Noncompetitive antagonism and inverse agonism as mechanism of action of nonpeptidergic antagonists at primate and rodent CXCR3 chemokine receptors

Article abstract of DOI:10.1124/jpet.107.134783

The chemokine receptor CXCR3 is involved in various inflammatory diseases, such as rheumatoid arthritis, multiple sclerosis, psoriasis, and allograft rejection in transplantation patients. The CXCR3 ligands CXCL9, CXCL10, and CXCL11 are expressed at sites

Full text of DOI:10.1124/jpet.107.134783

0022-3565/08/3252-544–555$20.00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2008 by The American Society for Pharmacology and Experimental Therapeutics
JPET 325:544–555, 2008
Vol. 325, No. 2
134783/3329867
Printed in U.S.A.
Noncompetitive Antagonism and Inverse Agonism as
Mechanism of Action of Nonpeptidergic Antagonists
at Primate and Rodent CXCR3 Chemokine Receptors
Dennis Verzijl, Stefania Storelli, Danny J. Scholten, Leontien Bosch, Todd A. Reinhart,
Daniel N. Streblow, Cornelis P. Tensen, Carlos P. Fitzsimons, Guido J. R. Zaman,
James E. Pease, Iwan J. P. de Esch, Martine J. Smit, and Rob Leurs
Leiden/Amsterdam Center for Drug Research, Division of Medicinal Chemistry, Vrije Universiteit Amsterdam, Amsterdam,
The Netherlands (D.V., S.S., D.J.S., L.B., C.P.F., I.J.P.d.E., M.J.S., R.L.); Department of Infectious Diseases and Microbiology,
Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania (T.A.R.); Vaccine and Gene Therapy
Institute, Oregon Health and Science University, Beaverton, Oregon (D.N.S.); Department of Dermatology, Leiden University
Medical Center, Leiden, The Netherlands (C.P.T.); N.V. Organon, Molecular Pharmacology Unit, Oss, The Netherlands
(G.J.R.Z.); and Leukocyte Biology Section, National Heart and Lung Institute, South Kensington Campus, Faculty of Medicine,
Imperial College London, London, United Kingdom (J.E.P.)
Received November 28, 2007; accepted February 11, 2008
ABSTRACT
The chemokine receptor CXCR3 is involved in various inflam- methylphenyl)-6,7-dihydro-5H-benzocyclohepten-8-yl]-
matory diseases, such as rheumatoid arthritis, multiple scle- carbonyl]amino]benzyl] tetrahydro-2H-pyran-4-aminium
rosis, psoriasis, and allograft rejection in transplantation pa- chloride (TAK-779). To understand the action of these CXCR3
tients. The CXCR3 ligands CXCL9, CXCL10, and CXCL11 are antagonists in various animal models of disease, the com-
expressed at sites of inflammation, and they attract CXCR3- pounds were also tested at rat and mouse CXCR3, as well as at
bearing lymphocytes, thus contributing to the inflammatory CXCR3 from rhesus macaque, which was cloned and charac-
process. In this study, we characterize five nonpeptidergic terized for the first time in this study. Except for TAK-779, all
compounds of different chemical classes that block the ac- compounds show slightly lower affinity for rodent CXCR3 than
tion of CXCL10 and CXCL11 at the human CXCR3, i.e., the for primate CXCR3. In addition, we have characterized the
3H-pyrido[2,3-d]pyrimidin-4-one derivatives N-1R-[3-(4-et- molecular mechanism of action of the various antagonists at
hoxy-phenyl)-4-oxo-3,4-dihydro-pyrido[2,3-d]pyrimidin-2-yl]- the human CXCR3 receptor. All tested compounds act as non-
ethyl-N-pyridin-3-ylmethyl-2-(4-fluoro-3-trifluoromethyl- competitive antagonists at CXCR3. Moreover, this noncompe-
phenyl)-acetamide (VUF10472/NBI-74330) and N-1R-[3-(4- titive behavior is accompanied by inverse agonistic properties
ethoxy-phenyl)-4-oxo-3,4-dihydro-pyrido[2,3-d]pyrimidin-2- of all five compounds as determined on an identified constitu-
yl]-ethyl-N-pyridin-3-ylmethyl-2-(4-trifluoromethoxy-phenyl)- tively active mutant of CXCR3, CXCR3 N3.35A. It is interesting
acetamide (VUF10085/AMG-487), the 3H-quinazolin-4-one to note that all compounds except TAK-779 act as full inverse
decanoic acid {1-[3-(4-cyano-phenyl)-4-oxo-3,4-dihydro-quinazolin- agonists at CXCR3 N3.35A. TAK-779 shows weak partial in-
2-yl]-ethyl}-(2-dimethylamino-ethyl)-amide (VUF5834), the verse agonism at CXCR3 N3.35A, and it probably has a
imidazolium compound 1,3-bis-[2-(3,4-dichloro-phenyl)-2- different mode of interaction with CXCR3 than the other two
oxo-ethyl]-3H-imidazol-1-ium bromide (VUF10132), and the classes of small-molecule inverse agonists.
quaternary ammonium anilide N,N-dimethyl-N-[4-[[[2-(4-
This work was supported in part by grants from The Nederlandse Organi-
statie voor Wetenschappe-lijk Onderzoek (NWO; to M.J.S.). S.S. was supported
Chemokines are secreted peptides that are important me-
diators in inflammation. They are classified into four families
based on the number and position of conserved N-terminal
cysteine residues, i.e., CC, CXC, CX3C, and XC chemokines
(Murphy et al., 2000). Chemokines bind to a subset of G
protein-coupled receptors (GPCRs) of class A, which are
named based on their specific chemokine preferences (Mur-
by an NWO-Stichting Technische Wetenschappen Pioneer grant (to R.L.),
Dutch Top Institute Pharma (to D.J.S. and L.B.), Dutch Cancer Society Grant
RUL 2000-2168 (to C.P.T.), and the National Institutes of Health Grant
AI060422 (to T.A.R.).
S.S. and D.J.S. contributed equally to this work.
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.107.134783.
ABBREVIATIONS: GPCR, G protein-coupled receptor; CAM, constitutively active mutant; FBS, fetal bovine serum; PCR, polymerase chain
reaction; HEK, human embryonic kidney; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline; BSA, bovine serum
albumin; InsP, inositol phosphates; TBS, Tris-buffered saline; PLC, phospholipase C; TM, transmembrane; WT, wild type; HxR, histamine Hx
receptor.
544

Nonpeptidergic Antagonists of the CXCR3 Chemokine Receptor
545
phy et al., 2000). The chemokine receptor CXCR3 is mainly CXCR3 from rhesus macaque was cloned, characterized,
expressed on activated T-helper 1 cells, but also on B cells and subjected to a detailed pharmacological analysis using
and natural killer cells (Qin et al., 1998). CXCR3 is activated the nonpeptidergic compounds. Moreover, we constructed
by the interferon-␥-inducible chemokines CXCL9, CXCL10, and characterized a constitutively active mutant (CAM) of
and CXCL11, with CXCL11 having the highest affinity CXCR3, which was used to further determine the inverse
(Loetscher et al., 1996; Cole et al., 1998). Upon activation, agonistic properties of the small-molecule compounds.
CXCR3 activates pertussis toxin-sensitive G proteins of the
G␣i class and it mediates, e.g., chemotaxis, calcium flux, and
activation of kinases such as p44/p42 mitogen-activated pro-
tein kinase and Akt (Smit et al., 2003).
Materials and Methods
Materials. Dulbecco’s modified Eagle’s medium and trypsin were
purchased from PAA Laboratories GmbH (Paschen, Austria), RPMI
CXCR3 and its ligands are up-regulated in a variety of
inflammatory diseases, implying a role for CXCR3 in, e.g., 1640 medium with GlutaMAX-I and 25 mM HEPES, nonessential
amino acids, sodium pyruvate, and 2-mercaptoethanol were from
Sigma-Aldrich (St. Louis, MO), penicillin and streptomycin were
obtained from Lonza (Verviers, Belgium), fetal bovine serum (FBS)
was purchased from Integro B.V. (Dieren, The Netherlands), and
certified FBS was from Invitrogen (Paisley, UK). myo-[2-3H]Inositol
(10–20 Ci/mmol) was purchased from GE Healthcare (Chalfont St.
Giles, UK). 125I-CXCL10 (2200 Ci/mmol) and 125I-CXCL11 (2000 Ci/
mmol) were obtained from PerkinElmer Life and Analytical Sciences
(Boston, MA). Chemokines were obtained from PeproTech (Rocky Hill,
NJ). TAK-779 was obtained from the National Institutes of Health
AIDS research and reference reagent program.
rheumatoid arthritis (Qin et al., 1998), multiple sclerosis
(Sørensen et al., 1999), transplant rejection (Hancock et al.,
2000), atherosclerosis (Mach et al., 1999), and inflammatory
skin diseases (Flier et al., 2001). The role of CXCR3 in cancer
is two-fold: CXCR3 may be involved in the metastasis of
CXCR3-expressing cancer cells (Walser et al., 2006), whereas
expression of CXCL10 (Luster and Leder, 1993) or CXCL11
(Hensbergen et al., 2005) at tumor sites may attract CXCR3-
expressing immune cells that help control tumor growth and
metastasis.
DNA Constructs. The cDNA of human CXCR3 inserted in
pcDNA3 (Loetscher et al., 1996) was amplified by PCR and inserted
into pcDEF3 (a gift from Dr. Langer, Robert Wood Johnson Medical
School, Piscataway, NJ).
Several animal models have been developed for CXCR3,
among which a murine model of metastatic breast cancer
(Walser et al., 2006), a murine model of renal cell carcinoma
(Pan et al., 2006), and an arthritis model in Lewis rats (Salomon
et al., 2002). In a mouse rheumatoid arthritis model, TAK-779,
a small-molecule antagonist with affinity for CCR5, CCR2b,
and CXCR3, inhibits the development of arthritis by down-
regulating T-cell migration, indicating that targeting chemo-
kine receptors in models of inflammation is feasible and effec-
tive (Baba et al., 1999; Yang et al., 2002; Gao et al., 2003).
Several classes of small-molecule compounds targeting
CXCR3 have recently been described, including 4-N-aryl-
[1,4]diazepane ureas (Cole et al., 2006), 1-aryl-3-piperidin-
4-yl-urea derivatives (Allen et al., 2007), quinazolin-4-one,
3H-pyrido[2,3-d]pyrimidin-4-one derivatives (Heise et al.,
2005; Storelli et al., 2005; Johnson et al., 2007; Storelli et al.,
2007), and the above-mentioned quaternary ammonium an-
ilide TAK-779 (Gao et al., 2003). So far, no detailed informa-
tion is available on the molecular mechanism of action of
A cDNA containing the rhesus macaque (Macaca mulatta; Gen-
Bank accession no. EU313340) CXCR3 open reading frame was
obtained from rhesus macaque peripheral blood mononuclear cells
that were stimulated overnight with phytohemagglutinin-P plus
phorbol myristate acetate. Total RNA was extracted using TRIzol
(Invitrogen), and cDNA was generated by reverse transcription with
avian myeloblastosis virus reverse transcriptase (Promega, Madison,
WI) and oligo(dT) primer. PCR was then performed with Taq polymer-
ase (Promega) and primers TRCXCR3F2 (5Ј-AGCCCAGCCATGGTC-
CTTG-3Ј) and TRCXCR3R2 (5Ј-CCTCACAAGCCCGAGTAGGA-3Ј).
The resulting PCR product was spin-column purified (QIAGEN GmbH,
Hilden, Germany), and it was ligated to pGEM-T vector (Promega).
Later, the cDNA was subcloned into pcDEF3.
A cDNA containing the Rattus norvegicus CXCR3 open reading
frame was cloned from an F344 heart allograft that was transplanted
into a Lewis rat at 28 days after transplantation (Streblow et al.,
2003). Total RNA was prepared from 0.25 g of rat tissue using the
these small-molecule antagonists at the CXCR3 receptor, TRIzol method. First-strand cDNA was synthesized from 2 ␮g of
total RNA using 1.0 ␮M oligo(dT)-T7 primer [GGCCAGTGAATTG-
TAATACGACTCACTATAGGG(A)24] and 200 U of SuperScript III
reverse transcriptase (Invitrogen). Doubled-stranded cDNA was gen-
erated by the addition of second-strand buffer (Invitrogen) according
to the manufacturer’s protocol, and it was purified by phenol/chloro-
form extraction. PCR was performed with Taq DNA polymerase and
primers ratCXCR3Fwd: ATAGGATTCATGTACCTTGAGGTCAGT-
GAACGTCA and ratCXCR3Rev: ATAGAATTCTTACAAGCCCAAG-
TAGGAGGCCTCAGT. The resulting PCR fragment was cloned into
pGEM-Teasy (Promega), and it was subcloned into pcDEF3.
The cDNA of mouse CXCR3 was a kind gift from Dr. Luster
(Harvard Medical School, Charlestown, MA) (Lu et al., 1999), and it
was subcloned into pcDEF3. The chimeric G protein G␣qi5 (pcDNA1-
HA-mG␣qi5) was a gift from Dr. Conklin (University of California at
San Francisco, San Francisco, CA) (Coward et al., 1999). Other
plasmids that were used were pcDNA3.1-CXCR1, pcDEF3-CXCR2,
pcDNA3-CXCR4, pcDEF3-CCR1, pcDNA3.1-CCR2, pcDEF3-H1, and
pCIneo-H3(445).
despite the general notion that small molecule ligands prob-
ably will not have overlapping binding sites with the chemo-
kines, which are supposed to bind mainly to the N terminus
and extracellular receptor loops (Murphy et al., 2000). More-
over, although most compounds have been tested on human
CXCR3 using in vitro assays, little or no information on their
affinity for CXCR3 of other species is available. Especially in
view of rodent models of inflammatory diseases it is impor-
tant to know the relative affinities of the compounds for the
receptors of different species.
In this study, we report on the molecular characterization
of the 3H-pyrido[2,3-d]pyrimidin-4-one derivatives VUF10472
(NBI-74330) (Heise et al., 2005; Storelli et al., 2007) and
VUF10085 (AMG-487) (Johnson et al., 2007; Storelli et al.,
2007); the 3H-quinazolin-4-one VUF5834 (Storelli et al.,
2005; Johnson et al., 2007); the imidazolium compound
VUF10132 (Axten et al., 2003); and the quaternary ammo-
nium anilide TAK-779 (Baba et al., 1999) at CXCR3 of hu-
man (Loetscher et al., 1996), rat (Wang et al., 2000), and
mouse (Tamaru et al., 1998; Lu et al., 1999). In addition,
Synthesis of Small-Molecule Compounds. Synthesis of
VUF5834 (Storelli et al., 2005), VUF10472, and VUF10085 (Storelli
et al., 2007) has been described previously.
Synthesis of VUF10132 was adapted from the procedure described

546
Verzijl et al.
by Axten et al. (2003). In brief, 1-(3,4-dichloro-phenyl)-2-imidazol-1-
Phospholipase C Activation. Twenty-four hours after transfec-
yl-ethanone (0.19 g; 0.75 mmol) and 2-bromo-1-(3,4-dichloro-phenyl)- tion, HEK293T cells were labeled overnight in Earle’s inositol-free
ethanone (0.20 g; 0.75 mmol) were stirred in acetonitrile (50 ml) minimal essential medium (Invitrogen) supplemented with 10% fetal
overnight at room temperature. The mixture was diluted with ether,
bovine serum, penicillin, and streptomycin and with myo-[2-3H]-
and the solid product was filtered, washed with ether, and dried to inositol (1 ␮Ci/ml). Cells were washed with inositol phosphates
afford VUF10132 as white solid. Yield: 0.31 g (78%). 1H NMR (di- (InsP) assay buffer [20 mM HEPES, pH 7.4, 140 mM NaCl, 5 mM
methyl sulfoxide) ␦: 6.20 (s, 4H), 7.67 to 7.80 (m, 4H), 7.88 to 8.12 (m,
3H), 8.23 to 8.25 (m, 2H), 9.12 (s, 1H).
KCl, 1 mM MgSO4, 1 mM CaCl2, 10 mM glucose, and 0.05% (w/v)
BSA], and then they were incubated for 2 h in InsP assay buffer
supplemented with 10 mM LiCl in the presence or absence of indi-
cated ligands. The incubation was stopped by aspiration of the assay
buffer and addition of ice-cold 10 mM formic acid. After incubation
for 90 min at 4°C, [3H]InsP were isolated by anion exchange chro-
matography (Dowex AG1-X8 columns; Bio-Rad) and counted by liq-
uid scintillation.
Chemotaxis. Twenty-four hours after transfection, chemotaxis of
L1.2 cells toward 10 nM CXCL10 was determined in the presence or
absence of small-molecule compounds, using 5-␮m pore ChemoTx
96-well plates (Neuro Probe, Inc., Gaithersburg, MD). In brief,
ChemoTx plates were blocked using RPMI 1640 medium with
GlutaMAX-I and 25 mM HEPES supplemented with 1% (w/v) BSA.
Chemokine and compound dilutions were made in RPMI 1640 me-
dium with GlutaMAX-I and 25 mM HEPES supplemented with 0.1%
(w/v) BSA, and samples were added to the wells. L1.2 cells were
added on top of the membrane and incubated for 5 h in a humidified
chamber at 37°C. The number of migrated cells per well was deter-
mined using a hemocytometer.
Enzyme-Linked Immunosorbent Assay. Forty-eight hours af-
ter transfection, HEK293T cells were washed with TBS, and then
they were fixed with 4% paraformaldehyde in PBS. After blocking
with 1% skim milk in 0.1 M NaHCO3, pH 8.6, cells were incubated
with mouse monoclonal anti-CXCR3 antibody (MAB160; R&D Sys-
tems, Minneapolis, MN) in TBS containing 0.1% BSA, washed three
times with TBS, and then incubated with goat anti-mouse horserad-
ish peroxidase-conjugated secondary antibody (Bio-Rad). Later, cells
were incubated with substrate buffer containing 2 mM o-phenylene-
diamine (Sigma-Aldrich), 35 mM citric acid, 66 mM Na2HPO4, and
0.015% H2O2, pH 5.6. The reaction was stopped with 1 M H2SO4, and
absorption at 490 nm was determined.
Data Analysis. Nonlinear regression analysis of the data and
calculation of Kd and Ki values was performed using Prism 4.03
(GraphPad Software Inc., San Diego, CA). The pKb values in the PLC
activation assay were calculated using the Cheng-Prusoff equation
pKb ϭ IC50/(1 ϩ [agonist]/EC50) (Cheng and Prusoff, 1973).
Cell Culture and Transfection. HEK293T cells were grown at
5% CO2 at 37°C in Dulbecco’s modified Eagle’s medium supple-
mented with 10% fetal bovine serum, penicillin, and streptomycin.
HEK293T cells were transfected with 2.5 ␮g of cDNA encoding
CXCR3 supplemented with 2.5 ␮g of pcDNA1-HA-mG␣qi5 [for phos-
pholipase C (PLC) activation experiments and ELISA] or 2.5 ␮g of
pcDEF3 by using linear polyethyleneimine (mol. wt. 25,000; Poly-
sciences, Warrington, PA). The total amount of DNA in gene-dosing
experiments was kept constant at 5 ␮g by addition of pcDEF3. In
brief, 5 ␮g of DNA in total was diluted in 250 ␮l of 150 mM NaCl.
Later, 30 ␮g of polyethyleneimine in 250 ␮l of 150 mM NaCl was
added to the DNA solution, and then the solution was incubated for
10 min at room temperature. The mixture was added to adherent
HEK293T cells in 100 mM tissue culture dishes. The next day, cells
were trypsinized, resuspended into culture medium, and plated in
the appropriate poly-L-lysine (Sigma-Aldrich)-coated assay plates.
For membrane preparation, cells were harvested 48 h after transfec-
tion from the tissue culture dishes in which they were transfected.
Murine pre-B L1.2 cells were grown in RPMI 1640 medium with
GlutaMAX-I and 25 mM HEPES, supplemented with 10% heat-
inactivated certified FBS, penicillin, streptomycin, glutamine, non-
essential amino acids, 2-mercaptoethanol, and sodium pyruvate.
L1.2 cells were transfected with 10 ␮g/20 million cells using a Bio-
Rad Gene Pulser II (330 V and 975 ␮F; Bio-Rad, Hemel Hempstead,
UK), and they were grown in culture medium supplemented with 10
mM sodium butyrate.
Chemokine Binding. Cell membrane fractions from transiently
transfected HEK293T (CXCR3 and CXCR4) or COS-7 (CXCR2) cells
were prepared as follows. Cells were washed with ice-cold PBS,
detached using ice-cold PBS containing 1 mm EDTA, and centrifuged
at 1500g for 10 min at 4°C. Later, cells were washed with PBS and
centrifuged at 1500g for 10 min at 4°C. The pellet was resuspended
in ice-cold membrane buffer (15 mM Tris, pH 7.5, 1 mM EGTA, 0.3
mM EDTA, and 2 mM MgCl2), and then it was homogenized by 10
strokes at 1100 to 1200 rpm using a Teflon-glass homogenizer and
rotor. The membranes were subjected to two freeze-thaw cycles using
liquid nitrogen, and they were centrifuged at 40,000g for 25 min at
4°C. The pellet was rinsed with ice-cold Tris-sucrose buffer (20 mM
Tris, pH 7.4, and 250 mM sucrose), and then it was resuspended in
the same buffer and frozen in liquid nitrogen. Protein concentration
was determined using a Bio-Rad protein assay. Membranes were
incubated in 96-well plates in binding buffer (50 mM HEPES, pH 7.4,
1 mM CaCl2, 5 mM MgCl2, 100 mM NaCl, and 0.5% BSA for CXCR3
and CXCR4 radioligand binding assays and 50 mM Na2/K-phosphate
buffer, pH 7.4, and 0.5% BSA for CXCR2 radioligand binding assay)
with approximately 50 pM 125I-chemokine and displacer for 2 h at
room temperature. Later, membranes were harvested via filtration
through Unifilter GF/C plates (PerkinElmer Life and Analytical
Results
Nonpeptidergic CXCR3 Antagonists. A selection of
CXCR3-targeting small-molecule antagonists from two differ-
ent structural classes were synthesized as described previously,
and then they were subjected to detailed pharmacological anal-
ysis: quinazolinone-derived 3H-pyrido[2,3-d]pyrimidin-4-one
compounds VUF10472 (NBI-74330) (Heise et al., 2005; Storelli
et al., 2007) and VUF10085 (AMG 487) (Johnson et al., 2007;
Storelli et al., 2007), the 3H-quinazolin-4-one VUF5834 (Storelli
et al., 2005; Johnson et al., 2007), and the imidazolium com-
Sciences) that were pretreated with 1% polyethylenimine, and then pound VUF10132 (Axten et al., 2003) (Fig. 1). The well de-
they were washed three times with ice-cold wash buffer (50 mM
HEPES, pH 7.4, 1 mM CaCl2, 5 mM MgCl2, and 500 mM NaCl for
CXCR3 and CXCR4 and 50 mM Na2/K-phosphate buffer, pH 7.4,
scribed CCR5 antagonist TAK-779 (Baba et al., 1999) has been
reported to show affinity for mouse CXCR3 (Gao et al., 2003);
therefore, it was included in our set of small-molecule com-
pounds as well (Fig. 1).
Characterization of Human CXCR3. Binding studies
were performed with radiolabeled CXCL10 and CXCL11 on
membranes prepared from HEK293T cells transiently trans-
fected with cDNA encoding human CXCR3 (Loetscher et al.,
1996). Homologous displacement of the radioligands with
for CXCR2). Radioactivity was measured using
(PerkinElmer Life and Analytical Sciences).
a MicroBeta
Histamine H1 Receptor Binding. Cell homogenates from
COS-7 cells transiently transfected with cDNA encoding the human
histamine H1 receptor were incubated with 7 nM [3H]mepyramine
and 10 ␮M CXCR3 antagonist in H1 buffer (50 mM Na2/K-phosphate
buffer, pH 7.4) for 30 min at room temperature, and then they were
harvested via filtration using ice-cold H1 buffer as described above. unlabeled chemokines resulted in pKd values Ϯ S.E.M. of

Nonpeptidergic Antagonists of the CXCR3 Chemokine Receptor
547
Fig. 1. Structures of small-nonpeptidergic compounds targeting CXCR3. The structures of NBI-74330; N-1R-[3-(4-ethoxy-phenyl)-4-oxo-3,4-
dihydro-pyrido[2,3-d]pyrimidin-2-yl]-ethyl-N-pyridin-3-ylmethyl-2-(4-fluoro-3-trifluoromethyl-phenyl)-acetamide (VUF10472), AMG-487; N-1R-
[3-(4-ethoxy-phenyl)-4-oxo-3,4-dihydro-pyrido[2,3-d]pyrimidin-2-yl]-ethyl-N-pyridin-3-ylmethyl-2-(4-trifluoromethoxy-phenyl)-acetamide
(VUF10472), decanoic acid {1-[3-(4-cyano-phenyl)-4-oxo-3,4-dihydro-quinazolin-2-yl]-ethyl}-(2-dimethylamino-ethyl)-amide (VUF5834), 1,3-bis-
[2-(3,4-dichloro-phenyl)-2-oxo-ethyl]-3H-imidazol-1-ium bromide (VUF10132), and N,N-dimethyl-N-[4-[[[2-(4-methylphenyl)-6,7-dihydro-5H-
benzocyclohepten-8-yl]carbonyl]amino]benzyl] tetrahydro-2H-pyran-4-aminium chloride (TAK-779) are shown.
Fig. 2. Characterization of human CXCR3. A, homologous displacement binding at human CXCR3. Binding experiments were performed with
approximately 50 pM 125I-CXCL10 (E) or 125I-CXCL11 (F) and increasing concentrations of homologous unlabeled chemokine on membranes from
HEK293T cells transfected with cDNA encoding human CXCR3. Data are presented as percentage of total binding. pKd values for CXCL10 and
CXCL11 were 9.8 Ϯ 0.1 (n ϭ 11) and 10.6 Ϯ 0.1 (n ϭ 7), respectively. B, activation of PLC by human CXCR3. HEK293T cells were transfected with
cDNA encoding human CXCR3 and G␣qi5. After 48 h, [3H]InsP accumulation was determined in the presence of increasing concentrations of CXCL10
(E) or CXCL11 (F). Data are presented as a percentage of the maximal response obtained with 100 nM CXCL11. pEC50 values for CXCL10 and
CXCL11 were 7.5 Ϯ 0.1 (n ϭ 8) and 8.5 Ϯ 0.1 (n ϭ 12), respectively.
TABLE 1
Properties of chemokine ligands at CXCR3
The pKd values and Bmax values Ϯ S.E.M. are shown for human CXCL10 and human CXCL11 at the various CXCR3, calculated from homologous radioligand binding
experiments. pEC50 values Ϯ S.E.M. were obtained using the PLC activation assay.
Human CXCL10
Human CXCL11
Species
pKd
n
Bmax
n
pEC50
n
pKd
n
Bmax
n
pEC50
n
Human
N3.35A
Rhesus
Rat
9.8 Ϯ 0.1
10.2 Ϯ 0.2
10.0 Ϯ 0.1
10.2 Ϯ 0.2
10.1 Ϯ 0.1
11
4
358 Ϯ 53
327 Ϯ 91
163 Ϯ 7
631 Ϯ 141
658 Ϯ 127
6
4
2
7
3
7.5 Ϯ 0.1
7.5 Ϯ 0.0
7.4 Ϯ 0.1
8.8 Ϯ 0.1
8.3 Ϯ 0.1
8
2
3
3
5
10.6 Ϯ 0.1
11.3 Ϯ 0.3
10.8 Ϯ 0.3
10.7 Ϯ 0.2
10.7 Ϯ 0.3
7
3
3
4
4
407 Ϯ 126
564 Ϯ 242
255 Ϯ 114
1170 Ϯ 644
1882 Ϯ 858
4
3
4
3
4
8.5 Ϯ 0.1
8.7 Ϯ 0.1
8.5 Ϯ 0.1
9.5 Ϯ 0.1
9.2 Ϯ 0.2
12
4
3
3
7
3
3
3
Mouse

548
Verzijl et al.
Fig. 3. Snakeplot of CXCR3. Amino acids that were mutated to generate constitutively active mutants (T2.56, N3.35, and D3.49) are indicated as bold
circled residues. Amino acids that differ between human and rhesus macaque CXCR3 (T83 in intracellular loop 1 and V6.39 are both Ala in rhesus
macaque CXCR3) are shown as gray circles. Residues that are different between primate and rodent (i.e., rat and/or mouse) CXCR3 are indicated as
black circles. Human and rhesus macaque CXCR3 have an additional Cys (C37) in their amino terminus compared with rat and mouse CXCR3. Helix
8 at the membrane-proximal part of the carboxyl terminus and potential palmitoylation of Cys338 are shown as well.
9.8 Ϯ 0.1 for CXCL10 and 10.6 Ϯ 0.1 for CXCL11 (Fig. 2A; sequences of human and rodent CXCR3 are apparent (Fig. 3),
Table 1), in accordance with the values and rank order for we set out to characterize CXCR3 of rhesus macaque, mouse,
these chemokines reported in literature (Cole et al., 1998; and rat. CXCR3 of mouse (Tamaru et al., 1998; Lu et al., 1999)
Heise et al., 2005).
and rat (Wang et al., 2000) have been cloned and described
In HEK293 cells, human CXCR3 does not couple efficiently previously. Here, we report the cloning and characterization of
to the PLC-InsP pathway (Wang et al., 2000). Therefore, rhesus macaque CXCR3 from peripheral blood mononuclear
CXCR3 was transiently cotransfected with cDNA encoding cells (GenBank accession no. EU313340; see Materials and
the chimeric G protein G␣qi5 (Coward et al., 1999), after Methods). The deduced amino acid sequence of rhesus macaque
which robust PLC activation could be observed upon stimu- CXCR3 was found to be 99% identical to human CXCR3. The
lation with CXCL10 and CXCL11. Using this functional as- Thr83 residue in the first intracellular loop of human CXCR3 is
say, pEC50 values of 7.5 Ϯ 0.1 and 8.5 Ϯ 0.1 were obtained for Ala83 in rhesus macaque CXCR3, and Val259 at position 6.391 in
CXCL10 and CXCL11, respectively (Fig. 2B; Table 1). TM6 of human CXCR3 is Ala259 in rhesus macaque CXCR3
CXCL10 acted as a partial agonist, with an intrinsic activity (Fig. 3).
of 0.48 Ϯ 0.8 compared with the full agonist CXCL11 (Fig.
2B; n ϭ 5).
After expression in HEK293T cells, CXCR3 from rhesus,
Characterization of Rhesus Macaque, Rat, and Mouse
CXCR3. Because the affinities of small-molecule compounds
targeting chemokine receptors may differ among species, e.g.,
for TAK-779 at human and mouse CCR5 (Baba et al., 1999; Gao
1 GPCR numbering according to Ballesteros and Weinstein (1995), where
the first digit refers to the transmembrane helix, and the second number
indicates the position of the residue relative to the most conserved amino acid,
which is assigned position 50 in that helix (i.e., N1.50, D2.50, R3.50, W4.50,
et al., 2003), and significant differences between the amino acid P5.50, P6.50, and P7.50).

Nonpeptidergic Antagonists of the CXCR3 Chemokine Receptor
549
rat, and mouse showed similar binding affinities for human CXCL10, whereas 10.3 Ϯ 0.8% (n ϭ 3) of the cells migrated to
CXCL10 and CXCL11 as human CXCR3 (Table 1). Bmax 10 nM CXCL10 in the absence of inhibitors (Fig. 5E, inset).
values of rodent CXCR3 were increased compared with hu- Both compounds completely inhibited 10 nM CXCL10-induced
man CXCR3, approximately 2-fold higher for CXCL10 and 3- migration at a concentration of 10 ␮M, with pIC50 values of
to 4-fold higher for CXCL11 (Table 1) when equal amounts 6.8 Ϯ 0.2 (n ϭ 2) and 5.8 Ϯ 0.1 (n ϭ 3) for VUF10085 and
of cDNA were transfected. The Bmax value for CXCL10 at VUF5834, respectively (Fig. 5E).
rhesus CXCR3 was 2-fold lower than for human CXCR3,
Next, the affinity of the small-molecule compounds for
whereas human and rhesus CXCR3 had comparable Bmax CXCR3 of the different species was determined using 125I-
values for CXCL11 (Table 1). Rhesus CXCR3 stimulated CXCL10. The pKi values of the compounds tested at rhesus
PLC with similar maximal effect (Emax) and similar EC50 CXCR3 were identical to the values found for human CXCR3
values for human CXCL10 and CXCL11 as human CXCR3 (Table 2). Likewise, the affinities of the compounds tested at
(Fig. 4, A and B; Table 1). In contrast, the rodent CXCR3 rat and mouse CXCR3 were identical to each other. However,
species had approximately 10-fold higher potencies and VUF10472, VUF10085, VUF5834, and VUF10132 showed an
showed increased Emax for human CXCL10 compared with approximately 4-fold lower affinity for the rodent CXCR3
human CXCR3 (Fig. 4, A and B; Table 1). The Emax for (mouse and rat) compared with the primate CXCR3 (human
human CXCL11 at rat CXCR3 was also increased com- and rhesus). Only for TAK-779 no difference in affinity be-
pared with the effect of CXCL11 at CXCR3 of the other tween CXCR3 from the different species was observed.
species (Fig. 4B).
Because the nonpeptidergic compounds showed slightly
Behavior of Nonpeptidergic Compounds at CXCR3. lower affinity for rodent CXCR3 and the previous experi-
The panel of small-molecule antagonists inhibited 125I-CXCL10 ments were performed with human chemokines, we also de-
binding to human CXCR3, with Ki values ranging between 4 termined the ability of the antagonists to inhibit mouse
nM for VUF10472 and 1.3 ␮M for TAK-779 (Fig. 5A; Table 2). CXCL10-induced signaling at mouse CXCR3. The pEC50 of
VUF10085, VUF5834, and VUF10132 showed intermediate Ki mouse CXCL10 at mouse CXCR3 (8.4 Ϯ 0.1; n ϭ 3) was
values of 20, 158, and 251 nM, respectively (Fig. 5A; Table 2). comparable with the pEC50 of human CXCL10 at mouse
Subsequently, the ability of the small molecules to inhibit CXCR3 (8.3 Ϯ 0.1; n ϭ 5). The different antagonists inhibited
CXCL10-induced PLC activation through human CXCR3 was mouse CXCL10-induced activation of mouse CXCR3 (mea-
determined. All compounds reduced CXCL10-mediated (10–50 sured as PLC stimulation), with Kb values ranging from 4 nM
nM) PLC activation, with Kb values ranging from 6 nM for for VUF10472 to 400 nM for TAK-779. The Kb values ob-
VUF10472 to 800 nM for TAK-779 (Fig. 5B; Table 2). A good tained under these conditions are within 2-fold of the Kb
correlation (r2 ϭ 0.997) was observed between the measured values for inhibition of human CXCL10 at human CXCR3
pKi values and pKb values determined against CXCL10 (Fig. (Fig. 5F; Table 2).
5C). For the most potent compound, VUF10472, inhibition of
To determine whether the antagonists show competitive or
CXCL11-induced (5 nM) PLC activation was also determined, noncompetitive behavior, a Schild analysis was performed.
resulting in a Kb value of 12.6 nM (Fig. 5D). This value corre- HEK293T cells expressing human CXCR3 and G␣qi5 were
sponds well to the value obtained against CXCL10 (Fig. 5D; incubated with increasing concentrations of CXCL10 (Fig.
Table 2). Finally, the ability of the small-molecule compounds to 6A) or CXCL11 (Fig. 6B) in the absence or presence of differ-
inhibit chemotaxis was investigated. The murine pre-B cell line ent concentrations of VUF10472. A decreased maximal effect
L1.2 was transiently transfected with cDNA encoding human combined with a rightward shift of the curves was observed
CXCR3, and migration of cells toward CXCL10 in the presence in the presence of VUF10472, indicating noncompetitive an-
or absence of VUF10085 and VUF5834 was determined. tagonist behavior. The mechanism of antagonism of the other
CXCR3-transfected L1.2 cells did not migrate in the absence of classes of small-molecule compounds was investigated using
Fig. 4. Activation of PLC by primate and rodent CXCR3. A, CXCL10-induced activation of PLC. HEK293T cells were transfected with cDNA encoding
human, rhesus, rat, or mouse CXCR3 and G␣qi5. After 48 h, [3H]InsP accumulation was determined in the presence of increasing concentrations of
human CXCL10. pEC50 values for CXCL10 were 7.5 Ϯ 0.1 (human CXCR3; n ϭ 8), 7.4 Ϯ 0.1 (rhesus CXCR3; n ϭ 3), 8.8 Ϯ 0.1 (rat CXCR3; n ϭ 3),
and 8.3 Ϯ 0.1 (mouse CXCR3; n ϭ 5). B, CXCL11-induced activation of PLC. HEK293T cells were transfected with cDNA encoding human, rhesus,
rat, or mouse CXCR3 and G␣qi5. After 48 h, [3H]InsP accumulation was determined in the presence of increasing concentrations of human CXCL11.
pEC50 values for CXCL11 were 8.5 Ϯ 0.1 (human CXCR3; n ϭ 12), 8.5 Ϯ 0.1 (rhesus CXCR3; n ϭ 3), 9.5 Ϯ 0.1 (rat CXCR3; n ϭ 3), and 9.2 Ϯ 0.2 (mouse
CXCR3; n ϭ 3). Data are presented per receptor as a percentage of [3H]InsP accumulation in the absence of chemokine.

550
Verzijl et al.
Fig. 5. Characterization of small nonpeptidergic compounds at human CXCR3. A, displacement of 125I-CXCL10. Binding experiments were performed
with approximately 50 pM 125I-CXCL10 and increasing concentrations of VUF10472 (f), VUF10085 (Ⅺ), VUF5834 (F), VUF10132 (E), or TAK-779
(Œ) on membranes from HEK293T cells transfected with cDNA encoding human CXCR3. Data are presented as percentage of total binding. pKi values
were 8.4 Ϯ 0.1 (VUF10472; n ϭ 7), 7.7 Ϯ 0.0 (VUF10085; n ϭ 18), 6.8 Ϯ 0.1 (VUF5834; n ϭ 3), 6.6 Ϯ 0.0 (VUF10132; n ϭ 4), and 5.9 Ϯ 0.0 (TAK-779;
n ϭ 6). B, inhibition of CXCL10-induced PLC activation. HEK293T cells were transfected with cDNA encoding human CXCR3 and G␣qi5. After 48 h,
CXCL10-induced (10–50 nM) [3H]InsP accumulation was determined in the presence of increasing concentrations of VUF10472 (f), VUF10085 (Ⅺ),
VUF5834 (F), VUF10132 (E), or TAK-779 (Œ). pKb values were 8.2 Ϯ 0.1 (VUF10472; n ϭ 5), 7.7 Ϯ 0.1 (VUF10085; n ϭ 9), 7.0 Ϯ 0.1 (VUF5834; n ϭ
5), 6.7 Ϯ 0.1 (VUF10132; n ϭ 5), and 6.1 Ϯ 0.1 (TAK-779; n ϭ 4). C, relationship between pKi and pKb of small-molecule compounds. pKb values
obtained using CXCL10 were plotted against pKi values obtained against 125I-CXCL10 (r2 ϭ 0.997). D, inhibition of CXCL11-induced PLC activation
by VUF10472. HEK293T cells were transfected with cDNA encoding human CXCR3 and G␣qi5. After 48 h, CXCL11-induced (5 nM) [3H]InsP
accumulation was determined in the presence of increasing concentrations of VUF10472 (f). The pKb value of VUF10472 was 7.9 Ϯ 0.0 (n ϭ 3). For
comparison, inhibition of CXCL10-induced (50 nM) PLC activation by VUF10472 is shown as well (Ⅺ). Data are represented as percentage of
stimulation in the absence of antagonist. pKb values were calculated using the Cheng-Prusoff equation (Cheng and Prusoff, 1973). E, inhibition of
CXCL10-induced chemotaxis. L1.2 cells were transfected with cDNA encoding human CXCR3. After 24 h, inhibition of CXCL10-induced (10 nM)
chemotaxis was determined. pIC50 values were 6.8 Ϯ 0.2 (VUF10085; n ϭ 2) and 5.8 Ϯ 0.1 (VUF5834; n ϭ 3). Data are presented as percentage of
migrated cells in the absence of antagonist. Inset, CXCL10-induced chemotaxis of L1.2 cells. Data are presented as percentage of total L1.2 cells that
migrated to 10 nM CXCL10 in the absence of antagonist (n ϭ 3). F, inhibition of mouse CXCL10-induced PLC activation through mouse CXCR3.
HEK293T cells were transfected with cDNA encoding mouse CXCR3 and G␣qi5. After 48 h, mouse CXCL10-induced (20 nM) [3H]InsP accumulation
was determined in the presence of increasing concentrations of VUF10472 (f), VUF10085 (Ⅺ), VUF5834 (F), VUF10132 (E), or TAK-779 (Œ). pKb
values were 8.4 Ϯ 0.1 (VUF10472; n ϭ 4), 7.7 Ϯ 0.1 (VUF10085; n ϭ 4), 7.4 Ϯ 0.2 (VUF5834; n ϭ 3), 6.6 Ϯ 0.1 (VUF10132; n ϭ 4), and 6.4 Ϯ 0.1
(TAK-779; n ϭ 4).

Nonpeptidergic Antagonists of the CXCR3 Chemokine Receptor
551
TABLE 2
Properties of small nonpeptidergic compounds at CXCR3
pKi values Ϯ S.E.M. were generated in heterologous radioligand binding experiments using human 125I-CXCL10. pKb values were obtained at human CXCR3 with human
CXCL10 and at mouse CXCR3 with mouse CXCL10 in the PLC activation assay, and pIC50 values were generated using human CXCR3 N3.35A in the PLC activation assay.
pKi
pKb
pIC50
hCXCR3
hCXCL10
mCXCR3
mCXCL10
hCXCR3
N3.35A
Human
n
N3.35A
n
Rhesus
n
Rat
n
Mouse
n
n
n
n
VUF10472
VUF10085
VUF5834
VUF10132
TAK-779
8.4 Ϯ 0.1
7.7 Ϯ 0.0
6.8 Ϯ 0.1
6.6 Ϯ 0.0
5.9 Ϯ 0.0
7
18
3
4
6
7.9 Ϯ 0.2
7.3 Ϯ 0.1
6.4 Ϯ 0.0
5.9 Ϯ 0.1
5.9 Ϯ 0.0
3
3
3
3
3
8.4 Ϯ 0.1
7.5 Ϯ 0.0
6.8 Ϯ 0.0
6.6 Ϯ 0.1
5.9 Ϯ 0.1
3
3
3
3
3
7.8 Ϯ 0.0
6.9 Ϯ 0.1
6.2 Ϯ 0.1
6.0 Ϯ 0.1
5.8 Ϯ 0.0
3
10
6
6
6
7.9 Ϯ 0.1
7.0 Ϯ 0.1
6.2 Ϯ 0.1
6.0 Ϯ 0.1
5.9 Ϯ 0.0
3
3
3
3
3
8.2 Ϯ 0.1
7.7 Ϯ 0.1
7.0 Ϯ 0.1
6.7 Ϯ 0.1
6.1 Ϯ 0.1
5
9
5
5
4
8.4 Ϯ 0.1
7.7 Ϯ 0.1
7.4 Ϯ 0.2
6.6 Ϯ 0.1
6.4 Ϯ 0.1
4
4
3
4
4
8.1 Ϯ 0.1
8.0 Ϯ 0.1
7.3 Ϯ 0.0
6.6 Ϯ 0.1
5.9 Ϯ 0.1
3
3
3
3
3
Fig. 6. Noncompetitive behavior of nonpeptidergic compounds at human CXCR3. A, Schild analysis of VUF10472 using CXCL10 as agonist. HEK293T
cells were transfected with cDNA encoding human CXCR3 and G␣qi5. After 48 h, CXCL10-induced [3H]InsP accumulation was determined in the
absence (f) or in the presence of 10 nM (Ⅺ) or 30 nM (F) VUF10472 (n ϭ 2–3). B to D, Schild analysis using CXCL11 as agonist. HEK293T cells were
transfected with cDNA encoding human CXCR3 and G␣qi5. After 48 h, CXCL11-induced [3H]InsP accumulation was determined in the absence (f) or
in the presence of 30 nM (F), 100 nM (E), 1 ␮M (Œ), 10 ␮M (‚), or 100 ␮M (ࡗ) VUF10472 (B; n ϭ 6), VUF10132 (C; n ϭ 6), or TAK-779 (D; n ϭ 7).
Data are presented as percentage of chemokine-induced (1 ␮M) stimulation in the absence of small-molecule compound.
only CXCL11. VUF10132 (Fig. 6C) and TAK-779 (Fig. 6D) compared with 71% inhibition of CXCR3-mediated response.
both clearly showed noncompetitive antagonistic behavior The other CXCR3 antagonists did not significantly inhibit the
similar to that of VUF10472.
tested chemokine or histamine receptors (Table 3). In addition,
Specificity of the Nonpeptidergic CXCR3 Antagonists. the compounds were tested in radioligand binding studies on
The CXCR3 antagonists were tested against a small panel of membranes expressing CXCR2, CXCR4, or the histamine H1
chemokine and histamine receptors to determine their specific- receptor. At the tested concentration (10 ␮M), the CXCR3 an-
ity. HEK293T cells expressing the different GPCRs were incu- tagonists typically inhibit 125I-CXCL10 binding to CXCR3 with
bated with their respective agonists in the presence or absence 80 to 100% (Fig. 5A). Besides TAK-779, which inhibited binding
of CXCR3 antagonists, and activation of PLC was determined. of [3H]mepyramine to the histamine H1 receptor with only 41%
The percentage inhibition of agonist-induced PLC activation by at 10 ␮M, no significant inhibition of radioligand binding to
the CXCR3 antagonists is shown in Table 3. As expected, the CXCR2, CXCR4, or the H1 receptor by the other CXCR3 antag-
CXCR3 antagonists (10 ␮M) showed 50 to 100% inhibition of onists was observed.
CXCL11-induced (100 nM) PLC activation through CXCR3,
Nonpeptidergic Antagonists Are Inverse Agonists at
and the CCR5/CCR2/CXCR3 antagonist TAK-779 inhibited a Constitutively Active Mutant of CXCR3. Inverse ago-
CCL2-induced activation of CCR2 with 95%. Unexpectedly, nism has recently been shown to be an important molecular
VUF10132 inhibited CCL5-induced CCR1 activation with 61%, mechanism of action of small-molecule antagonists at a va-

552
Verzijl et al.
TABLE 3
Selectivity of CXCR3 antagonists
HEK293T cells coexpressing indicated human GPCRs and G␣qi5 were stimulated with indicated chemokines (100 nM) or histamine (10 ␮M) in the presence of CXCR3
antagonists (10 ␮M). The percentage of inhibition of agonist-induced PLC activation is shown. Experiments were performed in duplicate and repeated two (CXCR2) or three
times. Radioligand binding studies were performed on membranes of HEK293T (CXCR4) or COS-7 cells (CXCR2 and histamine H1 receptor). The percentage of inhibition
of specific radioligand binding by the CXCR3 antagonists (10 ␮M) is shown. Experiments were performed in triplicate and repeated two (CXCR2) or three times.
Inhibition of Agonist-Induced PLC Activation
Inhibition of Specific Binding
CXCR3
CXCL11
CXCR1
CXCL8
CXCR2
CXCL1
CXCR4
CXCL12
CCR1
CCL5
CCR2
CCL2
H1R
Histamine
H3R
CXCR2
CXCR4
H1R
Histamine 125I-CXCL8 125I-CXCL12 ͓3H͔Mepyramine
%
VUF10472 103 Ϯ 8
VUF10085 100 Ϯ 13
Ϫ19 Ϯ 16
Ϫ7 Ϯ 17
Ϫ7 Ϯ 29
Ϫ2 Ϯ 10 Ϫ33 Ϯ 5
Ϫ8 Ϯ 25
8 Ϯ 1 Ϫ7 Ϯ 5
6 Ϯ 2 1 Ϯ 8
13 Ϯ 3 Ϫ19 Ϯ 5
24 Ϯ 2
Ϫ6 Ϯ 7
Ϫ3 Ϯ 2
Ϫ3 Ϯ 13
61 Ϯ 4
11 Ϯ 3
10 Ϯ 4
Ϫ1 Ϯ 4
Ϫ1 Ϯ 11
95 Ϯ 6
Ϫ12 Ϯ 24 Ϫ14 Ϯ 2
Ϫ37 Ϯ 11 Ϫ33 Ϯ 14 Ϫ15 Ϯ 3
11 Ϯ 17 Ϫ35 Ϯ 12 Ϫ13 Ϯ 7
6 Ϯ 15
17 Ϯ 11
18 Ϯ 7
0 Ϯ 7
4 Ϯ 12
Ϫ9 Ϯ 32
30 Ϯ 13
20 Ϯ 10
28 Ϯ 11
30 Ϯ 6
VUF5834
VUF10132
TAK-779
67 Ϯ 9
71 Ϯ 6
53 Ϯ 4
0 Ϯ 2
Ϫ10 Ϯ 26
Ϫ3 Ϯ 8
Ϫ3 Ϯ 15 Ϫ12 Ϯ 6
Ϫ21 Ϯ 14
Ϫ8 Ϯ 4 Ϫ2 Ϯ 16 Ϫ12 Ϯ 4
41 Ϯ 7
riety of GPCR family members (Parra and Bond, 2007). Hu- T2.56P, significant constitutive activity was shown, whereas
man chemokine receptors generally do not show high levels no constitutive activity was found for the CXCR3 D3.49V
of constitutive activity. However, constitutively active che- mutant (data not shown). In view of the signal-to-noise ratio,
mokine receptor mutants have been described that signal in we choose CXCR3 N3.35A for further characterization. Upon
the absence of ligands. Examples are CXCR2 D3.49V (Burger transfection of HEK293T cells with increasing amounts of
et al., 1999), mutation of N3.35 of CXCR4 (Zhang et al., cDNA encoding CXCR3 WT or CXCR3 N3.35A, an increase in
2002), and mutation of T2.56 in CCR2 and CCR5 (Arias et al., receptor expression was detected in ELISA experiments (Fig.
2003). To determine whether the studied CXCR3 antagonists 7A). When similar amounts of cDNA were transfected,
act as neutral antagonists or inverse agonists at CXCR3, CXCR3 WT was expressed at higher levels than CXCR3
analogous CAMs of CXCR3 were constructed (Fig. 3). Upon N3.35A (Fig. 7A). As expected, CXCR3 WT did not activate
expression of the CXCR3 mutants N3.35A, N3.35S, and PLC in the absence of ligands (Fig. 7B). Even though CXCR3
Fig. 7. Characterization of the constitutively active mutant CXCR3 N3.35A. A, expression of CXCR3 WT and CXCR3 N3.35A. HEK293T cells were
transfected with cDNA encoding G␣qi5 and increasing amounts of cDNA encoding human CXCR3 WT or CXCR3 N3.35A. Mock cells were transfected
with cDNA encoding G␣qi5 and pcDEF3 empty vector. After 48 h, CXCR3 expression was determined in an ELISA assay. Data are presented as
percentage signal obtained in mock-transfected cells (n ϭ 2–3). B, constitutive activity of CXCR3 N3.35A. HEK293T cells were transfected with cDNA
encoding G␣qi5 and increasing amounts of cDNA encoding human CXCR3 WT or CXCR3 N3.35A. Mock cells were transfected with cDNA encoding
G␣qi5 and pcDEF3 empty vector. After 48 h, [3H]InsP accumulation was determined in the absence of chemokines. Data are presented as percentage
of [3H]InsP in mock-transfected cells (n ϭ 3). C, chemokine-induced PLC activation by CXCR3 N3.35A. HEK293T cells were transfected with cDNA
encoding CXCR3 N3.35A and G␣qi5. After 48 h, [3H]InsP accumulation was determined in the presence of increasing concentrations CXCL10 or
CXCL11. pEC50 values for CXCL10 and CXCL11 were 7.5 Ϯ 0.0 (n ϭ 2) and 8.7 Ϯ 0.1 (n ϭ 4), respectively. Data are presented as percentage of the
maximal response obtained with CXCL11 (100 nM). D, inverse agonism at CXCR3 N3.35A. HEK293T cells were transfected with cDNA encoding
CXCR3 N3.35A and G␣qi5. After 48 h, [3H]InsP accumulation was determined in the presence of increasing concentrations VUF10472 (f), VUF10085
(Ⅺ), VUF5834 (F), VUF10132 (E), or TAK-779 (Œ). pIC50 values (n ϭ 3) were 8.1 Ϯ 0.1 (VUF10472), 8.0 Ϯ 0.1 (VUF10085), 7.3 Ϯ 0.0 (VUF5834), 6.6 Ϯ
0.1 (VUF10132), and 5.9 Ϯ 0.1 (TAK-779). Data are presented as percentage of CXCR3 N3.35A-mediated constitutive [3H]InsP accumulation.

Nonpeptidergic Antagonists of the CXCR3 Chemokine Receptor
553
N3.35A was expressed at lower levels than CXCR3 WT, tutively active receptor underlies the pathogenesis. To date,
CXCR3 N3.35A showed a marked increase of PLC activation there are no reports about constitutive activity of CXCR3,
upon transfection of increasing amounts of cDNA (Fig. 7B).
and there are only few examples of constitutive activity of
The affinities of CXCL10 (pKd of 10.2 Ϯ 0.2) and CXCL11 chemokine receptors. Constitutive activity is a function of the
(pKd of 11.3 Ϯ 0.3) for CXCR3 N3.35A were slightly higher relative stoichiometry of receptors and G proteins (Kenakin,
than for CXCR3 WT (Table 1). CXCL10 and CXCL11 acti- 1996), and chemokine receptors, including CXCR3, are often
vated CXCR3 N3.35A over basal levels in a similar manner up-regulated under inflammatory conditions (Rabin et al.,
as CXCR3 WT, with pEC50 values of 7.5 Ϯ 0.0 (n ϭ 2) and 1999; Murphy et al., 2000). Hence, under pathophysiological
8.7 Ϯ 0.1 (n ϭ 4), respectively (Fig. 7C; Table 1). Subse- conditions constitutive activity of chemokine receptors might
quently, the affinity of the small-molecule antagonists at become apparent; therefore, the use of inverse agonists ben-
CXCR3 N3.35A was determined using 125I-CXCL10. The af- eficial. To study the relative efficacy of the nonpeptidergic
finity of all compounds, except TAK-779, was slightly re- antagonists, we generated mutants of CXCR3, i.e., CXCR3
duced for CXCR3 N3.35A compared with CXCR3 WT (Table N3.35A, N3.35S, D3.39V, and T2.56P, based on CAMs of
2). Finally, the effect of the small-molecule antagonists on chemokine receptors reported in literature (Burger et al.,
the constitutive activation of PLC was investigated. 1999; Zhang et al., 2002; Arias et al., 2003). All generated
VUF10472, VUF10085, VUF5834, and VUF10132 all acted mutants, except for CXCR3 D3.49V with a mutation in the
as full inverse agonists, with IC50 values ranging between 8 conserved DRY motif at the cytoplasmatic end of TM3, dis-
nM for VUF10472 and 251 nM for VUF10132 (Fig. 7D; Table played basal signaling. Mutation of the DRY motif of CXCR2
2). In contrast, TAK-779 acted as a partial inverse agonist, to VRY confers constitutive activity to CXCR2 (Burger et al.,
with an intrinsic activity of Ϫ0.42 Ϯ 0.0 and an IC50 of 1.3 1999), whereas the same mutation for CXCR1 (Burger et al.,
␮M (n ϭ 3) (Fig. 7D; Table 2).
1999) or CCR5 (Lagane et al., 2005) did not result in consti-
tutive activity. In contrast, mutation of the VRY motif to
DRY in the highly constitutively active chemokine receptor
ORF74 encoded by Kaposi’s sarcoma-associated herpesvirus
Discussion
CXCR3 has attracted considerable attention as a new drug did not diminish its constitutive activity but even increased it
target due to its involvement in a variety of serious disorders, (Ho et al., 2001). Therefore, mutation of D3.49 to Val does not
including cancer (Walser et al., 2006), atherosclerosis (Mach seem to be a universal switch for constitutive activity in
et al., 1999), inflammatory disorders such as rheumatoid chemokine receptors. In the same way, mutation of T2.56 in
arthritis (Qin et al., 1998), skin diseases (Flier et al., 2001), the conserved TXP motif resulted in a CAM for CXCR3 (this
and transplant rejection (Hancock et al., 2000). After the study), CCR5, and CCR2, but not for CCR1, CCR3, CCR4,
recognition of CXCR3 as a potential attractive drug target, CXCR2, and CXCR4 (Arias et al., 2003). Mutation of N3.35 in
several small-molecule CXCR3 antagonists have recently the N(L/F)Y motif in TM3 of CXC chemokine receptors re-
been identified (Gao et al., 2003; Heise et al., 2005; Storelli et sulted in CAMs for both CXCR3 (this study) and CXCR4
al., 2005; Cole et al., 2006; Allen et al., 2007; Johnson et al., (Zhang et al., 2002). The nonpeptidergic compounds acted as
2007; Storelli et al., 2007). In this study, we selected five full inverse agonists at CXCR3 N3.35A, except for TAK-779,
nonpeptidergic antagonists from four different structural which only partially inhibited constitutive signaling at the
classes, and we studied their mechanism of action at the highest concentration used (Fig. 7D). Because VUF10132
human CXCR3. In addition, the interaction of these antago- and TAK-779 have similar affinities for CXCR3 N3.35A (Ta-
nists with human, rhesus macaque, rat, and mouse CXCR3 ble 2), it seems that TAK-779 lacks certain structural fea-
was investigated.
tures needed for full inverse agonism at CXCR3. It is inter-
All tested nonpeptidergic antagonists inhibited CXCL10- esting to note that TAK-779 acts as a full inverse agonist at
induced activation of PLC, with pKb values that correlated CCR5 (Lagane et al., 2005). As expected for a receptor in the
very well with their affinity, with a rank order VUF10472 Ͼ active state (Kenakin, 1996), the affinities of the agonists
VUF10085 Ͼ VUF5834 Ͼ VUF10132 Ͼ TAK-779 (Table 2). CXCL10 and CXCL11 for CXCR3 N3.35A were increased
Subsequently, the mechanism of action of the antagonists compared with CXCR3 WT (Table 1). The full inverse ago-
was explored using Schild analysis. The dose-response curves nists VUF10472, VUF10085, VUF5834, and VUF10132,
for CXCL10 and CXCL11 did not reach the maximal response which are predicted to have a higher affinity for an inactive
in the presence of VUF10472 (Fig. 6), as has been shown receptor conformation (Kenakin, 1996), all showed reduced
previously by others for human and mouse CXCR3 (Heise et affinity for CXCR3 N3.35A compared with CXCR3 WT (Table
al., 2005; Jopling et al., 2007). Antagonists from other struc- 2). In contrast, the affinity of the weak partial inverse agonist
tural classes also decreased the maximum response of TAK-779 did not change, indicating a different mode of in-
CXCL11, in a manner that follows the rank order of their teraction with CXCR3 compared with the other structural
affinities. This indicates that all tested nonpeptidergic an- classes of compounds.
tagonists behave as noncompetitive antagonists for the PLC
activation by the endogenous agonists of CXCR3.
If antagonists are to be tested in animal models, detailed
information on the interaction of the compound with the recep-
Antagonists of GPCRs can be classified as neutral antag- tor of that specific species is required, because species differ-
onists or (partial) inverse agonists (Kenakin, 1996; Parra and ences may occur. We therefore tested the nonpeptidergic antag-
Bond, 2007). Although both antagonist classes are able to onists on rhesus macaque CXCR3, which was cloned in this
block an agonist-induced response by occupying the receptor, study, as well as on rat and mouse CXCR3. Human and rhesus
there is an ongoing debate about the exact clinical benefits or macaque show 99% amino acid identity (Fig. 3). Consistent with
disadvantages of each class (Parra and Bond, 2007). Treat- this high homology, no differences in affinity or potency of
ment with inverse agonists may be beneficial when a consti- CXCL10 and CXCL11, or in the affinity of the nonpeptidergic

554
Verzijl et al.
antagonists for the two receptors were found. Likewise, the difference in affinity observed between human and rodent
protein sequences of rat and mouse CXCR3 are 96% identical, CXCR3 imply that VUF10472/NBI-74330, VUF10085/AMG-
and no significant differences in affinity of the endogenous 487, and VUF5834 are useful in rodent models of CXCR3-
agonists or the nonpeptidergic compounds are observed be- mediated pathogenesis. Interestingly, it was found that the
tween the CXCR3 of these rodent species. However, lower se- nonpeptidergic antagonists act as inverse agonists at a con-
quence identity is found when human and rhesus macaque stitutively active CXCR3 mutant.
CXCR3s are compared with rat and mouse CXCR3. Approxi-
mately 85% identity exists between the primate (human and
rhesus macaque) and rodent (rat and mouse) species (Fig. 3).
Nevertheless, the affinities of CXCL10 and CXCL11 found for
rodent CXCR3 were comparable with the affinities found for
Acknowledgments
We thank Kamonchanok Sansuk and Anne O. Watts for help with
some of the experiments.
primate CXCR3 (Table 1). Although the affinities of CXCL10
and CXCL11 were comparable, the affinities of VUF10472,
VUF10084, VUF5834, and VUF101032 were only approxi-
References
Allen DR, Bolt A, Chapman GA, Knight RL, Meissner JW, Owen DA, and Watson RJ
(2007) Identification and structure-activity relationships of 1-aryl-3-piperidin-4-
yl-urea derivatives as CXCR3 receptor antagonists. Bioorg Med Chem Lett 17:697–
mately 4-fold lower for rodent CXCR3 than for primate CXCR3
701.
(Table 2). In addition, their pKb values against hCXCL10/
Arias DA, Navenot JM, Zhang WB, Broach J, and Peiper SC (2003) Constitutive
hCXCR3 and mCXCL10/mCXCR3 are comparable (Table 2;
Fig. 5, B and F), indicating the usefulness of these compounds
in mouse models. In contrast, the affinity of TAK-779 was equal
for CXCR3 of all tested species, again indicating that TAK-779
interacts with CXCR3 in another manner than the other small-
molecule antagonists. TAK-779, developed as a CCR5 antago-
nist, has been thoroughly investigated because of its potential
use as a human immunodeficiency virus-entry inhibitor. TAK-
779 shows high affinity for human CCR5 and to a lesser extent
for human CCR2b (Baba et al., 1999). Furthermore, TAK-779
was reported to bind mouse CCR5, as well as mouse CXCR3
(Gao et al., 2003). It is remarkable that TAK-779 shows a more
than 100-fold higher affinity for human CCR5 than for mouse
CCR5 (Baba et al., 1999; Gao et al., 2003). This species selec-
tivity of TAK-779 for CCR5 is not observed for human and
mouse CXCR3, which both bind TAK-779 with affinities around
1 ␮M (Table 2). It seems that TAK-779 has nanomolar affinity
for human CCR5 and micromolar affinity for human CXCR3,
whereas it has only micromolar affinity for mouse CCR5 and
mouse CXCR3. This should be kept in mind when analyzing
and extrapolating data from rodent models, because at a certain
effective concentration of TAK-779 different receptors will be
occupied in humans and mice. Although in mice the observed
effects would probably be mediated by a combination of block-
age of CCR5 and CXCR3, in humans the effects of TAK-779
would be mostly through inhibition of CCR5 with low nanomo-
lar affinity. The nonpeptidergic CXCR3 antagonists investi-
gated here only showed a 3- to 4-fold species selectivity for
CXCR3 of primates versus rodents in radioligand binding
studies, whereas no species difference was observed in func-
tional studies. When testing for selectivity against a panel
of GPCRs, VUF10132 inhibited CCR1-mediated signaling,
and it should therefore be optimized with respect to its affin-
ity and specificity for CXCR3. In contrast, we observed that
VUF10472, VUF10085, and VUF5834 are selective for
CXCR3. In line with our findings, VUF10472/NBI-74330 has
been reported previously not to affect chemotactic responses
by human H9 T-cell lymphoma cells in response to CXCL12
and CCL19 and not to interfere with calcium mobilization
induced by lysophosphatidic acid or radioligand binding to
several GPCRs (Heise et al., 2005).
activation of CCR5 and CCR2 induced by conformational changes in the conserved
TXP motif in transmembrane helix 2. J Biol Chem 278:36513–36521.
Axten JM, Foley JJ, Kingsbury WD, and Sarau HM (2003) inventors; SmithKline
Beechham Corporation, assignee. Imidazolium CXCR3 inhibitors. World patent
WO031101970. 2003 Dec 11.
Baba M, Nishimura O, Kanzaki N, Okamoto M, Sawada H, Iizawa Y, Shiraishi M,
Aramaki Y, Okonogi K, Ogawa Y, et al. (1999) A small-molecule, nonpeptide CCR5
antagonist with highly potent and selective anti-HIV-1 activity. Proc Natl Acad Sci
U S A 96:5698–5703.
Ballesteros JA and Weinstein H (1995) Integrated methods for the construction of
three dimensional models and computational probing of structure-function rela-
tions in G-protein coupled receptors. Methods Neurosci 25:366–428.
Burger M, Burger JA, Hoch RC, Oades Z, Takamori H, and Schraufstatter IU (1999)
Point mutation causing constitutive signaling of CXCR2 leads to transforming
activity similar to Kaposi’s sarcoma herpesvirus-G protein-coupled receptor. J Im-
munol 163:2017–2022.
Cheng Y and Prusoff WH (1973) Relationship between the inhibition constant (K1)
and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an
enzymatic reaction. Biochem Pharmacol 22:3099–3108.
Cole AG, Stroke IL, Brescia MR, Simhadri S, Zhang JJ, Hussain Z, Snider M,
Haskell C, Ribeiro S, Appell KC, et al. (2006) Identification and initial evaluation
of 4-N-aryl-[1,4]diazepane ureas as potent CXCR3 antagonists. Bioorg Med Chem
Lett 16:200–203.
Cole KE, Strick CA, Paradis TJ, Ogborne KT, Loetscher M, Gladue RP, Lin W, Boyd
JG, Moser B, Wood DE, et al. (1998) Interferon-inducible T cell alpha chemoat-
tractant (I-TAC): a novel non-ELR CXC chemokine with potent activity on acti-
vated T cells through selective high affinity binding to CXCR3. J Exp Med 187:
2009–2021.
Coward P, Chan SD, Wada HG, Humphries GM, and Conklin BR (1999) Chimeric G
proteins allow a high-throughput signaling assay of Gi-coupled receptors. Anal
Biochem 270:242–248.
Flier J, Boorsma DM, van Beek PJ, Nieboer C, Stoof TJ, Willemze R, and Tensen CP
(2001) Differential expression of CXCR3 targeting chemokines CXCL10, CXCL9,
and CXCL11 in different types of skin inflammation. J Pathol 194:398–405.
Gao P, Zhou XY, Yashiro-Ohtani Y, Yang YF, Sugimoto N, Ono S, Nakanishi T,
Obika S, Imanishi T, Egawa T, et al. (2003) The unique target specificity of a
nonpeptide chemokine receptor antagonist: selective blockade of two Th1 chemo-
kine receptors CCR5 and CXCR3. J Leukoc Biol 73:273–280.
Hancock WW, Lu B, Gao W, Csizmadia V, Faia K, King JA, Smiley ST, Ling M,
Gerard NP, and Gerard C (2000) Requirement of the chemokine receptor CXCR3
for acute allograft rejection. J Exp Med 192:1515–1520.
Heise CE, Pahuja A, Hudson SC, Mistry MS, Putnam AL, Gross MM, Gottlieb PA,
Wade WS, Kiankarimi M, Schwarz D, et al. (2005) Pharmacological characteriza-
tion of CXC chemokine receptor 3 ligands and a small molecule antagonist.
J Pharmacol Exp Ther 313:1263–1271.
Hensbergen PJ, Wijnands PG, Schreurs MW, Scheper RJ, Willemze R, and Tensen
CP (2005) The CXCR3 targeting chemokine CXCL11 has potent antitumor activity
in vivo involving attraction of CD8ϩ T lymphocytes but not inhibition of angiogen-
esis. J Immunother 28:343–351.
Ho HH, Ganeshalingam N, Rosenhouse-Dantsker A, Osman R, and Gershengorn MC
(2001) Charged residues at the intracellular boundary of transmembrane helices 2
and 3 independently affect constitutive activity of Kaposi’s sarcoma-associated
herpesvirus G protein-coupled receptor. J Biol Chem 276:1376–1382.
Johnson M, Li AR, Liu J, Fu Z, Zhu L, Miao S, Wang X, Xu Q, Huang A, Marcus A,
et al. (2007) Discovery and optimization of a series of quinazolinone-derived
antagonists of CXCR3. Bioorg Med Chem Lett 17:3339–3343.
Jopling LA, Watt GF, Fisher S, Birch H, Coggon S, and Christie MI (2007) Analysis
of the pharmacokinetic/pharmacodynamic relationship of a small molecule CXCR3
antagonist, NBI-74330, using a murine CXCR3 internalization assay. Br J Phar-
macol 152:1260–1271.
Kenakin T (1996) The classification of seven transmembrane receptors in recombi-
nant expression systems. Pharmacol Rev 48:413–463.
Lagane B, Ballet S, Planchenault T, Balabanian K, Le Poul E, Blanpain C, Perch-
erancier Y, Staropoli I, Vassart G, Oppermann M, et al. (2005) Mutation of the
DRY motif reveals different structural requirements for the CC chemokine recep-
tor 5-mediated signaling and receptor endocytosis. Mol Pharmacol 67:1966–1976.
Loetscher M, Gerber B, Loetscher P, Jones SA, Piali L, Clark-Lewis I, Baggiolini M,
In summary, we characterized three classes of small non-
peptidergic and noncompetitive CXCR3 antagonists at
CXCR3 of four different species, with VUF10472 being the
most potent compound at human, rhesus, rat, and mouse
CXCR3. The observed selectivity profile and relatively small

Nonpeptidergic Antagonists of the CXCR3 Chemokine Receptor
555
and Moser B (1996) Chemokine receptor specific for IP10 and mig: structure,
function, and expression in activated T-lymphocytes. J Exp Med 184:963–969.
Lu B, Humbles A, Bota D, Gerard C, Moser B, Soler D, Luster AD, and Gerard NP
(1999) Structure and function of the murine chemokine receptor CXCR3. Eur
J Immunol 29:3804–3812.
Luster AD and Leder P (1993) IP-10, a -C-X-C- chemokine, elicits a potent thymus-
dependent antitumor response in vivo. J Exp Med 178:1057–1065.
Mach F, Sauty A, Iarossi AS, Sukhova GK, Neote K, Libby P, and Luster AD (1999)
Differential expression of three T lymphocyte-activating CXC chemokines by hu-
man atheroma-associated cells. J Clin Invest 104:1041–1050.
Murphy PM, Baggiolini M, Charo IF, Hebert CA, Horuk R, Matsushima K, Miller
LH, Oppenheim JJ, and Power CA (2000) International union of pharmacology.
XXII. Nomenclature for chemokine receptors. Pharmacol Rev 52:145–176.
Pan J, Burdick MD, Belperio JA, Xue YY, Gerard C, Sharma S, Dubinett SM, and
Strieter RM (2006) CXCR3/CXCR3 ligand biological axis impairs RENCA tumor
growth by a mechanism of immunoangiostasis. J Immunol 176:1456–1464.
Parra S and Bond RA (2007) Inverse agonism: from curiosity to accepted dogma, but
is it clinically relevant? Curr Opin Pharmacol 7:146–150.
Qin S, Rottman JB, Myers P, Kassam N, Weinblatt M, Loetscher M, Koch AE, Moser
B, and Mackay CR (1998) The chemokine receptors CXCR3 and CCR5 mark
subsets of T cells associated with certain inflammatory reactions. J Clin Invest
101:746–754.
Storelli S, Verdijk P, Verzijl D, Timmerman H, van de Stolpe AC, Tensen CP, Smit
MJ, De Esch IJ, and Leurs R (2005) Synthesis and structure-activity relationship
of 3-phenyl-3H-quinazolin-4-one derivatives as CXCR3 chemokine receptor antag-
onists. Bioorg Med Chem Lett 15:2910–2913.
Storelli S, Verzijl D, Al-Badie J, Elders N, Bosch L, Timmerman H, Smit MJ, De
Esch IJ, and Leurs R (2007) Synthesis and structure-activity relationships of
3H-quinazolin-4-ones and 3H-pyrido[2,3-d]pyrimidin-4-ones as CXCR3 receptor
antagonists. Arch Pharm (Weinheim) 340:281–291.
Streblow DN, Kreklywich C, Yin Q, De La Melena VT, Corless CL, Smith PA,
Brakebill C, Cook JW, Vink C, Bruggeman CA, et al. (2003) Cytomegalovirus-
mediated upregulation of chemokine expression correlates with the acceleration of
chronic rejection in rat heart transplants. J Virol 77:2182–2194.
Tamaru M, Tominaga Y, Yatsunami K, and Narumi S (1998) Cloning of the murine
interferon-inducible protein 10 (IP-10) and its specific expression in lymphoid
organs. Biochem Biophys Res Commun 251:41–48.
Walser TC, Rifat S, Ma X, Kundu N, Ward C, Goloubeva O, Johnson MG, Medina JC,
Collins TL, and Fulton AM (2006) Antagonism of CXCR3 inhibits lung metastasis
in a murine model of metastatic breast cancer. Cancer Res 66:7701–7707.
Wang X, Li X, Schmidt DB, Foley JJ, Barone FC, Ames RS, and Sarau HM (2000)
Identification and molecular characterization of rat CXCR3: receptor expression
and interferon-inducible protein-10 binding are increased in focal stroke. Mol
Pharmacol 57:1190–1198.
Yang YF, Mukai T, Gao P, Yamaguchi N, Ono S, Iwaki H, Obika S, Imanishi T,
Tsujimura T, Hamaoka T, et al. (2002) A non-peptide CCR5 antagonist inhibits
collagen-induced arthritis by modulating T cell migration without affecting anti-
collagen T cell responses. Eur J Immunol 32:2124–2132.
Zhang WB, Navenot JM, Haribabu B, Tamamura H, Hiramatu K, Omagari A, Pei G,
Manfredi JP, Fujii N, Broach JR, et al. (2002) A point mutation that confers
constitutive activity to CXCR4 reveals that T140 is an inverse agonist and that
AMD3100 and ALX40-4C are weak partial agonists. J Biol Chem 277:24515–
24521.
Rabin RL, Park MK, Liao F, Swofford R, Stephany D, and Farber JM (1999)
Chemokine receptor responses on T cells are achieved through regulation of both
receptor expression and signaling. J Immunol 162:3840–3850.
Salomon I, Netzer N, Wildbaum G, Schif-Zuck S, Maor G, and Karin N (2002)
Targeting the function of IFN-gamma-inducible protein 10 suppresses ongoing
adjuvant arthritis. J Immunol 169:2685–2693.
Smit MJ, Verdijk P, van der Raaij-Helmer EM, Navis M, Hensbergen PJ, Leurs R,
and Tensen CP (2003) CXCR3-mediated chemotaxis of human T cells is regulated
by a Gi- and phospholipase C-dependent pathway and not via activation of MEK/
p44/p42 MAPK nor Akt/PI-3 kinase. Blood 102:1959–1965.
Sørensen TL, Tani M, Jensen J, Pierce V, Lucchinetti C, Folcik VA, Qin S, Rottman
J, Sellebjerg F, Strieter RM, et al. (1999) Expression of specific chemokines and
chemokine receptors in the central nervous system of multiple sclerosis patients.
J Clin Invest 103:807–815.
Address correspondence to: Dr. Rob Leurs, Leiden/Amsterdam Center of
Drug Research, Faculty of Science, Vrije Universiteit Amsterdam, De Boele-
laan 1083, 1081 HV Amsterdam, The Netherlands. E-mail: r.leurs@few.vu.nl