Chemokines and chemokine receptors play critical roles in directing the migration of alloreactive donor T cells into graft-vs-host disease (GVHD) target organs. However, blockade of GVHD by antagonist Ab against chemokine receptors remains an elusive goal. Using a mouse model of human GVHD, we demonstrate that in vivo administration of anti-CXCR3 Ab for 21 days (long-term), but not for 7 days (short-term), inhibits alloreactive CD8+ T cell-mediated GVHD. During a graft-vs-host reaction, infused donor CD8+ T cells generate two subsets of potent inducers of GVHD: CXCR3+CD8+ and CXCR3CD8+ T cells. Compared with CXCR3+CD8+ T cells, CXCR3CD8+ T cells produce less granzyme B, Fas ligand, IFN-γ, and TNF-α. Interestingly, stimulation with either dendritic cells or IL-2 induces a dynamic conversion between CXCR3+CD8+ and CXCR3CD8+ T cells. Short-term anti-CXCR3 Ab treatment inhibits only CXCR3+CD8+ T cell-mediated GVHD, but not the disease induced by CXCR3CD8+ T cells. Prolonged in vivo administration of anti-CXCR3 Ab significantly reduces the infiltration of alloreactive CD8+ T cells into GVHD target organs and inhibits GVHD mediated by either CXCR3+CD8+ or CXCR3CD8+ T cells. Thus, we have established a novel and effective approach with the potential to give rise to new clinical methods for preventing and treating GVHD after allogeneic hematopoietic stem cell transplantation.

Graft-vs-host disease (GVHD)4 remains the major toxic outcome of allogeneic hematopoietic stem cell transplantation (HSCT) (1, 2, 3). In mouse models of human GVHD directed against minor histocompatibility Ags (miHAs), infused donor T cells recognize miHAs presented by APCs and are rapidly activated to proliferate and differentiate into alloreactive effector T cells (4, 5, 6, 7, 8). These alloreactive effector T cells are then recruited into the skin, intestine, and liver, mediating the host tissue damage and subsequently the disease (2, 9, 10, 11). It is well understood that chemokines direct the migration and infiltration of T cells via their cognate chemokine receptors (9, 12, 13). Previous studies have demonstrated that the expression of CCR1, CCR2, CCR5, and CCR7 on peripheral blood T cells in patients after allogeneic HSCT is associated with human GVHD (14, 15). In experimental mouse GVHD models, donor T cells lacking CCR10 or CCR2 have reduced ability to infiltrate into GVHD target organs and reduced capability to cause severe GVHD as compared with wild-type (Wt) T cells (16, 17). Other studies have demonstrated that donor T cells lacking CCR1 or CCR6 are also much less virulent than Wt T cells in inducing GVHD due to their reduced ability to produce IFN-γ and TNF-α, and to infiltrate the intestine and skin (18, 19). However, GVHD remains the major complication in these models. Thus, multiple chemokine receptors play redundant roles in mediating the migration and tissue infiltration of alloreactive T cells during the GVHD process.

Previous studies indicate that neutralizing Ab against chemokine receptors may be used to reduce GVHD morbidity and mortality. In vivo administration with Ab against CCR5 reduces liver infiltration of alloreactive CD8+ T cells and ameliorates GVHD (20). In contrast, donor T cells lacking CCR5 cause augmented graft-vs-host (GVH) responses as compared with Wt donor T cells, leading to accelerated morbidity (21, 22). These observations raise an important question as to whether CCR5 can be a target for blocking of GVHD. Recently, CXCR3 has been shown to be correlated with GVHD (9, 14, 23). Our studies and those of others have demonstrated that CXCR3 is induced in both effector and effector memory T cells (24, 25, 26, 27, 28). Infusion of donor T cells derived from CXCR3−/− mice causes reduced gastrointestinal tract and liver damage (29). However, it remains to be established whether in vivo administration of CXCR3 antagonist(s) can be used for GVHD prevention and treatment.

Using a well-described allogeneic HSCT mouse model that is closely related to human GVHD directed against miHAs, we investigated the effect of neutralizing anti-CXCR3 Ab on alloreactive CD8+ T cell-mediated GVHD. We found that during GVH reactions, infused donor CD8+ T cells generate two subsets of GVHD potent inducers: CXCR3+CD8+ and CXCR3CD8+ T cells. Although short-term anti-CXCR3 Ab treatment for 7 days is sufficient to inhibit CXCR3+CD8+ T cell-mediated GVHD, prolonged treatment with anti-CXCR3 Ab for 21 days is necessary for effectively blocking CXCR3CD8+ T cell-mediated GVHD. In response to dendritic cells (DCs) or IL-2 stimulation, dynamic conversion occurs between these CXCR3+CD8+ and CXCR3CD8+ T cells. These findings have significant implications for further development of new clinical approaches for prevention and treatment of GVHD by the targeting of chemokine receptor(s) of alloreactive T cells.

B6/SJL (H-2Db, CD45.1+) and C3H.SW (H-2Db, CD45.2+ and Ly9.1+) mice were purchased from Jackson ImmunoResearch Laboratories. Animals were kept in a specific pathogen-free facility at the Chinese Academy of Sciences. Animal care and use were in compliance with institutional guidelines. Drinking water for bone marrow transplant recipients was supplemented with neomycin sulfate and polymyxin B (Sigma-Aldrich), as previously described (30).

All recombinant cytokines, including IL-2, IL-4, IL-7, IL-15, GM-CSF, TNF-α, and stem cell factor; chemokine rCXCL10; and PE-conjugated anti-CXCR3 Ab (220803) were purchased from R&D Systems. Microbead-conjugated Abs were purchased from Miltenyi Biotec. All other Abs used for immunofluorescence staining were obtained from BD Pharmingen. MHC-I Ab (28-14-8), rabbit anti-mouse CD8 Ab, and CTLA-4-Ig fusion protein were provided by Y. Guo (Second Military Medical University, Shanghai, China). B6 mouse-derived EL-4 leukemic cells (H-2Db) were obtained from American Type Culture Collection. EL-4 cells were grown in RPMI 1640 (Life Technologies) containing 10% FBS (Life Technologies), penicillin G (100 U/ml), and streptomycin (100 μg/ml) as complete medium.

T cell-depleted bone marrow cells (TCD BM) were prepared by depleting CD4+ and CD8+ T cells using CD4 and CD8 microbeads, followed by magnetic separation, as previously described (6, 30). CD8+ T cells were purified from spleens, lymph nodes, and/or livers by magnetic selection with CD8 microbeads. In some experiments, these CD8+ T cells were further sorted for CD44lowCD62Lhigh naive CD8+ T cells by FACS (FACSAria; BD Biosciences), as previously described (30, 31). Mature DCs were prepared, as previously described (32). Briefly, c-Kit+ hematopoietic cells were isolated by FACS from B6/SJL bone marrow and cultured for 6 days in complete medium with stem cell factor (10 ng/ml), GM-CSF (25 ng/ml), and IL-4 (10 ng/ml). CD11c+ cells were magnetically sorted from this 6-day culture using anti-CD11c Ab-conjugated microbeads and stimulated with GM-CSF (10 ng/ml) plus TNF-α (50 ng/ml) for an additional 2–3 days to induce mature DCs (B6/SJL DCs).

The purity of sorted cells in this study was consistently more than 98%, as revealed by immunofluorescence analysis.

Lethally irradiated mice underwent allogeneic bone marrow transplantation, as previously described (30). Briefly, B6/SJL mice were irradiated with 9.5 Gy administered in three fractions from a 137Cs source (MDS Nordion). C3H.SW CD8+ T cells (4 × 105 to 2 × 106) or C3H.SW CD44lowCD62Lhigh naive CD8+ T cells (2 × 106) plus C3H.SW TCD BM (5 × 106) were transplanted, via the tail vein, into the irradiated B6/SJL mice. In some experiments, anti-CXCR3 Ab was administered to B6/SJL recipient mice after transplantation via the tail vein at days 0, 3, and 7 (short-term) or at days 0, 3, 7, 10, 14, 18, and 21 (long-term). Recipient mice were monitored for clinical signs of GVHD and survival, as previously established (31, 33). Mice were sacrificed, and specimens of skin, intestine, and liver were taken for histopathologic assessment of GVHD. Slides were coded without reference to mouse type and prior treatment status and examined systematically by a pathologist (34, 35). To analyze the therapeutic effect of anti-CXCR3 Ab on GVHD, anti-CXCR3 Ab was administered to B6/SJL recipient mice on days 7, 10, 14, 16, 20, 22, 24, and 27 after transplantation. For adoptive transfer assay, donor-derived CD8+ T cells were isolated from the spleens and livers of B6/SJL mice receiving C3H.SW CD8+ T cells plus TCD BM on day 14 after transplantation. The cells were then separated into CXCR3CD8+ and CXCR3+CD8+ T cell populations by FACS and adoptively transferred into secondary B6/SJL recipient mice accompanied by TCD BM.

Donor-derived CXCR3CD8+ or CXCR3+CD8+ T cells were recovered from recipient mice after transplantation, and stimulated in complete medium for 3–5 days with B6/SJL DCs (T cell/DC ratio = 3:1) plus cytokine mixture, including IL-2 (5 ng/ml), IL-7 (3 ng/ml), and IL-15 (3 ng/ml). In some experiments, these T cells were stimulated with B6/SJL DCs, IL-2, IL-7, IL-15, or medium alone, and with B6/SJL DCs in the presence of CTLA-4-Ig, or neutralizing Ab to MHC-I or IL-2, with IgG as control. For the protein recycling assay, the stimulated cells with B6/SJL DCs or IL-2 were collected, washed twice, and reincubated for 1 h in complete medium to redistribute the internalized protein on the cell surface (36). The cells were then analyzed for cell surface or total levels of CXCR3 using surface or intracellular staining with anti-CXCR3 Ab and FACS.

Donor-derived CXCR3CD8+ or CXCR3+CD8+ T cells (1 × 105 cells/well) in 96-well U-bottom plates with complete medium were stimulated with B6/SJL DCs (1 × 104 cells/well) that had been exposed to 20 Gy of γ-ray irradiation. Cells were cultured for 96 h, and 0.5 μCi/well [3H]thymidine was added 18 h before the end of culture. The cells were then harvested onto glass fiber mats for measurement of [3H]thymidine incorporation based upon scintillation counting.

CFSE labeling was performed, as previously described (27). In brief, freshly isolated donor-derived CD8+ T cells were labeled with 2.5 μM CFSE (Molecular Probes). These CFSE-labeled cells were then transplanted into irradiated B6/SJL mice together with C3H.SW TCD BM or syngeneic B6/SJL TCD BM. The cells infused in spleen and liver of recipients were harvested at the indicated time after transplantation for FACS analysis. In some experiments, the expression of CXCR3 was analyzed on donor-derived CD8+ T cells by gating on the CFSE+ and/or CD45.2+ cells to distinguish from the de novo thymus-developed naive CD8+ T cells.

Donor-derived CD8+ T cells were freshly isolated and incubated with EL-4 cells in serum-free medium ex vivo 15 (BioWhittaker) for 12 h. The supernatant from each well was then collected for measurement of released lactate dehydrogenase using a nonradioactive cytotoxicity detection kit, according to manufacturer’s instructions (Promega).

The CDC assay was described previously (37, 38). In brief, CXCR3+CD8+ T cells at 1 × 106/ml were incubated with anti-CXCR3 Ab (dose ranging from 0.08 to 10.0 μg/ml) together with 2% normal rabbit serum without heat inactivation as a source of complement, and with anti-CD8 Ab (10.0 μg/ml) as control. After 1-h incubation at 37°C, cells were stained with propidium iodide (PI) for flow cytometric analysis. CDC activity was calculated by dividing the number of PI-positive cells by the total number of input T cells. Negligible cell loss of up to 5% as determined by trypan blue exclusion and cell counting was observed with various Abs.

The chemotaxis analysis in vitro was performed using a commercially available 96-well modified Boyden chamber chemotaxis system (ChemoTx; NeuroProbe), as previously described (39). In brief, 30 μl of chemotaxis medium (RPMI 1640, 0.1% BSA) containing CXCL10 (50 ng/ml) was added to the lower chamber, and CXCR3+CD8+ T cells (5 × 104 cells in 50 μl) were placed into the upper chamber, separated from the lower by a membrane. Chemotaxis was then allowed to proceed at 37°C in a 5% CO2 incubator. To examine the effect of anti-CXCR3 Ab, CXCR3+CD8+ T cells were incubated with a range of anti-CXCR3 Ab concentrations (0.4–50 μg/ml) for 30 min before and during the chemotaxis assay. Three hours later, cells in the lower chamber were recovered and enumerated. The ratio of migrating cells was calculated by dividing the number of cells in the lower chamber by the total input cells, which were delivered onto the upper chamber before migration.

Total RNA was extracted from the sorted donor CD8+ T cell subsets using the guanidium thiocyanate-phenol-chloroform method modified for TRIzol (Invitrogen Life Technologies). cDNA of all tested chemokine receptors was quantified through the quantitative real-time PCR technique, all with GAPDH as control. Quantitative real-time PCR was performed using a SYBR Green PCR mix in an ABI Prism 7900HT (Applied Biosystems). Thermocycler conditions included an initial holding at 50°C for 2 min, then 95°C for 10 min; this was followed by a two-step PCR program, as follows: 95°C for 15 s and 60°C for 60 s for 40 cycles. Data were collected and quantitatively analyzed on an ABI PRISM 7900 sequence detection system (Applied Biosystems). The primer sequences were designed using the Primer Express Software version 2.0 provided with the ABI Prism 7900HT (Table I). The GAPDH gene was used as an endogenous control to normalize for differences in the amount of total RNA in each sample. All values were expressed as fold increase or decrease relative to the expression of GAPDH. The mean value of the replicates for each sample was calculated and expressed as cycle threshold (CT; cycle number at which each PCR reaches a predetermined fluorescence threshold, set within the linear range of all reactions). The amount of gene expression was then calculated as the difference (ΔCT) between the CT value of the sample for the target gene and the mean CT value of that sample for the endogenous control (GAPDH). Relative expression was calculated as the difference (ΔΔCT) between the ΔCT values of the test sample and of the control sample. Relative expression of genes of interest was calculated and expressed as 2−ΔΔCT.

Table I.

Oligonucleotides used in quantitative real-time PCR

GenePrimer (5′→3′)
CCR2 Forward AAGGGTCACAGGATTAGGAAGGT 
 Reverse CACGGCATAATCATAACGTTCTG 
CCR3 Forward TTGCAGGACTGGCAGCATT 
 Reverse TGTCTTCTTCACCCTCTGGATAG 
CCR4 Forward GTCACAGACACCACCCAGGAT 
 Reverse TCCCCAAATGCCTTGATACC 
CCR5 Forward GGTGGAGGAGCAGGGACAA 
 Reverse TGTGTCCGTCCCTTTGCAT 
CCR6 Forward ACGAGGAGGACCATGTTGTGA 
 Reverse CAGGCCCAGAACTCCAAGAG 
CCR7 Forward CTACAGCGGCCTCCAGAAGA 
 Reverse CCATCTGGGCCACTTGGAT 
CCR8 Forward TGCCTCGATGGATTACACGAT 
 Reverse TGCCCCTGAGGAGGAACTCT 
CCR10 Forward GGGTACGATGAGGAGGCCTAT 
 Reverse AGGGAGACACTGGGTTGGAA 
CXCR2 Forward GCCATTGTACATGCCACAAGTAC 
 Reverse ATGGGCAGGGCCAGAATTAC 
CXCR3 Forward GCACCAGCCAAGCCATGTA 
 Reverse ATAATCGTAGGGAGAGGTGCTGTT 
GenePrimer (5′→3′)
CCR2 Forward AAGGGTCACAGGATTAGGAAGGT 
 Reverse CACGGCATAATCATAACGTTCTG 
CCR3 Forward TTGCAGGACTGGCAGCATT 
 Reverse TGTCTTCTTCACCCTCTGGATAG 
CCR4 Forward GTCACAGACACCACCCAGGAT 
 Reverse TCCCCAAATGCCTTGATACC 
CCR5 Forward GGTGGAGGAGCAGGGACAA 
 Reverse TGTGTCCGTCCCTTTGCAT 
CCR6 Forward ACGAGGAGGACCATGTTGTGA 
 Reverse CAGGCCCAGAACTCCAAGAG 
CCR7 Forward CTACAGCGGCCTCCAGAAGA 
 Reverse CCATCTGGGCCACTTGGAT 
CCR8 Forward TGCCTCGATGGATTACACGAT 
 Reverse TGCCCCTGAGGAGGAACTCT 
CCR10 Forward GGGTACGATGAGGAGGCCTAT 
 Reverse AGGGAGACACTGGGTTGGAA 
CXCR2 Forward GCCATTGTACATGCCACAAGTAC 
 Reverse ATGGGCAGGGCCAGAATTAC 
CXCR3 Forward GCACCAGCCAAGCCATGTA 
 Reverse ATAATCGTAGGGAGAGGTGCTGTT 

The cDNA encoding the N-terminal extracellular portion of CXCR3 was obtained by PCR, using full-length cDNA and a set of oligonucleotides (5′-ATGTACCTTGAGGTTAGTGAACG-3′ and 5′-GAAATCCTGTGGGCAGGGC-3′) as a template and primers. The resulting fragment was digested with XhoI and NotI and subcloned into a GST fusion protein expression vector, p-GEX-4T-1, which was predigested with XhoI and NotI. Expression and purification of the GST fusion protein were performed, as described previously (20).

Two New Zealand White rabbits were immunized with 100 μg of GST fusion protein in CFA (Sigma-Aldrich), eight times at biweekly intervals. One week after the final immunization, rabbits were bled and sera were obtained and fractionated into IgG using a column packed with protein A-agarose (Pharmacia Biotech). A portion of the IgG fraction was further digested with pepsin (Sigma-Aldrich), and the (Fab′)2 fragments were obtained by sequential chromatography using protein A affinity and gel filtration columns, as described previously (20).

Binding specificity and neutralizing function of the generated anti-CXCR3 Ab were examined and verified using FACS and chemotaxis assays.

Cells were harvested, pelleted by centrifugation, and resuspended in lysis buffer. Equal amounts of protein (20 μg) were loaded onto a 5% acrylamide stacking gel and resolved on SDS-PAGE using a 10% separating gel. Following transfer of separated proteins, nitrocellulose membranes were blocked and probed overnight at 4°C with rabbit anti-mouse CXCR3 Ab (H-95; Santa Cruz Biotechnology). The membranes were then probed for 1 h at room temperature with goat anti-rabbit peroxidase-conjugated IgG (Santa Cruz Biotechnology), and the immunoreactivity was detected using chemiluminescence. To quantify CXCR3 proteins, each band density was normalized to that of actin protein.

Leukemia was induced by the injection of 5 × 103 EL-4 leukemic cells into peritoneal cavities of lethally irradiated B6/SJL mice at day 7 after allogeneic bone marrow transplantation, as previously described (31). To examine the effect of anti-CXCR3 Ab, long-term anti-CXCR3 Ab treatment was performed in these EL-4 cell-injected B6/SJL recipient mice, with IgG treatment as control. Survival of the recipients was observed to assay GVL.

Significant differences were evaluated using an independent-samples t test, except that multiple treatment groups were compared within individual experiments by ANOVA or the Kruskal-Wallis test. The Kaplan-Meier (log rank test) was used for survival analysis. Values of p less than 0.05 were considered significant.

To determine whether in vivo blocking of CXCR3 inhibited GVHD directed to miHA, we administered anti-CXCR3 Ab to lethally irradiated B6/SJL (CD45.1) recipient mice receiving C3H.SW (CD45.2) CD8+ T cells plus C3H.SW TCD BM. We found that in vivo administration of recipient mice with anti-CXCR3 Ab from days 0 to 21 after transplantation (termed long-term treatment) inhibited GVHD in B6/SJL mice receiving donor CD8+ T cells plus TCD BM, with all of these mice surviving for over 60 days after transplantation (Fig. 1,A, p < 0.05). Histologic examination showed that long-term anti-CXCR3 Ab treatment inhibited the infiltration of inflammatory cells in the liver, intestine, and skin as compared with that for controls treated with IgG (Fig. 1,B). In contrast, as compared with control IgG-treated B6/SJL recipients, treatment with anti-CXCR3 Ab from days 0 to 7 after transplantation (termed short-term treatment) did not affect the development of GVHD in these B6/SJL recipient mice (Fig. 1). These results indicate that in vivo long-term administration of anti-CXCR3 Ab effectively inhibits GVHD.

FIGURE 1.

Inhibition of CD8+ T cell-mediated GVHD requires prolonged anti-CXCR3 Ab treatment. Lethally irradiated B6/SJL recipient mice (CD45.1+) were i.v. injected with C3H.SW CD8+ T cells (CD45.2+; 2 × 106) plus C3H.SW TCD BM (5 × 106) at day 0. These recipients were treated with anti-CXCR3 Ab for the short-term (i.e., on days 0, 3, and 7) or for the long-term (i.e., on days 0, 3, 7, 10, 14, 18, and 21). The effect of anti-CXCR3 Ab on inhibition donor CD8+ T cell-mediated GVHD was assayed. A, The clinical GVHD signs and survival were monitored over time after transplantation. Symbols represent the following: TCD BM, n = 9 (×); TCD BM plus CD8+ T cells plus control IgG treatment, n = 9 (○); and TCD BM plus CD8+ T cells plus anti-CXCR3 Ab treatment for the short-term, n = 11 (▴) or for the long-term, n = 11 (▪). B, Murine livers, skins, and intestines were collected on day 35 after the aforementioned transplantation and were sectioned for histologic staining with H & E; original magnification ×100. Results are representative of three independent experiments. ∗, p < 0.05.

FIGURE 1.

Inhibition of CD8+ T cell-mediated GVHD requires prolonged anti-CXCR3 Ab treatment. Lethally irradiated B6/SJL recipient mice (CD45.1+) were i.v. injected with C3H.SW CD8+ T cells (CD45.2+; 2 × 106) plus C3H.SW TCD BM (5 × 106) at day 0. These recipients were treated with anti-CXCR3 Ab for the short-term (i.e., on days 0, 3, and 7) or for the long-term (i.e., on days 0, 3, 7, 10, 14, 18, and 21). The effect of anti-CXCR3 Ab on inhibition donor CD8+ T cell-mediated GVHD was assayed. A, The clinical GVHD signs and survival were monitored over time after transplantation. Symbols represent the following: TCD BM, n = 9 (×); TCD BM plus CD8+ T cells plus control IgG treatment, n = 9 (○); and TCD BM plus CD8+ T cells plus anti-CXCR3 Ab treatment for the short-term, n = 11 (▴) or for the long-term, n = 11 (▪). B, Murine livers, skins, and intestines were collected on day 35 after the aforementioned transplantation and were sectioned for histologic staining with H & E; original magnification ×100. Results are representative of three independent experiments. ∗, p < 0.05.

Close modal

We next investigated the mechanism by which long-term treatment with CXCR3 Ab inhibited GVHD. Donor CD44lowCD62LhighCD8+ naive T cells that did not express CXCR3 were highly purified from normal C3H.SW mice (Fig. 2,A). We found that infused donor naive CD8+ T cells began expressing CXCR3 by day 3 after transplantation, and that expression peaked by day 7 (68.0 ± 7.8%), and declined by day 28 (4.3 ± 1.1%). Interestingly, a substantial population of CXCR3+CD8+ T cells increased again by day 35 (22.8 ± 2.2%) after transplantation (Fig. 2,A). In parallel, there were more donor-derived CXCR3CD8+ T cells than CXCR3+CD8+ T cells in the spleens between days 14 and 35, and in the livers between days 7 and 28 after transplantation (Fig. 2, B and C). These CXCR3CD8+ T cells peaked in number by day 14 (21.3 ± 3.3 × 105), and declined to 5.3 ± 0.5 × 105 by day 35 in the spleens (Fig. 2,B). Similarly, they peaked by day 14 (132.0 ± 18.7 × 104) and declined by day 28 (33.1 ± 14.5 × 104) in the livers, with only a slight increase by day 35 (40.0 ± 4.4 × 104; Fig. 2 C). These data suggest that both CXCR3+CD8+ and CXCR3CD8+ T cells are generated during the GVHD process, and that CXCR3CD8+ T cells are the dominant population throughout the disease course.

FIGURE 2.

CXCR3+CD8+ and CXCR3CD8+ T cells are generated throughout the course of GVHD. Splenic and hepatic CD8+ T cells from B6/SJL recipient mice receiving C3H.SW CD44lowCD62Lhigh naive CD8+ T cells plus TCD BM were magnetically isolated at indicated times after transplantation. The frequency of CXCR3 expression on the recovered CD8+ T cells was shown by gating on donor CD45.2+CD8+ cells using FACS (A). The absolute numbers of donor-derived CXCR3+CD8+ and CXCR3CD8+ T cells recovered from the spleens (B) and livers (C) were determined. Data are expressed as means ± SD of six mice for each group. Results represent three independent experiments.

FIGURE 2.

CXCR3+CD8+ and CXCR3CD8+ T cells are generated throughout the course of GVHD. Splenic and hepatic CD8+ T cells from B6/SJL recipient mice receiving C3H.SW CD44lowCD62Lhigh naive CD8+ T cells plus TCD BM were magnetically isolated at indicated times after transplantation. The frequency of CXCR3 expression on the recovered CD8+ T cells was shown by gating on donor CD45.2+CD8+ cells using FACS (A). The absolute numbers of donor-derived CXCR3+CD8+ and CXCR3CD8+ T cells recovered from the spleens (B) and livers (C) were determined. Data are expressed as means ± SD of six mice for each group. Results represent three independent experiments.

Close modal

We next asked whether donor-derived CXCR3CD8+ T cells were able to induce GVHD. Donor-derived CXCR3CD8+ and CXCR3+CD8+ T cells were isolated from B6/SJL mice receiving donor CD8+ T cells plus TCD BM at day 14 after transplantation (termed day 14 CXCR3CD8+ T cells and day 14 CXCR3+CD8+ T cells, respectively). Both day 14 CXCR3+CD8+ and CXCR3CD8+ T cells caused GVHD upon adoptive transfer into secondary B6/SJL recipients (Fig. 3). As few as 4 × 105 of either day 14 CXCR3+CD8+ or day 14 CXCR3CD8+ T cells were sufficient to induce lethal GVHD (Fig. 3). Short-term anti-CXCR3 Ab treatment effectively inhibited the production of GVHD in B6/SJL mice receiving day 14 CXCR3+CD8+ T cells, but not in B6/SJL mice receiving day 14 CXCR3CD8+ T cells (Fig. 4,A). This inhibition of GVHD was accompanied by a reduction in the number of donor-derived CD8+ T cells by day 14 after adoptive transfer in the spleens and livers of secondary B6/SJL mice receiving day 14 CXCR3+CD8+ T cells. This is as compared with IgG-treated controls (11.3 ± 4.2 × 105 vs 30.9 ± 6.2 × 105 in spleen, 3.4 ± 0.5 × 105 vs 8.8 ± 2.4 × 105 in liver; p < 0.05; Fig. 4 B). These data suggest that the generation of alloreactive CXCR3CD8+ T cells accounts for the inability of short-term anti-CXCR3 Ab treatment to inhibit GVHD.

FIGURE 3.

Both CXCR3CD8+ and CXCR3+CD8+ T cell populations are competent to induce GVHD. Donor CD45.2+CD8+ T cells were recovered on day 14 after transplantation from spleen and liver of B6/SJL recipients. These cells were further separated into CXCR3CD8+ and CXCR3+CD8+ T cell subsets. These isolated day 14 CXCR3CD8+ or CXCR3+CD8+ T cells plus C3H.SW TCD BM were adoptively transferred into lethally irradiated secondary B6/SJL recipient mice. Survival was observed over time after the adoptive transfer. Symbols represent the following: TCD BM (×); TCD BM plus 2 × 106 (▪), 8 × 105 (▾), or 4 × 105 (▾) day 14 CXCR3CD8+ T cells; and TCD BM plus 2 × 106 (□), 8 × 105 (▿), or 4 × 105 (⋄) day 14 CXCR3+CD8+ T cells. Results are representative of nine mice for each group of three independent experiments.

FIGURE 3.

Both CXCR3CD8+ and CXCR3+CD8+ T cell populations are competent to induce GVHD. Donor CD45.2+CD8+ T cells were recovered on day 14 after transplantation from spleen and liver of B6/SJL recipients. These cells were further separated into CXCR3CD8+ and CXCR3+CD8+ T cell subsets. These isolated day 14 CXCR3CD8+ or CXCR3+CD8+ T cells plus C3H.SW TCD BM were adoptively transferred into lethally irradiated secondary B6/SJL recipient mice. Survival was observed over time after the adoptive transfer. Symbols represent the following: TCD BM (×); TCD BM plus 2 × 106 (▪), 8 × 105 (▾), or 4 × 105 (▾) day 14 CXCR3CD8+ T cells; and TCD BM plus 2 × 106 (□), 8 × 105 (▿), or 4 × 105 (⋄) day 14 CXCR3+CD8+ T cells. Results are representative of nine mice for each group of three independent experiments.

Close modal
FIGURE 4.

Long-term anti-CXCR3 Ab treatment inhibits CXCR3CD8+ T cell-mediated GVHD. Treatment with anti-CXCR3 Ab for the short-term (A) or long-term (C) resulted in survival times observed over time after adoptive transfer in secondary B6/SJL recipients receiving day 14 CXCR3CD8+ or CXCR3+CD8+ T cells plus TCD BM. Symbols represent the following: TCD BM (×); TCD BM plus day 14 CXCR3CD8+ T cells plus control IgG treatment (•); TCD BM plus day 14 CXCR3CD8+ T cells plus anti-CXCR3 Ab treatment for the short-term (▴) or for the long-term (▪); and TCD BM plus day 14 CXCR3+CD8+ T cells plus treatment with control IgG (○) or with anti-CXCR3 Ab for the short-term (▵). The absolute numbers of donor-derived CD8+ T cells recovered from spleens and livers in these secondary recipients were calculated on day 14 (B) and day 35 (D). Data are expressed as means ± SD. Results are representative of nine mice for each group of three independent experiments. ∗, p < 0.05.

FIGURE 4.

Long-term anti-CXCR3 Ab treatment inhibits CXCR3CD8+ T cell-mediated GVHD. Treatment with anti-CXCR3 Ab for the short-term (A) or long-term (C) resulted in survival times observed over time after adoptive transfer in secondary B6/SJL recipients receiving day 14 CXCR3CD8+ or CXCR3+CD8+ T cells plus TCD BM. Symbols represent the following: TCD BM (×); TCD BM plus day 14 CXCR3CD8+ T cells plus control IgG treatment (•); TCD BM plus day 14 CXCR3CD8+ T cells plus anti-CXCR3 Ab treatment for the short-term (▴) or for the long-term (▪); and TCD BM plus day 14 CXCR3+CD8+ T cells plus treatment with control IgG (○) or with anti-CXCR3 Ab for the short-term (▵). The absolute numbers of donor-derived CD8+ T cells recovered from spleens and livers in these secondary recipients were calculated on day 14 (B) and day 35 (D). Data are expressed as means ± SD. Results are representative of nine mice for each group of three independent experiments. ∗, p < 0.05.

Close modal

Indeed, long-term treatment with anti-CXCR3 Ab effectively inhibited GVHD in these secondary B6/SJL mice receiving day 14 CXCR3CD8+ T cells, with 67% of them surviving as opposed to 22% surviving after short-term anti-CXCR3 Ab treatment (Fig. 4,C, p < 0.05). Furthermore, by day 35 after this adoptive transfer, a significant reduction in donor-derived CD8+ T cells was found in the livers of long-term anti-CXCR3 Ab-treated secondary B6/SJL mice (1.4 ± 0.4 × 105) as compared with those of the short-term anti-CXCR3 Ab-treated group (4.6 ± 1.2 × 105) or the IgG-treated group (7.1 ± 1.8 × 105) (Fig. 4 D). In contrast, donor-derived CD8+ T cells were significantly increased in the spleens of long-term anti-CXCR3 Ab-treated B6/SJL mice as compared with control mice receiving IgG treatment (23.8 ± 5.4 × 105 vs 13.1 ± 2.8 × 105). We found that when compared with control IgG treatment, the long-term anti-CXCR3 Ab treatment increased CD4+ T cells in the spleens of these secondary recipients (data not shown), and affected neither the percentage of CD4+CD25+Foxp3+ regulatory T cells (4.5 ± 1.5% vs 4.8 ± 1.7%), nor the ratio of alloreactive CD44highCD62Llow effector/memory CD8+ T cells vs CD44lowCD62Lhigh naive-like CD8+ T cells (CD44highCD62Llow, 82.6 ± 8.4% vs 87.8 ± 12.4%; CD44lowCD62Lhigh, 2.1 ± 0.7% vs 1.7 ± 0.4%, respectively). These results suggest that long-term anti-CXCR3 Ab treatment is essential to inhibition of the infiltration of alloreactive CD8+ T cells in GVHD target organs.

We further examined the effect of anti-CXCR3 Ab on the survival, proliferation, and migration of alloreactive CD8+ T cells. CFSE-labeled donor day 14 CXCR3+CD8+ T cells were adoptively transferred into lethally irradiated secondary B6/SJL mice. As compared with control IgG treatment, we found that long-term anti-CXCR3 Ab treatment did not affect the proliferation of infused donor CD8+ T cells (Fig. 5,A), but resulted in more donor-derived CD8+ T cells undergoing apoptotic death in the spleens of these recipients between days 1 and 21 after transplantation (Fig. 5,B). Ex vivo chemotaxis assay and CDC assay showed that anti-CXCR3 Ab inhibited the migration of day 14 CXCR3+CD8+ T cells in response to their ligand CXCL10 (Fig. 5,C) and lysed day 14 CXCR3+CD8+ T cells in the presence of fresh complement (Fig. 5 D). Thus, prolonged anti-CXCR3 Ab treatment is essential for inhibition of alloreactive CD8+ T cell-induced GVHD and acts by reducing migration into the GVHD target tissue and perhaps by in vivo killing of CXCR3+CD8+ T cells.

FIGURE 5.

The effect of anti-CXCR3 Ab on alloreactive CD8+ T cells. Treatment with anti-CXCR3 Ab for the long-term or with control IgG resulted in the proliferation (A) and apoptosis (B) of donor-derived CD8+ T cells as analyzed at indicated times after adoptive transfer in secondary B6/SJL recipients receiving day 14 CXCR3+CD8+ T cells plus TCD BM. C, The migratory ability of day 14 CXCR3+CD8+ T cells to respond to CXCL10 (50 ng/ml) was examined in the presence of the indicated dose of anti-CXCR3 Ab. D, CDC to day 14 CXCR3+CD8+ T cells mediated by anti-CXCR3 Ab with 2% normal rabbit serum was observed using PI exclusion assay. Data are expressed as the means ± SD of triplicate cultures. Results are representative of three independent experiments. ∗, p < 0.05.

FIGURE 5.

The effect of anti-CXCR3 Ab on alloreactive CD8+ T cells. Treatment with anti-CXCR3 Ab for the long-term or with control IgG resulted in the proliferation (A) and apoptosis (B) of donor-derived CD8+ T cells as analyzed at indicated times after adoptive transfer in secondary B6/SJL recipients receiving day 14 CXCR3+CD8+ T cells plus TCD BM. C, The migratory ability of day 14 CXCR3+CD8+ T cells to respond to CXCL10 (50 ng/ml) was examined in the presence of the indicated dose of anti-CXCR3 Ab. D, CDC to day 14 CXCR3+CD8+ T cells mediated by anti-CXCR3 Ab with 2% normal rabbit serum was observed using PI exclusion assay. Data are expressed as the means ± SD of triplicate cultures. Results are representative of three independent experiments. ∗, p < 0.05.

Close modal

We next characterized both alloreactive CXCR3+CD8+ and CXCR3CD8+ T cells. Quantitative real-time PCR analysis showed that day 14 CXCR3CD8+ T cells expressed 10-fold less CXCR3 mRNA than did day 14 CXCR3+CD8+ T cells, and expressed equal or significantly lower levels of other chemokine receptors, including CCR2, CCR3, CCR5, CCR8, and CXCR2 (Fig. 6,A). We found that unlike donor CD8+ naive T cells, these day 14 CXCR3CD8+ T cells showed decreased expression of CD127 (IL-7R), but expressed higher levels of CD122, granzyme B, and Fas ligand (FasL) (Fig. 6,B). Like day 14 CXCR3+CD8+ T cells, they were able to produce TNF-α and IFN-γ (Fig. 6,B). Interestingly, there were more annexin V+ cells in the day 14 CXCR3+CD8+ T cell population than in that of the day 14 CXCR3CD8+ T cells (Fig. 6,B). Both alloreactive CD8+ T cell subsets were able to vigorously proliferate in response to host DCs in vitro (Fig. 6,C). However, day 14 CXCR3+CD8+ T cells were more potent cytolytic killers than day 14 CXCR3CD8+ T cells (Fig. 6 D). These data suggest that donor day 14 CXCR3CD8+ T cells do not express more of the other chemokine receptors than do alloreactive CXCR3+CD8+ T cells, are less cytolytic than day 14 CXCR3+CD8+ T cells, and that both populations are Ag-activated T cells.

FIGURE 6.

Characterization of alloreactive CXCR3CD8+ T cells. Day 14 CXCR3CD8+ and CXCR3+CD8+ T cells were obtained by sorting using FACS. A, Expression of chemokine receptors in these cells was identified by quantitative real-time PCR, with a panel of donor CD8+ T cells from normal C3H.SW mice as controls. These results were normalized to endogenously expressed GADPH in the same sample and were expressed as 2−ΔΔCT, as described in Materials and Methods. B, The sorted cells were stained with Abs, as indicated. The values in the histograms represent the percentages of cells in each CD8+ T cell subset. C, Day 14 CXCR3CD8+ or CXCR3+CD8+ T cells were stimulated with irradiated allogeneic B6/SJL DCs. T cell proliferation was measured by incorporation of [3H]thymidine. D, These sorted cells were incubated with EL-4 cells for 12 h. The supernatant was measured for released lactate dehydrogenase to test the cytolytic activity. Data are expressed as the means ± SD of triplicate cultures. Results are representative of five independent experiments. ∗, p < 0.05.

FIGURE 6.

Characterization of alloreactive CXCR3CD8+ T cells. Day 14 CXCR3CD8+ and CXCR3+CD8+ T cells were obtained by sorting using FACS. A, Expression of chemokine receptors in these cells was identified by quantitative real-time PCR, with a panel of donor CD8+ T cells from normal C3H.SW mice as controls. These results were normalized to endogenously expressed GADPH in the same sample and were expressed as 2−ΔΔCT, as described in Materials and Methods. B, The sorted cells were stained with Abs, as indicated. The values in the histograms represent the percentages of cells in each CD8+ T cell subset. C, Day 14 CXCR3CD8+ or CXCR3+CD8+ T cells were stimulated with irradiated allogeneic B6/SJL DCs. T cell proliferation was measured by incorporation of [3H]thymidine. D, These sorted cells were incubated with EL-4 cells for 12 h. The supernatant was measured for released lactate dehydrogenase to test the cytolytic activity. Data are expressed as the means ± SD of triplicate cultures. Results are representative of five independent experiments. ∗, p < 0.05.

Close modal

We next determined whether alloreactive CXCR3CD8+ T cells were able to generate CXCR3+ T cells. We found that by 12 h after adoptive transfer, CXCR3 appeared on the surface of infused day 14 CXCR3CD8+ T cells recovered from the spleen (22.9 ± 5.1%) and the liver (26.4 ± 3.4%). These proportions were further increased to 33.9 ± 4.2% in spleen and 67.2 ± 4.3% in liver by 24 h (Fig. 7,A). Ex vivo stimulation with host DCs and IL-2 significantly up-regulated the expression of CXCR3 on the surface of day 14 CXCR3CD8+ T cells (Fig. 7, B and C). Addition of CTLA-4-Ig, anti-MHC-I neutralizing Ab, or anti-IL-2 neutralizing Ab markedly decreased the percentage of CXCR3+ to 21.0 ± 2.2%, 26.8 ± 1.0%, or 8.7 ± 3.1%, respectively (Fig. 7,D). Both mRNA and protein of CXCR3 were significantly increased in these day 14 CXCR3CD8+ T cells stimulated ex vivo with DCs or IL-2 (Fig. 8, A and B). These results explain the finding that day 14 CXCR3CD8+ T cells are able to differentiate to highly cytolytic CXCR3+CD8+ T cells mediating host tissue injury. Interestingly, a substantial proportion of infused day 14 CXCR3+CD8+ T cells rapidly reduced their surface expression of CXCR3 by 12 h after adoptive transfer (Fig. 7,A). Donor CXCR3CD8+ T cells were still detected by 21 days and later on after the adoptive transfer of donor day 14 CXCR3+CD8+ T cells (data not shown). Both DCs and IL-2 ex vivo stimulated the decrease of CXCR3 mRNA and protein in day 14 CXCR3+CD8+ T cells (Fig. 8, A and B). Previous studies have shown that both internalization and new synthesis account for regulation of the chemokine receptor number on the cell surface (36, 39, 40). Indeed, flow cytometric analysis has shown in this study that both DCs and IL-2 induce the internalization of surface CXCR3 in day 14 CXCR3+CD8+ T cells (Fig. 8 C). Taken together, these results suggest that the surface expression of CXCR3 can be dynamically regulated on alloreactive CD8+ T cells, which phenomenon may account for the necessity of long-term anti-CXCR3 treatment in GVHD inhibition.

FIGURE 7.

The dynamic expression of CXCR3 on alloreactive CD8+ T cells is regulated by DCs and IL-2. Highly purified day 14 CXCR3CD8+ or CXCR3+CD8+ T cells plus C3H.SW TCD BM were adoptively transferred into secondary B6/SJL recipients. After 12 and 24 h, donor-derived CD8+ T cells were recovered from spleens and livers in these secondary recipients and were analyzed for CXCR3 expression (A). Day 14 CXCR3CD8+ and CXCR3+CD8+ T cells were stimulated ex vivo in the presence of B6/SJL DCs (T cell/DC ratio = 3:1) and cytokine mixture, including IL-2, IL-7, and IL-15, for 3 days. Expression of CXCR3 was assayed on the cultured cells. Histograms are shown after gating on CD45.2+CD8+ T cells (B). Day 14 CXCR3CD8+ T cells were stimulated with B6/SJL DCs, IL-2, IL-7, IL-15, or B6/SJL DCs plus cytokine mixture (C), and with B6/SJL DCs in the presence of CTLA-4-Ig, or neutralizing Ab to MHC-I or IL-2 for 3 days, with IgG as control (D). Levels of CXCR3 expression on the cultured CD8+ T cells were assayed by FACS. Data are expressed as the means ± SD of six mice per group. Results represent four independent experiments. ∗, p < 0.05.

FIGURE 7.

The dynamic expression of CXCR3 on alloreactive CD8+ T cells is regulated by DCs and IL-2. Highly purified day 14 CXCR3CD8+ or CXCR3+CD8+ T cells plus C3H.SW TCD BM were adoptively transferred into secondary B6/SJL recipients. After 12 and 24 h, donor-derived CD8+ T cells were recovered from spleens and livers in these secondary recipients and were analyzed for CXCR3 expression (A). Day 14 CXCR3CD8+ and CXCR3+CD8+ T cells were stimulated ex vivo in the presence of B6/SJL DCs (T cell/DC ratio = 3:1) and cytokine mixture, including IL-2, IL-7, and IL-15, for 3 days. Expression of CXCR3 was assayed on the cultured cells. Histograms are shown after gating on CD45.2+CD8+ T cells (B). Day 14 CXCR3CD8+ T cells were stimulated with B6/SJL DCs, IL-2, IL-7, IL-15, or B6/SJL DCs plus cytokine mixture (C), and with B6/SJL DCs in the presence of CTLA-4-Ig, or neutralizing Ab to MHC-I or IL-2 for 3 days, with IgG as control (D). Levels of CXCR3 expression on the cultured CD8+ T cells were assayed by FACS. Data are expressed as the means ± SD of six mice per group. Results represent four independent experiments. ∗, p < 0.05.

Close modal
FIGURE 8.

Synthesis and internalization are important for the convertible expression of CXCR3 on alloreactive CD8+ T cells. CD8+ T cells were sorted from day 14 CXCR3CD8+ and CXCR3+CD8+ T cells stimulated with B6/SJL DCs and IL-2 for 3 days, respectively. Their expression of CXCR3 was calculated through quantitative real-time PCR by comparison with GAPDH and was expressed as 2−ΔΔCT, with a panel of cultured day 14 CXCR3CD8+ T cells with medium alone as control; data are expressed as means ± SD (A), and through Western blot using rabbit anti-mouse CXCR3 Ab, with the expression of actin as a normalized control (B). These CD8+ T cells were also subjected to a protein recycling assay, as described in Materials and Methods. Data are expressed as means ± SD of total and surface expression of CXCR3 molecule in CD8+ T cells with or without recycling, as determined using FACS with intracellular and surface staining with anti-CXCR3 Ab (C). Results are representative of three independent experiments. ∗, p < 0.05.

FIGURE 8.

Synthesis and internalization are important for the convertible expression of CXCR3 on alloreactive CD8+ T cells. CD8+ T cells were sorted from day 14 CXCR3CD8+ and CXCR3+CD8+ T cells stimulated with B6/SJL DCs and IL-2 for 3 days, respectively. Their expression of CXCR3 was calculated through quantitative real-time PCR by comparison with GAPDH and was expressed as 2−ΔΔCT, with a panel of cultured day 14 CXCR3CD8+ T cells with medium alone as control; data are expressed as means ± SD (A), and through Western blot using rabbit anti-mouse CXCR3 Ab, with the expression of actin as a normalized control (B). These CD8+ T cells were also subjected to a protein recycling assay, as described in Materials and Methods. Data are expressed as means ± SD of total and surface expression of CXCR3 molecule in CD8+ T cells with or without recycling, as determined using FACS with intracellular and surface staining with anti-CXCR3 Ab (C). Results are representative of three independent experiments. ∗, p < 0.05.

Close modal

Finally, we examined whether long-term anti-CXCR3 Ab treatment had the therapeutic effect of blocking ongoing GVHD. Anti-CXCR3 Ab was administered from days 7 to 28 after transplantation to B6/SJL mice receiving donor C3H.SW CD8+ T cells. All recipient mice administered control IgG developed typical GVHD, with ∼70% of them dying from the disease. By contrast, ∼80% of mice receiving long-term anti-CXCR3 Ab treatment survived, with a reduced clinical score for GVHD (Fig. 9,A). Interestingly, long-term administration of anti-CXCR3 did not affect the frequency of cytolytic T cells against EL-4 leukemia cells (Fig. 9,B) or the median survival time of EL-4 cell-challenged B6/SJL mice that received donor CD8+ T cells plus TCD BM as compared with IgG treatment control (Fig. 9 C). Thus, long-term administration of anti-CXCR3 Ab can inhibit ongoing GVHD, while retaining the GVL effect of alloreactive CD8+ T cells.

FIGURE 9.

Therapeutic effect of long-term treatment with anti-CXCR3 Ab. Lethally irradiated B6/SJL recipient mice (CD45.1+) were injected i.v. with C3H.SW CD8+ T cells plus TCD BM. A, These recipients were administrated with anti-CXCR3 Ab on days 7, 10, 14, 16, 20, 22, 24, and 27 after transplantation, with IgG as control. The therapeutic effect of anti-CXCR3 Ab on donor CD8+ T cell-mediated GVHD was assayed by observation of clinic GVHD signs and survival. B, The recipients were treated with anti-CXCR3 Ab from days 0 to 21 (long-term) after transplantation. Donor-derived CD8+ T cells were recovered from these recipient mice on day 35 after transplantation, and examined for cytolysis of EL-4 cells. C, These recipient mice treated with anti-CXCR3 Ab for long-term were in vivo challenged with EL-4 cells on day 7 after transplantation. Survival was monitored. Symbols represent the following: TCD BM (×); TCD BM plus CD8+ T cells plus treatment with anti-CXCR3 Ab from days 7 to 27 after transplantation (□), or with IgG (•); donor-derived CD8+ T cells from recipient mice treated with anti-CXCR3 Ab for the long-term (▾), or with IgG (○); TCD BM plus EL-4 cells (▿); TCD BM plus CD8+ T cells plus EL-4 cells plus treatment with anti-CXCR3 Ab for the long-term (▴), or with IgG (▵); and TCD BM plus CD8+ T cells plus anti-CXCR3 Ab treatment for the long-term (▪). Results are representative of nine mice for each group of three independent experiments. ∗, p < 0.05.

FIGURE 9.

Therapeutic effect of long-term treatment with anti-CXCR3 Ab. Lethally irradiated B6/SJL recipient mice (CD45.1+) were injected i.v. with C3H.SW CD8+ T cells plus TCD BM. A, These recipients were administrated with anti-CXCR3 Ab on days 7, 10, 14, 16, 20, 22, 24, and 27 after transplantation, with IgG as control. The therapeutic effect of anti-CXCR3 Ab on donor CD8+ T cell-mediated GVHD was assayed by observation of clinic GVHD signs and survival. B, The recipients were treated with anti-CXCR3 Ab from days 0 to 21 (long-term) after transplantation. Donor-derived CD8+ T cells were recovered from these recipient mice on day 35 after transplantation, and examined for cytolysis of EL-4 cells. C, These recipient mice treated with anti-CXCR3 Ab for long-term were in vivo challenged with EL-4 cells on day 7 after transplantation. Survival was monitored. Symbols represent the following: TCD BM (×); TCD BM plus CD8+ T cells plus treatment with anti-CXCR3 Ab from days 7 to 27 after transplantation (□), or with IgG (•); donor-derived CD8+ T cells from recipient mice treated with anti-CXCR3 Ab for the long-term (▾), or with IgG (○); TCD BM plus EL-4 cells (▿); TCD BM plus CD8+ T cells plus EL-4 cells plus treatment with anti-CXCR3 Ab for the long-term (▴), or with IgG (▵); and TCD BM plus CD8+ T cells plus anti-CXCR3 Ab treatment for the long-term (▪). Results are representative of nine mice for each group of three independent experiments. ∗, p < 0.05.

Close modal

We have established a new approach in which prolonged in vivo administration of anti-CXCR3 Ab blocks alloreactive T cell-mediated GVHD. We found that both donor-derived CXCR3+CD8+ and CXCR3CD8+ T cells develop in GVHD recipients and that both are virulent inducers of GVHD. Although short-term anti-CXCR3 Ab treatment inhibits day 14 CXCR3+CD8+ T cell-mediated GVHD, long-term anti-CXCR3 Ab treatment is required for blockade of GVHD induced by day 14 CXCR3CD8+ T cells. It is likely that anti-CXCR3 Ab treatment not only inhibits the migration of alloreactive CD8+ T cells, but also kills CXCR3+CD8+ T cells in vivo and ex vivo, perhaps via a mechanism involving CDC. This process then leads to the reduction of alloreactive T cells in GVHD target organs, and thus to GVHD inhibition, without mitigating GVL activity. As compared with day 14 CXCR3+CD8+ T cells, day 14 CXCR3CD8+ T cells do not express higher levels of granzyme B, FasL, IFN-γ, TNF-α, and other tested chemokine receptors, including CCR2, CCR3, CCR5, CCR8, and CXCR2. After stimulation by DCs and IL-2, these day 14 CXCR3+CD8+ T cells down-regulate their surface expression of CXCR3, whereas day 14 CXCR3CD8+ T cells rapidly become CXCR3+. This dynamic expression of CXCR3 results in conversion between CXCR3CD8+ and CXCR3+CD8+ T cells during GVH reactions. Thus, dynamic expression of CXCR3 on alloreactive CD8+ T cells hampers the efficacy of short-term treatment of anti-CXCR3 Ab to prevent and treat GVHD. Prolonged in vivo administration of anti-CXCR3 Ab to block continually generated alloreactive CXCR3+CD8+ T cells is critical to achievement of efficacious GVHD inhibition.

Previous studies have demonstrated that Ag and IL-2 induce the expression of CXCR3 on T cells (40, 41, 42). In mouse models of ocular inflammation, the CXCR3 expressed on Th1 cells profoundly changed at different stages of activation, with Ag being the critical regulator (40). Our findings together with these observations indicate that alloreactive CXCR3CD8+ and CXCR3+CD8+ T cells may not represent a distinct cell subset, but rather the activated T cells at different activation and differentiation stages. In our GVHD model, we found that alloreactive CXCR3+CD8+ T cells first peak at day 7 after transplantation in B6/SJL mice with ongoing GVHD. They are virulent inducers of GVHD, but are susceptible to short-term anti-CXCR3 Ab treatment. Interestingly, significantly more alloreactive CXCR3CD8+ T cells than CXCR3+CD8+ T cells were detected in the livers between days 7 and 28, and in the spleens between days 14 and 35 in B6/SJL recipients after transplantation. These cells expressed lower levels of CD127, and produced less IFN-γ, TNF-α, granzyme B, and FasL than did CXCR3+CD8+ T cells. Furthermore, CXCR3CD8+ T cells rapidly increased the expression of CXCR3 after their adoptive transfer into secondary B6/SJL recipients. All these data suggest that alloreactive CXCR3CD8+ T cells are indeed relatively undifferentiated T cells with the ability to generate CXCR3+CD8+ T cells, leading to induction of the subsequent host-tissue injury. However, this may not explain why some of the alloreactive CXCR3+CD8+ T cells down-regulate CXCR3 expression in response to DCs and IL-2.

Down-regulation of CXCR3 on activated T cells has been previously demonstrated in numerous studies (36, 39, 40, 41, 42, 43). Data from these studies indicate that upon cell-cell contact, and stimulation of TCR and cytokines, T cells rapidly down-regulate their surface CXCR3 expression to facilitate their selective migration into different organs and to enhance their proinflammatory activities. We found that in response to persistent host miHAs, the reduction of CXCR3 on day 14 CXCR3+CD8+ T cells can be observed as soon as 12 h after their adoptive transfer. Ex vivo culture showed that host DCs and IL-2 induce the reduction of CXCR3 in day 14 CXCR3+CD8+ T cells. Some previous studies have shown that host APCs are essential for initiation of CD8+ T cell-mediated GVHD, but are eliminated within 1 wk after preparative conditioning (27). Other studies suggest that donor-derived APCs are required for maximal GVHD, but not for GVL (44). We found that ex vivo blocking of the interaction of DC and T cells with CTLA4-Ig significantly inhibits the up-regulation of CXCR3 on the surface of alloreactive CXCR3CD8+ T cells. This suggests that costimulatory signals, such as that of CD86, as derived from donor DCs, would suffice to increase the expression of CXCR3, resulting in the conversion of CXCR3CD8+ T cells to CXCR3+CD8+ T cells. Furthermore, both internalization of CXCR3 and inhibited CXCR3 synthesis were seen to be responsible for DC- and IL-2-induced CXCR3 reduction in these CXCR3+CD8+ T cells. Our findings are in agreement with previous studies indicating that the decline of CXCR3 on the cell surface is probably accounted for by its internalization (36, 39) and down-regulation of its production (40, 41). These data suggest that alloreactive CXCR3CD8+ T cells can be generated from fully activated and/or differentiated alloreactive CXCR3+CD8+ T cells. Thus, CXCR3CD8+ T cells may represent less differentiated CD8+ T cells and may be derived from converted CXCR3+CD8+ T cells.

It is possible that the inability of short-term anti-CXCR3 Ab treatment to prevent alloreactive CXCR3CD8+ T cell-mediated GVHD might be associated with their expression of some other chemokine receptors, such as CCR2, CCR5, CCR6, CCR7, and CCR10. All these chemokine receptors have been shown to play roles in regulating alloreactive T cell-mediated GVHD (15, 16, 17, 18, 19, 20). Data from those studies indicate that donor T cells from CCR2−/− and CCR10−/− mice have reduced ability to infiltrate GVHD target organs (16, 17), and that CCR1−/− and CCR6−/− T cells are less virulent with reduced secretion of GVHD-related inflammatory cytokines (18, 19). However, GVHD remains a problem in these recipient mice. Moreover, transfer CCR5−/− T cells induced virulent GVHD with increased mortality (21, 22). Together with our findings that CXCR3CD8+ T cells do not express relatively more of other chemokine receptors, it is likely that these chemokine receptors do not play central roles in controlling alloreactive T cell-mediated GVHD.

Data from recent studies have demonstrated that CXCR3 is associated with the development of GVHD (9, 14, 45). Donor T cells derived from CXCR3 gene-deficient mice are much less virulent than Wt T cells in induction of lethal GVHD (29). A recent study further suggests that CXCL10-CXCR3 interactions play an important role in the pathogenesis of acute GVHD in the skin of patients receiving allogeneic HSCT (23). However, whereas studies using gene-deficient animals can definitively point to the involvement of a particular target, it remains to develop CXCR3 antagonist agents to inhibit GVHD for use in clinical applications. In this study, we establish, for the first time, that prolonged anti-CXCR3 Ab treatment for 21 days prevents GVHD. Most notably, we find that long-term anti-CXCR3 Ab treatment reduces the infiltration of donor-derived CD8+ T cells into GVHD target tissues and reduces subsequent inflammatory injury. Interestingly, this prolonged anti-CXCR3 Ab treatment does not affect the proliferation and engraftment of infused donor CD8+ T cells. Furthermore, we demonstrate that prolonged anti-CXCR3 Ab treatment also inhibits ongoing GVHD and preserves an alloreactive CD8+ T cell-mediated GVL effect. Therefore, it is likely that inhibition of GVHD by in vivo administration of anti-CXCR3 Ab enhances the reconstitution of donor T cells after allogeneic HSCT without impairing the GVL effect.

In summary, we have established a new method for successful inhibition of alloreactive T cell-mediated GVHD by specific targeting of CXCR3 using our newly generated neutralizing anti-CXCR3 Ab and a distinctive Ab-administration regimen. This method has potential for development of new clinical approaches to prevention and treatment of GVHD after HSCT. Furthermore, investigation into the finding that conventional short-term blockade with neutralizing Ab against CXCR3 fails to inhibit alloreactive T cell-mediated GVHD will have significant implications for re-evaluation of the effect of neutralizing Abs against other chemokine receptors in the treatment of T cell-mediated immunopathologies.

We express our gratitude to Dr. Qiang Liu (Department of Pathology, Renji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China) for help in the histopathologic assessment; Drs. Jian Zhang and Ling Qiu in FACS assay; Dr. Hongmei Li (Soochow University) in excellent statistical analysis; and Drs. Fang Zhao and Yichuan Xiao for kind assistance. We highly appreciate Dr. Sheri M. Skinner (Baylor College of Medicine, Houston, TX) for critical review of the manuscript.

The authors have no financial conflict of interest.

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.

1

This work was supported by grants from the National Natural Science Foundation of China (NSF30670911 and NSF 30873045), the programs of the Science and Technology Commission of Shanghai Municipality (03JC14085, 06DZ19020, and 074319102), and the Shanghai Leading Academic Discipline Project (T0206).

4

Abbreviations used in this paper: GVHD, graft-versus-host disease; CDC, complement-dependent cytotoxicity; CT, cycle threshold; DC, dendritic cell; FasL, Fas ligand; GVH, graft-vs-host; GVL, graft-vs-leukemia; HSCT, hematopoietic stem cell transplantation; miHA, minor histocompatibility Ag; PI, propidium iodide; TCD BM, T cell-depleted bone marrow cells; Wt, wild type.

1
Welniak, L. A., B. R. Blazar, W. J. Murphy.
2007
. Immunobiology of allogeneic hematopoietic stem cell transplantation.
Annu. Rev. Immunol.
25
:
139
-170.
2
Ferrara, J. L., R. Levy, N. J. Chao.
1999
. Pathophysiologic mechanisms of acute graft-vs.-host disease.
Biol. Blood Marrow Transplant.
5
:
347
-356.
3
Goker, H., I. C. Haznedaroglu, N. J. Chao.
2001
. Acute graft-versus-host disease: pathobiology and management.
Exp. Hematol.
29
:
259
-277.
4
Korngold, R., J. Sprent.
1982
. Features of T cells causing H-2-restricted lethal graft-vs.-host disease across minor histocompatibility barriers.
J. Exp. Med.
155
:
872
-883.
5
Via, C. S..
1991
. Kinetics of T cell activation in acute and chronic forms of murine graft-versus-host disease.
J. Immunol.
146
:
2603
-2609.
6
Shlomchik, W. D., M. S. Couzens, C. B. Tang, J. McNiff, M. E. Robert, J. Liu, M. J. Shlomchik, S. G. Emerson.
1999
. Prevention of graft versus host disease by inactivation of host antigen-presenting cells.
Science
285
:
412
-415.
7
Duffner, U. A., Y. Maeda, K. R. Cooke, P. Reddy, R. Ordemann, C. Liu, J. L. Ferrara, T. Teshima.
2004
. Host dendritic cells alone are sufficient to initiate acute graft-versus-host disease.
J. Immunol.
172
:
7393
-7398.
8
Zhang, Y., W. D. Shlomchik, G. Joe, J. P. Louboutin, J. Zhu, A. Rivera, D. Giannola, S. G. Emerson.
2002
. APCs in the liver and spleen recruit activated allogeneic CD8+ T cells to elicit hepatic graft-versus-host disease.
J. Immunol.
169
:
7111
-7118.
9
Wysocki, C. A., A. Panoskaltsis-Mortari, B. R. Blazar, J. S. Serody.
2005
. Leukocyte migration and graft-versus-host disease.
Blood
105
:
4191
-4199.
10
Kunkel, E. J., E. C. Butcher.
2002
. Chemokines and the tissue-specific migration of lymphocytes.
Immunity
16
:
1
-4.
11
Beilhack, A., S. Schulz, J. Baker, G. F. Beilhack, C. B. Wieland, E. I. Herman, E. M. Baker, Y. A. Cao, C. H. Contag, R. S. Negrin.
2005
. In vivo analyses of early events in acute graft-versus-host disease reveal sequential infiltration of T cell subsets.
Blood
106
:
1113
-1122.
12
Rollins, B. J..
1997
. Chemokines.
Blood
90
:
909
-928.
13
Luster, A. D..
1998
. Chemokines: chemotactic cytokines that mediate inflammation.
N. Engl. J. Med.
338
:
436
-445.
14
Jaksch, M., M. Remberger, J. Mattsson.
2005
. Increased gene expression of chemokine receptors is correlated with acute graft-versus-host disease after allogeneic stem cell transplantation.
Biol. Blood Marrow Transplant.
11
:
280
-287.
15
Yakoub-Agha, I., P. Saule, S. Depil, J. B. Micol, C. Grutzmacher, F. Boulanger-Villard, F. Bauters, J. P. Jouet, J. P. Dessaint, M. Labalette.
2006
. A high proportion of donor CD4+ T cells expressing the lymph node-homing chemokine receptor CCR7 increases incidence and severity of acute graft-versus-host disease in patients undergoing allogeneic stem cell transplantation for hematological malignancy.
Leukemia
20
:
1557
-1565.
16
Faaij, C. M., A. C. Lankester, E. Spierings, M. Hoogeboom, E. P. Bowman, M. Bierings, T. Revesz, R. M. Egeler, M. J. van Tol, N. E. Annels.
2006
. A possible role for CCL27/CTACK-CCR10 interaction in recruiting CD4 T cells to skin in human graft-versus-host disease.
Br. J. Haematol.
133
:
538
-549.
17
Terwey, T. H., T. D. Kim, A. A. Kochman, V. M. Hubbard, S. Lu, J. L. Zakrzewski, T. Ramirez-Montagut, J. M. Eng, S. J. Muriglan, G. Heller, et al
2005
. CCR2 is required for CD8-induced graft-versus-host disease.
Blood
106
:
3322
-3330.
18
Choi, S. W., G. C. Hildebrandt, K. M. Olkiewicz, D. A. Hanauer, M. N. Chaudhary, I. A. Silva, C. E. Rogers, D. T. Deurloo, J. M. Fisher, C. Liu, et al
2007
. CCR1/CCL5 (RANTES) receptor-ligand interactions modulate allogeneic T-cell responses and graft-versus-host disease following stem-cell transplantation.
Blood
110
:
3447
-3455.
19
Varona, R., V. Cadenas, L. Gomez, A. C. Martinez, G. Marquez.
2005
. CCR6 regulates CD4+ T-cell-mediated acute graft-versus-host disease responses.
Blood
106
:
18
-26.
20
Murai, M., H. Yoneyama, A. Harada, Z. Yi, C. Vestergaard, B. Guo, K. Suzuki, H. Asakura, K. Matsushima.
1999
. Active participation of CCR5+CD8+ T lymphocytes in the pathogenesis of liver injury in graft-versus-host disease.
J. Clin. Invest.
104
:
49
-57.
21
Wysocki, C. A., S. B. Burkett, A. Panoskaltsis-Mortari, S. L. Kirby, A. D. Luster, K. McKinnon, B. R. Blazar, J. S. Serody.
2004
. Differential roles for CCR5 expression on donor T cells during graft-versus-host disease based on pretransplant conditioning.
J. Immunol.
173
:
845
-854.
22
Welniak, L. A., Z. Wang, K. Sun, W. Kuziel, M. R. Anver, B. R. Blazar, W. J. Murphy.
2004
. An absence of CCR5 on donor cells results in acceleration of acute graft-vs-host disease.
Exp. Hematol.
32
:
318
-324.
23
Piper, K. P., C. Horlock, S. J. Curnow, J. Arrazi, S. Nicholls, P. Mahendra, C. Craddock, P. A. Moss.
2007
. CXCL10-CXCR3 interactions play an important role in the pathogenesis of acute graft-versus-host disease in the skin following allogeneic stem cell transplantation.
Blood
110
:
3827
-3832.
24
Sallusto, F., D. Lenig, C. R. Mackay, A. Lanzavecchia.
1998
. Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes.
J. Exp. Med.
187
:
875
-883.
25
Piali, L., C. Weber, G. LaRosa, C. R. Mackay, T. A. Springer, I. Clark-Lewis, B. Moser.
1998
. The chemokine receptor CXCR3 mediates rapid and shear-resistant adhesion-induction of effector T lymphocytes by the chemokines IP10 and Mig.
Eur. J. Immunol.
28
:
961
-972.
26
Zhang, Y., G. Joe, E. Hexner, J. Zhu, S. G. Emerson.
2005
. Alloreactive memory T cells are responsible for the persistence of graft-versus-host disease.
J. Immunol.
174
:
3051
-3058.
27
Zhang, Y., G. Joe, E. Hexner, J. Zhu, S. G. Emerson.
2005
. Host-reactive CD8+ memory stem cells in graft-versus-host disease.
Nat. Med.
11
:
1299
-1305.
28
Zhang, C., J. Lou, N. Li, I. Todorov, C. L. Lin, Y. A. Cao, C. H. Contag, F. Kandeel, S. Forman, D. Zeng.
2007
. Donor CD8+ T cells mediate graft-versus-leukemia activity without clinical signs of graft-versus-host disease in recipients conditioned with anti-CD3 monoclonal antibody.
J. Immunol.
178
:
838
-850.
29
Duffner, U., B. Lu, G. C. Hildebrandt, T. Teshima, D. L. Williams, P. Reddy, R. Ordemann, S. G. Clouthier, K. Lowler, C. Liu, et al
2003
. Role of CXCR3-induced donor T-cell migration in acute GVHD.
Exp. Hematol.
31
:
897
-902.
30
Zhang, Y., J. P. Louboutin, J. Zhu, A. J. Rivera, S. G. Emerson.
2002
. Preterminal host dendritic cells in irradiated mice prime CD8+ T cell-mediated acute graft-versus-host disease.
J. Clin. Invest.
109
:
1335
-1344.
31
Zhang, Y., G. Joe, J. Zhu, R. Carroll, B. Levine, E. Hexner, C. June, S. G. Emerson.
2004
. Dendritic cell-activated CD44hiCD8+ T cells are defective in mediating acute graft-versus-host disease but retain graft-versus-leukemia activity.
Blood
103
:
3970
-3978.
32
Zhang, Y., A. Harada, J. B. Wang, Y. Y. Zhang, S. Hashimoto, M. Naito, K. Matsushima.
1998
. Bifurcated dendritic cell differentiation in vitro from murine lineage phenotype-negative c-kit+ bone marrow hematopoietic progenitor cells.
Blood
92
:
118
-128.
33
Johnson, M. L., E. R. Farmer.
1998
. Graft-versus-host reactions in dermatology.
J. Am. Acad. Dermatol.
38
:
369
-392.
34
Cooke, K. R., W. Krenger, G. Hill, T. R. Martin, L. Kobzik, J. Brewer, R. Simmons, J. M. Crawford, M. R. van den Brink, J. L. Ferrara.
1998
. Host reactive donor T cells are associated with lung injury after experimental allogeneic bone marrow transplantation.
Blood
92
:
2571
-2580.
35
Hill, G. R., K. R. Cooke, T. Teshima, J. M. Crawford, J. C. Keith, Jr, Y. S. Brinson, D. Bungard, J. L. Ferrara.
1998
. Interleukin-11 promotes T cell polarization and prevents acute graft-versus-host disease after allogeneic bone marrow transplantation.
J. Clin. Invest.
102
:
115
-123.
36
Sauty, A., R. A. Colvin, L. Wagner, S. Rochat, F. Spertini, A. D. Luster.
2001
. CXCR3 internalization following T cell-endothelial cell contact: preferential role of IFN-inducible T cell α chemoattractant (CXCL11).
J. Immunol.
167
:
7084
-7093.
37
Dechant, M., G. Vidarsson, B. Stockmeyer, R. Repp, M. J. Glennie, M. Gramatzki, J. G. J. van de Winkel, T. Valerius.
2002
. Chimeric IgA antibodies against HLA class II effectively trigger lymphoma cell killing.
Blood
100
:
4574
-4580.
38
Hussain, S. A., C. M. Cheney, A. J. Johnson, T. S. Lin, M. R. Grever, M. A. Caligiuri, D. M. Lucas, J. C. Byrd.
2007
. Mcl-1 is a relevant therapeutic target in acute and chronic lymphoid malignancies: down-regulation enhances rituximab-mediated apoptosis and complement-dependent cytotoxicity.
Clin. Cancer Res.
13
:
2144
-2150.
39
Dagan-Berger, M., R. Feniger-Barish, S. Avniel, H. Wald, E. Galun, V. Grabovsky, R. Alon, A. Nagler, A. Ben-Baruch, A. Peled.
2006
. Role of CXCR3 carboxyl terminus and third intracellular loop in receptor-mediated migration, adhesion and internalization in response to CXCL11.
Blood
107
:
3821
-3831.
40
Chen, J., B. P. Vistica, H. Takase, D. I. Ham, R. N. Fariss, E. F. Wawrousek, C. C. Chan, J. A. DeMartino, J. M. Farber, I. Gery.
2004
. A unique pattern of up- and down-regulation of chemokine receptor CXCR3 on inflammation-inducing Th1 cells.
Eur. J. Immunol.
34
:
2885
-2894.
41
Sallusto, F., E. Kremmer, B. Palermo, A. Hoy, P. Ponath, S. Qin, R. Forster, M. Lipp, A. Lanzavecchia.
1999
. Switch in chemokine receptor expression upon TCR stimulation reveals novel homing potential for recently activated T cells.
Eur. J. Immunol.
29
:
2037
-2045.
42
Hodge, D. L., W. B. Schill, J. M. Wang, I. Blanca, D. A. Reynolds, J. R. Ortaldo, H. A. Young.
2002
. IL-2 and IL-12 alter NK cell responsiveness to IFN-γ-inducible protein 10 by down-regulating CXCR3 expression.
J. Immunol.
168
:
6090
-6098.
43
Reckamp, K. L., R. A. Figlin, N. Moldawer, A. J. Pantuck, A. S. Belldegrun, M. D. Burdick, R. M. Strieter.
2007
. Expression of CXCR3 on mononuclear cells and CXCR3 ligands in patients with metastatic renal cell carcinoma in response to systemic IL-2 therapy.
J. Immunother.
30
:
417
-424.
44
Matte, C. C., J. Liu, J. Cormier, B. E. Anderson, I. Athanasiadis, D. Jain, J. McNiff, W. D. Shlomchik.
2004
. Donor APCs are required for maximal GVHD but not for GVL.
Nat. Med.
10
:
987
-992.
45
Mapara, M. Y., C. Leng, Y. M. Kim, R. Bronson, A. Lokshin, A. Luster, M. Sykes.
2006
. Expression of chemokines in GVHD target organs is influenced by conditioning and genetic factors and amplified by GVHR.
Biol. Blood Marrow Transplant.
12
:
623
-634.