Online Enrichment Combined with High Performance Liquid Chromatography for Quantitation of Trace-level Chloramphenicol in Milk

Abstract

With growing concerns over food safety, more accurate, simpler, and faster detection methods are urgently required. An online enrichment system with a molecularly imprinted polymer (MIP) for chloramphenicol (CAP) was coupled to HPLC using a flow-injection pump and developed and validated for the analysis of trace levels of CAP in milk. Compared with direct injection into the HPLC, the developed analytical method had a lower limit of detection (0.01 µg L⁻¹) and an enrichment factor of 364 because of the specific adsorption of CAP on the MIP. The linear range of the method was 0.05 µg L⁻¹ to 10 mg L⁻¹ (r₂ > 0.99). After online purification and enrichment, the trace amounts of CAP in milk could be quantified, with limit of detection in milk of 0.1 µg kg⁻¹ and recoveries ranging from 82.6% to 98.1% at different spiked levels of CAP in raw milk, and commercial liquid milk.

Introduction

Chloramphenicol (CAP) is a broad spectrum antibiotic that is widely used in animals for the treatment of diseases because of its low cost, high potency, and activity (Schirmer and Meisel 2006). The use of CAP may lead to drug residues being present in animal source foods (Huang et al., 2006). CAP can cause aplastic anemia and gray syndrome in humans. Its use is prohibited in food-producing animals in many countries, including China and the United States, and it is on the European Union's group “A” list with a “zero tolerance residue limit” in edible tissue. With growing concerns over food safety, more accurate, simpler, and faster detection methods are urgently required. Many analytical methods have been developed for effective detection of CAP residues in foods using biological techniques such as microbiological (Singer and Katz 1985), enzymatic (Yamato et al., 1990), immunological (Shen and Jiang, 2005; Tao et al., 2016) assays and instrumental analyses such as sensor (Hummert et al., 1995, Park et al., 2004; Ebarvia et al., 2015), and chromatographic (Analytical and Bioanalytical ChemistryAllen 1985, Posyniak et al., 2003; Bogusz et al., 2004; Huang et al., 2006) techniques. Traditional biological techniques are strict with the detection environment (temperature, cleanliness, etc.) and usually require skillful operators. Further-more, biological techniques have inevitable deficiencies in automated analysis and widespread popularity. Chromatographic methods using LC (Gude et al., 1995; Posyniak et al., 2003), GC (Pfenning et al., 2000; Ding et al., 2005), GC-MS (Nagata and Oka, 1996; Impens et al., 2003; Shen and Jiang, 2005), and LC-MS (Huang et al., 2006; Wang et al., 2008; Pan et al., 2015; Gaugain et al., 2016; Kikuchi et al., 2017) have been reported for the determination of CAP residues in animal tissues. The most common method is HPLC equipped with UV detection and LC-MS. Although LC-MS can provide good sensitivity and accuracy, it is not available in all laboratories because the equipment is expensive and requires skilled operators. Although HPLC is a good choice for routine analysis, its sensitivity is not exceptional (Wang et al., 2008). Consequently, ongoing research has focused on improving the sensitivity of HPLC. Pretreatment methods such as liquid-liquid extraction (Russell et al., 2000; Nováková et al., 2004; Choi et al., 2006) and solid-phase extraction (SPE) (Laganà et al., 2004; Quintana et al., 2004; Weigel et al., 2004; Veach et al., 2015; Armenta et al., 2016), can be used to improve the limit of detection (LOD). However, conventional liquid-liquid extraction methods are time-consuming and requires large volumes of organic solvent.

Molecularly imprinted polymers (MIPs) have been developed and used for sample pretreatment to improve the sensitivity for detecting CAP (Song et al., 2014; Amjadi et al., 2016). Microspheres based on MIPs prepared by aqueous suspension polymerization have been used as offline SPE sorbent combined with HPLC-UV for detection of CAP in milk and shrimp, with the improved sensitivity but the LOD was not reported (Shi et al., 2007). A hybrid MIP with surface-enhanced Raman spectroscopy was reported for the detection of CAP in milk, with a LOD of 0.1 µg mL⁻¹ (Xie et al., 2017). A magnetic sorbent based on a MIP was prepared for the extraction of CAP from honey samples, and effectively lowered the LOD to 0.047 µg kg⁻¹ when combined with HPLC-MS (Alizadeh et al. 2012). All of these methods use offline SPE coupled to analytical apparatus, and differ in their advantages and limitations regarding their specificity, sensitivity, interference of matrix compounds, and expense of the apparatus.

The aim of this research was to develop a simple, reliable, and sensitive method to detect trace levels of CAP in milk. A MIP specific for CAP was synthesized and used as a SPE sorbent to establish a new analytical method through online coupling of SPE to HPLC.

Materials and Methods

Chemicals and materials    CAP, 4-vinyl pyridine (4-VP), ethylene glycol dimethacrylate (EGDMA), thiamphenicol, florfenicol, sulfadiazine and sulfanilamide oxazole were obtained from Sigma-Aldrich (St. Louis, MO, USA). 2, 2-Azobisisobutyronitrile, tetrahydrofuran (THF), acetic acid, acetonitrile, acetone methanol and ethyl acetate were obtained from Merck (Darmstadt, Germany). All reagents were of the highest available purity and of at least analytical grade. Water (18 MΩ cm⁻¹) was prepared using a Milli-Q Academic Water System (Millipore, Billerica, MA, USA) and used throughout this work. Raw milk was obtained from a farm in Beichen (Tianjin, China). Pasteurized milk and ultra high temperature treated milk were bought from a local supermarket (Tianjin, China).

Instruments    A HPLC system (Shimadzu, Kyoto, Japan) with an UV detector was used for the separation and quantitation of CAP. A Model FIA-3110 flow injection system (Jitian Instruments, Beijing, China) was used for online pre-concentration. Polytetrafluoroethylene tubes (0.5 mm) were used for all connections. A Cary 50-Bio UV spectrophotometer (Varian, New Mexico, USA) was used to measure the adsorption capacity of CAP-MIP.

Synthesis of the CAP-MIP    The CAP-MIP was synthesized by adding CAP (324 mg), 4-VP (0.21 mL), EGDMA (3.8 mL), and THF (5 mL) to a 100-mL tube. After sonication for 2 min, the tube was sealed and kept in the dark for 22 h. Next, the initiator 2, 2-azobisisobutyronitrile (60 mg) was added. Sonication was performed for an additional 30 min and nitrogen was bubbled through the solution for 15 min. Then, the tube was sealed and placed in a water bath at 50°C for 48 h. The polymer was Soxhlet-extracted with methanol to remove unreacted reagents and vacuum-dried at 65°C for 8 h, then crushed and sieved through a 200-mesh sieve. The rigid polymer was sonicated twice with 100 mL of methanol/acetic acid (9:1, v/v) for 15 min each time, and then dried in a vacuum oven at 65°C for 8 h. A non-imprinted polymer was synthesized by the same method but without CAP. The schematic was shown in Fig. 1-A.

Fig. 1.

Schematic of the instruments for online enrichment combined with high performance liquid chromatography for studying CAP. (A) CAP-MIP. (B) Load position of the six-port injector valve. (C) Injection position of the six-port injector valve.

Adsorption capacity and selectivity tests    CAP, thiamphenicol, florfenicol, sulfadiazine, and sulfanilamide oxazole were used to evaluate the adsorption and selectivity of the prepared CAP-MIP. Thirty milligrams of the MIP or nonimprinted polymer (NIP) was mixed with each of standard solution of the five antibiotics (15 mg L⁻¹, 5 mL), and the mixture was shaken for 1 h. The analytes remaining in the supernatant were quantitated by UV spectrophotometry at their maximum absorption wavelengths.

Online MIP-SPE-HPLC detection of CAP    To evaluate the applicability of the synthesized CAP-MIP for online SPE-HPLC detection of CAP, a cylindrical micro-column (1.5 cm × 4 mm i.d.) filled with 120 mg of the CAP-MIP was assembled. To ensure uniformity of the packing, the cartridge was preconditioned with water and methanol in turn at a flow rate of 1.0 mL min⁻¹ for 4 h. Online SPE pre-concentration was coupled to HPLC for detection of trace levels of CAP (Fig. 1-B and C). First, 25 mL of the sample solution was loaded onto the SPE micro-column with loading flow rate of 2.6 mL min⁻¹ while the HPLC injector valve was in the “load” position. CAP was concentrated by the CAP-MIP and non-target substances were sent to the waste (Fig. 1-B). Next, the six-port injector valve was switched to the “inject” position, CAP was back-eluted by the mobile phase at a flow rate of 1.0 mL min⁻¹ for 80 s, and then separated by the chromatographic separation column (Fig. 1-C). HPLC separations were performed on a C₁₈ analytical column (25 cm × 4.6 mm, 5 µm, Thermo Scientific, MA, USA) at 35°C, using an isocratic elution with acetonitrile/water (25:75, v/v) at a flow rate of 1.0 mL min⁻¹. The detection wavelength was 278 nm. The areas of the peaks at 11.5 min were used for the calculations.

Sample preparation    Milk (5 mL) was pipetted into a 15-mL centrifuge tube and centrifuged at 4000 ×g at 15°C for 10 min. The fat layer was discarded. Ethyl acetate (3 mL) was added and the tube was vortex mixed at 800 rpm for 2 min. The upper layer was pipetted into a new tube and dried under a stream of nitrogen gas. The residue was dissolved in 5 mL of an acetonitrile/water mixture (25:75, v/v). The resulting solution was filtered through an organic microfiber filter (0.22 µm) and then transferred into a 25-mL calibrated flask and diluted with water at pH 4.5 for the analysis.

For recovery studies, aliquots (5 mL) of milk in centrifuge tubes were spiked with different volumes of a CAP solution in methanol. Each sample was thoroughly mixed and let stand for at least 15 min, and then treated using the procedure detailed above.

Results and Discussion

Optimization of the CAP-MIP synthesis conditions    The MIP was designed as an effective and specific sorbent for CAP in solution. It was also designed to tolerate the high pressure in HPLC to allow for online enrichment and quantitation of trace-level CAP. Based on screening experiments, 4-VP and EGDMA were selected as the functional monomer and cross-linker, respectively, for CAP. Many factors can affect the adsorption ability of the synthesized MIP, including the ratios of CAP to 4-VP and CAP to EGDMA, the reaction temperature and time, and the solvent. We investigated methanol, acetonitrile, acetone, and THF as the solvent. Molar ratios of 1:1, 1:2, 1:4, and 1:8 for CAP to 4-VP, and 1:5, 1:10, 1:20, and 1:30 for CAP to EGDMA were trialed to optimize the CAP-MIP synthesis conditions. The adsorption abilities of polymers synthesized under different conditions were compared using static adsorption tests. The adsorption of CAP to the MIP was optimized when the molar ratio of CAP, 4-VP, and EGDMA was 1:2:20 (Fig. 2), THF was used as the solvent, and thermal activation was performed in a water bath at 50°C for 48 h.

Fig. 2.

(A) Effects of different solvents on the adsorption of CAP with a molar ratio of CAP:4-VP: EGDMA of 1:2:30. (B) Effects of different molar ratios of CAP to 4-VP on the adsorption of CAP with a molar ratio of CAP:EGDMA of 1:30 and THF as the solvent. (C) Effects of different molar ratios of CAP to EGDMA on the adsorption of CAP with a molar ratio of CAP to 4-VP of 1:2 and THF as the solvent.

Adsorption and selectivity tests (Table 1) showed that compared with the NIP, CAP-MIP specifically adsorbed CAP rather than its analogues thiamphenicol and florfenicol. The CAP-MIP did not specifically adsorbed sulfadiazine and sulfanilamide oxazole. Therefore, the prepared MIP was highly specific to CAP. The adsorption capacity of the CAP-MIP was 2.27±0.07 mg g⁻¹.

Table 1. Specificity of CAP-MIPs

Analytes	The adsorption ratio (%) (n=3)
	CAP-MIP (mean±SD)	NIP (mean±SD)
Chloramphenicol	90.8±2.8	21.1±1.5
Thiamphenicol	59.2±2.1	25.9±1.7
Florfenicol	25.1±0.9	23.3±1.2
Sulfadiazine	21.5±0.7	20.8±0.9
Sulfanilamide oxazole	20.9±0.6	26.2±0.7

Optimization of the enrichment conditions    To establish an analytical method for online enrichment and rapid detection of CAP, the prepared CAP-MIP was packed into a micro-column and coupled to HPLC with a flow injection pump.

Influence of pH on the enrichment    The pH played an important role in the enrichment effect. To ensure efficient adsorption of CAP, sample solution pH values between 2.0 and 9.0 were investigated. The pH was adjusted with HCl or NaOH. When the pH was increased from 2.0 to 4.5, the peak area for CAP increased. Further increases in the pH resulted in decreases in the peak area. These results validated our hypothesis that the pH could affect the polymer's surface electric charge and also determine the molecular form of CAP in solution. The optimum pH for the loaded sample was 4.5.

The effect of the sample loading flow rate on enrichment    Theoretically, the sample flow rate is an important factor for the enrichment efficiency, with lower flow rates resulting in higher enrichment because the analyte has sufficient time to contact with the MIP surface. We optimized the loading flow rate using a CAP standard solution (1 µg L⁻¹) and flow rates between 1.8–4.8 mL min⁻¹. We found that the sample loading flow rate did not greatly affect the enrichment. Taking into consideration the adsorption efficiency and analysis time, a sampling loading flow rate of 2.6 mL min⁻¹ was chosen for further studies.

Optimization of the elution conditions    Elution is the most important procedure in online enrichment. Traditionally, two six-point valves and two pumps are used in online analysis for elution of the target analyte from the sorbent by the mobile phase. The purpose of this research was to simplify the components used for elution. In our set-up, we used a flow injection instrument coupled with HPLC with no additional components. For this application, the composition of the mobile phase was optimized to maintain the elution efficiency and optimum peak shape. At higher organic solvent ratios, the CAP peak eluted earlier, which would cause many problems in actual samples. If the ratio of organic solvent was too low, the peak shape was poor. After optimization, an acetonitrile/water ratio of 25:75 (v/v) was selected for further studies.

The elution time must be optimized for elution efficiency and method sensitivity. Our results showed that the CAP peak area increased as the elution time was increased up to 80 s, and no improvement was observed with further increases in the elution time. Therefore, 80 s was chosen as the optimum elution time.

The online enrichment capacity of the CAP-MIP    The loading volume affects adsorption, and larger loading volumes can improve the adsorption capacity but also reduce the limit of detection (LOD). To reduce the analysis time, a loading volume of 25 mL was used in this study. Under the optimum conditions (sample pH = 4.5, sample flow rate = 2.6 mL min⁻¹, and elution time = 80 s), we measured the adsorption capacity. CAP solutions with different concentrations were loaded onto a column packed with 120 mg of the CAP-MIP. The peak area of CAP increased with increasing CAP concentration and was still increasing when the concentration had reached 10 mg L⁻¹. Considering the predicted CAP content in real samples, it was clear that there was no need to improve the adsorption capacity. The adsorption capacity of 120 mg of the MIP for CAP was calculated as 250 µg.

Analytical quality control parameters    After optimization of the analytical conditions, serial dilution of CAP was investigated using the established system. After online analysis of CAP standard solution, the LOD was calculated as 0.01 µg L⁻¹ based on baseline noise near the analyte peak (3 times of the signal-to-noise ratio). The sensitivity of our online enrichment method was 364 times that of direct injection (20-µL sample) into the HPLC. This could be attributed to the high enrichment capacity of the MIP and large sample loading volume.

To test the linear range of the established analytical method, various concentrations of CAP (0.05 µg L⁻¹, 0.5 µg L⁻¹, 5 µg L⁻¹, 50 µg L⁻¹, 1 mg L⁻¹, and 10 mg L⁻¹) were analyzed to obtain calibration curves. A good correlation coefficient (0.996) was obtained for the linear regression curve, and the relative standard deviations ranged from 3.62% to 6.89%. Compared with the offline enrichment method, online enrichment has higher sensitivity and precision, which results from the simplified sample preparation process and reduction in detecting manipulation. Hence, the linearity of online enrichment is better than that achieved in comparable offline enrichment methods. Tests on the application lifetime of the online enrichment microcolumn showed that it could be reused more than 10 times (data not shown).

Analysis of milk samples    Raw milk, commercial ultra-high temperature treated milk, and pasteurized milk were analyzed using the developed method. CAP was not detected in these samples (Fig. 3). The LOD calculated as three times of the signal-to-noise ratio of the baseline near the target peak in blank milk was 0.1 µg kg⁻¹. To study the recovery of the established method, the samples were spiked with CAP at 0.5 µg kg⁻¹, 1.0 µg kg⁻¹, and 10 µg kg⁻¹. Under the optimum conditions, the recoveries were at least 82.6% for the raw milk and commercial milk and the relative standard deviation (RSD) was less than 6.86% (Table 2).

Fig. 3.

Chromatogram from online enrichment of milk samples. (A). Row milk, (B). commercial ultra-high temperature treated milk, and (C). pasteurized milk. spiked with 0.5 µg kg⁻¹ CAP (a) and without CAP (b).

Table 2. Recovery of CAP in three kinds of milk samples (n=3)

Sample	Spiked level (µg kg−1)	Detected content (mean±SD, µg kg−1)	Recovery (%)	RSD (%)
Ultra high temperature treated milk	0.5	0.437±0.03	87.5	6.86
	1.0	0.887±0.046	88.7	5.07
	10	9.81±0.45	98.1	4.59
Pasteurized milk	0.5	0.425±0.025	85.0	5.88
	1.0	0.923±0.03	92.3	3.25
	10	9.75±0.29	97.5	2.97
Raw milk	0.5	0.413±0.027	82.6	6.54
	1.0	0.903±0.035	90.3	3.88
	10	9.54±0.46	95.4	4.82

Conclusions

We synthesized a MIP specific to CAP and successfully applied it to SPE enrichment for online coupling to HPLC and rapid detection of CAP. The established method is much more sensitive than methods employing HPLC with UV detection for the determination of CAP, which generally with a LOD at the micrograms-per-liter level. This performance is a result of online enrichment using the CAP-MIP, with high enrichment efficiency and simple pretreatment process for real samples. The developed method is precise and sensitive for the quantitation of trace levels of CAP in milk samples, and can meet the increasing demand for highly sensitivity detection of the banned drug CAP.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements    The authors are grateful for financial support from the National Key Research and Development Program of China (2017YFC1600402), the Tianjin Municipal Science and Technology Commission (Project No. 16PTSYJC00130), the National Natural Science Foundation of China (Project No. 31501566) and the International Science and Technology Cooperation Program of China (Project No. 2014DFR30350).

References

•  Alizadeh,  T.,  Ganjali,  M.R.,  Zare.  M., and  Norouz.  P. (2012). Selective determination of chloramphenicol at trace level in milk samples by the electrode modified with molecularly imprinted polymer. Food Chem., 130, 1108-1114.
•  Amjadi,  M.,  Jalili.  R., and  Manzoor.  J. L. (2016). A sensitive fluorescent nanosensor for chloramphenicol based on molecularly imprinted polymer-capped CdTe quantum dots. Luminescence, 31, 633-639.
• Analytical and Bioanalytical Chemistry  Allen,  E. H. (1985). Review of chromatographic methods for chloramphenicol residues in milk, eggs, and tissues from food-producing animals. J. Assoc. Off. Anal. Chem., 68, 990-999.
•  Armenta,  S.,  de la Guardia,  M.,  Abad-Fuentes,  A.,  Abad-Somovilla.  A., and  Esteve-Turrilla.  F. A. (2016). Highly selective solid-phase extraction sorbents for chloramphenicol determination in food and urine by ion mobility spectrometry. Anal. Bioanal. Chem., 408, 8559-8567.
•  Bogusz,  M. J.,  Hassan,  H.  Al-Enazi,  E.  Ibrahim.  Z. and  Al-Tufai.  M. (2004). Rapid determination of chloramphenicol and its glucuronide in food products by liquid chromatography-electrospray negative ionization tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 807, 343-356.
•  Choi,  D. W.,  J. Y.  Kim,  S. H.  Choi,  H. S.  Jung,  H. J.  Kim,  S. Y.  Cho,  C. S.  Kang., and  S. Y.  Chan. (2006). Identification of steroid hormones in pomegranate (Punica granatum) using HPLC and GC-mass spectrometry. Food Chem., 96, 562-571.
•  Ding,  S.,  J.  Shen,  S.  Zhang,  H.  Jiang, and  Z.  Su (2005). Determination of chloramphenicol residue in fish and shrimp tissues by gas chromatography with a microcell electron capture detector. J. AOAC Int., 88, 57-60.
•  Ebarvia,  B. S.,  I. E.  Ubando, and  F. B.  Sevill (2015). Biomimetic piezoelectric quartz crystal sensor with chloramphenicol-imprinted polymer sensing layer. Talanta, 144, 1260-1265.
•  Gaugain,  M.,  M.-P.  Chotard,  D.  Hurtaud-Pessel, and  E.  Verdo (2016). Comprehensive validation of a liquid chromatography-tandem mass spectrometry method for the confirmation of chloramphenicol in urine including stability of the glucuronide conjugate and efficiency of deconjugation. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci., 1011, 145-150.
•  Gude,  T.,  A.  Preiss, and  K.  Rubac (1995). Determination of chloramphenicol in muscle, liver, kidney and urine of pigs by means of immunoaffinity chromatography and gas chromatography with electron-capture detection. J. Chromatogr. B Biomed. Sci. Appl., 673, 197-204.
•  Huang,  J.-F.,  H.-J.  Zhang, and  Y.-Q.  Fen (2006). Chloramphenicol extraction from honey, milk, and eggs using polymer monolith microextraction followed by liquid chromatography-mass spectrometry determination. J. Agr. Food Chem., 54, 9279-9286.
•  Hummert,  C.,  B.  Luckas, and  H.  Siebenlis (1995). Determination of chloramphenicol in animal tissue using high-performance liquid chromatography with a column-switching system and ultraviolet detection. J. Chromatogr. B Biomed. Sci. Appl., 668, 53-58.
•  Impens,  S.,  W.  Reybroeck,  J.  Vercammen,  D.  Courtheyn,  S.  Ooghe,  K.  De Wasch,  W.  Smedts, and  H.  De Brabande (2003). Screening and confirmation of chloramphenicol in shrimp tissue using ELISA in combination with GC-MS2 and LC-MS2. Anal. Chim. Acta, 483, 153-163.
•  Kikuchi,  H.,  T.  Sakai,  R.  Teshima,  S.  Nemoto, and  H.  Akiyam (2017). Total determination of chloramphenicol residues in foods by liquid chromatography-tandem mass spectrometry. Food Chem., 230, 589-593.
•  Laganà,  A.,  A.  Bacaloni,  I.  De Leva,  A.  Faberi,  G.  Fago, and  A.  Marin (2004). Analytical methodologies for determining the occurrence of endocrine disrupting chemicals in sewage treatment plants and natural waters. Anal. Chim. Acta, 501, 79-88.
•  Nagata,  T. and  H.  Ok (1996). Detection of residual chloramphenicol, florfenicol, and thiamphenicol in yellowtail fish muscles by capillary gas chromatography-mass spectrometry. J. Agr. Food Chem., 44, 1280-1284.
•  Nováková,  L.,  P.  Solich,  L.  Matysová, and  J.  Ších (2004). HPLC determination of estradiol, its degradation product, and preservatives in new topical formulation Estrogel HBF. Anal. Bioanal. Chem., 379, 781-787.
•  Pan,  X.-D.,  P.-G.  Wu,  W.  Jiang, and  B.-j  M. (2015). Determination of chloramphenicol, thiamphenicol, and florfenicol in fish muscle by matrix solid-phase dispersion extraction (MSPD) and ultra-high pressure liquid chromatography tandem mass spectrometry. Food Control, 52, 34-38.
•  Park,  I.-S.,  D.-K.  Kim,  N.  Adanyi,  M.  Varadi, and  N.  Ki (2004). Development of a direct-binding chloramphenicol sensor based on thiol or sulfide mediated self-assembled antibody monolayers. Biosens. Bioelectron., 19, 667-674.
•  Pfenning,  A. P.,  J. E.  Roybal,  H. S.  Rupp,  S. B.  Turnipseed,  S. A.  Gonzales, and  J. A.  Hurlbu (2000). Simultaneous determination of residues of chloramphenicol, florfenicol, florfenicol amine, and thiamphenicol in shrimp tissue by gas chromatography with electron capture detection. J. AOAC Int., 83, 26-30.
•  Posyniak,  A.,  J.  Zmudzki, and  J.  Niedzielsk (2003). Evaluation of sample preparation for control of chloramphenicol residues in porcine tissues by enzyme-linked immunosorbent assay and liquid chromatography. Anal. Chim. Acta, 483, 307-311.
•  Quintana,  J. B.,  J.  Carpinteiro,  I.  Rodríguez,  R. A.  Lorenzo,  A. M.  Carro, and  R.  Cel (2004). Determination of natural and synthetic estrogens in water by gas chromatography with mass spectrometric detection. J. Chromatogr. A, 1024, 177-185.
•  Russell,  J. A.,  R. K.  Malcolm,  K.  Campbell, and  A. D.  Woolfso (2000). High-performance liquid chromatographic determination of 17β-estradiol and 17β-estradiol-3-acetate solubilities and diffusion coefficients in silicone elastomeric intravaginal rings. J. Chromatogr. B Biomed. Sci. Appl., 744, 157-163.
•  Schirmer,  C. and  H.  Meise (2006). Synthesis of a molecularly imprinted polymer for the selective solid-phase extraction of chloramphenicol from honey. J. Chromatogr. A, 1132, 325-328.
•  Shen,  H.-Y. and  H.-L.  Jian (2005). Screening, determination and confirmation of chloramphenicol in seafood, meat and honey using ELISA, HPLC-UVD, GC-ECD, GC-MS-EI-SIM and GCMS-NCI-SIM methods. Anal. Chim. Acta, 535, 33-41.
•  Shi,  X.,  A.  Wu,  S.  Zheng,  R.  Li, and  D.  Zhan (2007). Molecularly imprinted polymer microspheres for solid-phase extraction of chloramphenicol residues in foods. J. Chromatogr. B, 850, 24-30.
•  Singer,  C. J. and  S. E.  Kat (1985). Microbiological assay for chloramphenicol residues. J. Assoc. Off. Anal. Chem., 68, 1037-1041.
•  Song,  X.,  S.  Xu,  L.  Chen,  Y.  Wei, and  H.  Xion (2014). Recent advances in molecularly imprinted polymers in food analysis. J. Apll. Polym. Sci., 131.
•  Tao,  X.,  S.  Zhou,  X.  Yuan, and  H. L. (2016). Determination of chloramphenicol in milk by ten chemiluminescent immunoassays: influence of assay format applied. Anal. Methods, 8, 4445-4451.
•  Veach,  B. T.,  C. A.  Baker,  J. H.  Kibbey,  A.  Fong,  B. J.  Broadaway, and  C. P.  Drak (2015). Quantitation of Chloramphenicol and Nitrofuran Metabolites in Aquaculture Products Using Microwave-Assisted Derivatization, Automated SPE, and LC-MS/MS. J. AOAC Int., 98, 588-594.
•  Wang,  S.,  W.  Huang,  G.  Fang,  J.  He, and  Y.  Zhan (2008). On-line coupling of solid-phase extraction to high-performance liquid chromatography for determination of estrogens in environment. Anal. Chim. Acta, 606, 194-201.
•  Weigel,  S.,  R.  Kallenborn, and  H.  Hühnerfus (2004). Simultaneous solid-phase extraction of acidic, neutral and basic pharmaceuticals from aqueous samples at ambient (neutral) pH and their determination by gas chromatography-mass spectrometry. J. Chromatogr. A, 1023, 183-195.
•  Xie,  Y.,  M.  Zhao,  Q.  Hu,  Y.  Cheng,  Y.  Guo,  H.  Qian, and  W.  Ya (2017). Selective detection of chloramphenicol in milk based on a molecularly imprinted polymer-surface-enhanced Raman spectroscopic nanosensor. J. Ranman Spectrosc., 48, 204-210.
•  Yamato,  S.,  H.  Sugihara, and  K.  Shimad (1990). An Enzymatic Assay of Chloramphenicol Coupled with Fluorescence Reaction. Chem. Pharm. Bull., 38, 2290-2292.