Quantification of electrochemically generated iodine-containing metabolites using inductively coupled plasma mass spectrometry

Article abstract of DOI:10.1021/ac801878k

For the risk assessment of drug candidates, the identification and quantification of their metabolites is required. The majority of analytical techniques is based on calibration standards for quantification of the metabolites. As these often are not readily available, the use of inductively coupled plasma mass spectrometry (ICPMS) is an attractive alternative for drugs containing heteroatoms. In this work, the online coupling of electrochemistry (EC), liquid chromatography (LC), and ICPMS is presented. The antiarrhythmic agent amiodarone, which contains two iodine atoms, is oxidized in an electrochemical flow-through cell under N-dealkylation and deiodination. The metabolites that are generated at different EC potentials are identified by electrospray ionization (ESI) mass spectrometry, compared to those from rat liver microsomal incubations and quantified by ICPMS. Phase-optimized LC, a recent approach for high-performance isocratic separations, is used to avoid the ICPMS calibration problems known to occur with gradient separations. The potential of the complementary use of ESI-MS and ICPMS for the qualitative and quantitative analysis of drug metabolites becomes apparent in this work.

Full text of DOI:10.1021/ac801878k

Anal. Chem. 2008, 80, 9769–9775
Quantification of Electrochemically Generated
Iodine-Containing Metabolites Using Inductively
Coupled Plasma Mass Spectrometry
†
Wiebke Lohmann, Bj o¨ rn Meermann, Ines M o¨ ller, Andy Scheffer, and Uwe Karst*
Institut f u¨ r Anorganische and Analytische Chemie, Westf a¨ lische Wilhelms-Universit a¨ t M u¨ nster, Corrensstrasse 30,
4
8149 M u¨ nster, Germany
For the risk assessment of drug candidates, the identifica-
tion and quantification of their metabolites is required.
The majority of analytical techniques is based on calibra-
tion standards for quantification of the metabolites. As
these often are not readily available, the use of inductively
coupled plasma mass spectrometry (ICPMS) is an attrac-
tive alternative for drugs containing heteroatoms. In this
work, the online coupling of electrochemistry (EC), liquid
chromatography (LC), and ICPMS is presented. The
antiarrhythmic agent amiodarone, which contains two
iodine atoms, is oxidized in an electrochemical flow-
through cell under N-dealkylation and deiodination. The
metabolites that are generated at different EC potentials
are identified by electrospray ionization (ESI) mass spec-
trometry, compared to those from rat liver microsomal
incubations and quantified by ICPMS. Phase-optimized
LC, a recent approach for high-performance isocratic
separations, is used to avoid the ICPMS calibration
problems known to occur with gradient separations. The
potential of the complementary use of ESI-MS and ICPMS
for the qualitative and quantitative analysis of drug
metabolites becomes apparent in this work.
The method of choice for the analysis of drug metabolites is
nowadays electrospray ionization mass spectrometry (ESI-MS),
which provides valuable information on the molecular mass of an
analyte as well as structural information in fragmentation experi-
ments. However, for a reliable quantification with ESI-MS, refer-
ence substances are required. This is due to the fact that the
ionization efficiency is different for every single compound. In the
ideal case, the coeluting reference substance bears an isotopic
label to enable the use as an internal standard, which compensates
for temporary fluctuations in the ionization efficiency, as com-
monly observed in ESI-MS. However, in most cases, reference
substances for drug metabolites are not available without major
effort, which is due to the development of complex synthesis
routes and effective cleanup procedures. Therefore, the best way
is to use an complementary detection method, in which the
quantification is independent of the specific compound. One of
4
,5
these methods is nuclear magnetic resonance spectroscopy,
which, however, suffers from insensitivity so that relatively high
concentrations of the respective metabolite are required. Alter-
1
4
3
natively, when a C or H radioactively labeled drug is used for
metabolism studies, the metabolites can be quantified using
6
radiochemical detection techniques. The response of these
techniques is independent of the specific compound, and good
detection limits can be obtained. Nevertheless, the handling of
radioactive compounds is problematic, and the synthesis is time-
consuming and expensive. A promising alternative, which provides
good sensitivity as well as unproblematic handling conditions, is
the use of inductively coupled plasma mass spectrometry (ICPMS)
Since almost 500 years ago, when Paracelsus made his famous
1
statement “sola dosis facit venenum”, it is known that the
concentration of a substance in the body determines its effects.
Surprisingly, the reliable and robust quantification of drug
metabolites is still a challenge today. However, it is important to
know which concentration of a metabolite is present in the body.
For example, already minor amounts of a carcinogenic metabolite
may cause severe health effects, while less toxic metabolites do
not have any influence when they are present at the same
concentration. Some metabolites, such as the paracetamol me-
tabolite N-acetyl-p-benzoquinoneimine, are harmless in low con-
centrations but may cause severe side effects when the drug is
7
detection after liquid chromatographic separation. Even if ICPMS
detection is limited to drugs containing metal atoms (mostly Pt,
Au, or Ru), halogens (I, Br, Cl) or s with restrictions s other
nonmetal atoms (S, P), it is a valuable alternative for metabolism
studies of drugs containing one or more of these atoms. Moreover,
the analysis of phase II metabolites, which result from conjugation
of phase I metabolites with sulfur-containing glutathione or sulfuric
2
,3
administered in high concentrations.
(
3) Potter, W. Z.; Davis, D. C.; Mitchell, J. R.; Jollow, D. J.; Gillette, J. R.; Brodie,
B. B. J. Pharmacol. Exp. Ther. 1973, 187, 203–210
(4) Pauli, G. F.; Jaki, B. U.; Lankin, D. C. J. Nat. Prod. 2005, 68, 133–149
(5) Martino, R.; Gilard, V.; Desmoulin, F.; Malet-Martino, M. J. Pharm. Biomed.
Anal. 2005, 38, 871–891
.
*
To whom correspondence should be addressed. E-mail: uk@
.
uni-muenster.de. Fax: +49-251-83 36013.
†
Current address: Nieders ¨a chsisches Landesamt f u¨ r Verbraucherschutz and
.
Lebensmittelsicherheit (LAVES), Futtermittelinstitut Stade, Heckenweg 4-6, 21680
Stade, Germany.
(6) Zhu M., Zhang D., Skiles G. L. In Identification and Quantification of Drugs,
Metabolites and Metabolizing Enzymes by LC-MS; Chowdhury S. K., Ed.;
(
(
1) Paracelsus, Septem Defensiones, 1538.
2) Jollow, D. J.; Mitchell, J. R.; Potter, W. Z.; Davis, D. C.; Gillette, J. R.; Brodie,
Elsevier: Amsterdam, The Netherlands, 2005; pp 195-224
(7) Gammelgaard, B.; Packert Jensen, B. J. Anal. At. Spectrom. 2007, 22, 235–
249
.
B. B. J. Pharmacol. Exp. Ther. 1973, 187, 195–202
.
.
1
0.1021/ac801878k CCC: $40.75  2008 American Chemical Society
Analytical Chemistry, Vol. 80, No. 24, December 15, 2008 9769
Published on Web 11/17/2008

8
acid, is possible as well. Coupling of analytical separation
techniques such as liquid chromatography (LC), gas chromatog-
raphy, or capillary electrophoresis is possible, and ICPMS features
low limits of detection and large linear dynamic ranges. Comple-
mentary analysis with ESI-MS can provide structural information.
The hyphenation of liquid chromatography and ICPMS (LC/
ICPMS) has already been employed for quantifying halogenated
9
,10
metabolites, as for example metabolites of bromobenzoic acids
Figure 1. Structure of the antiarrhythmic agent AM and its cyto-
chrome P450 (CYP)-catalyzed metabolism to AM-Et.
11
and iodobenzoic acids or phase I and II metabolites of diclofenac
by Cl, Cl, and S detection. However, although reports on
the hyphenation of electrochemical methods with ICPMS for
3
5
37
32
12
In this work, the potential of EC/LC/ICPMS for the quantifica-
tion of the online electrochemically generated drug metabolites
shall be investigated. Therefore, the antiarrhythmic agent amio-
darone (AM, Figure 1) was selected as a model compound. One
molecule of AM contains two atoms of iodine, thus showing a
good response in ICPMS. AM undergoes extensive metabolism
in the liver under N-deethylation, hydroxylation, O-dealkylation,
13-17
online sample preparation do exist,
nothing is so far reported
about coupling an electrochemical (EC) flow-through cell for the
simulation of drug metabolism to an LC/ICPMS system (EC/
LC/ICPMS). The online coupling of EC with organic MS in
1
8,19
general is summarized in two recent review articles.
Electro-
chemistry for the simulation of oxidative drug metabolism was
first reported in the early 1980s, when the N-dealkylation of the
tricyclic antidepressant imipramine at a Pt electrode in a batch
After the introduction of the online
coupling of an electrochemical flow-through cell to a mass
the instrumentation was successively
improved, the setup was upgraded by a liquid chromatographic
separation (EC/LC/MS), and a large number of pharmaceutical
compounds were used for investigating the potential of EC/LC/
37
deiodination, and glucuronidation, with N-deethylation being the
3
8,39
most important biotransformation pathway (Figure 1).
In an
2
0,21
in vivo study in humans, the plasma concentration of N-de-
ethylamiodarone (AM-Et) in patients receiving 200 mg of AM
orally per day was determined to be 970 ± 347 ng/mL, while AM
was present in a concentration of 11 200 ± 435 ng/mL and other
reactor was reported.
22-24
spectrometer (EC/MS),
40
25
metabolites were found only in minor amounts. The quantifica-
tion of iodine-containing metabolites, which are formed upon EC
oxidation of AM in the flow-through cell, shall be performed using
the EC/LC/ICPMS setup.
2
6-36
MS to simulate oxidative drug metabolism.
(
8) Abou-Shakra, F. R.; Sage, A. B.; Castro-Perez, J.; Nicholson, J. K.; Lindon,
J. C.; Scarfe, G. B.; Wilson, I. D. Chromatographia 2002, 55, S9-S13
9) Packert Jensen, B.; Smith, C. J.; Bailey, C. J.; Rodgers, C.; Wilson, I. D.;
Nicholson, J. K. Rapid Commun. Mass Spectrom. 2005, 19, 519–524
10) Smith, C. J.; Shillingford, S.; Edge, A. M.; Bailey, C. J.; Wilson, I. D.
Chromatographia 2008, 67, 673–678
.
EXPERIMENTAL SECTION
(
Chemicals and Materials. Amiodarone hydrochloride and
magnesium chloride hexahydrate were obtained from Sigma-
Aldrich Chemie GmbH (Steinheim, Germany). Formic acid,
potassium dihydrogenphosphate, and dipotassium hydrogenphos-
phate were purchased from Fluka Chemie GmbH (Buchs,
Switzerland) and ammonium formate from Acros Organics (Geel,
Belgium). NADPH was ordered from AppliChem GmbH (Darm-
stadt, Germany). Acetonitrile for HPLC and ammonium hydroxide
solution (25%) were obtained from Merck KGaA (Darmstadt,
Germany). All chemicals were used in the highest quality
available. Water used for HPLC was purified using a Milli-Q
Gradient A 10 system and filtered through a 0.22-µm Millipak 40
filter (Millipore, Billerica, MA). Rat liver microsomes (male
Sprague-Dawley rats, pooled, protein content 20 mg/mL) were
purchased from BD Biosciences (Erembodegem, Belgium).
Instrumentation. The electrochemical equipment was ob-
tained from ESA Biosciences Inc. (Chelmsford, MA). It consisted
of a model 5021 conditioning cell and a Coulochem II potentiostat.
.
(
(
.
11) Packert Jensen, B.; Smith, C. J.; Bailey, C. J.; Rodgers, C.; Wilson, I. D.;
Nicholson, J. K. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2004,
809, 279–285.
(
12) Corcoran, O.; Nicholson, J. K.; Lenz, E. M.; Abou-Shakra, F.; Castro-Perez,
J.; Sage, A. B.; Wilson, I. D. Rapid Commun. Mass Spectrom. 2000, 14,
2
377–2384
13) Bings, N. H.; Stefanka, Z.; Mallada, S. R. Anal. Chim. Acta 2003, 479,
03–214
14) Baca, A. J.; De La Ree, A. B.; Zhou, F.; Manson, A. Z. Anal. Chem. 2003,
5, 2507–2511
15) Liljegren, G.; Forsgard, N.; Zettersten, C.; Pettersson, J.; Svedberg, M.;
Herranen, M.; Nyholm, L. Analyst 2005, 130, 1358–1368
16) Zhou, F. TrAC, Trends Anal. Chem. 2005, 24, 218–227
17) Clark, W. J.; Park, S. H.; Bostick, D. A.; Duckworth, D. C.; Van Berkel,
G. J. Anal. Chem. 2006, 78, 8535–8542
18) Diehl, G.; Karst, U. Anal. Bioanal. Chem. 2002, 373, 390–398
19) Karst, U. Angew. Chem., Int. Ed. 2004, 43, 2476–2478
20) Shono, T.; Toda, T.; Oshino, N. Drug Metab. Dispos. 1981, 9, 481–482
21) Shono, T.; Toda, T.; Oshino, N. J. Am. Chem. Soc. 1982, 104, 2639–2641
22) Hambitzer, G.; Heitbaum, J. Anal. Chem. 1986, 58, 1067–1070
23) Volk, K. J.; Lee, M. S.; Yost, R. A.; Brajter-Toth, A. Anal. Chem. 1988, 60,
20–722
24) Zhou, F.; Van Berkel, G. J. Anal. Chem. 1995, 67, 3643–3649
25) Iwahashi, H.; Ishii, T. J. Chromatogr., A 1997, 773, 23–31
26) Getek, T. A.; Korfmacher, W. A.; McRae, T. A.; Hinson, J. A. J. Chromatogr.,
A 1989, 474, 245–256
27) Jurva, U.; Wikstr o¨ m, H. V.; Bruins, A. P. Rapid Commun. Mass Spectrom.
000, 14, 529–533
28) Jurva, U.; Wikstr o¨ m, H. V.; Weidolf, L.; Bruins, A. P. Rapid Commun. Mass
Spectrom. 2003, 17, 800–810
29) Van Leeuwen, S. M.; Blankert, B.; Kauffmann, J. M.; Karst, U. Anal. Bioanal.
Chem. 2005, 382, 742–750
30) Lohmann, W.; Karst, U. Anal. Bioanal. Chem. 2006, 386, 1701–1708
31) Madsen, K. G.; Olsen, J.; Skonberg, C.; Hansen, S. H.; Jurva, U. Chem.
Res. Toxicol. 2007, 20, 821–831
32) Lohmann, W.; Karst, U. Anal. Chem. 2007, 79, 6831–6839
.
(
(
(
2
.
7
.
.
(
(
.
.
(
(
(
(
(
(
.
.
.
.
.
7
.
(33) Johansson, T.; Weidolf, L.; Jurva, U. Rapid Commun. Mass Spectrom. 2007,
(
(
(
.
21, 2323–2331.
.
(34) Jurva, U.; Holmen, A.; Groenberg, G.; Masimirembwa, C.; Weidolf, L. Chem.
Res. Toxicol. 2008, 21, 928–935.
.
(35) Madsen, K. G.; Skonberg, C.; Jurva, U.; Cornett, C.; Hansen, S. H.; Johansen,
(
(
(
T. H. Chem. Res. Toxicol. 2008, 21, 1107–1119.
(36) Lohmann, W.; Karst, U. Anal. Bioanal. Chem. 2008, 391, 79–96.
(37) Shayeganpour, A.; El-Kadi, A. O. S.; Brocks, D. R. Drug Metab. Dispos.
2
.
.
2006, 34, 43–50
(38) Trivier, J. M.; Libersa, C.; Belloc, C.; Lhermitte, M. Life Sci. 1993, 52,
PL91-PL96
(39) Soyama, A.; Hanioka, N.; Saito, Y.; Murayama, N.; Ando, M.; Ozawa, S.;
Sawada, J. Pharmacol. Toxicol. 2002, 91, 174–178
(40) Ha, H. R.; Bigler, L.; Wendt, B.; Maggiorini, M.; Follath, F. Eur. J. Pharm.
Sci. 2005, 24, 271–279
.
.
.
(
(
.
.
.
(
.
.
9770 Analytical Chemistry, Vol. 80, No. 24, December 15, 2008

The EC cell contains a glassy carbon working electrode, a Pd
counter electrode, and a Pd/H reference electrode. The HPLC
aqueous eluent and acetonitrile as organic eluent (50/50, v/v).
The column oven was tempered at 40 °C, and the autosampler
was cooled at 5 °C. A combination of 10-mm ProntoSIL 100-5-C18
SH 2 and 180-mm ProntoSIL 100-5-C18 EPS 2 as optimum
stationary phase for the separation of the EC oxidation products
and a combination of 30-mm ProntoSIL 100-5-C18 SH 2 and 180-
mm ProntoSIL 100-5-C18 EPS 2 for the separation of the microso-
mal incubation products were used.
2
separation was performed using an Agilent 1200 series HPLC
system comprising a G1379B microvacuum degasser, a G1376A
capillary pump, a G1329A autosampler, a G1316A column oven,
and a G1315B diode array detector (Agilent Technologies, Wald-
bronn, Germany). For EC/ESI-MS and EC/LC/ESI-MS measure-
ments, a QTRAP mass spectrometer from Applied Biosystems
(
Darmstadt, Germany) was used. The software used for control-
MS Conditions. For the ramping experiments using the EC/
ESI-MS setup, the following settings on the QTRAP instrument
using the linear ion trap in positive enhanced full scan mode (m/z
ling the HPLC was ChemStation (Agilent), and the QTRAP was
controlled by Analyst 1.4.1 (Applied Biosystems). For EC/LC/
ICPMS measurements, the same EC and HPLC instrumentation
was used as for the EC/LC/ESI-MS analysis, and an Elan 6000
ICPMS (PerkinElmer Sciex, Rodgau, Germany) was used for the
detection and was controlled by Elan software version 3.0 Hotfix3.
Data analysis was performed by DAx 32 bits Data Acquisition and
Data Analysis Software (Van Mierlo Software Consultancy, Eind-
hoven, The Netherlands).
200-1000) were used: nebulizer gas, 50 psi; dry gas/heating gas,
0
psi; temperature, off; ion spray voltage, 5000 V; declustering
potential, 60 V; entrance potential, 10 V; CAD gas pressure, “high”;
collision energy, 10 V. For the EC/LC/ESI-MS measurements,
the dry gas/heating gas was set to 70 psi and the temperature to
400 °C. Measurements concerning the POPLC optimization were
performed using the selected ion monitoring mode with a dwell
time of 50 ms for each m/z. On the Elan 6000 ICPMS instrument,
m/z 127 for iodine was observed, using the following settings: rf
power, 1400 W; plasma gas flow, 15 L/min; nebulizer gas flow
rate, 0.5 L/min; dwell time, 100 ms; sweeps, 10; O gas flow, 110
2
mL/min; auxiliary gas flow, 1.0 L/min. The skimmer cone as well
Electrochemical Oxidation. For the electrochemical oxida-
tion, AM in ammonium formate solution (20 mM, pH adjusted to
7
.4 with ammonium hydroxide solution) and acetonitrile (25/75,
v/v) were pumped through the EC conditioning cell model 5021
using a syringe pump model 74900 (Cole Parmer, Vernon Hills,
IL) at a flow rate of 10 µL/min. For EC/ESI-MS and EC/LC/
ICPMS experiments, the concentration of AM was 50 µM; for EC/
LC/ESI-MS, an AM concentration of 100 µM was used. For
ramping experiments, the effluent of the EC cell was directed into
the ESI source of the mass spectrometer. The potential at the
working electrode of the EC cell was ramped from 0 to 1500 mV
as the sampler cone were made of platinum, a Peltier cooled (∼5
°C) cyclonic spray chamber was used, and the sample introduction
was performed using a Meinhard nebulizer (PerkinElmer Sciex,
Rodgau, Germany).
RESULTS AND DISCUSSION
Electrochemical Oxidation of Amiodarone. To obtain first
information on the electrochemical oxidation behavior of AM, a
setup comprising an EC flow-through cell and a mass spectrom-
eter was employed. A solution containing 50 µM AM and 20 mM
aqeous ammonium formate and acetonitrile (25/75, v/v) to achieve
a sufficient conductivity of the solution as well as low adsorption
of the analytes at the working electrode was pumped with a
syringe pump at a constant flow rate through the EC cell. The
outlet of the cell was directly connected to the ESI source of the
mass spectrometer. The potential at the working electrode was
2
versus Pd/H in steps of 100 mV. For each potential, the system
was allowed to equilibrate for 6 min before recording mass spectra.
For EC/LC/ESI-MS and EC/LC/ICPMS measurements, the
potential was not ramped, but one EC potential for each measure-
ment was selected. The effluent from the EC cell was collected
in a 10 µL loop, which was mounted on a 10-port switching valve.
After the equilibration time of the total system of 6 min, the EC
effluent was injected onto the HPLC column by switching the 10-
port valve.
Rat Liver Microsomal Incubations. For rat liver microsomal
incubation of AM, the incubation mixture (total volume, 300 µL)
contained 1.3 mg/mL microsomal protein, 50 µM AM, 1.2 mM
NADPH, and 0.5 mM MgCl
After 5 min preincubation of microsomes and AM at 37 °C,
NADPH and MgCl were added to start the reaction. As negative
2
ramped between 0 and 1500 mV versus Pd/H in steps of 100
mV. Mass spectra were continuously recorded in dependence on
the EC oxidation potential.
2
in 50 mM phosphate buffer (pH 7.4).
2
At 0 mV versus Pd/H , AM itself is virtually the only detectable
compound with m/z 646 (Figure 2). Even though electrochemical
oxidation is known to occur also in the electrospray, only
negligible amounts of deiodinated AM with m/z 520 and 394 are
observed. As can be seen from the mass voltammogram resulting
after applying a potential at the EC cell, AM starts to oxidize in
2
control, incubations were made without NADPH to prevent
cytochrome P450 catalysis. The incubations were carried out for
2
h at 37 °C and were terminated by adding 300 µL of ice-cold
acetonitrile to the incubation mixture for the precipitation of
proteins. After 5 min, the samples were centrifuged and the
supernatant was directly used for LC/ESI-MS and LC/ICPMS
analysis.
HPLC Conditions. HPLC separation of the oxidation products
of AM was performed using phase-optimized liquid chromatog-
raphy (POPLC). Therefore, a POPLC-Kit 250-5 (Bischoff Analy-
sentechnik and -ger ¨a te GmbH, Leonberg, Germany) was used.
The optimization was performed according to the supplier’s
protocol. A flow rate of 0.5 mL/min was selected, with 10 mM
ammonium formate solution containing 0.02% formic acid as
2
the range of 400 mV versus Pd/H . The main oxidation product
is a species with m/z 618, corresponding to a mass loss of 28 Da
compared to AM. This mass loss can be assigned to the
elimination of one N-ethyl group, resulting in AM-Et. N-Dealky-
lation is a reaction commonly observed in EC oxidations of
2
1
N-alkylated compounds. Further oxidation of AM yields the
respective bis-N-dealkylated species with m/z 590 (AM-2Et), with
the oxidation starting in the range of 600 mV versus Pd/H2.
Furthermore, deiodination can be observed upon oxidation.
Already at low potentials, mono- and bis-deiodinated AM with m/z
520 and 394 (AM-I and AM-2I, respectively), can be observed.
Analytical Chemistry, Vol. 80, No. 24, December 15, 2008 9771

Figure 2. Mass voltammogram of AM. Mass spectra of AM after oxidation in the EC cell (equilibration time 6 min) were recorded depending
on the EC oxidation potential (vs Pd/H2). The potential was changed in steps of 100 mV.
Figure 3. Electrochemical oxidation pathway of AM.
The respective mono- and bis-deethylated species of the mono-
deiodinated AM (m/z 492 (AM-Et-I) and 464 (AM-2Et-I)) as well
as of the bis-deiodinated AM (m/z 366 (AM-Et-2I) and 338 (AM-
in the mobile phase. Since gradient elution is typically performed
under RP-HPLC conditions, the response of an early-eluting
compound is different from the response of a later-eluting species
with the same number of heteroatoms. Therefore, isocratic
separation conditions are preferred for a reliable quantification
with ICPMS. However, for most separation problems, isocratic
conditions are not the method of choice due to peak broadening
and long elution times when compounds with large differences
in polarity need to be separated. A recently introduced approach
was therefore selected, which allows the isocratic separation of
complex mixtures. In POPLC, the stationary phase can be tailored
for the separation problem, instead of the mobile phase. Due to
the avoidance of gradient elution, the use of POPLC should in
principle be well-suited for the combination with ICPMS. For AM
and its oxidation products as well as for the products from the
microsomal incubation, the optimization procedure was performed
initially with ESI-MS detection to allow the identification of the
peaks. In contrast to the EC/MS experiments, the effluent from
the EC cell was not directly introduced into the ESI source of the
mass spectrometer but directed into a 10 µL injection loop, which
2Et-2I)) were detected as well. The signal with m/z 547 was
assigned to O-dealkylated AM, which is also known from the
37
literature. Hence, the electrochemical oxidation pathway of AM,
which is shown in Figure 3, is in good accordance to the
3
7-40
biotransformation in the body.
Development of an Isocratic POPLC-Based Separation.
For the quantification of the dealkylated AM metabolites, ICPMS
was selected instead of ESI-MS, since the quantification in ICPMS
is independent of the nature of the respective species. Under
constant ionization conditions, the signal intensity in ICPMS is
only dependent on the number of heteroatoms of interest in a
certain species as well as on the concentration. Since the number
of heteroatoms in a compound is known in most cases, the
concentration of a compound can directly be calculated from the
signal intensity using a calibration function. A disadvantage of LC/
ICPMS is the dependence of the signal intensity on other than
analyte parameters, such as the concentration of organic solvent
9772 Analytical Chemistry, Vol. 80, No. 24, December 15, 2008

Figure 5. LC/ESI(+)-MS chromatogram of the POPLC-based
separation of the microsomal incubation mixture of amiodarone. The
TIC and extracted ion traces of m/z 646 (AM), 618 (AM-Et), 662
(
hydroxylated AM), and 634 (hydroxylated AM-Et) are shown.
metabolites could be detected. Different isomers of compounds
with m/z 662 and 634, respectively, were observed. All of these
species eluted at short retention times in the beginning of the
chromatographic separation. Therefore, these metabolites must
have a higher polarity than AM and AM-Et. The mass gain of 16
Da compared to AM and AM-Et indicated the introduction of one
atom of oxygen, which is in most cases due to hydroxylation in
CYP-catalyzed reactions. Hydroxylation of AM resulting in species
with m/z 662 and of AM-Et resulting in species with m/z 634 is
also in good accordance with shorter retention times and the
formation of different isomeric compounds. Even though the EC
oxidation was not able to mimic all CYP-catalyzed reactions, the
main reaction s being N-dealkylation to AM-Et s could be
simulated successfully.
Figure 4. LC/ESI(+)-MS chromatograms of the POPLC-based
separation of AM and its main oxidation products after oxidation of
AM at 1100 mV vs Pd/H2.
was mounted on a 10-port switching valve. By switching the valve,
the loop content was flushed by the HPLC pumps onto the POPLC
column. This injection technique is described elsewhere in more
3
2
2
detail. A potential of 1100 mV versus Pd/H was selected for
the electrochemical oxidation. The result from the calculation was
that the best combination of column segments was 10 mm of C18
material and 180 mm of C18-EPS material for the EC oxidation
mixture. The separation of AM and its oxidation products after
2
oxidation at 1100 mV versus Pd/H is shown in Figure 4. AM
EC/LC/ICPMS Coupling. With the optimized isocratic
separations using POPLC, the EC/LC/ICPMS system was devel-
oped. The setup is related to that for EC/LC/ESI-MS, which was
elutes at ∼18 min, and the retention times of the oxidation
products are in good accordance to the theoretically expected
elution order, with deethylated compounds being more polar than
nondeethylated species and deiodinated metabolites being more
polar than the respective iodinated species.
32
described earlier. The 50 µM AM was pumped with a constant
flow rate through the EC cell using a syringe pump. Potentials of
0
2
, 350, 600, 850, 1100, and 1350 mV versus Pd/H were applied at
Rat Liver Microsomal Incubations. To verify whether the
electrochemically generated metabolites of AM are identical with
those that are formed under cytochrome P450 (CYP) catalysis,
the results of the electrochemical oxidation were compared with
standard in vitro experiments: AM was incubated with rat liver
microsomes in the presence of NADPH for 2 h. Negative controls
were performed without using NADPH. The incubation mixtures
were analyzed by POPLC/ESI-MS in positive ion electrospray
mode. Since additional species compared to those from the EC
oxidation were formed upon microsomal incubation, an indepen-
dent POPLC optimization procedure was performed, with 30 mm
of C18 material and 180 mm of C18-EPS material resulting as an
optimum combination. The resulting chromatograms are shown
in Figure 5. The main peaks in the TIC have m/z values of 646
and 618, respectively, corresponding to AM and its mono-N-
deethylation product. Hence, and in accordance with literature
data, N-deethylation is the main oxidative biotransformation
pathway of AM. However, no bis-deethylated AM was detectable.
In contrast to the EC/LC/ESI-MS experiments, however, further
the working electrode. Injection was performed by connecting the
outlet of the EC cell to a 10-port switching valve and directing
the effluent from the cell into an injection loop, which was
mounted on the valve as was already described for the EC/LC/
ESI-MS measurements. By switching the valve, the loop content
was transferred onto the POPLC column. The EC/LC/ICPMS
setup is illustrated in Figure 6.
In the mass spectrometer, the iodine trace with m/z 127 was
monitored. Figure 7 shows the ICPMS chromatograms of AM
after electrochemical oxidation and POPLC separation on 10 mm
of C18 material and 180 mm of C18-EPS material at different EC
2
potentials (0, 350, 600, 850, 1100, and 1350 mV vs Pd/H ). Since
the iodine trace was recorded, all signals appearing in an LC/
ICPMS chromatogram arise from a component containing iodine.
However, by using ICPMS, all additional qualitative information
on the nature of the eluting iodine-containing compounds is lost.
Nevertheless, due to the previously recorded LC/ESI-MS chro-
matograms, providing information on the molecular weight of a
Analytical Chemistry, Vol. 80, No. 24, December 15, 2008 9773

Table 1. Percentage of the Peak Areas of AM, AM-Et,
AM-2Et, AM-I, AM-2Et-I, and Inorganic Iodine Species
on the Total Amount of Iodine-Containing Species
a
(Equal to 100%) at Different Potentials (vs Pd/H2)
0
mV 350 mV 600 mV 850 mV 1100 mV 1350 mV
AM
99.1
97.4
97.8
98.0
34.0
35.3
36.5
35.3
1.3
2.1
1.0
1.4
1.5
0.6
6.4
5.0
5.1
5.5
0.8
2.0
2.2
2.2
2.1
0.1
9.8
8.8
8.8
9.1
0.6
0.5
0.3
0.6
0.6
0.5
0.2
0.9
0.7
0.6
0.7
0.1
1.9
2.6
2.2
2.2
0.4
3.6
4.2
3.9
3.9
0.3
0.2
0.3
0.3
0.3
0.0
93.2
0.1
0.2
0.1
0.1
0.0
0.4
0.3
0.3
0.3
0.0
2.1
3.0
3.2
2.8
0.6
3.5
2.9
2.9
3.1
0.4
0.2
0.2
0.2
0.2
0.0
93.7
9
9.1
9.1
9
mean
abs SD
AM-Et
99.1 97.7
0.0
0.9
0.3
0.8
0.8
0.7
0.8
0.1
62.5
61.1
59.8
61.1
1.4
0
0
.9
.9
Figure 6. Schematic drawing of the EC/LC/ICPMS setup for the
quantification of amiodarone and its metabolites.
mean
abs. SD
AM-2Et
0.9
0.1
0.3
0
0
.4
.4
mean
abs SD
AM-I
0.4
0.1
0.4
0.5
0.5
0.5
0.1
0.6
0
0
.5
.5
mean
abs SD
AM-2Et-I
0.5
0.1
0
0
.5
.5
mean
0.5
0.0
79.2
abs SD
inorganic
1.2
2.8
iodine species
0
0
.9
.8
2.7
2.8
2.8
0.1
82.5
82.0
81.2
1.8
91.5
92.4
92.4
0.9
93.5
93.3
93.5
0.2
mean
abs SD
1.0
0.2
a
The peak areas of triplicate measurements (first three rows for
each compound) as well as the mean values (fourth row, boldface)
and the standard deviations (fifth row) are listed.
Quantification of Electrochemically Generated Metabo-
lites. As can be seen from the EC/LC/ICPMS chromatograms,
the yield of the oxidation to deethylated and deiodinated species
increased when higher oxidation potentials were applied. To obtain
insight into the amount of the formed oxidation products, all peak
areas of iodine-containing compounds (“iodine balance”) were
summed up and set to 100% for each chromatogram. The peak
areas of the single iodine-containing species were related to the
total amount of iodine-containing compounds. The peak areas of
AM and its oxidation products resulting from triplicate injections
as well as the means and the standard deviations are listed in
Table 1. It has to be considered that, in ICPMS, all compounds
containing two atoms of iodine (AM, AM-Et, AM-2Et) show the
same response, which is twice as high as the response of species
containing only one iodine atom (AM-I, AM-Et-I, AM-2Et-I). Thus,
the concentration of a compound containing one iodine atom, such
as AM-I, is twice as high as the concentration of a compound
containing two iodine atoms, such as AM, if both compounds have
the same peak area.
Figure 7. EC/LC/ICPMS chromatograms (m/z 127 for iodine) of AM
before oxidation (0 mV) and after oxidation at 350, 600, 850, 1100,
and 1350 mV vs Pd/H2. Each chromatogram is averaged from three
single measurements.
compound (Figure 4), the qualitative peak assignment could be
performed. AM itself is the latest eluting peak, which can be seen
at 0 mV versus Pd/H
3 min was detectable, corresponding to AM-Et resulting from
oxidation of AM in solution. AM starts to oxidize to AM-Et at ∼400
mV versus Pd/H as was observed from the EC/MS experiments.
2
. Only a small peak with a retention time of
1
2
At higher potentials, deiodinated and deethylated species were
formed. Additionally, peaks with retention times lower than 5 min
were detected. These peaks were not detected in LC/ESI-MS
measurements, thus presumably being inorganic iodine species
resulting from decomposition of AM, which are rarely observable
2
For the lower oxidation potentials (0 and 350 mV vs Pd/H ),
AM itself is the main iodine-containing compound in the chro-
matogram, with a percentage of more than 97% on the total iodine
in LC/ESI-MS. The first peak with a retention time of 2 min could
2
content for both cases. At 600 mV versus Pd/H , a considerable
-
be assigned to IO
3
by comparing the retention time with that of
amount of AM was oxidized to AM-Et, which shows an iodine
percentage of 61%. The remaining iodine content is mostly
attributed to AM (35.3%), with minor amounts of inorganic iodine
species and AM-I. At higher oxidation potentials, the percentage
an authentic standard. The same procedure was used to assign
the second peak with a retention time of 2.5 min to the
-
-
combination of the species I , I
2 3
, and I .
9774 Analytical Chemistry, Vol. 80, No. 24, December 15, 2008

different numbers of iodine atoms is not required in this case,
since deiodination was not observed and all detected species thus
contain the same number of iodine atoms.
Table 2. Percentage of the Peak Areas of AM, AM-Et,
and Hydroxylated Metabolites on the Total Amount of
Iodine-Containing Species (Equal to 100%) in the
Absence and in the Presence of NADPHa
CONCLUSIONS
without NADPH
99.3
with NADPH
In this work, the first online coupling of electrochemistry, liquid
chromatography, and inductively coupled plasma-mass spectrom-
etry was introduced. In a complementary approach with EC/LC/
ESI-MS, both the qualitative and the quantitative analysis of
heteroatom-containing drug metabolites are possible. In contrast
to the conventionally used electrospray ionization mass spectrom-
etry, ICPMS allows the quantification of metabolites without
reference substances being available. This is due to the fact that
all compounds having the same number of a heteroatom show
the same response in LC/ICPMS under isocratic separation
conditions.
AM
82.6
86.4
83.9
84.3
1.9
11.4
9.0
9
9
9.3
9.3
mean
abs SD
AM-Et
99.3
0.0
0.7
0
0
.7
.7
10.6
10.3
1.2
mean
abs SD
hydroxylated metabolites
0.7
0.0
6.0
4
5
.6
.5
mean
5.4
abs SD
0.7
This was utilized for the relative quantification of the metabo-
lites of the antiarrhythmic agent amiodarone, which contains two
iodine atoms. Qualitative EC/LC/ESI-MS and LC/ESI-MS mea-
surements showed that amiodarone was electrochemically oxi-
dized under loss of its N-ethyl groups and/or iodine, while
microsomal incubation yielded mainly monodeethylamiodarone
and hydroxylated products. An isocratic separation method was
developed using phase-optimized liquid chromatography and ESI-
MS detection. With the optimized chromatographic separation,
the amount of the formed oxidation products relatively to the total
iodine content could be calculated in dependence on the EC
oxidation potential using the EC/LC/ICPMS coupling. As ex-
pected, with increasing EC potential, the amount of oxidation
products increased, with monodeethylamiodarone as main oxida-
tion product at moderate potentials and inorganic iodine species
as main oxidation products at higher potentials.
a
The peak areas of triplicate measurements (first three rows for
each compound) as well as the mean values (fourth row, boldface)
and the standard deviations (fifth row) are listed.
of AM on the iodine balance drastically decreases, and the main
oxidation products are inorganic iodine species such as IO
, and I
observed at these potentials. Bis-deiodinated metabolites are not
detectable with ICPMS due to the absence of iodine atoms in the
molecules.
-
-
3
, I ,
-
I
2
3
. Nevertheless, AM-Et, AM-I, and AM-2Et were also
Quantification of Metabolites from Microsomal Incuba-
tions. For the quantification of metabolites that were formed upon
the microsomal incubations, the amount of total iodine-containing
species was set to 100% as described before. All peak areas of
single compounds were referred to the iodine balance. As
expected, in the absence of NADPH, AM itself is literally the only
compound observed, with a peak area of more than 99% on the
total iodine content (Table 2). This is due to the fact that the
cytochrome P450 enzymes, which are responsible for the AM
metabolism in microsomes, only work in presence of an electron-
donating system such as NADPH. When NADPH is added, ∼16%
of AM was converted into oxidative metabolites, being AM-Et and
hydroxylated metabolites as were already observed in LC/ESI-
MS measurements. Since the chromatographic resolution of the
hydroxylated metabolites is not sufficiently high to allow the
determination of the peak area of every metabolite, the integration
was performed on the sum of the hydroxylation products. As can
be seen from Table 2, the hydroxylated metabolites account for
These results show that LC/ICPMS is well suited for the
quantification of both electrochemically generated metabolites and
metabolites from microsomal incubations. Combined with LC/
ESI-MS, qualitative information on the drug metabolites is
accessible.
ACKNOWLEDGMENT
Financial support by the Fonds der Chemischen Industrie
(Frankfurt/Main, Germany), the Deutsche Bundesstiftung Um-
welt (Osnabrück, Germany) and the Deutsche Forschungsge-
meinschaft (Bonn Germany) is gratefully acknowledged.
Received for review September 5, 2008. Accepted October
5
.4% of the total iodinated species in the chromatogram, while
AM-Et contributes 10.3% to the iodine content in the microsomal
incubation extract. A differentiation between compounds with
29, 2008.
AC801878K
Analytical Chemistry, Vol. 80, No. 24, December 15, 2008 9775