The approximately 50 known chemokines are classified in distinct subfamilies: CXC, CC, CX3C, and C. Although the signaling of chemokines often is promiscuous, signaling events between members of these distinct chemokine classes are hardly observed. The only known exception so far is the murine CC chemokine ligand (CCL)21 (secondary lymphoid tissue chemokine, Exodus-2, 6Ckine), which binds and activates the murine CXC chemokine receptor CXCR3. However, this exception has not been found in humans. In this study, we provide evidence that human CCL21 is a functional ligand for endogenously expressed CXCR3 in human adult microglia. In absence of CCR7 expression, CCL21 induced chemotaxis of human microglia with efficiency similar to the CXCR3 ligands CXC chemokine ligand 9 (monokine induced by IFN-γ) and CXC chemokine ligand 10 (IFN-γ-inducible protein-10). Because human CCL21 did not show any effects in CXCR3-transfected HEK293 cells, it is indicated that CXCR3 signaling depends on the cellular background in which the CXCR3 is expressed.

In the peripheral immune system chemokines regulate migration of leukocytes expressing specific chemokine receptors (1). Based on conserved cysteine residues, chemokines and their receptors are subdivided into four families: CXC, CC, C, and CX3C chemokines (2). Chemokine signaling can be promiscuous and a variety of chemokines activate more than one chemokine receptor. However, this promiscuity is only observed within chemokine families, signaling between members of different chemokine families is no common phenomenon (3, 4). The only known exception so far concerns CC chemokine ligand (CCL)3 21 that binds and activates mouse CXCR3 (5, 6, 7). In the periphery, CCL21 and the related chemokine CCL19 (EBV-induced molecule-1 ligand chemokine) are known to be homeostatic chemokines that activate CCR7, whereas CXCR3 is activated by CXC chemokine ligand (CXCL)9, CXCL10, and CXCL11 chemokines that belong to the inflammatory chemokines (3, 4).

It is clear that chemokines and their receptors are also expressed in the CNS. Here chemokines, at least partially, control the infiltration of blood leukocytes into the brain and are involved in different brain pathologies (8, 9, 10).

Recently, we have described the induction of CCL21 expression in ischemic mouse neurons and the effects of CCL21 in microglia, suggesting the involvement of CCL21 in the communication between damaged neurons and microglia (11, 12, 13). Surprisingly, murine microglia responded to CCL21 stimulation in the absence of CCR7 via the alternative receptor CXCR3 (13, 14, 15), confirming results previously obtained in a recombinant expression system (5).

The validity of CCL21 as a neuronally derived activator of microglia in humans was challenged because activity of human CCL (hCCL)21 on human CXCR (hCXCR)3 has not been found (6, 7). However, none of these studies were conducted in (CCR7 lacking) primary human cells. Therefore, we studied possible effects of human CCL21 in primary adult microglia.

Human brain tissue was obtained by rapid autopsy. All patients or their next of kin had given written consent for autopsy during life, and for use of their brain tissue for research purposes. Microglial cell cultures were obtained as described previously (16). To induce proliferation of microglial cells, recombinant human GM-CSF (stock solution 300 μg/ml; Leucomax, Novartis, The Netherlands) was added (final concentration 25 ng/ml). Microglial cells immunostained (>90–98%) with anti-Mac1 (CD11b), LeuM5 (CD11c), LeuM3 (CD14), KP1 (CD68), and FcγRI (CD64), whereas they were negative for glial fibrillary acidic protein (GFAP). HEK293 cells were maintained in DMEM containing 10% FCS with 0.01% penicillin and 0.01% streptomycin in a humidified atmosphere (5% CO2) at 37°C.

Full-length murine CXCR (mCXCR)3 was obtained from total RNA extracted from murine mixed glial cell cultures as described (17). Primers for the full-length sequence of mCXCR3 were designed (accession no. AF045146): forward primer: 5′-GCAAGTTCCCAACCACAAGT-3′; backward primer: 5′-CGAGCTAGCCCAACTACCAGGAGGA-3′. The PCR product was cloned into pCRII.1 (Invitrogen, Breda, The Netherlands), sequenced, and subcloned into pCDNA3.1 for transfection experiments. Plasmids containing hCXCR3 were a kind gift of C. Tensen (Vrije Universiteit Medical Center, Amsterdam, The Netherlands) and B. Moser (Theodor-Kocher Institute, Bern, Switzerland) (18). One microgram of the plasmids was transfected with 6 μl of Fugene (Roche Molecular Biochemicals, Almere, The Netherlands) in HEK293 cells. Stable transfected cells were selected with G418 500 μg/ml. Mock transfections were performed with pcDNA 3.1 containing the reversed CXCR3 sequence.

Cells were seeded on poly-l-lysine-coated glass cover slides and fixed in 4% paraformaldehyde for 30 min. After rinsing in PBS and preincubation in PBS containing 0.3% Triton X-100 (PBS+) and 5% normal goat serum, slides were incubated in PBS+ with 2% normal goat serum and human CD11b (Mac-1; Chemicon International, Hofheim, Germany) or mouse anti hCXCR3 (R&D Systems, Abingdon, U.K.) for 2 h and visualized with a Cy3-labeled goat anti-mouse secondary Ab. Control experiments for all immunocytochemical stainings were done by incubating cells with an isotype-matched primary Ab against GFAP (Chemicon International) or in the absence of primary Abs.

RNA preparation and reverse transcriptase were performed as described previously (17). Two microliters of cDNA were amplified in the PCR using the following two primer pairs for CCR7: forward, 5′-ACCAATGAAAAGCGTGCTG-3′; backward, 5′-CGCCGATGAAGGCGTACAA-3′ (977 bp), and forward, 5′-TGAGGTCACGGACGATTACA-3′; backward, 5′-CATGCCACTGAAGAAGCTCA-3′ (355 bp) with an annealing temperature of 56°C and 35 cycles. As a control for cDNA quality, experiments with primers for the housekeeping gene GAPDH: forward, 5′-CATCCTGCACCACCAACTGCTTAG-3′; backward, 5′-GCCTGCTTCACCACCTTCTTGATG-3′ were done in parallel with an annealing temperature of 60°C and 28 cycles.

The cellular distribution of CXCR3 mRNA was investigated by in situ hybridization experiments as described earlier (17) using CXCR3 digoxigenin-conjugated sense and antisense probes.

Migration of microglia and CXCR3-transfected HEK293 cells in response to CXCL9, CXCL10, CCL19, CCL21, and CX3C chemokine ligand (CX3CL)1 (R&D Systems) was determined as described earlier (17). For experiments in the presence of CXCL10, microglia were first pretreated with 10 nM CXCL10 for 10 min at 37°C and thereafter the chemotaxis was measured in the presence of 10 nM CXCL10 in the lower and the upper chamber of the microchamber. Microglial cells (1.5 × 104) or 3 × 104 transfected HEK293 cells were applied per well (50 μl) in the upper chamber. Migrated cells were counted and data are presented as percentages of control migration.

Statistical analysis and comparison of the different groups was done with the Tukey ANOVA test for multiple comparison (SPSS, Chicago, IL) with a significance level of 0.05.

The presence of CXCR3 in stably transfected HEK293 cells and microglia was confirmed by immunocytochemistry (Fig. 1) and in situ hybridization (Fig. 2). Stably transfected HEK293 cells showed intense immunostaining for hCXCR3 (Fig. 1,A) and mCXCR3 (Fig. 1,B). Nontransfected HEK293 cells gave negative staining results (not shown). In cultured microglia, >92% of the cells were found positive for CD11b confirming the identity of these cells (Fig. 1,C). Parallel stainings for CXCR3 resulted in >95% CXCR3-positive cells (Fig. 1 D). Control experiments with an isotype-matched (IgG1) Ab against GFAP (Chemicon International), or without primary Abs did not show staining (data not shown). As already shown for CXCR3 and other chemokine receptors in glial cells, immunostaining for CXCR3 was predominantly located in the cytoplasm with minor staining the membrane (17, 19, 20).

FIGURE 1.

CXCR3 protein expression in CXCR3-transfected HEK293 cells and cultured human microglial cells. HEK293 cells transfected with human or mouse CXCR3 revealed positive staining (A and B, respectively). CD11b staining in >92% of the cells identified them as microglia (C). A pronounced CXCR3 signal was detected in human microglia (D). Control experiments with an isotype-matched Ab against GFAP or without primary Abs yielded no signal (not shown). Scale bars = 50 μm.

FIGURE 1.

CXCR3 protein expression in CXCR3-transfected HEK293 cells and cultured human microglial cells. HEK293 cells transfected with human or mouse CXCR3 revealed positive staining (A and B, respectively). CD11b staining in >92% of the cells identified them as microglia (C). A pronounced CXCR3 signal was detected in human microglia (D). Control experiments with an isotype-matched Ab against GFAP or without primary Abs yielded no signal (not shown). Scale bars = 50 μm.

Close modal
FIGURE 2.

CCR7 and CXCR3 mRNA expression in human microglia and HEK293 cells. CCR7 expression in human microglia (mic) and HEK293 cells (HEK) was investigated by RT-PCR (A). Two different primer sets (a and b) for CCR7 were used, resulting in a product of 977 or 355 bp, respectively. No CCR7 mRNA expression was observed in human microglia or HEK293 cells whereas lymph node (LN) tissue showed bands of the expected size. The quality of all cDNAs was verified by using GAPDH primers in parallel experiments. nc: negative control, MM: m.w. marker, the highlighted band is 500 bp. In situ hybridization revealed the expression of CXCR3 mRNA in human microglia (see arrowheads for the dark hybridization signal) (B) and hCXCR3-transfected HEK293 cells (C). Hybridization with sense probes did not yield a signal (C, inset). Scale bars = 10 μm.

FIGURE 2.

CCR7 and CXCR3 mRNA expression in human microglia and HEK293 cells. CCR7 expression in human microglia (mic) and HEK293 cells (HEK) was investigated by RT-PCR (A). Two different primer sets (a and b) for CCR7 were used, resulting in a product of 977 or 355 bp, respectively. No CCR7 mRNA expression was observed in human microglia or HEK293 cells whereas lymph node (LN) tissue showed bands of the expected size. The quality of all cDNAs was verified by using GAPDH primers in parallel experiments. nc: negative control, MM: m.w. marker, the highlighted band is 500 bp. In situ hybridization revealed the expression of CXCR3 mRNA in human microglia (see arrowheads for the dark hybridization signal) (B) and hCXCR3-transfected HEK293 cells (C). Hybridization with sense probes did not yield a signal (C, inset). Scale bars = 10 μm.

Close modal

Using RT-PCR experiments, we were not able to detect expression of CCR7 in both microglia and CXCR3-transfected HEK293 cells (Fig. 2). As a positive control to the two different sets of primers, lymph node tissue was used where pronounced expression of CCR7 mRNA was found (Fig. 2). In a recent publication, we already demonstrated CXCR3 mRNA expression in human microglia cells by means of RT-PCR (17) which was further confirmed by in situ hybridization analysis. CXCR3-transfected HEK293 served as a positive control for the used probe. Pronounced CXCR3 mRNA expression was detected in transfected HEK293 (Fig. 2,C), whereas the sense control appeared was negative (Fig. 2,C, inset). In human cultured microglia, CXCR3 mRNA expression was found in the small rim of cytoplasm around the nucleus (Fig. 2 B, see arrowheads), a signal that was absent with the sense probe (not shown).

The functional activity of CXCR3 in murine microglia in response to CXCL9 (10) and CCL21 was demonstrated recently (13, 15, 17). Similar as murine microglia, human microglia migrated in response to hCCL21 in a concentration-dependent manner (Fig. 3,A). The maximum migratory response was found, between 0.1 and 1 nM, which was comparable with migratory responses for CXCL9 or CXCL10 (Fig. 3,A) (17). Moreover, CCL21-induced chemotaxis was sensitive to pertussis toxin, indicating the involvement of Gi-coupled chemokine receptors (Fig. 3,A). Again, similar to our findings in murine microglia, hCCL19 had no effect in human microglia (Fig. 3 A).

FIGURE 3.

Effects of CXCR3 ligands on the migratory behavior of human microglia and CXCR3-transfected HEK293 cells. A, Concentration-dependent chemotaxis of human microglia in response to CCL21, pretreatment with pertussis toxin (PTX) inhibited the migratory response to CCL21. No typical concentration-dependent migration was observed in response to the CCR7 ligand CCL19. The maximum migratory response observed with CCL21 was comparable to those for other CXCR3 ligands CXCL9 and CXCL10. B, The migratory response of 1 nM CCL21 was significantly inhibited by the presence of 10 nM CXCL10. No effect of 10 nM CXCL10 was found on the migratory response of 1 nM CX3CL1. C, HEK293 cells transfected with mCXCR3 migrated toward both CXCL9, CXCL10 (data not shown), and CCL21 in a dose-dependent manner. D, hCXCR3-transfected HEK293 cells showed chemotaxis in response to CXCL9 and CXCL10 (data not shown) but not to CCL21. The graphs provide the results from a typical experiment; similar results were obtained in at least three independent experiments. Data are given in mean ± SEM, for experiments with human microglia: n = 4 and with transfected HEK293: n = 6. ∗, Significant different from control values (p ≤ 0.05).

FIGURE 3.

Effects of CXCR3 ligands on the migratory behavior of human microglia and CXCR3-transfected HEK293 cells. A, Concentration-dependent chemotaxis of human microglia in response to CCL21, pretreatment with pertussis toxin (PTX) inhibited the migratory response to CCL21. No typical concentration-dependent migration was observed in response to the CCR7 ligand CCL19. The maximum migratory response observed with CCL21 was comparable to those for other CXCR3 ligands CXCL9 and CXCL10. B, The migratory response of 1 nM CCL21 was significantly inhibited by the presence of 10 nM CXCL10. No effect of 10 nM CXCL10 was found on the migratory response of 1 nM CX3CL1. C, HEK293 cells transfected with mCXCR3 migrated toward both CXCL9, CXCL10 (data not shown), and CCL21 in a dose-dependent manner. D, hCXCR3-transfected HEK293 cells showed chemotaxis in response to CXCL9 and CXCL10 (data not shown) but not to CCL21. The graphs provide the results from a typical experiment; similar results were obtained in at least three independent experiments. Data are given in mean ± SEM, for experiments with human microglia: n = 4 and with transfected HEK293: n = 6. ∗, Significant different from control values (p ≤ 0.05).

Close modal

Using CXCR3-deficient mice, it was demonstrated that mCXCR3 is the responsible chemokine receptor for murine CCL21 in microglia (15). To further address this also in human microglia, we first tried to block CXCR3 signaling by Ab treatment. However, our hCXCR3 Ab did not influence the migration of hCXCR3-transfected HEK203 cells in response to CXCL9 and 10 (data not shown) and was therefore not suitable. Next, a desensitization approach was performed because we have shown previously, that CCL21 and CXCL10 cross-desensitize their signaling in mice, indicating that both ligands activate the same receptor (13). It is shown here that the effect of CCL21 was absent in microglia in the presence of 10 nM CXCL10 (Fig. 3,B), whereas the chemotactic activity of another chemokine, CX3CL1, was not affected (Fig. 3 B). These results strongly suggest that CXCR3 is desensitized by the presence of 10 nM CXCL10 and the lack of a CCL21 response indicates that CXCR3 is the responsible chemokine receptor for CCL21 in human microglia.

HEK293 cells were transfected with murine or human CXCR3 and investigated for possible responses to CCL21 and other CXCR3 ligands. In agreement with previously published findings, it was found that HEK293 cells transfected with mCXCR3 responded to CCL21 and CXCL9 in a concentration-dependent manner (Fig. 3,C). HEK293 cells transfected with hCXCR3 migrated when exposed to CXCL9 or CXCL10, but no migratory response was observed to CCL21 (Fig. 3 D) (5, 6). Migration of mock-transfected HEK293 was never observed (data not shown).

Pharmacological differences of chemokine receptors expressed in different expression systems have already been published (see CCR5 signaling in various cells; Refs. 21, 22, 23). Our results suggest that the pharmacology of hCXCR3 depends on its cellular background. Because we did not detect a microglia-specific splice variant for CXCR3 in mice and humans (data not shown), other mechanisms must account for the here-presented difference. Post-translational modifications (i.e., palmitoylation), which may influence signaling cascades are known for chemokine receptors (24, 25). Moreover, interactions with regulator of G protein signaling protein may influence chemokine receptor functioning (26) or chemokine receptor heterodimerization might change the pharmacology of hCXCR3 in microglia as it was shown for CCR2 (27).

Hence, the mechanism accounting for the effects of hCCL21 in human microglia requires further investigation. However, the presented results show that hCCL21 induces functional responses in human microglia similar to those in murine cells. This shows for the first time functional activity of hCXCR3 in response to hCCL21. Because hCCL21 expression was recently detected in cerebrospinal fluid of patients with neurodegenerative diseases (28), it is suggested that hCCL21 is involved in neuroimmune signaling also in humans.

1

Part of this project was supported by the Dutch Foundation “Stichting MS Research” Grants 95-237 and 02-510her.

3

Abbreviations used in this paper: CCL, CC chemokine ligand; CXCL, CXC chemokine ligand; hCCL, human CCL; hCXCR, human CXCR; GFAP, glial fibrillary acidic protein; mCXCR, murine CXCR; CX3CL, CX3C chemokine ligand.

1
Ebnet, K., D. Vestweber.
1999
. Molecular mechanisms that control leukocyte extravasation: the selectins and the chemokines.
Histochem. Cell Biol.
112
:
1
.
2
Rossi, D., A. Zlotnik.
2000
. The biology of chemokines and their receptors.
Annu. Rev. Immunol.
18
:
217
.
3
Mantovani, A..
1999
. The chemokine system: redundancy for robust outputs.
Immunol. Today
20
:
254
.
4
Murphy, P. M., M. Baggiolini, I. F. Charo, C. A. Hebert, R. Horuk, K. Matsushima, L. H. Miller, J. J. Oppenheim, C. A. Power.
2000
. International union of pharmacology. XXII. Nomenclature for chemokine receptors.
Pharmacol. Rev.
52
:
145
.
5
Soto, H., W. Wang, R. M. Strieter, N. G. Copeland, D. J. Gilbert, N. A. Jenkins, J. Hedrick, A. Zlotnik.
1998
. The CC chemokine 6Ckine binds the CXC chemokine receptor CXCR3.
Proc. Natl. Acad. Sci. USA
95
:
8205
.
6
Jenh, C. H., M. A. Cox, H. Kaminski, M. Zhang, H. Byrnes, J. Fine, D. Lundell, C. C. Chou, S. K. Narula, P. J. Zavodny.
1999
. Cutting edge: species specificity of the CC chemokine 6Ckine signaling through the CXC chemokine receptor CXCR3: human 6Ckine is not a ligand for the human or mouse CXCR3 receptors.
J. Immunol.
162
:
3765
.
7
Wang, X., X. Li, D. B. Schmidt, J. J. Foley, F. C. Barone, R. S. Ames, H. M. Sarau.
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
.
8
Glabinski, A. R., R. M. Ransohoff.
1999
. Chemokines and chemokine receptors in CNS pathology.
J. Neurovirol.
5
:
3
.
9
Asensio, V. C., C. Kincaid, I. L. Campbell.
1999
. Chemokines and the inflammatory response to viral infection in the central nervous system with a focus on lymphocytic choriomeningitis virus.
J. Neurovirol.
5
:
65
.
10
Mennicken, F., R. Maki, E. B. de Souza, R. Quirion.
1999
. Chemokines and chemokine receptors in the CNS: a possible role in neuroinflammation and patterning.
Trends Pharmacol. Sci.
20
:
73
.
11
Streit, W. J., S. A. Walter, N. A. Pennell.
1999
. Reactive microgliosis.
Prog. Neurobiol.
57
:
563
.
12
Bruce-Keller, A. J..
1999
. Microglial-neuronal interactions in synaptic damage and recovery.
J. Neurosci. Res.
58
:
191
.
13
Biber, K., A. Sauter, N. Brouwer, S. C. Copray, H. W. Boddeke.
2001
. Ischemia-induced neuronal expression of the microglia attracting chemokine secondary lymphoid-tissue chemokine (SLC).
Glia
34
:
121
.
14
Biber, K., M. W. Zuurman, I. M. Dijkstra, H. W. Boddeke.
2002
. Chemokines in the brain: neuroimmunology and beyond.
Curr. Opin. Pharmacol.
2
:
63
.
15
Rappert, A., K. Biber, C. Nolte, M. Lipp, A. Schubel, B. Lu, N. P. Gerard, C. Gerard, H. W. Boddeke, H. Kettenmann.
2002
. Secondary lymphoid tissue chemokine (CCL21) activates CXCR3 to trigger a Cl current and chemotaxis in murine microglia.
J. Immunol.
168
:
3221
.
16
De Groot, C. J., L. Montagne, I. Janssen, R. Ravid, P. Van Der Valk, R. Veerhuis.
2000
. Isolation and characterization of adult microglial cells and oligodendrocytes derived from postmortem human brain tissue.
Brain Res. Brain Res. Protoc.
5
:
85
.
17
Biber, K., I. Dijkstra, C. Trebst, C. J. De Groot, R. M. Ransohoff, H. W. Boddeke.
2002
. Functional expression of CXCR3 in cultured mouse and human astrocytes and microglia.
Neuroscience
112
:
487
.
18
Hensbergen, P. J., E. M. van der Raaij-Helmer, R. Dijkman, R. C. van der Schors, G. Werner-Felmayer, D. M. Boorsma, R. J. Scheper, R. Willemze, C. P. Tensen.
2001
. Processing of natural and recombinant CXCR3-targeting chemokines and implications for biological activity.
Eur. J. Biochem.
268
:
4992
.
19
Guillemin, G. J., J. Croitoru-Lamoury, D. Dormont, P. J. Armati, B. J. Brew.
2003
. Quinolinic acid upregulates chemokine production and chemokine receptor expression in astrocytes.
Glia
41
:
371
.
20
Rezaie, P., G. Trillo-Pazos, I. P. Everall, D. K. Male.
2002
. Expression of β-chemokines and chemokine receptors in human fetal astrocyte and microglial co-cultures: potential role of chemokines in the developing CNS.
Glia
37
:
64
.
21
Blanpain, C., I. Migeotte, B. Lee, J. Vakili, B. J. Doranz, C. Govaerts, G. Vassart, R. W. Doms, M. Parmentier.
1999
. CCR5 binds multiple CC-chemokines: MCP-3 acts as a natural antagonist.
Blood
94
:
1899
.
22
Oppermann, M., M. Mack, A. E. Proudfoot, H. Olbrich.
1999
. Differential effects of CC chemokines on CC chemokine receptor 5 (CCR5) phosphorylation and identification of phosphorylation sites on the CCR5 carboxyl terminus.
J. Biol. Chem.
274
:
8875
.
23
Nibbs, R. J., J. Yang, N. R. Landau, J. H. Mao, G. J. Graham.
1999
. LD78β, a non-allelic variant of human MIP-1α (LD78α), has enhanced receptor interactions and potent HIV suppressive activity.
J. Biol. Chem.
274
:
17478
.
24
Blanpain, C., V. Wittamer, J. M. Vanderwinden, A. Boom, B. Renneboog, B. Lee, E. Le Poul, L. El Asmar, C. Govaerts, G. Vassart, et al
2001
. Palmitoylation of CCR5 is critical for receptor trafficking and efficient activation of intracellular signaling pathways.
J. Biol. Chem.
276
:
23795
.
25
Percherancier, Y., T. Planchenault, A. Valenzuela-Fernandez, J. L. Virelizier, F. Arenzana-Seisdedos, F. Bachelerie.
2001
. Palmitoylation-dependent control of degradation, life span, and membrane expression of the CCR5 receptor.
J. Biol. Chem.
276
:
31936
.
26
Moratz, C., V. H. Kang, K. M. Druey, C. S. Shi, A. Scheschonka, P. M. Murphy, T. Kozasa, J. H. Kehrl.
2000
. Regulator of G protein signaling 1 (RGS1) markedly impairs Gi α signaling responses of B lymphocytes.
J. Immunol.
164
:
1829
.
27
Mellado, M., J. M. Rodriguez-Frade, A. J. Vila-Coro, S. Fernandez, A. Martin de Ana, D. R. Jones, J. L. Toran, A. Martinez.
2001
. Chemokine receptor homo- or heterodimerization activates distinct signaling pathways.
EMBO J.
20
:
2497
.
28
Pashenkov, M., M. Soderstrom, H. Link.
2003
. Secondary lymphoid organ chemokines are elevated in the cerebrospinal fluid during central nervous system inflammation.
J. Neuroimmunol.
135
:
154
.