Control of chemoselectivity in Dieckmann ring closures leading to tetramic acids

Article abstract of DOI:10.1039/c1ob05816a

An efficient strategy for the control of the chemoselectivity in Dieckmann ring closures leading to tetramic acids derived from serine and α-methyl serine is reported, and this provides pathways to diversely substituted systems from a common starting material.

Full text of DOI:10.1039/c1ob05816a

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 6663
www.rsc.org/obc
PAPER
Control of chemoselectivity in Dieckmann ring closures leading to tetramic
acids
a
a
b
a,b
Yong-Chul Jeong, Muhammad Anwar, Tuan Minh Nguyen, Benjamin Song Wei Tan,
b,c
a
Christina Li Lin Chai* and Mark G. Moloney*
Received 24th May 2011, Accepted 20th June 2011
DOI: 10.1039/c1ob05816a
An efficient strategy for the control of the chemoselectivity in Dieckmann ring closures leading to
tetramic acids derived from serine and a-methyl serine is reported, and this provides pathways to
diversely substituted systems from a common starting material.
Diverted synthesis based on natural products possessing the
desired biological activity is an attractive route for the discovery
1–4
of new drugs; the tetramic acid nucleus, found in various
bioactive natural products such as reutericyclin and equisetin
and their derivatives, provides an important scaffold for drug
5,6
design. In this regard, we have developed effective methodology
for the enantioselective synthesis of highly functionalized tetramic
acids via biomimetic Dieckmann or aldol ring closures; thus,
the serine-derived oxazolidine 1a may be acylated to cis-2a, and
Fig. 1
no interconversion to the 2,5-trans isomer, although evidence
for rotameric equilibrium around the N-acyl unit from peak
broadening was observed; VT and NOE analysis at 60 C in CDCl
suggested that the conformation shown in
Fig. 1 was preferred, an arrangement which presumably minimises
steric interactions. This outcome is similar to that reported for the
N-formyl system 9, in which only the cis isomer was reported.
However, if N-acylation of oxazolidine 1a is conducted with TFAA
at -40 C, both cis- and trans-diastereomers 8a,b were formed, and
the stereochemistry of these were established by NOE analysis
Fig. 2). Equilibrium product ratios of between 9 : 1 and 99 : 1
correspond to energy differences of 1.3–2.7 kcal mol , and this
was confirmed by molecular modeling calculations (Spartan 04 for
Windows, using DFT [B3LYP/6-31G*] basis set) which indicated
that the cis/trans energy difference DE(cis-trans) for the N-formyl, N-
acetyl and N-trifluoroacetyl series was -0.9, -0.8 and -1.9 kcal
mol respectively (Table 1, entries 7–9), clearly favouring the
cis outcome as observed experimentally. However, although full
selectivity for the trans-N-acyloxazolidines was not possible, these
results suggested that neither would their formation be impossible.
Next, we found that appropriate adjustment of the conditions
enabled tetramate 6a to be obtained directly from tetramate 4a,
itself available in high yield from oxazolidine 2a, as shown in
Scheme 1. Thus, hydrolysis of tetramate 4a with sodium hydroxide
gave the crude carboxylic acid 4c in high yield, although this
compound was found to be unstable to silica column chromatog-
raphy. Decarboxylation of acid 4c under various conditions such
as refluxing in toluene with or without a catalytic amount (5 mol%)
of para-toluenesulfonic acid, or refluxing in wet acetonitrile, gave
tetramate 6a in moderate yield (18–30%). However, hydrolysis of
7–9
efficiently closed to tetramate 4a (Scheme 1) in a process in
which the t-butyl group simultaneously controls chemo- and
◦
10
3
stereoselectivity. However, along with the major product 4a
73%), minor amounts of the alternative products 5a (arising
◦
and 90 C in DMSO-d
6
(
by C-2 epimerisation followed by Dieckmann ring closure (vide
infra)) and of ent-5a (arising by Dieckmann ring closure of minor
amounts of trans-3a (vide infra) present in cis-2a) were also
obtained (12 and 3.5% respectively). Although these products
were obtained with high stereocontrol, the yields of the minor
products were clearly not synthetically useful and we sought to
modify this route by changing the reagents, conditions and the
substrates themselves to provide selective and efficient access to
any of tetramates 4, 5 or 6, systems which we have recently shown
1
3
◦
(
-
1 14
11,12
to be of interest as templates for drug discovery.
Firstly, in order to better understand the N-acylation of
oxazolidine 1a, its reaction with the less hindered acylating agents,
acetyl chloride and trifluoroacetyl anhydride, at room temperature
was examined; exclusive formation of the cis-oxazolidines 7 and
-
1
8a (Fig. 1) was observed, the structures of which were confirmed
by NOE analysis (Fig. 2). Room temperature VT and NOE
analysis of N-acetyl 7 confirmed the structure as 2,5-cis- with
a
Chemistry Research Laboratory, University of Oxford, Mansfield Rd,
University of Oxford, Oxford, OX1 3TA, U.K. E-mail: mark.moloney@
chem.ox.ac.uk; Tel: +44 1865 275656
b
Institute of Chemical and Engineering Sciences, Agency for Science,
Technology and Research (A*STAR), 8 Biomedical Grove, Neuros, #07-
0
1, Singapore, 138665
c
Department of Pharmacy, National University of Singapore, 18 Science
Drive 4, Singapore, 117543. E-mail: christina_chai@ices.a-star.edu.sg;
Tel: +65 6799-8522
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6663–6669 | 6663

View Article Online
Scheme 1
Fig. 2
Table 1 Compound ratio and energy differences of selected compounds (see Scheme 1 and Fig. 1)
-1 b
Calculated energy difference (kcal mol )
a
Entry
Compound
Observed cis : trans ratio
AM1
PM3
c
c
1
2
3
4
5
6
7
8
9
1a
50 : 50
40 : 60
35 : 65
85 : 15
70 : 30
20 : 80
99 : 1
99 : 1
95 : 5
—
2.346
4.250
2.406
1.616c
c
1b
1.909
c
c
1c
1.409
c
c
2a : 3a
2b : 3b
2c : 3c
7
-1.622
-2.531c
c
-1.994
-1.642
c
c
0.246
1.948
-1.608
-1.968
0.293
-1.941
-0.821
-0.904
8
9
d
d
1
1
1
0
1
2
(4a - 13)
(5a - 14)
(4b - 5a)
-3.042
-6.248
e
e
—
—
-3.583
-6.190
f
f
-4.643
-5.420
a
1
b
Determined by comparison of the integral of proton on C(2) in H NMR spectrum. Calculated by using semi-empirical AM1 and PM3 at ground state
c
d
e
in Spartan 02. Heat of formation (cis form- trans form). Heat of formation difference between 4 and 13. Heat of formation difference between 5a and
1
f
4. Heat of formation difference between 4b and 5a.
6
664 | Org. Biomol. Chem., 2011, 9, 6663–6669
This journal is © The Royal Society of Chemistry 2011

View Article Online
tetramate 4a with lithium hydroxide gave tetramate 6a directly
in 30% yield, without isolation of acid 4c. The yield via this route
of pivalaldehyde in the presence of triethylamine. Acylation of
oxazolidine (±)-1c with DCC and DMAP gave oxazolidines (±)-
2c and (±)-3c with a cis to trans ratio of 20 : 80, an outcome
which is inverted relative to 1a,b. This material when treated with
potassium tert-butoxide in dry THF at reflux gave only tetramate
(±)-5b in 72% yield, and of interest is that no other diastereomers
of this product could be isolated from the reaction mixture. The
relative stereochemistries of key compounds were established by
(
26% over 2 steps from 1a) is significantly improved over our initial
8,15
report. Moreover, we found that cyclisation of oxazolidine 2a
with 2.2 equivalents of potassium tert-butoxide in dry butanol
also gave tetramate 6a directly in 25% yield, in a sequence in
which the Dieckmann cyclisation, hydrolysis and decarboxylation
all occurred in one pot. When wet butanol was used as solvent,
8
1
the yield increased even further to 40%. Further excess amounts
H NOE experiments (Fig. 2). Tetramate (±)-5b could be readily
(
3.3 equivalents) of potassium tert-butoxide or 2.2 equivalents of
converted to (±)-6b by hydrolysis and decarboxylation in 76%
yield.
sodium methoxide instead of potassium tert-butoxide, however,
gave poorer yields of 15 and 7% respectively. Nonetheless, this
yield for tetramate 6a (35% yield overall from 1a) is improved by
more than 4-fold over our initial report.
In order to develop this process further, we sought to control the
diastereoselectivity of the initial acylation of the key oxazolidine
In order to explain the above results, the energy difference of
the heat of formation of cis- and trans-oxazolidines was calculated
using semi-empirical AM1 and PM3 in Spartan 04 (Table 1). In the
calculation results for oxazolidines 1a–c, which exist in equilibrium
between the cis and trans forms in solution, the trans form was
found to be more stable than the cis form (Table 1, entries 1–3),
but for acylated oxazolidines cis-2a–c and trans-3a–c, although cis-
2a and 2b are more stable than the corresponding trans isomers,
for the a-methyl serine derivative, it is trans-3c which is more stable
(entries 4–6), consistent with experimental observations. The pre-
ferred cis stereochemical outcome for oxazolidines 2a–b permits
an all-trans configuration of the five membered ring (as shown in
Fig. 3, and assuming the envelope conformations as indicated in
Fig. 1), in which the nitrogen is slightly pyramidalised, and which
places all three contiguous bulky groups in pseudoequatorial
arrangements which minimise steric interactions; however, for the
1,3-trans products 3a,b, adjacent pairs of cis substituents cannot
avoid steric interactions in both possible conformers. The switch to
preference for trans-3c is a result of the presence of the additional
8
1
a (Scheme 1) by modifying the steric bulk of the C-5 ester
function, expecting improved 2,5-trans selectivity which would in
turn provide access to tetramates of type 5. Although oxazolidine
1
mono-methyl malonate using dicyclohexylcarbodiimide (DCC)
in the presence of a catalytic amount of dimethylaminopyridine
a is formed as a 1 : 1 mixture of diastereomers, acylation with
(
DMAP) gives the acylated oxazolidine (88% yield) as an 85 : 15
7–9
mixture of cis-2a and trans-3a isomers (Scheme 1). However,
for the slightly more bulky ethyl ester 1b, acylation with ethyl
malonyl chloride in the presence of triethylamine gave oxazolidines
2
b and 3b (75% yield) only as a 70 : 30 mixture in favour of the
16
cis isomer (Scheme 2). Treatment of this mixture with DMAP
at room temperature gave not only tetramate ent-5a (arising by
cyclisation of the minor isomer trans-3b) by precipitation in a
solution of ethyl acetate and petroleum ether with improved yield
1
2
methyl group (R = R = Me) (Fig. 3) whose bulk can then
be more readily accommodated on the oxazolidine ring. These
calculations indicate that in the acylation reaction of oxazolidines
1a–c, the favoured products are also the thermodynamically more
stable ones; the facile equilibrium between cis- and trans-1a–
c is critical in this outcome, and it has been recently reported
(
25%) but also cis-oxazolidine 2b by column chromatography
of the extraction residue (>97% de) (Scheme 2). As expected,
the pure cis-2b could be readily converted to tetramate 4b using
the standard cyclisation methodology (Scheme 2). This outcome
confirms that although oxazolidines 1 can freely equilibrate, after
N-acylation, the products cis-2b and trans-3b cannot interconvert
and will cyclise with different chemoselectivities, giving isomeric
tetramates 4b or ent-5a respectively.
In an attempt to bias the trans selectivity of this process even
further, an a-methyl analogue was examined (Scheme 1). Thus, a
3
17
that this equilibrium is catalysed by water. On the other hand,
configurationally stable related 2-trifluoromethyl-1,3-oxazolidines
derived from serine have recently been reported and exploited as
18
unusual pseudoprolines.
In the subsequent Dieckmann cyclisation of cis-oxazolidines
2a–c using potassium tert-butoxide, an equilibrium of enolates
10–12 is established after deprotonation (Fig. 3), with the closure
of enolate 11 leading to tetramates 4a,b and the closure of enolate
5 : 65 cis and trans mixture of oxazolidine (±)-1c was prepared
from commercially available a-methyl-DL-serine via esterification
with thionyl chloride in methanol, followed by condensation
Scheme 2
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6663–6669 | 6665

View Article Online
Fig. 3
2 to the alternative system 5. The cyclised tetramates 13 (R =
1
the reactions leading to tetramate products are unlikely to be
thermodynamically controlled (for example, 4a and 5a do not
equilibrate under basic conditions), the magnitude of the relative
rate constants of the ring closures are likely to be reflected in the
calculated stabilities of the products shown in Table 1. In the case
of trans-oxazolidines 3a–c, formation of enolate 15 is followed by
ring closure to give the tetramates of type 5, in which the bulky
t-butyl group naturally ends up on the exo-face of the bicyclic ring,
but now as the enantiomeric series; for 3c, this is the only possible
course.
In conclusion, closure of cis-oxazolidines 2 to tetramates of
type 4 is most readily achieved in good yield using the Dieckmann
reaction conditions (KOt-Bu (1.1 eq), dry t-BuOH, reflux), while
use of alternate conditions (KOt-Bu (2.2 eq), wet t-BuOH, reflux)
gives the simple bicyclic tetramate 6 directly. On the other hand,
-
1
H) and 14 are much less stable by about 3–6 kcal mol than the
corresponding tetramates 4a and 5a (Table 1, entries 10 and 11),
since the bulky t-butyl group ends up on the more hindered concave
endo-face of the bicyclic system, and therefore 13 and 14 are not
formed by closure of enolates 10 and 11 respectively. However, of
interest is that tetramate 4b is more stable than tetramate 5a by 4.6
-
1
kcal mol (entry 12), consistent with the experimental observation
that it is formed preferentially from oxazolidine 2b, although both
tetramates 4a and 5a as the major and minor forms respectively are
formed from oxazolidine 2a where the energy difference between
them is much lower. When DMAP is used as a mild base in the
closure of cis-2b and trans-3b, only the most acidic C(6) position
is deprotonated, giving enolates 12 and 15 and cyclisation forms
the more stable 5a and ent-5a respectively (Fig. 3). Although
6
666 | Org. Biomol. Chem., 2011, 9, 6663–6669
This journal is © The Royal Society of Chemistry 2011

View Article Online
tetramates of type 5 are obtainable in good yield from trans-
oxazolidines 3 by the use of similar conditions (KOt-Bu (1.1 eq),
dry THF, reflux). Thus, by exploiting steric effects around the
periphery of an oxazolidine template derived from serine, it is
possible to control chemoselectivity and therefore the pathway
of competing Dieckmann cyclisation reactions, giving efficient
synthetic routes to variously functionalised tetramic acids from
a common starting material.
minor isomer; 13.9 (CH
C(CH ), 43.5 (NCOCH
(C-5), 78.5 (C-4), 98.7 (C-2), 167.4 (NCO), 167.7 (CO
2
CH
3
), 21.5 (CCH
3
), 27.1 C(CH
3
)
3
), 39.2
), 66.1
3
)
3
2
), 53.4 (CO
2
CH ), 61.3 (CH
2
2
CH
3
2
CH ), 172.0
3
+
+
+
(CO
for C15
2
CH
3
);m/z (ES )338.19(M+ Na );HRMS(M+ Na )Found
Na 338.1576, calculated 338.1572.
H
25NO
6
(±)-(3S*,7aS*)-Ethyl 3-(tert-butyl)-7-hydroxy-7a-methyl-5-oxo-
1,3,5,7a-tetrahydropyrrolo[1,2-c]oxazole-6-carboxylate (±)-5b
To a solution of oxazolidine (±)-3c (1.2 g, 3.6 mmol) in dry THF
Experimental
t
(25 ml) was added KO Bu (0.42 g, 3.73 mmol) and solution was
heated at reflux for 4 h, cooled to room temperature and parti-
tioned between ether (15 ml) and water (2 ¥ 15 ml). The aqueous
layer was acidified with 2 M HCl and extracted with ethyl acetate
(3 ¥ 20 ml). The combined ethyl acetate extracts were washed with
Oxazolidines 1a–c, 2a–c and 3a–c were prepared using the
published procedures.
8,19
(
±)-Methyl 2-(tert-butyl)-5-methyloxazolidine-5-carboxylate 1c
brine, dried over MgSO
was purified by flash column chromatography (EtOAc : MeOH,
: 1) giving (±)-5b as a light yellow solid (0.78 g, 72% yield). d
4
and evaporated in vacuo. The residue
To a-methyl-DL-serine ester hydrochloride (2.0 g, 11.8 mmol) in
petrol (100 ml) was added triethylamine (2.46 ml, 17.7 mmol) and
pivaldehyde (1.54 ml, 14.2 mmol). The mixture was heated at reflux
for 20 h with continuous removal of water using a Dean–Stark
head then filtered and washed with ether (30 ml). The combined
filtrates were concentrated in vacuo to give a crude 35 : 65 cis and
trans mixture of oxazolidine (±)-1c (1.6 g, 68%) as a colourless
6
H
(
400 MHz, CDCl
3
) 0.97 (9H, s, C(CH
3
)
3
), 1.34 (3H, t, J 7.1 Hz,
), 3.39 (1H, d, J 8.1 Hz, CHH), 3.86
CH ), 4.64
) 14.1, 20.8, 25.5, 34.6,
CH
(
2
CH
3
), 1.55 (3H, s, CH
3
1H, d, J 8.1 Hz, CHH), 4.33 (2H, q, J 7.1 Hz, CH
2
3
(
1H, br s, CHt-Bu). d
C
(100 MHz, CDCl
3
+
6
+
1.6, 71.0, 72.2, 97.5, 167.3, 189.3, 206.1. m/z (ES ) 306.16 (M
oil. d
.45 (3H, s, CCH
CO CH ), 3.94 (1H, d, J 8.5 Hz, CHH), 4.15 (1H, s, CHt-Bu).
minor isomer; 0.92 (9H, s, C(CH ), 1.42 (3H, s, CCH ),3.52
1H, d, J 8.2 Hz, CHH), 3.74 (3H, s, CO CH ), 4.09 (1H, d, J 8.2
Hz, CHH), 4.29 (1H, s, CHt-Bu). d (100 MHz, CDCl ) major
isomer; 23.3 (CH ), 25.0 C(CH ), 33.9 C(CH ), 52.4 (CO CH ),
5.1 (C-5), 75.0 (C-4), 98.7 (C-2), 175.2 (CO ); minor isomer; 23.0
), 52.2 (CO CH ), 65.2 (C-5),
); m/z (ES ) 202.16 (M + H ),
H
(400 MHz, CDCl
3
) major isomer; 0.97 (9H, s, C(CH
3
)
3
),
+
+
Na ) HRMS (M + Na ) Found for C14
H
21NO Na 306.1315,
5
1
3
), 3.49 (1H, d, J 8.5 Hz, CHH), 3.75 (3H, s,
calculated 306.1312.
2
3
3
)
3
3
Synthesis of (2R,5R)-2-tert-butyl-7-ethoxycarbonyl-6-hydroxy-8-
oxo-1-aza-3-oxabicyclo[3.3.0]oct-6-ene (ent-5a) and (2R,5S)-2-
tert-butyl-1-ethoxycarbonylacetyl-4-ethoxycarbonyl-1,3-
oxazolidine (cis-2b) (Scheme 2)
(
2
3
C
3
3
3
)
3
3
)
3
2
3
6
2
(
7
2
CH
3
), 24.6 C(CH
3
)
3
), 33.0 C(CH
3
)
3
2
3
To a mixture of cis-2b and trans-3b (ratio of cis : trans = 70 : 30,
+
+
4.6 (C-4), 98.7 (C-2), 175.4 (CO
24.14 (M + Na ); HRMS (M + H ) Found 202.1446, calculated
202.1438.
2
6.0 g, 19.0 mmol) in dichloromethane (50 ml) was added 4-
+
+
◦
(dimethylamino)pyridine (2.3 g, 19.0 mmol) at 0 C. The mixture
for C10
H
20NO
3
was stirred for 1 day at room temperature. The crude reaction
mixture was washed with acidic water (1 N HCl) and the organic
(
±)-(2S*,5S*)- and (±)-(2S*,5R*)-Methyl 2-(tert-butyl)-1-(3-etho-
layer was dried by using MgSO . After removing dichloromethane,
4
xy-3-oxopropanoyl)-5-methyloxazolidine-5-carboxylate (2c and 3c)
the crude was dissolved in ethyl acetate and slowly added petrol.
The white solid ent-5a (1.3 g, 4.76 mmol, 25% yield) was obtained
by filtration. The filtered solution was concentrated in vacuo
followed by flash column chromatography gave cis-2b (3.9 g, 65%
yield).
To a solution of (±)-methyl 2-(tert-butyl)-4-methyloxazolidine-4-
carboxylate 1c (0.74 g, 3.7 mmol), DCCI (0.84 g, 4.1 mmol) and
DMAP (24 mg, 0.20 mmol) in DCM (15 ml) at 0 C was added
◦
ethyl hydrogen malonate (0.54 g, 4.05 mmol) in DCM (10 ml).
◦
The mixture was stirred at 0 C for 15 min and then at room
Synthesis of (2R,5R)-2-tert-butyl-6,8-dioxo-1-aza-3-oxabicyclo-
temperature for 4 h. The reaction mixture was filtered, the residue
was washed with DCM (3 ¥ 10 ml). The combined filtrates were
evaporated in vacuo and purified by flash column chromatography
[
3.3.0]octane 6a from 4a
A mixture of ester 4a (260 mg, 1.02 mmol) and lithium hydroxide
monohydrate (85 mg, 2.04 mmol) in tetrahydrofuran (12 ml),
methanol (3 ml) and water (3 ml) was heated at reflux for 15 h
then cooled to room temperature. The mixture was extracted with
ethyl acetate and acidic water (1 N HCl) then the organic layer was
(
EtOAc : petrol, 1 : 3) to give the oxazolidines (±)-2c and 3c (0.84
g, 72% yield) in a ratio of 1 : 4 as a colourless oil. d (400 MHz,
CDCl ) major isomer; 0.67–0.95 (9H, s, C(CH ), 1.17 (3H, t,
J 7.2 Hz, CH CH ), 1.60 (3H, s, CH ), 3.28 (1H, d, J 15.5 Hz,
CHH), 3.40 (1H, d, J 15.5 Hz, CHH), 3.63–3.78 (4H, m, CHH
and CO CH ), 4.08 (2H, q, J 7.2 Hz, CH CH ), 4.32 (1H, br d, J
.4 Hz, CHH), 5.48 (1H, br s, CHt-Bu); minor isomer; 1.62 (3H,
H
3
)
3 3
2
3
3
dried by using MgSO
by flash column chromatography gave tetramic acid 6a (60 mg,
0% yield).
4
and evaporated in vacuo. The purification
2
3
2
3
6
3
s, CH
3
), 0.92 (9H, s, C(CH
3
)
3
), 1.42 (3H, s, CCH
CH ), 4.09 (1H, d, J 8.2 Hz,
(100 MHz, CDCl ) major isomer;
), 26.4 C(CH ), 38.9 C(CH ), 43.1
), 61.2 (CH CH ), 65.8 (C-5), 78.5 (C-
), 98.3 (C-2), 166.9 (NCO), 167.7 (CO CH ), 173.6 (CO CH );
3
),3.52 (1H, d,
J 8.2 Hz, CHH), 3.74 (3H, s, CO
CHH), 4.29 (1H, s, CHt-Bu). d
3.9 (CH CH ), 20.5 (CCH
NCOCH ), 52.8 (CO CH
2
3
Synthesis of (2R,5R)-2-tert-butyl-6,8-dioxo-1-aza-3-
oxabicyclo[3.3.0]octane 6a via 4c
C
3
1
(
4
2
3
3
3
)
3
)
3 3
2
2
2
2
3
A mixture of ester 4a (500 mg, 1.96 mmol) and sodium hydroxide
(40 mg, 4.00 mmol) in tetrahydrofuran (20 ml), methanol (5 ml)
2
3
2
3
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6663–6669 | 6667

View Article Online
and water (5 ml) was heated at reflux for 8 h then cooled to
room temperature. The mixture was extracted with ethyl acetate
and acidic water (1 N HCl) then the organic layer was dried
a colourless oil (111 mg, 91% yield). R
f
0.37 (EA : CH
); nmax/cm (film) 1744 (s, C O), 1667
(400 MHz, CDCl , at r.t.) 0.91 (s, br, 9H, C(CH ),
2.19 (s, 3H, H-9), 3.79 (s, br, 3H, CO CH ), 4.08 (s, br, 1H, CHH),
CH )), 5.32
, at 67 C) 0.98 (s, 9H,
Me), 4.14 (t, J = 8.5
2
Cl , 1 : 9);
2
2
0
-1
[a] -33.2 (c 1.0 in CHCl
3
D
(s, C O); d
H
3
)
3 3
by using MgSO
acid 4c (460 mg, 1.91 mmol, 97% yield). A solution of crude
carboxylic acid 4c (460 mg, 1.91 mmol) in wet CH CN (15 ml)
4
. Evaporation in vacuo gave crude carboxylic
2
3
4.46 (s, br, 1H, CHH), 4.54 (s, br, 1H, 1H, CH(CO
2
3
t
◦
3
(s, br, 1H, CH Bu); d
H
(500 MHz, CDCl
3
was heated at reflux for 5 h then cooled to room temperature.
The mixture was extracted with ethyl acetate and acidic water
C(CH
3
)
3
), 2.22 (s, 3H, H-9), 3.83 (s, 3H, CO
2
Hz, 1H, CHH), 4.49 (dd, J = 3.8, 8.5 Hz, 1H, CHH), 4.68 (s, br,
◦
(
1 N HCl) then the organic layer was dried by using MgSO
4
and
1H, H-5), 5.33 (s, 1H, H-2); d
0.86 (s, 9H, t-Bu), 2.13 (s, 3H, H-9), 3.73 (s, 3H, CO
(t, J = 8. Hz, 1H, CHH), 4.33 (dd, J = 3.2, 8.5 Hz, 1H, CHH),
4.96 (dd, J = 3.5, 7.9 Hz, 1H, H-5), 5.18 (s, 1H, CHt-Bu); d
(125 MHz, CDCl , at r.t.) 23.49 (C-9), 25.69 (t-Bu), 37.55 (Ct-Bu),
H
(500 MHz, DMSO-d
6
, at 90 C)
evaporated in vacuo. Purification by flash column chromatography
gave tetramic acid 6a (115 mg, 0.57 mmol, 30% yield). Tetramic
acid 6a was also obtained in toluene with (22% yield) or without
2
Me), 4.06
C
(
18% yield) catalytic amount (5.0 mol%) of p-toluenesulfonic acid
3
monohydrate from crude carboxylic acid 4c.
52.68 (C-7), 59.88 (C-5), 68.08 (C-4), 96.62 (C-2), 170.22, 171.87;
+
m/z (ESI+) 252.12 ([M + Na] , 94%); HRMS (FI+) calculated for
+
Synthesis of (2R,5R)-2-tert-butyl-6,8-dioxo-1-aza-3-oxabicyclo-
C
11
H
19NO
4
([M + Na] ) 229.1314, found 229.1319.
[3.3.0]octane 6a from 2a
t
(2R,5S)-Methyl 2-(tert-butyl)-1-(2,2,2-trifluoroacetyl)-
oxazolidine-5-carboxylate 8a
A solution of oxazolidine 2a (1.0 g, 3.32 mmol) and KOBu (0.82 g,
7
t
.31 mmol) in wet Bu OH (40 ml) was heated at reflux for 15 h then
cooled to room temperature. The mixture was extracted with ethyl
acetate and acidic water (1 N HCl) then the organic layer was dried
At r.t.: To a solution of the oxazolidine 1a (50 mg, 0.27 mmol,
1
.0 eq.) in CH
2
2
Cl was added a solution of TFAA (56 ml, 0.40
by using MgSO . Purification by flash column chromatography
4
mmol, 1.5 eq.) and pyridine (65 ml, 0.80 mmol, 3.eq.) at r.t. The
reaction mixture was stirred at r.t for 3.5 h, after which it was
quenched with water, washed with 10% aq. HCl and extracted with
gave tetramic acid 6a (260 mg, 1.33 mmol, 40% yield). Tetramic
t
acid 6a was also obtained with 2.2 equivalents of KOBu in dry
t
t
t
Bu OH (25% yield), 3.3 equivalents of KOBu in wet Bu OH (15%
yield) and 2.2 equivalents of NaOMe in benzene and methanol
CH
MgSO
2
Cl
2
. The combined organic layer was dried with anhydrous
, filtered and solvents evaporated to give the crude material
4
(
7% yield along with 13% yield of 4a) from oxazolidine 2a.
which was purified on silica gel via flash column chromatography
eluent: 20% ethyl acetate–petroleum ether) to afford the product
(
(
2R,5R)-2-tert-Butyl-5-methyl-6,8-dioxo-1-aza-3-
as a colourless oil (60 mg, 80% yield).
oxabicyclo[3.3.0]octane (±)-6b
◦
At -40 C: To a solution of the oxazolidine 1a (50 mg,
0
.27 mmol, 1.0 eq.) in CH
2
Cl was added a solution of TFAA
2
A solution of tetramic acid (±)-5b (0.21 g, 0.74 mmol) in wet
(
56 ml, 0.40 mmol, 1.5 eq.) and pyridine (65 ml, 0.80 mmol, 3.eq.)
CH CN (10 ml) was heated at reflux for 18 h, cooled to room
3
◦
◦
at -40 C. The reaction mixture was stirred at -40 C overnight,
quenched with water, washed with 10% aq. HCl and extracted with
CH
MgSO
temperature and partitioned between EtOAc (20 ml) and 2 M
HCl (20 ml). The aqueous layer was extracted with EtOAc (20
ml) and the combined organic extracts were washed with brine
2
Cl
2
. The combined organic layer was dried with anhydrous
, filter and solvents evaporated to give the crude product,
4
(
25 ml), dried over MgSO and evaporated in vacuo. The was
4
which was purified on silica gel via flash column chromatography
eluent: 20% ethyl acetate–petroleum ether) to afford the pure cis-
isomer (25 mg, 33% isolated yield) and trans-isomer (23 mg, 31%
purified by flash column chromatography (EtOAc : petrol; 1 : 1)
increasing polarity to EtOAc giving (±)-6b as a white solid (0.12 g,
(
◦
7
1
3
d
4
6% yield). m.p. 95 C; d
H
(400 MHz, CDCl
3
) 1.00 (s, 9H, CMe ).
3
2
0
isolated yield). R
c 0.4 in CHCl ); nmax/cm (film) 1758 (s, C O), 1707 (s, C O);
(400 MHz, CDCl ) 0.98 (s, 9H, C(CH ), 3.81 (s, 3H, CO CH ),
.06 (t, J = 8.3 Hz, 1H, CHH), 4.56 (dd, J = 2.3, 9.1 Hz, 1H, CHH),
f
0.58 (PE : EA, 4 : 1); colourless oil; [a] -15.6
.56 (s, 3H, CH
.57 (d, J = 11.5 Hz, C7), 3.84 (d, J = 7.5 Hz, C4), 4.98 (s, 1H, C2);
(100 MHz, CDCl ); d 22.4 (CH ), 25.6 ((CH ), 34.9 (CMe ),
4.8 (C7), 71.1 (C4), 75.1 (C5), 95.8 (C2), 173.1 (C8), 206.2 (C6);
3
), 3.14 (d, J = 11.5 Hz, C7), 3.54 (d, J = 7.5 Hz, C4),
D
-
1
(
d
4
3
H
3
3
)
3
2
3
C
3
3
3
)
3
3
-
-
-
4.76 (d, J = 6.3 Hz, 1H, CH(CO
2
CH
3
)), 5.32 (s, 1H, CHt-Bu); d
C
m/z (ES ); 210.13 (M - H ); HRMS (M - H ); calculated for
; 210.1136; found; 210.1134.
(100 MHz, CDCl
3
) 25.80 (t-Bu), 37.30 (-Ct-Bu), 52.89 (-CO
2
Me),
C
11
H
16NO
3
1
5
9
8.44 (C-5), 69.13 (C-4), 98.82 (C-2), 115.79 ( JC–F = 286.3 Hz, C-
2
), 158.93 ( JC–F = 36.3 Hz, C-8), 168.70 (C-6); m/z (ESI+) 306.09
(
2R,5S)-Methyl 1-acetyl-2-(tert-butyl)-oxazolidine-5-carboxylate
+
(
(
[M + Na] , 100%); HRMS (ESI+) calculated for C11
[M + Na] ) 306.0924, found 306.0922.
H
16
F
3
NO Na
4
7
+
To a solution of the oxazolidine (100 mg, 0.53 mmol, 1.0 eq.)
in CH
2
Cl was added a solution of acetyl chloride (57 ml, 0.80
2
(
5
2R,5S)-Methyl 2-(tert-butyl)-1-(2,2,2-trifluoroacetyl)oxazolidine-
-carboxylate 8b
mmol, 1.5 eq.) and pyridine (130 ml, 1.60 mmol, 3.eq.) at r.t. The
reaction mixture was stirred at r.t. for 3.5 h, after which it was
quenched with water, washed with 10% aq. HCl and extracted with
2
D
0
R 0.52 (PE : EA, 4 : 1); Colourless oil; [a] -33.5 (c = 0.6 in
1
f
-
CH
2
Cl
2
. The combined organic layer was dried with anhydrous
, filtered and solvents evaporated to give the crude material,
CHCl
MHz, CDCl
(dd, J = 1.5, 9.3 Hz, 1H, CHH), 4.50 (dd, J = 7.3, 9.3 Hz, 1H,
CHH), 4.76 (dt, J = 1.8, 7.1 Hz, 1H, CH(CO CH )), 5.63 (s, 1H,
3
); nmax/cm (film) 1753 (s, C O), 1704 (s, C O); d
H
(400
MgSO
4
3
) 0.98 (s, 9H, C(CH ), 3.82 (s, 3H, CO CH ), 4.27
3
)
3
2
3
which was purified on silica gel via flash column chromatography
eluent: ethyl acetate–CH Cl , 2% to 5%) to afford the product as
(
2
2
2
3
6
668 | Org. Biomol. Chem., 2011, 9, 6663–6669
This journal is © The Royal Society of Chemistry 2011

View Article Online
CHt-Bu); d
C
(125 MHz, CDCl
3
) 26.21 (t-Bu), 39.57 (-Ct-Bu), 53.46
6 R. Schobert and A. Schlenk, Bioorg. Med. Chem., 2008, 16, 4203–
1
4221.
(
-CO
2
Me), 60.04 (C-5), 71.39 (C-4), 97.72 (C-2), 115.88 ( JC–F
=
7
M. Anwar and M. G. Moloney, Tetrahedron Lett., 2007, 48, 7259–
2
286.3 Hz, C-9), 158.12 ( JC–F = 36.3 Hz, C-8), 170.83 (C-6); m/z
7
262.
+
(
ESI+) 306.09 ([M + Na] , 100%); HRMS (ESI+) calculated for
8 M. D. Andrews, A. G. Brewster, K. M. Crapnell, A. J. Ibbett, T. Jones,
M. G. Moloney, K. Prout and D. Watkin, J. Chem. Soc., Perkin Trans.
+
C
11
H
16
F
3
NO
4
Na ([M + Na] ) 306.0924, found 306.0917.
1
, 1998, 223–235.
9
M. D. Andrews, A. G. Brewster and M. G. Moloney, Synlett, 1996,
Acknowledgements
612–614.
1
1
1
0 D. Seebach, A. R. Sting and M. Hoffmann, Angew. Chem., Int. Ed.
Engl., 1996, 35, 2708–2748.
This work was generously supported by the Korea Research Foun-
dation Grant funded by the Korean Government (MOEHRD)
1 C. A. Holloway, C. J. Matthews, Y.-C. Jeong, M. G. Moloney, C. F.
Roberts and M. Yaqoob, Chem. Biol. Drug. Des., 2011, 78, 229–235.
2 Y.-C. Jeong and M. G. Moloney, Synlett, 2009, 2487–2491.
(
KRF-2006-352-C00044), Galapagos SASU and University of
Oxford IUIF Fund. BTSW gratefully acknowledges the award
of an ASTAR Graduate Scholarship.
13 D. Seebach and J. D. Aebi, Tetrahedron Lett., 1984, 25, 2545–2548.
1
1
4 N. Isaacs, Physical Organic Chemistry, Longman, London, 1995.
5 M. D. Andrews, A. Brewster and M. G. Moloney, Tetrahedron:
Asymmetry, 1994, 5, 1477–1478.
References
16 D. Seebach, J. D. Aebi, M. G. Coquaz and R. Naef, Helv. Chim. Acta,
987, 70, 1194–1216.
17 M. S. Anson, H. F. Clark, P. Evans, M. E. Fox, J. P. Graham, N. N.
Griffiths, G. Meek, J. A. Ramsden, A. J. Roberts, S. Simmonds, M. D.
Walker and M. Willets, Org. Process Res. Dev., 2011, 15, 389–397.
18 G. Chaume, O. Barbeau, P. Lesot and T. Brigaud, J. Org. Chem., 2010,
75, 4135–4145.
1
1
2
3
S. Danishefsky, Nat. Prod. Rep., 2010, 27, 1114–1116.
I. Paterson and E. A. Anderson, Science, 2005, 310, 451–453.
M. A. Koch and H. Waldmann, Drug Discovery Today, 2005, 10,
4
71.
4
5
R. B. Breinbauer, I. R. Vetter and H. Waldmann, Angew. Chem., Int.
Ed., 2002, 41, 2878.
19 M. Anwar, A. R. Cowley and M. G. Moloney, Tetrahedron: Asymmetry,
2010, 21, 1758–1770.
B. J. L. Royles, Chem. Rev., 1995, 95, 1981–2001.
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 6663–6669 | 6669