A chiral enantioseparation generic strategy for anti-Alzheimer and antifungal drugs by short end injection capillary electrophoresis using an experimental design approach

Article abstract of DOI:10.1002/chir.22777

The present study describes a generic strategy using capillary electrophoretic (CE) method for chiral enantioseparation of anti-Alzheimer drugs, namely, donepezil (DON), rivastigmine (RIV), and antifungal drugs, namely, ketoconazole (KET), Itraconazole (ITR), fluconazole (FLU), and sertaconazole (SRT) in which these drugs have different basic and acidic properties. Several modified cyclodextrins (CDs) were applied for enantioseparation of racemates such as highly sulfated α, γ CDs, hydroxyl propyl-β-CD, and Sulfobutyl ether-β-CD. The starting screening conditions consist of 50-mM phosphate-triethanolamine buffer at pH?2.5, an applied voltage of 15?kV, and a temperature of 25°C. The CE strategy implemented in the separation starts by screening prior to the optimization stage in which an experimental design is applied. The design of experiment (DOE) was based on a full factorial design of the crucial two factors (pH and %CD) at three levels, to make a total of nine (3²) experiments with high, intermediate, and low values for both factors. Evaluation of the proposed strategy pointed out that best resolution was obtained at pH?2.5 for five racemates using low percentages of HS-γ-CD, while SBE-β-CD was the most successful chiral selector offering acceptable resolution for all the six racemates, with the best separation at low pH values and at higher %CD within 10-min runtime. Regression study showed that the linear model shows a significant lack of fit for all chiral selectors, anticipating that higher orders of the factors are most likely to be present in the equation with possible interactions.

Full text of DOI:10.1002/chir.22777

Received: 19 July 2017
DOI: 10.1002/chir.22777
Revised: 18 September 2017
Accepted: 22 September 2017
R E G U L A R A R T I C L E
A chiral enantioseparation generic strategy for anti‐
Alzheimer and antifungal drugs by short end injection
capillary electrophoresis using an experimental design
approach
Ahmed M. Abdel‐Megied1
| Rasha S. Hanafi2
| Hassan Y. Aboul‐Enein3
1 Pharmaceutical Analytical Chemistry
Department, Faculty of Pharmacy and
Pharmaceutical Manufacturing,
Kafrelsheikh University, Kafrelsheikh
City, Egypt
2 Pharmaceutical Chemistry Department,
Faculty of Pharmacy and Biotechnology,
German University in Cairo—GUC, Cairo,
Egypt
3 Pharmaceutical and Medicinal Chemistry
Department, Pharmaceutical and Drug
Industries Research Division, National
Research Center, Giza, Egypt
Abstract
The present study describes a generic strategy using capillary electrophoretic
(CE) method for chiral enantioseparation of anti‐Alzheimer drugs, namely,
donepezil (DON), rivastigmine (RIV), and antifungal drugs, namely, ketocona-
zole (KET), Itraconazole (ITR), fluconazole (FLU), and sertaconazole (SRT) in
which these drugs have different basic and acidic properties. Several modified
cyclodextrins (CDs) were applied for enantioseparation of racemates such as
highly sulfated α, γ CDs, hydroxyl propyl‐β‐CD, and Sulfobutyl ether‐β‐CD.
The starting screening conditions consist of 50‐mM phosphate‐triethanolamine
buffer at pH 2.5, an applied voltage of 15 kV, and a temperature of 25°C. The
CE strategy implemented in the separation starts by screening prior to the opti-
mization stage in which an experimental design is applied. The design of exper-
iment (DOE) was based on a full factorial design of the crucial two factors (pH
and %CD) at three levels, to make a total of nine (32) experiments with high,
intermediate, and low values for both factors. Evaluation of the proposed strat-
egy pointed out that best resolution was obtained at pH 2.5 for five racemates
using low percentages of HS‐γ‐CD, while SBE‐β‐CD was the most successful
chiral selector offering acceptable resolution for all the six racemates, with
the best separation at low pH values and at higher %CD within 10‐min runtime.
Regression study showed that the linear model shows a significant lack of fit for
all chiral selectors, anticipating that higher orders of the factors are most likely
to be present in the equation with possible interactions.
Correspondence
Hassan Y. Aboul‐Enein, Pharmaceutical
and Medicinal Chemistry Department,
Pharmaceutical and Drug Industries
Research Division, National Research
Center, Dokki, Giza 12622, Egypt.
Email: haboulenein@yahoo.com
Funding information
Faculteit Geneeskunde en Farmacie,
Farmaceutisch instituut, Dienst
Analytische Scheikunde en,
Farmaceutische Technologie (FABI)
Laarbeeklaan 103, B‐1090) Brussels,
Belgium.
KEYWORDS
azole derivatives, donepezil, enantiomers, experimental design, highly sulphated α,γ‐cyclodextrins,
rivastigmine
Abbreviations: BGE, Background electrolyte; CDs, Cyclodextrins; EOF, Electroosmotic flow; HS‐CD, Highly sulfated Cyclodextrins; SBE‐ β‐CD,
Sulfobutyl Ether β‐ Cyclodextrin; TEA, Triethanolamine
Chirality. 2017;1–12.
wileyonlinelibrary.com/journal/chir
© 2017 Wiley Periodicals, Inc.
1

2
ABDEL‐MEGIED ET AL.
inhibitors of cytochrome P450 (CYP) of the fungus.6 They
have a broad and potent antifungal effect, but most of the
azole antifungal drugs have side effects such as KET which
limits its clinical applications due to the hepatic
toxicity.7 These azoles are chiral drugs although they are
employed clinically as a racemic mixture 1:1 mixture of
the dioxolane ring in cis configuration as shown in Table 1.
In the last years, special emphasis has been placed on
the synthesis of enantiomerically pure compounds,
particularly in the pharmaceutical industry enantiomers.8
Enantioseparation of racemates is considered one of the
attractive topics in the field of pharmaceutical analysis
as optical purity control needs rapid and highly efficient
analytical methods.9 The FDA stated its policy statement
for the method development and determination of chiral
drugs: The stereoisomeric of a drug with a chiral center
should be known as well as the composition of the
material used in pharmacologic, toxicological and clinical
studies.10 DON enantiomers have shown different degrees
of inhibition against acetylcholine in vivo, S‐DON having
a 2.2‐fold higher affinity than R‐DON while both RIV
enantiomers have the same inhibitory activity.11,12 S‐KET
was 2‐4 fold more potent inhibitor of CYP450 than R‐KET
and therefore has stronger antifungal activity in vivo
studies.13 ITR has three chiral centers that give rise to eight
1 | INTRODUCTION
Alzheimer disease is
a neurodegenerative disorder
characterized by progressive loss of memory followed by
complete dementia which characterized by loss of memory
and progressive deficits in different cognitive domains.1
Rivastigmine (RIV) [(S)‐N‐ethyl‐3‐[(1‐dimethyl‐amino)
ethyl]‐N‐methyl‐phenyl‐carbamate hydrogen tartrate] is
a carbamate derivative of physostigmine which is able to
react covalently with the active site of the enzyme.2 The
chemical structures of the drugs under study are shown in
Table 1. Although its exact mechanism of action is still
unknown, it is thought to exert its therapeutic effect by a
reversible and noncompetitive acetylcholine esterase
inhibitor, selectively inhibiting its activity in certain brain
areas, also enhance choline acetyl transferase activity that
stimulates acetylcholine synthesis.3,4 A structural feature
of DON is the presence of a chiral center adjacent to a
carbonyl group which makes each enantiomer easily
racemizable via keto‐enol intermediate.5 On the other hand,
the antifungal drugs, azole compounds have been used
for nearly 40 years. The compounds under investigation in
this work are ketoconazole (KET), itraconazole (ITR),
fluconazole (FLU), and sertaconazole (SRT); these azoles
affect the synthesis of ergosterol because they are potent
TABLE 1 Chemical structures for drugs under investigation
Acidic drugs
Basic drugs

ABDEL‐MEGIED ET AL.
3
stereoisomers. It was demonstrated that 2S, 4R, 2S‐ITR
and 2S, 4R, 2R‐ITR possess the greatest antiangiogenic
and antifungal activity.14 Obviously, it was proven in some
cases that the therapeutic effect is limited to one of the
enantiomers. Furthermore, the pharmacologically inactive
eutomer may cause undesirable side effects, in some cases,
antagonist, and even toxic effect.15 The method development
strategy of enantiseparation in capillary electrophoresis (CE)
is considered a time‐consuming task, since finding the
appropriate chiral selector is “trial and error” process and
usually performed with a one‐variable approach, in which
the analyst examines the variables that can separate one
from one influence while the other factors are constantly
executed.16 It is very hard to predict the selectivity of a chiral
selector towards a certain enantiomer. So, the affinity of all
selectors has to be examined one at a time. Nowadays, the
design of experiment (DOE) is not considered fancy or
complex to apprehend anymore. Many software tools and
guided approaches are available to allow the multivariate
approach which involves the simultaneous change of
several variables. CE is suitable for the use of experimental
design in which the experimental conditions could be varied
from one experiment to another. Up to date, there are only a
few studies that report the use of experimental designs in
the field of chiral enantioseparation in CE.17-23 The
separation strategies are therefore valuable since they
should propose a limited set of experimental conditions,
which maximize the probability to separate numerous
pharmaceuticals with different molecular structures and
properties.24 Literature survey revealed that reported
techniques for estimation of the drugs under investigation
were HPLC,1,25-30 TLC,31-33 CE,34-43 and LC‐MS/MS.5,44-49
This work was aimed to investigate the use of charged
CDs for a generic strategy in order to achieve the
separation of enantiomeric drug substances. Additionally,
to explore the conditions rapidly at which the separation
of the enantiomers is sufficiently good with acceptable
analysis time. The initial set of experiments was performed
to establish basic requirements and followed by a full
factorial design for variable selection. This approach can
also be considered as the first step for method optimization.
It should be clear that the purpose of the DOE is not only to
find the optimal separation but also to explore the
experimental domain in such a way that a good opportunity
to be implemented by a limited number of experiments.
2 | MATERIALS AND METHODS
2.1 | Materials
All racemic drugs were purchased from Sigma Aldrich
(St. Louis, Missouri). HS‐α‐CD and HS‐γ‐CD were
purchased from Beckman Coulter (Fullerton, California).
Hydroxypropyl‐β‐CD and sulfobutylether‐β‐CD were
purchased from Cyclolab R&D (Budapest, Hungary).
Orthophosphoric acid (85% w/w) and triethanolamine
were obtained from Merck (Darmstadt, Germany).
Ultra‐pure water resistivity >18 MΩ cm‐1 at 25°C and
TOC < 5 ppb was obtained from Milli‐Q UF‐Plus system
(Millipore, Bedford, Massachusetts).
2.2 | Instrumentation
A Beckman P/ACE MDQ Capillary Electrophoresis System
equipped with a diode array detector (Beckman Coulter,
Fullerton, California) with a BeckmanPACE station Version
1.1 used for data acquisition using fused‐silica capillaries
(Polymicro Technologies, Phoenix, Arizona) of 50‐μm ID
cut to total lengths 31.2 cm with an effective separation
FIGURE 1 Representative CE electropherogram of the enantiomeric separation for A, Ketoconazole; B, Itraconazole; C, Flueconazole; D,
Sertaconazole; E, Donepezil; F, Rivastigmine (100 μg mL‐1) for each drug under the following conditions: BGE consist of HS‐γ‐CD (10% m/v)
in 50 mM phosphate buffer (pH 2.5), voltage 25 KV, temperature 25 °C and injection pressure 50 m bar for 10 sec

4
ABDEL‐MEGIED ET AL.
FIGURE 2 Representative CE electropherogram of the enantiomeric separation for A, Ketoconazole; B, Itraconazole; C, Flueconazole; D,
Sertaconazole; E, Donepezil; F, Rivastigmine (100 μg mL‐1) for each drug under the following conditions: BGE consist of SBE (2.5% m/v) in 50
mM phosphate buffer (pH 2.5), voltage 30 KV, temperature 25 °C and injection pressure 50 m bar for 10 sec
TABLE 2A The 32 full factorial design and the measured response results (resolution) for investigated drugs
Coded
factors
Real
factors
Rs
KET
ITR
FLU
SRT
DON
RIV
pH %CD pH %CD
HS‐γ‐ HS‐α‐ HS‐γ‐ HS‐α‐ HS‐γ‐ HS‐α‐ HS‐γ‐ HS‐α‐ HS‐γ‐ HS‐α‐ HS‐γ‐ HS‐α‐
Experiment
CD
CD
CD
CD
CD
CD
CD
CD
cd
CD
CD
CD
1
−1 −1
2.5 2.5
2.5 10
5.5 2.5
5.5 10
0.922 1.629 2.80
0.899 1.165 1.81
1.581 3.12
1.141 3.31
0.0
2.81
2.801 3.176 0.0
0.0
0.0
2
3
4
5
6
7
8
9
−1
1
1
−1
1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.321 2.998 3.176 0.451 0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.391 0.989 0.575 0.260 0.0
0.451 1.143 0.710 0.0 0.0
0.999 1.014 4.80 1.528 3.33
0.400 2.832 0.790 0.391 0.0
0.0
0.250 0.0
0.0 0.0
3.541 3.67
0.0 0.0
0.671 0.421 0.930 0.0
0.377 0.40 2.90 0.0
0.0
0.0
1
0.0
0.0
0.0
−1
1
0
2.5
5.5
4
5
2.80
0.0
0.341 0.0
0.0 0.0
0
5
0
−1
1
2.5
10
5
0.687 1.116 1.11
0.790 4.010 1.81
0.789 1.81
0.451 0.0
1.11
1.09
0
4
0
0
4
0.780 4.297 0.987 0.991 0.0
1.537 0.689 0.342 3.467 0.0
TABLE 2B The 32 full factorial design and the measured response results (resolution) for investigated drugs with SBE‐β‐CD
Coded factors Real factors Rs (SBE‐β‐CD)
Experiment
pH
%CD
pH
%CD
KET
ITR
FLU
SRT
DON
RIV
1
−1
−1
2.5
2.5
10
2.5
10
5
1.95
1.11
1.51
1.91
2.33
0.0
2.98
3.31
4.98
0.43
0.31
2.87
0.41
0.98
1.21
0.99
2
3
4
5
6
7
8
9
−1
1
1
2.5
8.5
8.5
2.5
8.5
5.5
5.5
5.5
3.67
0.0
1.92
0.0
1.81
0.0
3.59
0.0
−1
1
1
0.0
0.0
0.0
0.0
0.0
−1
1
0
0.95
0.0
0.79
0.0
0.89
0.0
0.99
0.0
1.14
0.0
0
5
0
−1
1
2.5
10
5
0.34
0.88
0.65
0.22
0.45
0.35
0.65
0.78
1.03
0.91
1.87
1.32
0.97
1.70
1.00
0
0
0

ABDEL‐MEGIED ET AL.
5
length 21.0 cm was employed. The detection wavelength
was fixed at 220 nm.
for 3 min. Then, it was rinsed for 2 min with phosphate
buffer (pH 2.5) and equilibrated for 2 min with chiral
selector solution. At the end of the day, the capillary
was flushed with 0.1 M NaOH for 5 min and finally with
water for 10 min.
2.3 | Instrumental conditions
The separations were performed using 50‐mM phosphate‐
triethanolamine buffer (pH 2.5) with varying concentra-
tions of CDs. The activation and preconditioning step for
a new capillary were performed by rinsing with 1.0 M
NaOH, 0.1 M NaOH, ultra‐pure water, and finally BGE
solution. Each preconditioning step was performed at a
pressure of 60 psi for 10 min at temperature 20°C. After
each run, the capillary was first washed with 0.1 N NaOH
2.4 | Analysis procedures
Stock solutions of 1.0 mg mL‐1 of the racemic investigated
drugs were prepared in the least amount of methanol and
diluted with water/BGE solution for method development
studies. Standard solutions of 100 μg mL‐1 of each
FIGURE 3 A, Main effect plot for the effect of pH and %CD on the resolution of racemates using HS‐γ‐CD. B, Contour plots for the
resolution of six racemates on HS‐γ‐CD. The X‐axis shows the pH and the Y‐axis shows the %CD

6
ABDEL‐MEGIED ET AL.
FIGURE 3 Continued
compound was obtained by appropriate dilution with
BGE solution.
at three levels, to obtain total nine (32) experiments with
high, intermediate, and low values for both factors.
2.5 | Short‐end injection
3 | RESULTS AND DISCUSSION
The samples were injected at the end of the capillary near
to the detector; this could be performed by applying neg-
ative pressure. For successful separation when using the
short‐end of the capillary, the CE equipment was oper-
ated in the positive polarity mode which means that the
detection electrode was the cathode.
Enantioseparation of racemates is a unique challenge. In
a symmetric environment, the physical and chemical
properties of optical encounters (except the rotation of
polarized light) are all identical.21,50 The drugs under
investigation are basic and acidic drugs possessing pKa
values ranging from 2.1 to 8.9 for weak acids and bases
as shown in Table 1. Protonation of the piperidinic nitro-
gen of DON and the tertiary amino group of RIV were
obtained at acidic pH values. Since ionoselective selective
interaction (the dissociated form complexes with the CD)
is expected to occur at low pH values where the analytes
are in their charged form. So pH ranging from 2.5 to 5.0
was selected to represent the levels of pH in the DOE.
Modified CDs were more preferable than natural ones
because they provide a number of advantages in terms
of solubility, increased cavity depth, and the occurrence
of additional combinations that can interact and stabilize
the CD‐drug complex.50 For screening phase, it is prefera-
ble to start by HS‐CDs when developing new chiral
enantioseparation strategy. HS‐CDs have a strong electro-
phoretic mobility toward the positive electrode in the CE
environment that may be due the negatively charged
groups. In addition, HS‐CDs with phosphate‐TEA buffer
(pH 2.5) have been selected to enhance solubility and to
generate mobility by protonation of the amine buffers.
2.6 | Long‐end injection
Long‐end injection or conventional injection means the
injection of the analytes was performed at the end of the
capillary away from the detector using positive pressure.
Thus, the electrophoretic separation was done using
negative polarity voltage in which the capillary must be
thermostatted as the short‐end was outside the capillary
cartilage.
2.7 | Experimental design
DOE and optimization were performed using Minitab17
software (State College, Pennsylvania). The screening
stage aims to achieve rapid and acceptable separations
for all compounds under investigation. Screening strategy
is based on the use of highly sulfated (HS‐α‐CD) and
(HS‐γ‐CD). The design of experiment was based on full
factorial design of the crucial two factors (pH and %CD)

ABDEL‐MEGIED ET AL.
7
Adding TEA to the buffers helps to suppress the EOF in
order to increase the enantioselectivity. Under these con-
ditions, the basic compounds (which are strongly cationic
at low pH) would interact with the hydrophobic cavity as
well sonically with the negatively charged sulfates. At pH
from 2.5 to 3, zwitterionic analytes will be positively
charged and behave similarly like basic compounds, while
acidic compounds will be primarily protonated and
interact with the hydrophobic cavity of the HS‐CDs.
Compounds comprising two rings in their chemical
structures (KET, SRT, DON, and ITR) have highly enantio-
meric resolution values upon using HS‐CDs presumably
due to higher affinity and better fit in the hydrophobic
cavity of the CD. It was observed that a high affinity for
enantioseparation of all analytes occurred when
performed at low CD concentrations, while drugs having
large chain substituent tend to show the lowest resolution
(RIV, FLU). This could be attributed to the steric
hindrance that may be produced by those chains which
does not allow the molecule to be quite fit with the outer
negatively charged CDs. However, experiments should be
performed in the order: HS‐γ‐CD then HS‐α‐CD as shown
Figure 1.
After the resolution (Rs) for each drug has been calcu-
lated, some optimization steps were needed except for cases
where Rs exceeded 1.5 meaning baseline acceptable separa-
tion. While in case of 0 < Rs < 1.5 after screening, where the
peaks are only partially separated, some optimization steps
FIGURE 4 A, Main effect plot for the effect of pH and %CD on the resolution of racemates on HS‐α‐CD. B, Contour plots for the resolution
of 6 racemates on HS‐α‐CD. The X‐axis shows the pH and the Y‐axis shows the %CD

8
ABDEL‐MEGIED ET AL.
FIGURE 4 Continued
were performed to increase the resolution. Finally, when
Rs = 0 after screening, a limited number of additional exper-
iments are done; however, if separation does not improve,
mixtures of HS‐CDs with neutral CDs where tried because
they generally yield to better results. However, it was found
that SBE‐β‐CD led to the best enantioeparation in most of
the investigated drugs as shown Figure 2.
∑Yð0Þ ∑Yð−1Þ
Ex½0;
¼
−
;
(2)
−1ꢀ
N=3
N=3
in which ΣY 1, ΣY(0), and ΣY(‐1) represent the sum of the
responses were the factor х is at level 1, 0, and −1, respec-
tively, and N is the number of design experiments.
However, the separation also depends on other factors
such as the EOF, the ionic strength and the co‐ion of the
BGE, the presence of additives such as TEA. These are
secondary parameters, which were not examined in the
experimental design, and kept constant.
3.1 | Modeling the separation
Factorial experiments are based on varying all factors
simultaneously at a limited number of factor levels, where
the most influential factors, their ranges of influence, and
factor interactions are not yet known. DOE allows exper-
iments to be performed over the whole range of the factor
space. They show a high degree of accuracy in exchange
with a minimum experimental effort, enabling factor
interactions to be detected. The concentration of CDs
and pH are considered the most crucial parameters that
influence in the chiral separation for drugs under investi-
gation. The influence of two parameters, each at three
levels, on the resolution of investigated drugs was studied.
Nine experiments were carried out in duplicate (32) with
high, low, and intermediate values for both factors. The
full factorial design and their responses (Rs) for these
compounds are shown in Table 2A, B. For each factor,
three effects can be estimated, for the intervals between
the levels [1,0], [0,‐1], and [1,‐1]. Only two of those three
effects are independent calculated (Equations 1 and 2
3.2 | Hydroxy propyl‐β‐CD (HP‐β‐CD)
For acidic and basic compounds, lower neutral CD
concentrations were expected to be optimal because of
their lower affinity for the charge selector. Almost no peaks
were detected for HP‐β‐CD (50‐100 mM) and no baseline
separations were obtained. It was found that screening
with neutral CDs does not result in sufficient separation;
it is advised to investigate with charged CDs. Here,
negative polarity is applied since the EOF is very low.
3.3 | Highly sulfated gamma‐CD
(HS‐γ‐CD)
While there was no separation for RIV at any pH and
%CD, a study of the main effect showed that %CD was
of little effect on the resolution of KET and DON, with
higher effect on the resolution of FLU and has maximal
effect on the resolution of SRT and ITR. The best separa-
tion was obtained at intermediate values of %CD for those
∑Yð1Þ ∑Yð0Þ
Ex½1;
¼
−
;
(1)
0ꢀ
N=3
N=3

ABDEL‐MEGIED ET AL.
9
two racemates (ITR and SRT). On the other hand, pH has
pronounced effect on Rs of all drugs (except RIV as previ-
ously mentioned), the higher the pH, the lower the Rs.
Best Rs is obtained at pH 2.5 for all five racemates as
shown in Figure 3A. Contour plots clearly show that
max. Rs were obtained at low pH values. %CD has a
variable effect on Rs but is generally best when kept at
low values with different thresholds as shown in
Figure 3B. At a desirability (>0.6) the optimum conditions
for all drugs were pH = 2.5 and %CD = 2.5, except for
KET were Rs was best at 10 %CD. For the various drugs
under investigation accuracy of prediction lies indifferent
confidence intervals and has different desirability in
terms of whether was maximum Rs achieved for the
various drugs.
3.4 | Highly sulfated alpha‐CD (HS‐α‐CD)
While there was no separation for RIV or FLU, at any of
the studied levels of all factors, the main effect study
showed that for DON, the Rs improved at low values of
pH and low or intermediate values of %CD. For KET,
the Rs improved at low values of pH and %CD. For
ITR, intermediate pH shows better resolution than low
and high values, and at low %CD. For SRT, intermediate
FIGURE 5 A, Main effect plot for the effect of pH and %CD on the resolution of racemates on SBE‐β‐CD. B, Contour plots for the resolution
of 6 racemates on SBE‐β‐CD. The X‐axis shows the pH and the Y‐axis shows the %CD

10
ABDEL‐MEGIED ET AL.
FIGURE 5 Continued
values of pH and of %CD show the best resolution as
shown in Figure 4A. Contour plots control of %CD and
pH is critical in case of KET and DON where Rs drops
to very low values (<0.5) at pH higher than 4.0 if %CD
over than 5% (blue areas). A very limited space in the
design (dark green) offers an acceptable resolution for
these two drugs; either at intermediate %CD coupled
with low pH or vice versa. SRT has a better design space
with respect to the large robust area (dark green) at
intermediate values of %CD and pH. FLU and RIV
racemates did not resolve at any of the studied condi-
tions as shown in Figure 4B.The optimum conditions
%CD was 2.5 for all drugs, while it varied for the pH.
In the case of KET and DON, pH optimally is 2.5 to
achieve good resolution while it should be 5.5 for
SRT and ITR. The accuracy of estimation lies in different
confidence intervals and has different desirability
in terms of whether was maximum Rs achieved for the
various drugs.
upper left‐hand side of the space, at lowest pH and
highest %CD. Those two factors are more critical to
RIV, SRT, and KET than the others since Rs < 0.5
cannot be prevented at low %CD coupled with high
pH. Consequently, the optimization plot proved again
that the best conditions of pH and %CD for all drugs
were 2.5 and 10, respectively. For the various drugs
under study, accuracy of prediction lies indifferent
confidence intervals and have different desirability in
terms of whether was maximum Rs achieved for the
various drugs.
4 | CONCLUSION
Design of experiment–based screening and optimization
strategies for the separation of six racemic compounds
were studied using four different chiral selectors and
had been successfully applied using full factorial design
approach (32) that provide high efficiencies with analysis
time less than 10 min. The selection of the experimental
factors to be fixed and of others to be varied is considered
crucial step for the success of this CE strategy. Among
HP‐β‐CD, HS‐γ‐CD, HS‐α‐CD, SBE‐ β‐CD, the later was
unique leading to good resolution for all six racemates at
low pH values and higher %CD. In addition, the response
surfaces obtained with this chiral selector were able to
visually classify the behavior of the enantiomers accord-
ing to the enatiomeric resolution pattern into two main
groups; the first comprises RIV, SRT, and KET, and the
second ITR, FLU, and DON. However, the regression
results indicate that the linear model shows a significant
3.5 | Sulfobutyl ether β‐cyclodextrin
(SBE‐β‐CD)
SBE‐β‐CD was the most successful chiral selector
offering Rs for all the six racemates. In all cases, pH
and %CD had a dramatic effect on the separation. The
worst separation was always obtained at high pH values
and at low %CD as shown in Figure 5A. The contour
plots interestingly show similar resolution pattern two
groups of drugs: for RIV, SRT and KET, at one hand,
and ITR, FLU, and DON on the other hand as shown in
Figure 5B. In all cases, the best resolution occurs at the

ABDEL‐MEGIED ET AL.
11
9. Maier NM, Franco P, Lindner W. Separation of enantiomers:
needs, challenges, perspectives. J Chromatogr A. 2001;906:3‐33.
lack of fit for all chiral selectors, anticipating that higher
orders of the factors are most likely to be present in the
equation with possible interactions. For future work, this
strategy will be tested for the development of enantio-
meric separations of other acidic and basic drugs under
the same experimental conditions. One improvement that
we should investigate is if a central point will be added to
the two‐level designs can help to find good conditions
occurring at intermediate levels.
10. FDA, Policy. Statement for the development of new stereoiso-
meric drugs. Washington, DC avialble at https://www.fda.gov/
Drugs/GuidanceComplianceRegulatoryInformation/Guidances/
ucm122883.htm; 1992.
11. Lu J, Wan L, Zhong Y, et al. Stereoselective metabolism of
donepezil and steady‐state plasma concentrations of S‐donepezil
based on CYP2D6 polymorphisms in the therapeutic responses
of Han Chinese patients with Alzheimer's disease. J Pharmacol
Sci. 2015;129:188‐195.
ACKNOWLEDGMENTS
12. Bolognesi ML, Bartolini M, Cavalli A, et al. Design, synthesis,
and biological evaluation of conformationally restricted
rivastigmine analogues. J Med Chem. 2004;47:5945‐5952.
This research was supported by Faculteit Geneeskunde en
Farmacie, Farmaceutisch institut, Dienst Analytische
Scheikunde en, Farmaceutische Technologie (FABI),
Laarbeeklaan, Brussels, Belgium. We are thankful to Prof.
Dr. Yvan Vander Heyden who provided expertise that
greatly assisted the research.
13. Dilmaghanian S, Gerber J, Filler S, Sanchez A, Gal J.
Enantioselectivity of inhibition of cytochrome P450 3A4
(CYP3A4) by ketoconazole: testosterone and methadone as
substrates. Chirality. 2004;16:79‐85.
14. Shi W, Nacev BA, Bhat S, Liu JO. Impact of absolute stereochem-
istry on the antiangiogenic and antifungal activities of
itraconazole. ACS Med Chem Lett. 2010;1:155‐159.
ORCID
15. Tomaszewski J, Rumore MM. Stereoisomeric drugs: FDA's
policy statement and the impact on drug development.
Drug Dev Ind Pharm. 1994;20:119‐139.
Ahmed M. Abdel‐Megied http://orcid.org/0000-0003-0180-
0918
Rasha S. Hanafi http://orcid.org/0000-0002-8196-5108
16. Francotte ER. Enantioselective chromatography as a powerful
alternative for the preparation of drug enantiomers.
J Chromatogr A. 2001;906:379‐397.
Hassan Y. Aboul‐Enein
7009
http://orcid.org/0000-0003-0249-
17. Dejaegher B, Vander Heyden Y. The use of experimental design
REFERENCES
in separation science. Acta Chromatographica. 2009;21:161‐201.
1. Radwan MA, Abdine HH, Al‐Quadeb BT, Aboul‐Enein HY,
Nakashima K. Stereoselective HPLC assay of donepezil enantio-
mers with UV detection and its application to pharmacokinetics
in rats. J Chromatogr B Biomed Sci Appl. 2006;830:114‐119.
18. Dejaegher B, Durand A, Vander HY. Experimental design in
method optimization and robustness testing. New Jersey, USA:
John Wiley & Sons; 2010.
19. Ali I, Aboul‐Enein HY, Gupta VK. Nanochromatography and
Nanocapillary Electrophoresis: Pharmaceutical and Environmen-
tal Analyses. Hoboken, USA: John Wiley & Sons; 2009.
2. Brunton LL, Parker KL. Goodman & Gilman's Manual of
Pharmacology and Therapeutics. McGraw‐Hill Companies, Inc:
CA, USA; 2008.
20. Ali I, Suhail M, Al‐Othman ZA, Alwarthan A, Aboul‐Enein HY.
Enantiomeric resolution of multiple chiral centres racemates by
capillary electrophoresis. Biomed Chromatogr. 2016;30:683‐694.
3. Muller T. Rivastigmine in the treatment of patients with
Alzheimer's disease. Neuropsychiatr Dis Treat. 2007;3:211‐218.
4. Sugimoto H, Yamanishi Y, Iimura Y, Kawakami Y. Donepezil
hydrochloride (E2020) and other acetylcholinesterase inhibitors.
Curr Med Chem. 2000;7:303‐339.
21. Al‐OthmanZA,Al‐WarthanA,AlamSD,AliI.Enantio‐separation
of drugs with multiple chiral centers by chromatography and
capillary electrophoresis. BiomedChromatogr. 2014;28:1514‐1524.
5. Matsui K, Oda Y, Nakata H, Yoshimura T. Simultaneous
determination of donepezil (aricept) enantiomers in human
plasma by liquid chromatography‐electrospray tandem mass
spectrometry. J Chromatogr B Biomed Sci Appl. 1999;729:147‐155.
22. Ali I, Al‐Othman ZA, Al‐Warthan A, Asnin L, Chudinov A.
Advances in chiral separations of small peptides by capillary elec-
trophoresis and chromatography. J Sep Sci. 2014;37:2447‐2466.
23. Ali I, Al‐Othman ZA, Al‐Warthan A, A‐Eh Y. Recent trends in
chiral separations by nano liquid chromatography and nano
capillary electrophoresis. Curr Chromatogr. 2014;1:81‐89.
6. Crego AL, Marina ML, Lavandera JL. Optimization of the sepa-
ration of a group of antifungals by capillary zone electrophoresis.
J Chromatogr A. 2001;917:337‐345.
24. Ates H, Mangelings D, Vander Heyden Y. Fast generic chiral
separation strategies using electrophoretic and liquid chromato-
graphic techniques. J Pharm Biomed Anal. 2008;48:288‐294.
7. Khoza S, Moyo I, Ncube D. Comparative hepatotoxicity of flucon-
azole, ketoconazole, itraconazole, terbinafine, and griseofulvin in
rats. J Toxic. 2017;1:1‐9.
25. Huang Q, Zhang K, Wang Z, Wang C, Peng X. Enantiomeric
determination of azole antifungals in wastewater and sludge by
liquid chromatography‐tandem mass spectrometry. Anal
Bioanal Chem. 2012;403:1751‐1760.
8. Aboul‐Enein HY, Wainer IW. The Impact of Stereochemistry on
Drug Development and Use: Chemical Analysis series. New York,
USA: John Wiley & Sons; 1997.

12
ABDEL‐MEGIED ET AL.
26. Hamdy DA, Brocks DR. Nonlinear stereoselective pharmacoki-
netics of ketoconazole in rat after administration of racemate.
Chirality. 2009;21:704‐712.
40. Lucangioli SE, Tripodi V, Masrian E, Scioscia SL, Carducci CN,
Kenndler E. Enantioselective separation of rivastigmine by
capillary electrophoresis with cyclodextrines. J Chromatogr A.
2005;1081:31‐35.
27. Hamdy DA, Brocks DR. A stereospecific high‐performance
liquid chromatographic assay for the determination of ketocona-
zole enantiomers in rat plasma. Biomed Chromatogr.
2008;22:542‐547.
41. Kavalirova A, Pospisilova M, Karlicek R. Enantiomeric analysis
of rivastigmine in pharmaceuticals by cyclodextrin‐modified
capillary zone electrophoresis. Anal Chim Acta. 2004;525:43‐51.
28. Thienpont A, Gal J, Aeschlimann C, Félix G. Studies on
stereoselective separations of the “azole” antifungal drugs keto-
conazole and itraconazole using HPLC and SFC on silica‐based
polysaccharides. Analusis. 1999;27:713‐718.
42. Gotti R, Cavrini V, Pomponio R, Andrisano V. Analysis and
enantioresolution of donepezil by capillary electrophoresis.
J Pharm Biomed Anal. 2001;24:863‐870.
43. Wan Ibrahim WA, Arsad SR, Maarof H, Sanagi MM, Aboul‐Enein
HY. Chiral separation of four stereoisomers of ketoconazole drugs
using capillary electrophoresis. Chirality. 2015;27:223‐227.
29. Srinivasu MK, Rao BM, Reddy BS, Kumar PR, Chandrasekhar
KB, Mohakhud PK. A validated chiral liquid chromatographic
method for the enantiomeric separation of rivastigmine hydro-
gen tartarate, a cholinesterase inhibitor. J Pharm Biomed Anal.
2005;38:320‐325.
44. Bhatt J, Subbaiah G, Kambli S, et al. A rapid and sensitive liquid
chromatography‐tandem mass spectrometry (LC‐MS/MS)
method for the estimation of rivastigmine in human plasma.
J Chromatogr B Biomed Sci Appl. 2007;852:115‐121.
30. Barth T, Conti R, Pupo MT, Okano LT, Bonato PS. Chiral HPLC
analysis of donepezil, 5‐O‐desmethyl donepezil and 6‐O‐
desmethyl donepezil in culture medium: application to fungal
biotransformation studies. Anal Bioanal Chem. 2012;404:257‐266.
45. Pommier F, Frigola R. Quantitative determination of
rivastigmine and its major metabolite in human plasma by liquid
chromatography with atmospheric pressure chemical ionization
tandem mass spectrometry. J Chromatogr B Biomed Sci Appl.
2003;784:301‐313.
31. Saysin S, Liawruangrath B, Liawruangrath S. High‐performance
thin‐layer chromatographic determination of ketoconazole in
pharmaceutical formulations. J Cosmet Sci. 2010;61:367‐376.
46. Thomas S, Shandilya S, Bharati A, Paul SK, Agarwal A, Mathela
CS. Identification, characterization and quantification of new
impurities by LC‐ESI/MS/MS and LC‐UV methods in
rivastigmine tartrate active pharmaceutical ingredient. J Pharm
Biomed Anal. 2012;57:39‐51.
32. Abdel‐Moety EM, Khattab FI, Kelani KM, AbouAl‐Alamein AM.
Chromatographic determination of clotrimazole, ketoconazole
and fluconazole in pharmaceutical formulations. Farmacote-
rapia. 2002;57:931‐938.
33. Radosław J, Ekiert Krzek J, Rzeszutko W. Evaluation of a TLC
densitometric method for analysis of azole antifungal agents.
Chromatographia. 2008;67:995‐998.
47. Arumugam K, Chamallamudi MR, Gilibili RR, et al. Develop-
ment and validation of a HPLC method for quantification of
rivastigmine in rat urine and identification of a novel metabolite
in urine by LC‐MS/MS. Biomed Chromatogr. 2011;25:353‐361.
34. Velikinac I, Cudina O, Jankovic I, Agbaba D, Vladimirov S.
Comparison of capillary zone electrophoresis and high
performance liquid chromatography methods for quantitative
determination of ketoconazole in drug formulations.
Farmacoterapia. 2004;59:419‐424.
48. Shah HJ, Kundlik ML, Pandya A, et al. A rapid and specific
approach for direct measurement of donepezil concentration in
human plasma by LC‐MS/MS employing solid‐phase extraction.
Biomed Chromatogr. 2009;23:141‐151.
35. Castro‐Puyana M, Crego AL, Marina ML. Enantiomeric separation
of ketoconazole and terconazole antifungals by electrokinetic
chromatography: rapid quantitative analysis of ketoconazole in
pharmaceutical formulations. Electrophoresis. 2005;26:3960‐3968.
49. Apostolou C, Dotsikas Y, Kousoulos C, Loukas YL. Quantitative
determination of donepezil in human plasma by liquid chroma-
tography/tandem mass spectrometry employing an automated
liquid‐liquid extraction based on 96‐well format plates. Applica-
tion to a bioequivalence study. J Chromatogr B Biomed Sci Appl.
2007;848:239‐244.
36. Wei Y, Li J, Zhu C, Hao A, Zhao M. 2‐O‐(2‐hydroxybutyl)‐beta‐
cyclodextrin as a chiral selector for the capillary electrophoretic
separation of chiral drugs. Anal Sci. 2005;21:959‐962.
50. Beesley TE, Scott RPW. Chiral Chromatography. New York,
USA: John Wiley & Sons; 1998.
37. Castro‐Puyana M, Crego AL, Marina ML, Garcia‐Ruiz C.
Enantioselective separation of azole compounds by EKC. Rever-
sal of migration order of enantiomers with CD concentration.
Electrophoresis. 2007;28:2667‐2674.
How to cite this article: Abdel‐Megied AM,
Hanafi RS, Aboul‐Enein HY. A chiral
enantioseparation generic strategy for anti‐
Alzheimer and antifungal drugs by short end
injection capillary electrophoresis using an
experimental design approach. Chirality. 2017;1–12.
https://doi.org/10.1002/chir.22777
38. Breadmore MC, Thormann W. Capillary electrophoresis evi-
dence for the stereoselective metabolism of itraconazole in
man. Electrophoresis. 2003;24:2588‐2597.
39. Jimidar MI, Van Ael W, Shah R, Redlich D, De Smet M. Fast
method development and rapid analysis using a screening
approach for enantiomeric separations in capillary electrophore-
sis. J Capill Electrophor Microchip Technol. 2003;8:101‐110.