Two epimerisations in the formation of oxetanes from L-rhamnose: Towards oxetane-containing peptidomimetics

Article abstract of DOI:10.1016/S0957-4166(00)00360-8

The methyl group in (R)-3,5-O-benzylidene-L-rhamnono-1,4-lactone 2, prepared from L-rhamnose 1 in 41% yield without need for chromatography, is axial and provides a good example of Mills' rule that the 'O-inside' conformations are, in general, more stable than alternative 'H-inside' conformations. The lactone 2 may be converted in two steps in an overall 57% yield to the rhamnono-oxetane 3, which should be useful in generating oxetane dipeptide isosteres and oxetane β-amino acids and in determining the value of the oxetane ring in inducing secondary structure in small peptidomimetics. Methyl 2,4-anhydro-(S)-3,5-O-benzylidene-L-rhamnonate 11, in which both the phenyl and methyl groups are equatorial, is slightly more thermodynamically stable than 3, providing a rare exception to Mills' rule. X-ray crystal structures of 2 and 11 are reported. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00360-8

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 11 (2000) 4113–4125
Two epimerisations in the formation of oxetanes from
-rhamnose: towards oxetane-containing peptidomimetics
L
Stephen W. Johnson,a Donald Angus,a Claude Taillefumier,a John H. Jones,a
David J. Watkin,b Emma Floyd,b J. Grant Buchananc and George W. J. Fleeta,*
aDyson Perrins Laboratory, Oxford Centre for Molecular Sciences, South Parks Road, Oxford OX1 3QY, UK
bChemical Crystallography Department, Oxford University, Parks Road, Oxford OX1 3PD, UK
cDepartment of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK
Received 15 August 2000; accepted 11 September 2000
Abstract
The methyl group in (R)-3,5-O-benzylidene-
L
-rhamnono-1,4-lactone 2, prepared from -rhamnose 1 in
L
41% yield without need for chromatography, is axial and provides a good example of Mills’ rule that the
‘O-inside’ conformations are, in general, more stable than alternative ‘H-inside’ conformations. The
lactone 2 may be converted in two steps in an overall 57% yield to the rhamnono-oxetane 3, which should
be useful in generating oxetane dipeptide isosteres and oxetane b-amino acids and in determining the
value of the oxetane ring in inducing secondary structure in small peptidomimetics. Methyl 2,4-anhydro-
(S)-3,5-O-benzylidene- -rhamnonate 11, in which both the phenyl and methyl groups are equatorial, is
L
slightly more thermodynamically stable than 3, providing a rare exception to Mills’ rule. X-ray crystal
structures of 2 and 11 are reported. © 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction
Unnatural biopolymers1 that adopt well-defined conformations might be designed to possess
biological activities similar to those of proteins and peptides (foldamers).2 For example, b-amino
acids provide well-defined secondary structural features in sequences far shorter than those of
proteinogenic amino acids;3 a hetero-oligomer of some cyclopentane b-amino acids mimics
natural antibiotics.4 There are many potential clinical applications of such materials.5
Carbohydrate scaffolds offer a large family of monomers with the ability to tune predisposi-
tion towards specific secondary structures. Pyranose sugar amino acids have been shown to exert
predictable conformational effects in peptidic systems due to the rigid chain conformations
imposed in a glucose framework by the all-equatorial substituents; when such systems were
incorporated into cyclic hexapeptide analogues of somatostatin, a g-amino acid served as a
* Corresponding author.
0957-4166/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00360-8

4114
S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
b-turn mimetic while a b-amino acid induced a g-turn.6 A pyranose amino acid has been
incorporated as a dipeptide isostere in the design of a peptide which is a farnesyl transferase
inhibitor.7 The tetrahydrofuran (THF) ring is less flexible than many pyranoses. 5-Azidomethyl-
tetrahydrofuran-carboxylic acid scaffolds (building blocks for d-oligopeptides) provide a set of
dipeptide isosteres which are predisposed towards different secondary structures, depending on
the stereochemistry of the substituents around the THF ring;8 different diastereomers form
b-turn9 and left-handed helical10 solution structures. A THF sugar amino acid was incorporated
into Leu-enkephalin as a GlyꢀGly substitute to provide opioids equipotent with the natural
peptide; CD studies indicated that such carbopeptoids may have a b-turn-like structure.11 THF
amino acids have been incorporated into a functional cation channel with a biomimetic channel
entrance and exit; four THF amino acids continue the gramicidin b-helix.12
As yet, there are no reports of the incorporation of oxetane scaffolds into carbopeptides. For
such studies, short and efficient routes to a range of oxetane amino acids are necessary. This
paper describes the conversion of -rhamnose 1 into a crystalline benzylidene-protected lactone
L
2 on a multigram scale with no need for column separation (Scheme 1); the lactone 2 is
transformed in two steps to the protected oxetane 3 in an overall yield of 57%, even though the
ring closure takes place with overall retention of configuration at C-2 of the lactone. The new
stereogenic centre on 2 results in the methyl group being axial and provides a good example of
Mills’ rule13 that the ‘O-inside’ conformations are in general more stable than alternative
‘H-inside’ conformations. The protected oxetane 3 should act as a convenient divergent
intermediate for the formation of the diastereomeric dipeptide isosteres 4 and of the b-azido
acids 5. The X-ray crystal structures of the benzylidene lactone 2 and of the benzylidene oxetane
11 clearly show the C-6 methyl group in axial and equatorial environments, respectively.
Scheme 1.
2. Synthesis and discussion
2.1. Formation of the benzylidene acetal 2
Oxidation of -rhamnose 1 with bromine water gives a mixture of 1,4- and 1,5-rhamnono-lac-
L
tones;14 treatment of the crude mixture of lactones, still substantially contaminated with
inorganic salts, with benzaldehyde allowed the isolation of the highly crystalline benzylidene

S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4115
lactone 2 in an overall yield of 41% from -rhamnose 1 on a multigram scale. The structure of
L
2 was firmly established by X-ray crystallographic analysis (Fig. 1). There are two possible
epimeric benzylidene lactones in which the phenyl group is equatorial in the six-membered acetal
ring. The isolated R-2 product has the Me group axial, and so it might be expected to be
disfavoured relative to S-epi-2 in which both the phenyl and methyl groups are equatorial.
Figure 1. X-ray crystal structure of (R)-3,5-O-benzylidene-L-rhamnono-1,4-lactone 2 (grey=C, red=O, blue=H)
The conformation about the C-4 to C-5 bond is very different for the two diastereoisomers 2
and S-epi-2, and is shown in the Newman projections (Fig. 2). In the case of 2, both the H-4
and H-5 protons are located in equatorial positions and the observed vicinal coupling constant
J4,5=1 Hz is small. If S-epi-2 had been formed a much larger trans-diaxial coupling would have
been expected. The configuration of the benzylic carbon atom in benzylidene-protected sugar
derivatives was investigated as long ago as 1954.15 A review by Mills13 in 1955 discussed the
conformation of bicyclic diacetals with cis-ring junctions (as in this case) in terms not only of
the axial and equatorial positions of residues, but also in consideration of the possibilities of
‘H-inside’ and ‘O-inside’ conformations; Mills predicted that in general the ‘O-inside’ conforma-
tion is more stable than the ‘H-inside’ conformation, but that the situation may be clouded
somewhat by the nature of other substituents on the ring.

4116
S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
Figure 2.
Further work by Angyal,16 Foster and co-workers,17 and Buchanan18 extended and supported
these general conclusions. The formation of 2 with an axial methyl group in preference to an
alternative equatorial substituent in S-epi-2 is in accordance with Mills’ analysis.
2.2. Formation of the oxetane ring 3
The synthesis of the oxetane 3 required introduction of a good leaving group at C-2 of the
lactone 2.19 Esterification of the alcohol 2 with trifluoromethanesulfonic (triflic) anhydride in the
presence of pyridine in THF afforded the stable triflate 6 in 70% yield (Scheme 2). Treatment
of the triflate 6 with potassium carbonate in methanol allowed the formation of the oxetane 3
in a yield of 77% on a 1 g scale. This procedure gives efficient access to the benzylidene
derivative 3 in three steps from rhamnose.
Scheme 2. (i) (CF3SO2)2O, pyridine, THF; (ii) K2CO3, MeOH; (iii) HCl, MeOH; (iv) CF3COOCs, MeCOEt; (v)
PhCH(OMe)2, TsOH, DMF
Alternatively, reaction of the rhamnono-triflate 6 with caesium trifluoroacetate in butanone20
gave the epimeric -glucono-lactone 8 in 90% yield. Esterification of 8 with triflic anhydride gave
L

S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4117
the corresponding (R)-benzylidene triflate 9 in good yield, provided the work-up was conducted
rapidly; however, it was clear that the triflate 9 was unstable to silica chromatography, during
which it decomposed and, to some extent, unexpectedly formed the epimeric (S)-benzylidene
triflate 10. The combined yield and ratio of the two triflates 9 and 10 depended on the work-up
procedures (Fig. 3).
Figure 3.
During the triflation reaction of the glucono-lactone 8, an NMR of the crude product showed
that only the expected triflate 9 was present. Various ratios of the two triflates were found
depending on the time spent on the column, and in all cases purification by flash chromatogra-
1
phy reduced the yield considerably. The H NMR spectrum of 9 showed that J2,3=1.4 Hz and
J4,5=2.3 Hz. This is clearly consistent with none of the hydrogens being trans-diaxially arranged
in the (R)-epimer 9; in contrast, the spectrum of 10 gave J2,3=9.6 Hz and J4,5=10.1 Hz,
consistent with these sets of hydrogens being trans-diaxially arranged. Unsuccessful attempts
were made to equilibrate the two acetal epimers in acidic solution (acidified with TFA and/or
silica), and by dry loading the compound onto silica.
When the crude triflate 9 was treated with potassium carbonate in methanol, a mixture of the
(R)-benzylidene oxetane 3 (54%) and the (S)-benzylidene oxetane 11 (9%) was obtained.
Reaction of the relatively stable pure (S)-benzylidene triflate 10 under the same conditions gave
the pure (S)-benzylidene oxetane 11 in 75% yield without any of the epimeric oxetane 3 being
formed. The structure of the (S)-epimer 11 was firmly established by X-ray crystallographic
analysis (Fig. 4).
Methanolysis of the benzylidene acetal 3 with a solution of hydrogen chloride in methanol
gave the diol 7 in 54% yield. Reprotection was achieved by treatment of 7 with benzaldehyde
dimethyl acetal21 in DMF in the presence of tosic acid; analysis of the reaction mixture after 3
1
and 6 days by H NMR spectroscopy revealed in each case a 1:4 ratio for the PhCH singlet at
l 5.72 for (R)-3 relative to the PhCH singlet at l 6.16 for (S)-11. This evidence suggests that the
(S)-epimer 11 is more stable in the ‘H-inside’ conformation than the ‘O-inside’ (R)-epimer 3.
This may be because, with an oxetane ring rather than a pyran ring attached to the 1,3-dioxane,
the HꢀH steric interactions in the ‘H-inside’ conformation are much smaller (Fig. 5); as a result,
the axial versus equatorial considerations of the substituents play a much more dominant role
in deciding the overall stability. If this is the case, it is a rare example of change in the ring
cis-fused to the benzylidene ring causing a change in the conformational/configurational
preference of the molecule.16
The formation of the oxetane ring in 3 by an intramolecular SN2 reaction from the
rhamnono-lactone triflate 6 (with overall retention of configuration at C-2) is considerably more
efficient overall than from the epimeric triflate 9 (with inversion of configuration at C-2).
Oxetane-2-carboxylates are formed in high yield from carbohydrate lactone triflates provided

4118
S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
Figure 4. X-ray crystal structure of methyl 2,4-anhydro-(S)-3,5-O-benzylidene-
blue=H)
L-rhamnonate 11 (grey=C, red=O,
Figure 5.
that the oxygen substituent at C-3 is trans to the carboxylate at C-2.22 It may be that the
difficulty in SN2 reactions caused by a b-oxygen substituent means the late transition state for
the closure reflects the crowding in the incipient product; cyclisations which lack the neighbour-
ing oxygen may not show such strong preferences for 2,3-trans-oxetanes.23 For the efficient
formation of 3 from 6 (Scheme 3), ring opening of the lactone 6 by methanol would give the
open chain triflate 12, which does not close to 13; there was no evidence for the formation of
any of the all cis-substituted 13 in any of the experimental conditions reported in this paper.
Subsequent epimerisation of 12 to 14 would then allow ring closure to the 2,3-trans-oxetane 3.
Alternatively it is just possible that epimerisation of 6 to 9 precedes ring opening, giving 14
directly.

S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4119
Scheme 3.
3. Summary
This paper reports the formation of an easily crystallised benzylidene
L
-rhamnono-lactone 2
which may be prepared from -rhamnose 1 without any column chromatography; formation of
L
2 in which the C-6 methyl group is axial, rather than the alternative epimer in which it would
be equatorial, provides a good example of Mills’ rule13 that the ‘O-inside’ conformations are in
general more stable than alternative ‘H-inside’ conformations. The lactone 2 may be converted
in two steps in an overall 57% yield to the rhamnono-oxetane 3 in which the formation of the
oxetane occurs with overall retention of configuration at C-2 of the sugar moiety. The
benzylidene oxetane 11 in which both the phenyl and methyl groups are equatorial is slightly
more thermodynamically stable than 3, providing a rare exception to Mills’ rule. The oxetane 3
should be useful in generating oxetane dipeptide isosteres and oxetane b-amino acids in order to
determine the value of the oxetane ring in inducing secondary structure in small peptido-
mimetics.
4. Experimental
Dichloromethane was distilled from calcium hydride; pyridine was distilled from calcium
,
hydride and stored over a dried 4 A molecular sieve. THF and DMF were purchased dry from
the Aldrich Chemical Company in Sure-Seal™ bottles capped with Oxford-Caps™. Hexane
refers to the fraction of petroleum ether which boils in the range 60–80°C. All other solvents
were used as supplied (Analytical or HPLC grade), without further purification. Reactions
performed under an atmosphere of nitrogen were maintained by an inflated balloon. A buffer of
pH 7 was prepared by dissolving KH2PO4 (85 g) and NaOH (14.5 g) in distilled water (950 ml).
All other reagents were used as supplied, without further purification. Thin layer chromatogra-
phy (TLC) was performed on aluminium backed sheets coated with 60 F254 silica. Sheets were

4120
S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
developed using a spray of 0.2% w/v cerium (IV) sulfate and 5% ammonium molybdate in 2 M
sulfuric acid. Flash chromatography was performed using C60 40/60 silica. Melting points were
recorded on a Kofler hot block. NMR spectra were recorded on either a Bruker DPX 200 (1H:
200 MHz and 13C: 50.3 MHz) or a Bruker DPX 400 (1H: 400 MHz and 13C: 100.6 MHz)
spectrometer in the deuteriated solvent stated. All chemical shifts (l) are quoted in ppm and
coupling constants (J) in Hz. Residual signals from the solvents were used as an internal
reference. Infrared spectra were recorded on a Perkin–Elmer 1750 IR Fourier Transform
spectrophotometer using thin films on NaCl plates. High resolution mass spectra (HRMS m/z)
were recorded on a micromass Autospec 500 OAT spectrometer using the technique of chemical
ionisation (NH3, CI). Low resolution mass spectra were recorded on a VG PLATFORM (APCI,
positive or negative as stated). Optical rotations were recorded on a Perkin–Elmer 241
polarimeter with a path length of 1 dm. Concentrations are quoted in g/100 ml. The wavelength
at which the rotations were measured corresponds to the sodium D line. Elemental analyses
were performed either by the microanalysis service of the Inorganic Chemistry Laboratory,
Oxford, or by Elemental Microanalysis Limited, Okehampton, Devon. Except where otherwise
stated, crystalline compounds were obtained by concentration or complete evaporation of
eluants obtained from flash chromatography.
4.1. (R)-3,5-O-Benzylidene-
L
-rhamnono-1,4-lactone 2
Barium carbonate (86.6 g, 440 mmol) was added to a stirred solution of
L
-rhamnose
monohydrate 1 (40.2 g, 220 mmol) in water (300 ml) at 0°C. Bromine (17 ml, 330 mmol) was
added to the solution (3×5.67 ml portions at 15 min intervals). The reaction mixture was then
stirred at 0°C for 1 h, and then allowed to warm to room temperature. After 15 h at room
temperature, TLC (ethyl acetate:methanol, 9:1) revealed that the starting material (RF 0.21) had
been completely consumed, and replaced by a major product (RF 0.34) and a minor product (RF
0.48). The reaction mixture was then filtered through Celite, and the solid was washed with
water. Nitrogen was bubbled through the filtrate until the solution was decolourised, and the
solvent was removed. The solid was extracted with refluxing acetone (12×250 ml), and the
acetone was then removed to give a white solid (18.2 g), still containing inorganic impurities,
which was used without further purification. This crude lactone mixture (11.5 g) in benzaldehyde
(130 ml) was treated with concentrated hydrochloric acid (40 ml); the reaction mixture was
stirred at room temperature for 2.4 h, and TLC (ethyl acetate:hexane, 2:1) revealed the
formation of a major product (RF 0.23). Approximately half the solvent was then removed, and
ether added until the product crystallised out; the solid was collected by filtration and washed
with ether; concentration of the ether washings gave the crystalline product (12.3 g). The solvent
was then removed from the concentrated washings, and the residue purified by flash chromatog-
raphy (ethyl acetate:hexane, 2:1) to give further product (1.90 g). The crystalline material was
combined to give (R)-3,5-O-benzylidene- -rhamnono-1,4-lactone 2 as a white crystalline solid,
L
(14.3 g, 41%), m.p. 175–177°C; [h]D22 −48.7 (c 0.99 in MeCN); wmax (NaCl) 3353 cm−1 (OꢀH),
1775 cm−1 (CꢁO); lH (CD3CN, 200 MHz) 1.49 (d, 3H, H-6, J5,6=7.2 Hz), 3.85 (d, 1H, ꢀOH,
J2,OH=9.2 Hz), 4.13 (dd, 1H, H-4, J3,4=2.2 Hz, J4,5=1.0 Hz), 4.47 (q, 1H, H-5, J5,6=7.2 Hz),
4.56 (dd, 1H, H-2, J2,OH=9.1 Hz, J2,3=4.1 Hz), 4.79 (dd, 1H, H-3, J2,3=4.0 Hz, J3,4=2.1 Hz),
5.89 (s, 1H, PhCH), 7.37–7.44 (m, 5H, Ph); lC (CD3CN, 50.3 MHz) 15.2 (C-6), 70.5 (C-5), 72.5
(C-2), 72.6 (C-3), 73.6 (C-4), 92.4 (PhCH), 127.3, 129.1, 129.9 (Ph), 139.0 (Cipso), 176.2 (C-1).
Found: C, 62.13; H, 5.68; C13H14O5 requires: C, 62.39; H, 5.64%.

S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4121
4.2. (R)-3,5-O-Benzylidene-2-O-trifluoromethanesulfonyl- -rhamnono-1,4-lactone 6
L
Triflic anhydride (17 ml, 0.10 mol) was added dropwise with rapid stirring under an
atmosphere of nitrogen to a solution of the benzylidene lactone 2 (14.9 g, 59.5 mmol) and
pyridine (22.6 ml, 0.28 mol) in THF (200 ml) at −78°C. After 20 min, a white precipitate started
to form, and the reaction mixture was allowed to warm to 0°C for ca. 2 h. When no starting
material remained (TLC, ethyl acetate:hexane, 2:1) the mixture was diluted with
dichloromethane (300 ml), washed with 0.1 M hydrochloric acid (50 ml), water (70 ml), dried
(MgSO4) and reduced to dryness. The solvent was removed and the residue was purified by flash
chromatography (ethyl acetate:hexane, 1:2) to give the triflate 6 as a white solid (15.9 g, 70%),
m.p. 139–140°C; [h]2D2 −63.9 (c 0.98 in MeCN); wmax (NaCl) 1795.8 cm−1 (CꢁO); lH (CD3CN, 200
MHz) 1.52 (d, 3H, H-6, J5,6=7.4 Hz), 4.35 (dd, 1H, H-4, J3,4=2.1 Hz, J4,5=0.9 Hz), 4.58 (q,
1H, H-5, J5,6=7.4 Hz), 5.21 (dd, 1H, H-3, J2,3=3.8 Hz, J3,4=1.6 Hz), 5.83 (d, 1H, H-2, J2,3=4.2
Hz), 5.94 (s, 1H, CHPh), 7.41 (m, 5H, Ph); lC (CD3CN, 50.3 MHz) 14.8 (C-6), 70.3 (C-5), 71.4
(C-3), 74.1 (C-4), 81.5 (C-2), 92.4 (PhCH), 127.1, 129.3, 130.2 (Ph), 138.3 (Cipso), 169.0 (C-1).
Found: C, 43.85; H, 3.69; C14H13F3O7S requires: C, 43.98; H, 3.43%.
4.3. Methyl 2,4-anhydro-(R)-3,5-O-benzylidene- -rhamnonate 3
L
The triflate 6 (1.77 g, 4.62 mmol) in methanol (35 ml, HPLC grade) was cooled to −78°C and
potassium carbonate (702 mg, 5.1 mmol) was added. The reaction mixture was stirred for 10
min at −78°C, after which the −78°C bath was substituted for a −30°C bath. The temperature
was then allowed to warm up to −10°C, which took 20 min (TLC in ethyl acetate:hexane, 1:1
showed that both the starting material and the product had RF 0.48). The reaction mixture was
then filtered through a plug of silica gel; the silica was rinsed with ethyl acetate and the solvent
was removed to give a residue that was purified by flash chromatography (ethyl acetate:hexane,
1:2) to give the benzylidene oxetane 3 as a white solid (0.94 g, 77%), m.p. 115–116°C; [h]2D3 +3.0
(c 1.01 in CHCl3); wmax (NaCl) 1764 cm−1 (CꢁO); lH (CDCl3, 200 MHz) 1.38 (d, 3H, H-6,
J5,6=7.0 Hz), 3.83 (s, 3H, CO2Me), 4.43 (dq, 1H, H-5, J4,5=2.1 Hz, J5,6=7.1 Hz), 4.69 (ddd,
1H, H-4, J2,4=1.0 Hz, J3,4=4.9 Hz, J4,5=2.3 Hz), 4.93 (dd, 1H, H-3, J2,3=2.4 Hz, J3,4=5.0 Hz),
5.08 (dd, 1H, H-2, J2,3=2.5 Hz, J2,4=0.9 Hz), 5.72 (s, 1H, PhCH), 7.36–7.57 (m, 5H, Ph); lC
(CDCl3, 50.3 MHz) 16.2 (C-6), 52.7 (CO2CH3), 70.8 (C-5), 72.1 (C-3), 79.4 (C-4), 84.1 (C-2),
92.3 (PhCH), 126.3, 128.6, 129.2 (Ph), 138.3 (Cipso), 170.4 (C-1). Found: C, 63.68; H, 6.07;
C14H16O5 requires: C, 63.63; H, 6.10%.
This procedure is the most efficient for the formation of 3; other experiments in which the
oxetane 3 is formed are described below.
4.4. (R)-3,5-O-Benzylidene-6-deoxy- -glucono-1,4-lactone 8
L
Caesium trifluoroacetate (11.1 g, 45.5 mmol) was added to a solution of the rhamnono-triflate
6 (7.20 g, 18.8 mmol) in butanone (100 ml); the reaction mixture was heated for 2 h at 60°C
when TLC (ethyl acetate:hexane, 1:1) indicated that the starting material (RF 0.59) had been
replaced by a major product (RF 0.48). The solvent was removed, and the residue purified by
flash chromatography (ethyl acetate:hexane, 1:1) to give -glucono-1,4-lactone 8 (4.27 g, 90%) as
L
a white crystalline solid; m.p. 145–148°C; [h]2D2 −84.1 (c 1.15 in MeCN); wmax (NaCl) 3259.1 cm−1
(OꢀH), 1767.6 cm−1 (CꢁO); lH (CD3CN, 200 MHz) 1.51 (d, 3H, H-6, J5,6=7.2 Hz), 4.11 (s, 1H,

4122
S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
H-2), 4.45 (dd, 1H, H-4, J3,4=2.5 Hz, J4,5=1.3 Hz), 4.54 (dq, 1H, H-5, J4,5=0.6 Hz, J5,6=7.2
Hz), 4.63 (dd, 1H, H-3, J2,3=0.6 Hz, J3,4=2.6 Hz), 5.89 (s, 1H, PhCH), 7.35–7.41 (m, 5H, Ph);
lC (CD3CN, 50.3 MHz) 15.3 (C-6), 70.1 (C-5), 73.8 (C-2), 75.5 (C-3), 77.5 (C-4), 92.8 (PhCH),
127.1, 129.2, 130.0 (Ph), 139.0 (Cipso), 175.6 (C-1). Found: C, 62.10; H, 5.78; C13H14O5 requires:
C, 62.39; H, 5.64%.
4.5. (R)-3,5-O-Benzylidene-6-deoxy-2-O-trifluoromethanesulfonyl-
L
-glucono-1,4-lactone 9 and
(S)-3,5-O-benzylidene-6-deoxy-2-O-trifluoromethanesulfonyl- -glucono-1,4-lactone 10
L
Pyridine (0.40 ml, 5.0 mmol) was added to a solution of the glucono-lactone (0.24 g, 0.97
mmol) in THF (10 ml) under nitrogen at −78°C. Triflic anhydride (0.33 ml, 2.0 mmol) was
added dropwise, and the mixture was stirred at −78°C for a further 15 min, before warming to
0°C. After 3.75 h, TLC (ethyl acetate:hexane, 2:1) revealed that the starting material (RF 0.47)
had been partially replaced by one product (RF 0.7). The reaction mixture was recooled to
−78°C; more pyridine (0.08 ml, 1.0 mmol) and triflic anhydride (0.17 ml, 1.0 mmol) were added.
After 15 min, the reaction mixture was warmed to 0°C; TLC revealed that all the starting
material had been consumed after 40 min. The reaction mixture was diluted with
dichloromethane (20 ml), washed with 0.1 M hydrochloric acid (10 ml), and then water (10 ml).
The organic fraction was dried (MgSO4), filtered and the solvent was removed. The residue was
dissolved in ethyl acetate, filtered through a plug of silica, and the solvent was removed to give
the (R)-3,5-O-benzylidene triflate 9 (0.36 g, 98%), m.p. 70–72°C; wmax (NaCl) 1803 cm−1 (CꢁO);
lH (CD3CN, 200 MHz) 1.52 (d, 3H, H-6, J5,6=7.0 Hz), 4.57 (dq, 1H, H-5, J4,5=2.0 Hz,
J5,6=7.0 Hz), 4.64 (dd, 1H, H-4, J3,4:J4,5:2.6 Hz), 5.16 (dd, 1H, H-3, J2,3=1.4 Hz, J3,4=2.8
Hz), 5.32 (bs, 1H, H-2), 5.97 (s, 1H, PhCH), 7.38–7.49 (m, 5H, Ph); lC (CD3CN, 50.3 MHz)
15.5 (C-6), 69.4, 73.2, 78.0, 82.1 (C-2, C-3,C-4, C-5), 93.7 (PhCH), 127.1, 129.3, 130.3 (Ph),
138.4 (Cipso), 168.4 (C-1).
In a closely analogous procedure starting from 3,5-O-benzylidene-6-deoxy- -glucono-1,4-lac-
L
tone (0.20 g, 0.79 mmol), the crude material 9 was purified by flash chromatography (ethyl
acetate:hexane, 1:4) to give the 3,5-O-(R)-benzylidene triflate 9 (0.07 g, 24%) and the (S)-3,5-O-
benzylidene triflate 10 (0.014 g, 5%), m.p. 84–87°C; [h]2D2 −44.2 (c 0.995 in CHCl3); wmax (NaCl)
1806 cm−1 (CꢁO); lH (CD3CN, 200 MHz) 1.40 (d, 3H, H-6, J5,6=6.0 Hz), 4.08 (dq, 1H, H-5,
J4,5=10.0 Hz, J5,6=6.2 Hz), 4.45 (dd, 1H, H-4, J3,4=8.0 Hz, J4,5=10.2 Hz), 5.22 (dd, 1H, H-3,
J2,3=9.4 Hz, J3,4=8.0 Hz), 5.95 (s, 1H, PhCH), 6.28 (d, 1H, H-2, J2,3=9.8 Hz), 7.40–7.51 (m,
5H, Ph); lC (CD3CN, 50.3 MHz) 19.1 (C-6), 73.1 (C-4), 2×74.8 (C-3, C-5), 78.8 (C-2), 96.9
(PhCH), 127.1, 129.4, 130.4 (Ph), 137.5 (Cipso), 167.0 (C-1). Found: C, 43.58; H, 3.73;
C14H15F3O7S requires: C, 43.98; H, 3.43%.
Due to the instability of 9 and its partial decomposition and epimerisation to 10, the ratio of
the benzylidene derivatives 9 and 10, and yields of products, depend on the conditions under
which 9 is handled. In contrast 10 is relatively stable and easy to handle.
4.6. Methyl 2,4-anhydro-(R)-3,5-O-benzylidene-
L
-rhamnonate 3 and methyl 2,4-anhydro-(S)-
3,5-O-benzylidene- -rhamnonate 11
L
4.6.1. From the triflate of the (R)-benzylidene gluconolactone 9
Potassium carbonate (0.46 g, 3.32 mmol) was added to a solution of crude (R)-benzylidene
triflate 9 (0.63 g, 1.66 mmol) in methanol (14 ml, HPLC grade) at −78°C under nitrogen. The

S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4123
solution was stirred for 1.3 h, and then diluted with dichloromethane (150 ml). The reaction
mixture was washed with water (3×75 ml), the aqueous fractions combined and further extracted
with dichloromethane (75 ml). All the organic fractions were combined, dried (MgSO4), filtered
and the solvent removed. The residue was purified by flash chromatography (ethyl ace-
tate:hexane, 1:2) to give two products, the 3,5-O-(R)-benzylidene oxetane 3 (RF 0.46, 0.236 g,
54%) (for data see above), and the 3,5-O-(S)-benzylidene oxetane 11 (RF 0.33, 0.038 g, 9%), m.p.
62–63°C; [h]2D2 −33.9 (c 1.01 in CHCl3); wmax (NaCl) 1754.5 cm−1 (CꢁO); lH (CDCl3, 400 MHz)
1.42 (d, 3H, H-6, J5,6=6.4 Hz), 3.82 (s, 3H, CO2Me), 4.44 (dq, 1H, H-5, J4,5=6.2 Hz, J5,6=6.3
Hz), 4.55 (ddd, 1H, H-4, J2,4=1.0 Hz, J3,4=7.4 Hz, J4,5=6.2 Hz), 5.05 (dd, 1H, H-3, J2,3=6.2
Hz, J3,4=7.4 Hz), 5.53 (d, 1H, H-2, J2,3=6.0 Hz), 6.16 (s, 1H, PhCH), 7.36–7.41 (m, 3H, Ph),
7.50–7.53 (m, 2H, Ph); lC (CDCl3, 100.6 MHz) 20.17 (C-6), 52.57 (CO2CH3), 70.54 (C-3), 77.79
(C-5), 79.31 (C-4), 81.67 (C-2), 97.10 (PhCH), 126.32, 128.48, 129.39 (Ph), 137.27 (Cipso), 170.44
(C-1).
4.6.2. From the triflate of the (S)-benzylidene gluconolactone 10
Potassium carbonate (0.10 g, 0.72 mmol) was added to a solution of the (S)-benzylidene
triflate 10 (0.17 g, 0.45 mmol) in methanol (3.5 ml, HPLC grade) at −78°C under nitrogen. The
solution was stirred for 1 h, when TLC (ethyl acetate:hexane, 1:1) showed that the starting
material (RF 0.71) had been replaced by a major product (RF 0.57). The solution was diluted in
dichloromethane (40 ml), washed with water (2×20 ml), dried (MgSO4), filtered and the solvent
removed to give the 3,5-O-(S)-benzylidene oxetane 11 (0.088 g, 74%), identical to the material
above.
4.6.3. From the oxetane diol 7
Benzaldehyde dimethyl acetal (0.17 ml, 1.08 mmol) and tosic acid (2 mg) were added to a
solution of oxetane diol 7 (0.032 g, 0.18 mmol) in DMF (1 ml) under nitrogen at room
temperature. After 6 days, TLC (ethyl acetate:hexane, 1:1) revealed that some of the starting
material (RF 0.21) had been transformed into two products (RF 0.5 and 0.64). The solution was
neutralised over sodium carbonate, filtered and the solvent removed. The residue (0.050 g) was
1
shown by H NMR analysis to contain a 1:4 mixture of 3,5-O-(R)-benzylidene:3,5-O-(S)-ben-
zylidene methyl-2,4-anhydro-
L
-rhamnonate.
4.7. Methyl 2,4-anhydro- -rhamnonate 7
L
The benzylidene-protected oxetane 3 (0.417g, 1.58 mmol) was added to a solution of acetyl
chloride (0.16 ml) in methanol (16 ml) at room temperature. After 1 h, TLC (ethyl ace-
tate:hexane, 1:1) revealed the absence of starting material (RF 0.5) and the presence of a major
product (RF 0.25). The solution was neutralised with sodium carbonate, filtered, and the solvent
was removed. The crude material was purified by flash chromatography (ethyl acetate:hexane,
2:1), giving methyl-2,4-anhydro-
-rhamnonate (0.17 g, 54%); m.p. 52–54°C; [h]2D2 +84.6 (c 0.94 in
L
CHCl3); wmax (NaCl) 1738 cm−1 (CꢁO), 2956 cm−1 (CꢀH), 3412 cm−1 (OꢀH); lH (CDCl3, 200
MHz) 1.31 (d, 3H, H-6, J5,6=7.0 Hz), 3.15 (bs, 1H, OH (C-5)), 3.81 (s, 3H, CO2Me), 3.41 (bm,
1H, H-5), 4.52 (d, 1H, OH (C-3), J3,OH=10.0 Hz), 4.59 (ddd, 1H, H-4, J2,4=0.9 Hz, J3,4=6.8
Hz, J4,5=4.0 Hz), 4.83 (ddd, 1H, H-3, J3,OH=9.9 Hz, J2,3=5.1 Hz, J3,4=6.7 Hz), 5.03 (dd, 1H,
H-2, J2,3=4.8 Hz, J2,4=0.8 Hz); lC (CDCl3, 50.3 MHz) 17.5 (C-6), 52.6 (CO2CH3), 69.5 (C-5),
71.6 (C-3), 86.7 (C-2), 86.9 (C-4), 171.1 (C-1); MS (APCI+) m/z: 117 (100%), 177: (MH+, 30%).
Found: C, 47.24; H, 6.76; C7H12O5 requires: C, 47.73; H, 6.87%.

4124
S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4.8. X-ray crystal structure analysis
The relative configurations of the stereogenic centres in the lactone 2 and the oxetane 11 were
established by X-ray single crystal structure analysis. Cell dimensions and intensity data were
measured with an Enraf–Nonius Mach3 Diffractometer, and Lorentz, polarisation and psi scan
absorption corrections were applied. All calculations were carried out on a 486PC computer. All
25
non-hydrogen atoms were located by SIR-9224 and refined using CRYSTALS
.
Illustrations were
produced using CAMERON.
26 The hydroxyl hydrogen atom was located from a difference Fourier
map, all other hydrogen atoms were seen in the difference density map but placed geometrically.
Non-hydrogen atoms were refined anisotropically using atomic scattering factors from Interna-
tional Tables.27 Larsen extinction correction and Chebychev weighting were used. Structural
data for 2 and 11 have been deposited at the Cambridge Crystallographic Data Centre.28
4.8.1. Crystal data for lactone 2
Crystallised from ethyl acetate/hexane to give colourless prisms, C13H14O5, M 250.25. Mono-
3
,
,
clinic P21; a, 5.244(2); b, 9.360(5); c, 12.077(7) A; h, 90; i, 99.96(5); k, 90°; V, 583.8 A ; Z=2;
crystal size, 0.2×0.2×0.8 mm; v, 0.924 mm−1; diffractometer, Nonius MACH3; Cu–Ka, 1.54180
,
A; T, 190 K; scan type, 2q/V; h, −1 to 6; k, −3 to 11; l, −15 to 14. Refinement on F, 1209
−3
−3
,
,
reflections [I>3|(I)]; R, 0.064; wR, 0.080; S, 0.952; Dzmax, 0.32 (e A ); Dzmin, −0.30 (e A );
(D/s)max, 0.000161.
4.8.2. Crystal data for oxetane 11
Crystallised from ethyl acetate/hexane to give colourless blocks, C14H16O5; M, 264.28.
,
Orthorhombic P212121; a, 4.870(2); b, 10.189(5); c, 27.382(7) A; h, 90; i, 90; k, 90°; V, 1358.5
3
−1
,
A ; Z=4; crystal size, 0.5×0.9×0.9 mm; v, 0.821 mm ; diffractometer, Nonius MACH3;
,
Cu–Ka, 1.54180 A; T, 293 K; scan type, 2q/V; h, 0–6; k, 0–12; l, −34 to 0. Refinement on F,
1631 reflections [I>3|(I)]; R, 0.042; wR, 0.055; S, 0.855; Dzmax, 0.13 e A ; Dzmin, −0.18 e A ;
−3
−3
,
,
(D/s)max, 0.0013.
Acknowledgements
An EPSRC ROPA post-doctoral fellowship supported this work.
References
1. Barron, A. E.; Zuckerman, R. N. Curr. Opin. Chem. Biol. 1999, 3, 681; Kirshenbaum, K.; Zuckerman, R. N.;
Dill, K. A. Curr. Opin. Struct. Biol. 1999, 9, 530.
2. Gellmann, S. H. Acc. Chem. Res. 1998, 31, 173.
3. Fisk, J. D.; Powell, D.R.; Gellman, S.H. J. Am. Chem. Soc. 2000, 122, 5443; Chun, Y. J.; Huck, B. R.;
Christianson, L. A.; Stanger, H. E.; Krauthauser, S.; Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 2000, 122,
3995; Wang, X.; Espinosa, J. F.; Gellman, S. H. J. Am. Chem. Soc. 2000, 122, 4821; Appella, D. H.;
Christianson, L. A.; Karle, I. L.; Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 1996, 118, 13071; Appella,
D. H.; Barchi, J. J.; Durell, S. R.; Gellman, S. H. J. Am. Chem. Soc. 1999, 121, 2309; Seebach, D.; Matthews,
J. L. J. Chem. Soc., Chem. Commun. 1997, 2015.
4. Porter, E. A.; Wang, X.; Lee, H.-S.; Weisblum, B.; Gellman, S. H. Nature 2000, 404, 565.

S. W. Johnson et al. / Tetrahedron: Asymmetry 11 (2000) 4113–4125
4125
5. Werder, M.; Hauser, H.; Abele, S.; Seebach, D. Helv. Chim. Acta 1999, 82, 1774.
6. von Roedern, E. G.; Lohof, E.; Hessler, G.; Hoffmann, M.; Kessler, H. J. Am. Chem. Soc. 1996, 118, 10156.
7. Overkleeft, H. S.; Verhelst, S. H. L.; Pieterman, E.; Meeuweno-ord, N. J.; Overhand, M.; Cohen, L. H.; van de
Marcel, G. A.; van Boom, J. H. Tetrahedron Lett. 1999, 40, 4103.
8. Smith, M. D.; Fleet, G. W. J. J. Pept. Sci. 1999, 5, 425.
9. Smith, M. D.; Claridge, T. D. W.; Tranter, G. E.; Sansom, M. S. P.; Fleet, G. W. J. J. Chem. Soc., Chem.
Commun. 1998, 2041; Hungerford, N. L.; Claridge, T. D. W.; Watterson, M. P.; Aplin, R. T.; Moreno, A. M.;
Fleet, G. W. J. J. Chem. Soc., Perkin Trans. 1 2000, in press.
10. Long, D. D.; Hungerford, N. L.; Smith, M. D.; Brittain, D. E. A.; Marquess, D. G.; Claridge, T. D. W.; Fleet,
G. W. J. Tetrahedron Lett. 1999, 40, 2195.
11. Chakraborty, T. K.; Jayaprakash, S.; Diwan, P. V.; Nagaraj, R.; Jampani, S. R. B.; Kunwar, A. C. J. Am. Chem.
Soc. 1998, 120, 12962.
12. Schrey, A.; Vescovi, A.; Knoll, A.; Rickert, C.; Koert, U. Angew. Chem., Int. Ed. 2000, 39, 900.
13. Mills, J. A. Adv. Carbohydr. Chem. 1955, 10, 1.
14. Jackson, E. L.; Hudson, C. S. J. Am. Chem. Soc. 1930, 52, 1270; Mantell, S. J.; Fleet, G. W. J.; Brown, D. J.
Chem. Soc., Chem. Commun. 1991, 1563.
15. Buchanan, J. G. Chem. Ind. 1954, 1484.
16. Angyal, S. J.; Kondo, Y. Carbohydr. Res. 1980, 81, 35.
17. Baggett, N.; Duxbury, J. M.; Foster, A. B.; Webber, J. M. Carbohydr. Res. 1965, 1, 22; Baggett, N.;
Mosihuzzaman, M.; Webber, J. M. Carbohydr. Res. 1969, 11, 263.
18. Chamberlain, N. L.; Edwards, I. A. S.; Stadler, H. P.; Buchanan, J. G.; Thomas, W. A. Carbohydr. Res. 1981,
90, 131.
19. Fleet, G. W. J.; Austin, G. N.; Peach, J. M.; Prout, K.; Son, J. C. Tetrahedron Lett. 1987, 28, 4741.
20. Bell, A. A.; Pickering, L.; Finn, M.; De la Fuente, C.; Krulle, T. M.; Davis, B. G.; Fleet, G. W. J. Synlett 1997,
1077.
21. Brockway, C.; Kocienski, P.; Pant, C. J. Chem. Soc., Perkin Trans. 1 1984, 875.
22. Witty, D. R.; Fleet, G. W. J.; Vogt, K.; Wilson, F. X.; Wang, Y.; Storer, R.; Myers, P. L.; Wallis, C. J.
Tetrahedron Lett. 1990, 31, 4787; Wang, Y.; Fleet, G. W. J.; Wilson, F. X.; Storer, R.; Myers, P. L.; Wallis, C.
J.; Doherty, O.; Watkin, D. J.; Vogt, K.; Witty, D. R.; Peach, J. M. Tetrahedron Lett. 1991, 32, 1675 and
references cited therein.
23. Izawa, T; Nakayama, K; Nishiyama, S; Yamamura, S.; Kato, K; Takita, T. J. Chem Soc., Perkin Trans. 1 1992,
3003; Witty, D. R.; Fleet, G. W. J.; Choi, S.-S.; Vogt, K.; Wilson, F. X.; Wang, Y.; Storer, R.; Myers, P. L.;
Wallis, C. J. Tetrahedron Lett. 1990, 31, 6927.
24. Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Polidori, G. J. Appl.
Cryst. 1994, 27, 435.
25. Watkin, D. J.; Prout, C. K.; Carruthers, J. R.; Betteridge, P. W. CRYSTALS Issue 10; Chemical Crystallography
Laboratory: University of Oxford, Oxford, 1996.
26. Watkin, D. J.; Prout, C. K.; Pearce, L. J.; CAMERON, Chemical Crystallography Laboratory: University of
Oxford, Oxford.
27. International Tables for Crystallography; Kluwer Academic: Dordrecht, 1992; Vol. C.
28. The atomic coordinates for 2 and 11 are available on request from the Cambridge Crystallographic Data Centre,
University Chemical Laboratory, Lensfield Road, Cambridge CB2 1EW; the crystallographic numbering system
differs from that used elsewhere in the text. Any request should be accompanied by the full literature citation for
this paper.
.