The use of receptor antagonists for chemokines is an alternative approach to blocking chemokine actions and has the potential to provide novel therapeutics. We determined the receptor antagonist properties of murine N-terminally truncated secondary lymphoid tissue chemokine (SLC)/6Ckine/CCR ligand 21 analogs and evaluated the preventive effects of SLC antagonists on chronic graft-vs-host disease (GVHD) in a murine model by blocking the homing of donor CCR7-expressing T cells into the recipient’s lymphoid organs. SLC analogs truncated >4 aa residues from the N terminus showed a loss of chemotaxis and Ca2+ influx of CCR7-expressing cells and also inhibited SLC-stimulated chemotaxis and SLC-induced Ca2+ influx completely. To determine whether SLC antagonist inhibits the development of chronic GVHD, chronic GVHD was induced by injecting DBA/2 spleen cells into (C57BL/6 × DBA/2) F1 mice. Total numbers of spleen cells and host B cells, serum levels of IgE, and of total IgG and IgG1 of anti-DNA Abs in SLC antagonist-treated GVHD mice were significantly lower than those in control PBS-treated GVHD mice. This was due to a reduction in the levels of activated donor CD4+ T cells and a decrease in IL-4 production, resulting in a reduction in the numbers of activated host B cells. Therefore, our results suggest that SLC antagonist has beneficial effects for the prevention of chronic GVHD.
Graft-vs-host disease (GVHD)3 is a well-recognized complication of allogeneic bone marrow transplantation and is a major impediment to its overall therapeutic success (1, 2). In recent years, peripheral blood stem cells have largely replaced bone marrow as a peripheral source of allogeneic stem cells because of their relative ease of collection, quicker engraftment kinetics, and economic advantages. The incidence of chronic GVHD has been reported to be higher after peripheral blood stem cell transplantation (PBSCT) than after bone marrow transplantation, whereas the incidence of acute GVHD after PBSCT is not significantly higher (3, 4, 5, 6, 7). Acute GVHD is a rapidly progressive, unrelenting, systemic illness characterized by immunosuppression, cachexia, and tissue injury of the skin, liver, and intestinal mucosa. The histopathology of acute GVHD is characterized by mononuclear cell infiltrates and epithelial injury in target organs. The pathogenesis of acute GVHD involves the development of a Th1 type cell-mediated immune response in which IFN-γ is thought to play a prominent role (8, 9). In contrast, chronic GVHD has a more indolent course, and has more diverse pathologic manifestations. The clinical presentation can resemble that of systemic lupus erythematosus, and is characterized by the development of immune complex disease and autoantibody formation. Although this form of GVHD is thought to be mediated by a Th2 type humoral immune response, the mechanism of chronic GVHD is not fully understood.
Murine chronic GVHD resembling human chronic GVHD induced by PBSCT can be induced by the transfer of DBA/2 (H-2d) mouse spleen cells into (C57BL/6 × DBA/2) F1 (B6D2F1) (H-2b/d) mice. In this model, mice exhibit a systemic autoimmune disorder characterized by splenomegaly, host B cell hyperactivation, autoantibody production, and immune complex deposition (10, 11). In addition, mice with chronic GVHD exhibit increases in the number of total spleen cells, including host B cells, and in the levels of serum IgE, and of total IgG and IgG1 of anti-DNA Abs after chronic GVHD induction (12). It has been postulated that alloreactive donor (DBA/2) CD4+ T cells recognize MHC class II molecules on the B6D2F1 host cells, and then provide help to host B cells for autoantibody production (10, 13).
Chemokines are often chemotactic and mitogenic for specific immune cells. Several new lymphocyte-specific chemokines, which attract naive and memory T cells, B cells, NK cells, and dendritic cells, have been reported (14, 15). Secondary lymphoid- tissue chemokine (SLC)/6Ckine/CCR ligand 21 and EBI-1-ligand chemokine (ELC)/macrophage inflammatory protein (MIP)-3β/CCR ligand 19 are strongly expressed in T cell zones of lymphoid organs, and SLC has been localized to high endothelial venules (HEVs) and lymphatic endothelium, suggesting that SLC and ELC play important roles in the homing of lymphocytes and dendritic cells to lymphoid tissues (16, 17, 18, 19, 20, 21, 22). The chemokine receptor CCR7, the functional receptor for SLC and ELC, is expressed in CD4+ and CD8+ T cells as well as in B cells, but not in NK cells, monocytes, or neutrophils in peripheral blood. CCR7 expression is up-regulated in B cells infected by EBV, CD4+ T cells infected by human herpesvirus-6 or human herpesvirus-7, mature dendritic cells, and adult T cell leukemia cells (23, 24, 25). plt mice, which lack SLC and have reduced ELC production, have been reported to have a phenotype resembling that of CCR7-deficient mice (26, 27). Both mutant mice show marked disturbances in the distribution of T, B, and dendritic cells, resulting in impaired T cell priming and T cell-dependent B cell response. Therefore, SLC and ELC have potent chemotactic activity for CD4+ and CD8+ T cells, B cells, and mature dendritic cells and potentially play a role in T cell priming by colocalization of Ag-presenting dendritic cells and T cells.
The use of receptor antagonists for chemokines is an alternative approach to blocking chemokine actions and has the potential to provide novel therapeutics. In several reports, N-terminally truncated or modified chemokine analogs have been described to act as receptor antagonists (28, 29, 30, 31, 32, 33, 34, 35, 36). We describe the receptor antagonist properties of murine N-terminally truncated SLC analogs and evaluate the preventive effects of SLC antagonist treatment on chronic GVHD model mice.
Materials and Methods
Cells and mice
CCR7-transfected L1.2 cells (L1.2/CCR7) were kindly provided by O. Yoshie (Kinki University School of Medicine, Osaka, Japan). This transfectant was maintained in RPMI 1640 medium supplemented with 10% heat-inactivated FCS (Life Technologies, Gaithersburg, MD). DBA/2 (H-2d) and (C57BL/6 × DBA/2)F1 (B6D2F1) female mice, 6–8 wk of age, were purchased from Charles River Breeding Laboratories (Atsugi, Kanagawa, Japan) and housed for 2 wk before GVHD induction.
Abs and reagents
FITC-conjugated anti-mouse I-Ab (Aβb) (AF6-120.1) and FITC-anti-mouse H-2Kb (AF6-88.5) mAbs were purchased from BD PharMingen (San Diego, CA). PE-conjugated anti-mouse CD4 (KT6), PE-anti-mouse CD8 (KT15), and PE-anti-mouse B220 (RA3.6B2) mAbs were obtained from Immunotech (Marseilles, France). Recombinant mouse SLC (6Ckine) was purchased from R&D Systems (Minneapolis, MN).
Construction of expression vectors encoding functional murine SLC chemokine and their N-terminal truncated analogs
To amplify the fragments containing functional coding regions, the following specific primers of murine SLC were used: 5′-AGTGATGGAGGGGGTCAGGAC-3′ (forward) and 5′-CAGGCGGGCTACTGGGCTATC-3′ (reverse) (37). Forward primers for each N-terminal truncated SLC analog were prepared with 21 oligonucleotide sequences corresponding to each truncated analog.
Total cellular RNAs were extracted from spleens of DBA/2 mice as described previously (38). The above fragments were amplified by RT-PCR by using an RNA PCR kit (Takara Shuzo, Kyoto, Japan) as described previously (39). After confirming the entire nucleotide sequence, the fragment was cloned into the KpnI site of pThioHis A vector (Invitrogen, Carlsbad, CA) after this site was end-filled to give a blunt end.
Purification of functional murine SLC chemokine protein and their analogs
The functional murine SLC chemokine protein and their analogs were purified as described previously (40). The in-frame pThioHis A-chemokine (SLC or their analogs) was transfected into Escherichia coli, TOP 10 strain, and His-patch-thioredoxin fusion proteins were prepared by using the His-Patch ThioFusion Expression system (Invitrogen) and purified on a ProBond resin column following the manufacturer’s instructions. After the fusion proteins had been digested with Enterokinase Max (Invitrogen), the complete chemokine proteins were purified by reverse-phase HPLC (RESOURCE RPC; Amersham Pharmacia Biotech, Piscataway, NJ). The protein concentration was determined with a BCA kit (Pierce, Rockford, IL), and the purity was analyzed by SDS-PAGE and silver staining. The purity of chemokine proteins was >95%.
Chemotactic assays for spleen cells were performed in polycarbonate-membrane, 6.5-mm diameter, 5-μm pore size transwell cell culture chambers (Costar, Cambridge, MA), as described previously (25). Aliquots (100 μl) of cells (5 × 106/ml) suspended in RPMI 1640/0.5% BSA were added to the upper chambers. Either SLC or analog, to produce a final concentration of 5 μg/ml, was added to the lower wells. The cells were allowed to migrate for 2 h at 37°C in a 5% CO2 incubator, after which the filters were fixed with 1% glutaraldehyde in PBS for 30 min and stained with 0.5% toluidine blue overnight. Cell migration was quantified by counting cells in each lower chamber and cells adhering to the bottom part of the polycarbonate filter. For blocking assays, after reacting with each analog at a final concentration of 1 or 5 μg/ml at 37°C for 30 min, the cells were added to the upper chambers, and chemokine was added to the lower wells at a final concentration of 5 μg/ml. Each assay was performed in triplicate.
Analysis of cytosolic-free calcium
L1.2/CCR7 cells (4 × 105) were loaded with 12.5 mg/ml Fluo-3AM (Nacalai Tesque, Kyoto, Japan) in PBS with 0.38 mg/ml Pluronic F127 (Molecular Probes, Eugene, OR) at 37°C for 30 min. After washing with PBS, the cells were resuspended in 25 mM HEPES, 140 mM NaCl, 10 mM glucose, 1.8 mM CaCl2, 1 mM MgCl2, and 3 mM KCl, pH 7.3. The fluorescence was monitored at 7-s intervals over 150 s, after addition of the test sample. Maximum Ca2+ levels were established using Fluo-3AM (designated 100% saturation) for each set of measurements by addition of 5 mM ionomycin (Sigma-Aldrich, St. Louis, MO). Ca2+ desensitization was performed by addition of one ligand, and then after 150 s the cells were treated with either the same or a different ligand.
Receptor binding of analogs was assayed by competition with 125I-labeled SLC. First, murine SLC (10 μg) was labeled with monoiodinated Bolton-Hunter reagent (specific activity 2200 Ci/mmol; DuPont, Wilmington, DE) at 4°C for 30 min. To determine the binding kinetics, L1.2/CCR7 cells (5 × 106) in 200 μl of binding buffer (RPMI 1640 medium, 0.5 mg/ml BSA, 50 mM HEPES, and 0.1% NaN3) were incubated with 125I-labeled SLC and varying concentrations of unlabeled competitor (analog) at 4°C for 30 min. The cells were pelleted through a mixture of diacetylphthalate and dibutylphthalate, and the cell-associated radioactivity was counted (total binding). Nonspecific binding was determined in the presence of a 100-fold concentration of unlabeled ligand and was subtracted from the total binding. Dissociation constants (Kd values) were determined by Scatchard analysis.
Preparation of murine cells and induction of chronic GVHD
Single-cell suspensions were prepared in RPMI 1640 from the spleens of normal DBA/2 parental donors. Cell suspensions were filtered through sterile mesh and washed. After the erythrocytes had been lysed in hemolysis buffer (including 144 mM NH4Cl and 17 mM Tris-HCl, pH 7.2), DBA/2 spleen cells were suspended at 90 × 106 viable cells/0.25 ml in RPMI 1640 medium. Chronic GVHD was induced by the injection of 90 × 106 DBA/2 (H-2d) mice parental cells into the tail vein of a normal, unirradiated B6D2F1 (H-2b/d) mouse. In the control PBS-treated and the SLC antagonist-treated groups, 90 × 106 DBA/2 spleen cells per mouse were injected into B6D2F1 recipients after incubation with PBS and murine SLC antagonist at a concentration of 50 μg/ml in RPMI 1640 medium supplemented with 2% FCS, at room temperature for 1 h, rotating with ROTATOR RT-50 (TAITEC, Saitama, Japan), respectively.
To determine whether in vivo administration of SLC antagonist inhibits the development of chronic GVHD, we designed the following administration groups: group 1, i.v. injection of 30 μg/mouse four times, 2 h before, just before, 2 h after, and 6 h after GVHD induction; and group 2, i.v. injection of 60 μg/mouse twice, 2 and 6 h after GVHD induction. These mice were observed for 16 days after GVHD induction.
Flow cytometric analysis of parental cell engraftment
The spleens of GVHD mice were harvested at 1, 5, 8, and 16 days after inoculation. Single-cell suspensions of splenocytes were prepared, and the number of cells per spleen was counted. Flow cytometric analysis was performed as described previously (41). After the erythrocytes had been removed with hemolysis buffer, the splenocytes were stained with FITC-conjugated anti-H-2b to distinguish parental from B6D2F1 cells. To further identify the splenocyte populations, anti-CD4, anti-CD8, and anti-B220 conjugated to PE were used. Host-B cell MHC class II expression is shown as the mean channel fluorescence for host Ia on B220+ lymphocytes using FITC-conjugated anti-I-Ab. Donor T cell engraftment is shown as the number of CD4+ or CD8+ lymphocytes that did not stain positively for MHC class II of the opposite parent. Data were collected on a FACScan flow cytometer and analyzed with CellQuest software (BD Biosciences, San Jose, CA).
For immunohistochemical analysis, cells were labeled with biotin sulfo-NHS-LC-biotin (Pierce) at a final concentration of 1 mg/ml in PBS with 5% FCS at 37°C for 30 min. After washing with PBS, cells (90 × 106/mouse) were suspended in RPMI 1640/2% FCS with or without murine SLC antagonist and injected into the tail veins of B6D2F1 mice. After 5 h, the mice were sacrificed, and their spleens were embedded in OCT (Tissue Tek, Elkhart, IN), snap-frozen on dry ice, and then stored at −70°C. Cryostat sections (5-μm thick) on slides were dried overnight and washed three times in PBS. The sections were incubated at room temperature for 15 min in 3% H2O2 to inhibit endogenous peroxidase activity, and then incubated with streptavidin/HRP (DAKO, Carpinteria, CA) at room temperature for 1 h. They were developed with a 3-amino-9-ethylcarbazole (AEC) substrate chromogen system (AEC as chromogen; DAKO). The sections were counterstained with hematoxylin.
Serum titers of total IgG and IgG1 of anti-DNA Abs were assessed by ELISA. The 96-well Maxisorp plates (Nalge Nunc International, Roskilde, Denmark) were incubated overnight at 4°C with 50 μl of calf thymus DNA (Sigma-Aldrich) at a concentration of 5 μg/ml in PBS. After blocking with 2% BSA, plates were incubated with 50 μl of diluted serum samples for 2 h at room temperature. After washing three times with PBS containing 0.1% Tween 20, HRP-labeled goat anti-mouse IgG or IgG1 was added to each well and the plates were kept at room temperature for 2 h. After the plates had been washed with PBS containing 0.1% Tween 20, the color was developed for 15 min in 100 μl of 3,3′,5,5′-tetramethylbenzidine (Sigma-Aldrich), stopped by adding 100 μl of 1 N HCl. The plates were read at 450 nm with IMMUNO-MINI (Nalge Nunc International) within 30 min.
Quantification of murine IL-2, IFN-γ, and IL-4
Spleen cells (5 × 106) were cultured in 1 ml of RPMI 1640 medium supplemented with 10% FCS, 5 mM 2-ME, and 10 μg/ml Con A (Sigma-Aldrich). The culture supernatants were harvested after 24 and 48 h and analyzed for murine IL-2, IFN-γ, and IL-4 with ELISA kits (Endogen, Webura, MA). Each sample was assayed in triplicate and the mean values were calculated.
Statistical analysis was performed using Student’s t test.
Chemotactic activity and Ca2+ influx induced by murine N-terminally truncated SLC analogs
To examine the role of the N-terminal region of murine SLC in determining function, we designed a series of analogs that were shortened at the N terminus (Fig. 1). As shown in Fig. 2, at a concentration of 5 μg/ml, three N-terminally truncated analogs, murine (m)SLC-1, mSLC-2, and mSLC-3, had lower chemotactic activity for L1.2/CCR7 cells (58 ± 11%, 55 ± 12%, and 42 ± 8%, respectively; p < 0.01) than that of the complete form of murine SLC. Another three analogs, mSLC-4, mSLC-5, and mSLC-6, which were truncated >4 aa residues from the N terminus, failed to induce chemotactic activity. Similar results were obtained with mouse spleen cells as the target cells (data not shown).
Transient intracellular Ca2+ mobilization was assayed as a second measure of function. The complete form of murine SLC was the most efficient in inducing cytosolic Ca2+ mobilization among the analogs tested in L1.2/CCR7 cells (Fig. 2). Induction of cytosolic Ca2+ mobilization by the N-terminally truncated analogs generally correlated with the results of chemotaxis. The mSLC-1, mSLC-2, and mSLC-3 induced a lower response (42 ± 11%, 38 ± 8%, and 30 ± 14%, respectively; p < 0.01) than that of the complete form, whereas mSLC-4, mSLC-5, and mSLC-6 did not induce significant cytosolic Ca2+ at a concentration of up to 50 μg/ml. From these findings, we suggest that those analogs truncated >4 aa residues from the N terminus showed loss of ability to induce chemotaxis and Ca2+ influx.
Blocking of chemotaxis and Ca2+ influx by murine N-terminally truncated SLC analogs
Next, we examined whether the three analogs, mSLC-4, mSLC-5, and mSLC-6, worked as antagonists. The mSLC-4, mSLC-5, and mSLC-6 inhibited SLC-stimulated chemotaxis (95 ± 5%, 90 ± 5%, and 92 ± 7% reduction, respectively; p < 0.01) and SLC-induced Ca2+ influx (97 ± 3%, 93 ± 5%, and 97 ± 3% reduction, respectively; p < 0.01) at a concentration of 5 μg/ml (Fig. 3).
The N-terminally truncated analogs were tested for CCR7 binding by competition for 125I-labeled SLC. The mSLC-1, mSLC-2, and mSLC-3 had a >4 times higher Kd (25, 66, and 96 nM, respectively) than that of the complete form of murine SLC (6.6 nM) when analyzed for binding to L1.2/CCR7 cells. The three inactive analogs, mSLC-4, mSLC-5, and mSLC-6, had a Kd (41, 164, and 84 nM, respectively) >6 times higher than that of the complete form of murine SLC. From these results, we decided to use mSLC-4 as an antagonist for the chronic GVHD experiment. We examined whether mSLC-4 induced internalization of the CCR7 receptor on L1.2/CCR7 cells, using anti-CCR7 mAb, CCR7.6B3 (25), which did not compete with the binding of mSLC. It was found that mSLC-4 did not induce internalization of the CCR7 receptor after binding to the receptor (data not shown).
SLC antagonist prevents the development of chronic GVHD
To determine whether SLC antagonist inhibits the development of chronic GVHD, chronic GVHD was induced by injecting DBA/2 spleen cells into B6D2F1 mice. In a preliminary experiment, we examined the optimum concentration of mSLC-4 for blocking the chemotaxis of DBA/2 spleen cells using in vitro chemotactic assays. At a concentration of >20 μg/ml for 90 × 106 DBA/2 spleen cells, the decrease of chemotaxis reached a plateau (data not shown). Therefore, we used mSLC-4 at a concentration of 50 μg/ml for this experiment. Furthermore, we compared cell viability, proliferation in response to the T cell mitogen Con A, and cytokine production between PBS-treated and SLC antagonist-treated splenocytes 5 h after treatment. There were no differences in cell viability (>90%) or Con A-stimulated proliferation between these two groups (data not shown). In addition, we also confirmed that there was no difference in production of cytokines such as IL-2, IL-4, and IFN-γ after Con A stimulation between these two groups (data not shown). After incubation with mSLC-4, spleen cells were injected into the tail veins of B6D2F1 mice. On day 16 after the induction, control PBS-treated mice exhibited the following findings typical of chronic GVHD, as described previously (10): splenomegaly, an increase in numbers of host B cells, and elevated serum levels of IgE and of IgG and IgG1 of anti-DNA Abs. The total number of spleen cells on day 16 after GVHD induction increased 1.8-fold: the number of cells on day 16 after GVHD induction was (105 ± 12) × 106 vs before GVHD induction (58 ± 8) × 106 (p < 0.05) (Fig. 4). The number of host B cells increased 2.2-fold: the number of cells on day 16 after GVHD induction was (62 ± 10) × 106 vs before GVHD induction (28 ± 6) × 106 (p < 0.05). In contrast, host CD4+ and CD8+ T cells possessed only 9.0% ((5.2 ± 0.8) × 106) and 7.0% ((4.0 ± 0.5) × 106) of total spleen cells, respectively, and their numbers had increased 1.4-fold ((7.3 ± 1.2) × 106 (p < 0.05)) and 1.2-fold ((4.6 ± 0.5) × 106 (p < 0.05)) on day 16 after GVHD induction. Therefore, the increase in numbers of spleen cells after GVHD induction is mainly due to the increase in numbers of host B cells. GVHD mice treated with SLC antagonist exhibited signs of chronic GVHD, but these signs were reduced compared with those of control mice. On day 16 after induction, total numbers of spleen cells and host B cells in SLC antagonist-treated GVHD mice were reduced to 67% ((70 ± 8) × 106 (p < 0.05)) and 40% ((25 ± 8) × 106 (p < 0.05)), respectively, of the values in the control PBS-treated GVHD mice, whereas there was no significant difference in the numbers of either CD4+ ((6.3 ± 0.7) × 106 vs (7.3 ± 1.2) × 106) or CD8+ ((4.4 ± 0.3) × 106 vs (4.6 ± 0.5) × 106) host T cells between SLC antagonist-treated and control PBS-treated GVHD mice.
Next, we compared the serum levels of IgE, and of IgG and IgG1 of anti-DNA Abs between SLC antagonist-treated GVHD mice and control PBS-treated GVHD mice. On day 16 after induction, the serum level of IgG anti-DNA Abs in SLC antagonist-treated GVHD mice was reduced to 35% of that in control PBS-treated GVHD mice (OD: SLC-antagonist-treated GVHD mice, 0.40 ± 0.12 vs control PBS-treated GVHD mice, 1.15 ± 0.12 (p < 0.05)) (Fig. 5). In addition, the serum IgG1 of anti-DNA Abs and IgE levels in SLC antagonist-treated GVHD mice were also reduced to 31% (OD: 0.041 ± 0.020 vs 0.132 ± 0.032 (p < 0.05)) and 52% (1.15 ± 0.20 vs 2.22 ± 0.28 μg/ml (p < 0.05)) of those in control PBS-treated GVHD mice, respectively. These findings showed that SLC antagonist ameliorated chronic GVHD. Furthermore, we confirmed these findings by analyzing survival 16 days after transplantation. As shown in Fig. 6, SLC-antagonist-treated GVHD mice showed a significant increase in survival compared with control PBS-treated GVHD mice (median survival: >80 vs 52 days, respectively; p < 0.05).
We examined the infiltration of donor spleen cells into spleens of B6D2F1 mice by injecting labeled donor spleen cells. To localize donor lymphocytes within the recipient’s spleen, we took advantage of cell biotinylation and immunohistological staining techniques. Frozen sections of spleens were stained with streptavidin/HRP and binding was visualized by AEC, resulting in a red color, and the sections were then counterstained with hematoxylin. As shown in Fig. 7, the control PBS-treated donor cells were spread diffusely within the periarteriolar lymphoid sheaths (PALS). In contrast, the number of SLC antagonist-treated donor cells in the PALS were reduced markedly (by 83%, p < 0.01). These findings suggest that addition of SLC antagonist inhibited the development of chronic GVHD by reducing the infiltration of CCR7-positive donor lymphocytes into the PALS of the recipient’s spleen.
SLC antagonist treatment changes the balance of donor CD4+ and CD8+ T cell numbers
To examine the mechanism by which the SLC antagonist prevents the development of chronic GVHD, we compared subsets of donor lymphocytes in the spleen between control PBS-treated GVHD mice and SLC antagonist-treated GVHD mice. Donor CD4+ T cells in control PBS-treated GVHD mice were activated and were increased in number, 9-fold ((3.5 ± 0.8) × 106 (p < 0.05)) on day 5 after GVHD induction, 11-fold ((4.4 ± 0.6) × 106 (p < 0.05)) on day 8, and 8-fold ((3.3 ± 1.1) × 106 (p < 0.05)) on day 16, compared with those on day 1 after GVHD induction ((0.4 ± 0.2) × 106) (Fig. 8). In contrast, donor CD8+ T cells in PBS-treated GVHD mice were not activated as efficiently and increased in number <2-fold ((1.2 ± 0.8) × 106, (1.6 ± 0.7) × 106, and (0.8 ± 0.4) × 106 on days 5, 8, and 16 after GVHD induction, respectively), compared with those on day 1 after GVHD induction ((0.8 ± 0.6) × 106). The numbers of donor CD4+ T cells in the spleens of SLC antagonist-treated GVHD mice were reduced to 51% ((1.8 ± 0.8) × 106), 50% ((2.2 ± 1.0) × 106), and 57% ((1.9 ± 1.0) × 106) on days 5, 8, and 16, respectively (p < 0.05), of the values in control PBS-treated GVHD mice. In contrast, over the 16 days after GVHD induction, there were no significant differences in the numbers of donor CD8+ T cells in the spleen between SLC antagonist-treated GVHD mice and control PBS-treated GVHD mice.
Next, we compared cytokine production of spleen cells between SLC antagonist-treated GVHD mice and control PBS-treated GVHD mice on day 8 after GVHD induction. The amount of IL-4 produced by spleen cells of SLC antagonist-treated GVHD mice was significantly lower than that of control PBS-treated GVHD mice, whereas there were no differences in IFN-γ and IL-2 production between SLC antagonist-treated GVHD mice and control PBS-treated GVHD mice (Table I). From these findings, the amelioration of chronic GVHD by treatment with an SLC antagonist appears to be due to a reduction in the numbers of activated donor CD4+ T cells and a decrease in IL-4 production, resulting in a reduction in the numbers of activated host B cells.
|.||IL-4 (pg/ml) .||IL-2 (pg/ml) .||IFN-γ (ng/ml) .|
|Control||535 ± 171||375 ± 124||11 ± 6|
|SLC antagonist||305 ± 92b||340 ± 85||8 ± 5|
|.||IL-4 (pg/ml) .||IL-2 (pg/ml) .||IFN-γ (ng/ml) .|
|Control||535 ± 171||375 ± 124||11 ± 6|
|SLC antagonist||305 ± 92b||340 ± 85||8 ± 5|
All experiments were done with groups of six mice. Similar results were obtained in one additional experiment. The statistical analysis was performed using Student’s t test.
p < 0.05 vs control PBS-treated GVHD mice.
In vivo administration of SLC antagonist ameliorated chronic GVHD
We examined whether in vivo administration of SLC antagonist prevented the development of chronic GVHD. For this experiment, we designed the following two different administration groups: group 1, i.v. injection of 30 μg/mouse four times, 2 h before, just before, 2 h after, and 6 h after GVHD induction; and group 2, i.v. injection of 60 μg/mouse twice, 2 and 6 h after GVHD induction. As shown in Fig. 9,A, total numbers of spleen cells and host B cells in group 1 were reduced to 71 and 67%, respectively (p < 0.05), of the values in the control PBS-treated GVHD mice on day 16 after induction, whereas there were no significant differences in the numbers of spleen cells and host B cells between group 2 and control PBS-treated GVHD mice. Amelioration of chronic GVHD in group 1 was also supported by findings such as a reduction in the numbers of donor CD4+ T cells in the recipient’s spleen (Fig. 9,B) and a decrease in the serum levels of IgE, and of IgG and IgG1 of anti-DNA Abs (Fig. 9 C). These results indicate that SLC antagonist administered after transplantation cannot prevent the development of chronic GVHD sufficiently, since donor T cells enter the lymphoid organs early after injection.
This is the first report to demonstrate the receptor antagonist properties of murine N-terminally truncated SLC analogs and to show that SLC antagonist inhibits the development of chronic GVHD in a murine model. Our findings are as follows: 1) SLC analogs truncated >4 aa from the N terminus failed to induce chemotaxis and Ca2+ influx by CCR7-expressing cells and also inhibited SLC-stimulated chemotaxis and SLC-induced Ca2+ influx completely; 2) the numbers of total spleen cells and host B cells and serum levels of IgE, and of IgG and IgG1 of anti-DNA Abs, in SLC antagonist-treated GVHD mice were significantly lower than those of control PBS-treated GVHD mice; and 3) this was due to a reduction in the numbers of activated donor CD4+ T cells and a decrease of IL-4 production, resulting in a reduced number of activated host B cells.
Truncation or modification of the few N-terminal amino acids of chemokines has been reported to lead to significant changes in functional activity and receptor binding (28, 29, 30, 31, 32, 33, 34, 35, 36). Deletion of the pyroglutamate residue at the N terminus of the complete form of monocyte chemoattractant protein (MCP)-1 results in at least a 50-fold decrease of agonistic activity on monocytes and basophils (28, 29). Deletion of the first two amino acids results in almost total loss of activity, although surprisingly, activity is regained on deletion of further amino acids. In the case of IL-8, the N-terminal residues 4, 5, and 6 are essential for receptor binding and triggering function (35, 36). RANTES loses agonistic potency and becomes a potent antagonist of chemokine binding when the first amino acid residue is modified artificially by addition of methionine or treatment with aminooxypentane (30, 32, 33). The naturally cleaved forms of MCP-2 and RANTES are also devoid of bioactivity (31, 34). The function of chemokines has also been further clarified from crystal structure determination and nuclear magnetic resonance data (42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56). In several chemokines such as IL-8, MCP-1, MCP-2, MCP-3, eotaxin, RANTES, MIP-1α, MIP-1β, fractalkine, and stromal cell-derived factor-1, it has been observed that the N-terminal region is essential for functional activity and that the loop immediately following the first two cysteines in the sequence, as well as the N-terminal region, plays an important role in receptor binding. We found that the N-terminally truncated murine SLC had lower activity and receptor binding than the complete form of murine SLC. The three analogs, mSLC-1, mSLC-2, and mSLC-3, which were truncated up to three amino acids from the N terminus, had only a 2-fold lower potency than the complete form of murine SLC. With further deletions from the mSLC-3 analog, biological activity was lost but receptor binding was retained with mSLC-4, mSLC-5, and mSLC-6. These findings show that at least four amino acid residues preceding the first cysteine are essential for the biological activity of murine SLC. Human SLC gives similar results to murine SLC (our unpublished observations).
Chronic GVHD remains the most common complication of allogeneic stem cell transplantation (SCT). This is probably related to changes in clinical SCT practice. The use of unrelated donors and related but nonhuman leukocyte Ag (non-HLA)-identical donors is expanding. The use of donor lymphocyte infusion to treat relapsed disease or to achieve full donor chimerism has resulted in the development of chronic GVHD. Furthermore, there is evidence to suggest that patients receiving allogeneic PBSCT have a lower incidence of acute GVHD but an equally high or a higher incidence of chronic GVHD than comparable patients receiving marrow grafts (3, 4, 5, 6, 7). Although the reason for this is unknown, it may be due to the transfer of a significantly larger dose of mature, immunocompetent T cells. Until now, two mechanisms have been suggested for the pathogenesis of chronic GVHD (1, 2). One is alloreactivity to minor histocompatibility Ags, and the other is a role of postthymic CD4+ T cells. The T cell precursors may undergo aberrant “thymic education” after SCT—a process that effectively makes them autoreactive. However, the exact pathogenesis of chronic GVHD remains ambiguous. Murine chronic GVHD resembling human chronic GVHD induced by PBSCT can be induced by the transfer of DBA/2 (H-2d) mouse spleen cells into (C57BL/6 × DBA/2) F1 (B6D2F1) (H-2b/d) mice (10). This model is a chronic disorder characterized by a process in which donor CD4+ T cells become activated, secrete predominantly Th2-associated cytokines, and stimulate autoreactive B cells to differentiate into autoantibody-secreting cells. Serum Ig levels, including IgE levels, are elevated in these mice, whereas the function of in vitro CD8+ CTL is reduced and donor anti-host CTL are not detected. An autoimmune disease resembling human systemic lupus erythematosus also develops in this model.
It has been reported that CCR7 and its ligands, SLC and ELC, play a central role in the homing and traffic of T cells and dendritic cells into secondary lymphoid organs such as the peripheral lymph nodes, spleen, and Peyer’s patches (14, 15, 16, 17, 18, 19, 20, 21, 22). Lymphocyte transmigration from peripheral blood to secondary lymphoid organs occurs through HEVs and is a multistep process involving selectin-supported rolling, followed by a triggering event, and then firm integrin-mediated adhesion (57). SLC is highly expressed by HEVs and is active in inducing integrin-mediated adhesion of lymphocytes (18, 20). Once lymphocytes have crossed HEVs in lymph nodes and Peyer’s patches, T cells localize efficiently in T cell-rich areas, and B cells in B cell-rich areas. SLC, in addition to being expressed by HEVs, is expressed by stromal cells within T cell-rich areas, while ELC is produced by macrophages, by dendritic cells, and by some nonhemopoietic cells in T cell-rich areas (19, 20, 21, 22). In both CCR7-deficient mice and plt mice, the organization of cells in T cell-rich areas is severely disturbed, and there is also defective entry of cells across HEVs (26, 27). Therefore, CCR7 and its ligands are not only required for guiding T cells through HEVs, but also for directing T cells to their corresponding functional microenvironment once they have entered the lymphoid organs. On the basis of these results, we aimed at ameliorating chronic GVHD using an SLC antagonist that blocked the homing of donor CCR7-positive T cells into secondary lymphoid organs in which donor CD4+ T cell-host B cell interaction is mainly elicited. Infiltration into the PALS of the recipient’s spleen of CCR7-positive donor lymphocytes incubated with SLC antagonist was markedly reduced, compared with infiltration in control GVHD mice. Consequently, the number of activated donor CD4+ T cells and IL-4 production in SLC antagonist-treated GVHD mice were reduced significantly, resulting in a decrease in the numbers of activated autoreactive host B cells. This was also proven from the finding that serum levels of IgE, and of IgG and IgG1 of anti-DNA Abs, in SLC antagonist-treated GVHD mice were significantly lower than those in control GVHD mice.
Several studies have reported the role of chemokines and their receptors in the pathogenesis of GVHD (58, 59, 60). In addition to production of MIP-1α by donor T cells, MIP-1α expression is significantly increased in acute GVHD target organs such as the liver, lung, and spleen after transfer of allogeneic lymphocytes compared with syngeneic lymphocytes (58, 59). Treatment with anti-MIP-1α or anti-CCR5 Abs reduces liver damage in GVHD (58). These results suggest that MIP-1α-induced migration of CCR5-expressing CD8+ T cells plays a significant role in the occurrence of acute GVHD. In a murine sclerodermatous GVHD model, early elevated cutaneous mRNA expression of TGF-β1, elevated CC chemokines, MCP-1, MIP-1α, and RANTES preceded subsequent skin and lung fibrosis, suggesting that TGF-β1-producing donor mononuclear cells and these chemokines might be important in the early pathogenesis of sclerodermatous GVHD (60). Therefore, these findings and our present results indicate that receptor antagonists for chemokines could be considered as therapeutic target for the prevention and treatment of GVHD.
Finally, there are several problems in applying the findings obtained from this model to the treatment of GVHD in humans. T cell depletion of the donor graft can dramatically reduce the incidence of GVHD (reviewed in Refs. 1 and 2). However, this reduction in GVHD did not translate into improved overall survival because of delayed immune reconstitution and unexpectedly high rates of graft failure and disease recurrence. Because SLC antagonist-treated donor T cells cannot traffic sufficiently to areas that contain donor-reactive host T cells, treatment with SLC antagonist may affect immune reconstitution and graft failure. In human allogeneic transplantation, recipients are treated with a cytotoxic regimen with additional whole-body irradiation to reduce the frequency of graft rejection. However, this treatment is associated with increased frequency and severity of GVHD with histopathological damage. In contrast, the model in this study is a chronic GVHD model with little to no histopathological damage, in which unirradiated B6D2F1 mice are infused with DBA/2 spleen cells. Therefore, although the mechanism of chronic GVHD induced in this mouse model is not necessarily identical with that of chronic GVHD in humans, SLC antagonist may provide a new potentially useful approach for the prevention of chronic GVHD.
This work was supported in part by a grant-in-aid from the Ministry of Education, Science and Culture of Japan.
Abbreviations used in this paper: GVHD, graft-vs-host disease; SCT, stem cell transplantation; PBSCT, peripheral blood SCT; SLC, secondary lymphoid tissue chemokine; ELC, EBI-1-ligand chemokine; HEV, high endothelial venule; MIP, macrophage inflammatory protein; AEC, 3-amino-9-ethylcarbazole; m, murine; PALS, periarteriolar lymphoid sheath; MCP, monocyte chemoattractant protein.