Natural zinniol derivatives from Alternaria tagetica. Isolation, synthesis, and structure-activity correlation.

Article abstract of DOI:10.1021/jf010641t

Two novel phytotoxins, 8-zinniol methyl ether (5) and 8-zinniol acetate (6), in addition to 6-(3',3'-dimethylallyloxy)-4-methoxy-5-methylphthalide (2), 5-(3',3'-dimethylallyloxy)-7-methoxy-6-methylphthalide (3), and the novel metabolites 8-zinniol 2-(phenyl)ethyl ether (4) and 7-zinniol acetate (7) have been identified as natural zinniol derivatives from the organic crude extract of Alternaria tagetica culture filtrates. Using zinniol as the starting material, phytotoxin 5 was synthesized, together with a number of synthetic intermediates (8-13). Both natural and synthetic zinniol derivatives were evaluated in the leaf-spot bioassay against marigold leaves (Tagetes erecta).

Full text of DOI:10.1021/jf010641t

J. Agric. Food Chem. 2002, 50, 1053−1058 1053
Natural Zinniol Derivatives from Alternaria tagetica. Isolation,
Synthesis, and Structure−Activity Correlation
M. MARCELA GAMBOA-ANGULO,† FABIOLA ESCALANTE-EROSA,†
KARLINA GARCIÄA-SOSA,† FAÄ TIMA ALEJOS-GONZAÄ LEZ,†
GUILLERMO DELGADO-LAMAS,‡ AND LUIS M. PEN˜ A-RODRIÄGUEZ*,†
Grupo de Qu´ımica Orga´nica, Centro de Investigacio´n Cient´ıfica de Yucata´n, Calle 43 No. 130,
Colonia Chuburna´, Me´rida, Yucata´n, Mexico 97200, and Instituto de Qu´ımica, Universidad Nacional
Auto´noma de Me´xico, Circuito Exterior, Ciudad Universitaria, Me´xico D.F. 04510
Two novel phytotoxins, 8-zinniol methyl ether (5) and 8-zinniol acetate (6), in addition to 6-(3′,3′-
dimethylallyloxy)-4-methoxy-5-methylphthalide (2), 5-(3′,3′-dimethylallyloxy)-7-methoxy-6-methylphtha-
lide (3), and the novel metabolites 8-zinniol 2-(phenyl)ethyl ether (4) and 7-zinniol acetate (7) have
been identified as natural zinniol derivatives from the organic crude extract of Alternaria tagetica
culture filtrates. Using zinniol as the starting material, phytotoxin 5 was synthesized, together with a
number of synthetic intermediates (8-13). Both natural and synthetic zinniol derivatives were evaluated
in the leaf-spot bioassay against marigold leaves (Tagetes erecta).
KEYWORDS: Alternaria tagetica; Tagetes erecta; marigold; phytotoxin; zinniol; zinniol derivatives; leaf-
spot bioassay
INTRODUCTION
Zinniol (1) is a phytotoxic tetraketide first isolated from
Alternaria zinniae (1) and later detected in culture filtrates from
Phoma macdonaldii (2), Alternaria dauci (3), Alternaria solani,
Alternaria porri, Alternaria carthami, Alternaria macrospora
(4), Alternaria cichorii (5), and Alternaria tagetica (6). The total
synthesis of 1 has been reported following two different
routes: one using a disulfonic acid as the starting material (7)
and another using a sequence of Diels-Alder and Alder-
Rickert reactions with dimethylacetylenedicarboxylate (DMAD)
structure (8). The synthetic preparation of zinniol has confirmed
both its structure and its phytotoxic activity.
The phytotoxic damage produced by zinniol in detached
leaves of host plants is similar to that caused by the corre-
sponding fungal pathogens in the field (4, 6). It has been reported
that the presence of the two hydroxymethyl groups in the
structure of 1 is essential for the expression of its phytotoxic
activity (3). Structure-activity studies carried out by preparing
and testing zinniol derivatives can improve our knowledge of
the structural requirements needed for the full expression of
biological activity and our understanding of the role played by
this phytotoxin in the plant-pathogen interaction.
Figure 1. Sructure of Alternaria tagetica metabolites.
the isomeric phthalides 2 and 3, 8-zinniol 2-(phenyl)ethyl ether
(4), 8-zinniol methyl ether (5), 8-zinniol acetate (6), and
7-zinniol acetate (7). In addition, the synthesis of phytotoxin 5
by chemical transformation of zinniol and the biological
evaluation of a number of intermediate derivatives (8-13)
generated during the synthetic process are reported.
We have previously reported the isolation of 1 as the major
phytotoxic component in the organic crude extract from culture
filtrates of A. tagetica (9). Here we describe a number of
metabolites structurally related to zinniol (Figure 1), including
MATERIALS AND METHODS
General Experimental Procedures. TLC, IR, EIMS, and H, 13C,
1
and 2D NMR instrumentation and experimental procedures have been
described previously (9, 10). GC-EIMS analyses were carried out using
a Hewlett-Packard 5890 gas chromatograph coupled to a Hewlett-
Packard 5971A mass selective detector. Conditions for GC analyses
were as follows: 0.5 µL of 0.1% sample; Ultra 1 column (Hewlett-
Packard, cross-linked methyl silicone gum, 25 m × 0.32 mm i.d., 0.52
* Author to whom correspondence should be addressed (telephone +52-
999-981-3923; fax +52-999-981-3900; e-mail lmanuel@cicy.mx).
† Centro de Investigacio´n Cient´ıfica de Yucata´n.
‡ Instituto de Qu´ımica Universidad Nacional Auto´noma de Me´xico.
10.1021/jf010641t CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/02/2002

1054 J. Agric. Food Chem., Vol. 50, No. 5, 2002
Gamboa-Angulo et al.
Table 1. 1H NMR Data (δ, m, J ) Hz) for Metabolites 1, 1a, and 4−7 in CDCl3 at 500 MHz
proton
1
1aa
4
5
6
7a
6
7
8
9
6.68, br s
4.68, s
4.76, s
6.72, s
5.14, s
5.22, s
3.74, s
6.71, s
4.55, d, 6.5
4.63, s
6.73, s
4.6, br s
4.59, s
6.79, s
4.7, s
5.25, s
3.74, s
6.68, s
5.2, s
4.73, ABq
3.81, s
3.78, s
3.65, s
3.72, s
10
1′
2′
2.16, s
2.17, s
2.15, s
2.16, s
2.16, s
2.17, s
4.52, br d, 6.5
5.49, ddd, 6.5,
1.5, 1.5
4.54, br d, 6.6
5.47, ddd, 6.6,
1.2, 1.5
4.53, br d, 6.5
5.48, ddd, 6.5,
1.5, 1.5
4.54, d, 6.5
5.55, ddd, 6.5,
1.5, 1.5
4.55, d, 6.6
5.49, ddd, 6.6,
1.5, 1.5
4.53, d, 6.42
5.47, ddd, 6.3,
1.8, 1.5
3′
4′
5′
1′′
2′′
Ar−H
1.8, s
1.74, s
1.8, s
1.79, d, 1
1.73, d, 1
3.8, dd, 7, 7.5
2.9, dd, 7,7
7.18−7.29, m
1.79, s
1.74, s
3.44, s
1.78, s
1.74, s
2.05, s
1.79, d, 0.9
1.75, s
2.1, s
1.75, s
2.07, s
2.05, s
a 300 MHz.
Table 2. 13C NMR Data for Metabolites 1, 1a, and 4−7 in CDCl3 at
125 MHz
µm film thickness); flow rate, 1 mL/min (He); temperature program,
T1 ) 180 °C, T2 ) 290 °C, gradient ) 10 °C/min; detector temperature,
300 °C.
HPLC Instrumentation and Analyses. (1) A Waters Delta prep
4000 preparative chromatography system (Millipore Corp.) equipped
with a Waters 486 UV detector for semipreparative separations and a
Waters 996 photodiode array detector was used for analytical studies.
Control of the equipment, data acquisition, and processing and
management of chromatographic information were performed by the
Millennium 2000 software program (Waters). Conditions for HPLC
analyses were as follows: Microporasil normal phase column (Waters,
10 µ, 300 × 3.9 mm i.d.), using hexane/EtOAc 85:15 (method a) or
CHCl3/EtOAc 90:10 (method b); flow rate ) 1.0 mL/min; and UV
detection at 260 nm.
(2) A Milton Roy CM-4000 series chromatograph was equipped with
a Milton Roy SM-4000 UV detector and a Milton Roy CS-4100
integrator. Conditions for HPLC analyses were as follows: Hypersil
ODS C18 column (Alltech, 5 µ, 250 × 4.6 mm i.d.); gradient elution
using CH3CN/H2O 10:90 to 90:10 (15 min, method c); flow rate ) 1.0
mL/min; and UV detection at 254 nm.
Isolation and Purification of Metabolites 1)6. Details of the
microbiological work, leaf-spot assay, and fractionation of the organic
crude extract by solvent partition and VLC purification (13 fractions)
have been described previously (9, 10). HPLC separations were carried
out using a preparative column (300 × 19.5 mm i.d.) packed with the
same microporasil normal phase as that used for analytical studies. After
VLC of the organic crude extract, purification of fraction 5 by HPLC
(method a, flow ) 18 mL/min) yielded metabolite 2 in pure form (0.73
mg/L of culture filtrate). Fraction 6, after successive HPLC purifications
(method a, flow of 18 and 15 mL/min), produced pure 3 (0.21 mg/L)
and 4 (0.17 mg/mL). A similar purification of fraction 7 (method a,
flow of 15 mL/min) resulted in the isolation of 5 (0.28 mg/mL) as a
single component. Finally, purification of phytotoxic fraction 9, using
the same column but method b and a flow of 16 mL/min, yielded 6
(0.40 mg/L) and 7 (0.19 mg/L).
C no.
1
1aa
4
5
6
7a
1
2
3
4
5
6
7
138.9
124.7
158.2
120.0
157.7
109.0
64.7
137.7
120.7
159.1
119.7
158.4
109.3
64.1
140.3
121.1
158.3
119.3
158.0
109.0
64.4
140.3
121.2
158.4
119.4
158.1
109.1
64.5
139.5
118.5
159.1
119.7
158.7
108.0
63.1
139.5
125.3
157.8
121.0
157.8
109.4
64.6
8
56.8
58.4
64.8
66.5
58.7
56.8
9
61.9
61.7
61.7
61.8
61.7
61.8
10
1′
2′
3′
4′
5′
1′′
2′′
9.2
65.3
119.9
137.6
25.8
9.3
65.3
119.7
134.4
25.7
18.2
170.9
21.0
170.6
21.0
9.2
65.3
120.0
137.5
25.8
18.2
71.4
36.4
9.2
65.3
119.9
137.5
25.8
9.3
65.4
119.8
137.7
25.8
18.3
171.1
21.2
9.3
65.4
119.8
137.7
25.8
18.2
170.9
21.2
18.2
18.2
58.0
138.6b
128.9b (2)
128.4b (2)
126.3b
a 75 MHz. b Ar−C.
(HPLC, method c) 18.27 min; tR (GC) 8.23 min; IR (film) νmax 3433
(OH), 1604 (CdC), 1214 (C-O), 1114 (C-O-C) cm-1; EIMS (70
eV), m/z 280 (M+, 8), 212 (M+ - C5H8, 36), 194 (M+ - C5H8 - H2O,
1
10), 180 (100), 69 (C5H8 + H, 25); H NMR data, see Table 1; 13C
NMR data, see Table 2.
8-Zinniol acetate (6) was obtained as a yellow oil: Rf 0.27 (CH2-
Cl2/EtOAc 9:1); tR (HPLC, method b) 14.9 min; IR (film) νmax 3435
(OH), 1739 (CdO), 1604 (CdC), 1232 (C-O), 1115 (C-O-C) cm-1
;
-
EIMS (70 eV), m/z 308 (M+, 7), 240 (M+ - C5H8, 12), 222 (M+
6-(3′,3′-Dimethylallyloxy)-4-methoxy-5-methylphthalide (2) was
obtained as a white amorphous powder: Rf 0.24 (CHCl3) and 0.39
(hexane/EtOAc 8:2); tR (HPLC, method a) 9.47 min; IR, EIMS, and
1H and 13C NMR spectra were identical to those reported in the literature
(9-11).
C5H8 - H2O, 5), 180 (100), 162 (180 - H2O, 32), 69 (C5H8 + H, 42);
1H NMR data, see Table 1; 13C NMR data, see Table 2.
7-Zinniol acetate (7) was obtained as a pale yellow oil: Rf 0.25
(CH2Cl2/EtOAc 9:1); tR (HPLC, method b) 17.2 min; IR (film) νmax
3435 (OH), 1739 (CdO), 1604 (CdC), 1232 (C-O), 1115 (C-O-C)
cm-1; EIMS (70 eV), m/z 308 (M+, 7), 240 (M+ - C5H8, 15), 222
(M+ - C5H8 - H2O, 5), 180 (100), 162 (180 - H2O, 11), 69 (C5H8 +
5-(3′,3′-Dimethylallyloxy)-7-methoxy-6-methylphthalide (3) was
obtained as a colorless oil: Rf 0.1 (CHCl3) and 0.32 (hexane/EtOAc
8:2); tR (HPLC, method a) 13.3 min; IR, EIMS, and 1H and 13C NMR
spectra were identical to those reported in the literature (10, 12).
8-Zinniol 2-(phenyl)ethyl ether (4) was obtained as a colorless
oil: Rf 0.55 (C6H6/acetone 95:5); tR (HPLC, method a) 14.5 min; IR
(film) νmax 3445 (OH), 3020 (Ar-H), 1605 (CdC), 1214 (C-O), 1148
(C-O-C) cm-1; EIMS (70 eV), m/z 370 (M+, 8), 302 (M+ - C5H8,
8), 284 (M+ - C5H8 - H2O, 8), 180 (100), 105 (C8H9, 22), 69 (C5H9,
1
H, 26); H NMR data, see Table 1; 13C NMR data, see Table 2.
Oxidation of Zinniol (1). A solution of 1 (616.8 mg, 2.31 mmol)
in acetone (485 mL) was stirred with an excess of a stock chromic
acid solution (11.0 g of chromium trioxide in 10 mL of concentrated
sulfuric acid diluted to 50 mL with water) at room temperature for 5
min (1). After dilution with water, the oxidation crude product was
obtained by ethyl ether extraction of the aqueous suspension (three
times, 2:1). The combined organic layer were washed with water and
1
18); H NMR data, see Table 1; 13C NMR data, see Table 2.
8-Zinniol methyl ether (5) was obtained as a white amorphous
powder: Rf 0.3 (hexane/acetone 8:2); tR (HPLC, method a) 19.1 min,

Phytotoxins from Alternaria
J. Agric. Food Chem., Vol. 50, No. 5, 2002 1055
saturated sodium chlorure solution. The solvents were then removed,
and the mixture of three products was filtered through a silica gel bed
and separated by preparative HPLC (method a) to yield 2 (168.2 mg,
28%), 3 (176.7 mg, 29%), and 8 (117.9 mg, 18%), in pure form.
Zinniol anhydride (8) was obtained as a white amorphous pow-
der: Rf 0.52 (CHCl3); tR (HPLC, method a) 4.4 min; IR (film) νmax
2939 (C-H), 1838 (anh CdO), 1772 (anh CdO), 1614 (CdC), 1214
(C-O), 1114 (C-O-C) cm-1; EIMS (70 eV), m/z 276 (M+, 3), 208
saturated NH4Cl solution, diluted with water, and extracted with EtOAc
(three times, 2:1, 1:1, 1:1, v/v). The organic layer was washed with
water and brine, dried (Na2SO4), and filtered. The solvent was removed
under reduced pressure to produce 11 (14.1 mg, 57%). The residual
aqueous layer was neutralized with H2SO4 (2 N) and extracted with
EtOAc (three times, 2:1, 1:1, 1:1, v/v). The organic layer was washed
with water and brine, dried, and concentrated to yield 12 (5 mg, 21%).
Methyl-2-methoxy-3-methyl-4-(3′,3′-dimethylallyloxy)-6-meth-
oxymethyl benzoate (11) was obtained as a pale yellow oil: Rf 0.44
(hexane/acetone 8:2); tR (GC) 8.75 min; IR (CHCl3) νmax 3010 (Ar-
H), 1727 (CdO), 1602 (CdC), 1114 (C-O-C) cm-1; EIMS (70 eV),
m/z 308 (M+, 1), 240 (M+ - C5H8, 61), 208 (240 - CH3O - H, 100),
1
(M+ - C5H8, 5), 164 (M+ - C5H8 - CO2, 8), 69 (C5H9, 100); H
NMR (300 MHz, CDCl3) δ 7.15 (1H, s, H-6), 5.47 (1H, ddd, J ) 6.6,
6.6, 1.5 Hz, H-2′), 4.68 (2H, br d, J ) 6.6 Hz, H-1′), 4.16 (2H, s,
H-9), 2.21 (3H, s, H-10), 1.82 (3H, br d, J ) 0.9 Hz, H-4′), 1.75 (3H,
s, H-5′); 13C NMR (75 MHz, CDCl3) δ 163.5 (C-8), 164.5 (C-7), 160.9
(C-3), 157.9 (C-5), 139.6 (C-1), 131.9 (C-3′), 118.2 (C-2′), 112.1 (C-
4), 103.1 (C-6), 66.4 (C-1′), 62.3 (C-9), 25.8 (C-4′), 18.3 (C-5′), 9.6
(C-10).
1
193 (208 - CH3, 66); H NMR (300 MHz, CDCl3) δ 6.71 (1H, s,
H-6), 5.47 (1H, ddd, J ) 6.5, 6.5, 1.5 Hz, H-2′), 4.54 (2H, d, J ) 6.5
Hz, H-1′), 4.46 (1H, br s, H-7), 3.89 (3H, s, H-2′′), 3.76 (3H, s, H-9),
3.56 (3H, s, H-1′′), 2.14 (3H, s, H-10), 1.79 (3H, s, H-4′), 1.74 (3H, s,
H-5′); 13C NMR (75 MHz, CDCl3) δ 168.4 (C-8), 159.1 (C-3), 157.0
(C-5), 137.8 (C-3′), 135.6 (C-1), 119.7 (C-2′), 119.6 (C-4), 119.1 (C-
2), 106.8 (C-6), 72.4 (C-7), 65.4 (C-1′), 61.9 (C-9), 58.3 (C-1′′), 52.0
(C-2′′), 25.8 (C-4′), 18.3 (C-5′), 9.0 (C-10).
Hydrolysis (13) and Methylation (14) of 6-(3′,3′-Dimethylallyl-
oxy)-4-methoxy-5-methylphthalide. A mixture of 2 (25 mg, 0.095
mmol), MeOH (10 mL, Aldrich), and an excess of K2CO3 (Merck)
was heated under reflux for 7 h. The solvent was removed in vacuo.
The residue was combined with 4 mL of dry THF (Aldrich) under
nitrogen, and then a suspension of 180 mg of NaH (80% in mineral
oil, 6 mmol, Aldrich), previously washed with dry THF, was added.
The suspension was stirred at room temperature for 30 min, and 300
µL of CH3I (4.82 mmol, Baker) was added slowly. The mixture was
kept at room temperature until no starting material could be detected
by TLC (43 h). The reaction mixture was quenched with saturated NH4-
Cl solution, diluted with water, and extracted with EtOAc (three times,
2:1, 1:1, 1:1, v/v). The organic layer was washed with water to
neutralization, then with brine, followed by treatment with Na2SO4 and
filtration. The solvent was finally removed in vacuo, yielding 9 (12.2
mg, 42%). The remaining aqueous layer was neutralized with H2SO4
(2 N) and extracted with EtOAc (three times, 2:1, 1:1, 1:1, v/v). The
organic layer was washed with water and brine, dried, concentrated,
and filtered through a silica gel microcolumn (hexane/acetone 8:2) to
yield 10 (3.7 mg, 13%).
2-Methoxy-3-methyl-4-(3′,3′-dimethylallyloxy)-6-methoxymeth-
ylbenzoic acid (12) was obtained as a white solid: Rf 0.12 (hexane/
acetone 8:2); tR (GC) 9.68 min; IR (CHCl3) νmax 3400-2600 (COOH),
3008 (Ar-H), 1715 (CdO), 1603 (CdC), 1114 (C-O-C) cm-1; EIMS
(70 eV), m/z 294 (M+, 2), 226 (M+ - C5H8, 13), 208 (226 - H2O,
1
10), 193 (208 - CH3, 100), 179 (11), 165 (12), 69 (C5H9, 23); H
NMR (300 MHz, CDCl3) δ 7.07 (1H, s, H-6), 5.47 (1H, ddd, J ) 6.6,
6.6, 1.5 Hz, H-2′), 4.81 (1H, s, H-7), 4.62 (2H, d, J ) 6 Hz, H-1′),
3.86 (3H, s, H-9), 3.48 (3H, s, H-1′′), 2.17 (3H, s, H-10), 1.80 (3H, br
d, J ) 1.2 Hz, H-4′), 1.77 (3H, s, H-5′); 13C NMR (75 MHz, CDCl3)
δ 166.0 (C-8), 161.0 (C-3), 158.3 (C-5), 142.1 (C-1), 138.3 (C-3′),
119.3 (C-4 and C-2′), 118.5 (C-2) 107.0 (C-6), 72.9 (C-7), 65.5 (C-
1′), 62.5 (C-9), 58.6 (C-1′′), 25.8 (C-4′), 18.3 (C-5′), 9.0 (C-10).
7-Zinniol Methyl Ether (13). A suspension of LiAlH4 (26 mg, 0.68
mmol) in dry THF (5 mL), under N2, in an ice bath, was stirred while
a solution of 11 (10 mg, 0.32 mmol) in THF was added dropwise.
After 5 min, the excess reagent was eliminated by successive additions
of THF/water (9:1) until a white precipitate appeared. The reaction
mixture was filtered, and the filtrate was extracted with ethyl acetate
(three times, 2:1, 1.1, 1:1, v/v). The organic layer was washed with
water and brine, dried, and evaporated to afford the crude reduction
product, which was filtered through silica gel (hexane/acetone) giving
13 (4.1 mg, 46%) in pure form: Rf 0.3 (hexane/acetone 8:2); tR (GC)
8.21 min; tR (HPLC, method c) 19.09 min; IR (CHCl3) νmax 3468 (OH),
3009 (Ar-H), 1606 (CdC), 1116 (C-O-C) cm-1; EIMS (70 eV),
m/z 212 (M+ - C5H8, 2), 194 (M+ - C5H11 - H2O, 5), 180 (100), 163
(30), 69 (C5H9, 100); 1H NMR (300 MHz, CDCl3) δ 6.64 (1H, s, H-6),
5.49 (1H, ddd, J ) 6.6, 6.6, 1.2 Hz, H-2′), 4.67 (2H, s, H-8), 4.52 (2H,
d, J ) 6 Hz, H-1′), 4.52 (2H, s, H-7), 3.81 (3H, s, H-9), 3.43 (3H, s,
H-1′′), 2.18 (3H, s, H-10), 1.79 (3H, br d, J ) 0.9 Hz, H-4′), 1.74 (3H,
s, H-5′); 13C NMR (75 MHz, CDCl3) δ 158.4 (C-3), 157.3 (C-5), 137.5
(C-3′), 135.3 (C-1), 126.1 (C-2), 120.5 (C-4), 120.0 (C-2′), 109.4 (C-
6), 74.4 (C-7), 65.4 (C-1′), 61.9 (C-9), 56.9 (C-8), 25.8 (C-4′), 18.2
(C-5′), 9.2 (C-10).
Methyl-3-methoxy-4-methyl-5-(3′,3′-dimethylallyloxy)-2-meth-
oxymethyl benzoate (9) was obtained as a yellow oil: Rf 0.49 (hexane/
acetone 8:2); tR (GC) 8.88 min; IR (CHCl3) νmax 3015 (Ar-H), 1730
(CdO), 1607 (CdC), 1118 (C-O-C) cm-1; EIMS (70 eV), m/z 225
(M+ - C5H8 - CH3, 9), 193 (M+ - C5H8 - H2O, 100), 179 (3), 69
1
(C5H9, 71); H NMR (300 MHz, CDCl3) δ 7.14 (1H, s, H-6), 5.47
(1H, ddd, J ) 6.6, 6.6, 1.2 Hz, H-2′), 4.71 (2H, s, H-8), 4.54 (2H, d,
J ) 6.6 Hz, H-1′), 3.9 (3H, s, H-1′′), 3.76 (3H, s, H-9), 3.39 (3H, s,
H-2′′), 2.19 (3H, s, H-10), 1.79 (3H, s, H-4′), 1.75 (3H, s, H-5′); 13C
NMR (75 MHz, CDCl3) δ 168.1 (C-7), 158.8 (C-3), 157.3 (C-5), 137.9
(C-3′), 129.8 (C-1), 124.6 (C-2), 124.2 (C-4), 119.6 (C-2′), 109.3 (C-
6), 65.7 (C-8), 65.4 (C-1′), 61.9 (C-9), 58.2 (C-2′′), 52.1 (C-1′′), 25.8
(C-4′), 18.3 (C-5′), 9.5 (C-10).
3-Methoxy-4-methyl-5-(3′,3′-dimethylallyloxy)-2-methoxymeth-
yl benzoic acid (10) was obtained as a white solid: Rf 0.14 (hexane/
acetone 8:2); tR (GC) 10.54 min; IR (CHCl3) νmax 3400-2600 (COOH),
3008 (Ar-H), 1716 (CdO), 1603 (CdC), 1114 (C-O-C) cm-1; EIMS
(70 eV), m/z 226 (M+ - C5H8, 10), 211 (M+ - C5H8 - CH3, 18), 193
(211 - H2O, 100), 165 (11), 69 (C5H9, 61); 1H NMR (300 MHz, CDCl3)
δ 7.23 (1H, s, H-6), 5.47 (1H, ddd, J ) 6.6, 6.6, 1.2 Hz, H-2′), 4.71
(2H, s, H-8), 4.58 (2H, d, J ) 6.6 Hz, H-1′), 3.74 (3H, s, H-9), 3.49
(3H, s, H-1′′), 2.21 (3H, s, H-10), 1.80 (3H, br d, J ) 1.2 Hz, H-4′),
1.75 (3H, s, H-5′); 13C NMR (75 MHz, CDCl3) δ 168.9 (C-7), 158.3
(C-3), 158.1 (C-5), 138.2 (C-3′), 131.3 (C-1), 125.1 (C-2), 121.1 (C-
4), 119.4 (C-2′), 110.7 (C-6), 66.1 (C-7), 65.3 (C-1′), 62.0 (C-9), 58.0
(C-1′′), 25.8 (C-4′), 18.3 (C-5′), 9.7 (C-10).
8-Zinniol Methyl Ether (5). A solution of 9 (4 mg, 0.013 mmol)
in anhydrous THF (3 mL) was treated with LiAlH4 (12.4 mg, 0.32
mmol) for 2 h as described above, to give 5 (1.9 mg, 49%).
Phytotoxicity Evaluation. Natural and synthetic metabolites were
tested for phytotoxic activity using the leaf-spot assay (15) on marigold
leaves (0.1 mg/application).
Phytotoxic metabolites (necrotic area in mm2): 1 (0.62), 5 (0.15), 6
Hydrolysis and Methylation of 5-(3′,3′-Dimethylallyloxy)-7-
methoxy-6-methylphthalide. A solution of 3 (21 mg, 0.08 mmol) in
MeOH (10 mL) was heated under reflux with an excess of K2CO3 for
24 h. The reaction mixture was concentrated under vacuum, and the
resulting material was combined with dry THF (4 mL) and a suspension
of NaH (208 mg, 6.93 mmol, 80% in mineral oil). The mixture was
stirred at room temperature for 30 min, and then CH3I (200 µL, 3.2
mmol) was added dropwise. Once no more starting material could be
detected by TLC (44 h), the reaction mixture was quenched with a
(0.10), 10 (0.49), 12 (0.16) and 13 (0.13).
RESULTS AND DISCUSSION
Bioassay-guided VLC and preparative HPLC purifications
of the phytotoxic organic crude extract from A. tagetica resulted
in the purification and identification of six metabolites (2-7),
two of them (5, 6) phytotoxic.

1056 J. Agric. Food Chem., Vol. 50, No. 5, 2002
Gamboa-Angulo et al.
The isomeric phthalides 6-(3′,3′-dimethylallyloxy)-4-meth-
oxy-5-methylphthalide (2) and 5-(3′,3′-dimethylallyloxy)-7-
methoxy-6-methylphthalide (3) were identified by comparing
their spectroscopic data with those reported in the literature (11,
12) and by direct comparison with authentic samples. Phthalide
2 has been reported previously from A. porri (12) and A. solani
(16), whereas 3 has been reported only from A. porri (11). Both
isomers were first reported as the main products resulting from
the oxidation of zinniol (1).
proton singlet at δ 3.44 in the 1H NMR, together with a carbon
signal at δ 58.0 in the 13C NMR, confirmed the presence of an
additional methoxyl group in the structure of 5. Whereas the
chemical shift for C-7 (δ 64.5) in the 13C NMR of 5 was very
similar to that of zinniol (δ 64.7), the corresponding C-8 signal
in the spectrum of 5 appeared downfield (δ 66.5) relative to
that of zinniol (δ 56.8), suggesting that the new methoxyl group
was located at the C-8 position in the structure of 5. This was
3
confirmed by the long-range, J interaction through oxygen
observed in the HMBC of 5 between the methyl group protons
at δ 3.44 and the carbon signal at δ 66.5 (C-8). Phytotoxic
metabolite 5 was thus identified as 8-zinniol methyl ether.
The remaining metabolites (4-7) were obtained in pure form
after preparative HPLC purifications of various fractions. Their
1
IR, EIMS, and H NMR spectra showed close similarities to
zinniol (1, 9). These included a single aromatic proton,
confirming the presence of a pentasubstituted aromatic ring, two
methyl groups, one attached to the aromatic ring and the other
as part of an aryl methyl ether, and signals corresponding to
the dimethylallyloxy side chain. The presence of the charac-
teristic base peak fragment at m/z 180 in the MS spectra of
metabolites 4-7, and their lower polarity on TLC when
compared to 1, suggested that these derivatives had one or both
of the benzylic alcohols substituted. The nature and location of
the substituents on each metabolite were established following
careful analysis of their MS, 1H NMR (Table 1), and 13C NMR
(Table 2) data, in addtition to the results from HMQC and
HMBC experiments.
The IR spectra of metabolites 6 and 7 were identical, both
presenting an intense absorption at 1739 cm-1 characteristic for
an ester carbonyl group. As expected, their EIMS spectra also
showed the same parent ion peak at m/z 308, where the
difference of 42 units from zinniol suggested an acetylated
structure. This was supported by a fragment at m/z 240 (M+ -
C5H8), originating from the loss of the dimethylallyloxy chain
side, which after losing the elements of acetic acid, produced
the characteristic base peak fragment at m/z 180. Further support
for an acetylated zinniol structure came from the three proton
methyl singlets at δ 2.16 in the 1H NMR spectra of 6 and 7 and
carbon signals at δ 171.1 and 170.9 in the 13C NMR spectra.
This suggested that metabolites 6 and 7 were isomeric monoacety-
lated derivatives of 1. Although the HMQC experiment of
metabolite 6 showed a direct correlation of the methylene
protons at δ 5.25 and 4.70 with the carbon signals at δ 58.7
and 63.1, respectively, the HMQC experiment of 7 showed a
direct correlation of the protons at δ 5.20 and 4.73 with the
carbon signals at δ 64.6 and 56.8. A comparison of the chemical
shift for both benzylic methylenes in the 13C NMR spectra of
6 and 7 with those previously described for 1, 4, and 5 (Table
2) made it possible to identify 6 as 8-zinniol acetate and 7 as
the isomeric 7-zinniol acetate. This assignment was further
supported by the long-range interactions observed in the HMBC
experiment of 6 between the methylene protons at δ 4.70 and
the aromatic carbon at δ 108.0 (C-6) and between the methylene
protons at δ 5.25 and carbon signals at δ 118.5 (C-2) and 158.7
(C-3) and the carbonyl carbon at δ 171.1 (carbonyl acetate).
To rule out the possibility of 6 and 7 being formed during the
isolation process, zinniol (1) was subjected to the same processes
involved in working up the culture filtrate of A. tagetica,
resulting only in the recovery of starting material. Additionally,
treatment of 1 with acetic anhydride, at room temperature in
the absence of pyridine, produced a mixture of both monoacety-
lated and diacetylated derivatives together with the starting
material. The monoacetylated derivatives were identical on TLC
to the natural derivatives 6 and 7. It was noted that only the
monoacetate 6 induced damage in marigold leaves when tested
in the leaf-spot assay.
The EIMS spectrum of 4 showed a molecular ion peak at
m/z 370 which corresponded to a difference of 104 units with
1
respect to that of zinniol. Additionally, the H NMR spectrum
of 4 showed five additional protons in the aromatic region,
which appeared as multiple signal (centered at δ 7.23) typical
of a monosubstituted aromatic ring; the spectrum also showed
two doublets of doublets at δ 3.80 and 2.90 assigned to the
protons of two methylene groups, respectively, one of which
(δ 3.80) showed the typical chemical shift for a methylene-
bearing oxygen. The presence of a second aromatic ring in the
structure of 4 was confirmed by 12 sp2 carbons signals in its
13C NMR spectrum (6 methine and 6 quaternary, Table 1),
whereas the two new methylene carbon signals appeared at δ
71.4 and 36.4. These data were constant with the presence of
an oxygenated 2-(phenyl)ethyl fragment in the structure of 4,
and this fragment can readily explain the 104 amu molecular
weight difference from zinniol.
To establish the correct location of the 2-(phenyl)ethyl
fragment in structure 4, it was necessary to first establish the
correct assignment of the various methylene groups in its
structure. The HMQC experiment of 4 showed a direct cor-
relation between the methylene protons at δ 4.63, 4.55, 3.80,
and 2.90 and the carbon signals at δ 64.8, 64.4, 71.4, and 36.4,
respectively. A long-range interaction observed in the HMBC
experiment of 4, between the aromatic proton at δ 6.71 and the
carbon signal at δ 64.4, allowed the latter to be assigned to
C-7 and the carbon signal at δ 64.8 to C-8. Similarly, long-
range interactions between the methylene protons at δ 4.63 (H-
8) and the carbon signals at δ 140.3 (C-1), 121.1 (C-2), and
158.3 (C-3), and, through oxygen, with δ 71.4 (C-1′′), indicated
that the phenylethyl residue was located on C-8 via an ether
linkage. This was supported by interactions observed between
the methylene protons at δ 3.80 and the carbon signals at δ
36.4 (C-2′′) and 138.6 (C-3′′), and, through oxygen, with δ 64.8
(C-8). On the basis of these data, the new metabolite 4 was
identified as 8-zinniol 2-(phenyl)ethyl ether.
To date, a number of metabolites structurally related to zinniol
have been reported. These include structures where the benzylic
alcohols have been modified to form lactone (17-19) or lactam
rings (5, 20) or to produce dimeric structures (9). These have
no phytotoxicity to the host. However, this is the first report of
natural zinniol derivatives having most of the original zinniol
structure intact.
To confirm its proposed structure, metabolite 5 was prepared
in the laboratory using zinniol as the starting material. Jones
oxidation of 1 afforded the expected mixture of isomeric
phthalides 2 and 3 together with a third product identified as
zinniol anhydride (8, Scheme 1) (1, 8). The identity of 8 was
established by the intense bands observed at 1838 and 1772
Phytotoxic metabolite 5 was obtained as single component
on TLC. A parent ion peak at m/z 280 in its EIMS suggested a
zinniol structure with an additional methylene moiety. A three-

Phytotoxins from Alternaria
J. Agric. Food Chem., Vol. 50, No. 5, 2002 1057
Scheme 1. Oxidation of Zinniol
Scheme 2. Synthesis of 8-Zinniol Methyl Ether (5)
Scheme 3. Synthesis of 7-Zinniol Methyl Ether (13)
cm-1 in its IR spectrum and by the presence of two quaternary
carbon signals at δ 164.5 and 163.5 in its 13C NMR spectum,
in addition to the absence of both proton and carbon signals
characteristic of the benzylic methylene groups. The anhydride
8 has been reported as an intermediate during the synthesis of
phthalides 2 and 3 (8). However, this is the first report of 8
being obtained as an oxidation of zinniol product.
Treatment of phthalide 2 under alkaline conditions led to the
formation of a more polar product, the expected hydroxy acid
(2a) (13). However, after workup of the reaction mixture, only
starting material could be recovered, suggesting a relactonization
of the hydroxy acid. To prevent this reversible reaction, once
hydrolysis was shown to be complete by TLC, the solvent was
removed under vacuum and the residue methylated. The reaction
product contained a mixture of the ester 9 and its acid 10
(Scheme 2) (14). Similar treatment of 3 (Scheme 3) yielded
the corresponding ester 11 and acid 12. In this case, the
hydrolysis product 3a proved to be more stable than 2a, because
it could be recovered and analyzed by TLC and IR, showing
the characteristic absorptions at 1715 and 1760 cm-1 due to
the hydroxy acid and cyclization product mixture.
3 nor zinniol anhydride (8) showed significant activity. As
expected, lower phytotoxic activity was detected with the
monosubstituted derivatives 5 and 13, thus confirming previous
reports that both benzylic alcohols in zinniol are essential for
the expression of phytotoxicity (3). However, the latter statement
was seriously challenged when the monosubstituted derivatives
4 and 7 did not damage marigold leaves, and both carboxylated
methyl ether derivatives 10 and 12 showed significant phytotoxic
activity. This activity was readily lost upon esterification to
produce 9 and 11. These results indicate that there might be a
number of factors controlling the expression of phytotoxic
activity in these molecules, including the type, size, and position
of the various functional groups.
In summary, we have reported the isolation and identification
of six metabolites from the culture filtrates of A. tagetica. All
metabolites were structurally related to zinniol, and four of them
were identified as new natural products (4-7), whereas five
new compounds (9-13) were synthetically prepared. Five of
the products tested, both natural and synthetic derivatives (5,
6, 10, 12, and 13), showed phytotoxicity on marigold leaves.
All derivatives were less phytotoxic than zinniol (1) with the
exception of the acid derivative 10, which showed a similar
activity. The synthetic preparation of 8-zinniol methyl ether (5)
and its isomer 7-zinniol methyl ether (13), using zinniol as
starting material, confirmed the structure for the new natural
product 5.
Reduction of 9 using LiAlH4 produced a zinniol methyl ether,
which was identical by TLC and HPLC to the natural metabolite
5 (Scheme 2). Similarly, reduction of 11 afforded 7-zinniol
1
methyl ether (13) as the main component (Scheme 3). H and
13C NMR data for the isomeric methyl ether 13 were very
similar to those of 5; significant diferences were observed only
in their 13C NMR spectra, where the chemical shift values of
C-1, C-2, C-7, and C-8 in 13 are δ 135.3, 126.1, 74.4, and 56.9,
respectively, and δ 140.3, 121.2, 64.5, and 66.5 in 5.
ACKNOWLEDGMENT
We thank Isabel Cha´vez, Hector R´ıos, Beatriz Quiroz, Roc´ıo
Patin˜o, Luis Velasco-Ibarra, and Javier Pe´rez-Flores (Instituto
de Qu´ımica, UNAM) for recording the NMR, IR, and MS
spectra and Mariel Avila-Mart´ınez (CICY), Mar´ıa Yolanda R´ıos,
The phytotoxic activity of all products (1-13), both natural
and synthetic, was evaluated in the leaf-spot assay. Although
zinniol (1) caused the strongest leaf necrotic lesions, neither its
diacetylated derivative (1a) nor the isomeric phthalides 2 and

1058 J. Agric. Food Chem., Vol. 50, No. 5, 2002
Gamboa-Angulo et al.
methylphthalide from Alternaria porri. Biosci., Biotechnol.,
Biochem. 1992, 56, 986.
Moustapha Bah, and Rogelio Pereda-Miranda (UNAM) for
technical assistance.
(13) Rios, M. Y. Aproximaciones sinte´ticas y correlaciones qu´ımicas
de cicloaductos dime´ricos derivados del Z-ligusltilido. Doctoral
Thesis, Universidad Nacional Auto´noma de Me´xico, 1997.
(14) Van Maarschalkerwaart, A. H.; Willard, N. P.; Pandit, U. K.
Synthesis of carbohydrate obtaining crown ethers and their
application as catalysis in asymmetric Michel additions. Tetra-
hedron 1992, 48, 8825-8840.
(15) Pen˜a-Rodr´ıguez, L. M.; Armingeon, N. A.; Chilton, W. S.
Phytotoxins from a Bipolaris pathogen of Johnson grass. J. Nat.
Prod. 1988, 51, 821-828.
LITERATURE CITED
(1) Starratt, A. N. Zinniol: a major metabolite of Alternaria zinniae.
Can. J. Chem. 1968, 46, 767-770.
(2) Sugawara, F.; Strobel, G. Zinniol, a phytotoxin, is produced by
Phoma macdonaldii. Plant Sci. 1986, 43, 19-23.
(3) Barasch, I.; Mor, H.; Netzer, D.; Kashman, Y. Production of
zinniol by Alternaria dauci and its phytotoxic effect on carrot.
Physiol. Plant Pathol. 1981, 19, 7-16.
(4) Cotty, P. J.; Mishagi, I. Zinniol production by Alternaria species.
Phytopathology 1984, 74, 785-788.
(5) Stierle, A.; Hershenhorn, J.; Strobel, G. Zinniol-related phyto-
toxins from Alternaria cichorii. Phytochemistry 1993, 32, 1145-
1149.
(6) Cotty, P. J.; Mishagi, I.; Hine, R. Production of zinniol by
Alternaria tagetica and its phytotoxic effect on Tagetes erecta.
Phytopathology 1983, 73, 1326-1328.
(7) Martin, J. A.; Vogel, E. The synthesis of zinniol. Tetrahedron
1980, 36, 791-794.
(8) Hariprakasha, H. K.; Subba Rao, G. S. R. Synthesis based on
cyclohexadienes: Part 6sTotal synthesis of some naturally
ocurring phthalides from Alternaria species. Indian J. Chem.
1998, 37B, 851-856.
(9) Gamboa-Angulo, M. M.; Alejos-Gonza´lez, F.; Escalante-Erosa,
F.; Garc´ıa-Sosa, K.; Delgado-Lamas, G.; Pen˜a-Rodr´ıguez, L. M.
Novel dimeric metabolites from Alternaria tagetica. J. Nat. Prod.
2000, 63, 1117-1120.
(10) Gamboa-Angulo, M. M.; Garc´ıa-Sosa, K.; Alejos-Gonza´lez, F.;
Escalante-Erosa, F.; Delgado-Lamas, G.; Pen˜a-Rodr´ıguez, L. M.
Tagetolone and tagetonolone: two phytotoxic polyketides from
Alternaria tagetica. J. Agric. Food Chem. 2001, 49, 1228-1232.
(11) Suemitsu, R.; Ohnishi, K.; Morikawa, Y.; Nagamoto, S. Zinni-
midinne and 5-(3′,3′-dimethylallyloxy)-7-methoxy-6-methylphtha-
lide from Alternaria porri. Phytochemistry 1995, 38, 495-497.
(12) Suemitsu, R.; Ohnishi, K.; Horiuchi, M.; Morikawa Y. Isolation
and identification of 6 (3′,3′-dimethylallyloxy)-4-methoxy-5-
(16) Gamboa-Angulo, M. M.; Alejos-Gonza´lez, F.; Pen˜a-Rodr´ıguez,
L. M. Homozinniol, a new phytotoxic metabolite from Alternaria
solani. J. Agric. Food Chem. 1997, 45, 282-285.
(17) Fujita, M.; Yamada, M.; Nakajima, S.; Kawai, K.; Nagai, M.
O-Methylation effect on the carbon-13 nuclear magnetic reso-
nance signals of ortho-disubstituted phenols and its application
to structure determination of new phthalides from Aspergillus
silVaticus. Chem. Pharm. Bull. 1984, 32, 2622-2627.
(18) Suemitsu, R.; Ohnishi, K.; Morikawa, Y.; Ideguchi, I.; Uno, H.
Porritoxinol, a phytotoxin of Alternaria porri. Phytochemistry
1994, 35, 603-605.
(19) Ichihara, A.; Tazaki, H.; Sakamura, S. The structure of zinnolide,
a new phytotoxin from Alternaria solani. Agric. Biol. Chem.
1985, 49, 2811-2812.
(20) Suemitsu, R.; Ohnishi, K.; Horiuchi, M.; Morikawa, Y.; Sakaki,
Y.; Matsumoto, Y. Structure of porriolide, a new metabolite from
Alternaria porri. Biosci., Biotechnol., Biochem. 1993, 57, 334-
335.
Received for review May 16, 2001. Revised manuscript received October
31, 2001. Accepted November 2, 2001. Financial support for the project,
through grants from PADEP-UNAM (Nos. 5356 and 5370) and
CONACyT (4871-E9406), is greatly appreciated. M.M.G.A. thanks
CONACyT for a graduate student fellowship.
JF010641T