EC-QCL mid-IR transmission spectroscopy for monitoring dynamic changes of protein secondary structure in aqueous solution on the example of β-aggregation in alcohol-denaturated α-chymotrypsin
Anal Bioanal Chem
EC-QCL mid-IR transmission spectroscopy for monitoring dynamic changes of protein secondary structure in aqueous solution on the example of β-aggregation in alcohol-denaturated α-chymotrypsin
Mirta R. Alcaráz 0 1 2
Andreas Schwaighofer 0 1 2
Héctor Goicoechea 0 1 2
Bernhard Lendl 0 1 2
0 Laboratorio de Desarrollo Analítico y Quimiometría, FBCB, Universidad Nacional del Litoral-CONICET , Ciudad Universitaria, 3000 Santa Fe , Argentina
1 Institute of Chemical Technologies and Analytics, Vienna University of Technology , Getreidemarkt 9/164-UPA, 1060 Vienna , Austria
2 Bernhard Lendl
In this work, a novel EC-QCL-based setup for midIR transmission measurements in the amide I region is introduced for monitoring dynamic changes in secondary structure of proteins. For this purpose, α-chymotrypsin (aCT) acts as a model protein, which gradually forms intermolecular β-sheet aggregates after adopting a non-native α-helical structure induced by exposure to 50 % TFE. In order to showcase the versatility of the presented setup, the effects of varying pH values and protein concentration on the rate of βaggregation were studied. The influence of the pH value on the initial reaction rate was studied in the range of pH 5.8-8.2. Results indicate an increased aggregation rate at elevated pH values. Furthermore, the widely accessible concentration range of the laser-based IR transmission setup was utilized to investigate β-aggregation across a concentration range of 5-60 mg mL−1. For concentrations lower than 20 mg mL−1, the aggregation rate appear s to be independent of concentration. At higher values, the reaction rate increases linearly with protein concentration. Extended MCR-ALS was employed to obtain pure spectral and concentration profiles of the temporal transition between α-helices and intermolecular β-sheets. Comparison of the global solutions obtained by the modelled data with results acquired by the laserbased IR transmission setup at different conditions shows excellent agreement. This demonstrates the potential and versatility of the EC-QCL-based IR transmission setup to monitor dynamic changes of protein secondary structure in aqueous solution at varying conditions and across a wide concentration range.
Quantum cascade laser; Infrared spectroscopy; Multivariate curve resolution-alternating least squares; Protein secondary structure; Aggregation; 2; 2; 2-Trifluoroethanol
Parts of this work have been presented at Euroanalysis 2015 in Bordeaux,
Prof. Dr. Bernhard Lendl had the honor of holding the BRobert Kellner
Lecture^ at the Euroanalysis conference held in Bordeaux in September
Infrared spectroscopy is a powerful and established analytical
method to study the structure of proteins [
]. The most
prominent absorption feature of proteins in the mid-IR region is the
amide I band (1600–1700 cm−1) which is induced by
vibrations of the peptide group. It is commonly used for analysis of
protein secondary structure [
], because differing patterns of
hydrogen bonding, dipole-dipole interactions and geometric
orientations in the α-helices, β-sheets, turns and random coil
structures induce different frequencies of the C=O vibrations
that can be correlated with the respective secondary structural
]. For adsorption studies or investigation of protein
thin films, the attenuated total reflectance (ATR) technique is
most commonly employed, while transmission measurements
are routinely used for spectra acquisition of proteins in
An experimental limitation to investigations of protein
secondary structure in aqueous solutions with state-of-the-art
Fourier transform infrared (FT-IR) spectrometers is
constituted by the low feasible path lengths of transmission cells. This
constraint originates from the combination of two factors: the
high molar absorption coefficient of the HOH bending band of
water near 1645 cm−1 that overlaps with the protein amide I
band and the low emission power provided by the thermal
light sources (globars) that are used in FT-IR spectrometers.
As a consequence, path lengths most commonly used for IR
transmission measurements of proteins in aqueous solutions
are in the range of 7 μm to avoid total IR absorption in the
region of the HOH bending band. This limitation comes along
with laborious cell and sample handling as well as the need for
high protein concentration (>10 mg mL−1) [
With the introduction of quantum cascade lasers (QCL), a
significant step was made towards resolving the restrictions
due to low-power light sources in mid-IR spectroscopy [
These new light sources provide emission powers that are
several orders of magnitude higher than thermal IR sources
and even offer brilliance values higher than reached by
]. QCLs are unipolar lasers based on inter
subband transitions of electrons within the semiconductors
conduction band. Contrary to conventional semiconductor lasers,
in QCLs, the emission wavelength range is decoupled from
the band gap of the available semiconductor materials and
depends primarily on the thicknesses of the semiconductor
layers in the nanometer range. Only a few years ago, a new
generation of QCLs, external-cavity QCLs (EC-QCLs),
became commercially available, which are operated at room
temperature and combine high emission power with a spectral
tuning range of a few hundred wavenumbers. This large
accessible spectral range permits the analysis of liquid samples,
where absorption bands generally are broad and often show
overlapping spectral features. The high emission power
enables increased optical path lengths for transmission
measurements even in the presence of strong absorbers such as water
and promise benefits in terms of robustness and sensitivity.
Thus, this type of QCL has been increasingly used for liquid
phase samples and has been successfully applied for analysis
of complex mixtures of analytes in aqueous solution in online
process monitoring [
] and for medical applications [
Most recently, EC-QCL-based IR transmission measurements
have been accomplished for the analysis of protein secondary
structure . It has been shown that the protein spectra
recorded with the laser-based setup show excellent
comparability with spectra acquired by FT-IR spectroscopy. Identification
of spectral features of different secondary structures at protein
concentrations as low as 2.5 mg mL−1 has been achieved
through rigorous application of an advanced data processing
protocol which was established to overcome noise issues
resulting from mechanical imperfections of the tuning
mechanism and the fine structure of the EC-QCL emission curve.
IR spectroscopy is frequently used for studying dynamic
changes of protein secondary structure. Alterations of
secondary structure can be induced by changing external conditions
such as pH, temperature, pressure, co-solvents, surfactants, or
chaotropic agents, which is often accompanied by protein
denaturation. The most prominent change of secondary structure
is the transition of α-helix to β-sheet resulting from protein
aggregation. While various types of denaturation are
accompanied by disruption of α-helices and formation of β-sheets, turns
and polyproline type II helices, exposure to alcohol induces and
stabilizes α-helical domains [
]. The effect of alcohols,
particularly those substituted with fluorine, on proteins has been
extensively studied during the last years; however, the physical
mechanisms by which it affects protein conformation are still
unclear. In the case of 2,2,2-trifluoroethanol (TFE), its low
dielectric constant (one third of water for pure solvent) is believed
to weaken solvophobic interactions that stabilize the native
structure of proteins, and simultaneously strengthen
electrostatic interactions such as intermolecular hydrogen bonds, thereby
stabilizing local secondary structures, particularly the α-helix
]. In addition, it was suggested that fluorine-substituted
alcohols form large micelle-like clusters of alcohol molecules,
resulting in a high local alcohol concentration. The strong
electron-withdrawing effect of fluorine atoms makes TFE a
better hydrogen bond donor, but a poorer acceptor, compared
to water. Upon binding to these hydrophobic clusters, proteins
and peptides undergo conformational transitions [
the detailed effects of TFE on protein conformation have found
to be diverse and strongly dependent on the concentration
ranges of the protein and the co-solvent [
α-Chymotrypsin (aCT) is a predominantly β-sheet protein
folded in two antiparallel β-barrel domains with a molecular
weight of 25 kDa and an isoelectric point (pI) at 8.4 [
The predisposition of aCT towards amyloid fibril aggregation
at intermediate TFE concentrations (15–35 %) and low
protein concentrations (0.025–0.5 mg mL−1, 1–20 μmol L−1) has
been investigated by turbidimetric, dynamic light scattering,
thermodynamic, intrinsic fluorescence and quenching studies
]. It has been shown that the interaction greatly
depends on solution conditions such as pH , TFE
], protein concentration  and temperature [
]. Amyloid fibril formation is promoted at intermediate TFE
concentrations, but at higher concentrations it is suppressed
due to pronounced stabilization of non-native α-helical
structures . For high protein concentrations (20–30 mg mL−1,
800–1200 μmol L−1), it has been shown that exposure of aCT
to high TFE concentrations (50 %) leads to instantaneous
formation of non-native α-helical structures by employing
FT-IR and CD spectroscopy. This fast transition is succeeded
by gradual formation of intermolecular β-sheet aggregates. At
these conditions, proteins consisting of native α-helical
secondary structure do not show β-sheet aggregation after
contact with TFE [
Multivariate curve resolution-alternating least square
(MCR-ALS) is a widespread iterative soft-modelling method
introduced by Tauler in 1995 [
]. This technique allows
obtaining information about multicomponent systems by
discriminating individual contributions of underlying
]. Nowadays, MCR-ALS has demonstrated to be a
powerful chemometric tool to overcome different chemical
problems in several analytical fields. Due to its flexibility
and robustness, this method has been successfully used in
combination with various analytical techniques, such as
], electrophoresis [
], flow analysis [
infrared spectroscopy [
]. For analysis of complex data
matrices, extended MCR-ALS is applied in order to
significantly decrease the ambiguity of the resolution and overcome
problems associated with rank-deficiency. In this study,
extended MCR-ALS is applied allowing to analyse multiple
experiments simultaneously and a global solution is achieved
The aim of this work is to establish the recently introduced
EC-QCL mid-IR transmission setup as a tool for monitoring
dynamic changes of protein secondary structure. To this end,
aCT was selected as a model protein which shows a gradual
transition from α-helix to intermolecular β-sheet after
exposure to TFE. In order to showcase the potential and versatility
of the presented setup, the effects of varying pH values and
protein concentration on the rate of β-aggregation were
investigated. The wide accessible concentration range of the
laserbased IR transmission setup was employed to study
βaggregation across a concentration range of 5–60 mg mL−1
(0.2–2.4 mmol L−1). Furthermore, the influence of the pH
value on the initial reaction rate was studied in the range of
pH 5.8–8.2. Extended MCR-ALS was used to obtain pure
spectral and concentration profiles of the temporal transition
between α-helices and intermolecular β-sheets.
Materials and methods
Reagents and samples
Sodium phosphate monobasic dihydrate p.a. (NaH2PO4 2H2O)
was purchased from Fluka (Buchs, Switzerland), sodium
phosphate dibasic dihydrate (Na2HPO4 2H2O) BioUltra, for
molecular biology, sodium hydroxide solution 50 % in water
(NaOH), hydrochloric acid 37 % (HCl) ACS reagent and 2,2,
2-trifluoroethanol ReagentPlus ≥99 % (TFE), were obtained
from Sigma-Aldrich (Steinheim, Germany). α-Chymotrypsin
from bovine pancreas (≥85 %) (aCT) was obtained by
SigmaAldrich (Steinheim, Germany) and used as purchased.
Ultrapure water (18 MΩ cm) used for preparation of all
solutions was obtained with a Milli-Q water purification system
from Millipore (Bedford, USA).
All aCT protein solutions were prepared by dissolving an
appropriate amount of lyophilized protein powder directly in
1.0 mL of a TFE/16.0 mmol L−1 phosphate buffer (50:50)
mixture solution and placed immediately into the flow cell
of the EC-QCL setup. The pH of the 16.0 mmol L−1
phosphate buffer solutions was adjusted with NaOH or HCl prior
to the mixing with TFE.
A set of seven samples (protein concentration: 20 mg mL−1)
was prepared at different pH values ranging between 5.8 and
8.2. Furthermore, a set of seven samples at pH 6.6 with final
concentrations ranging between 5 and 60 mg mL−1 of aCT was
prepared. pH measurements were carried out with a pH 330i
(Wissenschaftlich-Technische Werkstätten GmbH, Weilheim,
Germany) potentiometer equipped with a Sentix® 61
(Wissenschaftlich-Technische Werkstätten GmbH, Weilheim,
Germany) combined glass electrode and temperature probe.
EC-QC laser setup
IR measurements were performed on a custom-made
ECQCL setup equipped with a quantum cascade laser (Daylight
Solutions Inc., San Diego, USA) with spectral tuning range
between 1729.30 and 1565.06 cm−1, a temperature-controlled
38-μm path length flow cell and a thermoelectrically cooled
MCT detector (Infrared Associates Inc., USA; MCT-7-TE3)
as depicted in Fig. 1. The laser was thermoelectrically cooled
and was operated in pulsed mode at a repetition rate of
100 kHz and a pulse width of 500 ns. The laser head
temperature was set to 18 °C for all measurements. A gold plated
offaxis parabolic mirror (focal length: 43 mm) was used to focus
the MIR light on the detector operating at −60 °C with a
1 × 1 mm element size and a detectivity of D* = 4 × 109 cm
Hz0.5 W−1 at 9.2 μm. The measured signal was processed by a
two-channel boxcar integrator and digitized by a NI DAQ
9239 24-bit ADC (National Instruments Corp., Austin,
USA) at a sampling rate of 16 kHz. The whole setup was
controlled by a LabView-based GUI 11.0 (National
Instruments Corp., Austin, USA, 2011) with server–client
program structure [
All measurements were conducted at 25 °C using a
custom-made, temperature-controlled flow cell equipped with
two MIR transparent CaF2 windows and 38 μm-thick spacer.
To reduce the influence of water vapour, the setup was placed
in a housing of polyethylene foil and constantly flushed with
dry air. Each single beam spectrum consisting of 24,000 data
points was recorded during the tuning time of 1.5 s. A total of
20 scans were recorded for background and sample single
beam spectra (total measurement time for 20 scans: 100 s).
The single beam spectra of the corresponding solvent were
taken as reference and recorded under identical conditions as
sample spectra. To minimize the spectral noise originating
from the spectral mismatch of successive scans, the data
processing routine based on Correlation Optimized Warping
(COW) was employed to align spectra of repeated scans as
well as of the background with the sample spectrum as
described earlier . At last, minor Fourier filtering was
applied to remove residual noise in the final absorbance spectra.
OPUS 7.2 (Bruker Optik GmbH, Ettlingen, Germany,
2012) was used for spectral evaluation.
MCR-ALS is a soft-modelling technique that focuses on
bilinear decomposition of a data matrix D into two submatrices
containing chemically meaningful information of
contributions of the pure compounds involved in the system [
The decomposition is the result of the validity of
BeerLambert’s Law and is achieved by iterative optimization
according to the expression,
D ¼ C
ST þ E
where C contains the profiles referred to the abundance of the
qualitative pure responses and ST comprises the pure
instrumental responses of the components in the system; E contains
the residuals of the model [
One of the most intriguing characteristics of MCR-ALS
resolution is its operation without prior information about
the system under study. However, additional knowledge can
be included in order to achieve chemically meaningful
component profiles. Decomposition of D is obtained by iteratively
optimizing the initial estimates of either C or S using the
available knowledge about the system [
]. This information
is introduced through the implementation of chemical or
mathematical constraints, such as non-negativity, unimodality,
normalization and closure, among others [
Here, protein β-aggregation was monitored at varying
protein concentrations and pH values in the spectral range
between 1710 and 1585 cm−1 during a time period of 240 min.
The corresponding time-absorption spectra matrix for one
measurement run consisted of 38 × 4600 data points for the
temporal and spectral dimension, respectively. For MCR-ALS
Results and discussion
analysis, unfiltered spectra obtained after sample-background
alignment were used to build the time-absorption spectra
matrix. Baseline correction based on a multidimensional
extension of the asymmetric least squares method proposed by
] was applied prior to performing MCR-ALS.
All data sets belonging to the measurement series
investigating either the pH- or concentration dependence of
βaggregation were merged in two individual augmented
column-wise data matrices D by appending the
timeabsorption spectra matrices related to each experiment in the
column direction. In this way, DpH contained the
pHdependent experiments, and Dconc was built with the
concentration-dependent experiments. Prior to MCR-ALS
resolution, determination of the number of compounds in each
data matrix D was carried out using singular value
decomposition (SVD). Initial time-evolution estimations were obtained
using a routine based on the simple iterative self-modelling
approach (SIMPLISMA) methodology [
optimization was carried out applying different constraints, i.e.
nonnegativity in both modes, unimodality in the temporal mode
and normalization in spectral mode. After decomposition, the
column profiles of matrix C and the row profiles of S were
associated with the temporal evolution and pure spectra
profiles of α-helix and β-sheet conformation of the protein,
Data processing and analysis as well as MCR-ALS were
performed in MATLAB R2014b (MathWorks, Inc., Natick,
MA, 2014). MCR-ALS algorithms are available online at
Effect of TFE on the IR spectra of α-chymotrypsin
Employing the EC-QCL setup, mid-IR transmission spectra
were recorded of aCT in aqueous buffer solution and after
exposure to 50 % TFE/buffer solution. In Fig. 2, the IR
absorbance and second-derivative spectra are shown. The IR
absorbance spectrum of native aCT shows a band maximum at
1638 cm−1 and a shoulder at 1680 cm−1, characteristic for
the low- and high-frequency components in β-sheet
secondary structure [
]. Upon exposure to 50 % TFE/buffer
solution, the maximum of the amide I band changes to 1654 cm−1,
typical for the formation of α-helical structures [
effect is in accordance with earlier studies describing the
generation of α-helical secondary structure in proteins after
exposure to TFE [
]. The TFE-induced transition from native
β-sheet secondary structure to α-helix takes place in the time
range of milliseconds  and is not directly observable with
the employed setup.
It was observed that the TFE-induced α-helical secondary
structure is not stable over time, as previously reported for
Fig. 2 a Time-dependent IR absorbance and b second-derivative spectra
of 20 mg mL−1 α-chymotrypsin in 50 % TFE/buffer solution, pH 7.8 at
25 °C (solid lines). The spectra were recorded at time periods between 2
and 240 min (times as indicated in the graph) after the protein was
dissolved in TFE/buffer. Blue solid lines show the spectrum of aCT with
TFE-induced α-helical structure. Green solid lines indicate the spectrum
of the protein after gradual formation of intermolecular β-sheets. Grey
dashed lines represent spectra of the native protein in aqueous buffer.
Black arrows illustrate directions of absorbance changes as a function
proteins exhibiting native β-rich secondary structure [
Figure 2 shows the gradual change of the IR spectrum of
aCT over a time period of 240 min. The intensity of the band
at 1654 cm−1 decreases as bands at 1623 and 1697 cm−1
emerge. This arising spectral pattern is commonly attributed
to intermolecular antiparallel β-sheets aggregates, frequently
occurring in thermally denatured proteins [
Investigations of proteins exhibiting predominantly
αhelical secondary structure such as bovine serum albumin
and myoglobin did not reveal spectral changes indicating
intermolecular β-sheet formation over a comparable period of
time (data not shown).
pH dependence of β-aggregation
TFE-induced formation of intermolecular β-sheets of
20 mg mL−1 aCT in 50 % TFE/buffer solution was
investigated in the range of pH 5.8–8.2. It is noteworthy that under the
applied conditions (pH values and protein concentration
range), no visible precipitation and gelation or increase of
turbidity of the protein solution was observed within 24 h.
That observation strongly suggests that the effects observed
under the conditions of the present study differ from classical
amyloid fibril aggregation as investigated in numerous studies
]. This is in accordance with earlier reports which
found that at high TFE concentrations, proteins appear in an
aggregation deficient non-native state (also called TFE-state),
that is not prone to form amyloid-like fibrils [
For analysis of the temporal progression of the evolving
βsheet content at different pH values, the intensity change
relative to the baseline of the absorbance at 1623 cm−1 was
evaluated (Fig. 3a). The temporal profiles clearly show the
strong pH dependence of the β-sheet formation since the
change of absorbance is higher at elevated pH values. For
quantitative assessment of this behaviour, the initial rate was
calculated as the slope of the absorbance changes throughout
the first four measurement points (2–8 min) against time.
Figure 4 shows the effect of varying pH values on the initial
rate of β-sheet aggregation. The initial rate shows a sigmoidal
progression with low values for mildly acidic pH and high
values for basic pH, with the transition point at approximately
pH 7.0. For higher pH values, the reaction appears to be
finished within 4 h, whereas aggregation is not completed at
lower pH values within the observed time period.
In general, proteins show lowest solubility at solution pH
values near their isoelectric points. Under these conditions,
proteins possess both positively and negatively charged
groups, leading to an anisotropic charge distribution on the
protein surface generating possible dipoles. Thus,
proteinprotein interactions could be highly attractive, rendering
assembly processes such as aggregation energetically
]. In aqueous buffer solution, no
βaggregation could be observed in the investigated pH range.
Apparently, the electrostatic effect destabilizing the protein
conformation is intensified by the addition of TFE. Here, this
aspect is reflected by the increase of the initial rate of β-sheet
formation at pH values close to the pI of the protein. In a
previous study, a comparable change of aggregation
behaviour near the pI was observed for amyloid fibril aggregation at
intermediate TFE concentrations, i.e. with aCT appearing in
the fibril aggregation prone state .
The reproducibility of the system was evaluated by
monitoring the β-aggregation of a triplicate of 20 mg mL−1 aCT
solution at pH 7.0. These particular experimental conditions
were selected since at this pH and concentration level, the
highest variation was expected, as shown in Fig. 4. The
coefficient of variation of the initial rate at these conditions was
d e t e r m i n e d t o b e 4 . 2 % , w h i c h c e r t i f i e s ex c e l l e n t
reproducibility of the method for monitoring protein
aggregation with an EC-QCL setup.
β-aggregation monitored at different protein
To showcase the wide accessible concentration range of the
laser-based IR transmission setup, β-sheet formation of aCT
in 50 % TFE/buffer solution was monitored in a range
between 5 and 60 mg mL−1 of protein at pH 6.6. Again, the
initial rate was evaluated by calculating the slope of the first
four measurement points (2–8 min) of the temporal
progression against time. Figure 5 shows the effect of varying protein
concentrations on the initial rate of β-aggregation. For protein
concentrations between 5 and 20 mg mL−1, the values for the
initial rate remain constant, whereas at higher values, the
reaction rate increases linearly with protein concentration.
The results suggest that at low protein concentration, the
aggregation rate is independent from concentration. The
characteristic change of the slope at 20 mg mL−1 indicates
the value of the critical concentration. This behaviour has been
found for the aggregation characteristics of numerous proteins
]. For protein concentrations higher than the critical
concentration, the initial rate continually increases. This tendency
agrees with results of an earlier study, where higher rates of
change for β-aggregation with increasing protein
concentrations have been found for aCT in the presence of 50 % TFE at
pH 7.4 [
]. Generally, protein aggregation increases with
higher protein concentration due to the higher probability of
protein-protein association [
Extended MCR-ALS analysis
Detailed chemometric analysis of the two QCL-IR
transmission datasets of pH - and concentration-dependent
β-aggregation was performed using extended MCR-ALS. For this study,
MCR-ALS was chosen because it offers the possibility to
reveal the pure spectral profiles for the compounds involved
Fig. 5 Initial rate of β-aggregation for different aCT concentrations in
50 % TFE/buffer at pH 6.6 as analysed by (red squares) evaluation of the
IR absorbance spectra and (blue circles) MCR-ALS
in the reaction as well as their temporal progressions without
any prior knowledge of the system. Other chemometric
methods such as parallel factor analysis (PARAFAC) also
allow obtaining loadings with chemical interpretation about the
system in terms of pure spectral profiles of the involved
components. However, since in the current study the temporal
evolution for different parameters is not identical, this
algorithm is not applicable, due to a lack of trilinearity present in
these datasets [
]. On the other hand, algorithms based on
partial least square (PLS) are generally suitable for
secondorder calibration data analysis, but the results obtained as
loadings and scores do not provide an approximation to the
pure constituent profiles [
]. Thus, these algorithms are not
applicable when spectral information is desired.
In the extended variant of MCR-ALS, multiple matrices are
analysed simultaneously to reduce resolution ambiguities and
rank-deficiency problems. Seven individual time-resolved IR
measurements at different pH values were combined to an
augmented data matrix to obtain the temporal evolution of
β-aggregation (Fig. 6a) as well as spectral profiles (Fig. 6b)
of the individual protein secondary structures elements
involved in the process. The noise level of the spectral profiles
appears low, in particular when considering that MCR-ALS
modelling has been performed with spectra that have not been
Fourier filtered. Values for lack of fit (LOF, 1.8 %) and
percentage of explained variance (R2, 99.98 %) indicate good
description of the experimental data by the MCR-ALS model.
Here, three components were identified, two of them were
involved in the aggregation process and one was referred to
as instrumental noise. The obtained spectral profiles of pure
spectra are consistent with band maxima and shapes of the
associated protein secondary structure components. One
component features a band maximum at 1654 cm−1 and is assigned
to α-helical secondary structure [
]. Its decline of absorbance
along the reaction time is in accordance with the decrease of
intermolecular β-sheet conformation, respectively. Dashed grey lines
indicate the instrumental noise of the system
α-helical content during β-aggregation (see Fig. 6a). The
second component shows a strong band at 1621 cm−1 and a
weaker band at 1695 cm−1, which is the typical spectral
pattern associated with antiparallel intermolecular β-sheets [
A third component was attributed to instrumental noise. It
does not show any characteristic features in the spectral profile
or specific absorbance changes in the temporal profile.
Temporal profiles of the component assigned to
intermolecular β-sheets at different pH values are plotted in Fig. 3b.
Comparison of the modelled data with temporal progressions
obtained by evaluation of IR spectra shows good agreement,
which is, moreover, reflected in the similar trend regarding the
pH dependence of initial rates (see Fig. 4).
The procedure of extended MCR-ALS was also performed
for the measurement series studying the influence of initial
protein concentration on β-aggregation. Values for LOF and
R2 are 1.3 and 99.98 %, respectively, and prove excellent
quality of the MCR-ALS analysis. Temporal and spectral
profiles (see Electronic Supplementary Material Fig. S1) show
three components similar to the results of MCR-ALS
modelling of the pH-dependent study. The compatible outcome
obtained by extended MCR-ALS analyses of two independent
measurement series demonstrates the ruggedness of bilinear
decomposition. Also here, the initial rates of β-aggregation
were computed from modelled data and show a matching
tendency with values obtained from evaluation of
experimental IR spectra for increasing protein concentrations (see
Recently, an EC-QCL-based IR transmission setup has been
introduced for application in secondary structure analysis of
proteins in aqueous solution. The high optical power provided
by the laser light source enables transmission measurements
using higher path lengths (up to 38 μm) and protein
concentrations as low as 2.5 mg mL−1. Using this laser-based setup,
protein spectra recorded under static conditions showed
excellent comparability with spectra acquired by FT-IR
The aim of the present work is to employ the
EC-QCLbased IR transmission setup to accomplish monitoring of
dynamic changes in protein secondary structure. The gradual
formation of intermolecular β-sheet aggregates after inducing
non-native α-helical structures in aCT by exposure to TFE
was monitored at varying pH values and protein
concentrations. It has been shown that the initial reaction rate increases
close to the pI of aCT, attributed to higher attractive
electrostatic interactions under these conditions. The rate of
βaggregation has been found to increase linearly at protein
concentration levels above the critical value. The observed results
agree with the suggestion that TFE reduces hydrogen bonds
formed between proteins and surrounding water molecules,
thus inducing protein conformations that are compact and
maximise intermolecular hydrogen bonding. The gradual
formation of β-sheet aggregates from primarily generated
αhelical structures induced by TFE is in accordance with earlier
studies that found that α-helical structures represent the
kinetically favoured state and intermolecular β-sheets constitute
the thermodynamically preferred state of aCT under the
investigated solvent conditions [
Extended MCR-ALS analysis of pH and
concentrationdependent measurements show similar spectral profiles,
demonstrating the high quality of the IR spectra obtained by the
EC-QCL-based IR transmission setup at varying conditions
and the general robustness of this chemometric technique.
Concentration profiles obtained by the MCR-ALS model
show good comparability with evaluation of IR spectra.
The present study demonstrates the high potential and great
versatility of the laser-based IR transmission setup to monitor
dynamic changes of protein secondary structure in aqueous
solution. The use of a transmission flow cell with a path length
approximately four times higher than usually employed for
conventional FT-IR spectrometers facilitates experiments in
flow-through configuration. Flow injection analysis (FIA) as
well as sequential injection analysis (SIA) are established
techniques that could be coupled to the presented laser-based
IR transmission setup for denaturation studies of proteins and
investigations of protein-ligand binding. For monitoring
dynamic events on shorter time scale, stopped flow mixing is
feasible. For time-resolved measurements, the wavenumber
tuning rate of the EC-QCL constitutes the limiting factor.
Typical rates for broadband tuning of commercially available
lasers are in the range of 100 cm−1 s−1. Significantly higher
time-resolution can be obtained by keeping the emission
wavenumber of the laser constant throughout one experiment
and incrementally recording the absorbance at individual
wavenumbers along the reaction time during multiple
measurement repetitions. Rearrangement of the recorded data
obtains the set of time-resolved spectra. A prerequisite for this
approach is that the observed sample reaction is repeatable as
well as reproducible and following from that, the initiation of
reagent mixing by an automated flow injection technique.
Acknowledgments Open access funding provided by TU Wien
(TUW). Financial support was provided by the Austrian research funding
association (FFG) under the scope of the COMET programme within the
research project BIndustrial Methods for Process Analytical Chemistry—
From Measurement Technologies to Information Systems (imPACts)^
(contract #843546). This work was also partially supported by the
University Relations Grant provided by Agilent Technologies (no.
3375). M.R.A. gratefully acknowledges the financial support provided
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
Open Access This article is distributed under the terms of the Creative
C o m m o n s A t t r i b u t i o n 4 . 0 I n t e r n a t i o n a l L i c e n s e ( h t t p : / /
creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided you give
appropriate credit to the original author(s) and the source, provide a link
to the Creative Commons license, and indicate if changes were made.
1. Barth A . Infrared spectroscopy of proteins . Biochim Biophys Acta . 2007 ; 1767 : 1073 - 101 .
2. Fabian H , Mäntele W. Infrared spectroscopy of proteins. Handbook of vibrational spectroscopy . Hoboken: John Wiley & Sons, Ltd; 2006 .
3. Bal Ram S. Basic aspects of the technique and applications of infrared spectroscopy of peptides and proteins. Infrared analysis of peptides and proteins . ACS Symposium Series . 750 : American Chemical Society; 1999 . p. 2 - 37 .
4. Faist J , Capasso F , Sivco DL , Sirtori C , Hutchinson AL , Cho AY . Quantum cascade laser . Science . 1994 ; 264 ( 5158 ): 553 - 6 .
5. Weida MJ , Yee B . Quantum cascade laser-based replacement for FTIR microscopy . P Soc Photo-Opt Ins . 2011 ; 7902 : 79021C .
6. Siegmann-Hegerfeld T , Genner A , Brandstetter M , Miltner M , Lendl B , Harasek M. Online monitoring of Toc contaminations in clean-in-place processes for optimized process control, increased process efficiency and quality . Chem Eng Trans . 2013 ; 35 : 979 - 84 .
7. Brandstetter M , Volgger L , Genner A , Jungbauer C , Lendl B . Direct determination of glucose, lactate and triglycerides in blood serum by a tunable quantum cascade laser-based mid-IR sensor . Appl Phys B-Lasers O. 2013 ; 110 ( 2 ): 233 - 9 .
8. Brandstetter M , Genner A , Anic K , Lendl B . Tunable external cavity quantum cascade laser for the simultaneous determination of glucose and lactate in aqueous phase . Analyst . 2010 ; 135 ( 12 ): 3260 - 5 .
9. Brandstetter M , Lendl B. Tunable mid -infrared lasers in physical chemosensors towards the detection of physiologically relevant parameters in biofluids . Sens Actuators B-Chem . 2012 ; 170 : 189 - 95 . Alcaráz MR , Schwaighofer A , Kristament C , Ramer G , Brandstetter M , Goicoechea H , et al. External-cavity quantum cascade laser spectroscopy for mid-IR transmission measurements of proteins in aqueous solution . Anal Chem . 2015 ; 87 ( 13 ): 6980 - 7 .
11. Matsuo K , Sakurada Y , Tate S , Namatame H , Taniguchi M , Gekko K. Secondary-structure analysis of alcohol-denatured proteins by vacuum-ultraviolet circular dichroism spectroscopy . Proteins . 2012 ; 80 ( 1 ): 281 - 93 .
12. Uversky VN , Narizhneva NV , Kirschstein SO , Winter S , Lober G . Conformational transitions provoked by organic solvents in betalactoglobulin: can a molten globule like intermediate be induced by the decrease in dielectric constant? Fold Des . 1997 ; 2 ( 3 ): 163 - 72 .
13. Hong DP , Hoshino M , Kuboi R , Goto Y. Clustering of fluorinesubstituted alcohols as a factor responsible for their marked effects on proteins and peptides . J Am Chem Soc . 1999 ; 121 ( 37 ): 8427 - 33 .
14. Buck M. Trifluoroethanol and colleagues: cosolvents come of age. Recent studies with peptides and proteins . Q Rev Biophys . 1998 ; 31 ( 3 ): 297 - 355 .
15. Wright HT . Comparison of crystal-structures of chymotrypsinogena and alpha-chymotrypsin . J Mol Biol . 1973 ; 79 ( 1 ): 1 - 11 .
16. Wilcox PE . Chymotrypsinogens-chymotrypsins . In: Perlmann GE , Lorand L , editors. Methods in Enzymology . Volume 19 . Cambridge: Academic Press; 1970 . p. 64 - 108 .
17. Rezaei-Ghaleh N , Ebrahim-Habibi A , Moosavi-Movahedi AA , Nemat-Gorgani M . Role of electrostatic interactions in 2,2,2- trifluoroethanol-induced structural changes and aggregation of alpha-chymotrypsin . Arch Biochem Biophys . 2007 ; 457 ( 2 ): 160 - 9 .
18. Pallares I , Vendrell J , Aviles FX , Ventura S. Amyloid fibril formation by a partially structured intermediate state of alpha-chymotrypsin . J Mol Biol . 2004 ; 342 ( 1 ): 321 - 31 .
19. Khodarahmi R , Soori H , Amani M. Study of cosolvent-induced alpha-chymotrypsin fibrillogenesis: does protein surface hydrophobicity trigger early stages of aggregation reaction? Protein J. 2009 ; 28 ( 7-8 ): 349 - 61 .
20. Harrison RS , Sharpe PC , Singh Y , Fairlie DP . Amyloid peptides and proteins in review . In: Amara SG , Bamberg E , Fleischmann B , Gudermann T , Hebert SC , Jahn R , et al., editors. Reviews of Physiology, Biochemistry and Pharmacology . Vol 159 . Reviews of Physiology Biochemistry and Pharmacology . 1592007 . p. 1 - 77 .
21. Dong A , Matsuura J , Manning MC , Carpenter JF . Intermolecular beta-sheet results from trifluoroethanol-induced nonnative alphahelical structure in beta-sheet predominant proteins: infrared and circular dichroism spectroscopic study . Arch Biochem Biophys . 1998 ; 355 ( 2 ): 275 - 81 .
22. Tauler R . Multivariate curve resolution applied to second order data . Chemometr Intell Lab . 1995 ; 30 : 133 - 46 .
23. Ruckebusch C , Blanchet L. Multivariate curve resolution: a review of advanced and tailored appilcations and challenges . Anal Chim Acta . 2013 ; 765 : 28 - 36 .
24. Alcaráz MR , Siano GG , Culzoni MJ , Munoz de la Pena A , Goicoechea HC . Modeling four and three-way fast highperformance liquid chromatography with fluorescence detection data for quiatitation of fluoroquinolones in water samples . Anal Chim Acta . 2014 ; 809 : 37 - 46 .
25. Alcaráz MR , Vera Candioti L , Culzoni MJ , Goicoechea HC . Ultrafast quantitation of six quinolones in water samples by second-order capillary electrophoresis data modeling with multivariate curve resolution-alternating least squares . Anal Bioanal Chem . 2014 ; 406 ( 11 ): 2571 - 80 .
26. Alcaráz MR , Schenone AV , Culzoni MJ , Goicoechea HC . Modeling of second-order spectrophotometric data generated by a pH-gradient flow injection technique for the determination of doxorubicin in human plasma . Microchem J . 2014 ; 112 : 25 - 33 .
27. Kuligowski J , Quintas G , Tauler R , Lendl B , de la Guardia M. Background correction and multivariate curve resolution of online liquid chromatography with infrared spectrometric detection . Anal Chem . 2011 ; 83 ( 12 ): 4855 - 62 .
28. Diewok J , De Juan A , Tauler R , Lendl B . Quantitation of mixtures of diprotic organic acids by FT-IR flow titrations and multivariate curve resolution . Appl Spectrosc . 2002 ; 56 ( 1 ): 40 - 50 .
29. Diewok J , de Juan A , Maeder M , Tauler R , Lendl B . Application of a combination of hard and soft modeling for equilibrium systems to the quantitative analysis of pH-modulated mixture samples . Anal Chem . 2003 ; 75 ( 3 ): 641 - 7 .
30. De Juan A , Tauler R . Chemometrics applied to unravel multicomponent processes and mixtures. Revisiting latest trends in multivarate resolution . Anal Chim Acta . 2003 ; 500 : 195 - 210 .
31. Tauler R , Maeder M , de Juan A . Multiset data analysis: extended multivariate curve resolution . In: Brown SD , Tauler R , Walczak B , editors. Comprehensive chemometrics: chemical and biochemical data analysis. Oxford: Elsevier; 2009 . p. 473 - 505 .
32. Wagner C , Genner A , Ramer G , Lendl B. Advanced Total Lab Automation System (ATLAS). Modeling, programming and simulations using LabVIEW; 2011 .
33. de Juan A , Rutan S , Tauler R . Two-way data analysis: multivariate curve resolution-iterative resolution methods . In: Brown SD , Tauler R , Walczak B , editors. Comprehensive chemometrics: chemical and biochemical data analysis. Oxford: Elsevier; 2009 . p. 325 - 44 .
34. Muñoz de la Peña A , Goicoechea HC , Escandar GM , Olivieri AC , editors. Fundamentals and analytical applications of multiway calibration . Amsterdam: Elsevier; 2015 .
35. De Juan A , Tauler R . Multivariate curve resolution (MCR) from 2000: progress in concepts and applications . CRC Crit Rev Anal Chem . 2006 ; 363 - 4 : 163 - 76 .
36. Eilers P , Boelens H . Baseline correction with asymmetric least squares smoothing; 2005 :[ 1 - 24 pp.]. Available from: http:// zanran_storage.s3.amazonaws.com/www.science.uva.nl/ ContentPages/443199618.pdf.
37. Windig W , Guilment J . Interactive self-modeling mixture analysis . Anal Chem . 1991 ; 63 ( 14 ): 1425 - 32 .
38. Dong A , Huang P , Caughey WS . Protein secondary structures in water from second-derivative amide I infrared spectra . Biochemistry . 1990 ; 29 ( 13 ): 3303 - 8 .
39. Jasanoff A , Fersht AR . Quantitative-determination of helical propensities from trifluoroethanol titration curves . Biochemistry . 1994 ; 33 ( 8 ): 2129 - 35 .
40. Lu H , Buck M , Radford SE , Dobson CM . Acceleration of the folding of hen lysozyme by trifluoroethanol . J Mol Biol . 1997 ; 265 ( 2 ): 112 - 7 .
41. Filosa A , Wang Y , Ismail AA , English AM . Two-dimensional infrared correlation spectroscopy as a probe of sequential events in the thermal unfolding of cytochromes c . Biochemistry . 2001 ; 40 ( 28 ): 8256 - 63 .
42. Murayama K , Tomida M . Heat-induced secondary structure and conformation change of bovine serum albumin investigated by Fourier transform infrared spectroscopy . Biochemistry . 2004 ; 43 ( 36 ): 11526 - 32 .
43. Rezaei-Ghaleh N , Amininasab M , Nemat-Gorgani M . Conformational changes of alpha-chymotrypsin in a fibrillationpromoting condition: a molecular dynamics study . Biophys J . 2008 ; 95 ( 9 ): 4139 - 47 .
44. Chi E , Krishnan S , Randolph T , Carpenter J . Physical stability of proteins in aqueous solution: mechanism and driving forces in nonnative protein aggregation . Pharm Res . 2003 ; 20 ( 9 ): 1325 - 36 .
45. Wang W , Li N , Speaker S. External factors affecting protein aggregation . In: Wang W , Roberts C , editors. Aggregation of therapeutic proteins. Hoboken: John Wiley & Sons, Inc; 2010 . p. 119 - 205 .
46. Olivieri AC , Escandar GM . Partial least-squares with residual bilinearization . In: Olivieri AC , Escandar GM , editors. Practical three-way calibration . Boston: Elsevier; 2014 . p. 157 - 95 .
47. Chirgadze YN , Nevskaya NA . Infrared spectra and resonance interaction of amide-I vibration of the antiparallel-chain pleated sheet . Biopolymers . 1976 ; 15 ( 4 ): 607 - 25 .