Antimycobacterials from Lovage Root ( Ligusticum officinale Koch)

PHYTOTHERAPY RESEARCH Phytother. Res. (2012) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ptr.4823 Antimycobacterials from Lovage Root (Ligusticum officinale Koch) Juan David Guzman,1,2 Dimitrios Evangelopoulos,1 Antima Gupta,1 Jose M. Prieto,2 Simon Gibbons2* and Sanjib Bhakta1* 1 Mycobacteria Research Laboratory, Department of Biological Sciences, Institute of Structural and Molecular Biology, Birkbeck, University of London, Malet Street, London WC1E 7HX, UK 2 Department of Pharmaceutical and Biological Chemistry, UCL School of Pharmacy, 29-39 Brunswick Square, London WC1N 1AX, UK The n-hexane extract of Lovage root was found to significantly inhibit the growth of both Mycobacterium smegmatis mc2155 and Mycobacterium bovis BCG, and therefore a bioassay-guided isolation strategy was undertaken. (Z)-Ligustilide, (Z)-3-butylidenephthalide, (E)-3-butylidenephthalide, 3-butylphthalide, a-prethapsenol, falcarindiol, levistolide A, psoralen and bergapten were isolated by chromatographic techniques, characterized by NMR spectroscopy and MS, and evaluated for their growth inhibition activity against Mycobacterium tuberculosis H37Rv using the whole-cell phenotypic spot culture growth inhibition assay (SPOTi). Cytotoxicity against RAW 264.7 murine macrophage cells was employed for assessing their degree of selectivity. Falcarindiol was the most potent compound with a minimum inhibitory concentration (MIC) value of 20 mg/L against the virulent H37Rv strain; however, it was found to be cytotoxic with a half-growth inhibitory concentration (GIC50) in the same order of magnitude (SI < 1). Interestingly the sesquiterpene alcohol a-prethapsenol was found to inhibit the growth of the pathogenic mycobacteria with an MIC value of 60 mg/L, being more specific towards mycobacteria than mammalian cells (SI ~ 2). Colony forming unit analysis at different concentrations of this phytochemical showed mycobacteriostatic mode of action. Copyright © 2012 John Wiley & Sons, Ltd. Keywords: Ligusticum officinale Koch; Lovage; tuberculosis; cytotoxicity; a-prethapsenol. INTRODUCTION More than 8.5 million new cases of tuberculosis (TB) and 1.1 million deaths were estimated globally in 2010 according to the World Health Organization report (WHO, 2011). TB remains a global health emergency for several reasons, namely the appearance of multi-drug and extensively drug-resistant strains (M/XDR-TB), and immunosuppression-associated HIV/AIDS epidemic. Novel, safer and more effective drugs with no interaction with antiretroviral therapies are required for treating drug-resistant strains. Moreover, there are no specific drugs designed for treating latent TB. The TB Alliance sponsorship is putting a huge effort towards bringing a complete pipeline of anti-TB drugs; however, more research in early stages of drug discovery is necessary to fuel the pipeline. Natural products are an outstanding source of bioactive chemical scaffolds which can potentially lead to novel therapeutics for a wide array of diseases (Newman and Cragg, 2012). Several interesting antimycobacterials have been isolated from natural sources such as hirsutellones, manzamines or saringosterol 24-epimers and many other classes (Guzman et al., 2012). Large anti-TB bioprospecting screening programmes are * Correspondence to: Professor Simon Gibbons, Department of Pharmaceutical and Biological Chemistry, UCL School of Pharmacy, 29–39 Brunswick Square, London WC1N 1AX, UK; Dr Sanjib Bhakta, Mycobacteria Research Laboratory, Department of Biological Sciences, Institute of Structural and Molecular Biology, Birkbeck, University of London, Malet Street, London WC1E 7HX, UK. E-mail: [email protected]; [email protected] Copyright © 2012 John Wiley & Sons, Ltd. currently in progress, and there is a renewed interest in natural sources for finding novel antimycobacterials (Ashforth et al., 2010). The genus Ligusticum belongs to the Apiaceae plant family, a large and economically important group of herbaceous plants commonly used in traditional medicine in America, Europe and Asia. The species Ligusticum porteri, Ligusticum officinale and Ligusticum chuanxiong are medicinally important representatives of each respective region (Zschocke et al., 1998). Lovage (Ligusticum officinale Koch syn L. levisticum) is a traditional culinary ingredient in Britain for making soups, for aromatizing meats and as a seasonal winter beverage as a cordial. This species has been widely used for medicinal purposes in Europe from Mediterranean Greece and Italy to Scandinavia (Andrews, 1941). Even today in rural areas, the root is employed as a diuretic, spasmolytic and as a carminative material (Toulemonde et al., 1987). The major phytochemical compounds of Apiaceae species are the phthalide lactones and their non-aromatic dihydroand tetrahydro- derivatives, which can exist principally as monomers or dimers (Beck and Chou, 2007). Ligustilide is a common dihydrophthalide present in Ligusticum species displaying interesting biological properties notably acting as a powerful neuroprotective agent (Wu et al., 2011). Further important phyto-constituents of Ligusticum species are the polyacetylenes such as falcarindiol and falcarinol, which have been previously isolated from Lovage root and found to be mycobacterial growth inhibitors (Schinkovitz et al., 2008b). In this work, a bioassay-guided isolation of antimycobacterial constituents of Lovage root was undertaken in an attempt to isolate potentially active novel antitubercular compounds. Received 31 May 2012 Revised 23 July 2012 Accepted 25 July 2012 J. D. GUZMAN ET AL. MATERIALS AND METHODS Microbial strains, cells and culture media. Mycobacterium smegmatis mc2155 (ATCC 700084), Mycobacterium bovis BCG Pasteur (ATCC 35734), Mycobacterium tuberculosis H37Rv (ATCC 27294) and mouse RAW264.7 macrophage cells (ATCC TIB71) were used in this study. Middlebrook 7H9, Middlebrook 7H10 media, oleic acid, albumin, dextrose and catalase (OADC) and ADC supplements were purchased from BD Diagnostics. All other media and reagents were purchased from Sigma-Aldrich unless otherwise stated. Plant material, fractionation and purification of natural products. Ligusticum officinale Koch (batch number 37129) roots (490 g) obtained from Proline Botanicals Ltd, were ground using an electric mill, and the material was then exhaustively extracted using a Soxhlet apparatus with 2.5 L of n-hexane, 2.5 L of chloroform and finally 2.5 L of methanol. After filtration and removal of solvents under reduced pressure using a rotary evaporator, 7.5 g of an oily n-hexane extract, 1.1 g of chloroform extract and 13 g of methanol extract were obtained. The n-hexane extract (7.2 g) was fractionated using vacuum liquid chromatography eluting with hexane and ethyl acetate mixtures. Eighteen fractions named H1 to H18 were collected. Fraction H5, H6, H7 and H8 were chromatographed on solid-phase extraction (SPE) cartridge packed with normal phase silica (10 g, Strata, Phenomenex) eluting with hexane, ethyl acetate mixtures. Thereafter, the fractions were further fractionated by column chromatography (CC) or preparative thin-layer chromatography (PTLC) in silica gel. A schematic representation of the fractionation and purification is shown in Fig 1B. Compounds 1 and 2 were obtained as colorless amorphous solids from fraction H5S1 by CC eluting with toluene/ ethyl acetate 9:1 (v/v). Compounds 3 and 4 were also isolated as amorphous colorless solids from fraction H5S2 by PTLC eluting with toluene/ethyl acetate 95:5 (v/v). Compound 5 was obtained as colorless crystals by CC from fraction H6S2 eluting with toluene/acetone 96:4 (v/v). Compound 6 was isolated as yellow oil directly from fraction H7 by SPE in the hexane/ethyl acetate 8:2 (v/v) fraction. Compounds 7, 8 and 9 were obtained from H8S3 fraction by PTLC eluting with 100 mL of toluene/ethyl acetate 9:1 (v/v) containing five drops of formic acid. Mono and bidimensional 1H and 13C NMR spectroscopy (Bruker Avance 400) and mass spectrometry (Thermo Navigator) were employed for chemical characterisation of the isolated chemical products. Antimycobacterial testing. At each step of the chemical fractionation of the extract, the spot culture growth inhibition assay (SPOTi) was performed as previously described (Evangelopoulos and Bhakta, 2010). Initially, the n-hexane, chloroform and methanol extracts were tested against both M. smegmatis mc2155 and M. bovis BCG at 200, 100, 50, 25, 10 and 0 mg/L concentration. Thereafter, each sub-fraction that was obtained was tested only against M. bovis BCG at 100, 50, 25, 10, 5 and 0 mg/L. Briefly, extracts were dissolved in DMSO at 200 g/L concentration, while fractions and compounds were dissolved at 100 g/L. A one thousand-fold dilution was prepared from the stock and dispensed in 6 or 24-well plates (5 or 2 mL, respectively). Then, 5 mL or 2 mL of Copyright © 2012 John Wiley & Sons, Ltd. Figure 1. Antimycobacterial activity and chemical fractionation of the roots of Ligusticum officinale. (A) SPOTi assay of the three extracts of Lovage root against M. bovis BCG. (B) Schematic representation of the bioassay-guided fractionation of the extracts. The number under the fraction code represents its MIC value in mg/L against M. bovis BCG. Vacuum liquid chromatography (VLC), solid-phase extraction (SPE), column chromatography (CC) and preparative thin-layer chromatography (PTLC) were employed for fractionation/purification. (C) Chemical structures of compounds 1–9. This figure is available in colour online at wileyonlinelibrary. com/journal/ptr. molten Middlebrook 7H10 (MB7H10) agar supplemented with 0.5% glycerol and 10% OADC, were added to each well, and the plates were allowed to stand for 10 min without a lid. Then, 5 or 2 mL of an appropriately diluted (around 106 colony forming unit (CFU)/mL) mid-log phase liquid culture of the mycobacteria grown in Middlebrook 7H9 (MB7H9) broth, was carefully dispensed into the middle of each well. The plates were incubated at 37 C for two days for M. smegmatis mc2155 and for two weeks for M. bovis BCG. Minimum inhibitory concentrations (MICs) were determined visually after the incubation period. All the pure compounds were Phytother. Res. (2012) ANTIMYCOBACTERIALS FROM LOVAGE tested against Mycobacterium tuberculosis H37Rv using a similar procedure in a biosafety level III (BSL-3) laboratory. Isoniazid was included in all experiments as a positive control at 10, 1, 0.1, 0.05, 0.01 and 0 mg/L concentrations and DMSO was included as a negative control at 0.1% (v/v). Cytotoxicity assay. Mouse macrophage cells (RAW 264.7) were cultured according to a previously described method (Gupta and Bhakta, 2012). The cytotoxicity assay was performed in 96-well cell culture black flat-bottom plates (Costar; Appleton Woods) in triplicate. First, 16 mL of 50 g/L stock solutions of isolated compounds were placed in triplicate in the first row containing 200 mL of RPMI-1640 medium, and then twofold serially diluted. To each well, 100 mL of diluted macrophage cells (5  105 cells/mL) was added. After 48 h of incubation, the macrophage cells were washed twice with PBS, and fresh RPMI-1640 medium was added to each well. Plates were then treated with 30 mL of freshly prepared 0.01% resazurin filtered-sterilized solution and incubated for 16 h at 37 C. The change in color and the fluorescence intensity was recorded at lex = 560 nm, lem = 590 nm using a fluorometer microplate reader (OPTIMA, BMG LABTECH GmbH). The half growth inhibitory concentrations (GIC50) were determined based on fluorescence intensity, and the selectivity index (SI) was calculated as the ratio between the GIC50 against RAW 264.7 macrophages and the MIC against M. tuberculosis H37Rv for all of the compounds. Growth curve and CFU counting. Prethapsenol 5 was further evaluated against M. bovis BCG growing in liquid culture media. To four rolling bottles containing 90 mL of MB7H9 and 10 mL of ADC enrichment were added 50, 100 and 150 mL of a 50 g/L stock of compound 5 and completed with 150, 100 and 50 mL of DMSO respectively. A control was included by adding 200 mL of pure DMSO. A 1:100 inoculum dilution (around 107 cells) of a mid-log phase liquid culture of M. bovis BCG were added to each rolling culture, and then the four bottles were incubated at 37 C in a roller incubator set at 2 rpm. The optical density at 600 nm (OD600) was measured every day for two weeks. After 4 days, when the DMSO control reached 1.8 absorbance units (mid-exponential phase), an aliquot of the culture was diluted to 10 5, 10 6, 10 7 and 10 8 with MB7H9, and spread into Petri dishes containing solidified MB7H10 agar enriched with 10% OADC. The plates were sealed and incubated for three weeks at 37 C. Thereafter, emerging cell colonies were counted and the mean and standard deviation calculated among the three experimental replicates. Stability experiment. Eight microliters of a-prethapsenol (5) dissolved in DMSO at 50 g/L were added to 1992 mL of MB7H9 medium containing 0.5% glycerol, 0.05% Tween 80 and 10% ADC, to achieve a final concentration of 200 mg/L. A pure sample of a-prethapsenol was prepared by diluting an aliquot of 10 mL of the DMSO stock with 990 mL of methanol to achieve a concentration of 500 mg/L. A sample of supplemented MB7H9 was also included. All three samples were dispensed into 2 mL HPLC vials. The analysis was performed on an Agilent 1100 HPLC instrument using a reverse phase C-18 column (Phenomenex 10 cm, 4.7 mm, 5 mm) and a gradient mobile phase of water and acetonitrile, going from 90% water and 10% acetonitrile to 100% acetonitrile at 30 minutes after injection. The flow rate was 1.0 mL/min, the volume injected was 10 mL and a photodiode array detector was used continuously at 220 nm. Samples were maintained at 37 C and analyzed on day 0, 4, 14 and 28. The amount of a-prethapsenol was determined by calculating the area under the peak of each HPLC chromatogram. RESULTS The bioassay-guided isolation of chemical entities from the roots of L. officinale was carried out by evaluating the activity against M. bovis BCG at each step of chemical fractionation process. Initially, only the lowest polarity solvent extract obtained from the roots showed activity (Fig 1A). More polar chloroform and methanol extracts were inactive at the highest concentration tested (200 mg/L). The n-hexane extract completely inhibited both M. smegmatis mc2155 and M. bovis BCG growth at 100 mg/L. Upon vacuum liquid chromatography fractionation, the antimycobacterial activity principally concentrated into three fractions (H5-H7) having MIC values of 25 mg/L (Fig 1B). Further fractionation and purification of H5 led to the isolation of four monomers of ligustilides, namely (Z)-ligustilide (1), (Z)3-butylidenephthalide (2), (E)-3-butylidenephthalide (3) Table 1. MIC values against two slow-growing mycobacterial species and cytotoxicity and selectivity against the murine macrophage cell line RAW 264.7 Compound 1 2 3 4 5 6 7 8 9 Isoniazid MIC M. tuberculosis H37Rv (mg/L) MIC M. bovis BCG (mg/L) GIC50 Macrophages RAW264.7 (mg/L) Selectivity index (GIC50/MICH37Rv) >100 >100 >100 100 60 20 >100 >100 >100 0.1 >100 >100 >100 100 60 5 >100 100 100 0.1 250 250 250 200 125 5 200 100 100 3000a <2.5 <2.5 <2.5 2 2.1 0.25 <2 <1 <1 30000 a Value taken from reference (Gupta and Bhakta, 2012) Copyright © 2012 John Wiley & Sons, Ltd. Phytother. Res. (2012) J. D. GUZMAN ET AL. and 3-butylphthalide (4). Their NMR and MS spectra were in agreement with previously reported data (Miyazawa et al., 2004; Schinkovitz et al., 2008a). a-Prethapsenol (5), a sesquiterpenoid recently isolated for the first time from the roots of Ligusticum grayi (Cool et al., 2010), was obtained from H6. More than 300 mg of pure falcarindiol (6) were recovered from H7. Unambiguous identification was achieved by NMR and MS spectra analysis and comparison with reported data (Lechner et al., 2004). A dimer of (Z)-ligustilide, levistolide A (7) and two furanocoumarins psoralen (8) and bergapten (9) were obtained from fraction H8 and characterized by NMR and MS spectra (Kaouadji et al., 1986; Masuda et al., 1998). The chemical structures of the isolated compounds are depicted in Fig 1C. All of the ligustilides (1–4) displayed low potency against slow-growing mycobacterial species having MIC values ≥ 100 mg/L (Table 1). The saturated chain of compound 4 slightly improved the activity in comparison with both geometric configurations of the alkenes 2–3. Half-growth inhibition concentrations (GIC50) against macrophage RAW 264.7 cells were also comparable for the ligustilides with values between 200 and 250 mg/L. Their specificity against mycobacteria in relation to the mammalian cell line was low (SI < 2.5). The sesquiterpene alcohol a-prethapsenol (5) displayed an MIC value of 60 mg/L against both M. bovis BCG and M. tuberculosis H37Rv. At higher concentrations of 5, the growth of the bacteria was completely inhibited resulting in clear wells on the SPOTi assay. The cytotoxicity profile of a-prethapsenol indicated that it had certain selectivity towards killing mycobacteria in comparison with the macrophage cell line with an SI value of 2.1. Falcarindiol (6) was obtained in high yields from Lovage roots (0.1% w/w) and as found previously by several authors (Deng et al., 2008; Schinkovitz et al., 2008b) it is a potent inhibitor of M. tuberculosis growth displaying in this study an MIC value of 5 mg/L against M. bovis BCG and 20 mg/L against M. tuberculosis H37Rv. However, falcarindiol was found to be toxic towards mammalian RAW264.7 cells, being able to kill half the population at a concentration of 5 mg/L. The SI was therefore less than 1, suggesting the possibility of a specific mammalian metabolic interference. Levistolide A (7) was inactive against mycobacteria showing an MIC value higher than 100 mg/L. This dimer had a RAW264.7 macrophage GIC50 value of 200 mg/L. Furanocoumarins psoralen 8 and bergapten 9 were both inhibitory to M. bovis BCG at 100 mg/L, and their cytotoxicity towards murine macrophages was in the same order of magnitude as the MIC suggesting poor selectivity. The sesquiterpene alcohol a-prethapsenol (5) was selected for further study against M. bovis BCG in a liquid culture-based assay. The growth curve under different concentrations of 5 was constructed (Fig 2A). An extended lag phase of growth was observed when the concentration was equal or higher than the MIC. At 75 mg/L, four more days of incubation were required for the bacteria to attain the exponential growth observed in the untreated control (Fig 2A). A minor delay of the lag phase was also noted at a concentration of 50 mg/L concentration of 5. CFU counts however showed only a minor decrease in cell viability when treated with a concentration > MIC, and after removal of the pressure of the compound, high numbers of bacteria were able to grow (Fig 2B). Initially, it was Copyright © 2012 John Wiley & Sons, Ltd. Figure 2. Effect of different concentrations of a-prethapsenol (5) on the growth of M. bovis BCG and stability experiment of a-prethapsenol. (A) Growth curve of M. bovis BCG under three different concentrations of 5. The concentrations 0, 25, 50 and 75 mg/L are represented as circles, squares, triangles and diamonds, respectively. (B) Number of colony forming units of M. bovis BCG treated with different concentrations of 5 at day 4. Each bar represents the mean CFU  SD (C) HPLC chromatograms of MB7H9, pure a-prethapsenol and incubated in MB7H9 media at 37 C analyzed at different time points. thought that a-prethapsenol might be chemically unstable in MB7H9 media at 37 C, and therefore a stability experiment was undertaken. After one month of incubation in MB7H9 at usual growth conditions, the same amount of the small molecule was measured by reverse phase HPLC (Fig 2C), indicating high stability under these conditions. DISCUSSION The ligustilide monomers 1–4 and dimer 7 isolated in this study from Lovage root were found to be weak inhibitors of mycobacterial growth and only compound 4 showed inhibition at the highest concentration tested. Phytother. Res. (2012) ANTIMYCOBACTERIALS FROM LOVAGE (Z)-Ligustilide 1 isolated from Osha root (Ligusticum porteri) has been found to inhibit the growth of Staphylococcus aureus at 128 mg/L, showing potentiation of norfloxacin against an efflux-mediated S. aureus drugresistant strain (Cégiéla-Carlioz et al., 2005). Other bacteria and viruses (Beck and Stermitz, 1995) but also cancer cell lines (Kan et al., 2008) have been reported to be inhibited at high concentration of 1, and all these results point out a general mechanism of action affecting different types of cells. However, there is the possibility that other ligustilide-related compounds appearing in the complex fraction H-5/H-6 at lower concentration could be responsible for the high activity of these fractions. In particular, we observed by NMR analysis a small fraction containing a 1:1 mixture of an epoxyligustilide and (Z)-butylidenephtalide which was more active than 2 alone, but the low quantity of this fraction prevented full chemical and biological characterisation. Interestingly, the novel sesquiterpene skeleton of a-prethapsenol (5) was found to be active against both M. bovis BCG and M. tuberculosis H37Rv strains. No closely related structures have been shown to possess antimicrobial effects. Other sesquiterpenoids such as the benzoylated dihydroagarofuranoids isolated from Microtropis species (Chen et al., 2007), the sesquiterpenelactones based on germacrane, eudesmane or guaiane skeletons (Fischer et al., 1998; Vongvanich et al., 2006) or farnesol (Rajab et al., 1998) have shown significant antitubercular activity, but they share little structural similarity to 5. In other words, this compound might be affecting the bacteria through a particular unknown biochemical mechanism while being less active against eukaryotic cell lines. Related compounds have been found in L. grayi (Cool et al., 2010), and a systematic anti-tubercular screening of these entities may underpin more potent natural products. Falcarindiol (6) showed high potency but little selectivity. It is considered to be more active towards eukaryotic cells than bacteria, a not very appealing result for TB-drug discovery research; however, this compound may have an essential role to play in future cancer-drug discovery, and efforts in this direction are being made (Zaini et al., 2012). Furanocoumarins 8 and 9 showed low antimycobacterial activity and selectivity suggesting a non-specific mechanism of action, although other coumarins such as umbelliferone, scopoletin, xanthyletin and (S)-marmesin have been found to be active with MIC values in the range 40–60 mg/L (Chiang et al., 2010). M. bovis BCG cells were able to grow exponentially after one week of incubation with concentrations of the prethapsane alcohol 5 above its MIC. This natural product was found to be stable when dissolved in MB7H9 media and maintained under the conditions of incubation, and therefore it is reasonable to think that the concentration of 5 remains unchanged during the assay, although a bio-modification mechanism cannot be ruled out at this point. It seems sensible to suggest that BCG growth was delayed by 5, without notable bacterial killing as the CFU counting remained in the same order of magnitude for all treatments, indicating a shift towards the viable but non-replicating physiological state of mycobacteria. The bacteriostatic effect of 5 was consistent with the lag-phase extension seen in the growth curve because onset of exponential growth phase is dependent on sufficient supply of essential biomolecules. It is known that in the lag phase, bacterial cells prepare primordial molecules (nucleosides triphosphates, sugars, metals and cofactors) and enzymes for sustained cell division (Rolfe et al., 2012), and therefore we hypothesize that 5 may interfere with one or several steps of this transient adaptation. Acknowledgements The authors would like to acknowledge Bloomsbury Colleges, University of London for a Research Studentship to JDG and Medical Research Council, UK for a New Investigators Research Grant award (code G0801956) to SB. The authors are grateful to Professor Simon Croft for his generous support with the TB culture facility at the London School of Hygiene and Tropical Medicine. Conflict of Interest The authors have declared that there is no conflict of interest. REFERENCES Andrews AC. 1941. Alimentary use of Lovage in the Classical period. Isis 33: 514–518. Ashforth EJ, Fu C, Liu X, et al. 2010. Bioprospecting for antituberculosis leads from microbial metabolites. Nat Prod Rep 27: 1709–1719. Beck JJ, Chou S-C. 2007. The structural diversity of phthalides from the Apiaceae. J Nat Prod 70: 891–900. Beck JJ, Stermitz FR. 1995. Addition of methyl thioglycolate and benzylamine to (Z)-ligustilide, a bioactive unsaturated lactone constituent of several herbal medicines. An improved synthesis of (Z)-ligustilide. J Nat Prod 58: 1047–1055. Cégiéla-Carlioz P, Bessière J-M, David B, Mariotte A-M, Gibbons S, Dijoux-Franca M-G. 2005. Modulation of multi-drug resistance (MDR) in Staphylococcus aureus by Osha (Ligusticum porteri L., Apiaceae) essential oil compounds. Flavour Frag J 20: 671–675. Chen J-J, Chou T-H, Peng C-F, Chen I-S, Yang S-Z. 2007. Antitubercular dihydroagarofuranoid sesquiterpenes from the roots of Microtropis fokienensis. J Nat Prod 70: 202–205. Chiang C-C, Cheng M-J, Peng C-F, Huang H-Y, Chen I-S. 2010. A novel dimeric coumarin analog and antimycobacterial constituents from Fatoua pilosa. Chem Biodivers 7: 1728–1736. Copyright © 2012 John Wiley & Sons, Ltd. Cool LG, Vermillion KE, Takeoka GR, Wong RY. 2010. Irregular sesquiterpenoids from Ligusticum grayi roots. Phytochemistry 71: 1545–1557. Deng S, Wang Y, Inui T, et al. 2008. Anti-TB polyynes from the roots of Angelica sinensis. Phytother Res 22: 878–882. Evangelopoulos D, Bhakta S. 2010. Rapid methods for testing inhibitors of mycobacterial growth. Antibiotic Resistance Protocols, Methods in Molecular Biology 642: 279. Fischer NH, Lu T, Cantrell CL, Castaneda-Acosta J, Quijano L, Franzblau SG. 1998. Antimycobacterial evaluation of germacranolides in honour of professor G.H. Neil Towers 75th birthday. Phytochemistry 49: 559–564. Gupta A, Bhakta S. 2012. An integrated surrogate model for screening of drugs against Mycobacterium tuberculosis. J Antimicrob Chemother 67: 1380–1391. Guzman JD, Gupta A, Bucar F, Gibbons S, Bhakta S. 2012. Antimycobacterials from natural sources: ancient times, antibiotic era and novel scaffolds. Front Biosci 17: 1861–1881. Kan WLT, Cho CH, Rudd JA, Lin G. 2008. Study of the antiproliferative effects and synergy of phthalides from Angelica sinensis on colon cancer cells. J Ethnopharmacol 120: 36–43. Phytother. Res. (2012) J. D. GUZMAN ET AL. Kaouadji M, De Pachtere F, Pouget C, Chulia AJ, Lavaitte S. 1986. Three additional phthalide derivatives, an epoxymonomer and two dimers, from Ligusticum wallichii rhizomes. J Nat Prod 49: 872–877. Lechner D, Stavri M, Oluwatuyi M, Pereda-Miranda R, Gibbons S. 2004. The anti-staphylococcal activity of Angelica dahurica (Bai Zhi). Phytochemistry 65: 331–335. Masuda T, Takasugi M, Anetai M. 1998. Psoralen and other linear furanocoumarins as phytoalexins in Glehnia littoralis. Phytochemistry 47: 13–16. Miyazawa M, Tsukamoto T, Anzai J, Ishikawa Y. 2004. Insecticidal effect of phthalides and furanocoumarins from Angelica acutiloba against Drosophila melanogaster. J Agric Food Chem 52: 4401–4405. Newman DJ, Cragg GM. 2012. Natural products as sources of new drugs over the 30 years from 1981 to 2010. J Nat Prod 75: 311–335. Rajab MS, Cantrell CL, Franzblau SG, Fischer NH. 1998. Antimycobacterial activity of (E)-phytol and derivatives: A preliminary structure-activity study. Planta Med 64: 2–4. Rolfe MD, Rice CJ, Lucchini S, et al. 2012. Lag phase is a distinct growth phase that prepares bacteria for exponential growth and involves transient metal accumulation. J Bacteriol 194: 686–701. Schinkovitz A, Pro SM, Main M, et al. 2008a. Dynamic nature of the ligustilide complex. J Nat Prod 71: 1604–1611. Copyright © 2012 John Wiley & Sons, Ltd. Schinkovitz A, Stavri M, Gibbons S, Bucar F. 2008b. Antimycobacterial polyacetylenes from Levisticum officinale. Phytother Res 22: 681–684. Toulemonde B, Paul F, Noleau I. 1987. Phtalides from Lovage (L. officinale Koch). In Flavour Science and Technology, Martens M, Dalett A, Russwurm H (eds). John Wiley and Sons: New York; 89–94. Vongvanich N, Kittakoop P, Charoenchai P, Intamas S, Sriklung K, Thebtaranonth Y. 2006. Antiplasmodial, antimycobacterial, and cytotoxic principles from Camchaya calcarea. Planta Med 72: 1427–1430. WHO. 2011. Global tuberculosis control WHO report 1–8. Wu X-m, Qian Z-m, Zhu L, et al. 2011. Neuroprotective effect of ligustilide against ischaemia-reperfusion injury via up-regulation of erythropoietin and down-regulation of RTP801. Br J Pharmacol 164: 332–343. Zaini RG, Brandt K, Clench MR, Le Maitre CL. 2012. Effects of bioactive compounds from carrots (Daucus carota L.), polyacetylenes, beta-carotene and lutein on human lymphoid leukaemia cells. Anticancer Agents Med Chem 12: 640–652. Zschocke S, Liu J-H, Stuppner H, Bauer R. 1998. Comparative study of roots of Angelica sinensis and related umbelliferous drugs by thin layer chromatography, high-performance liquid chromatography, and liquid chromatography – mass spectrometry. Phytochem Anal 9: 283–290. Phytother. Res. (2012)