Journal of Chromatography B, 969 (2014) 53–59
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Determination of natamycin in rabbit cornea by high-performance liquid chromatography–tandem mass spectrometry with protective soaking extraction technology Zhou Tianyang a , Zhu Ling b , Xia Huiyun a , He Jijun a , Zhang Junjie a,∗ a b
Henan Eye Institute, Henan Eye Hospital, No. 7 Weiwu Road, Zhengzhou 450003, China School of Pharmaceutical Science, Zhengzhou University, No. 100 Science Road, Zhengzhou 450001, China
a r t i c l e
i n f o
Article history: Received 28 February 2014 Accepted 3 August 2014 Available online 12 August 2014
a b s t r a c t A new selective and sensitive high-performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS) method was developed for the quantification of natamycin (NAT) in rabbit corneas with amphotericin B as the internal standard (IS). The cornea samples were processed by a simple and protective methanol soaking extraction technology. The NAT could be extracted completely from rabbit cornea after 24 h of soaking with methanol under a mild condition. Chromatographic separation was performed on a C18 column (2.1 mm × 50 mm, 3.5 m) using mobile phase with ammonium acetate buffer (pH 4.5; 4.0 mM):acetonitrile (40:60, v/v) at a flow rate of 0.25 ml/min. Quantification was performed using the transitions 666.2 → 503.2 m/z for NAT and 924.5 → 906.6 m/z for IS by positive ion electrospray ionization in multiple reaction monitoring mode. The assay was validated over a concentration range of 8.64 ng/ml to 843 ng/ml with lower limit of detection of 4.32 ng/ml. The method was validated with respect to linearity, accuracy, precision, recovery, stability and extracting efficiency. The extraction recovery of NAT from cornea samples was approximately 100% with the new methanol soaking extraction procedure. The method has been successfully applied to the ocular pharmacokinetic studies of NAT eye drops in the cornea of Japanese white rabbit. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Fungal keratitis is a serious, infectious ocular disease that frequently occurs among people engaged in agricultural and other outdoor work [1]. It has become the most common infectious corneal disorder that results in blindness in certain regions [2–5]. Topical natamycin is the most commonly used medication for filamentous fungi keratitis, and is the only available commercialized antifungal medication for this disease [6–9]. Undoubtedly, the cornea is the most important therapeutic target of antifungal drugs. Information on the pharmacokinetics of NAT in the cornea is essential for setting the clinical dosage. 14 C-labeled NAT has been used to evaluate the corneal penetration of topical NAT [10]. But it is known that radioassay cannot distinguish parent drugs and metabolites. And the isotope effects and radioactive effects in vivo studies may significantly affect the analysis results. The application of LC–MS/MS is currently
∗ Corresponding author. Tel.: +86 371 65580919; fax: +86 371 65952907. E-mail address: [email protected] (Z. Junjie). http://dx.doi.org/10.1016/j.jchromb.2014.08.005 1570-0232/© 2014 Elsevier B.V. All rights reserved.
considered the method of choice for supporting pharmacokinetic and toxicokinetic studies due to its selectivity, sensitivity and short runtime [11]. It is a very competitive candidate for the quantification of NAT in rabbit cornea. LC–MS/MS also possesses good reproducibility, and broad applicability for the determination of most organic chemicals at suitable concentration levels. The bioanalytical methods for quantification of NAT in plasma, tears, wine, and other foodstuffs using LC–MS/MS have been reported in a few papers [11–14]. Obviously, the more widely used LC–MS/MS is more adequate to estimate the pharmacokinetics of NAT in the rabbit cornea. However, we have not found any reported papers on the determination of NAT in the cornea by LC–MS/MS. The stability of NAT in water or organic solvents is substantially affected by UV irradiation, extreme pH, heat exposure, oxidation, and presence of metal ions [15–17]. And the concentration of NAT in the cornea is very low [10]. A protective and efficient extraction method is essential to obtain high extraction recoveries and accuracy. But most of the reported methods involved troublesome solid phase extraction (SPE) methods [11–14]. They are not quite suitable for extracting NAT from the cornea because of the low extraction recoveries from biological samples [11,12]. Soaking extraction
54
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
method is often used to extract active substance from herb or animal tissues. Maybe suitable solvent can extracted NAT from cornea quickly and completely under mild conditions. Hence, the aim of present work was to develop a specific, sensitive, accurate and robust LC–MS/MS method for estimation of NAT in rabbit cornea. Methanol soaking extraction was used to replace the solid-phase extraction for the sample preparation. The validated method has been used to evaluate NAT pharmacokinetics in the cornea of Japanese white rabbits after topical administration of NAT ophthalmic suspension (5%). 2. Materials and methods 2.1. Materials NAT analytical standard was purchased from Dr. Ehrenstorfer GmbH (Germany). USP-grade amphotercin B (internal standard, IS) was purchased from Amresco Company (USA). High-performance liquid chromatography (HPLC)-grade methanol and acetonitrile were purchased from Tedia Company Inc. (Fairfield, USA). Ammonium acetate and glacial acetic acid (AR) were purchased from Zhengzhou Paini Chemical Reagent Factory (Zhengzhou, China). Ultrapure water was obtained from a Milli-Q PLUS PF water purification system. NAT ophthalmic suspension (5%, 15 ml, Natacyn® ) is a commercially available product obtained from Alcon Laboratories, Inc. (Fort Worth, USA). All solutions and samples containing NAT and IS were protected from light. 2.2. Chromatographic and mass spectrometric conditions An Alliance® HPLC system (2695 separation module) equipped with pump systems, autosampler, and temperature-controlled vial tray (Waters, USA) was used in this study. The mobile phase comprised ammonium acetate buffer (pH 4.5, 4.0 mM):acetonitrile (40:60, v/v). The flow rate was 0.25 ml/min (isocratic elution). A C18 column (3.5 m, 2.1 mm × 50 mm, Xterra MS, Waters, USA) was used for chromatographic separations at 35 ◦ C. The samples (5 l) were injected through auto-sampler on to the LC–MS/MS system. Solvent delay times were set to start at 0 min and end at 1.5 min. MS detection was performed on a Quattro PremierTM XE Micromass MS Technologies triple quadrupole mass spectrometer with a ZSprayTM electrospray ionization source (Waters, Milford, USA). The source was operated in the positive (ES+) mode and the capillary voltage was maintained at 3.0 kV. The cone and extractor voltage were kept at 20.0 and 3.0 V, respectively. The source temperature was set at 100 ◦ C. Ultrapure nitrogen was used as desolvation gas with a flow of 800 l/h and the desolvation temperature was set at 350 ◦ C. The collision energy was set at 10 V and Argon was used as the collision gas at a flow of 0.21 ml/min. The detection and quantification of analytes were performed using the multiple reaction monitoring (MRM) mode with mass transitions 666.2 → 503.2 m/z for NAT and 924.5 → 906.6 m/z for IS. The data were acquired using MassLinx Version 4.1 Mass Spectrometry Software (Waters, Milford, USA). 2.3. Preparation of stock and standard solutions The stock solutions of NAT (211 g/ml) and IS (70.4 g/ml) were prepared in methanol and methanol:DMSO (1:1, v/v), respectively. The stock solutions were further diluted with methanol to produce working standard solutions (8.64, 21.6, 54.0, 135, 337, and 843 ng/ml for NAT and 1.41 g/ml for IS) and quality control samples (QCs). The NAT working standard solutions (100 l) were pipetted into centrifuge tubes and dried by nitrogen flow at 35 ◦ C. Calibration standards were prepared by adding a 100 l aliquot of pooled methanol extracting solution of blank cornea and a 20 l
aliquot of IS working standard solutions to obtain a concentration range of 8.64, 21.6, 54.0, 135, 337, and 843 ng/ml. Subsequently, the tubes were vortexed for 30 s and centrifuged for 8 min at 3500 revolutions per minute for analysis. QCs (21.6, 135, and 337 ng/ml) were prepared in a similar manner with appropriate working standard solution. 2.4. Sample preparation Cornea samples were cut up and immersed in 0.5 ml methanol for 24 h at 4 ± 2 ◦ C under nitrogen protection. A 20 l aliquot of IS working standard solution was added to 100 l of methanol extracting solution. The mixture was vortexed for 30 s and centrifuged for 8 min at 3500 revolutions per minute. The supernatant was pipetted into an autosampler vial for LC–MS/MS analysis. The pooled blank extracting solution (PBES) were prepared by adding 10 times volume of methanol (v/w) to blank cornea and storing at 4 ± 2 ◦ C for 24 h under nitrogen protection. PBES was used to prepare the calibration standards, QCs and other spiked samples. 2.5. Method validation The method was validated for specificity, sensitivity, linearity, recovery, matrix effect, precision, accuracy, stability, and dilution integrity. The extracting efficiency of methanol was also evaluated to set the “soaking time” of the cornea samples. The specificity was investigated by analyzing processed blank cornea from six individual rabbits. Specificity was established by the lack of interfering peaks at the retention time for the NAT and IS. Linearity was tested at six different concentrations, covering a range of 8.64–843 ng/ml. The calibration curves were established by plotting the peak area ratio of analyte and IS versus concentration. The lower limit of quantification (LLOQ) of the assay was assessed as the lowest concentration on the calibration curve that can be quantitatively determined within ±20% accuracy and precision. The LLOQ was established based on five replicates. For estimating the extraction recovery, a 10 mg aliquot of blank rabbit cornea was immersed in 100 l of methanol spiked with analyte at three concentration levels (21.6, 135, and 337 ng/ml) and processed according to soaking procedure. The controls were prepared by adding a 10 l aliquot of deionized water to 100 l of PBES spiked with analyte at the same concentration levels. The extraction recovery of NAT from rabbit cornea was calculated by comparing the areas of extracted samples against the controls at corresponding concentration. These experiments were performed in five replicates. The matrix effect was evaluated by comparing the corresponding peak areas of the spiked PBES to those of the standard solutions evaporated directly and reconstituted in mobile phase. These experiments were performed in three replicates at three concentration levels (21.6, 135, and 337 ng/ml) for NAT and at a single concentration of 235 ng/ml for IS. Precision and accuracy of this analytical method were determined using QCs in five replicates of 21.6, 135, and 337 ng/ml of NAT in PBES for three consecutive days. The intra- and inter-batch precisions were calculated according to the relative standard deviation (% RSD). Accuracy was calculated according to the degree of closeness of back-calculated concentration value using the calibration curve to nominal concentration value (% bias = (Cobserved − Cnominal )/Cnominal ) × 100). The precision and accuracy determined at each concentration levels should be within ±15%. The stability of the stock solution was evaluated at −70 ± 10 ◦ C for 2 months and at 4 ± 2 ◦ C for 72 h. The stability of the NAT in rabbit cornea were evaluated by adding a 10 mg aliquot of blank
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
cornea to the spiked tube and storing at −70 ± 10 ◦ C for 4 weeks. The stability of NAT in PBES was evaluated at 4 ± 2 ◦ C for 72 h at the same concentration levels as the matrix effect experiment. The QCs were used for the bench-top stability study, which was performed at ambient temperature for 24 h.
55
cornea were calculated using DAS software Ver.2.1.1 (Shanghai BioGuider Medicinal Technology Co., Ltd., China). 3. Results and discussion 3.1. Method development
2.6. Dilution integrity Sample dilution was required if the concentrations of the study sample were expected to be higher than the upper limit of quantification (ULOQ). The dilution integrity experiment was performed at 2.0-fold higher ULOQ concentration. This dilution integrity stock was subsequently mixed with PBES by factors 3 and 5. Six replicates of 1/3 and 1/5 concentrations were prepared and processed along with a freshly prepared calibration curve. The deviation of the mean back-calculated concentration values should be in the range of 85% to 115% according to their nominal values. 2.7. Extracting efficiency Six healthy male Japanese white rabbits (Henan Kangda Laboratory Animals Co., Ltd., China) weighing 2.0 to 2.5 kg were used for evaluating the extracting efficiency. The animal studies were carried out as per the guidelines of the local ethical committee on animal experimentation. The rabbits were accommodated in cages kept in a light-controlled (alternate night and day cycles, 12 h each) and air-conditioned room and fed a standard laboratory diet and water ad libitum. NAT ophthalmic suspension (50 l) was instilled into the lower conjunctival cul-de-sac of eyes (both sides). After instillation, the eyelids were kept closed for 10 s to prevent the loss of the instilled solution. The central cornea epithelium of half animals was carefully removed two hours before topical administration of NAT. To remove a appropriate area of cornea epithelium, a piece of soft round filter paper ( 6 mm) wetted by 50% ethanol was put on the center of cornea for a minute after application of topical anesthesia (Tetracaine hydrochloride, 0.5%). The central cornea epithelium contacted the ethanol filter paper were easily removed with a blunt instrument. The animals were sacrificed by injecting pentobarbital sodium solution (4%, w/v) via the ear vein at 15 min post dose. Subsequently, the eyes were rinsed with saline and the corneas were collected, weighed, and processed as described above. A 50 l aliquot of extracting solution was collected at 24, 48, and 72 h after soaking. The extracting efficiency was evaluated by comparing the concentrations of NAT in the extracting solution at different soaking times. The statistical analysis was performed using the paired-sample t test by the SPSS Statistics Ver.17.0. 2.8. Application of the method A total of 36 healthy male Japanese white rabbits (Henan Kangda Laboratory Animals Co., Ltd., China) weighing 2.0 to 2.5 kg were used for the ocular pharmacokinetics study. NAT ophthalmic suspension (50 l) was topically administrated and the cornea epithelium of half animals was also removed before instillation as described above. The treated animals were sacrificed at 15, 30, 60, 120, 240, and 360 min post dose. All animals were given free access to food and water during the study. Each time point involved three animals with intact epithelium and three animals with debrided corneas. Subsequently, the eyes were rinsed with saline, and the corneas were collected and weighed. All samples were stored at −70 ± 10 ◦ C before processing. QCs were distributed among calibrators and unknown samples in the analytical run. Concentration–time curve of cornea was plotted according to the determined values. Pharmacokinetic parameters of NAT in the
The stability of NAT in water or organic solvents is substantially affected by UV irradiation, extreme pH, heat exposure, oxidation, and presence of metal ions [15–17]. A suitable sample processing method involving simple procedures and protective measures is essential to obtain satisfactory accuracy and precision. Therefore, the “one-step” soaking extraction method with methanol under nitrogen protection was used for sample preparation. The mean extraction recovery of such “one-step” method was higher than 90%, whereas the reported extraction recovery of SPE method was approximately 75% [11,12]. Though the methanol soaking extraction method was simple and efficient, it would introduce more impurities compared with SPE method, including salts and electrolytes. The origin and mechanism of matrix effects are not understood fully and there are many possible sources of ion suppression [13]. The effects of salt or electrolyte concentration on analyte response observed through electrospray ionization MS are undoubtedly among the most important and definite factors [18]. The “solvent delay times” were set to start at 0 min and end at 1.5 min to prevent the salts and electrolytes in methanol extracting solution entering in the MS detector. The analyte response was sensitive and stable during the whole experiment. The optimization of MS parameters and MS transitions was performed according to the reported papers [13,19]. Amphotericin B was selected as internal standard due to its structural similarities with NAT [19]. The capillary voltage, cone voltage, desolvation temperature, desolvation gas flow, collision gas flow and collision energy were optimized thoroughly with flow injection analysis (FIA) mode. Detection and quantification of analytes were performed using the MRM mode. Both the analyte and internal standard have ability to accept proton and generate [M + H]+ ions (Fig. 1). The fragments were selected at m/z 503.2 and 906.6 as the most prominent fragments for NAT and IS, respectively (Fig. 1). The fragments of at m/z 503.2 and 906.6 for NAT and IS were both stable in the collision cell at low collision energy. Higher collision energy would split these two ions into smaller fragments and the signal intensity of ions decreased obviously. The water and acetonitrile combination were used as mobile phase according to the reported paper [11]. Though chromatographic retention and the spray droplets in the ESI interface could be improved by high organic content in the mobile phase [11], the chromatographic peak shape of NAT was becoming abnormal when the proportion of acetonitrile was higher than 60%. To get more stable and sensitive signal response, mobile phase modifiers such as acetic acid, formic acid, ammonium formate, ammonium acetate and their combinations were also studied. The ammonium acetate (4.0 mM) which could enhance the target ion abundance and maintained stable retention and ionization behavior was used as buffer of the mobile phase. 3.2. Validation procedures 3.2.1. Specificity and selectivity Chromatograms of blank cornea samples from six individual rabbits lacked co-eluting peaks of NAT and IS at their respective retention times. Typical chromatograms of blank cornea sample, spiked PBES and pharmacokinetic sample were shown in Fig. 2. 3.2.2. Calibration curve The peak area ratio of NAT to IS was linear over a NAT concentration range of 8.64 ng/ml to 843 ng/ml. The linear equation
56
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
Fig. 1. MS/MS spectra of (a) IS and (b) NAT.
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
57
Fig. 2. Typical MRM chromatograms of (a) methanol extracting solution of blank cornea at MRM 924.5/906.6 for IS and 666.2/503.2 for NAT, (b) PBES spiked with NAT (135 ng/ml) and IS (235 ng/ml), (c) pharmacokinetic sample at 15 min.
was “y = 0.0039x + 0.0036”, where “y” is peak area ratio of NAT to IS and x is concentration of NAT. The correlation coefficients (R) of the calibration curves for spiked samples was greater than 0.9999. The weighting factor 1/x0 was used to calculate correlation coefficient, slope and intercept. The standard deviation values for correlation coefficient (R), intercept and slope were 0.013, 0.0065 and 0.00002, respectively. The LLOQ value was 8.64 ng/ml, with a signal-to-noise ratio of 13. The mean relatively recovery of the LLOQ was 103.4%, and the RSD was 12.0% (n = 5).
3.2.3. Matrix effect The quotient of the peak area (A) of the spiked and control samples was used as an indicator of ion suppression or signal enhancement, which was calculated as [(Aspiked /Acontrol × 100) − 100]%. The indicator of matrix effect for NAT at 21.6, 135, and 337 ng/ml concentration levels was −2.2%, −6.2%, and −4.4%, respectively. Thus, no significant ion enhancement or suppression was observed.
3.2.4. Recovery The extraction recoveries of NAT from the spiked cornea samples at low (21.6 ng/ml), medium (135 ng/ml), and high (337 ng/ml) concentrations levels were 103.1 ± 12.8%, 92.7 ± 8.6%, and 94.6 ± 7.8%, respectively (n = 3). 3.2.5. Accuracy and precision Accuracy and precision data for intra-batch and inter-batch spiked samples were presented in Table 1. The intra-batch precision (% RSD) and accuracy (% bias) values for developed method were from 2.1% to 11.2% and −7.1% to 4.6%, respectively. The inter-batch
precision (% RSD) and accuracy (% bias) values were from 6.0% to 7.4% and −2.2% to 1.2%, respectively. 3.2.6. Stability The stability of the stock solution was calculated by comparing the peak area of the initial solution and that of the final solution using independent-sample t test. No significant variation for the stock solution of NAT and IS was observed after 2 months at −70 ± 10 (P > 0.05, n = 3) and after 72 h at 4 ± 2 ◦ C (P > 0.05, n = 3). The stability data of NAT in the PBES at 4 ± 2 ◦ C, −70 ± 10 ◦ C and bench-top were presented in Table 2. 3.2.7. Dilution study The mean back-calculated concentrations of 1/3 and 1/5 diluted samples for NAT were 99.3% and 97.5% according to their nominal values, respectively. RSDs were 4.8% and 6.0%, respectively. 3.3. Extracting efficiency To evaluate the extracting efficiency of methanol, the cornea samples collected from the treated rabbits were used as the NAT-containing cornea samples. The concentration of NAT in the extracting solution should stay at a certain level with the soaking time going on if the NAT has been extracted completely. The concentrations of NAT in the extracting solution at different immersion times were presented in Table 3. No significant variation (P > 0.05, n = 6) was observed between different soaking times. 3.4. Application of the method The rabbit cornea samples were processed and determined along with the QCs. The assay sensitivity and specificity were sufficient to accurately characterize the pharmacokinetics of NAT in
58
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
Table 1 Intra-batch and inter-batch assay precision and accuracy for NAT. Nominal conc. (ng/ml)
Intra-batch (n = 5)
Inter-batch (n = 15)
1st day
21.6 135 337
2nd day
3rd day
%Bias
%RSD
%Bias
%RSD
%Bias
%RSD
%Bias
%RSD
−3.3 −7.1 2.1
11.2 9.8 8.5
1.0 0.0 −1.9
5.4 8.4 5.6
−4.3 −4.6 4.6
2.8 3.2 2.1
−2.2 −0.8 1.2
7.4 8.9 6.0
Table 2 Stability of NAT in the cornea and PBES (n = 3). Nominal conc. (ng/ml)
Stability
Mean
SD
Precision (%RSD)
21.6
In cornea (−70 ◦ C, 4W) In PBES (4 ◦ C,72 h) Bench-top (24 h) In cornea (−70 ◦ C, 4W) In PBES (4 ◦ C,72 h) Bench-top (24 h) In cornea (−70 ◦ C, 4W) In PBES (4 ◦ C,72 h) Bench-top (24 h)
19.8 21.0 20.5 140.7 134.7 141.6 348.0 359.2 336.9
1.82 1.44 0.45 7.2 11.0 5.5 8.5 17.8 7.5
9.2 6.9 2.2 5.1 8.2 3.9 2.5 5.0 2.2
135
337
Accuracy (%bias) −8.3 −2.8 −5.1 4.2 −0.2 4.9 3.3 6.6 −0.0
Fig. 3. Concentration–time profile of NAT in rabbit cornea following topical administration of NAT ophthalmic suspension (5%). Table 3 Variation of NAT concentration in extracting solutions at different soaking time (t test, n = 6). N
Concentration of NAT(ng/ml) Debrided cornea
1 2 3 4 5 6
Intact cornea
24 h
48 h
72 h
185.3 213.6 169.5 155.7 193.8 122.9
165.2 198.5 169.6 136.9 211.3 133.9
177.4 220.3 175.8 149.8 215.8 126.9
24 h
48 h
72 h
P > 0.05 32.2 27.2 20.6 16.5 34.9 40.1
26.8 26.2 19.8 14.5 39.2 41.9
29.6 27.9 23.6 16.5 33.6 40.3
P > 0.05
rabbit cornea. No study samples with concentrations above the ULOQ were observed. The ocular pharmacokinetic profile of NAT ophthalmic suspension was similar to that of the sustained release formulation
as shown in Fig. 3. The NAT ophthalmic suspension flocculated to irregular clumps quickly after instillation. The drug clumps adhered to the conjunctival sac or the eyelid for a long time (Fig. 4), and such adherence strongly affected the pharmacokinetic profile of NAT ophthalmic suspension. However, the prolonged drug resident time failed to enhance the drug absorption substantially because of the low dissolving speed of NAT from the drug clumps and the relatively fast irrigation of the tear fluid [20,21]. The maximum concentration (Cmax ) of NAT in the intact cornea was 275.8 ± 158.2 ng/g and the area under the concentration curve (AUC(0–360 min) ) was 58,879.1 min ng/g. The Cmax of NAT in the debrided cornea was 1225.4 ± 355.8 ng/g, and the AUC(0–360 min) was 161,377.0 min ng/g. The corneal epithelium also affected the absorption of NAT in the cornea after topical administration. The half-life (t1/2 ) of NAT in rabbit cornea couldn’t be calculated accurately because of the irregular absorption of the ophthalmic suspension.
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
59
Fig. 4. Flocculation of NAT ophthalmic suspension (5%) on the ocular surface of rabbit at 30 min post dose.
4. Conclusion The current validated LC–MS/MS bioanalytical method for quantification of NAT in rabbit cornea was specific, sensitive, robust, precise and accurate. The sample preparation method based on methanol soaking extraction was simple, mild and efficient. The recoveries of the extracted cornea samples were higher and consistent over the investigated concentration range. From the results of all the validation parameters and applicability of the assay, we can conclude that the newly developed method can be useful for ocular pharmacokinetic studies of NAT with desired precision and accuracy along with high-throughput. Acknowledgment The authors Zhou Tianyang and Zhang Junjie are thankful to Mr. Zhou Jichun of Henan Provincial Institute of Food and Drug Control for his assistance in the animal experiment. References [1] P.A. Thomas, Eye 17 (2003) 852–862. [2] S. Xuguang, W. Zhixin, W. Zhiqun, L. Shiyun, L. Ran, Am. J. Ophthalmol. 143 (2007) 131–133. [3] R.A. Oechsler, T.M. Yamanaka, P.J. Bispo, J. Sartori, M.C. Yu, A.S. Melo, D. Miller, A.L. Hofling-Lima, Clin. Ophthalmol. 7 (2013) 1693–1701.
[4] F. Mohd-Tahir, A. Norhayati, I. Siti-Raihan, M. Ibrahim, Interdiscip. Perspect. Infect. Dis. 2012 (2012) 851563. [5] L. Pradhan, S. Sharma, S. Nalamada, S.K. Sahu, S. Das, P. Garg, Indian J. Ophthalmol. 59 (2011) 512–514. [6] Sonal S. Tuli, Clin Ophthalmol. 5 (2011) 275–279. [7] X.H. Dong, W.J. Gao, X.P. He, Int. J. Ophthalmol. 5 (2012) 143–146. [8] V.N. Prajna, P.S. Lalitha, J. Mascarenhas, T. Krishnan, M. Srinivasan, C.M. Vaitilingam, C.E. Oldenburg, A. Sy, J.D. Keenan, T.C. Porco, N.R. Acharya, T.M. Lietman, Br. J. Ophthalmol. 96 (2012) 1440–1441. [9] P. Lalitha, R. Vijaykumar, N.V. Prajna, A.W. Fothergill, J. Clin. Microbiol. 46 (2008) 3477–3478. [10] D.M. O’Day, W.S. Head, R.D. Robinson, J.A. Clanton, Curr. Eye Res. 5 (1986) 877–882. [11] Yashpal Singh Chhonker, Devendra Kumar, Pankaj Shrivastava, Deepak Kumar, Rajbir Singh, Hardik Chandasana, Rabi Sankar Bhatta, J. Pharm. Anal. 3 (2013) 144–148. [12] R.S. Bhatta, H. Chandasana, C. Rathi, D. Kumar, Y.S. Chhonker, G.K. Jain, J. Pharm. Biomed. Anal. 54 (2011) 1096–1100. [13] L.M. Repizo, L.D. Martinez, R.A. Olsina, S. Cerutti, J. Raba, Anal. Bioanal. Chem. 402 (2012) 965–973. [14] F. Fuselli, C. Guarino, A. La Mantia, L. Longo, A. Faberi, R.M. Marianella, J. Chromatogr. B 906 (2012) 9–18. [15] J.L. Koontz, J.E. Marcy, W.E. Barbeau, S.E. Duncan, J. Agric. Food Chem. 51 (2013) 7111–7114. [16] K. Thoma, N. Kübler, E. Reimann, Pharmazie 52 (1997) 362–373. [17] J.M. Gutteridge, A.H. Thomas, A. Cuthbert, J. Appl. Biochem. 5 (1983) 53–58. [18] T.L. Constantopoulos, G.S. Jackson, C.G. Enke, J. Am. Soc. Mass. Spectrom. 10 (1999) 625–634. [19] Niranjan M. Deshpande, Manish G. Gangrade, Maharudra B. Kekarea, Vikas V. Vaidya, J. Chromatogr. B 878 (2010) 315–326. [20] Indu Pal Kaur, Shilpa Kakkar, Expert Opin. Drug Delivery 7 (2010) 1303–1327. [21] J. Barar, A.R. Javadzadeh, Y. Omidi, Expert Opin. Drug Delivery (2008) 567–581.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Determination of natamycin in rabbit cornea by high-performance liquid chromatography–tandem mass spectrometry with protective soaking extraction technology Zhou Tianyang a , Zhu Ling b , Xia Huiyun a , He Jijun a , Zhang Junjie a,∗ a b
Henan Eye Institute, Henan Eye Hospital, No. 7 Weiwu Road, Zhengzhou 450003, China School of Pharmaceutical Science, Zhengzhou University, No. 100 Science Road, Zhengzhou 450001, China
a r t i c l e
i n f o
Article history: Received 28 February 2014 Accepted 3 August 2014 Available online 12 August 2014
a b s t r a c t A new selective and sensitive high-performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS) method was developed for the quantification of natamycin (NAT) in rabbit corneas with amphotericin B as the internal standard (IS). The cornea samples were processed by a simple and protective methanol soaking extraction technology. The NAT could be extracted completely from rabbit cornea after 24 h of soaking with methanol under a mild condition. Chromatographic separation was performed on a C18 column (2.1 mm × 50 mm, 3.5 m) using mobile phase with ammonium acetate buffer (pH 4.5; 4.0 mM):acetonitrile (40:60, v/v) at a flow rate of 0.25 ml/min. Quantification was performed using the transitions 666.2 → 503.2 m/z for NAT and 924.5 → 906.6 m/z for IS by positive ion electrospray ionization in multiple reaction monitoring mode. The assay was validated over a concentration range of 8.64 ng/ml to 843 ng/ml with lower limit of detection of 4.32 ng/ml. The method was validated with respect to linearity, accuracy, precision, recovery, stability and extracting efficiency. The extraction recovery of NAT from cornea samples was approximately 100% with the new methanol soaking extraction procedure. The method has been successfully applied to the ocular pharmacokinetic studies of NAT eye drops in the cornea of Japanese white rabbit. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Fungal keratitis is a serious, infectious ocular disease that frequently occurs among people engaged in agricultural and other outdoor work [1]. It has become the most common infectious corneal disorder that results in blindness in certain regions [2–5]. Topical natamycin is the most commonly used medication for filamentous fungi keratitis, and is the only available commercialized antifungal medication for this disease [6–9]. Undoubtedly, the cornea is the most important therapeutic target of antifungal drugs. Information on the pharmacokinetics of NAT in the cornea is essential for setting the clinical dosage. 14 C-labeled NAT has been used to evaluate the corneal penetration of topical NAT [10]. But it is known that radioassay cannot distinguish parent drugs and metabolites. And the isotope effects and radioactive effects in vivo studies may significantly affect the analysis results. The application of LC–MS/MS is currently
∗ Corresponding author. Tel.: +86 371 65580919; fax: +86 371 65952907. E-mail address: [email protected] (Z. Junjie). http://dx.doi.org/10.1016/j.jchromb.2014.08.005 1570-0232/© 2014 Elsevier B.V. All rights reserved.
considered the method of choice for supporting pharmacokinetic and toxicokinetic studies due to its selectivity, sensitivity and short runtime [11]. It is a very competitive candidate for the quantification of NAT in rabbit cornea. LC–MS/MS also possesses good reproducibility, and broad applicability for the determination of most organic chemicals at suitable concentration levels. The bioanalytical methods for quantification of NAT in plasma, tears, wine, and other foodstuffs using LC–MS/MS have been reported in a few papers [11–14]. Obviously, the more widely used LC–MS/MS is more adequate to estimate the pharmacokinetics of NAT in the rabbit cornea. However, we have not found any reported papers on the determination of NAT in the cornea by LC–MS/MS. The stability of NAT in water or organic solvents is substantially affected by UV irradiation, extreme pH, heat exposure, oxidation, and presence of metal ions [15–17]. And the concentration of NAT in the cornea is very low [10]. A protective and efficient extraction method is essential to obtain high extraction recoveries and accuracy. But most of the reported methods involved troublesome solid phase extraction (SPE) methods [11–14]. They are not quite suitable for extracting NAT from the cornea because of the low extraction recoveries from biological samples [11,12]. Soaking extraction
54
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
method is often used to extract active substance from herb or animal tissues. Maybe suitable solvent can extracted NAT from cornea quickly and completely under mild conditions. Hence, the aim of present work was to develop a specific, sensitive, accurate and robust LC–MS/MS method for estimation of NAT in rabbit cornea. Methanol soaking extraction was used to replace the solid-phase extraction for the sample preparation. The validated method has been used to evaluate NAT pharmacokinetics in the cornea of Japanese white rabbits after topical administration of NAT ophthalmic suspension (5%). 2. Materials and methods 2.1. Materials NAT analytical standard was purchased from Dr. Ehrenstorfer GmbH (Germany). USP-grade amphotercin B (internal standard, IS) was purchased from Amresco Company (USA). High-performance liquid chromatography (HPLC)-grade methanol and acetonitrile were purchased from Tedia Company Inc. (Fairfield, USA). Ammonium acetate and glacial acetic acid (AR) were purchased from Zhengzhou Paini Chemical Reagent Factory (Zhengzhou, China). Ultrapure water was obtained from a Milli-Q PLUS PF water purification system. NAT ophthalmic suspension (5%, 15 ml, Natacyn® ) is a commercially available product obtained from Alcon Laboratories, Inc. (Fort Worth, USA). All solutions and samples containing NAT and IS were protected from light. 2.2. Chromatographic and mass spectrometric conditions An Alliance® HPLC system (2695 separation module) equipped with pump systems, autosampler, and temperature-controlled vial tray (Waters, USA) was used in this study. The mobile phase comprised ammonium acetate buffer (pH 4.5, 4.0 mM):acetonitrile (40:60, v/v). The flow rate was 0.25 ml/min (isocratic elution). A C18 column (3.5 m, 2.1 mm × 50 mm, Xterra MS, Waters, USA) was used for chromatographic separations at 35 ◦ C. The samples (5 l) were injected through auto-sampler on to the LC–MS/MS system. Solvent delay times were set to start at 0 min and end at 1.5 min. MS detection was performed on a Quattro PremierTM XE Micromass MS Technologies triple quadrupole mass spectrometer with a ZSprayTM electrospray ionization source (Waters, Milford, USA). The source was operated in the positive (ES+) mode and the capillary voltage was maintained at 3.0 kV. The cone and extractor voltage were kept at 20.0 and 3.0 V, respectively. The source temperature was set at 100 ◦ C. Ultrapure nitrogen was used as desolvation gas with a flow of 800 l/h and the desolvation temperature was set at 350 ◦ C. The collision energy was set at 10 V and Argon was used as the collision gas at a flow of 0.21 ml/min. The detection and quantification of analytes were performed using the multiple reaction monitoring (MRM) mode with mass transitions 666.2 → 503.2 m/z for NAT and 924.5 → 906.6 m/z for IS. The data were acquired using MassLinx Version 4.1 Mass Spectrometry Software (Waters, Milford, USA). 2.3. Preparation of stock and standard solutions The stock solutions of NAT (211 g/ml) and IS (70.4 g/ml) were prepared in methanol and methanol:DMSO (1:1, v/v), respectively. The stock solutions were further diluted with methanol to produce working standard solutions (8.64, 21.6, 54.0, 135, 337, and 843 ng/ml for NAT and 1.41 g/ml for IS) and quality control samples (QCs). The NAT working standard solutions (100 l) were pipetted into centrifuge tubes and dried by nitrogen flow at 35 ◦ C. Calibration standards were prepared by adding a 100 l aliquot of pooled methanol extracting solution of blank cornea and a 20 l
aliquot of IS working standard solutions to obtain a concentration range of 8.64, 21.6, 54.0, 135, 337, and 843 ng/ml. Subsequently, the tubes were vortexed for 30 s and centrifuged for 8 min at 3500 revolutions per minute for analysis. QCs (21.6, 135, and 337 ng/ml) were prepared in a similar manner with appropriate working standard solution. 2.4. Sample preparation Cornea samples were cut up and immersed in 0.5 ml methanol for 24 h at 4 ± 2 ◦ C under nitrogen protection. A 20 l aliquot of IS working standard solution was added to 100 l of methanol extracting solution. The mixture was vortexed for 30 s and centrifuged for 8 min at 3500 revolutions per minute. The supernatant was pipetted into an autosampler vial for LC–MS/MS analysis. The pooled blank extracting solution (PBES) were prepared by adding 10 times volume of methanol (v/w) to blank cornea and storing at 4 ± 2 ◦ C for 24 h under nitrogen protection. PBES was used to prepare the calibration standards, QCs and other spiked samples. 2.5. Method validation The method was validated for specificity, sensitivity, linearity, recovery, matrix effect, precision, accuracy, stability, and dilution integrity. The extracting efficiency of methanol was also evaluated to set the “soaking time” of the cornea samples. The specificity was investigated by analyzing processed blank cornea from six individual rabbits. Specificity was established by the lack of interfering peaks at the retention time for the NAT and IS. Linearity was tested at six different concentrations, covering a range of 8.64–843 ng/ml. The calibration curves were established by plotting the peak area ratio of analyte and IS versus concentration. The lower limit of quantification (LLOQ) of the assay was assessed as the lowest concentration on the calibration curve that can be quantitatively determined within ±20% accuracy and precision. The LLOQ was established based on five replicates. For estimating the extraction recovery, a 10 mg aliquot of blank rabbit cornea was immersed in 100 l of methanol spiked with analyte at three concentration levels (21.6, 135, and 337 ng/ml) and processed according to soaking procedure. The controls were prepared by adding a 10 l aliquot of deionized water to 100 l of PBES spiked with analyte at the same concentration levels. The extraction recovery of NAT from rabbit cornea was calculated by comparing the areas of extracted samples against the controls at corresponding concentration. These experiments were performed in five replicates. The matrix effect was evaluated by comparing the corresponding peak areas of the spiked PBES to those of the standard solutions evaporated directly and reconstituted in mobile phase. These experiments were performed in three replicates at three concentration levels (21.6, 135, and 337 ng/ml) for NAT and at a single concentration of 235 ng/ml for IS. Precision and accuracy of this analytical method were determined using QCs in five replicates of 21.6, 135, and 337 ng/ml of NAT in PBES for three consecutive days. The intra- and inter-batch precisions were calculated according to the relative standard deviation (% RSD). Accuracy was calculated according to the degree of closeness of back-calculated concentration value using the calibration curve to nominal concentration value (% bias = (Cobserved − Cnominal )/Cnominal ) × 100). The precision and accuracy determined at each concentration levels should be within ±15%. The stability of the stock solution was evaluated at −70 ± 10 ◦ C for 2 months and at 4 ± 2 ◦ C for 72 h. The stability of the NAT in rabbit cornea were evaluated by adding a 10 mg aliquot of blank
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
cornea to the spiked tube and storing at −70 ± 10 ◦ C for 4 weeks. The stability of NAT in PBES was evaluated at 4 ± 2 ◦ C for 72 h at the same concentration levels as the matrix effect experiment. The QCs were used for the bench-top stability study, which was performed at ambient temperature for 24 h.
55
cornea were calculated using DAS software Ver.2.1.1 (Shanghai BioGuider Medicinal Technology Co., Ltd., China). 3. Results and discussion 3.1. Method development
2.6. Dilution integrity Sample dilution was required if the concentrations of the study sample were expected to be higher than the upper limit of quantification (ULOQ). The dilution integrity experiment was performed at 2.0-fold higher ULOQ concentration. This dilution integrity stock was subsequently mixed with PBES by factors 3 and 5. Six replicates of 1/3 and 1/5 concentrations were prepared and processed along with a freshly prepared calibration curve. The deviation of the mean back-calculated concentration values should be in the range of 85% to 115% according to their nominal values. 2.7. Extracting efficiency Six healthy male Japanese white rabbits (Henan Kangda Laboratory Animals Co., Ltd., China) weighing 2.0 to 2.5 kg were used for evaluating the extracting efficiency. The animal studies were carried out as per the guidelines of the local ethical committee on animal experimentation. The rabbits were accommodated in cages kept in a light-controlled (alternate night and day cycles, 12 h each) and air-conditioned room and fed a standard laboratory diet and water ad libitum. NAT ophthalmic suspension (50 l) was instilled into the lower conjunctival cul-de-sac of eyes (both sides). After instillation, the eyelids were kept closed for 10 s to prevent the loss of the instilled solution. The central cornea epithelium of half animals was carefully removed two hours before topical administration of NAT. To remove a appropriate area of cornea epithelium, a piece of soft round filter paper ( 6 mm) wetted by 50% ethanol was put on the center of cornea for a minute after application of topical anesthesia (Tetracaine hydrochloride, 0.5%). The central cornea epithelium contacted the ethanol filter paper were easily removed with a blunt instrument. The animals were sacrificed by injecting pentobarbital sodium solution (4%, w/v) via the ear vein at 15 min post dose. Subsequently, the eyes were rinsed with saline and the corneas were collected, weighed, and processed as described above. A 50 l aliquot of extracting solution was collected at 24, 48, and 72 h after soaking. The extracting efficiency was evaluated by comparing the concentrations of NAT in the extracting solution at different soaking times. The statistical analysis was performed using the paired-sample t test by the SPSS Statistics Ver.17.0. 2.8. Application of the method A total of 36 healthy male Japanese white rabbits (Henan Kangda Laboratory Animals Co., Ltd., China) weighing 2.0 to 2.5 kg were used for the ocular pharmacokinetics study. NAT ophthalmic suspension (50 l) was topically administrated and the cornea epithelium of half animals was also removed before instillation as described above. The treated animals were sacrificed at 15, 30, 60, 120, 240, and 360 min post dose. All animals were given free access to food and water during the study. Each time point involved three animals with intact epithelium and three animals with debrided corneas. Subsequently, the eyes were rinsed with saline, and the corneas were collected and weighed. All samples were stored at −70 ± 10 ◦ C before processing. QCs were distributed among calibrators and unknown samples in the analytical run. Concentration–time curve of cornea was plotted according to the determined values. Pharmacokinetic parameters of NAT in the
The stability of NAT in water or organic solvents is substantially affected by UV irradiation, extreme pH, heat exposure, oxidation, and presence of metal ions [15–17]. A suitable sample processing method involving simple procedures and protective measures is essential to obtain satisfactory accuracy and precision. Therefore, the “one-step” soaking extraction method with methanol under nitrogen protection was used for sample preparation. The mean extraction recovery of such “one-step” method was higher than 90%, whereas the reported extraction recovery of SPE method was approximately 75% [11,12]. Though the methanol soaking extraction method was simple and efficient, it would introduce more impurities compared with SPE method, including salts and electrolytes. The origin and mechanism of matrix effects are not understood fully and there are many possible sources of ion suppression [13]. The effects of salt or electrolyte concentration on analyte response observed through electrospray ionization MS are undoubtedly among the most important and definite factors [18]. The “solvent delay times” were set to start at 0 min and end at 1.5 min to prevent the salts and electrolytes in methanol extracting solution entering in the MS detector. The analyte response was sensitive and stable during the whole experiment. The optimization of MS parameters and MS transitions was performed according to the reported papers [13,19]. Amphotericin B was selected as internal standard due to its structural similarities with NAT [19]. The capillary voltage, cone voltage, desolvation temperature, desolvation gas flow, collision gas flow and collision energy were optimized thoroughly with flow injection analysis (FIA) mode. Detection and quantification of analytes were performed using the MRM mode. Both the analyte and internal standard have ability to accept proton and generate [M + H]+ ions (Fig. 1). The fragments were selected at m/z 503.2 and 906.6 as the most prominent fragments for NAT and IS, respectively (Fig. 1). The fragments of at m/z 503.2 and 906.6 for NAT and IS were both stable in the collision cell at low collision energy. Higher collision energy would split these two ions into smaller fragments and the signal intensity of ions decreased obviously. The water and acetonitrile combination were used as mobile phase according to the reported paper [11]. Though chromatographic retention and the spray droplets in the ESI interface could be improved by high organic content in the mobile phase [11], the chromatographic peak shape of NAT was becoming abnormal when the proportion of acetonitrile was higher than 60%. To get more stable and sensitive signal response, mobile phase modifiers such as acetic acid, formic acid, ammonium formate, ammonium acetate and their combinations were also studied. The ammonium acetate (4.0 mM) which could enhance the target ion abundance and maintained stable retention and ionization behavior was used as buffer of the mobile phase. 3.2. Validation procedures 3.2.1. Specificity and selectivity Chromatograms of blank cornea samples from six individual rabbits lacked co-eluting peaks of NAT and IS at their respective retention times. Typical chromatograms of blank cornea sample, spiked PBES and pharmacokinetic sample were shown in Fig. 2. 3.2.2. Calibration curve The peak area ratio of NAT to IS was linear over a NAT concentration range of 8.64 ng/ml to 843 ng/ml. The linear equation
56
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
Fig. 1. MS/MS spectra of (a) IS and (b) NAT.
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
57
Fig. 2. Typical MRM chromatograms of (a) methanol extracting solution of blank cornea at MRM 924.5/906.6 for IS and 666.2/503.2 for NAT, (b) PBES spiked with NAT (135 ng/ml) and IS (235 ng/ml), (c) pharmacokinetic sample at 15 min.
was “y = 0.0039x + 0.0036”, where “y” is peak area ratio of NAT to IS and x is concentration of NAT. The correlation coefficients (R) of the calibration curves for spiked samples was greater than 0.9999. The weighting factor 1/x0 was used to calculate correlation coefficient, slope and intercept. The standard deviation values for correlation coefficient (R), intercept and slope were 0.013, 0.0065 and 0.00002, respectively. The LLOQ value was 8.64 ng/ml, with a signal-to-noise ratio of 13. The mean relatively recovery of the LLOQ was 103.4%, and the RSD was 12.0% (n = 5).
3.2.3. Matrix effect The quotient of the peak area (A) of the spiked and control samples was used as an indicator of ion suppression or signal enhancement, which was calculated as [(Aspiked /Acontrol × 100) − 100]%. The indicator of matrix effect for NAT at 21.6, 135, and 337 ng/ml concentration levels was −2.2%, −6.2%, and −4.4%, respectively. Thus, no significant ion enhancement or suppression was observed.
3.2.4. Recovery The extraction recoveries of NAT from the spiked cornea samples at low (21.6 ng/ml), medium (135 ng/ml), and high (337 ng/ml) concentrations levels were 103.1 ± 12.8%, 92.7 ± 8.6%, and 94.6 ± 7.8%, respectively (n = 3). 3.2.5. Accuracy and precision Accuracy and precision data for intra-batch and inter-batch spiked samples were presented in Table 1. The intra-batch precision (% RSD) and accuracy (% bias) values for developed method were from 2.1% to 11.2% and −7.1% to 4.6%, respectively. The inter-batch
precision (% RSD) and accuracy (% bias) values were from 6.0% to 7.4% and −2.2% to 1.2%, respectively. 3.2.6. Stability The stability of the stock solution was calculated by comparing the peak area of the initial solution and that of the final solution using independent-sample t test. No significant variation for the stock solution of NAT and IS was observed after 2 months at −70 ± 10 (P > 0.05, n = 3) and after 72 h at 4 ± 2 ◦ C (P > 0.05, n = 3). The stability data of NAT in the PBES at 4 ± 2 ◦ C, −70 ± 10 ◦ C and bench-top were presented in Table 2. 3.2.7. Dilution study The mean back-calculated concentrations of 1/3 and 1/5 diluted samples for NAT were 99.3% and 97.5% according to their nominal values, respectively. RSDs were 4.8% and 6.0%, respectively. 3.3. Extracting efficiency To evaluate the extracting efficiency of methanol, the cornea samples collected from the treated rabbits were used as the NAT-containing cornea samples. The concentration of NAT in the extracting solution should stay at a certain level with the soaking time going on if the NAT has been extracted completely. The concentrations of NAT in the extracting solution at different immersion times were presented in Table 3. No significant variation (P > 0.05, n = 6) was observed between different soaking times. 3.4. Application of the method The rabbit cornea samples were processed and determined along with the QCs. The assay sensitivity and specificity were sufficient to accurately characterize the pharmacokinetics of NAT in
58
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
Table 1 Intra-batch and inter-batch assay precision and accuracy for NAT. Nominal conc. (ng/ml)
Intra-batch (n = 5)
Inter-batch (n = 15)
1st day
21.6 135 337
2nd day
3rd day
%Bias
%RSD
%Bias
%RSD
%Bias
%RSD
%Bias
%RSD
−3.3 −7.1 2.1
11.2 9.8 8.5
1.0 0.0 −1.9
5.4 8.4 5.6
−4.3 −4.6 4.6
2.8 3.2 2.1
−2.2 −0.8 1.2
7.4 8.9 6.0
Table 2 Stability of NAT in the cornea and PBES (n = 3). Nominal conc. (ng/ml)
Stability
Mean
SD
Precision (%RSD)
21.6
In cornea (−70 ◦ C, 4W) In PBES (4 ◦ C,72 h) Bench-top (24 h) In cornea (−70 ◦ C, 4W) In PBES (4 ◦ C,72 h) Bench-top (24 h) In cornea (−70 ◦ C, 4W) In PBES (4 ◦ C,72 h) Bench-top (24 h)
19.8 21.0 20.5 140.7 134.7 141.6 348.0 359.2 336.9
1.82 1.44 0.45 7.2 11.0 5.5 8.5 17.8 7.5
9.2 6.9 2.2 5.1 8.2 3.9 2.5 5.0 2.2
135
337
Accuracy (%bias) −8.3 −2.8 −5.1 4.2 −0.2 4.9 3.3 6.6 −0.0
Fig. 3. Concentration–time profile of NAT in rabbit cornea following topical administration of NAT ophthalmic suspension (5%). Table 3 Variation of NAT concentration in extracting solutions at different soaking time (t test, n = 6). N
Concentration of NAT(ng/ml) Debrided cornea
1 2 3 4 5 6
Intact cornea
24 h
48 h
72 h
185.3 213.6 169.5 155.7 193.8 122.9
165.2 198.5 169.6 136.9 211.3 133.9
177.4 220.3 175.8 149.8 215.8 126.9
24 h
48 h
72 h
P > 0.05 32.2 27.2 20.6 16.5 34.9 40.1
26.8 26.2 19.8 14.5 39.2 41.9
29.6 27.9 23.6 16.5 33.6 40.3
P > 0.05
rabbit cornea. No study samples with concentrations above the ULOQ were observed. The ocular pharmacokinetic profile of NAT ophthalmic suspension was similar to that of the sustained release formulation
as shown in Fig. 3. The NAT ophthalmic suspension flocculated to irregular clumps quickly after instillation. The drug clumps adhered to the conjunctival sac or the eyelid for a long time (Fig. 4), and such adherence strongly affected the pharmacokinetic profile of NAT ophthalmic suspension. However, the prolonged drug resident time failed to enhance the drug absorption substantially because of the low dissolving speed of NAT from the drug clumps and the relatively fast irrigation of the tear fluid [20,21]. The maximum concentration (Cmax ) of NAT in the intact cornea was 275.8 ± 158.2 ng/g and the area under the concentration curve (AUC(0–360 min) ) was 58,879.1 min ng/g. The Cmax of NAT in the debrided cornea was 1225.4 ± 355.8 ng/g, and the AUC(0–360 min) was 161,377.0 min ng/g. The corneal epithelium also affected the absorption of NAT in the cornea after topical administration. The half-life (t1/2 ) of NAT in rabbit cornea couldn’t be calculated accurately because of the irregular absorption of the ophthalmic suspension.
Z. Tianyang et al. / J. Chromatogr. B 969 (2014) 53–59
59
Fig. 4. Flocculation of NAT ophthalmic suspension (5%) on the ocular surface of rabbit at 30 min post dose.
4. Conclusion The current validated LC–MS/MS bioanalytical method for quantification of NAT in rabbit cornea was specific, sensitive, robust, precise and accurate. The sample preparation method based on methanol soaking extraction was simple, mild and efficient. The recoveries of the extracted cornea samples were higher and consistent over the investigated concentration range. From the results of all the validation parameters and applicability of the assay, we can conclude that the newly developed method can be useful for ocular pharmacokinetic studies of NAT with desired precision and accuracy along with high-throughput. Acknowledgment The authors Zhou Tianyang and Zhang Junjie are thankful to Mr. Zhou Jichun of Henan Provincial Institute of Food and Drug Control for his assistance in the animal experiment. References [1] P.A. Thomas, Eye 17 (2003) 852–862. [2] S. Xuguang, W. Zhixin, W. Zhiqun, L. Shiyun, L. Ran, Am. J. Ophthalmol. 143 (2007) 131–133. [3] R.A. Oechsler, T.M. Yamanaka, P.J. Bispo, J. Sartori, M.C. Yu, A.S. Melo, D. Miller, A.L. Hofling-Lima, Clin. Ophthalmol. 7 (2013) 1693–1701.
[4] F. Mohd-Tahir, A. Norhayati, I. Siti-Raihan, M. Ibrahim, Interdiscip. Perspect. Infect. Dis. 2012 (2012) 851563. [5] L. Pradhan, S. Sharma, S. Nalamada, S.K. Sahu, S. Das, P. Garg, Indian J. Ophthalmol. 59 (2011) 512–514. [6] Sonal S. Tuli, Clin Ophthalmol. 5 (2011) 275–279. [7] X.H. Dong, W.J. Gao, X.P. He, Int. J. Ophthalmol. 5 (2012) 143–146. [8] V.N. Prajna, P.S. Lalitha, J. Mascarenhas, T. Krishnan, M. Srinivasan, C.M. Vaitilingam, C.E. Oldenburg, A. Sy, J.D. Keenan, T.C. Porco, N.R. Acharya, T.M. Lietman, Br. J. Ophthalmol. 96 (2012) 1440–1441. [9] P. Lalitha, R. Vijaykumar, N.V. Prajna, A.W. Fothergill, J. Clin. Microbiol. 46 (2008) 3477–3478. [10] D.M. O’Day, W.S. Head, R.D. Robinson, J.A. Clanton, Curr. Eye Res. 5 (1986) 877–882. [11] Yashpal Singh Chhonker, Devendra Kumar, Pankaj Shrivastava, Deepak Kumar, Rajbir Singh, Hardik Chandasana, Rabi Sankar Bhatta, J. Pharm. Anal. 3 (2013) 144–148. [12] R.S. Bhatta, H. Chandasana, C. Rathi, D. Kumar, Y.S. Chhonker, G.K. Jain, J. Pharm. Biomed. Anal. 54 (2011) 1096–1100. [13] L.M. Repizo, L.D. Martinez, R.A. Olsina, S. Cerutti, J. Raba, Anal. Bioanal. Chem. 402 (2012) 965–973. [14] F. Fuselli, C. Guarino, A. La Mantia, L. Longo, A. Faberi, R.M. Marianella, J. Chromatogr. B 906 (2012) 9–18. [15] J.L. Koontz, J.E. Marcy, W.E. Barbeau, S.E. Duncan, J. Agric. Food Chem. 51 (2013) 7111–7114. [16] K. Thoma, N. Kübler, E. Reimann, Pharmazie 52 (1997) 362–373. [17] J.M. Gutteridge, A.H. Thomas, A. Cuthbert, J. Appl. Biochem. 5 (1983) 53–58. [18] T.L. Constantopoulos, G.S. Jackson, C.G. Enke, J. Am. Soc. Mass. Spectrom. 10 (1999) 625–634. [19] Niranjan M. Deshpande, Manish G. Gangrade, Maharudra B. Kekarea, Vikas V. Vaidya, J. Chromatogr. B 878 (2010) 315–326. [20] Indu Pal Kaur, Shilpa Kakkar, Expert Opin. Drug Delivery 7 (2010) 1303–1327. [21] J. Barar, A.R. Javadzadeh, Y. Omidi, Expert Opin. Drug Delivery (2008) 567–581.
