Benzothiazole-Based Neutral Ratiometric Fluorescence Sensor for Amyloid Fibrils

Article abstract of DOI:10.1002/chem.201602981

Early detection of amyloid fibrils is very important for the timely diagnosis of several neurological diseases. Thioflavin-T (ThT) is a gold standard fluorescent probe for amyloid fibrils and has been used for the last few decades. However, due to its positive charge, ThT is incapable of crossing the blood–brain barrier and cannot be used for in vivo imaging of fibrils. In the present work, we synthesized a neutral ThT derivative, 2-[2’-Me,4’-(dimethylamino)phenyl]benzothiazole (2Me-DABT), which showed a strong affinity towards the amyloid fibrils. On association with the amyloid fibrils, 2Me-DABT not only showed a large increase in its emission intensity, but also, unlike ThT, a large blueshift in its emission spectrum was observed. Thus, unlike ThT, 2Me-DABT is a potential candidate for the ratiometric sensor of the amyloid fibrils. Detailed photophysical properties of 2Me-DABT in amyloid fibrils and different solvent media were studied to understand its sensory activity. Fluorescence resonance energy transfer (FRET) studies suggested that the sites of localization for ThT and 2Me-DABT in amyloid fibrils are not same and their average distance of separation in amyloid fibrils was determined. The experimental data was nicely supported by molecular docking studies, which confirmed the binding of 2Me-DABT in the inner core of the amyloid fibrils.

Full text of DOI:10.1002/chem.201602981

DOI: 10.1002/chem.201602981
Full Paper
&
Fluorescent Amyloid Probe
Benzothiazole-Based Neutral Ratiometric Fluorescence Sensor for
Amyloid Fibrils
Aruna K. Mora,[a] Sushant Murudkar,[a] A. Alamelu,[a, c] Prabhat K. Singh,[a]
Subrata Chattopadhyay,[b] and Sukhendu Nath*[a]
Abstract: Early detection of amyloid fibrils is very important
for the timely diagnosis of several neurological diseases. Thi-
oflavin-T (ThT) is a gold standard fluorescent probe for amy-
loid fibrils and has been used for the last few decades. How-
ever, due to its positive charge, ThT is incapable of crossing
the blood–brain barrier and cannot be used for in vivo imag-
ing of fibrils. In the present work, we synthesized a neutral
ThT derivative, 2-[2’-Me,4’-(dimethylamino)phenyl]benzothia-
zole (2Me-DABT), which showed a strong affinity towards
the amyloid fibrils. On association with the amyloid fibrils,
2Me-DABT not only showed a large increase in its emission
intensity, but also, unlike ThT, a large blueshift in its emission
spectrum was observed. Thus, unlike ThT, 2Me-DABT is a po-
tential candidate for the ratiometric sensor of the amyloid fi-
brils. Detailed photophysical properties of 2Me-DABT in amy-
loid fibrils and different solvent media were studied to un-
derstand its sensory activity. Fluorescence resonance energy
transfer (FRET) studies suggested that the sites of localiza-
tion for ThT and 2Me-DABT in amyloid fibrils are not same
and their average distance of separation in amyloid fibrils
was determined. The experimental data was nicely support-
ed by molecular docking studies, which confirmed the bind-
ing of 2Me-DABT in the inner core of the amyloid fibrils.
Introduction
vivo experiments. Several imaging techniques, like, magnetic
resonance imaging (MRI),[11,12] positron emission tomography
(PET),[13,14] and single-photon emission computed tomography
(SPECT)[15,16] have been used to study amyloid fibrils deep
inside the tissue. Indeed, the first human clinical trial to study
amyloidosis was performed by a PET technique using a 18F-la-
beled radiotracer.[13] However, the use of these techniques is
limited by several factors, such as expensive instruments, the
radiotracer’s short lifetime, complicated data analysis, and,
most importantly, exposure to radioactivity. Optical imaging of
amyloid fibrils using fluorescent tracers has attracted signifi-
cant attention in recent years because of its simplicity, cost ef-
fectiveness, faster data accumulation, simple data analysis, and
high sensitivity. Several small organic and inorganic molecules
have been developed for the optical imaging of the amyloid fi-
brils. For example, Uchida et al. used Hoechst 33342 for detect-
ing amyloid fibrils in APP transgenic mice.[17] Sensors based on
fluorescein,[18] BODIPYs,[19,20] curcumin,[21] thiophene,[22] binu-
clear metal complexes,[23] etc were used for amyloid fibril de-
tection. Interestingly, gold metal nanoparticles[24] and silicon
nanoarrays[25] were also developed for the same purpose.
Amyloidosis, a pathological condition, arises from the deposi-
tion of protein agglomerates on organs and tissues in the
human body, and is responsible for several diseases, like, Par-
kinson’s disease, Alzheimer’s disease, Huntington’s disease,
type II diabetes, etc.[1–3] Such protein agglomerates, generally
known as amyloid fibrils, are insoluble in water and remarkably
resistant towards proteolysis.[4,5] Early detection of amyloidosis
and knowledge of its progress is very important to prevent for-
mation of amyloid fibrils in the human body and to develop
suitable therapeutics.
Since its discovery by Rudolph Virchow in the year of 1854,[6]
several biochemical and biophysical techniques have been
used extensively to study amyloid fibrils. Further, light scatter-
ing,[7] X-ray scattering,[8] FTIR,[9] and AFM,[10] etc. have been
used to compliment the biophysical and biochemical results.
However, all such studies were limited to the in vitro and ex
[a] A. K. Mora, S. Murudkar, A. Alamelu, Dr. P. K. Singh, Dr. S. Nath
Radiation & Photochemistry Division, Bhabha Atomic Research Centre
Trombay, Mumbai 400 085 (India)
Intensive efforts notwithstanding, thioflavin-T (ThT) still re-
mains the exclusive fluorescent probe for amyloid fibrils.[26–28]
ThT, a nonemissive species in water, shows a very large in-
crease (>103 times) in its emission yield on binding with the
amyloid fibrils.[26] The increased emission yield of ThT has been
attributed to the restriction in the nonradiative torsional
motion around its central CꢀC single bond in the amyloid fi-
brils.[29,30] ThT lacks protein-specificity as it binds to amyloid fi-
brils made up from all classes of proteins. Nonetheless, it has
E-mail: snath@barc.gov.in
sukhendunath@yahoo.co.in
[b] Dr. S. Chattopadhyay
Bio-organic Division, Bhabha Atomic Research Centre
Trombay, Mumbai 400 085 (India)
[c] A. Alamelu
Present address: Cardomom Planters Association College
Bodinayakanur, Tamil Nadu (India)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201602981.
Chem. Eur. J. 2016, 22, 1 – 9
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
been used for several decades for its high association constant
with the amyloid fibrils, and large sensitivity due to the enor-
mous increase in its emission intensity on association with
amyloid fibrils. The actual mechanism for such a strong associ-
ation of ThT with amyloid fibrils is still not clearly understood.
It is believed that the molecular shape of ThT matches with
that of some of the grooves in the fibrillar channel and on its
surface.[31] Such shape complementarity between ThT and the
amyloid structure along with the hydrophobic stabilization is
believed to be mainly responsible for the strong association
between ThT and amyloid fibrils. However, ThT cannot be em-
ployed for in vivo imaging due to its positive charge, which
makes it highly hydrophilic and prevents it from crossing the
blood–brain barrier (BBB).[32] Further, the spectral position of
ThT remains largely unchanged on association with amyloid fi-
brils, which makes it inappropriate for the imaging studies due
to the large background signal.[30,33] Moreover, ThT has been
recently shown to promote amyloid b aggregation.[34] Such
limitations of ThT have encouraged the development of neu-
tral fluorescent sensors for amyloid fibrils.[32,35,36] However, sen-
sors with molecular shapes similar to ThT are anticipated to
provide extra selectivity due to structural complementarity be-
tween the fibrils and the probes. Thus, several neutral ThT de-
rivatives were synthesized that showed a high affinity towards
amyloid fibrils.[32,35–37] Derivatives based on radio-labeled bezo-
thiazoleaniline (BTA) moieties are the most prolific and promis-
ing amyloid sensors, and have been used for the PET imaging
of amyloid fibrils.[32,37] However, due to lack of suitable photo-
physical properties, such derivatives could not be used for the
detection of amyloid fibrils using sensitive fluorescence tech-
niques. Klunk et al. reported that BTA derivatives did not show
any change in their emission spectrum on association with
amyloid fibrils.[32] Instead, emission intensity of one of the BTA
derivatives, which is extensively used for PET amyloid imaging,
was reduced in the amyloid fibrillar solution.[37] Hence, a neutral
ThT derivative, which can show a significant increase in its
emission intensity along with the shift in the spectral position
on association with amyloid fibrils, needs to be developed for
efficient detection of these neurotoxic protein agglomerates.
In the present study, we synthesized a neutral analogue of
ThT, 2-[2’-Me,4’-(dimethylamino)phenyl]benzothiazole (2Me-
DABT, see Scheme 1 for the chemical structure), which showed
a large change in its photophysical properties on association
with amyloid fibrils made from insulin. The interaction of 2Me-
DABT with insulin amyloid fibrils was studied using both
steady-state and time-resolved fluorescence techniques. It is
shown that unlike ThT, 2Me-DABT showed a significant blue-
shift in the emission spectrum along with a largely increased
emission intensity due to its association with the amyloid fi-
brils, which suggests its potential application as a ratiometric
amyloid fibril sensor. Fluorescence resonance energy transfer
(FRET) and molecular docking studies were also performed to
identify its binding sites in amyloid fibrils.
Results and Discussion
2Me-DABT in aqueous solution showed a relatively narrow ab-
sorption band with a maxima at approximately 300 nm, which
was largely unchanged in the presence of insulin amyloid fi-
brils (see Figure S1 in the Supporting Information). The emis-
sion spectra of 2Me-DABT, recorded in aqueous solution and in
the presence of different concentrations of insulin fibrils, are
shown in Figure 1. 2Me-DABT showed a broad emission band
with maxima at 500 nm in aqueous solution. It is evident from
Figure 1 that the addition of insulin fibrils to the aqueous solu-
tion of 2Me-DABT led to a substantially increased (ꢁ65 times)
emission intensity along with a significant hypsochromic shift
in the emission spectrum. The inset of Figure 1 shows the
peak intensity normalized emission spectra of 2Me-DABT in the
presence of different concentrations of amyloid fibrils. The
emission maxima of 2Me-DABT changed from 500 nm in aque-
ous solution to 445 nm in amyloid fibrils. The observed large
fluorescence enhancement along with large blueshift in the
emission spectra clearly indicated that 2Me-DABT interacts
strongly with amyloid fibrils. It is noteworthy that despite
having a strong interaction with amyloid fibrils, unlike ThT, the
fibrillation kinetics of insulin remained unaffected by the pres-
ence of 2Me-DABT (see Figure S2 in the Supporting Informa-
tion). Thus, the present result suggested that 2Me-DABT may
be an efficient fluorescence sensor for the amyloid fibrils. In
particular, the large shift in the emission spectra is advanta-
geous for fluorescence imaging applications due to the rela-
tively low background signal. This may also improve the detec-
tion limit significantly.
Such a large shift in the emission spectra of 2Me-DABT upon
binding with amyloid fibrils can make it a ratiometric sensor
for the amyloid fibrils. It is worth mentioning that the emission
spectral position of ThT remains unchanged on association
Figure 1. Emission spectra (lex =340 nm) of 2Me-DABT in the presence of
different concentrations of insulin amyloid fibrils (0–30 mm). Inset: peak in-
tensity normalized emission spectra of 2Me-DABT in amyloid fibrils.
Scheme 1. Molecular structure of (2-[2’-Me,4’-(dimethylamino)phenyl]benzo-
thiazole) (2Me-DABT).
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
with amyloid fibrils[26,30] and hence can act only as a single in-
tensity based sensor. Such a single-intensity-based estimation
is prone to error caused by the fluctuation in probe concentra-
tion and its environment, drift in the opto-electronic setup,
etc. during the measurement.[38] However, the ratio of emission
intensities of the probe at two different wavelengths is
immune to such changes during the measurements.[38] Ratio-
metric measurements are particularly very important in micro-
scopic studies of cells, for which the concentration and envi-
ronment of probes as well as the optical path fluctuates con-
tinuously during the course of the measurement. It is also im-
portant to mention that the light used for the excitation of
2Me-DABT can generate autofluorescence from cells[39] and can
interfere with the measurements. However, due to the very
high emission quantum yield of 2Me-DABT in amyloid fibrils
(68%), it is quite expected that the emission intensity generat-
ed from 2Me-DABT in amyloid fibrils will be significantly higher
than that of cell autofluorescence. Hence, in microscopic stud-
ies using 2Me-DABT as a fibrillar probe, the interference from
the cell autofluorescence is expected to be negligible.
Figure 2. Double reciprocal plot for the variation in emission intensity of
2Me-DABT with the concentration of insulin amyloid fibrils. The two straight
lines are the fit for the experimental data by using Equation (1).
DABT in aqueous solution showed nonexponential decay kinet-
ics. All transient decay traces were fitted with a biexponential
decay function. The average lifetime (tav) of 2Me-DABT in aque-
ous solution was estimated to be 2.9 ns. It is to be noted from
Figure 3A that the emission transient decay became gradually
slower on addition of amyloid fibrils. The variation in the aver-
age lifetimes of 2Me-DABT with amyloid fibril concentrations
are presented in Figure 3B. It is evident from Figure 3B that
the average excited-state lifetime of 2Me-DABT was increased
from 2.9 ns in water to 4.3 ns in the presence of 25 mm amyloid
fibrils. Such changes in the excited-state lifetime certainly indi-
cated a strong interaction of 2Me-DABT with amyloid fibrils. As
time-resolved measurements are inherently ratiometric in
nature, the changes in the lifetime of 2Me-DABT can also be
used for the estimation of amyloid fibrils. However, it is to be
noted that the steady-state emission intensity of 2Me-DABT
was changed by approximately 65 times due to its association
with amyloid fibrils, whereas its lifetime changed only by
1.5 times. Hence, for the present system, the emission-intensi-
ty-based detection of amyloid fibrils will be more sensitive
than the lifetime-based method.
To quantify the interaction between 2Me-DABT and insulin
amyloid fibrils, binding constants were estimated by using the
following modified Benesi–Hildebrand equation for 1:1 com-
plexation:[40,41]
1
DIf
1
1
¼
þ
ð1Þ
KbðIC ꢀ IOÞ½Fꢂo ðIC ꢀ IOÞ
in which Io and IC are the emission intensities of 2Me-DABT in
water and when completely bound with amyloid fibrils, respec-
tively. DIf is the change in the emission intensity due to the ad-
dition of fibrils. [F]0 is the total concentration of fibrils in the
solution and Kb is the binding constant. The double reciprocal
plot for the variation in the emission intensity of 2Me-DABT
with the concentration of amyloid fibrils is shown in Figure 2.
The nonlinearity observed in Figure 2 clearly indicated two
modes of binding for 2Me-DABT with amyloid fibrils. The ex-
perimental data were fitted with Equation (1) at two different
concentration regions as shown in Figure 2. The binding con-
stants thus estimated from the fitting, were found to be 1.0ꢁ
106 and 4.1ꢁ104 mꢀ1. Earlier, ThT was reported to follow two
modes of binding, one in the fibrillar channel and the other in
the grooves of the fibrillar surface.[42–44] Due to structural com-
plementarity between ThT and the fibrillar channel, the bind-
ing in the inner channel of the fibrils is very strong compared
to that on the surface. Considering the similar chemical struc-
ture of ThT and 2Me-DABT, in analogy, we also propose that
the stronger binding mode corresponds to the 2Me-DABT
bound to the inner channel of the fibril and the weaker mode
is due to the binding of 2Me-DABT onto the fibrillar surface.
This proposition was also supported by our molecular docking
studies (vide infra).
To comprehend the reason for the observed changes in the
emission characteristics of 2Me-DABT due to its interaction
with amyloid fibrils, emission properties of 2Me-DABT in differ-
ent solvents were investigated in detail. The emission spectra
of 2Me-DABT recorded in acetonitrile/water solvent mixtures
are shown in Figure 4A. A substantial decrease in the emission
intensity of 2Me-DABT was noticed with the increase in the sol-
vent polarity. The peak intensity normalized emission spectra
of 2Me-DABT in solvent mixtures are shown in the inset of Fig-
ure 4A. The emission spectrum of 2Me-DABT showed a large
bathochromic shift on increasing the solvent polarity. Thus, the
emission maxima of 2Me-DABT in acetonitrile (dielectric con-
stant, e=37.5) is 463 nm and changed to 495 nm in a 95%
water/acetonitrile mixture (e=76).[45,46] Due to the extremely
low solubility of 2Me-DABT in water, its photophysical proper-
ties in neat water could not be studied. On changing the sol-
vent from neat acetonitrile to a 95% water/acetonitrile mix-
ture, 2Me-DABT showed a Stokes’s shift (1391 cmꢀ1), which was
quite large when compared to commonly used solvatochromic
Time-resolved emission measurements were performed to
understand the effect of amyloid fibrils on the excited-state dy-
namics of 2Me-DABT. Figure 3A shows measured emission
transient decays of 2Me-DABT in the presence of different con-
centrations of amyloid fibrils. The emission transient of 2Me-
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 3. A) Emission transient decays for 2Me-DABT in the presence of dif-
ferent concentrations of insulin amyloid fibrils: 1) 0, 2) 0.6, 3) 4.5, 4) 12, and
5) 20 mm. The dashed curve represents the instrument response function
(IRF) of the TCSPC instrument. B) Variation in the average lifetime of 2Me-
DABT with insulin amyloid fibril concentration.
Figure 4. A) Emission spectra of 2Me-DABT in acetonitrile in the presence of
different amounts of water. Inset: the peak normalized emission spectra of
2Me-DABT in acetonitrile/water solvent mixtures. B) Variation in the emission
quantum yield of 2Me-DABT with the percentage of water in acetonitrile/
water mixtures.
dyes, like coumarin 153 (1031 cmꢀ1). Thus, present results indi-
cated that 2Me-DABT is a strong solvatochromic dye and may
be a potential candidate to probe the micropolarity of differ-
ent complex chemical and biological environments. Figure 4B
crease in the emission at 490 nm due to ThT bound to amyloid
fibrils.[30] The emission spectrum for only ThT in fibrillar media
is also shown in Figure 5 for comparison. The emission intensi-
ty of only ThT in amyloid fibrils was quite low when excited at
340 nm. However, the emission of ThT at 490 nm is significantly
enhanced in the presence of 2Me-DABT in amyloid fibrils. The
variation in the emission intensities at 445 and 490 nm as
a function of ThT concentration are also shown in the inset of
Figure 5. The large increase in the emission intensity at 490 nm
at the cost of that at 445 nm clearly indicated an efficient
energy transfer from the photoexcited 2Me-DABT to ThT in the
fibrillar media. The energy-transfer process is further supported
by the fact that there was a substantial overlap between the
emission spectrum of 2Me-DABT and the absorption spectrum
of ThT in the fibrillar media (see Figure S3 in the Supporting In-
formation).
shows the variation in the emission quantum yield (f) of 2Me-
f
DABT with solvent polarity. Thus, the emission yield of 2Me-
DABT was decreased substantially with the increase in the
water content in mixed solvents. This result clearly indicated
that nonradiative deactivation of 2Me-DABT became more
prominent in the high-polarity solvent media.
From above results it can be inferred that a decrease in the
polarity around 2Me-DABT led to a blueshift in its emission
spectrum and a simultaneous increase in its emission yield.
Due to their extremely complex structure, amyloid fibrils can
have different hydrophobic pockets to accommodate small
molecules. For example, Nile red on its association with amy-
loid fibrils experienced much lower polarity when compared to
the bulk water, which is reflected by the blueshift in its emis-
sion spectra.[47] Thus, it can be inferred that the micropolarity
at the site of solubilization of 2Me-DABT in amyloid fibrils is
much less as compared to the bulk water. Such a low polarity
around 2Me-DABT in the fibrillar media resulted in a significant
blueshift in its emission spectra with a concomitant large in-
crease in its emission yield.
Observed energy transfer between the photoexcited 2Me-
DABT to ThT clearly indicated that ThT could not replace the
2Me-DABT from amyloid fibrils. Rather, the added ThT occupied
some other suitable available sites in amyloid fibrils without re-
placing the preoccupied 2Me-DABT. In fact, it is quite expected
that amyloid fibrils made from large proteins, like insulin, will
have large numbers of binding sites such that even if some of
the sites are occupied by one probe, the other probe can find
another energetically suitable site without replacing the first
probe. Additionally, our studies further confirmed that the
binding sites for ThT and 2Me-DABT are proximal enough for
an efficient FRET between them, which is further supported by
our molecular docking studies discussed in the later part.
The FRET between photoexcited 2Me-DABT and ThT in the
fibrillar solution has also been studied by time-resolved fluo-
rescence measurements. The emission transient decay of 2Me-
As 2Me-DABT has a very similar molecular structure to ThT, it
was important to know whether these two amyloid probes
have similar or independent binding sites in the insulin amy-
loid fibrils. For this, we investigated the emission characteristics
of 2Me-DABT in a fibrillar solution in the presence of different
concentrations of ThT. There was a substantial decrease in the
emission intensity of 2Me-DABT at 445 nm, on addition of ThT
to the fibrillar solution (Figure 5). Such a decrease in the emis-
sion intensity of 2Me-DABT is also accompanied by a large in-
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
ration between 2Me-DABT and ThT was calculated to be 41.2 ꢂ
(see the Supporting Information for detailed calculations). Such
a large separation between ThT and 2Me-DABT further sug-
gested that these two amyloid probes did not share the same
localization site in the amyloid fibrils, as indicated by the re-
sults of the steady-state emission studies.
The knowledge of binding sites of the probes is very impor-
tant to understand their sensory activity and for future sensor
development. Hence, we performed the blind molecular dock-
ing studies to find out the location of 2Me-DABT in amyloid fi-
brils. Due to lack of detail structural information on insulin
amyloid fibrils, in the present docking studies, we used the fi-
brillar structure for 42 residue amyloid-b protein (Ab1-42) deter-
mined by solid-state NMR (PDB ID: 2MXU).[49] The use of the
Ab1-42 fibril rather than fibril of the most abundant 40 residue
amyloid-b protein (Ab1–40) in the present work was motivated
from the fact that the protofibril of the Ab–42 protein is much
more toxic and showed more propensity towards aggregation
compared to the Ab1-40 protein.[50,51] Docking of 2Me-DABT in
Ab1-42 fibril was performed with several initial positions of
ligand with respect to the host fibril. The initial docking con-
sisted of 1000 runs, and all the results were clustered with
a RMSD value of 2 ꢂ. After clustering, two primary binding
sites for 2Me-DABT in Ab1-42 fibrils were identified. These two
most-probable and lowest-energy binding conformations in-
volved localization of 2Me-DABT in the inner channel of fibrils
(Figure 7A). The binding energies for mode-1 (with a popula-
tion 750 out of 1000) and mode-2 (with a population 180 out
of 1000) were estimated to be 6.87 and 6.39 kcalmolꢀ1, respec-
tively. Different interactions operating between 2Me-DABT and
the amino acids of amyloid fibrils for most populated mode of
binding (mode-1) are shown in Figure 7B. It was found that
the binding of 2Me-DABT in the amyloid fibrils was mainly sup-
ported by the p–p stacking interaction between the aromatic
rings of 2Me-DABT and the nearby histidine moiety (His14).
Further, the p–p interaction between the aromatic rings of
2Me-DABT with amide bonds of amyloid fibrils was also opera-
tive in the binding process. Besides, the hydrophobic interac-
tion with the CH2 groups of the peptide chain also favored the
incorporation of 2Me-DABT in the inner-core region of the fi-
brils. In addition, binding at the external grooves of the fibrils
was also observed in the present system (see Figure S4 in the
Supporting Information). The population of such surface
groove binding was extremely low (6 out of 1000) with a rela-
tively low binding energy (5.45 kcalmolꢀ1). Importantly, unlike
ThT, binding of 2Me-DABT at two ends of the fibrils could not
be observed in our docking studies.[52] The results are in agree-
ment with the simulation results reported by Wu et al., who
showed that the probability of binding at two ends of the
beta-sheet structure is much higher for ThT compared to its
neutral analogues.[53]
Figure 5. Emission spectra (lex =340 nm) of 2Me-DABT in insulin amyloid fi-
brillar solution in the presence of different concentrations of ThT (0–3 mm).
The dashed curve shows the emission spectrum (lex =340 nm) of 3 mm ThT
in the fibrillar solution. Inset: variation in the emission intensity at (circles)
445 and (triangles) 490 nm with ThT concentration in the fibrillar solution.
DABT in fibrillar solution became faster with the increase in
ThT concentration (cf. Figure 6). It is important to mention that
no ThT emission at 445 nm from fibrillar solution was observed
on photoexcitation with 340 nm light. Hence, emission decays
recorded at 445 nm were solely from the photoexcited 2Me-
DABT bound to amyloid fibrils. It is evident from Figure 6 that
the addition of ThT to the fibrillar solution led to quenching of
the excited state of 2Me-DABT. Thus, the average lifetime of
2Me-DABT in insulin amyloid fibrils was shortened from 4.3 to
1.7 ns in the presence of 3 mm ThT. Overall, the time-resolved
emission studies further supported the energy transfer from
photoexcited 2Me-DABT to ThT molecules in the fibrillar solu-
tion. The efficiency (E) for the FRET between photoexcited
2Me-DABT and ThT in the fibrillar solution was calculated using
the following equation and found to be 60.5%:[48]
htDA
i
ð2Þ
E ¼ 1ꢀ
htDi
in which htDAi and htDi are the amplitude-weighted lifetimes
[cf. Eq. (5)] of fibril-bound 2Me-DABT in the presence and ab-
sence of ThT, respectively. Using this E value, the average sepa-
To further identify the relative location of 2Me-DABT and ThT
in amyloid fibrils, we performed molecular docking using the
fibril with docked 2Me-DABT (mode-1) as the initial host struc-
ture. The most populated and the strongest mode of binding
showed that the added ThT is also localized in the inner core
of the fibril. The relative position of ThT with respect to 2Me-
Figure 6. Emission transient decays for insulin amyloid fibril-bound 2Me-
DABT in the presence of different ThT concentrations (0–3 mm). The dashed
curve represents the IRF.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
2Me-DABT upon binding to amyloid fibrils makes it a potential
ratiometric sensor and hence can provide high sensitivity in
the detection of amyloid fibrils by fluorescence imaging tech-
niques. Detailed spectroscopic studies indicated that the large
changes in the emission properties of 2Me-DABT in amyloid fi-
brils was due to low polarity of the hydrophobic pockets in
amyloid fibrils in which 2Me-DABT resided. FRET and molecular
docking studies confirmed that the binding locations for ThT
and 2Me-DABT are distinct and separated from each other by
an average distance of approximately 40 ꢂ. Further, as a neutral
molecule, 2Me-DABT may be more efficient for in vivo imaging
of amyloid fibrils. The knowledge on the binding sites and
mode of interaction between probe and fibrils, obtained from
the present study, will help in understanding the interaction of
benzothiazole-based small molecules with amyloid fibrils.
Experimental Section
The ground-state absorption and the steady-state emission spectra
were recorded using a Jasco spectrophotometer (model V-530)
and Hitachi spectrofluorometer (model F4500), respectively. Fluo-
rescence quantum yield (f) values were estimated by a compara-
f
tive method[48] using 4’,6-diamidino-2-phenylindole (DAPI) in di-
Figure 7. A) The two most probable binding sites for 2Me-DABT in amyloid
fibrils obtained from the molecular docking studies. B) A close look of the
binding pocket for mode-1. The dashed lines indicated different types of in-
teraction involved between 2Me-DABT and fibril.
methyl sulfoxide as the reference (f ꢁ0.58).[54] In steady-state
f
emission studies, 2Me-DABT was excited at 340 nm to avoid the
emission from insulin protein. The time-resolved emission measure-
ments were performed using a time-correlated single-photon
counting (TCSPC)-based spectrometer from IBH, UK, as described
earlier.[55] In time-resolved experiments, samples were excited with
a 340 nm LED source. The instrument response function (IRF) of
the TCSPC instrument was 1.3 ns. All decay traces were fitted by
following multiexponential function.
DABT in the amyloid fibrils is shown in Figure 8. Our docking
studies showed that ThT was separated from 2Me-DABT by
40.2 ꢂ. The distance between 2Me-DABT and ThT, obtained
from the molecular docking studies is very close to the values
obtained from the FRET studies. Our docking studies further
supported the fact that the site of localization of 2Me-DABT
and ThT are quite different as proposed in the previous sec-
tion.
X
i
IðtÞ ¼
i aieꢀt=t
ð3Þ
in which, ti and ai are the time constant and relative contribution
of the ith decay component to the measured decay traces, respec-
tively. Time (tav) and amplitude (hti)-weighted average lifetimes
were calculated using the following equations.[48]
Conclusions
P
i ait2i
The newly synthesized neutral benzothiazole derivative, 2Me-
DABT, binds strongly with the amyloid fibrils resulting in
a large increase in its emission intensity along with a large
spectral shift. Thus, unlike most extensively used amyloid sen-
sors, such as ThT, a large shift in the emission wavelength of
P
ð4Þ
ð5Þ
tav
¼
i aiti
Siaiti
Siai
hti ¼
Bovine insulin was obtained from Sigma. ThT was purchased as its
chloride salt from Sigma and purified by recrystalization from
a methanol/water mixture. Double-distilled water (conductivity
<0.1 mScmꢀ1) was used for the sample preparation. All chemicals
used for the synthesis were of analytical grade. Due to extremely
low solubility of 2Me-DABT in neat water, its aqueous solution was
prepared by using a concentrated stock solution in methanol such
that the methanol content in the final solution was <0.5% by
volume. In all experiments, absorbance of 2Me-DABT solution was
in the range 0.01–0.02 at 300 nm.
Insulin fibrils were prepared by following the procedure reported
in the literature.[30,56] Insulin solution (2 mgmLꢀ1) in 20% acetic
acid (pH 1.6) was heated at 708C for 24 h without stirring. The
completion of the fibrillation process was monitored through the
standard ThT fluorescence assay. Fibrils, thus prepared in the acidic
Figure 8. Binding pose of ThT in the amyloid fibril with incorporated 2Me-
DABT. The solid line indicates the distance between 2Me-DABT and ThT.
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
medium, were diluted 12-fold with Tris-HCl buffer and the pH was
adjusted to 7.4 with aqueous NaOH solution.
A cubical box with dimensions of 100ꢁ100ꢁ100 ꢂ3 with 0.6 ꢂ grid
spacing was created at the center of the fibril. The size of the grid
box was made in such that the ligand can access all possible bind-
ing sites in fibrils. Each docking involved 200 independent runs
with
a
maximum number of 5ꢁ106 energy evaluation and
Synthesis of 2Me-DABT
27000 generations. The Lamarckian genetic algorithm (LGA)[62]
method was applied to find the docked conformations of the
ligand with the lowest energy. Results of all 1000 runs were clus-
tered with root mean square deviation (RMSD) <2 ꢂ and ranked
according to their binding energies. For docking with ThT, fibril
with a docked 2Me-DABT was used as the host molecule. A similar
procedure as used for the docking of 2Me-DABT with fibril was fol-
lowed.
The preparation of 2Me-DABT is outlined in Scheme 2.
Acknowledgements
We are thankful to Prof. A. Datta, Indian Institute of Technolo-
gy, Bombay, for his help in quantum chemical calculations. A.A.
is thankful to the Indian Science Academies (IASC, INSA &
NASI) for the summer research fellowship. S.M. thanks the
Board of Research in Nuclear Sciences (BRNS) for a scholarship.
Scheme 2. Synthesis of 2Me-DABT.
4-(Dimethylamino)-2-methylbenzaldehyde (I)
Phosphorous oxychloride (0.78 mL, 8.1 mm) was added dropwise
under an argon atmosphere to a cooled (ꢀ108C) and stirred solu-
tion of N,N-3-trimethylaniline (1.1 mL, 7.4 mm) and N,N-dimethylfor-
mamide (0.7 mL, 8.9 mm) in dichloromethane (40 mL). The reaction
mixture was brought to room temperature, stirred for 4 h, treated
with aqueous 2m sodium hydroxide, and stirred for 4 h followed
by the extraction with ethyl acetate. The organic extract was
washed with water and brine, dried, concentrated in vacuo, and
the residue purified by column chromatography (silica gel) to give
Keywords: amyloid fibrils · biosensors · fluorescent probes ·
thioflavin-T · time-resolved spectroscopy
[1] J. D. Sipe, Annu. Rev. Biochem. 1992, 61, 947–975.
[2] L. C. Serpell, Biochim. Biophys. Acta Mol. Basis Dis. 2000, 1502, 16–30.
[3] F. Chiti, C. M. Dobson, Annu. Rev. Biochem. 2006, 75, 333–366.
[4] D. R. Booth, M. Sunde, V. Bellotti, C. V. Robinson, W. L. Hutchinson, P. E.
Fraser, P. N. Hawkins, C. M. Dobson, S. E. Radford, C. C. Blake, M. B.
Pepys, Nature 1997, 385, 787–793.
1
the desired aldehyde (0.5 g, 41.6%). H NMR (200 MHz, CDCl3): d=
2.36 (s, 3H), 2.9 (s, 6H), 6.80–6.84 (d, 2H, J=8.0 Hz), 7.65 (d, 1H,
J=8.0 Hz), 10.14 ppm (s, 1H).
[5] S. J. Hubbard, Biochim. Biophys. Acta Protein Struct. Mol. Enzymol. 1998,
1382, 191–206.
[6] J. D. Sipe, A. S. Cohen, J. Struct. Biol. 2000, 130, 88–98.
[7] A. Lomakin, D. S. Chung, G. B. Benedek, D. A. Kirschner, D. B. Teplow,
Proc. Natl. Acad. Sci. USA 1996, 93, 1125–1129.
[8] C. L. Oliveira, M. A. Behrens, J. S. Pedersen, K. Erlacher, D. Otzen, J. Mol.
Biol. 2009, 387, 147–161.
[9] G. Zandomeneghi, M. R. H. Krebs, M. G. McCammon, M. Fꢃndrich, Pro-
tein Sci. 2009, 13, 3314–3321.
[10] S. Guo, B. B. Akhremitchev, Biomacromolecules 2006, 7, 1630–1636.
[11] D. Yanagisawa, H. Taguchi, N. F. Ibrahim, S. Morikawa, A. Shiino, T. Inu-
bushi, K. Hirao, N. Shirai, T. Sogabe, I. Tooyama, J. Alzheimers Dis. 2014,
39, 617–631.
[12] M. Higuchi, N. Iwata, Y. Matsuba, K. Sato, K. Sasamoto, T. C. Saido, Nat.
Neurosci. 2005, 8, 527–533.
2Me-DABT
2-Aminothiophenol (0.49 g, 4.0 mm) was added to a stirred solu-
tion of I (3.1 mm) in dimethyl sulfoxide (25 mL) and the reaction
mixture was heated at 1808C for 15 min. After cooling to room
temperature, the mixture was poured into water and extracted
with ethyl acetate. The organic layer was washed with water and
brine, and dried. Removal of the solvent in vacuo produced
a yellow solid, which was purified by column chromatography
1
(silica gel) to give pure 2Me-DABT (0.61 g, 74%). H NMR (200 MHz,
CDCl3): d=2.39 (s, 3H), 2.73 (s, 6H), 7.02–7.10 (m, 2H), 7.29–7.33
(m, 1H), 7.36–7.48 (m, 1H), 7.87 (d, 1H, J=8 Hz), 8.03 (d, 1H, J=
8 Hz), 8.24 ppm (d, 1H, J=8 Hz).
[13] W. E. Klunk, H. Engler, A. Nordberg, Y. Wang, G. Blomqvist, D. P. Holt, M.
Bergstrçm, I. Savitcheva, G.-F. Huang, S. Estrada, B. Ausꢄn, M. L. Deb-
nath, J. Barletta, J. C. Price, J. Sandell, B. J. Lopresti, A. Wall, P. Koivisto,
G. Antoni, C. A. Mathis, B. Lꢅngstrçm, Ann. Neurol. 2004, 55, 306–319.
[14] A. Nordberg, Lancet Neurol. 2004, 3, 519–527.
[15] W. Zhen, H. Han, M. Anguiano, C. A. Lemere, C.-G. Cho, P. T. Lansbury, Jr.,
J. Med. Chem. 1999, 42, 2805–2815.
[16] Y. Wang, W. E. Klunk, M. L. Debnath, G.-F. Huang, D. P. Holt, L. Shao, C. A.
Mathis, J. Mol. Neurosci. 2004, 24, 55–62.
[17] Y. Uchida, H. Takahashi, Neurosci. Lett. 2008, 448, 279–281.
[18] D. Allsop, L. Swanson, S. Moore, Y. Davies, A. York, O. M. A. El-Agnaf, I.
Soutar, Biochem. Biophys. Res. Commun. 2001, 285, 58–63.
[19] M. Ono, H. Watanabe, H. Kimura, H. Saji, ACS Chem. Neurosci. 2012, 3,
319–324.
[20] A. Ojida, T. Sakamoto, M.-A. Inoue, S.-H. Fujishima, G. Lippens, I. Hama-
chi, J. Am. Chem. Soc. 2009, 131, 6543–6548.
[21] C. Ran, X. Xu, S. B. Raymond, B. J. Ferrara, K. Neal, B. J. Bacskai, Z. Medar-
ova, A. Moore, J. Am. Chem. Soc. 2009, 131, 15257–15261.
Computational methods
AutoDock suite (version 4.2)[57] implemented in the AutoDock Tools
(version 1.5.6)[58] was used for the molecular docking studies. The
structure of Ab1–42 fibril was obtained from the PDB data bank
(PDB ID: 2MXU). The structures of 2Me-DABT and ThT were energy-
optimized by quantum chemical calculations using a Gaussian 03
package.[59] B3LYP functional[60,61] and 6-311+ +g(d,p) basis func-
tion were used to optimize the chemical structures by DFT. The
energy-optimized structure of 2Me-DABT and ThT are shown in Fig-
ure S5 in the Supporting Information and used for blind docking
studies. For docking studies, fibril was considered as a rigid host
molecules and 2Me-DABT as a flexible ligand allowing all torsional
motion in the molecule. Molecular docking was performed with
five different initial locations of the ligand with respect to the fibril.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[22] R. Chandra, M.-P. Kung, H. F. Kung, Bioorg. Med. Chem. Lett. 2006, 16,
1350–1352.
[46] A. Barik, S. Nath, H. Pal, J. Chem. Phys. 2003, 119, 10202–10208.
[47] R. Mishra, D. Sjçlander, P. Hammarstrçm, Mol. BioSyst. 2011, 7, 1232–
1240.
[48] J. R. Lackowicz, Principles of fluorescence spectroscopy, 3rd ed., Springer,
New York, 2006.
[23] V. Sathish, E. Babu, A. Ramdass, Z.-Z. Lu, M. Velayudham, P. Thanasekar-
an, K.-L. Lu, S. Rajagopal, Talanta 2014, 130, 274–279.
[24] C. Wang, D. Liu, Z. Wang, Chem. Commun. 2012, 48, 8392–8394.
[25] P. Gagni, L. Sola, M. Cretich, M. Chiari, Biosens. Bioelectron. 2013, 47,
490–495.
[49] Y. Xiao, B. Ma, D. McElheny, S. Parthasarathy, F. Long, M. Hoshi, R. Nussi-
nov, Y. Ishii, Nat. Struct. Mol. Biol. 2015, 22, 499–505.
[26] H. LeVine III, Protein Sci. 2008, 2, 404–410.
[27] M. Groenning, J. Chem. Biol. 2010, 3, 1–18.
[28] M. Biancalana, S. Koide, Biochim. Biophys. Acta Proteins Proteomics 2010,
1804, 1405–1412.
[29] A. Srivastava, P. K. Singh, M. Kumbhakar, T. Mukherjee, S. Chattopadyay,
H. Pal, S. Nath, Chem. Eur. J. 2010, 16, 9257–9263.
[50] L. M. Luheshi, G. G. Tartaglia, A.-C. Brorsson, A. P. Pawar, I. E. Watson, F.
Chiti, M. Vendruscolo, D. A. Lomas, C. M. Dobson, D. C. Crowther, PLoS
Biol. 2007, 5, e290.
[51] J. Davis, W. E. Van Nostrand, Proc. Natl. Acad. Sci. USA 1996, 93, 2996–
3000.
[52] C. Wu, M. T. Bowers, J.-E. Shea, Biophys. J. 2011, 100, 1316–1324.
[53] C. Wu, Z. Wang, H. Lei, Y. Duan, M. T. Bowers, J.-E. Shea, J. Mol. Biol.
2008, 384, 718–729.
[30] P. K. Singh, M. Kumbhakar, H. Pal, S. Nath, J. Phys. Chem. B 2010, 114,
2541–2546.
[31] C. Rodrꢆguez-Rodrꢆguez, A. Rimola, L. Rodrꢆguez-Santiago, P. Ugliengo,
A. ꢇlvarez-Larena, H. Gutiꢄrrez-De-Terꢈn, M. Sodupe, P. Gonzꢈlez-Duarte,
Chem. Commun. 2010, 46, 1156–1158.
[32] W. E. Klunk, Y. Wang, G. F. Huang, M. L. Debnath, D. P. Holt, C. A. Mathis,
Life Sci. 2001, 69, 1471–1484.
[33] A. A. Maskevich, V. I. Stsiapura, V. A. Kuzmitsky, I. M. Kuznetsova, O. I. Po-
varova, V. N. Uversky, K. K. Turoverov, J. Proteome Res. 2007, 6, 1392–
1401.
[34] M. D’Amico, M. G. Di Carlo, M. Groenning, V. Militello, V. Vetri, M. Leone,
J. Phys. Chem. Lett. 2012, 3, 1596–1601.
[35] R. L. Yona, S. Mazeres, P. Faller, E. Gras, ChemMedChem 2008, 3, 63–66.
[36] C. A. Mathis, B. J. Bacskai, S. T. Kajdasz, M. E. McLellan, M. P. Frosch, B. T.
Hyman, D. P. Holt, Y. Wang, G.-F. Huang, M. L. Debnath, W. E. Klunk,
Bioorg. Med. Chem. Lett. 2002, 12, 295–298.
[37] A. Lockhart, L. Ye, D. B. Judd, A. T. Merrittu, P. N. Lowe, J. L. Morgenstern,
G. Hong, A. D. Gee, J. Brown, J. Biol. Chem. 2005, 280, 7677–7684.
[38] G. Grynkiewicz, M. Poenie, R. Y. Tsien, J. Biol. Chem. 1985, 260, 3440–
3450.
[39] F. Crespi, A. C. Croce, S. Fiorani, B. Masala, C. Heidbreder, G. Bottiroli,
Lasers Surg. Med. 2004, 34, 39–47.
[40] M. L. Benesi, J. H. Hildebrand, J. Am. Chem. Soc. 1949, 71, 2703–2707.
[41] S. Murudkar, A. K. Mora, S. Jakka, P. K. Singh, S. Nath, J. Photochem. Pho-
tobiol. A 2014, 295, 17–25.
[54] T. Hard, P. Fan, D. R. Kearns, Photochem. Photobiol. 1990, 51, 77–86.
[55] A. K. Mora, P. K. Singh, S. Nath, J. Phys. Chem. B 2016, 120, 4143–4151.
[56] K. J. Robbins, G. Liu, V. Selmani, N. D. Lazo, Langmuir 2012, 28, 16490–
16495.
[57] G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew, D. S. Good-
sell, A. J. Olson, J. Comput. Chem. 2009, 30, 2785–2791.
[58] M. F. Sanner, J. Mol. Graph. Model. 1999, 17, 57–61.
[59] Gaussian 03, Revision C.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsu-
ji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Na-
kajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakr-
zewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G.
Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford,
CT, 2004.
[42] M. Groenning, M. Norrman, J. M. Flink, M. van de Weert, J. T. Bukrinsky,
G. Schluckebier, S. Frokjaer, J. Struct. Biol. 2007, 159, 483–497.
[43] R. Sabatꢄ, I. Lascu, S. J. Saupe, J. Struct. Biol. 2008, 162, 387–396.
[44] P. K. Singh, A. K. Mora, S. Nath, Chem. Commun. 2015, 51, 14042–14045.
[45] Dielectric constants (ems) of the mixed solvents were calculated by
using the equation ems =v1e1 +v2e2, in which vi and ei are the volume
fraction and the dielectric constant of the constituent solvents; see ref-
erence [46].
[60] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[61] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[62] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart, R. K.
Belew, A. J. Olson, J. Comput. Chem. 1998, 19, 1639–1662.
Received: June 22, 2016
Published online on && &&, 0000
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
Bound and determined: Neutral thiofla-
vin-T derivative, 2Me-DABT, showed
a large increase in its emission intensity
as well as a significant blue shift in the
emission spectrum due to its associa-
tion with amyloid fibrils (see figure). Mo-
lecular docking studies confirmed the
binding of 2Me-DABT in the inner core
of the amyloid fibrils.
&
Fluorescent Amyloid Probe
A. K. Mora, S. Murudkar, A. Alamelu,
P. K. Singh, S. Chattopadhyay, S. Nath*
&& – &&
Benzothiazole-Based Neutral
Ratiometric Fluorescence Sensor for
Amyloid Fibrils
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ