Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117

Contents lists available at SciVerse ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

journal homepage: www.elsevier.com/locate/jpba

Immunomagnetic molecular probe with UHPLC–MS/MS: A promising way for

reliable bronchial asthma diagnostics based on quantification of cysteinyl
leukotrienes
Kamila Syslová a , Adéla Böhmová a , Elvan Demirbağ a , Kateřina Šimková a , Marek Kuzma b ,
Daniela Pelclová c , Vratislav Sedlák d , Petr Čáp e , Pavel Martásek c , Petr Kačer a,∗
a
Institute of Chemical Technology, Technicka 5, 166 28 Prague 6, Czech Republic
b
Institute of Microbiology, Videnska 1083, 142 20 Prague 4, Czech Republic
c
1st Medicinal Faculty, Charles University, Na Bojisti 1, 120 00 Prague 2, Czech Republic
d
University Hospital Hradec Kralove, Sokolska 581, 500 05 Hradec Kralove, Czech Republic
e
Na Homolce Hospital, 120 00 Prague, Czech Republic

a r t i c l e i n f o a b s t r a c t

Article history: A sensitive and precise method for simultaneous quantification of cysteinyl leukotrienes (=cys LTs) –
Received 28 December 2012 leukotriene C4 (=LTC4 ), leukotriene D4 (=LTD4 ) and leukotriene E4 (=LTE4 ) – essential biomarkers of
Received in revised form 26 March 2013 bronchial asthma present in exhaled breath condensate (=EBC) was developed. An immunomagnetic
Accepted 28 March 2013
molecular probe was prepared by anchoring cysteinyl leukotrienes antibody on the surface of function-
Available online 8 April 2013
alized monodispersed magnetic particles and used to selectively isolate cys LTs from biological matrices
– EBC, plasma and urine. Immobilization and the immunoaffinity capture procedures were optimized
Keywords:
to maximize the amount of separated cys LTs, which were detected “off-beads” after acidic elution by
Cysteinyl leukotrienes
Immunomagnetic separation
UHPLC–ESI-MS/MS operated in a multiple reaction monitoring mode. The developed method was char-
UHPLC–MS acterized with high precision ≤13.6% (intra-day precision determined as RSD) and ≤14.5% (inter-day
Bronchial asthma precision determined as RSD), acceptable accuracy ≤18.5% (determined as RE), and high recovery of
Biological matrices immunoseparation (≥93.1%) in aforementioned biological matrices. The applicability of the method was
demonstrated on EBC, plasma and urine clinical samples of patients with various subtypes of bronchial
asthma (occupational, steroid-resistant, moderate with and without corticosteroids therapy) and healthy
subjects where reasonable differences in cys LTs concentration levels were found. Combining extremely
selective immunomagnetic separation with highly sensitive and precise detection step, the developed
method was used to aid diagnosis, predict the most effective therapy, and monitor the response to treat-
ment. The detection of elevated inflammatory mediators (cys LTs) in EBC of subjects with relatively
asymptomatic asthma and normal pulmonary function tests could offer a novel way for monitoring the
lung inflammation and perhaps initiating treatment in an earlier stage.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction flow rate test. In addition, methacholine or histamine challenge

tests can evaluate the airways hyperactivity, typical for asthma,
Bronchial asthma, defined as a chronic inflammatory disorder and fractional exhaled nitric oxide (FENO) test measures the level
of the airways, is one of the top public health problems glob- of eosinophilic inflammation in the airways. Unfortunately, this
ally affecting both children and adults. Therefore, proper and early non-invasive test is negatively influenced in smokers. At present,
diagnostics and treatment is necessary. However, the diagnostics the quantification of inflammation in the lungs may be based on
can be rather difficult. Currently, there is no precise physiologi- invasive (open lung biopsy [1–3], bronchoalveolar lavage [2,4]) or
cal, immunological, or histological test for diagnosing asthma. In semi-invasive methods (method of induced sputum [5,6]) and the
order to rule out other possible conditions, the lung functions are measurement of inflammatory markers in plasma and urine, which
generally tested first in an asthma diagnosis process. Tests to mea- are likely to reflect systemic rather than lung inflammation. A rela-
sure the lung functions include spirometry test and peak expiratory tively novel method is the analysis of EBC, collected during normal
tidal breathing, which reflects the composition of extracellular lung
fluid. The measurements of metabolite products in the EBC are
∗ Corresponding author. Tel.: +420 220 444 158; fax: +420 220 444 340. absolutely non-invasive and repeatable with a good potential to
E-mail addresses: [email protected], [email protected] (P. Kačer). become the preferred alternative. New approaches are based on

0731-7085/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jpba.2013.03.026
 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117 109

Fig. 1. Arachidonic acid derivatives – potential biomarkers of bronchial asthma.

attempts of identifying robust biomarkers which could be utilized (cys LTs, leukotriene B4 (=LTB4 )) in EBC of asthma patients. Elevated
in establishing the diagnosis of asthma [7–9]. The former studies concentrations of 8-iso-prostaglandin F2␣ , a marker of oxidative
investigated the predictive value of EBC pH for asthma, the latter stress [18], have been reported in inflammatory respiratory dis-
researched hydrogen peroxide, nitrogen oxides (FENO is nowadays eases, including asthma [19], COPD [20], and cystic fibrosis [21],
diagnostically/clinically used) [10], arachidonic acid derivatives, and in healthy smokers [22]. Therefore, measurements of these
cytokines and others. Analysis of breath volatile compounds by mediators are considered potentially effective in the establishment
electronic nose has been proposed as a new technique for assessing of asthma diagnosis in a large cohort. Leukotrienes are derived
airway inflammation in respiratory medicine [11]. Arachidonic acid from arachidonic acid, which is also the precursor of prostaglandins
derivatives, especially cys LTs, have shown the most consistent (Fig. 1). There are two families of leukotrienes (LTB4 and cys LTs –
results for the diagnosis of asthma [12]. Various inflammatory LTC4 , LTD4 , LTE4 and metabolites of LTE4 – leukotriene F4 (LTF4 ) and
markers present in EBC have been investigated as possible asthma N-acetyl leukotriene E4 (N-acetyl-LTE4 )). LTB4 acts primarily when
biomarkers and several independent studies [13–17] have indi- the inflammation is dependent on neutrophils such as cystic fibro-
cated the presence of elevated levels of arachidonic acid derivatives sis, inflammatory bowel disease, and psoriasis. The second group
110 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117

(cys LTs) is concerned especially with eosinophil and mast cell solution in water, Penta, Czech Republic), formic acid (LC–MS
induced bronchoconstriction in asthma. They bind to highly selec- grade, Fisher Chemical, USA), sodium phosphate monobasic
tive receptors on bronchial smooth muscle and other airway tissue. monohydrate (≥99%, Fluka, Switzerland), sodium phosphate
Leukotrienes, in particular cys LTs, have a potentially important dibasic dihydrate (≥99.5%, Riedel de Haën, Germany), sodium
role in asthma biomarker-based diagnostics [14,23,24]. Concentra- chloride (99%, Aldrich, USA), potassium chloride (99%, Penta, Czech
tions of cys LTs in EBC are elevated in patients with asthma when Republic), sodium phosphate (99%, Riedel de Haën, Germany),
compared to healthy controls and their levels often decrease after urea (99%, Aldrich, USA), bovine serum albumin (=BSA) (fraction
directed anti-inflammatory therapy. Measurement of leukotrienes V, Sigma, USA), tris(hydroxymethyl)aminomethan (=Tris) (99.9%,
in EBC and urine could be useful for asthma phenotyping [25] and Aldrich, USA), hydrochloric acid (p.a., Penta, Czech Republic),
assessing the effect of anti-inflammatory treatment in patients with glycine (99%, Aldrich, USA), sodium azide (99%, Aldrich, USA) and
asthma [26]. Tween 20 (99%, Cayman Chemical, USA).
In the past, immunochemical and analytical techniques like RIA, The following buffers were used: Buffer A: 0.1 M Na–phosphate
ELISA and GC–MS were used for their concentration level eval- buffer pH 7.4 (0.262 g NaH2 PO4 ·H2 O and 1.442 g Na2 HPO4 ·2H2 O
uation [12,27–30]. During the last decade, LC–MS has become a was dissolved in water and volume adjusted to 100 ml); Buffer
promising technique for trace analysis of various compounds con- B: phosphate buffered saline (PBS) pH 7.4 with 0.1% (w/v) BSA
tained in complex biological matrices (plasma, urine, cerebrospinal (0.88 g NaCl and 0.8 g BSA was dissolved in 0.01 M Na–phosphate
fluid, bronchoalveolar lavage fluid, etc.) and it has been success- pH 7.4 and volume was adjusted to 100 ml); Buffer C: 0.2 M Tris
fully applied for the detection and quantification of EBC markers (2.42 g) with 0.1% (w/v) BSA was dissolved in distilled water, pH
[8,12–14,17,27]. Since the biomarkers are often detected in EBC at was adjusted with 1 M hydrochloric acid to 8.5 and volume was
levels as low as picograms per ml, concentration steps (e.g. solid adjusted to 100 ml. Buffer D: 100 mM glycine (pH was adjusted to
phase extraction, lyophilization, immunoseparation, etc.) are nec- 2.5 with 1 M HCl).
essary in order to achieve such a relatively low detection limit. The
mentioned methods vary primarily in their selectivity related to 2.2. Materials
the given analytes. The highest selectivity for cys LTs is achieved by
immunoaffinity separation, which is determined by a highly spe- Dynabeads® M-280, Tosylactivated superparamagnetic
cific reaction between an antigen (cys LTs) and an antibody (anti-cys polystyrene beads coated with a polyurethane layer were obtained
LTs antibody). In this study, biomarker extraction from biological from Dynal (Norway). Physical characteristics of beads were as
matrices (EBC or urine or plasma) is provided by magnetic parti- follows: diameter: 2.8 ± 0.2 ␮m; surface area: 4–8 m2 /g; active
cles with anchored anti-cys LTs antibodies. By using an external chemical functionality: 50–70 ␮mol/g; density: 1.3 g/cm3 ; con-
magnetic field, the produced antigen–antibody complex can be centration: 2 × 109 beads/ml (approx. 30 mg/ml). Prior to use, the
separated from the solution. The covalent bond between a magnetic beads were washed twice in Buffer A. Magnetic particle concen-
particle and an antibody allows one to perform the regeneration of trator (Dynal, Norway) was used to facilitate supernatant removal
immunobeads without any significant loss of the binding capac- and washing steps.
ity. For the quantification and detection of biomarkers, a highly
selective and sensitive method by UHPLC–ESI-MS/MS was applied. 2.3. Preparation of anti-cys LTs coated magnetic particles
For its extremely high degree of selectivity, the multiple reaction
monitoring (MRM) mode was used together with stable-isotope Dynabeads® suspension (30 ␮l) was separated using magnetic
dilution assay for its high precision of quantification. The analyti- concentrator and after 4 min the supernatant was removed. Out of
cal procedure was optimized, validated and analytically tested on magnetic field the beads were washed twice with Buffer A (200 ␮l)
real EBC, urine and plasma samples collected from patients diag- and subsequently separated from the supernatant again. Anti-cys
nosed with bronchial asthma and on the control group of healthy LTs antibody (0.9 ␮g) was brought to washed beads, vortexed for
subjects. 1 min and incubated under shaking at 37 ◦ C for 24 h. After magnetic
separation, the supernatant was removed and the particles were
washed twice with Buffer B (400 ␮l) at 4 ◦ C for 5 min. Free tosyl
2. Experimental groups were blocked by Buffer C (400 ␮l) at 37 ◦ C for 4 h. The par-
ticles were washed in Buffer B (400 ␮l) at 4 ◦ C for 5 min. The beads
2.1. Chemicals and reagents were washed in 1% solution of Tween 20 (400 ␮l) at 25 ◦ C for 5 min
in order to reduce the aggregation of particles and eliminate non-
All chemicals and reagents were of commercial origin and specific binding of antibodies onto the magnetic particle surface
the source was as follows: mixture of cys LTs contains LTC4 , (removing of the physisorbed portion of antibodies). Finally, the
LTD4 , LTE4 , LTF4 and N-acetyl-LTE4 (purity of each individual immunomagnetic beads were washed in Buffer B (400 ␮l) at labo-
substance ≥98%, Cayman Chemical, USA), cys LTs mono- ratory temperature for 5 min and were stored in Buffer A (400 ␮l)
clonal antibody (=anti-cys LTs) developed in mouse (Abcam, with 0.1% sodium azide (w/v) at a temperature of 4 ◦ C.
UK), LTC4 polyclonal antibody developed in rabbit (Bretin
Pharma, France), LTD4 polyclonal antibody developed in rab- 2.4. Clinical sample collection
bit (Bretin Pharma, France), LTE4 polyclonal antibody developed in
rabbit (Bretin Pharma, France), 5S-hydroxy-6R-(S-glutathionyl)- 2.4.1. Exhaled breath condensate
7E,9E,11Z,14Z-d5 -eicocatetraenoic-19,19,20,20,20-d5 acid (= The commercially available condenser EcoScreen (Jaeger,
LTC4 -d5 ) (≥97%, ≥99% deuterated form, Cayman Chemical, Germany) was used to collect the samples of EBC. All clinical sam-
USA), 5S-hydroxy-6R-(S-cysteinylglycinyl)-7E,9E,11Z,14Z-d5 - ples were collected between 8 and 12 a.m. The subjects (patients
eicocatetraenoic-19,19,20,20,20-d5 acid (=LTD4 -d5 ) (≥97%, ≥99% diagnosed with various sub-types of bronchial asthma and the con-
deuterated form, Cayman Chemical, USA), 5S-hydroxy-6R-(S- trol group of healthy non-smokers) were encouraged to perform a
cysteinyl)-7E,9E,11Z,14Z-d5 -eicocatetraenoic-19,19,20,20,20-d5 tidal breathing through a mouthpiece connected to the condenser
acid (=LTE4 -d5 ) (≥97%, ≥99% deuterated form, Cayman Chemical, (−20 ◦ C) for 5–10 min (volume of exhaled air was held at a constant
USA), acetonitrile (LC/MS grade, Fluka, Switzerland), water (LC/MS level of 120 l) while wearing a nose-clip. The acquired EBC volumes
grade, Fluka, Switzerland), ammonium hydroxide (28% NH3 of samples ranged between 1 and 2 ml. The EBC sample (1.0 ml) was
 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117 111

immediately spiked with 100 pg of each deuterium-labelled inter- Table 1

MRM transitions, collision energies (CE) and voltage on S-lens for cys LTs and their
nal standard (LTC4 -d5 , LTD4 -d5 and LTE4 -d5 ; concentration of each
internal standards.
cys LTS was 100 pg/1 ml EBC) and subsequently frozen to −80 ◦ C.
The storage period did not exceed 1 month. Anatyte MRM transmissions (m/z → m/z) CE (eV) S-lens (V)
The contamination of EBC by saliva was determined by activity LTC4 624.1 → 351.2 17 58
of salivary ␣-amylase (detail of determination contamination by LTC4 -d5 629.1 → 356.2 17 58
saliva is described in [31]). Maximal ␣-amylase activity in all sam- LTD4 495.2 → 477.3 16 52
LTD4 -d5 500.2 → 482.3 16 52
ples did not exceed 0.1% of the saliva activity and thus the absence
LTE4 438.2 → 333.1 16 47
of the saliva contamination in EBC samples was sufficiently proved. LTE4 -d5 443.2 → 338.1 16 47

2.4.2. Urine
The collection of human urine samples was carried out between autosampler (Thermo Scientific, USA). The mobile phase acetoni-
8 and 12 a.m. Urine was centrifuged (3000 rpm, i.e. 500 × g; time trile and water with 0.1% formic acid in a ratio of 70:30 (v/v) was
of centrifugation was 5 min) and the internal standards (100 pg of used for isocratic elution at a flow rate of 400 ␮l/min. The injec-
each internal deuterium standard – LTC4 -d5 , LTD4 -d5 and LTE4 -d5 ) tion volume was 10 ␮l. The UHPLC analysis was carried out using a
were added to the sample (0.2 ml). The sample was frozen at a tem- Kinetex C18 2.7 ␮m 100 Å column (50 mm × 2.1 mm) connected to
perature of −80 ◦ C. The concentration of creatinine was determined a Kinetex C18 pre-column (Phenemonex, USA).
in a separate portion of urine using Creatinine Assay Kit (Cayman A triple quadrupole mass spectrometer TSQ Vantage (Thermo
Chemical Company, USA). Scientific, USA) was used as a detector operating with electrospray
ionization in negative mode (ESI− ). For the measurement of cys
LTs and their deuterium-labelled analogues (LTC4 -d5 , LTD4 -d5 and
2.4.3. Plasma
LTE4 -d5 ), tandem mass spectrometry (MS/MS) was used, where the
Blood samples were collected between 8 and 12 a.m. by
parallel scan of particularly detected cys LTs was set according to
venipuncture and immediately placed into tubes containing EDTA
Table 1. The conditions on the mass spectrometer were optimized
as an anticoagulant. The samples were centrifuged (3000 rpm, i.e.
and were as follows: spray voltage −2500 V, temperature of ion
500 × g; time of centrifugation was 10 min) at a temperature of 4 ◦ C.
transfer tube 350 ◦ C, temperature of H-ESI vaporizer 350 ◦ C, pres-
The plasma fraction (0.5 ml) was removed, the internal standards
sure of sheath gas (nitrogen) 35 psi, flow of auxiliary gas (nitrogen)
(100 pg of each internal standard – LTC4 -d5 , LTD4 -d5 and LTE4 -d5 )
10 arbitrary units. The data were acquired and processed using the
were added, and the samples were immediately frozen at −80 ◦ C.
software Xcalibur 2.1.0 (Thermo Scientific, USA).

2.5. Preparation of samples for LC–ESI-MS/MS analysis

2.7. Validation of assay procedure
(“pre-treatment method”)
The analytical method was validated for the following set of
2.5.1. Exhaled breath condensate
parameters: accuracy, precision, lower limit of detection (LLOD),
Anti-cys LTs coated magnetic beads (approx. 1 mg) stored in
lower limit of quantification (LLOQ) and linearity.
Buffer A with sodium azide (0.1% solution – w/v) were washed
two times with Buffer B (400 ␮l) and re-suspended in 1 ml of EBC
2.7.1. Standard preparation and calibration procedure
(contained 100 pg of each deuterium labelled cys LT/1 ml EBC). The
The cys LTs stock solution (20 ng/ml) was prepared by diluting
suspension was incubated under shaking for 1 h and the super-
biomarker standard in mobile phase (water with 0.1% formic acid:
natant was removed. The beads were washed five times in Buffer B
acetonitrile) and was used further for the preparation of a series of
(400 ␮l). The supernatant was removed and the beads were put
calibration samples with the following leukotriene concentrations:
into Buffer D (50 ␮l, 1 min) which caused elution of the cys LT
5, 10, 25, 50, 100, 250 and 500 pg/50 ␮l (each calibration sample was
bound from complex with antibody. The supernatant was removed
spiked with 250 pg of each ISs (LTC4 -d5 , LTD4 -d5 and LTE4 -d5 )). The
immediately and the sample prepared in this way was analyzed by
peak area of particular cys LT with the assigned IS was used as the
UHPLC–ESI-MS/MS. The beads were regenerated for next immuno-
function to quantify the concentration.
magnetic separation by washing five times in Buffer B (400 ␮l). The
regenerated beads were stored at 4 ◦ C in Buffer A (400 ␮l) with 0.1%
2.7.2. Accuracy, precision, recovery, LLOD and LLOQ
sodium azide (w/v).
Accuracy, intra-day precision and inter-day precision were
determined at three different levels of cys LTs (LLOQ = low level;
2.5.2. Plasma medium, and high level). Accuracy was evaluated by comparing
PBS (0.5 ml) was added to the sample of plasma (0.5 ml plasma the mean calculated values (n = 5 for each concentration level) with
containing the internal standards–100 pg of each cys LT). The pre- the respective nominal concentration values of cys LTs added to the
treatment method followed the same protocol as in the case of EBC biological matrix. Intra-day precision was determined by measur-
samples (see Section 2.5.1). ing 5 replicates per concentration level, on the same day. Inter-day
precision was assessed by analysing 5 replicates per concentration
2.5.3. Urine level, on five consecutive days.
Urine samples (0.2 ml; concentration of internal standards: LLOD was defined as the lowest concentration of the analyte
100 pg of each cys LT/0.2 ml urine) were centrifuged (3000 rpm, i.e. which gave a peak response with a signal to noise ratio (S/N) of at
500 × g; time of centrifugation was 3 min) prior to pre-treatment. least three. LLOQ was defined as the lowest concentration of the
Pre-treatment method used the same protocol as in the case of EBC analyte: (1) which could be reliably quantified with a precision not
samples (see Section 2.5.1). exceeding 20% of the relative standard deviation (RSD) and accu-
racy not deviating more than ±20% of the actual value and (2) which
2.6. UHPLC–ESI-MS/MS analysis gave a peak response with a S/N of at least 10 [24]. Both LLOQ
and LLOD were determined empirically using consecutive dilu-
The UHPLC system consisted of a quaternary pump with an tions of cys LTs in free matrix of biological fluid (repeat separation
Accela 1250 degasser (Thermo Scientific, USA) and Open Accela of cys LTs from EBC, plasma, or urine by immunoextraction with
112 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117

immunoparticles). The linearity of the method was investigated by on the reaction between antigen and antibody offer a highly spe-
using linear regression analysis [32]. cific way of isolating a biomarker from complex biological matrices
For the extraction recovery, cys LTs (at three concentration lev- such as exhaled breath condensate, plasma, cerebrospinal fluid,
els: LLOQ, medium and high level) and their internal standards etc. Anchoring the antibodies on magnetic beads offers not only
(LTC4 -d5 , LTD4 -d5 and LTE4 -d5 , amount 100 pg) were added to high selectivity towards desired biomarker(s), but the magnetic
cys-LTs-free matrix and then separated with immunomagnetic par- support also allows easy and rapid separation and the chance of
ticles by following a procedure described in Section 2.5. its repeated use is often described as excellent [33]. A general
scheme of the method is shown in Supplementary data (Fig. S1).
2.7.2.1. Exhaled breath condensate. The EBC matrix for validation The first step of the procedure was the preparation of the immuno-
was prepared by combining EBC samples from healthy subjects. magnetic probe by functionalization of magnetic particles with
The concentration of cys LTs was determined by UHPLC–ESI-MS/MS an antibody against cys LTs (=anti-cys LTs). The second step was
methods. An aliquot (1 ml) of the EBC matrix was spiked with immunomagnetic separation, in which the clinical sample contain-
three different amounts of cys LTs (LLOQ = low, medium = 50 pg, ing cys LTs (EBC, plasma and urine) was used for the separation
and high = 100 pg of cys LTs). These samples were pre-treated by anti-cys LTs-coated magnetic particles. After the immunocap-
by the immunomagnetic separation procedure and quantified by ture, the immunoparticles were removed from the biomatrix by
UHPLC–ESI-MS/MS. means of external magnetic field and immediately washed in order
to remove non-binding components. The concentrated and purified
2.7.2.2. Plasma. Plasma was prepared by combining plasma sam- cys LTs bound to the particles were then eluted with appropriate
ples from healthy subjects and the concentration levels of cys LTs solvent, and subsequently detected by UHPLC–ESI-MS/MS. How-
were determined by UHPLC–MS/MS analysis. Different amount of ever, it was also possible to analyze them after the elution by
cys LTs (LLOQ, 25 and 50 pg) were added to the plasma (0.5 ml). traditional immunochemical methods like ELISA.
These samples were submitted to the immunomagnetic pre-
treatment procedure and quantified by UHPLC–ESI-MS/MS. 3.1. Immunomagnetic separation

Due to the very low concentration of cys LTs present in biolog-

2.7.2.3. Urine. For the method validation, urine matrix was used.
ical samples, their separation and pre-concentration prior to the
It was prepared by combining urine samples and the concentra-
UHPLC–ESI-MS/MS analysis is highly desirable. The separation was
tion levels of cys LTs were determined by UHPLC–MS/MS. Different
performed using magnetic beads with anchored anti-cys LTs anti-
amount of cys LTs (LLOQ, 25 and 50 pg) were added to the urine
body. The preparation of anti-cys LTs-coated magnetic microbeads
matrix. These samples were submitted to the immunomagnetic
was realized according to a procedure described in the experimen-
pre-treatment procedure and quantified by UHPLC–ESI-MS/MS.
tal part (see Section 2.3). The first step of the procedure was the
preparation of the affinity probe by functionalization of monodis-
2.8. Clinical study perse, hydrophobic magnetic beads with mouse antibody against
cysteinyl leukotrienes (anti-cys LTs). The antibodies were immobi-
The objective of the clinical study was to verify the possibility lized through their primary amino groups on the surface of the
of using immunomagnetic separation as a pre-treatment method magnetic beads activated by p-toluenesulfonyl (=tosyl) reactive
in combination with UHPLC–ESI-MS/MS for the evaluation of clin- groups. Those experiments employed affinity-purified monoclonal
ical samples. EBC, urine and plasma samples were collected from anti-cys LTs, which were anchored without alternation of their
persons with various sub-types of bronchial asthma (mild to severe bioactivity. The immobilization procedure was optimized in order
occupational asthma (30 subjects), hard-to-treat asthma (32 sub- to achieve maximal density of anti-cys LTs on the surface of the
jects), moderate with (28 subjects) and without corticosteroids magnetic beads. Further details are described in Section 3.1 of Sup-
therapy (25 subjects)). The control group was represented by 35 plementary data. In the course of the immobilization procedure,
subjects with similar average age (45–54 years, male) and gender two anchoring mechanisms could be distinguished: covalent bond-
without bronchial asthma diagnosis (healthy subjects). The sam- ing (chemisorption of anti-cys LTs) and non-covalent bonding. The
ples collected during the clinical study were worked up by the chemisorbed fraction of the antibody needs a longer reaction time
above-mentioned immunomagnetic separation and the concentra- but, thanks to the chemical bonding to the surface, it is stable and
tion levels of cys LTs were determined by UHPLC–ESI-MS/MS. more convenient for the repeatable use of anti-cys LTs particles
The study was carried out according to the Helsinki Declara- because of their uniform properties. Also, the influence of other
tion. The Ethics Committee of the 1st Faculty of Medicine, Charles parameters such as pH, coating time and temperature was inves-
University, approved all examinations and tests, and all of the study tigated, and their influence on the quality of prepared anti-cys LTs
subjects gave their written informed consent for all tests and exam- immunomagnetic microbeads was evaluated on the basis of the
inations. recovery parameter in the first and second cycle of immunomag-
netic separation. On the basis of the results given in Table 2 it
2.9. Statistical analysis was found that the optimal preparation conditions were as fol-
lows: the ratio of antibodies to number of magnetic microbeads
The acquired data were statistically analyzed using the software was 60 ␮g of anti-cys LTs to 6 × 107 (0.9 mg) microbeads, coat-
Statistica, version 7.0. The statistical analyses were materialized ing temperature 37 ◦ C, coating time 24 h and pH 8.8. Under these
using Student’s t-test, where the p values were considered signifi- conditions, prepared anti-cys LTs immunomagnetic beads carried
cant if p < 0.05. approx. 45% of chemisorbed anti-cys LTs and 55% of the physisorbed
ones. The physisorbed portion is usable only in the first cycle of
3. Results and discussion immunomagnetic separation since only the chemisorbed part per-
sists on the surface of the beads and is available for the next cycle
A method based on combined use of immunoaffinity capture on of immunoseparation after elution and regeneration of immuno-
magnetic particles and UHPLC–ESI-MS/MS for rapid detection of magnetic beads. The other alternative is the application of one
cys LTs (LTC4 , LTD4 and LTE4 ), the most prominent biomarkers of additional elution step in the course of the preparation procedure.
bronchial asthma present in EBC, was developed. Methods based The step consisted in washing the beads by 1% solution of Tween 20,
 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117 113

Table 2
Optimization of condition for coated antibodies on magnetic particles.

Amount antibody (␮g) Amount particles Temperature (◦ C) Time of coating (h) pH Detected amount of cys LTs
7
12 10 37 24 8.8 1.24 ng
60 107 37 24 8.8 5.38 pg
100 107 37 24 8.8 5.74 ng
300 107 37 24 8.8 6.82 ng
60 107 4 24 8.8 4.38 ng
60 107 25 24 8.8 5.03 ng
60 107 37 18 8.8 4.98 ng
60 107 37 30 8.8 5.44 ng
60 107 37 24 8.4 5.22 ng
60 107 37 24 9.2 4.42 ng

which removes the physisorbed amount of anti-cys LTs and enables there was not a sufficient amount of free antibodies on the sur-
the preparation of immunomagnetic microbeads with uniform sep- face of immunobeads to bind the excess quantity of cys LTs. With
aration ability in several consecutive separation cycles. In such a regard to the accuracy of the method, 0.9 mg of the immunobeads
way, the prepared molecular probe for immunofishing of cys LTs was used during the clinical analyses, which corresponded to the
was used for the evaluation of clinical samples. immunoseparation proceeding in the linear dependence area. After
Immunomagnetic isolation of cys LTs by anti-cys LTs-coated the immunoseparation, the separated biomarkers (cys LTs) were
magnetic beads was realized by the procedure described in the eluted from the substrate–antibody complex by Buffer D, which
experimental part (see Section 2.5). Further details are described did not denature the protein antibody and allowed the regen-
in Section 3.1 of Supplementary data. eration of the magnetic particles with anchored antibodies. By
The immunomagnetic extraction step was optimized to obtain using the described procedure, it was possible to separate more
conditions exhibiting high reproducibility and precision. The than 90% of the cys LTs originally contained in the clinical sample
experiments were performed with 0.9 mg of immunomagnetic (recovery parameter). The ratio of nonspecific reactions with the
beads and a cys LTs-free matrix containing externally added cys immunoaffinity sorbent was negligible.
LTs. The amount of beads was selected based on the binding capac- Immunomagnetic microbeads with coated antibody (anti-
ity and expected levels of cys LTs in matrix (e.g. quantity of cys LTs cys LTs) were compared with a mixture of immunomagnetic
in EBC clinical samples ranged in the interval of 20–150 pg/ml of microbeads with three different specific antibodies – anti-LTC4 ,
EBC). The time dependence of the immunoseparated cys LTs levels LTD4 and LTE4 . The prepared immunobeads were compared on
at a ratio of immunomagnetic beads: cys LTs (0.9 mg:100 pg in 1 ml the basis of determined specificity of the antibodies – see Table 3.
of EBC) is shown in Fig. 2. With prolongation of the immunosepa- The magnetic immunobeads with an anti-cys LTs antibody demon-
ration time, the separated amount of cys LTs from EBC increased strated worse specificity to LTE4 (65%), which is comparable to
but decomposition of the substances occurred because of their the specificity of its metabolite N-acetyl-LTE4 (60%). Antibodies
known low temperature stability (temperature of immunosepara- against individual leukotrienes (anti-LTC4 , anti-LTD4 and anti-LTE4 )
tion ∼25 ◦ C) [34]. After achieving the maximum separated amount show a high cross reaction with the leukotriene of the fifth row
(30 min), a degradation process started to prevail and therefore (e.g. LTC5 , LTD5 and LTE5 ), but these leukotrienes are not involved
the optimal period for the immunoseparation of cys LTs from EBC in the pathophysiology of bronchial asthma. Other eicosanoids
was determined as 30 min. An equally important parameter was demonstrate low levels of cross-reactions with all types of tested
the ratio of the quantities of immunomagnetic beads to cys LTs. antibodies.
Fig. 3 shows the dependence of the levels of immunoseparated cys
LTs on their original levels in the matrix (EBC, urine and plasma), 3.2. UHPLC–ESI-MS/MS analysis
which conspicuously suggest that the dependencies had the char-
acter of an adsorption isotherm, where two different areas could be The determination of picogram levels of cys LTs in EBC
described separately. In the first area with the linear dependence, (alternatively urine or plasma) is a tractable problem when the
the quantity of the antigen–antibody complex produced grew with
an increasing amount of biomarkers (cys LTs) in the matrix (Fig. 3
– enlarged working area), whereas in the second (non-linear) area,

Fig. 3. Efficiency of cys LTs immunomagnetic separation – EBC (), urine (♦) and
Fig. 2. Time dependence extraction of cys LTs from sample. plasma ().
114 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117

Fig. 4. UHPLC–ESI-MS/MS chromatograms for cys LTs. (A) EBC, (B) plasma, and (C) urine.

UHPLC–MS/MS method is employed. Using UHPLC–MS/MS offers (e.g. LTF4 : tR = 1.0 min and N-acetyl LTE4 : tR = 1.6 min). The mass-
not only high selectivity, sensitivity, accuracy and precision, but spectrometric detection was realized using negative electrospray
also high throughput, which is important for a method designated ionization (ESI− ) in a very selective MRM mode. The collision
for routine practice. energy (CE) and other parameters related to the mass spectrometer
The UHPLC conditions were optimized with respect to the lim- were optimized in order to acquire maximal yield of the dissocia-
itations of the ionization method (electrospray ionization, ESI) tion and thereby high sensitivity of the developed method (see the
and also with the emphasis on high detector sensitivity (Kine- experimental part – Section 2.6). With regard to the sensitivity of
tex C18 (2.7 ␮m); mobile phase acetonitrile: 0.1% formic acid in the biomarkers to several factors, it is essential to label the sample
water 70:30 (v/v); isocratic elution at a flow of 400 ␮l/min). The immediately after the collection with an isotope-labelled internal
LC method allowed the separation of monitored individual cys standard (addition of 100 pg of LTC4 -d5 , LTD4 -d5 and LTE4 -d5 to 1 ml
LTs (LTC4 : tR = 0.5 min, LTD4 : tR = 0.7 min and LTE4 : tR = 1.2 min; of EBC; 200 pg of LTC4 -d5 , LTD4 -d5 and LTE4 -d5 to 1 ml of plasma
dead time of column: tR = 0.3 min, Fig. 4) and other eicosanoids and 500 pg of LTC4 -d5 , LTD4 -d5 and LTE4 -d5 to 1 ml of urine). This
 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117 115

Table 3
Specificity of immunomagnetic bead.

Analyte Antibody

anti-cys LTs anti-LTC4 anti-LTD4 anti-LTE4 Mix of anti-LTC4 ,

LTD4 and LTE4

LTC4 100% 100% 32% 2% 100%

LTD4 100% 46% 100% 31% 100%
LTE4 65% 11% 13% 100% 100%
LTB4 <0.1% <0.1% <0.1% <0.1% <0.1%
LTF4 12% 9% 14% 11% 11%
N-Acetyl LTE4 60% 28% 25% 20% 24%
LTB5 <0.1% <0.1% <0.1% <0.1% <0.1%
LTC5 67% 100% 28% 1% 44%
LTD5 53% 48% 100% 19% 56%
LTE5 36% 2% 31% 100% 44%
Arachidonic acid <0.1% <0.1% <0.1% <0.1% <0.1%
Prostaglandin E2 <0.1% <0.1% <0.1% <0.1% <0.1%
Prostaglandin D2 <0.1% <0.1% <0.1% <0.1% <0.1%

allows a highly precise quantification together with monitoring the 3.5. Human study
changes in sample composition which occur during its processing
(“stable-isotope dilution assay”). In this study, EBC, plasma and urine were acquired from a group
of subjects diagnosed with various subtypes of bronchial asthma
(mild to severe occupational asthma, steroid-resistant asthma with
3.3. Calibration procedure corticosteroid therapy, moderate with and without corticosteroids

The calibration graphs for cys LTs were constructed for all
matrices – EBC, plasma and artificial urine (all matrices were Table 5
originally free of cys LTs). The calibration was created from arti- Validation parameters (accuracy, precision and recovery) for methods combination
ficially prepared samples with increasing amounts of cys LTs and immunomagnetic separation and LC–MS/MS analysis.
their corresponding internal standards (100 pg of particular inter- Concentration Precision RSD (%) Accuracy, RE (%) Recovery (%)
nal standard in 1 ml of EBC, 0.5 ml of plasma or 0.2 ml of artificial
urine) after the pre-treatment with immunomagnetic beads with Intra-day Inter-day

anti-LTC4 , anti-LTD4 and anti-LTE4 antibodies. The samples were Exhaled breath condensate
immediately analyzed by UHPLC–ESI-MS/MS in the MRM mode. LTC4
LLOD 9.5 9.8 −10.5 95.6
The calibration graphs were formed using the ratio of a particu-
50 pg/ml 8.7 9.2 −9.5 96.4
lar biomarker peak area to the corresponding internal standard 100 pg/ml 8.1 8.8 −8.6 96.1
peak area (axis x) versus the biomarker concentration in the mobile LTD4
phase (axis y). The calibration curve was evaluated by the least LLOD 9.2 9.5 −10.4 95.8
square regression in the form y = ax + b. The results are summa- 50 pg/ml 7.8 8.3 −9.2 96.7
100 pg/ml 7.4 7.8 −8.3 97.1
rized in Table 4. All cys LTs exhibited a linear dependence within LTE4
the range starting at the LLOQ up to 5000 pg with the Pearson’s LLOD 8.5 8.9 −10.3 95.5
correlation coefficients (R2 ) higher than 0.997 for all analytes (the 50 pg/ml 7.8 8.3 −9.7 96.6
acceptable value of R2 for each calibration curve was R2 ≥ 0.99). 100 pg/ml 7.2 7.5 −8.4 96.8

Plasma
LTC4
3.4. Method validation LLOD 10.1 10.4 −15.1 95.2
25 pg/0.5 ml 9.5 9.9 −13.3 95.6
50 pg/0.5 ml 8.9 9.4 −12.6 96.2
The validation parameters of the method, such as the accuracy
LTD4
(RE) and intra and inter-day precision (RSD), were assessed by LLOD 10.3 10.7 −17.7 93.6
analyzing three various concentration levels of the cys LTs, which 25 pg/ml 9.6 9.8 −14.4 94.1
were determined five times by the developed method consisting of 50 pg/ml 8.7 9.3 −12.5 94.8
separation with immunomagnetic beads with anti-LTC4 , anti-LTD4 LTE4
LLOD 10.0 10.6 −17.9 93.1
and anti-LTE4 antibodies and UHPLC–ESI-MS/MS in an indepen- 25 pg/ml 9.2 9.7 −14.8 95.3
dent series of spiked EBC, urine and plasma samples. The values 50 pg/ml 8.3 8.8 −11.4 95.6
of accuracy and precision for each cys LT and a body fluid were
Urine
evaluated and are listed in Table 5. The values of LLOD and LLOQ LTC4
are summarized in Table 6. LLOD 13.6 14.5 −18.5 93.2
25 pg/mg creatinine 12.6 13.5 −17.4 94.7
50 pg/mg creatinine 10.6 11.2 −15.2 95.5
LTD4
Table 4 LLOD 12.5 13.1 −16.2 93.4
Linear regression analysis of calibration curve (n = 10) for cys LTs. 25 pg/mg creatinine 11.4 12.2 −14.9 94.7
50 pg/mg creatinine 10.2 10.8 −13.2 95.4
Substance Slope Intercept SD RSD (%) R2
LTE4
LTC4 234.42 −14.52 6.268 2.674 ≥0.9978 LLOD 11.7 12.5 −15.5 93.5
LTD4 265.31 −15.52 6.742 2.541 ≥0.9984 25 pg/mg creatinine 10.4 11.0 −13.7 95.1
LTE4 282.65 −12.53 6.578 2.311 ≥0.9980 50 pg/mg creatinine 9.8 10.3 −12.3 95.7
116 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117

Table 6
Validations parameters for methods consist of immunomagnetic separation and
LC–MS/MS detection – lower limit of detection (LLOD) and quantification (LLOQ).

Matrix LTC4 LTD4 LTE4

LLOD (pg/ml) EBC 3 2 2

LLOQ (pg/ml) EBC 6 5 4

LLOD (pg/0.5 ml) Plasma 5 4 4

LLOQ (pg/0.5 ml) Plasma 7 6 6

LLOD (pg/ml) Urine 6 5 5

LLOQ (pg/ml) Urine 8 7 7

therapy) and a control group of healthy subjects, and the con-

centration values of cysteinyl leukotrienes were determined. All
groups were of a similar average age and equal gender (45–54 years,
male subjects). The control group was limited to individuals with-
out bronchial asthma and other respiratory disorders. All samples
were collected between 8 and 12 a.m. Table 7 and Fig. S4 depict
the results of the clinical study. In the EBC matrix, substantial dif- Fig. 5. Levels of cysteinyl leukotrienes (LTC4 – ; LTD4 – ♦; LTE4 – ) in exhaled
ferences in the levels of LTC4 and LTE4 were observed among the breath condensate during bronchoprovocation test after administration of his-
groups of subjects. No significant difference in the LTD4 levels for tamine.
the diagnosis was found, although its average level was higher in
the subjects with bronchial asthma (24.4 ± 8.5 pg/ml) compared
4. Conclusion
to controls (16.4 ± 4.9 pg/ml). However, the confidence intervals
were overlapping and this parameter thus seemed statistically
An analytical method was developed for a parallel, rapid and
insignificant. The variations observed in matrices of both clinical
precise detection and quantification of cysteinyl leukotrienes in
groups were relatively tenuous; nevertheless, the values of mean
three different biological matrices (EBC, urine and plasma). The
concentrations demonstrated a statistical significance. The highest
assay incorporated a pre-treatment step for a rapid and effective
statistically significant difference in the concentrations levels of cys
pre-concentration (immunomagnetic particles with anchored anti-
LTs for patients and the control group was seen for sums of cys LTs in
cys LTs antibodies) of markers present in biological samples in low
EBC (see Fig. S4D). On the other hand, cys LTs in plasma and urine
concentrations. This step complied with the requirements of a rapid
in the patients and controls group did not statistically differ (see
and mild isolation prior to subsequent analysis that was compati-
Fig. S4E–L).
ble with the limited stability of cys LTs. Magnetic and mechanical
The developed diagnostic method was also applied in a clini-
properties of the developed immunomagnetic tool also allowed its
cal experiment, the aim of which was to verify the possibility of
repeatable use for sample pre-concentration. UHPLC–ESI-MS/MS
monitoring the course of an allergic disease or therapeutic proce-
was used as the detection method, where the MRM mode was
dure. A non-specific bronchoprovocation test was performed in a
used for its extremely high degree of sensitivity and selectivity.
group of volunteers (10) who were administered histamine, and
Stable-isotope dilution assay was employed for its high precision
EBC was collected both before the test and subsequently in 10-
in quantification, which has been established as the technique of
min intervals. From the results obtained (Fig. 5) it is clear that
choice for trace analyses of various compounds present in complex
the subject responded to histamine by increased production of
biological matrixes. The combination of mass spectrometry detec-
cys LTs, whose levels finally returned to the initial concentration.
tion with liquid chromatography separation was selected to retain
The test showed that this method has the potential for mon-
the analytes from the solvent front. The analytical procedure was
itoring the level of inflammation [35] and the effectiveness of
optimized and validated. The precision values for cys LTs in the
the elimination from the exposure to the allergen or therapeutic
EBC, urine and plasma ranged in an interval of 7.2–13.6% (intra-
procedures.
day precision, determined as RSD), and recovery values ranged in
Table 7
an interval of 93.1–97.1%. The performed clinical study confirmed
Results of the clinical study (95% confidence interval). the possibility of differential diagnosis of bronchial asthma using
the developed pre-treatment method for the determination of cys
LTC4 LTD4 LTE4
LTs in EBC. The simplicity, rapidity, and robustness of the whole
Exhaled breath condensate (pg/ml) method make it suitable for use as a high-throughput assay in
Occupational asthma 62–101 14–45 43–126
routine clinical biochemical laboratories.
Steroid-resistant asthma 54–78 18–43 48–87
Moderate asthma without therapy 45–83 22–29 54–93
Moderate asthma with therapy 32–58 18–25 38–72
Acknowledgement
Controls 23–37 13–19 34–48

Plasma (pg/ml) This work was financially supported by the EU structural

Occupational asthma 53–108 21–38 42–58
funds – “Operational Programme Prague – Competitiveness” (Grant
Steroid-resistant asthma 54–84 20–33 41–55
Moderate asthma without therapy 54–86 25–32 42–52 CZ.2.16/3.1.00/22197). The authors wish to acknowledge with grat-
Moderate asthma with therapy 52–77 24–28 41–47 itude the financial support by the research project of the Charles
Controls 49–63 23–27 37–45 University in Prague (No. P28/1LF/6).
Urine (pg/mg creatinine)
Occupational asthma 11–19 9–17 25–44
Steroid-resistant asthma 13–21 10–17 23–48
Appendix A. Supplementary data
Moderate asthma without therapy 11–16 11–15 26–39
Moderate asthma with therapy 12–15 10–14 25–34 Supplementary data associated with this article can be found, in
Controls 9–13 9–13 22–31 the online version, at http://dx.doi.org/10.1016/j.jpba.2013.03.026.
 K. Syslová et al. / Journal of Pharmaceutical and Biomedical Analysis 81–82 (2013) 108–117 117

References [20] P. Montuschi, P.J. Barnes, G. Ciabattoni, Measurement of 8-isoprostane in

exhaled breath condensate, Methods Mol. Biol. 594 (2010) 73–84.
[1] M.T. Chuang, J. Raskin, D.J. Krellenstein, A.S. Teirstein, Bronchoscopy in diffuse [21] V. Lucidi, G. Ciabattoni, S. Bella, P.J. Barnes, P. Montuschi, Exhaled 8-isoprostane
lung disease: evaluation by open lung biopsy in nondiagnostic transbronchial and prostaglandin E(2) in patients with stable and unstable cystic fibrosis, Free
lung biops, Ann. Otol. Rhinol. Laryngol. 96 (1987) 654–657. Radic. Biol. Med. 45 (2008) 913–919.
[2] N.N. Jarjour, S.P. Peters, R. Djukanovic, W.J. Calhoun, Investigative use of bron- [22] M. Malerba, P. Montuschi, Non-invasive biomarkers of lung inflammation in
choscopy in asthma, Am. J. Respir. Crit. Care Med. 157 (1998) 692–697. smoking subjects, Curr. Med. Chem. 19 (2012) 187–196.
[3] P.K. Jeffery, A. Laitinen, P. Venge, Biopsy markers of airway inflammation and [23] P. Montuschi, M.L. Peters-Golden, Leukotriene modifiers for asthma treatment,
remodelling, Respir. Med. 94 (2000) S9–S15. Clin. Exp. Allergy 40 (2010) 1732–1741.
[4] H.Y. Reynolds, Use of bronchoalveolar lavage in humans – past necessity and [24] P. Montuschi, Leukotrienes, antileukotrienes and asthma, Mini Rev. Med. Chem.
future imperative, Lung 178 (2000) 271–293. 8 (2008) 647–656.
[5] R.H. Green, Ch.E. Brightling, S. McKenna, B. Hargadon, D. Parker, P. Bradding, A.J. [25] P. Montuschi, Toward a personalized pharmacotherapy of respiratory diseases,
Wardlaw, I.D. Pavord, Asthma exacerbations and sputum eosinophil counts: a Front. Pharmacol. 1 (2010) 131.
randomised controlled trial, Lancet 360 (2002) 1715–1721. [26] P. Montuschi, P.J. Barnes, New perspectives in pharmacological treatment of
[6] O. Holz, J. Kips, H. Magnussen, Update on sputum methodology, Eur. Respir. J. mild persistent asthma, Drug Discov. Today 16 (2011) 1084–1091.
16 (2000) 355–359. [27] P. Čáp, J. Chládek, F. Pehal, M. Malý, V. Petrů, P.J. Barnes, P. Montuschi, Gas chro-
[7] I. Horváth, J. Hunt, P.J. Barnes, Exhaled breath condensate: methodological matography/mass spectrometry analysis of exhaled leukotrienes in asthmatic
recommendations and unresolved questions, Eur. Respir. J. 26 (2005) 523–548. patients, Thorax 59 (2004) 465–470.
[8] A. Motta, D. Paris, D. Melck, G. de Laurentiis, M. Maniscalco, M. Sofia, P. Mon- [28] D. Tsikas, Application of gas chromatography–mass spectrometry and gas
tuschi, Nuclear magnetic resonance-based metabolomics of exhaled breath chromatography–tandem mass spectrometry to assess in vivo synthe-
condensate: methodological aspects, Eur. Respir. J. 39 (2012) 498–500. sis of prostaglandins, thromboxane, leukotrienes, isoprostanes and related
[9] P. Montuschi, D. Paris, D. Melck, V. Lucidi, G. Ciabattoni, V. Raia, C. Calabrese, A. compounds in humans, J. Chromatogr. B: Biomed. Sci. Appl. 717 (1998)
Bush, P.J. Barnes, A. Motta, NMR spectroscopy metabolomic profiling of exhaled 201–245.
breath condensate in patients with stable and unstable cystic fibrosis, Thorax [29] M. Takamoto, T. Yano, T. Shintani, S. Hiraku, A highly sensitive and selective
67 (2012) 222–228. method for the determination of leukotriene B4 in human plasma by nega-
[10] P. Montuschi, C. Mondino, P. Koch, G. Ciabattoni, P.J. Barnes, G. Baviera, Effects tive ion chemical ionization/gas chromatography/tandem mass spectrometry,
of montelukast treatment and withdrawal on fractional exhaled nitric oxide J. Pharm. Biomed. Anal. 13 (1995) 1465–1472.
and lung function in children with asthma, Chest 132 (2007) 1876–1881. [30] Y.J. Bai, X.Y. Gao, J.Q. Lu, H.G.i. Zhang, A LC–MS-based method for quantification
[11] P. Montuschi, N. Mores, A. Trové, C. Mondino, P.J. Barnes, The electronic nose of biomarkers from serum of allergic rats, Molecules 15 (2010) 3356–3365.
in respiratory medicine, Respiration 85 (2013) 72–84. [31] P. Montuschi (Ed.), Analysis of Exhaled Breath Condenstae: Methodological
[12] U. Hatipoglu, I. Rubinstein, New Perspectives in Monitoring Lung Inflammation. Issues in New Perspectives in Monitoring Lung Inflammation, Analysis of
Analysis of Exhaled Breath Condensate, CRC Press, Boca Raton, FL, 2004. Exhaled Breath Condensate, CRC Press, Florida, USA, 2004, ISBN 0-415-32465-3,
[13] Z. Csoma, S.A. Kharitonov, B. Balint, A. Busch, N.M. Wilson, P.J. Barnes, Increased pp. 11–30.
leukotrienes in exhaled breath condensate in childhood asthma, Am. J. Respir. [32] D. Tsikas, A proposal for comparing methods of quantitative analysis of endoge-
Crit. Care Med. 166 (2002) 1345–1349. nous compounds in biological systems by using the relative lower limit of
[14] S. Kazani, E. Israel, Exhaled breath condensates in asthma: diagnostic and ther- quantification (rLLOQ), J. Chromatogr. B 877 (2009) 2244–2251.
apeutic implications, J. Breath Res. 4 (2010) 047001. [33] G. Schlosser, P. Kačer, M. Kuzma, Z. Szilágyi, A. Sorrentino, C. Manzo, R. Pizzano,
[15] M. Kupczyk, S. Lundstrom, B. Dahlén, D. Balgoma, C.E. Wheelock, S.-E. Dahlén, L. Malorni, G. Pocsfalvi, Coupling immunomagnetic separation on magnetic
Lipidimediators in severe asthma, Eur. Respir. Mon. 51 (2011) 218–235. beads with matrix-assisted laser desorption ionization-time of flight mass
[16] M.U. Sahiner, E. Birben, S. Erzurum, C. Sackesen, O. Kalayci, Oxidative stress in spectrometry for detection of staphylococcal enterotoxin B, Appl. Environ.
asthma, World Allergy Organ. J. 4 (2011) 151–158. Microbiol. 73 (2007) 6945–6952.
[17] P. Montuschi, P.J. Barnes, Exhaled leukotrienes and prostaglandins in asthma, [34] K. Syslova, P. Kacer, B. Vilhanova, M. Kuzma, P. Lipovova, Z. Fenclova, J.
J. Allergy Clin. Immunol. 109 (2002) 615–620. Lebedova, D. Pelclova, Determination of cysteinyl leukotrienes in exhaled
[18] P. Montuschi, D. Currò, E. Ragazzoni, P. Preziosi, G. Ciabattoni, Anaphylaxis breath condensate: method combining immunoseparation with LC–ESI-
increases 8-iso-prostaglandin F2␣ release from guinea-pig lung in vitro, Eur. J. MS/MS, J. Chromatogr. B 879 (2011) 2220–2228.
Pharmacol. 365 (1999) 64–95. [35] P. Klusackova, J. Lebedova, P. Kacer, M. Kuzma, M. Brabec, D. Pelclova, Z.
[19] M. Barreto, M.P. Villa, C. Olita, S. Martella, G. Ciabattoni, P. Mon- Fenclova, T. Navratil, Leukotrienes and 8-isoprostane in exhaled breath conden-
tuschi, 8-Isoprostane in exhaled breath condensate and exercise-induced sate in bronchoprovocation tests with occupational allergens, Prostaglandins
bronchoconstriction in asthmatic children and adolescents, Chest 135 (2009) Leukot. Essent. Fatty Acids 78 (2008) 281–292.
66–73.