- Email: [email protected]
Analysis of etimicin sulfate by liquid chromatography with pulsed amperometric detection
Journal of Chromatography A, 1115 (2006) 202–207
Analysis of etimicin sulfate by liquid chromatography with pulsed amperometric detection Lingling Xi a , Genfu Wu b , Yan Zhu a,∗ a
Department of Chemistry, Xixi Campus, Zhejiang University, Hangzhou 310028, China b College of Life Sciences, Zhejiang University, Hangzhou 310058, China
Received 13 January 2006; received in revised form 24 February 2006; accepted 28 February 2006 Available online 4 April 2006
Abstract A new method for determination of etimicin’s (ETM) purity and content is developed by liquid chromatography (LC) and pulsed amperometric detection (PAD). A reversed-phase ion-pair LC method with pulsed amperometric detection on a gold electrode after post-added NaOH is described. The mobile phase consisted of an aqueous solution containing 0.033 mol L−1 oxalic acid, 0.012 mol L−1 heptafluorobutyric acid, and 210 mL L−1 acetonitrile with pH 3.40 adjusting by dilute NaOH solution. The total analysis time was not more than 30 min. The effects of the different chromatographic parameters on the separation were also investigated. A number of commercial samples of etimicin sulfate were analyzed using this method. © 2006 Elsevier B.V. All rights reserved. Keywords: Etimicin; Pulsed amperometric detection; Liquid chromatography
1. Introduction Etimicin (ETM), which is mainly used as the sulfate, is a new semi-synthetic, water soluble aminoglycoside antibiotic obtained by chemical modification of gentamycin C1a (Fig. 1) [1], It is active against both Gram-positive and Gramnegative bacteria, including strains which are resistant to other aminoglycosides and similar as netilmicin [2–4]. The otoand nephro-toxicity of etimicin are substantially lower than those of other aminoglycisides antibiotics and even lower than netilmicin [3,4]. Nevertherless, etimicin still has a narrow therapeutic range and it must be careful to monitor the levels in the blood. Etimicin is the 1-N-ethyl derivative of gentamycin C1a. Therefore, gentamycin C1a can be expected to be contained as a possible impurity in the samples. The 3 -N-ethyl (ETM-1), and 1,3 -N-ethyl (ETM-2) derivatives of gentamycin C1a , and some intermediates such as 3,2 ,6 -N-ethanoyl-gentamycin C1a (P1 ) and 1-N-ethyl-3,2 ,6 -N-ethanoyl-gentamycin C1a (P2 ) can be also formed during synthesis of etimicin and they pos-
∗
Corresponding author. Tel.: +86 571 88273637; fax: +86 571 88273637. E-mail address: [email protected] (Y. Zhu).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.02.093
sessed weakly antibacterial activity [1]. As can be seen, neither etimicin nor its related substances contain a significant UV absorbing chromophore. Microbiological assay which is not able to distinguish between the main components and the impurities in the drug, reversed-phase liquid chromatography with pre-column derivatization with o-phthalaldehyde (OPA) [5] and with 1-fluoro-2,4-dinitrobenzene [6] in which derivatized etimicin was used as internal standard, has been described. However, no LC method has been described to analyze etimicin sulfate as a drug substance and to determine possible impurities. Evaporative light-scattering detection (ELSD) is prescribed in The Ph. Chinese for the determination of etimicin content [7], which was not able to determine possible impurities in the drug, either. Moreover, a volatile mobile phase is required for ELSD detection and it is low sensitive in detection. A wide variety of methods for the analysis of aminoglycoside antibiotics have been published over the years, including microbiogical assay [8], immunoassays [9,10], capillary electrophoresis [11], thin layer chromatography [12], gas–liquid chromatography [13], and high-performance liquid chromatography with pre-column derivatization with detection of UV [14,15], electrochemical [16] and mass spectrometry [17,18]. Etimicin has a structure of aminoglycoside and the methods mentioned above should also be applicable to
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
203
Institute of Microbiology (Wuxi, China). Three commercial samples of etimicin were provided by Hangzhou Aida Pharmaceutical Co. Ltd. (Hangzhou, China), Hailan Aike Pharmaceutical Co. Ltd. (Hailan, China), Changzhou Fangyuan Pharmaceutical Co. Ltd. (Changzhou, China), respectively. Oxalic acid and sodium hydroxide were of analytical grade. Methanol and acetonitrile were of HPLC-grade. 2.2. Apparatus
Fig. 1. Structure of some etimicin components.
it. Since pre-column derivatization is cumbersome, time consuming and gives some problems with quantitation or results in unstable derivatives, electrochemical detection especially pulsed amperometric detection (PAD) is the better, pulsed amperometric detection has been demonstrated for the sensitive detection of numerous aminoglycoside antibiotics [19–24]. But up to now, the determination of etimicin with PAD has not been reported. To our knowledge, no paper has been published describing the composition of commercial etimicin samples. Like other aminoglycoside antibiotics [19–24], etimicin is polybasic cations at low pH, which are high polar species, and it can interact with the ion-pair agents to form ion-pair compounds which are less polar than the parent aminoglycosides and are amenable to separation by the preferred techniques of reversed-phase HPLC. In this work an ion-pair chromatography combined with PAD is described. The mobile phases that were investigated and further optimized. The quality of separation on different stationary phases was compared. Finally, the method has been applied to analyze some commercial samples of etimicin. 2. Experiment 2.1. Reagents, reference substances, and samples Distilled deionizes water of 18.2 M cm was used throughout. Heptafluorobutyric acid (Alfa Aesar, Heysham, Lancs). Standard etimicin and gentamycin C1a were obtained from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). The 3-N -ethyl and 1,3 -N-ethyl derivatives of gentamycin C1a (ETM-1, ETM-2), 3,2 ,6 -N-ethanoyl-gentamycin C1a (P1 ) and 1-N-ethyl-3,2 ,6 N-ethanoyl-gentamycin C1a (P2 ) were obtained from Jiangsu
Chromatographic analyses were carried out using a P680 HPLC Pump (Dionex, Sunnyvale, CA, USA), with a fixed loop of 20 L. Sodium hydroxide was added post-column using a pneumatic device, PC10 Pneumatic controller from Dionex (Sunnyvale, CA, USA). The ZORBAX Rx-C8 column (250 mm × 4.6 mm I.D.) from Agilent (Palo Alto, CA, USA). German Century SIL C18 column (AQ 5 m, 150 mm × 4.6 mm I.D.) from Dianlian Jiangshen Separating Science and Technology Company (Dianlian, China). AcclaimTM 120 C18 column ˚ 150 mm × 4.6 mm I.D.) from Dionex (Sunnyvale, (5 m120 A, CA, USA). The temperature of the column was maintained at 35 ◦ C. ED50A Electrochemical Detector from Dionex (Sunnyvale, CA, USA) was equipped with a gold working electrode with a diameter of 3 mm, an Ag/AgCl reference and stainless steel counter electrode. The cell of the pulsed amperometric detector was placed in the air keeping the temperature at 35 ◦ C. All the instrument control and data collection were performed by a Dionex Chromeleon 6.5. 2.3. Chromatography The mobile Phase is consisted of an aqueous solution containing 0.033 mol L−1 of oxalic acid, 0.012 mol L−1 heptafluorobutyric acid, 210 mL L−1 acetonitrile, adjust the apparent pH to 3.4 with dilute sodium hydroxide solution. Filter the mobile phase through a 0.45 m filter and sonicated before use. The flow rate was 1.0 mL min−1 . All substances to be analyzed were dissolved in the mobile phase. To allow pulsed amperometric detection, 0.52 mol L−1 NaOH was added postcolumn (0.5 mL min−1 ) through a mixing-tee from a nitrogen pressurized reservoir (30 psi) and mixed in a packed reaction coil (Dionex, 375 L), linking to the electrochemical cell. The flow rate for the addition of the base is not critical, but it should be reproducible between runs and must be pulse-free. It was necessary to raise the pH of the mobile phase to approximately 13 to improve the sensitivity of the detection [25]. The 0.52 mol L−1 NaOH solution was made starting from a 2.62 mol L−1 aqueous solution, which was pipetted into N2 degassed water to avoid carbonates that foul the electrodes. For this reason, it is advisable to pipette the NaOH solution from the center of the bottle and to use only two-thirds of the bottle. The time and voltage parameters for the pulsed amperometric detector were set as follows: E1 , E2 , and E3 were, respectively +0.12, +0.70, and −0.60 V with the assigned pulse durations t1 : 0–0.40 s, t2 : 0.41–0.60 s, and t3 : 0.61–1.00 s, integration of the signal was done between 0.20 and 0.40 s.
204
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
Fig. 2. Influence of the pH of the mobile phase on the capacity factors. Sta˚ 150 mm × 4.6 mm I.D.), tionary: AcclaimTM 120 C18 column (5 m 120 A, Methanol was used to determine t0 .
3. Results and discussion 3.1. Selection of chromatographic condition The influence of the different chromatographic parameters on the separation of etimicin and its related substances was evaluated using the retention times and resolution. The influence of the amount of acetonitrile in the mobile phase was examined between 130 and 240 mL L−1 . The retention times decrease on increasing the amount of acetonitrile and small differences in the acetonitrile content have significant consequences. To check the influence of the amount of oxalic acid, the content of oxalic acid in mobile phase was varied and the following contents were used: 0.025, 0.028, 0.033, and 0.036 mol L−1 . In the range examined, the amount of oxalic acid in the mobile phase has no significant effect on the retention times. The influence of the pH of the mobile phase examined using buffer solution with pH 2.5 and 4.5. Little change is observed between pH3.0 and 4.0 (Fig. 2). With further decrease or with further increase in the pH, the retention times decrease. Therefore, an appropriate and acidic mobile phase is necessary because the etimicin molecules must be positively charged to interact with the anionic heptafluorobutyric acid. The influence of the amount of heptafluorobutyric acid was investigated in the range 0.0074–0.015 mol L−1 has no significant effect on the retention, With further increase in the amount of heptafluorobutyric acid, the retention times increase much. Considering the retention times, resolution and sensitivity of detection, the mobile phase consisted of an aqueous solution containing 0.033 mol L−1 oxalic acid, 0.012 mol L−1 heptafluorobutyric acid and 210 mL L−1 acetonitrile, adjust the apparent pH to 3.4 with dilute sodium hydroxide solution. The influence of the column temperature was examined at 35, 40, and 45 ◦ C. As expected, the retention times of the components decrease when the column temperature is increased. However, the resolution between the peaks also decreases with higher column temperatures. Using the developed type of mobile phase above, three columns were examined at 35 ◦ C: ZORBAX Rx -C8 column (250 mm × 4.6 mm I.D.), German Century SIL C18 column (AQ 5 m, 150 mm × 4.6 mm I.D.) and AcclaimTM 120 C18 column
Fig. 3. Voltammetric response at Au electrode in 0.10 M NaOH: scan rate, 100 mV s−1 . (. . .) residual response, (—) 0.50 mg mL−1 etimicin sample.
˚ 150 mm × 4.6 mm I.D.). The latter gave clearly bet(5 m120A, ter overall resolution and was the only column that allowed the separation of etimicin from two other derivatives of gentamycin C1a (ETM-1 and ETM-2). 3.2. Waveform optimization and electrode stability The applied potentials and pulse durations can be chosen to optimize sensitivity, reproducibility, and selectivity for the samples of interest. In most cases, the optimum settings are those, which produce the largest chromatographic peak for the species of interest. Since current is sampled at E1 , the setting for E1 has the largest effect on the current response, and hence the peak height. Current-potential voltammetric response curves were used to screen for potential detector operation conditions. These voltammograms are illustrated in Fig. 3. The oxidation at the gold electrode surface by etimicin components occurred between 0.00 and 0.40 V. The reference solution (0.1 M NaOH) shows significantly less anodic current in this region. The electrochemical response of the reference solution was believed to be due to formation of surface oxide at E > +0.40 V and oxygen evolution at E > +0.70 V. The cyclic voltammetric data indicated that liquid chromatography with pulsed amperometric detection of etimicin should be operated with the Edet potential set between 0.00 and +0.40 V. The detector response was then used to optimize the amperometric response of the PAD unit. In the potential range from 0.02 to 0.37 V, E1 was optimized by varying in steps of 0.05 Volt between sample injections and compared the peak height. Using standard solution of etimicin, the response of the compound was measured as a function of potential Edet . Fig. 4 shows a plot of background, etimicin response versus the measurement potential, Edet . The maximum difference between background response and the response of etimicin occurred at approximately 0.12 V. This was the voltage used for Edet in the final method conditions. E1 = 0.12 V the signal-to-noise ratio of the detection is 3250 which is greater than at other detection voltages. The time and voltage parameters for the pulsed amperometrical detector were set as follows: E1 , E2 and E3 were, respectively
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
205
working electrode after about 60 analyses to obtain a good repeatability. After the electrode is cleaned with fine polishing compound, it is sonicated in water for 10 min. It is advisable to wipe the counter and reference electrodes at the same time with a tissue to remove deposited substances. It takes about 2 h to obtain a stable baseline with a freshly polished electrode. 3.3. Quantitative aspects of the LC method
Fig. 4. Plot of detector response (nC) as measurement potential (V) is varied.
+0.12, +0.70, and −0.60 V with the assigned pulse durations t1 : 0–0.40 s, t2 : 0.41–0.60 s, and t3 : 0.61–1.00 s, integration of the signal was done between 0.20 and 0.40 s. These parameters are the same as used before for similar components, like other aminoglycoside antibiotics [19–24]. Although the sequence of the potentials theoretically cleans the electrode surface, it is necessary to polish the gold
For the determination of the impurities a 20 g sample was examined by injecting 20 L of a 1.0 mg mL−1 etimicin sulfate sample solution (Fig. 5). For this quantity the limit of detection for P1 was 0.02% mg mL−1 (4 ng), as determined at a signal-to-noise ratio of 3. The limit of quantitation was 0.06% mg mL−1 (12 ng) (RSD = 10.3, n = 4). Fig. 6 shows the chromatogram of etimicin and its synthetic raw material GMC1a standard solutions. The linearity of etimicin and its related substances was examined in the concentration range from 0.005 to 0.125 mg mL−1 standard solutions. The results are shown in Table 1. Where y = peak area, x = concentration of standard solutions injected (mg mL−1 ), r = coefficient of correlation. The repeatability was checked by analyzing 0.05 mg mL−1 standard
Fig. 5. Typical chromatogram of a commercial sample (concentration: 1.0 mg mL−1 ) of etimicin sulfate obtained with pulsed amperometric detection. Stationary: ˚ 150 mm × 4.6 mm I.D.). (1) P2 , (2) P1 , (3) GMC1a , (4) ETM-1, (5) ETM-2, (6) ETM, (7) unknown-1, and (8) unknown-2. AcclaimTM 120 C18 column (5 m 120 A,
Fig. 6. Chromatogram of ETM standard solution (concentration: 0.125 mg mL−1 ) and GMC1a standard solutions (concentration: 0.125 mg mL−1 ). (1) GMC1a and (2) ETM.
206
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
Table 1 Linearity of etimicin components standard Components
Range (mg/mL)
Regression equation
r
RSD (%) (n = 6)
P2 P1 ETM-1 ETM-2 ETM GMC1a
0.005–0.125 0.005–0.125 0.005–0.125 0.005–0.125 0.005–0.125 0.005–0.125
y = 411.56x − 0.8613 y = 389.25x − 0.4826 y = 374.29x − 0.5273 y = 427.52x − 1.2458 y = 433.07x − 1.3366 y = 365.17x − 0.2465
0.9983 0.9994 0.9989 0.9992 0.9990 0.9991
0.23 2.54 1.49 0.38 0.86 1.50
Table 2 Composition of etimicin sulfate samples (%), relative to etimicin
4. Conclusions
Samples P2
The method using ion-pair liquid chromatography allowed separating eight components of etimicin. The total time of analysis was not more than 30 min. It is the first time that quantitative results are reported for so many etimicin components with pulsed amperometric detection after post-adding alkali. Pulsed amperometric detection suffers from some stability problems and some experience is required to obtain a good reproducibility, but it allows for sensitive detection and determines the possible impurities without derivatization.
1 2 3
P1
GMC1a ETM-1 ETM-2 Unknown-1 Unknown-2
0.39 0.04 0.57 0.44 0.06 0.39 1.54 0.20 0.45
0.04 0.15 0.25
1.38 1.69 2.60
0.07 0.08 0.07
0.07 0.08 0.07
Table 3 The recovery datum of ETM, GMC1a in the etimicin sulfate samples Samples 1 2 3
P2 (%) 105.23 98.23 103.25
P1 (%) 100.25 99.63 99.45
GMC1a (%)
ETM-1 (%)
ETM-2 (%)
ETM (%)
106.58 105.43 106.88
103.25 101.29 97.58
99.86 98.47 102.41
101.77 100.30 102.33
Acknowledgements
solution of each compound six times, The RSDs of the peak area are shown in Table 1. From the results we can find that the differences between the response factors of etimicin and its related substances negligible. So, the HPLC–PAD method can be applied to the analysis of impurities, the control of the ratio of multi-components drug and the determination of new substances by using another substance as reference, etc. 3.4. Analysis of commercial samples Three commercial etimicin sulfate samples, obtained from different manufactures, were analyzed using the described method. The obtained composition of the samples is shown in Table 2. All minor components are expressed as relative amounts of etimicin, using chromatograms obtained with a 5% (m V−1 ) dilution (0.05 mg mL−1 ) of the examined commercial sample. From the results ETM-2 was found to be the major impurity, which was not detected using other methods. As can be seen, the composition of etimicin samples is dependent on the origin. Different batches from the same manufacturer had a similar composition. Table 3 shows the recovery datum of P2 , P1 , GMC1a , ETM-1, ETM-2, and ETM in the commercial etimicin sulfate samples. Using chromatograms obtained with a 5% (m V−1 ) dilution (0.05 mg mL−1 ) of the examined commercial etimicin sample and 0.05 mg mL−1 standard solutions of P2 , P1 , GMC1a , ETM1, ETM-2, and ETM, respectively.
This is a project supported by National Natural Science Foundation of China (No. 20375035) and by Zhejiang Provincial Natural Science Foundation of China (No. Z404105), and Sponsored by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry and Zhejiang Provincial. References [1] J. Liu, X.-L. Hu, Y.-Q. Shen, J. Fan, M. Zhao, Chin. J. Pharm. 31 (2000) 358. [2] E. Adams, D. Pueling, M. Rafiee, J. Hoogmartens, J. Chromatogr. A 812 (1998) 151. [3] Y. Liao, S.-H. Dai, Chin. J. Antibiotics 20 (1995) 198. [4] Y. Liao, S.-H. Dai, Chin. J. Antibiotics 20 (1995) 365. [5] H.-Q. Shi, J. Wang, Chin. J. Pharm. 34 (2003) 290. [6] M.-J. Zhou, G.-L. Wei, Y.-P. Liu, Y.-M. Sun, S.-H. Xiao, L. Lu, C.-X. Liu, D.-F. Zhong, J. Chromatogr. B 798 (2003) 43. [7] Chin. Pharm. 2 (2005) 740. [8] L.D. Sabath, J.I. Casey, P.A. Ruch, L.L. Stumpf, M.J. Finland, Lab. Clin. Med. 78 (1971) 457. [9] S.A. Bromn, D.R. Newkirk, R.P. Hualer, G.G. Smith, K.J. Sugimoto, Assoc. Off. Anal. Chem. 73 (1990) 479. [10] J.M. Andrews, R.J. Wise, Antimicrob. Chem. 14 (1984) 509. [11] C.L. Flurer, K.A. Wolnik, J. Chromatogr. B 663 (1994) 259. [12] C.P. Bhogte, V.B. Patravale, P.V. Devarajan, J. Chromatogr. B 694 (1997) 443. [13] J.W. Mayhew, S.L. Gorbach, J. Chromatogr. 151 (1978) 133. [14] L.T. Wong, A.R. Beaubien, A.P. Pakuts, J. Chromatogr. 231 (1982) 145. [15] D.M. Barends, J.S.F. Vandersandt, A. Hulshoff, J. Chromatogr. B 182 (1980) 201. [16] L.A. Kaine, K.A. Wolnik, J. Chromatogr. B 674 (1994) 255. [17] L.G. Mclaughlin, J.D. Henion, P.J. Kijik, Biol. Mass. Spectrom. 23 (1994) 417.
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207 [18] M. Cherlet, S.D. Baere, P.D. Backer, J. Mass. Spectrom. 35 (2000) 1342. [19] E. Adams, W. Roelants, R.D. Paepe, E. Roets, J. Hoogmartens, J. Pharm. Biomed. Anal. 18 (1998) 689. [20] D. Debremaeker, E. Adams, E. Nadal, B. Van Hove, E. Roets, J. Hoogmartens, J. Chromatogr. A 953 (2002) 123. [21] E. Adams, R. Schepers, L.W. Gathu, R. Kibaya, E. Roets, J. Hoogmartens, J. Pharm. Biomed. Anal. 15 (1997) 505.
207
[22] E. Adams, J. Dalle, E.D. Bie, I.D. Smedt, E. Roets, J. Hoogmartens, J. Chromatogr. A 766 (1997) 133. [23] E. Adams, R. Schepers, E. Roets, J. Hoogmartens, J. Chromatogr. A 741 (1996) 233. [24] E. Adams, G.V. Vaerenbergh, E. Roets, J. Hoogmartens, J. Chromatogr. A 819 (1998) 93. [25] J.A. Statler, J. Chromatogr. 527 (1990) 244.
Analysis of etimicin sulfate by liquid chromatography with pulsed amperometric detection Lingling Xi a , Genfu Wu b , Yan Zhu a,∗ a
Department of Chemistry, Xixi Campus, Zhejiang University, Hangzhou 310028, China b College of Life Sciences, Zhejiang University, Hangzhou 310058, China
Received 13 January 2006; received in revised form 24 February 2006; accepted 28 February 2006 Available online 4 April 2006
Abstract A new method for determination of etimicin’s (ETM) purity and content is developed by liquid chromatography (LC) and pulsed amperometric detection (PAD). A reversed-phase ion-pair LC method with pulsed amperometric detection on a gold electrode after post-added NaOH is described. The mobile phase consisted of an aqueous solution containing 0.033 mol L−1 oxalic acid, 0.012 mol L−1 heptafluorobutyric acid, and 210 mL L−1 acetonitrile with pH 3.40 adjusting by dilute NaOH solution. The total analysis time was not more than 30 min. The effects of the different chromatographic parameters on the separation were also investigated. A number of commercial samples of etimicin sulfate were analyzed using this method. © 2006 Elsevier B.V. All rights reserved. Keywords: Etimicin; Pulsed amperometric detection; Liquid chromatography
1. Introduction Etimicin (ETM), which is mainly used as the sulfate, is a new semi-synthetic, water soluble aminoglycoside antibiotic obtained by chemical modification of gentamycin C1a (Fig. 1) [1], It is active against both Gram-positive and Gramnegative bacteria, including strains which are resistant to other aminoglycosides and similar as netilmicin [2–4]. The otoand nephro-toxicity of etimicin are substantially lower than those of other aminoglycisides antibiotics and even lower than netilmicin [3,4]. Nevertherless, etimicin still has a narrow therapeutic range and it must be careful to monitor the levels in the blood. Etimicin is the 1-N-ethyl derivative of gentamycin C1a. Therefore, gentamycin C1a can be expected to be contained as a possible impurity in the samples. The 3 -N-ethyl (ETM-1), and 1,3 -N-ethyl (ETM-2) derivatives of gentamycin C1a , and some intermediates such as 3,2 ,6 -N-ethanoyl-gentamycin C1a (P1 ) and 1-N-ethyl-3,2 ,6 -N-ethanoyl-gentamycin C1a (P2 ) can be also formed during synthesis of etimicin and they pos-
∗
Corresponding author. Tel.: +86 571 88273637; fax: +86 571 88273637. E-mail address: [email protected] (Y. Zhu).
0021-9673/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2006.02.093
sessed weakly antibacterial activity [1]. As can be seen, neither etimicin nor its related substances contain a significant UV absorbing chromophore. Microbiological assay which is not able to distinguish between the main components and the impurities in the drug, reversed-phase liquid chromatography with pre-column derivatization with o-phthalaldehyde (OPA) [5] and with 1-fluoro-2,4-dinitrobenzene [6] in which derivatized etimicin was used as internal standard, has been described. However, no LC method has been described to analyze etimicin sulfate as a drug substance and to determine possible impurities. Evaporative light-scattering detection (ELSD) is prescribed in The Ph. Chinese for the determination of etimicin content [7], which was not able to determine possible impurities in the drug, either. Moreover, a volatile mobile phase is required for ELSD detection and it is low sensitive in detection. A wide variety of methods for the analysis of aminoglycoside antibiotics have been published over the years, including microbiogical assay [8], immunoassays [9,10], capillary electrophoresis [11], thin layer chromatography [12], gas–liquid chromatography [13], and high-performance liquid chromatography with pre-column derivatization with detection of UV [14,15], electrochemical [16] and mass spectrometry [17,18]. Etimicin has a structure of aminoglycoside and the methods mentioned above should also be applicable to
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
203
Institute of Microbiology (Wuxi, China). Three commercial samples of etimicin were provided by Hangzhou Aida Pharmaceutical Co. Ltd. (Hangzhou, China), Hailan Aike Pharmaceutical Co. Ltd. (Hailan, China), Changzhou Fangyuan Pharmaceutical Co. Ltd. (Changzhou, China), respectively. Oxalic acid and sodium hydroxide were of analytical grade. Methanol and acetonitrile were of HPLC-grade. 2.2. Apparatus
Fig. 1. Structure of some etimicin components.
it. Since pre-column derivatization is cumbersome, time consuming and gives some problems with quantitation or results in unstable derivatives, electrochemical detection especially pulsed amperometric detection (PAD) is the better, pulsed amperometric detection has been demonstrated for the sensitive detection of numerous aminoglycoside antibiotics [19–24]. But up to now, the determination of etimicin with PAD has not been reported. To our knowledge, no paper has been published describing the composition of commercial etimicin samples. Like other aminoglycoside antibiotics [19–24], etimicin is polybasic cations at low pH, which are high polar species, and it can interact with the ion-pair agents to form ion-pair compounds which are less polar than the parent aminoglycosides and are amenable to separation by the preferred techniques of reversed-phase HPLC. In this work an ion-pair chromatography combined with PAD is described. The mobile phases that were investigated and further optimized. The quality of separation on different stationary phases was compared. Finally, the method has been applied to analyze some commercial samples of etimicin. 2. Experiment 2.1. Reagents, reference substances, and samples Distilled deionizes water of 18.2 M cm was used throughout. Heptafluorobutyric acid (Alfa Aesar, Heysham, Lancs). Standard etimicin and gentamycin C1a were obtained from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). The 3-N -ethyl and 1,3 -N-ethyl derivatives of gentamycin C1a (ETM-1, ETM-2), 3,2 ,6 -N-ethanoyl-gentamycin C1a (P1 ) and 1-N-ethyl-3,2 ,6 N-ethanoyl-gentamycin C1a (P2 ) were obtained from Jiangsu
Chromatographic analyses were carried out using a P680 HPLC Pump (Dionex, Sunnyvale, CA, USA), with a fixed loop of 20 L. Sodium hydroxide was added post-column using a pneumatic device, PC10 Pneumatic controller from Dionex (Sunnyvale, CA, USA). The ZORBAX Rx-C8 column (250 mm × 4.6 mm I.D.) from Agilent (Palo Alto, CA, USA). German Century SIL C18 column (AQ 5 m, 150 mm × 4.6 mm I.D.) from Dianlian Jiangshen Separating Science and Technology Company (Dianlian, China). AcclaimTM 120 C18 column ˚ 150 mm × 4.6 mm I.D.) from Dionex (Sunnyvale, (5 m120 A, CA, USA). The temperature of the column was maintained at 35 ◦ C. ED50A Electrochemical Detector from Dionex (Sunnyvale, CA, USA) was equipped with a gold working electrode with a diameter of 3 mm, an Ag/AgCl reference and stainless steel counter electrode. The cell of the pulsed amperometric detector was placed in the air keeping the temperature at 35 ◦ C. All the instrument control and data collection were performed by a Dionex Chromeleon 6.5. 2.3. Chromatography The mobile Phase is consisted of an aqueous solution containing 0.033 mol L−1 of oxalic acid, 0.012 mol L−1 heptafluorobutyric acid, 210 mL L−1 acetonitrile, adjust the apparent pH to 3.4 with dilute sodium hydroxide solution. Filter the mobile phase through a 0.45 m filter and sonicated before use. The flow rate was 1.0 mL min−1 . All substances to be analyzed were dissolved in the mobile phase. To allow pulsed amperometric detection, 0.52 mol L−1 NaOH was added postcolumn (0.5 mL min−1 ) through a mixing-tee from a nitrogen pressurized reservoir (30 psi) and mixed in a packed reaction coil (Dionex, 375 L), linking to the electrochemical cell. The flow rate for the addition of the base is not critical, but it should be reproducible between runs and must be pulse-free. It was necessary to raise the pH of the mobile phase to approximately 13 to improve the sensitivity of the detection [25]. The 0.52 mol L−1 NaOH solution was made starting from a 2.62 mol L−1 aqueous solution, which was pipetted into N2 degassed water to avoid carbonates that foul the electrodes. For this reason, it is advisable to pipette the NaOH solution from the center of the bottle and to use only two-thirds of the bottle. The time and voltage parameters for the pulsed amperometric detector were set as follows: E1 , E2 , and E3 were, respectively +0.12, +0.70, and −0.60 V with the assigned pulse durations t1 : 0–0.40 s, t2 : 0.41–0.60 s, and t3 : 0.61–1.00 s, integration of the signal was done between 0.20 and 0.40 s.
204
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
Fig. 2. Influence of the pH of the mobile phase on the capacity factors. Sta˚ 150 mm × 4.6 mm I.D.), tionary: AcclaimTM 120 C18 column (5 m 120 A, Methanol was used to determine t0 .
3. Results and discussion 3.1. Selection of chromatographic condition The influence of the different chromatographic parameters on the separation of etimicin and its related substances was evaluated using the retention times and resolution. The influence of the amount of acetonitrile in the mobile phase was examined between 130 and 240 mL L−1 . The retention times decrease on increasing the amount of acetonitrile and small differences in the acetonitrile content have significant consequences. To check the influence of the amount of oxalic acid, the content of oxalic acid in mobile phase was varied and the following contents were used: 0.025, 0.028, 0.033, and 0.036 mol L−1 . In the range examined, the amount of oxalic acid in the mobile phase has no significant effect on the retention times. The influence of the pH of the mobile phase examined using buffer solution with pH 2.5 and 4.5. Little change is observed between pH3.0 and 4.0 (Fig. 2). With further decrease or with further increase in the pH, the retention times decrease. Therefore, an appropriate and acidic mobile phase is necessary because the etimicin molecules must be positively charged to interact with the anionic heptafluorobutyric acid. The influence of the amount of heptafluorobutyric acid was investigated in the range 0.0074–0.015 mol L−1 has no significant effect on the retention, With further increase in the amount of heptafluorobutyric acid, the retention times increase much. Considering the retention times, resolution and sensitivity of detection, the mobile phase consisted of an aqueous solution containing 0.033 mol L−1 oxalic acid, 0.012 mol L−1 heptafluorobutyric acid and 210 mL L−1 acetonitrile, adjust the apparent pH to 3.4 with dilute sodium hydroxide solution. The influence of the column temperature was examined at 35, 40, and 45 ◦ C. As expected, the retention times of the components decrease when the column temperature is increased. However, the resolution between the peaks also decreases with higher column temperatures. Using the developed type of mobile phase above, three columns were examined at 35 ◦ C: ZORBAX Rx -C8 column (250 mm × 4.6 mm I.D.), German Century SIL C18 column (AQ 5 m, 150 mm × 4.6 mm I.D.) and AcclaimTM 120 C18 column
Fig. 3. Voltammetric response at Au electrode in 0.10 M NaOH: scan rate, 100 mV s−1 . (. . .) residual response, (—) 0.50 mg mL−1 etimicin sample.
˚ 150 mm × 4.6 mm I.D.). The latter gave clearly bet(5 m120A, ter overall resolution and was the only column that allowed the separation of etimicin from two other derivatives of gentamycin C1a (ETM-1 and ETM-2). 3.2. Waveform optimization and electrode stability The applied potentials and pulse durations can be chosen to optimize sensitivity, reproducibility, and selectivity for the samples of interest. In most cases, the optimum settings are those, which produce the largest chromatographic peak for the species of interest. Since current is sampled at E1 , the setting for E1 has the largest effect on the current response, and hence the peak height. Current-potential voltammetric response curves were used to screen for potential detector operation conditions. These voltammograms are illustrated in Fig. 3. The oxidation at the gold electrode surface by etimicin components occurred between 0.00 and 0.40 V. The reference solution (0.1 M NaOH) shows significantly less anodic current in this region. The electrochemical response of the reference solution was believed to be due to formation of surface oxide at E > +0.40 V and oxygen evolution at E > +0.70 V. The cyclic voltammetric data indicated that liquid chromatography with pulsed amperometric detection of etimicin should be operated with the Edet potential set between 0.00 and +0.40 V. The detector response was then used to optimize the amperometric response of the PAD unit. In the potential range from 0.02 to 0.37 V, E1 was optimized by varying in steps of 0.05 Volt between sample injections and compared the peak height. Using standard solution of etimicin, the response of the compound was measured as a function of potential Edet . Fig. 4 shows a plot of background, etimicin response versus the measurement potential, Edet . The maximum difference between background response and the response of etimicin occurred at approximately 0.12 V. This was the voltage used for Edet in the final method conditions. E1 = 0.12 V the signal-to-noise ratio of the detection is 3250 which is greater than at other detection voltages. The time and voltage parameters for the pulsed amperometrical detector were set as follows: E1 , E2 and E3 were, respectively
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
205
working electrode after about 60 analyses to obtain a good repeatability. After the electrode is cleaned with fine polishing compound, it is sonicated in water for 10 min. It is advisable to wipe the counter and reference electrodes at the same time with a tissue to remove deposited substances. It takes about 2 h to obtain a stable baseline with a freshly polished electrode. 3.3. Quantitative aspects of the LC method
Fig. 4. Plot of detector response (nC) as measurement potential (V) is varied.
+0.12, +0.70, and −0.60 V with the assigned pulse durations t1 : 0–0.40 s, t2 : 0.41–0.60 s, and t3 : 0.61–1.00 s, integration of the signal was done between 0.20 and 0.40 s. These parameters are the same as used before for similar components, like other aminoglycoside antibiotics [19–24]. Although the sequence of the potentials theoretically cleans the electrode surface, it is necessary to polish the gold
For the determination of the impurities a 20 g sample was examined by injecting 20 L of a 1.0 mg mL−1 etimicin sulfate sample solution (Fig. 5). For this quantity the limit of detection for P1 was 0.02% mg mL−1 (4 ng), as determined at a signal-to-noise ratio of 3. The limit of quantitation was 0.06% mg mL−1 (12 ng) (RSD = 10.3, n = 4). Fig. 6 shows the chromatogram of etimicin and its synthetic raw material GMC1a standard solutions. The linearity of etimicin and its related substances was examined in the concentration range from 0.005 to 0.125 mg mL−1 standard solutions. The results are shown in Table 1. Where y = peak area, x = concentration of standard solutions injected (mg mL−1 ), r = coefficient of correlation. The repeatability was checked by analyzing 0.05 mg mL−1 standard
Fig. 5. Typical chromatogram of a commercial sample (concentration: 1.0 mg mL−1 ) of etimicin sulfate obtained with pulsed amperometric detection. Stationary: ˚ 150 mm × 4.6 mm I.D.). (1) P2 , (2) P1 , (3) GMC1a , (4) ETM-1, (5) ETM-2, (6) ETM, (7) unknown-1, and (8) unknown-2. AcclaimTM 120 C18 column (5 m 120 A,
Fig. 6. Chromatogram of ETM standard solution (concentration: 0.125 mg mL−1 ) and GMC1a standard solutions (concentration: 0.125 mg mL−1 ). (1) GMC1a and (2) ETM.
206
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207
Table 1 Linearity of etimicin components standard Components
Range (mg/mL)
Regression equation
r
RSD (%) (n = 6)
P2 P1 ETM-1 ETM-2 ETM GMC1a
0.005–0.125 0.005–0.125 0.005–0.125 0.005–0.125 0.005–0.125 0.005–0.125
y = 411.56x − 0.8613 y = 389.25x − 0.4826 y = 374.29x − 0.5273 y = 427.52x − 1.2458 y = 433.07x − 1.3366 y = 365.17x − 0.2465
0.9983 0.9994 0.9989 0.9992 0.9990 0.9991
0.23 2.54 1.49 0.38 0.86 1.50
Table 2 Composition of etimicin sulfate samples (%), relative to etimicin
4. Conclusions
Samples P2
The method using ion-pair liquid chromatography allowed separating eight components of etimicin. The total time of analysis was not more than 30 min. It is the first time that quantitative results are reported for so many etimicin components with pulsed amperometric detection after post-adding alkali. Pulsed amperometric detection suffers from some stability problems and some experience is required to obtain a good reproducibility, but it allows for sensitive detection and determines the possible impurities without derivatization.
1 2 3
P1
GMC1a ETM-1 ETM-2 Unknown-1 Unknown-2
0.39 0.04 0.57 0.44 0.06 0.39 1.54 0.20 0.45
0.04 0.15 0.25
1.38 1.69 2.60
0.07 0.08 0.07
0.07 0.08 0.07
Table 3 The recovery datum of ETM, GMC1a in the etimicin sulfate samples Samples 1 2 3
P2 (%) 105.23 98.23 103.25
P1 (%) 100.25 99.63 99.45
GMC1a (%)
ETM-1 (%)
ETM-2 (%)
ETM (%)
106.58 105.43 106.88
103.25 101.29 97.58
99.86 98.47 102.41
101.77 100.30 102.33
Acknowledgements
solution of each compound six times, The RSDs of the peak area are shown in Table 1. From the results we can find that the differences between the response factors of etimicin and its related substances negligible. So, the HPLC–PAD method can be applied to the analysis of impurities, the control of the ratio of multi-components drug and the determination of new substances by using another substance as reference, etc. 3.4. Analysis of commercial samples Three commercial etimicin sulfate samples, obtained from different manufactures, were analyzed using the described method. The obtained composition of the samples is shown in Table 2. All minor components are expressed as relative amounts of etimicin, using chromatograms obtained with a 5% (m V−1 ) dilution (0.05 mg mL−1 ) of the examined commercial sample. From the results ETM-2 was found to be the major impurity, which was not detected using other methods. As can be seen, the composition of etimicin samples is dependent on the origin. Different batches from the same manufacturer had a similar composition. Table 3 shows the recovery datum of P2 , P1 , GMC1a , ETM-1, ETM-2, and ETM in the commercial etimicin sulfate samples. Using chromatograms obtained with a 5% (m V−1 ) dilution (0.05 mg mL−1 ) of the examined commercial etimicin sample and 0.05 mg mL−1 standard solutions of P2 , P1 , GMC1a , ETM1, ETM-2, and ETM, respectively.
This is a project supported by National Natural Science Foundation of China (No. 20375035) and by Zhejiang Provincial Natural Science Foundation of China (No. Z404105), and Sponsored by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry and Zhejiang Provincial. References [1] J. Liu, X.-L. Hu, Y.-Q. Shen, J. Fan, M. Zhao, Chin. J. Pharm. 31 (2000) 358. [2] E. Adams, D. Pueling, M. Rafiee, J. Hoogmartens, J. Chromatogr. A 812 (1998) 151. [3] Y. Liao, S.-H. Dai, Chin. J. Antibiotics 20 (1995) 198. [4] Y. Liao, S.-H. Dai, Chin. J. Antibiotics 20 (1995) 365. [5] H.-Q. Shi, J. Wang, Chin. J. Pharm. 34 (2003) 290. [6] M.-J. Zhou, G.-L. Wei, Y.-P. Liu, Y.-M. Sun, S.-H. Xiao, L. Lu, C.-X. Liu, D.-F. Zhong, J. Chromatogr. B 798 (2003) 43. [7] Chin. Pharm. 2 (2005) 740. [8] L.D. Sabath, J.I. Casey, P.A. Ruch, L.L. Stumpf, M.J. Finland, Lab. Clin. Med. 78 (1971) 457. [9] S.A. Bromn, D.R. Newkirk, R.P. Hualer, G.G. Smith, K.J. Sugimoto, Assoc. Off. Anal. Chem. 73 (1990) 479. [10] J.M. Andrews, R.J. Wise, Antimicrob. Chem. 14 (1984) 509. [11] C.L. Flurer, K.A. Wolnik, J. Chromatogr. B 663 (1994) 259. [12] C.P. Bhogte, V.B. Patravale, P.V. Devarajan, J. Chromatogr. B 694 (1997) 443. [13] J.W. Mayhew, S.L. Gorbach, J. Chromatogr. 151 (1978) 133. [14] L.T. Wong, A.R. Beaubien, A.P. Pakuts, J. Chromatogr. 231 (1982) 145. [15] D.M. Barends, J.S.F. Vandersandt, A. Hulshoff, J. Chromatogr. B 182 (1980) 201. [16] L.A. Kaine, K.A. Wolnik, J. Chromatogr. B 674 (1994) 255. [17] L.G. Mclaughlin, J.D. Henion, P.J. Kijik, Biol. Mass. Spectrom. 23 (1994) 417.
L. Xi et al. / J. Chromatogr. A 1115 (2006) 202–207 [18] M. Cherlet, S.D. Baere, P.D. Backer, J. Mass. Spectrom. 35 (2000) 1342. [19] E. Adams, W. Roelants, R.D. Paepe, E. Roets, J. Hoogmartens, J. Pharm. Biomed. Anal. 18 (1998) 689. [20] D. Debremaeker, E. Adams, E. Nadal, B. Van Hove, E. Roets, J. Hoogmartens, J. Chromatogr. A 953 (2002) 123. [21] E. Adams, R. Schepers, L.W. Gathu, R. Kibaya, E. Roets, J. Hoogmartens, J. Pharm. Biomed. Anal. 15 (1997) 505.
207
[22] E. Adams, J. Dalle, E.D. Bie, I.D. Smedt, E. Roets, J. Hoogmartens, J. Chromatogr. A 766 (1997) 133. [23] E. Adams, R. Schepers, E. Roets, J. Hoogmartens, J. Chromatogr. A 741 (1996) 233. [24] E. Adams, G.V. Vaerenbergh, E. Roets, J. Hoogmartens, J. Chromatogr. A 819 (1998) 93. [25] J.A. Statler, J. Chromatogr. 527 (1990) 244.