Abstract
Arachidonic acid (AA) is a polyunsaturated essential fatty acid and a precursor for eicosanoids. It is metabolized by cyclooxygenases, lipoxygenases, and cytochrome P450 (P450) enzymes, which convert AA into hydroxyeicosatetraenoic acids (HETEs) and epoxyeicosatrienoic acids (EETs), chiral eicosanoids with distinct biological activities. Although racemic HETEs and EETs have been studied in cardiovascular diseases, the enantiospecific roles of their enantiomers and the enantioselectivity of P450 enzymes remain largely unexplored. This study aimed to investigate the enantioselective metabolism of AA by human recombinant P450 enzymes, focusing on the formation of R/S-HETEs and (R, S)/(S,R)-EETs. Metabolites were analyzed using liquid chromatography electrospray ionization mass spectrometry. CYP1A2 exhibited the highest activity in forming R-midchain HETEs, followed by CYP3A4. CYP2C19 was the most active enzyme in producing R-subterminal HETEs, with CYP1A2 and CYP1A1, CYP4F3B, and CYP2E1 ranking second. Similarly, CYP2C19 showed the highest activity in generating S-midchain and S-subterminal HETEs, with CYP3A4, CYP2C8, CYP1A1, and CYP1A2 contributing to varying degrees. For EETs, CYP2C19 and CYP1A2 primarily catalyzed the formation of both (R, S)/(S, R)-EETs. These findings emphasize the significant roles of CYP2C19 and CYP1A2 in the regio- and stereoselective metabolism of HETEs and EETs, highlighting their contributions to lipid signaling and potential physiological implications.
Significant Statement
This work highlights the importance of profiling P450 with respect to their enantioselectivity in arachidonic acid metabolism. The findings indicate that major P450 differ in the magnitude of their hydroxyeicosatetraenoic acid and epoxyeicosatrienoic acid formation rates, which is a significant for studying diseases that is known to be influenced by alterations in these pathways. Altered enantioselectivity could have implications in diseases such as hypertension, cancer, inflammation, and cardiovascular disorders.
Key words: Epoxyeicosatrienoic acids, Hydroxyeicosatetraenoic acids, CYP2C19, CYP1A2, Enantioselectivity, LC-MS/MS
1. Introduction
Arachidonic acid (AA) is a polyunsaturated essential fatty acid and the precursor of variety of eicosanoids. The process of biosynthesis is initiated following the activation of specific cell surface that stimulates phospholipase A2 resulting in the release of AA from the cell membranes.1 The released AA is a substrate for 3 main enzymes namely cyclooxygenase, lipoxygenases, and cytochrome P450s (P450).1 Cyclooxygenases involve in forming prostaglandins, prostacyclin, and the thromboxane, whereas lipoxygenases contribute to the formation of leukotrienes.2 P450s are unique group of enzymes characterized by having hemothiolate that plays a role in the biotransformation processes.3 P450 enzymes are well known as the main pathway for clearing exogenous substances, such as drugs, in organisms. P450s are affected by genetic polymorphisms altering drug metabolism, leading to variation in response in both efficacy and safety of drugs.4 The role of P450 in endogenous fatty acid metabolism is attracting more attention due to the fact that the end products exert multifaceted biologic effects. To date, it is known that AA is metabolized by different P450 enzymes to epoxy and hydroxy AA metabolites. The epoxy AA metabolites are 5,6-, 8,9-, 11,12-, and 14,15-epoxyeicosatrienoic acids (EETs)5,6 and hydroxy metabolites are 5-, 8-, 9-, 11-, 12-, 15-, 16-, 17-, 18-, 19-, and 20-hydroxyeicosatetraenoic acids (HETEs).3,7 With the except of 20-HETE, epoxy and hydroxy AA metabolites are chiral eicosanoids.5,6 HETEs are further classified into 3 categories: midchain and subterminal, and terminal HETEs; the midchain HETEs are 5-, 8-, 9-, 11-, 12-, and 15-HETEs whereas subterminal includes 16-, 17-, 18-, and 19-HETEs.3,7 There is more focus on the midchain HETEs due to the recent studies demonstrating their role in the pathogenesis of cardiac hypertrophy, inflammation, diabetes, diabetic-induced cardiomyopathy, and liver diseases.2,8, 9, 10, 11, 12 Subterminal HETEs have been reported to exhibit cytoprotective roles. We reported previously that subterminal HETEs modulated by CYP1B1 protect against cardiac hypertrophy.13, 14, 15 However, abundant literature reports the renal and cardioprotective effect of EETs16, 17, 18 indicating the importance maintaining the EETs in biological system. It is worth noting that most studies reported levels of racemate HETEs and EETs rather than individual isomers in cardiovascular diseases.11,16, 17, 18, 19, 20 This is often due to limitations in analytical techniques for separating and quantifying the individual isomers of HETEs and EETs in biological samples leading to a scarcity of reports on the specific roles of each stereoisomer of midchain and subterminal HETEs, as well as EETs in medical conditions. Previous liquid chromatography-tandem mass spectrometry (LC-MS/MS) methods for HETEs were largely nonstereospecific, and no method could simultaneously analyze both midchain and subterminal HETEs.21, 22, 23, 24, 25 To address this, we developed and validated a method for the quantitative analysis of all HETE enantiomers and regioisomers, which we demonstrated using human liver microsome (HLM) incubations with AA.7 Existing chiral EET assays typically rely on derivatization, which can alter stereochemistry and extend run times26,27; to overcome this, we developed an LC-MS/MS method for underivatized EET enantiomers using gradient elution, providing a practical and reliable method that preserves native stereochemistry. It has been reported that enantiomers of HETEs have differential activities on enzymes13,14 and hence contribute differently to biological activities. Recently, 16-HETE and 17-HETE were reported to enantioselectively induce cellular hypertrophy.13,28 In addition, stereoselective effect of EETs was reported. For example, the 5S(6R)-EET induced more than 2.5-fold on depolarizing transepithelial voltage than 5R(6S)-EET.29 Thus, profiling P450s for their metabolic activity to form (R/S)-HETEs and (R, S)/(S, R)-EETs are becoming increasingly critical area of research. As a result, the aim of this study was to examine the human recombinant P450 enzymes in their role of AA metabolism, particularly focusing on their stereoisomerism formation using validated chiral LC-electrospray ionization-MS/MS assays.
2. Materials and methods
2.1. Materials
Human CYP1A2-, CYP2B6-, CYP2C8-, CYP2C19-, CYP2J2, CYP2E1, and CYP3A4- Baculosomes PLUS containing insect cell microsomes supplemented with NADPH-P450 reductase and human cytochrome b5 were developed by Invitrogen (Thermo Fisher Scientific). CYP1A1-CYP1B1-, CYP4F12-, CYP4F3A-, and CYP4F3B-Supersomes containing insect cell microsomes supplemented with NADPH-P450 reductase were obtained from Gentest (Discovery Life Sciences). The standard hydroxy-metabolites (5-, 8-, 9-, 11-, 12-, 15-, 16-, 17-, 18-, and 19-HETE, as well as internal standard 20-HETEd6) were purchased from Cayman Chemical. The racemic standards of (5,6-EET, 8,9-EET, 11,12-EET, 14,15-EET) and the deuterated internal standards (11,12)-EET-d11 were also obtained commercially from Cayman Chemical. High-performance liquid chromatography (HPLC)-grade acetonitrile, methanol, isopropyl alcohol, and NADPH were purchased from Millipore-Sigma. The ethyl acetate and glacial acetic acid were from Sigma Aldrich, whereas the HPLC water was obtained from Fisher Scientific Co.
The standard (±) 5, 8, 9, 11, 12, 15, 16, 17, 18, and 20- HETE and 20-HETE-D6 were purchased from Cayman Chemical.
2.2. AA-P450 coincubation
The incubation buffer was 100 mM potassium phosphate (pH 7.4) supplemented with 3 mM magnesium chloride hexahydrate. Conditions were optimized to achieve linearity with time and protein concentration. Each sample contained P450 enzyme with NADPH-P450 reductase, cytochrome b5, and AA (75 μM). The detailed incubation conditions are depicted in Table 1. For the human recombinant enzymes that are not supplemented with cytochrome b5 (CYP1B1, CYP1A1, and CYP1A2), cytochrome b5 was added to the incubation to achieve a 1:1 molar ratio with the P450 enzyme. Background controls were also prepared containing all components of reaction mixture but lacking NADPH. Incubations were carried out in a shaking water bath at 37 °C (90 rpm). The mixture was pre-equilibrated for 5 minutes followed by initiation of the reaction with 2 mM NADPH. Finally, the reaction was terminated with 600 μL of ice-cold acetonitrile containing the internal standards (20-HETE-d6 and 11-12-d11). The samples were extracted with ethyl acetate solvent. The supernatant was removed and dried under vacuum speed-vac and the residue of were reconstituted with 100 μL acetonitrile for HETEs whereas EETs were reconstituted with 100 μL of acetonitrile and methanol (70:30%).
Table 1.
Concentration of P450 and time for incubations
| Concentration for HETEs pmol/mL |
Concentration for EETs pmol/mL |
Time min |
|
|---|---|---|---|
| CYP2C19 | 50 | 25 | 15 |
| CYP3A4 | 100 | 50 | 20 |
| CYP2B6 | 100 | 50 | 20 |
| CYP1A2 | 100 | 50 | 15 |
| CYP1A1 | 100 | 50 | 15 |
| CYP1B1 | 100 | 50 | 20 |
| CYP2E1 | 100 | 50 | 20 |
| CYP2J2 | 100 | 50 | 20 |
| CYP2C8 | 100 | 50 | 20 |
| CYP4F12 | 50 | 25 | 20 |
| CYP4F3A | 50 | 25 | 20 |
| CYP4F3B | 50 | 25 | 20 |
2.3. LC-electrospray ionization-MS/MS operating condition
The HETE metabolites were analyzed using previously validated chiral assay,7 whereas the EETs were quantified separately using another validated method.30 Briefly, LC was performed using a mobile phase with gradient elution, consisting of an organic phase of acetonitrile, methanol, and isopropyl alcohol (88:6:6, v/v, with 0.1% acetic acid) and an aqueous phase of water containing 0.1% acetic acid for HETEs. For EETs, mobile phase A consisted of HPLC-grade water with 0.005% diethylamine and 0.01% formic acid, and mobile phase B was a mixture of acetonitrile, methanol, and isopropyl alcohol (82:10:8 v/v) with 0.005% diethylamine and 0.01% formic acid. Both assays used a chiral stationary phase REFLEC C-AMYLOSE A column (5 μm, 250 × 4.6 mm) (Regis Technologies Inc) with a KrudKatcher Ultra HPLC inline filter and a flow rate of 0.5 mL/min. The MS parameters were as follows: interface voltage, 3 kV; current, 0.8 μA; interface temperature, 300 °C; nebulizing gas flow, 1.6 L/min; drying gas flow, 10 L/min; heating gas flow, 10 L/min; desolvation temperature, 250 °C; and heat block temperature, 400 °C. Other MS/MS conditions such as precursor and product ions for HETEs and EETs are reported elsewhere.7,30 The concentration of an unknown metabolites formed by each P450 enzyme was determined using calibration curve method. The calibration curves were constructed using standards at concentrations ranging from 0.01 to 0.6 μg/mL per enantiomer. The separation of enantiomers is shown in Supplemental Figs 1 and 2.
2.4. Data analysis
The formation rates of the metabolites were determined by dividing the concentration of the metabolite to the corresponding P450 amount and time of incubation. The mean ± SD was used for description of the data. Ranking according to formation rates were determined by ordinary one-way ANOVA followed by Tukey’s multiple comparisons test using GraphPad PRISM version 10.3.1 (GraphPad).
3. Results
In this study, we studied 12 human recombinants P450 enzymes for their AA-stereoisomerism metabolizing profile. We observed that most of tested P450 enzymes mediated AA metabolism, but with variable stereoselectivity and regioselectivity.
3.1. Total formation rates of (R/S)-midchain HETEs by human recombinant P450
Among the tested P450, CYP1A2 yielded the highest activity of (R)-midchain formation, followed by CYP3A4, CYP2C19, and CYP2C8 ranked the third (Fig. 1A). The total formation rates of (R)-midchain HETEs by CYP1A2, CYP3A4, CYP2C19, and CYP2C8 were 1477.0 ± 75.5, 544.1 ± 51.6, 482.1 ± 30.5, and 401.8 ± 8.7 fmol/pmol CYP per min, respectively. Among the R-midchain HETEs, 12R-HETE was the most abundant HETE formed by CYP1A2 (1287.5 ± 62 fmol/pmol CYP per min), followed by 11R-HETE by CYP2C8 (164.4 ± 3.2 fmol/pmol CYP per min), and 9R-HETE by CYP3A4 (142.1 ± 14.6 fmol/pmol CYP per min) (Fig. 2A; Table 2). However, CYP2C19 showed the highest activity in forming (S)-midchain HETEs, followed by CYP3A4. CYP2C8 and CYP1B1 ranked third in activity. The total formation rates of (S)-midchain HETEs by CYP2C19, CYP3A4, CYP2C8, and CYP1B1 were 1170.7 ± 139.1, 756.3 ± 65.8, 378.4 ±18.7, and 321.9 ±19.7 fmol/pmol CYP per min, respectively (Fig. 1B). Among the S-HETEs formed, 5S-HETE was the most formed midchain S-enantiomer by CYP2C19 (631.0 ± 83.9 fmol/pmol CYP per min) followed by 15S-HETE formed by CYP2C8 (239.8 ± 15.6 fmol/pmol CYP per min), 9S-HETE by CYP3A4 (198.7 ± 16.0 fmol/pmol CYP per min), and 8S-HETE by CYP1B1 (155.3 ± 18.7 fmol/pmol CYP per min), (Fig. 2B; Table 2). The majority of P450s that contributed to overall formation rates of midchain exhibited enantioselectivity profiles favoring S-midchain HETEs over R-midchain HETE (Table 3).
Fig. 1.
Metabolic profile of human recombinant P450 enzymes in AA metabolism: formation of R/S-midchain and R/S-subterminal HETEs. The incubation conditions and the method of analysis are explained under Materials and methods. Results are presented as mean and SD, based on at least 3 individual experiments. We used one-way ANOVA followed by Tukey’s multiple comparisons test to compare the formation rates across the most active P450 enzymes; P < .05 indicated significantly different activities in the order. The letters in the figure represent statistical rankings, enzymes with different letters are significantly different from each other, and sharing the same letter indicates no statistical difference.
Fig. 2.
Heatmap of metabolic profile of human recombinant P450 enzymes showing AA metabolism to R/S-HETEs. The incubation conditions and the method of analysis are explained under Materials and methods. The results are shown based on the color scale at the bottom of the image.
Table 2.
Formation rates of (R/S)-midchain HETEs by human recombinant P450
The data are presented as mean ± SD, n = 3, unit, fmol/pmol CYP per min.
| 5 (R)-HETE |
8 (R)-HETE |
9 (R)-HETE |
11 (R)-HETE |
12 (R)-HETE |
15 (R)-HETE |
|
|---|---|---|---|---|---|---|
| Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | |
| CYP1A1 | BQL | BQL | BQL | BQL | 46.3 ± 6.2 | BQL |
| CYP1A2 | 97.4 ± 12.4 | 9.5 ± 0.9 | 19.5 ± 1.1 | 63.1 ± 3.6 | 1287.5± 62.0 | BQL |
| CYP1B1 | 68.2 ± 2.6 | 82.8 ± 11.8 | 41.2 ± 0.9 | 11.4 ± 0.5 | 60.8 ± 0.8 | BQL |
| CYP2B6 | BQL | 16.9 ± 0.3 | 51.6 ± 0.3 | 27.4 ± 1.3 | 23.1± 0.7 | BQL |
| CYP2C8 | BQL | 31.0 ± 0.9 | 27.0 ± 0.7 | 164.4 ± 3.2 | 116.3 ± 3.3 | 63.1± 1.0 |
| CYP2C19 | BQL | 105.7 ± 8.5 | 134.7 ± 9.2 | 105.8 ± 8.5 | 135.8 ± 9.9 | BQL |
| CYP2E1 | BQL | 33.8 ± 1.6 | 54.1 ± 3.2 | 36.4 ± 0.5 | 34.2 ± 3.2 | BQL |
| CYP2J2 | BQL | 19.3 ± 0.4 | 27.4 ± 0.7 | 16.6 ± 0.2 | 25.0 ± 0.1 | BQL |
| CYP3A4 | 84.0 ± 12.9 | 76.7 ± 5.9 | 142.1 ± 14.6 | 92.1 ± 4.8 | 98.2 ± 7.8 | 51.0 ± 6.4 |
| CYP4F3A | BQL | 64.2 ± 0.9 | BQL | 73.6 ± 2.4 | BQL | BQL |
| CYP4F3B | BQL | 17.4 ± 0.7 | BQL | BQL | BQL | BQL |
| CYP4F12 | BQL | BQL | BQL | BQL | BQL | BQL |
| 5 (S)-HETE |
8 (S)-HETE |
9 (S)-HETE |
11 (S)-HETE |
12 (S)-HETE |
15 (S)-HETE |
|
|---|---|---|---|---|---|---|
| Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | |
| CYP1A1 | BQL | BQL | BQL | 92.4± 12.6 | 87.3 ± 8.3 | 99.1± 26.3 |
| CYP1A2 | 36.0 ± 9.4 | 56.7± 3.9 | 39.7 ± 1.8 | 48.6 ± 3.8 | BQL | 32.3 ± 4.3 |
| CYP1B1 | 32.8 ± 0.9 | 155.3 ± 18.7 | 38.7 ± 1.0 | 75.6 ± 3.1 | BQL | 60.8 ± 0.8 |
| CYP2B6 | BQL | 11.6 ± 0.3 | 44.5 ± 0.3 | 32.2 ± 1.3 | 20.4 ± 0.7 | 8.3 ± 2.0 |
| CYP2C8 | BQL | 16.6 ± 0.9 | 22.7 ± 1.2 | 65.5 ± 2.7 | 33.7 ± 0.7 | 239.8 ± 15.6 |
| CYP2C19 | 631.0 ± 83.9 | 70.8 ± 9.7 | 155.9 ± 9.2 | 31.9 ± 6.7 | 133.0 ± 26.8 | 148.0 ± 9.3 |
| CYP2E1 | BQL | 31.0 ± 2.3 | 37.2 ± 4.2 | 30.6 ± 0.8 | 29.9 ± 3.6 | 31.7 ± 5.4 |
| CYP2J2 | BQL | 19.1 ± 0.3 | 22.8 ± 0.8 | 26.2 ± 0.3 | 18.8 ± 0.5 | 4.3 ± 0.7 |
| CYP3A4 | 141.3 ± 17.4 | 112.7± 9.9 | 198.7 ± 16.0 | 153.3 ± 11.4 | 32.2 ± 4.9 | 118.2 ± 6.6 |
| CYP4F3A | 126.9 ± 30.0 | 80.7 ± 3.7 | BQL | BQL | BQL | BQL |
| CYP4F3B | BQL | BQL | BQL | 9.8 ± 3.0 | BQL | BQL |
| CYP4F12 | BQL | BQL | BQL | BQL | BQL | BQL |
BQL, below quantifiable level.
Table 3.
Enantioselectivity of human P450 in midchain and subterminal HETEs
| P450 | Enantioselective (Midchain HETEs) | Midchain HETEs | Enantioselective (Subterminal HETEs) | Subterminal HETEs |
|---|---|---|---|---|
| CYP1A2 | Yes | R>S | No | S=R |
| CYP3A4 | Yes | S>R | ||
| CYP2C19 | Yes | S>R | Yes | S>R |
| CYP2C8 | No | S=R | Yes | S>R |
| CYP1B1 | Yes | S>R | ||
| CYP2E1 | No | S=R | Yes | S>R |
| CYP1A1 | Yes | S>R | Yes | R>S |
| CYP4F3A | Yes | R>S | ||
| CYP4F3B | Yes | R>S |
3.2. Total formation rates of (R/S)-subterminal HETEs by human recombinant P450s
Among the tested P450s, CYP2C19 exhibited the highest of R-subterminal HETEs formation, followed by CYP1A2, CYP1A1, CYP2E1, and CYP4F3B, with CYP4F3A ranking third (Fig. 1C). The formation rates of R-subterminal HETEs mediated by CYP2C19, CYP1A2, CYP1A1, CYP4F3B, CYP2E1, and CYP4F3A were 2963.2 ± 93.9, 760.5 ± 55.4, 698.8 ± 47.9, 698.8±15.7, 691.2 ± 22.5, and 427.2 ± 10.6 fmol/pmol CYP per min, respectively. Moreover, 19R-HETE was the predominant R-subterminal HETE formed by many P450s. CYP2C19 produced the highest amount of 19R-HETE (2790.5 ± 90.1 fmol/pmol CYP per min), followed by CYP4F3B (623.2 ± 6.3 fmol/pmol CYP per min) and CYP1A2 (482.7 ± 54.0 fmol/pmol CYP per min). In contrast, 18R-HETE was the lowest R-subterminal HETEs by CYP4F12 (4.4 ± 0.2 fmol/pmol CYP per min), followed by 17R-HETE formed by CYP3A4 (5.2 ± 0.8 fmol/pmol CYP per min) and 18R-HETE by CYP1B1 (8.1± 0.4 fmol/pmol CYP per min) (Fig. 2A; Table 4).
Table 4.
Formation rates of (R/S)-subterminal HETEs by human recombinant P450
The data are presented as mean ± SD, n = 3, unit, fmol/pmol CYP per min.
| 16 (R)-HETE |
17 (R)-HETE |
18 (R)-HETE |
19 (R)-HETE |
|
|---|---|---|---|---|
| Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | |
| CYP1A1 | 182.7 ± 11.2 | 222.2 ± 13.9 | 293.9 ± 22.9 | BQL |
| CYP1A2 | 65.8 ± 2.3 | 88.9 ± 1.3 | 123.1 ± 3.8 | 482.7 ± 54.0 |
| CYP1B1 | 50.5 ± 1.2 | 21.4 ± 0.4 | 8.1 ± 0.4 | BQL |
| CYP2B6 | 14.4 ± 0.2 | 10.7 ± 0.3 | 11.3 ±0.3 | 27.5 ± 0.9 |
| CYP2C8 | 65.1 ± 2.2 | 52.1 ± 1.3 | 30.3 ± 2.0 | 119.5 ± 4.7 |
| CYP2C19 | 54.8 ± 0.7 | 117.9 ± 4.6 | BQL | 2790.5 ± 90.1 |
| CYP2E1 | 23.5 ± 0.3 | 36.8 ± 0.6 | 158.9 ± 6.4 | 472.0 ± 15.5 |
| CYP2J2 | 19.8 ± 0.3 | 25.7 ± 0.8 | 25.3 ± 1.1 | 156.6 ± 12.9 |
| CYP3A4 | 50.0 ± 2.3 | 5.2 ± 0.8 | 31.2 ± 2.0 | BQL |
| CYP4F3A | 43.9± 0.6 | BQL | BQL | 331.2 ± 10.1 |
| CYP4F3B | 49.8 ±10.4 | 11.7 ± 0.1 | 14.1± 0.4 | 623.2 ± 6.3 |
| CYP4F12 | BQL | 11.5 ± 0.1 | 4.4 ± 0.2 | BQL |
| 16 (S)-HETE |
17 (S)-HETE |
18 (S)-HETE |
19 (S)-HETE |
|
|---|---|---|---|---|
| Mean ± SD | Mean ± SD | Mean ± SD | Mean ± SD | |
| CYP1A1 | 78.7 ± 3.3 | 55.6 ± 1.2 | 257.3 ± 11.5 | 1507.6 ± 53.1 |
| CYP1A2 | 8.3 ± 1.4 | 18.3 ± 0.5 | 76.2 ± 1.7 | 946.3 ± 245.8 |
| CYP1B1 | 88.6 ± 2.5 | 14.3 ± 0.2 | 12.7 ± 0.2 | 108.5 ± 0.5 |
| CYP2B6 | 14.2 ± 0.2 | 10.8 ± 0.3 | 10.6 ± 0.2 | 31.6 ± 1.2 |
| CYP2C8 | 43.8 ± 1.1 | 79.0 ± 2.5 | 31.2 ± 0.8 | 143.6 ± 4.1 |
| CYP2C19 | 27.0 ± 1.2 | BQL | 134.5 ± 27.0 | 5199.5 ± 219.9 |
| CYP2E1 | 26.5 ± 0.4 | 35.0 ± 0.5 | BQL | 907.5 ± 13.4 |
| CYP2J2 | 17.3 ± 0.2 | 26.0 ± 0.7 | 32.8 ± 0.8 | 115.9 ± 8.2 |
| CYP3A4 | 46.9 ± 2.9 | 3.1 ± 0.4 | BQL | 156.4 ± 3.7 |
| CYP4F3A | BQL | BQL | 57.2 ± 0.1 | BQL |
| CYP4F3B | 53.9 ± 1.1 | 17.4 ± 0.2 | 11.8 ± 0.6 | 545.8 ± 13.2 |
| CYP4F12 | BQL | 19.1 ± 0.0 | 53.2 ± 2.4 | 82.9 ± 2.9 |
BQL, below quantifiable level.
Similarly, among the tested P450 enzymes, CYP2C19 demonstrated the highest activity for S-subterminal HETE formation, followed by CYP1A1, whereas CYP1A2 and CYP2E1 ranked third (Fig. 1D). The formation rates of S-subterminal HETEs by CYP2C19, CYP1A1, CYP1A2, and CYP2E1 were 5361 ± 193.1, 1899.1 ± 67.7, 1049.1 ± 243.0, and 969.0 ± 13.0 fmol/pmol CYP per min, respectively. Regarding the S-subterminal HETEs, 19S-HETE was the highest HETE formed by CYP2C19 (5199.5 ± 219.9 fmol/pmol CYP per min) followed by CYP1A1 (1507.6 ± 53.1 fmol/pmol CYP per min) and CYP1A2 (946.3 ± 245.8 fmol/pmol CYP per min). On contrary, CYP3A4 formed the lowest S-HETE, 17S-HETE (3.1 ± 0.4), followed by 16S-HETE by 1A2 (Fig. 2B; Table 4). Enantioselectivity was observed among the key P450s involved in the formation of subterminal HETEs, with some P450s showing a preference for either S- or R-subterminal HETEs (Table 3).
3.3. Stereoselectivity of human recombinant P450s in formation of (R/S)-HETEs
The formation of HETE enantiomers showed a definite trend across different P450s for each HETE regioisomer (Table 5). CYP1A2 produced 73% of 5R-HETE and 27% of 5S-HETE, whereas CYP2C19 and CYP4F3A produced 100% 5S-HETE. Interestingly, CYP1A1 did not form detectable 5-HETE enantiomers. CYP1A2 favored the 8S-HETE isomer, producing 86%, whereas CYP2C19 contributed 60% to 8RS-HETE. In comparison, CYP4F3B only formed 8R-HETE. For 9-HETE, CYP3A4 and other P450s, such as CYP1B1 and CYP2B6, showed balanced profiles for both enantiomers, whereas CYP1A2 favored 9S-HETE, producing 67%. CYP2C8 and CYP2C19 exhibited higher formation rates of 11R-HETE than 11S-HETE, at 72% and 77%, respectively, compared with CYP1A1 and CYP4F3B, which exclusively formed 100% 11S-HETE. For 12-HETE, CYP1B1 produced 100% of the R-isomer. Similarly, CYP2C8 and CYP3A4 generated higher amounts of 12R-HETE than 12S-HETE, at 78% and 75%, respectively. In contrast, CYP1A1, CYP1A2, CYP1B1, CYP2B6, CYP2C19, CYP2E1, and CYP2J2 formed 100% of 15S-HETE, whereas CYP3A4 also favored the 15S-HETE enantiomer, producing 70% R- and S-HETE. The enantioselectivity of the HETE isomers of subterminal was analyzed across the tested P450s, and we found significant variation in percentages of the R- and S-enantiomers (Table 5). To begin with 16-HETE, some P450s showed a higher contribution to R-enantiomer formation; for example, CYP1A2 produced 16R-HETE at a rate of 89%. CYP2B6 showed a balanced profile of 50%. Notably, CYP4F3A was unique in forming 100% of 16R-HETE, with no S-enantiomer detected. In contrast, CYP4F12 showed no detectable levels of either enantiomer. The enantioselectivity of 17-HETE was also balanced. Several enzymes such as CYP1A1 and CYP1A2 exhibited a pattern of forming higher percentages of the R-enantiomer at 80% and 83%, respectively. However, some P450s favored the 17S-HETE including CYP4F12 at 62%. A few P450s showed minimal enantioselective formation, for example, CYP2B6. For 18-HETE, CYP2C19 and CYP4F3A exclusively formed 18S-HETE at 100%, whereas CYP2E1 and CYP3A4 showed the opposite, with 100% 18R-HETE. Other P450s such as CYP1A1 showed balanced formation of the 2 enantiomers. The enantioselectivity was also observed for 19-HETE. CYP1A1, CYP1B1, CYP3A4, and CYP4F12 exclusively formed 19S-HETE. Conversely, CYP4F3A entirely formed 19R-HETE. Only a few P450s such as CYP2B6 exhibited more balanced distributions.
Table 5.
Stereoselectivity of human recombinant P450 in formation of (R/S)-HETEs
| 5-HETE |
8-HETE |
9-HETE |
11-HETE |
12HETE |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| R | S | R | S | R | S | R | S | R | S | |
| CYP1A1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100 | 35 | 65 |
| CYP1A2 | 73 | 27 | 14 | 86 | 33 | 67 | 57 | 43 | 100 | 0 |
| CYP1B1 | 68 | 32 | 35 | 65 | 52 | 48 | 13 | 87 | 100 | 0 |
| CYP2B6 | 0 | 0 | 59 | 41 | 54 | 46 | 46 | 54 | 53 | 47 |
| CYP2C8 | 0 | 0 | 65 | 35 | 54 | 46 | 72 | 28 | 78 | 22 |
| CYP2C19 | 0 | 100 | 60 | 40 | 46 | 54 | 77 | 23 | 51 | 49 |
| CYP2E1 | 0 | 0 | 52 | 48 | 59 | 41 | 54 | 46 | 53 | 47 |
| CYP2J2 | 0 | 0 | 50 | 50 | 55 | 46 | 39 | 61 | 57 | 43 |
| CYP3A4 | 37 | 63 | 40 | 60 | 42 | 58 | 38 | 62 | 75 | 25 |
| CYP4F3A | 0 | 100 | 44 | 56 | 0 | 0 | 100 | 0 | 0 | 0 |
| CYP4F3B | 0 | 0 | 100 | 0 | 0 | 0 | 0 | 100 | 0 | 0 |
| CYP4F12 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 15HETE |
16-HETE |
17-HETE |
18-HETE |
19-HETE |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| R | S | R | S | R | S | R | S | R | S | |
| CYP1A1 | 0 | 100 | 70 | 30 | 80 | 20 | 53 | 47 | 0 | 100 |
| CYP1A2 | 0 | 100 | 89 | 11 | 83 | 17 | 62 | 38 | 34 | 66 |
| CYP1B1 | 0 | 100 | 36 | 64 | 60 | 40 | 39 | 61 | 0 | 100 |
| CYP2B6 | 0 | 100 | 50 | 50 | 50 | 50 | 51 | 49 | 47 | 53 |
| CYP2C8 | 21 | 79 | 60 | 40 | 40 | 60 | 49 | 51 | 45 | 55 |
| CYP2C19 | 0 | 100 | 67 | 33 | 100 | 0 | 0 | 100 | 35 | 65 |
| CYP2E1 | 0 | 100 | 47 | 53 | 51 | 49 | 100 | 0 | 34 | 66 |
| CYP2J2 | 0 | 100 | 54 | 46 | 50 | 50 | 44 | 57 | 57 | 43 |
| CYP3A4 | 30 | 70 | 52 | 48 | 63 | 37 | 100 | 0 | 0 | 100 |
| CYP4F3A | 0 | 0 | 100 | 0 | 0 | 0 | 0 | 100 | 100 | 0 |
| CYP4F3B | 0 | 0 | 48 | 52 | 40 | 60 | 54 | 46 | 53 | 47 |
| CYP4F12 | 0 | 0 | 0 | 0 | 38 | 62 | 8 | 92 | 0 | 100 |
The data are presented as %.
3.4. Total formation rates of (R, S)- and (S, R)-EETs enantiomers by human recombinant P450s
The majority of the tested human recombinant P450s formed (R, S)-EETs. CYP2C19 and CYP1A2 exhibited the highest activity in (R, S)-EET formation, followed by CYP2B6, CYP1A1, and CYP2E1. CYP3A4, CYP4F12, and CYP2J2 ranked third (Fig. 3A). The rates of CYP2C19 and CYP1A2 were 42,250.9 ± 3522.8 and 28,558.2 ± 4662.7 fmol/pmol CYP per min, respectively. The formation rates of (R, S)-EETs for CYP2B6, CYP1A1, CYP2E1, CYP3A4, CYP4F12, and CYP2J2 were 1697.4 ± 524.9, 1325.2 ± 81.5, 1277.7 ± 244.4, 346.9 ± 19.4, 437.6 ± 15.5, and 391.9 ± fmol/pmol CYP per min, respectively. The 5R(6S)-EET was the highest formed (R, S)-EET by CYP2C19 followed by CYP1A2, with formation rate of 29,468.3 ± 3146.8 and 21,339.9 ± 3861.5 fmol/pmol CYP per min, respectively. The second highest (R, S)-EET formed was 14R(15S)-EET by CYP2C19 (7519.0 ± 266.7). In addition, CYP2C19 and CYP1A2 played a major role in (R, S)-EETs formation of both 8(9)- and 11(12)-EETs as shown in Fig. 4A and Table 6. Among the tested P450s, most of the P450s contributed to the overall formation of (S, R)-EETs at different profiles. CYP1A2 and CYP2C19 showed the highest formation rates of (S, R)-EETs at 38,676.5± 6519.4 and 27,490.4 ± 2275.0 fmol/pmol CYP per min, Fig. 3B. CYP2B6 was the second, followed by CYP2E1 and CYP1A1. The rates of (S, R)-EETs for CYP2B6, CYP2E1, and CYP1A1 were 4056.1± 1447.0, 1116.0 ± 238.5, and 1108.3 ± 77.2 fmol/pmol CYP per min. The formation rates of (S, R) EETs varied significantly across different regioisomers. Among these, 5S(6R)-EET exhibited the highest activity, with values of 35,840.9 ± 6292.1 and 22,129.0 ± 2044.2 fmol/pmol CYP per min, by CYP1A2 and CYP2C19, respectively. CYP2C19 showed moderate activity in 8S(9R)-EET with its peak formation at 2556.5 ± 141.8, whereas CYP1A2 displayed slightly lower activity in (S, R) isomer of 14(15)-EET formation with a maximum of 1713.2 ± 200.3 fmol/pmol CYP per min. These results highlight the regioselective nature of P450 enzymes in metabolizing AA into enantiomer-specific EETs, with notable differences in activity levels across regioisomers. This variability underscores the importance of studying individual EET isomers and their roles in biological processes (Fig. 4B; Table 6).
Fig. 3.
The metabolic profile of human recombinant P450 enzymes in AA metabolism to form (R, S)/(S, R)-EETs. The incubation conditions and the method of analysis are explained under Materials and methods. Results are presented as mean and SD, n = 3. We used one-way ANOVA followed by Tukey’s multiple comparisons test to compare the formation rates across CYP enzymes; P < .05 indicated significantly different activities in the order. The letters in the figure represent statistical rankings, and enzymes with different letters are significantly different from each other.
Fig. 4.
Heatmap of metabolic profile of human recombinant P450 enzymes showing AA metabolism to (R, S)/(S, R)-EETs. The incubation conditions and the method of analysis are explained under Materials and methods. The results are shown based on the color scale at the bottom of the image.
Table 6.
Formation rates of (R, S)-EET enantiomers by human recombinant P450
The data are presented as mean ± SD, n = 3; unit, fpmole/pmole CYP per min.
| 14 (15)-EET |
11 (12)-EET |
|||
|---|---|---|---|---|
| (S, R) | (R, S) | (S, R) | (R, S) | |
| CYP1A1 | BQL | BQL | 211.4 ± 30.9 | 238.5 ± 19.7 |
| CYP1A2 | 1713.2 ± 200.3 | 2458.9 ± 304 | 404.5 ± 26.4 | 4342.3 ± 540.4 |
| CYP1B1 | BQL | BQL | 104.7 ± 5.8 | 113.5 ± 4.9 |
| CYP2B6 | 122.7 ± 4.8 | 174.0 ± 7.08 | 151.6 ± 13.5 | 178.2 ± 16.7 |
| CYP2C8 | BQL | BQL | BQL | BQL |
| CYP2C19 | 1148.7 ± 42.2 | 7519.0 ± 266.7 | 1656.2 ± 80.3 | 775.3 ± 6.6 |
| CYP2E1 | 182.1 ± 1.4 | 338.1 ± 20.7 | 104.0 ± 6.0 | 237.8 ± 4.5 |
| CYP2J2 | 137.4 ± 1.7 | 117.1 ± 2.6 | 101.7 ± 1.5 | 150.2 ± 5.2 |
| CYP3A4 | 98.9 ± 6.8 | 109.1 ± 9.4 | 102.5 ± 2.7 | 121.9 ± 8.3 |
| CYP4F3A | BQL | BQL | BQL | BQL |
| CYP4F3B | 176.4 ± 2.0 | BQL | BQL | BQL |
| CYP4F12 | BQL | BQL | 215.1 ± 9.6 | 221.8 ± 16.0 |
| 8 (9)-EET |
5 (6)-EET |
|||
|---|---|---|---|---|
| (S, R) | (R, S) | (S, R) | (R, S) | |
| CYP1A1 | BQL | BQL | 896.9 ± 60.2 | 1086.7 ± 79.4 |
| CYP1A2 | 717.9 ± 129.6 | 417.1± 37.2 | 35,840.9 ± 6292.1 | 21,339.9 ± 3861.5 |
| CYP1B1 | BQL | BQL | BQL | BQL |
| CYP2B6 | 248.7 ± 17.7 | 254.0 ± 28.7 | 3533.3 ± 1429.9 | 1091.2 ± 488.6 |
| CYP2C8 | BQL | BQL | BQL | BQL |
| CYP2C19 | 2556.5 ± 141.8 | 4488.3 ± 207.7 | 22,129.0 ± 2044.2 | 29,468.3 ± 3146.8 |
| CYP2E1 | 108.8 ± 2.3 | 123.2 ± 4.0 | 721.1 ± 231.7 | 578.6 ± 219.1 |
| CYP2J2 | BQL | 123.9 ± 0.4 | BQL | BQL |
| CYP3A4 | BQL | 115.8 ± 2.5 | BQL | BQL |
| CYP4F3A | BQL | BQL | BQL | BQL |
| CYP4F3B | BQL | BQL | BQL | BQL |
| CYP4F12 | 239.2 ± 7.4 | 215.8 ± 5.3 | BQL | BQL |
BQL, below quantifiable level.
3.5. Stereoselectivity of human recombinant P450s in formation of (R, S)/(S, R) enantiomers of EETs
The formation of (R, S) and (S, R) enantiomers of EETs showed significant differences across the various P450s, highlighting how different enzymes handle AA metabolism in a regio- and stereoselective manner (Table 7). CYP2B6 form of regioisomer 5(6)-EET (S, R) at 76% while CYP1A2 was next, forming 63%. Some P450s such CYP2C19 showed a relatively balanced production with 43% (S, R) and 57% (R, S). For the 8(9)-EET, CYP1A2 formed more of the (S, R) enantiomer, contributing 63% of the product. CYP3A4 and CYP2J2 also showed completely formed (R, S) form, with 100% and no (S, R) detected. In the case of 11(12)-EET, CYP2B6 produced nearly percentages of both enantiomers, with 54% (R, S) and 46% (S, R) whereas CYP2E1 showed a preference for (R, S), with 70% (R, S) and 30% (S, R). In terms of 14(15)-EET, CYP4F3B exclusively produced the (S, R) form of 14(15)-EET, with 100% (S, R).
Table 7.
Stereoselectivity of human recombinant P450 in formation of (R, S)/(S, R) enantiomers of EETs
| 5 (6)-EET |
8 (9)-EET |
11 (12)-EET |
14 (15)-EET |
|||||
|---|---|---|---|---|---|---|---|---|
| (R, S) | (S, R) | (R, S) | (S, R) | (R, S) | (S, R) | (R, S) | (S, R) | |
| CYP1A1 | 55 | 45 | 0 | 0 | 53 | 47 | 0 | 0 |
| CYP1A2 | 37 | 63 | 37 | 63 | 91 | 9 | 59 | 41 |
| CYP1B1 | 0 | 0 | 0 | 0 | 52 | 48 | 0 | 0 |
| CYP2B6 | 24 | 76 | 51 | 49 | 54 | 46 | 59 | 41 |
| CYP2C8 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| CYP2C19 | 57 | 43 | 64 | 36 | 32 | 68 | 87 | 13 |
| CYP2E1 | 45 | 55 | 53 | 47 | 70 | 30 | 65 | 35 |
| CYP2J2 | 0 | 0 | 100 | 0 | 60 | 40 | 46 | 54 |
| CYP3A4 | 0 | 0 | 100 | 0 | 54 | 46 | 52 | 48 |
| CYP4F3A | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| CYP4F3B | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100 |
| CYP4F12 | 0 | 0 | 47 | 53 | 51 | 49 | 0 | 0 |
The data are presented as %.
4. Discussion
In this study, we demonstrated the AA-metabolizing activity of 12 recombinant human P450 enzymes of different families and subfamilies to conduct a comprehensive profiling in their enantioselectivity in forming HETEs and EETs. Although we tried to include all important AA-metabolizing enzymes in our investigation, there were some P450 families such as CYP4As that were not commercially available. The human recombinant P450 isoforms used in this study exhibit differential amino acid sequence homology, with intrafamily isoforms, such as CYP1A1 and CYP1A2, sharing high-sequence identity (70%), whereas interfamily isoforms, eg, CYP2B6 and CYP2C19, display markedly lower homology (59%).31 We found that CYP2 family, particularly, CYP2C subfamily, CYP2C19 isoform, was one of the most unique P450 enzymes as it formed the highest (R/S)-enantiomers of subterminal HETEs, as well as S-enantiomer of midchain HETEs. CYP2C19 has shown the highest activity of AA metabolism to enantiomers of 19-HETE. 19(S)-HETE was the most abundant enantiomer formed by CYP2C19 followed by 19(R)-HETE. This is in agreement with a previous study that reported CYP2C19 as the enzyme with the highest AA-metabolizing activity for HETE formation, particularly 19-HETE. However, that study focused on profiling P450s in terms of their involvement in forming racemic HETE.3 Also, we reported that 19-HETE was the most abundant HETE formed by HLMs, with S-enantiomer formed greater than 19(R)-HETE.7 In addition, CYP1 family, particularly, CYP1A2 isoform, has shown highest activity in forming the isomer of 5S(6R)-EET, next to CYP2C19 which also showed to form their corresponding 5R(6S)-EET. The results of the EETs and HETEs agreed that CYP1A2 and CYP2C19 to be major important human P450s that play a role in AA metabolism. The enantioselectivity of CYP2C19 for 14R(15S)-EET and the near-racemic composition of 11(12)-EET are consistent with a previous study on human recombinant P450s which also reported the racemic composition of 11(12)-EET by human CYP1B1.27 There are few factors that could explain the enantioselectivity associated with an enzyme. These factors include the active site geometry, which varies in shape and the amino acid residue of the enzyme, and the active site volume. The orientation of the substrate could favor a chemical bond to be accessible to the enzyme leading to enzyme activity.3 Also, the electronic configuration in the active sites where heme ion in P450 enzymes along with the surrounded amino acids could influence the orientation of chemical bond and could lead to differential activity. For example, it has been reported that heme moiety structurally organizes the active site cavity of CYP1A2 and facilitates substrate binding through its influence on nearby protein residues and interactions with the ligand.32 CYP2 family especially CYP2A for example has been shown to have higher degree of structural flexibility.32 The existence of small, planar active site cavities, as well as critical residues such as Asn297, allows substrates to be precisely positioned and oriented within the enzyme.32 In addition to the structural factors, genetic polymorphism can influence on P450 expression and their activity leading to differential processing of enantiomers, leading to variations in the formation or metabolism of AA. For examples, CYP1A2∗1F variant increases CYP1A2 enzyme activity in smokers, whereas variants such as CYP2C92/3 decrease its metabolic capacity. Additionally, polymorphisms such as CYP2C19∗2 result in complete enzyme inactivation.4 The highlighting enantioselectivity of P450 aids the understanding of profile of the isomer in a biological system. For example, endogenous 19-HETE enantiomers show enantiospecific properties in their effects. 19 (S)-HETE inhibits CYP1B1 more strongly than R-enantiomer (Ki = 37.3 and 89.1 nM, respectively).33 Similarly, 17(S)-HETE was consistently higher as CYP1B1 allosteric activator of human recombinant CYP1B1 compared with 17(R)-HETE.28 In addition, the effects of (16)R-HETE were more potent compared with 16(S)-HETE in activating human recombinant CYP1B1.13 Beyond P450 interaction, 5(R)-HETE is a more potent neutrophil chemoattractant than 5(S)-HETE.34 12(S)-HETE induces angiogenesis and tumor metastasis, whereas 12(R)-HETE does not exhibit these properties.35,36 Interestingly, 12(R)-HETE exerts a strong inhibitory effect in renal and cardiac tissue through Na+/K+-ATPase blockage.37 This stereoselectivity is not limited to HETEs; EETs also display enantiomer-specific biological effects. The observed enantioselectivity underscores the significance of stereochemistry in drug development and evaluation. A well known example is thalidomide, which exists as (R)- and (S)-enantiomers. (R)-enantiomer was intended for therapeutic purposes, but in vivo racemization converted it to the teratogenic (S)-form, leading to teratogenicity.33 This shows the importance of evaluating both stereoisomers for safety, not only efficacy. Other examples include ibuprofen, where only (S)-ibuprofen is active, and the R-isomer is inactive.38 However, some of the R-isomer is converted to S-ibuprofen, demonstrating the complexity of stereochemistry in drug metabolism and activity.38 AA concentration of 75 μM was justified based on existing literature that reported higher levels of unesterified AA in vivo. Although the plasma levels of AA are typically in the nanomolar range,39 tissue levels have been reported to be micromolar to millimolar range. For example, the levels of unesterified AA have been measured at 13–44 μM in the umbilical cord and intervillous space,40 approximately 60 μM in the skin,41 and up to 250 μM in the liver.42 For this reason, AA concentrations of 50–100 μM are commonly used in published studies examining P450-mediated AA metabolism in incubations.43 To assess the impact of substrate concentration on enzyme behavior and stereoselectivity, we selected 3 and 10 μM AA because 3 μM is below the reported Km values for most human P450 enzymes, whereas 10 μM approaches these values (Supplemental Table 1). When stereoselectivity at 3 μM was compared with that at 75 μM, most HETEs retained their original enantiomeric preferences; however, several products including 16 HETE from CYP1A2 and 8, 16, and 17-HETEs from CYP2C19 became racemic, and others were below the detection limit, consistent with reduced catalytic efficiency under substrate limited conditions (Supplemental Table 2A). The enantiomeric distribution of EETs was also largely consistent with that at 75 μM, indicating that stereoselectivity is predominantly concentration independent (Supplemental Table 2B). At 10 μM AA, CYP1A2 generally maintained its stereochemical profile relative to 75 μM, with only modest differences such as a slight reduction in the R preference for 16 HETE and minor shifts in EET formation (Supplemental Table 2, A and B). The main exception was 11(12) EET, which showed reduced stereoselectivity at the lower substrate concentration (Supplemental Table 2B). CYP2C19 displayed similar stereochemical trends to those observed at 75 μM, maintaining stereoselectivity for 9, 15, and 19 HETE and 14(15) EET, whereas 11(12) EET exhibited diminished stereochemical bias at low substrate levels (Supplemental Table 2, A and B). Additionally, several metabolites were below the detection limit under these conditions (Supplemental Table 2A). Overall, most stereochemical trends observed at 75 μM were preserved. These results indicate that increasing AA concentrations indeed promoted a more stereoselective formation of AA metabolites, mimicking the release of AA from cell membranes under pathological conditions. The relative contributions of individual P450 to AA metabolism were estimated by combining in vitro formation rates with reported hepatic abundances (Supplemental Table 3). The abundance of P450 was obtained from the literature.44,45 Although CYP1A2 and CYP2C19 exhibited high in vitro formation rates per pmol of enzyme (70.63 and 79.72 pmol/pmol per min), respectively, their activities scaled to hepatic abundance were 2442.0 and 1116.55 pmol/mg per min, consistent with their significant contributions to HETE and EET formation (Supplemental Table 3). In contrast, CYP3A4, despite a lower per-enzyme formation rate (2.14 pmol/pmol per min), displayed substantial scaled activity (304.07 pmol/mg per min) due to its high liver expression (Supplemental Table 3). Contributions from other isoforms were comparatively lower, reflecting either reduced intrinsic activity or diminished hepatic expression. These findings highlight that both catalytic efficiency and enzyme abundance must be considered to accurately evaluate the physiological relevance of individual P450 isoforms in AA metabolism. The AA degradation half-life was derived from the depletion curve, and the formation rates of EETs and HETEs were incorporated to quantify their respective contributions to overall metabolic activity (Supplemental Table 4 and Supplemental Fig. 3). This integrated approach enhances the interpretation of metabolic activity by linking substrate disappearance with the production of metabolites. To ensure comparable enzyme-substrate dynamics between HLMs and recombinant P450 isoforms, (AA) concentration used maintained AA-to-enzyme ratio. For instance, CYP2C19 incubations employed 75 μM AA with 50 pmol/mL enzyme for HETEs, yielding an AA/P450 ratio of 1.5 μM/pmol. This is approximately equivalent to the 2 μM AA used in HLM incubations, which corresponds to 2–3 pmol of CYP2C19/mL. By adjusting the substrate concentration in this manner, the depletion study preserves physiologically relevant substrate exposure relative to enzyme content, allowing the resulting kinetic data to be meaningfully compared across systems. Our study has several limitations. First, we used human recombinant P450 enzymes in an incubation system that might not completely reflect the in vivo complexity. Second, we attempted to include the major isoforms, but not all human P450 enzymes or polymorphic variants were included. Third, the metabolite formation and stereoselectivity were assessed at a single AA concentration. Consequently, the physiological relevance of these findings should be interpreted cautiously. Future studies exploring a broader range of AA concentrations and determining intrinsic clearance (Vmax/Km) will be important to provide a more predictive assessment of in vivo metabolism.
This study investigated AA metabolism by human P450 enzymes, emphasizing the enantioselective formation of HETEs and EETs. This study investigated AA metabolism by human P450 enzymes, emphasizing the enantioselective formation of HETEs and EETs. Knowing the stereochemistry of these molecules is crucial as it aids identification of isomers that are involved in specific diseases. Our study contributed to identifying the enantioselective patterns. Although it might not always be practical to inhibit only one enantiomer directly, exploring and understanding the P450 isoforms that preferentially produce the harmful versus beneficial enantiomer could lead to opportunities for selective isoform inhibition, enzyme engineering, or drug design that could shift the metabolic balance or modulate cofactors or conditions that influence enantioselectivity.
Conflict of interest
The authors declare no conflicts of interest.
Acknowledgments
Financial support
This work was supported by a grant from the Canadian Institutes of Health Research (CIHR), [Grant CIHR PS 168846].
Data availability
All relevant data are included in this paper or can be obtained from the corresponding author upon reasonable request.
CRediT authorship contribution statement
Fadumo Ahmed Isse: Conceptualization, Investigation, Data curation, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. Ahmed A. El-Sherbeni: Conceptualization, review & editing. Ayman O.S. El-Kadi: Conceptualization, and Supervision.
Footnotes
This article has supplemental material available at dmd.aspetjournals.org.
Supplemental material
References
- 1.Wang T., Fu X., Chen Q., et al. Arachidonic acid metabolism and kidney inflammation. Int J Mol Sci. 2019;20(15):3683. doi: 10.3390/ijms20153683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.ElKhatib M.A.W., Isse F.A., El-Kadi A.O.S. Effect of inflammation on cytochrome P450-mediated arachidonic acid metabolism and the consequences on cardiac hypertrophy. Drug Metab Rev. 2023;55(1-2):50–74. doi: 10.1080/03602532.2022.2162075. [DOI] [PubMed] [Google Scholar]
- 3.El-Sherbeni A.A., El-Kadi A.O. Repurposing resveratrol and fluconazole to modulate human cytochrome P450-mediated arachidonic acid metabolism. Mol Pharm. 2016;13(4):1278–1288. doi: 10.1021/acs.molpharmaceut.5b00873. [DOI] [PubMed] [Google Scholar]
- 4.Zordoky B.N., El-Kadi A.O. Effect of cytochrome P450 polymorphism on arachidonic acid metabolism and their impact on cardiovascular diseases. Pharmacol Ther. 2010;125(3):446–463. doi: 10.1016/j.pharmthera.2009.12.002. [DOI] [PubMed] [Google Scholar]
- 5.Wei S., Brittin J.J., Falck J.R., et al. Chiral resolution of the epoxyeicosatrienoic acids, arachidonic acid epoxygenase metabolites. Anal Biochem. 2006;352(1):129–134. doi: 10.1016/j.ab.2006.01.005. [DOI] [PubMed] [Google Scholar]
- 6.Kiss L., Röder Y., Bier J., Weissmann N., Seeger W., Grimminger F. Direct eicosanoid profiling of the hypoxic lung by comprehensive analysis via capillary liquid chromatography with dual online photodiode-array and tandem mass-spectrometric detection. Anal Bioanal Chem. 2008;390(2):697–714. doi: 10.1007/s00216-007-1718-9. [DOI] [PubMed] [Google Scholar]
- 7.Isse F.A., Alammari A.H., El-Sherbeni A.A., Brocks D.R., El-Kadi A.O.S. The enantioselective separation and quantitation of the hydroxy-metabolites of arachidonic acid by liquid chromatography-tandem mass spectrometry. Prostaglandins Other Lipid Mediat. 2023;165 doi: 10.1016/j.prostaglandins.2022.106701. [DOI] [PubMed] [Google Scholar]
- 8.Reddy M.A., Thimmalapura P.R., Lanting L., Nadler J.L., Fatima S., Natarajan R. The oxidized lipid and lipoxygenase product 12(S)-hydroxyeicosatetraenoic acid induces hypertrophy and fibronectin transcription in vascular smooth muscle cells via p38 MAPK and cAMP response element-binding protein activation. Mediation of angiotensin II effects. J Biol Chem. 2002;277(12):9920–9928. doi: 10.1074/jbc.M111305200. [DOI] [PubMed] [Google Scholar]
- 9.Shoieb S.M., El-Ghiaty M.A., Alqahtani M.A., El-Kadi A.O.S. Cytochrome P450-derived eicosanoids and inflammation in liver diseases. Prostaglandins Other Lipid Mediat. 2020;147 doi: 10.1016/j.prostaglandins.2019.106400. [DOI] [PubMed] [Google Scholar]
- 10.Isse F.A., El-Sherbeni A.A., El-Kadi A.O.S. The multifaceted role of cytochrome P450-Derived arachidonic acid metabolites in diabetes and diabetic cardiomyopathy. Drug Metab Rev. 2022;54(2):141–160. doi: 10.1080/03602532.2022.2051045. [DOI] [PubMed] [Google Scholar]
- 11.Maayah Z.H., El-Kadi A.O. The role of mid-chain hydroxyeicosatetraenoic acids in the pathogenesis of hypertension and cardiac hypertrophy. Arch Toxicol. 2016;90(1):119–136. doi: 10.1007/s00204-015-1620-8. [DOI] [PubMed] [Google Scholar]
- 12.Shoieb S.M., Alammari A.H., Levasseur J., Silver H., Dyck J.R.B., El-Kadi A.O.S. Ameliorative role of fluconazole against abdominal aortic constriction-induced cardiac hypertrophy in rats. J Cardiovasc Pharmacol. 2022;79(6):833–845. doi: 10.1097/fjc.0000000000001258. [DOI] [PubMed] [Google Scholar]
- 13.Hidayat R., Shoieb S.M., Mosa F.E.S., et al. 16R-HETE and 16S-HETE alter human cytochrome P450 1B1 enzyme activity probably through an allosteric mechanism. Mol Cell Biochem. 2024;479(6):1379–1390. doi: 10.1007/s11010-023-04801-4. [DOI] [PubMed] [Google Scholar]
- 14.Shoieb S.M., El-Kadi A.O.S. S-Enantiomer of 19-hydroxyeicosatetraenoic acid preferentially protects against angiotensin II-induced cardiac hypertrophy. Drug Metab Dispos. 2018;46(8):1157–1168. doi: 10.1124/dmd.118.082073. [DOI] [PubMed] [Google Scholar]
- 15.Shoieb S.M., El-Sherbeni A.A., El-Kadi A.O.S. Identification of 19-(S/R)hydroxyeicosatetraenoic acid as the first endogenous noncompetitive inhibitor of cytochrome P450 1B1 with enantioselective activity. Drug Metab Dispos. 2019;47(2):67–70. doi: 10.1124/dmd.118.084657. [DOI] [PubMed] [Google Scholar]
- 16.Imig J.D. Epoxyeicosanoids in hypertension. Physiol Res. 2019;68(5):695–704. doi: 10.33549/physiolres.934291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Alsaad A.M.S., Zordoky B.N.M., Tse M.M.Y., El-Kadi A.O.S. Role of cytochrome P450-mediated arachidonic acid metabolites in the pathogenesis of cardiac hypertrophy. Drug Metab Rev. 2013;45(2):173–195. doi: 10.3109/03602532.2012.754460. [DOI] [PubMed] [Google Scholar]
- 18.El-Sherbeni A.A., El-Kadi A.O. Alterations in cytochrome P450-derived arachidonic acid metabolism during pressure overload-induced cardiac hypertrophy. Biochem Pharmacol. 2014;87(3):456–466. doi: 10.1016/j.bcp.2013.11.015. [DOI] [PubMed] [Google Scholar]
- 19.Elkhatali S., Maayah Z.H., El-Sherbeni A.A., et al. Inhibition of mid-chain HETEs protects against angiotensin II-induced cardiac hypertrophy. J Cardiovasc Pharmacol. 2017;70(1):16–24. doi: 10.1097/fjc.0000000000000494. [DOI] [PubMed] [Google Scholar]
- 20.Maayah Z.H., El-Kadi A.O.S. 5-, 12- and 15-Hydroxyeicosatetraenoic acids induce cellular hypertrophy in the human ventricular cardiomyocyte, RL-14 cell line, through MAPK- and NF-κB-dependent mechanism. Arch Toxicol. 2016;90(2):359–373. doi: 10.1007/s00204-014-1419-z. [DOI] [PubMed] [Google Scholar]
- 21.Tacconelli S., Fullone R., Dovizio M., et al. Pharmacological characterization of the biosynthesis of prostanoids and hydroxyeicosatetraenoic acids in human whole blood and platelets by targeted chiral lipidomics analysis. Biochim Biophys Acta Mol Cell Biol Lipids. 2020;1865(12) doi: 10.1016/j.bbalip.2020.158804. [DOI] [PubMed] [Google Scholar]
- 22.Neilson A.P., Ren J., Hong Y.H., et al. Effect of fish oil on levels of R- and S-enantiomers of 5-, 12-, and 15-hydroxyeicosatetraenoic acids in mouse colonic mucosa. Nutr Cancer. 2012;64(1):163–172. doi: 10.1080/01635581.2012.630168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mesaros C., Blair I.A. Targeted chiral analysis of bioactive arachidonic acid metabolites using liquid-chromatography-mass spectrometry. Metabolites. 2012;2(2):337–365. doi: 10.3390/metabo2020337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mazaleuskaya L.L., Salamatipour A., Sarantopoulou D., et al. Analysis of HETEs in human whole blood by chiral UHPLC-ECAPCI/HRMS. J Lipid Res. 2018;59(3):564–575. doi: 10.1194/jlr.D081414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hong S.H., Han J.E., Ko J.S., Do S.H., Lee E.H., Cho M.H. Quantitative determination of 12-hydroxyeicosatetraenoic acids by chiral liquid chromatography tandem mass spectrometry in a murine atopic dermatitis model. J Vet Sci. 2015;16(3):307–315. doi: 10.4142/jvs.2015.16.3.307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kiss L., Bier J., Röder Y., Weissmann N., Grimminger F., Seeger W. Direct and simultaneous profiling of epoxyeicosatrienoic acid enantiomers by capillary tandem column chiral-phase liquid chromatography with dual online photodiode array and tandem mass spectrometric detection. Anal Bioanal Chem. 2008;392(4):717–726. doi: 10.1007/s00216-008-2308-1. [DOI] [PubMed] [Google Scholar]
- 27.Mesaros C., Lee S.H., Blair I.A. Analysis of epoxyeicosatrienoic acids by chiral liquid chromatography/electron capture atmospheric pressure chemical ionization mass spectrometry using [13C]-analog internal standards. Rapid Commun Mass Spectrom. 2010;24(22):3237–3247. doi: 10.1002/rcm.4760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Isse F.A., Alammari A.H., El-Sherbeni A.A., El-Kadi A.O.S. 17-(R/S)-hydroxyeicosatetraenoic acid (HETE) induces cardiac hypertrophy through the CYP1B1 in enantioselective manners. Prostaglandins Other Lipid Mediat. 2023;168 doi: 10.1016/j.prostaglandins.2023.106749. [DOI] [PubMed] [Google Scholar]
- 29.Sakairi Y., Jacobson H.R., Noland T.D., Capdevila J.H., Falck J.R., Breyer M.D. 5,6-EET inhibits ion transport in collecting duct by stimulating endogenous prostaglandin synthesis. Am J Physiol. 1995;268(5):F931–F939. doi: 10.1152/ajprenal.1995.268.5.F931. [DOI] [PubMed] [Google Scholar]
- 30.Isse F.A., Helal S., El-Sherbeni A.A., Brocks D.R., El-Kadi A.O.S. Quantification of Epoxyeicosatrienoic acids enantiomers: the development of reliable and practical liquid chromatography mass spectrometry assay. J Chromatogr B Analyt Technol Biomed Life Sci. 2024;1247 doi: 10.1016/j.jchromb.2024.124346. [DOI] [PubMed] [Google Scholar]
- 31.Lewis D.F., Watson E., Lake B.G. Evolution of the cytochrome P450 superfamily: sequence alignments and pharmacogenetics. Mutat Res. 1998;410(3):245–270. doi: 10.1016/s1383-5742(97)00040-9. [DOI] [PubMed] [Google Scholar]
- 32.Gay S.C., Roberts A.G., Halpert J.R. Structural features of cytochromes P450 and ligands that affect drug metabolism as revealed by X-ray crystallography and NMR. Future Med Chem. 2010;2(9):1451–1468. doi: 10.4155/fmc.10.229. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Tokunaga E., Yamamoto T., Ito E., Shibata N. Understanding the thalidomide chirality in biological processes by the self-disproportionation of enantiomers. Sci Rep. 2018;8(1):17131. doi: 10.1038/s41598-018-35457-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Bittleman D.B., Casale T.B. 5-Hydroxyeicosatetraenoic acid (HETE)-induced neutrophil transcellular migration is dependent upon enantiomeric structure. Am J Respir Cell Mol Biol. 1995;12(3):260–267. doi: 10.1165/ajrcmb.12.3.7873191. [DOI] [PubMed] [Google Scholar]
- 35.Guo Y., Zhang W., Giroux C., et al. Identification of the orphan G protein-coupled receptor GPR31 as a receptor for 12-(S)-hydroxyeicosatetraenoic acid. J Biol Chem. 2011;286(39):33832–33840. doi: 10.1074/jbc.M110.216564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Helal S.A., Gerges S.H., El-Kadi A.O.S. Enantioselectivity in some physiological and pathophysiological roles of hydroxyeicosatetraenoic acids. Drug Metab Rev. 2024;56(1):31–45. doi: 10.1080/03602532.2023.2284110. [DOI] [PubMed] [Google Scholar]
- 37.Masferrer J.L., Rios A.P., Schwartzman M.L. Inhibition of renal, cardiac and corneal (Na(+)-K+)ATPase by 12(R)-hydroxyeicosatetraenoic acid. Biochem Pharmacol. 1990;39(12):1971–1974. doi: 10.1016/0006-2952(90)90617-t. [DOI] [PubMed] [Google Scholar]
- 38.Evans A.M. Comparative pharmacology of S(+)-ibuprofen and (RS)-ibuprofen. Clin Rheumatol. 2001;20(suppl 1):S9–S14. doi: 10.1007/bf03342662. [DOI] [PubMed] [Google Scholar]
- 39.Brash A.R. Arachidonic acid as a bioactive molecule. J Clin Invest. 2001;107(11):1339–1345. doi: 10.1172/jci13210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Benassayag C., Mignot T.M., Haourigui M., et al. High polyunsaturated fatty acid, thromboxane A2, and alpha-fetoprotein concentrations at the human feto-maternal interface. J Lipid Res. 1997;38(2):276–286. doi: 10.1016/S0022-2275(20)37440-X. [DOI] [PubMed] [Google Scholar]
- 41.Hammarström S., Hamberg M., Samuelsson B., Duell E.A., Stawiski M., Voorhees J.J. Increased concentrations of nonesterified arachidonic acid, 12L-hydroxy-5,8,10,14-eicosatetraenoic acid, prostaglandin E2, and prostaglandin F2alpha in epidermis of psoriasis. Proc Natl Acad Sci U S A. 1975;72(12):5130–5134. doi: 10.1073/pnas.72.12.5130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Edpuganti V., Mehvar R. UHPLC-MS/MS analysis of arachidonic acid and 10 of its major cytochrome P450 metabolites as free acids in rat livers: effects of hepatic ischemia. J Chromatogr B Analyt Technol Biomed Life Sci. 2014;964:153–163. doi: 10.1016/j.jchromb.2013.08.008. [DOI] [PubMed] [Google Scholar]
- 43.El-Sherbeni A.A., El-Kadi A.O. Characterization of arachidonic acid metabolism by rat cytochrome P450 enzymes: the involvement of CYP1As. Drug Metab Dispos. 2014;42(9):1498–1507. doi: 10.1124/dmd.114.057836. [DOI] [PubMed] [Google Scholar]
- 44.Lang D., Radtke M., Bairlein M. Highly variable expression of CYP1A1 in human liver and impact on pharmacokinetics of riociguat and granisetron in humans. Chem Res Toxicol. 2019;32(6):1115–1122. doi: 10.1021/acs.chemrestox.8b00413. [DOI] [PubMed] [Google Scholar]
- 45.Chen Y., Liu L., Nguyen K., Fretland A.J. Utility of intersystem extrapolation factors in early reaction phenotyping and the quantitative extrapolation of human liver microsomal intrinsic clearance using recombinant cytochromes P450. Drug Metab Dispos. 2011;39(3):373–382. doi: 10.1124/dmd.110.035147. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All relevant data are included in this paper or can be obtained from the corresponding author upon reasonable request.