Effective purification and enrichment of polypeptide antibiotics in animal tissues is always a challenge, due to the co-extraction of other endogenous peptides which usually interfere their final determination.
Trang 1Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/chroma
Xuqin Songa, b, Esther Turielc, Jian Yanga, Antonio Martín-Estebanc, ∗, Limin Heb, ∗
a Laboratory of Animal Genetics, Breeding and Reproduction in the Plateau Mountainous Region (Ministry of Education), College of Animal Science, Guizhou
University, Guiyang, Guizhou 550025, China
b National Reference Laboratory of Veterinary Drug Residues (SCAU), College of Veterinary Medicine, South China Agricultural University, Guangzhou 510642,
China
c Departamento de Medio Ambiente y Agronomía, INIA, CSIC, Carretera de A Coruña km 7.5, Madrid 28040, Spain
Article history:
Received 20 April 2022
Revised 31 May 2022
Accepted 31 May 2022
Available online 1 June 2022
Keywords:
Polypeptide antibiotics
Molecular imprinting
Imprinted stationary phase
Sample preparation
Animal tissue
a b s t r a c t
Effectivepurificationandenrichmentofpolypeptideantibioticsinanimaltissuesisalwaysachallenge, duetotheco-extractionofotherendogenouspeptideswhichusuallyinterferetheirfinaldetermination
Inthisstudy,amolecularlyimprintedcolumnwaspreparedbypackingpolymyxinE-imprintedparticles intoa100 mm × 4.6mm i.d.HPLC column.The as-preparedimprintedcolumns wereable to toler-ate 100%aqueousphase and exhibitedgood stabilityand highcolumn efficiency Polypeptides antibi-oticswithsimilarmolecularsizeorspatialstructuretopolymyxinEwerewellretainedbytheimprinted column,suggestingclassselectivity.Afteroptimizationofmobilephaseconditionsofimprintedcolumn, polypeptideantibioticsinanimaltissueextractswereenrichedandcleanedupbyin-linemolecularly im-printedsolid-phaseextraction,allowingthescreeningoftargetanalytesincomplexsamplesatlow con-centrationlevelsbyUVdetection.Eluatefractionfromtheimprintedcolumnwascollected,andfurther driedandre-dissolvedwithmethanol-0.5%formicacidaqueoussolution(80:20,v/v)forfinalLC-MS/MS analysis.Analysiswasaccomplishedusingmultiplereactionmonitoring(MRM)inpositiveelectrospray ionizationmodeand analytesquantifiedusingthematrix-matchedexternalcalibrationcurves.The re-sultsshowedhighcorrelationcoefficientsfortargetanalytesinthelinearrangeof2∼ 200μgkg−1.At fourdifferentconcentrationlevels(limitofquantification,50,100and 200μgkg−1),recoveriesoffour polypeptideantibioticsinswine,cattleandchickenmusclesrangedfrom66.7to94.5%withrelative stan-darddeviationslowerthan16.0%.Thelimitsofdetection(LOD)were2.0∼ 4.0μg/kg,dependingupon theanalyteandsample.Comparedwithaconventionalpretreatmentmethod,theimprintedcolumnwas abletoremovemoreimpuritiesandtosignificantlyreducematrixeffects,allowingtheaccurateanalysis
ofpolypeptideantibiotics
© 2022TheAuthors.PublishedbyElsevierB.V ThisisanopenaccessarticleundertheCCBY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/)
1 Introduction
Polypeptide antibiotics (PPTs) are a class of antibiotics isolated
from Bacillus, Streptomyces or Actinomyces, and consist of a cyclic
polypeptide structure formed by 4 ∼ 16 amino acids The an-
tibacterial mechanism of PPTs differs depending upon the antibi-
otic For instance, bacitracin (BTC) and virginiamycin (VGM) disrupt
the bacterial cell wall, while polymyxins affect the bacterial mem-
∗ Corresponding authors
E-mail addresses: amartin@inia.csic.es (A Martín-Esteban),
liminokhe@scau.edu.cn (L He)
branes Due to the favorable antibacterial effect, PPTs are widely used in animal husbandry to treat many bacterial infections, such
as dysentery, mastitis, enteritis, etc [1] PPTs like polymyxins, BCT and VGM are often added at subtherapeutic level to animal feed as growth promoters for animals Although PPTs are beneficial to an- imal production, their long-term or illegal addition to feed could cause drug residues in animal derived food and further threaten human health through the food chain [2] In addition, PPTs as a last resort against multiple drug resistance infection have been threat- ened by drug-resistant bacteria It has been reported that the long- term addition of VGM to chicken feed could increase drug resis- tance rate of Escherichia coli from 27% to 70% [3] Moreover, after
https://doi.org/10.1016/j.chroma.2022.463192
0021-9673/© 2022 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ )
Trang 2avoparcin (a glycopeptide antibiotic) use as antimicrobial growth
promoter, vancomycin (VCM) resistance was common detected in
intestinal enterococci, not only in exposed animals, but also in sur-
rounding hospitals [4] The recent emergence of MCR-1 resistance
genes in bacteria has also been associated with the excessive con-
sumption of polymyxins in animal husbandry [5]
In order to curb the rise of antimicrobial resistance, many coun-
tries and governments have adopted measures Glycopeptide an-
tibiotics, that are extremely important to humans and could pose
a serious threat to public health safety if used, were firstly banned
for the treatment of animal diseases and growth promotion [ 6, 7]
The application of other peptide antibiotics in animal production
is severely restricted as well BTC and VGM were banned as ani-
mal feed additives by the European Communities in 1998 [8] PPTs
such as polymyxin E (PME), VGM and BTC were allowed to be
added in feed at 4 ∼ 100 mg kg −1 as antimicrobial growth pro-
moters in China [9] However, the Ministry of Agriculture of China
announced prohibition of all antimicrobial growth promoters ex-
cept for Chinese medicine since 2020 [10] Furthermore, the max-
imum residues limits (MRL) of PPTs in the latest standards have
decreased from 50 ∼ 500 μg kg −1to 50 ∼ 300 μg kg −1in animal
food [11]
Considering the weak UV and fluorescence absorption of some
peptides, high performance liquid chromatography (HPLC) coupled
to evaporative light-scattering or mass spectrometry detectors, par-
ticularly liquid chromatography tandem mass spectrometry (LC-
MS/MS), are the techniques mainly used for PPTs analysis [ 12, 13]
Due to the high polarity of PPTs, their extraction is usually per-
formed using mixtures of acetonitrile (ACN) or methanol (MeOH)
and acidic water, resulting in the insufficient precipitation of pro-
teins and fats in biological samples Accordingly, the development
of a proper cleanup protocol is very important to reduce com-
plex matrix interferences and improve the accuracy of the analysis
Although various solid-phase extraction (SPE) cartridges such as
HLB, C 18 and Strata-X could be currently used to enrich and clean
up PPTs from animal tissues, but these sorbents led to the co-
extraction of impurities and serious matrix effects [ 2, 13, 14] It has
been reported that the absorption behavior of endogenous peptide
disruptors in animal tissues is similar to that of PPTs, and thus
they are co-extracted onto SPE cartridges [15], disrupting the accu-
rate characterization and quantification of drug residue would be
affected even by highly sensitive LC-MS/MS Nowadays, on-line/in-
line sample pretreatments are preferred over traditional SPE since
on-line/in-line SPE integrates sample loading, washing and elu-
tion, which greatly simplifies the pretreatment process, reducing
loss of analytes as well as eventual sample contamination [16–
19] Besides, the volume of sample extract injected into the an-
alytical instrument is very small and the consumption of organic
solvents is quite low, which is consistent with the principles of
green chemistry As a highly stable and durable recognition ma-
terial, molecularly imprinted polymers (MIP) have become an al-
ternative to selectively extract trace drugs from complex matri-
ces The use of MIPs in on-line/in-line SPE procedures (so called
MISPE) allows to simplify SPE steps, improving selective recogni-
tion and reducing matrix effects and it has been successfully ap-
plied in the on-line analysis of several veterinary drugs such as
tetracyclines [20], sulfonamides [21] and quinoxaline [22] in ani-
mal tissues, milk and eggs and in the in-line analysis of pheny-
lurea herbicides in vegetable samples [23], thiabendazole in fruits
[24]and fluoroquinolones in soils [25]
The aim of this study was to develop a selective analyt-
ical method by in-line MISPE for enrichment and purification
of selected polypeptide antibiotics in animal tissues (vancomycin
(VCM), teicoplain (TEC), polymyxin B (PMB) and bacitracin A
(BTCA)) prior their determination by LC-MS/MS The MIP parti-
cles of polymyxin E were packed in an HPLC column and used as
stationary phase After the optimization of mobile phase (loading, washing and elution conditions), the imprinting column was able
to in-line enrich analytes from complex matrix and finally com- bined with LC-MS/MS to detect PPTs
2 Material and methods
2.1 Reagents and chemicals
HPLC grade reagents including ACN, MeOH and formic acid (FA) were provided by Fisher Scientific (Fairlawn, NJ, USA) Ul- trapure water was obtained by a Millipore MilliQ equipment Trichloroacetic acid (TCA) was obtained from Guangzhou Chemical Reagent Factory (Guangzhou, China) Oasis HLB SPE cartridge (60
mg, 3 mL) was purchased from Waters Co (Milford, MA, USA)
2.2 Standards and stock solutions
The reference standards of VCM, BTCA and TEC were pur- chased from National Institutes for Food and Drug Control (Beijing, China) Daptomycin, PME and PMB (containing two major compo- nents of PMB1 and PMB2) were obtained from Dr Ehrenstrofer GmbH (Augsburg, Germany) Enrofloxacin, sulfadimidine and vir- giniamycin was available from TRC (Toronto, Canada) The purity
of each standard was higher than 84.7% Stock standard solution (1
mg mL −1) was prepared by weighing each reference standard into
a 10 mL brown volumetric flask and diluting with appropriate sol- vent as follows: daptomycin (DAP), enrofloxacin, sulfadimidine and virginiamycin (VGM) dissolved with MeOH; VCM and TEC with wa- ter; PMB, PME and BTCA with 0.1% FA aqueous solution Each stock solution should be stored at -20 °C for a maximum of 3 months Mixed standard working solutions at 10 μg mL −1 concentration were prepared by diluting stock solution with the mixed solution
of MeOH and 0.1% FA aqueous solution (50:50, v/v), which should
be stored at -20 °C during no more than a month
2.3 Preparation of imprinted chromatographic column
The MIP particles of polymyxin E were obtained by precipita- tion polymerization according to our previous study [26] After re- moving the template, the MIP particles were washed with water and MeOH for three times and dried in an oven at 60 °C under vac- uum for 24 h An amount of 3 g MIP particles were dispersed into
45 mL of HPLC grade isopropanol After sonication for 20 min, MIP particles were packed into an empty column (100 mm × 4.6 mm i.d.) under high pressure (80 0 0 psi) provided by a CP12 liquid chromatographic column packing machine (Scientific Systems Inc (SSI), CA, USA) Finally, the MIP column was rinsed with MeOH at
a flow rate of 0.2 mL min −1 for 24 h
2.4 Chromatographic performance of MIP column
The mobile phases including MeOH, ACN, water or FA in wa- ter were tested The chromatographic performance of MIP column, including background pressure and stability, were evaluated using polymyxin E as the target analyte The retention factor ( k) and the- oretical plates number (N) of both MIP and non-imprinted (NIP) columns were calculated as following equations:
k=t R − t0
t0
where, t R (min) is the retention time of polymyxin E; t 0 (min) is the void time of solvent peak
N=5.54
t R
w1/2
2
=16
t
R w
2
Trang 3where, w 1/2(min) is the peak width at half peak height; w (min)
is the peak width at peak base
2.5 Sample preparation based on in-line separation using HPLC
Muscle samples including beef, pork and chicken were obtained
from local supermarkets (Guangzhou, China) After homogeniza-
tion, 2 g samples were accurately weighed into a 50 mL polypropy-
lene centrifuge tube and spiked with appropriate standard working
solutions for incubation at room temperature for 30 min The ex-
traction was performed with 5 mL of ACN-10% TCA in water (1:1,
v/v) through sonication for 5 min and shaking for 20 min Af-
ter centrifugation, the supernatant was collected and the residue
was re-extracted following the same above extraction procedure
All the supernatants were combined and the final volume was
adjusted to 10 mL A volume of 2 mL of the extract was col-
lected and evaporated to near dryness The residues were dissolved
in 0.5 mL of MeOH-0.5% FA aqueous solution (80:20, v/v) before
in-line MISPE by HPLC-UV Once the chromatographic run time
reached 8 min, the target fraction of analytes was started and col-
lected into a 5 mL test tube until a run time of 10.5 min Then,
the eluate fraction of analytes at their retention time was collected
and evaporated to dryness at 45 °C Finally, the residues were re-
dissolved with 0.2 mL of the above reconstitution solution and an-
alyzed by LC-MS/MS
2.6 HPLC-UV and LC-MS/MS conditions
The chromatographic performance of MIP column and corre-
sponding in-line MISPE procedures were performed by HPLC-UV
The mobile phase includes ACN (A), 50% ACN in water (B) and 50%
ACN in water containing 0.02% FA (C) The in-line MISPE onto the
imprinted column was carried out with the following gradient elu-
tion program: 0 ∼6 min for loading, 100% A; 6.1 ∼7 min for wash-
ing, 100% B; 7.1 ∼11 min for eluting, 100% C; 11.5 ∼16 min for recon-
ditioning for the next run The flow rate was 0.75 mL min −1 and
the injection volume was 100 μL Target analytes were monitored
at 205 nm
Due to the trace amounts of polypeptides residues in ani-
mal tissues, final sample analysis was performed by LC-MS/MS A
Shimadzu HPLC system (Shimadzu, Kyoto, Japan) and an Applied
Biosystems Sciex Triple Quad 5500 triple-quadrupole mass spec-
trometer were used to detect the analytes Chromatographic and
mass conditions such as mass parameters, mobile phase and elu-
tion program were the same as our previous report [12] Briefly,
a Phenomenex Kinetex Biphenyl column (50 mm × 2.1 mm i.d.,
2.6 μm, Phenomenex, Torrance, CA) was used to separate the an-
alytes The mobile phase consisted of 0.1% FA in ACN solution (A)
and 0.1% FA in water solution (B) with the following gradient elu-
tion: 0 min, 6% A; 2 min, 6% A; 5 min, 40% A; 14 min, 70% A;
14.1 min, 6% A; 18 min, 6% A The mass conditions were acquired
in multiple reaction monitoring (MRM) mode The tune parameters
were carried out as follows: ionspray voltage, 5500 V; nebulizing
gas pressure, 55 psi; auxiliary gas,50 psi; curtain gas, 40 psi; ion
source temperature, 600 °C; entrancepotential, 10 V and collision
cell exit potential, 12 V At least two product ions of each com-
pound were monitored under the ESI +mode and the mass param-
eters are given in Table S1
2.7 Method validation
Eluate fraction of analytes at their retention time (8 to
10.5 min) was collected and further analyzed by LC-MS/MS Under
optimum conditions, method parameters for validation including
linearity, accuracy, precision and sensitivity were assessed Matrix-
matched standard solutions (0.5, 1, 2, 5, 10, 20, 50, 80, 100 and
200 ng/mL) were prepared to plot calibration curves Mean recov- eries of polypeptide antibiotics from pork, beef and chicken at the spiked concentrations of limit of quantification (LOQ), 50, 100 and
200 μg kg −1were calculated
3 Results and discussion
3.1 Optimization of mobile phase
Chromatographic conditions have a significant impact on the retention performance and impurity removal capability of the im- printed column during in-line MISPE Generally, in-line MISPE con- sists of three steps [23–25]: first, during the loading step, an or- ganic solvent (eg ACN) is prioritized as the mobile phase, since the co-extracted fat-soluble impurities are eluted rapidly by the or- ganic mobile phase, while target analytes are retained well on the MIP column with specific recognition Thereafter, a mixture of wa- ter and organic solvent is commonly used as the washing solution
to reduce the interference of complex matrix but without disrupt- ing the specific interactions between analytes and MIP Finally, a suitable eluent is selected to elute the target analytes still retained
by the MIP In order to make sure that polypeptides could be selec- tively bound to MIP column whereas impurities co-extracted were removed as much as possible, the in-line MISPE conditions (load- ing, washing and elution) should be optimized through adjusting the composition of mobile phase and the elution program In this study, several mobile phase compositions, including MeOH, ACN, MeOH-water/acid water, ACN-water/acid water were tested The results showed that a very high baseline background and an im- portant peak tailing of polymyxin E were observed when MeOH was used in the mobile phase, which could be explained by the strong UV absorption of MeOH at low UV wavelengths Accord- ingly, MeOH was discarded for further experiments and the ef- fect of the presence of ACN in the mobile phase on the reten- tion/separation of polymyxin E was studied It was observed that polymyxin E was completely retained onto the MIP column, mak- ing necessary the addition of formic acid to the component C of mobile phase (see Section 2.6) in order to disrupt the hydrogen bonding interactions occurring between target analyte and binding sites, thus allowing the elution of polymyxin E [23] As shown in Fig 1A, the retention time and peak shape of polymyxin E were not significantly affected by the increase of FA concentration, but, high FA concentration caused baseline instability due to FA strong absorption at low UV wavelengths Thus, a 0.02% FA was chosen as optimum to be present in the component C of the mobile phase to elute polymyxin E Besides, it was observed that polymyxin E was eluted faster with the increase of ACN concentration in component
C of the mobile phase ( Fig.1B), which suggests that MIP column exhibited also a reversed-phase retention mechanism alongside hy- drogen bonding interactions as mentioned above
3.2 Chromatographic performance and selectivity of MIP column
The chromatographic parameters affecting MIP column perfor- mance, including background pressure, stability and retention fac- tor, were examined Considering the commonly used solvents in HPLC analysis, the background pressure of MIP column in MeOH, ACN and water was evaluated at flow rates ranging from 0.2 to 1.0 mL min −1 As presented in Fig S1-A, the background pres- sures in different solvents are in the order: water >MeOH >ACN The highest is the water with a background pressure of 125 bar, in- dicating that the MIP column could tolerate 100% aqueous phase The reproducibility of retention times, thus the performance sta- bility of MIP column, was assessed by injecting the standard so- lution of polymyxin E for eight times in a row and the chro- matograms obtained are shown in Figure S1-B The retention time
3
Trang 4Fig 1 Effect of different percentage of FA concentration (A) and ACN (B) in component C of mobile phase on the retention of polymyxin E ACN: acetonitrile; FA: formic
acid
Fig 2 HPLC-UV chromatograms of polymyxin E on MIP column and NIP column Chromatographic conditions: see Experimental section
was 9.90 min with a relative standard deviation (RSD) of 0.04%,
whereas the theoretical plate number was 3298 with the RSD of
9.84%, demonstrating the excellent stability of the MIP column,
which would allow to be routinely used in HPLC analysis
Under the optimal HPLC conditions, the selectivity was inves-
tigated by comparing the retention of polymyxin E on MIP and
non-imprinted (NIP) columns The results showed ( Fig 2) that
polymyxin E was well retained and eluted from the MIP column
with sharp peak, while the retention time of polymyxin E on NIP
was 8.8 min with serious peak tailing The theoretical plate num-
ber obtained from NIP column was 136, much lower than that
from MIP column The stronger retention of polymyxin E on MIP
than on NIP, as well as the different peak shape demonstrate the
presence of selective binding sites on MIP column It is important
to point out that, strictly speaking, the chromatographic param-
eters measured (k and N) are accurate only for isocratic elution
conditions and for chromatographic Gaussian peaks and thus the
reported values can only be used for comparison purposes of MIP
and NIP columns under the experimental conditions indicated in
the present paper
The cross-reactivity was estimated by analyzing five different
polypeptide antibiotics (TEC, VCM, VGM, PMB and DAP) and two
antimicrobials (enrofloxacin and sulfadimidine) with a large con-
sumption in animal production Each compound was analyzed at
its optimal UV wavelength As illustrated in Fig 3, polypeptide
antibiotics, except for VGM, were well retained and eluted from the MIP column with a rather symmetrical peak shape and neg- ligible tailing, while enrofloxacin and sulfadimidine were not re- tained with retention times lower than 2 min This result con- firmed the existence of imprinted cavities, which can well match the polymyxin E shape, size and functional groups Since TEC, VCM, DAP and PMB have large molecular weight and complex cyclic polypeptide structure similar to polymyxin E, all of them were well retained by the imprinted column On the contrary, although VGM belongs to polypeptides, its structure is a large lactone ring with
a molecular weight lower than 600 Da, which significantly differs from that of polymyxin E (template) Enrofloxacin and sulfonamide whose molecular structures are quite different from template could not be retained on the MIP column Therefore, these results re- vealed that specific binding sites on the MIP contribute to the re- tention of analytes, allowing the determination of several polypep- tides simultaneously
3.3 Preparation of animal tissue extracts
Proteins in animal tissues could strongly interact with polypep- tide antibiotics, and thus low pH of the extraction solvent (TCA, sulfuric acid and hydrochloric acid) is utilized to extract the analytes Several studies have confirmed that the mixture of ACN/MeOH and 10% TCA aqueous solution is able to quantitatively
Trang 5Fig 3 HPLC-UV chromatograms of polypeptide antibiotics, enrofloxacin, and sulfadimidine on MIP column Chromatographic conditions: see Experimental section
recover polypeptides from biological samples [ 27, 28] Accordingly,
MeOH-10% TCA aqueous solution (1:1, v/v) and different propor-
tions of ACN in 10% TCA aqueous solutions for the extraction of
polypeptide antibiotics from animal tissue samples were tested As
shown in Fig S2, MeOH-10% TCA aqueous solution was not able
to completely disrupt the interactions of target analytes with sam-
ple matrix, leading to recoveries lower than 80 % for TEC However,
ACN-10% TCA aqueous solution (1:1, v/v) allowed to quantitatively
extract all the analytes under study reaching recoveries higher than
95%, and thus was selected for further experiments Furthermore,
since it was necessary to dry and reconstitute sample extracts, the
effect of different ratios of MeOH:0.5% formic acid in water as re-
constitution solution was evaluated It was observed that the re-
coveries of analytes increased with the presence of MeOH Finally,
an 8:2 (v/v) ratio was the optimum, providing recoveries higher
than 91 % for all the target analytes
3.4 In-line MISPE of polypeptide antibiotics from animal tissue
extracts
Sample extracts were injected in the chromatographic system
for the in-line MISPE of polypeptide antibiotics under optimum
conditions Polymyxin E was not included in this study since it was
used as template for MIP preparation It is well-known that, even
after exhaustive washing of MIPs, template leaking might occur
which would compromise the accurate determination of polymysin
E at trace level in real samples Fig 4 shows the LC-UV chro-
matogram at 205 nm obtained in the analysis of non-spiked and
spiked pork sample at the concentration of 200 μg kg −1 As can
Fig 4 HPLC-UV chromatograms obtained after the injection of spiked at 200 μg
kg −1 in pork sample and non-spiked pork sample directly onto the imprinted col- umn Chromatographic conditions: see Experimental section
be observed, the mixture of polypeptide antibiotics was unambigu- ously detected in the spiked sample thanks to the high selectivity provided by the MIP Target analytes were recognized and eluted free of co-extractives from the MIP column, with retention times from 8 to 10.5 min, whereas the matrix interferences were rapidly eluted, allowing the detection of polypeptide antibiotics at very low concentration level in the pork sample extract without any
5
Trang 6Table 1
Recovery and precision of polypeptide antibiotics in animal muscles ( n = 5)
Average recovery (RSD) a, %
Compound ( μg kg −1 ) Intra-batch Inter-batch Intra-batch Inter-batch Intra-batch Inter-batch Vancomycin LOQ 71.3(9.7) 75.5(11.6) 73.5(11.5) 69.7(10.8) 68.1(8.5) 68.2(6.2)
50 78.1(11.3) 79.9(9.1) 78.8(5.9) 84.4(8.9) 83.0(6.6) 81.3(7.2)
100 89.0(3.2) 88.1(5.6) 85.7(11.2) 88.9(12.2) 80.6(5.4) 83.6(6.0)
200 83.9(3.3) 82.2(6.4) 90.5(3.3) 86.9(6.2) 89.2(2.3) 82.1(10.6) Teicoplain LOQ 69.1(9.9) 72.1(13.8) 66.6(11.1) 66.7(13.4) 83.8(7.8) 81.5(7.7) A2-1 50 75.4(9.6) 81.8(12.1) 76.1(11.5) 76.0(14.3) 82.9(6.7) 85.1(7.5)
100 81.0(14.3) 85.7(9.1) 94.5(2.8) 86.7(10.3) 77.4(7.6) 77.0(8.2)
200 82.4(6.4) 85.5(7.1) 85.9(7.5) 84.2(7.5) 77.4(7.1) 85.0(10.6) Teicoplain LOQ 69.5(10.2) 71.6(9.5) 77.4(13.0) 73.5(12.6) 77.0(5.8) 79.7(6.3) A2-2&2-3 50 68.2(11.1) 76.7(13.5) 75.2(11.8) 75.7(16.0) 69.4(6.7) 71.9(8.2)
100 89.4(13.2) 90.8(11.2) 83.1(7.3) 86.7(11.8) 83.9(4.4) 78.2(8.2)
200 87.2(6.4) 86.5(7.2) 84.3(3.7) 82.3(9.1) 81.7(8.8) 81.2(12.3) Polymyxin LOQ 71.0(9.1) 79.5(11.9) 86.6(4.5) 81.9(6.6) 86.5(4.0) 80.8(10.7) B1 50 79.1(4.9) 78.6(5.6) 90.0(3.0) 85.3(7.9) 81.3(4.2) 83.6(4.7)
100 85.2(1.7) 84.0(3.2) 85.1(10.5) 83.9(12.3) 81.6(10.8) 85.6(6.9)
200 79.6(2.4) 80.1(4.4) 88.4(4.7) 86.3(5.4) 80.8(10) 81.3(8.0) Polymyxin LOQ 88.1(7.5) 85.9(11.1) 84.3(4.6) 79.7(9.9) 86.5(10.5) 79.4(15.0) B2 50 80.0(6.6) 80.5(6.0) 85.4(2.4) 87.2(7.4) 84.3(4.6) 83.8(3.5)
100 89.0(3.0) 88.3(4.2) 86.8(9.2) 89.9(11.1) 84.6(8.4) 87.0(5.7)
200 83.4(2.6) 83.4(3.9) 85.2(3.0) 87.5(3.5) 82.6(12.4) 82.8(9.6) Bacitracin A LOQ 71.1(4.5) 74.1(7.3) 69.2(13.9) 74.5(11.6) 84.1(3.4) 83.9(6.0)
50 83.5(5.2) 82.8(3.8) 89.6(3.0) 88.1(4.9) 83.7(6.8) 84.6(5.4)
100 89.6(2.7) 88.5(2.9) 87.2(1.5) 88.1(9.5) 78.0(8.9) 83.5(7.5)
200 84.6(2.8) 83.9(2.8) 89.2(2.1) 90.8(3.0) 83.9(6.1) 85.7(8.2)
a RSD, relative standard deviation
sample clean-up However, due to the slight differences observed
in the retention times (see Fig 3), it was not possible to resolve
target analytes when the mixture was injected into the MIP col-
umn Thus, from the obtained results, it can be concluded that the
in-line MISPE procedure is suitable for the screening of polypeptide
antibiotics in crude animal tissue extracts at the concentration lev-
els required without any other sample clean-up In this sense, only
those positive samples would be subjected to further analysis by
fraction collection and subsequent LC-MS/MS analysis as described
below
3.5 Analysis of polypeptide antibiotics from in-line MISPE extracts by
LC-MS/MS and method validation
As indicated in Section 2.5, the eluate fraction of analytes at
their retention time (8 to 10.5 min) was collected and further an-
alyzed by LC-MS/MS Under optimum conditions, method parame-
ters for validation including linearity, accuracy, precision and sen-
sitivity were assessed Matrix-matched standard solutions (0.5, 1,
2, 5, 10, 20, 50, 80, 10 0 and 20 0 ng/mL) were prepared to plot
calibration curves Good linearities at the spiked concentrations
of 5 ∼200 ng mL −1 for VCM and TEC, and 2 ∼200 ng mL −1 for
BTCA and PMB were observed with correlation coefficients ( ) bet-
ter than 0.99 Obtained recoveries and their corresponding relative
standard deviation (RSD) are summarized in Table1 Mean recov-
eries of polypeptide antibiotics from pork, beef and chicken at the
spiked concentrations of limit of quantification (LOQ), 50, 100 and
200 μg kg −1 were within 66.7% ∼94.5% interval with RSDs lower
than 14.3%, which are within the acceptable limits of residue anal-
ysis The typical MRM chromatograms acquired under ESI mode
from spiked pork at 100 μg kg −1 are depicted in Fig.5 The limit
of detection (LOD) and LOQ ranged from 2 to 4 μg kg −1 and 5 to
12.5 μg kg −1, respectively, which are lower than the MRLs recom-
mended by the Ministry of Agriculture of China All results have
demonstrated that proposed method is accurate, reliable and sen-
sitive for the monitoring of polypeptides residues in animal tissues
( Table2)
3.6 Matrix effect
Matrix effect (ME) caused by the co-extracted impurities could suppress or enhance the mass spectrum response of analytes un- der ESI mode The matrix effect was calculated by the following equation:
ME (% )
= Slope of matrix matched standard curve − Slope of standard curve
Slope of standard curve
× 100%
where the slope of standard curve is obtained by the linear fit- ting equation of pure standard solutions (without matrix) The pos- itive and negative values of ME represent the signal enhancement and signal suppression, respectively In addition, the values of ME within the range of ±0∼20%,±20∼50% and >±50% mean the soft, medium and strong matrix effect, respectively As demonstrated
in Fig.6, except that BTCA in chicken (-23.1%) and pork matrix (- 25.3%) showing medium matrix effect, other target analytes exhib- ited slight signal suppression as the matrix effects were soft (less than ±0∼20%). Many studies confirmed that strong signal suppres- sion for polypeptides was observed in biological sample matrices such as chicken (-41 ∼-52%) [2], pork (-24 ∼-86%) [13]and fish ( <
85%) [14], even though the SPE strategy was used for sample pu- rification in the cited above studies Considering severe matrix ef- fects for polypeptides, the developed approach of in-line molecu- larly imprinted solid-phase extraction provided less matrix inter- ference, which could become a better alternative for sample pu- rification
3.7 Comparison with literature method
The developed method was compared with the literature method [29], involving in the extraction with ACN-10% TCA aque- ous solution and the purification with Oasis HLB SPE cartridge As exhibited in Fig S3, there were serious impurity interferences af- ter extraction using a conventional SPE cartridge A relatively clean chromatogram was observed when only 1 mL of the extract was
Trang 7Fig 5 Typical MRM chromatograms obtained from spiked (100 μg kg −1 ) (A) and non-spiked pork sample extracts (B) from the elute fraction obtained by in-line MISPE Chromatographic conditions: see Experimental section
Table 2
The calibration curve, linearity, limit of detection, limit of quantitation of the developed method for polypeptide antibiotics in animal muscles
Compound Muscle Calibration curve Correlationcoefficient ( r 2 ) Linear range(ng mL −1 ) LOD a( μg kg −1 ) LOQ b( μg kg -1 )
a LOD, limit of detection
b LOQ, limit of quantitation
7
Trang 8Fig 6 Matrix effects of chicken, beef and pork matrices on the response of target
analytes ( n = 3)
directly introduced into the LC-MS/MS, but the existence of many
impurities, especially from TCA, would not only reduce the effi-
ciency of ion formation but also would affect the long-term service
life of analytical instrument In sharp contrast, there were few im-
purity interferences after the in-line extraction using MIP column
with the matrix effect less than -20%, suggesting very week matrix
suppression effects Consequently, the proposed analytical method
based on in-line MISPE might be a convenient alternative for the
determination of polypeptides in animal foodstuffs
3.8 Applications to real samples
To demonstrate the feasibility and practicability of the devel-
oped method, 60 muscle samples (20 chicken, 20 beef and 20
pork) collected from different provinces were analyzed All muscle
samples were free from VGM and TEC, perhaps because these two
compounds are expensive and more importantly, they have been
banned in food producing animal for a long time The PMB and
BTCA were detected in 1 chicken and 4 pork samples, indicating
that they are still used in animal production to prevent bacterial
infection, or even use as growth promoters However, the concen-
trations of PMB and BTCA detected in muscle samples were below
the corresponding MRLs The highest concentration of BTCA de-
tected was 43.6 μg kg −1 in pork sample and the lowest was only
3.9 μg kg −1of PMB in chicken Since the PMB and BTCA have low
oral absorption, the residues in animal tissues are low when drugs
are rationally used and withdrawal time are enough
4 Conclusions
In this study, a molecularly imprinted column was prepared us-
ing polymysin E-imprinted particles as stationary phase for the
in-line extraction of polypeptide antibiotics from animal food-
stuffs The imprinted column exhibited excellent stability and high
enough column efficiency The polypeptide antibiotics with simi-
lar structure to polymyxin E were all well-retained on the MIP
column, further demonstrating that the MIP has class-specific
recognition abilities, allowing the selective enrichment of several
polypeptide antibiotics simultaneously from complex samples Af-
ter optimization, the imprinted-stationary phase was successfully
employed for the screening of polypeptide antibiotics from animal
muscles sample extracts by HPLC-UV without any further clean-up
at the concentration levels required Finally, sample extracts were
collected and further analyzed by LC-MS/MS Compared to conven-
tional SPE, the developed method is able to efficiently remove ma-
trix interferences, allowing to meet current requirements for anal-
ysis of trace amounts of sevral polypeptides in terms of precision
and accuracy The efficient purification, automated operation and
rapid separation provided by the present developed method open new bright and exciting research avenues in the analysis of antibi- otics in complex matrices
Declaration of Competing Interest
The authors declare that they have no known competing finan- cial interests or personal relationships that could have appeared to influence the work reported in this paper
CRediT authorship contribution statement Xuqin Song: Methodology, Validation, Writing – original draft,
Funding acquisition Esther Turiel: Conceptualization, Supervi- sion, Writing – review & editing Jian Yang: Methodology
Antonio Martín-Esteban: Conceptualization, Methodology, Super-
vision, Writing – review & editing, Resources, Funding acquisition
Limin He: Conceptualization, Supervision, Writing – review & edit-
ing, Resources, Funding acquisition
Acknowledgments
The authors are grateful to the assistance from Myriam Díaz-Álvarez This research was funded by the National Science Foundation of China (grant number 31572562), the Key Pro- gram of Guangzhou Science and Technology Plan (grant number 201804020019) and the Special Funds of the National Natural Sci- ence Foundation of Guizhou University ([2020]25)
References
[1] I.L Tsai, H.Y Sun, G.Y Chen, S.W Lin, C.H Kuo, Simultaneous quantification
of antimicrobial agents for multidrug-resistant bacterial infections in human plasma by ultra-high-pressure liquid chromatography–tandem mass spectrom- etry, Talanta 116 (2013) 593–603
[2] J Boison, S Lee, J Matus, A multi-residue method for the determination of seven polypeptide drug residues in chicken muscle tissues by LC-MS/MS, Anal Bioanal Chem 407 (2015) 4065–4078
[3] F Aarestrup, A Seyfarth, H.D Emborg, K Pedersen, R Hendriksen, F Bager, Ef- fect of abolishment of the use of antimicrobial agents for growth promotion on occurrence of antimicrobial resistance in fecal enterococci from food animals
in Denmark, Antimicrob Agents Chemother 45 (2001) 2054–2059 [4] A.E van den Bogaard, E.E Stobberingh, Epidemiology of resistance to antibi- otics: links between animals and humans, Int J Antimicrob Agents 14 (20 0 0) 327–335
[5] Y.Y Liu, Y Wang, T.R Walsh, L.X Yi, R Zhang, J Spencer, Y Doi, G Tian,
B Dong, X Huang, L.F Yu, D Gu, H Ren, X Chen, L Lv, D He, H Zhou, Z Liang, J.H Liu, J Shen, Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecu- lar biological study, Lancet Infect Dis 16 (2016) 161–168
[6] Announcement No 560 Veterinary Drug Local Standard Revocation Cata- logue http://www.moa.gov.cn/gk/zcfg/nybgz/20 0806/t20 080606 _ 1057240.htm Ministry of Agriculture and Rural Affairs of the People’s Republic of China March 2008
[7] US Food and Drug Administration, Code of Federal Regulations Title 21, Section 530.41: Drugs prohibited for extralabel use in animals, March (2021) https://www.ecfr.gov/current/title-21/chapter-I/subchapter- E/part- 530/ subpart- E/section- 530.41
[8] Commission of the European communities, council regulation 2821/98, Off J Eur Commun Legis L351 (2012) 4
[9] Announcement No 168 Specifications for the Use of Feed Drug Additives July 2001 http://www.moa.gov.cn/gk/tzgg _ 1/gg/200210/t20021028 _ 18792.htm Ministry of Agriculture and Rural Affairs of the People’s Republic of China [10] Announcement No 194 January 2020 http://www.moa.gov.cn/nybgb/2019/ 201907/202001/t20200103 _ 6334292.htm Ministry of Agriculture and Rural Af- fairs of the People’s Republic of China
[11] Ministry of Agriculture and Rural Affairs of China, the National Health Com- mission of China, and the State Administration of Market Regulation, Docu- ment GB-31650-2019 National Food Safety Standard-Maximum Residue Limits for Veterinary Drugs in Foods, September (2019) http://www.moa.gov.cn/xw/ zwdt/201910/t20191008 _ 6329518.htm
[12] X Song, Q Huang, Y Zhang, M Zhang, J Xie, L He, Rapid multiresidue analysis
of authorized/banned cyclopolypeptide antibiotics in feed by liquid chromatog- raphy–tandem mass spectrometry based on dispersive solid-phase extraction,
J Pharm Biomed Anal 170 (2019) 234–242 [13] A Kaufmann, M Widmer, Quantitative analysis of polypeptide antibiotic residues in a variety of food matrices by liquid chromatography coupled to tandem mass spectrometry, Anal Chim Acta 797 (2013) 81–88
Trang 9[14] Y Xu, X Tian, C Ren, H Huang, X Zhang, X Gong, H Liu, Z Yu, L Zhang,
Analysis of colistin A and B in fishery products by ultra performance liquid
chromatography with positive electrospray ionization tandem mass spectrom-
etry, J Chromatogr B 899 (2012) 14–20
[15] E Wan, C Ho, D Sin, Y.C Wong, Detection of residual bacitracin A, colistin A,
and colistin B in milk and animal tissues by liquid chromatography tandem
mass spectrometry, Anal Bioanal Chem 385 (2006) 181–188
[16] S Rodriguez-Mozaz, M.J Lopez de Alda, D Barceló, Advantages and limitations
of on-line solid phase extraction coupled to liquid chromatography–mass spec-
trometry technologies versus biosensors for monitoring of emerging contami-
nants in water, J Chromatogr A 1152 (2007) 97–115
[17] A.L Heffernan, K Thom pson, G Eaglesham, S Vijayasarathy, J.F Mueller,
P.D Sly, M.J Gomez, Rapid, automated online SPE-LC-QTRAP-MS/MS method
for the simultaneous analysis of 14 phthalate metabolites and 5 bisphenol ana-
logues in human urine, Talanta 151 (2016) 224–233
[18] C Moser, D Zoderer, G Luef, M Rauchenzauner, L Wildt, A Griesmacher,
C Seger, Simultaneous online SPE-LC-MS/MS quantification of six widely
used synthetic progestins in human plasma, Anal Bioanal Chem 403 (2011)
961–972
[19] C Ramirez, C Wang, P Gardinali, Fully automated trace level determination
of parent and alkylated PAHs in environmental waters by online SPE-LC-AP-
PI-MS/MS, Anal Bioanal Chem (2013) 406
[20] Y.K Lv, J.Q Zhang, Z.Y Guo, W Zhang, H.W Sun, Determination of tetracyclines
residues in egg, milk, and milk powder by online coupling of a precolumn
packed with molecular imprinted hybrid composite materials to RP-HPLC-UV,
J Liq Chromatogr Relat Technol 38 (2015) 1–7
[21] J He, S Wang, G Fang, H Zhu, Y Zhang, Molecularly imprinted poly-
mer online solid-phase extraction coupled with high-performance liquid
chromatography −UV for the determination of three sulfonamides in pork and
chicken, J Agric Food Chem 56 (2008) 2919–2925
[22] Z.J Duan, L.P Fan, G.Z Fang, J.H Yi, S Wang, Novel surface molecularly imprinted sol–gel polymer applied to the online solid phase extraction of methyl-3-quinoxaline-2-carboxylic acid and quinoxaline-2-carboxylic acid from pork muscle, Anal Bioanal Chem 401 (2011) 2291–2299
[23] F.G Tamayo, A Martin-Esteban, Selective high performance liquid chromatog- raphy imprinted-stationary phases for the screening of phenylurea herbicides
in vegetable samples, J Chromatogr A 1098 (2005) 116–122 [24] E Turiel, J.L Tadeo, P.A.G Cormack, A Martin-Esteban, HPLC imprinted-sta- tionary phase prepared by precipitation polymerisation for the determination
of thiabendazole in fruit, Analyst 130 (2005) 1601–1607 [25] E Turiel, A Martín-Esteban, J.L Tadeo, Molecular imprinting-based separation methods for selective analysis of fluoroquinolones in soils, J Chromatogr A
1172 (2007) 97–104 [26] X Song, E Turiel, L He, A Martín-Esteban, Synthesis of molecularly imprinted polymers for the selective extraction of polymyxins from environmental water samples, Polymers 12 (2020) 131
[27] Q Fu, X Li, K Zheng, Y Ke, Y Wang, L Wang, F Yu, X Xia, Determination of colistin in animal tissues, egg, milk, and feed by ultra-high performance liquid chromatography-tandem mass spectrometry, Food Chem 248 (2017) 166–172 [28] H Kumar, D Kumar, E Nepovimova, D Oulkar, A Kumar, R Mohammad, S Bu- dakoti, N Upadhyay, R Verma, K Kuca, Determination of colistin B in chicken muscle and egg using ultra-high-performance liquid chromatography-tandem mass spectrometry, Int J Environ Res Public Health 18 (2021) 12
[29] D Zhang, J.-A Park, D.-S Kim, N.-H Kim, S.-K Kim, K.-S Cho, D Jeong, J.-H Shim, A.M Abd El-Aty, H.-C Shin, Simultaneous detection of bacitracin and polymyxin B in livestock products using liquid chromatography with tan- dem mass spectrometry, J Sep Science 38 (2015) 2371–2380
9