RANTES and MIP-1α Activate Stats in T Cells*

The chemokines RANTES (regulated on activation, normal T cell expressed and secreted) and MIP (macrophage inflammatory protein)-1α have been implicated in regulating T cell functions. RANTES-induced T cell activation is apparently mediated via two distinct signal transduction cascades: one linked to recruitment of pertussis toxin-sensitive G proteins and the other linked to protein-tyrosine kinase activation. In this report, we identified that the transcription factors Stat1 and Stat3 (for signal transducers and activators of transcription) are rapidly activated in T cells in response to RANTES and MIP-1α. Nuclear extracts from MOLT-4 and Jurkat T cells treated with RANTES or MIP-1α contain tyrosine-phosphorylated Stat1:1 and Stat1:3 dimers that exhibit DNA-binding activity. We demonstrated that RANTES and MIP-1α treatment of Jurkat cells resulted in transcriptional activation of a Stat-inducible gene, c-fos, with kinetics consistent with Stat activation by these chemokines. RANTES and MIP-1α mediate their effects via shared chemokine receptors (CCRs): CCR1, CCR4, and CCR5. Our data revealed a concordance between chemokine-induced Stat activation and c-fos induction and CCR4 and CCR5 expression. These findings indicate that chemokine-mediated activation of G-protein-coupled receptors leads to signal transduction that invokes intracellular phosphorylation intermediates used by other cytokine receptors.

Many cytokines and growth factors mediate their effects via activation of a common signal transduction pathway, the STAT pathway. Binding of the ligand to its specific transmembrane receptor results in receptor aggregation, which may involve single or multiple receptor chains. Receptor aggregation leads to the catalytic activation of receptor-associated cytoplasmic protein-tyrosine kinases, termed janus kinases (JAK) and phosphorylation-activation of latent monomeric signal transducers and activators of transcription, Stat proteins. Six Stat proteins have been identified to date. Receptor-associated phosphorylated Stats then dimerize via SH2-phosphotyrosyl interactions and translocate to the nucleus, where they bind to specific promoter sequences, thereby regulating gene expression (reviewed in Ref. 29). All Stat proteins, with the possible exception of Stat2, differentially bind to more than ten related DNA elements, which fit the consensus TTNNNNNAA (consensus STAT recognition element). Conserved structural motifs within the cytoplasmic domains among the cytokine receptors have been implicated as Jak and STAT recognition sites (30,31). Although Jak-STAT signaling is likely a common feature of all cytokines, recent reports suggest that STAT activation is not necessarily exclusively mediated via Jak association with the cytoplasmic domains of receptors that constitute the cytokine receptor superfamily. Angiotensin II binding to its cognate seven transmembrane, G-protein-coupled receptor, activates Stat1, Stat2, and Stat3 (32)(33)(34).
Apart from their potent chemotactic activities, there is accumulating evidence that CC chemokines are also capable of stimulating T cells in vitro (35)(36)(37). At M concentrations, RANTES-induced signaling in T cells is mediated by at least two distinct signaling cascades: one associated with recruitment of G proteins and the other to protein-tyrosine kinase activation (35). This RANTES-induced tyrosine kinase activation has been functionally linked to T cell proliferation, upregulation of the IL-2 receptor, and the production of cytokines. At these micromolar doses, RANTES will induce the tyrosine kinase activity of the zeta-associated protein (ZAP)-70 and the focal adhesion kinase (FAK) pp125 FAK (36). In the presence of anti-CD3 monoclonal antibody, CC chemokines, at nanomolar doses, exert costimulatory effects on T cells (37). This chemokine enhancement of T cell activation, in combination with T cell receptor-mediated signals, is also associated with proliferation and IL-2 production (37). In this report, we investigated the potential involvement of Stat proteins in chemokine-induced tyrosine phosphorylation in T cells. In a similar manner * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ To whom correspondence should be addressed: to angiotensin II, RANTES and MIP-1␣ can activate chemokine receptor (CCR)-mediated STAT signaling in T cells. Our data indicate that both chemokines, at nanomolar doses, induce the rapid tyrosine phosphorylation and activation of Stat1 and Stat3. In addition, at nanomolar concentrations, both chemokines induce the gene expression of the Stat-inducible protooncogene c-fos.
Cell Extracts-Nuclear extracts were prepared as described previously (38). Briefly, cells were washed twice with ice-cold phosphatebuffered saline that contained 1 mM Na 3 VO 4 and 5 mM NaF and once with hypotonic buffer. Following incubation for 10 min in hypotonic buffer at 10 8 cells/ml, supplemented with 0.2% Triton X-100, cells were disrupted by repeated passage through a 25-gauge needle and centrifuged at 12,000 ϫ g for 20 s. The pellet was incubated in high salt buffer at 2.5 ϫ 10 8 cells/ml for 30 min and clarified by centrifugation at 12,000 ϫ g for 20 min, and the supernatant was supplemented with 0.05% Triton X-100. Nuclear fractions that yielded 4.5-6.7 g of protein/10 6 cells, based on the Bradford method for protein determination (Bio-Rad Labs., CA.) were aliquoted and stored at Ϫ70°C. Hypotonic buffer contained 12 mM Hepes (pH 7.9), 4 mM Tris (pH 7.9), 0.6 mM EDTA, 10 mM KCl, 5 mM MgCl 2 , 1 mM Na 3 VO 4 , 1 mM Na 4 P 2 O 7 , 1 mM NaF, 0.6 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 10 mg/ml aprotinin, 2 mg/ml leupeptin, and 2 mg/ml pepstatin A. The buffer that contained 300 mM KCl and 20% glycerol constituted high salt buffer.
Oligonucleotides-Double-stranded oligodeoxynucleotides, representing the sis-inducing element (SIE) of the c-fos promoter and a mutant IFN-stimulated response element (ISRE), were synthesized. The sequences are: SIE, 5Ј-ATTTCCCGTAAATCCC-3Ј, and mutant ISRE, 5Ј-CCTTCTGAGGCCACTAGAGCA-3Ј. These oligodeoxynucleotides were synthesized with SalI-compatible linkers at the 5Ј terminus (TCGAC). Gel-purified oligonucleotides were mixed with their respective complements, heated to 65°for 15 min, and annealed at room temperature for 18 h. Double-stranded elements were used directly in competition experiments.
Antibodies-Monoclonal antibodies against Stat1 and Stat3 were purchased from Transduction Laboratories, KY. Anti-phosphotyrosine (Tyr(P)) monoclonal antibody, clone 4G10 was purchased from Upstate Biotechnology Inc., NY. For supershift studies, polyclonal antisera against Stat1 and Stat2 were a gift from C. Schindler (Columbia University College of Physicians & Surgeons, NY) and Stat3 antiserum was a gift from D. Levy (NYU School of Medicine). Polyclonal antisera against CCR1 was a gift from R. Horuk, Berlex Biosciences, CA. Preimmune rabbit IgG (ICN Biomedicals, Inc., CA.) was used as a control Incubation with either medium alone or secondary and tertiary reagents alone resulted in superimposable negative cytograms, represented as the profile in panel A and positive cytogram (ϩCCR1 antisera) represented in panel B. MOLT-4 CCR1 expression was likewise confirmed (data not shown).

FIG. 2. Identification of RANTES-inducible STAT complexes by GDAC.
Actively growing MOLT-4 cells were incubated with or without RANTES (6 nM). Nuclear extracts were prepared and analyzed for DNA-binding STAT complexes using GDAC. Eluates from genomic DNA were resolved by SDS-PAGE (7%) and analyzed by Western blotting. Blots were probed with antisera to Stat1, Stat2, Stat3, Stat5, and Stat6. Lysates from whole cell extracts from fibroblast (FL) and Jurkat (JL) cells that had not undergone GDAC served as positive controls, as did a nucelar extract from IFN-Con 1 -treated MOLT-4 cells that had undergone GDAC. As indicated, a nuclear extract from untreated MOLT-4 cells that had undergone GDAC served as the negative control.
for immunoprecipitation.
Mobility Shift Assay-10 g of nuclear extract from untreated or chemokine-treated cells were analyzed using the electrophoretic mobility shift assay (EMSA), by a modification of the procedure described previously (41). Briefly, extracts were incubated with or without doublestranded oligodeoxynucleotides corresponding to the c-fos SIE or a mutant SIE, in the presence of 1.5 g of poly(dI-dC)⅐poly(dI-dC), in EMSA buffer for 30 min at room temperature (final volume 30 ml). Protein-DNA complexes were resolved on a 4.5% polyacrylamide gel using 0.5 ϫ Tris-borate-EDTA as running buffer. For supershift experiments, 1.0 l of polyclonal antisera to Stat1, Stat2, Stat3, anti-phosphotyrosine (4G10), or preimmune sera were incubated with protein extracts for 30 min at 4°prior to the addition of DNA. EMSA buffer contained 13 mM Hepes (pH 7.9), 65 mM NaCl, 0.15 mM EDTA (pH 8.0), 0.06 mM EGTA (pH 8.0), 1.0 mM dithiothreitol, and 5% Ficoll.
RNA Purification, Gel Electrophoresis and Northern Hybridization-Poly(A) ϩ RNA extraction and Northern hybridization procedures have been described elsewhere (42). A 1.3 kbp v-fos cDNA insert in plasmid pFBH-1 was used.

RESULTS AND DISCUSSION
Chemokine-inducible Stat Activation-RANTES and MIP-1␣ both bind to the CC chemokine receptors designated CCR1, CCR4, and CCR5. Target T cells for study were chosen initially based on CCR1 expression determined by flow cytometric analysis of anti-CCR1 antibody binding to native CCR1 on cells (Fig. 1). The lack of availability of specific antibodies for the shared receptors CCR4 and CCR5 precluded identification of their cell surface expression. Our analyses revealed that both the MOLT-4 and Jurkat T cell lines express CCR1.
Subsequently, we undertook studies to examine whether, in a similar manner to angiotensin II, RANTES can activate Stats in MOLT-4 cells. We employed a procedure that we have developed, GDAC, to assay for chemokine-inducible Stat activation (40). GDAC does not require prior knowledge of target DNA elements. Briefly, nuclear extracts from RANTES-treated cells were mixed with genomic DNA bound to cellulose. The mixture was allowed to equilibrate, following which DNA-binding complexes were eluted in high salt buffer. Eluted fractions were resolved by SDS-PAGE, and Stat proteins were detected using anti-Stat immunoblots. Using this procedure, we identified that RANTES induces Stat1-and Stat3-containing DNAbinding complexes in MOLT-4 cells within 30 min (Fig. 2). Stat2, Stat5, and Stat6 were not detected in the DNA-binding complexes.
Since both RANTES and MIP-1␣ bind to CCR1 (and CCR4 and CCR5), we reasoned that MIP-1␣ might also invoke Stat1 and Stat3 activation in MOLT-4 cells. All combinations of homo-and heterodimers of Stat1 and Stat3 will bind to the high affinity c-fos SIE recognition element, m67 (38). Accordingly, we examined the kinetics of activation of both RANTES-and possibly MIP-1␣-inducible Stat1-and Stat3-containing complexes in a standard mobility shift assay. MOLT-4 cells were treated with MIP-1␣ for varying times (15 min to 2 h), then nuclear extracts were analyzed in a gel mobility shift assay. As shown in Fig. 3, MIP-1␣ rapidly induced SIE-binding activities in MOLT-4 cells, within 15 min, that were no longer detectable 2 h after treatment. Similarly, RANTES and MIP-1␣ rapidly induced SIE-binding activities in Jurkat cells, by 15 min, that were likewise no longer detectable after 2 h (Fig. 3).
The apparent contradiction in kinetics of chemokine-induced Stat activation observed using the two different approaches, GDAC and gel mobility shift assay, may be attributable to the differences in DNA recognition elements employed in the two procedures. Specifically, GDAC allows for detection of STAT complexes that bind with relatively high affinity to any number of different target genomic DNA elements. Moreover, GDAC allows for DNA recognition by STAT complexes in the context of any potential accessory factors that are also present in the nuclear extract and that may be involved in DNA-binding. By contrast, gel mobility shift assays invoke the use of a specified DNA target element, in this case the SIE, thereby restricting the scope of STAT complex binding. The delay in chemokineinduced Stat activation observed by GDAC compared with the gel shift assay, may reflect both the low abundance of activated STAT complexes in the cell extracts and the low abundance of SIE-like recognition elements in the genomic DNA used for GDAC. With increased time, chemokine-induced STAT complexes will accumulate, enhancing the likelihood of detection by GDAC. GDAC identification of DNA-binding Stat-containing complexes in 2-h-induced nuclear extracts may reflect STAT complexes that are induced with slower kinetics and that recognize DNA elements distinct from the SIE element employed in the gel shift assay. Moreover, these Stat-containing complexes may be associated with DNA-binding adapter proteins.
We confirmed the specificity of RANTES-and MIP-1␣-induced SIE-binding activities using unlabeled competitor SIE DNA and a nonspecific oligonucleotide element (a mutant interferon-stimulated response element) in the gel shift assays (Fig. 4). Specifically, both RANTES and MIP-1␣-inducible specific SIE-binding activities are present in nuclear extracts from both MOLT-4 and Jurkat cells. Using anti-Stat antibodies in a gel mobility supershift assay, we observed that anti-Stat1 antibodies recognized both SIE-binding complexes in chemokinetreated nuclear extracts (Fig. 5, A and B), whereas anti-Stat3 antibodies only recognized one complex (Fig. 5A). We infer that the chemokine-induced SIE binding activities correspond to the STAT complexes Stat1:1 and Stat1:3. The mobilities of antibody-supershifted STAT-SIE complexes vary according to the charge and size of the resultant complexes. Inclusion of antisera to Stat2 resulted in the appearance of a slow migrating band in both untreated (data not shown) and chemokinetreated cells. The mobilities of the chemokine-induced STAT-DNA complexes were, however, unaffected by anti-Stat2 antisera. We infer that Stat2 is not a constituent of the chemokineinduced STAT complexes and that there are constituents in the anti-Stat2 antisera that interact with DNA-binding factors in cells to invoke a non-Stat-specific SIE-containing complex. Additionally, the Stat1:1 and Stat1:3 complexes may be supershifted with anti-phosphotyrosine antibody 4G10, confirming that, in common with other cytokine-induced STAT complexes, the chemokine-induced STAT complexes are phosphorylated on tyrosine residues (Fig. 5C).
Chemokine-inducible c-fos Gene Expression-To address the biological consequence of chemokine-inducible Stat activation, we examined whether chemokine treatment leads to the transcriptional regulation of a Stat-inducible gene. The promoter of the proto-oncogene c-fos contains a Stat binding DNA element (43). Indeed, a modified version of the SIE of the c-fos promoter was used as the recognition element for Stat1-and Stat3containing STAT complexes in our gel shift assays.
Chemokine receptor expression is tightly regulated; we have observed that both Jurkat and MOLT-4 cells transiently express chemokine receptors. We have shown that PMA treatment for 16 -18 h induces the gene expression of chemokine receptors (data not shown), without affecting Stat activation. Accordingly, PMA-treated Jurkat cells were exposed to RAN-TES or MIP-1␣ in time course studies, then poly(A) ϩ RNA extracted and probed for c-fos gene expression by Northern hybridization. Our results, shown in Fig. 6A, reveal that both RANTES and MIP-1␣ induce c-fos gene expression within 2 h, which is absent by 6 h. The kinetics of c-fos induction are consistent with the kinetics of chemokine-inducible Stat activation. Interestingly, cyclohexamide treatment of cells 1 h prior to chemokine treatment lead to a superinduction of c-fos gene expression at 6 h. These data imply that c-fos gene expression is under the control of a protein synthesis-dependent pathway.
RANTES and MIP-1␣ mediate their effects by shared receptors: CCR1, CCR4, and CCR5. We conducted experiments to determine the concordance between chemokine-induced c-fos gene induction and expression of specific species of CCRs. The data in Fig. 6B show a correlation between CCR4 and CCR5 gene expression and chemokine-inducible c-fos gene expression.
Examination of the intracellular loops of CCR1 reveals a putative Stat3 recognition sequence, YRLQ (residues 311-314) (31). Close examination of this conserved region within the carboxyl-terminal intracellular loop of the different CCRs reveals a similar Stat recognition motif in CCR4 (YILQ), as well as conserved tyrosine residues in the other receptors. More- over, a highly conserved motif in the second intracellular loop of the eight CCRs contains a tyrosine residue in all but CCR7: DRYI/LAI/VVH/Q. Interestingly, this tyrosine-containing motif is present in the angiotensin II receptor that is associated with ligand-induced Stat activation. Furthermore, angiotensin II induces similar Stat complexes as RANTES and MIP-1␣ (33). Viewed together, these data suggest that chemokine-induced signal transduction may be mediated, in part, via ligand activation of intracellular domains of the CCR associated with Stat activation.
Antigen-independent activation of T cells by cytokines may be important for recruiting effector T cells at the site of an immune response and in maintaining the clonal size of memory T cells in the absence of antigenic stimulation (28). Although there is evidence to suggest that RANTES and MIP-1␣ can costimulate T cell activation at nanomolar doses (37), the evidence to date for antigen-independent CC chemokine-induced signaling in T cells, mediated by protein-tyrosine kinase activation, is restricted to RANTES (35). In this study, micromolar concentrations of RANTES induced T cell activation, characterized by the up-regulation of the IL-2 receptor ␣ chain, IL-2 and IL-5, and T cell proliferation. Our data indicate that, at nanomolar concentrations, RANTES and MIP-1␣ will trigger protein-tyrosine kinase activation, which in this instance is associated with Stat activation. The biological consequence of this Stat activation in terms of T cell functions remains to be elucidated. It is likely that the protein-tyrosine kinases associated with RANTES-induced T cell proliferation are distinct from those associated with RANTES-and MIP-1␣-induced Stat activation. Additionally, subtle structural differences among the receptors that mediate chemokine-induced protein-tyrosine kinase activation may define which kinases are activated and hence which signaling pathways are invoked. The existence of multiple signaling pathways associated with tyrosine-phosphorylated intermediates suggests a complexity related to regulation of T cell functions.
This report represents original findings with regard to chemokine activation of Stat signaling pathways. Based on our GDAC studies, it is intriguing to speculate that chemokineinduced Stat1-and Stat3-containing DNA-binding complexes may accumulate in T cells and bind to DNA elements that are distinct from the consensus STAT recognition element, TT-NNNNNAA. Such novel promoter elements might provide the basis for grouping gene families, thereby identifying potential gene targets of chemokine action. The immediate challenge, however, is to determine the kinases that mediate Stat phosphorylation-activation in this system and the role of G-proteins in Stat activation. These studies are currently in progress.