Synthesis of melampolides and cis,cis-germacranolides as natural herbicide models

Article abstract of DOI:10.1016/j.tet.2004.06.128

The comparative study between the theoretical molecular properties of the starting materials and the yields in the transformation of melampolides to cis,cis-germacranolides using SeO₂/t-BuOOH (TBHP) as oxidant allows to establish a feasible relationship with their values of dipolar moment. Conditions for this transformation are optimized and some mechanistic considerations are made based in this finding. Cluster analysis of the phytotoxic activity of the melampolides and cis,cis-germacranolides obtained shows that the activity is greatly influenced by the spatial shape of the backbone, prevailing upon other factors such as the presence of reactive functional groups. Graphical abstract.

Full text of DOI:10.1016/j.tet.2004.06.128

Tetrahedron 60 (2004) 8477–8488
Synthesis of melampolides and cis,cis-germacranolides as natural
herbicide models
*
a,*
a
Francisco A. Mac ´ı as, Ra u´ l F. Velasco, Jos e´ A. Alvarez, Diego Castellano
b
a
´
a
and Juan C. G. Galindo
a
Grupo de Alelopat ´ı a, Departamento de Qu ´ı mica Org a´ nica, Facultad de Ciencias, Universidad de C a´ diz, C/ Rep u´ blica Saharaui s/n,
Apdo. 40, 11510 Puerto Real (C a´ diz), Spain
b
Departamento de Qu ´ı mica F ´ı sica, Facultad de Ciencias, Universidad de C a´ diz, C/ Rep u´ blica Saharaui s/n, Apdo. 40,
1510 Puerto Real (C a´ diz), Spain
1
Received 18 May 2004; revised 25 June 2004; accepted 30 June 2004
Available online 7 August 2004
Abstract—The comparative study between the theoretical molecular properties of the starting materials and the yields in the transformation
of melampolides to cis,cis-germacranolides using SeO /t-BuOOH (TBHP) as oxidant allows to establish a feasible relationship with their
2
values of dipolar moment. Conditions for this transformation are optimized and some mechanistic considerations are made based in this
finding. Cluster analysis of the phytotoxic activity of the melampolides and cis,cis-germacranolides obtained shows that the activity is greatly
influenced by the spatial shape of the backbone, prevailing upon other factors such as the presence of reactive functional groups.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
metal-catalyzed oxidation methods with alkyl-hydroperox-
ides as stoichiometric oxidant have been developed and are
Selenium dioxide is a soft and widely used reagent useful to
get access to allylic alcohols and unsatured carbonyl
systems from alkenes, among others. The mechanism is
thought to proceed through a first ‘ene’ reaction rate-
determining step, followed by a [2,3] sigmatropic rearrange-
commonly used. Among them, the system SeO /tert-butyl-
2
hydroperoxide (TBHP) has found extensive application in
natural products synthesis as a soft and stereoselective
reactive.
3
,4
1
Currently, we are interested in developing new models of
agrochemicals based on phytotoxic allelochemicals. The
ment, and breakdown of the resultant selenium ester. This
5
subject has been recently revisited, so as kinetic isotope
effects and theoretical calculations strongly supports the
idea of an ene concerted reaction with SeO itself as active
increasing incidence of resistant weeds to important
herbicides classes such as s-triazines and dinitroanilines
6
7
2
in addition to environmental concerns are pulling-on
research on this field. As an alternative, Allelopathy is
able to offer new lead compounds to be used as herbicides
and several classes of compounds such as phenolics,
oxidant rather than an ene-concerted process with a selenous
C
ester, or even an electrophilic attack by HSeO . However,
2
the isolation of some minor products suggests the presence
of a minor pathway involving a stepwise ene reaction
through reversible formation of a zwitterion followed by
8
alkaloids, and terpenes have been reported as phytotoxic.
2
Among them, germacranolides (Fig. 1) constitute a very
interesting group because of the high number of compounds
reported and their wide spectrum of biological activities that
rate-limiting proton transfer. In spite of its usefulness,
some environmental and safety concerns have arisen
because wastes contain high amounts of the metal. Thus,
9
10
11
includes pharmacological, cytotoxic, fungicidal, anti-
1
2
13
bacterial, and allelopathic activities. Consequently, an
important effort to synthesize the germacranolide skeleton
has been made. These efforts have been centred, mainly,
*
Part 15 in the Series Natural Products as Allelochemicals. For Part 14 see:
Mac ´ı as, F. A.; Chinchilla, D.; Molinillo, J. M. G.; Mar ´ı n, D.; Varela, R. M.;
14
Torres, A. Tetrahedron 2003, 59, 1679–1683.
in trans,trans-germacranolides, but little has been done
about the chemistry and biological activity in the other
subgroups.
Keywords: Allylic oxidation mechanism; Synthesis of sesquiterpenes;
Melampolides; cis,cis-Germacranolides; Structure–activity relationship
study (SAR).
*
Corresponding author. Tel.: C34-956-016-370; fax: C34-956-016-193;
e-mail: famacias@uca.es
Going on with our previous work in sesquiterpene lactones
(SL) as potential lead herbicides
4
,5,15
here in we report the
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.06.128

8478
F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
Figure 1. Main germacranolides backbones.
diasterospecific synthesis of several SL with melampolide
and cis,cis-germacranolide skeletons using modified reac-
skeletons, even though the yields obtained for cis,cis-
germacranolides were low (around 10%).
5
a
1
7
tion conditions in the SeO –TBHP system suitable for use
2
in a Structure–Activity Relationship (SAR) study of their
phytotoxicity.
2
.1. Synthesis and molecular modeling
In order to overcome these low yields and aiming also to
perform a SAR study melampolides 2–8 were used as starting
materials to obtain the corresponding cis,cis-germacranolides
2
. Results and discussion
(Table 1) with different yields. Conditions for optimal yields
were set to as SeO /TBHP (0.5:2 molar ratio) in a reflux of
2
Metal-catalyzed oxidations with alkyl hydroperoxides can
be divided into two categories depending on if the
intermediate involved in the oxygen-transfer step is a
peroxometal or an oxometal. In other words, the initial
oxidizing agent could be the selenium dioxide itself, or the
selenium-TBHP adduct. It is generally believed that this
reaction proceeds through an initial ene reaction of SeO2
with the olefin, followed by a [2,3] sigmatropic rearrange-
ment, the resulting Se(II) being reoxidized by TBHP, and
additional data supporting this vision has been published
DCM using compound 4. These conditions allow to minimize
the amount of the 14-oxo-derivative. Best results were
obtained with the 14-acetoxy- and 14-chloro-derivatives 4
and5respectively,followedbytheoxo-melampolide3andthe
hydroxy-melampolide 2. The 14-TBDMS derivative 6 gave
the 14-oxoderivative 3 (53% yield) due to deprotection of the
silyl group caused by the acid medium generated by the
selenium salts and subsequent oxidation of the free hydroxyl.
No change in the stereochemistry of the D(4–5) double bond
was observed. Finally, lower or no yields were obtained with
the epoxy-derivatives 7 and 8 (Table 1).
1
6
(
is needed (typically, 2 mol%).
Fig. 2). In this case, only a catalytic amount of the metal
7
1
Theoretical DH 8 values as obtained using PM3 calculations
r
(
1
Table 2) were in all cases about K42 Kcal/mol. This
8
When this reaction is carried out using the natural
germacranolide costunolide (1) as substrate the amount of
SeO2 needed increases and the stereochemistry of the
double bond changes in the final product from E to Z
result supports the idea that the different reactivity observed
for each substrate should not be related with the difference
of energy between starting and final products. Conse-
quently, factors affecting the transition states should arise as
determining to explain the experimental data.
(
compound 2). In this case, a different mechanism with a
bulky peroxometal intermediate t-BuOO–SeO H has been
3
1
7
proposed (Fig. 3).
1
7
According to the mechanism proposed by Haruna and Ito
three transition states will consecutively take place during
This mechanism should certainly explain why the methyl
group at C-10 reacts before the methylene position at C-9 in
the germacranolide skeleton, instead of the accepted order
the reaction (Fig. 4): (a) a cyclic intermediate corresponding
to the ene reaction step E ; (b) the eclipsed conformation of
#
1
1
6
CH OCH OCH. The change in the stereochemistry of
the selenium substituent and the double bond during the
rotation step E ; and (c) that corresponding to the
2
3
#
the double bond takes place through a rotation before the
sigmatropic rearrangement occurs.
2
#
sigmatropic rearrangement E . The conformational flexi-
3
bility of the backbone in the different substrates and end
products are mostly the same and are not influenced by
substituents at the C-14 position. Otherwise, the reacting
part of the melampolide (e.g. the double bond at C-4, C-5
and the C-15 methyl group) is the same for all starting
compounds (2–8). Consequently, the activation energies
corresponding to the rotation and rearrangement steps have
to be similar in all cases and will not determine the different
yields observed. Thus, we propose as a feasible working
The reaction also yields as side-products the corresponding
14-oxoderivatives due to oxidation of the resulting hydroxyl
group. This reaction has demonstrated to be useful in getting
access to the melampolide and cis,cis-germacranolide
hypothesis that those factors affecting the energy of the first
transition state (E , ene reaction) would be crucial to explain
#
1
the different results obtained and will also be in good
agreement with the literature.
1
6
2
Figure 2. General mechanism of the SeO allylic oxidation.
A correlation between theoretical dipolar moments obtained

F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
8479
17
2
Figure 3. Mechanism proposed for the SeO -TBHP allylic oxidation.
from PM3 calculations and the experimental yields can be
observed(Table 1, Fig. 5): the lowerthe dipolarmomentis, the
higher the resulting yield. This correlation would be in good
agreement with the hypothesis presented: the ene reaction is a
concerted process where no new charges are generated; lower
dipolar moments of starting material lead to lower dipolar
moments in the transition state, and thus, transition states with
lower polarities will be better dissolved in low polar solvents
such as those used for these oxidations (DCM: mZ1.60 D;
THF: mZ1.63 D). Accordingly, the solvation enthalpy
contribution will be diminished and, consequently, the
activation energy will also be lowered, thus favoring the
reaction pathway. Compounds 4 and 5 present the lower
dipolar moments and the higher yields while the free alcohol 2
and the 14-oxoderivative 3 render lower amounts of the
cis,cis-germacranolide. Finally, the epoxiderivative 8, with
the highest value of dipolar moment, does not react at all
Compound 7 constitutes an exception to this correlation
(Fig. 5), as the theoretical value of the dipolar moment does
not match with the reactivity. However, experimental Rf
values obtained by TLC show that real polarity of this
compound should correspond to a value similar to those of
compounds 2 and 8, thus matching the correlation proposed.
The starting melampolide 2 was obtained from costunolide
(1) using SeO /TBHP, along with the aldehyde 3 as side
2
5
a,17
product.
The acetylated derivative 4 was used as starting
material for cis,cis-germacranolide large scale synthesis
since it provided the best yields. Selective epoxidation using
MCPBA at the C1–C10 (18 and 19) or the C4–C5 (15)
double bonds in the cis,cis-germacranolide backbone was
achieved through the introduction of a bulky steric
demanding tert-butyl-dimethyl-silyl (TBDMS) group at
C15 (16). However, steric hindrance induced by the C14-
acetoxyl group in 16 did not allow direct epoxidation at C1–
C10. So, selective deprotection of the acetoxy derivative at
(Table 1).
1
9
C14 was carried out using magnesium methoxide,
yielding the corresponding de-acetylated derivative (17)
and traces of the enantiomerically pure Michael adduct at
the unsatured lactone ring (17b). The stereochemistry of
compound 17b was unequivocally assigned as 11S through
NOESY experiments. Epoxidation of the resulting free-
hydroxyl compound 17 yielded both alpha (18) and beta
a
2
Table 1. Melampolide to cis,cis-germacranolides conversion using SeO /TBHP as oxidant system
R
Starting material
Final product
Yield (%)
Dipolar moment (m, D)
CH OH
CHO
2
2
3
4
5
6
10
11
12
13
3
10
24
57
43
53
4.34
4.45
3.78
4.04
4.70
CH
CH
CH
2
2
2
OAc
Cl
OTBDMS
b
1b,10b-Epoxy-derivatives
CH
CH
2
OH
OAc
7
8
14
—
3
—
3.87
4.49
2
a
SeO :TBHP molar relation (0.5:2), DCM, reflux.
2
The reaction product was the carbaldehyde melampolide, with no change in the stereochemistry.
b

8480
F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
18
Table 2. Theoretical DH
f
8 and DH
r
8 (Kcal/mol) for some germacranolides as obtained from PM3 calculations
Starting material
DH
f
8 (Kcal/mol)
End product
DH
f
8 (Kcal/mol)
DH
r
8 (Kcal/mol)
2
3
4
5
7
K94.8
K82.3
K136.6
K61.2
K106.8
10
11
12
13
14
K137.1
K124.4
K177.1
K101.2
K153.4
K42.3
K42.1
K40.5
K40.0
K46.6
(
19) epoxide isomers. Finally, epoxidation of the double
bond at C4–C5 only afforded one isomer (15) due to
hindrance of the alpha face of the molecule by the lactone
ring (Fig. 6). Epoxidation in the same position of the
melampolide framework is not possible due to steric
restrictions caused by the spatial disposition of the molecule.
2
.2. Bioassays and SAR studies
The bioactive properties of sesquiterpene lactones have
been attributed for a long time to the electrophilic properties
of the a-methylene-g-lactone moiety because it easily reacts
with nucleophilic groups contained in important biomole-
Figure 5. Correlation between dipolar moments of starting materials and
the yields in the SeO /TBHP allylic oxidation.
2
2
0
cules. A long discussion whether nucleophilic additions at
the double bond of the lactone ring is the main factor
causing bioactivity on sesquiterpene lactones is still on the
literature. However, this feature might not be the only
responsible of their bioactivity and we have recently
reported positive bioactive sesquiterpene lactones without
framework; and (iii) importance of halogen-containing
compounds, very common in pesticide and insecticide
2
1
synthetic designs. Consequently, compounds depicted in
Figure 6 were synthesized according to the methodology
described above and tested. Compounds 1 and 22–26 were
4
previously tested and have been introduced here by means
of comparison and cluster analysis.
4
,15a
this functional group.
The changes introduced in the
basic melampolide and cis,cis-germacranolide frameworks
attend to analyze the influence in the activity that the
following criteria might have: (i) influence of reactive
groups other than the a-methylene-g-lactone group; (ii)
influence of the spatial arrangement of the basic carbon
Bioassays were carried out using the monocots Triticum
aestivum L. and Allium cepa L., and the dicots Lactuca
sativa L. and Lepidium sativum L. as target species,
2
2
according to the methodology reported by our group and
Figure 4. Activated states according to the proposed reaction mechanism.

F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
Table 3. H NMR data for listed compounds (400 MHz in CDCl , signal of the residual CHCl centered at d 7.25 ppm)
8481
1
a
3
3
H
1
3
4
5
6
8
9
10
11
12
12b
5.45
2.40 dddd
2.60
2.51 ddd
2.83 ddd
6.32 brd
5.35
2.87
2.07 m
1.82
2.24 ddd
2.64
5.69
6.37
4.58
6.48 dd
b
2.40 m
5.48
2.12
1.95
5.59
2.15
1.99
5.42
2.11
1.94
2.90
2.32 dddd
1.19
2.24 ddd
2.16
5.24 d
4.59
2.81
2.16 m
1.62
2.06
4.48 m
5.49 dd
b
2.20
6.56
2.76
2.54
5.54
2.27
2.55
2.24
2.67
5.49
5.18
2.72
2.07
1.73
2.61
2.22
5.39
6.11
4.38
4.55
3.98
b
b
b
b
2.02 m
1.56
b
b
b
b
c
c
2
2
3
3
5
6
7
8
8
9
9
1
1
1
1
1
a
b
2.26 m
b
b
b
b
b
b
a
b
2.52
c
2.24 ddd
c
2.32 m
2.04 ddd
2.28 m
4.97 brd
4.52 dd
2.26 ddddd 2.57
1.87 m
2.16
5.00
1.85 ddd
1.58
5.00
1.84 m
2.15
5.04
2.46 m
2.62 m
5.50 d
5.07
c
c
2.66 m
2.18 ddd
5.38 brd
4.62
5.48
5.19
4.54
4.58
2.43
4.58
2.59
2.73
2.28
1.67
1.92
1.33
5.42
2.71 m
2.08 dddd
1.72
2.44
2.42
1.67
2.80
a
b
a
b
3a
3b
4
4
5
2.86 dddd
1.44 dddd
2.26
1.54
2.17
2.07
5.39
6.11
4.38
2.36 m
1.57
2.33 dddd
1.50
c
2.38 m
c
c
c
c
d
d
2.03
1.97
2.18
2.01
2.60
2.30
c
1.95 m
c
c
d
d
1.37 ddd
5.41
6.18
2.49 m
5.58
6.25
5.41 d
6.08 d
9.39 d
5.46
6.18
4.13
3.99
1.84
5.40
6.16
4.14
4.00
1.84
5.58
6.27
4.10 d
4.05
3.99
6.17
9.83 s
4.43
3.91
9.42
0
4.55 d
1.83
4.43
9.38
1.83 d
1.85 s
1.91
4.04
H
1
13
5.61 dd
14
3.10 m
15
16
17
5.48
2.17
2.50
2.18
2.54
5.44 brd
5.18
2.68
2.07 dddd
1.71 dddd
2.58 ddd
2.31
5.56
6.26
4.10
4.04
3.96
17 b
5.48
2.12
2.46
2.17
2.56
5.51 d
5.07
2.20
2.09
1.63
2.25
2.52
3.63
3.57
4.05
4.00
3.98
18
19
20
5.73 dd
2.19
2.42
2.28 ddd
b
1.88
3.01 d
4.28
2.86
2.00 dddd
1.80 dddd
2.47 ddd
2.24
5.54
2.17
2.52
2.20 m
b
2.61
5.45 dd
5.17
2.70
2.05 m
1.72
2.58 m
2.25 m
5.58
6.27
4.59
3.17
1.17
2.43
2.15
2.20
5.42
5.04
2.76
1.97
1.76
1.44
2.38
5.64
6.31
3.90
3.70
4.07
3.09
2.14
2.61
2.03
2.03
5.49
5.19
2.82
2.03
1.74
2.14
2.61
5.58
6.27
3.81
3.55
4.00
6.47 ddd
2.43 dddd
2.26 dddd
2.06 dd
b
2.21 m
b
b
b
2
2
3
3
5
6
7
8
8
9
9
1
1
1
1
1
a
2.03
1.76
2.03
1.76
b
2.54 m
2.26 m
2.67 m
5.49 d
5.16 dd
2.66 m
2.10 dddd
1.74 dddd
2.63 ddd
2.44 m
5.60 d
b
b
b
b
a
b
c
b
b
c
b
c
c
b
5.52 dd
5.20
2.83
2.03 m
1.76 m
2.65 m
2.21 ddd
5.59
6.28
3.80
3.59
4.03
4.93
4.58
1.67
1.99 ddd
1.42
2.86 dddd
2.33
2.71 dd
2.81 dd
9.43
a
b
a
b
3a
3b
4
4
5
5.68
6.35
4.62
4.44
6.39 d
4.09 d
4.02 d
3.99 s
0
4.43
3.96 s
3.78 d
1.86
4
1
3
: CH
3
–CO, d 2.03, 3H, s; 6: (CH
5 , d 3.55, d, 1H; CH –CO, d 2.03, 3H, s; 16–19: (CH
3
)C–, d 0.89, 9H, s; CH
3
–, d 0.05, 3H, s; CH
3
–, d 0.04, 3H, s; 8: CH
–, d 0.06, 3H, s; CH –, d 0.05, 3H, s; 16: CH
–, d 0.05, 3H, s; 20: CH –CH –N, d 1.74, 4H, brs; H-11, d 2.31, 1H, m; CH
3
–CO, d 2.09, 3H, s; 12, 12b: CH
3
–CO, d 2.03, 3H, s; 15:
0
3
3
)C–, d 0.90, 9H, s; CH
3
3
3
–CO, d 2.07, 3H, s; 17b: OCH
3
–, d
–34, 3H; (CH
3
)C–, d 0.89, 9H, s; CH
3
–, d 0.06, 3H, s; CH
3
2
2
2
–CH –N, d 2.61, 4H,
2
brs.
a
Multiplicities are not repeated if identical with the preceding column.
Signals might be interchanged within the same column.
Signals might be interchanged within the same column.
Signals might be interchanged within the same column.
b
c
d
parameters analyzed were germination and root and shoot
elongation.
for similar sesquiterpene lactones with trans,trans-germa-
cranolide structure, some conclusions can be established:
1
5a
Melampolides show, in general, low levels of phytotoxicity:
only the root length of the dicots lettuce and cress was
inhibited (Table 5). However, all values are below 20%
inhibition. On the other hand, the aldehydes 3 and 9, and the
chlorinated compound 5 show good levels of stimulation of
wheat germination. No significant activities where detected
in onion.
2
.2.1. Influence of the conformation. Cluster analysis (or
complete linkage analysis) is a powerful statistic tool used
to compare and group data in categories attending to their
similar bioactivity. As a result, a hierarchy tree is obtained
23
with those elements with lower differences of bioactivity
among them falling closer or into the same subgroup. In our
particular case, we combined data obtained for germination
and growth of all three skeletal types (e.g. trans,trans-
germacranolides, melampolides, and cis,cis-germacrano-
lides) assayed with the four target species (lettuce, cress,
onion, and wheat). The resulting grouping tree (Fig. 7)
clearly separates previously tested trans,trans-germacrano-
lides (compounds 1, 22–24, 26, Fig. 7, group F ) from
1
melampolides and cis,cis-germacranolides (group F ). The
2
profiles of activity of the elements included in F category
1
are similar among them, and quite different from F . a
2
Regarding cis,cis-germacranolides (Table 5), only the
isomeric epoxides 18 and 19 present significant phytotoxic
activity on lettuce germination. Other groups or substituents
do not seem to increase these effects. The aldehyde 11 and
the chlorinated derivative 13 present good levels of
stimulation of wheat germination. Finally, the epoxide 18
and the acetoxy derivative 12 are the only cis,cis-
germacranolides showing a weak inhibition on root lettuce,
similar to those of melampolides.
1
5a
general decreasing order of activity for germacranolides can
be set to as trans,trans-germacranolidesOmelampolideO
cis,cis-germacranolides.
2
4
According to these results and those previously published

8482
F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
Figure 6. Germacranolides tested in the bioassays.
There is a clear correlation between the lost of activity and the
change in the spatial disposition of the carbon framework—
caused by the different stereochemistry in the two double
bonds of the macrocycle. While trans,trans-germacranolides
present a ‘double-crown’ like conformation (typical theoreti-
3
˚
cal molecular volume VZ313 A ), melampolides adopt a
3
˚
‘
twisted‘ disposition (V(2)Z318 A ), and cis,cis-germacra-
3
˚
nolides has a ‘boat-like’ conformation (V(10)Z335 A ) (Fig.
). Since all compounds present similar volumes, total steric
demands are mostly the same for all of them. Thus, if the
21
8
Figure 8. Minimum energy conformers as obtained by PM3 calculations.
Figure 7. Complete linkage analysis (cluster) of the germacranolides tested.

F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
Table 4. C NMR data for listed compounds (100 MHz in CDCl , signal of the residual CHCl centered at d 77.0 ppm)
8483
13
a
3
3
C
3
4
5
6
8
9
10
11
12
129.2
24.6
26.3
140.1
126.5
77.9
12b
129.5
23.8
24.6
141.0
150.0
77.3
b
b
1
2
3
4
5
6
7
8
9
153.4 d
129.9
24.9
38.1
138.2
125.3
80.6
130.4
24.4
38.1
139.6
125.3
80.3
125.1
24.7
38.5
62.6
27.3
34.3
71.4
29.7
35.5
125.5
152.9
b
24.2 t
37.1 t
b
c
b
b
b
24.3
26.5
25.8
25.6
b
b
d
b
137.4 s
126.1 d
80.7 d
45.5 d
140.4
125.0
80.8
142.4
124.3
79.3
141.3
126.1
79.6
140.1
125.4
78.1
138.4
126.3
77.9
b
45.3
25.1
24.0
45.4
25.1
22.8
45.5
25.2
23.8
46.0
24.1
24.5
60.6
44.1
25.1
17.4
43.8
31.8
43.6
31.6
22.1
43.8
31.7
24.4
42.9
32.0
22.8
b
b
26.1 t
22.0 t
b
b
c
b
b
24.2
139.4 s
b
d
1
1
1
1
1
1
0
144.7 s
139.2 s
170.2 s
119.3 t
195.6 d
16.9 q
135.8
139.8
170.7
118.8
137.6
138.2
170.2
119.2
48.9
138.4
140.2
170.5
118.6
66.7
52.2 d
139.2
171.1
119.3
203.8
16.9
143.6
139.4
170.1
123.3
195.3 d
66.3
134.6
139.5
170.1
122.7
134.3
138.1
170.4
123.9
67.2
d
1
2
3
4
5
139.6
169.8
119.2
67.1
139.4
b
c
c
c
170.1
122.6
66.2
67.5 t
17.1
67.3 t
66.3
17.2
17.2
17.4
66.2 t
194.1 d
C
1
2
3
4
5
6
7
8
9
13
130.1 d
14
60.7
28.7
15
16
17
125.8
24.2
17b
18
62.0
22.4
19
60.9
30.3
20
128.2
21.6
27.1
129.6
24.6
26.1
139.8
124.4
78.1
43.9
31.7
24.6
134.5
139.6
170.8
122.5
67.6
66.3
125.2
24.4
26.0
140.0
125.1
79.7
42.7
31.2
23.9
139.6
48.9 d
177.0
71.3
66.5
66.0
152.5
25.4
37.3
136.1
126.7
80.6
46.0
26.1
22.5
145.4
46.5 d
171.1
53.2
195.3 d
17.0 q
b
24.6 t
26.2 t
b
b
b
b
23.6
26.3
140.1
124.4
78.3
44.1
31.8
26.9
23.5
142.1 s
125.4 d
77.9 d
43.7 d
31.2 t
144.5
123.1
79.2
29.9
28.9
26.8
63.9
139.2 s
169.7
122.3
66.2
125.5
124.3
78.2
44.0
29.7
24.7
63.2
139.4
169.7
122.3
67.2
125.0
77.2
44.1
28.7
24.6
63.2
139.2
169.7
122.6
67.2
61.5
80.6
39.1
32.3
b
b
b
b
b
23.9 t
25.1
24.4
c
c
1
1
1
1
1
1
0
136.2 s
139.5 s
170.3 s
122.8 t
48.8 t
135.5
138.3
170.6
123.8
66.8
139.2
139.8
170.3
122.4
66.3
1
2
3
4
5
b
b
66.2 t
64.3
63.0
66.3
64.3
64.2
a
b
4
: d 20.9 q (CH
3
–CO–); d 170.3 s (CH
3
b–CO–); 6: d K5.2 q (CH
–CO–); 12b: d 20.9 q (CH
–Si); d K5.3 q (CH –Si); d 18.4 s [(CH C–Si)]; d 25.9 q [(CH
O–); 20: d 23.8 t [CH –CH –N]; d 54.7 t [CH –CH –N].
3
–Si); d K5.3 q (CH
–CO–); d 169.5 s (CH
C–Si)]; 16: d 20.9 q (CH
3
–Si); d 25.9 q [(CH
3
)
3
C–Si)]; 8: d 20.7 q (CH
–CO–); 15: d 20.9 q (CH –CO–); d 169.4 s (CH
–CO–); d 170.0 s (CH
3
–CO–); d 170.5 s (CH
3
–
b
a
CO–); 12: d 20.9 q (CH
3
–CO–); d 171.1 s (CH
3
3
3
3
3
–CO–); 16,
3
–CO–); 17b:
a
17, 17b, 18, 19: d K5.3 q (CH
3
3
3
)
3
3
)
3
3
d 59.2 q (CH
3
2
2
2
2
a
Multiplicities are not repeated if identical with the preceding column.
Signals might be interchanged within the same column.
Signals might be interchanged within the same column.
Signals might be interchanged within the same column.
b
c
d
change in the shape of the backbone leads also to a change in
the orientation of the reactive exocyclic double bond in the
lactone ring (Fig. 8), this might be correlated with the decrease
observed in the phytotoxicity.
unsaturated carbonyl system at C-14 (compounds 3 and
11) represents another possible reacting center for Michael
nucleophilic additions. However, comparison of data from
alcohols 2 and 10 and their corresponding aldehydes 3 and
11 show no increment in the phytotoxic activity. This is
indicative that steric hindrance or stereoelectronic effects
2
.2.2. Functional group influence. A second a,b-
Table 5. Bioactivity shown by melampolides and cis,cis-germacranolides in the Petri dish bioassay
Lettuce
Cress
R
Wheat
Onion
G
R
S
G
S
G
R
S
G
R
S
Melampolides
2
3
4
5
7
9
(K)
(K)
(K)
(K)
(K)
(C)
(K)
(K)
(K)
(K)
(K)
(C)
(C)
C
(K)
C
(C)
(C)
(K)
(K)(C)
(C)
(K)(C)
(C)
(K)
(K)
(C)
(C)
(C)
C
(K)
C
(K)
(C)
(C)
(K)
C
(C)
(C)
(C)
cis,cis-Germacranolides
12
13
15
16
17
18
19
(K)
(K)
(C)
(K)
(C)
(C)
(K)
(K)
(C)
(C)
(C)
(K)
—
(K)
(K)
—
C
(C)
(C)
(C)
C
(K)
(K)
K(C)
(C)
(C)
(C)
(C)
(K)
(K)
(K)
(K)
(K)
(K)
(K)(C)
(C)
(K)
(K)
(K)
(K)
(K)
(K)
(K)
—
—
—
(K)(C)
(C)
(C)
—
G: Germination; R: Root Length; S: Shoot Length; blank: No active; (C), (K): stimulatory or inhibitory values below 20% at the maximum concentration
tested (100 mM); C, K: stimulatory or inhibitory values between 20–40%.

8484
F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
dos not allow such a reaction. This hypothesis has been
further tested trying to use pyrrolidine as molecular probe
resembling amino groups of biomolecules. Reaction of
pyrrolidine with compounds 3 and 11 led to the expected
Michael adducts at the lactone ring (compounds 20 and 21,
respectively). However, no reaction products could be
obtained in the unsaturated aldehyde system, thus support-
ing this idea. Note also that the non-conjugated aldehyde 9
remains inactive.
New nucleophilic reactive centers do not necessarily add
additional biological inhibitory effects, unless they are quite
accessible. In this sense, easily accessible oxirane rings
seem to induce higher inhibiting activities (cis,cis-germa-
cranolides 18 and 19).
4. Experimental
4
.1. General
Halogen atoms are commonly present in pesticide formu-
lation, where they have been usually introduced to increase
All reagents and solvents were used as obtained from
commercial suppliers, excepting compound 1. Costunolide
(1) was isolated by column chromatography from Costus
Resin Oil (Pierre Chauvet, S. A.) and purified by
recristallization from hexane/ethyl acetate mixtures. Sol-
vents were distilled from glass prior to use. Column
chromatography was performed on silica gel (35–75
mesh) and TLC analysis was carried out using aluminum-
packed precoated silica gel plates. For HPLC, LiChrosorb
silica 60 was used in the normal-phase mode with a
differential refractometer (RI) in a Hitachi L-6020 HPLC
2
5
the activity. Otherwise, brominated and chlorinated
derivatives are also frequently found in marine natural
products showing high levels of biological activity.
2
6
However, the introduction of halogen atoms or even an
epoxide ring do not add any increase of the phytotoxicity
(
compounds 5, 7, 8, 14, 15, 18, 19), except for the isomeric
epoxides cis,cis-germacranolide 18 and 19 which show an
important increment of the phytotoxic activity, specially on
lettuce germination. The different activity shown among
epoxides at the C1–C10 in melampolides (7) and cis,cis-
germacranolides (18 and 19) could relay on the different
spatial disposition of the backbones, allowing an easier
access in the last ones to the oxirane ring.
1
13
instrument. H and C NMR spectra were recorded using a
Varian UNITY-400 spectrometer at 400 MHz and
100 MHz, respectively, and using CDCl as solvent. The
3
1
resonance of residual chloroform at d 7.25 ppm in the H
H
and d 77.00 ppm for CDCl in the C spectra were used as
C 3
1
3
A deep look into the second group F , allows to differentiate
2
three new subgroups: F22 includes only cis,cis-germacra-
nolides with a TBDMS substituent at C-15, F23 groups both
cis,cis-germacranolides and melampolides possessing a
chlorine or carbaldehyde groups, and hydroxyl or acetoxyl
containing compounds fall into the F21 category. Thus, the
F22 subgroup resembles the sterically demanding TBDMS
cis,cis-germacranolide derivatives.
internal references. Mass spectra were obtained by using a
VG 1250 or a VG AUTOSPEC instruments at 70 eV. IR
spectra were recorded on a Mattson 5020.
Soulangianolide A (2). 100 mg of costunolide (1) were
dissolved in DCM (10 mL) and strongly stirred at room
temperature with selenium dioxide (SeO , 24 mg,
2
0
2
.5 equiv.). Then, t-butyl hydroperoxide (TBHP, 0.1 mL,
equiv.) was drop wise. After 1 h, filtering the reaction
3
. Conclusions
mixture through silica gel and the solvent evaporated under
vacuum stopped the reaction. The crude of reaction was
purified by CC using hexane/ethyl acetate mixtures as
eluant, yielding 2 (62%) as a colorless oil and trace amounts
of the 14-oxo-melampolide 3. All spectral and physical data
of 2 were in full agreement with those reported in the
We report an optimal relation SeO /TBHP to obtain cis,cis-
2
germacranolides from melampolides as 1:4. Such a ratio is
in good accordance with the proposed catalytic nature of
SeO in this reaction. Otherwise, to explain the change in
the stereochemistry of the double bonds it is necessary to
adopt a peroxometal intermediate reactive species in
contrast to recently reported hypothesis, at least for this
2
5
a
literature for 14-hydroxymelampolide.
2
4.1.1. 14-Oxomelampolide (3). nmax (KBr) 1763 (carbonyl
K1
system.
group), 1679 (aldehyde group), 1625 (double bond) cm
HRMS: found [M] 246.12549, C H O requires
;
C
1
5
18
3
C
246.12560; FAB 246.137 [M] ; EIMS, m/z (rel. int.): 231
Also, a feasible relationship between the theoretical dipolar
moment and the yield is observed, which mirrors the
influence of the solvation energy in the first rate-determin-
ing ene reaction step. Compounds having low dipolar
moments are better solvated in low polar solvents such as
DCM or THF, thus favoring better yields.
C
1
(1.1); H NMR data, see Table 3, J (Hz):
[MKCH
]
3
1,2aZ1,2bZ8.4; 3a,3bZ13.1; 3a,2bZ11.7; 5,6Z10.3;
6,7Z9.6; 8a,8bZ14.1; 8a,9bZ12.8; 8a,9aZ6.7; 8a,7Z
3.5; 8b,7Z11.5; 8b,9bZ5.9; 8b,9aZ1.9; 13a,7Z3.1;
1
3
13b,7Z3.5; 14,9Z1.7; 15,5Z1.4; C NMR data, see
Table 4.
Regarding the SAR study, it can be concluded as a general
trend the decreasing order of activity trans,trans-germacra-
nolidesOmelampolidesOcis,cis-germacranolides. Such an
order has been found to positively correlate with the
progressive change in the carbon framework of the decalyne
4.2. Acetylation
All acetylations were carried out by dissolving the
compound in dry pyridine and adding an excess of acetic
anhydride. After 24 h stirring, the reaction mixture was
1
5a
system and it is in good agreement with previous results.
Cluster analysis has demonstrated to be a powerful tool to
analyze profiles of activity and to group those compounds
with similar bioactivity.
washed with a saturated aqueous solution of CuSO to
4
remove the excess of pyridine, yielding the corresponding
acetyl derivatives quantitatively.

F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
8485
4
.2.1. 14-Acetoxymelampolide (4). nmax (KBr) 1768
carbonyl group), 1730 (acetoxy carbonyl group), 1667
4.5. Epoxidations
(
(
K1
double bond) cm ; HRMS: found [MC1] 291.15991,
C
All epoxidations were carried out as follows: to a
solution of the starting compound (0.262 mmol) in
sodium acetate buffered dried THF (8 mL) another
solution of m-CPBA (0.290 mmol) in dried THF (4 ml)
was drop wise while stirring. Reaction was monitored by
TLC until no starting materials could be observed. Then,
the reaction was stopped and the work-up was as
follows: the reaction mixture was extracted with NaOH
(aq) 5% (2!) and the organic phase washed with
distilled water (2!). All the aqueous phases were re-
extracted with ethyl acetate, and the combined organic
phases were dried on anhydrous sodium sulfate. After
separation by CC, (1R,10R)-1b,10b-epoxy-14-hydroxy-
melampolide 7 was obtained from 2 in a crystalline form
(62%). Spectral and physical data were in full agreement
C H O requires 291.15963; EIMS, m/z (rel. int.): 291
1
[
Table 3, J (Hz): 1,2aZ1,2bZ7.6; 5,6Z10.2; 6,7Z9.4;
7 22 4
C
C
MC1] (2.9), 231 [MKAcOH] (100); H NMR data, see
1
8
1
1
a,8bZ14.3; 8a,9bZ13.1; 8a,9aZ6.1; 8a,7Z3.8; 8b,7Z
2.0; 8b,9bZ4.5; 8b,9aZ3.0; 13a,7Z3.4; 13b,7Z3.2;
4,14 Z12.4; 15,5Z1.3; C NMR data, see Table 4.
0
13
4.2.2. (1R,10S)-14-Acetoxy-1,10-epoxymelampolide (8).
nmax (KBr) 1765 (carbonyl group), 1744 (acetoxy carbonyl
K1
C
group), 1678 (double bond) cm ; HRMS: found [M]
06.14681, C H O requires 306.14673; FAB 307.147
3
[
1
7 22 5
C
MC1] ; EIMS, m/z (rel. int.): 247 [MKOAc] (36.0); H
C
1
NMR data, see Table 3, J (Hz): 1,2aZ4.3; 1,2bZ10.1;
2
3
1
9
a,2bZ14.5; 2a,3aZ1.6; 2a,3bZ6.1; 3a,3bZ12.7;
a,2bZ12.7; 5,6Z10.5; 6,7Z9.8; 8a,8bZ17; 8b,7Z
2.4; 8b,9bZ4.7; 8b,9aZ3.1; 9a,9bZ15.1; 9b,8aZ13.8;
1
5a
with those reported in the literature.
Compound 15
was obtained from 12 with a 47% yield, and compounds
18 (42%) and 19 (56%) were obtained from 17.
0
13
b,8bZ4.7; 13a,7Z3.1; 13b,7Z3.5; 14,14 Z12.3;
C
NMR data, see Table 4.
4
.5.1. (4R,5S)-14-Acetoxy-4,5-epoxy-15-hydroxy-cis,cis-
germacranolide (15). nmax (KBr) 3483 (OH), 1762
carbonyl group), 1738 (acetoxy carbonyl group), 1685
4
.3. Chlorination of 14-hydroxymelampolide (2)
40 mg of compound 2 were dissolved in 4 ml of dried
(
K1
C
4
(double bond) cm ; HRMS: found [MC1] 323.15048,
requires 323.14946; EIMS, m/z (rel. int.): 323
[MC1] (3.8), 263 [MKAcO] (46.7), 245 [MKAcO–
pyridine, followed by addition of tosyl chloride (TsCl,
14 mg, 1.2 equiv.) with stirring. After 22 h, the pyridine
was removed under vacuum, and the crude was purified by
C H O
17 20 6
C
C
4
C
1
H2O] (68.3); H NMR data, see Table 3, J (Hz): 1,2aZ
1,2bZ8.1; 3a,3bZ14.1; 3a,2bZ11.3; 3a,2aZ7.4; 5,6Z
CC, yielding 14-chloromelampolide 5 (80%).
9
.5; 6,7Z3.5; 8a,8bZ14.6; 8a,9bZ12.9; 8a,7Z5.0;
4
bonyl group), 1669 (double bond), 666 (C–Cl) cm
.3.1. 14-Chloromelampolide (5). nmax (KBr) 1763 (car-
;
8a,9aZ3.8; 8b,7Z12.9; 8b,9bZ4.9; 8b,9aZ3.5;
9a,9bZ14.8; 9b,8aZ12.8; 13a,7Z2.3; 13b,7Z2.7;
14,14 Z12.7; 15,15 Z12.7; C NMR data, see Table 4.
K1
C
HRMS: found [M] 266.10748:266.10453, C H ClO
0
0
13
1
5
19
2
requires 266.10736:268.10441; FAB 266.097:268.564
[
C C
M] ; EIMS, m/z (rel. int.): 251 [MKCH ] (3.0), 231
4.5.2. (1S,10S)-15-t-Butyldimethylsilyloxy-1,10-epoxy-
14-hydroxy-cis,cis-germacranolide (18). nmax (KBr)
3455 (OH), 1763 (carbonyl group), 1658 (double bond)
3
C
1
[
MKCl] (4); H NMR data, see Table 3, J (Hz): 1,2aZ
1
6
3
1
,2bZ9.3; 3a,3bZ3a,2bZ13.0; 3a,2aZ1.9; 5,6Z10.6;
,7Z9.4; 8a,8bZ14.0; 8a,9bZ13.8; 8a,9aZ5.8; 8a,7Z
.8; 8b,7Z11.9; 8b,9bZ4.7; 8b,9aZ2.6; 13a,7Z3.1;
K1
C
cm ; HRMS: found [MC1] 395.22762, C H O Si
1 34 5
2
C
requires 395.22538; EIMS m/z (rel. int.): 395 [MC1]
0
13
C
1
3b,7Z3.5; 14,14 Z12.3; 15,5Z1.0; C NMR data, see
(8.8), 337 [MKHC(CH ) ] (3.6); H NMR data, see
3 3
Table 4.
Table 3, J (Hz): 1,2aZ11.0; 1,2bZ3.2; 2a,2bZ14.9;
a,3aZ2a,3bZ4.3; 5,6Z10.1; 6,7Z3.6; 8a,8bZ15.4;
2
4
.4. Silylation of 14-hydroxymelampolide (2)
8a,9bZ9.6; 8a,7Z3.7; 8a,9aZ2.1; 9a,9bZ15.1;
0
9
a,8bZ9.2; 13a,7Z2.3; 13b,7Z2.6; 14,14 Z12.4; 14,8Z
1
3
200 mg of 2 were dissolved in N,N-dimethylformamide
N,N-DMF, 4 mL), followed by addition of t-butyldimethyl-
5.1; 15,5Z1.6; 15’,5Z1.9; C NMR data, see Table 4.
(
silyl chloride (TBDMSCl, 273 mg, 2 equiv) and dry
collidine (0.24 mL, 2 equiv.). After 24 h, the reaction was
stopped by addition of water (4 mL) and the reaction
mixture extracted with AcOEt (3!). The combined organic
phases were dried with anhydrous sodium sulfate yielding
4.5.3. (1R,10R)-15-t-Butyldimethylsilyloxy-1,10-epoxy-
14-hydroxy-15-cis,cis-germacranolide (19). nmax (KBr)
3477 (OH), 1764 (carbonyl group), 1660 (double bound)
K1
C
cm ; HRMS: found [MC1] 395.22516, C H O Si
2
1 34 5
C
requires 395.22538; EIMS m/z (rel. int.): 395 [MC1]
C
1
1
4-t-butyldimethylsilyloxy-melampolide 6 (99%).
(8.8), 337 [MKHC(CH ) ] (3.6); H NMR data, see
3 3
Table 3, J (Hz): 2a,3aZ2a,3bZ4.4; 2a,2bZ15.9; 5,6Z
0
13
4.4.1. 14-[tert-Butyl-dimethylsilyloxy]melampolide (6).
nmax/cm : 1744 (carbonyl group), 1670 (double bond);
11.3; 6,7Z4.3; 13a,7Z2.3; 13b,7Z2.5; 14,14 Z12.4;
NMR data, see Table 4.
C
K1
C
HRMS: found [M] 362.22759, C H O Si requires
2
1 34 3
C
62.22772; FAB 361.232 [MK1] ; EIMS, m/z (rel. int.):
05 [MKC(CH ) ] (28.1), 277 [MKC(CH ) Si] (4.9);
3 3 3 3
H NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ9.3; 5,6Z
3
4.5.4. (1R,10R)-1,10-Epoxy-14-hydroxymelampolide (7)
ring opening. 184 mg of compound 7 were dissolved in
10 ml of THF, followed by addition of boron trifluoride
dihydrate (BF $2H O, 0.15 mL, 2 equiv.). After 4 h, the
reaction was stopped by adding water, the reaction mixture
extracted with ethyl acetate (3!), and the combined organic
phases were dried on anhydrous sodium sulfate. The crude
C
C
3
1
1
0.6; 6,7Z9.4; 8a,8bZ14.0; 8a,9bZ13.8; 8a,9aZ5.8;
a,7Z3.8; 8b,7Z11.9; 8b,9bZ4.7; 8b,9aZ2.6; 13a,7Z
3
2
8
3
0
13
.1; 13b,7Z3.5; 14,14 Z12.3; 15,5Z1.0; C NMR data,
see Table 4.

8486
F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
of reaction was purified by CC to afford (1R,10S)-1-
hydroxy-14-oxomelampolide 9 (53%). The configuration at
carbons C1 and C10 was assigned based on positive
correlations found in the bidimensional NOESY experiment
8a,9aZ8a,7Z4.9; 8b,7Z12.9; 8b,9bZ4.9; 8b,9aZ3.7;
13
0
13a,7Z2.3; 13b,7Z2.7; 14,14 Z12.5; C NMR data, see
Table 4.
(H-5, H-7, H-10, H-1) that were in full agreement with the
minimum energy conformer obtained by PM3 calculations.
4.6.4. 14-Acetoxy-15-oxo-cis,cis-germacranolide (12b).
nmax (KBr) 1768 (aldehyde group), 1747 (carbonyl group),
K1
1
731 (acetoxy carbonyl group), 1689 (double bond) cm
;
C
4
(
.5.5. (1R,10S)-1-Hydroxy-14-oxomelampolide (9). nmax
KBr) 3463 (OH), 1759 (carbonyl group), 1722 (aldehyde
HRMS: found [M] 304.13065, C H O requires
15 20 5
C
304.13107; EIMS, m/z (rel. int.): 304 [M] (1.4), 261
K1
C
C
1
group), 1665 (double bond) cm ; HRMS: found [MC1]
65.14103, C H O requires 265.14398; EIMS m/z (rel.
int.): 265 [MC1] (15.0), 247 [MKH O] (46.9), 229
[MKAcO] (1.7); H NMR data, see Table 3, J (Hz):
1,2aZ1,2bZ7.8; 2a,2bZ2a,3bZ14.7; 5,6Z9.2; 6,7Z
3.7; 8a,8bZ14.5; 8a,9bZ12.2; 8a,9aZ5.5; 8b,7Z12.6;
8b,9bZ4.9; 8b,9aZ5.4; 9a,9bZ14.9; 13a,7Z3.1;
2
1
5 20 4
C
C
2
C
1
[
MKH O–CO] (89.8); H NMR data, see Table 3, J (Hz):
2
0
13
1
9
1
,10Z11.8; 3a,3bZ3b,2aZ12.7; 5,6Z6,7Z10.1;
a,9bZ13.6; 9b,8aZ13.6; 9b,8bZ4.9; 13a,7Z3.2;
3b,7Z3.4; C NMR data, see Table 4.
13b,7Z3.4; 14,14 Z12.9; 15,5Z1.3; C NMR data, see
Table 4.
1
3
4
.6.5. 14-Chloro-15-hydroxy-cis,cis-germacranolide
(13). nmax (KBr) 3461 (OH), 1761 (carbonyl group), 1687
4.6. Oxidation of melampolides to cis,cis-
germacranolides
K1
C
(double bond) cm ; HRMS: found [MKCl] 247.13336,
C H O Cl requires 247.13342; EIMS, m/z (rel. int.): 282
1
5 19 3
C
C
C
Selenium dioxide (SeO , 0.5 equiv.) and TBHP (2 equiv.)
2
[M] (1.0), 247 [MKCl] (46.7), 251 [MKCH OH]
2
1
were added to a DCM solution (7 mL) of compounds 2–8
(80 mg each). The reaction mixture was heated until reflux
and strongly stirred over 8 h. Then, the reaction was stopped
(34.2); H NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ9.7;
5,6Z9.6; 6,7Z3.9; 8a,8bZ14.5; 8a,9bZ12.4; 8a,7Z4.9;
8a,9aZ4.9; 8b,7Z12.7; 8b,9bZ4.8; 8b,9aZ3.7; 9a,9bZ
0
13
as it was for compound 2, yielding 10 (10%), 11 (24%), 12
(
14.7; 13a,7Z2.5; 13b,7Z2.7; 14,14 Z11.4; C NMR data,
see Table 4.
57%), 13 (43%), and 14 (3%), respectively. Compound 8
did not react and compound 6 run deprotection of the silyl
ether group and subsequent oxidation of the free hydroxyl
group to yield 3 (53%). It was also possible to detect trace
amounts of the 15-oxo-derivative 12b using 4 as starting
material. The low yields obtained for the rest of starting
materials might be the cause for not detecting analogue
compounds in the other systems assayed.
4.6.6. 14,15-Dihydroxy-1b,10b-epoxide-cis,cis-germa-
cranolide (14). nmax (KBr) 3504 (OH), 1763 (carbonyl
K1
C
group), 1660 (double bond) cm ; HRMS: found [M]
280.13112, C H O requires 280.13108; FAB 280.659
[M] ; EIMS, m/z (rel. int.): 252 [MKH O] (4.0), 244
1
5 20 5
C
C
2
C
1
[MK2H O] (6.0); H NMR data, see Table 3, J (Hz):
2
5
,6Z9.3; 6,7Z4.4; 9a,9bZ14.5; 9a,8bZ9a,8aZ9b,8bZ
0
13
4
.6.1. 14,15-Dihydroxy-cis,cis-germacranolide (10). nmax
KBr) 3406 (OH), 1755 (carbonyl group), 1658 (double
4.3; 13a,7Z2.3; 13b,7Z2.6; 14,14 Z12.4; C NMR data,
see Table 4.
(
bond) cm ; HRMS: found [MKH O] 246.12497,
K1
C
2
C
3
C
C H O requires 246.12504; FAB 264.352 [M] ;
4.7. Silylation of 12
1
5
18
C
EIMS, m/z (rel. int.): 246 [MKH O] (5.0), 228 [MK
2
C
1
2
1
8
9
H O] (6.0); H NMR data, see Table 3, J (Hz): 1,2aZ
,2bZ7.1; 5,6Z9.5; 6,7Z3.7; 8a,8bZ14.3; 8a,9bZ12.4;
a,7Z8a,9aZ4.4; 8b,7Z12.5; 8b,9bZ4.7; 8b,9aZ3.6;
30 mg of 12 were dissolved in N,N-dimethylformamide
(3 mL), followed by addition while stirring of TBDMSCl
(2 equiv., 25 mg) and collidine (2 equiv., 0.04 mL). After
24 h, the reaction was stopped as for compound 2 to afford
the corresponding 14-acetoxy-15-t-butyldimethylsilyloxy-
cis,cis-germacranolide 16 (99%).
2
0
13
a,9bZ14.4; 13a,7Z2.4; 13b,7Z2.7; 14,14 Z12.6;
C
NMR data, see Table 4.
4.6.2. 15-Hydroxy-14-oxo-cis,cis-germacranolide (11).
nmax (KBr) 3444 (OH), 1757 (carbonyl group), 1749
4.7.1. 14-Acetoxy-15-t-butyldimethylsilyloxy-cis,cis-ger-
macranolide (16). nmax (KBr) 1767 (carbonyl group), 1730
K1
aldehyde group), 1681 (double bond) cm ; HRMS:
(
found [MC1] 263.13237, C H O requires 263.12833;
C
K1
(acetoxy carbonyl group), 1682 (double bond) cm
;
1
5 18 4
C
C
C
EIMS, m/z (rel. int.): 263 [MC1] (18.8), 245 [MKH O]
2
100); H NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ7.1;
HRMS: found [MC1] 420.23107, C H O Si requires
21 30 5
1
C
420.23320; EIMS, m/z (rel. int.): 420 [M] (3.8), 361 [MK
(
C C
AcO] (27.9), 303 [MKAcO–HC(CH ) ] (12.6);
1
5
1
,6Z9.5; 6,7Z3.8; 8a,8bZ14.9; 8b,9bZ4.8; 8b,9aZ2.5;
3a,7Z2.4; 13b,7Z2.7; 14,1Z1.3; C NMR data, see
H
3
3
1
3
NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ7.8; 5,6Z9.5;
6,7Z3.7; 8a,8bZ14.2; 8b,7Z12.6; 8b,9bZ4.7; 8b,9aZ
Table 4.
0
13
4
see Table 4.
.0; 13a,7Z2.3; 13b,7Z2.7; 14,14 Z12.4; C NMR data,
4
(
.6.3. 14-Acetoxy-15-hydroxy-cis,cis-germacranolide
12). nmax (KBr) 3426 (OH), 1747 (carbonyl group), 1739
K1
(acetoxy carbonyl group), 1636 (double bond) cm
HRMS: found [M] 306.14678, C H O requires
3
2
;
4.8. Deacetylation of 16
C
1
7
22
5
C
06.14673; FAB 307.153 [MC1] ; EIMS, m/z (rel. int.):
30 mg of 16 were dissolved in dried methanol (1 mL) under
nitrogen atmosphere. Afterwards, magnesium methoxide
(0.08 mL, dilution 7.4% wt. in MeOH) was added. After
24 h, the reaction was stopped by adding water, the reaction
C
75 [MKCH OH] (1.0), 246 [MKCH OH–AcOH]
C
2
2
1
(
7
27.0); H NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ
.5; 5,6Z9.6; 6,7Z3.8; 8a,8bZ14.5; 8a,9bZ12.5;

F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
8487
C
[M] 333.194748, C H NO requires 333.194009; EIMS,
19 26 4
mixture extracted with AcOEt (4!), and the combined
organic phases washed with brine (4!) and dried with
anhydrous sodium sulphate. The solvent was removed in
vacuum. 15-t-Butyldimethylsilyloxy-14-hydroxy-cis,cis-
germacranolide 17 was obtained with a 74% yield, along
with traces of the corresponding Michael adduct 17b.
C
m/z (rel. int.): 333 [M] (5.0), 317 [MKO] (100.0), 288
C
C
C
1
[MKO–CHO] (10.7), 261 [MKC H N] (6.9); H NMR
4
9
dH (400 MHz, CDCl3) 9.40 (d, 1H, JZ1.4 Hz, H-14), 5.18
(brd, 1H, JZ9.8 Hz, H-6), 5.65 (d, 1H, JZ10.1 Hz, H-5),
6.54 (dd, 1H, JZ8.3 Hz, H-1).
2
2
4.8.1. 15-t-Butyldimethylsilyloxy-14-hydroxy-cis,cis-ger-
macranolide (17). n (KBr) 3442 (OH), 1755 (carbonyl
4.10. Germination and seedling growth bioassays
max
K1 C
group) cm ; HRMS: found [M] 378.22259, C H O Si
Seeds of Lactuca sativa L. cv. Roman (lettuce), Lepidium
sativum L. cv. Com u´ n (cress), Allium cepa L. cv.
Valenciana (onion), and Triticum aestivum L. cv. Cortex
2
1 34 4
C
requires 378.22264; FAB 379.124 [MC1] ; EIMS, m/z
C
rel. int.): 361 [MKH O] (3.8), 323 [MKHC(CH3)3]
C
(
2
1
´
(
5.1); H NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ7.2;
(wheat) were obtained from FITO, S.L. (Barcelona, Spain).
5
,6Z9.6; 6,7Z3.7; 8a,8bZ14.3; 8b,7Z12.7; 8b,9aZ4.2;
All undersized or damaged seeds were discarded and the
assay seeds were selected for uniformity. Bioasays were
carried out in 9 cm Ø plastic Petri dishes, using Whatman #1
filter paper as support. The general procedure for seedling
bioassay was as follows: 25 seeds of each species were
placed per dish, excepting Triticum aestivum (10 seeds per
dish), with 5 mL of test soln, and incubated in the dark at
25 8C. Four replicates for each concentration were set up.
Germination and growth time varied for each plant species:
L. sativum, 3 days; L. sativa and T. aestivum, 5 days; and
0
13
13a,7Z2.4; 13b,7Z2.7; 14,14 Z12.6; C NMR data, see
Table 4.
4
1
.8.2. (11S)-15-t-Butyldimethylsilyloxy-11,13-dihydro-
4-hydroxy-13-methoxy-cis,cis-germacranolide (17b).
nmax (KBr) 3446 (OH), 1768 (carbonyl group), 1658
(
K1 C
double bond) cm ; FAB 409.226 [MK1] ; EIMS, m/z
C C
rel. int.): 393 [MKOH] (4.7), 353 [MKC(CH ) ] (3.0);
H NMR data, see Table 3, J (Hz): 1,2aZ1,2bZ6.6; 5,6Z
(
3
3
1
K2
9
8
1
.2; 6,7Z5.2; 8a,8bZ14.1; 8b,7Z12.9; 8b,9aZ4.0;
Allium cepa, 7 days. Test mother solutions (10 M) were
prepared using dimethyl sulfoxide (DMSO) and then diluted
0
b,9bZ4.6; 9a,9bZ14.9; 9b,8aZ4.0; 13,13 Z9.3;
0
13
K4
to 10 M using 10 mM MES (2-[N-morpholino]ethane-
3,11Z5.5; 13 ,11Z3–7; C NMR data, see Table 4.
The stereochemistry at C-11 was assigned based on positive
nOe correlations found in the bidimensional NOESY
experiment between H-6 and H-13.
sulphonic acid). Following solutions were obtained by
dilution maintaining the 0.5% DMSO percentage. Parallel
controls were performed. All pH values were adjusted to 6.0
before bioassay. All products were purified prior to the
bioassay using HPLC equipped with a refractive index
detector. Minimum degree of purity was of 99% as extracted
from the chromatograms.
4.9. Michael additions of pyrrolidine to compounds 3
and 11
0
.15 mmol each of compounds 3 and 11 were dissolved in
dry THF (0.5 mL) and the solution cooled in the freezer for
5 min. Pyrrolidine solution (21 mg, 0.3 mmol) in dry THF
0.5 mL) was added to the cooled solution and the mixture
4.11. Statistical treatment
1
(
Germination and root and shoot length were tested by
Welch’s test, the differences between test solutions and
2
8
kept at room temperature during 20 h. The solvent was
evaporated in vacuum and the crude of reaction purified on a
Al O column chromatography using DCM/MeOH (97:3)
controls being significant with P!0.01. Cluster analysis
2
9
was performed using the Statistica package. The analyses
were recorded to all compounds tested using as variables
germination index, and root and shoot growth.
2
3
27
as eluant, to yield 75% of 20 and 21 in each case. The
stereochemistry of the adduct at C-11 was established as R
for compound 20 since a positive nOe effect was observed
for H-11 when the signal corresponding to H-6 was
irradiated (% nOe: 2.8% for H-11, 2.8% for H-9, 1.2% for
H-15, and 3.1 for H-8b). No positive nOe effects could be
observed on the side chain signals when H-5 was irradiated
4.12. Molecular modeling
Minimum energy conformations and molecular properties
were obtained using MMX and PM3 calculations (PCMO-
DEL ver 6.0, Serena Software, Bloomington, IN; MOPAC,
ver. 6.00), respectively. Conformers were obtained using the
randomize command in PCMODEL and the local minimum
energy structures obtained were used for further semiem-
piric minimization with MOPAC using PM3 method. For
semiempiric calculations the parameters PRECISE, GEO-
(
% nOe: 1.2% for H-1, 6.3% for H-2b, and 8.4% for H-7).
4
.9.1. (11R)-14-Oxomelampolide pyrrolidine mono
adduct (20). nmax (KBr) 1768 (carbonyl group), 1658
K1 C
double bond) cm ; HRMS: found [M] 317.196936,
(
C H NO requires 317.199094; EIMS, m/z (rel. int.): 317
1
9
26
3
C
M] (100.0), 288 [MKCHO] (9.1), 246 [MKC H N]
C
C
[
OK, and TZ86400 were used. Theoretical DH 8 values
f
4
8
(
4.6); data, see Table 3, J (Hz): 1,2aZ9.1; 1,2bZ7.1;
produced by MOPAC allowed to discriminate among
conformers and to obtain the minimum energy conformer
in each case.
2
2
8
1
a,2bZ12.4; 2a,3bZ6.0; 2a,3aZ3.1; 2b,3aZ7.0;
b,3bZ2.1; 5,6Z6,7Z10.1; 8a,8bZ14.2; 8a,9bZ12.5;
0
b,9bZ1.9; 9a,9bZ13.9; 13,13 Z13.2; 13,11Z4.0;
0
13
3 ,11Z5.2; C NMR data, see Table 4.
Acknowledgements
4.9.2. 14-Oxo-15-hydroxy-cis,cis-germacranolide pyrro-
lidine mono adduct (21). nmax (KBr) 3446 (OH), 1768
(
This research was supported by the Ministerio de Ciencia y
Tecnolog ´ı a, Spain (MCYT; Project No. AGL2001-3556).
K1
carbonyl group), 1658 (double bond) cm ; HRMS: found

8488
F. A. Mac ´ı as et al. / Tetrahedron 60 (2004) 8477–8488
R. F. Velasco acknowledges fellowships from Fundaci o´ n
Caja de Madrid, Universidad de C a´ diz, and Ministerio de
´
Educaci o´ n y Ciencia. We thank FITO, S.A. and SYN-
GENTA, for providing seeds and herbicides, respectively.
L.; Hern a´ ndez-Bautista, B. E.; Pelayo-Benavides, H. R.;
Calera, M.; Fern a´ ndez-Luiselli, E. ACS Symposium Ser.
1995, 582, 224–241.
14. For a review see: Minnaard, A. J.; Winjberg, J. B. P. A.; de
Groot, A. Tetrahedron 1999, 55, 2115.
1
5. (a) Mac ´ı as, F. A.; Galindo, J. C. G.; Castellano, D.; Velasco,
R. F. J. Agric. Food Chem. 1999, 47, 4407. (b) Mac ´ı as, F. A.;
Galindo, J. L. G.; Molinillo, J. M. G.; Castellano, D.
Phytochemistry 2000, 54, 165.
References and notes
1
6. Lempers, H. E. B.; Rippoll e` s i Garc ´ı a, A.; Sheldon, R. A.
J. Org. Chem. 1998, 63, 1408.
1
. Sharpless, K. B.; Lauer, R. F. J. Am. Chem. Soc. 1972, 94,
154.
7
1
1
7. Haruna, M.; Ito, K. J. Chem. Soc., Chem. Commun. 1981, 483.
8. MOPAC VS 6.00, Stewart, J. J. P.; Seiler, F. J. Research
Laboratory: US Air Force Academy, USA, 1990.
2. Singleton, D. A.; Hang, C. J. Org. Chem. 2000, 65, 7554.
3. Snider, B. B.; Shi, B. Tetrahedron 1999, 55, 14823.
4
. Mac ´ı as, F. A.; Galindo, J. C. G.; Castellano, D.; Velasco,
R. F. J. Agr. Food Chem. 2000, 48, 5288.
1
9. Xu, Y. C.; Bizuneh, A.; Walker, C. Tetrahedron Lett. 1996, 37,
455.
0. (a) Galindo, J. C. G.; Hern a´ ndez, A.; Dayan, F. E.; Tellez,
M. R.; Mac ´ı as, F. A.; Paul, R. N.; Duke, S. O. Phytochemistry
5
. (a) Mac ´ı as, F. A.; Galindo, J. C. G.; Massanet, G. M. M.
Phytochemistry 1992, 31, 1969. (b) Vyvyan, J. R. Tetrahedron
2
2002, 58, 1631.
1
999, 52, 805. (b) Schmidt, T. J. Bioorg. Med. Chem. 1997, 5,
45. (c) Hwang, D.; Fischer, N. H.; Jang, B. C.; Tak, H.; Kim,
6
. Gressel, J. In Weed Physiology Vol. II Herbicide Physiology,
Duke, S. O. ed., CRC Press: Boca Ratan FL. 1985; pp 159–
6
J. K.; Lee, W. Biochem. Biophys. Res. Commun. 1996, 226,
10.
1
89.
. Mudge, L. C.; Gosett, B. J.; Murphy, T. R. Weed Sci. 1984, 32,
9.
8
7
8
9
2
2
2
1. Torsell, K. B. G. Natural Products Chemistry; Swedish
Pharmaceutical Society: Apotekarsocieteten, 1997 p 58.
2. Mac ´ı as, F. A.; Castellano, D.; Molinillo, J. M. G. J. Agr. Food
Chem. 2000, 48, 2512.
5
. Waller, G. R. Allelochemicals. Role in Agriculture and
Forestry. ACS Symposium Series 1987 p 330.
. (a) Mittra, S.; Datta, A. P.; Singh, S. K.; Singh, A. Acta
Pharmacol. Sin. 2000, 21, 1106. (b) Jain, N. K.; Kulkarni, S. K.
J. Ethnopharmacol. 1999, 68, 251.
3. (a) Yuan, Q.; Yan, S. G.; Chai, G. Q.; Zhang, S. X.; Zheng,
C. Z. Acta Chim. Sinica 1999, 57, 100. (b) Nagasaki, K.;
Tarutani, K.; Yamaguchi, M. J. Plankton Res. 1999, 21, 2219.
1
0. (a) Park, E. J.; Kim, J. Planta Med. 1998, 64, 752. (b)
Beekman, C.; Woerdenbag, H. J.; van Uden, W.; Pras, N.;
Konings, A. W. T.; Wikstroem, H. V.; Schmidt, T. J. J. Nat.
Prod. 1997, 60, 252.
24. Velasco, R. F. Ph.D. Dissertation, University of Cadiz, 2003.
25. Primo-Y u´ fera, E.; Carrasco-Dori e´ n, J. M. Qu ´ı mica Agr ´ı cola II.
Plaguicidas y Fitorreguladores, Alhambra, Ed.; 1980, 511.
26. Clark, W. D.; Corbett, T.; Valeriote, F.; Crews, P. J. Am.
Chem. Soc. 1997, 9, 9285.
1
1. Wedge, D. E.; Galindo, J. C. G.; Mac ´ı as, F. A. Phytochemistry
2000, 53, 747.
1
2. (a) Fran c¸ ois, G.; Passreiter, C. M.; Woerdenbag, H. J.; van
Looveren, M. Planta Med. 1996, 62, 126. (b) Taylor, R. S. L.;
Towers, G. H. N. Phytochemistry 1998, 47, 631.
3. (a) Mac ´ı as, F. A.; Oliva, R.; Varela, R. M.; Torres, A.;
Molinillo, J. M. G. Phytochemistry 1999, 52, 613. (b) Anaya,
27. Hejchman, E.; Haugwitz, R. D.; Cushman, M. J. Med. Chem.
1995, 38, 3407.
28. Zar, J. H. Statistical Analysis; Prentice Hall, Inc: Englewood
Cliffs, NJ, 1984.
1
29. StatSoft Inc., Release 4.5, 1993.