Metabolite Profiling of Meridianin C In Vivo of Rat by UHPLC/Q-TOF MS

Meridianin C (MC), as a marine alkaloid, is a potent protein kinase inhibitor which exhibits good anticancer activity. However, the in vivo metabolism of MC has not been described to date. In this study, an ultra-high performance liquid chromatography/quadrupole time-of-flight mass spectrometry (UHPLC/Q-TOF MS) method is employed to investigate the in vivo metabolites of MC in rats. Plasma, bile, urine, and feces are collected after a single oral dose of MC. Protein precipitation, solid phase extraction (SPE), and ultrasonic extraction methods are used to prepare samples. Based on the mass spectral fragmentation patterns, elution order, and retrieving literatures, a total of 13 metabolites of MC were detected and tentatively identified, utilizing MetaboLynx software. The metabolic pathways of MC in rats include N- or O-glucuronidation, O-sulfation, N-hydroxylation, dihydroxylation, and trihydroxylation. The relative content of the metabolites in each kinds of biological samples is also evaluated. This study will help to understand the in vivo properties of MC for the future usage.


Introduction
It is universally acknowledged that the discovery and development of drugs is a time-consuming and costly procedure [1]. Absorption, distribution, metabolism, and excretion (ADME) of compound is one of the important considerations in the drug discovery process and can help investigate the drug after administration or injection. And so, ADME characteristics play an important role in drug design. However, drug discoveries and developments are commonly terminated in case of preclinical or clinical prodrug toxicity. Biotransformation of compound in vivo is thought to be one of the factors of drug toxicity [2]. Nowadays, the prodrug screening approach has been performed at a very early stage of drug discovery even drug development. Drugs after oral administration may always be biotransformed by intestinal flora and various metabolizing enzymes to different kinds of phase I and phase II metabolites, which are related to the parent drug. Furthermore, drug metabolic pathways are key factors in optimizing prodrugs for optimal pharmacokinetics properties.
Metabolic pathways of prodrug can help finding latent active metabolites and minimizing potential safety due to biotransformation of reactive or toxic metabolites [3]. e LC/MS technology has been used as an important tool in drug discovery and drug development, owing to its selectivity, sensitivity, and speed. e quadrupole time-offlight mass spectrometry systems can easily provide highresolution mass and accurate mass data, which can efficiently help to calculate the molecular of metabolites. In the past few years, ultrahigh performance liquid chromatography (UHPLC) has become the common approach of ingredients separation for the LC/MS system in the cause of the drug metabolic study, for providing higher throughput, sensitivity, and/or improved selectivity than the traditional HPLC method. So, UHPLC combined with quadrupole time-of-flight mass spectrometry systems has become a commonly technology of the drug metabolism study [4].
Natural products have played close attention in drug discovery because of their easy degradation, environmental friendliness, and unique mechanisms of function [5][6][7][8][9]. Meridianins A-G ( Figure 1) are natural products isolated from marine organisms. Since the isolation of meridianins A-E in 1998 [10], a total of seven of these alkaloids have been isolated (meridianins A-G) [11]. Meridianin alkaloids exhibit a lot of biological activities, such as anticancer [5,10], protein kinase inhibitory activity [11], antimalarial activity [12,13], antituberculosis activity [13], anti-Alzheimer's disease activity [14], and so on. Especially, meridianin C has been studied showing dramatical anticancer activities [15,16]. To our knowledge, meridianins A-G are still not been used as drugs in clinical, and the in vivo metabolisms have not been studied yet. In the present study, UHPLC/Q-TOF MS is used to characterize and semiquantify the metabolites of MC in rat plasma, bile, urine, and feces for purpose of providing evidence of the course of meridianins in vivo.

Chemicals and Reagents.
MC was prepared in our lab. e structure was determined using 1 H NMR (Supplementary Figure S1) and mass spectrometry (MS) methods. e HPLC grade formic acid and acetonitrile were prepared form ermo Fisher Scientific Co. Ultrapure water (18.2 MQ) was prepared with a Milli-Q water purification system (Millipore, France). [17]. Male Sprague-Dawley rats (220-250 g) were invited from the SPF (Beijing) Biotechnology Co., Ltd. (Beijing, China). ey were adapted to new circumstances for one to seven days. e rats were fasted overnight before the drug metabolic study and only allowed free access to water.

Animal Experiments
Meridianin C powder was dispersed in 0.5% sodium carboxymethylcellulose (Na-CMC) solution at concentration of 10 mg/mL and then sonicated until a homogeneous suspension is obtained, which would be used for oral administration. Plasma samples were collected from the suborbital vein to obtain blank plasma. After oral administration of meridianin C at a dose of 100 mg/kg bodyweight to two rats, blood was collected from the suborbital vein at the time points of 0.5, 1, 2, 4, 8, 12, and 24 h, and all the blood samples were immediately centrifuged at 3,500 rpm for 10 min at 4°C to obtain supernatant plasma and stored at −80°C until analysis. Two rats were used for urine and feces sample collection. Both urine and feces were collected for 0-24 h after oral administration of MC at the dose of 100 mg/kg bodyweight. Urine samples were collected every two to four hours and stored at −20°C for later analysis. Feces samples were also stored at −20°C after drying. Two rats were used for bile collection in the period of 0-24 h. After collecting blank bile for a certain amount, rats were orally administrated with the same dosage of MC. e experimental protocol was approved by the Animal Experiment Ethics Review Committee of Jiangsu Vocational College of Medicine (LLSQ-2021-031106). [17] 2.3.1. Plasma. First, 200 μL of mixed plasma was diluted with 600 μL methanol, vortex-mixed for 2 min, centrifuged at 8,000 rpm for 10 min to precipitate proteins, and the supernatants were evaporated to dryness transferred in another tube. Second, the residue was reconstituted with 200 μL methanol-water (80 : 20, v/v) and centrifuged at 14,000 rpm for 10 min. Finally, the supernatants were passed through a 0.22 μm membrane filter, and 1 μL sample was injected into the UHPLC/Q-TOF MS system for analysis.

Urine.
e solid phase extraction (SPE) method was used for handling urine samples with Bond Elut C18 cartridges (3 mL, 500 mg). Each cartridge was activated with 2 mL methanol and conditioned by 2 mL water. en, 100 μL of the spiked urine samples was loaded on to the cartridges. en, 2 mL water and 2 mL methanol were added separately to the SPE cartridges to elute the matrix and the metabolites. e methanol elution was collected and dried. en, the residue was redissolved with 200 μL methanol-water (80 : 20, v/v) and centrifuged at 14,000 rpm for 10 min. After the supernatants were passed through a 0.22 μm membrane filter, 1 μL samples of the filtrates was injected into the UHPLC/Q-TOF MS system for analysis.

Bile.
Bile samples were also prepared with the SPE method, and the procedure was similar to urine sample, except that 100 μL bile residue was redissolved with 150 μL methanol-water (80 : 20, v/v).

Feces.
First, the spiked drying feces were grinded into fines. Weighted 1 g feces sample into a 50 mL tube and added 20 mL methanol-water (80 : 20, v/v). Second, 15 min ultrasonic extraction was used for feces sample preparation and then centrifuged at 3,500 rpm for 10 min to participate impurity. A curtain amount of the supernatants were continually centrifuged at 14,000 rpm for 10 min and passed through a 0.22 μm membrane filter and used for mass analysis.

UHPLC and MS
Conditions. e chromatography separation was carried out using a Waters ACQUITY BEH C18 column (2.1 × 100 mm, 1.7 μm, Waters, Milford, MA, USA) at the column temperature of 40°C. e mobile phase consisted of (A) water containing 0.1% formic acid and (B) acetonitrile containing 0.1% formic acid with a gradient elution as follows: 0-1 min, 5% A; 1-10 min, 5%-95% A; and 10-12 min, 95% A. e flow rate was maintained at 0.4 mL/ min, and 1.0 μL of the control and MC-containing biological samples were injected into the system at each run.
Metabolite identification of MC in rat was performed on Waters quadrupole time-of-flight (Q-TOF) Xevo G2 (Waters, Manchester, UK) equipped with an electrospray ionization (ESI) source. e capillary voltage of the ESI source was set at 3.0 kV in the positive ionization mode. e cone voltage was set at 40 eV. Source and desolvation temperatures were set at 120°C and 450°C, respectively. e nebulization gas flow was set at 800 L/h. All data were acquired with the MS E method in the resolution mode with 6 eV as low-collision energy and 20-60 eV ramp as high-collision energy. Both the molecular ions and the mass fragment ions could be acquired in a single run. Mass data were collected over the range of m/z 100-1500 in both functions. Leucineenkephalin (m/z 556.2771) was used as real-time lock mass.

Mass Spectrometric Analysis of MC.
e structures of meridianins A-G are shown in Figure 1. ere are five main fragment ions of MC in the high-resolution spectrum ( Figure 2). e speculated formula, observed and calculated mass, and mass errors are given in Table 1. e detected mass errors ranged from 0 to −5.3 mDa, indicating a good accuracy. e fragment ion structures are shown in Figure 2. Some fragments displayed a few ion clusters, and only one ion is given. [17]. A novel strategy was proposed for the systematic screening and characterization of MC metabolites. First, the MSE data of both the blank control samples and the biological samples containing MC were acquired using UHPLC-Q/TOF MS. Second, with the help of MetaboLynx TM software, the molecular weights, the elemental compositions, and the proposed metabolic pathways of the metabolites derived from the accurate mass measurements can be automatically predicted. Finally, the structures of metabolites were justified based on the mass difference, the characteristic fragments of MC metabolites, and available literatures data. e presence of the isotope pattern resulting from bromine atoms means that it is not lost via the metabolism.

Metabolites Identification of MC
Both Br 79 and Br 81 isotopes were monitored by the mass spectrometer.
e corresponding high-resolution mass spectrum displays identical isotopic cluster in an 1 : 1 ratio with two Da difference, thus confirming the presence of one bromine atom in the structure (such as m/z 289.01/291.01 in Figure 2). At the same time, the high-resolution mass spectrum without the isotopic cluster in 1 : 1 ratio indicates that the molecular has lost the bromine atom in the fragmentation process (such as m/z 209.08, 193.07, 169.08, and 140.05 in Figure 2). e extracted ion chromatograms (EICs) of the prototypes and metabolites in MC-containing rat biological samples are shown in Figure 3, which were generated with MetaboLynx software and also be confirmed by manually extracting in the matrix (Supplementary Figure S2). e parameters of the 14 metabolites (including the prototype) are given in Table 2.

Identification of Parent Drug in Rat Biological Samples (M10).
e metabolite M10 displayed the same molecular ion at 289.01/291.01 (C 12 H 9 BrN 4 , retention time 4.21 min), and the fragment ions were similar to those of MC. M10 was identified as the parent drug MC by comparing the retention time and accurate mass spectra with that of the authentic standard.

Hydroxylated Metabolites (M3, M6, and M12).
M3, M6, and M12 displayed the same molecular ion at m/z 305.00 (C 12 H 9 BrN 4 O, retention times 3.24, 3.61 and 4.47 min), which was 16 Da (O) higher than the protonated molecule of MC. e characteristic fragment ions at m/z 289/291 indicate the loss of O atom. erefore, M3, M6, and M12 were identified as hydroxylated metabolites of MC. By retrieving literature [18], the most probable conjugate cite was at the N position. e diagnostic ion of M3 and M6 at m/ z 185 was 16 Da higher than the characteristic ion at m/z 169 of MC ( Figure 2). us, it was proposed that the O atom conjugated at the A or B rings. 1-N-site in B ring was the most proposed conjugated site [19], and thereby, M3 with a relative higher amount should be 1-N-hydroxylate MC. M12, without the ion of m/z 185, was speculated to be 2′-Nhydroxylate MC. e diagnostic ion of M2 at m/z 185 was 16 Da higher than the characteristic ion at m/z 169 of MC (Figure 2), and thereby, the O atom conjugating at the A or B rings was put forward. 1-N-site in B ring was the most proposed conjugated site [18,19]. While the absence of m/z 185 suggested that the O atom might conjugate at C ring, the most proposed conjugated site might be at 2′-N-in C ring. e subsequent fragmentation ions at m/z 170, 143, and so on are similar to those of MC, which was undergoing an easy loss of O atom followed by subsequent fragmentation. e characteristic fragment ion of M11 at m/z 198 generated a relatively high response, which was the absence of M2. It is tentatively deduced that the fragment m/z 198 was generated based on a rearrangement owing to fragmentation, when O atom conjugated at 2′-N position, as shown in Figure 1. Meanwhile, the 2′-N-O-S bond in C ring speculated easier break-up, encountering a high collision energy than 1-N-O-S bond in B ring because the extraction ion chromatograph of protonated molecular ion of M11 was hard to see. e calculated 1-octanol/water partition coefficient (CLog P) value was an important parameter for the metabolite identification, as an ingredient with a higher CLog P value may be eluted later when separated in the reversed-phase column.

Journal of Analytical Methods in Chemistry 5
176 Da were indicating a glucuronide conjugate. In the high-resolution channel, they displayed the same fragment ion at m/z 289/291, and so, they were speculated to be glucuronide metabolites of MC. Among them, 1-N-O-and 2′-N-O-were the two possible conjugated sites [20][21][22]. We tentatively deduced M4 and M7 to be MC-1-N-glucuronidation and MC-2′-N-glucuronidation by their elution order.  (hydroxylation + glucuronide conjugation) + 2× alcohols dehydration, with the mass difference of 1.4 mDa. It is obvious that the skeleton of MC could not lose two H 2 O, and so, the alcohols dehydration may happen at the conjugated glucuronide site. Given this, further study is needed to justify this presumption.

Relative Content of Metabolites.
e peak area was calculated using MetaboLynx (Figure 3). e content of each metabolite was displayed as the percentage of each extraction ion chromatograph peak areas relative to the total peak areas in plasma, urine, bile, and feces sample. As the biological samples at each time were spiked together in a certain period of time, the relative content could give some useful message for the ADME study of MC in the future. As given in Table 3, about 75.8% of MC was displayed as prototype in plasma, and 11.7%, 9.1%, and 57.7% of MC were displayed in bile, urine, and feces, respectively. About 51.5% hydroxylated and glucuronide metabolites, as well as 20.9% hydroxylated and sulfate metabolite, existed in bile. e relative amount of metabolites in urine was similar to those in bile. Fecal samples contained the most kinds of metabolites, and hydroxylated metabolites were the major phase I metabolites, occupying nearly 20% metabolites. 13.0% hydroxylated and sulfate metabolites were the major phase II metabolites.

Result of Analysis.
e UHPLC/Q-TOF MS method, combined with MetaboLynx, was employed for rapid analysis of the metabolites of MC in vivo of rats. Acetonitrile and water containing 0.1% formic acid as the mobile phase and collision energy at 20-60 eV displayed good mass fragments of the metabolites. MetaboLynx can rapidly excavate the metabolites by comparing the blank control sample with the samples collected after administration of MC, which will dramatically reduce the mental work extraction ion chromatograph one by one and increase the accuracy and integrity. As the prototype contains bromine, Journal of Analytical Methods in Chemistry we set the parameters of the isotopic mode, which will reduce the processing time from a few hours to a few minutes. is strategy can also be applied to evaluate the other compound metabolisms containing bromine or chlorine.
A total of 14 metabolites including the parent compound MC were found in vivo of rat. e prototype of MC displayed a high concentration in plasma, bile, urine, and feces, proposing a higher bioavailability of MC after oral administration (Table 3). At the same time, natural products in vivo mainly undergo the phase II metabolism including glucuronide and sulfate conjugates to increase polarity and decrease toxicity, then excreting out of the body. e glucuronide or sulfate conjugation at MC detected the minor metabolism, compared with glucuronide or sulfate metabolites after hydroxylation of MC, which could boil down to no hydroxyl groups in the parent skeleton ( Figure 1). It could be inferred that N-site was not active as O-site encountering the conjugated metabolism [21]. Furthermore, no sulfate conjugated at N-site metabolites was detected in all of the matrices, except a few N-glucuronide conjugation, so we inferred that the N-sulfate conjugation could hardly happen in vivo of rat [23]. e feces sample detected the most kinds of metabolites, and most of them also existed in other biological matrices. A relatively high amount of hydroxylated metabolites (almost 20%) of MC in feces suggests that MC was first transformed by intestinal bacteria, and then, MC and these phase I metabolites were absorbed into the blood, suffering further metabolism including N-, O-glucuronidation, and O-sulfation. Meanwhile, some phase II metabolisms including hydroxylation + sulfation (13.0%), glucuronide conjugation (7.6%), and hydroxylation + glucuronide conjugation (1.9%) were also detected in feces, which was quite different from those in the bile, suggesting that MC might undergo some first-pass   Journal of Analytical Methods in Chemistry metabolisms in the intestine after administration. However, this is contradicted to the relatively high bioavailability of MC in plasma. To our surprise, the amount of metabolites in plasma was quite different from those in urine and bile. e pharmacokinetics study of MC in plasma was needed in the future. A total of almost 88% phase II metabolites of MC were detected in bile. us, it can be inferred that MC also undergoes some first-pass metabolisms in the liver after absorption. e relative amount of metabolites in bile is similar to those in urine and different from those in feces. We tentatively speculated that in addition to hepatoenteric circulation, MC may encounter not only the phase I but also some phase II metabolisms happening in intestine with the metabolic enzymes and bacteria in the intestine, which catalyzes the drug's metabolic response and ultimately affects drug absorption.

Conclusions
In the present study, the UHPLC/Q-TOF MS method was used to explore the metabolites of MC in rat, after oral administration of MC, with the help of MetaboLynx software. A total of 14 metabolites were detected in rats and were tentatively identified based on the mass spectral fragmentation patterns, elution order, or retrieving literatures. N-or O-glucuronidation, O-sulfation, N-hydroxylation, dihydroxylation, and trihydroxylation were the major metabolisms of MC in vivo. Meanwhile, the relative content of each metabolite in each biological matrix was evaluated. Relatively high amount MC was displayed as prototype in plasma, proposing a high bioactivity of MC after oral administration. is study provides an experimental evidence for the metabolic process of MC and its family compounds in vivo and also provides a simple method for the rapid study of the metabolic process of bromine or chlorine containing compounds in vivo.

Data Availability
e data used to support the findings of this study are included within the manuscript and are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
Guozhe Zhang performed and designed the experiments, analyzed the data, and wrote the manuscript. Linxia Xiao helped with the data analysis. Liang Qi conceived and designed the experiments. All authors approved the final manuscript.