Design, synthesis, and functionalization of dimeric peptides targeting chemokine receptor CXCR4

Article abstract of DOI:10.1021/jm2009716

The chemokine receptor CXCR4 is a critical regulator of inflammation and immune surveillance, and it is specifically implicated in cancer metastasis and HIV-1 infection. On the basis of the observation that several of the known antagonists remarkably shar

Full text of DOI:10.1021/jm2009716

ARTICLE
pubs.acs.org/jmc
Design, Synthesis, and Functionalization of Dimeric Peptides
Targeting Chemokine Receptor CXCR4
Oliver Demmer,† Ingrid Dijkgraaf,‡ Udo Schumacher,§ Luciana Marinelli,|| Sandro Cosconati,||
Eleni Gourni,‡ Hans-J€urgen Wester,‡ and Horst Kessler†,^,*
†Institute for Advanced Study and Center of Integrated Protein Science at the Department Chemie, Technische Universit€at M€unchen,
Lichtenbergstrasse 4, D-85748 Garching, Germany
‡Lehrstuhl f€ur Pharmazeutische Radiochemie, Technische Universit€at M€unchen, Walther-Meissner-Strasse 3, 85748 Garching, Germany
§Institute for Anatomy II: Experimental Morphology, Universit€atsklinikum Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg,
Germany
Dipartimento di Chimica Farmaceutica e Tossicologica, Universitꢀa di Napoli “Federico II”, Via D. Montesano, 49-80131 Napoli, Italy
^Chemistry Department, Faculty of Science, King Abdulaziz University, P.O. Box 80203, Jeddah 21589, Saudi Arabia
S Supporting Information
b
ABSTRACT: The chemokine receptor CXCR4 is a critical
regulator of inflammation and immune surveillance, and it is
specifically implicated in cancer metastasis and HIV-1
infection. On the basis of the observation that several of
the known antagonists remarkably share a C2 symmetry
element, we constructed symmetric dimers with excellent
antagonistic activity using a derivative of a cyclic pentapep-
tide as monomer. To optimize the binding affinity, we
investigated the influence of the distance between the
monomers and the pharmacophoric sites in the synthesized constructs. The affinity studies in combination with docking
computations support a two-site binding model. In a final step, 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA)
was introduced as chelator for (radio-)metals, thus allowing to exploit these compounds as a new group of CXCR4-binding peptidic
probes for molecular imaging and endoradiotherapeutic purposes. Both the DOTA conjugates and some of their corresponding
metal complexes retain good CXCR4 affinity, and one 68Ga labeled compound was studied as PET tracer.
’ INTRODUCTION
melanoma, brain cancers): primary tumor growth, migration of
cancer cells, and establishment of metastases. In result CXCR4 is
overexpressed in more than 70% of cancers, generating the need
for personalized treatment through a combination of diagnosis
and treatment.
As part of our ongoing effort to develop high affinity CXCR4
binding ligands for molecular imaging and endoradiotherapeutic
purposes, we were interested in introducing DOTA, a chelator
often used for a wide variety of medically interesting metal ions.7
Because functionalization of a given ligand is often accompanied
by a significant reduction in binding affinity, we explored again
the concept of dimerization (multimerization) to compensate for
potential impairment of binding affinity.8ꢀ11
CXCR4 is one of the most prominent members of the
chemokine receptor family and a central component of the
communication pathways in the body. This seven-transmem-
brane G-protein coupled receptor (GPCR) and its natural ligand,
the stromal cell-derived factor (CXCL12; formerly known as
SDF1) are part of a signaling system involved in cell migration
of organogenesis, hematopoiesis, inflammation, and immune
response.1 Because of this central role, the CXCR4ꢀCXCL12
interaction is involved in various severe diseases like cancer, HIV,
and autoimmune diseases that are seemingly independent at first
glance. The therapeutic potential of CXCR4 antagonists has
already been shown for the treatment of HIV infection, cancer,
and rheumatoid arthritis.2ꢀ5 Other potential therapeutic uses of
CXCR4 antagonists have been described for asthma, the mobi-
lization of stem cells for stem cell transplantations, and the
attenuation of pain, and they are also discussed for the treatment
of neurological diseases.6 Our focus is to find an approach for
personalized medicine for cancer by providing tools for diagnosis
and treatment. CXCR4 is involved in three fundamental stages of
various cancer types (e.g., in lung, breast, prostate, ovarian, colon,
The concept of polyvalency, and herein more specifically
dimers, is particularly interesting in the case of CXCR4 due to
the peculiar structure of the receptor which was first hypothe-
sized to contain two neighboring binding sites, one responsible
for the affinity and the other for signaling function.12
Received: July 20, 2011
Published: September 12, 2011
r
2011 American Chemical Society
7648
dx.doi.org/10.1021/jm2009716 J. Med. Chem. 2011, 54, 7648–7662
|

Journal of Medicinal Chemistry
ARTICLE
Scheme 1. Dimeric CXCR4 Antagonists 1ꢀ3
Scheme 2. the Original Cyclic Pentapeptide FC131 4 and
Our Derivative 7 Used for Dimerization
using very long spacers with length approximately from 20 to
80 Å can have positive effects on binding affinity.28 In this study,
ꢁ
evidence was provided that two CXCR4 receptors can be
addressed by two ligands in one large molecule when they are
separated by a long spacer. In contrast, our studies of ligand
oriented design focus on dimers connected by shorter spacers to
mimic known ligands. Here two (different) binding sites in the
same monomer can be addressed (see below).
After further modifications of 4, including an N-methylation
for affinity enhancement and the elucidation of a side chain
suitable for additional modifications, we selected cyclo(D-Tyr1-
D-[NMe]Orn2-Arg3-Nal4-Gly5) (D-[NMe]Orn = Nα-methyl-
D-ornithine; Nal = L-3-(2-naphthyl)alanine) 7 for our dimer-
ization studies.29,30 Because the side chain of D-[NMe]Orn2 is
the “least important residue” in this peptide and is quite
amenable to acylation, we used D-[NMe]Orn2 as anchor point
for dimerization.31
Herein we describe the synthesis of dimers and evaluation of
the optimal distance between the monomeric units. Subsequent
comparison of the affinity improvement with other C2 symmetric
CXCR4 ligands in combination with the exchange of supposedly
important pharmacophoric residues gives hints on a different
binding mode originating from the design of 4. Finally DOTA is
introduced to the dimers as chelator allowing preparation of
(radio-)metal ion complexes of peptides with promising poten-
tial for therapeutic and diagnostic applications. The recent
advent of the first and unique X-ray structures of CXCR4
receptor alone and in complex with the antagonist small mole-
cule IT1t and the cyclic peptide CVX15 give us the opportunity
to model the interaction of the cyclic peptide monomer 7 and the
dimeric peptide 10 with the receptor.19
On the other hand, several known CXCR4 antagonists like
the FDA approved drug Mozobil (AMD3100 1; Scheme 1)
contain a C2 symmetry element,13 and its success story has
sparked the development of further excellent ligands.14 Not only
1 and its derivatives exhibit this C2 symmetry element but
also other small molecules (e.g., dipicolylamine-zinc(II)
compounds; 2) as well as peptidic dimers (e.g., S1D 3) that
have been found independently (Scheme 1).15,16 Although the
structure and thus its exact symmetry has not been fully revealed,
the orphan drug CTCE-9908 is also a dimeric peptidic CXCR4
antagonist.5,17 While many of the above-cited ligands would well
fit in the “two-site model” of the receptor, it is well-known that
CXCR4 can form homo- and heterodimers, and this has to be
considered as an alternative explanation why some symmetric
dimers have superior binding properties in comparison to their
monomers.18 However, the spacing of the ligand-binding sites in
the crystal structure of dimeric β2-adrenergic receptor matches
the ∼40 Å distance, and this was recently demonstrated to be
approximately the same in the CXCR4 dimers,19 thus suggesting
that the formation of a functional 2:2 SDF-1:CXCR4 com-
plex might be plausible, while for smaller ligands the same is
unimaginable and the “two-site model” of the receptor seems to
be the most reliable hypothesis.
However, the functional importance of dimerization remains
incompletely characterized, although a considerable body of data
suggests that it has important in vivo pharmacological effects.
Besides these CXCR4-specific considerations for the construc-
tion of dimeric ligands, dimers and higher multimers have been
used in other fields of medicinal chemistry.10,11,20,21 Especially in
molecular imaging, we and others could demonstrate the power
of the multivalency approach to improve target/ligand interac-
tion and thus imaging contrast.8,22
As a starting point for our ligand development, we have chosen
4 (FC131, Scheme 2), which was developed by Fujii et al. by
downsizing of polyphemusin II.23ꢀ25 It combines high CXCR4
affinity with the commonly low toxicity of peptides and a high
stability toward enzymatic degradation associated with cyclic
pentapeptides.25,26 Additionally, this class of ligands are believed
to have less side effects when used as pharmaceuticals as they are
inverse agonists, especially in comparison with 1, which is a
partial agonist.4,27 In a very recent study, it could be shown that
’ RESULTS
Synthesis. Monomeric, N-methylated cyclic pentapeptides
were synthesized according to established procedures.32 Glycine
was attached to the resin to avoid racemization during the
cyclization step and to allow for easier cyclization due to turn
preformation caused by the N-terminal D-amino acid.33 The
peptide chain was elongated until ornithine with standard Fmoc
solid phase peptide synthesis (SPPS). Nα-2-Nitrobenzenesul-
fonyl (nosyl; Ns) protected Nδ-Boc-D-ornithine was synthesized
as building block because the Ns group activates the Nα for fast
on-resin N-methylation.32,34 Nα-Alloc-Nδ-Fmoc-D-ornithine was
used in branched peptides 6, 8, and 9 to modify the side chain.
Alloc was cleaved after acylation of the side chain and the Nα
7649
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
Scheme 3. Synthesis of the Nα-Methylated, Branched, Cyclic Pentapeptidesa
a Reagents: (i) 20% piperidine, NMP; (ii) Fmoc-AA, TBTU, HOBT, DIEA, NMP; (iii) Pd(PPh3)4, PhSiH3, DCM; (iv) NsCl, collidine, NMP;
(v) MeOH, DIAD, PPh3, THF; (vi) DBU, 2-mercaptoethanol, NMP; (vii) Fmoc-AA, HATU, HOAt, DIEA, NMP; (viii) 20% HFIP, DCM; (ix) DPPA,
NaHCO3, DMF; (x) 95% TFA, 2.5% TIPS, 2.5% H2O.
Scheme 4. Construction of Diacidic Building Blocks Shown Exemplary for the Homo-Bβ-Aspartic Acid Linker
reprotected with Ns and then N-methylated (Scheme 3). After
cleavage of the Ns group, D-tyrosine was attached using HATU
and subsequently the linear peptide was cyclized and deprotected
to yield the pure peptide after HPLC separation.
methylene group via the ArndtꢀEistert homologization.35 The
protected aspartic acid was reacted with diazomethane, and the
silver-catalyzed Wolff rearrangement gave the homobeta-amino
acid (Scheme 4). For both aspartic acids, the second acid group
was protected with allylbromide so that this allyl group can be
cleaved orthogonally to the t-Bu groups of tris(t-Bu)DOTA. The
free DOTA moiety was then obtained in the final acidic deprotec-
tion step with HCl. The peptidic linkers were constructed by
attaching the aspartic acid building blocks to the resin followed by
standard SPPS to attach 6-aminohexanoic acid and tris(t-
Bu)DOTA and Pd-catalyzed allyl deprotection before the linker
was cleaved from the resin and purified by HPLC (see Supporting
Information).
The monomeric peptide units were coupled with the DOTA
labeled linkers described above, deprotected, and if required also
transformed into the corresponding In3+ chelate in an one-pot
reaction (see Supporting Information).
Biological Results. Compound 5 was resynthesized and
tested to ensure the comparability of our testing system with
the one used in literature (Table 1).29 The inactive monomer
6 was generated by exchanging the crucial arginine residue
isosterically with the uncharged citrulline. It also carries the
Peptide monomers were dimerized by coupling in DMF
solution, with DIEA and 0.5 equiv of the corresponding diacid.
The distance between the monomers was varied by using linear
diacids with 3ꢀ16 carbon atoms. In contrast to this procedure,
dimer 11 with one active and one inactive monomer was built by
coupling the inactive monomer with a 10-fold excess of glutaric
acid to suppress dimerization. After HPLC purification, the active
monomer was attached in a second step in equimolar amounts.
To allow the introduction of functional moieties into the
dimeric peptides, trifunctional linkers were used. Two diacid
building blocks were synthesized to vary the distance between
the two peptidic units that contain an additional amine for the
introduction of further spacing and DOTA units. They were
constructed on solid support containing tris(t-Bu)DOTA con-
nected via an 6-aminohexanoyl unit to aspartic acid or homobeta-
aspartic acid, respectively. The symmetric homobeta-aspartic acid
unit was built from the standard L-aspartic acid used for SPPS
containing Fmoc and t-Bu protection by introducing the
7650
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
Table 1. Monomeric and Dimeric CXCR4 Ligands and IC50
Valuesa
The dimers additionally containing the DOTA moiety 21ꢀ28
have a wide range of affinities ranging from nearly 10 nM to
almost 100 nM (Table 2) depending on the spacer and the
chelation with indium. With the exception of the dimers 23 and
24, the indium chelates show a higher CXCR4 affinity than their
free DOTA counterparts. The most active compounds, 25ꢀ28,
contain an additional glycine spacer which corresponds in
combination with the DOTA conjugated spacer to the length
found in the best dimer 15.
25 was labeled with 68Ga, and the labeled tracer was used for
initial in vivo evaluation in OH1 h-SCLC tumor bearing nude
mice. Biodistibution studies carried out 60 min after the intra-
venously injection of the labeled compound into the tail vain of
mice (Table 3). Attention being focused on the experimental
tumor and on CXCR4 bearing organs, such as spleen and liver,
because it is known that high CXCR4 mRNA expression is
observed. At 60 min pi, the liver was the organ with the highest
accumulation of radioactivity (44.3 ( 5.5%ID/g).36 Relatively
high accumulation was also observed in lung (2.1 ( 0.3%ID/g),
spleen (4.0 ( 0.6%ID/g), and kidney (3.3 ( 0.5%ID/g). The
accumulation of the radioactivity in the tumor (2.1 ( 0.5%ID/g)
was higher than in the blood (1.9 ( 0.3%ID/g) and the muscle
(0.4 ( 0.1%ID/g). Co-injection of a natGa containing CXCR4
specific PET tracer (compound 2c ref 31) resulted in lower
tumor uptake with a reduction of approximately 50%, demon-
strating retention of 68Ga-25 in the experimental tumor by
CXCR4-binding. The increased kidney uptake under blocking
conditions also confirms competition of tracer binding to CXCR4
receptors by the cold peptide. On the basis of the suboptimal
biodistribution, specificity of binding of the new tracer to CXCR4
receptors was demonstrated in a competition study by coinjection
of an excess of cold receptor ligand in one mouse. In such cases, i.
e. when the biodistribution of a tracer is found to be suboptimal,
we suggest demonstrating further in vivo data, such as metabolic
stability or specificity of binding, only on a very small number or
cohort of animals and to refrain from aiming to quantify further
in vivo values with limited or no further scientific impact with high
accuracy. Although only one animal has been used, a quantitative
blockade is always a clear indicator for high specificity of binding.
As a consequence of reducing the uptake in the tumor by
blockade, the majority of the tracer is now directed and retained
in organs known to typically retain (basic) peptides, such as
somatostatins, i.e. the kidneys and the spleen.
L
n
compd
R
m
IC50 [nM]
5
6b
Gua
Ac
2 ( 2
1000
7
H
6 ( 1
60 ( 20
67 ( 5
4 ( 2
19 ( 7
12 ( 8
4 ( 1
3 ( 1
2 ( 1
4 ( 7
33 ( 17
8 ( 5
6 ( 2
14 ( 2
8
Ac
9
Gma
10
11c
12
13
14
15
16
17
18
19
20
0
0
0
0
0
0
0
0
1
1
1
3
3
2
4
6
8
11
14
1
2
3
a Mean value of at least two 2 experiments. b inactive monomer (i).
c mixed dimer with one active (a) and one inactive monomer (i).
acetylated ornithine moiety to stay as close as possible to the
acylated motif present in the dimers. In 7, the less important
arginine was exchanged by ornithine to conserve the number of
carbon atoms in the side chain and simultaneously generate an
amine that allows acylation and dimerization having an IC50 of
6 nM. Modification of the ornithine side chain by acylation
impairs the binding affinity so that 8 and 9 are more than 10-fold
less active than 7.
The dimers 10ꢀ20 were used to elucidate the optimal spacer
lengths, with 11 being an exception as it contains one active and
one inactive peptide unit which was used to investigate the
binding mode. The optimal distance of the dimers is not well-
defined, with a broad range of 5ꢀ10 carbon atoms in the spacer
and an affinity of 2ꢀ4 nM. 15 has the best binding affinity with
2 nM, which is more than 30 times more active than the acetylated
monomer 8 and had 3-fold higher activity than the monomer 7
from which it was synthesized. The mixed dimer 11 has an IC50
of about 20 nM, which is almost in the middle of the affinity range
between the less affine monomers 8 and 9 and the better dimers
with two active peptide moieties. The dimers with an additional
glycine spacing unit 18ꢀ20 do not reach the activity level of the
most active dimers but are still 10 times more affine than the
acylated monomers.
In parallel with biodistribution studies, PET imaging of 68Ga-25
was performed to visualize the uptake in the various organs. The
PET images were in accordance with the data from the ex vivo
biodistibution. As revealed by organ analyses, high accumulation
of 68Ga-25 was obtained in the liver, however, still allowing
visualization of the tumor.
’ DISCUSSION
To compare the effect of dimerization, compounds 8 and 9
were synthesized having an acylated amine at the ornithine Nδ of
the monomer. Compound 8 is less affine by a factor of 10 in
comparison to 7, with the free amine showing that the basic
group at this position has positive effects, even so, it is the most
unimportant side chain. To be able to exclude effects of the
spacing unit toward binding affinity, compound 9 was synthe-
sized carrying a glutaric amide at the ornithine side chain. This
compound shows even less affinity toward CXCR4 than the
7651
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
Table 2. Dimeric DOTA Labeled Peptides and Their IC50 Valuesa
compd
X
n
M
IC50 [nM]
21
22
23
24
25
26
27
28
0
0
1
1
0
0
1
1
98 ( 13
76 ( 22
31 ( 14
81 ( 8
39 ( 2
15 ( 3
17 ( 7
13 ( 5
In3+
In3+
In3+
In3+
Gly
Gly
Gly
Gly
a Mean value of at least 3 experiments.
Table 3. Biodistribution of 68Ga-25 in OH1 Bearing-Tumor
Nude Mice 60 min pia,b
acetylated peptide, demonstrating that the spacer itself does not
contribute beneficially to binding affinity (Table 1).
In comparison with the acetylated monomer 8, all dimers,
connected by a diacid with less than 16 carbon atoms, show an
enhanced affinity (Table 1). Dimer 10 spaced by glutaric acid has
a more than 10-fold higher affinity than monomer 9 with just
glutaramide attached, showing that not the spacer but the second
peptide unit is responsible for enhanced affinity. The optimal
distance between the two peptides has a broad spacer range of
5ꢀ10 carbon atoms with an IC50 of 2ꢀ4 nM and no distinct
minimum, suggesting some flexibility of the dimers. However,
when the distance is further increased to 13 or 16 atoms, the
affinity becomes lower.
We further investigated if two identical binding pockets are
addressed by the dimers by exchanging an important pharma-
cophoric group of the monomer. Arg3 of 8 was substituted by
L-citrulline (Cit), which leads to a complete loss of affinity of the
monomer 6. However, the combination of one active with one
inactive cyclopentapeptide gives the asymmetric dimer 11,
which is less affine than the corresponding symmetric dimer
10 but exhibits more than three times better affinity than
compound 9, which carries the same spacing group but without
a second peptide unit. Hence, the guanidine group of the
second peptide is beneficial for binding affinity but not as
essential as in the first binding pocket. This in turn shows that
both binding pockets are not identical, as the second one can
also bind cyclic pentapeptides but does not require the same
unblockeda
blockeda,b
organ
blood
(n = 5)
(n = 1)
1.87 ( 0.29
1.14 ( 0.19
2.11 ( 0.27
44.31 ( 5.56
0.81 ( 0.29
3.98 ( 0.60
3.26 ( 0.51
1.48 ( 0.20
2.1
2.5
heart
lung
11.0
43.0
1.4
liver
pancreas
spleen
15.6
34.8
2.6
kidney
adrenal
glands
stomach
intestine
muscle
1.04 ( 0.39
1.30 ( 0.42
0.37 ( 0.07
2.08 ( 0.48
1.81 ( 0.22
0.05 ( 0.01
0.63 ( 0.09
4.1
3.0
1.0
OH1 tumor
tumor/heart
tumor/liver
tumor/
kidney
1.1
0.5
0.02
0.03
tumor/
5.96 ( 2,36
0.5
muscle
a Data are expressed as % ID/g tissue ( SD. b Blocking was achieved by
coinjection of 100 μg/mouse of a natGa-tracer (compound 2c in ref 31).
7652
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
Scheme 5. Schematic Comparison of Polyphemusin II (Upper Part) and a Dimeric Ligand (Lower Part)a
a Aromatic amino acids are highlighted in red and basic in blue. The boxed regions of polyphemusin II are assumed to correspond to the arrangement of
the pharmacophoric groups in the dimer in this model.
pharmacophoric groups. This also shows that the enhancement
of affinity is not due to the increase of local concentration of
active peptide monomers because the mixed dimer 11 should
then have an affinity similar to monomer 9 with just the
glutaramide spacing unit. Additionally, the only slightly higher
affinity of dimers in comparison with the monomer 8 would
rather fit with a binding mode of the second peptide moiety to a
subsite on the receptor than to the simultaneous binding of two
identical binding pockets.10,20
An explanation for this behavior is revealed when the origins of
the starting cyclic pentapeptide 4 and its derivatives used for
dimerization are considered (Scheme 5). Both polyphemusin II
and its shortened analogue T140 are rich in basic and aromatic
amino acids of which the four most important residues were
combined with a glycine to yield 4 (three of these residues are
boxed on the left side of Scheme 5).23 Therefore, the simplest
explanation why dimerization leads to a better affinity for
derivatives of polyphemusin II is that the surrounding area of
the initially addressed binding pocket of 4 has additional sites for
beneficial interaction with aromatic and basic groups which are
addressed by the second peptide unit of the dimer with a more
unspecific binding mode. This hypothesis is further supported by
many other CXCR4 antagonists that typically consist of at least
one basic and one aromatic moiety.37 Therefore peptides 8 and 9
are less affine as they cannot address these other binding sites for
additional beneficial interaction because of the missing second
peptide unit. This binding mode with one main binding pocket
and a smaller contribution to receptor affinity from the second
peptide unit with a subsite differs from the binding mode of 1. In
this case, both cyclen moieties are essential for affinity as the
monomer does not show any affinity.3
To further support this hypothesis, additional docking studies
were performed. Analysis of the recently published X-ray crystal
structures of CXCR4 receptor bound to an antagonist small
molecule IT1t (PDB codes 3ODU, 3OE6, 3OE8, and 3OE9)
and a cyclic peptide CVX15 (PDB code 3OE0) reveals that the
receptor conformations are substantially identical with CVX15
occupying the great majority of the binding cleft and the low-
molecular-weight IT1t ligand occupying just a portion of the
same site.19 Because the CVX15 and our compounds are both
cyclic analogues of the horseshoe crab peptide polyphemusin II,
molecular docking studies of the NMR solution structure of
monomeric cyclopentapeptides 7 (see Experimental Procedures
for further details) were attained on the structure with PDB code
3OE0.
In line with the rational design behind the synthesis of
polyphemusin II-mimic small peptide 7, the best binding pose
suggested by the docking software Glide strongly resembles the
experimental binding mode of CVX15 (Figure 1a).38 In parti-
cular, the ligand Tyr1 makes contacts with F189, Y190, and
V196, while the Orn2 residue points outward establishing an
ionic interaction with D187 in extracellular loop II (ELII)
(Figure 1b). The latter interaction explains why its acetylated
analogue 8 and compound 9, featuring a glutaric amide at the its
side chain, are less potent binders than 7 (Table 1). Notably, the
projection of Orn2 toward the external part of the receptor is also
consistent with the choice of this residue as the attachment point for
the synthesis of dimeric compounds (10ꢀ20). Other important
interactions are observed for Arg3 residue, which establishes a
7653
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
Figure 1. Binding conformation of 7 in the CXCR4 crystal structure as calculated by using Glide. The receptor is represented as light-green sticks and
ribbons. The cocrystal polyphemusin analogues (a) and the ligand (a and b) are represented as magenta and orange sticks, respectively. H-bonds are
represented as yellow dashed lines. For clarity reasons, only interacting residues are shown.
compounds 10 would bind site two in the same fashion as that
of the corresponding monomeric form (see above). Thus, the
dimeric compound 10 was constructed starting from the
calculated 7/CXCR4 complex by adding the corresponding
linker and the second cyclopentapeptide (Table 1). The built
complex was then subjected to a Monte Carlo search of all the
energetically feasible conformations of both the linker and
peptidic portions in the receptor context. This simulation
resulted in two main conformation families in which the peptide
occupying site one alternatively points toward ELII, TMII, and
TMII or toward TMIV and TM5 (Figure 2). In both cases, the
intrinsic flexibility of the CXCR4 site one would suggest that
this receptor region could plastically adapt to the different
dimeric cyclopentapeptides depending on their linker length
(see Table 1). Additionally, in both suggested conformations,
the Arg3 residue of the peptide occupying site one forms
Coulombic interactions with negatively charged regions of the
receptor, and this would be in line with SAR data, demonstrat-
ing that substitution of Arg3 with L-citrulline leads to a loss in
affinity (compound 11).
DOTA moieties were introduced via trifunctional linkers to
enable molecular imaging and endoradiotherapeutic uses. Their
corresponding indium chelates were prepared as the structure of
these complexes is closer to the finally desired (radio-)pharma-
ceuticals than the nonchelated DOTA compounds. Additionally,
previous studies on CXCR4 antagonists have already shown the
importance of metal ion chelates toward binding affinity in
comparison with their nonchelated analogues.16,40 All eight pos-
sible permutations of the peptide with and without an additional
glycine spacer, the two linkers, as well as the free DOTA unit, and
its indium chelate were synthesized and tested for their affinity
(Table 2).
As expected, the dimers lose affinity through the introduction
of the DOTA moiety but are suited for imaging and treatment of
CXCR4 related diseases. For every pair of metal-chelated and
nonchelated dimer, there is a significant difference in binding
affinity showing the importance of metal ions which is often
observed.31,41 With exception of the dimers 23 and 24, the
indium chelates show better affinity, which also supports our
hypothesis for the binding mode that positive charges are
beneficial toward receptor binding. Furthermore, the DOTA
conjugated dimers with an additional glycine spacer have the best
Figure 2. Alternative binding conformations (a and b) of 10 in the
CXCR4 binding site. The ligand is represented as orange sticks, while
receptor as light-green ribbons and a surface are colored according to its
electrostatic potential from red (negative) to blue (positive).
H-bond with T117 and an ionic interaction with D171 while the
adjacent Nal4 residue is embedded in a hydrophobic pocket made
up by Y190, F199, Q200, and H203 (Figure 1b).
Subsequently, molecular modeling studies were also per-
formed to unravel the binding mode of the dimeric compounds
10. Because the CXCR4 receptor was crystallized in a dimeric
form, we first investigated whether our dimeric compounds could
occupy the two binding sites of the receptor dimer. Such an
hypothesis was immediately confuted by the distance between the
two sites (≈ 40 Å), which cannot be spanned even by dimer 17,
which bears the longest linker. Truly, this was already suggested
by the binding data, which demonstrated that compounds
10ꢀ20 do not bind two identical receptor clefts. The only
other possibility is provided by the existence of the two
neighboring binding sites in the CXCR4 monomer (namely
site one and two) as previously suggested by NMR experiments
and radioligand binding assays and recently confirmed by X-ray
crystallography.12,39 While site two is in the fully solved trans-
membrane bundle region, site one is located in the extracellular
portion (N-terminus and extracellular loops), which unfortu-
nately has been only partially solved. Thus, while the interac-
tions of our peptides with site two can be exactly modeled, only
a general picture of the binding of 10 to site one can be depicted.
Beside the above-described intrinsic difficulties, docking calcu-
lations of the dimeric ligands would be highly error prone due to
the extreme flexibility of both the compounds and of the N-term-
inal fragment. Therefore, in our modeling studies, we started from
the assumption that one cyclopentapeptide (Table 1) in
7654
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
’ CONCLUSION
Starting from an acetylated monomeric CXCR4 antagonist
with an IC50 of 60 nM, several peptidic dimers were synthe-
sized by optimizing the distance between the monomers
which enhanced the affinity 30-fold. A possible explanation
for the enhanced binding affinity is a subsite binding of the
second peptide unit near to the main binding pocket. This is
suggested by our structural elucidation of the pharmacophoric
groups in the dimers and a comparison of the dimeric struc-
ture with the 18 amino acid containing polyphemusin II,
which served for the development of the original cyclic
pentapeptide. This hypothesis could be further supported by
molecular modeling using the recently published crystal
structure of CXCR4. Eight DOTA labeled dimers were
synthesized using two specially designed linker building
blocks. By introduction of the DOTA moiety in combination
with a metal-ion, dimers with a high potential for molecular
imaging and endoradiotherapeutic purposes were designed.
Subsequently 68Ga-25 was synthesized and evaluated in
respect to its imaging characteristics which showed subopti-
mal biodistribution due to its high lipophilicity.
’ EXPERIMENTAL PROCEDURES
Chemicals and Instruments. All commercially available chemical
reagents were used without further purification. Technical solvents were
distilled before use.
Tritylchloride-polystyrene-resins (TCP resins) were purchased from
PepChem and amino acid derivatives from Iris Biotech GmbH, Nova-
Biochem, Merck, Bachem, Neosystem, and Aldrich, while all other che-
micals were bought from Aldrich, Fluka, and Merck if not stated otherwise.
N-Methylpyrrolidone (NMP) was obtained from BASF and used
without further distillation. Dry solvents were purchased from Aldrich,
Fluka, and Merck. Dry dichloromethane was distilled from calciumhy-
dride under argon and kept over 4 Å molecular sieve. Water for RP-
HPLC was filtered through a 0.22 μm filter (Millipore, Millipak40).
RP-HPLC analyses were performed using using an Amersham
Figure 3. (A) PET dynamic imaging (0ꢀ110 min pi) of OH1 tumor
bearing nude mouse using 68Ga-25, (maximum intensity projection,
MIP). (B) Timeꢀactivity curves for tumor, heart, liver, kidney, and
muscle.
affinities as the distance between the two monomers corresponds
to the dimer with the best affinity 15. The affinities of the two
best compounds 26 and 28 are not only about 4-fold higher than
the acylated monomer 8 but they are additionally suited for
imaging purposes showing the beneficial impact of the dimeric
approach.
The main goal of this study was the synthesis of a library of
peptides with high affinity toward CXCR4 with a view to use
them as potential radiotracers for noninvasive molecular detec-
tion of primary tumors as well as targeted metastases. 25 was
labeled with 68Ga and evaluated in first in vivo studies in tumor
bearing mice. Although a tumor uptake of up to 2% ID/g
(Table 3) allowed visualization of the human tumor xenograft
by PET imaging, the high liver accumulation of 68Ga-25 renders
it an unsuitable radiotracer for the detection of primary tumors
and their metastases, particularly in the liver and the surrounding
organs (Figure 3).
In summary, we succeeded in the development of DOTA
labeled multimeric peptides that bind with high affinity to CXCR4
receptors. Initial in vivo studies with a 68Ga-labeled dimer for
mapping CXCR4 receptors that are directly involved in organ
specific metastasis allowed delineation of a CXCR4-expressing
human xenograft but also demonstrated high unspecific uptake
by the liver. Although this phenomenon was already observed
for other radiolabeled CXCR4, ligands it is currently unclear
whether this unsuitable behavior could be overcome by a further
increase of the hydrophilicity of the dimer.
€
Pharmacia Biotech Akta Basic 10F equipped with an an Omnicrom
YMC column (4.6 mm ꢁ 250 mm, 5 μm C18, 1 mL/min). The eluent
was a linear gradient from water (0.1% trifluoroacetic acid (TFA)) to
acetonitrile (ACN; 0.1% TFA) over 30 min and detection at 220 and
254 nm. The retention time (Rt) of the analytical RP-HPLC is given in
min, with the gradient in percentage of acetonitrile. Purities were
determined at 220 nm with the Unicorn software package and are given
relative to their starting compound.
Semipreparative RP-HPLC was done on a Beckman System Gold
equipped with high pressure module 125, UV-detector 166, and using an
Omnicrom ODS-A C18 (120 Å, 5 μm, 250 mm ꢁ 20 mm) column in
combination with the same solvents as stated above.
NMR spectra were recorded on a Bruker Avance 250 or Bruker DMX
500 at 298K. The chemical shifts are reported in ppm on the δ scale
relative to the solvent signal used. 13C NMR-spectra were recorded using
1H-broad band decoupling. Pulse programs were taken from the Bruker
library or written by members of our group. Samples were prepared in
tubes with a diameter of 5 mm using 0.5 mL of deuterated solvent with a
final concentration of approximately 20ꢀ50 mM. The resulting spectra
were processed on a workstation using Bruker TOPSPIN 1.3 software.
ESI mass spectra were recorded on a Finnigan LCQ in combination
with a Agilent/HP 1100 RP-HPLC system using a Omnicrom YMC
ODS-A C18 column (120 Å, 3 μm, 125 mm ꢁ 2 mm) with a flow rate of
0.2 mL/min. The eluent was a linear gradient from water to acetonitrile
with 0.1% formic acid over 20 min with detection at 220 nm.
7655
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
All tested compounds exhibited g95% purity determined with RP-
HPLC-(MS).
Removal of Acid Labile Side Chain Protecting Groups.
Cyclized peptides were stirred in a solution of TFA, water, and TIPS
(95:2.5:2.5) at RT for 1 h or until no more protected peptide could be
observed by ESI-MS and precipitated in diethylether and washed two
more times.
Removal of DOTA t-Bu Groups. To the coupling solution with
the dimerized peptides the same volume of conc HCl was added on an
ice bath under vigorous stirring. The deprotection was carried out at RT
and monitored for completeness by ESI-MS every 30 min and stopped
by neutralizing with conc NH4OH on an ice bath.
Chelation of Indium with DOTA Ligands. The solution with
the deprotected DOTA-dimers was treated with InCl3 (5 equiv)
dissolved in 5 M aqueous NH4Cl of the same volume as the total
deprotection solution. After 15 min of stirring at RT, the solution was
subjected to HPLC purification.
Loading of TCP-Resin. Peptide synthesis was carried out using
TCP-resin (0.9 mmol/g) following standard Fmoc-strategy. Fmoc-
Xaa-OH (1.2 equiv) were attached to the TCP resin with
N,N-diisopropylethylamin (DIEA; 2.5 equiv) in anhydrous DCM
(0.8 mL/g resin) at room temperature for 1 h. The remaining trityl
chloride groups were capped by addition of 1 mL/g (resin) of a solution
of MeOH, DIEA (5:1; v:v) for 15 min. The resin was filtered and washed
5 times with DCM and 3 times with MeOH. The loading capacity was
determined by weight after drying the resin under vacuum and ranged
from 0.4 to 0.9 mmol/g.
On-Resin Fmoc Deprotection. The resin-bound Fmoc peptide
was treated with 20% piperidine in NMP (v/v) for 10 min and a second
time for 5 min. The resin was washed 5 times with NMP.
Standard Amino Acid Coupling. A solution of Fmoc-Xaa-OH
(2 equiv), 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium tetra-
fluoroborate (TBTU) (2 equiv), hydroxybenzotriazole (HOBt; 2 equiv),
and DIEA (5.2 equiv) in NMP (1 mL/g resin) was added to the resin-
bound free amine peptide and shaken for 60 min at room temperature
and washed 5 times with NMP.
o-2-Nitrobenzenzesulfonyl (nosyl, o-Ns) Protection. A so-
lution of o-Ns-Cl (5 equiv) and collidine (10 equiv) in NMP (1 mL/g
resin) was added to the resin-bound free amine peptide and shaken for
15 min at room temperature. The resin was washed 3 times with NMP
and 3 times with dry THF.
N-Methylation under Mitsunobu Conditions. A solution of
triphenylphosphine (5 equiv), diisopropyl azodicarboxylate (DIAD;
5 equiv) and MeOH (10 equiv) in dry THF (1 mL/g resin) was added
to the resin-bound o-Ns protected peptides and shaken for 10 min at
room temperature. The resin was filtered off and washed 3 times with
dry THF and 3 times with NMP.
On-Resin o-Ns Deprotection. For o-Ns deprotection, the resin-
bound o-Ns-peptides were stirred in a solution of mercaptoethanol (10
equiv) and DBU (5 equiv) in NMP (1 mL/g resin) for 5 min. The
deprotection procedure was repeated one more time, and the resin was
washed 5 times with NMP.
Amino Acid Coupling to Hindered Amines. A solution of
Fmoc-Xaa-OH (2 equiv), 2-(7-aza-1H-benzotriazole-1-yl)-1,1,3,3-tetra-
methyluronium hexafluorophosphate (HATU; 2 equiv), 1-hydroxy-7-
azabenzotriazole (HOAt; 2 equiv), and DIEA (4 equiv) in NMP(1 mL/g
resin) was added to the resin-bound peptides and shaken for 3 h at room
temperature and washed 5 times with NMP.
Allyloxycarbonyl (Alloc) and Allyl Deprotection. Pd(PPh3)4
(0.125 equiv) in dry DCM (0.5 mL/g resin) was added to the resin-
bound Alloc peptide followed by an addition of phenylsilane in dry
DCM (0.5 mL/g resin) and shaken for 1 h. The resin was washed 5 times
with DCM.
Peptide Cleavage from the Resin. For complete cleavage of the
peptides from the resin, they were treated three times with a solution of
DCM and hexafluoroisopropanol (HFIP; 4:1; v:v) at room temperature
for half an hour and the solvent evaporated under reduced pressure.
Cyclization. To a solution of peptide in DMF (1 mM peptide
concentration) and NaHCO3 (5 equiv) diphenylphosphoryl azide
(DPPA; 3 equiv) was added at room temperature (RT) and stirred
overnight or until no linear peptide could be observed by ESI-MS. The
solvent was evaporated to a small volume under reduced pressure and
the peptides precipitated in saturated NaCl solution and washed two
times in HPLC grade water.
Receptor Binding Assays. Competition studies were performed
on Jurkart cells using cyclo(-D-Tyr1[125I]-Arg2-Arg3-Nal4-Gly5) (125I-
CPCR4) as radioligand. In brief, cells were resuspended in PBS/0.2%
BSA. A total of 200 μL of the suspension containing 400000 Jurkat cells
were incubated with 25 μL of the tracer solution (containing 3.1 kBq,
approximately 0.1 nM) and 25 μL of the test peptides at different
concentrations of 10ꢀ11 to 10ꢀ5 M. Nonspecific binding was determined
in the presence of 1 μM cold cyclo(-D-Tyr1Arg2-Arg3-Nal4-Gly5). After
shaking for 2 h at RT, the incubation was terminated by centrifugation at
1300 rpm for 5 min. Cell pellets were washed twice with cold PBS. Cell
bound radioactivity was determined by using a 1480 Wizard3 gamma-
counter from Wallac (Turku, Finland). Experiments were repeated 2ꢀ3
times in triplicate. IC50 values of the compounds were calculated by
nonlinear regression using GraphPad Prism (GraphPad Prism 4.0 Soft-
ware, Inc., San Diego, CA, USA). Each data point is the average of three
determinations.
Labeling with 68Ga. 68Ga-25 was selected for an initial in vivo
study. For this purpose, Gallium-68 (e+ = 89%, t1/2 = 68.1 min, Eβ+max
=
1.90 MeV) was eluted from a commercially available Ge-68/Ga-68
generator (iThemba, South Africa) by diluted HCl. The fraction with the
highest activity, approximately 1.2 mL with >80% of the entire eluted
activity, was used for labeling of 25 using a commercially available fully
automated labeling module (Gallelut-Synthesizer, Scintomics GmbH,
F€urstenfeldbruck, Germany). After the pH of the eluate was adjusted
with a suitable amount of HEPES (4-(2-hydroxyethyl)-1-piperazi-
neethanesulfonic acid) (930 μL, 600 mg HEPES/0.5 mL H2O), 35 μg
(15 nmol) of the peptide was added. After reaction for 5 min at 95 ꢀC,
quality control of the product was carried out by radio thin layer
chromatography (TLC) on Silica gel 60-plates using 0.5 μL of product
solution on two different TLC systems: (a) eluent TLC-1, 0.1 M sodium
citrate (5.89 g trisodium citrate dihydrate in 200 mL ultrapure water);
eluent TLC-2, (1/1, v/v) methanol/1 M ammonium acetate (15.46 g
ammonium acetate in 200 mLultrapure water). Using the TLC method 1,
the product and Ga-colloid stay at the starting point, whereas free Ga3+
moves with front. Using the TLC method 2, uncomplexed Ga-species
stay at starting point, whereas the labeled peptide moves with the front.
The quality control of the labeled peptide was also achieved by RP-
HPLC. HPLC analysis was performed on a Semi RP18 Multosphere
Column (250 mm ꢁ 100 mm) applying a linear gradient system at a
5 mL/min flow rate from 49% B to 60% B in 20 min, where solvent A =
aqueous ammonium formate 0.2 M and solvent B = MeOH. The
detection of the peptides was performed via a UV detector at 220 nm.
Biosdistribution Studies. Athymic nude mice (approximately 30 g)
were obtained from Charles River, Germany. The animals were inocu-
lated subcutaneously into the right flank with OH1 h-SCLC cells (5 ꢁ
106 cells/animal). Tumors were allowed to grow for three weeks. Tissue
distribution studies of the 68Ga labeled tracer carried out after the
intravenous administration of 0.1 mL [∼70 μCi (∼2.6 MBq), 0.7 μg of
total peptide] of the radiolabeled product via the tail vein and the
Dimerization by Acylation in Solution. Fully deprotected
peptides were stirred with HATU (1.1 equiv) and DIEA (2.2 equiv)
and the corresponding acid (1 equiv) in DMF (10 mM peptide
concentration) for 30 min at RT. The solution was directly purified
by HPLC separation.
7656
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
animals (n = 5) were sacrificed 1 h pi. Subsequently, the tissues and
organs were weighted, the radioactivity was counted in a 1480 Wizard3
gamma-counter from Wallac (Turku, Finland), and the % ID/g of each
organ or tissue was calculated. Specificity of binding was demonstrated
by coinjection of a natGa-tracer (compound 2c, ref 31) (100 μg/mouse;
n = 1).31
47, 28, 26. Rt (10ꢀ100%): 21.9 min. m/z calcd for C24H26N2O6, 438.18;
found, 461.5 [M + Na+].
(9H-Fluoren-9-yl)methyl (R)-1-(Carbonyl)-3-(tert-butoxycarbonyl)-
propan-2-ylcarbamate. The product was obtained similar to a general
procedure given in literature.35 Nα-Fmoc-aspartic acid(tBu)OH (2.0 g;
4.9 mmol; 1equiv) was dissolved in anhydrous THF (20 mL). NEt3
(0.74 mL; 5.4 mmol; 1.1 equiv) and ethylchloroformate (0.52 mL; 5.4
mmol; 1.1 equiv) were added sequentially at ꢀ15 ꢀC. Stirring was
continued for 15 min, and then the solution was allowed to warm up to
0 ꢀC. In the meantime, N-methylnitrosourea (2.5 g; 24.3 mmol; 5 equiv)
was stirred in ice-cold Et2O (20 mL) and 40% KOH (20 mL; ice-cold)
was added dropwise until complete dissolution. The yellow diazo-
methane solution in Et2O was added dropwise at 0 ꢀC to the amino
acid solution, and it was then allowed to warm up to RT and stirred for
another 2.5 h. Excess diazomethane was decomposed by dropwise
addition of HOAc. The solution was washed with satd NaHCO3, satd
NH4Cl, and brine. The organic layer was dried (Na2SO4) and evapo-
rated under reduced pressure. The resulting diazo ketone was dissolved
in water/dioxane (1:5; v/v; 160 mL). After addition of silver benzoate
(0.12 g; 0.5 mmol; 0.1 equiv), the mixture was sonicated in an ultrasound
bath until complete conversion (30 min) monitored by TLC (MeOH/
DCM, 1:20; Rf, 0.1ꢀ0.2). After evaporation of dioxane under reduced
pressure, the solution was acidified with 5% HCl and the precipitate
extracted with EtOAc (three times). The organic layer was dried
(Na2SO4) and evaporated under reduced pressure and the crude
product purified by flash chromatography (MeOH/DCM, 1:20; Rf,
PET-Camera Imaging. Mice (∼30 g) were anaesthetized using
isoflurane anesthesia and injected with 80ꢀ90 μCi (∼3 MBq, 1.4 μg of
total peptide) of 68Ga-25 via the tail vein in a volume of 300 μL of PBS.
For the blocking experiments, mice were coinjeced with 50 μL (100 μg)
of natGa-tracer (compound 2c, ref 31) complex in a total volume of
300 μL. PET scans were performed using an Inveon Siemens PET
scanner, and dynamic imaging was performed immediately after the
time of injection with acquisition times: 5 ꢁ 60 s, 5 ꢁ 300 s, 6 ꢁ
600 s, 1 ꢁ 120 s, and a total duration being 110 min (Figure 3A). Analysis
of tracer kinetics was performed by drawing a circular region of interest
(ROI) in the tumor, heart, liver, kidney, and muscle. The average of the
accumulated radioactivity in the ROI was calculated in Bq/mL and time
activity curves were plotted, representing the radioactivity in the organs of
interest versus time (Figure 3B). For blocking experiments, dynamic
images were obtained using acquisition times of 5 ꢁ 60 s, 5 ꢁ 300 s,
and 1 ꢁ 600 s, with a overall sorter imaging period of 40 min. All images
were reconstructed by a two-dimensional ordered subsets expectation
maximum (2D-OSEM) algorithm, and no correction was applied for
attenuation. Images analysis was done using Inveon software. The results
were calculated as Bq/mL.
1
0.1ꢀ0.2) to yield 1.3 g (3.1 mmol; 63% yield). H NMR (250 MHz,
Solution Synthesis. General Amine Protection Procedure. To a
0.2 M solution of amino acid and Na2CO3 (0.5 M), the same volume of a
0.2 M reagent solution in THF was added and stirred at RT for 1 h. The
THF was evaporated under reduced pressure, the aqueous phase washed
once with diethylether and acidified with conc HCl to pH 1, and the
product extracted with EtOAc. The combined organic layers were dried
(Na2SO4), filtered, concentrated, and dried in vacuo.
DMSO-d6): δ 12.2 (s, br, 1H,), 7.90 (d, 2H), 7.69 (dd, 2H), 7.42 (t, 2H),
7.33 (m, 3H), 4.27 (m, 3H), 3.59 (m, 1H), 2.41 (m, 4H), 1.38 (s, 9H).
13C NMR (75 MHz, DMSO-d6): 172.49, 170.24, 144.35, 141.19,
128.07, 127.51, 125.63, 120.56, 80.39, 65.80, 60.20, 47.17, 45.81,
28.13. Rt (10ꢀ100%): 23.5 min. m/z calcd for C24H27NO6, 425.18;
found, 448.1 [M + Na+].
Boc Deprotection. First 3ꢀ8 mmol of Boc protected amino acid were
dissolved in 10 mL of DCM and 5 mL of TFA added slowly. Then the
solution was stirred at RT for 45 min and the solvent evaporated in vacuo
to yield the crude product ready for reprotection.
(9H-Fluoren-9-yl)methyl (R)-1-((Allyloxy)carbonyl)-3-(tert-butoxy-
carbonyl)propan-2-ylcarbamate. The product was obtained similar
to a procedure given in literature.42 (9H-Fluoren-9-yl)methyl (R)-1-
(carbonyl)-3-(tert-butoxycarbonyl)propan-2-ylcarbamate (0.98 g; 2.3
mmol; 1 equiv) was stirred with allyl bromide (5.52 mL; 6.4 mmol;
2.8 equiv) and DIEA (0.78 mL; 4.6 mmol; 2 equiv) in ACN (4.6 mL)
at 45 ꢀC for 1 h. The reaction was monitored by TLC MeOH/DCM
(1:20; v/v). The solution was allowed to reach RT. After addition of
ethylacetate (EtOAc; 20 mL), the organic layer was washed with satd
KHSO4, satd NaHCO3, and half-satd NaCl, dried (Na2SO4), and
evaporated under reduced pressure to yield 0.77 g (1.7 mmol; 74%
yield). 1H NMR (250 MHz, CDCl3): δ 7.76 (d, 2H,), 7.58 (d, 2H), 7.40
(t, 2H), 7.31 (m, 2H), 5.91 (m, 1H), 5.63 (d, 1H), 5.32 (m, 1H), 5.24
(d, 1H), 4.60 (d, 2H), 4.36 (m, 3H), 4.21 (m, 1H), 2.66 (m, 4H), 1.45
(s, 9H). 13C NMR (75 MHz, CDCl3): 143.91, 141.29, 131.84, 127.67,
127.04, 125.08, 119.95, 118.57, 66.86, 65.38, 47.21, 45.13, 39.26, 38.08,
28.05. Rt (10ꢀ100%): 27.6 min. m/z calcd for C27H31NO6, 465.22;
found, 488.3 [M + Na+].
(9H-Fluoren-9-yl)methyl (S)-1-((Allyloxy)carbonyl)-3-(carbonyl)pro-
pan-2-ylcarbamate. The product was obtained similar to a procedure
given in literature.42 (9H-Fluoren-9-yl)methyl (R)-1-((allyloxy)carbonyl)-
3-(tert-butoxycarbonyl)propan-2-ylcarbamate (0.77 g; 1.65 mmol) was
dissolved in DCM (4 mL), and TFA (2 mL) was added and stirred for
2 h at RT. After evaporation to dryness, the solid was dissolved in satd
NaHCO3, washed with ether, and acidified (pH 2) with HCl (5%) to
form a white precipitate that was extracted two times with EtOAc. The
organic layer was washed with acidified water (HCl, pH 1), dried
(Na2SO4), and evaporated under reduced pressure to yield 0.58 g (1.4
mmol; 85% yield). 1H NMR (250 MHz, CDCl3): δ 7.75 (d, 2H,), 7.68
(s, 1H), 7.57 (d, 2H), 7.40 (t, 2H), 7.31 (m, 2H), 5.89 (m, 1H), 5.66 (d,
1H), 5.28 (m, 2H), 4.60 (d, 2H), 4.38 (m, br, 2H), 4.22 (m, 1H), 2.62
Nα-Ns-Nδ-Boc-D-Ornithine. Nδ-Boc-D-ornithine (1.51 g, 6.5 mmol)
was protected with o-nitrobenzenesulfonylchloride (NsCl) (1.44 g, 6.5
mmol) and gave a slightly yellow, sticky oil as sufficiently pure product
(2.43 g, 90%). 1H NMR (250 MHz, DMSO-d6): δ 12.72 (s, 1H), 8.46
(d, 1H), 8.01 (m, 1H), 7.93 (m, 1H), 7.86 (m, 2 H), 6.78 (t, 1 H), 3.85
(m, 1H), 2.85 (m, 2 H), 1.38 (m, 13H). 13C NMR (125 MHz, DMSO-d6):
172.3, 147.7, 134.4, 133.8, 132.9, 130.3, 124.5, 110.7, 77.9, 56.2, 29.7,
28.7, 26.4. Rt (10ꢀ100%): 18.6 min. m/z calcd for C16H23N3O8S,
417.12; found, 440.1 [M + Na+].
Nα-Alloc-Nδ-Boc-D-Ornithine. Nδ-Boc-L-ornithine (0.49 g, 2.1 mmol)
was protected with allyl chloroformate (0.22 mL, 2.1 mmol) and gave a
slightly yellow, sticky oil as sufficiently pure product (0.53 g, 80%). 1H
NMR (250 MHz, DMSO-d6): δ 12.52 (s, 1H), 7.49 (d, 1H), 6.78 (t,
1H), 5.91 (br m, 1H), 5.30 (dd, 1 H), 5.19 (dd, 1 H), 4.48 (m, 2H), 3.91
(br m, 1H), 2.91 (m, 2H), 1.81ꢀ1.40 (br m, 4H), 1.38 (s, 9H). 13C
NMR (63 MHz, DMSO-d6): 174.4, 156.5, 156.1, 134.1, 117.4, 77.9,
65.1, 60.2, 54.1, 28.8, 26.7, 14.6. Rt (10ꢀ100%): 16.7 min. m/z calcd for
C14H24N2O6, 316.16; found, 339.3 [M + Na+].
Nα-Alloc-Nδ-Fmoc-D-Ornithine. Nα-Alloc-Nδ-Boc-D-ornithine (0.36 g,
1.68 mmol) was subjected to Boc deprotection and subsequently re-
protected with Fmoc-OSu (0.567 g, 1.68 mmol) and gave a white foam
as sufficiently pure product (0.61 g, 89%). 1H NMR (500 MHz, DMSO-
d6): δ 12.5 (s, 1H,), 7.9 (d, 2H), 7.7 (d, 2H), 7.5 (d, 1H), 7.4 (t, 2H),
7.32 (t, 2H), 7.28 (m, 1H), 5.9 (m, 1H), 5.3 (d, 1H), 5.2 (d, 1H), 4.5 (d,
2H), 4.3 (d, 2H), 4.2 (t, 1H), 3.9 (m, 1H), 3.0 (d, 2H), 1.7 (m, 1H), 1.5
(m, 3H). 13C NMR (125 MHz, DMSO-d6): 174, 156.0, 155.9, 144, 141,
133, 128, 127.0, 126.9, 125.1, 125.0, 120.1, 119.9, 65, 64, 53.54, 53.50,
7657
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
(m, 4H). 13C NMR (75 MHz, CDCl3): 143.69, 141.33, 131.60, 127.77,
127.10, 125.00 120.01, 118.87, 67.25, 65.65, 47.13, 44.47, 37.87, 37.64.
Rt (10ꢀ100%): 22.0 min. m/z calcd for C23H23NO6, 409.15; found,
432.2 [M + Na+].
experiments (150 ms mixing time) were recorded. 8k and 512 data
points were recorded in the direct and indirect dimension,
respectively.51 For the E-COSY experiment, 2k data points were
recorded in the indirect dimension.
Nα-Fmoc-L-aspartic Acid (t-Bu) Allylester. The product was obtained
similar to a procedure given in literature.43 Nα-Fmoc-L-aspartic acid (β-t-
Bu ester) (0.103 g; 2.50 mmol; 1equiv) was stirred with allyl bromide
(6.0 mL; 7.0 mmol; 2.8 equiv) and DIEA (0.78 mL; 5.0 mmol; 2 equiv)
in ACN (5.0 mL) at 45 ꢀC for 100 min. The reaction was monitored by
TLC MeOH/DCM (1:20; v/v). The solution was allowed to reach RT.
After addition of EtOAc (40 mL), the organic layer was washed with satd
KHSO4, satd NaHCO3, and half-satd NaCl, dried (Na2SO4), and
evaporated under reduced pressure to yield 1.01 g (2.24 mmol; 90%
yield). 1H NMR (250 MHz, CDCl3): δ 7.74 (d, 2H,), 7.58 (m, 2H), 7.38
(t, 2H), 7.29 (t, 2H), 5.89 (m, 1H), 5.81 (d, 1H), 5.32 (d, 1H), 5.23 (d,
1H), 4.56 (m, 3H), 4.37 (m, 3H), 4.23 (m, 1H), 2.86 (m, 2H), 1.42 (s,
9H). 13C NMR (75 MHz, CDCl3): 170.62, 170.00, 155.98, 143.91,
143.73, 141.29, 131.49, 127.71, 127.07, 125.13, 119.98, 118.80, 81.89,
67.28, 66.30, 50.61, 47.10, 37.79, 28.03. Rt (10ꢀ100%): 27.8 min. m/z
calcd for C26H29NO6, 451.20; found, 474.3 [M + Na+].
Molecular Dynamics Simulation. The GROMACS 4.0.352 software
package (www.gromacs.org), was used to perform the MD simulations.
All scripts for subsequent analysis were packaged with GROMACS. The
OPLS-AA force field was used for the molecular dynamic simulations.
Rigid SPC water was the water model used. Remaining solute bonds were
constrained by the SHAKE algorithm and temperature and pressure
control was executed by Berendsen coupling.53 A cubic simulation box
with periodic boundary conditions was employed, along with PME
(rcoulomb = 1.1 nm) for electrostatic and with a cut-off distance (rvdw
=
1.1 nm) for Lennard-Jones nonbonding interactions. All of the MD
simulations were established and performed using the following proce-
dure. The molecules were built with SYBYL (SYBYL 7.3, Tripos
International, 1699 South Hanley Rd., St. Louis, Missouri 63144, USA),
placed at the center of a cubic simulation box, and a steepest descent
energy minimization in vacuo was conducted. The box was then solvated
with water molecules. Steepest decent energy minimization was used to
remove bad van der Waals contacts between atoms. This was followed by
a series of equilibration steps of 50 fs starting at 50 K (position restraints
at 250000) and going in 50 K steps toward 300 K (no position restraints)
while lowering the position restraints in each step by a factor of 10 except
for the last step, where it was lowered by a factor of 25. These short
simulation was performed with temperature coupling (τT = 0.1 ps; Tref is
equal to the simulated temperature) and no pressure coupling and used
the output of the previous step as input. An additional equilibration step
at 300 K with temperature coupling (τT = 0.1 ps; Tref = 300 K) and
isotropic pressure coupling (τP = 0.5 ps with reference pressure of
1.01325 bar and 4.5 ꢁ 10ꢀ5 compressibility) was done for 100 fs.
Following the short equilibrations the long simulations for analysis were
run for a total simulation time of 100ꢀ150 ns with otherwise the same
specifications as the last equilibration step.
Nα-Fmoc-L-aspartic Acid (OH) Allylester. The product was obtained
similar to a procedure given in literature.43 Nα-Fmoc-L-aspartic acid (β-t-
Bu) allylester (1.01 g; 2.23 mmol) was dissolved in DCM (4 mL), and
TFA (2 mL) was added and stirred for 1 h at RT. After evaporation to
dryness, the solid was dissolved in satd NaHCO3, washed with ether, and
acidified (pH 2) with HCl (5%) to form a white precipitate that was
extracted two times with EtOAc. The organic layer was washed with
acidified water (HCl, pH 1), dried (Na2SO4), and evaporated under
reduced pressure to yield 0.77 g; 1.94 mmol; 87% yield. 1H NMR (250
MHz, DMSO): δ 7.90 (m, 2H,), 7.83 (m, 1H), 7.70 (d, 2H), 7.42 (m,
2H), 7.32 (m, 2H), 5.87 (m, 1H), 5.30 (d, 1H), 5.18 (m, 1H), 4.58 (d,
2H), 4.45 (m, 1H), 4.26 (m, 3H), 2.70 (m, 2H). 13C NMR (75 MHz,
CDCl3): 171.85, 171.29, 144.22, 141.20, 132.72, 128.11, 127.540, 125.66
120.59, 118.03, 66.24, 65.52, 51.02, 47.07, 36.32. Rt (10ꢀ100%): 22.0
min. m/z calcd for C22H21NO6, 395.14; found, 418.1 [M + Na+].
Glutaric Acid Mono Amide (Sodium Salt). The ammonium salt was
synthesized via a modified protocol found in literature and then replaced
with sodium ions.44 Glutaric anhydride (0.50 g; 4.4 mmol) was dissolved
in DCM (10 mL) and insoluble glutaric acid filtered off. Aqueous
ammonium hydroxide was heated to 70 ꢀC and the gas bubbled through
the stirred DCM solution. After precipitation of a white solid, the
ammonia was heated for another 30 min and the DCM solution stirred
overnight. The white solid was filtered, washed two times with DCM
(10 mL), and dried to yield 0.58 g (3.9 mmol; 89%) of glutaric acid mono
amide ammonium salt. The ammonium salt was dissolved in water
(10 mL) and treated with an equimolar amount of sodium hydroxide
(0.16 g) and lyophilized to yield the sodium salt. 1H NMR (250 MHz,
DMSO-d6): δ 7.31 (s, 1H,), 6.63 (s, 1H), 2.03 (q, 4H), 1.65 (m, 2H).
1,4,7-Tris(tert-butoxycarbonylmethyl)-1,4,7,10-tetraazacyclodode-
cane-10-acetic Acid. (Tris(t-Bu)DOTA) was synthesized according to
procedures given in the literature.45
NMR Spectroscopy and Computational Studies. NMR
Spectroscopy. Nuclear magnetic resonance (NMR) experiments were
recorded with a 20 mM concentrated sample in aqueous 50 mM
acetate buffer pH 4.5 at 280K on a Bruker DMX600 spectrometer
(Bruker Biospin, Rheinstetten, Germany). Spectra were referenced
relative to external DSS.46 Water suppression was achieved with a
WATERGATE sequence.47 1H,1H-TOCSY (80 ms mixing time) and
1H,1HꢀCOSY experiments were used for chemical shift assignment of
all proton resonances using standard procedures.48 Sequential assign-
ment was accomplished by throughbond connectivities from hetero-
nuclear multibond correlation (HMBC) spectra.49 The 1H,1H E-COSY
experiment was used for the extraction of homonuclear coupling
constants.50 For the extraction of distance information, ROESY
Docking. The crystal structure with PDB code 3OE0 was downloaded
from the RCSB Protein Data Bank.
This file contains a single unit of the CXCR4 receptor cocrystallized
with the cyclic peptide CVX15, which is structurally similar with the
cyclopentapeptides antagonist used in this study. The 3OE0.pdb crystal
structure was prepared using the “Protein Preparation Wizard” panel of
Schr€odinger 2010 molecular modeling package.54 In particular, using the
“preprocess and analyze structure” tool, the bond orders were assigned,
all the hydrogen atoms were added, the disulfide bonds were assigned,
and all the water molecules in a distance greater than 5 Å from any
heterogroup were deleted. Using Epik 2.0 a prediction of the hetero-
groups ionization and tautomeric states was performed.54 An optimiza-
tion of the hydrogen-bonding network was performed using the “H-bond
assignment” tool. Finally, using the “impref utility”, the positions of the
hydrogen atoms were optimized by keeping all the heavy atoms in place.
Glide is a grid-based ligand docking with energetics approach and
searches for favorable interactions between ligands and receptors. The
shape and properties of the receptor are represented on a grid by
different sets of fields that provide progressively more accurate scoring of
the ligand pose. These fields are generated as preprocessing steps in the
calculation and hence need to be computed only once for each receptor.
For the grid generation of the CXCR4 receptor, the binding site was
defined using the native CVX15 ligand cocrystallized with the receptor.
This box gives a more precise measure of the effective size of the search
space. However, ligands can move outside this box during grid mini-
mization. The Cartesian coordinates of the inner box, X, Y, and Z length
were set to 30 Å.
These grids were used to dock both the trans and cis conformations
of peptide 7 as experimentally determined by NMR, distance geo-
metry, and subsequent molecular dynamics (restrained MD). The
7658
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
conformational space of the ligand is defined by Glide by several lowest-
energy poses that are subjected to a Monte Carlo procedure that
examines nearby torsional minima. This procedure is needed in some
cases to properly orient peripheral groups and occasionally alters
internal torsion angles. The default value (1.00) for the van der Waals
radii scaling factor was chosen, which means no scaling for the nonpolar
atoms was performed (no flexibility was simulated for the receptor). In
the present study, the extra-precision (XP) mode of GlideScore function
was used to score the obtained binding poses. The force field used for the
docking was the OPLS-2005.55
Conformational Search. The 7/CXCR4 complex obtained from
docking studies was used as a starting point to build the 10/CXCR4
complex, which was then subjected to conformational sampling calcula-
tions. In particular, the structure of 10 was built and optimized with
MacroModel within the Schr€odinger Maestro package, keeping also the
second cyclopentapeptide in its trans conformation. Macromodel supplies
eight search algorithms: torsional sampling, serial torsional sampling,
systematic torsional sampling, mixed torsional/low-mode sampling, low-
mode sampling, serial low-mode sampling, large-scale low-mode sam-
pling, and mixed torsional/large-scale low-mode sampling. Among them,
the large-scale low-mode sampling method was used as it implements
techniques to reduce the memory requirements so thatit can be applied to
large systems such as proteinꢀligand complexes. The calculations were
performed with OPLS-2005 force field in a water solvent model. The
energy window for saving structures was set to 5.02 kcal/mol. Each
conformational search included 10000 iterations. The rmsd cutoff value
was set at 0.5 Å to avoid retrieving redundant conformations. When each
search was finished, 100 representative ligand/receptor conformations
were retained by the ligand heavy-atom rmsd analysis. In this way, lower
energy and relatively diverse ligand conformation ensembles were generated.
Pictures of the modeled ligand/receptor complexes were rendered with the
UCSF Chimera package from the Resource for Biocomputing, Visualization,
and Informatics at the University of California, San Francisco.56
diphenylphosphoric acid azide; DOTA, (1,4,7,10-tetraazacyclo-
dodecane-1,4,7,10-tetraacetic acid); GPCR, G-protein coupled
receptor; HATU, O-(7-Azabenzotriazol-1-yl)-N,N,N0,N0,-tetra-
methyluroniumhexafluorophosphat; HFIP, hexafluoroisopropa-
nol; HIV, human immunodeficiency virus; HOAt, 1-hydroxy-7-
azabenzotriazol; HOBT, 1-hydroxybenzotriazol; MIP, maximum
intensity projection; Nal, L-3-(2-naphthyl)alanine; NMP, N-methyl-
2-pyrrolidone; NMR, nuclear magnetic resonance; Ns, Nα-
2-nitrobenzenesulfonyl; NsCl, Nα-2-nitrobenzenesulfonylchlor-
ide; PET, positron emission tomography; SPPS, solid phase
peptide synthesis; TBTU, O-(1H-benzotriazol-1-yl)-N,N,N0,N0-
tetramethyluroniumtetra-fluoroborate; TFA, trifluoroacetic acid;
TIPS, triisopropylsilane
’ REFERENCES
(1) (a) Nagasawa, T.; Kikutani, H.; Kishimoto, T. Molecular cloning
and structure of a pre-B-cell growth-stimulating factor. Proc. Natl. Acad.
Sci. U.S.A. 1994, 91, 2305–2309. (b) Oberlin, E.; Amara, A.; Bachelerie,
F.; Bessia, C.; Virelizier, J. L.; Arenzana-Seisdedos, F.; Schwartz, O.;
Heard, J. M.; Clark-Lewis, I.; Legler, D. F.; Loetscher, M.; Baggiolini, M.;
Moser, B. The CXC chemokine SDF-1 is the ligand for LESTR/fusin
and prevents infection by T-cell-line-adapted HIV-1. Nature 1996,
382, 833–835. (c) Tashiro, K.; Tada, H.; Heilker, R.; Shirozu, M.;
Nakano, T.; Honjo, T. Signal Sequence Trap—A Cloning Strategy for
Secreted Proteins and Type-I Membrane-Proteins. Science 1993, 261,
600–603. (d) Bleul, C. C.; Farzan, M.; Choe, H.; Parolin, C.; Clark-Lewis,
I.; Sodroski, J.; Springer, T. A. The lymphocyte chemoattractant SDF-1 is
a ligand for LESTR/fusin and blocks HIV-1 entry. Nature 1996,
382, 829–833. (e) Busillo, J. M.; Benovic, J. L. Regulation of CXCR4
signaling. Biochem. Biophys. Acta: Biomembranes 2007, 1768, 952–963. (f)
Burger, M.; Hartmann, T.; Krome, M.; Rawluk, J.; Tamamura, H.; Fujii,
N.; Kipps, T. J.; Burger, J. A. Small peptide inhibitors of the CXCR4
chemokine receptor (CD184) antagonize the activation, migration, and
antiapoptotic responses of CXCL12 in chronic lymphocytic leukemia B
cells. Blood 2005, 106, 1824–1830. (g) Murdoch, C. CXCR4: chemokine
receptor extraordinaire. Immunol. Rev. 2000, 177, 175–184. (h) Klein,
R. S.; Rubin, J. B. Immune and nervous system CXCL12 and CXCR4:
parallel roles in patterning and plasticity. Trends Immunol. 2004, 25,
306–314.
(2) (a) Tamamura, H.; Fujii, N. The therapeutic potential of CXCR4
antagonists in the treatment of HIV infection, cancer metastasis and
rheumatoid arthritis. Expert Opin. Ther. Targets 2005, 9, 1267–1282.
(b) De Clercq, E. New approaches toward anti-HIV chemotherapy.
J. Med. Chem. 2005, 48, 1297–1313. (c) Tsutsumi, H.; Tamamura, H.;
Fujii, N. Inhibitors of the chemokine receptor CXCR4: Chemotherapy
of AIDS, metastatic cancer, leukemia and rheumatoid arthritis. Lett. Drug
Des. Discovery 2007, 4, 20–26.
(3) De Clercq, E. Potential clinical applications of the CXCR4
antagonist bicyclam AMD3100. Mini-Rev. Med. Chem. 2005, 5, 805–824.
(4) Burger, J. A.; Kipps, T. J. CXCR4: a key receptor in the crosstalk
between tumor cells and their microenvironment. Blood 2006, 107,
1761–1767.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental details and char-
b
acterization data of synthesized compounds and bioconjugates.
This material is available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Phone: ++49 + 89/289-13300. Fax: ++49 + 89/289-13210.
E-mail: Kessler@tum.de. Address: Institute for Advanced Study,
Technische Universit€at M€unchen, Lichtenbergstrasse 2a, D-85748
Garching, Germany.
’ ACKNOWLEDGMENT
(5) Wong, D.; Korz, W. Translating an antagonist of chemokine
receptor CXCR4: from bench to bedside. Clin. Cancer Res. 2008, 14,
7975–7980.
We thank M. Wolff and B. Cordes, both Department
Chemie and M. Beschorner, A. Alke, K. McGuire, S. Reder,
Dr. M. Schottelius, and Dr. J. Notni, all Lehrstuhl f€ur Pharma-
zeutische Radiochemie, for their assistance. Part of this work was
supported by the Center of Integrated Protein Science Munich
(CIPS), the Federal Ministry of Education and Research
(MOBITEC/MOBITUM 01EZ0826), and the German Research
Foundation (SFB 824/1-2009).
(6) (a) Khan, A.; Greenman, J.; Archibald, S. J. Small molecule
CXCR4 chemokine receptor antagonists: developing drug candidates.
Curr. Med. Chem. 2007, 14, 2257–2277. (b) Levesque, J. P.; Winkler,
I. G. Mobilization of hematopoietic stem cells: state of the art. Curr.
Opin. Organ Transplant. 2008, 13, 53–58. (c) White, F. A.; Wilson, N. M.
Chemokines as pain mediators and modulators. Curr. Opin. Anaesthesiol.
2008, 21, 580–585. (d) Bhangoo, S. K.; Ren, D.; Miller, R. J.; Chan,
D. M.; Ripsch, M. S.; Weiss, C.; McGinnis, C.; White, F. A. CXCR4
chemokine receptor signaling mediates pain hypersensitivity in associa-
tion with antiretroviral toxic neuropathy. Brain Behav. Immunol. 2007,
21, 581–591. (e) Savarin-Vuaillat, C.; Ransohoff, R. M. Chemokines and
’ ABBREVIATIONS USED
AA, amino acid; D-[NMe]Orn, Nα-methyl-D-ornithine; DIAD,
diisopropylazodicarboxylate; DIEA, diisopropylethylamine; DPPA,
7659
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
chemokine receptors in neurological disease: raise, retain, or reduce?
Neurotherapeutics 2007, 4, 590–601. (f) Mines, M.; Ding, Y.; Fan,
G. H. The many roles of chemokine receptors in neurodegenerative
disorders: emerging new therapeutical strategies. Curr. Med. Chem.
2007, 14, 2456–2470. (g) Mꢁelik-Parsadaniantz, S.; Rostꢀene, W.
Chemokines and neuromodulation. J. Neuroimmunol. 2008, 198, 62–68.
(7) (a) Aime, S.; Castelli, D. D.; Crich, S. G.; Gianolio, E.; Terreno, E.
Pushing the Sensitivity Envelope of Lanthanide-Based Magnetic Reso-
nance Imaging (MRI) Contrast Agents for Molecular Imaging Applica-
tions. Acc. Chem. Res. 2009, 42, 822–831. (b) Shokeen, M.; Anderson,
C. J. Molecular Imaging of Cancer withCopper-64 Radiopharmaceuticals
and Positron Emission Tomography (PET). Acc. Chem. Res. 2009,
42, 832–841. (c) De Jong, M.; Breeman, W. A. P.; Kwekkeboom, D. J.;
Valkema, R.; Krenning, E. P. Tumor Imaging and Therapy Using Radi-
olabeled Somatostatin Analogues. Acc. Chem. Res. 2009, 42, 873–880.
(d) Major, J. L.; Meade, T. J. Bioresponsive, Cell-Penetrating, and
Multimeric MR Contrast Agents. Acc. Chem. Res. 2009, 42, 893–903.
(e) Tweedle, M. F. Peptide-Targeted Diagnostics and Radiotherapeutics.
Acc. Chem. Res. 2009, 42, 958–968. (f) Schottelius, M.; Laufer, B.; Kessler,
H.; Wester, H. J. Ligands for Mapping alpha(v)beta(3)-Integrin Expres-
sion in Vivo. Acc. Chem. Res. 2009, 42, 969–980.
(16) Tamamura, H.; Ojida, A.; Ogawa, T.; Tsutsumi, H.; Masuno,
H.; Nakashima, H.; Yamamoto, N.; Hamachi, I.; Fujii, N. Identification
of a new class of low molecular weight antagonists against the chemokine
receptor CXCR4 having the dipicolylamineꢀzinc(II) complex struc-
ture. J. Med. Chem. 2006, 49, 3412–3415.
(17) Faber, A.; Roderburg, C.; Wein, F.; Saffrich, R.; Seckinger, A.;
Horsch, K.; Diehlmann, A.; Wong, D.; Bridger, G.; Eckstein, V.; Ho,
A. D.; Wagner, W. The many facets of SDF-1 alpha, CXCR4 agonists
and antagonists on hematopoietic progenitor cells. J. Biomed. Biotechnol.
2007, 10.
(18) (a) Hereld, D.; Jin, T. Slamming the DOR on chemokine
receptor signaling: Heterodimerization silences ligand-occupied CXCR4
and delta-opioid receptors. Eur. J. Immunol. 2008, 38, 334–337. (b)
Levoye, A.; Balabanian, K.; Baleux, F.; Bachelerie, F.; Lagane, B. CXCR7
heterodimerizes with CXCR4 and regulates CXCL12-mediated G protein
signaling. Blood 2009, 113, 6085–6093.
(19) Wu, B.; Chien, E. Y.; Mol, C. D.; Fenalti, G.; Liu, W.; Katritch,
V.; Abagyan, R.; Brooun, A.; Wells, P.; Bi, F. C.; Hamel, D. J.; Kuhn, P.;
Handel, T. M.; Cherezov, V.; Stevens, R. C. Structures of the CXCR4
Chemokine GPCR with Small-Molecule and Cyclic Peptide Antago-
nists. Science 2010, 330, 1066–1071.
(8) Thumshirn, G.; Hersel, U.; Goodman, S. L.; Kessler, H. Multi-
meric cyclic RGD peptides as potential tools for tumor targeting: Solid-
phase peptide synthesis and chemoselective oxime ligation. Chem.—Eur.
J. 2003, 9, 2717–2725.
(20) Kiessling, L. L.; Gestwicki, J. E.; Strong, L. E. Synthetic multi-
valent ligands in the exploration of cellꢀsurface interactions. Curr. Opin.
Chem. Biol. 2000, 4, 696–703.
(21) Gazit, A.; Osherov, N.; Gilon, C.; Levitzki, A. Tyrphostins. 6.
Dimeric benzylidenemalononitrile tyrophostins: potent inhibitors of EGF
receptor tyrosine kinase in vitro. J. Med. Chem. 1996, 39, 4905–4911.
(22) (a) Poethko, T.; Schottelius, M.; Thumshirn, G.; Hersel, U.;
Herz, M.; Henriksen, G.; Kessler, H.; Schwaiger, M.; Wester, H. J. Two-
step methodology for high-yield routine radiohalogenation of peptides:
F-18-labeled RGD and octreotide analogs. J. Nucl. Med. 2004,
45, 892–902. (b) Li, Z. B.; Chen, K.; Chen, X. 68Ga-labeled multimeric
RGD peptides for microPET imaging of integrin alpha(v)beta(3)
expression. Eur. J. Nucl. Med. Mol. Imaging 2008, 35, 1100–1108.
(23) (a) Miyata, T.; Tokunaga, F.; Yoneya, T.; Yoshikawa, K.;
Iwanaga, S.; Niwa, M.; Takao, T.; Shimonishi, Y. Antimicrobial peptides,
isolated from horseshoe crab hemocytes, tachyplesin II, and polyphe-
musins I and II: chemical structures and biological activity. J. Biochem.
(Tokyo) 1989, 106, 663–668. (b) Tamamura, H.; Xu, Y. O.; Hattori, T.;
Zhang, X. Y.; Arakaki, R.; Kanbara, K.; Omagari, A.; Otaka, A.; Ibuka, T.;
Yamamoto, N.; Nakashima, H.; Fujii, N. A low-molecular-weight
inhibitor against the chemokine receptor CXCR4: A strong anti-HIV
peptide T140. Biochem. Biophys. Res. Commun. 1998, 253, 877–882.
(24) (a) Masuda, M.; Nakashima, H.; Ueda, T.; Naba, H.; Ikoma, R.;
Otaka, A.; Terakawa, Y.; Tamamura, H.; Ibuka, T.; Murakami, T.;
Koyanagi, Y.; Waki, M.; Matsumoto, A.; Yamamoto, N.; Funakoshi,
S.; Fujii, N. A Novel Anti-HIV Synthetic Peptide, T-22 ([Tyr5,12,Lys7]-
Polyphemusin-II). Biochem. Biophys. Res. Commun. 1992, 189, 845–850.
(b) Tamamura, H.; Waki, M.; Imai, M.; Otaka, A.; Ibuka, T.; Waki, K.;
Miyamoto, K.; Matsumoto, A.; Murakami, T.; Nakashima, H.; Yamamoto,
N.; Fujii, N. Downsizing of an HIV-cell fusion inhibitor, T22 ([Tyr(5,12),
Lys(7)]-polyphemusin II), with the maintenance of anti-HIV activity and
solution structure. Bioorg. Med. Chem. 1998, 6, 473–479. (c) Tamamura,
H.; Arakaki, R.; Funakoshi, H.; Imai, M.; Otaka, A.; Ibuka, T.; Nakashima,
H.; Murakami, T.; Waki, M.; Matsumoto, A.; Yamamoto, N.; Fujii, N.
Effective lowly cytotoxic analogs of an HIV-cell fusion inhibitor, T22
([Tyr5,12,Lys7]-polyphemusin II). Bioorg. Med. Chem. 1998, 6, 231–238.
(d) Tamamura, H.; Omagari, A.; Oishi, S.; Kanamoto, T.; Yamamoto, N.;
Peiper, S. C.; Nakashima, H.; Otaka, A.; Fujii, N. Pharmacophore identi-
fication of a specific CXCR4 inhibitor, T140, leads to development of
effective anti-HIV agents with very high selectivity indexes. Bioorg. Med.
Chem. Lett. 2000, 10, 2633–2637.
(9) (a) Poethko, T.; Schottelius, M.; Thumshirn, G.; Herz, M.;
Haubner, R.; Henriksen, G.; Kessler, H.; Schwaiger, M.; Wester, H. J.
Chemoselective pre-conjugate radiohalogenation of unprotected mono-
and multimeric peptides via oxime formation. Radiochim. Acta 2004,
92, 317–327. (b) Vagner, J.; Handl, H. L.; Monguchi, Y.; Jana, U.; Begay,
L. J.; Mash, E. A.; Hruby, V. J.; Gillies, R. J. Rigid linkers for bioactive
peptides. Bioconjugate Chem. 2006, 17, 1545–1550. (c) Wester, H. J.;
Kessler, H. Molecular targeting with peptides or peptideꢀpolymer
conjugates: Just a question of size? J. Nucl. Med. 2005, 46, 1940–1945.
(10) Kessler, H.; Schudok, M.; Haupt, A. In Peptides; Jung, G., Bayer,
E., Eds.; Walter de Gruyter: T€ubingen, Germany, 1988; pp 664ꢀ666.
(11) (a) Mammen, M.; Choi, S. K.; Whitesides, G. M. Polyvalent
interactions in biological systems: implications for design and use of
multivalent ligands and inhibitors. Angew. Chem. Int. Ed. 1998, 37,
2755–2794. (b) Kiessling, L. L.; Gestwicki, J. E.; Strong, L. E. Synthetic
multivalent ligands as probes of signal transduction. Angew. Chem., Int.
Ed. Engl. 2006, 45, 2348–2368.
(12) Crump, M. P.; Gong, J. H.; Loetscher, P.; Rajarathnam, K.;
Amara, A.; Arenzana-Seisdedos, F.; Virelizier, J. L.; Baggiolini, M.; Sykes,
B. D.; Clark-Lewis, I. Solution structure and basis for functional activity
of stromal cell-derived factor-1; dissociation of CXCR4 activation from
binding and inhibition of HIV-1. EMBO J. 1997, 16, 6996–7007.
(13) (a) De Clercq, E.; Yamamoto, N.; Pauwels, R.; Balzarini, J.;
Witvrouw, M.; De Vreese, K.; Debyser, Z.; Rosenwirth, B.; Peichl, P.;
Datema, R.; et al. Highly potent and selective inhibition of human
immunodeficiency virus by the bicyclam derivative JM3100. Antimicrob.
Agents Chemother. 1994, 38, 668–674. (b) De Clercq, E. The AMD3100
story: the path to the discovery of a stem cell mobilizer (Mozobil).
Biochem. Pharmacol. 2009, 77, 1655–1664.
(14) Zhan, W. Q.; Liang, Z. X.; Zhu, A. Z.; Kurtkaya, S.; Shim, H.;
Snyder, J. P.; Liotta, D. C. Discovery of small molecule CXCR4
antagonists. J. Med. Chem. 2007, 50, 5655–5664.
(15) (a) Ojida, A.; Mito-oka, Y.; Inoue, M.; Hamachi, I. First artificial
receptors and chemosensors toward phosphorylated peptide in aqueous
solution. J. Am. Chem. Soc. 2002, 124, 6256–6258. (b) Loetscher, P.;
Gong, J. H.; Dewald, B.; Baggiolini, M.; Clark-Lewis, I. N-Terminal
peptides of stromal cell-derived factor-1 with CXC chemokine receptor 4
agonist and antagonist activities. J. Biol. Chem. 1998, 273, 22279–22283.
(c) Liang, X. CXCR4, inhibitors and mechanisms of action. Chem. Biol.
Drug Des. 2008, 72, 97–110. (d) Heveker, N.; Tissot, M.; Thuret, A.;
Schneider-Mergener, J.; Alizon, M.; Roch, M.; Marullo, S. Pharmacolo-
gical properties of peptides derived from stromal cell-derived factor 1: study
on human polymorphonuclear cells. Mol. Pharmacol. 2001, 59, 1418–1425.
(25) Fujii, N.; Oishi, S.; Hiramatsu, K.; Araki, T.; Ueda, S.; Tamamura,
H.; Otaka, A.; Kusano, S.; Terakubo, S.; Nakashima, H.; Broach, J. A.;
Trent, J. O.; Wang, Z. X.;Peiper, S. C. Molecular-size reduction ofa potent
CXCR4-chemokine antagonist using orthogonal combination of confor-
mation- and sequence-based libraries. Angew. Chem. Int. Ed. 2003, 42,
3251–3253.
7660
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
(26) Haubner, R.; Kuhnast, B.; Mang, C.; Weber, W. A.; Kessler, H.;
Wester, H. J.; Schwaiger, M. [F-18]Galacto-RGD: Synthesis, radiolabel-
ing, metabolic stability, and radiation dose estimates. Bioconjugate Chem.
2004, 15, 61–69.
duodenally absorbable CXC chemokine receptor 4 antagonist, KRH-
1636, exhibits a potent and selective anti-HIV-1 activity. Proc. Natl. Acad.
Sci. U.S.A. 2003, 100, 4185–4190. (b) Fujii, N.; Nakashima, H.;
Tamamura, H. The therapeutic potential of CXCR4 antagonists in the
treatment of HIV. Expert Opin. Invest. Drug 2003, 12, 185–195.
(38) Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.;
Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. Extra
precision Glide: Docking and scoring incorporating a model of hydro-
phobic enclosure for protein-ligand complexes. J. Med. Chem. 2006, 49,
6177–6196.
(39) (a) Clark-Lewis, I.; Kim, K. S.; Rajarathnam, K.; Gong, J. H.;
Dewald, B.; Moser, B.; Baggiolini, M.; Sykes, B. D. Structureꢀactivity
relationships of chemokines. J. Leukocyte Biol. 1995, 57, 703–11.
(b) Wells, T. N.; Power, C. A.; Lusti-Narasimhan, M.; Hoogewerf,
A. J.; Cooke, R. M.; Chung, C. W.; Peitsch, M. C.; Proudfoot, A. E.
Selectivity and antagonism of chemokine receptors. J. Leukocyte Biol.
1996, 59, 53–60. (c) Gupta, S. K.; Pillarisetti, K.; Thomas, R. A.; Aiyar,
N. Pharmacological evidence for complex and multiple site interaction of
CXCR4 with SDF-1alpha: implications for development of selective
CXCR4 antagonists. Immunol. Lett. 2001, 78, 29–34.
(40) (a) Gerlach, L. O.; Jakobsen, J. S.; Jensen, K. P.; Rosenkilde,
M. R.; Skerlj, R. T.; Ryde, U.; Bridger, G. J.; Schwartz, T. W. Metal ion
enhanced binding of AMD3100 to Asp262 in the CXCR4 receptor.
Biochemistry 2003, 42, 710–717. (b) Khan, A.; Silversides, J. D.; Madden,
L.; Greenman, J.; Archibald, S. J. Fluorescent CXCR4 chemokine receptor
antagonists: metal activated binding. Chem. Commun. (Cambridge, UK)
2007, 416–418. (c) Khan, A.; Nicholson, G.; Greenman, J.; Madden, L.;
McRobbie, G.; Pannecouque, C.; De Clercq, E.; Ullom, R.; Maples, D. L.;
Maples, R. D.; Silversides, J. D.; Hubin, T. J.; Archibald, S. J. Binding
Optimization through Coordination Chemistry: CXCR4 Chemokine
Receptor Antagonists from Ultrarigid Metal Complexes. J. Am. Chem.
Soc. 2009, 131, 3416–3417. (d) Heppeler, A.; Froidevaux, S.; M€acke,
H. R.; Jermann, E.; Bꢁehꢁe, M.; Powell, P.; Hennig, M. Radiometal-Labelled
Macrocyclic Chelator-Derivatised Somatostatin Analogue with Superb
Tumour-Targeting Properties and Potential for Receptor-Mediated Inter-
nal Radiotherapy. Chem. Eur. J. 1999, 5, 1974–1981. (e) Liu, S.; He, Z.;
Hsieh, W.-Y.; Fanwick, P. E. Synthesis, Characterization, andX-rayCrystal
Structure of In(DOTA-AA) (AA = p-Aminoanilide): A Model for
111In-Labeled DOTA-Biomolecule Conjugates. Inorg. Chem. 2003, 42,
8831–8837.
(41) Reubi, J. C.; Schar, J. C.; Waser, B.; Wenger, S.; Heppeler, A.;
Schmitt, J. S.; M€acke, H. R. Affinity profiles for human somatostatin
receptor subtypes SST1-SST5 of somatostatin radiotracers selected for
scintigraphic and radiotherapeutic use. Eur. J. Nucl. Med. 2000, 27,
273–282.
(42) Hamzꢁe, A.; Hernandez, J. F.; Martinez, J. Synthesis of (R) and
(S) enantiomers of Fmoc-protected 1,2,4-oxadiazole-containing beta3-
amino acids from Fmoc-(R)-beta-HAsp(OtBu)-OH. Tetrahedron Lett.
2003, 44, 6079–6082.
(43) Herzner, H.; Kunz, H. Spacer-separated sialyl LewisX cyclo-
peptide conjugates as potential E-selectin ligands. Carbohydr. Res. 2007,
342, 541–557.
(44) Castonguay, R.; Lherbet, C.; Keillor, J. W. Mapping of the active
site of rat kidney gamma-glutamyl transpeptidase using activated esters
and their amide derivatives. Bioorg. Med. Chem. 2002, 10, 4185–4191.
(45) Mizukami, S.; Takikawa, R.; Sugihara, F.; Hori, Y.; Tochio, H.;
Walchli, M.; Shirakawa, M.; Kikuchi, K. Paramagnetic relaxation-based
F-19 MRI probe to detect protease activity. J. Am. Chem. Soc. 2008,
130, 794–795.
(27) (a) Zhang, W. B.; Navenot, J. M.; Haribabu, B.; Tamamura, H.;
Hiramatu, K.; Omagari, A.; Pei, G.; Manfredi, J. P.; Fujii, N.; Broach, J. R.;
Peiper, S. C. A point mutation that confers constitutive activity to CXCR4
reveals that T140 is an inverse agonist and that AMD3100 and ALX40ꢀ4C
axe weak partial agonists. J. Biol. Chem. 2002, 277, 24515–24521. (b) Trent,
J. O.; Wang, Z. X.; Murray, J. L.; Shao, W.; Tamamura, H.; Fujii, N.; Peiper,
S. C. Lipid bilayer simulations of CXCR4 with inverse agonists and weak
partial agonists. J. Biol. Chem. 2003, 278, 47136–47144. (c) Wang, Z.-x.;
Tamamura, H.; Frilot, N.; Broach, J.; Fujii, N.; Peiper, S. Screening and
characterization of cyclic pentapeptide CXCR4 antagonists/inverse agonists
using a pheromone responsive reporter gene in Saccharomyces cerevisiae:
Utility of G protein coupled receptor constitutively active mutants. In
Chemokine Biology—Basic Research and Clinical Application; Neote, K. L.,
Gordon, L., Moser, B., Eds.; Birkh€auser: Basel, 2007; pp 61ꢀ77.
(28) Tanaka, T.; Nomura, W.; Narumi, T.; Masuda, A.; Tamamura,
H. Bivalent Ligands of CXCR4 with Rigid Linkers for Elucidation of the
Dimerization State in Cells. J. Am. Chem. Soc. 2010, 132, 15899–901.
(29) Ueda, S.; Oishi, S.; Wang, Z. X.; Araki, T.; Tamamura, H.;
Cluzeau, J.; Ohno, H.; Kusano, S.; Nakashima, H.; Trent, J. O.; Peiper,
S. C.; Fujii, N. Structureꢀactivity relationships of cyclic peptide-based
chemokine receptor CXCR4 antagonists: disclosing the importance of
side-chain and backbone functionalities. J. Med. Chem. 2007, 50, 192–198.
(30) Demmer, O.; Koglin, N.; Kessler, H.; Wester, H. J. Cyclic
peptides for targeting CXCR4 chemokine receptor expression. J. Pept.
Sci. 2006, 12, 172–172.
(31) Demmer, O.; Gourni, E.; Schumacher, U.; Wester, H. J.; Kessler,
H. PET imaging of CXCR4 receptors in cancer by a new optimized
ligand. ChemMedChem, 2011, DOI 10.1002/cmdc.201100320
(32) Demmer, O.; Dijkgraaf, I.; Schottelius, M.; Wester, H. J.; Kessler,
H. Introduction of functional groups into peptides via N-alkylation. Org.
Lett. 2008, 10, 2015–2018.
(33) (a) Demmer, O.; Frank, A. O.; Kessler, H. Design of Cyclic
Peptides. In Design of Peptides and Proteins for Biopharmaceutical
Applications: Applications for Therapeutic Agents and Biomedical Research;
Jensen, K. J., Ed.; Wiley: New York, 2009; pp 133ꢀ176. (b) Kessler, H.;
Kutscher, B. Peptide Conformations 0.40. Cyclic Hexapeptide Analogs
of Thymopoietin—Synthesis and Conformational Analysis. Liebigs Ann.
Chem. 1986, 914–931.(c) Gilon, C.; Mang, C.; Lohof, E.; Friedler, A.;
Kessler, H. Synthesis of Peptides and Peptidomimetics. In Houben-Weyl:
Methods of Organic Chemistry; Goodman, M., Felix, L., Moroder, L.,
Toniolo, C., Eds.; Thieme Verlag: Stuttgart, 2003; pp 461ꢀ542.
(34) Biron, E.; Chatterjee, J.; Kessler, H. Optimized selective N-methy-
lation of peptides on solid support. J. Pept. Sci. 2006, 12, 213–219.
(35) M€uller, A.; Vogt, C.; Sewald, N. Synthesis of Fmoc-beta-
homoamino acids by ultrasound-promoted Wolff rearrangement. Synth-
esis (Stuttgart) 1998, 837–841.
(36) (a) Moepps, B.; Frodl, R.; Rodewald, H. R.; Baggiolini, M.;
Gierschik, P. Two murine homologues of the human chemokine
receptor CXCR4 mediating stromal cell-derived factor 1 alpha activation
of G(12) are differentially expressed in vivo. Eur. J. Immunol. 1997,
27, 2102–2112. (b) Hanaoka, H.; Mukai, T.; Tamamura, H.; Mori, T.;
Ishino, S.; Ogawa, K.; Iida, Y.; Doi, R.; Fujii, N.; Saji, H. Development of
a In-111-labeled peptide derivative targeting a chemokine receptor,
CXCR4, for imaging tumors. Nucl. Med. Biol. 2006, 33, 489–494.
(c) Jacobson, O.; Weiss, I. D.; Szajek, L.; Farber, J. M.; Kiesewetter, D. O.
Cu-64-AMD3100—a novel imaging agent for targeting chemokine receptor
CXCR4. Bioorg. Med. Chem. 2009, 17, 1486–1493. (d) Nimmagadda, S.;
Pullambhatla, M.; Stone, K.; Green, G.; Bhujwalla, Z. M.; Pomper, M. G.
Molecular Imaging of CXCR4 Receptor Expression in Human Cancer
Xenografts with [Cu-64]AMD3100 Positron Emission Tomography. Cancer
Res. 2010, 70, 3935–3944.
(46) Wishart, D. S.; Bigam, C. G.; Yao, J.; Abildgaard, F.; Dyson,
H. J.; Oldfield, E.; Markley, J. L.; Sykes, B. D. 1H, 13C and 15N chemical
shift referencing in biomolecular NMR. J. Biomol. NMR 1995,
6, 135–140.
(47) Piotto, M.; Saudek, V.; Sklenar, V. Gradient-tailored excitation
for single-quantum NMR spectroscopy of aqueous solutions. J. Biomol.
NMR 1992, 2, 661–665.
(37) (a) Ichiyama, K.; Yokoyama-Kumakura, S.; Tanaka, Y.; Tanaka,
R.; Hirose, K.; Bannai, K.; Edamatsu, T.; Yanaka, M.; Niitani, Y.;
Miyano-Kurosaki, N.; Takaku, H.; Koyanagi, Y.; Yamamoto, N. A
(48) (a) Kessler, H.; Seip, S. NMR of Peptides. In Two-Dimensional
NMR-Spectroscopy, Applications for Chemists and Biochemists; Croasmun,
7661
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662

Journal of Medicinal Chemistry
ARTICLE
W. R., Carlson, M. K., Eds.; VCH: New York, 1994; pp 619ꢀ654.
(b) Kessler, H.; Schmitt, W. Peptides and Polypeptides. In Encyclopedia
of Nuclear Magnetic Resonance; Grant, D. M., Harris, R. K., Eds.; J. Wiley
& Sons: New York, 1996; pp 3527ꢀ3537.
(49) (a) Bax, A.; Summers, M. F. H-1 and C-13 Assignments from
Sensitivity-Enhanced Detection of Heteronuclear Multiple-Bond Con-
nectivity by 2d Multiple Quantum Nmr. J. Am. Chem. Soc. 1986,
108, 2093–2094. (b) Bermel, W.; Wagner, K.; Griesinger, C. Proton-
Detected C,H Correlation Via Long-Range Couplings with Soft Pulses—
Determination of Coupling-Constants. J. Magn. Reson. 1989, 83,
223–232. (c) Kessler, H.; Schmieder, P.; Kock, M.; Kurz, M. Improved
Resolution in Proton-Detected Heteronuclear Long-Range Correlation.
J. Magn. Reson. 1990, 88, 615–618.
(50) Griesinger, C.; Sorensen, O. W.; Ernst, R. R. Practical Aspects
of the E-Cosy Technique—Measurement of Scalar Spin Spin Coupling
Constants in Peptides. J. Magn. Reson. 1987, 75, 474–492.
(51) (a) Bothner-By, A. A.; Stephens, R. L.; Lee, J. M.; Warren,
C. D.; Jeanloz, R. W. Structure Determination of a Tetrasaccharide—
Transient Nuclear Overhauser Effects in the Rotating Frame. J. Am.
Chem. Soc. 1984, 106, 811–813. (b) Kessler, H.; Griesinger, C.;
Kerssebaum, R.; Wagner, K.; Ernst, R. R. Separation of Cross-Relaxation
and J-Cross-Peaks in 2D Rotating-Frame NMR Spectroscopy. J. Am.
Chem. Soc. 1987, 109, 607–609.
(52) (a) Berendsen, H. J. C.; van der Spoel, D.; van Drunen, R.
GROMACS: A message-passing parallel molecular dynamics implemen-
tation. Comput. Phys. Commun. 1995, 91, 43–56. (b) Lindahl, E.; Hess,
B.; van der Spoel, D. GROMACS 3.0: A package for molecular
simulation and trajectory analysis. J. Mol. Model. 2001, 7, 306–317.
(53) Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.;
Dinola, A.; Haak, J. R. Molecular Dynamics with Coupling to an External
Bath. J. Chem. Phys. 1984, 81, 3684–3690.
(54) Schr€odinger Suite 2009 Protein Preparation Wizard, Epik
version 2.0; Impact version 5.5; Prime version 2.1; Schr€odinger, LLC:
New York, NY, 2009.
(55) Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.;
Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. Extra
precision glide: docking and scoring incorporating a model of hydro-
phobic enclosure for protein-ligand complexes. J. Med. Chem. 2006,
49, 6177–6196.
(56) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.;
Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF Chimera—a
visualization system for exploratory research and analysis. J. Comput.
Chem. 2004, 25, 1605–12.
7662
dx.doi.org/10.1021/jm2009716 |J. Med. Chem. 2011, 54, 7648–7662