Anal Bioanal Chem (2017) 409:7191–7199

DOI 10.1007/s00216-017-0681-3

RESEARCH PAPER

Development and validation of a simple, rapid and sensitive

LC-MS/MS method for the measurement
of urinary neurotransmitters and their metabolites
Jingya Yan 1 & Unnikrishnan Kuzhiumparambil 2 & Sushil Bandodkar 3 & Nadia Solowij 4 &
Shanlin Fu 1

Received: 26 July 2017 / Revised: 13 September 2017 / Accepted: 27 September 2017 / Published online: 13 October 2017
# Springer-Verlag GmbH Germany 2017

Abstract Neurotransmitters play crucial roles in physiologi- measuring the neurotransmitter concentrations in urine of
cal functions and their imbalances have demonstrated associa- healthy donors. Furthermore, the undertaken stability experi-
tion in the pathology of several diseases. The measurement of ments indicated that acidified urine specimens allowed the
neurotransmitters possesses a great potential as a significant analytes to be stable for prolonged durations in comparison
clinical tool. This study presents the development and valida- to those untreated. The study also reveals the performance of
tion of an LC-MS/MS method for simultaneous quantification the method is unaffected by the absence of expensive deuter-
of multi-class neurotransmitters associated with dopamine, ated reference standards under the experimental conditions
tryptophan and glutamate-γ-aminobutyric acid pathways. A employed which further simplifies the analytical procedures
total of ten neurotransmitters and their metabolites (dopamine, and provides a significant cost saving for running the assay.
epinephrine, metanephrine, tryptophan, serotonin, kynurenic
acid, kynurenine, anthranilic acid, GABA, glutamic acid) were
determined based on a simple and rapid ‘dilute and shoot’
method using minimal urine volume. The chromatographic Keywords Neurotransmitter . Liquid chromatography . Mass
separation was achieved using a Poroshell 120 Bonus-RP LC spectrometry . Urine
Column in combination with a gradient elution within an 8.5-
min time frame. The method exhibited good sensitivity as the
limits of quantification ranged between 0.025 and 0.075 μg/ Introduction
mL with acceptable matrix effects (< ± 14.5%), no carryover
and good linearity (r2 > 0.98). The accuracy and precision for Neurotransmitters are endogenous substances which act as
all analytes were within tolerances, at < ± 9.9% mean relative primary chemical messengers, released from neurons to target
error (MRE) and < 8.6% relative standard deviation (RSD), cells by relaying, amplifying and modulating signals. The ma-
respectively. The method was successfully applied in jor neurotransmitters that function at central and peripheral
levels are classified into two categories: monoamine and ami-
no acid neurotransmitters [1]. Monoamine neurotransmitters
* Shanlin Fu
[email protected]
include dopamine, serotonin, epinephrine and norepinephrine
[2]. Amino acid neurotransmitters consist of glutamic acid,
1
Centre for Forensic Science, University of Technology Sydney,
tryptophan, γ-aminobutyric acid (GABA), alanine, aspartic
PO Box 123, Broadway, Ultimo, NSW 2007, Australia acid and taurine [3].
2
Climate Change Cluster, University of Technology Sydney,
Neurotransmitters play essential roles in regulating the
PO Box 123, Broadway, Ultimo, NSW 2007, Australia body’s nervous, immune and cardiovascular system [4].
3
Department of Clinical Biochemistry, The Children’s Hospital at
These neuroactive metabolites have been found to be impli-
Westmead, Locked Bag 4001, Westmead, NSW 2145, Australia cated with a wide range of diseases including cancer tumours,
4
School of Psychology and Illawarra Health and Medical Research
neurodegenerative, psychiatric and psychological disorders
Institute, University of Wollongong, Wollongong, NSW 2522, [5]. Recent research shows that the imbalance of tryptophan
Australia metabolism is closely linked with the development of lung
7192 J. Yan et al.

and breast cancer [6]. Studies which examined urinary neuro- extraction technique as it offers high selectivity, precision and
transmitter concentrations in psychological disorders indicat- extraction yields of above 90% [21]. Liquid-phase micro-ex-
ed that depressive symptoms were in correlation with the in- traction (LPME) is a relatively newly developed extraction pro-
crease of catecholamines in the body [7–9]. The imbalances of cess described to have low solvent consumption [22]. As the
glutamate and GABA metabolism have been closely associ- extraction of hydrophilic compounds in the organic phase is
ated with the occurrence of seizures such as epilepsy [10]. The more difficult, most studies focus on extracting hydrophobic
crucial involvement of neurotransmitters in neurological, en- compounds [23]. To overcome the challenges involved in
docrinological and immunological functions has directed LPME, Jiang et al. [22] developed a novel method to solve
great attention to researchers and clinicians towards these mol- the collection of acceptor solutions through the solidification
ecules and metabolites as diagnostic indicators for monitoring of floating organic drop. Another established sample prepara-
disease states and therapeutic interventions [11, 12]. Hence, tion method for biological samples is sample dilution. The ad-
there are high demands for the development of validated ana- vantages of this technique are the removal of extensive clean-up
lytical methods for the quantification of neurotransmitter con- procedures, ability to prevent sample deterioration, time effi-
centrations in biological matrices. ciency and cost-effectiveness [3, 24, 25]. Sample dilution is
Neurotransmitters have been measured in several biologi- mainly used in the sample preparation of CSF or dialysates
cal matrixes, including urine, cerebral spinal fluid (CSF), plas- [26]. Although this technique has not been widely adapted, it
ma, saliva, serum and platelets [13]. The CSF matrix provides possesses potential as an alternative to existing sample prepa-
the best representation of CNS neurotransmitter levels. ration methods.
However, measuring neuroactive compounds in the CSF re- This study aimed to develop and validate an analytical LC-
flects major drawbacks in the sample collection and prepara- MS/MS method that can simultaneously detect and quantify
tion stages causing interpretation difficulties [14]. In literature, several neurotransmitters and their metabolites with minimal
blood and urine are the most common matrices employed in sample preparation steps and reduced running costs. There has
neurotransmitter measurements [15]. Urine is described as been minimal literature on multiclass neurotransmitter analy-
non-invasive in nature, enabling inexpensive methods, being sis and quantification, particularly for metabolites of the tryp-
largely available and simple compared to other matrices such tophan pathway. The uniqueness of this analytical method is
as plasma and blood [16]. For this study, urine was selected demonstrated in the optimization of an original sample prep-
based on the advantages and simplicity of this matrix. It is aration procedure which involves sample dilution avoiding
known that an analyte concentration in urine is susceptible traditional extensive extraction clean-up and derivatisation
to variation due to hydration status; this variation, however, procedures and the removal of the use of expensive deuterated
can be corrected by normalising the analyte concentration to internal standard (IS) reference materials. It is expected that
the creatinine level in urine if necessary [17]. the innovations of this methodology have not only great re-
Traditionally, quantitative analysis of neurotransmitters in search potential but also enable the practicality of neurotrans-
biological samples is performed by radioenzymatic or immu- mitter quantifications within a clinical setting.
nological assays, fluorimetry and gas chromatography [18].
Developments and research throughout the past two decades
have resulted in these procedures to be superseded by greatly
selective and sensitive chromatographic methods coupled to Materials and method
mass spectrometry detection systems [19]. GC-MS is
characterised as a potent technique in the identification of neu- Chemicals and reagents
roactive compounds whereby analytes are derivatised to attain
volatility and stability. The rise of simple and economical LC Anthranilic acid, dopamine hydrochloride, epinephrine hydro-
techniques coupled with tandem mass spectrometry significant- chloride, L -glutamic acid, GABA, kynurenic acid, L -
ly decreases run times, improves detection limits, analytical kynurenine, metanephrine, serotonin hydrochloride and L-
sensitivity and reproducibility of biological preparations [20]. tryptophan were purchased as powders from Sigma Aldrich
One of the most challenging aspects to account for in the (Sydney, Australia). D3-Dopamine, D5-glutamic acid, D6-
quantitative measurement of neurotransmitters is sample prep- GABA, D5-kynurenic acid and D3-metanephrine were sup-
aration. The determination of neurotransmitters in biological plied by CDN Isotopes (Quebec, Canada). HPLC grade ace-
samples is challenging due to the high possibility of interfer- tonitrile was purchased from Honeywell (Sydney, Australia).
ences, chemical instability issues and low concentrations of Formic acid and acetic acid were supplied by RCI Labscan
analytes present in biological matrices [3]. Generally, common (Bangkok, Thailand). Three millilitre syringes were purchased
sample preparation requires lengthy separation and extraction from Beckton Dickinson (Sydney, Australia) and 0.22 μm
procedures prior to instrumental analysis. Currently, solid- hydrophilic syringe filters were supplied by Micro-Analytix
phase extraction (SPE) is the most frequently employed Pty Ltd. (Taren Point, NSW, Australia).
Development and validation of a simple, rapid and sensitive LC-MS/MS method for the measurement of urinary... 7193

Urine collections 5 μg/mL mixed IS solution in 0.2 M acetic acid was prepared
from 100 μg/mL individual stock solutions of the five internal
Surine, commercial synthetic urine devoid of endogenous standards (D3-dopamine, D5-glutamic acid, D6-GABA, D5-
neurotransmitters, was purchased from Sigma Aldrich kynurenic acid, D3-metanephrine).
(Sydney, Australia). Authentic urine specimens were obtained
from ten healthy donors of Asian and Caucasian origin includ- Sample preparation
ing six male and four female participants in the age group of
20–40 years. Participants were each provided with a sterile Six calibration points were prepared by spiking, into 2 mL
urine specimen container during mid-day for urine collection. micro-centrifuge tubes, 1 mL of synthetic blank urine with
The collected urine samples were immediately frozen in a the mixed standard solution at concentrations of 0.50, 1.25,
−20 °C freezer and analysed within a week. 2.50, 3.75, 6.00 and 7.50 μg/mL, respectively. Three quality
control (QC) samples were similarly prepared at 0.20 (low
Liquid chromatography QC), 2.75 (medium QC) and 5.00 μg/mL (high QC). These
samples were made into 2 mL by diluting with acetonitrile,
The chromatographic separation of analytes was achieved on vortexed for 60 s, precipitated in ice for 10 min followed by
an Agilent Po roshell 1 20 Bonus-RP LC column centrifugation for 10 min at a temperature of 2 °C and velocity
(2.1 mm × 100 mm 2.7 μm particle size) using the Agilent of 5600 g. Subsequently, 1 mL of the mixture was pipetted
1290 infinity liquid chromatography system coupled to the from the micro-centrifuge tube and diluted with 0.25 mL of
Agilent 6490 triple quadrupole mass spectrometer. A gradient the mixed IS solution and 1.25 mL of 0.2 M acetic acid. The
elution was performed at a flow rate of 0.18 mL/min with two solution was filtered through a 0.22 μm hydrophilic syringe
mobile phases. Mobile phase A was composed of water con- filter. The filtrate was subjected to LC-MS/MS analysis.
taining 0.2% (v/v) formic acid and mobile phase B comprised Authentic urine was similarly prepared resulting in an overall
of acetonitrile with 0.1% (v/v) formic acid. The gradient elution five-fold dilution into 0.2 M acetic acid for LC-MS/MS
are as follows: 100% A (0 min) ➔ 100% A (hold 2 min) ➔ analysis.
95% B (linear increase in 0.5 min) ➔ 95% B (hold 2.5 min) ➔ For method validation with the use of the internal standards,
100% A (linear decrease in 0.25 min) ➔ 100% A (hold 3 min). peak area ratios of analyte to its deuterated IS counterpart were
used. For the analytes without their corresponding deuterated
Mass spectrometry standards, D5-kynurenic acid was used to quantify anthranilic
acid, tryptophan and kynurenine. D3-Dopamine and D3-
The Agilent 6490 triple quadrupole (QQQ) LC/MS detector metanephrine were used for epinephrine and serotonin, respec-
was operated in the multiple reaction monitoring (MRM) mode tively. For method validation without the use of the internal
using positive electrospray ionisation (ESI+) mode. The source standards, the absolute peak areas of analytes were used.
conditions were as follows: gas temperature was 290 °C, flow
rate of 660 L/h, fragmentor voltage 120 Vand capillary voltage Method validation
of 3500 V. Initially, the MS was acquired in the targeted MS/
MS full scan data mode. The precursor ion was selected in the Method validation was performed in accordance with pub-
first quadrupole and subsequently fragmented in the high- lished procedures [27]. Linearity was assessed using the linear
energy collisional dissociation cell. Following fragmentation, coefficient of determination statistical measure known as the
the product ions were obtained in subsequent quadrupoles. R-squared value. A linear association is established when the
When the precursor ions and the collision energy of their most detected signal and concentration of an analyte are proportion-
abundant product ions were identified for each analyte, the al (r2 > 0.98). The limit of detection (LOD) and limit of quan-
QQQ parameters were set to MRM mode. A total of two tification (LOQ) were determined by spiking decreasing con-
MRM transitions were used for each analyte. centrations of the mixed stock solution into blank synthetic
urine. The LOD is defined as the lowest concentration point at
Preparation of standard solutions which the instrument exhibits signal-to-noise (S/N) ratio equal
to 3. LOQ concentration of an analyte is defined as the lowest
Standard stock solutions (500 μg/mL) of all ten analytes were concentration reliably quantified and fulfils the criteria of not
individually prepared in 0.2 M acetic acid. A 25 μg/mL mixed exceeding ± 20% mean relative error (MRE) and < 20% rel-
standard solution containing the analytes was generated by ative standard deviation (RSD).
obtaining aliquots of each individual stock solution. The so- Accuracy and precision were evaluated simultaneously
lutions were stored in −20 °C freezer prior to usage. The using different concentration levels (QC samples) with five
mixed standard solution was diluted appropriately with replicates on five different days. Accuracy was measured
0.2 M acetic acid to prepare the working solution series. A using the %MRE. Precision was measured by the %RSD.
7194 J. Yan et al.

For the method to be deemed accurate and precise, the %MRE autosampler at a maintained constant temperature of 4 °C for
and %RSD values must fall within the maximum acceptable 48 h. The freeze-thaw stability was conducted by assessing
range of ± 20 and < 20 respectively. authentic urine specimens over three cycles. The urine speci-
Matrix effects were examined through the assessment of mens were frozen at −20 °C and thawed at room temperature.
analyte enhancement or suppression. As a ‘dilute and shoot’ Long-term stability was performed by testing untreated and
approach was used and no extractions were involved, standard acidified authentic urine stored at −20 °C over 90 days.
addition method was considered appropriate [28] and was Stability for all experiments was measured by re-analysing
applied, whereby authentic human urine specimens were the specimens stored in their designated conditions. The ab-
spiked at 2.75 and 5.00 μg/mL of analytes. Primarily, both solute peak areas must not exhibit a difference greater than
the spiked and the original samples of individual authentic 20% to the initial peak area to be regarded as stable.
urine sets were prepared using the dilution procedure de- For the assessment of carryover, a high concentration,
scribed in ‘Sample preparation’ section and analysed using 10 μg/mL mixed standard solution and 0.2 M acetic acid were
the method without IS. The differences in concentration be- prepared. The 0.2 M acetic acid was injected immediately
tween the spiked and the original sample represent the mea- after a high concentration and this process was performed in
sured concentration of the spiked analytes. The following replicates. A method is free from carryover when the acetic
equation was used for the determination of the percentage acid sample does not give a peak area higher than the LOD
matrix effect (%ME) where target concentration was the and a signal to noise (S/N) ratio larger than 3.
spiked concentration:
 
Measured concentration−target concentration
%ME ¼  100 Results and discussion
Target concentration

In relation to %ME, a value of 0 indicates there are no Method optimisation

matrix effects involved, > 0 means that ion enhancement is
present and < 0 implies ion suppression. The acceptable range The qualitative and quantitative analysis of neurotransmitters
of ion suppression or enhancement is within ± 15%ME. were achieved through determination of retention time, pre-
Stability experiments were conducted to assess the stability cursor ion and a minimum of two product fragment ions of
of the ten analytes under different experimental conditions. each compound with acceptable ion ratios within ± 30% of
The comparison of absolute peak areas of freshly prepared those for the reference standards. The MRM transition data
working solutions with samples stored under various condi- acquired in the positive ion mode are recorded in Table 1.
tions was of interest. Post-preparative stability was performed Individual 1 μg/mL standards were analysed to obtain the
by re-analysing the QC samples after being left in the retention times of each analyte. The analytes all eluted within

Table 1 MRM transitions of ten

analytes and five deuterated Analyte Precursor 1st product ion 2nd product ion Ratio of 2nd/1st
internal standards including ion [M + H]+ product ions
precursor and product ions. The
first product ion transitions were m/z m/z Collision m/z Collision
used as the quantifying ions energy (eV) energy (eV)
Tryptophan 205 188 5 118 29 0.35
Kynurenine 209 192 5 94 17 0.39
Kynurenic acid 190 144 25 116 37 0.44
Anthranilic acid 138 120 9 92 21 0.26
Glutamic acid 148 130 5 84 17 0.27
GABA 104 87 9 69 13 0.31
Dopamine 154 137 25 119 9 0.36
Epinephrine 184 166 9 107 21 0.25
Metanephrine 198 180 9 148 21 0.22
Serotonin 177 160 9 115 33 0.29
D5-Kynurenic acid 195 149 17 121 37 0.35
D5-Glutamic acid 153 135 9 88 13 0.39
D6-GABA 110 93 9 92 9 0.29
D3-Dopamine 157 139 9 93 25 0.41
D3-Metanephrine 201 183 9 151 13 0.28
Development and validation of a simple, rapid and sensitive LC-MS/MS method for the measurement of urinary... 7195

Fig. 2 Representative extracted ion chromatograms of the LOQ and

Fig. 1 Representative extracted ion chromatograms of the LOQ and authentic urine for the amino acid analytes (from top to bottom): (1)
authentic urine for the amine analytes (from top to bottom): (1) synthetic synthetic urine blank, (2) anthranilic acid in surine, (3) anthranilic acid
urine blank, (2) dopamine in surine, (3) dopamine in human urine, (4) in human urine, (4) glutamic acid in surine, (5) glutamic acid in human
kynurenine in surine, (5) kynurenine in human urine, (6) epinephrine in urine, (6) tryptophan in surine, (7) tryptophan in human urine, (8) GABA
surine, (7) epinephrine in human urine, (8) serotonin in surine, (9) sero- in surine, (9) GABA in human urine, (10) Kynurenic acid in surine and
tonin in human urine, (10) metanephrine in surine and (11) metanephrine (11) kynurenic acid in human urine
in human urine
7196 J. Yan et al.

Table 2 Linearity, LOD and

LOQ of the validated method Analyte Correlation coefficient Correlation coefficient LOD (μg/mL) LOQ (μg/mL)
(r2) with IS (r2) without IS

Tryptophan 0.998 0.997 0.005 0.025

Kynurenine 0.990 0.986 0.005 0.025
Kynurenic acid 0.997 0.985 0.025 0.075
Anthranilic acid 0.994 0.989 0.025 0.050
Glutamic acid 0.999 0.998 0.050 0.075
GABA 0.996 0.988 0.050 0.075
Dopamine 0.993 0.981 0.050 0.075
Epinephrine 0.995 0.992 0.050 0.075
Metanephrine 0.998 0.997 0.050 0.075
Serotonin 0.997 0.995 0.050 0.075

an 8.5-min window with glutamic acid as the first to elute and working range generated correlation coefficients (r2) higher
kynurenic acid eluting last. The chromatograms of the ten than 0.98 using linear regression (Table 2). The linearity re-
analytes at their LOQ concentrations are illustrated in Figs. 1 sults obtained from both approaches, i.e. with and without the
and 2. use of internal standards, were comparable.
The simple dilution combined with filtration sample prep- The LOD and LOQ of the analytes were in the range of
aration method possesses advantages in the LC-MS/MS anal- 0.005–0.050 and 0.025–0.075 μg/mL respectively (Table 2).
ysis of neurotransmitters and their metabolites. The simplicity This indicated a highly sensitive method was developed for
of the procedure does not require prolonged preparation dura- the simultaneous quantification of ten neurotransmitters and
tions. The nature of ‘dilute and shoot’ preparation procedures their metabolites.
allows minimal volumes of urine to be used due to the absence The results for accuracy and precision for the QC samples
of extraction processes involved. Although a sample volume with and without IS are compared in Tables 3 and 4 respec-
of 1 mL was used in this study for the convenience, it is tively. There are no noticeable differences between the two
expected that the urine volume could go much lower in theory approaches; both approaches afforded accurate and precise
if other reagents and solutions are proportionally reduced at analytical methods. For the method without IS, the accuracy
the same time during sample preparation. These are extremely ranged from −9.9% to +8.1% MRE, and the precision ranged
beneficial for fast-paced laboratory settings with high de- from 1.7% to 8.3% RSD. These values had met the acceptable
mands in the analysis of samples. method validation criteria, indicating that the method perfor-
mance was not compromised by the omission of the expensive
Method validation internal standards.
The validation of linearity, accuracy and precision showed
The linearity of the method was from the LOQ to 7.50 μg/mL that the method maintained high analytical performance in the
for each analyte. The six selected calibration points of the absence of internal standards; this is extremely advantageous

Table 3 Accuracy results at

different QC concentration levels Analyte Accuracy (%MRE) Accuracy (%MRE)
with and without internal With IS Without IS
standards (n = 5)
0.20 μg/mL 2.75 μg/mL 5.00 μg/mL 0.20 μg/mL 2.75 μg/mL 5.00 μg/mL
Tryptophan −3.7 3.9 −5.1 4.9 5.8 6.3
Kynurenine 4.6 2.8 3.3 −5.3 −3.4 4.3
Kynurenic acid 5.5 −3.1 −1.7 −7.2 4.7 −2.4
Anthranilic acid −4.1 −5.4 2.2 3.7 7.3 1.8
Glutamic acid 5.7 7.5 −2.8 6.6 −8.2 −3.3
GABA −6.4 6.3 −4.9 8.1 −7.1 −6.1
Dopamine −6.8 −5.7 −3.0 −7.8 −7.6 5.4
Epinephrine 3.9 −3.6 5.9 4.3 3.4 6.8
Metanephrine 7.0 3.9 7.5 −8.5 −5.3 −9.9
Serotonin −5.2 −3.0 4.2 5.7 4.0 −4.9
Development and validation of a simple, rapid and sensitive LC-MS/MS method for the measurement of urinary... 7197

Table 4 Precision results at

different QC concentration levels Analyte Inter-day precision (%RSD) Inter-day precision (%RSD)
with and without internal With IS Without IS
standards (n = 5)
0.20 μg/mL 2.75 μg/mL 5.00 μg/mL 0.20 μg/mL 2.75 μg/mL 5.00 μg/mL
Tryptophan 5.8 6.0 2.4 7.3 3.9 2.5
Kynurenine 5.1 3.7 6.3 4.8 2.2 4.0
Kynurenic acid 3.4 4.2 3.8 4.6 3.4 4.9
Anthranilic acid 6.2 4.9 5.9 6.9 2.1 5.2
Glutamic acid 7.0 3.6 6.1 8.3 1.7 3.0
GABA 4.7 5.6 4.4 5.4 5.2 3.7
Dopamine 3.3 3.5 5.2 6.7 2.6 6.8
Epinephrine 3.5 5.7 7.8 3.2 5.0 4.3
Metanephrine 3.9 4.3 5.4 3.5 6.2 2.7
Serotonin 4.6 2.7 8.6 5.1 2.8 5.4

as it significantly reduces the running cost in the operation of metabolites was minimised through the addition of acetic acid
this method. and storage of samples on ice during the preparation process.
As the developed sample preparation procedure does not The ten analytes in working solutions were found to main-
involve an extraction process, ion suppression or enhance- tain stability for 48 h in the autosampler at a maintained con-
ment were assessed through the standard addition method. stant temperature of 4 °C. As laboratories or clinical work-
The results for the matrix effects investigated at 2.75 and places may require the use of stock solutions over an extended
5.00 μg/mL of the method are outlined in Table 5. All analytes period, it is useful to understand the duration and conditions
were found to be within tolerances (± 15% ME); therefore, the under which each analyte is able to maintain stability when
method is not affected by ionisation enhancement or suppres- stored for a lengthy period. For this reason, freeze-thaw and
sion due to matrix influence. Additionally, this indicates the long-term stability experiments were subjected to extensive
selectivity of the method as target analytes were not subject to analysis. A summary of the results for the assessed analytes
interference of other endogenous substances in human urine in these two studies are outlined in Table 6. The metabolites
matrix. were found to maintain stability for at least three freeze-thaw
The chemical instability of neurotransmitters and their me- cycles when stored at −20 °C. Long-term storage of urine
tabolites is one of the most challenging fields in quantitative showed analytes were stable for at least 90 days in treated
analysis. The main causes of sample degradation due to chem- authentic urine with acetic acid. On the other hand, the major-
ical instabilities are daylight, temperature and dissolved oxy- ity of the analytes in untreated authentic urine were found to
gen. In order to prevent degradation of neuroactive metabo- remain stable for up to 30 days upon excretion and immediate
lites, stabilising precautions are required to be conducted dur- storage in a −20 °C freezer. Lu et al. [30], in a study of
ing storage, sampling and analysis [29]. The degradation of
Table 6 Freeze-thaw and long-term stability of acidified authentic
Table 5 Matrix effect results at 2.75 and 5.00 μg/mL concentration urine (n = 5)
levels (n = 10)
Analyte Freeze-thaw stability Long-term stability
Analyte %ME spiked at %ME spiked at (% difference) after (% difference) after
2.75 μg/mL 5.00 μg/mL 3 cycles 90 days

Tryptophan 5.6 8.3 Tryptophan 7.4 8.7

Kynurenine −13.1 −9.2 Kynurenine −10.5 13.3
Kynurenic acid 11.5 −9.9 Kynurenic acid −8.1 −14.9
Anthranilic acid −10.2 7.8 Anthranilic acid 19.2 −18.6
Glutamic acid −7.8 −5.2 Glutamic acid −13.8 12.1
γ-Aminobutyric acid −9.1 −8.5 GABA 12.7 10.3
Dopamine 4.5 −7.7 Dopamine −9.5 13.7
Epinephrine −8.4 8.9 Epinephrine 18.1 −19.8
Metanephrine −12.2 −6.5 Metanephrine 16.3 11.9
Serotonin −14.5 −10.4 Serotonin −11.6 −12.6
7198 J. Yan et al.

Table 7 Average concentrations of all analytes in freshly prepared determined in an 8.5-min window. The method was validated
authentic urine (n = 10)
in accordance with standard laboratory protocols and guide-
Analyte Average concentration ± SD (μg/mL) lines. This included the assessment of an analytical criterion
investigating the linearity range, LOD, LOQ, accuracy, preci-
Tryptophan 6.310 ± 0.390 sion, matrix effect, carryover and stability. The method was
Kynurenine 0.790 ± 0.048 further validated by performing analysis of neurotransmitters
Kynurenic acid 0.887 ± 0.042 in authentic human urine. Another notable finding is that the
Anthranilic acid 1.580 ± 0.051 method performance was not compromised in the absence of
Glutamic acid 3.050 ± 0.150 deuterated internal standards when both approaches (with and
GABA 0.543 ± 0.036 without internal standards) were compared under the experi-
Dopamine 0.468 ± 0.028 mental conditions employed. The simple sample preparation
Epinephrine 0.095 ± 0.008 procedures together with a short chromatographic run time
Metanephrine 0.181 ± 0.012 offer a significant advantage over many other published
Serotonin 0.210 ± 0.014 methods when lengthy sample preparation and derivatisation
steps are involved for a busy clinical laboratory to handle
volume samples and to shorten turn-around time. The omis-
tryptophan pathway metabolites, reported that freshly prepared sion of expensive deuterated internal standards in the method
authentic urine with 0.1% ascorbic acid was stable for three will provide a significant cost-saving for any laboratory in-
freeze-thaw cycles and at least 30 days in a −20 °C freezer. volved in performing the analysis and will be particularly
The stability findings obtained in this study were found to beneficial for those laboratories under financial constraint. It
be consistent with those published in the literature. In addition, is important to note that a laboratory should conduct careful
it has also been reported that amine neuroactive molecules are matrix effect assessment using its specific specimen sets es-
stable for a year in acidified urine and storage at −20 °C [31]. pecially when IS are not used. The developed method presents
Hence, it is highly recommended for urine to be treated with a useful addition to the analytical tools available for investi-
acidification upon collection in order to maintain the stability gating the links between neurotransmitter imbalance and a
of neuroactive metabolites for prolonged durations. range of health conditions. The study further emphasises the
The gradient elution profile used was found to be efficient importance of storing samples under acidic conditions due to
in flushing out any remaining analytes between runs as blank instability of these neurotransmitter analytes under normal
responses were zero or significantly less than LOD peak areas, physiological pH and provides basis to guide researchers in
supporting the method being free of carryover. experimental design and particularly on how samples ought to
be collected, stored and analysed.
Application of the validated method in the determination
of neurotransmitters and their metabolites in human urine Compliance with ethical standards Ethics approval for collecting hu-
man urine samples was obtained from the UTS Human Research Ethics
The physiological concentrations of the neurotransmitters Committee (Ethics Approval No. UTS HREC 2010268A). Informed con-
assessed in the method are outlined in Table 7 with represen- sent was obtained from all donors who provided urine samples.
tative chromatograms presented in Figs. 1 and 2. For all
Conflict of interest The authors declare that they have no conflict of
analytes, these values were found to be within the reported interest.
reference ranges of various studies [32, 33]. This demonstrates
that the method is able to produce reliable quantitative data
when compared with values that have been reported using References
different methods.
1. Cai HL, Zhu RH, Li HD, Zhang J, Li LF. MultiSimplex optimiza-
tion of chromatographic separation and dansyl derivatization con-
Conclusion ditions in the ultra performance liquid chromatography-tandem
mass spectrometry analysis of risperidone, 9-hydroxyrisperidone,
monoamine and amino acid neurotransmitters in human urine. J
An analytical method using liquid chromatography triple Chromatogr B. 2011;879(21):1993–9. https://doi.org/10.1016/j.
quadrupole mass spectrometry was developed and validated jchromb.2011.05.006.
for the detection and quantification of ten neurotransmitters 2. Greco S, Danysz W, Zivkovic A, Gross R, Stark H. Microdialysate
associated with three neurotransmitter pathways in human analysis of monoamine neurotransmitters—a versatile and sensitive
LC–MS/MS method. Anal Chim Acta. 2013;771:65–72. https://
urine. Sample preparation of neurotransmitters and their me- doi.org/10.1016/j.aca.2013.02.004.
tabolites was achieved by a simple dilution method in an acid- 3. Bicker J, Fortuna A, Alves G, Falcao A. Liquid chromatographic
ic medium followed by filtration. The analytes of interest were methods for the quantification of catecholamines and their
Development and validation of a simple, rapid and sensitive LC-MS/MS method for the measurement of urinary... 7199

metabolites in several biological samples—a review. Anal Chim 20. Li XS, Li S, Wynveen P, Mork K, Kellermann G. Development and
Acta. 2013;768:12–34. https://doi.org/10.1016/j.aca.2012.12.030. validation of a specific and sensitive LC-MS/MS method for quan-
4. Sakaguchi Y, Yoshida H, Hayama T, Itoyama M, Todoroki K, tification of urinary catecholamines and application in biological
Yamaguchi M, et al. Selective liquid-chromatographic determina- variation studies. Anal Bioanal Chem. 2014;406(28):7287–97.
tion of native fluorescent biogenic amines in human urine based on https://doi.org/10.1007/s00216-014-8120-1.
fluorous derivatization. J Chromatogr A. 2011;1218(33):5581–6. 21. Cakal C, Ferrance JP, Landers JP, Caglar P. Microchip extraction of
https://doi.org/10.1016/j.chroma.2011.05.076. catecholamines using a boronic acid functional affinity monolith.
5. Fuertig R, Ceci A, Camus SM, Bezard E, Luippold AH, Hengerer Anal Chim Acta. 2011;690(1):94–100. https://doi.org/10.1016/j.
B. LC-MS/MS-based quantification of kynurenine metabolites, aca.2011.02.009.
tryptophan, monoamines and neopterin in plasma, cerebrospinal 22. Jiang L, Chen Y, Chen Y, Ma M, Tan Y, Tang H, et al.
fluid and brain. Bioanalysis. 2016;8(18):1903–17. https://doi.org/ Determination of monoamine neurotransmitters in human urine
10.4155/bio-2016-0111. by carrier-mediated liquid-phase microextraction based on solidifi-
6. Chuang SC, Fanidi A, Ueland PM, Relton C, Midttun O, Vollset SE, cation of stripping phase. Talanta. 2015;144:356–62. https://doi.
et al. Circulating biomarkers of tryptophan and the kynurenine path- org/10.1016/j.talanta.2015.06.068.
way and lung cancer risk. Cancer Epidemiol Biomark Prev. 23. Gu Q, Shi X, Yin P, Gao P, Lu X, Xu G. Analysis of catecholamines
2014;23(3):461–8. https://doi.org/10.1158/1055-9965.epi-13-0770. and their metabolites in adrenal gland by liquid chromatography
7. Mooney JJ, Samson JA, Hennen J, Pappalardo K, McHale N, tandem mass spectrometry. Anal Chim Acta. 2008;609(2):192–
Alpert J, et al. Enhanced norepinephrine output during long-term 200. https://doi.org/10.1016/j.aca.2008.01.017.
desipramine treatment: a possible role for the extraneuronal mono- 24. Willemsen JJ, Ross HA, Lenders JWM, Sweep CGJ. Stability of
amine transporter (SLC22A3). J Psychiatr Res. 2008;42(8):605–11. urinary fractionated metanephrines and catecholamines during collec-
https://doi.org/10.1016/j.jpsychires.2007.07.009. tion, shipment, and storage of samples. Clin Chem. 2007;53:268–72.
8. Moret C, Briley M. The importance of norepinephrine in depres- 25. Roberts NB, Higgins G, Sargazi M. A study on the stability of
sion. Neuropsychiatr Dis Treat. 2011;7(Suppl 1):9–13. https://doi. urinary free catecholamines and free methyl-derivatives at different
org/10.2147/NDT.S19619. pH, temperature and time of storage. Clin Chem Lab Med.
9. Hughes JW, Watkins L, Blumenthal JA, Kuhn C, Sherwood A. 2010;48(1):81–7. https://doi.org/10.1515/cclm.2010.017.
Depression and anxiety symptoms are related to increased 24-
26. Hubbard KE, Wells A, Owens TS, Tagen M, Fraga CH, Stewart CF.
hour urinary norepinephrine excretion among healthy middle-
Determination of dopamine, serotonin, and their metabolites in pe-
aged women. J Psychosom Res. 2004;57(4):353–8. https://doi.
diatric cerebrospinal fluid by isocratic high performance liquid chro-
org/10.1016/j.jpsychores.2004.02.016.
matography coupled with electrochemical detection. Biomed
10. Albrecht J, Zielinska M. Mechanisms of excessive extracellular glu-
Chromatogr. 2010;24(6):626–31. https://doi.org/10.1002/bmc.1338.
tamate accumulation in temporal lobe epilepsy. Neurochem Res.
27. Toxicology SWGfF. Scientific Working Group for Forensic
2017;42(6):1724–34. https://doi.org/10.1007/s11064-016-2105-8.
Toxicology (SWGTOX) standard practices for method validation
11. Sokolowska I, Ngounou Wetie AG, Wormwood K, Thome J, Darie
in forensic toxicology. J Anal Toxicol. 2013;37(7):452–74. https://
CC, Woods AG. The potential of biomarkers in psychiatry: focus on
doi.org/10.1093/jat/bkt054.
proteomics. J Neural Transm. 2015;122(1):9–18. https://doi.org/10.
1007/s00702-013-1134-6. 28. Malachová A, Sulyok M, Beltrán E, Berthiller F, Krska R.
12. Mamas M, Dunn WB, Neyses L, Goodacre R. The role of metab- Optimization and validation of a quantitative liquid chromatogra-
olites and metabolomics in clinically applicable biomarkers of dis- phy–tandem mass spectrometric method covering 295 bacterial and
ease. Arch Toxico. 2011;85(1):5–17. https://doi.org/10.1007/ fungal metabolites including all regulated mycotoxins in four mod-
s00204-010-0609-6. el food matrices. J Chromatogr A. 2014;1362:145–56. https://doi.
13. Marc DT, Ailts JW, Campeau DC, Bull MJ, Olson KL. org/10.1016/j.chroma.2014.08.037.
Neurotransmitters excreted in the urine as biomarkers of nervous 29. van de Merbel NC, Hendriks G, Imbos R, Tuunainen J, Rouru J,
system activity: validity and clinical applicability. Neurosci Nikkanen H. Quantitative determination of free and total dopamine in
Biobehav Rev. 2011;35(3):635–44. https://doi.org/10.1016/j. human plasma by LC–MS/MS: the importance of sample preparation.
neubiorev.2010.07.007. Bioanalysis. 2011;3(17):1949–61. https://doi.org/10.4155/bio.11.170.
14. Rodan LH, Gibson KM, Pearl PL. Clinical use of CSF neurotrans- 30. Lu H, Yu J, Wang J, Wu L, Xiao H, Gao R. Simultaneous quanti-
mitters. Pediatr Neurol. 2015;53(4):277–86. https://doi.org/10. fication of neuroactive dopamine serotonin and kynurenine path-
1016/j.pediatrneurol.2015.04.016. way metabolites in gender-specific youth urine by ultra perfor-
15. Hinz M, Stein A, Trachte G, Uncini T. Neurotransmitter testing of mance liquid chromatography tandem high resolution mass spec-
the urine: a comprehensive analysis. Open Access J Urol. 2010;2: trometry. J Pharm Biomed Anal. 2016;122:42–51. https://doi.org/
177–83. https://doi.org/10.2147/OAJU.S13370. 10.1016/j.jpba.2016.01.031.
16. Lepschy M, Rettenbacher S, Touma C, Palme RG. Excretion of 31. Chan ECY, Ho PCL. Chromatographic measurements of catechol-
catecholamines in rats, mice and chicken. J Comp Physiol B. amines and metanephrines. In: Bertholf R, Winecker R, editors.
2008;178(5):629–36. https://doi.org/10.1007/s00360-008-0254-z. Chromatographic methods in clinical chemistry and toxicology.
17. Panuwet P, Hunter RE, D’Souza PE, Chen X, Radford SA, Cohen Hoboken, NJ: John Wiley & Sons, Ltd; 2007. p. 101–26.
JR, et al. Biological matrix effects in quantitative tandem mass 32. Li XS, Li S, Kellermann G. Simultaneous extraction and determina-
spectrometry-based analytical methods: advancing biomonitoring. tion of monoamine neurotransmitters in human urine for clinical rou-
Crit Rev Anal Chem. 2016;46(2):93–105. https://doi.org/10.1080/ tine testing based on a dual functional solid phase extraction assisted
10408347.2014.980775. by phenylboronic acid coupled with liquid chromatography-tandem
18. Peaston RT, Weinkove C. Measurement of catecholamines and their mass spectrometry. Anal Bioanal Chem. 2017;409(11):2859–71.
metabolites. Ann Clin Biochem. 2004;41(Pt 1):17–38. https://doi. https://doi.org/10.1007/s00216-017-0231-z.
org/10.1258/000456304322664663. 33. Kaspar H, Dettmer K, Chan Q, Daniels S, Nimkar S, Daviglus ML,
19. Eisenhofer G, Peitzsch M, McWhinney BC. Impact of LC-MS/MS et al. Urinary amino acid analysis: a comparison of iTRAQ-LC-
on the laboratory diagnosis of catecholamine-producing tumors. MS/MS, GC-MS, and amino acid analyzer. J Chromatogr B.
Trends Anal Chem. 2016;84:106–16. https://doi.org/10.1016/j. 2009;877(20–21):1838–46. https://doi.org/10.1016/j.jchromb.
trac.2016.01.027. 2009.05.019.