Diethylation labeling combined with UPLC/MS/MS for simultaneous determination of a panel of monoamine neurotransmitters in rat prefrontal cortex microdialysates

Article abstract of DOI:10.1021/ac801339z

The primary challenge associated with the development of an LC/MS/MS-based assay for simultaneous determination of biogenic monoamine neurotransmitters such as norepinephrine (NE), dopamine (DA), serotonin (5-HT), and normetanephrine (NM) in rat brain microdialysates is to improve detection sensitivity. In this work, a UPLC/MS/MS-based method combined with a diethyl labeling technique was developed for simultaneous determination of a panel of monoamines in rat prefrontal cortex microdialysates. The chromatographic run time is 3.5 min/sample. The limits of detection of the UPLC/MS/MS-based method for NE, DA, 5-HT/ and NM, with/without diethyl labeling of monoamines, are 0.005/0.4 (30/2367 pM), 0.005/0.1 (33/653 pM), 0.005/0.2 (28/1136 pM), and 0.002/0.2 ng/mL (11/1092 pM), respectively. Diethyl labeling of amino groups of monoamines affords 20-100 times increased detection sensitivity of corresponding native monoamines during the UPLC/MS/MS analysis. This could result from the following: (1) improved fragmentation patterns; (2) increased hydrophobicity and concomitantly increased ionization efficiency in ESI MS and MS/MS analysis; (3) reduced matrix interference. This labeling reaction employs a commercially available reagent, acetaldehyde-d₄, to label the amine groups on the monoamines via reductive animation. It is also simple, fast (～25-min reaction time), specific, and quantitative under mild reaction conditions. Data are also presented from the application of this assay to monitor the drug-induced changes of monoamine concentrations in rat prefrontal cortex microdialysate samples followed by administration of SKF 81297, a selective D₁ dopamine receptor agonist known to elevate the extracellular level of the neurotransmitters DA and NE in the central nervous system.

Full text of DOI:10.1021/ac801339z

Anal. Chem. 2008, 80, 9195–9203
Diethylation Labeling Combined with UPLC/MS/MS
for Simultaneous Determination of a Panel of
Monoamine Neurotransmitters in Rat Prefrontal
Cortex Microdialysates
Chengjie Ji,*,† Wenlin Li,‡ Xiao-dan Ren,† Ayman F. El-Kattan,† Rouba Kozak,† Scott Fountain,‡
and Christopher Lepsy†
Pfizer Global Research & Development, Groton Laboratories, Pfizer Inc., Groton, Connecticut 06340, and La Jolla
Laboratories, Pfizer Inc., La Jolla, California 92121
The primary challenge associated with the development
of an LC/MS/MS-based assay for simultaneous determi-
nation of biogenic monoamine neurotransmitters such as
norepinephrine (NE), dopamine (DA), serotonin (5-HT),
and normetanephrine (NM) in rat brain microdialysates
is to improve detection sensitivity. In this work, a UPLC/
MS/MS-based method combined with a diethyl labeling
technique was developed for simultaneous determination
of a panel of monoamines in rat prefrontal cortex microdi-
alysates. The chromatographic run time is 3.5 min/
sample. The limits of detection of the UPLC/MS/MS-
based method for NE, DA, 5-HT/ and NM, with/without
diethyl labeling of monoamines, are 0.005/0.4 (30/2367
pM), 0.005/0.1 (33/653 pM), 0.005/0.2 (28/1136
pM), and 0.002/0.2 ng/mL (11/1092 pM), respectively.
Diethyl labeling of amino groups of monoamines affords
20-100 times increased detection sensitivity of corre-
sponding native monoamines during the UPLC/MS/MS
analysis. This could result from the following: (1) im-
proved fragmentation patterns; (2) increased hydropho-
bicity and concomitantly increased ionization efficiency
in ESI MS and MS/MS analysis; (3) reduced matrix
interference. This labeling reaction employs a commer-
cially available reagent, acetaldehyde-d4, to label the
amine groups on the monoamines via reductive amina-
tion. It is also simple, fast (∼25-min reaction time),
specific, and quantitative under mild reaction conditions.
Data are also presented from the application of this assay
to monitor the drug-induced changes of monoamine
concentrations in rat prefrontal cortex microdialysate
samples followed by administration of SKF 81297, a
selective D1 dopamine receptor agonist known to elevate
the extracellular level of the neurotransmitters DA and
NE in the central nervous system.
norepinephrine (NE), and normetanephrine (NM) in the rat brain
is an important step in the process of discovery and evaluation of
new drugs designed to treat disorders of the central nervous
system, which are related to these biogenic monoamine neu-
rotransmitters.1 In vivo microdialysis is currently one of the main
techniques used to continuously monitor the effects of drugs that
act on extracellular pools of rat brain monoamines and their
metabolites.2 However, basal levels of monoamines in rat brain
microdialysates are typically in the low picogram per milliliter
range, which makes analytical measurement a challenge.
Various methods for determination of 5-HT, DA, NE, and their
3-10
metabolites have been reported.
Among them, the most
frequently used methods for monitoring NE, DA, 5-HT, and their
metabolites are high-performance liquid chromatography coupled
with electrochemical detection (HPLC-EC)3,6 and HPLC coupled
with fluorescence detection (HPLC-FL).8-10 HPLC-EC offers a
relatively simple and highly sensitive analytical method. However,
it is difficult to use this technique for simultaneous determination
of all four monoamines in microdialysis samples.8 Alternatively,
HPLC-FL provides a means of simultaneous determination of
monoamines in mircodialysis samples. In addition to the need for
multiple sample treatment steps, the lengthy HPLC run time
(more than 40 min/analytical run) is a drawback of this technique
when it is used for simultaneous determination of a panel of
monoamines in microdialysis samples.9,10 Recently, Jung and co-
workers have reported a novel approach for simultaneous deter-
mination of biogenic monoamines in rat brain striatum within a
(1) Chi, J. D.; Odontiadis, J.; Franklin, M. J. Chromatogr., B: Biomed. Sci. Appl.
1999, 731, 361–367
(2) Chaurasia, C. S.; Chen, C. E.; Ashby, C. R., Jr J. Pharm. Biomed. Anal.
1999, 19, 413–422
(3) Kissinger, P. T.; Bruntlett, C. S.; Shoup, R. E. Life Sci. 1981, 28, 455–465
(4) Matsumoto, M.; Togashi, H.; Yoshioka, M.; Hirokami, M.; Morii, K.; Saito,
H. J. Chromatogr. 1990, 526, 1–10
(5) Michael, D.; Travis, E. R.; Wightman, R. M. Anal. Chem. 1998, 70, 586A–
592A
(6) Nagatsu, T. J. Chromatogr. 1991, 566, 287–307
(7) Perez, X. A.; Andrews, A. M. Anal. Chem. 2005, 77, 818–826
(8) Yoshitake, T.; Kehr, J.; Todoroki, K.; Nohta, H.; Yamaguchi, M. Biomed.
Chromatogr. 2006, 20, 267–281
.
.
.
.
.
.
.
Simultaneous determination of changes in basal levels of
biogenic monoamines such as serotonin (5-HT), dopamine (DA),
.
(9) Yoshitake, T.; Kehr, J.; Yoshitake, S.; Fujino, K.; Nohta, H.; Yamaguchi,
* To whom correspondence should be addressed. E-mail: Chengjie.ji@
M. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2004, 807, 177–
pfizer.com.
183
(10) Yoshitake, T.; Yoshitake, S.; Fujino, K.; Nohta, H.; Yamaguchi, M.; Kehr,
J. J. Neurosci. Methods 2004, 140, 163–168
.
† Groton Laboratories.
‡ La Jolla Laboratories.
.
10.1021/ac801339z CCC: $40.75 2008 American Chemical Society
Published on Web 10/25/2008
Analytical Chemistry, Vol. 80, No. 23, December 1, 2008 9195

short analysis time of less than 10 min.11 Capillary HPLC coupled
with photoluminescence following electron transfer (PFET) offers
detection limits comparable to those of HPLC-EC and HPLC-FL
and overcomes the shortcomings associated with HPLC-CE (e.g.,
electrode fouling) and HPLC-FL (e.g., need for a pretreatment
step before sample injection).
In contrast to HPLC-CE, HPLC-FL and capillary HPLC-PFET,
in which analytes can only be identified by retention time
matching, LC/MS/MS can provide high specificity due to the
additional structure information obtained using tandem mass
spectrometric analysis. However, without a pretreatment step, LC/
MS/MS cannot offer adequate sensitivity for simultaneous mea-
surement of monoamines in rat brain microdialysates. A recently
reported LC/MS/MS-based assay by Hows and co-workers was
not able to determine basal levels of NE and 5-HT in rat brain
microdialysis due to its poor sensitivity.12
Chemical derivatization of amine groups on amino acids with
different chemical reagents has been shown to enhance amino
acid detection by ESI MS analysis.13-16 Differential isotopic
dimethyl labeling of N-terminal peptides with d(0)- and d(2)-, or
d(0), 12C-, and d(2), 13C-formaldehyde combined with LC-ESI or
LC-MALDI was used successfully in quantitative proteomics.17-20
N-Terminal dimethyl labeling was also reported for peptide
sequencing in combination with database searching and de novo
peptide sequencing of neuropeptides directly from tissue extract
without any genomic information.21-23 More recently, Guo and
co-workers reported an approach for quantitative profiling of
amine-containing metabolites in biological samples using LC-ESI
MS in conjunction with stable-isotope dimethyl labeling.24 The
labeling is carried out by using reductive amination chemistry.25
This chemical labeling has been reported to be simple, fast (less
than 10-min reaction time), specific, and provide high yields under
mild reaction conditions when used to label peptides and amine-
contained metabolites.17,24
the rat prefrontal cortex microdialysates following administration
of SKF-81297, a selective D1 dopamine receptor agonist known
to elevate the extracellular level of the neurotransmitter DA and
NE in the central nervous system.
EXPERIMENTAL SECTION
Chemicals and Reagents. (±)-Normetanephrine- R, ꢀ, ꢀ-d3
(d3-NM) HCl (98.7 atom % D), serotonin-a,a,b,b-d4 (d4-5-HT)
creatinine sulfate complex, and 2-(3,4-dihydroxyphenyl)ethyl-
1,1,2,2-d4-amine (d4-DA) HCl were purchased from C/D/N
Isotopes Inc. (Quebec, Canada). Sodium acetate, acetic acid,
ammonium acetate, HPLC-grade water, and methanol were
obtained from Mallinckrodt Becker (Phillipsburg, NJ). Formic acid
(96%) was analytical grade and obtained from Acros Organics.
Perfusion Fluid CNS was purchased from CMA Microdialysis Inc.
(North Chelmsford, MA). All other chemicals and reagents were
purchased from Sigma (St. Louis, MO) unless otherwise stated.
Reductive Amination. The derivatization of NE, DA, 5-HT,
and NM was adopted and modified from Guo et al.24 To profile
the derivatized monoamines, 10 µL of acetate buffer (1 M, pH 5)
and 20 µL of cyanoborohydride coupling buffer (containing 0.02
M sodium phosphate, pH 7.5, containing 0.2 M sodium chloride
and 3.0 g/L sodium cyanoborohydride) was added to four different
1.5-mL tubes, each containing 40 µL of 0.4 µg/mL NE, DA, 5HT,
and NM dissolved in HPLC-grade water, repectively. Then, the
mixture solution was vortexed and another 10 µL of 20% acetal-
dehyde-d4 was added to each tube. The final mixture solution was
vortexed and incubated at 37 °C for 25 min, followed by LC/MS
and LC/MS/MS analysis as described in the following sections.
To examine the effect of the concentration of cyanoborohydride
on the labeling efficiency, 0, 5, 10, 20, 30, and 40 µL of
cyanoborohydride coupling buffer were added into different wells
of a 96-well plate. Each well contained 30 µL of standard solution,
with a concentration of 1 ng/mL for each monoamine of interest,
which was prepared by diluting the stock solution using artificial
CSF/preservative solution (5 µM EDTA and 100 µM citric acid)
(5:1, v/v). Then the solution in each well was brought up to the
same volume by adding 35, 30, 20, 10, and 0 µL of artificial CSF/
preservative solution (5 µM EDTA and 100 µM citric acid) (5:1,
v/v), followed by the addition of 20 µL of 1 M acetate buffer and
10 µL of 25% acetaldehyde-d4. The resulted mixture was vortexed
and incubated at 37 °C for 25 min and finally analyzed using the
optimized LC-MRM method. To examine the effect of reaction
time on the labeling efficiency, 200 µL of freshly prepared labeling
reagent mixture was added into 300 µL of the same standard
solution that was used to check the effect of the concentration of
cyanoborohydride on the labeling efficiency. Then, the combined
solution was vortexed and incubated in the UPLC sample manage-
ment system, which had been set at 37 °C for 30 min before
conducting this experiment. The reaction solution was analyzed
using optimized UPLC-MRM method (see below) every 5 min
for 60 min. For the real sample analysis, to increase the
throughput, 25 µL of freshly prepared calibration standard solu-
tions, QC samples, and rat prefrontal cortex microdialysate
samples were added into different wells of a 96-well plate. Then,
10 µL of internal standard solution was added, followed by the
addition of 20 µL of freshly prepared labeling reagent mixture,
into each well containing standards, QCs, or rat prefrontal cortex
microdialysate samples. The resulting solutions in the 96-well plate
In this work, we report the successful combination of UPLC/
MS/MS with diethyl labeling of each primary amine group on
monoamine neurotransmitters, via reductive amination, for simul-
taneous measurement of a panel of monoamine neurotramitters
in rat prefrontal cortex microdialysates. A practical example of
the developed method is presented in which elevated levels of
NE, DA, and 5-HT, NM were detected over a 4.5-h time course in
(11) Jung, M. C.; Shi, G.; Borland, L.; Michael, A. C.; Weber, S. G. Anal. Chem.
2006, 78, 1755–1760
(12) Hows, M. E.; Lacroix, L.; Heidbreder, C.; Organ, A. J.; Shah, A. J. J. Neurosci.
Methods 2004, 138, 123–132
(13) Johnson, D. W. Rapid Commun. Mass Spectrom. 2001, 15, 2198–2205
(14) Liu, Z.; Minkler, P. E.; Lin, D.; Sayre, L. M. Rapid Commun. Mass Spectrom.
2004, 18, 1059–1065
(15) Qulrke, J. M. E.; Adams, C. L.; Van Berkel, G. J. Anal. Chem. 1994, 66,
1302–1315
(16) Yang, W. C.; Mirzaei, H.; Liu, X.; Regnier, F. E. Anal. Chem. 2006, 78,
4702–4708
(17) Hsu, J. L.; Huang, S. Y.; Chow, N. H.; Chen, S. H. Anal. Chem. 2003, 75,
6843–6852
(18) Ji, C.; Guo, N.; Li, L. J. Proteome Res. 2005, 4, 2099–2108
(19) Ji, C.; Li, L. J. Proteome Res. 2005, 4, 734–742
(20) Ji, C.; Li, L.; Gebre, M.; Pasdar, M.; Li, L. J. Proteome Res. 2005, 4, 1419–
1426
(21) Fu, Q.; Li, L. Anal. Chem. 2005, 77, 7783–7795
(22) Ji, C.; Lo, A.; Marcus, S.; Li, L. J. Proteome Res. 2006, 5, 2567–2576
(23) Hsu, J. L.; Huang, S. Y.; Shiea, J. T.; Huang, W. Y.; Chen, S. H. J. Proteome
Res. 2005, 4, 101–108
(24) Guo, K.; Ji, C.; Li, L. Anal. Chem. 2007, 79, 8631–8638
(25) Means, G. E.; Feeney, R. E. Biochemistry 1968, 7, 2192–2201
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9196 Analytical Chemistry, Vol. 80, No. 23, December 1, 2008

were centrifuged, vortexed, and incubated at 37 °C for 25 min,
followed by LC-MRM analyses. The labeling reagent mixture was
prepared on the ice bath immediately before chemical labeling
by mixing 600 µL of cyanoborohydride coupling buffer (containing
0.02 M sodium phosphate, pH 7.5, 0.2 M sodium chloride and 3.0
g/L sodium cyanoborohydride) with 200 µL of 25% acetaldehyde-
d4 in 0.75 M acetate buffer (pH 5). Caution: because sodium
cyanoborohydride is a highly toxic chemical that will produce
hydrogen cyanide gas when exposed to acid, the reductive
amination was carried out in a fume hood.
Chromatography. A Waters Acquity Ultra Performance LC
(UPLC) system equipped with binary solvent management and
sample management with a programmable column manager
(Waters, Milford, MA) was used for liquid chromatography. In
brief, 20 µL of the final solution after reductive amination was
injected onto a 1.7-µm particle size, 2.1 × 100 mm Acquity UPLC
BEH phenyl column (Waters). The column temperature was set
at 35 °C. Solvent A consisted of 0.2% formic acid and 20 mM
ammonium acetate in water, and solvent B consisted of methanol.
Gradient was maintained at 7% B for 0.5 min, increased from 7%
B to 70% B in 2.7 min, then decreased back to the initial
composition with 0.1 min, and held it for 0.2 min. The eluates
were electrosprayed into an API 4000 mass spectrometer at a flow
rate of 0.3 mL/min.
Preparation of Calibration Standards and IS. A 10 µg/mL
stock solution of the NE, DA, 5-HT, and NM standards was
prepared in HPLC-grade water/methanol (v/v ) 80/20) in a glass
volumetric flask and stored at a -20 °C freezer. To characterize
the analytical performance of UPLC/MS/MS-based methods for
quantitative measurement of nonderivatized and d8-diethyl-labeled
monoamines, this stock solution was serially diluted using artificial
CSF/preservative solution (5 µM EDTA and 100 µM citric acid)
(5:1, v/v) to prepare standards at 100, 50, 25, 10, 5, 2.5, 1, 0.4, 0.2,
0.1, 0.04, 0.02, 0.01, 0.005, 0.002, and 0.001 ng/mL for each
monoamine of interest. For the analysis of rat brain prefrontal
cortex microdialysate samples, the stock solution was serially
diluted on the day of assay using artificial CSF/preservative
solution (5 µM EDTA and 100 µM citric acid) (5:1, v/v) to prepare
calibration standards at 1, 0.4, 0.2, 0.1, 0.04, 0.02, and 0.01 ng/
mL for each monoamine of interest. Two IS stock solutions were
prepared in HPLC-grade water/methanol (v/v ) 80/20): one for
monoamine analysis using UPLC/MS/MS without derivatization,
which contains 4 µg/mL d4-DA, 10 µg/mL d4-5-HT, and 20 µg/
mL d3-NM; the other for monoamine analysis using UPLC/MS/
MS with diethyl labeling, which contains 200 ng/mL d4-DA, 200
ng/mL d4-5-HT, and 40 ng/mL d3-NM. These internal standard
stock solutions were diluted 100 times on the day of the assay
using artificial CSF/preservative solution (5 µM EDTA and 100
µM citric acid) (5:1, v/v).
Mass Spectrometry. An API 4000 triple quadrupole mass
spectrometer (MDS-Sciex, Toronto, Canada) controlled by the
Analyst 1.4.2 software (Applied Biosystems, Foster City, CA) was
used in this study. In all conditions, the following parameters
remained the same: electrospray capillary voltage was 4.6 kV;
source temperature was 550 °C; curtain gas was set at 35; and
nebulizing, desolvation gas flow rates were set at 85. For LC/MS
and LC/MS/MS analyses of derivatized and nonderivatizated NE,
NM, DA, and 5-HT, declustering potential was set at 20 eV. For
LC-MRM analysis of nonderivatizated and derivatized NE, NM,
DA, and 5-HT, declustering potential was set at 20 and 35 V,
respectively. CAD was set at 8. The resolutions of Q1 and Q3
were set at unit. The dwell time was set at 20 ms for each SRM
transition.
Data Analysis. Nonderivatized and derivatized NE, DA, 5HT,
NM, and their IS were detected specifically by monitoring UPLC
elution times and ion pairs corresponding to the precursor and
specific product ion mass/charge ratios (m/z) on a triple-
quadrupole mass spectrometer operated in the positive ion SRM
mode. Data acquisition and integration of the peak areas of the
SRM signal for NE, DA, NM, and 5HT and their stable labeled
internal standards were performed using the standard quantitative
analysis package of the Analyst 1.4.2 software (Applied Biosys-
tems). All regression analysis and sample quantification can be
performed using the Analyst 1.4.2 software or the regression utility
of the Watson LIMS database software. The calibration curve was
Microdialysis. MAB 4.PES microdialysis probes (200-µm o.d.,
3 mm long) were equipped with hollow fiber dialysis membrane
(6000 Da cutoff polyether sulfone; SciPro Inc.) and fused-silica
inlet/outlet lines (180-µm o.d. made from PEEK). The microdi-
alysis probe was implanted into the rat prefrontal cortex, a brain
region innervated by dopamine neurons, over a period of 30 min
to the following coordinates: 2.9 mm anterior to bregma, 0.5 mm
lateral from midline, and 0.6 mm below dura. The probe was
perfused with aCSF at 1.5 µL/min for at least 3 h before samples
were collected for analysis. SKF-81297 (a selective D1 dopamine
receptor antagonist (1.0 mg/kg)) was prepared in 20% cyclodextrin
and delivered through subcutaneous injection. Four consecutive
dialysates were collected from the prefrontal cortex of a freely
moving rat at an interval of 30 min to monitor basal levels of
monoamines. Six consecutive dialysates were collected followed
by drug administration. All brain dialysates were collected in vials
containing a volume of 5 µM EDTA and 100 µM citrate equal to
20% of the intended sample volume. The typical volume of the
dialysis samples was 45 µL unless otherwise noted. The dialysate
samples were stored in an ice bath for immediate analysis or
frozen at -80 °C for future use. Note: all experiments involving
animals were carried out in compliance with national legislation
and subject to local ethical review.
RESULTS AND DISCUSSION
Reductive Amination of Monoamines. Reductive amination
of primary and secondary amine groups in peptides and proteins
with formaldehyde has been previously reported.17-19,25 More
recently, Guo et al. has reported that there are several attractive
features in using differential dimethyl labeling for relative
quantification of primary and secondary amine-containing
metabolites in biological samples.24 These features include the
following: (1) the experimental conditions for reductive ami-
nation are extremely mild, (2) a high yield in a relatively short
constructed based on the response ratio of peak area (Panalyte
/
PIS) versus nominal standard concentrations by either least-
squares linear regression or quadratic regression using a weight-
ing factor of either 1/concentration or 1/concentration2 (see Table
3 for details). Concentrations of NE, DA, NM, and 5-HT were
determined using the response ratio from unknown samples and
the linear regression curve.
Analytical Chemistry, Vol. 80, No. 23, December 1, 2008 9197

Scheme 1. Reductive Amination of DA, NE, and NM with Acetaldehyde-d4
reaction time (10 min), and (3) the reaction is easily completed
without any special reagents or reaction equipment. In this
study, we extended the application of reductive amination in
the enhancement of the detection limit of a conventional LC/
MS/MS for simultaneously quantitative measurement of a panel
of monoamines (NE, DA, NM, 5-HT) in the rat prefrontal cortex
microdialysates. We investigated labeling amine groups on the
monoamine neurotransmitters with formaldehyde, formaldehyde-
d2, 13C-formaldehyde-d2, 13C-formaldehyde, acetaldehyde, ac-
etaldehyde-d3, and acetaldehyde-d4 via reductive amination.
Finally, acetaldehyde-d4 was selected based on the following
two criteria: (1) no possible interference peaks were present
from the used labeling reagents or other endogenous bio-
chemicals in microdialysis samples and (2) low-level back-
ground noise. Schemes 1 and 2 show chemical reaction process
of diethyl labeling NE, DA, 5-HT, and NM with acetaldehyde-
d4. An intermediate Schiff base is formed at the initial step of
this reaction process. Two main possible reaction pathways
have been proposed (see Scheme 1(I),(II) and Scheme 2(I),(II)).
In pathway I, the electron-donating abilities of the benzene rings
attack carbon cations to form a six-member ring, which is
stable, while in pathway II, the intermediate Schiff base is
reduced by sodium cyanoborohydride. Then, the resulted
secondary amines in pathways I and II form Schiff bases again,
which are subsequently reduced by sodium cyanoborohydride
to produce products A and B. Comparing the LC/MS spectra
of diethyl-labeled NE, DA, NM, and 5-HT (Figure 2E-H), we
can see, for NM, NE, and 5-HT, the main products are produced
with the addition of 64 Da following pathway I; while, for DA,
two products, following pathways I and II with the addition of
electronic accepting ability of CH3 replaced the H of one of
the phenol group making the formation of another six-member
ring at the decreasing rate, DA . NE g NM. The electron-
donating ability of 10-member ring in 5-HT is weaker than that
of the 6-member ring in DA. Therefore, the main product for
5-HT derivatizated with acetaldehyde-d4 is also B with the
addition of 64 Da.
We also noted that the diethyl labeling reaction is pH dependent
as described in the literature.24 Acetate buffer (pH 5) was found to
provide the optimal condition resulting in the highest yield possible
for all four monoamines. Since the acetaldehyde is very volatile even
at the room temperature, the pure acetaldehyde-d4 was chilled before
opening and then diluted with 1 M acetate buffer (pH 5) at the
volume ratio of 1:3. The resultant acetaldehyde-d4 concentration (∼5
M) used in this study is ∼100 times higher than that of sodium
cyanoborohydride (47 mM) in cyanoborohydride coupling buffer.
Therefore, only the effect of the reducing reagent concentration was
examined and 25-min reaction time was used (see Figure 1A). We
can see that the labeling efficiency for NE, 5-HT, and NM increased
as the concentration of the reducing reagent increased, especially
in the low-concentration range of cyanoborohydride (below 9.44
mM). The labeling efficiency for NE, 5-HT, and NM is close to the
maximum when the concentration of cyanoborohydride is above 9.44
mM. However, the labeling efficiency for DA to form diethylated
DA with an additional six-member ring increased as the concentration
of cyanoborohydride increased in the low-concentration range (below
2.36 mM), but the labeling efficiency for DA to produce the same
product reduced when continued to increase the concentration of
cyanoborohydride to above 2.36 mM. The explanation for this is that
the rate of reduction of the imine groups formed at the initial step of
the reaction (see Scheme 1) was affected by the concentration of
cyanoborohydride in the reaction mixture. At the low concentration
of cyanoborohydride, the reduction of the imine groups is slow, which
results in the increased chance to form a six-member ring and
62 ([M + H]+ ) 216.2 in Figure 2H) and 64 Da ([M + H]+
)
218.1 in Figure 2H), are almost equally formed. The possible
reasons are as follows: the steric effect of the additional OH
groups being added to the ꢀ-carbon of NM and NE, and the
9198 Analytical Chemistry, Vol. 80, No. 23, December 1, 2008

Scheme 2. Reductive Amination of 5-HT with Acetaldehyde-d4
consequently form diethylated DA with an additional six-member
ring. However, after the concentration of cyanoborohydride reaches
a certain point, the higher the concentration of cyanoborohydride in
the reaction mixture, the higher the rate of reduction of imine groups,
which results in a decreased chance to form diethylated DA with
the addition of a six-member ring. To consider the labeling efficiency
for all four monoamines to produce the precursor ions of interest
and the dilution factor due to the volume of cyanoborohydride
coupling buffer being added, in this study, 12.8 mM was chosen as
the final cyanoborohydride concentration in the reaction mixture to
label amino groups on the monoamine neurotransmitters. Another
key element affecting labeling efficiency is the reaction time, which
was also investigated in this study (see Figure 1B). We can see that
the labeling reaction is almost completed for all four monoamines at
the incubation time of 25 min at 37 °C using the selected reaction
condition (see Experimental Section for details).
Figure 1. Effects of reducing reagent concentration (A) and reaction time (B) on reaction efficiency of diethyl labeling. The reaction was conducted
at 37 °C for 25 min in (A), and 12.8 mM cyanoborohydride concentration was used in (B).
Analytical Chemistry, Vol. 80, No. 23, December 1, 2008 9199

Figure 2. ESI MS spectra of nonderivatized (A) NM, (B) 5-HT, (C) DA, and (D) NE; and d8-diethyl-labeled (E) NM, (F) 5-HT, (G) DA, and (H)
NE.
In summary, diethyl labeling provides a simple means of
labeling amino groups on monoamine neurotransmitters, such as
DA, NE, 5-HT, and NM, via reductive amination. The labeling
chemistry is relatively fast (∼25 min) and specific under mild
reaction conditions with minimal manipulation. And the resultant
diethylated products are very stable, at least within 72 h at a
storage temperature of -4 °C, which was checked (data not
shown).
LC ESI MS and LC ESI MS/MS Analysis of Nonderiva-
tized and Diethyl-Labeled DA, NE, NM, and 5-HT. Figure 2
shows the mass spectra of nonderivatized and d8-diethyl-labeled
DA, NE, NM, and 5-HT at the same volume and molar injection
for each monoamine. By comparing the peak intensities of
molecular ions of each pair of nonderivatized and d8-diethyl-
labeled monoamine, we can see the peak intensities of molecular
ions are increased ∼6-15 times due to diethyl labeling, which
resulted in increased hydrophobicity, concomitantly increased
retention time on a phenyl RP column (see Table 1), and increased
ionization efficiency in ESI MS analysis. A similar phenomenon
has also been reported in the literature for derivatized amino acids
using protonated methyl acetimidate26 and an N-hydroxysuccin-
imide ester of N-alkylnicotinic acid,16 since protonated hydropho-
bic analytes are more likely to migrate to the surface layer of the
Table 1. Retention Time of Nonderivatized and
D8-Diethyl-Labeled DA, NE, 5-HT, and NM on a Phenyl
Column Using the Same Reserved Phase Gradient As
Described in the Experimental Section
retention time (min)
monoamine
nonderivatized
d8-diethyl labeled
DA
1.25
1.05
1.95
1.21
2.17
1.98
2.66
2.38
NE
5-HT
NM
eletrosprayed droplets and become gas-phase ions during the
solvent evaporation. In addition, the diethyl-labeled monoamines
are more stable than their unlabeled counterparts, resulting in
less in-source fragmentation due to the applied declustering
potential (see Figure 2). This, in turn, contributes to the increased
peak intensities for protonated molecular ions of diethyl labeled
monoamines in the ESI MS analysis.
Figure 3 shows the MS/MS spectra of nonderivatized and d8-
diethyl-labeled DA, NE, 5-HT, and NM. We can see that the MS/
MS spectra of d8-diethyl-labeled NE, 5-HT, and NM were more
simple than those that were not derivatized while the simplicity
of MS/MS spectra of nonderivatized and d8-diethyl-labeled DA
remains similar. This may be due to the changed fragmentation
pattern resulting from diethyl labeling of monoamines. Reducing
(26) Shortreed, M. R.; Lamos, S. M.; Frey, B. L.; Phillips, M. F.; Patel, M.;
Belshaw, P. J.; Smith, L. M. Anal. Chem. 2006, 78, 6398–6403
.
9200 Analytical Chemistry, Vol. 80, No. 23, December 1, 2008

Figure 3. ESI MS/MS spectra of nonderivatized (A) NM, (B) 5-HT, (C) DA, and (D) NE; and d8-diethyl-labeled (E) NM, (F) 5-HT, (G) DA, and
(H) NE.
Table 2. API 4000 Mass Spectrometry Conditions for MRM Transitions of Nonderivatized and D8-Diethyl-Labeled
DA, NE, 5-HT, NM, and Their Internal Standards
collision cell
exit potential
(V)
targe
compound
precursor
ion (m/z)
product
ion (m/z)
declustering
potential (V)
collision
energy (eV)
nonderivatized
DA
154.1
170.1
177.1
184.1
187.1
181.1
158.1
216.1
234.1
241.1
248.1
251.1
245.1
220.1
91.1
107
20
20
20
20
20
20
20
35
35
35
35
35
35
35
36
29
36
28
28
36
36
42
22
24
22
22
24
42
15
15
15
15
15
15
15
15
15
15
15
15
15
15
NE
5-HT
NM
d3-NM
d4-5-HT
d4-DA
DA
NE
5-HT
NM
d3-NM
d4-5-HT
d4-DA
115
134.1
137.1
119
91.1
62.1
216.1
94.1
230.1
233.1
96.1
64.1
d8-diethyl labeled
the number of ions in the MS/MS spectra of diethyl-labeled
monoamines could result in the increased peak intensities of the
observed product ions. This could be another explanation for the
increased detection sensitivity for diethyl labeled monoamines in
the LC-MRM analysis (see below).
Analytical Performance of Nonderivatized and d8-Diethyl-
Labeled DA, NE, NM, and 5-HT. The UPLC/MS/MS method
was developed for quantitative measurement of DA, NE, 5-HT,
and NM in the rat prefrontal cortex microdialysates. The detection
and quantification limits of the UPLC/MS/MS-based method for
nonderivatized and d8-diethyl-labeled monoamines were charac-
terized by analyzing a series of standard mixtures, each mixture
containing four monoamine neurotransmitters at different con-
centrations, using the optimized MRM transitions as listed in Table
2. Supporting Information Figure 1 shows the representative
chromatograms of nonderivatized and d8-diethyl-labeled DA, NE,
5-HT, and NM. Table 3 lists the characterized analytical perfor-
mance results of UPLC/MS/MS-based methods for nonderiva-
tized and d8-diethyl-labeled monoamines. We can see that the
sensitivity of the UPLC/MS/MS-based method for quantitative
Analytical Chemistry, Vol. 80, No. 23, December 1, 2008 9201

Table 3. Analytical Performance of UPLC/MS/MS-Based Methods for Nonderivatized and d8-Diethyl-Labeled DA, NE,
5-HT. and NM
nonderivatized
d8-diethyl labeled
limit of
quantification
(ng/mL)
limit of
quantification
(ng/mL)
name of
monoamines
dynamic range
(ng/mL)
fit for
progression
weighting
factor
linearity
DA
0.1
1
0.2
0.4
0.01
0.01-1
0.01-1
0.01-1
0.01-1
0.9997
0.9999
0.9998
0.9999
linear
1/x
NE
0.01
linear
1/x
5-HT
NM
0.005
0.002
quadratic
linear
1/x*x
1/x
measurement of monoamines has been increased ∼20-100 times
due to diethyl labeling, which results in the following: (1)
increased hydrophobicity and, concomitantly, increased ionization
efficiency in ESI MS/MS analysis; (2) changed fragmentation
pattern. This increased sensitivity might also be due to the
reduced matrix effects resulting from LC separation and reduced
background noise in the MRM chromatogram because of changed
transition.
presented in Supporting Information Figure 2. All the solutes were
eluted in 3.5 min. Therefore, the total chromatographic analysis
time for all 10 rat prefrontal cortex mocrodialysate samples was
less than 1 h (3.5 min × 10 ) 35 min). The high sensitivity of this
method, which could result in the requirement of small sample
volume, combined with a short UPLC analysis time, will enable
the consecutive analysis of the rat prefrontal cortex microdialysate
samples in a high temporal resolution if it is needed.
Analysis of Rat Prefrontal Cortex Microdialysates. Diethyl
labeling combined with UPLC/MS/MS was applied to monitor
the drug-induced changes of the monoamine concentrations in
rat prefrontal cortex microdialysates. Figure 4 shows the concen-
tration changes of the monoamines upon the injection of 1 mg/
kg SKF 81297. The dopamine level was gradually increased by
∼3 times at 90 min after SKF 81297 was subcutaneously injected
and gradually relaxed to a value still higher than the basal level
of dopamine at 180 min after drug treatment. The concentration
of NE, 5-HT, and NM changed with the same time course as that
of DA. The similar modulation of DA and NE was observed in
the rat prefrontal cortex microdialysates by using an LC-EC
method after the injection of 1 mg/kg SKF 81297 (data not
shown). In one of the selected rats, average basal levels of DA,
NE, 5-HT, and NM (mean ± SEM) were 0.058 ± 0.011, 0.063 ±
0.008, 0.017 ± 0.001, and 0.013 ± 0.002 ng/mL, respectively, while
high concentrations of the same compounds after subcutaneous
injection of 1 mg/kg SKF 81297 were 0.151, 0.249, 0.075, and 0.048
ng/mL. Concentrations were not corrected for in vitro probe
recovery. Representative chromatograms from the analysis are
CONCLUSIONS
In this work, we presented a high-sensitivity and high-
throughput UPLC/MS/MS method in combination with diethyl
labeling for quantitative measurement of a panel of monoamines,
such as DA, NE, 5-HT, and NM, in rat prefrontal cortex microdi-
alysates. The use of UPLC/MS/MS enabled us to separate and
quantify four monoamines of interest in the rat brain microdialy-
sate in a 3.5-min chromatographic run. To the best of our
knowledge, this has been the most efficient chromatographic
separation method for simultaneous measurement of DA, NE,
5-HT, and NM in the rat brain microdialysate samples. Although
diethyl labeling is a shortcoming as a pretreatment step before
sample injection for this UPLC/MS/MS-based method, it provides
∼20-100 times increased sensitivity of a UPLC/MS/MS-based
approach for quantitative measurement of DA, NE, 5-HT, and NM.
Therefore, it enables the use of a UPLC/MS/MS-based method
for simultaneous measurement of DA, NE, 5-HT, and NM in the
rat prefrontal cortex microdialysates, in which monoamine con-
centrations have been reported to be very low and difficult to
measure using LC/MS/MS-based approach.12 The simple and
rapid labeling reaction is carried out under very mild conditions
with minimum sample and reagent manipulation. The labeling
reagents are commercially available, and the labeling process is
very robust and reliable. The resulting diethylated products are
very stable at least within 72 h at the storage of -4 °C, which can
be set for an autosampler in most of bioanalytical laboratories.
Therefore, the throughput of the sample preparation could be
highly improved by carrying out chemical derivatization in one
or more 96-well plates. In the future, a similar practice could also
be applied to analyze other primary or secondary amine-containing
biochemicals in the biological matrix if high sensitivity is needed.
Comparedwithotherreportedtechniques,suchasHPLC-EC,27-29
HPLC-FL,9 and HPLC-PFET,11 for simultaneous measurement of
(27) Heidbreder, C. A.; Lacroix, L.; Atkins, A. R.; Organ, A. J.; Murray, S.; West,
Figure 4. Concentration changes of the monoamines in the rat
prefrontal cortex microdialysates after injection of SKF-81297 at the
dosage of 1 mg/kg. Injection time of SKF-81297 is marked as time 0.
Results were expressed as percentage increases above baseline
(average of 4 untreated points, 100%). n ) 2 rats.
A.; Shah, A. J. J. Neurosci. Methods 2001, 112, 135–144
(28) Vicente-Torres, M. A.; Gil-Loyzaga, P.; Carricondo, F.; Bartolome, M. V.
J. Neurosci. Methods 2002, 119, 31–36
(29) Zhang, W.; Cao, X.; Xie, Y.; Ai, S.; Jin, L.; Jin, J. J. Chromatogr., B: Anal.
Technol. Biomed. Life Sci. 2003, 785, 327–336
.
.
.
9202 Analytical Chemistry, Vol. 80, No. 23, December 1, 2008

monoamine neurotransmitters in the rat brain microdialysates,
diethyl labeling combined with UPLC/MS/MS provides compa-
rable or even lower concentration detection limits. The high
specificity and high throughput of this UPLC/MS/MS-based
approach will enable its wide application for simultaneous mea-
surement of monoamine neurotransmitters in the pharmaceutical
industries, where monoamine neurotransmitters have been used
as biomarkers in the discovery and development of new drugs to
treat the disorders of central and peripheral nervous systems.
MS hardware and software installation and configuration. We also
thank David Johnson for the discussion on using LC-EC for
quantitative measurement of DA and NE in rat prefrontal cortex
microdialysates and sharing the observed modulation of DA and
NE in rat prefrontal cortex microdialysates after subcutaneous
administration of 1 mg/kg SKF 81297. C.J. and W.L. made equal
experimental contributions.
SUPPORTING INFORMATION AVAILABLE
Additional information as noted in text. This material is
available free of charge via the Internet at http://pubs.acs.org.
ACKNOWLEDGMENT
We thank Drs. Nalini Sadagopan, Gabriella Szekely-Klepser,
and Sarah Grimwood for their support and scientific discussions.
We thank Dr. Qinhua Huang for discussing and reviewing the
proposed chemical process and Dr. Jiun-Tang Huang for UPLC/
Received for review June 30, 2008. Accepted September
4, 2008.
AC801339Z
Analytical Chemistry, Vol. 80, No. 23, December 1, 2008 9203