Full Paper
opportunities for structural variation.[6] We report here full de-
tails of the development of this process, from the first exam-
ples of Stille cross-coupling in this ynamide cascade (M=SnR3),
to the evolution and subtleties of the Suzuki alternative, in-
cluding the first use of potassium alkenyltrifluoroborate salts[7]
in any carbopalladation/cross-coupling sequence (M=BF3K).[8]
We also discuss the development of a reductive cyclization[9]
that affords monocyclic amidodienes 5—products which
through [4+2] cycloadditions (including hetero-Diels–Alder re-
actions with dienophiles 6) lead to heterobicyclic frameworks
7, of potential use in medicinal chemistry. Finally, alongside
some mechanistic observations and the exploration of
equivalent reactions of ynhydrazides, the formation of an
unprecedented 7,4-fused ring enamide 8 by a formal 4-endo-
trig cyclization is described.[10]
Results and Discussion
A prerequisite for the realization of these processes was the
synthesis of ynamides featuring a bromoalkene substituent.
Among many methods for ynamide synthesis,[11] we were par-
ticularly attracted to Hsung’s robust copper-catalyzed coupling
of amides and bromoalkynes, which has been shown to
tolerate a bromoalkene.[11f] Our work thus began with the
preparation of a range of bromoalkenyl amides (Scheme 2). To
prepare the ‘parent’ bromoalkene sulfonamide 9a, we initially
employed a two-directional allylation of aqueous glyoxal with
2,3-dibromopropene[12] using the procedure of Otera,[13] which
gave an intermediate diol as an inconsequential mixture of dia-
stereomers in quantitative yield. Periodate cleavage of this diol
followed by in situ reduction provided alcohol 10a, a com-
pound that proved somewhat unstable toward prolonged stor-
age or distillation. A superior method for the preparation of
10a involved direct bromoallylation of aqueous formaldehyde,
from which 10a was obtained in quantitative yield and suffi-
cient purity to be employed directly in subsequent chemistry.
Otera’s chemistry also proved suitable for the bromoallylation
of other aldehydes, leading to the secondary alcohols 10b–c,
which together with 10a were converted to the sulfonamides
9a–c by Mitsunobu reaction with TsBocNH, then tert-butoxy-
carbonyl (Boc) deprotection using trifluoroacetic acid. We also
targeted the preparation of other amides, and after some
optimization found that amine salt 11 could be prepared from
10a by tosylation, azide displacement, and Staudinger
reduction. This was converted to the trifluoroacetyl, Boc,
and methoxycarbonyl derivatives 9d–f.
Scheme 2. a) 2,3-dibromopropene (2.3 equiv), Sn (2.3 equiv), HBr (48% aq.,
few drops), Et2O/H2O (1:1); b) NaIO4, MeOH/pH 7 buffer (5:1); NaBH4, 08C!
RT; c) 2,3-dibromopropene (1.3 equiv), Sn (1.3 equiv), HBr (48% aq., few
drops), Et2O/H2O (1:1); d) TsBocNH, PPh3, DIAD, THF, 08C!RT; e) CF3CO2H,
CH2Cl2, 08C; f) TsCl, py, CH2Cl2; g) NaN3, DMSO; h) H2S, pyridine/H2O (1:1);
AcOH; i) Boc2O, Et3N, THF; j) MeOCOCl, py, CH2Cl2; k) TFAA, py, CH2Cl2; l) BBr3,
CH2Cl2; AcOH; m) TBSCl, imid., CH2Cl2; n) TsCl, Et3N, CH2Cl2; o) 2,3-dibromo-
propene, 50% NaOH (aq.), Bu4NHSO4 (0.1 equiv), tol/H2O (1:1); p) TBAF, THF,
08C; r) 2,3-dibromopropene, 25% NaOH (aq.), Bu4NHSO4 (0.1 equiv), Bu4NI
(0.1 equiv), tol/H2O (1:1). Procedures i)–k) were preceded by basic extraction
of 11 using NaOH/CH2Cl2. DIAD=diisopropyl azodicarboxylate; TFAA =tri-
fluoroacetic anhydride; TBAF=tetra-n-butylammonium fluoride.
We next turned to the synthesis of bromoalkyne coupling
partners for ynamide formation, which were prepared from the
corresponding alkynes either by lithiation/bromine quench
(13a–e, Scheme 3), or preferably and more mildly using N-bro-
mosuccinimide (NBS) and catalytic silver(I) nitrate (13a–f).[14]
The bromoalkynyl indole 13g was conveniently accessed from
3-formylindole 16 via vinyl dibromide formation,[15] then elimi-
nation of HBr using potassium hexamethyldisilazide (KHMDS),
a strategy that also served well for the formation of the Roche
ester-derived bromoalkyne 13h.
With a view to the synthesis of six- and seven-membered
azacycles such as tetrahydroquinolines and benzazepines, we
prepared sulfonamides 9g and 9h in two steps from 10g and
10h by Mitsunobu amination, then bromoboration/protode-
borylation (with in situ Boc deprotection). Sulfonamides 9i and
9j feature an additional nitrogen substituent in the tether, and
thus represent precursors to diazepines and diazocanes; these
were prepared from amino alcohols 10i and 10j, respectively,
using equivalent chemistry. A notable feature of these latter
sequences was the requirement for phase-transfer conditions
to install the 2-bromoallyl group on the sulfonamides.
With a range of amides and bromoalkynes in hand, ynamide
synthesis was addressed using the Hsung method (catalytic
CuSO4/1,10-phenanthroline, K3PO4, toluene, 808C, Table 1).
These conditions indeed proved successful for the formation
of ynamides 3a–h from sulfonamide 9a, with the bromoalkene
surviving unscathed; these ynamides were delivered in good
to excellent yields in all cases, including the indole-substituted
ynamide 3 f, and more complex examples featuring additional
functionality and stereogenic centres (e.g. 3c, 3g, 3h). An ex-
ception was TMS-ynamide 3d, where we experienced varying
degrees of desilylation under the reaction conditions. Disap-
Chem. Eur. J. 2015, 21, 12627 – 12639
12628 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim