Enantiomeric separation and simulation study of eight anticholinergic drugs on an immobilized polysaccharide-based chiral stationary phase by HPLC

Article abstract of DOI:10.1039/C8NJ00685G

The enantiomeric separation of eight anticholinergic drugs was first systematically examined on a derivative polysaccharide chiral stationary phase (CSP), i.e. Chiralpak ID in the normal phase mode. Except for scopolamine hydrobromide and benzhexol hydroc

Full text of DOI:10.1039/C8NJ00685G

NJC
View Article Online
View Journal
PAPER
Enantiomeric separation and simulation study of
eight anticholinergic drugs on an immobilized
polysaccharide-based chiral stationary phase by HPLC
a
Cite this:DOI: 10.1039/c8nj00685g
Meng Li,a Bo Zhang,a Jia Yu,a Jian Wang*b and Xingjie Guo
*
The enantiomeric separation of eight anticholinergic drugs was first systematically examined on a
derivative polysaccharide chiral stationary phase (CSP), i.e. Chiralpak ID in the normal phase mode.
Except for scopolamine hydrobromide and benzhexol hydrochloride, the other six analytes including
atropine sulfate, phencynonate, dipivefrine hydrochloride, tropicamide, homatropine methylbromide and
oxybutynin were either completely or partially separated under the optimized mobile phase conditions
with resolutions of 3.98, 2.52, 2.02, 2.14, 1.80 and 0.41, respectively. The influences of organic modifier
types and content, and base/acid additives on enantiomeric separation were evaluated and optimized.
Furthermore, a simulation study was also used to explain the chiral recognition mechanisms of this class
of drug enantiomers on Chiralpak ID for the first time. The modeling data were in agreement with the
chromatographic results concerning enantioselectivity.
Received 8th February 2018,
Accepted 4th June 2018
DOI: 10.1039/c8nj00685g
rsc.li/njc
As the previous article reported, it was confirmed that the two
enantiomers of atropine possess different pharmacological
1. Introduction
The enantiomeric separation of chiral drugs has become increasingly activities.10 Additionally, it was found that the (+)- and (ꢀ)-isomers
important due to the different pharmacodynamic and/or pharmaco- of benzhexol differed in the affinity of acetylcholine receptors.11
kinetic properties of the two enantiomers.1–4 The diverse stereo- Therefore, keeping in mind these facts, the enantiomeric
selectivity of an individual enantiomer in vivo might cause a range of separation of this class of drug enantiomers was very important
side effects to the human body. Hence, in order to improve the safety and necessary.
of clinical medication and provide convenient research into the
At present there are only a few papers that have studied the
biological properties of each enantiomer, the development of chiral enantiomeric separation of some anticholinergic drugs by
analytical methods is an urgent need within the pharmaceutical capillary electrophoresis (CE) and HPLC.12–16 Therein, atropine
industry.
and benzhexol were completely separated with different b-cyclo-
Up until now high performance liquid chromatography (HPLC) dextrin derivatives as chiral additives by CE. Furthermore, the
using chiral stationary phases (CSPs) was considered as the best enantiomers of oxybutynin were baseline resolved on two coated
technique for analysis and separation of drug enantiomers. Among amylose-based columns (AmyCoat and Chiralpak AD) with
the CSPs employed in HPLC, polysaccharide-based CSPs, especially n-hexane–2-propanol–DEA (80 : 20 : 0.1, v/v) as the mobile phase.
amylose derivatives, as chiral selectors were the most successful Beyond that, however, almost no paper has been concerned with
in terms of enantiomeric separation. They possessed advan- the enantiomeric separation of the anticholinergic drugs using
tages in CSPs such as robustness and exhibited a wide range of immobilized amylose-based CSPs.
applications.5–7 However, the chiral recognition mechanisms of
these CSPs were not yet completely elucidated.8
In recent years, in order to determine the chiral recognition
mechanisms of the CSPs at the molecular level, several com-
Anticholinergic drugs, usually known as the cholinoceptor putational simulations focusing on the enantiomeric separation,
antagonists, are mainly used in clinics for spasmolysis, mydriasis particularly by molecular docking, have been developed.17–21 The
and dilated bronchi.9 Most of this class of drug, like atropine, have modeling technique could gain better insight into the binding
been commonly formulated and administered as racemates. energy between each enantiomer and the CSP, and clarify the types
of intermolecular interactions. Therefore, it was efficient and
a School of Pharmacy, Shenyang Pharmaceutical University, Shenyang,
accurate to explain the chiral recognition process. For instance,
Liaoning Province, 110016, P. R. China. E-mail: gxjhyz@gmail.com
b School of Pharmaceutical Engineering, Shenyang Pharmaceutical University,
I. Ali et al.16 applied simulation studies to investigate the
enantioseparation of four chiral analytes on the AmyCoat chiral
selector. The significant difference in the binding affinity and
Shenyang, Liaoning Province, 110016, P. R. China.
E-mail: jianwang@syphu.edu.cn
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
Fig. 1 The chemical structures of eight analytes.
the specific interactions for separation were found. M. F. Alajmi 2.2 Chromatographic conditions
et al.22 studied the separation mechanisms of DL-leucine-DL-
The experiments were performed on a Shimadzu LC-10A HPLC
system (Shimadzu, Japan) equipped with an LC-10AT pump
and an SPD-10A UV-vis Detector. The detection signals and data
were collected and processed by the Sepu3000 software (Hangzhou,
China). The stationary phase was the commercially available
Chiralpak ID (250 mm ꢁ 4.6 mm, 5 mm) column purchased
from Daicel Chiral Technologies (Shanghai, China). The mobile
phase systems consisted of n-hexane–ethanol/2-propanol/1-propanol.
The prepared mobile phase was filtered and de-gassed before use.
The column was maintained at 25 1C, and the injection volume was
20 mL. The flow rate was 1.0 mL minꢀ1 and the detection wavelength
was set at 230 nm.
tryptophan dipeptide on a tris-(3,5-dimethylphenyl carbamate)
amylose chiral selector using a modeling method. However, a
thorough search of the literature confirmed that a simulation
study for the chiral separation of anticholinergic drugs with
immobilized amylose-based CSPs has not been reported until now.
Herein, we explored for the first time an efficient HPLC analytical
method for the enantiomeric separation of eight analytes including
atropine sulfate, phencynonate, dipivefrine hydrochloride, tropic-
amide, homatropine methylbromide, oxybutynin, scopolamine
hydrobromide and benzhexol hydrochloride (Fig. 1). The tested
column was Chiralpak ID (Fig. 2), which was a commercially
immobilized polysaccharide-based CSP, namely, amylose tris-
(3-chlorophenylcarbamate). In addition, the influences of alcohol-
modifying agents and basic/acidic additives on enantiomeric
separation were evaluated and optimized in detail. Furthermore,
molecular docking using AutoDock was carried out to ascertain the
possible chiral recognition mechanisms of Chiralpak ID for the
first time. The results of these experiments are given herein.
2.3 Preparation of sample solutions
All eight analytes studied were dissolved in ethanol to prepare
stock solutions of 1 mg mLꢀ1, and were then diluted to a
suitable concentration, respectively. All solutions were filtered
through an organic nylon membrane of 0.45 mm pore size.
2.4 Calculations
The chromatographic parameters of retention factor (k), separation
factor (a) and resolution (Rs) were calculated as follows:
k = (tR ꢀ t0)/t0, where tR and t0 represented the retention times
of the analytes and the unretained solutes, respectively; a = k2/k1,
where k1 and k2 were the retention factors of the first and second
eluted enantiomers; Rs = 2(t2 ꢀ t1)/(W1 + W2), where t1 and t2
were the retention times of the successively eluted enantiomers,
and W1 and W2 were the peak widths of the first and second
eluted enantiomers, respectively.
2. Experimental
2.1 Chemicals and reagents
The standard substances of atropine sulfate, phencynonate,
dipivefrine hydrochloride, tropicamide, homatropine methyl-
bromide, oxybutynin, scopolamine hydrobromide and benzhexol
hydrochloride were obtained from the Chinese Food and Drug
Inspection Institute (Beijing, China). HPLC grade n-hexane, ethanol,
2-propanol and 1-propanol were purchased from Concord
Technology (Tianjin, China). Formic acid (FA) and diethylamine
2.5 Simulation study
(DEA) were of analytical grade and supplied from Shandong The system of Intels Xeon E5-2670 with Red hat 6.4 was used for
Yuwang Industrial (Shandong, China).
molecular docking of the eight drug enantiomers. The Marwin
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
Fig. 2 The 2D (A) and 3D (B) structures of the amylose tris-(3-chlorophenylcarbamate) chiral selector (Chiralpak ID).
Sketch software was used to draw the structures of the studied previous study, only tropicamide was baseline resolved under
enantiomers and Chiralpak ID. The structures were transferred the n-hexane–alcohol systems without the addition of DEA or
to 3D and saved in mol2 file format. Sybyl 6.9.1 based on the DEA and FA. As the eight analytes were basic, a small amount of
molecular mechanics minimization was used to optimize the base (0.1% DEA) was added to the mobile phase to reduce the
structures. After that, using AutoDock 4.2, the enantiomers and peak tail and to obtain a good peak shape. The results revealed
the CSP were prepared by assigning Gastegier charges, merging that the enantiomeric separation of some analytes, such as
nonpolar hydrogen atoms and saving in PDBQT file format. The atropine sulfate, homatropine methylbromide and tropicamide,
automated molecular docking was done with AutoDock 4.2, and was significantly improved. To optimize further the chiral HPLC
the grid box size of 120 ꢁ 120 ꢁ 120 points with 0.75 Å spacing was conditions, a mixture of 0.1% DEA and 0.1% FA was used to
applied. One hundred independent docking runs were used for study the chiral separation of the eight analytes. It was shown
each enantiomer and the CSP for the lowest free energy of the that a total of four analytes including atropine sulfate, phencyno-
fastening conformation from the largest cluster, which was saved nate, dipivefrine hydrochloride and tropicamide were completely
in PDBQT file format. Finally, Discovery Studio Visualizer was used separated with an Rs larger than 1.5, and one analyte, i.e.
for molecular display.
oxybutynin, was partially separated. However, it was interesting
to note that the enantioselectivity of homatropine methyl-
bromide was not observed, indicating that different acidic and
basic additives might result in different chiral separation.
To some extent, the ability of the enantiomeric separation of
2-propanol and 1-propanol was better than that of ethanol.
3. Results and discussion
3.1 Chromatographic separation of eight drug enantiomers
As a general rule, Chiralpak ID is mainly used to resolve drug Nevertheless, the exception was atropine sulfate which achieved
enantiomers under normal phase mode. Thus, the enantiomeric the largest resolution using ethanol as the alcohol modifier.
separation of eight analytes was evaluated with the alcohol- From Table 1, we concluded that five of the eight analytes were
modifying agents ethanol, 2-propanol and 1-propanol. In our baseline resolved, one was partially separated and two were not
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
Table 1 Enantiomeric separation of the six analytes on Chiralpak ID under optimized conditions
Analytes
k1
a
Rs
Mobile phase
Atropine sulfate
Phencynonate
Dipivefrine hydrochloride
Tropicamide
Homatropine methylbromide
Oxybutynin
2.82
1.75
2.48
1.59
0.95
1.59
1.35
1.21
1.20
1.15
1.14
1.10
3.98
2.52
2.02
2.14
1.80
0.41
n-Hexane–ethanol–DEA–FA (60 : 40 : 0.1 : 0.1, v/v/v/v)
n-Hexane–2-propanol–DEA–FA (75 : 25 : 0.1 : 0.1, v/v/v/v)
n-Hexane–2-propanol–DEA–FA (75 : 25 : 0.1 : 0.1, v/v/v/v)
n-Hexane–ethanol–DEA–FA (60 : 40 : 0.1 : 0.1, v/v/v/v)
n-Hexane–1-propanol–DEA (85 : 15 : 0.1, v/v/v)
n-Hexane–2-propanol–DEA–FA (80 : 20 : 0.1 : 0.1, v/v/v/v)
Flow rate: 1.0 mL minꢀ1; column temperature: 25 1C; detection wavelength: 230 nm; injection volume: 20 mL.
separated under the optimized chromatographic conditions.
On the basis of the above studies, we selected the optimal
The k1 values of atropine sulfate, phencynonate, dipivefrine mobile phase composition of each analyte to study the effect of
hydrochloride, tropicamide, homatropine methylbromide and alcohol content on enantioseparation. For the six separated
oxybutynin were 2.82, 1.75, 2.48, 1.59, 0.95 and 1.59, respectively. drugs, we found that the k and Rs values increased with a
The values of a and Rs were 1.35, 1.21, 1.20, 1.15, 1.14 and decrease in alcohol content in the mobile phase, indicating that the
1.10, and 3.98, 2.52, 2.02, 2.14, 1.80 and 0.41, respectively. The chiral separation was strongly influenced by the polarity of the
chiral separation of the above six analytes was acceptable because mobile phase. However, the a value remained basically unchanged.
the magnitudes of a and Rs were larger than 1.0 and 1.5,
respectively.
The flow rate, which varied from 0.6 to 1.0 mL minꢀ1, was
also evaluated and it was observed that a decrease in the flow
rate resulted in a prolonged retention time and better separation.
However, the longer analysis times could result in a larger
occurrence rate of wide peaks and low column efficiency. Thereby,
1.0 mL minꢀ1 was chosen as the best flow rate. Moreover, the
column temperature (25–40 1C) was investigated to optimize the
chromatographic conditions. The conclusion was that a higher
column temperature harmed the chiral separation of the studied
analytes.
As a result of the experiments, optimized enantiomeric
separation conditions of eight drug enantiomers were confirmed
and reported (Table 1), and the corresponding chromatograms
are shown in Fig. 3.
3.2 Simulation study of eight drug enantiomers with
Chiralpak ID
Fig. 3 Typical chromatograms of chiral separation of atropine sulfate,
phencynonate, dipivefrine hydrochloride, tropicamide, homatropine
methylbromide and oxybutynin on Chiralpak ID under the optimal chromato-
graphic conditions.
Molecular docking of atropine sulfate, phencynonate, dipivefrine
hydrochloride, tropicamide, homatropine methylbromide, oxy-
butynin, scopolamine hydrobromide and benzhexol hydrochloride
Table 2 The modeling results of the enantiomers of the eight studied analytes with Chiralpak ID
Binding affinity/energy
Number of
hydrogen bonds
Number of
hydrophobic interactions
(kcal molꢀ1
)
DER–S
a
Analytes
Enantiomers
Atropine sulfate
R
S
R
S
R
S
R
S
R
S
R
S
R
S
R
S
ꢀ7.68
ꢀ7.10
ꢀ7.47
ꢀ7.73
ꢀ6.71
ꢀ6.93
ꢀ6.49
ꢀ6.73
ꢀ6.92
ꢀ7.05
ꢀ6.76
ꢀ6.81
ꢀ6.42
ꢀ6.39
ꢀ7.73
ꢀ7.72
2
1
2
1
6
3
2
2
2
5
2
2
1
3
0
0
1
2
1
4
2
3
2
5
2
1
2
1
2
2
2
6
ꢀ0.58
0.26
Phencynonate
Dipivefrine hydrochloride
Tropicamide
0.22
0.24
Homatropine methylbromide
Oxybutynin
0.13
0.05
Scopolamine hydrobromide
Benzhexol hydrochloride
ꢀ0.03
ꢀ0.01
a
Difference in the binding energies of enantiomers.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
with amylose tris-(3-chlorophenylcarbamate) (Chiralpak ID) was in the binding energies of the (R)- and (S)-enantiomers with CSP
carried out to investigate the chiral recognition mechanisms. The (DER–S) was calculated. If the absolute value of DER–S was larger,
modeling results are outlined below in detail.
it would contribute more to the difference in retention time of
3.2.1 Molecular docking calculations. The binding energies the two enantiomers. Thus, the enantiomeric separation might
for the best scoring, i.e. the lowest binding affinity, conformation be easier to achieve. From Table 2 it could be seen that the
of the (R)- and (S)-enantiomers with CSP are shown in Table 2. DER–S values of the eight analytes mentioned above were ꢀ0.58,
The different values of the binding energies of the drug enantiomers 0.26, 0.22, 0.24, 0.13, 0.05, ꢀ0.03 and ꢀ0.01 kcal molꢀ1
,
with CSP depended on the various stereochemical structures of the respectively. Among them, the atropine sulfate enantiomers
drugs, which were maintained by different types and numbers showed the biggest energy difference (DER–S = ꢀ0.58 kcal molꢀ1),
of intermolecular interaction forces. Meanwhile, the results also whereas the DER–S values of the scopolamine hydrobromide
implied that the stability of the enantiomer in combination with (ꢀ0.03 kcal molꢀ1) and benzhexol hydrochloride (ꢀ0.01 kcal molꢀ1
)
CSP was different. The more negative the binding energy, the enantiomers were much smaller than those of any other analytes.
stronger the interaction was that was established between the Therefore, it was hypothesized that atropine would have the larger
enantiomers and CSP.23
resolution. On the other hand, the smaller resolution would be
In order to further elaborate on the relationship between the acquired for the scopolamine hydrobromide and benzhexol
enantiomeric separation and the binding energy, the difference hydrochloride. According to the previous enantiomeric separation
Fig. 4 The 3D docking positions of the two enantiomers of the studied analytes with CSP: (A) atropine sulfate, (B) phencynonate, (C) dipivefrine
hydrochloride, (D) tropicamide, (E) homatropine methylbromide, (F) oxybutynin, (G) scopolamine hydrobromide and (H) benzhexol hydrochloride. Purple
represents the (R)-enantiomer of the analytes, green represents the (S)-enantiomer and the green dotted lines represent the hydrogen bonds.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
results, atropine sulfate was completely separated with the largest and the hydrogen of the amino group in atropine and the
Rs of 3.98 and no chiral separation for scopolamine hydrobromide oxygen of the CQO group in CSP, and the hydrogen of the –OH
and benzhexol hydrochloride was found on Chiralpak ID. These group of atropine and the oxygen of the –O– group in CSP,
results were consistent with our expectation.
respectively. For scopolamine hydrobromide, a hydrogen bond
3.2.2 The hydrogen bond interaction. The three-dimensional for the (R)-enantiomer was generated by the hydrogen of the
docking conformations of the eight drug enantiomers interacting amine group in scopolamine and the oxygen of the CQO group
with CSP are shown in Fig. 4. It is clear that the drug enantiomers in CSP, while for the (S)-enantiomer, the hydrogen of the –OH
are stable at different positions in the chiral grooves of CSP through group in scopolamine and the oxygen atoms of two different
hydrogen bonds and other interactions. From Table 3 and Fig. 4, it –O– groups in CSP and the oxygen of the CQO group in
can be seen that the structures of carbonyl (CQO), hydroxyl (–OH) scopolamine and hydrogen of the –NH group in CSP formed
and amino groups near the chiral carbon atom of the drug three hydrogen bonds. Additionally, the CQO groups in the
enantiomers could form stronger hydrogen bonds with the oxybutynin, phencynonate and dipivefrine hydrochloride molecules
amino, CQO and ether (–O–) groups of CSP. These interactions did not form hydrogen bonds with CSP. We inferred that the larger
were found between the polar atoms and were called conventional or longer side chains near the carbonyl might affect the access to
hydrogen bonds.
the chiral grooves of CSP, resulting in a greater distance to the
For the eight chiral drugs, the type and number of hydrogen amine groups of the CSP (Fig. 4). For benzhexol hydrochloride, a
bonds were significantly different due to the difference in the hydrogen bond was not found and its enantiomeric separation was
stereochemical structure of the drug enantiomers. For example, not obtained at the same time. These results were enough to
the CQO and –OH groups of three structurally similar analytes show that the hydrogen bond played an important role in chiral
including atropine sulfate, homatropine methylbromide and separation of this set of drug enantiomers.
scopolamine hydrobromide formed hydrogen bonds with the
3.2.3 The hydrophobic interaction. The hydrophobic inter-
amino and –O– groups of CSP, respectively. However, the action can also be observed from Table 4. Three kinds of
number and type of hydrogen bonds between the (R)- or (S)- hydrophobic interaction including pi hydrophobic, alkyl hydro-
enantiomers and CSP were clearly different. Taking atropine phobic and mixed pi–alkyl hydrophobic were observed and
sulfate and scopolamine hydrobromide as examples, it was were present in different enantiomers of the studied analytes
found that for (R)- and (S)-atropine sulfate, two and one and CSP. It can be seen that the distance of all hydrophobic
hydrogen bonds were formed between the oxygen of the CQO interactions was larger than 3.65 Å, indicating that the hydro-
group in atropine and the hydrogen of the –NH– group in CSP phobic interaction of the enantiomers and CSP was weaker to
Table 3 The hydrogen bonds of the studied analytes with CSP
Analyte
Enantiomer
Name
Type
Distance (Å)
Atropine sulfate
R
S
:h1i:H319–d:RES1:O4, d:RES1:H42–:h1i:O165a
d:RES1:H32–:h1i:O103
Conventional hydrogen bond
Conventional hydrogen bond
1.61, 2.49
1.91
Phencynonate
R
S
:h1i:H267–d:RES1:O14, d:RES1:H58–:h1i:O103
d:RES1:H58–:h1i:O90
Conventional hydrogen bond
Conventional hydrogen bond
2.85, 1.80
2.03
Dipivefrine
hydrochloride
R
:h1i:H267–d:RES1:O3, d:RES1:H30–:h1i:O103,
d:RES1:H31–:h1i:O103, d:RES1:H31–:h1i:O114,
d:RES1:H29–:h1i:O90, d:RES1:H29–:h1i:O125
:h1i:H267–d:RES1:O3, d:RES1:H30–:h1i:O125,
d:RES1:H29–:h1i:O103
Conventional hydrogen bond
2.55, 2.03, 2.19,
2.17, 2.14, 2.59
S
Conventional hydrogen bond
1.95, 1.97, 2.15
Tropicamide
R
S
d:RES1:H26–:h1i:O72, d:RES1:H22–:h1i:O42
:h1i:H267–d:RES1:O21, d:RES1:H41–:h1i:O154
Conventional hydrogen bond
Conventional hydrogen bond
1.94, 2.42
2.21, 1.92
Homatropine
methylbromide
R
S
:h1i:H267–d:RES1:O19, d:RES1:H42–:h1i:O103
:h1i:H267–d:RES1:O20, :h1i:H294–d:RES1:O19,
d:RES1:H42–:h1i:O90, d:RES1:H42–:h1i:O103,
d:RES1:H42–:h1i:O125
Conventional hydrogen bond
Conventional hydrogen bond
2.68, 2.17
2.65, 3.09, 1.94,
2.48, 2.56
Oxybutynin
R
S
:h1i:H289–d:RES1:O11, d:RES1:H38–:h1i:O165
:h1i:H267–d:RES1:O9, d:RES1:H38–:h1i:O125
Conventional hydrogen bond
Conventional hydrogen bond
1.78, 2.86
2.47, 2.13
Scopolamine
hydrobromide
R
S
d:RES1:H41–:h1i:O137
Conventional hydrogen bond
Conventional hydrogen bond
1.85
2.37, 2.10, 2.91
:h1i:H267–d:RES1:O4, d:RES1:H33–:h1i:O103,
d:RES1:H33–:h1i:O114
Benzhexol
hydrochloride
R
S
—
—
—
—
—
—
a
d:RES1 represents the drug enantiomers, and :h1i represents CSP.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018

View Article Online
NJC
Paper
Table 4 The hydrophobic interaction of the studied analytes with CSP
Analytes
Enantiomers Name
Types
Distance (Å)
Atropine sulfate
R
S
d:RES1–:h1i:Cl71a
Pi–alkyl
Pi–alkyl, pi–alkyl
5.07
4.14, 5.32
d:RES1–:h1i:Cl29, d:RES1–:h1i:Cl61
Phencynonate
R
S
d:RES1–:h1i:Cl29
Pi–alkyl
4.38
5.14, 4.98,
4.04, 5.19
:h1i:Cl112–d:RES1, :h1i–d:RES1, d:RES1–:h1i:Cl29,
d:RES1–:h1i:Cl61
Alkyl, pi–alkyl, pi–alkyl, pi–alkyl
Dipivefrine
hydrochloride
R
S
d:RES1:C17–:h1i, d:RES1–:h1i:Cl29
Pi–sigma, pi–alkyl
Pi–sigma, pi–pi T-shaped, pi–alkyl 3.74, 5.96, 5.19
3.65, 4.46
d:RES1:C16–:h1i, :h1i–d:RES1, d:RES1–:h1i:Cl29
Tropicamide
R
S
d:RES1–:h1i:Cl164, d:RES1–:h1i
Pi–alkyl, pi–alkyl
Pi–sigma, pi–pi stacked,
pi–pi T-shaped, pi–alkyl, pi–alkyl
4.20, 5.42
3.90, 4.57, 5.81,
4.45, 5.19
d:RES1:C18–:h1i, :h1i–d:RES1, :h1i–d:RES1,
d:RES1–:h1i:Cl29, d:RES1–:h1i:Cl61
Homatropine
methylbromide
R
S
d:RES1–:h1i:Cl29, d:RES1–:h1i:Cl61
d:RES1–:h1i:Cl61
Pi–alkyl, pi–alkyl
Pi–alkyl
4.30, 5.22
4.65
Oxybutynin
R
S
d:RES1–:h1i:Cl71, d:RES1–:h1i:Cl102
d:RES1–:h1i:Cl29
Pi–alkyl, pi–alkyl
Pi–alkyl
4.40, 4.72
4.76
Scopolamine
hydrobromide
R
S
:h1i–d:RES1, d:RES1–:h1i:Cl206
Pi–pi stacked, pi–alkyl
Pi–alkyl, pi–alkyl
3.88, 4.41
4.44, 5.36
d:RES1–:h1i:Cl29, d:RES1–:h1i:Cl61
Benzhexol
hydrochloride
R
S
:h1i:Cl29–d:RES1, :h1i–d:RES1:C8
Alkyl, pi–alkyl
4.73, 4.73
4.05, 4.26, 5.29,
5.10, 5.09, 4.85
:h1i:Cl19–d:RES1:C8, :h1i–d:RES1:C8, :h1i–d:RES1:C8, Alkyl, pi–alkyl, pi–alkyl, pi–alkyl,
:h1i–d:RES1:C8, :h1i–d:RES1, d:RES1–:h1i:Cl71
pi–alkyl, pi–alkyl
a
d:RES1 represents the drug enantiomers, and :h1i represents CSP.
some extent. The type and number of hydrophobic interactions were evaluated and optimized by changing the mobile phase as
were related to the structure of the drug enantiomers. The pi–alkyl well as the base and acid additives. Under the optimized
interactions of atropine sulfate, homatropine methylbromide, conditions, five out of eight of the drug enantiomers were
scopolamine hydrobromide and tropicamide were mainly baseline resolved, and one was partially separated. Additionally,
formed by the enantiomers with Pi orbitals and CSP with Cl a simulation study was first used to illustrate the chiral recognition
atoms. Interestingly, for (R)- and (S)-tropicamide, the pyridine mechanisms, and the docking results were in accordance with the
ring in the molecule generated additional pi–alkyl and pi–pi chromatographic parameters regarding enantioselectivity. The
hydrophobic interactions with CSP, respectively. A number of hydrogen bonds and hydrophobic interactions played a crucial
different pi–alkyl interactions were developed between the role for the enantiomeric separation.
structures of cyclopentane in the phencynonate molecule and
the piperidine ring in the benzhexol hydrochloride molecule
with the phenyl moiety of CSP, respectively. Moreover, due to
Conflicts of interest
the presence of methyl in dipivefrine hydrochloride, a pi–sigma
hydrophobic interaction was found.
There are no conflicts to declare.
From the modeling results, except for the stronger hydrogen
bond interaction, it was important to note that the hydrophobic
interaction was also key for the chiral separation. A large
References
1 J. W. Oh, T. Q. Trung, K. S. Sin, J. S. Kang and K. H. Kim,
Chirality, 2007, 19, 528–535.
2 S. Yuan and Q. L. Zeng, Chromatographia, 2017, 80, 1171–1177.
3 F. Jamali, R. Mehvar and F. M. Pasutto, J. Sep. Sci., 1989,
78(9), 695–715.
4 C. Alvarez, J. A. Sanchez-Brunete, S. Torrado-Santiago,
R. Cadorniga and J. J. Torrado, Chromatographia, 2000,
52(7–8), 455–458.
amount of pi–alkyl (–Cl) hydrophobic interactions between the
enantiomers and CSP clearly demonstrated that the substituents
on the benzene ring of CSP were crucial for enantiomeric
separation. Therefore, in light of these facts, we have sufficient
reason to think that the interactions mentioned above make a
significant contribution to the enantiomeric separation of the
studied analytes.
5 E. Yashima, J. Chromatogr. A, 2001, 906, 105–125.
6 S. Materazzo, S. Carradori, R. Ferretti, B. Gallinella, D. Secci
and R. Cirilli, J. Chromatogr. A, 2014, 1327, 73–79.
4. Conclusion
´ ´ ´
7 R. Geryk, K. Kalıkova, J. Vozka, D. Plecita, M. G. Schmid and
In this paper, the enantiomeric separation of eight anticholinergic
drugs on a polysaccharide-based CSP was first investigated in
normal phase mode. The conditions of enantiomeric separation
´
E. Tesarova, J. Chromatogr. A, 2014, 1363, 155–161.
8 R. W. Stringham, Adv. Chromatogr., 2007, 38, 257–290.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018
New J. Chem.

View Article Online
Paper
NJC
9 Q. D. You, Chem. Ind. Press, 2008.
10 L. Mateus, S. Cherkaoui, P. Christen and J. L. Veuthey,
J. Chromatogr. A, 2000, 868, 285–294.
17 W. H. Li, C. H. Liu, G. G. Tan, X. R. Zhang, Z. Y. Zhu and
Y. F. Chai, Int. J. Mol. Sci., 2012, 13, 710–725.
18 C. Fernandes, M. E. Tiritan, Q. Cass, V. Kairys, M. X. Fernandes
and M. Pinto, J. Chromatogr. A, 2012, 1241, 60–68.
11 R. B. Barlow, J. Pharm. Pharmacol., 1971, 23(2), 90–97.
12 G. H. Du, S. Y. Lu, L. H. Liu, J. W. Xie, B. H. Zhong and 19 I. Ali, M. N. Lone, M. Suhail, Z. A. AL-Othmanb and
K. L. Liu, Chin. J. Pharm. Anal., 2007, 27(9), 1395–1398. A. Alwarthan, RSC Adv., 2016, 6(17), 14372–14380.
13 X. L. Lin, G. B. Li, W. Q. Jiang, Y. B. Chu, P. Wu and 20 C. Fernandes, A. Palmeira, A. Santos, M. E. Tiritan,
Y. F. Guan, Anal. Chim. Acta, 2001, 431, 41–48. C. Afonso and M. M. Pinto, Chirality, 2013, 25, 89–100.
14 M. M. Hefnawy and J. T. Stewart, Anal. Lett., 1998, 31(4), 659–667. 21 H. Y. Aboul-Enein, I. Ali, M. Laguerre and G. Felix, J. Liq.
15 S. I. Tahara, A. Okayama, Y. Kitada, T. Watanabe, H. Nakazawa, Chromatogr. Relat. Technol., 2002, 25, 2739–2748.
K. Kakehi and Y. Hisamatu, J. Chromatogr. A, 1999, 848, 22 M. F. Alajmi, A. Hussain, M. Suhail, S. D. Mukhtar, D. R. Sahoo,
465–471. L. Asnin and I. Ali, Chirality, 2016, 28(9), 642–648.
16 I. Ali, Z. A. Al-Othman, A. Al-Warthan, S. D. Alam and 23 Y. Q. Yao, P. L. Song, X. L. Wen, M. D. Deng, J. Wang and
J. A. Farooqi, Chirality, 2014, 26, 136–143. X. J. Guo, J. Sep. Sci., 2017, 40, 2999–3007.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2018