Interferon-inducible T cell alpha chemoattractant (I-TAC): a novel non-ELR CXC chemokine with potent activity on activated T cells through selective high affinity binding to CXCR3

Abstract

Chemokines are essential mediators of normal leukocyte trafficking as well as of leukocyte recruitment during inflammation. We describe here a novel non-ELR CXC chemokine identified through sequence analysis of cDNAs derived from cytokine-activated primary human astrocytes. This novel chemokine, referred to as I-TAC (interferon-inducible T cell alpha chemoattractant), is regulated by interferon (IFN) and has potent chemoattractant activity for interleukin (IL)-2-activated T cells, but not for freshly isolated unstimulated T cells, neutrophils, or monocytes. I-TAC interacts selectively with CXCR3, which is the receptor for two other IFN-inducible chemokines, the IFN-gamma-inducible 10-kD protein (IP-10) and IFN-gamma- induced human monokine (HuMig), but with a significantly higher affinity. In addition, higher potency and efficacy of I-TAC over IP-10 and HuMig is demonstrated by transient mobilization of intracellular calcium as well as chemotactic migration in both activated T cells and transfected cell lines expressing CXCR3. Stimulation of astrocytes with IFN-gamma and IL-1 together results in an approximately 400,000-fold increase in I-TAC mRNA expression, whereas stimulating monocytes with either of the cytokines alone or in combination results in only a 100-fold increase in the level of I-TAC transcript. Moderate expression is also observed in pancreas, lung, thymus, and spleen. The high level of expression in IFN- and IL-1-stimulated astrocytes suggests that I-TAC could be a major chemoattractant for effector T cells involved in the pathophysiology of neuroinflammatory disorders, although I-TAC may also play a role in the migration of activated T cells during IFN-dominated immune responses.

Figure 1

Nucleotide sequence and predicted amino acid sequence of human I-TAC. The deduced amino acid sequence is indicated in single letter codes below the cDNA sequence, with the stop codon indicated by an asterisk. The putative signal peptide is underlined. The polyadenylation signal is of a rare type and is marked with a double underline. The sequence is available from EMBL/GenBank/DDBJ under accession number AF030514.

Figure 2

Sequence alignment and phylogenetic relationship of I-TAC with other known CXC chemokines. (A) Amino acid alignment of I-TAC with CXC chemokines. The alignment was generated with the program Megalign (DNA Star, Madison, WI) using the Clustal method and then manually aligned to obtain the maximum amino acid similarity. Amino acid residues conserved in all sequences are boxed and shaded. Unshaded boxed regions indicate amino acid residues conserved between I-TAC and at least two other CXC chemokines. (B) Phylogenetic tree showing the relationships between I-TAC and known CXC chemokines. The alignment above was used to generate a phylogenetic tree by the Megalign program and the extent of similarity between sequences is indicated below the phylogenetic tree.

Figure 2

Sequence alignment and phylogenetic relationship of I-TAC with other known CXC chemokines. (A) Amino acid alignment of I-TAC with CXC chemokines. The alignment was generated with the program Megalign (DNA Star, Madison, WI) using the Clustal method and then manually aligned to obtain the maximum amino acid similarity. Amino acid residues conserved in all sequences are boxed and shaded. Unshaded boxed regions indicate amino acid residues conserved between I-TAC and at least two other CXC chemokines. (B) Phylogenetic tree showing the relationships between I-TAC and known CXC chemokines. The alignment above was used to generate a phylogenetic tree by the Megalign program and the extent of similarity between sequences is indicated below the phylogenetic tree.

Figure 3

Northern blot analysis of the tissue distribution of I-TAC. Multiple Tissue Northern Blot filters (Clontech), containing 2 μg poly(A)⁺ mRNA/lane, were hybridized with a ³²P-labeled human I-TAC cDNA probe. Conditions for hybridization and subsequent washings were as indicated in Materials and Methods. Autoradiography was performed at −70°C with intensifying screen for 24–72 h. Migration of molecular weight markers in kilobases are shown between the blots.

Figure 4

Chemotactic response of activated T cells to I-TAC. PHA-stimulated T cells grown in the presence of IL-2 for 10–15 d were used in chemotaxis assays as described in Materials and Methods. Each assay was done in triplicate and the cells migrating in response to the indicated concentration of chemokine were scored in five to eight high powered fields (HPF). Error bars were determined by first calculating the average number of cells in the HPF and then determining the standard deviation within the three different wells.

Figure 5

Mobilization of intracellular calcium in activated T cells. (A) Concentration-dependent [Ca²⁺]_i changes induced by I-TAC in IL-2–stimulated T cell blasts. I-TAC was added at 50, 10, 5, and 0.5 nM to INDO-1–loaded cells and [Ca²⁺]_i-dependent fluorescence changes were recorded. The data shown is from one representative experiment of at least four separate experiments using cells from different donors. (B) Dose- response curves of I-TAC, IP-10, and HuMig on IL-2–stimulated T cell blasts. Chemokine was added at 0.25, 0.5, 5, 10, 50, and 500 nM to INDO-1–loaded cells and intracellular [Ca²⁺] changes were monitored as above. These data are compiled from two separate experiments using cells from different donors.

Figure 6

Cross-desensitization of IL-2–stimulated T cell blasts. Chemokine concentrations used were 1,000 nM for IP-10, 500 nM for HuMig, and 50 nM for I-TAC. INDO-1–loaded cells were exposed sequentially to the indicated chemokines. These data are from one representative experiment of four separate experiments using cells from different donors.

Figure 7

Chemotaxis of stable CXCR3 transfectants. 300-19 transfectants expressing CXCR3 (clone MLRA-A5) were challenged with I-TAC (squares), HuMig (circles), and IP-10 (diamonds). Chemotaxis was done as previously described (15). The incubation time for chemotaxis was 2 h. I-TAC responses to nontransfected 300-19 cells are also shown (triangles).

Figure 8

Mobilization of intracellular calcium in stable CXCR3 transfectants. Concentration-dependent [Ca²⁺]_i changes induced by I-TAC, HuMig, or IP-10 on INDO-1–loaded 300-19/CXCR3 transfectants were determined as described above. Doses were as indicated.

Figure 9

Cross-desensitization of 300-19/CXCR3 transfectants. Chemokine concentrations used were 1,000 nM for IP-10, 500 nM for HuMig, and 50 nM for I-TAC. INDO-1–loaded cells were exposed sequentially to the indicated chemokines, and intracellular [Ca²⁺] changes were monitored as above.

Figure 10

I-TAC binding to HEK293 cells stably expressing CXCR3. CXCR3 transfectants were incubated with radiolabeled I-TAC and various concentrations of unlabeled chemokines. Radiolabeled I-TAC bound to cells was determined as indicated in Materials and Methods. The top panel shows the displacement of radiolabeled I-TAC by unlabeled I-TAC, and the Scatchard analysis used to derive affinities for the two binding sites. The lower panels are the curves showing the displacement of radiolabeled I-TAC by IP-10 and HuMig. The maximum amount of I-TAC bound was 5976 ± 90 cpm. Parental HEK293 cells bound 1551 ± 48 cpm.