Observations arising from a Beckmann rearrangement-Mannich cyclization approach to the azepinobisindole alkaloid iheyamine A

Article abstract of DOI:10.1016/j.tet.2017.05.093

An overview of an iterative Beckmann rearrangement-Mannich cyclization approach to the azepinobisindole alkaloid iheyamine A is described. In a preliminary model study, the (E)-oxime 10 underwent Beckmann rearrangement to give the bisindolylacetamide 4 followed by an intramolecular Mannich cyclization affording 2-(indolin-2-yl)indole 5 containing the heterocyclic framework of the iheyamine alkaloids. However, the 2-(indolin-2-yl)indole 5 could not be converted into the azepinobisindole core of iheyamine A. When the same Beckmann-Mannich approach was applied toward the natural product itself, a result was obtained that contrasted the model study. The (E)-oxime 3 did not undergo Beckmann rearrangement, but instead an intramolecular Mannich cyclization whereby the electron rich C4 site attacked the intermediate iminum ion, generating the 4-(indolin-2-yl)indole 25 bearing the heterocyclic framework of the slime mould pigment arcyriacyanin A. Although this route did not result in the synthesis of iheyamine A being accomplished, some interesting observations related to the venerable Beckmann rearrangement and Mannich cyclization reactions are described.

Full text of DOI:10.1016/j.tet.2017.05.093

Tetrahedron 73 (2017) 4355e4362
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Observations arising from a Beckmann rearrangement-Mannich
cyclization approach to the azepinobisindole alkaloid iheyamine A
Ashley C. Lindsay, Jonathan Sperry*
School of Chemical Sciences, University of Auckland, 23 Symonds St., Auckland, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
An overview of an iterative Beckmann rearrangement-Mannich cyclization approach to the azepinobi-
sindole alkaloid iheyamine A is described. In a preliminary model study, the (E)-oxime 10 underwent
Beckmann rearrangement to give the bisindolylacetamide 4 followed by an intramolecular Mannich
cyclization affording 2-(indolin-2-yl)indole 5 containing the heterocyclic framework of the iheyamine
alkaloids. However, the 2-(indolin-2-yl)indole 5 could not be converted into the azepinobisindole core of
iheyamine A. When the same Beckmann-Mannich approach was applied toward the natural product
itself, a result was obtained that contrasted the model study. The (E)-oxime 3 did not undergo Beckmann
rearrangement, but instead an intramolecular Mannich cyclization whereby the electron rich C4 site
attacked the intermediate iminum ion, generating the 4-(indolin-2-yl)indole 25 bearing the heterocyclic
framework of the slime mould pigment arcyriacyanin A. Although this route did not result in the syn-
thesis of iheyamine A being accomplished, some interesting observations related to the venerable
Beckmann rearrangement and Mannich cyclization reactions are described.
Received 13 April 2017
Received in revised form
20 May 2017
Accepted 30 May 2017
Available online 31 May 2017
Keywords:
Iheyamine
Beckmann rearrangement
Mannich cyclization
Alkaloid
© 2017 Elsevier Ltd. All rights reserved.
1. Introduction
commence with a Mannich cyclization followed by an intra-
molecular Beckmann rearrangement.
Iheyamines A (1) and B (2) were isolated from the ascidian
Polycitorella sp. collected off Iheya island, Okinawa (Fig. 1).1 The
unique azepinobisindole scaffold present in these alkaloids
inspired our research group to recently complete a total synthesis
of iheyamine A using a route featuring a cross-Mannich reaction to
forge the 2,20-bisindole bond.2 Prior to this successful synthesis, a
distinct synthetic route to iheyamine A was pursued, the details of
which are reported herein.
2. Results and discussion
In order to gauge the viability of the proposal outlined in
Our initial plan was to develop a procedure to simultaneously
integrate nitrogen into the indole 3-position and build the azepine
heterocycle in a one-pot operation (Scheme 1). Along these lines, it
was predicted that under acidic conditions, (E)-oxime 3 would
undergo
a
stereospecific Beckmann rearrangement3 to form
bisindolylacetamide 4 followed by an intramolecular Mannich
cyclization4,5 to give the 2-(indolin-2-yl)indole 5. Given that 5
contained the complete skeleton of iheyamine A, its subsequent
transformation into the natural product was predicted to require
facile redox chemistry. The route in Scheme 1 depicts the Beck-
mann rearrangement occurring first; the process could also
* Corresponding author.
E-mail address: j.sperry@auckland.ac.nz (J. Sperry).
Fig. 1. Iheyamines A (1) and B (2).
http://dx.doi.org/10.1016/j.tet.2017.05.093
0040-4020/© 2017 Elsevier Ltd. All rights reserved.

4356
A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
Scheme 1, a model study was initiated targeting the azepinobi-
sindole core 6 of iheyamine A (Scheme 2). The known bis(indolyl)
ketone 76 was converted to the thermodynamically favoured (E)-8,
with the oxime geometry confirmed by NOE analysis.7 Given that
both the Beckmann rearrangement and Mannich cyclization can be
promoted by TFA,3,4,8 we subjected (E)-8 to TFA at room tempera-
ture in an effort to effect the proposed Beckmann-Mannich cascade,
which led to the bisindolylacetamide 9 rapidly being formed. The
product 9 is a result of the Beckmann rearrangement proceeding
with the desired regioselectivity (migration of the anti-indole
heterocycle), as confirmed by NOE analysis.7 Upon subjecting the
bisindolylacetamide 9 to TFA at 100 ꢀC, the intramolecular Mannich
reaction occurred to give the desired 2-(indolin-2-yl)indole 10.
Somewhat disappointingly, although the Beckmann rearrangement
and Mannich cyclization both proceeded in neat TFA, we
were unable to effect both of these reactions in a one-pot operation.
For example, when the oxime 8 had undergone Beckmann rear-
rangement to 9 (as indicated by TLC analysis), heating the reaction
to initiate the Mannich cyclization led to degradation.
The desired regiochemical outcome for the Mannich reaction (9
to 10) is worthy of comment and can be attributed to the choice of
an amide (9) as the substrate for this reaction. This feature of the
synthetic plan was guided by Bremner and co-workers' attempted
biomimetic synthesis of the N-methyl iheyamine A core 11 (Scheme
3, A).9 Bremner's approach required CeC bond migration to occur
during the acid-mediated Plancher rearrangement10,11 of spirocycle
11, generating the carbocation 12 and hence the desired N-meth-
ylazepinobisindole 13. However, CeN bond migration was fav-
oured, resulting in the isomeric azepinobisindole 15. In this
instance, we posit the CeC bond migration did not occur as
carbocation 12 is not resonance stabilised by the adjacent amine,
which would be protonated under the reaction conditions.
Conversely, CeN bond migration generated carbocation 14 stabi-
lised by two aromatic systems. In our approach (Scheme 3, B),
protonation of bisindolylacetamide 9 initiated an intramolecular
Mannich cyclization to give the spirocycle 16 analogous to
Bremner's intermediate 11. In this instance, 16 undergoes CeC bond
migration to give the desired 2-(indolin-2-yl)indole 10 via the
carbocation 17, which would experience resonance stabilization
from the adjacent amide. CeN bond migration in 16 is also less
favoured as carbocation 18 is only stabilised by one indole
(compared to two in carbocation 14) and as such, the undesired
regioisomer 19 was not observed.
With the complete heterocyclic framework of the iheyamines
assembled, we set out to convert 2-(indolin-2-yl)indole 10 into the
azepinobisindole 6 using standard redox transformations (Scheme
4). This proved much harder than anticipated; attempted reduction
of the lactam in 10 failed to give 20 when using a variety of reducing
agents including lithium aluminium hydride, diisobutylaluminium
hydride, borane and alane, a result that was attributed to the
instability of compound 10. It was thought that dehydrogenation of
the indoline in 10 would give a more stable 2,20-bisindole 21, that
upon lactam reduction would enable access to the azepinobisindole
6. Despite careful treatment of 10 with one equivalent of DDQ, we
were unable to prevent 22 being rapidly formed as a result of facile
aromatization; the 2,20-bisindole 21 was never observed. Subject-
ing 22 to the reducing agents described previously led to degra-
dation and all attempts to access 6 from 22 via the imidoyl
chloride12 also failed.
The Beckmann-Mannich approach toward the synthesis of
Scheme 1. Proposed Beckmann-Mannich approach to iheyamine A (1).

A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
4357
Scheme 2. [A] Synthesis of 2-(indolin-2-yl)indole 10 by sequential Beckmann rearrangement-Mannich cyclization; [B] key NOE correlations in 8, 9 and 10.
iheyamine A was undertaken regardless of its failure on a model
system (Scheme 5). Indole-3-acetic acid (IAA) was converted to the
acid chloride that was added to a freshly prepared solution of 5-
methoxyindolylmagnesium bromide to give the unstable bis(in-
dolyl)ketone 24, which was readily transformed into (E)-oxime 3,
the geometry of which was confirmed by NOE analysis.8 Although
the yield of 3 is only 15% from IAA, greater than 500 mg batches of 3
could be quickly prepared. It was assumed that 3 would undergo a
Beckmann rearrangement in an analogous fashion to the model
study and provide bisindolylacetamide 4. Upon subjecting oxime 3
to TFA at room temperature, the reaction was much slower than in
the model study (18 h vs 1 h). A single product was formed, but it
was readily apparent that it was not the bisindolylacetamide 4. A
sharp singlet at in the 1H NMR spectrum at 10.54 ppm inferred the
oxime was still present. Furthermore, the 1H NMR spectrum con-
tained peaks indicative of an indoline, suggesting a Mannich
cyclization had occurred. However, the indole C2eH peak was still
present in the 1H NMR spectrum, suggesting the product was 4-
(indolin-2-yl)indole 25, which was subsequently confirmed by X-
ray analysis (Fig. 2).13 Interestingly, compound 25 possesses the
same heteroaromatic framework present in the slime mould
pigment arcyriacyanin A,14 an inhibitor of protein tyrosine kinase
and protein kinase C. The transformation of 3 into 25 is postulated
to be the first reported intramolecular formation of a 2,40-bisindole
bond.15
product 25 via 27. In Path B, attack by the indole C3 would form the
spirocyclic intermediate 28, which following rearrangement gen-
ꢀ
erates the 4-(indolin-2-yl)indole 25. Szantay and co-workers have
proposed that the synthesis of azepino[5,4,3-cd]indoles from
tryptamines and formaldehyde follows the latter mechanism.16
3. Conclusions
To conclude, an overview of a Beckmann rearrangement-
Mannich cyclization approach towards iheyamine A is presented.
In a preliminary model study targeting the azepinobisindole core of
the natural product, the (E)-oxime 10 underwent a stereoselective
Beckmann rearrangement to give the bisindolylacetamide 4 fol-
lowed by an intramolecular Mannich cyclization to give the 2-
(indolin-2-yl)indole
5 containing the complete heterocyclic
framework of the iheyamines. Unfortunately, the 2-(indolin-2-yl)
indole 5 could not be converted into the azepinobisindole core of
iheyamine A despite attempting several different redox trans-
formations. When the same Beckmann-Mannich approach was
applied to the natural product itself, a result that contrasted the
model study was obtained. The (E)-oxime 3 did not undergo
Beckmann rearrangement, but instead an intramolecular Mannich
cyclization whereby the electron rich C4 site attacked the inter-
mediate iminum ion, generating the 4-(indolin-2-yl)indole 25
bearing the scaffold present in the slime mould pigment arcyria-
flavin A. Although these studies did not result in a successful syn-
thesis of iheyamine A, the results we encountered ultimately
guided the subsequent total synthesis of iheyamine A.2
The formation of 4-(indolin-2-yl)indole 25 is rationalised in
Scheme 6. Protonation of 3 generates iminium ion 26 which could
feasibly proceed through two separate mechanistic pathways.
Direct attack from the electron rich C4 site (path A) would give the

4358
A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
Scheme 3. [A] CeN bond migration in 119; [B] CeC bond migration in 16.
3.1. Experimental section
29.9 ppm) chloroform (d 77.1 ppm), DMSO (d 39.5 ppm) or meth-
anol (
d
49.0 ppm) peaks. 13C NMR values are reported as chemical
and assignment. 1H NMR shift values are reported as
All reactions were carried out in oven-dried glassware under a
nitrogen atmosphere unless otherwise stated. Analytical thin layer
chromatography was performed using 0.2 mm silica plates and
compounds were visualized under 365 nm ultraviolet irradiation
followed by staining with either alkaline permanganate or etha-
nolic vanillin solution. Melting points were recorded on a melting
point apparatus and are uncorrected. Infrared (IR) spectra were
recorded on a Perkin Elmer Spectrum 100 FT-IR spectrometer using
a diamond ATR sampling accessory and absorption maxima are
expressed in wavenumbers (cmꢁ1). NMR spectra were recorded as
indicated on an NMR spectrometer operating at 500, 400 and
300 MHz for 1H nuclei and 125, 100 and 75 MHz for 13C nuclei.
Chemical shifts are reported in parts per million (ppm) relative to
shift
d
chemical shift d, relative integral, multiplicity (s, singlet; d, doublet;
t, triplet; q, quartet; m, multiplet), coupling constant (J in Hz) and
assignment. Assignments are made with the aid of DEPT 90, DEPT
135, COSY, NOESY and HSQC experiments. High resolution mass
spectra were obtained by electrospray ionization in positive ion
mode at a nominal accelerating voltage of 70 eV on a microTOF
mass spectrometer.
3.1.1. (E)-1,2-Di(indol-3-yl)ethanone oxime (8)
Hydroxylamine hydrochloride (65 mg, 0.94 mmol) was added to
a stirred solution of bis(indolyl)ketone 76 (87 mg, 0.32 mmol) in
pyridine (1 mL) and the resulting mixture was stirred at r.t. for 18 h.
Ethyl acetate (20 mL) was added and the mixture washed with
copper sulfate solution (10% aq., 5 ꢂ 20 mL). The organic extract
was dried (Na2SO4), filtered and concentrated in vacuo. Purification
by flash chromatography on silica gel eluting with hexanes-ethyl
the tetramethylsilane peak recorded as
solvent, or the residual acetone ( 2.05 ppm), chloroform (
7.24 ppm), DMSO ( 2.50 ppm) or methanol ( 3.31 ppm) peaks. The
13C NMR values were referenced to the residual acetone (
d 0.00 ppm in CDCl3/TMS
d
d
d
d
d

A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
4359
Scheme 4. Attempted conversion of 10 into azepinobisindole 6.
Scheme 5. Unexpected formation of 25 from oxime (E)-3.

4360
A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
m, 3 ꢂ CH), 6.96 (1 H, t, J 7.5, CH), 4.18 (2 H, s, CH2); dC (100 MHz,
(CD3)2SO) 153.7 (C), 136.7 (C), 135.9 (C), 127.1 (C), 126.4 (CH), 124.7
(C), 123.5 (CH), 122.5 (CH), 121.7 (CH), 120.7 (CH), 119.6 (CH), 118.8
(CH),118.1 (CH), 112.2 (C), 111.4 (CH), 111.2 (CH), 111.0 (C), 21.7 (CH2).
3.1.2. N,2-Di(indol-3-yl)acetamide (9)
A solution of oxime 8 (252 mg, 0.87 mmol) in trifluoroacetic acid
(2 mL) was stirred for 1 h at r.t. The reaction mixture was diluted
with ethyl acetate (20 mL), washed with saturated sodium
hydrogen carbonate (3 ꢂ 20 mL), dried (Na2SO4), filtered and
concentrated in vacuo. Purification by flash chromatography on
silica gel eluting with hexanes-ethyl acetate (1:1) gave the title
compound (102 mg, 0.35 mmol, 41%) as a pale orange oil. HRMS [ESI,
(M þ Na)þ] found 312.1109 [C18H15N3O þ Na]þ requires 312.1107;
vmax/cmꢁ1 (neat): 3395, 1892, 1702, 1419; dH (400 MHz, (CD3)2SO)
11.38 (1 H, br s, NH), 10.65 (1 H, br s, NH), 10.61 (1 H, br s, NH), 7.96
(1 H, d, J 2.6, CH), 7.72 (1 H, d, J 8.1, CH), 7.65 (1 H, d, J 8.1, CH), 7.35
(1 H, d, J 8.0, CH), 7.27 (1 H, d, J 8.0, CH), 7.07e7.01 (2 H, m, 2 ꢂ CH),
6.98e6.92 (3 H, m, 3 ꢂ CH), 4.10 (2 H, s, CH2); dC (100 MHz,
(CD3)2SO) 149.7 (C), 136.1 (C), 135.6 (C), 128.7 (CH), 127.4 (C), 125.6
(C), 123.0 (CH), 122.1 (CH), 121.0 (CH), 120.8 (CH), 119.1 (CH), 118.7
(CH), 118.2 (CH), 111.5 (C), 111.4 (CH), 111.2 (CH), 108.5 (C), 31.2
(CH2).
Fig. 2. ORTEP representation of 25.13
acetate (1:1) gave the title compound (79 mg, 0.27 mmol, 87%) asþa
brown solid, M.p. 179.5e181.8 ꢀC (charred); HRMS [ESI, (M þ H) ]
found 290.1281 [C18H15N3O þ H]þ requires 290.1288; vmax/cmꢁ1
(neat) 3395, 3057, 1705, 1241, 1043; dH (400 MHz, (CD3)2SO) 11.20
(1 H, br s, NH), 10.75 (1 H, br s, NH), 10.74 (1 H, s, OH), 8.15 (1 H, d, J
8.1, CH), 7.79 (1 H, d, J 8.1, CH), 7.74 (1 H, d, J 2.8, CH), 7.35 (1 H, d, J
7.8, CH), 7.30 (1 H, d, J 7.8, CH), 7.17 (1 H, d, J 2.3, CH), 7.11e7.00 (3 H,
3.1.3. ( )-7,7a,12,12a-Tetrahydro-5H-azepino[3,2-b:4,5-b0]diindol-
6(13H)-one (10)
A solution of bisindolylacetamide 9 (84 mg, 0.3 mmol) in tri-
fluoroacetic acid (2 mL) was stirred at 100 ꢀC for 18 h. The reaction
Scheme 6. Mechanistic considerations for the formation of 4-(indolin-2-yl)indole 25.

A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
4361
was cooled, diluted with ethyl acetate (10 mL) and carefully neu-
tralised with saturated sodium hydrogen carbonate (20 mL). The
organic phase was washed with saturated sodium hydrogen car-
bonate (3 ꢂ 20 mL), dried (Na2SO4), filtered and concentrated in
vacuo. Purification by flash chromatography on silica gel eluting
with ethyl acetate-hexanes (1:1) gave the title compound (22 mg,
0.08 mmol, 26%) as a brown oil. HRMS [ESI, (M þ Na)þ] found
312.1118 [C18H15N3O þ Na]þ requires 312.1107; vmax/cmꢁ1 (neat)
3240, 2924, 1682, 1454, 740; dH (400 MHz, (CD3)2CO) 10.33 (1 H, br
s, NH), 9.65 (1 H, s, NH), 7.98 (1 H, d, J 7.9, CH), 7.34 (1 H, dd, J 7.9, 0.8,
CH), 7.16 (1 H, d, J 7.0, CH), 7.12e7.09 (1 H, m, CH), 7.03e6.99 (1 H, m,
CH), 6.92e6.88 (1 H, m, CH), 6.64 (1 H, t, J 7.0, CH), 6.54 (1 H, d, J 7.9,
CH), 5.40 (1 H, br s, NH), 5.06 (1 H, d, J 7.9, CH), 3.86e3.83 (1 H, m,
CH), 3.53 (1 H, dd, J 17.0, 5.4, CH2), 2.96 (1 H, dd, J 17.0, 5.4, CH2); dC
(100 MHz, (CD3)2CO) 148.9 (C), 128.5 (CH), 126.1 (CH), 123.9 (CH),
123.2 (CH), 123.1 (CH), 120.94 (2 ꢂ C), 120.90 (C), 119.32 (C), 119.30
(C), 119.2 (C), 111.9 (CH), 110.4 (2 ꢂ CH), 60.5 (CH), 41.7 (CH), 23.8
(CH2).
acetate (1:1) gave the title compound (584 mg, 1.72 mmol, 16% froþm
IAA) as a brown solid, M.p. 184.4e185.8 ꢀC; HRMS [ESI, (M þ Na) ]
found 342.1211 [C19H17N3O2 þ H]þ requires 342.1213; vmax/cmꢁ1
(neat) 3402. 2914, 1620, 1484, 1423; dH (400 MHz, (CD3)2SO) 11.07
(1 H, br s, NH), 10.74 (1 H, br s, NH), 10.69 (1 H, s, OH), 7.78 (1 H, d, J
8.0, CH), 7.70 (1 H, d, J 2.7, CH), 7.66 (1 H, d, J 2.7, CH), 7.31 (1 H, d, J
8.0, CH), 7.26 (1 H, d, J 8.8, CH), 7.15 (1 H, d, J 2.2, CH), 7.03e7.01 (1 H,
m, CH), 6.97e6.95 (1 H, m, CH), 6.75 (1 H, dd, J 8.8, 2.7, CH), 4.16
(2 H, s, CH2), 3.73 (3 H, s, Me); dC (100 MHz, (CD3)2SO) 153.9 (2 ꢂ C),
135.9 (C), 131.7 (C), 127.1 (C), 127.0 (CH), 125.1 (C), 123.5 (CH), 120.7
(CH), 118.8 (CH), 118.1 (CH), 112.0 (CH), 111.9 (C), 111.8 (CH), 111.2
(CH), 111.0 (C), 104.3 (CH), 55.3 (Me), 21.7 (CH2).
3.1.6. ( )-10-Methoxy-4,4a,9,9a-tetrahydrocyclohepta[1,2-b:5,6,7-
c0d0]diindol-3(1H)-one oxime (25)
A solution of oxime 3 (189 mg, 0.529 mmol) in trifluoroacetic
acid (4 mL) was stirred at r.t. for 18 h. The reaction mixture was
carefully neutralised (sat. sodium hydrogen carbonate) and
extracted with ethyl acetate (3 ꢂ 10 mL). The organic extracts were
combined, dried (Na2SO4), filtered and concentrated in vacuo. Pu-
rification by flash chromatography on silica gel eluting with
toluene-acetone (7:3) gave the title compound (134 mg, 0.42 mmol,
71%) as a brown solid, M.p. 156.1e159.3 ꢀC (charred); HRMS [ESI,
(M þ Na)þ] found 342.1219 [C19H17N3O2 þ H]þ requires 342.1213;
vmax/cmꢁ1 (neat) 3147, 3048, 1608, 1461, 1227; dH (400 MHz,
(CD3)2SO) 11.48 (1 H, br s, NH), 10.54 (1 H, s, OH), 8.39 (1 H, d, J 3.0,
CH), 7.41 (1 H, d, J 8.7, CH), 7.18 (1 H, d, J 7.5, CH), 7.04 (1 H, d, J 8.7,
CH), 6.96 (1 H, t, J 7.5, CH), 6.70e6.63 (2 H, m, 2 ꢂ CH), 5.60 (1 H, d, J
2.6, CH), 5.17 (1 H, dd, J 7.1, 2.6, CH), 3.90 (3 H, s, Me), 3.49 (1 H, t, J
8.0, CH2), 2.98 (1 H, dd, J 14.0, 10.5, CH2), 1 ꢂ NH not observed; dC
(100 MHz, (CD3)2SO) 153.0 (C), 150.4 (C), 148.7 (C), 131.3 (CH), 130.8
(C), 128.9 (C), 128.2 (C), 127.0 (CH), 124.1 (C), 123.7 (CH), 117.9 (CH),
111.1 (CH), 109.3 (CH), 107.6 (C), 107.4 (CH), 61.4 (CH), 56.5 (Me),
43.4 (CH), 36.3 (CH2).
3.1.4. 5H -Azepino[3,2-b:4,5-b0]diindol-6(13H)-one (22)
To a solution of 2-(indolin-2-yl)indole 10 (23 mg, 0.08 mmol) in
dioxane (1 mL) at 0 ꢀC was carefully added 2,3-dichloro-5,6-
dicyano-1,4-benzoquinone (17 mg, 0.07 mmol) portionwise. The
resulting reaction mixture was warmed to r.t. and stirred for 5 min,
then diluted with ethyl acetate (10 mL). The organic phase was
washed with saturated sodium hydrogen carbonate (5 ꢂ 10 mL),
dried (Na2SO4), filtered and concentrated in vacuo. Purification by
flash chromatography on silica gel eluting with ethyl acetate-
hexanes (2:1) gave the title compound (7 mg, 0.02 mmolþ, 32%) as
a red solid, M.p. 270.2e274.5 ꢀC; HRMS [ESI, (M þ H) ] found
286.0977 [C18H11N3O þ H]þ requires 286.0975; vmax/cmꢁ1 3373,
2955, 2916, 2166, 1736; dH (400 MHz, (CD3)2SO) 11.95 (1 H, br s,
NH), 11.70 (1 H, br s, NH), 8.91 (1 H, d, J 8.0, CH), 8.41 (2 H, d, J 8.0,
2 ꢂ CH), 7.86e7.82 (2 H, t, J 9.2, 2 ꢂ CH), 7.61e7.53 (2 H, m, 2 ꢂ CH),
7.39e7.31 (2 H, m, 2 ꢂ CH); dC (100 MHz, (CD3)2SO) 177.7 (C), 152.8
(C), 140.32 (C), 140.29 (C), 126.7 (CH), 126.6 (CH), 125.21 (C), 125.18
(C), 124.4 (CH), 122.8 (C), 122.2 (C), 121.1 (CH), 121.0 (C), 120.9 (2 ꢂ
CH), 119.9 (CH), 112.8 (CH), 112.2 (CH).
Acknowledgements
We thank the Royal Society of New Zealand for the award of a
Rutherford Discovery Fellowship (J.S) and the University of Auck-
land for a Doctoral Scholarship (A. C. L).
3.1.5. (E)-2-(Indol-3-yl)-1-(5-methoxyindol-3-yl)ethanone oxime
(3)
Thionyl chloride (0.83 mL, 11.43 mmol) was added dropwise to a
stirred suspension of indole-3-acetic acid (IAA, 2 g, 11.4 mmol) in
toluene (20 mL) at r.t. The reaction mixture was stirred at r.t. for
30 min, 40 ꢀC for 1.5 h, cooled to r.t. and concentrated in vacuo to
give indole-3-acetyl chloride.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2017.05.093.
To a solution of methylmagnesium bromide (3.4 mL, 1 M in THF)
in ether (10 mL) at r.t was added a solution of 5-methoxyindole
(1.5 g, 10.2 mmol) in ether (10 mL) dropwise. The reaction
mixture was cooled to ꢁ10 ꢀC and a solution of indole-3-acetyl
chloride in ether (20 mL) was added quickly in one portion. The
reaction mixture was warmed to r.t. and stirred for 16 h. Water
(100 mL) was added and the mixture extracted with ethyl acetate
(3 ꢂ 100 mL). The organic phases were dried (Na2SO4), filtered and
concentrated in vacuo. Purification by flash chromatography on
silica gel eluting with hexanes-ethyl acetate (1:1) gave the ketone
24 (731 mg, 2.85 mmol, 25%) as a brown solid that was used
immediately in the next step.
Hydroxylamine hydrochloride (500 mg, 7.2 mmol) was added to
a stirred solution of bis(indolyl)ketone 22 (731 mg, 2.85 mmol) in
pyridine (10 mL) and the reaction mixture stirred at r.t. for 18 h.
Ethyl acetate (50 mL) was added and the mixture washed with a
solution of copper sulfate (10%; 5 ꢂ 50 mL). The organic extract was
dried (Na2SO4), filtered and concentrated in vacuo. Purification by
flash chromatography on silica gel eluting with hexanes-ethyl
References
1. Sasaki T, Ohtani II, Tanaka J, Higa T. Tetrahedron Lett. 1999;40:303.
2. Lindsay AC, Leung IKH, Sperry J. Org Lett. 2016;18:5404.
3. For the synthesis of 3-amidoindoles using the Beckmann rearrangement, see:
(a) Albini FM, Oberti R, Caramella P. J Chem Res, Synop. 1983;1:4;
(b) Qu J, Kumar N, Alamgir M, Black DS. Tetrahedron Lett. 2009;50:5628;
(c) Augustine JK, Kumar R, Bombrun A, Mandal AB. Tetrahedron Lett. 2011;9:
1074.
4. For intermolecular examples, see: (a) Smith GF, Walters AE. J Chem Soc. 1961:
940;
(b) Bergman J, Koch E, Pelcman B. Tetrahedron Lett. 1995;36:3945;
(c) Hao X, Liu S. Tetrahedron Lett. 2011;52:5640;
(d) Xu M, An R, Huang T, Hao X, Liu S. Tetrahedron Lett. 2016;57:1247.
5. For intramolecular examples, see: (a) Gilbert EJ, Van Vranken DL. J Am Chem
Soc. 1996;118:5500;
(b) Chisholm JD, Van Vranken D. J Org Chem. 2000;65:7541;
(c) Bergman J, Koch E, Pelcman B. J Chem Soc, Perkin Trans. 2000;1:2609.
6. Kamenskii AB, Smushkevich YI, Livshits AI, Suvorov NN. Chem Heterocycl
Compd. 1980;16:741.
7. See supporting information for full details.
8. For a mechanistic study on the TFA-mediated Beckmann rearrangement, see:
Ronchin L, Vavasori A J Mol Catal A Chem. 2009;313:22.
9. Bremner J, Sengpracha W, Skelton BW. Synth Comm. 2008;38:1931.

4362
A.C. Lindsay, J. Sperry / Tetrahedron 73 (2017) 4355e4362
10. (a) Plancher G. Gazz Chim Ital. 1898;28:374;
fax: þ44(1223)336033; or deposit@ccdc.cam.ac.uk].
(b) Plancher G. Atti Accad Lincei. 1900;9:115;
(c) Boyd-Barrett HS. J Chem Soc. 1932:321.
14. (a) Gill M, Steglich W. Progr Chem Org Nat Prod. 1987;51:216;
(b) Steglich W. Pure Appl Chem. 1989;61:281.
11. For some recent applications of the Plancher rearrangement, see: (a)
Caballero E, Avendano C, Menendez JC. J Org Chem. 2003;68:6944;
(b) Liu KG, Robichaud AJ, Lo JR, Mattes JF, Cai Y. Org Lett. 2006;8:5769;
(c) Eisenbeis SA, Phillips JR, Rescek D, Oyola-Cintron Y. Tetrahedron Lett.
2010;51:4303;
(d) Loh C,C, Raabe G, Enders D. Chem Eur J. 2012;18:13250.
12. Alper H, Tanaka M. Synthesis. 1978:781.
13. Crystallographic data (reference number CCDC 1521668) can be obtained free
of charge via www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cam-
bridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
15. For the synthesis of 2,40-bisindoles, see: (a) Itoh S, Ogino M, Haranou S, et al.
J Am Chem Soc. 1995;117:1485;
(b) Murase M, Watanabe K, Kurihara T, Tobinaga S. Chem Pharm Bull. 1998;46:
889;
(c) Mayer G, Hinze C, Polborn K, Steglich W. Aus J Chem. 2004;57:625;
(d) Kraus GA, Guo H. Org Lett. 2008;10:3061;
(e) Liu F, Movassaghi M. Tetrahedron Lett. 2015;56:2995.
ꢀ
16. Novak L, Hanania M, Kovacs P, Rohaly J, Kolonits P, Szantay C. Heterocycles.
1997;45:2331.