Structure-activity studies on a library of potent calix[4]arene-based PDGF antagonists that inhibit PDGF-stimulated PDGFR tyrosine phosphorylation

Article abstract of DOI:10.1039/b515483a

Platelet-derived growth factor (PDGF) and its receptor PDGFR are required for tumor growth and angiogenesis, so disruption of the PDGF-PDGFR interaction should lead to starvation of tumors and reduction of tumor growth. Potent PDGF antagonists have been discovered through the synthesis of a series of calix[4]arene-based compounds that are designed to bind to the three-loop region of PDGF. The effect of lower-rim alkylation, linker and number of interacting head groups on the calix[4]arene scaffold on PDGF affinity and cellular activity has been investigated. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b515483a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Structure–activity studies on a library of potent calix[4]arene-based PDGF
antagonists that inhibit PDGF-stimulated PDGFR tyrosine phosphorylation
Huchen Zhou,a De-an Wang,b,c Laura Baldini,a Eileen Ennis,a Rishi Jain,a Adam Carie,b,c
Sa¨ıd M. Sebti*b,c and Andrew D. Hamilton*a
Received (in Pittsburgh, PA) 1st November 2005, Accepted 21st March 2006
First published as an Advance Article on the web 10th May 2006
DOI: 10.1039/b515483a
Platelet-derived growth factor (PDGF) and its receptor PDGFR are required for tumor growth and
angiogenesis, so disruption of the PDGF–PDGFR interaction should lead to starvation of tumors and
reduction of tumor growth. Potent PDGF antagonists have been discovered through the synthesis of a
series of calix[4]arene-based compounds that are designed to bind to the three-loop region of
PDGF. The effect of lower-rim alkylation, linker and number of interacting head groups on the
calix[4]arene scaffold on PDGF affinity and cellular activity has been investigated.
our strategies of synthetic protein surface antagonists25–27 to
Introduction
the design of PDGF antagonists. Synthetic PDGF antagonists
Platelet-derived growth factor (PDGF) is a mitogen that plays
have the potential advantages of stability toward protease and
a key role in cell proliferation, angiogenesis, wound healing,
nuclease cleavage, smaller size and the convenience of systematic
chemotaxis, and inhibition of apoptosis.1–3 In malignant diseases
modifications. Over the past several years we have developed a
calixarene-based system28,29 that represents the only family of non-
natural antagonists of PDGF.
The design of synthetic molecules that can bind to a protein
surface and disrupt physiologically important protein–protein
interactions remains a major challenge.30–34 In contrast to the
binding of substrates to well-defined enzyme active site cavities,
protein–protein interfaces are much less defined.35–37 They involve
large and lightly featured domains that in many cases cover an
such as cancer, PDGF or its receptors are often found to be
overexpressed and required for tumor growth.4–7 PDGF has also
been shown to induce high levels of vascular endothelial growth
factor (VEGF), an angiogenesis inducer that is required for the
initiation of the formation of new blood vessels.8,9 To elicit its
biological activity, PDGF binds to PDGFR, a receptor tyrosine
kinase (RTK).10 As a result of binding, the receptor dimerizes and
undergoes autophosphorylation that triggers the recruitment of a
series of signaling proteins and the activation of the corresponding
signal transduction pathways.11
Based on the understanding of this biological process, dis-
ruption of the PDGF–PDGFR interaction should impede the
critical angiogenesis process thus leading to starvation of tumors
and reduction of tumor growth.12–16 To this end, there have been
several studies on antagonists against PDGFR or PDGF. Known
PDGFR antagonists include antibodies against PDGFR,17,18
peptides corresponding to the PDGF binding area19,20 and small
molecule inhibitors of receptor dimerization and tyrosine kinase
activity.12,13 Gleevec,21 the Abl tyrosine kinase inhibitor, which
also inhibits PDGFR tyrosine kinase, has been validated in
clinical trials and approved recently by the FDA as a new
cancer therapeutic. In contrast, only a few PDGF antagonists
are known, such as anti-PDGF antibodies,22 soluble extracellular
domains of PDGFR23 and PDGF-binding DNA aptamers,24 and
these have had only marginal success. We have been expanding
2
38
˚
area of 1200–2000 A . The X-ray crystal structure (Fig. 1)
of the homodimer of PDGF-B (PDGF-BB) shows that it is a
dimer of polypeptide chains that folds into two antiparallel pairs
of b-strands and contains five intra- and inter-chain disulfide
linkages. The three loop regions I–III are exposed to solvent,
loop I being highly disordered and thus missing in the solved
structure.
aDepartment of Chemistry, Yale University, New Haven, CT 06520, USA.
E-mail: Andrew.Hamilton@yale.edu; Fax: +1 203 432 3221; Tel: +1 203
432 5570
bDrug Discovery Program, H. Lee Moffitt Cancer Center & Research
Institute, Tampa, FL, 33612, USA. E-mail: Sebti@moffitt.usf.edu; Fax: +1
813 979 6748; Tel: +1 813 979 6734
cDepartments of Interdisciplinary Oncology and Biochemistry & Molecular
Biology, H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL,
33612, USA
Fig. 1 X-Ray crystal structure of recombinant human PDGF-BB dimer
and sequence of PDGF-B monomeric chain. Adapted from ref. 38.
2376 | Org. Biomol. Chem., 2006, 4, 2376–2386
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Fig. 2 SAR studies involved systematic modifications of the calix[4]arene scaffold, shown as the shaded domains on the structure of GFB 204. The
structure of GFB111 is also shown.
We have previously reported a PDGF antagonist that has 4
Results and discussion
peptidic loops attached to calix[4]arene, GFB111 (Fig. 2).28,29 This
Design and syntheses of PDGF antagonists
molecule was shown to disrupt the PDGF–PDGFR interaction
and have anti-angiogenic and anti-proliferative activity against
human tumors. To overcome the drawbacks of GFB111, its
large size, difficulty of synthesis and aggregation in water, we
have recently discovered a second-generation PDGF antagonist
GFB204 (Fig. 2).39 Simple acyclic isophthalate groups function-
alized with an acidic and a hydrophobic group replaced the
peptide loops, thus significantly reducing the molecular weight
without diminishing activity. GFB204 was found to block PDGF
receptor phosphorylation and Erk1/2 activation with an IC50 of
0.19 lM. Further evaluation of its antitumor activity showed
that it inhibited tumor growth in xenografted nude mice (Fig. 3).
GFB204 was also shown to inhibit angiogenesis by a mechanism
involving PDGF as well as VEGF.
Our approach is based on the calix[4]arene scaffold that bears
different functionalities complementary to the three binding loops
of PDGF-BB. The three loops critically involved in PDGFR
binding are composed of hydrophobic (loop I), hydrophobic
and cationic (loop II) and cationic (loop III) residues. Epitope
mapping and mutagenesis studies40–45 have shown that of the
three loops, the residues in loop I are most important in the
binding to PDGFR. In our previous two generations of calixarene
derivatives, the combination of hydrophobic and acidic groups28
turned out to be essential to achieve optimal binding affinity. In
this study, we report the synthesis of a third-generation library in
order to evaluate more comprehensively the key structure–activity
relationships.
Calix[4]arenes with different numbers of isophthalate groups
were synthesized by reacting the acyl chloride form of the
tetraacid derivative of calix[4]arene (1) with monobenzyl mono-
t-butyl 5-aminoisophthalate (2) in the presence of methanol,
followed by preparative thin layer chromatography to obtain the
products with one to four isophthalate groups, with the remaining
carboxylates blocked as methyl esters (3b–6b). The necessity of
the aromatic isophthalate frame was assessed with compound 7b,
which contains in its place a dialkyl amide, and was prepared
by coupling benzyl ((t-butoxycarbonylmethyl)amino)acetate to
the tetraacid chloride of calix[4]arene, followed by either benzyl
ester deprotection (H2, Pd–C) or t-butyl ester removal (TFA)
(Fig. 4).
Fig. 3 Antitumor activity of GFB204 in the nude mouse tumor xenograft
model. ᭜: control animals that received a vehicle. ꢀ: treated animals that
were injected with GFB204 (10 mg kg−1) daily.
To investigate the effect on affinity of different linker groups
between the isophthalate and calix[4]arene components, com-
pounds 8b–12b were prepared. Compound 8b came from the
coupling of tetraacid chloride calix[4]arene to 1-benzyl 3-t-
butyl 5-(2-aminoacetylamino)isophthalate. The preparation of 9b
involved the treatment of tetraamino calix[4]arene with triphos-
gene to generate the tetra(carbamoyl chloride) that subsequently
reacted with aniline 2 to give the urea derivative 9b. The
tetra(diazo)calix[4]arene 10b was obtained by reacting the parent
Here, we report an investigation of the effect of a wide range
of structure variations of GFB204 on PDGF affinity and cellular
activity including different numbers of isophthalate groups, varia-
tion of the linkage between isophthalate and calix[4]arene scaffold
and substituents on the lower rim hydroxyls (Fig. 2). We have
derived a structure–activity analysis based on the direct binding of
PDGF to these antagonists as well as cell-based assays, providing
valuable insights into PDGF recognition and directing us to a
third-generation PDGF antagonist.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2376–2386 | 2377
©

View Article Online
Fig. 4 (a)–(f): Synthetic routes for PDGF antagonists 3b–6b, 7b, 10b–12b and 15b. (g)–(i): Structures of PDGF antagonists 8b, 13b and 14b.
calix[4]arene with the diazonium salt derived from 2 by treatment
with sodium nitrite under acidic conditions. During this reaction,
the tris(diazo)calix[4]arene 12b was also obtained after recrystal-
lization. As a close analog of the amide linker, a sulfonamide group
was introduced into compound 11b by coupling the corresponding
tetrasulfonyl chloride calix[4]arene with aniline 2.
The role of alkylation at the lower rim was exemplified by
compounds 10b, 13b, 14b and 15b. Compounds 13b and 14b were
synthesized similarly to GFB204 except that the lower rim was
alkylated with either 2-methoxyethyl bromide or methyl iodide.
In the preparation of 15b, the lower rim hydroxyls were first
protected by benzyl groups that were later removed along with
the isophthalate benzyl ester by hydrogenation.
Binding affinity of PDGF-BB antagonists
There is a single tryptophan residue in each PDGF-B monomer
near binding loop I, which provides a fluorescent probe for
detecting antagonist binding to this region. The dissociation
constants were derived from the change of fluorescent intensity of
tryptophan upon binding by the antagonists. The titration curves
were fitted using Scientist software to a 1 : 1 binding equation, as
described in the Experimental section.
First, an array of antagonists with one to four isophthalate
groups was investigated to understand the optimal number of in-
teracting moieties needed for efficient binding. A gradual decrease
in binding affinity was observed when the number of isophthalate
2378 | Org. Biomol. Chem., 2006, 4, 2376–2386
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 1 The dissociation constants of PDGF-BB with antagonists and
the corresponding IC50 values for the inhibition of PDGFR auto-
phosphorylation
groups was reduced from four to one, as in compounds 6b–3b.
The tetra(isophthalate) calix[4]arene 6b has the highest affinity
(Kd = 9.3 nM) in this series, while the tris(isophthalate) 5b has a
slightly weakened affinity (Kd = 21.1 nM). The affinity decreased
further to Kd = 137 nM when, as in compound 3b, there is
only a single isophthalate arm remaining (Fig. 5). The affinity
of tetra-substituted diazocalix[4]arene 10b (Kd = 22.6 nM) toward
PDGF-BB is only marginally larger than that of tri-substituted 12b
(Kd = 23.8 nM). These observations suggest that the three-armed
calix[4]arenes may be comparably active in vivo since, despite their
slightly weakened affinity, they may have the advantages of a
significantly lower molecular weight compared to their four-armed
analogs.
Compound names
Kd/nM
IC50/lM
GFB204
17.3
68.0
137
39.4
21.1
9.26
—
0.19 0.06
10
>10
1 (GFB248)
3b (GFB224)
4b (GFB223)
5b (GFB239)
6b (GFB210)
7b (GFB221)
8b (GFB222)
9b (GFB227)
10b (GFB229)
11b (GFB234)
12b (GFB247)
13b (GFB237)
14b (GFB210)
15b (GFB238)
>10
0.43 0.35
0.35 0.31
2.32 1.81
0.6 0.45
>10
0.11 0.04
>10
0.48 0.1
0.58 0.13
0.35 0.31
>10
—
—
22.6
13.7
23.8
46.9
31.3
10.0
Fig. 5 Fluorescence titrations of PDGF-BB with compounds 3b–6b and
1. From top to bottom: Kd/nM: 3b (137), 1 (68), 4b (39.4), 5b (21.1), 6b
(9.26).
Our binding target is the three loops of PDGF-BB that contain
both hydrophobic and cationic groups and are critical for its
binding to PDGFR. The most efficient binding is seen with
those synthetic molecules containing several isophthalate residues
bearing both hydrophobic and anionic groups. These binding
domains provide a number of potential points of interaction,
which should lead to stronger affinity. However, this effect would
likely be saturated at a certain point due to limited surface
contact and increased steric congestion. Based on our evaluation
of Kd and IC50 values of the above compounds, the three-armed
calix[4]arenes such as 5b and 12b appear to be comparably effective
PDGF-BB antagonists (Table 1 and Fig. 6).
Second, the effect of the lower rim alkyl groups on the binding
affinity was investigated. In our initial design, n-butyl groups at the
lower rim were introduced to constrain the calix[4]arene scaffold
into a cone conformation in which the phenyls are unable to invert
through the cavity of the ring.46 The butylated compound GFB204
exists in a well-defined cone conformation, as indicated by the pair
of doublets from the bridging methylene protons in the 1H NMR
spectrum. When the n-butyl group was replaced by a methoxyethyl
group in compound 13b, the cone conformation was retained while
the binding affinity Kd decreased slightly from 17.3 nM (GFB204)
to 46.9 nM (13b).
Fig. 6 GFB204, 5b and 10b inhibit PDGF-BB stimulation of PDGFR
tyrosine phosphorylation, Erk1/Erk2 and Akt phosphorylation. NIH 3T3
cells were treated with increasing concentrations of GFB204, 5b or 10b
for 5 min prior to stimulation with PDGF-BB (10 ng mL−1) for 10 min.
The cells were then lysed and processed for SDS/PAGE Western blotting
with an antibody specific for phosphotyrosine for PDGFR tyrosine
phosphorylation, phospho-Erk1/2 or phospho-Akt. Actin was blotted
as a control.
partial cone intermediate, only the cone conformation is observed
by NMR at room temperature. We rationalized that tetra-hydroxy
calix[4]arene derivatives should also give satisfactory affinity to
PDGF-BB since the prevalent conformation is cone. Indeed, the
tetra-hydroxy calix[4]arene 15b binds to PDGF-BB with a high
affinity of Kd = 10 nM. Therefore, the alkylation of the lower rim
is not required for the calix[4]arene to pre-organize into a favorable
cone shape that is complementary to the protruding loop region of
PDGF-BB. Removal of four lower rim n-butyl groups significantly
reduced the molecular weight of the molecules, which is desired
to improve solubility and bioavailability. The other tetra-hydroxy
compound 10b also showed a high affinity (Kd = 22.6 nM) as did
the tetramethyl compound 14b (Kd = 31.3 nM), further suggesting
that a low barrier to conformational interconversion also gives an
advantage for potency.
Extensive NMR and computational studies have shown that
a calix[4]arene with four free hydroxyl groups at the lower rim
is stabilized in its cone conformation by a circular array of four
hydrogen bonds.46 Although the cone is in equilibrium with the
Third, the role of the linkage between isophthalate and cal-
ixarene subunits was studied. The affinity was retained when the
initial amide linker was substituted by sulfonamide (11b) or diazo
(10b). Compared to the partial double bond characteristics of
ꢀ
inverted cone conformation (DG= = 15 kcal mol−1) through a
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2376–2386 | 2379
©

View Article Online
an amide, the sulfonamide linker should have more rotational
freedom as a single bond, while the diazo bond should be more
constrained. The comparable Kd values of the three compounds
(GFB204: 17.3 nM; 11b: 13.7 nM; 10b: 22.6 nM) suggested that
different degrees of rigidity in the linker can be tolerated without
significant loss of affinity toward PDGF-BB. This is advantageous,
as it provides a wide choice of linkers that could be further selected
for optimal in vivo properties.
The difference of the chemical shift (Dd/ppm) of the pair of
bridging methylene protons is a marker of the planarity of the
cone conformation in calix[4]arenes.46 The smaller the Dd value the
more planar is the cone, reaching zero ppm when it is completely
planar. A survey of the PDGF-BB antagonists reported in this
paper showed a wide range of Dd values (0.70–1.32 ppm). This
observation suggested that although the resting states of the
calix[4]arene scaffolds change significantly due to their varying
planarity, upon binding to PDGF-BB they may adjust to an
optimal conformation without any major energy barrier due to
the conformational flexibility of the calix[4]arene scaffold.
competitive cellular activity. This observation has led us to design
and synthesize new antagonists based on three-folded symmetry.47
The linker between the isophthalate and calix[4]arene scaffolds
had an insignificant effect on the affinity, but notably influenced
the cellular efficacy. Compounds 10b, GFB204 and 8b (with
diazo, amide and glycine linkers, respectively) all have satisfactory
activity in cell culture. In contrast, compounds 11b and 9b with
sulfonamide and urea linkers gave very low activity. The alkylation
at the lower rim hydroxyls was found to be unnecessary because the
four OH groups form an array of hydrogen bonds to stabilize the
calix[4]arene scaffold into a cone conformation that is potentially
well pre-organized to interact with the loop regions of PDGF-
BB. To probe the structure of the complex, we have used the
ligand–protein docking program GOLD which searches for the
optimal docking of a flexible ligand to a rigid protein framework.
Using the GoldScore fitness function, docking of the diazo-
calixarene (10b) to PDGF-BB showed a cone-shaped calixarene
binding to a large surface area of the loop region of PDGF-BB
(Fig. 7).
PDGF-BB antagonists block PDGF-BB induced PDGF receptor
tyrosine phosphorylation and Erk1/2 activation
The calix[4]arene-based antagonists were shown to block the
subsequent autophosphorylation of PDGFR and activation of
downstream kinases Erk1 and Erk2 in NIH 3T3 cells, as described
in the Experimental section. The IC50 values provided a direct
evaluation of the cellular activity of our PDGF-BB antagonists
(Table 1 and Fig. 6). Three of the compounds, GFB204, 10b
and 5b, showed potent IC50 values: 0.19 lM, 0.11 lM and
0.43 lM respectively. These three compounds are also among those
with the highest PDGF-BB affinity from the fluorescent binding
experiments, with Kd values of 17.3 nM, 22.6 nM and 21.1 nM,
respectively.
To our surprise, compound 15b had a very low activity (IC50
>10 lM), although its affinity for PDGF-BB was high (Kd
=
10.0 nM). This may be due to non-specific interaction of the
calix[4]arene derivatives within the cellular assay. The cell-based
assay provided a useful tool to screen the preliminary specificity
and physiological stability of the initial leads. For example,
compound 11b showed excellent binding affinity (Kd = 13.7 nM)
but only gave poor cellular activity (IC50 >10 lM) (Table 1).
The isophthalate group appears to be necessary for high cellular
activity, since compound 7b, which lacks the isophalate aromatic
ring, showed much reduced activity. This is probably due to the
decreased hydrophobicity when the four aromatic isophthalate
groups were removed.
Fig. 7 The docking between PDGF-BB and diazo-calixarene (10).
Docking was carried out using GOLD (GoldScore fitness function) and
the graph was generated using WebLab Viewer Pro.
These observations are now guiding us in new strategies for the
design of molecules with reduced molecular weight and elevated
in vivo activity.
Conclusion
To understand the structure–activity relationship of calix[4]arene
derivatives towards PDGF-BB binding, a library of compounds
that address different aspects of the structural variation of the par-
ent GFB204 was prepared and their dissociation constants towards
PDGF and their ability to inhibit PDGF-stimulated PDGFR tyro-
sine phosphorylation were obtained through fluorescent titrations
and cell-based assays, respectively. Three isophthalate arms turned
out to be sufficient to obtain good binding affinity and also to give
Experimental
Materials and instruments
PDGF-BB was purchased from Cell Sciences. All solvents were
purchased from Mallinckrodt or Aldrich and all reagents were
2380 | Org. Biomol. Chem., 2006, 4, 2376–2386
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
purchased from Aldrich unless otherwise stated and used without
further purification. All moisture-sensitive reactions were carried
out under nitrogen atmosphere. Purification by column chro-
matography was carried out using silica gel (230–400 mesh size).
Analytical thin layer chromatography (TLC) was conducted using
Baker 0.25 mm silica gel pre-coated glass plates with fluorescent
indicator active at UV245. Preparative TLC was conducted using
Analtech 1000 mm silica gel pre-coated plates with fluorescent
indicator active at UV245. 1H NMR and 13C NMR spectra were
acquired on either Bruker DPX 400 or DPX 500 series spectrom-
eters. Chemical shifts are expressed as parts per million using
solvent as internal standard. Preparative HPLC was performed
on a Waters 600E controller in conjunction with Waters 490E
multi-wavelength UV detector. Analytical HPLC was performed
on a Rainin HP controller with a Rainin UV detector, both
attached to a Dell Optiplex PC running Varian Star Workstation
software. Nominal mass spectrometry data was obtained by Dr
Walter McMurray at the W. M. Keck Foundation Biotechnology
Resource Laboratory of Yale University. High resolution mass
spectrometry data was obtained by the Washington University
Resource for Biomedical and Bio-organic Mass Spectrometry.
Fluorescence titration experiments were carried out on a Photon
Technology International spectrofluorometer equipped with a
LPS-220B lamp power supply, a MD-5020 motor driver and an
814 photomultiplier detection system.
were determined by measuring the length (l) and the width (w)
and calculating the volume (V = lw2/2). The statistical significance
between control and treated animals was evaluated using Student’s
t-test.
Inhibition of growth factor-dependent receptor tyrosine
phosphorylation by GFBs
Starved NIH 3T3 cells were pretreated with GFBs for 5 min
before stimulation with PDGF-BB (10 ng mL−1) for 10 min. The
cells were then harvested and lysed, and proteins from the lysates
were separated by SDS-PAGE and transferred to nitrocellulose.
Membranes then were blotted with anti-phospho-tyrosine anti-
body (4G10, Upstate Biotechnology, Lake Placid, NY, USA) for
activated PDGFR.28 Phosphotyrosine PDGFRs were quantified
using a Bio-Rad Model GS-700 Imaging Densitometer (Bio-Rad
Laboratories Inc., Hercules, CA, USA).
Inhibition of growth factor-mediated stimulation of
phosphorylation of Erk1/2 and Akt by GFBs
Starved NIH 3T3 cells were pretreated with the indicated concen-
tration of GFBs for 5 min, before 10 min stimulation with PDGF-
BB (10 ng mL−1). Cell lysates were run on SDS-PAGE gels, and
then transferred to nitrocellulose and Western blotted with anti-
phosphorylated Erk1/Erk2 (Cell Signaling Technologies) and
anti-phosphorylated Akt as described previously by us.28
Fluorescence tritration
Synthesis
Lyophilized PDGF-BB (10 lg) was reconstituted by addition
of deionized water (100 lL). This stock solution was diluted to
5 mM phosphate buffer pH 7.4 to reach a final concentration of
160 nM. To a quartz cuvette that contained 500 lL of 160 nM
PDGF-BB solution, one of the antagonists was added in 32 lM ×
1 lL aliquots and incubated at room temperature for 3 min before
measurement. The excitation wavelength was 275 nm and the
fluorescence of tryptophan at 340 nm was monitored after each
addition. A blank buffer was titrated with the same antagonist in
the same manner to get a background to be subtracted before
curve fitting. The titration curve was fitted to an equimolar
binding model using the Scientist program. The equations are
F = (F0/(1.6 × 10−7))P + f 2KPL, P = 1.6 × 10−7/(1 + KL)
and conc. = (L + KPL). The initial values and constraints of
parameters are f 2 = 0, 0 < L < conc., 0 < P < 1.6 × 10−7 and
K = 1.0 × 106. F: fluorescence intensity at any titration point;
F0: fluorescence intensity at the beginning; P: unbound PDGF;
f 2: unit fluorescence intensity of bound protein; K: association
constant; L: unbound antagonist.
General procedure for removal of benzyl group. For example,
the fully protected compound 5a (5 mg, 0.0028 mmol) and
palladium on carbon (10%, 0.5 mg) in methanol–ethyl acetate
(2 mL, 1 : 1) was stirred under 1 atm H2 for 8 h. The catalyst
was filtered off and the filtrate was concentrated in vacuo to yield
the deprotected product 5b (3.5 mg, 83%) as a white solid. 5b:
1H NMR (400 MHz, CD3OD): d = 10.16 (s, 2H, NH), 9.77
(s, 1H, NH), 8.27–8.59 (m, 9H, Ar), 7.76 (s, 2H, Ar calix),
7.71 (s, 2H, Ar calix), 7.20 (s, 2H, Ar calix), 7.14 (s, 2H, Ar
calix), 4.59 (t, J = 13.2 Hz, 4H, Hax of ArCH2Ar), 4.19 (t,
J = 6.5 Hz, 4H, OCH2CH2CH2CH3), 3.93 (t, J = 6.5 Hz,
4H, OCH2CH2CH2CH3), 3.53 (s, 3H, COCH3), 3.43 (t, J =
13.2 Hz, 4H, Heq of ArCH2Ar), 1.97 (m, 8H, OCH2CH2CH2CH3),
1.63 (s, 18H, C(CH3)3), 1.56 (s, 9H, C(CH3)3), 1.44 (m, 8H,
OCH2CH2CH2CH3), 1.06 (m, 12H, OCH2CH2CH2CH3). HR
ESI-MS m/z: calcd [M + 2Na]2+ 770.8205, found 770.8196.
General procedure for removal of t-butyl group. For example,
the fully protected compound 8a (15 mg, 0.007 mmol), triethyl-
silane (5 lL, 0.035 mmol), and trifluoroacetic acid (150 lL,
2.1 mmol) in 3 mL of CH2Cl2 was stirred overnight. The
deprotected product 8b (11 mg, 80%) as a white solid was obtained
with satisfactory purity after evaporation of the solvent followed
by addition of diethyl ether and evaporation to dryness. 8b:
1H NMR (400 MHz, CD3OD): d = 8.32 (s, 8H, Ar), 8.25 (s,
4H, Ar), 7.38–7.24 (m, 28H, Ph and Ar calix), 5.28 (s, 8H,
CH2Ph), 4.50 (d, J = 13.3 Hz, 4H, Hax of ArCH2Ar), 3.98
(br s, 16H, COCH2 and OCH2CH2CH2CH3), 3.30 (d, J =
13.3 Hz, 4H, Heq of ArCH2Ar), 1.93 (m, 8H, OCH2CH2CH2CH3),
1.51 (m, 8H, OCH2CH2CH2CH3), 1.03 (t, J = 7.3 Hz, 12H,
Antitumor activity in the nude mouse tumor xenograft model
Nude mice (Charles River, Wilmington, MA) were maintained
in accordance with the Institutional Animal Care and Use Com-
mittee (IACUC) procedures and guidelines. A-549 cells (ATCC)
were harvested and resuspended in PBS. A-549 cells were injected
subcutaneously (s.c.) into the right and left flanks (10 × 106 cells
per flank) of 8 week old female nude mice. When tumors reached
about 200 mm3, animals were dosed intraperitoneally (i.p.) with
0.1 mL once daily. Control animals received a vehicle, whereas
treated animals were injected with GFB204. The tumor volumes
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2376–2386 | 2381
©

View Article Online
OCH2CH2CH2CH3). 13C NMR (125 MHz, CDCl3–CD3OD, 1 :
1): d = 168.3 (s, CO), 167.9 (s, CO), 167.1 (s, CO), 165.3 (s, CO),
159.3 (s, Ar), 138.4 (s, Ar), 135.3 (s, Ph), 134.6 (s, Ar), 131.3 (s,
Ar), 130.5 (s, Ar), 128.1 (d, Ph), 127.8 (d, Ph), 127.7 (d, Ph),
127.5 (d, Ar), 126.9 (s, Ar), 125.7 (d, Ar), 124.7 (d, Ar), 124.3
(d, Ar), 74.8 (t, OCH2CH2CH2CH3), 66.7 (t, OCH2Ph), 43.7 (t,
COCH2), 31.9 (t, OCH2CH2CH2CH3), 30.5 (t, ArCH2Ar), 18.9 (t,
OCH2CH2CH2CH3), 13.4 (q, OCH2CH2CH2CH3). HR ESI-MS
m/z: calcd [M − H]− 1643.6650, found 1643.7494.
Monobenzyl mono-t-butyl 5-aminoisophthalate (2). To a solu-
tion of monomethyl 5-nitroisophthalate in 40 mL of dry methylene
chloride (3 g, 13 mmol) was added oxalyl chloride (5.1 g, 40 mmol)
and 100 lL of DMF, and then the solution was stirred for 3 h.
The solution was evaporated in vacuo and added to t-BuOH (19 g,
260 mmol), DMAP (1.63 g, 13 mmol) and DIEA (17 g, 130 mmol)
then stirred overnight. The solution was evaporated, and dissolved
in 50 mL of methylene chloride and washed with 1 N NaOH (3 ×
50 mL) and 1 N HCl (3 × 50 mL). The organic extracts were
concentrated and dried in vacuo to afford monomethyl mono-t-
butyl 5-nitroisophthalate (2.5 g, 68%) as a brown solid. 1H NMR
(500 MHz, CDCl3): d = 8.97 (s, 1H), 8.93 (s, 1H), 8.89 (s, 1H),
4.00 (s, 3H), 1.65 (s, 9H). 13C NMR (125 MHz, CDCl3): d = 164.1,
162.4, 148.1, 135.4, 134.1, 131.9, 127.7, 127.4, 83.0, 52.7, 27.8.
FAB-MS m/z: calcd. [M + H]+ 282.1, found 282.4.
To the above product (2.4 g, 8.5 mmol) was added 1 N LiOH
(15 mL), followed immediately by THF (15 mL). After stirring
vigorously for 2 h, the solution was poured into methylene chloride
(60 mL), and 1 N HCl (60 mL) was added. The organic extracts
were combined and evaporated in vacuo to afford mono-t-butyl 5-
nitroisophthalate (2.2 g, 96%) as a light yellow powder. 1H NMR
(500 MHz, DMSO-d6): d = 8.78 (s, 1H), 8.72–8.70 (m, 2H), 1.60
(s, 9H). 13C NMR (125 MHz, DMSO-d6): d = 165.2, 163.0, 148.5,
135.0, 134.8, 133.5, 127.7, 127.0, 83.1, 28.0. FAB-MS m/z: calcd.
[M + H]+ 268.1 found 268.2.
The obtained powder (2.1 g, 7.8 mmol) was taken up in MeOH
(200 mL) and added to a suspension of 10% Pd/C (200 mg) in
H2O (5 mL), and stirred for 3 h under 1 atm of H2. The catalyst
was filtered off and the filtrate evaporated in vacuo to provide
mono-t-butyl 5-aminoisophthalate (1.74 g, 94%) as a white solid.
1H NMR (500 MHz, CD3OD): d = 7.88 (s, 1H), 7.53 (s, 1H), 7.48
(s, 1H), 1.61 (s, 9H). 13C NMR (125 MHz, DMSO-d6): d = 167.18,
149.10, 131.72, 118.13, 117.61, 117.37, 116.97, 80.52, 27.7.
This product (2 g, 8.4 mmol) was taken up in DMF (50 mL) and
cooled to 0 ◦C, with stirring. Then Cs2CO3 (2.7 g, 8.4 mmol) and
benzyl bromide (1.44 g, 8.4 mmol) were added, and the solution
stirred for 5 h at RT. After extraction with methylene chloride,
the organic layer was washed with H2O and the combined organic
extract was evaporated in vacuo. The residue was chromatographed
(hexanes–EtOAc, 3 : 1) to provide monobenzyl mono-t-butyl 5-
aminoisophthalate (2) (1.8 g, 66%) as a white powder. 2): 1H NMR
(500 MHz, CDCl3): d = 8.10 (s, 1H), 7.59 (s, 1H), 7.55 (s, 1H),
7.44–7.32 (m, 5H), 5.35 (s, 2H), 1.58 (s, 9H). 13C NMR (125 MHz,
CDCl3): d = 166.1, 165.1, 145.3, 136.1, 133.7, 131.6, 128.8, 128.5,
128.4, 121.9, 120.9, 120.4, 81.8, 67.1, 28.4. FAB-MS m/z: calcd.
[M + H]+ 328.1, found 328.3.
chloride (5 mL) was added oxalyl chloride (0.65 mL, 7.2 mmol)
and 1 drop of dry DMF. After stirring overnight, the reaction
mixture was evaporated in vacuo to give the acyl chloride. A
solution of 2 (235 mg, 0.72 mmol) and dry pyridine (64 lL,
0.72 mmol) in dry methylene chloride (7 mL) were added and
stirred overnight. After evaporation in vacuo, the residue was
purified by column chromatography (hexanes–acetone, 4 : 1) to
afford the fully protected precursor of GFB204 (168 mg, 67%).
The t-butyl groups were removed as described previously to obtain
GFB204 as a white solid. GFB204: 1H NMR (400 MHz, acetone-
d6): d = 9.40 (s, 4H, NH), 8.49 (s, 4H, Ar), 8.47 (s, 4H, Ar), 8.26
(s, 4H, Ar), 7.56 (s, 8H, Ar calix), 7.47–7.28 (m, 20H, Ph), 5.32 (s,
8H, CH2Ph), 4.63 (d, J = 13.6 Hz, 4H, Hax of ArCH2Ar), 4.07
(t, J = 7.2 Hz, 8H, OCH2CH2CH2CH3), 3.45 (d, J = 13.6 Hz,
4H, Heq of ArCH2Ar), 1.98 (m, 8H, OCH2CH2CH2CH3),
1.55 (m, 8H, OCH2CH2CH2CH3), 1.04 (t, J = 9.0 Hz, 12H,
OCH2CH2CH2CH3). 13C NMR (400 MHz, acetone-d6): d =
166.73 (CO), 165.83 (CO), 165.73 (CO), 160.74 (4◦, Ar), 140.78
(4◦, Ar), 137.11 (4◦, Ar), 135.86 (4◦, Ar), 132.01 (4◦, Ar),
131.54 (4◦, Ar), 129.34 (3◦, Ar), 129.14 (3◦, Ar), 129.09 (3◦, Ar),
129.04 (4◦, Ar), 128.97 (3◦, Ar), 126.07 (3◦, Ar), 125.93 (3◦, Ar),
125.51 (3◦, Ar), 75.95 (PhCH2), 67.45 (OCH2CH2CH2CH3),
33.05
(OCH2CH2CH2CH3),
31.85
(ArCH2Ar),
20.05
(OCH2CH2CH2CH3), 14.39 (OCH2CH2CH2CH3). FAB-MS
m/z: calcd. [M − H]− 1835.7 found 1835.2.
To obtain tetrakis(butoxy)calix[4]arene tetra(mono-t-butyl
isophthalate) (6b) from the same precursor as that of GFB204,
the benzyl groups were removed according to the procedure
described previously, to provide compound 6b as a white solid.
6b: 1H NMR (400 MHz, acetone-d6): d = 9.38 (s, 4H, NH),
8.47 (s, 4H, Ar), 8.40 (s, 4H, Ar), 8.20 (s, 4H, Ar), 7.60 (s,
8H, calix Ar), 4.63 (d, J = 13.2 Hz, 4H, Hax of ArCH2Ar),
4.10 (br t, 8H, OCH2CH2CH2CH3), 3.47 (d, J = 13.2 Hz, 4H,
Heq of ArCH2Ar), 2.01 (m, 8H, OCH2CH2CH2CH3), 1.57 (m,
8H, OCH2CH2CH2CH3), 1.53 (s, 9H, C(CH3)3), 1.06 (t, J =
7.2 Hz, 12H, OCH2CH2CH2CH3). 13C NMR (400 MHz, acetone-
d6): d = 167.39 (CO), 165.87 (CO), 165.07 (CO), 160.61 (4◦,
Ar), 140.38 (4◦, Ar), 135.85 (4◦, Ar), 133.32 (4◦, Ar), 132.23
(4◦, Ar), 129.40 (4◦, Ar), 129.12 (3◦, Ar), 125.93 (3◦, Ar),
125.69 (3◦, Ar), 125.36 (3◦, Ar), 81.67 (4◦, C(CH3)3), 76.03 (2◦,
OCH2CH2CH2CH3), 33.10 (2◦, OCH2CH2CH2CH3), 31.83 (2◦,
ArCH2Ar), 28.27 (1◦, C(CH3)3), 20.07 (2◦, OCH2CH2CH2CH3),
14.41 (1◦, OCH2CH2CH2CH3). HR ESI-MS m/z: calcd [M +
2Na]2+ 873.3569, found 873.3577.
Tetrakis(butoxy)calix[4]arene mono(mono-t-butyl isophthalate)
tris(methoxy) (3b), tetrakis(butoxy)calix[4]arene bis(mono-t-butyl
isophthalate) bis(methoxy) (4b), tetrakis(butoxy)calix[4]arene
tris(mono-t-butyl isophthalate) mono(methoxy) (5b). To a stirred
suspension of tetrakis(butoxy)calix[4]arene tetracarboxylic acid
(1) (30 mg, 0.036 mmol) in dry CH2Cl2 (2 mL) was added oxalyl
chloride (190 lL, 2.18 mmol) and 1 drop of dry DMF. After
stirring overnight the solvent was evaporated, the resulting residue
was dried in vacuo and then dissolved in dry CH2Cl2 (1 mL). A
solution of 2 (35 mg, 0.108 mmol), methanol (5 lL, 0.108 mmol)
and triethylamine (40 lL, 0.288 mmol) in dry CH2Cl2 (1 mL)
was added dropwise and the reaction mixture was stirred for
12 h. The reaction mixture was diluted with CH2Cl2 (3 mL),
washed with 0.5 N HCl (5 mL) and H2O (2 × 3 mL), dried
Tetrakis(butoxy)calix[4]arene tetrakis(monobenzyl isophthalate)
(GFB204). To
a solution of tetrakis(butoxy)calix[4]arene
tetracarboxylic acid (1) (100 mg, 0.12 mmol) in dry methylene
2382 | Org. Biomol. Chem., 2006, 4, 2376–2386
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
over Na2SO4 and concentrated in vacuo. The crude product was
separated by preparative TLC (CH2Cl2–methanol, 100 : 1) to
obtain the fully protected compounds 4a (10 mg, 19%), 6a and
a mixture of 3a and 5a as white solids. 4a is the mixture of
1,2- and 1,3-substituted isomers, and no attempt was made to
separate them. The mixture of 3a and 5a was further purified
by preparative TLC (hexanes–acetone, 4 : 1) to obtain 3a (8 mg,
6.82 (s, 8H, Ar), 5.15 (s, 8H, CH2Ph), 4.39 (d, J = 13.1 Hz, 4H,
Hax of ArCH2Ar), 3.98 (br s, 8H, OCH2CH2CH2CH3), 3.89 (br s,
16H, NCH2COOBz and NCH2COOt-Bu), 3.13 (d, J = 13.1 Hz,
4H, Heq of ArCH2Ar), 1.86 (m, 8H, OCH2CH2CH2CH3), 1.41
(m, 8H, OCH2CH2CH2CH3), 1.38 (s, 36H, OCCH3), 0.98 (t,
J = 7.3 Hz, 12H, OCH2CH2CH2CH3). 13C NMR (100 MHz,
DMSO-d6, 90 ◦C): d = 169.9 (s, CO), 168.0 (s, CO), 167.1 (s,
CO), 156.7 (s, Ar), 135.4 (d, Ar), 133.2 (s, Ph), 128.1 (s, Ar), 127.6
(d, Ph), 127.2 (2d, Ph), 126.7 (s, Ar), 80.6 (s, C(CH3)3), 80.4 (t,
OCH2CH2CH2CH3), 73.9 (t, OCH2Ph), 65.4 (2t, NCH2), 30.9
(t, OCH2CH2CH2CH3), 30.0 (t, ArCH2Ar), 27.1 (q, C(CH3)3),
18.1 (t, OCH2CH2CH2CH3), 12.9 (q, OCH2CH2CH2CH3). After
removal of the t-butyl ester groups as previously described,
compound 7b was obtained as a white solid. HR ESI-MS m/z:
7a: calcd [M + Na]+ 1891.9129, found 1891.9429; 7b: calcd [M −
H]− 1643.6650, found 1643.7494.
1
19%) and 5a (10 mg, 17%). 3a: H NMR (400 MHz, CDCl3):
d = 8.52 (s, 1H, Ar), 8.41 (s, 1H, Ar), 8.38 (s, 1H, Ar), 7.87
(s, 1H, NH), 7.46–7.31 (m, 9H, Ph and Ar calix), 7.17 (s, 2H,
Ar calix), 7.09 (s, 2H, Ar calix), 5.37 (s, 2H, CH2Ph), 4.45 (d,
J = 13.5 Hz, 2H, Hax of ArCH2Ar), 4.44 (d, J = 13.6 Hz, 2H,
Hax of ArCH2Ar), 3.99 (t, J = 7.6 Hz, 4H, OCH2CH2CH2CH3),
3.89 (t, J = 7.3 Hz, 2H, OCH2CH2CH2CH3), 3.86 (t, J =
7.3 Hz, 2H, OCH2CH2CH2CH3), 3.82 (s, 6H, COCH3), 3.54
(s, 3H, COCH3), 3.27 (d, J = 13.5 Hz, 2H, Heq of ArCH2Ar),
3.26 (d, J = 13.6 Hz, 2H, Heq of ArCH2Ar), 1.90–1.81 (m, 8H,
OCH2CH2CH2CH3), 1.58 (s, 9H, C(CH3)3), 1.47–1.39 (m, 8H,
OCH2CH2CH2CH3), 1.01–0.96 (m, 12H, OCH2CH2CH2CH3).
4a: 1H NMR (400 MHz, CDCl3): d = 8.52 (s, 2H, Ar), 8.43
(s, 2H, Ar), 8.37 (s, 2H, Ar), 8.02 (s, 2H, NH), 7.42 (d, 4H,
J = 6.9 Hz, Ph), 7.35–7.22 (m, 14H, Ph and Ar calix), 5.34 (s,
4H, CH2Ph), 4.51–4.44 (m, 4H, Hax of ArCH2Ar), 3.98–3.92 (m,
8H, OCH2CH2CH2CH3), 3.60 (s, 6H, COCH3), 3.32–3.26 (m,
4H, Heq of ArCH2Ar), 1.91–1.83 (m, 8H, OCH2CH2CH2CH3),
1.56 (s, 18H, C(CH3)3), 1.47–1.41 (m, 8H, OCH2CH2CH2CH3),
0.99 (t, J = 7.3 Hz, 12H, OCH2CH2CH2CH3). 5a: 1H NMR
(400 MHz, CDCl3): d = 8.30–8.50 (m, 9H, Ar), 7.87 (s, 2H,
NH), 7.70 (s, 1H, NH), 7.28–7.48 (m, 21H, Ph and Ar calix),
7.17 (s, 1H, Ar calix), 7.09 (s, 1H, Ar calix), 5.33 (s, 4H, CH2Ph),
5.32 (s, 2H, CH2Ph), 4.52 (dd, J1 = 13.6 Hz, J2 = 4.8 Hz, 4H,
Hax of ArCH2Ar), 4.05 (t, J = 7.6, 4H, OCH2CH2CH2CH3),
3.91 (m, 4H, OCH2CH2CH2CH3), 3.52 (s, 3H, COCH3), 3.34
(dd, J1 = 13.6 Hz, J2 = 4.8 Hz, 4H, Heq of ArCH2Ar), 1.90
(m, 8H, OCH2CH2CH2CH3), 1.56 (s, 18H, C(CH3)3), 1.54 (s,
9H, C(CH3)3), 1.50 (m, 8H, OCH2CH2CH2CH3), 1.02 (m, 12H,
OCH2CH2CH2CH3). The deprotected 3b, 4b and 5b were obtained
after removal of the benzyl ester groups from 3a, 4a and 5a
respectively, as previously described. ESI-MS m/z: 3b: calcd [M +
Na]+ 1108.5034, found 1108.5051; 4b: calcd [M − H]− 1289.6,
found 1289.7; 5b: calcd [M + 2Na]2+ 770.8205, found 770.8196.
Tetrakis(butoxy)calix[4]arene tetrakis(monobenzyl mono-t-butyl-
isophthalate-glycine-linked) (8a). To a stirred suspension of 1
(24 mg, 0.029 mmol) in dry CH2Cl2 (2 mL) was added oxalyl
chloride (150 lL, 1.74 mmol) and 1 drop of dry DMF. After
5 h the solvent was evaporated, the resulting residue was dried
in vacuo and then dissolved in dry CH2Cl2 (2 mL). A solution of
1-benzyl 3-t-butyl 5-(2-aminoacetylamino)isophthalate (134 mg,
0.35 mmol) (obtained by coupling of 2 with N-Fmoc-glycine via
the acyl chloride, followed by removal of Fmoc by diethylamine)
in dry CH2Cl2 (1 mL) was added dropwise and the reaction
mixture was stirred for 7 h. The solvent was evaporated and
the residue was purified by crystallization from CH2Cl2–methanol
(1 : 1) to yield the fully protected product 8a as a white solid
1
(45 mg, 68%). 8a: H NMR (400 MHz, CDCl3–MeOD, 1 : 1):
d = 8.37 (s, 4H, Ar), 8.33 (s, 4H, Ar), 8.25 (s, 4H, Ar), 7.40–
7.19 (m, 28H, Ph and Ar calix), 5.30 (s, 8H, CH2Ph), 4.45 (d,
J = 13.4 Hz, 4H, Hax of ArCH2Ar), 4.28 (s, 8H, COCH2), 3.94
(br s, 8H, OCH2CH2CH2CH3), 3.26 (d, J = 13.4 Hz, 4H, Heq
of ArCH2Ar), 1.88 (m, 8H, OCH2CH2CH2CH3), 1.52 (s, 36H,
C(CH3)3), 1.46 (m, 8H, OCH2CH2CH2CH3), 0.99 (t, J = 7.3 Hz,
12H, OCH2CH2CH2CH3). 13C NMR (125 MHz, CDCl3–MeOD,
1 : 1): d = 168.2 (s, CO), 167.8 (s, CO), 165.4 (s, CO), 164.6
(s, CO), 159.4 (s, Ar calix), 138.4 (s, Ar), 135.3 (s, Ph), 134.7 (s,
Ar calix), 132.6 (s, Ar), 130.6 (s, Ar), 128.1 (d, Ph), 127.9 (d,
Ph), 127.7 (d, Ph), 127.5 (d, Ar calix), 126.9 (s, Ar calix), 125.4
(d, Ar), 124.4 (d, Ar), 124.0 (d, Ar), 81.7 (s, C(CH3)3), 74.8 (t,
OCH2CH2CH2CH3), 66.7 (t, OCH2Ph), 43.7 (t, COCH2), 31.9 (t,
OCH2CH2CH2CH3), 30.6 (t, ArCH2Ar), 27.5 (q, C(CH3)3), 18.9
(t, OCH2CH2CH2CH3), 13.5 (q, OCH2CH2CH2CH3). ESI-MS
m/z: 8a: calcd [M + H]+ 2290.02, found 2289.96. The t-butyl ester
groups were then removed to obtain the deprotected product 8b.
Tetrakis(butoxy)calix[4]arene tetrakis(N-benzylacetate N-
acetate) (7b). To a stirred suspension of 1 (26 mg, 0.03 mmol)
in dry CH2Cl2 (2 mL) was added oxalyl chloride (167 lL,
1.92 mmol) and 1 drop of dry DMF. After 5 h the solvent
was evaporated, the resulting residue dried in vacuo and
then dissolved in dry CH2Cl2 (2 mL). A solution of benzyl
((tert-butoxycarbonylmethyl)amino)acetate (53 mg, 0.19 mmol)
(obtained by coupling glycine t-butyl ester hydrochloride to
benzyl bromoacetate in the presence of NEt3 in dry THF) and
diisopropylethylamine (33 lL, 0.19 mmol) in dry CH2Cl2 (1 mL)
was added dropwise and the reaction mixture was stirred for 2 h.
The reaction was then quenched by the addition of H2O (2 mL)
and the organic layer was separated, washed with H2O (2 mL),
dried over Na2SO4 and concentrated in vacuo. The crude product
was purified by preparative TLC (CH2Cl2–methanol, 97 : 3) to
obtain the fully protected 7a as a colorless oil (32 mg, 53%). 7a:
Tetrakis(butoxy)calix[4]arene tetrakis(mono-t-butylisophthalyl
ureido) (9b). To a solution of tetrakis(butoxy)calix[4]arene
tetraamino (50 mg, 0.069 mmol) in toluene (2 mL) was added
triphosgene (27 mg, 0.089 mmol), and the mixture was stirred
at 110 ◦C for 4 h. After cooling the reaction to 30 ◦C, a
solution of 2 (99 mg, 0.30 mmol) and diisopropylethyl amine
(52 lL, 0.30 mmol) in toluene (1.5 mL) was added. The mixture
was stirred overnight at room temperature and the solvent was
evaporated. The residue was dissolved in CH2Cl2 (5 mL), washed
with H2O (2 × 3 mL), dried over Na2SO4 and concentrated
in vacuo. The residue was purified by column chromatography
◦
1H NMR (400 MHz, DMSO-d6, 90 C): d = 7.35 (m, 20H, Ph),
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2376–2386 | 2383
©

View Article Online
(CH2Cl2–methanol, 50 : 1) to yield the fully protected product 9a
as a white solid (58 mg, 40%). The benzyl groups were removed
as described previously to obtain deprotected 9b as a white solid
1) (436.6 mg, 0.66 mmol) in dry CH2Cl2 (10 mL), chlorosulfonic
acid (1.85 mL, 28.0 mmol) was added at 0–10 C. After stirring
◦
at room temperature for 3.5 h, the reaction was poured onto ice,
extracted with CH2Cl2, washed with brine, dried over Na2SO4
and dried in vacuo to obtain tetrakis(methoxyethoxy)calix[4]arene
tetrakis(sulfonylchloride) (697 mg, 99.7%). To a solution of
tetrakis(methoxyethoxy)calix[4]arene tetrakis(sulfonylchloride)
(63 mg, 0.06 mmol) in dry CH2Cl2 (5 mL) was added 2
(122 mg, 0.37 mmol) in dry CH2Cl2 (1 mL) and dry pyridine
(32 lL 0.37 mmol). After stirring overnight, the reaction was
concentrated in vacuo and the residue was purified by column
chromatography (hexanes–acetone, 7 : 3 to 6 : 4) to obtain fully
protected 11a as a white solid (72.4 mg, 73%). After removal of
the t-butyl groups as described previously, 11b was obtained as a
1
(11 mg, 78%). 9b: H NMR (400 MHz, CDCl3–CD3OD, 1 : 1):
d = 8.02 (s, 4H, Ar), 7.99 (s, 4H, Ar), 7.82 (s, 4H, Ar), 6.66 (br s,
8H, Ar calix), 4.28 (d, J = 13.6 Hz, 4H, Hax of ArCH2Ar), 3.72
(br s, 8H, OCH2CH2CH2CH3), 3.99 (d, J = 13.6 Hz, 4H, Heq
of ArCH2Ar), 1.75 (m, 8H, OCH2CH2CH2CH3), 1.39 (s, 36H,
C(CH3)3), 1.28 (m, 8H, OCH2CH2CH2CH3), 0.83 (t, J = 7.6 Hz,
12H, OCH2CH2CH2CH3). 13C NMR (100 MHz, CDCl3–CD3OD,
1 : 1): d = 167.6 (s, CO), 164.9 (s, CO), 160.3 (s, Ar calix), 153.3 (s,
CO), 139.3 (s, Ar), 134.9 (s, Ar calix), 132.0 (s, Ar), 131.9 (s, Ar),
130.9 (s, Ar calix), 123.9 (d, Ar), 122.9 (s, Ar), 122.8 (d, Ar), 119.3
(d, Ar calix), 81.3 (s, C(CH3)3), 74.6 (t, OCH2CH2CH2CH3), 31.7
(t, OCH2CH2CH2CH3), 30.6 (t, ArCH2Ar), 27.3 (q, C(CH3)3),
18.9 (t, OCH2CH2CH2CH3), 13.4 (q, OCH2CH2CH2CH3). ESI-
MS m/z: 9a: calcd [M + H]+ 2121.97, found 2122.18; 9b: calcd
[M − H] 1759.77, found 1759.74.
1
white solid. 11b: H NMR (400 MHz, acetone-d6): d = 8.35 (t,
J = 1.6 Hz, 4H, Ar), 8.09 (dd, J = 2.4, 1.6 Hz, 4H, Ar), 8.04
(dd, J = 2.4, 1.6 Hz, 4H, Ar), 7.53–7.32 (m, 20H, Ph), 7.26 (s,
8H, Ar calix), 5.41 (s, 8H, CH2Ph), 4.50 (d, J = 13.6 Hz, 4H,
Hax of ArCH2Ar), 4.13 (t, J = 4.4 Hz, 8H, CH2CH2OCH3), 3.62
(t, J = 4.4 Hz, 8H, CH2CH2OCH3), 3.17 (d, J = 13.6 Hz, 4H,
Heq of ArCH2Ar), 3.10 (s, 12H, OCH3). 13C NMR (400 MHz,
acetone-d6): d = 166.3 (CO), 165.5 (CO), 160.9 (Ar calix), 139.8
(Ar), 137.1 (Ph), 136.3 (Ar calix), 134.5 (Ar or Ar calix), 132.8
(Ar or Ar calix), 132.5 (Ar or Ar calix), 129.4 (Ph), 129.2 (Ph),
129.1 (Ph), 128.2 (Ar calix), 126.7 (Ar), 126.0 (Ar), 125.6 (Ar),
74.4 (CH2Ph), 72.1 (CH2CH2OCH3), 67.7 (CH2CH2OCH3), 58.3
(OCH3), 31.4 (ArCH2Ar). ESI-MS m/z: 11a: calcd [M + Na]+
2235.69, found 2235.71; 11b: calcd [M − H]− 1987.45, found
1987.67.
Calix[4]arene tetrakis(monobenzyl isophthalyl diazo) (10b) and
calix[4]arene tris(monobenzyl isophthalyl diazo) (12b). To a so-
lution of 2 (263 mg, 0.80 mmol) in 1 N HCl (1.6 mL) and THF
(1.6 mL), NaNO2 (58 mg, 0.84 mmol) was added. The mixture
was cooled to 0 ◦C and stirred for 20 min. NaBF4 (171 mg,
1.56 mmol) was added and the solution was stirred for 10 min.
This solution was then added to a solution of calix[4]arene (76 mg,
0.18 mmol) in methanol–DMF (1.6 mL, 1 : 1). After stirring for
2 h the reaction was quenched by the addition of H2O (5 mL) and
extracted with CH2Cl2. The organic layer was washed with H2O
(3 mL), dried over Na2SO4 and concentrated in vacuo. The residue
was dissolved in CH2Cl2, and methanol was added to precipitate an
orange solid. This solid was crystallized from CH2Cl2 + MeOH
three times to obtain the fully protected product 10a (100 mg,
32%). The three filtrate fractions were combined and evaporated
to obtain 12a (65 mg, 25%). The t-butyl groups were removed
Tetrakis(methoxyethoxy)calix[4]arene tetrakis(monobenzyl iso-
phthalate) (13b). To a solution of tetrakis(methoxyethoxy)alix-
[4]arene tetracarboxylic acid (synthesized in a manner similar to
1) (168 mg, 0.20 mmol) in dry CH2Cl2 was added oxalyl chloride
(1.06 mL, 12.1 mmol) and 1 drop of dry DMF. After stirring
overnight, the solvent was evaporated, and the residue was dried
in vacuo and then dissolved in dry CH2Cl2 (4 mL). A solution
of 2 (392 mg, 1.2 mmol) in dry CH2Cl2 (1 mL) was added,
followed by diisopropylethylamine (278 lL, 1.6 mmol) and the
reaction mixture was stirred for 12 h. It was poured into water,
extracted with CH2Cl2, washed with water, dried over MgSO4
and concentrated in vacuo. The residue was purified by column
chromatography (ethyl acetate–hexanes–CH2Cl2, 7 : 3 : 1) to
obtain the fully protected product 13a as a white solid (142 mg,
40%). After removal of the t-butyl groups as previously described,
compound 13b was obtained as a white solid. 13b: 1H NMR
(400 MHz, acetone-d6): d = 9.41 (s, 4H, NH), 8.49 (br, 8H, Ar),
8.26 (br, 4H, Ar), 7.57–7.27 (m, 28H, Ph and Ar calix), 5.31 (s, 8H,
CH2Ph), 4.74 (d, J = 13.6 Hz, 4H, Hax of ArCH2Ar), 4.26 (t, J =
4 Hz, 8H, CH2CH2OCH3), 3.87 (t, J = 4 Hz, 8H, CH2CH2OCH3),
3.42 (d, J = 13.6 Hz, 4H, Heq of ArCH2Ar), 3.40 (s, 12H, OCH3).
13C NMR (400 MHz, acetone-d6): d = 171.5 (CO), 171.0 (CO),
170.5 (CO), 161.8 (Ar calix), 141.7 (Ar), 138.2 (Ph), 137.0 (Ar,
calix), 133.2(Ar, Ph or Ar calix), 132.7 (Ar, Ph or Ar calix), 130.4
(CH of Ph), 130.3 (Ar, Ph or Ar calix), 130.1 (CH of Ph), 130.0
(CH of Ph), 127.2 (CH of Ar), 127.0 (CH of Ar), 126.6 (CH of
Ar), 75.6 (CH2Ph), 73.7 (CH2CH2OCH3), 68.5 (CH2CH2OCH3),
59.8 (OCH3), 32.7 (ArCH2Ar). ESI-MS m/z: 13b: calcd [M − H]−
1843.58, found 1843.63.
1
as described previously to obtain 10b and 12b. 10b: H NMR
(400 MHz, DMSO-d6): d = 8.48 (s, 12H, Ar), 7.90 (s, 8H, Ar
calix), 7.45 (d, J = 7.2 Hz, 8H, Ph), 7.35–7.27 (m, 12H, Ph), 5.37
(s, 8H, CH2Ph), 4.38 (br s, 4H, Hax of ArCH2Ar), 3.67 (br s, 4H,
Heq of ArCH2Ar). 13C NMR (125 MHz, CDCl3): d = 165.9 (s,
CO), 164.4 (s, CO), 160.0 (s, Ar calix), 152.5 (s, Ar), 144.4 (s, Ar
calix), 135.6 (s, Ph), 132.5 (s, Ar), 131.1 (d, Ar calix), 130.5 (s, Ar),
130.0 (d, Ar calix), 128.4 (d, Ph), 128.1 (2d, Ph), 126.5 (d, Ar),
126.0 (d, Ar), 124.3 (d, Ar), 66.7 (t, OCH2Ph), 31.6 (t, ArCH2Ar).
1
12a: H NMR (500 MHz, DMSO-d6): d = 10.30 (br, 3H, NH),
8.56–8.69 (m, 9H, Ar), 7.77–7.87 (m, 6H, Ar calix), 7.30–7.46 (m,
16H, Ph and Ar calix), 7.23 (s, 1H, Ar calix), 7.21 (s, 1H, Ar calix),
5.38–5.42 (m, 6H, CH2Ph), 4.38 (br t, 4H, Hax of ArCH2Ar), 3.88
(br d, 2H, Heq of ArCH2Ar), 3.75 (br d, 2H, Heq of ArCH2Ar),
1
1.61 (m, 27H, C(CH3)3). 12b: H NMR (500 MHz, DMSO-d6):
d = 8.55 (m, 9H, Ar), 7.87 (s, 2H, Ar calix), 7.84 (d, J = 1.6 Hz,
2H, Ar calix), 7.77 (d, J = 1.6 Hz, 2H, Ar calix), 7.29–7.46 (m,
16H, Ph and Ar calix), 7.09 (s, 1H, Ar calix), 7.07 (s, 1H, Ar calix),
5.34–5.39 (m, 6H, CH2Ph), 4.40 (br, 4H, Hax of ArCH2Ar), 3.80
(br, 2H, Heq of ArCH2Ar). ESI-MS m/z: 10b: calcd [M − H]−
1551.42, found 1551.48.
Tetrakis(methoxyethoxy)calix[4]arene tetrakis(monobenzyl iso-
phthalate sulfonamide-linked) (11b). To a solution of tetrakis-
(methoxyethoxy)calix[4]arene (synthesized in a manner similar to
2384 | Org. Biomol. Chem., 2006, 4, 2376–2386
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Tetrakis(methoxy)calix[4]arene tetrakis(mono-t-butyl isophtha-
late) (14b). To a stirred suspension of tetrakis(methoxy)-
calix[4]arene tetracarboxylic acid (synthesized in a manner similar
to 1) (20 mg, 0.03 mmol) in dry CH2Cl2 (2 mL) was added oxalyl
chloride (160 lL, 1.83 mmol) and 1 drop of dry DMF. After 5 h the
solvent was evaporated, the resulting residue was dried in vacuo
and dissolved in dry CH2Cl2 (1 mL). A solution of 2 (118 mg,
0.36 mmol) in dry CH2Cl2 (1 mL) was added dropwise and the
reaction mixture was stirred for 12 h. The solvent was evaporated
and the residue was purified by column chromatography using
CH2Cl2–methanol (50 : 1) to yield the fully protected product 14a
as a white solid (47 mg, 82%). After removal of the benzyl groups
as described previously, 14b was obtained as a white solid. 14b:
1H NMR (400 MHz, CD3OD): d = 8.72 (s, 1H, Ar paco (partial
cone conformation)), 8.68 (s, 1H, Ar paco), 8.65 (s, 1H, Ar paco),
8.59 (s, 1H, Ar paco), 8.47 (s, 4H, Ar cone), 8.37 (s, 4H, Ar cone),
8.35–8.34 (3s, 4H, Ar paco), 8.28 (s, 2H, Ar paco), 8.22 (s, 4H, Ar
cone), 8.16 (s, 2H, Ar paco), 8.10 (s, 2H, Ar calix paco), 7.88 (s,
2H, Ar calix paco), 7.74 (s, 2H, Ar calix paco), 7.59 (s, 8H, Ar calix
cone), 4.53 (d, J = 13.2 Hz, 4H, ArCH2Ar cone), 4.22 (d, 2H, J =
13.5 Hz, ArCH2Ar paco), 3.98 (s, 12H, OCH3 cone), 3.95 (br s, 4H,
ArCH2Ar paco), 3.82 (s, 9H, OCH3 paco), 3.49 (d, J = 13.2 Hz,
4H, ArCH2Ar cone), 3.39 (d, 2H, J = 13.5 Hz, ArCH2Ar paco),
3.20 (s, 3H, OCH3 paco), 1.65 (s, 9H, C(CH3)3 paco), 1.61 (s, 9H,
C(CH3)3 paco), 1.52 (s, 18H, C(CH3)3 paco), 1.51 (s, 36H, C(CH3)3
cone). 1H NMR (400 MHz, DMSO-d6, 90 ◦C): d = 10.09 (s, 4H,
NH), 8.56 (s, 4H, Ar), 8.51 (s, 4H, Ar), 8.13 (s, 4H, Ar), 7.82 (br s,
8H, Ar calix), 3.95 (br s, 8H, ArCH2Ar), 3.51 (br s, 9H, OCH3),
1.55 (s, 36H, C(CH3)3). 13C NMR (100 MHz, DMSO-d6, 90 ◦C):
d = 165.0 (s, CO), 163.7 (s, CO), 163.0 (s, CO), 160.0 (s, Ar), 133.2
(s, Ar), 131.7 (s, Ar), 128.4 (s, Ar), 124.5 (d, Ar), 124.3 (d, Ar),
123.9 (d, Ar), 80.6 (s, C(CH3)3), 27.2 (q, C(CH3)3). ESI-MS m/z:
14a: calcd [M + H]+ 1893.74, found 1894.09; 14b: calcd [M − H]−
1531.54, found 1531.53.
Acknowledgements
We thank the National Institutes of Health (GM 35208 and
CA106829) for financial support of this work.
References
1 C. H. Heldin, A. Ostman and L. Ronnstrand, Biochim. Biophys. Acta,
1998, 1378(1), F79–F113.
2 M. Tallquist and A. Kazlauskas, Cytokine Growth Factor Rev., 2004,
15, 205.
3 S. M. Jones and A. Kazlauskas, FEBS Lett., 2001, 490, 110.
4 L. Seymour, D. Dajee and W. R. Bezwoda, Breast Cancer Res. Treat.,
1993, 26, 247.
5 R. Henriksen, K. Funa, E. Wilander, T. Backstrom, M. Ridderheim
and K. Oberg, Cancer Res., 1993, 53, 4550.
6 J. H. Yu, C. Ustach and H. R. C. Kim, J. Biochem. Mol. Biol., 2003,
36, 49.
7 I. F. Dunn, O. Heese and P. M. Black, J. Neuro–Oncol., 2000, 50, 121.
8 R. Kumar, J. Yoneda, C. D. Bucana and I. J. Fidler, Int. J. Oncol., 1998,
12, 749.
9 D. Wang, H. J. Huang, A. Kazlauskas and W. K. Cavenee, Cancer Res.,
1999, 59, 1464.
10 P. Rockwell and N. I. Goldstein, Mol. Cell. Differ., 1995, 3, 315.
11 C. H. Heldin and B. Westermark, Physiol. Rev., 1999, 79, 1283.
12 W. D. Klohs, D. W. Fry and A. J. Kraker, Curr. Opin. Oncol., 1997, 9,
562.
13 A. Levitzki and A. Gazit, Science, 1995, 267, 1782.
14 K. Matsuno, T. Selshi, T. Nakajima, M. Ichimura, N. A. Giese, J. C.
Yu, S. Oda and Y. Nomoto, Bioorg. Med. Chem. Lett., 2003, 13, 3001.
15 P. Traxler, Expert Opin. Ther. Targets, 2003, 7, 215.
16 A. Levitzki, Cytokine Growth Factor Rev., 2004, 15, 229.
17 A. Ostman and C. H. Heldin, Adv. Cancer Res., 2001, 80, 1.
18 V. Ramakrishnan, M. A. Escobedo, L. J. Fretto, J. J. Seroogy, J. E.
Tomlinson and D. L. Wolf, Growth Factors, 1993, 8, 253.
19 D. M. Brennand, U. Dennehy, V. Ellis, M. F. Scully, P. Tripathi, V. V.
Kakkar and G. Patel, FEBS Lett., 1997, 413, 70.
20 U. Engstrom, A. Engstrom, A. Ernlund, B. Westermark and C. H.
Heldin, J. Biol. Chem., 1992, 267, 16581.
21 J. Zimmermann, Chimia, 2002, 56, 428.
22 G. A. Ferns, E. W. Raines, K. H. Sprugel, A. S. Motani, M. A. Reidy
and R. Ross, Science, 1991, 253, 1129.
23 D. S. Duan, M. J. Pazin, L. J. Frettoo and L. T. Williams, J. Biol. Chem.,
1991, 266, 413.
24 L. S. Green, D. Jellinek, R. Jenison, A. Ostman, C. H. Heldin and N.
Janjic, Biochemistry, 1996, 35, 14413.
25 Y. Hamuro, M. C. Calama, H. S. Park and A. D. Hamilton, Angew.
Chem., Int. Ed. Engl., 1997, 36, 2680.
26 H. S. Park, Q. Lin and A. D. Hamilton, J. Am.Chem. Soc., 1999, 121,
8.
Calix[4]arene tetrakis(mono-t-butyl isophthalate) (15b). To a
stirred suspension of tetrakis(benzyloxy)calix[4]arene tetracar-
boxylic acid (synthesized in a manner similar to 1) (25 mg,
0.026 mmol) in dry CH2Cl2 (2 mL) was added oxalyl chloride
(90 lL, 1.56 mmol) and 1 drop of dry DMF. After stirring
overnight the solvent was evaporated, the resulting residue was
dried in vacuo and then dissolved in dry CH2Cl2 (2 mL). A
solution of 2 (51 mg, 0.156 mmol) in dry CH2Cl2 (1 mL) was
added, followed by pyridine (15 lL, 0.156 mmol), and the reaction
mixture was stirred for 12 h. The reaction was poured into water,
then extracted with CH2Cl2, and washed with brine and dried over
Na2SO4. The solvent was evaporated and the residue was purified
by column chromatography (hexanes–acetone, 4 : 1) to yield 15a
as a white solid. (25 mg, 45%). After removal of the benzyl ester
groups as described previously, 15b was obtained as a white solid.
15b: 1H NMR (400 MHz, acetone-d6 and drops of CD3OD): d =
8.61 (s, 4H, Ar), 8.53 (s, 4H, Ar), 8.23 (s, 4H, Ar), 7.84 (s, 8H,
Ar calix), 4.50 (br, 4H, Hax of ArCH2Ar), 3.60 (br, 4H, Heq of
ArCH2Ar), 1.52 (s, 36H, C(CH3)3). 13C NMR (400 MHz, CDCl3):
d = 165.9 (s, acetone-d6 and drops of CD3OD): d = 168.9 (CO),
166.1 (CO), 158.7 (Ar), 140.7 (Ar), 133.8 (Ar), 132.8 (Ar), 131.3
(Ar), 129.7 (Ar), 126.9 (Ar), 126.6 (Ar), 126.3 (Ar), 82.7 (C(CH3)3),
33.3 (ArCH2Ar), 28.3 (C(CH3)3). HR ESI-MS m/z: 15b: calcd
[M + Na]+ 1499.4747, found 1499.4722.
27 H. S. Park, Q. Lin and A. D. Hamilton, Proc. Natl. Acad. Sci. U. S. A.,
2002, 99, 5105.
28 M. A. Blaskovich, Q. Lin, F. L. Delarue, J. Sun, H. S. Park, D. Coppola,
A. D. Hamilton and S. M. Sebti, Nat. Biotechnol., 2000, 18, 1065.
29 S. M. Sebti and A. D. Hamilton, Oncogene, 2000, 19, 6566.
30 K. K. Clark-Ferris and J. Fisher, J. Am. Chem. Soc., 1985, 107, 5007.
31 M. A. Fazal, B. C. Roy, S. G. Sun, S. Mallik and K. R. Rodgers, J. Am.
Chem. Soc., 2001, 123, 6283.
32 A. L. Banerjee, M. Swanson, B. C. Roy, X. Jia, M. K. Haldar, S. Mallik
and D. K. Srivastava, J. Am. Chem. Soc., 2004, 126, 10875.
33 M. R. Arkin and J. A. Wells, Nat. Rev. Drug Discovery, 2004, 3, 301.
34 T. Berg, Angew. Chem., Int. Ed., 2003, 42, 2462.
35 T. A. Larsen, A. J. Olson and D. S. Goodsell, Struct. Fold. Des., 1998,
6, 421.
36 F. B. Sheinerman, R. Norel and B. Honig, Curr. Opin. Struct. Biol.,
2000, 10, 153.
37 A. V. Veselovsky, Y. D. Ivanov, A. S. Ivanov, A. I. Archakov, P. Lewi
and P. Janssen, J. Mol. Recognit., 2002, 15, 405.
38 C. Oefner, A. D. Arcy, F. K. Winkler, B. Eggimann and M. Hosang, M.,
EMBO J., 1991, 11, 3921.
39 J. Sun, D. Wang, R. K. Jain, A. Carie, S. Paquette, E. Ennis, M. A.
Blaskovich, L. Baldini, D. Coppola, A. D. Hamilton and S. M. Sebti,
Oncogene, 2005, 24, 4701.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 2376–2386 | 2385
©

View Article Online
40 N. Giese, W. J. LaRochelle, M. May-Siroff, K. C. Robbins and S. A.
Aaronson, Mol. Cell. Biol., 1990, 10, 5496.
41 J. M. Clements, L. J. Bawden, R. E. Bloxidge, G. Catlin, A. L. Cook,
A. H. Drummond, R. M. Edwards, A. Fallon, D. R. Green, P. G.
Hellewell, P. M. Kirwin, P. D. Nayee, S. J. Richardson, D. Brown,
S. B. Chahwala, M. Snarey and D. Winslow, EMBO J., 1991, 10,
4113.
43 W. J. LaRochelle, J. H. Pierce, M. May-Siroff, N. Giese and S. A.
Aaronson, J. Biol. Chem., 1992, 267, 17074.
44 J. Kreysing, A. Ostman, M. van de Poll, G. Backstrom and C. H.
Heldin, FEBS Lett., 1996, 385, 181.
45 D. Schilling, J. D. Reid, A. Hujer, D. Morgan, E. Demoll, P. Bumer,
R. A. Fenstermarker and D. M. Kaetzel, Biochem. J., 1998, 333, 637.
46 F. C. J. M. van Veggel, in Calixarenes in Action, ed. L. Mandolini and
R. Ungaro, Imperial College Press, London, 2000, p. 11.
47 T. Mecca and A. D. Hamilton, unpublished work.
42 A. Ostman, M. Andersson, U. Hellman and C. H. Heldin, J. Biol.
Chem., 1991, 266, 10073.
2386 | Org. Biomol. Chem., 2006, 4, 2376–2386
This journal is
The Royal Society of Chemistry 2006
©