Stereocontrol by quaternary centres: A stereoselective synthesis of (-)-luminacin D

Article abstract of DOI:10.1002/chem.201304776

Very high diastereoselectivity can be achieved by 1,3-chelation-controlled allylation of aldehydes that possess a non-chelating α-ether substituent, even if the α-position is a quaternary centre and/or a spiro-epoxide. This reaction was used as a key step in an enantioselective synthesis of the angiogenesis inhibitor luminacin D.

Full text of DOI:10.1002/chem.201304776

DOI: 10.1002/chem.201304776
Communication
&
Total Synthesis
Stereocontrol by Quaternary Centres: A Stereoselective Synthesis
of (À)-Luminacin D
Nathan Bartlett,[a] Leona Gross,[a] Florent Pꢀron,[a] Daniel J. Asby,[a] Matthew D. Selby,[b]
Ali Tavassoli,[a] and Bruno Linclau*[a]
There is comparatively little information about the mode of
Abstract: Very high diastereoselectivity can be achieved
by 1,3-chelation-controlled allylation of aldehydes that
action or biological function of luminacin D. Given that it is the
most potent member of this family in several of the originally
reported assays,[3] there is significant potential and need for an
possess a non-chelating a-ether substituent, even if the
a-position is a quaternary centre and/or a spiro-epoxide.
approach that enables the synthesis of sufficient quantities of
This reaction was used as a key step in an enantioselective
this molecule to enable further research into its cellular mode
synthesis of the angiogenesis inhibitor luminacin D.
of action.
There are a few syntheses of the luminacins reported,[7] how-
ever, each with shortcomings in terms of length and/or unse-
Natural products continue to be a robust source of novel ther-
apeutics, with approximately 50% of currently approved anti-
cancer drugs being natural products or their derivatives.[1] The
luminacin family of natural products, discovered from the fer-
mentation broth of the soil bacterium Streptomyces sp.,[2] con-
tains several members that have shown promising anticancer
activity in multiple assays and cell lines. Two members of this
family, luminacin D (1a) and luminacin C2 (1b, also known as
UCS15A), have been shown to be potent inhibitors of angio-
genesis in several in vitro assays.[3] Luminacin D was also
shown to inhibit the proliferation of several cancer cell lines.[3]
Additional studies with luminacin C2 have shown it to be
a protein–protein interaction inhibitor that targets Src signal
transduction by inhibiting the SH3 domain-mediated interac-
tions of Src kinase with its targets, thus preventing the Src-spe-
cific tyrosine phosphorylation of numerous proteins.[4] Src kin-
ases play a key role in the signalling and regulation of multiple
processes associated with cancer, such as cell migration, cell
adhesion, extracellular matrix sensing, cell cycle timing, as well
as several poorly understood events necessary for angiogene-
sis.
lective reaction steps. A particular concern is the epoxide intro-
duction, with three total syntheses featuring a late stage-epox-
idation step with very low, or undesired selectivity.[7b–d,8] Be-
cause of this, we sought to develop a synthetic approach in
which an enantiopure epoxide intermediate is assembled first,
to then utilise its stereochemistry for diastereoselective com-
pletion of the aliphatic portion, from which the luminacins and
the migracins can be synthesised.
Herein we report a successful total synthesis of (À)-lumina-
cin D using this strategy, and report on the excellent diastereo-
control possible by allylation of aldehydes having a-oxygenat-
ed centres, including quaternary centres, under 1,3-chelation
conditions. We also unambiguously show that this type of al-
dehyde addition is consistent with the Cornforth–Evans (CE)
model of stereoinduction.
Hence, conventional disconnection leads to the aliphatic
fragment 2 (Scheme 1). Its construction was envisaged by
spontaneous hemiacetal ring closure, syn-aldol reaction, and
by the key step, diastereoselective allylation controlled by 1,2-
induction of the quaternary epoxide centre of the enantiopure
intermediate 3. This chelation-controlled allylation step was in-
spired by previous work from our group showing excellent
levels of diastereocontrol exerted by all-C quaternary stereo-
centres for allylstannation of 2,2-disubstituted malonalde-
hydes.[9] However, a 1,3-dialdehyde group (c.f. 3b) is not desir-
able in the present case, as it contains four diastereotopic alde-
hyde faces, and our efforts were directed to investigating 3a
and 3c as substrates for the allylation reaction.[10] The synthesis
of the enantiopure epoxides 3a,c was envisioned from sub-
strates 4 and 5.
Luminacin C2 was further demonstrated to inhibit the inva-
sion and metastasis of model breast cancer cell lines in vitro,
by inhibition of the protein–protein interaction of the Src-ho-
mology domain of cortactin with AMAP1.[5] The recent report
that two structurally related compounds, named migracin A
and B (1c), inhibit the migration of a breast cancer cell line,[6]
provides further evidence for the anticancer potential of this
molecule, or its derivatives.
Starting from 6, synthesised in three steps from methyl
acrylate,[7c,11] benzylation and reduction gave 4 as a substrate
for a Sharpless epoxidation, which was followed by alcohol oxi-
dation to give 3a (Scheme 2). The 3-oxopropionate substrate
3c was synthesised from enantiopure 7, which was obtained
in one step from the corresponding menthyl sulfinyl ester.
Two-step Knoevenagel condensation led to 5 as the E isomer
[a] N. Bartlett, L. Gross, F. Pꢀron, Dr. D. J. Asby, Dr. A. Tavassoli, Dr. B. Linclau
Department of Chemistry, University of Southampton
Highfield, Southampton SO17 1BJ (UK)
E-mail: bruno.linclau@soton.ac.uk
[b] M. D. Selby
UCB, 216 Bath Road, Slough, Berkshire, SL1 WE (UK)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304776.
Chem. Eur. J. 2014, 20, 3306 – 3310
3306
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
10–13 was proved by X-ray crys-
tallography of a common deriva-
tive.[14]
The stereochemical outcome
was rationalised as shown in
Figure 1. Chelation, involving the
aldehyde
and
benzyloxy
groups,[15] gives rise to two inter-
converting half-chair structures I
and II. Nucleophilic attack is ex-
pected to proceed via a chair-
like transition state, as indicated.
Scheme 1. Retrosynthetic analysis, with a 1,3-chelation-mediated allylation as key step.
Scheme 3. Diastereoselective allylations of a-epoxyaldehydes.
Scheme 2. Synthesis of the allylation substrates.
only, which was then subjected to Fernandez de la Pradilla’s
vinyl sulfoxide epoxidation methodology.[12] Although replica-
tion of their conditions[12b] led to a low yield and selectivity,
lowering the temperature and concentration of the reaction
gave good diastereocontrol. Interestingly, the phenyl sulfoxide
containing 8 was obtained in a 16:1 ratio of (inseparable) dia-
stereomers,[13] but the corresponding p-tolyl sulfoxide contain-
ing 9 was obtained in a lower ratio (7:1). The relative stereo-
chemistry of 8 was determined by X-ray crystallography after
hydrolysis of the ester group.[14] Finally, lithiation of the sulfox-
ide and treatment with DMF gave 3c, albeit in a moderate
yield. As this operation removes the enantiopure sulfoxide
group, 3c was then obtained in a 7:1 enantiomeric ratio.
The key allylation reactions are shown in Scheme 3. Al-
though treatment of 3 with 3 equiv of MgBr2·OEt2 at À408C[9]
led to epoxide opening to form a bromohydrin (data not
shown), reducing the number of equivalents to 1.6, and the
temperature to À788C, fully suppressed this side reaction.
Thus, allylation of 3a led to a mixture of diastereomers 10 and
11 in a good yield and ratio, but the better results were ob-
tained with 3c, giving isomer 12 virtually exclusively in excel-
lent yield. Interestingly, treatment with allyl magnesium bro-
mide gave a low selectivity (d.r. 7:3),[14] and the use of BF3·OEt2
gave decomposition products. The relative stereochemistry of
Figure 1. Proposed transition states for reaction of 3, with I/III correspond-
ing to a Cornforth–Evans-type, and II/IV to a polar Felkin–Anh-type model.
Stereocontrol is then further determined by the CÀC and CÀO
substitution at the quaternary centre. With opposite orienta-
tions of the epoxide CÀO and carbonyl dipoles, the transition
state following attack to I is akin to a Cornforth–Evans (CE)-
type stabilisation, whereas that following attack to II, with the
CÀO bond perpendicular to the plane of the carbonyl group, is
comparable to a polar Felkin–Anh (PFA)-type.[16,17] From a steric
point of view, both substituents, when in the pseudo-axial po-
sition, are expected to hinder nucleophilic attack from that
side. The exact model for stereoinduction by non-chelating CÀ
O substituents has been subject to debate,[16] but the experi-
ments here show that the reaction via I, leading to 10, repre-
sents the lowest energy transition state and indicating a CE-
Chem. Eur. J. 2014, 20, 3306 – 3310
www.chemeurj.org
3307
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
type stabilisation is operating. Molecular modelling showed
that half-chair I is more stable than II by 11 kJmolÀ1.[14] We be-
lieve that the epoxide group is not involved in the MgBr2·OEt2
mediated chelation, as this would be expected to result in
poor diastereoselectivity.[18]
late concentrations was epoxide opening observed. The result-
ing thioester was then reduced to give the aldehyde 17, which
was arylated using the aryl bromide 18[14] in excellent yield.
At this stage the tert-butyl ester was reduced to the corre-
sponding aldehyde group, which after desilylation led to for-
mation of the hemiacetal ring 20. After considerable experi-
mentation, it was found that high-yielding debenzylation at
the primary benzylic position was only possible after prior oxi-
dation of the secondary benzylic alcohol. This was achieved in
a chemoselective manner using the Dess–Martin periodinane.
Hydrogenolysis was then followed by a further oxidation to
give luminacin D (1). All spectral data and the optical rotation
fully corresponded with the data provided by Wakabayashi.[2,14]
We next assessed the effect of 1 on the proliferation of
a model breast cancer cell line (MCF-7). Our sample of lumina-
cin D caused a dose-dependent reduction in the proliferation
of MCF-7 cells, with an IC50 of 55(Æ4) mm,[14] in-line with report-
ed values for other epithelial cell lines.[3,5]
Similar considerations can be made for the reaction of 3c,
with chelation between the two carbonyl groups giving two
interconverting “open book” structures[9,10a,e] III and IV. Apart
from the different steric environments between a half-chair
and open book conformation, the stereoselectivity compared
to 3a is thought to be enhanced due to the absence of 1,2A
strain in III, compared to IV.[19] This is corroborated by model-
ling, which showed that III is much more stable than IV, by
43 kJmolÀ1 [14]
.
To complete the total synthesis, inversion of the alcohol con-
figuration is required (Scheme 4). This is achieved by a Mitsuno-
bu/deprotection process, in which the use of chloroacetic
acid[20] gave superior results. Protection as silyl ether, followed
by ozonolysis with a phosphine-mediated reduction sequence
led to the required aldehyde 14. It was found that residual
phosphorous impurities could be efficiently removed by treat-
ment with Merrifield resin and NaI.[21]
The high stereoinduction provided by the quaternary epox-
ide centre, in a 1,3-chelation context, clearly is of wider signifi-
cance for stereoselective synthesis. We further investigated the
generality of this process by synthesising a simpler model
compound 21a, in which the quaternary epoxy centre is re-
placed by an acyclic tertiary (non-chelating)[22] silyl ether
(Table 1).
The b-triethylsilyloxy group in 14 is expected to impart the
desired aldehyde facial selectivity required for the introduction
of the next stereocentre.[17] Indeed, treatment of 14 with the
boron enolate derived from 15a led to a product having the
desired relative stereochemistry as the major isomer, but in
a moderate 4:1 ratio.[14] The diastereomeric ratio was improved
dramatically by employing a matched double diastereo-differ-
entiation process featuring enantiopure oxazolidinone 15b.
Pleasingly, enolate facial induction imposed by the chiral auxili-
ary is stereodominant, which allowed the removal of the dia-
stereomeric aldol product that was formed by reaction of the
minor enantiomer of 14, after silyl protection of the alcohol
groups. Hence, 16 was obtained as an enantiopure diastereo-
mer.
As expected, allylation under chelation conditions proceed-
ed with excellent diastereoselectivity (entry 1), in contrast to al-
lylation under BF3·OEt2 activation, which does not involve che-
lation (entry 2). Treatment of 21a with allylmagnesium bro-
mide gave low selectivity (entry 3). The selectivity is rational-
ised by reaction of the half-chair Va, via a chair-like CE-type
transition state (Figure 2). Somewhat surprisingly, the stereoin-
duction provided by the larger OTBDMS group is higher com-
pared to the epoxide group. Chelation involving the OTBDMS
group, which would lead to a much less selective allylation, is
ruled out given the high diastereoselectivity obtained. Interest-
ingly, allylstannation of the corresponding 2-methylated ana-
logue only gives a 2.6:1 ratio of products,[15c] with the major
Removal of the auxiliary was achieved by a fully chemoselec-
tive ethyl thiolate mediated displacement. Only at higher thio-
Scheme 4. Completion of the luminacin D synthesis. a) PPh3, DIAD, ClCH2COOH, THF; b) MeOH; c) imidazole, DCM; d) PPh3; Merrifield resin, NaI; e) DIPEA,
DCM, À788C; f) imidazole, DCM; g) THF; h) DCM; i) THF; j) toluene, À788C; k) THF; l) NaHCO3, DCM; m) THF/AcOH.
Chem. Eur. J. 2014, 20, 3306 – 3310
www.chemeurj.org
3308
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
In conclusion, we report that 1,3-chelation-controlled allyla-
tions of aldehydes containing a non-chelating a-ether substitu-
ent proceed with excellent diastereoselectivity, even when the
a position is a quaternary centre or a spiro-epoxide. The rela-
tive stereochemistry of the major reaction product unambigu-
ously points towards a contributing CE-type stabilisation of the
transition state. 1,2A strain was also shown to have a beneficial
effect on the diastereoselectivity. The allylation reaction was
exploited as a key step in a successful synthesis of (À)-lumina-
cin D. Other notable features in that synthesis included an ep-
oxide introduction in the first stage of the synthesis, and the
high diastereoselectivity achieved in constructing the densely
functionalised aliphatic fragment. Furthermore, we have vali-
dated the anticancer activity of the synthesised sample. Fur-
ther investigations to widen the scope of chelation-controlled
additions on aldehydes 21, as well as a large scale synthesis of
the aliphatic fragment to provide additional quantities of lumi-
nacin D and to achieve the synthesis of migracin A and migra-
cin B for biological evaluation, are underway.
Table 1. Investigation of stereoinduction by a-quaternary centres.
Entry
R
Lewis acid
Yield [%][a]
Ratio 22/23[b]
1
2
3
4
5
6
Me
Me
Me
H
H
H
MgBr2·OEt2
BF3·OEt2
none[c]
MgBr2·OEt2
BF3·OEt2
none[c]
87
57
87
53
81
81
>95:5
22:78
40:60
>95:5
25:75[24]
50:50[24]
1
[a] Isolated yield. [b] Determined by H NMR spectroscopy before chroma-
tography. [c] Allyl magnesium bromide was used.
Acknowledgements
N.B. thanks Pfizer for a CASE studentship; L.G. thanks the ERAS-
MUS office and University of Hamburg for support; F.P. is grate-
ful to the European Community (INTERREG IVa channel pro-
gramme, IS:CE-Chem, project 4061) for a studentship.
Figure 2. Proposed transition states for formation of 22 and 23, and compar-
ison with 3-benzyloxy-2-methylpropionaldehyde.
product isomer arising via transition state VII[23] as shown.
Hence, somewhat counter intuitively, the presence of the a-
OTBDMS group serves to increase the selectivity, which is a fur-
ther indication of the operating CE-type stabilisation. Castle
et al. recently demonstrated that analogous ketones (instead
of aldehydes) react similarly with excellent diastereoselectiv-
ity.[24] The 1,2A strain involving the keto group will further bene-
fit diastereoselectivity.
Keywords: allylation
luminacin D · stereoinduction
·
chelation
·
enantioselectivity
·
[1] D. J. Newman, G. M. Cragg, J. Nat. Prod. 2012, 75, 311–335.
[2] N. Naruse, R. Kageyama-Kawase, Y. Funahashi, T. Wakabayashi, Y. Wata-
nabe, T. Sameshima, K. Dobashi, J. Antibiot. 2000, 53, 579–590.
[3] T. Wakabayashi, R. Kageyama-Kawase, N. Naruse, Y. Funahashi, K. Yoshi-
matsu, J. Antibiot. 2000, 53, 591–596.
We also investigated the behaviour of glyceraldehyde 21b
(entries 4–6), for which we found, to our surprise, only a single
precedent as substrate under comparable conditions.[26,27]
MgBr2·OEt2 mediated allylstannation was again very selective,
to give syn-diastereomer 22b. With no a-methyl group, stereo-
induction can only be explained by a CE (Vb) versus PFA (VIb)
competition. The relative configuration of the major isomer
22b unambiguously shows that for additions to aldehyde 21b
under 1,3-chelation, the stereocontrol instilled by a non-chelat-
ing a-OTBDMS group follows a Cornforth–Evans-type model,
and also that it leads to a different stereochemical outcome
compared to attack to VII.
The CE and PFA models can only be distinguished by prod-
uct outcome for carbonyl additions in which a conformational
restraint is imposed on the orientation of the a-stereocentre of
the electrophile.[16a] This distinction has been demonstrated by
Evans and Marco, who exploited destabilising syn-pentane in-
teractions in an aldol Zimmermann–Traxler transition state.[16,28]
The examples shown above represent the first cases in which
this conformational restriction is imposed by 1,3-chelation and
which does not involve a cyclic transition state that includes
the nucleophile.
[4] a) S. V. Sharma, C. Oneyama, Y. Yamashita, H. Nakano, K. Sugawara, M.
Hamada, N. Kosaka, T. Tamaoki, Oncogene 2001, 20, 2068–2079; b) C.
Oneyama, H. Nakano, S. V. Sharma, Oncogene 2002, 21, 2037–2050.
[5] S. Hashimoto, M. Hirose, A. Hashimoto, M. Morishige, A. Yamada, H.
Hosaka, K.-I. Akagi, E. Ogawa, C. Oneyama, T. Agatsuma, M. Okada, H.
Kobayashi, H. Wada, H. Nakano, T. Ikegami, A. Nakagawa, H. Sabe, Proc.
Natl. Acad. Sci. USA 2006, 103, 7036–7041.
[6] Y. Arai, H. Iinuma, Y. Ikeda, M. Igarashi, M. Hatano, N. Kinoshita, T. Ukaji,
S. Simizu, K. Umezawa, J. Antibiot. 2013, 66, 225–230.
[7] a) K. Tatsuda, S. Nakano, F. Narazaki, Y. Nakamura, Tetrahedron Lett.
2001, 42, 7625–7628; b) J. B. Shotwell, E. S. Krygowski, J. Hines, B. Koh,
E. W. D. Huntsman, H. W. Choi, J. S. Schneekloth, Jr., J. L. Wood, C. M.
Crews, Org. Lett. 2002, 4, 3087–3089; c) R. Jogireddy, M. E. Maier, J. Org.
Chem. 2006, 71, 6999–7006; d) D. Oehlrich, S. M. E. Vidot, M. W. Davies,
G. J. Clarkson, M. Shipman, Tetrahedron 2007, 63, 4703–4711; e) F. Fang,
C. Johannes, Y. Yao, X. Zhu, WO03/057685, 2003.
[8] Ref. [7e] describes an apparent diastereoselective epoxidation (no ratio
was provided) on a racemic allylic alcohol intermediate, leading to the
undesired diastereomer.
[9] B. Linclau, E. Cini, C. S. Oakes, S. Josse, M. Light, V. Ironmonger, Angew.
Chem. 2012, 124, 1258–1261; Angew. Chem. Int. Ed. 2012, 51, 1232–
1235.
[10] Examples of selective additions to a-formyl esters and amides under
chelation control, with stereochemical bias provided by a central chiral
centre: a) K. Prantz, J. Mulzer, Chem. Eur. J. 2010, 16, 485–506; b) K.
Prantz, J. Mulzer, Angew. Chem. 2009, 121, 5130–5133; Angew. Chem.
Int. Ed. 2009, 48, 5030–5033; c) T. K. M. Shing, Y. Y. Yeung, Angew. Chem.
Chem. Eur. J. 2014, 20, 3306 – 3310
www.chemeurj.org
3309
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
2005, 117, 8195–8198; Angew Chem. Int. Ed. 2005, 44, 7981–7984; d) P.
Saravanan, E. J. Corey, J. Org. Chem. 2003, 68, 2760–2764; e) F. Soucy, L.
Grenier, M. L. Behnke, A. T. Destree, T. A. McCormack, J. Adams, L. Pla-
mondon, J. Am. Chem. Soc. 1999, 121, 9967–9976; f) E. J. Corey, W. Li, T.
Nagamitsu, Angew. Chem. 1998, 110, 1784–1787; Angew. Chem. Int. Ed.
1998, 37, 1676–1679.
[20] M. Saiah, M. Bessodes, K. Antonakis, Tetrahedron Lett. 1992, 33, 4317–
4320.
[21] B. H. Lipshutz, P. A. Blomgren, Org. Lett. 2001, 3, 1869–1871.
[22] a) G. E. Keck, S. Castellino, Tetrahedron Lett. 1987, 28, 281–284; b) D. A.
Evans, B. D. Allison, M. G. Yang, C. E. Masse, J. Am. Chem. Soc. 2001, 123,
10840–10852.
[23] NMR studies of VII showed that the methyl group is pseudo-equatorial
(G. E. Keck, S. Castellino, J. Am. Chem. Soc. 1986, 108, 3847–3849), but
the reactive conformer may resemble as shown in Figure 2 (chair-like
transition state).
[24] B. M. Loertscher, P. R. Young, P. R. Evans, S. L. Castle, Org. Lett. 2013, 15,
1930–1933.
[25] Reaction of 2,3-dibenzylglyceraldehyde under these conditions leads to
similar selectivities, see a) K. Mead, T. L. Macdonald, J. Org. Chem. 1985,
50, 422–424; b) M. T. Reetz, K. Kesseler, J. Org. Chem. 1985, 50, 5434–
5436.
[11] A. B. Charette, B. Cote, J. Am. Chem. Soc. 1995, 117, 12721–12732.
[12] a) R. Fernandez de La Pradilla, S. Castro, P. Manzano, M. Martin-Ortega, J.
Priego, A. Viso, A. Rodriguez, I. Fonseca, J. Org. Chem. 1998, 63, 4954–
4966. The epoxidation of acyclic 2-acyl vinylsulfoxides was reported to
proceed with low diastereoselectivity: b) R. Fernandez de La Pradilla, S.
Castro, P. Manzano, J. Priego, A. Viso, J. Org. Chem. 1996, 61, 3586–
3587.
[13] The phenyl derivatives were racemic, the p-tolyl ones were enantio-
pure; see ref. [14].
[14] See the Supporting Information for details.
[26] Propargylation of 3-methoxy-2-(triisopropylsilyl)oxypropionaldehyde
under TiCl4 chelation conditions proceeded with excellent selectivity.
However, a CE-type transition state was not discussed: W. Felzmann, D.
Castagnolo, D. Rosenbeiger, J. Mulzer, J. Org. Chem. 2007, 72, 2182–
2186.
[27] Reaction of 2,3-dibenzylglyceraldehyde under chelation conditions is
also very selective, but both 1,2- and 1,3-chelation can occur, and both
are expected to give the same syn-adduct with high selectivity. For
21b, only 1,3-chelation is possible. See: A. Bernardi, S. Cardani, L. Co-
lombo, G. Poli, G. Schimperna, C. Scolastico, J. Org. Chem. 1987, 52,
888–891.
[15] a) M. T. Reetz, A. Jung, J. Am. Chem. Soc. 1983, 105, 4833–4835; b) S.-I.
Kiyooka, C. H. Heathcock, Tetrahedron Lett. 1983, 24, 4765–4768; c) G. E.
Keck, D. E. Abbott, Tetrahedron Lett. 1984, 25, 1883–1886.
[16] a) D. A. Evans, S. J. Siska, V. J. Cee, Angew. Chem. 2003, 115, 1803–1807;
Angew. Chem. Int. Ed. 2003, 42, 1761–1765; b) V. J. Cee, C. J. Cramer,
D. A. Evans, J. Am. Chem. Soc. 2006, 128, 2920–2930, and references
therein.
[17] Reviews: a) M. T. Reetz, Angew. Chem. 1984, 96, 542–555; Angew. Chem.
Int. Ed. Engl. 1984, 23, 556–569; b) M. T. Reetz, Acc. Chem. Res. 1993, 26,
462–468; c) A. S. Cieplak, Chem. Rev. 1999, 99, 1265–1336; d) P. Wipf,
J.-K. Jung, Chem. Rev. 1999, 99, 1469–1480; e) A. Mengel, O. Reiser,
Chem. Rev. 1999, 99, 1191–1223.
[28] a) J. Alberto Marco, M. Carda, S. Diaz-Oltra, J. Murga, E. Falomir, H.
Roeper, J. Org. Chem. 2003, 68, 8577–8582; b) S. Diaz-Oltra, M. Carda, J.
Murga, E. Falomir, J. A. Marco, Chem. Eur. J. 2008, 14, 9240–9254.
[18] a) H. Urabe, O. O. Evin, F. Sato, J. Org. Chem. 1995, 60, 2660–2661;
b) B. J. Albert, A. Sivaramakrishnan, T. Naka, N. L. Czaicki, K. Koide, J. Am.
Chem. Soc. 2007, 129, 2648–2659.
[19] For an example in which similar diastereoselectivity is due to allylic
strain considerations, see refs. [10a,e].
Received: December 5, 2013
Published online on February 12, 2014
Chem. Eur. J. 2014, 20, 3306 – 3310
www.chemeurj.org
3310
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim