Communications
3 in high yield (86%) and enantioselectivity (96% ee). The
oped by Seebach and Hintermann.[10] This chiral oxazolidi-
none impressively demonstrated its usefulness in the synthesis
of discodermolide by chemists at Novartis.[11] The benefits,
including higher selectivity and crystallinity of the intermedi-
ates (with the drawback of increased molecular weight), were
also of great use in the reactions described in this commu-
nication. An enantioselective alkylation of the Li enolate of
10[10] with tigloyl bromide[12] gave 11 in high yield (92%) and
excellent selectivity (d.r. > 97:3, Scheme 2). Cleavage (and
recycling) of the chiral auxiliary by LiAlH4 followed by a
Swern oxidation gave aldehyde 12[13] in 98% yield over two
steps. This chiral aldehyde was then transformed in a boron-
mediated aldol reaction using ent-10 to give the syn-aldol 13.
The selectivity of this reaction (87:13 for the desired isomer,
separable by flash chromatography) is less than perfect but
comparable to reactions of similar substrates in the litera-
ture.[2c] Transformation of compound 13 to the Weinreb amide
was easily accomplished using Al(CH3)3, as were subsequent
TBS protection and reduction by DIBAH to the aldehyde
(85% over three steps). Another boron-mediated aldol
reaction gave hydroxy amide 14 featuring an all-syn config-
uration with excellent stereoselectivity (d.r. > 97:3). The
direct reduction of the auxiliary-bound imide 14 to the
aldehyde 15 was possible by LAH in toluene; this surprising
reaction exemplifies the strengths of the DIOZ auxiliary.
Wittig reaction gave the a,b-unsaturated ester, which was
transformed through reduction and subsequent oxidation to
the a,b-unsaturated aldehyde 16. X-ray crystallographic
analysis of 16 (m.p. 75–778C) allowed for the unambiguous
determination of the configuration of the newly formed
stereogenic centers. The transformation to the vinyl iodide 17
following Takai[14] was possible in excellent yield and stereo-
selectivity.
observed diastereoselectivity of only ca. 5:1 was a conse-
quence of epimerization under the reaction conditions. This
low selectivity was not a problem, as the acetal 3 (as a mixture
of diastereoisomers) was transformed in the presence of acid
in iPrOH[2o] to the thermodynamically more stable, configura-
tionally homogenous product 5[2k] (after deprotection). This
alkyne was hydrozirconated by using Schwartzꢀs reagent and
then transmetalated in situ to give the vinyl zinc species.
Subsequent Negishi cross-coupling with the readily available
dibromide 6[7] gave the trisubstituted vinyl bromide 7 in 81%
yield. Interestingly, the addition of small amounts of DIBAH
consistently resulted in higher yields. However, compound 7
displayed the wrong configuration of the trisubstituted olefin;
as a result a stereoinversion was required in this synthesis.
Negishi and co-workers recently reported that similarly
substituted haloalkenes undergo cross-coupling under inver-
sion (and not under retention).[8] Therefore, vinyl bromide 7
was allowed to react under Pd catalysis with dimethylzinc, and
we observed in the NMR spectrum a clean inversion at the
double bond to the cis compound 8 (68% yield, d.r. > 97:3).
This is even more remarkable, as the reversal of reagents, that
is, first reaction of dibromide 6 with dimethylzinc followed by
the dihydropyran derivative starting from 5, led to low yields
and a mixture of isomers. The mechanism of this stereo-
inversion in the Negishi cross-coupling reaction remains
unknown; in the literature s-bound Pd–allenyl species were
postulated as intermediates.[8] In the context of our research,
the clean inversion of 7 to 8 was of great use. Removal of the
terminal protecting group and transformation of the hydroxy
group to the iodide 9 was carried out under standard
conditions.
We chose an Evans aldol strategy for the synthesis of the
second fragment,[9] but opted for the DIOZ auxiliary (4-
isopropyl-5,5-diphenyloxazolidin-2-one), which was devel-
Having both fragments at hand, we chose to merge them
using
a
procedure developed by Marshall et al.[2g,s]
Scheme 2. a) LDA, THF then tigloyl bromide, 92%, d.r.>97:3; b) LAH, ether, quant. c) Swern oxidation, 99%; d) Bu2BOTf, Et3N, CH2Cl2, then 12,
77%, d.r. 87:13; e) CH3ONHCH3·HCl, Al(CH3)3, CH2Cl2, 86%; f) TBSOTf, 2,6-lutidine, 99%; g) DIBAH, quant.; h) ent-10, Bu2BOTf, Et3N, CH2Cl2,
then aldehyde, 61%, d.r. >97:3; i) LiAlH4, toluene, 83%; j) (carbethoxyethylidene)triphenylphosphorane, toluene, 99%, k) DIBAH, THF, 93%;
l) MnO2, CH2Cl2 86%; m) CrCl2, CHI3, THF, quant., d.r. >97:3.
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8707 –8710
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Take advantage of blue reference links
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&