Food Anal. Methods (2011) 4:497–504 DOI 10.1007/s12161-011-9195-3
Extraction and Determination of Oxytetracycline Hydrochloride and Oxolinic Acid in Fish Feed by Derivative Spectrophotometry of First Order M. Inés Toral & Sandra L. Orellana & César A. Soto & Pablo Richter
Received: 4 October 2010 / Accepted: 5 January 2011 / Published online: 18 January 2011 # Springer Science+Business Media, LLC 2011
Abstract In this work is proposed the extraction and determination of oxytetracycline (OTC) and oxolinic acid (OA) in fish feed by first-derivative spectrophotometry. The extractions are carried out by parallel modality, where OTC is extracted in the presence of OA in a sample, and in another is extracted OA in the presence of OTC, and the sequential modality, where OTC is extracted first and then followed by OA in a single feed sample. A phosphate buffer, pH 7.0, was selected for OTC extraction and acetonitrile for OA extraction. These solvents were used in both extraction modalities. The extraction percentages obtained by parallel mode are better than those obtained by sequential extraction. In both cases, the limits of extraction were 25 mg kg−1 for OTC and 10 mg kg−1 for OA. However, it is proposed to work with the parallel extraction for its efficiency, accuracy, precision, and less time requirement. Keywords Oxytetracycline . Oxolinic acid . Extractions . Derivative spectrophotometry . Feed
M. I. Toral (*) : S. L. Orellana Department of Chemistry, Faculty of Sciences, University of Chile, 653, Santiago, Chile e-mail:
[email protected] C. A. Soto Department of Inorganic and Analytical Chemistry, Faculty of Chemical Sciences, University of Concepción, 160-C, Concepción, Chile P. Richter Department of Inorganic and Analytical Chemistry, Faculty of Chemical and Pharmaceutical Sciences, University of Chile, 233, Santiago, Chile
Introduction Oxytetracycline (OTC), an antibacterial broad-spectrum antibiotic belonging to the family of tetracyclines (TCAs), is widely used in aquaculture. Its mechanism of action is to inhibit the synthesis of bacterial proteins being fixed to 30S ribosome subunit (Chopra and Roberts 2001). Its structure presents a hydronaphthacene skeleton containing four fused rings, showing acid–base properties due to the presence of amino and hydroxyl groups. In general, chelating properties of TCAs are due to the presence of ketone and enol groups, and their antibacterial activity and pharmacokinetic properties are influenced by the chelation of metal ions (Arias et al. 2005). Oxolinic acid (OA) is another synthetic antibiotic used regularly in the salmon industry. It belongs to the family of quinolones, which have in common the presence of a 4-quinolone-3-carboxylate ring. The first action site described for these antibiotics was the bacterial enzyme DNA gyrase, a bacterial topoisomerase II. Quinolones inhibit some of the catalytic activities of DNA gyrase in bacteria (Hooper 1999). In this context, quinolones exert their toxicity on the bacterial cell, stabilizing the double strand of DNA that has been broken by the DNA gyrase so that the subsequent ligation cannot occur. Also, they show acid–base properties by possessing carboxylic acid group with a pKa close to 7.0 (Luetzhoft et al. 2000; Turiel et al. 2003). In the last decades, a significant and progressive increase in the use of antibiotics has been observed for the aquaculture production due to increased rate of bacterial diseases and their use as a prophylactic agent. The use of medicated feed (pellets) plays an important role in the salmon production usually because the antibiotic is delivered in the feed. The formulation of the basic feed for fish is protein, carbohydrates, lipids, water, fiber, cations, and
498
ash (Rigos et al. 2004). OTC and OA are added separately or together in fish feed, and the ratio variation depends on the disease, fish prophylaxis, and nutritional needs. However, the regulatory requirements of different countries are considered. In fish feed, the common dosages of OTC and OA are between 55 to 220 mg kg−1 and 10 to 25 mg kg−1, respectively (Aoyama et al. 1991; Ellingsen et al. 2002). Several methods have been reported for individual or simultaneous determination of these drugs in animal muscle tissue (Hernández-Arteseros et al. 2002; Pouliquen et al. 1997; Oka and Matsumoto 2000), biological fluids (Oka and Matsumoto 2000), and environmental matrices (Pouliquen et al. 2007; Halling-Sørensen et al. 2003); however, in the feed matrices have only been reported in the individual determination of OTC (Fernández-González et al. 2002; Wang et al. 2005) and OA (Saad et al. 2002), or in combination with other drugs (Thiex and Larson 2009; Wang et al. 2008; Weng et al. 2003; Caballero et al. 2002; Galarini et al. 2009; Pecorelli et al. 2003), but a simultaneous determination between OTC and OA has not been reported. These reported methods used various analytical techniques for the determination of these antibiotics, including TLC (Weng et al. 2003), capillary electrophoresis (Saad et al. 2002), and spectrofluorimetry (Fernández-González et al. 2002; Wang et al. 2005), HPLC (Pouliquen et al. 1997; Halling-Sørensen et al. 2003; Wang et al. 2008; Galarini et al. 2009), and also, they used sophisticated instrumentation of high cost and specific detectors such as LC–MS–MS (Blanchflower et al. 1997). The aim of this study is to develop a simple method and easy to implement in a control laboratory for the extraction and determination of OTC and OA in fish feed using derivative spectrophotometry in order to eliminate mutual interference among the drugs. Counting on these alternative methods will bring advantages such as the decrease in the number of samples that must be evaluated by more sophisticated method such as LC or ICP, thereby, saving costs and time.
Food Anal. Methods (2011) 4:497–504
with each solvent. Other concentration ranges were prepared by appropriate dilution with the same solvent. Apparatus A Shimadzu UV-1603 spectrophotometer with 10 mm quartz cells was used for absorbance and derivative absorption spectra measurement. For all solutions, the first-derivative spectra were recorded on a range of 190–500 nm versus solvent, using: slit wide values 2.0 nm, sampling intervals 0.2 nm, and scan speed of 480 nm min−1. The spectral data were processed by Shimadzu software kit ver. 3.7 (P/N 20660570-04). An analytical balance, R200D, from Sartorius was used for weightings. A centrifuge, Rotofix 32, from Hettich Zentrifugen was used in all extraction process. The apparatus Orion Digital Research Ion-Analyzer 701, Vacuum filter System Millipore, containing a Glass 47-mm Filter Holder, XX10 047 00, vacuum-filtering flask of 1 L, and vacuum/pressure pump, 220 V, 50 Hz, XX55 220 50, with filter of pore size 0.45 μm, was used to separate the solids from liquids using Whatman filter paper no. 5C of cotton cellulose. Ultrapure water system device Milli-Q was used to obtain the water used in this work. Procedure for Parallel and Sequential Extraction in Simulated Mixtures with Different Mass Ratios An enriched fish feed sample was prepared starting from 20 g of blank feed with 5 mg of each of the drugs (250 mg kg−1 of OTC and OA), which was homogenized in Ultra-Turrax T-25 for 5 min. Then, extraction is performed in parallelly or sequentially according to the following procedure (Scheme 1). Different fractions of the sample were weighed and added to various quantities of blank feed in order to obtain solid dilutions with concentrations between 125 and 25 mg kg−1. The samples were prepared in OA/OTC ratios between 1:1 and 1:6. The extraction processes were carried out using 5 g of different samples.
Materials and Methods Reagents and Standards All chemicals were analytical reagent grade. Na2HPO4, NaH2PO4, EDTA, and solvents were purchased from Merck®. OTC hydrochloride 95% and OA 98.9% were purchased from Sigma-Aldrich®. Buffer solution of sodium dihydrogen phosphate/sodium monoacid phosphate pH 7.0 was prepared by dissolving 2.71 g of Na2HPO4 and 3.71 g of NaH2PO4 in 1,000 mL of deionized water. Stock solutions of OTC and OA 5.0×10−4 mol L−1 were prepared by dissolving 23.00±0.01 mg of OTC in buffer solution HPO4−2/H2PO4−0.1 mol L−1, pH 7.0, and 13.05± 0.01 mg of OA in acetonitrile and then diluted to 100 mL
Procedure for Simultaneous Determination of OTC and OA in Simulated Mixtures with Diverse Mass Ratios For each sample, five replicates were made. Then, the firstderivative spectra of each replicate were evaluated, and the concentration of each compound was calculated, allowing the determination of the corresponding values for recovery and relative standard deviation (RSD). Procedure of Photostability of OTC and OA Photostability studies were carried out using OTC and OA stock individual solutions with a concentration of 4.0×10−5 mol L−1 for each compound in buffer phosphate (pH 7.0) and
Food Anal. Methods (2011) 4:497–504
499
SEQUENTIAL EXTRACTION 5 g Sample of fish feed that containing “x” mg OTC and “y” mg OA was homogenized in a mortar during 1h. Then the following steps were carried out.
1. Add 20 mL of buffer solution HPO4-2 / H2PO4– 0.1 mol L-1 pH 7.0 and 4 mL of .EDTA 0.1 mol L-1. 2. Shake in magnetic stirrer for 40 min. 3. Centrifuge at 1.395 g for 5 min and filter with vacuum. The obtained supernatant contains OTC.
4. To residual solid, add 25 mL of acetonitrile. 5. Shake in magnetic stirrer for 40 min. 6. Centrifuge at 1.395 g for 5 minutes and filter with vacuum. The obtained supernatant contains OA.
PARALLEL EXTRACTION 5 g Sample of fish feed containing “x” mg OTC and “y” mg OA.
5 g Sample of fish feed containing “x” mg OTC and “y” mg OA.
1. Add 20 mL of buffer solution HPO4-2 / H2PO4– 0.1 mol L-1 pH 7.0 and 4 mL of EDTA 0.1 mol L-1. 2. Shake in magnetic stirrer for 40 min. 3. Centrifuge at 1.395 g for 5 min and filter with vacuum. The obtained supernatant contains OTC.
1. To sample add 25 mL of acetonitrile. 2. Shake in magnetic stirrer for 40 min. 3. Centrifuge at 1.395 g for 5 min and filter with vacuum. The obtained supernatant contains OA.
Scheme 1 Procedure for sequential and parallel extraction of OTC and OA from samples of fish feed. In all cases, the UV–Vis spectra and its first derivatives for both drugs were obtained. Then, the
acetonitrile (ACN), respectively. The solutions were exposed to direct light, indirect light, and darkness, and the firstderivative spectra were evaluated nine times during a period of 24 h and then were evaluated every 24 h until 5 days.
derivative units (DU) were obtained at 266.8 nm and 382.2 nm for the OA and OTC, respectively
(a)
H3C OH H3C
The selection of the solvent was carried out considering the following properties of both drugs: polarity, solubility, interaction with the matrix, stability, and spectral behavior of drugs in the solvent. Both drugs are structurally different (Fig. 1); however, both molecules have strong absorption in the ultraviolet–visible spectrophotometry (UV–Vis) due to the presence of aromatic rings and carboxylic groups. OTC presents an extensive π−π * conjugation, due to phenol groups, and a chelating site, which includes the β-diketone system (positions 11 and 12), ceto-enol groups (positions 1 and 3), and carboxamide (position 2) A ring (Chopra and Roberts 2001). This conjugation system has a strong absorption between 200 and 450 nm in the zero-order UV
OH OH
6
5
4
D
C
B
A
10
11
12
1
OH
O
OH
7
8
Results and Discussion
CH3 N
9
3 2
NH2
OH
(b)
O
O O C
O
5
4
6
3
7
O
O
8
1
OH
2
N C 2H5
Fig. 1 Molecular structure for a oxytetracycline (OTC) and b oxolinic acid (OA)
500
spectrum, where two principal bands are resolved at 270 and 360 nm. Similarly, OA also has a strong absorption between 190 and 400 nm, due to 4-oxo-1,4-dihydroquinoline skeleton (4-quinolone-3-carboxylate ring). This is evident in the UV spectrum where there are two principal bands at 270 and 350 nm. According to the literature, OTC presents three pKa values, the first corresponds to the C3-OH (pKa1 3.2), the second to the C4-N (CH3)2 (pKa2 7.5), and the third to the C10-OH (pKa3 8.9; Qiang and Adams 2004), while OA has a pKa value of 6.9. This difference in pKa values, related to the corresponding structures, shows that OTC has a wider range of solubility in polar solvents than OA, which is the basis for selective extraction. Solvent Effect in the Solubility of OTC and OA The effect on the solubility of each antibiotic at 1.0×10−3 mol L−1 was studied in order to provide comprehensive information about the solubility of each antibiotic. The solubility of each antibiotic was separately and qualitatively evaluated in solvents with different polarities (dielectric constant " between 78.5 and 4.0), in acidic, basic, and neutral solutions, according to the experimental conditions. The results are shown in Table 1. In concordance with its structures, the OTC has different functional groups that confer polar and acid–base characteristics; these properties allow this compound to be soluble in most aqueous solutions and polar organic solvents with ε higher than 20. Oppositely, OA presents low solubility in water, methanol, ethanol, acetone, dichloromethane, and ethyl acetate because its polar characteristics are minor. However,
Food Anal. Methods (2011) 4:497–504
in aprotics solvents such as ACN or dimethylsulphoxide (DMSO), OA shows good solubility at room temperature. According to Table 1, in ACN and DMSO, both drugs are soluble. However, DMSO was discarded since it presents absorption bands in the UV–Vis region. Oppositely, in ACN, both drugs present well-defined bands and good linearity (Fig. 2). However, when the extraction of these drugs from the matrix of fish feed was assessed, it was found that only OA is quantitatively extracted in ACN, which indicates that OTC interacts strongly with the matrix components. Based on these results, ACN was selected to extract OA (Fig. 2b). Moreover, according to solubility results, the OTC extraction from feed can be realized using phosphate buffer at pH 4.0 and 7.0 in presence of EDTA solution because in both cases, only OTC is soluble. On the other hand, in this condition, the spectra obtained are the same in the absence and presence of fish feed (Fig. 2c). However, pH 7.0 was selected because OTC stability is favored, the spectra are well-defined, and the absorbance is a function of OTC concentration. In summary, considering the solvent study and the effect of the matrix on solubility and spectral behavior, buffer pH 7.0 was selected for OTC extraction and ACN for OA extraction. Photostability of OTC and OA To establish the photostability of OTC in phosphate buffer pH 7.0 and OA in ACN at room temperature, solutions of each compound were exposed to direct light, indirect light, and darkness during 5 days. All solutions were evaluated by derivative spectra every 3 h in the first 24 h and then
Table 1 Solubility of OTC and OA 1.0×10−3 mol L−1 Solvents Water Dimethylsulfoxide Acetonitrile Methanol Ethanol Acetone Dichloromethane Ethyl acetate Chloroform HCl 1.0×10−2 mol L−1 NaOH 1.0×10−2 mol L−1 NH3 1.0×10−2 mol L−1 NaOH 1.0×10−2 mol L−1 in methanol NaOH 1.0×10−2 mol L−1 in Ethanol Buffer pH 7.0 HPO4−2/H2PO4−in presence of EDTA 0.1 mol L−1 Buffer pH 4.0 HPO4−2/H2PO4− in presence of EDTA 0.1 mol L−1
Dielectric constant (ε), 25 °C
OTC
OA
78.5 48.9 37.5 32.7 24.6 20.7 8.9
Soluble Soluble Soluble Soluble Soluble Soluble Not soluble
Not soluble Soluble Soluble Not soluble Not soluble Not soluble Not soluble
6.0 4.7
Not soluble Not soluble Soluble Soluble Soluble Soluble Soluble Soluble Soluble
Not soluble Not soluble Not soluble Soluble Soluble Soluble Soluble Not soluble Not soluble
– – – – – – –
Food Anal. Methods (2011) 4:497–504
501 1.5
were evaluated every 24 h, until 5 days. In all cases, the derivatives’ spectra were not altered; this fact is indicative that the drugs are stable and are not degraded.
Absorbance
Considering the absorption spectra of OTC and OA in ACN (Fig. 2b) and in phosphate buffer pH 7.0 in the presence of EDTA (Fig. 2c), it is possible to observe that the absorbance value depends on the concentrations of drug. Also, a good linear relationship between the response and concentration in the studied range was found (Table 2). Similarly, it was observed that the absorption spectra of OA in ACN present proportionality between concentration and absorbance. Despite the possibility of using UV–Vis spectrophotometry, the first-order derivative spectrophotometric technique was chosen in order to ensure that there is no interference from traces of a drug against the other and minimize the matrix effect. In addition, with this technique, it is possible to work with moderately turbid samples. This fact is important because the supernatant after centrifugation is cloudy, and spectrophotometric measurement is performed directly in the extract.
OTC
1.0
C
A
1.5
(b)
OTC OA
1.0
D
C
0.5
0.0
Optimization of Spectral Variables
Smoothing Factor Selection. Using the first derivative, the smoothing factor was varied at 2,000, 4,000, 8,000, and 16,000 points. These values are defined by default in the software according to wavelength range, where the classic sweep of the spectrum is realized. When the smoothing factor increases, there is a decrease of the derivative signal, but the noise decreases in a higher extent, and consequently, the signal to noise ratio (S/N) increases. A value of 16,000 was selected because in these conditions, the highest S/N was found, and the bands were sufficiently resolved.
B
0.5
A 200
B 250
300
350
400
450
0.6
(c)
A
0.5
Absorbance
Derivative Order. From zero-order spectra, the first and second derivative spectra of OTC and OA in selected solvents were obtained. In Fig. 3(I) are shown the derivative spectra of these drugs in ACN. In Fig. 3(II) are shown the derivative spectra of the same drugs in phosphate buffer pH 7.0 in presence of EDTA; therefore, the first and second derivatives could be used for analytical purposes. When the derivative order increases, sensitivity concomitantly decreases (Toral et al. 2009). In this context, when the first-order derivative was used, the simultaneous determination of OTC and OA can be easily carried out because the spectrum shows well-defined areas for determining each analyte with high sensitivity. Thus, the first derivative was selected. Higher order derivatives were discarded for not presenting analytical advantages, since the noise increases.
OA
D
0.0
Absorbance
Spectral Behavior and Selection of Spectral Variables
(a)
B
0.4 0.3 0.2
C
0.1
D
0.0 -0.1 200
250
300
350
400
450
500
Wavelength / nm Fig. 2 Zero-order spectra of OTC and OA in acetonitrile. a Absence fish of feed. b Presence of fish feed. OTC: A, B; OA: C, D. Drug concentrations, 2.0 and 4.0×10−5 mol L−1. c Zero-order spectra of OTC and OA in HPO4−2/H2PO4−0.1 mol L−1 buffer solution (pH 7.0) in the presence of EDTA 0.1 mol L−1 and in absence and presence of fish feed. OTC, A absence and B presence; OA, C absence and D presence. Drug concentrations, 2.0×10−5 mol L−1
Selection of the Scale Factor. The scale factor is a function of the software that magnifies the spectrum to obtain a better reading of the analytical signal. This factor was studied between 1 and 108, being 10,000 as the selected amplification factor. It is necessary to highlight that this factor only favors the reading but not
502
Food Anal. Methods (2011) 4:497–504
Table 2 Analytical parameters for OA and OTC
Analytical parameters
OA
OTC
Linear regression equation(C/mol L−1) Correlation coefficient (r) Blank standard deviation (DU) DL/mol L−1 QL/mol L−1 Determination range/mol L−1
DU1 = 1.11×107 C − 5.89 0.9991 2.70 7.290×10−7 2.43×10−6 2.43×10−6 −0.8×10−4
DU2=3.05×106 C − 1.06 0.9992 5.86 5.770×10−6 1.88×10−5 1.88×10−5 −0.1×10−3
wavelengths (λ) were discarded because OTC signals are low. Consequently, the graphical method was used to quantify OTC, selecting a λ value of 382.2 nm, since in the range of 360 to 425 nm, OA shows no signal (DU), and OTC signals present a high sensitivity. According to this analysis, the simultaneous determination of OTC and OA can be carried out at 382.2 and 266.8 nm, respectively. To confirm the wavelengths used, the extractions were performed in parallel mode for both drugs, showing that OTC has a marked effect of matrix in the zero-order spectra; however, in the wavelength range in which both drugs are determined, the signals of first derivative are not altered.
the sensitivity of the method since the noise is also amplified. Selection of Analytical Wavelength. For the first derivative spectra, the following facts were considered for the wavelength selection: the analytical signals must not be affected by the matrix, the signal must be higher (derivative units (DU)), and a high S/N ratio must be obtained. Figure 3(I) shows spectra of each drug in ACN showing a zero crossing of OTC at 266.8 nm for the quantitative determination of OA. For the OTC determination, there are three zero crossing points at 260.1, 288.4, and 328.5 nm (Fig. 3(II)); however, these analytical
100
I (a)
400
OTC OA
266.8nm
200
I (b) OTC OA
50
DU
0 0 -200 -400
-50
-600 -100 400
II (a)
60
II (b)
300
OTC OA
40
OTC OA
200
DU
20 100
382.2 nm
0
0 -100
-20
-200
-40
-300
-60 200
250
300
350
400
450
Wavelength / nm Fig. 3 I Derivative spectra of OTC and OA in acetonitrile. a First derivative. b Second derivative. Concentrations, OTC (2.0 to 10.0×10−5 mol L−1) and OA (1.0 to 5.0×10−5 mol L−1). II Derivative spectra of OTC and OA in HPO4−2/H2PO4−0.1 mol L−1 buffer solution (pH 7.0) in
500
200
250
300
350
400
450
500
Wavelength / nm presence of EDTA 0.1 mol L−1. a First derivative. b Second derivative. Concentrations, OTC (2.0 to 10.0×10−5 mol/L) and OA (1.0 to 5.0× 10−5 mol L−1)
Food Anal. Methods (2011) 4:497–504
600
503
(a)
400 50 mg/kg 100 mg/kg 125 mg/kg
200
200
DU
DU
400
(b)
266.8nm
300 mg/kg 600 mg/kg 750 mg/kg
382.2 nm
0
0 -200
-200
-400
-400 200
250
300
350
400
450
500
200
250
Wavelength /nm
300
350
400
450
500
Wavelength /nm
Fig. 4 Parallel extraction of OTC (phosphate buffer pH 7.0) and OA (in acetonitrile) from samples of fish feed in relation 1:6, respectively. a Determination of OA at 266.8 nm. b Determination of OTC at 382.2 nm
Analytical Parameters Using the first derivative spectra, a smoothing factor of 16,000, and scaling factor of 10,000, calibration graphs were constructed for the determination of OA at λ= 266.8 nm and OTC at λ=382.2 nm. The equations for determination of each drug were obtained using the solution of least squares linear regression (Table 2). The analytical features of the method were obtained according to International Conference on Harmonization (ICH 1997) criteria to calculate the detection limits (DL) and quantifi-
cation limits (QL) which were (3.3 σ/S) and (10 σ/S), respectively; where S is the slope of the calibration curve and σ is the standard deviation for the response of 11 blanks. The range of determination was defined between the QL and the loss of linearity between the DU and the resulting concentration of each drug. The repeatability, expressed as relative standard deviation (RSD), was obtained by using nine standard samples containing each drug at 2.0×10−5, 3.0×10−5, and 4.0×10−5 mol L−1 and three replicates at each concentration. The RSDs were 1.9% and 1.8% for OTC and OA, respectively.
Table 3 Recovery for the determination and extraction of OTC and OA in fish food spiked with the drugs Antibiotic
OTC OTC OA OA OTC OA OTC OA OTC OA OTC OA OTC OA a
n=20
Spiked concentration/mg kg−1
Parallel extractiona recovery (%)
Sequential extractiona recovery (%)
OTC or OA 75 55 25 10 OTC/OA (1:1)
98±4 102±4 98±4 95±5
98±4 97±4 92±4 91±4
106±4 97±5 103±3 98±4
123±4 89±5 110±4 91±4
98±4 97±4
98±4 84±4
98±4 99±5
99±4 80±5
99±4 89±5
99±4 73±5
25 25 50 50 OTC/OA (3:1) 75 25 OTC/OA (7:1) 175 25 OTC/OA (7:1) 75 10
504
Validation and Application of the Method The sequential and parallel extraction modalities for the developed methods were applied in synthetic samples, spiked with OA and OTC in different proportions between 1:1 and 1:6, the latter being the most frequently administered to salmon. In all cases, 20 replicates will be carried out in order to validate the method (Cinquina et al. 2003). The spectra obtained for parallel extraction for OA/OTC (1:6) at different concentrations are shown in Fig. 4. Similar spectra bands were obtained when the sequential extraction was carried out; however, the recovery decreases when the concentration of OTC increases. Table 3 shows the results of the sample in parallel and sequential modes using phosphate buffer pH 7.0 for OTC and ACN for OA showing different relationships. The extraction percentages obtained by parallel mode are better than by sequential extraction. Despite that the extraction limit in sequential extraction for both drugs was 25 mg kg−1, the parallel extraction is proposed for its efficiency, accuracy, and precision, besides that, it requires less time.
Conclusions This work proposes the extraction and simultaneous determination of OTC and OA in fish feed by derivative spectrophotometry. This determination is possible to carry out using extraction by sequential and parallel modes, both are applicable to different OA/OTC ratios used in fish feed. Furthermore, it is possible to establish that the proposed method is an easy and rapid method for the extraction and determination of both drugs. The extraction efficiency is higher in parallel mode and requires less total time in the protocol compared with the sequential method. The determination of antibiotics in fish feed with simple methods and easy to implement is very important because it is necessary to maintain control of the amount of drugs added to the pellet, to avoid development of bacterial resistance to these drugs and possible contamination of the waters around salmon farms. The results of these developed alternative methods could be complementary with those confirmatory methods, a fact by which this research work will deliver a significant scientific contribution, which will allow projecting these methods towards the salmon industry and associate laboratories.
Food Anal. Methods (2011) 4:497–504 Acknowledgements The authors are grateful to the National Fund for Development of Sciences and Technology (FONDECYT), Project 1070905 and 1100103, for the financial support.
References Arias M, García-Falcón M, García-Ríos L, Mejuto J, Rial-Otero R, Simal-Gándara J (2005) J Food Eng 78:69 Aoyama R, McErlane K, Erber H, Kitts D, Burt H (1991) J Chromatogr 588:181 Blanchflower W, McCracken R, Haggan A, Kennedy D (1997) J Chromatogr B 692:351 Caballero R, Torres-Lapasio J, García-Álvarez-Coque M, RamisRamos G (2002) Anal Lett 35:687 Chopra I, Roberts M (2001) Microbiol Mol Biol Rev 65:232 Cinquina A, Longo F, Anastasi G, Giannetti L, Cozzani R (2003) J Chromatogr A 987:227 Ellingsen O, Midttun B, Rogstad A, Syvertsen C, Samuelsen O (2002) Aquaculture 209:19 Fernández-González R, García-Falcón M, Simal-Gandara J (2002) Anal Chim Acta 455:143 Galarini R, Fioroni L, Angelucci F, Tobo G, Cristofani E (2009) J Chromatogr A 1216:8158 Halling-Sørensen B, Lykkeberg A, Ingerslev F, Blackwell P, Tjørnelund J (2003) Chemosphere 50:1331 Hernández-Arteseros J, Barbosa J, Compañó R, Prat M (2002) J Chromatogr A 945:1 Hooper D (1999) Drugs 58:6 Luetzhoft H, Vaes W, Freidig A, Halling-Sorensen B, Hermens J (2000) Environ Sci Technol 34:4989 Oka H, Matsumoto H (2000) J Chromatogr A 882:109 Pecorelli I, Galarini R, Bibi R, Casciarri E, Floridi A (2003) Anal Chim Acta 483:81 Pouliquen H, Gouelo D, Larhantec M, Pilet N, Pinault L (1997) J Chromatogr B 702:157 Pouliquen H, Delépée R, Larhantec M, Morvan M, Le Bris H (2007) Aquaculture 262:23 Qiang Z, Adams C (2004) Water Res 38:2874 Rigos G, Nengas I, Alexis M, Troisi G (2004) Aquat Toxicol 69:281 Saad B, Mohamad R, Mohamed N, Lawrence G, Jab S, Idiris M (2002) Food Chem 78:383 Turiel E, Bordin G, Rodríguez A (2003) J Chromatogr A 1008: 145 Thiex N, Larson R (2009) J AOAC Int 92:2 Toral M, Saldías M, Soto C, Orellana S (2009) Quim Nova 32:257 Wang H, Hou F, Jiang Ch (2005) J Lumin 113:94 Wang L, Yang H, Zhang Ch, Mo Y, Lu X (2008) Anal Chim Acta 619:54 Weng N, Sun H, Roets E, Hoogmartens J (2003) J Pharm Biomed Anal 33:85 ICH (1997) ICH Q2B: validation of analytical procedures: methodology. International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use, Geneva, Switzerland