A simple, highly regioselective, one-pot stereoselective synthesis of tertiary α-hydroxyesters: a tandem oxidation/benzilic ester rearrangement

Article abstract of DOI:10.1016/j.tetlet.2006.06.107

This letter describes a simple, highly regioselective, stereoselective one-pot tandem oxidation/benzilic ester rearrangement protocol for the conversion of α-hydroxyketones to tertiary α-hydroxyesters.

Full text of DOI:10.1016/j.tetlet.2006.06.107

Tetrahedron Letters 47 (2006) 6049–6052
A simple, highly regioselective, one-pot stereoselective synthesis
of tertiary a-hydroxyesters: a tandem oxidation/benzilic
ester rearrangement
*
´
Carolina Sılva Marques, Nuno Moura and Anthony J. Burke
´
Departamento de Quı´mica and Centro de Quı´mica de Evora, Universidade de Evora, Rua Roma˜o Romalho 59, 7000 Evora, Portugal
´
´
Received 15 March 2006; revised 19 June 2006; accepted 21 June 2006
Available online 10 July 2006
Abstract—This letter describes a simple, highly regioselective, stereoselective one-pot tandem oxidation/benzilic ester rearrangement
protocol for the conversion of a-hydroxyketones to tertiary a-hydroxyesters.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Several years ago, Casiraghi and co-workers used a
Lederer–Manasse-type phenoxide addition to a range
of chiral pyruvate esters to give o-hydroxyatrolactic
esters.10 Dihydroxylation of a,b-unsaturated esters has
also been used.11
Tertiary a-hydroxy carboxylic acids and esters are very
important structural units present in a plethora of natural
products and medicinal compounds, or in the key inter-
mediates leading to these compounds. For example, the
a-hydroxy carboxylic acid unit is present in complex nat-
ural products like integerrimine1 and monocrotaline.2
The a-hydroxyester unit is present in oxybutynin, a widely
used muscarinic receptor antagonist for the treatment of
urinary incontinence.3 Topotecan and irnotecan are two
compounds which contain the tertiary a-hydroxyester
unit, which are the active ingredients for the treatment
of ovarian and colorectal cancer, respectively.4 It is also
present in the glucocorticoid receptor binders reported
recently by a team at GlaxoSmithKline.5
In this letter, we report a simple one-pot tandem
oxidation/benzilic ester rearrangement protocol for
the synthesis of tertiary a-hydroxyesters from
a-hydroxyketones.
2. Discussion
Some time ago, we were interested in using a-diketone 2
(Scheme 1) as a model substrate in order to probe the
Despite the importance of this functional group, there
are quite a limited number of methods available for
the direct stereoselective synthesis of the tertiary a-
hydroxy carboxylic acid or ester unit. Most methods
rely on the addition of organometallic reagents to acti-
vated ketones to achieve this objective, like the methods
reported by Loupy and Monteux,6 Fennie et al.7 and the
group of Senanayake3 where Grignard reagents were
added to a-ketoester substrates. Wang et al.8 obtained
a-hydroxy carboxylic acids by the addition of allyltri-
chlorosilane to a-oxocarboxylic acids. The homo-aldol
reaction was also recently exploited for this purpose.9
O
OMe
Ph
O
OMe
Ph
[O]
Ph
Ph
OH
O
1
2
Cu(OAc)2, MeOH, rt
70%(2 Steps)
HO Ph
HO Ph
Ph
Ph
+
MeO2C
MeO2C
MeO
H
OMe
H
syn-3b
anti-3a
*
Corresponding author. Tel.: +351 266745310; fax: +351
266744971; e-mail: ajb@dquim.uevora.pt
Scheme 1.
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.06.107

6050
C. S. Marques et al. / Tetrahedron Letters 47 (2006) 6049–6052
mechanism of a certain benzilic acid rearrangement.12,13
Our strategy to synthesise 2 was to treat a-hydroxy-
ketone 1 (obtained via acidic methanolysis of chalcone
epoxide12 as a mixture of anti- and syn-isomers12) with
a suitable oxidation agent. Many oxidation agents were
used without effect, but when Cu(OAc)2 was used in
methanol instead of isolating a-diketone 2, we obtained
the tertiary a-hydroxyesters anti-3a and syn-3b in a ratio
of 1.8:114 and with a combined yield of 70%.
OMe O
Me
Ph
OH
Me
17
To access b-chloro-a-hydroxyketones 10 and 14 we used
the procedure of Xu et al.20 which employed TMSCl as
the source of chloride nucleophile.
We believe that a-diketone 2 was indeed generated but
readily suffered a benzylic ester rearrangement15 due to
the reaction of diketone 2 with methoxide ion.16 The
rearrangement was completely regioselective. To con-
firm that indeed an a-diketone was generated in this
reaction, we conducted the reaction in the same way
as before, but added 1,2-diaminobenzene to trap the
intermediate a-diketone 2. This method had been used
previously by other workers to establish the generation
of a-diketones from a-hydroxybenzofuranones in solu-
tion.17 As expected, quinoxaline 4 was the only product
obtained (Scheme 2), thus confirming the generation of
a-diketone intermediates in this reaction.
a-Hydroxyketones 8, 9, 11, 12 and 16 were evaluated in
a series of room temperature tandem oxidation/benzilic
ester rearrangements (Table 2).
Although, the type of substrate we could access limited
this preliminary study (which we hope to extend in the
near future), we nevertheless obtained some important
information relating to this reaction procedure. It was
found that in all cases the in situ benzilic ester rearrange-
ment was 100% regioselective, with the nucleophile
selectively attacking carbonyl-2. The yields were gener-
ally satisfactory for the two steps. Some reasonable pre-
liminary diastereoselectivities were obtained with a
maximum of just under 3:1, and it implied that the size
of the 1-phenyl ring and the substituent in the 3-position
influence the diastereoselectivity.
Gratified by these results, we decided to conduct a pre-
liminary study to get an idea of the scope of this reaction
and to access the influence of both electronic, stereo-
chemical factors, and the solvent on the reaction effi-
ciency, the regioselectivity and the diastereoselectivity.
The following family of a-hydroxyketones 8–16 were
prepared by selective ring cleavage of epoxy ketones
5–7, which were obtained from the corresponding
chalcones via the Weitz–Scheffer epoxidation reaction18
(Table 1).
In the case of b-chloro ketones 10 and 14 no benzilic ester
rearrangement occurred. It appears that the presence of a
chlorine in the 3-position inhibited the reaction, as only
the starting material was obtained. This could be, for
the following reasons: (1) the electron withdrawing chlo-
rine prevents 1,2-migration of the CH(Cl)Ph group or (2)
the intermediate a-diketone exists predominantly in its
enol form (Scheme 3) thus preventing the benzilic ester
rearrangement from occurring. b-Acetoxyketone 15
failed to give any tertiary a-hydroxyester product, giving
unidentified products instead, perhaps from more
favourable competing side reactions. These results
appear to indicate that substrates containing electron
withdrawing groups in the migrating unit have little
propensity to undergo the benzilic ester rearrangement.
As an additional study we looked at the effect of temper-
ature on the reaction. The tandem oxidation/benzilic
Roberts and co-workers19 recently reported an efficient
way of accessing a-hydroxy-b-methylketones 9 and 13
by methylating the corresponding a,b-unsaturated epox-
ides with the active methylating reagent formed by
reacting Me3Al with water. We applied this procedure
to obtain a-hydroxyketones 9 and 13. A secondary
product that could not be removed despite our best
1
efforts contaminated a-hydroxy-b-methylketone 13. H
NMR analysis appeared to identify the secondary com-
pound as a single diastereomer (it was not possible to
determine its relative configuration) of a-hydroxy-b-
methylketone 17, presumably obtained via electrophilic
methylation of the 1-phenyl ring of either the epoxide
starting compound or one of the a-hydroxy-b-methyl-
ketone diastereomeric products. For this reason, a-
hydroxy-b-methylketone 13 was not used for the tandem
oxidation/benzilic ester rearrangement.
ester rearrangement on a-hydroxyketone
1 using
CuOAc2 in methanol at reflux (1.8 h then 4 h at rt) gave
the mixture of anti-3a and syn-3b with a total isolated
yield of 66%. The diastereomeric ratio was not
determined.
We feel that this one-pot transformation of appropriate
a-hydroxyketones to tertiary a-hydroxyesters fits a
number of the requirements listed by Sharpless and
co-workers21 for consideration as a click chemistry pro-
cess. For instance, it is modular, atom economical
(there are no by-products), the reaction conditions
are simple, one reagent serves as the solvent and the
starting materials are readily available.22 We are cur-
rently investigating this reaction procedure to access
other important target compounds and to obtain an
asymmetric version.
OMe
O
OMe
Ph
Cu(OAc)2 (1 equiv), MeOH
N
N
Ph
Ph
NH2
Ph
OH
(1.25 equiv)
4 (34%)
1
NH2
r.t., 5h
Scheme 2.

C. S. Marques et al. / Tetrahedron Letters 47 (2006) 6049–6052
6051
Table 1.
O
X
O
X
O
Reagent(s)
O
+
Ar
Ph
Ar
Ph
Ar
Ph
OH
syn
OH
anti
5 Ar = Ph
6 Ar = 2-MeOC6H4
7 Ar = 2,4-(MeO)2C6H3
8 Ar = Ph, X = OEt
9 Ar = Ph, X = Me
10 Ar = Ph, X = Cl
11 Ar = 2-MeOC6H4, X = OMe
12 Ar = 2-MeOC6H4, X = OEt
13 Ar = 2-MeOC6H4, X = Me
14 Ar = 2-MeOC6H4, X = Cl
15 Ar = 2-MeOC6H4, X = OAc
16 Ar = 2,4-(MeO)2C6H3, X = OMe
Entry
Epoxyketone
Reagent(s)
Product
Reaction time (h)
Yielda (%)
anti:synb
1
2
3
4
5
6
7
8
9
5
5
5
6
6
6
6
6
7
EtOH/H2SO4
Me3Al/H2O
TMSCl
MeOH/H2SO4
EtOH/H2SO4
Me3Al/H2O
TMSCl
8
9
6
3
19
24
42
4
19
24
24
100
92
70
91
50
c
2.8:1
2:1
20:1
3.8:1
3.5:1
1.5:1
1:20
3.1:1
syn only
10
11
12
13
14
15
16
61
37
90
AcOH
MeOH/H2SO4
a For both diastereomers.
b Determined by 1H NMR spectroscopy, except for entry 8 where the diastereomeric ratio is based on isolated yields.
c Could not be calculated due to the presence of an inseparable side product.
Table 2.
HO Ar
O
X
HO Ar
Cu(OAc)2, R1OH, rt
Ph
Ph
R1O2C
R1O2C
+
Ar
Ph
X
H
H
X
OH
8, 9, 11,12,17
syn-18a-23a
anti-18b-23b
Entry
a-Hydroxyketone
R1
X
Ar
Product
Reaction time (h)
Yield (2 steps)a (%)
anti:synb,c
1
2
3
4
5
6
8
9
Me
Me
Me
Et
Me
Me
OEt
Me
OMe
OMe
OEt
Ph
Ph
18
19
20
21
22
23
72
45
44
48
48
94
37
42
35
63
49
49
2.8:1
2:1
2.9:1
2.6:1
2.1:1
2.9:1
11
11
12
17
2-MeOC2H4
2-MeOC2H4
2-MeOC2H4
2,4-(MeO)2C2H4
OMe
a For both diastereomers.
b Determined by 1H NMR spectroscopy.
c The assignment of the anti-isomer as the major isomer was made on the basis of the literature precedents.12,13
References and notes
O
Cl
O
Cl
1. White, J. M.; Ohira, S. J. Org. Chem. 1986, 51, 5492.
2. Vedejs, E.; Ahmad, S.; Larsen, S. D.; Westwood, S. J.
Org. Chem. 1987, 52, 3937.
Ar
Ph
Ar
Ph
OH
O
3. Senanayake, C. H.; Fang, K.; Grover, P.; Bakale, R. P.;
Vandenbossche, C. P.; Wald, S. A. Tetrahedron Lett. 1999,
40, 819.
Scheme 3.
4. Subrahmanyam, D.; Sarma, V. M.; Venkateswarlu, A.;
Sastry, T. V. R. S.; Kulakarni, A. P.; Rao, D. S.; Reddy,
K. V. S. R. K. Biorg. Med. Chem. 1999, 7, 2013.
5. Barker, M.; Clackers, M.; Demaine, D. A.; Humphreys,
D.; Johnston, M. J.; Jones, H. T.; Pacquet, F.; Pritchard,
J. M.; Salter, M.; Shanahan, S. E.; Skone, P. A.; Vinader,
V. M.; Uings, I.; McLay, I. M.; Macdonald, S. J. F. J.
Med. Chem. 2005, 48, 4507.
Acknowledgements
´
´
We wish to thank the Centro de Quımica de Evora for
financial support of this work. The personnel of the
NMR service at CACTI (University of Vigo, Spain)
are acknowledged for 1H NMR and 13C NMR analyses.
The personnel of the mass spectrometry unit and the
microanalysis service at CACTI are acknowledged for
mass spectrometric and elemental analyses.
6. Loupy, A.; Monteux, D. A. Tetrahedron 2002, 65, 1541.
7. Fennie, M. W.; DiMauro, E. F.; O’Brien, E. M.; Venka-
tachalam, A.; Kozlowski, M. C. Tetrahedron 2005, 61, 6249.

6052
C. S. Marques et al. / Tetrahedron Letters 47 (2006) 6049–6052
8. Wang, Z.; Xu, G.; Wang, D.; Pierce, M. E.; Confalone, P.
N. Tetrahedron Lett. 2000, 41, 4523.
15. March, J. Advanced Organic Chemistry, 4th ed.; Wiley
Interscience, 1992; p 1080.
9. Juhl, K.; Gathergood, N.; Jørgensen, K. A. Chem.
Commun. 2000, 2211.
16. This type of reaction has previously been reported for a
cortisone a-diketone derivative: Lewbart, M. L.; Mattox,
V. R. J. Org. Chem. 1963, 28, 1773.
17. Brady, B. A.; Geoghegan, M.; O’Sullivan, W. I. J. Chem.
Soc. Perkin Trans. 1 1989, 1557.
18. Weitz, E.; Scheffer, A. Ber. 1921, 54, 2327.
19. Carde, L.; Davies, H.; Geller, T. P.; Roberts, S. M.
Tetrahedron Lett. 1999, 41, 5421.
20. Xu, L.; Li, C.; Xia, C.; Zhao, P. Tetrahedron Lett. 2004,
45, 1.
10. Casiraghi, G.; Bigi, F.; Casnati, G.; Sartori, G.; Soncini, P.;
Fava, G. G.; Belicchi, M. F. J. Org. Chem. 1988, 53, 1779.
11. Nicolaou, K. C.; Yue, E. W.; La Greca, S.; Nadine, A.;
Yang, Z.; Leresche, J. E.; Tsuri, T.; Naniwa, Y.; De
Riccardis, F. Chem. Eur. J. 1995, 1, 467.
12. Burke, A. J. Ph.D. Thesis, National University of Ireland,
1993.
13. Burke, A. J.; Schmalle, H. W.; Brady, B. A.; O’Sullivan,
W. I. Acta Cryst. 2000, C56, 484.
21. Kolbe, H. C.; Finn, M. G.; Sharpless, K. B. Angew.
Chem., Int. Ed. 2001, 40, 2005.
22. In fact, the preceding ring-opening reaction is a classic
example of a click chemical process (Ref. 21).
14. The major isomer was established to be the anti-isomer
from a derivatisation study on a glycolic acid precursor of
known relative configuration (Ref. 12).