Biomimetic synthesis, antibacterial activity and structure-activity properties of the pyroglutamate core of oxazolomycin

Article abstract of DOI:10.1039/c2ob00042c

Biomimetic intramolecular aldol reactions on oxazolidine templates derived from serine may be used to generate densely functionalised pyroglutamates, which are simpler mimics of the right hand side of oxazolomycin. Some of the compounds from this sequence exhibit in vivo activity against S. aureus and E. coli, suggesting that pyroglutamate scaffolds may be useful templates for the development of novel antibacterials, and cheminformatic analysis has been used to provide some structure-activity data.

Full text of DOI:10.1039/c2ob00042c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 3472
www.rsc.org/obc
PAPER
Biomimetic synthesis, antibacterial activity and structure–activity properties
of the pyroglutamate core of oxazolomycin†
Plamen Angelov, Yui Kwan Sonia Chau, Paul J. Fryer, Mark G. Moloney,* Amber L. Thompson and
Paul C. Trippier
Received 6th January 2012, Accepted 28th February 2012
DOI: 10.1039/c2ob00042c
Biomimetic intramolecular aldol reactions on oxazolidine templates derived from serine may be used to
generate densely functionalised pyroglutamates, which are simpler mimics of the right hand side of
oxazolomycin. Some of the compounds from this sequence exhibit in vivo activity against S. aureus and
E. coli, suggesting that pyroglutamate scaffolds may be useful templates for the development of novel
antibacterials, and cheminformatic analysis has been used to provide some structure–activity data.
The oxazolomycin class of natural products (Fig. 1), which
includes oxazolomycins A–C 1a–c, 16-methyloxazolomycin 1d,
curromycin A and B 1e–f and KSM2690 1g–h and lajollamycin,
is of interest for its structural uniqueness and wide biological
activity1 (lajollamycin, for instance, exhibits potent antibacterial
activity against both drug-sensitive and -resistant Gram-positive
strains (MIC values < 20 μg mL−1),2 and 16-methyloxazolomy-
cin is active against B. subtilis at 5 μg mL−1 although this drops
to 50 μg mL−1 for KSM-2690B1) and significant progress in the
understanding of their biosynthesis has recently been reported.3,4
Although several total syntheses have now been achieved,5–7 the
highly functionalised-γ-lactam ring portion remains a particular
synthetic challenge. Access to the lactam–lactone spirocyclic
system has been the focus of some attention, and Mohapatra has
recently reported the use of auxiliary-controlled aldol chemistry
to construct the substituted pyroglutamate ring,8 while Taylor
has reported an approach based upon modification of proline.9
However, the synthesis of an intermediate suitable for incorpor-
ation of the middle fragment, by including the necessary chiral
quaternary centre at C-3 (oxazolomycin numbering), has proved
to be much more challenging, and in addition to our own work
in this area,10–12 the groups of Donohoe,13 Pattenden,14 Ohfune
and Soloshonok15 and Wang16 have all made important contri-
butions. We report here the successful achievement of an intra-
molecular aldol reaction on a functionalised oxazolidine, leading
directly to highly functionalised pyroglutamates with three con-
tiguous chiral centres and which are simpler mimics of the more
complex spirolactam–lactone system of the oxazolomycins.
Seebach’s concept of self-reproduction of stereocentres is used
as a key element to control the chemo- and stereo-selectivity of
the reaction.17 Other groups have also found merit in a similar
strategy based on aldol closure for the synthesis of salinos-
poramide A and its analogues,18,19 by direct adaptation of our
original report.12
At the outset, we required an entry to γ-alkoxymalonamides, a
compound class for which there is a surprising paucity of effec-
tive synthetic methodology. Our first attempt to access these
systems is outlined in Scheme 1; although conversion of glycolic
acid 2 to the corresponding silyl ethers 3 could be achieved
in good yield under standard conditions (RCl, imidazole, DMF)
(3, R = TBDMS (98%), TBDPS (80%), TIPS (64%)), their
further conversion according to Scheme 1 proved to be proble-
matic. Thus, apart from the reaction of the PMB protected
material (3, R = PMB), which was quantitative, the silyl ethers 3
did not give good conversion to the corresponding β-ketoesters 4
(yield, R = TBDMS (19%), TBDPS (0%), TIPS (29%)). Depro-
tection of both the PMB and TIPS systems was possible (yield
5, R = PMB (65%), TIPS (11%)), but only the TIPS acid 5
allowed access to the desired oxazolidine 6 (R = TIPS) by
DCC–DMAP coupling, and even then in poor yield (11%).
Although this multi-step sequence offered considerable flexi-
bility, given the low yields, an alternative was required.
Of recent approaches leading to functionalised β-ketoesters,20
methods for the synthesis of functional acetoacetates by con-
densation of 3,3-dimethylacryloyl chloride21 or methoxyacetyl
chloride22 with Meldrum’s acid 7a were of particular interest;
Meldrum’s acid is enjoying a renaissance of interest as a nucleo-
phile.23 We found that acylation of methyl Meldrum’s acid 7b
using 3,3-dimethylacryloyl and methoxyacetyl chlorides pro-
ceeded efficiently (Scheme 2), but we could not open either of
the products 8a or 8b with alcohols under a variety of con-
ditions. Suspecting that the bulk of the α-methyl substituent was
likely to be the source of the problem, we examined acylation of
Meldrum’s acid 7a using methoxyacetyl chloride. We found in
this case that acylation also efficiently generated the desired
Department of Chemistry, Chemistry Research Laboratory, The
University of Oxford, 12 Mansfield Road, Oxford, OX1 3TA, UK.
E-mail: mark.moloney@chem.ox.ac.uk
†Electronic supplementary information (ESI) available. CCDC 860455
and 860456. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2ob00042c
3472 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 1
Scheme 1
product 8c, predominantly as the enol form (keto–enol = 1 : 4),
and that this material was not fully stable to silica purification;
in keeping with the literature report,22 it was in fact readily con-
verted using tert-butanol in toluene at reflux to the tert-butyl
ester 9a in 79% yield. This material was easily deprotected using
trifluoroacetic acid, to give the corresponding acid 10a predomi-
nantly in the keto form (keto–enol = 4 : 1) in 89% yield. In order
to conveniently vary the alkoxy substituent, we used a literature
procedure for the direct substitution of alkoxides on the enolate
derived from ethyl 4-chloroacetoacetate (Scheme 2).24–27 By this
means, we were able to access acetoacetates 11a,b in good yields
(in the case of 11a, concomitant transesterification was
obtained), which were found to exist as a mixture of keto–enol
tautomers, and these were readily saponified (NaOH, H2O) to
the β-ketoacids 10b,c in good yield (Scheme 2).
13a–c (Scheme 3). Reaction of these half-malonamides with
sodium methoxide efficiently gave the aldol products 14a–c and
15a–c, typically as a mixture of epimers, in favour of the diaster-
eomer in which the more bulky alkoxymethyl group was located
in the exo-position at C-6. The stereochemistry of the major pro-
ducts 15a–c was readily established by nOe analysis (Fig. 2).30
We believe that the chemoselectivity of this process results
from initial deprotonation at the malonamide residue, which
equilibrates with the C-5 position prior to ring closure, leading
to a bicyclic system in which the bulky substituent resides on
the less hindered convex exo-face;12,31,32 this is consistent with
the outcomes in related reports.18,19 These materials were easily
deprotected using the Corey–Reichard protocol,33 to give the
inseparable pyroglutaminols 16 and 17a in 86% yield (ratio
1 : 6). Deprotection of 15b gave the desired lactam 17b along
with 18 which could not be separated (3 : 2 ratio) in a yield of
82%. In the case of 15c, concomitant tert-butyl ether deprotec-
tion was also observed, and the resulting diol underwent intra-
molecular lactonisation to give lactone–lactam 19, whose
structure was assigned by careful HMBC analysis; this
These acids were used to acylate oxazolidine 12, prepared
from L-serine methyl ester hydrochloride and pivaldehyde
according to the Seebach protocol,28,29 with EDAC (1-(3-
dimethylaminopropyl)-3-ethylcarbodiimide methiodide) in the
presence of a catalytic amount of DMAP, to the cis-oxazolidines
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3473

View Article Online
Scheme 2
Scheme 3
compound is of interest since it comprised the core structure of
neooxazolomycin.34 The major diastereomers 15a–c from this
sequence possess the correct relative stereochemistry for oxazo-
lomycin 1, and this outcome effectively demonstrated that the
presence of the additional alkoxy functionality in malonamides
13a–c, even with inclusion of quite bulky groups such as the
tert-butyl group, did not impede the desired aldol reaction.
However, critical to the further development of this strategy was
the incorporation of the C-2 methyl group (Fig. 1) which is
invariably present in all known members of the oxazolomycin
family.
On the basis of earlier work,11,32 we expected that deprotona-
tion of acetoacetates 9a, 11a, and 11b with potassium tert-
butoxide followed by reaction with methyl iodide would permit
3474 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 2
methylation to give methylacetoacetates 9b, 11c, and 11d, and
stereochemistry in these systems. On this basis, the isomers 22a
(minor) with C(7)Me at 1.41 and C(7)H at 2.43, and 23a
(major), with C(7)Me at 1.06 and C(7)H at 3.50, were assigned
as shown in Scheme 3. This approach provides direct access to
polysubstituted pyroglutamates related to the right hand side of
oxazolomycin. That the additional methyl substituent has such a
pronounced effect on the diastereoselectivity of the ring closing
process (at least compared to the outcome for the desmethyl
series 13) is a reflection of the congested nature of the bicyclic
system, which possess three contiguous chiral centres, two
of which are quaternary. We did not examine equilibration
phenomena of the diastereomeric products, and whether the
product distribution is a result of kinetic and thermodynamic
control is therefore not clear.
Having demonstrated the feasibility of conducting efficient
aldol reactions according to Fig. 1, it was of interest to improve
this approach, and on the basis of recently published work,35–37
we expected that direct opening of acylated Meldrum’s acids 8c,
d with oxazolidine 12 should be possible; if so, this would
remove two steps from the synthesis. In the event, when the two
starting materials 8c or 8d and 12 were simply heated at 60 °C
in acetonitrile for 2 h, the desired amides 13a,b were readily
formed and in good yield. In keeping with earlier observations,
aldol cyclisation of the ketoamides 13a,b proceeded smoothly
at r.t. in the presence of 1.5 equiv. of sodium methoxide in
methanol, giving the products as mixtures of diastereoisomers
14a,b and 15a,b (14a : 15a = 1 : 4 and 14b : 15b = 1 : 5), the
this strategy in fact worked very efficiently (Scheme 2). De-
protection under appropriate conditions gave the corresponding
acids 10d–f. Repetition of the sequence outlined above
(Scheme 3) gave the corresponding oxazolidines 20a–c as a
mixture diastereoisomers, albeit in lower yield in two cases than
the desmethyl series, probably as a result of the additional steric
bulk in this system, but these products were again readily
cyclised by aldol reaction using sodium methoxide in dry metha-
nol; this reaction proceeded most efficiently for the benzyl 20b
and tert-butyl 20c systems. In all cases, more than one diastereo-
mer was formed, not all of which were separable, for which the
stereochemistry of most was readily assigned by nOe analysis
(Fig. 2). However, in the case of the cyclisation of oxazolidine
20a, the several product diastereomers 21a, 22a and 23a which
were formed could only be assigned by reference to chemical
1
shift patterns. Thus, it was noted that in the H NMR spectrum
of diastereomers 21b, 21c and 23b, all possessing the 7S-stereo-
chemistry as established by nOe analysis, that the C(7) methine
protons had relatively higher chemical shifts (both δ 3.2) while
the C(7) methyl groups had relatively lower chemical shifts
(both δ 1.1). On the other hand, diastereomers 22b and 22c with
the 7R-stereochemistry, exhibited the C(7) methine proton with a
relatively low chemical shift (δ 2.4) while the C(7) methyl group
had a relatively higher chemical shift (δ 1.4); this outcome corre-
sponds with the less highly functionalised systems examined
earlier,12 and provides a useful, if indirect, method for assigning
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3475

View Article Online
Scheme 4
major ones 15a,b having the correct relative stereochemistry for
oxazolomycin (Scheme 3) as determined on the basis of NOE
experiments (Fig. 2).
delivery from the less hindered face. Indeed, treatment of 24a,b
with 2 equiv KHMDS followed by t-BuOH gave nearly clean
21a,b in 45 and 46% isolated yield respectively (Scheme 4), and
identical to the material prepared by direct aldol reaction
(Scheme 3). This represents a significant improvement over the
direct cyclisation, and not least in significantly enhanced diaster-
Having successfully prepared the desired lactams 15a,b, we
investigated the introduction of a methyl group by double depro-
tonation and diastereoselective enolate alkylation. Initial reaction
of 15a with 2 equiv. LDA followed by 1 equiv. of iodomethane
which gave the desired C-methylated products 21a and 24a with
high diastereoselectivity, but only in moderate yield, and no
competing O-methylation was observed. However, by using
an excess of the alkylating reagent (3 equiv.), the yield of the
C-alkylated products was significantly improved (60–80%) and
no O-alkylation was observed (Scheme 3). This process could be
easily replicated with the benzyl substrate 15b, although in both
these reactions, 15–30% of unreacted starting materials 15a,b
was recovered. The stereochemical outcome of this process was
determined on the basis of NOE experiments (Fig. 2) and
showed that the electrophilic attack occurs predominantly from
the less hindered convex side of the bicyclic system, as might be
expected; we have previously shown that alkylation, nucleophilic
additions and reductions in these bicyclic systems is strongly
influenced by the concave nature of the substrate.38–40 Although
the configuration of the major products 24a,b is incorrect for the
natural product, the high diastereoselectivity of the enolate alky-
lation suggested that it may be possible to invert the configur-
ation through double deprotonation and subsequent proton
1
eoselectivity. In the H NMR spectrum of diastereomers 21a,b
and 24a,b, NMR patterns were consistent with those described
above; thus, for 21a,b, both possessing the 7S-stereochemistry,
the C(7) methine protons had relatively higher chemical shifts
(both δ 3.2) while the C(7) methyl groups had relatively lower
chemical shifts (δ 1.1–1.2). On the other hand, diastereomers
24a and 24b with the 7R-stereochemistry, exhibited the C(7)
methine proton with a lower chemical shift (δ 2.8) while the C
(7) methyl group had a relatively higher chemical shift (δ 1.4).
With 21a,b in hand, we proceeded with their deprotection
using Corey–Reichard aminal cleavage33 to give the correspond-
ing acids 25a,b in good yield (80 and 88% respectively) along
with minor amounts of acetals 26a,b (up to 20%) (Scheme 4).
The stereochemical assignments were confirmed by single-
crystal X-ray diffraction of the intermediate 25a,41 which indi-
cated a conformation in which the relatively more bulky C(4)
Me, C(3)CH2OMe and C(2)CH2OH groups occupy pseudoequa-
torial positions (Fig. 3). That selective manipulation of 21a,b
was possible was demonstrated by ester hydrolysis (LiOH)
which gave pyroglutamates 27a,b, with the opposite absolute
3476 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 3 X-ray structures for 25a and 28a (thermal ellipsoids shown at 25% probability level).
Table 1 Cheminformatica and bioassayb of compounds 14–19 and 21–28 against S. aureus and E. coli
Bioactivityb
S. aureus
E. coli
Compound
LogPa
PSA
%PSA
Zone size/mm
Relative potencyc
Zone size/mm
Relative potencyc
1a
1b
1c
1d
1e
1f
1g
1h
1.61
1.61
1.61
2.03
1.72
2.10
2.03
2.03
2.66
0.24
1.97
1.97
0.40
2.14
1.30
−2.40
—
−2.52
0.79
2.69
1.84
0.79
2.51
2.51
1.84
2.51
2.51
0.95
2.69
−1.74
−0.01
−1.91
0.45
171.7
171.7
171.7
171.7
180.1
171.7
171.7
171.7
191.5
85.3
85.3
85.3
85.3
85.3
85.3
105
17.9
17.9
17.9
17.1
16.6
16.5
17.1
17.1
18.4
17.6
14.4
14.4
17.6
14.4
14.6
30.3
—
39.2
16.6
13.8
14.0
16.6
13.8
13.8
14.0
13.8
13.8
16.7
13.8
27.9
21.8
34.1
20.4
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
3.8
—
—
0.91
—
—
—
—
—
—
1.7
—
2.6
—
1.7
3.8
—
—
—
—
—
—
0.88
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Lajollamycin
14a/15a
14b
14b
15a
15b
15c
16a/17a
17b+18
19
21a
21b
21c
22a/23a
22b
22b
22c
23b
23b
24a
—
Inactive
15
—
12
—
0.038
0.087
0.095
—
0.08
—
—
—
—
—
—
0.035
—
0.087
0.095
0.035
0.055
0.070
—
—
—
0.06
0.06
0.07
18
Inactivee
Inactive
13
14e
Inactive
19
Inactive
Inactive
Inactive
Inactive
Inactive
Inactive
11d
Inactive
Inactive
Inactive
Inactive
Inactive
12
—
—
85.3
85.3
85.3
85.3
85.3
85.3
85.3
85.3
85.3
85.3
85.3
105.1
105.1
116.1
111.2
12d
Inactive
13
Inactive
18
Inactive
14e
11d
12d
15
15
Inactivee
Inactive
Inactive
Inactive
11
12e
Inactive
Inactive
11
18
20.5
18
24b
25a
25b
27a
Inactive
13
28b
a LogP, PSA and MSA calculated using Marvin; %PSA = (PSA/MSA) × 100.48 b Hole plate bioassay using 100 μL of 4 mg mL−1 solution (6 : 4
DMSO–H2O) unless otherwise indicated.49 c Expressed as zone size per mg ml−1, relative to cephalosporin C standard. d Halo only. e Concentration
2 mg mL−1 (6 : 4 DMSO–H2O).
but the correct relative configuration for the right-hand fragment
of oxazolomycin, or alternatively acetylation of 25a,b gave
esters 28a,b in excellent yield. Confirmation of the structure of
lactam 28b was confirmed by single-crystal X-ray diffraction
(Fig. 3).41 Despite the hindered nature of 21a,b and 25a,b,
noteworthy is the efficiency with which both of these processes
proceeded.
Bioassay of these compound libraries against S. aureus and
E. coli using the hole-plate method gave the activities indicated
in Table 1; although not giving quantitative MIC values, these
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3477

View Article Online
in vivo assays give outcomes enabling rapid assessment of
the effect of structural variations of the fragment modification.
The value of such forward chemical genetic (phenotypic)
screens in these cell based systems is that hits are automatically
selected for their combination of activity and cell-permeability
properties, and without any bias or pre-selection of the most
optimal target of a pathway, although it of course gives no in-
formation about mode of action or target identity.42,43 Of
significance is that only a few compounds in this library showed
antibacterial activity against S. aureus, although more were
active against E. coli. Alkoxy derivatives 14b, 15b, 21c, 22b,c,
25b and 28b were the only ones with activity against S. aureus,
and of the compounds with the correct relative configuration
for oxazolomycin, only 21c, 25b and 28b were active. All the
active compounds have cLogP values in the range 1.8–2.5, and
%PSA values close to 14, and with a potency of some 1–4%
relative to the cephalosporin reference standard. However, 25b
and 28b are clearly different, being very much more polar, with
cLogP values in the range −0.01 to 0.5, and %PSA values
close to 21. For E. coli, different diastereomers exhibited differ-
ent activities, and in some cases these were very marked,
especially for the benzyloxymethyl systems 21b, 22b, 23b
and 24b; actives had cLogP values in the range 1.8–2.7 and
%PSA values in the range 13.8–17.6, although their potencies
were also substantially lower than the cephalosporin C control.
However, compounds 25a,b and 28b are clearly different, being
very much more polar and generally significantly more active,
with cLogP values in the range −1.9 to 0.5, and %PSAvalues as
high as 34. The methoxymethyl compounds 14a, 15a, 21a, 22a
and 23a, which might be considered to mimic structurally most
closely the oxazolomycin right hand side, were all inactive or
only weakly active, with cLogP values in the range 0.24–0.79
and %PSA values of 16.6–17.6, that is, outside the range
of identified actives above, and we have recently reported that
other pyroglutamate systems are similarly inactive, unless conju-
gated to a longer middle fragment mimic which attenuates
cLogP.44,45 Of interest is that the most active compounds had
cheminformatic values which correlated with the corresponding
values for oxazolomycins 1a–h, which have cLogP values in
the range 1.7–2.1 and %PSA values in the range 16.5–17.9
(Table 1). The better activity of the benzyloxymethyl systems
14b, 15b, 21b, 22b and 23b, which also exhibit higher cLogP
values than the methoxymethyl and tert-butyloxymethyl
systems, is probably as a result of improved cell membrane per-
meability, and it would seem likely that introduction of signifi-
cant bioactivity into compounds like 21a with the correct relative
configuration for oxazolomycin will require adjustment of their
cLogP values at least. That this might be feasible is illustrated
by the fact that the highly polar deprotected systems 16–18 were
inactive, but the less highly polar 25b and 28a,b, all of which
possess the correct relative stereochemistry for oxazolomycin,
exhibit strong bioactivity against E. coli. The polar surface
area parameter (PSA), which correlates the presence of polar
atoms with membrane permeability and therefore gives an indi-
cation of drug transport properties,46 has been reported to have
an optimal value of 70 < PSA < 120 Å2 for a non-CNS orally
absorbable drug,47 and of interest is that the active compounds
14, 15, 21, 22, 23, 25 and 28 had values in the range 85–116.
Two-fold dilution (to 2 mg mL−1) of the most active compounds
14b, 22b, and 23b still gave activity for E. coli, although not for
S. aureus. However, it is important to note that the compounds
reported herein belong to the enantiomeric series relative to
the natural product, and therefore conclusions about the pharma-
cophoric grouping responsible for antibacterial activity cannot be
made.
The efficiency of the biomimetic aldol ring closures disclosed
herein is remarkable,3 and although this synthesis has resulted
in the construction of the opposite enantiomeric series relative
to oxazolomycin, the correct absolute configuration would
be available using D-serine as a precursor. We have recently
demonstrated that the intrinsic antibacterial activity of simple
pyroglutamates50,51 and tetramates is low,52 but that homologa-
tion to longer chain side-units restores activity.44,45,53 We have
also shown that a change as small as the introduction of a methyl
substituent, achieved using threonine as the starting amino
acid, improves bioactivity in tetramates,11 and it would there-
fore appear from the work described here that bioactivity may
be critically dependent on appropriate functionality around the
pyroglutamate ring periphery. Moreover, the absence of
the β-lactone unit does not fully abolish activity, and work to
establish the biological effect of its introduction is currently
underway. Given the difficulties in the development of antibac-
terials54 and the recently discussed urgent need for the identifi-
cation of novel strategies for expanding the antibacterial drug
development pipeline,55,56 particularly using fragment-based
approaches,57 and the renaissance of interest in natural product
lead structures,58–69 the oxazolomycins may offer inspiration
for unusual chiral fragments for antibacterial development,
in much the same way as we have shown for simpler
pyrrolidinones.70
Experimental
(2R,5R,6R)-1-Aza-2-(tert-butyl)-6-hydroxy-6-(methoxymethyl)-
5-methoxycarbonyl-3-oxa-8-oxobicyclo[3.3.0]octane 15a
The ketoamide 13a (3.02 g, 10 mmol) was dissolved in MeOH
(100 mL) and to the solution was added NaOMe (0.81 g,
15 mmol). The mixture was stirred overnight at r.t. under nitro-
gen; the reaction was quenched with sat. aq. NH4Cl (200 mL)
and extracted with Et2O (4 × 100 mL). The combined organic
layers were dried (MgSO4) and the solvent was evaporated under
reduced pressure to give 2.72 g of 4 : 1 mixture of 15a and 14a.
Column chromatography on silica gel with petrol–Et2O 1 : 1
(increasing polarity to 1 : 2) allowed partial separation of the dia-
stereoisomers and gave 1.44 g of 15a as white crystalline solid,
and 1.3 g mixed fraction which was re-columned with the next
batch. Rf 0.30 (Et2O); [α]2D5 = +37.4° (c = 1, MeOH); m.
p. 110–112 °C; νmax (film) 1733, 1699; δH (400 MHz, CDCl3)
0.88 (9H, s, t-Bu), 2.41 (1H, d, J = 16, C(7)HH), 3.19 (1H, d,
J = 16 Hz, C(7)HH), 3.35 (5H, m, CH2OCH3), 3.46 (1H,
s, OH), 3.79 (3H, s, COOCH3), 4.21 (1H, d, J = 8 Hz, C(4)HH),
4.53 (1H, d, J = 8 Hz, C(4)HH), 4.92 (1H, s, C(2)H); δC
(100 MHz, CDCl3) 24.9, 36.3, 45.6, 52.6, 59.4, 67.8, 74.2, 79.1,
96.2, 171.4, 175.6; m/z (ESI+) 324 ([M + Na]+, 25%) 625
([2M + Na]+, 100%); HRMS: found 324.1407 for C14H23NNaO6
[M + Na]+, calculated 324.1418.
3478 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
(2R,5R,6S)-1-Aza-2-(tert-butyl)-6-hydroxy-6-(benzyloxymethyl)-
5-methoxycarbonyl-3-oxa-8-oxobicyclo[3.3.0]octane 14b and
(2R,5R,6R)-1-aza-2-(tert-butyl)-6-hydroxy-6-(benzyloxymethyl)-
5-methoxycarbonyl-3-oxa-8-oxobicyclo[3.3.0]octane 15b
COOCH3), 4.13 (1H, d, J = 8.8 Hz, C(4)HH), 4.54 (1H, d, J =
8.8 Hz, C(4)HH) and 4.88 (1H, s, C(2)H); δC (100 MHz;
CDCl3) 24.9 (C(CH3)3), 27.2 (C(CH3)3), 36.3 (C(CH3)3), 45.5
(C(7)), 52.5 (COOCH3), 63.7 (CH2OC(CH3)3), 67.7 (C(4)), 74.0
(C(6)), 79.2 (C(5)), 95.82 (C(2)), 171.4 and 175.5 (2 × CvO);
The ketoamide 13b (2.70 g, 7.15 mmol) was dissolved in
MeOH (75 mL) and to the solution was added NaOMe (0.58 g,
10.72 mmol). The mixture was stirred overnight at r.t. under
nitrogen. The reaction was quenched with sat. aq. NH4Cl
(150 mL) and extracted with Et2O (3 × 75 mL). The combined
organic layers were dried (MgSO4) and the solvent was evapor-
ated under reduced pressure to give 2.51 g of crude 5 : 1 mixture
of 15b and 14b. Column chromatography on silica gel with 3 : 1
petrol–EtOAc (increasing polarity to 2 : 1) allowed partial separ-
ation of the diastereoisomers and gave 1.40 g of 15b, 0.65 g
mixed fraction and 0.15 g of 14b. Rf 0.20 (petrol–EtOAc 2 : 1);
m.p. 114–116 °C; νmax (film) 3367, 2951, 1747, 1698; δH
(400 MHz, CDCl3) 0.88 (9H, s, t-Bu), 2.41 (1H, d, J = 16 Hz, C
(7)HH), 3.19 (1H, d, J = 16 Hz, C(7)HH), 3.34 (1H, s, OH),
3.43 (2H, s, CH2OBn), 3.67 (3H, s, COOCH3), 4.21 (1H, d, J =
8 Hz, C(4)HH), 4.50 (2H, AB quartet, J = 12 Hz, OCH2Ph),
4.59 (1H, d, J = 8 Hz, C(4)HH), 4.91 (1H, s, C(2)H), 7.27–7.40
(5H, m, Ph); δC (100 MHz, CDCl3) 24.9, 36.3, 45.6, 52.6, 67.8,
71.8, 73.8, 79.2, 96.1, 127.9, 128.2, 128.6, 136.7, 171.3, 175.4;
HRMS: found 400.1716 for C20H27NNaO6+ [M + Na]+, requires
400.1731. Data for 14b: Rf 0.15 (petrol–EtOAc 2 : 1)
m.p. 104–106 °C; νmax (cm−1, neat): 3365, 2960, 1731, 1692;
δH (400 MHz, CDCl3) 0.89 (9H, s, t-Bu), 2.37 (1H, d, J = 16
Hz, C(7)HH), 3.22 (1H, s, OH), 3.37 (1H, d, J = 16 Hz, C(7)
HH), 3.38 (1H, d, J = 8 Hz, CH2OBn), 3.59 (1H, d, J = 8 Hz,
CH2OBn), 3.73 (1H, d, J = 8 Hz, CH2O), 3.81 (3H, s,
COOCH3), 4.56 (2H, AB quartet, J = 12 Hz, OCH2Ph), 4.73
(1H, d, J = 8 Hz, CH2O), 4.80 (1H, d, J = 8 Hz, CH2O),
7.27–7.40 (5H, m, Ph); δC (100 MHz, CDCl3) 25.0, 36.1, 45.2,
52.5, 69.67, 72.4, 73.77, 79.3, 95.5, 128.1, 128.5, 128.7, 136.2,
153.7, 171.0, 173.3; HRMS: found 400.1725 for
C20H27NNaO6+ [M + Na]+, requires 400.1731.
m/z (ESI+) 709 ([2M
+
Na]+, 100%); HRMS (ESI+)
C17H29NNaO6+ requires 366.1887, found 366.1884; [α]2D0 + 31.2
(c = 0.88 in MeOH). Data for 14c: Rf 0.25 (2 : 1 petrol–EtOAc);
δH (400 MHz; CDCl3) 0.90 (9H, s, C(CH3)3), 1.22 (9H, s,
C(CH3)3), 2.37 (1H, d, J = 16.2, C(7)HH), 3.23 (1H, br, s, OH),
3.29 (1H, d, J = 9.4 Hz, CHHOC(CH3)3), 3.41 (1H, d, J = 16.2
Hz, C(7)HH), 3.53 (1H, d, J = 9.4 Hz, CHHOC(CH3)3), 3.81
(3H, s, COOCH3), 3.84 (1H, d, J = 9.7 Hz, C(4)HH), 4.69 (1H,
d, J = 9.7 Hz, C(4)HH), 4.83 (1H, s, C(2)H).
(2R,3R)-Methyl 3-hydroxy-2-(hydroxymethyl)-3-
(methoxymethyl)-5-oxopyrrolidine-2-carboxylate 17a and
(2R,3S)-methyl 3-hydroxy-2-(hydroxymethyl)-3-
(methoxymethyl)-5-oxopyrrolidine-2-carboxylate 16
To a solution of 15a (0.090 g, 0.30 mmol) in 2% HCl in trifluor-
oethanol (3 mL) was added 1,3-propanedithiol (0.030 mL,
0.30 mmol). The mixture was left to stir overnight and then the
solvent removed in vacuo. The crude product was purified by
column chromatography on silica gel (increasing polarity from
petrol (40–60)–ethyl acetate 1 : 1 to ethyl acetate–methanol
4 : 1), affording 16 and 17a in 6 : 1 dr as a colourless oil (0.06 g,
86%); Rf = 0.52 (EtOAc–MeOH, 4 : 1); νmax (film) 3662, 2961,
1748; Data for 17a: δH (400 MHz; CDCl3) 2.36 (1H, d, J =
17.2 Hz, C(5)H), 2.69 (1H, d, J = 17.2 Hz, C(5)H) 3.30 (3H, s,
CH2OCH3), 3.33 (2H, m, CH2OCH3), 3.76 (3H, s, CO2CH3),
3.90 (1H, d, J = 11.4 Hz, CH2OH), 4.10 (1H, d, J = 11.4 Hz,
CH2OH); δC (100 MHz; MeOD) 42.0 (C(5)), 51.9 (CO2CH3),
58.4 (CH2OH), 64.2 (CH2OCH3), 72.4 (C(3)), 74.8 (C(2)), 79.1
(CH2OCH3), 171.7 (CO2Me), 177.7 (C(5)); Data for 16: δH
(400 MHz; MeOD) 2.30 (1H, d, J = 17.2 Hz, C(5)H), 2.79 (1H,
d, J = 17.2 Hz, C(5)H) 3.31 (3H, s, CH2OCH3), 3.42 (2H, m,
CH2OCH3), 3.58 (1H, d, J = 9.6 Hz, CH2OH), 3.70 (1H, d, J =
9.6 Hz, CH2OH), 3.77 (3H, s, CO2CH3); δC (100 MHz; CDCl3)
42.0 (C(5)), 51.9 (CO2CH3), 58.4 (CH2OH), 64.2 (CH2OCH3),
72.4 (C(3)), 74.8 (C(2)), 79.1 (CH2OCH3), 171.7 (CO2Me),
177.7 (C(5)); m/z (ESI+) 234 ([M + H]+, 80%) 467 ([2M + H]+,
(2R,5R,6R)-1-Aza-2-(tert-butyl)-6-hydroxy-6-(tert-
butoxymethyl)-5-methoxycarbonyl-3-oxa-8-oxobicyclo[3.3.0]
octane 15c and (2R,5R,6S)-1-aza-2-(tert-butyl)-6-hydroxy-6-(tert-
butoxymethyl)-5-methoxycarbonyl-3-oxa-8-oxobicyclo[3.3.0]
octane 14c
+
100%); HRMS (ESI+) Found 256.0794 for C9H15NNaO6
[M + Na]+, requires 256.0792.
To a solution of oxazolidine 13c (0.46 g, 1.34 mmol) in dry
methanol (10 mL) was added NaOMe (0.073 g, 1.35 mmol) and
solution was stirred for 24 h at r.t. and then partitioned between
diethyl ether (20 mL) and aq. NH4Cl (20 mL). The diethyl ether
layer was washed with brine (20 mL), dried over MgSO4,
filtered and the solvent was evaporated in vacuo to give crude
product as a light brown sticky oil (0.35 g), which was purified
by flash column chromatography (3 : 1 petrol–EtOAc) to give
a mixture of epimeric alcohols 15c (0.25 g, 55%) and 14c
(0.062 g, 13%) as white solid and colourless thick oil, respect-
ively. Data for 15c: Rf 0.40 (2 : 1 petrol–EtOAc); νmax (film)/
cm−1 3253, 2976, 1682, 1366, 1192, 1086; δH (400 MHz;
CDCl3) 0.86 (9H, s, C(CH3)3), 1.14 (9H, s, C(CH3)3), 2.39 (1H,
d, J = 15.6 Hz, C(7)HH), 3.16 (1H, d, J = 15.6 Hz, C(7)HH),
3.31 (2H, AB quartet, J = 9.0 Hz, CH2OC(CH3)3), 3.75 (3H, s,
(2R,3R)-Methyl 3-hydroxy-2-(hydroxymethyl)-3-
(benzyloxymethyl)-5-oxopyrrolidine-2-carboxylate 17b and
(3R,4R)-methyl 3-amino-4-benzyloxymethyl-4-hydroxy-6-
oxotetrahydro-2H-pyran-3-carboxylate 18
To a solution of alcohol 15b (0.13 g, 0.33 mmol) in 2% HCl in
trifluoroethanol (2 mL) was added 1,3-propanedithiol (0.036 g,
0.33 mmol) and mixture was stirred for 20 h at r.t. The solvent
was then evaporated in vacuo to give mixture of 17b and 18
(0.088 g, 85%) in a ratio of 3 : 2, respectively, as a pale brown
thick oil: Rf 0.72 (4 : 1 EtOAc–MeOH with a drop of AcOH);
νmax (film)/cm−1 2925, 2361, 2329, 1736, 1682, 1313, 1235,
1461, 1377, 722;); m/z (ESI+) 332 ([M + Na]+, 99%); HRMS
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3479

View Article Online
(ESI+) C15H19NNaO6+ requires 332.1105, found 332.1106; Data
for 17b: δH (400 MHz; D2O) 2.50–2.60 (2H, AB quartet, J =
17.4 Hz, C(4)HH), 3.38–3.41 (2H, AB quartet, J = 10.1 Hz,
C(3)CHH), 3.52 (3H, s, COOCH3), 3.82 (1H, d, J = 11.7 Hz,
CHHOH), 3.98 (1H, d, J = 11.7 Hz, CHHOH), 4.37–4.41 (2H,
AB quartet, J = 12.0 Hz, OCHHPh) and 7.24–7.33 (ArH); δC
(100 MHz; D2O) 41.9 (C(4)), 53.4 (COOCH3), 63.2 (C(2)CH2),
72.5 (C(3)CH2), 73.6 (OCH2Ph), 79.5 (C(3)), 128.6 (para-CH),
128.7 and 129.0 (ortho- and meta-CH), 137.4 (ipso-CH), 172.5
and 177.5 (2 × CvO); Data for 18: δH (400 MHz; D2O) 2.26
(1H, d, J = 17.7 Hz, C(4)HH), 2.73 (1H, d, J = 17.7 Hz,
C(4)HH), 3.38–3.41 (2H, AB quartet, J = 10.1 Hz, C(3)CHH),
3.69 (3H, s, COOCH3), 3.64–3.75 (2H, m, C(7)HH), 4.34–4.41
(2H, m, OCHHPh); δC (100 MHz; D2O) 42.2 (C(4)), 53.4
(COOCH3), 59.2 (C(2)), 64.0 (C(7)), 72.5 (C(3)CH2), 73.3
(C(3)), 73.6 (OCH2Ph), 128.6 (para-CH), 128.7 and 129.0
(ortho- and meta-CH), 137.4 (ipso-CH), 171.5 and 177.5
(2 × CvO).
(3 mL). The reaction mixture was stirred at 0 °C for 15 min and
then at r.t. for 5 h. The crude reaction mixture was partitioned
between DCM (20 mL) and aq. NH4Cl (20 mL) and the organic
layer was dried over MgSO4, filtered and concentrated to give a
brown oil (0.18 g). Purification by column chromatography (3 : 1
petrol–EtOAc) gave cis-oxazolidine 20b, (0.044 g, 18%) as
a pale yellow thick oil: Rf 0.29 (3 : 1 petrol–EtOAc); δH
(400 MHz; CDCl3) 0.89 (9H, s, C(CH3)3), 1.34 (3H, d, CH3),
3.71–3.80 (4H, m, COOCH3 and C(2)′H), 4.02–4.10 and 4.30
and 4.50–4.65 (4H, m, C(4)HH and C(4)′H2), 4.28 (2H, s, C(6)′
H2), 5.32 (1H, s, C(2)H), 7.27–7.34 (5H, m, ArH); δC
(100 MHz; CDCl3) 12.8 (CH3), 25.7 (C(CH3)3), 37.7 (C(CH3)3),
52.8 (COOCH3), 67.9 (C(4)), 72.9 (C(4)′), 73.4 (C(6)′), 96.9
(C(2)), 127.7 (para-CH), 128.3, 128.5 (ortho- and meta-CH),
136.8 (ipso-CH), 169.7, 171.8 and 202.4 (3 × CvO); m/z (ESI+)
+
805 ([2M + Na]+, 99%); HRMS (ESI+) C21H29NNaO6 requires
414.1887 found 414.1887.
(2R,5S)-Methyl 1-(4′-(tert-butoxy)-2′-methyl-3′-oxobutanoyl)-2-
(tert-butyl)oxazolidine-5-carboxylate 20c
(1R, 5R)-2-Aza-5-hydroxy-1-hydroxymethyl-3,8-dioxo-7-
oxabicyclo[3.3.0]octane 19
To a stirred solution of oxazolidine 1229 (0.26 g, 1.41 mmol),
DMAP (0.015 g, 0.12 mmol), EDAC (0.32 g, 1.69 mmol) in
DCM (7 mL) at 0 °C was added a solution of 4-(tert-butoxy)-2-
methyl-3-oxobutanoic acid 10f (0.32 g, 1.69 mmol) in DCM
(3 mL). The reaction mixture was stirred at 0 °C for 15 min and
then at r.t. for 5 h. The crude reaction mixture was partitioned
between DCM (30 mL) and aq. NH4Cl (30 mL) and the organic
layer was dried over MgSO4, filtered and the solvent was evapor-
ated in vacuo to give brown oil (0.43 g). Purification by column
chromatography (3 : 1 petrol–EtOAc) gave cis-oxazlidine 20c
(0.095 g, 16%) as a pale yellow thick oil: Rf 0.4 (2 : 1 petrol–
EtOAc); νmax (film)/cm−1 2975, 2361, 1745, 1667, 1366, 1192,
1104, 894; δH (400 MHz; CDCl3) 0.94 (9H, s, C(CH3)3), 1.22
(9H, s, C(CH3)3), 1.25 (3H, m, CH3), 3.78 (3H, s, COOCH3),
3.87–4.10 (2H, m) and 4.10–4.30 (2H, m) and 4.40–4.62 (2H,
m, C(5)H, C(2)′H, C(4)′H2 and C(4)HH), 5.35 (1H, s, C(2)H);
δC (100 MHz; CDCl3) 25.7 (C(CH3)3), 27.4 (C(CH3)3),
52.8 (COOCH3), 68.0 (C(4)), 97.0 (C(2)); m/z (ESI+) 737
To a solution of alcohol 15c (0.075 g, 0.22 mmol) in 2% HCl in
trifluoroethanol (1.5 mL) was added 1,3-propanedithiol (0.024 g,
0.22 mmol) and the mixture stirred for 20 h at r.t. The solvent
was then evaporated in vacuo to give 19 (0.034 g, 83%) as a
pale yellow thick oil: Rf 0.66 (4 : 1 EtOAc–MeOH with a drop of
AcOH); νmax (film)/cm−1 2924, 2854, 1779, 1692, 1461, 1377,
1273, 723; δH (500.3 MHz; acetone-d) 2.65–2.72 (2H, AB
quartet, J = 17.2 Hz, C(7)HH), 3.95 (2H, s, C(2)CH2), 4.38 (1H,
d, J = 9.5 Hz, C(5)HH), 4.42 (1H, d, J = 9.5 Hz, C(5)HH), 5.28
(br, s, C(6)OH); δC (125 MHz; acetone-d) 43.5 (C(7)), 59.9 (C
(2)CH2), 67.5 (C(2)), 76.7 (C(5)), 78.5 (C(6)), 173.9 (C(3)) and
175.3 (C(8)); m/z (ESI+) 210 ([M + Na]+, 70%); HRMS (ESI+)
C7H9NNaO5+ requires 210.0378, found 210.0380.
(2R,5R)-Methyl 2-(tert-butyl)-1-(4′-methoxy-2′-methyl-3′-
oxobutanoyl)oxazolidine-5-carboxylate 20a
+
([2M + Na]+, 100%); HRMS (ESI+) C18H31NNaO6 requires
To a solution of oxazolidine 1229 (1.19 g, 6.4 mmol) in DCM
(30 mL) was added DMAP (0.053 g, 0.43 mmol) and EDC·HCl
(1.22 g, 6.4 mmol). The mixture was stirred for 15 min at 0 °C
and a solution of acid 10d (0.927 g, 6.35 mmol) in DCM
(10 mL) was added dropwise. The mixture was stirred at 0 °C
for a further 15 min and then at r.t. for 3 h. The reaction mixture
was washed with sat. aq. NaH2PO4 (3 × 30 mL) and the aqueous
layer extracted with DCM (3 × 30 mL). The combined organic
extracts were washed with brine, dried over MgSO4, filtered and
the solvent removed in vacuo, affording 20a (1.34 g, 68%) as an
orange oil, which was used without further purification.
380.2044, found 380.2044.
(2R,5R,6R,7S)-Methyl 2-(tert-butyl)-6-hydroxy-6-
(methoxymethyl)-7-methyl-5-methoxycarbonyl-3-oxa-8-
oxobicyclo[3.3.0]octane 21a and (2R,5R)-Methyl 2-(tert-butyl)-6-
hydroxy-6-(methoxymethyl)-7-methyl-5-methoxycarbonyl-3-
oxa-8-oxobicyclo[3.3.0]octane 22a/23a
To a solution of crude 20a (1.34g, 4.24 mmol) in dry methanol
(30 mL) was added NaOMe (0.34 g, 6.34 mmol). The reaction
mixture was left to stir at r.t. for 24 h. The contents were then
poured into sat. aq. NH4Cl (50 mL) and extracted with diethyl
ether (3 × 30 mL). The combined organic extracts were washed
with brine, dried over MgSO4, filtered and the solvent removed
in vacuo, affording the crude product. The crude product was
purified via column chromatography on silica gel (petrol
(40–60)–EtOAc, 3 : 1), first affording 21a (0.079 g, 6%) as a
white solid and then 22a/23a (0.16 g, 12%) in a 2 : 1 dr as a
(2R,5S)-Methyl 1-(4′-(benzyloxy)-2′-methyl-3′-oxobutanoyl)-2-
(tert-butyl)oxazolidine-5-carboxylate 20b
To a stirred solution of oxazolidine 1229 (0.098 g, 0.53 mmol),
DMAP (0.0057 g, 0.046 mmol), EDAC (0.12 g, 0.63 mmol) in
DCM (7 mL) at 0 °C was added a solution of 4-(benzyloxy)-2-
methyl-3-oxobutanoic acid 11c (0.14 g, 0.63 mmol) in DCM
3480 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
colourless solid; Data for 21a: Rf 0.46 (Et2O); [α]2D5 = +16.4°
(c = 1, MeOH); m.p. 146–148 °C; νmax (cm−1, neat) 3345, 2966,
1736, 1696; δH (200 MHz; CDCl3) 0.88 (9H, s, C(CH3)3), 1.12
(3H, d, J = 7.2 Hz, CH3), 3.20 (1H, q, J = 7.2 Hz, C(7)H), 3.34
(1H, br s, C(6)(OH)), 3.35 (3H, s, CH2OCH3), 3.36 (2H, m,
CH2OCH3), 3.78 (3H, s, CO2CH3), 4.15 (1H, d, J = 8.5 Hz,
C(4)H), 4.55 (1H, d, J = 8.5 Hz, C(4)H), 4.90 (1H, s, C(2)H);
δH (400 MHz, C6D6) 1.16 (9H, s, t-Bu), 1.27 (3H, d, J = 8 Hz,
CH3CH), 2.74 (1H, s, OH), 2.86 (3H, s, CH3O), 2.95 (2H, AB
quartet, J = 8 Hz, CH2OCH3), 3.21 (1H, q, J = 8 Hz, CH3CH),
3.38 (3H, s, COOCH3), 4.29 (1H, d, J = 8 Hz, C(4)H), 4.82
(1H, d, J = 8 Hz, C(4)H), 5.24 (1H, s, C(2)H); δC (100 MHz;
CDCl3) 7.2 (CH3), 24.9 (C(CH3)3), 36.3 (C(CH3)3, 45.8 (C(7)),
52.6 (CO2CH3), 59.4 (CH2OCH3), 68.2 (C(4)), 73.6
(CH2OCH3), 78.0 (C(6)), 81.1 (C(5)), 96.2 (C(2)), 171.6
(CO2Me), 178.0 (C(8)O); δC (100 MHz, C6D6) 7.58 (CH3),
25.28 (C(CH3)3), 36.61 (C(CH3)3, 45.95 (C(7)), 51.85
(CO2CH3), 58.70 (CH2OCH3), 68.62 (C(4)), 73.51 (CH2OCH3),
78.32 (C(6)), 81.11 (C(5)), 96.61 (C(2)), 171.82 (CO2Me),
177.60 (C(8)O); m/z (ESI)+ 338 ([M + Na]+, 50%) 658 ([[M +
Na]+, 100%); HRMS (ESI+) Found 338.1574 for
13%) as a thick colourless oil and 23b (0.0047g, 10%) as thick
colourless oil. Data for 21b: Rf 0.30 (petrol–EtOAc 2 : 1);
m.p. 100–102 °C; νmax (film) 3406, 2952, 1746, 1703; δH
(400 MHz, CDCl3) 0.88 (9H, s, t-Bu), 1.11 (3H, d, J = 7 Hz,
CH3CH), 2.97 (1H, s, OH), 3.22 (1H, q, J = 7 Hz, CH3CH),
3.45 (2H, AB quartet, J = 10 Hz, CH2OBn), 3.66 (3H, s,
COOCH3), 4.16 (1H, d, J = 9 Hz, CH2O), 4.46 (1H, d, J = 12
Hz, CH2Ph), 4.55 (1H, d, J = 12 Hz, CH2Ph), 4.61 (1H, d, J = 9
Hz, CH2O), 4.89 (1H, s, NCHO), 7.27–7.40 (5H, m, Ph); δC
(100 MHz, CDCl3) 7.3 (CH3), 24.9 (C(CH3)3), 36.3 (C(CH3)3,
45.9 (C(7)), 52.5 (CO2CH3), 68.3 (C(4)), 71.2, 73.9
(CH2OCH3), 78.1 (C(6)), 81.2 (C(5)), 96.2 (C(2)), 127.9, 128.2,
128.6, 136.7, 171.5 (CO2Me), 177.9 (C(8)); HRMS: found
414.1876 for C21H29NNaO6 [M + Na]+, requires 414.1887.
+
Data for 22b: Rf 0.53 (1 : 1 petrol–EtOAc); δH (500.3 MHz;
C6D6) 1.23 (9H, s, C(CH3)3), 1.60 (3H, d, J = 7.5 Hz, C(7)
CH3), 2.32 (1H, q, J = 7.5 Hz, C(7)H), 2.94 (br, s, OH), 2.85
(1H, d, J = 9.8 Hz, CHHOCH2Ph), 3.04 (1H, d, J = 9.8 Hz,
CHHOCH2Ph), 3.44 (1H, d, J = 9.1 Hz, C(4)HH), 3.48 (3H, s,
COOCH3), 4.06 (2H, s, OCH2Ph), 5.00 (1H, d, J = 9.1 Hz, C(4)
HH), 5.02 (1H, s, C(2)H), 7.13–7.25 (5H, m, ArH); δC
(125 MHz; CDCl3) 12.3 (C(7)CH3), 25.6 (C(CH3)3), 36.0
(C(CH3)3), 49.5 (C(7)), 51.9 (COOCH3), 71.2 (C(4)), 73.3
(CH2OCH2Ph), 80.0 (OCH2Ph), 96.9 (C(2)), 127.8 (para-CH),
128.6, 128.7 (ortho- and meta-CH), 137.5 (ipso-CH), 171.4 and
179.1 (2 × CvO); m/z (ESI+) 805 ([2M + Na]+, 99%); HRMS
(ESI+) C21H29NNaO6+ requires 414.1887, found 414.1891. Data
for 23b: Rf 0.47 (1 : 1 petrol–EtOAc); δH (500 MHz; C6D6) 1.08
(3H, d, J = 7.4 Hz, CH3), 1.19 (9H, s, C(CH3)3), 3.11 (2H, s,
CH2OCH2Ph), 3.53 (3H, s, COOCH3), 3.65 (1H, q, J = 7.4 Hz,
C(7)H), 3.73 (1H, d, J = 9.4 Hz, C(4)HH), 4.00 (1H, d, J = 11.9
Hz, OCHHPh), 4.02 (1H, d, J = 11.9 Hz, OCHHPh), 5.02 (1H,
d, J = 9.4 Hz, C(4)HH), 5.09 (1H, s, C(2)H), 7.13–7.24 (5H, m,
ArH); δC (125 MHz; CDCl3) 7.9 (CH3), 25.4 (C(CH3)3), 36.5
(C(CH3)3), 47.7 (C(7)), 52.0 (COOCH3), 69.6 (C(4)), 70.5
(CH2OCH2Ph), 73.4 (OCH2Ph), 95.9 (C(2)), 127.9 (para-CH),
128.7, 128.8 (ortho- and meta-CH), 137.5 (ipso-CH), 171.3 and
175.3 (2 × CvO); m/z (ESI+) 805 ([2M + Na]+, 99%); HRMS
(ESI+) C21H29NNaO6+ requires 414.1887, found 414.1892.
+
C15H25NNaO6 [M + Na]+, requires 338.1574; Data for 22a/
23a (major isomer); Rf = 0.13 (petrol (40–60)–EtOAc, 3 : 1);
νmax (film) 3440, 2973, 1655; δH (400 MHz; CDCl3) 0.89 (9H,
s, CC(CH3)3), 1.06 (3H, d, J = 7.6 Hz, CH3), 3.37 (2H, m,
CH2OCH3), 3.38 (3H, s, CH2OCH3), 3.50 (1H, m, C(7)H), 3.68
(1H, d, J = 9.6 Hz, C(4)H), 3.79 (3H, s, CO2CH3), 4.72 (1H, d,
C(4)H), 4.82 (1H, s, C(2)H); δC (100 MHz; CDCl3) 7.62 (CH3),
25.0 (C(CH3)3), 36.1 (C(CH3)3), 47.4 (C(7)), 52.5 (CO2CH3)
59.0 (CH2OCH3) 65.3 (C(4)H), 71.6 (CH2OCH3), 77.3 (C(6)),
79.7 (C(5)), 95.4 (C(2)), 171.1 (CO2CH3), 175.7 (C(8)O); Data
for 22a/23a (minor isomer) Rf = 0.13 δH (400 MHz; CDCl3)
0.91 (9H, s, C(CH3)3), 1.41 (3H, d, J = 7.8 Hz, CH3), 2.43 (1H,
q, J = 7.8 Hz, C(7)H), 3.27 (1H, d, J = 9.9 Hz, CH2OCH3), 3.39
(3H, s, CH2OCH3), 3.43 (1H, d, J = 9.9 Hz, CH2OCH3), 3.78
(3H, s, CO2CH3) 3.94 (1H, d, J = 14.4 Hz, C(4)H), 4.11 (1H, d,
J = 14.4 Hz, C(4)H), 4.72 (1H, s, C(2)H); δC (100 MHz;
CDCl3) 12.1 (CH3), 25.1 (C(CH3)3), 35.6 (C(CH3)3), 49.4
(C(7)), 52.4 (CO2CH3) 59.0 (CH2OCH3) 62.0 (C(4)H), 76.0
(CH2OCH3), 77.0 (C(6)), 80.3 (C(5)), 96.1 (C(2)), 171.1
(CO2CH3), 175.7 (C(8)O); m/z (ESI)+ 338 ([M + Na]+, 65%)
658 ([2M + Na]+, 100%); HRMS (ESI+) Found 338.1575 for
C15H25NNaO6+ [M + Na]+, requires 338.1574.
(2R,5R,6R,7S) And (2R,5R,6S, 7R)-1-aza-2-(tert-butyl)-6-
hydroxy-7-methyl-6-(tert-butoxymethyl)-5-methoxycarbonyl-3-
oxa-8-oxobicyclo[3.3.0]octane 21c and 22c
To a solution of oxazolidine 20c (0.071 g, 0.20 mmol) in dry
methanol (5 mL) was added NaOMe (0.011 g, 0.200 mmol) and
solution was stirred for 24 h at r.t. and then partitioned between
diethyl ether (10 mL) and NH4Cl (10 mL). The diethyl ether
layer was washed with brine (20 mL), dried over MgSO4,
filtered and the solvent was evaporated in vacuo to give crude
product as a light brown sticky oil (0.052 g), which was purified
by flash column chromatography (5 : 1 petrol–EtOAc) to give a
mixture of isomers 21c (0.012 g, 17%) and 22c (0.027 g, 38%)
both as white solid. Data for 21c: Rf 0.34 (3 : 1 petrol–EtOAc);
δH (500 MHz; CDCl3) 0.88 (9H, s, C(CH3)3), 1.12 (3H, d, J =
7.2 Hz, C(7)CH3), 1.17 (9H, s, C(CH3)3), 3.04 (br, s, OH), 3.23
(1H, q, J = 7.2 Hz, C(7)H), 3.34 (2H, s, CH2OC(CH3)3), 3.76
(3H, s, COOCH3), 4.10 (1H, d, J = 8.8 Hz, C(4)HH), 4.57
(2R,5R,6R,7S), (2R,5R,6S,7R)- And (2R,5R,6S,7S)-1-Aza-2-(tert-
butyl)-6-hydroxy-7-methyl-6-(benzyloxymethyl)-5-
methoxycarbonyl-3-oxa-8-oxobicyclo[3.3.0]octane 21b, 22b and
23b
To a solution of oxazolidine 20b (0.048 g, 0.12 mmol) in dry
methanol (5 mL) was added NaOMe (0.0067 g, 0.12 mmol) and
solution was stirred for 24 h at r.t. and then partitioned between
diethyl ether (10 mL) and aq. NH4Cl (10 mL). The ether layer
was washed with brine (20 mL), dried over MgSO4, filtered and
the solvent was evaporated in vacuo to give crude product as a
light brown sticky oil (0.052 g), which was purified by flash
column chromatography (5 : 1 petrol–EtOAc) to give a mixture
of isomers 21b, (0.014 g, 30%) as white solid, 22b (0.0062 g,
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3481

View Article Online
(1H, d, J = 8.8 Hz, C(4)HH), 4.88 (1H, s, C(2)H); δC
(125 MHz; CDCl3) 7.4 (CH3), 24.9 (C(CH3)3), 25.6 (CH2OC
(CH3)3), 36.3 (C(CH3)3), 46.1 (C(7)), 52.1 (COOCH3),
63.1 (CH2OC(CH3)3), 68.0 (C(4)), 95.9 (C(2)), 171.6 and 178.0
(2 × CvO); m/z (ESI+) 737 ([2M + Na]+, 98%); HRMS (ESI+)
(50 mL). The mixture was extracted with Et2O (3 × 50 mL), the
combined organic layers were dried (MgSO4) and the solvents
were evaporated under reduced pressure to give crude material
containing 21a and 24a in 20 : 1 ratio (by NMR). Column
chromatography on silica gel with 2 : 1 Et2O–petrol as the eluent
gave 21a (430 mg, 46%) as crystalline solid. Data as above.
+
C18H31NNaO6 requires 380.2044, found 380.2040. Data for
22c: Rf 0.20 (3 : 1 petrol–EtOAc); δH (500.3 MHz; C6D6) 0.88
(9H, s, CH2OC(CH3)3), 1.29 (9H, s, C(CH3)3), 1.73 (3H, d, J =
7.7 Hz, C(7)CH3), 2.46 (1H, q, J = 7.7 Hz, C(7)H), 2.91 (1H, d,
J = 9.4 Hz, CHHOC(CH3)3), 3.13 (1H, d, J = 9.4 Hz, CHHOC
(CH3)3), 3.15 (br, s, OH), 3.53 (3H, s, COOCH3), 3.63 (1H, d,
J = 9.2 Hz, C(4)HH), 5.04 (1H, d, J = 9.2 Hz, C(4)HH), 5.15
(1H, s, C(2)H); δC (125 MHz; C6D6) 12.5 (CH3), 25.7
(C(CH3)3), 27.0 (C(CH3)3), 36.4 (C(CH3)3), 49.8 (C(7)), 51.8
(COOCH3), 66.0 (CH2OC(CH3)3), 71.4 (C(4)), 96.8 (C(2)),
171.5 and 179.2 (2 × CvO); m/z (ESI+) 737 ([2M + Na]+,
(2R,5R,6R,7R)-2-(tert-Butyl)-6-hydroxy-6-(benzyloxymethyl)-5-
methoxycarbonyl-7-methyl-3-oxa-8-oxobicyclo[3.3.0]octane 24b
A solution of 15b (1.38 g, 3.65 mmol) in THF (40 mL) under
nitrogen was cooled to −10 °C in methanol–ice bath and LDA
(4.46 mL of a 1.8 mol l−1 solution, 8.03 mmol) was slowly
added. The mixture was stirred for 30 min and then iodomethane
(0.57 mL, 9.20 mmol) was added. The reaction mixture was left
to stir for 90 more minutes, gradually reaching r.t., quenched
with sat. aq. NH4Cl and extracted with diethyl ether (4 ×
50 mL). The combined organic layers were dried (MgSO4) and
the solvents were evaporated under reduced pressure. The
residue was re-dissolved in small amount of diethyl ether and
dry-loaded onto silica gel. Column chromatography with 3 : 1
petrol–EtOAc, increasing polarity to 2 : 1, gave 475 mg of mixed
fraction of 24b and 21b (5 : 2 by NMR), 24b (415 mg) as
viscous oil and the starting material 15b (410 mg). Rf 0.26
(petrol–EtOAc 2 : 1); νmax (film) 3418, 2957, 1730, 1716; δH
(400 MHz, C6D6) 1.18 (9H, s, t-Bu), 1.44 (3H, d, J = 8 Hz,
CH3CH), 2.85 (1H, q, J = 8 Hz, CH3CH), 3.26 (1H, d, J = 9,
CH2OBn), 3.29 (3H, s, COOCH3), 3.40 (1H, d, J = 9 Hz,
CH2OBn), 3.44 (1H, s, OH), 4.18 (2H, AB quartet, J = 12 Hz,
CH2Ph), 4.42 (1H, d, J = 9 Hz, C(4)H), 4.78 (1H, d, J = 9 Hz, C
(4)H), 5.20 (1H, s, C(2)H), 7.20–7.35 (5H, m, Ph); δC
(100 MHz, C6D6) 13.0 (CH3), 25.5 (C(CH3)3), 36.1 (C(CH3)3,
51.7 (CO2CH3), 54.1 (CH2OCH3), 69.4 (C(4)), 71.1
(CH2OCH3), 73.7, 78.8 (C(6)), 78.9 (C(6)), 97.0 (C(2)), 137.6,
172.0 (CO2Me), 179.9 (C(8)); HRMS: found 414.1874 for
C21H29NNaO6+ [M + Na]+, requires 414.1887.
+
35%); HRMS (ESI+) C18H31NNaO6 requires 380.2044, found
380.2039.
(2R,5R,6R,7R)-2-(tert-Butyl)-6-hydroxy-6-(methyloxymethyl)-5-
methoxycarbonyl-7-methyl-3-oxa-8-oxobicyclo[3.3.0]octane 24a
A solution of 15a (1.22 g, 4.05 mmol) in THF (40 mL) under
nitrogen was cooled to −10 °C in a methanol–ice bath and LDA
(4.95 mL of a 1.8 mol l−1 solution, 8.91 mmol) was slowly
added. The mixture was stirred for 30 min and then iodomethane
(0.63 mL, 10.13 mmol) was added. The reaction mixture was
left to stir for 90 more minutes, gradually reaching r.t. The reac-
tion was quenched with sat. aq. NH4Cl and extracted with
diethyl ether (4 × 50 mL). The combined organic layers were
dried (MgSO4) and the solvents were evaporated under reduced
pressure. The residue was re-dissolved in small amount of
diethyl ether and dry-loaded onto silica gel. Column chromato-
graphy with 1 : 1 petrol–Et2O, increasing polarity to neat Et2O,
gave a mixed fraction of 24a and 21a (10 : 4 by NMR)(213 mg),
24a (780 mg) and the starting material 15a (180 mg). Data
for 24a: Rf 0.40 (Et2O); [α]2D5 = +36.3° (c = 1, MeOH); m.p.
102–104 °C; νmax (film) 3420, 2958, 1740, 1716; δH (400 MHz,
C6D6) 1.20 (9H, s, t-Bu), 1.43 (3H, d, J = 8 Hz, CH3CH), 2.79
(1H, q, J = 8 Hz, CH3CH), 2.89 (3H, s, CH3O), 3.05 (1H, s,
OH), 3.08 (1H, d, J = 9 Hz, CH2OCH3), 3.18 (1H, d, J = 9 Hz,
CH2OCH3), 3.33 (3H, s, CO2CH3), 4.39 (1H, d, J = 9 Hz, C(4)
H), 4.75 (1H, d, J = 9 Hz, C(4)H), 5.22 (1H, s, C(2)H); δC
(100 MHz, C6D6) 12.9 (CH3), 25.5 (C(CH3)3), 36.1 (C(CH3)3,
51.7 (CO2CH3), 54.0, 58.7 (CH2OCH3), 69.4 (C(4)), 73.0
(CH2OCH3), 78.6 (C(6)), 97.2 (C(2)), 171.9 (CO2Me), 179.7
(2R,5R,6R,7S)-2-(tert-Butyl)-6-hydroxy-6-(benzyloxymethyl)-5-
methoxycarbonyl-7-methyl-3-oxa-8-oxobicyclo[3.3.0]octane 21b
A solution of 24b (445 mg, 1.14 mmol) in THF (15 mL) under
nitrogen was cooled to 0 °C in a water–ice bath. The solution
was magnetically stirred and KHMDS (5 mL of a 0.5 mol l−1
solution, 2.5 mmol) was added over a period of 2–3 min and the
mixture left to stir for 15 more minutes. t-BuOH (0.5 mL) was
added, followed by sat. aq. NH4Cl (20 mL). The mixture was
extracted with Et2O (3 × 30 mL), the combined organic layers
were dried (MgSO4) and the solvents were evaporated under
reduced pressure to give 320 mg of crude material containing
21b and 24b in 20 : 1 ratio (by NMR analysis). Column chrom-
atography on silica gel with 2 : 1 petrol–EtOAc as the eluent
gave 200 mg of 21b (45%) as crystalline solid. Data as above.
+
(C(8)); HRMS: found 338.1570 for C15H25NNaO6 , [M + Na]+
requires 338.1574.
(2R,5R,6R,7S)-2-(tert-Butyl)-6-hydroxy-6-(methoxymethyl)-5-
methoxycarbonyl-7-methyl-3-oxa-8-oxobicyclo[3.3.0]octane 21a
A solution of 24a (927 mg, 2.9 mmol) in THF (60 mL) under
nitrogen was cooled to 0 °C in a water–ice bath and KHMDS
(12.9 mL of a 0.5 mol l−1 solution, 6.47 mmol) was added over
a period of 3–5 min while stirring. The cooling bath was
removed and the mixture was left to stir for 15 more minutes.
t-BuOH (0.5 mL) was added, followed by sat. aq. NH4Cl
(2R,3R,4S)-3-Hydroxy-2-hydroxymethyl-3-methoxymethyl-4-
methyl-5-oxo-pyrrolidine-2-carboxylic acid methyl ester 25a
To a solution of 21a (460 mg, 1.46 mmol) in 2,2,2-trifluoroetha-
nol (15 mL) was added 1,3-propanedithiol (0.22 mL, 2.2 mmol)
3482 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
and conc. aq. HCl (0.18 mL, 2.2 mmol). The mixture was left to
stir at r.t. and after 24 h the solvent was removed under reduced
pressure. The residue was re-dissolved in small amount of DCM
and dry-loaded onto silica gel. Column chromatography on silica
gel with Et2O, increasing polarity to 10 : 1 Et2O–MeOH, gave
30 mg (6%) of 26a (eluted with Et2O) and 320 mg (88%) of 25a
(eluted with 10 : 1 Et2O–MeOH). Rf 0.23 (Et2O–MeOH 10 : 1).
J = 5 Hz, CH3CH), 2.68 (1H, q, J = 5 Hz, CH3CH), 3.28 (3H, s,
OCH3), 3.47 (1H, d, J = 9 Hz, CH2OCH3), 3.54 (1H, d, J = 9
Hz, CH2OCH3), 3.95 (1H, d, J = 12 Hz, CH2OH), 4.20 (1H, d, J
= 12 Hz, CH2OH), 7.09 (1H, s, NH); δC (100 MHz, acetone-d6)
7.4, 42.6, 58.4, 63.8, 71.4, 73.9, 80.4, 173.0, 177.9; HRMS:
found 232.0819 for C9H14NO6− [M − H]−, requires 232.0827.
[α]D25 = − 60.5° (c = 1, MeOH); m.p. 157–159 °C; νmax (cm−1
,
neat) 3315, 3270, 1710, 1650; δH (400 MHz, CD3OD) 1.05 (3H,
d, J = 7 Hz, CH3CH), 2.77 (1H, q, J = 7 Hz, CH3CH), 3.28 (1H,
d, J = 10 Hz, CH2OCH3), 3.29 (3H, s, OCH3), 3.47 (1H, d, J =
10 Hz, CH2OCH3), 3.74 (3H, s, COOCH3), 3.83 (1H, d, J = 11
Hz, CH2OH), 4.20 (1H, d, J = 11 Hz, CH2OH); δC (100 MHz,
CD3OD) 6.4, 42.0, 52.0, 58.2, 63.7, 72.2, 72.5, 80.4, 172.2,
(2R,3R,4S)-3-Benzyloxymethyl-3-hydroxy-2-hydroxymethyl-4-
methyl-5-oxopyrrolidine-2-carboxylic acid 27b
To a solution of 25b (180 mg, 0.56 mmol) in 1 : 1 : 1 THF–
MeOH–H2O (15 mL) was added LiOH·H2O (47 mg,
1.11 mmol) and the mixture was stirred for 36 h at r.t. The sol-
vents were evaporated under reduced pressure and to the viscous
aqueous residue was added sat. aq. soln. of NaCl (20 mL). The
pH was then brought to 2–3 by careful addition of conc. HCl
and the acidic solution was extracted with EtOAc (5 × 20 mL).
The combined organic layers were dried (MgSO4) and the
solvent was evaporated under reduced pressure to give 27b as
solid foam (170 mg, 98%). m.p. 90–100 °C (dec.). νmax (film)
3295, 2946, 1687; δH (400 MHz, CD3OD) 1.07 (3H, d, J = 8
Hz, CH3CH), 2.83 (1H, q, J = 8 Hz, CH3CH), 3.48 (1H, d, J = 8
Hz, CH2OBn), 3.65 (1H, d, J = 8 Hz, CH2OBn), 3.88 (1H, d, J
= 10 Hz, CH2Ph), 4.23 (1H, d, J = 10 Hz, CH2Ph), 4.44 (1H, d,
J = 12 Hz, CH2OH), 4.57 (1H, d, J = 12 Hz, CH2OH),
7.27–7.36 (5H, m, Ph); δC (100 MHz, CD3OD) 6.9, 42.9, 63.8,
71.1, 72.4, 73.2, 80.4, 127.6, 127.7, 128.3, 138.3, 173.3, 179.9;
179.8; HRMS: found 270.0948 for C10H17NNaO6 [M + Na]+,
+
requires 270.0948.
(2R,3R,4S)-3-Benzyloxymethyl-3-hydroxy-2-hydroxymethyl-4-
methyl-5-oxopyrrolidine-2-carboxylic acid methyl ester 25b
To a solution of 21b (300 mg, 0.77 mmol) in 2,2,2-trifluoroetha-
nol (10 mL) was added 1,3-propanedithiol (0.12 mL,
1.16 mmol) and conc. aq. HCl (0.10 mL, 1.16 mmol). The
mixture was left to stir at r.t. and after 24 h the solvent was
removed under reduced pressure. The residue was re-dissolved in
small amount of DCM and dry-loaded onto silica gel. Column
chromatography with Et2O, increasing polarity to 10 : 1 Et2O–
MeOH, gave 56 mg (18%) of 26b (eluted with Et2O) and
200 mg (80%) of 25b (eluted with 10 : 1 Et2O–MeOH). Rf 0.38
(Et2O–MeOH 10 : 1); νmax (film) 3324, 2950, 1730, 1688; δH
(400 MHz, CDCl3) 1.08 (3H, d, J = 7 Hz, CH3CH), 2.84 (1H, q,
J = 7 Hz, CH3CH), 3.42 (1H, d, J = 12 Hz, CH2OBn), 3.56 (3H,
s, COOCH3), 3.56 (1H, d, J = 12 Hz, CH2OBn), 4.00 (1H, app
t, J = 8 Hz, CH2OH), 4.09 (1H, dd, J = 12, 8 Hz, CH2OH), 4.17
(1H, dd, J = 12, 8 Hz, CH2OH), 4.32 (1H, s, OH), 4.48 (2H, AB
quartet, J = 12 Hz, CH2Ph), 7.18 (1H, s, NH), 7.27–7.37 (5H,
m, Ph); δC (100 MHz, CDCl3) 7.1, 42.3, 52.7, 63.2, 70.3, 71.3,
73.4, 81.2, 127.6, 127.9, 128.5, 137.2, 171.7, 179.1; HRMS:
HRMS: found 308.1136 for C15H18NO6 [M − H]−, requires
−
308.1140.
(2R,3R,4S)-2-Acetoxymethyl-3-hydroxy-3-methoxymethyl-4-
methyl-5-oxopyrrolidine-2-carboxylic acid methyl ester 28a
To a magnetically stirred solution of 25a (400 mg, 1.62 mmol),
Et3N (0.24 mL, 1.7 mmol) and DMAP (12 mg, 0.1 mmol) in
THF (50 mL) was added acetyl chloride (0.12 mL, 1.7 mmol)
and the mixture was left to stir at r.t. After 40 min there was still
starting material visible on the TLC, so additional amount of
Et3N–DMAP–AcCl (0.24 mL, 12 mg and 0.12 mL respectively)
were added and 30 min later the reaction was quenched with aq.
soln. of NH4Cl and a few drops of conc. HCl. The mixture was
extracted with ethyl acetate, the combined organic layers were
dried (MgSO4) and the solvents were evaporated under reduced
pressure to give crude 28a, which was purified by column
chromatography on silica gel with 20 : 1 Et2O–MeOH as the
eluent. Pure product 28a was obtained as a solid foam (375 mg,
80%) along with 40 mg of the starting material 25a. νmax (film)
3302, 1730, 1698; δH (400 MHz, CDCl3) 1.12 (3H, d, J = 7 Hz,
CH3CH), 2.06 (3H, s, COCH3), 2.67 (1H, q, J = 7 Hz, CH3CH),
3.32 (1H, d, J = 10 Hz, CH2OCH3), 3.32 (3H, s, OCH3), 3.42
(1H, d, J = 10 Hz, CH2OCH3), 3.71 (1H, s, OH), 3.75 (3H, s,
COOCH3), 4.42 (1H, d, J = 11 Hz, CH2OAc), 4.73 (1H, d, J =
11 Hz, CH2OAc), 6.82 (1H, s, NH); δC (100 MHz, CDCl3) 7.3,
20.8, 42.0, 52.9, 59.1, 66.4, 70.1, 72.8, 80.5, 170.7, 178.0;
HRMS: found 312.1055 for C12H19NNaO7+ [M + Na]+, requires
312.1054.
+
found 346.1252 for C16H21NNaO6 [M + Na]+, requires
346.1261.
(2R,3R,4S)-3-Hydroxy-2-hydroxymethyl-3-methoxymethyl-4-
methyl-5-oxopyrrolidine-2-carboxylic acid 27a
To a solution of 25a (440 mg, 1.78 mmol) in 1 : 1 : 1 THF–
MeOH–H2O (30 mL) was added LiOH·H2O (150 mg,
3.56 mmol) and the mixture was stirred overnight at r.t. The sol-
vents were evaporated under reduced pressure and to the viscous
aqueous residue was added sat. aq. soln. of NaCl (20 mL). The
pH was then brought to 2–3 by careful addition of conc. HCl
and the acidic solution was extracted with THF (5 × 20 mL).
The combined organic layers were dried (MgSO4) and the
solvent was evaporated under reduced pressure to give a highly
viscous oily residue, which was stirred for one night in Et2O.
The amorphous residue was dried under reduced pressure to give
27a as solid foam (340 mg, 82%). Rf 0.3 (EtOAc–MeOH 1 : 1);
[α]2D5 = −42.1° (c = 1, MeOH); m.p. 110–120 °C (dec.); νmax
(film) 3293, 2939, 1682; δH (400 MHz, acetone-d6) 1.05 (3H, d,
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3483

View Article Online
24 C. B. Troostwijk and R. M. Kellogg, J. Chem. Soc., Chem. Commun.,
1977, 932–933.
25 D. Seebach and M. Eberle, Synthesis, 1986, 37–40.
26 J. L. Van der Bann, J. W. F. K. Barnick and F. Bickelhaupt, Tetrahedron,
1978, 34, 223–231.
27 B. Simoneau and P. Brassard, Tetrahedron, 1986, 42, 3767–3774.
28 D. Seebach, J. D. Aebi, M. G. Coquaz and R. Naef, Helv. Chim. Acta,
1987, 70, 1194–1216.
29 D. Seebach and J. D. Aebi, Tetrahedron Lett., 1984, 25, 2545–2548.
30 E. L. Bentz, R. Goswami, M. G. Moloney and S. M. Westaway, Org.
Biomol. Chem., 2005, 3, 2872–2882.
31 Y.-C. Jeong, M. Anwar, T. M. Nguyen, B. S. W. Tan, C. L. L. Chai and
M. G. Moloney, Org. Biomol. Chem., 2011, 9, 6663–6669.
32 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. 1,
1998, 223–235.
33 E. J. Corey and G. A. Reichard, J. Am. Chem. Soc., 1992, 114, 10677–
10678.
34 K. Takahashi, M. Kawabata, D. Uemura, S. Iwadare, R. Mitomo,
F. Nakano and A. Matsuzaki, Tetrahedron Lett., 1985, 26, 1077–1078.
35 P. Angelov, Synlett, 2010, 1273–1275.
(2R,3R,4S)-2-Acetoxymethyl-3-benzyloxymethyl-3-hydroxy-4-
methyl-5-oxo-pyrrolidine-2-carboxylic acid methyl ester 28b
To a stirred solution of 25a (130 mg, 0.4 mmol), Et3N (0.14 mL,
1 mmol) and DMAP (12 mg, 0.1 mmol) in DCM (10 mL) at r.t.
was added acetyl chloride (0.07 mL, 1 mmol). After 15 min, the
reaction mixture was washed with dilute aq. HCl. The aqueous
layer was extracted once with DCM, the combined organic
layers were dried (MgSO4) and the solvent was evaporated under
reduced pressure to give practically clean 28b as a crystalline
solid. Yield (132 mg, 90%). Rf 0.13 (Et2O); m.p. 146–148 °C;
νmax (film) 3340, 1741, 1707; δH (400 MHz, CDCl3) 1.12 (3H,
d, J = 6 Hz, CH3CH), 2.06 (3H, s, COCH3), 2.23 (1H, s, OH),
2.71 (1H, q, J = 6 Hz, CH3CH), 3.42 (1H, d, J = 10 Hz,
CH2OBn), 3.50 (1H, d, J = 10 Hz, CH2OBn), 3.61 (3H, s,
COOCH3), 4.44 (1H, d, J = 11 Hz, CH2OAc), 4.51 (2H, AB
quartet, J = 12 Hz, CH2Ph), 4.77 (1H, d, J = 11 Hz, CH2OAc),
6.91 (1H, s, NH), 7.27–7.38 (5H, m, Ph); δC (100 MHz, CDCl3)
7.4, 20.8, 42.3, 52.9, 66.3, 70.3, 70.7, 73.6, 80.5, 127.8, 128.1,
128.6, 136.9, 170.7, 178.1; HRMS: found 388.1357 for
C18H23NNaO7+ [M + Na]+, requires 388.1367.
36 I. Ivanov, S. Nikolova, P. Angelov, S. Statkova-Abeghe and
E. Kochovska, ARKIVOC, 2007 (xv), 11–17.
37 F. Xu, J. D. Armstrong, G. X. Zhou, B. Simmons, D. Hughes, Z. Ge and
E. J. J. Grabowski, J. Am. Chem. Soc., 2004, 126, 13002–13009.
38 J. H. Bailey, D. Cherry, J. Dyer, M. G. Moloney, M. J. Bamford,
S. Keeling and R. B. Lamont, J. Chem. Soc., Perkin Trans. 1, 2000,
2783–2792.
39 J. Dyer, S. Keeling, A. King and M. G. Moloney, J. Chem. Soc., Perkin
Trans. 1, 2000, 2793–2804.
References
40 J. H. Bailey, A. T. J. Byfield, P. J. Davis, A. C. Foster, M. Leech,
M. G. Moloney, M. Muller and C. K. Prout, J. Chem. Soc., Perkin
Trans. 1, 2000, 1977–1982.
41 Crystallographic data (excluding structure factors) have been deposited
with the Cambridge Crystallographic Data Centre (CCDC 860455 &
860456).
42 A. A. Puri and M. Bogyo, ACS Chem. Biol., 2009, 4, 603–616.
43 J. A. Frearson and I. T. Collie, Drug Discovery Today, 2009, 14, 1150–
1158.
44 C. L. Bagwell, M. G. Moloney and A. L. Thompson, Bioorg. Med.
Chem. Lett., 2008, 18, 4081–4086.
45 C. L. Bagwell, M. G. Moloney and M. Yaqoob, Bioorg. Med. Chem.
Lett., 2010, 20, 2090–2094.
46 P. Ertl, B. Rohde and P. Selzer, J. Med. Chem., 2000, 43, 3714–3717.
47 P. Ertl, in Molecular Drug Properties, ed. R. Mannhold, Wiley-VCH,
Weinheim, 2007, pp. 111–126.
1 M. G. Moloney, P. C. Trippier, M. Yaqoob and Z. Wang, Curr. Drug Dis-
covery Technol., 2004, 1, 181–199.
2 R. R. Manam, S. Teisan, D. J. White, B. Nicholson, J. Grodberg,
S. T. C. Neuteboom, K. S. Lam, D. A. Mosca, G. K. Lloyd and
B. C. M. Potts, J. Nat. Prod., 2005, 68, 240–243.
3 C. Zhao, J. M. Coughlin, J. Ju, D. Zhu, E. Wendt-Pienkowski, X. Zhou,
Z. Wang, B. Shen and Z. Deng, J. Biol. Chem., 2010, 285, 20097–20108.
4 C. H. Zhao, J. H. Ju, S. D. Christenson, W. C. Smith, D. F. Song,
X. F. Zhou, B. Shen and Z. X. Deng, J. Bacteriol., 2006, 188, 4142–
4147.
5 A. S. Kende, K. Kawamura and R. J. DeVita, J. Am. Chem. Soc., 1990,
112, 4070–4072.
6 E. O. Onyango, J. Tsurumoto, N. Imai, K. Takahashi, J. Ishihara and
S. Hatakeyama, Angew. Chem., Int. Ed., 2007, 46, 6703–6705.
7 K. Eto, M. Yoshino, K. Takahashi, J. Ishihara and S. Hatakeyama, Org.
Lett., 2011, 13, 5398–5401.
48 Marvin was used for drawing, displaying and characterizing chemical
structures, substructures and reactions, Marvin 5.2.1, 2009, ChemAxon
(http://www.chemaxon.com). Calculator Plugins were used for structure
property prediction and calculation, Marvin 5.2.5.1, 2009, ChemAxon
(http://www.chemaxon.com).
8 D. K. Mohapatra, D. Mondal, R. G. Gonnade, M. S. Chorghade and
M. K. Gurjar, Tetrahedron Lett., 2006, 47, 6031–6035.
9 J. P. N. Papillon and R. J. K. Taylor, Org. Lett., 2000, 2, 1987–1990.
10 M. G. Moloney and M. Yaqoob, Tetrahedron Lett., 2008, 49, 6202–6204.
11 M. Anwar and M. G. Moloney, Tetrahedron Lett., 2007, 48, 7259–7262.
12 M. D. Andrews, A. G. Brewster and M. G. Moloney, Synlett, 1996, 612–
614.
13 T. J. Donohoe, J. Y. K. Chiu and R. E. Thomas, Org. Lett., 2007, 9, 421–
424.
14 N. J. Bennett, J. C. Prodger and G. Pattenden, Tetrahedron, 2007, 63,
6216–6231.
49 Bioassay of products:70–72 Microbiological assays were performed by the
hole-plate method with the test organism Staphylococcus aureus N.C.T.
C. 6571 or E. coli X580. Solutions (100 μl) of the compounds to be
tested (4 mg ml−1) were loaded into wells in bioassay plates, and incu-
bated overnight at 37 °C. The diameters of the resultant inhibition zones
were measured ( 1 mm), and relative potency estimated by reference to
standards prepared with cephalosporin C; this is expressed as zone diam-
eter per M, of the analyte relative to cephalosporin C standard.
50 N. Chandan and M. G. Moloney, Org. Biomol. Chem., 2008, 6, 3664–
3666.
15 T. Yamada, K. Sakaguchi, T. Shinada, Y. Ohfune and V. A. Soloshonok,
Tetrahedron: Asymmetry, 2008, 19, 2789–2795.
16 H. L. Teng, F. L. Luo, H. Y. Tao and C. J. Wang, Org. Lett., 2011, 13,
5600–5603.
51 T. Hill, P. Kocis and M. G. Moloney, Tetrahedron Lett., 2006, 47, 1461–
1463.
17 D. Seebach, A. R. Sting and M. Hoffmann, Angew. Chem., Int. Ed. Engl.,
1996, 35, 2708–2748.
52 Y.-C. Jeong and M. G. Moloney, Synlett, 2009, 2487–2491.
53 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.
18 T. Ling, B. C. Potts and V. R. Macherla, J. Org. Chem., 2010, 75, 3882–
3885.
19 T. Ling, V. R. Macherla, R. R. Manam, K. A. McArthur and
B. C. M. Potts, Org. Lett., 2007, 9, 2289–2292.
54 D. J. Payne, M. N. Gwynn, D. J. Holmes and D. L. Pompliano, Nat. Rev.
Drug Discovery, 2006, 6, 29–40.
55 C. Morel and E. Mossialos, Br. Med. J., 2010, 340, 1115–1118.
56 A. D. So, N. Gupta and O. Cars, Br. Med. J., 2010, 340, 1091–1092.
57 S. Barelier, J. Pons, K. Gehring, J.-M. Lancelin and I. Krimm, J. Med.
Chem., 2010, 53, 5256–5266.
20 L. Dubinsky, L. M. Jarosz, N. Amari, P. Krief, V. V. Kravchenko,
B. P. Krom and M. M. Meijler, Chem. Commun., 2009, 7378–7380.
21 P. P. Mahulikar and R. B. Mane, J. Chem. Res., 2006, 15–18.
22 J. Lloyd, H. J. Finlay, K. Atwal, A. Kover, J. Prol, L. Yan, R. Bhandaru,
W. Vaccaro, T. Huynh, C. S. Huang, M. Conder, T. Jenkins-West, H. Sun,
D. Li and P. Levesque, Bioorg. Med. Chem. Lett., 2009, 19, 5469–5473.
23 B. M. Trost, M. Osipov, P. S. J. Kaib and M. T. Sorum, Org. Lett., 2011,
13, 3222–3225.
58 V. P. Gullo, J. McAlpine, K. S. Lam, D. Baker and F. Petersen, J. Ind.
Microbiol. Biotechnol., 2006, 33, 523–531.
3484 | Org. Biomol. Chem., 2012, 10, 3472–3485
This journal is © The Royal Society of Chemistry 2012

View Article Online
59 A. M. Rouhi, Chem. Eng. News, 2003, 81, 77–91.
60 R. B. Breinbauer, I. R. Vetter and H. Waldmann, Angew. Chem., Int. Ed.,
2002, 41, 2878–2890.
61 A. Ganesan, Curr. Opin. Biotechnol., 2004, 15, 584–590.
62 M. A. Koch and H. Waldmann, Drug Discovery Today, 2005, 10, 471–
483.
66 J. T. Njardarson, C. Gaul, D. Shan, X.-Y. Huang and S. J. Danishefsky,
J. Am. Chem. Soc., 2004, 126, 1038–1040.
67 F. vonNussbaum, M. Brands, B. Hinzen, S. Weigand and D. Habich,
Angew. Chem., Int. Ed., 2006, 45, 5072–5129.
68 F. E. Koehn and G. T. Carter, Nat. Rev. Drug Discovery, 2005, 4, 206–
220.
63 S. Shang and D. S. Tan, Curr. Opin. Chem. Biol., 2005, 9, 248–
258.
64 A. Nören-Müller, I. Reis-Correa, H. Prinz, C. Rosenbaum, K. Saxena,
H. J. Schwalbe, D. Vestweber, G. Cagna, S. Schunk, O. Schwarz,
H. Schiewe and H. Waldmann, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
10606–10611.
69 D. J. Newman, J. Med. Chem., 2008, 51, 2589–2599.
70 M. Anwar, A. R. Cowley and M. G. Moloney, Tetrahedron: Asymmetry,
2010, 21, 1758–1770.
71 B. Smith, S. C. Warren, G. G. F. Newton and E. P. Abraham, Biochem. J.,
1967, 103, 877–890.
72 J. E. Baldwin, A. J. Pratt and M. G. Moloney, Tetrahedron, 1987, 43,
2565–2575.
65 R. H. Baltz, J. Ind. Microbiol. Biotechnol., 2006, 33, 507–513.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3472–3485 | 3485