Fluorescence solvatochromism and modulated anticholinergic activity of novel coumarin compounds sequestered in human serum albumin nanocavities

Article abstract of DOI:10.1039/c9nj03293b

A coumarin compound, 3,3′-methylenebis(4-hydroxy-2H-chromen-2-one) (MHC), and its substituted derivative, 3,3′-(phenylmethylene)bis(4-hydroxy-2H-chromen-2-one) (MHCB), possess potent anticholinergic activities, which were found to be reduced significantly in the presence of human serum albumin (HSA). The molecular interactions responsible for sequestering MHC and MHCB were explored by steady-state and time-resolved fluorescence using the modulated solvatochromic behavior of the probes as a fluorescent marker. A quantitative description of the different solvent parameters responsible for the notable solvent-dependent photo-physical properties of the investigated systems was extracted from multiple linear regression analysis of the experimental data using the Kamlet-Taft and Catalán formalisms. A series of complimentary studies involving circular dichroism measurement and molecular docking calculations revealed that the binding of both coumarin derivatives resulted in the stabilization of the α-helical structure of human serum albumin (HSA). However, significant differences in binding mechanism were noted in terms of the strength, mode and principal forces responsible for the spontaneous association of the probes into the protein-binding domain.

Full text of DOI:10.1039/c9nj03293b

NJC
View Article Online
View Journal | View Issue
PAPER
Fluorescence solvatochromism and modulated
anticholinergic activity of novel coumarin
compounds sequestered in human serum
albumin nanocavities†
Cite this: New J. Chem., 2019,
43, 18713
Mostofa Ataur Rohman, ‡ Prayasee Baruah,‡ Deboshika Bhattacharjee,
B. Myrboh and Sivaprasad Mitra
*
A coumarin compound, 3,30-methylenebis(4-hydroxy-2H-chromen-2-one) (MHC), and its substituted
derivative, 3,30-(phenylmethylene)bis(4-hydroxy-2H-chromen-2-one) (MHCB), possess potent anticholinergic
activities, which were found to be reduced significantly in the presence of human serum albumin (HSA). The
molecular interactions responsible for sequestering MHC and MHCB were explored by steady-state and
time-resolved fluorescence using the modulated solvatochromic behavior of the probes as a fluorescent
marker. A quantitative description of the different solvent parameters responsible for the notable solvent-
dependent photo-physical properties of the investigated systems was extracted from multiple linear
regression analysis of the experimental data using the Kamlet–Taft and Catalan formalisms. A series of
´
complimentary studies involving circular dichroism measurement and molecular docking calculations
revealed that the binding of both coumarin derivatives resulted in the stabilization of the a-helical
structure of human serum albumin (HSA). However, significant differences in binding mechanism were
noted in terms of the strength, mode and principal forces responsible for the spontaneous association of
the probes into the protein-binding domain.
Received 25th June 2019,
Accepted 1st November 2019
DOI: 10.1039/c9nj03293b
rsc.li/njc
In addition, this group of compounds shows a very high
fluorescence quantum yield. Hence, they behave as promising
Introduction
Coumarins, which are fragment organic compounds from the fluorescent tags for probing a series of different complex
benzopyrone class, are naturally found in many plants and widely chemical as well as biological media because of their strong
used in certain perfumes and fabric conditioners.1–3 Since their polarity-dependent Stokes shift and substantial change in
first isolation in 1820, the coumarin group of compounds have dipole moment on excitation.19–21 Studies on the interaction of
been extensively studied due to their industrial, pharmaceutical, bioactive compounds with biologically relevant macromolecules
antimicrobial, antioxidant and anticancer properties.4–10 The are important in the scientific and medical fields since they
improved pharmacokinetic behaviour in coumarins is believed provide the basic foundation for their application in the phar-
to be due to their rapid absorption and metabolism in the body.11 maceutical sciences.22,23 In a recent report, the development of
Further, coumarins have also been identified as active therapeutic a series of multispectral methods to exploit the interaction of
agents having anticoagulant activity,12 immuno-stimulatory HSA with two coumarin derivatives, 3-(furan-2-yl)-6-methyl-2H-
properties for treating chronic infections,13 anti-inflammatory chromen-2-one (FM) and 6-chloro-3-(furan-2-yl)-2H-chromen-2-
property,12 and potent edema protective functions.14
one (CM), was presented.24
Fluorescent coumarins are also popular as dye materials15 and in
Recently, a substituted coumarin derivative, namely 3,30-
sensory applications to detect several metal cations.16,17 Reports are methylenebis(4-hydroxy-2H-chromen-2-one) (MHC) was reported
also available to justify their use as efficient dopant candidates as potential inhibitor for acetylcholinesterase (AChE).25 In fact,
for the generation of organic light-emitting diodes (OLEDs).18 enzymatic hydrolysis data revealed that the AChE inhibition
property of MHC is comparable even with the FDA approved
Centre for Advanced Studies in Chemistry, North-Eastern Hill University,
cholinergic drug donepezil, opening up the possibility of its use
as potential drug for the treatment of Alzheimer’s Disease (AD).
However, the effectiveness of pharmacologically screened drug
Shillong 793 022, India. E-mail: smitra@nehu.ac.in, smitranehu@gmail.com
† Electronic supplementary information (ESI) available: Detailed protocol for the
estimation of anticholinergic activity (ST1), Tables S1–S6 and Fig. S1–S5. See DOI:
molecules is strongly dependent on their lipophilicity and/or
hydrogen bond (HB) formation ability. Further, molecular docking
10.1039/c9nj03293b
‡ Equal contributors to the work.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18713

View Article Online
Paper
NJC
calculations revealed the importance of hydrogen bond formation
in the binding process of MHC in the peripheral anionic site (PAS)
of AChE. Therefore, it is imperative to analyse the nature of the
solute–solvent interaction quantitatively in MHC, which has been
screened as a potential AChE inhibitor (AChEI).
The solvent-dependent fluorescence behaviour of differently
substituted coumarin derivatives has also attracted significant
attention recently since different spectroscopic parameters
such as position of the fluorescence maximum, fluorescence
yield and the excited-state lifetime of the investigated coumarin
compounds are found to be strongly dependent on the polarity
and hydrogen bonding ability of the solvent and also on the
nature and position of substitution in the parent coumarin
scaffold. Therefore, understanding the solvent-dependent
photophysical behaviour of coumarin derivatives and identifying
the quantitative contribution of microscopic solvent properties
are still an active area of research.
Scheme 1 Synthetic route and structural representation of the coumarin
compounds investigated in the present study.
tendency of human serum albumin (HSA). Therefore, the results
on the spectral modulation and nature of the molecular inter-
action of both systems are also presented in the presence of a
series of heterogeneous media such as cyclodextrin and HSA.
Results and discussion
The nature of different types of interactions results in a
marked change in the excited state in comparison with the
ground state, specifically due to the extensive electron density
redistribution upon excitation.26 Normally, the description of
the non-specific interaction of organic solutes with solvent
involves polarity and/or polarizability. The extent of interaction
depends on the solvent dielectric constant (e), refractive index
(n) and the dipole moment (m) of the solute in its Franck–
Condon (FC) and/or relaxed fluorescent (RF) states. The corres-
ponding spectral indicator such as absorption (nabs) or fluorescence
(nfluo) maximum (in wavenumber) of the solute varies linearly with
solvent polarity function, which is the orientation polarizability
(Df ) according to the Lippert–Mataga relation.27,28
Estimation of anticholinergic activity and its modulation in HSA
The kinetic parameters for the normal and inhibited enzyme
hydrolysis pathways in the presence of the investigated coumarin
compounds were obtained via the Michaelis–Menten (MM) analysis
based on Scheme 2. Both coumarin compounds were found to
inhibit acetylcholinesterase (AChE) activity in a non-competitive
manner as apparent from the MM analysis. The comparative kinetic
data of AChE inhibition and its modulation in the presence of HSA
for both systems is shown in Fig. 1. The MM constant (Km)
remained practically unchanged, whereas, a notable alteration was
seen in the Vmax value. Also, the values of a and a0 were found to be
equal (where a = 1 + ([EI])/([E]) and a0 = 1 + ([IES])/([ES]), respectively)
since the inhibitor has equal probabilities of binding to the enzyme
(E) and the enzyme–substrate (ES) complex (see ST1 in ESI†). In the
case of non-competitive inhibition, inhibitors have been found to
attach to the peripheral anionic site (PAS) of AChE, thereby posing
no interference to the substrate binding catalytic anionic site
(CAS).36 The estimated IC50 values (Table 1) obtained from the
modified Hill equation justifies both investigated coumarins as
potent AChEIs, which have bright promise for possible utilization
as a therapeutic avenue towards Alzheimer’s disease (AD). The
greater AChE inhibition potency for MHC and MHCB in
A deviation in the observable spectroscopic behaviour from
the traditional equations for the non-specific interactions mentioned
above often indicates the presence of specific interactions of
the solute with the solvent, mostly arising due to the formation
of hydrogen bonds (HB) between the solute and the solvent.
Solvatochromic studies quantifying the specific solute–solvent
interaction were considered by employing ET(30), a uni-parametric
scale developed by Dimroth and Reichard,29 which represents the
HB donation acidity (a) of protic solvents. A more general multi-
parametric linear solvation energy relationship (LSER), formalized
first by Kamlet, Taft and coworkers30 and further developed by
Catalan,31 includes both a and solvent HB acceptance basicity (b).
´
A series of recent literature reported both treatments to quantify
the solute–solvent interaction for many organic heterocycles.32–35
In the present investigation, the anticholinergic activity of
3,30-(phenylmethylene)bis(4-hydroxy-2H-chromen-2-one) (MHCB)
is estimated and compared with that of MHC (structure shown in
Scheme 1) described earlier.25 A comprehensive description of
the photophysical and fluorescence spectral behaviour of both
compounds is presented in a series of different solvents both in
´
terms of the KT and Catalan models. Interestingly, MHCB
possess very close structural resemblance with warfarin, one of
the most extensively studied fluorescent biomarkers for serum
albumin proteins. Also, the AChE inhibitory efficiency of the
investigated coumarins was shown to decrease significantly in
Scheme 2 Mechanistic representation of the normal and inhibited enzymatic
activity of AChE. The competitive and uncompetitive inhibition paths are
represented by (A) and (B), respectively, while the non-competitive inhibition
involves both the pathways. The rate constants and the dissociation constants
serum albumin matrix, which signifies the strong sequestering of individual steps are represented by k and K, respectively.
18714 | New J. Chem., 2019, 43, 18713--18725
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
Fig. 1 (a) Hydrolysis curve (scattered points) for AChE activity (solid
squares) and its inhibition in the presence of 5 nM (open circle) or 10 nM
(solid circle) of MHCB in phosphate buffer solution of pH = 8.0. Inset
shows the same experiment in HSA matrix. The solid line represents non-linear
regression of the experimental data in each case. [AChE] = 0.079 U mLꢁ1
.
(b) Modified Hill analysis of AChE inhibition by MHC (circle) and MHCB (square)
in aqueous buffer (solid points) and in the presence of HSA (empty points).
Inset shows the corresponding IC50 values.
comparison with structurally similar chromone derivatives37 is
consistent with well-established structure activity relationship (SAR)
reviews that the replacement of the coumarin skeleton with a
chromone moiety results in the loss of AChE inhibition potency.38
Human serum albumin has been established as an effective
bio-mimicking medium for AChE inhibition since the potencies of
drugs have been found to be greatly modulated in the presence of
HSA compared with aqueous buffer. Similarly, in the present
investigation, HSA was found to decrease the inhibition potency
of both MHC and MHCB considerably. For example, the estimated
IC50 value in buffer medium for MHCB is 98.12 ꢀ 7.9 compared to
80.16 ꢀ 1.3 nM for MHC. However, in HSA matrix, this value
changed to 117.13 and 112.14 nM, respectively. This shows that
although in buffer medium the potency of MHC is higher, it is
rendered almost equal for both compounds in HSA medium.
Interestingly the relative change in IC50 value39 was estimated to
be 39% for MHC in comparison with only 20% for MHCB. These
results indicate that the sequestering ability of HSA towards MHC
is much stronger than that in the case of MHCB and in excellent
agreement with the fluorescence titration results described later.
Fig. 2 Steady-state fluorescence excitation (left panel) and emission
(right panel) spectra of MHC (a) and MHCB (b) in selected solvents. Solvent
abbreviations: 1,4-dioxane (DIO), cyclohexane (CHX), toluene (TOL), tetra-
hydrofuran (THF), dimethyl sulfoxide (DMSO), acetonitrile (ACN), methyl
alcohol (MeOH), and n-butanol (BuOH).
other hand, the corresponding emission spectral peak appears
in the range of 344–404 nm and 385–557 nm in different
solvents for MHC and MHCB, respectively. The excitation
spectra obtained by monitoring the fluorescence at the respective
emission maximum presented unstructured profiles and appear
within the corresponding absorption spectral range in both
cases. The representative fluorescence emission and excitation
spectral profiles for both MHC and MHCB in some selected
solvents are shown in Fig. 2. Careful inspection of the emission
spectral profiles reveals multi-dimensional solvent dependence
for both systems. For example, the broad emission peak for
MHC in water with a full width at half maximum (FWHM) of
5221 cmꢁ1 shows the largest bathochromic shift (lmax = 404 nm),
while the corresponding spectrum in cyclohexane is extremely
narrow (FWHM = 3987 cmꢁ1) and blue-shifted (lmax = 350 nm).
The emission peak maximum shows a continuous red shift with
Steady-state spectral properties and solvent effect
The absorption profiles for both MHC and MHCB show a structured an increase in polarity of the medium. In contrast, for MHCB, the
pattern and appear within the spectral range of 270–360 nm in emission peak position in water and cyclohexane appears at
all the solvents studied here (shown in ESI,† Fig. S1). On the 345 and 543 nm, respectively. There is no straightforward
Table 1 Kinetic data of AChE hydrolysis and the effect of different concentrations of MHCB as an inhibitor (I) on various parameters in aqueous buffer
media and serum albumin matrixa
Km/mM
From Michaelis–Menten equation
In buffer medium
Vmax/nM sꢁ1
a
a0
IC50 (nM)
nH
From modified Hill equation
System
[I] = 0 nM
[I] = 5 nM
[I] = 10 nM
185 ꢀ 20
817 ꢀ 24
1.0
1.2
1.3
1.0
1.2
1.3
98.12 ꢀ 7.9 (80.16 ꢀ 7.1)
1.04 ꢀ 0.08 (1.51 ꢀ 0.1)
188 ꢀ 23 (174 ꢀ 24)
166 ꢀ 19 (166 ꢀ 31)
692 ꢀ 21 (617 ꢀ 24)
623 ꢀ 31 (527 ꢀ 37)
In HSA matrix
[I] = 0 nM
[I] = 5 nM
178 ꢀ 17
828 ꢀ 23
1.0
1.1
1.2
1.0
1.1
1.2
117.21 ꢀ 12.3 (110.34 ꢀ 11.6)
1.10 ꢀ 0.11 (1.51 ꢀ 0.1)
190 ꢀ 21 (178 ꢀ 17)
198 ꢀ 19 (174 ꢀ 22)
730 ꢀ 26 (667 ꢀ 18)
685 ꢀ 39 (619 ꢀ 22)
[I] = 10 nM
a
Corresponding data for MHC is given in parenthesis.25
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18715

View Article Online
Paper
NJC
Table 2 Various spectroscopic properties of the investigated chromone derivatives measured in a series of solventsa
Solvent
labs/nm
n
abs/cmꢁ1
lem/nm
n
em/cmꢁ1
Stokes shift/cmꢁ1
ff/10ꢁ3
t
av/ns
kr/107, sꢁ1
Sknr/108, sꢁ1
For MHC
Cyclohexane
Toluene
1,4-Dioxane
Acetonitrile
Methanol
Tetrahydrofuran
DMSO
285
298
301
295
301
300
297
301
300
293
35 088
33 557
33 223
33 898
33 223
33 333
33 670
33 223
33 333
34 130
344
350
360
360
360
372
380
404
353
355
29 070
28 571
27 778
27 778
27 778
26 882
26 316
24 752
28 329
28 169
6018
4986
5445
6121
5445
6452
7354
8470
5005
5961
15.42
2.29
18.86
2.00
0.85
0.76
9.00
68.00
3.02
6.44
8.76
9.13
0.39
2.34
2.44
0.64
1.34
0.84
0.51
2.65
0.20
0.03
4.80
0.09
0.03
0.10
0.70
8.10
0.60
0.20
1.12
1.09
25.15
4.25
4.10
15.61
7.42
11.09
19.50
3.70
Water
Ethyl acetate
n-Butanol
For MHCB
Cyclohexane
Toluene
1,4-Dioxane
Acetonitrile
Methanol
Tetrahydrofuran
DMSO
328
326
338
334
332
340
337
324
340
332
30 488
30 675
29 586
29 940
30 120
29 412
29 674
30 864
29 412
30 120
543
557
466
493
487
459
490
385
465
484
18 416
17 953
21 459
20 284
20 534
21 786
20 408
25 974
21 505
20 661
12 072
12 722
8127
9656
9587
7625
9265
4890
7906
9459
0.46
0.31
2.50
0.42
0.27
0.53
2.30
1.00
1.70
1.30
9.54
7.69
0.72
0.51
1.34
1.19
0.57
0.47
0.91
0.62
0.005
0.004
0.300
0.083
0.020
0.044
0.400
0.200
0.200
0.200
1.04
1.29
13.81
19.72
7.43
8.37
17.53
21.40
10.95
16.18
Water
Ethyl acetate
n-Butanol
a
Corresponding solvent properties are listed in the ESI (Table S1).
correlation between emission maximum and solvent polarity in
this case. However, the emission spectral profiles show a distinctive
shape within these two extremes and the intensity ratio of these two
peaks varies significantly in different solvents. The results indicate
that despite their similar structural motif, the solvent response
to the corresponding fluorescence states of MHC and MHCB is
distinctly different. In addition to the absorption and emission
peak position (labs and lem, respectively), other spectral para-
meters such as Stokes shift (Dnss) and fluorescence yield (ff) are
also listed in Table 2 for both systems.
Fig. 3 Time-resolved fluorescence decay profile (open circles) and simu-
lated data (solid line) of MHC in toluene (left) and MHCB in cyclohexane
(right). IRF indicates the instrument response function. Visual inspection of
the distribution of weighted residuals and autocorrelation function (ac)
confirms the applicability of 3- and 2-exp decay model for simulation of
experimental data in the respective cases.
Time-resolved fluorescence behaviour
Extensive fluorescence lifetime measurements were performed
for both MHC and MHCB in different solvents by monitoring
the fluorescence at their emission maximum. For both systems,
the accepted fitting model and the magnitude of tav is strongly
dependent on the nature of the solvent (Fig. S2, ESI†). For
example, although the experimental traces can be adequately
expressed by a single exponential decay function for MHCB in
most of the solvents studied here (with a decay time in the
range of 0.5–1.3 ns), a two-exponential model was necessary to
reproduce the data in non-interacting solvents such as cyclo-
hexane and/or toluene (Fig. 3). On the other hand, MHC gives a
two-exponential decay in most of the solvents. However, a
minimum of a three-exponential decay function is necessary
to fit the data in cyclohexane and/or toluene with acceptable
statistical parameters. The complete results for the fluorescence
decay fitting along and statistical parameters for both compounds
in different solvents are given in the ESI† (Table S2).
observed that for all the systems, the magnitude of Sknr is quite
high and almost two orders of magnitude larger than the rate
constants for the radiative processes (kr). Thus, the non-
radiative decay process predominates in the deactivation of
the excited state of both coumarins. Overall, the steady-state
and time-resolved fluorescence data of the investigated
coumarin compounds show significant solvent dependence.
However, there is no straightforward correlation between this
solvatochromism with a single physico-chemical property
such as polarity and/or viscosity of the solvent. Therefore, it
is believed that the interaction of these probes with solvent
needs to be modeled by considering both the non-specific and
specific interactions.
The calculated average lifetime (tav) values for all the systems
are also presented in Table 1. Further, estimation of the radiative
Fluorescence studies in mixed solvents
(kr) and total non-radiative (Sknr) decay rate constants in each case The importance of a specific solvent effect in controlling the excited-
was performed using the known values of ff and tav. It was state photo-physical behaviour of the investigated compounds was
18716 | New J. Chem., 2019, 43, 18713--18725
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
further justified by monitoring the fluorescence in a series of mixed hydrogen bonded complexes of MHCB with DMSO and water,
solvents. The fluorescence property showed complex changes in giving the emission spectral peak at 485 and 385 nm, respectively.
different compositions of binary solvent mixtures comprising
The fluorescence decays were collected using a fixed excitation
one or more of the following signatures: (i) momentous change wavelength (lexc = 295 nm) at the magic angle for both com-
in fluorescence intensity, (ii) shift in fluorescence spectral pounds in different mixed solvent compositions. The emission
position, and (iii) notable difference in the width of the fluorescence was monitored at the maximum wavelength of the corresponding
spectrum. Some representative examples of MHC in dimethyl spectrum in the case of MHC. However, time-resolved data was
sulfoxide (DMSO)/water mixture is shown in Fig. 4(a), while the collected both at 385 and 485 nm at each solvent composition for
corresponding figures in DMSO/methanol and water/acetonitrile MHCB. Some of the representative decay profiles are shown in
mixtures are given in the ESI† (Fig. S3). It is interesting to note that Fig. 4c and in the ESI† (Fig. S3). The notable deviation in the
while the fluorescence intensity of MHC in DMSO underwent steady decay profiles with the solvent composition confirms the signifi-
quenching with an increase in the mole fraction of methanol cant solvent effect on the excited-state relaxation process in both
presumably due to the formation of solvent hydrogen bonding, compounds. The details of the fluorescence decay fitting and
the introduction of water in an acetonitrile solution resulted in the corresponding average lifetimes (tav) are listed in the ESI†
an increase in intensity. Therefore, it is obvious that the specific (Table S3). Interestingly, analogous to the fluorescence intensity
interaction through hydrogen bond formation is not the only variation pattern, the tav values of MHC also increased until the
factor controlling the spectral intensity. Further, the width of mole fraction of water reached ca. 0.85 and then it decreased
the fluorescence spectrum underwent a significant increase in monotonically. A similar pattern was also observed for the
water compared with that in acetonitrile solution. On the other fluorescence lifetime of MHCB obtained by monitoring the emission
hand, the addition of water in the DMSO solution of MHC led to at 485 nm. The rather unusual behaviour in the DMSO/water
an increase in fluorescence intensity at a low water concen- mixture with a high water content (with xDMSO E 0.15) is quite
tration until it reached a maximum and then further decreased well known40,41 and consistent with the observation of an
at a higher water proportion (inset Fig. 4a). A gradual red shift in abrupt increase in solvent hydrophobicity under this condition
the fluorescence maximum was also noted during this process. estimated from molecular dynamics (MD) simulation.42 The
However, this situation was far more complex and interesting present results indicate that the fluorescing states representing
for MHCB (Fig. 4b). The fluorescence emission maximum of the relaxation of MHC and MHCB at 485 nm are relatively
MHCB in DMSO appeared at 485 nm. On increasing the mole nonpolar in nature. However, the MHCB fluorescence at 385 nm
fraction of water, the intensity of this band decreased with a originates from a species that is significantly different in nature
concomitant increase in the intensity of a new peak at 385 nm and mostly controlled by a specific solvent interaction. Overall,
(inset Fig. 4b). The presence of a clear iso-emissive point at both the steady-state and time-resolved fluorescence studies in
445 nm indicates the equilibrium between the two differently mixed solvents further reaffirm the multi-faceted mode of
solvent action on the relaxation spectroscopic behaviour of the
investigated coumarin compounds.
Quantitative estimation of the contribution from solvent
parameters
The spectroscopic behaviour of the coumarin compounds depends
strongly on the nature of the solvent. To describe this solvato-
chromism through quantitative estimation of different solvent
parameters, all the quantities listed in Table 1 were subjected to
multiple regression analysis based on the Kamlet–Taft (KT) and
´
Catalan models. The analysis results are displayed in the ESI†
(Tables S4 and S5, respectively, ESI†), and Fig. 5 shows the
percentage contribution of each of these parameters towards
different spectroscopic properties in a variety of solvents.
It should be noted that for both systems, specific solvent
interactions characterized by solvent hydrogen bond donation
acidity (a) and acceptance basicity (b) contribute strongly in
addition to the solvent dipolarity/polarizability contribution
(p* parameter). For both MHC and MHCB, the qualitative
contribution of the different solvent parameters towards the
photophysical behaviour of the absorbing species (characterized
by nmabasx) is almost equivalent. The only significant difference is
the relatively higher contribution (ca. 50%) of the p* parameter
in MHC compared with the mere 15% in MHCB. However, in
addition to the significant contribution of b (30–40%) in emission
Fig. 4 Fluorescence emission spectra of MHC (a) and MHCB (b) in DMSO/
water mixture with changing mole fraction (x) of water as 0 (1), 0.35 (2),
0.57 (3), 0.80 (4), 0.92 (5) and 1.00 (6). Inset shows the intensity variation of
emission maximum in the respective case. (c) Time-resolved fluorescence
decay traces obtained by monitoring MHCB fluorescence at 385 nm with a
variation in water content in DMSO/water mixture. (d) Variation in average
fluorescence decay time (tav) for MHC and MHCB with an increase in the
mole fraction of water.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18717

View Article Online
Paper
NJC
´
properties give relatively better correlation with the Catalan formal-
ism than the KT analysis, the qualitative results of the different
solvatochromic behaviours are almost similar in both cases.
Fluorescence modulation in cyclodextrin nanocavities
Cyclodextrins (CDx) are extensively used as cyclic components
for the construction of supramolecular architectures because of
their well-defined ring structure and affinity for different classes
of compounds of varying size and shape. The unique behaviour of
CDx to encapsulate organic compounds inside their hydrophobic
central cavity make them potential candidates as extremely efficient
molecular vehicles for drug delivery.43,44 Furthermore, binding of a
fluorescent guest in the interior of the cavity of CDx renders an
opportunity to study different photophysical properties in tailored
environmental conditions. The results discussed in the previous
sections indicate that the spectral features of both MHC and MHCB
depend strongly on various solvent parameters, including hydrogen
bonding ability. Therefore, it will be interesting to monitor the
fluorescence of these compounds in the nanocavities of CDx.
The effect of b-CDx (consisting of a macrocyclic ring of seven
glucose subunits joined by a-1,4 glycosidic bonds) on the
spectral properties of the investigated systems was studied by
keeping the concentration of the probe fixed and changing the
concentration of the added b-CDx. The absorption spectra of
none of these compounds showed notable changes upon the
addition of cyclodextrin, except a minor increase in absorption
intensity presumably due to the improved dissolution of the
probes in the presence of b-CDx. However, the fluorescence
spectra underwent drastic changes and interestingly, the spectral
response of the investigated coumarin derivatives towards an
increase in b-CDx concentration was distinctly different (Fig. 6).
For example, the fluorescence intensity of MHCB increased with
a significant red-shift in peak position from 396 nm to 420 nm in
the presence of b-CDx. Typically, caging of a series of guest
molecules in the b-CDx cavity led to an increase in fluorescence
intensity due to the restricted motion of the encapsulated probe
with reduced non-radiative decay pathways. Therefore, the effect
of cyclodextrin on the intensity and position of the fluorescence
spectrum in the case of MHCB seems compatible with the inclu-
sion of the probe into the cavity. The binding constant (K) was
estimated to be 132.40 ꢀ 2.45 Mꢁ1 from the slope and intercept
of linear variation in 1/(I ꢁ I0) against 1/[b-CDx] (inset, Fig. 6b)
Fig. 5 Percentage contribution of solvent parameters obtained from the
Kamlet–Taft (top panel) and Catalan (bottom panel) analysis for the
´
different spectroscopic properties of MHC (a and c) and MHCB (b and d).
The properties listed here are absorption and emission peak (nabs and nem
,
respectively), Stokes shift (Dnss), quantum yield (ff), average fluorescence
lifetime (tav), radiative and total non-radiative rate constant (kr and Sknr).
max
em
spectral maximum (n ), the major contribution originates from
the p* parameter (50–55%), even though it acts in exactly the
opposite way for MHC and MHCB. For example, KT analysis
predicts that the fluorescence maximum will shift towards a
max
em
lower wavelength (increase in the value of n ) in highly polar
solvents such as water for MHCB, whereas, it is exactly opposite
in the case of MHC. These predictions correlate well with the
fluorescence data of the investigated systems given in Fig. 2 and
Table 2. Despite the similar structural motifs of MHC and
MHCB, the striking difference in the solvent response towards
their photophysical behaviour is intriguing. For example, while
the variation in fluorescence yield (ff) constitutes ca. 20%
contribution from the p* parameter in the case of MHC, it is
almost insignificant for MHCB. Further, the contribution of the
a and b parameters towards ff in these systems is contradictory.
As discussed before, the non-radiative rate constant (Sknr) plays
a dominant role in the excited-state deactivation process for
both coumarins. Interestingly, all three solvent parameters
contribute in the same direction towards the change in Sknr for
both systems, even though in a significantly different amount. It
should be noted here that while the contribution from the solvent
a and b parameters changes Sknr in an analogous way for MHC
and MHCB, the contribution of the p* term is almost two and
half times more in the former.
Thus, the importance of the p* term, particularly in the
regression analysis of the spectral behaviour of these coumarin
derivatives needs further exploration. Unfortunately, quantitative
separation of the solvent dipolarity and polarizability effect
cannot be achieved in the KT formalism. However, the recent
´
Fig. 6 Variation in fluorescence intensity (lexc = 305 nm) of MHC (a) and
MHCB (b) with an increase in b-CDx concentration (along the arrow
direction). Inset shows the corresponding SV and BH plots. (c) Time-
resolved decay traces of MHC and MHCB (open and solid circles, respec-
tively) together with the simulated data (solid line) with 2-exp decay
function in the presence of 7.2 and 4.8 mM b-CDx, respectively. Inset:
Variation in tav of MHCB with an increase in [b-CDx].
development of the Catalan four-parameter solvent scale separates
these two quantities and gives more detailed information on the
non-specific behaviour of the solvatochromic interaction. The results
of the regression analysis of all the properties of MHC and MHCB
using the Catalan solvent scale are also included in Fig. 5 (lower
panel). It should be noted here that although the individual spectral
´
18718 | New J. Chem., 2019, 43, 18713--18725
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
for 1 : 1 complex formation between b-CDx and MHCB using the Sequestration of the cholinergic drugs in human serum
Benesi–Hildebrand (BH) relation45 given by eqn (1).
albumin matrix
The excited state photophysical behaviour of environment-
sensitive fluorescence probes is known to be markedly affected
in protein nanocavities. Further, the investigated systems were
identified as potential anti-cholinergic candidates for the treat-
ment of AD. However, the AChE inhibitory activity of these
compounds including the Food and Drug Administration (FDA)
approved AD drugs shows a significant diminution in the
presence of serum albumin matrix.25,39 Therefore, it is imperative
to understand the nature of the interaction of these compounds
with human serum albumin (HSA). In this section, we describe
the fluorescence modulation and the nature of the binding
interaction of the coumarin probes with HSA using a variety of
spectroscopic techniques and molecular docking calculation.
1
1
1
1
¼
þ
ꢂ
(1)
I ꢁ I0
I
1 ꢁ I0 KðI1 ꢁ I0Þ ½CDꢃ
where I0 and IN are the fluorescence intensities of the free and
fully complexed MHCB, respectively.
On the other hand, the fluorescence intensity of MHC
decreased continuously in the b-CDx environment without
any significant shift in peak position (Fig. 6a). It should be
noted here that normally a fluorescent molecule experiences a
relatively non-polar environment in an encapsulated complex
in comparison with the bulk solution. It was already confirmed
from the KT analysis (described in the previous section) that
the fluorescence yield (ff) of MHC decreases with a reduction
in polarity (given by p* parameter), which is consistent with the
present observation in the b-CDx environment. As confirmed
from the time-resolved measurements (see below), the decrease
in MHC fluorescence intensity in the presence of cyclodextrin is
due to the ground state complex formation (static quenching).
The association constant (Ka) under this situation can be
approximated as the Stern–Volmer (SV) quenching constant
(KS) for static cases,46 which is related with the fluorescence
intensity and lifetime in the absence (given by I0 and t0,
respectively) and presence (I and t, respectively) of a certain
quencher concentration [Q] as:
Quenching of intrinsic HSA fluorescence in presence of MHC
and MHCB
The absorption spectral maximum of HSA appears at 280 nm.
Excitation at this wavelength resulted in a broad intrinsic
protein fluorescence within the 290–450 nm range originating
from aromatic residues such as tyrosine (tyr) and tryptophan
(trp) moieties, with almost equal fluorescence yield.46 The
addition of both MHC and MHCB to the HSA solution resulted
in a steady decrease in the protein fluorescence intensity
(Fig. 7) due to the interaction of these coumarin probes with
the aromatic residues in the protein binding domain (PBD).
Equivalent results were also observed at other temperatures for
both systems (Fig. S4 in the ESI†). Interestingly, the protein
fluorescence spectral profile became broad with two distinctly
visible emission maxima at ca. 305 and 335 nm with the addition
of both MHC and MHCB. Although the spectral position of the
first band remained practically unchanged and can be considered
I0
I
t0
t
¼ 1 þ KS½Qꢃ;
¼ 1
(2)
The calculated value from the slope of the linear SV plot (inset,
Fig. 6a) is 53.51 ꢀ 1.63 Mꢁ1, which indicates that the extent of
interaction in this case is about two and half times less in
magnitude compared with that of MHCB.
The time-resolved measurement in the b-CDx environment
for both compounds showed a significant deviation from the
single-exponential decay functions observed in aqueous environ-
ment. In addition to a similar time constant corresponding to the
free (uncomplexed) fluorophore in the bulk aqueous phase, one
extra component with a relatively longer decay time, presumably
arising due to the formation of a host–guest complex, contributes
about 30% of the total decay of MHC (Table S6, ESI†). Interest-
ingly, the corresponding contribution from the encapsulated
component is as high as 70% in the case of MHCB. This indicates
that the propensity of complex formation with b-CDx in MHCB is
relatively higher than that in MHC, which is further justified from
the magnitude of the binding constants in the respective cases
estimated from the steady-state experiments. Some of the
representative decay traces are shown in Fig. 6c. It should be
noted that while the magnitude of tav showed steady increase
from 6.36 to 7.03 ns for MHCB with an increase in the b-CDx
concentration from 1.4 to 4.8 mM (until saturation in steady
state intensity was achieved), it remained practically insensitive
at tav = 5.66 ꢀ 0.06 ns in the case of MHC even at [b-CDx] = 33.6 mM
(inset, Fig. 6c). The constant value of tav further reaffirms that the
quenching of the MHC fluorescence with an increase in b-CDx
concentration occurs through ground state complex formation.
Fig. 7 Quenching of intrinsic HSA (1.2 mM) fluorescence in the presence of
increasing MHC (a) and MHCB (b) concentration along the arrow direction.
Inset shows the linear variation of the Stern–Volmer (SV) plot in each case.
Temperature variation of binding constant (Kb) and SV constant (KSV) values are
shown for MHCB obtained from double-log (c) and SV (d) plots, respectively.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18719

View Article Online
Paper
NJC
Table 3 Stern–Volmer quenching constant (KSV), binding constant (Kb) and number of binding sites (n) calculated from the modulation of intrinsic
protein (HSA) fluorescence data in the presence of varying MHC and MHCB concentration at different temperaturesa
MHC
MHCB
Temp./K Ksv/105
log kb
n
DG
DH, DS
Ksv/105
log kb
n
DG
DH, DS
298
303
308
313
318
7.42 ꢀ 0.31 7.26 ꢀ 0.22 1.29 ꢀ 0.04 ꢁ41.37 DH = 80.56, 2.66 ꢀ 0.06 5.04 ꢀ 0.07 0.92 ꢀ 0.01 ꢁ28.30 DH = ꢁ10.94,
7.08 ꢀ 0.31 7.04 ꢀ 0.17 1.25 ꢀ 0.03 ꢁ40.71 DS = ꢁ0.13 2.49 ꢀ 0.07 4.94 ꢀ 0.12 0.90 ꢀ 0.02 ꢁ28.60 DS = 0.06
6.93 ꢀ 0.30 6.83 ꢀ 0.10 1.21 ꢀ 0.02 ꢁ40.06
6.50 ꢀ 0.26 6.39 ꢀ 0.13 1.13 ꢀ 0.02 ꢁ39.40
6.60 ꢀ 0.34 6.48 ꢀ 0.16 1.14 ꢀ 0.03 ꢁ38.74
2.35 ꢀ 0.06 4.89 ꢀ 0.10 0.90 ꢀ 0.01 ꢁ28.89
2.24 ꢀ 0.05 4.95 ꢀ 0.09 0.92 ꢀ 0.02 ꢁ29.18
2.00 ꢀ 0.07 4.85 ꢀ 0.13 0.90 ꢀ 0.03 ꢁ29.47
a
The mean values (ꢀstandard deviation) were estimated from three independent set of measurements; KSV (Mꢁ1) values are obtained from straight
line fitting of protein Trp fluorescence quenching data using eqn (3), whereas log Kb and n values were obtained from double log plot (eqn (4)); the
values of DH (kJ molꢁ1) and DS (kJ molꢁ1 ꢁ1) obtained from the van’t Hoff Plot (eqn (5)) were used to calculate DG (kJ molꢁ1); and correlation
K
coefficient (R2) of straight line fitting is Z0.98 in each case.
to originate from the tyrosinate moiety, the emission position at and n are shown in Fig. 7 and the ESI† (Fig. S5). The corres-
335 nm (due to the tryptophan residue) showed a certain shift. The ponding values estimated from the intercept and slope of the
results indicate that with addition of the coumarin derivatives, double-log analysis for both systems are presented in Table 3.
there is a notable change in the protein secondary structure (see The average number of binding sites for the coumarin derivatives
later) which results in varying degrees of exposure of the trp is close to one, corresponding to the single binding site of HSA at
residue to the bulk aqueous phase. The Stern–Volmer (SV) relation subdomain IIA (Sudlow site I). However, the magnitude of n is
for the protein fluorescence quenching data can be written in the higher for MHC in comparison with MHCB, suggesting the
form of eqn (3) at different temperatures.
relatively stronger interaction of the former with HSA. This
observation is consistent with relatively higher values of KSV
and Kb in MHC obtained from the fluorescence titration data
and the complimentary results obtained from CD and molecular
docking calculation (see below).
I0
¼ 1 þ KSV½Qꢃ ¼ 1 þ kqt0½Qꢃ
(3)
I
where I0 and I are the fluorescence intensities at 335 nm before
and after the addition of the quencher (MHC or MHCB),
respectively, kq is the bimolecular quenching constant, t0 is
the fluorescence lifetime of HSA in the absence of any quencher,
and [Q] is the concentration of the quencher. The values of SV
constant (KSV) calculated from the slope of the linear SV plots
It is interesting to note that the binding constant values for
the investigated coumarin derivatives with HSA are moderate
and vary within the order of 107–105
M
ꢁ1. Since HSA plays a
significant role as a carrier protein, drug binding with HSA is
an active screening test to estimate therapeutic possibilities.49
Similar studies are common even for other biological macro-
molecules such as DNA.50 In general, the role of HSA is to
shield bound drugs from oxidation and release them at specific
targets to produce therapeutic effects. Therefore, compounds
with higher affinity for HSA allow easier drug transport in vivo.
However, as discussed before, the strong sequestration of
cholinergic compounds with HSA reduces the active fraction
of the drug to inhibit AChE activity, and therefore, is no longer
considered to be potent for the treatment of AD. A subtle
balance between these two requirements with moderate affinity
to HSA, equivalent to that estimated for the investigated
coumarin derivatives in the present study, seems to be suitable
to optimize the efficiency of anti-cholinergic drugs in vivo.
Considering the enthalpy (DH0) and entropy (DS0) change to
be independent of temperature, all the thermodynamic para-
meters and spontaneity (given by the free energy change, DG) of
the ligand binding process were obtained from the following
set of equations by performing the fluorescence titration
experiment at different temperatures.
and listed in Table 3 are 7.42 ꢀ 0.31 and 2.66 ꢀ 0.06 (105, Mꢁ1
)
for MHC and MHCB, respectively, at 298 K. The magnitude of
kq estimated from the known value of t0 = 4.2 ꢀ 0.1 ns (which
was measured independently by exciting the protein solution at
295 nm47) is 1.77 and 0.63 (1014, Mꢁ1 ꢁ1), respectively. This
s
quantity is almost 3–4 orders of magnitude higher than the
maximum possible value of kq (E1 ꢂ 1010
M
ꢁ1 sꢁ1) for a purely
diffusion-controlled (dynamic) quenching process, which suggests
the formation of a ground state complex between HSA and the
coumarin derivatives (static mechanism).
Characterizing the association of MHC and MHCB in the
protein binding domain
The binding constant (Kb) of the investigated coumarins (ligand)
with the protein and the number of binding sites (n) of the protein
were determined from the fluorescence titration data using eqn (4)
at different temperatures by considering (a) an equilibrium
between the unbound and ligand-bound protein molecules and
(b) the presence of similar but independent binding sites in the
protein surface.48
DH0
2:303R
1
DS0
2:303R
log Kb ¼ ꢁ
ꢂ
þ
;
DG0 ¼ DH0 ꢁ TDS0 (5)
ꢀ
ꢁ
I0 ꢁ I
T
log
¼ log Kb þ n log½ligandꢃ
(4)
I
Interestingly, the results given in Fig. 7c and d and in
Some of the representative plots where the data was fitted Table 3 indicate that the magnitude of KSV and Kb decreases
with the above equation to calculate the magnitudes of log Kb with an increase in temperature for both compounds, further
18720 | New J. Chem., 2019, 43, 18713--18725 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
confirming the hypothesis of ground state complex formation
(static mechanism) for fluorescence quenching discussed earlier.
Nonetheless, the negative free energy change (DG0) for both cases
indicates the spontaneous binding process of these compounds
to HSA. The type of non-covalent interactions responsible for the
binding of organic molecules on the protein surface is often
characterized by the sign and magnitude of thermodynamic
entities such as DS0 and DH0.51 A positive DS value is mostly
indicative of a hydrophobic mechanism in drug–protein inter-
Fig. 8 Circular dichroism (CD) spectra of HSA in the presence of varying
action. On the other hand, negative entropy and enthalpy concentration of MHC (a) and MHCB (b). The protein–ligand concentration
ratio increases along the arrow direction as 1 : 0 (i), 1 : 5 (ii) and 1 : 10 (iii).
Inset shows the change in a-helix content of the protein calculated from
the CD data in each case.
changes indicate the importance of van der Waals force and
hydrogen bond formation. From the results displayed in Table 3,
can be inferred that the hydrophobic interaction plays the
leading role in MHCB binding with has, whereas, van der Waals
force and/or hydrogen bond formation becomes relevant in the
case of MHC. The difference in predominant forces controlling
the binding for both coumarin compounds with a similar
structural motif is presumably the primary factor for the variation
in their binding thermodynamics with has, as discussed above.
Characterizing the major forces for binding: molecular docking
calculation
The blind docking results indicate that MHC sits in sub-domain
IIA of drug site 1, whereas, MHCB binds to sub-domain IB of
HSA. For targeted docking in MHC, the grid box was centred at
22.147 (x-centre), 32.155 (y-centre) and 36.265 (z-centre) with the
corresponding size of 26, 28 and 28, respectively. On the other
hand, a grid box value of 43.028, 31.207 and 17.906 for the x-, y-,
and z-centres with a corresponding size of 28, 42 and 34,
respectively, was set for MHCB in domain IB of HSA. From
the best posed docked structure given in Fig. 9 and the corres-
ponding energy parameters (Table 4), it was observed that MHC
binds more favourably to the hydrophobic cavity in subdomain
IIA with DG = ꢁ43.93 kJ molꢁ1 (corresponding to log K = 7.69) in
comparison with MHCB (DG = ꢁ41.39 kJ molꢁ1, log K = 7.25) in
domain IB of HSA. The trend and values of the estimated
Identification of protein secondary structure: circular
dichroism measurement
Circular dichroism (CD) is a powerful tool to investigate the
change in the secondary structure of the proteins when they
interact with small molecules. The CD spectra of HSA exhibited
two negative bands in the UV region at 208 and 222 nm
(contributed by p–p* and n–p* transfer of a-helical structure
of protein, respectively). The results of CD were converted into
mean residual ellipticity (MRE) using the following relation:
Observed CD ðmdegreeÞ
MRE ¼
(6)
Cpnl
where Cp is the molar concentration of the protein, n is the
number of amino acid residues and l is the path length. The
a-helical contents of free and complexed protein were calculated
from the mean MRE values at 208 nm (MRE208) using eqn (7).52
½MRE208 ꢁ 4000ꢃ
a-Helix ð%Þ ¼
ꢂ 100
(7)
½33 000 ꢁ 4000ꢃ
where 4000 is the MRE of the form and random coil conformation
cross and 33 000 is the MRE value of a pure helix at 208 nm. As
shown in the CD spectra of HSA in the absence and presence of
the coumarin compounds (Fig. 8), the intensity of the negative
band increases without any shift in the band maximum and the
calculated a-helix content of HSA increases when it is bound
to MHC or MHCB. The increase in ellipticity suggests that
on complexation, both coumarin compounds stabilize the
secondary structure of HSA.53,54 This extra stabilization may
arise because the hydroxyl group present in MHC and MHCB
can bind with the amino acid residues of HSA to stabilize the
protein helix structure.55 The relatively higher stabilization in
the case of MHC (inset, Fig. 8) in comparison with that of
MHCB is consistent with the stronger interaction of the former
with HSA, as revealed in the fluorescence titration experiment
discussed before.
Fig. 9 Best docked poses of MHC and MHCB with HSA. The Ligplot
images on the right-hand side depict the hydrogen bonding and hydro-
phobic interactions of the various residues of HSA with the compounds.
Dotted lines represent hydrogen bonds, whereas, circular spikes show
hydrophobic interactions.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18721

View Article Online
Paper
NJC
Table 4 Molecular docking calculation results for the binding interaction of MHC (A) and MHCB (B) with HSA
Hydrogen bonding interaction
Binding affinity/kJ molꢁ1
nHB
Residues involved
RHB/Å
Hydrophobic interaction
Val216, Lys212, Ala213
A
B
ꢁ43.93
6
O2ꢄ ꢄ ꢄN(Leu331)
O5ꢄ ꢄ ꢄOD2(Asp324)
O5ꢄ ꢄ ꢄO(Arg209)
4.82
4.85
4.44
2.80
4.19
4.56
O6ꢄ ꢄ ꢄNH1(Arg209)
O6ꢄ ꢄ ꢄNH2(Arg209)
O6ꢄ ꢄ ꢄNZ(Lys351)
ꢁ41.39
4
O5ꢄ ꢄ ꢄO(Leu115)
O6ꢄ ꢄ ꢄNE(Arg186)
O6ꢄ ꢄ ꢄOD2(Asp183)
O6ꢄ ꢄ ꢄO(Leu182)
3.10
3.91
3.85
3.95
Arg114, Arg117, Val116, Ile142, His146,
Arg145, Leu185, Tyr161, Tyr138, Met123
binding parameters match well with that obtained from the although about 20% of the solvatochromic response of the fluores-
fluorescence experiments (DG = ꢁ41.37 kJ molꢁ1 for MHC and cence yield was contributed through the solvent polarity and/or
DG = ꢁ28.30 kJ molꢁ1 for MHCB). It should be noted here that polarizability (p*) term in MHC, it was almost insignificant for
molecular docking calculation gives a rough approximation of MHCB. Despite their similar structural architecture, the solvent
the binding sites. Further, since the larger protein structure is HB donation acidity (a) and acceptance basicity (b) act differently
static and motion is allowed only to the smaller-sized drug for most of the spectral properties of the investigated systems. The
molecules, extremely accurate determination of the energy modulation in fluorescence response in the hydrophobic cavity of
parameters is not expected. Despite these inherent limitations, cyclodextrin or at the protein binding domain in HSA is consistent
the calculated negative values of the affinities indicate spontaneous with the solvatochromic analysis. Titration of the intrinsic
interaction in both coumarin derivatives investigated in this study tryptophan fluorescence in conjunction with molecular docking
and augurs well with the values obtained from the fluorescence calculation reveals the strong sequestration tendency of HSA for
experiments.
both compounds and the results corroborate nicely with the
The amino acid residues Val216, Lys212, and Ala213 of site I relative decrease in AChE inhibition efficiency in the respective
are involved in the hydrophobic interaction to stabilize the cases. These findings can help in understanding the binding
complex formation for MHC. On the other hand, the ligand and transport properties of coumarinyl compounds in human
MHCB interacts hydrophobically with Arg114, Arg117, Val116, plasma at the molecular level, and therefore, can lead to the
Ile142, His146, Arg145, Leu185, Tyr161, Tyr138, and Met123. development of better AChE inhibitors.
Moreover, MHC forms a total of six hydrogen bonds (HB) with
amino acid residues of HSA, of which one HB with Leu331
Materials and methods
having a bond length (rHB) of 4.82 Å, one with Asp324 (rHB
=
Synthesis and characterization MHC and MHCB
4.85 Å), three with Arg209 (rHB = 4.44, 2.80 and 4.19 Å) and one
with Lys351 (rHB = 4.56 Å). Similarly, MHCB also forms four
hydrogen bonds with other amino acid residues of HSA, namely
Leu115 Arg186, Asp183 and Leu182 with rHB = 3.10, 3.91, 3.85
and 3.95 Å, respectively. In both ligand–HSA complexes, it has
been well observed that the hydroxyl oxygen of the ligand formed
more hydrogen bonds with amino acid residues compared with the
ketonic oxygen of the ligands to stabilize the complex formation.
All commercially available chemicals and reagents used for the
synthesis of the desired coumarin derivatives were purchased
from Sigma Aldrich and Merck and were used without further
purification. The synthesis of the compounds involved the
earlier reported procedure56 with slight modification and is
given in Scheme 1. In brief, to a stirred solution of aldehyde
1 (2.0 mmol) and 4-hydroxy coumarin 2 (4.0 mmol) in 5 mL
Millipore water, Ni nanoparticles (NPs) (25 mg mmolꢁ1) were
added at room temperature. The reaction mixture was stirred at
the same temperature for 3 h. After completion of the reaction
Conclusions
The anticholinergic activity of the investigated coumarin com- (monitored by thin layer chromatography, TLC), the reaction
pounds, MHC and MHCB, is comparable with some of the FDA mixture was filtered through a Buchner funnel, and washed
approved drugs for the treatment of AD. However, the in vitro with water. The residue was recrystallized from 5 mL ethanol to
assessment of their AChE inhibition activity in the presence afford pure compounds 3a and 3b. The purity of the products
of the HSA matrix was found to decrease substantially as evidenced was confirmed by infrared (IR), nuclear magnetic resonance
by the 12–14% increase in IC50 value for both systems, primarily (NMR) and mass spectroscopy (data given below). Melting
due to the sequestration of the drugs in the HSA cavity through a points were determined in open capillary tubes and are uncorrected.
combination of hydrophobic and hydrogen bonding (HB) inter- IR spectra were recorded in potassium bromide (KBr) pellets on a
actions. Quantitative estimation of the HB effect was modeled PerkinElmer Spectrum 400 FTIR instrument, and the frequen-
with multiparametric regression of several spectroscopic responses cies are expressed in cmꢁ1. 1H-NMR and 13C-NMR spectra were
´
based on the Kamlet–Taft and Catalan models. It was found that recorded on a Bruker Avance II-400 spectrometer in DMSO-d6/CDCl3
18722 | New J. Chem., 2019, 43, 18713--18725 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
(chemical shifts in d with TMS as an internal standard). Mass spectral range with that of a standard (quinine bisulfate in 0.5 M
spectral data was obtained with a WATERS (ZQ-4000) mass spectro- H2SO4 solution, fsf = 0.54658) following the protocol described
meter. All reactions were monitored by thin layer chromatography earlier.46 The relative experimental error in the measurement of
(TLC) using pre-coated aluminium sheets (silica gel 60 F254, 0.2 mm ff was estimated to be ꢀ10%.59
thickness).
Time-resolved fluorescence decay measurements were per-
3,30-Methylenebis(4-hydroxy-2H-chromen-2-one) (3a, MHC). formed in a fluorescence spectrometer (QM-40, PTI, USA) equipped
White powder, mp: 218–220 1C; IR nmax (KBr): 3380, 3045, 1640, with a TCSPC fluorescence lifetime detection unit (PM-3). The
1615, 1579, 1293, 1084, 756 cmꢁ1; H NMR (CDCl3, 400 MHz): sample was excited by a 295 nm nano-LED supplied by PTI. The
1
dH (ppm) 11.52 (s, 1H, OH), 11.24 (s, 1H, OH), 7.98–7.80 (m, 2H, experimentally obtained decay traces were expressed as a sum of
Ar-H), 7.59–7.55 (m, 2H, Ar-H), 7.35–7.33 (m, 2H, Ar-H), 6.95– exponentials (eqn (8)) and fitted with the iterative deconvolution
6.84 (m, 2H, Ar-H), 2.52 (s, 2H, CH2); 13C NMR (100 MHz, CDCl3 + method based on the Levenberg–Marquardt algorithm60 with
DMSO-d6): dC (ppm) 164.49, 160.55, 151.88, 130.33, 128.12, 128.01, reference to the instrument response function (IRF), collected at
127.34, 126.20, 124.01, 123.76, 115.89, 113.45, 23.06; MS (ES+) the excitation wavelength using a scattering solution. The
calcd for C19H12O6: 336.06, found m/z 337.09 (M + H)+.
success of the fitting model was checked by noting the reduced
3,30-(Phenylmethylene)bis(4-hydroxy-2H-chromen-2-one) (3b, chi-square (w2) and Durbin–Watson (DW) parameter along with
MHCB). White crystal, mp: 228–230 1C; IR nmax (KBr): 3399, visual inspection of the residual distribution in the whole
3070, 1659, 1618, 1569, 1282, 1093, 757 cmꢁ1; 1H NMR (CDCl3, fitting range.
ꢂ
ꢃ
400 MHz): dH (ppm) 11.48 (s, 1H, OH), 11.26 (s, 1H, OH), 7.12–8.02
(m, 13H, Ar-H), 6.04 (s, 1H, 4H); 13C NMR (100 MHz, CDCl3): dC
(ppm) 165.83, 164.62, 152.53, 152.29, 135.19, 132.88, 128.65, 126.89,
126.48, 124.92, 124.40, 116.83, 116.65, 105.63, 103.89, 36.17; MS
(ES+) calcd for C25H16O6: 412.09, found m/z 413.12 (M + H)+.
X
ꢁt
FðtÞ ¼
ai exp
(8)
ti
i
where ai is the amplitude of the ith component associated with
fluorescence lifetime ti such that Sai = 1. For non-exponential
fitted functions, the average lifetime (tav) was expressed in terms
of fractional contribution ( fi) of each decay time to the steady
Solvents and reagents for spectral measurements
state intensity with the following equation.46
A series of ten organic solvents, all highest grade available with
varying solvent parameters (listed in Table S1, ESI†) was
procured from Sigma-Aldrich and used fresh to make 3–5 mM
concentration of the probe necessary for spectroscopic measure-
ments. Purified water, collected from an Elix 10 water purification
system (Millipore India Pvt. Ltd) was used to make the aqueous
solutions. Analytically pure b-cyclodextrin (b-CDx) and human
serum albumin (fraction V, lyophilized powder form of HSA) were
also used as received from Sigma-Aldrich. Analytically best grade
available reagents were used for measuring the enzyme activity of
AChE, the details of which are mentioned elsewhere.25
P
aiti2
X
i
P
tav
¼
fiti ¼
(9)
aiti
i
i
Further, estimation of the radiative (kr) and total non-radiative
(Sknr) decay rate constants in each case was done with the
known values of ff and tav using the following set of relations,
and the values are given in the table.
ff
tav
ð1 ꢁ ff Þ
kr ¼
;
Sknr
¼
(10)
tav
Instruments and data analysis
Variation in solvatochromic data was checked with the three-
Estimation of the acetylcholinesterase (AChE) inhibition activity parameter-based Kamlet–Taft (KT) or four-parameter-based
´
of the investigated coumarin compounds was done in a Synergy Catalan model. The detailed protocol for multiple regressions
H1 hybrid meter plate reader instrument (BioTek) using 96-well with the KT relation is given elsewhere.32–34 Additionally,
plates following the protocol reported earlier.25,37 A brief descrip- we also analyzed the spectroscopic properties with the four-
31,35
´
tion of the adopted procedure and data handling techniques is also parameter Catalan scale
with the intention to isolate the
given in the ESI† (ST1). Cuvette-based UV-vis absorption and steady- independent contribution from solvent polarizability (SP) and
state fluorescence emission/excitation spectra were collected in solvent dipolarity (SdP).
PerkinElmer Lambda25 and Hitachi F4500 spectrophotometers,
Molecular docking calculation
respectively. All experiments were carried out at room temperature
(298 ꢀ 2 K) with 10 mm quartz cuvettes. For the fluorescence Molecular docking calculation was performed to disclose the
experiments, a 5 nm bandpass was used in both the emission/ binding site and identify several interactions involved between
excitation side and all the spectra were corrected for photomultiplier the amino acid residues of HSA and the guest to stabilize the
response within the measured wavelength range. Each of the HSA–ligand complex. For the molecular docking calculation,
fluorescence spectra of both probes in heterogeneous media, taken the 3D structure of HSA (PDB ID: 1AO6) was retrieved from the
as the average of three independent measurements, was further Protein Data Bank (www.rcsb.org). The structure was prepared for
corrected for possible signal attenuation due to the presence of docking by the removal of all heteroatoms, including ions and
substances other than the fluorophore using the method described water molecules, addition of polar hydrogen and assignment of
elsewhere.57 Fluorescence quantum yields (ff) were calculated Kollman charges. The coumarin compounds (MHC and MHCB)
by comparing the total intensity under the whole fluorescence were modelled using ChemDraw and minimized using the MMF4
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18723

View Article Online
Paper
NJC
force field.61 The docking studies were executed with the help of
AutoDock Vina and AutoDock tools-1.5.6 (ADT)62 and fashioned
using Chimera63 using the protocol of Lamarckian Genetic
Algorithm (LGA).64 The number of docking runs was set to 50.
The conformational cluster analysis was done using the maximum
RMS tolerance of 2.00 and the lowest energy conformer was chosen
for sketching the docked poses.
17 J. Wang, X. Qian and J. Cui, J. Org. Chem., 2006, 71, 4308–4311.
18 M. T. Lee, C. K. Yen, W. P. Yang, H. H. Chen, C. H. Liao,
C. H. Tsai and C. H. Chen, Org. Lett., 2004, 6, 1241–1244.
19 J. M. Hierrezuelo and C. C. Ruiz, J. Phys. Chem. A, 2012, 116,
12476–12485.
20 S. Nad and H. Pal, J. Phys. Chem. A, 2003, 107, 501–507.
21 M. Homocianu, A. Airinei and D. O. Dorohoi, J. Adv. Res.
Phys., 2011, 2, 1–9.
22 Y. Yue, S. Zhao, Y. Sun, X. Yan, J. Liu and J. Zhang, J. Lumin.,
2017, 187, 333–339.
Conflicts of interest
23 A. Garg, D. M. Manidhar, M. Gokara, C. Malleda, C. S. Reddy
and R. Subramanyam, PLoS One, 2013, 8, e63805.
24 Y. Yue, Z. Wang, Z. Wang, Y. Zhang and J. Liu, J. Mol. Struct.,
2018, 1169, 75–80.
There are no conflicts to declare.
Acknowledgements
25 P. Baruah, G. Basumatary, S. O. Yesylevskyy, K. Aguan, G. Bez
and S. Mitra, J. Biomol. Struct. Dyn., 2019, 37, 1750–1765.
26 I. Matei, S. Ionescu and M. Hillebrand, Solute-solvent
hydrogen bond formation in the excited state. Experimental
and theoretical evidence, in Hydrogen bonding and transfer
in the excited state, ed. K.-L. Han and G.-J. Zhao, John Wiley
& Sons Ltd, Chichester, UK, 2010, ch. 4.
The authors acknowledge the support from the Dept. of Science
& Technology (DST), Govt. of India to the Chemistry Department
through FIST program (SR/FST/CSI-194/2008).
Notes and references
27 F. D. Lewis and W. Weigel, J. Phys. Chem. A, 2000, 104,
8146–8153.
1 T. S. Reddy and A. R. Reddy, Dyes Pigm., 2013, 96, 525–534.
2 J. R. Hoult and M. Paya, Gen. Pharmacol., 1996, 27, 713–722.
3 E. C. Gaudino, S. Tagliapietra, K. Martina, G. Palmisano and
G. Cravotto, RSC Adv., 2016, 6, 46394–46405.
4 S. Widmer, M. Dorrestijn, A. Camerlo, S. K. Urek, A. Lobnik,
C. E. Housecroft, E. C. Constableb and L. J. Scherer, Analyst,
2014, 139, 4335–4342.
28 T. S. Singh, S. Mitra, A. K. Chandra, N. Tamai and S. Kar,
J. Photochem. Photobiol., A, 2008, 197, 295–305.
29 J. Zou, Q. Yu and Z. Shang, J. Chem. Soc., 2001, 2, 1439–1443.
30 M. J. Kamlet, J. M. Abboud, M. H. Abraham and R. W. Taft,
J. Org. Chem., 1983, 48, 2877–2887.
´
31 J. Catalan, J. Phys. Chem. B, 2009, 113, 5951–5960.
5 F. J. Duarte, Tunable Laser Optics, New York, Elsevier-
Academic, 2003.
6 K. P. Carter, A. M. Young and A. E. Palmer, Chem. Rev., 2014,
114, 4564–4601.
7 X. M. Peng, G. L. Damu and C. Zhou, Curr. Pharm. Des.,
2013, 19, 3884–3930.
8 M. Mladenovic, N. Vukovic, N. Niciforovic, S. Sukdolak and
S. Solujic, Molecules, 2009, 14, 1495–1512.
32 N. S. Moyon, A. K. Chandra and S. Mitra, J. Phys. Chem. A,
2010, 114, 60–67.
33 N. S. Moyon, P. M. Gashnga, S. Phukan and S. Mitra, Chem.
Phys., 2013, 421, 22–31.
34 N. S. Moyon and S. Mitra, J. Phys. Chem. A, 2011, 115, 2456–2464.
35 M. A. Rohman, P. Baruah, S. O. Yesylevskyy and S. Mitra,
Chem. Phys., 2019, 517, 67–79.
36 H. Dvir, I. Silman, M. Harel, T. L. Rosenberry and J. L. Sussman,
Chem. – Biol. Interact., 2010, 187, 10–22.
37 P. Baruah, M. A. Rohman, S. O. Yesylevskyy and S. Mitra,
Bioimpacts, 2019, 9, 85–96.
38 P. Anand, B. Singh and N. Singh, Bioorg. Med. Chem., 2012,
20, 1175–1180.
39 M. M. Islam, A. B. Gurung, A. Bhatacharjee, K. Aguan and
S. Mitra, Chem. – Biol. Interact., 2016, 249, 1–9.
40 M. A. Rohman, D. Sutradhar, S. Sangilipandi, K. Mohan
Rao, A. K. Chandra and S. Mitra, J. Photochem. Photobiol., A,
2017, 341, 115–126.
9 A. S. Al-Ayed and N. Hamdi, Molecules, 2014, 19, 911–924.
10 K. V. Sashidhara, A. Kumar, M. Kumar, J. Sarkar and
S. Sinha, Bioorg. Med. Chem. Lett., 2010, 20, 7205–7211.
11 O. Pelkonen, H. Raunio, A. Rautio, M. Pasanen and
M. A. Lang, The Metabolism of Coumarin. Coumarins: Biology,
Applications and Mode of Action, Chichester, John Wiley &
Sons, 1997, pp. 67–92.
12 J. I. Weitz, Blood Coagulation and Anticoagulant, Fibrinolytic,
and Antiplatelet Drugs, in Goodman & Gilman’s: The Pharma-
cological Basis of Therapeutics, 12th edn, 2011, ISBN: 978-0-07-
175352-4.
41 J. Kuchlyan, D. Banik, A. Roy, N. Kundu and N. Sarkar,
J. Phys. Chem. B, 2014, 118, 13946–13953.
42 S. Banerjee, S. Roy and B. Bagchi, J. Phys. Chem. B, 2010, 114,
12875–12882.
43 S. Jacob and A. B. Nair, Drug Dev. Res., 2018, 79, 201–217.
44 G. Tiwari, R. Tiwari and A. K. Rai, J. Pharm. BioAllied Sci.,
2010, 2, 72–79.
13 R. D. Thornes, Ir. Med. J., 1983, 76, 225.
14 J. R. Casley-Smith, in Coumarin in the Treatment of Lymphoedema
and Other High-Protein Oedemas. Coumarins: Biology, Applications
and Mode of Action, ed. R. O’Kennedy and R. D. Thornes, John
Wiley & Sons, Chichester, 1997, vol. 46, pp. 143–184.
15 R. M. Christie, K. M. Morgan and M. S. Islam, Dyes Pigm.,
2008, 76, 741–747.
45 B. D. Wang, J. Hai, Z. C. Liu, Q. Wang, Z. Y. Yang and
S. H. Sun, Angew. Chem., Int. Ed., 2010, 49, 4576–4579.
16 W. Lin, L. Yuan, J. Feng and X. Cao, Eur. J. Org. Chem., 2008,
2689–2692.
18724 | New J. Chem., 2019, 43, 18713--18725 This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019

View Article Online
NJC
Paper
46 J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, 55 J. N. Miller, Proc. Anal. Div. Chem. Soc., 1979, 16, 203–208.
Singapore, 2006.
56 D. Bhattacharjee, S. K. Sheet, S. Khatua, K. Biswas, S. Joshi
and B. Myrboh, Bioorg. Med. Chem., 2018, 26, 5018–5028.
57 B. Valeur and M. N. Berberan-Santos, Molecular fluorescence:
principles and applications, Wiley-VCH, 2012.
47 M. A. Rohman, K. Saha and S. Mitra, J. Biomol. Struct. Dyn.,
2019, 37, 4737–4746.
48 V. K. Sonu, M. M. Islam, M. A. Rohman and S. Mitra,
Colloids Surf., B, 2016, 146, 514–522.
58 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991–1024.
49 Z. Sharif-Barfeh, S. Beigoli, S. Marouzi, A. S. Rad, A. Asoodeh 59 S. Phukan, M. Saha, A. K. Pal, A. C. Bhasikuttan and S. Mitra,
and J. Chamani, J. Solution Chem., 2017, 46, 488–504. J. Photochem. Photobiol., A, 2015, 303, 67–79.
50 A. A. Moosavi-Movahedi, A. R. Golchin, K. Nazari, J. Chamani, 60 P. R. Bevington, Data Reduction and Error Analysis for the
A. A. Saboury, S. Z. Bathaie and S. Tangestani-Nejad,
Thermochim. Acta, 2004, 414, 233–241.
Physical Sciences, McGraw-Hill, New York, 1969.
61 T. Halgren, J. Comput. Chem., 1996, 17, 490–519.
51 P. D. Ross and S. Subramanian, Biochemistry, 1981, 20, 3096–3102. 62 D. Seeliger and B. L. de Groot, J. Comput.-Aided Mol. Des.,
52 P. B. Kandagal, S. Ashoka, J. Seetharamappa, S. M. T. Shaikh, 2010, 24, 417–422.
Y. Jadegoud and O. B. Ijare, J. Pharm. Biomed. Anal., 2006, 41, 63 E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch,
393–399.
D. M. Greenblatt, E. C. Meng and T. E. Ferrin, J. Comput.
Chem., 2004, 25, 1605–1612.
´
53 V. D. Porta, E. Bramanti, B. Campanella, M. R. Tine and
C. Duce, RSC Adv., 2016, 6, 72386–72398.
54 R. W. Woody, Circular Dichroism and the Conformational Analysis
of Biomolecules, Plenum Press, New York, 1996, pp. 25–67.
64 M. Fasano, S. Curry, E. Terreno, M. Galliano, G. Fanali,
P. Narciso, S. Notari and P. Ascenzi, IUBMB Life, 2005, 57,
787–796.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2019 New J. Chem., 2019, 43, 18713--18725 | 18725