Design, synthesis, and biological evaluation of scaffold-based tripeptidomimetic antagonists for CXC chemokine receptor 4 (CXCR4)

Article abstract of DOI:10.1016/j.bmc.2014.07.004

Structure-activity relationship studies of the cyclopentapeptide CXCR4 antagonists (cyclo(-l-/d-Arg¹-Arg²-2-Nal ³-Gly⁴-d-Tyr⁵-)) suggest that the l-/d-Arg ¹-Arg²-2-Nal³

Full text of DOI:10.1016/j.bmc.2014.07.004

Accepted Manuscript
Design, Synthesis, and Biological Evaluation of Scaffold-Based Tripeptidomi-
metic Antagonists for CXC Chemokine Receptor 4 (CXCR4)
Zack G. Zachariassen, Stefanie Thiele, Erik A. Berg, Pernille Rasmussen,
Torgils Fossen, Mette M. Rosenkilde, Jon Våbenø, Bengt Erik Haug
PII:
S0968-0896(14)00507-0
DOI:
http://dx.doi.org/10.1016/j.bmc.2014.07.004
BMC 11698
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
28 March 2014
26 June 2014
2 July 2014
Please cite this article as: Zachariassen, Z.G., Thiele, S., Berg, E.A., Rasmussen, P., Fossen, T., Rosenkilde, M.M.,
Våbenø, J., Haug, B.E., Design, Synthesis, and Biological Evaluation of Scaffold-Based Tripeptidomimetic
Antagonists for CXC Chemokine Receptor 4 (CXCR4), Bioorganic & Medicinal Chemistry (2014), doi: http://
dx.doi.org/10.1016/j.bmc.2014.07.004
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design, Synthesis, and Biological Evaluation of Scaffold-Based Tripeptidomimetic
Antagonists for CXC Chemokine Receptor 4 (CXCR4)
a
b
c
c
c
Zack G. Zachariassen, Stefanie Thiele, Erik A. Berg, Pernille Rasmussen, Torgils Fossen,
b
,a
,c
Mette M. Rosenkilde, Jon Våbenø,* Bengt Erik Haug*
a
Department of Pharmacy, Faculty of Health Sciences, UiT The Arctic University of Norway,
Breivika, NO-9037 Tromsø, Norway.
b
Laboratory for Molecular Pharmacology, Department of Neuroscience and Pharmacology,
Faculty of Health and Medical Sciences, University of Copenhagen, Blegdamsvej 3, DK-
2
200 Copenhagen, Denmark.
c
Department of Chemistry and Centre for Pharmacy, University of Bergen, Allégaten 41,
NO-5007 Bergen, Norway.
Corresponding author for the review process:
Bengt Erik Haug. Phone: +47 55 58 34 68; e-mail: Bengt-Erik.Haug@farm.uib.no
Post-publication corresponding authors:
*
Bengt Erik Haug. Phone: +47 55 58 34 68; e-mail: Bengt-Erik.Haug@farm.uib.no
or
*
Jon Våbenø. Phone: +47 77 62 09 09; e-mail: jon.vabeno@uit.no
Notes:
The authors declare no competing financial interest.
Keywords:
CXCR4 antagonist, cyclopentapeptide, peptidomimetic, scaffold
1

Abstract
Structure-activity relationship studies of the cyclopentapeptide CXCR4 antagonists (cyclo(-L-
1
2
3
4
5
1
2
3
/
D-Arg -Arg -2-Nal -Gly -D-Tyr -)) suggest that the L-/D-Arg -Arg -2-Nal tripeptide
sequence contained within these cyclopentapeptides serves as a recognition motif for peptidic
CXCR4 antagonists. Starting by dissecting the cyclopentapeptide structure and reintroducing
cyclic constraints in a stepwise manner, we here report a novel class of scaffold-based
tripeptidomimetic CXCR4 antagonists based on the D-Arg-Arg-2-Nal motif. Biological
testing of the prototype compounds showed that they represent new peptidomimetic hits;
importantly, the modular nature of the scaffold provides an interesting starting point for
future ligand optimization.
Graphical Abstract
2

1
. INTRODUCTION
CXC chemokine receptor 4 (CXCR4) is a peptidergic GPCR with the 68-residue peptide
1
,2
CXC chemokine ligand 12 (CXCL12) as its only endogenous ligand. In addition to the
developmental and physiological role of CXCL12/CXCR4, CXCR4 has been shown to be
involved in a number of pathological conditions, including HIV, cancer, and rheumatoid
3
arthritis. Consequently, CXCR4 has emerged as an attractive drug target, and several small-
3
,4
molecule CXCR4 antagonists have been described in the literature over the last decade,
1
2
including a series of cyclic pentapeptides based on the amino acid sequence L-/D-Arg -Arg -
3
4
5
1
2
-Nal -Gly -D-Tyr (2-Nal = L-3-(2-naphthyl)alanine), i.e. the L-Arg epimer 1 (FC131) and
1
5
the D-Arg epimer 2 (FC092) (Figure 1).
Figure 1. Structures of the lead cyclopentapeptide antagonists 1 and 2.
6
Small cyclic peptides, such as cyclopentapeptides, are known to mimic peptide turns, i.e.
structural motifs where the peptide backbone folds back on itself to form a pseudo-cyclic
structure. Such turns appear to be a universal ligand recognition element for peptidergic
7
GPCRs; thus, the cyclopentapeptide CXCR4 antagonists represent an excellent starting point
for design of novel CXCR4 antagonists based on smaller turn-mimicking scaffolds. Such
scaffolds have to maintain the 3D-orientation of the pharmacophoric groups of the parent
8
peptide, resulting in a so-called topographical (or type-III) peptidomimetic. With respect to
size, a recent review points out that three amino acids represent an optimal size for small-
9
molecule peptidomimetics.
4
The Gly residue in the cyclopentapeptide CXCR4 antagonists (Figure 1) was originally
5
introduced for synthetic reasons and thus can be considered as a spacer. Through SAR
5
studies, it has become clear that the D-Tyr side chain is not essential for biological
1
0,11
activity,
and we have in a recent molecular docking study found that this may stem from
5
11
the lack of a defined binding pocket for the D-Tyr side chain. Even if our modeling
3

5
suggested that D-Tyr of 1 is partly solvent exposed, the phenolic hydroxyl group was found
3
2 11
5
to take part in hydrogen bonding to Glu . However, as the D-Phe analogue was found to
be only 2-fold less potent, this H-bond is not essential for antagonistic activity; indeed, the
5
Gly analogue of 1 was found to be only one order of magnitude less potent than 1 in our
1
1
1
2
assay system. Collectively, these observations imply that the remaining L-/D-Arg -Arg -2-
3
Nal tripeptide fragment serves as a recognition motif for peptidic CXCR4 antagonists, which
motivates further studies of both flexible and constrained small molecules containing this
motif. Through dissection of the cyclopentapeptide structure and a stepwise reintroduction of
cyclic constraints, we here report the design, synthesis, and biological evaluation of a novel
class of scaffold-based tripeptidomimetic CXCR4 antagonists based on the D-Arg-Arg-2-Nal
motif.
2
2
. RESULTS AND DISCUSSION
.1. General Design Considerations.
As SAR studies of the cyclopentapeptide CXCR4 antagonists (Figure 1) have demonstrated
1
2,13
that position 2 (L-Arg) is very sensitive to structural modifications,
we decided to keep L-
2
Arg throughout this study. Similarly, we have recently shown that replacement of L-2-Nal in
position 3 with aromatic/alicyclic analogs results in significant reduction of the antagonistic
1
1
potency, and therefore used a 2-naphthyl group with the appropriate spacer length for all
compounds. In contrast, position 1 has been shown to be relatively tolerant to structural
modifications, both with respect to stereochemistry (L- or D-arginine) and the chemical nature
1
3,14
1
of the side chain.
Even if the originally discovered L-Arg epimer 1 displays somewhat
1
higher affinity than the D-Arg epimer 2 (Figure 1; IC -values of 0.004 and 0.008 μM,
5
0
5
respectively), subsequent SAR studies have shown that cyclopentapeptide analogs
1
1
containing D-Arg in many cases are more active than the corresponding L-Arg epimers. For
example, the most active cyclopentapeptide CXCR4 antagonist reported to date is the N-
1
1
2
3
4
5
methylated D-Arg epimer FC122 (cyclo(-N-Me-D-Arg -Arg -2-Nal -Gly -D-Tyr -)), which
1
showed 8-fold higher affinity than the corresponding L-Arg epimer (IC -values of 0.003 and
5
0
1
0
0
.023 μM, respectively). Moreover, head-to-tail cyclization of peptides is known to be
facilitated by incorporation of a D-amino acid in an all-L sequence due to a turn-inducing
1
5
1
effect. For these reasons, we decided to focus on the D-Arg epimers in the present study,
1
2
3
4
5
using the lead cyclopentapeptide 2 (cyclo(-D-Arg -Arg -2-Nal -Gly -D-Tyr -), Figure 2A) as
starting point.
4

2
.2. Biological Evaluation.
The antagonistic potency of the synthesized compounds 2–14 (Figures 2 and 6) on human
1
3
CXCR4 was determined by a functional assay as previously described and is shown in
Table 1; the EC -value of the known lead compound 2 was 0.52 μM.
5
0
Table 1. Antagonistic potency of compounds 2–14 on human CXCR4.
a
b
Compd
log EC ± SEM
EC (μM)
5
0
50
c
2
-6.28 ± 0.09
-4.24 ± 0.35
-4.07 ± 0.24
-4.36 ± 0.10
> -4
0.52
58
3
4
5
6
7
8
9
86
44
>100
>100
>100
>100
60
> -4
> -4
> -4
1
0
-4.22 ± 0.06
> -4
1
1
>100
>100
80
1
1
2
3
4
> -4
-4.10 ± 0.31
-4.20 ± 0.12
1
64
a
For structures, see Figures 2 and 6.
b
c
Values represent the mean of at least three independent experiments performed in duplicates.
5
Known compound.
5

A
D-Tyr5
Gly4
NH HN
NH HN
O
O
OH
O
-Nal3
H
H
2
N
N
NH2
NH
O
O
D-Arg1
Arg2
HN
FC092 (2)
H N
NH
NH
2
B
R'
O
R
H
H
N
N
NH2
O
NH
HN
3
4
: R = R' = H
: R = NHCOMe, R' = CONH2
H N
NH
2
C
5: cyclo(-D-Arg1-Arg2-2-Nal3-Gly4-Gly5-)
O
H
N
n
O
NH
HN
-Nal3
H
H
2
N
N
NH2
NH
O
O
6
: n = 4
D-Arg1
Arg2
7: n = 5
HN
8
9
: n = 3
: n = 2
H N
NH
2
10: n = 1
Figure 2. Structures of (A) the lead cyclopentapeptide 2, (B) the tripeptidic compounds 3–4,
and (C) the macrocyclic compounds 5–10.
2
.3. Design and SAR.
2
.3.1. Linear Tripeptidic Compounds.
Taking a minimalist approach, 2 was initially dissected to the linear peptidic derivative 3
1
2
3
(
Figure 2B) in order to determine the activity of the isolated Arg -Arg -2-Nal motif. The N-
1
2
3
acetylated D-Arg -Arg -2-Nal tripeptide amide 4 (Figure 2B) was also included to study the
role of the two flanking amide groups. Compound 3 (EC = 58 μM) displayed 112-fold
5
0
lower potency than 2, reflecting the extensive dissection of the cyclopentapeptide structure.
This finding is consistent with literature data for similar linear tripeptidic CXCR4 antagonists
1
6,17
originating from an Arg-Arg-Nal motif;
representative structures are shown in Figure 3.
In the same way as 3, these compounds were based on a central L-arginine, and were found to
have moderate potency (anti-HIV assay), typically in the order of 100-fold reduction relative
to the parent cyclopentapeptide.
6

16,17
Figure 3. Structures of previously reported linear tripeptidic CXCR4 antagonists.
The N- and C-terminal capped linear tripeptide 4 showed somewhat lower activity (EC = 86
5
0
μM) than 3; thus, the terminal amide groups of 4 do not contribute favorably to activity. The
activity of 4 relative to 2 (165-fold reduction) is consistent with SAR studies by Fujii et al.
1
2
which showed that the linear N- and C-terminal capped pentapeptide Ac-D-Arg -L-Arg -L-2-
3
4
5
Nal -Gly -D-Tyr -NH was 173-fold less potent (anti-HIV assay) than the parent
2
5
cyclopentapeptide 2.
2
.3.2. Macrocyclic Compounds.
The analogs 3 and 4 are quite flexible, which is generally considered as an undesirable
1
2
3
feature, and a macrocyclic constraint was reintroduced to force the D-Arg -Arg -2-Nal motif
4
5
into a more restricted conformation (Figure 2C). Use of a Gly -Gly dipeptide spacer to give
a simplified cyclopentapeptide (5) resulted in a slight increase in potency (EC = 44 μM)
50
4
5
relative to 4. Since the Gly -Xaa amide bond was not found to be involved in direct binding
1
1
interactions with the receptor in our recently proposed binding mode, 5 was further
4
5
4
simplified by replacing Gly -Gly with a 5-aminopentanoic acid hydrocarbon spacer (6).
This modification, which essentially is an ethylene amide bond isostere (Ψ[CH CH ])
2
2
replacement, resulted in loss of activity in our assay (EC > 100 μM). This shows that the
5
0
4
5
planar and polar Gly -Gly amide bond in 5 contributes favorably to activity, either by
geometrical or solvation effects.
7

Next, we employed a ring expansion/contraction strategy by using ω-amino carboxylic acid
spacers of different length (7–10, Figure 2C). Extension of the hydrocarbon spacer in 6 (15-
membered ring) to give a 16-membered ring (7) or contraction to give a 14-membered ring
(
8) did still not give any measurable activity (EC > 100 μM for both compounds). Similarly,
50
the 13-membered ring (9) was inactive (EC > 100 μM); however, the further constrained
5
0
cyclotetrapeptide 10 (12-membered ring) was equipotent (EC = 60 μM) with the linear
5
0
tripeptidic compound 3.
2
.3.3. Bicyclic Tripeptidomimetics.
Encouraged by the activity of 10, we set out to develop a further constrained turn-mimicking
1
2
3
scaffold capable of presenting the side chains and backbone of the D-Arg -Arg -2-Nal
fragment of the cyclopentapeptide 2 in the required 3D-orientation. The key to successful
development of such topographical peptidomimetics is knowledge of the bioactive
conformation of the parent peptide, in this case the cyclopentapeptide ligands. Based on an
extensive exploration of the conformational space for a series of cyclopentapeptide CXCR4
antagonists from the literature, we have previously reported a 3D pharmacophore model that
describes the spatial arrangement of the pharmacophoric side chains as well as the bioactive
1
8
conformation of the cyclopentapeptide backbone.
1
9
Searching through the extensive literature on turn-mimetics (see ref for a review), we were
2
0
21,22
intrigued by the tripeptide-derived 3,6,8-trisubstituted bicyclic structure A (Figure 4A),
which contains two endocyclic amide bonds. The synthesis of A was first reported by
2
1
Vojkovsky et al. who suggested it as a potential peptide-turn motif; however, no biological
applications of A have yet been reported. In order to elucidate whether this scaffold would be
suitable for our purpose, a structural comparison of low-energy conformations of A with our
3
D pharmacophore model was undertaken. This showed that scaffold A is able to orient the
side chains in a similar way as the parent cyclopentapeptide (Figure 4B).
A
B
H
N
R3
S
N8
3
6
O
R1
R3
R2
O
scaffold A
R1
R2
8

Figure 4. (A) Structure of scaffold A, and (B) superimposition of a low-energy conformation
of A (grey carbon atoms) and the bioactive backbone conformation of the cyclopentapeptide
1
8
CXCR4 antagonists (green carbon atoms) as defined by our 3D pharmacophore model.
RMSD = 0.160 Å (cf. section 4.2).
Scaffold A contains three stereocenters, where two (C3 and C6) are defined by the building
blocks (see Scheme 4 in Chemistry section). The bridge-head stereocenter (C9a) is formed in
the final cyclization step, and the stereochemical outcome has been shown to be governed by
2
3
the configuration at C3 (see Chemistry section); thus, two diastereomeric scaffolds A and
A’ (Figure 5A) can be prepared. Interestingly, structural comparison of the expected
bioactive conformation of A (shown in Figure 4B) with low-energy conformations of A’
showed that A’ can adopt an almost identical conformation with respect to the orientation of
the two amide bonds and the three side chains (Figure 5B).
A
H
N
H
N
R3
S
R3
S
N
N
O
R1
O
R1
R2
O
R2
O
scaffold A
scaffold A'
B
R3
R1
R2
Figure 5. (A) Structures of the diastereomeric scaffolds A and A’, and (B) superimposition
of low-energy conformations of A (grey carbon atoms) and A’ (green carbon atoms). RMSD
=
0.036 Å (cf. section 4.2).
Thus, we decided to pursue tripeptidomimetic compounds 11 and 12 (based on scaffold A
and A’, respectively; Figure 6) by using L- and D-cysteine as building blocks, resulting in an
1
1
amide group in the R side chain. In order to introduce the same arginine R side chain as the
parent cyclopentapeptide, we also adapted the synthesis to allow for preparation of 13 and 14
(Figure 6), which in contrast to 11 and 12 do not carry this amide bond.
9

H
N
H
N
S
S
N
O
N
O
H
N
H
N
NH2
NH
NH2
NH
O
N
O
N
H
H
O
O
HN
HN
11
12
HN
NH2
HN
NH2
H
N
H
N
S
S
N
N
H
N
H
N
NH2
NH
NH2
NH
O
O
O
O
HN
HN
13
14
HN
NH2
HN
NH2
Figure 6. Structures of the bicyclic target compounds 11–14.
Interestingly, compounds 11 and 12 were inactive, while compounds 13 and 14 (EC = 80
5
0
and 64 μM, respectively) showed activity similar to 3 and 10. Clearly, the amide bond in the
1
R side-chain of 11 and 12 is unfavorable for the biological activity, an observation that
1
justifies the synthetic effort put into the R building block 33 that was used for 13 and 14 (see
Scheme 7, Chemistry section). Thus, 13 and 14 represent new peptidomimetic hit compounds
for CXCR4 antagonism.
The scaffold-based tripeptidomimetics 13 and 14 are significantly less potent (154- and 123-
fold, respectively) than the optimized cyclopentapeptide 2 that they are based on. This was an
expected consequence of the rather extensive structural changes that were needed in order to
arrive at downsized structures. To our knowledge, there are only two other examples of
scaffold-based tripeptidomimetic CXCR4 antagonists in the literature (Figure 7): Niida et al.
2
4
used a 1,3,6-trisubstituted 1,6-dihydropyridin-2-one scaffold, while Ueda et al. have
reported a series of achiral CXCR4 antagonists based on a 1,2,5-trisubstituted indole
2
5
scaffold. Also for these compounds, a significant drop in activity compared to the parent
cyclopentapeptides was observed; the same group has typically reported an IC -value of
5
0
0
.008 μM for the cyclopentapeptide 2, which means that the affinity reduction for the
dihydropyridin-2-one based compound (IC = 15.1 μM) and the indole-based compound
5
0
(
IC = 3.0 μM) was 1888- and 375-fold, respectively. These numbers reflect the general
50
complexity of the initial “scaffold jump” for prototype compounds, as also seen in the present
study.
1
0

Figure 7. Structures of previously reported scaffold-based tripeptidomimetic CXCR4
2
4,25
antagonists.
The moderate activity of our prototype bicyclic tripeptidomimetics 13 and 14 also means that
binding mode studies (typically performed by combining site-directed receptor mutagenesis
studies and molecular docking) are not expected to provide reliable data for these
compounds. Thus, further SAR studies, aimed at optimizing both the scaffold core and the
side chains, are currently in progress.
2
.4. Chemistry.
The macrocyclic compounds 2 and 5–10 were prepared by head-to-tail macrolactamization of
linear precursors (Scheme 1) that were obtained through Fmoc-based solid phase peptide
synthesis (SPPS). The linear precursors were prepared using either a preloaded Fmoc-Gly
trityl resin (2, 5, and 10) or a 2-chlorotrityl chloride resin for the loading of the N-Fmoc ω-
aminocarboxylic acids (6–9). Cleavage from the resin was facilitated using
hexafluoroisopropanol and the side-chain protected peptides were cyclized using PyBOP
followed by global deprotection using TFA.
1
1

Scheme 1. Synthesis of 2 and 5–10. Reagents and conditions: (a) N-Fmoc ω-amino-
carboxylic acid, DIPEA, CH Cl , rt, 1 h; (b) MeOH, DIPEA, CH Cl , rt, 2 x 5 min (c) HFIP,
2
2
2
2
CH Cl , rt, 15 min, then 2 x 10 min; (d) PyBOP, DIPEA, DMF/CH Cl , rt, 24 h; (e)
2
2
2
2
TFA/TIS/H O, rt, 2 h.
2
Compound 3 was synthesized (Scheme 2) by coupling protected arginine 15 with 2-
naphthalene-2-yl)ethanamine to give 16, which in turn was Fmoc-deprotected and coupled
(
with N-Boc 5-aminopentanoic acid to give 17. Global deprotection facilitated by TFA
followed by guanidinylation of the primary amine gave 3.
Scheme 2. Synthesis of 3. Reagents and conditions: (a) 2-(naphthalene-2-yl)ethanamine
hydrochloride, HBTU, DIPEA, DMF, rt, 20 h; (67%); (b) 2-ethanolamine, DMF, rt, 3 h; (c)
N-Boc 5-aminopentanoic acid, HATU, DIPEA, DMF, 24 h; (d) TFA/TIS/H O, rt, 2 h; (e)
2
1
H-pyrazole-1-carboxamidine hydrochloride, DIPEA, DMF, rt, 48 h (21% over four steps).
The linear tripeptide 4 was prepared by Fmoc-based SPPS on an Fmoc-NH-Rink amide resin
Scheme 3) followed by acetylation of the N-terminal using acetic anhydride. Cleavage from
(
the resin and global deprotection mediated by TFA gave the desired N-acetylated tripeptide
amide 4.
1
2

Scheme 3. Synthesis of 4. Reagents and conditions: (a) Ac O, DIPEA, DMF, rt, 30 min; (b)
2
TFA/TIS/H O, rt, 3 h.
2
The key step in the synthesis of the bicyclic tripeptidomimetics 11–14 (Scheme 4) is the
spontaneous formation of the 6,6-fused bicyclic ring system A upon treatment of acetal C
2
1,22
with TFA.
The resulting aldehyde condensates with the backbone amide nitrogen to give
the N-acyliminium ion intermediate B, which subsequently undergoes nucleophilic attack
2
1,22
from the deprotected thiol, resulting in the formation of the desired bicycle A.
The
2
3
cyclization occurs stereoselectively, and the configuration at the bridge-head (C9a) is
1
1
dependent on the configuration at C3 (R substituent), and in the absence of a R substituent
on the configuration at C6.
Scheme 4. Scaffold and retrosynthetic strategy.
Synthesis of the linear precursor C requires access to the three building blocks D, E and F.
1
For the synthesis of target compounds 11 and 12, the R -side chain (incorporated through
building block F) was introduced as guanidinylated glycine 20 (Scheme 5). This material was
prepared by guanidinylation of glycine methyl ester hydrochloride (18) using N,N-di-Boc-
1
H-pyrazole-1-carboxamidine followed by hydrolysis of the methyl ester of the resulting 19
2
6,27
using LiOH in a mixture of water and acetone.
Scheme 5. Synthesis of carboxylic acid 20. Reagents and conditions: (a) N,N-di-Boc-1H-
pyrazole-1-carboxamidine, DIPEA, DMF, rt, 14 h (85%); (b) LiOH, H O/acetone, rt, 1.5 h
2
(99%).
Synthesis of the bicyclic core (Scheme 6) commenced with the alkylation of 2-(naphthalene-
-yl)ethanamine (21) with bromoacetaldehyde dimethyl acetal in refluxing THF to give
2
1
3

secondary amine 22. This amine was in turn coupled with protected arginine (15) to give 23
in high yield. Further Fmoc-deprotection and coupling with appropriately protected L-
cysteine gave 24, which was submitted to another Fmoc-deprotection and then coupled with
carboxylic acid 20 to give the linear precursor 25. This material was treated with TFA,
thioanisole and water to facilitate global deprotection, leading to formation of the
acyliminium ion intermediate that after nucleophilic attack by the thiol gave 11. The
diastereomeric 12 was prepared by coupling of Fmoc-D-Cys(Trt)-OH to Fmoc-deprotected
2
3 to give intermediate 26 (see Experimental section), which in turn was converted to linear
precursor 27 followed by deprotection and cyclization to give 12.
Scheme 6. Synthesis of the bicyclic tripeptidomimetic 11. Reagents and conditions: (a)
BrCH CH(OMe) , THF, reflux, 24 h (30%); (b) Fmoc-Arg(Pbf)-OH (15), HATU, DIPEA,
2
2
DMF, rt, 16 h (83%); (c) Et HN, CH Cl , rt, 2 h; (d) Fmoc-L-Cys(Trt)-OH, HATU, DIPEA,
2
2
2
CH Cl , rt, 22 h (49% over two steps); (e) Et HN, CH Cl , rt, 2 h; (f) 20, HATU, DIPEA,
2
2
2
2
2
CH Cl , rt, 18 h (57% over two steps); (g) TFA/thianisole/H O, rt, 2.5 h (13% after HPLC
2
2
2
purification).
The configuration of the newly formed stereocenters at the bridge-head carbon atoms (C9a)
1
in 11 and 12 (see Figure 5) was determined using the 2D H ROESY experiment (see
supplementary data for detailed ROESY spectra for 11 and 12). The known configurations of
C6 (S for both 11 and 12) and C3 (R for 11 and S for 12) were used as prerequisites for
determination of the configuration of C9a. The strong cross-peaks observed at δ 5.09/4.71
2
(
H9a/H3), δ 5.09/1.73 (H9a/Hβ arginine R ) and the medium strong cross-peak at δ 5.09/1.44
2
(
H9a/Hγ arginine R ) observed in the 2D ROESY spectrum of 11 confirmed the (S)
1
4

configuration of C9a in 11. Moreover, the strong cross-peak observed at δ 5.00/4.72
H9a/H3) in the 2D ROESY spectrum of 12 and the presence of only very weak cross-peaks
(
2
2
at δ 5.00/1.64 (H9a/Hβa arginine R ) and δ 5.00/1.52 (H9a/Hβb arginine R ) confirmed the
R) configuration of C9a in 12.
(
The F building block required for 13 and 14 required a multi-step synthesis (Scheme 7). 2-
Oxopiperidine derivative 28 was selectively reduced using freshly prepared Ca(BH ) , and
4
2
2
8
after an acidic work-up, alcohol 29 was isolated in good yield. The alcohol was next taken
2
8
through a carbodiimide-mediated dehydration to give ,-unsaturated lactam 30. Use of N-
3-dimethylaminopropyl)-N′-ethylcarbodiimide (EDC) in refluxing toluene gave up to 73%
(
yield, while DCC gave the product in quantitative yield. In both cases, CuI was added to
activate the carbodiimide. Next, the lactam was Boc-protected to give 31 followed by
Michael addition of triphenylmethane thiol to give racemic 32, with excellent yield in both
steps. Finally, the Boc-protected lactam was hydrolyzed using LiOH to give racemic
carboxylic acid 33 in high yield.
Scheme 7. Synthesis of building block 33. Reagents and conditions. (a) CaCl , NaBH ,
2
4
MeOH, 0 C, 2 h, then rt, 16 h (76%); (b) DCC, CuI, toluene, 110 C, 1 h (quant); (c) Boc O,
2
Et N, DMAP, CH Cl /DMF, rt, 14 h (94%); (d) Ph CSH, Et N, CH Cl , rt, 14 h (95%); (e)
3
2
2
3
3
2
2
1
M aq. LiOH, THF, rt, 1 h (94%).
Assembly of the linear precursors for 13 and 14 (Scheme 8) was carried out by Fmoc-
deprotecting 23 and subsequent coupling of the resulting material with carboxylic acid 33, to
give 34 as an inseparable mixture of diastereoisomers. Linear precursor 34 was globally
deprotected and cyclized, and the amino group was guanidinylated to give 13 and 14.
1
5

Scheme 8. Synthesis of 13 and 14. Reagents and conditions. (a) Et HN, CH Cl , rt, 3 h; (b)
2
2
2
3
3, HBTU, DIPEA, CH Cl , rt, 17 h (85% over two steps); (c) TFA/thioanisole/H O, rt, 1.5
2 2 2
h; (d) 1H-pyrazole-1-carboxamidine hydrochloride, DIPEA, DMF, rt, 72 h (7% for 13, 11%
for 14, over two steps and after HPLC purification).
RP-HPLC analysis after guanidinylation showed two distinct peaks, which were separable by
semi-prep RP-HPLC. NMR analysis of the two isolated products clearly showed that they
were single diastereoisomers. For the two isolated products, only the configuration at C6 (S)
1
was known, and the configuration of C3 and C9a was determined using the 2D H ROESY
experiment (see supplementary data). For both 13 and 14, strong cross-peaks between H9a
and H3 (at δ 5.14/2.74 and at δ 4.66/2.38 for 13 and 14, respectively) were observed,
indicating similar overall geometry to that of 11 and 12. The presence of a cross-peak
2
between H9a and Hγ arginine R was observed only in the ROESY spectrum of 13
suggesting that the configuration of C9a for this compound is S, whereas the configuration of
C9a for the stereoisomer 14 is R. Since H9a and H3 are cis in both 13 and 14, the
configuration at C3, which stems from the racemic carboxylic acid 33, is R in 13 and S in 14.
It is interesting to note that S configuration of C9a in 11 and 13 leads to a downfield shift of
H6 when compared with the analogous R isomer. Grimes et al. speculated that a solvent
1
mediated hydrogen bond between the R -side chain amide NH and the ring carbonyl group
1
could contribute to stabilizing a conformation in which the R substituent is in an equatorial
23
position, which favors the observed stereoselectivity of the cyclization. However, for the
1
formation of 13 and 14, the absence of a R amide NH did not influence on the
stereoselectivity in the cyclization step.
3
. CONCLUSIONS
In this work we have demonstrated that new scaffold-based tripeptidomimetic CXCR4
antagonists can be rationally designed from cyclopentapeptide CXCR4 antagonists. The
bicyclic compounds reported herein represent an interesting class of new tripeptidomimetic
CXCR4 antagonists, and although the prototype compounds showed moderate activity, they
1
6

serve as useful leads for further optimization. The peptidomimetic scaffold we have
employed is constructed from three building blocks, each containing one of the
pharmacophoric groups, and therefore allows for synthetic access to a range of target
molecules. We envision that further SAR studies involving the three different binding groups
will afford new and optimized CXCR4 antagonists.
4
4
. EXPERIMENTAL SECTION
.1. Chemistry.
All reagents and starting materials were purchased from Sigma-Aldrich and used as delivered
1
3
unless otherwise stated. Cyclic peptide 2 was prepared as previously described. The
preparation of 4–10 is described in the supplementary data. Anhydrous toluene, CH Cl and
2
2
THF were obtained from an anhydrous solvent delivery system (SDS-800 from mBraun) at
the Department of Chemistry, University of Bergen. Analyses using thin layer
chromatography were performed on Alugram® SIL G/UV254 0.20 mm layer plates from
Machery-Nagel or on aluminum sheets with Merck silica gel (60 F ). TLC plates were
2
54
visualized using either ultraviolet light or by immersing the plate in 2% solution of ninhydrin
in ethanol containing 10 drops of concentrated sulphuric acid pr 100 mL followed by heating.
Purification by flash column chromatography was performed using J.T Baker Silica Gel or
Merck 60 Kieselgel (230-400 mesh). All final compounds were purified using semi-
preparative RP-HPLC eluting with mixtures of acetonitrile and H O (both containing 0.1%
2
TFA). Fractions of equal purity were pooled and lyophilized. All tested compounds were
analyzed by RP-HPLC and found to be of >95% purity (UV 220 nm).
4
.1.1. (9H-Fluoren-9-yl)methyl (S)-(1-((2-(naphthalen-2-yl)ethyl)amino)-1-oxo-5-(3-
((2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-yl)sulfonyl)guanidino)pentan-2-
yl)carbamate (16)
HBTU (945 mg, 2.4 mmol) and DIPEA (0.66 mL, 3.7 mmol) were added to a stirred solution
of Fmoc-Arg(Pbf)-OH (809 mg, 1.2 mmol) in dry DMF (3 mL) under an argon atmosphere.
The mixture was stirred at room temperature for 30 min before a solution of 2-(naphthalen-2-
yl)ethan-1-amine hydrochloride (518 mg, 2.5 mmol) in DMF (2 mL) was added drop-wise to
the reaction mixture, and stirring continued for 20 h. The solvent was evaporated and the
residue was partitioned between EtOAc (20 mL) and H O (10 mL). The organic phase was
2
washed with H O (2 x 10 mL), 5% KHCO (20 mL) and brine (20 mL), dried over MgSO ,
2
3
4
1
7

filtered and evaporated. The crude product was purified by flash chromatography on silica
gel (EtOAc/hexane; gradient 1:1 to 9:1) to give the title compound as a white solid (706 mg,
1
6
7
4
2
1
1
1
4
7%). R (EtOAc/hexane 9:1) = 0.23; H NMR (400 MHz, CDCl ) δ = 7.75 – 7.67 (m, 5H),
f
3
.60 – 7.49 (m, 3H), 7.41 – 7.32 (m, 4H), 7.31 – 7.26 (m, 2H), 7.26 – 7.23 (m, 1H), 4.34 –
.20 (m, 2H), 4.17 – 4.06 (m, 2H), 3.66 – 3.45 (m, 2H), 3.24 – 3.08 (m, 2H), 2.97 – 2.92 (m,
H), 2.90 – 2.87 (m, 2H), 2.81 – 2.79 (m, 2H), 2.53 (s, 3H), 2.46 (s, 3H), 2.05 (s, 3H), 1.48 –
1
3
.40 (m, 8H); C NMR (101 MHz, CD OD) δ = 174.9, 173.4, 165.3, 160.3, 158.67, 158.66,
3
58.4, 145.7, 145.5, 143.0, 143.0, 139.9, 138.3, 135.4, 134.1, 134.0, 129.5, 129.20, 129.18,
29.0, 128.78, 128.74, 128.6, 127.4, 126.8, 126.6, 126.5, 121.3, 118.9, 88.1, 68.2, 62.0, 56.6,
4.3, 42.0, 39.3, 37.4, 36.9, 32.1, 30.9, 29.0, 21.3, 20.0, 18.8, 14.9, 13.0; HRMS (ESI): m/z
+
[
M + H] calcd for C H N O S: 802.3633; found: 802.3639.
4
6
52
5
6
4
.1.2. (S)-5-Guanidino-2-(5-guanidinopentanamido)-N-(2-(naphthalen-2-
yl)ethyl)pentanamide (3)
To a solution of the Fmoc-protected amine 16 (610 mg, 0.77 mmol) in DMF (6 mL) was
added 2-ethanolamine (6 mL) and the mixture was stirred for 3 h at room temperature under
HPLC monitoring. The solvent was removed in vacuo, the residue was dissolved in EtOAc,
washed with H O (2 x 10 mL), saturated NaHCO (2 x 10 mL) and brine (15 mL). The
2
3
solvent was evaporated and the crude product (500 mg) was used in the next step without
+
further purification. HRMS (ESI): m/z [M + H] calcd for C H N O S: 580.2952; found:
3
1
41
5
4
5
80.2958.
HATU (590 mg, 1.5 mmol) and DIPEA (0.41 mL, 2.3 mmol) were added to a stirring
solution of 5-((tert-butoxycarbonyl)amino)pentanoic acid (338 mg, 1.5 mmol) in dry DMF (3
mL) under an argon atmosphere. The mixture was stirred at room temperature for 30 min
before a solution of the crude product from the previous step (451 mg, 0.78 mmol) in DMF
(1.5 mL) was added drop wise to the reaction mixture, and stirring continued for 24 h. The
reaction mixture was partitioned between EtOAc (30 mL) and H O (20 mL). The aqueous
2
layer was extracted with EtOAc (2 x 15 mL) and the combined organic layer washed with
1
0% citric acid (15 mL), 5% KHCO , (15 mL) and brine (15 mL), and dried over MgSO .
3 4
Filtration and removal of the solvent under reduced pressure gave the crude product (520
mg). Purification by flash chromatography (EtOAc/hexane; gradient 1:1 to pure EtOAc,
followed by EtOAc/MeOH; gradient 9:1 to 8:2) afforded 17 as colorless foam (358 mg)
+
which was judged to be of sufficient purity for the next step. HRMS (ESI): m/z [M + H]
calcd for C H N O S: 779.4160; found: 779.4168.
4
1
59
6
7
1
8

The Boc-protected amine 17 (358 mg, 0.46 mmol) was dissolved in a mixture of TFA, TIS
and H O (95:2.5:2.5, 15 mL) and the resulting solution was stirred at room temperature. The
2
reaction was monitored using analytical RP-HPLC and all starting material was found to be
consumed after 2 h. The TFA solution was evaporated and the residue was precipitated by
addition of cold diethyl ether (10 mL) and cooled overnight. The ether was decanted and the
residue was dried in vacuo to give the crude product (403 mg), which was used in the next
+
step without further purification. HRMS (ESI): m/z [M + H] calcd for C H N O :
2
3
34
6
2
4
27.2816; found: 427.2816.
To a stirring solution of the crude primary amine (403 mg, 0.61 mmol) in DMF (3 mL) was
added 1H-pyrazole-1-carboxamidine hydrochloride (453 mg, 3.0 mmol) and DIPEA (0.53
mL, 3.1 mmol) and the resulting mixture was stirred under nitrogen atmosphere for 48 h. The
reaction was monitored using analytical RP-HPLC. After the solvent had been evaporated,
the residue was precipitated by addition of cold diethyl ether, washed with ether, purified by
1
preparative RP-HPLC, and lyophilized (129 mg, 21% over four steps). H NMR (400 MHz,
CD OD): δ = 7.59 – 7.51 (m, 3H), 7.42 (s, 1H), 7.24 – 7.10 (m, 3H), 4.01 (dd, J = 8.3, 5.7,
3
1
7
1
4
H), 3.39 – 3.23 (m, 2H), 2.90 (t, J = 6.7, 2H), 2.80 (m, 2H), 2.73 (t, J = 7.0, 2H), 2.00 (t, J =
1
3
.0, 2H), 1.51 – 1.09 (m, 8H); C NMR (101 MHz, CD OD): δ = 175.9, 174.3, 158.84,
3
58.81, 138.0, 135.2, 133.9, 129.3, 128.8, 128.7, 128.54, 128.48, 127.2, 126.6, 54.7, 42.2,
+
2.0, 41.8, 36.7, 36.1, 30.4, 29.5, 26.4, 23.8; HRMS (ESI): m/z [M + H] calcd for
C H N O : 469.3034; found: 469.3033.
24
36
8
2
2
6
4
.1.3. Methyl (N,N'-bis(tert-butoxycarbonyl)carbamimidoyl)glycinate (19)
To a slurry of glycine methyl ester hydrochloride (1.11 g, 8.8 mmol) in DMF (30.5 mL) was
added N,N-di-Boc-1H-pyrazole-1-carboxamidine (1.02 g, 3.3 mmol) and DIPEA (1.1 mL, 6.3
mmol). The mixture was stirred overnight (14 h) at room temperature and turned into a
yellow slurry. The reaction mixture was partitioned between EtOAc (25 mL) and H O (20
2
mL) and the aqueous layer was further extracted three times with EtOAc (25 mL), before the
combined organic layers were dried over anhydrous MgSO . The solvent was evaporated
4
under reduced pressure to give the crude product as a light yellow solid (1.4 g). Further
purification using flash column chromatography (Hexanes/EtOAc, 3:1) gave the title
1
compound as a colourless solid (932 mg, 85 %). R (Hexanes/EtOAc, 3:1) = 0.27; H NMR
f
(
400 MHz, CDCl ) δ = 11.44 (s, 1H), 8.87 (s, 1H), 4.25 (d, J = 4.9, 2H), 3.78 (s, 3H), 1.51 (s,
3
1
3
9
H), 1.50 (s, 9H); C NMR (100 MHz, CDCl ) δ = 170.2, 163.6, 156.3, 153.3, 83.7, 79.9,
3
1
9

+
5
3
2.8, 43.0, 28.6, 28.4; HRMS (ESI): m/z [M + H] calcd for C H N O : 332.1822; found:
14 26 3 6
32.1821.
2
7
4
.1.4. (N,N'-bis(tert-Butoxycarbonyl)carbamimidoyl)glycine (20)
LiOH·H O (166 mg, 4.0 mmol) was added to a solution of methyl ester 19 (330 mg, 1 mmol)
2
in acetone/H O (6:1, 4.5 mL) at room temperature. TLC monitoring showed full consumption
2
of the starting material after 1.5 h, and the reaction mixture was cooled to 0 °C before a 0.5
M aqueous HCl solution was added drop-wise until pH 3. The mixture was allowed to warm
to room temperature and extracted with EtOAc (3 x 20 mL). Drying over anhydrous Na SO
2
4
and evaporation of the solvent gave the title compound as a colourless solid (312 mg, 99 %).
1
R (Hexanes/EtOAc/AcOH, 60:40:1) = 0.20; H NMR (400 MHz, CDCl ) δ = 11.37 (s, 1H),
f
3
1
3
8
1
.87 (s, 1H), 4.18 (s, 2H), 1.51 (s, 9H), 1.49 (s, 9H); C NMR (100 MHz, CDCl ) δ = 171.7,
3
+
62.2, 157.0, 153.1, 84.4, 80.9, 44.3, 28.5, 28.3; HRMS (ESI): m/z [M + H] calcd for
C H N O : 318.1665; found: 318.1657.
13
23
3
6
4
.1.5. 2,2-Dimethoxy-N-(2-(naphthalen-2-yl)ethyl)ethan-1-amine (22)
2
-(Napthalene-2-yl)ethyl-1-amine (21) (322 mg, 1.9 mmol) and bromoacetataldehyde
dimethyl acetal (0.2 mL, 1.7 mmol) were dissolved in dry THF (4 mL), and the mixture was
heated at a gentle reflux for 24 h. After the solvent was evaporated, the residue was dissolved
in CH Cl (20 mL) and the organic phase was washed with saturated NaHCO (2 x 10 mL)
2
2
3
and brine. The mixture was concentrated and purified by flash chromatography on silica gel
1
(
EtOAc) to give the title compound (130 mg, 30%) as yellow oil. R (EtOAc) = 0.14; H
f
NMR (400 MHz, CDCl ): δ = 7.78 (t, J = 8.6, 3H), 7.64 (s, 1H), 7.49 – 7.37 (m, 2H), 7.33 (d,
3
1
3
J = 8.4, 1H), 4.45 (t, J = 5.4, 1H), 3.34 (s, 6H), 2.96 (s, 4H), 2.78 (d, J = 5.4, 2H); C NMR
(
100 MHz, CDCl ): δ = 137.7, 133.8, 132.4, 128.3, 127.9, 127.7, 127.5, 127.2, 126.2, 125.5,
3
+
1
04.2, 54.2, 51.39, 51.36, 36.8; HRMS (ESI): m/z [M + H] calcd for C H NO : 260.1645;
1
6
22
2
found: 260.1646.
4
.1.6. (9H-Fluoren-9-yl)methyl (S)-(1-((2,2-dimethoxyethyl)(2-(naphthalen-2-
yl)ethyl)amino)-1-oxo-5-(3-((2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-
yl)sulfonyl)guanidino)pentan-2-yl)carbamate (23)
HATU (183 mg, 0.48 mmol) and DIPEA (84 μL, 0.48 mmol) were added to a stirred solution
of Fmoc-Arg(Pbf)-OH (292 mg, 0.45 mmol) in dry DMF (1.2 mL) under an argon
atmosphere. The mixture was stirred at room temperature for 30 min before amine 22 (125
2
0

mg, 0.48 mmol) in DMF (0.6 mL) was added drop-wise to the reaction mixture, and stirring
continued for 16 h. The reaction mixture was partitioned between H O (10 mL) and EtOAc
2
(
20 mL). The organic phase was further washed with H O (2 x 10 mL) and brine (2 x 10 mL),
2
dried over MgSO , filtered and evaporated. The crude product was purified by flash
4
chromatography on silica gel (EtOAc/hexane 8.5:1.5) to give the title compound as white
1
13
foam (334 mg, 83 %, retains EtOAc). R (EtOAc/hexane 8.5:1.5) = 0.28; H NMR and C
f
+
NMR gave no useful information due to rotamers; HRMS (ESI): m/z [M + H] calcd for
+
C H N O S: 890.4157; found: 890.4166; m/z [M + Na] calcd for C H N O SNa:
50
60
5
8
50 59
5
8
9
12.3977; found: 912.3976.
4
.1.7. (9H-Fluoren-9-yl)methyl ((7S,10R)-3-methoxy-5-(2-(naphthalen-2-yl)ethyl)-6,9-
dioxo-7-(3-(3-((2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-
yl)sulfonyl)guanidino)propyl)-13,13,13-triphenyl-2-oxa-12-thia-5,8-diazatridecan-10-
yl)carbamate (24)
Et NH (1.5 mL) was added to a solution of the Fmoc-protected amine 23 (261 mg, 0.29
2
mmol) in CH Cl (3 mL) and the mixture was stirred at room temperature for 2 h with TLC
2
2
monitoring. The solvent was removed under reduced pressure, to give a light yellow foam.
The crude product (230 mg) was used directly in the next step.
To a solution of crude Fmoc-deprotected 23 (0.29 mmol) in CH Cl (2 mL), were added
2
2
Fmoc-Cys(Trt)-OH (183 mg, 0.31 mmol), HATU (112 mg, 0.30 mmol) and DIPEA (0.14
mL, 0.8 mmol), and the mixture was stirred for 22 h at room temperature. After evaporation
of the solvent, the residue was partitioned between EtOAc (15 mL) and H O (50 mL). The
2
aqueous layer was extracted with EtOAc (2 x 15 mL), and the combined organic layer was
washed with 1M KHSO (30 mL), H O (30 mL), saturated NaHCO (30 mL), brine (30 mL)
4
2
3
and was dried over anhydrous MgSO . Removal of the solvent under reduced pressure gave
4
the crude product as an orange colored foam (419 mg). Purification by flash column
chromatography (EtOAc/hexanes, 8:2) gave 24 as a colorless foam (194 mg, 49 % over 2
1
steps). H NMR showed that the product retained EtOAc, and the yield is adjusted
1
13
accordingly. R (EtOAc/hexane, 8:2) = 0.16; H NMR and C NMR gave no useful
f
+
information due to rotamers; HRMS (ESI): m/z [M + Na] calcd for C H N O S Na:
7
2
78
6
9 2
1
257.5169; found: 1257.5172.
4
6
.1.8. tert-Butyl ((7S,10R)-3-methoxy-19,19-dimethyl-5-(2-(naphthalen-2-yl)ethyl)-
,9,12,17-tetraoxo-7-(3-(3-(2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-
2
1

ylsulfonyl)guanidino)-propyl)-10-(tritylthiomethyl)-2,18-dioxa-5,8,11,14,16-
pentaazaicosan-15-ylidene)carbamate (25)
Et NH (0.73 mL) was added to a solution of the Fmoc-protected amine 24 (181 mg, 0.15
2
mmol) in CH Cl (1.5 mL) at room temperature. TLC monitoring showed no sign of the
2
2
starting material after 2 h and the solvent was removed under reduced pressure to give a light
yellow foam. The crude product (183 mg) was used directly in the next step. HRMS (ESI):
+
m/z [M + H] calcd for C H N O S : 1013.4669; found: 1013.4672.
5
7
69
6
7 2
To a solution of carboxylic acid 20 (183 g, 0.31 mmol) in CH Cl (1 mL), were added HATU
2
2
(60 mg, 0.16 mmol), DIPEA (0.07 mL, 0.40 mmol) and Fmoc-deprotected 24 (65 mg, 0.2
mmol) and the resulting yellow reaction mixture was allowed to stir for 18 h at room
temperature. The solvent was evaporated and the residue was partitioned between EtOAc (10
mL) and H O (10 mL). The aqueous layer was extracted with EtOAc (2 x 10 mL), and the
2
combined organic layer was washed with 1M KHSO (20 mL), H O (20 mL), saturated
4
2
NaHCO (20 mL), brine (20 mL) and was dried over anhydrous MgSO . Removal of the
3
4
solvent under reduced pressure gave the crude product as a yellow foam (203 mg).
Purification by flash column chromatography (EtOAc/hexanes, 9:1) gave the title compound
1
as a light yellow foam (101 mg, 57 % over 2 steps). H NMR showed that the product
1
retained EtOAc, and the yield is adjusted accordingly. R (EtOAc/hexanes, 9:1) = 0.21; H
f
1
3
+
NMR and C NMR gave no useful information due to rotamers; HRMS (ESI): m/z [M + H]
calcd for C H N O S : 1312.6145; found: 1312.6146.
7
0
90
9
12 2
4
4
.1.9. 2-Guanidino-N-((3R,6S,9aS)-6-(3-guanidinopropyl)-8-(2-(naphthalen-2-yl)ethyl)-
,7-dioxohexahydro-2H,6H-pyrazino[2,1-b][1,3]thiazin-3-yl)acetamide (11)
Linear precursor 25 (92 mg, 0.07 mmol) was dissolved in a mixture of TFA, thioanisole and
distilled H O (90:5:5, 8 mL) and stirred for 2.5 h at room temperature. After evaporation of
2
the solvent under reduced pressure the brown residue was cooled to 0 °C. Addition of cold
ether resulted in precipitation and the crude solid was purified by preparative RP-HPLC, and
1
lyophilized to give 11 as a fluffy white solid (7.2 mg, 13%). H NMR (600 MHz, CD OD): δ
3
=
7.82 – 7.79 (m, 3H), 7.70 (s, 1H), 7.47 – 7.41 (m, 3H), 5.09 (dd, J = 7.4, 4.0, 1H), 4.95 (dd,
J = 9.3, 6.1, 1H), 4.71 (t, J = 8.8, 1H), 4.03 – 3.97 (m, 3H), 3.81 (dd, J = 13.3, 7.5, 1H), 3.64
(dd, J = 13.4, 4.0, 1H), 3.53 (m, 1H), 3.11 – 3.01 (m, 6H), 1.79 – 1.66 (m, 2H), 1.48 – 1.40
1
3
(
m, 2H); C NMR (150.9 MHz, CD OD) δ = 170.0, 169.6, 168.9, 159.5, 158.5, 137.3, 135.1,
3
1
4
33.9, 129.3, 128.7, 128.5, 128.44, 128.40, 127.2, 126.7, 57.3, 52.27, 52.26, 51.4, 49.5, 44.7,
2
+
1.6, 34.4, 29.8, 27.1, 26.3; HRMS (ESI): m/z [M + 2H] (z = 2) calcd for C H N O S:
26
37
9
3
2
2

+
2
27.6365; found: 227.6359; HRMS (ESI): m/z [M + H] calcd for C H N O S: 554.2656;
26 36 9 3
found: 554.2659.
4
.1.10. (9H-fluoren-9-yl)methyl ((7S,10S)-3-methoxy-5-(2-(naphthalen-2-yl)ethyl)-6,9-
dioxo-7-(3-(3-((2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-
yl)sulfonyl)guanidino)propyl)-13,13,13-triphenyl-2-oxa-12-thia-5,8-diazatridecan-10-
yl)carbamate (26)
Et NH (7 mL) was added to a solution of the Fmoc-protected amine 23 (308 mg, 0.33 mmol)
2
in CH Cl (7 mL) and the reaction was stirred for 3 h at room temperature with HPLC
2
2
monitoring. The solvent was removed under reduced pressure and the crude product (320
+
mg) was used directly in the next step. HRMS (ESI): m/z [M + H] calcd for C H N O S:
3
5
50
5
6
6
68.3476; found: 668.3488.
Fmoc-D-Cys(Trt)-OH (364 mg, 0.62 mmol), HATU (236 mg, 0.62 mmol) and DIPEA (0.25
mL, 1.4 mmol) were dissolved in CH Cl (2 mL) and the resulting mixture was stirred for 30
2
2
min at room temperature, before a solution of crude Fmoc-deprotected 23 (320 mg, 0.48
mmol) in CH Cl (3 mL) was added. The reaction mixture was stirred for 24 h at room
2
2
temperature under anhydrous conditions (CaCl ). After evaporation of the solvent the residue
2
was partitioned between EtOAc (20 mL) and H O (40 mL). The aqueous layer was extracted
2
with EtOAc (2 x 15 mL), and the combined organic layer was washed with 1M KHSO (30
4
mL), H O (30 mL), saturated NaHCO (30 mL), brine (30 mL) and was dried over anhydrous
2
3
MgSO . Removal of the solvent under reduced pressure gave the crude product as a light red
4
colored foam (710 mg). Purification by flash column chromatography (EtOAc/hexanes, 8:2)
gave the title compound as a colorless foam (237 mg, 45% over two steps). Rf
+
(
EtOAc/hexane, 8:2) = 0.19; HRMS (ESI): m/z [M + H] calcd for C H N O S :
7
2
78
6
9 2
1
235.5344; found: 1235.5334.
4
6
.1.11. tert-Butyl ((7S,10S)-3-methoxy-19,19-dimethyl-5-(2-(naphthalen-2-yl)ethyl)-
,9,12,17-tetraoxo-7-(3-(3-(2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-
ylsulfonyl)guanidino)-propyl)-10-(tritylthiomethyl)-2,18-dioxa-5,8,11,14,16-
pentaazaicosan-15-ylidene)carbamate (27)
Et NH (6 mL) was added to a solution of the Fmoc-protected amine 26 (237 mg, 0.19 mmol)
2
in CH Cl (6 mL) and the reaction was stirred for 3 h at room temperature with HPLC
2
2
monitoring. The solvent was removed under reduced pressure and the crude product (277
2
3

+
mg), was used directly in the next step. HRMS (ESI): m/z [M + H] calcd for C H N O S :
5
7
69
6
7 2
1
013.4664; found: 1013.4663
To a solution of carboxylic acid 20 (150 mg, 0.47 mmol) in CH Cl (2 mL), were added
2
2
HATU (156 mg, 0.41 mmol) and DIPEA (0.15 mL, 0.82 mmol). The mixture was stirred at
room temperature under anhydrous environment (CaCl ) for 15 min before a solution of the
2
Fmoc-deprotected amine 26 (277 mg, 0.27 mmol) in CH Cl (2 mL) was added drop-wise to
2
2
the reaction mixture, and stirring continued for 24 h. After evaporation of the solvent the
residue was partitioned between EtOAc (20 mL) and H O (30 mL). The aqueous layer was
2
extracted with EtOAc (2 x 15 mL), and the combined organic layer was washed with 1M
KHSO (30 mL), H O (30 mL), saturated NaHCO (30 mL), brine (30 mL), and was dried
4
2
3
over anhydrous MgSO . Removal of the solvent under reduced pressure gave the crude
4
product (352 mg). Purification by flash column chromatography (EtOAc/hexanes, 9:1) gave
the title compound (93 mg, 37% over 2 steps). R (EtOAc/hexane, 9:1) = 0.45; NMR analyses
f
+
did not give useful information due to formation of rotamers; HRMS (ESI): m/z [M + H]
calcd for C H N O S : 1312.6145; found: 1312.6141.
7
0
89
9
12 2
4
4
.1.12. 2-guanidino-N-((3S,6S,9aS)-6-(3-guanidinopropyl)-8-(2-(naphthalen-2-yl)ethyl)-
,7-dioxohexahydro-2H,6H-pyrazino[2,1-b][1,3]thiazin-3-yl)acetamide (12)
Linear precursor 27 (76 mg, 0.06 mmol) was dissolved in a mixture of TFA, thioanisole and
distilled H O (90:5:5, 7 mL) and stirred for 2 h. After evaporation of the solvent under
2
reduced pressure the brown residue was cooled to 0 °C. Addition of cold ether resulted in
precipitation and the crude product was purified by preparative RP-HPLC, and lyophilized to
1
give the title compound as a fluffy white solid (1.5 mg, 3%). H NMR (600 MHz, CD OD): δ
3
=
7.84 – 7.78 (m, 3H), 7.70 (s, 1H), 7.46 – 7.39 (m, 3H), 5.00 (dd, J = 11.0, 5.4, 1H), 4.72
(dd, J = 11.9, 6.3, 1H), 4.64 (t, J = 8.0, 1H), 4.15 – 4.08 (m, 1H), 3.99 (q, J = 29.4, 17.3, 2H),
3
2
1
1
4
5
.70 – 3.55 (m, 3H), 3.36 (dd, J = 10.6, 6.3, 1H), 3.12 – 3.04 (m, 2H), 3.01 – 2.96 (m, 1H),
.90 – 2.85 (m, 1H), 2.73 (t, J = 11.3, 1H), 1.67 – 1.58 (m, 1H), 1.56 – 1.49 (m, 1H), 1.31 –
1
3
.24 (m, 2H); C NMR (150.9 MHz, CD OD) δ = 170.5, 169.9, 169.3, 159.4, 158.4, 137.2,
3
35.0, 133.9, 129.3, 128.7, 128.6, 128.5, 128.4, 127.4, 126.8, 57.6, 52.8, 51.4, 48.9, 48.7,
+
4.6, 41.4, 34.7, 30.7, 28.5, 26.3; HRMS (ESI): m/z [M + H] calcd for C H N O S;
2
6
35
9
3
54.2656; found: 554.2659.
2
4

28
4
.1.13. 3-(Hydroxymethyl)piperidin-2-on (29)
To a stirred suspension of ethyl 2-oxopiperidine-3-carboxylate (28) (1.00 g, 5.8 mmol) and
anhydrous CaCl (702 mg, 6.3 mmol) in dry MeOH (12 mL) in an ice/H O bath, NaBH (501
2
2
4
mg, 13 mmol) was added in one portion, and stirring was continued for 2 h at 0 °C. After 16
h of additional stirring at room temperature, the solvent was evaporated, and a 3N aqueous
citric acid solution was added in small portions until all solid material was dissolved (pH = 2-
3
). The solution was extracted with CH Cl (6 x 50 mL), dried over Na SO and filtered.
2 2 2 4
Evaporation of the solvent gave the crude product (643 mg), which was purified by flash
column chromatography on silica gel (5 % MeOH in EtOAc) to give the title compound as
1
white solid (573 mg, 76 %). R (EtOAc) = 0.11; H NMR (400 MHz, CDCl ): δ = 5.99 (bs,
f
3
1
H), 4.07 (d, J = 8.5, 1H), 3.81 – 3.60 (m, 2H), 3.39 – 3.21 (m, 2H), 2.56 – 2.42 (m, 1H),
1
3
1
.97 – 1.71 (m, 3H), 1.56 – 1.39 (m, 1H); C NMR (100 MHz, CDCl , 25°C): δ = 176.2,
3
+
6
4.9, 43.2, 42.0, 24.1, 22.3; HRMS (ESI): m/z [M + H] calcd for C H NO : 130.0868;
6
12
2
found: 130.0850.
2
8
4
.1.14. 3-Methylenepiperidin-2-one (30)
N,N’-Dicyclohexylcarbodiimide (DCC) (346 mg, 1.7 mmol) was added to a stirring solution
of alcohol 29 (167 mg, 1.3 mmol) in dry toluene (2 mL). The mixture was heated at 110 °C
(oil bath) and CuI (23 mg, 0.12 mmol) was added and the mixture was stirred for 70 min
before it was cooled to room temperature, after which H O (1.6 mL) was added and stirring
2
continued for 1 h. Et O (3.4 mL) was added, and the mixture was filtered. The aqueous phase
2
was separated and extracted four times with 30 mL portions of CH Cl . The combined
2
2
extracts were dried over K CO filtered and evaporated to give the title compound as a white
2
3,
1
solid (144 mg, 100 %). R (EtOAc) = 0.17; H NMR (400 MHz, CDCl ): δ = 6.91 (bs, 1H),
f
3
6
1
.21 (bs, 1H), 5.37 – 5.26 (m, 1H), 3.40 (td, J = 6.2, 2.6, 2H), 2.67 – 2.51 (m, 2H), 1.95 –
1
3
.79 (m, 2H); C NMR (100 MHz, CDCl ): δ = 166.6, 138.1, 122.5, 43.2, 30.4, 23.7; HRMS
3
+
(
ESI): m/z [M + H] calcd for C H NO: 112.0762; found: 112.0779.
6
10
4
.1.15. tert-Butyl 3-methylene-2-oxopiperidine-1-carboxylate (31)
Et N (1.2 mL, 8.8 mmol) and DMAP (1.07 g, 8.8 mmol) were added to a stirring solution of
3
lactam 30 (971 mg, 8.7 mmol) in dry CH Cl /DMF (10:1, 27 mL). Boc O (3.83 g, 17 mmol)
2
2
2
in CH Cl (12 mL) was added drop-wise, and the reaction mixture was stirred at room
2
2
temperature for 14 h. The mixture was concentrated and purified by flash column
chromatography on silica gel (EtOAc/hexane 1:4) to give the title compound (1.72 g, 94 %)
2
5

1
as transparent thick oil. R (EtOAc/hexane 1:4) = 0.34; H NMR (400 MHz, CDCl ): δ = 6.33
f
3
(d, J = 1.4, 1H), 5.41 (d, J = 0.7, 1H), 3.78 – 3.67 (m, 2H), 2.58 (t, J = 6.1, 2H), 1.95 – 1.84
1
3
(
m, 2H), 1.55 (s, 9H); C NMR (100 MHz, CDCl ): δ = 164.8, 153.3, 138.8, 125.1, 83.1,
3
+
4
6.9, 29.5, 28.2, 22.8; HRMS (ESI): m/z [M + Na] calcd for C H NO Na: 234.1102;
1
1
17
3
found: 234.1101.
4
.1.16. tert-Butyl 2-oxo-3-(tritylthiomethyl)piperidine-1-carboxylate (32)
To a stirred solution of lactam 31 (1.65 g, 7.8 mmol) in dry CH Cl (33 mL) was added
2
2
Ph CSH (2.16 g, 7.8 mmol) and Et N (1.1 mL, 7.8 mmol) and the resulting mixture was
3
3
stirred at room temperature for 12 h. The reaction mixture was washed with H O,
2
concentrated and purified by flash column chromatography on silica gel (hexane/EtOAc 8:2)
1
to give the title compound as white foam (3.60 g, 95 %). R (hexane/EtOAc 4:1) = 0.18; H
f
NMR (400 MHz, CDCl ): δ = δ 7.45 – 7.20 (m, 15H), 3.76 – 3.66 (m, 1H), 3.34 (ddd, J =
3
1
1
2.9, 7.6, 5.1, 1H), 2.88 (dd, J = 12.7, 4.6, 1H), 2.28 (dd, J = 12.7, 8.6, 1H), 2.09 – 1.97 (m,
1
3
H), 1.95 – 1.81 (m, 1H), 1.75 – 1.65 (m, 2H), 1.49 (s, 9H), 1.33 – 1.26 (m, 1H); C NMR
(
100 MHz, CDCl ): δ = 172.8, 152.9, 145.0, 129.9, 128.1, 126.9, 83.2, 67.1, 45.1, 43.6, 33.0,
3
+
2
5
8.2, 25.5, 21.7; HRMS (ESI): m/z [M + Na] calcd for C H NO SNa: 510.2073; found:
3
0
33
3
10.2070.
4
.1.17. 5-(tert-Butoxycarbonylamino)-2-((tritylthio)methyl)pentanoic acid (33)
To a stirred solution of lactam 32 (3.60 g, 7.4 mmol) in THF (37 mL), was added LiOH (1.0
M, 15 mL, 15 mmol). The reaction was monitored by TLC and all starting material was
consumed after 1h. The THF was evaporated and the aqueous residue was acidified to pH 4
with 10% citric acid. The mixture was extracted with CH Cl (4 x 50 mL), dried over
2
2
MgSO , filtered and evaporated to give the title compound as white foam (3.52 g, 94%). Rf
4
1
(
hexane/EtOAc 6:4) = 0.20; H NMR (400 MHz, CDCl ): δ = 7.46 – 7.17 (m, 15H), 4.47 (bs,
3
1
1
1
H), 2.97 (bs, 2H), 2.61 (dd, J = 12.5, 8.1 Hz, 1H), 2.22 (dd, J = 12.6, 6.1 Hz, 1H), 2.11 –
1
3
.99 (m, 1H), 1.50 – 1.33 (s and m, 11H), 1.26 (m, 2H); C NMR (100 MHz, CDCl ): δ =
3
79.8, 156.2, 144.8, 129.9, 128.2, 127.0, 79.5, 67.2, 44.9, 40.3, 33.3, 29.0, 28.6, 27.5; HRMS
+
(
ESI): m/z [M + Na] calcd for C H NO SNa: 528.2184; found: 528.2174.
3
0
35
4
2
6

4
1
.1.18. tert-Butyl ((SR)-5-(((S)-1-((2,2-dimethoxyethyl)(2-(naphthalen-2-yl)ethyl)amino)-
-oxo-5-(3-((2,2,4,6,7-pentamethyl-2,3-dihydrobenzofuran-5-
yl)sulfonyl)guanidino)pentan-2-yl)amino)-5-oxo-4-((tritylthio)methyl)pentyl)carbamate
34)
(
Et HN (1.9 mL, 18 mmol) was added to a stirring solution of 23 (334 mg, 3.8 mmol) in
2
CH Cl (3.9 mL) at room temperature. The reaction was monitored by TLC and deemed
2
2
completed after 3 h. The mixture was evaporated to dryness and the residue used in the
following step.
To a stirred solution of carboxylic acid 33 (182 mg, 0.36 mmol) in dry CH Cl (2.1 mL),
2
2
were added HBTU (142 mg, 0.38 mmol) and DIPEA (66 μL, 0.38 mmol) and the mixture
was stirred for 30 min. Fmoc-deprotected amine 23 (250 mg out of 344 mg crude product,
0
.38 mmol) in dry CH Cl (2.1 mL) was added and stirring continued for 17 h. The solvent
2 2
was evaporated to give red foam, which was partitioned between EtOAc (20 mL) and H O
2
(
10 mL). The organic phase was separated and washed with 5% KHSO (10 mL) and brine
4
(10 mL). The organic phase was further concentrated and purified by flash chromatography
on silica gel (EtOAc/hexane 3:1) to give an inseparable mixture of the two diastereoisomers
of 34 as white foam (352 mg, 85 %). The product retained EtOAc and the yield is adjusted
+
accordingly. R (EtOAc/hexane, 3:1) = 0.29; HRMS (ESI): m/z [M + H] calcd for
f
+
C H N O S : 1155.5658; found: 1155.5667; HRMS (ESI): m/z [M + Na] calcd for
65
83
6
9 2
C H N O S Na: 1177.5477; found: 1177.5474.
65
82
6
9 2
4
.1.19. 1,1'-(((3R,6S,9aS)-8-(2-(Naphthalen-2-yl)ethyl)-4,7-dioxohexahydro-2H,6H-
pyrazino[2,1-b][1,3]thiazine-3,6-diyl)bis(propane-3,1-diyl))diguanidine (13)
The diastereoisomeric mixture of linear precursor 34 (925 mg, 0.8 mmol) was dissolved in a
mixture of TFA, thioanisole and H O (90:5:5, 28 mL) and the resulting mixture was stirred at
2
room temperature for 90 min. The TFA mixture was evaporated and the crude product was
precipitated by addition of cold diethyl ether. The ether was drained off and the residue was
dried in vacuo to give the crude product as a reddish solid (422 mg). The crude product (210
mg, approx. 0.43 mmol) in dry DMF (3.5 mL) was added 1H-pyrazole-1-carboxamidine
hydrochloride (92 mg, 0.63 mmol) and DIPEA (0.11 mL, 0.63 mmol), and the mixture was
stirred under argon atmosphere for 72 h during which the reaction was monitored by RP-
HPLC. Diethyl ether (30 mL) was then added and the mixture was cooled at 4 °C and stirred
for an additional hour, resulting in the precipitation of a white solid (230 mg). The crude
product was purified by preparative HPLC to give the title compound as a fluffy white solid
2
7

1
(
11 mg, 7%). H NMR (600 MHz, CD OD) δ = 7.82 – 7.77 (m, 3H), 7.69 (s, 1H), 7.47 – 7.40
3
(m, 3H), 5.14 (dd, J = 5.4, 4.6, 1H), 4.00 (dt, J = 13.6, 7.7, 1H), 3.70 (dd, J = 13.9, 4.0, 1H),
3
.58 – 3.51 (m, 2H), 3.17 (t, J = 7.0, 2H), 3.10 – 3.98 (m, 5H), 2.78 – 2.73 (m, 1H), 2.58 (t, J
12.0, 1H), 1.91 – 1.85 (m, 1H), 1.68 – 1.56 (m, 4H), 1.46 – 1.41 (m, 1H), 1.40 – 1.35 (m,
=
1
3
3
1
4
H); C NMR (150.9 MHz, CD OD) δ = 173.8, 169.4, 158.7, 158.5, 137.3, 135.0, 133.8,
3
29.3, 128.7, 128.51, 128.49, 128.45, 127.2, 126.7, 56.6, 52.9, 50.9 (HSQC), 49.3 (HSQC),
+
3.8, 42.4, 41.6, 34.4, 30.2, 28.9, 28.6, 27.5, 26.2; HRMS (ESI): m/z [M + H] calcd for
C H N O S: 539.2911; found: 539.2906.
27
39
8
2
4
.1.20. 1,1'-(((3S,6S,9aR)-8-(2-(Naphthalen-2-yl)ethyl)-4,7-dioxohexahydro-2H,6H-
pyrazino[2,1-b][1,3]thiazine-3,6-diyl)bis(propane-3,1-diyl))diguanidine (14)
Bicycle 14 was prepared as described for 13. The crude was purified by preparative HPLC to
1
give the title compound as a fluffy white solid (16 mg, 11%). H NMR (600 MHz, CD OD) δ
3
=
7.84 – 7.80 (m, 2H), 7.78 (d, J = 7.7, 1H), 7.67 (d, 1H), 7.47 – 7.43 (m, 3H), 4.68 – 4.62
(
m, 2H), 4.08 – 4.02 (m, 1H), 3.68 (dt, J = 13.6, 5.9, 1H), 3.60 (d, J = 9.1, 2H), 3.19 – 3.14
m, 3H), 3.08 (t, J = 6.7, 2H), 3.00 – 2.95 (m, 1H), 2.94 – 2.90 (m, 1H), 2.55 (t, J = 11.4,
(
1
–
1
5
H), 2.41 – 2.36 (m, 1H), 1.85 – 1.79 (m, 1H), 1.65 – 1.57 (m, 3H), 1.53 – 1.46 (m, 1H), 1.45
1
3
1.39 (m, 1H), 1.33 – 1.25 (m, 2H); C NMR (150.9 MHz, CD OD) δ = 174.1, 170.4,
3
58.7, 158.5, 137.3, 134.9, 133.8, 129.3, 128.73, 128.70, 128.42, 128.41, 127.4, 126.8, 57.1,
3.7, 49.0 (HSQC), 48.9 (HSQC), 42.8, 42.4, 41.5, 34.6, 30.7, 29.9, 28.8, 27.7, 26.3; HRMS
+
(
ESI): m/z [M + H] calcd for C H N O S: 539.2911; found: 539.2905.
2
7
39
8
2
4
.2. Molecular modeling.
The conformational search for scaffolds A and A’ was performed using the MacroModel
2
9
application as integrated in Schrödinger Suite 2012, running on a Dell Precision 390
workstation. The OPLS_2005 force field and the GB/SA solvent model for water were used
for the energy calculations. The Truncated Newton Conjugate Gradient (TNCG) algorithm
was used for the energy minimizations, with maximum 500 iterations and a convergence
threshold of 0.012 kcal/mol×Å (0.05 kJ/mol×Å). The Monte Carlo Multiple Minimum
(MCMM) search algorithm was used for torsional sampling (1000 steps), and the energy
cutoff was set to 5 kcal/mol (21 kJ/mol) above the lowest energy conformation. Mirror image
conformations were retained, and redundant conformations were eliminated by using an
RMSD cutoff (all atoms) of 0.5 Å. For both scaffolds, this resulted in four low-energy
conformations, which were superimposed on the backbone of our previously reported 3D
2
8

pharmacophore model for cyclopentapeptide CXCR4 antagonists using five atoms: the α-
1
2
3
1
2
carbon atoms of Arg , Arg , and 2-Nal and the β-carbon atoms of Arg and Arg .
4
4
.3. Biological Studies.
.3.1. Transfections and tissue culture.
COS-7 cells were grown in Dulbecco's modified Eagle's medium (DMEM) with Glutamax
Invitrogen, UK) supplemented with 10% fetal bovine serum (FBS), 180 units/mL penicillin
and 45 μg/mL streptomycin (PenStrep) at 37 °C in a 10% CO /90% humidified atmosphere.
(
2
3
0,31
Transfection of cells was carried out by the calcium phosphate precipitation method.
Briefly, plasmid DNA (20 µg of receptor cDNA and 30 μg of the chimeric Gα to Gα signal-
i
q
converting G protein GαΔ6qi4myr) were mixed with TE-buffer (10 mM Tris-HCl, 2 mM
Na EDTA, pH 7.5) and 30 μL calcium chloride (2 M) to a total volume of 480 μL, and then
2
added to the same amount of HEPES buffered saline (280 mM NaCl, 50 mM HEPES, 1.5
mM Na HPO , pH 7.2). Precipitation was allowed for 45 min at room temperature, after
2
4
which the precipitate together with 300 μL chloroquine (2 mg/mL) in 10 mL culture media
6
was added to the 6 × 10 COS-7 cells seeded the day before. Transfection was stopped after 5
h by replacing media, and cells were incubated overnight.
4
.3.2. Functional assay.
The potency was measured using a scintillation proximity-based inositol-phosphate
3
2
5
accumulation assay (SPA-IP). In brief, one day after transfection COS-7 cells (0.35 × 10
3
cells/well) were incubated for 24 h with [ H]-myo-inositol (5 μL/mL, 2 μCi/mL) in 0.1 mL of
growth medium per well in a 96-well plate. The following day, cells were washed twice in
PBS and were incubated in 0.1 mL of Hank's balanced salt solution (Invitrogen, U.K.)
supplemented with 10 mM LiCl at 37 °C in the presence of various concentrations of ligands
for 90 min. Cells were extracted by addition of 50 µL of 10 mM formic acid to each well,
3
followed by incubation on ice for 30−60 min. The [ H]inositol-phosphates in the formic acid
cell lysates were quantified by Ysi-poly-D-Lys coated SPA beads. Briefly, 35 μL of cell
extract was mixed with 80 μL of a SPA bead suspension (12.5 μg/μL in H O) in a PicoPlate-
2
9
6 white plate. Plates were sealed, agitated for at least 30 min and centrifuged (5 min, 1500
rpm). SPA beads were allowed to settle and react with the extract for 8 h before radioactivity
was determined using a Packard Top Count NXT™ scintillation counter (PerkinElmer, MA,
USA). All determinations were made in duplicate. This readout has earlier been used
1
3,33
effectively for CXCR4 and other chemokine receptors.
2
9