Cyclobotryoxide, a Phytotoxic Metabolite Produced by thePlurivorous Pathogen Neofusicoccum australeAnna Andolfi,† Lucia Maddau,‡ Alessio Cimmino,† Benedetto T. Linaldeddu,‡ Antonio Franceschini,‡
Salvatorica Serra,‡ Sara Basso,† Dominique Melck,§ and Antonio Evidente*,†
†Dipartimento di Scienze del Suolo, della Pianta, dell’Ambiente e delle Produzioni Animali, Universita di Napoli Federico II, ViaUniversita 100, 80055 Portici, Italy‡Dipartimento di Agraria, Sezione di Patologia Vegetale ed Entomologia, Universita degli Studi di Sassari, Via E. De Nicola 1, 07100Sassari, Italy§Istituto di Chimica Biomolecolare del CNR, Comprensorio Olivetti, Edificio 70, Via Campi Flegrei 34, 80078 Pozzuoli, Italy
ABSTRACT: Two isolates of Neofusicoccum australe belonging to ITS haplotypesH4 and H1 and associated with grapevine cordon dieback and branch dieback ofPhoenicean juniper, respectively, have been shown to produce in vitro structurallydifferent secondary metabolites. From the strain BOT48 of N. australe (haplotypeH4) a new cyclohexenone oxide, namely, cyclobotryoxide, was isolated togetherwith 3-methylcatechol and tyrosol. Cyclobotryoxide was characterized as(1S,5R,6S)-5-hydroxy-3-methoxy-4-methyl-7-oxabicyclo[4.1.0]hept-3-en-2-one by spectroscopic, optical, and chemical methods.The strain BL24 (haplotype H1) produced tyrosol along with botryosphaerone D and (3S,4S)-3,4,8-trihydroxy-6-methoxy-3,4-dihydro-1(2H)-naphthalenone. The metabolites obtained from both strains were tested at four concentrations on leaves ofgrapevine cv. Cannonau, holm oak, and cork oak by the leaf puncture assay. Cyclobotryoxide proved to be the most phytotoxiccompound. Tyrosol and cyclobotryoxide were also tested on detached grapevine leaves at concentrations of 0.25 and 0.5 mg/mL.Only cyclobotryoxide was found to be active in this bioassay.
originally thought to be native to the southern hemisphere,1,2
has recently been reported on a variety of native and exotichosts worldwide.3−7 In Italy N. australe was primarily foundassociated with drupe rot of olives8 and declining grapevineplants.9 Phylogenetically a one bp difference in the sequence ofthe ITS region of rDNA was observed between the Italianisolates of N. australe and the ex-type isolate (CMW6837) usedin the original description.8,9 Recently Sakalidis et al.,10 re-evaluating GenBank ITS rDNA sequence data of N. australe,found the occurrence of 13 ITS haplotypes for this fungalspecies, of which two were dominant (H1 and H4), while theothers were more rare. On the basis of the findings of thisextensive study, Italian strains of N. australe fit within thehaplotype H4.However, in a study still in progress aimed at clarifying the
causes of decline affecting Juniperus spp. in several natural areasin Sardinia (Italy), a large collection of N. australe strains wereisolated from juniper trees showing a progressive dieback ofshoots and branches. Phylogenetically most of the juniper’sstrains of N. australe fit mainly within the haplotype H4, butsome strains belong to the haplotype H1. These findingssuggest that N. australe is part of the endemic mycobiota ofnative tree species in Sardinia, in addition to being a pathogenof agricultural crops.Pathogenicity tests conducted on grapevine and juniper
plants using Sardinian strains of N. australe belonging to bothH4 and H1 haplotypes have revealed the occurrence of
significant variations in virulence among the strains of thisplurivorous pathogen. The nature of symptoms caused by N.australe on grapevine and Phoenicenan juniper (Figure 1A andB) suggests that phytotoxic metabolites may be involved in thehost−pathogen interaction. On the other hand, it is well-knownthat Botryosphaeriaceae species are able to produce in vitro aplethora of bioactive secondary metabolites, some of which maycontribute to the onset of disease symptoms.11−15 In particular,Botryosphaeriaceae species associated with declining grapevineare able to produce hydrophilic and lipophilic bioactivemetabolites in liquid culture.16 So far, phytotoxins have beenisolated from Neofusicoccum parvum17 and Diplodia seriata,18
although their role on symptom disease expression needs to befully elucidated.While new polyketides along with other known secondary
metabolites were isolated from one isolate of N. australebelonging to haplotype H12,7 no data on the production ofphytotoxic metabolites by others N. australe haplotypes havebeen published. Given that significant variation in the virulencebetween the isolates was observed, it is reasonable to assumethat some variation on the secondary metabolite profiling couldoccur among isolates of this species. Thus, the objective of thisresearch was to isolate and characterize the phytotoxicmetabolites produced in vitro by two strains of N. australebelonging to H1 and H4 haplotypes isolated from Phoeniceanjuniper and grapevine, respectively.
■ RESULTS AND DISCUSSIONThe two strains of N. australe, belonging to haplotypes H4 andH1, were fermented in liquid Cazapek medium containing 2%corn meal. The cultures were exhaustively extracted withEtOAc, and the organic extracts were purified by combinedcolumn and TLC chromatography as detailed in theExperimental Section, yielding five metabolites (1, 3−6, Figure2). Metabolites 1, 3, and 4 were produced by haplotype H4,while metabolites 4, 5, and 6 were identified for haplotype H1.All the metabolites were isolated as homogeneous amorphoussolids and by preliminary 1H NMR investigation appeared tobelong to different classes of natural compounds.The compound named cyclobotryoxide (1) had a molecular
weight of 170 as deduced from its HRESIMS spectrum,
corresponding to a molecular formula of C8H10O4, indicatingfour hydrogen deficiencies. The same spectrum showed thesodium cluster [M + Na]+ at m/z 193.1509, and when the samespectrum was recorded in negative mode, the pseudomolecularion [M − H]− at m/z 169 appeared. The UV spectrum showedan absorption maximum at 275 nm, consistent with thetheoretical one (272) calculated applying the Woodward rulesfor compound containing a suitable α,β,γ-trisubstituted α,β-unsaturated carbonyl group.19 The IR spectrum showed bandscharacteristic of hydroxy and conjugated carbonyl groups.20
The 1H NMR spectrum of 1 (Table 1) showed the presenceof a doublet of doublets (J = 3.0 and 1.0 Hz) and a doublet (J =
3.0 Hz) at δ 3.79 and 3.45, respectively, typical of the protons(H-6 and H-1) of a cis-disubstituted oxirane ring.19 Theresonance at δ 4.88 was assigned to a proton (H-5) of asecondary hydroxylated carbon. It coupled in the COSYspectrum21 with the geminal hydroxy group observed as adoublet (J = 11.0 Hz) at δ 3.60, with H-6 and long-range (J < 1
Figure 1. Grapevine and juniper diseases symptoms associated withNeofusicoccum australe strain BOT48 (A) and BL24 (B), respectively.
Figure 2. Structures of cyclobotryoxide and its acetyl derivative, 3-methylcatechol, tyrosol, botryosphaerone D, and (3S,4S)-3,4,8-trihydroxy-6-methoxy-3,4-dihydro-1(2H)-naphthalenone (1−6) pro-duced by N. australe from grapevine and juniper.
Table 1. 1H and 13C NMR Data and HMBC Correlations forCyclobotryoxide (1) in CDCl3
aThe chemical shifts are in δ values (ppm) from TMS. b2D 1H,1H(COSY) and 13C,1H (HSQC) NMR experiments delineated thecorrelations of all the protons and the corresponding carbons.cMultiplicities were assigned by DEPT spectrum. dThe shift isvariable; this value was observed in CDCl3 at room temperature.
Journal of Natural Products Article
dx.doi.org/10.1021/np300512m | J. Nat. Prod. 2012, 75, 1785−17911786
Hz) with a vinyl methyl group [CH3−C(4)]. The latter wasobserved as a broad singlet at δ 1.26. Furthermore, the 1HNMR spectrum showed a singlet at δ 4.05 assigned to amethoxy group.19
These partial structures were confirmed by data of the 13CNMR spectrum (Table 1), which showed the presences ofsignals due to an α,β-unsaturated α-oxygenated carbonyl group(δ 194.3, 165.5, 112.0), for the carbonyl (C-2) and twoquaternary olefinic (C-3 and C-4) carbons. The methoxy andmethyl groups resonated at δ 56.1 and 7.9 and were located atC-3 and C-4, respectively, on the basis of the couplingsobserved in the HMBC spectrum.21 The three oxygenatedsecondary carbons were observed at typical chemical shiftvalues of δ 60.9, 54.2, and 52.1, and they were assigned by thecoupling observed in the HSQC spectrum21 to C-5, C-6, andC-1, respectively.22
Thus, chemical shifts were assigned to all the protons andcarbons, and cyclobotryoxide was formulated as 5-hydroxy-3-methoxy-4-methyl-7-oxabicyclo[4.1.0]hept-3-en-2-one.This structure was confirmed by all the couplings and
correlations observed in the HMBC and NOESY21 spectra. TheNOESY spectrum showed, besides the expected correlationsbetween H-1 and H-6 and the methoxy and the methyl groups,also those between the latter and the proton of the hydroxygroup at C-5 and between H-5 and H-6. The latter correlationwas in agreement with an inspection of a Dreiding model of 1,as both relative configurations of C-5 (R* or S*) could generatean NOE effect between H-6 and H-5. However, only therelative R* configuration could be assigned to C-5, as it isconsistent with the dihedral angle near 90° observed betweenH-5 and H-6, which, in agreement with the literature, had avery low value (1.0 Hz) measured for their coupling in the 1HNMR spectrum.19
The absolute configuration of 1 was determined bycomparison of its ECD spectrum with those of the structurallyrelated epoxycyclohexenones.23 In fact, its ECD spectrum(Figure 3A) showed negative and positive Cotton effects at 310and 276 nm, respectively. These effects were similar to relatedepoxycyclohexenones such as sphaeropsidone, terremutin, andpanepoxydon.11,23
On the basis of these results 1 was formulated as (1S,5R,6S)-5-hydroxy-3-methoxy-4-methyl-7-oxabicyclo[4.1.0]hept-3-en-2-one.The structure assigned to cyclobotryoxide was further
supported by preparation of derivative 2 by acetylation. Thespectroscopic data of this derivative were fully consistent withthe structure assigned to 1. The IR spectrum of 2 showed theabsence of a hydroxy group and the presence of a band typicalof an acetyl carbonyl group at 1747 cm−1.20 Its 1H NMRspectrum differed from that of 1 essentially in the downfieldshift (Δ δ 1.30) of H-5 observed as a broad singlet at δ 6.18 andthe absence of the doublet at δ 3.60 of the hydroxy proton andthe presence of the singlet of the acetyl group at δ 2.19. TheESIMS spectrum of 2 recorded in the positive mode showedthe sodium cluster [M + Na]+ at m/z 235.Metabolite 3 showed a molecular weight of 124, as deduced
from its ESIMS spectrum recorded in the positive and negativemodes, and was identified as 3-methylcatechol on the basis ofits spectroscopic data, which were the same as previouslyreported.24 3-Methylcatechol, a minor metabolite of toluene,showed toxicity in human and mammal cells.25−27 It was alsoisolated, together with other phenolic compounds, from the
fermentation liquid of the endophytic fungus Penicillium sp.GT6105 from Kandelia candel.28
Metabolite 4 was isolated from the organic extracts of bothstrains of N. australe and was identified as tyrosol, on the basisof its 1H NMR spectrum, which was similar to that previouslyreported.29 The ESIMS spectrum showed the sodium cluster[M + Na]+ and the pseudomolecular ion [M − H]− at m/z 161and 137, respectively, when it was recorded in the positive andnegative modes. Tyrosol is a phytotoxic metabolite producedby plants and fungi including Botryosphaeriaceae species.15,29
Recently, tyrosol was also identified as a phytotoxic metaboliteproduced by N. parvum.17 This fungus was isolated togetherwith four other Botryosphaeriaceae from declining grapevinesin Spain.16
From the organic extract of N. australe isolated from juniper,metabolites 5 and 6 were obtained. The molecular weights of 5and 6 were determined to be 252 and 224, respectively, asdeduced from their ESI and APCIMS spectra recorded,respectively, in the positive and negative modes.The chemical shifts and multiplicities of all protons and
corresponding carbons of 5 and 6 were assigned by thecouplings observed in their COSY, HMBC, and HSQC spectra.Comparison of these and reported data identified 5 and 6 asbotryosphaerone D7 and (3S,4S)-3,4,8-trihydroxy-6-methoxy-3,4-dihydro-1(2H)-naphthalenone.30
The absolute configuration of botryosphaerone D (5), notpreviously reported, was also determined. Its ECD spectrum(Figure 3B) was recorded under the same conditions andcompared to that of 6.30 On the basis of coincident negativeand positive Cotton effects at 214, and 239 and 280 nm,
Figure 3. ECD spectra of cyclobotryoxide (A) and botryosphaerone D(B) recorded in EtOH.
Journal of Natural Products Article
dx.doi.org/10.1021/np300512m | J. Nat. Prod. 2012, 75, 1785−17911787
compound 5 was formulated as (3S,4S)-7-ethyl-3,4,8-trihy-droxy-6-methoxy-3,4-dihydro-1(2H)-naphthalenone. Botryos-phaerone D (5), together with three other new naphthalenonepolyketides, had previously been isolated from N. australe strainZJ12-1A, a fungus obtained from sterilized root epidermis ofSonneratia apetala in China. This compound did not show anyantimicrobial and cytotoxic activities.7 Finally, (3S,4S)-3,4,8-trihydroxy-6-methoxy-3,4-dihydro-1(2H)-naphthalenone (6)was reported as a new naphthalenone produced from wood-decaying fungus Phaeosphaeria sp.30
Cyclobotryoxide was characterized as a new naturallyoccurring compound belonging to a family of epoxycyclohex-enones that have been isolated from bacteria, fungi, higherplants, and mollusks. These compounds have shown interestingbiological properties such as antifungal, antibacterial, antitu-moral, phytotoxic, and enzyme-inhibitory activities.23 Further-more, sphaeropsidone and epi-sphaeropsidone, belonging to thesame family of natural compounds, were previously isolatedfrom Diplodia cupressi,31 a pathogenic fungus of cypress.The isolation of naphthalenone polyketides as 5 and 6 was
not a surprise. In fact, they belong to a group of well-knownphytotoxins produced by pathogenic fungi for different agrariancrops such as scytalone, isosclerone, botrytone, and cis- andtrans-3,4dihydro-2,4,8-trihydroxynaphthalenones.32,33
When assayed on the leaves of different plant species atdifferent concentrations, compounds 1 and 3−6 showedvarying degrees of phytotoxicity. The toxicity data obtainedin the leaf-puncture bioassay on holm oak, cork oak, andgrapevine cv. Cannonau leaves are summarized in Table 2.Cyclobotryoxide (1) appeared to be the most active metabolite.It had remarkable toxicity in a range of concentrations from0.25 to 1.0 mg/mL, causing necrotic lesions on leaves of allspecies tested. At the lower concentrations its toxicity wasprogressively reduced. A marked decrease of activity wasobserved for its acetyl derivative (2) even at the highestconcentration (1 mg/mL) tested, confirming that structuralmodifications could alter biological activity. At the lower doses,compound 2 was inactive on cork and holm oak. Compound 3
showed moderate activity on grape leaves, while on the otherspecies tested it was weakly active or inactive. Compound 4,produced by both N. australe isolates, exhibited moderatephytotoxicity on grapevine leaves, whereas on the other speciestested it was weakly active or inactive. Compound 5, isolated asthe main metabolite of haplotype H4, appeared to be much lesstoxic than 1 in this bioassay, even at the highest concentration.In addition, compound 6 did not show any phytotoxicity either.Compounds 1 and 4 were also tested on grapevine detached
leaves at 0.25 and 0.5 mg/mL. In this bioassay cyclobotryoxidecaused symptoms at 0.5 mg/mL on the foliar lamina consistingof irregular internerval or marginal necrotic spots, slowlybecoming coalescent, followed by distortion and withering ofthe leaf lamina. No visible symptoms were induced by tyrosol.The present study provides the first report on the expression
of phytotoxic secondary metabolites by two N. australe isolatesbelonging to two of the 13 haplotypes of this species proposedby Sakadilis et al.10 The metabolites characterized emphasizeddifferent metabolic profiles of the two N. australe haplotypes(H1 and H4) examined. Such diversity is also reflected in thetoxic potential shown by each of the metabolites isolated. Thisis not surprising, considering that differences in aggressivenesswithin N. australe isolates correlate with genetic variability aspreviously reported.34,35 A different degree of virulence hasbeen observed for fungal pathogens, and in some cases thesedifferences were correlated with their different capacity toproduce virulence factors involved, fully or in part, in thesymptom expression.36−38
An aspect worthy of discussion concerns the nature of thesymptoms induced by N. australe on grapevine. To date, fieldobservations have shown that this pathogen is chiefly involvedin the etiology of canker and dieback of grapevine trunk andcordon. Therefore, it is interesting to understand if thephytotoxins secreted by this pathogen could also be involvedin the onset of foliar symptoms on the host plant as alreadyreported for other grapevine pathogens. It is likely that thesymptomatology associated with infection of N. australe ongrapevine has not yet been clarified. In this regard, our research
Table 2. Phytotoxicity of Compounds 1−6 Evaluated at Different Concentrations on Three Plant Species [Area Lesions (mm2)± SE]
6 cork oak n.a.a n.a. n.a. n.a.holm oak n.a. n.a. n.a. n.a.grapevine n.a. n.a. n.a. n.a.
an.a. = no activity.
Journal of Natural Products Article
dx.doi.org/10.1021/np300512m | J. Nat. Prod. 2012, 75, 1785−17911788
group has planned further investigations aimed at clarifying thistopic. Furthermore, the current work contributes to expandingthe knowledge on bioactive metabolites produced byBotryosphaeriaceae. On the other hand, it is worth pointingout that most of the metabolites so far isolated fromBotryosphaeriaceae associated with grapevine trunk diseasedo not seem to be structurally correlated with those producedby Eutypa lata. This information, previously pointed out byMahoney et al.,39 could allow Eutypa dieback to bediscriminated from other grapevine diseases. In addition, newscenarios of investigation could be opened up for tyrosol, ametabolite frequently isolated from species of Botryosphaer-iaceae. Is there a strategy of regulation of virulence factors inBotryosphaeriaceae as also reported for other fungi?40 If so, istyrosol a quorum-sensing molecule as observed in Candidaalbicans?41 This is a new challenge for the future.
■ EXPERIMENTAL SECTIONGeneral Experimental Procedures. Optical rotations were
measured in CHCl3, unless otherwise noted, on a Jasco polarimeter,while the ECD spectra were recorded on a JASCO J-815 spectrometerin EtOH; IR spectra were recorded as glassy films on a Perkin-ElmerSpectrum One FT-IR spectrometer, and UV spectra were taken inMeOH solution on a Perkin-Elmer Lambda 25 UV/vis spectropho-tometer; 1H and 13C NMR spectra were recorded at 600 and at 125MHz, respectively, in CDCl3, unless otherwise noted, on Brukerspectrometers. The same solvent was used as internal standard. DEPT,COSY-45, HSQC, HMBC, and NOESY experiments21 wereperformed using standard Bruker microprograms. HRESIMS spectrawere recorded on a Waters Micromass Q-TOF Micro; ESI andAPICMS spectra were recorded on an Agilent Technologies 6120Quadrupole LC/MS instrument. Analytical and preparative TLC wereperformed on silica gel (Merck, Kieselgel 60, F254, 0.25 and 0.5 mm,respectively) or reversed-phase (Whatman, KC18 F254, 0.20 mm)plates. The spots were visualized by exposure to UV radiation (253) orby spraying first with 10% H2SO4 in MeOH and then with 5%phosphomolybdic acid in EtOH, followed by heating at 110 °C for 10min. Column chromatography was performed on a silica gel column(Merck, Kieselgel 60, 0.063−0.200 mm).Fungal Strains, Culture Medium, and Growth Conditions.
Two strains of N. australe were examined in this study. The strainBL24 of N. australe (haplotype H1) was originally isolated from asymptomatic branch of Phoenicean juniper collected in a natural areaon Caprera Island (Italy). The strain BOT48 of N. australe (haplotypeH4) was isolated from wood tissues of symptomatic grapevine cordoncollected in a vineyard in northern Sardinia.9 Both strains wereidentified on the basis of morphological characters and analysis ofinternal transcribed spacer (ITS) rDNA. Fungal DNA extraction, PCRamplification reactions, and DNA sequencing were carried out asreported by Linaldeddu et al.9
Representative sequences of both isolates were deposited inGenBank: BL24 (ITS: accession number JX312356); BOT48 (ITS:accession number HQ011406).Pure cultures of both strains were maintained on potato-dextrose
agar (Fluka, Sigma-Aldrich Chemic GmbH, Buchs, Switzerland) andstored at 4 °C in the collection of the Dipartimento di Agraria, Sezionedi Patologia Vegetale ed Entomologia, University of Sassari, Italy.Extraction and Purification of Phytotoxins from N. australe
Strain BOT48. The fungus was grown in 2 L Erlenmeyer flaskscontaining 400 mL of Czapek medium amended with corn meal (pH5.7). Each flask was seeded with 5 mL of a mycelia suspension andthen incubated at 25 °C for 4 weeks in darkness. The culture filtrates(5.7 L) were acidified to pH 4 with 2 M HCl and extractedexhaustively with EtOAc. The organic extracts were combined, driedwith Na2SO4, and evaporated under reduced pressure to give a brown-red oil residue (4 g), having a high phytotoxic activity on grapevineleaves by the puncture leaf bioassay at 4 mg/mL. The residue wassubmitted to a bioassay-guided fractionation through column
chromatography on silica gel, eluted with CHCl3/i-PrOH (95:5, v/v). Eleven homogeneous fraction groups were collected and screenedfor their phytotoxic activity. The residue (154.1 mg) of the fifthfraction was further purified by CC on silica gel, eluted with petroleumether/Me2CO (7:3). The residue (38.4 mg) of the fifth fraction wasfurther purified by TLC on silica gel, eluent n-hexane/Me2CO (7:3),yielding a homogeneous, amorphous solid [1, 23.7 mg, Rf 0.53, eluentn-hexane/Me2CO (7:3), Rf 0.50, eluent CHCl3/i-PrOH (95:5)],which was named cyclobotryoxide. The residue (100.7 mg) of the sixthfraction from the initial column was further purified by CC on silicagel, eluted with CHCl3/MeOH (97:3). The residue (40.8 mg) of thethird fraction was purified by TLC on silica gel, eluent n-hexane/Me2CO (7:3), yielding a further quantity of 1 (8.7 mg, total 5.7 mg/L). The residue (20.1 mg) of the fourth fraction was purified by twosuccessive TLC steps on silica gel, eluted with CHCl3/i-PrOH (95:5)and EtOAc/n-hexane (55:45), yielding a homogeneous, amorphoussolid identified, as below reported, as 3-methylcatechol [3, 4.4 mg, 0.8mg/L, Rf 0.50, eluent CHCl3/i-PrOH (92:8), Rf 0.61, eluent EtOAc/n-hexane (55:45)]. The residue (86.9 mg) of the eighth fraction of theinitial column was purified by two successive steps using CC and TLCon silica gel, both eluted with CHCl3/i-PrOH (95:5), yielding ahomogeneous, amorphous solid that was identified as tyrosol [4, 1.7mg, 0.31 mg/L, Rf 0.23, eluent CHCl3/i-PrOH (95:5), Rf 0.48, eluentEtOAc/n-hexane (6:4)].
−1 3346, 1712, 1640, 1613, 1462, 1243; 1H and 13C NMR,Table 1; HRESIMS (+) spectrum m/z 193.1509 [C8H10O4Na, calcd193.1518, M + Na]+; ESI (−) m/z 169 [M + H]−.
3-Methylcathecol (3): UV λmax nm (log ε) 292 (3.59); IR νmaxcm−1 3318, 1503, 1201; 1H and 13C NMR were similar to thosereported;24 ESIMS (+) m/z 147 [M + Na]+; ESIMS (−) m/z 123 [M− H]−.
Tyrosol (4): 1H NMR was similar to data reported;29 ESIMS (+)m/z 161 [M + Na]+; ESIMS (−) m/z 137 [M − H]−.
Acetylation of Cyclobotryoxide (2). Cyclobotryoxide (1, 5.0mg), dissolved in pyridine (30 μL), was converted into thecorresponding 5-O-acetyl derivative by acetylation with Ac2O (30μL) at room temperature for 10 min. The reaction was stopped byaddition of MeOH, and an azeotrope formed by addition of benzenewas evaporated in a N2 steam. The oily residue (6.0 mg) was purifiedby TLC on silica gel, eluent n-hexane/Me2CO (6:4), yieldingderivative 2 as a homogeneous solid (4.1 mg, Rf 0.39). 2: UV λmaxnm (log ε) 272 (3.98); IR νmax cm
Extraction and Purification of Phytotoxins by N. australeStrain BL24. The culture filtrate (7 L) of this strain, grown in thesame cultural conditions of haplotyte H4, was acidified to pH 4 with 2M HCl and then exhaustively extracted with EtOAc. The organicextracts were combined, dried (Na2SO4), and evaporated underreduced pressure to yield a brown solid residue (400 mg). The residuewas tested at concentrations of 4 mg/mL as described below, wasfound to be phytotoxic against grapevine leaves, and was submitted tobioassay-guided fractionation through column chromatography onsilica gel, eluted with CHCl3/i-PrOH (95:5). Ten homogeneousfraction groups were collected and screened for their phytotoxicactivity. The residue (11.3 mg) of the sixth fraction was purified byTLC on silica gel, eluent CHCl3/i-PrOH (95:5), yielding ahomogeneous, amorphous solid identified as tyrosol [4, 2.8 mg, 0.5mg/L Rf 0.23, eluent CHCl3/i-PrOH (95:5), Rf 0.48, eluent EtOAc/n-hexane (6:4)]. The residue (20.6 mg) of the seventh fraction of thefirst CC was further purified by two successive steps by TLC on silicagel, eluent CHCl3/i-PrOH (95:5), and by reversed-phase TLC, eluentEtOH/H2O (6:4), yielding two homogeneous, amorphous solidsidentified as botryosphaerone D [5, mg 4.8, 0.7 mg/L Rf 0.24, eluentCHCl3/i-PrOH (95:5), Rf 0.56, eluent EtOH/H2O (6:4)] and(3S,4S)-3,4,8-trihydroxy-6-methoxy-3,4-dihydro-1(2H)-naphthalenone
Journal of Natural Products Article
dx.doi.org/10.1021/np300512m | J. Nat. Prod. 2012, 75, 1785−17911789
−1 3494, 3405, 2958,2927, 2855, 1729, 1636, 1578, 1459, 1432, 1403, 1375, 1360, 1291,1164, 1128, 1075, 1031, 843); 1H and 13C NMR data were very similarto those already reported;30 ESIMS (−) m/z 223 [M − H]−, 207 [M− OH]−; APCIMS (+) m/z 225 [M + H]+.Phytotoxic Activity. The organic extracts and the pure
compounds were assayed on holm oak (Quercus ilex L.), cork oak(Q. suber L.), and grapevine leaves (Vitis vinifera L.) by leaf punctureassay. The organic extracts were tested at 1, 2, and 4 mg/mL. Purecompounds (1−6) were assayed at 1, 0.5, 0.25, and 0.125 mg/mL.Extracts, fractions, and compounds were dissolved in MeOH anddiluted with distilled H2O up to the assay concentrations (the finalcontent of MeOH was 4%). Droplets of the test solutions (20 μL)were applied on the axial side of leaves that had previously been needlepunctured. Droplets (20 μL) of MeOH in distilled H2O (4%) wereapplied on leaves as control. Each treatment was repeated twice. Theleaves were then kept in a moist chamber to prevent the droplets fromdrying. Leaves were observed daily and scored for symptoms after 1week. The effect of the toxins on the leaves, consisting of necroticspots surrounding the puncture, was observed up to 15 days. Lesionswere estimated using APS Assess 2.0 software42 following the tutorialsin the user’s manual. The lesion size was expressed in mm2.Cyclobotryoxide (1) and tyrosol (4) were also tested on detached
leaves of grapevine cv. Cannonau (red vine). The leaves with theirpetioles were immersed in 1 mL of toxic solution until its completeabsorption and then transferred to distilled H2O. Toxicity symptomswere recorded 48 h later. Both compounds were tested at 0.25 and 0.5mg/mL.
NotesThe authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
The NMR spectra were recorded in the laboratory of ChimicaBiomolecolare del CNR, Pozzuoli. This work was supported inpart by grants from Fondazione Banco di Sardegna, ParcoNazionale dell'Arcipelago di La Maddalena and Italian Ministryof University and Research. Contribution DISSPAPA N. 273.
■ REFERENCES(1) Slipper, B.; Fourie, G.; Crous, P. W.; Coutinho, T. A.; Wingfield,B. D.; Wingfield, M. J. Mycologia 2004, 96, 1030−1104.(2) Burgess, T. I.; Sakalidis, M. L.; Hardy, G. S. J. Aust. Ecol. 2006, 31,559−566.(3) Armengol, J.; Gramaje, D.; Perez-Sierra, A.; Landeras, E.;Alzugaray, R.; Martos, S.; Luque, J. Plant Dis. 2008, 92, 980.
(4) Amponsah, N. T.; Jones, E. E.; Ridgway, H. J.; Jaspers, M. V. Aust.Plant Dis. Notes 2009, 4, 6−8.(5) Espinoza, J. G.; Briceno, E. X.; Chavez, E. R.; Urbez-Torres, J. R.;Latorre, B. A. Plant Dis. 2009, 93, 1187−1194.(6) Inderbitzin, P.; Bostock, R. M.; Trouillas, F. P.; Michailides, T. J.Mycologia 2010, 102, 1350−1368.(7) Xu, Y. X.; Lu, C. H.; Zheng, Z. H.; Shen, Y. N. Helv. Chim. Acta2011, 94, 897−902.(8) Lazzizera, C.; Frisullo, S.; Alves, A.; Phillips, A. J. L. Plant Pathol.2008, 57, 948−956.(9) Linaldeddu, B. T.; Scanu, B.; Schiaffino, A.; Serra, S. Phytopathol.Mediterr. 2010, 49, 417−420.(10) Sakalidis, M. L.; Hardy, G. E. J., St.; Burgess, T. I. Australas.Plant Pathol. 2011, 40, 510−521.(11) Evidente, A.; Sparapano, L.; Fierro, O.; Bruno, G.; Giordano, F.;Motta, A. Phytochemistry 1998, 48, 1139−1143.(12) Evidente, A.; Maddau, L.; Spanu, E.; Franceschini, A.; Lazzaroni,S.; Motta, A. J. Nat. Prod. 2003, 66, 313−315.(13) Sparapano, L.; Bruno, G.; Fierro, O.; Evidente, A. Phytochemistry2004, 65, 189−198.(14) Evidente, A.; Masi, M.; Linaldeddu, B. T.; Franceschini, A.;Scanu, B.; Cimmino, A.; Andolfi, A.; Motta, A.; Maddau, L.Phytochemistry 2012, 77, 245−250.(15) Venkatasubbaiah, P.; Sutton, T. B.; Chilton, W. S.Phytopathology 1991, 81, 243−247.(16) Martos, S.; Andolfi, A.; Luque, J.; Mugnai, L.; Surico, G.;Evidente, A. Eur. J. Plant Pathol. 2008, 121, 451−461.(17) Evidente, A.; Punzo, B.; Andolfi, A.; Cimmino, A.; Melck, D.;Luque, J. Phytopathol. Mediterr. 2010, 49, 74−79.(18) Djoukeng, J. D.; Polli, S.; Larignon, P.; Abou-Mansour, E. Eur. J.Plant Pathol. 2009, 124, 303−308.(19) Pretsch, E.; Buhlmann, P.; Affolter, C. Structure Determination ofOrganic Compounds-Table of Spectral Data, 3rd ed.; Spinger-Verlag:Berlin, 2000; pp 161−243, 385−404.(20) Nakanishi, K.; Solomon, P. H. Infrared Absorption Spectroscopy,2nd ed.; Holden Day: Oakland, 1977; pp 17−44.(21) Berger, S.; Braun, S. 200 and More Basic NMR Experiments: APractical Course, 1st ed.; Wiley-VCH: Weinheim, 2004.(22) Breitmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy; VCH:Weinheim, 1987; pp 183−280.(23) Mennucci, B.; Claps, M.; Evidente, A.; Rosini, C. J. Org. Chem.2007, 72, 6680−6691.(24) Ahamad, P. Y. A.; Kunhi, A. A. M.; Divakar, S. J. Microbiol.Biotechnol. 2001, 17, 371−377.(25) Shen, Y. Arch. Environ. Contam. Toxicol. 1998, 34, 229−34.(26) Barreto, G. E. S.; de Oliveira, E. S.; de Villar, F. A. G. A.; dosSantos, G. S.; Lima, R. M. F.; Costa, S. L.; de Fatima Dias Costa, M.;Egito, E. S. T.; El-Bacha, R. S. Braz. Archiv. Biol. Technol. 2007, 50,839−849.(27) Nakai, N.; Murata, M.; Nagahama, M.; Hirase, T.; Tanaka, M.;Fujikawa, T.; Nakao, N.; Nakashima, K.; Kawanishi, S. Free Radical Res.2003, 37, 69−76.(28) Li, X.; Yao, Y.-H.; Zheng, Y.-N.; Lin, W.-H.; Sattler, I.; Deng, Z.-W. Zhongguo Tianran Yaowu 2007, 5, 20−23.(29) Capasso, R.; Cristinzio, G.; Evidente, A.; Scognamiglio, F.Phytochemistry 1992, 31, 4125−4128.(30) Pittayakhajonwut, P.; Sohsomboon, P.; Dramae, A.;Suvannakad, R.; Lapanun, S.; Tantichareon, M. Planta Med. 2008,74, 281−286.(31) Evidente, A.; Maddau, L.; Scanu, B.; Andolfi, A.; Masi, M.;Motta, A.; Tuzi, A. J. Nat. Prod. 2011, 74, 757−763.(32) Andolfi, A.; Mugnai, L.; Luque, J.; Surico, G.; Cimmino, A.;Evidente, A. Toxins 2011, 3, 1569−1605.(33) Cimmino, A.; Villegas-Fernandez, A. M.; Andolfi, A.; Melck, D.;Rubiales, D.; Evidente, A. J. Agric. Food Chem. 2011, 59, 9201−9206.(34) Van Niekerk, J. M.; Crous, P. W.; Fourie, P. H.; Halleen, F.Mycologia 2004, 4, 781−798.(35) Taylor, A.; Hardy, G. E. J., St.; Wood, P.; Burgess, T. Australas.Plant Pathol. 2005, 34, 187−195.
Journal of Natural Products Article
dx.doi.org/10.1021/np300512m | J. Nat. Prod. 2012, 75, 1785−17911790
(36) Apoga, D.; Åkesson, H.; Jansson, H. B.; Odham, G. Eur. J. PlantPathol. 2002, 108, 519−526.(37) Reino, J. L.; Hernandez-Galan, R.; Duran-Patron, R.; Collado, I.G. J. Phytopathol. 2004, 152, 563−566.(38) Strange, R. N. Nat. Prod. Rep. 2007, 24, 127−144.(39) Mahoney, N.; Molyneux, R. J.; Smith, L. R.; Schoch, T. K.;Rolshausen, P. E.; Gubler, W. D. J. Agric. Food Chem. 2005, 53, 8148−8155.(40) Albuquerque, P.; Casadevall, A. Med. Mycol. 2012, 50, 337−345.(41) Chen, H.; Fujita, M.; Feng, Q.; Clardy, J.; Fink, G. R. Proc. Natl.Acad. Sci. U. S. A. 2004, 101, 5048−5052.(42) Lamari, L. Assess Image Analysis Software for Plant DiseaseQuantification; APS Press: St. Paul, MN, 2002.
Journal of Natural Products Article
dx.doi.org/10.1021/np300512m | J. Nat. Prod. 2012, 75, 1785−17911791
