intracellular domains of cxcr3 that mediate cxcl9, cxcl10, and cxcl11 function

Intracellular Domains of CXCR3 that Mediate CXCL9, CXCL10, and CXCL11 Function

Richard A. Colvin, Gabriele S.V. Campanella, Jieti Sun, and Andrew D. Luster

Center for Immunology and Inflammatory Diseases Division of Rheumatology, Allergy, and ImmunologyMassachusetts General HospitalBoston, MA 02129

Running title: CXCR3 functional domains

Please address all correspondence to: Andrew D. Luster, MD, PhDCenter for Immunology and Inflammatory DiseasesMassachusetts General Hospital 149 Thirteenth Street room 8031Charlestown, MA 02129Phone: (617) 726-5710Fax: (617) 726-5651E-mail: [email protected]

Keywords: CXCR3, chemokine, chemotaxis, GPCR, internalization, signal transduction, CXCL10, CXCL9, CXCL11, IP-10, MIG, I-TAC, dynamin, arrestin

JBC Papers in Press. Published on May 17, 2004 as Manuscript M403595200

Copyright 2004 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on Novem

ber 23, 2018http://w

ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Abbreviations: PBS – phosphate buffered saline; BSA – bovine serum albumin; ERK –extracellular signal-related protein kinase; GPCR – G protein-coupled receptor; PCR –polymerase chain reaction; ANOVA – analysis of variance


- 2 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Summary

The chemokine receptor CXCR3 is a G protein-coupled receptor found predominantly

on T cells that is activated by 3 ligands: CXCL9 (Mig), CXCL10 (IP-10), and CXCL11

(I-TAC). Previously, we have found that of the three ligands, CXCL11 is the most potent

inducer of CXCR3 internalization and is the physiologic inducer of CXCR3

internalization after T cell contact with activated endothelial cells. We have therefore

hypothesized that these three ligands transduce different signals to CXCR3. In light of

this hypothesis, we sought to determine if regions of CXCR3 are differentially required

for CXCL9, CXCL10, and CXCL11 function. Here we identified two distinct domains

that contributed to CXCR3 internalization. The carboxyl-terminal domain and ß-

arrestin1 were predominantly required by CXCL9 and CXCL10, and the third

intracellular loop was predominantly required by CXCL11. Chemotaxis and calcium

mobilization induced by all three CXCR3 ligands were dependent on the CXCR3

carboxyl-terminus and the DRY sequence in the third trans-membrane domain. Our

findings demonstrate that distinct domains of CXCR3 mediate its functions and suggest

that the differential requirement of these domains contributes to the complexity of the

chemokine system.


- 3 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Introduction

Chemokines, or chemoattractant cytokines, are a family of small (8-10 kDa)

proteins that play an important role in the recruitment and activation of leukocytes (1).

By inducing the migration of leukocytes, they play a critical role in innate immunity as

well as the development of an adaptive immune response and the maintenance of chronic

inflammation. Chemokines induce their biological effects by binding to seven trans-

membrane-spanning G protein-coupled receptors (GPCRs) (1). Approximately 50

chemokines have been described that interact with approximately 16 GPCR chemokine

receptors, implying that there is redundancy in the chemokine system. However, this

apparent redundancy has not been supported by in vivo studies that have instead

demonstrated that individual chemokines that activate the same receptor can have unique

functions in vivo (2). This may be related to differential chemokine expression in vivo

and/or differential receptor activation by different chemokine ligands (3).

CXCR3 is expressed on the surface of a number of cell types, including activated

T cells and NK cells, and subsets of inflammatory dendritic cells, macrophages, and B

cells (4-6). CXCR3 is a pertussis toxin sensitive G protein-coupled receptor, indicating

that it is coupled to the Gi class of heterotrimeric G proteins (7). Activation of

chemokine receptors has been reported to induce a number of signaling pathways as well

as the activation of integrins, G protein related kinases (GRKs), and the binding of β↑

arrestin (8-12).


- 4 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

The down regulation of a receptor’s response is important in its overall function.

Chemokine receptors are down regulated in at least two ways. The most rapid way is the

uncoupling of the receptor from the G protein, which prevents further activation by

ligand, a process referred to as desensitization (13). Chemokine receptors are also down

regulated by internalization, or endocytosis, a process that removes the receptor from the

cell surface resulting in a more prolonged unresponsiveness to the ligand. The

internalized receptor may be degraded in lysozomes or recycled back to the cell surface

(14). In addition, it has recently been proposed that following internalization, GPCR

complexes can transduce unique signaling information (15,16). The internalization of

many GPCRs, including at least two chemokine receptors, CXCR4 and CCR5, is

mediated by the phosphorylation of carboxyl-terminal serine and threonine residues by

the GRKs. This phosphorylation event induces ß-arrestin2 binding to the cytoplasmic

tail, which targets the receptor to clathrin coated pits in a dynamin- dependent manner

(8-10).

Previously, it has been shown that chemokine receptors can be differentially

activated by different ligands. For example, CXCL8 (IL-8) activation of CXCR2 results

in higher levels of receptor internalization than CXCL7 (NAP2) (17). It is not clear,

however, whether CXCL8 and CXCL7 induce CXCR2 internalization through different

pathways or if the effect is due to differences in ligand potency. Additionally, synthetic

derivatives of CCL5 (RANTES) can induce CCR5 internalization and can mobilize


- 5 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

calcium, but cannot activate chemotaxis (18). These findings suggest that the activation

of a chemokine receptor by different ligands may lead to different signals.

Three ligands are known to activate CXCR3 – CXCL10 (IP-10), CXCL9 (Mig),

and CXCL11 (I-TAC) (4,19). Each of these ligands is induced by interferon-γ (IFN-γ)

and is produced in Th1-type immune responses (19-21). Although these three ligands

are all induced by IFN-γ, they appear to mediate distinct biological phenomena in vivo

(22-24). This may be related to differential expression of these ligands as has been seen

in cardiac and skin allograft rejection (22,24-29), atherosclerosis (30), host response to

infection (31), and inflammatory skin diseases (32). Alternatively, the different

biological outcomes may also be related to the differential activation of CXCR3 by

CXCL10, CXCL9, and CXCL11.

In this regard, our previous work has shown that CXCL11 is the physiologic

inducer of CXCR3 internalization following T cell contact with IFN-γ activated

endothelial cells, even though these cells produce greater amounts of CXCL10 and

CXCL9 than CXCL11 (33). We also found that CXCL11 was the most potent inducer of

CXCR3 internalization. Therefore, we speculated that these three ligands transduce

different signals to CXCR3 (33). To test this hypothesis, we mutated CXCR3 in the

intracellular domains to determine if CXCR3 is differentially activated by CXCL10,

CXCL9, and CXCL11. We specifically tested the effects of mutations on the receptor,

including ligand binding, receptor phosphorylation, and receptor internalization; effects


- 6 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

of the mutations on downstream signaling pathways, including Erk phosphorylation and

calcium mobilization; and finally, the effects of the mutations on chemotaxis.


- 7 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Experimental Procedures

Reagents.

Recombinant CXCL9, CXCL10, and CXCL11 were purchased from Peprotech (Rock

Hill, NJ). The PE-conjugated anti-CXCR3 antibody 1C6 was purchased from R&D

Systems (Minneapolis, MN). The bovine-pre-prolactin-FLAG plasmid was a gift of Dr.

Israel Charo (University of California at San Francisco). The dynamin and ß-arrestin

dominant negative constructs were gifts of Dr. Marc Caron (Duke University Medical

Center, Durham, NC).

Plasmids and mutagenesis.

All receptors used in this study are derived from human CXCR3. A cDNA encoding

CXCR3 was inserted into the Kpn1 and EcoR1 restriction sites in the multi-cloning site

of pcDNA3.1 (Invitrogen, Carlsbad, CA). The Tail(-)-CXCR3 truncation was

constructed by amplifying the cDNA encoding CXCR3 with the appropriate primers.

Point mutations were introduced into CXCR3 by using the Quikchange Mutagenesis Kit

by Stratagene (LaJolla, CA) and oligonucleotides encoding the specific changes.

Chimeric genes encoding the bovine-pre-prolactin signal sequence and the FLAG tag

followed by CXCR3 were constructed by ligating the pre-prolactin-FLAG genes to

CXCR3 at a Sal1 site. All constructs were confirmatory sequenced.


- 8 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Tissue culture.

300-19 is a pre-B cell leukemia cell line known to functionally express chemokine

receptors following stable transfection (34). At baseline, 300-19 cells functionally

express CXCR4 allowing for a positive control in assays of chemokine function (34).

300-19 cells were cultured in RPMI (Cellgro and Mediatech, Herndon, VA) with 10%

fetal calf serum, 100 U/ml penicillin and 100 µg/ml streptomycin (Mediatech), and 2 mM

L-glutamine (Mediatech) (complete RPMI). HEK293 cells were cultured in Dulbecco’s

modification of minimal essential medium (DMEM) (Mediatech), with 10% fetal calf

serum, 100 U/ml penicillin and 100 µg/ml streptomycin (Mediatech), and 2 mM L-

glutamine (Mediatech).

Stable transfection.

1x107 300-19 cells were incubated with 10 µg of linearized CXCR3/pcDNA3.1

constructs for 10 minutes on ice and electroporated using a Bio-RAD Gene Pulser II

(Bio-Rad, Hercules, CA) at 250 V and 975 microfarads in a 0.2 cm gap electrode cuvette

(Bio-Rad). Following the electroporation, the cells were grown in complete RPMI for 24

hours, and then placed into complete RPMI plus 80 µg/ml G418 (Mediatech) for

selection.

Enrichment of CXCR3 or CXCR3 mutant expressing cells.


- 9 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

5 x 106 transfected cells were stained with 3 µl of CXCR3 antibody 1C6 conjugated to

PE (R&D Systems, Minneapolis, MN) in10% goat serum in phosphate buffered saline

(PBS) for 30 minutes at 4° C. The stained cells were washed and then incubated with

microbeads coupled to an anti-PE antibody and CXCR3 expressing cells were positively

selected over a MACS LS column (Miltenyi Biotec, Auburn, CA) and cultured in

complete RPMI without G418.

Cell surface expression of CXCR3 and CXCR3 mutants.

Cultured cells were resuspended in 100 µl of flow cytometry buffer (PBS without calcium

and magnesium containing 1% bovine serum albumin (BSA) and 0.1% sodium azide)

and 10% goat serum. They were incubated for 5 minutes at room temperature. The anti-

CXCR3 antibody, 1C6, conjugated to PE (R&D Systems) was added to the cells, which

were then incubated at 4°C for 30 minutes. The cells were washed twice in PBS and

subsequently fixed by resuspension in PBS with 2% paraformaldehyde. Receptor cell

surface expression was measured on a FACSCalibur flow cytometer and the data

analyzed using CellQuest (BD Biosciences, San Jose, CA) or FlowJo (San Carlos, CA).

Receptor binding assays.

Binding assays were performed as previously reported (35). Briefly, 400,000 wild-type-


- 10 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

or mutant-CXCR3/300-19 cells were placed into 96-well tissue culture plates in a total

volume of 150 µl of binding buffer (0.5% BSA, 5 mM MgCl2, 1 mM CaCl2, 50 mM

HEPES pH 7.4). 0.04 nM of 125I labeled CXCL10 (New England Nuclear, Boston, MA)

or CXCL11 (Amersham Biosciences, Piscataway, NJ) and increasing amounts of

unlabeled CXCL10 or CXCL11 (Peprotech, Rocky Hill, NJ) were added to the cells and

incubated for 90 minutes at room temperature with shaking. The cells were transferred to

96-well filter plates (Millipore, Billerica, MA) pre-soaked in 0.3% polyethyleneimine

and washed three times with 200 µl binding buffer supplemented with 0.5 M NaCl. The

plates were dried and the radioactivity was measured after the addition of scintillation

fluid in a Wallac Microbeta scintillation counter (Perkin Elmer Life Sciences, Boston,

MA).

Receptor internalization.

Wild-type-CXCR3/300-19 or mutant-CXCR3/300-19 cells (250,000) were incubated

with various concentrations of CXCL10, CXCL11, or CXCL9 for 15 or 30 minutes as

indicated at 37°C. At the end of the incubations, ice-cold flow cytometry buffer was

added and cells were analyzed for cell surface expression of CXCR3 using the PE

conjugated CXCR3 antibody 1C6 as above.

Chemotaxis.


- 11 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Chemotaxis assays of 300-19 cells were performed in 96 well Neuroprobe chemotaxis

chambers with 5 µM pore size polycarbonate membranes (Neuroprobe, Gaithersburg,

MD). 31 µl of RPMI containing 1% BSA and chemokines were placed in the bottom

chamber of the device per the manufacturers directions. 25,000 cells were layered onto

the top of the membrane in RPMI containing 1% BSA. The chambers were then

incubated at 37°C for 5 hours. After washing the top of the filter with deionized water,

the chambers were subjected to centrifugation at 1,500 rpm for 5 minutes. The filters

were removed and media was aspirated. The chambers were then frozen at -80°C for at

least one hour. 20 µl of CyQuant dye mix (Molecular Probes, Eugene, OR) was added to

each well of the Neuroprobe chamber. Following a 2-hour incubation period, the

fluorescence was measured in a CytoFluor fluorescent plate reader (Applied Biosystems,

Foster City, CA). For each experiment, a cellular titration curve was completed to make

sure that the fluorescence reading was in the linear range of the CyQuant dye and the

background fluorescence was subtracted from the readings for each sample. The

chemotactic index was determined by dividing the fluorescence at each chemokine

concentration by the fluorescence when no chemokine was added. The CyQuant results

have been compared favorably with direct counting of the migrated cells. Chemotaxis

data was analyzed using analysis of the variance (ANOVA) allowing the comparison of

the chemotactic curves rather than individual points.


- 12 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Calcium flux.

5x106 wild-type- or mutant-CXCR3/300-19 cells were resuspended in 2 ml of RPMI

with 1% bovine serum albumin. 15 µl of Fura-2 (Molecular Probes, Eugene, OR) was

added and the cells were incubated at 37°C for 20 minutes. The cells were washed twice

in PBS and resuspended in 2 ml of calcium flux buffer (145 mM NaCl, 4 mM KCl, 1 mM

NaHPO4, 1.8 mM CaCl2, 25 mM HEPES, 0.8 mM MgCl2 and 22 mM glucose).

Fluorescence readings were measured at 37° C in a DeltaRAM fluorimeter (Photon

Technology International, Lawrenceville, NJ). Intracellular calcium concentrations were

recorded as the excitation fluorescence intensity emitted at 510 nM in response to

sequential excitation at 340 nm and 380 nm and are presented as the relative ratio of

fluorescence at 340/380.

Receptor Phosphorylation.

HEK293 cells transiently transfected with pcDNA3.1 with wild-type-CXCR3 or Tail(-

)-CXCR3 that had been constructed to include a bovine pre-prolactin signal sequence

followed by DNA sequences encoding the FLAG epitope tag. Forty-eight hours

following transfection, the media was changed to phosphate free DMEM with 1% BSA.

The cells were incubated for 10 minutes and 150 mCi of 32P inorganic phosphate (NEN

Life Sciences, Boston, MA) was added to each well. Cells were incubated for 3 hours


- 13 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

followed by stimulation by 100 nM CXCL10 or CXCL11. Extracts were made and

immunoprecipitations were performed as described previously (36). Briefly, cells were

lysed in 0.15 M NaCl, 20 mM HEPES pH 7.4, 5 mM EDTA, 3 mM EGTA, 4 mg/ml

dodecyl-ß-maltoside, 0.2 mg/ml cholestoryl hemisuccinate, plus protease inhibitors and

phosphatase inhibitors. Following lysis, extracts were cleared by centrifugation, pre-

cleared using protein-G-sepharose beads, and proteins were immunoprecipitated using

M2-anti-FLAG antibody preconjugated to agarose beads (Sigma, St. Louis, MO).

Following overnight immunoprecipitation, beads were washed five times using

radioimmunoassay precipitation (RIPA) buffer (50 mM tris-HCl pH 7.4, 1% NP-40,

0.25% Na-deoxycholate, 150 mM NaCl, 1 mM EDTA) and proteins were resolved on a

10% SDS-PAGE gel. Gels were dried on a vacuum gel-drier and subsequently exposed

to autoradiography film.

Transient Transfection.

HEK293 cells were plated onto 6 well tissue culture plates in Dulbecco’s modified

essential medium (DMEM). When the cells were approximately 50% confluent they were

transfected with pcDNA3.1 with the indicated CXCR3 gene inserted using Fugene 6

(Roche Diagnostics, Indianapolis, IN) according to the manufacturer’s directions. Total

transfected DNA was kept constant using pcDNA3.1 without an insert.


- 14 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Results

Truncation of the carboxyl-terminus. A mutant of CXCR3, called Tail(-)-CXCR3,

was generated to determine the role of the carboxyl-terminus in CXCR3 function (Table

1). Eight amino acids of the predicted carboxyl-terminus were retained in order to ensure

surface expression, while removing any putative signaling elements. A population of

300-19 cells that stably and highly expressed Tail(-)-CXCR3 was positively selected.

Tail(-)-CXCR3 was expressed on the cell surface of 300-19 cells at similar levels to

wild-type-CXCR3 (Table 2). This suggested that CXCR3 is unlikely to require

palmitoylation of carboxyl-terminal cysteines to achieve cell surface expression as has

been seen for CCR5 (37). In our binding assays, we could not determine an IC50 value

for Tail(-)-CXCR3/300-19 cells to CXCL10 and CXCL11 despite the detection of some

chemokine binding in the absence of competitor (Table 2). CXCL9 binding assays were

not performed as there is no commercially available 125I labeled preparation of this

chemokine.

CXCL10, CXCL9, and CXCL11 were tested for their ability to induce internalization

of wild-type-CXCR3 and Tail(-)-CXCR3 (Figure 1). The ability of CXCL10 and

CXCL9 to induce internalization of Tail(-)-CXCR3 was dramatically reduced compared

to wild-type-CXCR3 (p<0.01). In contrast, CXCL11 induced internalization of Tail(-)-


- 15 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

CXCR3 as efficiently as wild-type-CXCR3. In chemotaxis assays, Tail(-)-

CXCR3/300-19 cells did not migrate to CXCL10 or CXCL9, and migration to CXCL11

was dramatically reduced (Figure 1). Upon stimulation with CXCL10, CXCL9, or

CXCL11 (10 nM) there was no evidence of calcium mobilization in Tail(-)-

CXCR3/300-19 cells (Figure 2) or Erk-phosphorylation (data not shown). These data

demonstrate that the carboxyl-terminus of CXCR3 was essential for CXCL10-,

CXCL9-, and CXCL11-induced chemotaxis, calcium mobilization, and Erk

phosphorylation, but was not essential for CXCL11-induced receptor internalization.

CXCR3 Phosphorylation. Carboxyl-terminal serines and threonines have been

implicated in regulating GPCR desensitization and internalization, including CXCR4 and

CCR5 (38-40). Upon receptor activation, these carboxyl-terminal serines and threonines

are phosphorylated by GRKs, which allows ß-arrestin binding and can lead to both G

protein uncoupling and internalization by targeting receptors to clathrin coated pits

(8,41). Wild-type-CXCR3 and Tail(-)-CXCR3 were immunoprecipitated from extracts

made from HEK293 cells transiently transfected with plasmids encoding FLAG-tagged

versions of these receptors (Figure 3). Forty-eight hours after transfection, the cells were

metabolically labeled with 32P inorganic phosphate and stimulated with 100 nM

CXCL10 or CXCL11. A low level of baseline phosphorylation was seen in 3 of 5

experiments for wild-type-CXCR3. Wild-type-CXCR3 phosphorylation was enhanced


- 16 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

after five minutes of stimulation by CXCL10 and CXCL11. However, neither CXCL10

nor CXCL11 stimulation enhanced the phosphorylation of Tail(-)-CXCR3 and

constitutive phosphorylation of Tail(-)-CXCR3 was not seen. Five weeks after exposure

to film there were no bands present at the appropriate size in lanes from extracts of Tail(-

)-CXCR3/300-19 cells or untransfected 300-19 cells either before or after stimulation

with CXCL10 or CXCL11. In order to determine if the phosphorylated protein was

CXCR3, Western blots were performed using the anti-CXCR3 antibody 1C6. These

experiments revealed that the bands on western blots from CXCR3 transfected HEK293

cells migrated identically to the radiolabelled immunoprecipitated proteins on the SDS-

PAGE gels (Figure 3). Untransfected HEK293 cells revealed no signal at this location

(Figure 3). Interestingly, CXCR3 migrated at almost twice its expected size on SDS-

PAGE gels suggesting the presence of post-translational modifications. Additionally,

immunoprecipitation of extracts from CXCL10- and CXCL11-stimulated and

unstimulated wild-type-CXCR3/300-19 cells with a control anti-CCR7 antibody did

not detect bands at this size (data not shown). These data demonstrate that both CXCL10

and CXCL11 induced the phosphorylation of CXCR3, and that receptor phosphorylation

may be required for CXCL10-induced internalization.

Carboxyl-terminal serine and threonine substitution mutant. To further dissect the

role of the CXCR3 carboxyl-terminus in ligand mediated signaling, a serine/threonine


- 17 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

mutant, ST(-)-CXCR3, was constructed that contained alanine substitutions of the serine

and threonine residues in the carboxyl-terminus (Table 1). A population of cells that

stably and highly expressed ST(-)-CXCR3 was positively selected. Cell surface

expression of ST(-)-CXCR3 was similar to wild-type-CXCR3 on 300-19 cells and

ST(-)-CXCR3/300-19 cells bound CXCL10 and CXCL11 similarly to wild-type-

CXCR3/300-19 cells as determined by IC50 values (Table 2).

CXCL10 and CXCL9 induced less internalization of ST(-)-CXCR3 than wild-type-

CXCR3 (p<0.01), demonstrating that the carboxyl-terminal serines and threonines

contributed to maximal CXCL10- and CXCL9-mediated internalization (Figure 4). The

alteration of the CXCR3 carboxyl-terminal serines and threonines, however, had no

effect on CXCL11-induced internalization (Figure 4). These data are consistent with the

Tail(-)-CXCR3 data and suggest that CXCL11 induced an internalization pathway that

does not require the carboxyl-terminal serine and threonine residues. Of note, wild-

type-CXCR3 receptor internalization was similar after 15 minutes and 5 hours of

stimulation with CXCL10, CXCL9, and CXCL11 (data not shown). Peak calcium

mobilization was similar for wild-type-CXCR3/300-19 cells and ST(-)-CXCR3/300-

19 cells following CXCL10, CXCL9, and CXCL11 stimulation (Figure 2). However, the

free intracellular calcium concentration remained elevated in the ST(-)-CXCR3/300-19

cells, suggesting a prolonged signal following chemokine stimulation (Figure 2). ST(-)-

CXCR3 was desensitized to repeated doses of CXCL10 and CXCL11 similarly to wild-


- 18 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

type-CXCR3 (data not shown).

Stimulation of ST(-)-CXCR3/300-19 cells resulted in enhanced chemotaxis to

CXCL10, CXCL9, and CXCL11 when compared to wild-type-CXCR3/300-19 cells

(Figure 4). The dose-response for chemokine-induced migration was similar for both

ST(-)-CXCR3/300-19 cells and wild-type-CXCR3/300-19 cells. However, the number

of migrating cells was significantly greater for ST(-)-CXCR3/300-19 cells in response

to CXCL10, CXCL9, and CXCL11, but not to the control CXCL12 (Figure 4 and data

not shown). These differences were statistically significant (p<0.01). Although

statistically significant, the differential migration to CXCL11 was less pronounced than

that to CXCL10 or CXCL9 (Figure 4). Of note, 300-19 cells expressing CXCR3 with

subsets of the serine and threonine substitutions, (e.g. S349AS350AS351AS355AS356A,

S349AS350AS351AS364A, or S349AS350AS351AS358AT360AS361A), also

demonstrated significantly higher chemotactic activity compared to wild-type-CXCR3

(data not shown).

Effect of dynamin and ß-arrestin on CXCR3 internalization. The requirement of the

carboxyl-terminus serines and threonine for CXCL10- and CXCL9-induced CXCR3

internalization suggested that ß-arrestin might play a role in CXCR3 internalization.

Dynamin is a small GTPase that is important for ß-arrestin-dependent internalization

(15). In order to evaluate whether both CXCL10- and CXCL11-induced CXCR3


- 19 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

internalization required dynamin and ß-arrestin1 or ß-arrestin2, plasmids encoding

dominant negative dynamin (K44A), ß-arrestin1 (V53D), or ß-arrestin2 (V54D) (42)

were transiently co-transfected with CXCR3 into HEK293 cells. Forty-eight hours after

transfection, the cells were harvested and the ability of CXCL10 and CXCL11 to induce

CXCR3 internalization was tested (Figure 5). CXCR3 internalization was always about

50% less in HEK293 cells than in 300-19 cells or primary lymphocytes; however, the

dynamin K44A and ß-arrestin1 V53D dominant negative mutants significantly reduced

CXCL10-induced CXCR3 internalization (p<0.01 and p<0.05), but had little effect on

CXCL11-induced CXCR3 internalization. The ß-arrestin2 V54D dominant negative had

no significant effect on CXCL10- or CXCL11-induced internalization. These results

suggest that CXCL10 required dynamin and ß-arrestin1 to induce CXCR3 internalization

in HEK293 cells, while CXCL11-induced CXCR3 internalization proceeded in a

dynamin- and ß-arrestin-independent manner.

CXCR3 intracellular loop 3. In order to define a region of CXCR3 that mediated

CXCL11-induced internalization, the third intracellular loop of CXCR3 was replaced

with the third intracellular loop of CXCR1 and called i3-CXCR1-CXCR3. i3-CXCR1-

CXCR3 surface expression on 300-19 cells was slightly elevated compared to wild-type

CXCR3, although CXCL10 and CXCL11 binding to i3-CXCR1-CXCR3/300-19 cells

was similar compared to wild-type-CXCR3/300-19 cells (Table 2). CXCL11-induced


- 20 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

internalization of i3-CXCR1-CXCR3 was significantly reduced in 300-19 cells

compared to wild-type-CXCR3, while CXCL10- and CXCL9-induced internalization

was not significantly different (Figure 6). These data suggest that the third intracellular

loop of CXCR3 was required for maximal internalization induced by CXCL11 but not

CXCL9 or CXCL10. CXCL10-, CXCL9-, and CXCL11-induced i3-CXCR1-

CXCR3/300-19 chemotaxis was slightly diminished compared to wild-type-

CXCR3/300-19 (Figure 6). Calcium mobilization following stimulation with CXCL10,

CXCL9, and CXCL11 was slightly reduced compared to wild-type-CXCR3 (Figure 2).

Additionally, following stimulation with CXCL11, i3-CXCR1-CXCR3/300-19 cells

were desensitized to a second dose of CXCL11 similarly to wild-type-CXCR3/300-19

cells (data not shown).

DRY site mutation: R149N. It has previously been shown that Gαi dependent

chemokine receptors require an aspartate-arginine-tyrosine (DRY) sequence in the third

trans-membrane domain for inducing chemotaxis (43). The DRY sequence has also been

shown to be important for the internalization of some G-protein coupled receptors

following activation (44).

In order to further define the role of the CXCR3 DRY sequence in ligand-

induced internalization, chemotaxis, and calcium mobilization, a receptor mutant was

generated that changed the DRY sequence to aspartate-asparagine-tyrosine (DNY).


- 21 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

(Table 1).

R149N-CXCR3 was expressed at slightly reduced levels compared to wild-type

CXCR3 on 300-19 cells (Table 2). R149N-CXCR3/300-19 cells bound CXCL10 with a

similar IC50 value compared to wild-type-CXCR3/300-19 cells (Table 2).

Internalization of this receptor following stimulation with CXCL10, CXCL9, or CXCL11

was similar to internalization of wild-type-CXCR3, suggesting that this sequence played

no role in CXCL10-, CXCL9-, or CXCL11-induced CXCR3 internalization (Figure 7).

However, CXCL10-, CXCL9-, and CXCL11-induced chemotaxis was reduced to

background levels, suggesting that this sequence was essential for CXCR3 mediated

chemotaxis (Figure 7). Similar to chemotaxis, CXCL10-, CXCL9-, and CXCL11-

induced calcium mobilization and Erk-phosphorylation to were abolished by this

substitution (Figure 2 and data not shown).

Additional intracellular domain mutants. Additional intracellular loop mutants

were constructed to determine the role of the first intracellular loop of CXCR3 in

chemotaxis and receptor internalization following ligand stimulation. Alanines were

substituted for the serines and threonines in the first intracellular loop using site-directed

mutagenesis. S80A-CXCR3, T83A-CXCR3, and T90A-CXCR3, had no effect on

receptor expression, induction of chemotaxis, or internalization by CXCL10 and

CXCL11 (data not shown).


- 22 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

The sequence LLLRL332-336 in the carboxyl-terminus is similar to the adaptin-

2 binding site that has been demonstrated to be important for CXCR2 internalization (45).

Alanines were substituted for leucines 332-334. LLL332-334AAA-CXCR3 was

expressed at levels comparable to wild-type-CXCR3 in 300-19 cells (Table 2).

LLL332-334AAA-CXCR3/300-19 cells bound CXCL10 and CXCL11 similarly to

wild-type-CXCR3/300-19 cells (Table 2). Additionally, CXCL10-, CXCL9-, and

CXCL11-induced receptor internalization (Figure 7) and calcium mobilization for

LLL332-334AAA-CXCR3/300-19 cells was similar to wild-type-CXCR3/300-19

cells (Figure 2). However, CXCL10-, CXCL9-, and CXCL11-induced chemotaxis was

reduced in LLL332-334AAA-CXCR3/300-19 cells compared to wild-type-

CXCR3/300-19 cells (Figure 7). These data demonstrate that the LLLRL motif in the

carboxyl terminus of CXCR3 participates in regulating chemotaxis but not receptor

internalization or calcium mobilization.


- 23 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Discussion

One of the complexities of the chemokine system is that different ligands for the

same receptor can induce different biological effects. In this study, we have explored the

structural basis for these effects in the CXCR3 receptor ligand system.

We have now shown that different domains of CXCR3 are required for CXCL10-

, CXCL9-, and CXCL11-induced function. In summary, the DRY site was essential for

CXCL10-, CXCL9-, and CXCL11-induced chemotaxis, calcium mobilization, and Erk

phosphorylation but not for CXCR3 internalization; the CXCR3 carboxyl-terminus was

essential for CXCL10-, CXCL9-, and CXCL11-induced chemotaxis, calcium

mobilization, and Erk phosphorylation; the CXCR3 carboxyl-terminus and ß-arrestin1

were required for CXCL10-induced receptor internalization, while the more effective

CXCL11-induced internalization was independent of the carboxyl-terminus and ß-

arrestin1; the third intracytoplasmic loop was required for maximal CXCL11-induced

internalization; and residues LLL332-334 played a role in mediating CXCR3 ligand-

induced chemotaxis (Figure 8).

Carboxyl-terminus. CXCR3 truncated at the carboxyl-terminus clearly showed

that this region was essential for maximal chemokine binding, chemotaxis, calcium

mobilization, and MAPK activation, but was not important for receptor expression.


- 24 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Although ligand binding to Tail(-)-CXCR3 was severely diminished, signaling through

this receptor was not completely abrogated, as demonstrated by the ability of CXCL11 to

induce efficient internalization of this receptor. CXCL10 and CXCL9 induced only

minimal internalization of Tail(-)-CXCR3, demonstrating that these chemokines

differentially activate CXCR3 leading to distinct biological outcomes. One possibility is

that CXCL9 and CXCL10 did not bind Tail(-)-CXCR3 while CXCL11 did. It is possible

that Tail(-)-CXCR3 is predominantly in the uncoupled conformation that has been

shown to bind CXCL11 but not CXCL9 or CXCL10 (46). Binding data were not helpful

to differentiate these possibilities, as there was only minimal binding of CXCL10 and

CXCL11 to Tail(-)-CXCR3. Previously it had been shown that chemokine receptors can

be activated by ligands that show no detectable binding (47,48). It has been speculated

that this results from a chemokine-chemokine receptor interaction with an off rate that is

faster than the sensitivity of competitive binding assays (47).

Carboxyl-terminal serines and threonine. Mutation of the CXCR3 carboxyl-

terminal serines and threonine resulted in a receptor (ST(-)-CXCR3) that was

internalized less than wild-type-CXCR3 when stimulated with CXCL9 and CXCL10.

This mutant also exhibited enhanced migration to CXCL10, CXCL9, and CXCL11,

which may have resulted from the prolonged signaling following chemokine stimulation

as seen in the calcium mobilization data.

Third intracellular loop. The substitution of the third intracellular loop of CXCR3


- 25 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

with that of CXCR1 (i3-CXCR1-CXCR3) resulted in a receptor that was internalized

less than wild-type-CXCR3 upon stimulation with CXCL11 but not CXCL9 and

CXCL10, suggesting that the third intracellular loop of CXCR3 was required for maximal

CXCL11-induced internalization and that this region of CXCR3 was differentially

activated by the CXCR3 ligands.

DRY motif. We found the DRY sequence in the third trans-membrane domain of

CXCR3 to be important for chemotaxis, calcium mobilization, and Erk phosphorylation,

but not for receptor internalization. This sequence has been previously shown to be

important for signaling by Gαi-coupled chemokine receptors (49). Consistent with

pertussis toxin data, our data also support that the DRY sequence was not necessary for

CXCR3 internalization (33).

LLLRL motif. It has been reported that CXCL8-induced CXCR2 internalization

can proceed through a carboxyl-terminal serine- and threonine-independent manner that

is dependent on the adaptin-2 binding motif (LLKIL) in the CXCR2 carboxyl terminus

(17,45). The CXCR3 carboxyl-terminus contains the similar sequence LLLRL. This

motif cannot be responsible for CXCL11-induced CXCR3 internalization as its absence

did not effect Tail(-)-CXCR3 internalization. Consistent with this, substitution of the

LLL332-334 residues with alanine residues had no effect on CXCL11-induced CXCR3

internalization. This substitution also had no effect on CXCL9- and CXCL10-induced

CXCR3 internalization. However, this sequence appears to play a role in CXCR3-


- 26 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

mediated chemotaxis as CXCL10, CXCL9, and CXCL11-induced less chemotaxis in

LLL332-334AAA-CXCR3/300-19 cells compared to wild-type-CXCR3/300-19 cells.

CXCR3 function and ß-arrestins. Many G protein-coupled receptors are

internalized after activation through a process that involves G protein-related kinase

phosphorylation of carboxyl-terminal serine and threonine residues. This leads to ß-

arrestin binding, which uncouples the receptor from the G proteins, and subsequent

dynamin- and clathrin-mediated internalization (50).

Our data demonstrated that the dynamin and ß-arrestin1 dominant negatives

impaired CXCL10-, but not CXCL11-induced internalization in transiently transfected

HEK293 cells. Although CXCR3 internalization was not as robust in HEK293 cells as in

300-19 cells, the data show a clear difference between CXCL10- and CXCL11-induced

CXCR3 internalization. These data are consistent with our findings that CXCL10 and

CXCL11 require two distinct domains of CXCR3. It is noteworthy that the ß-arrestin1

and not the ß-arrestin2 dominant negative protein affected CXCR3 internalization, as

previously it was shown that ß-arrestin2 was required for CXCL12-induced CXCR4

internalization (39,51).

In summary, our results indicate that the ligands of CXCR3 can preferentially

activate distinct internalization pathways. CXCL10 and CXCL9 predominantly induce a


- 27 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

carboxyl-terminal dependent pathway, whereas CXCL11 predominantly induces a

carboxyl-terminal independent pathway. Since we have previously shown that CXCL11

is the more potent and physiological inducer of CXCR3 internalization, the tail-

independent mechanism is potentially more important (33). This is reminiscent of the

tail- and dynamin-independent process that has been described for CXCR2

internalization (52). The relative role of the internalization pathways in the in vivo

function of CXCL10, CXCL9, and CXCL11 as well as the role of the CXCR3 carboxyl-

terminus in mediating chemotaxis is of great interest in understanding CXCR3 function

in inflammatory processes.


- 28 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Acknowledgements

This work was supported by NIH grant K08 AI50147 to RAC and NIH grants PO1

DK50305 and R01 CA69212 to ADL. We would like to thank Josephine Leung for

technical assistance, Dr. William Hipkin from Schering-Plough for technical advice and

Dr. Robert Gerszten for helpful discussions.


- 29 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

References1. Luster, A. D. (1998) N Engl J Med 338, 436-4452. Proudfoot, A. E. (2002) Nat Rev Immunol 2, 106-1153. Elsner, J., Mack, M., Bruhl, H., Dulkys, Y., Kimmig, D., Simmons, G., Clapham,

P. R., Schlondorff, D., Kapp, A., Wells, T. N., and Proudfoot, A. E. (2000) J Biol Chem 275, 7787-7794

4. Loetscher, M., Gerber, B., Loetscher, P., Jones, S. A., Piali, L., Clark-Lewis, I., Baggiolini, M., and Moser, B. (1996) J Exp Med 184, 963-969

5. Loetscher, M., Loetscher, P., Brass, N., Meese, E., and Moser, B. (1998) Eur J Immunol 28, 3696-3705

6. Qin, S., Rottman, J. B., Myers, P., Kassam, N., Weinblatt, M., Loetscher, M., Koch, A. E., Moser, B., and Mackay, C. R. (1998) J Clin Invest 101, 746-754

7. Neptune, E. R., and Bourne, H. R. (1997) Proceedings of the National Academy of Sciences of the United States of America 94, 14489-14494

8. Aramori, I., Ferguson, S. S., Bieniasz, P. D., Zhang, J., Cullen, B., and Caron, M. G. (1997) EMBO J 16, 4606-4616

9. Sun, Y., Cheng, Z., Ma, L., and Pei, G. (2002) J Biol Chem10. Fong, A. M., Premont, R. T., Richardson, R. M., Yu, Y. R., Lefkowitz, R. J., and

Patel, D. D. (2002) Proc Natl Acad Sci U S A 99, 7478-748311. Wong, M., Uddin, S., Majchrzak, B., Huynh, T., Proudfoot, A. E., Platanias, L.

C., and Fish, E. N. (2001) J Biol Chem 276, 11427-1143112. Murdoch, C., and Finn, A. (2000) Blood 95, 3032-304313. Loetscher, P., Seitz, M., Clark-Lewis, I., Baggiolini, M., and Moser, B. (1994)

FASEB J 8, 1055-106014. Moro, O., Lameh, J., and Sadee, W. (1993) J Biol Chem 268, 6862-686515. Daaka, Y., Luttrell, L. M., Ahn, S., Della Rocca, G. J., Ferguson, S. S., Caron, M.

G., and Lefkowitz, R. J. (1998) J Biol Chem 273, 685-68816. Luttrell, L. M., Ferguson, S. S., Daaka, Y., Miller, W. E., Maudsley, S., Della

Rocca, G. J., Lin, F., Kawakatsu, H., Owada, K., Luttrell, D. K., Caron, M. G., and Lefkowitz, R. J. (1999) Science 283, 655-661

17. Feniger-Barish, R., Ran, M., Zaslaver, A., and Ben-Baruch, A. (1999) Cytokine11, 996-1009

18. Vila-Coro, A. J., Mellado, M., Martin de Ana, A., Martinez, A. C., and Rodriguez-Frade, J. M. (1999) J Immunol 163, 3037-3044

19. Cole, K. E., Strick, C. A., Paradis, T. J., Ogborne, K. T., Loetscher, M., Gladue, R. P., Lin, W., Boyd, J. G., Moser, B., Wood, D. E., Sahagan, B. G., and Neote, K. (1998) J Exp Med 187, 2009-2021


- 30 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

20. Farber, J. M. (1990) Proc Natl Acad Sci U S A 87, 5238-524221. Luster, A. D., and Ravetch, J. V. (1987) J Exp Med 166, 1084-109722. Hancock, W. W., Gao, W., Csizmadia, V., Faia, K. L., Shemmeri, N., and Luster,

A. D. (2001) J Exp Med 193, 975-98023. Khan, I. A., MacLean, J. A., Lee, F. S., Casciotti, L., DeHaan, E., Schwartzman,

J. D., and Luster, A. D. (2000) Immunity 12, 483-49424. Zhang, Z., Kaptanoglu, L., Haddad, W., Ivancic, D., Alnadjim, Z., Hurst, S.,

Tishler, D., Luster, A. D., Barrett, T. A., and Fryer, J. (2002) J Immunol 168,3205-3212

25. Kobayashi, H., Novick, A. C., Toma, H., and Fairchild, R. L. (2002)Transplantation 74, 387-395

26. Dufour, J. H., Dziejman, M., Liu, M. T., Leung, J. H., Lane, T. E., and Luster, A. D. (2002) J Immunol 168, 3195-3204

27. el-Sawy, T., Fahmy, N. M., and Fairchild, R. L. (2002) Curr Opin Immunol 14,562-568

28. Hancock, W. W., Lu, B., Gao, W., Csizmadia, V., Faia, K., King, J. A., Smiley, S. T., Ling, M., Gerard, N. P., and Gerard, C. (2000) J Exp Med 192, 1515-1520

29. Miura, M., Morita, K., Kobayashi, H., Hamilton, T. A., Burdick, M. D., Strieter, R. M., and Fairchild, R. L. (2001) J Immunol 167, 3494-3504

30. Mach, F., Sauty, A., Iarossi, A. S., Sukhova, G. K., Neote, K., Libby, P., and Luster, A. D. (1999) J Clin Invest 104, 1041-1050

31. Amichay, D., Gazzinelli, R. T., Karupiah, G., Moench, T. R., Sher, A., and Farber, J. M. (1996) J Immunol 157, 4511-4520

32. Flier, J., Boorsma, D. M., van Beek, P. J., Nieboer, C., Stoof, T. J., Willemze, R., and Tensen, C. P. (2001) J Pathol 194, 398-405

33. Sauty, A., Colvin, R. A., Wagner, L., Rochat, S., Spertini, F., and Luster, A. D. (2001) J Immunol 167, 7084-7093

34. Arai, H., Tsou, C. L., and Charo, I. F. (1997) Proc Natl Acad Sci U S A 94,14495-14499

35. Campanella, G. S., Lee, E. M., Sun, J., and Luster, A. D. (2003) J Biol Chem 278,17066-17074

36. January, B., Seibold, A., Whaley, B., Hipkin, R. W., Lin, D., Schonbrunn, A., Barber, R., and Clark, R. B. (1997) J Biol Chem 272, 23871-23879

37. Blanpain, C., Wittamer, V., Vanderwinden, J. M., Boom, A., Renneboog, B., Lee, B., Le Poul, E., El Asmar, L., Govaerts, C., Vassart, G., Doms, R. W., and Parmentier, M. (2001) J Biol Chem 276, 23795-23804

38. Roland, J., Murphy, B. J., Ahr, B., Robert-Hebmann, V., Delauzun, V., Nye, K. E., Devaux, C., and Biard-Piechaczyk, M. (2003) Blood 101, 399-406

39. Kraft, K., Olbrich, H., Majoul, I., Mack, M., Proudfoot, A., and Oppermann, M. (2001) J Biol Chem 276, 34408-34418


- 31 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

40. Oppermann, M., Mack, M., Proudfoot, A. E., and Olbrich, H. (1999) J Biol Chem274, 8875-8885

41. Haribabu, B., Richardson, R. M., Fisher, I., Sozzani, S., Peiper, S. C., Horuk, R., Ali, H., and Snyderman, R. (1997) J Biol Chem 272, 28726-28731

42. Herskovits, J. S., Burgess, C. C., Obar, R. A., and Vallee, R. B. (1993) J Cell Biol122, 565-578

43. Arai, H., and Charo, I. F. (1996) J Biol Chem 271, 21814-2181944. Arora, K. K., Cheng, Z., and Catt, K. J. (1997) Mol Endocrinol 11, 1203-121245. Fan, G. H., Yang, W., Wang, X. J., Qian, Q., and Richmond, A. (2001)

Biochemistry 40, 791-80046. Cox, M. A., Jenh, C. H., Gonsiorek, W., Fine, J., Narula, S. K., Zavodny, P. J.,

and Hipkin, R. W. (2001) Mol Pharmacol 59, 707-71547. Ahuja, S. K., Lee, J. C., and Murphy, P. M. (1996) J Biol Chem 271, 225-23248. Xanthou, G., Williams, T. J., and Pease, J. E. (2003) Eur J Immunol 33, 2927-

293649. Neer, E. J. (1995) Cell 80, 249-25750. Ferguson, S. S., Downey, W. E., 3rd, Colapietro, A. M., Barak, L. S., Menard, L.,

and Caron, M. G. (1996) Science 271, 363-36651. Cheng, Z. J., Zhao, J., Sun, Y., Hu, W., Wu, Y. L., Cen, B., Wu, G. X., and Pei,

G. (2000) J Biol Chem 275, 2479-248552. Richardson, R. M., Marjoram, R. J., Barak, L. S., and Snyderman, R. (2003) J

Immunol 170, 2904-2911


- 32 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Figure Legends.

Figure 1. Receptor internalization and chemotaxis of Tail(-)-CXCR3. Wild-type-

CXCR3/300-19 cells and Tail(-)-CXCR3/300-19 cells were activated with the CXCR3

ligands CXCL10, CXCL9, and CXCL11. (A) Receptor internalization. Wild-type-

CXCR3/300-19 (black bars) or Tail(-)-CXCR3/300-19 cells (gray bars) were exposed

to increasing amounts of CXCL10, CXCL9, or CXCL11 for 30 minutes. Subsequently,

the cell surface expression of wild-type-CXCR3 or Tail(-)-CXCR3 was measured by

flow cytometry and compared to unstimulated cells. The X-axis represents the indicated

chemokine concentration used to stimulate the cells. The Y-axis represents the receptor

expression relative to the unstimulated receptor expression normalized to 1. The data

shown represent the mean internalization of three experiments with error bars

representing the standard error of the mean. Using two-way analysis of variance

(ANOVA), p<0.01 for CXCL10 and CXCL9-induced internalization. There was no

statistical difference for CXCL11-induced internalization. (B) Chemotaxis. The bars

represent the migration of wild-type-CXCR3/300-19 cells (black bars) or Tail(-)-

CXCR3/300-19 cells (gray bars) across a Neuroprobe membrane to CXCL10, CXCL9,

or CXCL11. The X-axis represents the chemokine concentration and the Y-axis

represents the chemotactic index. The data shown are the mean of 2 samples and the

experiment shown is representative of 5 separate experiments.


- 33 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Figure 2. Calcium mobilization assays of CXCR3 and CXCR3 mutants. The top row

contains the calcium mobilization plots for wild-type-CXCR3/300-19 cells loaded with

Fura-2 and activated with 10 nM CXCL10, CXCL9, or CXCL11 as indicated. The

curves show the ratio of the 510 nm emissions after activation at 340 nm and 380 nm.

Subsequent rows contain the calcium mobilization plots of the mutant-CXCR3/300-19

cells as indicated to the same treatments. The plots are representative of at least three

experiments and are all shown on the same scale. Lines for wild-type-CXCR3, ST(-)-

CXCR3, and LLL-CXCR3 represent the baseline.

Figure 3. Phosphorylation and Western Analysis of wild-type-CXCR3 and Tail(-)-

CXCR3. (A) Phosphorylation. An autoradiogram of a 10% SDS-PAGE gel showing

immunoprecipitations of HEK293 cell extracts from cells transfected with wild-type-

Flag-CXCR3, Tail(-)-Flag-CXCR3, and untransfected HEK-293 cells. Cells were

either unstimulated or stimulated with 100 nM CXCL10 or 100 nM CXCL11.

Immunoprecipitation was performed using M2-anti-FLAG antibody conjugated to

agarose beads. (B) Western Blotting. Extracts from the wild-type-CXCR3/300-19 and

300-19 cells were run on 10% SDS-PAGE gels, transferred to nitrocellulose

membranes, blotted with the anti-CXCR3 antibody 1C6 developed by

chemiluminescence, and detected by autoradiography.


- 34 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Figure 4. Receptor internalization and chemotaxis of ST(-)-CXCR3. (A) Receptor

internalization. Wild-type-CXCR3/300-19 cells or ST(-)-CXCR3/300-19 cells were

exposed to increasing amounts of CXCL10, CXCL9, or CXCL11 for 15 minutes.

Subsequently, the cell surface expression of wild-type-CXCR3 (black bars) or ST(-)-

CXCR3 (gray bars) was measured by flow cytometry and compared to unstimulated cells.

The X axis represents the indicated chemokine concentration used to stimulate the cells.

The Y axis represents the receptor expression relative to the unstimulated receptor

expression normalized to 1. The data shown represent the mean internalization of three

experiments with error bars representing the standard error. Using ANOVA, the

differences were significant to P<0.01 for CXCL10 and CXCL9. (B) Chemotaxis.

Migration of wild-type-CXCR3/300-19 cells (black bars) or ST(-)-CXCR3/300-19

cells (gray bars) was measured in Neuroprobe chemotaxis chambers. The X-axis

represents the chemokine concentration and the Y-axis represents the chemotactic index.

The data shown are the means of 5 experiments, each done in duplicate. The differences

between wild-type-CXCR3/300-19 cells and ST(-)-CXCR3/300-19 cells using the

means of all 5 experiments were significant at p<0.01 for CXCL10, CXCL9, and

CXCL11 by ANOVA.

Figure 5. The effect of dynamin, ß-arrestin1, and ß-arrestin2 on CXCR3 internalization.

HEK293 cells were transiently co-transfected with CXCR3/pcDNA3.1 and either


- 35 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

dynamin dominant negative K44A/pCDNA3.1, ß-arrestin1 dominant negative V53D, ß-

arrestin2 dominant negative V54D (gray bars), or pcDNA3.1 as control (black bars).

Wild-type-CXCR3 cell surface expression was measured by flow cytometry after

stimulation with increasing amounts of CXCL10 or CXCL11. The X-axis represents the

indicated chemokine concentration used to stimulate the cells. The Y-axis represents the

receptor expression relative to the unstimulated receptor expression normalized to 1.

p<0.01 (dynamin dominant negative) and p<0.05 (ß-arrestin1 dominant negative) for

CXCL10-induced internalization using ANOVA. There was no statistical difference for

CXCL11-induced internalization or internalization in the presence of the ß-arrestin2

dominant negative. The data shown represent the mean internalization of four

experiments with error bars representing the standard error of the mean.

Figure 6. Receptor internalization and chemotaxis of i3-CXCR1-CXCR3. (A) Receptor

internalization. Wild-type-CXCR3/300-19 cells or i3-CXCR1-CXCR3/300-19 cells

were exposed to increasing amounts of CXCL10, CXCL9, or CXCL11 for 15 minutes.

Subsequently, the cell surface expression of wild-type-CXCR3 (black bars) or i3-

CXCR1-CXCR3 (gray bars) was measured by flow cytometry and compared to

unstimulated cells. The X axis represents the indicated chemokine concentration used to

stimulate the cells. The Y axis represents the receptor expression relative to the

unstimulated receptor expression normalized to 1. The data shown represent the mean


- 36 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

internalization of four experiments with error bars representing the standard error of the

mean. Using ANOVA, the differences between wild-type-CXCR3/300-19 cells and i3-

CXCR1-CXCR3/300-19 cells using the means of all four experiments were not

significant for CXCL10 and CXCL9, and significant for CXCL11 (p<0.01). (B)

Chemotaxis. Migration of wild-type-CXCR3/300-19 cells (black bars) or i3-CXCR1-

CXCR3/300-19 cells (gray bars) was measured in Neuroprobe chemotaxis chambers.

The X-axis represents the chemokine concentration and the Y-axis represents the

chemotactic index. The data shown are the mean of 2 samples and the experiment shown

is representative of 3 separate experiments. The differences in chemotactic index were

statistically significant for CXCL10 (p<0.01) and CXCL9 (p<0.01) but not for CXCL11

(p=0.058) by ANOVA.

Figure 7. Receptor internalization and chemotaxis of R149N-CXCR3 and LLL332-

334AAA-CXCR3. (A) Receptor internalization. Wild-type-CXCR3/300-19 (black

bars) or mutant-CXCR3/300-19 cells (gray bars) were exposed to increasing amounts of

CXCL10, CXCL9, or CXCL11 for 30 minutes. Subsequently, the cell surface expression

of wild-type-CXCR3 or mutant-CXCR3 was measured by flow cytometry and

compared to unstimulated. The X axis represents the indicated chemokine concentration

used to stimulate the cells. The Y axis represents the receptor expression relative to the

unstimulated receptor expression normalized to 1. The data shown represent the mean


- 37 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

internalization of three experiments with error bars representing the standard error of the

mean. p values are indicated on each chart. (B) Chemotaxis. Migration of wild-type-

CXCR3/300-19 cells (black bars) or mutant-CXCR3/300-19 cells (gray bars) was

measured in Neuroprobe chemotaxis chambers. The X-axis represents the chemokine

concentration and the Y-axis represents the chemotactic index. The data shown are the

mean of 2 samples and the experiment shown is representative of 3 separate experiments.

The differences in chemotactic index were statistically significant, p<0.01 for R149N-

CXCR3/300-19 cells, and p<0.05 for LLL332-334-CXCR3/300-19 cells, by ANOVA.

Figure 8. CXCR3 schematic. Domains required for activation by CXCL10 and CXCL9

are shown on the left and CXCL11 on the right. Black filled circles represent sites of

mutation that affected CXCR3 function. Italicized functions represent functions for

domains that were required differentially by the CXCR3 ligands.


- 38 -

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Richard A. Colvin, Gabriele S.V. Campanella, Jieti Sun and Andrew D. Lusterfunction

Intracellular domains of CXCR3 that mediate CXCL9, CXCL10, and CXCL11

published online May 17, 2004J. Biol. Chem.

10.1074/jbc.M403595200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

by guest on Novem


ww

.jbc.org/D

ownloaded from

http://www.jbc.org/lookup/doi/10.1074/jbc.M403595200

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;M403595200v1&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2004/05/17/jbc.M403595200.citation

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=early/2004/05/17/jbc&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/early/2004/05/17/jbc.M403595200.citation

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/

intracellular domains of cxcr3 that mediate cxcl9, cxcl10, and cxcl11 function

Documents