A focused sulfated glycoconjugate Ugi library for probing heparan sulfate-binding angiogenic growth factors

Article abstract of DOI:10.1016/j.bmcl.2012.08.001

A library of small molecule heparan sulfate (HS) mimetics was synthesized by employing the Ugi four-component condensation of d-mannopyranoside-derived isocyanides with formaldehyde as the carbonyl component and a selection of carboxylic acids and amines,

Full text of DOI:10.1016/j.bmcl.2012.08.001

Bioorganic & Medicinal Chemistry Letters 22 (2012) 6190–6194
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A focused sulfated glycoconjugate Ugi library for probing heparan
sulfate-binding angiogenic growth factors
Ligong Liu a,b, , Caiping Li a, Siska Cochran a, Daniel Feder c, Luke W. Guddat c, Vito Ferro a,c,
⇑
⇑
a Progen Pharmaceuticals Limited, Brisbane, Qld 4076, Australia
b Institute for Molecular Bioscience, The University of Queensland, Brisbane, Qld 4072, Australia
c School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Qld 4072, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 10 August 2012
A library of small molecule heparan sulfate (HS) mimetics was synthesized by employing the Ugi four-
component condensation of -mannopyranoside-derived isocyanides with formaldehyde as the carbonyl
D
component and a selection of carboxylic acids and amines, followed by sulfonation. The library was used
to probe the subtle differences surrounding the ionic binding sites of three HS-binding angiogenic growth
factors (FGF-1, FGF-2 and VEGF). Each compound features 3 or 4 sulfo groups which serve to anchor the
ligand to the HS-binding site of the protein, with a diverse array of functionality in place extending from
C-1 or C-6 to probe for adjacent favorable binding interactions. Selectivity of binding to these proteins
was clearly observed and supported by molecular docking calculations.
Keywords:
Growth factors
Ugi condensation
Heparan sulfate mimetics
Glycoconjugates
Inhibitors
Ó 2012 Elsevier Ltd. All rights reserved.
Finding more efficient approaches to accelerate lead discovery
and optimization has been a continuing theme for drug discov-
ery in recent years. Out of many outstanding strategies, multi-
component reactions have attracted significant attention from
both the academic and industrial sectors due to their efficiency
in the generation of chemical diversity.1 The Ugi four-component
condensation is the most important isocyanide-mediated multi-
component reaction, and has been explored in many drug
discovery and development applications2 including in the glyco-
mics area.3 Some years ago the potential of the Ugi reaction for
targeting angiogenic heparan sulfate (HS)-binding growth factors
such as fibroblast growth factors 1 and 2 (FGF-1 and -2) and
vascular endothelial growth factor (VEGF) was demonstrated.4,5
Herein we report a focused library approach utilizing the Ugi
reaction for probing the ligand–protein binding sites in order
to identify higher affinity small molecule leads for these HS-
binding growth factors.
The FGFs and VEGF are important mediators of angiogenesis and
thus are attractive targets for drug discovery.6 They initiate cell sig-
naling cascades that lead to angiogenesis by initial formation of ter-
nary complexes with HS and their cognate receptors. Inhibiting
angiogenesis by targeting the HS-binding site with HS mimetics
and thus blocking the formation of these ternary complexes is a via-
ble therapeutic strategy for cancer.7 The FGFs have been widely
studied and a number of X-ray crystal structures are available,
including in complex with heparin-derived oligosaccharides,8 and
in ternary complex with its receptor (FGFR) and a heparin-derived
oligosaccharide.9 The HS-binding sites of these proteins share a
common feature, that is, a densely charged, cationic, shallow
groove.
Over the past decade, we have explored the use of small mole-
cule HS mimetics as potential inhibitors of FGF- and/or VEGF-med-
iated angiogenesis for the development of novel cancer
therapeutics.10 Such approaches avoid the inherent synthetic com-
plexity associated with natural HS oligosaccharide sequences and
have the potential for modifying biological activity by simple
chemical modifications. Among the various approaches, we repor-
ted10e the use of a panel of linked sulfated cyclitols containing the
core structure 3 (Fig. 1) to probe various HS-binding proteins.
These studies indicated that ligand selectivity towards such pro-
teins could be influenced by subtle changes in the size and orien-
tation of the spacers between the two highly charged centres,
which is in agreement with the generally accepted view that differ-
ent domains or structural motifs of HS bind to different proteins to
mediate their effects.11
In the current design strategy a simple ionic binding motif on a
monosaccharide scaffold that could act to ‘anchor’ the ligand to the
HS-binding site on the target protein was envisaged. The Ugi reac-
tion would then be utilized to decorate the scaffold, via various
linkers, with a diverse range of functional groups in order to probe
for additional favorable binding interactions. This concept allows
for the rapid assembly of a library of small molecules for use as a
tool to map the structural tolerances on the surrounding surface
of the ionic binding site of these related proteins, with the aim of
⇑
Corresponding authors. Tel.: +61 7 3346 2988; fax: +61 7 3346 2101 (L.L.);
tel.: +61 7 3346 9598; fax: +61 7 3365 4299 (V.F.).
E-mail addresses: l.liu@imb.uq.edu.au (L. Liu), v.ferro@uq.edu.au (V. Ferro).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.08.001

L. Liu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6190–6194
6191
Figure 1. Library design based on Man-6-isocyanide (1; series 1) and Man-1-isocyanide (2; series 2) and the core structure motif of linked cyclitols (3). X = SO3Na.
improving affinity and selectivity for each individual protein. This
approach also seeks to reduce the overall charge of the ligands by
minimizing the number of sulfo groups in the HS-binding anchor
and by introducing mostly hydrophobic (aliphatic or aromatic)
groups via the Ugi reaction.
sulfonation step. Purification of the multi-charged products was
not always straightforward and sometimes required a combination
of several purification techniques, such as silica gel or size-exclu-
sion column chromatography and solid phase extraction (SPE).
For those compounds with an extra carboxylate group, which
was introduced as the methyl ester during the Ugi reaction step,
the crude sulfated product was further subjected to a sodium
hydroxide mediated hydrolysis step prior to the standard purifica-
tion. The purified product fractions were individually analysed by
capillary electrophoresis (CE), pooled and lyophilized to give the fi-
nal sulfated material as an amorphous powder (purity P 98% by
CE).
The binding affinities of the library for FGF-(1,2) and VEGF were
measured using a surface plasmon resonance (SPR) solution affin-
ity assay (Table 1).14 The affinities for the non-Ugi elaborated par-
ent compounds10f,15 trisulfate 15 and tetrasulfate 16 are also
provided for comparison. The measured Kd values clearly showed
the varied preference of each protein towards the ligands, with
the affinities spanning 38-, 7- and 104-fold differences for FGF-1,
FGF-2 and VEGF, respectively. For FGF-1, either an aromatic group
or an extra negative charge is preferred to an aliphatic group for in-
To limit the number of possible structures, we selected only two
monosaccharide scaffolds 1 and 2 based on D-mannopyranose with
extension from the more readily accessible positions C-1 and C-6
(Fig. 1). These common scaffolds were designed to contain the iso-
cyanide component for the Ugi reaction whilst the carboxylic acid
and amine components could be selected from a vast range of com-
mercially available structures. The simplest carbonyl component,
formaldehyde, was chosen as the fourth partner to avoid the com-
plexity of generating an additional diastereomeric centre. The
selection of D-mannose was also based on existing internal pro-
grams involving this sugar as well as the presence of a pair of
cis-oriented vicinal hydroxyl groups which can be sulfonated, a
feature common to the linked cyclitol core structure 3.
The preparation of mannoside-6-isocyanide 9 is outlined in
Scheme 1. By following standard functional group transformations,
the alcohol 412 was converted into the corresponding amine 7 in
54% overall yield. Subsequent formylation and dehydration affor-
ded 9 in high yield (93%). The corresponding C1-isocyanide 10
was prepared as reported earlier.4 The structures of the intermedi-
ates and products were characterized by 1H and/or 13C NMR spec-
troscopy.13 Both building blocks 9 and 10 showed characteristic
signals for isocyanide carbon at 157.63 and 157.57 ppm, respec-
tively, in 13C NMR spectra.
creased affinity. Additional charged groups either in the D-manno-
side ring or as an attached functional group did not always
correlate with improved affinity. For example, 1Aa-1Ba-1Ca-1Ig-
1Ih-1Ii showed lower affinities compared with their corresponding
series 2 compounds. 1He (37
lM) not only binds more tightly than
its Series 2 analogue 2He (171
lM) but is also one of the best com-
pounds out of the panel for FGF-1. Although extra charged groups
The components for the Ugi library construction are illustrated
in Scheme 2. Although formaldehyde has not been used frequently
in the Ugi reaction, we found in general that the yields using this
carbonyl source were good. In a few cases, elevated temperatures
were required to increase the yield (such as entry 9, Table 1).
The presence of free hydroxyl groups was generally well tolerated
during the Ugi reaction except in the case of glycolic acid, which
gave an unidentified mixture. However, protection of the hydroxyl
group as the benzyl ether gave the desired Ugi adduct in good yield
(69%, ⁄⁄⁄entry 13, Table 1). Global debenzylation was carried out
in quantitative yield by hydrogenolysis with 20% Pd(OH)2 on char-
coal as catalyst. The crude polyol, after drying under high vacuum
in the presence of phosphorus pentoxide, was used directly for the
(motifs g, h, i) had a greater influence when delivered in Series 2
(20–44
groups that result in the tightest binding compound (2Ca at
11.8 M) and provide more useful SAR information (the homo-
logues 2Ba and 2Aa at 24.8 and 97.4 M, respectively). The affini-
lM) compared with Series 1 (111–338 lM), it is aromatic
l
l
ties of the ligands towards FGF-2 showed the smallest variation,
with the best two containing either a flexible aromatic group
(113 lM for 2Ca) or an extra negative charge (118 lM for 2Ig).
For VEGF, ligand binding affinities followed a similar trend to
FGF-1, which may imply that there is some similarity in their
HS-binding domains. However, one striking difference was the
intolerance of VEGF to cyclic aliphatic rings (1–3 mM for D, E, F,
G), while an aromatic group (28.8–109 lM for motif A, B, C, d, e)
Scheme 1. Reagents and conditions: (a) MsCl (1.5 equiv), DMAP (0.11 equiv), Et3N (3 equiv), DCM (0.06 M), 0–18 °C, 18 h, 100%; (b) NaN3 (5 equiv), DMF (0.2 M), 90 °C, 3 h,
63%; (c) (i) PPh3 (1.3 equiv), THF (0.2 M), 18 °C, 1 h, (ii) H2O (20 equiv), 18 °C, 18 h, 85%; (d) Ac2O (30 equiv), HCO2H (0.06 M), 18 °C, 3.5 h; (e) POCl3 (2 equiv), TEA (4 equiv),
DCM (0.05 M), 0 °C, 2 h, 93% (two steps).

6192
L. Liu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6190–6194
Scheme 2. Reagents and conditions: (a) HCHO (1 equiv, 36.5–38.0% w/w, aqueous solution stabilized with 10% MeOH), R1NH2 (1 equiv), R2CO2H (1 equiv), MeOH (0.4 M), 18
or 60 °C, 24 h, 36–75%; (b) 20% Pd(OH)2 on charcoal, MeOH, 50 PSI, 2 h, 100%; (c) (i) SO3ꢀMe3N or SO3ꢀpyridine complex (3 equiv per hydroxyl), DMF (0.04 M), 60 °C, 18 h, (ii)
pH 9, 1 M NaOH, 0 °C, evaporated to dryness, (iii) only for series with methyl ester (g0 and h0), 1 M NaOH, 18 °C, 18 h, (iv) Bio-Gel P-2 (0.1 M NH4HCO3) and/or LH-20 (DI water)
column chromatography, silica gel chromatography, or SPE with Strata-X cartridge or Strata-NH2 cartridge, (v) ion-exchange resin column (AGÒ-50W-X8, Na+ form), (vi)
freeze–dry (R1 and R2 see Table 1).
Table 1
Yield of the Ugi library production and dissociation constants measured for sulfated products binding to FGF-(1,2), VEGF
Entry
Yield (%)a
Sulfated product (steps b,c)
Structural Code
Kd
(
l
M)b
Ugi reaction (step a)
FGF-1
FGF-2
VEGF
1
2
3
4
5
6
7
8
64
38
42
60
82
62
49
56
45
40
32
51
45
24
15
25
63
67
18
52
73
46
29
41
68
14
57
28
28
51
34
—
1Aa
1Ba
1Ca
1Da
1Ea
1Fa
1Ga
1He
1Ja
1If
1Ig
1Ih
1Ii
1Ib
1Ic
1Id
2Aa
2Ba
2Ca
2He
2Ig
2Ih
2Ii
196 29
140 12
224 30
310 50
370 40
265 28
243 21
37 10
480 70
192 16
680 60
76#
77#
109#
413
8
1700#
3000#
1200#
1000#
45.3 1.1
280 210
512.0 2.8
207#
410 70
480 90
540 30
430 100
240 50
ND
291 16
480 120
240 80
480 80
ND
400 110
240 40
287 27
113 20
837#
9
56c,d
69e
65f
66f
52d
76, 97d
82
95 19
13
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
174#
338 18
160 60
111 26
104#
545#
366
ND
9
1100#
596#
67
500 400
97 19
25 16
11.8 2.1
170 50
43 14
81#
24, 74d
67
402#
76.6#
29#
12, 45d
37
91#
65f
75f
51d
—
122
5
40#
44 13
188 18
157 25
210 15
340 150
81#
20
6
50#
15
16
343 19
130 50
630 100
—
—
198#
a
Unoptimized percentage yields. In the cases of low yield for final sulfated product, the reason was the generation of undersulfated by-products, which required tedious
purification by chromatography and/or SPE. Experimental details for the Ugi and sulfonation reactions have been reported previously.4
b
The average and SD of at least two measurements. #Data from a single measurement only. ND: no data available.
60 °C, 18 h (normal Ugi conditions: rt 18 h).
Sulfate (OSO3Na) in final product was introduced at this step as hydroxyl (OH).
Sulfate (OSO3Na) in final product was introduced at this step as benzyl ether (OBn).
c
d
e
f
Carboxylate (CO2Na) in final product was introduced at this step as methyl ester (CO2Me).

L. Liu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6190–6194
6193
(an allyl group is generally present at C4). The additional flexibility
of the Ugi-derived side chains would also appear to contribute sig-
nificantly to the improved affinity (see below).
Previous molecular docking studies on FGF-1 and FGF-2 with
small molecule HS-mimetics have shown that the predicted loca-
tions of bound sulfo groups display good overlap with the positions
observed for sulfo groups from co-crystallized heparin-derived oli-
gosaccharides.10c,15 Molecular docking calculations were thus per-
formed to examine the binding modes of compound 2Ca with FGF-
1 and FGF-2 (Fig. 3). Docking studies were undertaken with Mole-
gro Virtual Docker (MVD) using the MolDock SE algorithm.16 The
ligand search space was confined to an 11 Å sphere originating
from the ligand binding sites of FGF-1 (PDB code 2AXM) and
FGF-2 (PDB code 1BFB) based on the corresponding ligand in the
crystal structure. The location of the sulfate groups of the bound li-
gands in the crystal structures of these complexes were used as a
template to score higher for poses that resemble this binding
mode. Ligands and water molecules were omitted from the coordi-
nate files before docking. In the docking of the complex between
2Ca and FGF-1, the C6 sulfate group in the inhibitor forms hydro-
gen bonds with the nitrogen atom (3.0 Å) in the side-chain amide
Figure 2. Structures of the most potent compounds from each series. R = SO3Na.
was highly active with a negatively charged functional group
ranked next (40–283 lM for J, f, g, h, i). It is interesting to note
there are exceptions when an aromatic group or a negative charge
is delivered at less than ideal positions such as in 1Ib, 1Ic and 2Aa
or 1Ii (402–1100 lM). As the linker group between the sulfated D-
mannoside ring and the probing functional group is varied from
Series 1 and 2 (5–6 atoms vs 7–8 atoms), VEGF displays a large spa-
tial tolerance for aromatics, for example, similar activities (45–
109
lM) among 1Aa-1Ba-1Ca-1He and 2Ba-2Ca-2He). An extra
negative charge clearly improves the affinity, as it does in the
D
-
mannoside ring of Series 2. However, it is not always the case that
increased charge in the molecule results in higher affinity, with the
best lead out of either series bearing an aromatic motif (1He at
of Gln 127 and with the e-amino group of Lys 118 (3.1 Å). The lat-
ter nitrogen atom also forms a hydrogen bond (2.9 Å) with the C2
sulfate group. The other two oxygen atoms in the C2 sulfate form
hydrogen bonds with the protein; (i) between the backbone amide
of Lys 113 (2.7 Å) and (ii) between the nitrogen of the side-chain
amide of Asn 18 (2.8 Å). A third sulfate group (C4) forms a hydro-
gen bond to the backbone amide of Lys 128 (2.7 Å). The remainder
of the inhibitor forms van der Waals contacts with the side-chain
of Arg 122 and the aromatic ring forms p–cation interactions with
the side-chain guanidino group of this residue.
In the docking calculations for the complex between 2Ca and
FGF-2, the C2 sulfate group forms two hydrogen bonds (both
45.3
We have recently reported the synthesis of an alternative li-
brary of HS mimetics based on the -mannopyranose scaffold in
lM and 2Ca at 29 lM, Fig. 2).
D
which diversity was introduced at C6 via click chemistry (CuI-cat-
alyzed Huisgen azide–alkyne cycloaddition reaction) or a click-
Swern oxidation–Wittig reaction sequence.10f Members of the ear-
lier library in general had poorer affinities for FGF-1, VEGF and
especially FGF-2 compared with the current library, most likely
as a consequence of possessing one less sulfate group per molecule
Figure 3. Docking of compound 2Ca to FGF-1 (A, B) and FGF-2 (C, D). Both docking calculations show that the inhibitor can bind to the FGFs primarily through its sulfate
groups, forming hydrogen bonds and electrostatic interactions between the negatively charged sulfate oxygen atoms and the conserved residues (Lys 118/126, Gln 127/135,
Ala 129/137, Asn 18/28). The side chain of the inhibitor forms van der Waals interactions with both FGFs. The molecular surface of the protein (top) is shown in grey and
inhibitor surface is shown in transparent green.

6194
L. Liu et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6190–6194
3. (a) Murphy, P. V.; Bradley, H.; Tosin, M.; Pitt, N.; Fitzpatrick, G. M.; Glass, W. K.
J. Org. Chem. 2003, 68, 5692; (b) Bradley, H.; Fitzpatrick, G.; Glass, W. K.; Kunz,
H.; Murphy, P. V. Org. Lett. 2001, 3, 2629; (c) Park, W. K. C.; Auer, M.; Jaksche,
H.; Wong, C.-H. J. Am. Chem. Soc. 1996, 118, 10150; (d) Sutherlin, D. P.; Stark, T.
M.; Hughes, R.; Armstrong, R. W. J. Org. Chem. 1996, 61, 8350; (e) Lockhoff, O.;
Frappa, I. Comb. Chem. High Throughput Screening 2002, 5, 361; (f) Lockhoff, O.
Angew. Chem., Int. Ed. 1998, 37, 3436; (g) Bedjeguelal, K.; Bienayme, H.;
Dumoulin, A.; Poigny, S.; Schmitt, P.; Tam, E. Bioorg. Med. Chem. Lett. 2006, 16,
3998; (h) ChapmanTimothy, M.; DaviesIeuan, G.; Gu, B.; BlockTimothy, M.;
ScopesDavid, I. C.; HayPhilip, A.; CourtneyStephen, M.; McNeillLuke, A.;
SchofieldChristopher, J.; DavisBenjamin, G. J. Am. Chem. Soc. 2005, 127, 506;
(i) Chapman, T. M.; Davies, I. G.; Gu, B.; Block, T. M.; Scopes, D. I. C.; Hay, P. A.;
Courtney, S. M.; McNeill, L. A.; Schofield, C. J.; Davis, B. G. J. Am. Chem. Soc. 2005,
127, 506.
4. Liu, L.; Li, C. P.; Cochran, S.; Ferro, V. Bioorg. Med. Chem. Lett. 2004, 14, 2221.
5. Zhang, J.; Rivers, G.; Zhu, Y.; Jacobson, A.; Peyers, J.; Grundstrom, G.; Burch, P.;
Hussein, S.; Marolewski, A.; Herlihy, W.; Rusche, J. Bioorg. Med. Chem. 2001, 9,
825.
6. For reviews see (a) Bikfalvi, A. Bull. Cancer 2007, 94, F193; (b) Conrad, H. E.
Heparin-binding Proteins; Academic Press: San Diego, 1998; (c) Galzie, Z.;
Kinsella, A. R.; Smith, J. A. Biochem. Cell Biol. 1997, 75, 669.
7. For reviews see (a) Rusnati, M.; Presta, M. Curr. Pharm. Des. 2007, 13, 2025; (b)
Pluda, J. M. Semin. Oncol. 1997, 24, 203; (c) Kessler, T.; Fehrmann, F.; Bieker, R.;
Berdel, W. E.; Mesters, R. M. Curr. Drug Targets 2007, 8, 257; (d) Ellis, L. M.;
Hicklin, D. J. Nat. Rev. Cancer 2008, 8, 579.
8. (a) DiGabriele, A. D.; Lax, I.; Chen, D. I.; Svahn, C. M.; Jaye, M.; Schlessinger, J.;
Hendrickson, W. A. Nature 1998, 393, 812; (b) Faham, S.; Hileman, R. E.; Fromm,
J. R.; Linhardt, R. J.; Rees, D. C. Science 1996, 271, 1116.
9. (a) Pellegrini, L.; Burke, D. F.; von Delft, F.; Mulloy, B.; Blundell, T. L. Nature
2000, 407, 1029; (b) Schlessinger, J.; Plotnikov, A. N.; Ibrahimi, O. A.;
Eliseenkova, A. V.; Yeh, B. K.; Yayon, A.; Linhardt, R. J.; Mohammadi, M. Mol.
Cell 2000, 6, 743.
10. (a) Cochran, S.; Li, C.; Fairweather, J. K.; Kett, W. C.; Coombe, D. R.; Ferro, V. J.
Med. Chem. 2003, 46, 4601; (b) Johnstone, K. D.; Karoli, T.; Liu, L.; Dredge, K.;
Copeman, E.; Li, C. P.; Davis, K.; Hammond, E.; Bytheway, I.; Kostewicz, E.; Chiu,
F. C.; Shackleford, D. M.; Charman, S. A.; Charman, W. N.; Harenberg, J.; Gonda,
T. J.; Ferro, V. J. Med. Chem. 2010, 53, 1686; (c) Liu, L.; Bytheway, I.; Karoli, T.;
Fairweather, J. K.; Cochran, S.; Li, C.; Ferro, V. Bioorg. Med. Chem. Lett. 2008, 18,
344; (d) Ferro, V.; Dredge, K.; Liu, L.; Hammond, E.; Bytheway, I.; Li, C.;
Johnstone, K.; Karoli, T.; Davis, K.; Copeman, E.; Gautam, A. Semin. Thromb.
Hemost. 2007, 33, 557; (e) Freeman, C.; Liu, L. G.; Banwell, M. G.; Brown, K. J.;
Bezos, A.; Ferro, V.; Parish, C. R. J. Biol. Chem. 2005, 280, 8842; (f) Liu, L.; Li, C.;
Cochran, S.; Jimmink, S.; Ferro, V. ChemMedChem 2012, 7, 1267; (g) Ferro, V.;
Liu, L.; Johnstone, K. D.; Wimmer, N.; Karoli, T.; Handley, P.; Rowley, J.; Dredge,
K.; Li, C. P.; Hammond, E.; Davis, K.; Sarimaa, L.; Harenberg, J.; Bytheway, I. J.
Med. Chem. 2012, 55, 3804.
2.6 Å) with the nitrogen atoms in the backbone amide of residues
Ala 137 and Lys 136. The C2 and the C3 sulfates also interact with
the positively charged e-amino group of Lys 126 (2.7 Å), and the C3
sulfate group additionally forms a hydrogen bond with a nitrogen
atom (3.0 Å) in the side-chain amide of Asn 28. The side-chain gua-
nidino group of Arg 121 is wedged between the other two sulfate
groups (C4 and C6) forming ionic interactions (3.0–3.2 Å). The side
chain of 2Ca binds to FGF-2 through van der Waals interactions
with the side-chains of residues Lys 130 and Gln 135.
In conclusion, by harnessing the powerful Ugi four component
condensation, a library of HS mimetics was quickly assembled
and used to probe the binding environment of some angiogenic
HS-binding growth factors. The affinities measured for these
monosaccharide derivatives are close to those generally observed
for polysulfated di- to tetrasaccharides,10a,c especially those leads
bearing fewer charged groups. A clear trend for protein preference,
particularly an aromatic group for FGF-1 and VEGF, was observed.
These observations are generally in agreement with previous stud-
ies on modified heparin polysaccharides which showed that
increasing lipophilic modifications can improve affinity for VEGF
and FGF-1 but not FGF-2.17 Future studies should focus on optimiz-
ing the sulfation pattern around the monosaccharide motif as well
as investigating a combination of more than one hydrophobic
group to maximize non-ionic binding contributions with an expec-
tation to achieve higher affinity small molecule ligands.
Acknowledgements
We thank Drs. Tommy Karoli, Jon Fairweather, Ian Bytheway,
Robert Don and Edward Hammond for useful discussions.
Supplementary data
Supplementary data (copies of NMR spectra for intermediates
and final compounds. Experimental details and characterization
data for selected compounds) associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.bmcl.2012.08.001.
11. (a) Capila, I.; Linhardt, R. J. Angew. Chem., Int. Ed. 2002, 41, 390; (b) Harmer, N. J.
Biochem. Soc. Trans. 2006, 34, 442; (c) Biswas, C. Trends Glycosci. Glycotechnol.
1992, 4, 53.
12. Bernlind, C.; Oscarson, S.; Widmalm, G. Carbohydr. Res. 1994, 263, 173.
13. 1H and 13C NMR data for compound 9: 1H NMR (400 MHz, CDCl3, 7.27): 7.42-
7.29 (m, 15 H, 3 ꢁ Ph), 5.01, 4.61 (ABq, JAB = 11.2 Hz, PhCH2), 4.78, 4.72 (ABq,
JAB = 12.0, PhCH2), 4.77 (d, 1 H, J1,2 = 1.6, H1), 4.65, 4.63 (ABq, JAB = 11.6, PhCH2),
3.93 (dd, 1H, J3,4 = 9.2, J2,3 = 2.8, H3), 3.85 (dd, 1 H, J4,5 = 9.6, H4), 3.83 (dd, 1 H,
H2), 3.73 (ddd, 1H, J5,6a = 7.2, J5,6b = 2.0, H5), 3.69 (dd, 1 H, J6a,6b = 15.2, H6b),
3.55 (dd, 1 H, H6a), 3.37 (s, 3H, CH3O). 13C (100 MHz, CDCl3, 77.0): 157.57 (NC),
138.00, 137.94, 137.83, 128.47, 128.362, 128.355, 128.34, 127.98, 127.89,
127.72, 127.68, 127.65, 127.57, 98.99 (C1), 79.90, 75.17, 74.96, 74.15, 72.72,
71.89, 69.60, 54.93, 43.10.
References and notes
1. For recent reviews see (a) Dömling, A.; Wang, W.; Wang, K. Chem. Rev. 2012,
112, 3083; (b) Colombo, M.; Peretto, I. Drug Discovery Today 2008, 13, 677; (c)
Ganem, B. Acc. Chem. Res. 2009, 42, 463; (d) Banfi, L.; Guanti, G.; Riva, R.; Basso,
A. Curr. Opin. Drug Discov. Devel. 2007, 10, 704.
2. (a) Akritopoulou-Zanze, I. Curr. Opin. Chem. Biol. 2008, 12, 324; (b) Hulme, C.;
Dietrich, J. Mol. Divers. 2009, 13, 195; (c) Dömling, A. Chem. Rev. 2006, 106, 17;
(d) Dömling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000, 39, 3168; (e) Dömling, A.
Curr. Opin. Chem. Biol. 2002, 6, 306; (f) Hulme, C.; Gore, V. Curr. Med. Chem.
2003, 10, 51; (g) Weber, L. Curr. Med. Chem. 2002, 9, 2085.
14. Cochran, S.; Li, C. P.; Ferro, V. Glycoconjugate J. 2009, 26, 577.
15. Bytheway, I.; Cochran, S. J. Med. Chem. 2004, 47, 1683.
16. Thomsen, R.; Christensen, M. H. J. Med. Chem. 2006, 49, 3315.
17. Huang, L.; Fernandez, C.; Kerns, R. J. Bioorg. Med. Chem. Lett. 2007, 17, 419.