Journal of Chromatography B, 877 (2009) 991–994

Contents lists available at ScienceDirect

Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb

Determination of lincomycin in urine and some foodstuffs by flow injection analysis coupled with liquid chromatography and electrochemical detection with a preanodized screen-printed carbon electrode Mei-Hsin Chiu a , Hsueh-Hui Yang b , Chi-Ho Liu c , Jyh-Myng Zen a,∗ a

Department of Chemistry, National Chung Hsing University, Taichung 402, Taiwan Department of Research, Buddhist Tzu Chi General Hospital, General Education Center, Tzu Chi College of Technology, Hualien, Taiwan c Animal Health Research Institute Council of Agriculture, Executive Yuan, Taiwan b

a r t i c l e

i n f o

Article history: Received 23 September 2008 Accepted 27 February 2009 Available online 5 March 2009 Keywords: Lincomycin Screen-printed carbon electrode Solid-phase extraction HPLC

a b s t r a c t An electroanalytical method for the determination of lincomycin in feeds, honey, milk and urine was demonstrated in this study. The procedure employed a solid-phase extraction for the isolation of lincomycin from real samples. The antibiotic residues were subsequently analyzed by high-performance liquid chromatography (HPLC) coupled with a disposable electrochemical sensor. The use of a disposable sensor together with the application of solid-phase extraction is attractive in practical application and should be useful in fast screening assay. The electroanalysis of lincomycin was first investigated using a preanodized screen-printed carbon electrode (SPCE*). Note that the SPCE* holds the advantages of low cost and easy to handle. The analytical parameters, such as, preanodization potential, preanodization time, solution pH, detection potential, cartridge, wash solution, elute solution and mobile phase, were further studied in detail. Under optimized conditions, the linear detection range for lincomycin is up to 1 mM (correlation coefficient = 0.999) with a detection limit of 0.08 ␮M (S/N = 3) and a quantification limit of 0.27 ␮M (S/N = 10). The applicability of the method was successfully demonstrated in real sample analysis. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Lincomycin produced from Streptomycin lincolnensis is a wellestablished antibiotic drug used in human and veterinary medicine. It is commonly administered to poultry, dairy cattle and honey bee to cure the infection caused by Gram-positive pathogens. The risk of drug residues present in food products (milk and honey) may cause undesired effects on consumers such as allergic reactions or bacterial resistance. The tolerance levels of lincomycin residue set by the US Food and Drug Administration are 0.34 ␮M for milk and 0.23 ␮M for the edible tissues of chicken and swine. Therefore, it is important to develop an efficient analytical method for the routine analysis of lincomycin in agricultural and food samples. Various methods have been reported for the quantification of lincomycin, including microbiological assay method [1], chemical assay [2], electrochemiluminescence detection [3], gas chromatography [4], high-performance liquid chromatography (HPLC) with UV [5], electrochemical [6,7] and mass spectrometry detectors [8–11]. Since lincomycin, as shown in Fig. 1, possesses only a

∗ Corresponding author. Tel.: +886 4 22850864; fax: +886 4 22854007. E-mail address: [email protected] (J.-M. Zen). 1570-0232/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jchromb.2009.02.066

weak UV absorbance in the low wavelength range (200–220 nm), derivatization is thus required by excluding use of UV or fluorescence detection. A more reliable LC–MS technique, on the other hand, always requires expensive instrumentation and skilled technician. Electrochemical detection holds the potential to develop as screening assays because of its simplicity and low cost. The electroanalysis of lincomycin has been investigated using a variety of electrodes including boron-doped diamond [12,13], multi-wall carbon nanotube modified glassy carbon [14], copper [15] and gold electrodes [6,7]. Nevertheless these electrodes are in general expensive and are thus not designed for single-use purpose. In this study, a disposable electrochemical sensor, i.e., preanodized screen-printed carbon electrode (SPCE*), is developed for the determination of lincomycin with good stability and sensitivity. Note that important advantages of SPCE include low cost (thus disposable), easy for mass production and flexible in design [16,17]. To develop a screening method capable of detecting lincomycin residues in real samples, solid-phase extraction was applied followed by HPLC analysis. After optimizing the experimental parameters, the proposed method possesses many advantages such as low detection limit, fast response, low cost and simplicity. It is successfully applied for the determination of lincomycin in feeds, honey, milk and urine.

992

M.-H. Chiu et al. / J. Chromatogr. B 877 (2009) 991–994

Fig. 1. Structure of lincomycin.

2. Experimental

Fig. 2. Effects of (A) preanodization potential and (B) preanodization time to a bare SPCE on the peak current of 100 ␮M lincomycin.

2.1. Chemicals and reagents Acetonitrile, methanol, sodium phosphate, sodium phosphate monobasic, lincomycin hydrochloride, phosphoric acid, sodium hydroxide and all other chemicals were purchased from Sigma (St. Louis, MO, USA) and used as received. Aqueous solution was prepared with Millipore deionized water throughout this investigation. A 10-mM stock solution of lincomycin was prepared with mobile phase and stored under refrigeration. Carrier and sample solution were prepared by suitable dilution of the stock solution before experiments. 2.2. Apparatus and procedure All experiments were carried out with a CHI 832a electrochemical workstation (Austin, TX, USA). The three-electrode system consists of an SPCE working electrode with a geometric area of 0.196 cm2 , an Ag/AgCl reference (CHI 134, Austin, TX, USA) and a platinum auxiliary electrode. To achieve the SPCE* for each experiment, the SPCE working electrode was treated by the following steps: (1) scanning in the potential range of 0.0 to +1.3 V in 0.1 M, pH 7 phosphate buffer solution (PBS) by cyclic voltammetry and (2) applying +2.0 V for 180 s at the electrode in the same solution. The flow injection analysis (FIA) system consists of a Cole-Parmer microprocessor pump drive (PM-92E, BAS, Austin, TX, USA), a Rheodyne model 7125-sample injection valve (20 ␮l loop) with interconnecting Teflon tube and a Zensor SF-100 thin-layer detecting electrochemical cell specifically designed for SPCE (Zensor R&D, Taichung, Taiwan) [18]. The sample clean-up was achieved by solid-phase extraction on C1, C18 (500 mg/3 ml, CHROM EXPERT, Sacramento, CA, USA) and HLB OASIS column (10 mg/1 ml, Milford, MA, USA). The HPLC separation was performed at room temperature on a Waters analytical column Symmetry C8 (100 mm × 2.1 mm I.D., particle size 5 ␮m).

with a lincomycin standard solution over a concentration range of 1–50 ␮M. Samples were purified using solid-phase extraction method as described in later section. 3. Results and discussion 3.1. Electrochemical behavior of lincomycin In our earlier studies, we have reported exclusive applications of disposable preanodized screen-printed carbon electrode (designated as SPCE*) with improved electrochemical activity [19–23]. During preanodization, the SPCE tends to become more porous together with surface reorientation to generate edge plane sites and surface carbonyl functionalities. The electrocatalytic activity thereby increases by the substantial increase in the generation of edge plane sites. The effects of preanodization potential and preanodization time to the treatment of SPCE on the anodic peak current (ipa ) of lincomycin were first evaluated by cyclic voltammetry. As shown in Fig. 2, the peak current was found to reach a maximum at a preanodization potential of +2.0 V vs. Ag/AgCl and a preanodization time of 300 s. However, taking into account of reproducibility and stability, the optimum preanodization time was chosen as 180 s. Fig. 3 illustrates cyclic voltammograms of 100 ␮M lincomycin with the corresponding background voltammograms in 0.1 M, pH 7 PBS at a bare SPCE and the SPCE*. As can be seen, an anodic oxidation signal is noticed at around +1.1 V on a bare SPCE. As to the SPCE*, the oxidation peak potential (Epa ) shifts negatively to +0.9 V with a large increase in current magnitude due to the electrocatalytic effect at the SPCE*.

2.3. Sample preparation The real samples studied include pig feed, milk, honey and human urine. The pig feed was received from Animal Health Research Institute Council of Agriculture, Taiwan. Milk and honey samples were obtained from a local supermarket. Pig feed was first grounded into powder and a portion (1 g) was weighed into a 50-ml glass jar and mixed with deionized water. After that, the sample was ultrasonicated for 10 min followed by centrifugation for 5 min at 4000 rpm. Then, the supernatant was filtered using a 0.45-␮m filter paper. For other real samples 1 ml (or 1 g) of real samples were mixed with 9 ml deionized water and the solution was directly filtered. For the recovery studies, real samples were spiked

Fig. 3. CV responses for (A) the SPCE and (B) SPCE* in pH 7, 0.1 M PBS in the absence (dash line) and presence (solid line) of 100 ␮M lincomycin at a scan rate of 50 mV s−1 .

M.-H. Chiu et al. / J. Chromatogr. B 877 (2009) 991–994

993

Table 1 Recoveries of real sample obtained by adding standard to the extract of solid-phase extraction. Sample

Spiked (␮M)

Found (␮M) ± S.D.

Feed Honey Milk Urine

100 100 50 50

103.26 100.02 48.82 48.22

a

Fig. 4. Hydrodynamic voltammograms for 20 ␮M lincomycin in 0.1 M, pH 6 PBS carrier solution at a flow rate of 0.4 ml min−1 .

The effects of scan rate and pH on the ipa of lincomycin were further evaluated. A slope of 0.5 from the log(ipa ) vs. log(scan rate) plot suggests that a diffusion-controlled process is involved in the oxidation of lincomycin at the SPCE*. The influence of pH on the oxidation of lincomycin was studied in the pH range of 4.7–8. The Epa was found to shift to a more negative potential with the increase of pH and the slope of close to −59 mV/pH indicating an equal proton/electron transfer process. Although the ipa reaches maximum at pH 7, for highly sensitive determination without losing of reproducibility in the detection of lincomycin, pH 6 was adopted in the following experiment. The advantage of using the SPCE* for the detection of lincomycin is very obvious based on the above study. 3.2. Solid-phase extraction In real sample analysis, the components of feeds, food and urine are usually very complicated and therefore the analytical results may be interfered. Solid-phase extraction is consequently required in the lincomycin assay. Different cartridges and type of wash solution (organic solvent/water) for the retention of lincomycin was evaluated in this study. The C1 column is more effective to achieve sample clean-up by solid-phase extraction than those of C18 and HLB columns. Lincomycin is strongly adsorbed onto the C1 column when the concentration of acetonitrile in wash solutions is lower than 5%. It is therefore promising to use this wash solution to elute salts and other interferences that originated from feeds, foods and urine. On the other hand, increasing concentration of methanol leads to a higher amount of eluted lincomycin. A very good recovery (99.2%, n = 3) was achieved with 100% of methanol. Overall, real samples were effectively purified by using a C1 extraction cartridge. Samples (1 ml) were loaded onto the cartridge pre-conditioned with 2 ml water and 2 ml acetonitrile prior to use. Part of the sample matrix was removed by washing with 2 ml water and 6 ml 5% acetonitrile. Finally, lincomycin was eluted from the cartridge with 1 ml 100% methanol and the extract was evaporated to dryness. The residue was reconstituted with 1 ml mobile phase, acetonitrile:methanol:PBS, pH 6 (5:20:75) prior to analysis.

± ± ± ±

1.21 2.15 0.14 0.17

a

Recovery (%)

103.26 100.02 97.64 96.44

(Found/spiked) × 100%, n = 3.

sidering the S/N ratio and R.S.D. values in analytical application, an applied potential of +0.95 V was selected for FIA experiments. As to the effect of flow rate, the FIA responses increased with the increase of flow rate from 0.1 to 0.6 ml min−1 and started to decrease as the flow rate was higher than 0.6 ml min−1 . Although 0.6 ml min−1 seems suitable for the measurement, at a flow rate of 0.4 ml min−1 the HPLC column shows a better reproducibility. A flow rate of 0.4 ml min−1 was thus used in the subsequent studies. Under the optimum conditions, i.e., detection potential = +0.95 V and flow rate = 0.4 ml min−1 , the amperometric current responses were found to have a linear range between 0.05 ␮M and 1 mM with the slope and correlation coefficient of 0.034 and 0.996 ␮A ␮M−1 , respectively. 3.4. HPLC analysis An examination was made to find how the ratios of water, acetonitrile and methanol in the mobile phase influenced in the separation of lincomycin. The effect on retention time and peak width of lincomycin is obvious. When using acetonitrile:water (10:90) as the mobile phase, the retention time of analyte is too short (68.1 s). On the other hand, substituting acetonitrile with methanol at the same ratio caused a broad peak (peak width 333.5 s) with a poor resolution. Acceptable retention time (265.8 s) and peak width (67.1 s) obtained by choosing a mixture of acetonitrile:methanol:PBS, pH 6 (5:20:75) as the mobile phase. Meanwhile, since organic solvent may change the electrochemical behavior of lincomycin on the SPCE*, the mobile phase of acetonitrile:methanol:PBS, pH 6 (5:20:75) was used as the electrolyte to examine the electrochemical behavior of lincomycin. The fact that virtually the same cyclic voltammogram with good repeatability in the detection of 100 ␮M lincomycin (R.S.D. = 2.84%, n = 10)

3.3. FIA system To optimize the amperometric response, parameters affecting the hydrodynamic behavior of lincomycin including the applied potential and flow rate were studied. Fig. 4 shows the hydrodynamic voltammograms of lincomycin in 0.1 M, pH 6 PBS carrier solution. As can be seen, the responses for lincomycin increase with the increase of applied potential between +0.8 and +1.2 V. However, the higher applied potentials also resulted in higher background currents. Con-

Fig. 5. HPLC chromatograms of (A) lincomycin standard and (B) the top solid trace is for blank, the middle dashed for honey sample spiked with 100 ␮M standard, and the bottom dotted trace for feed sample spiked with 100 ␮M standard. Flow rate = 0.4 ml min−1 , applied potential = 0.95 V, mobile phase = acetonitrile:methanol:PBS, pH 6 (5:20:75).

994

M.-H. Chiu et al. / J. Chromatogr. B 877 (2009) 991–994

Table 2 Recoveries of real sample obtained by adding standard to the real samples. Sample

Spiked (␮M)

Found (␮M) ± S.D.

Feed

50 5 2 1

49.54 4.809 1.937 0.926

± ± ± ±

0.71 0.087 0.016 0.017

99.08 96.17 96.85 92.55

Honey

20 5 2 1

18.77 4.73 1.89 1.040

± ± ± ±

0.14 0.011 0.013 0.003

93.83 94.67 94.68 104.14

Mlik

50 20 5 2 1

51.79 19.73 4.740 1.950 0.940

± ± ± ± ±

0.44 0.13 0.007 0.007 0.009

103.58 98.65 95.08 97.62 94.39

50 20 5 2 1

50.90 19.88 4.93 2.04 1.020

± ± ± ± ±

0.51 0.96 0.015 0.028 0.008

101.79 99.38 98.63 101.89 101.58

Urine

a

a

Recovery (%)

(Found/spiked) × 100%, n = 3.

was observed in aqueous solution clearly indicates that the chosen mobile phase is suitable in electrochemical detection. The linearity of signal for the determination of lincomycin using HPLC coupled with electrochemical detector was next examined. The responses of SPCE* toward lincomycin were found linear in the concentration range up to 1000 ␮M with slope (␮A/␮M), intercept (␮A) and correlation coefficient of 0.0077, 0.0006 and 0.999, respectively. The LOD and LOQ values were calculated as 0.08 ␮M and 0.27 ␮M at signal-to-noise ratios of 3 and 10, respectively. 3.5. Analysis of real samples The developed method was finally verified on the determination of lincomycin in real samples of feeds, honey, milk and urine. The accuracy of the detection was first studied by adding the standard to the extract of solid-phase extraction. The obtained results are summarized in Table 1. As can be seen, the results of good recoveries ranging from 96.44% to 103.26% (n = 3, R.S.D. = 0.29–2.15%) confirm that using the SPCE* for the determination of lincomycin after HPLC in real samples is promising. The accuracy of the solidphase extraction was further checked by adding the standard to the real sample before solid-phase extraction. The results are summarized in Fig. 5 and Table 2. Again, the recoveries of lincomycin from the feeds, honey, milk and urine were satisfactory with values ranging from 92.55% to 104.14% (R.S.D. = 0.14–4.84%, n = 3). The inter-day precision for all the real samples are in the range of 2.31–3.70 (R.S.D. = 2.43–3.81%, n = 27). Most importantly, these results pro-

vide a sufficient evidence for the feasibility of using the solid-phase extraction for the sample preparation in the determination of lincomycin in real samples. In conclusion a rapid and inexpensive method was shown for the determination of lincomycin by solid-phase extraction and FIA/HPLC with electrochemical detection. The method was proved to be suitable for the selective measurements of lincomycin in feeds, honey, milk and urine. The accuracy of the method was demonstrated for lincomycin determination in various real samples with high recoveries. Since the SPCE* holds the advantages of low cost and easy to handle, the proposed approach offers a good possibility for use in screening assay and routine analysis of lincomycin. Acknowledgements The authors gratefully acknowledge financial support from the National Science Council of Taiwan. This work is supported in part by the Ministry of Education, Taiwan under the ATU plan. References [1] T.E. Eble, M.J. Weinstein, G.H. Wagman (Eds.), Antibiotics: Isolation, Separation and Purification, Elsevier, Amsterdam, 1978, p. 231. [2] G.C. Prescott, J. Pharm. Sci. 55 (1966) 423. [3] X. Zhao, T. You, H. Qiu, J. Yan, X. Yang, E. Wang, J. Chromatogr. B 810 (2004) 137. [4] W. Luo, B. Yin, C.Y.W. Ang, L. Rushing, H.C. Thompson Jr., J. Chromatogr. B 687 (1996) 405. [5] J. Olˇsovská, M. Jelínková, P. Man, M. Kobˇerská, J. Janata, M. Flieger, J. Chromatogr. A 1139 (2007) 214. [6] W.R. LaCourse, C.O. Dasenbrock, J. Pharm. Biomed. Anal. 19 (1999) 239. [7] J. Szúnyog, E. Adams, K. Liekens, E. Roets, J. Hoogmartens, J. Pharm. Biomed. Anal. 29 (2002) 213. [8] M. Douˇsa, Z. Sikaˇc, M. Halama, K. Lemr, J. Pharm. Biomed. Anal. 40 (2006) 981. [9] T.S. Thompson, D.K. Noot, J. Calvert, S.F. Pernal, J. Chromatogr. A 1020 (2003) 214. [10] C. Benetti, R. Piro, G. Binato, R. Angeletti, G. Biancotto, Food Addit. Contam. 23 (2006) 1099. [11] T.S. Thompson, D.K. Noot, J. Calvert, S.F. Pernal, Mass Spectrom. 19 (2005) 309. [12] K. Boonsong, S. Chuanuwatanakul, N. Wangfuengkanagul, O. Chailapakul, Sens. Actuators B 108 (2005) 627. [13] O. Chailapakul, P. Aksharanandana, T. Frelink, Y. Einaga, A. Fujishima, Sens. Actuators B 80 (2001) 193. [14] Y. Wu, S. Ye, S. Hu, J. Pharm. Biomed. Anal. 41 (2006) 820. [15] X. Fang, X. Liu, J. Ye, Y. Fang, Anal. Lett. 29 (1996) 1975. [16] J. Wang, Acc. Chem. Res. 35 (2002) 811. [17] J.-M. Zen, A.S. Kumar, in: C.A. Grimes, E.C. Dickey, M.V. Pishko (Eds.), Encyclopedia of Sensors, vol. 9, American Scientific Publisher, 2006, p. 33. [18] C.-T. Hsu, H.-H. Chung, H.-J. Lyuu, D.-M. Tsai, A.S. Kumar, J.-M. Zen, Anal. Sci. 22 (2006) 35. [19] C.-C. Yang, A.S. Kumar, J.-M. Zen, Anal. Biochem. 338 (2005) 278. [20] J.-C. Chen, H.-H. Chung, C.-T. Hsu, D.-M. Tsai, A.S. Kumar, J.-M. Zen, Sens. Actuators B 110 (2005) 364. [21] J.-C. Chen, A.S. Kumar, H.-H. Chung, S.-H. Chien, M.-C. Kuo, J.-M. Zen, Sens. Actuators B 115 (2006) 473. [22] K.S. Prasad, J.-C. Chen, C. Ay, J.-M. Zen, Sens. Actuators B 123 (2007) 715. [23] K.S. Prasad, G. Muthuraman, J.-M. Zen, Electrochem. Commun. 10 (2008) 559.