Determination of polypeptide antibiotics in animal tissues using liquid chromatography tandem mass spectrometry based on in-line molecularly imprinted solid-phase extraction

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.

Journal of Chromatography A 1673 (2022) 463192 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Determination of polypeptide antibiotics in animal tissues using liquid chromatography tandem mass spectrometry based on in-line molecularly imprinted solid-phase extraction Xuqin Song a,b, Esther Turiel c, Jian Yang a, Antonio Martín-Esteban c,∗, Limin He b,∗ 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 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 b a r t i c l e i n f o Article history: Received 20 April 2022 Revised 31 May 2022 Accepted 31 May 2022 Available online June 2022 Keywords: Polypeptide antibiotics Molecular imprinting Imprinted stationary phase Sample preparation Animal tissue a b s t r a c t 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 In this study, a molecularly imprinted column was prepared by packing polymyxin E-imprinted particles into a 100 mm × 4.6 mm i.d HPLC column The as-prepared imprinted columns were able to tolerate 100% aqueous phase and exhibited good stability and high column efficiency Polypeptides antibiotics with similar molecular size or spatial structure to polymyxin E were well retained by the imprinted column, suggesting class selectivity After optimization of mobile phase conditions of imprinted column, polypeptide antibiotics in animal tissue extracts were enriched and cleaned up by in-line molecularly imprinted solid-phase extraction, allowing the screening of target analytes in complex samples at low concentration levels by UV detection Eluate fraction from the imprinted column was collected, and further dried and re-dissolved with methanol-0.5% formic acid aqueous solution (80:20, v/v) for final LC-MS/MS analysis Analysis was accomplished using multiple reaction monitoring (MRM) in positive electrospray ionization mode and analytes quantified using the matrix-matched external calibration curves The results showed high correlation coefficients for target analytes in the linear range of ∼ 200 μg kg−1 At four different concentration levels (limit of quantification, 50, 100 and 200 μg kg−1 ), recoveries of four polypeptide antibiotics in swine, cattle and chicken muscles ranged from 66.7 to 94.5% with relative standard deviations lower than 16.0% The limits of detection (LOD) were 2.0 ∼ 4.0 μg/kg, depending upon the analyte and sample Compared with a conventional pretreatment method, the imprinted column was able to remove more impurities and to significantly reduce matrix effects, allowing the accurate analysis of polypeptide antibiotics © 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/) Introduction Polypeptide antibiotics (PPTs) are a class of antibiotics isolated from Bacillus, Streptomyces or Actinomyces, and consist of a cyclic polypeptide structure formed by ∼ 16 amino acids The antibacterial mechanism of PPTs differs depending upon the antibiotic 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 animal 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 threatened by drug-resistant bacteria It has been reported that the longterm addition of VGM to chicken feed could increase drug resistance 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/) X Song, E Turiel, J Yang et al Journal of Chromatography A 1673 (2022) 463192 avoparcin (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 surrounding hospitals [4] The recent emergence of MCR-1 resistance genes in bacteria has also been associated with the excessive consumption of polymyxins in animal husbandry [5] In order to curb the rise of antimicrobial resistance, many countries and governments have adopted measures Glycopeptide antibiotics, 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 animal 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 ∼ 100 mg kg−1 as antimicrobial growth promoters in China [9] However, the Ministry of Agriculture of China announced prohibition of all antimicrobial growth promoters except for Chinese medicine since 2020 [10] Furthermore, the maximum residues limits (MRL) of PPTs in the latest standards have decreased from 50 ∼ 500 μg kg−1 to 50 ∼ 300 μg kg−1 in 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, particularly liquid chromatography tandem mass spectrometry (LCMS/MS), are the techniques mainly used for PPTs analysis [12,13] Due to the high polarity of PPTs, their extraction is usually performed using mixtures of acetonitrile (ACN) or methanol (MeOH) and acidic water, resulting in the insufficient precipitation of proteins and fats in biological samples Accordingly, the development of a proper cleanup protocol is very important to reduce complex matrix interferences and improve the accuracy of the analysis Although various solid-phase extraction (SPE) cartridges such as HLB, C18 and Strata-X could be currently used to enrich and clean up PPTs from animal tissues, but these sorbents led to the coextraction 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 accurate characterization and quantification of drug residue would be affected even by highly sensitive LC-MS/MS Nowadays, on-line/inline sample pretreatments are preferred over traditional SPE since on-line/in-line SPE integrates sample loading, washing and elution, 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 analytical 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 material, molecularly imprinted polymers (MIP) have become an alternative to selectively extract trace drugs from complex matrices The use of MIPs in on-line/in-line SPE procedures (so called MISPE) allows to simplify SPE steps, improving selective recognition and reducing matrix effects and it has been successfully applied in the on-line analysis of several veterinary drugs such as tetracyclines [20], sulfonamides [21] and quinoxaline [22] in animal tissues, milk and eggs and in the in-line analysis of phenylurea herbicides in vegetable samples [23], thiabendazole in fruits [24] and fluoroquinolones in soils [25] The aim of this study was to develop a selective analytical 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 particles 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 combined with LC-MS/MS to detect PPTs 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) Ultrapure 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, mL) was purchased from Waters Co (Milford, MA, USA) 2.2 Standards and stock solutions The reference standards of VCM, BTCA and TEC were purchased from National Institutes for Food and Drug Control (Beijing, China) Daptomycin, PME and PMB (containing two major components of PMB1 and PMB2) were obtained from Dr Ehrenstrofer GmbH (Augsburg, Germany) Enrofloxacin, sulfadimidine and virginiamycin 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 solvent as follows: daptomycin (DAP), enrofloxacin, sulfadimidine and virginiamycin (VGM) dissolved with MeOH; VCM and TEC with water; PMB, PME and BTCA with 0.1% FA aqueous solution Each stock solution should be stored at -20 °C for a maximum of 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 precipitation polymerization according to our previous study [26] After removing the template, the MIP particles were washed with water and MeOH for three times and dried in an oven at 60 °C under vacuum for 24 h An amount of 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 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 water 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 theoretical plates number (N) of both MIP and non-imprinted (NIP) columns were calculated as following equations: k= tR − t0 t0 where, tR (min) is the retention time of polymyxin E; t0 (min) is the void time of solvent peak N = 5.54 tR w 1/2 = 16 tR w X Song, E Turiel, J Yang et al Journal of Chromatography A 1673 (2022) 463192 where, w1/2 (min) is the peak width at half peak height; w (min) is the peak width at peak base 200 ng/mL) were prepared to plot calibration curves Mean recoveries 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 calculated 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 homogenization, g samples were accurately weighed into a 50 mL polypropylene centrifuge tube and spiked with appropriate standard working solutions for incubation at room temperature for 30 The extraction was performed with mL of ACN-10% TCA in water (1:1, v/v) through sonication for and shaking for 20 After 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 mL of the extract was collected 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 min, the target fraction of analytes was started and collected into a mL test tube until a run time of 10.5 Then, the eluate fraction of analytes at their retention time was collected and evaporated to dryness at 45 °C Finally, the residues were redissolved with 0.2 mL of the above reconstitution solution and analyzed by LC-MS/MS 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 imprinted column during in-line MISPE Generally, in-line MISPE consists of three steps [23–25]: first, during the loading step, an organic solvent (eg ACN) is prioritized as the mobile phase, since the co-extracted fat-soluble impurities are eluted rapidly by the organic mobile phase, while target analytes are retained well on the MIP column with specific recognition Thereafter, a mixture of water and organic solvent is commonly used as the washing solution to reduce the interference of complex matrix but without disrupting 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 selectively bound to MIP column whereas impurities co-extracted were removed as much as possible, the in-line MISPE conditions (loading, 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 important 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 Accordingly, MeOH was discarded for further experiments and the effect of the presence of ACN in the mobile phase on the retention/separation of polymyxin E was studied It was observed that polymyxin E was completely retained onto the MIP column, making 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 hydrogen bonding interactions as mentioned above 2.6 HPLC-UV and LC-MS/MS conditions The chromatographic performance of MIP column and corresponding 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 elution program: 0∼6 for loading, 100% A; 6.1∼7 for washing, 100% B; 7.1∼11 for eluting, 100% C; 11.5∼16 for reconditioning 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 animal 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 spectrometer were used to detect the analytes Chromatographic and mass conditions such as mass parameters, mobile phase and elution 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 analytes The mobile phase consisted of 0.1% FA in ACN solution (A) and 0.1% FA in water solution (B) with the following gradient elution: min, 6% A; min, 6% A; 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 compound were monitored under the ESI+ mode and the mass parameters are given in Table S1 3.2 Chromatographic performance and selectivity of MIP column The chromatographic parameters affecting MIP column performance, including background pressure, stability and retention factor, 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 pressures in different solvents are in the order: water>MeOH>ACN The highest is the water with a background pressure of 125 bar, indicating that the MIP column could tolerate 100% aqueous phase The reproducibility of retention times, thus the performance stability of MIP column, was assessed by injecting the standard solution of polymyxin E for eight times in a row and the chromatograms obtained are shown in Figure S1-B The retention time 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 Matrixmatched standard solutions (0.5, 1, 2, 5, 10, 20, 50, 80, 100 and X Song, E Turiel, J Yang et al Journal of Chromatography A 1673 (2022) 463192 Fig 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 HPLC-UV chromatograms of polymyxin E on MIP column and NIP column Chromatographic conditions: see Experimental section was 9.90 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 investigated 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 with serious peak tailing The theoretical plate number 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 parameters 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 consumption 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 negligible tailing, while enrofloxacin and sulfadimidine were not retained with retention times lower than This result confirmed 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 revealed that specific binding sites on the MIP contribute to the retention of analytes, allowing the determination of several polypeptides simultaneously 3.3 Preparation of animal tissue extracts Proteins in animal tissues could strongly interact with polypeptide 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 X Song, E Turiel, J Yang et al Journal of Chromatography A 1673 (2022) 463192 Fig 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 proportions 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 sample 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 reconstitution solution was evaluated It was observed that the recoveries 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 Fig 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 column Chromatographic conditions: see Experimental section 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 shows the LC-UV chromatogram at 205 nm obtained in the analysis of non-spiked and spiked pork sample at the concentration of 200 μg kg−1 As can be observed, the mixture of polypeptide antibiotics was unambiguously 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 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 X Song, E Turiel, J Yang et al Journal of Chromatography A 1673 (2022) 463192 Table Recovery and precision of polypeptide antibiotics in animal muscles (n = 5) Compound Vancomycin Teicoplain A2-1 Teicoplain A2-2&2-3 Polymyxin B1 Polymyxin B2 Bacitracin A a Spiked (μg kg−1 ) Average recovery (RSD) a , % Beef Pork Intra-batch Inter-batch Intra-batch Inter-batch Chicken Intra-batch Inter-batch LOQ 50 100 200 LOQ 50 100 200 LOQ 50 100 200 LOQ 50 100 200 LOQ 50 100 200 LOQ 50 100 200 71.3(9.7) 78.1(11.3) 89.0(3.2) 83.9(3.3) 69.1(9.9) 75.4(9.6) 81.0(14.3) 82.4(6.4) 69.5(10.2) 68.2(11.1) 89.4(13.2) 87.2(6.4) 71.0(9.1) 79.1(4.9) 85.2(1.7) 79.6(2.4) 88.1(7.5) 80.0(6.6) 89.0(3.0) 83.4(2.6) 71.1(4.5) 83.5(5.2) 89.6(2.7) 84.6(2.8) 69.7(10.8) 84.4(8.9) 88.9(12.2) 86.9(6.2) 66.7(13.4) 76.0(14.3) 86.7(10.3) 84.2(7.5) 73.5(12.6) 75.7(16.0) 86.7(11.8) 82.3(9.1) 81.9(6.6) 85.3(7.9) 83.9(12.3) 86.3(5.4) 79.7(9.9) 87.2(7.4) 89.9(11.1) 87.5(3.5) 74.5(11.6) 88.1(4.9) 88.1(9.5) 90.8(3.0) 68.1(8.5) 83.0(6.6) 80.6(5.4) 89.2(2.3) 83.8(7.8) 82.9(6.7) 77.4(7.6) 77.4(7.1) 77.0(5.8) 69.4(6.7) 83.9(4.4) 81.7(8.8) 86.5(4.0) 81.3(4.2) 81.6(10.8) 80.8(10) 86.5(10.5) 84.3(4.6) 84.6(8.4) 82.6(12.4) 84.1(3.4) 83.7(6.8) 78.0(8.9) 83.9(6.1) 68.2(6.2) 81.3(7.2) 83.6(6.0) 82.1(10.6) 81.5(7.7) 85.1(7.5) 77.0(8.2) 85.0(10.6) 79.7(6.3) 71.9(8.2) 78.2(8.2) 81.2(12.3) 80.8(10.7) 83.6(4.7) 85.6(6.9) 81.3(8.0) 79.4(15.0) 83.8(3.5) 87.0(5.7) 82.8(9.6) 83.9(6.0) 84.6(5.4) 83.5(7.5) 85.7(8.2) 75.5(11.6) 79.9(9.1) 88.1(5.6) 82.2(6.4) 72.1(13.8) 81.8(12.1) 85.7(9.1) 85.5(7.1) 71.6(9.5) 76.7(13.5) 90.8(11.2) 86.5(7.2) 79.5(11.9) 78.6(5.6) 84.0(3.2) 80.1(4.4) 85.9(11.1) 80.5(6.0) 88.3(4.2) 83.4(3.9) 74.1(7.3) 82.8(3.8) 88.5(2.9) 83.9(2.8) 73.5(11.5) 78.8(5.9) 85.7(11.2) 90.5(3.3) 66.6(11.1) 76.1(11.5) 94.5(2.8) 85.9(7.5) 77.4(13.0) 75.2(11.8) 83.1(7.3) 84.3(3.7) 86.6(4.5) 90.0(3.0) 85.1(10.5) 88.4(4.7) 84.3(4.6) 85.4(2.4) 86.8(9.2) 85.2(3.0) 69.2(13.9) 89.6(3.0) 87.2(1.5) 89.2(2.1) RSD, relative standard deviation 3.6 Matrix effect sample clean-up However, due to the slight differences observed in the retention times (see Fig ), it was not possible to resolve target analytes when the mixture was injected into the MIP column 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 levels 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 Matrix effect (ME) caused by the co-extracted impurities could suppress or enhance the mass spectrum response of analytes under 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 fitting equation of pure standard solutions (without matrix) The positive 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 exhibited slight signal suppression as the matrix effects were soft (less than ±0∼20%) Many studies confirmed that strong signal suppression for polypeptides was observed in biological sample matrices such as chicken (-41∼-52%) [2], pork (-24∼-86%) [13] and fish (