Sugar-derived Ras inhibitors: Group epitope mapping by NMR spectroscopy and biological evaluation

Article abstract of DOI:10.1002/ejoc.200600132

Novel inhibitors of Ras protein activation have been found, containing a bicyclic core derived from D-arabinose and benzyl and phenylhydroxylamine moieties. NMR studies (trNOE, saturation-transfer difference, STD) of the binding between these molecules an

Full text of DOI:10.1002/ejoc.200600132

FULL PAPER
DOI: 10.1002/ejoc.200600132
Sugar-Derived Ras Inhibitors: Group Epitope Mapping by NMR Spectroscopy
and Biological Evaluation
Francesco Peri,*[a] Cristina Airoldi,[a] Sonia Colombo,[a] Silvia Mari,[b]
Jesús Jiménez-Barbero,[b] Enzo Martegani,[a] and Francesco Nicotra[a]
Keywords: NMR spectroscopy / Inhibitors / Medicinal chemistry / Drug design / Epitope mapping
Novel inhibitors of Ras protein activation have been found,
synthesized and tested to confirm this hypothesis, and no in-
containing a bicyclic core derived from -arabinose and ben- teraction with Ras in vitro, nor biological activity in mamma-
D
zyl and phenylhydroxylamine moieties. NMR studies
(trNOE, saturation-transfer difference, STD) of the binding
between these molecules and human p21 h-Ras are reported.
A pharmacophore mapping indicates that both the benzyl
and the phenylhydroxylamine moieties are essential for pro-
tein binding. Molecules lacking one of these groups were
lian cells was observed. Our studies led to the development
of molecules that selectively inhibit Ras-dependent cellular
growth in mammalian cells.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
Introduction
Inhibitors 1–4 (Figure 1) present two benzyl groups and a
phenylhydroxylamino group linked to a rigid scaffold. We
previously reported on the conformational properties of the
bicyclic core derived from the natural monosaccharide -
arabinose:[4] it presents the proper spatial disposition to ac-
commodate pharmacophore groups for interaction with
Ras protein. According to molecular modelling and dock-
ing studies, compounds 1–4 are all able to bind Ras with
the orientations of the aromatic and phenylhydroxylamine
moieties similar to those adopted by a number of inhibitors
developed by Schering-Plough.[5] All inhibitors seem to
adopt a similar bound conformation despite the different
configurations of the C-2 atoms of the bicycles and the dif-
ferent natures (amide or sulfonamide) of the C-2 linkers.
This hypothesis is supported by the experimental observa-
tion that all the molecules show similar affinities for Ras,
and inhibit with very similar potency both guanine nucleo-
tide exchange and GDP dissociation from p21 h-Ras. Dock-
ing analysis, in agreement with data previously reported by
Schering-Plough,[6] points out that the phenylhydroxyl-
amine moiety of these inhibitors plays an important role in
the interaction with the protein. The hydroxylamine group
binds in all cases with the Mg++ ion that is also coordinated
by the β-phosphate group of GDP and interacts through
hydrogen bonds with Thr58. We also found important in-
teractions between the amide or sulfonamide groups of the
inhibitors and Ras residues Gly10/Gly60. The benzyl ether
on C-4 is placed in a hydrophobic cavity of Ras and inter-
acts with residues Val9, Met72, Gln99 and Ile100. On the
contrary, the benzyl ether on C-2Ј points outward from the
binding site and has no important interactions with Ras
(Figure 2).
Mutated forms of the Ras proteins are present in about
30% of human tumours.[1] Point mutations of Ras (k-Ras)
generate constitutively active proteins responsible for un-
controlled cellular growth and proliferation leading to on-
cogenesis. Ras activation is also known to be involved in
cancer cell survival after radiation or chemotherapy in a
number of different tumour types, thus contributing to tu-
mour re-insurgence after therapy.[2] There is, therefore, a
great deal of interest in the search for selective inhibitors of
Ras activation as potential anticancer drugs. In this context,
we have prepared and tested novel molecules (compounds
1–4, Figure 1) that are capable of inhibiting nucleotide ex-
change and guanosine diphosphate (GDP) dissociation
from human p21 h-Ras, and of blocking growth and prolif-
eration on normal and k-Ras-transformed mammalian
cells.[3] Interestingly, the inhibition effect of molecules 2 and
4 on cellular growth and proliferation of k-Ras-transformed
mouse fibroblasts were more pronounced than on normal
cells, thus indicating a specificity action toward Ras-medi-
ated signalling.
[a] Department of Biotechnology and Biosciences, University of
Milano-Bicocca,
20126 Milano, Italy
Fax: +39-02-64483565
E-mail: francesco.peri@unimib.it
[b] Departamento de Estructura y Función de Proteínas Centro de
Investigaciones Biológicas, CSIC,
Ramiro de Maeztu 9, 28040 Madrid, Spain
E-mail: jjbarbero@cib.csic.es
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3707

F. Peri, et al.
FULL PAPER
Figure 1. Chemical structures of compounds 1–10 and SCH-54292.
previously adopted by Schering. This compound was used
as a reference in NMR binding studies and biological tests.
The comparison of the binding properties and biological
activities of compounds 1–10 will provide further important
information about the structure-activity relationship in this
family of inhibitors.
Results
Synthesis
Compounds 5–10 were prepared according to the syn-
thetic pathway of Scheme 1. The common bicyclic iodide
intermediate 13 was obtained by the iodo-promoted cycli-
sation of the allyl-C-arabinofuranoside 12. Only the β-an-
omer reacted, and the α-anomer was recovered as reported
elsewhere.[3,4] Nucleophilic displacement of iodine with tet-
rabutylammonium azide afforded 14, which was converted
into the corresponding amine 15 by treatment with tri-
phenylphosphane. Reaction of 15 with p-nitrobenzenesulfo-
nyl chloride, followed by reduction of the aromatic nitro
Figure 2. Compound 1 docked into the binding cavity of Ras using
the GLIDE program (GLIDE 2.7, Schrodinger LLC, 2003). While
the benzyl group on C-4 of inhibitor is inserted into a binding
pocket, the benzyl group on C-2Ј protrudes from the protein.
We present here a second generation of inhibitors, group with hydrazine and Pd/C, yielded compound 9 as a
namely molecules 5–10, structurally very similar to the pre- mixture of diastereoisomers. Alternatively, the condensation
viously tested compounds 1–4. Compounds 5–8 have a free of 15 with p-nitrobenzoic acid in the presence of N,N-di-
hydroxy group on C-2Ј instead of the benzyl group, 9 and isopropylcarbodiimide (DIC) and N-hydroxybenzotriazole
10 have free hydroxy groups on both C-4 and C-2Ј. These (HOBt), followed by reduction with hydrazine and Pd/C,
molecules have been designed for two main reasons: to in- afforded the diastereomeric mixture 10.
vestigate the importance of the aromatic (benzyl) rings in
Intermediate 14 was O-benzylated by using benzyl bro-
the interaction with Ras, and to have greater water solubil- mide and sodium hydride to obtain 16. The C-2Ј benzyl
ity compared to compounds 1–4. The inhibitor SCH-54292 ether of 16 was selectively converted into an O-acetate by
was also prepared by optimizing the synthetic procedure treatment with Ac2O/TFA, and the acetyl group was then
3708
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720

Sugar-Derived Ras Inhibitors: NMR Spectroscopy and Biological Evaluation
FULL PAPER
Scheme 1. Reagents and conditions: a) AcCl, MeOH, 18 h, 91%; b) BTSFA, CH3CN, 100 °C, 3 h, then ATMS, TMSOTf, CH3CN, room
temperature, 1 h, then H2O, 97%; c) NIS, THF, 90 °C, 10 min, 58%; d) Bu4NN3, DMF, 70 °C, 72 h, 90%; e) PPh3, THF/H2O, 70 °C,
14 h, 85%; f) p-nitrobenzenesulfonyl chloride, pyridine, 0 °C, 6 h, 64%; g) DIC, HOBt, DIPEA, p-nitrobenzoic acid, DMF, room tempera-
ture, 24 h, 75.5%; h) NH2NH2, Pd/C, THF, 0 °C, 45 min, 98%; i) NaH, BnBr, DMF, room temperature, 15 min, 91%; l) Ac2O/TFA (4:1,
v/v), room temperature, 90 min, then MeONa, MeOH, room temperature, 30 min, 85%; m) PPh3, THF/H2O, 70 °C, 14 h, 94%; n) TEA,
p-nitrobenzenesulfonyl chloride, CH2Cl2, 0 °C, 5 h, 82%; o) DIC, HOBt, DIPEA, p-nitrobenzoic acid, DMF, room temperature, 24 h,
70%; p) NH2NH2, Pd/C, THF, 0 °C, 45 min, 92%. BSTFA = N,O-bis(trimethylsilyl)trifluoroacetamide, ATMS = allyltrimethylsilane,
TMSOTf = trimethylsilyl trifluoromethanesulfonate, NIS = N-iodosuccinimide, TEA = triethylamine, DIC = N,N-diisopropylcarbodi-
imide, HOBt = N-hydroxybenzotriazole, DIPEA = diisopropylethylamine.
Scheme 2. Reagents and conditions: a) PPh3, DIAD, –80 °C, 1 h, then room temperature, 18 h, 84%; b) K2CO3, dry CH3OH, 24 h, 95%;
c) NH2NH2, Pd/C, THF, 0 °C, 45 min, 93%. DIAD = diisopropyl azodicarboxylate.
removed with sodium methoxide in methanol. The azide by a Mitsunobu-type glycosylation (Scheme 2).[7] Unlike
was reduced using triphenylphosphane in THF/water, af- previously published methods[5] based on the use of tetra-
fording amine 17. Reaction of the amino group of 17 with acetylbromoglucose as the glycosyl donor, this glycosyla-
p-nitrobenzenesulfonyl chloride allowed the introduction of tion reaction was efficient and almost totally stereoselective
the p-nitrophenylsulfonamide moiety (20 and 21). Alterna- in favour of the β-N-glycoside, yielding 24 in 84% yield.
tively, condensation with p-nitrobenzoic acid yielded amides Removal of the acetyl protecting groups was accomplished
22 and 23. Amide and sulfonamide diastereomers were sep- by mild basic hydrolysis (Na2CO3 in MeOH) in order to
arated at this point of the synthesis by flash chromatog- avoid base-promoted cleavage of the N–glycoside bond.
raphy on silica gel. Finally, reduction of the aromatic nitro The final compound SCH-54292 was obtained after re-
group with hydrazine and Pd/C led to the final phenylhy- duction of the nitro group with hydrazine and Pd/C.
droxylamine compounds 5–8.
Ras–Ligand Binding Studies by NMR
For the synthesis of compound SCH-54292, the critical
N-glycosylation step of tetraacetylglucose with 4-nitro-N-
For ligands that exchange between the free and bound
[2-(2-naphthyloxy)ethyl]benzenesulfonamide was achieved states at a moderately fast rate, trNOESY experiments pro-
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3709

F. Peri, et al.
FULL PAPER
vide a helpful means to study their conformations at the vestigate the interactions in solution between compounds
receptor binding sites. This is usually the case for carbo- 5–10 and the Ras-GDP complex. p21 h-Ras was incubated
hydrates (or glycomimetics) when they are bound to their in a [D11]-Tris buffer, containing 10% CD3OD, 100 m
receptors.[8] trNOESY experiments were performed to in- NaCl, 5 m MgCl2, an amount of GDP equimolar to the
Figure 3. A) NOESY spectrum of compound 5, mixing time = 800 ms; B) trNOESY spectrum of compound 5-Ras-GDP mixture, mixing
time = 200 ms. The spectral region between δ = 1.7 and 4.7 ppm, which contains all signals of protons on the bicyclic scaffold, is shown.
Both the samples were dissolved in a [D11]-Tris buffer at pH = 7.3, containing 10% CD3OD, 100 m NaCl, and 5 m MgCl2. Total
sample volume was 450 µL.
3710
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720

Sugar-Derived Ras Inhibitors: NMR Spectroscopy and Biological Evaluation
FULL PAPER
Table 1. Selective T1 and T1ρ values for compounds 5–8.
Sample
Hx
Hbenz
Hm
H-2
H-1
Selective T1
5
2.07 0.02
0.76 0.03
2.17 0.01
0.75 0.02
2.11 0.01
0.81 0.02
2.45 0.03
1.30 0.02
2.37 0.01
1.08 0.01
2.48 0.01
0.95 0.01
2.46 0.02
0.94 0.03
2.75 0.04
1.59 0.03
2.32 0.01
0.96 0.03
2.27 0.02
0.77 0.01
2.38 0.02
0.83 0.03
2.56 0.05
1.30 0.01
1.14 0.01
0.54 0.01
1.22 0.01
0.59 0.01
1.01 0.01
0.45 0.01
1.13 0.01
0.76 0.01
0.58 0.01
0.32 0.01
0.50 0.01
0.31 0.01
0.58 0.01
0.32 0.02
0.57 0.01
0.31 0.02
5+Ras-GDP
6
6+Ras-GDP
7
7+Ras-GDP
8
8+Ras-GDP
T1ρ
5
0.95 0.01
0.51 0.02
1.11 0.01
0.55 0.01
1.20 0.00
0.47 0.01
1.13 0.02
0.72 0.02
1.16 0.01
0.49 0.01
1.63 0.01
0.39 0.01
1.72 0.01
0.42 0.01
2.09 0.03
0.75 0.01
1.06 0.02
0.42 0.02
1.35 0.01
0.41 0.01
1.37 0.01
0.46 0.02
1.55 0.02
0.79 0.04
0.82 0.03
0.31 0.01
0.78 0.02
0.25 0.01
0.72 0.01
0.26 0.01
0.53 0.03
0.43 0.02
0.26 0.01
0.14 0.01
0.26 0.01
0.19 0.02
0.27 0.00
0.16 0.02
–
5+Ras-GDP
6
6+Ras-GDP
7
7+Ras-GDP
8
8+Ras-GDP
–
protein, and compounds 5–10. An optimized ligand/protein Table 2. Selective T1 and T1ρ values for SCH-54292.
molar ratio of 20 was used in all experiments. NOESY spec-
tra of compounds 5–10 in the free state were obtained with
mixing times of 800 ms. In trNOESY experiments, the opti-
mal value of 200 ms for the mixing time was used. In con-
trast with what was observed in the free state, negative
NOEs (trNOEs) were observed for molecules 5–8 as a con-
sequence and proof of interaction with the protein (Fig-
ure 3). Under the same experimental conditions, com-
Sample
Hm
Hf
T1
H-2
pounds 9 and 10 did not show any trNOE signals, probably
indicating their lack of interaction with the Ras-GDP com-
plex. The specific binding of all new compounds to the Ras-
GDP complex was also investigated by the analysis of the
variations of the selective T1 and T1ρ of several ligand pro-
tons measured in the free state and in the presence of the
Ras-GDP complex (Table 1).
SCH-54292
2.52 0.08 1.24 0.05
SCH-54292+Ras-GDP 1.97 0.05 0.96 0.03
–
–
T1ρ
SCH-54292
1.64 0.08 0.97 0.04
0.60 0.06
0.60 0.06
SCH-54292+Ras-GDP 1.33 0.04 0.62 0.07
Selective T1 and T1ρ values of free compounds 5–8 were
1
significantly larger than those determined in the presence between the H spectra and STD spectra of compounds 5–
of the protein. This effect was evident in all protons ana- 8 (Figure 4) pointed out that the major interactions with
lyzed, and it was particularly strong for the aromatic ones. Ras involved both the benzyl and the phenylhydroxylamine
As also observed in the trNOESY experiments (see above), moieties of the ligands. In particular, the signals of the Hm
the measurements of the selective T1 and T1ρ values for and Hx (phenylhydroxylamine) and of the Ha and Hb pro-
compounds 9 and 10 provided values with no significant tons (benzyl) were clearly observed in all STD experiments.
variations in the absence or presence of the Ras-GDP com- In analogy with the previously described experiments, the
plex. For the reference compound, SCH-54292, the aro- STD experiments recorded for 9 and 10 did not show any
matic protons showed a decrease in selective T1 and T1ρ signals. Moreover, the STD spectrum of the reference com-
values when passing from the free to the bound state, while pound SCH-54292 indicated the interaction of both naph-
no change of T1 values was observed for protons of the thyl and phenylhydroxylamine groups with the protein,
sugar moiety (Table 2). Interestingly, the magnitude of the while no interaction with the sugar moiety was detected. As
variations of both selective T1 and T1ρ values was always outlined in previous studies,[6] in this case the glucose por-
much larger for compounds 5–8 than for SCH-54292.
tion functions to increase water solubility and does not play
Saturation-transfer difference (STD) experiments[9] are a any role in the binding with Ras. Thus, all the NMR spec-
very helpful means to detect ligand binding to receptors; troscopic data for 5–8 and the parent compound seem to
therefore STD experiments were performed on the same li- indicate major interactions of the aromatic residues with
gand/protein mixtures with the aim of confirming the inter- Ras, while the bicyclic moiety acts as scaffold to provide the
action, and of studying which region of the ligand directly proper orientation of the interacting residues, along with
interacts with Ras (epitope mapping).[9] The comparison the required solubility.
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3711

F. Peri, et al.
FULL PAPER
Figure 4. A) 1H NMR of compound 5 with Ras-GDP complex; B) STD spectrum of compound 5 with Ras-GDP complex: ligand/protein
ratio, 20:1, number of scan (NS) = 1360, on-resonance frequency δ = 0.8 ppm, off-resonance frequency δ = 40 ppm, total saturation time
= 2 s. Spectra were recorded on the same sample, dissolved in a [D11]-Tris buffer at pH = 7.3, containing 10% CD3OD, 100 m NaCl,
and 5 m MgCl2. Total sample volume was 450 µL.
Figure 5. (A) C-Cdc25Mm-stimulated nucleotide exchange on p21 hRas. The values are expressed as a percentage of the control exchange
rate. (B) C-Cdc25Mm-stimulated dissociation of p21 h-Ras·MANT-GDP complexes. The second column represents nucleotide dissociation
due to Ras intrinsic activity. The values are expressed as a percentage of the control dissociation rate. The inhibitors (5–10 and SCH-
54292) were added at a final concentration of 100 µ.
3712
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720

Sugar-Derived Ras Inhibitors: NMR Spectroscopy and Biological Evaluation
FULL PAPER
Biology
In vitro Experiments on p21 h-Ras
Compounds 5–10 were initially tested for their ability to
inhibit the C-Cdc25Mm-stimulated nucleotide exchange on
purified human Ras protein (p21 h-Ras). To this purpose, a
modified version of Lenzen’s method was used.[10] The C-
Cdc25Mm-stimulated guanine nucleotide exchange was
monitored using the fluorescent 2Ј(3Ј)-O-(N-methylanthra-
niloyl)-GTP (MANT-GTP). p21 h-Ras was incubated with
MANT-GTP in the absence and in the presence of 100 µ
putative inhibitors 5–10. SCH-54292 was used as a positive
control under the same experimental conditions. The ex-
change reaction was started by the addition of C-Cdc25Mm
.
SCH-54292, 5 and 6 were the most active compounds, exhi-
biting 90% (6) or complete (5 and SCH-54292) inhibition
of the exchange. Lower activities were observed for 7 and
8, (75% and 35% inhibition of nucleotide exchange, respec-
tively), while compounds 9 and 10 were totally inactive
(Figure 5A).
Figure 6. Inhibition test in mammalian cells. (A) Normal NIH3T3,
and (B) NIH3T3 k-Ras mouse fibroblasts were seeded into 60 mm
dishes and grown for 1 d: Three dishes were allowed to continue
growing without addition of the inhibitor (o), while compounds 5
(ᮀ), 6 (∆), 7 (᭿) and 8 (᭡) were added to the other dishes at a
final concentration of 100 µ. At different time points sample cells
were collected for determination of cell number.
The mechanism of action of these inhibitors was also
investigated in a nucleotide-dissociation assay (Figure 5B)
by measuring the release of fluorescent nucleotide by the
p21 h-Ras·MANT-GDP complex in the presence of GDP
and exchange factor C-Cdc25Mm
.
The dissociation rate of p21 h-Ras in complex with
MANT-GDP was partially reduced in the presence of com-
pounds SCH-54292, 5 and 6, and drastically reduced in the
presence of compound 7. Compounds 8, 9 and 10 only
slightly inhibited the GDP dissociation. Both the nucleotide
exchange and dissociation experiments strongly suggest the
importance of the inhibitor’s benzyl group for the interac-
tion with Ras: molecules 9 and 10, which lack this group,
did not show any interaction with the protein.
In order to study the in vivo inhibition of the Ras-medi-
ated signalling, preliminary experiments were performed to
measure the level of activation of MAPK after addition of
the inhibitors 5, 6, 7 and 8 to normal NIH3T3 fibroblasts
growing in 10% serum. As shown in Figure 7, a pro-
nounced decrease in phospho-MAPK expression was ob-
served after 2 h at a final concentration of 100 µ for com-
pound 7, a less pronounced decrease was observed for com-
Effect of Ras Inhibitors on Mammalian Cells
In order to investigate a specific effect of Ras-mediated pound 8, while no difference in phospho-MAPK expression
signalling in vivo, inhibition of mammalian cell growth by was observed for compounds 5 and 6 compared with the
compounds 5, 6, 7 and 8 was evaluated both in normal cells control. Since compounds 5 and 6 were the most potent
and in cells transformed by k-Ras (Arg12).[11] Compounds inhibitors in vitro, the fact that they are less active than 7
5 and 6 inhibited only slightly the growth of both cell lines and 8 in cell growth inhibition and completely inactive in
when added at a concentration of 100 µ, while growth was decreasing the phospho-MAPK expression suggests that
completely blocked in both cell lines using compounds 7 the polar sulfonamide group can prevent the crossing of the
and 8 at a same concentration (Figure 6).
cell membrane.
Figure 7. Assay of MAPK activation. Lysates (14 µg of total proteins) were separated by SDS-PAGE, transferred to nitrocellulose, and
immunodecorated with anti-p42/44 MAPK antibody and anti-phospho-p42/44 MAPK antibody.
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3713

F. Peri, et al.
FULL PAPER
Conclusions
The data that we have presented here allow some general
considerations on the structure-activity relationship of bicy-
clic Ras inhibitors. The epitope mapping through STD-
NMR experiments for compounds 5–8 clearly points out
that both the phenylhydroxylamino and the benzyl moieties
of the ligands directly interact with Ras. On the contrary,
the protons on the bicyclic scaffold do not interact directly
with the protein. Both the benzyl and the phenylhydrox-
Scheme 3. Hydrogen numbering in the bicyclic scaffold.
Methyl D-Arabinofuranoside (11): To a stirred suspension of -ar-
abinose (5 g, 33.304 mmol) in dry CH3OH (200 mL), acetyl chlo-
ride was added dropwise at room temperature under argon. After
24 h, Amberlite IRN78 OH– resin was added to neutralise the acid,
ylamino groups can therefore be considered pharmacoph- and the mixture was stirred for 5 min. The resin was removed by
filtration, and the solvent was evaporated. The product was puri-
fied by flash column chromatography (9.75:0.25, EtOAc/CH3OH)
affording 11 (4.96 g, 90% yield) as a yellow oil (mixture of α and
β anomers).[12]
ores. The observation that compounds 9 and 10, which lack
both benzyl groups, are totally inactive in inhibiting nucleo-
tide exchange and dissociation and Ras-dependent cellular
proliferation, supports our hypothesis that the phenylhy-
droxylamino group must be accompanied by another aro-
matic moiety to have binding and biological activity.
The -arabinose-derived bicyclic scaffold has the func-
tion of orienting the pharmacophore groups without di-
rectly interacting with the protein. STD-NMR experiments
showed that the inhibitor orientation in the binding pocket
of Ras was not strongly affected by the C-2 configuration
and the nature (amide or sulfonamide) of the C-2 linker.
However, these molecules have different behaviors in in vi-
tro tests on p21 h-Ras: while 5 and 6 are more potent in
blocking nucleotide exchange, compound 7 is the strongest
inhibitor of the GDP dissociation. Moreover, only amide-
containing molecules 7 and 8 are active in inhibiting cellular
growth in both normal and k-Ras-transformed mammalian
cells, while sulfonamides 5 and 6 are inactive. These data
suggest that, while interacting more strongly with purified
Ras, the sulfonamide moiety can prevent cellular uptake,
thus decreasing their potency in cells. We are presently in-
vestigating the molecular mechanism of action of these in-
hibitors, with particular attention to their interactions with
effectors and exchange factors.
1-(D-Arabinofuranosyl)-2-propene (12): To a stirred solution of
methyl -arabinofuranoside (11) (4.5 g, 27.41 mmol) in dry
CH3CN (9 mL), BTSFA (16.3 mL, 61.670 mmol) was added at
100 °C under argon (balloon). After 3 h, the reaction mixture was
cooled to room temperature. ATMS (6.5 mL, 41.115 mmol) and
trimethylsilyl trifluoromethanesulfonate (TMSOTf) (2.5 mL,
13.705 mmol) were added at 0 °C, and the reaction mixture was
stirred at room temperature for 1 h. Water (120 mL) was added
slowly to hydrolyze TMS ethers, the mixture was neutralised with
1  NaOH, and concentrated in vacuo. The product was purified
by flash column chromatography (1:9, petroleum ether/EtOAc)
providing 12 (4.65 g, 97%) as a light green oil (mixture of α and β
anomers; α/β = 1:1.5, as determined from the integration ratio of
the 1H NMR signals). HRMS (FT-ICR): C8H14O4Na: calcd.
197.0790; found 197.0811 [M+Na]+. The following signal attri-
butions were obtained from the spectra of the mixture. α Anomer:
1H NMR (CD3OD): δ = 5.88 (tdd, J = 16.0, 6.9, 2.1 Hz, 1 H), 5.12
(ddd, J = 17.2, 3.6, 1.5 Hz, 1 H), 5.06 (m, 1 H), 3.95 (m, 1 H), 3.79
(m, 3 H), 3.68 (dd, J = 11.7, 3.4 Hz, 1 H), 3.60 (dd, J = 11.8,
5.4 Hz, 1 H), 2.39 (m, 2 H) ppm. 13C NMR (100 MHz, CD3OD):
δ = 135.5, 117.4, 84.7, 83.6, 81.6, 78.8, 63.3, 38.7 ppm. β Anomer:
1H NMR (CD3OD): δ = 5.87 (tdd, J = 17.2, 10.2, 7.0 Hz, 1 H),
5.13 (ddd, J = 17.2, 3.5, 1.5 Hz, 1 H), 5.03 (m, 1 H), 3.97 (m, 1
H), 3.95 (dd, J = 7.0, 3.2 Hz, 1 H), 3.82 (dd, J = 3.0 Hz, 1.0 Hz, 1
H), 3.75 (ddd, J = 4.8, 3.8, 2.6 Hz, 1 H), 3.67 (dd, J = 11.4, 3.8 Hz,
1 H), 3.64 (dd, J = 11.4, 4.9 Hz, 1 H), 2.40 (m, 2 H) ppm. 13C NMR
(CD3OD): δ = 135.9, 117.1, 87.3, 82.5, 80.3, 78.5, 63.5, 34.3 ppm.
Experimental Section
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-4-Hydroxy-5-(hy-
droxymethyl)-2-(iodomethyl)hexahydrofuro[3,2-b]furan (R-13 and S-
13): To a stirred solution of 12 (mixture of anomers, 4 g, 23 mmol)
in dry THF (200 mL), NIS (7.7 g, 34.4 mmol) was added at 90 °C
under argon. After 10 min, the β anomer had reacted completely
according to TLC analysis (eluent: 3:7, petroleum ether/EtOAc),
while the α anomer had not reacted, as expected.[3,4] The mixture
was cooled to room temperature, Na2S2O3 was added to reduce
excess iodine, and the suspension was vigorously stirred until it
became colourless. The product was purified by flash column
chromatography (2:8, petroleum ether/EtOAc) to give 13 (4 g, 58%)
as a light green oil [mixture of diastereomers, (R)/(S) = 2.5:1, as
determined from the integration ratio of the 1H NMR signals]. The
α anomer was recovered unreacted. The following signal attri-
Chemistry
General Procedures: All solvents were dried with molecular sieves
(4 Å, Fluka) for at least 24 h prior to use. When dry conditions
were required, the reactions were performed under argon. Thin-
layer chromatography (TLC) was performed on silica gel 60 F254
plates (Merck) with UV detection, or by using a developing solu-
tion of concd. H2SO4/EtOH/H2O (5:45:45), followed by heating at
180 °C. Flash column chromatography was performed on silica gel
230–400 mesh (Merck). Mixtures of petroleum ether (boiling range
40–60 °C) and ethyl acetate were used as eluents. 1H and 13C NMR
spectra were recorded with a Varian 400 MHz Mercury instrument
at 300 K unless otherwise stated. Chemical shifts are reported in
ppm downfield from TMS as internal standard; hydrogen number-
ing is shown in Scheme 3. Mass spectra were recorded with a Fou- butions were obtained from the spectra of the mixture. HRMS (FT-
rier Transform Ion Cyclotron Resonance (FT-ICR) instrument
(model APEXII, Bruker Daltonics), equipped with a 4.7 T cryom-
agnet (Magnex). Optical rotations were measured at ambient tem-
perature, using the sodium-D line, with a P3002 electronic polarim-
eter (A. Krüss, Germany).
ICR): C8H13IO4Na: calcd. 322.9756; found 322.9743 [M+Na]+.
C8H13IO4 (300.1): calcd. C 32.02, H 4.37; found C 32.11, H 4.30.
R-13: 1H NMR (CD3OD): δ = 4.73 (bt, J = 4.3 Hz, 1 H, 6a-H),
4.47 (dd, J = 4.2, 1.6 Hz, 1 H, 3a-H), 4.04 (m, 1 H, 2-H), 3.94 (dd,
J = 5.3, 1.4 Hz, 1 H, 4-H), 3.70 (m, 2 H, 5-H, 2Јa-H), 3.58 (dd, J
3714
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720

Sugar-Derived Ras Inhibitors: NMR Spectroscopy and Biological Evaluation
FULL PAPER
= 12.5, 7.1 Hz, 1 H, 2Јb-H), 3.32 (m, 2 H, 1Јa-H, 1Јb-H), 2.28 (dd,
J = 13.6, 5.3 Hz, 1 H, 1a-H), 1.67 (ddd, J = 13.7, 9.6, 4.6 Hz, 1 H,
1b-H) ppm. 13C NMR (CD3OD): δ = 92.6, 87.9, 84.3, 79.2, 78.6,
S-16): To a stirred solution of compound 14 (mixture of dia-
stereomers, 700 mg, 3.25 mmol) in dry DMF, benzyl bromide
(1.55 mL, 13.018 mmol) and NaH (60% in oil, 520 mg,
13.01 mmol) were added in three portions at room temperature
over a period of 20 min. After 15 min, the reaction was quenched
by adding ethanol, and the solvents were evaporated. The product
was purified by flash column chromatography (9.25:0.75, petro-
leum ether/EtOAc) obtaining 16 (1.25 g, 90.5%) as a yellow oil
[mixture of diastereomers, (R)/(S) = 2:1 as determined from the
integration ratio of the 1H NMR signals]. HRMS (FT-ICR):
C22H25N3O4Na: calcd. 418.1743; found 418.1735 [M+Na]+.
C22H25N3O4 (418.17): calcd. C 66.82, H 6.37, N, 10.63; found C
1
63.0, 40.5, 9.7 ppm. S-13: H NMR (CD3OD), selected signals: δ
= 4.35 (dd, J = 4.3, 1.1 Hz, 1 H, H-3a), 4.20 (m, 1 H, H-2), 3.63
(dd, J = 12.5, 7.0 Hz, 1 H, 2Јb-H), 2.03 (dd, J = 14.0, 4.4 Hz, 1 H,
1b-H) ppm. 13C NMR (CD3OD): δ = 93.7, 89.3, 84.5, 82.1, 78.5,
63.2, 38.8, 10.3 ppm.
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-2-(Azidomethyl)-4-
hydroxy-5-(hydroxymethyl)-hexahydrofuro[3,2-b]furan (R-14 and S-
14): To a stirred solution of 13 (mixture of diastereomers, 2.2 g,
7.33 mmol) in dry DMF (35 mL) tetrabutylammonium azide[3] was
added at 70 °C under argon. After 72 h, the reaction mixture was
concentrated in vacuo without heating. The product was purified
by flash column chromatography (2:8, petroleum ether/EtOAc) re-
covering 14 (1.41 g, 90%) as a yellow oil [mixture of diastereomers,
1
67.53, H 7.01, N 11.21. R-16: H NMR (CDCl3): δ = 7.4–7.2 (m,
10 H, aromatic protons), 4.79 (bt, J = 5.2 Hz, 1 H, 6a-H), 4.71 (m,
1 H, 3a-H), 4.7–4.5 (2ABq, 4 H, benzyl), 4.29 (m, 1 H, 2-H), 4.03
(ddd, J = 6.2, 6.2, 3.8 Hz, 1 H, 5-H), 3.86 (dd, J = 6.2, 1.2 Hz, 1
H, 4-H), 3.61 (dd, J = 10.5, 3.8 Hz, 1 H, 2Јa-H), 3.59 (dd, J = 10.5,
6.2 Hz, 1 H, 2Јb-H), 3.52 (dd, J = 13.0, 3.5 Hz, 1 H, 1Јa-H), 3.24
(dd, J = 13.0, 3.5 Hz, 1 H, 1Јb-H), 2.19 (dd, J = 13.4, 5.2 Hz, 1 H,
1a-H), 1.78 (ddd, J = 13.4, 10.2, 6.3 Hz, 1 H, 1b-H) ppm. 13C
NMR (CDCl3): δ = 88.72, 85.60, 83.53, 83.35, 77.70, 73.41, 72.10,
1
(R)/(S) = 2.5:1, as determined from the integration ratio of the H
NMR signals]. HRMS (FT-ICR): C8H13N3O4Na: calcd. 238.0804;
found 238.0789 [M+Na]+. C8H13N3O4 (238.08): calcd. C 44.65, H
6.09, N 19.53; found C 44.59, H 6.12, N 19.49. R-14: 1H NMR
(CD3OD): δ = 4.73 (bt, J = 4.4 Hz, 1 H, 6a-H), 4.44 (dd, J = 4.2,
1 H, 1.6 Hz, 3a-H), 4.28 (m, 1 H, 2-H), 3.97 (dd, J = 5.3, 1.4 Hz,
1 H, 4-H), 3.73–3.69 (m, 2 H, H-2Јa, 5-H), 3.59 (dd, J = 12.5,
7.0 Hz, 1 H, 2Јb-H), 3.48 (dd, J = 13.1, 3.3 Hz, 1 H, 1Јa-H), 3.21
(dd, J = 13.1, 5.4 Hz, 1 H, 1Јb-H) 2.10 (dd, J = 13.4, 5.4 Hz, 1 H,
1a-H), 1.78 (ddd, J = 13.4, 10.0, 4.7 Hz, 1 H, 1b-H) ppm. 13C
NMR (CD3OD): δ = 92.3, 88.0, 84.2, 79.1, 78.6, 63.1, 54.7,
36.7 ppm. S-14: 1H NMR (CD3OD): δ = 4.69 (bt, J = 5.0 Hz, 1 H,
6a-H), 4.32 (d, J = 4.1 Hz, 1 H, 3a-H), 4.18 (m, 1 H, 2-H), 4.04
(d, J = 5.5 Hz, 1 H, 4-H), 3.73–3.69 (m, 2 H, 2Јa-H, 5-H), 3.65
(dd, J = 11.8, 6.4 Hz, 1 H, 2Јb-H), 3.49 (dd, J = 12.9, 4.0 Hz, 1 H,
1Јa-H), 3.27 (dd, J = 12.9, 3.8 Hz, 1 H, 1Јb-H), 2.25 (ddd, J = 14.1,
8.4, 6.0 Hz, 1 H, 1a-H) 1.88 (dd, J = 14.1, 5.3 Hz, 1 H, 1b-H) ppm.
13C NMR (CD3OD): δ = 93.4, 89.7, 84.3, 81.0, 78.3, 63.2, 55.8,
36.4 ppm.
1
70.18, 53.60, 35.82 ppm. S-16: H NMR (CDCl3): δ = 7.4–7.2 (m,
10 H, aromatic protons), 4.72 (bt, J = 4.4 Hz, 1 H, 6a-H), 4.65–
4.45 (m, 5 H, 3a-H, benzyl), 4.24 (m, 1 H, 2-H), 4.03 (ddd, J = 6.0,
3.7 Hz, 1 H, 5-H), 3.94 (bd, J = 6.0 Hz, 1 H, 4-H), 3.65 (dd, J =
10.3, 3.0 Hz, 1 H, 2Јa-H), 3.58 (dd, J = 10.3, 5.9 Hz, 1 H, 2Јb-H),
3.48 (dd, J = 12.5, 8.1 Hz, 1 H, 1Јa-H), 3.16 (dd, J = 12.5, 4.0 Hz,
1 H, 1Јb-H), 2.20 (ddd, J = 14.1, 8.4, 5.6 Hz, 1 H, 1a-H) 1.91 (m,
1 H, 1b-H) ppm. 13C NMR (CDCl3): δ = 90.00, 85.40, 85.03, 83.63,
80.23, 73.64, 72.57, 70.00, 55.28, 35.80 ppm.
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-2-(Aminomethyl)-4-
(benzyloxy)-5-methoxyhexahydrofuro[3,2-b]furan (R-17 and S-17):
A solution of compound 16 (740 mg, 1.87 mmol) in 4:1 Ac2O/TFA
(16 mL) was stirred at room temperature for 90 min. The reaction
was quenched by adding a mixture of ice and 1  NaOH, and the
product was extracted with AcOEt. After the usual workup, the
crude mixture was recovered as a yellow oil. The crude mixture was
dissolved in dry MeOH (30 mL) and metallic Na was added in a
catalytic amount under argon. The solution was stirred at room
temperature for 30 min. Amberlite IRA-120 H+ resin was added,
and the mixture was stirred for 10 min. Then the resin was removed
by filtration, and the solvent was evaporated. After the usual
workup and chromatography (5.5:4.5, petroleum ether/AcOEt), a
mixture of (2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-2-(azi-
domethyl)-4-(benzyloxy)-5-methoxyhexahydrofuro[3,2-b]furan[4]
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-2-(Aminomethyl)-4-
hydroxy-5-(hydroxymethyl)-hexahydrofuro[3,2-b]furan (R-15 and S-
15): A solution of compound 14 (mixture of diastereomers, 500 mg,
2.35 mmol) in THF (20 mL) was treated with triphenylphosphane
(1.86 g, 6.97 mmol) and H2O (1.7 mL, 93 mmol), and was stirred
at 60 °C overnight. The reaction mixture was concentrated in
vacuo. The product was purified by flash column chromatography
(8:2, EtOAc/CH3OH, 1% TEA) affording 15 (377 mg, 85%) as a
yellow oil [mixture of diastereomers, (R)/(S) = 2:1, as determined
from the integration ratio of the 1H NMR signals]. HRMS (FT-
ICR): C8H15NO4Na: calcd. 212.0899; found 212.0911 [M+Na]+.
C8H15NO4 (212.08): calcd. C 50.78, H 7.99, N 7.40; found C 50.82,
H 7.90, N 7.49. R-15: 1H NMR (CD3OD): δ = 4.71 (bt, J = 4.6 Hz,
1 H, 6a-H), 4.42 (dd, J = 4.3, 1.8 Hz, 1 H, 3a-H), 4.09 (m, 1 H, 2-
H), 3.94 (dd, J = 5.8, 1.8 Hz, 1 H, 4-H), 3.88–3.57 (m, 3 H, 2Јa-H,
2Јb-Hb, 5-H), 2.81 (dd, J = 13.1, 3.9 Hz, 1 H, 1Јa-H), 2.68 (dd, J
= 13.1, 6.8 Hz, 1 H, 1Јb-H), 2.09 (dd, J = 13.4, 5.1 Hz, 1 H, 1a-
H), 1.61 (ddd, J = 13.5, 10.3, 4.8 Hz, 1 H, 1b-H) ppm. 13C NMR
(CD3OD): δ = 91.9, 87.6, 84.1, 80.3, 78.6, 63.0, 45.8, 37.2 ppm. S-
1
[(R)/(S) = 2:1, as determined from the integration ratio of the H
NMR signals] was provided as a colourless oil (463 mg, 84.5% for
the two steps). To a solution of this diastereomeric mixture
(450 mg, 1.473 mmol) in THF (15 mL), triphenylphosphane
(1.18 g, 4.420 mmol) and water (1 mL, 59 mmol) were added, and
the reaction mixture was stirred at 60 °C overnight. The reaction
mixture was concentrated in vacuo. The product was purified by
flash column chromatography (7.5.2.5, EtOAc/MeOH with 1%
TEA Ǟ 1:1, EtOAc/MeOH with 1% TEA), affording 17 (377 mg,
94%) as colourless oil [mixture of diastereomers, (R)/(S) = 2:1, as
determined from the integration ratio of the 1H NMR signals].
HRMS (FT-ICR): C15H21NO4Na: calcd. 302.1368; found 302.1354
[M+Na]+. C15H21NO4 (302.13): calcd. C 64.50, H 7.58, N 5.01;
found C 64.45, H 7.53, N 5.08. R-17: 1H NMR (CD3OD): δ =
7.34–7.32 (m, 5 H, aromatic protons), 4.71 (t, J = 4.5 Hz, 1 H, 6a-
H), 4.68, 4.55 (ABq, J = 11.7 Hz, 2 H, benzyl), 4.59 (dd, J = 1.1,
4.3 Hz, 1 H, 3a-H), 4.09 (m, 1 H, 2-H), 3.90–3.79 (m, 2 H, 4-H, 5-
H), 3.69 (dd, J = 11.8, 3.2 Hz, 1 H, 2Јa-H), 3.58 (dd, J = 11.7,
1
15: H NMR (CD3OD), selected signals: δ = 4.68 (bt, J = 4.4 Hz,
1 H, 6a-H), 4.28 (d, J = 4.3 Hz, 1 H, 3a-H), 2.85 (dd, J = 13.2,
6.2 Hz, 1 H, 1Јa-H), 2.27 (ddd, J = 14.2, 8.4, 6.0 Hz, 1 H, 1a-H)
1.85 (dd, J = 14.4, 5.9 Hz, 1 H, 1b-H) ppm. 13C NMR (CD3OD):
δ = 92.9, 89.8, 84.4, 81.7, 77.8, 63.0, 46.3, 36.3 ppm.
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-2-(Azidomethyl)-4-
benzyloxy-5-(benzyloxymethyl)-hexahydrofuro[3,2-b]furan (R-16 and
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3715

F. Peri, et al.
5.4 Hz, 1 H, 2Јb-H), 2.78 (dd, J = 13.2, 3.9 Hz, 1 H, 1Јa-H), 2.66 (2S,3aS,4R,5R,6aS)- and (2R,3aS,4R,5R,6aS)-4-(Benzyloxy)-5-
FULL PAPER
(dd, J = 13.2, 6.6 Hz, 1 H, 1Јb-H), 2.08 (dd, J = 13.4, 5.0 Hz, 1 H,
1a-H), 1.61 (ddd, J = 13.5, 10.3, 4.7 Hz, 1 H, 1b-H) ppm. 13C
NMR (CD3OD): δ = 139.1, 129.2, 128.8, 128.6, 128.6, 89.6, 86.3,
methoxy-2-[(4-nitrophenyl)sulfonamidomethyl]hexahydrofuro[3,2-b]-
furan (20 and 21): To a stirred solution of compound 17 (15 mg,
0.053 mmol) in dry CH2Cl2 (0.7 mL), TEA (9 µL, 0.064 mmol) and
86.0, 84.6, 80.9, 72.89, 63.1, 45.9, 37.1 ppm. S-17: 1H NMR p-nitrobenzensulfonyl chloride were added at 0 °C under argon. Af-
(CD3OD), selected signals: δ = 4.44 (d, J = 4.0 Hz, 1 H, 3a-H),
2.24 (ddd, J = 14.0, 8.3, 6.1 Hz, 1 H, 1a-H), 1.83 (dd, J = 13.9,
4.9 Hz, 1 H, 1b-H) ppm. 13C NMR (CD3OD) selected signals: δ =
139.06, 90.1, 87.9, 85.6, 84.9, 82.7, 72.94, 63.2, 46.7, 36.3 ppm.
ter 5 h, the reaction was quenched by adding MeOH. The solvents
were evaporated in vacuo and, after flash chromatography (1:1, pe-
troleum ether/EtOAc), pure compounds 20 [(S) diastereomer, 5 mg,
20%] and 21 [(R) diastereomer, 15 mg, 60%] were purified and sep-
arated as pale yellow oils. 20: 1H NMR (CDCl3): δ = 8.31, 8.04
(AAЈXXЈ, J = 8.8 Hz, 4 H, aromatic protons), 7.30–7.22 (m, 5 H,
aromatic protons), 6.38 (dd, J = 6.8, 2.8 Hz, 1 H, NH), 4.65, 4.52
(ABq, J = 11.8 Hz, 2 H, benzyl), 4.65 (dd, J = 5.5, 3.5 Hz, 1 H,
6a-H), 4.40 (d, J = 3.4 Hz, 1 H, 3a-H), 4.31–4.26 (m, 1 H, 2-H),
4.07 (d, J = 5.4 Hz, 1 H, 4-H), 4.00–3.94 (m, 2 H, 2Јa-H, 5-H),
3.72 (dd, J = 11.6, 2.0 Hz, 1 H, 2Јb-H), 3.18 (dt, J = 12.6, 3.0 Hz,
1 H, 1Јa-H), 3.10 (ddd, J = 12.6, 6.9, 3.4 Hz, 1 H, 1Јb-H), 2.63 (br.
s, 1 H, OH), 2.24 (ddd, J = 14.6, 9.4, 5.6 Hz, 1 H, 1a-H), 1.99 (dd,
J = 14.6, 4.6 Hz, 1 H, 1b-H) ppm. 13C NMR (CDCl3): δ = 149.65,
145.6, 137.1, 128.4, 128.2, 127.9, 127.5, 124.1, 89.2, 86.5, 83.5, 83.0,
77.9, 72.4, 61.3, 46.0, 33.9 ppm. [α]2D0 = +14.6 (c = 0.5, CHCl3).
HRMS (FT-ICR): C21H24N2O8S Na: calcd. 487.1151; found
487.1143 [M + Na]+. C21H24N2O8S (464.2): calcd. C 54.30, H 5.21,
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-4-Hydroxy-5-meth-
oxy-2-[(4-nitrophenyl)sulfonamidomethyl]hexahydrofuro[3,2-b]furan
(18): To a stirred solution of compound 15 (180 mg, 0.951 mmol)
in dry pyridine (13.5 mL), p-nitrobenzensulfonyl chloride was
added at 0 °C under argon. After 6 h, the reaction was quenched
by adding CH3OH. The solvents were evaporated in vacuo and,
after flash chromatography (1:9, petroleum ether/EtOAc), 18
(228 mg, 64 %) was recovered as pale yellow oil [diastereomeric
mixture, (R)/(S) = 3:1, as determined from the integration ratio of
1
the H NMR signals). HRMS (FT-ICR): C14H18N2O8SNa: calcd.
397.0682; found 397.0674 [M + Na]+. C14H18N2O8S (374.2): C
44.92, H 4.85, N 7.48; found C 44.9, H 4.89, N 7.39. R-18: 1H
NMR (CD3OD): δ = 8.39, 8.07 (AAЈXXЈ, J = 8.8 Hz, 4 H, aro-
matic protons), 4.65 (bt, J = 4.5 Hz, 1 H, 6a-H), 4.28 (dd, J = 4.3,
1.8 Hz, 1 H, 3a-H), 4.06 (m, 1 H, 2-H), 3.81 (dd, J = 5.5, 1.5 Hz,
1 H, 4-H), 3.68–3.61 (m, 2 H, 2Јa-H, 5-H), 3.50 (dd, J = 12.5,
7.1 Hz, 1 H, 2Јb-H), 3.15 (dd, J = 13.6, 3.9 Hz, 1 H, 1Јa-H), 3.02
(dd, J = 13.6, 5.8 Hz, 1 H, 1Јb-H), 2.04 (dd, J = 13.5, 5.2 Hz, 1 H,
1a-H), 1.65 (ddd, J = 13.5, 10.1, 4.8 Hz, 1 H, 1b-H) ppm. 13C
NMR (CDCl3): δ = 151.1, 147.9, 129.2, 125.2, 92.1, 87.8, 84.0,
78.6, 78.5, 63.0, 47.2, 37.0 ppm. S-18: 1H NMR (CD3OD) selected
signals: δ = 4.62 (bt, J = 4.6 Hz, 1 H, 6a-H), 4.22 (d, J = 4.0 Hz,
1 H, 3a-H), 3.95 (d, J = 5.3 Hz, 1 H, 4-H), 3.54 (dd, J = 11.7,
5.8 Hz, 1 H, 2Јb-H), 2.18 (ddd, J = 14.1, 8.4, 5.8 Hz, 1 H, 1a-H),
1.84 (dd, J = 14.1, 4.8 Hz, 1 H, 1b-H) ppm. 13C NMR (CDCl3),
selected signals: δ = 129.2, 93.2, 89.7, 84.4, 80.4, 78.1, 63.2,
36.3 ppm.
1
N 6.03; found: C 54.22, H 5.29, N 5.97. 21: H NMR (CDCl3): δ
= 8.32, 8.03 (AAЈXXЈ, J = 8.6 Hz, 4 H, aromatic protons), 7.37–
7.29 (m, 5 H, aromatic protons) 5.33 (t, J = 6.1 Hz, 1 H, NH), 4.72
(bt, J = 4.5 Hz, 1 H, 6a-H), 4.63, 4.50 (ABq, J = 11.6 Hz, 2 H,
benzyl), 4.55 (dd, J = 4.4, 1.5 Hz, 1 H, 3a-H), 4.12 (m, 1 H, 2-H),
3.86–3.76 (m, 3 H, 2Јa-H, 4-H, 5-H), 3.60 (dd, J = 12.0, 4.8 Hz, 1
H, 2Јb-H), 3.27 (ddd, J = 13.0, 6.2, 3.3 Hz, 1 H, 1Јa-H), 3.03 (td,
J = 12.9, 3.0 Hz, 1 H, 1Јb-H), 2.10 (dd, J = 13.5, 5.0 Hz, 1 H, 1a-
H), 2.05 (br. s, 1 H, OH), 1.68 (ddd, J = 13.5, 10.3, 4.7 Hz, 1 H,
1b-H) ppm. 13C NMR (CDCl3): δ = 149.8, 145.5, 137.1, 128.4,
128.0, 127.8, 127.5, 124.3, 88.8, 84.4, 84.1, 83.0, 76.8, 72.2, 62.1,
45.9, 35.7 ppm. [α]2D0: +7.8 (c 0.5, CHCl3). HRMS (FT-ICR):
C21H24N2O8S Na: calcd. 487.1151; found 487.1164 [M +Na]+.
C21H24N2O8S (487.11): calcd. C 54.30, H 5.21, N 6.03; found C
54.26, H 5.25, N 6.00.
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-4-Hydroxy-5-meth-
oxy-2-[(4-nitrophenyl)carboxamidomethyl]hexahydrofuro[3,2-b]furan
(19): To a stirred solution of compound 15 (200 mg, 1.057 mmol) in
dry DMF (16 mL), HOBt (214 mg, 1.585 mmol), DIPEA (545 µL,
3.171 mmol) and p-nitrobenzoic acid (212 mg, 1.268 mmol) were
added under argon. DIC (246 µL, 1.585 mmol) was added at 0 °C,
and the reaction mixture was stirred at room temperature for 24 h.
The solvent was evaporated in vacuo, and the product was purified
by flash column chromatography (8.5:1.5, toluene/CH3OH) afford-
ing 19 (270 mg, 75.5 %) as a pale yellow oil [mixture of dia-
stereomers, (R)/(S) = 2:1, as determined from the integration ratio
(2S,3aS,4R,5R,6aS)- and (2R,3aS,4R,5R,6aS)-4-(Benzyloxy)-5-
methoxy-2-[(4-nitrophenyl)carboxamidomethyl]hexahydrofuro[3,2-
b]furan (22 and 23): To a stirred solution of compound 17 (200 mg,
0.716 mmol) in dry DMF (10 mL), HOBt (145 mg, 1.074 mmol),
DIPEA (368 µL, 2.148 mmol) and 4-nitrobenzoic acid (144 mg,
0.86 mmol) were added under argon. DIC (166 µL, 1.074 mmol)
was added at 0 °C, and the reaction mixture was stirred at room
temperature for 24 h. The solvent was evaporated in vacuo and,
after flash chromatography (4.5:5.5, toluene/EtOAc), pure com-
pounds 22 [(S) diastereomer, 59 mg, 20%] and 23 [(R) diastereomer,
of the 1H NMR signals]. HRMS (FT-ICR): C15H18N2O7Na: calcd. 150 mg, 50%] were purified and separated as pale yellow oils. 22:
361.1012; found 361.1033 [M+Na]+. C15H18N2O7 (338.6): calcd. C 1H NMR (CDCl3): δ = 8.23, 8.01 (AAЈXXЈ, J = 8.9 Hz, 4 H, aro-
53.25, H 5.36, N 8.28; found C 53.19, H 5.29, N 8.36. R-19: 1H matic protons), 7.65 (dd, J = 6.8, 2.8 Hz, 1 H, NH), 7.35–7.28 (m,
NMR (CD3OD): δ = 8.27, 7.99 (AAЈXXЈ, J = 8.9 Hz, 4 H, aro-
matic protons), 4.72 (bt, J = 4.7 Hz, 1 H, 6a-H), 4.47 (dd, J = 4.4,
1.9 Hz, 1 H, 3a-H), 4.29 (m, 1 H, 2-H), 3.96 (dd, J = 5.8, 1.8 Hz,
5 H, aromatic protons), 4.69 (dd, J = 5.0, 3.3 Hz, 1 H, 6a-H), 4.66,
4.53 (ABq, J = 11.7 Hz, 2 H, benzyl), 4.41 (m, 2 H, 2-H, 3a-H),
4.09 (d, J = 5.0 Hz, 1 H, 4-H), 4.02 (m, 1 H, 5-H), 3.80 (m, 2 H,
1 H, 4-H), 3.74–3.65 (m, 3 H), 3.62–3.47 (m, 2 H, 1Јa-H, 1Јb-H), 1Јa-H, 2Јa-H), 3.69 (dd, J = 11.9, 2.9 Hz, 1 H, 2Јb-H), 3.53 (td, J
2.17 (dd, J = 13.4, 5.2 Hz, 1 H, 1a-H), 1.68 (ddd, J = 13.5, 10.1,
= 14.3, 3.0 Hz, 1 H, 1Јb-H), 2.82 (br. s, 1 H, OH), 2.27 (ddd, J =
4.8 Hz, 1 H, 1b-H) ppm. 13C NMR (CDCl3): δ = 167.8, 150.6, 14.4, 9.0, 5.6 Hz, 1 H, 1a-H), 2.07 (dd, J = 14.4, 4.8 Hz, 1 H, 1b-
140.9, 129.4, 124.3, 92.0, 87.6, 83.9, 78.5, 78.4, 62.9, 44.4, H) ppm. 13C NMR (CDCl3): δ = 166.3, 149.1, 140.2, 137.2,
37.6 ppm. S-19: 1H NMR (CD3OD), selected signals: δ = 4.69 (bt,
128.514, 128.35, 127.8, 127.5, 123.2, 88.5, 86.8, 83.8, 83.2, 78.6,
J = 5.2 Hz, 1 H, 6a-H), 4.12 (d, J = 5.1 Hz, 1 H, 3a-H), 2.28 (ddd, 72.3, 61.6, 43.4, 33.7 ppm. [α]2D0 = –18.4 (c = 0.5, CHCl3). HRMS
J = 14.1, 8.5, 5.8 Hz, 1 H, 1a-H), 1.95 (dd, J = 14.1, 5.2 Hz, 1 H, (FT-ICR): C22H24N2O7Na: calcd. 451.1481; found 451.1498
1b-H) ppm. 13C NMR (CDCl3): δ = 168.1, 150.5, 141.2, 129.6,
124.2, 93.0, 89.7, 84.5, 80.0, 77.9, 62.9, 45.2, 36.0 ppm.
[M+Na]+. C22H24N2O7 (428.4): calcd. C 61.67, H 5.65, N 6.54;
found C 61.64, H 5.61, N 6.59. 23: 1H NMR (CDCl3): δ = 8.26,
3716
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720

Sugar-Derived Ras Inhibitors: NMR Spectroscopy and Biological Evaluation
FULL PAPER
7.94 (AAЈXXЈ, J = 8.7 Hz, 4 H, aromatic protons), 7.35–7.28 (m,
5 H, aromatic protons), 6.77 (t, J = 4.9 Hz, 1 H, NH) 4.75 (bt, J
= 4.5 Hz, 1 H, 6a-H) 4.68, 4.54 (ABq, J = 11.6 Hz, 2 H, benzyl)
ditions described for compound 5 and, after purification by flash
column chromatography (9:1, toluene/MeOH), afforded 7 (62 mg,
92 %) as a pale yellow amorphous solid. 1H NMR (9:1, D2O/
4.63 (d, J = 4.4 Hz, 1 H, 3a-H), 4.28 (m, 1 H, 2-H), 3.89 (m, 2 H, CD3OD): δ = 7.60, 6.99 (AAЈXXЈ, J = 8.7 Hz, 4 H, aromatic pro-
4-H, 5-H), 3.82 (dd, J = 12.0, 2.2 Hz, 1 H, 2Јa-H), 3.80 (m, 1 H, tons), 7.28 (m, 5 H, aromatic protons), 4.72 (m, 1 H, 6a-H), 4.61,
1Јa-H), 3.65 (dd, J = 12.0, 4.0 Hz, 1 H, 2Јb-H), 3.49 (td, J = 13.8, 4.48 (ABq, J = 11.6 Hz, 2 H, benzyl), 4.52 (bd, J = 4.2 Hz, 1 H,
5.9 Hz, 1 H, 1Јb-H), 2.50 (br. s, 1 H, OH), 2.19 (dd, J = 13.5, 3a-H), 4.27 (m, 1 H, 2-H), 3.84 (m, 2 H, 4-H, 5-H), 3.61 (dd, J =
4.9 Hz, 1 H, 1a-H), 1.64 (ddd, J = 13.6, 10.5, 4.7 Hz, 1 H, 1b-H)
12.3, 3.6 Hz, 1 H, 2Јa-H), 3.54 (dd, J = 13.9, 7.1 Hz, 1 H, 1Јa-H),
ppm. 13C NMR (CDCl3): δ = 165.3, 149.5, 139.6, 137.2, 128.3, 3.51 (m, 1 H, 2Јb-H), 3.38 (dd, J = 14.0, 3.8 Hz, 1 H, 1Јb-H), 2.27
128.1, 127.8, 127.5, 123.6, 88.8, 84.5, 84.2, 83.1, 77.4, 72.1, 62.2,
(ddd, J = 14.5, 8.0, 6.3 Hz, 1 H, 1a-H), 1.86 (dd, J = 14.6, 4.0 Hz,
1 H, 1b-H) ppm. 13C NMR (CD3OD): δ = 170.1, 156.0, 139.1,
129.3, 129.2, 128.8, 128.6, 126.4, 113.3, 90.4, 87.9, 85.8, 84.9, 80.7,
73.0, 63.3, 45.3 36.3 ppm. [α]2D0 = –3.91 (c = 0.7, DMSO). HRMS
(FT-ICR): C22H26N2O6Na: calcd. 437.1689; found 437.1678
[M+Na]+.C22H26N2O6 (414.3): calcd. C 63.76, H 6.32, N 6.76;
found C 63.73, H 6.29, N 6.81.
43.0, 36.1 ppm. [α]2D0 = –17.2 (c = 0.5, CHCl3). HRMS (FT-ICR):
C
22H24N2O7Na: calcd. 451.1481; found 451.1476 [M + Na]+.
C22H24N2O7 (428.2): calcd. C 61.67, H 5.65, N 6.54; found C
61.69, H 5.60, N 6.57.
(2S,3aS,4R,5R,6aS)-4-(Benzyloxy)-2-{[4-(hydroxyamino)phenyl]-
sulfonamidomethyl}-5-methoxyhexahydrofuro[3,2-b]furan (5): To a
stirred solution of compound 20 (40 mg, 0.086 mmol) in THF
(3.7 mL), Pd/C (3.7 mg) was added at 0 °C. After 15 min, the sus-
pension was treated with hydrazine hydrate (8.4 µL, 0.172 mmol)
and stirred at 0 °C for 45 min. The reaction was quenched by add-
ing acetone. The Pd/C was removed by filtration, and the solvents
were concentrated in vacuo. The product was purified by flash col-
umn chromatography (9.5:0.5 toluene/CH3OH) to give 5 (35.5 mg,
92 %) as a pale yellow amorphous solid. 1H NMR (9:1, D2O/
CD3OD): δ = 7.62, 7.00 (AAЈXXЈ, J = 8.8 Hz, 4 H, aromatic pro-
tons), 7.38–7.32 (m, 5 H, aromatic protons), 4.64 (bt, J = 5.0 Hz,
1 H, 6a-H), 4.59, 4.47 (ABq, J = 11.6 Hz, 2 H, benzyl), 4.48 (bd,
J = 4.7 Hz, 1 H, 3a-H), 4.11 (m, 1 H, 2-H), 3.76 (m, 1 H, 5-H)
3.61 (bd, J = 6.1 Hz, 1 H, 4-H), 3.49 (dd, J = 12.2, 3.4 Hz, 1 H,
2Јa-H), 3.26 (dd, J = 12.2, 6.6 Hz, 1 H, 2Јb-H), 2.93 (dd, J = 13.3,
3.9 Hz, 1 H, 1Јa-H), 2.87 (dd, J = 13.3, 8.0 Hz, 1 H, 1Јb-H), 2.18
(ddd, J = 14.4, 8.7, 5.9 Hz, 1 H, 1a-H), 1.75 (dd, J = 14.6, 4.0 Hz,
1 H, 1b-H) ppm. 13C NMR δ = 156.7, 139.0, 131.05, 129.2, 129.1,
128.8, 128.6, 113.1, 90.4, 87.7, 85.9, 84.7, 80.6, 72.9, 63.4, 48.8,
36.3 ppm. [α]2D0 = –12.65 (c = 0.7, DMSO). HRMS (FT-ICR):
(2R,3aS,4R,5R,6aS)-4-(Benzyloxy)-2-{[4-(hydroxyamino)phenyl]-
carboxamidomethyl}-5-methoxyhexahydrofuro[3,2-b]furan (8): The
procedure was carried out as described for the preparation of 5,
starting from 23 (90 mg, 0.209 mmol) and, after purification by
flash column chromatography (9:1, toluene/CH3OH), afforded 8
(82 mg, 95%) as pale yellow amorphous solid. 1H NMR (9:1, D2O/
CD3OD): δ = 7.65, 7.02 (AAЈXXЈ, J = 8.8 Hz, 4 H, aromatic pro-
tons), 7.32–7.26 (m, 5 H, aromatic protons), 4.74 (m, 1 H, 6a-H),
4.63, 4.53 (ABq, J = 11.6 Hz, 2 H, benzyl), 4.62 (m, 1 H, 3a-H),
4.23 (m, 1 H, 2-H), 3.82 (dd, J = 6.4, 1.4 Hz, 1 H, 4-H), 3.79 (m,
1 H, 5-H), 3.65 (dd, J = 12.3, 3.1 Hz, 1 H, 2Јa-H), 3.51 (dd, J =
13.3, 5.8 Hz, 1 H, 2Јb-H), 3.48 (m, 1 H, 1Јa-H), 3.43 (dd, J = 14.2,
6.4 Hz, 1 H, 1Јb-H), 2.11 (dd, J = 13.9, 5.3 Hz, 1 H, 1a-H), 1.68
(ddd, J = 14.2 10.2, 4.9 Hz, 1 H, 1b-H) ppm. 13C NMR (CD3OD):
δ = 170.0, 156.1, 139.0, 129.2, 129.2, 128.8, 128.6, 126.3, 113.3,
89.7, 86.0, 86.0, 84.5, 79.1, 72.9, 63.1, 44.1, 37.6 ppm. [α]2D0
=
–13.59 (c = 0.7, DMSO). HRMS (FT-ICR): C22H26N2O6Na: calcd.
437.1689; found 437.1698 [M+Na]+. C22H26N2O6 (414.3): calcd. C
63.76, H 6.32, N 6.76; found C 63.75, H 6.28, N 6.83.
C
21H26N2O7SNa: calcd. 473.1358; found 473.1367 [M + Na]+.
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-4-Hydroxy-2-{[4-(hy-
droxyamino)phenyl]sulfonamidomethyl}-5-methoxyhexahydrofuro-
[3,2-b]furan (R-9 and S-9): The procedure was carried out as de-
scribed for the preparation of 5, starting from 18 (100 mg,
0.267 mmol) and, after purification by flash column chromatog-
raphy (7.5:2.5, toluene/MeOH), afforded 9 (95 mg, 98%) as pale
yellow amorphous solid [mixture of diastereomers, (R)/(S) = 3:1,
C21H26N2O7S (450.6): calcd. C 55.99, H 5.82, N 6.22; found C
56.07, H 5.77, N 6.29.
(2R,3aS,4R,5R,6aS)-4-(Benzyloxy)-2-{[4-(hydroxyamino)phenyl]-
sulfonamidomethyl}-5-methoxyhexahydrofuro[3,2-b]furan (6): Com-
pound 21 (50 mg, 0.107 mmol) was allowed to react under the con-
ditions described for compound 5 to afford compound 6 (45 mg,
92 %) as a pale yellow amorphous solid. 1H NMR (9:1, D2O/
CD3OD): δ = 7.68, 7.02 (AAЈXXЈ, J = 8.1 Hz, 4 H, aromatic pro-
tons), 7.39–7.34 (m, 5 H, aromatic protons), 4.67 (bt, J = 5.0 Hz,
1 H, 6a-H), 4.60, 4.48 (ABq, J = 11.5 Hz, 2 H, benzyl), 4.42 (bd,
J = 4.7 Hz, 1 H, 3a-H), 3.95 (m, 1 H, 2-H), 3.75 (m, 1 H, 5-H)
3.65 (bd, J = 5.7 Hz, 1 H, 4-H), 3.59 (dd, J = 12.5, 3.4 Hz, 1 H,
2Јa-H), 3.42 (dd, J = 12.5, 5.8 Hz, 1 H, 2Јb-H), 3.14 (dd, J = 13.7,
3.9 Hz, 1 H, 1Јa-H), 2.97 (dd, J = 13.7, 5.3 Hz, 1 H, 1Јb-H), 1.95
(dd, J = 14.3, 3.6 Hz, 1 H, 1a-H), 1.61 (ddd, J = 14.3, 8.7, 5.9 Hz,
1 H, 1b-H) ppm. 13C NMR (CD3OD): δ = 155.6, 138.6, 129.9,
128.8, 128.4, 128.2, 128.1, 112.1, 88.4, 85.7, 85.6, 83.2, 77.9, 71.6,
62.2, 46.7, 36.9 ppm. [α]2D0: –4.75 (c 0.5, DMSO). HRMS (FT-ICR):
C21H26N2O7SNa: calcd. 473.1358; found 473.1342 [M + Na]+.
C21H26N2O7S (450.6): calcd. C 55.99, H 5.82, N 6.22; found C
56.04, H 5.76, N 6.25.
1
as determined from the integration ratio of the H NMR signals].
1
R-9: H NMR (D2O): δ = 7.70, 7.09 (AAЈXXЈ, J = 8.8 Hz, 4 H,
aromatic protons), 4.70 (m, 1 H, 6a-H), 4.29 (dd, J = 4.3, 1.8 Hz,
1 H, 3a-H), 4.02 (m, 1 H, 2-H), 3.82 (dd, J = 5.8, 1.7 Hz, 1 H, 4-
H), 3.71 (ddd, J = 6.4, 6.2, 3.5 Hz, 1 H, 5-H), 3.65 (dd, J = 12.3,
3.5, 3.5 Hz, 1 H, 2Јa-H), 3.49 (dd, J = 12.2, 6.6 Hz, 1 H, 2Јb-H),
3.11 (dd, J = 14.1, 3.6 Hz, 1 H, 1Јa-H), 3.02 (dd, J = 13.6, 5.8 Hz,
1 H, 1Јb-H), 1.98 (dd, J = 14.0, 5.4 Hz, 1 H, 1a-H), 1.64 (ddd, J =
14.3, 10.2, 4.8 Hz, 1 H, 1b-H) ppm. 13C NMR (CD3OD): δ = 156.6,
131.3, 129.1, 113.1, 92.1, 87.7, 84.1, 78.6, 78.5, 63.2, 47.2,
1
37.2 ppm. S-9: H NMR (D2O), selected signals: δ = 4.33 (dd, J =
4.2, 0.8 Hz, 1 H, H-3a), 4.11(m, 1 H, H-2) 3.85 (dd, J = 5.5, 0.7 Hz,
1 H, 4-H), 3.75 (m, 1 H, 5-H), 3.60 (dd, J = 12.1, 3.7 Hz, 1 H, 2Јa-
H), 3.42 (dd, J = 12.1, 6.7 Hz, 1 H, 2Јb-H), 2.21 (ddd, J = 14.3,
8.5, 5.8 Hz, 1 H, 1a-H), 1.75 (dd, J = 14.6, 4.6 Hz, 1 H, H-1b)
ppm. 13C NMR (CD3OD) selected signals: δ = 131.1, 129.2, 113.0,
93.1, 89.6, 84.4, 80.4, 78.3, 63.5, 36.5 ppm. HRMS (FT-ICR):
(2S,3aS,4R,5R,6aS)-4-(Benzyloxy)-2-{[4-(hydroxyamino)phenyl]- C14H20N2O7SNa: calcd. 383.0889; found 383.0877 [M + Na]+.
carboxamidomethyl}-5-methoxyhexahydrofuro[3,2-b]furan (7): Com-
pound 22 (70 mg, 0.163 mmol) was allowed to react under the con-
C14H20N2O7S (360.2): calcd. C 46.66, H 5.59, N 7.77; found C
46.59, H 5.63, N 7.81.
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3717

F. Peri, et al.
(2R,3aS,4R,5R,6aS)- and (2S,3aS,4R,5R,6aS)-2-{[4-(Hy- SCH-54292: The procedure was carried out as described for the
FULL PAPER
droxyamino)phenyl]carboxamidomethyl}-5-methoxyhexahydrofuro-
[3,2-b]furan (R-10 and S-10): The procedure was carried out as de-
scribed for the preparation of 5, starting from 19 (120 mg,
0.355 mmol) and, after purification by flash column chromatog-
raphy (7.5:2.5, toluene/CH3OH), pure 10 (113 mg, 98%) was ob-
tained as a pale yellow amorphous solid [mixture of diastereomers,
(R)/(S) = 2:1, as determined from the integration ratio of the 1H
NMR signals]. HRMS (FT-ICR): C15H20N2O6Na: calcd. 347.1219;
found 347.1234 [M+Na]+. C15H20N2O6 (324.3): calcd. C 55.55, H
6.22, N 8.64; found C 55.63, H 6.29, N 8.57. R-10: 1H NMR (D2O):
δ = 7.65, 7.02 (AAЈXXЈ, J = 8.8 Hz, 4 H, aromatic protons), 4.77
(bt, J = 4.6 Hz, 1 H, 6a-H), 4.49 (dd, J = 4.3, 1.7 Hz, 1 H, 3a-H),
4.29 (m, 1 H, 2-H), 3.98 (dd, J = 5.8, 1.7 Hz, 1 H, 4-H), 3.76 (ddd,
preparation of 5, starting from 25 (40 mg, 0.075 mmol) and, after
purification by flash column chromatography (9:1, toluene/
CH3OH), afforded SCH-54292 (36.5 mg, 93%) as a pale yellow
1
amorphous solid. H NMR (9:1, D2O/CD3OD): δ = 7.88 (d, J =
7.3 Hz, 1 H, c2-H), 7.86 (d, J = 8.8 Hz, 1 H, c3-H), 7.84 (d, J =
8.4 Hz, 1 H, c1-H), 7.82, 7.06 (AAЈXXЈ, J = 8.8 Hz, 4 H, aromatic
protons), 7.53 (t, J = 7.6 Hz, 1 H, d-H), 7.43 (t, J = 7.5 Hz, 1 H),
7.30 (d, J = 2.4 Hz, 1 H, f-H), 7.16 (dd, J = 8.9, 2.6 Hz, 1 H, f-H),
5.07 (d, J = 9.2 Hz, 1 H, 1-H), 4.33 (m, 2 H, jc-H, jd-H), 3.78 (ddd,
J = 16.0, 8.0, 5.0 Hz, 1 H, ja-H), 3.67 (t, J = 9.0 Hz, 1 H, 2-H),
3.60 (dd, J = 12.3, 2.1 Hz, 1 H, 6a-H), 3.59 (ddd, J = 16.0, 8.0,
5.0 Hz, 1 H, jb-H), 3.58 (t, J = 8.9 Hz, 1 H, 3-H), 3.51 (dd, J =
12.3, 5.4 Hz, 1 H, 6b-H), 3.41 (m, 1 H, 5-H), 3.33 (m, 1 H, 4-H)
J = 6.2, 6.2, 3.4 Hz, 1 H, 5-H), 3.71 (dd, J = 12.2, 3.4 Hz, 1 H, ppm. 13C NMR ([D6]DMSO): δ = 155.7, 155.4, 134.1, 129.4, 129.2,
2Јa-H), 3.58 (dd, J = 12.2, 6.5 Hz, 1 H, 2Јb-H), 3.49 (m, 2 H, 1Јa- 128.5, 127.5, 127.4, 126.6, 126.5, 123.8, 118.4, 111.3, 106.6, 86.7,
H, 1Јb-H), 2.14 (dd, J = 14.4, 5.2 Hz, 1 H, 1a-H), 1.72 (ddd, J =
14.4, 10.2, 4.9 Hz, 1 H, 1b-H) ppm. 13C NMR (CD3OD): δ = 169.0,
155.0, 128.2, 125.4, 112.2, 91.0, 86.6, 82.9, 77.9, 77.6, 62.0, 43.1,
36.6 ppm. S-10: 1HNMR (D2O), selected signals: δ = 4.49 (dd, J =
4.3, 1.7 Hz, 1 H, 3a-H), 4.05 (bd, J = 5.2 Hz, 1 H, 4-H), 3.82 (ddd,
J = 6.5, 5.4, 4.0 Hz, 1 H, 5-H), 3.68 (dd, J = 12.1, 3.8 Hz, 1 H,
2Јa-H), 3.58 (dd, J = 12.1, 6.7 Hz, 1 H, 2Јb-H), 3.53 (m, 2 H, 1Јa-
H, 1Јb-H), 2.32 (ddd, J = 8.3, 6.2 Hz, 1 H, 1a-H) 1.86 (dd, J =
14.4, 5.7 Hz, 1 H, 1b-H) ppm. 13C NMR (CD3OD): δ = 169.1,
154.9, 128.2, 125.4, 112.2, 92.0, 88.7, 83.5, 79.4, 77.1, 62.3, 44.2,
35.4 ppm.
78.3, 77.2, 70.1, 69.1, 66.2, 60.4, 40.8 ppm. [α]2D0 = –42.6 (c = 0.5,
DMSO). MS: m/z = 521.2 [M+H]+, 544.3 [M+H+Na]+. HRMS
(FT-ICR): C24H28N2O9SNa: calcd. 543.1413; found 543.1432
[M+Na]+. C24H28N2O9S (520.1): calcd. C 55.38, H 5.42, N 5.38;
found C 55.58, H 5.89, N 4.76.
NMR Binding Studies
All NMR spectra were recorded at 293 K with a Bruker Avance
500 MHz spectrometer.
Sample Preparation: For the experiments with the free ligand, com-
pounds 9 and 10 were dissolved in a [D11]-Tris buffer at pH = 7.3,
containing 100 m NaCl and 5 m MgCl2; for compounds 5–8,
and SCH-54292, 10% CD3OD was added. COSY, TOCSY, and
HSQC experiments were performed by using the standard se-
quences. A mixing time of 800 ms was employed for NOESY ex-
periments. For the binding experiments, p21 h-Ras, expressed and
purified as described previously,[13] was dissolved in 405 µL of the
same [D11]-Tris buffer, containing an amount of GDP equimolar
to the protein, and transferred into a 5 mm NMR tube; 45 µL of
the ligand solution (molecules 5–8 and SCH-54292 dissolved in
CD3OD, molecules 9 and 10 dissolved in D2O) were added slowly,
and the mixture was incubated at 4 °C overnight.
Compound 24: To a stirred solution of 4-nitro-N-[2-(2-naphthyl-
oxy)ethyl]benzensulfonamide[5] (500 mg, 1.344 mmol) and 2,3,4,6-
O-tetraacetylglucopyranose (608 mg, 1.747 mmol) in dry THF
(13.4 mL), PPh3 (528.8 mg, 2.016 mmol) was added under argon.
The reaction mixture was cooled to –80 °C and, after 15 min,
DIAD (390.5 µL, 2.016 mmol) was added very slowly. The reaction
mixture was stirred at –80 °C for 1 h, and then it was allowed to
reach room temperature and stirred overnight. The solvent was
evaporated in vacuo, and the product was purified by flash column
chromatography (9.5:0.5, toluene/EtOAc) giving 24 (840.5 mg,
84%) as a yellow crystalline solid. 1H NMR (CDCl3): δ = 8.23,
8.05 (AAЈXXЈ, J = 8.6 Hz, 4 H, aromatic protons), 7.73–7.66 (m,
3 H,aromatic protons), 7.42 (bt, J = 6.9 Hz, 1 H, aromatic pro-
tons), 7.32 (bt, J = 7.0 Hz, 1 H, aromatic protons), 7.00 (s, 1 H,
aromatic proton), 6.93 (bd, J = 8.9 Hz, 1 H, aromatic proton), 5.45
(d, J = 8.6 Hz, 1 H), 5.31 (m, 2 H), 5.07 (t, J = 9.3 Hz, 1 H), 4.18–
4.04 (m, 4 H), 3.82–3.79 (m, 2 H), 3.59–3.54 (m, 1 H), 2.04 (s, 3
H, CH3), 2.02 (s, 3 H, CH3), 2.00 (s, 3 H, CH3), 1.91 (s, 3 H, CH3)
ppm. 13C NMR (CDCl3): δ = 170.3, 170.0, 169.6, 169.4, 155.8,
150.3, 145.3, 134.5, 129.7, 129.3, 129.0, 127.8, 126.9, 126.8, 124.4,
124.1, 118.5, 107.0, 85.6, 74.5, 73.7, 68.5, 68.0, 66.2, 61.9, 43.8,
21.0 ppm. [α]2D0 = –12.8 (c = 0.5, DMSO). HRMS (FT-ICR):
C32H34N2O14SNa: calcd. 725.1628; found 725.1635 [M +Na]+.
C32H34N2O14S (702.4): calcd. C 54.70, H 4.88, N 3.99; found: C
54.65, H 4.92, N 3.95.
Binding Experiments: trNOESY experiments were carried out with-
out saturation of the residual HDO signal. Optimized mixing times
of 200 and 250 ms and molar ratios between 15:1 and 30:1 of com-
pound/protein were employed. In relaxation-edited experiments, T1
and T1ρ values were calculated for the same ligand protons in the
free state and in the presence of p21 h-Ras-GDP complex. Selective
1H T1 relaxation times were measured using the standard inversion
recovery method with selective pulses at the resonance of interest.
T1ρ experiments used a spin-lock period of ca. 4000 Hz. In both
cases, 14 different delays were used to monitor the decay of the
magnetization. The experiments were repeated for the free mole-
cules and in the presence of Ras-GDP. The relaxation delays for
selective T1 were chosen to cover values between 0.1 and 3 s, while
those for T1ρ covered values between 0.05 and 2 s. All the spectra
were processed using the program Mestre-C. T1 and T1ρ values
were extracted from fitting the integral of a given proton signal as
a function of the relaxation delay, according to a single exponential
decay: y = –a·exp(–x/b), where b = T1 or T1ρ. STD experiments
were performed without saturation of the residual HDO signal for
molar ratios between 15:1 and 50:1 of compound/protein. A train
of Gaussian-shaped pulses of 50 ms each was employed, with a
total saturation time of the protein envelope of 2 s. MaxB1 field
strength: 50 Hz. An off-resonance frequency of δ = 40 ppm and
on-resonance frequencies between δ = 0.8 and –1.5 ppm (protein
aliphatic signals region) were applied. In all cases, line broadening
Compound 25: To a stirred solution of 24 (68 mg, 0.096 mmol) in
dry CH3OH (0.5 mL), K2CO3 (27 mg, 0.193 mmol) was added un-
der argon at room temperature. After a few minutes, a yellow pre-
cipitate appeared and the suspension was stirred at room tempera-
ture for 24 h. The reaction mixture was filtered, and the solid resi-
due was dissolved in boiling methanol and then filtered while hot
to eliminate salts. The filtrate was concentrated in vacuo and the
product was purified by flash column chromatography (8.5:1.5, tol-
uene/CH3OH) to provide 25[5] (49 mg, 95%) as a yellow amorphous
solid. The resulting product was exclusively the β-anomer.
3718
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720

Sugar-Derived Ras Inhibitors: NMR Spectroscopy and Biological Evaluation
of ligand protons was monitored before and after every binding
FULL PAPER
Acknowledgments
experiment in order to check our compound’s stability.
Work supported in part by the European Community’s Human
Potential Programme under contract HPRN-CT-2002-00173 (Gly-
cidic Scaffolds), and in part by the FIRB (design, preparation and
biological evaluation of new organic molecules as potential innov-
ative drugs). We thank Dr. Chiara Marinzi for helpful discussion
and manuscript reading.
Biology
Expression and Isolation of Proteins: The C-Cdc25Mm (GEF from
mouse; portion of CDC25Mm that contains the catalytic domain of
the protein)[13,14] was expressed in Escherichia coli using the pGEX-
2T expression vector, and was affinity-purified by using Glutathi-
one Sepharose 4B resin. p21 h-Ras was expressed in E. coli using
the pQE-30 expression vector, and was affinity-purified as a 6×His-
tagged protein by using Ni-NTA resin.[13]
[1] A. Wittinghofer, H. Waldmann, Angew. Chem. Int. Ed. 2000,
39, 4192–4214 and references cited therein.
Measurement of C-Cdc25Mm-Stimulated Guanine Nucleotide Ex-
change on p21 h-Ras: To investigate the ability of putative Ras in-
hibitors to inhibit or to reduce the C-Cdc25Mm-stimulated nucleo-
tide exchange on purified human Ras proteins, we used a technique
described by Lenzen et al.[10] with some modifications. This ap-
proach utilizes guanine nucleotides carrying an N-methylanthra-
niloyl fluorophore (MANT-GDP or MANT-GTP). p21 h-Ras
(100 n) and MANT-GTP (0.5 µ) were incubated in buffer B
(40 m Hepes, pH = 7.5, 2 m DTT, 100 µ MgCl2) in the absence
and in the presence of the inhibitors. The exchange reaction was
started by the addition of C-Cdc25Mm (25 n), and then monitored
at an excitation wavelength of 370 nm and an emission wavelength
of 450 nm with a Perkin–Elmer Luminescence Spectrometer. Mea-
surements were taken every second. The slope of each curve was
calculated.
[2] a) M. D. Sklar, Science 1988, 239, 645–647; b) A. C. Miller, D.
Samid, Int. J. Cancer 1995, 60, 249–254; c) E. J. Bernhard, E. J.
Stanbridge, S. Gupta, Cancer Res. 2000, 60, 6597–6600.
[3] F. Peri, C. Airoldi, S. Colombo, E. Martegani, A. S.
van Neuren, M. Stein, C. Marinzi, F. Nicotra, ChemBiochem
2005, 6, 1839–1848.
[4] F. Peri, R. Bassetti, E. Caneva, L. De Gioia, B. La Ferla, M.
Presta, E. Tanghetti, F. Nicotra, J. Chem. Soc., Perkin Trans. 1
2002, 638–644.
[5] A. G. Taveras, S. W. Remiszewsky, R. J. Doll, D. Cesarz, E. C.
Huang, P. Kirschmeier, B. N. Pramanik, M. E. Snow, Y.-S.
Wang, J. D. del Rosario, B. Vibulbhan, B. B. Bauer, J. E.
Brown, D. Carr, J. Catino, C. A. Evans, V. Girijavallabhan, L.
Heimark, L. James, S. Liberles, C. Nash, L. Perkins, M. M.
Senior, A. Tsarbopoulos, A. K. Ganguly, R. Aust, E. Brown,
D. Delisle, S. Fuhrman, T. Hendrickson, C. Kissinger, R. Love,
W. Sisson, E. Villafranca, S. E. Webber, Bioorg. Med. Chem.
1997, 5, 125–133.
[6] A. K. Ganguly, Y.-S. Wang, B. N. Pramanik, R. J. Doll, M. E.
Snow, A. G. Taveras, S. Remiszewski, D. Cesarz, J. del Rosario,
B. Vibulbhan, J. E. Brown, P. Kirschmeier, E. C. Huang, L.
Heimark, A. Tsarbopoulos, V. M. Girijavallabhan, Biochemis-
try 1998, 37, 15631–15637.
Measurement of Dissociation Rate: We used the method described
by Lenzen et al.[10] with some modifications to investigate the abil-
ity of Ras inhibitors to influence the C-Cdc25Mm-stimulated disso-
ciation rate of p21 h-Ras·MANT-GDP complexes. The complex
p21 h-Ras·MANT-GDP (200 n), obtained as described by Lenzen
et al.[10], and an excess of GDP (500 µ) were incubated in buffer
B (40 m Hepes, pH = 7.5, 2 m DTT, 100 µ MgCl2) in the ab-
sence and in the presence of the inhibitors. The dissociation reac-
tion was started by addition of C-Cdc25Mm (100 n), and then
monitored at an excitation wavelength of 370 nm and an emission
wavelength of 450 nm with a Perkin–Elmer Luminescence Spec-
trometer. Measurements were taken every second. The slope of
each curve was calculated assuming a single exponential decay.
[7] J. J. Turner, N. Wilschut, H. S. Overkleeft, W. Klaffke, G. A.
van der Marel, J. H. van Boom, Tetrahedron Lett. 1999, 40,
7039–7042.
[8] a) For a detailed account of the application of NMR to the
study of carbohydrates and their interactions see: NMR Spec-
troscopy of glycoconjugates (Eds.: J. Jiménez-Barbero, T. Pe-
ters), Wiley-VCH, Weinheim, 2002; for applications of
trNOESY experiments to the analysis of the bioactive confor-
mation of carbohydrates, see, for instance: b) V. L. Bevilacqua,
D. S. Thomson, J. H. Prestegard, Biochemistry 1990, 29, 5529–
5537; c) V. L. Bevilacqua, Y. Kim, J. H. Prestegard, Biochemis-
try 1992, 31, 9339–9349; d) H. Kogelberg, D. Solis, J. Jiménez-
Barbero, J. Curr. Opin. Struct. Biol. 2003, 13, 646–653; for ap-
plications of trNOESY experiments to the analysis of the bi-
oactive conformation of glycomimetics, see, for instance: e) A.
Bernardi, D. Arosio, L. Manzoni, D. Monti, H. Posteri, D.
Potenza, S. Mari, J. Jimenez-Barbero, Org. Biomol. Chem.
2003, 1, 785–792; f) L. M. Mikkelsen, M. J. Hernáiz, M.
Martín-Pastor, T. Skrydstrup, J. Jiménez-Barbero, J. Am.
Chem. Soc. 2002, 124, 14940–14951.
Cell Cultures and Growth Conditions: Normal NIH3T3 and k-Ras
(Arg12) transformed[11] mouse fibroblasts (from Dr. P. Bossu,
Dompé, L’Aquila, Italy) were grown on 60 mm plastic dishes in
Dulbecco’s modified Eagle’s medium supplemented with 10% new-
born calf serum (Gibco), penicillin (100 units/mL), and streptomy-
cin (100 µg/mL). Cells were detached by treating them with 0.05%
trypsin and EDTA (0.15 m), and growth was monitored by count-
ing the cell number/mL with a Coulter Counter ZM.
Assay of MAPK Activation: Cells were scraped, and ice-cold Lysis
buffer (25 m HEPES, pH = 7.5, 150 m NaCl, 1% NP-40, 0.25%
Na deoxycholate, 10% glycerol, 25 m NaF, 10 m MgCl2, 1 m
EDTA, 1 m Na vanadate, one tablet of Protease Inhibitor Mix-
ture from Roche Applied Science in 50 mL of extraction medium)
was added. The lysates were transferred to a microcentrifuge tube
on ice and centrifuged. Proteins were separated by SDS-PAGE,
transferred to nitrocellulose, and immunodecorated with anti-p42/
44 MAPK antibody and anti-phospho-p42/44 MAPK antibody
(Cell Signalling Technology, Beverly, MA, USA). Bound antibodies
were revealed with the ECL Western blotting analysis system
(Amersham Pharmacia Biotech).
[9] a) M. Mayer, B. Meyer, Angew. Chem. Int. Ed. 1999, 38, 1784–
1788; b) J. Klein, R. Meinecke, M. Meyer, B. Meyer, J. Am.
Chem. Soc. 1999, 121, 5336 –5337; c) M. Vogtherr, T. Peters, J.
Am. Chem. Soc. 2000, 122, 6093 –6099; d) A. Bernardi, D.
Arosio, D. Potenza, I. Sanchez-Medina, S. Mari, F. J. Cañada,
J. Jimenez-Barbero, Chem. Eur. J. 2004, 10, 4395–4405. For
new applications of STD experiments, employing living cells,
see: e) S. Mari, D. Serrano-Gómez, F. J. Cañada, A. L. Corbi,
J. Jiménez-Barbero, Angew. Chem. Int. Ed. 2005, 44, 296; f) B.
Claasen, M. Axmann, R. Meinecke, B. Meyer, J. Am. Chem.
Soc. 2005, 127, 916–919.
[10] C. Lenzen, R. H. Cool, A. Wittinghofer, Methods Enzymol.
1995, 255, 95–109.
Eur. J. Org. Chem. 2006, 3707–3720
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3719

F. Peri, et al.
FULL PAPER
[11] M. V. Milburn, L. Tong, A. M. de Vos, A. Brunger, Z. Yamai-
zumi, S. Nishimura, S. H. Kim, Science 1990, 247, 939–945.
[12] A. S. Seriani, R. Barker, J. Org. Chem. 1984, 49, 3292–3300.
[13] P. Coccetti, L. Mauri, L. Alberghina, E. Martegani, A. Par-
meggiani, Biochem. Biophys. Res. Commun. 1995, 206, 253–
259.
[14] E. Martegani, M. Vanoni, R. Zippel, P. Coccetti, R. Brambilla,
C. Ferrari, F. Sturani, L. Alberghina, EMBO J. 1992, 11, 2151–
2157.
Received: February 15, 2006
Published Online: June 21, 2006
3720
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3707–3720