Biomarker Discovery for Cytochrome P450 1A2 Activity Assessment in Rats, Based on Metabolomics

Cytochrome P450 1A2 (CYP1A2) is one of the major CYP450 enzymes (CYPs) in the liver, and participates in the biotransformation of various xenobiotics and endogenous signaling molecules. The expression and activity of CYP1A2 show large individual differences, due to genetic and environmental factors. In order to discover non-invasive serum biomarkers associated with hepatic CYP1A2, mass spectrometry-based, untargeted metabolomics were first conducted, in order to dissect the metabolic differences in the serum and liver between control rats and β-naphthoflavone (an inducer of CYP1A2)-treated rats. Real-time reverse transcription polymerase chain reaction and pharmacokinetic analysis of phenacetin and paracetamol were performed, in order to determine the changes of mRNA levels and activity of CYP1A2 in these two groups, respectively. Branched-chain amino acids phenylalanine and tyrosine were ultimately focalized, as they were detected in both the serum and liver with the same trends. These findings were further confirmed by absolute quantification via a liquid chromatography–tandem mass spectrometry (LC-MS/MS)-based targeted metabolomics approach. Furthermore, the ratio of phenylalanine to tyrosine concentration was also found to be highly correlated with CYP1A2 activity and gene expression. This study demonstrates that metabolomics can be a potentially useful tool for biomarker discovery associated with CYPs. Our findings contribute to explaining interindividual variations in CYP1A2-mediated drug metabolism.


Quantification of phenacetin and paracetamol
For sample preparation, an aliquot of 80 μL plasma sample was pipetted into a 1.5 mL Eppendorf tube, followed by addition of 10 μL internal standard working solution (1 μg/mL pseudoephedrine hydrochloride, IS-1), 10 μL methanol, and 50 μL saturated NaHCO3. Analytes were extracted with 800 μL ethyl acetate by vortexing for 3 min, and then centrifuged at 8000 rpm for 10 min at 4 ℃. A total of 750 μL aliquot of the organic layer was evaporated to dryness under a gentle stream of nitrogen at 37 ℃, and the resulting residue was reconstituted in 80 μL methanolwater (20:80, v/v). After 10 min of centrifugation (14000 rpm, 4 ℃), 5 μL of the supernatant was injected into the LC-MS system for analysis.
The analytical conditions were as follows: column temperature, 35 ℃; autosampler temperature, 15 ℃; and flow rate, 0.3 mL/min. The gradient elution program was set as follows.
Mobile phase A (0.1% formic acid) and mobile phase B (methanol) were set at 0 min, 15% B; 5 min, 70% B; and 6.5 min, 70% B. The ESI source was set in positive ionization mode; selected ion monitoring (SIM) mode (m/z 180 for phenacetin, m/z 152 for paracetamol and m/z 166 for IS-1) was used. The detector voltage was 1.5 kV, the heat block temperature was 200 ℃; and the desolvation line temperature was 250 ℃; nitrogen was used as nebulizing gas, with a flow rate of 1.5 L/min. The calibration standard ranges used for phenacetin and paracetamol were 5-8000 μg/L and 10-8000 μg/L, respectively.

Untargeted metabolomics analysis
Serum and liver sample pretreatment, GC-MS and LC-MS analysis, data preprocessing, and metabolite identification were all based on our previous studies [1][2][3].

Sample pretreatment
For frozen liver samples, liver homogenates were prepared first. The same part of the left lobe of the liver from each rat was taken for tissue sample preparation. Ten volumes of pre-cold methanol were added to approximately 100 mg liver samples, followed by homogenization three times (5.5 m/s for 30 s) with 60 s intervals between each step. After two centrifugations (14,000 rpm, 4 ℃, 10 min), the supernatant was removed for metabolomic analysis. Serum samples were thawed at room temperature. For GC-MS analysis, 100 μL methanol was added to a 10 μL aliquot of serum or liver homogenate, and the mixture was thoroughly vortex-mixed for 15 min. After two centrifugations (14,000 rpm, 4 ℃, 10 min), 80 μL supernatant was transferred to a brown glass vial, and then oximated with 25 μL MOX (10 mg/mL in pyridine) at 1200 rpm for 90 min at 37 ℃. After vacuum drying (Labconco CentriVap, Kansas, MO, United States), the residue was silylated with 120 μL MSTFA:ethyl acetate (1:4, v/v) by incubating at 37 ℃ for 120 min, and then the supernatant was separated for GC-MS analysis.

GC-MS analysis
GC-MS analysis was performed on GCMS-QP2010 Ultra (Shimadzu Inc., Kyoto, Japan) equipped with a Rtx-5MS capillary column (30 m × 0.25 mm ID, 0.25 μm, Restek, USA). Helium was employed as the carrier gas at a flow rate of 1mL/min. The oven temperature was initially set at 70 ℃ for 3 min, followed by an increase to 320 ℃ (10 ℃/min), and maintained at 320 ℃ for 2 min. The temperature of the injector, transfer line, and ion source were set at 250, 250, and 200 ℃, respectively. The mass spectrometer was operated in electron impact mode with the energy of 70 eV. Data acquisition was performed in full san mode with a 45-600 mass to charge ratio (m/z) range.
A 1 μL sample was injected, with the split ratio of 50:1. GCMS solution version 2.7 (Shimadzu Inc., Kyoto, Japan) was used for spectra acquisition and data processing.

LC-MS analysis
LC-MS analysis was performed on an ultra-fast liquid chromatography (UFLC) system coupled with ion trap/time-of-flight hybrid mass spectrometry (IT/TOF-MS, Shimadzu Inc., Kyoto, Japan). Chromatographic separation was achieved by a Phenomenex Kinetex C18 column (100 × 2.1 mm, 2.6 μm, Phenomenex, United States). The column temperature was set at 40 ℃. The gradient elution with 0.4 mL/min flow rate (phase A: 0.1% formic acid, phase B: acetonitrile) was carried out from 95% A to 5% A within 20 min and maintained at 5% A for 3 min. For mass analysis, ESI was set in both positive and negative ion mode with a 100-1000 m/z san range. The TOF analyzer detector voltage was 1.8 kV, and the interface voltage was set at 4.5 kV and -3.5 kV for positive and negative mode, respectively. The curved desorption line and heat block temperature were both set at 200 ℃. Nitrogen was used as the nebulizing gas, with a flow rate of 1.5 L/min. A 5 μL sample was injected for analysis. LCMS solution version 3.0 (Shimadzu Inc., Kyoto, Japan) was used for spectra acquisition and data processing.

Data preprocessing
Each chromatogram obtained from GC-MS and LC-MS analysis was processed for peak deconvolution and alignment using Profiling Solution version 1.1 (Shimadzu, Kyoto, Japan), followed by background-peak-filtering, 80% rule, limitation of QCs, missing data imputation, and normalization. The details of each step were as follows [3]: (1) Background-peak filtering: each chromatogram was checked against the solvent blanks (inserted randomly in the analytical batch) to exclude possible sources of contamination, such as instrumental contamination or reagent impurities.
(2) 80% rule: retained the variables which were detectable in more than 80% samples in at least one group to minimize the effect of the missing values.
(3) QC sample limitation: removed the variables with RSD values higher than 30% in QC samples.
(4) Missing data imputation and normalization: replaced the missing values with a half of the minimum value found in the dataset. After the total area normalization for each sample, a resulting matrix was obtained and then prepared for further differential features screening and metabolite identification.

Metabolites identification
For GC-MS analysis, metabolites were preliminarily identified by a comparison of mass spectra and intensities with those available in National Institute of Standards and Technology (NIST 11) library. To minimize false discovery rates, only those peaks with similarity more than 75% were assigned for compound names and considered reliable. Some of the metabolites were further confirmed by standard compounds available in our lab.        CYP1A2 mRNA level was calculated using the 2 -∆Ct method and normalized by β-actin expression.

Tables
Data are presented as mean ± SD, and n = 8 for each group. Ct: cycle threshold; △Ct = Ct of CYP1a2 -Ct of β-actin; C: control group; BNF: β-naphthoflavone treatment group.   a Correlation coefficients of Spearman correlation analysis between differential metabolites and metabolic ratio. b Correlation coefficients of Spearman correlation analysis between differential metabolites and mRNA level of CYP1A2. c Change trends of differential metabolites based on area normalization data in untargeted metabolomics. ↓decreasing change trend after β-naphthoflavone administration. ↑increasing change trend after β-naphthoflavone administration. The value of the metabolic ratio reflected the activity of CYP1A2, and a higher value represented a higher activity.
CYP1A2 mRNA expression was calculated using the 2 -∆Ct method. Ct: cycle threshold; △Ct = Ct of CYP1a2 -Ct of β-actin. Data are presented as mean ± SD; n = 8 for each group. C: control group; BNF: β-naphthoflavone treatment group. Data are presented as mean ± SD; n = 8 for each group. C: control group; BNF: β-naphthoflavone treatment group.