Aldol reaction of butane-2,3-diacetal protected methyl glycerate

Article abstract of DOI:10.1016/j.tetasy.2011.01.004

The addition of aldehydes to butane-2,3-diacetal protected glycerates has been investigated. The reaction was shown to be diastereoselective by ¹H NMR spectroscopy thus confirming the ability of the diacetal to influence the stereochemical outc

Full text of DOI:10.1016/j.tetasy.2011.01.004

Tetrahedron: Asymmetry 22 (2011) 149–152
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Aldol reaction of butane-2,3-diacetal protected methyl glycerate
Simon E. Fern a, Peter Heath b, Alexander J. A. Cobb a,
⇑
a School of Pharmacy, University of Reading, Whiteknights, Reading, Berks RG6 6AD, UK
b Chemical Analysis Facility (CAF), University of Reading, Whiteknights, Reading, Berks RG6 6AD, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 November 2010
Accepted 12 January 2011
The addition of aldehydes to butane-2,3-diacetal protected glycerates has been investigated. The reaction
was shown to be diastereoselective by 1H NMR spectroscopy thus confirming the ability of the diacetal to
influence the stereochemical outcome of the new stereogenic center, the configuration of which was
determined by Mosher’s ester analysis.
Ó 2011 Elsevier Ltd. All rights reserved.
Dedicated to Professor Steven V. Ley CBE
FRS on the occasion of his 65th birthday
1. Introduction
the locked nucleic acid, through the aldol addition of BDA methyl
ester 1 to an aldehyde to generate aldol product 2 (Scheme 1).4
1,2-Diacetal units are an extremely useful and a dynamic class
of structure in organic synthesis.1 Their ease of preparation and ri-
gid structure have seen them being utilized in a wide range of syn-
thetic applications, particularly in the construction of natural
products. Of particular interest to us is their ability to access qua-
ternary stereogenic centers with an excellent stereocontrol. In par-
ticular we wished to utilize butane-2,3-diacetal (BDA) protected
methyl glycerate 12 for the construction of a novel class of locked
nucleic acid (LNA) (Scheme 1).3 The synthesis of LNAs is often ham-
pered by the requirement for multi-step transformations3c and we
envisaged that the use of an asymmetric synthesis using BDA
would provide a more direct route to this class of molecule. Our
aim, therefore, was to generate the quaternary stereocenter in
We found, however, that this initial aldol step has not been
extensively investigated. Although Ley et al. detail the addition of
a range of electrophiles to ester 1, including acetone, no aldehydes
were reported in detail.4a Pohmmakotr et al. have utilized the
enantiomer of ester 1 in an aldol process, but interestingly re-
ported that no diastereoselectivity was observed.4b
Therefore, we decided to study this reaction in much more de-
tail in order to understand what effect, if any, the butane-diacetal
framework might have upon the formation of the new stereocenter
(Scheme 2).
OMe
OH
Me
O
O
Me
R2
R1O
OMe
Li-Amide base
O
OMe
Me
THF, -78 o
C
O
OMe
OH
OMe
2
O
O
O
O
Me
OMe
R1O
Me
Me
O
O
O
O
R2
R1O
+
Me
O
+
O
Me
OMe
R1O
H
R2
OMe
Me
OMe
OH
H
R2
1a : R1 = Me
1b : R1 = Ph
O
O
Me
OMe
R2
R1O
2
1
1c : R1 = C6F6
O
H
N
O
H
O
O
3
O
N
R2
HO
Scheme 2. Aldol reaction of butane-2,3-diacetal glycerates with aldehydes.
HO
Locked Nucleic Acid
2. Results and discussion
Scheme 1. Proposed synthesis of a locked nucleic acid (LNA) utilizing butane-2,3-
diacetal chemistry.
Our investigation began by studying the effect of the BDA ester
upon the reaction. We, therefore, tested methyl glycerate 1a along-
side phenyl 1b and perfluorophenyl 1c glycerates using benzalde-
hyde as an electrophile and standard enolate forming conditions
⇑
Corresponding author.
E-mail address: a.j.a.cobb@reading.ac.uk (A.J.A. Cobb).
0957-4166/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2011.01.004

150
S. E. Fern et al. / Tetrahedron: Asymmetry 22 (2011) 149–152
(Scheme 2 and Table 1, entries 1–3). Under these conditions, we
found that methyl ester (Table 1, entry 1) gave us a 2.1:1 diastereo-
meric ratio of aldol products 2:3,5 suggesting that butane-2,3-diac-
etal backbone does exert some stereochemical preference on the
aldol reaction. The phenyl ester gave a slightly improved ratio of
2.7:1, but to the detriment of the yield (Table 1, entry 2), while
the perfluorophenyl ester gave no reaction at all (Table 1, entry
3). In all cases, and as has been previously observed, none of the
diastereoisomer, whereby the ester group lies in the equatorial po-
sition was observed.
In each case, the formation of (R)-secondary alcohol was fa-
vored, as determined by Mosher’s ester analysis, and in roughly
similar diastereoselectivities. Changes in the electronic nature of
the aromatic aldehyde had little effect on diastereoselectivity com-
pared to benzaldehyde, but usefully, the non-aromatic aldehyde
hexanal also reacted to form the predicted major diastereoisomer.
The yields of the reactions using hexanal and p-anisaldehyde
(entries 6 and 7) were significantly improved via the addition of
HMPA, as were (though not significantly) the diastereoselectivities,
although this additive did not seem to enhance the reaction of
other electrophiles.
We also used the BDA aldehyde 62 as an electrophile and only
obtained the single diastereoisomer 2h, though in a modest 34%
yield (Scheme 3), again favoring the (R)-isomer.4c
Table 1
Optimization study
Entry
Ester
Base
Aldehyde
Product
Yield (%)
dra
1
2
3
4
5
6
7
1a
1b
1c
1a
1a
1a
1a
LDA
LDA
LDA
(À)-5
(+)-5
(À)-5
(+)-5
PhCHO
PhCHO
PhCHO
PhCHO
PhCHO
PhCH@CHCHO
PhCH@CHCHO
2a/3a
2b/3b
2c/3c
2a/3a
2a/3a
2d/3d
2d/3d
66
56
0
64
60
40
51
2.1:1
2.7:1
—
3:2
7:4
2:1
3:2
OMe
O
Me
O
OMe
Me
O
Me
H
O
6
O
OMe
O
Me
OMe
MeO
LDA, THF, -78 oC
a
Based on isolated yield.
In an attempt to improve the diastereoselectivity, we then
Me
MeO
Me
decided to see the influence that the chiral bases might have on
the reaction. Commonly used chiral lithium amides 4 and 5
(Fig. 1) were applied under the same conditions, but in all cases
the diastereoselectivity was adversely affected, although the major
diastereoisomer remained the same (Table 1, entries 4–7).
O
O
OMe
O
HO
OMe
O
O
Me
H
MeO
OMe
2h
Me Me
Me Me
Me
Ph
N
H
Ph
Ph
N
H
Ph
Scheme 3. A bulky aldehyde appears to only give one diastereoisomer.
4
5
In order to explain the diastereoselectivity of the process, we
first needed to know the nature of the enolate, and this was
achieved by trapping it with trimethyl silyl chloride and examina-
tion of the resulting 2D NOE NMR signals. This suggested that it
was the Z-enolate, which was formed (Fig. 2).
Figure 1. Chiral lithium amide bases.
We, therefore, proceeded to test a variety of aldehydes using the
original conditions and saw that in each case, a degree of diastere-
oselectivity was obtained. In every case, the diastereoisomers were
separable (Table 2).
OMe
O
Me
Table 2
Substrate scope
MeO
Me
OMe
O
OMe
OH
O
Si Me
Me
Me
O
O
Me
R1
Me
OMe
MeO
O
OMe
Figure 2. Relevant 2D NOE correlations showing the enolate geometry.
LDA, THF
O
2
Me
This geometry is in agreement with the predicted outcome
using the Ireland model6 (see Fig. 3) and generates an enolate with
concave and convex character. The facial selectivity of the aldol
addition may well be a consequence of this structural feature, as
the electrophile is much more likely to prefer an approach from
the more open convex face (Fig. 3).
O
-78 o
C
O
Me
OMe
1a
MeO
+
OMe
O
OH
Me
O
H
R1
O
Me
OMe
R1
MeO
O
Convex Face
3
Preferred
OMe
Entry
Aldehyde
Yield (%)
Product
dra
O
H
MeO
LiO
1
2
3
4
PhCHO
66
56
60
40
46
92
95
2a:3a
2d:3d
2e:3e
2f:3f
2g:3g
2e:3e
2f:4f
2.1:1
1.4:1
2.3:1
1.7:1
2.3:1
3:1
OMe
Li
O
OMe
PhCH@CHCHO
n-C6H11CHO
p-MeOC6H4
p-F-C6H4
n-C6H11CHO
p-MeOC6H4
O
O
N
O
MeO
Me
Me
MeO
Concave Face
Hindered
5
6b
7b
2:1
Figure 3. Deprotonation of the ester gives the Z-enolate, which is concave/convex
in character and where the electrophile is likely to prefer approach from the least
hindered face.
a
Based on isolated yield.
With 10% v/v HMPA.
b

S. E. Fern et al. / Tetrahedron: Asymmetry 22 (2011) 149–152
151
The preference for compound 2 over compound 3 can be ex-
plained by invoking the Zimmerman–Traxler model for lithium-
amide mediated aldol reactions, whereby there are two possible
transition states.7 The first, requires the R-group of the aldehyde
(in Fig. 4, this is a phenyl-group) to lie over the sterically crowded
BDA group. In terms of the transition state, this is the equatorial
position (Fig. 4A), and this appears quite unfavorable. The second
possibility is where the R-group points away from the BDA frame-
work but is forced to adopt an axial position within the transition
state. Although this gives rise to a 1,3-diaxial interaction between
it and the OMe group of the ester (Fig. 4B), it appears more favor-
able than the former option.
hydes to generate both a new secondary alcohol and a quaternary
center. The new stereocenter is to a certain extent, controlled by
the BDA group, but conveniently where two diastereoisomers are
produced, they are easily separable. We are continuing in our ef-
forts to utilize these adducts in the synthesis of novel locked nu-
cleic acids.
4. Experimental
4.1. Typical experimental procedure
To a solution of diisopropyl amine (1.1 equiv) in anhydrous tet-
rahydrofuran (0.9 mL mmolÀ1
) was added drop-wise n-BuLi
(1.1 equiv, 1.6 M in hexanes) under an inert atmosphere at
À78 °C. The resulting mixture was stirred for 10 min at À78 °C, be-
fore being transferred under an inert atmosphere via cannula to a
À78 °C solution of the butane-2,3-diacetal methyl glycerate
(1 equiv) in anhydrous tetrahydrofuran (8 mL mmolÀ1). After
10 min at this temperature, the aldehyde (4.8 equiv) was added
dropwise. The resulting mixture was stirred at À78 °C for 3.5 h,
whereupon it was quenched with satd ammonium chloride. The
quenched reaction was allowed to warm to room temperature
and ether was added. The resulting mixture was washed with
water and brine and the organic layer dried over MgSO4, filtered,
and concentrated under reduced pressure to give a residue, which
was purified by flash column chromatography (ethyl acetate/hex-
anes, 1:3.4).
A
O
OMe
Li
Li
O
O
H
O
MeO
Me
O
O
3a
Me
OMe
)(
Ph
Me
O
OMe
Me
O
OMe
OMe
vs
B
O
OMe
Li
O
H
O
Ph
Li
O
O
4.1.1. (2R)- and (2S)-2-(Hydroxy-phenyl-methyl)-5,6-dimethyl-
5,6-dimethoxy-[1,4]-dioxane-2-carboxylic acid methyl ester 2a
and 3a
Ph
O
MeO
Me
OMe
2a
Me
O
OMe
Me
Following the above general procedure, butane-2,3-diacetal
methyl glycerate was reacted with benzaldehyde in the presence
of LDA. (R)-Isomer 2a: 1H NMR (400 MHz, CDCl3): 1.22 (3H, s,
CH3), 1.37 (3H, s, CH3), 3.19 (3H, s, OCH3), 3.24 (3H, s, OCH3),
3.59 (3H, s, CO2CH3), 3.76 (1H, d, J 12, CHH), 4.00 (1H, d, J 12,
CHH), 4.77 (1H, s, CH), 7.24–7.36 (5H, m, ArH). 13C NMR
(100 MHz, CDCl3): 17.69, 17.74, 48.13, 50.56, 51.97, 59.94, 75.64,
97.98, 99.74, 126.86, 127.96, 128.30, 137.89, 171.24.
O
OMe
OMe
Me
Figure 4. Although there is a 1,3-diaxial interaction between the phenyl group of
the aldehyde and the OMe group of the ester (B), it is still slightly more favorable
than if the phenyl group were to lie directly over the BDA group (A).
This explanation might also go some way to explaining the
excellent diastereoselectivity when using aldehyde 2h. The in-
creased steric bulk of this aldehyde, makes the corresponding tran-
sition state shown in Figure 4A extremely unfavorable.
Additionally, it has previously been shown that organometallic
addition to this aldehyde leads to predominantly the same syn-
(R)-product, as a result of Felkin control.8 It is possible that this,
along with our proposed mode of selectivity, leads to a matched
chirality effect and this could explain the formation of a single
diastereoisomer.
[
a
]
D = À107.1 (c 0.7, CHCl3). HRMS (m/z) Calcd for C17H24O7:
340.1522. Found: 363.1413 (MNa+). (S)-Isomer 3a: 1.24 (3H, s,
CH3), 1.37 (3H, s, CH3), 3.23 (3H, s, OCH3), 3.32 (3H, s, OCH3),
3.65 (3H, s, CO2CH3), 3.74 (1H, d, J 12, CHH), 4.16 (1H, d, J 12,
CHH), 4.92 (1H, s, CH), 7.20–7.35 (5H, m, ArH). 13C NMR
(100 MHz, CDCl3): 17.68, 17.72, 48.24, 50.44, 52.08, 56.53, 76.68,
97.74, 100.17, 126.52, 128.24, 128.40, 135.66, 171.56.
[a
]
D = À76.2 (c 0.6, CHCl3). HRMS (m/z) Calcd for C17H24O7:
340.1522. Found: 363.1412 (MNa+).
Finally, another advantage of the BDA group is the ease with
which it can be removed; this was demonstrated by exposing sub-
strate 2a to p-toluenesulfonic acid, whereupon triol 6 was obtained
in 92% yield (Scheme 4).
Acknowledgments
We wish to thank a Cancer Research UK and the University of
Reading (Reading Endowment Trust Fund) for funding this work
(S.E.F.). We also wish to thank Dr. Geoff Brown for useful NMR
discussions.
OMe
Me
OH
O
OH
p-TSA
O
O
Me
OMe
Ph
OMe
Ph
MeO
MeOH,
reflux
HO
References
OH
O
92%
6
1. For reviews see: (a) Ley, S. V.; Polara, A. J. Org. Chem. 2007, 72, 5943; (b) Ley, S.
V.; Sheppard, T. D.; Myers, R. M.; Chorghade, M. S. Bull. Chem. Soc. Jpn. 2007, 80,
1451; Also see: (c) Michel, P.; Ley, S. V. Angew. Chem., Int. Ed. 2002, 41, 3898; For
related chemistry, see: (d) Knudsen, K. R.; Stepan, A. F.; Michel, P.; Ley, S. V. Org.
Biomol. Chem. 2006, 4, 1471; (e) Bridgwood, K. L.; Tzschucke, C. C.; O’Brien, M.;
Wittrock, S.; Goodman, J. M.; Davies, J. E.; Logan, A. W. J.; Hüttl, M. R. M.; Ley, S.
V. Org. Lett. 2008, 10, 4537; (f) Ley, S. V.; Dixon, D. J.; Guy, R. T.; Palomero, M. A.;
Polara, A.; Rodrígez, F.; Sheppard, T. D. Org. Biomol. Chem. 2004, 2, 3618; (g)
Dixon, D. J.; Guarna, A.; Ley, S. V.; Polara, A.; Rodrígez, F. Synthesis 2002, 1973.
2. Michel, P.; Ley, S. V. Angew. Chem., Int. Ed. 2002, 41, 3898.
Scheme 4. BDA deprotection.
3. Conclusion
In conclusion, we have demonstrated that the BDA framework
has the ability to control an intermolecular aldol process with alde-

152
S. E. Fern et al. / Tetrahedron: Asymmetry 22 (2011) 149–152
3. For reviews see: (a) Kaur, H.; Babu, B. R.; Maiti, S. Chem. Rev. 2007, 107, 4672; (b)
Cobb, A. J. A. Org. Biomol. Chem. 2007, 5, 3260; (c) Mathé, C.; Périgaud, C. Eur. J.
Org. Chem. 2008, 1489.
4. (a) Ley, S. V.; Michel, P.; Trapella, C. Org. Lett. 2003, 5, 4553. It is mentioned in a
footnote to this reference, that when unspecified aldehydes are used, a 2:1
diastereomeric ratio of products is obtained; (b) Pohmakotr, M.; Kambutong, S.;
Tuchinda, P.; Kuhakarn, C. Tetrahedron 2008, 64, 6315; Related processes
include: (c) Areces, P.; Carrasco, E.; Light, M. E.; Plumet, J. Synlett 2009, 2500; (d)
Ley, S. V.; Dixon, D. J.; Guy, R. T.; Rodríguez, F.; Sheppard, T. D. Org. Biomol. Chem.
2005, 3, 4095; (e) Dixon, D. J.; Ley, S. V.; Polara, A.; Sheppard, T. Org. Lett. 2001, 3,
3749. And Ref.1g
5. Absolute stereochemistry determined by Mosher’s Ester analysis: see: (a) Dale, J.
A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512; (b) Hoye, T. R.; Jeffrey, C. S.;
Shao, F. Nat. Protocols 2007, 2, 2451.
6. Ireland, R. E.; Willard, A. K. Tetrahedron Lett. 1975, 16, 3975.
7. Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957, 79, 1920.
8. Boyer, J.; Allenbach, Y.; Ariza, X.; Garcia, J.; Georges, Y.; Vicente, M. Synlett 2006,
1895.
.