Electrochemical oxidation by square-wave potential pulses in the imitation of oxidative drug metabolism

Article abstract of DOI:10.1021/ac200897p

Electrochemistry combined with mass spectrometry (EC-MS) is an emerging analytical technique in the imitation of oxidative drug metabolism at the early stages of new drug development. Here, we present the benefits of electrochemical oxidation by square-wa

Full text of DOI:10.1021/ac200897p

ARTICLE
pubs.acs.org/ac
Electrochemical Oxidation by Square-Wave Potential Pulses in the
Imitation of Oxidative Drug Metabolism
Eslam Nouri-Nigjeh, Hjalmar P. Permentier, Rainer Bischoff, and Andries P. Bruins*
Analytical Biochemistry and Mass Spectrometry Core Facility, Department of Pharmacy, University of Groningen, A. Deusinglaan 1,
9713 AV Groningen, The Netherlands
S Supporting Information
b
ABSTRACT: Electrochemistry combined with mass spectro-
metry (ECÀMS) is an emerging analytical technique in the
imitation of oxidative drug metabolism at the early stages of new
drug development. Here, we present the benefits of electro-
chemical oxidation by square-wave potential pulses for the
oxidation of lidocaine, a test drug compound, on a platinum
electrode. Lidocaine was oxidized at constant potential and by
square-wave potential pulses with different cycle times, and the
reaction products were analyzed by liquid chromatographyÀ
mass spectrometry [LCÀMS(/MS)]. Application of constant
potentials of up to +5.0 V resulted in relatively low yields of
N-dealkylation and 4-hydroxylation products, while oxidation
by square-wave potential pulses generated up to 50 times more of the 4-hydroxylation product at cycle times between 0.2 and 12 s
(estimated yield of 10%). The highest yield of the N-dealkylation product was obtained at cycle times shorter than 0.2 s. Tuning of
the cycle time is thus an important parameter to modulate the selectivity of electrochemical oxidation reactions. The N-oxidation
product was only obtained by electrochemical oxidation under air atmosphere due to reaction with electrogenerated hydrogen
peroxide. Square-wave potential pulses may also be applicable to modulate the selectivity of electrochemical reactions with other
drug compounds in order to generate oxidation products with greater selectivity and higher yield based on the optimization of cycle
times and potentials. This considerably widens the scope of direct electrochemistry-based oxidation reactions for the imitation of
in vivo oxidative drug metabolism.
he imitation of oxidative drug metabolism by cytochrome
P450s (P450) at the early stages of new drug development
dissolution of the surface oxide to restore the reactivity of the
fresh surface.6,8
T
requires fastand accurateanalyticaltechniques.1,2 Electrochemistry
combined with mass spectrometry (ECÀMS) is emerging as
a versatile analytical technique capable of the generation and
identification of many known in vivo metabolites.3À5 Although
different oxidation products may be generated at different
potentials, constant potential oxidations often lack in selectivity
when it comes to generation of defined metabolites. In addition,
electrode fouling and/or passivation associated with constant
potential oxidation of organic compounds may attenuate surface
reactivity and thereby reduce product yield.6
A second function of square-wave potential pulses is claimed
to involve the generation of intermediate reactive oxygen species
during the oxidation of water on the Pt electrode. Production of
OH radicals adsorbed on the Pt electrode may participate in
O-transfer reactions to the preadsorbed organic molecules on the
surface. Square-wave potential pulses have been applied success-
fully to achieve anodic detection of numerous polar aliphatic
compounds.9
The aim of pulsed potential electrochemical detection is
sensitive and reproducible current measurement, but the reaction
products are not characterized by spectrometric methods.10À14
Synthetic organic electrochemistry, on the other hand,15 aims at
the production and characterization of specific molecules. In this
field, there have been no reports thus far of the use of square-
wave potential pulses.
Platinum electrodes have been widely used for organic com-
pound oxidation and detection due to their excellent conductiv-
ity and chemical stability even at high positive potentials.7
Johnson and co-workers showed that electrochemical detection
of various organic compounds on Pt electrodes can be signifi-
cantly improved using square-wave potential pulses, as well as
other potential waveforms, with cycle times between 0.2 and 2 s.
Adsorbed molecules generated during oxidation are usually
desorbed during the anodic formation of a surface oxide. Because
the formation of surface oxide can attenuate surface reactivity, a
subsequent negative potential step quickly allows the cathodic
The electrochemical production of drug metabolites by means
of square-wave potential pulses might allow both higher yields
Received: January 14, 2011
Accepted: June 6, 2011
Published: June 06, 2011
r
2011 American Chemical Society
5519
dx.doi.org/10.1021/ac200897p Anal. Chem. 2011, 83, 5519–5525
|

Analytical Chemistry
ARTICLE
Scheme 1. In Vivo Metabolites of Lidocaine Due to Oxidative Metabolism by Cytochrome P450sa
a Refs 5, 16, and 17.
through surface renewal and access to additional products
through O-transfer reactions. In the present study, lidocaine, a
local anesthetic drug, was used as a test compound. The in vivo
metabolites of lidocaine result from N-dealkylation, N-oxidation,
and aromatic and benzylic hydroxylation (Scheme 1).5,16,17
Direct electrochemical oxidation of lidocaine on a porous graphite
electrode gave only the N-dealkylated product.18 N-Oxidation of
lidocaine was observed in the presence of electrochemically
generated hydrogen peroxide.19 Here, we show that square-wave
potential pulses with different cycle times lead to a marked
change in the distribution of lidocaine reaction products as analyzed
and identified by liquid chromatographyÀmass spectrometry
[LCÀMS(/MS)].
electrodes. The reference electrode was placed in the working
compartment. All experiments were performed at ambient
temperature. For deaeration and aeration, argon and synthetic
air, respectively, were bubbled at 25 mL/min via a sparge tube
(MW-4145, BASi) through the 1 mL solution during all experi-
ments (4 mL of solution was used for the experiments presented
in Supporting Information Figure S2). The auxiliary compart-
ment was always under argon purging. In the case of pulsed
potential experiments, the upper and lower potential steps within
one cycle were of equal duration (e.g., 1 s steps for 2 s cycle time).
Solutions containing 10 mM lidocaine and 0.1 M TBAP
dissolved in ACN/H2O 99/1 (v/v) (0.1 M TBAP used as
electrolyte to provide sufficient conductivity for electrochemical
experiments) were subjected to constant potential oxidation and
to oxidation by square-wave potential pulses with continuous
gas flow, for 30 min prior to LCÀMS analysis. Samples were
collected from the working compartment and diluted 100 times
in water containing 10 μM acetaminophen, as an internal
standard for LCÀMS signal normalization, immediately after
the batch oxidations and stored at room temperature until
LCÀMS analysis.
’ EXPERIMENTAL PROCEDURES
Reagents. Tetrabutylammonium perchlorate (TBAP, 86893)
and lidocaine (L7757) were purchased from Sigma-Aldrich.
Water was purified by a Maxima Ultrapure water system
(ELGA, High Wycombe, Bucks, U.K.). Ultrapure HPLC grade
acetonitrile (ACN) was purchased from Merck. H218O with 97
atom % 18O (H218O, 329878) was purchased from Sigma-Aldrich.
3-Hydroxylidocaine (CAS no. 34604-55-2) was purchased from
Toronto Research Chemicals Inc.
Electrode Preparation. The surface of the platinum disk
working electrode was polished with a lapping sheet (Micromesh
grade 3200) prior to each experiment. After mechanical polishing
the surface was washed with ethanol and air-dried.
Electrochemical Measurements. Electrochemical experi-
ments were performed with a homemade potentiostat controlled
by a MacLab system (ADInstruments, Castle Hill, NSW,
Australia) and EChem v.1.52 software (eDAQ, Denistone East,
NSW, Australia). A two-compartment electrochemical cell was
constructed by using a porous Vycor tip with Teflon heat shrink
(MF-2064, Bioanalytical Systems (BASi), West Lafayette, IN,
U.S.A.) to separate working and auxiliary half-cells. The working
electrode was a 2 mm diameter platinum disk (MF-2013, BASi),
and the auxiliary electrode was a platinum wire (MW-4130,
BASi). Potentials were measured against a silver wire pseudo-
reference electrode (MF-2017, BASi), to eliminate the possibility
of chloride contamination of the working solution during
prolonged electrolysis in the presence of conventional reference
LCÀMS Analysis. LCÀMS experiments were carried out on
an LC Packings Ultimate HPLC system (LC Packings, Amsterdam,
The Netherlands) coupled to an API 365 triple quadrupole mass
spectrometer (MDS Sciex, Concord, ON, Canada) upgraded to
EP10+ (Ionics, Bolton, ON, Canada) with electrospray ioniza-
tion in the positive mode in the TurboIonSpray source. The MS
spectra were acquired between m/z 100 and 600 (step size 1.0 amu,
dwell time 1 ms).
A C18 reversed-phase column (GraceSmart RP 18 5 μm,
2.1 mm  150 mm; Grace Davison, Lokeren, Belgium) was
used at a flow rate of 200 μL/min: solvent A, H2O/ACN 95/5
(v/v) with 0.1% formic acid; solvent B, ACN/H2O 95/5 (v/v)
with 0.1% formic acid. An amount of 5 μL of a diluted oxidation
product mixture was injected, and a linear gradient of 5À50% B
in 20 min was used for elution. Peak heights were normalized
with respect to the peak height of the acetaminophen internal
reference compound. In case the ion counting detector of the
MH+ ion was saturated, the 13C1 isotope intensity was taken, and
the corresponding monoisotopic 12C ion intensity was calculated
using the theoretical isotope distribution.
5520
dx.doi.org/10.1021/ac200897p |Anal. Chem. 2011, 83, 5519–5525

Analytical Chemistry
ARTICLE
Figure 1. Relative MH+ ion intensities of the N-dealkylation (a and b), the 4-hydroxylation (c and d), and the N-oxidation (e and f) products of
lidocaine from a solution of 10 mM lidocaine in 0.1 M tetrabutylammonium perchlorate in acetonitrile/water 99/1 (v/v) at different oxidation potentials
under argon and air atmosphere, respectively. Ion intensities were normalized relative to the intensity of the signal for acetaminophen, which was added
as internal standard to all LCÀMS analyses. Experiments were performed in triplicate.
N-Dealkylation, N-oxidation, and aromatic and benzylic
hydroxylation products were identified by LCÀMS/MS analysis,
and in the case of the aromatic hydroxylation at the 4-position
further evidence was obtained by coinjection with the 3-hydrox-
ylidocaine as reference compound, as described before.19
proposed reaction mechanism, which does not involve dissolved
molecular oxygen and proceeds via direct electron transfer from
the tertiary amine to the electrode followed by deprotonation to
give an iminium intermediate that, after hydrolysis and intra-
molecular rearrangement, leads to the N-dealkylation product.21
Hydroxylation at the 4-position of the aromatic ring in
lidocaine was observed at potentials of +2.0 V or more (Figure 1,
parts c and d), although overall yield was low. The 4-hydroxyla-
tion product is most likely generated through an anodic sub-
stitution reaction by water molecules, which is initiated by a two-
electron transfer to the electrode simultaneously with the
formation of a Wheland-type intermediate, resulting in the aro-
matic hydroxylation product after deprotonation (Scheme 2).22,23
Although unsubstituted aromatic rings can be oxidized at higher
positive potentials,20 the electron-donating amide and the two
methyl substituents presumably decrease the oxidation potential
of the aromatic ring in lidocaine and direct the hydroxylation
reaction toward the 4-position through resonance stabilization of
the charge of the radical cation intermediate. Whereas direct
electrochemical oxidation is apparently unable to generate
the 3-hydroxylation product, in vivo oxidative metabolism by
P450 produces both the 3- and the 4-hydroxylation products
(Scheme 1) indicating that in vivo aromatic hydroxylation does
not involve a radical cation intermediate, but that it proceeds via
an oxygen insertion mechanism as proposed for reaction of the
oxo-ferryl radical cation.24 The range of aromatic hydroxyla-
tion reactions by direct electrochemistry is thus restricted in
’ RESULTS AND DISCUSSION
Constant Potential Oxidation. In a previous study, N-deal-
kylation of lidocaine was achieved by online ECÀMS in a porous
graphite flow-through cell with a potential ramp from 0 to +1.8 V
in a solution of ACN/H2O 50/50 (v/v).18,20 The amount of
N-dealkylation product was shown to decrease above +0.8 V in a
potential ramp experiment, although no other oxidation products
were observed at higher potentials.18 The oxidation conditions in
our current batch cell with a Pt working electrode at potentials
below +2.0 V resulted in N-dealkylation and N-oxidation. At
potentials of +2.0 V and higher, 4-hydroxylation, as well as small
amounts of doubly oxidized products (mainly 3,4-dihydrox-
ylation), was observed. Figure 1 shows the distribution of
oxidation products as a function of the applied potential on
the Pt working electrode when purged with argon and air,
respectively.
The N-dealkylation product was barely detected at potentials
above +2.0 V, both under argon and under air atmosphere
(Figure 1, parts a and b). The N-dealkylation yield was not
affected by the presence of air, which is consistent with the
5521
dx.doi.org/10.1021/ac200897p |Anal. Chem. 2011, 83, 5519–5525

Analytical Chemistry
ARTICLE
Scheme 2. Proposed Anodic Substitution Mechanism
Initiated by the Abstraction of Two Electrons from the
Aromatic Ring Followed by Substitution with Water
Proceeding through a Wheland-Type Intermediatea
a Ref 23.
comparison to P450 due to the different nature of the reaction
mechanism.
In contrast to N-dealkylation and 4-hydroxylation, formation
of the N-oxidation product is strongly dependent on the
presence of dissolved molecular oxygen in the solution with a
sharp increase in yield between +4.0 and +5.0 V (Figure 1, parts e
and f). N-Oxidation has been reported to result from the reaction
between lidocaine and electrochemically generated hydrogen
peroxide,19 through peroxide intermediate formation and
decomposition.25 Hydrogen peroxide is generated in the cathodic
compartment of the counter electrode and reaches the anodic
compartment by diffusion through the porous frit membrane
that separates both compartments, as previously reported.19
Stable isotope labeling was used to reveal the source of the
oxygen atom in the N-oxidation product. According to the
proposed N-oxidation mechanism, hydrogen peroxide is generated
by reduction of molecular oxygen and the oxygen atom should
originate from dissolved molecular oxygen. Since the only
oxygen sources in our experiment are molecular oxygen and
water, we replaced H216O by H218O, to test this hypothesis.
LCÀMS analysis showed that the N-oxidation product incorporated
only 16O confirming that the oxygen atom in the N-oxidation
product originated from dissolved molecular oxygen. The low
amount of N-oxidation product observed under argon atmo-
sphere (Figure 1e) may therefore be explained by incomplete
removal of residual oxygen during argon purging.
Oxidation by Square-Wave Potential Pulses. The absolute
yield of N-dealkylation and 4-hydroxylation products under
constant potential conditions was below 1%. Yields were esti-
mated by comparison of the MH+ ion intensities of the products
with that of lidocaine under the assumption that the electrospray
ionization efficiencies for lidocaine and its oxidation products
are similar. In order to affect the kinetics of the oxidation
reaction, we modified constant potential oxidation by adding a
lower potential step to generate a square-wave potential pulse
with variable cycle time.
Figure 2. (a) Relative MH+ ion intensities of the 4-hydroxylation
product of lidocaine measured by LCÀMS after electrochemical oxida-
tion with square-wave potential pulses alternating between an upper
potential of +3.0 V and lower potentials ranging from À2.0 to +3.0 V
with a fixed cycle time of 2 s. Ion intensities were normalized relative to
the intensity of the signal for acetaminophen, which was added as
internal standard to all LCÀMS analyses. Experiments were performed
in triplicate. Ion intensities are plotted on a logarithmic scale. (b) Linear
stripping voltammograms recorded at a scan rate of 1 V/s from +1.0 to
0.0 V after constant potential oxidation at +3.0 V (Pt electrode) for 0,
0.01, 0.05, 0.10, 0.5, and 1.0 s in the absence of lidocaine (the arrow
shows the respective traces from top to bottom).
varied between +3.0 and À2.0 V with a cycle time of 2 s. The yield
of the 4-hydroxylation product at different lower potentials
shows that alternating between +3.0 V and a lower potential of
+0.5 V or less leads to a more than 10-fold increase in 4-hydro-
xylation product, indicating that the surface properties of the
electrode are significantly affected by potential steps below
+0.5 V (Figure 2a). It was suggested previously that, although
surface reactivity could be attenuated by formation of an inert
platinum oxide layer during oxidation, a subsequent reduction
step can restore the reactivity of the fresh surface.6,8 To test the
oxide layer formation under our condition and its subsequent
reduction, the surface was oxidized at +3.0 V with times ranging
from 0, 0.01, 0.05, 0.1, 0.5, to 1.0 s in the absence of lidocaine
followed by linear stripping voltammograms recorded from +1.0
to 0.0 V at a scan rate of 1 V/s (Figure 2b). A fresh surface (0 s)
did not show any reduction peak across the recorded potential
region that implies the absence of oxide layer on the fresh surface.
However, application of the oxidation potential even for 0.01 s
led to a negative current peak at approximately +0.5 V indicating
In view of the results presented in Figure 1, parts c and d, an
upper oxidation potential of +3.0 V was selected for generation of
the 4-hydroxylation product, while the lower potential step was
5522
dx.doi.org/10.1021/ac200897p |Anal. Chem. 2011, 83, 5519–5525

Analytical Chemistry
ARTICLE
Figure 3. Relative MH+ ion intensities of the N-dealkylation (a and b), 4-hydroxylation (c and d), and N-oxidation (e and f) products of lidocaine from a
solution of 10 mM lidocaine in 0.1 M tetrabutylammonium perchlorate in acetonitrile/water 99/1 (v/v) at varying cycle times (plotted logarithmically)
of square-wave potential pulses alternating between +3.0 and À1.0 V under argon and air atmosphere, respectively. Ion intensities were normalized
relative to the intensity of the signal for acetaminophen, which was added as internal standard to all LCÀMS analyses. Experiments were performed in
triplicate.
rapid formation of an oxide layer on the Pt surface at +3.0 V, and
its subsequent reduction at potentials lower than +0.5 V. In
addition, reduction of the presumed oxide layer at +0.5 V
coincides with the potential at which we observe a dramatic
increase in the yield of 4-hydroxylation product of lidocaine
(Figure 2a). This indicates strongly that reduction of the oxide
layer on the Pt electrode surface at or below +0.5 V reactivates
the electrode surface for the next pulse cycle leading to increased
yields.
On the basis of the above results, we selected À1.0 V as the
lower potential level and monitored product distribution as a
function of cycle time ranging from 0.02 to 200 s by LCÀMS
(Figure 3). N-Dealkylation was only detected at cycle times
shorter than 0.2 s (Figure 3, parts a and b). Since N-dealkylation
was nearly absent upon constant potential oxidation at +3.0 V
(Figure 1, parts a and b), repeated and fast switching of the
surface potential apparently facilitated electron transfer from the
tertiary amine moiety of lidocaine to the electrode. The yield of
the 4-hydroxylation product also showed a strong dependence
on cycle time, with an optimum at cycle times around 1 s in-
dependent of an argon or air atmosphere (Figure 3, parts c and d).
Whereas the yield of the 4-hydroxylation product was low at a
constant potential of +3.0 V or at short cycle times, it in-
creased about 50-fold (estimated total yield of 10%) at cycle
times of 1 s (see Figures 3c and 1c). The selectivity of pulsed
electrochemical oxidation of lidocaine can thus be directed at the
N-dealkylation or at the 4-hydroxylation product by selection of
the cycle time.
The cause of this change in selectivity is currently unclear.
Orientation of lidocaine on the working electrode, the lifetime of
different reactive intermediates of oxidizedlidocaine, and recovery
of a fresh Pt surface may all be controlled by the potential levels of
the square-wave potential pulses and may have their effect on the
competition between N-dealkylation and 4-hydroxylation.
A change of atmosphere from argon to air did not affect the
yield of the 4-hydroxylation product (Figure 3, parts c and d),
whereas a higher yield of the N-dealkylation product was observed
under air atmosphere, possibly becauseoxygenreductiongenerates
5523
dx.doi.org/10.1021/ac200897p |Anal. Chem. 2011, 83, 5519–5525

Analytical Chemistry
ARTICLE
was much lower using a glassy carbon electrode. This supports
further that this reaction requires a noble metal surface that can
be reactivated through redox cycling (Supporting Information
Figure S1). In addition, a linear increase in the generation of
4-hydroxylation product was observed under pulsed potential
condition over a 30 min period, suggesting that successful surface
reactivation was achieved under pulse condition (Supporting
Information Figure S2).
’ CONCLUSIONS
Constant potential oxidation of lidocaine generates low yields
of N-dealkylation and 4-hydroxylation products. By varying the
cycle time of square-wave potential pulses and the voltage of the
lower potential step, we increased the yield of 4-hydroxylidocaine
by about 50-fold relative to constant potential conditions result-
ing in an overall yield of approximately 10% at a cycle time of 1 s,
whereas N-dealkylation was favored at short cycle times below
0.2 s. Pulsed potentials are thus effective in modulating electro-
chemical oxidation reactions that are initiated by direct electron
transfer. We show that the Pt electrode surface is rapidly
passivated under oxidative conditions and that it is regenerated
during the low-potential step of the square-wave pulse. How
cycle time affects the selectivity of the reaction (N-dealkylation at
short pulse times versus 4-hydroxylation at longer pulse times)
remains to be elucidated. It is conceivable that the lidocaine
molecule reorients itself on the electrode during pulsing and that
different parts of the molecule are in contact with the electron-
transferring surface. Further experiments are needed to study this
in greater detail. Square-wave potential pulses may be applicable
to other drug compounds in order to generate oxidation products
with greater selectively and higher yield based on optimization of
cycle times and potentials. This could widen the scope of direct
electrochemistry-based oxidation reactions for the imitation of
in vivo drug metabolism.
Figure 4. LCÀMS analysis of lidocaine oxidation products in the pre-
sence H218O with square-wave potential pulses alternating between +3.0
and À1.0 V and a cycle time of 2 s under air atmosphere. The extracted
ion chromatograms correspond to the MH+ ions of the 4-hydroxylation
and the N-oxidation products at m/z 251 (solid line, 16O) and 253
(dashed line, 18O).
superoxide anions that may be sufficiently basic for proton ab-
straction to form the iminium intermediate, facilitating generation
of the N-dealkylation product.21 Generation of the N-oxidation
product requires molecular oxygen (see Figure 1, parts e and f),
but it is also dependent on cycle time (Figure 3, parts e and f). A
lower yield of the N-oxidation product was observed at cycle
times between 0.2 and 12 s. It is intriguing that a minimum in
N-oxidation yield coincides with the optimum for the 4-hydroxyla-
tion reaction suggesting a competition between the electrochemical
generation of hydrogen peroxide and oxidation of the aromatic
ring for subsequent reaction with water.
’ ASSOCIATED CONTENT
The anodic substitution mechanism (see Scheme 2) implies
that the oxygen atom in 4-hydroxylated lidocaine originates from
water. To verify this we replaced H216O by H218O during oxida-
tion of lidocaine using square-wave potential pulses. The ex-
tracted ion chromatograms (Figure 4) show that the major part
of the 4-hydroxylation product contained 18O, although about
15À20% of 16O incorporation was observed, which is likely due
to a remaining 3% of H216O in the commercial H218O vial and to
residual water from the atmosphere. This result is in agreement
with the proposed mechanism. Producing the N-oxidation product
using square-wave potential pulses in the presence of H218O
resulted in approximately 30% incorporation of 18O, whereas the
remainder contained 16O in accordance with the proposed me-
chanism through which molecular oxygen is reduced to hydrogen
peroxide. The 30% of 18O incorporation in the N-oxidation
product under pulsed potential conditions might be due to
oxidation of H218O to 18O2 followed by reduction to H218O2.
The only observed hydroxylation product of lidocaine upon
pulsed potential oxidation was 4-hydroxylidocaine, whereas the
Fenton reaction, which leads to generation of hydroxyl radicals,
results additionally in 3-hydroxylation and benzylic hydroxylation.19
The absence of the latter hydroxylation products indicates
strongly that no adsorbed OH radicals are generated on the Pt
electrode during water oxidation. Pulsed potential oxidation of
lidocaine on a gold electrode showed the same response as on the
Pt electrode, whereas the yield of the 4-hydroxylation product
S
Supporting Information. Additional information as noted
b
in text. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: a.p.bruins@rug.nl. Phone: +31-50-3633262. Fax: +31-
503638347.
’ ACKNOWLEDGMENT
The Dutch Technology Foundation (STW) is gratefully
acknowledged for financial support (Grant 07047). Further
financial support was obtained from Astra Zeneca (Mølndal,
Sweden) and Organon (Oss, The Netherlands; currently a
subsidiary of Merck & Co. Inc., Whitehouse Station, NJ, U.S.A.).
LC equipment was provided by LC Packings, Amsterdam, The
Netherlands (now part of Dionex, Sunnyvale, CA, U.S.A.).
’ REFERENCES
(1) Lohmann, W.; Karst, U. Anal. Bioanal. Chem. 2009, 394,
1341–1348.
(2) Lohmann, W.; D€otzer, R.; G€utter, G.; Van Leeuwen, S. M.; Karst,
U. J. Am. Soc. Mass Spectrom. 2009, 20, 138–145.
5524
dx.doi.org/10.1021/ac200897p |Anal. Chem. 2011, 83, 5519–5525

Analytical Chemistry
ARTICLE
(3) Baumann, A.; Lohmann, W.; Schubert, B.; Oberacher, H.; Karst,
U. J. Chromatogr., A 2009, 1216, 3192–3198.
(4) Madsen, K. G.; Olsen, J.; Skonberg, C.; Hansen, S. H.; Jurva, U.
Chem. Res. Toxicol. 2007, 20, 821–831.
(5) Johansson, T.; Weidolf, L.; Jurva, U. Rapid Commun. Mass
Spectrom. 2007, 21, 2323–2331.
(6) Hoekstra, J. C.; Johnson, D. C. Anal. Chim. Acta 1999, 390,
45–54.
(7) Panizza, M.; Cerisola, G. Chem. Rev. 2009, 109, 6541–6569.
(8) Johnson, D. C.; LaCourse, W. R. Anal. Chem. 1990, 62, 589–597.
(9) Johnson, D. C.; Dobberpuhl, D.; Roberts, R.; Vandeberg, P.
J. Chromatogr. 1993, 640, 79–96.
(10) Williams, D. G.; Johnson, D. C. Anal. Chem. 1992, 64,
1785–1789.
(11) Angerstein-Kozlowska, H.; Conway, B. E.; Sharp, W. B. A.
J. Electroanal. Chem. 1973, 43, 9–36.
(12) Tilak, B. V.; Conway, B. E.; Angerstein-Kozlowska, H.
J. Electroanal. Chem. 1973, 48, 1–23.
(13) Austin, D. S.; Johnson, D. C.; Hines, T. G.; Berti, E. T. Anal.
Chem. 1983, 55, 2222–2226.
(14) Johnson, D. C.; Feng, J.; Houk, L. L. Electrochim. Acta 2000,
46, 323–330.
(15) Lund, H.; Hammerich, O. Organic Electrochemistry, 4th ed.;
Marcel Dekker, Inc.: New York, 2001.
(16) Oda, Y.; Imaoka, S.; Nakahira, Y.; Asada, A.; Fujimori, M.;
Fujita, S.; Funae, Y. Biochem. Pharmacol. 1989, 38, 4439–4444.
(17) Thomas, J.; Meffin, P. J. Med. Chem. 1972, 15, 1046–1049.
(18) Jurva, U.; Wikstr€om, H. V.; Bruins, A. P. Rapid Commun. Mass
Spectrom. 2000, 14, 529–533.
(19) Nouri-Nigjeh, E.; Permentier, H. P.; Bischoff, R.; Bruins, A. P.
Anal. Chem. 2010, 82, 7625–7633.
(20) Jurva, U.; Wikstr€om, H. V.; Weidolf, L.; Bruins, A. P. Rapid
Commun. Mass Spectrom. 2003, 14, 800–810.
(21) Smith, P. J.; Mann, C. K. J. Org. Chem. 1969, 34, 1821–1826.
(22) Eberson, L. J. Am. Chem. Soc. 1967, 89, 4669–4677.
(23) Eberson, L.; Nyberg, K. Acc. Chem. Res. 1973, 6, 106–112.
(24) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. Rev. 2004,
104, 3947–3980.
(25) Oswald, A. A.; Guertin, D. L. J. Org. Chem. 1963, 28, 651–657.
5525
dx.doi.org/10.1021/ac200897p |Anal. Chem. 2011, 83, 5519–5525