Pii: s0731-7085(98)00268-4

Journal of Pharmaceutical and Biomedical Analysis Determination of tetracycline and its major degradation products by liquid chromatography with fluorescence A. Pena a,*, A. Carmona b, A. Barbosa b, C. Lino a, I. Silveira a, B. Castillo b a Laboratory of Bromatology, Faculty of Pharmacy, Uni6ersity of Coimbra, 3000 Coimbra, Portugal b Laboratory of Instrumentals Techniques, Department of Analytical Chemistry, Faculty of Pharmacy, Complutense Uni6ersity, Received 15 May 1998; received in revised form 4 September 1998; accepted 13 September 1998 Abstract
A liquid chromatographic method of tetracycline and its major degradation products on a C -reversed phase column with acidic mobile phase and fluorescence detection is described. The quantification limit, measured as theamount of sample that gave a signal ten times the peak-to-peak noise of the baseline, was: 0.25 ng for tetracycline(TC) and epitetracycline (ETC), 25 ng for and 4-epianhydrotetracycline (EATC) and 50 ng for anhydrotetracycline(ATC) of injected standard. By means of this liquid chromatography (LC) assay TC, ETC, EATC and ATC as maindegradation products of tetracycline, can be separated and determined with good sensitivity and specificity within 15min. 1998 Elsevier Science B.V. All rights reserved.
Keywords: Tetracycline; Epitetracycline; Epianydrotetracycline; Anydrotetracycline; Liquid chromatography; Fluorescence detection 1. Introduction
tetracycline (ATC) and 4-epianhydrotetracycline(EATC). Tetracycline used in feed additives are Tetracycline antibiotic is widely prescribed in even less pure. The source of the tetracycline used animal husbandry. It is used for the prevention in feeds is the residual tetracycline found in the and control of disease and as feed additives to dried, ground mycelial powder harvested from promote weight gain and increase feed conversion Veterinary pharmaceutical formulations of te- stored under adverse conditions, e.g. high temper- tracycline (TC) contain small amounts of impuri- ature and humidity [2]. It must be realized that ties namely 4-epitetracycline (ETC), anhydro- degradation products of tetracyclines can alsooccur in the stomach [3].
It is important to determine not only the tetra- * Corresponding author. Fax: + 351-39-27126.
cycline and its major degradation products in 0731-7085/98/$ - see front matter 1998 Elsevier Science B.V. All rights reserved.
PII: S 0 7 3 1 - 7 0 8 5 ( 9 8 ) 0 0 2 6 8 - 4 A. Pena et al. / J. Pharm. Biomed. Anal. 18 (1998) 839 – 845 pharmaceutical formulations but also in biologi- pseudo-first-order kinetics, leading to ATC at cal and food samples. Next to a decrease of very low pH. The epimerization of ATC, and the potency, degradation can lead to toxic degrada- dehydration of the ETC lead to the formation of tion products. This is already proven for EATC The toxic effects of ATC were attributed to the Permitted concentrations of these impurities in relative position of the dimethylamino group (on pharmaceutical formulations are fixed by the Eu- Several papers dealing with the liquid chro- Since the residues in biological and food sam- matographic determination of tetracyclines and ples, are products of metabolism they consist of their degradation products have been published the parent drug and all compounds derived from [11,12]. Those reports deal mainly with the deter- it, such as free metabolites. The maximum residue mination of tetracyclines in pharmaceutical prepa- levels (LMR) in foods for TC, established by the rations, where relatively high concentrations are European Community (EC) [7] are for the parent involved, but it is also important determine their compound and its epimer. Therefore, it is also presence in biological and food samples at resid- very important to develop methods for its deter- ual levels. Therefore, it was essential to introduce The stability of TC is poor under strong acidic Fluorescence detection of tetracyclines is more and alkaline conditions with reversible formation specific and also in many cases more sensitive than UV detection normally used [13], and its a (ETC) in weak acid (pH 3) and to anhydro-TC important tool in the analysis of tetracycline under strong acidic (below pH 2) conditions [8].
residues in biological and food samples.
Epimerization on carbon-4 in tetracycline, In this paper a LC method with fluorescence which is a reversible first-order process, occurs detection, according to the Haagsma method [14] most rapidly between pH 3 and 5. The presence of validated in our laboratory for the residue analy- a hydroxyl group at C favors dehydration and sis of oxytetracycline (OTC), tetracycline (TC) aromatization of the C-ring of tetracycline follow and chlortetracycline (CTC), was tested for the Fig. 1. Structures of tetracycline and its major related substances.
A. Pena et al. / J. Pharm. Biomed. Anal. 18 (1998) 839 – 845 analysis of TC, ETC, EATC and ATC. The pre- phases in well-ventilated chemical fume hood.
sented method utilizes the fluorescence produced when tetracyclines reacted with magnesium ions.
1. Water was purified by distillation and passage Most tetracyclines fluorescence in the presence through Milli-Q system (Millipore). The water of magnesium ions, and this fluorescence is in- was filtered through a 0.45 mm filter under tensified by the addition of a base such as vacuum and degassed by ultrasonication.
2. All the solvents were LC grade and were pur- Tetracycline was found to fluorescence more if chased from Carlo Erba (Italy). Oxalic acid, heated to produce anhydro-salts before complex- magnesium acetate, boric acid, potassium hy- ation [16]. During dehydration, two double droxide and sodium hydroxide were analytical bonds are added to the tetracycline nucleus, reagent grade chemical obtained from Merck producing stable anhydro forms and increasing fluorescence. Therefore, this property can be 3. Tetracycline (TC) was obtained from Sigma used for a more sensitive detection of the anhy- dro forms, measured as anhydro-TC – magne- purchased from European Pharmacopoeia. In- dividual stock standard solutions of TC, ETC, which gives fast separation and determination with good sensitivity and selectivity, of TC, into a volumetric flask, and was stored at − 20°C in brown glass vials for a maximum tetracycline, suitable for their precise routine period of one month. Tetracycline was cor- analysis in biological and food samples.
The working solutions were a mixture of the four compounds prepared by a serial dilution of 2. Experimental
the stocks and were stored in brown glass vials at4°C. These solutions were prepared immediately The mobile phases used for analysis containing model 7125 loop injector (Rheodyne, Cotati, aqueous oxalic acid solution (pH 2.0; 0.01 M) and CA) and a Perkin Elmer LS-3B fluorescence de- 20 – 40% of acetonitrile. The mobile phases were tector operated at an excitation wavelength of filtered through a 0.45 mm filter under vacuum, 385 nm and an emission wavelength of 500 nm were used. The spectral band width was 10 nm The reagent post-column was prepared daily, for both excitation and emission. Mobile phase dissolving 6.0 g of magnesium acetate, 6.1 g boric flow 0.8 ml min−1. The derivatization reagent acid and 2.5 g of potassium hydroxide in 950 ml was delivered at a flow rate of 0.45 ml min−1.
of water, adjusted with 1 M sodium hydroxide The results were recorded on a 3390A chro- and make up to 1 l. Its important to follow the order of addition of these reagents, because pre- column used was a Chromspher C , 100-3 mm cipitations may occur. This solution was filtered through a 0.45 mm filter under vacuum and de- gassed by ultrasonication, was held in a brownglass bottle.
All glassware was cleaned with Extran MA 03, Merck, Germany; 10% v/v, rinsed in concentrated Caution: tetracyclines are irritants and tetracy- acid-dichromate solution, washed thoroughly with cline itself is a possible teratogen. Handle tetra- tap water, rinsed with deionized water and dried cyclines standards with care. Prepare mobile A. Pena et al. / J. Pharm. Biomed. Anal. 18 (1998) 839 – 845 Table 1Linear regression of the assay response for TC, ETC, EATC and ATC 3. Results
these solutions: they were always kept at 4°C andprotected from light, immediately after the injec- The standard curves were tested for linearity in tion in the chromatographic system, they were not the range of 2.5 – 25 ng for ETC and TC, and allowed to stand in the laboratory at room tem- 50 – 250 ng for EATC and ATC, of injected perature. Under these conditions no appreciable decomposition was observed in these solutions for The linear regressions of the assay response for approximately 1 working day (8 – 12 h).
TC, ETC, EATC and ATC are shown in Table 1.
The good repeatability mentioned in the cali- The correlation coefficients for the regression bration is an indication for the good stability of this compounds during the chromatographic greater than 0.993, showing the standard curve to be linear within the range of standards used.
The quantification limits, measured as that amount of sample that gave a signal ten times the 4. Discussion
peak-to-peak noise of the baseline, were: 0.25 ngfor TC and ETC, 25 ng for EATC and 50 ng for The quantification limits achieved by the present method are in agreement with the MRL The quantification of small amounts of impuri- established by the European Comminity for TC ties can be realized owing to the good separation and its epimer in foods and in the same chromato- graphic run we can also averiguate the presence of the toxic EATC and ATC. Since the anhydro- analysing five standard solutions at two levels of TC – magnesium complexes are highly fluorescent concentration (ETC and TC at 25 and 2.5 ng and the method is more sensitive for these compounds EATC and ATC at 250 and 50 ng), whilst be- permitting the separation of very small amounts tween-day precision were determined by analysing the same standard solutions on 5 successive days.
can be separated from TC, according to the Eu- We obtained coefficients of variation between 2.4 ropean Pharmacopoeia permitted concentrations and 3.5% and 9.0 and 4.1% respectively, showing Chromatography of anhydrotetracyclines is The individual stock standard solutions of TC, some what more difficult to achieve than that the ETC, EATC and ATC were prepared in methanol parent compounds because of the lower polarity and stored at − 20°C in brown glass vials. We did not observe significant alterations over a maxi- A study on reversed-phase LC of TC and its mum period of 1 month. The working solutions degradation products using acid mobile phases were a mixture of the four compounds prepared was first published by Knox et al. [17].
by a serial dilutions of the stocks with methanol, The method presented utilizes a C Chromo- in brown glass vials, and were prepared daily, spher column at room temperature with acetoni- immediately before use. Some care was taken with trile – 0.01 M oxalate buffer (pH 2) as the mobile A. Pena et al. / J. Pharm. Biomed. Anal. 18 (1998) 839 – 845 phase. Tetracyclines form chelate complexes with and ATC altering the concentration of organic ions at i-diketones (C –C ) and carboxyamide (C ) [18] and bind with silanol groups in the We have tried several modifications of the per- stationary phase [19]. In an acidic medium (pH centage of acetonitrile in the mobile phase, so 1 – 2.5) the tetracycline molecule is fully proto- those anhydrotetracyclines eluted after the tetra- nated and exists in its cationic form [20,21] and cycline with good resolution, permitting also the can be paired with a suitable anion such as ox- separation of tetracycline and his epimer.
alate [22]. On the other hand, all tetracyclines had The isocratic analysis using acetonitrile – oxalic the best asymmetric values at pH 2 [23].
acid solution (pH 2.0; 0.01 M) (30:70, v/v) as Following elution, tetracyclines are derivatized mobile phase, allows the separation of the four with Mg2+ ions at room temperature to produce a highly fluorescent derivative. The fluorescence With 20% of acetonitrile the resolution between detector is set with an excitation wavelength of 385 nm and an emission wavelength of 500 nm.
eluted very late with bad resolution.
The development of the fluorimetric method An increase of the acetonitrile concentration to was based upon experience obtained with the 40% finally enable the elution of EATC and ATC analysis of OTC, TC and CTC. The fluorescence but with this mobile phase ETC overlapped with response is dependent upon pH. For a maximum fluorescence a pH greater than 8 is essential, and Several variations of the ratio of acetonitrile in is associated with the ionized form of the pheno- the acid mobile phases did not yield substantially lic-i-diketone site of these molecules. The pH is better separations, and some improvement was adjusted with the addition of the post-column achieved by the use of gradient elution as we can reagent, prepared in alkaline solution at pH 9.
observe in Fig. 2, showing greater resolution of The fluorescence response is also dependent these compounds. The gradient was chosen based upon the flow rate of derivatization reagent. In in this work experience, to allow for optimum our study the fluorescence response reached a separation. We have applied the follow gradient: maximum at 0.45 ml min−1. Post-column deriva- from 0 to 5 min 20% of acetonitrile and 80% of tization does not directly affect the chromato- oxalic acid (pH 2.0; 0.01 M), from 5 to 16 min graphic properties of the tetracyclines, however 40% of acetonitrile and 60% of oxalic acid (pH the reaction chemistry must be rapid on the chro- 2.0; 0.01 M) and at 17 min 30% of acetonitrile matographic time scale in order to preserve the and 70% of oxalic acid (pH 2.0; 0.01 M), at a Post-column derivatization has also the advan- tage that a separate sample treatment step is notrequired and the analytes are better separated 5. Conclusions
from interferences prior to derivatization.
Preliminar work was carried out on isocratic Methanol was chosen as the universal solvent analysis, making several variations of the ratio of for the tetracyclines, because of its ability to organic modifier and aqueous oxalic acid in the dissolve the tetracyclines and its miscibility with aqueous and organic solvents. Methanol was also The composition of acid mobile phases must be chosen because aqueous solvents tend to acceler- established very well, because complete separation ate degradation of tetracyclines compounds.
of TC and ETC is obtained only with mobile Because many reversed-phase materials are un- phases which are too weak to eluted the more stable at pH lower than 2 – 3 it was necessary to strongly retained EATC and ATC within a rea- flush the columns with a neutral solvent (e.g.
sonable time. The retention decrease directly with water – acetonitrile 50:50) for 1 h at the end of concentration of acetonitrile in the mobile phase.
each working day [22]. This practice contributed We can obtain shorter retention times of EATC markedly to the prolonging of the column life.
A. Pena et al. / J. Pharm. Biomed. Anal. 18 (1998) 839 – 845 The epimer of TC is always eluted first, while suitable for analysing the impurities of TC, show- its anhydro-forms are most retained in agreement ing better resolution (R ) values, we obtained with the polarity of the tetracycline compounds.
good symmetrical peaks using a Chromspher C Accordingly to polarity of the chromatographic column. Our results show that even though a C column was used, we obtained good performance.
eluted with good resolution only with a high This method enables good separations of TC, percentage of acetonitrile in the mobile phase.
With this LC method we can differentiate be- The methodology reported herein use a simple tween the tetracycline and their major degrada-tion products, at residue level within 15 min. Thus solvent system containing a low concentration of it is possible to ascertain whether the tetracyclines buffer, avoiding the drawbacks related with its use determined in biological and food samples are the for the chromatographic system. Its also allows intact molecule originally applied or some fluores- analysis of the tetracyclines and its degradation cent degradation and food products. The high products with the same chromatographic column.
degree of selectivity achieved in using a post- Although the analysis was performed at pH 2, column reaction and fluorescence detection, that this system was proven not to cause any epimer- is less prone to interference from other com- ization of TC. Formation of EATC and ATC due pounds in the sample matrix, has considerable to partial degradation of ETC and TC in the potential as a basis for the development of a strong acidic mobile phase also was not observed.
method for the determination of these compoundsin biological and food samples, since we will canminimize the complex time consuming sampleextraction and clean-up procedures.
References
[1] R.B. Ashworth, J. Assoc. Anal. Chem. 68 (1985) 1013 – [2] R.F. Lindawer, D.M. Cohen, K.P. Munnely, Anal.
[3] H.J.E.M. Reewijk, V.R. Tjaden, J. Chromatogr. 353 [4] H. Oka, M. Suzuki, J. Chromatogr. 314 (1984) 303 – 311.
[5] A. Azalos, Chromatographia 10 (1985) 313 – 323.
[6] European Pharmacopeia, 3rd edn., 1997.
[7] Commission Regulation no. 281/96, Off. J. Eur. Com- mun., L37/9 – L37/11, 14 February, 1996.
[8] H. Oka, H. Nakagawa, K.-I. Harada, J.D. Mac Neil, in: AOAC Int. (Eds.), Chemical Analysis for Antibiotic usedin Agriculture, AOAC Int., 1995, pp. 332 – 346.
[9] N.H. Khan, P. Wera, E. Roets, J. Hoogmartens, J. Liq.
Chromatogr. 13 (1990) 1351 – 1374.
[10] H.F dos Santos, W.B.de Almeida, M.C. Zerner, J.
[11] N.H. Khan, P. Wera, J. Hoogmartens, J. Liq. Chro- [12] C. Hendrix, E. Roets, J. Crommen, J de Beer, E. Por- queras, W. Van der Bossche, J. Hoogmartens, J. Liq.
Chromatogr. 16 (1993) 3321 – 3329.
Fig. 2. Chromatogram in gradient analysis indicated. ETC-Tr [13] J.H. Knox, J. Jurand, J. Chromatogr. 110 (1975) 103 – 3.56; TC-Tr 4.75; EATC-Tr 12.67 and ATC-Tr 13.83.
A. Pena et al. / J. Pharm. Biomed. Anal. 18 (1998) 839 – 845 [14] R.J. McCracken, W.J. Blanchflower, S.A. Haggan, D.G.
[19] C. Bogert, A.M. Kroon, J. Pharm. Sci. 70 (1981) 186 – Kennedy, Analyst 120 (1995) 1763 – 1766.
[15] N. Haagsma, P. Scherpenisse, in: N. Haagsma, A. Ruiter, [20] C.R. Stephens, K. Murai, K. Brunings, R.B Woodward, J.
P.B. Czedik-Eysenberg (Eds.), Proceedings of the Eu- Am. Chem. Soc. 78 (1956) 4155 – 4158.
roresidue II Conference Veldohoven, 1993, pp. 342 – 346.
[21] J.H. Knox, J. Jurand, J. Chromatogr. 186 (1979) 763 – 782.
[16] H. Poiger, Ch. Schlatter, Analyst 101 (1976) 808 – 814.
[22] F. Kramer-Hraczynska, J. Chromatogr. Sci. 29 (1991) [17] D. Hall, J. Pharm. Pharmacol. 28 (1976) 420 – 422.
[23] H. Oka, K. Uno, K.-I. Harada, K. Yasada, M. Suzuki, J.
[18] Y.Y. Lee, W. Evrett, J. Am. Chem. Soc. 103 (1981) 5221.

Source: https://estudogeral.sib.uc.pt/bitstream/10316/5822/1/file1698b204e7a341d0b6865c058380b3cb.pdf