Mechanism and structure studies of cinnamaldehyde/cyclodextrins inclusions by computer simulation and NMR technology

Article abstract of DOI:10.1016/j.carbpol.2018.04.055

This work aims to explore the inclusion mechanism and structure of cinnamaldehyde (CNMA) and cyclodextrins (CDs), and to provide some theoretical information for the application of CNMA and its inclusion. In this study, we prepared three kinds of inclusion and investigated the mechanism and structure by theory and experiment. Molecular docking and dynamical simulations presented a stable 1:1 inclusion complex and the visual structure model. The structural features indicated that the benzene ring of CNMA was enclosed in the hydrophobic cavity of CDs, which were consistent with the results of ¹H NMR, 2D-ROESY, Fourier transform infrared spectroscopy. The inclusion mechanism studies showed that the inclusion process was driven mainly by enthalpy with the binding constant following the order of DM (dimethyl) > HP (hydroxypropyl) > β-CD. Moreover, the inclusion complex showed an advantageous water solubility and dissolution rate compared with CNMA.

Full text of DOI:10.1016/j.carbpol.2018.04.055

Accepted Manuscript
Title: Mechanism and structure studies of
cinnamaldehyde/cyclodextrins inclusions by computer
simulation and NMR technology
Authors: Qiaomei Sun, Peixiao Tang, Ludan Zhao, Hongyu
Pu, Yuanming Zhai, Hui Li
PII:
S0144-8617(18)30441-7
DOI:
Reference:
https://doi.org/10.1016/j.carbpol.2018.04.055
CARP 13509
To appear in:
Received date:
Revised date:
Accepted date:
26-1-2018
11-4-2018
14-4-2018
Please cite this article as: Sun, Qiaomei., Tang, Peixiao., Zhao, Ludan.,
Pu, Hongyu., Zhai, Yuanming., & Li, Hui., Mechanism and structure studies
of cinnamaldehyde/cyclodextrins inclusions by computer simulation and NMR
technology.Carbohydrate Polymers https://doi.org/10.1016/j.carbpol.2018.04.055
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Mechanism and structure studies of cinnamaldehyde/cyclodextrins
inclusions by computer simulation and NMR technology
Qiaomei Suna, Peixiao Tanga*, Ludan Zhaoa, Hongyu Pua, Yuanming Zhaib*, and Hui
Lia
a School of Chemical Engineering, Sichuan University, Chengdu 610065, China
b Analytical & Testing Center, Sichuan University, Chengdu 610064, China
*Corresponding author. Address: College of Chemical Engineering, Sichuan University, Chengdu,
Sichuan, China. Tel.: +86 028 85405220; Fax: +86 028 85401207.
E-mail address: tangpeixiao@126.com (Peixiao Tang)；yuanmingzhai@scu.edu.cn (Yuanming
Zhai)
Highlights
 Inclusion mechanisms and structures of CNMA-CDs were investigated.
 Docking and molecule dynamics predicted detailed inclusion information.
 The inclusion abilities followed the order: DM > HP > β-CD.
 The benzene ring of CNMA inserted into hydrophobic cavity of CDs.
 The solubility and dissolution rate of CNMA were improved after inclusion.
Abstract
This work aims to explore the inclusion mechanism and structure of cinnamaldehyde
1

(CNMA) and cyclodextrins (CDs), and to provide some theoretical information for
the application of CNMA and its inclusion. In this study, we prepared three kinds of
inclusion and investigated the mechanism and structure by theory and experiment.
Molecular docking and dynamical simulations presented a stable 1:1 inclusion
complex and the visual structure model. The structural features indicated that the
benzene ring of CNMA was enclosed in the hydrophobic cavity of CDs, which were
1
consistent with the results of H-NMR, 2D-ROESY, Fourier transform infrared
spectroscopy. The inclusion mechanism studies showed that the inclusion process was
driven mainly by enthalpy with the binding constant following the order of DM
(dimethyl) > HP (hydroxypropyl) > β-CD. Moreover, the inclusion complex showed
an advantageous water solubility and dissolution rate compared with CNMA.
Key words: Cinnamaldehyde; Cyclodextrin; Inclusion mechanism; Structure;
Molecule dynamical simulation
1. Introduction
Cinnamaldehyde (CNMA, Fig. 1) is a component of essential oils derived from
cassia, which is utilized in several products, including perfumes, essences, and food
preservatives. Moreover, the 50% lethal concentration of CNMA was 62 μM for
human breast cancer cell MCF-7 indicating potential anticancer bioactivity (Nagle et
al., 2012). Administration 20 mg/kg orally for 2 months showed significant
improvement for diabetic rats, which showed the antidiabetes bioactivity of CNMA
2

(Anand, Murali, Tandon, Murthy, & Chandra, 2010). And antithrombus bioactivity of
CNMA was revealed when the mortality rate of collagen-epinephrine-induced acute
pulmonary thromboembolism effectively reduced with oral administration 250 mg/kg
in mice (Huang, Wang, Luo, Xie, & Shi, 2007). However, CNMA is highly volatile,
oil liquid and therefore exhibits poor aqueous solubility and stability under ambient
temperature. This property is the main factor hindering the application and
development of CNMA. Several studies have explored various methods, including
α-cyclodextrin inclusion, corn starch inclusion, and microcapsules, to improve the
solubility properties of CNMA (Chun, Jo, Bjrapha, Choi, & Min, 2015; Makwana,
Choudhary, Dogra, Kohli, & Haddock, 2014; Tian et al., 2013).
Cyclodextrin (CD) is an efficient material to improve physicochemical properties
of guest through the inclusion due to its biocompatibility, relatively non-toxicity, and
low price (Carrier, Miller, & Ahmed, 2007). The inclusion technology has been
widely used in food industries. For example, researchers have investigated the
inclusion structure and properties between food additives and cyclodextrins (Castro et
al., 2016; Fernándezgarcía & Pérezgálvez, 2017; Zeynep, Semran Ipek, Engin, &
Tamer, 2016). These literatures have used cyclodextrin to improve the stability and
release rate of food additives in different ways, to expand the application range and
field of food additives. Hence, the studies of the mechanism and structure of the
inclusion is of great importance to clarify and define the inclusion process.
Many methods ranging from computer simulations to experimental approaches
3

are available for characterizing the structure and mechanism between guest and
cyclodextrins (Abdolmaleki, Ghasemi, & Ghasemi, 2017; Li et al., 2015). Nuclear
magnetic resonance (NMR) is the most widely used because it provides reliable
1
chemical information (Pessine, Calderini, & Alexandrino, 2012). H-NMR, NOE
experiments (ROESY and NOESY), and solid-state NMR are the common methods
used for studying inclusions in the literatures (Ferro et al., 2017; Liscieplak,
Sitkowski, & Kolodziejski, 2014). Moreover, the development of computer
technologies has rendered computer simulation a key strategy for inclusion research,
including docking (Zhang, Liu, Yang, Chen, & Jiao, 2017) and dynamic simulation
(Li et al., 2015). Therefore, the combination of experimental methods and the
theoretical simulation could benefit to obtain reliable information about inclusion
mechanism and structural characteristic.
In this study, β-cyclodextrin (β-CD), hydroxypropyl-β-cyclodextrin (HP-β-CD),
and dimethyl-β-cyclodextrin (DM-β-CD) (Fig. 1) were chosen to prepare an inclusion
complex with CNMA. This study aimed to gain insight into the structure and
mechanism of the inclusion complex by performing computer simulations and
experiments. Molecular docking and dynamics simulation were conducted to
intuitively predict the combination mode and the inclusion structure. The inclusion
mechanism (including inclusion ratio, apparent binding constant, and thermodynamic
parameters) were determined through NMR and isothermal titration calorimetry (ITC).
The characterization of structure and morphology of the CNMA-CDs inclusion
4

complex were studied by 2D-ROESY and solid-state characterization method. And,
the encapsulation efficiency, aqueous solubility, and dissolution were investigated for
all inclusion complexes. The results of this study would contribute to the application
and development of CNMA in the food and pharmaceutical industries.
Fig. 1. Structure and atom labels of CDs and CNMA.
2. Experimental
2.1. Materials and reagents
CNMA was obtained from J&K Scientific Ltd. (Beijing, China). β-CD (FW =
1135, purity ≥ 98%), DM-β-CD (DS = 2.0, FW = 1331.39, purity ≥ 98%), and
HP-β-CD (DS = 1.0, FW= 1541.54, purity ≥ 98%) were acquired from Chengdu
Kelong Chemical Co., Ltd. (Chengdu, China). DMSO-d6 was purchased from J&K
Scientific Ltd. (Beijing, China). The other reagents and chemicals used were of
analytical reagent grade. Tri-distilled water was used throughout the experiment.
2.2. Preparation of inclusion complexes
5

To prepare the inclusion complexes, β-CD/DM-β-CD/HP-β-CD was combined
with CNMA at a 1:1 molar ratio (Ponce Cevallos, Buera, & Elizalde, 2010). The
mixed solution intermittently reacted ultrasonically for 0.5 h and magnetically stirred
at 60 ℃ for 8 h. The reaction process maintained the sealing state to prevent the
volatilization loss of CNMA. Subsequently, the reaction solution was filtered to
remove solid residues and fully frozen in a refrigerator at -20 ℃. The final complexes
obtained by drying in a laboratory vacuum freeze dryer for two days, and the
temperature of the condensing plate was -60 ℃. The inclusion complexes were
protected from light in a dry place. For comparison, corresponding binary physical
mixtures with the same ratio at 1:1 were prepared by simply mixing CNMA and CD
for 10 min by using a vortex blender.
2.3. Molecular docking and dynamics simulations
The 3D structure of CNMA was built and optimized from ChemBio Office 2012.
The 3D cyclodextrin structures were downloaded from Cambridge Structural
Database (Ref. code: BCDEXD03 for β-CD; Ref. code: BOYFOK03 for DM-β-CD),
and all attached water molecules were removed. The structure of HP-β-CD was
established on the basis of β-CD, and optimized by ChemBio3D Ultra through MM2
molecular mechanics method. Molecular docking and MD simulations were
performed on YASARA v16.7.22 with AMBER14 force field (Krieger, Darden,
Nabuurs, Finkelstein, & Vriend, 2004). Partial atomic charges of CNMA were
computed using the AM1-BCC model, and counter ions (Na+ or Cl-) were added to
6

obtain a charge-neutral system. Simulations in explicit water were conducted at a
constant temperature (298.15 K) in pH 7.4 after initial energy minimization
procedures. Periodic boundary conditions were applied to the system. The long-range
Coulomb interactions were applied using particle-mesh Ewald (PME) summation and
a cut-off of 8.0 Å. The system with the lowest energy was analyzed within the
docking results, and chosen to run MD simulations. Simulations were carried out
using a pre-defined macro (md_run) within YASARA package. Multiple time steps
were used in the simulation: 1.25 fs for the intramolecular forces, 2.5 fs for the
intermolecular forces, and simulation data were collected every 100 ps. Analysis of
MD data was conducted using pre-defined macro (md_analyze, and
md_analyzebindenergy) within YASARA package.
2.4. NMR experiment preparations
1H-NMR spectra were obtained by using a Bruker Advance 600 Hz spectrometer
(Germany) with a 5 mm probe in DMSO-d6 at 25 ℃. For Job’s plot, the total
concentration constants at 10 mM, and different volumes of 10 mM solutions of
1
CNMA and CDs were mixed to obtain the H NMR spectra. To determine the
apparent binding constant (Scott’s plot), constant CDs (2 mM) were mixed with
different concentrations of CNMA to attain final CNMA/CDs ratios ranging from 0.0
to 4.0. All samples were interacted to equilibrate before the measurement. 2D ROESY
was recorded by a Bruker Advance 600 Hz spectrometer (Germany) at 25 ℃ with a
mixing time of 300 ms.
7

2.5. Characterization of inclusion complexes
The Fourier transform infrared (FT-IR) spectra of CNMA, CDs, and their
complexes were recorded on a Nicolet 6700 spectrometer (Thermo Fisher Scientific,
USA) at a scanning scope ranging from 4000 cm−1 to 400 cm−1. The spectra were
obtained by averaging 64 scans with a resolution of 4 cm-1. Potassium bromide
palettes were used for all spectra.
Differential scanning calorimetry (DSC) curves were recorded with a TA Q200
thermal analyzer (TA Instruments Co., New Castle, DE, USA) in the range of -50 ℃
to 200 ℃ with the heating rate of 10 ℃/min. Thermogravimetric analysis (TGA) was
conducted using a thermogravimetric analyzer (TG 209F1 Iris, NETZSCH, Germany)
from 30 ℃ to 400 ℃. The heating rate was 10 ℃/min. All samples were sealed in
closed alumina crucibles with an empty reference pan under nitrogen flow.
Scanning electron microscopy (SEM) images of the complexes were examined
by a JSM-7500F scanning electron microscope (JEOL, Japan). The samples were
sputter-coated with a thin gold layer prior to analysis at 5.0 kV.
2.6. ITC analysis
ITC measurements were performed at 298 K using an ITC200 isothermal
titration calorimeter (MicroCal; USA). A 2 mM CNMA solution was placed in a
sample cell and titrated by 20 mM CD solution at 500 rpm and at 25 ℃. Twenty
successive 2 μL aliquot injections (with an initial injection of 0.4 μL) were conducted
with a 40 μL auto-syringe filled with CDs. All solutions were degassed before
8

measurement to eliminate air bubbles. The resulting corrected injection heats were
plotted through model-fitting by using the MicroCal Origin software provided with
the instrument. Each experiment was performed in triplicates.
2.7. Encapsulation efficiency, solubility, and dissolution tests
Given that the filter residue reflects the amount of unbound CNMA in the
inclusion complex, the encapsulation efficiency of CNMA in the complexes was
calculated by the following formula in triplicates:
( )
Encapsulation efficiency %
=
Total amount of drug−Amount of unbound drug × 100
(1)
Total amount of drug
The water solubility of the complex was evaluated by preparing a saturated
solution (Huang et al., 2014). The test was conducted by adding excess CNMA into
CD aqueous solutions (0–20 mM), which was then stirred for 24 h at 25 ℃ to achieve
equilibrium. The solutions were filtered through a 0.45 μm hydrophilic membrane and
then detected at 280 nm using a UV spectrophotometer (TU-1901, Beijing Purkinje
General Instrument Co., Ltd., China) after being appropriately diluted. Each
experiment was repeated thrice.
In vitro dissolution analyses were performed using a ZRC-8D dissolution tester
(Chuangxing, Tianjin, China) with the rotate speed of 100 rpm at 37 ℃. Powdered
samples (30 mg) of CNMA and inclusion complexes containing 30 mg of CNMA
were separately added to tri-distilled water (1000 mL) at the same time. Aliquots (5
mL) of the sample solutions were removed at every fixed time, and filtered through a
9

0.45 μm hydrophilic membrane. An equal amount of water was replenished in a
timely manner to maintain the solvent volume and avoid the dilution effect. All
samples were diluted appropriately to measure the concentration with a UV
spectrophotometer.
3. Results and discussion
3.1. Theoretical analysis of the formation process of inclusions
Computer simulation can provide insight into the inclusion between guests and
hosts, and verify the experimental results in theory (Chen, Chang, & Gilson, 2004;
Yang et al., 2017). In the present work, molecular docking studies were conducted to
obtain a visual image of the inclusion of CNMA with CDs (Fig. 2A). The benzene
ring of CNMA inserted into the cyclodextrin cavity, and bound with the primary
hydroxyl groups of CD molecules through hydrophobic effect. Hydrogen bonding
was formed between the carbonyl moiety and hydroxyls of β-CD to stabilize the
inclusion structure. Thus, the shape and structure of CDs can suitably accommodate
guest molecules to form a clathrate structure.
10

Fig. 2. (A) Representative docking conformations of (a) CNMA-β-CD, (b)
CNMA-HP-β-CD, and (c) CNMA-DM-β-CD system. Hydrophobic effects are
signified by green-fine lines, and hydrogen bonds are shown as yellow dash lines. (B)
RMSDs values of three inclusion systems obtained by molecular dynamical
simulations to judge system stability. (a) CNMA-β-CD system, (b) CNMA-HP-β-CD
system, and (c) CNMA-DM-β-CD system.
In order to predict the stoichiometric ratio between CNMA and CDs, 1:1 and 1:2
clathrate structures were built by molecular docking to perform molecular dynamics
simulations (MD). The MD was performed within the most optimal conformation
with pH (7.4), temperature (298K), and water solvent. In 1:2 inclusion models (Fig.
S1), the structure of a 1:2 inclusion is unstable. The inclusion model became 1:1
11

stoichiometric ratio with the passage of time, resulting in a stable binding structure.
However, the inclusion complexes with a 1:1 stoichiometry were stable in MD
process (Fig. S2). Thus, the host and guest molecules were likely to be a 1:1 clathrate
structure.
The inclusion complexes with a 1:1 stoichiometry were further analyzed for
obtaining more detailed kinetic information. The root mean square deviation (RMSD)
is an index that measures the structural drift from the initial coordinates and the
atomic fluctuation in the process of MD simulation (Fujiwara, & Amisaki, 2006). The
RMSD values in CNMA with β-CD, HP-β-CD, DM-β-CD systems reached plateau at
about 50 ps, 40 ns, and 33 ns, respectively (Fig. 2B). The RMSD values calculated
from the trajectory data were 1.36±0.58, 2.85±0.22, and 2.16±0.35 Å for β-CD,
HP-β-CD, DM-β-CD systems, respectively. Hence, CNMA-β-CD system exhibited
the smallest atomic fluctuation. And, the binding energy simulations revealed that the
average binding energy of the three inclusion systems followed the order:
CNMA-DM-β-CD > CNMA-HP-β-CD > CNMA-β-CD (Table 1), and the binding
energy of the systems reached balance at 50 ns indicating that the inclusion system
reached a stable state (Fig. S3). The binding energy results exhibited that the
modification of the group could improve the inclusion capacity of β-CD (Zhang et al.,
2017).
12

Table 1 Information obtained from molecular dynamics simulations of free CNMA,
free CDs, and inclusion complexes after equilibrium was achieved (50 ns).
Average binding
Bond energy
(KJ/mol)
--
Coulomb
(KJ/mol)
-90463
-92018
-90818
-89269
-91948
-90442
-88925
VdW
(KJ/mol)
13035
12655
12333
12360
12542
12188
12232
System
energy (KJ/mol)
CNMA
β-CD
--
--
12495
13953
14510
12549
14107
14549
HP-β-CD
--
DM-β-CD
--
CNMA-β-CD
CNMA-HP-β-CD
CNMA-DM-β-CD
38.38
41.87
60.48
In order to insight into the binding properties and the system stabilities, a
detailed analysis was conducted on the MD data (Table 1). The results showed that
the bond energy was maximal in DM-β-CD inclusion, implying that CNMA had the
strongest bonding force in DM-β-CD inclusion. Moreover, there are intermolecular
forces on all atoms, and the perturbation of the Coulomb and Van der Waals (VdW)
interaction also revealed the changes of the simulation system. Coulomb force is a
very important part of the force field between charged atoms. The strong Coulomb
force indicated a strong electrostatic force and a compact structure in inclusion system
(CNMA-DM-β-CD > CNMA-HP-β-CD > CNMA-β-CD) compared with free CDs
and CNMA. The VdW force is a parameter that can reflect the melting points of the
compound (Rao & Sunkada, 2007). Thus, inclusion complex should have a higher
melting point, which was confirmed by thermal analysis.
13

Structural information could also gain from the molecular dynamics simulations.
To visualize the structural deviations of the complex, no free water was retained in the
cavity after complexation with CNMA (Fig. S4). As shown in Fig. S2, the trajectory
analysis showed that the simulated configurations were structurally similar to the
initial form, implying that CNMA steadily bound with CDs. The benzene ring moiety
of CNMA consistently remained in the hydrophobic cavity of CDs, and this finding
agreed with the docking results. Slight changes in the CD structure were observed in
β-CD and DM-β-CD, whereas a slightly more evident change was noted in HP-β-CD,
which indicated that CNMA-HP-CD system was perturbed.
3.2. Characterization of inclusion complex by NMR spectroscopy
3.2.1. 1H NMR experiments
In this work, the formation of CNMA-CDs inclusion complex was clearly
demonstrated by NMR, and the process of inclusion was elucidated preliminarily. The
1H NMR spectra of CNMA in the absence and presence of CDs are presented in Fig.
3. The hydrogen chemical shift (δ) and chemical shift differences (Δδ) of CDs and
CNMA were shown in Table 2 and Table S1, respectively. In Table 2, inclusion with
CNMA exhibited small changes in both β-CD and HP-β-CD, and a significant effect
on the chemical shifts of protons of DM-β-CD. The chemical shift of H3, H5, and H6
protons of CDs were changed obviously, indicating that these protons mainly
contributed to the inclusion with CNMA. This result was consistent with the previous
14

results (Li et al., 2015). The chemical shift of H4,6 and H7 protons of CNMA
contributed to the inclusion with CDs due to their evident changes (Table S1).
Compared with the free CNMA and CDs, the chemical shifts of the inclusion
complexes were observed, which indicated that CNMA had reacted with CDs.
Table 2 Chemical shifts (δ) and chemical shift differences (Δδ) of CDs and their
corresponding inclusion complexes with CNMA in DMSO-d6.
Hydrogens
β-CD
H1
H2
H3
H4
3.284
3.287
0.003
-
H5
H6
H7
H4,7,8
H9
δ
δ
4.809
4.811
0.002
5.695
5.692
3.333
3.335
0.002
3.403
3.400
4.431
4.435
0.004
4.513
4.507
3.599
3.601
0.002
3.730
3.736
3.535
3.537
0.002
3.593
3.590
-
-
-
CNMA-β-CD
-
-
-
Δδ
δ
-
-
-
HP-β-CD
-
3.204
1.005
CNMA-HP-β-CD
δ
-
-
3.205
1.007
Δδ
δ
-0.003 -0.003 -0.006
-
0.006 -0.003
-
0.001
0.002
DM-β-CD
5.773
5.783
3.477
3.432
4.720
4.773
3.357
3.352
3.633
3.697
3.379
3.557
0.178
3.228
3.232
0.004
-
-
-
-
-
-
CNMA-DM-β-CD
δ
Δδ
0.010 -0.045 0.053 -0.005 0.064
15

Fig. 3. 1H NMR spectrum of pure CNMA, pure CDs, and inclusion complexes in
DMSO-d6 at room temperature.
3.2.2. Inclusion mode of the inclusion complex
Job’s plot was performed to judge the stoichiometry of the inclusion complex
between CNMA and CDs (Yee et al., 2017). Fig. 4A showed the chemical shifts of
H-5 (CDs) for a CD mole fraction (rCD) ranging from 0.0 to 1.0. The observed
significant chemical shift was due to the formation of an inclusion complex. It was
noteworthy that the shapes of the Job's plots were not symmetrical and the maximum
chemical shift difference was 0.4 ~ 0.5, which indicated that the complex of CNMA
16

with CDs could be a co-existing mixture through 1:1 and 1:2 stoichiometry (Huang,
1982). However, it had been proved that a 1:2 stoichiometry complex was unstable
and negligible by the molecule docking results, thus, a 1:1 stoichiometry complex was
chosen as a stable state.
Fig. 4. (A) Job’s plot of three inclusion systems. 2D ROESY spectra of (B)
CNMA-β-CD, (C) CNMA-HP-β-CD, and (D) CNMA-DM-β-CD system.
3.2.3. Inclusion ability of the inclusion complex
The apparent binding constant (Kb) can be calculated according to the Scott’s
plot when the guest and the host form a 1:1 inclusion (Choi et al., 2017)
[퐶푁푀퐴]
[퐶푁푀퐴]
1
=
+
퐾 ∆훿
푏
푚푎푥
(2)
∆훿
∆훿
표푏푠
푚푎푥
where ∆훿표푏푠 is the observed chemical shift difference of CD protons with and
17

without CNMA, and ∆훿푚푎푥 is the chemical shift difference at saturation. The results
fitted by Eq. 2 were shown in Fig. 5. According to the slope and intercept, the
apparent binding constants of the three systems could be obtained as follows: H5
(2930 M-1) ≈ H6 (2915 M-1) > H3 (1409 M-1) > H2 (493 M-1) in CNMA-β-CD system,
H5 (16754 M-1) > H9 (3652 M-1) > H2 (2829 M-1) > H3 (1125 M-1) in
CNMA-HP-β-CD system, and H6 (17522 M-1) > H5 (3300 M-1) > H7 (2658 M-1) >
H2,4 (1825 M-1) in CNMA-DM-β-CD system. The inclusion ability followed the
ordered with DM-β-CD > HP-β-CD > β-CD, and H5 and H6 played the main roles in
the interaction process. DM-β-CD had the highest binding constant, which was the
most suitable for forming a complex with CNMA. These results agreed with the
average binding energy obtained from molecular dynamics simulations.
This result can be explained by two possible reasons: one is the cavity size and
compact structure of cyclodextrin, and the other one is its hydrophobic nature. For the
first reason, given that the guest has a small molecular weight and a plane structure,
the modified β-CD cavity structure is more compact and more conducive to an
inclusion with the guest. For the second reason, since hydrophobic effect is the main
driving force of inclusion, the inclusion ability of methyl modified β-CD is stronger
than that modified by hydroxyl, and the hydrophobic group of CNMA is mainly
responsible for the combination.
18

Fig. 5. Scott’s plot for three inclusion system with CD concentration set at 2 mM and
variable CNMA concentration at pH 7.4 at room temperature. (A) is for CNMA-β-CD
system. (B) is for CNMA-HP-β-CD system. (C) is for CNMA-DM-β-CD system.
3.2.4. Structure of the inclusion complex
In order to further identify the structure and inclusion modes of the complexes in
the solution, 2D-ROESY was conducted in D2O at room temperature. The spectra
presented the NOE cross-peaks between the protons for spatial contacts within 4 Å
(Schneider, Hacket, Rüdiger, & Ikeda, 1998). Fig. 4 showed the 2D-ROESY spectra
of the three systems. In CNMA-β-CD system, the cross peak correlations of
H1,2,3,4,6,7 (CNMA) and H2,3,4,5,6 (β-CD) were observed to determine the entire
CNMA molecule inserted into the β-CD cavity. H3,5 (DM-β-CD) and H1,3,5
(HP-β-CD) displayed cross peak correlations with H1,2,3,4,6,7 (CNMA), indicating
that the CNMA molecule penetrated the hydrophobic cavity. This inclusion mode
wherein the entire guest molecule interacted with CDs may be due to two possible
reasons. One is the small molecular weight of CNMA. The other is that aromatic
compounds typically have a pronounced interaction with the CD hydrophobic cavity
19

because of the aromatic ring (Schneider, Hacket, Rüdiger, & Ikeda, 1998). These
results confirmed the interactions between the guest and host molecules and that the
benzene ring of CNMA plays an important role in inserting into the hydrophobic
cavities of CDs.
3.3. Solid-state characterization of inclusion complexes
3.3.1. FT-IR spectra
Changes in the FT-IR spectra may be provide insight into the potential
guest–CDs interaction (Almarzouqi, Elwy, Shehadi, & Adem, 2009). The strong peak
of CNMA at ~1620 cm−1, which was attributed to the stretching vibrations of C=O
groups, appeared in the inclusion complex (Fig. 6A). However, several peaks of the
vibrations of the aromatic ring framework around at 1700 - 2000 cm−1 and two peaks
of the stretching vibrations of C-H of the aldehyde group at ~2700 and ~2800 cm−1
completely disappeared compared with physical mixture in the three systems. Thus,
the inclusion complexes formed between CNMA and CDs, with the aromatic moiety
and aldehyde group of CNMA included in the CD cavities.
3.3.2. DSC and TGA analysis
DSC and TGA are useful tools for identifing the interaction by using
thermograms (Maeda, Yu, & Nakayama, 2013; Marques, Hadgraft, & Kellaway, 1990;
Taupitz, Dressman, Buchanan, & Klein, 2013). The endothermic peak around -19 ℃,
which is the melting point of CNMA, was observed and increased to a higher
temperature in the inclusion complex in DSC curves (Fig. 6B). The inclusion process
20

improves the thermal stability of CNMA. In Fig. 6C, the TGA results of CNMA and
its inclusion complex were shown. The mass loss of less than 100 °C can be
associated to water loss of CDs. The significant mass loss around 130 °C in the
HP/DM-β-CD inclusions was mainly due to the evaporation of the CNMA from the
inclusion complex. Then, the decomposition of CDs occurred at a higher temperature.
These changes in the curves confirmed the formation of inclusion complexes. These
results indicated that the thermal stability of CNMA was modified after complexation
with CDs, particularly in β-CD. The results agreed with the VdW forces that found in
molecular dynamics simulation results.
3.3.3. SEM
SEM is an ideal method for measuring surface morphology, including particle
size, surface roughness, and surface texture of substances (Gong et al., 2016; Tang et
al., 2017). The SEM images of the samples are displayed in Fig. S5. The size and
surface morphologies were changed by the inclusion. The CNMA-β-CD complexes
showed a relatively more homogeneous shape with a rough and fluffy texture, an
irregular surface, and smaller sizes. Morphological analysis revealed that
CNMA-HP-β-CD and CNMA-DM-β-CD displayed an irregular blocky form, which
was markedly different from the shapes of free CDs. The morphological differences
of the materials confirmed the formation of a new solid phase.
21

Fig. 6. (A) IR spectra of (a) pure CNMA, (b) pure β-CD, (c) physical mixture of
CNMA and β-CD, (d) CNMA-β-CD, (e) pure HP-β-CD, (f) physical mixture of
CNMA and HP-β-CD, (g) CNMA-HP-β-CD, (h) pure DM-β-CD, (i) physical mixture
of CNMA and DM-β-CD, and (j) CNMA-DM-β-CD. (B) DSC curves of pure CNMA,
pure CDs, and inclusion complex. (C) TGA curves of pure CDs and inclusion
complex. (D) Release curves of CNMA in the pure state and inclusion complex in
water at 37℃.
3.4. Thermodynamic properties of the inclusion
Detailed thermodynamic information, including the stability constant of complex
(K), enthalpy change (ΔH), entropy change (ΔS), and changes of Gibbs free energy
22

(ΔG), were obtained by using the ITC method (Kitamura, Nakatani, Takaha, & Okada,
1999). The thermodynamic parameters were calculated by the van’t Hoff equation:
∆G = −RTln K
∆G = ∆H − T∆S
(3)
(4)
Fig. 7 showed the isothermal calorimetry titration curves of CNMA with three types
of CDs. Results obtained from the nonlinear least-squares fitting and the
aforementioned equations were presented in Fig. 7. ΔG and ΔH were negative in all
three systems, demonstrating that the inclusion processes were spontaneous,
exothermic (Belica et al., 2014). Moreover, the positive ΔS indicated that the degree
of disorder was larger in the system, which contribute to the spontaneity of the
reactions (Passos et al., 2013). The inclusion process were driven by both enthalpy
and entropy, however, the enthalpic effects of binding dominate over entropic effects
because of |ΔH| > |TΔS| (Belica et al., 2014). The binding constant (K) followed the
order of β < HP < DM in the inclusion progress. A larger binding constant
corresponded to an easier inclusion. The results were consistent with Scott’s plot. The
slight differences in the data may be due to the differences in sample preparation
(deuterium generation reagent etc.) and instruments.
23

Fig. 7. Isothermal calorimetry titration curves of CNMA with (A) CNMA-β-CD, (B)
CNMA-HP-β-CD, and (C) CNMA-DM-β-CD at 298 K.
3.5. Analysis of encapsulation efficiency, solubility, and dissolution
The encapsulation efficiency and solubility enhancements in water were
summarized in Table 3. The encapsulation abilities for CNMA followed the order:
β-CD < HP-β-CD < DM-β-CD, and HP-β-CD and DM-β-CD showed similar
encapsulation abilities. β-CD slightly affected the solubility of CNMA, and DM-β-CD
increased the solubility of CNMA to a greater extent than HP-β-CD did. The
dissolution rates of all samples were high. Free CNMA dissolved slowly (90% after
40 min) because of its poor water solubility (Fig. 6D). All samples demonstrated a
very high cumulative release percentage of over 90%. The reason for the increased
solubility and accelerated dissolution rate after inclusion was the structural features of
the CDs, which include a hydrophobic cavity and a hydrophilic outer surface.
24

Table 3 Encapsulation efficiency and solubility enhancement of inclusion complexes.
Inclusion complex
CNMA
Solubility in water (mM)
0.675±0.003
Encapsulation efficiency（%）
--
CNMA-β-CD
1.593±0.004
29.346±0.873
46.477±0.748
53.623±1.022
CNMA-HP-β-CD
CNMA-DM-β-CD
15.864±0.008
21.127±0.006
Values are presented as mean ± SD (n = 3).
4. Conclusion
In this work, the structural model and binding mechanism of the inclusion
complex CNMA-CDs were gained from computer simulation and correlated with
experimental data. CNMA and CDs formed a 1:1 inclusion complex. DM-β-CD
showed the strongest binding ability in the inclusion process with CNMA. The
benzene ring of CNMA inserted into the hydrophobic cavities of CDs, and H5,6 of
CDs and the benzene ring of CNMA played the main roles in the interaction, which
were the primary structure features in CNMA-CDs inclusion process. The complexes
of CNMA with CDs showed improved solubility, stability, and dissolution rate
compared with the pure drug. The research provides some theoretical and
experimental information for the application of CNMA and its inclusion complexes.
Acknowledgments
This work was supported by the Large-scale Science Instrument Shareable Platform
Construction of Sichuan Province (Grant No. 2015JCPT0005-15010102) and the
25

Large-scientific Instruments Sharing Platform Ability Construction of Sichuan
Province (Grant No. 2016KJTS0037).
References
Abdolmaleki, A., Ghasemi, F., & Ghasemi, J. B. (2017). Computer-aided drug design
to explore cyclodextrin therapeutics and biomedical applications. Chemical Biology &
Drug Design, 89(2), 257-268.
Almarzouqi, A. H., Elwy, H. M., Shehadi, I., & Adem, A. (2009). Physicochemical
properties of antifungal drug-cyclodextrin complexes prepared by supercritical carbon
dioxide and by conventional techniques. Journal of Pharmaceutical & Biomedical
Analysis, 49(2), 227-233.
Anand, P., Murali, K. Y., Tandon, V., Murthy, P. S., & Chandra, R. (2010).
Insulinotropic effect of cinnamaldehyde on transcriptional regulation of pyruvate
kinase, phosphoenolpyruvate carboxykinase, and GLUT4 translocation in
experimental diabetic rats. Chemico-biological interactions, 186(1), 72-81.
Belica, S., Jeziorska, D., Urbaniak, P., Buko, V. U., Zavodnik, I. B., & Pałecz, B.
(2014). Calorimetric and spectroscopic characterization of complexes between
β-cyclodextrin or heptakis (2,6-di-O-methyl)-β-cyclodextrin and sertraline
hydrochloride in aqueous solution. Journal of Chemical Thermodynamics, 70(70),
160-167.
Carrier, R. L., Miller, L. A., & Ahmed, I. (2007). The utility of cyclodextrins for
enhancing oral bioavailability. Journal of Controlled Release, 123(2), 78-99.
Castro, D. O., Tabary, N., Martel, B., Gandini, A., Belgacem, N., & Bras, J. (2016).
Effect of different carboxylic acids in cyclodextrin functionalization of cellulose
nanocrystals for prolonged release of carvacrol. Materials Science & Engineering C,
69, 1018-1025.
Chen, W., Chang, C. E., & Gilson, M. K. (2004). Calculation of Cyclodextrin Binding
26

Affinities: Energy, Entropy, and Implications for Drug Design. Biophysical Journal,
87(5), 3035-3049.
Chun, J. Y., Jo, Y. J., Bjrapha, P., Choi, M. J., & Min, S. G. (2015). Antimicrobial
Effect of α- or β-Cyclodextrin Complexes with Trans-Cinnamaldehyde Against
Staphylococcus aureus and Escherichia coli. Drying Technology, 33(3), 377-383.
Fernándezgarcía, E., & Pérezgálvez, A. (2017). Carotenoid:β-cyclodextrin stability is
independent of pigment structure. Food chemistry, 221, 1317-1321.
Ferro, M., Castiglione, F., Pastori, N., Punta, C., Melone, L., & Panzeri, W. (2017).
Dynamics and interactions of ibuprofen in cyclodextrin nanosponges by solid-state
NMR spectroscopy. Beilstein Journal of Organic Chemistry, 13, 182-194.
Fujiwara, S., & Amisaki, T. (2006). Molecular dynamics study of conformational
changes in human serum albumin by binding of fatty acids. Proteins-structure
Function & Bioinformatics, 64(3), 730–739.
Gong, L., Li, T., Chen, F., Duan, X., Yuan, Y., Zhang, D., & Jiang, Y. (2016). An
inclusion complex of eugenol into β-cyclodextrin: Preparation, and physicochemical
and antifungal characterization. Food chemistry, 196, 324-330.
Huang, C. Y. (1982). Determination of binding stoichiometry by the continuous
variation method: the job plot. Methods in Enzymology, 87(87), 509-525.
Huang, J., Wang, S., Luo, X., Xie, Y., & Shi, X. (2007). Cinnamaldehyde reduction of
platelet aggregation and thrombosis in rodents. Thrombosis Research, 119(3),
337-342.
Huang, Z., Tian, S., Ge, X., Zhang, J., Li, S., & Li, M. (2014). Complexation of
chlorpropham with hydroxypropyl-β-cyclodextrin and its application in potato sprout
inhibition. Carbohydrate Polymers, 107(107), 241-246.
Chou, J.M., Park, K., Lee, B., Jeong, D., Dindulkar, S.D., & Choi, Y. (2017).
Solubility and bioavailability enhancement of ciprofloxacin by induced oval-shaped
mono-6-deoxy-6-aminoethylamino-β-cyclodextrin. Carbohydrate Polymers, 163,
118-128.
27

Kitamura, S., Nakatani, K., Takaha, T., & Okada, S. (1999). Complex formation of
large‐ring cyclodextrins with iodine in aqueous solution as revealed by isothermal
titration calorimetry. Macromolecular Rapid Communications, 20(12), 612-615.
Krieger, E., Darden, T., Nabuurs, S. B., Finkelstein, A., & Vriend, G. (2004). Making
optimal use of empirical energy functions: Force‐field parameterization in crystal
space. Proteins-structure Function & Bioinformatics, 57(4), 678–683.
Li, S., Wang, L., Jiang, J., Tang, P., Wang, Q., Wu, D., & Li, H. (2015). Investigations
of bisacodyl with modified β-cyclodextrins: Characterization, molecular modeling,
and effect of PEG. Carbohydrate Polymers, 134, 82-91.
Liscieplak, A., Sitkowski, J., & Kolodziejski, W. (2014). Comparative proton nuclear
magnetic resonance studies of amantadine complexes formed in aqueous solutions
with three major cyclodextrins. Journal of Pharmaceutical Sciences, 103(1), 274-282.
Maeda, H., Yu, O., & Nakayama, H. (2013). Inclusion complexes of melatonin with
modified cyclodextrins. Journal of Inclusion Phenomena & Macrocyclic Chemistry,
78(1-4), 217-224.
Makwana, S., Choudhary, R., Dogra, N., Kohli, P., & Haddock, J. (2014).
Nanoencapsulation and immobilization of cinnamaldehyde for developing
antimicrobial food packaging material. LWT - Food Science and Technology, 57(2),
470-476.
Marques, H. M. C., Hadgraft, J., & Kellaway, I. W. (1990). Studies of cyclodextrin
inclusion complexes. I. The salbutamol-cyclodextrin complex as studied by phase
solubility and DSC. International Journal of Pharmaceutics, 63(3), 259-266.
Nagle, A. A., Gan, F. F., Gavin, J., Choon-Leng, S., Geoffrey, W., & Eng-Hui, C.
(2012). Induction of Tumor Cell Death through Targeting Tubulin and Evoking
Dysregulation of Cell Cycle Regulatory Proteins by Multifunctional
Cinnamaldehydes. PloS one, 7(11), e50125.
Passos, J. J., Sousa, F. B. D., Mundim, I. M., Bonfim, R. R., Melo, R., & Viana, A. F.,
et al. (2013). Double continuous injection preparation method of cyclodextrin
28

inclusion compounds by spray drying. Chemical Engineering Journal, 228(28),
345-351.
Pessine, F. B. T., Calderini, A., & Alexandrino, G. L. (2012). Review: Cyclodextrin
Inclusion Complexes Probed by NMR Techniques. InTech, 237-264
Ponce Cevallos, P. A., Buera, M. P., & Elizalde, B. E. (2010). Encapsulation of
cinnamon and thyme essential oils components (cinnamaldehyde and thymol) in
β-cyclodextrin: effect of interactions with water on complex stability. Journal of Food
Engineering, 99(1), 70-75.
Rao, S. P., & Sunkada, S. (2007). Making sense of boiling points and melting points.
Resonance, 12(6), 43-57.
Schneider, H. J., Hacket, F., Rüdiger, V., & Ikeda, H. (1998). NMR Studies of
Cyclodextrins and Cyclodextrin Complexes. Chemical Reviews, 98(5), 1755-1785.
Tang, P., Yang, H., Tang, B., Wu, D., Du, Q., Xu, K., & Li, H. (2017).
Dimethyl-β-cyclodextrin/salazosulfapyridine inclusion complex-loaded chitosan
nanoparticles for sustained release. Carbohydrate Polymers, 156, 215-222.
Taupitz, T., Dressman, J. B., Buchanan, C. M., & Klein, S. (2013).
Cyclodextrin-water soluble polymer ternary complexes enhance the solubility and
dissolution behaviour of poorly soluble drugs. Case example: itraconazole. European
Journal of Pharmaceutics & Biopharmaceutics, 83(3), 378-387.
Tian, Y., Zhu, Y., Bashari, M., Hu, X., Xu, X., & Jin, Z. (2013). Identification and
releasing characteristics of high-amylose corn starch-cinnamaldehyde inclusion
complex prepared using ultrasound treatment. Carbohydrate Polymers, 91(2),
586-589.
Yang, J., Tang, K., Qin, G., Chen, Y., Peng, L., & Wan, X., et al. (2017). Hydrogen
bonding energy determined by molecular dynamics simulation and correlation to
properties of thermoplastic starch films. Carbohydrate Polymers, 166, 256-263.
Yee, E. M., Hook, J. M., Bhadbhade, M. M., Vittorio, O., Kuchel, R. P., & Brandl, M.,
et al. (2017). Preparation, characterization and in vitro biological evaluation of (1:2)
29