CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 221 Bioactive Constituents from Turkish Pimpinella Species by Nurhayat Tabanca a ) b ), Erdal Bedir a ) c ), Daneel Ferreira a ) d ), Desmond Slade a ), David E. Wedge e ), Melissa R. Jacob a ), Shabana I. Khan a ), Nese Kirimer b ), K. Husnu Can Baser b ), and Ikhlas A. Khan* a ) d ) a ) National Center for Natural Products Research, Research Institute of Pharmaceutical Sciences, School of Pharmacy, The University of Mississippi, University, MS 38677, USA (phone: ‡ 1-662-915-7821; fax: ‡ 1-662-915-7062; e-mail: ikhan@olemiss.edu) b ) Department of Pharmacognosy, Faculty of Pharmacy, Anadolu University, TK-26470, Eskisehir, Turkey c ) Department of Bioengineering, Faculty of Engineering, Ege University, TK-35100, Bornova, Izmir, Turkey d ) Department of Pharmacognosy, School of Pharmacy, The University of Mississippi, University, MS 38677, USA e ) USDA-ARS-NPURU, The University of Mississippi, University, MS 38677, USA A new phenylpropanoid, 4-(3-methyloxiran-2-yl)phenyl 2-methylbutanoate (1), a new trinorsesquiterpene, 4-(6-methylbicyclo[4.1.0]hept-2-en-7-yl)butan-2-one (2), and eight known compounds (3 ± 10) were isolated from the essential oils of several Pimpinella species growing in Turkey. The structures of the new compounds were determined by 1D- and 2D-NMR analyses. The absolute configuration of 1 was established via comparison of its optical rotation with that of epoxypseudoisoeugenyl 2-methylbutyrate (11), the absolute configuration of which was determined by chemical degradation and an appropriate Mosher ester formation. Direct bioautography revealed antifungal activity of 1 and 11 against Colletotrichum acutatum, C. fragariae, and C. gloesporioides. Subsequent evaluation of antifungal compounds in a 96-well microtiter assay showed that compounds 1 and 11 produced the most-significant growth inhibition in Phomopsis spp., Colletotrichum spp., and Botrytis cinerea. Compounds 1 and 6 displayed antimicrobial activities against Mycobacterium intracellulare, with IC50 values of 2.78 and 1.29 mm, respectively. Introduction. ± Due to increasing microbial resistance in medicine and agriculture, discovery of new antimicrobial substances is an important research objective. Particularly desirable is the discovery of antimicrobial agents with novel modes of action. In addition, environmentally safe agrochemicals with low mammalian toxicities are a major public concern. Natural-product leads offer an efficient approach to discovering and optimizing new pharmaceuticals and agrochemicals for disease control. Therefore, we began a study to evaluate several natural products from Pimpinella for potential use as disease-control agents. The widely known Pimpinella (Umbelliferae) species P. anisum has been used as a popular aromatic herb and spice since ancient times. It is cultivated in Turkey, Asia, South Africa, Europe, and America. Its fruit has been used in cooking and for medicinal purposes, e.g., as a carminative, antispasmodic, antiseptic, expectorant, stomachic, and for treatment of gastrointestinal disturbances, bronchial asthma, insomnia, and persistent cough. Aniseed fruit oil is used in the pharmaceutical industry to mask odors and flavors [1 ± 6]. P. major and P. saxifraga, known as Pimpinellae Radix, are used as an antitussive and mild expectorant in bronchitis, as well as for infections of the upper respiratory tract [7]. Biological activities of Pimpinella species are mainly attributed to phenylpropanoid derivatives, and (E)-pseudoisoeugenol  2005 Verlag Helvetica Chimica Acta AG, Zürich 222 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) (ˆ 2-methoxy-4-(prop-1-enyl)phenol) has been found only in the Pimpinella genus [8 ± 10]. Besides these characteristic phenylpropanoids, a number of C12 trinorsesquiterpenes, such as geijerene (ˆ 4-ethenyl-4-methyl-3-(1-methylethenyl)cyclohexene) and azulene, occur in considerable amounts in Pimpinella oils. The occurrence of phenylpropanoids and trinorsesquiterpenes are significant since the genus Pimpinella can be clearly separated from all the other Umbelliferae genera investigated so far [11 ± 15]. Here, we describe the isolation and characterization of compounds 1 ± 10 from Pimpinella species: the new phenylpropanoid 1 (from P. saxifraga), the four known phenylpropanoids 5 ± 8 (from P. peucedanifolia, P. aurea, P. rhodantha, and P. anisum, respectively), one new (2) and one known (3) trinor-type sesquiterpene (from P. tragium ssp. lithophila), the two known guaiane-type sesquiterpenes 4 and 9 (from P. peucedanifolia), and the known caryophyllene-type sesquiterpene 10 (from P. kotschyana). The antifungal and antimicrobial activities of some of these compounds were evaluated. Results and Discussion. ± Phytochemical studies of Pimpinella oils resulted in the isolation of compounds 1 ± 10. The known compounds 3 ± 10 were identified by comparison of their spectroscopic data with reported values [16 ± 23]. The formula of the new compound 1 was deduced to be C14H18O3 from HR-ESI-MS (m/z 235.1347 ([M ‡ H]‡ )) and 1H- and 13C-NMR data. From the 1H- and 13C-NMR spectra, a disubstituted aromatic ring was identified (AA'XX' system: d(H) 7.03 (d, J ˆ 8.5 Hz, 2 H), 7.26 (d, J ˆ 8.5 Hz, 2 H); d(C) 121.5 (d, 2 C), 126.4 (d, 2 C), 135.0 (s), CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 223 150.4 (s)). In addition, the 1H- and 13C-NMR spectra of 1 showed characteristic signals of a 2-methylbutanoate moiety (d(H) 2.62 (ddd, J ˆ 14.0, 7.0, 7.0 Hz, 1 H), 1.62 (ddd, J ˆ 7.0, 7.0, 6.5 Hz, 2 H), 1.28 (d, J ˆ 7.0 Hz, 3 H), 1.01 (m, 3 H); d(C) 12.0 (q), 16.9 (q), 27.1 (t), 41.4 (d), 174.8 (s)), together with those of a 3-methyloxiran-2-yl unit (d(H) 3.57 (d, J ˆ 2.0 Hz, 1 H), 3.00 (m, 1 H), 1.44 (d, J ˆ 5.0 Hz, 3 H); d(C) 59.1 (d), 59.2 (d), 18.2 (q)). The full assignment of the 1H- and 13C-NMR signals of 1 was secured by G-DQFCOSY, G-HMQC, and G-HMBC experiments (Fig. 1, a), and by comparison with the NMR data of phenylpropanoid derivatives previously isolated from Pimpinella species [7] [11]. From these data, the structure of 1 was identified as trans-4-(3-methyloxiranyl2-yl)phenyl 2-methylbutanoate. Compound 1 and the known epoxyisoeugenol derivative 11 (see below) had almost identical optical rotations, i.e., [a] 25 D ˆ ‡ 26.6 vs. ‡ 26.0 (c ˆ 0.9 vs. 1.0, CHCl3 ), respectively. It is, thus, reasonable to assume that these two compounds possess identical absolute configurations at their three common stereogenic centers. In view of the absence of configurational data for epoxypseudoisoeugenol derivatives, we embarked on assigning the absolute configuration of this class of compounds based on the readily available epoxypseudoisoeugenyl 2-methylbutyrate (11) [10]. Fig. 1. a) Selected HMBC (C ! H) correlations of compound 1. b) Major spin system (from DQF-COSY experiments), selected HMQC correlations, and key NOE contacts of compound 2 As shown in the Scheme, hydrolysis of 11 in refluxing MeOH containing NaOH afforded 2-(2-hydroxy-1-methoxypropyl)-4-methoxyphenol (12) and (‡)-(2S)-methylbutanoic acid (13). The latter compound was identified by comparison of its optical rotation ([a] 25 D ˆ ‡ 19.1 (c ˆ 0.6, CHCl3 )) with that of a commercially available sample. The substituted phenol 12 was not isolated, but first methylated to 14. The latter was then transformed into the (R)- and (S)-Mosher esters 15r and 15s, respectively, by treatment of two portions of ca 3.0 mg each with ( )-(R)- and (‡)-(S)-3,3,3-trifluoro2-methoxy-2-phenylpropanoyl chloride, respectively [24]. Comparison of the 1H-NMR data of 15r and 15s indicated shielding of the Me(3) doublet in 15r (d(H) 1.20) compared to 15s (d(H) 1.30); and in the latter, the H C(1) doublet was shielded (d(H) 4.80) relative to that of 15r (d(H) 4.90). These chemical-shift differences confirmed the (2R)-configuration of 14 (see the Scheme). Since cleavage of the trans-oxiranyl moiety under alkaline conditions must necessarily occur via an SN2 mechanism in an anti fashion under inversion of configuration at C(1'') in 11, the absolute configuration at C(1), in 14 is (S). Based on these considerations, compound 11 possesses a (1''R,2''R)- 224 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) Scheme. Derivatization of Compound 11 for Stereochemical Assignment oxiranyl moiety. Hence, the absolute configuration of both 11 and the new phenylpropanoid analog 1 should be (1''R,2S,2''R). The molecular formula of 2 was determined as C12H18O by HR-ESI-MS (m/z 179.1505 ([M ‡ H]‡ )). Inspection of the 1H-NMR data of 2 showed two Me singlets at d(H) 1.08 and 2.13, the latter arising from an Ac group. Two resonances in the upfield region suggested a cyclopropanyl ring (d(H) 1.01 (m, 1 H), 0.62 (dd, J ˆ 5.0, 4.5 Hz, 1 H)). Additionally, a disubstituted CˆC bond (d(H) 5.34 (m) and 6.00 (m)) and four CH2 groups (d(H) 1.80, 1.40, 1.82, 1.94, 1.59, 1.67 (6m, 6 H); 2.49 (t, J ˆ 7.5 Hz, 2 H) were observed. In the 13C-NMR spectrum of 2, the resonances for the CˆC and CˆO groups (d(C) 121.7 (d), 128.9 (d), and 208.3 (s), respectively) supported the 1H-NMR evidence. Of the four degrees of unsaturation indicated by the molecular formula, two were attributed to one disubstituted CˆC bond and one CˆO group, the remaining degrees of unsaturation indicating the molecule to be bicyclic. The combined use of G-DQF-COSY and G-HMQC spectra of 2 permitted the assignment of one major spin system between CH2(3) and CH2(5') (see Fig. 1, b). To correlate the remaining fragments (an Ac and a Me group) with the above spin system, gradient long-range 1H,13C-NMR and G-HMBC experiments were performed. The long-range correlations revealed the molecule to be a carabrane-type norsesquiterpene derivative [25] [26]. The relative configuration of 2 was resolved by NOESY experiments. The strong cross-peaks observed between H C(1') and Me C(6') implied a syn (b) relationship, and the extremely weak NOESY signals between H C(7') and both H C(1') and Me C(6') revealed that H (7') was on the a-face of the molecule. Comparison of the 18.8 (c ˆ 1.0, CHCl3 )), with those of carabrane-type optical rotation of 2 ([a] 25 D ˆ sesquiterpenoids that possess identical stereocenters [25 ± 27] further confirmed these configurational assignments ((1'S,6'R,7'S)). On the basis of all this evidence, the structure of 2 was deduced to be 4-(6-methylbicyclo[4.1.0]hept-2-en-7-yl)butan-2-one (traginone). The following known compounds were isolated from Pimpinella species: dictamnol (3) from P. tragium ssp. lithophila; 4,6-guaiadien (4), 4-(1-propenyl)phenylisobutyrate (5), and alismol (9) from P. peucedanifolia; 4-(3-methyloxiran-2-yl)phenyltiglate (6) CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 225 from P. aurea; 2-methoxy-4-(1-propenyl)phenyltiglate (7) from P. rhodantha; pseudisoeugenol 2-methylbutyrate (8) from P. anisum; and 12-hydroxy-b-caryophyllene acetate (10) from P. kotschyana (Table 1). These compounds were identified by comparison of their 1H- and 13C-NMR data with literature values [16 ± 23]. For 9 and 10, the MS values previously reported [21 ± 23] were confirmed, while those for 3 ± 8 are reported for the first time. As far as we know, compounds 3 ± 5 and 9 have not been found before in Pimpinella. Table 1. Investigated Pimpinella Plant Materials. Abbreviations: R, root; A, aerial parts without fruit; F, fruit. Species Collection site Collection date Plant part Yield [%] a ) No. b ) P. saxifraga l. P. tragium Vill. ssp. lithophila (Schischkin) Tutin P. peucedanifolia Fischer ex Ledeb P. aurea dc. P. rhodantha Boiss. P. anisum l. P. kotschyana Boiss. Kars: Sarikamis Eskisehir: Turkmen Mountain Mus: Solhan July 2001 June 2000 0.17 0.22 0.17 0.21 13924 13872 July 2001 R A R F Van: Ercis Giresun: Tamdere Herbalist shop Konya: Eregli July 2001 August 2001 August 2001 July 2001 F R F F 5.05 0.10 2.49 1.02 13912 13932 14128 13886 13899 a ) Weight percent of isolated essential oils calculated on a moisture-free basis. b ) Registration number of sample deposited at the Herbarium of the Faculty of Pharmacy, Anadolu University, Eskisehir, Turkey ( ESSE ). Direct bioautography via thin-layer chromatography (TLC) on silica gel revealed antifungal activities of 1 and 11 against Collectotrichum acutatum, C. fragariae, and C. gloesporioides (Table 2). The occurrence of antifungal compounds was evidenced by the presence of clear zones (where fungal mycelical or reproductive stroma were absent) with a dark background on the TLC plate. This is the first report of the antifungal activity of 1 and 11 against C. acutatum, C. fragariae, and C. gloesporioides. Table 2. Autobiographically Determined Mean Inhibitory Zones (in mm) of Selected Colletotrichum Species upon Exposure to Compounds 1 and 11 (at two different loadings). The inhibitory zones were defined as clear zones free of any fungal growth or reproductive stroma. The data are compared with those of the commercial fungicidal standards Benomyl, Captan, Cypodinil, and Azoxystobin, which were applied as pure compounds in EtOH. The natural products 2 and 3 were not active in these tests. Compound 1 11 Benomyl Captan Cyprodinil Azoxystobin C. acutatum C. fragariae C. gloeosporioides 2 mg 4 mg 2 mg 4 mg 2 mg 4 mg 80 80 19.7  0.7 14.7  0.7 30.3  0 24.8  0.7 11  1.4 11  1.4 ± ± ± ± 80 10  0 19.7  0.7 14.7  0.7 30.8  0.7 20.7  0.7 15.5  0.7 12.5  0.7 ± ± ± ± 80 7  1.4 20.2  0 9.61  0.69 30.3  0 30.3  0 11.5  0.7 10  0 ± ± ± ± A 96-well microtiter assay with a dose-response format showed that 1 and 11 were fungicidal against Phomopsis spp., Colletotrichum spp. and Botrytis cinerea. Fusarium oxysporium showed a high level of chemical resistance to all compounds tested. Compounds 2 and 3 showed weak activity against all fungi tested. Compound 11 226 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) appeared to be the most active drug toward species with a therapeutic threshold of ca. 3.0 mm for Phomopsis and Colletotrichum fragariae (Fig. 2). At 0.3 mm, 11 was the most active and caused 94.5% growth inhibition of P. obscurans (Fig. 2, a), but was less active against P. viticola (61.3% inhibition, not shown). Compound 11 was more effective at reducing growth in P. obscurans than the commercial fungicides Azoxystrobin, Cyporodinil, or Captan. At 3.0 mm, 11 was also the most active compound against C. fragariae (96.7% growth inhibition; Fig. 2, b). At 30.0 mm, 11 showed weak antifungal activity and effected 40.0% growth inhibition in B. cinerea, 54.1% in C. acutatum, and 73.7% in C. gloeosporioides after 48 h each (Figs. 3 and 4, a). Fig. 2. Mean growth inhibition [%] of a) Phomopsis obscurans (after 120 h) and b) Colletotrichum fragariae (after 48 h) upon exposure to 1 ± 3 and 11 Compound 1 appeared to be the second most-active drug across fungal species with a therapeutic threshold of ca. 30.0 mm in Phomopsis, Botrytis, and Colletotrichum species. At 30.0 mm, 1 showed 100% growth inhibition of all test organisms (Figs. 2, 3, and 4, b) except Fusarium oxysporium (65.1%). Compound 1 was less effective than Azoxystrobin, Cyporodinil, or Captan at reducing growth of most fungi in these experiments. In contrast to 1 and 11, the trinorsesquiterpenes 2 and 3 exhibited CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 227 Fig. 3. Mean growth inhibition [%] of a) Colletotrichum acutatum and b) C. gloeosporioides upon 48-h exposure to 1 ± 3 and 11 negligible antifungal activities. Our results, thus, indicate that phenylpropanoids may serve as useful natural-product leads with potential application in plant protection and fungal-disease control. Compounds 1 ± 3, 5, 6, and 8 ± 10 were evaluated for their antimalarial and antimicrobial activities (Table 3). None of the compounds showed antimalarial activity against Plasmodium falciparum D6 and W2 clones. In the antimicrobial screening, the tested compounds were inactive against Candida albicans, Cryptococcus neoformans, Staphylococcus aureus, methicillin-resistant S. aureus, Pseudomonas aeruginosa, and Aspergillus fumigatus. However, 1 and 6 showed potent activity against Mycobacteria intracellulare, with IC50 values of 2.78 and 1.29 mm, respectively; they were also tested against other Mycobacteria species (Table 3). From compounds 1 and 6, a new class of antimycobacterial agents may, thus, be developed. The antimycobacterial activity of 11 has already been reported [10]; it was highly active against M. intracellulare, with an IC50 value of 5.68 mm. Owing to the benzylic nature of the epoxides 1, 6, and 11 C(1'')- 228 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) Fig. 4. Mean growth inhibition [%] of Botrytis cinerea upon exposure to 1 ± 3 and 11, as determined after 48 h (a) and 72 h (b), respectively. atom, these compounds should be powerful electrophiles that may react with nucleophilic centers of the biological system of the affected microorganisms, which would rationalize the observed antimycobacterial activities. Table 3. Antimycobacterial Activities of Compounds 1 and 6 Relative to the Standard Drug Ciprofloxacin. Compounds 2, 3, 5, and 8 ± 10 were also tested, but turned out to be hardly active (> 50% growth at maximum concentration). All compounds were tested up to a maximum concentration of 86 mm, except Ciprofloxacin (15 mm). Compound 1 6 Ciprofloxacin M. intracellulare M. smegmatis M. aurum IC50 a ) IC50 MIC IC50 MIC IC50 MIC n.a. n.a. 1.89 2.99 1.29 0.10 42.73 21.55 0.47 4.27 2.16 0.75 21.37 10.78 ± 2.78 1.29 0.75 MIC b ) 10.68 5.39 3.78 c n.a. ) n.a. 0.91 a ) Concentration affording 50% growth inhibition. highest test concentration. b M. phlei ) Minimum inhibitory concentration. c ) Not active at CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 229 Experimental Part 1. General. Thin-layer chromatography (TLC): Merck silica gel GF 254 (precoated plates); hexane/Et2O 90 : 10, 80 : 20, or 70 : 30, visualization by spraying with vanillin/H2SO4 . High-performance flash chromatography (HPFC; system from Biotage, Inc., Dyax Corp.): Biotage columns Si-12-M column (150  12 mm i.d.; 9 g KP-Sil silica gel (40 ± 63 mm); 2.0 ml/min) and Si-25-M (150  25 mm i.d.; 40 g KP-Sil silica gel (40 ± 63 mm); 2.0 ml/min). Optical rotation: Rudolph Research Analytical digital polarimeter; 1- and 5-cm cell tubes. UV Spectra: Hewlett-Packard 8452A spectrophotometer; lmax (log e) in nm. IR Spectra: Ati Mattson Genesis Series FTIR spectrometer; in cm 1. 1H- and 13C-NMR-Spectra: Bruker Avance DRX-500 (500 and 125 MHz, resp.) and Bruker DRX-300 (300 and 75 MHz, resp.) spectrometers; chemical shifts d in ppm rel. to SiMe4 (ˆ 0 ppm) as internal standard, coupling constants J in Hz. GC/MS: HP 5890 gas chromatograph linked to an HP 5970 mass spectrometer (EI, 70 eV) equipped with an automatic injector and a DB-1 capillary column (20 m  0.18 mm i.d., 0.25-mm film thicknesses); 1 ml/min He carrier gas; GC oven temp.: 708 for 3 min, 70 ± 1808 at 58/min, then 2208 for 10 min, and 220 ± 2408 at 18/min; mass range: m/z 40 ± 550. ESI-FT-MS (Electrospray-ionization Fourier-transformation mass spectrometry): Bruker BioApex FT-MS, pos. mode; in m/z (rel. %). 2. Plant Material. The air-dried plants, collection sites, and yields of essential oils are given in Table 1. Voucher specimens have been deposited at the Herbarium of the Faculty of Pharmacy, Anadolu University in Eskisehir, Turkey (ESSE). 3. Extraction and Isolation. Fruits (F), aerial parts without fruits (A), and roots (R) of the different species and subspecies were crushed separately and then steam-distilled (Clevenger-type apparatus) for 3 h to afford the corresponding essential oils. The percentage yields of the oils calculated on a moisture-free basis are given in Table 1. The essential oils were analyzed with a HP G1800 GC/MSD system. After GC/MS analysis, the constituents 1 ± 10 were not identifiable in the Wiley GC/MS Library and the in-house Baser Library of Essential Constituents. The separation of these compounds was, thus, necessary (see below). The isolated compounds were re-analyzed by GC/MS to confirm their identity with the constituents of the essential oils. The following GC/MS retention times (tR , [min]) were found for 1 ± 10: 23.4, 14.9, 16.8, 17.8, 19.4, 25.1, 26.0, 24.5, 21.1, and 24.3, resp. 3.1. The essential oil of P. saxifraga (150 mg) obtained from roots was subjected to HPFC (Biotage SI-12-M; 1. hexane (60 ml), 2. hexane/Et2O 99 : 1 ! 90 : 10 ! 85 : 15 ! 80 : 20 ! 70 : 30, 30 ml each, 3. hexane/Et2O 60 : 40 (60 ml)) to yield 15 fractions (Fr. 1A ± 1O, 3 ml each). Fr. 1F (3.0 mg) was subjected to prep. RP-HPLC (Ultracarb 5-ODS-30; 250  10.0 mm; H2O/MeCN 90 : 10 ! 10 : 90 in 25 min, 2.5 ml/min; UV detection at 210 nm), which afforded compound 1 (1.6 mg). 3.2. The essential oil of P. tragium ssp. lithophila (200 mg) obtained from the aerial parts without fruits was subjected to HPFC (Biotage SI-25-M; 1. hexane (30 ml), 2. hexane/Et2O 98 : 2 ! 90 : 10 ! 80 : 20 ! 70 : 30, 30 ml each, 3. hexane/Et2O 60 : 40 (90 ml)) to give ten fractions (Fr. 2A ± 2J, 3 ml each). Fr. 2D and 2G gave 2 (5 mg) and 3 (20 mg), resp. 3.3. The essential oil of P. tragium ssp. lithophila (100 mg) obtained from roots was subjected to HPFC (Biotage SI-12-M; 1. hexane (30 ml), 2. hexane/Et2O 99 : 1 ! 90 : 10 ! 80 : 20 ! 70 : 30, 30 ml each, 3. hexane/ Et2O 60 : 40 (60 ml)) to yield 15 fractions: Fr. 3A ± 3O, 3 ml each. Fr. 3A gave 4 (1.0 mg). 3.4. The essential oil of P. peucedanifolia (100 mg) obtained from fruits was subjected to HPFC (Biotage SI12-M; 1. hexane (100 ml), 2. hexane/Et2O 99 : 1 ! 90 : 10, and 85 : 15 ! 80 : 20 ! 70 : 30, 30 ml each, 3. hexane/ Et2O 60 : 40 (60 ml)) to afford 14 fractions: Fr. 4A ± 4O (3 ml each). Fr. 4B gave 5 (5.0 mg). Fr. 4G (2.2 mg) was further purified by gel filtration (Sephadex LH-20, MeOH) to afford 9 (1.0 mg). 3.5. The essential oil of P. aurea (100 mg) obtained from fruits was subjected to HPFC (Biotage SI-12-M; 1. hexane (40 ml), 2. hexane/Et2O 99 : 1 ! 90 : 10 ! 80 : 20 ! 70 : 30, 30 ml each, 3. hexane/Et2O 60 : 40 (60 ml)) to give 15 fractions: Fr. 5A ± 5O (3 ml each). Fr. 5D afforded 6 (1.0 mg). 3.6. The essential oil of P. rhodantha (50 mg) obtained from roots was subjected to HPFC (Biotage Si-12-M; 1. hexane (20 ml), hexane/Et2O 99 : 1 ! 90 : 10 ! 88 : 12 ! 85 : 15 ! 80 : 20 ! 70 : 30, 30 ml each, 3. hexane/Et2O 60 : 40 (60 ml)) to yield 16 fractions: Fr. 6A ± 6P (3 ml each). Fr. 6F afforded 7 (1.0 mg). 3.7. The essential oil of P. anisum (200 mg) obtained from fruits was subjected to HPFC (Biotage Si-25-M; 1. hexane (70 ml), hexane/Et2O 99 : 1 ! 90 : 10 ! 80 : 20 ! 70 : 30 (30 ml each), 3. hexane/Et2O 60 : 40 (60 ml)) to yield 14 fractions: Fr. 7A ± 7N (3 ml each). Fr. 7E gave 8 (3.0 mg). 3.8. The essential oil of P. kotchyana (220 mg) obtained from fruits was subjected to HPFC (Biotage SI-25M; 1. hexane (20 ml), 2. hexane/Et2O 99 : 1 ! 90 : 10 ! 88 : 12 ! 85 : 15 ! 80 : 20 ! 70 : 30 (30 ml each), 3. hexane/Et2O 60 : 40 (60 ml)) to yield 16 fractions: Fr. 8A ± 8P (3 ml each). Fr. 8D afforded 10 (5.0 mg). 230 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 4-[(2R,3R)-3-Methyloxiran-2-yl]phenyl (2S)-2-Methylbutanoate ((‡)-1). Colorless oil. [a] 25 D ˆ 26.6 (c ˆ 0.9, CHCl3 ). UV (EtOH): 240 (1.25). IR (KBr): 3412, 2075, 1643, 448. 1H-NMR (500 MHz, CDCl3 ) 1): 7.26 (d, J ˆ 8.5, H C(3'), H C(5')); 7.03 (d, J ˆ 8.5, H C(2'), H C(6')); 3.57 (d, J ˆ 2.0, H C(1'')); 3.00 (m, H C(2'')); 2.62 (ddd, J ˆ 14, 7.0, 7.0, H C(2)); 1.62 (ddd, J ˆ 7.0, 7.0, 6.5, CH2(3)); 1.44 (d, J ˆ 5.0, Me C(2'')); 1.28 (d, J ˆ 7.0, MevC(2)); 1.01 (m, Me(4)). 13C-NMR (125 MHz, CDCl3 )1): 174.8 (s, C(1)); 150.4 (s, C(1')); 135.0 (s, C(4')); 126.4 (d, C(3'), C(5v)); 121.5 (d, C(2'), C(6')); 59.1 (2d, C(1''), C(2'')); 41.4 (d, C(2)); 27.1 (t, C(3)); 18.2 (q, Me C(2'')); 16.9 (q, Me C(2)); 12.0 (q, C(4)). GC/MS: 234 (2, M‡ ), 150 (13), 149 (9), 133 (9), 121 (7), 107 (15), 106 (26), 105 (9), 85 (22), 84 (18), 78 (14), 77 (21), 57 (100), 51 (20), 43 (8), 41 (25). ESI-MS: 235.1347 (M‡, C14H18O ‡3 ; calc. 234.13). Hydrolysis of Epoxypseudoisoeugenol 2-Methylbutyrate (ˆ 4-Methoxy-2-[(2R,3R)-3-methyloxiran-2-yl]phenyl (2S)-2-Methylbutanoate; 11). Compound 11 (20.0 mg) was refluxed for 5 min in MeOH (10 ml) containing 1.25% of NaOH. After cooling to r.t., H2O (40 ml) was added, and this mixture was extracted with Et2O (3  50 ml). a) The org. layer was washed with H2O (3  25 ml) and dried (Na2SO4 ). The remaining residue (14.6 mg) was methylated with diazomethane (Diazald procedure), which afforded an oily mixture (12.2 mg). After purification by CC (1.0 g of SiO2 ; hexanes/AcOEt 8 : 2), 1-(2,5-dimethoxyphenyl)-1methoxypropan-2-ol (14; 6.5 mg) was obtained. b) The aq. layer was acidified with 5% aq. HCl and extracted with Et2O (3  50 ml). The resulting org. layer was washed with H2O (3  25 ml) and dried (Na2SO4 ) to afford (2S)-butanoic acid (13; 3.8 mg), which was identified by GC/MS, optical rotation, and comparison with an authentic, commercial sample. Data of 14. 1H-NMR (500 MHz, (D5 )pyridine)1): 7.59 (br. s, H C(6')); 6.96 (2m, H C(3'), H C(4')); 4.86 (d, J ˆ 6.5, H C(1)); 4.24 (m, J ˆ 13, 6.5, H C(2)); 3.72, 3.70, 3.29 (3s, 3 MeO); 1.32 (d, J ˆ 6.5, Me(3)). GC/ MS: 226 (10, M‡, C12H18O ‡4 ), 181 (100), 182 (19), 167 (7), 151 (23), 137 (5), 121 (12), 108 (4), 91 (6), 77 (7), 65 (5), 45 (71), 29 (5). ‡ ‡ Data of 13. Colorless oil. [a] 25 D ˆ ‡ 19.0 (c ˆ 0.6, CHCl3 ). GC/MS: 102 (M , C5H10O 2 ), 87 (23), 74 (100), 73 (14), 57 (43), 55 (15), 45 (23), 41 (56), 39 (21), 29 (47), 27 (41), 26 (7). Preparation of Mosher Esters 15. In an NMR tube, compound 14 (3.0 mg) in (D5 )pyridine was treated with (‡)-(S)-a-methoxy-a-(trifluoromethyl)phenylacetyl chloride (MTPA; 30 ml) 2 ). The tube was shaken carefully, and then left at r.t. to afford the (S)-MTPA derivative 15s. Similarly, a second portion of 14 (3.2 mg) was reacted with (R)-MTPA (30 ml) to afford 15r (see the Scheme). 1H-NMR data for 15s and 15r were recorded from the reaction mixtures (in (D5 )pyridine). Then, they were further purified by CC (SiO2 ; hexane/AcOEt 8 : 2), affording pure 15s (2.5 mg) and 15r (2.8 mg), resp., which were analyzed in CDCl3 by NMR. 1 H-NMR Data of 15s. a) At 500 MHz in (D5 )pyridine (in situ; relevant signals only): 7.60 (d, J ˆ 6.5, H C(6')); 7.46 (2m, H C(3'), H C(4')); 5.78 (m, H C(2)); 4.81 (d, J ˆ 7.0, H C(1)); 3.71, 3.70, 3.46 (3s, 3 MeO); 1.30 (d, J ˆ 6.5, Me(3)). b) At 500 MHz in CDCl3 (after purification; relevant signals only): 7.46 (d, J ˆ 7.0, H C(6')); 7.30 (2m, H C(3'), H C(4')); 5.39 (m, H C(2)); 4.53 (d, J ˆ 7.0, H C(1)); 3.71, 3.65, 3.49 (3s, 3 MeO); 1.14 (d, J ˆ 6.5, Me(3)). 1 H-NMR Data of 15r. a) At 500 MHz in (D5 )pyridine (relevant signals only): 7.60 (d, J ˆ 6.5, H C(6')); 7.45 (2m, H C(3'), H C(4')); 5.72 (m, H C(2)); 4.88 (d, J ˆ 7.5, H C(1)); 3.70, 3.69, 3.45 (3s, 3 MeO); 1.21 (d, J ˆ 6.5, Me(3)). b) At 500 MHz in CDCl3 (relevant signals only) 7.48 (d, J ˆ 8.0, H C(6')); 7.32 (2m, H C(3'), H C(4')); 5.34 (m, H C(2)); 4.63 (d, J ˆ 8.0, H C(1)); 3.72, 3.63, 3.52 (3s, 3 MeO); 1.04 (d, J ˆ 6.5, Me(3)). 4-(6-Methylbicyclo[4.1.0]hept-2-en-7-yl)butan-2-one (2; Traginone). Colorless oil. [a] 25 18.8 (c ˆ 1.0, D ˆ CHCl3 ). UV (EtOH): 214 (0.93). IR (KBr): 3714, 3431, 1594, 998, 489. 1H-NMR (500 MHz, CDCl3 )1): 6.00 (m, H C(2')); 5.34 (m, H C(3')); 2.49 (t, J ˆ 7.5, CH2(3)); 2.13 (s, Me(1)); 1.94, 1.82 (2m, CH2(4')); 1.80, 1.40 (2m, CH2(5')); 1.67, 1.59 (2m, CH2(4)); 1.08 (s, Me C(6')); 1.01 (m, H C(7')); 0.62 (dd, J ˆ 4.5, 5.0, H C(1')). 13 C-NMR (125 MHz, CDCl3 ): 208.3 (s, C(2)); 128.9 (d, C(2'); 121.7 (d, C(3')); 43.9 (t, C(3)); 30.1 (q, C(1)); 27.4 (d, C(7')); 27.2 (t, C(5')); 25.1 (d, C(1')); 24.0 (s, C(6')); 23.5 (t, C(4)); 21.6 (t, C(4')); 20.0 (q, Me C(6')). ESIMS: 179.1505 (M‡, C12H18O‡ ; calc. 178.14). GC/MS: 178 (2, M‡ ), 161 (5), 145 (9), 131 (3), 120 (35), 119 (14), 118 (14), 107 (13), 105 (72), 93 (26), 92 (35), 91 (65), 79 (59), 77 (36), 65 (16), 55 (19), 53 (15), 43 (100), 41 (34). 1 13 Dictamnol (3). Colorless oil. [a] 25 C-NMR spectral data were D ˆ ‡ 53.3 (c ˆ 0.3, CHCl3 ). The H- and consistent with those reported [16]. GC/MS (C12H18O): 178 (2, M‡, C12H18O‡ ), 163 (6), 160 (15), 145 (27), 131 (19), 120 (32), 119 (15), 118 (18), 117 (26), 115 (14), 107 (16), 105 (63), 93 (32), 92 (37), 91 (75), 85 (3), 79 (540, 77 (21), 71 (39), 65 (21), 55 (19), 53 (21), 43 (100), 41 (53). 1) 2) Trivial atom numbering, see chemical formulae and the Scheme. Systematic name: (‡)-(R)-3,3,3-trifluoro-2-methyoxy-2-phenylpropanoyl chloride. CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 231 4,6-Guaiadien (4). Colorless oil. [a] 25 46.6 (c ˆ 0.6, CHCl3 ). The 1H- and 13C-NMR spectral data were D ˆ consistent with those reported [17]. GC/MS: 204 (59, M‡, C15H ‡24 ), 189 (58), 175 (5), 162 (14), 161 (79), 147 (27), 133 (39), 121 (16), 119 (68), 107 (32), 105 (93), 95 (19), 93 (35), 91 (88), 81 (32), 79 (40), 77 (45), 67 (23), 66 (27), 65 (23), 55 (46), 53 (28), 43 (37), 42 (20), 41 (100). 4-(1-Propenyl)phenyl Isobutyrate (5). Colorless oil. The 1H- and 13C-NMR spectral data were consistent with those reported [18]. GC/MS: 204 (7, M‡, C13H16O ‡2 ), 161 (2), 145 (10), 135 (12), 134 (100), 133 (41), 119 (5), 105 (12), 103 (12), 93 (11), 91 (11), 79 (14), 77 (17), 71 (11), 55 (11), 43 (56), 41 (26). 1 13 4-(3-Methyloxiranyl)phenyltiglate (6). Colorless oil. [a] 25 D ˆ ‡ 16.6 (c ˆ 0.5, CHCl3 ). The H- and C-NMR spectral data were consistent with those reported [19]. GC/MS: 232 (3, M‡, C14H16O ‡3 ), 107(14), 84 (6), 83 (100), 77 (7), 69 (9), 55 (75), 53 (7), 43 (25), 41 (8). 2-Methoxy-4-(1-propenyl)phenyltiglate (7). Colorless oil. The 1H- and 13C-NMR spectral data were consistent with those reported [19] [20]. GC/MS: 246 (5, M‡, C15H18O ‡3 ), 220 (6), 187 (2), 177 (6), 164 (22), 159 (8), 147 (7), 135 (6), 119 (6), 107 (11), 91 (17), 83 (100), 77 (14), 67 (12), 55 (95), 53 (13), 43 (23), 41 (23). 1 2-Methoxy-4-(1-propenyl)phenyltiglate (8). Colorless oil. [a] 25 D ˆ ‡ 22.0 (c ˆ 1.0, CHCl3 ). The H- and 13 C-NMR spectral data were consistent with those reported [19]. GC/MS [17]: 248 (5, M‡, C15H20O ‡3 ), 165 (12), 164 (100), 149 (23), 135 (3), 121 (3), 105 (5), 91 (8), 77 (6), 66 (5), 55 (3), 57 (32), 41 (14). 1 13 Alismol (9). Colorless oil. [a] 25 D ˆ ‡ 8.0 (c ˆ 0.5, CHCl3 ). The H-NMR, C-NMR, and mass spectral data were consistent with those reported [21] [22]. 12-Hydroxy-b-caryophyllene Acetate (10). Colorless oil. [a] 25 30.0 (c ˆ 0.8, CHCl3 ). The 1H-NMR, D ˆ 13 C-NMR, and mass spectral data were consistent with those reported [23]. Inoculum Preparation. Conidia were harvested from cultures 7 ± 10 days old by flooding plates with sterile distilled H2O (5 ml) and dislodging the conidia by softly brushing the colonies with an L-shaped glass rod. Conidial suspensions were filtered through sterile Miracloth (Calbiochem-Novabiochem Corp., La Jolla, CA) to remove mycelia. Conidia concentrations were determined photometrically [28] [29] from a standard curve based on the percent of transmittance at 625 nm. The suspensions were then adjusted with sterile distilled H2O to a concentration of 1.0  106 conidia/ml. Bioautography. The conidia of Colletotrichum fragariae, C. acutatum, and C. gloeosporioides suspensions were adjusted to a concentration of 3.0  105 conidia/ml with liquid potato-dextrose broth (PDB; Difco, Detroit, MI) and 0.1% Tween-80. With a 50-ml chromatographic sprayer, each glass TLC plate (Silica Gel GF Uniplate, Analtech, Inc., Newark, DE; with fluorescent indicator at 250 nm) was sprayed lightly (to a dampness) three times with the conidial suspension. Inoculated plates were placed in a moisture chamber (30  13  7.5 cm, model 398-C; Pioneer Plastics, Inc., Dixon, KY) and incubated in a growth chamber at 24  18 during a 12-h period with light (60  5 mmol ´ m 2 ´ s 1). Inhibition of fungal growth was measured 4 d after treatment. The sensitivity of each fungal species to the test compounds was determined by comparing the sizes of inhibitory zones. Means and standard deviations of inhibitory-zone sizes were used to evaluate antifungal activity of solvent fractions and pure compounds. Bioautography experiments were performed multiple times using both dose-response and non-dose-response formats. The fungicide technical standards Benomyl, Cyprodinil, Azoxystrobin, and Captan (Chem Service, Inc., West Chester, PA) were used as controls at 2 mg. Microtiter Assay. A standardized 96-well microtiter-plate assay developed by Wedge and Kuhajek [29] was used to evaluate antifungal activities of Pimpinella species towards Botrytis cinerea, Colletotrichum acutatum, C. fragariae, C. gloeosporioides, and Fusarium oxysporum. Captan, Azoxystrobin, and Cyprodinil, which represent three different modes of action, were used as commercial fungicide standards. Each fungus was challenged in a dose-response format with test compounds, where the final-treatment concentrations were 0.3, 3.0, and 30.0 mm. Untreated microtiter plates (Nunc MicroWell, Roskilde, Denmark) were covered with a plastic lid and incubated in a growth chamber as described above. Fungal growth was then evaluated by measuring the absorbance of each well at 620 nm using a microplate photometer (Packard Spectra Count, Packard Instrument Co., Downers Grove, IL). Each compound was evaluated in duplicate at three different doses (0.3, 3.0, and 30.0 mm). Sixteen wells with broth and inoculum served as positive controls, eight wells with solvent at the appropriate concentrations and broth without inoculum were used as negative controls. Mean absorbances and standard errors were used to evaluate fungal growth after 48 and 72 h, except for P. obscurans and P. viticola (120 h). The mean values (n ˆ 4) for concentration-dependent inhibition (%) of each fungus for each test compound rel. to untreated positive controls (n ˆ 32) were used to evaluate fungal-growth inhibition. Treatments were arranged as a split-plot design replicated twice in time. Antimicrobial and in vitro Antimalarial Assays. These assays were performed according to procedures reported previously [10]. 232 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) We are grateful to Prof. Dr. Z. Aytac (Faculty of Science & Letter, Department of Biology, Gazi University, Ankara, Turkey) for identification of the plant specimens and for help with collection of plant materials. We acknowledge Prof. Dr. B. Yildiz, Dr. A. Sokmen, Dr. E. Sonmez, Dr. M. Ekici, Dr. H. Akan, Research Assistant F. Karaveliogullari, Dr. S. Celik, Dr. T. Tugay, Research Assistant T. Arabaci, Dr. T. Degirmenci, Mr. I. Yaman, Mr. I. Koklu, Mr. I. Yilmaz, and Mr. A. Koruk for help with collection of plant materials. We thank Dr. B. Demirci and Dr. T. Ozek for GC/MS analyses of the essential oils, and Mr. F. Wiggers and Dr. C. Dunbar for recording NMR spectra, as well as Dr. Srinivas Pullelal for valuable asistance. We thank Jessie L. Robertson and Dewayne Harries for running antifungal assays. We also thank M. Wright and J. Trott for performing the antimicrobial and antimalarial tests. This research was supported by the Anadolu University Research Fund, Eskisehir, Turkey (AUAF 00 03 34) and, in part, by the United States Department of Agriculture (Agricultural Research Service Specific Cooperative Agreement No. 58-6408-7-012). REFERENCES [1] P. M. Santos, A. C. Figueriredo, M. M. Oliveira, J. G. Barroso, L. G. Pedro, S. G. Deans, A. K. M. Younus, J. J. C. Scheffer, Phytochemistry 1998, 48, 455. [2] P. M. Santos, A. C. Figueiredo, M. M. Oliveira, J. G. Barroso, L. G. Pedro, S. G. Deans, A. K. M. Younus, J. J. C. Scheffer, Biotechnol. Lett. 1999, 21, 859. [3] T. Ishikawa, E. Fujimata, J. Kitajima, Chem. Pharm. Bull. 2002, 50, 1460. [4] V. M. Rodrigues, P. T. V. Rosa, M. O. M. Marques, A. J. Petenate, M. A. Meireles, J. Agric. Food Chem. 2003, 51, 1518. : : [5] K. H. C. Baser, Tõbbi ve Aromatik Bitkilerin Ilaç ve Alkollü Içki Sanayinde Kullanõmõ, Istanbul Ticaret Odasõ Yayõnlarõ, Istanbul, 1997, p. 33 ± 35. [6] J. A. Duke,  Handbook of Medicinal Herbs, CRS Press; London, 1985, p. 374 ± 375. [7] I. Bohn, K. H. Kubeczka, W. Schultza, Planta Med. 1989, 55, 489. [8] J. Reichling, R. Martin, B. Kemmerer, Plant Cell Tissue Organ Culture 1995, 43, 131. [9] N. Andarwulan, K. Shetty, Food Biotechnol. 2000, 14, 1. [10] N. Tabanca, E. Bedir, N. Kirimer, K. H. C. Baser, S. I. Khan, M. R. Jacob, I. A. Khan, Planta Med. 2003, 69, 933. [11] R. Martin, J. Reichling, H. Becker, Planta Med. 1985, 3, 198. [12] A. Velasco-Negueruela, M. J. Perez-Alonso, P. L. Perez de paz, C. G. Vallejo, J. Pala-Paul, A. Inigo, Flavour Fragr. J. 2002, 17, 468. [13] A. Velasco-Negueruela, M. J. Perez-Alonso, P. L. Perez de Paz, J. Pala-Paul, J. Sanz, J. Chromatogr., A. 2003, 1011, 214. [14] V. Dev, A. T. Bottini, J. Nat. Prod. 1987, 50, 968. [15] A. T. Bottini, V. Dev, D. J. Garfagnoli, C. S. Mathela, A. B. Melkani, A. A. Miller, N. S. Sturm, Phytochemistry 1986, 25, 207. [16] N. Takeuchi, T. Fujita, K. Goto, N. Morisaki, N. Osone, S. Tobinaga, Chem. Pharm. Bull. 1993, 41, 923. [17] H. D. Friedel, R. Matusch, Helv. Chim. Acta 1987, 70, 1616. [18] K. H. Kubeczka, A. Bartsch, I. Ullmann, in Atherische Ole, Ed. K. H. Kubeczka, George Thieme Verlag, Stuttgart, 1982, p. 158 ± 187. [19] R. Kleiman, R. D. Plattner, D. Weisleder, J. Nat. Prod. 1988, 51, 249. [20] V. Dev, A. T. Bottini, J. Nat. Prod. 1987, 50, 968. [21] W. A. Koning, N. Bulow, C. Fricke, S. Melching, A. Rieck, H. Muhle, Phytochemistry 1996, 43, 629. [22] G. P. Peng, G. Tian, X. F. Huang, F. C. Lou, Phytochemistry 2003, 63, 877. [23] H. J. Williams, G. Moyna, S. B. Vinson, A. I. Scott, A. A. Bell, R. B. Stipanovic, Robert, Nat. Prod. Lett. 1997, 11, 25. [24] B. N. Su, E. J. Park, Z. H. Mbwambo, B. D. Santarsiero, A. D. Mesecar, H. H. S. Fong, J. M. Pezzuto, A. D. Kinghorn, J. Nat. Prod. 2002, 65, 1278. [25] M. Yoshikawa, T. Murakami, H. Matsuda, Chem. Pharm. Bull. 1998, 46, 1186. [26] H. Matsuda, T. Morikawa, K. Ninomiya, M. Yoshikawa, Tetrahedron 2001, 57, 8443. [27] M. Ohshiro, M. Kuroyanagi, A. Ueno, Phytochemistry 1990, 29, 2201. [28] A. Espinel-Ingrof, T. M. Kerkering, J. Clin. Microbiol. 1991, 29, 393. [29] D. E. Wedge, J. M. Kuhajek, SAAS Bull. Biochem. Biotechnol. 1998, 11, 1. Received August 26, 2004