A catalytic asymmetric bioorganic route to enantioenriched tetrahydro- and dihydropyranones

Article abstract of DOI:10.1021/ja043925d

A conceptually novel approach to hetero Diels-Alder adducts of carbonyl compounds is described using as the key steps an antibody-mediated kinetic resolution of hydroxyenones followed by a ring-closure process. Various β-hydroxyenones proved to be very good substrates for antibodies 84G3- and 93F3-catalyzed retro-aldol reactions, allowing the preparation of highly enantiomerically enriched (up to 99% ee) precursors of pyranones. An attractive feature of this methodology is the possibility to convert these acyclicenantioenriched β-hydroxyenones into tetrahydropyranones by a conventional Michael-type addition procedure or into the corresponding dihydropyranones using an alternative palladium-catalyzed oxidative ring closure. For the palladium-mediated cyclization, a biphasic system has been implemented that allows the direct preparation of enantiopure dihydropyranones from the corresponding racemic aldol precursors using a sequential antibody-resolution/palladium-cyclization strategy, without isolation of the intermediate enantioenriched hydroxyenones. This bioorganic route is best applied to the preparation of hetero Diels-Alder adducts otherwise derived from less nucleophilic dienes and unactivated dienophiles.

Full text of DOI:10.1021/ja043925d

Published on Web 01/15/2005
A Catalytic Asymmetric Bioorganic Route to Enantioenriched
Tetrahydro- and Dihydropyranones
Charles Baker-Glenn,† Neil Hodnett,† Maud Reiter,† Sandrine Ropp,†
Rachael Ancliff,‡ and Ve´ronique Gouverneur*,†
Contribution from the Chemistry Research Laboratory, UniVersity of Oxford,
Mansfield Road, Oxford OX1 3TA (UK), and GlaxoSmithKline, Medicines Research Centre,
Gunnels Wood Road, SteVenage SG1 2NY (UK)
Received October 6, 2004; E-mail: veronique.gouverneur@chemistry.ox.ac.uk
Abstract: A conceptually novel approach to hetero Diels-Alder adducts of carbonyl compounds is described
using as the key steps an antibody-mediated kinetic resolution of hydroxyenones followed by a ring-closure
process. Various â-hydroxyenones proved to be very good substrates for antibodies 84G3- and 93F3-
catalyzed retro-aldol reactions, allowing the preparation of highly enantiomerically enriched (up to 99% ee)
precursors of pyranones. An attractive feature of this methodology is the possibility to convert these acyclic-
enantioenriched â-hydroxyenones into tetrahydropyranones by a conventional Michael-type addition
procedure or into the corresponding dihydropyranones using an alternative palladium-catalyzed oxidative
ring closure. For the palladium-mediated cyclization, a biphasic system has been implemented that allows
the direct preparation of enantiopure dihydropyranones from the corresponding racemic aldol precursors
using a sequential antibody-resolution/palladium-cyclization strategy, without isolation of the intermediate
enantioenriched hydroxyenones. This bioorganic route is best applied to the preparation of hetero Diels-
Alder adducts otherwise derived from less nucleophilic dienes and unactivated dienophiles.
natural compounds.1 Only more recently have Jacobsen et al.
developed a metal-catalyzed enantioselective HDA reaction
Introduction
The hetero Diels-Alder (HDA) reaction between carbodienes
and carbonyl compounds has been the focus of extensive studies
and has emerged as an important target for asymmetric catalysis.
Successes reported in this area have mostly involved the reaction
of electron-rich dienes such as 1-methoxy-3-(trimethylsilyloxy)-
butadiene (Danishefsky’s diene) with both activated and unac-
tivated dienophiles, including ketones. The products resulting
from these reactions are dihydropyranones, which are attractive
substrates for the synthesis of carbohydrates and numerous other
allowing the use of less nucleophilic dienes, bearing fewer than
two oxygen substituents, combined with unactivated carbonyl
compounds. These metal-catalyzed asymmetric HDA reactions
allow ultimate access to structurally diverse tetrahydropyranones.
Such transformations are successfully performed in the presence
of chiral tridentate chromium(III) catalysts leading to enantio-
enriched all syn products possessing up to three stereocenters.2
Nature also appears to have used the Diels-Alder reaction in
the biosynthesis of several secondary metabolites,3 and therefore
it is not surprising that numerous biocatalysts have been prepared
for the assembly of novel carbocyclic and heterocyclic products.
The immune system has proven to be a rich source of unique
Diels-Alderase antibodies as a result of ingenious approaches
to elicit catalysts that are even able to promote disfavored
reaction pathways preferentially over the other possible reaction
outcomes.4 More recently, modified RNA fragments have also
been shown to be effective Diels-Alderase catalysts that seem
to operate through Lewis acid-type catalysis, a different
mechanistic mode to catalytic antibodies.5 Surprisingly, none
of these biopolymer catalysts were generated for hetero Diels-
Alder reaction of carbonyl compounds, with the exception of a
polyclonal antibody preparation that was shown to catalyze the
cycloaddition of a carbodiene with ethyl glyoxylate with
moderate rate acceleration.6 In this article, we describe a
conceptually novel approach to both enantioenriched dihydro-
† University of Oxford.
‡ GlaxoSmithKline.
(1) (a) Danishefsky, S. J. Chemtracts: Org. Chem. 1989, 273. (b) Danishefsky,
S. J. Aldrichimica Acta 1986, 19, 59. (c) Boger, D. L. In ComprehensiVe
Organic Synthesis; Trost B. M., Ed.; Pergamon Press: Oxford, 1991; Vol.
5, p 451. (d) Waldmann, H. Synthesis 1994, 535. (e) Jorgensen, K. A.
Angew. Chem., Int. Ed. 2000, 39, 3558. (f) Jørgensen, K. A. Eur. J. Org.
Chem. 2004, 2093. For recent reports of catalytic asymmetric HDA reactions
of aldehydes involving Danishefsky’s diene and its analogues, see: (g)
Schaus, S. E.; Brånalt, J.; Jacobsen, E. N. J. Org. Chem. 1998, 63, 403.
(h) Simonsen, K. B.; Svenstrup, N.; Roberson, M.; Jørgensen, K. A. Chem.-
Eur. J. 2000, 6, 123. (i) Yamada, T.; Kezuka, S.; Mita, T.; Ikeno, T.
Heterocycles 2000, 52, 104. (j) Leveque, L.; Le Blanc, M.; Pastor, R.
Tetrahedron Lett. 2000, 41, 5043. (k) Kezuka, S.; Mita, T.; Ohtsuki, N.;
Ikeno, T.; Yamada, T. Chem. Lett. 2000, 824. (l) Mihara, J.; Aikawa, K.;
Uchida, T.; Irie, R.; Katsuki, T. Heterocycles 2001, 54, 395. (m) Aikawa,
K.; Irie, R.; Katsuki, T. Tetrahedron 2001, 57, 845. (n) Doyle, M. P.;
Phillips, I. M.; Hu, W. J. Am. Chem. Soc. 2001, 123, 5366. (o) Thompson,
C. F.; Jamison, T. F.; Jacobsen, E. N. J. Am. Chem. Soc. 2001, 123, 9974.
(p) Liu, P.; Jacobsen, E. N. J. Am. Chem. Soc. 2001, 123, 10772. (q)
Kezuka, S.; Mita, T.; Ohtsuki, N.; Ikeno, T.; Yamada, T. Bull. Chem. Soc.
Jpn. 2001, 74, 1333. (r) Long, J.; Hu, J.; Shen, X.; Ji, B.; Ding, K. J. Am.
Chem. Soc. 2002, 124, 10. (s) Wang, B.; Feng, X.; Huang, Y.; Liu, H.;
Cui, X.; Jiang, Y. J. Org. Chem. 2002, 67, 2175. (t) Kii, S.; Hashimoto,
T.; Maruoka, K. Synlett 2002, 931. (u) Du, H.; Long, J.; Hu, J.; Li, X.;
Ding, K. Org. Lett. 2002, 4, 4349. (v) Yamashita, Y.; Saito, S.; Ishitani,
H.; Kobayashi, S. J. Am. Chem. Soc. 2003, 125, 3793.
(2) Dossetter, A. G.; Jamison, T. F.; Jacobsen, E. N. Angew. Chem., Int. Ed.
1999, 38, 2398.
9
10.1021/ja043925d CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 1481-1486
1481

A R T I C L E S
Baker-Glenn et al.
Scheme 1. Aldolase Antibody Route to Hetero Diels-Alder
pyranones and tetrahydropyranones based on the use of aldolase
I antibody 84G3 or antibody 93F3. These antibodies were raised
by reactive immunization against a hapten featuring both a
sulfone and a 1,3-diketone functionality.7 The key step involves
the kinetic resolution of structurally diverse enone-derived aldol
products with aldolase antibody 84G3 or 93F3, followed by the
cyclization of these various enantioenriched aldol products into
one or the other class of hetero Diels-Alder adducts. This
approach allows easy access to numerous adducts, including
products otherwise derived from the combination of less
nucleophilic dienes with unactivated aldehydes. We present
herein full details of our studies, including an investigation of
the scope and limitation of the aldolase I antibody route to
various enantioenriched aldol products derived from enones and
a discussion of how the ring-closure process controls the degree
of oxidation of the cyclized product, leading either to enantio-
enriched tetrahydropyranones or dihydropyranones.
Adducts
Specifically, we hypothesized that aldolase I antibodies are
potential catalysts for the preparation of HDA adducts of
carbonyl compounds, according to a stepwise pathway where
an antibody-mediated aldol kinetic resolution will deliver
enantioenriched aldol products and a ring closure on the
â-carbon of the hydroxyenone would complete the cyclo-
condensation. This hypothesis is based on the mechanistic
similarities between the first step of a stepwise hetero Diels-
Alder reaction of carbonyl compounds and an aldolization
process. In this strategy, the presence of a leaving group such
as the methoxy group on the sp2-hydridized carbon positioned
â to the carbonyl in the aldol products will lead upon cyclization
to the corresponding dihydropyranones. In the absence of this
group, the enantioenriched tetrahydropyranones will be obtained
instead (Scheme 1).
Results and Discussion
Two mechanistic pathways can operate for Lewis acid-
mediated hetero Diels-Alder reactions of Danishefsky-type
carbodienes with carbonyl compounds.8 These two pathways
are a concerted nonsynchronous mechanism by analogy with
the catalyzed all-carbon Diels-Alder reaction or the formation
of a Mukaiyama-aldol intermediate, followed by an intra-
molecular Michael-type addition. To a great extent, the mech-
anism that operates is a function of the catalyst and solvent
system along with the structural features of the reactants.
Preparation of Enantioenriched â-Hydroxyenones. To test
this hypothesis, we investigated the capacity of aldolase I
antibodies 84G3 and 93F3 to deliver enantioenriched aldol
products derived from R,â-unsaturated methyl ketones and
various aldehydes (Table 1). We first undertook preliminary
work on selected synthetic forward aldol reactions in the
presence of these aldolase I antibodies. These experiments
revealed that only traces or no aldol products were formed
following extended incubation with the antibody.9 We therefore
turned our attention to the antibody-catalyzed kinetic resolution
of aldol product (()-1a, and we found that the treatment of
aldol (()-1a with aldolase antibody 84G3 or antibody 93F3
resulted in the rapid formation of anisaldehyde. Gratifyingly,
the retro-aldolization did not go beyond 50% conversion,
auguring an efficient kinetic resolution process (entry 1). Given
our encouraging results in the resolution of aldol (()-1a, we
set out to study the range of â-hydroxyenones that could be
resolved using antibodies 84G3 and 93F3. We focused our study
on aldol products derived from the combination of seven
representative enones and six aldehydes. In total, 12 racemic
aldols were prepared in one step from the corresponding donor
enones and acceptor aldehydes. Indeed, the direct aldol reaction
between various aldehydes and lithium enolates was successful
for the preparation of all racemic aldols (()-1a-l with excellent
chemical yields.10 The racemic aldols (()-1a-l were then
subjected to kinetic resolution with antibody 84G3 and/or 93F3
in PBS at pH 7.4 and at room temperature. Analysis by high-
performance liquid chromatography (HPLC) indicated that a
reaction is taking place with most aldol products. Conversion
(3) (a) Laschat, S. Angew. Chem., Int. Ed. Engl. 1996, 35, 289. (b) Pohnert,
G. ChemBioChem 2001, 2, 873. (c) Tremblay, M. R.; Dickerson, T. J.;
Janda, K. D. AdV. Synth. Catal. 2001, 343, 577. (d) Ose, T.; Watanabe,
K.; Mie, T.; Honma, M.; Watanabe, H.; Yao, M.; Oikawa, H.; Tanaka, I.
Nature 2003, 422, 185. (e) Stocking, E. M.; Williams, R. M. Angew. Chem.,
Int. Ed. 2003, 42, 3078.
(4) (a) Hilvert, D.; Hill, K. W.; Nared, K. D.; Auditor, M. T. M. J. Am. Chem.
Soc. 1989, 111, 9261. (b) Xu, J. A.; Deng, Q. L.; Chen, J. G.; Houk, K.
N.; Bartek, J.; Hilvert, D.; Wilson, I. A. Science 1999, 286, 2345. (c) Chen,
J.; Deng, Q.; Wang, R.; Houk, K. N.; Hilvert, D. ChemBioChem 2000, 1,
255. (d) Heine, A.; Stura, E. A.; Yli-Kauhaluoma, J. T.; Gao, C. S.; Deng,
Q. L.; Beno, B. R.; Houk, K. N.; Janda, K. D.; Wilson, I. A. Science 1998,
279, 1934. (e) Gouverneur, V. E.; Houk, K. N.; de Pascual-Teresa, B.;
Beno, B.; Janda, K. D.; Lerner, R. A. Science 1993, 262, 204. (f) Braisted,
A. C.; Schultz, P. G. J. Am. Chem. Soc. 1990, 112, 7430. (g) Romesberg,
F. E.; Spiller, B.; Schultz, P. G. Science 1998, 279, 1929.
(5) (a) Tarasow, T. M.; Eaton, B. E. Cell. Mol. Life Sci. 1999, 55, 1463. (b)
Tarasow, T. M.; Tarasow, S. L.; Tu, C.; Kellogg, E.; Eaton, B. E. J. Am.
Chem. Soc. 1999, 121, 3614. (c) Tarasow, T. M.; Eaton, B. E.; Biopolymers
1998, 48, 29.
(6) Hu, Y. J.; Ji, Y. Y.; Wu, Y. L.; Yang, B. H.; Yeh, M. Bioorg. Med. Chem.
Lett. 1997, 7, 1601.
(7) (a) Wagner, J.; Lerner, R. A.; Barbas, C. F., III. Science 1995, 270, 1797.
(b) Bjornestedt, R.; Zhong, G.; Lerner, R. A.; Barbas, C. F., III. J. Am.
Chem. Soc. 1996, 118, 11720. (c) Zhong, G.; Hoffmann, T.; Lerner, R. A.;
Danishefsky, S.; Barbas, C. F., III. J. Am. Chem. Soc. 1997, 119, 8131. (d)
Barbas, C. F., III; Heine, A.; Zhong, G.; Hoffman, T.; Gramatikova, S.;
Bjornestedt, R.; List, B.; Anderson, J.; Stura, E. A.; Wilson, I. A.; Lerner,
R. A. Science 1997, 278, 2085. (e) Hoffman, T.; Zhong, G.; List, B.; Shabat,
D.; Anderson, J.; Gramatikova, S.; Lerner, R. A.; Barbas, C. F., III. J. Am.
Chem. Soc. 1998, 120, 2768. (f) List, B.; Shabat, D.; Zhong, D.; Turner, J.
M.; Li, A.; Bui, T.; Anderson, J.; Lerner, R. A.; Barbas, C. F., III. J. Am.
Chem. Soc. 1999, 121, 7283. (g) Zhong, D.; Lerner, R. A.; Barbas, C. F.,
III. Angew. Chem., Int. Ed. 1999, 111, 3957. (h) Sinha, S. C.; Sun, J.;
Miller, G. P.; Wartmann, M.; Lerner, R. A. Chem.-Eur. J. 2001, 7, 1691.
(i) Sinha, S. C.; Sun, J.; Miller, G. P.; Barbas, C. F., III; Lerner, R. A.
Org. Lett. 1999, 10, 1623. (j) Zhong, D.; Shabat, D.; List, B.; Anderson,
J.; Sinha, S. C.; Lerner, R. A.; Barbas, C. F., III. Angew. Chem., Int. Ed.
1998, 37, 2481. (k) Maggiotti, V.; Resmini, M.; Gouverneur, V. Angew.
Chem., Int. Ed. 2002, 41, 1012. (l) Baker-Glenn, C.; Ancliff, R.;
Gouverneur, V. Tetrahedron 2004, 60, 7607.
(8) (a) Danishefsky, S. J.; Larson, E.; Askin, D.; Kato, N. J. Am. Chem. Soc.
1985, 107, 1246. (b) Roberson, M.; Jepsen, A. S.; Jorgensen, K. A.
Tetrahedron 2001, 57, 907. (c) Corey, E. J.; Cywin, C. L.; Roper, T. D.
Tetrahedron Lett. 1992, 33, 6907. (d) Schaus, S. E.; Brånalt, J.; Jacobsen,
E. N. J. Org. Chem. 1998, 63, 403. (e) Mujica, M. T.; Afonso, M. M.;
Galindo, A.; Palenzuela, J. A. Tetrahedron 1996, 52, 2167. (f) Ghosh, A.
K.; Mathivanan, P.; Cappiello, J.; Krishnan, K. Tetrahedron: Asymmetry
1996, 7, 2165. (g) Qian, C.; Wang, L. Tetrahedron Lett. 2000, 41, 2203.
(9) A mixture of para-nitrobenzaldehyde and a large excess (700 equiv) of
MVK, penten-3-one, or 4-methylbutanone was incubated in the presence
of 10 mol % of ab38C2 or ab84G3. No trace of the desired aldol product
was observed by HPLC upon extended reaction time when MVK or penten-
3-one were used as donors. Only traces of aldol product could be detected
for the reaction of para-nitrobenzaldehyde with 4-methylbutanone in the
presence of ab84G3.
(10) See Supporting Information for full experimental details.
9
1482 J. AM. CHEM. SOC. VOL. 127, NO. 5, 2005

Bioorganic Route to Tetrahydro- and Dihydropyranones
A R T I C L E S
Table 1. â-Hydroxyenones Obtained via Kinetic Resolution with 84G3 or 93F3
a 10 mol % antibody, room temperature, PBS/MeCN 9/1, pH 7.4. b 4 mol % antibody, room temperature, pH 7.4, all ee measured at about 50% conversion.
c 20 mol % antibody for this experiment. d 1 mol % ab 84G3, PBS/toluene, room temperature. e ee measured at 60% conversion. f ee measured at 75%
conversion.
slowed considerably at approximately 50% for most aldol
products, suggesting that these reactions were highly enantio-
selective (entries 2-10), but went far beyond 50% for aldol
products 1k and 1l (entries 11 and 12). It should be noted that,
upon incubation of these two substrates 1k and 1l with
antibodies 84G3 or 93F3, no aldehyde that should have resulted
from a retro-aldol reaction or other products such as the target
cyclized adducts were observed in the antibody crude mixture.
The data also suggested that the antibodies do not accept aldol
products derived from R,â-unsaturated ketones possessing long
chain substituents â to the carbonyl group (entry 4). The
presence of an ethyl or a propyl group in place of a methyl
group on that position slowed the antibody-catalyzed retro-
aldolization considerably (entries 1-3), and no reaction was
observed with the aldol (()-1d (entry 4). The aldol products
derived from 4-methylbutenone were all accepted by antibodies
84G3 and 93F3 (entries 5-8). These experiments revealed that
84G3 and 93F3 tolerated more structural variations for the part
of the molecule originally derived from the acceptor, although
faster conversions were observed for aldols derived from the
less electrophilic aldehydes as expected for a retro-aldolization
process.
To ensure that catalysis proceeded as anticipated and to probe
the synthetic scope of these reactions, the unconverted aldol
products were recovered at approximately 50% conversion and
studied using chiral phase HPLC. Comparison of the HPLC trace
of the recovered aldol products with those of racemic standards
indicated that the catalyst was highly enantioselective for most
substrates, allowing the recovery of the (R)- or (S)-aldols with
ee values ranging from 40 to 99%. For substrate 1b recovered
with an enantiomeric excess of only 40%, a control experiment
revealed that this apparent lack of efficiency was not due to
the inability of the antibody to discriminate between the two
enantiomers but was the result of partial racemization of the
substrate upon the extended reaction time required to reach 50%
conversion. For substrates not prone to racemize under the
9
J. AM. CHEM. SOC. VOL. 127, NO. 5, 2005 1483

A R T I C L E S
Baker-Glenn et al.
Scheme 2. Cross-Metathesis of â-Hydroxyenone 1a with Dodecene
Table 2. Kinetic Parameters for Selected Retro-Aldol Reactions
reaction conditions, the lower ee values were obtained for aldol
products derived from nonconjugated aldehydes. In contrast,
excellent levels of enantiocontrol were observed for aldol
products derived from various conjugated aldehydes. We also
found that the aldol products 1k and 1l recovered at 60 and
75% conversion, respectively, were racemic mixtures, suggesting
that these antibody-mediated reactions were nonspecific. For
these two substrates, we suggest that a Michael addition
involving a reactive nucleophilic residue of the antibody
competes with the desired retro-aldolization process. Indeed,
observation that the aldol compound 1l possessing the more
exposed Michael-type acceptor could not be kinetically resolved
supports this argument. For compound 1k, we hypothesize that
a Michael-type addition process followed by elimination of
methanol also led to a covalently modified antibody, suggesting
that this substrate acted as an irreversible inhibitor for antibody
84G3 and 93F3. Control experiments supported this hypothesis
as the presence of aldol compounds 1k or 1l inhibited the
antibody-catalyzed retro-aldolization of compound (()-1a.
To assign the absolute configurations of the recovered aldol
products, the enantiopure products 1f, 1h, and 1j were synthe-
sized by independent chemical asymmetric syntheses based on
the use of Evans-type chiral oxazolidines.10 Compound 1g was
prepared enantiopure from the corresponding hydroxyester
obtained by Baker’s yeast reduction of the ketoester.10 The
absolute configurations of all other compounds were assigned
by analogy.
1
substrates
Ab
k
cat (min-
)
KM (µM)
(()-1a
(()-1a
(()-1b
(()-1g
(()-1h
(()-1h
(()-1j
(()-1j
84G3
93F3
84G3
84G3
84G3
93F3
84G3
93F3
0.45
0.40
0.03
0.006
2.7
4.1
2.2
3.2
46
51
116
352
59
66
40
37
therefore occurred upon metathesis. The formation of compound
1d also suggests that capping the double bond of 1a with a
methyl group does not prevent its reaction with dodecene. This
sequential retro-aldolization/cross-metathesis process is ex-
tremely versatile and allows numerous structural variations,
allowing the preparation of various enantioenriched aldol
products that could be regarded as valuable precursors for the
synthesis of many natural product targets.
Kinetic Data. To gain more detailed information on the
antibody-catalyzed transformations, kinetic studies were per-
formed at room temperature on selected substrates (Table 2).
For these reactions, the background retro-aldolization reaction
was undetectable even after an extended period of time.
Determination of kcat/kuncat was therefore not possible. The
antibody-catalyzed transformations leading to enantioenriched
aldol compounds follow Michaelis-Menten kinetics with some
of these reactions rapidly catalyzed by the antibody as reflected
with kcat values up to 4.1 min-1. As shown in Table 2, the
efficiency with which the aldol compounds 1h and 1j derived
from 4-methoxycinnamaldehyde are processed by antibodies
84G3 and 93F3 is remarkable, a highly desirable feature since
both the allylic alcohol and the enone may be used for further
functional manipulation. The antibody-catalyzed retro-aldol
reactions were inhibited by the addition of an equimolar amount
of 2,4-pentadione as expected for a covalent catalytic mechanism
involving the reactive amine programmed in ab84G3 and
ab93F3. The rate constants suggest that the best rate accelera-
tions are obtained for aldol products derived from the less
reactive aldehydes. These kinetic data suggest that this bio-
organic approach will be the most efficient for the preparation
of adducts derived from unactivated aldehydes that are more
difficult to obtain using metal-catalyzed hetero Diels-Alder
reactions.
Cyclization into Tetrahydropyranones. The enantioenriched
aldol products 1f-h obtained by kinetic resolution in the
presence of antibody 84G3 or 93F3 were subsequently converted
into the corresponding tetrahydropyranones (Table 3). A major
concern for the cyclization step is the potential sensitivity of
the aldol products 1f-h to elimination or epimerization.
Common reagents such as TFA or Amberlyst-15 failed to afford
the cyclized compound or led to full or partial racemization.
Direct cyclization of the recovered aldol products (R)-1f and
(R)-1g was accomplished successfully in the presence of
TMSOTf and iPr2NEt, affording the desired HDA adducts (R)-
The data summarized in Table 1 demonstrated that the
aldolase type I antibodies 84G3 and 93F3 were able to
kinetically resolve various â-hydroxyenones but cannot mediate
their cyclization.
Expanding the Repertoire of Enantioenriched â-Hydroxy-
enones. Because racemic aldol products derived from long chain
enones are not accepted by antibodies 84G3 or 93F3, the need
to overcome this limitation became apparent and led us to
examine different possibilities to find a general solution to this
problem. We discovered that subjecting the aldol product
(S)-1a to a cross-metathesis reaction11 was the most efficient
way to elongate the alkenyl chain in a single-step process
without epimerization or other undesirable side reactions
(Scheme 2). We undertook the resolution of aldol product
(()-1a, and this procedure allowed us to prepare a sample of
enantioenriched (S)-1a with an ee of 91% as verified by chiral
HPLC.10 This compound (S)-1a was used as a starting material
for the preparation of aldol product 1d that could not be resolved
using antibody 84G3 or 93F3. Upon cross-metathesis with 3
equiv of dodecene in the presence of 10 mol % of the
Ru-catalyst 3, the desired aldol product (S)-1d was obtained
with 85% yield and 90% ee. No detectable epimerization
(11) For recent reviews on cross-metathesis, see: (a) Connon, S. J.; Blechert,
S. Angew. Chem., Int. Ed. 2003, 42, 1900. (b) Chatterjee, A. K.; Morgan,
J. P.; Scholl, M.; Grubbs, R. H. J. Am. Chem. Soc. 2000, 122, 3783. (c)
Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J. Am. Chem.
Soc. 2003, 125, 11360.
9
1484 J. AM. CHEM. SOC. VOL. 127, NO. 5, 2005

Bioorganic Route to Tetrahydro- and Dihydropyranones
A R T I C L E S
Table 3. Cyclization of Enantioenriched Aldol Products into Tetrahydropyranones
aldol
ee (%)
conditions
product
ee (%)
(R)-1f R1 ) CH2Ph(p-OMe)
(R)-1g R1 ) CH2CH2Ph
(S)-1h R1 ) CHdCHPh(p-OMe)
83a
72
>99
a
a
b
(R)-2f R1 ) CH2Ph(p-OMe)
(R)-2g R1 ) CH2CH2Ph
(R)-2h R1 ) CH2CH2Ph(p-OMe)
83
72
>99
Table 4. Ring Closure of (()-1a in Different Solvent Systems
entry
solvent
reoxidant (mol %)
additive
T/
°
C
t/h
Pd cat (mol %)
conv. %a (yield %)
1
2
3
4
5
6
7
DME
PBS
CuCl (10)
CuCl (10)
CuCl (10)
CuCl (10)
CuCl (18)
CuCl (20)
CuCl (20)
Na2HPO4
50
50
50
50
50
50
50
20
21.5
17
21.5
18.5
72
PdCl2 (10)
PdCl2 (10)
PdCl2 (10)
PdCl2 (10)
PdCl2 (10)
PdCl2 (10)
PdCl2 (10)
100(50)
50
50
100
65
-
-
-
-
-
-
DMF/PBS (1/1)
DMF/PBS (1/9)
toluene/PBS (1/1)
toluene/PBS (1/1)
toluene/H2O (1/1)
100(68)
83
21.5
1
a Consumption of starting material by H NMR; 10 mol % only. PBS ) phosphate-buffered saline pH 7.4.
2f and (R)-2g with enantiomeric excesses of 72 and 83%,
respectively, suggesting that no racemization had occurred under
these conditions. However, the direct cyclization was not feasible
for substrate (S)-1h because a competitive elimination occurred
instead. For this substrate, a two-step sequence involving
chemoselective hydrogenation of the allylic alcohol, followed
by cyclization, afforded the hetero Diels-Alder adduct (R)-2h
with an enantiomeric excess greater than 99%.
racemization. To avoid, if at all possible, the isolation of the
antibody-resolved aldol products before the cyclization, we
adapted our original procedure using (()-1a as model compound
(Table 4). We discovered that the oxidative cyclization of aldol
(()-1a proceeded using PBS as the reaction solvent, albeit more
slowly, auguring the possibility to establish a one-pot antibody/
PdCl2 process to access the desired enantioenriched dihydro-
pyranones directly from the racemic aldol precursors (entry 2).
We therefore investigated the compatibility of this reaction with
various solvent systems, including solvent mixtures that could
possibly tolerate the simultaneous presence of the two catalysts.
This study revealed that the reaction is best performed in a
mixture of DMF/PBS (1/9) (entry 4) or a biphasic mixture of
toluene/PBS (1/1) (entry 6). Under biphasic conditions, the
reaction is slower and requires up to 20 mol % CuCl and 20
mol % PdCl2 as well as extended reaction time to reach
completion. We were pleased to find that under these modified
conditions, no racemization was observed if the starting aldol
product is enantioenriched. For the biphasic solvent system, the
use of PBS in place of water is advantageous as reflected by
the higher conversion into the desired adduct (entries 6 and 7).
As an attempt to implement a one-pot retroaldolization/
cyclization process, we studied the compatibility of the antibody
with palladium dichloride. As matters transpired, it became
apparent that although the antibody tolerated well the presence
Cyclization into Dihydropyranones. The major issue cloud-
ing our initial strategy to access dihydropyranones in addition
to tetrahydropyranones was the inability of antibodies 84G3 and
93F3 to kinetically resolve aldol products derived from 4-meth-
oxybutenone. This limitation suggested the need to develop an
alternative procedure for the ring-closure process if one wants
to broaden the scope of this antibody route for the preparation
of enantioenriched dihydropyranones. To restore the oxidation
state of the sp2-hybridized carbon positioned â to the ketone
group, we reasoned that an oxidative ring closure can be used
as an alternative to the intramolecular Michael-type addition
reaction described above. We have recently reported that a
palladium(II)-mediated oxidative cyclization was effective for
the preparation of structurally diverse 2,3-dihydro-4H-pyran-
4-ones from the corresponding â-hydroxyenones.12 The best
results were obtained using palladium(II) chloride/copper(I)
chloride/oxygen in DME, which are the reaction conditions
traditionally applied for a Wacker-type oxidation. Under these
conditions, a series of â-hydroxyenones were converted into
the corresponding dihydropyranones with good yields. The use
of enantiopure aldol precursors also allowed the preparation of
the corresponding dihydropyranones without any detectable
of PdCl2
for the kinetic resolution process, the palladium-
mediated oxidative cyclization did not proceed in the presence
of the protein.13 Inspired by the antibody-route to multigram
(13) For another example of a palladium(II)-catalyzed Wacker reaction inhibited
by an aldolase I antibody, see page 6 of the Supporting Information of:
Finn, M. G.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 1998,
120, 2963.
(12) Reiter, M.; Ropp, S.; Gouverneur, V. Org. Lett. 2004, 6, 91.
9
J. AM. CHEM. SOC. VOL. 127, NO. 5, 2005 1485

A R T I C L E S
Baker-Glenn et al.
Scheme 3. Direct Conversion of (()-1a into Enantiopure
Dihydropyranone (S)-2a
compounds. This strategy represents the first aldolase I antibody
route to enantioenriched HDA adducts of carbonyl compounds
via a stepwise pathway with an aldol product as the key
intermediate. The substrate limitation resulting from the inability
of antibodies 84G3 and 93F3 to kinetically resolve racemic aldol
products derived from long chain enones was solved using a
cross-metathesis reaction with the appropriate olefinic partner.
A remarkable feature of this approach is that it is particularly
efficient for the preparation of adducts otherwise derived from
less activated dienes and dienophiles for which only limited
successes were obtained using the more traditional metal-
catalyzed hetero Diels-Alder reaction. Another key feature of
this antibody approach is the possibility to access both tetra-
hydropyranones and a representative dihydropyranone. This was
achieved by modifying the reaction conditions for the ring-
closure process. Applications of this route to hetero Diels-Alder
adducts for the synthesis of natural products are now in progress.
scale preparation of other enantioenriched aldol products,14 we
finally opted for a biphasic PBS/toluene or PBS/chlorobenzene
system using the following procedure. In a typical reaction, a
solution of the antibody in PBS was added to a solution of the
racemic aldol in either toluene or chlorobenzene. The mixture
was shaken, and when the desired ee was reached as monitored
by HPLC, the reaction mixture was cooled, allowing separation
of the organic layer containing the enantioenriched aldol product
from the frozen aqueous antibody solution. To the organic layer
was then added PBS, PdCl2, and CuCl, and the reaction was
heated at 50 °C and left stirring until complete conversion of
the aldol product into the desired dihydropyranone. The cyclized
product was easily recovered from the organic phase, and the
ee was measured by HPLC. In this process, the antibody solution
could be thawed for reuse. The procedure was successfully
applied for the preparation of enantiopure adduct (S)-2a from
the corresponding racemic aldol precursor. The best result was
obtained by increasing the amount of both CuCl and PdCl2
(up to 60% mol) relative to the substrate, allowing quantita-
tive conversion of the enantioenriched aldol product into the
desired dihydropyranone (S)-2a with no detectable epimerization
(Scheme 3).
Acknowledgment. This study was supported in part by the
EPSRC (GR/S68095/01), the Royal Society (Research Grant
RSRG 23219B) and the British Council (Chevening scholarship,
M.R.). N.H. and S.R. thank the Leverhulme Trust for a Research
Fellowship (F/08 341/B), and C.B.G. is grateful to GlaxoSmith-
Kline for a full case studentship. We thank Professor R. A.
Lerner for providing generous quantities of ab84G3 and 93F3,
Dr. P. Procopiou (GSK) for helpful discussions and Dr A. R.
Cowley for the X-ray data.
Supporting Information Available: Full Experimental Sec-
tion (PDF) including procedures, spectroscopic data for all new
compounds, full description of independent asymmetric syn-
theses for absolute configuration assignment including X-ray
data, antibody screening, procedure for the determination of
enantiomeric excesses, and detailed antibody kinetics. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Conclusions
In summary, we have developed a bioorganic approach for
the preparation of hetero Diels-Alder adducts of carbonyl
(14) (a) Sinha, S. C.; Barbas, C. F., III; Lerner, R. A. Proc. Natl. Acad. Sci.
U.S.A. 1998, 14603. (b) Sinha, S. C.; Sun, J.; Miller, G. P.; Wartmann,
M.; Lerner, R. A. Chem.-Eur. J. 2001, 7, 1692. (c) Turner, J. M.; Bui, T.;
Lerner, R. A.; Barbas, C. F., III; List, B. Chem.-Eur. J. 2000, 6, 2772.
JA043925D
9
1486 J. AM. CHEM. SOC. VOL. 127, NO. 5, 2005