Iminosugar-ferrocene conjugates as potential anticancer agents

Article abstract of DOI:10.1039/c2ob25727k

We prepared a series of new iminosugar-ferrocene hybrids displaying potent inhibition of fucosidase (bovine kidney) and inactivation of MDA-MB-231 breast cancer cells proliferation at low micromolar concentrations. The synthetic route brought to light an unprecedented isomerisation of a 2-ethanalylpyrrolidine.

Full text of DOI:10.1039/c2ob25727k

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 5592
www.rsc.org/obc
PAPER
Iminosugar–ferrocene conjugates as potential anticancer agents†
Audrey Hottin,a Faustine Dubar,b Agata Steenackers,c Philippe Delannoy,c Christophe Biotc and
Jean-Bernard Behr*a
Received 12th March 2012, Accepted 8th June 2012
DOI: 10.1039/c2ob25727k
We prepared a series of new iminosugar–ferrocene hybrids displaying potent inhibition of fucosidase
(bovine kidney) and inactivation of MDA-MB-231 breast cancer cells proliferation at low micromolar
concentrations. The synthetic route brought to light an unprecedented isomerisation of a
2-ethanalylpyrrolidine.
Introduction
Iminosugars are natural or synthetic carbohydrate mimics dis-
playing potent affinities towards glycoenzymes.1,2 Among the
hundreds of known iminosugars, pyrrolidines like 1–3 (Fig. 1),
which share a specific fucose-like substituent distribution, are
highly potent nanomolar inhibitors of alpha-L-fucosidase
(AFU).3 This particular enzyme is associated with many dis-
Fig. 1 Structures of iminosugars 1–4.
orders including inflammation or viral infection.4 Moreover,
increasing fucosylation contributes to several abnormal charac-
teristics of tumor cells, regarding adhesion and growth, and AFU
levels are significantly increased in certain malignant tissues.4e,f
It is not firmly established whether the overexpression of AFU is
a cause or a consequence of cancer development. However, it
seemed rational to explore the possibility of using the fucose-
binding protein AFU as a target for the selective delivery of a
cytotoxic molecule towards cancer cells. Ferrocene (Fc) is a
remarkable pharmacophore, exhibiting physicochemical proper-
ties that accommodate biological uses.5 Whereas Fc has no bio-
logical effect by itself, some ferrocene conjugates were shown to
display antitumor, antimalarial or antifungal properties.6
Fc moiety with a fuco-configured polyhydroxy-pyrrolidine
playing the role of the drug-carrier targeting AFU.
Results and discussion
The anticipated strategy for the synthesis of organometallic 4
relied on a reductive amination between a ferrocenylamine and a
protected polyhydroxy-pyrrolidine featuring an ethanalyl side
chain. We wished to prepare first a series of target compounds,
varying the length of the flexible linker connecting Fc to the
iminosugar in order to establish structure–activity relationships.
To this aim, we envisioned introducing ferrocenylamines 5a–c of
general structure Fc(CH2)nNH2 [5a: n = 1; 5b: n = 2; 5c: n = 3]
into our synthetic plan. Satisfactorily, these compounds were
easily prepared following reported procedures.7 The required
aldehydo-pyrrolidine, in turn, could be obtained from unsatu-
rated precursors, as exemplified by former examples from the
literature.8 Thus, we focused our first efforts on the preparation
of C-2 allyl-pyrrolidine 8a (Scheme 1), featuring a (2S) configur-
ation at the pseudo-anomeric position. Indeed, this definite
configuration would provide adequate orientation of the C-2 sub-
stituent to induce strong binding to AFU, the (2R) isomers being
usually much less active as fucosidase inhibitors.3c
Combining our efforts towards innovative fucosidase inhibi-
tors and ferrocene conjugates, we designed Fc–iminosugar
hybrids 4a–c (Fig. 1) as a new class of potential anticancer
agents. Their structures result from the association of a cytotoxic
aUniversité de Reims Champagne-Ardenne, Institut de Chimie
Moléculaire de Reims, CNRS UMR 7312, UFR des Sciences Exactes et
Naturelles, 51687 Reims Cedex 2, France. E-mail: jb.behr@univ-reims.fr;
Fax: +33 326913166; Tel: +33 326913238
bUniversité Lille Nord de France, Université Lille 1, Unité de Catalyse
et Chimie du Solide, CNRS UMR 8181, 59652 Villeneuve d’Ascq Cedex,
France
cUniversité Lille Nord de France, Université Lille 1, Unité de
Glycobiologie Structurale et Fonctionnelle, CNRS UMR 8576, IFR 147,
59650 Villeneuve d’Ascq Cedex, France
The synthesis of 8a started with glycosylamine 6, prepared in
a mere 5 steps from D-ribose according to a reported route.3b
Allyl-magnesium chloride reacted efficiently with 6, affording
ring-opening products 7a and 7b in pure form after separation
†Electronic supplementary information (ESI) available: Spectroscopic
data for compounds 4, 8–10, 13–16. See DOI: 10.1039/c2ob25727k
5592 | Org. Biomol. Chem., 2012, 10, 5592–5597
This journal is © The Royal Society of Chemistry 2012

View Online
reduction was effected yielding alcohol 11b as the sole product,
which features the (2R) configuration. Obviously, the formation
of intermediate aldehyde 10b resulted from an unexpected iso-
merisation of the pre-formed (2S) isomer 10a during the reaction
or purification. These observations suggest a critical role of silica
in this intriguing transformation. We thus performed the period-
ate cleavage under more standard conditions, using EtOH–water
as the solvent and powdered sodium periodate as the reagent.
Gratifyingly, only aldehyde 10a was formed and isolated almost
pure after extraction with CH2Cl2. The structure of 10a was
correlated with known alcohol 11a after reduction with sodium
borohydride, which ascertained the (2S) configuration.12
However, aldehyde 10a was also prone to isomerisation in a
CDCl3 solution (t1/2 = 24 h at room temperature), though much
more slowly than in the presence of silica. At the same time,
when (2R) configurated allyl-pyrrolidine 8b was subjected to the
same sequence of reactions (osmylation–periodate cleavage) the
expected aldehyde 10b was obtained quantitatively as the sole
product. This unusual transformation of 10a into 10b has no pre-
cedent in the literature and further experiments were conducted
to determine a possible mechanism for this transformation
(Scheme 2). Firstly, we prepared the all-trans aldehyde A by
treating the known diol precursor13 with sodium periodate under
the conditions stated above. As expected, aldehyde A epimerised
to B quantitatively after 3 days at room temperature in a CDCl3
solution. No isomerization occurred with B under the same con-
ditions. This transformation seems thus general and could be
applicable to other ethanalyl-pyrrolidines. Furthermore, when 10a
was subjected to epimerisation in the presence of deuterated
water, proton exchange occurred strictly at the exo-cyclic enoliz-
able position and not at C-2, affording deuterated 10b (Scheme 2).
As a consequence, the transformation of 10a into 10b implies
the breaking of the C–N bond, which could reasonably occur via
elimination. This step could be catalysed by traces of acid
Scheme 1 Synthesis of ethanalylpyrrolidines.
by silica gel chromatography (83% yield). However, the addition
occurred with modest facial stereoselectivity (60% de). Determi-
nation of the configuration at the newly created stereocenter in
7a and 7b was performed after subsequent transformation into
the corresponding pyrrolidines 8a and 8b respectively (a process
effected with MsCl and resulting in inversion at C-5) and proved
the major isomer to have the expected (2S) configuration. Hence,
the nucleophilic attack at the anomeric carbon of 6 is anti-selec-
tive, a result that is in agreement with previous observations on
the addition of organometallics to isopropylidene-protected
ribosyl-amines.3b,9 Unfortunately, we failed to increase the
stereoselectivity by using a Barbier-type reaction with indium
and allyl bromide as the allylating agent.10 Under these con-
ditions, the reaction did not afford the expected aminoalcohol 7
but mainly unwanted degradation products.
Next, the synthesis of the required aldehyde from allyl-pyrroli-
dine 8a was experienced, following a standard osmylation/oxi-
dation procedure.11 When subjected to the Upjohn conditions
(cat. OsO4, N-methyl morpholine N-oxide as co-oxidant), allyl-
pyrrolidine 8a afforded diol 9a (dr, 1 : 1) in only 32% yield. The
use of commercial alpha AD-mix as an osmiumVIII source led to
significantly better results, the reaction affording 9a (dr, 6 : 4) in
79% isolated yield. The oxidative cleavage of diol 9a was then
studied (Scheme 1). The use of silica-supported sodium period-
ate has proven successful for such a transformation, and has the
advantage of being carried out in CH2Cl2.11 However, when a
solution of 9a was stirred in the presence of a suspension of
silica supported NaIO4, the diol reacted within 1 h to give a
1
1 : 1 mixture of two aldehydes, as deduced from the H-NMR
spectrum of the crude reaction mixture. Surprisingly, after purifi-
cation on silica gel only one aldehyde remained, which was iso-
lated in 82% yield. To confirm the stereochemistry, a NaBH4
Scheme 2 Proposed mechanism accounting for epimerization of
2-ethanalylpyrrolidines.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 5592–5597 | 5593

View Online
Table 1 Antifucosidase activity of compounds 4a–c, 15, 16a
Entry
Compound
IC50 b (μM)
1
2
3
4
5
4a
4b
4c
15
16
1.6 0.1
1.6 0.2
1.2 0.1
43.8 4.2
4.5 0.5
a Inhibition of fucosidase (bovine kidney) at
2
mM substrate
concentration. b Determined according to Dixon plots by assaying five
concentrations of each inhibitor.
Scheme 3 Coupling and deprotection.
present either on silica or in CDCl3. A subsequent aza-Michael
conjugate addition could account for the re-formation of the
pyrrolidine ring. Interestingly, such additions were previously
shown to be 3,4-syn selective,14 which is in agreement with the
results obtained either with A or with 10a.
Fig. 2 Effect of ferrocene–iminosugar conjugates 4a–c on
MDA-MB-231 breast cancer cells proliferation.
Nevertheless, the assistance of the neighboring amino group
might act as a driving force, thus we turned to a t-butylcarbamate
protecting group at nitrogen, to prevent this undesired epimerisa-
tion. Hydrogenolysis of diol 9a followed by reaction with
di-tert-butyl dicarbonate afforded diol 12 in 39% yield for the
two steps (Scheme 3).
bulky Fc functionality did not impede binding of the pyrrolidine
ligand to fucosidase. The Fc containing 4a was even a better
inhibitor than its phenyl counterpart 16 (IC50 = 4.5 μM). As
anticipated, the (2R) configuration led to a less potent inhibitor:
compound 15 (IC50 = 43.8 μM) was at least 10-fold less active
than the other iminosugars towards fucosidase. The anti-proli-
ferative effect was next analyzed using the hormone-independent
breast cancer cell line MDA-MB-231, previously shown to be
sensitive to ferrocene derivatives.4g,15 Cell growth data are
reported in Fig. 2 and are expressed as percentage of untreated
control. To our delight, compounds 4a–c had a significant effect
on MDA-MB-231 breast cancer cells proliferation in a concen-
tration-dependent manner (Fig. 2). Here again, the best results
were obtained with Fc–iminosugar 4c, which displayed 77%
growth inhibition at 50 μM and led to almost complete inacti-
vation at 100 μM. Interestingly, 4c was up to 10-fold more
potent as a cytotoxic agent than the reference drug cisplatin
(50% growth inhibition at 226 μM) towards the studied cell
lines.16 In return, compounds 15 and 16 showed no antiprolifera-
tive activity up to the maximum tested concentration of 100 μM
(data not reported in Fig. 2).
Aldehyde 13 was the only product obtained by oxidative clea-
vage of 12, whatever the source of NaIO4 used, and proved
stable either on silica gel or in solution. The stereochemistry of
13 was unequivocally confirmed by chemical correlation after its
transformation (NaBH4, then 1 M HCl) to the known iminosugar
1.12 Reductive amination was performed next with aldehyde 13
and amines 5a–c respectively, by stirring both reagents in
CH2Cl2 in the presence of MgSO4 as a dehydrating agent. After
filtration and evaporation, MeOH was added and the resulting
imine was thoroughly reduced with sodium borohydride to
afford ferrocenyliminosugars 14a–c in 68–70% yield after purifi-
cation. Final deprotection was achieved by stirring overnight
with aqueous HCl, and target compounds 4a–c were obtained in
a pure form after neutralization and chromatography over silica
gel. A similar sequence of reactions afforded compound 15
(structure shown in Scheme 3), the C-2 epimer of 4a that features
the “wrong” (2R) configuration. Finally, we prepared also com-
pound 16, a non-Fc analogue of 4a, by using benzylamine in the
reductive amination with aldehyde 13. Iminosugar 16 could play
the role of a control molecule to evaluate the influence of Fc on
either antifucosidase or antiproliferative activity.
Conclusions
Preliminary enzymatic and anticancer evaluations were per-
formed on Fc–iminosugars 4a–c, 15, 16. All the compounds dis-
played micromolar inhibition towards fucosidase from bovine
kidney (Table 1), 4c being more active (IC50 = 1.2 μM) than 4a
(IC50 = 1.6 μM) or 4b (IC50 = 1.6 μM). The presence of the
The synthesis of iminosugar–ferrocene conjugates was achieved
by reductive amination of an N-Boc protected ethanalylpyrroli-
dine with ferrocenylamines. The carbamate protection proved
crucial to prevent epimerisation at C-2 of the pyrrolidine ring.
Such a spontaneous isomerisation was observed with the N-Bn
5594 | Org. Biomol. Chem., 2012, 10, 5592–5597
This journal is © The Royal Society of Chemistry 2012

View Online
3
protected analogues, and could occur via a elimination/aza
Michael sequence. The iminosugar–ferrocene hybrids displayed
potent inhibition towards fucosidase and showed a significant
effect on the growth of MDA-MB-231 breast cancer cells pro-
liferation. These first results demonstrate the potential of ferro-
cene–iminosugars as anticancer agents. Work is under progress
to prepare a new set of Fc–iminosugars and non-Fc analogues to
elucidate the exact mechanism of action of this novel family of
anticancer agents and to study the possibility of targeting selec-
tively cancer cells through recognition by fucosidase.
3H, JHH = 6.0 Hz, 6-H), 1.33 (s, 3H, iPr), 1.56 (s, 3H, iPr),
1.76–1.97 (m, 1H, 1′-H), 2.22–2.42 (m, 1H, 1′-H), 2.92 (quint,
3
3
1H, JHH = 6.0, 5-H), 3.08 (dd, JHH = 10.5, 3.2 Hz, 2-H), 3.55
2
2
(d, 1H, JHH = 14.0 Hz, CH2-Ph), 3.91 (d, 1H, JHH = 14.0 Hz,
CH2-Ph), 4.43 (d, 1H, JHH = 6.5 Hz, 3-H), 4.52 (t, 1H, JHH
3
3
=
6.5 Hz, 4-H), 4.64–5.10 (m, 2H, 3′-H), 5.53–5.74 (m, 1H, 2′-H),
7.16–7.45 (m, 5H, Ar); 13C NMR (62.5 MHz; CDCl3) δ 12.6 (6-
C), 25.5, 26.4 (iPr), 28.9 (1′-C), 50.2 (CH2-Ph), 58.6 (5-C), 63.4
(2-C), 81.6, 82.0 (3,4-C), 111.4 (iPr-C), 117.1 (3′-C), 126.6 and
128.1 (Ar-C), 135.2 (2′-C), 138.6 (Ar-C); HRMS-ESI+ (m/z):
[M + H]+ calcd for C18H26NO2 288.1964; found 288.1968.
8b (70%, yellow oil): Rf = 0.53 (Et2O–petroleum ether: 2 : 8);
[α]2D0 = −58.2 (c 1, CHCl3); IR (film) νmax 702, 997, 1124,
1208, 1239, 1369, 1378, 2933, 2985; 1H NMR (CDCl3,
Experimental section
General considerations
3
250 MHz) δ 1.15 (d, 3H, JHH = 6.0 Hz, 6-H), 1.31 (s, 3H, iPr),
All reactions were performed under argon. The reagents and sol-
vents were commercially available in high purity and used as
received. Silica gel F254 (0.2 mm) was used for TLC plates,
detection being carried out by spraying with an alcoholic solu-
tion of phosphomolybdic acid, p-anisaldehyde or an aqueous
solution of KMnO4 (2%)–Na2CO3 (4%), followed by heating.
Flash column chromatography was performed over silica gel M
9385 (40–63 μm) Kieselgel 60. NMR spectra were recorded on
1.53 (s, 3H, iPr), 2.17–2.50 (m, 4H, 2,5,1′-H), 3.75 (s, 2H, CH2-
3
3
Ph), 4.40 (t, 1H, JHH = 5.5 Hz, 4-H), 4.49 (t, 1H, JHH
=
5.5 Hz, 3-H), 5.00–5.07 (m, 1H, 3′-Hb), 5.10–5.17 (m, 1H,
3′-Ha), 5.75–6.02 (m, 1H, 2′-H), 7.15–7.38 (m, 5H, Ar); 13C
NMR (62.5 MHz; CDCl3) δ 13.4 (6-C), 26.1 (iPr), 26.6 (iPr),
32.4 (1′-C), 53.3 (CH2-Ph), 62.2 (5-C), 67.3 (2-C), 79.5 (3-C),
81.1 (4-C), 111.2 (iPr-C), 116.9 (3′-C), 127.2, 128.5 and
129.3 (Ar-C), 136.0 (2′-C), 138.6 (Ar-C); HRMS-ESI+ (m/z):
[M + H]+ calcd for C18H26NO2 288.1964; found 288.1956.
1
Bruker AC 250 (250 MHz for H, 62.5 MHz for 13C) or 600
1
(600 MHz for H, 150 MHz for 13C) spectrometers. Chemical
shifts are expressed in parts per million (ppm) and were cali-
brated to the residual solvent peak. Coupling constants are in Hz
and splitting pattern abbreviations are: br, broad; s, singlet; d,
doublet; t, triplet; q, quartet; m, multiplet. IR spectra were
recorded with an IRTM plus MIDAC spectrophotometer and are
expressed in cm−1. Optical rotations were determined at 20 °C
with a Perkin-Elmer Model 241 polarimeter in the specified sol-
vents. High resolution mass spectra (HRMS) were performed on
a Q-TOF Micro micromass positive ESI (CV = 30 V).
Synthesis of diol 9a. A suspension of alpha AD-mix (3.3 g) in
acetone–distilled water (50 : 50, 20 ml) was added to allylpyrro-
lidine 8a (450 mg, 1.568 mmol) at 0 °C. The resulting mixture
was left to react at 4 °C for 35 h. Na2SO3 (2 g) was then added
and the mixture was stirred for an additional 45 min. The sol-
ution was extracted with EtOAc (3 × 20 ml). The combined
organic layers were dried (MgSO4), filtered, evaporated and the
residue was purified by silica gel chromatography (EtOAc) to
yield diol 9a (395 mg, 79%, yellow oil) as a mixture of two dia-
stereomers (79%, orange oil): Rf = 0.24 (EtOAc); IR (film) νmax
734, 1057, 1377, 2933, 2982; HRMS-ESI+ (m/z): [M + H]+
calcd for C18H27NO4 322.2018; found 322.2034; major isomer:
1H NMR (CDCl3, 250 MHz) δ 1.20 (d, 1H, J = 6.8 Hz, 6-H),
1.32 (s, 3H, iPr), 1.36–1.54 (m, 1H, 1′-Ha), 1.56 (s, 3H, iPr),
1.64–1.72 (m, 1H, 1′-Hb), 2.96–3.12 (m, 1H, 3′-Ha), 3.20–3.59
(m, 3H, 2,5-H and 3′-Hb), 3.85 (dd, 2H, J = 78.7, 10.1 Hz, CH2-
Ph), 4.51–4.70 (m, 2H, 3,4-H), 7.23–7.40 (m, 5H, Ar);
13C NMR (62.5 MHz; CDCl3) δ 10.71 (6-C), 24.55 (iPr), 26.28
(iPr), 29.14 (1′-C), 51.86 (CH2-Ph), 57.95 (5-C), 63.20 (2-C),
66.07 (3′-C), 70.71 (2′-C), 81.86 (4-C), 84.49 (3-C), 112.26 (iPr-
C), 128.58, 128.61 and 129.34 (3 × Ar), 138.77 (Ar-C); minor
isomer: 1H NMR (CDCl3, 250 MHz) δ 1.21–1.28 (m, 1H,
1′-Ha), 1.38 (d, 1H, J = 6.9 Hz, 6-H), 1.32 (s, 3H, iPr),
1.36–1.54 (m, 1H, 1′-Hb), 1.62 (s, 3H, iPr), 3.23–3.29 (m, 2H,
2′-H), 3.43–3.52 (m, 3H, 2,3′-H), 3.85 (dd, 2H, J = 78.7,
10.1 Hz, CH2-Ph), 4.35–4.37 (m, 1H, 3-H), 4.59–4.63 (m, 1H,
4-H), 7.23–7.40 (m, 5H, Ar). 13C NMR (62.5 MHz; CDCl3)
δ 11.09 (6-C), 23.53 (iPr), 26.49 (iPr), 30.91 (1′-C), 53.22
(CH2-Ph), 58.30 (5-C), 65.31 (2-C), 66.70 (3′-C), 71.23 (2′-C),
83.55 (4-C), 86.25 (3-C), 111.63 (iPr-Cq), 129.07, 129.34,
129.90 (3 × Ar), 139.08 (Ar-Cq).
General procedure for the synthesis of allylpyrrolidines 8a, 8b
Allylmagnesium chloride (15 ml of a commercial 2 M solution
in THF, 30 mmol) was added to a stirred solution of glycosyl-
amine 6 (2.059 g, 7.88 mmol) in THF (20 ml) at 0 °C, and the
resulting mixture was left to react at rt for 7 h. Saturated NH4Cl
was then added and the solution was extracted with Et2O (3 ×
20 ml). The combined organic layers were dried (MgSO4),
evaporated and aminoalcohols 7a, 7b were separated and
purified by FC (Et2O–petroleum ether, 5 : 5) to yield 7a (Rf =
0.46, 1.545 g, 64%) and 7b (Rf = 0.26, 0.467 g, 19%) as yellow
oils. To a solution of pure amino alcohol (1.068 g, 3.50 mmol)
in pyridine (4.5 ml) and THF (4.5 ml) at 0 °C, MsCl (664 μl,
8.72 mmol, 2.45 equiv.) was added dropwise. The mixture was
stirred for 2 h at 0 °C and for an additional 2 h at room tempera-
ture. Then a saturated solution of NH4Cl and Et2O were succes-
sively added at 0 °C and the resulting organic phase was
separated. The aqueous layer was extracted with Et2O and the
combined organic layers were dried over MgSO4 and concen-
trated. The residue was purified by silica gel chromatography
(Et2O–petroleum ether: 2 : 8) to yield pure allylpyrrolidine.
8a (86%, yellow oil): Rf = 0.56 (Et2O–petroleum ether: 2 : 8);
[α]2D0 = 63.2 (c 1.1, CHCl3); IR (film) νmax 700, 1064, 1161,
1210, 1378, 2933, 2984; 1H NMR (CDCl3, 250 MHz) δ 1.12 (d,
Synthesis of aldehyde 13. A mixture of diol 9a (392 mg,
1.22 mmol), HCO2NH4 (538 mg, 8.54 mmol, 7 equiv.) and
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 5592–5597 | 5595

View Online
Pd/C 10% (392 mg) in MeOH (12 ml) was stirred at 60 °C for
1.5 h. Filtration over celite and purification by silica gel chromato-
graphy (DCM–MeOH: 9 : 1) yielded the N-debenzylated diol
(158 mg, 56%, colorless oil), which was added to a solution of
Et3N (243 μl, 2.392 mmol, 3.5 equiv.) and Boc2O (164 mg,
0.766 mmol, 1.1 equiv.) in CH2Cl2 (7 ml) at 0 °C. The mixture
was allowed to warm slowly to room temperature and was stirred
for 24 h. Distilled water (20 ml) was then added and the aqueous
layer was extracted with DCM (3 × 20 ml). The combined
organic layers were dried (MgSO4), filtered and evaporated under
reduced pressure. The crude residue was purified by silica gel
column chromatography (EtOAc) to afford 12 (158 mg, 70%, col-
ourless oil). To this diol 12 (75 mg, 0.226 mmol) in 4.5 ml of a
2 : 1 EtOH–water solution was added NaIO4 (141 mg,
0.661 mmol, 2.9 equiv.). The solution was stirred at rt for 1 h,
then water (10 ml) and DCM were added. After separation of the
layers, the aqueous phase was washed with DCM (3 × 10 ml).
The combined organic layers were dried (MgSO4), filtered and
evaporated to afford the crude aldehyde 13 (67 mg, 100%, yellow
oil) which was used in the next step without further purification.
4a (98%): [α]2D0 = −17.4 (c 1, MeOH); 1H NMR (D2O,
500 MHz) δ 1.25 (d, 3H, J = 6.8 Hz, 6-H), 1.93–2.13 (m, 2H,
1′-H), 3.11 (t, 2H, J = 7.7 Hz, 2′-H), 3.31 (td, 1H, J = 2 × 7.9,
6.8 Hz, 2-H), 3.50–3.55 (m, 1H, 5-H), 4.02–4.09 (m, 2H, 3-H,
4-H), 4.11 (s, 2H, NCH2Fc), 4.26–4.46 (m, 9H, Fc); 13C NMR
(125 MHz; CD3OD) δ 13.8 (6-C), 31.1 (1′-C), 46.8 (2′-C), 48.9
(NCH2Fc), 56.6 (5-C), 61.5 (2-C), 69.8, 69.8, 70.5, 70.6 (Fc),
74.4 (4-C), 79.0 (3-C), 82.3 (Cq-Fc); HRMS-ESI+ (m/z):
[M + H]+ calcd for C18H27FeN2O2 359.1422; found 359.1419.
4b (73%): [α]2D0 = −7.2 (c 0.5, MeOH); 1H NMR (D2O,
500 MHz) δ 1.33 (d, 1H, J = 6.8 Hz, 6-H), 2.12–2.22 (m, 2H,
1′-H), 2.80 (t, 2H, J = 7.2 Hz, NCH2CH2Fc), 3.16–3.28 (m, 4H,
2′-H and NCH2CH2Fc), 3.43 (q, 1H, J = 7.5 Hz, 2-H),
3.68–3.77 (m, 1H, 5-H), 4.12 (t, 1H, J = 3.2 Hz, 4-H), 4.17 (dd,
1H, J = 7.5, 3.2 Hz, 3-H), 4.21–4.33 (m, 9H, Fc); 13C NMR
(125 MHz; CD3OD) δ 12.7 (6-C), 28.0 (NCH2CH2Fc), 29.1
(1′-C), 46.3, 50.0 (2′-C, NCH2CH2Fc), 57.9 (5-C), 60.1 (2-C),
68.9, 69.3, 69.7 (Fc), 73.4 (4-C), 78.0 (3-C), 84.5 (Cq-Fc);
HRMS-ESI+ (m/z): [M
+
H]+ calcd for C19H29FeN2O2
373.1578; found 373.1578.
1
4c (97%): [α]2D0 = −12.9 (c 0.86, MeOH); H NMR (D2O,
600 MHz) δ 1.34 (d, 3H, J = 6.7 Hz, 6-H), 1.88–1.95 (m, 2H,
NCH2CH2CH2Fc), 2.18–2.21 (m, 2H, 1′-H), 2.47 (t, 2H, J =
General procedure for the synthesis of iminosugar–ferrocene
conjugates 14a–c, 4a–c, 15, 16
7.5
Hz,
NCH2CH2CH2Fc),
3.05–3.13
(m,
2H,
To a solution of aldehyde 13 (30 mg, 0.100 mmol) in DCM (2 ml),
MgSO4 (119 mg, 1.00 mmol, 10 equiv.) and the ferrocenylamine 5
(1.2 equiv.) were successively added. The solution was stirred at rt
for 5 h. After filtration and concentration, the resulting material was
dissolved in MeOH (2 ml) and NaBH4 (5 mg, 0.130 mmol,
1.3 equiv.) was added at 0 °C. This solution was stirred and left to
warm to rt overnight. Saturated NH4Cl and EtOAc were succes-
sively added at 0 °C and the resulting organic layer was separated.
The aqueous layer was extracted with EtOAc and the combined
organic layers were dried over MgSO4 and concentrated. The
residue was purified by silica gel chromatography (EtOAc–MeOH:
19 : 1 → 8 : 2) to yield pure ferrocenyliminosugar 14.
14a (68%, yellow oil): Rf = 0.25 (EtOAc–MeOH: 19 : 1);
[α]2D0 = +31.9 (c 0.84, CHCl3); IR (film) νmax 1027, 1107, 1144,
1171, 1244, 1373, 1689, 2934, 2981; HRMS-ESI+ (m/z):
[M + H]+ calcd for C26H38FeN2O4 499.2259; found 499.2250.
14b (70%, yellow oil): Rf = 0.43 (EtOAc–MeOH: 19 : 1);
[α]2D0 = +28.9 (c 1, CHCl3); IR (film) νmax 1026, 1107, 1143,
1171, 1245, 1373, 1690, 2932, 2979; HRMS-ESI+ (m/z):
[M + H]+ calcd for C27H40FeN2O4 513.2416; found 513.2410.
14c (70%, yellow oil): Rf = 0.23 (EtOAc–MeOH: 8 : 2);
[α]2D0 = +24.4 (c 1, CHCl3); IR (film) νmax 754, 1025, 1107,
1373, 1688, 2932; HRMS-ESI+ (m/z): [M + H]+ calcd for
C28H42FeN2O4 527.2572; found 527.2575.
NCH2CH2CH2Fc), 3.14–3.28 (m, 2H, 2′-H), 3.47 (q, 1H, J =
7.5 Hz, 2-H), 3.72–3.77 (m, 1H, 5-H), 4.12–4.14 (m, 1H, 4-H),
4.16–4.18 (m, 1H, 3-H), 4.18–4.26 (m, 9H, Fc); 13C NMR
(126 MHz; CD3OD) δ 12.6 (6-C), 27.7, 29.2, 29.3 (1′-C,
NCH2CH2CH2Fc), 46.1, 49.9 (2′-C, NCH2CH2CH2Fc), 58.1
(5-C), 59.8 (2-C), 68.5, 69.1, 69.6 (3 × Fc), 73.5 (4-C), 78.1
(3-C), 88.3 (Cq-Fc); HRMS-ESI+ (m/z): [M + H]+ calcd for
C20H31FeN2O2 387.1735; found 387.1752.
15 (100%): [α]D20 = 2.4 (c 0.33, H2O); 1H NMR (D2O,
600 MHz) δ 1.15 (d, 3H, J = 6.8 Hz, 6-H), 1.76 (td, 1H, J =
15.3, 7.8 Hz, 1′-Ha), 1.94 (dt, 1H, J = 14.0, 7.8 Hz, 1′-Hb), 2.96
(t, 2H, J = 7.8 Hz, 2′-H), 3.11 (q, 1H, J = 6.4 Hz, 2-H), 3.19
(quint, 1H, J = 6.4 Hz, 5-H), 3.94 (s, 2H, NCH2Fc), 4.13 (t, 1H,
J = 5.2 Hz, 3-H), 4.22 (t, 1H, J = 5.2 Hz, 4-H), 4.25–4.44 (m,
9H, Fc); 13C NMR (126 MHz, MeOD) δ 14.4 (6-C), 27.1 (1′-C),
46.8 (2′-C), 49.0 (3′-C), 56.7 (5-C), 59.6 (2-C), 69.8, 70.0, 70.7
(3 × Fc), 73.6 (3-C), 73.8 (4-C); ESI-HRMS: calcd for
C18H27FeN2O2 ([M + H]+) 359.1422; found 359.1430.
1
16 (100%): [α]2D0 = −31.6 (c 0.5, MeOH); H NMR (D2O,
600 MHz) δ 1.19 (d, 1H, J = 6.8 Hz, 6-H), 1.82–1.94 (m, 1H,
1′-Ha), 1.94–2.07 (m, 1H, 1′-Hb), 2.91–3.06 (m, 2H, 2′-H), 3.24
(dd, 1H, J = 6.1, 7.8 Hz, 2-H), 3.36–3.44 (m, 1H, 5-H),
3.95–4.02 (m, 2H, 3,4-H), 4.02–4.08 (m, 2H, 3′-H), 7.40–7.49
(m, 5H, Ar); 13C NMR (63 MHz, MeOD) δ 12.99 (6-C), 30.34
(1′-C), 45.49 (2′-C), 51.99 (3′-C), 55.66 (5-C), 59.09 (2-C),
73.51 (4-C), 77.53 (3-C), 129.23, 129.53, 129.73 (Ar), 134.58
(Ar-Cq); ESI-HRMS: calcd for C18H27FeN2O2 ([M + H]+)
251.1760; found 251.1768.
The carbamate protecting group at nitrogen induced very large
NMR signals (see copies of NMR spectra in the ESI†). Conse-
quently, compounds 14a–c were fully characterized in their
deprotected form after treatment of a solution of ferrocenyl-
iminosugar 14 (0.0684 mmol) in MeOH (1 ml) with 1 M HCl
(1 ml). The mixture was stirred at 40 °C overnight. After com-
pletion of the reaction the solution was neutralized with Amber-
lyst® A-26 (OH−) and evaporated. Purification by silica gel
Acknowledgements
column chromatography (DCM–MeOH: 8 : 2
MeOH–NH4OH: 6 : 4 : 1) yielded ferrocenyliminosugar 4 as a
yellow film.
→
CHCl3–
Financial support from the Ministry of Higher Education and
Research (MESR), CNRS, Conseil Regional Champagne
Ardenne, Conseil General de la Marne, and EU-programme
5596 | Org. Biomol. Chem., 2012, 10, 5592–5597
This journal is © The Royal Society of Chemistry 2012

View Online
E. A. Hillard and G. Jaouen, Med. Chem. Commun., 2010, 1, 149–151.
For reports on ferrocene–carbohydrate conjugates, see: (c) M. J. Adam
and L. D. Hall, Can. J. Chem., 1980, 58, 1188–1197; (d) J. M. Casas-
Solvas, A. Vargas-Berenguel, L. F. Capitan-Vallvey and F. Santoyo-
Gonzales, Org. Lett., 2004, 6, 3687–3690; (e) C. L. Ferreira, C. B. Ewart,
C. A. Barta, S. Little, V. Yardley, C. Martins, E. Polishchuk, P. J. Smith,
J. R. Moss, M. Merkel, M. J. Adam and C. Orvig, Inorg. Chem., 2006,
45, 8414–8422; (f) J. M. Casas-Solvas, E. Ortiz-Salmeron,
J. J. Gimenez-Martinez, L. Garcia-Fuentes, L. F. Capitan-Vallvey,
F. Santoyo-Gonzales and A. Vargas-Berenguel, Chem.–Eur. J., 2009, 15,
710–725.
6 (a) D. R. Van Staveren and N. Metzler-Nolte, Chem. Rev., 2004, 104,
5931–5985; (b) C. Biot, F. Nosten, L. Fraisse, D. Ter-Minassian,
J. Khalife and D. Dive, Parasite, 2011, 18, 207–214.
7 (a) For the synthesis of 5a: S. C. B. Gnoatto, A. Dassonville-Klimpt,
S. Da Nascimento, P. Galéra, K. Boumediene, G. Gosmann, P. Sonnet
and S. Moslemi, Eur. J. Med. Chem., 2008, 43, 1865–1877;
(b) For the synthesis of 5b: L. D. Wescott and D. L. Mattern, J. Org.
Chem., 2003, 68, 10058–10066; (c) For the synthesis of 5c: I. U. Khand,
T. Lanez and P. L. Pauson, J. Chem. Soc., Perkin Trans. 1, 1989, 2075–
2078.
FEDER to the PlAneT CPER project is gratefully
acknowledged. E. Le Monnier and A. Kotland are sincerely
acknowledged for their participation in some aspects of this
project. J.-B. B. warmly thanks Dr Nicolas Blanchard (ENSC
Mulhouse) for inspiring discussions on the mechanism of
epimerization.
Notes and references
1 (a) Iminosugars as Glycosidase Inhibitors: Nojirimycin and Beyond, ed.
A. E. Stütz, Wiley-VCH, Weinheim, Germany, 1999; (b) Iminosugars:
From Synthesis to Therapeutic Applications, ed. P. Compain and
O. R. Martin, Wiley-VCH, Weinheim, Germany, 2007.
2 (a) R. J. Nash, A. Kato, C.-Y. Yu and G. W. J. Fleet, Future Med. Chem.,
2011, 3, 1513–1521; (b) G. Horne, F. X. Wilson, J. Tinsley,
D. H. Williams and R. Storer, Drug Discovery Today, 2011, 16, 107–118;
(c) T. M. Wrodnigg, A. J. Steiner and B. Ueberbacher, Anti-Cancer
Agents Med. Chem., 2008, 8, 77–85.
3 (a) C. Chevrier, D. LeNouën, M. Neuburger, A. Defoin and C. Tarnus,
Tetrahedron Lett., 2004, 45, 5363–5366; (b) A. Kotland, F. Accadbled,
K. Robeyns and J.-B. Behr, J. Org. Chem., 2011, 72, 2224–2227;
(c) E. Moreno-Clavijo, A. T. Carmona, Y. Vera-Ayoso, A. Moreno-
Vargas, C. Bello, P. Vogel and I. Robina, Org. Biomol. Chem., 2009, 7,
1192–1202.
4 (a) I. Robina, A. J. Moreno-Vargas, A. T. Carmona and P. Vogel, Curr.
Drug Metab., 2004, 5, 329–361; (b) E. Shimanek, E. J. McGarvey,
J. A. Jablonowski and C.-H. Wong, Chem. Rev., 1998, 98, 833–862;
(c) M. Padro, J. A. Castillo, L. Gomez, J. Joglar, P. Clapes and C. de
Bolos, Glycoconjugate J., 2010, 27, 277–285; (d) E. Moreno-Clavijo,
A. T. Carmona, A. J. Moreno-Vargas, L. Molina and I. Robina, Curr.
Org. Synth., 2011, 8, 102–133; (e) J.-J. Wang and E.-H. Cao, Clin. Chim.
Acta, 2004, 347, 103–109; (f) M. Shah, S. Telang, G. Raval, P. Shah and
P. S. Patel, Cancer, 2008, 113, 336–346; (g) For a review article on the
role of fucosyl-oligosaccharides in human breast cancer, see
J. J. Listinsky, G. P. Siegal and C. M. Listinsky, Am. J. Transl. Res.,
2011, 3, 292–322.
8 (a) A. M. Palmer and V. Jäger, Eur. J. Org. Chem., 2001, 2547–2558;
(b) J.-B. Behr and G. Guillerm, Tetrahedron: Asymmetry, 2002, 13,
111–113.
9 A. Rajender, J.-P. Rao and B. V. Rao, Tetrahedron: Asymmetry, 2011, 22,
1306–1311.
10 J.-B. Behr, A. Hottin and A. Ndoye, Org. Lett., 2012, 14, 1536–1539.
11 I. Gautier-Lefebvre, J.-B. Behr, G. Guillerm and N. S. Ryder, Bioorg.
Med. Chem. Lett., 2000, 10, 1483–1486.
12 C. Chevrier, D. LeNouen, A. Defoin and C. Tarnus, Carbohydr. Res.,
2011, 346, 1202–1211.
13 (a) J.-B. Behr, A. Erard and G. Guillerm, Eur. J. Org. Chem., 2002,
1256–1262; (b) J.-B. Behr, A. Gainvors-Claisse and A. Belarbi, Nat.
Prod. Res., 2006, 20, 1308–1314.
14 E. Moreno-Clavijo, A. T. Carmona, A. J. Moreno-Vargas and I. Robina,
Tetrahedron Lett., 2007, 48, 159–162.
15 M. Görmen, P. Pigeon, S. Top, E. A. Hillard, M. Huché, C. G. Hartinger,
F. de Montigny, M. A. Plamont, A. Vessières and G. Jaouen, ChemMed-
Chem, 2010, 5, 2039–2050.
5 (a) G. Gasser, I. Ott and N. Metzler-Nolte, J. Med. Chem., 2011, 54,
3–25; (b) M. Görmen, P. Pigeon, S. Top, A. Vessières, M.-A. Plamont,
16 D. R. Ciocca, S. A. W. Fuqua, S. Lock-Lim, D. O. Toft, W. J. Welch and
W. L. McGuire, Cancer Res., 1992, 52, 3648–3654.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 5592–5597 | 5597