Miconidin Acetate and Primin as Potent 5-Lipoxygenase Inhibitors from Brazilian Eugenia hiemalis (Myrtaceae)

Abstract

This paper describes the isolation and identification of primin and miconidin acetate as metabolites from the flower bud extract of Eugenia hiemalis as well as the anti-inflammatory activity of miconidin acetate by inhibition of 5-lipoxygenase. Miconidin acetate inhibited leukotriene B₄ formation catalyzed by the human recombinant enzyme (IC₅₀ = 0.3 ± 0.17 µM) more than primin (IC₅₀ = 1.4 ± 0.6 µM) and zileuton (IC₅₀ = 1.1 ± 0.7 µM). Miconidin acetate (20 µM) inhibited LTB₄ formation to an extent of 59 ± 12 % in vitro using a cell-based assay, comparable to the positive control zileuton (69 ± 12 % inhibition at a concentration of 10 µM). The binding modes of miconidin acetate were further evaluated in silico by molecular docking to the human 5-lipoxygenase crystal structure. The hydroxyl group was predicted to form a hydrogen bond with the terminal Ile676, while the pentyl moiety occupied the hydrophobic substrate channel. The obtained results show that flower buds of E. hiemalis are an interesting source of anti-inflammatory compounds, mainly of miconidin acetate.

Key words

Eugenia hiemalis - Myrtaceae - miconidin acetate - primin - anti-inflammatory - 5-lipoxygenase

Eugenia is the largest genus in the Myrtaceae family from tropical America [1], growing from Mexico and the Caribbean to Argentina [2]. The plants of this genus are evergreen trees or shrubs with usually edible fruits and are known to be rich in volatile oils [3], [4]. Several species are used in folk medicine as anti-inflammatory [5], diuretic, digestive [6], and antimicrobial remedies [7].

Eugenia hiemalis Cambess., commonly known as “guamirim”, is a tree that grows in Brazil, Argentina, Uruguay, and Paraguay [8]. There are a few reports on its chemical composition and biological activity. Nevertheless, no report of its traditional use was found for this species. A benzoquinone (primin) and a monoacetyl derivative of the correspondent hydroquinone were isolated from the extract of leaves collected in Southern Brazil [9]. Both compounds showed antiapoptotic activity towards KG1a cells [10]. Moreover, three galloyl arbutins (hyemalosides A–C) along with nine phenolic compounds were isolated from the leavesʼ extract from plants growing in Paraguay. Some of these compounds inhibited HIV-1 RNase H in vitro [11]. The essential oil from the leaves contained sesquiterpene hydrocarbons as the major components [12], [13].

Natural products have long been used in folk medicine for the treatment of inflammatory conditions such as fevers, pain, and arthritis. These bioactive plant-derived substances represent a wide structural diversity. Some examples include curcumin, resveratrol, baicalein, boswellic acid, betulinic acid, ursolic acid, and oleanolic acid, which act on several inflammatory targets like cyclooxygenases (COX-1 and COX-2), 5-lipoxygenase (5-LOX), cytokines, chemokines, or interleukins [14].

Previous work evaluating primin detected a strong selective inhibition of COX-2 over COX-1 [15]. Furthermore, primin was similarly as potent as the 5-LOX inhibitor and reference compound zileuton, which is clinically used for asthma treatment. Docking analysis showed that the benzoquinone structure was essential for this activity [16]. This paper describes the isolation of miconidin acetate (MA) (1) and primin (2) ([Fig. 1]) by chromatographic fractionation of the flower bud extract from E. hiemalis, their anti-inflammatory activities via 5-LOX inhibition, and a docking analysis of MA binding to the target. Both compounds were identified and confirmed by spectral data and comparison to the literature. MA was the major metabolite and constitutes a hydroquinone derivative related to primin.

Fig. 1 Chemical structures of MA (1) and primin (2).

In the search for protein targets mediating a possible anti-inflammatory effect of MA, a similarity ensemble approach (SEA) was pursued using the publicly available SEA search tool [17], [18]. The structure of MA was uploaded as smiles code. MA was then compared to structurally similar molecule ensembles from the ChEMBL16 binding database. If the similarity was high, MA was predicted to act on the same biological targets at the same concentration as the respective molecules from the ChEMBL. We considered all predictions with E-values ≤ − 4, as suggested by Lounkine et al. [19]. Among the 54 predicted targets for MA, the 8th-ranked target, rat 5-lipoxygenase (5-LOX), was linked to inflammation. Overall, 5-LOX from different species was proposed five times as a target by the SEA tool.

MA inhibited leukotriene B₄ (LTB₄) formation in the human recombinant enzyme (IC₅₀ = 0.3 ± 0.17 µM) more potently than primin (IC₅₀ = 1.4 ± 0.6 µM) and zileuton (IC₅₀ = 1.1 ± 0.7 µM). Primin previously inhibited LTB₄ biosynthesis in a cell-based assay with an IC₅₀ = 4.0 µM (IC₅₀ of zileuton was 4.1 µM) [16]. Therefore, MA was also evaluated for 5-LOX inhibition in vitro using a cell-based assay and detecting LTB₄ with an ELISA kit [20]. MA (20 µM) inhibited LTB₄ formation to an extent of 59 ± 12 %, while the positive control zileuton (10 µM) inhibited 5-LOX by 69 ± 12 %. The inhibitory activity in the cell-free assay demonstrated direct interferences of MA and primin with 5-LOXʼs catalytic activity. In the case of cell-based assays, tested compounds can actually interact with several targets within the 5-LOX activation and reaction cascade [21]. In our case, the activity of MA in the cell-based assay demonstrated mainly its ability to enter the cells.

To further elucidate the mode of inhibition, MA was docked into an X-ray crystal structure of human 5-LOX (PDB entry 3o8 y) [22] using GOLD 5.2 (CCDC, GB). As shown in [Fig. 2], MA was predicted to occupy part of the active site going by the iron ion. It formed a hydrogen bond with the iron-coordinating residue Ile676 and hydrophobic contacts in the tunnel leading to the catalytic center.

Fig. 2 The substrate arachidonic acid docked into the active site of 5-LOX (a). Predicted binding pose and protein-ligand interactions of MA and its target 5-LOX (b). Chemical interactions are color-coded: yellow – hydrophobic contact; green – hydrogen bond donor. The surface of the binding site is displayed according to aggregated hydrophilicity (blue)/hydrophobicity (yellow).

Miconidin has been considered the biosynthetical precursor of primin [23], [24], and the isolation of MA as a major compound in the floral buds suggests that this metabolite is a storage form for primin, which could be involved in protection against microorganisms and herbivore attacks during the flowering stage. MA was identified chromatographically as a minor compound from Primula obconica Hance (Primulaceae) leaves [25]. Nevertheless, its isolation as a major natural product was just reported for E. hiemalis. Furthermore, no quinones or related compounds were reported so far for the Eugenia genus, except for E. hiemalis leaves [9] and flower buds.

Despite its relatively simple structure, MA and primin showed that they are potent 5-LOX inhibitors with similar activities compared to the reference inhibitor zileuton, both in cell-based and human recombinant enzyme-based LTB₄ formation assays. These results, together with the molecular docking study, suggest that the 3D structure of MA is important for its ability to access the catalytic center of 5-LOX.

Material and Methods

General experimental procedures: NMR spectra were obtained by Bruker DRX 400, Bruker AMX 500, and Varian XL 300 spectrometers for the ¹H NMR, ¹³C NMR, correlation spectroscopy (COSY), heteronuclear multiple-quantum correlation (HMQC), and heteronuclear multiple-bond correlation (HMBC). The mass spectrum was recorded by a spectrometer MS 50 (Kratos). IR spectra were obtained by a Prestige-21 Shimadzu, and UV-Vis spectra by Perkin Elmer Lambda 10.

Plant material: The aerial parts of E. hiemalis were collected and identified in April 2014 in Porto Alegre (Rio Grande do Sul, Brazil) by Dr. J. A. Jarenkow (Department of Botany, Federal University of Rio Grande do Sul, Brazil). A voucher specimen (ICN 127 910) was deposited in the herbarium of the Instituto de Ciências Naturais (Federal University of Rio Grande do Sul).

Extraction and isolation: The flower buds (27 g) were extracted with dichloromethane (0.25 L) by maceration for 7 days. The extract was concentrated under reduced pressure affording 0.5 g of CH₂Cl₂ extract that was fractionated in a gravity chromatography column on silica gel (40–63 µm, Sigma-Aldrich; 2.5 × 40 cm) with hexane-ethyl acetate (100/0 to 0/100, flow rate 1 mL/min) to give 1 (138.2 mg) and 2 (43.6 mg).

In vitro 5-lipoxygenase assay: Inhibition of 5-LOX was determined in two assays: a cell-based assay [20] and a cell-free assay using human recombinant enzyme (modified method of Albert et al.) [26]. Human neutrophil granulocytes for the cell-based assay were isolated from buffy coat (50 mL) obtained from healthy donors. Dextran solution was used for sedimentation, and subsequent lysis and washing were performed. Isolated cells were diluted to a final concentration of 4500 cells/µL. The incubation mixture consisted of 225 µL of cell suspension, 10 µL of 2 mM CaCl₂, 10 µL of 10 µM eicosatetraenoic acid, 5 µL of tested substances dissolved in DMSO, 10 µL of 21 µM calcium ionophor A23187, and 5 µL of 120 µM arachidonic acid. The reaction was stopped after 10 min incubation at 37 °C and the concentration of LTB₄ was measured using a commercial LTB₄ ELISA kit (Enzo Life Sciences) according to the manufacturerʼs instructions. Absorbance relative to the LTB₄ concentration was measured at 405 nm using a Tecan Infinite M200 (Tecan Group). The results are expressed as percentage of inhibition of LTB₄ formation against untreated samples (blanks).

Human recombinant 5-LOX (Cayman Chemical) was used for the cell-free assay. 5-LOX (1 unit/reaction) was added to 180 µL of incubation mixture consisting of phosphate buffer saline (pH 7.4), 1 mM of Na₂EDTA, and 200 of µM ATP. After the addition of the test substances (10 µL) dissolved in DMSO (or pure DMSO in case of blank), the mixture was incubated for 10 min at 4 °C. Then 5 µL of 80 mM CaCl₂ and 5 µL of 800 µM arachidonic acid were added and the mixture was incubated for 10 min at 37 °C. The reaction was terminated by the addition of 10 µL of 10 % formic acid. All samples were diluted 1 : 15 in assay buffer and the main product of reaction, LTB₄, was quantified using an LTB₄ ELISA kit (Enzo Life Sciences) according to the manufacturerʼs instructions. At least four concentrations were used for the calculations of the IC₅₀ values. Three independent experiments with at least two replicates were used for the calculations of the inhibition curves. IC₅₀ values were determined by regression analyses using Microsoft Excel.

Docking analysis: A putative binding mode for MA was calculated employing molecular docking using GOLD 5.2 software (www.ccdc.cam.ac.uk). Before docking, MA was geometrically optimized using Biovia DiscoveryStudio (www.biovia.com). The binding site was defined in a 10-Å radius around the iron ion in the catalytically active center. For all other options, default settings were used.

Supporting information

Spectral and physicochemical data of MA (1) and primin (2), and also a graphic of the inhibition of LTB₄ formation in the human recombinant enzyme by these compounds, are available as Supporting Information.

Acknowledgements

G. A. Z. is thankful to CAPES for the MSc. fellowship. The authors thank Prof. Vassilios Roussis (University of Athens) for the NMR spectra of MA.

Conflict of Interest

The authors declare no conflict of interest.

• References

• 1 Govaerts R, Sobral M, Ashton P, Barrie F, Holst BK, Landrum LR, Matsumoto K, Mazine FF, Lughadha EN, Proneça C, Soares-Silva LH, Wilson PG, Lucas E. World checklist of Myrtaceae. London: Kew Royal Botanic Gardens Publishing; 2008
• 2 Landrum LR, Kawasaki ML. The genera of Myrtaceae in Brazil: an illustrated synoptic treatment and identification Keys. Brittonia 1997; 49: 508-536
• 3 Flores G, Dastmalchi K, Paulino S, Whalen K, Dabo AJ, Reynertson KA, Foronjy RF, DʼArmiento JM, Kennelly EJ. Anthocyanins from Eugenia brasiliensis edible fruits as potential therapeutics for COPD treatment. Food Chem 2012; 134: 1256-1262
• 4 Cole RA, Haber WA, Setzer AN. Chemical composition of essential oils of seven species of Eugenia from Monteverde, Costa Rica. Biochem Syst Ecol 2007; 35: 877-886
• 5 Slowing K, Carretero E, Villar A. Anti-inflammatory activity of leaf extracts of Eugenia jambos in rats. J Ethnopharmacol 1994; 43: 9-11
• 6 Ferro E, Schinini A, Maldonado M, Rosner J, Hirschmann GS. Eugenia uniflora leaf extract and lipid metabolism in Cebus apella monkeys. J Ethnopharmacol 1988; 24: 321-325
• 7 Souza GC, Haas APS, von Poser GL, Schapoval EES, Elisabetsky E. Ethnopharmacological studies of antimicrobial remedies in the south of Brazil. J Ethnopharmacol 2004; 90: 135-143
• 8 Kawasaki ML. Flora da Serra do Cipó, Minas Gerais, Myrtaceae. Bolm Botânica Univ S Paulo 1989; 11: 121-170
• 9 Falkenberg MB. Quinones and other metabolites from Eugenia hiemalis Cambessèdes and Paramyrciaria glazioviana (Kiaerskou) Sobral (Myrtaceae) [thesis]. Bonn: University of Bonn; 1996
• 10 Efferth T, Rücker G, Falkenberg M, Manns D, Olbrich A, Fabry U, Osieka R. Detection of apoptosis in KG-1a leukemic cells treated with investigational drugs. Arzneimittelforschung 1996; 46: 196-200
• 11 Bokesch HR, Wamiru A, Le Grice SFJ, Beutler JA, McKee TC, McMahon JB. HIV-1 ribonuclease H inhibitory phenolic glycosides from Eugenia hyemalis . J Nat Prod 2008; 71: 1634-1636
• 12 Zatelli GA, Zimath P, Tenfen A, Cordova CMM, Scharf DR, Simionatto EL, Alberton MD, Falkenberg M. Antimycoplasmic activity and seasonal variation of essential oil of Eugenia hiemalis Cambess. (Myrtaceae). Nat Prod Res DOI: 10.1080/14786419.2015.1091455. advance online publication 1 October 2015
• 13 Apel MA, Sobral M, Schapoval ES, Henriques AT, Menut C, Bessiere JM. Chemical composition of the essential oils of Eugenia hyemalis and Eugenia stigmatosa. Part VI: section biflore. J Essent Oil Res 2004; 16: 437-439
• 14 Gautam R, Jachak SM. Recent developments in anti-inflammatory natural products. Med Res Rev 2009; 29: 767-820
• 15 Landa P, Kutil Z, Temml V, Vuorinen A, Malik J, Dvorakova M, Marsik P, Kokoska L, Pribylova M, Schuster D, Vanek T. Redox and non-redox mechanism of in vitro cyclooxygenase inhibition by natural quinones. Planta Med 2012; 78: 326-333
• 16 Landa P, Kutil Z, Temml V, Malik J, Kokoska L, Widowitz U, Pribylova M, Dvorakova M, Marsik P, Schuster D, Bauer R, Vanek T. Inhibition of in vitro leukotriene B4 biosynthesis in human neutrophil granulocytes and docking studies of natural quinones. Nat Prod Commun 2013; 8: 105-108
• 17 Keiser MJ, Roth BL, Armbruster BN, Ernsberger P, Irwin JJ, Shoichet BK. Relating protein pharmacology by ligand chemistry. Nat Biotechnol 2007; 25: 197-206
• 18 Keiser MJ, Roth BL, Armbruster BN, Ernsberger P, Irwin JJ, Shoichet BK. SEA. Available at. http://sea.bkslab.org Accessed February 17, 2016
• 19 Lounkine E, Keiser MJ, Whitebread S, Mikhailov D, Hamon J, Jenkins JL, Lavan P, Weber E, Doak AK, Côté S, Shoichet BK, Urban L. Large-scale prediction and testing of drug activity on side-effect targets. Nature 2012; 486: 361-367
• 20 Adams M, Kunert O, Haslinger E, Bauer R. Inhibition of leukotriene biosynthesis by quinolone alkaloids from the fruits of Evodia rutaecarpa . Planta Med 2004; 70: 904-908
• 21 Werz O, Steinhilber D. Development of 5-lipoxygenase inhibitors-lessons from cellular enzyme regulation. Biochem Pharmacol 2005; 70: 327-333
• 22 Gilbert NC, Bartlett SG, Waight MT, Neau DB, Boeglin WE, Brash AR, Newocomer ME. The structure of human 5-lipoxygenase. Science 2011; 14: 217-219
• 23 Connolly M, Cune JM, Dauncey E, Lovell CR. Primula obconica – is contact allergy on the decline?. Contact Dermatitis 2004; 51: 167-171
• 24 Krebs M, Christensen LP. 2-methoxy-6-pentyl-1,4-dihydroxybenzene (miconidin) from Primula obconica: a possible allergen?. Contact Dermatitis 1995; 33: 90-93
• 25 Horper W, Marner F. Phenols and quinones from leaves of Primula obconica . Nat Prod Lett 1995; 6: 163-170
• 26 Albert D, Zündorf I, Dingermann T, Müller WE, Steinhilber D, Werz O. Hyperforin is a dual inhibitor of cyclooxygenase-1 and 5-lipoxygenase. Biochem Pharmacol 2002; 64: 1767-1775