Use of a structural analogue v rsus a stable isotope
labeled internal standard for t e quantification of
angiotensin IV in rat brain dialysates using nano-liquid
ss
old
ceut
anua
ides
id c
ver
s kn
les,
epti
er, w
. As
f a s
stable isotope labeled (SIL) analogue. Linearity of the method was improved when either of the
Neurope
pounds
signals in
effects of
techniqu
concentr
efficiency
is requir
nated to
requirem
The combination of nanospray ionization with a triple
simul-
As pre-
if target
anoLC-
lutions,
eptides,
ment of
nization
fect the
accuracy and precision of the method. An internal standard
RAPID COMMUNICATIONS IN MASS SPECTROMETRY
Rapid Commun. Mass Spectrom. 2007; 21: 1187–1195
Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.2950ders (FWO-Vlaanderen) and the Koningin Elisabeth Stichting.quadrupole mass spectrometer, used in the selected reaction
monitoring (SRM) mode, is suitable for quantitative and
selective analysis. Miniaturized LC results in higher
responses than regular LC if the same amount of analyte
is injected. However, the injection volumes on miniaturized
columns are limited so that the samples often need to be
preconcentrated. This can be performed by applying the
(IS) may correct for these matrix effects and for random
variations associated with the LC/MS analysis, such as
variations in injection volumes, fluctuations of the ionization
process and the ion transmission and detector sensitivity of
the mass spectrometer.3 Consequently, the use of an IS is
expected to improve the accuracy and precision of the
analytical method.
*Correspondence to: Y. Michotte, Department of Pharmaceutical
Chemistry, Drug Analysis and Drug Information, Vrije Univer-
siteit Brussel, Laarbeeklaan 103, 1090 Brussels, Belgium.
E-mail: ymichot@vub.ac.be
Contract/grant sponsor: The Research Fund (OZR) from the
Vrije Universiteit Brussel, the Fund for Scientific Research Flan-proposed ISs were applied. Only when using the SIL-IS could the repeatability of injection and the
method’s precision and accuracy be improved. Finally, both ISs were able to correct for degradation
of Ang IV in dialysates, prolonging the possible storage period of the samples. We conclude that the
structural analogue is not suited as an IS and that the application of a SIL analogue is indispensable
when quantifying Ang IV in dialysates using nanoLC/ESI-MS/MS detection. Copyright# 2007 John
Wiley & Sons, Ltd.
ptides constitute a large family of bioactive com-
participating in the modulation or transmission of
the central nervous system.1 To explore the central
neuropeptides, the in vivo microdialysis sampling
e is a powerful tool. Because of the low extracellular
ations of neuropeptides and their low microdialysis
, a highly sensitive and selective analysis technique
ed. Nano-liquid chromatography (nanoLC) hyphe-
tandem mass spectrometry (MS/MS) fulfills the
ents to quantify these low concentrated peptides.
column-switching technique, which also allows
taneous desalting of the microdialysis samples.
viously described, the latter technique is necessary
analytes need to be quantified in dialysates using n
ESI-MS/MS detection.2
Although dialysates are protein-free aqueous so
they are complex mixtures of small molecules and p
which can induce ionization suppression or enhance
the analyte of interest when applying electrospray io
(ESI). Dialysate matrix effects can profoundly afchromatography/tandem ma
Katrien Lanckmans, Sophie Sarre, Ilse Sm
Research Group Experimental Pharmacology, Department of Pharma
Vrije Universiteit Brussel, Brussels, Belgium
Received 18 December 2006; Revised 26 January 2007; Accepted 26 J
Quantifying low concentrations of neuropept
sensitive analysis technique, such as nano-liqu
mass spectrometry (nanoLC/ESI-MS/MS). Howe
due to matrix effects. Indeed, ESI-MS detection i
dialysates are complex mixtures of small molecu
influence the ionization efficiency of the neurop
the samples. In the study reported in this pap
standard (IS) can correct for these matrix effects
angiotensin IV (Ang IV). We compared the use ospectrometry
ers and Yvette Michotte*
ical Chemistry, Drug Analysis and Drug Information,
ry 2007
in microdialysates requires a selective and
hromatography/electrospray ionization tandem
, we observed reduced accuracy of the method
own to be sensitive to matrix effects. Moreover,
peptides and other matrix compounds that can
de of interest and the stability of the peptide in
e investigated whether the use of an internal
a model compound for neuropeptides we used
tructural analogue (norleucine1-Ang IV) with ae
hCopyright # 2007 John Wiley & Sons, Ltd.

Only a few papers describe the use of microdialysis
performance of the nanoLC/MS/MS assay for the quanti-
from NeoMPS (Strasbourg, France). Norleucine1-Ang IV
micro-switching module (SwitchosTM, LC Packings) loaded
clean-up, the compounds of interest were eluted in back-
Air Liquide) was used as collision gas and the dwell time and
1188 K. Lanckmans et al.(Nle-Tyr-Ile-His-Pro-Phe) was custommade by Prof. Tourwe´
(Vrije Universiteit Brussel, Brussels, Belgium). The SIL
analogue of Ang IV (Val-Tyr-Ile (13C6,
15N)-His-Pro-Phe,
98% purity) was synthesized by Sigma-Aldrich (St. Louis,
MO, USA).
Acetonitrile (ACN, LC grade) and formic acid (FA,
analytical-reagent grade) were supplied by Acros Organics
(NJ, USA). The mQ water used for preparing solutions was
generated using a Seralpur Pro 90 CN purification system
(Seral, Ransbach-Baumbach, Germany). CaCl2.6H2O was
bought from Fluka (Buchs, Switzerland) and NaCl and KCl
from Merck (Darmstadt, Germany).
In all experiments, the drugs were dissolved in modified
Ringer’s solution (147mM NaCl, 4mM KCl and 2.3mM
CaCl2.6H2O). Unless otherwise stated, dialysates spiked
with Ang IV were analyzed immediately after preparation
because of stability issues (see below).
Column-switching nanoLC
The mobile phase was pumped at a flow rate of 300 nL/min
by an UltimateTM quaternary low-pressure gradient pump
(LC Packings, Dionex, Amsterdam, The Netherlands). Injec-
tions of 10mL were made in partial mode (20mL loop) using
a refrigerated (158C, precautionary measure) micro-
autosampler (FamosTM, LC Packings). The needle wash sol-
vent was awater/ACN (50:50 v/v) solution. After injection, a
Copyright # 2007 John Wiley & Sons, Ltd.fication of neuropeptides in dialysates. We compared the use
of a structural analogue with that of a stable isotope labeled
(SIL) analogue as IS. It is generally accepted that SIL stan-
dards yield better assay performance data for quantitative
LC/MS analysis because they behave similarly to the target
analyte during sample preparation and also in the ion source
of the mass spectrometer. However, SIL standards are not
readily available, they are expensive and sometimes they do
not properly correct for matrix effects.8,9 Moreover, the use of
deuterated compounds can result in different retention times
due to differences in physicochemical properties.10 As struc-
tural analogue we chose the metabolically more stable Ang
IV analogue, norleucine1-Ang IV (Nle1-Ang IV) and as SIL
analogue a 13C6,
15N-labeled analogue. Validation para-
meters, such as linearity, repeatability of injections, accuracy
and precision, were compared. Finally, the stability of Ang IV
in dialysates was investigated.
EXPERIMENTAL
Chemicals
Angiotensin IV (Val-Tyr-Ile-His-Pro-Phe) was purchasedsampling followed by nanoLC/MS/MS to detect neuropep-
tides in vivo.4–7 However, in these valuable papers, little
attention was paid to quantification and validation in terms
of precision and accuracy. Furthermore, the use of an IS was
not considered either.
In this study, we used angiotensin IV (Ang IV), an effector
peptide of the renin-angiotensin system, as a model com-
pound to investigate whether an IS can improve the overallinter-channel delay were respectively 300 and 20ms. The
source and analyzer settings were as follows: no desolvation
and cone gas were used; source temperature 808C; extractor
voltage 3.00V; RF lens voltage 0.00V; entrance and
exit voltages of the collision cell, respectively, 1.00 and
1.00V. The pressure in the gas cell fluctuated between 3.50
and 3.75
103mbar. The resolutions of the first and second
quadrupoles were, respectively, 15.0/15.0 (HM/LM) and
13.5/13.5 (HM/LM) and the ion energies were 0.6 and 0.7.
The multiplier voltage was 650V. The following transitions
were monitored: 775.3> 513.3 (Ang IV); 782.4> 520.4
(SIL-Ang IV) and 789.3> 527.3 (Nle1-Ang IV).
All measurements were controlled by Masslynx1 version
4.0 operating software (Micromass-Waters). Integration of
the chromatograms after smoothing (window size¼ 3;
number of smooths¼ 2) was performed by Quanlynx
(Masslynx).
Microdialysis
The microdialysis experiments were carried out on freely
moving rats, according to the national guidelines on animal
experimentation and were approved by the Ethical Com-
mittee for Animal Experiments of the Faculty of Medicine
Rapid Commun. Mass Spectrom. 2007; 21: 1187–1195flush mode and directed to the C18 separation column
(15 cm
75mm i.d., 5mmparticles, PepmapTM, LC Packings).
Gradient elution was performed using solvent A (water/
ACN/FA 98:2:0.1 v/v/v) and solvent B (water/ACN/FA
20:80:0.1 v/v/v). The solvents for the mobile phase
were degassed for 15min on a Branson 2000 ultrasonic
bath (Danbury, CT, USA) after preparation and by helium
(Alphagaz 21, Air Liquide, Liege, Belgium) during the
analyses. The linear LC gradient for separation was 100% A
for 5min (loading period), 0–100% B in 8min, 100% B for
5min, 100–0% B in 0.5min and 100%A for 16.5min. The total
sample analysis time was 35min.
The LC system was connected to an UltimateTM UV
detector, equipped with a U-Z ViewTM nano flow cell (3 nL)
(LC Packings) for troubleshooting purposes. All LC parts
were controlled by Chromeleon1 software (Dionex, Amster-
dam, The Netherlands).
Nanospray MS/MS detection
The outlet of the LC system was hyphenated to the nano-
source of a Quattro PremierTM triple quadrupole mass
spectrometer (Micromass-Waters, Manchester, UK), using
distal coated Picotip1 nanospray emitters (10mm i.d.,
New Objective, MA, USA). Detection was performed in ESI
positive mode and quantification was executed in selected
reaction monitoring (SRM) mode. Argon (Alphagaz 21,the samples for 5min onto a C18 precolumn (5mm
300mm
i.d., 5mm particles, PepmapTM, LC Packings) at a flow rate of
30mL/min, to trap the compounds of interest and to direct
the hydrophilic contaminants (mainly salts) to waste. The
loading solvent, delivered by the SwitchosTM pump, consis-
ted of water/ACN/FA (98:2:0.1 v/v/v). After the sampleDOI: 10.1002/rcm