Application of Ugi reactions in synthesis of divalent neoglycoconjugates: Evidence that the sugars are presented in restricted conformation

Article abstract of DOI:10.1021/ol016294q

(Equation presented) The Ugi reaction has been used to prepare divalent galactose derivatives. NMR analysis shows that a divalent neoglycoconjugate, where the glycopeptides are bridged by a terephthaloyl group, is an 83:17 mixture of two conformers; the amide groups of the major isomer have E-anti conformations. The spatial relationship and the relative orientation of the sugars are restricted, which may have consequences for the recognition of this and related structures in biological systems.

Full text of DOI:10.1021/ol016294q

ORGANIC
LETTERS
2
001
Vol. 3, No. 17
629-2632
Application of Ugi Reactions in
Synthesis of Divalent
2
Neoglycoconjugates: Evidence That the
Sugars Are Presented in Restricted
Conformation
†
†
†
‡
Helena Bradley, Geraldine Fitzpatrick, W. Kenneth Glass, Horst Kunz, and
,
†
Paul V. Murphy*
Chemistry Department, Conway Institute of Biomolecular and Biomedical Research,
UniVersity College Dublin, Dublin 4, Ireland, and Institut f u¨ r Organische Chemie,
Johannes Gutenburg-UniVersit a¨ t, Duesbergweg 10-14, D-55128 Mainz, Germany
paul.V.murphy@ucd.ie
Received June 18, 2001
ABSTRACT
The Ugi reaction has been used to prepare divalent galactose derivatives. NMR analysis shows that a divalent neoglycoconjugate, where the
glycopeptides are bridged by a terephthaloyl group, is an 83:17 mixture of two conformers; the amide groups of the major isomer have E-anti
conformations. The spatial relationship and the relative orientation of the sugars are restricted, which may have consequences for the recognition
of this and related structures in biological systems.
4
5
Oligosaccharides and proteins, heterogeneously coated on
cell surfaces, provide a complex environment where a variety
of functions take place. These include cell-cell interactions
involved in the immune response, inflammation, cell signal-
ing and infection by pathogens. The design, synthesis, and
development of carbohydrate- and glycoconjugate-based drug
molecules that interfere with the binding events at cell
inhibitors, and xenotransplantation. Strategies that have
been adopted for the purpose of discovery of therapeutics
6
from carbohydrates include combinatorial synthesis, the
synthesis of carbohydrate mimetics and of carbohydrates
modified with hydrophobic groups that increase affinity for
7
(
3) (a) Liebe, B.; Kunz, H. Angew. Chem., Int. Ed. Engl. 1997, 36, 618.
b) Keil, S.; Claus, C.; Dippold, W.; Kunz, H. Angew. Chem., Int. Ed. 2001,
40, 366 and references therein. (c) Deshpande, P.; Danishefsky, S. J. Nature
(
1
surfaces is an emerging therapeutic area. Much recent work,
1
997, 387, 167. (d) Hummel, G.; Schmidt, R. R. Tetrahedron Lett. 1997,
for example, has focused on development of novel inhibitors
38, 1173.
2
of carbohydrate-selectin recognition. Other areas of interest
(
4) (a) von Itzstein, M.; Wu, W.-Y.; Kok, G. B.; Pegg, M. S.; Dyason,
3
J. C.; Jin, B.; Phan, T. V.; Smythe, M. L.; White, H. F.; Oliver, S. W.;
Colman, P. M.; Varghese, J. N.; Ryan, D. M.; Woods, J. M.; Bethell, R.
C.; Hotham, V. J.; Cameron, J. M. Nature 1993, 363, 418. (b) Kamitakahara,
H.; Suzuki, T.; Nishigori, N.; Suzuki, Y.; Kanie, O.; Wong, C.-H. Angew.
Chem., Int. Ed. 1998, 37, 1524.
include the development of cancer vaccines, influenza
†
University College Dublin.
Johannes Gutenburg-Universit a¨ t.
‡
(
1) Wong, C.-H. Acc. Chem. Res. 1999, 32, 376.
(5) Liu, B.; Roy, R. J. Chem. Soc., Perkin Trans. 1 2001, 773.
(6) Schweizer F.; Hindsgaul, O. Curr. Opin. Chem. Biol. 1999, 3, 291
and references therein.
(2) Simanek, E. E.; McGarvey, G. J.; Jablonski, J. A.; Wong, C.-H. Chem.
ReV. 1998, 98, 833 and references therein.
1
0.1021/ol016294q CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/02/2001

8
the receptor by subsite-assisted binding. These approaches
have met with some success, yet there is still considerable
difficulty in discovery of high affinity ligands for carbohy-
drate receptors, as the monovalent epitopes tend to bind their
receptors weakly (millimolar range). Carbohydrates on cell
surfaces are displayed in clusters (or multivalent arrays), and
it is most likely that cooperative effects are responsible for
increasing the strength of binding of these ligands to their
Scheme 2
9
receptors and the resulting potency of these compounds.
Multivalent carbohydrates can also affect the nature of the
biological response that results.10 The development of
strategies for synthesis of multivalent (di-, tri- and higher
order) carbohydrate displays is required for the development
of therapeutics from carbohydrates and for providing com-
pounds, which are structurally well defined, for studies of
the mechanisms of action of these ligands.
isocyanoacetate gave the simple divalent galactose dimer 3.
The acetate groups can be removed using NaOMe/MeOH
without degradation of the product to give the homodimeric
glycoconjugate 4.
Multicomponent reactions such as the Ugi reaction (Scheme
1) have generated much interest because of their synthetic
The Ugi reaction of 1, formaldehyde, methyl isocyano-
acetate (2.0 equiv of each reagent), and terephthalic acid (1.0
equiv) gave the dimeric carbohydrate 5 (Scheme 3). The
acetates can again be removed to give 6.
Scheme 1
Scheme 3
potential, their utility in combinatorial chemistry, and for
11
the generation of molecular diversity. The potential of the
reaction in carbohydrate chemistry and biology has been
recognized by other researchers.12
We have decided to investigate synthetic routes to divalent
and higher order multivalent carbohydrates that will also
facilitate introduction of diversity. Divalent carbohydrates
13
can show improved biological activity, and therefore we
have initially investigated the suitability of the Ugi reaction
for convergent one-pot synthesis of dimeric carbohydrates.
Thus galactose amine 1 can be used to prepare conjugate
2
by condensation with succinic anhydride in the presence
of diisopropylethylamine in dichloromethane (56%, Scheme
). The Ugi reaction of 1, 2, formaldehyde, and methyl
2
Having prepared 4 and 6, we wanted to establish whether
there are any preferred solution structures because of the
potential relevance of conformation to the recognition of
(
7) Murphy, P. V.; Hubbard, R. E.; Manallack, D. T.; Wills, R. E.;
Montana, J. G.; Taylor, R. J. K. Bioorg. Med. Chem. 1998, 6, 2421.
8) Arya, P.; Kutterer, K. M. K.; Qin, H.; Roby, J.; Barnes, M. L.; Kim,
J. M.; Roy, R. Bioorg. Med. Chem. Lett. 1998, 8, 1127.
9) (a) Kiessling, L. L.; Gestwicki, J. E.; Strong, L. E. Curr. Opin. Chem.
divalent and multivalent ligands in biological systems.14
A
(
number of carbohydrate derivatives 7-9 related to those
(
found in the divalent compounds were also prepared (Scheme
Biol. 2000, 4, 696. (b) Mammen, M.; Choi, S. K. Whitesides, G. M. Angew.
Chem., Int. Ed. 1998, 37, 2755.
1
5
4) to aid the structural studies.
(
10) Gestwicki, J. E.; Strong, L. E.; Kiessling, L. L. Chem. Biol. 2000,
, 583.
11) Domling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000, 39, 33168 and
references therein.
12) (a) Oertel, K.; Zech, G.; Kunz, H. Angew. Chem., Int. Ed. 2000,
9, 1431 and references therein. (b) Park, W. K. C.; Auer, M.; Jaksche, H.
NMR analyses of â-glycosyl amido derivatives have been
7
1
6,17
(
reported previously.
However, the conformation and
O has
configuration of the unprotected carbohydrates in D
2
(
3
J. Am. Chem. Soc. 1996, 118, 10150. (c) Sutherlin, D. P.; Stark, T. M.;
Armstrong, R. W. J. Org. Chem. 1996, 61, 8350. (d) Tsai, C.-Y.; Park, W.
K. C.; Weitz-Schmidt, G.; Ernst, B.; Wong, C.-H. Bioorg. Med. Chem. Lett.
(14) Roy, R. Das, S. K. Santoyo-Gonzalez, F. Hernandez-Mateo, F.; Dam,
T. K.; Brewer, C. F. Chem. Eur. J. 2000, 6, 1757.
(15) Experimental protocols were as described previously; see: Retail-
leau, L.; Laplace, A.; Fensterbank, H.; Larpent, C. J. Org. Chem. 1998,
63, 608;
(16) The R-glycosides have not been reported. Conventions for describing
the configurations and conformations of the â-derivatives have been
described previously; see: Avalos, M.; Babiano, R.; Dur aj n, C. J.; Jim ej nez,
J. L.; Palacios, J. C. J. Chem. Soc., Perkin Trans. 2, 1992, 2205 and
references therein.
1
998, 8, 2333. (e) Lockhoff, O. Angew. Chem., Int. Ed. 1998, 37, 3436. (f)
Ziegler, T.; Gerling, S.; Lang M. Angew. Chem., Int. Ed. 2000, 39, 2109.
Nunns, C. L.; Spence, L. A.; Slater, M. J.; Berrisford, D. J. Tetrahedron
Lett. 1999, 40, 9341.
(13) Kogan, T. P.; Dupre, B.; Bui, H.; McAbee, K. L.; Kassir, J. M.;
Scott, I. L.; Hu, X.; Vanderslice, P.; Beck, P. J.; Dixon, R. A. F. J. Med.
Chem. 1998, 41, 1099.
2630
Org. Lett., Vol. 3, No. 17, 2001

were consistent with the structure proposed. The signals in
1
13
Scheme 4a
the H- and C NMR spectra of unprotected 4, 6, and 7-9
in D O at 10-40 °C were not broadened, and this simplified
2
the determination of the preferred solution structures. The
resonances for the anomeric proton were generally doublets
and had coupling constants consistent with assignment of
the â-configuration; the signals and NOE enhancements
4
observed for the ring protons confirmed that the
C
1
conformation is favored for the pyranose in each case. The
NMR spectra, recorded in D O, of 4, 7, and 8 clearly show
two signal sets (Table 1); NOE and chemical shift data
a
2
2 3
Reagents and conditions: (a) BzCl, Na CO , MeOH, 0 °C, 1
h; (b) Ac
2 2 3
O, Na CO , MeOH, 0 °C, 1 h.
not been studied. The H and 13C NMR spectra for most of
the known derivatives show two signal sets, and this has
been attributed to observation of the Z- and E-isomers of
the â-D-glycosylamide. Babiano and co-workers and others
have found that an antiperiplanar (anti) rather than syn-
periplanar (syn) conformation is generally preferred (see
Figure 1). They have established a set of rules based on
1
a
Table 1. Selected NMR Data for Glycosylamides
δ H-1 (J 1-2, Hz)
δ C-1
entry
compd
E-anti
Z-anti
E-anti
Z-anti
4b
6c
7
8
9
1
2
3
4
5
5.06 (8.0)
4.90 (9.0)
4.98 (9.0)
4.71 (9.0)
4.90 (9.0)
5.45 (9.0)
89.0
88.5
87.5
88.0
88.5
85.3
5.40 (9.0)
5.63 (9.5)
82.5
82.7
a
NMR were recorded in D2O from 10 to 40 °C. b Data corresponds to
the galactose residue adjacent to the tertiary amide. c Data given for the
major conformational isomer.
indicate that the anti- rather than syn-conformation is
preferred. Therefore the signal sets can be assigned to Z-
and E-anti isomers, and their respective proportions at
equilibrium can be determined by integration of anomeric
1
proton signals in the H NMR spectra (4, 30:70; 7, 45:55; 8,
1
5:85). The NMR spectra for 6 (and also 9) indicate that
there is one major isomer present in solution; the E-anti
conformation can be assigned to this isomer on the basis of
1
the chemical shift of the anomeric proton ( H NMR, δ 4.98,
d, 40 °C). Also, a series of 1D NOE, 2D NOESY, and 2D
ROESY experiments for 6 showed strong NOE enhance-
ments (Figure 1) for the major isomer consistent with the
amides adopting E-anti conformations. The presence of a
minor isomer, where one amide adopts the E-anti and the
other adopts the Z-anti conformation (Figure 2), can be
Figure 1. There are strong NOE enhancements for 6 between the
aromatic protons and H-1 and also between H-2 and the methylene
group adjacent to the anomeric center.
chemical shifts in H and 13C spectra that can be used to
1
assign the Z or E configuration; these have proven very useful
16
herein. Thus, for the glycosylamides, the chemical shift of
the anomeric proton of the Z-isomer will be greater than that
of the E-isomer as a result of deshielding caused by the
carbonyl group; the chemical shift of the anomeric carbon
of the Z-isomer will be less than that of the E-isomer. In
our experience, the protected products such as 5 can exhibit
fluxional behavior, presumably through the interconversion
of rotamers, and this resulted in broad signals in the NMR
Figure 2. The 10 lowest energy conformations of 8 predicted by
a Monte Carlo conformational search are shown; the pyranose ring
atoms have been superimposed.
1
13
spectra at room temperature. The signals in H and C NMR
spectra of 5 recorded in C N at 100 °C did coalesce and
5 5
D
(17) Related tertiary amides that have been prepared previously include
N-alkylglucosyl(meth)acrylamides as surfactants and amphiphilic glycolipid
analogues as stimulators of specific immune responses against antigens;
see: Lockhoff, O.; Sadler, P. Carbohydr. Res. 1998, 314, 13 and ref 15.
detected from the NMR data as there are two anomeric
signals ( H NMR, δ 5.90 and δ 4.90, 40 °C) that show
1
18
Org. Lett., Vol. 3, No. 17, 2001
2631

clear cross-peaks with the anomeric signal at δ 4.98 in the
NOESY and ROESY spectra; these cross-peaks have op-
posite sign to the NOE enhancements supporting exchange
between the isomers.19
Conformational searches using Macromodel 6.0 were
20
carried out for 8 (Figure 2). The low energy structures
found comprised Z- and E-anti isomers.21
On the basis of our interpretation of the data obtained we
can propose that the major conformational isomer adopted
by 6 in D O is either 6a or 6b (Figure 3). Both of these
2
isomers contain a C-2 axis of symmetry; for 6a the
carbohydrates are in a cis arrangement, and for 6b they are
trans. Calculations similar to those carried out for 8 indicate
that 6b may be more stable, but there is no experimental
evidence to support this. Further work is necessary to
establish the structure of the preferred conformer, if indeed
there is one. However, the preliminary data indicate that for
6
there is a significant population of isomers where the
topographical arrangement between the carbohydrates is
restricted. The biological recognition of this and related
amides should be of interest and is under investigation.
In summary, multicomponent Ugi reactions have been used
for convergent synthesis of divalent carbohydrate derivatives.
Further work will be required to improve the yields of the
reactions and to simplify the purification procedures. These
compounds exhibit interesting conformational preferences.
An interesting goal will now be to design and synthesize
conformationally diverse libraries of divalent and higher
order multivalent ligands where each member will have a
restricted, yet different, topographical arrangement between
the recognition components. Such libraries will be useful for
further studies related to the mechanism of action of
Figure 3. NMR indicates that an 83:17 mixture of E/E (6a/6b)
and Z/E isomers (6c/6d) is present at 40 °C in D O.
2
multivalent ligands. They will also aid the development of
potent carbohydrate-based therapeutics. This work is cur-
rently underway.
Acknowledgment. The authors thank Enterprise Ireland,
Pfizer Pharmaceuticals Inc., Co. Cork for an Irish American
Partnership; DAAD; HRB/Wellcome Trust, Dr. Martin
Walsh and Professor Richard J. K. Taylor.
(
18) At 10 °C these signals are doublets at δ 5.58 and δ 4.60.
(19) Exchange between Z- and E-anti conformations has been observed
Supporting Information Available: Experimental pro-
cedures for synthesis of 4 and 6; 1D and 2D NMR spectra
for 4 and 6-9; and X-ray crystallography statistics for
aromatic tertiary amides from CCDC. This material is
available free of charge via the Internet at http://pubs.acs.org.
also for 8 from NOE and variable temperature NMR experiments.
2
(20) Calculations were carried out on an O silicon graphics workstation
using Macromodel 6.0. Monte Carlo conformational searches were carried
out using the SUMM method and the all-atom amber Force Field.
(21) It was noted that the aromatic ring was not planar to the carbonyl
group in the calculated structures. X-ray crystallographic data indicate this
is preferred. See Supplementary Information for statistics on data deposited
at the CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.
OL016294Q
2632
Org. Lett., Vol. 3, No. 17, 2001