http://www.biomed.cas.cz/mbu/folia/ Folia Microbiol. 54 (3), 179–187 (2009) Hydroxylated Anthraquinones Produced by Geosmithia species E. STODŮLKOVÁa, M. KOLA ÍKa, Z. K ESINOVÁa, M. KUZMAa, M. ŠULCa, P. MANa, P. NOVÁKa, P. MARŠÍKb, P. LANDAb, J. OLŠOVSKÁa, M. CHUDÍČKOVÁa, S. PAŽOUTOVÁa, J. ČERNÝc, J. BELLAd, M. FLIEGERa * aDepartment of Biogenesis and Biotechnology, Institute of Microbiology, Academy of Sciences of the Czech Republic v.v.i., 142 20 Prague 4, Czech Republic bLaboratory of Plant Biotechnologies, Joint Laboratory of the Institute of Experimental Botany, Academy of Sciences of the Czech Republic v.v.i., and Research Institute of Crop Production v.v.i., 166 00 Prague 6, Czech Republic cDepartment of Cell Biology, Faculty of Science, Charles University, 128 00 Prague 2, Czech Republic dSchool of Chemistry, The University of Edinburgh, Edinburgh EH9 3JJ, UK Received 12 January 2009 ABSTRACT. Geosmithia fungi are little known symbionts of bark beetles. Secondary metabolites of lilac colored species G. lavendula and other nine Geosmithia species were investigated in order to elucidate their possible role in the interactions of the fungi with environment. Hydroxylated anthraquinones (yellow, orange, and red pigments), were found to be the most abundant compounds produced into the medium during the submerged cultivation. Three main compounds were identified as 1,3,6,8-tetrahydroxyanthraquinone (1), rhodolamprometrin (1-acetyl-2,4,5,7-tetrahydroxyanthraquinone; 2), and 1-acetyl-2,4,5,7,8-pentahydroxyanthraquinone (3). Compounds 2 and 3 (representing the majority of produced metabolites) inhibited the growth of G+-bacteria Staphylococcus aureus and Bacillus subtilis with minimum inhibitory concentration of 64–512 µg/mL. Anti-inflammatory activity detected as inhibition of cyclooxygenase-2 was found only for compound 3 at 1 and 10 µg/mL. Compound 2 interfered with the morphology, compound 3 with cell-cycle dynamics of adherent mammalian cell lines. Abbreviations AQ(s) COX-1 COX-2 CZD DAPI LOD anthraquinone(s) cyclooxygenase-1 cyclooxygenase-2 Czapek–Dox (medium) 4´,6-diamidino-2-phenylindole limit of detection LOQ MEA MIC(s) PBS UPLC limit of quantification malt extract (medium) minimal inhibitory concentration(s) phosphate-buffered saline ultra-performance liquid chromatography Filamentous fungi of the genus Geosmithia (Ascomycota:Hypocreales) are little known but regular and worldwide distributed associates of bark beetles (Coleoptera:Curculionidae, Scolytinae) (Kola ík et al. 2007, 2008) (Fig. 1). Fungi interact with their host plant in numerous ways (Six 2003) but the nature of these interactions, in case of Geosmithia, is still unknown. The identification of extrolytes is one of the approaches providing the insight into the ecology of Geosmithia. Such compounds exhibit numerous biological activities which make them also candidates for further pharmacological or biotechnological investigations and other applications. Geosmithia species exhibit phenotype highly similar to the genus Penicillium, but lack green colony colors. The most distinctive is the complex of species, including G. lavendula, having lilac to red colored conidia, a feature extremely rare among filamentous fungi (Kola ík et al. 2007). During the first screening of secondary metabolites of G. lavendula, hydroxylated anthraquinones (yellow, orange, and red pigments) were found as the most abundant and distinctive compounds produced into the medium during submerged cultivation. AQs, an important group of >170 pigments widely distributed in nature, represents the largest group of natural quinones distributed in fungi, lichens, bacteria, and plants or animals (Gill 2001; Pankewitz and Hilker 2006; Thomson 1997). Hydroxylated AQs are involved in interactions of their producers with the environment and can serve as devices against radiation (Edwards 2004), competing organisms (Chrysayi-Tokousbalides and Kastanias 2003) and predation (Hilker and Köpf 1994; Avery et al. 1997; Ganapaty et al. 2004). *Corresponding author; fax +420 241 062 347; e-mail flieger@biomed.cas.cz . 180 E. STODŮLKOVÁ et al. Vol. 54 Natural AQs are characterized by a large structural variety and exhibit very diverse effects with broad pharmacological activities and low toxicity (Müller 2000). They possess anti-inflammatory, antiviral (Semple et al. 2001; Schinazi et al. 1990), antimicrobial (Izhaki 2002), antifungal (Chrysayi-Tokousbalides and Kastanias 2003; Rath et al. 1995), antimalarial (Sittie et al. 1999), hypotensive, analgetic (Younos et al. 1990), antioxidant (Tripathi et al. 1997), moderate antitumor, cytostatic (Gálvez et al. 1996), antileukemic, and mutagenic (Ismail et al. 1997), astringent, and purgative effects (Bruneton 1999). Another important feature of naturally produced AQs is their great biotechnological potential. Many of them are known as fabric dyes, additives to mordant, and also as stains used in histology (Hobson and Wales 1998; Avwioro et al. 2005). Fig. 1. Left: Gallery system of fig bark beetle (Hypoborus ficus) on common fig (Ficus carica) with mycelium and spores of G. lavendula (lilac) and Geosmithia sp. (white) (Croatia 2003; galleries were mechanically exposed and cultivated for 14 d at 25 °C); bar = 10 mm. Right: Colonies of G. lavendula (marginal lines) and Geosmithia sp. representing the same species as CCF 3660 used here (central line) (MEA, 14 d, 25 °C). Here we report on the isolation, structure determination, chemical characterization, and production as well as the evaluation of bioactivity of colored metabolites produced by G. lavendula and other nine Geosmithia species having lilac to red colored conidia and/or producing the same pigment into the agar media. This is the first report on the production of secondary metabolites by Geosmithia species. MATERIALS AND METHODS Fungal strains. G. lavendula (CCM 8366) was isolated from Hypoborus ficus (Coleoptera:Scolytinae) feeding on Ficus carica in Torino di Sangro, Abruzzo Region (Italy), 2004. Species identification was done based on the identity of phenotype and ITS1–5.8–ITS2 region of ribosomal DNA (deposited under GenBank no. AM421122) with the type strain of G. lavendula. The molecular-genetic analysis was done according to Kola ík et al. (2007). Other tested strains producing red or yellow pigments into agar plate are specified in Table I. All strains are deposited in the Culture Collection of Fungi, Prague (code CCF), Czech Collection of Microorganisms, Brno (code CCM) or in the Institute of Microbiology, Academy of Sciences of the Czech Republic, Prague (codes MK and RJ). Cultivation conditions. The stock cultures of monosporic strains were maintained on malt agar slants (malt extract 20 g/L, agar 20 g/L) and cultivated on CZD (in g/L: sucrose 30, agar 20, NaNO3 3, K2HPO4 1, MgSO4 0.5, KCl 0.5, Fe2(SO4)3 0.01; pH 6.5) or MEA (in g/L: malt extract 20, glucose 20, peptone 1). Submerged cultivations were done in 250 mL Erlenmeyer flasks on a rotary shaker (200 rpm) for 14 or 21 d at 24 °C in the dark. Pigment extraction. Fermentation broth of G. lavendula was centrifuged and extracted 3× with equal volume of AcOEt–CH3COOH 20 : 1 (V/V). The pooled extracts were dried over anhydrous Na2SO4, filtered, and evaporated to dryness under reduced pressure. Column chromatography. The dry mixture prepared by trituration of silica gel and crude AcOEt extract was chromatographed using silica gel column (Kieselgel 60, 70–230 mesh ASTM; Merck, Germany) washed with n-C6H14–AcOEt 3 : 2 (V/V) followed by n-C6H14–AcOEt–CH3COOH 3 : 2 : 1 (V/V/V) and finally with n-C6H14–AcOEt–CF3COOH 6 : 4 : 3 (V/V/V) as eluents. The compounds eluted in the following sequence: 1 (yellow), 2 (orange), 3 (red). 2009 ANTHRAQUINONES PRODUCED BY Geosmithia spp. 181 182 E. STODŮLKOVÁ et al. Vol. 54 UPLC. An ACQUITY UPLC system (Waters, USA) equipped with 2996 PDA detector was used for method development and for quantitative determination of all three compounds 1, 2, and 3 in fermentation broth. The analyses were run on BEH Shield RP18 column (50 × 2.1 mm, particle size 1.7 µm; Waters) at a flow rate of 0.4 mL/min with the following mobile phases: A, 13 mmol/L CF3COOH and B, CH3CN, isocratic elution in A : B (73 : 27; V/V); column temperature 35 °C, UV detection 225 nm, data sample rate 20 pts/s, filter constant 0.5, injection volume 3 µL, analysis time 17 min. Standard solutions (1 mg/mL) of crystallized 2 (diluted in MeOH) and powdered 3 (MeOH–Me2SO 9 : 1, V/V) were prepared in triplicates for quantitative analyses at concentrations of 1.95, 3.9, 7.8, 15.6, 31.25, 62.5, 125, 250, 500, and 1000 µg/mL. The calibration graphs were constructed by plotting the integrating peak areas vs. concentration. The following linear regression equations and correlation coefficients were obtained: compound 2, y = 3.54 · 104x – 5.92 · 104; r = 0.99985; compound 3, y = 3.18 · 104x – 1.87 · 104; r = 0.99980 for calibration graphs on scale 1.95–1 000 µg/mL. The LOD (µg/mL) and LOQ (µg/mL) were: compound 2; LOD 2.35, LOQ 7.13; compound 3, LOD 2.25, LOQ 6.81. To quantify compound 1, linear regression curve of compound 2 was used due to insufficient quantity of purified sample. Analysis of AQs in fermentation broth. For analysis of AQs production, AcOEt–CH3COOH extracts (2 mL, in triplicates) were analyzed after 1, 3, 7, 11, 15, and 21 d. Mass spectrometry. Experiments were performed on a commercial 9.4T APEX-Ultra FTMS instrument (Bruker Daltonics, USA). The instrument operated in negative ion mode. Spectra were collected over the mass range 150–2000 m/z at 1M data points resulting in maximum resolution of 2 × 105 at 400 m/z. One µL of each sample was dissolved in 0.1 mL of MeOH–H2O (1 : 1) and introduced to MS by direct infusion via electrospray ion source. The flow rate was 1.5 µL/min and the temperature of drying gas (nitrogen) was set to 230 °C. The species of interest were isolated in the gas phase with a 3.0 m/z window and fragmentation was induced by dropping the potential of the collision cell (12 V). The accumulation time was set to 0.1 and 0.2 s (0.5 s for MS–MS), the cell was opened for 1200 µs, and 8 experiments were collected for one spectrum. The instrument was externally calibrated using singly-charged arginine clusters resulting in sub-ppm accuracy. The spectra were apodized using sin apodization with one zero fill. NMR spectra were measured on a VarianUNITY Inova-400 spectrometer (399.87 and 100.55 MHz, respectively) and Bruker Avance 800 (800.01 and 201.16 MHz, respectively) equipped with TCI CryoProbe optimized for 1H and 13C observation in (CD3)2SO at 303 and 330 K, respectively. Residual signal of solvent was used as an internal standard (δH 2.500 ppm, δC 39.60 ppm). The structures were evaluated based on 1H NMR, 13C NMR, COSY, HMQC, HSQC, HMBC, gHMBCAD, INADEQUATE, and ROESY spectra, which were measured using standard manufacturers’ software (Varian, USA, and Bruker BioSpin, Germany, respectively). 1H NMR spectrum was zero filled to fourfold data points and multiplied by window function (two-parameter double-exponential Lorentz–Gauss function) before Fourier transformation to improve resolution. A line broadening (1 Hz) was applied to the 13C NMR spectrum and 4× zero-filled before Fourier transformation. As the compounds are proton-poor, the assignment was based upon heteronuclear long-range correlation experiments (HMBC, gHMBCAD) and particularly 2D 13C–13C correlated experiment (INADEQUATE) carried out on high sensitivity CryoProbe leading to carbon spin–spin connectivities through the whole spin system and consequently enabling the assignment of all carbon signals. Antimicrobial activity of colored compounds 2 and 3 was tested in vitro against the following microbial strains purchased from Oxoid (UK): yeast (Candida albicans ATCC 10231), G+- (Bacillus subtilis ATCC 6633; Staphylococcus aureus subsp. aureus ATCC 25923), and G–-bacteria (Escherichia coli ATCC 25922; Pseudomonas aeruginosa ATCC 27853) cultivated in Mueller–Hinton broth (Oxoid). Nystatin and ciprofloxacin (both Sigma-Aldrich, Czech Republic) were used as a positive control of microbial strain susceptibility. Compound 1 was not tested due to its insufficient quantity. In vitro antimicrobial activity was determined from the Mueller–Hinton fermentation broth by microdilution method using 96-well microtitre plates. Each well was inoculated with 5 µL of bacterial or yeast suspension at CFU concentration of 107/mL and incubated at 37 °C for 1 d (2 d for the yeast). Wells containing 3 % Me2SO were assayed as a blank control and did not inhibit any microbial strains. Compounds 2 and 3 were dissolved in Tris-buffered saline (pH 7.6) using a final concentration of 3 % Me2SO. Two-fold serial dilutions of each tested substance were subsequently carried out starting from the concentration of 512 mg/mL. The growth of microorganisms was observed as absorbance A600 by the UV–VIS spectrophotometer Helios epsilon (Spectronic Unicam, UK). MICs were calculated based on the 2009 ANTHRAQUINONES PRODUCED BY Geosmithia spp. 183 growth control (pigment-free) and were the lowest pigment concentration that resulted in 80 % reduction. All samples were tested in triplicate. COX-1 and COX-2 assays. Ovine COX-1 (specific activity 72 900 U/mg) and COX-2 (specific activity 5416 U/mg) were purchased from Cayman Chemicals (USA); 1-14C-arachidonic acid (specific activity 1.78 GBq/mmol) was from PerkinElmer (USA); scintillation solution MonoflowTM 2 was from National Diagnostics (USA). The slightly modified assay designed according to Noreen et al. (1998) was used. COX-1 (0.35 mg protein) and COX-2 (0.88 mg protein) enzyme solution (10 µL) was activated with Tris-HCl (pH 8.0) reaction mixture (60 µL) containing l-epinephrine (1 mmol/L), reduced glutathione (1 nmol/L), and hematin (1 µmol/L) on ice for 5 min. Then compounds 2 and 3 dissolved in 10 µL of EtOH were added into the enzyme mixtures and reaction was started by the addition of 1-14C-arachidonic acid (1.85 kBq). After a 15-min incubation at 37 °C, the reaction was terminated by adding 40 µL of 4 mol/L HCOOH. Resulting arachidonic acid metabolic products were extracted by CHCl3 (200 µL), evaporated in a stream of nitrogen to dryness, dissolved in 10 µL of EtOH, and analyzed by HPLC equipped with flow-through beta-RAM Detector (Model 3B; IN/US Systems, USA). The positive and negative control was represented by indomethacin and boiled enzymes for all experiments. The corresponding volume of EtOH was used as a solvent control. All experiments were performed in triplicate. Bioactivity assays on HeLa cells. Cells were cultivated in D-MEM medium supplemented with 10 % FCS (Gibco–Invitrogen, USA) grown on glass cover slips (up to 50 % density) in 6-well plates (Nunc; Thermo Fisher Scientific, USA), treated with 2 and 3 dissolved in Me2SO (stock solutions 10 mmol/L) for various times and with various concentrations. Wells containing Me2SO only (1 %) were assayed as a blank control – no alteration of cellular morphology and physiology was observed. All samples were tested in triplicate. Treated cells (35 % CO2, 7 °C) were fixed (3.7 % paraformaldehyde in PBS, 20 min, room temperature), permeabilized (1 % Triton X-100 in PBS), blocked (1 % bovine serumalbumin in PBS) and stained with Phalloidin-Alexa567, anti-tubulin antibody followed by secondary GAM-Alexa488. Staining of the nuclei was performed during mounting in Mowiol–DAPI. Specimens were visualized and images processed using Cell® microscope (Olympus, Japan). For the quantification, individual cells on cover slips were manually counted and evaluated, 1500 cells for each condition. All fluorescent reagents (including secondary antibodies) were purchased from Molecular Probes (Invitrogen, USA). Chemicals. All reagents and solvents (HPLC gradient grade) were purchased from Sigma unless otherwise stated. RESULTS The mixture of secondary metabolites produced to CZD fermentation broth by a submerged culture of G. lavendula was extracted using AcOEt. Combined extracts were dried over Na2SO4, and evaporated to dryness yielding the crude dark red pigments. This crude residue was subjected to silica gel column chromatography and further purified by crystallization or repeatedly separated by column chromatography into pure compounds. The structural determinations of the compounds were done by mass spectrometry, NMR and UV– VIS spectroscopy. The compounds were characterized as anthraquinones, viz. 1,3,6,8-tetrahydroxyanthraquinone (1), 1-acetyl-2,4,5,7-tetrahydroxyanthraquinone (rhodolamprometrin; 2), and 1-acetyl-2,4,5,7,8-pentahydroxyanthraquinone (3) (Table II). Compound 1 gave deprotonized ion at m/z 271.0247 (C14H7O6) on the FT-ICR mass spectrometer. Molecular formula, together with MS–MS spectra and the 1H and 13C NMR data (Table III), matched 1,3,6,8-tetrahydroxyanthraquinone. The UV–VIS spectra had characteristic absorption maxima for hydroxyanthraquinones at 221, 291, and 449 nm. Compound 2 was obtained in crystalline form as orange needle-shaped crystals. The molecular formula determined by negative ion mode FTMS (C16H9O7; m/z 313.0349) as well as the fragmentation pattern matched the results from NMR analysis. The UV–VIS spectra showed absorption maxima at 223, 263, 292, and 459 nm. These results confirmed the structure of rhodolamprometrin (1-acetyl-2,4,5,7-tetrahydroxyanthraquinone). The negative mass spectrum of powdered red compound 3 gave rise to C16H9O8, [M–H]– 329.0299. Such composition was in perfect agreement with the structure of 1-acetyl-2,4,5,7,8-pentahydroxyanthraquinone, as confirmed by both MS–MS and NMR. The UV–VIS spectra had absorption maxima at 231, 262, 314, and 502 nm. 184 E. STODŮLKOVÁ et al. Vol. 54 Quantitative UPLC determination of extracellular AQs showed that 3 could be detected on CZD medium of submerged cultures already after the 1st day. Its production increased steadily with cultivation time, reaching a maximum concentration of 104 mg/mL after 21 d. The concentration of 2 had a similar pattern as for 3 but with lower maximum (62.5 µg/mL). The production of 1 was also recorded but only in low quantity (<3 µg/mL). Crude AcOEt extract prepared from dry filtrate of fungal biomass (30 g) consisting mainly of spores, fungal debris, and mycelium contained additional 622 µg/g of compound 3, 13 µg/g of 2, and traces of 1. The production of AQs was strongly suppressed when G. lavendula was cultivated on MEA medium and only traces of 2 and 3 (<1 µg/mL) were determined at the end of the cultivation (21 d). No measurable amount of 1 was detected. Table III. The 1H and 13C chemical shifts of compounds 1–3a 1 2 3 Position 1 2 3 4 4a 5 6 7 8 8a 9 9a 10 10a 1-Ac 1-CO 2-OH 4-OH 5-OH 7-OH 8-OH δC δH, J δC 108.74 d 165.09 s 108.16 d 164.27 s 108.76 s – – – – – 181.45 s 135.03 s 188.67 s – – – – – – – – 7.126 d (2.4) 126.69 s 161.26 s 108.20 d 163.67 s 107.97 s 164.11 s 108.33 d 165.14 s 108.83 d 134.58* s 181.95 s 130.85* s 188.63 s 108.59 s 30.73 q 201.24 s – – – – – 6.957 d (2.4) 11.334 br s 12.199 br s δH, J 6.655 s 6.602 d (2.3) 7.051 d (2.3) 2.382 s nd 12.425 br s 12.120 br s nd – δC 125.91 s 160.82 s 108.46 d 163.65 s 108.53 s 159.43 s 110.18 d 156.84 s 149.65 s 112.26 s 186.40 s 130.52 s 186.25 s 103.99 s 30.95 q 201.27 s – – – – – δH, J 6.694 s 6.637 s 2.424 s nd 12.699 br s 12.574 br s nd *Assigned by comparison with 1 and 3. as – singlet, br s – broad singlet, d – doublet, q – quadruplet; J – coupling constant (Hz); nd – not determined. Additionally, nine strains of Geosmithia spp. (Table I) were assessed for production of pigments on both, CZD and MEA media except strains RJ0258 and RJ0363 that did not grow on CZD medium. UPLC analysis of pigments isolated from 21-d-old fermentation broth revealed that the highest content of AQs was in the strain CCF 3655 cultivated on MEA medium (compound 3, 17.8 µg/mL; 2, 2.6 µg/mL) followed by strain CCF 3660 where content of 3 and 2 reached 5.36 and 1.57 µg/mL, respectively. Traces of 1 were detected only in CCF 3660 and RJ0258 cultivated also on MEA medium. Much lower production levels of 2 and 3 were observed for these strains on CZD medium. The other tested strains produced none or very low quantity of 2 and 3. The strains RJ0363, CCF 3645, and MK 1653 produced no AQs either on CZD or MEA medium despite the fact that the fungal colonies showed colored surroundings on CZD solid medium. The antibacterial activity of 2 and 3 was examined in the presence of four bacterial and one yeast strains, ciprofloxacin and nystatin being used as a control (Table IV). Both substances showed clear inhibition against G+-bacteria, whereas no activity was found for G–-strains and yeast. The growth of S. aureus was strongly inhibited in the presence of 3 (MIC 64 µg/mL) compared to 2 (MIC 512 µg/mL), while both 2 and 3 suppressed growth of B. subtilis at the same MICs (i.e. 256 µg/mL). Anti-inflammatory activity of both compounds was tested on inhibition of COX-1 and COX-2, two key enzymes of prostaglandin biosynthesis. In case of COX-2, a slight inhibition of the enzyme activity (12 and 10 %) was observed for 3 at 1 and 10 µg/mL, respectively. Compound 2 showed no effect at any concentration tested (i.e. 1–100 µg/mL). Biological activity assay performed on mammalian cell lines suggested specific modulation of cell physiology. Human epithelial cell line HeLa was treated with 2 and 3. After a 1-d treatment with 2 (10 and ANTHRAQUINONES PRODUCED BY Geosmithia spp. 185 2009 100 µmol/L), actin cytoskeleton was transformed into patches which resembled the effect of latrunculin. Cells rounded up, partly detached, but did not die for at least 2 d. A representative phenotype is shown in Fig. 2A. Treatment (1 d) with compound 3 (10 and 100 µmol/L) led to alterations in the dynamics of the cell cycle. The most obvious phenotype was accumulation of abnormal metaphase–anaphase transition mitotic phase, which is normally only transient and rare. For the results of cell cycle quantification see Table V, for representative alterations of the mitotic spindle and chromosome mislocalization Fig. 2B. Table IV. The antimicrobial activity (MICa, µg/mL) of compounds 2 and 3 and reference antibiotics ciprofloxacin (CIP) and nystatin (NYS) Microorganism 2 3 CIP NYS 256 512 256 64 2 1 – – G+ Bacillus subtilis Staphylococcus aureus G– Escherichia coli Pseudomonas aeruginosa >512 >512 >512 >512 0.03 0.5 – – Yeast Candida albicans >512 >512 – 4 a≥80 %. Fig. 2. Effect of 2 (A) and 3 (B) on HeLa cells; treatment with 100 µmol/L, 1 d, nuclei stained by DAPI; magnification ×40. A: Actin visualized using phalloidin Alexa-567; cells typically round up, detach and form abnormal foci of polymerized actin, resembling the effect of latrunculin. B: tubulin visualized using anti-tubulin Alexa-488; cells produce abnormal mitoses with abundant transitional metaphase–anaphase stage. DISCUSSION The isolation and structural characterization of AQs produced by a submerged culture of G. lavendula revealed a closely related family of AQs exhibiting very similar UV–VIS spectra with clear shift in the region of 200–300 nm. The NMR and MS data of 1 were identical to those previously published for 1,3,6,8-tetrahydroxyanthraquinone isolated from Aspergillus versicolor (Berger 1980), Leptographium spp. (Ayer et al. 1989) and Chaetomium globosum (Wijeratne et al. 2006). Compound 2 is known as 1-acetyl-2,4,5,7-tetrahydroxyanthraquinone (rhodolamprometrin) reported in the crinoid Comatula solaris (Francesconi 1980) and a mutant strain of Trichoderma viride (Betina et al. 1986) as an indicator-like pigment. Both 1 and 2 were also found in extracts from the crinoids Heterometra savi, Lamprometra kluzingeri (Erdman and Thomson 1972), and Oxycomanthus japonicus (Takahashi et al. 2002). There are only two reports concerning acetylated pentahydroxyanthraquinone. The first one is a trace crinoid pigment from Comactinia meridionalis (Rideout and Sutherland 1985). However, its presumable structure of 8-acetyl-1,2,4,5,7-pentahydroxyanthraquinone was suggested based on MS data only. The second report described red anthraquinone, draculone, isolated from the tropical lichen Melanotheca cruenta (Mathey et al. 2002) and determined as a 2-acetyl-1,3,4,6,8-pentahydroxyanthraquinone. 186 E. STODŮLKOVÁ et al. Vol. 54 Inhibition activities (Table IV) are in good agreement with previously published data of various AQs, e.g., catenarin, chrysophanol, physcion, and emodin (Hatano et al. 1999), which exhibited considerable antibacterial activity against B. subtilis and other G+-bacteria. Lunatin and bisanthraquinone cytoskyrin A isolated from Curvularia lunata were active against B. subtilis, S. aureus, and E. coli (Jadulco et al. 2002). Table V. Effect of 1-acetyl-2,4,5,7,8-pentahydroxyanthraquinone (3) on cell cyclea Cell cycle 100 µmol/L 10 µmol/L Control Prophase Metaphase Meta–anaphase hybrid Anaphase Telophase Apoptosis 1.33 2.59 2.79 0.40 1.00 2.99 0.59 1.07 0.06 0.40 1.36 0.47 0.74 0.92 0 0.31 1.23 0.12 Dividing cells, % 8.1 3.5 3.4 aHeLa cells were treated (1 d), stained with DAPI, counted and analyzed by Cell® microscope (for each treatment approximately 1500 nuclei were evaluated); percentages of relevant mitotic phases (including meta–anaphase hybrid stage and apoptosis) scored by typical morphological features. Biological activity of 2 and 3 identified on HeLa cells could be interpreted as another level of functional complexity of Geosmithia-derived pigments. In our case, the effect of 2 resembles the effect of latrunculins or tolytoxin. It is obvious that the AQ can enter the cell and dramatically change its morphology. Direct or indirect effects on actin monomers or polymers, spectrum of affected cell types and phylogenetic distribution of targeted organisms are under investigation. Compound 2 affects cell cycle of HeLa cells by blocking metaphase–anaphase transition; this could be mediated by direct de- or stabilization of the mitotic spindle or via modulation of checkpoint or motor proteins involved particularly in this stage. Unfortunately, there are only very limited number of suitable inhibitors dissecting molecular mechanisms during mitosis to compare with. Those already published do not exhibit an identical phenotype. One of the closest, hesperadin (Aurora B inhibitor), reveals a role for Aurora B in correcting kinetochore–microtubule attachment and in maintaining the spindle assembly checkpoint (Hauf et al. 2003). Another two – with star-like distribution of mitotic spindle and chromosomes – Ispinesib and Monoastrol, inhibit KSP (also known as HsEg5), a kinesin that plays an essential role in the formation of a bipolar mitotic spindle required for cell cycle progression through mitosis (Lad et al. 2008; Mayer et al. 1999). Yet another molecular explanation of the observed effect on metaphase–anaphase transition is inhibition of caspase-related proteinase separase or activation of securin, both responsible for regulated cleavage of cohesin complex (required to maintain cohesion between sister chromatids, from their synthesis in S phase until the end of metaphase). There are no specific low-molar-mass inhibitors of this machinery yet identified. It is tempting to speculate that one of our compounds could affect this particular step in mitotic regulation. The function of AQs in fungi like Geosmithia has not yet been established. This unique mixture of pigments can protect the spores from environmental stress and thus enhance their survival as is known in lichens (Edwards 2004). It could be also useful for the adaptation of these fungi, which are transmitted mostly on insect vector’s surfaces (Kola ík et al. 2008). We also showed that produced AQs are able to inhibit certain organisms, which could be important in the competition with microbes detrimental for the host beetle. Identified AQs could be a part of repellent activity towards host beetle predators. Alterations of actin (stabilization in case of phalloidin or disruption using latrunculins) or tubulin cytoskeleton is used quite often as a poisoning or food disguising strategy (Haefner 2003). Further experiments will be necessary to get an insight into the unique connections and mutual interactions among fungi as producers of secondary compounds like AQs and their natural predators like birds, insects, and microbes. Thanks are due to the Institutional Research Concept AV 0Z502 00510, Academy of Sciences of the Czech Republic, project no. KAN 200 200 651, Ministry of Education, Youth and Sports of the Czech Republic MSM 00216 20828, Centre of Molecular and Cellular Immunology 1M 68378 05001, MSM 00216 20858, and LC 7017, Grant Agency of the Czech Republic 203/05/P575, and the Charles University Grant Agency 205/2004 for their financial contribution. Authors are greatly indebted to R. Jankowiak (Agricultural University of Cracow, Poland) for providing strains RJ0363 and RJ0258. 2009 ANTHRAQUINONES PRODUCED BY Geosmithia spp. 187 REFERENCES AVERY M.L., HUMPHREY J.S., DECKER D.G.: Feeding deterrence of anthraquinone, anthracene, and anthrone to rice-eating birds. J.Wild.Manag. 61, 1359–1365 (1997). AVWIORO O.G., AWOYEMI F.A., ODUOLA T.: A novel natural collagen and muscle stain from Morinda lucida extracts. Internat.Med.J. 4, 44–48 (2005). AYER W.A., BROWNE L.M., LIN G.: Metabolites of Leptographium wageneri, the causative agent of black stain root disease of conifers. J.Nat.Prod. 52, 119–129 (1989). BERGER Y.: 1,3,6,8-Tetrahydroxyanthraquinone from Aspergillus versicolor. Phytochemistry 19, 2779–2780 (1980). BETINA V., SEDMERA P., VOKOUN J., PODOJIL M.: Anthraquinone pigments from a conidiating mutant of Trichoderma viride. Experientia 42, 196–197 (1986). BRUNETON J.: Pharmacognosy, Phytochemistry, Medical Plants, 2nd ed. Intercept Ltd.–Lavoisier Publ., Paris 1999. CHRYSAYI-TOKOUSBALIDES M., KASTANIAS M.A.: Cynodontin: a fungal metabolite with antifungal properties. J.Agric.Food Chem. 51, 4920–4923 (2003). EDWARDS H.G.M., COCKELL C.S., NEWTON E.M., WYNN-WILLIAMS D.D.: Protective pigmentation in UVB-screened Antarctic lichens studied by Fourier transform Raman spectroscopy: an extremophile bioresponse to radiation stress. J.Raman Spectrosc. 35, 463–469 (2004). ERDMAN T.R., THOMSON R.H.: Naturally occurring quinones. Anthraquinones in crinoids Heteromera savignii and Lamprometra kluzingeri. J.Chem.Soc.Perkin Trans. I, 1291–1292 (1972). FRANCESCONI K.A.: Pigments of some echinoderms collected from Western Australian waters. Austral.J.Chem. 33, 2781–2784 (1980). GÁLVEZ J., GOMEZ-LECHÓN M.J., GARCÍA-DOMENECH R., CASTELL J.V.: New cytostatic agents obtained by molecular topology. Bioorg.Med.Chem.Lett. 6, 2301–2306 (1996). GANAPATY S., THOMAS P.S., FOTSO S., LAATSCH H.: Antitermitic quinones from Diospyros sylvatica. Phytochemistry 65, 1265–1271 (2004). GILL M.: The biosynthesis of pigments in Basidiomycetes. Austral.J.Chem. 54, 721–734 (2001). HAEFNER B.: Drugs from the deep: marine natural products as drug candidates. Drug Discov.Today 536–544 (2003). HATANO T., UEBAYASHI H., ITO H., SHIOTA S., TSUCHIYA T., YOSHIDA T.: Phenolic constituents of Cassia seeds and antibacterial effects of some naphthalenes and anthraquinones on methicillin-resistant Staphylococcus aureus. Chem.Pharm.Bull. 47, 1121–1127 (1999). HAUF S., COLE R.W., LATERRA S., ZIMMER C., SCHNAPP G., WALTER R., HECKEL A., VAN MEEL J., RIEDER C.L., PETERS J.M.: The small molecule hesperadin reveals a role for Aurora B in correcting kinetochore–microtubule attachment and in maintaining the spindle assembly checkpoint. J.Cell Biol. 161, 281–294 (2003). HILKER M., KÖPF A.: Evaluation of the palatability of chrysomelid larvae containing anthraquinones to birds. Oecologia 100, 421–429 (1994). HOBSON D.K., WALES D.S.: “Green” dyes. J.Soc.Dyers Colour. 114, 42–44 (1998). ISMAIL N.H., ALI A.M., AIMI N., KITAJIMA M., TAKAYAMA H., LAJIA N.H.: Anthraquinone from Morinda elliptica. Phytochemistry 45, 1723–1725 (1997). IZHAKI I.: Emodin – a secondary metabolite with multiple ecological functions in higher plants. New Phytol. 155, 205–217 (2002). JADULCO R., BRAUERS G., EDRADA R.A., EBEL R., WRAY V., SUDARSONO, PROKSCH P.: New metabolites from sponge-derived fungi Curvularia lunata and Cladosporium herbarum. J.Nat.Prod. 65, 730–733 (2002). KOLA ÍK M., KOSTOVČÍK M., PAŽOUTOVÁ S.: Host range and diversity of the genus Geosmithia (Ascomycota:Hypocreales) living in association with bark beetles in the Mediterranean area. Mycol.Res. 111, 1298–1310 (2007). KOLA ÍK M., KUBÁTOVÁ A., HULCR J., PAŽOUTOVÁ S.: Geosmithia fungi are highly diverse and consistent bark beetle associates: evidence from their community structure in temperate Europe. Microb.Ecol. 55, 65–80 (2008). LAD L., LUO L., CARSON J.D., WOOD K.W., HARTMAN J.J., COPELAND R.A., SAKOWICZ R.: Mechanism of inhibition of human KSP by ispinesib. Biochemistry 47, 3576–3585 (2008). MATHEY A.P., SPITELLERA P., STEGLICHA W.: Draculone, a new anthraquinone pigment from the tropical lichen Melanotheca cruenta. Z.Naturforsch. 57, 565–567 (2002). MAYER T.U., KAPOOR T.M., HAGGARTY S.J., KING R.W., SCHREIBER S.L., MITCHISON T.J.: Small molecule inhibitor of mitotic spindle bipolarity identified in a phenotype-based screen. Science 286, 971–974 (1999). MÜLLER K.: Current status and recent developments in anthracenone antipsoriatics. Curr.Pharm.Design 6, 901–918 (2000). NOREEN Y., RINGBOM T., PERERA P., DANIELSON H., BOHLIN L.: Development of a radiochemical cyclooxygenase-1 and -2 in vitro assay for identification of natural products as inhibitors of prostaglandin biosynthesis. J.Nat.Prod. 61, 2–7 (1998). PANKEWITZ F., HILKER M.: Defensive components in insect eggs: are anthraquinones produced during egg development? J.Chem.Ecol. 32, 2067–2072 (2006). RATH G.M., NDONZAO M., HOSTETTMANN K.: Antifungal anthraquinones from Morinda lucida. Internat.J.Pharmacogn. 33, 107–114 (1995). RIDEOUT J.A., SUTHERLAND M.D.: Pigments of marine animals. XV. Bianthrones and related polyketides from Lamprometra palmata gyges and other species of crinoids. Austral.J.Chem. 38, 793–808 (1985). SCHINAZI R.F., CHU C.K., BABU J.R., OSWALD B.J., SAALMANN V., CANNON D.L., ERIKSSON B.F.H., NASR M.: Anthraquinones as a new class of antiviral agents against human immunodeficiency virus. Antiviral Res. 13, 265—272 (1990). SEMPLE S.J., PYKE S.M., REYNOLDS G.D., FLOWER R.L.P.: In vitro antiviral activity of the anthraquinone chrysophanic acid against poliovirus. Antiviral Res. 49, 169–178 (2001). SITTIE A.A., LEMMICH E., OLSEN C.E., HVIID L., KHARAZMI A., NKRUMAH F.K., CHRISTENSEN S.B.: Structure–activity studies: in vitro antileishmanial and antimalarial activities of anthraquinones from Morinda lucida. Planta Med. 65, 259–261 (1999). SIX D.L.: Bark beetle–fungus symbiosis, pp. 99–116 in K. Bourtzis, T.A. Miller (Eds): Insect Symbiosis. CRC Press, Baton Rouge (FL, USA) 2003. TAKAHASHI D., MAOKA T., TSUSHIMA M., FUJITANI K., KOZUKA M., MATSUNO T., SHINGU T.: New quinone sulfates from the crinoids Tropiometra afra macrodiscus and Oxycomanthus japonicus. Chem.Pharm.Bull. 50, 1609–1612 (2002). THOMSON R.H.: Naturally Occurring Quinones. IV. Recent Advances. Blackie Academic and Professional, New York 1997. TRIPATHI Y.S., SHARMA M., MANICKAM M.: Rubiadin, a new antioxidant from Rubia cordifolia. Indian J.Biochem.Biophys. 34, 302– 306 (1997). WIJERATNE E.M.K., TURBYVILLE T.J., FRITZ A., WHITESELL L., GUNATILAKA A.A.L.: A new dihydroxanthenone from a plant-associated strain of the fungus Chaetomium globosum demonstrates anticancer activity. Bioorgan.Med.Chem. 14, 7917–7923 (2006). YOUNOS C., ROLLAND A., FLEURENTIN J., LANHERS M., MISSLIN R., MARTIER F.: Analgetic and behavioral effects of Morinda citrofolia. Planta Med. 56, 430–434 (1990).