Stereopentads derived from a sequence of Mukaiyama aldolization and free radical reduction on α-methyl-β-alkoxy aldehydes: A general strategy for efficient polypropionate synthesis

Article abstract of DOI:10.1021/jo8021583

(Chemical Equation Presented) In a stereodivergent manner, all 16 diastereomeric stereopentads 7-22 were synthesized starting with α-methyl-β-alkoxy aldehydes 25 and 27. We designed an approach based on a sequence of a Mukaiyama aldolization with enoxysil

Full text of DOI:10.1021/jo8021583

Stereopentads Derived from a Sequence of Mukaiyama Aldolization
and Free Radical Reduction on r-Methyl-ꢀ-alkoxy Aldehydes: A
General Strategy for Efficient Polypropionate Synthesis
Jean-Franc¸ois Brazeau, Philippe Mochirian, Michel Pre´vost, and Yvan Guindon*
Institut de recherches cliniques de Montre´al (IRCM), Bio-organic Chemistry Laboratory, 110 AVenue des
Pins Ouest, Montre´al, Que´bec, Canada H2W 1R7, Department of Chemistry, UniVersite´ de Montre´al,
C.P. 6128, succursale Centre-Ville, Montre´al, Que´bec, Canada H3C 3J7, and Department of Chemistry,
McGill UniVersity, 801 Sherbrooke Street West, Montre´al, Que´bec, Canada H3A 2K6
yVan.guindon@ircm.qc.ca
ReceiVed September 26, 2008
In a stereodivergent manner, all 16 diastereomeric stereopentads 7-22 were synthesized starting with
R-methyl-ꢀ-alkoxy aldehydes 25 and 27. We designed an approach based on a sequence of a Mukaiyama
aldolization with enoxysilane 24 followed by a hydrogen transfer reaction. Recent advancements concerning
these reactions are described, and novel key intermediates are characterized in the aldol step. The synthesis
of C(1)-C(11) fragment 60 of zincophorin, which contains a synthetically challenging stereopentad unit,
is described attesting the usefulness of our strategy.
Introduction
For more than a century, isolated natural products of the
polyketide family have attracted the interest of the scientific
community. The biological properties of representative com-
pounds (see Figure 1) of this class, as well as their structural
complexity, have justified research in synthetic organic chem-
istry but also in biochemistry, genetics, and pharmacology.1
Polypropionate motifs (stereotriads, stereotetrads, stereopentads,
etc.), consisting of a repeating sequence of stereogenic centers
bearing methyl and hydroxyl functionalities, are common
components of these polyketide natural products. Synthesizing
these contiguous stereogenic centers is challenging, considering
the number of possible stereoisomers, thus the need for
stereocontrolled approaches.
As illustrated in Figure 2, the methyl-hydroxyl-methyl array
having three stereogenic centers was defined as stereotriad2
(diastereoisomers 3-6) subunits while adding another propi-
onate unit leads to stereopentads. Sixteen stereoisomers (7-22)
FIGURE 1. Structure of rifamicyn (1) and zincophorin (2).
are thus possible if one starts with a substrate possessing a
unique stereogenic center.
In the past, the most common strategies for the synthesis of
these stereotriads involved a carbon-carbon bond forming aldol
reaction (titanium,3 tin,3 and boron enolates3,4). Just as important
are methodologies based on additions of crotylboranes,5 cro-
tylboronates,6 crotylsilanes,7 allenyl stannane,8 and allyltitanates9
(1) (a) Rohr, J. Angew. Chem., Int. Ed. 2000, 39, 2847. (b) Davies-Coleman,
M. T.; Garson, M. J. Nat. Prod. Rep. 1998, 15, 477. (c) Henkel, T.; Brunne,
R. M.; Mu¨ller, H.; Reichel, F. Angew. Chem., Int. Ed. 1999, 38, 643. (d) Omura,
S. Macrolide Antibiotics; Academic Press Inc.: New York, 1984.
(2) (a) Hoffmann, R. W. Angew. Chem., Int. Ed. Engl. 1987, 26, 489. (b)
Hoffmann, R. W.; Dahmann, G.; Andersen, M. W. Synthesis 1994, 629.
64 J. Org. Chem. 2009, 74, 64–74
10.1021/jo8021583 CCC: $40.75 2009 American Chemical Society
Published on Web 12/08/2008

Mukaiyama Aldolization and Free Radical Reduction
FIGURE 2. Stereotriads and the corresponding polypropionate stereopentads.
to aldehydes, which showed a high level of stereoselectivity,
advancing significantly our collective capacity of creating
stereogenic centers in acyclic molecules. They have been used
successfully for the assembly of the four stereotriads. Other
approaches based on the stereoselective transformations of cyclic
and acyclic precursors produced these molecules, as well as
higher analogues, attesting to the challenges and opportunities
that the synthesis of polypropionate motifs and the related
natural products provides.10
Yet, no single strategy has been reported leading directly to
all 16 stereopentad diastereoisomers. Many of the approaches
listed before used chiral reagents and took advantage of double
asymmetric induction. Not surprisingly, mismatched scenarios
were at times noticed. For instance, the addition of the well-
known crotylboronate and the crotylborane on stereotriad
aldehydes led to the 2,3-anti-3,4-anti polypropionate stereo-
pentads with poor stereoselectivity.5,6,11 To overcome this
problem, Roush elegantly took advantage of the silicon hyper-
valence using crotyltrifluorosilanes for the elaboration of R,ꢀ-
anti-polypropionate stereopentads from 2,3-anti-ꢀ-hydroxy al-
dehydes.12 Unfortunately, aldehydes having a 2,3-syn relationship
led to less impressive results. As for Marshall, he proposed a
compelling strategy based on the reactions of chiral allenylzinc
reagents with the propionate stereotriads previously obtained
through the methodology developed in his group.13 Only the
R,ꢀ-anti stereopentad isomers can be obtained via this route.
Six out of the eight possible polypropionate were constructed
selectively, poor diastereoselectivity being noted with 2,3-anti-
aldehydes and (M)-allenylzinc. Paterson also proposed a sys-
tematic approach to stereopentads that combines three steps:
an aldol addition with enol borinates, a hydroxyl-directed ketone
reduction, followed by a hydroboration of the terminal alkene.4
(10) Other very interesting approaches for propionate and polypropionate
synthesis have also been developed over the years. See: (a) Kishi, Y. Tetrahedron
1981, 23, 3873. (b) Masamune, S.; Choy, W. Aldrichimica Acta 1982, 15, 47.
(c) Danishefsky, S. J. Aldrichimica Acta 1986, 19, 59. (d) Hanessian, S.
Aldrichimica Acta 1989, 22, 3. (e) Heathcock, C. H. Aldrichimica Acta 1990,
23, 99. (f) Szymoniak, J.; Lefranc, H.; Mo¨ıse, C. J. Org. Chem. 1996, 61, 3926.
(g) Reggelin, M.; Brenig, V. Tetrahedron Lett. 1996, 6851. (h) Arjona, O.;
Menchaca, R.; Plumet, J. J. Org. Chem. 2001, 66, 2400. (i) Breit, B.; Zahn,
S. K. J. Org. Chem. 2001, 66, 4870. (j) Calter, M. A.; Guo, X.; Liao, W. Org.
Lett. 2001, 3, 1499. (k) Kiyooka, S.-I. Tetrahedron: Asymmetry 2003, 14, 2897.
(l) Torres, E.; Chen, Y.; Kim, C. I.; Fuchs, P. L. Angew. Chem., Int. Ed. 2003,
42, 3124. (m) Falder, L. D.; Carreira, E. M. Org. Lett. 2004, 6, 2485. (n) Vogel,
P.; Gerber-Lemaire, S.; Carmona, A. T.; Meilert, K. T.; Schwenter, M.-E. Pure
Appl. Chem. 2005, 77, 131. (o) Solsona, J. G.; Romea, P.; Urpi, F.; Vilarrasa,
J. J. Org. Chem. 2005, 70, 6533. (p) Shimp, H. L.; Micalizio, G. C. Org. Lett.
2005, 7, 5111. (q) Chau, A.; Paquin, J.-F.; Lautens, M. J. Org. Chem. 2006, 71,
1924. (r) Shen, X.; Wasmuth, A. S.; Zhao, J.; Zhu, C.; Nelson, S. G. J. Am.
Chem. Soc. 2006, 128, 7438. (s) Sheengut, V.; Ward, D. E. J. Org. Chem. 2007,
72, 7805. (t) Parker, K. A.; Wang, P. Org. Lett. 2007, 9, 4793. (u) Da´vila, W.;
Torres, W.; Prieto, J. A. Tetrahedron 2007, 67, 8218. (v) Arikan, F.; Li, J.;
Menche, D. Org. Lett. 2008, 10, 3521.
(3) (a) Evans, D. A.; Clark, J. S.; Mettemich, R.; Novack, V. J.; Sheppard,
G. S. J. Am. Chem. Soc. 1990, 112, 866. (b) Evans, D. A.; Ng, H. P.; Clark,
J. S.; Rieger, D. L. Tetrahedron 1992, 48, 2127.
(4) (a) Paterson, I. Pure Appl. Chem. 1992, 64, 1821. (b) Paterson, I.;
Channon, J. A. Tetrahedron Lett. 1992, 33, 797. (c) Paterson, I.; Norcross, R. D.;
Ward, R. A.; Romea, P.; Lister, A. L. J. Am. Chem. Soc. 1994, 116, 11287. (d)
Yeung, K.-S.; Paterson, I. Chem. ReV. 2005, 105, 4327.
(5) (a) Brown, H. C.; Bhat, K. S.; Randad, R. S. J. Org. Chem. 1987, 52,
3701. (b) Brown, H. C.; Bhat, K. S.; Randad, R. S. J. Org. Chem. 1989, 54,
1570.
(6) (a) Roush, W. R.; Palkowitz, A. D.; Ando, K. J. Am. Chem. Soc. 1990,
112, 6348. (b) Roush, W. R.; Ando, K.; Powers, D. B.; Palkowitz, A. D.;
Halterman, R. L. J. Am. Chem. Soc. 1990, 112, 6339. (c) Roush, W. R.; Grover,
P. T. J. Org. Chem. 1995, 60, 3806.
(7) (a) Jain, N. F.; Takenaka, N.; Panek, J. S. J. Am. Chem. Soc. 1996, 118,
12475. (b) Panek, J. S.; Jain, N. F. J. Org. Chem. 1998, 63, 4572.
(8) (a) Marshall, J. A.; Perkins, J. F.; Wolf, M. A. J. Org. Chem. 1995, 60,
5556. (b) Marshall, J. A. Chem. ReV. 2000, 100, 3163.
(9) (a) BouzBouz, S.; Cossy, J. Org. Lett. 2001, 3, 3995. (b) BouzBouz, S.;
Cossy, J. Org. Lett. 2003, 5, 3029. This group has also studied the use of
oxymercuration for propionate synthesis. See: (c) Defosseux, M.; Blanchard,
N.; Meyer, C.; Cossy, J. Tetrahedron 2005, 61, 7632, and references therein.
(11) Masamune and Hoffmann work with the complex enol borane and
crotylboronate: (a) Short, R. P.; Masamune, S. Tetrahedron Lett. 1987, 28, 2841.
(b) Tanimoto, N.; Gerritz, S. W.; Sawabe, A.; Noda, T.; Filla, S. A.; Masamune,
S. Angew. Chem., Int. Ed. Engl. 1994, 33, 673. (c) Andersen, M. W.;
Hilderbrandt, B.; Hoffmann, R. W. Angew. Chem., Int. Ed. Engl. 1991, 30, 97–
99. (d) Andersen, M. W.; Hildebrandt, B.; Dahmann, G.; Hoffmann, R. W. Chem.
Ber. 1991, 124, 2127.
(12) (a) Chemler, S. R.; Roush, W. R. J. Org. Chem. 2003, 68, 1319. (b)
Chemler, S. R.; Roush, W. R. Tetrahedron Lett. 1999, 40, 4643. (c) Chemler,
S. R.; Roush, W. R. J. Org. Chem. 1998, 63, 3800.
(13) Marshall, J. A.; Schaaf, G. M. J. Org. Chem. 2001, 66, 7825.
J. Org. Chem. Vol. 74, No. 1, 2009 65

Brazeau et al.
FIGURE 3. Mukaiyama and free radical hydrogen transfer approach to polypropionate construction.
SCHEME 1
Twelve out of the 16 isomers were obtained directly.14 All of
the above-mentioned approaches provided important conceptual
advancements to the field while pointing out the difficulties of
creating stereogenic centers on acyclic molecules which possess
a complex stereochemical environment.
Our group designed a conceptually different strategy in which
no mismatched scenarios should be observed.15 Central to our
strategy is a substrate control approach involving the two
reactions depicted in Figure 3. The first one, an aldol-like
Mukaiyama reaction, involves a mixture of E and Z tetrasub-
stituted enoxysilanes bearing a bromide atom and an aldehyde
in the presence of a Lewis acid. In this first step, we are only
concerned about forming the carbon-carbon bond and creating
the stereogenic center at C(3). Their 3,4-relative stereochemistry
would be determined by the nature of the Lewis acid used and
the transition states thus favored, as illustrated in Figure 3. The
use of monodentate Lewis acid would lead to 3,4-syn isomers
while bidentate Lewis acid to 3,4-anti isomers.
The second set of reactions used in this sequence involves
the creation of a carbon-centered free radical at C(2), followed
by a hydrogen transfer reaction. Prior to our studies, free radical
intermediates were never seriously considered in the synthesis
of acyclic complex molecules. Over the years, we found that
Lewis acids are crucial in controlling the anti16 and syn17 relative
stereochemistry between C(3) and the newly created C(2)
stereogenic center. We observed that the presence of a hetero-
atom on the secondary carbon C(3) and of an ester at C(1) was
critical for the diastereoselectivity.16a
(Scheme 1) for the synthesis of polypropionate stereopentads
7-22.18 Our investigation addresses the fundamental stereo-
chemical issues originating from R,ꢀ-substituent effects in
Mukaiyama aldol reactions and radical reductions. Interestingly,
the success of the aldolization step is highly dependent on the
stoichiometry and the nature of the Lewis acid used, especially
under chelated control conditions. An application of our new
strategy is outlined in the elaboration of the C(1)-C(11)
fragment of zincophorin (Figure 1).
Results and Discussion
Development of the Mukaiyama Reaction. The enoxysilane
24 used in the aldol Mukaiyama has three important charac-
teristics. First, it contains a bromide-carbon bond which, in
the aldol product, will serve as a source of the carbon-centered
free radical for the second step of our sequence. Second, the
enoxysilane is tetrasubstituted and thus sterically congested. This
will provide a significant advantage in the synthesis of stereo-
pentads, whereas the 1,2-induction will be the stereocontrolling
factor in the aldol step (vide infra).19 Finally, it is important to
note that a mixture of E- and Z-enoxysilanes is used, the
stereochemistry at C(2) in the aldol product being irrelevant
for the second step of our sequence. The enoxysilanes can
be readily synthesized starting from the commercially available
2-bromopropionate 23 by using a standard enolization procedure
(Scheme 1). The reaction can be performed on a significant scale
(>250 mmol), and the enoxysilane 24, present in a 4:1 E:Z
selectivity, is stable for months when stored at -20 °C under
inert atmosphere.
Herein, we report the full details of our studies aiming at an
iterative sequence using aldehydes derived from stereotriads 3-6
(14) The others were constructed by switching the terminal hydroxy group
of isomers previously obtained. See ref 4.
(15) (a) Guindon, Y.; Houde, K.; Pre´vost, M.; Cardinal-David, B.; Landry,
S. R.; Daoust, B.; Bencheqroun, M.; Gue´rin, B. J. Am. Chem. Soc. 2001, 123,
8496. (b) Guindon, Y.; Pre´vost, M.; Mochirian, P.; Gue´rin, B. Org. Lett. 2002,
4, 1019. (c) Cardinal-David, B.; Brazeau, J.-F.; Katsoulis, I. A.; Guindon, Y.
Curr. Org. Chem. 2006, 10, 1939.
(16) (a) Guindon, Y.; Yoakim, C.; Gorys, V.; Ogilvie, W. W.; Delorme, D.;
Renaud, J.; Robinson, G.; Lavalle´e, J.-F.; Slassi, A.; Jung, G.; Rancourt, J.;
Durkin, K.; Liotta, D. J. Org. Chem. 1994, 59, 1166. (b) Guindon, Y.; Faucher,
A.-M.; Bourque, E.; Caron, V.; Jung, G.; Landry, S. R. J. Org. Chem. 1997, 62,
9276. (c) Guindon, Y.; Liu, Z.; Jung, G. J. Am. Chem. Soc. 1997, 119, 9289. (d)
Bouvier, J.-P.; Jung, G.; Liu, Z.; Gue´rin, B.; Guindon, Y. Org. Lett. 2001, 3,
1391.
(17) (a) Guindon, Y.; Lavalle´e, J.-F.; Llinas-Brunet, M.; Horner, G.; Rancourt,
J. J. Am. Chem. Soc. 1991, 113, 9701. (b) Guindon, Y.; Rancourt, J. J. Org.
Chem. 1998, 63, 6566.
As illustrated in Scheme 2, the aldehydes 3, 4, 5, 6, 30, and
34 were prepared from their corresponding ester precursors using
(18) (a) Mochirian, P.; Cardinal-David, B.; Gue´rin, B.; Pre´vost, M.; Guindon,
Y. Tetrahedron Lett. 2002, 43, 7067. (b) Guindon, Y.; Brazeau, J.-F. Org. Lett.
2004, 6, 2599.
(19) Evans reported a very interesting investigation about the stereochemical
consequences of R-alkyl-ꢀ-alkoxy aldehyde in Mukaiyama aldolization. See:
(a) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Yang, M. J. J. Am. Chem. Soc. 1996,
118, 4322. (b) Evans, D. A.; Allison, B. D.; Yang, M. C.; Masse, C. G. J. Am.
Chem. Soc. 2001, 123, 10840.
66 J. Org. Chem. Vol. 74, No. 1, 2009

Mukaiyama Aldolization and Free Radical Reduction
TABLE 1. Mukaiyama Aldolization with 2,3-syn Aldehydes 3, 4, and 30
a Conditions A: Reactions were conducted with 2.0 equiv of enoxysilane 24 followed by 1.2 equiv. BF3 ·OEt2 in CH2Cl2 (0.1 M) at -78 °C for 2 h,
unless noted otherwise. Conditions B: Aldehydes were precomplexed with 1.1 equiv of TiCl3(OiPr) followed by the addition of 2.0 equiv of
enoxysilane 24 in CH2Cl2 (0.1 M) at -78 °C for 2 h, unless noted otherwise. Conditions C: Aldehydes were precomplexed with 2.5 equiv of
TiCl3(OiPr) followed by the addition of 2.0 equiv of enoxysilane 24 in CH2Cl2 (0.1 M) at -78 °C for 2 h, unless noted otherwise. b Isolated yields.
c Product ratios were determined by 1H analysis (400 or 500 MHz) of the crude reaction mixture. d In all cases, products were separable by
chromatographic methods. e 20% of undesired 3,4-syn isomer was observed.
SCHEME 2
We were more concerned about the reactions involving those
aldehydes activated by bidentate Lewis acids. The substituents
at C(2) and C(3) being gauche to each other in these complexes,
the energy necessary for their formation should therefore be
increased. Many attempts to construct the Cram-chelated
Mukaiyama aldol products using bidentate Lewis acids such as
TiCl4, MgBr2 ·OEt2, Et2BOTf, SnCl4, Me2AlCl, etc. led to
disappointing results. Interestingly, as we focused on lowering
the titanium Lewis acidity, the use of TiCl3(OiPr) provided our
first positive result, the aldol products 36 being obtained in good
yield, albeit with modest Cram-chelate selectivity (Table 1, entry
2). Even more interesting was our observation that this drawback
could be overcome by increasing the stoichiometry of the Lewis
acid, as indicated by the impressive 3,4-anti stereoselectivity
favoring compounds 36 when 2.5 equiv of TiCl3(OiPr) is used
(entry 3). Similar results are obtained with aldehyde 4 (entry
6). The importance of the stoichiometry of the Lewis acid on
the reaction outcome will be further discussed (vide infra).
The aldehydes 5, 6, and 34 having a 2,3-anti relative
stereochemistry were then reacted with our enoxysilane 24. As
seen in Table 2, an excellent diastereoselectivity was noted when
BF3 ·OEt2 was used as the Lewis acid, a 3,4-syn relative
stereochemistry being induced (entries 1, 3, and 4). Interestingly,
under Cram-chelated conditions, opposing 1,2- and 1,3-induc-
tions are possible in these cases, both substituents being on the
opposite face of the chelate intermediate. Fortunately, as seen
in entry 5, an excellent ratio favoring the 3,4-anti products was
noted using an excess of TiCl3(Oi-Pr) and aldehyde 6.
Surprisingly, the Mukaiyama reaction of 2,3-anti-3,4-anti
aldehyde 5 in presence of an excess of TiCl3(OiPr) was selective
but much less efficient. The reaction time is longer, and the
presence of R,ꢀ-unsaturated aldehyde was noted. Fortunately,
in this case, the use of 1.1 equiv of TiCl4 was very efficient,
the major 3,4-anti products being obtained in good yield (entry
2). One should note that, under these conditions (1.1 equiv of
TiCl4), only degradation of the substrates was noted with
aldehydes 3, 4, and 6. As seen in Table 1 (entry 5) and Table
2 (entry 4), a silyloxyether can be used as the protecting group
a sequence of protection and reduction/oxidation.20 These esters
were obtained using our approach to the propionate stereotriads
asdescribedbeforeandusedinoursynthesisofthestereopentads.15a,b
The first group of aldehydes 3, 4, and 30 share a 2,3-syn
relative stereochemistry. Opposing 1,2- and 1,3-stereoinductions
are therefore possible when nucleophiles are added on such
aldehydes when activated by monodentate Lewis acid. However,
a high 3,4-syn diastereoselectivity was observed when BF3 ·OEt2
was added to these aldehydes followed with a mixture of
enoxysilane 24 (Table 1, entries 1, 4, and 5). This is consistent
with the predominance of 1,2-induction when sterically encum-
bered enoxysilane reagents are used.19
(20) The crude ꢀ-benzyl aldehydes so prepared were very clean by 1H NMR
analysis and were used directly in the Mukaiyama reactions without chromato-
graphic purification to avoid racemization. For all detailed procedures, see
Supporting Information.
J. Org. Chem. Vol. 74, No. 1, 2009 67

Brazeau et al.
TABLE 2. Mukaiyama Aldolization with 2,3-anti Aldehydes 5, 6, and 34
a Conditions A: Reactions were conducted with 2.0 equiv of enoxysilane 24 followed by 1.2 equiv of BF3OEt2 in CH2Cl2 (0.1 M) at -78 °C for 2 h,
unless noted otherwise. Conditions B: Aldehydes were precomplexed with 1.1 equiv of TiCl4 followed by the addition of 2.0 equiv of enoxysilane 24 in
CH2Cl2 (0.1 M) at -78 °C for 2 h, unless noted otherwise. Conditions C: Aldehydes were precomplexed with 2.5 equiv of TiCl3(OiPr) followed by the
addition of 2.0 equiv of enoxysilane 24 in CH2Cl2 (0.1 M) at -78 °C for 2 h, unless noted otherwise. b Isolated yields. c Product ratios were determined
by H analysis (400 or 500 MHz) of the crude reaction mixture. d In all cases, products were separable by chromatographic methods.
1
chemical entity could be identified: a mixture of different
complexes and free aldehyde seemed to be present. Most
importantly, no significant shift of the carbonyl signal was
observed.
The spectrum corresponding to aldehyde 3 in presence of
2.5 equiv of TiCl3(OiPr) is shown (Figure 4). A striking 18 ppm
downfield shift of the carbonyl is now noticed. Particularly
interesting are the shifts of the carbon bearing OBn groups. The
carbon C(3) bearing the secondary benzyloxy substituent now
lies 5 ppm downfield. As for the carbon bearing the primary
benzyloxy group, the interpretation is more difficult since the
signal is in close proximity to benzylic carbons. However, since
all these signals have been shifted downfield, one could conclude
that all the oxygens of aldehyde 3 are involved in the complex
when 2.5 equiv of TiCl3(OiPr) is used.
Our results indicated that, with 1 equiv of TiCl3(OiPr), a
mixture of complexes are noted in solution. The activated
aldehyde (monodentate, bidentate, or dimeric complexes) is
FIGURE 4. 13C NMR spectra of aldehyde 3 with 0.0 and 2.5 equiv of
TiCl3(OiPr) recorded at -20 °C.
present at a concentration below the sensitivity threshold of the
NMR spectrometer. The diastereoselectivity observed is poor,
suggesting that the Cram-chelate pathway sought after is not
predominant and that monodentate activation (Felkin-Anh) is
competitive. With 2.5 equiv of TiCl3(OiPr), the situation changes
as the oxygen of the benzylethers at C(3) and C(5) seemed to
be involved as the aldehyde.
Gau’s studies on titanium complexes are instructive on the
nature of the possible intermediates involved in our reactions.21
He noted the prevalence of complexes having six ligands on
the titanium center. He suggested, as well, that the binding
ability of various chemical entities to titanium was such as
i-PrO- > Cl- > THF > Et2O > PhCHO > RCO2Me. The
following scenario is therefore suggested and depicted in
Scheme 3. With 1 equiv, many complexes are in equilibrium.
One could then suggest that complex A be significantly
populated, both oxygens of the benzylethers being ligands to
the titanium center. However, this is a nonreactive intermediate,
the carbonyl being unactivated by the Lewis acid. Since the
of the primary alcohol in the BF3 ·OEt2 protocol (conditions
A). Unfortunately, this protecting group in 30 and 34 is not
compatible with the titanium-based Lewis acids. From the four
stereotriads depicted in Figure 2, we have thus successfully
synthesized, with high diastereoselectivities, the eight pairs of
desired tertiary bromides.
NMR Studies of the Chelation-Controlled Mukaiyama
Reaction. The peculiarity of the conditions necessary to achieve
the aldolization in presence of TiCl3(OiPr) led us to seek more
information on the nature of the intermediates implicated. Low-
temperature 13C NMR studies with 2,3-syn-3,4-anti aldehyde 3
were performed to better understand the impact of the stoichi-
ometry of TiCl3(OiPr) (see Figure 4). Since no difference in
the diastereomeric ratio is observed at -20 °C, the study was
performed in CD2Cl2 at this temperature as the spectra were
better defined.
The bottom spectrum in Figure 4 illustrates the 13C NMR of
the free aldehyde 3 in CD2Cl2 at the same concentration as the
reaction studied. The spectrum of aldehyde 3 with 1.1 equiv of
TiCl3(OiPr) was relatively complex, and no single discernible
(21) Gau, H.-M.; Lee, C.-S.; Lin, C.-C.; Jiang, M.-K.; Ho, Y.-C.; Kuo, C.-
N. J. Am. Chem. Soc. 1996, 118, 2936.
68 J. Org. Chem. Vol. 74, No. 1, 2009

Mukaiyama Aldolization and Free Radical Reduction
SCHEME 3
the case of aldehyde 48, the oxygen at C(5) is absent. In both
cases, the formation of the tridentate intermediate is therefore
not possible. One will note that the desired 3,4-anti adducts
could not be isolated, the elimination product of the starting
material being observed as the major product in both reactions.
In conclusion, an excess of TiCl3(OiPr) in presence of bisben-
zyloxy aldehydes led to a three- points chelate as an ate complex
in which the carbonyl was activated and where 1,2-induction
dominated.
Development of the Radical Reduction. We have developed
a strategy to control the stereochemical outcome of the second
step of our sequence: the hydrogen transfer reaction. Once again,
Lewis acids have a important role. Crucial to the success of
these reactions is the presence of ester, amide, or ketone R to
the secondary radicals.16 The delocalization of the radicals in
these functionalities induces the planarity of these tetrasubsti-
tuted “enol-like” radicals. The presence of a heteroatom on the
stereogenic center adjacent to the delocalized radicals was shown
to be important to control the diastereoselectivities sought after.
Using a mixture of C(2)-diastereomeric Mukaiyama adducts, a
2,3-syn or 2,3-anti relative stereochemistry could be established
in the reduced product, depending on the choice of Lewis acid
in the hydrogen transfer reaction.
ratio of aldol products is low, reactive complexes respectively
leading to syn and anti have to be implicated. Complex B
leading to the 3,4-anti product could obviously play a role.
Similarly, complexes such as C derived from A with another
equivalent of Lewis acid could lead to the 3,4-syn product.
However, with an excess of TiCl3(OiPr), the thermodynamically
preferred complex could involve both benzyloxy and the
carbonyl. In order to accommodate three new ligands, a chloride
has to be displaced, creating a cationic species stabilized by its
counteranion as the ate complex D, thus the need for more than
2 equiv of Lewis acid.22 Once the rigid three-points chelate is
formed, 1,2-induction dictates the stereochemical outcome of
the reaction, and because of the steric importance of the
enoxysilane 24, the 3,4-anti product is obtained.
First, in the presence of aluminum-based Lewis acid, a 2,3-
syn relative stereochemistry could be induced using the hydrogen
transfer reaction by taking advantage of the endocyclic effect
(Scheme 4).17 In this case, the Lewis acid first reacts with the
free hydroxyl group and a ligand is exchanged. Subsequently,
a cyclic complex is created involving the carbonyl of the ester
which will imbed the free radical center in a ring for the next
step. This leads to the lower energy transition state A depicted
in Scheme 4 with an attack from the top face by the trialkyltin
hydride giving the 2,3-syn motif. One would note that a free
hydroxyl group at C(3) could, through hydrogen bonding with
the oxygen of the carbonyl, allow such reaction pathway in the
absence of Lewis acid, albeit with low ratio.
A 2,3-anti relative stereochemistry in the stereopentad could
also be obtained using the hydrogen transfer reaction. The anti
diastereoselectivity depends, in this case, on the lowest energy
transition state depicted in B (Scheme 5), whereas minimization
of allylic 1,3-strain and intramolecular dipole-dipole are
achieved.16a,b As stated before, when ꢀ-hydroxy esters are used,
one has to be concerned by the existence of a competing
pathway leading to the 2,3-syn product. Hydrogen bonding
between the C(3) hydroxyl and the carbonyl needs to be
eliminated in order to maximize the anti diastereoselectivity.
Obviously, the introduction of a protecting group on the oxygen
at C(3) could achieve such goal with the downside of adding
steps to our process. We have discovered that Bu2BOTf was
an optimal solution to this problem.15,25 Indeed, this Lewis acid
reacts with the hydroxyl at C(3) to generate the dialkyl borinate
in situ in the presence of a base, such as DIEA. The resulting
product does not need to be isolated prior to the hydrogen
transfer reaction, and the borinate can be released easily in the
subsequent workup of the reaction. One should note that the
borinate does not activate intramolecularly the carbonyl as
demonstrated by the selective formation of the 2,3-anti product
when one uses this Lewis acid. An additional property justified
even more the use of the borinate. Indeed, we have noted that
poor anti diastereoselectivity was observed when R2 ) H, a
FIGURE 5. Structures of aldehydes 47 and 48.
To support these observations, we noted that the remarkable
shift of the aldehyde carbon was also observed (15 ppm) when
the complex was prepared from 1.0 equiv of TiCl3(OiPr)
followed by addition of 1.0 equiv of sodium tetrakis[(3,5-
trifluoromethyl)phenyl]borate (NaBARF). This salt is known
to be easily solvated.23 The BARF anion could thus act as the
counteranion of the ate complex and sodium chloride be
formed.24 The relevance of the three-points chelation intermedi-
ate is also supported by the poor efficiency of the Mukaiyama
aldol reactions with aldehydes 47 and 48 (Figure 5) in presence
of an excess of TiCl3(OiPr). The bulky silyl protecting group
of aldehyde 47 should prevent the chelation on this oxygen. In
(22) This intermediate is not necessarily the reactive specie since the Curtin-
Hammett principle can be evoked. However, many stereoselective nucleophilic
additions with TiCl4 and SnCl4 have been explained by models supported by
NMR spectroscopy. See: Keck, G. E.; Andrus, M. B.; Castellino, S. J. Am. Chem.
Soc. 1989, 111, 8136, and references therein.
(23) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066.
(24) Also used to support the formation of an “ate” complex with Me2AlCl
in Diels-Alder reactions. See: Castellino, S.; Wesley, J. D. J. Am. Chem. Soc.
1993, 115, 2986.
(25) Bouvier, J.-P.; Jung, G.; Liu, Z.; Gue´rin, B.; Guindon, Y. Org. Lett.
2001, 3, 1391.
J. Org. Chem. Vol. 74, No. 1, 2009 69

Brazeau et al.
SCHEME 4
SCHEME 5
scenario not faced in the context of this study. These ratios could
be increased by taking advantage of the exocyclic effect
(Scheme 5, transition state C).16c,d In this case, a ring (permanent
or temporary) is created by linking together the hydroxyls at
C(3) and C(5). We showed that boron-based Lewis acids could
generate such temporary rings by chelating preferentially the
C(3) and C(5) hydroxyl groups, even in the presence of the
carbonyl at C(1).25 As suggested in transition state C, an
optimized orientation of the substituents at C(4) leads to increase
energy differences between the syn and anti transition states,
favoring the 2,3-anti product. However, we think it is very
unlikely that such bidentate complexes would be formed in the
compounds studied herein because of steric effects, especially
in the case of the stereopentads. As stated before, the formation
of the borinate was nevertheless important to prevent the
endocyclic effect induced by a free hydroxyl.
Having the 4,5-syn aldol adducts 35-39 in hand, we
evaluated the free radical reductions of these substrates using
Et3B/O2 as radical initiator. The 2,3-syn products were obtained
when AlMe3 was first precomplexed with the R-bromo-ꢀ-
hydroxyesters prior to the hydrogen transfer reaction (Table 3,
entries 1, 3, 5, 7, and 9). Regardless of the substrate relative
stereochemistry, reductions under the endocyclic effect (Scheme
4) were highly efficient and 2,3-syn diastereoselective.
Boron-based Lewis acids were then evaluated, and impressive
results were also obtained. As illustrated in entries 2, 4, 6, 8,
and 10, very high 2,3-anti selectivity (>20:1) was observed,
suggesting that the acyclic stereocontrol (Scheme 5, transition
state B) was efficient in those cases.
and 44, excellent selectivities were also observed (entries 6 and
8). An extremely surprising result was noted for the 3,4-syn-
4,5-anti-5,6-syn tertiary bromides 43 and 45. Indeed, having
15 out of 16 diastereoisomers in hand, we were astonished to
find that the anti conditions led to a disappointing 1:1 ratio
between the C(2) and C(3) in the reduced products. As noted
before, simply changing the stereochemistry at C(5), as in 41,
gave an excellent diastereoselectivity. We are presently inves-
tigating the factors at the origin of the lack of stereoselectivity
noted for this particular polypropionate motif. One hypothesis
suggests that the folding of the chain in 43 and 45 is such that
it interferes with the reagent trajectory (Bu3SnH) from the
bottom face of the radical, raising the energy of transition state
B. Such putative conformational bias could be alleviated by
introducing other torsional strains in this substrate. Linking the
heteroatom at C(3) and C(5) through an acetonide was therefore
considered. Another reason to do so streams from our previous
studies on the exocyclic effect17b which suggested that the
presence of a six-membered cycle adjacent to the carbon-
centered radical (Scheme 5, path B) improved significantly the
2,3-anti ratios in such reactions. The bromides 53 and 54 were
therefore prepared, and the hydrogen transfer reaction was then
realized without Bu2BOTf (Table 5). We were extremely pleased
with the high diastereoselectivity noted favoring the desired 2,3-
anti polypropionates 55 and 57. In conclusion, we achieVe our
goal to design a general and diVergent strategy for the synthesis
of the 16 stereopentads.
Proof of Stereostructures. The proof of structures for all
the polypropionate was realized by converting the stereopentads
into their corresponding cyclic lactones for NMR and X-ray
studies to confirm the substitution pattern.26 An example of
which is illustrated in Scheme 6. The all-anti stereopentad 18
was transformed into lactone 59 via a hydrogenolysis process.
X-ray analysis confirmed the all-anti substitution pattern of
molecule 18.
Results of the stereoselective hydrogen transfer of the 4,5-
anti bromoesters 41-45 are summarized in Table 4. Again, as
with the 4,5-syn R-bromoesters, reactions were very selective
when AlMe3 was added prior to the reducing agent, the
endocyclic pathway being at the origin of the 2,3-syn selectivity
(Table 4, entries 1, 3, 5, 7, and 9).
Having 12 out of the 16 stereopentads in hand, we then turned
our attention to the 2,3-anti series using our boron-based Lewis
acid strategy. One will note that the “arduously accessible”8a
all-anti stereopentad 18 was obtained with excellent stereocon-
trol from the corresponding bromides 42a,b (entry 4). For 41
Synthesis of the C(1)-C(11) Fragment of Zincophorin.
Zincophorin27 is a natural product that has attracted the attention
of many groups because of its antibiotic profile and its
(26) See Supporting Information.
70 J. Org. Chem. Vol. 74, No. 1, 2009

Mukaiyama Aldolization and Free Radical Reduction
TABLE 3. Stereoselective Hydrogen Transfer Reactions with 4,5-syn-r-Bromoesters
a Conditions A: Substrates (0.1 M) were pretreated with AlMe3 for 45 min followed by Bu3SnH (1.5 equiv) in CH2Cl2 at -78 °C. Addition of air
and Et3B (0.2 equiv) every 30 min was done until the reaction was complete by TLC. Conditions B: Substrates (0.1 M) were pretreated with iPr2NEt
(1.5 equiv) and Bu2BOTf followed by Bu3SnH (1.5 equiv) in CH2Cl2 at -78 °C. Addition of air and Et3B (0.2 equiv) every 30 min was done until the
reaction was complete by TLC. b Isolated yields. Product ratios were determined by H analysis (400 or 500 MHz) of the crude reaction mixture. d In
all cases, products were separable by chromatographic methods.
c
1
SCHEME 6
SCHEME 7
mechanism of action.12,13,28 As seen in Scheme 7, Miyashita et
al. has recently synthesized the C(1)-C(11) fragment 60 of
zincophorin.28d The molecule 60 could be targeted starting from
polypropionate stereopentad 16 which was synthesized in its
enantiomerically pure version from Roche ester.
As depicted in Scheme 8, the stereopentad 16 was protected
with a TES group and then transformed into alcohol 61. Swern
oxidation furnished the aldehyde which was treated under
Horner-Emmons-Wadsworth conditions, thereby providing the
R,ꢀ-unsaturated ester 62. Chemoselective reduction of the
double bond was realized in the presence of pyridine. The ester
obtained was reduced and treated under the Swern conditions,
leading to the formation of aldehyde 63. Addition of BiBr3
followed by the silyl enol ether 24 furnished the desired 3,7-
anti-tetrahydropyrans 64a,b with good yield (79%) in a 1:1 ratio
at C(2).29 These radical precursors were used as a mixture and
(27) Grafe, U.; Schade, W.; Roth, M.; Radics, L.; Incze, M.; Ujszaszy, K. J.
Antibiot. 1984, 37, 836.
J. Org. Chem. Vol. 74, No. 1, 2009 71

Brazeau et al.
TABLE 4. Stereoselective Hydrogen Transfer Reactions with 4,5-anti-r-Bromoesters
a Conditions A: Substrates (0.1 M) were pretreated with AlMe3 for 45 min followed by Bu3SnH (1.5 equiv) in CH2Cl2 at -78 °C. Addition of air
and Et3B (0.2 equiv) every 30 min was done until the reaction was complete by TLC. Conditions B: Substrates (0.1 M) were pretreated with iPr2NEt
(1.5 equiv) and Bu2BOTf followed by Bu3SnH (1.5 equiv) in CH2Cl2 at -78 °C. Addition of air and Et3B (0.2 equiv) every 30 min was done until the
c
reaction was complete by TLC. b Isolated yields. Product ratios were determined by H NMR analysis (400 or 500 MHz) of the crude reaction mixture.
1
d In all cases, products were separable by chromatographic methods. e Combined yields.
TABLE 5. Stereoselective Hydrogen Transfer Reactions
b
a Isolated yields. Product ratios were determined by H NMR analysis (500 MHz) of the crude reaction mixture.
1
were subjected to a radical reduction under the classical
exocyclic effect in the presence of Ph3SnH/Et3B. This afforded
65 as a diastereomeric mixture (88:12) in a combined yield of
92% that was further separated by flash chromatography.
Completion of the C(1)-C(11) fragment of zincophorin was
done by reducing the ester, protecting the resulting hydroxyl
with a TIPS, followed by removal of the benzyloxy groups to
provide 60, the structure of which being confirmed by com-
parison to the spectroscopic data reported in the total synthesis
of Miyashita.
(28) (a) Song, Z.; Hsung, R. P.; Lu, T.; Lohse, A. G. J. Org. Chem. 2007,
72, 9722. (b) Song, Z.; Hsung, R. P. Org. Lett. 2007, 9, 2199. (c) Cossy, J.;
Meyer, C.; Defosseux, M.; Blanchard, N. Pure Appl. Chem. 2005, 77, 1131. (d)
Komatsu, K.; Tanino, K.; Miyashita, M. Angew. Chem., Int. Ed. 2004, 43, 4341.
(e) Defosseux, M.; Blanchard, N.; Meyer, C.; Cossy, J. J. Org. Chem. 2004, 69,
4626. (f) Defosseux, M.; Blanchard, N.; Meyer, C.; Cossy, J. Org. Lett. 2003,
5, 4037. (g) Cossy, J.; Blanchard, N.; Defosseux, M.; Meyer, C. Angew. Chem.,
Int. Ed. 2002, 41, 2144. (h) Guindon, Y.; Murtagh, L.; Caron, V.; Landry, S. R.;
Jung, G.; Bencheqroun, M.; Faucher, A.-M.; Gue´rin, B. J. Org. Chem. 2001,
66, 5427. (i) Booysen, J. F.; Holzapfel, C. W. Synth. Commun. 1995, 25, 1473.
(j) Cywin, C. L.; Kallmerten, J. Tetrahedron Lett. 1993, 34, 1103. (k)
Danishefsky, S. J.; Selnick, H. G.; Zelle, R. E.; DeNinno, M. P. J. Am. Chem.
Soc. 1988, 110, 4368.
72 J. Org. Chem. Vol. 74, No. 1, 2009

Mukaiyama Aldolization and Free Radical Reduction
SCHEME 8
128.18, 128.02, 127.96, 127.87, 127.84, 127.78, 127.6, 127.5, 127.4,
83.0, 75.4, 74.1, 73.5, 73.3, 68.1, 53.4, 37.3, 35.8, 25.0, 24.9, 11.6,
11.5, 11.4; MS (ESI) m/z 493 (MH+, 95), 475 (100), 369 (28), 277
(65); HRMS calcd for C25H34BrO5 (MH+) 493.1590, found
493.1598 (1.7 ppm). Diastereoisomer B: Colorless oil, Rf ) 0.17
(hexanes/EtOAc, 85:15); IR (neat) νmax ) 3529, 3062, 2949, 1735,
Conclusion
In this report, we have described the first direct synthesis of
all diastereomeric stereopentads. This integrated strategy could
be a valuable tool for the determination of stereochemistry of
unassigned structures of natural products, for the synthesis of a
polyketide library, and for the study of unnatural analogues of
important polyketides, which are all active areas of research.10,30
Our stereodivergent synthesis starts from readily available
materials which can be easily synthesized in both enantiomeri-
cally pure forms. Of crucial importance was the discovery of
the efficiency of TiCl3(OiPr) in Mukaiyama aldolization to form
a three-points-chelated intermediate. Central to our approach
was the usefulness of the free-radical-based reductions devel-
oped by our group. Finally, a first application of this strategy
was demonstrated with the stereoselective synthesis of a
C(1)-C(11) building block for the construction of zincophorin.
1
1453, 1267; H NMR (400 MHz, CDCl3) δ 7.35-7.26 (m, 10H),
4.62 (d, J ) 11.0 Hz, 1), 4.58 (d, J ) 10.9 Hz, 1H), 4.49 (s, 2H),
4.43 (d, J ) 5.8 Hz, 1H), 3.77 (s, 3H), 3.61 (dd, J ) 5.8, 5.8 Hz,
1H), 3.50 (m, 2H), 2.90 (d, J ) 5.8 Hz, 1H), 2.27-2.19 (m, 1H),
2.18-2.12 (m, 1H), 1.88 (s, 3H), 1.02 (d, J ) 6.9 Hz, 3H), 0.93
(d, J ) 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3) 172.2, 138.8,
138.7, 128.65, 128.61, 128.1, 128.0, 127.9, 127.8, 83.8, 75.6, 75.0,
73.34, 73.31, 63.8, 53.5, 36.1, 24.6, 12.2, 11.8; MS (ESI) m/z 493
(MH+, 100), 475 (90), 369 (28), 277 (55); HRMS calcd for
C25H34BrO5 (MH+) 493.1590, found 493.1590 (0.1 ppm).
General Procedure for the Mukaiyama ReactionsCram-
Chelate. (3R,4S,5R,6S)-Methyl-5,7-bis(benzyloxy)-2-bromo-3-
hydroxy-2,4,6-trimethylheptanoate (44a,b). To a cold (-78 °C)
solution of the appropriate aldehyde 6 (1 equiv) in dry CH2Cl2 (0.1
M) was added slowly a freshly prepared solution of TiCl3(OiPr)32
(0.8 M in dry CH2Cl2), and the mixture was stirred for 15 min at
-78 °C. Then, the bromoenoxysilane 24 (1.3 equiv) was added,
and the resulting solution was stirred at -78 °C until the aldehyde
was completely consumed, as determined by TLC. A saturated
aqueous solution of NH4Cl was poured into the reaction mixture.
After the aqueous layer was extracted with ether, the organic layer
was successively washed with a saturated aqueous solution of
NaHCO3 and brine. The organic layer was dried (MgSO4), filtered,
and concentrated. The residue was purified by flash chromatography
on silica gel using 15% EtOAc/hexanes to give the desired
compounds 44a,b (yield ) 72%). A 3:1 mixture of 2,3-diastere-
oisomers was observed on the basis of NMR data. Diastereoisomer
A: Colorless oil, Rf ) 0.25 (hexanes/EtOAc, 85:15); IR (neat) νmax
Experimental Section31
General Procedure for the Mukaiyama ReactionsFelkin-Anh.
(3S,4S,5R,6S)-Methyl-5,7-bis(benzyloxy)-2-bromo-3-hydroxy-
2,4,6-trimethylheptanoate (43a,b). To a cold (-78 °C) solution
of aldehyde 6 (1 equiv) in dry CH2Cl2 (0.1 M) was added
bromoenoxysilane 24 (1.3 equiv). The mixture was stirred for 1
min at -78 °C then BF3 ·OEt (1.5 equiv) was added slowly. The
resulting solution was stirred for 2 h until the aldehyde was
completely consumed, as determined by TLC. A saturated aqueous
solution of NH4Cl was poured into the reaction mixture. After the
aqueous layer was extracted with ether, the organic layer was
successively washed with a saturated aqueous solution of NaHCO3
and brine. The organic layer was dried (MgSO4), filtered, and
concentrated. The residue was purified by flash chromatography
on silica gel using 20% EtOAc/hexane to give the desired
compounds 43a,b (yield ) 94%). A 3:1 mixture of 2,3-syn:2,3-
anti was observed on the basis of NMR data. Diastereoisomer A:
1
) 3455, 2949, 1737, 1452, 1259; H NMR (400 MHz, CDCl3) δ
7.38-7.28 (m, 10H), 4.65 (d, J ) 11.4 Hz, 1H), 4.50 (d, J ) 11.4
Hz, 1H), 4.49 (d, J ) 11.9 Hz, 1H), 4.45 (d, J ) 11.9 Hz, 1H),
4.23 (d, J ) 9.1 Hz, 1H), 3.93 (br s, 1H), 3.82 (dd, J ) 3.0, 6.0
Hz, 1H), 3.79 (s, 3H), 3.37 (d, J ) 3.4 Hz, 2H), 2.16-2.09 (m,
1H), 2.06-1.99 (m, 1H), 1.87 (s, 3H), 0.95 (d, J ) 7.0 Hz, 3H),
0.77 (d, J ) 7.2 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 171.5,
138.3, 138.1, 128.7, 128.4, 128.3, 127.8, 127.7, 127.6, 82.0, 75.8,
73.8, 73.5, 73.1, 67.2, 52.9, 39.5, 35.4, 21.2, 12.6, 11.9; MS (ESI)
m/z 493 (MH+, 100), 475 (21), 345 (90), 323 (60); HRMS calcd
for C25H34BrO5 (MH+) 493.1590, found 493.1583 (-1.2 ppm).
Diastereoisomer B: Colorless oil, Rf ) 0.18 (hexanes/EtOAc, 85:
15); IR (neat) νmax ) 3437, 3030, 2933, 1743, 1453, 1256; 1H NMR
(400 MHz, CDCl3) δ 7.36-7.28 (m, 10H), 4.64 (d, J ) 11.4 Hz,
1), 4.54 (d, J ) 11.4 Hz, 1H), 4.49 (d, J ) 11.8 Hz, 1H), 4.45 (d,
Colorless oil, Rf ) 0.29 (hexanes/EtOAc, 85:15); IR (neat) νmax
)
3529, 3062, 2949, 1735, 1453, 1268; 1H NMR (400 MHz, CDCl3)
δ 7.35-7.27 (m, 10H), 4.57 (s, 2H), 4.48 (s, 2H), 4.34 (d, J ) 4.4
Hz, 1H), 3.69 (s, 3H), 3.58 (dd, J ) 4.0, 7.0 Hz, 1H), 3.45 (dd, J
) 7.6, 9.0 Hz, 1H), 3.37 (dd, J ) 5.6, 9.0 Hz, 1H), 2.90 (d, J )
4.4 Hz, 1H), 2.17-2.09 (m, 1H), 1.92-1.86 (m, 1H), 1.89 (s, 3H),
0.99 (d, J ) 6.9 Hz, 3H), 0.95 (d, J ) 6.9 Hz, 3H); 13C NMR (100
MHz, CDCl3) δ 171.8, 138.9, 138.6, 129.4, 129.4, 129.25, 129.20,
129.1, 129.07, 128.99, 128.94, 128.8, 128.64, 128.56, 128.22,
(29) Evans, P. A.; Cui, J.; Gharpure, S. J.; Hinkle, R. J. J. Am. Chem. Soc.
2003, 125, 11456.
(30) Smith, A. B., III; Freeze, B. S. Tetrahedron 2008, 64, 261.
J. Org. Chem. Vol. 74, No. 1, 2009 73

Brazeau et al.
J ) 11.9 Hz, 1H), 4.28 (d, J ) 4.2 Hz, 1H), 4.01 (dd, J ) 4.1, 8.4
Hz, 1H), 3.79-3.72 (m, 1H), 3.74 (s, 3H), 3.41-3.33 (m, 2H),
2.20-2.12 (m, 2H), 1.91 (s, 3H), 0.97 (d, J ) 7.0 Hz, 3H), 0.95
(d, J ) 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 172.0, 138.4,
138.1, 128.7, 128.6, 128.1, 128.0, 127.9, 83.1, 74.3, 73.8, 73.4,
53.4, 40.7, 36.1, 26.4, 14.7, 11.9; MS (ESI) m/z 493 (MH+, 100),
345 (55); HRMS calcd for C25H34BrO5 (MH+) 493.1590, found
493.1596 (1.3 ppm).
(0.2 equiv of a 1.0 M solution in hexane). The resulting solution
was stirred at -78 °C, and 0.2 equiv of Et3B then air were added
each 30 min until the reaction was judged complete by TLC (around
3 h). 1,4-Dinitrobenzene (0.2 equiv) was then added to the solution,
and the mixture was stirred for an additional 15 min at -78 °C. A
saturated aqueous solution of NH4Cl was poured into the reaction
mixture. After the aqueous layer was extracted with ether, the
organic layer was successively washed with saturated aqueous
solution of NaHCO3 and brine. The organic layer was dried
(MgSO4), filtered, and concentrated. To recover the resultant free
alcohol, addition of H2O2 (30%, 1.5 mL/mmol of starting material
in 5 mL of MeOH) in the reaction mixture was necessary. After
the resultant solution was stirred for 2 h at 0 °C, organic solvents
were then evaporated. The reaction mixture was quenched with
brine and extracted with CH2Cl2 (3×). The combined organic
extracts were dried over MgSO4, filtered, and concentrated under
reduced pressure. The crude product was dissolved in ether (0.5
M) and mixed vigorously for 2 h with KF/Celite (0.75 g/mmol of
substrate) to eliminate tin residues. The reaction mixture was filtered
and concentrated under reduced pressure. The residue was purified
by flash chromatography on silica gel using 25% EtOAc/hexanes
to give the desired compound 22 (yield ) 79%). The ratio of 2,3-
anti:2,3-syn was found to be >20:1 on the basis of NMR data:
General Procedure for Radical Reduction under Chelation-
ControlsThe Endocyclic Effect. (2S,3R,4S,5R,6S)-Methyl-5,7-
bis(benzyloxy)-3-hydroxy-2,4,6-trimethylheptanoate (21). To a
stirred solution of the R-bromoesters 44a,b (1 equiv) in dry CH2Cl2
(0.1 M) at -78 °C was added AlMe3 (2.0 M in toluene, 3.0 equiv).
The mixture was stirred for 1 h at the same temperature before
adding Bu3SnH (1.8 equiv) and Et3B (0.2 equiv of a 1.0 M solution
in hexane). The resulting solution was stirred at -78 °C, and 0.2
equiv of Et3B then air were added each 30 min until the reaction
was judged complete by TLC (around 3 h). 1,4-Dinitrobenzene (0.2
equiv) was then added to the solution, and the mixture was stirred
for an additional 15 min at -78 °C. A saturated aqueous solution
of NH4Cl was carefully poured into the reaction mixture. After the
aqueous layer was extracted with ether, the organic layer was
successively washed with saturated aqueous solution of NaHCO3
and brine. The organic layer was dried (MgSO4), filtered, and
concentrated. The crude product was dissolved in ether (0.5 M)
and mixed vigorously for 2 h with KF/Celite33 (0.75 g/mmol of
substrate) to eliminate tin residues. The reaction mixture was filtered
and concentrated under reduced pressure. The residue was purified
by flash chromatography on silica gel using 25% EtOAc/hexanes
to give the desired compound 21 (yield ) 78%). The ratio of 2,3-
syn:2,3-anti was found to be >20:1 on the basis of NMR data:
Colorless oil, Rf ) 0.15 (hexanes/EtOAc, 80:20); IR (neat) νmax
)
1
3498, 3030, 2972, 1737, 1455; H NMR (400 MHz, CDCl3) δ
7.36-7.26 (m, 10H), 4.63 (d, J ) 11.0 Hz, 1H), 4.59 (d, J ) 11.0
Hz, 1H), 4.53 (d, J ) 12.0 Hz, 1H), 4.46 (d, J ) 12.0 Hz, 1H),
4.09 (dd, J ) 4.3, 9.7 Hz, 1H), 3.70 (s, 3H), 3.67 (t, J ) 5.6 Hz),
3.45 (dd, J ) 6.4, 9.2 Hz, 1H), 3.38 (dd, J ) 5.4, 9.2 Hz, 1H),
3.25 (d, J ) 2.8 Hz, 1H), 2.58 (qd, J ) 7.1, 9.7 Hz, 1H), 2.22-2.12
(m, 1H), 1.85-1.78 (m, 1H), 1.06 (d, J ) 6.9 Hz, 3H), 0.98 (d, J
) 7.3 Hz, 3H), 0.96 (d, J ) 7.6 Hz, 3H); 13C NMR (100 MHz,
CDCl3) δ 177.0, 138.6, 128.6, 128.0, 127.9, 84.3, 75.9, 73.3, 73.3,
72.6, 52.0, 43.9, 36.6, 36.1, 14.0, 12.9, 10.5; MS (ESI) m/z 437
(M + Na, 100), 415 (MH+, 88), 397 (60); HRMS calcd for
C25H35O5 (MH+) 415.2484, found 415.2493 (2.0 ppm).
Colorless oil, Rf ) 0.17 (hexanes/EtOAc, 80:20); IR (neat) νmax
)
1
3476, 3029, 2973, 1735, 1455; H NMR (400 MHz, CDCl3) δ
7.35-7.27 (m, 10H), 4.65 (d, J ) 11.4 Hz, 1H), 4.50 (d, J ) 11.4
Hz, 1H), 4.49 (d, J ) 12.0 Hz, 1H), 4.45 (d, J ) 11.9 Hz, 1H),
4.04 (td, J ) 2.6, 9.4 Hz, 1H), 3.88 (br s, 1H), 3.77 (dd, J ) 2.4,
7.0 Hz, 1H), 3.71 (s, 3H), 3.37 (dd, J ) 1.6, 6.3 Hz, 1H), 3.36 (s,
1H), 2.64 (qd, J ) 3.7, 6.7 Hz, 1H), 2.17-2.11 (m, 1H), 1.97-1.88
(m, 1H), 1.14 (d, J ) 7.1 Hz, 3H), 0.94 (d, J ) 7.0 Hz, 3H), 0.86
(d, J ) 6.9 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 176.2, 138.5,
138.3, 128.7, 128.6, 128.0, 127.9, 83.0, 74.6, 74.3, 73.8, 73.3, 52.0,
42.3, 38.7, 35.9, 13.7, 11.7, 8.9; MS (ESI) m/z 415.1 (MH+, 100),
307.1 (25), 181.0 (50); HRMS calcd for C25H35O5 (MH+) 415.2484,
found 415.2479 (1.3 ppm).
Acknowledgment. The authors wish to express their gratitude
to the Natural Sciences and Engineering Research Council of
Canada (NSERC) and the Institut de recherches cliniques de
Montre´al (IRCM) for their financial support. Fellowship support
from NSERC PGS D and FQRNT B2 (J.-F.B) is gratefully
acknowledged.
General Procedure for Radical Reduction Acyclic Effect.
(2R,3RS,4S,5R,6S)-Methyl-5,7-bis(benzyloxy)-3-hydroxy-2,4,6-
trimethylheptanoate (22). To a stirred solution of the R-bro-
moesters 44a,b (1 equiv) in dry CH2Cl2 (0.1 M) at -78 °C were
added iPr2NEt and Bu2BOTf. The mixture was stirred for 1 h at
the same temperature before adding Bu3SnH (1.5 equiv) and Et3B
Supporting Information Available: Experimental proce-
dures and characterization data for compounds 5, 6, 15-22,
28, 31-34, 41-45, 47, 49-51, 53-59, and 60-65; determi-
nation of relative configuration for compounds 15-22; experi-
mental procedure for 13C NMR spectra of compound 3 at low
temperature with TiCl3(OiPr); CIF data for compound 59. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(31) See the Supporting Information for general methods and procedures.
(32) Solsona, J. G.; Romea, P. D.; Urpy´, F.; Vilarrasa, J. Org. Lett. 2003, 5,
519.
(33) Ando, T.; Yamawaki, J. Chem. Lett. 1979, 45.
JO8021583
74 J. Org. Chem. Vol. 74, No. 1, 2009