Glycosylated foldamers to probe the carbohydrate-carbohydrate interaction

Article abstract of DOI:10.1021/ja0614565

The synthesis of a triglycosylated helical foldamer based on a combination of cyclopentyl- and pyrrolidinyl-based amino acids is described. This structure is stable in water, maintaining as it does a series of carbohydrate units in proximity to one another, and represents the basis of a new approach to the study of carbohydrate-carbohydrate interactions. Copyright

Full text of DOI:10.1021/ja0614565

Published on Web 08/01/2006
Glycosylated Foldamers To Probe the Carbohydrate-Carbohydrate
Interaction
Graham L. Simpson,† Andrew H. Gordon,† David M. Lindsay,† Netnepa Promsawan,†
Matthew P. Crump,† Keith Mulholland,§ Barry R. Hayter,‡ and Timothy Gallagher*,†
School of Chemistry, UniVersity of Bristol, Bristol, BS8 1TS, U.K., AstraZeneca, Silk Road Business Park,
Macclesfield, SK10 2NA, U.K., and AstraZeneca, Mereside, Macclesfield, SK10 4TG, U.K.
Received March 2, 2006; E-mail: t.gallagher@bristol.ac.uk
On the basis of the seminal studies of Hakamori1a,b,2d and
Burger,1c,2c intercellular interactions involving complementary
carbohydrate motifs are now recognized to be multivalent in nature,
highly substrate specific, and associated with a number of key
biological processes.2 To date, study of this important recognition
process has exploited the multivalent component,3 with, for
example, synthetic micelles, monolayers, and glyconanoparticles
playing a key role.4 Nevertheless, the molecular detail of these weak,
sometimes Ca2+ dependent, interactions remains unclear, neces-
sitating the need for models to probe the nature of this phenomenon
in terms of individual carbohydrate (CHO) motifs.
Our proposal was to ligate carbohydrates onto a conformationally
defined helical scaffold to allow carbohydrate-carbohydrate in-
teractions to be studied within a controlled and nonmultiValent
environment. We propose to evaluate these interactions directly by
spectroscopic methods, which may also be associated within (and
reportable via) the scaffold core. In this way, complementary (and
sensitive) analytical methods could be employed to study this
phenomenon. A suitable core was reported by Gellman5 whose
mixed cyclopentyl/pyrrolidine “foldamers” display a 12-helical
secondary structure (i.e., 12-membered H-bonded ring CdO(i)-
HN(i + 3)5f) in aqueous solution. Positioning the CHO moieties
on a single face of the foldamer scaffold offers an efficient design
that maximizes the opportunity for intrascaffold CHO-based
interactions. To pursue this strategy we have targeted a hexapeptide
1, which is selectively glycosylated at i + 1, i +3, and i +4 on the
basis of a predicted 12-helical pattern (Figure 1(a) and (b)).
Peptide 1 is accessible via the requisite enantiomerically pure
glycosylated pyrrolidine trans-â-amino acid, with peptide assembly
exploiting cyclopentyl â-amino acids as spacers. In this paper we
describe the synthesis and characterization of the first glycosylated
foldamer which presents a carbohydrate surface which positions
and maintains model CHO units in close proximity to one another.6
Our synthetic strategy centers on the enantiomerically pure
pyrrolidine â-amino acid monomers (e.g., 7) glycosylated via a
sulfonamide-based5d linker (Scheme 1). Suitably protected trans-
3-aminopyrrolidine-4-carboxylate 3 was prepared from â-keto ester
27 utilizing a diastereoselective reductive amination/crystallization
procedure as reported by Gellman.5e Protecting group manipulations
and sulfonamide formation provided access to pyrrolidine 4 ready
for carbohydrate attachment through cross metathesis (CM).
Optimal CM conditions required crotyl-crotyl containing com-
ponents, and reactions involving 4 and four â-O-crotyl carbohy-
drates partners 5a, 5b, 5c, and 5d were employed. The latter were
based on galactose, glucose, glucosamine, and lactose, respectively,
and this led to the glycosylated monomers 6a-d in moderate yield.8
Figure 1. (a) Schematic â-peptide 12-helical scaffold; (b) helix periodicity
wheel and glycosylation sites (blue spheres) for hexapeptide 1.
Scheme 1. Synthesis of Glycosylated Pyrrolidine â-Amino Acida
a Conditions: (a) (R)-(+)-R-methyl benzylamine, AcOH, EtOH, 4 h; (b)
NaCNBH3, 75 °C, 16 h; (c) 4 M HCl in dioxane, EtOAc, 25% (3 steps);
(d) LiOH‚H2O, THF/MeOH/H2O, 0 °C; (e) BnBr, Cs2CO3, DMF, 86% (2
steps); (f) TFA, CH2Cl2; (g) CH3CH2dCHCH2SO2Cl, DIPEA, CH2Cl2, 0
°C, 79% (2 steps); (h) Grubbs 2nd generation catalyst, CH2Cl2, 40 °C, 24
h, 38-60%; (i) 4 M HCl in dioxane; (j) H2, 10% Pd/C, AcOH, room temp,
24 h, 86% (2 steps).
Completion of the glycosylated monomer synthesis was exemplified
with â-galactose-based monomer 6a via amino acid deprotection,
to give the O-acetyl protected galactosyl â-amino acid 7, which
has been used as a model to illustrate the foldamer-based concept
associated with a triglycosylated hexapeptide 1.
The peptide coupling strategy employed the glycosylated and
cyclopentyl derivatives 7 and 8 (and 10), respectively (Scheme 2).
We initially used a solid-phase synthesis strategy, but use of
standard Fmoc protocols (with a HATU/HOBt/NMM coupling
strategy) on Rink amide resin failed. This was because the hindered
(i + 3)/(i + 4) coupling step involving two glycosylated units (7
+ 9 to generate 11) did not go to completion. To overcome this
problem, a convergent solution phase approach based on a more
† University of Bristol.
§ AstraZeneca, Silk Road Business Park.
‡ AstraZeneca, Mereside.
9
10638
J. AM. CHEM. SOC. 2006, 128, 10638-10639
10.1021/ja0614565 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2. Synthesis of Glycosylated Hexapeptides 13 and 14a
Figure 2. CD data for hexapeptides 13 and 14 in methanol and water.
Data have been normalized for concentration and number of residues.
i
i
a Conditions: (a) ClCO2 Bu, NMM, -10 °C, 71%; (b) ClCO2 Bu, NMM,
Acknowledgment. We thank Professor Derek Woolfson and
Dr Maxim Ryadnov for advice and access to CD facilities and
AstraZeneca, EPSRC, and the Royal Thai Government for financial
support.
i
-10 °C, 43%; (c) ClCO2 Bu, NMM, -10 °C, 61%; (d) 20% piperidine in
DMF; (e) ClCO2 Bu, NMM, -10 °C, 74% (2 steps); (f) 20% piperidine in
DMF; (g) BzCl, Et3N, 0 °C, 80%; (h) MeOH/NH3, 88%. Nonadjacent NOEs
observed for peptides 13 and 14 in MeOH (see text) are shown with double
headed arrows. NMM ) N-methylmorpholine.
i
Supporting Information Available: Experimental details. This
material is available free of charge via the Internet at http://pubs.acs.org.
reactive mixed anhydride was employed. The coupling of 7 and 8
to give 9 was successful, as was the reaction of 9 with each of 7
and 10 to give tripeptides 11 and 12, respectively. A 3 + 3 coupling
of tripeptides 11 and 12 gave the triglycosylated hexapeptide 13,
capped at the N- and C-termini with benzoyl and primary amide,
respectively, and the carbohydrate moieties were deacetylated to
give the target triglycosylated hexapeptide 14. Both 13 and 14 were
purified by reverse phase chromatography, and assignments were
made by NMR (600 MHz) and ESI mass spectrometry.
2D NMR (NOESY and ROESY) analysis of O-protected peptide
13 in CD3OH showed multiple i and i + 2 NOEs between backbone
protons. Specifically CâH(i)-CRH(i + 2) and CâH(i)-NH(i + 2)
NOEs are consistent with the expected 12-helical pattern and
correlated well with Gellman’s sulfonylated pyrrolidine/cyclopentyl
foldamers, the structures of which have been assessed in detail.5d
Peptide 14 shows a very similar set of nonadjacent NOEs in CD3-
OH, and analysis of 14 in H2O also demonstrated a predominant
12-helical conformation.9
These results indicate that glycosylated peptide 14 is a 12-helical
foldamer under aqueous solution and that incorporation of carbo-
hydrate units (protected or unprotected) into this array does not
alter the secondary structure. These findings were corroborated by
circular dichroism (CD) studies (Figure 2). In MeOH, both peptides
13 and 14 showed a characteristic maximum at 204 nm and a
weaker minimum at 230 nm. In water the λmax of peptide 14 is
shifted to 202 nm and λmin to 225 nm.10
In conclusion, we have demonstrated a model system amenable
to the study of carbohydrate-carbohydrate interactions, which
complements multivalency-based approaches. Glycosylation of the
foldamer scaffold does not perturb the helical structure necessary
to maintain the carbohydrate units in close proximity, and the model
triglycosylated peptide 14 also serves as an important control for
future studies. Our next objective is to incorporate carbohydrate
moieties that are associated with an established and biologically
significant carbohydrate-carbohydrate interaction to study this
process in comparative isolation and to detail the nature of the
mechanisms involved.
References
(1) (a) Hakomori, S. Arch. Biochem. Biophys. 2004, 426, 173. (b) Hakomori,
S. Pure Appl. Chem. 1991, 63, 473. (c) Misevic, G. N.; Burger, M. M.
Experientia 1987, 43, 676.
(2) (a) Pincet, F.; Le Bouar, T.; Zhang, Y.; Esnault, J.; Mallet, J.-M.; Perez,
E.; Sinay¨, P. Biophys. J. 2001, 80, 13. (b) Rojo, J.; Morales, J. C.; Penade´s,
S. Top. Curr. Chem. 2002, 218, 45. (c) Bucior, I.; Burger, M. M.
Glycoconjugate J. 2004, 21, 111. (d) Hakomori, S. Glycoconjugate J. 2004,
21, 125.
(3) The Lewisx-Lewisx interaction has provided a focus in this area. (a)
Siuzdak, G.; Ichikawa, Y.; Caulfield, T. J.; Munoz, B.; Wong, C.-H.;
Nicolaou, K. C. J. Am. Chem. Soc. 1993, 115, 2877. (b) Geyer, A.; Gege,
C.; Schmidt, R. R. Angew. Chem., Int. Ed. Engl. 1999, 38, 1466. (c) Geyer,
A.; Gege, C.; Schmidt, R. R. Angew. Chem., Int. Ed. 2000, 39, 3246. (d)
Tromas, C.; Rojo, J.; de la Fuente, J. M.; Barrientos, A. G.; Garc´ıa, R.;
Penade´s, S. Angew. Chem., Int. Ed. 2001, 40, 3052. (e) Nishida, Y.;
Tsurumi, T.; Sasaki, K.; Watanabe, K.; Dohi, H.; Kobayashi, K. Org.
Lett. 2003, 5, 3775. (f) de la Fuente, J. M.; Eaton, P.; Barrientos, A. G.;
Menendez, M.; Penades, S. J. Am. Chem. Soc. 2005, 127, 6192.
(4) (a) Bregant, S.; Zhang, Y. M.; Mallet, J. M.; Brodzki, A.; Sinay¨, P.
Glycoconjugate J. 1999, 16, 757. (b) Esnault, J.; Mallet, J. M.; Zhang, Y.
M.; Sinay¨, P.; Le Bouar, T.; Pincet, F.; Perez, E. Eur. J. Org. Chem.
2001, 2, 253. (c) Gourier, C.; Pincet, F.; Perez, E.; Zhang, Y. M.; Mallet,
J. M.; Sinay¨, P. Glycoconjugate J. 2004, 21, 165. (d) Santacroce, P. V.;
Basu, A. Glycoconjugate J. 2004, 21, 89. (e) Herna´iz, M. J.; de la Fuente,
J. M.; Barrientos, A. G.; Penade´s, S. Angew. Chem., Int. Ed. 2002, 41,
1554. (f) de la Fuente, J. M.; Barrientos, A. G.; Rojas, T. C.; Rojo, J.;
Can˜ada, J.; Ferna´ndez, A.; Penade´s, S. Angew. Chem., Int. Ed. 2001, 113,
2317. (g) Barrientos, A. G.; de la Fuente, J. M.; Rojas, T. C.; Ferna´ndez,
A.; Penade´s, S. Chem.sEur. J. 2003, 9, 1909. (h) Rojo, J.; D´ıaz, V.; de
la Fuente, J. M.; Segura, I.; Barrientos, A. G.; Riese, H. H.; Bernade, A.;
Penade´s, S. Chembiochem 2004, 5, 291.
(5) (a) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173. (b) Appella, D. H.;
Barchi, J. J.; Durell, S. R.; Gellman, S. H. J. Am. Chem. Soc. 1999, 121,
2309. (c) Wang, X. F.; Espinosa, J. F.; Gellman, S. H. J. Am. Chem. Soc.
2000, 122, 4821. (d) Lee, H.-S.; Syud, F. A.; Wang, X.; Gellman, S. H.
J. Am. Chem. Soc. 2001, 123, 7721. (e) Lee, H.-S.; LePlae, P. R.; Porter,
E. A.; Gellman, S. H. J. Org. Chem. 2001, 66, 3597. (f) Cheng, R. P.;
Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001, 101, 3219.
(6) The lactose-lactose interaction has been reported to stabilize a peptide
array but only in hydrophobic solvents (polyfluorinated alcohols) and not
in buffer. Hasegawa, T.; Sasaki, T. Chem. Commun. 2003, 978.
(7) Blake, J.; Wilson, C. D.; Rapoport, H. J. Am. Chem. Soc. 1964, 86, 5293.
(8) Of a range of substrates tested, crotyl-crotyl partners led to less alkene
isomerization and limited the formation of homometathesis products.
(9) In total, three CâH(i)-CRH(i + 2) and four CâH(i)-NH(i + 2) NOEs
were observed in the central portion of 14 in water (see Supporting
Information). All possible CâH(i)-NH(i + 1) NOEs were observed,
consistent with the 12-helix.5d,f
(10) Titration of 14 with CaCl2 (from 0.1 to 100 mM of Ca2+) showed no
effect (by CD) on the secondary structure of peptide 14, and we infer
that no interactions are associated with this simple Gal-based model.
JA0614565
9
J. AM. CHEM. SOC. VOL. 128, NO. 33, 2006 10639