Spirodiketopiperazines of mannofuranose: Carbopeptoid α-amino acid esters at the anomeric position of mannofuranose

Article abstract of DOI:10.1016/S0957-4166(98)00206-7

Epimeric mannofuranose anomeric aminoesters are prepared from readily available azidolactones and can act as building blocks for the incorporation of mannofuranose units into peptide chains [carbopeptoids]; alternative acylating conditions allow either ra

Full text of DOI:10.1016/S0957-4166(98)00206-7

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 2137–2154
Spirodiketopiperazines of mannofuranose: carbopeptoid α-amino
acid esters at the anomeric position of mannofuranose
Juan C. Estevez,c Jonathan W. Burton,a Ramon J. Estevez,a,c Helen Ardron,a
Mark R. Wormald,b Raymond A. Dwek,b David Brown d and George W. J. Fleet a,∗
aDyson Perrins Laboratory, University of Oxford, South Parks Road, Oxford OX1 3QY, UK
bGlycobiology Institute, Biochemistry Department, Oxford University, South Parks Road, Oxford OX1 3QU, UK
cDepartamento de Química Orgánica, Universidade de Santiago de Compostela, e Sección de Alcaloides do C.S.I.C.,
15706 Santiago de Compostela, Spain
dGlaxoWellcome Medicines Research Centre, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY, UK
Received 25 April 1998; accepted 15 May 1998
Abstract
Epimeric mannofuranose anomeric aminoesters are prepared from readily available azidolactones and can act
as building blocks for the incorporation of mannofuranose units into peptide chains [carbopeptoids]; alternative
acylating conditions allow either rapid acylation of the more stable but kinetically hindered amine or reaction with
the less hindered but less stable amine to allow control of the anomeric configuration of the products. This is
exemplified by coupling of the aminoesters with glycine derivatives to give dipeptide equivalents, and subsequent
cyclization to spiro diketopiperazines. Anomers with the nitrogen function cis to the 2,3-diol are more stable than
those with nitrogen trans; mannofuranose derivatives are more stable than the mannopyranose isomers. © 1998
Published by Elsevier Science Ltd. All rights reserved.
1. Introduction
It has been suggested that pyranose amino acid fragments exemplified by 1 can be used as peptide-bond
linked carbohydrates (carbopeptoids) to provide extra diversity and carbohydrate structure in the search
for ligand lead and drug discovery.1 The same pyranose amino acid moiety 1 has been used as a flexible
β-turn mimic and other aminopyranose carboxylates may be able to act as dipeptide isosteres, perhaps
inducing β- and γ-turns.2 Pyranose amino acid fragments such as 2 have also been used in the synthesis of
peptide sugar hybrids.3 Other pyranose derivatives have been studied as novel oligonucleotide backbone
analogues,4 building blocks for combinatorial synthesis,5 and other peptide–carbohydrate hybrids.6 Some
∗
Corresponding author. E-mail: george.fleet@chem.ox.ac.uk
0957-4166/98/$19.00 © 1998 Published by Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00206-7

2138
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
reports of solid-phase synthesis using such monomers have appeared.7 Very few tetrahydrofuran related
amino acids have been reported,8 although the template properties of such materials would appear to
constitute at least as interesting a set of materials as those derived from pyranoses, in terms of both
structural diversity and conformational properties.
Amino acid moieties at the anomeric positions of sugars such as the mannofuranose aminoesters 3
are an example of a family of novel carbohydrate structures which could be incorporated into amide
libraries 4. Such structures are α,α-disubstituted amino acids and thus may induce secondary structures
in relatively short sequences; they are also attractive components for peptide library synthesis in that
they provide a novel set of amino acids with hydroxylic components. Additionally, such fragments
allow the generation of families of materials with the specific structural information of an individual
carbohydrate recognition site. We have reported in a number of preliminary and other communications
the synthesis of glucose,9 rhamnose,10 galactose11 and mannose12 analogues with an α-amino acid
moiety at the anomeric position of the sugar. The only other reports of such compounds are a different
approach to the synthesis of mannose components such as 4 by Dondoni,13 and in the synthesis of
aminosialic acid analogues.14 Such structures may be viewed as α-amino-C-glycosides of hexoses, so
that a seven carbon sugar is required for their synthesis; the readily available lactone 515 has been widely
used for the synthesis of mimics of mannose16 and for the synthesis of other highly functionalised
compounds.17 Many of these syntheses depend on the efficient conversion of 5 to the epimeric azides
6;18 this paper reports the conversion of the azides 6 to the epimeric amino esters 7, coupling of the
amines to a protected glycine to give the protected carbopeptoids 8, and their subsequent elaboration to
the spirodiketopiperazines 9. Some of this material has been published in a preliminary form.19
2. Results and discussion
The synthesis of the protected amines 7 is shown in Scheme 1. Hydrogenation of the azides 6β and 6α
in ethyl acetate in the presence of palladium black gave the amines 10β and 10α in 98% and 94% yields,
respectively. Oxidation of 6β by bromine in methanol, followed by addition of triethylamine, gave the
epimeric amines 7α and 7β [60% and 12% yields, respectively] together with the nitrile 11 [17% yield],
giving a combined yield of isolated oxidation products of 89%. Similar treatment of the epimeric amine
6α gave the same three products in the same proportions.
A possible mechanism for this novel oxidative ring contraction is shown in Scheme 2 [there is no
direct evidence for the existence of 13, 14 and 15 as discrete intermediates]. Initial bromination of 10
might give 12 from which base catalysed elimination of HBr would form the imine 13. It is possible that
the initial bromination occurs at C-2 of the lactone, but overall substitution of H+ by Br+ followed by

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2139
Scheme 1. (i) H2, Pd black, EtOAc; (ii) Br2, Et3N, AcONa, MeOH
HBr elimination would also form 13. The formation of the same proportion of products from the epimeric
amines 10 might then arise from the common intermediate 13, capable of forming all three of the isolated
products. Ring opening of iminolactone 13 by methanol would give an open chain hydroxyimine 14
which closes to form the epimeric amino esters 7α and 7β. Alternatively, further oxidation of the imine
13 by bromine would give the bromoimine 15. Subsequent base-induced nucleophilic attack on 15 would
then result in fragmentation to form the isolated nitrile 11, rather than in ring opening.
Scheme 2. Possible pathway for the formation of aminoesters 7 and nitrile 11
Both epimeric amino esters 7 can be obtained in pure form; these provided the first examples of
stable amino esters at the anomeric position of sugars. Their easy isolation probably owes much to the
level of protection of the carbohydrate, and the use of non-nucleophilic solvents to study the materials.
The 7β:7α ratio of 4.8:1 isolated in the reaction appears to be close to the equilibrium position; in
non-protic solvents, the equilibration is slow enough to allow the isolation of the individual anomers.
The major epimer 7β is less sterically hindered as the amino group is cis to the acetonide/diol group,
whereas with the minor epimer 7α the larger methyl ester group is more crowded. The assignment of
the anomeric configuration of the structures is based on the NMR studies described later in the paper
and is consistent with the evidence from the crystal structure and NMR data of related mannofuranose
spirohydantoins.20,21
The epimeric amino esters 7 equilibrate even in non-polar solutions; all attempts to isolate deprotected
equivalents of the aminoesters were unsuccessful. The open chain imine would be susceptible to nucleo-
philic attack to give a variety of products. However, acylation of the amines should give anomeric amides

2140
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
which are likely to be much more stable; electron-withdrawing groups stabilise tetrahedral adducts,
relative to trigonal C=X systems. Thus N-acylpyranosylamines are stable and isolable compounds which
do not easily anomerise and are not subject to ready hydrolysis at the anomeric position. This is even
more likely to be the case for N-acylated amino esters, such as 8 for example, in which both the N-acyl
functionality and the ester are both electron withdrawing groups and will thermodynamically destabilise
the open chain imine equivalent against the tetrahedral adducts. Such substitution will also kinetically
inhibit any equilibration process.
The less stable aminoester 7α has the less hindered amine, and it was found that alternative acylating
conditions could control the anomeric configuration of the product amide [Scheme 3]. Thus acylation
of the thermodynamically more stable anomer 7β with Z-glycine [Z=benzyloxycarbonyl], dicyclohexyl-
carbodiimide [DCC] and 1-hydroxybenzotriazole [HBT] in DMF gave 8α as the major product [72%
yield] with a small amount of 8β [8% yield]. A similar product ratio was obtained from the coupling of
the less stable aminoester 7α under the same conditions giving 8α [71% yield] and 8β [6% yield]. It thus
appears that, under these conditions, the amino esters equilibrate more rapidly than they are acylated, and
that the less hindered — but less stable — amine 7α is acylated faster. Similar results were obtained when
BOC-glycine [BOC=tert-butyloxycarbonyl] was coupled with 7β under the same conditions giving 16α
in 75% yield but 16β in only 8% yield, whereas with 7α, 16α was produced in 72% yield and 16β in
7% yield.
Scheme 3. (i) Z-Glycine or BOC-glycine, DCC, 1-hydroxybenzotriazole, DMF; (ii) Z-glycine, ClCO2Et, Et3N, CH2Cl2; (iii)
H2, Pd black, MeOH; (iv) tBuOK, THF
In marked contrast, equilibration of the anomeric amines 7 does not compete with the peptide coupling
when Z-glycine is activated as the mixed anhydride with ethyl chloroformate. Under these conditions,
where the more stable 7β is coupled, the major product is 8β [56% yield] with only traces of 8α being
formed. Thus the coupling reaction now occurs faster than equilibration to the more reactive anomeric
amine 7α. In no case was there any evidence to indicate that any of the coupled products underwent any
equilibration under the reaction conditions.
The epimeric Z-protected dipeptides 8 were converted to the corresponding diketopiperazines 18
without any equilibration of the anomers taking place. Hydrogenation of the Z-protecting group in 8β
gave the diketopiperazine 18β in 95% yield, indicating that the initially formed amine 17β spontaneously
cyclises. In contrast, hydrogenation of the anomeric Z-derivative 8α afforded the amine 17α isolated in
91% yield. The cyclisation of 17α requires attack by the amine on a very hindered ester carbonyl group;
treatment of 17α with potassium tert-butoxide in tetrahydrofuran induces clean cyclisation to 18α [89%
yield] with no contamination by anomeric equilibration.

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2141
All these acylated derivatives of 7 are stable to any epimerisation at the anomeric position under a wide
variety of conditions; thus the side chain acetonide can be removed from 8α and 16α by acetic acid in
methanol to form 19α and 20α in yields of 80% and 82% respectively. The diols 19α and 20α were
reprotected by reaction with 2,2-dimethoxypropane and acetone in the presence of p-toluenesulfonic acid
to give 8α and 16α in quantitative yields.
No anomeric equilibration of the diketopiperazine moiety occurred under a wide range of treatments
with acid. The diketopiperazines 18α and 18β can be partially deprotected by reaction with acetic acid in
methanol to afford 21α and 21β in 88% and 87% yields, respectively [Scheme 4]. Treatment of 21α and
21β with aqueous trifluoroacetic acid resulted in complete deprotection to 9α and 9β in 92% and 80%
yields. Furthermore the fully 9α — and partially 21α — deprotected furanoses can be converted to 18α
in excellent yields; 9β and 21β can be similarly deprotected to 18β. No anomeric equilibration — nor
any isomerisation to mannopyranose isomers — took place in any of these reactions, implying that there
is no ring opening of the furanose ring under these conditions.
Scheme 4. (i) tBuOK, DMF; (ii) AcOH:MeOH (1:1); (iii) Me2C(OMe)2, Me2CO, p-toluenesulfonic acid; (iv) CF3COOH:H2O
(1:1)
The relative stability of epimeric diketopiperazines was indicated by a remarkable series of equilibra-
tions of the α-anomers to the β-anomers by treatment with potassium tert-butoxide in dimethylform-
amide at 100°C. Thus equilibration under these conditions caused epimerisation of the fully protected
diacetonide 18α to 18β in 88% yield, of the monoacetonide 21α to 21β in 85% yield, and of the fully
protected diacetonide 9α to 9β in 90% yield. These experiments clearly demonstrate that the β-anomers
are thermodynamically more stable than the α-isomers. Moreover, although in 21 and 9 ring opening
and closing takes place there is no capture of the intermediate imines by the side-chain secondary
hydroxyl group to give mannopyranose isomers, and this strongly suggests that mannofuranose isomers
are thermodynamically more stable than mannopyranose isomers under these conditions.

2142
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
1
Fig. 1. Lower trace: the 1D H NMR spectrum of 8α, showing the resonance assignments. Upper trace: a trace through the
phase-sensitive 2D NOESY spectrum, 400 ms mixing time, at 7.17 ppm (corresponding to the Man NH resonance)
The anomeric configurations of 8, 9, 18 and 21 were established by NMR studies. The 1D NMR
spectra of 8α and 8β are shown in Figs 1 and 2, respectively.
In both compounds, an NOE is observed from the Man NH resonance to the Man C2H resonance. This
NOE is much larger in 8α than 8β. The average Man NH to Man C2H distances were calculated from
the NOE build-up curves by using the C2H to C4H NOE as an internal calibration (Table 1). This gives a
value of 2.4 Å for 8α and 3.8 Å for 8β. Figure 3 shows the variation in NH–C2H distance with rotation
around the C1–N bond for the two anomeric configurations of 8 (either cis or trans H–C2–C1–N). A
distance of 2.4 Å can only be obtained for the cis form, whilst 3.8 Å can only be obtained for the trans
form. Thus, in 8α the C2H proton is on the same side of the ring as the NH group (cis) whilst in 8β they
are on different sides (trans). This is confirmed by the observation of strong NOEs from the Man NH
resonance to two of the protecting group methyl resonances for 8β (Fig. 2).
The 1D NMR spectra of 9α and 9β are shown in Figs 4 and 5. In both spectra, four hydroxyl proton
resonances can be identified. For 9α, the hydroxyl resonances can be assigned to C2, C3, C5 and C6
confirming the presence of a furanose ring. A strong NOE is observed between N1H and C2H for 9α
(Fig. 4, giving an inter-proton distance of 2.7 Å), but not for 9β (Fig. 5). The N1H to C2H distances
determined by molecular modelling for the two epimers of 9 are 2.5 Å for cis and 3.9 Å for trans

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2143
1
Fig. 2. Lower trace: the 1D H NMR spectrum of 8β, showing the resonance assignments. Upper trace: a trace through the
phase-sensitive 2D NOESY spectrum, 400 ms mixing time, at 7.18 ppm (corresponding to the Man NH resonance)
Table 1
Calculated inter-proton distances for 8α and 8β based on NOE cross-peak intensity. The Man
C2H–C4H distances, used as the internal calibration, were determined by molecular modelling to
be 3.25 Å for cis H–C2–C1–N and 3.44 Å for trans H–C2–C1–N. The choice of which value to use
has little effect on the calculated distances for NH to C2H. The C2H–NH distances determined by
molecular modelling are shown in figure c as a function of the C2–C1–N–H torsion angle
H–C2–C1–N1. This confirms the cis H–C2–C1–N1 configuration for 9α and the trans configuration
for 9β.
The anomeric configurations of 18α, 18β, 21α, and 21β, (data not shown) were determined in an
identical fashion to 9α and 9β, giving cis H–C2–C1–N1 configurations for 18α and 21α and trans
H–C2–C1–N1 configurations for 18β and 21β.
It is thus clear that N-acylated derivatives of mannofuranose structures containing an α-amino acid
moiety at the anomeric position are stable configurationally other than under strongly basic or acidic
conditions; additionally, the difference in reactivity between the anomeric amino esters 7 allows control
of the configuration of the resulting acylated materials. Such compounds are suitable candidates for the
generation of amide libraries at the anomeric positions of carbohydrates.

2144
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
Fig. 3. Plots of NH to C2H distance (Å) versus torsion angle (rotation around the C1–N bond) for the two configurations of 8.
Lower line is for cis H–C2–C1–N (8α), upper line is for trans H–C2–C1–N (8β)
1
Fig. 4. Lower trace: the 1D H NMR spectrum of 9α, showing the resonance assignments. Upper trace: a trace through the
phase-sensitive 2D NOESY spectrum, 400 ms mixing time, at 4.25 ppm (corresponding to the C2H resonance). The negative
peaks are chemical exchange peaks involving the C7OH resonance

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2145
1
Fig. 5. Lower trace: the 1D H NMR spectrum of 9β, showing the resonance assignments. Upper trace: a trace through the
phase-sensitive 2D NOESY spectrum, 400 ms mixing time, at 4.61 ppm (corresponding to the C2H resonance). The negative
peaks are chemical exchange peaks involving the broad OH resonance at 4.5 ppm
3. Experimental
Melting points were recorded on a Kofler block and are corrected. Proton nuclear magnetic resonance
(δH) spectra were recorded on a Bruker AM 500 (500 MHz) spectrometer, or where stated, on a Varian
Gemini 200 (at 200 MHz). Carbon nuclear magnetic resonance (δC) spectra were recorded, on a Varian
Gemini 200 (at 50.3 MHz) or, where stated, on a Bruker AM 500 (125 MHz) spectrometer; multiplicities
were assigned using DEPT sequence. All chemical shifts are quoted on the δ-scale using residual solvent
as an internal standard. For the NOE studies, 1H NMR spectra on compounds 8α and 8β in CDCl3, 9α,
9β, 21α and 21β in DMSO and 18α and 18β in CD3CN were recorded on a Varian Unity 500, with a
probe temperature of 30°C. Resonance assignments were obtained from the 1D and phase-sensitive 2D
COSY spectra, referenced to the residual solvent signal (7.20 ppm for CDCl3, 2.49 ppm for DMSO or
1.93 ppm for CD3CN). In all cases except 9α, the carbon backbone could be followed unambiguously.
1
Unambiguous assignments for 9α were obtained using H–13C correlations from HMQC and HMBC
spectra. The only remaining ambiguities were in the stereo-specific assignments of the protecting group
methyl resonances, which could be determined by the pattern of NOEs between these resonances and the
mannose ring proton resonances. Phase-sensitive 2D NOESY spectra were recorded with mixing times
of 100 to 800 ms without any random variation. Inter-proton distances were determined from the NOE
cross-peak volumes using the two-spin approximation as previously described.22 Molecular modelling
was performed on a Silicon Graphics Indigo 2 workstation using Insight II and Discover software (MSI).
The variation in internuclear distance with torsion angle was determined as previously described.23
Infrared spectra were recorded on a Perkin–Elmer 1750 FT-IR spectrophotometer. Mass spectra were
recorded on VG Micromass 20-250, ZAB 1F or Trio-1 GC–MS (DB-5 column) spectrometers using
desorption chemical ionisation (DCI, NH3), chemical ionisation (CI, NH3), fast atom bombardment
(FAB) or electrospray techniques, as stated. Optical rotations were measured on a Perkin–Elmer 241
polarimeter with a path length of 1 dm. Concentrations are given in g/100 ml. Hydrogenations were

2146
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
executed at atmospheric pressure under an atmosphere of hydrogen gas maintained by an inflated balloon.
Microanalyses were performed by the microanalysis service of the Dyson Perrins Laboratory. Thin layer
chromatography (t.l.c.) was carried out on aluminium sheets coated with 60F254 silica or glass plates
coated with silica Blend 41. Plates were developed using a spray of 0.2% w/v cerium(IV) sulphate
and 5% ammonium molybdate in 2 M sulphuric acid or 0.5% ninhydrin in methanol (for amines).
Flash chromatography was carried out using Sorbsil C60 40/60 silica. Solvents and commercially
available reagents were dried and purified before use according to standard procedures; in particular,
dichloromethane was refluxed over, and distilled from, calcium hydride, methanol was purchased dry
in Aldrich Sure-SealTM bottles, pyridine was distilled from, and stored over, potassium hydroxide,
tetrahydrofuran was distilled, under nitrogen, from a solution dried with sodium in the presence of
benzophenone. Hexane and ethyl acetate were distilled before use to remove involatile fractions. All
solvents were removed in vacuo. The azidolactones 6 were prepared as previously described.18
3.1. 2-Amino-2-deoxy-3,4:6,7-di-O-isopropylidene-D-glycero-D-galacto-heptono-1,5-lactone 10α
A solution of azidolactone 6α (1.01 g, 3.2 mmol) in ethyl acetate (10 ml) was stirred under an
atmosphere of hydrogen in the presence of a catalytic amount of palladium black (ca. 8 mg). After 24
h the mixture was filtered through Celite. Crystallisation of the crude product by (ethyl acetate/hexane)
25
yielded aminolactone 10α as a white crystalline solid (900 mg, 98%), m.p. 124–127°C; [α]D +87.6
(c 0.8, CHCl3); νmax (KBr): 3363, 3304 (NH), 1756 cm−1 (C_O); δH (CDCl3): 1.38, 1.40, 1.43, 1.46
0
(12H, 4×s, 2×C(CH3)2), 3.96 (1H, br d, H-2), 4.12 (1H, dd, J6,7 4.2 Hz, J7,7 9.1 Hz, H-7), 4.16 (1H, dd,
J6,7 6.0 Hz, J7 ,7 9.1 Hz, H-70), 4.41 (2H, m, J5,6 8.4 Hz, J6,7 4.2 Hz, J6,7 6.0 Hz, H-6, H-5), 4.64 (1H,
dd, J3,4 1.9 Hz, J4,5 7.5 Hz, H-4), 4.93 (1H, dd, J2,3 8.6 Hz, J3,4 1.9 Hz, H-3); δC (CDCl3): 24.0, 25.0,
25.9, 26.9 (4×q, C(CH3)2), 54.8 (d, C-2), 66.7 (t, C-7), 70.9, 73.0, 76.1, 76.4 (4×d, C-3, C-4, C-5, C-6),
0
0
0
+
109.9, 110.0 (2×s, C(CH3)2), 170.1 (s, C-1); m/z (CI, NH3): 288 (M+H+, 100) 305, (M+NH4 , 30%).
Found: C, 54.67; H, 7.12; N, 4.62. C13H21O6N requires: C, 54.35; H, 7.37; N, 4.88%.
3.2. 2-Amino-2-deoxy-3,4:6,7-di-O-isopropylidene-D-glycero-D-talo-heptono-1,5-lactone 10β
A solution of azidolactone 6β (1.04 g, 3.3 mmol) in ethyl acetate (20 ml) was stirred under an
atmosphere of hydrogen in the presence of a catalytic amount of palladium black (ca. 8 mg). After
24 h the reaction mixture was filtered through Celite to give after crystallisation (ethyl acetate/hexane)
aminolactone 10β as a white crystalline solid (890 mg, 94%), m.p. 145–147°C (ethyl acetate/hexane);
[α]D25 +56.2 (c 1.1, CHCl3); νmax (KBr): 3374, 3319 (NH), 1752 cm−1 (C_O); δH (CDCl3): 1.38, 1.39,
1.43, 1.46 (12H, 4×s, 2×C(CH3)2), 3.60 (1H, br d, sharp in D2O, H-2), 4.03 (1H, dd, J4,5 1.7 Hz, J5,6
8.3 Hz, H-5), 4.08 (1H, dd, J6,7 3.9 Hz, J7,7 9.2 Hz, H-7), 4.15 (1H, dd, J6,7 6.1 Hz, J7 ,7 9.3 Hz, H-70),
0
0
0
0
4.41 (1H, ddd, J5,6 8.2 Hz, J6,7 3.9 Hz, J6,7 6.0 Hz, H-6), 4.69 (1H, dd, J3,4 7.8 Hz, J4,5 1.6 Hz, H-4),
4.77 (1H, dd, J2,3 2.9 Hz, J3,4 7.9 Hz, H-3); δC (CDCl3): 24.1, 24.8, 25.7, 26.8 (4×q, C(CH3)2), 53.3 (d,
C-2), 66.5 (t, C-7), 72.4, 72.6, 75.7, 76.4 (4×d, C-3, C-4, C-5, C-6), 109.9, 110.4 (2×s, C(CH3)2), 172.9
(s, C-1); m/z (CI, NH3): 288 (M+H+, 100%). Found: C, 54.31; H, 7.45; N, 4.53. C13H21O6N requires: C,
54.35; H, 7.37; N, 4.88%.

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2147
3.3. 2,3:5,6-Di-O-isopropylidene-4-methoxycarbonyl-mannononitrile 11, methyl 2-amino-2,5-anhydro-
3,4:6,7-di-O-isopropylidene-D-glycero-D-talo-heptonate 7β and methyl 2-amino-2,5-anhydro-3,4:6,7-
di-O-isopropylidene-D-glycero-D-galacto-heptonate 7α
Bromine (0.42 ml, 8.2 mmol) was added to a stirred solution of aminolactone 10β (2.36 g, 8.2
mmol) and sodium acetate (2.02 g, 24.6 mmol) in anhydrous methanol (40 ml). After 5 min t.l.c.
(ethyl acetate:hexane, 1:1) indicated the presence of no starting material and triethylamine (2.29 ml,
16.4 mmol) was added. After 24 h the reaction mixture was preabsorbed onto silica; purification by
flash chromatography (ethyl acetate:hexane, 1:1) yielded mannononitrile 11 as a white crystalline solid
20
(440 mg, 17%), m.p. 109–110°C (hexane/ether); [α]D +53.5 (c 0.8 CHCl3); νmax (KBr): 1761 cm−1
(C_O); δH (500 MHz and COSY, CDCl3): 1.36, 1.39, 1.48, 1.60 (12H, 4×s, 2×C(CH3)2), 3.85 (3H, s,
0
0
OCO2CH3), 3.96 (1H, dd, J5,6 6.1 Hz, J6,6 8.3 Hz, H-6), 4.12 (1H, m, J4,5 8.0 Hz, J5,6 6.1 Hz, J5,6 6.2
Hz, H-5), 4.16 (1H, dd, J5,6 6.2 Hz, J6,6 8.3 Hz, H-60), 4.31 (1H, dd, J2,3 5.1 Hz, J3,4 7,8 Hz, H-3),
4.89 (1H, d, J2,3 5.1 Hz, H-2), 5.12 (1H, t, J 8.0 Hz, H-4); δC (CDCl3): 25.2, 25.7, 25.9, 26.7, (4×q,
2×C(CH3)2), 55.5 (q, OCO2CH3), 66.1 (d, C-2), 67.4 (t, C-6), 74.0, 76.0, 77.5 (3×d, C-3, C-4, C-5),
111.0, 112.2 (2×s, 2×C(CH3)2), 116.8 (s, C-1), 155.0 (s, C_O); m/z (CI, NH3): 316 (M+H+, 60), 333
0
0
+
(M+NH4 , 100%). Found: C, 53.29; H, 6.81; N, 4.71. C14H21O7N requires: C, 53.33; H, 6.71; N, 4.44%.
20
Further elution of the column gave aminoester 7β as a colourless oil (1.57 g, 60%); [α]D −4.0 (c
0.5 CHCl3); νmax (KBr): 3430, 3347 (NH), 1738 cm−1 (C_O, ester); δH (CDCl3): 1.37, 1.39, 1.44, 1.54
(12H, 4×s, 2×C(CH3)2), 3.71 (1H, dd, J4,5 3.0 Hz, J5,6 7.9 Hz, H-5), 3.80 (3H, s, CO2CH3), 3.99 (1H,
dd, J6,7 4.4 Hz, J7,7 8.7 Hz, H-7), 4.08 (1H, dd, J6.7 6.3 Hz, J7 ,7 8.7 Hz, H-70), 4.35 (1H, ddd, J5,6 7.8
0
0
0
0
Hz, J6,7 4.4 Hz, J6,7 6.3 Hz, H-6), 4.86 (2H, m, H-3, H-4); δC (CDCl3): 24.2, 25.6, 25.7, 26.8 (4×q,
2×C(CH3)2), 52.5 (q, CO2CH3), 66.9 (t, C-7), 73.2, 78.7, 80.5, 81.2 (4×d, C-3, C-4, C-5, C-6), 93.0 (s,
C-2), 109.4, 112.8 (2×s, 2×C(CH3)2), 169.2 (s, C-1); m/z (CI, NH3): 318 (M+H+, 100%). Found: C,
52.86; H, 7.60; N, 4.21. C14H23O7N requires: C, 52.99; H, 7.31; N, 4.41%.
Further elution of the column gave aminoester 7α as a white crystalline solid (313 mg, 12%), m.p.
102–104°C (hexane/ether); [α]D20 +70.1 (c 0.5, acetonitrile); νmax (KBr): 3903, 3854 (NH), 1737 cm−1
(C_O, ester); δH (CDCl3): 1.32, 1.39, 1.44, 1.47 (12H, 4×s, 2×C(CH3)2), 3.80 (3H, s, CO2CH3), 4.10
(1H, dd, J6,7 4.3 Hz, J7,7 8.8 Hz, H-7), 4.15 (1H, dd, J6,7 6.1 Hz, J7,7 8.8 Hz, H-70), 4.20 (1H, dd, J4,5
3.5 Hz, J5,6 8.2 Hz, H-5), 4.51 (1H, ddd, J5,6 8.2 Hz, J6,7 4.3 Hz, J6,70 6.1 Hz, H-6), 4.52 (1H, d, J3,4 5.8
Hz, H-3), 4.87 (1H, dd, J3,4 5.8 Hz, J4,5 3.5 Hz, H-4); δC (d3 acetonitrile): 23.9, 24.4, 25.0, 26.0, (4×q,
2×C(CH3)2), 51.6 (q, CO2CH3), 66.5 (t, C-7); 73.0, 79.9, 80.3, 87.5 (4×d, C-3, C-4, C-5, C-6), 94.9 (s,
C-2), 108.7, 112.7 (2×s, 2×C(CH3)2), 170.6 (s, C-1); m/z (CI, NH3): 318 (M+H+, 100%). Found: C,
53.30; H, 7.60; N, 4.40. C14H23O7N requires: C, 52.99; H, 7.31; N, 4.41%.
0
0
0
Oxidation of the epimeric amine 10α (138 mg, 0.48 mmol) using a solution of bromine (0.48 mmol)
in dichloromethane under the same conditions as above gave the same three products in the same
proportions.
3.4. Methyl N-2-(benzyloxycarbonylglycylamino)-2,5-anhydro-3,4:6,7-di-O-isopropylidene-D-glycero-
D-talo-heptonate 8α and methyl N-2-(benzyloxycarbonylglycylamino)-2,5-anhydro-3,4:6,7-di-O-
isopropylidene-D-glycero-D-galacto-heptonate 8β
(a) From methyl β-aminoester 7β: A solution of DCC (1.46 g, 7.11 mmol), HBT (0.97 g, 7.11 mmol)
and Z-glycine (1.46 g, 7.11 mmol) in dry DMF (30 ml) was stirred at 0°C under nitrogen for 15 min.
The reaction was then allowed to warm to room temperature and stirred for a further 45 min. Over this
period a white precipitate of dicyclohexylurea gradually came out of the reaction mixture. The aminoester

2148
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
7β (1.66 g, 5.14 mmol) in dry DMF (18 ml) was added and the mixture was stirred for 16 h at room
temperature. The reaction mixture was filtered through Celite and the filtrate concentrated in vacuo to
give a residue that was dissolved in dichloromethane (50 ml) and extracted with water (2×30 ml). The
organic layers were combined and the solvents removed in vacuo to give a residue that was purified by
flash chromatography (ethyl acetate:hexane, 2:1) to give peptide 8α (1.92 g, 72% yield) as a white solid
that was crystallised from acetonitrile but small quantities of dicyclohexylurea remained. νmax (KBr):
3424, 3329 (N–H), 2986, 2953 (–CH), 1752, 1726, 1699 (C_O), 1543, 1215, 1068 cm−1; δH (500 MHz,
CDCl3): 1.23, 1.31, 1.38 (3×s, 12H, 4×CH3), 3.73 (s, 3H, –COOCH3), 3.81 (b, 2H, Gly CH2), 4.03 (m,
2H, Man C6H/H0), 4.32 (dd, J3,4 3.9 Hz, J4,5 7.8 Hz, 1H, Man C4H), 4.39 (m, 1H, Man C5H), 4.91 (dd,
0
J2,3 5.5 Hz, 1H, Man C3H), 4.96 (d, 1H, Man C2H), 5.06, 5.09 (2×d, JH,H 12.7 Hz, 2H, CH2–Ph), 5.28
(b, 1H, Gly NH), 7.17 (s, 1H, Man NH), 7.3 (b, 5H, C6H5).
Further elution gave peptide 8β (0.2 g, 7.5% yield) as a gum, νmax (film): 3396 (–NH), 2987, 2938
(–CH), 1760, 1730, 1702 (C_O), 1499, 1261, 1214, 1069 cm−1; δH (500 MHz, CDCl3): 1.28, 1.30, 1.36,
1.43 (4×s, 12H, 4×CH3), 3.70 (s, 3H, –COOCH3), 3.85 (b, 2H, Gly CH2), 3.96 (dd, J3,4 3.5 Hz, J4,5 9.2
Hz, 1H, Man C4H), 4.04 (m, 2H, Man C6H/H0), 4.27 (m, 1H, Man C5H), 4.71 (d, 1H, J2,3 6.1 Hz, Man
C2H), 4.75 (dd, 1H, Man C3H), 5.05 (s, 2H, CH2–Ph), 5.40 (b, 1H, Gly NH), 7.18 (s, 1H, Man NH),
7.3 (b, 5H, C6H5); δC (50.3 MHz, CDCl3): 24.77, 25.10, 25.67, 26.88 (q, –CH3), 44.38, 67.11, 67.29
(t, –CH2–), 53.08 (q, –COOCH3), 72.73, 79.85, 80.03, 83.26 (d, –CH–), 88.56, 109.43, 114.12, 136.05
(s, –C–), 128.05, 128.20, 128.50 (d, Ar–H), 156.57, 168.59, 169.10 (s, C_O); m/z (NH3, CI, %): 509
[(M+1)+, 10], 401 (30), 343 (72), 285 (20), 167 (21), 108 (85), 91 (100), 58 (38). Found: C, 56.85; H,
6.38; N, 5.63. C24H32N2O10 requires: C, 56.69; H, 6.34; N, 5.51%.
(b) From methyl α-aminoester 7α: A solution of DCC (185 mg, 0.9 mmol), HBT (123 mg, 0.9 mmol)
and BOC-Gly (185 mg, 0.9 mmol) in dry DMF (4 ml) was stirred at 0°C under nitrogen for 15 min. The
reaction was then allowed to warm up to room temperature and stirred for a further 45 min. Over this
period a white precipitate dicyclohexylurea gradually come out of solution. Aminoester 7α (210 mg, 0.66
mmol) dissolved in dry DMF (2 ml) was added and the mixture was stirred for 16 h at room temperature.
The reaction mixture was worked up as above and the crude products purified by flash chromatography
(ethyl acetate:hexane, 2:1) to give peptide 8α (239 mg, 71% yield) as a white solid that was crystallised
from acetonitrile but was contaminated with small quantities of dicyclohexylurea. Further elution gave
peptide 8β (20 mg, 6% yield).
3.5. Methyl N-2-(tert-butoxycarbonylglycylamino)-2,5-anhydro-3,4:6,7-di-O-isopropylidene-D-
glycero-D-talo-heptonate 16α and methyl N-2-(tert-butoxycarbonylglycylamino)-2,5-anhydro-3,4:6,7-
di-O-isopropylidene-D-glycero-D-galacto-heptonate 16β
(a) From methyl β-aminoester 7β: A solution of DCC (0.61 g, 2.97 mmol), HBT (0.40 g, 2.97 mmol)
and BOC-Gly (0.52 g, 2.97 mmol) in dry DMF (16 ml) was stirred at 0°C under nitrogen for 15 min. The
reaction was then allowed to warm up to room temperature and stirred for a further 45 min during which
time dicyclohexylurea gradually precipitated out of solution. Then aminoester 7β (0.71 g, 2.26 mmol)
in dry DMF (6 ml) was added and the mixture was stirred for 16 h at room temperature. The reaction
mixture was filtered through Celite and the filtrate evaporated in vacuo to give a residue that was dissolved
in dichloromethane (50 ml) and extracted with water (2×30 ml). The organic layers were evaporated in
vacuo to give a residue that was preabsorbed onto silica and purified by flash chromatography (ethyl
acetate:hexane, 2:1) to give peptide 16α (0.80 g, 75% yield) as a white solid that was crystallised
from acetonitrile but always contained small quantities of dicyclohexylurea. νmax (KBr): 3414, 3337
(N–H), 2987, 2937 (–CH), 1760, 1730, 1698 (C_O), 1550, 1369, 1213, 1164, 1087 cm−1; δH (500 MHz,

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2149
CDCl3): 1.31, 1.37, 1.45, 1.46 (4×s, 21H, 7×–CH3), 3.76–3.84 (m, 5H), 4.08–4.14 (m, 2H), 4.31 (m,
1H), 4.45–4.49 (m, 1H), 4.97–4.99 (m, 1H), 5.02–5.04 (m, 1H), 5.25 (bs, 1H, –NH), 7.56 (bs, 1H, –NH);
δC (50.3 MHz, CDCl3): 24.54, 25.12, 25.32, 26.86, 28.24 (q, –CH3), 44.48, 66.79 (t, –CH2–), 52.99 (q,
–COOCH3), 73.23, 80.82, 82.86, 87.16 (d, –CH–), 80.33, 93.48, 109.08, 113.99 (s, –C–), 156.57, 167.66,
169.86 (s, C_O); m/z (NH3, CI, %): 493 [(M+19)+, 2], 492 [(M+18)+, 8], 475 [(M+1)+, 3], 459 (8), 436
(16), 419 (21), 375 (24), 225 (100).
Peptide 16β (0.09 g, 8.6% yield) as a gum. νmax (film): 3397 (–NH), 2984, 2935 (–CH), 1758, 1699
(C_O), 1496, 1371, 1165, 1070 cm−1; δH (500 MHz, CDCl3): 1.37, 1.40, 1.44, 1.46, 1.57 (5×s, 21H,
7×–CH3), 3.78 (s, 3H, –COOCH3), 3.78–3.86 (m, 2H, –CH2–), 4.04 (dd, J3,4 4.4 Hz, J4,5 8.7 Hz, 1H,
0
0
H4), 4.10–4.16 (m, 2H, H6 and H6 ), 4.35 (ddd, J5,6 6.0 Hz, J5,6 4.6 Hz, 1H, H5), 4.78 (d, J2,3 5.8 Hz, 1H,
H2), 4.83 (dd, 1H, H3), 5.09 (bs, 1H, –NH), 7.30 (bs, 1H, –NH); δC (50.3 MHz, CDCl3): 24.77, 25.10,
25.78, 26.84, 28.18 (q, –CH3), 44.17, 67.30 (t, –CH2–), 53.02 (q, –COOCH3), 72.71, 79.81, 80.00, 83.33
(d, –CH–), 80.05, 88.46, 109.41, 114.06 (s, –C–), 155.72, 168.97, 169.06 (s, C_O); m/z (NH3, CI, %):
475 [(M+1)+, 4], 376 (12), 375 (75), 343 (18), 225 (100).
(b) From methyl α-aminoester 7α: A solution of DCC (86 mg, 0.42 mmol), HBT (57 mg, 0.42 mmol)
and N-tert-butoxycarbonylglycine (73 mg, 0.42 mmol) in dry DMF (2 ml) was stirred at 0°C under
nitrogen for 15 min. The reaction was then allowed to warm up to room temperature and stirred for
a further 45 min. Aminoester 7α (100 mg, 0.32 mmol) dissolved in dry DMF (1 ml) was added and
the mixture was stirred for 16 h at room temperature. The reaction mixture worked up as above and
the residue purified by flash chromatography (ethyl acetate:hexane, 2:1) to give peptide 16α (108 mg,
72% yield) as a white solid that was crystallised from acetonitrile contaminated with small quantities of
dicyclohexylurea, together with peptide 16β (10.5 mg, 7% yield).
3.6. Methyl N-2-(benzyloxycarbonylglycylamino)-2,5-anhydro-3,4:6,7-di-O-isopropylidene-D-glycero-
D-galacto-heptonate 8β and methyl N-2-(benzyloxycarbonylglycylamino)-2,5-anhydro-3,4:6,7-di-O-
isopropylidene-D-glycero-D-talo-heptonate 8α
A stirred solution of Z-glycine (185 mg, 0.945 mmol), and triethylamine (132 µl) in THF (6 ml) and
CH3CN (6 ml), at −15°C was treated with ethyl chloroformate (90 µl) and the reaction mixture left
until a precipitate formed. Then a solution of the aminoester 7β (300 mg, 0.945 mmol) and triethylamine
(132 µl) in DMF (6 ml) was slowly added. The resulting mixture stirred first at −10°C for 1 hour and
then at room temperature for 4 hours, concentrated in vacuo and the solid residue was dissolved in
dichloromethane. The organic solution was washed with 0.5 M solution of HCl, then 0.5 M solution
of sodium bicarbonate, dried (Na2SO4) and the solvent removed in vacuo to give a residue that was
preabsorbed onto silica and purified by flash chromatography (ethyl acetate:hexane, 1:1) to give peptide
8β (268.65 mg, 56% yield) and traces peptide 8α, both identical to previously prepared samples.
3.7. Methyl
N-2-(tert-butoxycarbonylglycylamino)-2,5-anhydro-3,4-O-isopropylidene-D-glycero-D-
talo-heptonate 20α
Acetic acid (10 ml) was added to a solution of 16α (0.5 g, 1.05 mmol) in methanol (10 ml), and
the solution was stirred at 50°C for 18 h. The reaction mixture was evaporated in vacuo and co-
evaporated with toluene (2×15 ml) to give a residue that was purified by flash chromatography (ethyl
acetate:methanol, 95:5) to afford the partially hydrolysed BOC-peptide 20α (0.38 g, 82% yield) as a
white solid, m.p. 224–225°C (acetonitrile); [α]D20: +98.5 (c, 1 in methanol); νmax (KBr): 3453, 3308
(–NH), 2990, 2931 (–CH), 1749, 1721, 1694 (C_O), 1543, 1276, 1164, 1080, 1062 cm−1; δH (500

2150
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
MHz, CD3OD): 1.30, 1.40, 1.44 (3×s, 15H, 5×–CH3), 3.61 (dd, 1H), 3.69–3.71 (m, 4H), 3.76 (dd, 1H),
3.87 (dd, 1H), 3.98–4.01 (m, 1H), 4.78–4.86 (m, 2H), 4.90–4.92 (dd, 1H); δC (50.3 MHz, CD3OD):
25.11, 25.95, 28.68 (q, –CH3), 44.24, 65.05 (t, –CH2–), 52.98 (q, –COOCH3), 70.35, 81.08, 81.64, 87.21
(d, –CH–), 80.72, 93.31, 114.73 (s, –C–), 158.34, 168.71, 172.31 (s, C_O); m/z (NH3, CI, %): 452
[(M+18)+, 2], 396 (3), 378 (5), 335 (6), 75 (100). Found: C, 49.92; H, 7.02; N, 6.46. C18H30N2O10
requires: C, 49.77; H, 6.91; N, 6.45%.
3.8. Reprotection of partially deprotected BOC-peptide 20α to fully protected 16α
2,2-Dimethoxypropane (1 ml) and p-toluenesulfonic acid (27 mg) were added to a solution of 20α (50
mg, 0.12 mmol) in analytical grade acetone (2 ml). The reaction mixture was stirred at room temperature
for 12 h and then sodium hydrogen carbonate (34 mg) was added and the reaction stirred for a further 5
min. The mixture was filtered through Celite and the filtrate evaporated in vacuo to give peptide 16α (54
mg, quantitative yield), identical to the material described above.
3.9. Methyl N-2-(benzyloxycarbonylglycylamino)-2,5-anhydro-3,4-O-isopropylidene-D-glycero-D-talo-
heptonate 19α
Acetic acid (4 ml) was added to a solution of the fully protected peptide 8α (100 mg, 0.2 mmol) in
methanol (4 ml), and the solution was stirred at 50°C for 18 h. The reaction mixture was evaporated
in vacuo and co-evaporated with toluene (2×10 ml) to give a residue that was purified by flash
chromatography (ethyl acetate:methanol, 95:5) to afford the partially hydrolysed peptide 19α (74 mg,
80% yield) as a white solid, m.p. 228–229°C (acetonitrile); [α]D20: +92.5 (c, 0.8 in methanol); νmax
(KBr): 3328 (–NH and –OH), 2948 (–CH), 1755, 1731, 1698 (C_O), 1551, 1264, 1233, 1078 cm−1
;
δH (500 MHz, CD3OD): 1.30, 1.40 (2×s, 6H, 2×–CH3), 3.62 (dd, 1H), 3.70–3.78 (m, 6H), 3.88 (dd,
1H), 3.98–4.02 (m, 1H), 4.77–4.91 (m, 2H), 5.09 (s, 2H), 7.30–7.37 (m, 5H, 5×Ar–H); δC (50.3 MHz,
CD3OD): 23.57, 24.43 (q, –CH3), 43.09, 63.57, 66.44 (t, –CH2–), 51.57 (q, –COOCH3), 68.88, 79.55,
80.23, 85.80 (d, –CH–), 91.92, 113.43, 136.89 (s, –C–), 127.71, 127.84, 128.27 (d, Ar–H), 157.89,
167.62, 171.04 (s, C_O); m/z (NH3, CI, %): 469 [(M+1)+, 6], 378 (27), 361 (45), 303 (100). Found:
C, 54.13; H, 6.03; N, 5.86. C21H28N2O10 requires: C, 53.83; H, 6.03; N, 5.98%.
3.10. Reprotection of 19α to fully protected Z-peptide 8α
2,2-Dimethoxypropane (1 ml) and p-toluenesulfonic acid (27 mg) was added to a solution of mono-
acetonide 19α (50 mg, 0.11 mmol) in analytical grade acetone (2 ml). The reaction mixture was stirred at
room temperature for 12 h; sodium hydrogen carbonate (34 mg) was then added and the reaction mixture
stirred for a further 5 min. The mixture was filtered through Celite and the filtrate evaporated in vacuo to
give peptide 8α (54 mg, quantitative yield) as a white solid, identical to the material above.
3.11. Methyl [N-(2-glycylamino)-2,5-anhydro-3,4:6,7-di-O-isopropylidene-D-glycero-D-talo-
heptonoyl] glycinate 17α
Palladium black (50 mg) was added to a solution of peptide 8α (0.5 g, 0.5 mmol) in methanol (30
ml) and the reaction mixture stirred for 2 h at room temperature under an atmosphere of hydrogen. The
reaction mixture was then filtered through Celite and the filtrate evaporated in vacuo to give a residue
that was purified by flash chromatography (chloroform:methanol, 85:15) to give aminoester 17α (336

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2151
mg, 91% yield) as a white solid, m.p. 155–157°C (chloroform); [α]D20: +100.8 (c, 0.25 in MeOH); νmax
(KBr): 3191 (–NH), 2989 (C–H), 1756, 1705 (C_O), 1556, 1375, 1266, 1214, 1070 cm−1; δH (500 MHz,
CDCl3): 1.32, 1.38, 1.46 (3×s, 12H, 4×–CH3), 3.41 (s, 2H), 3.82 (s, 3H, –COOCH3), 4.09–4.15 (m, 2H),
4.37 (dd, J3,4 3.8 Hz, J4,5 7.4 Hz, 1H, H4), 4.46–4.50 (m, 1H), 5.02 (dd, J2,3 5.8 Hz, 1H, H3), 5.15 (d,
1H, H2), 8.52 (bs, 1H, –NH); δC (50.3 MHz, CDCl3): 24.55, 24.75, 25.31, 26.71 (q, –CH3), 42.63, 66.61
(t, –CH2–), 52.89 (q, –COOCH3), 73.09, 80.45, 81.46, 86.39 (d, –CH–), 92.78, 109.19, 113.94 (s, –C–),
167.92, 170.60 (s, C_O); m/z (NH3, CI, %): 389 [(M+19)+, 2], 376 [(M+2)+, 14], 375 [(M+1)+, 100],
343 (45), 317 (46), 285 (60), 259 (56), 183 (48), 101 (73).
3.12. 1-(R)-(30,60-Diketopiperazine)-1,4-anhydro-2,3:5,6-di-O-isopropylidene-D-mannose 18α
(a) From methyl glycinate 17α: Potassium tert-butoxide (195 mg, 1.74 mmol) was added to a solution
of aminoester 17α (0.5 g, 1.34 mmol) in dry THF (250 ml) and the reaction mixture was stirred for 24 h at
room temperature under nitrogen. The solution was evaporated in vacuo to give a residue that was purified
by flash chromatography (ethyl acetate:methanol, 95:5) to give the protected diketopiperazine 18α (405
mg, 89% yield) as an amorphous solid, [α]D20: +113.5 (c, 1 in MeOH); νmax (KBr): 3259 (–NH), 2987,
2936 (–CH), 1693 (C_O), 1375, 1214, 1070 cm−1; δH (500 MHz, CD3CN): 1.28, 1.32, 1.40, 1.42 (4×s,
0
12H, 4×–CH3), 3.70 (dd, J5,6 4.1 Hz, J6,6 17.5 Hz, 1H, H6), 3.95–3.99 (m, 1H), 4.05–4.10 (m, 2H), 4.23
(dd, J4,5 7.3 Hz, J3,4 3.3 Hz, 1H, H4), 4.36–4.40 (m, 1H, H5), 4.86 (d, J2,3 5.7 Hz, 1H, H2), 4.90 (dd, 1H,
H3), 6.54 (bs, 1H, –NH), 7.15 (bs, 1H, –NH); δC (50.3 MHz, CD3CN): 25.01, 25.26, 25.54, 26.93 (q,
–CH3), 45.78, 67.14 (t, –CH2–), 73.73, 81.17, 82.05, 89.34 (d, –CH–), 92.52, 109.68, 114.83 (s, –C–),
164.77, 167.43 (s, C_O); m/z (NH3, CI, %): 344 [(M+2)+, 6], 343 [(M+1)+, 30], 327 (12), 286 (14), 285
(100). Found: C, 52.78; H, 6.55; N, 7.94. C15H22N2O7 requires: C, 52.63; H, 6.48; N, 8.18%.
(b) From monoacetonide 21α: To a solution of the monoacetonide 21α (26 mg, 0.08 mmol) in
analytical grade acetone (2 ml), 2,2-dimethoxypropane (1 ml) and p-toluenesulfonic acid (26 mg) was
added. The reaction mixture was stirred at room temperature for 9 h, then neutralised with sodium
hydrogen carbonate (52 mg), stirred for a further 5 min, filtered through Celite and the filtrate evaporated
in vacuo to give diketopiperazine 18α (29 mg, quantitative yield) as a white solid, identical to the material
above.
(c) From fully deprotected 9α: The unprotected diketopiperazine 9α (20 mg, 0.08 mmol) in analytical
grade acetone (2 ml) and 2,2-dimethoxypropane (1 ml) was stirred with p-toluenesulfonic acid (27 mg)
at room temperature for 5 h and then neutralised with sodium hydrogen carbonate (34 mg). The reaction
mixture was worked up as above to give the diacetonide 18α (24 mg, 90% yield) as a white solid.
3.13. 1-(S)-(30,60-Diketopiperazine)-1,4-anhydro-2,3:5,6-di-O-isopropylidene-D-mannose 18β
(a) From the Z-protected peptide 8β: Palladium black (5 mg) was added over a solution of peptide
8β (50 mg, 0.05 mmol) in methanol (3 ml) and the reaction mixture stirred for 2 h at room temperature
under an atmosphere of hydrogen. The reaction mixture was then filtered through Celite and the filtrate
evaporated in vacuo to give a residue that was purified by flash chromatography (chloroform:methanol,
85:15) to give diketopiperazine 18β (32 mg, 95% yield) as an amorphous white solid, [α]D20: +96 (c, 0.5
in MeOH); νmax (KBr): 3231 (–NH), 2988, 1695 (C_O), 1455, 1381, 1216, 1063, 1046 cm−1; δH (500
0
MHz, CD3CN): 1.30, 1.35, 1.36, 1.52 (4×s, 12H, 4×–CH3), 3.80 (dd, J5,6 3.8 Hz, J6,6 18.1 Hz, 1H,
H-6), 3.93–4.08 (m, 4H), 4.27–4.31 (m, 1H), 4.88 (dd, J2,3 6.0 Hz, J3,4 3.8 Hz, 1H, H-3), 4.94 (d, 1H,
H-2), 6.69 (bs, 1H, –NH), 6.82 (bs, 1H, –NH); δC (50.3 MHz, CD3CN): 24.47, 25.35, 26.01, 26.94 (q,
–CH3), 45.17, 66.99 (t, –CH2–), 73.98, 80.37, 81.29, 81.60 (d, –CH–), 88.29, 109.46, 118.27 (s, –C–),

2152
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
165.27, 167.27 (s, C_O); m/z (NH3, CI, %): 360 [(M+18)+, 5], 343 [(M+1)+, 76], 285 (100), 267 (48).
Found: C, 52.61; H, 6.44; N, 8.15. C15H22N2O7 requires: C, 52.63; H, 6.48; N, 8.18%.
(b) From equilibration of the α-anomer 18α: Potassium tert-butoxide (30 mg, 0.27 mmol) was added
to a solution of the epimeric diketopiperazine 18α (500 mg, 1.46 mmol) dissolved in dry DMF. The
reaction mixture was heated at 100°C under an atmosphere of nitrogen for 12 h. The solvent was then
removed in vacuo to give a residue that was purified by flash chromatography (ethyl acetate:MeOH, 95:5)
to afford diketopiperazine 18β (440 mg, 88% yield) identical to the material described above.
(c) From the monoacetonide 21β: p-Toluenesulfonic acid (27 mg) was added to a solution of
diketopiperazine 21β (20 mg, 0.07 mmol) in acetone (2 ml) and 2,2-dimethoxypropane (1 ml). The
reaction mixture was stirred at room temperature for 5 h and then sodium hydrogen carbonate (34 mg)
was added and the reaction stirred for a further 5 min. The mixture was filtered through Celite and the
filtrate evaporated in vacuo to give diketopiperazine 18β (23 mg, quantitative yield) as a white solid.
(d) From the deprotected spirodiketopiperazine 9β: To a solution of diketopiperazine 9β (20 mg, 0.08
mmol) in analytical grade acetone (2 ml), 2,2-dimethoxypropane (1 ml) and p-toluenesulfonic acid (26
mg) was added. Work-up as described above afforded the protected diketopiperazine 18β (21 mg, 80%
yield).
3.14. 1-(R)-(30,60-Diketopiperazine)-1,4-anhydro-2,3-O-isopropylidene-D-mannose 21α
A solution of diketopiperazine diacetonide 18α (250 mg, 0.7 mmol) in a mixture of acetic acid and
water (1:1) (20 ml) was stirred at 50°C for 2 h. The reaction mixture was evaporated in vacuo and the
residue co-evaporated twice with toluene (25 ml) and purified by flash chromatography (CHCl3:MeOH,
4:1) to give diketopiperazine 21α (195 mg, 88% yield) as a white solid, m.p. 125°C, decomposed
(methanol); [α]D20: +120.5 (c, 1 in MeOH); νmax (KBr): 3344 (–NH and –OH), 2940, 1691 (C_O),
1420, 1377, 1216, 1090 cm−1; δH (500 MHz, CD3OD): 1.31, 1.45 (2×s, 6H, 2×–CH3), 3.68 (dd, J5,6
0
0
4.8 Hz, J6,6 11.7 Hz, 1H, H-6), 3.75 and 4.08 (ABq, JAB 17.5 Hz, 2H, –CH2–), 3.79 (dd, J5,6 2.7 Hz,
1H, H-60), 4.05–4.08 (m, 1H, H-5), 4.18 (dd, J4,5 8.9 Hz, J3,4 3.3 Hz, 1H, H-4), 4.90 (d, J2,3 5.6 Hz,
1H, H-2), 5.01 (dd, 1H, H-3); δC (50.3 MHz, CD3OD): 23.57, 24.28 (q, –CH3), 44.56, 63.53 (t, –CH2–),
68.62, 80.42, 80.54, 88.14 (d, –CH–), 91.82, 114.19 (s, –C–), 166.46, 168.39 (s, C_O); m/z (NH3, CI,
%): 320 [(M+18)+, 20], 303 [(M+1)+, 96], 245 (100).
3.15. 1-(S)-(30,60-Diketopiperazine)-1,4-anhydro-2,3-O-isopropylidene-D-mannose 21β
(a) By epimerization of α-anomer 21α: The epimeric diketopiperazine 21α (200 mg, 0.7 mmol)
was dissolved in dry DMF (10 ml) and treated with potassium tert-butoxide (46 mg, 0.4 mmol). The
reaction mixture was heated at 100°C under an atmosphere of nitrogen for 12 h and then evaporated in
vacuo to give a residue that was purified by flash chromatography (ethyl acetate:MeOH, 85:15) to give
diketopiperazine 21β (170 mg, 85% yield) as a white solid, m.p. 216–218°C (methanol); [α]D20: +88.8
(c, 0.5 in MeOH); νmax (KBr): 3350 (N–H and –OH), 2940, 1686 (C_O), 1418, 1373, 1215, 1090 cm−1
;
0
δH (500 MHz, D2O): 1.41, 1.56 (2×s, 6H, 2×–CH3), 3.58 (dd, J5,6 5.5 Hz, J6,6 12.2 Hz, 1H, H-6), 3.74
(dd, J5,6 2.7 Hz, 1H, H-60), 3.89–3.92 (m, 1H, H-5), 3.98 and 4.19 (ABq, JAB 18.7 Hz, 2H, –CH2–),
0
4.05 (dd, 1H, H-4), 5.06–5.09 (m, 2H, H-3 and H-2); δC (50.3 MHz, D2O): 24.23, 25.56 (q, –CH3),
44.73, 63.88 (t, –CH2–), 69.85, 79.09, 80.92, 81.35 (d, –CH–), 88.37, 114.75 (s, –C–), 166.16, 169.92 (s,
C_O); m/z (NH3, CI, %): 320 [(M+18)+, 71], 303 [(M+1)+, 95], 245 (73), 183 (100).
(b) By hydrolysis of the diacetonide 18β: A solution of the diacetonide 18β (40 mg, 0.12 mmol)
in a mixture of acetic acid and water (1:1) (4 ml) was stirred at 50°C for 2 h. The reaction mixture

J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
2153
was evaporated in vacuo and the residue coevaporated twice with toluene (3 ml) and purified by flash
chromatography (CHCl3:MeOH, 80:20) to give diketopiperazine 21β (31 mg, 87% yield) identical to the
material above.
3.16. 1-(R)-(30,60-Diketopiperazine)-1,4-anhydro-D-mannose 9α
The monoacetonide 21α (150 mg, 0.48 mmol) was dissolved in a mixture of trifluoroacetic acid:water
(1:1) (6 ml) and stirred at room temperature for 1 h. The mixture was then evaporated in vacuo,
coevaporated twice with toluene (5 ml) and the residue crystallised from MeOH/ethyl acetate to give
diketopiperazine 9α (120 mg, 92% yield), m.p. 190°C, decomposed (methanol); [α]D20: +2 (c, 0.1 in
MeOH); νmax (KBr): 3277 (–NH and –OH), 2917, 1689 and 1655 (C_O), 1457, 1338, 1092, 1081
cm−1; δH (500 MHz, DMSO): 3.30 (bdd, J5,6 6.1 Hz, J6,6 11.3 Hz, 1H, C6H), 3.53 (bd, 1H, C6H), 3.71
0
(b, 1H, C5H), 3.77–3.85 (m, 2H, CH2), 3.90 (ddd, J2,3 4.5 Hz, J3,4 2.5 Hz, J3,3OH 11 Hz, 1H, C3H), 3.94
(dd, J4,5 8.8 Hz, 1H, C4H), 4.20 (b, 1H, C6OH), 4.25 (t, J2,2OH 4.5 Hz, 1H, C2H), 4.42 (b, 1H, C5OH),
5.46 (d, 1H, C2OH), 5.77 (d, 1H, C3OH), 8.50 (b, 1H, NH), 8.83 (s, 1H, N1H); δC (125 MHz, DMSO):
43.8 (t, CH2), 63.1 (t, C6), 68.6 (d, C5), 71.6 (d, C3), 79.5 (d, C2), 81.8 (d, C4), 90.4 (s, C1), 165.2,
166.3 (2×s, 2×CO); m/z (NH3, CI, %): 263 [(M+1)+, 19], 143 (81), 127 (100).
3.17. 1-(S)-(30,60-Diketopiperazine)-1,4-anhydro-D-mannose 9β
(a) By epimerization of the unprotected α-anomer 9α The epimeric diketopiperazine 9α (100 mg,
0.4 mmol) was dissolved in dry DMF (5 ml) and potassium tert-butoxide (45 mg, 0.4 mmol) added. The
reaction mixture was heated at 100°C under an atmosphere of nitrogen for 12 h and then evaporated in
vacuo to give a residue that was evaporated in vacuo and then coevaporated twice with toluene (5 ml)
and crystallised from MeOH/ethyl acetate/hexane to afford diketopiperazine 9β (90 mg, 90% yield) as a
white solid, m.p. 151–153°C (methanol); [α]D20: +10.8 (c, 0.5 in MeOH); νmax (KBr): 3389 and 3247
(–NH and –OH), 2937, 1693 (C_O), 1443, 1215, 1087, 1049 cm−1; δH (500 MHz, DMSO): 3.32 (dd,
0
0
J5,6 5.2 Hz, J6,6 11.2 Hz, 1H, C6H), 3.54 (dd, J5,6 1.9 Hz, 1H, C6H), 3.67 (m, 1H, C4H), 3.67, 3.90
0
(2×d, JH,H 17.8 Hz, 2H, CH2), 3.69 (m, 1H, C5H), 4.14 (m, 1H, C3H), 4.61 (d, J2,3 4.1 Hz, 1H, C2H),
4.29, 4.51, 5.36, 5.47 (4×b, 4H, 4×OH), 7.38 (s, 1H, N1H), 8.13 (b, 1H, NH); δC (125 MHz, DMSO):
44.0 (t, CH2), 62.9 (t, C6), 69.4 (d, C5), 70.8 (d, C3), 72.5 (d, C2), 80.2 (d, C4), 87.4 (s, C1), 164.2,
166.3 (2×s, 2×CO); m/z (NH3, CI, %): 280 [(M+18)+, 12], 263 [(M+1)+, 28], 245 (15), 143 (96), 138
(81), 127 (45), 120 (100). Found: C, 40.91; H, 5.52; N, 10.53. C9H14N2O7 requires: C, 41.22; H, 5.34:
N, 10.69%.
(b) By hydrolysis of the monoacetonide 21β: The monoacetonide 21β (20 mg, 0.07 mmol) was
dissolved in a mixture of trifluoroacetic acid:water (1:1) (1 ml) and stirred at room temperature for 1
h. The mixture was then evaporated in vacuo, coevaporated twice with toluene (2 ml) and the residue
crystallised from MeOH/ethyl acetate to give diketopiperazine 9β (14 mg, 80% yield) identical to the
material above.
Acknowledgements
This work has been supported by the Ministerio de Educacion y Ciencia, the Xunta de Galicia and
The Universidad de Santiago de Compostela, an EPSRC postdoctoral fellowship, and an EPSRC CASE
award.

2154
J. C. Estevez et al. / Tetrahedron: Asymmetry 9 (1998) 2137–2154
References
1. Nicolaou, K. C., Florke, H. M., Egan, G., Barth, T., Estevez, V. A., Tetrahedron Lett., 1995, 36, 1775.
2. von Roedern, E. G., Lohof, E., Hessler, G., Hoffmann, M., Kessler, H., J. Am. Chem. Soc., 1996, 118, 10156; von Roedern,
E. G., Kessler, H., Angew. Chem. Int. Ed. Engl., 1994, 334, 687.
3. Suhara, Y., Hildreth, J. E. K., Ichikawa, Y., Tetrahedron Lett., 1996, 37, 1575; Suhara, Y., Izumi, M., Ichikawa, M., Penno,
M. B., Ichikawa, Y., Tetrahedron Lett., 1997, 38, 7167.
4. Goodnow, R. A., Richou, A.-R., Tam, S., Tetrahedron Lett., 1997, 38, 3195; Goodnow, R. A., Tam, S., Preuss, D. L.,
McComas, W. W., Tetrahedron Lett., 1997, 38, 3199.
5. McDevitt, J. P., Lansbury, P. T., J. Am. Chem. Soc., 1996, 118, 3818.
6. Wessel, H. P., Mitchell, C. M., Lobato, C. M., Schmidt, G. Angew. Chem. Int. Ed. Engl., 1995, 34, 2712; Timmers, C. M.,
Turner, J. J., Ward, C. M., van der Marcel, G. A., Kouwijzer, M. L. C. E., Grootenhuis P. D. J., van Boom, J. H., Chem.
Eur. J., 1997, 6, 920.
7. Muller, C., Kitas, E., Wessel, H. P., J. Chem. Soc., Chem. Commun., 1995, 2425; Drouillat, B., Kellam, B., Dekany, G.,
Starr, M. S., Toth, I. Bioorg. Med. Chem. Lett., 1997, 17, 2247.
8. Poitout, L., le Merrer, Y., Depazay, J.-C., Tetrahedron Lett., 1995, 36, 6887.
9. Krülle, T. M., Watson, K. A., Gregoriou, M., Johnson, L. N., Crook, S., Watkin, D. J., Griffiths, R. C., Nash, R. J.,
Tsitsanou, K. E., Zographos, S. E., Oikonomakos, N. G., Fleet, G. W. J., Tetrahedron Lett., 1995, 36, 8291; Brandstetter,
T. W., de la Fuente, C., Kim, Y., Cooper, R. I., Watkin, D. J., Oikonomakos, N. G., Johnson, L. N., Fleet, G. W. J.,
Tetrahedron, 1996, 52, 10711.
10. Estevez, J. C., Smith, M. D., Lane, A. L., Crook, S., Watkin, D. J., Besra, G. S., Brennan, P. J., Nash, R. J., Fleet, G. W. J.,
Tetrahedron Asymm., 1996, 7, 387.
11. Brandstetter, T. W., Wormald, M. R., Dwek, R. A., Butters, T. D., Platt, F. M., Tsitsanou, K. E., Zographos, S. E.,
Oikonomakos, N. G., Fleet, G. W. J., Tetrahedron Asymm., 1996, 7, 157.
12. Estevez, J. C., Estevez, R. J., Ardron, H., Wormald, M. R., Brown, D., Fleet, G. W. J., Tetrahedron Lett., 1994, 35, 8885;
Estevez, J. C., Ardron, H., Wormald, M. R., Brown, D., Fleet, G. W. J., Tetrahedron Lett., 1994, 35, 8889.
13. Dondoni, A., Scherrmann, M. C., Marra, A., Delepine, J. L., J. Org. Chem., 1994, 59, 7517.
14. Sabesan, S., Tetrahedron Lett., 1997, 38, 3127.
15. Beacham, A. R., Bruce, I., Choi, S., Doherty, O., Fairbanks, A. J., Fleet, G. W. J., Skead, B. M., Peach, J. M., Saunders,
J., Watkin, D. J., Tetrahedron: Asymm., 1991, 2, 883.
16. Bruce, I., Fleet, G. W. J., di Bello, I. C., Winchester, B. G., Tetrahedron, 1992, 48, 10191.
17. Hui, A., Fairbanks, A. J., Nash, R. J., Lilley, P. M. de Q., Storer, R., Watkin, D. J., Fleet, G. W. J., Tetrahedron Lett., 1994,
35, 8895.
18. Fleet, G. W. J., Bruce, I., Girdhar, A., Haraldsson, M., Peach, J. M., Watkin, D. J., Tetrahedron, 1990, 46, 19.
19. Estevez, J. C., Long, D. D., Wormald, M. R., Dwek, R. A., Fleet, G. W. J., Tetrahedron Lett., 1995, 36, 8287.
20. Burton, J. W., Son, J. C., Fairbanks, A. J., Choi, S. S., Taylor, H., Watkin, D. J., Winchester, B., Fleet, G. W. J., Tetrahedron
Lett., 1993, 34, 6119.
21. Burton, J. W., Ardron, H., Estevez, J. C., Wormald, M. R., Dwek, R. A., Taylor, H., Watkin, D. J., Brown, D., Fleet, G. W.
J., in preparation.
22. Ardron, H., Butters, T. D., Platt, F. M., Wormald, M. R., Dwek, R. A., Fleet, G. W. J., Jacob, G. S., Tetrahedron Asymm.,
1993, 4, 2011.
23. Wooten, E. W., Edge, C. J., Bazzo, R., Dwek, R. A., Rademacher, T. W., Carbohydr. Res., 1990, 203, 13.