Pharmacokinetic Study of Ethambutol and Amikacin in Rat Lung Tissue Detected by UPLC-MS Method

Lower drug concentrations at tuberculosis (TB) infection sites may result in drug resistance. Understanding the tissue penetration of anti-TB drugs will offer new clues for optimizing drug dose. In this work, we performed pharmacokinetic study of ethambutol (EMB) and amikacin (AMK) in rat lung. Herein, a total of 160 rats were randomized to 16 time points, including before dosing (0h), and after dosing 0.25, 0.33, 0.5, 0.67, 1.0, 1.5, 2.0, 2.5, 3.0, 4.0, 8.0, 9.0, 12.0, 16.0 and 24.0h. The drug doses were 90mg/kg intragastric administration for EMB, and 0.20mg/kg intramuscular injection for AMK. The lung homogenate was precipitated with 15% trichloroacetic acid (TCA), diluted with 20% acetonitrile with 0.1% formic acid and 2% heptafluorobutyric acid. The compounds were detected by UPLC-MS/MS. The linear range detected by UPLC-MS/MS was from 10 to 5000 ng/mL (r 2 ≥ 0.99) for both drugs. The maximum tissue concentrations (C max ) were 39890 and 13855ng/mL at 2.25h for EMB and 0.40h for AMK, respectively. The terminal elimination half-life (t1/2 z ) was approximately of 3.43h and 14.56h, respectively. Their estimated tissue penetration ratios were 107% for EMB and 0.1% for AMK. In conclusion, in this work, a simple UPLC-MS method was developed to perform PK study of EMB and AMK in rat model. Although it is animal PK study, our work might provide a reference for the lung lesion concentration in TB patients. It is necessary to pay attention to AMK usage in severe pulmonary tuberculosis due to its low penetration. Pharmacokinetic Study of Ethambutol and Amikacin in Rat Lung Tissue Detected by UPLC-MS Method.


Introduction
Despite previous successes in management of tuberculosis (TB), the global emergence of multidrug-resistant TB (MDR-TB) and extensively drug-resistant (XDR)-TB has become highly problematic and has led to high mortality [1]. Poor outcome may be due to poor treatment adherence, lack of drugs and lower drug exposure, especially at the site of disease [2][3][4][5][6]. Therapeutic drug monitoring (TDM) of TB drugs has been used to optimize clinical treatment by measuring systemic drug levels in blood/plasma as a guide for individual dose adjustments [2,[7][8][9]. In MDR and XDR-TB treatment, first-line drugs such as EMB and isoniazid (INH) and second-line drugs are usually combined [10]. These second line drugs include several fluoroquinolone (FQ) compounds as well as the injectable drugs AMK, capreomycin (CAP), and kanamycin (KAN) [11]. According to previous reports, the concentration of some drugs in plasma is lower than the recommend ranges [8,11], and much lower in disease tissues including lung. Recently, researchers have tested drug concentrations in lung tissue [12,13], and lesion sites [13], as well as the dialysate [12][13][14] in the pulmonary cavity, to understand the drug penetration of lung tissue. It was found that clofazimine (CLA), moxifloxacin (MOX) and levofloxacin (LFX) has higher tissue penetration, RIF, PZA and linezolid (LZD) with median, and INH and KAN lower [12,13]. However, so far, the knowledge of the penetration of AMK and EMB in lung tissues was very limited. In this work, we developed an UPLC-MS/MS method to quantify AMK and EMB concentration in lung tissues, and carried out pharmacokinetic study in rat model.

Solution
The stock solutions of EMB and AMK, at a concentration of 1.0mg/mL, were prepared using 10% methanol, and stored at -80 ℃. The working solutions containing EMB and AMK were prepared by diluting the mixed stock solution with 20% methanol to obtain different concentrations (25,000, 22,500, 5000, 1000, 500, 100, 50ng/mL). The IS working solutions were prepared with 20% methanol to obtain the concentration of 0.25µg/mL for EMB-d4, and 5µg/mL for APM. All working solutions were stored at 4 ℃. The standard curves and the quality control (QC) samples containing EMB and AMK, were obtained by mixing the corresponding working solutions (40µL), IS (40µL) with 200µL of blank tissue homogenate.

Sample Preparation
Tissue homogenate (200µL), 40µL IS and 40µL 20% methanol were mixed, and protein precipitated by 480µL trichloroacetic acid (TCA) at the concentration of 15%. The mixture was vortexed for 3min and centrifuged at 12,000rpm at 4 ℃ for 10min. Then the supernatant (100µL) was diluted with 100µL solution (20% ACN, 0.1% FA and 2% HFBA, and transferred to an auto-sampler for analysis by UHPLC-MS/MS. The injected volume was 5µL.

Method Validation
Method validation was performed according to Chinese pharmacopoeia 2015, and FDA guidelines [15]. The method was validated for selectivity, carryover, linearity, accuracy, precision, matrix effect, recovery, and stability. The details were shown in the supplement materials.  The lung tissues were collected, homogenized in PBS (weight to volume of 1:5, w:v), and centrifuged at 12,000rpm for 10min.

Pharmacokinetic study in Rat
The supernatants were collected and stored at -80 ℃ until further analysis.

Data Analysis
All data were expressed as mean ± standard deviation (SD). The PK analysis was performed in DAS 2.0 software, and PK parameters from non-compartment model were collected, including Cmax (the maximal drug plasma concentration); T max (the time to reach C max ); AUC 0-t (the area under the plasma concentration-time curve from 0 to the time t); AUC 0-∞ (the area under the plasma concentrationtime curve from 0 to infinity); t 1/2 (the elimination half-life); CL (clearance), and Vd (apparent volume of distribution).

Incurred Sample Reanalysis
An incurred sample reanalysis (ISR) was carried out to evaluate the reproducibility of the analytical method. Two samples from each time point were randomly selected, except that before dosing, after dosing at 4.0, 12.0, and 24.0h. Not enough volume was left for the samples at 4.0 and 12.0h time points. The receipt standard for ISR was a percentage change less than 20% in 67% samples.

Supplement Methods
Selectivity and Carryover: To investigate the selectivity of the method, six different sources of blank rat lung tissue homogenate were used to make blank, blank spiked with EMB (IS, EMB-d4) and AMK (IS, APM) at the concentration of LLOQ. In addition, lung tissues from a rat after an administration of EMB and AMK for 3 h were prepared. The presence of interfering peaks in the retention times of the analytes and their ISs was evaluated. Carryover was checked by analyzing a blank matrix sample following HQC.
Linearity: Seven standard curve samples with a series of concentrations ranging from 10-5000ng/mL were prepared as described in "Preparation of standard curve and quality control solution" The analyte-to-IS peak ratios were plotted against the corresponding EMA or AMK concentrations. 3) The stability of freeze-thaw performed by storing at -80 ℃ and thawing at room temperature every 24h for three cycles, 4) Storage at -80 ℃ for 30 days.   in the range of 10-5000ng/mL for both EMB and AMK in rat lung homogenate. The lower limit of quantification (LLOQ) was both 10ng/mL. The intraday accuracy ranged from -5.52% to 1.8% for EMB and -6.6% to 7.08% for AMK, respectively. The interday accuracy ranged from 0.43% to 3.36% for EMB and -2.57% to 5%

UPLC-MS Method
for AMK, respectively. The intraday and interday precision of EMB and AMK were all less than 11.14%. These data indicated that the method had good precision and accuracy ( Table 1). The mean recovery rate of IS normalized EMB and AMK ranged from 98.62% to 106.12% and 97.01%-104.22%, respectively. The matrix effects were from 96.33%-104.14% for EMB, and 94.23%-104.38% for AMK ( Table 2). The pre-and post-preparative, freeze-thaw, and long-term stability were measured by comparing six replicates of two QC samples (low and high levels) of EMB and AMK. As shown in (Table 3), the stability results (RSD% and RE%) were all less than 15% in rat lung homogenate under six storage conditions: room temperature or 4 ℃ for 24h pre-or post-preparation, freeze-thaw, and -80 ℃ for 30 days.

Pharmacokinetic Results
The and AMK, respectively. In order to detect the accuracy of analysis results, we performed ISR of 24 samples. As shown in (Table 5), the differences between ISR and initial concentration were less than 20% of their mean value except 2 samples including one 30min time sample with 30.3% for EMB, and one 20min time sample with -21.7% for AMK.
TCA was chosen as it is dependent on the trichloro-group, and has higher extraction recovery for hydrophilic compounds (such as AMK) [17,21,26]. Furthermore, to improve the retention time of AMK (an aminoglycoside compound) in C18 column, an ion-pairing agent of HFBA was added. HFBA exhibited better retention of aminoglycosides on the analytical column and was also better at enhancing the electrospray process than other ion pair reagents [27,28]. The verified UPLC-MS/MS method was then used for quantitation in a PK study of EMB (i.p.) and AMK (i.m.).
Our results showed that EMB had very strong lung penetration (107%), however, AMK was very weak (0.1%). The C max of AMK in lung detected in this work was similar to that previously reported [29,30]. Our results indicate that the developed method was available to detect the concentration of EMB and AMK in the lung tissues from MDR-TB patients.

Conclusion
In this work, a simple UPLC-MS method was developed to perform PK study of EMB and AMK in rat model. Although it is animal PK study, our work might provide a reference for the lung lesion concentration in TB patients. Our study showed that the lung penetration of AMK was very weak. It is necessary to pay attention to AMK usage in severe pulmonary tuberculosis due to its low penetration.