Lentiquinones A, B, and C, Phytotoxic ... - ACS Publications

Nov 20, 2018 - derivatives, named lentiquinones A (1), B (2), and C (3), and the known lentisone. From the ...... Chenopodium album L. Chenopodiaceae...
1 downloads 0 Views 2MB Size
Article pubs.acs.org/jnp

Cite This: J. Nat. Prod. 2018, 81, 2700−2709

Lentiquinones A, B, and C, Phytotoxic Anthraquinone Derivatives Isolated from Ascochyta lentis, a Pathogen of Lentil Marco Masi,† Paola Nocera,† Maria C. Zonno,‡ Angela Tuzi,† Gennaro Pescitelli,§ Alessio Cimmino,† Angela Boari,‡ Alessandro Infantino,⊥ Maurizio Vurro,‡ and Antonio Evidente*,†

Downloaded via UNIV DE BARCELONA on January 11, 2019 at 19:54:13 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



Dipartimento di Scienze Chimiche, Università di Napoli Federico II, Complesso Universitario Monte S. Angelo, Via Cintia 4, 80126 Napoli, Italy ‡ Istituto di Scienze delle Produzioni Alimentari (ISPA), Consiglio Nazionale delle Ricerche, Via Amendola, 122/O, 70126 Bari, Italy § Dipartimento di Chimica e Chimica Industriale, Università di Pisa, Via Moruzzi 13, 56124 Pisa, Italy ⊥ Consiglio per la Ricerca in Agricoltura e l’Analisi dell’Economia Agraria (CREA), Centro di Ricerca Difesa e Certificazione, via C.G. Bertero 22, 00156, Rome, Italy S Supporting Information *

ABSTRACT: A strain of the pathogenic fungus Ascochyta lentis isolated from lentil (Lens culinaris) was studied to ascertain its capability to produce bioactive metabolites. From the culture filtrates were found three new anthraquinone derivatives, named lentiquinones A (1), B (2), and C (3), and the known lentisone. From the mycelium, four known analogues were identified, namely pachybasin (in larger amount), ω-hydroxypachybasin, 1,7-dihydroxy-3-methylanthracene-9,10-dione, and phomarin. Lentiquinones A−C were characterized by spectroscopic methods as 3,4,6-trihydroxy-8-methyl-2H-benzo[g]chromene5,10-dione, 2,3,4,5,10-pentahydroxy-7-methyl-3,4,4a,10-tetrahydroanthracen9(2H)-one, and its 2-epimer, respectively, and the relative configuration of the two latter compounds was deduced by X-ray diffraction data analysis. The absolute configuration of lentiquinones B and C was determined as (2R,3S,4S,4aS,10R) and (2S,3S,4S,4aS,10R), respectively, by electronic circular dichroism (ECD) in solution and solid state, and TDDFT calculations. When tested by using different bioassays, the novel compounds showed interesting activities. In particular, applied to punctured leaves of host and nonhost plants, the three new compounds and lentisone caused severe necrosis, with lentiquinone A being the most active among the new metabolites. On cress (Lepidium sativum), this latter compound proved to be particularly active in inhibiting root elongation. On Lemna minor all the compounds reduced the content of chlorophyll, with 1,7-dihyroxy-3methylanthracene-9,10-dione being the most active. The new compounds, together with lentisone, proved to have antibiotic properties.

A

phenols, named lathyroxins A and B, were isolated together with other known metabolites, namely p-hydroxybenzaldehyde, p-methoxyphenol, and tyrosol.5 Furthermore, from another strain of A. lentis grown in liquid culture, a novel anthraquinone named lentisone, and other known metabolites (namely pachybasin, tyrosol, and pseurotin A) were isolated.6 Thus, it seemed of interest to evaluate the production of bioactive compounds also by one of the reference strains used previously for genetic purposes, and to compare it with the metabolic profile of other closely related strains.4 This manuscript reports the isolation and the chemical and biological characterization of three new and five known bioactive anthraquinone derivatives produced by A. lentis and

scochyta blights are important plant diseases worldwide, and among crops, legumes are particularly infected by these fungal pathogens.1 For example, chickpea (Cicer arietinum L.) can be attacked by Ascochyta rabiei; fava bean (Vicia faba L.) by Ascochyta fabae; and lentil (Lens culinaris Medik.) by Ascochyta lentis.2 Several pathogens belonging to the Ascochyta genus proved to produce secondary phytotoxic metabolites involved in disease development and in symptom appearance.3 Recently, a morphological variant of A. lentis, named A. lentis var. lathyri,4 was isolated in Italy and described for the first time on the basis of morphological and genetic comparisons with some A. lentis reference strains. Considering the novelty of A. lentis var. lathyri and the lack of information about its eventual capability to produce bioactive secondary metabolites, the metabolic profile of one strain of this pathogen has been studied. In particular, two new phytotoxic monosubstituted © 2018 American Chemical Society and American Society of Pharmacognosy

Received: July 9, 2018 Published: November 20, 2018 2700

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

phenolic group, being linked by a hydrogen bond to the carbonyl (OC-5) of the adjacent quinone ring (B ring), appeared, as expected, as a sharp singlet at δ 11.99.16 Furthermore, the same spectra showed the presence of two singlets and a broad signal typical of an oxymethylene (H2C2), a benzylic methyl (Me-8) and another hydroxy group (HO3) at the typical chemical shift values of 4.01, 2.37, and 2.80, respectively,16 suggesting also the presence of a 1,2,3,4tetrasubstituted dihydro-2H-pyran ring (C ring) fused to the B ring. The couplings observed in the HSQC spectrum17 (Table 1) permitted the assignment of the signals observed in the 13C NMR spectrum (Table 1) at δ 123.6, 120.3, 67.2, and 22.4, to the protonated carbons C-7, C-9, C-2, and Me-8, respectively.18 The same spectrum showed 10 sp2 carbons, two of which were the quinone carbonyls assigned by the long-range coupling observed in the HMBC spectrum17 (Table 1) for C10 at δ 183.9 with H-9 and consequently C-5 was at δ 182.5. The remaining eight sp2 carbons belong to both the tetrasubstituted aromatic, quinone, and pyran rings (A, B, and C rings) and four of them (5a and 9a, and 4a and 10a) were the bridgehead carbons of the junction between the A and B, and B and C rings. Six of them were assigned on the basis of the other couplings observed in the HMBC spectrum. In fact, the couplings between C-3 and H2C-2 and HO-3, C-5a and HO-6, H-7 and Me-8, C-4 with HO-3, C-6 and HO-6, and H-7, C-8 with H-9 and H-7, C-9 with H-7 and Me-8, C-9a with Me-8, allowed the assignment of the signals at δ 141.5, 131.8, 161.3, 161.5, 123.8, 120.3, and 147.8 to C-3, C-5a, C-6, C-4, C-8, C-9, and C-9a, respectively, and of the remaining two sp2 carbons at δ 142.5 and 142.4 to C-10a and C-4a (which could be interchangeable). Thus, the signals were assigned to all the protons and the corresponding carbons of 1 as reported in Table 1, and lentiquinone A (1) was formulated as 3,4,6-trihydroxy-8methyl-2H-benzo[g]chromene-5,10-dione. The structure assigned to 1 was confirmed by all the other couplings observed in the HMBC data (Table 1) and the data from the HRESIMS spectrum. The latter showed the sodiated dimer [2 M + Na]+, the sodium adduct ion [M + Na] +, and the protonated ion [M + H]+ at m/z 571, 297, and 275.0559, respectively. Lentiquinone B (2) has a molecular formula of C15H16O6 as deduced by its HRESIMS and 13C NMR data, consistent with eight indices of hydrogen deficiency. Compounds 2 and 1 differ for one carbon and six hydrogens due to the different functionalization of the B and C rings, while the A ring is the same. This was confirmed by comparing the 1H and 13C NMR data (Table 1), which highlighted the nature of a hemiquinone and a trihydroxycyclohexene for the B and C rings, respectively. Indeed, its 13C NMR spectrum shows a single α,β-unsaturated carbonyl (C-9) at δ 185.7 and lacks any oxygenated methylene carbon for a pyran ring. The 1H NMR spectrum lacks the signal of a hydrogen-bounded phenolic group, while an olefinic proton was present, and the complexity of the region corresponding to the protons attached to the oxygenated carbons was significantly increased. In particular, the 13 C and 1 H NMR spectra showed a secondary hydroxylated carbon at δ 73.4 (d, C-10)/5.27 (d, J = 10.1 Hz). This proton showed a COSY correlation with H-4a of the adjacent tertiary carbon, which appeared at δ 2.94 as a complex signal due to its coupling also with H-4 (C ring) and its allylic and homoallylic coupling with H-1 and H-2, respectively.16,19

some preliminary information on their possible involvement in disease development and symptom appearance.



RESULTS AND DISCUSSION The fungal liquid culture was separated by filtration into the two main components, i.e., the culture filtrate and the mycelium. Both were subjected to extraction by organic solvents as reported in detail in the Experimental Section. Eight main metabolites were obtained: four from the culture filtrates (1−4) and four from the mycelium (5−8). From preliminary 1H NMR investigations, all these metabolites appeared to be structurally related, being anthraquinone derivatives and appearing as yellow-colored compounds. Three compounds isolated from the culture filtrates proved to be new and were named lentiquinones A, B, and C (1, 2, and 3, respectively). The other five metabolites proved to be known fungal and/or plant metabolites, namely lentisone (4), pachybasin (5), ω-hydroxypachybasin (6), 1,7-dihydroxy-3methylanthracene-9,10-dione (7), and phomarin (8). Pachybasin (5) proved to be the metabolite produced in larger amount by the fungus into the mycelium (7 mg/g of dried mycelium), whereas lentiquinone B (2) was the main metabolite in the culture filtrates (0.6 mg/L). The known metabolites were identified by comparing their observed and reported spectroscopic data; lentisone (4): Andolfi et al. (2013);6 for all the other metabolites: Sun et al. (2012).7 Specifically, for pachybasin (5): Bick and Rhee (1966),8 De Stefano et al. (1999),9 De Souza et al. (2006)10 and Xia et al. (2007);11 for ω-hydroxypachybasin (6): Imre et al. (1976),12 Kuo et al. (1995)13 and Nanthathong et al. (2012);14 for 1,7dihydroxy-3-methylanthracene-9,10-dione (7): de Souza et al. (2006);10 for phomarin (8): Bick and Rhee (1966)8 and de Souza et al. (2006).10

Lentiquinone A (1) has the molecular formula C14H10O6 as deduced by its HRESIMS and 13C NMR data consistent with 10 indices of hydrogen deficiency. Its 1H and 13C spectra showed spin systems in accord with those of a substituted anthraquinone. The data were also consistent with the bands of hydroxy, α,β-conjugated carbonyl, and aromatic groups observed in the IR spectrum15 as well as with the absorption maxima observed in the UV spectrum.16 The 1H and COSY17 data (Table 1) showed the typical spin system of a 1,2,3,5tetrasubstituted benzene ring (A ring) with two singlets of the meta-located protons at δ 7.37 (H-9) and 6.98 (H-7), the latter being shielded by the ortho-located phenolic group. The 2701

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products Table 1. 1H and

13

Article

C NMR Data of Lentiquinones A−C (1−3, Respectively)a,b 1

position

δCc

δH (J in Hz)

1 2

67.2 t

3 4

141.5 s 161.5 s

4a

142.4 sd

5 5a

182.5 s 131.8 s

6

161.3 s

7

123.6 d

8

123.8 s

8a 9 9a 10 10a Me-7 Me-8 HO-3 HO-6

120.3 147.8 183.9 142.5

4.01 (2H) s

d s s sd

22.4 q

2 HMBC

δCc

HO-3

138.2 d 71.2 d

H2-2, HO-3 HO-3

75.6 d 75.5 d 48.0 d

3

δH (J in Hz) 6.84 (1H) 4.24 (1H) 2.6) 3.53 (1H) 3.96 (1H)

6.98 (1H) s

7.37 (1H) s

2.37 (3H) s 2.80 (1H) br s 11.99 (1H) br s

HO-6, H-9, Me-8 H-9, H-7

H-7, Me-8 Me-8 H-9

122.6 d

139.2 s 185.7 s 130.7 s 73.4 d 131.5 s 19.6 q

δH (J in Hz)

H-4a H-4, H-3

136.8 d 67.1 d

6.88 (1H) dd (4.2, 2.8) 4.43 (1H) td (4.2, 1.8)

dd (10.1, 8.2) dd (10.1, 8.4)

H-4 H-10

73.0 d 71.4 d

3.74 (1H) dd (8.2, 4.2) 4.26 (1H) dd (8.2, 6.1)

H-1, H-3, H4 H-6, H-10

50.8 d

2.83 (1H) me (10.0, 6.1, 2.8, 1.8)

6.91 (1H) d (1.2)

H-8, Me-7

124.0 d 127.4 s

7.38 (1H) d (1.2)

H-6, H-8, Me-7 H-6, Me-7

120.3 d

Me-7 H-1, H-8 H-1 H-4 H-10 H-6, H-8

140.6 s 187.7 s 133.0 s 73.6 d 134.9 s 21.1 q

2.94 (1H) me (10.1, 8.4, 4.2, 2.6)

125.2 s 119.0 d

δCc

t (2.6) ddd (8.2, 4.2,

156.3 s HO-6, H-7, Me-8 HO-6, H-7

HMBC

5.27 (1H) d (10.1) 2.31 (3H) s

157.9 s

HMBC H-4 H-4 H-10, H3 H-10, H3 H-6

6.90 (1H) d (1.2)

H-8, Me-7 H-6, H-8

7.39 (1H) d (1.2)

H-6, Me-7 Me-7 H-8 H-10 H-4 H-4 H-6, H-8

5.28 (1H) d (10.0) 2.31 (3H) s

H-9, H-7

a The chemical shifts are in δ values (ppm) from TMS. b2D 1H, 1H (COSY) 13C, 1H (HSQC) NMR experiments delineated the correlations of all the protons and the corresponding carbons. cMultiplicities were assigned by DEPT spectrum. dThese attributions could be exchanged. eThis is a complex signal due to the coupling of H-4a with the proton (H-4 and H-10) of the adjacent secondary carbons C-4 and C-10 of C and B rings and also of its allylic and homoallylic coupling with H-1 and H-2.16,19

Thus, C-4a at δ 48.0 constituted one of bridgehead carbons between the B and C rings. H-4 resonates as a doublet of doublets (J = 10.1 and 8.4 Hz) at δ 3.96, being also coupled with H-3, the proton of the secondary hydroxylated carbon (C3) of the same C ring, which appeared as a doublet of doublets (J = 10.1 and 8.2 Hz) at δ 3.53. In turn, H-3 was coupled with H-2 of the adjacent hydroxylated carbon (C-2), which resonated as a doublet of doublets of doublets (J = 8.2, 4.2, and 2.6 Hz), at δ 4.24. H-2 also coupled with the adjacent olefinic protons (H-1) resonating as a triplet (J = 2.6 Hz) at δ 6.84, being also allylically coupled with H-4a. The 13C NMR spectrum, besides the cited carbonyl group, showed the olefinic protonated carbon, the signals of the other three protonated secondary carbons, the benzylic methyl, and those of the four aromatic and the olefinic sp2 carbons, which were attributed by the couplings observed in the HSQC and HMBC spectra. Thus, the chemical shifts were assigned to all the protons and the corresponding carbons of 2 as shown in Table 1, and lentiquinone B (2) was formulated as 2,3,4,5,10-pentahydroxy7-methyl-3,4,4a,10-tetrahydroanthracen-9(2H)-one. The structure assigned to 2 was confirmed by the other couplings observed in the HMBC data (Table 1) and the data from the HRESIMS spectrum. The latter showed the sodiated dimer [2M + Na]+ and the protonated molecular ion [M + H]+ at m/z 607 and 293.1013, respectively. The relative configuration of lentiquinone B was deduced by the 1H NMR coupling constants. In fact, as the ring C assumes the most stable half-chair conformation, the typical 3JH,H values

recorded for the coupling between H-4a and H-4 (J = 8.4 Hz), H-4 with H-3 (J = 10.1 Hz), and H-3 with H-2 (J = 8.2 Hz), demonstrated that all four protons occupy pseudoaxial positions16,19 and permitted assignment of the relative configuration of 2 as (2R*,3S*,4S*,4aS*,10R*). The structure and the relative configuration of lentiquinone B (2) were confirmed by X-ray diffraction data analysis carried out on the colorless block-shaped crystals obtained by the slow addition of water to a solution of 2 in MeOH, followed by slow evaporation of the aqueous alcohol mixture. An ORTEP view of lentiquinone B is shown in Figure 1. All bond lengths and angles are in the normal range. Crystal data and refinement details are reported in the Experimental Section. Compound 2 crystallizes in the orthorhombic P212121 space group with one molecule of 2 and one H2O solvent molecule contained in the asymmetric unit. The structure of lentiquinone B (2) comprises three sixmembered rings, whose bond lengths, pattern, and geometry disclose the aromatic nature of ring A, the hemiquinone nature of ring B, and the cyclohexene nature of ring C. In the ring systems the sp2 hybridization of C-9, C-9a, and C-1 and the planarity of ring A allow a near all-planar shape of the molecule. A degree of flexibility is found in the B and C rings, which adopts an envelope conformation of ring B (atom C-4a at the flap), and the half-chair conformation of ring C (C-3 and C-4 atoms up and down the mean plane C-1/C-2/C-4a/C-9a). Five stereogenic centers are present in the molecule with the relative configuration (2R*,3S*,4S*,4aS*,10*R) confirmed by 2702

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

Figure 1. ORTEP view of lentiquinone B (2) molecular structure with ellipsoids drawn at 50% probability level. Selected bond lengths and angles: C9−O6 = 1.223(3), C9a−C1− = 1.338(3)Å; C9a−C1− C2 = 123.6(2), C1−C9a−C9 = 118.19(19), C8a−C9−C9a = 118.08(17)°.

Figure 2. ORTEP view of lentiquinone C (3) molecular structure with ellipsoids drawn at 50% probability level. Selected bond lengths and angles: C9−O6 = 1.2297(14), C9a−C1− = 1.3465(16) Å; C9a− C1−C2 = 123.35(11), C1−C9a−C9 = 118.35(10), C8a−C9−C9a = 118.58(9)°.

the X-ray structure analysis. The methine H atoms at C-2, C-3, C-4, C-4a, and C-10 are in the axial positions with a mutually trans configuration. All the hydroxy groups are in the equatorial positions and are involved in intra- and intermolecular H-bonding that also includes the solvent H2O molecule (Figure S1, Supporting Information). Lentiquinone C (3) showed the same molecular formula C15H16O6 as 2, as deduced by its HRESIMS and 13C NMR data, and the same eight indices of hydrogen deficiency. Comparison of the 1H and 13C NMR data of 2 and 3 showed similar spin systems (Table 1). The most significant differences were the upfield shifts (Δδ 4.1) observed for C-2 in the 13C NMR spectrum of 3, and the significantly different value recorded for the coupling between H-2 with both H-3 (J = 4.2 Hz) and H-1 (J = 4.2 Hz), while the other J-couplings remained practically unchanged. The value measured for the coupling between H-2 and H-3 is typical for an axial− equatorial coupling in a half-chair conformation for a cyclohexene derivative.16,19 Thus, lentiquinone C (3) appears to be the 2-epimer of 2. The couplings observed in the COSY, HSQC, and HMBC data (Table 1) permitted the assignment of the chemical shifts for all the protons and the corresponding carbons of 3 as shown in Table 1, and the structure of lentiquinone C as (2S*,3S*,4S*,4aS*,10R*)-2,3,4,5,10-pentahydroxy-7-methyl3,4,4a,10-tetrahydroanthracen-9(2H)-one. The structure assigned to 3 was also supported by the data of its HRESIMS spectrum which showed the sodiated dimer [2M + Na]+ and the protonated molecular ion [M + H]+ at m/ z 607 and 293.1028, respectively. The structure and the relative configuration of lentiquinone C (3) were confirmed by the X-ray diffraction data analysis carried out on its orange block-shaped crystals obtained by the same procedure described for 2. An ORTEP view of lentiquinone C is shown in Figure 2. All bond lengths and angles are in the normal range. Crystal data and refinement details are reported in the Experimental Section. Lentiquinone C (3) crystallizes in the monoclinic P21 space group with one molecule contained in the asymmetric unit. Crystal structure analysis shows that the molecular geometry of lentiquinone C (3) is similar to lentiquinone B (2). The two molecules perfectly overlay each other, except the hydroxy O-1 atom at C-2 that in 3 is axial and in 2 equatorial (Figure 3). The X-ray diffraction data analysis clearly shows that 3 is the 2-epimer of 2, with (2S*,3S*,4S*,4aS*,10R*) relative config-

Figure 3. Superimposition of lentiquinone C (3, element colors) with lentiquinone B (2, orange) X-ray geometries.

uration. In the crystal packing of 2 all the hydroxy groups are involved in strong OH···O intra- and intermolecular hydrogen bonds and layers of parallel molecules are arranged in a tridimensional H-bonding pattern (Figure S2, Supporting Information). Attempts to assign the absolute configurations of 2 and 3 by X-ray diffraction data analysis were made according to literature methods applied to light-atoms structures using Mo Kα radiation.20−22 However, no certainty about the absolute configuration was reached because of a too large standard deviation for the calculated absolute structure parameters [Flack parameters: 0.0(2) for lentiquinone B and 0.14(12) for lentiquinone C].20 The absolute configurations of both lentiquinones B and C were defined via electronic circular dichroism (ECD) spectroscopy.23 The ECD spectra of 2 and 3 were measured in MeCN and, for 3, also in the solid state as a KCl pellet (Figures 4−6). The spectra show a rich pattern of bands expected for the extended conjugated chromophore. Using a consolidated computational procedure,24 the conformations of 2 and 3 were investigated by a conformational search with molecular mechanics (Merck molecular force field, MMFF), followed by geometry optimizations with density functional theory (DFT) run at the ωB97X-D/6-311+G(d,p) level and including a solvent model (SMD) for MeCN. Compound 2 has a dominant conformation in keeping with NMR (measured J-couplings) and X-ray diffraction data results, and a minor conformation with rotated OH groups. For compound 3 various conformers were found corresponding to the rotamerism of OH groups and ring flipping of ring C. However, the two most stable conformers featured the same ring conformation found in the X-ray geometry and similar to 2703

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

Figure 4. Experimental (solid blue line) and calculated (dotted black line) ECD spectrum of (2R,3S,4S,4aS,10R)-2. Experimental conditions: 2.89 mM solution in MeCN, 0.01 cm cell. Calculation level: CAM-B3LYP/def2-TZVP/SMD//ωB97X-D/6-311+G(d,p)/SMD; plotted as sum of Gaussians with exponential half-width of 0.32 eV, red-shifted by 17 nm, scaled by a factor 2.

Figure 6. Experimental (solid blue line) ECD spectrum of (2S,3S,4S,4aS,10R)-3 in the solid state (KCl pellet) compared with the spectrum calculated on the X-ray geometry (dotted black line) at the CAM-B3LYP/def2-TZVP level. Plotted as sum of Gaussians with exponential half-width of 0.35 eV, red-shifted by 18 nm, scaled by a factor 2.

2. ECD calculations were run with time-dependent DFT (TDDFT) using various functionals (B3LYP, CAM-B3LYP, M06) and the def2-TZVP basis set, with the SMD solvent model. The good agreement between experimental and calculated ECD spectra, especially using CAM-B3LYP (Figures 4 and 5), permitted assignment of the absolute configurations as (2R,3S,4S,4aS,10R)-2 and (2S,3S,4S,4aS,10R)-3.

expected for conjugated chromophores with strong dipoleallowed transitions like the present one.26 Inspection of the crystal packing of 3 confirms that such couplings are viable because of the skewed orientation of distinct molecules along the b axis. Assayed on punctured leaves of 10 plant species (including the host) belonging to eight botanic families, the compounds showed different activities (Table 2). Lentisone (4) was the most active compound, causing the widest necrosis to all the tested leaves. Lentiquinone A (1) was active too, but on average less active than lentisone, whereas lentiquinone B and C (2 and 3) were even less, but clearly active. The other tested compounds were modestly active or not active at all. In general, the different sensitivity of the leaves to the tested compounds seemed to be related to the characteristics of leaf surface. For instance, basil leaves have a tiny epidermis, and thus the clear necrosis probably are caused by the active compounds because they can readily diffuse into the parenchyma. The results confirmed those reported by Andolfi et al. (2013)6 as lentisone proved to be toxic in their assay, whereas pachybasin was not. On cress, lentiquinone A (1) had the highest activity, reducing almost 70% of rootlet growth compared to the control (Table 3). The other compounds had modest or negligible effects. In the bioassay on Lemna minor, 1,7 dihydroxy-3-methylantracene-9,10-dione was particularly active (together with ω-hydroxypachybasin), reducing the chlorophyll content in the fronds by over 80% compared to the control. All the other compounds had a lower effect on chlorophyll, except pachybasin that was inactive. In the assay on seeds of the parasitic weed Phelipanche ramosa, lentisone (4) caused the total inhibition of seed germination (Table 3). Lentiquinone A (1) reduced the germination percentage by 60%, compared to the control, whereas the other tested metabolites had modest or negligible effects. Considering its effectiveness at 10−3 M, 4 was also tested at lower concentrations. At 2 × 10−4 M and 4 × 10−5 M it still caused some phytotoxic effects in terms of reduction of seed germination and germ tube elongation (data not shown), whereas at 8 × 10−6 M it was inactive. Thus, regarding the phytotoxicity, it seems that the presence of a B-ring quinone moiety, as in 1 and 4, is an important

Figure 5. Experimental (solid red line) and calculated (dotted black line) ECD spectrum of (2S,3S,4S,4aS,10R)-3. Experimental conditions: 3.05 mM solution in MeCN, 0.01 cm cell. Calculation level: CAM-B3LYP/def2-TZVP/SMD//ωB97X-D/6-311+G(d,p)/SMD; plotted as sum of Gaussians with exponential half-width of 0.4 eV, red-shifted by 12 nm, scaled by a factor 1.5.

For compound 3, the quantity of crystals used for the X-ray diffraction experiment was large enough (about 50 μg) to apply a variant of the ECD analysis known as the solid-state ECD/TDDFT approach.25 Here, the ECD spectrum is measured in the solid state as KCl pellet, and the calculation is run with TDDFT on the X-ray geometry, avoiding any uncertainty related to the conformation. The solid-state experimental and calculated ECD spectra are shown in Figure 6. Their good agreement in the long-wavelength region confirms the aforementioned absolute configuration of 3. The discrepancy at shorter wavelength is most likely a consequence of intercrystalline exciton couplings, which are 2704

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

Table 2. Effect of Ascochyta lentis Metabolites (1−8) in the Leaf Puncture Assay compound plant species

family

Amaranthus retrof lexus L. Dittrichia viscosa (L.) Greuter Sonchus Chenopodium album L. Convolvulus arvensis L. Ocimum basilicum L. Lens culinaris L. Lupinus albus L. Setaria Solanum nigrum L.

1

Amaranthaceae Asteraceae Asteraceae Chenopodiaceae Convolvulaceae Lamiaceae Leguminosae Leguminosae Poaceae Solanaceae

a

2 3 2 4 3 4 3 4 3 4

2

3

4

5

6

7

8

1 1 0 2 2 2 2 2 1 1

1 1 0 3 2 2 1 4 2 1

3 3 3 4 4 4b 4 4 2d 4

0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 1 2 0 0

0 0 0 0 0 0

0 0 0 0 0 0

c

c

0 0 0

2 1 0

a

Dimension of leaf necrosis by a visual scale from 0 to 4 (0 = no effects; 4 = necrosis around 1 cm diameter). bNecrosis very wide, out of the scale. Only chlorosis. dNecrosis surrounded by an area of chlorophyll retention (green island).

c

Table 3. Effects of the Ascochyta lentis Metabolites (1−8) in Different Bioassays compound bioassay

plant species

1

2

3

4

5

6

7

8

rootlet elongationa chlorophyllb seed germinationc antibiosisd

Lepidium sativum Lemna minor Phelipanche ramosa Bacillus subtilis

32.7 77.7 40e 11

89.0 60.2 79 12

83.9 58.5 78 14

75.6 60.5 1 14

100.0 100.0 93 0

90.4 24.1 104e 0

90.8 17.6 106 0

102.3 53.6 106 0

a

Rootlet elongation (percentage in comparison to the control length). bChlorophyll content (percentage compared to the control content). cSeed germination (percentage compared to the control germination). dGrowth inhibition halo (diameter: mm). eGerminated tubes shorter than the control ones.

of four out of the nine tested strains, namely Verticillium dahlia, Penicillium allii, Rhizoctonia sp., and Phoma exigua. All the other compounds proved to be ineffective (data not shown) to all the tested fungi. In the antibiotic assay against Gram+ and Gram− bacteria, lentisone (4) and a lentiquinone C (3) were active against Bacillus subtilis, causing clear growth inhibition areas (Table 3). Lentiquinone B (2) was active too, but to a lesser extent; lentiquinone A (1) was even less active, although clearly active. The other four bioassayed compounds proved to be completely inactive at the tested concentration. Conversely, all the compounds were inactive against Escherichia coli (data not shown). For a SAR evolution the results of the antibiotic activity are similar to those reported for the phytotoxicity. The results obtained in our biological assays for the known compounds are in line with those reported in previous studies, in particular regarding pachybasin,6 which had no biological effects in all our assays. The eight isolated compounds (both the new and the known ones, either from liquid filtrate or from mycelium) proved to be anthraquinone derivatives except lentiquinone A (1) which is a 2H-benzo[g]chromene. Anthraquinones are the largest group of natural pigments of quinoid nature, produced by microbes and plants, and possess many different biological properties, e.g., antibacterial, insecticidal, fungicidal, herbicidal, and antiviral activities. However, their roles in plant−pathogen relationships have not yet been clearly defined. The improvement in methods for the identification of natural compounds permits the establishment of the structure of new anthraquinones (several hundreds of compounds have been so far described). This may offer new opportunities to discover natural agrochemicals to combat pathogenic fungi, pests, and weeds, as well as drugs with antiviral, anticancer, and immunomodulatory activity.34,35

structural facture to impart activity, while it is missing in lentiquinones B and C (2 and 3), which are less active. Also, the nature of the C-ring seems important for the phytotoxicity as the presence of a pyran C-ring, as in 1, compared to 4 decreased the activity. Metabolites 5−8, all quinones, probably did not show phytotoxicity due to the aromatization of the Cring. The relation between the quinone skeleton and its biological activity is already known. In fact, a similar structure−activity relationship was also found in a recent study carried out by some of the authors assaying the cyclohexene epoxide epiepoformin, a phytotoxic metabolite produced by the oak pathogen Diplodia quercivora,27 and some of its semisynthetic derivatives in an etiolated wheat coleoptile bioassay.28 Furthermore, similar SAR results were observed using some derivatives of sphaeropsidones, two epimeric phytotoxic cyclohexene epoxides produced by the canker inducing agent of Italian cypress (Cupressus sempervires L.) Diplodia cupressi,29 when their phytotoxic and antifungal activities were tested on nonhost plants and on five Phytophthora pathogenic species.30 The same results were also obtained from the SAR study based on their haustorium-inducing activity in Orobanche cumana, O. crenata, and Striga hermonthica.31 This hypothesis on the mode of action of lentisone and lentiquinones as well as that of epi-epoformin and sphaeropsidones is in full agreement with the results obtained using natural and synthetic quinones like sorghum xenognosin and dimethoxybenzoquinones for studying haustoria and the chemistry in host recognition parasitic angiosperms. Quinone/hydroquinone structures serve as cofactors in many metabolic pathways, playing critical chemical roles in oxidation/reduction processes.32,33 In the antifungal bioassay, only lentiquinone A proved to be modestly effective, by partially inhibiting the mycelial growth 2705

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

formed, and thus to protect itself by the reaction of the plant to the pathogen’s attack. Studies have been planned in order to study this hypothesis. Furthermore, studying those compounds could be helpful in the metabolic comparisons and the classification/distinction of strains of the pathogen morphologically and genetically similar, but attacking different hosts (Ascochyta fabae f. sp. lentis Gossen et al., anamorph (ATCC 96419)).40

The pathways of anthraquinone synthesis in plants and fungi are different. While in plants they are produced via shikimate and acetate-mevalonate pathways, in fungi they are synthesized primarily through the latter pathway.36,37 Secondary fungal metabolites formed from this pathway significantly differ in structure and biological properties. For example, anthraquinones of Trichoderma spp. inhibit the growth of many pathogenic soil bacteria and fungi (thus contributing to the antagonistic capability of these fungi) and stimulate plant growth. The group of anthraquinones produced by these fungi includes pachybasin (5), chrysophanol, emodin, ω-hydroxypachybasin (6), and 1,5- and 1,7-dihydroxy-3- hydroxymethyl-9,10 anthraquinones.38 Pachybasin was also produced, together with chrysophanol, by Trichoderma aureoviride9 and by Halorosellinia sp. isolated from Kandelia Woody tissue.11 More recently, T. aureoviride was shown to produce ω-hydroxypachybasin together with a new quinone, a new xanthone, and 11 known compounds.14 Pachybasin, ω-hydroxypachybasin, 1,7-dihydroxy-3-methylanthracene-9,10-dione and phomarin (5−8) were previously isolated also from Coniothyrium sp., an endophytic fungus isolated from Salsola oppostifolia, and exhibited antimicrobial activity against the fungus Microbotryum violaceum, the alga Chlorella f usca, and the bacteria E. coli and Bacillus megaterium.7 They were also isolated together with other anthraquinone pigments from the culture broth of Phoma foveata, which causes gangrene in potatoes;8 5, 7 and 8 were also isolated from the culture of the endophytic fungus Phoma sorghina, found in association with Tithonia diversifolia (Asteraceae).10 Many species of Phoma secrete anthraquinones with herbicidal properties39 whereas only a few are described for Ascochyta, a genus closely related to Phoma. ω-Hydroxypachybasin (6) was also isolated from the roots of Digitalis trojana together with 19 anthraquinones,12 and lately, together with 11 known compounds and seven anthraquinones from the roots of Rubia lanceolata Hayata.13 It is interesting to note that the metabolic profile of the A. lentis strain object of the current study was quite different when compared to that both of A. lentis var. lathyri5 and A. lentis6 although some similarities were observable. The only two metabolites produced by both strains of A. lentis were pachybasin and lentisone. However, whereas the latter was produced in relatively similar amounts by the two strains (1.3 against 0.2 mg/L, respectively), the production of pachybasin was different. Indeed pachybasin was extracted from the mycelium of our strain in large amounts (7.0 mg/g of dried mycelium), whereas it was produced only at 2.5 mg/L of culture filtrate by the other previously studied strains. Beside the new lentiquinones produced in the liquid culture, the other known anthraquinone derivatives produced in the mycelium of A. lentis, i.e., ω-hydroxypachybasin, 1,7-dihydroxy-3-methylanthracene-9,10-dione, and phomarin are not produced by the other two strains. The discovery of novel compounds could be of great help in the understanding of the mechanisms of disease development and in symptom appearance. Considering the phytotoxicity of the pool of compounds in the different bioassays, the possibility that the necrotrophic fungus could produce those metabolites in the first steps of host penetration to facilitate tissue colonization seems to be not so remote. However, another conceivable possibility is that the fungus produces those anthraquinones in order to inhibit the generation of reactive oxygen species (ROS) in the plant tissue it infects, reducing the levels of ROS once they are



EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotation was measured on a Jasco P-1010 digital polarimeter; IR spectra were recorded as deposit glass film on a Thermo Nicolet 5700 FT-IR spectrometer; UV spectra were measured in MeCN on a Jasco V-530 spectrophotometer; ECD spectra were recorded on a Jasco J-815 spectropolarimeter in MeCN with the following conditions: scanning speed of 50 nm min−1, step size of 0.1 nm, a bandwidth of 1 nm, response time of 1 s, accumulation of 8 scans. Solid-state ECD spectra were measured with the KCl pellet technique, described in Pescitelli et al. (2009).25 1H, 13C and 2D NMR spectra were recorded at 400 or 500 MHz in CDCl3 or methanol-d4 on Bruker and Varian instruments. The same solvents were also used as internal standards. Carbon multiplicities were determined by DEPT spectra.17 DEPT, COSY-45, HSQC, and HMBC experiments were performed using Bruker and Varian microprograms.17 HRESI and ESI spectra were recorded on a Waters Q-TOF Micro Mass and on an Agilent 6120 Quadrupole LC−MS instrument, respectively. Analytical, preparative, and reverse phase TLCs were carried out on silica gel (Merck, Kieselgel 60, F254, 0.25, 0.5 mm and RP-18 F254 s respectively) plates. The spots were visualized by exposure to UV radiation, or by spraying first with 10% H2SO4 in MeOH, and then with 5% phosphomolybdic acid in EtOH, followed by heating at 110 °C for 10 min. Column chromatography was performed using silica gel (Merck, Kieselgel 60, 0.063−0.200 mm). Fungal Strain. The strain of Ascochyta lentis used in this study was isolated from a diseased Lens culinaris L. plant and kindly supplied by Dr. W. Kaiser (USDA-ARS, Western Regional Plant Introduction Station, Washington State University) for the genetic studies. It is stored in the ATCC Collection as # ATCC96419, Mating type 1, and in the ISPA fungal collection as # ITEM 17454. The isolate was routinely grown and maintained in plates and slants containing potato-dextrose-agar. Production, Extraction, and Purification of the Culture Filtrates. The fungus was grown in 1 L Roux flasks containing 200 mL of a defined mineral41 for 4 weeks at 25 °C in the dark. The culture filtrate (7.2 L), obtained by filtration, was lyophilized and stored at −20 °C until use. It was redissolved in distilled water to 1/ 10 of its initial volume and extracted exhaustively with EtOAc. The combined organic extracts were dried with Na2SO4 and evaporated under reduced pressure. The residue (290 mg) was fractioned by column chromatography on silica gel eluted with CHCl3−i-PrOH (85:15), obtaining 18 groups of homogeneous fractions. The residue (21.3 mg) of the ninth fraction was further purified by preparative TLC eluted with n-hexane−acetone (6:4), to give nine fractions. The fourth fraction of the latter purification was further purified by reverse phase TLC eluted with MeOH−H2O (7:3), to afford a pure metabolite in the form of an amorphous solid, that was named lentiquinone A (1, 2.4 mg, 0.3 mg/L, Rf 0.3). The residue (20.2 mg) of the 13th fraction of the first column was purified by preparative TLC eluted with CHCl3−i-PrOH (85:15), giving an amorphous solid that was subsequently purified by analytical TLC eluted with CHCl3− MeOH (85:15), to give two crystalline solids named lentiquinone B (2, 0.6 mg/L, Rf 0.4) and C (3, 0.4 mg/L, Rf 0.35). The residue (15.4 mg) of the 12th fraction was purified by analytical TLC, using CHCl3−i-PrOH (9:1), yielding an amorphous solid, that was purified by reverse phase TLC eluted with MeOH-H2O (8:2), to afford an amorphous solid identified as lentisone (4, 1.2 mg, 0.2 mg/L, Rf 0.6). Lentiquinone A (1). UV λmax (log ε) 419 (3.08), 278 (3.40), 250 (3.39), 220 (3.39) nm; IR νmax 3362, 1660, 1640 cm−1; 1H and 13C 2706

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

NMR, see Table 1; HRESIMS (+) m/z 571 [2M + Na] +, 297 [M + Na] +, 275.0559 [M + H] + (calcd for C14H11O6 275.0550). Lentiquinone B (2). [α]25D: +87.4 (c = 0.5 MeOH); UV λmax (log ε) 338 (2.58), 285 (3.24), 254 (3.21), 220 (3.23) nm; ECD (CH3CN) λmax (Δε) 366 (−0.93), 336 (0), 294 (+1.15), 263 (0), 251 (−0.77), 236 (0), 226 (+1.23), 218 (0), 201 (−11.9) nm; IR νmax 3311, 1662, 1456 cm−1; 1H and 13C NMR, see Table 1; HRESIMS (+) m/z 607 [2M + Na]+, 293.1013 [M + H]+ (calcd for C15H17O6 293.1020). Lentiquinone C (3). [α]25D: +71.4 (c = 0.5 MeOH); UV λmax (log ε) 345 (2.56), 284 (3.25), 252 (3.24), 218 (3.29) nm; ECD (CH3CN) λmax (Δε) 366 (−1.25), 340 (0), 287 (+2.88), 258 (+3.17), 227 (+1.53), 222 (0), 202 (−18.8) nm; ECD (KBr) λmax (mdeg) 388 (−0.72), 353 (0), 311 (+0.75), 280 (0), 237 (−1.0), 203 (−3.5) nm; IR νmax 3315, 1662, 1450 cm−1; 1H and 13C NMR, see Table 1; HRESIMS (+) m/z 607 [2M + Na]+, 293.1028 [M + H]+, (calcd for C15H17O6 293.1020). Lentisone (4). 1H NMR (500 MHz, in methanol-d4), δ 7.45 (br s, H-7), 7.09 (br s, H-5), 4.89 (d, J = 3.0 Hz, H-1), 4.11 (ddd, J = 9.4, 5.7, 2.1 Hz, H-3), 3.99 (dd, J = 3.0, 2.1, H-2), 2.94 (dd, J = 19.5, 5.7, H-4a), 2.53 (dd, J = 19.5, 9.4 Hz, H-4b), 2.43 s (Me-6); ESI MS (+) m/z 291 [M + H]+. These data are in agreement with those previously reported.6 Extraction and Purification of the Mycelium. The mycelium (30.5 g) obtained after culture filtration was lyophilized and stored at −20 °C until use. It was suspended in EtOAc and subsequently filtered. The organic extract was evaporated under reduced pressure. The residue (800 mg) was fractionated by column chromatography on silica gel eluting with n-hexane−EtOAc (8:2), to give 12 groups of homogeneous fractions. The residue (234.2 mg) of the fourth and fifth fractions were combined and further purified by column chromatography on silica gel eluted with CH2Cl2−i-PrOH (99:1). From the second fraction of this latter column, a yellow amorphous solid compound was obtained, identified as pachybasin (5, 214.7 mg, 7 mg/g, Rf 0.8). The residue (8.7 mg) of the second fraction of the first column was purified by analytical TLC eluted with n-hexane− EtOAc (7:3), giving a yellow amorphous solid, that proved to be ωhydroxypachybasin (7, 1.2 mg, 0.04 mg/g, Rf 0.3). Two other bands that were combined and further purified by analytical TLC eluted with n-hexane−i-PrOH (9:1), to afford two yellow amorphous compounds 1,7-dihydroxy-3-methylanthracene-9,10-dione (7, 1.5 mg, 0.04 mg/g, Rf 0.7) and phomarin (8, 2.5 mg, 0.08 mg/g, Rf 0.7), respectively. Pachybasin (5). 1H NMR (500 MHz, in CDCl3), δ 12.55 (s, OH), 8.25 (dd, J = 8.2, 3.4 Hz, H-5 and H-8), 7.77 (m, H-6), 7.59 (m, H-4 and H-7), 7.06 (d, J = 2.5 Hz, H-2), 2.44 (s, OCH3), these data were very similar to that previously reported;7−11 ESIMS (+) m/z 239 [M + H]+. ω-Hydroxypachybasin (6). 1H NMR (500 MHz, in CDCl3), δ 12.64 (s, OH-1), 8.32 (m, H-5 and H-8), 7.62 (m, H-7, H-6 and H4), 7.36 (m, H-2), 4.84 (brs, CH2); 13C NMR (500 MHz, in CDCl3), δ 188.4 (s, C-9), 182.6 (s, C10), 163.2 (s, C-1), 151.1 (s, C-3), 133.8 (d, C-6), 134.4 (d, C-7), 133.4 (s, C-4a), 134.8 (s, C-8a), 133.7 (s, C10a), 127.1 (d, C-8), 127.1 (d, C-5), 121.4 (d, C-2), 121.4 (d, C-4), 117.5 (s, C-9a), 64.4 (t, CH2OH), these data were in agreement with those previously reported;7,12−14 ESIMS (+) m/z 255 [M + H]+. 1,7-Dihydroxy-3-methylanthracene-9,10-dione (7). 1H NMR (500 MHz, in CDCl3), δ 12.49 (brs, OH-1), 8.14 (d, J = 8.6 Hz, H-5), 7.60 (s, H-8), 7.57 (s, H-4), 7.15 (d, J = 8.6, H-6), 7.04 (s, H2), 2.78 (s, CH3); 13C NMR (500 MHz, in CDCl3), δ 163.2 (s, C-1), 123.7 (d, C-2), 149.7 (s, C-3), 120.7 (d, C-4), 133.9 (s, C-4a), 129.5 (d, C-5), 121.6 (d, C-6), 165.5 (s, C-7), 112.2 (d, C-8), 133.9 (s, C8a), 188.5 (s, C-9), 115.1 (s, C-9a), 181.5 (s, C-10), 126.6 (s, C-10a), 29.4 (q, CH3) these data were in agreement with those previously reported;7,10 ESIMS (+) m/z 255 [M + H]+. Phomarin (8). 1H NMR (500 MHz, in CDCl3), δ 12.70 (br s, OH-1), 8.24 (d, J = 8.6 Hz, H-8), 7.66 (d, J = 2.6 Hz, H-5), 7.63 (s, H-4), 7.22 (dd, J = 8.6, 2.6 Hz, H-7), 7.09 (s, H-2), 2.46 (s, CH3), these data were in agreement with those previously reported;7,8,10 ESIMS (+) m/z 255 [M + H]+.

Crystal Structure Determination of Lentiquinones B and C (2 and 3). Single crystals of lentiquinones B and C (2 and 3) suitable for X-ray structure analysis were obtained by slow addition of water to a solution of 2 (or 3) in MeOH, followed by slow evaporation of the aqueous alcohol mixture. One selected crystal of 2 (or 3) was mounted in flowing N2 at 173 K on a Bruker-Nonius KappaCCD diffractometer equipped with Oxford Cryostream apparatus (graphite monochromated Mo Kα radiation λ = 0.71073 Å, CCD rotation images, thick slices, φ and ω scans to fill the asymmetric unit. A semiempirical absorption correction (multiscan, SADABS) was applied. Both the two structures were solved by direct methods using the SIR97 program42 and anisotropically refined by the full matrix least-squares method on F2 against all independent measured reflections using the SHELXL-2016/6 program.43 One solvent crystallization water molecule was found in the asymmetric unit of 2. All the hydroxy H atoms in 2 and 3 and the water H atoms in 2 were located in difference Fourier maps and freely refined with Uiso(H) equal to 1.2Ueq of the carrier atom. All the other H atoms were placed in calculated positions and refined accordingly to a riding model (C−H distances in the range 0.95−1.00 Å and Uiso(H) equal to 1.2U eq of the carrier atom). Relative configuration: 2R,3S,4S,4aS,10R (compound 2) and 2S,3S,4S,4aS,10R (compound 3). For 2 and 3 the X-ray diffraction data were collected according to the literature methods recently reported to assign the absolute configuration in light-atom structures when Mo Kα radiation is used.21 For 2 and 3 it was not possible to obtain the calculated absolute structure factors at the level of precision necessary to definitively assign the absolute configuration.20,22 The absolute structure factors were calculated using the programs SHELXL2016/6 and PLATON-v30118. Crystallographic Data of 2. C15H16O6·H2O, M = 310.29, orthorhombic, P212121, a = 5.0230(13) Å, b = 6.942(3) Å, c = 40.656(6) Å, α = β = γ = 90°, V = 1417.7(7) Å3, T = 173(2) K, Z = 4, Dcalcd = 1.454 Mg/m3, crystal size 0.50 × 0.25 × 0.10 mm3, F(000) = 656, absorption coefficient 0.116 mm−1, reflections collected 29666, independent reflections 4820 [Rint = 0.0328], final R indices [I > 2σ(I)] R1 = 0.0509, wR2 = 0.1195, R indices (all data) R1 = 0.0571, wR2 = 0.1221. Flack x determined using 1563 quotients: 0.0(2). Parsons z: 0.0(2), Hooft y: 0.0(2), Bijvoet pairs 1941. Crystallographic Data of 3. C15H16O6, M = 292.28, monoclinic, P21, a = 5.5150(11) Å, b = 20.433(3) Å, c = 5.8530(7) Å, α = 90°, β = 100.075(15)°, γ = 90°, V = 649.39(18) Å3, T = 173(2) K, Z = 4, Dcalcd = 1.495 Mg/m3, crystal size 0.35 × 0.35 × 0.25 mm3, F(000) = 308, absorption coefficient 0.116 mm−1, reflections collected 31356, independent reflections 5691 [Rint = 0.0255], final R indices [I > 2σ(I)] R1 = 0.0341, wR2 = 0.0906, R indices (all data) R1 = 0.0380, wR2 = 0.0932. Flack x determined using 2455 quotients: 0.14(12). Parsons z: 0.17(12), Hooft y: 0.20(11), Bijvoet pairs 2773. The crystallographic data for 2 and 3 have been deposited in the Cambridge Crystallographic Data Centre with deposition numbers CCDC 1832451 and 1832452 respectively. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/. Biological Assays. Owing to different solubility, the seven compounds tested in different bioassays were dissolved in MeOH (lentiquinone A−C and lentisone, 1−4, respectively) or DMSO (pachybasin, ω-hydroxypachybasin, 1,7-dihidroxy-3-methylantharacene-9,10-dione and phomarin, 5−8, respectively). Suitable control treatments with the two solvents were prepared for comparison. Leaf Puncture Assay. The metabolites were tested at 2 μg/μL concentration on several plant species belonging to a number of botanic families, as reported in Table 2. Briefly, droplets (20 μL) of the solution containing each compound were applied to detached leaves punctured with a needle. Five replications were used for each plant species tested. Leaves were kept in a moistened chamber under continuous fluorescent lights at 25 °C. The eventual appearance of symptoms, consisting in circular necrosis or chlorosis, was observed 3 days after droplet application. Toxicity effects were expressed by using a visual scale from 0 (no symptoms) to 4 (wide necrosis up to 1 cm diameter) 2707

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

using the program SpecDis,48,49 applying the dipole-length rotational strengths formalism; the difference with dipole-velocity values was negligible in all cases.

Assay on Lepidium sativum Rootlet Elongation. Seeds of L. sativum were kept for 5 min in a hypochlorite solution (1%), then rinsed for some minutes under a continuous flux of tap water. They were placed on Petri dishes, on a wet paper disk, at 25 °C in the dark for 48 h, to germinate. Ten uniformly germinated and healthy seedlings were selected and transferred to each small Petri dish, containing the paper disk wetted with 2 mL of the test solution. Each compound was tested at 2 × 10−4 M concentration, in triplicate. Disks were kept under continuous fluorescent lights for 3 days. Thus, the rootlet length was measured, and the results expressed as reduction percentage in comparison to the respective control. Lemna minor Assay. A bioassay for detecting the effects of the metabolites on chlorophyll degradation was performed on Lemna minor by using a protocol previously described.44 All the compounds were tested at 2 μg/μL concentration, by using 100 μL per each pot. Chlorophyll content was determined as previously reported, and expressed as percentage of chlorophyll content reduction in comparison to the respective control. Assay on Seed Germination of the Parasitic Weed Phelipanche ramosa. The assay was carried out according to an experimental scheme previously described.45 The compounds were tested on conditioned P. ramosa seeds at 10 −3 M concentration, by contemporaneous application of the synthetic stimulant GR24. After 5 days, the number of germinated seeds was determined by direct observation under a stereomicroscope, and the results were expressed as germination percentage in comparison to the control. Visual estimation of the germination tubes was also performed. The experiment was carried out in duplicate. Antifungal Assay. For the antifungal bioassay, the following experiment was set up. PDA (150 μL) was added under sterile conditions to each well of a 96-well-plate. Five μL of solvent containing 10 μg of each metabolite was placed in each well. A fungal conidial suspension was obtained by placing a PDA plug (around 1 cm diameter) with the actively growing mycelia of each fungus in an Eppendorf tube containing 1 mL water and mixing the material by using a Vortex. Five μL of the fungal suspension was added to each plate. A total of nine fungal strains were tested with the eight compounds, with four replications each, namely, Verticillium dahliae, Fusarium solani, Aspergillus carbonarius, Penicillium allii, Colletotrichum fioriniae, Rhizoctonia sp., Phoma exigua, Alternaria alternata, and Cladosporium halotolerans. All the strains were made available from the ISPA fungal collection. The eventual appearance of the fungal colony overgrowing PDA was visually evaluated after 2 days. Antibiotic Assay. This assay was performed on two species by using a published protocol.46 Four μL of the solvent (MeOH or DMSO) containing each compound at 10−1 M was applied to the diskette. The eventual appearance of a growth inhibition halo around the diskette, revealing an antibiotic effect, was visually estimated 24 h after the application of the compounds. Computational Section. MMFF and preliminary DFT calculations were run with Spartan’16 (Wavenfunction, Irvine, CA, USA, 2016) with standard parameters and convergence criteria; DFT and TDDFT calculations were run with Gaussian’16 with default grids and convergence criteria.47 Conformational searches were run with the Monte Carlo algorithm implemented in Spartan’16 using Merck molecular force field (MMFF). All structures thus obtained were first optimized with the DFT method at the ωB97X-D/6-31G(d) level in vacuo, then reoptimized at the ωB97X-D/6-311+G(d,p) level in vacuo. Final optimizations were run at the ωB97X-D/6-311+G(d,p) level including the universal solvent model (SMD) for MeCN. The procedure afforded two minima for compound 2, the most stable of which had a population >98%, and three minima for compound 3, the two most stable of which had an overall population >99%. For the solid-state ECD/TDDFT approach, the X-ray geometry of 3 was refined by optimizing only the hydrogen atoms at ωB97X-D/631G(d) level in vacuo. TDDFT calculations were run with various functionals (B3LYP, CAM-B3LYP, M06) and def2-TZVP basis set, including SMD for MeCN. Average ECD spectra were computed by weighting component ECD spectra with Boltzmann factors at 300 K estimated from DFT internal energies. ECD spectra were generated



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.8b00556.



NMR, HRESIMS, IR and UV spectra of compounds 1, 2, and 3 (PDF)

AUTHOR INFORMATION

Corresponding Author

*Tel: +39 081 2539178. Fax: +39 081 674330. E-mail: [email protected]. ORCID

Marco Masi: 0000-0003-0609-8902 Paola Nocera: 0000-0002-2102-3629 Maria C. Zonno: 0000-0002-8245-1723 Gennaro Pescitelli: 0000-0002-0869-5076 Alessio Cimmino: 0000-0002-1551-4237 Angela Boari: 0000-0003-1444-019X Alessandro Infantino: 0000-0003-0048-1257 Maurizio Vurro: 0000-0001-6875-4093 Antonio Evidente: 0000-0001-9110-1656 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The work was carried out with academic grants from the Dipartimento Scienze Chimiche, Università di Napoli Federico II, Napoli, Italy and in the frame of Programme STAR 2017 financially supported by UniNA and Compagnia di San Paolo. Antonio Evidente is associated with “Istituto di Chimica Biomolecolare del CNR”, Pozzuoli, Italy. G. P. acknowledges the CINECA award under the ISCRA initiative, for the availability of high performance computing resources and support.



REFERENCES

(1) Rubiales, D.; Fondevilla, S. Front. Plant Sci. 2012, 3, 27. (2) Tivoli, B.; Baranger, A.; Avila, C. M.; Banniza, S.; Barbetti, M.; Chen, W.; Davidson, J.; Lindeck, K.; Kharrat, M.; Rubiales, D.; Sadiki, M.; Sillero, J. C.; Sweetingham, M.; Muehlbauer, F. J. Euphytica 2006, 147, 223−253. (3) Evidente, A.; Berestetskiy, A.; Cimmino, A.; Tuzi, A.; Superchi, S.; Melck, D.; Andolfi, A. J. Agric. Food Chem. 2009, 57, 11168− 11173. (4) Infantino, A.; Zaccardelli, M.; Costa, C.; Ozkilinc, H.; Habibi, A.; Peever, T. J. Plant Pathol. 2016, 98, 541−548. (5) Masi, M.; Nocera, P.; Boari, A.; Cimmino, A.; Zonno, M. C.; Infantino, A.; Vurro, M.; Evidente, A. J. Nat. Prod. 2018, 81, 1093− 1097. (6) Andolfi, A.; Cimmino, A.; Villegas-Fernández, A. M.; Tuzi, A.; Santini, A.; Melck, D.; Rubiales, D.; Evidente, A. J. Agric. Food Chem. 2013, 61, 7301−7308. (7) Sun, P.; Huo, J.; Kurtán, T.; Mándi, A.; Antus, S.; Tang, H.; Draeger, S.; Schulz, B.; Hussain, H.; Krohn, K.; Pan, W. Chirality 2013, 25, 141−148. (8) Bick, I. R. C.; Rhee, C. Biochem. J. 1966, 98, 112−116. (9) De Stefano, S.; Nicoletti, R.; Beato, O. Tabacco 1999, 7, 21−24.

2708

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709

Journal of Natural Products

Article

(10) De Souza Borges, W.; Pupo, T.; Novel, M. J. Braz. Chem. Soc. 2006, 17, 929−934. (11) Xia, X. K.; Huang, H. R.; She, Z. G.; Shao, C. L.; Liu, F.; Cai, X. L.; Vrijmoed, L. L. P.; Lin, Y. C. Magn. Reson. Chem. 2007, 45, 1006− 1009. (12) Imre, S.; Sar, S.; Thomson, R. H. Phytochemistry 1976, 15, 317−320. (13) Kuo, S. C.; Chen, P. R.; Lee, S. W.; Chen, Z. T. J. Chin. Chem. Soc. 1995, 42, 869−871. (14) Nanthaphong, K.; Rukachaisirikul, V.; Tadpetch, K.; Kaewpet, M.; Phongpaichit, S.; Preedanon, S.; Sakayaroj, J. Arch. Pharmacal Res. 2012, 35, 461−468. (15) Nakanishi, K.; Solomon, P. H. Infrared Absorption Spectroscopy, 2nd ed.; Holden Day: Oakland, 1977; pp 17−44. (16) Pretsch, E.; Buhlmann, P.; Affolter, C. Structure Determination of Organic CompoundsTables of Spectral Data, 3rd ed.; SpringerVerlag: Berlin, 2000; pp 161−243. (17) Berger, S.; Braun, S. 200 and More Basic NMR ExperimentsA Practical Course, 1st ed.; Wiley-VCH: Weinheim, 2004. (18) Breitmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy; VCH: Weinheim, 1987; pp 183−280. (19) Sternhell, S. Q. Rev., Chem. Soc. 1969, 23, 236−270. (20) Parsons, S.; Flack, H. D.; Wagner, T. Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2013, B69, 249−259. (21) Escudero-Adàn, E. C.; Benet-Buchholz, J.; Ballester, P. Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 2014, B70, 660−668. (22) Parsons, S. Tetrahedron: Asymmetry 2017, 28, 1304−1313. (23) Superchi, S.; Scafato, P.; Górecki, M.; Pescitelli, G. Curr. Med. Chem. 2018, 25, 287−320. (24) Pescitelli, G.; Bruhn, T. Chirality 2016, 28, 466−474. (25) Pescitelli, G.; Kurtán, T.; Flörke, U.; Krohn, K. Chirality 2009, 21, E181−E201. (26) Padula, D.; Di Pietro, S.; Capozzi, M. A. M.; Cardellicchio, C.; Pescitelli, G. Chirality 2014, 26, 462−470. (27) Andolfi, A.; Maddau, L.; Basso, S.; Linaldeddu, B. T.; Cimmino, A.; Scanu, B.; Deidda, A.; Tuzi, A.; Evidente, A. J. Nat. Prod. 2014, 77, 2352−2360. (28) Cala, A.; Masi, M.; Cimmino, A.; Molinillo, J. M. G.; Macias, F. A.; Evidente, A. Molecules 2018, 23, 1529−1531. (29) Evidente, A.; Sparapano, L.; Fierro, O.; Bruno, G.; Giordano, F.; Motta, A. Phytochemistry 1998, 48, 1139−1143. (30) Evidente, A.; Maddau, L.; Scanu, B.; Andolfi, A.; Masi, M.; Motta, A.; Tuzi, A. J. Nat. Prod. 2011, 74, 757−763. (31) Fernández-Aparicio, M.; Masi, M.; Maddau, L.; Cimmino, A.; Evidente, M.; Rubiales, D.; Evidente, A. J. Agric. Food Chem. 2016, 64, 5188−5196. (32) Jamison, D. S.; Yoder, J. I. Plant Physiol. 2001, 125, 1870− 1879. (33) Palmer, A. G.; Gao, R.; Maresh, J.; Erbil, W. K.; Lynn, D. G. Annu. Rev. Phytopathol. 2004, 42, 439−464. (34) Duval, J.; Pecher, V.; Poujol, M.; Lesellier, E. Ind. Crops Prod. 2016, 94, 812−833. (35) Fouillaud, M.; Venkatachalam, M.; Girard-Valenciennes, E.; Caro, Y.; Dufossé, L. Mar. Drugs 2016, 14, 64. (36) Dewick, P. Medicinal Natural Products: A Biosynthetic Approach; John Wiley & Sons: New York, 2002. (37) Osbourn, A. E.; Lanzotti, V. Plant-Derived Natural Products; Springer: Dordrecht, 2009. (38) Liu, S. Y.; Lo, C. T.; Shibu, M. A.; Leu, Y. L.; Jen, B. Y.; Peng, K. C. J. Agric. Food Chem. 2009, 57, 7288−7292. (39) Rai, M.; Deshmukh, P.; Gade, A.; Ingle, A.; Kövics, G. J.; Irinyi, L. Crit. Rev. Microbiol. 2009, 35, 182−196. (40) Kim, W.; Peever, T. L.; Park, J.-J.; Park, C.-M.; Gang, D. R.; Xian, M.; Davidson, J. A.; Infantino, A.; Kaiser, W. J.; Chen, W. Sci. Rep. 2016, 6, 20192. (41) Pinkerton, F.; Strobel, G. A. Proc. Natl. Acad. Sci. U. S. A. 1976, 73, 4007−4011.

(42) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119. (43) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, A64, 112−122. (44) Cimmino, A.; Andolfi, A.; Zonno, M. C.; Troise, C.; Santini, A.; Tuzi, A.; Vurro, M.; Ash, G.; Evidente, A. J. Nat. Prod. 2012, 75, 1130−1137. (45) Vurro, M.; Boari, A.; Pilgeram, A.; Sands, D. Biol. Control 2006, 36, 258−265. (46) Bottalico, A.; Capasso, R.; Evidente, A.; Randazzo, G.; Vurro, M. Phytochemistry 1990, 29, 93−96. (47) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnemberg, J. L.; Williams-Young, D.; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J. Gaussian 16, Revision A.03; Gaussian, Inc.: Wallingford, CT, 2016. (48) Bruhn, T.; Schaumloeffel, A.; Hemberger, Y.; Bringmann, G. Chirality 2013, 25, 243−249. (49) Bruhn, T.; Schaumlöffel, A.; Hemberger, Y.; Pescitelli, G. SpecDis, version 1.71; Berlin, 2017, https://specdis-software.jimdo. com/.

2709

DOI: 10.1021/acs.jnatprod.8b00556 J. Nat. Prod. 2018, 81, 2700−2709