Highly diastereoselective, tandem, three-component synthesis of tetrahydrofurans from ketoaldehydes via silylated β-lactone intermediates

Article abstract of DOI:10.1002/anie.200800235

(Chemical Equation Presented) Like falling dominoes! A novel tandem, three-component reaction is described that generates up to two C-C bonds, one C-O bond, and three additional stereocenters leading to substituted tetrahydrofuran units. This process involves a tandem Mukaiyama aldol-lactonization/reductive cyclization and proceeds via a silylated β-lactone intermediate. The method was applied to prepare the tetrahydrofuran fragment of colopsinol B. Py = 2-pyridyl.

Full text of DOI:10.1002/anie.200800235

Communications
DOI: 10.1002/anie.200800235
Tandem Synthesis
Highly Diastereoselective, Tandem, Three-Component Synthesis of
Tetrahydrofurans from Ketoaldehydes via Silylated b-Lactone
Intermediates**
T. Andrew Mitchell, Cunxiang Zhao, and Daniel Romo*
In memory of Albert I. Meyers
Processes that form multiple bonds and stereocenters in a
single reaction, without the isolation of intermediates, are
known as tandem, domino, multicomponent, or cascade
reactions.[1] They are powerful complexity-building reactions.
We have developed stereoselective routes to both cis[2] and
trans[3] b-lactones 5 through tandem Mukaiyama aldol-
lactonization (TMAL) processes[4] between thiopyridyl
ketene acetal 2 and aldehyde 1 (Scheme 1, pathway a). This
methodology has been utilized in total syntheses of
(ꢀ)-panclicin D,[5] orlistat and its congeners,[6] okinonellins,[7]
brefeldin A,[8] and belactosin C.[9] In the course of these
studies, we identified several by-products (e.g. b-chlorosilyl-
ester 4; Scheme 1, pathway b) that led us to propose the
silylated b-lactone intermediate 3 in the TMAL process. Thus
we considered methods for intercepting these intermediates
to enable the study of useful complexity-building processes.
Mead and Pillai have reported Lewis acid promoted
reductive cyclizations of simple keto-b-lactones for tetra-
hydrofuran (THF) synthesis. This approach suggested the
possibility of utilizing aldehyde substrates bearing pendant
ketones that could undergo reductive cyclization in the
TMAL process (Scheme 2).[10] The THF motif is commonly
found in natural products, and while several approaches
ꢀ
toward these heterocycles exist, many routes rely on C O
bond formation of relatively complex substrates or proceed
through oxocarbenium ions derived from O-glycosides.[11]
Herein we describe the development of a tandem, three-
component synthesis of THFs 9 from g-ketoaldehydes 6,
thiopyridyl ketene acetals 2, and silyl nucleophiles in which up
ꢀ
ꢀ
to two C C bonds, one C O bond, and three new stereocen-
ters are generated.
Scheme 2. Proposed three-component synthesis of tetrahydrofuran 9
from ketoaldehyde 6, thiopyridyl ketene acetal 2, and a silyl nucleophile
via a postulated silylated b-lactone intermediate 7.
Scheme 1. Synthesis of b-lactone 5 (pathway a) or b-chloro silyl ester 4
(pathway b) via a postulated silylated b-lactone intermediate 3 in the
tandem Mukaiyama aldol-lactonization process. Py=2-pyridyl.
Initially, we sought further evidence for the postulated
intermediacy of the silylated b-lactone 3 in the TMAL
process. Previously we reported a correlation between the size
of the silyl group of ketene acetals 2a–c and product
distribution leading to either b-lactone 5a (20–66%) or
b-chlorosilyl esters 4a–c (5–40%; Scheme 3).[5] However, in
[*] Dr. T. A. Mitchell, Dr. C. Zhao, Prof. D. Romo
Department of Chemistry, Texas A&M University
P.O. Box 30012, College Station
TX 77842-3012 (USA)
Fax: (+1)979-862-4880
E-mail: romo@tamu.edu
Homepage: http://www.chem.tamu.edu/rgroup/romo
[**] We thank the NIH (GM 069784) and the Welch Foundation (A-1280)
for funding this work. We thank Kay Morris and Dr. Ziad Moussa, as
well as Profs. Dan Singleton and Brian Connell for helpful
discussions.
Scheme 3. Effect of varying the silyl group of ketene acetals 2a–d on
the product distribution of the TMAL process with octanal (1a). SiR3:
a: TES (triethylsilyl); b: TBS (tert-butyldimethylsilyl); c: TIPS (triiso-
propylsilyl); d: TBDPS (tert-butyldiphenylsilyl), R1 =CH3(CH2)6.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5026
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 5026 –5029

Angewandte
Chemie
some cases the silyl esters 4a–c were accompanied by the
corresponding acids, thus raising the possibility of subsequent
formation of silyl esters by a silylating agent generated
in situ.[10b] Use of the more robust TBDPS ketene acetal 2d
gave the b-chloro silyl ester 4d exclusively, which was isolated
in 56% yield with no acid detected by NMR analysis of the
crude product. This result provides strong evidence for the
silylated b-lactone intermediate 3 (Scheme 1).
substantially improved yield (42%; Table 1, entry 6). A large
excess of ZnCl2 and Et3SiH led to 10a with a greatly improved
ratio to 11a but in a slightly decreased yield (38%; Table 1,
entry 7). Finally, optimal conditions were achieved at 08C
using 4.0 equivalents of ZnCl2 and 50 equivalents of Et3SiH to
deliver 10a which was isolated in 54% yield (Table 1,
entry 8). This overall yield corresponds to approximately
82% yield per step (over 4 steps), and includes subsequent
reduction of the initially formed silyl ester (cf. 9, Scheme 2).
Several substituted racemic g-ketoaldehydes (ꢁ )-6b–f
were also investigated and all gave similar overall yields for
10b–f as well as high diastereoselectivity (Table 2).
With this result in hand, we began the study of the
proposed tandem Mukaiyama aldol-lactonization/reductive
cyclization process (called “the tandem process” throughout)
by employing the more readily prepared TIPS (versus
TBDPS) ketene acetal (E)-2e (E/Z 4:1) and racemic
g-ketoaldehyde (ꢁ )-6a (see the reaction in Table 1). The
latter substrate was chosen to ensure high chelation-con-
trolled selectivity in the initial Mukaiyama aldol reaction[12]
and high facial selectivity in the subsequent oxocarbenium
reduction as predicted by the model of Woerpel and co-
workers.[13] The substrates, ketoaldehydes (ꢁ )-6, were pre-
pared by oxidation of the corresponding diol.[14] Building on
our previous extensive studies of the TMAL process, we
utilized precomplexation of the ketene acetal (E)-2e with
ZnCl2, which modulates the Lewis acidity and increases the
rate of the TMAL process.[15] Direct reduction of the crude
silyl ester (cf. 9, Scheme 2) to the primary alcohol with
DIBALH simplified the purification. Regardless of the
amount of Et3SiH used, our initial attempts unexpectedly
delivered furan 11a as the major product (Table 1, entries 1–
4). Although initial yields of THF 10a were unsatisfactory,
only one diastereomer was observed as anticipated, based on
chelation-controlled selectivity in the initial aldol/SN2 inver-
sion during b-lactone ring opening, and the Woerpel model
for oxocarbenium reduction. The relative configuration of
10a was confirmed by X-ray analysis of the corresponding
para-bromobenzoate.[16] Initial reaction at 08C with warming
to ambient temperature improved the ratio of 10a to 11a;
however, the furan was still the major product (Table 1,
entry 5). Increasing the amount of Lewis acid to 4.0 equiv-
alents ultimately delivered 10a as the major product in
Table 2: Synthesis of tetrahydrofurans 10b–f by the tandem process
from various g-ketoaldehydes (ꢁ)-6b–f.
Entry Ketoaldehyde Major adduct
10/11[a] Yield d.r.[a]
[%][b]
1
2
3
4
5
(ꢁ)-6b
(ꢁ)-6c
(ꢁ)-6d
(ꢁ)-6e
(ꢁ)-6 f
1.3:1
2.3:1
3.0:1
3.5:1
2.2:1
42
52
49
54
>19:1
>19:1
>19:1
>19:1
49[c] >19:1
Table 1: Optimization of the tandem process to THF 10a.
[a] Determined by 1H NMR spectroscopic analysis (300 MHz) of the
crude product. [b] Yield of isolated product (over 2 steps). [c] The
reaction mixture for the DIBALH reduction was slowly warmed from
ꢀ78!ꢀ308C over 6 h to prevent cleavage of the PMB group. PMB=
para-methoxybenzyl.
Entry
ZnCl2
[equiv]
(E)-2e
[equiv]
Et3SiH
[equiv]
T [8C]
10a/11a[a]
(% yield 10a)[b]
The tandem process was also successful with b-oxygen-
ated ketoaldehydes (for example 6g) and ketene acetal
(E)-2e (E/Z > 19:1) and gave good diastereoselectivity
(d.r. 9:1) for the desired THF 10g, albeit with a reduced
yield (Scheme 4). The stereochemical outcome of the initial
Mukaiyama aldol reaction was consistent with the model
proposed by Evans et al.[17] for additions to b-silyloxy
aldehydes, and was as observed in previous TMAL reac-
tions.[5,6]
1
2
3
4
5
6
7
8
2.0
2.0
2.0
2.0
2.0
4.0
8.0
4.0
2.0
2.0
2.0
2.0
2.0
1.2
1.2
1.2
0.0
2.0
23
23
23
23
0!23
0!23
0!23
0
only 11a (0)
1.0:3.5 (11)
1.0:3.5 (9)
1.0:3.5 (10)
1.0:2.0 (24)
2.0:1.0 (42)
9.0:1.0 (38)
6.2:1.0 (54)
10.0
50.0
10.0
10.0
100.0
50.0
[a] Determined by 1H NMR spectroscopic analysis (300 MHz) of the
crude product. [b] Yield of isolated product (over 2 steps). Bn=benzyl,
DIBALH= diisobutylaluminum hydride.
Pleasingly, allylsilane could also be utilized as the
nucleophilic component to deliver 10h as a single diastereo-
Angew. Chem. Int. Ed. 2008, 47, 5026 –5029
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5027

Communications
Subsequent chelation-controlled addition leads to the
observed high diastereoselectivity in the aldol step providing
ꢀ
silylated b-lactone 16. Subsequent alkyl C O cleavage with
inversion delivers 17. Oxocarbenium 17 adopts the stereo-
electronically favorable envelope conformation with the
benzyloxy substituent in a pseudoaxial orientation and
reduction occurs from the “inside” of the envelope, as
predicted by the Woerpel model,[13] to finally give silyl ester
18.
To highlight the utility of this methodology, we synthe-
sized the THF fragment of colopsinol B (19) (Scheme 7).[21]
Thus, a-benzyloxy-g-ketoaldehyde (ꢁ )-6 f was utilized, under
typical conditions with acetate ketene acetal 2c, to deliver the
Scheme 4. The tandem process with a b-silyloxy aldehyde (ꢁ)-6g.
mer, accompanied by increased quantities of 11a (Scheme 5).
ꢀ
Importantly, an additional C C bond is constructed with
excellent stereocontrol of the resulting quaternary center.
Scheme 5. Use of allyltrimethylsilane as the nucleophile leading to
formation of a quaternary carbon in the tandem process.
A mechanistic pathway that rationalizes the stereochem-
ical outcome of this tandem process is proposed based on the
current experimental evidence for a chelation-controlled
TMAL process,[12,15] as well as the Woerpel model for
oxocarbenium reductions[13] (Scheme 6). Precoordination of
ZnCl2 and ketene acetal 2e leads to the tetrahedral complex
12.[18] Initial monodentate coordination with the a-benzyloxy
aldehyde (ꢁ )-6a leads to a trigonal-bipyramidal complex
13,[19] which involves bidentate chelation of zinc(II) with the
thiopyridyl group.[20] A highly ordered, boatlike transition
state arrangement 14 is generated by ligand rearrangement
and leads to bidentate coordination of the aldehyde (ꢁ )-6a.
Scheme 7. Synthesis of the THF fragment 10i of colopsinol B (19) b y
employing the tandem process.
alcohol 10i as a mixture of diastereomers ( ꢂ 3:1.3:1, ꢂ 42%).
This outcome was expected because of the low diastereose-
lectivity in the initial aldol step. However, the major
diastereomer could be readily separated and was isolated in
23% yield ( ꢂ 70% per step over 4 steps). The configuration
of 10i was confirmed by coupling constant and nOe analysis,
and these data also correlated well with that reported for the
THF unit of colopsinol B (19).[16]
In summary, we have developed a diastereoselective,
three-component synthesis of substituted THF units that
employ g-ketoaldehydes, ketene acetals, and silyl nucleo-
philes and that proceeds through silylated b-lactone inter-
mediates. The overall efficiency for a-benzyloxy-g-ketoalde-
hydes is good, with typical yields in the range of 42 to 54%,
which corresponds to ꢂ 82% yield per step for the tandem,
three-component process and the subsequent reduction. The
observed high diastereoselectivity and relative stereochemi-
cal induction are consistent with a boat transition state
arrangement in the TMAL process, SN2 opening of the
intermediate b-lactone by the pendant ketone, and the
Woerpel model for “inside attack” of five-membered oxo-
carbenium ions. The rapid assembly of a THF fragment of
colopsinol B from simple starting materials demonstrates the
utility of this methodology.
Scheme 6. Mechanistic proposal for the tandem Mukaiyama aldol-
lactonization/reductive cyclization process. R=CH2C(O)CH3.
Received: January 16, 2008
Published online: May 27, 2008
5028 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 5026 –5029

Angewandte
Chemie
118, 2150; Angew. Chem. Int. Ed. 2006, 45, 2096; m) L. V. R.
Keywords: cyclization · Mukaiyama aldol · oxocarbenium ions ·
synthetic methods · tandem reactions
.
Reddy, A. D. Roy, R. Roy, A. K. Shaw, Chem. Commun. 2006,
3444; n) M. Barbasiewicz, M. Makosza, Synthesis 2006, 1190.
[12] H. W. Yang, D. Romo, J. Org. Chem. 1998, 63, 1344.
[13] a) A. Schmitt, H.-U. Reissig, Chem. Ber. 1995, 128, 871; b) C. H.
Larsen, B. H. Ridgway, J. T. Shaw, K. A. Woerpel, J. Am. Chem.
Soc. 1999, 121, 12208; c) C. H. Larsen, B. H. Ridgway, J. T. Shaw,
D. M. Smith, K. A. Woerpel, J. Am. Chem. Soc. 2005, 127, 15521
and references therein.
[1] For reviews on tandem reactions, see: a) Domino Reactions in
Organic Synthesis (Eds.: L. F. Tietze, H. P. Bell, G. Brasche),
Wiley-VCH, Weinheim, 2006; b) D. Enders, C. Grondal,
M. R. M. Huttl, Angew. Chem. 2007, 119, 1590; Angew. Chem.
Int. Ed. 2007, 46, 1570; c) K. C. Nicolaou, T. Montagnon, S. A.
Snyder, Chem. Commun. 2003, 551.
[14] S. E. Denmark, J. Fu, Org. Lett. 2002, 4, 1951.
[15] C. Zhao, Ph.D. Thesis, Texas A&M University (USA), 2001.
[16] See the Supporting Information for stereochemical proof.
[17] D. A. Evans, M. J. Dart, J. L. Duffy, M. G. Yang, J. Am. Chem.
Soc. 1996, 118, 4322.
[18] A homogeneous solution results after stirring a 1:1 ratio of
thiopyridyl ketene acetal and ZnCl2. 1H NMR spectroscopic
analysis supports the formation of a tetrahedral complex. See
Ref. [15]. Mechanistic studies of the TMAL reaction will be
described in a subsequent full paper.
[2] Y. Wang, C. Zhao, D. Romo, Org. Lett. 1999, 1, 1197.
[3] H. W. Yang, D. Romo, J. Org. Chem. 1997, 62, 4.
[4] a) T. Mukaiyama, K. Narasaka, K. Banno, Chem. Lett. 1973,
1011; b) T. Mukaiyama, K. Banno, K. Narasaka, J. Am. Chem.
Soc. 1974, 96, 7503 – 7509; c) C. H. Heathcock, Science 1981, 214,
395; d) T. Mukaiyama, Aldrichimica Acta 1996, 29, 59.
[5] H. W. Yang, C. Zhao, D. Romo, Tetrahedron 1997, 53, 16471.
[6] G. Ma, M. Zancanella, Y. Oyola, R. D. Richardson, J. W. Smith,
D. Romo, Org. Lett. 2006, 8, 4497.
[19] Although tetrahedral zinc(II) complexes are the most common,
several examples of trigonal-bipyramidal zinc(II) are known,
see: a) R. H. Prince in Comprehensive Coordination Chemistry:
Zinc and Cadmium, Vol. 5 (Ed.: G. Wilkinson), Pergamon, New
York, 1987, pp. 925 – 1045; b) R. R. Holmes in Progress in
Inorganic Chemistry, Vol. 32 (Ed.: S. J. Lippard), Wiley, New
York, 1984, pp. 119 – 236; c) W. N. Lipscomb, N. Strater, Chem.
Rev. 1996, 96, 2375; d) C. O. Rodriguez de Barbarin, N. A.
Bailey, D. E. Fenton, Q.-Y. He, J. Chem. Soc. Dalton Trans.
1997, 161; e) W. Oppolzer, R. N. Radinov, Tetrahedron Lett.
1988, 29, 5645.
[20] For examples of zinc(II) complexes containing a bidentate ligand
that forms a four-membered ring with the metal center, see:
a) M. Bonamico, G. Dessy, V. Fares, L. Scaramuzza, J. Chem.
Soc. Dalton Trans. 1972, 2515; b) C. C. Ashworth, N. A. Bailey,
M. Johnson, J. A. McCleverty, N. Morrison, B. Tabbiner, J.
Chem. Soc. Chem. Commun. 1976, 743; c) B. Becker, K.
Radacki, W. Wojnowski, J. Organomet. Chem. 1996, 521, 39;
d) K. A. Fraser, M. M. Harding, Acta Crystallogr. 1967, 22, 75.
[21] T. Kubota, M. Tsuda, M. Takahashi, M. Ishibashi, H. Naoki, J.
Kobayashi, J. Chem. Soc. Perkin Trans. 1 1999, 3483. To the best
of our knowledge, no studies toward the synthesis of this natural
product have been reported.
[7] W. D. Schmitz, B. Messerschmidt, D. Romo, J. Org. Chem. 1998,
63, 2058.
[8] Y. Wang, D. Romo, Org. Lett. 2002, 4, 3231.
[9] S. Cho, D. Romo, Org. Lett. 2007, 9, 1537.
[10] a) K. T. Mead, S. K. Pillai, Tetrahedron Lett. 1993, 34, 6997;
b) for extension of this methodology to more highly substituted
THFs using b-lactones by the TMAL process, see: T. A.
Mitchell, D. Romo, J. Org. Chem. 2007, 72, 9053.
[11] a) For a recent review of THF synthesis, see: J. P. Wolfe, M. B.
Hay, Tetrahedron 2007, 63, 261; b) For selected recent examples,
see: C. G. Nasveschuk, T. Rovis, J. Org. Chem. 2008, 73, 612;
c) S.-H. Kim, S.-J. Oh, P.-S. Ho, S.-C. Kang, K.-J. O, C.-M. Yu,
Org. Lett. 2008, 10, 265; d) F. Colobert, S. Choppin, L. Ferreiro-
Mederos, M. Obringer, S. L. Arrata, A. Urbano, M. C. Carreno,
Org. Lett. 2007, 9, 4451; e) A. Rajender, H.-J. Gais, Org. Lett.
2007, 9, 579; f) A. M. Bernard, A. Frongia, R. Guillot, P. P. Piras,
F. Secci, M. Spiga, Org. Lett. 2007, 9, 541; g) B. Alcaide, P.
Almendros, T. M. del Campo, Angew. Chem. 2007, 119, 6804;
Angew. Chem. Int. Ed. 2007, 46, 6684; h) L. V. Desai, M. S.
Sanford, Angew. Chem. 2007, 119, 5839; Angew. Chem. Int. Ed.
2007, 46, 5737; i) U. Jahn, D. Rudakov, Org. Lett. 2006, 8, 4481;
j) J. A. Marshall, A. M. Mikowski, Org. Lett. 2006, 8, 4375;
k) S. N. Chavre, H. Choo, J. H. Cha, A. N. Pae, K. I. Choi, Org.
Lett. 2006, 8, 3617; l) A. Blanc, F. D. Toste, Angew. Chem. 2006,
Angew. Chem. Int. Ed. 2008, 47, 5026 –5029
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 5029