The crocacins: Novel natural products as leads for agricultural fungicides

Article abstract of DOI:10.2533/000942903777678669

A short route to the synthesis of analogues of the antifungal but unstable natural product crocacin D was developed. A wide range of compounds were made in which the complex side chain was replaced by simple aromatic units, and many of them showed high activity in a mitochondrial beef heart respiration assay. Some of these compounds were active against plant pathogenic fungi of agricultural importance in glasshouse tests. In order to find compounds with greater stability than the natural crocacins, the photolabile (Z)-enamide was replaced with many different moieties, but none gave rise to useful fungicidal activity.

Full text of DOI:10.2533/000942903777678669

CROP PROTECTION RESEARCH
685
CHIMIA 2003, 57, No.11
Chimia 57 (2003) 685–691
© Schweizerische Chemische Gesellschaft
ISSN 0009–4293
The Crocacins: Novel Natural Products
as Leads for Agricultural Fungicides
Patrick J. Crowley*, Ian H. Aspinall, Kevin Gillen, Christopher R.A. Godfrey, Ingrid M. Devillers,
Gordon R. Munns, Olivia-Anabelle Sageot, Joe Swanborough, Paul A. Worthington, and John Williams
Abstract: A short route to the synthesis of analogues of the antifungal but unstable natural product crocacin
D was developed. A wide range of compounds were made in which the complex side chain was replaced by
simple aromatic units, and many of them showed high activity in a mitochondrial beef heart respiration assay.
Some of these compounds were active against plant pathogenic fungi of agricultural importance in
glasshouse tests. In order to find compounds with greater stability than the natural crocacins, the photola-
bile (Z)-enamide was replaced with many different moieties, but none gave rise to useful fungicidal activity.
Keywords: Agrochemical · Crocacin · Fungicide · Natural product · Respiration
1. Introduction
In 1994 Profs. Höfle and Reichenbach Septoria nodorum (wheat glume blotch).
at GBF in Braunschweig, Germany isolated We also confirmed their mode of action as
Natural products have played a significant the new compounds crocacins A, B, C, and inhibition of mitochondrial respiration at
role in the discovery of new agricultural D from the myxobacterium Chondromyces complex III in beef heart and the model fun-
fungicides because they often have interest- crocatus, and identified their structures as gus Neurospora crassa.
ing structures, novel biochemical modes of the novel (Z)-enamides 1, 2, 3, and 4 [2]
With their novel structures, interesting
action, and unusual biological activity. (Fig. 1). In further studies crocacins A (1) mode of action and significant fungicidal
However, they also frequently have highly and D (4)were shown to inhibit the electron activity, the crocacins were attractive leads
complex structures, are often unstable, and transport chain at complex III in a beef for synthesis, although ambitious targets as
are not always as biologically active as heart mitochondrial respiration assay, and agrochemicals due to their relatively com-
modern commercial pesticides. Neverthe- to inhibit the growth of several fungi in vit- plex structures. We therefore decided to
less their novel structures continue to in- ro. Crocacins B (2) and C (3) had no signif- embark on a programme of chemistry to see
spire chemists to synthesise simplified ana- icant biological activity.
if simpler, more active and more stable
logues with optimised properties such as a
The mode of action was of special in- compounds could be found.
broad spectrum of disease control, good terest as the important commercial fungi-
For practical use on crops in the field,
persistence of effect and high cost efficacy cides azoxystrobin (5) and kresoxim- fungicides must have reasonable stability
[1].
methyl (6) [3], analogues of the natural on leaf surfaces so that they can kill incom-
product strobilurin A (7), also inhibit mito- ing fungal spores over a period of 1–2
chondrial respiration in plant pathogens at weeks as well as those that are already pres-
complex III [4] (Fig. 2). Modelling studies ent. Both crocacin A (1) and D (4) were
suggested that the crocacins interacted with shown to have low levels of photostability,
the active site of complex III in a different with 50% of the parent compound being
manner to that of the strobilurins.
lost in 7 and 37 min, respectively, as thin
We obtained samples of crocacins A (1) films on glass slides in a simulated sunlight
and D (4) from Profs. Hofle and Reichen- test in the laboratory. This low photostabil-
bach and confirmed their activity against ity is not surprising in view of the number
certain fungi in vitro, and also demonstrat- of double bonds the crocacins contain, and
ed that they were active by foliar spray would undoubtedly lead to very poor per-
against some plant pathogenic fungi on sistence of fungicidal activity on plants, se-
small plants. Crocacin D (4) was slightly verely limiting the activity that these com-
more active than crocacin A (1). The dis- pounds could show in the field.
*Correspondence: Dr. P.J. Crowley
Syngenta
Jealott’s Hill International Research Centre
Bracknell
Berkshire, RG42 6EY, UK
Tel.: +44 1344 414840
Fax: +44 1344 413739
eases which were controlled included Plas-
There was no significant aqueous hy-
mopara viticola (vine downy mildew), drolysis of the glycine methyl ester group
Phytophthora infestans (potato late blight), of crocacin A (1) at pH 4 and 7 at 39 oC, but
Puccinia recondita (wheat brown rust), and addition of pig liver esterase under these
E-Mail: patrick.crowley@syngenta.com

CROP PROTECTION RESEARCH
686
CHIMIA 2003, 57, No.11
conditions led to rapid generation of the
acid, suggesting that the compounds might
be easily metabolised in plants or in fungi.
This paper will briefly discuss the range
of analogues made to overcome these defi-
ciencies and the chemical approaches to
them, and only limited reference will be
made to biological results. A more detailed
discussion of the structure activity relation-
ships, the extensive use of molecular mod-
elling, and X-ray crystallographic studies
will be published elsewhere.
Me Me
Me Me
NH2
N
H
O
N
H
COOR
OMe OMe
Me
O
OMe OMe
Me
O
1 R = Me Crocacin A
2 R = H Crocacin B
3 Crocacin C
Me Me
N
H
O
N
H
COOMe
OMe OMe
Me
O
4 Crocacin D
Fig. 1. Structures of crocacins A–D
2. Analogues of Crocacins A (1)
and D (4)
2.1. Design Strategy
N
N
Me
There were three major aims in the de-
sign of analogues (Fig. 3). The first was to
find compounds with improved fungicidal
potency on a wider spectrum of plant
pathogens. The second was to increase sta-
bility compared to the natural products so
that analogues would be able to exhibit fun-
gicidal activity under field conditions. The
third was to reduce the complexity of the
molecules so that they could be synthesised
economically. We chose to concentrate on
making analogues of crocacin D (4), which
was more active and simpler in structure
than crocacin A (1), having only one (Z)-
double bond in the central section of the
molecule, although a small number of cro-
cacin A analogues were made.
An essential element of this strategy
was to replace most, if not all, the double
bonds by more stable groups, in order to
achieve photostability. We hoped that this
would not only improve fungicidal activity
by increasing persistence, but would also
lead to simpler molecules.
O
O
O
OMe
MeOOC
CN
OMe
OMe
MeOOC
MeOOC
N
5
6
7
Azoxystrobin
Kresoxim-methyl
Strobilurin A
Fig. 2. Structures of strobilurin analogues and strobilurin A (7)
Increase photostability
Reduce hydrolysis
Me
Me
N
COOMe
O
N
H
H
OMe OMe
Me
O
d
a
c
e
b
Simplify
Fig. 3. Plan for design of analogues of crocacin D (4)
We speculated that the (Z)-enamide
glycine ester (fragments d and e in Fig. 3)
might be the key toxophore in crocacins A
(1) and D (4), and that the lipophilic side
chain with the styryl group (fragment a), the
array of four contiguous chiral centres
(fragment b) and the diene (fragment c) was
essentially ‘decoration’ which could be re-
placed by stable aromatic groups. The (Z)-
enamide was predicted to be quite unstable,
and would probably need to be replaced by
a more robust group. The glycine ester
(fragment e) was also a source of potential
degradation through hydrolysis by esteras-
es in plant or fungal cells, and might need
to be replaced by groups more resistant to
hydrolysis.
O
1
N
H
O
N
COOMe
Crocacin A
H
O
= double bond
8
Me Me
N
H
O
N
H
COOMe
OMe OMe
Me
O
(CH2)9
N
4
H
O
N
H
COOMe
Crocacin D
O
= single bond
9
Fig. 4. Analogues of crocacin A and D with double bonds replaced
Following the logic of these ideas, we
derived a set of target structures in which
much of the crocacin structure had been side chain was replaced by very much sim- 4). All compounds were tested on a cascade
changed. The (Z)-double bonds were re- pler groups. Examples of some of the ana- consisting of a beef heart mitochondrial
placed with benzene rings, and the whole logues made are compounds 8 and 9 (Fig. respiration assay, followed by application

CROP PROTECTION RESEARCH
687
CHIMIA 2003, 57, No.11
to fungi on leaf disks, with the most active
Overall we concluded that it is difficult nificantly. Simple esters can be prone to hy-
compounds then being tested on fungi on to mimic the rigidity of the (Z)-enamide drolysis by esterases in plants, and indeed
small plants. Unfortunately neither com- without increasing steric bulk, and that it is tests using maize cell and liquid fungal cell
pounds 8 or 9 or a range of close analogues likely that space in the active site is very re- cultures showed that the methyl ester of
showed activity in any of these tests.
stricted.
several analogues was cleaved very rapidly.
Some effort was therefore devoted to look-
ing for groups that might mimic the methyl
2.2. Variation of the Side Chain
2.4. Mimics of the Glycine Ester
We realised that in pursuing the highly
The glycine methyl ester was found ester but be more resistant to hydrolysis
simplified structures of analogues 8 and 9 to be important for activity, and simple (Table 3). The N-methylamide 25 was inac-
we may have made too many changes at changes had a marked effect on activity. For tive, but the methyl ketone 29 showed quite
once. We therefore decided to adopt a step- example, substitution on the glycine nitro- good activity. Certain heterocycles have
wise approach, changing one part of the gen or α-carbon atoms reduced activity sig- been shown to be good methyl ester re-
molecule at a time. We believed that the
(Z)-enamide glycine ester was the key tox-
ophore, so we opted to maintain this moiety
Table 1. Examples of side chain replacements
in our analogues, while replacing the com-
plex side chain with simplified aromatic
groups. If this strategy turned out to be suc-
cessful, we could then seek replacements
for the (Z)-enamide linker, and mimics for
the labile glycine ester.
R
NH
O
N
H
COOMe
O
Compound
No.
IC50, NADH
EC90
Plasmopara
viticola
Indeed, as we had hoped, the complex
side chain of crocacins A (1) and D (4)
proved not to be necessary for activity, and
could be greatly simplified (Table 1). In the
first analogues synthesised we replaced
the whole side chain with simple n-alkyl
chains. An example was compound 10
which was very active in the beef heart mi-
tochondrial respiration assay, although it
was not fungicidal. Benzamides substituted
with alkoxy groups such as compound 11
were similarly good inhibitors of respira-
tion, but were inactive on fungi. However,
substituting the benzamide with chains
containing aryl groups, such as compound
12, now produced activity on fungi on leaf
disks and on plants at a level similar to the
crocacins. Benzamides with a 4-substituent
were much more active than those substi-
tuted in the 3-position, and highest activity
was shown by 4-substituted benzyl ana-
logues such as compound 13.
R
Oxidase
[nanomolar]a
[ppm]b
n-C10H21
10
11
24
21
36
>100
>100
25
n-C6H11O
Me Me
4
OMe OMe
Me
O
12
13
9
25
10
Et
Br
17
aBeef heart mitochondrial preparation. bFoliar spray on small vines.
2.3. Replacement of the (Z)-Enamide
Since these analogues retaining the (Z)-
enamide group were shown to be quite un-
stable in laboratory tests, replacing it with
alternative but stable linkers was predicted
to improve fungicidal activity on plants by
increasing the persistence of the compound
on leaf surfaces. Many different types of
analogue were synthesised (Fig. 5), includ-
ing some with rings as linking groups such
as compounds 14–20, and some with flexi-
ble chains such as compounds 21–23. None
of these compounds were fungicidal and
only the benzoxazolone 20 had any signifi-
cant activity on the respiration assay (Table
2). The cis-cyclopropane 24 on the other
hand did have weak activity on fungi, but
its very poor level of inhibition of respira-
tion casts doubt on this as the mode of ac-
tion.
R
O
N
R
R
N
O
X
X
O
O
COOMe
COOMe
COOMe
14
15
16
X = O, C
H
N
R
R
N
O
R
N
O
COOMe
O
H
O
N
H
COOMe
O
COOMe
O
17
18
19
Fig. 5. Examples of (Z)-enamide replacements

CROP PROTECTION RESEARCH
688
CHIMIA 2003, 57, No.11
Table 2. Examples of (Z)-enamide replacements
placements [5], and several were tried. The
pyrazine 28 was the best, but although it
was resistant to hydrolysis, it was only
weakly active in the respiration assay and
on fungi.
R
Linker
COOMe
O
IC50, NADH
EC90
Plasmopara
viticola
Compound
No.
R
Linker
Oxidase
[nanomolar]a
3. Biological Activity of Analogues
[ppm]b
O
The most active analogues controlled a
similar spectrum of fungal diseases to the
crocacins A (1) and D (4), showing activity
in glasshouse tests principally against Plas-
mopara viticola and Phytophthora infes-
tans on small vine and tomato plants re-
spectively. They also showed high activity
on the beef heart respiration assay. How-
ever, many compounds that were good in-
hibitors of respiration were poor or inactive
fungicides. The reasons for this are not
clear, but may reflect either subtle differ-
ences between mammalian and fungal
complex III active sites, or differences be-
tween compounds in metabolism or stabili-
ty.
O
N
O
20
140
>100
Et
O
O
O
H
N
H
N
230
1600
400
>100
>100
>100
21
22
23
O
Et
Et
O
H
N
H
O
O
N
O
H
H
N
N
Et
Et
O
O
H
24
920
100
H
N
H
NH
Although the analogues were shown to
be more photochemically stable than the
natural products, they still had relatively
short half-lives on plant surfaces, probably
due to the photochemical instability of the
(Z)-enamide, and none were active enough
to progress further.
aBeef heart mitochondrial preparation. bFoliar spray on small vines.
Table 3. Examples of ester mimics
O
R
4. Synthetic Routes
N
COOR1
H
O
In order to make a range of analogues to
explore structure activity relationships, we
required a synthesis of (Z)-enamides. There
have been several published syntheses of
the complex side chain of the crocacins [6],
and recently the first total syntheses of both
crocacins A (1) and D (4) have been pub-
lished [7], using an elegant approach to the
(Z)-enamide. However, when our work be-
gan in 1995 there was no published chem-
istry in this area, and there were no good
general methods for synthesising (Z)-enam-
ides. The priority throughout this work was
to obtain enough compound to test as rap-
idly as possible, so most of the chemistry
was not optimised, and it is likely that the
yields could be improved considerably.
Several attempts were made to synthe-
sise a (Z)-enamide using a lithiation ap-
proach. Although alkylation of the model
compound 30 with simple alkyl halides to
give 31, following the work of the Tischlers
[8] worked well (Scheme 1), no useful
products could be obtained from alkylation
with groups that could be converted into an
ester, for example to give 32 or 33.
IC50, NADH
EC90
Plasmopara
viticola
Compound
No.
R
R1
Oxidase
[nanomolar]a
[ppm]b
880
210
>100
N
H
CONHMe
25
26
Et
Et
>100
N
H
COOMe
70
110
49
9
>100
100
30
N
COOMe
27
28
Me
Et
Et
N
N
H
N
(CH2)5
N
H
COMe
29
12
25
N
H
COOMe
Et
aBeef heart mitochondrial preparation. bFoliar spray on small vines.

CROP PROTECTION RESEARCH
689
CHIMIA 2003, 57, No.11
Due to the difficulties of introducing the electrolysis of caprolactam in ethanol using Protection of the amide nitrogen with Boc
(Z)-enamide we decided to look for starting a platinum anode. Elimination of ethanol to give compound 38 provided the neces-
points where the (Z)-double bond was al- then gave compound 35 [8]. Acylation of sary activation of the ring amide carbonyl
ready present, such as the 1,3,4,5-tetrahy- the amide nitrogen with acid chlorides was group, and reaction with glycine ester now
dro-2H-azepin-2-one (35) (Scheme 2). We best carried out in the presence of powdered occurred smoothly to give the Boc protect-
envisaged that acylation of compound 35 to molecular sieves, with no base. However, ed compound 39 in 70% yield. Unfortu-
give compound 36, followed by ring open- subsequent ring opening with glycine ester nately selective acylation of the Boc amide
ing with glycine ester, would give the ana- to give compound 41 or with hydroxide to group to give compound 40 could not be
logues of type 41 directly. The known 7- give compound 37 proved to be extremely achieved, and this approach was aban-
ethoxy-1,3,4,5,6,7-hexahydro-2H-azepin- low yielding due to preferential cleavage of doned.
2-one (34) was made in good yield by the exocyclic acyl group when R was alkyl.
LDA (2 Equiv.)/THF
N
N
H
H
n-BuI
80% (90% (Z))
O
O
31
30
LDA (2 Equiv.)/THF
N
OMe
H
MeO
Br
OMe
OMe
O
v. low yield
32
(1) n-BuLi (2 Equiv.)/THF
(2) CuI or ZnBr2
N
COOMe
H
Br
COOMe
O
33 v. low yield
Scheme 1. A lithiation approach to (Z)-enamides
-2e
LiOH
NaH/THF
RCOCl
40-60%
O
NH4Br/toluene
4A mol sieve
80%
O
N
COOH
EtO
N
N
N
EtOH
62%
N
H
O
H
O
R
H
O
H
R
O
34
35
36
37
(BOC)2O
DMAP/CH2Cl2
95%
HCl.H2N
COOMe
Et3N/CH2Cl2
N
O
O
O
HCl.H2N
COOMe
38
Et3N/CH2Cl2
70%
Deprotection
O
N
O
N
O
H
O
N
H
COOMe
O
N
O
N
H
COOMe
H
O
N
COOMe
O
O
R
R
H
39
40
41
Scheme 2. Synthesis and use of tetrahydroazepinone 35

CROP PROTECTION RESEARCH
690
CHIMIA 2003, 57, No.11
An alternative route outlined in Scheme
3 was then tried. The ring in compound 38
was opened quantitatively with sodium
methoxide to give compound 42, and reac-
tion of the amide anion generated using
LiN(SiMe3)2 with various acid chlorides
gave the imides 43, although only in
20–30% yield. Selective deprotection of the
Boc group with Mg(ClO4)2, again in poor
yield, followed by hydrolysis of the ester,
gave the key carboxylic acids 37. Simple
amide coupling with glycine ester then af-
forded the desired crocacin analogues 41.
Although this sequence gave some valuable
analogues where R was a variety of long
chain alkyl groups, the overall yields were
very low, and purification of the final prod-
ucts was difficult.
MgClO4
LiN(Me3Si)2
NaOMe
O
N
COOMe
H
O
N
O
COOMe
MeCN
20-60%
O
N
MeOH
quant.
o
-70 C, RCOCl
20-35%
O
O
O
R
O
42
43
38
HCl.H2N
COOMe
LiOH
O
N
COOMe
O
N
COOH
O
N
H
MeOH
40-60%
H
H
O
N
H
COOMe
EDC/DMAP
30-60%
R
R
R
44
37
41
Scheme 3. Synthesis of crocacin analogues with alkyl side chains
However, for analogues where R was an
aromatic group (Scheme 4), we found a sig-
nificantly shorter variation of this route,
which although giving only moderate
yields, was much more convenient and en-
abled many analogues to be made. The di-
rect acylation of the tetrahydroazepinone
35 with a benzoyl chloride could again be
achieved in good yield in the presence of
molecular sieves, with no base added. The
less reactive benzamide group in 36 was
now less susceptible to hydrolysis, and
treatment with hydroxide provided a work-
able yield of the ring opened product 37, al-
though variable amounts of acyl cleavage
to regenerate the benzoic acid 45 still oc-
curred. As a matter of practicality, it was
convenient to couple the mixture of the car-
boxylic acids 37 and 45 with glycine ester,
and then separate the desired amide product
41 from the unwanted compound 46. This
synthetic route to analogues also worked
well when R was an α,β-unsaturated group,
or a 2- or 3-pyridyl group.
Variations of this route were used to
make analogues 47 where the double bond
was replaced by a cyclopropyl group,
(Scheme 5), and also where the whole link-
ing group was replaced by a chain contain-
ing an oxygen atom as in compound 49. In
the latter case an aryl side chain in com-
pound 48 was again required to ensure
cleavage at the ring rather than the side
chain in the hydrolysis step.
HCl.H2N
COOH
COOMe
LiOH
RCOCl
O
N
O
N
H
+
O
N
H
COOMe
N
N
H
ClCH2CH2Cl
H2O/THF
quant.
R
O
Et3N/DMAP
CH2Cl2
O
Mol. sieve
Reflux
40-60%
H
O
R
R
+
50:50
35
41
36
37
30-60% overall
O
COOH
R
R
N
COOMe
45
H
46
O
R = e.g.
C5H11
N
Et
Et
Scheme 4. Synthesis of crocacin analogues with aromatic or unsaturated side chains
Me3SO+I-
NH2OH.HCl
TsCl/Py
(BOC)2O
88%
+
K2CO3/THF/
H2O
NaH/DMSO
73%
NaOAc/H2O
quant.
N
N
N
H
O
O
H
O
O
OH
3 : 5
separable
MeOOC
NH2.HCl
LiOH/THF/H2O
O
N
O
N
H
COOH
H
O
N
H
COOMe
EDC/DMAP
75%
N
quant.
O
O
O
O
O
A similar approach was also used to
make the benzo-analogues 50 (Scheme 6).
A Beckmann rearrangement of tetralone
oxime gave the tetrahydrobenzazepinone
50, which as the N-Boc derivative 51, was
reacted with glycine ester regioselectively
to compound 52. Deprotection then gave
the aniline 53, which was acylated with
a wide range of carboxylic acids RCOOH,
to give analogues 54.
RCOOH
TFA/CH2Cl2
98%
R
N
NH2
EDC/DMAP/
CH2Cl2
H
O
N
H
COOMe
O
N
H
COOMe
O
70-80%
47
O
O
RCOCl
H
N
LiOH
H
N
H
N
EDC
O
O
COOMe
O
COOH
O
DMAP
70-80%
60-70%
ClCH2CH2Cl
N
H
O
N
O
R
R
O
mol. sieve
o
60 C, 70-80%
O
R
49
48
The benzylbenzoic ester intermediates
required for the most active analogues were
Scheme 5. Synthesis of cyclopropane and flexible chain (Z)-enamide replacements

CROP PROTECTION RESEARCH
691
CHIMIA 2003, 57, No.11
made according to Scheme 7. Grignard re-
action of ethyl 4-iodobenzoate [10] worked
very well to give the key benzhydrol 55,
from which a range of benzylic side chain
acids could be made. Reduction to the ben-
zylbenzoic esters 57 was achieved in high
yields using an indium trichloride catalysed
reaction with dichlorodimethylsilane [11].
Methylation of the benzylic hydroxyl to
give 58 was most efficiently carried out us-
ing trimethylsilyldiazomethane in the pres-
ence of HBF4 [12]. Oxidation and Wittig re-
action to give the methylene compound 59,
followed by triethylsilane reduction gave
the methyl derivative 60.
PPA, 120oC
85%
(Boc)2O
HCl.H2N
COOMe
O
N
DMAP/
Et3N/CH2Cl2
quant.
N
COOMe
H
O
N
H
N
Et3N/CH2Cl2
55%
O
O
NOH
O
H
O
O
51
50
52
TFA
quant.
RCOOH/EDC
DMAP/
Et3N/CH2Cl2
NH2
R
N
O
N
COOMe
H
O
N
H
COOMe
40-60%
H
O
53
54
Scheme 6. Synthesis of crocacin analogues with the (Z)-enamide replaced by a benzene ring
OH
MeO
i-PrMgCl
CHO
5. Conclusions
I
COOEt
THF/-40 o
C
65%
COOEt
MeO
We have successfully made a range of
analogues of crocacins A (1) and D (4)
which were considerably simpler than the
natural products. Although a number of
these were active against plant pathogenic
fungi on plants, none were significantly
better, either in activity or stability, than the
natural products. Despite considerable ef-
fort we were unable to find stable replace-
ments for the central portion of the mole-
cule that gave rise to significant fungicidal
activity; the only analogues that were active
retained the photochemically vulnerable
(Z)-enamide group. We believe that space
in the active site where the (Z)-enamide fits
is probably restricted, so that any change to
this part of the structure reduces binding of
the whole molecule. Analogues of the cro-
cacins are therefore unlikely to be suffi-
ciently active or stable to make useful agri-
cultural fungicides.
55
Me3SiCHN2
aq. HBF4
61%
PCC
18-C-6
CHCl3
Me2SiCl2/
CH2Cl2
InCl3
94%
74%
OMe
O
COOEt
MeO
COOEt
MeO
COOEt
MeO
56
57
58
Ph3P+Me Br-
n-BuLi
50%
Et3SiH/
Me
TFA
57%
COOEt
MeO
COOEt
MeO
60
59
Scheme 7. Synthesis of benzylbenzoic acid side chains
[5] R.T. Lewis, T. Richard, A.M. Macleod,
K.J. Merchant, F. Kelleher, I. Sanderson,
R.H. Herbert, A. Cascieri, S. Sadowski,
R.G. Ball, K. Hoogsteen, J. Med. Chem.
1995, 38, 923; R. Jansen, P. Washausen,
B. Kunze, H. Reichenbach, G. Höfle, Eur.
J. Org. Chem. 1999, 1085; P. Krogsgaard-
Larsen, E. Falch, P. Sauerberg, S.B. Freed-
man, H.L. Lemboel, E. Meier, Trends in
Pharmacological Sciences, 1988, (Sup-
pl.), 69.
[6] a) J.T. Feutrill, M.J. Lilly, M.A. Rizzacasa,
Org. Lett. 2000, 2, 3365; b) T.K.
Chakraborty, S. Jayaprakash, Tet. Lett.
2000, 42, 497; c) L. Dias, L.G. de Oliveira,
Org. Lett. 2001, 3, 3951.
Acknowledgments
Prof. G Höfle and Prof. H. Reichenbach
(both of GBF, Braunschweig, Germany) for pro-
viding samples of crocacin A (1) and crocacin D
(4), Dr L. May and Mrs F. Dupen (both Syngen-
ta) for glasshouse fungicidal activity tests,
T.E.M. Fraser (Syngenta) for measurements of
physical properties and stabilities, and Miss J.
Phillips (Syngenta) for measurements of inhibi-
tion of mitochondrial beef heart respiration.
[7] a) T.K. Chakraborty, P. Laxman, Tet. Lett.
2002, 43, 2645; b) T.K. Chakraborty, P.
Laxman, Tet. Lett. 2003, 44, 4989.
[8] N. Tischler, M.H. Tischler, Tet. Lett. 1978,
3407.
[9] M. Mitzlaff, K. Warning (HoechstAG) DE
2718666, 1978.
[10] P. Knochel, L. Boymond, M. Rottlander,
G. Cahiez, Angew. Chem., Int. Ed. 1998,
37, 1701.
Received: September 15, 2003
[1] C.R.A. Godfrey, in ‘Agrochemicals from
Natural Products’, Ed. C.R.A. Godfrey,
Marcel Dekker, New York, 1994, p. 312.
[2] B. Kunze, R. Jansen, G. Höfle, H. Rei-
chenbach, J. Antibiot. 1994, 47, 881.
[3] J.M. Clough, C.R.A. Godfrey, J.R. God-
win, R.S.I. Joseph, C. Spinks, Pesticide
Outlook 1996, 4, 16.
[11] A. Baba, T. Miyai, M. Ueba, Synlett 1999,
182.
[12] T. Aoyama, T. Shioiri, Tet. Lett. 1990, 31,
5507.
[4] P.M. Wood, D.W. Hollomon, Pest Man-
agement Science 2003, 59, 499.