Dansyl C-glucoside as a novel agent against endotoxic shock

Article abstract of DOI:10.1002/cmdc.201000282

Come dansyl with me! A small library of naphthyl gluco derivatives was synthesised from methyl α-D-glucopyranoside without protection/ deprotection steps. One library member, dansyl C-glucoside 5, showed an extraordinary anti-inflammatory activity, protecting 100□% of mice from sepsis at a dose of 25 μg□kg^-1.

Full text of DOI:10.1002/cmdc.201000282

DOI: 10.1002/cmdc.201000282
Dansyl C-Glucoside as a Novel Agent Against Endotoxic Shock
Barbara La Ferla,[a] Valerio Spinosa,[a] Giuseppe D’Orazio,[a] Marco Palazzo,[b] Andrea Balsari,[c]
Anastasia A. Foppoli,[d] Cristiano Rumio,[b] and Francesco Nicotra*[a]
Dedicated to Prof. Saverio Florio on the occasion of his 70th birthday.
The biological role of carbohydrates in a variety of recognition
processes of pharmacological relevance has stimulated interest
in glycomimetics as potential drugs.[1] Relenza, Tamiflu, and
Glyset are some successful examples of approved drugs be-
longing to this class of compounds. In the field of C-glycosides,
which are glycomimetics where the anomeric oxygen is re-
placed with a carbon atom to give improved chemical and
metabolic stability at the anomeric centre, no relevant bioac-
tivity has been reported until now. We have found that the
dansyl C-glucoside 5 is a novel and potent agent against endo-
toxic shock, fully protecting mice from sepsis.
glyco-drugs arises from the need for protection/deprotection
steps and the control of stereochemistry. An orthogonal func-
tional group suitable for chemoselective derivatisation of a
polyhydroxylated compound like a sugar is the primary amino
group.
We identified 2-(a-d-glucopyranosyl)ethanamine 3 (Scheme
1) as a key C-glycosyl intermediate, and generated this com-
pound by stereoselective C-allylation[7] of methyl a-d-glucopyr-
anoside, followed by ozonolysis and reductive amination of
the obtained aldehyde (present as hemiacetal 1).[8]
Some of us found that, in a murine model of endotoxic
shock induced by LPS, oral administration of glucose
(2.5 gkgꢀ1) protects all mice tested from sepsis.[2] This protec-
tion is mediated by the activation of the sodium-dependent
glucose transporter-1 (SGLT-1). A major drawback of this treat-
ment is that a high concentration of glucose, which impacts
metabolism, must be used. Therefore, our goal was to identify
glucomimetics that, at pharmacological concentrations, were
able to achieve protection against LPS-induced damage while
avoiding the side effects associated with the administration of
a high concentration of glucose.
Very little is known about the structure of SGLT-1 and its
mechanism of action,[3–6] although some aryl b-glucosides have
been patented as SGLT-1 inhibitors. Therefore, the only infor-
mation available for the rational design of SGLT-1 ligands is
based on the structure of d-glucose (or d-galactose, which is
also transported) and an aromatic aglycon. In order to gener-
ate compounds that would mimic glucose but could not be
metabolised, a C-glucoside is the key intermediate of choice.
We also looked for a procedure that allows the stereoselective
generation of this C-glucosidic key intermediate, and then its
derivatisation to generate a library, avoiding protection and
deprotection steps. A strong limitation in the development of
Scheme 1. Reagents and conditions: a) BSTFA, CH3CN, 808C, 1h; then,
TMSOTf, Me3Si(allyl), RT, 12h; then O3, MeOH/H2O (1:1), ꢀ788C, 75 min; then
Me2S, 5 min;[8] b) AcONH4, NaCNBH3, MeOH, RT, 1h, (3) 55%, (4) 57% (two
steps); c) MeOH, THF, K2CO3, dansyl-Cl, RT, 2h, (5) 50%, (6) 58%; d) MeOH,
THF, K2CO3, naphthylsulphonyl-Cl, RT, 2h, 52%; e) MeOH, THF, K2CO3, naph-
thylcarboxy-Cl, RT, 2h, 46%.
The amino group of compound 3 was chemoselectively
functionalised in different ways in the presence of the free
sugar hydroxy groups. The choice of pharmacophores to be
linked to the amine in 3 was essentially based on the limited
data reported in the literature on molecules able to interact
with SGLT-1, such as phloridzin, a natural b-glucoside with an
aromatic aglycon.[9] We started to derivatise compound 3 by
reaction with fluorescent dansyl chloride, and luckily dansyl
C-glucoside 5 displayed a very interesting anti-inflammatory
activity in vitro (see below, Figures 1 and 2).
[a] Dr. B. La Ferla, V. Spinosa, G. D’Orazio, Prof. F. Nicotra
Department of Biotechnology and Bioscience & CINMPIS, Universitꢀ degli
studi di Milano Bicocca, Piazza della Scienza 2, 20126 Milano (Italy)
Fax: (+39)026-448-3565
E-mail: francesco.nicotra@unimib.it
[b] Dr. M. Palazzo, Prof. C. Rumio
Department of Human Morphology, Universitꢀ degli studi di Milano, Via
Mangiagalli 31, 20133 Milano (Italy)
[c] Prof. A. Balsari
Institute of Pathology, Universitꢀ degli studi di Milano, Via Mangiagalli 31,
20133 Milano (Italy)
[d] Dr. A. A. Foppoli
Encouraged by this result, and to identify some structural re-
quirements for biological activity, we made a number of signifi-
cant changes in the structure of compound 5. We generated
naphthyl sulphonamide 7, which lacks a dimethylamino group,
Department of Pharmaceutical Sciences “P. Pratesi”, Universitꢀ degli studi di
Milano, Via Mangiagalli 25, 20133 Milano (Italy)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201000282.
ChemMedChem 2010, 5, 1677 – 1680
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1677

MED
pound 13 (Scheme 2) was performed, once more
avoiding the protection/deprotection steps. Methyl
a-d-glucopyranoside was reduced with triethylsilane,
affording 1-deoxy-d-glucose 9. Then selective tosyla-
tion of the primary hydroxy group, treatment of the
tosylate 10 with NaN3, reduction of the obtained
azide 11 with H2/Pd(OH)2, and finally treatment of
the amine 12 with dansyl chloride, afforded com-
pound 13.
The synthesis of compound 17 required the inser-
tion of a longer spacer between the sugar and the
dansyl moiety. We started from a-allyl C-glucoside
14,[7] and exploited the cross-metathesis (CM) reac-
tion for elongation. To perform the CM on our unpro-
tected substrate, we exploited a recent protocol that
describes the use of conventional hydrophobic ruthe-
Figure 1. IL-8 production in response to LPS (1 mgmLꢀ1) and compounds 4, 5, 6, 7, 8, 9,
12, 17 (50 mgLꢀ1). 13+LPS was below the level of detection.
*
and naphthyl amide 8, which has an amide group in place of
the sulphonamide group (Scheme 1).
nium catalysts in homogeneous aqueous solvent mixtures.[11,12]
CM between a-allyl C-glucoside 14 and cis-2-butene-1,4-diph-
thalimide,[13] using commercial Hoyveda–Grubbs 2nd genera-
tion catalyst (MeOH/CHCl3 (1:1) at 408C for 36 h), afforded a
reasonable quantity of compound 15 (57%, 80% conversion,
8:2 E/Z). Catalytic hydrogenation followed by hydrazinolysis
generated amine 16, which was converted into the dansyl de-
rivative 17.
Taking into account that galactose is recognised and trans-
ported by SGLT-1,[10] we synthesised a galacto-derivative 6, ex-
ploiting the same procedure described for compound 5. We
also elongated the spacer between the dansyl moiety and the
sugar in compound 17 (Scheme 3), and changed the position
of the dansyl group in compound 13. The synthesis of com-
Figure 2. IL-8 production in response to LPS (1 mgmLꢀ1) and compounds 5, 6, 7, 8, 17 (50 mgLꢀ1 to 5 mgLꢀ1).
1678
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1677 – 1680

17 maintained a low efficacy, while 7 and 8 completely lost
their suppressive activity (Figure 2).
From the ELISA data, it can be argued that the presence of
the dimethylamino substituent on the naphthyl group is cru-
cial for anti-inflammatory activity. Notably, the length of the
spacer between the sugar residue and the naphthyl pharmaco-
phore is also relevant.
To determine the effective involvement of the glucose trans-
porter-1 (SGLT-1) in the observed reduction of HT29-derived
IL-8, we repeated the experiments with the most active com-
pound 5, using an SGLT-1 knockdown cell line. Using small in-
terfering RNA (siRNA), SGLT-1 expression was silenced in the
HT29 cell line. These cells were then treated with LPS, and the
level of IL-8 was measured by ELISA. As shown in Figure 3a
Scheme 2. Reagents and conditions: a) BSTFA, CH3CN, 808C, 1h; then,
TMSOTf, Et3SiH, RT, 12h;[7] b) TsCl, Py, 08C!RT, 5h, 80% (two steps); c) NaN3,
TBAI, DMF, 808C, 24h, 90%; d) H2, Pd(OH)2, MeOH, RT, 12h, quant.;
e) Dansyl-Cl, K2CO3, MeOH, THF, RT, 2h, 42%.
Scheme 3. Reagents and conditions: a) HG, cis-2-butene-1,4-diphthalimide,
CHCl3/MeOH (1:1), 408C, 36 h, 55%; b) H2, Pd(OH)2, MeOH, RT, 3 h; c) N2H4,
MeOH, 508C, 4 h, 78% (two steps); d) Dansyl-Cl, K2CO3, MeOH, THF, RT, 2h,
77%.
IL-8 production by the human cell line, HT29, was used as a
functional, preliminary screen of the anti-inflammatory activity
of members of the library. The HT29 cells were incubated for
18 h in complete culture media alone or in media containing
50 mgLꢀ1 of our compounds. Cells were then stimulated for
6 h with 1 mgmLꢀ1 of LPS from Salmonella enterica, serovar
abortus equi. Variations in IL-8 concentration in the culture
media were evaluated by ELISA (Figure 1).
Figure 3. a) IL-8 production in response to LPS (1 mgmLꢀ1) and compound 5,
50 mgLꢀ1, with or without siRNA (siR). b) Percent survival of mice treated
^
with LPS/GalNH2 and with (— —), or without (—x—) compound 5. The ani-
mals were monitored for four days after treatment. Experimental protocols
were approved by the Ethics Committee for Animal Experimentation of the
Instituto Nazionale Tumori (Milan, Italy) and carried out according to guide-
lines of the United Kingdom Co-ordinating Committee on Cancer Research
for animal welfare in experimental neoplasia (1998).
From this preliminary evaluation, it was clear that com-
pounds 4, 9, 12, simple nonmetabolisable glucose/galactose
derivatives lacking the naphthyl moiety, did not affect LPS-in-
duced IL-8 production, while for compounds 5, 6, 7, 8, 17,
each with a naphthyl group linked to the C-glycosidic append-
age, a significant reduction in IL-8 was observed. Compound
13 was actually toxic to the HT29 cells. The dose-dependent
activity of compounds 5, 6, 7, 8 and 17 was also investigated,
(last column), the activity of compound 5 is lost when SGLT-1
expression is knocked down, proving that the biological activi-
ty of compound 5 is mediated by its interaction with SGLT-1.
Finally, we investigated the biological activity of compound
5 in vivo. These studies were carried out with a mouse model
of lethal septic shock induced by intraperitoneal (i.p.) adminis-
tration of LPS and galactosamine (GalNH2).[14] One group of
mice was treated only with LPS (250 mgkgꢀ1) and GalNH2
decreasing the concentration from 50 mgLꢀ1 down to 5 mgLꢀ1
.
Compound 5 was able to maintain IL-8 suppression at all con-
centrations tested, even at 5 mgLꢀ1. The activity of compounds
6, 7, 8 and 17 decreased in a dose-dependent manner, but at
the lowest concentration tested (5 mgLꢀ1), compounds 6 and
ChemMedChem 2010, 5, 1677 – 1680
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
1679

MED
(1 gkgꢀ1); another group was injected with LPS (250 mgkgꢀ1),
GalNH2 (1 gkgꢀ1) and compound 5 (25 mgkgꢀ1), administrated
by oral gavage; a third group of mice was given only sterile
water; a final group was treated only with compound 5
(25 mgkgꢀ1). For each experimental group, 10 mice were used.
As expected, in the mice treated with only LPS/GalNH2, 100%
of the animals developed and died from septic shock within
24 h. In contrast, 100% of the mice survived in the group treat-
ed with LPS/GalNH2 and compound 5 (Figure 3b), and in the
two control groups (see Supporting Information). This finding
confirms that compound 5 can effectively inhibit bacteria-
induced inflammation and has the potential to be a life-saving
treatment for septic shock.
269, 15204–15209; d) A. K. Meinild, D. D. F. Loo, B. A. Hirayama, E. Gal-
lardo, E. M. Wright, Biochemistry 2001, 40, 11897–11904; e) M. Panayo-
tova-Heiermann, S. Eskandari, E. Turk, G. A. Zampighi, E. M. Wright, J.
Biol. Chem. 1997, 272, 20324–20327; M. Panayotova-Heiermann, D. D.
Loo, C. T. Kong, J. E. Lever, E. M. Wright, J. Biol. Chem. 1996, 271,
10029–10034.
[4] a) M. M. Raja, N. K. Tyagi, R. K. H. Kinne, J. Biol. Chem. 2003, 278, 49154–
49163; b) T. Puntheeranurak, B. Wimmer, F. Castaneda, H. J. Gruber, P.
Hinterdorfer, R. K. H. Kinne, Biochemistry 2007, 46, 2797–2804; c) B.
Wimmer, M. Raja, P. Hinterdorfer, H. J. Gruber, R. K. H. Kinne, J. Biol.
Chem. 2008, 284, 983–991.
[5] a) M. M. Raja, H. Kipp, R. K. H. Kinne, Biochemistry 2004, 43, 10944–
10951; b) T. Puntheeranurak, L. Wildling, H. J. Gruber, R. K. H. Kinne, P.
Hinterdorfer, J. Cell Sci. 2006, 119, 2960–2967; c) A. Kumar, N. K. Tyagi,
P. Goyal, D. Pandey, W. Siess, R. K. H. Kinne, Biochemistry 2007, 46,
2758–2766; d) N. K. Tyagi, A. Kumar, P. Goyal, D. Pandey, W. Siess,
R. R. H. Kinne, Biochemistry 2007, 46, 13616–13628; e) T. Puntheeranur-
ak, M. Kasch, X. B. Xia, P. Hinterdorfer, R. K. H. Kinne, J. Biol. Chem. 2007,
282, 25222–25230; f) B. A. Hirayama, D. D. F. Loo, A. Diez-Sampedro,
D. W. Leung, A. K. Meinild, M. Lai-Bing, E. Turk, E. M. Wright, Biochemistry
2007, 46, 13391–13406.
In conclusion, a new class of C-glucosides displaying strong
anti-inflammatory activity have been synthesised using
straightforward chemoselective procedures that did not re-
quire protection, including the Sakurai reaction and cross-
metathesis. These compounds can interact with SGLT-1, a
newly identified component in the inflammatory cascade. In
particular, compound 5 displays a very potent activity, and its
administration, at a dose of 25 mgkgꢀ1, results in a 100% sur-
vival of mice that are treated with LPS/GalNH2. The identifica-
tion of highly active anti-inflammatory compounds that act as
agonists for SGLT-1 is intriguing, as studies published examin-
ing this transporter have sought to block its activity to prevent
glucose absorption.[15]
[6] X. Xia, C. T. Lin, G. Wang, W. Fang, Arch.Biochem. Biophys. 2004, 425,
58–64.
[7] J. A. Bennek, G. A. Gray, J. Org. Chem. 1987, 52, 892–897.
[8] F. Cardona, B. La Ferla, J. Carbohydr. Chem. 2008, 27, 203–213.
[9] D. F. Diedrich, Arch. Biochem. Biophys. 1966, 117, 248.
[10] E. M. Wright, Am. J. Physiol. Gastrointest. Liver Physiol. 1998, 275, 879–
882.
[11] a) J. B. Binder, J. J. Blank, R. T. Raines, Org. Lett. 2007, 9, 4885–4888;
b) B. H. Lipshutz, G. T. Anguinaldo, S. Ghorai, K. Voigtritter, Org. Lett.
2008, 10, 1325–1328.
[12] a) B. Mohr, D. M. Lynn, R. H. Grubbs, Organometallics 1996, 15, 4317–
4325; b) T. A. Kirkland, D. M. Lynn, R. H. Grubbs, J. Org. Chem. 1998, 63,
9904–9909; c) D. M. Lynn, R. H. Grubbs, J. Am. Chem. Soc. 2001, 123,
3187–3193; d) D. M. Lynn, B. Mohr, R. H. Grubbs, L. M. Henling, M. W.
Day, J. Am. Chem. Soc. 2000, 122, 6601–6609; e) S. H. Hong, R. H.
Grubbs, J. Am. Chem. Soc. 2006, 128, 3508–3509; f) A. Michrowska, L.
Gulajski, Z. Kaczmarska, K. Mennecke, A, Kirschning, K. Grela, Green
Chem. 2006, 8, 685–688; g) J. P. Jordan, R. H. Grubbs, Angew. Chem.
2007, 119, 5244–5247; Angew. Chem. Int. Ed. 2007, 46, 5152–5155; h) Ł.
Gułajski, A. Michrowska, J. Naroznik, Z. Kaczmarska, L. Rupnicki, K. Grela,
ChemSusChem 2008, 1, 103–109.
Acknowledgements
The CINMPIS Consortium supported this work through a fellow-
ship awarded to V. Spinosa.
Keywords: C-glycosides · chemoselectivity · endotoxic shock ·
glycomimetics · sepsis
[13] A. K. Chatterjee, T.-L. Choi, D. P. Sanders, R. H. Grubbs, J. Am. Chem. Soc.
2003, 125, 11360–11370.
[14] C. Galanos, M. A. Freudenberg, W. Reutter, Proc. Natl. Acad. Sci. USA
1979, 76, 5939–5943.
[15] a) M. F. Albertoni Borghese, M. P. Majowicz, Drugs Future 2009, 34, 297–
305 and references there in; b) F. Castaneda, A. Burse, W. Boland, R. K.
Kinne, Int. J. Med. Sci. 2007, 4, 131–139.
[1] For recent reviews see: a) B. Ernst, J. Magnani, Nat. Rev. Drug Discovery
2009, 8, 661–677; b) L. Cipolla, B. La Ferla, C. Airoldi, C. Zona, A. Orsato,
N. Shaikh, L. Russo, F. Nicotra, Future Medicinal Chemistry 2010, 2, 587–
599.
[2] M. Palazzo, S. Gariboldi, L. Zanobbio, S. Selleri, G. F. Dusio, V. Mauro, A.
Rossini, A. Balsari, C. Rumio, J. Immunol. 2008, 181, 3126–3136.
[3] a) B. Lo, M. Silverman, J. Biol. Chem. 1998, 273, 29341–29351; b) A.
Diez-Sampedro, E. M. Wright, B. A. Hirayama, J. Biol. Chem. 2001, 276,
49188–49194; c) E. Turk, M. G. Martin, E. M. Wright, J. Biol. Chem. 1994,
Received: July 9, 2010
Published online on August 19, 2010
1680
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 1677 – 1680