ORIGINAL PAPER

Chiral HPLC analysis of donepezil, 5-O-desmethyl donepezil

and 6-O-desmethyl donepezil in culture medium: application
to fungal biotransformation studies
Thiago Barth & Raphael Conti & Mnica Tallarico Pupo &
Laura Tiemi Okano & Pierina Sueli Bonato
Received: 12 April 2012 / Revised: 4 May 2012 / Accepted: 7 May 2012 / Published online: 29 May 2012
#Springer-Verlag 2012
Abstract An high performance liquid chromatography
(HPLC) method for the enantioselective determination of
donepezil (DPZ), 5-O-desmethyl donepezil (5-ODD), and
6-O-desmethyl donepezil (6-ODD) in Czapek culture medi-
um to be applied to biotransformation studies with fungi is
described for the first time. The HPLC analysis was carried
out using a Chiralpak AD-H column with hexane/ethanol/
methanol (75:20:5, v/v/v) plus 0.3 % triethylamine as mobile
phase and UV detection at 270 nm. Sample preparation was
carried out by liquidliquid extraction using ethyl acetate as
extractor solvent. The method was linear over the concen-
tration range of 10010,000 ng mL
1
for each enantiomer of
DPZ (r0.9985) and of 1005,000 ng mL
1
for each enan-
tiomer of 5-ODD (r0.9977) and 6-ODD (r0.9951).
Within-day and between-day precision and accuracy evalu-
ated by relative standard deviations and relative errors,
respectively, were lower than 15 % for all analytes. The
validated method was used to assess DPZ biotransformation
by the fungi Beauveria bassiana American Type Culture
Collection (ATCC) 7159 and Cunninghamella elegans ATCC
10028B. Using the fungus B. bassiana ATCC 7159, a
predominant formation of (R)-5-ODD was observed while
for the fungus C. elegans ATCC 10028B, DPZ was bio-
transformed to (R)-6-ODD with an enantiomeric excess
of 100 %.
Keywords Beauveria bassiana
.
Cunninghamella elegans
.
Donepezil
.
Enantioseparation
.
5-O-Desmethyl donepezil
.
6-O-Desmethyl donepezil
Introduction
Donepezil (DPZ) (Fig. 1a), 2,3-dihydro-5,6-dimethoxy2-
[[1-(phenylmethyl)-4-piperidinyl]methyl]-1H-inden-1-one
[1], is a reversible and selective inhibitor of acetylcholines-
terase [2], therapeutically used for the treatment of Alz-
heimers disease [3]. DPZ is marketed as a racemic
mixture [4] and its enantiomers have different extents of
inhibition against acetylcholinesterase in vivo and in vitro
[5]. However, no information is available concerning the most
active enantiomer.
An in vivo study carried out in rats showed that O-deme-
thylation, N-dealkylation, and N-oxidation are the main reac-
tions involved in the biotransformation of DPZ. In addition,
this study showed the predominant formation of the metabo-
lite 6-O-desmethyl donepezil (6-ODD) (Fig. 1b) [6]. The in
vivo metabolism study of DPZ in humans suggests that the
isoenzymes involved in its metabolism are CYP2D6 and
CYP3A4. The major metabolites observed were 6-ODD, 5-
O-desmethyl donepezil (5-ODD) (Fig. 1c), DPZ-cis-N-oxide,
and the glucuronide forms of 6-ODD and 5-ODD [7]. The
6-ODD is the only metabolite that has pharmacological activ-
ity similar to DPZ [8]. No information is available concerning
the enantioselective metabolism of this drug.
T. Barth (*)
:
P. S. Bonato (*)
Departamento de Fsica e Qumica, Faculdade de Cincias
Farmacuticas de Ribeiro Preto, Universidade de So Paulo,
Av. Caf SN,
Ribeiro Preto, SP 14040-903, Brazil
e-mail: barththiago@yahoo.com.br
e-mail: psbonato@hotmail.com
R. Conti
:
M. T. Pupo
Departamento de Cincias Farmacuticas, Faculdade de Cincias
Farmacuticas de Ribeiro Preto, Universidade de So Paulo,
Ribeiro Preto, SP 14040-903, Brazil
L. T. Okano
Departamento de Qumica, Faculdade de Filosofia,
Cincias e Letras de Ribeiro Preto, Universidade de So Paulo,
Ribeiro Preto, SP 14040-901, Brazil
Anal Bioanal Chem (2012) 404:257266
DOI 10.1007/s00216-012-6107-3
The enantioselective determination of DPZ in biological
matrices is described in only few papers. In 1992, Haginaka
and Seyama [9] reported the resolution of DPZ enantiomers
in rat plasma by HPLC using a protein-based stationary
phase (ovomucoid, Ultron ES-OVM) and protein precipita-
tion for sample preparation. Matsui et al. [10] developed an
HPLC method for the enantioselective determination of
DPZ enantiomers in dog plasma with a protein-based sta-
tionary phase (avidin, Bioptick AV-1) and direct plasma
injection by column-switching procedure. In 1999, the same
group [5] reported another HPLC method for the enantiose-
lective determination of DPZ enantiomers in human plasma
using the same protein-based stationary phase and liquid
liquid extraction for sample preparation. In turn, Radwan et
al. [4] reported a method for the enantioselective determina-
tion of DPZ in rat plasma and pharmaceutical formulations
using liquid-liquid extraction and HPLC with polysaccharide-
based stationary phase (cellulose tris(3,5-dimethylphenylcar-
bamate)). More recently, DPZ enantiomers were determined
in rabbit plasma by capillary electrophoresis using sulphated-
-cyclodextrin as chiral selector and liquid-liquid extraction
for sample preparation [11]. Up to now, the simultaneous
enantioselective determination of DPZ, 5-ODD, and 6-ODD
in biological samples has not been reported.
The pharmacokinetic, pharmacodynamics, and toxico-
logical enantioselective properties of DPZ and especially
of its metabolites have been little studied. To perform these
studies, milligrams of metabolites in the enantiomerically
pure form are required. A strategy to obtain them is the use
of fungi in biotransformation processes [12]. The synthesis
of optically active compounds by using microbial models
offers advantages compared with chemical synthesis, be-
cause it can be highly enantiomeric and regio-selective
under mild conditions [13]. Biotransformation studies with
fungi may also provide information for further correlations
with enantioselective biotransformation in vivo, for simulat-
ing the mammalian metabolism [14, 15]. These considera-
tions justify the interest in developing an enantioselective
method for the simultaneous determination of DPZ, 6-ODD,
and 5-ODD in culture mediumto be used in biotransformation
studies of DPZ by fungi.
Experimental
Chemicals and reagents
The reference substances rac-donepezil hydrochloride, rac-
5-ODD, and rac-6-ODD were purchased from Toronto Re-
search Chemicals (North York, ON, Canada). Methanol and
ethanol were purchased from JT Baker (Phillipsburg, NJ,
USA), 2-propanol was obtained from Fisher Scientific (Fair
Lawn, NJ, USA) and hexane (n-hexane 95 %) was pur-
chased from Tedia (Fairfield, OH, USA), all of chromato-
graphic grade. Triethylamine was obtained from JT Baker
(Phillipsburg, NJ, USA). Ethyl acetate was obtained from
Tedia (Fairfield, OH, USA). The boric acid and potassium
dihydrogenphosphate were obtained from Merck (Darmstadt,
Germany), and sodiumtetraborate decahydrate was purchased
from JT Baker (Phillipsburg, NJ, USA). All these chemicals
were of analytical grade in the highest purity available. Water
was purified with a Milli-Q plus system (Millipore, Bedford,
MA, USA).
Reference substance solutions
The DPZ, 5-ODD, and 6-ODD stock solutions were pre-
pared at the concentration of 1 mg mL
1
. The working
Fig. 1 Chemical structures of
DPZ (a), 6-ODD (b), and
5-ODD (c). * Represent the
chiral center
258 T. Barth et al.
solutions of rac-DPZ were prepared at the concentrations of
4, 12, 40, 120, 300, and 400 g mL
1
while the rac-5-ODD
and rac-6-ODD solutions were 4, 12, 30, 60, 150, and
200 g mL
1
. All of them were prepared in methanol on a
free-base basis. The solutions were stored frozen at 20 C
and protected from light.
Liquid chromatographic conditions
The liquid chromatographic analyses were conducted using
a Shimadzu chromatograph (Kyoto, Japan), equipped with
an LC-10AS solvent pump unit, a SPD-10A UVvis detec-
tor operating at 270 nm. The system control was carried out
by a SCL-10A VP controller. Injections were performed
manually through a 50-L loop with a Rheodyne model
7125 injector (Cotati, CA, USA). A Shimadzu LC solution
software, version 1.22 SP1, was used for system control and
data acquisition. The resolution of DPZ, 5-ODD, and 6-ODD
enantiomers was carried out at room temperature (252 C)
on a Chiralpak AD-H column (1504.6 mm, 5 m par-
ticle size, Chiral Technologies, Exton, PA, USA) using
hexane/ethanol/methanol (75:20:5, v/v/v) plus 0.3 % trie-
thylamine as the mobile phase at a flowrate of 1.5 mL min
1
.
A CN guard column (44 mm, 5-m particle size, Merck,
Darmstadt, Germany) was used to protect the analytical
column.
Elution order determination
The enantiomers of DPZ, 5-ODD and 6-ODD were obtained
by semi-preparative analysis of the racemates under the
HPLC conditions described in the present paper (see Liquid
chromatographic conditions). Furthermore, the collected
enantiomers of DPZ were analyzed on a Chiralcel OD
column (254.6 mm, 10 m particle size) using hex-
ane/2-propanol/TEA (87:12.9:0.1, v/v/v) as mobile phase,
according to the conditions described by Radwan et al. [4].
Then, the retention times of the enantiomers from both
studies were compared and the elution order established.
In turn, the elution order of 5-ODD and 6-ODD enan-
tiomers was determined by obtaining the circular dichro-
ism (CD) spectra of the pure enantiomers of DPZ, 5-ODD,
and 6-ODD (200 to 400 nm) on a JASCO J-810 spectropo-
larimeter equipped with the temperature control apparatus
JASCO PTC-423S, using a quartz cuvette with a 1.0-cm
optical path at 25 C (Easton, MD, USA). Subtractions of
base line (mobile phase) were carried out for all analysis. The
standard scanning conditions were at a rate of 200 nm min
1
continuous mode, spectral bandwidth 1 nm, and a response of
0.5 s. The 5-ODD and 6-ODD enantiomer that pre-
sented the same cotton effect of (R)-DPZ was considered
(R) and the one that presented the same cotton effect of (S)-
DPZ was considered (S).
Extraction procedure
DPZ, 5-ODD, and 6-ODD were extracted from the Czapek
culture medium by a liquid-liquid extraction procedure.
Aliquots of 0.5 mL Czapek medium spiked with 25 L of
standard solutions of DPZ, 5-ODD, and 6-ODD or samples
obtained in the biotransformation process were transferred
to 10 mL glass tubes, and buffered with 0.5 mL of sodium
borate buffer 0.1 mol L
1
, pH 9. The samples were mixed by
vortex agitation for 20 s. Then, a 4-mL aliquot of extraction
solvent, ethyl acetate, was added. The tubes were shacked
for 15 min using a Vibrax VXR agitator (IKA, Staufen,
Germany) set at 1,500 rpm and then centrifuged at
1,800g for 5 min, at 4 C. The organic layers (3 mL) were
transferred to 10 mL conical glass tubes. The solvent was
evaporated to dryness under a stream of compressed air at
room temperature. The residues were dissolved in 100 L of
mobile phase and vortex mixed for 20 s, and then 50 L
were analyzed by the HPLC system.
Racemization assessment during extraction
and biotransformation procedures
DPZ may racemize, via a keto-enol intermediate since the
chiral center is adjacent to a carbonyl group [5, 9, 10]. So,
the influence of the biotransformation and extraction con-
ditions on the racemization of DPZ, 5-ODD and 6-ODD
was evaluated. Firstly, the enantiomers of DPZ, 5-ODD and
6-ODD were obtained by semi-preparative analysis of the
racemates under the HPLC conditions described in the pres-
ent paper (see Liquid chromatographic conditions). After
separation, the fractions containing each enantiomer were
evaporated to dryness under a stream of compressed air and
the residues were dissolved in methanol to obtain a final
concentration of approximately 50 g mL
1
. The enantio-
meric excess (ee) of these solutions was determined by the
equation ee0(AB/A+B)100, where A is the peak area of
the enantiomer in higher concentration and B is the peak
area of the enantiomer in lower concentration [16].
To determine the racemization in the biotransformation
conditions, 1-mL aliquots of DPZ, 5-ODD, and 6-ODD
enantiomers solutions were added to Erlenmeyer flasks con-
taining 100 mL of Czapek culture medium and submitted to
the same conditions used in the biotransformation procedure
(see Donepezil biotransformation procedure). Daily, dur-
ing the period of biotransformation (7 days), aliquots of
1 mL (n03) were collected. At the end of the biotransforma-
tion study, the samples were extracted and analyzed in tripli-
cate and the ee was determined. This ee values were compared
with the ee obtained after the enantiomer separation.
To evaluate the influence of sample pH during the ex-
traction procedure on the racemization, samples of 0.5 mL
of Czapek culture medium were spiked with 25 L of the
Chiral HPLC analysis of donepezil, 5-O-desmethyl donepezil and 6-O-desmethyl donepezil 259
solutions of each enantiomer of DPZ, 5-ODD, and 6-ODD
(n03). After that the pH of the samples were adjusted by the
addition of 0.5 mL of buffer solutions and the extraction was
performed. The evaluated pH values were 7.0 (sodium
phosphate buffer, 0.1 mol L
1
), 9.0 (sodium borate buffer,
0.1 mol L
1
), and 10.0 (sodium borate buffer, 0.1 mol L
1
).
The ee of the enantiomer solutions obtained after enantio-
mer separation was also used for comparison with the ee
determined during the evaluation of the pH effect on the
racemization.
Method validation
To determine the extraction recovery, aliquots of Czapek
culture medium (0.5 mL) were spiked with DPZ at the
concentrations of 300, 3,000, and 7,500 ng mL
1
for each
enantiomer (n03) and 300, 1,500, and 3,750 ng mL
1
for
each enantiomer of 5-ODD and 6-ODD and submitted to the
extraction procedure. Another set of samples were prepared
extracting 0.5 mL aliquots of Czapek medium and then
spiking the extract with the same amounts of DPZ, 5-
ODD, and 6-ODD enantiomers. The recovery was deter-
mined by the ratio of the areas of the samples with analytes
extracted and non-extracted and expressed as percentage of
the amount extracted.
Calibration curves were obtained by spiking 0.5 mLaliquots
of Czapek culture mediumwith 25 Lof standard solutions of
rac-DPZ, in the concentration range of 4400 g mL
1
,
resulting in concentrations of 10010,000 ng mL
1
for
each enantiomer. The calibration curves for rac-5-ODD
and rac-6-ODD were prepared similarly at the concen-
tration range of 1005000 ng mL
1
for each enantiomer.
The linearity of the calibration curves was determined
using the correlation coefficient (r) and the F test for lack-
of-fit (F
LOF
) using a p value of 0.05. MINITAB Release
version 14.1 (State College, PA, USA) was used to perform
the statistical calculations.
The sensitivity of the method was evaluated by determin-
ing the quantification limit (LOQ). The LOQ was defined as
the lowest enantiomer concentration that could be deter-
mined with accuracy and precision below 20 % [17] over
five analytical runs. The LOQ was determined by using
aliquots of Czapek culture medium (0.5 mL) spiked with
100 ng mL
1
of each enantiomer (the lowest point in the
calibration curve).
The precision and accuracy of the method were eval-
uated by within-day (n05) and between-day (n03)
assays using Czapek culture medium spiked with DPZ
at the concentrations of 300, 3,000, and 7,500 ng mL
1
for each enantiomer and 300, 1,500, and 3,750 ng mL
1
for
each enantiomer of 5-ODD and 6-ODD. The results obtained
were expressed as relative standard deviation (RSD, %) and
relative error (RE (%)).
Freeze-thaw cycle stability, short-term room temperature
stability and stability in the biotransformation conditions
were determined. To perform the freezethaw cycle stability,
three aliquots (n03) of samples prepared in Czapek culture
medium at the concentration of 300 and 7,500 ng mL
1
of
each enantiomer of DPZ and 300 and 3,750 ng mL
1
of each
enantiomer of 5-ODD and 6-ODD were stored at 20 C for
24 h and thawed at room temperature. When completely
thawed, the samples were refrozen for 12 h under the same
conditions. The freezethaw cycle was repeated twice, and
then the samples were analyzed on the third cycle. For the
determination of short-term room temperature stability, ali-
quots of samples prepared in Czapek culture medium at the
concentrations specified above were kept at room tempera-
ture (222 C) for 12 h and analyzed. To determine the
stability in the biotransformation conditions, an aliquot of
2 mg of rac-DPZ (free-base basis) dissolved in 1 mL of
sterile water was added to Erlenmeyer flasks containing
100 mL of Czapek medium (10 g mL
1
of each enantio-
mer), and submitted to the same conditions used in the
biotransformation procedure (see Donepezil biotransfor-
mation procedure). Daily, during the period of biotransfor-
mation (7 days), aliquots of 0.5 mL (n03) were analyzed.
The peak areas obtained from the stability studies were
compared with the peak areas obtained with freshly pre-
pared samples at the same concentration and were consid-
ered stable if the deviation (expressed as RE (%)) from the
fresh samples was within15 %.
The selectivity of the method was evaluated by analyzing
sterile Czapek medium and sterile Czapek medium added of
fungal mycelium under the conditions previously estab-
lished (see Donepezil biotransformation procedure).
Fungi
The fungi Beauveria bassiana American Type Culture Col-
lection (ATCC) 7159 and Cunninghamella elegans ATCC
10028B, purchased from the ATCC (Manassas, VA, USA)
were used in this biotransformation study.
Donepezil biotransformation procedure
Three discs of 0.5 cm of diameter (potato dextrose agar
plugs) containing the fungal mycelia were aseptically trans-
ferred to 9.0-cm diameter Petri dishes containing PDA me-
dium (potato, dextrose, and agar) and allowed to grow for
6 days at 30 C. Biotransformation was performed using a
two-stage fermentation protocol [18, 19]. In the first stage
(pre-culture), three 0.5-cm uniform discs were cut with a
transfer tube (Fischer Scientific, Pittsburgh, PA, USA) and
then inoculated in 50-mL Falcon tubes containing 10 mL of
pre-fermentative liquid broth (glucose, 10 g; tryptone soy
broth, 5 g; yeast extract, 3 g; and malt extract, 10 g; per litre
260 T. Barth et al.
and pH adjusted to 6.2 with 0.1 mol L
1
HCl solution). The
Falcon tubes were incubated for 4 days (96 h) at 30 C on a
rotatory shaker (New Brunswick Scientific Co., Inc., model
INNOVA 4300, Edison, NJ, USA) operating at 120 rpm.
In the second stage (biotransformation), the resulting myce-
lium was transferred into 250-mL Erlenmeyer flasks con-
taining 100 mL of Czapek medium [20] adjusted to pH 5.0
with a 1.0 mol L
1
HCl solution.
Moreover, 2 mg of DPZ (free-base basis) dissolved in
1 mL of sterile water was added to the flask. Control flasks
consisted of culture broth (Czapek) without DPZ and fungi,
sterile broth with fungal mycelium, and sterile broth with
DPZ (stability in biotransformation conditions). Biotrans-
formation experiments were carried out at 30 C, with shaking
at 120 rpm for 168 h. Aliquots of 0.5 mL were collected from
the culture flasks, submitted to extraction procedure and ana-
lyzed by HPLC.
The biotransformation kinetic studies were presented as
concentration versus collecting interval (hours) profiles. In
addition, the results obtained in the biotransformation process
were also expressed as ee. The efficiency of the biotransfor-
mation process was calculated (in percentage). It was per-
formed quantifying the amount of the 5-ODD and 6-ODD
metabolites in the culture medium and correlating this amount
with the initial amount of DPZ (time 0). The biotransforma-
tion procedure was performed in replicate (n02).
Results and discussion
Selection of the separation conditions
The chiral resolution of DPZ, 5-ODD, and 6-ODD was per-
formed using a Chiralpak AD-H column, under normal elu-
tion mode. The mobile phase was prepared using hexane/
2-propanol, hexane/2-propanol/methanol, hexane/ethanol, or
hexane/ethanol/methanol mixtures. Methanol was used to
reduce the analysis time and to improve the peak efficiency.
Triethylamine was added to these mobile phases in order to
reduce the interaction of the drugs with the silanol groups of
the silica support. Moreover, methanol and triethylamine were
important for the separation of (S)-DPZ (peak 1) and (S)-
5-ODD (peak 2) (Fig. 2a). The best separation was achieved
with a mobile phase consisting of hexane/ethanol/methanol
(75:20:5, v/v/v) plus 0.3 % triethylamine, at a flow rate of
1.5 mL min
1
and detection at 270 nm. Under these condi-
tions, the separation of all compounds was performed in
19 min with enough resolution (R
s
1.56) (Fig. 2a).
Elution order determination
The elution order for DPZ enantiomers on the Chiralpak
AD-H column was reversal than the observed by Radwan et
al. [4], which used a Chiralcel OD column. The chiral
selector for the Chiralpak AD-H and Chiralcel OD columns
are amylose tris (3,5-dimethylphenylcarbamate) and cellu-
lose tris(3,5-dimethylphenylcarbamate), respectively. So the
difference between them is the polysaccharides used to
prepare the derivative. The reversal of elution order by
changing the polysaccharide is well documented [2124].
Peak 1 corresponded to (S)-DPZ and peak 4 to (R)-DPZ
(Fig. 2a).
The chiral resolution and elution order of 5-ODD and 6-
ODD enantiomers is not available in the literature. There-
fore, to determine the elution order of DPZ metabolites, a
combination of chromatographic and CD methods was used
[25, 26]. Hence, the CD spectra of 5-ODD and 6-ODD
enantiomers were compared with the CD spectra of DPZ
enantiomers. For 5-ODD and 6-ODD, the first eluted enan-
tiomers showed cotton effect similar to the (S)-DPZ and
they were considered as (S)-5-ODD (peak 2) and (S)-6-
ODD (peak 3); the second eluted enantiomers of the metab-
olites showed cotton effect similar to the (R)-DPZ and they
were considered as (R)-5-ODD (peak 5) and (R)-6-ODD
(peak 6) (Fig. 2a). The CD spectra used for the determina-
tion of the elution order are presented in Fig. 3.
Method validation
Several solvents or mixtures of them were evaluated to
extract DPZ, 5-ODD, and 6-ODD enantiomers from Czapek
Fig. 2 (a) Representative chromatogram of blank Czapek culture
medium spiked with 3,000 ng mL
1
of each DPZ enantiomer and
1,500 ng mL
1
of each 5- and 6-ODD enantiomers. (b) Representative
chromatogram of blank Czapek culture medium. (c) Representative
chromatogram of Czapek culture medium incubated with the fungus C.
elegans ATCC 10028B during 7 days. (d) Representative chromatogram
of Czapek culture medium incubated with the fungus B. bassiana ATCC
7159 during 7 days. (1) (S)-DPZ, (2) (S)-5-ODD, (3) (S)-6-ODD, (4) (R)-
DPZ, (5) (R)-5-ODD, and (6) (R)-6-ODD. Chromatographic conditions
are described in Liquid chromatographic conditions
Chiral HPLC analysis of donepezil, 5-O-desmethyl donepezil and 6-O-desmethyl donepezil 261
culture medium using liquid-liquid extraction. Finally the
selected solvent was ethyl acetate. DPZ and its metabolites
are basic compounds and, moreover DPZ may racemize in
aqueous solution [5, 9, 10]. Thus, to obtain the best recovery
in the absence of racemization, the extraction of DPZ, 5-ODD
and 6-ODD, as well as the isolated enantiomers, was evaluat-
ed in pH 7.0, 9.0, and 10.
The best recovery results were observed at pH 10, but
racemization was observed under this pH that prohibits
its use. In turn, at pH 7 and 9, the racemization was
negligible, but the recovery value was higher at pH 9, so this
pH was selected for further validation of the method
(Table 1).
The recoveries of DPZ, 5-ODD and 6-ODD were approx-
imately 97, 95, and 92 % for both enantiomers, respectively.
The RSD (%) for the recoveries of all analytes were lower
than 10.1 % (Table 2).
Linear regression analyses were performed by plotting
the peak area of the enantiomer (y) versus theoretical enan-
tiomer concentrations (x). The method proved to be linear
over the concentration range of 10010,000 ng mL
1
for
each DPZ enantiomer, with correlation coefficient (r)
0.9985 (Table 2) and over the concentration range of 100
5,000 ng mL
1
for each 5-ODD and 6-ODD enantiomer,
with correlation coefficient (r)0.9951. The linearity of the
method was also confirmed by the lack-of-fit test (Table 2).
Fig. 3 CD spectra for the DPZ
(a), 5-ODD (b), and 6-ODD
(c) enantiomers
Table 1 Influence of the biotransformation process on the racemization and sample pH on the recovery and racemization
Analytes (n03)
a
ee initial
b
pH 7.0 pH 9.0 pH 10.0 Biotransformation
c
ee (%) Recovery (%) ee (%) Recovery (%) ee (%) Recovery (%) ee (%)
(S)-DPZ 99.30.3 99.10.4 71.04.4 98.80.9 95.24.4 90.21.1 96.55.3 96.50.1
(R)-DPZ 98.70.4 98.21.1 72.04.8 97.80.7 95.94.3 89.71.3 97.15.7 97.00.2
(S)-5-ODD 98.30.8 97.90.7 65.36.8 97.30.2 93.16.2 88.50.9 94.26.7 95.30.2
(R)-5-ODD 99.30.7 98.90.4 63.36.6 98.20.5 92.56.4 89.30.8 93.96.8 95.50.6
(S)-6-ODD 98.70.2 98.50.6 55.18.3 97.90.6 89.17.8 90.31.6 92.17.9 97.40.5
(R)-6-ODD 99.30.3 98.70.5 56.79.9 98.00.7 90.28.2 91.81.5 93.48.3 97.60.7
DPZ donepezil, 6-ODD 6-O-desmethyl-donepezil, 5-ODD 5-O-desmethyl-donepezil
a
Number of determinations
b
Determined after enantiomer purification
c
Determined after seven biotransformation days
262 T. Barth et al.
The lowest concentration quantified (LOQs) by the vali-
dated method was 100 ng mL
1
for all analytes. The RSD (%)
and RE (%) were lower than 20 % (Table 3).
The precision and accuracy of the method were evaluated
by within and between-day assays. The RSDs and relative
errors were lower than 15 % (Table 4).
The freezethaw and short-term room temperature stabil-
ity of DPZ, 5-ODD, and 6-ODD enantiomers in Czapek
culture medium were evaluated by comparison of the peak
areas of the stability test samples with the peak areas from
freshly prepared samples and showed RSDs and relative
errors lower than 15 % (Table 5). Moreover, DPZ was stable
during 168 h under the biotransformation conditions (RSD
and RE<15 %), and no degradation of DPZ could be ob-
served during this period. Daily, during the period of bio-
transformation (7 days), the controls with pure enantiomers
of DPZ, 5-ODD, and 6-ODD dissolved in Czapek culture
medium were extracted and analyzed by HPLC. The ee of
the pure enantiomers subjected to seven biotransformation
days were compared with the initial ee, i.e., after enantiomer
separation and negligible racemization was observed,
around 3 % (Table 1).
Concerning the selectivity evaluation, the studied fungi
did not produce any secondary metabolite that presented
retention time close to those of DPZ, 5-ODD, and 6-ODD
enantiomers (Fig. 2c, d), and the Czapek culture medium
does not supply any interfering peak (Fig. 2b).
Biotransformation of DPZ using fungi
The monitored biotransformation reaction of DPZ involves
the demethylation of aromatic methoxyl groups at positions
5 and 6 yelding the metabolites 5-ODD and 6-ODD, respec-
tively. The selected fungi to perform the O-demethylation
reactions were from the genus Beauveria and Cunningha-
mella. The use of these fungi in biotransformations were
reviewed [13, 27], and their ability to perform several kinds
of reactions, including O-demethylation, was described.
The biotransformation reactions of DPZ using B. bassi-
ana ATCC 7159 and C. elegans ATCC 10028B were fol-
lowed for 7 days (168 h). Aliquots were collected every 24 h
and analyzed by the HPLC method against to a calibration
curve obtained simultaneously. After extraction and analy-
sis, the analytes concentrations were obtained and graphs of
concentration vs. time of incubation were plotted (Fig. 4).
The O-demethylation reaction by the fungus B. bassiana
ATCC 7159 was observed with several compounds, such as,
anthracyclinone [28] -lactam derivative [29] and N-substi-
tuted 7-azanorbornanes [30]. The biotransformation of DPZ
by the fungus B. bassiana ATCC 7159 resulted in the
predominant formation of the metabolite 5-ODD (Fig. 4b).
In turn, the formation of 6-ODD enantiomers was also
observed but the amounts were lower than the LOQ of the
method. After 168 h of incubation, the maximal concentra-
tions of 5-ODD enantiomers observed in the culture medi-
um were 157 (2.3 % of biotransformation) and 733 ng mL
1
(8.3 % biotransformation) for (S)-5-ODD and (R)-5-ODD,
respectively. These concentrations correspond to an ee of
60.6 %of (R)-5-ODD(Fig. 4b). In addition, other unidentified
Table 2 Recovery and linearity of the method
Analyte Recovery Linearity
% RSD (%) Range (ng mL
1
) Linear equation r ANOVA lack-of-fit
F value p value
(S)-DPZ 97.8 0.2 10010,000 y0165.8x+1,154.5 0.9985 1.77 0.2
(R)-DPZ 96.8 0.7 10010,000 y0166.5x+2,545.3 0.9986 1.27 0.33
(S)-5-ODD 96.6 3.4 1005,000 y0149.1x2,769.6 0.9983 0.74 0.58
(R)-5-ODD 95.0 3.0 1005,000 y0144.9x1,013.6 0.9977 0.56 0.69
(S)-6-ODD 91.7 5.6 1005,000 y0164.3x+163.5 0.9970 0.15 0.96
(R)-6-ODD 93.9 7.1 1005,000 y0165.8x1,168.2 0.9951 0.22 0.92
DPZ donepezil, 6-ODD 6-O-desmethyl-donepezil, 5-ODD 5-O-desmethyl-donepezil, RSD relative standard deviation in percentage, r correlation
coefficient
Table 3 Limit of quantification of the method
Analyte Nominal
concentration
(ng mL
1
)
Obtained
concentration
(ng mL
1
)
Accuracy Precision
RE (%) RSD (%)
(S)-DPZ 100 102.6 2.6 5.3
(R)-DPZ 100 96.0 4.0 8.3
(S)-5-ODD 100 108.8 8.0 10.7
(R)-5-ODD 100 98.2 1.8 4.9
(S)-6-ODD 100 95.3 4.6 6.7
(R)-6-ODD 100 93.0 7.0 9.1
DPZ donepezil, 6-ODD 6-O-desmethyl-donepezil, 5-ODD 5-O-des-
methyl-donepezil, RE relative error in percentage, RSD relative stan-
dard deviation in percentage
Chiral HPLC analysis of donepezil, 5-O-desmethyl donepezil and 6-O-desmethyl donepezil 263
Table 4 Precision and accuracy of the method
Analyte Nominal Concentration
(ng mL
1
)
Within day (n05)
a
Between day (n03)
b
Concentration (ng mL
1
) RSD (%) RE (%) Concentration (ng mL
1
) RSD (%) RE (%)
(S)-DPZ 300 304.0 3.1 1.3 301.2 2.0 0.4
3,000 3,028.1 3.2 0.9 3,005.2 1.3 0.2
7,500 7,145.0 1.7 4.7 7,055.9 2.1 5.9
(R)-DPZ 300 308.1 7.6 2.7 308.8 2.9 2.9
3,000 3,068.5 3.0 2.3 2,992.9 2.3 0.2
7,500 7,252.9 2.9 3.3 7,032.1 3.1 6.2
(S)-5-ODD 300 306.6 8.6 2.2 293.0 4.7 2.3
1,500 1,588.4 2.0 5.9 1,503.9 6.7 0.3
3,750 3,840.2 4.5 2.4 3,612.3 6.5 3.7
(R)-5-ODD 300 294.6 8.9 1.8 293.0 1.9 2.3
1,500 1,596.6 4.6 6.4 1,509.2 6.6 0.6
3,750 3,847.1 3.6 2.6 3,621.7 6.9 3.4
(S)-6-ODD 300 310.1 3.6 3.4 304.1 4.8 1.4
1,500 1,578.7 2.0 5.2 1,510.6 5.4 0.7
3,750 3,465.7 2.9 7.6 3,452.1 2.4 7.9
(R)-6-ODD 300 284.8 4.6 5.1 280.9 3.5 6.3
1,500 1,612.7 2.9 7.5 1,529.8 5.1 2.0
3,750 3,555.3 3.1 5.2 3,514.4 3.9 6.3
DPZ donepezil, 6-ODD 6-O-desmethyl-donepezil, 5-ODD 5-O-desmethyl-donepezil, RSD relative standard deviation in percentage, RE relative
error in percentage
a
Number of determinations
b
Number of days
Table 5 Short-term room temperature and freeze/thaw stability of analytes
Analyte
(n06)
Fresh sample
concentration
(ng mL
1
)
Short-term room temperature (12 h) Freeze/thaw cycles
Determined concentration
(ng mL
1
)
RSD
(%)
RE
(%)
b
Determined concentration
(ng mL
1
)
RSD
(%)
RE
(%)
b
(S)-DPZ 289.0 286.7 2.5 0.8 275.3 2.5 4.7
6,884.8 6,743.4 2.5 2.0 6,942.1 1.9 0.8
(R)-DPZ 270.2 269.3 4.0 0.3 259.0 3.0 4.3
6,781.7 6,677.7 2.6 1.5 6,859.2 2.2 1.1
(S)-5-ODD 304.6 294.1 5.6 3.4 276.8 3.1 9.1
3,317.6 3,266.7 1.3 3.1 3,342 1.6 0.9
(R)-5-ODD 280.4 280.5 4.9 0.0 261.6 9.0 6.7
3,505.6 3,404.8 3.1 2.9 3,515.7 2.6 0.3
(S)-6-ODD 303.6 275.2 7.1 9.3 264.9 3.7 12.7
3,249.0 3,318.5 3.1 2.1 3,283.0 2.7 1.0
(R)-6-ODD 283.6 277.0 12.4 2.3 265.4 9.5 6.4
3,272.5 3,333.5 3.4 1.9 3,349.4 2.2 2.3
DPZ donepezil, 6-ODD 6-O-desmethyl-donepezil, 5-ODD 5-O-desmethyl-donepezil, n number of determinations, RSD relative standard deviation
in percentage, RE relative error in percentage
264 T. Barth et al.
metabolite was also observed in the biotransformation by this
fungus. This peak can be considered a possible metabolite of
DPZ that was not studied here (peak 7, Fig. 5a).
The fungi from the genus Cunninghamella were used to
perform O-demethylation reactions in several drugs such as
muraglitazar [31], naproxen [32], pantoprazole [33], pyril-
amine [34], and verapamil [35].
The biotransformation of DPZ by the fungus C. elegans
ATCC 10028B resulted in the formation of the metabolite
(R)-6-ODD, enantioselectively. The metabolite 6-ODD is
considered the active metabolite of DPZ [8] and consists
of the predominant metabolite formed in rats [6] and
humans [7]. The formation of (R)-6-ODD was firstly ob-
served after 72 h of incubation, but the maximal concentration
of (R)-6-ODD in the culture medium, was 1,520 ng mL
1
(15.1 %) in 168 h (Fig. 4a). This concentration corresponds to
an ee of 100 % of (R)-6-ODD (Fig. 4a), because the formation
of the enantiomer (S)-6-ODD was not observed. Therefore,
this fungus was found to be an excellent model for the pro-
duction of enantiomerically pure 6-ODDmetabolite. Figure 5b
shows typical chromatogram obtained fromthe biotransforma-
tion process employing the fungus C. elegans ATCC 10028B.
Other unidentified metabolites were also observed in the bio-
transformation by this fungus (peak 8 and 9, Fig. 5b). So the
biotransformation by the fungi B. bassiana ATCC7159 and C.
elegans ATCC 10028B deserve more studies for the isolation
and characterization of these unknown metabolites (Fig. 5a
(peak 7), b (peaks 8 and 9)).
Concluding remarks
A method for the enantioselective determination of DPZ, 5-
ODD and 6-ODD in Czapek culture medium was developed
and validated. This method is the first report for the simulta-
neous enantioselective analysis of DPZ, 5-ODD and 6-ODD,
Fig. 4 Concentration-time graphs for the biotransformation process of DPZ by fungi (a) C. elegans ATCC 10028B and (b) B. bassiana ATCC 7159.
The bars denote the standard deviation of replicate (n02)
Fig. 5 (a) Representative chromatograms for 168-h incubation of the
fungus B. bassiana ATCC 7159 with DPZ and of Czapek culture
medium incubated with B. bassiana ATCC 7159 (fungus control)
showing that this fungus did not produce any secondary metabolite
in the retention time of the analytes. (1) (S)-DPZ, (2) (S)-5-ODD, (3)
(S)-6-ODD, (4) (R)-DPZ, (5) (R)-5-ODD, (6) (R)-6-ODD, and (7)
unknown peak. (b) Representative chromatograms for the 96 h
incubation of the fungus C. elegans ATCC 10028B with DPZ and of
Czapek culture medium incubated with C. elegans ATCC 10028B
(fungus control) showing that this fungus did not produce any second-
ary metabolite in the retention time of the analytes. (1) (S)-DPZ, (4)
(R)-DPZ, (6) (R)-6-ODD, (8, 9) unknown peaks. Chromatographic
conditions are described in Liquid chromatographic conditions
Chiral HPLC analysis of donepezil, 5-O-desmethyl donepezil and 6-O-desmethyl donepezil 265
and it was successfully employed to study the biotransforma-
tion of DPZ by fungi. In addition, the biotransformation study
also demonstrates, for the first time, the enantioselectivity on
the biotransformation of DPZ, since the enantioselective bio-
transformation of this drug has not been reported yet for
mammals or other biotransformation model. In this study,
the predominant formation of (R)-5-ODD by B. bassiana
ATCC 7159 with an ee of 60.6 % and (R)-6-ODD by C.
elegans ATCC 10028B with an ee of 100 % were observed.
The biotransformation with the fungus C. elegans ATCC
10028B correlates with the metabolism in mammals and can
be used as an alternative to this model. Moreover, both fungi
can be also used to obtain these metabolites to be used in
toxicological and pharmacological studies.
Acknowledgments The authors are grateful to Fundao de Amparo
Pesquisa do Estado de So Paulo (FAPESP), Conselho Nacional de
Desenvolvimento Cientfico e Tecnolgico (CNPq), and Coordenao
de Aperfeioamento de Pessoal de Nvel Superior (CAPES) for finan-
cial support and for granting research fellowships.
Conflict of interest The authors have declared no conflict of interest.
References
1. Moffat AC, Osselton MD, Widdop B (2003) Clarkes analysis of
drugs and poisons, 3rd edn. Pharmaceutical Press, London
2. Geldmacher DS (2004) Expert Rev Neurother 4:516
3. Wilkinson D (2007) Psychiatry 7:914
4. Radwan MA, Abdine HH, Al-Quadeb BT, Aboul-Einein HY,
Nakashima K (2006) J Chromatogr B 830:114119
5. Matsui K, Oda Y, Nakata H, Yoshimura T (1999) J Chromatogr B
729:147155
6. Matsui K, Mishima M, Nagai Y, Yuzuriha T, Yoshimura T (1999)
Drug Metab Dispos 27:14061414
7. Tiseo PJ, Perdomo CA, Friedhoff LT (1998) Br J Clin Pharmacol
46:1924
8. Dooley M, Lamb HM (2000) Drug Aging 16:199226
9. Haginaka J, Seyama C (1992) J Chromatogr 577:95102
10. Matsui K, Oda Y, Ohe H, Tanaka S, Asakawa N(1995) J Chromatogr
A 694:209218
11. Lu YH, Zhang M, Meng Q, Zhang ZX (2006) Acta Pharm Sinic
41:471475
12. Venisetty VJ, Ciddi V (2003) Curr Pharm Biotechnol 4:123140
13. Asha S, Vidyavathi M (2009) Biotechnol Adv 27:1629
14. Smith RV, Rosazza JP (1983) J Nat Prod 46:7991
15. Azerad R (1999) Adv Biochem Eng Biotechnol 63:169218
16. Altria KD (1996) In: Shitani H, Polonsk J (eds) Quantitative
applications of the resolution of enantiomers by capillary electro-
phoresis. Blackie Academic and Professional, London
17. Guidance for Industry: Bioanalytical Method Validation (2001),
United States Food and Drug Administration, Silver Spring.
Available at http://www.fda.gov/downloads/Drugs/Guidance
ComplianceRegulatoryInformation/Guidances/UCM070107.pdf.
Accessed 15 Mar 2012
18. Borges KB, Borges WS, Pupo MT, Bonato PS (2008) J Pharm
Biomed Anal 46:945952
19. Barth T, Pupo MT, Borges KB, Okano LT, Bonato PS (2010)
Electrophoresis 31:15211528
20. Alviano CS, Fabiarz SR, Travassos LR, Angluster J, Souza W
(1992) Mycopathologia 119:1723
21. Wang T, Chen YW (1999) J Chromatogr A 855:411421
22. Wang T, Chen YW, Vailaya A (2000) J Chromatogr A 902:345
355
23. Kazusaki M, Kawabata H, Matsukura H (2000) J Liq Chrom Relat
Tech 23:28192828
24. Gaggeri R, Rossi D, Collina S, Manucci B, Baierl M, Juza M
(2011) J Chromatogr A 1218:54145422
25. Lmmerhofer M, Pell R, Mahut M, Richter M, Schiesel S, Zettl H,
Dittrich M, Schubert-Zsilavecz M, Lindner W (2010) J Chroma-
togr A 1217:10331040
26. Barth T, Simes RA, Pupo MT, Okano LT, Bonato PS (2011) J Sep
Sci 34:35783586
27. Grogan GJ, Holland HL (2000) J Mol Catal B: Enzym 9:132
28. Wu G, Gard A, Rosazza J (1980) J Antibiot 33:705710
29. Archelas A, Fourneron JD, Furstoss R (1988) Tetrahedron Lett
29:66116614
30. Olivo HF, Peeples TL, Ros MY, Velsquez F, Kim JW, Narang S
(2003) J Mol Catal B: Enzym 21:97105
31. Zhang D, Zhang H, Aranibar N, Hanson R, Huang Y, Cheng PT,
Wu S, Bonacorsi S, Zhu M, Swaminathan A, Humphreys G (2006)
Drug Metab Dispos 34:267280
32. Zhong DF, Sun L, Liu L, Huang HH (2003) Acta Pharm Sinic
24:442447
33. Xie ZY, Huang HH, Zhong DF (2005) Xenobiotica 35:467477
34. Hansen EB Jr, Cerniglia CE, Korfmacher WA, Miller DW, Heflich
RH (1987) Drug Metab Dispos 15:97106
35. Sun L, Huang HH, Liu L, Zhong DF (2004) Appl Environ Microbiol
70:272227277
266 T. Barth et al.
Copyright of Analytical & Bioanalytical Chemistry is the property of Springer Science & Business Media B.V.
and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for individual use.