Derivatisation of microcystin with a redox-active label for high-performance liquid chromatography/electrochemical detection

Article abstract of DOI:10.1039/b206384k

Microcystins are a group of low molecular weight, cyclic peptide hepatotoxins. The most common detection and quantitation methods for these toxins are liquid chromatography with UV or mass spectrometric detections, phosphatase inhibition assays and enzyme-linked immunosorbent assays. In addition, derivatisation of these toxins with organic fluorophores followed by CE/laser induced fluorescence detection and HPLC/chemiluminescence detection; and with luminescent lanthanide chelates for competition assays have also been reported. However, the use of an electrochemical-active unit as a tag for microcystins has never been explored. Since the sulfhydryl group of 6-ferrocenylhexanethiol (Fc-C6-SH) can undergo a facile addition reaction with the α,β-unsaturated carbonyl group, this compound has been used as a redox-active labelling agent for a derivative of microcystins, microcystin-LR (MC-LR). The conjugate, Fc-MC-LR, has been isolated by high-performance liquid chromatography with electrochemical detection. The peak height-concentration curve was linear in the test range 20-400 ng of MC-LR (r value for linear regression > 0.9987). The detection limit was determined to be ca. 18 ng MC-LR (S/N ≈ 3). Meanwhile, the conjugate Fc-MC-LR has also been characterised by positive-ion electrospray-ionisation mass spectrometry. Electrochemical studies show that the adduct displays a reversible ferrocenium/ferrocene couple at ca. -0.040 V vs. SCE (scan rate = 50 mV s^-1) in 0.1 M aqueous ammonium acetate-acetonitrile (55:45 v/v).

Full text of DOI:10.1039/b206384k

View Article Online / Journal Homepage / Table of Contents for this issue
Derivatisation of microcystin with a redox-active label for
high-performance liquid chromatography/electrochemical detection
Kenneth Kam-Wing Lo,* Dominic Chun-Ming Ng, Jason Shing-Yip Lau,
Rudolf Shiu-Sun Wu and Paul Kwan-Sing Lam*
Centre for Coastal Pollution and Conservation, City University of Hong Kong,
Tat Chee Avenue, Kowloon, Hong Kong, P. R. China. E-mail: bhkenlo@cityu.edu.hk;
Fax: +852 2788 7406; Tel: +852 2788 7231
Received 2nd July 2002, Accepted 16th September 2002
First published as an Advance Article on the web 6th December 2002
Microcystins are a group of low molecular weight, cyclic peptide hepatotoxins. The most common detection
and quantitation methods for these toxins are liquid chromatography with UV or mass spectrometric
detections, phosphatase inhibition assays and enzyme-linked immunosorbent assays. In addition, derivatisation
of these toxins with organic fluorophores followed by CE/laser induced fluorescence detection and HPLC/
chemiluminescence detection; and with luminescent lanthanide chelates for competition assays have also been
reported. However, the use of an electrochemical-active unit as a tag for microcystins has never been explored.
Since the sulfhydryl group of 6-ferrocenylhexanethiol (Fc-C6-SH) can undergo a facile addition reaction with
the a,b-unsaturated carbonyl group, this compound has been used as a redox-active labelling agent for a
derivative of microcystins, microcystin-LR (MC-LR). The conjugate, Fc-MC-LR, has been isolated by
high-performance liquid chromatography with electrochemical detection. The peak height-concentration curve
was linear in the test range 20–400 ng of MC-LR (r value for linear regression > 0.9987). The detection limit
was determined to be ca. 18 ng MC-LR (S/N ¼ 3). Meanwhile, the conjugate Fc-MC-LR has also been
characterised by positive-ion electrospray-ionisation mass spectrometry. Electrochemical studies show that the
adduct displays a reversible ferrocenium/ferrocene couple at ca. ꢀ0.040 V vs. SCE (scan rate ¼ 50 mV sꢀ1) in
0.1 M aqueous ammonium acetate–acetonitrile (55:45 v/v).
Microcystins are a group of low molecular weight, cyclic pep-
tide hepatotoxins produced by cyanobacterial species in
eutrophic lakes and drinking water reservoirs.1,2 To date,
about sixty different types of microcystins have been charac-
terised. Since these toxins represent significant hazards to
humans, livestock and wildlife, their levels in the environment
must be strictly monitored. There have been many reports on
the detection and quantitation of this class of toxins. The most
common one is liquid chromatography with UV or mass spec-
trometric detections.3–6 Phosphatase inhibition assays7,8 and
enzyme-linked immunosorbent assays (ELISA)9,10 are also
effective quantitation procedures. In addition, derivatisation
of these toxins with organic fluorophores followed by CE/laser
induced fluorescence detection11 and HPLC/chemilumines-
cence detection;12 and with luminescent lanthanide chelates
for competition assays have also been reported.13 Compared
to these analytical techniques, electrochemical detection of
microcystins has been receiving much less attention, mainly
because of the lack of a redox-active moiety on these toxin
molecules. In this regard, differential pulse polarography has
been used to study the binding of copper and zinc ions to
microcystin derivatives.14 The change in the electrochemical
potential and decrease in the height of metal polarogram peaks
have been correlated to the concentration of microcystins.
Besides, the interactions between microcystins and mercury(II)
ion, and other ions such as lead(II), have also been studied
using anodic stripping voltammetry.15 Meanwhile, direct elec-
trochemical (EC) detection in the liquid chromatographic
separation of microcystins, based on the oxidation of the argi-
nine and tyrosine residues of these toxins, has also been exam-
ined.16 Despite all these reports, to the best of our knowledge,
the use of an electrochemical-active unit as a tag for micro-
cystins has never been explored. In this paper, we report the
utilisation of a redox-active compound, 6-ferrocenylhexa-
nethiol (Fc-C6-SH), as a label for a derivative of these toxins,
microcystin-LR (MC-LR). The reactivity of Fc-C6-SH
towards the a,b-unsaturated ketone moiety has been demon-
strated using methyl vinyl ketone (MVK) as the model sub-
strate. The toxin-derivatisation procedure and HPLC/EC
separation of the conjugate, Fc-MC-LR, together with its char-
acterisation and electrochemical properties are also reported.
Experimental
Materials
All solvents for chemical synthesis were of analytical grade and
used without further purification. Ammonium acetate (Junsei,
Tokyo, Japan) was used as received. Water for aqueous buffers
was purified to 18.2 MO cm on a Milli-Q apparatus (Millipore,
USA). HPLC grade acetonitrile and trifluoroacetic acid (TFA)
were purchased from RDH (Seelze, Germany). A commercial
sample of MC-LR purchased from Calbiochem (La Jolla, CA,
USA) was used as the standard.
Synthesis
The label Fc-C6-SH was prepared from modified literature
methods.17–19 To a dichloromethane (40 ml) solution of ferro-
cene (Acros, Geel, Belgium) (1.99 g, 10.7 mmol) was added 6-
bromohexanoyl chloride (Acros, Geel, Belgium) (2.28 g, 10.7
mmol) dissolved in 5 ml CH2Cl2 . Aluminium chloride (Acros,
274
New J. Chem., 2003, 27, 274–279
DOI: 10.1039/b206384k
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
Geel, Belgium) (1.43 g, 10.7 mmol) was added to the mixture.
The colour of the suspension changed from orange to purple
and the suspension was stirred under an inert atmosphere of
nitrogen for 4 h at room temperature. The reaction was then
quenched by addition of 4 M HCl (100 ml). The aqueous layer
was separated and extracted with CH2Cl2 (100 ml ꢁ 3). The
organic layers were collected, dried over MgSO4 and evapo-
rated under vacuum to give a brown oil. The crude product,
6-bromohexanoylferrocene, was purified by column chromato-
graphy using silica gel as the stationary phase. The unreacted
starting material was eluted with petroleum ether (40–60 ꢂC)
–diethyl ether (95:5) and the product with petroleum ether
(40–60 ꢂC)–diethyl ether (90:10). Yield: 2.50 g, 6.9 mmol,
64%. Positive-ion ESI-MS: m/z ¼ 363, M+. IR (KBr) n/
cmꢀ1: 1668 (s, C=O).
To a mixture of 6-bromohexanoylferrocene (1.50 g, 4.1
mmol), zinc (Sigma–Aldrich, St. Louis, MO, USA) (8.09 g,
123.8 mmol) and mercury(II) chloride (Sigma–Aldrich, St.
Louis, MO, USA) (0.75 g, 2.8 mmol) in 20 ml toluene and
18 ml H2O was slowly added 14 ml concentrated HCl (12
M) solution. The mixture was stirred at room temperature
for 12 h, after which, the toluene layer was separated, washed
with water (100 ml ꢁ 3), dried over MgSO4 and evaporated
under vacuum to give a brown oil. The crude product, 1-
bromo-6-ferrocenylhexane, was purified by column chromato-
graphy on silica gel with n-hexane as the mobile phase.
Yield ¼ 0.78 g, 2.2 mmol, 54%. Positive-ion ESI-MS: m/z ¼
348, M+.
measurements revealed that the major microcystin derivative
present was MC-LR (m/z ¼ 996), while its counterparts
MC-YR (m/z ¼ 1045) and MC-RR (m/z ¼ 1038) were also
present in trace amounts. The toxin MC-LR was then further
purified by HPLC. The HPLC system consisted of a Waters
600 pump (Milford, MA, USA) equipped with Rheodyne
7725i injector (Rohnert Park, CA, USA) with a 50-ml sample
loop. The column was an Ultrasphere ODS column (250
mm ꢁ 4.6 mm, Beckman, Beckman, Fullerton, CA, USA).
The gradient mobile phases consisted of (A) 30% (v/v) aqu-
eous acetonitrile and 0.05% (v/v) TFA, and (B) 100% acetoni-
trile, with a linear gradient of 0–20% B from 0 to 20 min. The
flow rate was 1 ml minꢀ1. UV detection was performed with a
Waters 490E detector with the absorbance monitored at 238
nm. The fractions containing MC-LR were collected and the
solvent removed under reduced pressure. The sample was then
dissolved in ethanol (2 ml) to make up a stock solution. The
concentration of this MC-LR stock solution was determined
based on a calibration curve constructed by analysing commer-
cially available MC-LR standards.
Reaction of MVK and Fc-C6-SH
A mixture of Fc-C6-SH (0.2 g, 0.7 mmol) and MVK (Aldrich,
St. Louis, MO, USA) (39 mg, 0.6 mmol) in 5 ml CH3CN was
stirred at room temperature for 12 h. The solution was then
loaded to a silica gel column with petroleum ether as the elu-
ent. The fractions containing the product were collected and
the solvent was removed by rotary evaporation to give Fc-
MVK as a yellow oil. Yield ¼ 169 mg, 0.5 mmol, 82%. 1H
NMR (300 MHz, C6D6 , 298 K, relative to TMS) d 4.04 (s,
5H, ferrocenyl H’s), 4.00–3.98 (m, 4H, ferrocenyl H’s), 2.61
[t, J ¼ 7.33 Hz, 2H, SCH2CH2(CO)CH3], 2.33 (t, J ¼ 7.03
Hz, 2H, Fc-CH2CH2CH2CH2CH2CH2S), 2.25 (t, J ¼ 7.62
Hz, 2H, Fc-CH2CH2CH2CH2CH2CH2S), 2.20 [t, J ¼ 7.33
Hz, SCH2CH2(CO)CH3], 1.60 [s, 3H, SCH2CH2(CO)CH3],
1.55–1.21 (m, 8H, Fc-CH2CH2CH2CH2CH2CH2S). Positive-
ion ESI-MS: m/z ¼ 372, M+. IR (KBr) n/cmꢀ1: 1718 (s,
C=O).
1-Bromo-6-ferrocenylhexane was then converted to Fc-C6-
SH. A mixture of 1-bromo-6-ferrocenylhexane (0.73 g, 2.1
mmol) and thiourea (Junsei, Tokyo, Japan) (0.22 g, 2.9 mmol)
in 8 ml ethanol was refluxed for 4 h under an inert atmosphere
of nitrogen. Then, NaOH (0.01 g in 1.5 ml H2O) was added
and the solution was refluxed for another 2 h. The solution
was then cooled to room temperature and acidified to pH 2
with 0.1 M HCl. The mixture was extracted with ethyl acetate
(100 ml ꢁ 3) and the organic extracts were dried over MgSO4
and evaporated under vacuum to give a brown oil. Fc-C6-
SH was purified by column chromatography on silica gel using
petroleum ether as the eluent. The product appeared as a yel-
1
low oil. Yield ¼ 0.23 g, 0.8 mmol, 36%. H NMR (300 MHz,
CDCl3 , 298 K, relative to TMS) d 4.09 (s, 5H, ferrocenyl H’s),
4.04 (s, 4H, ferrocenyl H’s), 2.52 (q, J ¼ 7.33 Hz, 2H, CH2-
SH), 2.32 (t, J ¼ 7.33 Hz, 2H, Fc-CH2), 1.66–1.25 (m, 9H, Fc-
CH2CH2CH2CH2CH2CH2SH). Positive-ion ESI-MS: m/z ¼
302, M+. IR (KBr) n/cmꢀ1: 2561 (m, SH).
Labelling of MC-LR with Fc-C6-SH
Fc-C6-SH (10 ml) was added to 100 ml of a stock ethanolic
solution of MC-LR (0.794 mg mlꢀ1). To the solution was added
5% (w/v) aqueous Na2CO3 solution (10 ml). The suspension
was stirred at 60 ꢂC for 4 h. The solid precipitated was
removed by centrifugation. The supernatant was evaporated
to dryness under reduced pressure. The solid residue was then
resuspended in 600 ml water. The mixture was then centrifuged
again to remove the solid residue. The supernatant was diluted
with water to 1 ml. A 50-ml portion of this solution was applied
to HPLC equipped with an EC detector for analysis.
Isolation of MC-LR
Microcystis cell material was from a surface bloom collected
from a farm pond in China. The purification was based on a
modified reported procedure.20 The dried cells (10 g) were
mixed with 5% (v/v) aqueous acetic acid (200 ml) and the sus-
pension was stirred at room temperature for 1 h. The mixture
was centrifuged at 4620ꢁg for 15 min and the supernatant was
collected. The solid residue was extracted with 5% (v/v) aqu-
eous acetic acid (200 ml ꢁ 2). The supernatant fractions were
combined and applied to a reversed-phase silica gel (30 g)
(Chromatorex, Fuji Silysia Chem. Ltd., Nagoya, Japan) col-
umn. The column was washed with H2O (850 ml) and the tox-
ins were eluted with methanol (400 ml). The fractions
containing the toxins (monitored by positive-ion ESI-MS)
were collected and concentrated to ca. 20 ml by rotary eva-
poration. To the solution was added normal-phase silica gel
(230–400 mesh, silica gel 60, Merck, Darmstadt, Germany)
(3.0 g) and the suspension was evaporated to dryness. The
silica gel adsorbed with the toxins was then applied to a silica
gel column (15 g) of the same stationary phase using chloro-
form–methanol–water (65:25:5) as the eluent. The fractions
containing the toxins were collected. Positive-ion ESI-MS
HPLC/EC for Fc-MC-LR
The HPLC analysis of Fc-MC-LR was carried out using the
Ultrasphere ODS column described above, under an isocratic
condition. The mobile phase was 0.1 M aqueous ammonium
acetate–acetonitrile (55:45 v/v) at a flow rate of 1.0 ml minꢀ1
.
. The HPLC system consisted of a Waters 515 HPLC pump
(Milford, MA, USA) with a Rheodyne 7725i injector (Rohnert
Park, CA, USA) equipped with a 50-ml sample loop. The
detector was a Waters 464 pulsed electrochemical detector
(Milford, MA, USA) equipped with a thin-layer flow cell
(volume ¼ 4 ml, gasket thickness ¼ 0.0400). The working, coun-
ter and reference electrodes were a glassy carbon, stainless steel
and Ag/AgCl electrode, respectively. The operation potential
was +300 mV.
New J. Chem., 2003, 27, 274–279
275

View Article Online
Electrochemical studies of Fc-MVK and Fc-MC-LR
microcystins in the literature have employed sulfhydryl-
containing molecules such as cysteine,11 glutathione21 and 2-
aminoethanethiol22 to react with the toxin molecule, and to
further derivatise the primary-amine containing adducts with
amine-specific labels. However, these derivatisation procedures
involve two labelling steps and have reduced specificity
because the amine group commonly exists in biological sam-
ples compared to the characteristic a,b-unsaturated carbonyl
group of these toxins. In the present work, our strategy is to
incorporate a redox-active molecule specifically to the toxin
molecule, most preferably via a one-step reaction. The simple
molecule, Fc-C6-SH, is a very promising candidate based on
the following reasons: (1) it contains an electrochemically
active ferrocene group, which is well known to exhibit reversi-
ble electrochemistry; (2) the hexamethylene spacer can mini-
mise any potential steric hindrance between the toxin
molecule and the label; (3) the thiol moiety can undergo a
Michael addition reaction with the a,b-unsaturated carbonyl
group of the toxin molecule (Scheme 1). Using ferrocene as a
redox-active tag for biomolecules such as DNA, peptides
and proteins have been documented.23–31 Electrochemical stu-
dies of ferrocenylalkanethiols immobilised on gold electrodes
as self-assembled monolayers have also been investigated.17,18
However, to the best of our knowledge, utilisation of Fc-C6-
SH and related compounds as a derivatisation agent for bio-
logical molecules has never been explored.
Cyclic voltammetry experiments were performed on a CH
Instruments Electrochemical Workstation CHI750A (Austin,
TX, USA). For Fc-MVK, the electrochemical experiments
were carried out at room temperature with a two-compartment
glass cell with a working volume of 500 ml. The working elec-
trode was a glassy carbon electrode. A platinum gauze counter
electrode was accommodated in the working electrode com-
partment. A silver/silver nitrate electrode was used as the
reference electrode. The reference electrode compartment was
connected to the working electrode compartment via a Luggin
capillary. The solvent was acetonitrile with 0.1 M nBu4NPF6
(Sigma–Aldrich, St. Louis, MO, USA) as the supporting elec-
trolyte. For Fc-MC-LR, a micro-volume voltammetry cell
(Bioanalytical Systems, West Lafayette, IN, USA) with a
working volume of 100 ml was used. A carbon paste working
electrode, a platinum wire counter electrode and a silver/silver
chloride reference electrode were used.
Other instrumentation
1H NMR spectra were recorded on a Varian (Pal Alto, CA,
USA) Mercury 300 MHz NMR spectrometer at 298 K. Positive-
ion ESI mass spectra were recorded on a Perkin Elmer Sciex
(Concord, ON, Canada) API 365 mass spectrometer. IR
spectra were recorded on a Perkin Elmer (Shelton, CT, USA)
1600 series FT-IR spectrophotometer.
The molecule Fc-C6-SH was synthesised based on modified
literature procedures.17–19 Ferrocene was first functionalised
with a 6-bromohexanoyl substituent by the Friedel–Crafts acy-
lation.17,19 The carbonyl group was then reduced with Zn/Hg
in toluene–aqueous HCl.17,19 The bromine atom was finally
converted to the thiol group by the reaction with thiourea.17,18
The compound Fc-C6-SH was characterised by 1H NMR,
ESI-MS and IR.
Results and discussion
Design of redox-active label
The toxin MC-LR contains seven amino acid residues. It
appears that the carboxylic acid groups can be readily functio-
nalised, however, many studies have indicated that the car-
boxylic groups are inactive towards derivatisation due to
steric reasons.3,20 Since the toxin molecule contains an a,b-
unsaturated carbonyl moiety, which is active towards addition
reaction with a nucleophile, derivatisation procedures of
Isolation of MC-LR
Microcystins for derivatisation described in this work were
extracted from Microcystis cell material from a surface bloom
Scheme 1 Reaction scheme for the formation of Fc-MC-LR from MC-LR and Fc-C6-SH.
276
New J. Chem., 2003, 27, 274–279

View Article Online
collected from a farm pond in China and were purified accord-
ing to a modified literature procedure.20 Positive-ion ESI-MS
measurements indicated that the major microcystin derivative
present was MC-LR, which is also the most commonly found
microcystin derivative in nature.3,32 The extracts also con-
tained the MC-YR and MC-RR derivatives in trace amounts.
After a chromatographic separation on silica gel, the fractions
enriched with MC-LR were collected and further purified by
reversed-phase HPLC with UV detection. The toxin MC-LR
was eluted at 19.74 min (Fig. 1). The presence of MC-LR
was confirmed by positive-ion ESI-MS (Fig. 2). The fractions
containing this microcystin derivative were then combined
and the solvent removed under reduced pressure. The toxin
was eventually dissolved in ethanol to make up a stock
solution.
Fig. 2 Positive-ion ESI-MS of MC-LR.
Reaction of MVK and Fc-C6-SH
were collected, the presence of which was confirmed by posi-
tive-ion ESI-MS. The mass spectrum (Fig. 4) showed two
peaks at m/z ¼ 649 and 1298, attributable to the molecular
ions [Fc-MC-LR + 2H+]2+ and [Fc-MC-LR + H+]+, respec-
tively. From the control derivatisation experiments in which
no toxin samples were used, we found that the unreacted ferro-
cenyl derivatisation agent and its redox-active decomposed
products were eluted at ca. 3.44 and 5.02 min, much earlier
than Fc-MC-LR. Independent experiments show that the
labelling method is very repeatable. The peak height-concen-
tration curve was linear in the test range 20–400 ng MC-LR
(r value for linear regression > 0.9987). The detection limit
was determined to be ca. 18 ng MC-LR (S/N ¼ 3). Since the
environmentally relevant concentration of this toxin is in the
mg lꢀ1 scale, a pre-concentration step such as solid-phase
extraction is required for analysis of real samples using the cur-
rent method.3 On the other hand, the detection limit of the cur-
rent system is much lower than those of colorimetry methods
(ca. 1 mg)6 and visualisation reactions for the thin-layer chro-
matographic analysis (ca. 50–100 ng) of these toxins,21 but is
comparable to the typical detection limits of most HPLC/
UV detection systems.3,33,34 Meanwhile, this limit of detection
is relatively high compared to those of other detection and
quantitation methods for this class of toxins such as CE/MS
(ca. 4 pg),4 HPLC/MS detections (ca. 0.1 ng),6 ELISA (ca.
1–150 pg),8,9 and fluorescence and chemiluminescence detec-
tions (ca. 10–20 pg).11,12 Nevertheless, the use of a ferrocene-
containing labelling reagent for electrochemical detection of
microcystins still has its merits because it is a relatively in-
expensive method and the derivatisation procedure is simple,
direct and specific, because the use of a linker molecule such
The reactivity of Fc-C6-SH towards organic molecules con-
taining an a,b-unsaturated ketone moiety has been studied.
MVK was chosen as the model substrate. The reaction of Fc-
C6-SH and MVK was carried out in acetonitrile at room tem-
perature for 12 h. The adduct Fc-MVK, obtained in a good
yield after chromatographic purification, was characterised
1
by H-NMR, positive-ion ESI-MS and IR. In the 1H NMR
spectrum, the resonance signals for the vinyl protons of the
starting material MVK were not observed at d 6.27, 6.26 and
5.94. The S–H stretching absorption band at ca. 2561 cmꢀ1
in the IR spectrum also disappeared. The formation of a
thioether bond between the sulfhydryl group of Fc-C6-SH
and MVK was indicated by the triplets at d 2.61
[SCH2CH2(C=O)CH3] and 2.20 [SCH2CH2(C=O)CH3].
d
The positive-ion ESI measurement gave the adduct ion peak at
m/z ¼ 372.
Derivatisation of MC-LR with Fc-C6-SH
The derivatisation of MC-LR was carried out by reacting MC-
LR with Fc-C6-SH in methanol–sodium carbonate solution at
60 ꢂC for 4 h. The supernatant was isolated from the reaction
mixture by centrifugation and then evaporated to dryness. The
solid was redissolved into water for reversed-phase HPLC/EC
analysis. The operation potential of the electrochemical detec-
tor was +300 mV vs. Ag/AgCl and the mobile phase was
0.1 M aqueous ammonium acetate–acetonitrile (55:45 v/v). At a
flow rate of 1.0 ml minꢀ1, the adduct Fc-MC-LR was eluted at
9.56 min (Fig. 3). The fractions containing the labelled toxin
Fig. 1 HPLC purification of MC-LR with UV detection. Column:
Ultrasphere ODS 250 mm ꢁ 4.6 mm. Mobile phase A: 30% (v/v)
aqueous acetonitrile and 0.05% (v/v) TFA, B: 100% acetonitrile.
Flow rate: 1 ml minꢀ1. Linear gradient: 0–20% B from 0 to 20 min.
Detection: absorbance at 238 nm.
Fig. 3 HPLC separation of Fc-MC-LR with EC detection. Column:
Ultrasphere ODS 250 ꢁ 4.6 mm. Mobile phase: 0.1 M aqueous ammo-
nium acetate–acetonitrile (55:45 v/v). Flow rate: 1 ml minꢀ1. Isocratic
elution. Detection: operation potential at +300 mV vs. Ag/AgCl,
glassy carbon working electrode.
New J. Chem., 2003, 27, 274–279
277

View Article Online
Fig. 4 Positive-ion ESI-MS of Fc-MC-LR.
Fig. 6 Cyclic voltammogram of Fc-MC-LR in 0.1 M aqueous ammo-
.
nium acetate–acetonitrile (55:45 v/v) at 298 K. Sweep rate: 50 mV sꢀ1
Carbon paste working electrode.
as cysteine,11 glutathione21 or 2-aminoethanethiol22 is not
required. In addition, compared to phosphatase inhibition
assays,7,8 the current method also offers higher specificity ori-
ginated from the specific reaction between the thiol group of
Fc-C6-SH and the a,b-unsaturated carbonyl group of MC-LR.
toxin molecules are stable, and hence higher detection specifi-
city can be achieved.
Conclusions
Electrochemical properties of Fc-MVK and Fc-MC-LR
The electrochemical properties of both Fc-MVK and Fc-MC-
LR have been studied by cyclic voltammetry. At a sweep rate,
n, of 50 mV sꢀ1, Fc-MVK exhibits a reversible couple at ca.
+0.333 V vs. SCE in CH3CN (0.1 M nBu4PF6) at 298 K, attri-
butable to the ferrocenium/ferrocene couple (Fig. 5). The far-
adaic currents increase linearly with the square root of the
sweep rates from 10 to 500 mV sꢀ1, suggestive of a diffusion
controlled process.35 Meanwhile, Fc-MC-LR also displays a
reversible ferrocenium/ferrocene couple at ca. ꢀ0.040 V vs.
SCE (n ¼ 50 mV sꢀ1) in 0.1 M aqueous ammonium acetate–
acetonitrile (55:45 v/v) (Fig. 6). The linear relationship
between the faradaic currents and the square roots of the
sweep rates (from 10 to 500 mV sꢀ1) indicates the diffusion-
controlled nature of the electron transfer. It is interesting to
note that native MC-LR is not redox-active due to the lack
of an electrochemically active moiety. HPLC/EC detection
of MC-LR with an operation potential at +1.20 V vs. Ag/
AgCl has been reported.16 While the linear range of this
method (13–250 ng) is similar to that of our system, at such
a relatively high potential, the arginine residue of the toxin is
oxidised irreversibly. In contrary, the use of an electrochemi-
cally reversible ferrocene unit in the present work allows a
much lower operation potential at which amino acids of the
In summary, a redox-active label Fc-C6-SH has been designed
as a derivatisation agent for the toxin microcystins. Since the
sulfhydryl moiety of this compound is reactive towards the
a,b-unsaturated ketone moiety, as revealed by the facile reac-
tion with MVK, Fc-C6-SH has been used to derivatise a micro-
cystin derivative, MC-LR. The ferrocene-toxin conjugate
Fc-MC-LR has been analysed by HPLC coupled with EC
detection. Its electrochemical properties have also been inves-
tigated. It has been shown that Fc-C6-SH is a versatile redox-
active derivatisation reagent for MC-LR. Related studies using
other derivatisation agents are underway.
Acknowledgements
We thank the Hong Kong Research Grants Council (project
numbers 8730011 and CityU1103/00M) and the City Univer-
sity of Hong Kong (project number 7001283) for financial sup-
port. The blue-green algal sample was a gift from Ms. Xiaoyun
Shen. J. S. Y. L. acknowledges the receipt of a postgraduate
studentship and a Research Tuition Scholarship, both adminis-
tered by the City University of Hong Kong.
References
1
K.-I. Harada, Toxic Microcystis, ed. M. F. Watanabe, K.-I.
Harada, W. W. Carmichael and J. Fujiki, CRC Press, Boca
Raton, FL, 1996, p. 103.
2
T. N. Duy, P. K. S. Lam, G. R. Shaw and D. W. Connell, Rev.
Environ. Contam. Toxicol., 2000, 163, 113.
3
4
J. Meriluoto, Anal. Chim. Acta, 1997, 352, 227.
K. P. Bateman, P. Thibault, D. J. Douglas and R. L. White,
J. Chromatogr., A, 1995, 712, 253.
5
6
7
8
9
H. S. Lee, C. K. Jeong, H. M. Lee, S. J. Choi, K. S. Do, K. Kim
and Y. H. Kim, J. Chromatogr., A, 1999, 848, 179.
K. Kaya, T. Sano, H. Inoue and H. Takagi, Anal. Chim. Acta,
2001, 450, 73.
C. J. Ward, K. A. Beattie, E. Y. C. Lee and G. A. Codd, FEMS
Microbiol. Lett., 1997, 153, 465.
O. I. Fontal, M. R. Vieytes, J. M. V. Baptista de Sousa, M. C.
Louzao and L. M. Botana, Anal. Biochem., 1999, 269, 289.
J. McElhiney, L. A. Lawton and A. J. R. Porter, FEMS Micro-
biol. Lett., 2000, 193, 83.
Fig. 5 Cyclic voltammogram of Fc-MVK in CH3CN (0.1
M
nBu4PF6) at 298 K. Sweep rate: 50 mV sꢀ1. Glassy carbon working
electrode.
10 A. Zeck, A. Eikenberg, M. G. Weller and R. Niessner, Anal.
Chim. Acta, 2001, 441, 1.
278
New J. Chem., 2003, 27, 274–279

View Article Online
11 P. C. H. Li, S. Hu and P. K. S. Lam, Mar. Pollut. Bull., 1999,
39, 250.
24 T. Ihara, Y. Maruo, S. Takenaka and M. Takagi, Nucleic Acids
Res., 1996, 24, 4273.
12 H. Murata, H. Shoji, M. Oshikata, K.-I. Harada, M. Suzuki,
F. Kondo and H. Goto, J. Chromatogr., A, 1995, 693, 263.
13 P. Mehto, M. Ankelo, A. Hinkkanen, A. Mikhailov, J. E.
Eriksson, L. Spoof and J. Meriluoto, Toxicon, 2001, 39, 831.
14 A. V. Humble, G. M. Gadd and G. A. Codd, Water Res., 1997,
31, 1679.
25 T. Ihara, M. Nakayama, M. Murata, K. Nakano and M. Maeda,
Chem. Commun., 1997, 1609.
26 K. Di Gleria, C. M. Halliwell, C. Jacob and H. A. O. Hill, FEBS
Lett., 1997, 400, 155.
27 K. Di Gleria, D. P. Nickerson, H. A. O. Hill, L. L. Wong and
V. Fulo¨p, J. Am. Chem. Soc., 1998, 120, 46.
¨
15 F. Yan, M. Ozsoz and O. A. Sadik, Anal. Chim. Acta, 2000, 409,
247.
28 K. K. W. Lo, L. L. Wong and H. A. O. Hill, FEBS Lett., 1999,
451, 342.
16 J. Meriluoto, B. Kincaid, M. R. Smyth and M. Wasberg, J. Chro-
matogr., A, 1998, 810, 226.
29 A. E. Beilstein and M. W. Grinstaff, J. Organomet. Chem., 2001,
637, 398.
17 K. Uosaki, Y. Sato and H. Kita, Langmuir, 1991, 7, 1510.
18 A. S. Viana, L. M. Abrantes, G. Jin, S. Floate, R. J. Nichols and
M. Kalaji, Phys. Chem. Chem. Phys., 2001, 3, 3411.
19 C. J. Yu, H. Wang, Y. Wan, H. Yowanto, J. C. Kim, L. H.
Donilon, C. Tao, M. Strong and Y. Chong, J. Org. Chem.,
2001, 66, 2937.
20 K.-I. Harada, M. Suzuki, A. M. Dahlem, V. R. Beasley, W. W.
Carmichael and K. L. Rinehart, Jr., Toxicon, 1988, 26, 433.
21 A. Pelander, I. Ojanpera¨, K. Lahti, K. Ninivaara and E. Vuori,
Water Res., 2000, 34, 2643.
30 C. J. Yu, Y. J. Wan, H. Yowanto, J. Li, C. L. Tao, M. D. James,
C. L. Tan, G. F. Blackburn and T. J. Meade, J. Am. Chem. Soc.,
2001, 123, 11 155.
31 K. K. W. Lo, J. S. Y. Lau, D. C. M. Ng and N. Zhu, J. Chem.
Soc., Dalton Trans., 2002, 1753.
32 L. A. Lawton and P. K. J. Robertson, Chem. Soc. Rev., 1999, 28,
217.
33 K.-I. Harada, K. Matsuura, M. Suzuki, H. Oka, M. F. Watanabe,
S. Oishi, A. M. Dahlem, V. R. Beasley and W. W. Carmichael,
J. Chromatogr., 1988, 448, 275.
22 F. Kondo, Y. Ikai, H. Oka, M. Okumura, N. Ishikawa, K.
Harada, K. Matsuura, H. Murata and M. Suzuki, Chem. Res.
Toxicol., 1992, 5, 591.
23 G. Moorhead, R. W. MacKintosh, N. Morrice, T. Gallagher and
C. MacKintosh, FEBS Lett., 1994, 356, 46.
34 L. A. Lawton, C. Edwards and G. A. Codd, Analyst, 1994, 119,
1525.
35 A. J. Bard and L. R. Faulkner, Electrochemical Methods, Funda-
mentals and Applications, Wiley, New York, USA, 2nd edn.,
2001.
New J. Chem., 2003, 27, 274–279
279