(Received for publication, March 12, 1997, and in revised form, May 7, 1997)
From the Theodor-Kocher Institute, University of Bern, CH-3000 Bern
9, Switzerland and the Biomedical Research Center,
University of British Columbia, Vancouver,
British Columbia, V6T 1Z3 Canada
The nucleotide sequence for a putative chemokine receptor, termed TER1, ChemR1, or CKR-L1, was recently obtained by a polymerase chain reaction-based cloning technique. It encodes a protein of 355 amino acids that shows 32-45% sequence identity with human chemokine receptors. The gene was localized on human chromosome 3p21-24, the site for the genes for the five known CC chemokine receptors, suggesting that the natural ligand may be a CC chemokine. We have stably expressed this receptor in murine pre-B cells 300-19 and have tested their responsiveness to 20 human chemokines and some other potential agonists. The CC chemokine I-309 was the only agonist that selectively induced intracellular Ca2+ mobilization and chemotaxis in receptor-transfected 300-19 cells. Stromal cell-derived factor 1, which binds to murine CXCR4 expressed in parental as well as transfected 300-19 cells, served as positive control in the functional screening. The interaction of I-309 with TER1 was of high affinity as shown by 125I-I-309 binding (Kd of 1.2 nM) and transient [Ca2+]i changes at subnanomolar concentrations of agonist. Migration responses in receptor-transfected 300-19 cells was typically bimodal with maximal activity at 10 nM of I-309. These data demonstrate that TER1 (ChemR1 or CKR-L1) is the receptor for I-309, and we propose to call this receptor CCR8 in agreement with the current nomenclature for chemokine receptors. The expression of CCR8 in blood leukocytes and lymphocytes was analyzed by Northern blot. No transcripts were found in RNA from freshly isolated blood neutrophils, monocytes, cultured macrophages, and phytohemagglutinin-stimulated T lymphocytes, and a faint hybridization signal corresponding to the RNA species of 4 kb was obtained only with RNA from interleukin-2-treated T lymphocytes. CCR8 is unusual for its selectivity for a single chemokine, previously shown only for CXCR1 and CXCR4, which bind interleukin-8 and stromal cell-derived factor 1, respectively. Identification of the receptor for I-309 represents a significant progress in determining the function of I-309 in inflammation and disease.
Chemokines are produced locally at sites of inflammation and infection and regulate the recruitment of leukocytes and lymphocytes (1, 2). All chemokines contain four conserved cysteines linked by disulfide bonds, and two subfamilies, CXC and CC chemokines, are defined on the basis of the first two cysteines, which are separated by one amino acid or are adjacent. Generally, chemokines act on more than one type of leukocyte and in vitro responses include chemotaxis, enzyme release from intracellular stores, oxygen radical formation, shape change through cytoskeletal rearrangement, generation of lipid mediators, and induction of adhesion to endothelium or extracellular matrix proteins (1-3). The overlap in target cell selectivity for many of the currently 34 known human chemokines can be taken as a measure of their importance in the regulation of immunity.
Chemokines act via seven transmembrane domain receptors that
couple to Bordetella pertussis toxin-sensitive G-proteins
for signal transduction (1, 4, 5). Similar to chemokines, the nine
human chemokine receptors are divided into two subfamilies, the CXC
chemokine receptors (CXCR) and the CC chemokine receptors (CCR), based
on their selectivity for either CXC or CC chemokines. Ligand
cross-selectivity, i.e. CXCRs that bind CC chemokines or vice versa, is not observed. Chemokine receptors consist of
350-368 amino acids and sequence identity among members of the two
receptor subfamilies are 36-77 and 46-74%, respectively. Most
chemokine receptors recognize more than one chemokine and many
chemokines, including IL-8,1 RANTES,
MIP-1 and the MCPs, bind to more than one receptor.
Several reports have described cDNAs for so-called orphan chemokine
receptors for which no ligand has been identified. Burkitt's lymphoma-derived receptor 1 and monocyte-derived receptor 15 are amino-terminal splicing variants of a gene expressed in monocytes, lymphocytes, and tissue cells (6, 7). The Epstein-Barr virus-infected Burkitt's lymphoma receptor EBI1 shows 40% sequence identity with the
IL-8 receptors and is expressed exclusively in B and T cell lines (8).
The chemokine receptor-like protein CMKBRL1 (9), also known as V28
(10), shows higher similarity to known CCRs than CXCRs. The most recent
orphan receptor, referred to as TER1 (11), ChemR1 (12), or CKR-L1 (13),
shares 32-45% sequence identity with chemokine receptors. Its gene is
co-localized with the genes for all five known CC chemokine receptors
on human chromosome 3 band p21-24 (14).
Here, we demonstrate that TER1 (ChemR1 or CKR-L1) is the receptor for the CC chemokine I-309. In stably transfected mouse pre-B cells this receptor mediated intracellular Ca2+ mobilization and in vitro migration in response to I-309, whereas no other of 20 human chemokines tested was active. In accordance with the new nomenclature rules for chemokine receptors, we have named the I-309 receptor CCR8.
The CXC chemokines Mig, IP10, IL-8,
GRO, NAP-2, GCP-2, ENA78, SDF-1, PF4, the CC chemokines MCP-1,
MCP-2, MCP-3, MCP-4, MIP-1
, MIP-1
, RANTES, I-309, HCC-1, TARC,
and the neuropeptide NPY were chemically synthesized according to
established protocols (15). Somatostatin, substance P, and calcitonin
were purchased from Sigma. C5a and C3a were a gift of Dr. C. A. Dahinden (Clinical Immunology, University Hospital, Bern,
Switzerland).
CCR8 DNA was generated
by PCR with primers TER1SE and TER1AS corresponding to the nucleotide
sequence of TER1 (11) and cDNA from IL-2-stimulated PBL as
amplification template (16). The primers TER1SE,
5-TTATGTGTCTCTGTGACCAG, and TER1AS, 5
-TAGTCTTCATTGATCCTCAC, correspond to positions 327-346 and 1425-1444 in the published sequence and span the entire coding region for TER1 of 355 amino acids.
PCR amplifications were performed in total volumes of 30 µl,
containing 1 × PCR buffer including 1.5 mM
MgCl2 (Life Technologies, Inc.), 2 µl of cDNA
preparation, 0.4 µM primers, and 400 µM
dNTPs. To start DNA synthesis 1.2 units of Taq polymerase
(Life Technologies, Inc.) were added after cDNA denaturation and
allowed to progress for 30 cycles (1 min at 50 °C, 45 s at
72 °C, and 15 s at 95 °C) in a master cycler 5330 (Eppendorf-Netheler-Hinz, Inc.). The 1118-base pair DNA fragments were
cloned in Gene Scribe-Z vector pTZ18R (U. S. Biochemical Corp.),
sequenced, and subcloned into the mammalian expression vector SR
puro
(provided by Dr. F. Arenzana-Seisdedos, Pasteur Institut, Paris,
France). To generate stable transfectants, 5 × 106
mouse pre-B cells 300-19 in 400 µl of electroporation buffer were
transfected with 20 µg of linearized SR
puro-CCR8 plasmid DNA as
described (17). Puromycin-resistant 300-19 clones were established by
limited dilution culturing in the presence of 1.5 µg/ml puromycin
(Sigma) and screened for CCR8 expression by RNA Dot-blot analysis.
10-µg samples of total RNA were examined from freshly isolated human blood monocytes, neutrophils, lymphocytes, cultured macrophages, and activated T cells by Northern blot analysis as described (16, 18) using a 32P-labeled CCR8 DNA insert (109 cpm/µg DNA) corresponding to the 620-base pair fragment defined by positions 327 and 946 in the published sequence (11) as hybridization probe. As internal control, a 2-µg sample of total RNA from CCR8-transfected 300-19 cells was included.
125I-I-309 Binding to 300-19-CCR8 TransfectantsSynthetic I-309 was iodinated with Bolton and Hunter reagent (Amersham Corp.) as instructed by the supplier. Briefly, 0.4 nmol of peptide ligand in 25 µl of 0.1 M borate buffer, pH 8.5, were added to 1 mCi of 125I-labeled Bolton and Hunter reagent and incubated on ice for 25 min. To stop the labeling reaction, 200 µl of 0.2 M glycine, pH 8.5, were added, and the mixture was kept on ice for a further 5 min. Iodinated peptide was separated from 125I-labeled Bolton and Hunter reagent by gel filtration chromatography (Bio-Gel P-6 DG, Bio-Rad). Binding of 125I-I-309 to CCR8 transfected 300-19 cells and analysis of bound radioligand were performed as described (17). Bindability of three independently labeled 125I-I-309 preparations was 12-18%.
300-19-CCR8 Cell Responses to ChemokinesChanges in the cytosolic free Ca2+ concentration ([Ca2+]i) were measured in cells loaded with fura-2/AM by incubation for 25 min at 37 °C with 0.1 nmol of fura-2/AM per 106 cells in a buffer containing 136 mM NaCl, 4.8 mM KCl, 1 mM CaCl2, 5 mM glucose, and 20 mM HEPES, pH 7.4. After centrifugation, loaded cells were resuspended in the same buffer (3 × 106 cells/ml) and stimulated with the indicated chemokine at 37 °C, and the [Ca2+]i-related fluorescence changes were recorded (19). In vitro migration of CCR8-transfected and parental 300-19 cells was assessed in 48-well chambers (Neuro Probe, Cabin John, MD) using collagen IV-precoated polyvinylpyrrolidone-free polycarbonate membranes (Nucleopore) with 5-µm pores (18). RPMI 1640 supplemented with 20 mM HEPES, pH 7.4, and 1% pasteurized plasma protein solution (Swiss Red Cross Laboratory, Bern, Switzerland) was used to dissolve the chemokines (lower wells) and to dilute the cells (105 of 300-19-CCR8 cells/upper well). Following incubation for 120 min at 37 °C under CO2-buffered conditions, the membrane was removed, and the adherent cells were fixed, stained, and quantified in five randomly selected fields at 1,000-fold magnification in triplicate wells.
The orphan receptor TER1 shares 39-45% identical amino acids with CC chemokine receptors and only 32-34% with CXC chemokine receptors. The highest similarity of 45% is found with CCR4 (1, 4, 5). Many structural features that distinguish chemokine receptors from other G-protein-coupled serpentine receptors are found in TER1, notably the extended DRY motif in the second intracellular loop (DRYLA(V/I)VHA(T/V)), H(S/C)CXNPXXYXF in the seventh transmembrane domain, and YLLNLA(L/I)SDL in transmembrane domain 2, which is more conserved in CC chemokine receptors (1, 4, 5).
Receptor-mediated [Ca2+]i ChangesFor functional characterization, DNA containing the coding region for TER1 was generated by PCR and used for transfection of the mouse pre-B cell line 300-19. Cell clones with stable expression of TER1 mRNA (300-19-CCR8) were examined for responses to 20 human chemokines and other potential agonists (Table I), and changes in the intracellular Ca2+ concentration ([Ca2+]i) that are typical for chemokine receptor signaling were monitored. Of all proteins listed in Table I, only SDF-1 and I-309 were active in 300-19 cells expressing the receptor. I-309 induced typical transient [Ca2+]i changes in receptor-transfected but not in parental cells, demonstrating that the previous orphan receptor is selective for the CC chemokine I-309, which we have termed CCR8 in keeping with the current chemokine receptor nomenclature (Fig. 1a). The interaction of CCR8 with I-309 is of high affinity, as judged by the rapid increases in the [Ca2+]i changes and responses to subnanomolar concentrations of agonist. The CXC chemokine SDF-1 is selective for the chemokine receptor CXCR4 and known to induce chemokine responses in cells expressing the murine homologue of CXCR4 (20, 21). Thus, [Ca2+]i responses to SDF-1 in parental (not shown) and CCR8-transfected 300-19 cells that express murine CXCR4 are considered as positive control for functional integrity of these cells to chemokine stimulation. Selectivity of I-309 for CCR8 is further documented in Ca2+ cross-desensitization experiments (Fig. 1b). Stimulation of leukocytes with chemokines generally results in transient unresponsiveness to those chemokines, which bind to the same receptor (1). Stimulation of CCR8-expressing 300-19 cells with 100 nM I-309 completely abolished a second response to I-309. By contrast, stimulation with I-309 did not prevent a second response to 100 nM of SDF-1 and vice versa responsiveness to I-309 was not affected by initial stimulation with SDF-1, indicating that SDF-1 was not interacting with CCR8.
|
Chemotaxis
The prototypical response in vitro of
chemokines in leukocytes is chemotactic migration, which is routinely
measured in chemotaxis microchambers (1, 2). Results in Fig.
2 demonstrate that CCR8 was very efficient in mediating
transfected 300-19 cell migration in response to I-309, whereas, as
expected, no effects were seen in parental cells. Responses to
increasing concentrations of I-309 were bimodal, which is
characteristic for chemokines as opposed to agonists with chemokinetic
activity, and maximal migration was observed at 10 nM.
I-309 Binding to CCR8
300-19 cells expressing CCR8 bound
radiolabeled I-309 with high affinity (Fig. 3).
Scatchard analysis of the specifically bound 125I-I-309
revealed 1.5 × 104 binding sites per cell and a
binding constant (Kd) of 1.2 nM (inset
in Fig. 3). Unspecifically bound material was in the range of 30-60%
of total cell-associated radioactivity, which could not be reduced by
alternative radioiodination protocols or changes in the binding
conditions. No specific 125I-I-309 binding was observed
with parental (untransfected) 300-19 cells.
The cDNAs for I-309 and its murine homologue TCA3 were cloned as T cell activation-dependent gene products by subtractive hybridization (22, 23). I-309 and TCA3 were shown to be readily expressed in activated (concanavalin A, PHA, and phorbol myristate acetate) T cell and natural killer cell lines but not in resting T cell lines (22-24). Transcripts for I-309/TCA3 were not detected in B lymphoid and myeloid cell lines and in many organs, suggesting that activated T cells are the primary sites of I-309/TCA3 production.
Recombinant proteins were available for almost 10 years; however, reports on I-309 and TCA3 function are sparse and in part controversial. Chemotactic activity was described with monocytes and the monocytic cell line THP-1 and tissue cells including mesangial and smooth muscle cells (25-28). As opposed to many CC chemokines, I-309 and TCA3 were reported to be inactive on cultured T cell and natural killer cell lines (29, 30). Although no responses were seen with I-309 in human neutrophils (25), TCA3 was found to be an attractant of murine neutrophils (27, 31, 32). In addition to chemotaxis, TCA3 was reported to mediate enzyme release, oxygen metabolite production and adhesion to fibrinogen in neutrophils and macrophages, and stimulation of growth and adhesion to fibrinogen in smooth muscle and mesangial cells (26, 28, 31). Most recently, I-309 was found to prevent dexamethasone-induced apoptosis in cultured murine thymic lymphoma cells (33).
Receptor ExpressionExpression of CCR8 mRNA was examined
by Northern blot analysis in total RNA from human blood cells. As shown
in Fig. 4, none of the cells was positive for CCR8
transcripts, including freshly isolated blood neutrophils, monocytes,
macrophages, and PBL. As positive control, 2 µg of total RNA from a
CCR8-expressing 300-19 cell clone was included. The experimental
conditions were such that transcripts for other chemokine receptors
were readily detected, as previously reported for the IL-8 receptors
(CXCR1 and CXCR2) and the RANTES/MIP-1 receptors (CCR1) in
neutrophils (16, 17), CCR1 and the MCP-1 receptor (CCR2) in monocytes
(16), and CXCR4 in all types of leukocytes (34). Lack of detectable
CCR8 transcript expression in neutrophils and lymphocytes is in full
agreement with the reported inactivity of I-309 in these cells (25).
However, lack of CCR8 transcripts in monocytes is in striking contrast to the reported I-309-induced monocyte responses. This discrepancy was
not due to differences in the I-309 preparations used because our
synthetic I-309 was identical in sequence to the reported recombinant
protein (25), and synthetic and recombinant I-309 were equally potent
as stimuli of CCR8-expressing 300-19 cells. The reported chemotactic
activity on monocytes (25) is much weaker than the activity observed in
the present study on CCR8-expressing 300-19 cells, suggesting that
monocytes may express a receptor with lower affinity for I-309.
Except for CXCR4, which is ubiquitously expressed in circulating blood cells and blood cell precursors, freshly isolated PBL have low to undetectable levels of mRNA for any chemokine receptor and do not respond to the corresponding chemokines (16-18). However, culturing of PBL for several days in the presence of IL-2 results in expansion of CD45RO+ T cells with strong expression of receptors, as recently demonstrated for CCR1, CCR2, and CXCR3 (16, 18). By contrast, treatment of PBL with IL-2 did not affect CCR8 expression, and in addition, PHA treatment for 3 days failed to induce receptor expression (Fig. 4). Only faint hybridization signals corresponding to mRNA species of approximately 4 kilobases were occasionally observed in IL-2 cultured PBL, indicating donor to donor variation in CCR8 expression.
Here, we have shown that CCR8-transfected 300-19 cells responded to I-309 with high potency and efficacy, suggesting that natural target cells for I-309 might respond equally well. The primary target cells for chemokines are leukocytes and lymphocytes that express the corresponding receptors in quantities large enough for ready detection in total RNA by Northern blot analysis (1, 4, 5). Similar to CXCR3, the receptor for the CXC chemokines IP10 and Mig (18), we did not find CCR8 transcripts in freshly isolated blood neutrophils, monocytes, or lymphocytes, which therefore may not represent true targets for I-309 action. By contrast to CXCR3 and several other CC chemokine receptors (16, 18), culturing of PBL in the presence of IL-2 did not result in expression of similar levels of CCR8 transcripts. The faint CCR8 mRNA hybridization signals, as observed in the RNA from some IL-2-treated T cell cultures, suggests that I-309 may act on a subpopulation of activated T cells. Molecular probes for CCR8 will now facilitate the identification of the natural target cells and thus help to advance our understanding of I-309 function in inflammation and disease.
We thank Dr. P. Loetscher for providing RNA from monocytes and macrophages. Recombinant I-309 was a kind gift of Dr. J.-C. Renauld (Ludwig Institute for Cancer Research, Brussels, Belgium). Donor blood buffy coats were provided by the Swiss Central Laboratory Blood Transfusion Service, Schweizerisches Rotes Kreuz.