Concise synthesis of (±)-horsfiline and (±)-coerulescine by tandem cyclisation of iodoaryl alkenyl azides

Article abstract of DOI:10.1039/b208114h

A brief, efficient and economical synthesis of the spiropyrrolidinyloxindoles, horsfiline and coerulescine, has been achieved, starting from itaconic acid and, respectively, p-anisidine or o-iodoaniline. Tandem radical cyclisation of iodoaryl alkenyl azides is the key step in both syntheses.

Full text of DOI:10.1039/b208114h

View Article Online / Journal Homepage / Table of Contents for this issue
Concise synthesis of ( )-horsfiline and ( )-coerulescine by tandem
cyclisation of iodoaryl alkenyl azides†
Dimitrios E. Lizos and John A. Murphy*
Department of Pure and Applied Chemistry, University of Strathclyde, 295 Cathedral Street,
Glasgow, Scotland, UK G1 1XL. E-mail: John.murphy@strath.ac.uk
Received 19th August 2002, Accepted 5th September 2002
First published as an Advance Article on the web 29th November 2002
A brief, efficient and economical synthesis of the spiropyrrolidinyloxindoles, horsfiline and coerulescine, has been
achieved, starting from itaconic acid and, respectively, p-anisidine or o-iodoaniline. Tandem radical cyclisation of
iodoaryl alkenyl azides is the key step in both syntheses.
were identified to have potent and selective action on human
breast cancer cell lines.
Introduction
The recent isolation of the natural products1 spirotryprostatin
A 1 and spirotryprostatin B 2 (Scheme 1), and the discovery of
their activity (albeit mild activity) as cell cycle inhibitors has
added considerable impetus to research on the spiropyrrol-
idinyloxindole alkaloids. Total syntheses2 of the spiro-
tryprostatins A and B have been achieved, but it transpires
that synthetic intermediates prepared during Danishefsky’s
synthetic studies3 on these compounds, may be more interesting
than the spirotryprostatins themselves. Three compounds, 3–5,
Among these, tricyclic compound 3 is exciting since it is a
relatively uncomplicated potential lead for development of
novel anti-tumour agents. Also highly active was the pair of
diastereomers 4 and 5. The discovery of equal activity for the
diastereomeric pair shows that the stereochemistry of the spiro
centre in these molecules is not crucial for anti-tumour action.
Although the cellular site of action of these three compounds
has not yet been disclosed, if it is assumed that they act at a
common locus, as a working hypothesis, then the anti-tumour
activity of tricyclic spiropyrrolidinyloxindoles requires further
investigation, and therefore synthetic routes that provide easy
access to these classes of compounds become important.
It is not surprising therefore that there has been a significant
and increasing interest in the synthesis of the tricyclic natural
† Electronic supplementary information (ESI) available: 1H and 13C
NMR spectra are provided for the compounds; the preparations of
compounds 19, 20 and 21 and the conversion of 20 to 16 are also
included. See http://www.rsc.org/suppdata/ob/b2/b208114h/
Scheme 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 1 7 – 1 2 2
117

View Article Online
products, horsfiline4a–h,4j 6 (isolated by Bodo and co-workers4a)
and coerulescine4d,i–l 7 (isolated5 by Colegate and co-workers)
that bear the same tricyclic skeleton as 3. A range of synthetic
approaches has been used. With respect to horsfiline, Jones and
Wilkinson4b used a free radical cyclisation as the key step in
their synthesis. Lévy and co-workers produced4d two routes fea-
turing (i) oxidative rearrangement of a tetrahydrocarboline (see
also ref. 4a) and (ii) oxidation and formaldehyde-treatment of
an appropriately substituted tryptamine. In a synthesis of (Ϫ)-
horsfiline as well as, separately, the unnatural enantiomer,
Borschberg and co-workers4c studied the stereospecific conver-
sion of tryptamine derivatives firstly to tetrahydrocarbolines
and thence to spiropyrrolidinylindolones, while Palmisano et al.
prepared4e (Ϫ)-6 via the (chiral auxiliary-directed) cyclo-
addition of azomethine ylides onto 3-methyleneindolones.
Meanwhile, Fuji and co-workers used4f asymmetric nitroolefin-
ation of indolones to introduce the correct stereochemistry at
what would become the spiro-carbon, subsequently making
the pyrrolidine ring. More recently, in preparations of ( )-6,
Carreira’s group4g have ring-expanded a spiro-cyclopropyl-
indolone to the requisite spiropyrrolidine, and Kumar et al.
rearranged4j an (aziridinoethylidene)indolone to a pyrrolidinyl-
indolone.
Although 16a had now been produced, the process of pro-
tecting and later deprotecting the alkene detracted from the
directness of the synthetic route. With this in mind, we returned
to investigate further whether a clean reduction of 15a to 16a
could be found. Gratifyingly, success was achieved with LiBH4,
generated in situ from lithium chloride and sodium boro-
hydride.12 This afforded the desired product cleanly and in high
yield (83%). Conversion to the azide 17a (68%) was accom-
plished using diphenylphosphoryl azide, and this was followed
by tandem cyclisation to afford the tricyclic product 22a; this
was subjected to in situ methylation to yield 23a (60% over two
steps). Finally deprotection under Birch conditions afforded
horsfiline 6 (87%).
Significantly, the tandem radical cyclisation afforded solely
the desired tricyclic skeleton; no product arising from apparent
6-endo-trig cyclisation of the aryl radical onto the alkene was
seen. In studies related to the synthesis of these same alkaloids,
Jones et al. had shown13 how the balance of 5-exo to apparent
6-endo cyclisation is determined by kinetic and thermodynamic
factors. At lower temperatures, the 5-exo predominates,14 but
at higher temperatures, rearrangement to 6-endo can occur
through a neophyl rearrangement. In our case, the product of
the initial 5-exo cyclisation is likely to be trapped by the azide
group, preventing side-products from forming.
The same protocol was then followed for the synthesis of
coerulescine and with generally excellent yields, producing
intermediates 13b (92%), 15b (94%), 16b (84%), 17b (70%) and
23b (68%). We were a little concerned about the final deprotec-
tion step to afford coerulescine. Debenzylation of 23a in the
horsfiline case involves selective addition of solvated electrons
to the aromatic ring of the benzyl protecting group. It is entirely
reasonable that this should be a very selective process, since the
aromatic ring in the horsfiline skeleton is likely to be far more
electron-rich, and hence more difficult to reduce, than that of
the benzyl group, due to the electron-donating properties of the
methoxy and amide groups. With coerulescine, the absence of
the methoxy group should make the difference between the two
rings less pronounced, but the benzyl group should still bear
the more electron-poor ring. In practice, the selectivity was
maintained and the reaction proceeded in an excellent (83%)
yield.
It is seen that the iodoaryl alkenyl azide tandem radical
cyclisation reaction affords a rapid route to ( )-horsfiline in
20% yield from t-BOC-p-anisidine [18% from anisidine], and
( )-coerulescine in 29% yield from 2-iodoaniline. An acid
chloride derived from the economical itaconic acid is the key
partner reagent for both syntheses. The efficiency of this route
ranks with the highest reported.
From our perspective, the spiropyrrolidinyl-oxindole skele-
ton should be an ideal target for synthesis using the iodoaryl
azide tandem radical cyclisation strategy that we recently
exploited in the total synthesis of aspidospermidine6 and in the
formal total synthesis of vindoline.7 Thus tandem cyclisation of
a radical, 8, should afford the tricyclic system, 9, in a single
process. This paper reports8 the realisation of that goal.
Results and discussion
To synthesise horsfiline, dilithiated t-BOC-p-anisidine 10a
was iodinated9 with 1,2-diiodoethane, to give 11a (86%).
Deprotection of the t-BOC group gave 12a (89%), which was
reductively alkylated with benzaldehyde10 (94%). (Conversion
of the primary amine 12a to a secondary amine such as 13a is
necessary since the radical cyclisations about to be undertaken
on the amide substrates, e.g. 17 in Scheme 2 do not succeed4b
with secondary amides. We chose benzylation for this con-
version, because of the later easy removal of the benzyl group).
Condensation of the benzylamine 13a with the acid chloride11
14 of the monomethyl ester of itaconic‡ acid (in turn readily
available11 from commercially available itaconic acid) afforded
the amide 15a (94%). It was now necessary to reduce the ester
group in amide 15a to an alcohol, 16a, before conversion to
azide 17a and cyclisation. Initial attempts to reduce the ester to
alcohol 16a with DIBAL-H led to the desired product 16a as
the major product, but it was contaminated with a minor com-
pound that, based on examination of the NMR spectrum, we
propose to be 18, where unexpected reduction of the alkene
has occurred. Unfortunately the two compounds had similar
physical properties, so separation was not possible. To over-
come this problem, protection of the alkene double bond in 15a
was undertaken. Reaction of 15a with PhSeNa, formed in situ
by reduction of diphenyl diselenide with sodium borohydride,
surprisingly led to no reaction, and we attribute this to steric
hindrance to attack on the alkene by the large phenylseleno
group. To decrease the steric problems, PhSH was instead
reacted with ester 15a in the presence of base to give an
excellent yield of product 19 (96%). Reduction of 19 afforded
the desired alcohol 20 (55%) together with a small amount
of unexpected amine 21 (10%). Alcohol 20 was purified and
oxidized to the corresponding sulfoxide, which was directly
subjected to thermal elimination affording the alcohol 16a
(72%).
In mechanistic terms, this route shows greatest similarities
with the route of Jones and Wilkinson,4b since both use radical
cyclisations as the key steps (Scheme 3). However, the two routes
differ considerably in all other facets, (e.g. single cyclisation
versus tandem cyclisation) and most particularly in the nature
and reactivity of the final radicals, carbon radical 24 in their
case, versus nitrogen radical 25 in this case. Importantly,
in designing the synthesis of more complex spiropyrrolidinyl-
indolones, the diversity of efficient routes should now allow
access to a wide variety of substitution patterns of analogues
for anti-tumor testing.
Experimental section
Melting points were carried out on a Reichert hot stage appar-
atus and are uncorrected. Infra red spectra were obtained on a
Perkin Elmer 1720-X FTIR. Mass spectra were recorded on a
JEOL JMS-AX505 HA instrument using, electron impact
ionisation (EIϩ), chemical ionisation (CI), electrospray (ES) or
fast atom bombardment (FAB). Some low and high resolution
mass spectra were obtained on JLSX 102 and VG ZAB-E
instruments using peak-matching techniques at the EPSRC
‡ The IUPAC name for itaconic acid is methylenesuccinic acid.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 1 7 – 1 2 2
118

View Article Online
Scheme 2 Syntheses of horsfiline and coerulescine. Reagents and conditions: (a) t-BuLi, Et2O, Ϫ20 ЊC then ICH2CH2I, 86% for 11a; (b) TFA, DCM,
0 ЊC, 89% for 12a; (c) ZnCl2, PhCHO, MeOH, NaCNBH3, 94% for 13a, 92% for 13b; (d) Et3N, PhH, 94% for 15a; 94% for 15b; (e) PhSH, DBU, THF,
∆, 96% for 19; (f ) DIBAL-H, PhMe, Ϫ78 ЊC, 55% for 20; (g) m-CPBA, DCM, then PhMe, ∆, 72 h, 72% for 16a; (h) LiCI, NaBH4, EtOH, THF, 83%
for 16a, 84% for 16b; (i) PPh3, (PhO)2P(O)N3, DEAD, THF, 68% for 17a; 70% for 17b; (j) (Me3Si)3SiH, AIBN, PhH; (k) CH2O, NaCNBH3, MeCN,
yield over 2 steps: 60% for 23a, 68% for 23b; (1) Na, NH3, 86% for 6, 83% for 7.
Scheme 3 The different regiochemical outcomes of the two radical cyclisation approaches to spiropyrrolidinylindolones.
1
Mass Spectrometry Centre, Swansea. H NMR spectra were
recorded at 400 MHz on a Bruker DPX 400 spectrometer.
13C{H} NMR spectra were similarly recorded at 100.61 MHz
on a Bruker DPX400 machine. NMR experiments were carried
out in deuterochloroform with tetramethylsilane as internal
reference unless otherwise stated. 13C NMR spectra were
acquired in a broadband decoupled mode with the multi-
plicities obtained using a DEPT or Jmod sequence. Chemical
shifts (δ) are reported in parts per million (ppm) from tetra-
methylsilane. The following abbreviations are used for multi-
plicities: s = singlet; d = doublet; t = triplet; q = quartet; m =
multiplet, br = broad. Coupling constants (J) are reported in
Hertz (Hz).
tert-Butyl (2-iodo-4-methoxyphenyl)carbamate 11a9
Addition of tert-butyllithium in pentane (1.7 M, 39.6 cmϪ3
67.2 mmol, 2.5 equiv.) to a solution of tert-butyl (4-methoxy-
phenyl)carbamate (6.0 g, 26.9 mmol, 1.00 equiv.), in dry diethyl
ether (60 cmϪ3), was performed under a nitrogen atmosphere at
Ϫ20 ЊC, followed by stirring for 3 h at the same temperature. To
this mixture was added 1,2-diiodoethane (11.4 g, 40.3 mmol,
,
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 1 7 – 1 2 2
119

View Article Online
1.5 equiv.) in dry diethyl ether (80 cmϪ3) at Ϫ78 ЊC, and the
reaction was gradually warmed to room temperature and
stirred for 16 h. After addition of a saturated aqueous solution
of sodium thiosulfate (200 cmϪ3), the mixture was extracted
with diethyl ether. The organic extracts were then washed with
brine, dried over anhydrous sodium sulfate and evaporated
under reduced pressure. The residue was purified by column
chromatography (10 : 1 petroleum ether–ethyl acetate) to yield
tert-butyl (2-iodo-4-methoxyphenyl)carbamate 11a as colour-
less prisms (8.1 g, 86%); mp 48–49 ЊC (lit.9 mp 49–50 ЊC);
[Found (EI) Mϩ 349.0172. C12 H16INO3 requires M, 349.0175].
νmax (film)/cmϪ1 3399, 2976, 2933, 2834, 1725, 1511, 1367, 1250,
1161, 1037; δH 1.51 (9H, s, 3 × CH3), 3.72 (3H, s, CH3), 6.55
(1H, br s, NH), 6.84–6.87 (1H, dd, J 9.0, 2.8 ArH), 7.28 (1H, d,
J 2.8, ArH) 7.76 (1H, d, J 9.0, ArH); δC 28.5 (q), 55.9 (q), 80.9
(s), 90.5 (s), 115.1 (d), 122.3 (d), 123.9 (d), 132.6 (s), 153.3 (s),
156.3 (s). m/z (EI) 349 (29%, Mϩ), 293 (96), 249 (46), 166 (34),
106 (22) and 57 (100).
carbonylbut-3-enoic acid methyl ester 14 as a colourless oil
(10.8 g, 96%) bp 84–87 ЊC (11 mmHg) (Found: Mϩ 162.0080,
C6H7ClO3 requires M 162.0084); νmax (film)/cmϪ1 3004, 2955,
2847, 1740, 1638, 1437, 1343, 1206; δH 3.39 (2H, d, J 0.8, CH2),
3.72 (3H, s, CH3), 6.17 (1H, t, J 0.8, CH2), 6.70 (1H, s, CH2); δC
37.8 (t), 52.8 (q), 136.5 (t), 138.7 (s), 168.6 (s), 170.3 (s); m/z (EI)
163 ([MH]ϩ, 5%), 127 (60), 113 (95), 99 (50), 85 (100).
3-[Benzyl(2-iodo-4-methoxyphenyl)carbamoyl]but-3-enoic acid
methyl ester 15a
Triethylamine (1.6 cmϪ3, 11.4 mmol, 1.10 equiv.) was added
to a stirred solution of benzyl(2-iodo-4-methoxyphenyl)amine
(3.5 g, 10.3 mmol, 1.00 equiv.) in dry benzene (10 cmϪ3) at room
temperature. 3-Chlorocarbonylbut-3-enoic acid methyl ester
(2.18 g, 13.4 mmol, 1.30 equiv.) in dry benzene (10 cmϪ3) was
added dropwise. The resulting solution was stirred at rt for 30
min, then heated under reflux for 90 min. After cooling, the
solvent was removed under reduced pressure and the residue
dissolved in diethyl ether (50 cmϪ3). The ether was washed with
water, dried over anhydrous sodium sulfate and concentrated.
Column chromatography eluting with petroleum ether–ethyl
acetate 4 : 1 gave pure 3-[benzyl(2-iodo-4-methoxyphenyl)-
carbamoyl]but-3-enoic acid methyl ester 15a as a white solid mp
107–110 ЊC (4.5 g, 94%); Found C, 51.68; H, 4.36; N, 3.03.
C20H20INO4 requires C, 51.61; H, 4.33; N, 3.01%; [Found: (ES)
MHϩ 466.0521, C20H21INO4 requires MH 466.0515]; νmax (film)/
cmϪ1 3065, 2998, 2841, 1736, 1651, 1622, 1491, 1287, 1175; δH
3.06 (1H, d, J 16.5, CH2), 3.64–3.69 (4H, CH3 ϩ CH2), 3.78
(3H, s, CH3), 4.17 (1H, d, J 14.3, CH2), 5.16 (1H, s, CH2), 5.27
(1H, s, CH2), 5.77 (1H, d, J 14.3, CH2), 6.70 (1H, dd, J 2.8, 8.7,
ArH), 6.81 (1H, d, J 8.7, ArH), 7.23–7.31 (5H, m, ArH), 7.42
(1H, d, J 2.8, ArH); δC 40.1 (t), 52.0 (q), 52.1 (t), 55.7 (q), 100.1
(s), 114.4 (d), 123.2 (t), 125.0 (d), 127.6 (d), 128.4 (d), 129.6 (d),
132.0 (d), 136.5 (s), 136.8 (s), 137.5 (s), 159.0 (s), 169.2 (s), 171.3
(s); m/z (ES) 466 [(MH)ϩ, 60%], 340 (100).
4-Amino-3-iodoanisole 12a
Trifluoroacetic acid (21.2 cmϪ3, 274.9 mmol, 12.00 equiv.) was
added to a solution of tert-butyI (2-iodo-4-methoxyphenyl)-
carbamate (8.0 g, 22.9 mmol, 1.00 equiv.) in dry DCM
(120 cmϪ3) at 0 ЊC and the reaction was stirred at room temper-
ature for 2 h. Sodium hydroxide (2 M) was added followed by
hydrochloric acid (2 M) until neutral pH had been attained.
The two layers were separated and the aqueous one was
extracted with DCM. The organic layers were washed with
brine, dried over anhydrous sodium sulfate and the solvent was
removed in vacuo. The crude product was purified by column
chromatography (4 : 1 petroleum ether–ethyl acetate) giving
4-amino-3-iodoanisole 12a as a colourless oil (5.1 g, 89%);
[Found (ES): MHϩ 249.9732 C7H9INO, requires MH,
249.9729]; νmax (film)/cmϪ1 3431, 3348, 2939, 2830, 1596, 1493,
1232, 1037; δH 3.62 (2H, br s, NH2), 3.77 (3H, s, CH3), 6.74
(1H, d, J, 8.7, ArH), 6.82 (1H, dd, J 8.7, 2.7, ArH), 7.26 (1H, d,
J 2.7, ArH); δC 56.4 (q), 84.7 (s), 115.9 (d), 116.5 (d), 124.0 (d),
141.3 (s), 153.1 (s); m/z (CI) 267 (MNH4ϩ, 63%), 250 (Mϩ, 100).
Reduction of 3-[benzyl(2-iodo-4-methoxyphenyl)carbamoyl]but-
3-enoic acid methyl ester 15a using NaBH4 and LiCl
Lithium chloride (16.2 mg, 0.43 mmol, 2.00 equiv.) was added
Benzyl(2-iodo-4-methoxyphenyl)amine 13a
to
a solution of 3-[benzyl(2-iodo-4-methoxyphenyl)carb-
A stirred mixture of 4-amino-3-iodoanisole 12 (3 g, 12.00
mmol, 1.00 equiv.), zinc() chloride (1.97 g, 14.45 mmol, 1.20
equiv.) and benzaldehyde (1.47 cmϪ3, 14.45 mmol, 1.20 equiv.)
in methanol, (50 cmϪ3) was treated with sodium cyanoboro-
hydride (0.91 g, 14.45 mmol, 1.20 equiv.) and warmed at reflux
under nitrogen for 2 h. The cooled reaction mixture was diluted
with sodium hydroxide (1 M, 50 cmϪ3) and extracted with
diethyl ether. The combined ether layers were washed with
water, dried over anhydrous sodium sulfate and concentrated in
vacuo. Benzyl(2-iodo-4-methoxyphenyl)amine 13a was purified
by column chromatography on silica gel eluting with petroleum
ether–ethyl acetate 5 : 1 to give pure title compound as a yellow
oil (3.85 g, 94%); Found: C, 49.87; H, 4.15; N, 4.06. C14H14INO
requires C, 49.56; H, 4.16; N, 4.13%; (Found (ES): MHϩ
340.0196, C14H15INO requires MH 340.0198); νmax (film)/cmϪ1
3401, 3028, 2998, 2830, 1603, 1504, 1282, 1039; δH 3.75 (3H, s,
CH3), 4.33 (1H, br s, NH), 4.39 (2H, d, J 4.2, CH2), 6.54 (1H, d,
J 8.9, ArH), 6.84, (1H, dd, J 2.9, 8.9, ArH), 7.30–7.42 (6H, m,
ArH); δC 49.2 (t), 56.1 (q), 85.5 (s), 111.7 (d), 115.7 (d), 124.8
(d), 127.4 (d), 128.8 (d), 139.1 (s), 142.0 (s), 152.2 (s); m/z (CI)
340.1 ([MH]ϩ, 100%).
amoyl]but-3-enoic acid methyl ester 15a (100 mg, 0.21 mmol,
1.00 equiv.) in dry THF–EtOH (2 : 1 mixture, 3 cmϪ3). When it
had dissolved, NaBH4 (16.2 mg, 0.43 mmol, 2.00 equiv.) was
added and the reaction mixture was stirred at room temperature
for 16 h. Then the solvent was evaporated near dryness and
H2O (4 cmϪ3) was added. The organic phase was extracted
with ethyl acetate, washed with H2O, dried over anhydrous
sodium sulfate and concentrated under vacuum. The reaction
mixture was subjected to column chromatography eluting with
petroleum ether–ethyl acetate 1 : 1 to afford N-benzyl-2-(2-
hydroxyethyl)-N-(2-iodo-2-methoxyphenyl)acrylamide 16a as a
colourless oil (76 mg, 83%); [Found (ES): MHϩ 438.0563,
C19H21INO3 requires MH 438.0566]; νmax (film)/cmϪ1 3409,
2938, 1614, 1487, 1287, 1030; δH 2.01 (1H, br s, OH) 2.25–2.31
(1H, m, CH2), 2.48–2.54 (1H, m, CH2), 3.62–3.77 (5H, m, CH2
and CH3), 4.10 (1H, d, J 14.1, CH2), 5.10 (1H, s, CH2), 5.14
(1H, s, CH2), 5.71 (1H, d, J 14.1, CH2), 6.56 (1H, d, J 8.7 ArH),
6.68 (1H, dd, J 2.7, 8.7, ArH), 7.21–7.39 (5H, m, ArH), 7.40
(1H, d, J 2.7, ArH); δC 38.2 (t), 52.4 (t), 55.7 (q), 62.6 (t), 100.1
(s), 114.4 (d), 120.2 (t), 125.0 (d), 127.9 (d), 128.6 (d), 129.6 (d),
131.6 (d), 136.6 (s), 137.1 (s), 142.0 (s), 159.1 (s), 172.2 (s); m/z
(El) 438.1 [(MH)ϩ, 100%]
3-Chlorocarbonylbut-3-enoic acid methyl ester 1411
2-(2-Azidoethyl)-N-benzyl-N-(2-iodo-4-methoxyphenyl)-
acrylamide 17a
Thionyl chloride (7.1 cmϪ3, 97.2 mmol, 1.40 equiv.) was added
to a solution of 2-methylenesuccinic acid 4-methyl ester
(10 g, 69.4 mmol, 1.00 equiv.) in dry diethyl ether (8 cmϪ3
)
Triphenylphosphine (363 mg, 1.38 mmol, 1.10 equiv.), DEAD
(0.22 cmϪ3, 1.38 mmol, 1.10 equiv.) and diphenylphosphoryl
azide (0.3 cmϪ3, 1.38 mmol, 1.10 equiv.) were added sequen-
and the reaction mixture was refluxed gently until gas evolution
was essentially complete (15 min). Distillation gave 3-chloro-
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 1 7 – 1 2 2
120

View Article Online
tially to a solution of alcohol 16a in dry THF (12 cmϪ3), and the
solution was stirred at rt for 24 h. Then the solvent was evapor-
ated and the residue was subjected to column chromatography
using petroleum ether–ethyl acetate 5 : 1 to afford pure 2-(2-
azidoethyl)-N-benzyl-N-(2-iodo-4-methoxyphenyl) acrylamide
17a as a slightly yellow oil (396 mg, 68%); [Found (ES): MHϩ
463.0622, C19H20IN4O2 requires MH 463.0631]; νmax (film)/cmϪ1
3030, 2938, 2838, 2096, 1651, 1622, 1488, 1287, 1030; δH 2.34–
2.41 (1H, m, CH2), 2.55–2.62 (1H, m, CH2), 3.43 (2H, t, J 6.9,
CH2), 3.78 (3H, s, CH3), 4.09 (1H, d, J 14.1, CH2), 5.12 (1H, s,
CH2), 5.17 (1H, s, CH2), 5.71 (1H, d, J 14.1, CH2), 6.55 (1H, d,
J 8.7, ArH), 6.69 (1H, dd, J 2.7, 8.7, ArH), 7.21–7.35 (5H, m,
ArH), 7.41 (1H, d, J 2.7, ArH); δC 33.9 (t), 50.1 (t), 52.3 (t), 55.8
(q), 100.4 (s), 114.5 (d), 120.3 (t), 125.1 (d), 127.8 (d), 128.6 (d),
129.7 (d), 131.7 (d), 136.9 (s), 137.1 (s), 140.7 (s), 159.2 (s), 170.5
(s); m/z (CI) 463 [(MH)ϩ, 35%], 115 (100).
MeOH 9 : 1 afforded pure ( )-horsfiline as a white solid
(18.2 mg, 87%); mp 154–156 ЊC (lit.4g mp 156–157 ЊC) [Found
(ES): MHϩ 233.1287, C13H17N2O2 requires MH 233.1290]; νmax
(film)/cmϪ1 3400, 2944, 2849, 1704, 1604, 1491, 1304, 1209,
1031; δH 2.05–2.18 (1H, m, CH2), 2.38–2.44 (1H, m, CH2), 2.49
(3H, s, CH3), 2.77–2.84 (1H, m, CH2), 2.90 (1H, d, J 9.4, CH2),
2.97 (1H, d, J 9.4, CH2), 3.06–3.11 (1H, m, CH2), 3.80 (3H, s,
CH3), 6.72 (1H, dd, J 2.5, 8.4, ArH), 6.80 (1H, d, J 8.4, ArH),
7.06 (1H, d, J 2.5, ArH), 8.32 (1H, br s, NH); δC 38.3 (t), 41.9
(q), 54.3 (s), 56.1 (t), 56.8 (q), 66.2 (t), 110.0 (d), 110.6 (d), 112.9
(d), 133.6 (s), 137.4 (s), 156.5 (s), 182.8 (s); m/z (CI) 233 [(M ϩ
H)ϩ), 100%].
Benzyl(2-iodophenyl)amine 13b
12b (5 g, 22.80 mmol, 1.00 equiv.), was treated as for 12a and
afforded benzyl(2-iodophenyl)amine 13b as a yellow oil (6.5 g,
92%) [Found (ES): MHϩ 310.0095, C13H13IN requires MH
310.0093]; νmax (film)/cmϪ1 3400, 3060, 3027, 2854, 1589, 1502,
1318, 1004; δH 4.52 (2H, d, J 5.6, CH2), 4.74 (1H, br s, NH) 6.56
(1H, ddd, J 1.4, 7.3, 7.8, ArH), 6.64 (1H, dd, J 1.3, 8.1, ArH),
7.26 (1H, ddd, J 1.4, 7.3, 8.1, ArH), 7.39–7.48 (5H, m, ArH),
7.78–7.80 (1H, dd, J 1.4, 7.8, ArH); δC 48.5 (t), 85.5 (s), 111.1
(d), 119.0 (d), 127.4 (d), 127.5 (d), 128.9 (d), 129.6 (d), 138.8 (s),
139.2 (d), 147.3 (s); m/z (EI) 309 [(MH)ϩ, 18%], 180 (20), 91
(100) and 65 (22).
Tandem radical cyclisation of 2-(2-azidoethyl)-N-benzyl-N-
(2-iodo-4-methoxyphenyl)acrylamide 17a
2-(2-Azidoethyl)-N-benzyl-N-(2-iodo-4-methoxyphenyl)acryl-
amide 17a (295 mg, 0.64 mmol, 1.00 equiv.) was dissolved in
dry benzene (20 cmϪ3) and the solution was purged with argon
for 30 min. The solution was brought to reflux and tristrimeth-
ylsilylsilane (TTMSS) (0.22 cmϪ3, 0.70 mmol, 1.10 equiv.)
and AIBN (21 mg, 0.13 mmol, 0.2 equiv.) were added and
the mixture was refluxed for 5 h. The reaction mixture was
allowed to cool down to room temperature and the solvent was
removed in vacuo. Ethyl acetate (50 cmϪ3) was added and the
mixture extracted with hydrochloric acid (2 M). The aqueous
phase was basified with sodium hydroxide (2 M) and then it was
extracted with ethyl acetate. The organic phase was dried over
anhydrous sodium sulfate and concentrated in vacuo. The crude
1-benzyl-5-methoxy-1H-spiro[indole-3,3Ј-pyrrolidin]-2-one 22a
(150 mg, 0.49 mmol, 1.00 equiv.) from the previous step was
dissolved in acetonitrile (7 cmϪ3) and treated5f with formalde-
hyde [(0.24 cmϪ3, 2.92 mmol, 6.00 equiv.) 37% in water] and
NaCNBH3 (61 mg, 0.97 mmol, 2.00 equiv.) at rt. The mixture
was stirred for 15 min and neutralised with acetic acid. After
stirring was continued for another 1 h, the mixture was basified
with ammonia solution and the solvent was removed in vacuo.
The residue was extracted with ethyl acetate. The combined
organic extracts were washed with brine, dried over anhydrous
sodium sulfate and concentrated under reduced pressure.
Column chromatography eluting with ethyl acetate: acetone 4
: 1 gave pure 1-benzyl-5-methoxy-1Ј-methyl-1H-spiro[indole-
3,3Ј-pyrrolidin]-2-one 23a as a colourless oil (123 mg, 60% over
two steps) (Found MHϩ 323.1761, C20H23N2O2 requires MH
323.1759); νmax (film)/cmϪ1 2941, 2836, 2787, 1707, 1601, 1491,
1177, 1035; δH 2.12–2.20 (1H, m, CH2), 2.40–2.46 (1H, m, CH2),
2.51 (3H, s, CH3), 2.78–2.85 (1H, m, CH2), 2.91 (1H, d, J 9.4,
CH2), 3.01 (1H, d, J 9.4, CH2), 3.15–3.20 (1H, m, CH2), 3.77
(3H, s, CH3), 4.88 (2H, s, CH2), 6.59 (1H, d, J 8.5, ArH), 6.68
(1H, dd, J 2.6, 8.5, ArH), 7.11 (1H, d, J 2.6, ArH), 7.24–7.33
(5H, m, ArH); δC 38.4 (t), 42.0 (q), 44.1 (t), 53.9 (s), 56.1 (q),
56.8 (t), 66.2 (t), 109.3 (d), 110.6 (d), 112.5 (d), 127.4 (d), 127.8
(d), 129.0 (d), 135.6 (s), 136.2 (s), 137.0 (s), 156.6 (s), 180.2 (s);
m/z (CI) 323 [(MH)ϩ, 100%].
3-[Benzyl(2-iodophenyl)carbamoyl]but-3-enoic acid methyl ester
15b
Benzyl(2-iodophenyl)amine 13b (4.5 g, 14.55 mmol, 1.00 equiv.)
was reacted as for 13a and gave 3-[benzyl(2-iodophenyl)carb-
amoyl]but-3-enoic acid methyl ester 15b as a white solid mp
60–62 ЊC (5.96 g, 94%); [Found (EI): Mϩ 435.0339, C19H18INO3
requires M 435.0331]; νmax (film)/cmϪ1 2949, 1736, 1653, 1625,
1470, 1172; δH 3.06 (1H, d, J 16.5, CH2), 3.67–3.70 (4H, m, CH2
and CH3), 4.23 (1H, d, J 14.3, CH2), 5.15 (1H, s, CH2), 5.27
(1H, s, CH2), 5.78 (1H, d, J 14.3, CH2), 6.94–7.00 (2H, m,
ArH), 7.16–7.20 (1H, m, ArH), 7.24–7.30 (5H, m, ArH), 7.91
(1H, d, J 7.9, ArH); δC 40.0 (t), 52.0 (q), 52.5 (t), 99.8 (s), 123.4
(t), 127.7 (d), 128.4 (d), 129.0 (d), 129.4 (d), 129.6 (d), 132.0 (d),
136.4 (s), 136.6 (s), 140.2 (d), 144.7 (s), 168.9 (s), 171.3 (s); m/z
(EI) 435 (Mϩ, 5%), 376 (9), 308 (18), 180 (19) and 91 (100).
N-Benzyl-2-(2-hydroxyethyl)-N-(2-iodophenyl)acrylamide 16b
3-[Benzyl(2-iodophenyl)carbamoyl]but-3-enoic acid methyl
ester 15b (12.1 g, 27.8 mmol, 1.00 equiv.) was treated as for
15a to afford N-benzyl-2-(2-hydroxyethyl)-N-(2-iodophenyl)-
acrylamide 16b as a colourless oil (9.5 g, 84%) [Found (ES):
MHϩ 408.0461, C18H19INO2 requires MH 408.0460]; νmax (film)/
cmϪ1 3401, 2931, 1647, 1619, 1469, 1030; δH 2.27–2.31 (1H, m,
CH2), 2.50–2.57 (1H, m, CH2), 3.20 (1H, br s, OH), 3.74–3.81
(2H, m, CH2), 4.14 (1H, d, J 14.3, CH2), 5.10 (1H, s, CH2), 5.13
(1H, s, CH2), 5.70 (1H, d, J 14.3, CH2), 6.68 (1H, d, J 7.6, ArH),
6.97–7.00 (1H, m, ArH), 7.15–7.28 (6H, m, ArH), 7.92 (1H, dd,
J 1.4, 7.9, ArH); δC 38.3 (t), 52.2 (t), 62.8 (t), 99.9 (s), 120.7 (t),
127.9 (d), 128.7 (d), 128.9 (d), 129.6 (d), 129.7 (d), 131.6 (d),
136.6 (s), 140.4 (d), 141.9 (s), 144.4 (s), 171.9 (s); m/z (EI) 407
(2%, Mϩ), 376 (12), 280 (17), 180 (17), 91 (100) and 65 (25).
( )-Horsfiline 64g
2-(2-Azidoethyl)-N-benzyl-N-(2-iodophenyl)acrylamide 17b
Ammonia (ca. 5 cmϪ3) was condensed at Ϫ50 ЊC and sodium
(31 mg) was added with vigorous stirring. 1-Benzyl-5-methoxy-
1Ј-methyl-1H-spiro[indole-3,3Ј-pyrrolidin]-2-one 23a (29.4 mg,
0.09 mmol, 1.00 equiv.) was added in THF (1 cmϪ3). After 12
min, a small amount of NH4Cl was added and the ammonia
left to evaporate. Water (25 cmϪ3) was added and the reaction
was extracted with ethyl acetate. The organic layer was dried
over anhydrous sodium sulfate and the solvent was evaporated
in vacuo. Column chromatography on silica gel with DCM–
Alcohol 16b (480 mg, 1.18 mmol, 1.00 equiv.) was treated as
for alcohol 16a to afford pure 2-(2-azidoethyl)-N-benzyl-N-
(2-iodophenyl)acrylamide 17b as a slightly yellow oil (357 mg,
70%); [Found (EI): Mϩ 432.0446, C18H17IN4O requires M
432.0447]; νmax (film)/cmϪ1 3031, 2934, 2875, 2095, 1651, 1624,
1469, 1385; δH 2.37–2.41 (1H, m, CH2), 2.57–2.64 (1H, m, CH2),
3.44 (2H, t, J 6.7, CH2), 4.16 (1H, d, J 14.2, CH2), 5.12 (1H, s,
CH2), 5.17 (1H, s, CH2), 5.70 (1H, d, J 14.2, CH2), 6.70 (1H, d,
J 7.5, ArH), 6.96–7.00 (1H, m, ArH), 7.16–7.38 (6H, m, ArH),
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 1 7 – 1 2 2
121

View Article Online
7.90 (1H, dd, J 1.4, 7.8, ArH); δC 33.9 (t), 50.1 (t), 52.2 (t), 99.9
(s), 120.3 (d), 120.7 (t), 125.8 (d), 127.9 (d), 128.6 (d), 129.0 (d),
129.6 (d), 130.1 (d), 131.7 (d), 136.9 (s), 140.5 (d), 140.5 (s),
145.4 (s) 170.5 (s); m/z (CI) 432 (100%, Mϩ).
References
1 (a) C.-B. Cui, H. Kakeya, G. Okada, R. Onose and H. Osada,
J. Antibiot., 1996, 49, 527; (b) C.-B. Cui, H. Kakeya and H. Osada,
J. Antibiot., 1996, 49, 832; (c) C.-B. Cui, H. Kakeya and H. Osada,
Tetrahedron, 1996, 52, 12651.
2 (a) P. R. Sebahar and R. M. Williams, J. Am. Chem. Soc., 2000, 122,
5666; (b) H. Wang and A. Ganesan, J. Org. Chem., 2000, 65, 4685;
(c) L. E. Overman and M. D. Rosen, Angew. Chem., Int. Ed., 2000,
39, 4596; (d ) T. D. Bagul, G. Lakshmaiah, T. Kawabata and K. Fuji,
Org. Lett., 2002, 4, 249; (e) T. Lindel, Nachr. Chem., 2000, 48, 1498;
( f ) H. S. Wang and A. Ganesan, J. Org. Chem., 2000, 65, 4685;
(g) F. von Nussbaum and S. J. Danishefsky, Angew. Chem., Int. Ed.,
2000, 39, 2175.
Tandem radical cyclisation of 2-(2-azidoethyl)-N-benzyl-N-
(2-iodophenyl)acrylamide 17b
2-(2-Azidoethyl)-N-benzyl-N-(2-iodophenyl)acrylamide
17b
(160 mg, 0.37 mmol, 1.00 equiv.) when treated as for azide 17a
gave pure 1-benzyl-1Ј-methyl-1H-spiro[indole-3,3Ј-pyrrolidin]-
2-one 23b as a colourless oil (73.6 mg, 68% over two steps)
[Found (ES): MHϩ 293.1652, C19H21N2O requires MH
293.1654]; νmax (film)/cmϪ1 2940, 2838, 2786, 1711, 1611, 1486,
1467, 1359, 1183; δH 2.11–2.18 (1H, m, CH2), 2.41–2.46 (1H, m,
CH2), 2.50 (3H, s, CH3), 2.77–2.83 (1H, m, CH2), 2.89 (1H, d, J
9.4, CH2), 2.95 (1H, d, J 9.4, CH2), 3.09–3.14 (1H, m, CH2), 4.9
(2H, s, CH2), 6.71 (1H, d, J 7.7, ArH), 7.04 (1H, ddd, J 1.5, 7.3,
7.7, ArH), 7.14 (1H, ddd, J 0.8, 7.7, 7.7, ArH), 7.24–7.33
(5H, m, ArH), 7.44 (1H, dd, J 1.4, 7.8, ArH); δC 38.3 (t), 42.0
(q), 44.0 (t), 53.5 (s), 56.9 (t), 66.6 (t), 108.9 (d), 123.1 (d),
123.2 (d), 127.4 (d), 127.7 (d), 127.8 (d), 129.0 (d), 136.0 (s),
136.2 (s), 142.1 (s), 180.6 (s); m/z (EI) 292 [(M)ϩ, 30%], 235
(100), 91 (80).
3 S. Edmondson, S. J. Danishefsky, L. Sepp-Lorenzino and N. Rosen,
J. Am. Chem. Soc., 1999, 121, 2147.
4 (a) A. Jossang, P. Jossang, H. A. Hadi, T. Sévenet and B. Bodo,
J. Org. Chem., 1991, 56, 6527; (b) K. Jones and J. Wilkinson,
J. Chem. Soc., Chem. Commun., 1992, 1767; (c) C. Pellegrini,
C. Strässler, M. Weber and H.-J. Borschberg, Tetrahedron:
Asymmetry, 1994, 5, 1979; (d ) S.-I Bascop, J. Sapi, J.-Y. Laronze
and J. Lévy, Heterocycles, 1994, 38, 725; (e) G. Palmisano,
R. Annunziata, G. Papeo and M. Sisti, Tetrahedron: Asymmetry,
1996, 7, 1; ( f ) G. Lakshmaiah, T. Kawabata, M. Shang and K. Fuji,
J. Org. Chem., 1999, 64, 1699; (g) C. Fischer, C. Meyers and
E. M. Carreira, Helv. Chim. Acta, 2000, 83, 1175; (h) G. Cravotto,
G. B. Giovenzana, T. Pilati, M. Sisti and G. Palmisano, J. Org.
Chem., 2001, 66, 8447; (i) M. E. Kuehne, D. M. Roland and
R. Hafter, J. Org. Chem., 1978, 43, 3705; (j) U. K. Syam Kumar,
H. Ila and H. Junjappa, Org. Lett., 2001, 3, 4193; (k) M. Sonei,
K. Noguchi, R. Yamagami, Y. Kawada, K. Yamada and
F. Yamada, Heterocycles, 2000, 53, 7; (l ) S. E. V. Bell, R. F. C. Brown,
F. W. Eastwood and J. M. Horvath, Aust. J. Chem., 2000, 53, 183.
5 N. Anderton, P. A. Cockrum, S. M. Colegate, J. A. Edgar, K. Flower,
I. Vit and R. I. Willing, Phytochemistry, 1998, 48, 437.
6 B. Patro and J. A. Murphy, Org. Lett., 2000, 2, 3599.
7 S.-Z. Zhou, S. Bommezijn and J. A. Murphy, Org. Lett., 2002, 4, 443.
8 For preliminary report, see: D. Lizos, R. Tripoli and J. A. Murphy,
Chem. Commun., 2001, 2732.
( )-Coerulescine 7
1-Benzyl-1Ј-methyl-1H-spiro[indole-3,3Ј-pyrrolidin]-2-one 23b
(22 mg, 0.075 mmol, 1.00 equiv.) was treated as for 23a and
afforded pure ( )-coerulescine 7 as a pale yellow gum (12.7 mg,
83%); [Found (ES): MHϩ 203.1182, C12H14N2O requires MH
203.1184]; νmax (film)/cmϪ1 3385, 2969, 2793, 1711, 1619, 1471,
1336, 1196, 1047; δH 2.10–2.18 (1H, m, CH2), 2.39–2.45 (1H, m,
CH2), 2.50 (3H, s, CH3), 2.80–2.86 (1H, m, CH2), 2.89–2.96
(2H, m, CH2), 3.05–3.11 (1H, m, CH2), 6.88 (1H, d, J 7.7, ArH),
7.07 (1H, ddd, J 0.9, 7.2, 7.7, ArH), 7.20 (1H, ddd, J 1.2, 7.7,
7.7, ArH), 7.42 (1H, d, J 7.2, ArH), 8.16 (1H, br s, NH); δC 38.1
(t), 41.9 (q), 53.9 (s), 56.9 (t), 66.2 (t), 109.8 (d), 123.1 (d), 123.5
(d), 128.0 (d), 136.1 (s), 140.4 (s), 183.3 (s); m/z (EI) 202 (40%,
Mϩ), 145 (25), 130 (28), 57 (100) and 51 (12).
9 Y. Kondo, S. Kojima and T. Sakamoto, J. Org. Chem., 1997, 62,
6507.
10 L. F. Tietze and T. Grote, Chem. Ber., 1993, 126, 2733.
11 B. R. Baker, R. E. Schaub and J. H. Williams, J. Org. Chem., 1952,
17, 116.
12 S. E. Webber, K. Okano, T. L. Little, S. H. Reich, Y. Xin,
S. A. Fuhrman, D. A. Matthews, R. A. Love, T. F. Hendrickson,
A. K. Patick, J. W. Meador, R. A. Ferre, E. L. Brown, C. E. Ford,
S. L. Binford and S. T. Worland, J. Med. Chem., 1998, 41, 2786 note
that commercial lithium borohydride was not tried in this
experiment.
Acknowledgements
We thank: the Royal Society for a Leverhulme Senior Research
Fellowship to JAM; University of Strathclyde for a studentship
to D. L. and EPSRC Mass Spectrometry Service Swansea for
mass spectra.
13 K. Jones, S. A. Brunton and R. Gosain, Tetrahedron Lett., 1999, 40,
8935.
14 A. R. Lee, W. H. Huang, T. L. Lin, K. M. Shih and H. F. Lee,
J. Heterocycl. Chem., 1995, 32, 1.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 1 7 – 1 2 2
122