Enantioseparation and determination of flumequine enantiomers in multiple food matrices with chiral liquid chromatography coupled with tandem mass spectrometry

Article abstract of DOI:10.1002/chir.23125

The present work firstly described the enantioseparation and determination of flumequine enantiomers in milk, yogurt, chicken, beef, egg, and honey samples by chiral liquid chromatography-tandem mass spectrometry. The enantioseparation was performed under reversed-phase conditions on a Chiralpak IC column at 20°C. The effects of chiral stationary phase, mobile phase components, and column temperature on the separation of flumequine enantiomers have been studied in detail. Target compounds were extracted from six different matrices with individual extraction procedure followed by cleanup using Cleanert C18 solid phase extraction cartridge. Good linearity (R²>0.9913) was obtained over the concentration range of 0.125 to 12.5 ng g^-1 for each enantiomer in matrix-matched standard calibration curves. The limits of detection and limits of quantification of two flumequine enantiomers were 0.015-0.024 and 0.045-0.063 ng g^-1, respectively. The average recoveries of the targeted compounds varied from 82.3 to 110.5%, with relative standard deviation less than 11.7%. The method was successfully applied to the determination of flumequine enantiomers in multiple food matrices, providing a reliable method for evaluating the potential risk in animal productions.

Full text of DOI:10.1002/chir.23125

Received: 26 June 2019
DOI: 10.1002/chir.23125
Revised: 20 July 2019
Accepted: 4 August 2019
R E G U L A R A R T I C L E
Enantioseparation and determination of flumequine
enantiomers in multiple food matrices with chiral liquid
chromatography coupled with tandem mass spectrometry
Shuang Li | Beibei Liu | Mengyao Xue | Jia Yu | Xingjie Guo
Department Pharmaceutical Analysis,
Abstract
Institution Shenyang Pharmaceutical
The present work firstly described the enantioseparation and determination of
flumequine enantiomers in milk, yogurt, chicken, beef, egg, and honey sam-
ples by chiral liquid chromatography‐tandem mass spectrometry. The
enantioseparation was performed under reversed‐phase conditions on a
Chiralpak IC column at 20°C. The effects of chiral stationary phase, mobile
phase components, and column temperature on the separation of flumequine
enantiomers have been studied in detail. Target compounds were extracted
from six different matrices with individual extraction procedure followed by
cleanup using Cleanert C18 solid phase extraction cartridge. Good linearity
(R2>0.9913) was obtained over the concentration range of 0.125 to 12.5 ng g‐1
for each enantiomer in matrix‐matched standard calibration curves. The limits
of detection and limits of quantification of two flumequine enantiomers were
0.015‐0.024 and 0.045‐0.063 ng g‐1, respectively. The average recoveries of the
targeted compounds varied from 82.3 to 110.5%, with relative standard devia-
tion less than 11.7%. The method was successfully applied to the determination
of flumequine enantiomers in multiple food matrices, providing a reliable
method for evaluating the potential risk in animal productions.
University, Shenyang, P. R. China
Correspondence
Xingjie Guo and Jia Yu, Department
Pharmaceutical Analysis, Institution
Shenyang Pharmaceutical University,
Address 1 No. 103, Wenhua Road Shenhe
District 110016, Shenyang, Liaoning
Province, P. R. China.
Email: syykdx410@163.com;
yujia_yc@163.com
KEYWORDS
chiral separation, enantiomeric determination, flumequine, food matrices, LC‐MS/MS
1 | INTRODUCTION
to the growth of antibiotic‐resistant bacteria, allergic
reactions, or even toxic effects. Flumequine, a first‐
Fluoroquinolones (FQs) are synthetic antibiotics widely
used in veterinary medicine with the purposes of
inhibiting the growth of microorganisms, preventing or
treating infections, and promoting growth when used at
subtherapeutic doses.1 In recent years, the public concern
about the presence of antibiotic residues in food‐
producing animals has increased because they can lead
generation FQ, can selectively inhibit type II topoisomer-
ase and bacterial DNA gyrase, which is essential for
bacterial DNA replication and transcription in the
process of cell growth and division.2,3 Owing to the high
antimicrobial activity against
a
wide range of
Gram‐negative and Gram‐positive pathogens, flumequine
has become an indispensable part of the treatment of
infections in food‐producing animals such as cattle,
turkey, pig, and poultry. Usually, flumequine is adminis-
tered orally to food‐producing animals and distributed
Abbreviations:
CSP,
chiral
stationary
phase;
Na2EDTA,
acidethylenediamine tetraacetic acid disodium salt; TCA, trichloroacetic
Chirality. 2019;1–11.
wileyonlinelibrary.com/journal/chir
© 2019 Wiley Periodicals, Inc.
1

2
LI ET AL.
widely in the tissue. Research studies have shown that
high doses of flumequine can distort the embryonic
development in rats.4 There is no report indicating that
low doses of flumequine are harmful; however, the
widespread use of flumequine inevitably leads to the
significant and harmful residues in animal edible tissues
and later sold as food products, such as meat, milk, and
egg. Thus, concerning the amount of flumequine residue
found within food and food safety, it is necessary to
quantify and confirm residues of flumequine in animal
productions to protect the consumer.
determination of chiral drugs in various matrices. Never-
theless,
a
few papers have reported on the
enantioselective quantification of flumequine in different
matrices, including in rat plasma, sediment, and different
water samples.22-25 But, to our best knowledge, no
literature has been published on enantioselective deter-
mination of flumequine in multiple food matrices by
LC‐MS/MS.
Therefore, this study aimed to develop a new robust
method for the enantioselective determination of
flumequine in multiple matrices (milk, yogurt, honey,
Stereochemistry of chiral drugs has gotten widespread
attention because enantiomers of drugs always tend to
display quite different pharmacological and toxicological
activities. Many chiral drugs available for clinical use
have at least one chiral centre in chemical structure,
and most are used as racemates. In some cases, the
bioactivity is only presented in one of the enantiomers,
while the other enantiomer has no effect. Furthermore,
degradation products may also be enantioselective, which
may exhibit potential risk. Flumequine contains one
chiral centre and exists as a mixture of two enantiomers
(shown in Figure 1). Study has suggested that the
enantiomers of flumequine exhibit significant differences
in the antibacterial activity.5 Therefore, the sensitive and
reliable resolution and determination method of
flumequine enantiomers should be developed to
control the enantiomer occurrence and quantify the
contaminants in animal productions and accurately
evaluate the potential risk.
According to previous papers, a variety of analytical
methods for the racemic determination of flumequine in
animal productions (including milk, honey, egg, fish,
shrimp, chicken, pork, and beef) have been published
using capillary electrophoresis (CE),6,7 CE tandem mass
spectrometry (CE‐MS/MS),8,9 liquid chromatography‐
ultraviolet detection (LC‐UV),10-15 and liquid chromatog-
raphy tandem mass spectrometry (LC‐MS/MS).16-21
LC‐UV and LC‐MS have been widely utilized for
quantification analysis of flumequine. For example, Zhao
et al10 developed a method for 10 quinolones in swine,
chicken, and shrimp muscle by LC‐UV. The trace
analysis of flumequine and several other quinolones
in muscle and egg has been investigated using LC‐MS.16
In the present, there is a trend on enantioselective
egg, chicken, and beef) using
a
reversed‐phase
HPLC‐MS/MS with the Chiralpak IC column. The effects
of chiral stationary phase (CSP), mobile phase
composition, and the temperature on the separation and
retention of flumequine enantiomers have been
investigated in detail. To date, this paper is the first report
on the enantioselective analysis of flumequine in multiple
food matrices using chiral LC‐MS/MS. The method
developed was validated by its application to the analysis
of flumequine authentic samples. The developed method
was demonstrated to be sensitive, selective, and reliable
for the determination of flumequine enantiomers in
multiple food matrices, which could facilitate further
research in this area.
2 | MATERIALS AND METHODS
2.1 | Chemical and reagents
The racemic flumequine (purity>98.0%) was supplied
from the National Institute for Food and Drug Control
(Beijing, China). MS‐grade ammonium acetate, acetic acid,
formic acid, methanol, and acetonitrile were purchased
from Sigma‐Aldrich (Beijing, China). Analytical grade tri-
chloroacetic (TCA) acid (purity≥99.0%), ethylenediamine
tetraacetic acid disodium salt (purity≥98.0%), NaOH
(purity≥96.0%), n‐hexane, disodium hydrogen phosphate
dihydrate (purity≥99.0%), citric acid monohydrate
(purity≥98.0%), and potassium dihydrogen phosphate
(purity≥99.5%) were provided by Yuwang Technology
(Shandong, China). Ultrapure water was used throughout
our study. The Cleanert C18 cartridge (500 mg, 3 mL),
Cleanert PAX cartridge (60 mg, 3 mL), and Cleanert
PEP‐2 (60 mg, 3 mL) cartridge were purchased from Agela
Technologies (Tianjin, China). Standard stock solution of
flumequine (1 mg mL‐1) was prepared in pure methanol.
Working standard solutions were prepared daily by
serial dilution in methanol from the standard stock
solutions. All solutions were stored at 4°C prior to the
analysis.
FIGURE 1 Chemical structures of flumequine enantiomers

LI ET AL.
3
further clean‐up, which resembled the purification
described for milk and yogurt products.
2.2 | Sample preparation
The multiple blank samples were purchased from the
local market in Benxi (China). Preliminary analysis
showed that these samples were analyte‐free and thus
were used for the method validation. These samples were
homogenized and spiked with minimum volumes of solu-
tions containing the necessary concentrations of the
flumequine assayed. The samples were stored at 4°C for
24 hours, thus allowing the incorporation of the
flumequine on the matrix.
2.5 | Egg samples
Aliquots of 2 g of whole egg were placed into a 50 mL of
polypropylene centrifuge tube, and 8 mL of 1% acetic acid
in ethanol solution was added. The sample was shaken
for 5 minutes, and 500 μL of acetonitrile and 4 mL of
1% acetic acid in ethanol solution were added. Before
centrifugation step, the suspension was vortexed and
put into ultrasonic bath for 15 minutes; next, the mixture
was centrifuged at 4000 rpm for 15 minutes. The upper
layer was transferred into a new centrifuge tube, and
the extraction was repeated with 4 mL of 1% acetic acid
in ethanol solution. The entire supernatant was evapo-
rated under nitrogen stream at 45°C; the residue was
reconstituted with 1 mL of methanol. After mixing on a
vortexer, 2 mL of n‐hexane was added. After shaking
and sonication, the upper layer was discarded. And the
suspension was treated for further clean‐up, which was
analogous to one described for milk and yogurt products.
2.3 | Milk and yogurt samples
Aliquots of 2 g of milk (or yogurt) were transferred to 50
mL of polypropylene centrifuge tubes, and 2.5 mL of TCA
in acetonitrile (25%, w/v) was added. Then, 5‐mL
acidethylenediamine tetraacetic acid disodium salt
(Na2EDTA) solution (0.1 mM) was added to the tube
and pH was adjusted to 4 with NaOH solution (5 M).
The mixture was vortexed and left to settle in the dark
for 15 min. The solution was then ultrasonic bath for 15
minutes and directly centrifuged at 4000 rpm for
15 minutes. The supernatant was transferred to a new
glass tube, and the precipitate was rinsed and centrifuged
again with 2.5 mL of ultrapure water. The entire superna-
tant was used for further clean‐up. The clean‐up was
carried out using a Cleanert C18 cartridge (500 mg, 3
mL). The column was conditioned with 3 mL of metha-
nol and 3 mL of water. After application of the extract,
the cartridge was rinsed with 3 mL of water and vacuum
dried for 2 minutes. The flumequine was eluted from the
column with 3 mL 1% formic acid in methanol. The
eluate was evaporated to dryness at 45°C under a stream
of nitrogen. The dry residue was dissolved in 200‐μL
mobile phase and filtered through a 0.22‐μm syringe filter
before LC‐MS/MS analysis.
2.6 | Honey samples
An aliquot of 2 g homogenized honey was weighed and
placed into 50 mL of polypropylene centrifuge tube.
Then, 10 mL of 0.2 M ammonium acetate buffer (pH 7)
was added. Subsequently, 3 mL of 5% formic acid in ace-
tonitrile was added and the mixture was vortexed until
the honey was completely diluted. After sonication (15
min) and centrifugation (15 min, 4000 rpm), the entire
sample was used for further clean‐up. The purification
was similar to the one described for milk and yogurt
products.
2.7 | LC‐MS/MS
2.4 | Chicken and beef samples
The chromatographic analysis of the flumequine enantio-
mer was conducted using a Waters AcquityTM UPLC
system (Waters Corp., Milford, MA, USA), coupled to a
Micromass QuattromicroTM API mass spectrometer
(Waters, MA, USA) with an electrospray ionization
(ESI) interface. Six CSPs, including four cellulose‐based
columns (ChiralpakIB and Chiralpak IC, 250 mm × 4.6
mm, i.d. 5 μm; Chiralcel OD‐RH and Chiralcel OJ‐RH,
150 mm × 4.6 mm, i.d. 5 μm) and two amylose‐based chi-
ral columns (Chiralpak IA and Chiralpak ID, 250 mm ×
4.6 mm, i.d. 5 μm), have been evaluated for the chiral
separation of flumequine. All the seven chiral columns
were provided by Daicel Chiral Technologies (China)
Aliquots of 2 g of thawed and minced muscle tissues were
weighed and placed in a 50 mL of polypropylene
centrifuge tube. Then, 15 mL of 0.2‐M ammonium acetate
buffer (pH 7.4) was added. Subsequently, 2 mL of acetoni-
trile was added. The sample was allowed to stand for
15 minutes at room temperature and then sonicated (15
min) before centrifugation for 15 minutes at 4000 rpm.
The supernatant was transferred to another 50 mL of
centrifuge tube. About 2 mL of n‐hexane was added to
each tube, followed by 10 seconds of vortex mixing and
5 minutes of centrifugation at 4000 rpm. The top hexane
layer was discarded. The remaining solution was used for

4
LI ET AL.
Co., Ltd (Shanghai, China). The optimized chromato-
graphic condition for the enantioseparation of the
flumequine on Chiralpak IC column was achieved by
running the reversed phase condition consisting of 20%
aqueous phase (0.1% formic acid and 5 mM ammonium
formate in water) and 80% organic phase (0.1% formic
acid and 5 mM ammonium formate in acetonitrile).
Separation was performed under isocratic mode with a
flow rate of 0.6 mL min‐1. Column temperature and
autosampler temperature were set at 20 and 4°C, respec-
tively. The injection volume was 10 μL.
All data collected were processed using MassLynx 4.1
software (Waters Corp., Milford, MA, USA). Optimized
collision energy and other parameters were obtained based
on the direct injection of individual standard solutions at
the concentration of 100 ng mL‐1. The ionization source
conditions were as follows: capillary voltage of 3.0 kV
and source temperature of 150°C. Nitrogen gas was used
as the desolvation gas and set to a flow rate of 1000 L
hour−1 with a temperature of 500°C. The collision gas by
ultrahigh‐purity argon was held at 0.12 mL min−1. Detec-
tion was performed in multiple reaction monitoring
(MRM) mode and in the positive ionization mode. Two
MRM transitions were monitored for flumequine. The
most abundant product ion was selected for quantification,
while the other transition was used for confirmation.
Typical conditions were as follows: m/z 262 > 243.88 was
used for quantification and m/z 262 > 243.88 and m/z
262 > 201.88 were used for confirmation when the collision
energy was set at 28 and 12 eV, respectively. The optimized
cone voltage of flumequine was 36 V.
Chiralcel OJ‐RH) and two amylose‐based chiral columns
(Chiralpak IA and Chiralpak ID) was tested using a vari-
ety of reversed‐phase mobile phase combinations. The
initial column screening was conducted under the same
mobile condition, with a organic solvent of 80% methanol
or 60% acetonitrile at a flow rate of 0.6 mL min‐1. If
enantioseparation was unsatisfactory, optimization steps
by changing the mobile phase compositions were
considered. And mobile phase additives such as
ammonium acetate and formic acid were also tested for
good resolution. From the results obtained from the
selected six columns, Chiralpak IA (amylose tris‐(3,
5‐dimethylphenylcarbamate)), Chiralpak IB (cellulose
tris‐(3, 5‐dimethylphenylcarbamate)), Chiralpak ID
(amylose tris‐(3‐chlorophenylcarbamate)), and Chiralcel
OD‐RH (cellulose tris‐(3, 5‐dimethhylphenylcarbamate))
could be capable for partial separation of the enantio-
mers. Only Chiralpak IC and Chiralcel OJ‐RH column
were effective for baseline separation of the flumequine
enantiomers. Chiralpak IC is an immobilized column,
which is produced by immobilization of the cellulose
tris‐ (3,5‐dichlorophenylcarbamate) on silica gel surface.
While, Chiralcel OJ‐RH is a coated column, of which
the chiral selector is cellulose tris‐(4‐methylbenzoate).
Though the baseline separation could be achieved on
the two columns, Chiralpak IC column was preferable
because of the higher enantioselectivity and the shorter
retention times.
It can be seen that flumequines contain carboxyl,
carbonyl, phenyl moiety, fluorine atom, and nitrogen
atom. There might be hydrogen bonding, π‐π interaction,
and dipole‐dipole interaction between flumequine and
CSP during the chiral resolution process. For Chiralpak
IC, a chloro‐substituted CSP, the higher enantioselectivity
abilities may be due to the existence of chlorine atoms on
the phenyl moiety. Owing to the electron‐withdrawing
inductive effect, chlorine atoms might improve the acid
strength of the NH group (carbamate group) and contrib-
ute to the formation of hydrogen bonding with
flumequine.31 Besides, the weak π‐π interaction may exist
between phenyl moiety of CSP and the phenyl ring of
flumequine.
3 | RESULTS AND DISCUSSION
3.1 | Method development
3.1.1 | Effect of CSPs
According to previous studies, enantioselective separation
of chiral drugs is extensively carried out using a CSP.
Owing to the poor ionization and potential hazard, the
normal‐phase LC is generally considered to be incompat-
ible to MS. Hence, the reversed phase mode is applied to
determination. The selection of chiral column with differ-
ent stationary phases is a key factor to achieve efficient
enantioseparation. Among the various CSPs, the
polysaccharide‐based CSPs are extensively used, which
show good chiral recognition ability toward many chiral
compounds.26-30
3.1.2 | Effect of mobile phase composition
Composition of mobile phase plays an important role in
enantiomeric separation in terms of retention behaviours,
elution, and resolution.32 Acetonitrile and methanol as
the commonly used organic modifiers were tested in the
preliminary experiments. It was found that the baseline
separation of flumequine could not be achieved, and the
column pressure greatly increased when methanol was
In the preliminary experiments, the enantioseparation
of flumequine on four cellulose‐based columns
(Chiralpak IB and Chiralpak IC, Chiralcel OD‐RH and

LI ET AL.
5
used. Whereas, a satisfactory separation was obtained
when using acetonitrile as an organic modifier. As
increasing the percentage of acetonitrile, shorten reten-
tion times, and decreased resolution was observed, so
80% of acetonitrile was finally used. Generally, small
quantities of additives could remarkably improve peak
shapes and enhance MS response and resolutions. Differ-
ent ammonium acetate concentrations (2, 5, and 10 mM)
were investigated for this purpose. It was found that the
concentration of ammonium acetate exhibited no signifi-
cant effect on the enantioselective separation, whereas
the MS response was the highest when the buffer concen-
tration was 5 mM. Consequently, 5‐mM concentration of
ammonium acetate was selected. In addition, 0.1% formic
acid was added in mobile phases that could enhance the
[M+H]+ responses and improve sensitivity for the target
analytes. And lower flow rate resulted in better resolution
but longer analysis time. For the aim of good resolution
and reasonable analysis time, the flow rate was set at
0.6 mL min‐1. In consequence, the final mobile phase
was composed of 20% aqueous phase (0.1% formic acid
and 5mM ammonium formate in water) and 80% organic
phase (0.1% formic acid and 5mM ammonium formate in
acetonitrile) delivered with a flow rate of 0.6 mL min‐1.
R is the universal gas constant (8.314 J·mol‐1K‐1). The
ΔΔH° and ΔΔS° are on behalf of the differences of ΔH°
and ΔS° between the two enantiomers. Equation 1 pre-
dict that if plots of ln k versus 1/T were linear, the slope
and intercept were −ΔH°/R and ΔS°/R + lnΦ. Due to
the fact that the value of Φ is unknown, ΔS*/R will take
place of ΔS°/R + lnΦ. For a linear plot of ln α versus
1/T, the slope and intercept are ‐ΔΔH°/R and ΔΔS°/R.
In the present work, the effect of column temperature
on separation of flumequine enantiomers was carried out
on Chiralpak IC with step increasing of temperature from
20 to 40°C with 5°C increments using the optimized
mobile phase. With the increase of column temperature,
the capacity factor and separation factor decreased.
Hence, 20°C was chosen as the optimal temperature.
The calculated thermodynamic parameters are listed in
Table 1. The plots of ln k and ln α versus 1/T were linear
(linear correlation coefficient R2>0.99), suggesting
that the retention mechanism was independent of
temperature in the studied range and the enantioselective
interaction remained unchanged. The negative ΔH°
values revealed that the distribution of the two enantio-
mers between the mobile phase and the stationary phase
was exothermic. Moreover, both ΔΔH° and ΔΔS° were
negative showing that separation of flumequine was
enthalpy‐driven.34,36
3.1.3 | Effect of column temperature
Column temperature is regarded as a significant parameter
with reference to chiral separations.33 The thermodynamic
parameters could be calculated according to the capacity
factor (k), separation factor (α), and resolution (Rs) by
classical Van't Hoff equations,34-36 which assumes that
analyte retention is only due to enantioselective interac-
tions with the stationary phase and does not distinguish
between chiral and nonchiral interactions.37 The classical
Van't Hoff equations are exhibited as follows:
3.1.4 | Identification of enantiomeric
elution orders
Commercially available enantiopure standards of
flumequine were not available, and their elution order
on the Chiralpak IC column, which was utilized for chiral
separation in our work, has not been reported. The elu-
tion order of the flumequine enantiomers was deter-
mined according to previous literature, that
demonstrated S‐(‐)‐ flumequine was eluted first from the
Chiralcel OJ‐RH column.22 Firstly, we achieved the sepa-
ration of flumequine under the same condition as
depicted by Zhao et al.22 The first eluting peak from
Chiralcel OJ‐RH column was collected, then the solvent
was evaporated. The residue was dissolved in methanol
and injected into the Chiralpak IC column by optimum
condition. By comparing the retention time of single
enantiomer with the racemic flumequine, the enantiomer
ΔH° ΔS°
ΔH° ΔS*
lnk ¼ −
þ
þ lnΦ ¼ −
þ
;
(1)
(2)
RT
R
RT
R
ΔΔH° ΔΔS°
lnα ¼ −
þ
:
RT
R
In the classical Van't Hoff equations, ΔH° and ΔS°
represent the enthalpy and entropy changes of the ana-
lyte from the mobile phase to the stationary phases, Φ is
the phase ratio, T is the absolute temperature (K), and
TABLE 1 The thermodynamic parameters of flumequine enantiomers.
Analyte
ln k
R2
ΔH°, kJ/mol
ΔS*, J/K/mol
ΔΔH°, kJ/mol
−1.12
ΔΔS°, J/K/mol
−2.10
S‐enantiomer
R‐enantiomer
lnk1=954.83/T−2.2939
lnk2=1080/T−2.546
0.9950
0.9937
−7.86
−8.98
−19.07
−21.17

6
LI ET AL.
elution order was established as S‐(‐)‐flumequine
followed by its R‐(+)‐enantiomer.
3.1.8 | Extraction of egg samples
According to previous study, three extraction solutions
were preliminarily investigated, including acetonitrile,
TCA in acetonitrile (10%,w/v), and ammonium hydroxide
in acetonitrile (10%,v/v). But the recovery of all these
solutions was poor and the recovery provided by alkalized
acetonitrile approached zero indicating that alkalized
solvent was not efficient for flumequine extraction from
egg samples. Therefore, the extraction produce had to
be modified by adding acid in solvent to improve the
recovery. Finally, the sample preparation produce was
adopted 1% acetic acid in ethanol, providing a satisfactory
recovery for egg samples43 (Figure S1).
3.1.5 | Optimization of extraction
Food matrices (milk, yogurt, chicken, beef, honey, and
egg) are notably complex consisting of various endoge-
nous substance. For the purpose of developing a sensitive
method to analyse flumequine at trace concentration
level in food matrices, it is of great importance to isolate
the analytes from matrices as much as possible. Since
different food matrices were quite distinct and complex,
and in order to improve the extraction efficiency, individ-
ual extraction methods were applied to different food
matrices.
3.1.9 | Extraction of honey samples
Honey is a weakly acidic complex matrix that contains
sugars, proteins, and pigments. On account of the
published reports, the 0.03‐mM NaH2PO4 buffer, 0.1‐M
Na2EDTA solution, and 0.2‐M ammonium acetate buffer
were investigated as extraction solvent. Among the three
extraction solutions, 0.2‐M ammonium acetate buffer
showed best extraction efficiency and was chosen as
extraction solvent. (Figure S1)
3.1.6 | Extraction of milk and yogurt
samples
Dairy products contain abundant fats and proteins.
Therefore, it is necessary to precipitate the protein and
remove the fat to enhance the sensitivity. Based on the
published literatures,12,38,39 TCA in acetonitrile (25%,
w/v), TCA in methanol (25%, w/v) and Mcllvaine buffer
were evaluated as the extraction solvents. As indicated in
Figure S1, the extraction efficiency of TCA in acetonitrile
(25%, w/v) was better than others. In some cases, the
addition of EDTA could form the complexes with
the analyst, which was beneficial for extraction. Results
demonstrated that the recovery was obviously improved
when the EDTA was added. The pH of sample also be
tested, and the pH 4 provided higher recovery.
3.1.10 | Optimization of purification
Owing to the complexity of the food matrices, following
the extraction process, the purification step is essential
before the LC‐MS/MS analysis. Previously, SPE has been
widely applied for preconcentration and purification of
flumequine from different food matrices.10,16,19 Based on
the properties of the flumequine, three commonly used
cartridges, namely, Cleanert C18 (500 mg, 3 mL),
Cleanert PEP‐2 (60 mg, 3 mL), and Cleanert PAX (60
mg, 3 mL) have been selected to investigate the extraction
recovery. Results are revealed in Figure 2. It was found
3.1.7 | Extraction of chicken and beef
samples
To date, previous research studies showed many extrac-
tion solvents, which were effective in extracting meat
matrix,2,40-42 including acid‐acetonitrile, ammonium
acetate buffer, phosphate buffer solution, and dichloro-
methane. In this work, we investigated acid‐acetonitrile,
0.2 M ammonium acetate buffer, and 0.01 M phosphate
buffer as extraction solvents. It was found that the recov-
ery of 0.2 M ammonium acetate buffer was higher than
others (Figure S1). Based on this, 0.2‐M ammonium
acetate buffer was selected as the optimum extraction
procedure for chicken and beef samples.
FIGURE 2 Effect of the solid‐phase extraction cartridge on the
recoveries of the flumequine

LI ET AL.
7
that poor recoveries were provided by Cleanert PAX.
The recoveries of Cleanert C18 and Cleanert PEP‐2 were
very similar. Hence, considering satisfactory recoveries
and lower cost, a C18 cartridge was chosen in current
study. Because of the acidic character of flumequine,
acidic elution solvent helps to improve the extraction
recoveries. So 1% formic acid in methanol, 1% formic
acid in acetonitrile,and 1% formic acid in ethanol were
studied as elution solvent. The results illustrated in
Figure 3 indicated that methanol provided the highest
recovery, which may be due to the reason that
methanol can form strong hydrogen bonding with
analyte and then elute it from cartridge. Thus, 1%
formic acid in methanol was selected as optimum elu-
tion solvent.
linearity, limit of detection (LOD), and limit of
quantification (LOQ), accuracy, precision, matrix effect
(ME), and stability. Blank samples (free of the
flumequine studied) were used for the method
validation.
Ten blank samples (milk, yogurt, chicken, beef, eggs,
and honey) were extracted and analysed to evaluate the
specificity of the method, and it was shown that no inter-
ference was observed at the retention of target analytes.
The typical chromatograms of the blank sample and
spiked sample are acquired and compared in Figure 4.
It can be concluded that validated method is specific for
the target analytes.
The linearity of the method was performed using
matrix‐matched calibration curves by spiking standards
before extraction at six concentration levels ranging
from 0.125 to 12.5 ng g−1. Each level was prepared in
triplicate. The spiked samples were pretreated fol-
lowing the above‐mentioned procedures. Calibration
curves were generated using linear regression analysis,
and obtained linearity was accepted when the
correlation coefficients (R2) was higher than 0.99. The
3.1.11 | Method validation
The performance of the developed method was vali-
dated according to a conventional validation procedure
that included the following parameters: specificity,
FIGURE 3 Effect of the elution solvent
on the recoveries of the flumequine
FIGURE 4 Typical enantioselective liquid chromatography tandem mass spectrometry multiple reaction monitoring chromatograms of A,
blank milk sample and B, milk spiked with 0.25 ng g‐1

8
LI ET AL.
data are listed in Table 2. Satisfactory linearities
were obtained (R2 >0.9913) in the range of 0.125 to
12.5 ng g‐1.
The LOD of the method is defined as the lowest con-
centration that produced signal‐to‐noise (S/N) ratios of
3, and the LOQ is defined based on the S/N of 10. The
results showed that the LODs and LOQs for flumequine
ranged from 0.015 to 0.024 ng g‐1 and 0.045 to 0.063 ng
g‐1, respectively (Table 2).
The recovery assay was performed to investigate the
accuracy and precision (expressed as relative standard
deviation, RSD) of the method by spiking the analytes
into various matrices (before extraction). Three replicates
of the spiked samples at three different levels (0.25, 0.5,
and 5 ng g‐1 for each enantiomer) were prepared on
three successive days. All samples were extracted and
cleaned up according to the above‐mentioned proce-
dures. Method recovery was determined by comparing
the back‐calculated concentration with the nominal
value. Intraday precision was obtained by analyzing
the three levels of the spiked samples with three repli-
cates in the same day on the LC‐MS/MS, while interday
precision was obtained by repeating this experiments on
three consecutive days. The results are summarized in
Table 3. The relative recoveries S‐enantiomer of
flumequine varied from 83.7 to 110.5% in all studied
matrices, intraday RSD, and interday RSD were in the
range of 1.0 to 11.7% and 2.2 to 11.0%. The relative
recoveries of R‐enantiomer of flumequine varied from
82.3 to 108.3% in all studied matrices. And intraday
RSD and interday RSD were in the range of 0.6 to
10.2% and 1.4 to 10.2%.
TABLE 2 Correlation coefficient (R2), linearity range, matrix effect, LOQ, and LOD of flumequine enantiomers in food matrices
Matrix
Analyte
Linearity Range, ng g‐1
R2
LOD, ng g‐1
LOQ, ng g‐1
Matrix Effects, %
Milk
S‐enantiomer
R‐enantiomer
0.125–12.5
0.125–12.5
0.9924
0.9927
0.017
0.015
0.045
0.049
132.4
128.9
Yogurt
Chicken
Beef
S‐enantiomer
R‐enantiomer
0.125–12.5
0.125–12.5
0.9980
0.9975
0.018
0.021
0.048
0.055
131.6
130.7
S‐enantiomer
R‐enantiomer
0.125–12.5
0.125–12.5
0.9915
0.9913
0.015
0.017
0.050
0.046
126.8
129.0
S‐enantiomer
R‐enantiomer
0.125–12.5
0.125–12.5
0.9984
0.9978
0.020
0.019
0.056
0.050
131.4
127.3
Egg
S‐enantiomer
R‐enantiomer
0.125–12.5
0.125–12.5
0.9960
0.9950
0.020
0.018
0.063
0.059
139.1
143.7
Honey
S‐enantiomer
R‐enantiomer
0.125–12.5
0.125–12.5
0.9926
0.9937
0.021
0.024
0.052
0.055
95.4
97.8
Abbreviations: LOD: limit of detection; LOQ: limit of quantification.
TABLE 3 Recovery and RSDs of flumequine enantiomers in food matrices (n=3)
Recovery, % Intraday RSD, %
0.25 ng g‐1 0.5 ng g‐1 5 ng g‐1 0.25 ng g‐1 0.5 ng g‐1 5 ng g‐1 0.25 ng g‐1 0.5 ng g‐1 5 ng g‐1
Interday RSD, %
Matrix
Analyte
Milk
S‐enantiomer 103.2
R‐enantiomer 106.7
96.3
92.5
83.7
86.2
1.9
4.2
5.6
4.4
3.5
1.0
3.4
5.6
5.9
8.4
3.0
5.5
Yogurt
S‐enantiomer
R‐enantiomer
87.4
86.2
88.6
87.3
91.8
85.8
4.6
4.8
11.7
6.9
4.3
3.4
4.3
8.7
6.5
5.2
3.8
4.9
Chicken S‐enantiomer
R‐enantiomer
88.9
87.1
100.8
101.1
83.9
82.3
6.3
5.2
3.5
5.0
2.1
2.4
6.1
9.5
3.3
7.8
9.5
3.7
Beef
S‐enantiomer 110.5
R‐enantiomer 108.3
97.5
96.4
100.4
101.5
6.4
8.9
6.2
4.3
1.2
0.8
2.2
7.3
11.0
8.1
6.4
4.9
Egg
S‐enantiomer 102.6
R‐enantiomer 101.8
85.7
88.4
96.7
98.3
1.6
5.0
1.6
5.7
2.1
3.1
4.3
10.2
5.0
2.8
6.7
4.5
Honey
S‐enantiomer
R‐enantiomer
92.1
97.5
102.4
106.2
94.7
94.2
1.0
0.6
8.1
10.2
9.2
9.9
5.3
1.4
6.6
9.8
3.3
5.2
Abbreviation: RSDs: relative standard deviation.

LI ET AL.
9
It is well known that ME is a widely existing phenom-
sample matrix, individual extraction methods were
applied to different food matrices. The sample prepara-
tion procedure, including extraction and purification,
was assessed for higher extraction efficiencies. The
proposed method demonstrated good linearity and accu-
racy, and the LOD was at low ng g‐1 levels. In conclusion,
this method is suitable and reliable for monitoring the
flumequine enantiomers occurrence and quantifying
the residues in animal productions and evaluating the
potential risk.
enon when ESI is used, which is attributed to the
ionization competition from other matrix constituents.
Both signal enhancement and signal suppression can
influence the accuracy of and precision of the method.
Hence, it is essential to evaluate the ME. In the current
study, the ME was investigated for flumequine in
different matrices by comparing peak areas obtained in
postextraction spiked samples with that of corresponding
standard solution. If the ME equals to 100%, no ME is
present. In contract, the ME >100 % indicates an
ionization enhancement and the ME <100 % indicates
an ionization suppression. Table 3 gave the values of
ME for flumequine in six matrices. As seen, the ME
observed for honey was negligible ranging from 95.4 to
97.8%. However, in other five matrices, a quite strong
signal enhancement was observed with ME from 126.8
to 143.7%, due to the high complexity of the sample
matrix. When all these problems were considered
together, we adopted the matrix‐matched standard
curves, which could eliminate the effect of the
matrix and could meet the requirement of quantitative
analysis.
The stability of the enantiomers of flumequine was
evaluated in solvent and in matrix. The stock solutions
prepared in methanol were stored at 4°C for 3 months.
The stability of the stock solutions was tested monthly
by injection of a newly prepared working solution. Spiked
blank samples were also stored at 4°C, and then analysed
after 3, 7 and 14 days. The responses did not change
obviously, revealing good stability of flumequine in both
solvent and matrix.
ORCID
Xingjie Guo https://orcid.org/0000-0003-4722-0193
REFERENCES
1. Rodríguez‐Díaz RC, Fernández‐Romero JM, Aguilar‐Caballos
MP, Gómez‐Hens A. Chromatographic determination of
flumequine in food samples by post‐column derivatisation with
terbium (III). Anal Chim Acta. 2006;578(2):220‐226.
2. Zhao S, Jiang H, Li X, Mi T, Li C, Shen J. Simultaneous determi-
nation of trace levels of 10 quinolones in swine, chicken, and
shrimp muscle tissues using HPLC with programmable fluores-
cence detection. J Agric Food Chem. 2007;55:3829‐3834.
3. Schröder U, Machetzki A. Determination of flumequine,
nalidixic acid and oxolinic acid in shrimps by high‐
performance liquid chromatography with fluorescence detec-
tion. Eur Food Res Technol. 2007;225(5‐6):627‐635.
4. Huang Q. Teratogenicity of Flumequine to Rats. Chin J Vet
Drug. 1999;33:9‐11.
5. Balint J, Egri G, Fogassy E, et al. Synthesis, absolute configura-
tion and intermediates of 9‐fluoro‐6, 7‐dihydro‐5‐methyl‐1‐oxo‐
1H, 5H‐benzo [i, j] quinolizine‐2‐carboxylic acid (flumequine).
Tetrahedron: Asymmetry. 1999;10(6):1079‐1087.
3.1.12 | Application to real samples
6. Barrón D, Jiménez‐Lozano E, Bailac S, Barbosa J. Simultaneous
determination of flumequine and oxolinic acid in chicken tis-
sues by solid phase extraction and capillary electrophoresis.
Anal Chim Acta. 2003;477(1):21‐27.
For purpose of evaluating the potential residues of
flumequine in food matrices, the proposed method was
successfully applied to analysis of real samples. Six kinds
of matrices, a total of 36 real samples obtained from dif-
ferent local market (Benxi, China), were prepared and
analysed in accordance with above‐mentioned procedure.
Results showed that no flumequine was detected in any
the purchased food.
7. Ibarra IS, Rodriguez JA, Páez‐Hernández ME, Santos EM,
Miranda JM. Determination of quinolones in milk samples
using a combination of magnetic solid‐phase extraction and cap-
illary electrophoresis. Electrophoresis. 2012;33(13):2041‐2048.
8. Lara FJ, García‐Campaña AM, Alés‐Barrero F, Bosque‐Sendra
JM, García‐Ayuso LE. Multiresidue Method for the Determina-
tion of Quinolone Antibiotics in Bovine Raw Milk by Capillary
Electrophoresis‐Tandem Mass Spectrometry. Anal Chem.
2006;78:7665‐7673.
4 | CONCLUSION
9. Ruiz‐Viceo JA, Rosales‐Conrado N, Guillén‐Casla V, Pérez‐
Arribas LV, León‐González ME, Polo‐Díez LM. Fluoroquino-
lone antibiotic determination in bovine milk using capillary
liquid chromatography with diode array and mass spectrometry
detection. J Food Compos Anal. 2012;28(2):99‐106.
In this study, a novel and reliable enantioselective
method using chiral LC‐MS/MS for the determination of
flumequine enantiomers in six food matrices has been
successfully established and validated. A series of CSPs
were evaluated, and the chromatographic conditions
were also optimized. Duo to the high complexity of the
10. Zhao S‐J, Jiang H‐Y, Li X‐L, Mi T‐J, Li C, Shen J‐Z. Simulta-
neous Determination of Trace Levels of 10 Quinolones in
Swine, Chicken, and Shrimp Muscle Tissues Using HPLC with

10
LI ET AL.
Programmable Fluorescence Detection. J Agric Food Chem.
2007;55(10):3829‐3834.
Degradation Behavior of Flumequine Enantiomers in Maricul-
ture Pond Water. Chirality. 2016;28:646‐655.
24. Castrignanò E, Kannan A, Feil E, Kasprzyk‐Hordern B.
Enantioselective fractionation of fluoroquinolones in the aqueous
environment using chiral liquid chromatography coupled with
tandem mass spectrometry. Chemosphere. 2018;206:376‐386.
11. Zheng M‐M, Gong R, Zhao X, Feng Y. Selective sample pretreat-
ment by molecularly imprinted polymer monolith for the
analysis of fluoroquinolones from milk samples. J Chromatogr
A. 2010;1217(14):2075‐2081.
25. Xue M‐Y, Qin Y‐C, Gu X, et al. Determination of flumequine
enantiomers and 7‐hydroxy flumequine in water and sediment
by chiral HPLC coupled with hybrid quadrupole‐time‐of‐flight
mass spectrometery. Sci Rep. 2018;8(1):7582.
12. Herrera‐Herrera AV, Hernández‐Borges J, Rodríguez‐Delgado
MA. Fluoroquinolone antibiotic determination in bovine, ovine
and caprine milk using solid‐phase extraction and high‐
performance liquid chromatography‐fluorescence detection with
ionic liquids as mobile phase additives. J Chromatogr A.
2009;1216(43):7281‐7287.
26. Beridze N, Tsutskiridze E, Takaishvili N, Farkas T, Chankvetadze
B. Comparative Enantiomer‐Resolving Ability of Coated and
Covalently Immobilized Versions of Two Polysaccharide‐Based
Chiral Selectors in High‐Performance Liquid Chromatography.
Chromatographia. 2018;81(4):611‐621.
13. Jiménez V, Companyó R, Guiteras J. Validation of a method for
the analysis of nine quinolones in eggs by pressurized liquid
extraction and liquid chromatography with fluorescence detec-
tion. Talanta. 2011;85(1):596‐606.
27. Bajtai A, Lajkó G, Szatmári I, et al. Dedicated comparisons of
diverse polysaccharide‐ and zwitterionic Cinchona alkaloid‐
based chiral stationary phases probed with basicand ampholytic
indole analogs in liquid and subcritical fluidchromatography
mode. J Chromatogr A. 2018;1563:180‐190.
14. Jiménez V, Companyó R, Guiteras J. Analysis of quinolone anti-
biotics in eggs: preparation and characterization of a raw
material for method validation and quality control. Food Chem.
2012;134(3):1682‐1690.
28. Bouanini M, Belboukhari N, Menéndez JC, Sekkoum K, Cheriti
A, Aboul‐Enein HY. Chiral separation of novel iminonaringenin
derivatives. Chirality. 2018;30(4):484‐490.
15. Cui X‐Q, Zhang P‐J, Yang X‐L, et al. β‐CD/ATP composite mate-
rials for use in dispersive solid‐phase extraction to measure
(fluoro)quinolone antibiotics in honey samples. Anal Chim Acta.
2015;878:131‐139.
29. Gogaladze K, Chankvetadze L, Tsintsadze M, Farkas T,
Chankvetadze B. Effect of basic and acidic additives on the sep-
aration of some basic drug enantiomers on polysaccharide‐based
chiral columns with acetonitrile as mobile phase. Chirality.
2015;27(3):228‐234.
16. Loredana A, Visciano P, Stramenga A, et al. Development and
Validation of a Method for the Determination of Quinolones
in Muscle and Eggs by Liquid Chromatography‐Tandem Mass
Spectrometry. Food Anal Methods. 2016;8:2308‐2320.
30. Ali I, Suhail M, Alothman ZA, Alwarthan A. Chiral separation
and modeling of baclofen, bupropion, and etodolac profens
on amylose reversed phase chiral column. Chirality. 2017;
29(7):386‐397.
17. Rodríguez‐Gómez R, García‐Córcoles MT, Çipa M, Barrón D,
Navalón A, Zafra‐Gómez A. Determination of quinolone resi-
dues in raw cow milk. Application of polar stir‐bars and ultra‐
high performance liquid chromatography‐tandem mass spec-
trometry. Food Addit Contam Part A. 2018;35(6):1127‐1138.
31. Zhu B‐L, Yao Y‐Q, Deng M‐D, Jiang Z, Li Q. Enantioselective
separation of twelve pairs of enantiomers on polysaccharide‐
based chiral stationary phases and thermodynamic analysis of
separation mechanism. Electrophoresis. 2018;39(19):2398‐2405.
18. Hermo MP, Nemutlu E, Kir S, Barrón D, Barbosa J. Improved
determination of quinolones in milk at their MRL levels using
LC‐UV, LC‐FD, LC‐MS and LC‐MS/MS and validation in line
with regulation 2002/657/EC. Anal Chim Acta. 2008;
613(1):98‐107.
32. Douša M, Lehnert P, Adamusová H, Bosáková Z. Fundamental
study of enantioselective HPLC separation of tapentadol enan-
tiomers using cellulose‐based chiral stationary phase in normal
phase mode. J Pharm Biomed Anal. 2013;74:111‐116.
19. Blasco C, Picó Y. Development of an improved method for trace
analysis of quinolones in eggs of laying hens and wildlife species
using molecularly imprinted polymers. J Agric Food Chem.
2012;66:11005‐11014.
33. Persson BA, Andersson S. Unusual effects of separation condi-
tions on chiral separations. J Chromatogr A. 2001;906(1‐2):
195‐203.
20. Gajda A, Posyniak A, Zmudzki J, Gbylik M, Bladek T. Analysis
of quinolone antibiotics in eggs: Preparation and characteriza-
tion of a raw material for method validation and quality
control. Food Chem. 2012;135(2):430‐439.
34. Péter A, Vékes E, Armstrong DW. Effects of temperature on
retention of chiral compounds on a ristocetin a chiral stationary
phase. J Chromatogr A. 2002;958(1‐2):89‐107.
21. Lombardo‐Agüí M, García‐Campaña AM, Gámiz‐Gracia L, Cru-
ces‐Blanco C. Determination of quinolones of veterinary use in
bee products by ultra‐high performance liquid chromatography‐
tandem mass spectrometry using a QuEChERS extraction proce-
dure. Talanta. 2012;93:193‐199.
35. O'Brien T, Crocker L, Thompson R, et al. Mechanistic aspects of
chiral discrimination on modified cellulose. Anal Chem.
1997;69(11):1999‐2007.
36. Li M, Zhao Y, Zhou L, Yu J, Wang J, Guo X‐J. Study of the enan-
tiomeric separation of the anticholinergic drugs on two
immobilized polysaccharide‐based chiral stationary phases by
HPLC and the possible chiral recognition mechanisms. Electro-
phoresis. 2018;39(11):1361‐1369.
22. Hu Z‐H, Wang Y‐G, Jin H‐X, Ouyang X‐K, Wu W‐J. Analysis of
flumequine enantiomers in rat plasma by UFLC‐ESI‐MS/MS.
Chirality. 2016;28(11):737‐743.
23. Wang Y‐F, Gao X‐F, Jin H‐X, Wang Y‐G, Wu W‐J, Ouyang X‐K.
37. Szabó ZI, Foroughbakhshfasaei M, Noszál B, Tóth G.
Validation of a Chiral Liquid Chromatographic Method for the
Enantioseparation of racecadotril using polysaccharide‐

LI ET AL.
11
typechiral stationary phases in polar organic mode. Chirality.
2018;30(1):95‐105.
43. Zeng Z, Dong A, Yang G, Chen Z, Huang X. Simultaneous
determination of nine fluoroquinolones in egg white and egg
yolk by liquid chromatography with fluorescence detection. J
Chromatogr B: Anal Technol Biomed Life Sci. 2005;821:202‐209.
38. Herrera‐Herrera AV, Hernández‐Borges J, Rodríguez‐Delgado
MA, Herrero M, Cifuentes A. Determination of quinolone resi-
dues in infant and young children powdered milk combining
solid‐phase
extraction
and
ultra‐performance
liquid
chromatography‐tandem mass spectrometry. J Chromatogr A.
2011;1218(42):7608‐7614.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
39. Zhang H, Ren Y, Bao X. Simultaneous determination of (fluoro)
quinolones antibacterials residues in bovine milk using ultra
performance liquid chromatography‐tandem mass spectrometry.
J Pharm Biomed Anal. 2009;49(2):367‐374.
40. Cho HJ, Yi H, Cho SM, et al. Single‐step extraction followed by
LC for determination of (fluoro)quinolone drug residues in mus-
cle, eggs, and milk. J Sep Sci. 2010;33:1034‐1043.
How to cite this article: Li S, Liu B, Xue M, Yu J,
Guo X. Enantioseparation and determination of
flumequine enantiomers in multiple food matrices
with chiral liquid chromatography coupled with
tandem mass spectrometry. Chirality. 2019;1–11.
https://doi.org/10.1002/chir.23125
41. Chen G, Du Y. Screening of Danofloxacin Residue in Bovine
Tissue by Terbium‐Sensitized Luminescence on C18 Sorbent
Strips. J Agric Food Chem. 2011;59(4):1058‐1062.
42. Hernández‐Arteseros JA, Beltrán JL, Compañó R, Prat MD. Fast‐
scanning fluorescence spectroscopy as a detection system in
liquid chromatography for confirmatory analysis of flumequine
and oxolinic acid. J Chromatogr A. 2002;942(1‐2):275‐281.