Determination of metabolites of phloretin in rats using UHPLC-LTQ-Orbitrap mass spectrometry

Purpose: To study the metabolites of phloretin in vivo using ultra-high performance liquid chromatography linear ion trap-Orbitrap mass spectrometry (UHPLC-LTQ-Orbitrap). Methods: After administration of phloretin (50 mg/kg; oral route) to six rats, blood samples were taken from each animal. Each sample was then subjected to solid-phase extraction to prepare it for chromatographic/spectroscopic analysis. Finally, each sample was analyzed using UHPLC-LTQOrbitrap with a negative-mode electrospray ionization source. Results: Based on mass measurements, chromatographic retention times, and MS2 fragmentation ions, we detected and identified phloretin and 16 metabolites of the drug in vivo in rats. Metabolic reactions of phloretin included glucosylation and glucuronide conjugation, diglucuronide conjugation, glucosylation and sulfate conjugation, sulfate conjugation, glucuronide conjugation, and glucosylation and hydroxylation. Conclusion: The findings provide a better understanding of phloretin metabolism and metabolites, and new information about their effective forms, pharmacological actions, metabolic fate, and toxic actions in vivo.

Drug metabolites are typically characterized structurally in vivo and in vitro by liquid chromatography/mass spectrometry methods [11,12]. Due to the shorter times required and higher-yield separation and accurate resolution capacities, ultra-high-performance liquid chromatography (UHPLC) coupled with highresolution mass spectrometry (e.g., the UHPLClinear ion trap (LTQ)-Orbitrap) significantly contributes to the accurate and efficient characterization of drug metabolites [13,14].

EXPERIMENTAL Chemicals and reagents
Grace Pure TM SPE C18 phase extraction cartridges (200 mg/3 mL, 59 μm, 70 Å) were obtained from Grace Davison Discovery Science TM (Deerfield, IL, USA). HPLC-grade acetonitrile was obtained from Fisher (Fair Lawn, NJ, USA). Ultra-pure water was generated using a Milli-Q water purification system from Millipore (Billerica, MA, USA). All other reagents used in this study were analytical grade and commercially available.
The phloretin was purchased from Nanjing Spring and Autumn Biological Engineerin gCo. Ltd. Its structure was determined from UV, MS, 1 H-NMR, and 13 C-NMR results and from a comparison of those results with previously published data. HPLC analysis revealed that the phloretin purity was > 98 %. The structure is shown in Figure 1.

Animals and drug administration
Sprague-Dawley rats (six male rats; body weight range, 200 -250 g; Beijing Wei tong Li hua Experimental Animals Company, Beijing, China) were kept in controlled environmental conditions (relative humidity, 70 ± 5 %; ambient temperature, 24 ± 2 °C) and were supplied with food and water ad libitum during the 1 week of acclimation before the start of the experiment. The rats were then randomly assigned to one of two groups. Group I (n = 3) was the drug group; these rats were given phloretin. Group II (n = 3), was the control group (i.e., no phloretin given). Before the experiment commenced, all rats were fasted for 12 h but had free access to water.

Phloretin
was suspended in 0.5 % carboxymethylcellulose sodium (CMC-Na) aqueous solution. Two hours after phloretin was given via the oral route (50 mg/kg body weight), the group 1 rats were anesthetized using ether and then euthanized using decapitation. Each group 2 rats was given 0.5 % CMC-Na aqueous solution 2 mL via the oral route and was anesthetized and euthanized using the same procedures as the group 1 rats. A blood sample was withdrawn from each rat into heparinized centrifuge tubes and centrifuged (4000 rpm, 10 min) to obtain the plasma, which was stored at -20 °C until pretreatment and analyses.
The experiment was designed and performed in accordance with the guidelines established by Animal Experiments of Hunan University of Medicine. The study protocol was approved by the Animal Biomedical Ethical Committee of Hunan University of Medicine (approval no. kjdw-20171104-05) [15].

Sample preparation
A solid-phase extraction (SPE) method was used to pretreat each plasma sample. Each SPE cartridge was pretreated using successive rinses with water, methanol, and water (5 mL each). For each sample, 1 mL plasma was added to the SPE cartridge and flowed through by gravity. Water and then methanol (5 mL each) were then used to rinse the SPE cartridge. After collection, the methanol eluent was dried by evaporation at room temperature under N 2 gas. Acetonitrile/water (100 µL; 10:90, v/v) was used to re-dissolve the residue and the sample was then centrifuged (12,000 rpm, 4 °C, 30 min). A 2-µL supernatant sample was analyzed by injection into the UHPLC-LTQ-Orbitrap MS.
The spectral analysis used high-resolution electrospray ionization (ESI)-MS and MS/MS performed on the LTQ-Orbitrap mass spectrometer connected to the UHPLC instrument via the ESI interface. The negative ion mode was used during sample analysis. The tune method settings consisted of a nitrogen sheath gas flow rate of 30 arb, a nitrogen auxiliary gas flow rate of 5 arb, a spray voltage of 4.0 kV, a capillary temperature of 350 °C, a capillary voltage of 25 V, and a tube lens voltage of 110 V. Calibration was performed according to the manufacturer's guidelines to ensure an accurate mass analysis. The centroid mass spectra were acquired in the mass range of m/z 100-1000.
The Orbitrap mass analyzer resolution was set at 30,000 during the full-scan experiment. The total analytical time was minimized during the datadependent MS/MS scanning to avoid generation of fragmentation spectra of the targeted ions. The collision-induced dissociation collision energy was adjusted to 35% of maximum, and the precursor ion isolation width was m/z 2.0 Da. To prevent repetition, dynamic exclusion was enabled; the repeat count was set at 5, the dynamic repeat time was set at 30 sec, and the dynamic exclusion duration was set at 60 sec.

Peak selection and data processing
Data acquisition and processing were performed using a Thermo Xcaliber 2.1 workstation. To maximize fragment ion detection, the peaks with intensities >10,000 were selected for identification. Chemical formulas were calculated for the parent ions of the selected peaks based on accurate mass values using a molecular formula predictor. The parameters were C

Phloretin fragmentation pathway
The phloretin MS n fragmentation pattern was examined to assist with the characterization of metabolite structure. In negative ion mode, the parent ion had a deprotonated ion

Detection and determination of metabolite structure
In addition to phloretin (the parent drug), 17 metabolites were found and underwent chromatographic and mass spectrometric analyses (Table 1) after comparing the highresolution extracted ion chromatography results for the samples from the rats who were given phloretin with the results for the plasma samples from the control rats ( Figure 3).

Metabolites 7, 1, and 3
Metabolite 7 elution occurred at 13.07 min with the quasi-molecular m/z 435.12857 ions (0.06 ppm, C 21 H 23 O 10 ). It was 162 Da greater than phloretin, which suggested that metabolite 7 was a glucosylation product of the prototype drug. An ion present at m/z 273.07520 in the MS 2 spectra was the result of a 162 Da loss from the precursor ion at m/z 435. This result also suggested that it was a glucosylation conjugation product of phloretin. Based on these results, metabolite 7 was categorized as a glucosylation conjugation product of phloretin.

Metabolite 1 included a deprotonated molecular ion [M-H]
at m/z 611.16046 (-0.33 ppm, C 27 H 31 O 16 ). It was 176.03 Da greater than metabolite 7. This result suggested that metabolite 1 was a glucuronide conjugation product of metabolite 7. Therefore, it was categorized as a glucuronide and glucosylation conjugation product of phloretin.
Metabolite 3 elution occurred at 9.37 min with the quasi-molecular m/z 515.08502 ions (-0.71 ppm, C 21 H 23 O 13 S). It was 79.96 Da greater than metabolite 7. This result suggested that metabolite 3 was a sulfate conjugation product of metabolite 7. This conclusion was supported by the presence of characteristic ions at m/z 273 and m/z 167. Therefore, metabolite 3 was categorized as a sulfate and glucosylation conjugation product of phloretin.

Metabolite 12
Metabolite 12 elution was at 17.10 min with the quasi-molecular m/z 289.07062 ions (-0.15 ppm, C 15 H 13 O 6 ). It was 16 Da greater than phloretin. Therefore, metabolite 12 was identified as a mono-hydroxylated product of the prototype drug.

DISCUSSION
A UHPLC-LTQ-Orbitrap mass spectrometry method was used to analyze the metabolites of phloretin in vivo. To improve the separation results, mobile phase (e.g., methanol-water, acetonitrile-water, and acetonitrile-methanolwater) system optimization was performed. The results indicated that a mobile phase consisting of an acetonitrile-methanol (3:1/v:v) and formic acid (0.1%) aqueous solution improved the chromatographic peak resolution of the phloretin, its metabolites, and the endogenous components.
In this study, phloretin (a dihydrochalcone) was detected using ESI in negative ion mode [16]. ESI provided greater response intensities and reduced spectral interference. Some work on the identification of metabolites of phloretin has been performed [17][18]. For example, one study did find sulfate and phloretin conjugates in rat plasma [18]. However, more studies to identify phloretin metabolites should be performed.

CONCLUSION
Metabolites can possess pharmacological and toxic activities. Therefore, the early stages of drug investigations include drug metabolite characterization. In this study, phloretin and its metabolites were found and categorized in plasma from rats given phloretin per os. These results contribute to our understanding of phloretin's effective forms, metabolic fates, and pharmacological and toxic actions in vivo.