Studies on the metabolism and degradation of vancomycin in simulated in vitro and aquatic environment by UHPLC-Triple-TOF-MS/MS

Vancomycin is one of the most commonly used glycopeptide antiobiotics, and as such is an important emerging environmental contaminant. Pharmaceuticals and personal care products (PPCPs), such as antibiotics, are problematic since wastewater treatment processes are not completely effective at removing these chemical compounds. Since wastewater treatment processes are not completely effective, vancomycin occurs in surface water. Vancomycin and its metabolites in vivo and degradation products in aquatic environment may lead to undesirable ecological effects that threaten the environment or cause undesirable reactions that affect human health. We aimed to study vancomycin metabolism in vitro and its natural degradation in aquatic environment, as well as explore for related metabolites and degradation products. Accordingly, we established four systems, using a constant temperature oscillator at 37 °C for 10 days for vancomycin in activated rat liver microsomes (experimental system), inactivated rat liver microsomes (control system), phosphate buffer saline (PBS system) and pure water (pure water system), as well as an additional system of activated rat liver microsomes without vancomycin (blank system). The metabolism and degradation of vancomycin were studied using a high resolution and high sensitivity ultra-high performance liquid chromatography (UHPLC)-Triple-time of flight (TOF)-mass spectrometry (MS) method in positive ion mode. The compared result of activated rat liver microsomes system and inactivated rat liver microsomes system confirms that vancomycin is not metabolized in the liver. Vancomycin was degraded in the four non-blank incubation systems. The MetabolitePilot 2.0 software was used for screening the probable degradation products, as well as for establishing its associated degradation pathways. Eventually, four degradation products were identified and their chemical structures were deduced. The results of this study provide a foundation for evaluation of the effects of vancomycin and its degradation products on environmental safety and human health in the future.

The full-scan MS spectra and MS/MS spectra data were obtained in positive electrospray ionization (ESI) mode by Analyst TF 1.6 software (AB Sciex, USA). The optimized parameters of the high resolution mass spectrometer are shown in Table 1. Full scan mode was applied, the parent ions scan ranged from m/z 100 to 2000 Da with a 200 ms accumulation time and the product ions scan range from 100 to 1500 Da with a 70 ms accumulation time. Simultaneously, the calibration delivery system calibrated mass numbers by every five samples to acquire exact mass of ions. The running time of data independent acquisition (DIA) was 20 min.
Preparation of rat liver microsomes. Ten male Sprague-Dawley (SD) rats (250 ± 20 g) were provided by the Experimental Animal Research Center at Hebei Medical University. The animal experimentation were approved by the Ethics Committee of Hebei Medical University (Approval Number 2018012). Rat liver was removed immediately after decapitation and then weighed blotting the excess blood with a filter paper. Liver microsomes were prepared by differential centrifugation after the liver was repeatedly washed with a cold solution (10 mmol/L Tris-HCl, 250 mmol/L sucrose and 1 mmol/L ethylene diamine tetraacetic acid (EDTA) at pH 7.4), shredded and homogenized on ice. A full description of the procedures can be obtained elsewhere 32 . Eventually, the Lowery protein assay was used to measure protein concentrations in the rat liver microsome suspension 33 . Experimental design. In this study, five systems (experimental, control, blank, PBS and pure water systems) were established. The five systems each included 11 samples, in a final volume of 200 µL. All systems, except the blank system, contained a final vancomycin concentration of 60 mg/L 26 . The experimental, control and blank systems contained PBS and MgCl 2 at a final concentration of 3.3 mmol/L. The experimental and blank systems contained 1 mg/mL of activated rat liver microsome protein, and the control system contained 1 mg/ml of inactivated rat liver microsome protein. After preheating the incubation system at 37 °C for 5 min, NADPH at 1 mmol/L was added to the experimental and blank systems, while the same volume of pure water was added to the control system. Vancomycin was diluted in PBS and pure water in the PBS system and pure water system, respectively (details in Table 2). One sample from each system was placed directly in the freezer at −20 °C (0 h), and the remaining samples of the incubation systems were placed in a constant temperature oscillator at 37 °C for 1-10 days. One sample from each system was removed every 24 hours for ten consecutive days, adding 1 mL pre-cooled methanol immediatel yafter removal to the experimental, control and blank systems. All samples were stored at −20 °C until further analysis.

Pretreatment of samples.
All the samples were centrifuged (10,625 g for 10 min) including experimental, control and blank systems. The supernatant was completely transferred to a clean glass tube and dried with a slow stream of nitrogen at 37 °C. The dried substances were dissolved with 200 µL 30% methanol before determining.
Quality control (QC) sample. A QC sample of vancomycin in pure water was prepared with a concentration of 60 mg/L. The same QC samples were injected at the start of the run and then again every eleven samples.

Results and Discussion
Degradation of vancomycin. During the 1-10-day incubation period, the concentration of vancomycin (M0) decreased in the experimental, control, PBS and pure water systems (Fig. 1). A similar degradation rate in the experimental and control systems over the early time period of the incubation, when the microsomes are viable, may be attributed to the weak effect of activated rat liver microsomes cytochrome P450 enzymes on vancomycin. The results confirmed that vancomycin is not metabolized in the liver and most of the vancomycin ingested in the body is excreted in the urine. The most marked loss of vancomycin was in rat liver microsomes, which was degraded 50% over approximately 6 days. The time to 50% loss of vancomycin in PBS or in water was approximately 9 days and over 10 days, respectively, which showed that vancomycin can be degraded in an aquatic environment. Organics and mineral salt may be favorable to the degradation of vancomycin.
Mass spectrometry fragmentation patterns of vancomycin. The extracted ion chromatogram and the MS fragmentation patterns of vancomycin, reveal that vancomycin eluted at 13.04 min with a parent ion at m/z 724.7246 with two charges (Fig. 2a). The chemical structure of vancomycin is very complex, and includes aminoglycoside and polypeptide. Combined with the MS/MS spectra of vancomycin (Fig. 2b), the MS/ MS fragmentation behaviors were analyzed by inference and the 'Fragmentation Library function' in the Mass

Identification of the degradation products. The procedure to identify likely degradation products. The
Triple TOF MS/MS instrument, data independent acquisition (DIA) and various techniques to process the raw data were used to identify likely degradation products. First, the raw data from the experimental, control, PBS and pure water systems were compared with the blank system and a prepared standard solution of vancomycin using the MetabolitePilot 2.0 software, and the likely degradation products were acquired. Subsequently, the data for likely degradation products were acquired and confirmed using many data-mining functions in the PeakView software, such as extracted ion chromatogram (XIC), information dependent acquisition (IDA) explorer, mass calculator and formula finder. Ultimately, four likely degradation products (M1, M2, M3 and M4) of vancomycin (M0) were identified. The biotransformation pathway, molecular formula, retention time, theoretical and observed mass, mass error and score for vancomycin and each of its four likely degradation products are outlined in Table 3. For high quality and accurate mass spectra, the relative error (RE) between the theoretical and observed values for the ion mass were less than 5 ppm 38,39 . Accordingly, the inferred element composition of the four likely degradation products was acceptable, as the RE for vancomycin and the four likely degradation products was less than 5 ppm. Using the XIC function in the PeakView software, the extracted ion chromatograms of the experimental, control, blank, PBS and pure water systems were obtained by applying the characteristic ion at m/z 724.7245, 725.2171, 725.2167, 731.2164, 730.7239 of vancomycin and the four likely degradation products. The comparison between day 0 and day 5 of incubation is shown in Fig. 4. The chromatograms reveal that M1, M2, M3 and M4 were present in the experimental and control systems. Additionally, no degradation products were detected in the blank system, but M1, M2 and M4 were detected in the PBS system, and only M1 and M2 were detected in the pure water system. The contents of the degradation products. The relative standard deviation (RSD) of the vancomycin intensity in the QC samples was 14.41%, which was less than 15%, thus indicating that the repeatability and stability were acceptable 40,41 . Based on the definition that the total content of vancomycin and its four degradation products is 100%, the related content of vancomycin and its four degradation products were calculated using the normalization method. The trends for vancomycin and the four degradation products during the 1-10-day incubation period in the experimental, control, PBS and pure water systems are shown in Fig. 1. In the experimental and control systems, M1 and M2 increased with time over the first seven days and then decreased slightly from day 8. Compared with the other three systems, M1 and M2 had the fastest increase in the PBS system, with M2 increasing faster than M1. This may be due to the stimulation and inhibition of the formation of M1 and M2 by the mineral salt and organics, respectively. M3 was only produced in the experimental and control systems. Compared with the other three degradation products, M3 was the lowest produced in the first seven days and then exceeded M1 after seven days. M4 showed an obvious upward trend with the longer time in both the experimental and control systems, and increased at the fastest rate of the four degradation products. M4 showed a trend to increase in the first two days in the PBS system and then decreased slightly from day 3. The M4 content was the lowest in the PBS system, and was not found in the pure water system. A possible explanation for this is that organics were favorable to M3 and M4 production.
Structural deduction of the degradation products. The degradation products formed by biotransformation generally retain the basic skeleton and some substructures of the original parent molecule. This means that the degradation products may have similar cleavage laws to the raw drugs in HPLC-MS/MS analysis. Vancomycin contains two amino groups (one on the aminoglycoside moiety and the other on the polypeptide component). The MS/MS spectra of vancomycin and its four degradation products (Figs 2b, 5b, 6b, 7b and 8b) reveal that all the compounds contained the same fragment ions at m/z 100.07, 118.08 and 144.10. These fragment ions were produced from the fragmentation of the aminoglycoside fragmentation of vancomycin, thus the amino moiety in the aminoglycoside structure is consistent across all four degradation products. Based on the finding that vancomycin easily loses the CO part from the polypeptide under collision energy, we reasoned that the degradation products had similar principles and this provided the basis for deducing their chemical structures.
The retention time for M1 ([M + 2H] 2+ at m/z 725.2171) was 12.64 min (Fig. 5a). M1 was acquired by the biotransformation process of oxidative deamination to alcohol, which suggested that M1 was obtained by changing the amino group in the polypeptide structure of vancomycin. M1 lost part of the aminoglycoside and one CO to produce the characteristic product ion at m/z 1116.2826. The likely structure of the characteristic product ion at m/z 1116.2826 is shown in Fig. 5b.   (Fig. 6a). The biotransformation process for M2 involved the loss of CH 3 N, and oxidation and methylation. Vancomycin contains only one CH 3 N group within the polypeptide structure. M2 lost the part of aminoglycoside to produce the characteristic product ion at m/z 1144.2798 (Fig. 6b). M1 and M2 are isomers, and the polarity of M1 is greater than that of M2 according to the value of the CLogP. As a result, M1 eluted before M2.
Vancomycin lost a CH 3 N group and was N-acetylated to produce M3 ([M + 2H] 2+ at m/z 731.2164), which eluted at 13.90 min (Fig. 7a). The loss of CH 3 N was the same as for M2, and the position of the change in the amino moiety was the same as for M1. The likely characteristic product ion at m/z 1128.2832 was acquired by M3, which lost the aminoglycoside component and a CO (Fig. 7b).
M4 ([M + 2H] 2+ at m/z 730.7237) was obtained by the biotransformation process involving the loss of water and oxidation and methylation and eluted at 14.30 min (Fig. 8a). Vancomycin contains many hydroxyl groups, and two adjacent hydroxyl groups could be dehydrated to form ethers. The likely characteristic product ion at m/z 1099.2981 was acquired by M4, which lost the aminoglycoside and two CO moieties (Fig. 8b).
The Fragmentation Library Function in the Mass Frontier 5.0 software played a guiding role in the process to deduce the chemical structure of the four degradation products. The chemical structures of the four degradation products are illustrated in Fig. 9. The theoretical and observed MS and MS/MS mass of the four degradation products are listed in Table 4. All relative errors of the ion mass were less than 10 ppm.
The degradation pathway of vancomycin. In this study, a total of 4 metabolites were identified. The plausible degradation pathways of vancomycin are shown in Fig. 9. Additionally, the principle of the degradation of vancomycin into four metabolites was described in detail in the "Structural deduction of the degradation products" section above. The results showed that the amino group, methylamino group, adjacent hydroxyl group and hydrogens on the alkyl group in the polypeptide were the main degradation sites.

Conclusions
We used a high resolution and high sensitivity UHPLC-Triple-TOF-MS/MS method in positive ion mode to investigate the degradation of vancomycin in activated and inactivated rat liver microsomes, PBS and pure water systems. The compared result for activated rat liver microsomes system and inactivated rat liver microsomes system confirmed that the effect of activated rat liver microsomes cytochrome P450 enzymes on vancomycin is so weak that vancomycin is essentially not metabolized in the liver. A comparison of the degradation trends of vancomycin in the four systems, reveals that organics and mineral salt may be favorable to the degradation of vancomycin in aquatic environment. Ultimately, four likely novel degradation products were identified. The structural characteristics and cleavage of the four novel degradation products were deduced based on the mass spectra analysis of vancomycin. These finding could be a foundation for future research on the effects of vancomycin and its degradation products on environmental safety and human health.