Structure-activity relationships of cyclic peptide-based chemokine receptor CXCR4 antagonists: Disclosing the importance of side-chain and backbone functionalities

Article abstract of DOI:10.1021/jm0607350

Previously, we have identified a highly potent CXCR4 antagonist 2 [cyclo(-D-Tyr¹-Arg²-Arg³-Nal ⁴-Gly⁵-)] and its Arg² epimer 3 [cyclo(-D-Tyr¹-D-Arg²-Arg³-Nal ⁴-Gly⁵-)] by the screening of cyclic pentapeptide libraries that were designed based on the structure-activity relationship studies on 14-residue peptidic CXCR4 antagonist 1, In the present study, a new series of analogues of 2 and 3 were synthesized to evaluate the influences of peptide side-chain and backbone modification on bioactivities. Based on the Ala-scanning study, in which each residue in 2 and 3 was replaced with Ala having the identical chirality, substitution of Arg³ and Nal ⁴ [Nal = L-3-(2-naphthyl)alanine] with Ala (compounds 6, 7, 10, 11) led to significant loss of the potency, indicating these amino acids are more important contributors to the bioactivity. For the cyclic peptide backbone, several modifications including D/L-Ala or cyclic amino acids substitution at the Gly⁵ position and sequential N-methylation on amide nitrogens were conducted. Among the analogues, compounds 13 [cyclo(-D-Tyr ¹-Arg²-Arg³-Nal⁴-D-Ala ⁵-)] and 32 [cyclo(-D-Tyr¹-D-MeArg²-Arg ³-Nal⁴-Gly⁵-)] were close in potency to the most potent lead 2. NMR and conformational analysis indicated that both of these analogues favor the same backbone conformation as 2, whereas similar analysis of less potent analogues indicates that an altered backbone conformation is favored. The conformational analysis showed that steric repulsion by a 1,3-allylic strain-like effect across the planar peptide bond might contribute to the conformational preferences of cyclic pentapeptides.

Full text of DOI:10.1021/jm0607350

192
J. Med. Chem. 2007, 50, 192-198
Structure-Activity Relationships of Cyclic Peptide-Based Chemokine Receptor CXCR4
Antagonists: Disclosing the Importance of Side-Chain and Backbone Functionalities
Satoshi Ueda,† Shinya Oishi,† Zi-xuan Wang,‡ Takanobu Araki,† Hirokazu Tamamura,†,# Je´roˆme Cluzeau,† Hiroaki Ohno,†
Shuichi Kusano,| Hideki Nakashima,|| John O. Trent,§ Stephen C. Peiper,‡ and Nobutaka Fujii†,*
Graduate School of Pharmaceutical Sciences, Kyoto UniVersity, Sakyo-ku, Kyoto 606-8501, Japan, Department of Pathology,
Medical College of Georgia, Georgia 30912, Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental UniVersity,
Chiyoda-ku Tokyo 101-0062, Japan, St. Marianna UniVersity, School of Medicine, Miyamae-ku, Kawasaki 216-8511, Japan, and
James Graham Brown Cancer Center, UniVersity of LouisVille, Kentucky 40202
ReceiVed June 19, 2006
Previously, we have identified a highly potent CXCR4 antagonist 2 [cyclo(-D-Tyr1-Arg2-Arg3-Nal4-Gly5-)]
and its Arg2 epimer 3 [cyclo(-D-Tyr1-D-Arg2-Arg3-Nal4-Gly5-)] by the screening of cyclic pentapeptide libraries
that were designed based on the structure-activity relationship studies on 14-residue peptidic CXCR4
antagonist 1. In the present study, a new series of analogues of 2 and 3 were synthesized to evaluate the
influences of peptide side-chain and backbone modification on bioactivities. Based on the Ala-scanning
study, in which each residue in 2 and 3 was replaced with Ala having the identical chirality, substitution of
Arg3 and Nal4 [Nal ) L-3-(2-naphthyl)alanine] with Ala (compounds 6, 7, 10, 11) led to significant loss of
the potency, indicating these amino acids are more important contributors to the bioactivity. For the cyclic
peptide backbone, several modifications including D/L-Ala or cyclic amino acids substitution at the Gly5
position and sequential N-methylation on amide nitrogens were conducted. Among the analogues, compounds
13 [cyclo(-D-Tyr1-Arg2-Arg3-Nal4-D-Ala5-)] and 32 [cyclo(-D-Tyr1-D-MeArg2-Arg3-Nal4-Gly5-)] were close
in potency to the most potent lead 2. NMR and conformational analysis indicated that both of these analogues
favor the same backbone conformation as 2, whereas similar analysis of less potent analogues indicates that
an altered backbone conformation is favored. The conformational analysis showed that steric repulsion by
a 1,3-allylic strain-like effect across the planar peptide bond might contribute to the conformational preferences
of cyclic pentapeptides.
Introduction
fusion and subsequent entry of the viral genome into the target
cell.11,12 Recently, Mu¨ller et al. disclosed that CXCL12/CXCR4
interactions participate in breast cancer metastasis analogous
to programming directed migration in normal leukocytes and
progenitor cells.13 Expression of CXCR4 is enriched on the
surface of malignant primary breast cancer cells while CXCL12
is preferentially expressed in organs that are frequent sites of
metastasis in breast cancer, such as lung, liver, lymph nodes,
and bone marrow. The coordinate actions between an attractant
molecule and the corresponding receptor allow tumor cells to
spread specifically to distant organs that provide a supportive
niche. Furthermore, Nanki et al. reported that CXCL12/CXCR4
interactions might play a central role in memory T cell migration
into inflamed rheumatoid arthritis (RA) synovium and for
persisting inflammation at the affected site mediated by CD4+
T cells.14
Chemokines comprise a protein family of chemotactic factors
that bind G protein-coupled receptors.1,2 Engagement of chemo-
kine receptors by their ligands triggers changes of the receptor
conformation that lead to the initiation of a signaling cascade
involving G protein binding, protein kinase activation, Ca2+
mobilization from intracellular stores, and cytoskeletal rear-
rangement, eventually leading to directed cell migration toward
the gradients of the respective ligand.3,4 A chemokine receptor
CXCR4 and its endogenous ligand CXCL12 (stromal cell
derived factor-1, SDF-1) are partners in multiple important
functions in normal physiology involving the leukocyte chemo-
taxis in the immune system5 and progenitor cell migration during
embryologic development of the cardiovascular,6,7 hemopoietic,8
and central nervous systems.9,10 On the other hand, CXCR4 has
also multiple functions in pathologic physiology. CXCR4 serves
as a coreceptor for infection of T cell line-tropic (X4) strains
of the human immunodeficiency virus type 1 (HIV-1a). Fol-
lowing activation of the gp120 subunits of the envelope
glycoprotein by binding to CD4, CXCR4 leads to membrane
Thus, CXCR4 is considered as an important therapeutic target
for multiple diseases. Several potent CXCR4 antagonists have
been developed so far. Among them, a â-sheet-like 14-residue
peptide 1 was identified by potency optimization of a 18-residue
cyclic peptide isolated from horseshoe crabs (Figure 1).15 The
peptide 1 and its analogues were also characterized as HIV-1
entry inhibitors,16 anticancer-metastatic,17,18 and anti-RA agents.19
Several other low-molecular-weight CXCR4 antagonists such
as AMD310020,21 and KRH163622 have also been reported to
inhibit HIV-1 infection through CXCR4. Recently, we have
identified novel potent CXCR4 antagonists 2 and 3 by screening
of two orthogonal cyclic pentapeptide libraries,23 which were
designed based on the structure-activity relationship studies
on 1 (Figure 1).24 These peptides contain two arginine, one 3-(2-
naphthyl)alanine, and one tyrosine residue, that potentially
* Corresponding author. Tel: +81-75-753-4551, Fax: +81-75-753-4570,
E-mail: nfujii@pharm.kyoto-u.ac.jp
† Kyoto University.
‡ Medical College of Georgia.
# Tokyo Medical and Dental University.
| St. Marianna University.
§ University of Louisville.
a Abbreviations: Nal, L-3-(2-naphthyl)alanine; Pic, pipecolic acid; Sar,
sarcosine; HIV-1, human immunodeficiency virus type 1; RA, rheumatoid
arthritis; SA-MD, simulated annealing molecular dynamics; AMD3100, 1,1′-
[1,4-phenylenebis(methylene)]-bis(1,4,8,11-tetraazacyclotetradecane);KRH1636,
N-{(S)-4-guanidino-1-[(S)-1-naphthalen-1-yl-ethylcarbamoyl]butyl}-4-{[(py-
ridin-2-yl-methyl)amino]methyl}benzamide.
10.1021/jm0607350 CCC: $37.00 © 2007 American Chemical Society
Published on Web 12/29/2006

Cyclic Peptide-Based Chemokine Receptor CXCR4 Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 193
Scheme 1a
a Reagents: (i) o-nitrobenzenesulfonyl chloride, 2,4,6-collidine; (ii) MeOH, PPh3, DEAD; (iii) DBU, 2-mercaptoethanol; (iv) Fmoc-AA-OH, HATU,
HOAT, DIPEA; (v) piperidine.
Table 1. Biological Activities of 2, 3, and the Ala-Substituted
Derivatives
peptide
sequencea
IC50 (µM)b EC50 (µM)c
0.004 0.16
>1 115
0.063
>1
>1
0.008
2
4
5
6
7
3
8
9
10
11
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Ala-L-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-L-Ala-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-L-Arg-L-Ala-L-Nal-Gly-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Ala-Gly-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Ala-D-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-D-Ala-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-D-Arg-L-Ala-L-Nal-Gly-)
12
>120
>120
0.39
0.13
0.23
29
16
Figure 1. Structures of 1 and its downsized peptides 2 and 3. Bold
residues are the indispensable residues of 1 for the potent CXCR4-
antagonistic activity. Nal ) L-3-(2-naphthyl)alanine, Cit ) L-citrulline.
>1
60
>120
cyclo(-D-Tyr-D-Arg-L-Arg-L-Ala-Gly-) >1
a The substituted residues from the parent peptides 2 and 3 are designated
by underlining. b IC50 values for the cyclic pentapeptides are based on
inhibition of [125I]SDF-1 binding to CXCR4 transfectants of CHO cells.
c EC50 values are based on the inhibition of HIV-induced cytopathogenicity
in MT-4 cells. All data are the mean values for at least three independent
experiments.
correspond to the pharmacophore residues of the parent peptide
1. We25-27 and others28 have performed several modifications
on 2 including incorporation of (E)-alkene or reduced-amide
dipeptide isosteres and conformationally constrained amino acid
analogues, and fine-tuning of backbone ring structures. How-
ever, systematic modifications of 2 to design more potent
antagonists have not been reported so far. In order to clarify
the elements among the peptide side-chain and backbone
functional groups indispensable for the ligand binding to
CXCR4, chemical derivatization of 2 and 3 was conducted. In
this manuscript, we describe the details of structure-activity
relationship studies on cyclic pentapeptide-based CXCR4
antagonists 2 and 3 as well as identification of a more potent
CXCR4 antagonist.
32
(HFIP) in CH2Cl2 provided linear protected peptides, which
were cyclized with diphenylphosphoryl azide (DPPA) in DMF.
Final deprotection with TFA-H2O (95:5) followed by reverse-
phase HPLC purification afforded the cyclic peptides. All
peptides were identified with ion-spray mass spectrometry, and
the purity was more than 95% by analytical HPLC.
Results and Discussion
Identification of Indispensable Pharmacophore Function-
ality by Alanine-Scanning. Our first attempt was to identify
the minimal side-chain functional group requirement of 2 and
3 for CXCR4 antagonism. To evaluate the comparative signifi-
cance of side-chain functionality, each residue except for the
Gly residue of 2 and 3 was substituted with Ala. Because the
chirality of Ala was identical to that of the corresponding residue
in parent peptides, exhibition of similar conformations was
expected even after the substitution. All Ala-substituted peptides
4-11 showed significantly less CXCR4 antagonistic and anti-
HIV activities when compared to the parent peptides 2 and 3.
Ala3- or Ala4-substituted analogues 6, 7, 10, and 11 did not
show any CXCR4 antagonistic activity up to 1 µM (Table 1).
This indicates that all the side-chain functional groups are
important for the high CXCR4 antagonistic activity. On the other
hand, L/D-Ala2-substituted analogues, 5 and 9, and D-Ala1-
substituted analogue 8 maintained moderate activities (5: IC50
) 63 nM, EC50 ) 12 µM; 8: IC50 ) 130 nM, EC50 ) 29 µM;
9: IC50 ) 230 nM, EC50 ) 16 µM), although the potencies
were less than one-twentieth of the parent peptides. These data
suggest that the phenol group of D-Tyr1 and a guanidino group
of Arg2 do not play a critical role in receptor binding, while
guanidino group of Arg3 and naphthalene group of Nal4 are
indispensable for the ligand interaction with CXCR4. This
Chemistry. Synthesis on solid support of all peptides was
performed on 2-chlorotrityl [(2-Cl)Trt] resin in parallel using
usual Fmoc-based solid-phase peptide synthesis as described
in the Experimental Section. t-Bu and 2,2,4,6,7-pentameth-
yldihydrobenzofuran-5-sulfonyl (Pbf) groups were employed for
Tyr and Arg residue side-chain protection, respectively. L-Ala,
D-Ala, L-Pro, D-Pro, L-pipecolic acid (L-Pic), D-Pic, â-alanine
(â-Ala), and L-Nal were employed for the C-terminal residues
of 12/19, 13/20, 14/21, 15/22, 16/23, 17/24, 18/25, and 26/31,
respectively. The Gly residue was positioned at the C-terminal
of the other peptide resins to avoid potential epimerization
during the cyclization. For the preparation of N-methyl amino
acid-containing peptide resins, an N-methyl group was incor-
porated on the R-amino group by a site-selective method
reported by Miller et al.29,30 (Scheme 1). The R-amino group
was temporarily protected with an o-nitrobenzenesulfonyl (o-
Ns) group before N-methylation by the Mitsunobu reaction.
After removal of the o-Ns group by treatment with 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) and 2-mercaptoethanol,
the subsequent amino acids were coupled using O-(7-azaben-
zotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
(HATU)31 as an activating reagent. Treatment of protected
peptide resins with 20% (v/v) 1,1,1,3,3,3-hexafluoroisopropanol

194 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Ueda et al.
Figure 2. Superimposition of low-energy structures of 2 (purple) and 12 (green, left) or 13 (green, right).
Table 2. Biological Activities of the Gly5-Modified Analogues of 2 and
3
propensity is consistent with the previous structure-activity
relationship studies on 1, where Ala-substitutions at the Arg2-
Nal3 motif in 1 were more sensitive to anti-HIV activity as
compared to the substitution of the other important residues,
such as Tyr5 and Arg14.24 It is noteworthy that two L/D-Ala2-
substituted peptides 5 and 9 having opposite chiralities retained
moderate bioactivities. Recently, we have shown that L-Arg2
in 2 could be replaced by nonbasic amino acids such as L-Phe-
(4-F) (L-4-fluorophenylalanine) and D-MeAla without significant
loss of CXCR4 antagonistic activity.26,27 Hence, the functional
group or the spatial disposition of the Arg2 guanidino group
could be further optimized.
IC50
EC50
peptide
sequencea
(µM)b
(µM)c
2
12
13
14
15
16
17
18
3
19
20
21
22
23
24
25
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-L-Ala-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-D-Ala-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-L-Pro-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-D-Pro-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-L-Pic-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-D-Pic-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-â-Ala-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-L-Ala-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-D-Ala-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-L-Pro-)
0.004
0.17
0.011
>1
0.16
20
0.49
d
-
-
-
-
d
>1
d
>1
>1
0.047
0.008
0.092 10
0.011
> 1
d
3.0
0.37
0.67
Conformational Restriction of Cyclic Peptides by Modi-
fication of the Glycine Residue. In contrast to the finding that
all side-chain functional groups of 2 and 3 were important for
the high CXCR4 antagonistic activities, the Gly5 position had
many possibilities for further optimization. Since the absence
of a side chain in Gly5 possibly affected the conformational
flexibility of the peptide backbone, it was expected that use of
chiral or conformationally constrained amino acids would
potentially restrict the global conformations to decrease the
entropy losses upon the peptide binding on CXCR4. In order
to evaluate the structure-activity relationship at the Gly5
position, simple aliphatic amino acids were utilized for this
study. In our previous conformational studies on 2, characteristic
orientations of the amide carbonyl groups of D-Tyr,1 Arg,2 Arg3,
and Gly5 were observed in the low energy state.23 These
carbonyl oxygens were oriented away from the side chains of
the respective following amino acids. This could be attributed
to steric repulsion by a 1,3-allylic strain-like effect across the
planar peptide bond.33 On the other hand, the flexible Nal4 ψ
and Gly5 φ angles had been expected, since there were two
possibilities of Nal4 carbonyl oxygen orientation. However, the
carbonyl oxygen was directed away from the Gly5 pro-R
hydrogen atom in the calculated conformations, while the other
orientation was not exhibited. This implied that the incorporation
of a side chain having R-chirality (D-amino acid) at the Gly5
position could restrict the rotation of the Nal4-Gly5 peptide bond
plane. On the basis of this hypothesis, we introduced both
enantiomers of Ala, Pro, and Pic to the Gly5 position for
conformational restriction. In addition, â-Ala was utilized for
the optimization of carbon chain length at the Gly5 position.
d
-
-
-
-
d
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-D-Pro-) > 1
d
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-L-Pic-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-D-Pic-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-â-Ala-)
0.64
> 1
0.35
d
38
a The substituted residues from the parent peptide 2 and 3 are underlined.
b IC50 values for the cyclic pentapeptides are based on inhibition of
[
125I]SDF-1 binding to CXCR4 transfectants of CHO cells. c EC50 values
are based on the inhibition of HIV-induced cytopathogenicity in MT-4 cells.
d Not tested. All data are the mean values for at least three independent
experiments.
observed for D-Ala5-substituted analogues 13 and 20 (13: IC50
) 11 nM; EC50 ) 0.49 µM, 20: IC50 ) 11 nM; EC50 ) 0.67
µM) as compared to the corresponding L-Ala5-substituted
peptides 12 and 19, respectively. The bioactivities of D-Ala5-
substituted analogues were approximately half of the parent
peptides 2 and 3. This suggested that substitution with L-Ala5
resulted in significant conformational change, while D-Ala5
substitution kept the bioactive conformations of the parent
peptides 2 and 3. Simulated annealing molecular dynamics (SA-
MD) analysis demonstrated that the backbone conformation of
13 was similar to that of 2 but different from that of 12 (Figure
2). Local conformations around Nal4 and Gly5/D-Ala5 were very
similar between 2 and 13 as expected, while the conformation
of L-Ala5-substituted analogues 12 differed particularly in the
opposite orientation of the Nal4 carbonyl oxygen. These
calculated structures are consistent with the observed NOE data;
in L-Ala5-substituted peptide 13, strong NOE between Nal4 HR
and D-Ala5 HN indicates that these hydrogen atoms were oriented
into the same direction. On the other hand, the observed weak
NOE between Nal4 HR and L-Ala5 HN in peptide 12 indicates
these hydrogen atoms were oriented into the opposite directions.
The 1,3-pseudo allylic strain between the Nal4 carbonyl oxygen
and the R-methyl group of D/L-Ala5 could result in these
different conformational preferences between 12 and 13. This
L-Ala5-substituted analogues 12 and 19 showed less than 10-
fold lower CXCR4 antagonistic and anti-HIV activities of the
parent peptides 2 and 3, respectively (Table 2, 12: IC50 ) 170
nM; EC50 ) 20 µM, 19: IC50 ) 92 nM; EC50 ) 10 µM). By
contrast and as expected, more than 8-fold higher potencies were

Cyclic Peptide-Based Chemokine Receptor CXCR4 Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 195
Table 3. Biological Activities of N-Methyl Amino Acid-Containing
Analogues of 2 and 3
Peptides 26, 28, 29, and 30 derived from N-methyl amino acid-
scanning of peptide 2 showed more than 25-fold less CXCR4
antagonistic activity as compared to the parent peptide 2.
N-Methylated analogues 31, 34, and 35 also showed a significant
decrease in CXCR4 antagonism as compared to the parent
peptide 3. Use of L-MeArg3 in peptide 33 slightly decreased
the activity. N-Methylation at the Nal4 position caused a
remarkable decrease in CXCR4 antagonistic activity (29: IC50
) 250 nM; 34: IC50 ) 563 nM), suggesting that N-methylation
at the putative receptor-binding motif (Arg3-Nal4) is unfavorable
probably due to the absence of an amide proton. Previously,
we showed that replacement of the Arg3-Nal4 motif with the
corresponding (E)-alkene dipeptide isostere unit (L-Arg3-ψ[(E)-
CH)CH]-L-Nal4) or a reduced amide isostere unit (L-Arg3-ψ-
[CH2-NH]-L-Nal4) caused a significant loss of CXCR4 an-
tagonistic activity.25 These observations also indicate the
importance of the Arg3-Nal4 amide bond in both functional and
conformational aspects.
IC50
EC50
peptide
sequencea
(µM)b
(µM)c
2
26
27
28
29
30
3
31
32
33
34
35
cyclo(-D-Tyr-L-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-MeTyr-L-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-L-MeArg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-L-Arg-L-MeArg-L-Nal-Gly-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-MeNal-Gly-)
cyclo(-D-Tyr-L-Arg-L-Arg-L-Ala-Sar-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-MeTyr-D-Arg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-D-MeArg-L-Arg-L-Nal-Gly-)
cyclo(-D-Tyr-D-Arg-L-MeArg-L-Nal-Gly-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-MeNal-Gly-)
cyclo(-D-Tyr-D-Arg-L-Arg-L-Nal-Sar-)
0.004
0.128
0.023
0.099
0.250
0.167
0.008
0.157
0.003
0.021
0.563
0.256
0.16
d
-
1.399
9.534
d
-
-
d
0.39
d
-
0.088
0.782
d
-
-
d
a N-Methylated residues are underlined. b IC50 values for the cyclic
pentapeptides are based on inhibition of [125I]SDF-1 binding to CXCR4
transfectants of CHO cells. c EC50 values are based on the inhibition of
HIV-induced cytopathogenicity in MT-4 cells. d Not tested. All data are
the mean values for at least three independent experiments.
On the other hand, the L/D-MeArg2-substituded peptides 27
and 32 showed potent CXCR4 antagonistic and anti-HIV
activities (27: IC50 ) 23 nM, EC50 ) 1.4 µM; 32: IC50 ) 3.0
nM, EC50 ) 0.088 µM), indicating that N-methylation at Arg2
is not critical to antagonist activity. Interestingly, the D-MeArg2
substitution (32) led to approximately 2-fold increase of CXCR4
antagonistic activity and more than 40-fold anti-HIV activity
as compared with 3, and these activities were nearly equal to
those of 2. NMR and SA-MD calculations showed that major
conformer of 32 exhibited a backbone conformation similar to
2 but different from the parent peptide 3, particularly with
respect to the orientation of D-Tyr1 carbonyl oxygen (Figure
3).38 This conformation was also supported by strong NOEs
between [D-Arg2 HR and D-Arg2 HNMe] and [D-Arg2 HNMe and
Arg3 HN] in 32 (the corresponding strong NOEs were not
observed in the parent peptide 3; see Supporting Information).
It is possible that an unfavorable 1,3-pseudo-allylic strain-like
effect between the N-methyl group of D-MeArg2 and D-Tyr1
side chain of 32 induced the flip of D-Tyr1-D-Arg2 amide group
upon N-methylation of D-Arg2, resulting in the identical local
conformation around the D-Tyr1-D-MeArg2 dipeptide with the
D-Tyr1-L-Arg2 conformation of 2. Previously, we have shown
a similar local conformational change upon N-methylation at
the same position in D-Ala2-substituted analogues of 2; i.e. cyclo-
(-D-Tyr1-D-MeAla2-Arg3-Nal4-Gly5-) showed 5-fold higher CX-
CR4 antagonistic activity (IC50 ) 42 nM) than the nonmethyl-
ated peptide, cyclo(-D-Tyr1-D-Ala2-Arg3-Nal4-Gly5-) 9.27 It is
also noted that the lower bioactivities of peptide 27 compared
to 2 could be explained by the potential flip of the D-Tyr1-L-
MeArg2 amide bond orientation by N-methylation. These data
suggest that the amide proton of Arg2 has little contribution to
bioactivity, and the amide bond orientation between D-Tyr1 and
D-MeArg2 in 32 may contribute to its enhanced biological
function.
could explain the reason for the higher potencies of D-Ala5-
substituted analogues. The above information could serve for
the further optimization of the Gly5 position using D-amino acids
having side-chain functionality.
Our next approach was to restrict global conformations of
peptides 2 and 3 using cyclic amino acids such as L/D-Pro and
L/D-Pic. These amino acids can provide a fused ring structure
of cyclic peptides consisting of small and large rings. It was
expected that the limited φ angle flexibility of Pro and Pic could
contribute to the global conformational restriction.34 Covalent
linkage between the amide nitrogen and the side chain could
also produce a favorable orientation of the Nal4 carbonyl oxygen.
However, L/D-Pro5- and L/D-Pic5-substituted peptides 14-17,
21, 22, and 24 did not show CXCR4 inhibitory activities with
IC50 values lower than 1 µM. Even the most potent peptide 23
exhibited an IC50 of only 0.64 µM. This suggests that the
presence of cyclic amino acids at this position is sterically or
conformationally unfavorable for the peptide-CXCR4 interac-
tion. We could also assume that an amide proton at this position
is required for high activity. This was supported by the fact
that peptides having a sarcosin (Sar) at the Gly5 position
possessed less than one-thirtieth CXCR4 antagonistic activity
of the parent peptides (see the next section). â-Ala5-substituted
analogues 18 and 25 showed lower CXCR4 antagonistic and
anti-HIV activities (18: IC50 ) 47 nM, EC50 ) 3 µM; 25: IC50
) 350 nM, EC50 ) 38 µM), indicating that expansion of the
backbone ring size (16-membered ring) at this position is not
favorable. Recently, we showed that reduction in the size of
the backbone ring using γ-Nal [4-amino-5-(2-naphthyl)pentanoic
acid] or γ-(E)-Nal [(E)-4-amino-5-(2-naphthyl)pent-2-enoic
acid] unit (14-membered ring) instead of Nal4-Gly5 dipeptide
resulted in moderate to significant loss of CXCR4 antagonistic
activity.27 These observations suggest the importance of Gly5
as a spacer for appropriate spatial orientation of the CXCR4
antagonist pharmacophores.
Conclusion
Identification of a Novel Potent CXCR4 Antagonist
through N-Methyl Amino Acid-Scanning of 2 and 3. N-
Methylation of the peptide backbone has been shown to be a
valuable method in structure-activity relationship studies on
bioactive peptides.35-37 Substitutions with N-methyl amino acids
often cause an increase or decrease in potency and selectivity
of peptide ligands, providing useful information on the bioactive
conformation. Hence, every amide bond of 2 and 3 was replaced
sequentially with the corresponding N-methylated amide, and
the bioactivities of obtained peptides were evaluated (Table 3).
Our present Ala-scanning study has shown that all of the side-
chain functional groups contribute to high CXCR4 antagonistic
activity of peptides 2 and 3. In particular, Arg3 and Nal4 were
proven to be indispensable for CXCR4 antagonistic activity.
We have also shown that L-Ala substitution for Gly5 of 2 or 3
caused a remarkable decrease in CXCR4 antagonistic and anti-
HIV activities, while D-Ala substitution retained activity.
Conformational studies revealed that D-Ala-substituted analogue
13 adopted a backbone conformation similar to that of 2, which
allows the rationalization of the biological activity for these

196 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Ueda et al.
Figure 3. Superimposition of low-energy structures of 32 (green) and 2 (purple, left) or 3 (purple, right).
L-Pic-(2-Cl)Trt-resin for 16 and 23, Fmoc-D-Pic-(2-Cl)Trt-resin for
17 and 24, Fmoc-â-Ala-(2-Cl)Trt-resin for 18 and 25, Fmoc-L-Nal-
(2-Cl)Trt-resin for 26 and 31, and H-Gly-Trt(2-Cl)Trt-resin (0.75
mmol/g) for the other peptides. Fmoc-amino acids were coupled
using 1,3-diisopropylcarbodiimide (DIPCDI, 0.078 mL, 0.50 mmol)
and N-hydroxybenzotriazole hydrate (HOBt‚H2O, 76 mg, 0.50
mmol) in DMF (1.0 mL) for 1.5 h. For the coupling of Fmoc-
amino acid to the N-methyl amino acid, HATU (186 mg, 0.49
mmol) and 1-hydroxy-7-azabenzotriazole (HOAt, 68 mg, 0.50
mmol) were employed in place of DIPCDI/HOBt. The Fmoc group
was deprotected by treatment with 20% (v/v) piperidine-DMF for
20 min.
N-Methyl Modification of N-Terminal R-Amino Group on
Resin. Resin (0.10 mmol) was treated with o-nitrobenzenesulfonyl
chloride (66.5 mg, 0.30 mmol) and 2,4,6-collidine (0.066 mL, 0.50
mmol) in CH2Cl2 (1.0 mL) for 2 h at room temperature. After the
resin was washed (CH2Cl2 × 3, DMF × 3, and THF × 3), to a
suspension of the N-Ns-protected resin in anhydrous THF (1.0 mL)
were added MeOH (0.020 mL, 0.50 mmol), PPh3 (131 mg, 0.50
mmol), and diethyl diazodicarboxylate (0.227 mL, 0.50 mmol) at
0 °C. The mixture was shaken for 2 h at room temperature, followed
by washing the resin (THF × 3 and CHCl3 × 3). The N-methylated
resin was treated with DBU (0.075 mL, 0.50 mmol) and 2-mer-
captoethanol (0.070 mL, 1.0 mmol) for 1.5 h at room temperature
to give the protected peptide resin having an N-methyl amino acid
at the N-terminus.
Cleavage of Protected Peptides from the Resin and Cycliza-
tion. Protected peptide resin was treated with 20% (v/v) HFIP-
CH2Cl2 (10 mL) for 2 h. After filtration of the resin, the filtrate
was concentrated to provide the crude linear protected peptide. To
the solution of the residue in DMF (30 mL) were added DPPA
(0.539 mL, 0.25 mmol) and NaHCO3 (42.0 mg, 0.50 mmol) at -40
°C. After being stirred for 36 h at room temperature, the whole
was filtered, and the filtrate was concentrated to give the protected
cyclic peptide, which was subjected to solid-phase extraction (SPE)
over basic alumina in CHCl3-MeOH (9:1) to remove inorganic
salts derived from DPPA.
Deprotection of Protected Cyclic Peptide and HPLC purifi-
cation. Protected cyclic peptides were treated with 95% (v/v) TFA
solution (10 mL) for 2 h at room temperature. Concentration under
reduced pressure and purification by preparative HPLC gave cyclic
peptides.
series of analogues. In addition, through comprehensive N-
methyl-scanning of all residues in 2 and 3, the N-methylated
analogue 32 was characterized as one of the most potent cyclic
pentapeptide-based CXCR4 antagonists synthesized thus far. The
slight increase in CXCR4 antagonistic activity in 32 as compared
to its nonmethylated analogue 3 could be explained by the
favorable peptide bond orientation at the N-methylation site.
Conformational studies suggested that the high potency in these
series of compounds is due to the orientation of the backbone
amide bonds, although direct interaction of the amide functions
with the CXCR4 receptor is not clear. These results give
valuable insight for understanding the ligand-receptor interac-
tions and may also provide useful approaches for the design of
new low-molecular-weight CXCR4 antagonists.
Experimental Section
General. Exact mass (HRMS) spectra were recorded on a JEOL
JMS-01SG-2 or JMS-HX/HX 110A mass spectrometer. The ion-
spray mass spectrum was obtained with a Sciex APIIIIE triple
quadrupole mass spectrometer (Toronto, Canada). Optical rotations
were measured in water or 50% (v/v) water/AcOH solution with a
Horiba high-sensitive polarimeter SEPA-200. 1H NMR spectra were
recorded using a Bruker AM 600 or JEOL JNM-ECA600 spec-
trometer at 600 MHz frequency, or JEOL JNM-AL400 spectrometer
at 400 MHz frequency. Chemical shifts are calibrated to the solvent
signal (2.49 ppm for DMSO, or 4.65 ppm for H2O; s ) singlet, d
) doublet, dd ) double doublet, m ) multiplet). For HPLC
separations, a Cosmosil 5C18-ARII analytical column (Nacalai
Tesque, 4.6 × 250 mm, flow rate 1 mL/min) or a Cosmosil 5C18-
ARII preparative column (Nacalai Tesque, 20 × 250 mm, flow
rate 11 mL/min) was employed, and eluting products were detected
by UV at 220 nm. A solvent system consisting of 0.1% TFA in
water (v/v, solvent A) and 0.1% TFA in MeCN (v/v, solvent B)
was used for HPLC elution.
Preparation of Amino Acid-Loaded 2-Chlorotrityl Resin.
2-Chlorotrityl chloride resin (1.25 mmol/g, 0.63 mmol) was treated
with Fmoc-amino acid (0.69 mmol) and N,N-diisopropylethylamine
(DIPEA) (2.77 mmol) in CH2Cl2 (5.00 mL) for 1.5 h. After the
resin was washed with CH2Cl2, it was dried in vacuo. The loading
was determined by measuring at 290 nm UV absorption of the
piperidine-treated sample: Fmoc-L-Nal-(2-Cl)Trt-resin (0.72 mmol/
g); Fmoc-L-Pro-(2-Cl)Trt-resin (0.78 mmol/g); Fmoc-D-Pro-(2-Cl)-
Trt-resin (0.81 mmol/g); Fmoc-L-Pic-(2-Cl)Trt-resin (0.88 mmol/
g); Fmoc-D-Pic-(2-Cl)Trt-resin (0.84 mmol/g); Fmoc-â-Ala-(2-
Cl)Trt-resin (0.79 mmol/g).
Cell Culture. Human T-cell lines, MT-4 and MOLT-4 cells were
grown in RPMI 1640 medium containing 10% heat-inactivated fetal
calf serum, 100 IU/mL penicillin, and 100 µg/ mL streptomycin.
Virus. A strain of X4-HIV-1, HIV-1IIIB, was used for the anti-
HIV assay. This virus was obtained from the culture supernatant
of HIV-1 persistently infected MOLT-4/HIVIIIB cells and stored
at -80 °C until used.
Fmoc-Based Solid-Phase Peptide Synthesis. Protected peptide
resins were manually constructed on a 0.10 mmol scale on H-L-
Ala-(2-Cl)Trt-resin (0.89 mmol/g) for 12 and 19, H-D-Ala-(2-Cl)-
Trt-resin (0.90 mmol/g) for 13 and 20, Fmoc-L-Pro-(2-Cl)Trt-resin
for 14 and 21, Fmoc-D-Pro-(2-Cl)Trt-resin for 15 and 22, Fmoc-
Anti-HIV-1 Assay. Anti-HIV-1 activity was determined based
on the protection against HIV-1-induced cytopathogenicity in MT-4

Cyclic Peptide-Based Chemokine Receptor CXCR4 Antagonists
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2 197
cells. Various concentrations of test compounds were added to
HIV-1 infected MT-4 cells at multiplicity of infection (MOI) of
0.01 and placed in wells of a flat-bottomed microtiter tray (1.5 ×
104 cells/ well). After 5 days incubation at 37 °C in a CO2 incubator,
the number of viable cells was determined using the 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT).
structures. The root-mean-square deviation (rmsd) values for all
backbone structures of ten low-energy structures were below 0.23
Å.
Acknowledgment. This work was supported by Grant-in-
Aid for Scientific Research from the Ministry of Education,
Culture, Sports, Science, and Technology of Japan, and Health
and Labour Sciences Research Grants (Research on HIV/AIDS)
and Philip Morris USA Inc. and Philip Morris International.
Computation time was provided by the Supercomputer Labora-
tory, Institute for Chemical Research, Kyoto University. S.U.
and S.O. are grateful for the JSPS Research Fellowships for
Young Scientists.
[
125I]-SDF-1 Binding and Displacement. Stable CHO cell
transfectants expressing CXCR4 variant were prepared as describe
previously.39 CHO transfectants were harvested by treatment with
trypsin-EDTA, allowed to recover in complete growth medium
(MEM-R, 100 µg/ mL penicillin, 100 µg/mL streptomycin, 0.25
µg/mL amphotericin B, 10% (v/v)) for 4-5 h, and then washed in
cold binding buffer (PBS containing 2 mg/mL BSA). For ligand
binding, the cells were resuspended in binding buffer at 1 × 107
cell/mL, and 100 µL aliquots were incubated with 0.1 nM of [125I]-
SDF-1 (Perkin-Elmer Life Sciences) for 2 h on ice under constant
agitation. Free and bound radioactivities were separated by
centrifugation of the cells through an oil cushion, and bound
radioactivity was measured with gamma-counter (Cobra, Packard,
Downers Grove, IL). Inhibitory activity of test compounds was
determined based on the inhibition of [125I]-SDF-1 binding to
CXCR4 transfectants (IC50).
Supporting Information Available: Characterization data for
1
all new compounds, H NMR data of 2, 3, 12, 13, and 32, and
HPLC charts of representative compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Murphy, P. M.; Baggiolini, M.; Charo I. F.; Herbert, C. A.; Horuk,
R.; Matsushima, K.; Miller, L. H.; Oppenheim, J. J.; Power, C. A.
International union of pharmacology. XXII. Nomenclature for
chemokine receptors. Pharmacol. ReV. 2000, 52, 145-176.
(2) Mackay, C. R.; Chemokines: immunology’s high impact factors. Nat.
Immunol. 2001, 2, 95-101.
(3) Premack, B. A.; Schall, T. J. Chemokine receptors: gateways to
inflammation and infection. Nat. Med. 1996, 2, 1174-1178.
(4) Zlotnik, K.; Yoshie, O. Chemokines: a new classification system
and their role in immunity. Immunity 2000, 12, 121-127.
(5) Bleul, C. C.; Fuhlbrigge, R. C.; Casanovas, J. M.; Aiuti, A.; Springer,
T. A. A highly efficacious lymphocyte chemoattractant, stromal cell-
derived factor 1 (SDF-1). J. Exp. Med. 1996, 2, 1101-1109.
(6) Tachibana, K.; Hirota, S.; Iizasa, H.; Yoshida, H.; Kawabata, K.;
Kataoka, Y.; Kitamura, Y.; Matsushima, K.; Yoshida, N.; Nishikawa,
S.; Kishimoto, T.; Nagasawa, T. The chemokine receptor CXCR4 is
essential for vascularization of the gastrointestinal tract. Nature 1998,
393, 591-594.
(7) Nagasawa, T.; Hirota, S.; Tachibana, K.; Takakura, N.; Nishikawa,
S.; Kitamura, Y.; Yoshida, N.; Kikutani, H.; Kishimoto, T. Defects
of B-cell limphopoiesis and bone-marrow myelopoiesis in mice
lacking the CXC chemokine PBSF/SDF-1. Nature 1996, 382, 635-
638.
(8) Aiuti, A.; Webb, I. J.; Bleul, C.; Springer, T.; Gutierrez-Ramos, J.
C. The chemokine SDF-1 is a chemoattractant for human CD34+
hematopoietic progenitor cells and provides a new mechanism to
explain the mobilization of CD34+ progenitors to peripheral blood.
J. Exp. Med. 1997, 2, 111-120.
(9) Zhu, Y.; Yu, Y.; Zhang, X. C.; Nagasawa, T.; Wu, J. Y.; Rao, Y.
Role of the chemokine SDF-1 as the meningeal attractant for
embryonic cerebellar neurons. Nat. Neurosci. 2002, 5, 719-720.
(10) Stumm, R. K.; Zhou, C.; Ara, T.; Lazarini, F.; Dubois-Dalcq, M.;
Nagasawa, T.; Hollt, V.; Schulz, S. CXCR4 regulates interneuron
migration in the developing neocortex. J. Neurosci. 2003, 23, 5123-
5130.
(11) Feng, Y.; Broder, C. C.; Kennedy, P. E.; Berger, E. A. HIV-1 entry
co-factor: Functional cDNA cloning of a seven-transmembrane G
protein-coupled receptor. Science 1996, 272, 872-877.
(12) Oberlin, E.; Amara, A.; Bachelerie, F.; Bessia, C.; Virelizier, J. L.;
Arenzana-Seisdedos, F.; Schwartz, O.; Heard, J. M.; Clark-Lewis,
I.; Legler, D. L.; 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.
(13) Mu¨ller, A.; Homey, B.; Soto, H.; Ge, N.; Catron, D.; Buchanan, M.
E.; Mcclanahan, T.; Murphy, E.; Yuan, W.; Wagner, S. M.; Barrera,
J. L.; Mohar, A.; Vera´stegui, E.; Zlotnik, A. Involvement of
chemokine receptors in breast cancer metastasis. Nature 2001, 410,
50-56.
(14) Nanki, T.; Hayashida, K.; EI-Gabalawy, H. S.; Suson, S.; Shi, K.;
Girschick, H. J., Yavuz, S.; Lipsky, P. E. Stromal cell-derived factor-
1-CXC chemokine receptor interactions play a central role in CD4+
T cell accumulation in rheumatoid arthritis synovium. J. Immunol.
2000, 165, 6590-6598.
NMR Spectroscopy. The peptide sample was dissolved in
1
DMSO-d6 at a concentration of 5 mM. H NMR spectra of the
peptides were recorded at 300 K. The assignment of the proton
1
3
resonance was achieved by use of H-1H COSY spectra. J(HN,
HR) coupling constants were measured from one-dimensional
spectra. The mixing time for NOESY experiments was set at 200
ms. NOESY spectra were composed of 512 real points in the F2
dimension and 256 real points, which were zero-filled to 256 points
in the F1 dimension, with 144 scans per t1 increment. The cross-
peak intensities were evaluated by relative build-up rates of the
1
cross-peaks. 2, 3, 12, and 13 exhibited one set of signals in H
NMR spectra. On the other hand, two distinct sets of signals were
1
observed in H NMR spectra of 32 with relative populations of
69% and 31%, indicating the existence of two conformations. For
the minor conformer, NOESY spectra showed the NOE contact
between R protons of D-Tyr1 and D-MeArg2 which is characteristic
of the amide bond in a cis conformation. The major conformer did
not exhibit any sequential HR-HR NOEs, suggesting this conformer
adopt an all-trans conformation. The calculated structures of major
and minor conformers exhibited trans- and cis-D-Tyr1-D-MeArg2
amide bond, respectively, which was consistent with the observed
NOEs.
Calculation of Structures. The structure calculations were
performed on a Silicon Graphics Origin 2000 workstation with the
NMR refine program within the Insight II/Discover package using
the consistent valence force field (CVFF). Pseudoatoms were
defined for the CH3 protons of L-Ala5 of 12, D-Ala5 of 13, and
R
N-methyl protons of 32, and for all methylene protons of Nal4,
D-Tyr1, D/L-Arg2, and Arg3, prochirality of which were not identified
from 1H NMR data. The dihedral φ angle constraints were calculated
based on the Karplus equation: 3J(HN, HR) ) 6.7 cos2(θ - 60°) -
1.3 cos(θ - 60°) + 1.5. Lower and upper angle errors were set to
15°. The NOESY spectrum with a mixing time of 200 ms was used
for the estimation of the distances restraints between protons. The
NOE intensities were classified into three categories (strong,
medium, and weak) based on the number of contour lines in the
cross-peaks to define the upper-limit distance restraints (1.7, 3.5,
and 5.0 Å, respectively). The upper-limit restraints were increased
by 1.0 Å for the involved pseudoatoms. Lower bounds between
nonbonded atoms were set to their van der Waals radii (1.8 Å).
These distance and dihedral angle restraints were included with
force constants of 25-100 kcal/mol‚Å2 and 25-100 kcal/mol‚rad2,
respectively. The 50 initial structures generated by the NMR refine
program randomly were subjected to the simulated annealing
calculations. The final minimization stage was achieved until the
maximum derivative became less than 0.01 kcal/mol‚Å2 by the
steepest descents and conjugate gradients methods. Excellent
convergence was seen in the backbone structure of all calculated
(15) 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,

198 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 2
Ueda et al.
N. A novel anti-HIV synthetic peptide, T-22 ([Tyr5,12,Lys7]-polyphe-
musin II). Biochem. Biophys. Res. Commun. 1992, 189, 845-850.
(16) Tamamura, H.; Xu, Y.; Hattori, T.; Zhang, X.; Arakaki, R.; Kanbara,
K.; Omagari, A.; Otaka, A.; Ibuka, T.; Yamamoto, N.; Nakashima,
H.; Fujii, N. A low molecular weight inhibitor against the chemokine
redeptor CXCR4: a strong anti-HIV peptide T140. Biochem. Biophys.
Res. Commun. 1998, 253, 877-882.
(17) Tamamura, H.; Hori, A.; Kanzaki, N.; Hiramatsu, K.; Mizumoto,
M.; Nakashima, H.; Yamamoto, N.; Otaka, A.; Fujii, N. T140
analogues as CXCR4 antagonists identified as anti-metastatic agent
in the treatment of breast cancer. FEBS Lett. 2003, 550, 79-83.
(18) Takenaga, M.; Tamamura, H.; Hiramatsu, K.; Nakamura, N.;
Yamaguchi, Y.; Kitagawa, A.; Kawai, S.; Nakashima, H.; Fujii, N.;
Igarashi, R. A single treatment with microcapsules containing a
CXCR4 antagonist suppressed pulmonary metastasis of murine
melanoma. Biochem. Biophys. Res. Commun. 2004, 320, 226-232.
(19) Tamamura, H.; Fujisawa, M.; Hiramatsu, K.; Mizumoto, M.; Na-
kashima, H.; Yamamoto, N.; Otaka, A.; Fujii, N. Identification of a
CXCR4 antagonist, T140 analog, as anti-rheumatoid arthritis agent.
FEBS Lett. 2004, 569, 99-104.
(20) Schols, D.; Struyf, S.; Van Damme, J.; Este, J. A,; Henson, G.; De
Clarcq, E. Inhibition of T-tropic HIV strains by selective antago-
nization of the chemokine receptor CXCR4. J. Exp. Med. 1997, 186,
1383-1388.
(21) Donzella, G. A.; Schols, D.; Lin, S. W.; Este, J. A.; Nagashima, K.
A.; Maddon P. J. AMD3100, a small molecule inhibitor of HIV-1
entry via the CXCR4 co-receptor. Nat. Med. 1998, 4, 72-76.
(22) Ichiya, 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 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.
(23) 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.; Peiper, S. C. Molecular-size reduction of
a potent CXCR4-chemokine antagonist using orthogonal combination
of conformation- and sequence-based libraries. Angew. Chem. Int.
Ed. 2003, 42, 3251-3253.
(24) Tamamura, H.; Omagari, A.; Oishi, S.; Kanamoto, T.; Yamamoto,
N.; Peiper, S. C.; Nakashima, H.; Otaka, A.; Fujii, N. Pharmacophore
identification of a specific CXCR4 inhibitor, T140, lead to develop-
ment of effective anti-HIV agents with very high selectivity indexes.
Bioorg. Med. Chem. Lett. 2000, 10, 2633-2637.
(25) Tamamura, H.; Hiramatsu, K.; Ueda, S.; Wang, Z.; Kusano, S.;
Terakubo, S.; Trent, J. O.; Peiper, S. C.; Yamamoto, N.; Nakashima,
H.; Otaka, A.; Fujii, N. Stereoselective synthesis of [L-Arg-L/ D-3-
(2-naphthyl)alanine]-type (E)-alkene dipeptide isosteres and its
application to the synthesis and biological evaluation of pseudopeptide
analogues of the CXCR4 antagonist FC131. J. Med. Chem. 2005,
48, 380-391.
(26) Tamamura, H.; Esaka, A.; Ogawa, T.; Araki, T.; Ueda, S.; Wang,
Z.; Trent, J. O.; Tsutsumi, H.; Masuno, H.; Nakashima, H.;
Yamamoto, N.; Peiper, S. C.; Otaka, A.; Fujii, N. Structure-activity
relationship studies on CXCR4 antagonists having cyclic pentapeptide
scaffolds. Org. Biomol. Chem. 2005, 3, 4392-4394.
(27) Tamamura, H.; Araki, T.; Ueda, S.; Wang, Z.; Oishi, S.; Esaka, A.;
Trent, J. O.; Nakashima, H.; Yamamoto, N.; Peiper, S. C.; Otaka,
A.; Fujii, N. Identification of novel low molecular weight CXCR4
antagonists by structural tuning of cyclic tetrapeptide scaffolds. J.
Med. Chem. 2005, 48, 3280-3289.
(28) Marshall, G. R.; Wu, Y.; Heyden, N. V.; Ratner, L.; Nikiforovich,
G. V. Backbone modification of cyclic pentapeptide antagonist of
CXCR4. In Peptides 2004 (Proceedings of 3rd international and 28th
European peptide symposium); Flegel, M., Fridkin, M., Gilon, C.,
Slaninova, J., Eds.; Kenes International: Switzerland, 2005; pp 748-
749.
(29) Miller, S. C.; Scanlan, T. S. Site-selective N-methylation of peptides
on solid support. J. Am. Chem. Soc. 1997, 119, 2301-2302.
(30) Lin, X.; Dorr, H.; Nuss, J. M. Utilization of Fukuyama’s sulfonamide
protecting group for the synthesis of N-substituted R-amino acids
and derivatives. Tetrahedron Lett. 2000, 41, 3309-3313.
(31) Carpino, L. A.; 1-Hydroxy-7-azabenzotriazole. An efficient peptide
coupling additive. J. Am. Chem. Soc. 1993, 115, 4397-4398.
(32) Bollhagen, R.; Schmiedberger, M.; Barlos. K.; Grell. E. A new reagent
for the cleavage of fully protected peptides synthesised on 2-chlo-
rotrityl chloride resin. J. Chem. Soc., Chem. Commun. 1994, 2559-
2560.
(33) Haubner, R.; Finsinger, D.; Kessler, H. Stereoisomeric peptide
libraries and peptidomimetics for designing selective inhibitors of
the Rvâ3 integrin for a new cancer therapy. Angew. Chem. Int. Ed.
Engl. 1997, 36, 1374-1389.
(34) Toniolo, C. Conformationally restricted peptides through short range
cyclization. Int. J. Pept. Protein Res. 1990, 35, 287-300.
(35) Dechantsreiter, M. A.; Planker, Eckart.; Matha, B.; Lohof, E.;
Ho¨lzemann, G.; Jonczyk, A.; Goodman, S. L.; Kessler, H. N-
Methylated cyclic RGD peptides as highly active and selective Rvâ3
integrin antagonists. J. Med. Chem. 1998, 42, 3033-3040.
(36) Rajeswaran, W. G.; Hocart, S. J.; Murphy, W. A.; Taylor, J. E.; Coy,
D. H. Highly potent and subtype selective ligand derived by N-methyl
scan of a somatostatin antagonist. J. Med. Chem. 2001, 44, 1305-
1311.
(37) Erchegyi, J.; Hoeger, C. A.; Low, W.; Hoyer, D.; Waser, B.;
Eltschinger, V.; Schaer, J.; Cescato, R.; Reubi, J. C.; Rivier, J. E.
Somatostatin receptor 1 selective analogues: 2. NR-Methylated Scan.
J. Med. Chem. 2005, 48, 507-514.
(38) 32 exhibited two sets of signals in 1H NMR spectra. The signals
were assigned to trans- (69%) and cis- (31%) conformers at D-Tyr1-
D-MeArg2 amide bond. Since backbone conformation and side chain
disposition of cis-conformer of 32 was apparently different from the
parent peptides 2 and 3, we assumed that the major trans-conformer
could contribute to the bioactivities of 32, although contribution of
the minor cis-conformer cannot be ruled out.
(39) Navenot, J. M.; Wang, Z. X.; Trent, J. O.; Murray, J. L.; Hu, Q. X.;
DeLeeuw, L.; Moore, P. S.; Chang, Y.; Peiper, S. C. Molecular
anatomy of CCR5 engagement by physiologic and viral chemokines
and HIV-1 envelope glycoproteins; Differences in primary structural
requirements for RANTES, MIP-1R, and vMIP-II binding. J. Mol.
Biol. 2001, 59, 380-393.
JM0607350