S. Feld, Physical Medical Biology.
2000, 45, R1-R59.
D. Zhang, Y. Xie, M. F. Mrozek, C. Ortiz, V. J. Davisson, D. Ben- Amotz,
Analytical Chemistry. 2003, 75, 5703-5709.
K. Kneipp, H. Kneipp, V. B. Kartha, R. Manoharan, G. Deinum, I. Itzkan, R.
R. Dasari, M. S. Feld, Physical Review E: Statistical Physics. 1998, 57,
R6281-R6284.
P. W. Atkins, L. D. Barron, Molecular Physics. 1969, 16, 453-466.
L. D. Barron, M. P. Bogaard, A. D. Buckingham, Journal of the American
Chemical Society. 1973, 95, 603-605.
L. D. Barron, L. Hecht, E. W. Blanch, A. F. Bell, Progress in Biophysics and
Molecular Biology. 2000, 73, 1-49.
L. Ashton, C. Johannessen, R. Goodacre, Analytical Chemistry. 2011, 83,
7978-7983.
A. Lee, W. Anderson, R. Smith, H. Griffey, V. Mohan, Journal of Raman
Spectroscopy. 2001, 32, 795-802.
L. Rimai, T. Cole, J. L. Parsons, J. T. Hickmott, E. B. Carew, Biophysical
Journal. 1969, 9, 320-329.
R. F. Steiner, R. F. Beers, Polynucleotides: Natural and Synthetic Nucleic
Acids , Elsevier, Amsterdam, 1961, viii, pp. 305.
Y. Liao, Y. Meng, H. Lei, Y. Wang, Chinese Optics Letter. 2008, 6, 61-63.
A. F. Bell, L. Hecht, L. D. Barron, Journal of the American Chemical
Society. 1997, 116, 6006-6013.
101
Chapter 4
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
A. F. Bell, L. Hecht, L. D. Barron, Journal of Raman Spectroscopy. 1999, 30,
651-656.
A. F. Bell, L. Hecht, L. D. Barron, Chemistry A European Journal. 1997, 3,
1292-1298.
A. F. Bell, L. Hecht, L. D. Barron, Journal of Chemistry Society: Faraday
Transactions. 1997, 93, 553-562.
A. F. Bell, L. Hecht, L. D. Barron, Journal of the American Chemical
Society. 1998, 120, 5820-5821.
A. F. Bell, L. Hecht, L. D. Barron, Biospectroscopy. 1998, 4, 107-111.
G. Felsenfeld, D. R. Davies, A. Rich, Journal of the American Chemical
Society. 1957, 79, 2023-2024.
M. Mathlouthi, A. M. Seuvre, J. L. Koenig, Carbohydrate Research. 1984,
131, 1-15.
T. Ueda, K. Ushizawa, M. Tsuboi, Spectrochimica Acta. 1994, 50, 16611674.
E. Koglin, J. M. Sequaris, P. Valenta, Journal of Molecular Structure. 1980,
60, 421-425.
A. J. Hobro, PhD thesis, Structural Investigation of RNA through the
Application of Raman, Raman Optical Activity and Surface Enhanced
Spectroscopies, 2008, Manchester, University of Manchester, UK.
102
Chapter 4
Adenosine
AMP
Raman Intensity
ADP
ATP
A(2)MP
A(2,3)MP
A(3)MP
A(3,5)MP
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure 4.1: Raman spectra of adenosine (pH= 12.95), AMP (pH= 6.02), ADP (pH=
5.18), ATP (pH= 4.20), A(2)MP (pH= 3.13), A(2,3)MP (pH=5.54), A(3)MP (pH=
8.10) and A(3,5)MP (pH=6.67) in solution. The concentration for each sample was
100 mg/ml and laser power was 0.6 W at the sample.
103
Chapter 4
Adenosine
AMP
ROA Intensity
ADP
ATP
A(2)MP
A(2,3)MP
A(3)MP
A(3,5)MP
600
800
1000
1200
1400
-1
1600
1800
Wavenumber (cm )
Figure 4.2: ROA spectra of adenosine (pH= 12.95), AMP (pH= 6.02), ADP (pH=
5.18), ATP (pH= 4.20), A(2)MP (pH= 3.13), A(2,3)MP (pH=5.54), A(3)MP (pH=
8.10) and A(3,5)MP (pH=6.67) in solution. The concentration for each sample was
100 mg/ml and laser power at the sample was 0.6 W.
104
Chapter 4
Table 4.1: Raman band assignments of adenosine, AMP, ADP, ATP, A(2)MP,
A(2,3)MP, A(3)MP and A(3,5)MP [23-32].
Assignment
Adenosine
AMP
ADP
ATP
A(2)MP
A(2,3)MP
A(3)MP
A(3,5)MP
Backbone A bend
Ring breathing mode
OH bend
(C-O),(C-C)(R) str
“
“
“
Asym PO32- bend
“
Sym PO32- str
Sym PO32- str
Sym PO32- str
Sym PO32- str
Sym PO32- str
Phosphate vibration
Asym PO32- str
C-N, C-H str
Pyr ring
In-plane ring vibrations of adenine
residue
Imidazole ring str
C-N, C=C str
C-N, C-H
Ribose CH2 bend
C2H,N-C,C-H str
C-C, C-N (Imidazole ring)
C=C str, C=N str
„
C-C, C-N str
754
1230
1277
1326
1354
643
731
821
855
886
920
1007
1083
1180
1222
1259
1311
1342
640
731
827
855
884
1012
1045
1113s
1180
1225
1256
1311
1342
637
693
789
815
858
985
1092
1127s
1295
-
699
769
809
841
881
979
1012
1053
1100
1183
1222
1277
1306
-
699
772
850
881
912
976
1048
1081
1143
1183
1222
1275
1306
1347
696
766
847
881
946
976
1145
1183
1220
1275
1306
1347
631
693
743
807
830
881
918
976
1056
1145
1185
1220
1275
1306
1344
1392
1442
1475
1499
1524
1594
1660
1382
1427
1465
1487
1514
1584
1651
1382
1425
1465
1487
1514
1584
1653
1377
1430
1447
1477
1529
1582
1667
1347
1427
1455
1480
1550
1615
1397
1430
1455
1480
1550
1620
1397
1427
1455
1480
1550
1617
1392
1450
1477
1550
1608
*str= stretching, bend= bending, sym= symmetric, A-sym= anti-symetric, A=adenosine, pyr= pyrimidine.
105
Chapter 4
Table 4.2: ROA band assignments of adenosine, AMP, ADP, ATP,
A(3,5)MP,A(2,3)MP,A(2)MP and A(3)MP [25-32].
Assignment
Adenosine
AMP
ADP
ATP
A(2)MP
A(2,3)MP
A(3)MP
A(3,5)MP
Backbone A bending
“
Ring breathing mode
OH bending
Phosphate vib
“
+793
-
+694
+741
-758
+779
-
+697
+744
-761
+782
+842
+652
+702
+738
-
+679
-723
+749
-796
+842
+643
+708
+744
-773
+808
-848
+667
+735
-799
-
+673
+732
-779
+839
-
-885
-
+882
-
+888
-
+879
-905
+882
-911
+882
-914
-871
+914
+925
-
+922
-
+925
+992
+946
-984
+931
-
+942
+975
+936
+984
-
-
-1121
+1050
-
+1023
+1075
+1127
+1006
-1094
-1121
+1009
-1064
+1110
+1012
-1064
-
+1025
+1078
-1113
-
+1180
+1215
+1210
-
+1154
+1244
+1183
+1210
+1249
+1170
+1207
+1296
+1338
-1260
+1299
+1348
-1257
+1299
+1345
+1289
+1330
-1268
+1283
+1320
-1348
+1312
-1351
-1296
+1320
-
-1265
+1309
+1354
+1361
+1386
-1404
+1474
+1530
-1628
+1381
+1422
-1512
+1553
+1626
+1376
+1419
+1464
+1551
+1621
-1366
+1417
+1475
+1509
-1560
+1590
+1623
+1394
+1474
+1512
-1580
-
+1384
+1434
+1454
+1489
+1556
-1585
-
-1361
+1386
+1427
+1489
-1585
+1604
+1386
+1422
-1462
+1494
-1590
-
asym O-P-O vib
Adenosine base
Ribose moiety and phosphate
vib
Ribose moiety
Sym str PO32 vib
“
“
PO32 vib
Phosphate vib
Couples base and sugar ring
vib
C-N str
Pyr ring
C-N,C-H str
Ribose moiety and phosphate
group
C-N, C=C str
“
C-N, C-H str
C-C-H,N-C,C-H str
C-C, C-N (Imidazole ring)
C=C str,C=N str
“
C=N, C-C, C-N str
*str= stretching, bend= bending, sym= symmetric, A-sym= anti-symetric, A=adenosine, pyr= pyrimidin
106
Chapter 4
4.8 Supplementary Information
4.8.1 Colloid Preparation
Citrate-reduced colloids were prepared by reduction of silver nitrate by citrate ions
(Lee and Meisel method) [1], where 0.094 g of AgNO3 was dissolved in 500 ml of
H2O and heated to boiling point, then 10 ml of 1% trisodium citrate solution was
added drop wise to the mixture. Heating was continued for another hour with
constant stirring and then the solution was allowed to cool at room temperature.
Approximately 300 ml of green-grey solution was obtained at 0.5 M concentration.
All glassware used to prepare the colloids was washed prior to use with aqua regia
followed by gentle scrubbing with a 2% Helmanex solution and then thoroughly
rinsed. Distilled water was used for all colloid and sample preparations.
Sample Preparation for SERS Measurements
The sample, aggregating agent salt (K2SO4) and colloid concentration details are
given in the figure legend below. All samples were prepared to 1 ml, the sample
being left to sit for 60 min in order to obtain maximum SERS enhancement.
4.8.2 Surface enhanced Raman spectroscopy (SERS)
SERS spectra of adenosine and its ribonucleotides in the presence of silver
nanoparticles in an aqueous environment are shown in Figure S4.1. In Table S4.1 we
present SERS band assignments for the seven adenosine ribonucleotides studied.
Under the same experimental conditions, the SERS spectra of AMP, A(2)MP,
A(2,3)MP, A(3)MP and A(3,5)MP are all seen to be very similar to each other. This
postulates the possibility that all AMP derivative analytes interact with the metal
surface through the same sites of the adenosine moiety. The phosphate group for
107
Chapter 4
each adenosine monophosphate appears to be located too far away from the metal
surface to show any sensitivity to position. Thus, it is more difficult to differentiate
between phosphate group vibrational modes in these SERS spectra.
Previous studies of adenosine, AMP, ADP and ATP at metal surfaces, mainly gold
and silver, have also proposed that the binding site occur through adenine ring and
the adenine ring vibration dominate the spectra [2-6]. This interpretation is supported
by the fact that there are no obvious SERS bands from 1000-1100 cm-1, the region
corresponding to bands arising from distinctive phosphate group vibrations in the
Raman spectra of these species.
It was found that Raman and SERS spectra were sensitive to the number of
phosphate groups but not to their position around the ribose ring. Previous
publications for detection of phosphorylation using SERS have shown that there are
variations on results as they are very dependent on experimental conditions such as
aggregating salt and analyte concentration [7, 8]. There is also a time dependent
decline in colloidal based SERS signals during the measurement. Although it is
possible to differentiate between individual adenosine ribonucleotides, the phosphate
region barely shows reliably enhanced SERS bands [7,8].
108
Chapter 4
4.8.3 References
1.
2.
3.
4.
5.
6.
7.
8.
P.C. Lee, D. Meisel, Journal of Physical Chemistry. 1982, 86, 3391-3395.
T. Watanabe, O. Kawanami, H. Katoh, K. Honda, Y. Nishimura, M. Tsuboi,
Surface Science. 1985, 158, 341-351.
J. S. Suh, M. Moskovits, Journal of the American Chemical Society. 1986,
108, 4711-4718.
C. Otto, T. J. J. Van den Tweel, F. F. M. De Mul, J. Greve, Journal of Raman
Spectroscopy. 1986, 17, 289-298.
S. K. Kim, T. H. Joo, S. W. Suh, M.S. Kim, Journal of Raman Spectroscopy.
1986, 17, 381-386.
K. Itoch, K. Minami, T. Tsujino, M. Kim, Journal of Physical Chemistry.
1991, 95, 1339-1345.
J. Moger, P. Gribbon, A. Sewing, C. P. Winlove, Biochimica Et Biophysica
Acta. 2007, 91170, 12-918.
B. L. Mitchell, A. J. Patwardhan, S. M. Ngola, S. Chan, N. Sundararajan,
Journal of Raman Spectroscopy. 2008, 39, 380-388.
109
Chapter 4
Adenosine
AMP
SERS Intensity
ADP
ATP
A(2)MP
A(2,3)MP
A(3)MP
A(3,5)MP
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure S4.1: SERS spectra of adenosine ribonucleotides and adenosine in the
presence of silver citrate reduced colloid. Analyte concentrations were 1x10-5 mg/ml,
K2SO4 concentration was 0.020 M, data collection time was 50 seconds, and laser
power was 0.20 W at the laser.
110
Chapter 4
Table S4.1: SERS band assignments of adenosine, AMP, ADP, ATP, A(2)MP,
A(2,3)MP, A(3)MP and A(3,5)MP [2-6].
Assignment
Adenosine
AMP
ADP
ATP
A(2)MP
A(2,3)MP
A(3)MP
A(3,5)MP
Backbone A bend
-
-
616
616
-
-
-
-
Ring breathing mode
738
735
735
735
735
738
738
738
OH bend
-
-
776
-
-
-
-
-
(C-O),(C-C)(R) str
-
-
-
-
-
799
799
-
“
“
“
-
925
916
911
931
931
928
-
-
956
-
959
-
956
-
-
984
981
950
-
-
-
-
-
-
-
-
-
-
1031
1028
1034
1028
1026
1028
1034
-
-
-
-
1078
-
-
-
-
-
1129
-
1143
-
-
-
Phosphate vibration
Asym PO32 str
-
1178
-
1183
-
1181
-
-
-
-
-
C-N, C-H str
1260
1252
1249
1247
-
-
-
1255
Pyr ring
1335
1328
1325
1325
1273
1328
1328
-
In-plane ring vibrations
of adenine residue
-
-
-
-
1338
-
-
1335
Imidazole ring str
-
1361
1364
1364
1376
1361
-
1366
C-N, C=C str
1389
1394
1394
1394
-
1399
1397
1394
C-N, C-H
-
-
-
-
-
-
-
-
Ribose CH2 bend
1469
1465
-
-
1457
1462
1462
1462
C2H,N-C,C-H str
-
-
1472
1472
1495
-
-
-
C-C, C-N (Imidazole
ring)
-
1507
1512
1512
-
1502
1504
-
C=C str, C=N str
„
-
1557
1585
1575
1624
1578
1549
1563
1563
1558
C-C, C-N str
-
1659
1652
1652
1645
1640
1643
1638
Asym PO32 bend
“
Sym
PO32
-
Sym PO32
2-
Sym PO3
Sym
PO32
2-
Sym PO3
str
str
str
str
str
*str= stretching, bend= bending, sym= symmetric, A-sym= anti-symetric, A=adenosine, pyr= pyrimidine.
111
Chapter 5
Chapter 5
Study of Experimental and Computational
Raman and Raman Optical Activity (ROA)
Spectra of Cyclic and Linear L-Ala-L-Ala in
Solution
112
Chapter 5
5.0 Declaration
This chapter consists of one draft paper awaiting submission to Journal of Raman
Spectroscopy: S. Ostovar pour 1*, T. J. Dines 2, C. Levene 1, B.Z. Chowdhry 3 and
E.W. Blanch1, Journal of Raman spectroscopy. 2012
The manuscripts have been incorporated in a format identical to that for journal
submission, except for minor adjustments to incorporate them into this thesis. As
first author on this publication I carried out all of the associated experimental and
spectroscopic analysis. The calculation were carried out by T. J. Dines and B. Z.
Chowdhry and provided here for purpose of comparison with experimental results.
113
Chapter 5
Study of Experimental and Computational Raman and
Raman Optical Activity (ROA) Spectra of Cyclic and
Linear L-Ala-L-Ala in Solution
S. Ostovar pour 1*, T. J. Dines 2, C. Levene 1, B.Z. Chowdhry 3 and E.W. Blanch1
1
Faculty of Life Sciences, University of Manchester, MIB 131 Princess Street, Manchester, M1 7DN,
UK
2
Division of Electronic Engineering and Physics, University of Dundee, Dundee, DD1 4HN, UK
3
School of Science, University of Greenwich at Medway, Central Avenue, Chatham Maritime, Kent,
ME4 4TB, UK
E-mail: [email protected]; Fax: +44 (0)161 236 0409; Tel: +44 (0)161 306 5819
5.1 Abstract
Raman spectroscopy and Raman optical activity (which measures the small
difference between the intensities of left and right scattered polarized light from
chiral molecules) have been utilized to determine peptide conformations in solution
as a result of their sensitivity to molecular conformation. Short dipeptides such as
diketopiperazine (DKP) have been investigated past decades since both natural and
synthetic DKPs have a wide variety of desired biological effects including antitumour, anti-viral, anti-fungal and anti-bacterial activities. Density functional theory
(DFT) calculations were carried out using the Gaussian 09 program and the IEFPCM solvation method with the Karplus and York continuous surface charge
formalism, with water as the solvent. The starting geometries for geometry
optimization were based upon those obtained from the X-ray crystal structures of LAla-L-Ala. The calculated spectra of both linear and cyclic L-Ala-L-Ala are in good
agreement with our experimental ROA and Raman spectra; the comparison of which
114
Chapter 5
for both forms showed that ROA is more sensitive to structural changes where it
provided more marker bands. In particular, Raman bands at 1317, 1459 and 1524 cm1
for the cyclic form with their corresponding ROA bands at 1322, 1452 and 1521
cm-1 are not observed in the spectra for the linear form which suggests that these
bands are unique for the cyclic form of dipeptides. There are considerable differences
between the observed ROA bands for the cyclic and linear forms of dialanine that
reflect large differences in the vibrational modes of the polypeptide backbone upon
cyclization. In this study, the ROA spectrum of cyclic L-Ala-L-Ala has been reported
for the first time which demonstrated that ROA spectroscopy when utilised in
combination with computational modelling clearly provides a potential tool for
characterization of cyclic peptides.
5.2 Introduction
Cyclic dipeptides, particularly diketopiperazine (DKP), have been extensively
investigated during the last few decades as they are a large and important group of
compounds in medicinal chemistry due to their pharmacological activity [1]. In some
cases, cyclic compounds appear to have more interesting biological properties in
comparison to their linear equivalents [2]. The small size, hydrophobicity and lack of
charge exhibited by cyclic dipeptides render them more cell membrane permeable,
thereby promoting the bioavailability of these peptides [3]. They are also of interest
in studies on the thermodynamic behaviour of non-ionic compounds in water as a
result of sharing the capability of establishing hydrogen bonds with the solvent [4].
Because of their simplicity and limited conformational freedom along with a higher
probability of conformational homogeneity, cyclic dipeptides are readily used as
model compounds for longer peptide molecules [5]. Both natural and synthetic DKPs
115
Chapter 5
display a wide variety of biological activities including anti-tumour, anti-viral, antifungal, and anti-bacterial effects [6-8]. Due to their rigidity, chiral nature and varied
side chains they are also an attractive scaffold for drug design. Infrared, CD, NMR
and Raman spectra of linear and cyclic L-Ala-L-Ala and its N-deuterated form have
been obtained by several groups in attempts to characterize their structures in
solution [9-15].
Raman optical activity (ROA), which measures the small difference between the
intensities of Raman scattering using right and left circularly polarized light [16-18],
has been utilized in the determination of peptide conformation in solution as a result
of its sensitivity to molecular structure [19]. ROA provides more structural
information than Raman spectroscopy as ROA spectral details are directly responsive
to stereochemistry. Despite the considerable volume of work reported on linear
dialanine structures using both Raman and ROA [20-24], some important aspects of
their structures are not yet determined. Therefore, in this paper we have used both
Raman and ROA spectroscopies to investigate the differences in conformation
between cyclic and linear forms of L-Ala-L-Ala.
Experimental and calculated Raman and ROA spectra of linear L-Ala-L-Ala have
both been widely reported [25], and the Raman spectrum of cyclic L-Ala-L-Ala has
been presented [26] but not its ROA spectrum. The combination of quantum
mechanics/molecular mechanics modelling with experimental ROA spectroscopy
provides a uniquely sensitive tool for investigating the conformations of unusual
peptides. This approach was utilised by Bour and co-workers [25], who used the
B3LYP functional and the 6-31+G** Pople-type basis set, with the CPCM dielectric
correlation in the Gaussian version of the COSMO solvent model in order to better
characterize the spectra of the linear form [25]. Clusters of linear L-Ala-L-Ala with
116
Chapter 5
water molecules were obtained from molecular dynamics simulations that gave
qualitatively correct inhomogeneous broadening of Raman spectra lines but did not
bring a convincing improvement of ROA signals. There were also differences in the
calculated ROA intensities compared to the experimental case [25]. In this work, we
report the simulated and experimental ROA spectra for cyclic L-Ala-L-Ala for the
first time, as well as the ROA spectrum of the linear form, in order to improve
analysis of experimental spectra and so better understand the structural constraints
imposed by cyclization.
5.3 Experimental methods
Cyclic and linear L-Ala-L-Ala were purchased from Bachem Ltd. (Saffron Walden,
Essex, UK) and used without further purification. Deuterium oxide (99.98 atom %),
Na2HPO4 and NaH2PO4 were obtained from Sigma-Aldrich Ltd (Poole, Dorset, UK).
The buffer and sample concentrations used are given in the corresponding figure
captions. Samples for Raman and ROA spectroscopy were prepared by dissolving
lyophilized material into distilled and deionised phosphate buffer solution; where
low solubility was exhibited the sample was heated to 70 oC and then allowed to cool
to room temperature. Each solution was then centrifuged for 5 minutes at 3000 rpm
(1000 g) in order to minimize dust particles from the environment prior to loading
into a quartz microflourescence cell for spectroscopic measurement.
All Raman and ROA spectra were measured using a chiralRaman spectrometer
(BioTools Inc., Jupiter FL, USA) configured in the backscattering geometry and
operating with a wavelength of 532 nm and spectral resolution of 7 cm-1. The laser
power at the laser was 1.20 W and data collection times were 6-24 h.
117
Chapter 5
5.4 Computational methods
DFT calculations were carried out using the Gaussian 09 program [27] with the
B3LYP method [28, 29] and the AUG-cc-pVDZ basis set [30]. All calculations were
performed with the IEF-PCM solvation method [30] using the Karplus and York
continuous surface charge formalism [31, 32], with a Polarizability Continuum
Model (PCM) of the water solvent with default PCM parameters being used, except
that Pauling atomic radii were substituted for the default UFF radii. The starting
geometries for geometry optimisation were those previously obtained at the B3LYP/cc-pVDZ level for cyclic (L-Ala-L-Ala) [33], assuming C2 symmetry. The
starting geometries for linear L-Ala-L-Ala geometry optimisation were based upon
its reported X-ray crystal structure [34]. Vibrational spectra were calculated at the
optimised geometries and Raman and ROA activities were computed dynamically
for an excitation wavelength of 532 nm. Relative Raman and ROA intensities were
calculated from the computed Raman and ROA activities using the following
equations;
 Raman 
I fi
 ROA 
I fi



 fi 
4
 45a 2  7 2 

 hc fi  
 fi 1  exp  

 kT  


0
 fi 
4
 48G2 '  16 A2' 

 hc fi  
 fi 1  exp  

 kT  

0
where 0 = 18,797 cm-1 and T = 298.15 K.
In these equations the ytensor invariants are defined as follows:
a2 is the isotropic invariant of the electric-dipole/electricdipole polarizability tensor,
 is the symmetric anisotropic invariant of the electric-dipole/electricdipole
118
Chapter 5
polarizability tensor,
is the anisotropic invariant of the cross-product of the
electric-dipole/electricdipole polarizability tensor with the electric-dipole/magneticdipole polarizability tensor,
is the anisotropic invariant of the cross-product of the
electric-dipole/electricdipole polarizability tensor with the tensor A obtained by
contracting the electric-dipole/electric-quadrupole polarizability tensor with the
antisymmetric-unit tensor of Levi-Civita.
The Cartesian force constants obtained from the Gaussian 09 output were converted
to force constants expressed in terms of internal coordinates using a normal
coordinate analysis program derived from that of Schachtsneider [34]. A full set of
internal coordinates, including all bond angles and torsion angles, was reduced to a
set of 3N-6 symmetry-adapted internal coordinates. Normal coordinate analyses were
performed without scaling of force constants, producing potential energy
distributions for harmonic wavenumbers. Simulated Raman and ROA spectra were
constructed by convolution with a Lorentzian lineshape function of 10 cm-1 fwhm.
5.5 Results and Discussion
The atom numbering scheme along with computed molecular geometry of linear and
cyclic L-Ala-L-Ala are shown in Figure 5.1 and 5.2. The experimental and calculated
Raman and ROA spectra of both linear and cyclic L-Ala-L-Ala are shown in Figure
5.3, with their corresponding vibrational band assignments in Table 5.1. The band
lengths, angles and selected torsion angles are shown in Tables S5.1, S5.2 and S5.3
(refer to supplementary information). The predicted spectra of linear and cyclic LAla-L-Ala agree well with both experimental Raman and ROA spectra presented
here, in addition to matching those spectra reported in previous studies for the linear
form [20-25]. The calculated wavenumber values in some regions are slightly
119
Chapter 5
overestimated by less than 1%; the relative intensities in the spectra of linear L-AlaL-Ala are predicted accurately for most of the Raman and ROA bands, while those of
cyclic L-Ala-L-Ala are reasonably well predicted. To assist in gaining reliable
assignment of features, the Raman and ROA spectra of deuterated versions of these
dipeptides were also investigated and are presented in Figure 5.4. The Raman and
ROA band assignments for both types of L-Ala-L-Ala are listed in Table 5.2. The
experimental and calculated ROA spectra of the cyclic form of L-Ala-L-Ala are
reported here for the first time.
ROA and Raman spectra of peptides are often considered in terms of three distinct
regions: the backbone skeletal stretch region ~870-1150 cm-1 which originates from
Cα-C, Cα-Cβ, Cα-N stretch coordinates; the amide III region from ~1230-1340 cm-1 in
which bands are involved mainly from N-H in-plane deformations with Cα-N
stretching and contributions from Cα-H deformations; and the amide I region ~16301700 cm-1 which is associated mostly with C=O stretching [35]. The amide III region
is very important for ROA study as the coupling between N-H and Cα-H deformation
is very sensitive to geometry which provides information on ROA band structure
[35]. However, the spectra of cyclic dipeptides may contain larger features in other
spectral regions than are typically observed for the corresponding linear forms;
therefore here we discuss the features observed across a much broader spectral range.
A comparison of solution models in various isotopomers used in the calculation of
the Raman and ROA spectra of linear L-Ala-L-Ala was reported by Jalkanen et al.
[20]. Most recently, the conformational changes in ROA spectra of isotopic
substitution of C and N for linear L-Ala-L-Ala has been investigated by employing a
CPCM solvent model; however it did not bring a convincing improvement of ROA
signals when matched up to standard dielectric solvent correlation [25]. We discuss
120
Chapter 5
the differences between the spectra of the two dipeptides, as well as the level of
agreement between experiment and computation over the full spectral range
measured, in detail below.
400-800 cm-1 region
In previous reports of Raman and ROA spectra of linear L-Ala-L-Ala [25] only a few
peaks below 580 cm-1 were assigned while in the present study all of the vibrational
mode assignments from 400-800 cm-1 are well predicted that enables a full
vibrational mode analysis as shown in Table 5.1. Overall there is good agreement
between the simulated and experimental spectra of both dipeptides despite the over
prediction of ROA band intensities from 600-800 cm-1 for cyclic L-Ala-L-Ala which
also corresponds to the over prediction of the corresponding Raman intensities.
The peaks in the 400-800 cm-1 region of the Raman and ROA spectra for both
dipeptides are due to a variety of ring, C=O and N-H vibrations as well as to C-C
bending modes. The –ve/+ve ROA couplet at 433/473 cm-1 does not correspond to
any similar ROA band for linear L-Ala-L-Ala and is predicted as a +ve band in the
calculation that arises from the DKP ring vibration. This is the first ROA signature of
cyclic dipetide structures observed at low wavenumber.
In the experimental ROA spectrum of linear L-Ala-L-Ala two bands, one +ve and
one –ve, can be observed at 458 and 580 cm-1, respectively, that are due to C-N and
C-C stretching vibrations. For cyclic L-Ala-L-Ala there are strong negative ROA
peaks predicted at 678 and 616 cm-1 as well as a +ve peak at 597 cm-1 which
correspond to either very weak or negligible signals in the experimental ROA
spectrum of cyclic L-Ala-L-Ala. This may be explained by overlap of several of
these oppositely signed bands leading to their cancellation. The Raman band
121
Chapter 5
observed at 688 cm-1 is assigned to a combination of C-C stretching, C-C-N and
C=O bending that appears stronger for the cyclic form, which is also predicted by the
calculations, this band is shifted upwards by about 12 cm-1 upon deuteration for both
linear and cyclic L-Ala-L-Ala which verifies that there is a similar degree of
coupling in linear and cyclic forms of the C=O stretch vibration with N-H bending.
The calculated ROA bands at 473, 601, 720 and 790 cm-1 for the cyclic form are
mainly due to the DKP ring and amide VI vibrations which are either very weak or
negligible in the experimental ROA spectrum.
800-1400 cm-1 region
The relative intensities of cyclic and linear L-Ala-L-Ala bands in both experimental
Raman and ROA spectra are in excellent agreement with the calculated intensities.
Both simulated ROA spectra of linear and cyclic L-Ala-L-Ala demonstrate the best
level of correlation with the corresponding experimental spectra in the region from
800-1400 cm-1. This is significant since the extended amide III region, from 12301340 cm-1 is included which is a sensitive fingerprint of peptide conformation.
Therefore, differences in ROA spectra within this region are particularly useful for
investigating structural differences.
Additionally, the most notable differences between the experimental Raman spectra
of linear and cyclic L-Ala-L-Ala can be observed within this region. The observed
Raman bands for linear L-Ala-L-Ala are more intense in contrast to those for cyclic
L-Ala-L-Ala, for which the same sample concentration and conditions were applied.
The intense Raman peaks for linear L-Ala-L-Ala at ~885, 1281 and 1409 cm-1 that
are mainly assigned to C-N, C-C and C-H stretching also appear as intense bands
when dissolved in D2O, albeit with a 3 cm-1 upward shift. The corresponding ROA
122
Chapter 5
bands are observed at 885, 1283 and 1417 cm-1 while these bands are not present in
either the Raman or ROA spectra for the cyclic form. These latter vibrational modes
are thus greatly changed by cyclization and formation of the second peptide linkage.
The observed intensities in the region of 1000-1400 cm-1 appear to be sensitive to the
structural changes induced by cyclization in both Raman and ROA spectra as they
reduce by a factor of two for the cyclic form. The experimental ROA features for
linear L-Ala-L-Ala at 1025, 1078, 1097, 1121 and 1384 cm-1 reverse their signs for
the cyclic form. These bands are mainly assigned to CH3, C-C and C-N stretching
mode which indicates that stereochemical changes induced by cyclization for these
vibrational modes are indeed present. The only corresponding Raman bands for
cyclic and linear L-Ala-L-Ala appear at 1105 and 1097 cm-1. These sign changes in
band appearance illustrate the influence of cyclization on the vibrational coordinates
of the peptide group and emphasize that ROA is more sensitive than Raman
spectroscopy for studying the effects of cyclization upon conformation. Some
features in the experimental ROA spectra of linear and cyclic L-Ala-L-Ala are not
predicted by the computed spectra e.g. the bands at 1078 and 1327 cm-1 for linear LAla-L-Ala and 925 and 1080 cm-1 for the cyclic form. Nevertheless, there is a good
match between both calculated and experimental Raman and ROA spectra of cyclic
and linear L-Ala-L-Ala apart from the slight under prediction of ROA bands in the
region from 1000- 1150 cm-1, which mainly originate from C-C, C-H and C-N
vibrational modes. The different appearance of both Raman and ROA spectra for
cyclic and linear L-Ala-L-Ala in both the backbone skeletal stretch region ~ 10501200 cm-1 and the extended amide III region ~1250-1350 cm-1 suggests that the
backbone conformation is very different in these two dialanines.
123
Chapter 5
1400- 1800 cm-1 region
A number of spectral differences also occur within the high wavenumber region as a
consequence of cyclization. The positive ROA peak for linear L-Ala-L-Ala at 1417
cm-1 does not correspond to any peak in the spectra of the cyclic form in either the
hydrated or deuterated states. The ROA band due to CH3 bending vibrations, which
appears as a sharp and unusually intense negative peak at 1452 cm-1 for cyclic L-AlaL-Ala and flips its sign to a positive and smaller peak at 1457 cm-1 for linear L-AlaL-Ala in H2O, is up shifted by 5 cm-1 in D2O for the linear form but remains at the
same position for the cyclic form. The amide I Raman band at ~1650 cm-1 and amide
II mode at ~1550 cm-1 materialize in this region and two bands are present in the
experimental Raman spectra at 1524 and 1661 cm-1 for cyclic L-Ala-L-Ala but only
at 1642 cm-1 for linear L-Ala-L-Ala. The corresponding Raman peak for 1661 cm-1
in the cyclic form appears as two distinct bands at 1594 and 1668 cm-1 in D2O. There
are no corresponding Raman or ROA bands around 1524 cm-1 for the linear form
which postulates that this is a marker band for cyclic L-Ala-L-Ala. In the previous
study by Bour et al. [21] the amide II mode remained unobserved in the computed
spectra of both normal and labelled linear L-Ala-L-Ala whereas, in our present study
the Raman and ROA amide II bands are successfully simulated. The computed
Raman and ROA spectra for both linear and cyclic L-Ala-L-Ala show a similar peak
at 1656 cm-1 in the amide I region, originating from the C=O stretching mode, but
such a feature is present only in the experimental ROA spectrum of the linear form.
Interestingly, no such amide I ROA band appears for the cyclic form, possibly
because the two carbonyl groups have opposing orientation so that their individual
ROA signatures effectively cancel each other. More ROA bands can be observed
experimentally for this region in contrast with the Raman spectra which confirms the
124
Chapter 5
enhanced sensitivity of ROA over Raman spectra to the structural changes upon
cyclization for L-Ala-L-Ala. We also note that the observed bands in the ROA
spectrum of cyclic L-Ala-L-Ala in this region differ to those of the linear form that is
found for the other spectral regions. In addition, there is no ROA signal observed in
the amide II region for linear L-Ala-L-Ala, whereas there are small but clear +ve
amide II bands at 1477 and 1521cm-1 for the cyclic form.
5.6 Conclusion
Cyclization is an important structural modification of biological peptides, as is
shown by the increasing number of cyclic peptides that are now being discovered to
perform physiological function, but the conformational constraints imposed by this
process are not well understood. This study has identified several Raman and ROA
marker bands that identify cyclization in dialanine. ROA spectra, being sensitive to
stereochemistry, are found to be more sensitive to cyclization with ROA marker
bands being observed at 433, 790, 845, 987, 1067, 1304, 1322, 1407, 1452, 1521 and
1541 cm-1. Although less sensitive, Raman spectra also appear to contain bands that
identify cyclization, these principally being 989, 1183, 1317, 1459 and 1524 cm-1.
Even though explicit solvation was not utilized in the calculations of the Raman and
ROA spectra, we obtained a good level of agreement with experimental data that has
allowed the vibrational modes responsible for the spectral features to be identified.
The combination of ROA spectroscopy with computational modelling clearly
provides an incisive tool for characterizing cyclic peptides.
125
Chapter 5
Figure 5.1: The chemical structure with atom numbering scheme (left) and calculate
minimum energy conformation (right) of linear L-Ala-L-Ala.
H5
15
H
H12
H2
H 17
C
C4
3
O7
H 18
C
N1
C
6
H16
O14
C13
N8
10
H 20
C
11
H9
H19
Figure 5.2: The chemical structure with atom numbering scheme (left) and calculate
minimum energy conformation (right) of cyclic L-Ala-L-Ala.
126
Chapter 5
Exp. Linear L-Ala-L-Ala
9
4.5x10
Exp. Cyclic L-Ala-L-Ala
R
I +I
L
Calc. Linear L-Ala-L-Ala
9
2x10
Calc. Cyclic L-Ala-L-Ala
400
600
800
1000
1200
1400
1600
1800
Exp. Linear L-Ala-L-Ala
6
1.7x10
Exp. Cyclic L-Ala-L-Ala
R
I -I
L
Calc. Linear L-Ala-L-Ala
5
5.6x10
Calc. Cyclic L-Ala-L-Ala
400
600
800
1000
1200
1400
1600
1800
-1
Wavenumber (cm )
Figure 5.3: Experimental and computed Raman (top) and ROA (bottom) spectra of
linear (pH= 7.0) and cyclic L-Ala-L-Ala (pH= 7.0) in aqueous solution. The
concentration for each sample was 50 mg/ml and laser power was 0.6 W at the
sample. The marker bands that are induced upon cyclization are highlighted by
shading.
127
Chapter 5
Exp. Linear L-Ala-L-Ala
9
5x10
R
I +I
L
Calc. Linear L-Ala-L-Ala
Exp. Cyclic L-Ala-L-Ala
9
5x10
Calc. Cyclic L-Ala-L-Ala
400
600
800
1000
1200
1400
1600
1800
Exp. Linear L-Ala-L-Ala
6
2x10
Calc. Linear L-Ala-L-Ala
R
I -I
L
Exp. Cyclic L-Ala-L-Ala
6
1x10
Calc. Cyclic L-Ala-L-Ala
400
600
800
1000
1200
Wavenumber cm
1400
1600
1800
-1
Figure 5.4: Experimental and computed Raman (top) and ROA (bottom) spectra of
linear (pH= 7.0) and cyclic L-Ala-L-Ala (pH= 7.0) in D2O. The concentration for
each sample was 50 mg/ml and laser power was 0.6 W at the sample. The marker
bands that are induced upon cyclization are highlighted by shading.
128
Chapter 5
5.7 References:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
S. C. Brauns, P. Milne, R. Naude, M. Van de Venter, Anticancer Research.
2004, 24, 1713-1719.
P. J. Milne, A. L. Hunt, K. Rostoll, J. J. Van der Walt, C. J. M. Graz, Journal
of Pharmacy and Pharmacology. 1998, 50, 1331-1337.
C. B. Unal, M. D. Owen, W. R. Millington, Brain Research. 1997, 47, 52-59.
V. Crescenzi, A. Cesaro, E. Russo, International Journal of Peptide and
Protein Research. 1973, 5, 427-434.
M. J. O. Anteunis, Bulletin des Sociétés Chimiques Belges. 1978, 87, 627650.
K. McCleland, P. J. Milne, F. R. Lucieto, C. Frost, S. C. Brauns, M. V. D.
Venter, J. Du Plessis, K. Dyason, Journal of Pharmacy and Pharmacology.
2004, 56, 1143-1153.
S. W. Yang, T. M. Chan, J. Terracciano, D. Loebenberg, G. D. Chen, M.
Petal, V. Gullo, B. Pramani, M. Chu, Journal of Antibiotics. 2004, 57, 345347.
S. C. Brauns, G. Dealtry, P. Milne, R. Naude, M. Van De Venter, Anticancer
Research. 2005, 25, 4197-4202.
D. B. Davies, M. A. Khalad, Journal of the Chemical Society, Perkin
Transactions 2. 1976, 11, 1238-1244.
K. D. Kopple, V. Narutis, International Journal of Peptide and Protein
Research. 1981, 18, 33-40.
R. L. Bowman, M. Kellerman, W. C. Johnson, Biopolymers. 1983, 22, 10451070.
F. L. Bettens, R. P. A. Bettens, R. D. Brown, P. D. Godfrey, Journal of the
American chemical society. 2000, 122, 5856-5860.
Y. Zhu, M. Tang, X. Shi, Y. Zhao, International Journal of Quantum
Chemistry. 2007, 107, 745-753.
T. C. Cheam , S. Krimm, Spectrochimica Acta Part A: Molecular
Spectroscopy. 1984, 40, 481-501.
T. C. Cheam, S. Krimm, Spectrochimica Acta Part A: Molecular
Spectroscopy. 1988, 44, 185-208.
L. D. Barron, Molecular Light Scattering and Optical Activity. Cambridge
University Press. 2004.
F. Zhu, N. W. Isaacs, L. Hecht, L. D. Barron, Structure. 2005, 13, 1409-1419.
T. A. Keiderling, Current Opinion in Chemical Biology. 2002, 6, 682-688.
S. J. Ford, Z. Q. Wen, L. Hecht, L. D. Barron, Biopolymers. 1994, 34, 303313.
K. J. Jalkanen, R. M. Nieminen, M. Knapp-Mohammady, S. Suhai,
International Journal of Quantum Chemistry. 2003, 92, 239-259.
P. Bour, J. Kapitan, V. R. Baumruk, Journal of Physical Chemistry A. 2001,
105, 6362-6368.
M. Knapp-Mohammady, K. J. Jalkanen, F. Nardi, R. C. Wade, S. Suhai,
Chemical Physics. 1999, 240, 63-77.
A. F. Weir, A. H. Lowrey, R. W. Williams, Biopolymers. 2001, 58, 577-591.
P. Mukhopadhyay, G. R. Zuber, D. N. Beratan, Biophysical Journal. 2008,
95, 5574-5586.
J. Sebek, J. Kapitan, J. Sebestik, V. Baumruk, P. Bour, The Journal of
Physical Chemistry A. 2009, 113, 7760-7768.
129
Chapter 5
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
A. P. Mendham, T. J. Dines, M. J. Snowden, B. Z. Chowdhry, R. Withnall,
Journal of Raman Spectroscopy. 2009, 40, 1478-1497.
Gaussian 09, Revision A.1, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J.
B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V.
Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
A. D. Becke, Journal of Chemical Physics. 1993, 98, 5648-5652
C. Lee, W. Yang, R. G. Parr, Physical Review B. 1988, 37, 785-789.
T. H. Dunning, Journal of Chemical Physics. 1989, 90, 1007-1021.
M. T. Cances, V. Mennucci, J. Tomasi, Journal of Chemical Physics. 1997,
107, 3032-3041.
D. M. York, M. Karplus, Journal of Physical chemistry A. 1999, 103, 1106011079.
R. J. Fletterick, C. Tsai, R. E. Hughes, Journal of physical chemistry. 1971,
75, 918-922.
J. A. Schachtschneider, Vibrational Analysis of Polyatomic Molecules, Parts
V and VI, Technical Report Nos. 231 and 57, Shell Development Co.,
Houston TX, 1964 and 1965.
L. D. Barron, Current Opinion in Structural Biology. 2006, 16, 638-643.
130
Chapter 5
Table 5.1: Calculated and experimental wavenumber band assignments for Raman and ROA of cyclic and linear L-Ala-L-Ala in H2O
Column1
Column2 Column3 Column4 Column5 Column6 Column7
Experimental
Calculated
Band assignments
Linear
Cyclic
Linear
Cyclic
Linear
ROA
Raman
ROA
Raman
458 (+ve)
550 (+ve)
580 (-ve)
616 (+ve)
658 (-ve)
699 (+ve)
776 (-ve)
834 (+ve)
859 (-ve)
885 (+ve)
905 (-ve)
931 (+ve)
(25)
953 (-ve)
964 (+ve)
984 (+ve)
1009 (-ve)
1025(+ve)
-
Column8
Cyclic
561
577
660
678
760
767
840
878
914
-
421 A
456 B
484 A
597 A
600 B
613 A
678 A
684 B
780 B
786 A
844 B
963 B
(N5C4C10) (41), ip(CO) (11)
(N9C8C13) (57)
op(NH) (36),(N5C6) (44)
(C1C4) (10),(N5C6C8) (17), (CO2) (13),(CO2) (14)
(N5C4C1) (13), ip(CO) (12),(CO2) (23),(CO2) (10)
(C6C8) (17),(C6C8N9) (13), ip(CO) (20),(CO2) (10)
op(CO) (29),(CO2) (37)
op(CO) (33),(CO2) (25)
s(CO2) (10),(C1C4) (19), (CO2) (22)
(C8N9) (24), (C8C13) (16)
(C4N5) (10),ip(CH3) (18),op(CH3) (18)
-
ip(ring-1) (12), ip(ring-2) (58)
(CCH) (21),ip(CO) (29),op(CO) (15)
(NC) (18), ip(ring-1) (43), ip(ring-2) (12)
(NC) (18), ip(ring-1) (43), ip(ring-2) (12)
op(NH) (84) [amide V]
(CC) (31), ip(ring-1) (13), ip(CO) (14), op(NH) (35)
(NC) (10), (CC) (21), op(NH) (22), op(CO) (25)
(C3C4) (17), ip(ring-3) (32), op(CO) (37) [amide VI]
(CC) (10), ip(ring-3) (25), op(CO) (25)
op(CO) (47), ip(CH3) (18) [amide VI]
(C3C4) (11),(CC) (37), op(CH3) (12)
(NC) (18), ip(ring-3) (16), ip(CH3) (13), op(CH3)
956
994
1019
-
1001 A
1035 A
(C4C10) (13), (C6C8) (19),op(CH3′) (15)
(C8N9) (11), (C8C13) (30),ip(NH3) (30), op(CH3′) (16)
op(NH3) (36), ip(CH3′) (34)
-
(NC) (10), op(CH3) (54)
(C3C4) (31),ip(CH3) (20)
458
473
688
782
885
928
-
433(-ve)
473 (+ve)
601 (-ve)
720 (-ve)
790 (+ve)
845 (-ve)
-
473
485
616
688
761
793
844
925
-
402
454
-
956
995
1012
-
953 (-ve)
987 (+ve)
1023 (-ve)
-
989
-
131
Chapter 5
1053 (-ve)
1078(+ve)
1097 (-ve)
1121(+ve)
1175 (-ve)
1247(+ve)
1270 (-ve)
1283(+ve)
1343(+ve)
1384 (-ve)
1399 (-ve)
1417(+ve)
1437 (-ve)
1457(+ve)
1472 (-ve)
1568(+ve)
1628 (-ve)
1680(+ve)
1050
1105
1172
1281
1327
1343
1374
1409
1469
1642
-
1067 (+ve)
1080 (-ve)
1094 (+ve)
1129 (-ve)
1156 (+ve)
1304 (+ve)
1322 (-ve)
1351 (+ve)
1389 (+ve)
1407 (+ve)
1437 (+ve)
1452 (-ve)
1477 (+ve)
1492 (+ve)
1521 (+ve)
1541 (-ve)
1553 (+ve)
1567 (+ve)
1640 (+ve)
1654 (-ve)
1097
1186
1317
1386
1459
1524
1661
1051
1109
1120
1138
1163
1233
1280
1286
1335
1365
1369
1403
1410
1428
1456
1460
1471
1474
1495
1546
1589
1612
1650
1656
1061 B
1109 A
1129 B
1160 B
1194 A
1294 A
1314 B
1322 B
1332 A
1388 A,B
1395 A
1436 B
1452 B
1454 A
1459 B
1460 A
1477 B
1521 A
1652 B
1656 A
(C4N5) (13),(C4C10) (15),′(C4H11) (14),op(CH3) (20)
(C4C10) (27), ip(CH3) (17), op(CH3′) (10)
(C8N9) (17), (C8C13) (13),ip(NH3) (18)
op(NH3) (10),′(C8H14) (15), op(CH3′) (16)

(C4N5) (25),(N5C4C10) (11), op(CH3) (20)
op(NH3) (27), (N9C8C13) (10), ip(CH3′) (16)
(C6N5) (27), ip(NH) (29),′(C8H14) (11)
′(C4H11) (72), ip(CH3) (10)
(C4H11) (65),′(C8H14) (13)
(C8H14) (14),′(C8H14) (32)
s(CO2) (17),s(CH3) (59)
s(CO2) (40),s(CH3) (34)
s(CH3′) (88)
(C8H14) (45),s(CH3′) (11), as(CH3′) (21)
as(CH3) (89), ip(CH3) (10)
as(CH3) (84)
as(CH3′) (82), ip(CH3′) (13)
(C8H14) (16), as(CH3′) (62), op(CH3′) (10)
s(NH3) (90)
as(CO2) (91)
(C6N5) (30), ip(NH) (63)
as(NH3) (74)
(CO) (33), as(NH3) (45)
(CO) (32), as(NH3) (47)
132
(CCH) (12),ip(CH3) (26), op(CH3) (16)
(C3C4) (39),(CCH) (11),ip(CH3) (33)
(C3C4) (46),ip(CH3) (17)
(NC) (28), (CCH) (20), op(CH3) (24)
(NC) (31), (CCH) (21), op(CH3) (13)
(NC) (17), (CCH) (41),(CCH) (17)
(CCH) (49),(CCH) (20)
(CCH) (57),(CCH) (14)
(CCH) (64),(CCH) (16)
s(CH3) (84) (91)
(NC) (13), ip(NH) (40),s(CH3) (12) [amide III]
(NC) (31), (CC) (10),ip(NH) (15),as(CH3) (26)
(CO) (10), (NC) (22), ip(NH) (11),as(CH3) (41) [amide III]
as(CH3) (83)
as(CH3) (70)
as(CH3) (86)
(CO) (14), ip(NH) (36), as(CH3) (24) [amide II]
(CO) (17), (NC) (22), (CC) (12), ip(NH) (20) [amide II]
(CO) (60), (NC) (10), ip(NH) (15) [amide I]
(CO) (61), (NC) (12), ip(NH) (15) [amide I]
Chapter 5
Table 5.2: Calculated and experimental wavenumber band assignments for Raman and ROA of cyclic and linear L-Ala-L-Ala in D2O.
Experimental
Linear
Cyclic
ROA
Raman ROA
436
445 (-ve)
470 (-ve)
(17),op(ND) (30)
(14)
495 (-ve)
550 (+ve)
571
613 (+ve)
631
640 (+ve)
738 (+ve)
755
(20)
782
(13)
825 (-ve)
828
877 (+ve)
877
916 (+ve)
911
975 (+ve)
945
III]
1012 (+ve) 1003
ip(CH3) (25)
1048 (-ve) 1067 (+ve) 1067
1116(+ve) 1100
1159 (-ve) -
Calculated
Linear
Cyclic
Band assignments
Linear
Cyclic
Raman
461 (+ve)
-
431
443
-
420 B
473 A
(N9C8C13) (52)
(N5C4C10) (18),op(NH) (36),(N5C6) (20)
-
ip(CO) (13), op(ND) (73) [amide V]
(NC) (16), (CCH) (21), ip(ring-1) (13), ip(ring-2)
-
482
-
486 B
-
(CCH) (18),ip(CO) (20), op(ND) (26),op(CO)
504 (+ve)
601 (+ve)
622 (+ve)
729 (+ve)
607
676
694
-
558
629
661
754
498 A
590 A
661 A
678 B
757 B
(C1C4) (10),(N5C6C8) (19), (CO2) (10),(CO2) (20)
(C6C8) (24),ip(ND3) (15), ip(CO) (29) 
(N5C4C1) (21), (CO2) (36)
(C6C8C13) (13), op(CO) (60)
ip(ring-1) (34), op(NH) (46) [amide V]
(CC) (21), ip(CO) (50) [amide IV]
(NC) (10), (NC) (10), (CC) (29), op(CO) (26)
(C3C4) (19), op(CO) (36), ip(ring-3) (31) [amide VI]
(NC) (11), (CC) (25), ip(ND) (15), ip(ring-3)
770 (+ve)
-
767
-
763
811
786 A
812 B
(C1C4C10) (11),(CO2) (56)
(C8N9) (12),(C8C13) (26), ip(ND3) (14), op(ND3) (33)
op(CO) (47), ip(CH3) (18) [amide VI]
(CC) (16), ip(ND) (11), op(CO) (24), ip(ring-3) (10),ip(CH3)
822 (-ve)
885 (+ve)
919 (+ve)
-
882
942
833
856
877
908
950
917 B
931 A
ip(ND3) (29), (CO2) (12)
(C1C4) (13), ip(ND3) (11), op(ND3) (10), ip(CH3) (13), (CO2) (10)
(C8N9) (19),op(ND3) (19),ip(CH3′) (23)
(C4N5) (13), ip(CH3) (16),op(CH3) (19)
(C8N9) (11), (C6C8) (19),op(CH3′) (20)
(NC) (28), ip(ND) (15),op(CH3) (36)
(NC) (19), ip(ND) (41),op(CH3) (30) [amide
989 (+ve)
992
1012
1017 B
(C4C10) (19),ip(ND) (66)
(C3C4) (25), ip(ND) (14), ip(ring-3) (13),
1072 (+ve)
1091 (-ve)
1113 (+ve)
-
1037
1094
1151
1049
1078
1108
1126
1145
1156
1028 A
1083 A
1101 B
1118 A
1151 B
(C4N5) (12), (C4C10) (11),′(C4H11) (14), ip(CH3) (12), op(CH3) (17)(C3C4) (35), (CCH) (10), ip(CH3) (21)
(C8C13) (23),(C6C8N9) (10)
ip(ND) (21), (CCH) (10),op(CH3) (38)
s(ND3) (12), ip(CH3′) (11), op(CH3′) (28)
(C3C4) (20),ip(CH3) (22), op(CH3) (16)
(C4N5) (11), (C4C10) (18)
(C3C4) (34),ip(CH3) (29)
(C8N9) (13), (C8C13) (15),s(ND3) (20), as(ND3) (21)
(C4N5) (14),(N5C4C10) (16), ip(CH3) (15), op(CH3) (20)
(C3C4) (10),(NC) (11), (CCH) (18), ip(CH3) (14), op(CH3) (24)
133
Chapter 5
(16)
1204
-
-
1207
-
1165
1184
1194
-
1229 B
s(ND3) (26), as(ND3) (33)
as(ND3) (95)
(C8N9) (10),s(ND3) (18), as(ND3) (10), ip(CH3′) (10)
-
(NC) (25), ip(ND) (26), (CCH) (10),ip(CO)
(20)
1281 (+ve)
1330 (+ve)
1358 (-ve)
1379 (+ve)
1412 (+ve)
-
-
1244 (-ve)
-
-
1230 A
-
[amide III]
(NC) (20), ip(ND) (10), (CCH) (12), (CCH)
1281
1312
1333
1358
1371
1412
1462
1487
1556
1594
1668
1304 (+ve)
1340 (+ve)
1384 (+ve)
1452 (+ve)
1479 (+ve)
1502 (+ve)
1322
1386
1457
1507
1649
1286
1329
1351
1369
1400
1407
1411
1456
1457
1463
1473
1502
1546
1640
1309 A
1319 B
1335 A
1347 B
1388 B
1389 A
1441 B
1452 A
1458 A, B
1461 B
1495 A
1634 B
1637 A
′(C4H11) (73)
(C4H11) (38),′(C8H14) (37)
(C4H11) (35),′(C8H14) (30)
s(CO2) (20),s(CH3) (62)
s(CO2) (25),(C8H14) (31),s(CH3) (27)
s(CO2) (14),(C8H14) (44)
s(CH3) (93)
as(CH3) (90), ip(CH3) (10)
as(CH3) (48),as(CH3′) (29)
as(CH3) (35),as(CH3′) (39)
as(CH3′) (81), ip(CH3′) (13)
(C6N5) (32), (C6C8) (11), ip(ND) (11)
as(CO2) (91)
(CO) (72), (C6N5) (16)
(NC) (18), (CCH) (33),(CCH) (11)
as(CH3) (20), (CCH) (36),(CCH) (19)
(CCH) (60),(CCH) (16)
as(CH3) (36), (CCH) (12),(CCH) (10)
s(CH3) (94)
s(CH3) (96)
(NC) (54) [amide II]
as(CH3) (76)
as(CH3) (88),as(CH3) (89)
as(CH3) (80)
(NC) (35), (CC) (14) [amide II]
(CO) (72), (NC) (10) [amide I]
(CO) (75), (NC) (10) [amide I]
1462 (-ve)
1487 (+ve)
1668 (+ve)
1647 (+ve)
134
Chapter 5
5.8 Supplementary information
Table S5.1: Calculated and experimental bond lengths (Å) for cyclic and linear LAla-L-Ala.
Cyclic L-Ala-L-Ala
Calculated
X-raya
(N1C13)
r(N1H2)
r(C3N1)
r(C3C4)
r(C4H15)
r(C4H16)
r(H5C3)
r(C6C3)
r(O7C6)
r(C4H17)
1.3429
1.0147
1.4662
1.5281
1.0975
1.0976
1.1032
1.5249
1.244
1.0946
1.331
0.926
1.454
1.514
0.964
0.963
0.977
1.518
1.239
1.035
Calculated
X-raya
1.2661
1.5533
1.2628
1.0947
1.3381
1.4608
1.2445
1.5007
1.5351
1.0229
1.0269
1.096
1.0975
1.5345
1.0127
1.5285
1.097
1.0934
1.0223
1.0992
1.0966
1.0956
1.2422
1.5409
1.2279
0.8961
1.3462
1.4561
1.2252
1.4953
1.5306
0.8721
0.7652
0.8983
1.5181
0.9433
1.5231
1.1875
0.9581
0.866
1.131
1.0212
0.8012
Linear L-Ala-L-Ala
r(C1O2)
r(C1C4)
r(O3C1)
r(C4H11)
r(N5C6)
r(N5C4)
r(C6O7)
r(C8N9)
r(C8C6)
r(N9H17)
r(N9H16)
r(C10H20)
r(C10H19)
r(C10C4)
r(H12N5)
r(C13C8)
r(C13H22)
r(H14C8)
r(H15N9)
r(H18C10)
r(H21C13)
r(H23C13)
135
Chapter 5
Table S5.2: Calculated and experimental bond angles (o) for cyclic and linear L-AlaL-Ala.
Cyclic L-Ala-L-Ala
θ(C13N1H2)
θ(C13N1C3)
θ(N1C13O14)
θ(H2N1C3)
θ(N1C3C4)
θ(N1C3H5)
θ(N1C3C6)
θ(C3C4H15)
θ(C3C4H16)
θ(C4C3H5)
θ(C4C3C6)
θ(C3C4H17)
θ(H15C4H16)
θ(H15C4H17)
θ(H16C4H17)
θ(H5C3C6)
θ(C3C6N8)
θ(C3C6O7)
Calculated
115.69
126.49
122.29
116.57
109.82
109.05
111.29
109.65
110.75
108.71
112.26
110.05
108.95
108.28
109.13
105.55
116.78
120.93
Experimental
111.35
126.16
122.53
121.82
109.65
107.41
110.57
112.88
109.5
110.51
112.12
108.11
112.58
109.06
104.265
106.11
110.53
120.56
115.45
125.7
118.84
107.29
112.8
110.59
106.9
109.38
124.31
124.82
115.49
117.72
109.75
117.54
119.68
106.53
111.75
109.12
110.16
107
111.87
111.25
110.88
107.24
107.81
108.91
108.63
109.91
108.44
110.29
108.66
118.16
123.86
117.98
103.74
109.16
110.67
111.18
111.77
122.88
125.93
113.16
119.19
110.14
117.83
120.89
108.33
110.23
110.46
107.45
108.46
112.33
111
109.28
102.91
113.92
106.49
117.14
120.6
Linear L-Ala-L-Ala
θ(O2C1C4)
θ(O2C1O3)
θ(C4C1O3)
θ(C1C4H11)
θ(C1C4N5)
θ(C1C4C10)
θ(H11C4N5)
θ(H11C4C10)
θ(C6N5C4)
θ(N5C6O7)
θ(N5C6C8)
θ(C6N5H12)
θ(N5C4C10)
θ(C4N5H12)
θ(O7C6C8)
θ(N9C8C6)
θ(C8N9H17)
θ(C8N9H16)
θ(N9C8C13)
θ(N9C8H14)
θ(C8N9H15)
θ(C6C8C13)
θ(C6C8H14)
θ(H17N9H16)
θ(H17N9H15)
θ(H16N9H15)
θ(H20C10H19)
θ(H20C10C4)
θ(H20C10H18)
θ(H19C10C4)
θ(H19C10H18)
136
Chapter 5
θ(C4C10H18)
θ(C8C13H22)
θ(C13C8H14)
θ(C8C13H21)
θ(C8C13H23)
θ(H22C13H21)
θ(H22C13H23)
θ(H21C13H23)
110.85
110.8
110.85
110.42
109.19
109.21
108.73
108.43
104.99
107.07
112.21
108.22
110.12
117.46
105.4
108.44
Table S5.3: Calculated and experimental torsion angles (o) for cyclic and linear LAla-L-Ala.
Cyclic L-Ala-L-Ala

Calculated
(C10C13N1H2) 
175.65
(O14C13N1H2)

-3.77
(C10C13N1C3)

8.96
(O14C13N1C3) 
-170.46
(C13N1C3C4)

-154.91
(C13N1C3H5)

86.06
(C13N1C3C6)

-29.98
(N1C13C10N8)

17.2555
(N1C13C10C11) 
143.63
(N1C13C10H12) 
-98.09
(H2N1C3C4)

38.5
(H2N1C3H5)

-80.53
(H2N1C3C6)

163.44
(N1C3C4H15)

-56.84
(N1C3C4H16)

63.41
(N1C3C4H17)

-175.84
(N1C3C6N8)

20.08
(N1C3C6O7)

-160.5
(H5C3C4H15)

62.39
(C6C3C4H15)

178.78
(H5C3C4H16)

-177.35
(C6C3C4H16)

-60.96
(H17C4C3H5)

-56.61
(H17C4C3C6)

59.78
(C4C3C6O7)

-36.94
(H5C3C6O7)

81.33
Experimental
-
Linear L-Ala-L-Ala
τ (O2C1C4H11)
τ (O2C1C4N5)
τ (O2C1C4C10)
τ (H11C4C1O3)
τ (N5C4C1O3)
τ (C10C4C1O3)
τ (C1C4N5C6)
τ (C1C4N5H12)
τ (C1C4C10H20)
τ (C1C4C10H19)
τ (C1C4C10H18)
τ (H11C4N5C6)
45.15
162.6
-74.07
-136.17
-18.72
104.61
-92.69
79.58
56.62
176.36
-63.25
25
164.66
-76.74
44.64
-15.62
102.98
-135.64
-112.98
70.86
167.9
-55.26
0.87
137
Chapter 5
τ (H11C4N5H12)
τ (H11C4C10H20)
τ (H11C4C10H19)
τ (H11C4C10H18)
τ (O7C6N5C4)
τ (C8C6N5C4)
τ (C6N5C4C10)
τ (H12N5C6O7)
τ (H12N5C6C8)
τ (N5C6C8N9)
τ (N5C6C8C13)
τ (N5C6C8H14)
τ (H12N5C4C10)
τ (N5C4C10H20)
τ (N5C4C10H19)
τ (N5C4C10H18)
τ (O7C6C8N9)
τ (O7C6C8C13)
τ (O7C6C8H14)
τ (H17N9C8C6)
τ (H16N9C8C6)
τ (H15N9C8C6)
τ (C13C8N9H17)
τ (H14C8N9H17)
τ (C13C8N9H16)
τ (H14C8N9H16)
τ (H15N9C8C13)
τ (N9C8C13H22)
τ (N9C8C13H21)
τ (N9C8C13H23)
τ (H15N9C8H14)
τ (C6C8C13H22)
τ (C6C8C13H21)
τ (C6C8C13H23)
τ (H14C8C13H22)
τ (H21C13C8H14)
τ (H23C13C8H14)
-162.74
-61.34
58.41
178.8
-4.83
173.7
143.52
-177.08
1.45
151.25
-88.67
35.18
-44.22
-178.31
-58.56
61.82
-30.15
89.93
-146.22
-75.59
42.84
163.42
163.63
43.06
-77.94
161.49
42.64
57.89
-63.25
177.63
-77.93
-60.01
178.85
59.72
176.13
54.99
-64.14
-175.3
52.81
-170.35
-2.74
175.73
125.32
173.37
-8.16
165.4
-76.86
47.41
-50.84
-71.3
65.54
-16.05
101.69
-134.03
-56.2
56.86
175.59
-176.19
62.3
-63.13
175.36
55.61
62.08
-65.44
176.19
-65.9
-56.19
176.29
57.91
-178.79
53.69
-64.68
138
Chapter 6
Chapter 6
Combined Experimental and Computational
Study of Raman and Raman Optical Activity
(ROA) Spectra of Linear and Cyclic L-Ser-LSer in Solution
139
Chapter 6
6.0 Declaration
This chapter consists of one draft paper awaiting submission to Journal of Raman
Spectroscopy: S. Ostovar pour 1*, T. J. Dines 2, C. Levene 1, B.Z. Chowdhry 3 and
E.W. Blanch1, Journal of Raman spectroscopy. 2012
The manuscripts have been incorporated in a format identical to that for journal
submission, except for minor adjustments to incorporate them into this thesis. As
first author on this publication, I carried out all of the associated experimental and
spectroscopic analysis. The calculation were carried out by T. J. Dines and B. Z.
Chowdhry and provided here for purpose of comparison with experimental results.
140
Chapter 6
Combined Experimental and Computational Study of
Raman and Raman Optical Activity (ROA) Spectra of
Linear and Cyclic L-Ser-L-Ser in Solution
S. Ostovar pour, 1 T. J. Dines, 2 C. Levene, 1 B.Z. Chowdhry3 and E.W. Blanch1*
1
Faculty of Life Sciences, University of Manchester, MIB 131 Princess Street, Manchester, M1 7DN,
UK
2
Division of Electronic Engineering and Physics, University of Dundee, Dundee, DD1 4HN, UK
3
School of Science, University of Greenwich at Medway, Central Avenue, Chatham Maritime, Kent,
ME4 4TB, UK
*
Corresponding author: [email protected]; Fax: +44 (0)161 236 0409; Tel: +44 (0)161 306
5819
6.1 Abstract
A study of the conformations of cyclic and linear L-Ser-L-Ser in aqueous solution
has been carried out using Raman and Raman optical activity (ROA) spectroscopies
and quantum mechanical calculations. Raman and ROA spectra of linear and cyclic
L-Ser-L-Ser in H2O and D2O were measured and assignment of the observed bands
has been proposed from DFT calculations at the B3LYP/cc-pVDZ level assuming C2
symmetry in both cases. We found that ROA spectra are more sensitive than Raman
spectra to the structural changes induced by cyclization. Specifically, Raman bands
at 670, 1164 and 1317 cm-1 along with ROA bands at 682, 716, 782, 894, 1064,
1080, 1154, 1345, 1482 and 1683 cm-1 have been identified as marker bands for the
cyclic form. In addition, we observe an unusually intense amide II band in both the
Raman and ROA spectra at 1519 cm-1 for cyclic L-Ser-L-Ser that is not present in the
spectra of the linear form.
141
Chapter 6
6.2 Introduction
The cyclization of peptides has recently been the subject of a number of studies in
peptide chemistry in relation to their biological activity [1,2]. Their potential
application as antibiotic and anticancer drugs has been highlighted by several
publications [3-5]. Appropriate pharmaceutical delivery systems for these molecules
have not been fully understood untill recent times owing to the lack of available
information concerning their structural behaviour [3-5]. Whilst being the subject of
intrinsic interest and significant for the study of biologically active short dipeptides,
they also appear to be good model compounds for analysis of certain properties such
as secondary structure of similar residues within polypeptide chains [6], since
modelling of longer peptides at a similar level of accuracy would be very difficult
due to the high computational requirements.
Raman and Raman optical activity (ROA) spectroscopies have been previously
utilized for the determination of peptide conformation in aqueous environments.
ROA spectroscopy measures an intensity difference in the Raman scattering of rightand left-circularly polarized light from chiral molecules [7-10]. ROA spectroscopy is
particularly sensitive to conformational changes, thereby enabling characterization of
the natural behaviour of biological molecules in solution. However, the relationship
between the ROA spectrum and the molecular conformation is complex, therefore
theoretical simulations of ROA spectra are very useful for providing a more detailed
interpretation of the vibrational modes measured [11,12].
The combination of quantum mechanics/molecular mechanics modelling with
experimental ROA spectroscopy provides a uniquely sensitive tool for investigating
the conformations of unusual peptides. Previous X-ray, NMR, CD and IR studies
have been used to characterize the structure of cyclic L-Ser-L-Ser [13]. The Raman
142
Chapter 6
[14-15] and ROA [16] spectra of L-serine have been simulated using DFT and the
B3LYP hybrid functional in several studies in which the influence of side chain
conformation on the vibrational spectrum was investigated in ROA spectra [16],
though the authors made no comparison with experimental results. The Raman and
ROA spectra of tri-L-serine in aqueous solution were also investigated
experimentally and theoretically in the neutral and zwitterionic states [17]. The
spectra were calculated with DFT and B3LYP hybrid exchange along with a
consideration of implicit solvation effects using a Polarizability Continuum Model
(PCM). Due to the limitations in modelling of hydrogen bond interactions of water
with small peptides [17] inherent to the PCM, the experimental and computational
spectra obviously differed in some details [17].
To our knowledge, no experimental or calculated ROA spectra of linear and cyclic
L-Ser-L-Ser have been reported previously. Therefore, we report here the simulated
and experimental Raman and ROA spectra for linear and cyclic L-Ser-L-Ser for the
first time, in order to improve understanding of their conformations in solution and
the structural constraints imposed by cyclization. This is an extension of our recent
study on cyclization effects in dialanine (Chapter 5).
6.3 Experimental methods
Cyclic and linear L-Ser-L-Ser were purchased from Bachem Ltd. (Saffron Walden,
Essex, UK) and used without further purification. Deuterium oxide (99.98 atom %),
Na2HPO4 and NaH2PO4 were obtained from Sigma-Aldrich Ltd (Poole, Dorset, UK).
The phosphate buffer and sample concentrations are given in the corresponding
figure captions. Samples for Raman and ROA spectroscopy were prepared by
dissolving lyophilized material into the buffer solution; in any case of insolubility the
143
Chapter 6
sample was heated to 70 oC, and then allowed to cool to room temperature. Each
solution was centrifuged for 5 minutes at 3000 rpm (1000 g) in order to minimize the
presence of dust particles from the environment prior to loading into a quartz
microflourescence cell for spectroscopic measurement.
All Raman and ROA spectra were measured on a chiralRaman spectrometer
(BioTools Inc., Jupiter FL, USA) at a wavelength of 532 nm, configured in the
backscattering geometry and with a spectral resolution of 7 cm-1. The laser power
was 1.2 W and data collection times were 6-24 h.
6.4 Computational methods
DFT calculations were carried out using the Gaussian 09 program [18] with the
B3LYP method [19, 20] and the AUG-cc-pVDZ basis set [21]. All calculations were
performed with the IEF-PCM solvation method [22] using the Karplus and York
continuous surface charge formalism [23], with a Polarizability Continuum Model
(PCM) of the water solvent with default PCM parameters being used except that the
Pauling atomic radii were substituted for the default UFF radii. The starting
geometries for geometry optimization were those previously obtained at the
B3LYP/cc-pVDZ level for cyclic (L-Ser-L-Ser) [24], assuming C2 symmetry in both
cases. Vibrational spectra were calculated at the optimized geometries and Raman
and ROA activities were computed dynamically for an excitation wavelength of 532
nm. Relative Raman and ROA intensities were calculated from the computed Raman
and ROA activities using the equations:
144
Chapter 6
 Raman 
I fi
 ROA 
I fi



 fi 
4
 45a 2  7 2 

 hc fi  
 fi 1  exp  

 kT  


0
0
 fi 
4

 hc fi
 fi 1  exp  
 kT

 48G2 '  16 A2' 




where 0 = 18,797 cm-1 and T = 298.15 K.
In these equations the ytensor invariants are defined as follows:
a2 is the isotropic invariant of the electric-dipole/electricdipole polarizability tensor,
 is the symmetric anisotropic invariant of the electric-dipole/electricdipole
polarizability tensor,
is the anisotropic invariant of the cross-product of the
electric-dipole/electricdipole polarizability tensor with the electric-dipole/magneticdipole polarizability tensor,
is the anisotropic invariant of the cross-product of the
electric-dipole/electricdipole polarizability tensor with the tensor A obtained by
contracting the electric-dipole/electric-quadrupole polarizability tensor with the
antisymmetric-unit tensor of Levi-Civita.
The Cartesian force constants obtained from the Gaussian 09 output were converted
to force constants expressed in terms of internal coordinates using a normal
coordinate analysis program derived from those of Schachtsneider [24]. A full set of
internal coordinates, including all bond angles and torsion angles, was reduced to a
set of 3N-6 symmetry-adapted internal coordinates. Normal coordinate analyses were
performed without scaling of force constants, producing potential energy
distributions for harmonic wavenumbers. Simulated Raman and ROA spectra were
constructed by convolution with a Lorentzian lineshape function of 10 cm-1 fwhm.
145
Chapter 6
6.5 Results and Discussions
The atom numbering scheme along with computed molecular geometry of linear and
cyclic L-Ser-L-Ser are shown in Figure 6.1 and 6.2. The experimental and calculated
Raman and ROA spectra of both linear and cyclic L-Ser-L-Ser are shown in Figure
6.3, and their corresponding vibrational band assignments are presented in Table 6.1.
The bond lengths, angles and selected torsion angles are shown in Tables S6.1, S6.2
and S6.3 (refer to supplementary information). In order to enable reliable assignment
of spectral features, the Raman and ROA spectra of the deuterated versions of these
dipeptides were also investigated and are presented in Figure 6.4. The Raman and
ROA band assignments for both forms of L-Ser-L-Ser in D2O are listed in Table 6.2.
The observed Raman and ROA bands for this region are in reasonably good
agreement with previous measurements of the Raman spectrum of cyclic L-Ser-LSer and Raman and ROA measurements of L-serine and tri-L-serine [14,16,17,27].
Even without the modelling of explicit hydration the Raman spectra for both linear
and cyclic forms of L-Ser-L-Ser in H2O presented in Figure 6.3, are quite well
simulated, with most Raman bands being predicted in the correct position and with
the correct relative intensities. The most obvious, and expected, difference is due to
the presence of the large peak in each spectrum between 1600 and 1700 cm-1 that
originates from O-H bending motions of solvent water molecules. This difference is,
of course, due to the PCM not incorporating individual water molecules. In the case
of the ROA spectrum for linear L-Ser-L-Ser, shown in Figure 6.3, most bands are
correctly predicted in terms of position, sign and even relative intensity. The details
of the computed ROA spectrum of cyclic L-Ser-L-Ser in H2O are not in as close
agreement with those of the experimental ROA spectrum, particularly in terms of
146
Chapter 6
their relative intensities. However, even here most ROA features are correctly
predicted in terms of position and sign.
Although solvent interactions have been approximated using a PCM, the calculations
are clearly accurate enough to allow the origins of specific bands to be determined.
These are detailed in Tables 6.1 and 6.2, for linear and cyclic L-Ser-L-Ser in H2O
and D2O, respectively. The most significant structural information which is obtained
in terms of the vibrational modes is commonly considered in terms of the amide I, II
and III vibrations. The C=O stretching mode of peptides is largely responsible for
bands appearing in the amide I region from 1600-1700 cm-1. The amide II region at
~1510-1570 cm-1, arising from the combination of C-N and N-H in-plane bending is
not usually as rich in structural information for peptides and proteins in contrast to
the amide I and III regions. Commonly for peptides, the most informative region is
the amide III from ~1230- 1340 cm-1, in which bands originate from the coupling
between N-H deformations and Cα-H stretching [25,26]. However, the spectra of
cyclic dipeptides appear to contain larger features in other spectral regions than are
typically observed for the corresponding linear forms; therefore here we discuss the
features observed across a much broader spectral range.
In the Raman spectra of linear and cyclic L-Ser-L-Ser the bands at 1642 and 1671
cm-1 are both assigned to the amide I vibration. Upon deuteration these bands are
shifted, upwards in wavenumber for the linear form by 10 cm-1 and down shifted by
22 cm-1 in the case of the cyclic form, due to the amide I mode coupling with N-H
bending vibrations as shown previously [14,27]. The corresponding ROA bands for
the amide I region appear as a doublet of positive peaks at 1666 and 1638 cm-1 for
linear L-ser-L-ser that appears with a negative sign at 1642 cm-1 in the calculated
spectrum. The amide I band appears as a weak negative peak at 1683 cm -1 for the
147
Chapter 6
cyclic form which is shifted to lower wavenumber at 1654 cm-1 in D2O. The band at
1519 cm-1 in the Raman spectrum of cyclic L-Ser-L-Ser is assigned to a mixed
vibrational mode involving C-N, C-C, C=O, N-H groups and the amide II mode and
gives rise to a large positive peak at the same position in the ROA spectrum. No
corresponding band is found in the Raman spectrum of the linear form which
suggests that this band is unique for the cyclic form of dipeptides as it was shown
previously for cyclic L-Ala-L-Ala (Chapter 5). This Raman band shifts to lower
wavenumber upon deuteration by 20 cm-1, which confirms that this band has a
significant contribution from N-H deformations, as was observed previously for
other cyclic dipeptides [14,27]. The calculated Raman and ROA spectra for cyclic LSer-L-Ser are in good agreement with the experimental results for the amide II
region. The rest of the bands in the 1400- 1500 cm-1 region relate mainly to C-H
bending motions. Aside from these, the Raman bands at 1469 and 1404 cm-1 for the
linear and cyclic forms, respectively, can be assigned to vibrations that are mainly
from CH2 and N-H bending motions. The corresponding ROA bands appear as a ve/+ve couplet at 1417/1452 cm-1 for the linear form and appear as a negative doublet
at 1427 and 1457 cm-1 for cyclic L-Ser-L-Ser. The observed peak position shifts by
~3-10 cm-1 when going from H2O to D2O solution which confirms the contribution to
this band from the N-H vibrational mode.
The strong band at 1482 cm-1 in the ROA spectrum of cyclic L-Ser-L-Ser is mainly
assigned to the amide II vibration which is unusually intense compared to that
observed for most peptides, and does not appear for the linear form here. This
suggests that the appearance of the amide II region changes significantly when LSer-L-Ser undergoes cyclization where the molecule adopts a much more rigid
structure.
148
Chapter 6
For linear L-Ser-L-Ser the Raman bands at 1378 and 1320 cm-1 correspond to the
three +ve ROA bands at 1384, 1343 and 1304 cm-1 that are mainly assigned to C-OH and C-H bending. These bands shift downward by 3 cm-1 in the Raman spectra of
cyclic L-Ser-L-Ser and yield a +ve/-ve/+ve triplet of ROA peaks at 1394, 1345 and
1322 cm-1 that are assigned to CH2 wagging vibrations as well as the amide III mode,
as shown in Table 6.1. The Raman spectrum of linear L-Ser-L-Ser also contains two
bands at 1278 and 1252 cm-1 that give rise to two weak +ve/-ve couplets in the ROA
spectrum at 1283/1268 and 1252/1239 cm-1, respectively, which are correctly
predicted in the calculated spectra though with lower intensity and a more negative
bias in terms of ROA bandshape. These features arise from C=O, C-H and CH2
vibrations. In contrast, the ROA spectrum of cyclic L-Ser-L-Ser exhibits a +ve/+ve/ve triplet at 1294/1268/1239 cm-1 in this region, with the sign flipped for the peaks at
1294 and 1226 cm-1 in the deuterated sample’s ROA spectrum.
The observed bands within this region of vibrational spectra mainly originate from
N-Cα, C-C and C-O stretching along with CH2 rocking vibrations. The band at 1154
cm-1 in the Raman spectrum of linear L-Ser-L-Ser shifts upward by 10 cm-1 in the
spectrum of the cyclic form, whereas in the spectra measured in D2O it is shifted
downward by 27 cm-1. This band was also associated with the same vibration for the
cases of cyclic Gly-Gly and L-Ala-L-Ala [12, 25]. The –ve ROA band at 1064 cm-1
that mainly originates from C-O and CH2 stretching vibrations in the spectrum of
cyclic L-Ser-L-Ser changes its sign to +ve and moves to 1067 cm-1 for the linear
form, in agreement with the calculated spectra of both forms. The remaining bands
below 1000 cm-1 originate from a mixture of C-C-H, C-O, C-C, N-H, C-C-O and
CH2 vibrations but these only give rise to two strong ROA bands for linear L-Ser-LSer at 761 and 905 cm-1. More ROA bands are observed in the case of the cyclic
149
Chapter 6
form, in particular the -ve/+ve bands at 894/782 cm-1, presumably arising because the
cyclic form is more rigid.
The bands below 700 cm-1 belong to two amide group vibrations of N-H out of plane
bending and C=O bending modes with contributions from the ring (i.e. in the case of
cyclic form) and C-C bending. Other vibrations at this region that only occur for
cyclic L-Ser-L-Ser include ring deformations, ring stretching and O-H torsions.
Though these vibrations arise in this region, in the spectrum of cyclic L-Ser-L-Ser
we only observe a –ve ROA band at 717 cm-1 which is assigned to amide VI and ring
stretching modes.
6.6 Conclusions
Cyclic dipeptides possess interesting and beneficial biological activities, which
mostly have unknown biological functions. The application of Raman spectroscopy
and Raman optical Activity (ROA) to the study of bioactive peptides has matured
over the past decade to a high level of sophistication where it provides useful
information regarding the biological conformations of these species. In the present
study, several Raman and ROA marker bands have been identified when L-Ser-L-Ser
undergoes cyclization. ROA spectra, being sensitive to stereochemistry, are found to
be more sensitive to cyclization with ROA marker bands being observed at 1683,
1519, 1482, 1345, 1080, 1064, 894, 782 and 716 cm-1. Although less sensitive,
Raman spectra also appear to contain bands that identify cyclization, these
principally being at 1519, 1469, 1317, 1164, 845 and 670 cm-1.
It is unusual to observe strong signals in the amide II region for dipeptides as the
other amide regions generally provide much more structural information for proteins
and peptides. The amide II band appeared in the ROA spectrum of cyclic L-Ser-LSer at 1519 cm-1 which corresponds closely to its appearance for cyclic L-Ala-L-Ala
150
Chapter 6
at 1521 cm-1 (Chapter 5). It is apparent that this band is unique for cyclic dipeptides
since it was obtained in both Raman and ROA spectra of cyclic L-Ser-L-Ser and LAla-L-Ala. Other similarities were also observed in Raman and ROA spectra of
cyclic L-Ala-L-Ala, in particular, similarities within the Raman bands positions at
1317 and 1459 cm-1 and ROA bands at 790 and 1067 cm-1. Together, these studies
show that Raman and ROA spectra do provide sensitive markers of peptide
cyclization that can be used to easily differentiate them from conventional linear
peptides.
151
Chapter 6
6.7 References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
K. McCleland, P. J. Milne, F. R. Lucieto, C. Frost, S. C. Brauns, M. V. D.
Venter, J. Du Plessis, K. Dyason, Journal of Pharmacy and Pharmacology.
2004, 56, 1143-1153.
M. B. Martins, I. Carvalho, Tetrahedron. 2007, 63, 9923-9932.
J. A. Trischman, R. E. Oeffner, M. J. D. Luna, M. Kazaoka, Marine
Biotechnology. 2004, 6, 215-220.
S. W. Yang, T. M. Chan, J. Terracciano, D. Loebenberg, G. D. Chen, M.
Patel, V. Gullo, B. Pramani, M. Chu, Journal of Antibiotics. 2004, 57, 345347.
S. C. Brauns, P. Milne, R. Naude, M. Van de Venter, Anticancer Research.
2004, 24, 1713-1719.
L. D. Barron, Current Opinion in Structural Biology. 2006, 16, 638-643.
L. D Barron, L. Hecht, Bimolecular conformational studies with vibrational
Raman optical activity. In Biomolecular Spectroscopy; R. J. H. Clark, R. E.
Hester, Editors; Wiley: Chichester, 1993; Part B, pp 235.
T. B. Freedman, L. A. Nafie, T. A. Keiderling, Biopolymers. 1995, 37, 265279.
L. A. Nafie, Applied Spectroscopy. 1996, 50, 14A-26A.
L. D. Barron, L. Hecht, Vibrational Raman optical activity: From
fundamentals to biochemical applications. In Circular Dichroism, Principles
and Applications; K. Nakanishi, N. Berova, R. W. Woody, Editors; VCH
Publishers: New York, 1994; pp 179.
K. Ruud, A. J. Thorvaldsen, Chirallity. 2009, 21, E54-E67.
C. Hermann, K. Ruud, M. Reiher, Chemical Physics. 2008, 343, 200-209.
G. G. Fava, M. F. Belicchi, Acta Crystallographica. 1981, 1337, 625-629.
A. P. Mendham, T. J. Dines, M. J. Snowden, B. Z. Chowdhry, R. Withnall,
Journal of Raman Spectroscopy. 2009, 40, 1478-1497;
A. P. Mendham, T. J. Dines, M. J. Snowden, R. Withnall, B. Z. Chowdhry,
Journal of Raman Spectroscopy. 2009, 40, 1508-1520.
M. Pecul, Chemical Physics Letters. 2006, 427, 166–176.
V. W. Jurgensen, K. Jalkanen, Physical Biology. 2006, 3, S63–S79
Gaussian 09, Revision A.1, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J.
B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V.
Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
A. D. Becke, Journal of Chemical Physics. 1993, 98, 5648-5652
C. Lee, W. Yang, R. G. Parr, Physical Review B. 1988, 37, 785-789.
152
Chapter 6
21.
22.
23.
24.
25.
26.
27.
T. H. Dunning, Journal of Chemical Physics. 1989, 90, 1007-1021.
M. T. Cances, V. Mennucci, J. Tomasi, Journal of Chemical Physics. 1997,
107, 3032-3041.
D. M. York, M. Karplus, Journal of Physical chemistry A. 1999, 103, 1106011079.
J. A. Schachtschneider, Vibrational Analysis of Polyatomic Molecules, Parts
V and VI, Technical Report Nos. 231 and 57, Shell Development Co.,
Houston TX, 1964 and 1965.
R. Schweitzer-Stenner, Journal of Raman Spectroscopy. 2001, 32, 711-732.
N. G. Mirkin, S. Krimm, Journal of Molecular Structure. 1991, 242, 143160.
T. C. Cheam, S. Krimm, Spectrochimica Acta Part A: Molecular
Spectroscopy. 1984, 40, 481-501.
153
Chapter 6
Figure 6.1: The chemical structure with atom numbering scheme (left) and calculate
minimum energy conformation (right) of linear L-Ser-L-Ser.
H 21
O7
O 16
H
19 O
C
4
17H
9
6
C
N
8
C3
H15
H5
N
H2
C
10
1
H 22
C
13
H18
11
H 20
H12
C
O14
Figure 6.2: The chemical structure with atom numbering scheme (left) and calculate
minimum energy conformation (right) of cyclic L-Ser-L-Ser.
154
Chapter 6
Exp. Linear L-Ser-L-Ser
9
6x10
R
I +I
L
Calc. Linear L-Ser-L-Ser
Exp. Cyclic L-Ser-L-Ser
9
4.2x10
Calc. Cyclic L-Ser-L-Ser
500
600
700
800
900 1000 1100 1200 1300 1400 1500 1600 1700 1800
Exp. Linear L-Ser-L-Ser
Calc. Linear L-Ser-L-Ser
R
I -I
L
Exp. Cyclic L-Ser-L-Ser
Calc. Cyclic L-Ser-L-Ser
500
600
700
800
900 1000 1100 1200 1300 1400 1500 1600 1700 1800
-1
Wavenumber (cm )
Figure 6.3: Experimental and computed Raman (top) and ROA (bottom) spectra of
linear (pH= 7.0) and cyclic L-Ser-L-Ser (pH= 7.0) in aqueous solution. The
concentration for each sample was 50 mg/ml and laser power was 1.2 W at the laser.
The marker bands that are induced upon cyclization are highlighted by shading.
155
Chapter 6
Exp. Linear L-Ser-L-Ser
9
4.7x10
R
I +I
L
Calc. Linear L-Ser-L-Ser
Exp. Cyclic L-Ser-L-Ser
Calc. Cyclic L-Ser-L-Ser
9
1.1x10
500
600
700
800
900 1000 1100 1200 1300 1400 1500 1600 1700 1800
Exp. Linear L-Ser-L-Ser
7
5.7x10
R
I -I
L
Calc. Linear L-Ser-L-Ser
Exp. Cyclic L-Ser-L-Ser
5
1.7x10
Calc. Cyclic L-Ser-L-Ser
500
600
700
800
900 1000 1100 1200 1300 1400 1500 1600 1700 1800
-1
Wavenumber (cm )
Figure 6.4: Experimental and computed Raman (top) and ROA (bottom) spectra of
linear (pH= 7.0) and cyclic L-Ser-L-Ser (pH= 7.0) in D2O. The concentration for
each sample was 50 mg/ml and the laser power was 0.6 W at the sample. The marker
bands that are induced upon cyclization are highlighted by shading.
156
Chapter 6
Table 6.1: Calculated and experimental Raman and ROA bands for cyclic and linear L-Ser-L-Ser in H2O.
Experimental
Linear
ROA
Raman
Cyclic
ROA
1666(+ve)
1638(+ve)
1565(+ve)
1452(+ve)
1417(-ve)
1384(+ve)
1363(+ve)
1343(+ve)
1304(+ve)
1283(+ve)
1268(+ve)
1252(+ve)
1239(+ve)
1178(+ve)
1151(+ve)
1116(+ve)
1094(+ve)
1067(+ve)
1039(+ve)
1017(+ve)
964(+ve)
956(+ve)
905(-ve)
1683(-ve)
1519(+ve)
1482(+ve)
1457(-ve)
1427(-ve)
1394(+ve)
1345(-ve)
1322(+ve)
1294(+ve)
1268(+ve)
1239(-ve)
1154(+ve)
1116(-ve)
1083(+ve)
1064(-ve)
1048(+ve)
1025(+ve)
998(+ve)
922(+ve)
1642
1469
1404
1378
1320
1278
1252
1154
1080
1048
992
964
-
Calculated
Linear
Cyclic
Band assignments
Linear
Cyclic
1655
1642
1612
1562
1558
1494
1479
1466
1446
1429
1397
1375
1343
1314
1303
1300
1267
1251
1243
1195
1163
1143
1108
1087
1062
1040
1009
969
954
924
as(NH3) (89)
(CO) (66), (C6N5) (13)
as(NH3) (75)
(C6N5) (33), ip(NH) (53)
as(CO2) (91)
(CH2′) (98)
(CH2) (98)
s(NH3) (86)
(C13O22H23) (13), (CH2′) (73)
(C10O18H29) (18), (CH2) (58)
(C8H14) (19),′(C8H14) (41)
s(CO2) (56),(C1C4) (10)
′(C4H11) (26),′(C8H14) (23)
′(C4H11) (15),(C8H14) (23), (CH2′) (22)
(C4H11) (20),(CH2′) (38)
(C10O18H29) (12),(C4H11) (16),′(C4H11) (12),τ(CH2′) (17)
ip(NH) (16), (CH2) (35)
(C13O22H23) (29), (CH2) (18), (CH2′) (14)
-
(C6N5) (11), ip(NH) (23),(CH2) (15)
(C10O18H29) (43),(C4H11) (12),′(C4H11) (16), ω(CH2) (13)
op(NH3) (20), (CH2) (28)
(C4N5) (29),(CH2) (23)
(C8N9) (11),ip(NH3) (24),(C8H14) (18)
(C4C10) (27),(C10O18) (35)
(C8N9) (21),op(NH3) (10)
(C13O22) (83)
(C4N5) (12),(C10O18) (13),(CH2) (24) 
(C8C13) (14),op(NH3) (27), (CH2′)(32)
(C6C8) (24), ip(NH3) (11)
(C4C10) (17), (C10O18) (33),(CH2) (11)
(CO) (53), (NC) (15), ip(NH) (19) [amide I]
(CO) (53), (NC) (12), ip(NH) (19) [amide I]
-
-
(CO) (23), (NC) (18), (CC) (12), ip(NH) (17) [amide II]
(CO) (24), ip(NH) (31), (CH2) (33) [amide II]
(NC) (34), (CH2) (35)
(CH2) (91)
(NC) (15), ip(NH) (25), (CH2) (30), (CH2) (12) [amide III]
ip(NH) (22), (CH2) (83)
(COH) (22),ip(NH) (18),(CH2) (32) [amide III]
(COH) (20),(CCH) (12),(CH2) (32),(CH2) (26)
(COH) (14),(CH2) (43),(CH2) (20)
(NC) (10), (CCH) (12),(CCH) (36),(CCH) (15)
(NC) (10), (NC) (16), (CCH) (31),(CCH) (13)
-
(CCH) (39),(CCH) (16) 
(CCH) (45),(CCH) (16)
-
(COH) (27),(CCH) (14),(CH2) (44)
(COH) (18), (CCH) (10),(CCH) (21), (CH2) (36)
-
-
(NC) (29),(COH) (12), (CH2) (16)
(NC) (36), (C3C4) (14),(C-O) (11),ip(CO) (10)
(C3C4) (19),(C-O) (25),(CH2) (13)
(C3C4) (28),(C-O) (59)
(C-O) (40) ,(CCH) (10),(CH2) (10)
(NC) (10), (C3C4) (12),(C-O) (12), (CH2) (39)
-
(C3C4) (19),(C-O) (10), (CH2) (28)
ip(ring-3) (28), (CH2) (33)
Raman
1671
1519
1469
1404
1374
1317
1281
1239
1164
1080
992
952
-
1661
1658
1540
1493
1476
1472
1456
1414
1395
1382
1372
1332
1321
1283
1277
1224
1236
1153
1110
1067
1069
1024
980
946
935
157
Chapter 6
877(+ve)
(12)
761(+ve)
640(-ve)
534(+ve)
882
894(-ve)
-
884
853
(C8N9) (36), (C8C13) (18), (CH2′) (10)
816
784
670
-
782(+ve)
717(-ve)
682(+ve)
664(+ve)
631(-ve)
616(+ve)
577(-ve)
-
845
670
622
-
848
792
752
709
641
570
575
524
844
807
736
682
645
644
621
543
510
(C1C4) (15),(CO2) (12)
(C-O) (16),(CCH) (14),(CCH) (18),op(CO) (36)
-
(CC) (29), (NC) (12)
(CO2) (27), (CO2) (24)

(C6C8C13) (11), ip(CO) (18), op(CO) (32) 
(C3C4) (18), ip(ring-3) (32), op(CO) (32) [amide VI]
ip(CO) (16), op(CO) (30)

(C4C10) (10), (N5C4C1) (22),(CO2) (23), (CO2) (10), (CO2) (147) op(NH) (56), op(CO) (35) [amide V,VI]
-
(CCO) (10), op(NH) (77), op(CO) (12) [amide V]
-
(NC) (12), (CC) (38), op(NH) (23)
op(NH) (34),(N5C6) (30)
(CC) (11), ip(CO) (47), op(NH) (12) [amide IV]
(C1C4) (13),(N5C6C8) (13), (CO2) (15)
(CCO) (21),op(NH) (23),op(CO) (34)
(C4C10O18) (31), (CO2) (15)
(CCO) (23),op(NH) (12), op(CO) (20),op(ring-2) (10)
158
(CC) (14), (C3C4) (26),(C-O) (17), (CCH) (10),op(CO)
Chapter 6
Table 6.2: Calculated and experimental Raman and ROA bands for cyclic and linear L-Ser-L-Ser in D2O.
Experimental
Linear
ROA
Raman
Cyclic
ROA
Raman
1654(+ve)
1556(+ve)
1482(+ve)
1467(-ve)
1449(+ve)
1429(+ve)
1391(+ve)
1379(-ve)
1366(+ve)
1338(+ve)
1325(-ve)
1291(+ve)
1249(+ve)
1215(+ve)
1140(+ve)
1086(-ve)
1064(+ve)
1003(+ve)
987(+ve)
931(+ve)
-
1654(-ve)
1575(+ve)
1546(+ve)
1479
1457(-ve)
1439(+ve)
1417(+ve)
1384(+ve)
1348(+ve)
1320(+ve)
1294(-ve)
1268(+ve)
1226(+ve)
1202(+ve)
1167(+ve)
1132(+ve)
1108(-ve)
1048(-ve)
970(+ve)
922(+ve)
-
1649
1499
1469
1386
1338
1294
1270
1207
1127
1075
989
964
905
-
1654
1560
1469
1449
1427
1376
1358
1327
1294
1249
1212
1140
1094
1064
1006
992
928
908
-
Calculated
Linear Cyclic
Band assignments
Linear
1646
1558
1495
1493
1478
1429
1410
1376
1367
1332
1301
1287
1274
1252
1181
1175
1152
1144
1128
1096
1074
1046
1039
1015
985
931
912
896
ν (CO) (72), ν(C6N5) (17)
ν(CO) (72), ν(NC) (13) [amide I]
ν (CO) (68), ν(NC) (14) [amide I]
νas (CO2) (92)
ν (NC) (35), ν (CC) (14), δ(CH2) (12) [amide II]
ν (C6N5) (31), δip(ND) (11),δ (CH2′) (26)
ν (C6N5) (10), δ(CH2′) (74)
δ (CH2) (98)
δ (CH2) (98)
δ (CH2) (87), ν(CO) (12), ν(CC) (10), ν(NC) (49) [amide II]
ω (CH2′) (98)
νs (CO2) (21), ω (CH2) (57)
ω (CH2) (80)
δ (C8H14) (10), δ′(C8H14) (44), ω(CH2) (12)
ω (CH2) (88)
νs (CO2) (39),δ′(C8H14) (14),ω(CH2) (20)
ω (CCH) (25),τ(CH2) (33)
νs (CO2) (11),δ(C4H11) (15),δ′(C4H11) (40)
ν (NCα) (20), ω(CCH) (17),τ(CH2) (37)
δ(C8H14) (29), τ(CH2′) (422)
δ(CCH) (28),ω(CCH) (12),ρ(CCH) (11),τ(CCH) (10),τ(CH2) (16)
δ (CCH) (32),ρ(CCH) (12),τ(CH2) (17)
δ (C4H11) (23), δ′(C4H11) (14), τ(CH2) (19),τ(CH2′) (15)
δ (CCH) (16),τ(CH2) (32)
δ (C8H14) (23), δ′(C8H14) (16),τ(CH2′) (20)
δ (CCH) (26), τ(CH2) (27)
δ (C4H11) (28),τ(CH2) (56)
ν (NC) (17), ν(CC) (10), δip(ND) (13), ω(CCH) (13),τ(CCH) (19)
ν (NCα) (23), δip(ND) (28), δ(CCH) (12), δip(CO) (16) [amide III]
δas (ND3) (99)
δs (ND3) (16),δas(ND3) (66)
ν (C8N9) (13),ν(C6C13) (29),δs(ND3) (14)
ν (C4N5) (17),δas(ND3) (20), ρ(CH2′) (18)
δ (COD) (13), τ(CCH) (24),ρ(CH2) (39)
ν (C4N5) (15),δs(ND3) (11),δas(ND3) (17), ρ(CH2) (17)
δ (COD) (15), δip(ND) (19), ρ(CH2) (35)
ν (C4C10) (18),ν(C10O18) (25), ρ(CH2) (11)
δs (ND3) (25), ρ(CH2′) (14)
ν (C3C4) (26),ν(C-O) (58)
ν (C13O22) (60)
ν (C13O22) (16),δ(C10O18D29) (16)
ν (C6C8) (20),ν(C10O18) (18),δ(C8H14) (10)
ν (C6N5) (10), δip(ND) (44)
ν (C3C4) (11),ν(C-O) (12),δ(COD) (13), ρ(CCH) (13),δip(ring-3) (22)
δ (COD) (22), δip(ND) (18), ρ(CCH) (10)
ν (NCα) (23),ν(C-O) (11),δ(COD) (21),δip(ND) (21)
ν (C4N5) (14),ν(C10O18) (13),δ(C10O18D29) (16), δip(ND) (11), ρ(CH2)(15)
ν(NCα) (15), ν(C3C4) (22),ν(C-O) (18), δip(ND) (22) [amide III]
ν (C8N9) (11),δ(C13O22D23) (52)
ν (C1C4) (11),ν(C10O18) (19),δ(C10O18D29) (37)
-
1637
1635
1513
1477
1471
1394
1392
1348
1344
1316
1311
1280
1270
1238
1227
1132
1103
1067
988
984
959
952
-
159
Cyclic
Chapter 6
885(-ve)
845(+ve)
811(+ve)
784(+ve)
640(-ve)
556(-ve)
-
882
744
676
640
-
871(+ve)
845(-ve)
767(+ve)
697(+ve)
646(+ve)
601(+ve)
541(-ve)
519(-ve)
845
688
601
-
875
871
820
782
771
731
660
636
551
519
871
861
835
829
769
722
658
617
573
569
-
ρop (ND3) (16)
ν (C8N9) (20),ν(C1C4) (10),ρ(CH2′) (18)
ρip (NH3) (18), δip(CO) (10), δ(CO2) (10)
ν (C8N9) (10),ν(C8C13) (14),δ(C13O22D23) (11), ρop(ND3) (39)
δ (CO2) (23), ω(CO2) (31)
ν (C8C13) (14),δop(CO) (50)
ν(C6C8) (14),ρip(ND3) (19),δ(C8C13O22) (11),δip(CO) (19), δop(CO) (12)
ν (C1C4) (10), ν(C4C10) (11), δ(N5C4C1) (18),δ(CO2) (25), ω(CO2) (15)
ν (C1C4) (12),δ(N5C6C8) (14),δ(CO2) (12), ρ(CO2) (14)
δ (N5C4C10) (11),δ(C4C10O18) (31), ω(CO2) (15)
160
δ (COD) (30), ρ(CH2) (29)
δ(COD) (35),δip(ring-3) (11), ρ(CH2) (31)
ν (C3C4) (20), ν(C-O) (15), δ(CCH) (13), ρ(CCH) (11), δop(CO) (16)
ν(C-O) (10),δ(COD) (16),δ(CCH) (11), ρ(CCH) (14), δop(CO) (28)
ν(CC) (37), ν(NC) (13), δip(ND) (20)
ν(C3C4) (19), δip(ring-3) (26), δop(CO) (32)
ν(NCα) (10), ν(CC) (16),ν(NC) (12), ν(C3C4) (16),δ op(CO) (28) [amide VI]
ν (CC) (35), δip(CO) (19)
δ (CCO) (28),δop(ND) (11),δop(CO) (46) [amide VI]
δ(CCO) (13), δip(CO) (30), δop(ND) (29), δop(CO) (20) [amide IV]
-
Chapter 6
6.8 Supplementary information
Table S6.1: Calculated and experimental bond lengths (Å) for cyclic and linear LSer-L-Ser.
Linear L-Ser-L-Ser
Calculated bond lengths (Å)
r(C1O3)
r(O2C1)
r(C4N5)
r(C4C1)
r(N5H12)
r(C6N5)
r(C6C8)
r(O7C6)
r(C8C13)
r(C8H14)
r(N9H15)
r(N9C8)
r(C10O18)
r(C10C4)
r(H11C4)
r(C13O22)
r(C13H25)
r(H16N9)
r(H17N9)
r(H19O18)
r(H20C10)
r(H21C10)
r(O22H23)
r(H24C13)
1.2603
1.2647
1.4577
1.5598
1.0127
1.3369
1.5353
1.2429
1.5286
1.0932
1.0245
1.4955
1.4387
1.5268
1.096
1.4329
1.0977
1.0273
1.0218
0.9659
1.099
1.0974
0.9661
1.0976
Cyclic L-Ser-L-Ser
r(N1-C3)
r(H2-N1)
r(C3-C6)
r(C3-H5)
r(C4-H15)
r(C4-H17)
r(C4-C3)
r(C6-O7)
r(C6-N8)
r(O16-C4)
r(O16-H21)
1.4605
1.0153
1.5216
1.0995
1.0997
1.0994
1.5395
1.2439
1.3407
1.4285
0.9679
Table S6.2: Calculated bond angles (o) for cyclic and linear L-Ser-L-Ser.
Linear L-Ser-L-Ser
Calculated bond angles (o)
θ(O3C1O2)
θ(O3C1C4)
θ(O2C1C4)
θ(N5C4C1)
θ(C4N5H12)
θ(C4N5C6)
126.09
118.94
114.97
113.61
116.88
123.88
161
Chapter 6
θ(N5C4C10)
θ(N5C4H11)
θ(C1C4C10)
θ(C1C4H11)
θ(H12N5C6)
θ(N5C6C8)
θ(N5C6O7)
θ(C8C6O7)
θ(C6C8C13)
θ(C6C8H14)
θ(C6C8N9)
θ(C13C8H14)
θ(C13C8N9)
θ(C8C13O22)
θ(C8C13H25)
θ(C8C13H24)
θ(H14C8N9)
θ(H15N9C8)
θ(H15N9H16)
θ(H15N9H17)
θ(C8N9H16)
θ(C8N9H17)
θ(O18C10C4)
θ(C10O18H19)
θ(O18C10H20)
θ(O18C10H21)
θ(C10C4H11)
θ(C4C10H20)
θ(C4C10H21)
θ(O22C13H25)
θ(C13O22H23)
θ(O22C13H24)
θ(H25C13H24)
θ(H16N9H17)
θ(H20C10H21)
109.25
107.76
111.82
106.72
118.93
115.92
124.64
119.44
111.47
111.63
107.46
110.07
108.93
106.79
108.77
110.15
107.12
109.82
109.75
108.36
109.07
111.95
108.37
108.19
110.12
110.7
107.38
109.41
109.08
111.25
108.44
110.64
109.21
107.85
109.14
Cyclic L-Ser-L-Ser
θ(C3-N1-H2)
θ(N1-C3-C6)
θ(N1-C3-H5)
θ(N1-C3-C4)
θ(C3-N1-C13)
θ(H2-N1-C13)
θ(C6-C3-H5)
θ(C6-C3-C4)
θ(C3-C6-O7)
θ(C3-C6-N8)
θ(H5-C3-C4)
θ(H15-C4-C3)
θ(H15-C4-O16)
θ(H17-C4-C3)
θ(H17-C4-H15)
θ(H17-C4-O16)
θ(C3-C4-O16)
θ(O7-C6-N8)
θ(C4-O16-H21)
116.54
113.27
108.27
111.05
128.01
115.42
105.92
111.2
119.33
118.32
106.73
108.52
110.83
108.86
108.57
106.59
113.29
122.25
108.58
162
Chapter 6
Table S6.3: Calculated torsion angles (o) for cyclic and linear L-Ser-L-Ser
Linear L-Ser-L-Ser
Calculated torsion angles (o)
τ(O3C1C4N5)
τ(O3C1C4C10)
τ(O3C1C4H11)
τ(O2C1C4N5)
τ(O2C1C4C10)
τ(O2C1C4H11)
τ(H12N5C4C1)
τ(C6N5C4C1)
τ(C10C4N5H12)
τ(H11C4N5H12)
τ(C10C4N5C6)
τ(H11C4N5C6)
τ(C4N5C6C8)
τ(C4N5C6O7)
τ(N5C4C10O18)
τ(N5C4C10H20)
τ(N5C4C10H21)
τ(C1C4C10O18)
τ(C1C4C10H20)
τ(C1C4C10H21)
τ(H12N5C6C8)
τ(H12N5C6O7)
τ(N5C6C8C13)
τ(N5C6C8H14)
τ(N5C6C8N9)
τ(C13C8C6O7)
τ(H14C8C6O7)
τ(N9C8C6O7)
τ(C6C8C13O22)
τ(C6C8C13H25)
τ(C6C8C13H24)
τ(C6C8N9H15)
τ(C6C8N9H16)
τ(C6C8N9H17)
τ(O22C13C8H14)
τ(H25C13C8H14)
τ(H24C13C8H14)
τ(O22C13C8N9)
τ(H25C13C8N9)
τ(H24C13C8N9)
τ(C13C8N9H15)
τ(C13C8N9H16)
τ(C13C8N9H17)
τ(C8C13O22H23)
τ(H14C8N9H15)
τ(H14C8N9H16)
τ(H14C8N9H17)
τ(H19O18C10C4)
τ(O18C10C4H11)
τ(H20C10O18H19)
τ(H21C10O18H19)
τ(H20C10C4H11)
τ(H21C10C4H11)
τ(H23O22C13H25)
τ(H24C13O22H23)
-7.15
117.09
-125.76
172.54
-63.22
53.93
86.21
-87.41
-39.41
-155.77
146.97
30.61
175.65
-3.59
61.15
-58.94
-178.26
-65.5
174.41
55.09
2.15
-177.09
-85.71
37.84
155.01
93.57
-142.88
-25.71
-175.76
64.09
-55.57
157.23
36.92
-82.35
59.81
-60.34
-180
-57.36
-177.51
62.84
36.33
-83.97
156.76
161.52
-82.7
156.99
37.73
-173.4
177.75
-53.75
67.02
57.66
-61.66
-79.94
41.64
163
Chapter 6
Cyclic L-Ser-L-Ser

(C6-C3-N1-H2)
(H5-C3-N1-H2)
(C4-C3-N1-H2)
(C13-N1-C3-C6)
(N1-C3-C6-O7)
(N1-C3-C6-N8)
(C13-N1-C3-H5)
(C13-N1-C3-C4)
(N1-C3-C4-H17)
(N1-C3-C4-H15)
(N1-C3-C4-O16)
(C3-N1-C13-C10)
(C3-N1-C13-O14)
(H2-N1-C13-C10)
(H2-N1-C13-O14)
(O7-C6-C3-H5)
(N8-C6-C3-H5)
(O7-C6-C3-C4)
(N8-C6-C3-C4)
(C6-C3-C4-H17)
(C6-C3-C4-H15)
(C6-C3-C4-O16)
(H5-C3-C4-H17)
(H5-C3-C4-H15)
(H5-C3-C4-O16)
(H17-C4-O16-H21)
(C3-C4-O16-H21)
(H15-C4-O16-H21)
-171.89
-54.73
62.14
5.93
173.49
-6.34
123.09
-120.04
-174.81
-56.75
66.79
0.57
-179.26
178.42
-1.41
54.97
-124.86
-60.61
119.55
58.07
176.14
-60.33
-57.01
61.06
-175.41
178.27
-62.03
60.23
164
Chapter 7
Chapter 7
Conclusion and Future work
165
Chapter 7
7.0 Conclusion
Raman, ROA and SERS are spectroscopic techniques that constitute a powerful
analytical toolkit for deriving quantitative and qualitative biomolecular information.
Application of these techniques in combination is more advantageous, since they can
provide more structural information on the nature of biomolecules along with the
chirality of the molecules. As demonstrated in different chapters of this thesis, more
marker bands were obtained in the ROA spectra of the sugar, ribonucleotides and
short dipeptides in contrast to the Raman and SERS spectra, which confirm the
sensitivity of ROA to structural changes. However, ROA is a weak effect and limited
to biological molecules that are only available at relatively low concentrations. These
limitations could be overcome by employing SERS. As shown in this thesis,
obtaining SERS is more complex than measuring the parent Raman spectra since a
range of factors needs to be controlled in order to obtain reliable measurements.
Small changes in the experimental conditions can lead to a huge variation in the
SERS signals.
The reproducibility of SERS was one of the main issues for obtaining consistent
SEROA results as unstable substrates may cause a loss of signal enhancement over
time. In order to measure reliable SEROA spectra, it was crucial to stabilise the
aggregation process and indeed any other dynamic process occurring in the sample.
One solution to this problem was to suspend the colloidal particles in a gel and halt
or greatly slow down the aggregation process. The time-dependence of SERS affects
the optimum conditions drastically since the extent of the aggregation over the
measuring period influences the fluctuations in the bands measured and the
subsequent difficulty in controlling the reproducibility of spectra. This thesis presents
a number of studies confirming SEROA as a chiroptical technique by controlling the
166
Chapter 7
SERS experimental setup and its time-dependence. This was mainly achieved by
using citrate- and hydroxylamine-reduced silver colloids along with a polycarbophil
polymer as stabilising media in solution for studying SERS and SEROA of L- and Dribose and L- and D-tryptophan. The addition of polymer to the colloidal system was
shown to control the aggregation process in SERS solution along with protecting the
Ag nanoparticles from further aggregation. SEROA spectra measured for both
enantiomers of each of ribose and tryptophan clearly showed mirror image bands for
the majority of spectral regions where the actual signals were clearly distinguishable
from the background noise. This work presents the first observation of mirror image
SEROA bands, an essential step in the proof of this new technique [1].
Another system to provide enhancement for the ROA weak effect was that of silver
silica nanotags prepared by collaborators at the University of Strathclyde which act
as single plasmonic nanoparticle substrates. These silver nanoparticles which are
conjugated to benzotriazole dye molecules have been used as an approach to enhance
the sensitivity of SERS and allow colorimetric detection of analytes at relatively low
concentrations. It was shown in this thesis that they can also function as chiral
plasmonic nanomaterials by providing superchiral electromagnetic fields for
ultrasensitive detection of biomolecular conformation. The chiroptical activity of this
system was verified by measurement of SERROA spectra of the enantiomers of
ribose and tryptophan when attached to silver silica nanotags These spectra clearly
showed that SERROA bands of opposing sign were obtained from the two
enantiomers of each chiral molecule and that these were signatures of the chiroptical
nature of the interaction between these analytes and the surface plasmons of the dyetagged nanoprobes, as a result of a chiral influence on the SERRS spectrum of the
benzotriazole dye. Our findings demonstrated that chirality was induced into the
167
Chapter 7
achiral plasmonic surface of the substrate by binding to L- and D-enantiomeric
analytes. The SERROA effect measured through the interaction of the benzotriazole
dye molecules with the surface plasmons is fundamentally different to that leading to
the SEROA spectra measured for D- and L-ribose and tryptophan [1] adsorbed
directly onto the polymer-stabilised colloids. The direct interaction between the
chiral molecules and metal nanoparticles in that case was responsible for the
observed mirror image SEROA spectra which are mainly due to the field gradient
generated by the plasmon resonance. The observed chirality effect mainly originates
from dipolar interactions with chiral molecules. The present, SERROA, effect is a far
field effect where non-direct interactions between silver nanoparticles and analyte
occur and result in transmission of the chirality of the analyte to the achiral
benzotriazole tags attached by linker molecules to a silver surface. The mechanism
proposed as being responsible for these SERROA results is comparable to that
recently reported for a class of hybrid plasmonic nanomaterials [2] with indirect
adsorption of chiral molecules. Abdulrahman et al. [2] showed that a chiral response
was induced into the plasmonic resonance of the achiral nanostructure using circular
dichroism, so via measurement of electronic excitation. Here, we have observed a
similar response but through monitoring vibrational excitation. The chiral plasmonic
response from silver silica nanotags is a promising alternative which again validated
SERROA as a technique but through a different mechanism.
The detection and quantification of phosphorylated species in complex mixtures has
proven to be difficult due to the limited availability of suitable methods. It has been
established in this thesis that Raman and ROA spectroscopies are powerful probes,
both qualitatively and quantitatively, of phosphorylation [3]. Raman and ROA
spectra of adenosine and seven of its derivative ribonucleotides were measured and
168
Chapter 7
confirmed the sensitivity of both spectroscopic techniques to structural differences.
However, it was shown that ROA is more sensitive to the site and degree of
phosphorylation, with a considerable number of marker bands being identified for
these ribonucleotides. It was found that Raman spectra were sensitive to the number
of phosphate groups but not to their position around the ribose ring. The general
features of the ROA spectra varied dramatically for these adenosine nucleotides and
provided a fingerprint sensitive to both the number and position of phosphate groups.
Studies of the conformations of cyclic and linear L-Ser-L-Ser and L-Ala-L-Ala in
aqueous solution were carried out using Raman and ROA spectroscopies in
combination with quantum mechanical calculations performed by a collaborator at
the University of Dundee. The Raman and ROA spectra of the cyclic and linear
forms of L-Ala-L-Ala and L-Ser-L-Ser were measured in both H2O and D2O and
assignment of the observed bands was proposed from DFT calculations in both cases.
The calculated spectra of both linear and cyclic L-Ala-L-Ala and L-Ser-L-Ser were
in good agreement with our experimental ROA and Raman spectra; the comparison
of which for both forms showed that ROA is more sensitive to structural changes due
to cyclization of dipeptides as it provided more marker bands. Considerable
differences were noted between the observed ROA bands for the cyclic and linear
forms of dialanine and diserine that reflect large differences in the vibrational modes
of the polypeptide backbone upon cyclicization. This study demonstrated that ROA
spectroscopy when utilised in combination with computational modelling clearly
provides a potential tool for characterization of cyclic peptides.
169
Chapter 7
7.1 Future work
This thesis has highlighted the importance of optimization of the correct
experimental protocols for obtaining reliable SERS and SEROA spectra, particularly
the latter. By addition of a hydrogel polymer the aggregation of colloids was
controlled which greatly assisted in obtaining reproducible SEROA spectra. The
hydrogel colloidal system clearly has potential for the stereochemical analysis of a
wide range of biological molecules by SEROA spectroscopy. In the present thesis it
was only possible to use one type of hydrogel polymer, polycarbopol, other
hydrophilic polymers should also be investigated since the current polymer used in
this thesis was pH sensitive whereas other polymers are not and so may provide
better stabilising media for colloid suspension.
The reported theoretical configuration of SEROA is encouraging as the right
experimental setup can be predicted [4]. Bour et al. showed the dependence of CID
ratio on the distance and molecular orientation for SEROA of ribose [4]. Their
results were in agreement with the reported experiments in this thesis [1] and can
potentially provide a suggestion for future experimental setups. Much further work is
required to now understand the two mechanisms identified here for SER(R)OA and
there is a need for experimental and theoretical concepts to work synergistically if
the potential for SEROA as an analytical technique is to become fully realized. Other
classes of nanoparticles such as hollow gold nanospheres, silver triangles, gold
nanorods and gold/silver silica nanoshells provide a strong electromagnetic field that
increase SERS intensities without relying upon high conjunction potentials or ‘hot
spots’ that are characteristic of aggregated metal colloids. Since the nature of
absorption in these nanomaterials is very different, much further work needs to be
performed to investigate this new phenomenon, and to optimize it. This study already
170
Chapter 7
raises interesting possibilities for the enantioselective detection of chiral molecules,
and in particular biomolecules, hence extending the scope of nanoplasmonic devices.
The potential applications of nanoparticles as plasmonic nanomaterials remains
hindered due to limitations in colloidal synthesis to achieve monodisperse
nanoparticles which exhibit enhanced properties. An approach to overcome this
limitation is the rational design of colloidal heterostructured nanocrystals in which
the chemical composition and different reactants (this refers to monomers which
induce nucleation of nanocrystals and sustain their subsequent enlargement) are
spatially controlled ultimately leading to induction of anisotropic growth of metallic
branches and different nanoparticle junctions [5]. The use of such plasmonic
nanomaterials as powerful chirality nanoprobes upon attachment to protein and DNA
for in situ structural determination will be achieved by selecting appropriate nature,
dimensions, morphology and functionalization of the metallic nanostructured
surfaces [6,7]. The external chiral template provides a great advantage to solutionbased chiral nanomaterials for biological samples such as DNA, bacteria and viruses,
therefore offering a wide range of possible applications in biology and medicine.
171
Chapter 7
7.2 References
1.
2.
3.
4.
5.
6.
7.
S. Ostovar Pour, S. E. J. Bell, E. W. Blanch, Chemical Communications.
2011, 4754-4756.
N. A. Abdulrahman, Z. Fan, T. Tonooka, S. M. Kelly, N. Gadegaard, E.
Hendry, A. O. Govorov, M. Kadodwala, Nano Letters. 2012, 977-983.
S. Ostovar Pour, E. W. Blanch, Applied Spectroscopy. 2012, 289-293.
V. Novak, J. Sebestík, P. Bour, Journal of Chemical Theory and
Computation. 2012, 1714-1720.
L. Carbone, P. D. Cozzoli, NanoToday. 2010, 449-493.
E. Hendry, T. Carpy, J. Johnston, M. Popland, R.V. Mikhaylovskiy, A. J.
Lapthorn, S. M. Kelly, L. D. Barron, N. Gadegaard, M. Kadodwala, Nature
Nanotechnology. 2010, 783-787
M. F. Garcia-Parajo, Nature Photonics. 2008, 201-203.
172
Chapter 8
Chapter 8
Appendix
173
Chapter 8
8.0 Declaration
This chapter consists of one full paper by Catherine M. Templeton, Saeideh Ostovar
pour, Jeanette R. Hobbs, Ewan W. Blanch, Steven D. Munger, Graeme L. Conn,
Chemical Senses. 2011, 1-10.
This article has been reproduced in an unchanged format. This work was a
collaboration with Dr Graeme Conn’s group at the Emory University School of
Medicine, Atlanta, USA, and I am second author on this publication. In this paper I
measured Raman and ROA spectra of wild type and two mutants of monellin
(MNEI), a sweet tasting protein in order to investigate secondary structures changes
underlying the sweet taste receptor response. The rest of the work on this paper was
carried out by the other authors.
174