Letter pubs.acs.org/JAFC
Food Polyphenol Apigenin Inhibits the Cytochrome P450 Monoxygenase Branch of the Arachidonic Acid Cascade ABSTRACT: The product of cytochrome P450 monooxygenase (P450) ω-hydroxylation of arachidonic acid (AA), 20hydroxyeicosatetraenoic acid (HETE), is a potent vasoconstrictor. Utilizing microsomes as well as individual CYP4 isoforms we demonstrate here that flavonoids can block 20-HETE formation. Apigenin inhibits CYP4F2 with an IC50 value of 4.6 μM and 20HETE formation in human liver and kidney microsomes at 2.4−9.8 μM. Interestingly, the structurally similar naringenin shows no relevant effect on the formation of 20-HETE. Based on these in vitro data, it is impossible to evaluate if a relevant blockade of 20-HETE formation can result in humans from intake of polyphenols with the diet. However, the potency of apigenin is comparable to those of P450 inhibitors such as ketoconazole. Moreover, an IC50 value in the micromolar range is also described for the inhibition of CYP-mediated drug metabolism leading to food−drug interactions. The modulation of the arachidonic acid cascade by food polyphenols therefore warrants further investigation. KEYWORDS: eicosanoids, oxylipins, polyphenols, flavonoides, enzyme inhibition, food drug interaction
■
■
INTRODUCTION
Chemicals and Biological Materials. Oxylipin standards and internal standards, HET0016, ketoconazole, and arachidonic acid were obtained from Cayman Chemicals (local distributor: Biomol, Hamburg, Germany). Apigenin (≥97%), naringenin (≥95), and isocitric dehydrogenase (NADP) from Bacillus subtilis were purchased from Sigma-Aldrich (Schnelldorf, Germany). Human liver microsomes (HLM Pooled Male Donors, 20 mg/mL), rat liver microsomes (RLM Pooled Male Sprague−Dawley, 20 mg/mL), CYP4F2 (Supersomes Human CYP4F2 + P450 Reductase + b5 0.5 nmol, 7 mg/mL) and CYP4A11 (Supersomes Human CYP4A11 + P450 Reductase + b5 0.5 nmol, 7 mg/mL) were obtained from Corning Gentest (Kaiserslautern, Germany). Human kidney microsomes (HKM Mixed Gender Pooled, 20 mg/mL) were purchased from BioreclamationIVT (Frankfurt am Main, Germany). Microsomal Incubations and Quantification of 19-HETE and 20-HETE. Incubations were performed in a total volume of 200 μL of 0.1 M potassium phosphate buffer (pH 7.4) containing 100 μM arachidonic acid (ARA) and 10 μL of microsome solution (0.3−1 mg/ mL microsomal protein) as described9 with slight modifications. In brief, 10 μL of ARA 2 mM in ethanol was transferred into a 1.5 mL plastic tube (Sarstedt, Nuembrecht, Germany, article 72.690.001), evaporated under nitrogen, and dissolved in buffer containing microsomes. The inhibitor was added in DMSO (final concentration ≤ 1%), and the mixture was pre-incubated for 10 min at 37 °C. Reactions were started with the addition of 5 μL of NADPH generating system (concentrations in assay: 41 mM NADP+, 0.13 M MgCl2, 0.27 M isocitrate, 11 U/mL isocitrate dehydrogenase). Reactions were terminated after 20 min by liquid/liquid extraction with ethyl acetate (two times with 600 μL) and the addition of 5 pmol of 5-HETEd8 as internal standard (IS). The combined ethyl acetate phases were evaporated, resuspended in 50 μL of MeOH, and analyzed by LC-MS. LC-MS oxylipin analysis was carried out as described10 utilizing an SCIEX 3200 triple-quadrupole instrument. In brief, separation was carried out on a reversed phase column (2.1 mm × 100 mm, 1.8 μm particle) at a flow rate of 400 μL/min with a binary gradient of water and a mixture of acetonitrile, methanol, and acetic acid. Detection was carried out in scheduled selected reaction monitoring (SRM) following negative electrospray ionization. The
A large number of highly potent lipid mediatorsso-called oxylipinsare formed in the arachidonic acid (ARA) cascade.1 Whereas cyclooxygenase-catalyzed formation of prostaglandins and leukotriene formation by lipoxygenase have been intensively studied since the early 1970s, the importance of the cytochrome P450 monooxygenase catalyzed reactions became apparent in the past decades.2 In this third branch, ARA and other polyunsaturated fatty acids (PUFA) are converted to anti-inflammatory, analgesic, and vasodilatory epoxy-PUFA as well as hydroxy-PUFA. Terminal hydroxylation of ARA leads to 20-hydroxyeicosatetraenoic acid (HETE), a potent lipid hypertensive oxylipin. This lipid mediator regulates vasoconstriction as well as renal salt absorption. Thus, 20HETE plays an important role in blood pressure regulation and is implicated in the occurrence of pathophysiological hypertension.3 Induction and inhibition of P450 activity is among the bestinvestigated food−drug interactions.4−7 Modulating phase I and phase II drug metabolism in several ways, some food polyphenols inhibit P450 (for example, CYP1A2 or CYP3A4) and the relevance of these health effects is well documented, for example for the consumption of grapefruit and grapefruit juice.4,5 The active polyphenol naringenin inhibits not only the CYP3A family but also CYP2C9 and CYP2C19 with an IC50 below 5 μM.8 The same enzymes catalyze the formation of oxylipins in the third branch of the ARA cascade.3 However, no data are available on a modulation of P450-derived oxylipin formation by food ingredients. Using individual P450 isoforms as microsomal preparations from tissues important in endogenous 20-HETE formation (liver and kidney), we demonstrate in this study for the first time that food polyphenols could block the third branch of the ARA cascade. Inhibition of 20-HETE formation by the exemplary studied flavonoids apigenin and naringenin was compared with those of established P450 inhibitors. With a high specificity and almost similar potency in the micromolar range these effects could be of biological relevance. © 2016 American Chemical Society
MATERIALS AND METHODS
Received: Revised: Accepted: Published: 8973
October 10, 2016 November 11, 2016 November 14, 2016 November 14, 2016 DOI: 10.1021/acs.jafc.6b04501 J. Agric. Food Chem. 2016, 64, 8973−8976
Letter
Journal of Agricultural and Food Chemistry
Table 1. Inhibition of 20-HETE Formation in Human Liver Microsomes (HLM), Human Kidney Microsomes (HKM), and Rat Liver Microsomes (RLM) as well as CYP4F2 and CYP4A11a IC50, μM (95% CI) HET0016 RLM HLM HKM CYP4F2 CYP4A11
0.22 (0.09−0.53) 0.041 (0.028−0.060) 0.013 (0,006−0,031) 0.037 (0.017−0.080) 0.155 (0.073−0.33)
ketoconazole 6.9 5.4 43 0.83 >100
apigenin
(3.8−12.8) (3.7−8.1) (17−106) (0.34−2.0)
50 2.4 9.8 4.6 >100
naringenin
(16−154) (1.5−3.8) (5.7−16.6) (1.9−11.1)
>100 >100 >100 100: 54 ± 5% inhibition >100
Substrate (arachidonic acid) concentration was 100 μM. The inhibition was calculated on the basis of product formation (100 − % of control), and IC50 and 95% confidence intervals were calculated by fitting of the resulting dose response curves. a
Table 2. Inhibition of 19-HETE Formation in Human Liver Microsomes (HLM), Human Kidney Microsomes (HKM), and Rat Liver Microsomes (RLM) as well as CYP4F2 and CYP4A11a IC50, μM (95% CI) RLM HLM HKM CYP4F2 CYP4A11
HET0016
ketoconazole
apigenin
naringenin
ndb 1.0 (0.22−4.8) 0.050 (0.030−0.083) nd 0.285 (0.043−1.89)
13.1 (4.3−40.3) 6.1 (3.6−10.4) >100 nd >100
26 (8−87) 2.3 (0.88−5.9) >100 nd >100
>100 >100 >100 nd >100
Substrate (arachidonic acid) concentration was 100 μM. The inhibition was calculated on the basis of product formation (100 − % of control), and IC50 and 95% confidence interval were calculated by fitting of the resulting dose response curves. bnd, not determined, no formation of 19-HETE. a
Figure 1. Effect of specific CYP4 inhibitor HET0016, nonspecific P450 inhibitor ketoconazole, and food polyphenol apigenin on (A) 20-HETE and (B) 19-HETE formation in human liver microsomes. The inhibition was calculated on the basis of product formation (100 − % of control). limit of detection of the method was 10 nM for 20-HETE. Details about the method can be found in the Supporting Information. Inhibitory effects were calculated on the basis of 20-HETE and 19HETE formation, respectively. Two independent replicates were carried out for each compound/concentration. The combined data were fitted as dose response curve (GraphPad Prism 5.0 (GraphPad Software, San Diego, CA, USA)) for calculation of the IC50 values and 95% confidence interval (CI).
formed during incubation, which is >30-fold the LLOQ of the detection method, allowing a precise and robust measurement of inhibition. When RLM and HLM were incubated with the nonspecific P450 inhibitor ketoconazole, the formation of both 20-HETE and 19-HETE was reduced with increasing concentration. On the basis of the resulting dose response curves, IC50 values of 5.4 and 6.9 μM were calculated for HLM and RLM, respectively (Tables 1 and 2), based on 20-HETE formation. This potency is consistent with IC50 in the micromolar range observed for the conversion of other substrates by HLM, such as 8.7 μM for paclitaxel (CYP2C8) and 9 μM for diclofenac (CYP2C9), or individual P450, for example, 0.74 μM for CYP4F12 and 4.6 μM for CYP2J2.11 Whereas 19-HETE is the product of ARA conversion by several human P450 families including CYP4 and CYP2 and the isoforms 1A1, 1A2, and 2E1, among others, 20-HETE is dominantly formed by CYP4A and CYP4F subfamilies.3 Consistently, 20-HETE formation in liver and kidney micro-
■
RESULTS AND DISCUSSION Microsomal incubations in combination with an LC-MS readout were used to evaluate the inhibition of hydroxy-FA formation in the third branch of the ARA cascade by food ingredients. The ARA concentration was set to 100 μM in the range of the determined KM for 19/20-HETE formation in RLM (Supporting Information). The assay was terminated after 20 min during linear product formation, allowing sensitive detection of a modulation of enzyme velocity (Supporting Information). In the absence of inhibitor >1 μM of product was 8974
DOI: 10.1021/acs.jafc.6b04501 J. Agric. Food Chem. 2016, 64, 8973−8976
Letter
Journal of Agricultural and Food Chemistry †
somes could be potently blocked by nanomolar concentration of the specific inhibitor HET0016 (Figure 1; Table 1). The observed IC50 value of 13 nM in HKM is consistent with 8.9 nM reported by Miyata and Roman.12 In human renal and hepatic tissue CYP4F2 and CYP4A11 significantly contribute to 20-HETE formation.3,13 Consistently, 20-HETE formation by CYP4F2 and CYP4A11 was blocked by HET0016 with IC50 values of 37 and 155 nM, respectively (Table 1). The assay was then used to evaluate how the food polyphenols affect the third branch of the ARA cascade. The grapefruit flavonoids naringenin and apigenin found, for example, in celery or bell pepper were selected because they are known to inhibit P450 metabolism of drugs.6,7,8,14 At an initial screening concentration of 100 μM naringenin did not cause a relevant inhibition (
