Studies on hydrolytic and oxidative metabolic pathways of anhydroecgonine methyl ester (methylecgonidine) using microsomal preparations from rat organs

Article abstract of DOI:10.1021/tx0255828

During smoking of cocaine-base (crack), anhydroecgonine methyl ester (AEME, methyl-ecgonidine) is formed in large amounts as a pyrolysis product of cocaine and is absorbed in the lungs. The metabolism of AEME was studied in the present investigation using microsome preparations from rat liver, lung, kidney, and brain. Potential metabolites of AEME were synthesized and used as substrate to complement the experiments. Analysis of the incubation mixtures was performed using gas chromatography - mass spectrometry and nanoelectrospray multiple-stage mass spectrometry. Screening for metabolites was focused on postulated oxidative pathways, chemical and enzymatic hydrolysis, and ethanol dependent transesterification as known from cocaine metabolism. Enzymatic hydrolysis of AEME to anhydroecgonine (AE), which was inhibited by sodium fluoride, was found in all microsomal preparations. Liver microsomes exhibited the highest activity, brain microsomes the lowest. Anhydronorecgonine methyl ester (ANEME) and anhydroecgonine methyl ester N-oxide were identified as AEME metabolites of liver and lung microsomes only. In the presence of ethanol AEME was metabolized to anhydroecgonine ethyl ester and anhydronorecgonine ethyl ester. Further metabolism of AE or ANEME was not observed. No N-hydroxy-anhydronorecgonine derivatives were found which could represent precursors of cytotoxic metabolites as known to be formed from cocaine.

Full text of DOI:10.1021/tx0255828

Chem. Res. Toxicol. 2002, 15, 1543-1548
1543
Stu d ies on Hyd r olytic a n d Oxid a tive Meta bolic
P a th w a ys of An h yd r oecgon in e Meth yl Ester
(Meth ylecgon id in e) Usin g Micr osom a l P r ep a r a tion s fr om
Ra t Or ga n s
Anabel S. Fandin˜o, Stefan W. Toennes,* and Gerold F. Kauert
Institute of Forensic Toxicology, Center of Legal Medicine, J .W.Goethe University, Kennedyallee 104,
D-60596 Frankfurt/ Main, Germany, and Institute of Forensic Toxicology, Center of Legal Medicine,
J .W.Goethe University, Kennedyallee 104, D-60596 Frankfurt/ Main, Germany
Received J uly 19, 2002
During smoking of cocaine-base (crack), anhydroecgonine methyl ester (AEME, methyl-
ecgonidine) is formed in large amounts as a pyrolysis product of cocaine and is absorbed in
the lungs. The metabolism of AEME was studied in the present investigation using microsome
preparations from rat liver, lung, kidney, and brain. Potential metabolites of AEME were
synthesized and used as substrate to complement the experiments. Analysis of the incubation
mixtures was performed using gas chromatography-mass spectrometry and nanoelectrospray
multiple-stage mass spectrometry. Screening for metabolites was focused on postulated
oxidative pathways, chemical and enzymatic hydrolysis, and ethanol dependent transesteri-
fication as known from cocaine metabolism. Enzymatic hydrolysis of AEME to anhydroecgonine
(AE), which was inhibited by sodium fluoride, was found in all microsomal preparations. Liver
microsomes exhibited the highest activity, brain microsomes the lowest. Anhydronorecgonine
methyl ester (ANEME) and anhydroecgonine methyl ester N-oxide were identified as AEME
metabolites of liver and lung microsomes only. In the presence of ethanol AEME was
metabolized to anhydroecgonine ethyl ester and anhydronorecgonine ethyl ester. Further
metabolism of AE or ANEME was not observed. No N-hydroxy-anhydronorecgonine derivatives
were found which could represent precursors of cytotoxic metabolites as known to be formed
from cocaine.
urine sample from a cocaine user (6), however, it cannot
be excluded that this metabolite was an artifact produced
from norcocaine during analysis by gas chromatography.
In tr od u ction
During crack smoking, cocaine (COC)1 and its main
thermal breakdown product anhydroecgonine methyl
ester (AEME, methylecgonidine) are absorbed in the
lungs (1). Though there are some reports on potential
metabolic products of AEME, the metabolism of AEME
has not yet been studied. It is known that AEME can be
hydrolyzed enzymatically to anhydroecgonine (AE, ecgoni-
dine) in human liver homogenate (2). In vitro degradation
of AEME to AE has been observed in sheep plasma (3)
and in human plasma due to butyryl cholinesterase and
nonenzymatic processes (4). Metabolites of oxidative
pathways have also been identified: anhydroecgonine
methyl ester N-oxide (AEMENO) has recently been
detected in incubations with rat liver microsomes and
in blood and urine samples from crack users (5). Anhy-
dronorecgonine methyl ester (ANEME) was found in a
AEME may be metabolized by similar routes as COC.
The main in vivo metabolism of COC is its hydrolysis to
benzoylecgonine by liver carboxylesterase (in humans:
hCE-1). In the presence of ethanol, this enzyme also
catalyzes the ethyl transesterification to cocaethylene (7).
Dean et al. (8) reported that the supernatants of rat liver,
lung, kidney, and heart homogenates also exhibited
cocaine methyl esterase and ethyl transferase activities.
COC is also metabolized through oxidative pathways
leading to hepatotoxic metabolites through a multistep
cytochrome P450-dependent N-oxidative pathway [re-
viewed by Boelsterli and Go¨ldlin (9)]. Through this
pathway, COC is N-demethylated to norcocaine, which
is further oxidized to N-hydroxynorcocaine and to nor-
cocaine nitroxide. This pathway was observed following
COC injection in mice (10-12), COC incubation with rat,
hamster or mice liver microsomes (12-14), and rat brain
microsomes (15). Pellinen et al. (16) demonstrated that
the first step of cocaine bioactivation is catalyzed by the
P450 3A enzymes in both murine and human liver
microsomes and that cocaine-induced liver injury in mice
may be prevented by the administration of inhibitors of
these enzyme isoforms.
* To whom correspondence should be addressed. Telephone:
+49 69 6301 7561. Fax: +49 69 6301 7531. E-mail: toennes@
em.uni-frankfurt.de.
1
Abbreviations: AEME, anhydroecgonine methyl ester; AEME-d3,
anhydroecgonine methyl ester-d3; AE-d3, anhydroecgonine-d3; AE,
anhydroecgonine; ANEME, anhydronorecgonine methyl ester; ANE,
anhydronorecgonine; AEMENO, anhydroecgonine methyl ester N-
oxide; AENO, anhydroecgonine N-oxide; AEEE, anhydroecgonine ethyl
ester; ANEEE, anhydronorecgonine ethyl ester; COC, cocaine; MSTFA,
N-methyl-N-(trimethylsilyl)trifluoroacetamide; NaF, sodium fluoride;
nanoESI-MSn, nanoelectrospray multiple-stage mass spectrometry.
10.1021/tx0255828 CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/07/2002

1544 Chem. Res. Toxicol., Vol. 15, No. 12, 2002
Fandin˜o et al.
out for 60 min at 37 °C. The reaction was stopped by adding
500 µL of acetonitrile followed by centrifugation for 10 min at
14000g. The supernatant was used for GC/MS and nanoESI-
MSn analysis.
The amounts of AE formed in the presence of sodium fluoride
(NaF) were used as control for the nonenzymatic hydrolysis as
200 mM NaF completely inhibit the esterases according to Dean
et al. (8). Enzymatic hydrolysis was calculated by subtraction
of the nonenzymatic component of the reaction.
Tests for ethyl transesterification of AEME in rat liver
microsomes were also performed in the absence and presence
of 200 mM NaF with 5 mM AEME and 250 mM ethanol,
according to the optimal incubation conditions for COC as
determined by Boyer and Petersen (18).
Syn th esis of AEMENO a n d AENO. AEMENO and AENO
were synthesized according to the method of Cymerman and
Purushothaman (19) for synthesis of N-oxides from tertiary
amines. Due to the thermolability of these compounds, analysis
was performed by nanoESI-MSn, and the identity of the
synthesis products was confirmed by interpretation of the
product ion spectra as described by Fandin˜o et al. (5).
Syn th esis of AEEE. AEEE was synthesized according to the
method of Kozikowski et al. (20). To 3 g of COC was added 36
mL of 1 N HCl, and the solution was heated at 100 °C for 21 h.
The resulting solution was cooled to room temperature and
extracted three times with 9 mL of dichloromethane. The
organic phase was discarded and the aqueous phase was
lyophilized. POCl3 (12 mL) was added to the dry residue, and
the mixture was heated at 100 °C for 20 h. The excess POCl3
was removed under reduced pressure, and the receptacle with
the resulting dark oil was cooled using a mixture of liquid
nitrogen and ethanol (-40 °C). The dark oil was then carefully
treated with 7.5 mL of anhydrous ethanol and was stirred for
67 h. After the oil was warmed to room temperature, it was
stirred for another 20 min and concentrated under reduced
pressure. The residue was dissolved in 6 mL of water (alkalin-
ized with ammonia, pH 10) and extracted with dichloromethane.
The aqueous phase was discarded, and the dried organic phase
(sodium sulfate) was concentrated and distilled under reduced
pressure. The synthesized AEEE was purified and isolated by
high performance liquid chromatography [Bruker LC21 pump
(Bremen, Germany) with Merck-Hitachi UVD detector L-4000
(Darmstadt, Germany)] using a 16 × 250 mm Eurospher RP
18 column (100 Å, 7 µm particle size, Macherey & Nagel,
Dueren, Germany). Elution was performed with a gradient of
0.1% trifluoroacetic acid in water (A) and methanol (B) from 0
to 35% within 35 min and to 100% for another 10 min. Fractions
were collected during 7 min intervals. AEEE (absorption
maximum at 220 nm) was eluted between 28 and 35 min and
extracted with ether at pH 10. The identity of the substance
was confirmed by GC/MS analysis (Figure 1a).
Syn th esis of ANE, ANEME, a n d ANEEE. To 1 mg of
norcocaine in a 20 mL vial was added 2 mL of concentrated HCl,
and the sealed vial was heated at 110 °C for 24 h. After the
solution cooled to room temperature, it was extracted twice with
1 mL of dichloromethane to remove benzoic acid. The organic
phase was discarded, and the aqueous phase evaporated to
dryness under a stream of air. The identity of the product ANE
was confirmed after trimethylsilylation using MSTFA by GC/
MS analysis (Figure 1b).
For the preparation of ANEME or ANEEE, ANE was dis-
solved in methanol (for the synthesis of the methyl ester) or in
ethanol (for the synthesis of the ethyl ester), and 0.25 mL of
boron trifluoride ethyl etherate (BF3‚Et2O) was added. The
sealed vials were incubated at 50 °C for 4 h and after cooling to
room temperature were evaporated to dryness. The identity of
both substances was confirmed after trimethylsilylation using
MSTFA by GC/MS analysis (Figure 1, panels c and d).
Extr a ction P r oced u r e for th e An a lysis of AEME a n d Its
Meta bolites in Micr osom e In cu ba tion Mixtu r es. To the
microsome incubation mixture were added 4 mL of 0.1 M HCl
and 100 ng of internal standards (AE-d3, AEME-d3), and they
The aim of the present paper was a systematic study
of the metabolism of AEME, AE, and ANEME using
microsome preparations from liver, lung, kidney, and
brain of rats.
Exp er im en ta l P r oced u r es
Ch em ica ls, Refer en ce Sta n d a r d s. The solutions of the
reference standards (1 mg/mL acetonitrile) AEME, AE, and
norcocaine were from Cerilliant (Promochem, Wesel, Germany),
AEME-d3 and AE-d3 were from Lipomed (Arlesheim, Switzer-
land), and the derivatization reagent MSTFA was from Mach-
erey & Nagel (Dueren, Germany). All other reagents and organic
solvents were from Merck (Darmstadt, Germany), and bio-
chemicals were obtained from Sigma-Aldrich (Deisenhofen,
Germany) in the highest purity available.
Reference substances for the postulated metabolites ANEME,
ANE, AEMENO, AENO, AEEE, and ANEEE were synthesized
as described below.
Ap p a r a tu s. Gas chromatographic-mass spectrometric (GC/
MS) analysis was performed on a Hewlett-Packard (Waldbronn,
Germany) HP6890 GC equipped with an autosampler HP6890
ALS and interfaced to a HP5973 MSD. The GC conditions were
as follows: Macherey & Nagel OPTIMA-1-MS capillary column
(30 m × 250 µm i.d., 0.25 µm film thickness), helium as carrier
gas with a flow rate of 0.7 mL/min, splitless injection at 250 °C
injection port temperature, temperature program: 55 °C for 2
min, with 20 °C/min to 170 °C, with 12 °C/min to 310 °C and
hold for 5 min. The MS conditions were as follows: 280 °C
transferline temperature, 70 eV ionization energy, and 250 °C
ion source temperature. Data analysis was performed on a
Windows computer using HP ChemStation software (Rev.
B.01.00).
Nanoelectrospray multiple-stage mass spectrometric (na-
noESI-MSn) experiments were performed with a LCQ quadru-
pole ion-trap mass spectrometer (Finnigan MAT, San J ose, CA)
equipped with a nanoelectrospray ion source (Protana, Odense,
Denmark). Analysis was performed in the positive ion ESI mode,
using self-made and gold-coated glass capillaries (opening 1-5
µm), in which the liquid samples (4 µL) were introduced. The
capillaries were made with a Micropipet Puller Model P-97
(Sutter Instruments Co., Novato, CA), and gold covering was
performed with an EMITECH K550.
In cu ba tion s w ith Ra t Tissu es Micr osom es. Three male
Sprague-Dawley rats (3 months old) were anesthetized with
enflurane, and liver, kidneys, lungs, and brain were excised and
immediately frozen in liquid nitrogen. After homogenization of
the pooled organs in 3 vol of 0.1 M phosphate buffer (pH 7.4),
microsomes were isolated by centrifugation at 10000g (30 min)
and at 100000g (70 min), resuspended in 0.1 M phosphate buffer
(pH 7.4), and again centrifuged at 100000g (70 min). The pellets
formed were stored at -80 °C until use. Protein concentration
was determined using the Bio-Rad protein assay kit (Bio-Rad,
Munich, Germany) and applying bovine serum albumin as
standard. The microsome preparation from liver contained 7.9
mg of protein/mL, from kidneys 4.7 mg/mL, from lungs 2.8 mg/
mL, and from brain 2.1 mg/mL, respectively. Incubation of
AEME or its metabolites with microsomes was essentially
performed as described by Kraemer et al. (17). The incubation
mixture consisted of an amount of microsome preparation
equivalent to 0.75 mg of protein, 150 µM substrate (AEME, AE,
ANEME, or ANE as 1 mg/mL solution in acetonitrile), 1.2 mM
NADP, 2 units of isocitrate dehydrogenase, 5 mM isocitrate, and
5 mM magnesium chloride in 0.1 M phosphate buffer (pH 7.4)
up to a total volume of 500 µL. Incubations of AEME with liver
microsome preparation were carried out for 15, 30, and 60 min
at 37 °C to find out if product formation was linear. Incubations
of AEME were carried out for 60 min at 37 °C with microsome
preparations from each tissue with or without 200 mM sodium
fluoride (n ) 5 each) and incubations without substrate or
without microsomes were used as control. Incubations of AE,
ANEME, or ANE with liver microsome preparation were carried

Microsomal Metabolism of Anhydroecgonine Methyl Ester
Chem. Res. Toxicol., Vol. 15, No. 12, 2002 1545
F igu r e 1. Electron impact mass spectra of trimethylsilylated (a) anhydronorecgonine methyl ester, (b) anhydroecgonine ethyl ester,
(c) anhydronorecgonine ethyl ester, (d) anhydronorecgonine.
were vortex-mixed. Extraction was performed using 3 mL of
Bond Elut Certify HF 300 mg of solid-phase extraction car-
tridges from Varian (Darmstadt, Germany) and the extraction
robot RapidTrace from Zymark (Idstein, Germany). The extrac-
tion protocol was as follows: conditioning with 2 mL of methanol
and 3 mL of 0.1 M HCl, loading of the sample onto the column
at 1 mL/min, rinsing with 2 mL of 0.25 M acetic acid and 3 mL
of methanol at 1.5 mL/min, and elution of the analytes with 3
mL of freshly prepared methanol-ammonium hydroxide solution
(98:2, v/v) at 1 mL/min. The extracts were evaporated to dryness
using the Zymark TurboVap LV with 25 °C bath temperature,
and the residues were transferred with 2 × 100 µL acetone into
1.5 mL reaction tubes. The solutions were centrifuged for 20
min at 14000g, and the supernatant was transferred into
autosampler vials, and 50 µL of 0.1 M hydrochloric acid in
2-propanol was added and evaporated at 25 °C under a nitrogen
stream.
comparison of the incubation mixtures with and without NaF,
the relative responses were used (analyte area/AEME-d3 area).
Na n oESI-MSn An a lysis. For nanoESI-MSn analysis, the dry
extracts were dissolved in 100 µL of methanol/water (1:1, v/v)
containing 0.1% formic acid. The capillary and tube lens voltages
were optimized for analysis of AEMENO and were set to 13.54
and 5 V, respectively. The heated metal-transfer capillary was
held at 150°C, and the ionspray voltage at 1 kV. Positive-ion-
spectra were averaged over 20 scans, each scan consisting of
three microscans. For consecutive MSn experiments, the relative
collision energy in the ion-trap was varied between 20 and 25%
corresponding to the LCQ software settings defining the am-
plitude of the resonance excitation AC voltage, depending on
the chemical nature of the precursor ion. Analysis was per-
formed between m/z 50 and 400 in full scan mode.
Resu lts
Qu a lita tive a n d Qu a n tita tive Assa y of AEME a n d Its
Meta bolites w ith GC/MS. The dry extracts of the microsome
incubation mixtures were derivatized with 40 µL of MSTFA for
30 min at 90 °C and 1 µL was injected. GC/MS analysis was
first performed in SCAN mode for screening and in selected ion
monitoring (SIM) mode for quantifying of the following analytes
(internal standard first, quantifiers underlined): AEME-d3 m/z
155, 184, 125, AEME m/z 152, 181, 122, ANEME m/z 210, 224,
239, AEEE m/z 166, 195, 122, ANEEE m/z 224, 238, 253, AE-
d3 m/z 213, 227, 242, AE m/z 210, 224, 239, ANE m/z 268, 282,
297. For calibration, phosphate buffer (pH 7.4) was spiked with
AEME and AE to 0, 1, 2, 5, 10, 20, 30, 50, 100, 200, 500, and
1000 ng/mL. One milliliter was analyzed as described above,
and a linear regression analysis was performed (analyte area/
internal standard area). The responses were linear over the
range tested with regression coefficients being greater than
0.994. The limit of detection for AEME was 1 ng/mL and for
AE 5 ng/mL. The extraction yield was 81(3% for AEME (100
ng/mL, n ) 5) and 47 ( 7% for AE (100 ng/mL, n ) 5). The
intra-day precision for AEME analysis was 3% and for AE 5%
(250 ng/mL serum, n ) 5), and the corresponding accuracies
were 3% for AEME and 8% for AE. The inter-day precision for
AEME was 5% and for AE 9% (250 ng/mL serum, n ) 5 days).
Due to the lack of pure reference standards, a calibration with
ANEME, AEEE, ANEEE, and ANE was not possible. For
Micr osom a l Hyd r olytic Meta bolism . Nonenzymatic
hydrolysis of AEME to AE as exhibited in NaF-treated
controls was only 1.1 ( 0.1% (0.83 ( 0.09 nmol/incuba-
tion) in all tissue microsomes (n ) 20, 4 tissues with n )
5 each) after 60 min of incubation. The formation of AE
over 60 min of incubation with liver microsome prepara-
tion was linear (regression coefficient 0.993). The amount
of AE produced in any incubation sample was reproduc-
ible with variation coefficients between 4 and 13% for
AEME and between 4 and 11% for AE (n ) 5 each). The
amounts were significantly higher without NaF (p < 0.01)
than in NaF containing incubation mixtures which
indicated that all organs contained esterase activity.
However, marked differences in these enzyme activities
of the microsome preparations from the four organs were
noted (Table 1). The enzymatic activities per milligram
of microsomal protein decreased in the order liver > lung
> kidney > brain, where liver exhibited 83-fold more
activity than brain microsomes, lung 15-fold, and kidney
3-fold. If the protein yields of the microsome preparations
(liver 1.4-3.3-fold higher than the others) and the organ
weights (liver 7.2-8.7-fold higher than the others) are

1546 Chem. Res. Toxicol., Vol. 15, No. 12, 2002
Fandin˜o et al.
Ta ble 1. En zym a tica lly P r od u ced An h yd r oecgon in e (AE)
d u r in g In cu ba tion of An h yd r oecgon in e Meth yl Ester
w ith Ra t Tissu e Micr osom es (n ) 5 ea ch , AE
con cen tr a tion cor r ected for th e n on en zym a tica lly
p r od u ced AE of th e con tr ol in cu ba tion s w ith 200 m M
Na F )a
and m/z 168 (subsequent fragmentation gave a mixed
spectrum containing the product ions of AE and ANEME).
The ion with m/z 198 was not found in the controls or
kidney and brain microsome incubations but only with
liver and lung microsomes and was identified to be
AEMENO by its product ion spectrum. In the analysis
of liver microsome incubation mixtures with AE as
substrate using GC/MS and nanoESI-MSn, no ANE or
AENO were detectable.
As it is known for norcocaine, the metabolism to
N-hydroxynorcocaine and further oxidations are respon-
sible for the cocaine-induced hepatotoxicity (21). The
extracts of the incubation mixtures were screened for
further metabolites of AEME, however, neither in the
incubation mixtures with AEME nor in those with the
hypothetical precursors ANEME or ANE was further
metabolism to N-hydroxylated derivatives observed.
microsome
preparation
organ weight
(g)
protein yield
(mg/g)
enzymatic AE
(pmol/min/mg)
liver
lung
kidney
brain
50.00
5.85
6.94
5.75
20.5
8.5
14.3
6.5
748 ( 29
139 ( 6
30 ( 4
9 ( 3
a
The weights of the pooled organs from three rats and the
protein yields of the microsomal preparations are given for
estimation of relative hydrolyzing capacity of the four organs.
multiplied the much higher esterase content of liver is
evident (liver 2279-fold compared to brain, lung 21-fold
and kidney 9-fold).
Transesterification with ethanol is another aspect of
the hydrolytic metabolism and was studied with mi-
crosomes from liver only. In the incubation mixtures with
AEME and ethanol, the metabolites AE, ANEME, AEEE,
and ANEEE were identified using GC/MS, whereas in
the presence of the esterase inhibitor NaF, AEEE and
AE were hardly detectable. A quantification of the
identified metabolites was not possible due to the lack
of pure reference standards.
Micr osom a l Oxid a tive Meta bolism . In the incuba-
tion mixtures of kidney and brain microsomes, no me-
tabolites of oxidative pathways were detected. N-De-
methylation was observed with liver microsomes and
could be increased by inhibition of the hydrolytic me-
tabolism with NaF. In lung microsomes, N-demethylation
of AEME was observed in the presence of NaF only.
Analysis of thermolabile metabolites (e.g., N-oxides)
using nanoESI-MSn in the full scan mode revealed an
ion with m/z 198 in addition to m/z 182 (precursor AEME)
Discu ssion
During the process of crack smoking, the cocaine
pyrolysis product AEME is absorbed in considerable
amounts together with cocaine. The present paper de-
scribes the results of metabolic studies using rat liver,
lung, kidney, and brain microsome preparations. As
AEME contains some of the structural features similar
to cocaine, which are the target of metabolizing enzymes,
similar metabolic reactions could be expected. It is known
that COC undergoes hydrolysis and transesterification
(7, 8), aryl oxidation (22), N-demethylation, and further
N-oxidation (9).
In the incubation mixtures of rat liver and lung
microsomes, the AEME metabolites AE, ANEME, and
AEMENO were identified. In the incubation mixtures of
rat kidney and brain microsomes, only AE was detected,
indicating that these organs contained negligible mi-
crosomal oxidizing capacity for AEME. However, it
F igu r e 2. Metabolic pathways of anhydroecgonine methyl ester. Solid arrows show proven pathways, dashed arrows indicate
postulated pathways.

Microsomal Metabolism of Anhydroecgonine Methyl Ester
Chem. Res. Toxicol., Vol. 15, No. 12, 2002 1547
cannot be excluded that in specific cells (e.g., proximal
or distal tubular cells in the kidney) AEME may be
metabolized by P450 enzymes, which was not detected
under the conditions used (microsomes were prepared
from the whole organs). In the incubation mixtures of
liver microsomes with AE, ANEME, or ANE, only the
parent compound was detected, and no other metabolites
were observed. Figure 2 shows the proposed metabolic
pathways for AEME according to our study results.
The highest yield of the hydrolysis product AE was
found in liver microsome incubates. If microsomal protein
yield, organ weight, and esterase activity of the four
organs are considered, it appears that liver has more than
100-fold more hydrolyzing capacity than the lung, kidney,
or brain (cf. Table 1). The nonenzymatic hydrolysis at
physiological pH was negligible in comparison to the
enzymatic route. These findings are in agreement with
data from investigations of mammalian carboxylesterases
by Satoh and Hosokawa (23) who found that the highest
hydrolase activity of all tissues studied occurred in the
liver. In a study on the stability of AEME in human
serum (4), slow enzymatic and chemical hydrolysis of
AEME was found where 50% of the initial AEME was
hydrolyzed to AE within 5 days. It can be concluded that
from the four studied tissues, the liver plays the most
important role in the hydrolytic metabolism of AEME in
vivo.
Like cocaine is metabolized to cocaethylene in the
presence of ethanol AEME also undergoes ethyl trans-
esterification to form AEEE. The addition of the esterase
inhibitor NaF decreased the formation of AEEE, indicat-
ing that a microsomal esterase is involved. Whether the
new oxidative metabolite ANEEE is formed from ANEME
by ethyl transesterification or from AEEE by N-de-
methylation has still to be investigated.
The enzyme responsible for methyl ester cleavage of
cocaine in humans has been shown to be the hepatic
microsomal carboxylesterase isoform 1 (7). It was also
found that the removal of the N-methyl group decreases
the affinity of COC, benzoylecgonine, or cocaethylene to
the enzyme, since the kinetic inhibition constants (Ki)
were substantially higher for the N-demethylated deriva-
tives norcocaine, benzoylnorecgonine, and norcocaethyl-
ene. This can explain our findings that ANE, the hypo-
thetical hydrolysis product of ANEME and ANEEE, was
detected neither in incubates of AEME with rat tissue
microsomes with and without the addition of ethanol nor
in the incubate of ANEME with rat liver microsomes.
These results indicate that ANEME has a very low
affinity to microsomal esterases.
ing analysis (5). Therefore, it was not possible to perform
a material balance and assess the relative importance
of oxidation vs hydrolysis.
In the liver microsomal incubation mixtures with AE
as substrate, no oxidative metabolites (ANE, AENO, or
others) were detected, which confirms findings of
Thompson et al. (10) who reported that the polar me-
tabolites benzoylecgonine and ecgonine are N-demeth-
ylated by liver microsomes much more slowly than
cocaine.
In the present study, further oxidative metabolism of
ANEME was not observed in any of the incubation
mixtures; therefore, a metabolic pathway of AEME
leading to hepatotoxic metabolites, like that of cocaine,
could not be established. However, up to now nothing is
known on the toxicological properties of AEMENO.
Ack n ow led gm en t. The authors acknowledge the
support of Prof. Dr. M. Karas from the Division for
Instrumental Analytical Chemistry, J .W. Goethe Uni-
versity, Frankfurt/Main, Germany, in the nano-ESI-MSn
experiments and the help of Thomas Kraemer, Michael
Walther, and Markus Thiel with the microsome incuba-
tions.
Refer en ces
(1) Nakahara, Y., and Ishigami, A. (1991) Inhalation efficiency of free-
base cocaine by pyrolysis of ‘crack’ and cocaine hydrochloride. J .
Anal. Toxicol. 15, 105-109.
(2) Paul, B. D., McWhorter, L. K., and Smith, M. L. (1999) Electron
ionization mass fragmentometric detection of urinary ecgonidine,
a
hydrolytic product of methylecgonidine, as an indicator of
smoking cocaine. J . Mass. Spectrom. 34, 651-660.
(3) Scheidweiler, K. B., Shojaie, J ., Plessinger, M. A., Wood, R. W.,
and Kwong, T. C. (2000) Stability of methylecgonidine and
ecgonidine in sheep plasma in vitro. Clin. Chem. 46, 1787-1795.
(4) Fandino, A. S., Toennes, S. W., and Kauert, G. F. (2002) Studies
on in-vitro degradation of anhydroecgonine methyl ester (meth-
ylecgonidine) in human plasma. J . Anal. Toxicol. (in press).
(5) Fandino, A. S., Karas, M., Toennes, S. W., and Kauert, G. (2002)
Identification of anhydroecgonine methyl ester N-oxide, a new
metabolite of anhydroecgonine methyl ester, using electrospray
mass spectrometry. J . Mass. Spectrom. 37, 525-532.
(6) Zhang, J . Y., and Foltz, R. L. (1990) Cocaine metabolism in man:
identification of four previously unreported cocaine metabolites
in human urine. J . Anal. Toxicol. 14, 201-205.
(7) Brzezinski, M. R., Spink, B. J ., Dean, R. A., Berkman, C. E.,
Cashman, J . R., and Bosron, W. F. (1997) Human liver carboxy-
lesterase hCE-1: binding specificity for cocaine, heroin, and their
metabolites and analogs. Drug Metab. Dispos. 25, 1089-1096.
(8) Dean, R. A., Zhang, J ., Brzezinski, M. R., and Bosron, W. F. (1995)
Tissue distribution of cocaine methyl esterase and ethyl trans-
ferase activities: correlation with carboxylesterase protein. J .
Pharmacol. Exp. Ther. 275, 965-971.
(9) Boelsterli, U. A., and Go¨ldlin, C. (1991) Biomechanisms of cocaine-
induced hepatocyte injury mediated by the formation of reactive
metabolites. Arch. Toxicol. 65, 351-360.
(10) Thompson, M. L., Shuster, L., and Shawk, K. (1979) Cocaine-
induced hepatic necrosis in mice, the role of cocaine metabolism.
Biochem. Pharmacol. 28, 2389-2395.
(11) Misra, A. L., Pontani, R. B., and Vadlamani, N. L. (1979)
Metabolism of norcocaine, N-hydroxy norcocaine and cocaine-N-
oxide in the rat. Xenobiotica 9, 189-199.
(12) Shuster, L., Casey, E., and Welankiwar, S. S. (1983) Metabolism
of cocaine and norcocaine to N-hydroxynorcocaine. Biochem.
Pharmacol. 32, 3045-3051.
(13) Rauckman, E. J ., Rosen, G. M., and Cavagnaro, J . (1982)
Norcocaine nitroxide. a potential hepatotoxic metabolite of co-
caine. Mol. Pharmacol. 21, 458-463.
(14) Kloss, M. W., Cavagnaro, J ., Rosen, G. M., and Rauckman, E. J .
(1982) Involvement of FAD-containing monooxygenase in cocaine-
induced hepatotoxicity. Toxicol. Appl. Pharmacol. 64, 88-93.
(15) Kloss, M. W., Rosen, G. M., and Rauckman, E. J . (1984) Biotrans-
formation of norcocaine to norcocaine nitroxide by rat brain
microsomes. Psychopharmacology 84, 221-224.
From the oxidative metabolic pathways assumed for
cocaine, only N-demethylation to ANEME and N-oxida-
tion to AEMENO was observed. The amounts of ANEME
produced with liver microsomes were markedly larger
than with lung microsomes; in brain and kidney microso-
mal incubation mixtures no ANEME was detected, prob-
ably because the levels of the P450 in brain and of the
P450 3A isoform in kidney are only 10% of those in the
liver (15, 24). A quantitative determination of AEMENO
using nanoESI-MS in addition to the qualitative detec-
tion was not possible due to the large variation of
quantitative results (poor precision) therefore the amounts
of AEMENO produced in the two tissues could not be
compared. GC/MS results are also not reliable as the
thermal labile metabolite AEMENO is degraded to
AEME and carbomethoxycycloheptatriene isomers dur-

1548 Chem. Res. Toxicol., Vol. 15, No. 12, 2002
Fandin˜o et al.
(16) Pellinen, P., Honkakoski, P., Stenba¨ck, F., Niemitz, M., Alhava,
E., Pelkonen, O., Lang, M., and Pasanen, M. (1994) Cocaine
N-demethylation and the metabolism-related hepatotoxicity can
be prevented by cytochrome P450 3A inhibitors. Eur. J . Phar-
macol. 270, 35-43.
(17) Kraemer, T., Bickeboeller-Friedrich, J ., and Maurer, H. H. (2000)
On the metabolism of the amphetamine-derived antispasmodic
drug mebeverine: gas chromatography-mass spectrometry stud-
ies on rat liver microsomes and on human urine. Drug Metab.
Dispos. 28, 339-347.
mann, J . S. (1995) Chemistry and biology of the 2â-alkyl-3â-
phenyl analogues of cocaine: subnanomolar affinity ligands that
suggest a new pharmacophore model at the C-2 position. J . Med.
Chem. 38, 3086-3093.
(21) Ndikum-Moffor, F. M., Schoeb, T. R., and Roberts, S. M. (1998)
Liver toxicity from norcocaine nitroxide, and N-oxidative metabo-
lite of cocaine. J . Pharmacol. Exp. Ther. 284, 413-419.
(22) J indal, S. P., and Lutz, T. (1989) Mass spectrometric studies of
cocaine disposition in animals and humans using stable isotope-
labeled analogues. J . Pharm. Sci. 78, 1009-1014.
(18) Boyer, C. S., and Petersen, D. R. (1992) Enzymatic Basis for the
Transesterification of Cocaine in the Presence of Ethanol: Evi-
dence for the Participation of Microsomal Carboxylesterases. J .
Pharmacol. Exp. Ther. 260, 939-946.
(19) Cymerman, C. J ., and Purushothaman, K. K. (1970) An improved
preparation of tertiary amine N-oxides. J . Org. Chem. 35, 1721-
1722.
(23) Satoh, T. a. H. M. (1998) The mammalian carboxylesterases: from
molecules to function. Annu. Rev. Pharmacol. Toxicol. 38, 257-
288.
(24) Cummings, B. S., Zangar, R. C., Novak, R. F., and Lash, L. H.
(1999) Cellular distribution of cytochromes P-450 in the rat
kidney. Drug Metab. Dispos. 27, 542-548.
(20) Kozikowski, A. P., Saiah, M. K. E., J ohnson, K. M., and Berg-
TX0255828