Attempted synthesis of 3-hydroxy-2-octadecylindole. Proposed structural revision of previously prepared 3-hydroxy-2-octadecylindole and a proposed structure of fistulosin

Article abstract of DOI:10.1016/j.tet.2009.08.055

A synthetic route to 3-hydroxy-2-octadecylindole, the putative structure of the novel indole alkaloid fistulosin, starting from 2-nitro-1-iodobenzene was examined. Key steps include formation of a 1-stannylsubstituted 1-methoxy-1-alkene from a Fischer chr

Full text of DOI:10.1016/j.tet.2009.08.055

Tetrahedron 65 (2009) 8786–8793
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Attempted synthesis of 3-hydroxy-2-octadecylindole. Proposed structural
revision of previously prepared 3-hydroxy-2-octadecylindole
and a proposed structure of fistulosin
Ronald W. Clawson, Jr., Christopher A. Dacko, Robert E. Deavers, III, Novruz G. Akhmedov,
*
Bjo¨rn C.G. So¨derberg
C. Eugene Bennett Department of Chemistry, West Virginia University, Morgantown, West VA 26506-6045, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 June 2009
Received in revised form 18 August 2009
Accepted 19 August 2009
Available online 26 August 2009
A synthetic route to 3-hydroxy-2-octadecylindole, the putative structure of the novel indole alkaloid fis-
tulosin, starting from 2-nitro-1-iodobenzene was examined. Key steps include formation of a 1-stannyl-
substituted 1-methoxy-1-alkene from a Fischer chromium carbene, intermolecular Kosugi–Migita–Stille
coupling, and a palladium-catalyzed reductive N-heteroannulation. Demethylation of 3-methoxy-2-octa-
decylindole, a possible immediate precursor to 3-hydroxy-2-octadecylindole, was unsuccessful and gave
instead methyl 2-(1-oxononadecanyl)aminobenzoate. The structure of the isolated alkaloid was suggested
to be 2-(1-oxooctadecanyl)aminobenzoate by comparison of analytical data with a synthetic sample. In
addition, oxidation of 2-octadecylindole gave a 2,20-dimer, a compound identical to previously prepared
3-hydroxy-2-octadecylindole.
Keywords:
Palladium-catalyzed
Indoles
Reductive-annulation
Fistulosin
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
reported antifungal activity against a variety of fungi including
Fusarium oxysporum, Fusarium solani, and Penicillium roqueforti
A novel alkaloid named fistulosin has been isolated both from
the leek root (Allium porrum)1 and from the roots of the Welsh
onion (Allium fistulosum)2 by Tomita et al. For example, 51 mg of
fistulosin was obtained from 50 kg of roots of the Welsh onion. The
structure of fistulosin was proposed to be 3-hydroxy-2-octadecy-
lindole (1) based on EI-HRMS, IR, DEPT, 1H–1H COSY, 1H–13C COSY,
and selective INEPT experiments. Fistulosin was represented in the
keto form in this paper (Fig.1). 3-Hydroxyindoles are to date almost
unknown in nature and to our knowledge fistulosin represents the
only example of a fully characterized 2-alkyl-3-hydroxyindole
isolated to date.3 The unique nature of this compound and its
peaked our interest.
A more careful examination of the 1H and 13C NMR data given
revealed some discrepancies including an incorrect number of
hydrogens, the absence of one quarternary carbon if the compound
exists in the enol form, or alternatively the absence of the C2-
methine-proton if it is in the keto form. In addition, the experi-
mental exact mass (HRMS) was reported as m/z 385.6337 for Mþ
and the calculated m/z (for C26H43NO) was reported as m/z
385.6387. The calculated value appears to be in error since the exact
mass for this molecular formula is 385.3345. Considering the large
difference between the experimentally determined mass and the
calculated value (760 ppm), it seemed unlikely that this is the
correct molecular formula.
O
OH
After we initiated our study, a synthesis of 3-hydroxy-2-octa-
decylindole was reported by Nishida et al.4 The final step in their
synthesis was an acid-mediated deprotection of the corresponding
N-tosylated compound 2. In the event, compound 2 was treated
with a large excess of concentrated sulfuric acid at 0 ꢀC for 1.5 h to
give 1-keto (Scheme 1). According to the authors 1-ketowas readily
tautomerized to 1-enol by treatment with, again, a large excess of
sulfuric acid at 0 ꢀC for 1.5 h. Based on the significant difference in
spectral data for the synthetic compounds (1-enol/keto) compared
to the isolated material, the authors concluded that the structure of
fistulosin was not 3-hydroxy-2-octadecylindole. The stability of the
C18H37
C18H37
N
H
N
H
1-enol
1-keto
Figure 1.
* Corresponding author. Tel.: þ13042933435; fax: þ13042934904.
E-mail address: bjorn.soderberg@mail.wvu.edu (B.C.G. So¨derberg).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.08.055

R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
8787
O
O
OH
conc. H2SO4
conc. H2SO4
C18H37
C18H37
C18H37
N
N
Ts
N
H
H
2
1-keto
1-enol
Scheme 1.
synthesized 1-enol is surprising since a previous study of the parent
3-hydroxylindole indicated a very rapid enol/keto tautomeriza-
tion in CDCl3.5 In fact, the enol form was not observed in this solvent
but it is observed in DMSO-d6. 3-Hydroxyindole is also highly un-
stable and rapidly undergoes oxidative dimerization to form
indigo.6
B observed at
d 1.66 ppm and d 1.14 ppm, respectively. Finally, the
carbonyl resonances for all four peronatins were observed at
d
204.3–204.9 ppm.
Considering the NMR similarities between 1-keto/1-enol and
the peronatins, the ease of dimerization of 3-hydroxyindole to in-
digo, and the formation of the dimeric peronantins from indoles
under oxidative conditions, perhaps one or both of the compounds
obtained from N-detosylation of 2 are the corresponding 2,20-di-
mers (Fig. 3). The HRMS data and fragmentation patterns of 1-keto/
1-enol also support this proposal. The reported molecular ions
(Mþ) for 1-enol (m/z 385, 100%) and for 1-keto (m/z 385, 18%) may
actually be the (M/2)þ1þ peak seen from the symmetrical cleavage
of the corresponding dimer. A caveat is the possibility that the
presence of a long aliphatic chain presents sufficient obstruction for
dimerization to occur, however a related oxidative dimerization of
2-tert-butylindole has been reported.10
A small group of 2,20-dimeric indoles consisting of peronatins
A–B (Fig. 2), 7-hydroxy-70-methoxyperonatin B, and 7,70-dime-
thoxyperonatin B has been isolated from damaged fruit bodies of
Collybia peronata and Tricholoma scalpturatum.7 All four peronatins
exhibit very small Mþ ions (2–4%) in the mass spectra but have
much more pronounced M/2þ1þ (100%, EIMS) or M/2þ2þ ions
(CIMS). The structure of peronatins A–B was confirmed by synthesis
via oxidative dimerization of 2,4-dimethylindole (3) affording per-
onatin A under basic conditions and peronatin B under acidic con-
ditions (Scheme 2).8 Peronatin A can be isomerized to peronatin B
under acidic conditions. Although not observed, the authors pos-
tulated the formation of 2,4-dimethyl-3-hydroxyindole (4) as an
intermediate. The sensitivity of 2-alkyl-3-hydroxyindoles was also
observed upon base-mediated deacetylation of N,O-diacetyl 3-hy-
droxy-2-methylindole (5), also producing a 2,20-dimer (Scheme 2).9
O
O
O
O
R R
R R
H+
N
N
N
N
H
H
H
H
O
H
N
R = C18H37
O
H
N
H+
Figure 3.
N
H
N
H
O
O
Our studies into the synthesis of 3-hydroxy-2-octadecylindole
and related compounds are described herein. A revised structure of
the naturally occurring material named fistulosin is also proposed.
Peronatin A
Peronatin B
Figure 2.
R
O
H
N
OH
OAc
HO-
H2O2
N
H
N
H
N
H
N
Ac
O
R
3
4
5
Scheme 2.
The IR, 1H, and 13C NMR data reported by Nishida et al.4 for
the synthetic 1-keto and 1-enol are remarkably similar. The IR-
absorptions for the C]O and C]C–OH are assigned at
1675 cmꢁ1 and 1674 cmꢁ1, respectively. In the 1H NMR spectrum,
the only major differences are the broad singlets reported at
2. Results and discussion
We have recently developed a novel route to 3-alkoxysubstituted
indoles via a palladium-catalyzed reductive N-heteroannulation of
1-alkoxy-1-(2-nitrophenyl)-1-alkenes.11 This methodology was
successfully applied to the first synthesis of the unusual alkaloid
koniamborine.12 A short synthesis of a potential precursor to 3-hy-
droxy-2-octadecylindole employing this annulation methodology
and a Kosugi–Migita–Stille coupling as the key steps was envisioned
(Scheme 3). Cleavage of the O–R0 bond using the appropriate Lewis
acid would then afford 3-hydroxy-2-octadecylindole (fistulosin).
The required tin reagent was initially thought to be derived
from deprotonation of Z-1-ethoxy-1-eicosene (6). This compound
was readily prepared by lithiation of ethyl allyl ether with tert-
butyl lithium followed by alkylation with 1-iodoheptadecane
(Scheme 4).13 Unfortunately, we were unable to transform 6 to 7 by
d
4.70 ppm (1-keto) and 6.08 ppm (1-enol) integrating to one
proton and a significant chemical shift difference between the
first methylene of the side-chains. In addition, the carbon reso-
nances for the C-3 indole carbons were assigned for 1-keto to
d
202.6 ppm and for 1-enol to d 204.6 ppm. The latter C-3 res-
onance is significantly shifted downfield compared to the related
compounds 3-ethoxyindole (140.8 ppm) and 6-bromo-3-acyloxy-
indole (130.2 ppm).
The chemical shifts and general trends for synthetic 1-keto/1-
enol are similar to what is observed for the peronatins. The N–H
resonances for peronatin A and B were found at
6.12 ppm in the 1H NMR spectra. A significant shift difference was
also seen for the methyl groups in the 2-position of peronatin A and
d 4.38 ppm and
d
a
-lithiation and reaction with tributyltin chloride under a variety of
reaction conditions.14

8788
R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
OR'
OR'
C18H37
I
R3Sn
OR'
C18H37
Pd(0), CO
Pd(0)
+
C18H37
N-heteroannulation
coupling
N
H
NO2
NO2
Scheme 3.
of 2,3-cyclopentenindole affording
L
-aza-7,8-benzocyclooctene-2,6-
dione.17,18 Tryptophan is also metabolized in the liver in a similar
OEt
18H37
Bu3Sn
OEt
OEt
1) Base
2) Bu3SnCl
1) tBuLi, -78 °C
2) C17H35I, -40 °C
fashion to give N-formyl kynurenine and kynurenine.
C
C18H37
O
N
OMe
C18H37
SiO2, air
4 h, 87%
6 (88%)
Scheme 4.
7
OMe
H
N
H
O
C18H37
An alternative to a-deprotonation for the formation of 1-alkoxy-
1-stannylalkenes is the reaction of alkoxysubstituted Fischer car-
bene complexes with a trialkyltin triflate developed by McDonald
et al.15 The Fischer carbene complex 8 was prepared in a standard
fashion via metal–halogen exchange of 1-iodononadecane using
t-BuLi followed by reaction with chromium hexacarbonyl and tri-
methyloxonium tetrafluoroborate (Scheme 5). Carbene 8 was sen-
sitive to air and noticeably decomposed upon standing in air for
a few minutes. This complex was treated, immediately after iso-
lation, with tributyltin triflate in the presence of triethylamine to
give 9 in 82% yield as a w3:1 E/Z mixture. Kosugi–Migita–Stille
coupling of 9 with 2-nitro-1-iodobenzene gave the expected
product 10 in excellent yield. With the indole precursor 10 in hand,
it was subjected to the annulation conditions previously used to
prepare a number of 3-alkoxyindoles. Surprisingly, reaction of 10
with carbon monoxide in the presence of a catalytic amount of
bis(dibenzylideneacetone)palladium, bis(1,3-diphenylphosphino)-
propane, and 1,10-phenanthroline monohydrate did not effect
cyclization. However, changing to the simpler catalyst system
consisting of palladium diacetate-1,10-phenanthroline mono-
hydrate smoothly gave indole 11 in 95% yield.
11
12
Scheme 6.
It is interesting to note that a number of long-chained 2-(1-
oxoalkyl)aminobenzoic acids related to 12 have been isolated from
a variety of plant sources. For example, 2-(1-oxoeicosyl)amino-
benzoic acid and 2-(3-hydroxy-1-oxoeicosyl)aminobenzoic acid
were isolated from the aerial parts of Ononis natrix,19,20 2-(1-oxo-13-
docosenyl)aminobenzoic acid from Ononis viscosa subsp. breviflora21
and from Ononis natrix,22 ethyl 2-(1-oxodocosyl)aminobenzoate
from the roots of Gentiana tibetica,23 2-(1-oxodocosyl)aminobenzoic
acid from areal parts of Inula oculus-christi,24 and 2-(1-oxo-C20 to
C24)aminobenzoic acids from Inula japonica.25
The analytical data from isolated fistulosin and 12 are compared
in Table 1. Note again, that the reported data for the isolated ma-
terial has one proton too many. Apart from the methoxy-resonance,
compound 12 exhibits very similar proton and carbon chemical
shifts and multiplicities compared to the isolated compound. Based
on this, we postulated that the natural product may be structurally
closely related to 12.
Bu3Sn
OMe
OMe
Bu3SnOTf
NEt3
1) t-BuLi
2) Cr(CO)6
3) Me3OBF4
Pd(dba)2, PPh3
I
(CO)5Cr
C19H39
I
C19H39
C18H37
NO2
8 (52%)
9 (81%)
OMe
OMe
C18H37
C18H37
Pd(OAc)2, CO (6 atm), DMF
1,10-phenanthroline-H2O
N
NO2
10 (86%)
H
11 (95%)
Scheme 5.
Demethylation of related 3-methoxyindole-2-carboxylic acid es-
ters to 3-hydroxyindole-2-carboxylic acid esters has been reported
using BBr3–CH2Cl2.16 However, attempted cleavage of the methyl
ether in 11 to afford 3-hydroxy-2-octadecylindole (1-enol) under the
same reaction conditions did not produce any identifiable product.
The same result was obtained using avarietyof other commonly used
demethylation reagents including HBr, AlCl3, and TMSI. Upon treat-
ment of 11 with silica gel we were surprised to isolate methyl 2-(1-
oxononadecyl)aminobenzoate (12) in 88% yield (Scheme 6). Similar
oxidations of indoles have beenreported, forexample, auto-oxidation
Using 12 as a possible starting point to determine the true
structure of isolated fistulosin, the corresponding acid, 2-(1-oxo-
nonadecyl)aminobenzoic acid (13), was prepared (Scheme 7). In
the event, nonadecanoyl chloride was prepared from nonadecanoic
acid and thionyl chloride. The acid chloride was treated with
2-aminobenzoic acid in the presence of triethylamine to give 13.
Analytical data for this compound are seen in Table 1. The acid
proton (CO2H) in 13 was not observed in deuterated chloroform.
The NMR data for the compound isolated from the Welsh onion and
the leek root and the synthetic compound 13 were compared. Two

R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
8789
Table 1
main differences were observed, a quaternary carbon signal at
172.7 ppm in the 13C NMR spectrum for 13 absent in the isolated
material and a two proton multiplet (
1.64 ppm) in the 1H NMR
spectrum of the naturally occurring compound missing in the
spectrum of 13. We suggest that the signal at
1.64 ppm in the 1H
NMR spectrum is either water or some other impurity in the
sample. It is also possible that the authors did not observe the
carbonyl carbon at d 172.7 ppm due to its low intensity. In any case,
Analytical data for isolated fistulosin, 12, 13, and 14
d
d
Fistulosin
80–83 ꢀC
12
13
94–95 ꢀ
14
81–82 ꢀ
C
Mp
IR
d
C
d
1684
1709, 1694
1673
1673
1H
1H
1H
1H
1H
3H
2H
2H
2H
2H
10.8 (br s)
8.8 (d)
8.1 (d)
7.6 (t-like)
7.1 (t-like)
11.1 (br s)
8.74 (d)
8.02 (d)
7.54 (t)
7.07 (t)
3.93 (s)
2.44 (t)
1.75 (m)
11.0 (br s)
8.77 (d)
8.13 (dd)
7.60 (dt)
7.11 (dt)
10.9 (br s)
8.77 (d)
8.13 (dd)
7.60 (dt)
7.11 (dt)
the similarity between the remainder of the NMR data is too pro-
found not to suggest at least a close structural relationship. How-
ever, the two compounds were not identical since the melting
points of the synthetic 13 (94–95 ꢀC) and fistulosin (80–83 ꢀC) were
clearly different.
Tomita et al. based their proposed molecular formula of
C26H43NO on a m/z of 385 obtained in FD-MS. Possibly based on this
low-resolution data, a smaller area around this mass was examined
for HRMS and the Mþ was reported to be m/z 385.6337. As was
previously pointed out, the calculated m/z for C26H43NO is 385.3345
and this value is significantly different from the one determined
experimentally.
Long-chain 2-(1-oxoalkyl)aminobenzoic acids have been
shown to readily lose water when subjected to MS analysis. For
example, in the mass spectrum of 2-(1-oxodocosyl)aminobenzoic
acid24 the molecular ion (C29H49NO3) found at m/z 459.371 had
a 1% relative intensity but the fragment at m/z 441.359 (–H2O) had
a relative intensity of 12%. The same trend was observed for 2-(1-
oxoeicosyl)aminobenzoic acid19 with a molecular ion m/z 431 (2%)
and the peak from the loss of H2O at m/z 413 (7%). Based on these
observations, perhaps the m/z 385 reported by Tomita et al. was
the result of loss of water from a compound with the molecular
weight of 403. Interestingly but perhaps a serendipity, 2-(1-
oxooctadecanyl)aminobenzoic acid (14) has a molecular weight of
403. In order to compare analytical data, 2-(1-oxooctadecanyl)-
aminobenzoic acid (14) was prepared from octadecanoyl chloride
and 2-aminobenzoic acid as described for 13 (Scheme 7).The NMR
data for 14 are, not surprising, almost identical to 13 and very
similar to the natural product. In addition, the melting point ranges
were the same. Based on the close similarity between the data
presented in Table 1, we propose that the compound named fistu-
losin is actually 2-(1-oxooctadecanyl)aminobenzoic acid (14).
However, we have unfortunately not been able to obtain a sample of
fistulosin or copies of the original spectra of the natural product for
final comparison.
2.44 (t-like)
1.74 (m)
1.63 (m)
1.4 (m)
1.25
(br s, 28H)
0.88 (t-like)
2.47 (t)
1.77 (pent)
2.47 (t)
1.77 (m)
1.40 (pent)
1.34–1.24
(m, 28H)
0.88 (t)
1.40 (m)
1.34–1.24
(m, 26H)
0.88 (t)
1.37–1.15
(m, 30H)
0.88 (t)
3H
13C
172.3
168.7
141.7
134.7
130.8
122.2
120.3
114.7
52.3
38.7
31.9
29.2–29.7
25.5
22.7
172.7
172.1
142.1
135.6
131.7
122.2
120.6
113.9
172.7
172.2
142.2
135.7
131.8
122.6
120.6
113.8
171.5
142.5
135.8
131.9
122.7
120.7
113.7
38.9
32.1
29.0–29.9
25.9
23.4
38.7
31.9
29.2–29.9
25.5
22.7
38.7
31.9
29.2–29.7
25.5
22.7
14.3
14.1
14.1
14.1
All NMR data were obtained in CDCl3 at 500/125 MHz for fistulosin and at 600/
150 MHz for 12, 13, and 14.
O
1) SOCl2
2) Et3N,
OH
H
RCO2H
CO2H
NH2
N
O
R
13 (R = C18H35, 81%)
14 (R = C17H35, 86%)
Scheme 7.
R=NH2, KOt-Bu
C18H37
C18H37
Et3SiH
I
SiEt3
PdCl2(PPh3)2
Pd(dba)2, dppp
C18H37
NEt3, CuI
C18H37
PtO2
R=NO2
1,10-phen., CO
N
H
NO2
R
R
16 (R=NH2, 72%)
17 (R=NO2, 95%)
18 (95%)
15 (70%)
O
H
N
O
C18H37
C18H37
3-CPBA
DCM
C18H37
N
H
N
H
C18H37
O
N
H
19 (33%)
20 (36%)
Scheme 8.

8790
R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
We next decided to examine if the structure of synthetic 1-enol
of Chemical Engineering, College of Engineering and Mineral Re-
sources, West Virginia University.
was in fact the dimer shown in Figure 3. It was expected that
oxidative dimerization of 2-octadecylindole (15) would furnish the
2,20-dimer. Direct preparation of 2-octadecylindole (15) from
2-methylindole using Katritsky’s alkylation methodology26 failed to
give more than minor amounts of the expected product. Base-
mediated cyclizations of 2-(1-alkyn-1-yl)benzenamines has been
used in several cases to prepare 2-substituted indoles.27 However,
cyclization of 16 prepared via a Sonogashira coupling of 2-iodo-
benzeamine with 1-eicosyne did not give 15 upon reaction with t-
BuOK (Scheme 8). Indole 15 was instead prepared in a three-step
sequence. In the event, Sonogashira cross-coupling of 2-iodo-1-
nitrobenzene with 1-eicosyne gave the expected product 17. This
compound was hydrosilylated using triethylsilane in the presence of
platinum dioxide as the catalyst.28 The hydrosilylated product 18 was
then submitted to the N-heteroannulation conditions and we were
pleased to notice that not only did the cyclization occur, the silyl
group was also lost to give 15. It should be noted that this type of
substrate has not previously been examined as a substrate for the
reductive N-heteroannulation. Oxidative dimerization of 15 with
hydrogen peroxide, according to Van Vranken et al.,8 under basic or
acidic conditions gave complex reaction mixtures that proved to be
problematic to purify. However, simply changing the oxidant to
3-CPBA afforded a readily separated mixture of the 2,20-dimer 19 and
the corresponding 2,30-coupling product 20.10 Analytical data for the
2,20-dimer 19 were in all respects (NMR, IR, mp) identical to the data
for 1-enol prepared by Tomita et al. As was expected from HRMS
(APCI) analysis of 19, a very small MþHþ peak (relative intensity 1%)
was observed. Two much more pronounced peaks were observed for
M/2þ2Hþ (relative intensity 45%) and M/2þ (relative intensity 100%).
This is in concert with the data reported for the peronatins.7 We were
only able to prepare one of the diastereomeric 2,20-dimers thus, the
question whether or not the structure of synthetic 1-keto is correct
or if it is the missing 2,20-dimer remains unclear.
4.1.1. (Z)-1-Ethoxy-1-eicosene (6). Allyl ethyl ether (620 mL, 5.46
mmol) was stirred in THF (20 mL) at ꢁ78 ꢀC under a positive flow of
nitrogen. t-BuLi (14.4 mL, 1.7 M in pentane, 24.4 mmol) was slowly
added to the solution over 15 min. The resulting yellow solution was
stirred for an additional 30 min at ꢁ78 ꢀC before warming to ꢁ40 ꢀC
over 4 h. 1-Iodoheptadecane29 (2.00 g, 5.46 mmol) in THF (40 mL)
was quickly added via a cannula to the reaction mixture. The
resulting pale yellow solution was allowed to stir for an additional
30 min at ꢁ40 ꢀC before warming to ambient temperature over 3 h.
Water (2 mL) was carefully added to quench the reaction before
extracting with EtOAc (3ꢂ100 mL). The combined organic phases
were washed with brine (3ꢂ20 mL), dried (MgSO4), and filtered. The
solvents were removed under reduced pressure to give a white oily
reside that was purified by chromatography (hexanes) affording 6
(1.51 g, 4.65 mmol, 88%) as a colorless oil. 1H NMR (270 MHz)
d 5.91
(dt, J¼6.3, 1.4 Hz, 1H), 4.32 (q, J¼7.3 Hz, 1H), 3.76 (q, J¼7.1 Hz, 2H),
2.05 (m, 2H), 1.26 (br s, 32H), 0.88 (m, 6H); 13C NMR (67.5 MHz)
d
144.4 (ꢁ), 107.2 (ꢁ), 67.4 (þ), 32.0 (þ), 29.9 (þ), 29.7 (multiple
carbons, þ), 29.5 (þ), 29.4 (þ), 29.3 (þ), 23.9 (þ), 22.7 (þ), 15.3 (ꢁ),
14.1 (ꢁ); IR 1665, 1111 cmꢁ1. Anal. Calcd for C22H44O: C, 81.41; H,
13.66. Found: C, 81.39; H, 13.42.
4.1.2. 1-Iodononadecane30. A heterogeneous mixture of 1-non-
adecanol (10.00 g, 35.15 mmol) and triphenyl phosphite (37.0 mL,
140.6 mmol) was stirred in hexanes (80 mL) at 0 ꢀC for 20 min. Io-
dine (35.68 g, 140.6 mmol) was added in approximately three equal
portions over 30 min. The resulting black slurry was stirred for an
additional 1 h at 0 ꢀC. The reaction mixture was returned to ambient
temperature and stirred for 12 h. Water (300 mL) was carefully
added before extracting with hexanes (3ꢂ200 mL). The combined
organic phases were washed with a saturated solution of NaHSO3
(aqueous, 3ꢂ100 mL), dried (MgSO4), and filtered. Removal of the
solvent afforded a brown oil that was immediately purified by
chromatography (hexanes) to afford 1-iodononadecane as a white
solid (11.53 g, 29.24 mmol, 83%). Physical data (mp) and spectral data
(1H NMR) were in complete accordance with literature values.31
3. Conclusion
Evidence suggesting the true structure of fistulosin, isolated
from the leek root and the Welsh onion, to be 2-(1-oxooctadecyl-
amino)benzoic acid was presented. In addition, the structure of
synthetic 3-hydroxy-2-octadecylindole was shown to be the cor-
responding 2,20-dimer.
4.1.3. [(Methoxy)(nonadecyl)carbene]pentacarbonylchromium
(5). 1-Iodononadecane (11.53 g, 29.24 mmol) was dissolved in Et2O
(150 mL) under N2 in a 500-mL flask. The flask was cooled to ꢁ20 ꢀC
whereupon the iodide precipitated. t-BuLi (37.8 mL, 1.7 M in pen-
tane, 64.3 mmol) was quickly added via a cannula. The resulting
cloudy yellow solution was stirred for an additional 30 min at
ꢁ20 ꢀC before slowly warming to room temperature over 1 h. The
solution was stirred for an additional hour before rapid transfer via
a cannula to a 500-mL flask containing Cr(CO)6 (6.43 g, 29.2 mmol)
in Et2O (50 mL). The resulting dark brown solution was stirred
(18 h) where after the solvent was removed under reduced pres-
sure. Water (80 mL) was added to the brown residue and Me3OBF4
was added in portions until the solution was acidic (pH¼2). The
resulting yellow slurry was extracted with hexanes (3ꢂ200 mL)
and the combined organic phases were dried (MgSO4) and filtered.
Removal of the solvent afforded a yellow oily-solid that was im-
mediately purified by chromatography (hexanes) to afford 5 as
a bright yellow solid (7.57 g, 15.1 mmol, 52%).32 1H NMR (270 MHz)
4. Experimental section
4.1. General procedures
The chemical shifts are expressed in
d values relative to SiMe4
(0.0 ppm, 1H and 13C) or CDCl3 (77.0 ppm, 13C) internal standards.
Results of APT (Attached Proton Test) 13C NMR experiments are
shown in parentheses, where relative to CDCl3, (ꢁ) denotes CH3 or
CH and (þ) denotes CH2 or C.
Anhydrous benzene, N-methylpyrrolidinone, and N,N-dime-
thylformamide were used as received. Tetrahydrofuran (THF) and
diethyl ether were distilled from sodium benzophenone ketyl prior
to use. Hexanes, ethyl acetate, triethylamine, and dichloromethane
were distilled from calcium hydride. Chemicals prepared according
to literature procedures have been footnoted the first time used; all
other reagents were obtained from commercial sources and used as
received. All reactions were performed under a nitrogen atmos-
phere in oven-dried glassware unless otherwise stated. Solvents
were removed from reaction mixtures and products on a rotary
evaporator at water aspirator pressure. Chromatography was per-
d
4.76 (s, 3H), 3.29 (t, J¼7.7 Hz, 2H), 1.50 (m, 2H), 1.25 (br s, 32H),
0.88 (t, J¼6.9 Hz, 3H); 13C NMR (67.5 MHz)
d
363.7 (þ), 223.1 (þ),
216.4 (þ), 67.6 (ꢁ), 63.2 (þ), 32.0 (þ), 29.7 (multiple carbons, þ),
29.4 (þ), 29.3 (þ), 26.4 (þ), 22.7 (þ), 14.1 (ꢁ); IR 2062, 1925 cmꢁ1
.
formed on silica gel (35–75
Melting points were determined on a MelTemp and are un-
corrected. Elemental analyses were performed in the Department
m
m).
4.1.4. Tributyl (1-methoxyicos-1-enyl)stannane (6). Following puri-
fication, the carbene (5.31 g, 10.6 mmol) was immediately dissolved
in a solution of NEt3 (10 mL) in Et2O (30 mL). Bu3SnOTf (6.96 g,

R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
8791
15.9 mmol) was added and the resulting brown solution was stirred
for 4 days under a nitrogen atmosphere. The solvent and excess NEt3
were removed under reduced pressure and the brown oily residue
was purified by chromatography (hexanes/NEt3, 99:1) to afford 9
(5.17 g, 8.63 mmol, 81%) as a colorless oil. Spectral data from a 3:1 E/Z
purifying the product by chromatography (hexanes/EtOAc, 95:5)
affording 12 (98 mg, 0.23 mmol, 87%) as an oily yellow solid.1H NMR
(270 MHz)
d
11.1 (br s, 1H), 8.74 (d, J¼8.7 Hz, 1H), 8.02 (d, J¼8.2 Hz,
1H), 7.54 (t, J¼7.2 Hz, 1H), 7.07 (t, J¼7.2 Hz, 1H), 3.93 (s, 3H), 2.44 (t,
J¼7.4 Hz, 2H), 1.75 (m, 2H), 1.37–1.15 (m, 30H), 0.88 (t, J¼6.9 Hz, 3H);
mixture: E-isomer: 1H NMR (270 MHz)
(s, 3H), 2.18 (m, 2H), 1.90 (m, 2H), 1.25–1.60 (m, 44H), 0.99–0.83 (m,
18H); 13C NMR
165.6 (þ),112.1 (ꢁ), 54.8 (ꢁ), 32.0 (þ), 29.9 (multiple
d
5.23 (t, J¼7.5 Hz, 1H), 3.45
13C NMR (67.5 MHz)
d
172.3 (þ), 168.7 (þ), 141.7 (þ), 134.7 (ꢁ), 130.8
(ꢁ), 122.2 (ꢁ), 120.3 (ꢁ), 114.7 (þ), 52.3 (ꢁ), 38.7 (þ), 31.9 (þ), 29.7
(multiple carbons, þ), 29.5 (þ), 29.4 (þ), 29.2 (þ), 25.5 (þ), 22.7 (þ),
14.1 (ꢁ); IR 3266, 1709, 1694 cmꢁ1. Anal. Calcd for C27H45NO3: C,
75.13; H,10.51; N, 3.24. Found: C, 75.34; H,10.74; N, 2.89. HRMS (ESI)
calcd for C27H46NO3 (MþþH) 432.3472, found 432.3471.
d
carbons, þ), 29.7 (þ), 29.4 (þ), 29.1 (þ), 27.7 (þ), 22.7 (þ), 14.1 (ꢁ),
13.7 (ꢁ), 10.3 (þ); partial data for the Z-isomer: 1H NMR
d 4.54 (t,
J¼6.9 Hz,1H), 3.53 (s, 3H); 13C NMR (67.5 MHz)
163.2 (þ),123.8 (ꢁ),
d
59.4 (ꢁ), 27.8 (þ), 27.1 (þ), 25.1 (þ), 18.5 (þ), 10.9 (þ); IR 2921,
1463 cmꢁ1. Anal. Calcd for C33H68OSn: C, 66.10; H, 11.43. Found: C,
66.39; H, 11.79.
4.1.8. 2-(1-Oxononadecyl)aminobenzoic acid (13). A mixtureof non-
adecanoic acid (1.00 g, 3.35 mmol) and thionyl chloride (340 mL,
4.69 mmol) was heated at reflux for 8 h. Excess thionyl chloride was
removed under reduced pressure to give nonadecanoyl chloride as
an orange oil that was used without further purification. 2-Amino-
4.1.5. 1-(1-Methoxyicos-1-enyl)-2-nitrobenzene (10). A solution of
2-iodo-nitrobenzene (131 mg, 0.525 mmol), bis(dibenzylideneace-
tone)palladium (15 mg, 0.026 mmol), and triphenylphosphine
(28 mg, 0.105 mmol) was stirred in toluene (20 mL) for 5 min under
apositive flowofnitrogen. Totheyellowsolutionwasaddedstannane
9 (346 mg, 0.577 mmol) dissolved in toluene (5 mL). The yellow so-
lution was heated at reflux (68 h) whereupon a dark brown solution
was formed. The reaction was monitored using TLC (hexanes). The
solvent was removed under reduced pressure and the resulting black
viscous oil was purified by chromatography (hexanes/EtOAc, 95:5),
affording 10 (195 mg, 0.451 mmol, 86%) as a faint yellow oil. Spectral
benzoic acid (450 mg, 3.28 mmol) and NEt3 (500 mL, 3.61 mmol)
were stirred in CH2Cl2 (20 mL) for 5 min. A solution of freshly pre-
pared nonadecanoyl chloride (1.04 g, 3.28 mmol) in CH2Cl2 (30 mL)
was added slowlyin approximatelyfive equal portions. The resulting
solution was stirred for 5 h, water (100 mL) was added and the
stirring was continued 5 min. The phases were separated and the
aqueous portion was extracted with CH2Cl2 (3ꢂ15 mL). The com-
bined organic phases were dried (MgSO4), filtered, and the solvent
was removed. The crude solid was recrystallized from hexanes to
afford 13 (1.11 g, 2.67 mmol, 81%) as a white powdery solid. Mp: 94–
data from a 4:1 E/Z mixture: E-isomer: 1H NMR (270 MHz)
d 7.91 (dd,
J¼8.1, 1.4 Hz, 1H), 7.58 (dt, J¼7.3, 1.4 Hz, 1H), 7.49 (dd, J¼7.9, 1.6 Hz,
1H), 7.42 (dt, J¼7.5,1.6 Hz,1H), 4.82 (t, J¼7.6 Hz,1H), 3.62 (s, 3H),1.90
(q, J¼7.3, 2H), 1.40–1.15 (br m, 32H), 0.88 (t, J¼6.9 Hz, 3H); 13C NMR
95 ꢀC.1H NMR (600 MHz)
d
11.0 (br s,1H), 8.77 (d, J¼8.4 Hz,1H), 8.13
(dd, J¼7.8, 1.8 Hz, 1H), 7.60 (dt, J¼7.2, 1.8 Hz, 1H), 7.11 (dt, J¼7.8,
0.6 Hz, 1H), 2.47 (t, J¼7.8 Hz, 2H), 1.77 (p, J¼7.8 Hz, 2H), 1.40 (p,
J¼7.8 Hz, 2H), 1.34–1.24 (m, 28H), 0.88 (t, J¼7.2 Hz, 3H); 13C NMR
(67.5 MHz)
d
151.5 (þ),148.9 (þ),132.3 (ꢁ),131.9 (ꢁ),131.6 (þ),129.0
(ꢁ),124.2 (ꢁ),101.9 (ꢁ), 55.9 (ꢁ), 31.9 (þ), 29.7 (multiple carbons, þ),
29.6(þ), 29.4(þ),29.3(þ), 29.0(þ),25.3(þ), 22.7(þ),14.1(ꢁ);partial
(150 MHz) d 172.7, 172.1, 142.1, 135.6, 131.7, 122.6, 120.6, 113.9, 38.7,
31.9, 29.7, 29.6, 29.5, 29.3, 29.2, 25.5, 22.7, 14.1; IR 3336, 2918, 2850,
1673,1270,1165, 754 cmꢁ1. HRMS (ESI) calcd for C26H44NO3 (MþþH)
418.3321, found 418.3321.
data fortheZ-isomer: 1H NMR (270 MHz)
d
7.70 (dd, J¼8.1,1.1 Hz,1H),
4.97 (t, J¼7.5 Hz, 1H), 3.43 (s, 3H); 13C NMR (67.5 MHz)
d
150.2 (þ),
148.9 (þ),131.7 (ꢁ),130.9 (ꢁ),128.7 (ꢁ),123.5 (ꢁ),116.6 (ꢁ), 58.4 (ꢁ);
IR 2916, 1533 cmꢁ1. Anal. Calcd for C27H45NO3: C, 75.13; H, 10.51; N,
3.24. Found: C, 75.51; H, 10.71; N, 3.56.
4.1.9. 2-(1-Oxooctadecyl)aminobenzoic acid (14). Octadecanoic acid
(1.00 g, 3.52 mmol) and thionyl chloride (360 mL, 4.92 mmol) were
heated at reflux for 8 h. Excess thionyl chloride was removed under
reduced pressure to give octadecanoyl chloride as an orange oil that
was used without further purification. 2-Aminobenzoic acid
4.1.6. 3-Methoxy-2-octadecylindole (11). Compound 10 (445 mg,
1.03 mmol), palladium diacetate (14 mg, 0.062 mmol), and 1,10-
phenanthroline monohydrate (25.0 mg, 0.124 mmol) were dissolved
in DMF (2 mL) in a threaded ACE glass pressure tube. The tube was
fitted with a pressure head, and the solution was saturated with
carbon monoxide (four cycles to 6 atm of CO). The reaction mixture
was heated (120 ꢀC, 6 atm CO, 82 h). A majority of the solvent was
removed under vacuum and the brown residue was purified by
chromatography (hexanes/EtOAc/Et3N, 95:4:1) affording 11 (391 mg,
(461 mg, 3.36 mmol) and NEt3 (520 mL, 3.70 mmol) were stirred in
CH2Cl2 (30 mL) for 5 min. A solution of freshly prepared octade-
canoyl chloride (1.07 g, 3.70 mmol) in CH2Cl2 (20 mL) was added
slowly in approximately five equal portions. The resulting solution
was stirred for 5 h, water (100 mL) was added and the stirring was
continued for 5 min. The phases were separated and the aqueous
portion was extracted with CH2Cl2 (3ꢂ15 mL). The combined or-
ganic phases were dried (MgSO4), filtered, and the solvent was
removed. The crude solid was recrystallized from hexanes to afford
14 (1.17 g, 2.89 mmol, 86%) as a white solid. Mp: 81–82 ꢀC. 1H NMR
0.978 mmol, 95%) as an oily yellow solid.1H NMR (270 MHz)
d 7.58 (d,
J¼6.9 Hz, 1H), 7.44 (br s,1H), 7.24 (dd, J¼6.9, 2.2 Hz,1H), 7.11 (t, J¼7.1,
1.4 HZ, 1H), 7.06 (dt, J¼7.1, 1.4 Hz, 1H), 3.92 (s, 3H), 2.74 (t, J¼7.5 Hz,
2H), 1.66 (p, J¼6.9 Hz, 2H), 1.4–1.2 (br s, 30H), 0.88 (t, J¼6.9 Hz, 3H);
(600 MHz)
d
10.9 (br s, 1H), 8.77 (d, J¼8.4 Hz, 1H), 8.13 (dd, J¼7.8,
13C NMR (67.5 MHz)
d
135.9 (þ), 132.8 (þ), 127.0 (þ), 121.6 (þ), 121.1
1.8 Hz, 1H), 7.60 (dt, J¼7.2, 1.8 Hz, 1H), 7.11 (dt, J¼7.8, 0.6 Hz, 1H),
(ꢁ), 119.0 (ꢁ), 117.0 (ꢁ), 110.8 (ꢁ), 62.1 (ꢁ), 31.9 (þ), 29.7 (multiple
carbons, þ), 29.6 (þ), 29.4 (þ), 29.4 (þ), 29.4 (þ), 24.7 (þ), 22.6 (þ),
14.1 (ꢁ); IR 3477, 2920, 909, 731 cmꢁ1. Anal. Calcd for C27H45NO: C,
81.14; H, 11.35; N, 3.50. Found: C, 81.24; H, 11.50; N, 3.56. HRMS (ESI)
calcd for C27H46NO (MþHþ): 400.3574. Found: 400.3572.
2.47 (t, J¼7.8 Hz, 2H), 1.77 (m, 2H), 1.40 (m, 2H), 1.34–1.24 (m, 24H),
0.88 (t, J¼7.2 Hz, 3H); 13C NMR (150 MHz)
d 172.7, 172.2, 142.2,
135.7, 131.8, 122.6, 120.6, 113.8, 38.7, 31.9, 29.7, 29.6, 29.5, 29.3, 29.2,
25.5, 22.7, 14.1; IR 3336, 2918, 2850, 1673, 1270, 1165, 754 cmꢁ1
.
HRMS (ESI) calcd for C25H42NO3 (MþHþ) 404.3165, found 404.3161.
MS/MS (ESI) found 404.4 and 386.2.
4.1.7. Methyl 2-(1-oxononadecyl)aminobenzoate (12). 3-Methoxy-
2-octadecylindole (11) (104 mg, 0.260 mmol) and silica gel (ap-
proximately 500 mg) were stirred in reagent grade methanol
(10 mL) in a round-bottom flask on a rotary evaporator for 5 min. The
solvent was then removed at elevated temperature (boiling water
bath) and reduced pressure. The mixture was allowed to stand open
to the atmosphere at ambient temperature overnight before
4.1.10. 2-(1-Icosy-1-nyl)benzenamine (16). To a solution of 2-iodo-
benzenamine (1.00 g, 4.57 mmol) in NEt3 (10 mL) at ambient
temperature was added PdCl2 (41 mg, 0.228 mmol) and PPh3
(120 mg, 0.457 mmol). The resultant suspension was stirred for
10 min followed by addition of 1-eicosyne (2.12 g, 7.62 mmol) and
then CuI (44 mg, 0.228 mmol). This mixture was stirred for 24 h

8792
R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
under an N2 atmosphere at ambient temperature, followed by re-
moval of the solvent under reduced pressure. The resulting crude
product was purified by chromatography (hexanes/EtOAc, 95:5) to
afford 16 (1.21 g, 3.27 mmol, 72%) as a brown solid. Mp 39–41 ꢀC;
22.7, 14.1 (resonances overlap); IR (ATR) 3375, 2915, 2849, 1471,
746, 1456, cmꢁ1; HRMS (ESI) calcd for C26H44N (MþHþ) 370.3474,
found 370.3470.
1H NMR
d
7.23 (dd, J¼7.2,1.2 Hz,1H), 7.06 (dt, J¼7.8,1.2 Hz,1H), 6.67
4.1.14. 2-Octadecyl-2-(2-octadecyl-3-oxo-2,3-dihydro-1H-indol-2-yl)-
1,2-dihydro-3H-indol-3-one (19) and 2-octadecyl-2-(2-octadecyl-1H-
indol-3-yl)indolin-3-one (20). A solution of 3-chloroperoxybenzoic
acid (248 mg, 1.01 mmol) in dichloromethane (5 mL) was added
dropwise at ambient temperature to a stirred solution of 2-octa-
decyl-1H-indole (15) (310 mg, 0.838 mmol) in the same solvent
(2 mL). The mixture was stirred for 1 h, then poured into 5% NH4Cl
(15 mL) and extracted with CH2Cl2 (3ꢂ25 mL). The combined or-
ganic phases were dried (MgSO4), filtered, and the solvent was
removed under reduced pressure. The resulting crude product was
purified by chromatography (hexanes, then hexanes/EtOAc 98:2) to
afford 19 (107 mg, 0.139 mmol, 33 %) as a bright yellow solid and 20
(113 mg, 0.150 mmol, 36%) as a dark yellow solid. Spectral data for
(d, J¼8.4 Hz, 1H), 6.65 (dt, J¼7.2, 1.2 Hz, 1H), 4.15 (br s, 2H), 2.46 (t,
J¼7.2 Hz, 2H), 1.62 (pent, J¼7.2 Hz, 2H), 1.45 (pent, J¼7.2 Hz, 2H),
1.31–1.26 (br m, 28H), 0.88 (t, J¼7.2 Hz, 3H); 13C NMR
d 147.6, 132.0,
128.8, 117.8, 114.1, 109.0, 95.8, 31.9, 29.69, 29.66, 29.65, 29.63, 29.5,
29.4, 29.2, 29.0, 22.7, 19.6, 14.1 (resonances overlap, one resonance
missing); IR (ATR) 3490, 3467, 3390, 3371, 2915, 2847, 1607, 1467,
753, 719 cmꢁ1; HRMS (ESI) calcd for C26H44N (MþHþ) 370.3474,
found 370.3475.
4.1.11. 1-(1-Icosyn-1-yl)-2-nitrobenzene (17). To a solution of 2-
iodo-nitrobenzene (2.68 g, 10.8 mmol) in NEt3 (30 mL) at ambient
temperature was added Pd(PPh3)2Cl2 (377 mg, 0.537 mmol). The
resultant suspension was stirred for 10 min followed by addition of
1-eicosyne (5.00 g, 18.0 mmol) followed by CuI (102 mg,
0.537 mmol). This mixture was then stirred for 3 h under an N2 at-
mosphere at ambient temperature, followed by removal of all vola-
tiles under reduced pressure. The resulting crude product was
purified by column chromatography (hexanes/EtOAc, 97:3) to afford
17 (4.12 g, 10.3 mmol, 95%) as a brown solid. Mp 39–42 ꢀC; 1H NMR
19: mp 105–107 ꢀC; 1H NMR (600 MHz)
d
7.56 (d, J¼7.8 Hz, 2H), 7.48
(t, J¼8.4 Hz, 2H), 6.95 (d, J¼8.4 Hz, 2H), 6.78 (t, J¼6.6 Hz, 2H), 6.02
(s, 2H), 1.88 (m, 2H), 1.30–0.94 (br m, 66H), 0.88 (t, J¼6.6 Hz, 6H);
13C NMR (150 MHz)
d 204.5, 162.0, 138.0, 124.1, 121.6, 118.3, 111.9,
72.5, 29.8, 29.7, 29.7, 29.7, 29.6, 29.6, 29.5, 29.4, 29.4, 29.4, 29.3,
22.9, 22.7, 14.1 (several resonances overlap); IR (ATR) 3358, 2916,
2849, 1674, 1613, 741, cmꢁ1; HRMS (APCI) calcd for C52H85N2O2
(MþHþ, 1%) 769.6611, found 769.6613; C26H44NO (M/2þ2Hþ, 45%)
386.3423, found 386.3422; C26H42NO (M/2þ, 100%) 384.3266,
found 384.3265;
(600 MHz)
d
7.96 (dd, J¼8.4, 0.6 Hz, 1H), 7.57 (dd, J¼7.8, 1.2 Hz, 1H),
7.51 (dt, J¼7.2,1.2 Hz,1H), 7.38 (dt, J¼7.2,1.2 Hz,1H), 2.47 (t, J¼7.2 Hz,
2H), 1.63 (pent, J¼7.2 Hz, 2H), 1.46 (pent, J¼7.2 Hz, 2H), 1.22–1.36 (br
m, 28H), 0.88 (t, J¼7.2 Hz, 3H); 13C NMR (150 MHz)
d 150.1, 134.7,
132.4,127.7,124.3,119.4, 99.5, 75.9, 31.9, 29.7 (multiplecarbons), 29.6,
29.1, 28.3, 22.7,19.8,14.1; IR (ATR) 2917, 2848,1514,1339 cmꢁ1; HRMS
(ESI) calcd for C26H41NNaO2 (MþNaþ) 422.3035, found 422.3030.
4.1.15. Spectral data for 20. Mp 49–52 ꢀC; 1H NMR (600 MHz)
d 7.81
(br s, 1H), 7.77 (d, J¼8.4 Hz, 1H), 7.63 (d, J¼7.8 Hz, 1H), 7.46 (t,
J¼8.4 Hz, 1H), 7.24 (d, J¼8.4 Hz, 1H), 7.09 (t, J¼7.2 Hz, 1H), 7.03 (t,
J¼7.8 Hz,1H), 6.87 (d, J¼8.4 Hz,1H), 6.82 (t, J¼7.2,1H), 5.03 (br s,1H),
2.81 (m, 2H), 2.48 (dt,J¼12.6, 3.6 Hz, 1H), 2.23 (dt, J¼12.0, 4.2 Hz,
1H), 1.54 (m, 4H), 1.31–1.21 (br m, 60H), 0.88 (t, J¼6.6 Hz, 6H); 13C
4.1.12. Triethyl((E)-1-(2-nitrophenyl)icos-1-enyl)silane (18). PtO2
(278 mg, 1.23 mmol) and 17 (4.90 g, 12.26 mmol) were placed under
a N2 atmosphere. Triethylsilane (2.85 g, 24.5 mmol) was introduced
via syringe and the mixture was stirred at 60 ꢀC in an oil bath (3 h).
The resulting crude product was purified by chromatography (hex-
anes/EtOAc, 98:2) to afford 18 (5.37 g,10.4 mmol, 85%) as a yellow oil.
NMR (150 MHz) d 203.3, 160.1, 137.1, 136.8, 135.2, 133.5, 128.4, 127.3,
125.0, 121.2, 120.6, 119.6, 118.8, 112.1, 110.4, 108.9, 70.9, 38.1, 31.9,
30.8, 29.9, 29.72, 29.71, 29.70, 29.68, 29.66, 29.65, 29.64, 29.61,
29.56, 29.47, 29.41, 29.36, 28.4, 23.9, 22.7, 14.1 (several resonances
overlap); IR (ATR) 2917, 2850,1611,1486, 741, 718, cmꢁ1; HRMS (ESI)
calcd for C52H85N2O (MþHþ) 753.6656, found 753.6652.
1H NMR
d
7.90 (dd, J¼8.4, 1.2 Hz, 1H), 7.48 (dt, J¼7.8, 1.2 Hz, 1H), 7.30
(dt, J¼7.8,1.8 Hz,1H), 7.01(dd, J¼7.8,1.2 Hz,1H), 5.95 (t, J¼7.2 Hz,1H),
1.85–1.74 (m, 2H), 1.32–1.12 (m, 32H), 0.90–0.84 (m, 12H), 0.61–0.50
(m, 6H); 13C NMR
d
148.1,143.5,138.9,137.3,132.3,130.5,126.3,124.2,
Acknowledgements
31.9, 30.8, 29.7, 29.6, 29.5, 29.4, 29.1, 29.0, 22.7, 14.1, 7.2, 3.4 (reso-
nances overlap); IR (ATR) 2921, 2852, 1525, 1347, 718, 705 cmꢁ1
;
This work was supported by a research grant from the National
Science Foundation (CHE 0611096). NSF-EPSCoR (Grant #1002165R)
is gratefully acknowledged for the funding of a 600 MHz Varian
Inova NMR and a Thermo-Finnigan LTQ-FT Mass Spectrometer and
the NMR and MS facilities in the C. Eugene Bennett Department of
Chemistry at West Virginia University.
HRMS (ESI) calcd for C32H57NNaO2Si (MþNaþ) 538.4056, found
538.4050.
4.1.13. 2-Octadecyl-1H-indole (15). To a solution of 18 (2.03 g,
3.93 mmol), Pd(dba)2 (135 mg, 0.236 mmol), 1,3-bis-(diphenyl-
phosphino)propane (dppp) (97 mg, 0.236 mmol), and 1,10-phen-
anthroline monohydrate (94 mg, 0.472 mmol) were dissolved in
anhydrous DMF (8 mL) in a threaded ACE glass pressure tube. The
tube was fitted with a pressure head, and the solution was satu-
rated with carbon monoxide (four cycles of 6 atm of CO). The re-
action mixture was heated at 120 ꢀC (oil bath temperature) under
CO (6 atm) for 16 h. Water (30 mL) was added and the orange so-
lution was extracted with ethyl acetate (3ꢂ30 mL). The combined
organic phases were dried (MgSO4), filtered, and the solvent was
removed under reduced pressure. The resulting crude product was
purified by chromatography (hexanes/EtOAc, 98:2) to afford 15
References and notes
1. Tomita, F.; Tsuji, M.; Tanaka, A. Jpn. Kokai Tokkyo Koho, JP2000026413, 2000.
[CAN 132:74873].
2. Phay, N.; Higashiyama, T.; Tsuji, M.; Matsuura, H.; Fukushi, Y.; Yokota, A.;
Tomita, F. Phytochemistry 1999, 52, 271–274.
3. A few compounds have been suggested based solely on low-resolution MS data,
see: (a) Seo, J.-S.; Keum, Y.-S.; Cho, I. K.; Li, Q. X. Int. Biodeterior. Biodegrad. 2006,
58, 36–43; (b) Grifoll, M.; Selifonov, S. A.; Gatlin, C. V.; Chapman, C. P. J. Appl.
Environ. Microbiol. 1995, 61, 3711–3723; (c) Yamada, K.; Minoda, Y.; Umehara, K.
Jpn. Kokai Tokkyo Koho [JP 48091279; CAN 80:119229].
4. Terada, Y.; Arisawa, M.; Nishida, A. J. Org. Chem. 2006, 71, 1269–1272.
5. Capon, B.; Kwok, F. C. J. Am. Chem. Soc. 1989, 111, 5346–5356.
6. Tanoue, Y.; Terada, A.; Sakata, K.; Hashimoto, M.; Morishita, S.-I.; Hamada, M.;
Kai, N.; Nagai, T. Fish. Sci. 2001, 67, 726–729.
(1.01 g, 2.73 mmol, 70%) as a tan solid. Mp 73–74 ꢀC; 1H NMR
d 7.82
(br s, 1H), 7.51 (d, J¼7.2 Hz, 1H), 7.28 (d, J¼7.2 Hz, 1H), 7.10 (dt, J¼7.2,
1.2 Hz, 1H), 7.06 (dt, J¼7.2, 1.2 Hz, 1H), 6.23 (d, J¼2.4 Hz, 1H), 2.74 (t,
J¼7.8 Hz, 2H),1.71 (pent, J¼7.2 Hz, 2H),1.39 (m, 2H),1.35–1.18 (br m,
7. Pang, Z.; Sterner, O. J. Nat. Prod. 1994, 57, 852–857.
8. Stachel, S. J.; Nilges, M.; Van Vranken, D. L. J. Org. Chem. 1997, 62, 4756–4762.
9. Komai, T. Chem. Pharm. Bull. 1956, 4, 314–315.
10. Astolfi, P.; Greci, L.; Rizzoli, C.; Sgarabotto, P.; Marrosu, G. J. Chem. Soc., Perkin
Trans. 2 2001, 1634–1640.
28H), 0.88 (t, J¼7.2 Hz, 3H); 13C NMR
d 140.0, 135.8, 128.9, 120.9,
119.7, 119.6, 110.2, 99.5, 31.9, 29.7, 29.6, 29.5, 29.4, 29.3, 29.2, 28.3,

R.W. Clawson Jr. et al. / Tetrahedron 65 (2009) 8786–8793
8793
11. Clawson, R. W., Jr.; Deavers, R. E., III; Akhmedov, N. G.; So¨derberg, B. C. G. Tet-
rahedron 2006, 62, 10829–10834.
21. Barrero, A. F.; Carbrera, E.; Rodriguez, I.; Fernandez-Gallego, E. M. Phytochem-
istry 1994, 36, 189–194.
12. Clawson, R. W., Jr.; So¨derberg, B. C. G. Tetrahedron Lett. 2007, 48, 6019–6021.
13. A modification of the procedure by Evans, D. A.; Andrews, G. C.; Buchwalter, B.
J. Am. Chem. Soc. 1974, 96, 5560–5561.
14. Using the procedure of Gadwood, R. C.; Rubino, M. R.; Nagarajan, S. C.; Michel,
S. T. J. Org. Chem. 1985, 50, 3255–3260.
22. Barrero, A. F.; Sanchez, J. F.; Rodriguez, I. Phytochemistry 1990, 29, 1967–1969.
23. Tan, R. X.; Kong, L. D.; Wei, H. X. Phytochemistry 1998, 47, 1223–1226.
24. Malakov, P.; Papanov, G.; Ziesche, J. Phytochemistry 1982, 21, 2589–2590.
25. Qin, J. J.; Jin, H. Z.; Fu, J. J.; Hu, X. J.; Zhu, Y.; Shen, Y. H.; Yan, S. K.; Zhang, W. D.
Chin. Chem. Lett. 2008, 19, 556–558.
15. McDonald, F. E.; Schultz, C. C.; Chatterjee. J. Org. Chem. 1995, 14, 3628–3629.
16. Dropinski, J. F.; Akiyama, T.; Einstein, M.; Habulihaz, B.; Doebber, T.; Berger, J. P.;
Meinke, P. T.; Shi, G. Q. Bioorg. Med. Chem. Lett. 2005, 15, 5035–5038.
17. Witkop, B.; Patrick, J. B.; Rosenblum, M. J. Am. Chem. Soc. 1951, 73, 2641–2647.
18. For a review of this reaction and other related oxidative cleavages of the 2,3-
bond, see: Mentel, M.; Breinbauer, R. Curr. Org. Chem. 2007, 11, 159–176.
19. Nawasreh, M.; Abu Zarga, M.; Sabri, S.; Al-Khalil, S.; Bomeister, E.; Zeller, C.-P.;
Atta-ur-Rahman, A. J. Pharm. Sci. 2004, 18, 165–170.
26. Katritzky, A. R.; Akutagawa, K. J. Am. Chem. Soc. 1986, 108, 6808–6809.
27. Rodriguez, A. L.; Koradin, C.; Dohle, W.; Knochel, P. Angew. Chem., Int. Ed. 2000,
39, 2488–2490.
28. Hamze, A.; Provot, O.; Alami, M.; Brion, J. Org. Lett. 2005, 7, 5625–5628.
29. Lebel, H.; Jacobsen, E. N. J. Org. Chem. 1998, 63, 9624–9625.
30. Following the procedure for the preparation of 1-iodoheptadecane: Ref. 30.
31. Marukawa, K.; Takikawa, H.; Mori, K. Biosci. Biotechnol. Biochem. 2001, 65,
305–314.
20. Al-Khalil, S.; Masalmeh, A.; Abdalla, S.; Tosa, H.; Iinuma, M. J. Nat. Prod. 1995, 58,
760–763.
32. This compound decomposes rapidly at ambient temperature and was used
immediately after chromatography in the next step.