PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Immunol. Author manuscript; available in PMC 2014 January 15.
Published in final edited form as:
PMCID: PMC3891913
NIHMSID: NIHMS530520

Loss of B7-H1 expression by recipient parenchymal cells leads to expansion of infiltrating donor CD8+ T cells and persistence of GVHD

Abstract

Previous experimental studies have shown that acute graft-versus-host disease (GVHD) is associated with two waves of donor CD8+ T cell expansion. In the current studies, we used in vivo bioluminescent imaging (BLI), in vivo BrdU-labeling, and three different experimental GVHD systems to show that B7-H1 expression by recipient parenchymal cells controls the second wave of alloreactive donor CD8+ T cell expansion and the associated second phase of GVHD. Loss of B7-H1 expression by parenchymal cells during the course of GVHD was associated with persistent proliferation of donor CD8+ T cells in GVHD target tissues and continued tissue injury, whereas persistent expression of B7-H1 expression by parenchymal cells led to reduced proliferation of donor CD8+ T cells in GVHD target tissues and resolution of GVHD. These studies demonstrate that parenchymal cell expression of B7-H1 is required for tolerizing infiltrating T cells and preventing the persistence of GVHD. Our results suggest that therapies designed to preserve or restore expression of B7-H1 expression by parenchymal tissues in the recipient could prevent or ameliorate GVHD in humans.

Keywords: Hematopoietic cell transplantation, Graft versus host disease, Co-inhibitory molecule B7-H1, Anti-CD3, Lymphopenia

Introduction

We recently showed that donor CD8+ T cells transplanted into allogeneic recipients conditioned with 800 cGy TBI go through two waves of expansion, each associated with a distinct phase of GVHD. The first-wave expansion and the first-phase of GVHD usually begins at 5 to 10 days after transplantation and is not lethal. The second wave of expansion and the second phase of GVHD usually begins at 30 days after transplantation and is lethal (1). Zhang et al. have shown that memory CD8+ T cells isolated from recipients at 30 days after transplantation induce severe GVHD in secondary recipients, indicating that memory CD8+ T cells mediate the persistence of GVHD (2-3). In contrast, we have shown that donor CD8+ T cells in recipients conditioned with a CD3-specific antibody before transplantation go through only one wave of expansion, which is not associated with clinical evidence of GVHD (1). Previous studies have not defined the mechanisms that mediate or prevent the second wave of donor memory CD8+ T cell expansion and the second phase of GVHD.

B7-H1 (also called PD-L1) is constitutively expressed by hematopoietic cells, including resting T cells, B cells, DCs, and macrophages (4-7). B7-H1 protein is not normally expressed by parenchymal cells but can be induced by proinflammatory cytokines such as IFN-γ and TNF-α (8-11). Activated T cells upregulate expression of PD-1, and its interaction with B7-H1 induces T cell anergy, exhaustion, and apoptosis (5, 12). We and others reported that blockade of B7-H1 and PD-1 interaction augmented acute GVHD (13-14). B7-H1 expression by recipient hematopoietic cells and parenchymal cells induced alloreactive CD8+ T cell exhaustion and reduced graft-versus-leukemia (GVL) effects in TBI-conditioned recipients during the first 2 weeks after transplantation (15) and in recipients given delayed donor lymphocyte infusion (16-17). GVHD recipients usually have lymphopenia due to thymus damage and reduction of de novo developed T cells as well as deficient proliferation of donor T cells in the peripheral lymphoid tissues (18). However, it is still unknown how GVHD can persist in lymphopenic recipients and how parenchymal cell expression of B7-H1 regulates the GVHD development in the lymphopenic recipients.

In the current studies, we evaluated the role of B7-H1 expression by recipient hematopoietic cells and parenchymal cells in regulating the second wave of donor CD8+ T cell expansion and the second phase of GVHD, using three experimental GVHD systems, including TBI-conditioned wild-type (WT) recipients, anti-CD3-conditioned recipients with or without B7-H1 deficiency, and unconditioned Rag-2−/− recipients with or without parenchymal cell expression of B7-H1. We used in vivo bioluminescent (BLI) and in vivo BrdU labeling to evaluate the proliferation of injected donor CD8+ T cells in recipient lymphoid and GVHD target tissues. Our results demonstrate that loss of B7-H1 expression by recipient parenchymal cells during the course of GVHD permits the second wave of donor CD8+ T cell expansion infiltrating GVHD target tissues, leading to persistence of GVHD and resulting in death.

Materials and Methods

Mice

C57BL/6 (H-2b), BALB/c (H-2d), and Rag-2−/−BALB/c (H-2d) mice were purchased from NCI Laboratory (Frederick, Maryland) or Taconic Farm (Germantown, New York). Luciferase-transgenic (luc+) C57BL/6, congenic C57BL/6 (CD45.1) and B7-H1−/− BALB/c mice were generated as previously described (19-20). B7-H1−/−Rag-2−/− mice were generated by back-crossing B7-H1−/− BALB/c to Rag-2−/− BALB/c mice. Mice were age-matched for each experiment, and all were 8-12 weeks of age. All animals were maintained in a pathogen-free room at City of Hope Animal Research Facilities (Duarte, CA). Animal use protocols were approved by the institutional review committee.

Antibody production, conditioning of recipients, transplantation, assessment of GVHD and in vivo bioluminescent imaging (BLI)

Production of anti-CD3 mAb (145-2C11) and anti-B7-H1 mAb (10B5) has been previously described (1, 21). Recipients were conditioned with 800 cGy TBI from a 137Cs source 8 hours before transplantation. Alternatively, recipients were conditioned with one i.v. injection of anti-CD3 (5 μg/g) and three daily i.p. injections of vorinostat (40 μg/g) and busulfan (8.75 μg/g). BM and sorted CD8+ T cells from C57BL/6 donors were injected i.v. into BALB/c recipients. Recipients were monitored for donor CD8+ T cell expansion, clinical GVHD and survival. The assessment and scoring of clinical GVHD and the procedure for monitoring T cell expansion by in vivo BLI have been described previously (22).

Isolation of MNC from GVHD target tissues and flow cytometric analysis

Procedures for isolating mononuclear cells (MNC) from GVHD target tissues have been described previously (23-24). Abs to mouse CD3ε (145-2C11), TCRβ (H57-597), CD4 (RM4-5), CD8α (53-6.7), B220 (RA3-6B2), CD11b/Mac-1(M1/70), Gr-1(RB6-8C5), H-2b (AF6-88.5), H-2d (34-2-12), PD-1(CD279, J43), B7-H1(CD274, 1-111A), and IL-7α (CD127) were purchased from BD Pharmingen (San Jose, CA, USA), ebioscience (San Diego, CA) or R&D system (Minneapolis, MN USA). The four-laser MoFlo Immunocytometry System (DakoCytomation, DK-2600 Glostrup, Denmark) or a CyAn immunocytometer (Dako Cytomation, Fort Collins, CO) was used for flow cytometry, and data were analyzed with FlowJo software (Tree Star, Ashland, OR) as previously described (25-27).

In vivo BrdU labeling

Bromodeoxyuridine (BrdU) 2 mg in PBS was injected i.p. daily for three days followed by tissue collection and analysis of BrdU incorporation using the BrdU flow kit protocol (BD Pharmingen).

Proliferation Assays

Ex-vivo CD4+ T cell proliferation was measured as previously described (22, 25). Briefly, sorted CD4+ T cells (2×105) were incubated with irradiated (3 Gy) dendritic cells (1×105) in complete RPMI media containing 10% fetal bovine serum, penicillin/streptomycin, L-glutamine, and 2-mercaptoethanol at the bottom of a 96 U-well plate for 5 days. 3H-TdR (1μCi/well) was added to the culture 18 hours before harvest.

Histopathology

Tissue specimens were fixed in 10% formalin and embedded in paraffin. Sections stained with H&E, and histologic evidence of GVHD was assessed according to a scoring system described previously without knowledge of the assignment to experimental groups (22-23, 28). Pictures were taken with a Pixera (600CL) cooled charge-coupled device camera (Pixera, Los Gatos, CA).

Hepatocyte isolation and Real-Time PCR of B7-H1 and IFN-γR mRNA

Hepatocytes were isolated according to standard procedure previously described (29). Isolation of total tissue RNA and synthesis of first strand cDNA have been described previously (23-24). mRNA was quantified by real-time quantitative PCR, using Applied Biosystems 7300 Fast Real-Time PCR System (Applied Biosystems, Forest City, CA).

Primers for B7-H1 were:

  • Forward (F): 5’-TGCCCTTCAGATCACAGACGTCAA-3’;
  • Reverse (R): 5’-TGGCTGGATCCACGGAAATTCTCT-3’.

Primers for INF-γR were:

  • F: 5’-AATAAGGATCCTGTGGGCACTGCT-3’,
  • R: 5’-AAACTCACAGATGCTAGCCCACCT-3’.

Relative expression levels of genes were normalized within each sample to the housekeeping gene GAPDH (glyceraldehyde 3-phosphate dehydrogenase). Quantitative PCR results are calculated relative to expression levels before transplantation.

Measurement of cytokines

Culture supernatant cytokines (IL-2, IFN-γ, TNF-α, and IL-6) were measured with ELISA kits (BD Biosciences Pharmingen) as previously described (26).

Hydrodynamic injection of B7-H1 cDNA

B7-H1 plasmid was produced in Dr. Lieping Chen's laboratory. The procedure for hydrodynamic injection has been described previously (30). Injections of 1ml PBS containing 20ug B7-H1 plasmid or control vector were given i.v. within 10 seconds.

Statistical Analysis

Survival was evaluated by the log-rank test with GraphPad Prism version 4.0 (GraphPad Software). Means were compared by the unpaired two-tailed Student t test.

Results

The second wave of donor CD8+ T cell expansion and GVHD was associated with loss of B7-H1 expression by hepatocytes

Since activated T cells usually upregulate expression PD-1, and since B7-H1 interaction with PD-1 could tolerize the activated T cells, we tested the role of host tissue expression of B7-H1 in regulating donor CD8+ T cell expansion and persistence of GVHD in TBI-conditioned recipients with or without anti-CD3 pretreatment. BALB/c recipients were pretreated with anti-CD3 (5 μg/g) or PBS 7 days before transplantation. On day 0, recipients were exposed to 800 cGy TBI, and 8 hr later, they were given luc+CD8+ T cells (20 × 106) and wild-type BM cells (50 × 106) from C57BL/6 donors. Donor T expansion was monitored with in vivo BLI. In the absence of anti-CD3 treatment, donor CD8+ T cells went through two waves of expansion and caused two-phases of clinical GVHD (Fig. 1A-C), consistent with results reported previously (1). In anti-CD3-pretreated recipients, however, donor CD8+ T cells went through only one wave of expansion, which was associated with little evidence of clinical GVHD (Fig. 1A-C). Thus, anti-CD3 pretreatment prevented the second wave of donor CD8+ T cell expansion and the development of GVHD.

Fig. 1
Loss of B7-H1 expression by parenchymal tissues in TBI-conditioned recipients was associated with chronic inflammation

We compared hepatocyte expression of B7-H1 on days 0, 10, 20, 30, and 40 after transplantation in recipients conditioned with or without anti-CD3. As shown in Fig. 1D, at day 0 (before transplantation) hepatocytes did not show detectable levels of B7-H1 in either group, but 10 days later, hepatocytes upregulated expression of B7-H1 in both groups. In the absence of anti-CD3 treatment, expression of B7-H1 by hepatocytes decreased beginning at 20 days after transplantation and became undetectable by 40 days after transplantation. In anti-CD3 pretreated recipients, expression of B7-H1 persisted throughout the entire time period. At 40 days after transplantation, levels of B7-H1 expression were much lower in hepatocytes from recipients without anti-CD3 pretreatment compared to those that had received anti-CD3 (Figure 1D). These results suggested that down-regulated expression of B7-H1 in recipient parenchymal tissue contributes to the second wave expansion of donor CD8+ T cells and the second phase of GVHD.

Loss of B7-H1 expression was not caused by an intrinsic defect in hepatocytes but was associated with increased IL-6 production by infiltrating donor cells

Since IFN-γ can induce parenchymal cell expression of B7-H1 (8-10), and since we observed that induction of host parenchymal tissue expression of B7-H1 required donor T cell production of IFN-γ (23), recipients with and without anti-CD3 pretreatment were given a single injection of IFN-γ (150 μg) on day 30 after transplantation. 48 hours later, hepatocytes from recipients in both groups expressed high-levels of B7-H1 (Fig. 1E). These results indicate that down-regulated expression of B7-H1 by hepatocytes at 30 days after transplantation does not result from an intrinsic cellular defect and must be caused by extrinsic factors.

To identify the extrinsic factor(s), we compared the IFN-γ production by mononuclear cells infiltrating the liver and expression of IFN-γ receptor (IFN-γR) by hepatocytes in recipients with or without anti-CD3 pretreatment. No significant differences were found in the two groups (Fig. 1F). In evaluating production of IL-6, TNF-α, and IL-2 by infiltrating mononuclear cells, we found that IL-6 production was more than 5-fold higher in recipients without anti-CD3 pretreatment as compared to those with anti-CD3 pretreatment (p<0.01), with no differences in IL-2 or TNF-α production (Fig. 1G). IL-6 is associated with chronic tissue inflammation (31-32), and IL-6 can inhibit IFN-γ/IFN-γR signaling and potentially down-regulate B7-H1 expression (33). These results indicate that down-regulated expression of B7-H1 by hepatocytes is associated with chronic inflammation and increased IL-6 production.

Over-expression of B7-H1 by hepatocytes inhibited the second wave of donor CD8+ T cell expansion and ameliorated the second phase of GVHD

We used a gain-of-function experiment to test the role of tissue expression of B7-H1 in tolerizing infiltrating donor T cells. It has been shown that hydrodynamic injection of pCMV-luciferase cDNA could force hepatocytes to express luciferase protein(30). Therefore, we first tested whether hydrodynamic injection of B7-H1 cDNA could induce hepatocyte expression of B7-H1. B7-H1 cDNA or control plasmid was given via hydrodynamic injection into B7-H1−/− BALB/c mice. 24 hours after the injection, hepatocytes showed markedly upregulated expression of B7-H1 mRNA and cell surface protein (Sup-Fig.1).

We then tested whether hydrodynamic injection of B7-H1 cDNA could ameliorate ongoing GVHD. Accordingly, BALB/c recipients were conditioned with 800 cGy TBI without anti-CD3 preconditioning. The number of donor CD8+ T cells injected on day 0 was decreased from 20 × 106 to 5 × 106 in order to reduce the severity of GVHD and to improve the ability of the recipients to tolerate the hydrodynamic injection procedures. Hydrodynamic injections of B7-H1 cDNA or control plasmid were given 20 days after transplantation, when the recipients had recovered from the first phase of clinical GVHD. B7-H1 expression was upregulated 24 hours after injection of B7-H1 cDNA but not after injection of control cDNA (Fig. 2A). Compared to recipients treated with vector control, recipients treated with B7-H1 cDNA showed significantly reduced second-phase clinical GVHD (P<0.01, Fig. 2B). Recipients treated with B7-H1 cDNA also showed improved survival as compared to the control (92% versus 69%, P<0.05, Fig. 2C). 50 days after transplantation, donor-type CD8+ T cells were harvested from the liver and stimulated ex-vivo with immobilized anti-CD3. The proliferation of donor CD8+ T cells from the liver of recipients treated with B7-H1-cDNA was 50% lower than seen with CD8+ T cells from the control recipients (P<0.01, Fig. 2D). Less lymphocyte infiltration was consistently observed in the liver of recipients treated with B7-H1 compared to controls (Fig. 2E). These results support the notion that GVHD target tissue expression of B7-H1 can inhibit the expansion of infiltrating donor T cells and ameliorate the second phase of GVHD.

Fig. 2
Induction of hepatocyte expression of B7-H1 via hydrodynamic injection of B7-H1 cDNA inhibited tissue infiltrating donor CD8+ T cell expansion

Absence of B7-H1 expression allowed the second wave of donor CD8+ T cell expansion and exacerbated GVHD

In previous studies, we showed that donor CD8+ T cells did not cause GVHD in recipients prepared with a regimen of anti-CD3 with low-dose of Vorinostat (26). Under this GVHD preventative anti-CD3-based conditioning, donor CD8+ T cells showed a single wave expansion (Sup-Fig. 2). We compared wild-type (WT) and B7-H1−/− recipients conditioned with this GVHD preventative regimen to test whether expression of B7-H1 in the recipient controls donor CD8+ T cell expansion and prevents GVHD. In WT recipients, donor CD8+ T cells showed a slow and weak single wave of expansion (Fig. 3A & B). In B7-H1−/− recipients, donor CD8+ T expansion was rapid and strong, reaching an initial peak at approximately 13 days after transplantation and declining slightly thereafter. Donor CD8+ T cells then showed a second wave of expansion beginning at approximately 30 days after transplantation. By day 40 after transplantation, the BLI intensity of donor CD8+ T cells in B7-H1−/− recipients was more than 10-fold higher than in WT recipients (P<0.01).

Fig. 3
Expression of B7-H1 by recipient tissues prevented the second wave of donor CD8+ T cell expansion and the second phase of GVHD

The differences in CD8+ T cell expansion kinetics correlated with manifestations of GVHD. WT recipients showed little evidence of GVHD, and all survived for more than 80 days (Fig. 3C & D). B7-H1−/− recipients showed two phases of clinical GVHD that correlated with the two waves of donor CD8+ T cell expansion. The first phase of GVHD peaked at approximately 13 days after transplantation, as indicated by moderate diarrhea, ruffled fur, and weight-loss. The second phase of GVHD began at 35 days after transplantation, as indicated by severe diarrhea, hair-loss, and weight-loss, and most of the recipients died between days 40 and 60 after transplantation.

At 50 days after transplantation, we compared the yield of CD4+CD8+ double-positive (DP) thymocytes and the percentage and yield of donor-type CD8+ T cells in the spleen, liver, and skin of WT and B7-H1−/− recipients. The numbers of DP thymocytes and splenic MNC were approximately 40-fold higher in WT recipients than in B7-H1−/− recipients (P<0.01, Fig. 3E & F). While the yield of donor-type CD8+ T cells in the spleen was approximately 7-fold higher in WT recipients than in B7-H1−/− recipients, the number of donor-type CD8+ T cells in the liver and skin was approximately 2-5 fold lower in WT recipients than in B7-H1−/− recipients (P<0.01, Fig. 3G). Histopathological scores were consistently lower in WT recipients than in B7-H1−/− recipients (P<0.01, Fig. 3H and Sup-Fig. 3). These results demonstrate that host tissue expression of B7-H1 controls donor CD8+ T expansion and prevents GVHD in the recipients conditioned with a GVHD preventative anti-CD3-based regime. The strong in vivo BLI signal and the high yield of donor CD8+ T cells in GVHD target tissues but not in the spleen of B7-H1−/− recipients at 50 days after transplantation suggest that the absence of B7-H1 permits the proliferation of donor CD8+ T cells in GVHD target tissues.

Parenchymal tissue cell expression of B7-H1 prevented the second wave of donor CD8+ T cell expansion and the second phase of GVHD in Rag-2−/− recipients

We tested whether host parenchymal tissue cell expression of B7-H1 was required for prevention of the second wave of donor CD8+ T expansion and the second phase of GVHD, using unconditioned Rag-2−/− recipients in order to avoid any potential influence of TBI or anti-CD3 conditioning. First, we established chimeras that had expression of B7-H1 only in parenchymal tissues by transferring B7-H1−/−Rag-2−/− BM into Rag-2−/− mice. We then evaluated the effect of B7-H1 expression in parenchymal tissues by injecting blocking anti-B7-H1 mAb or control rat-IgG every other day for 30 days beginning on day 14 after transplantation. Recipients treated with control IgG showed only mild signs of GVHD and more than 90% survived for more than 80 days, but the recipients treated with anti-B7-H1 mAb showed progressive clinical GVHD, and almost all died by 40 to 60 days after transplantation (P<0.01, Fig. 4A & B). These results suggest that expression of B7-H1 by parenchymal cells in the recipient is required to control late-stage GVHD mediated by donor CD8+ T cells.

Fig. 4
Expression of B7-H1 by parenchymal tissues prevented the second wave of donor CD8+ T cell expansion and the second phase of GVHD in Rag-2−/− recipients

We also established chimeras that expressed B7-H1 in hematopoietic cells but not in parenchymal tissues by transferring Rag-2−/− BM cells into B7-H1−/−Rag-2−/− recipients. Control chimeras that expressed B7-H1 both in hematopoietic cells and in parenchymal tissues were established by transferring Rag-2−/− BM into Rag-2−/− recipients. In control chimeras, donor CD8+ T cells went through a weak single wave of expansion accompanied by minimal evidence of GVHD. More than 90% of these recipients survived for at least 80 days (Fig. 4 C – F). In chimeras with B7-H1 expression limited to hematopoietic cells, donor CD8+ T cells showed a rapid and progressive expansion accompanied by development of severe GVHD, and most of the recipients died within 40-60 days. These results demonstrate that expression of B7-H1 by parenchymal tissues in the recipient is required to control the second wave of donor CD8+ T cell expansion and to prevent the persistence of GVHD, whereas expression of B7-H1 by recipient hematopoietic cells is not sufficient for these effects.

The second wave of donor CD8+ T cell expansion occurred within GVHD target tissues

Further experiments were carried out to characterize differences in the behavior of donor CD8+ T cells according to the presence of GVHD in recipients conditioned with 800 cGy TBI or the absence of GVHD in recipients conditioned with anti-CD3-base regimen. The injected and the de novo developed donor-type T cells were distinguished by congenic marker CD45.2 (Fig. 5). At 35 days after transplantation, recipients with GVHD had marked reductions in percentage and yield of CD4+CD8+ thymocytes and splenic donor-type T cells, especially de novo developed T cells, as compared to recipients without GVHD (P<0.01, Fig. 5 & 6), indicating lymphopenia is associated with GVHD. In contrast, the percentage and yield of donor-type T cells, especially the injected donor CD8+ T cells, in the GVHD target tissues liver and skin was markedly higher in recipients with GVHD than in those without GVHD (P<0.01, Fig. 5 & 6). These results suggested that donor CD8+ T cells proliferate within GVHD target tissues in lymphopenic recipients during the second wave of expansion and the second phase of GVHD.

Fig. 5
Comparison of percentage of CD4+CD8+ thymocytes and percentage of injected donor CD8+ T cells in spleen and GVHD target tissues of recipients with and without GVHD
Fig. 6
Yield of CD4+CD8+ thymocytes and injected donor CD8+ T cells in spleen and GVHD target tissues in recipients with and without GVHD

To test this hypothesis, recipients with and without GVHD were i.p. injected with BrdU daily for 3 days beginning at 35 days after transplantation, and the percentages of BrdU+ proliferating cells among the injected donor-type CD8+ T cells were measured by flow cytometry. In recipients with GVHD, the injected donor CD8+ T cells showed little proliferation in the spleen and lymph nodes and much stronger proliferation in the GVHD target tissues such as liver, gut, and skin (P<0.01, Fig. 7 and Sup-Fig. 4). In recipients without GVHD, the injected donor CD8+ T cells showed much less proliferation in both the spleen and in GVHD target tissues (Fig. 7A & Sup-Fig.4). The rapid proliferation of donor CD8+ T cells infiltrating the liver of recipients with GVHD was associated with down-regulated expression of B7-H1 expression by hepatocytes. The slower proliferation of donor CD8+ T cells infiltrating the liver of recipients without GVHD was associated with persistent expression of B7-H1 by hepatocytes (Fig. 7B). These results indicate that proliferation of donor CD8+ T cells in GVHD target tissues correlates inversely with the level of B7-H1 expression.

Fig. 7
Comparison of proliferation of the injected donor CD8+ T cells in the spleen and GVHD target tissues in recipients with or without GVHD

Since it has been reported that tissue expression of B7-H1 induced donor CD8+ T cells into an exhausted state wherein the T cells up-regulate PD-1 and down-regulate of IL-7Rα (16), we also compared the expression of PD-1 and IL-7Rα by donor CD8+ T cells from the liver of recipients with or without GVHD on day 35 after transplantation. In both groups, the donor CD8+ T cells infiltrating the liver significantly upregulated expression of PD-1 and down-regulated expression of IL-7Rα as compared to donor CD8+ T cells before transplantation (P<0.01, Fig. 7C). The injected donor CD8+ T cells from recipients with GVHD expressed approximately 2 fold lower levels of PD-1 but 2-fold higher levels of IL-7Rα compared to those from recipients without GVHD (P<0.01, Fig. 7C). After ex-vivo stimulation with immobilized CD3 antibody, the injected CD8+ T cells isolated from the liver of recipients with GVHD proliferated 10-fold more vigorously than those from the liver of recipients without GVHD (P<0.01, Fig. 7D). These results indicate that loss of B7-H1 expression by hepatocytes in recipients with GVHD is associated with activation and expansion of infiltrating donor T cells and that persistent expression of B7-H1 by hepatocytes in recipients without GVHD is associated with donor CD8+ T cell tolerance.

Discussion

In this report, we have demonstrated that expression of B7-H1 by parenchymal cells in the recipient prevents the proliferation and expansion of donor CD8+ T cells infiltrating GVHD target tissues and aborts the persistence of GVHD. Loss of B7-H1 expression by parenchymal cells during the initial phase of GVHD leads to a second wave of alloreactive donor CD8+ T cell expansion in GVHD target tissues and the persistence of GVHD. Anti-CD3 treatment before transplantation maintained parenchymal cell expression of B7-H1, preventing the second wave of donor CD8+ T cell expansion and preventing the persistence of GVHD.

The second wave of donor CD8 + T cell expansion was consistent with proliferation of memory donor T cells

Previous reports by Zhang et al. and us showed that injected donor CD8+ T cells in TBI-conditioned recipients first expanded and then contracted ~two weeks after transplantation (1-2). The residual CD8+ T cells showed a memory-like phenotype of CD44hiCD122hiCD25lo, and adoptive transfer of these cells into secondary recipients induced severe GVHD (2). In the current studies, we showed that the donor CD8+ T cells had two waves of expansion in TBI-conditioned recipients. The first wave took place mainly in lymphoid tissues and then contracted. The second wave took place mainly in GVHD target tissues and was associated with lethal GVHD. This observation provides direct evidence that tissue infiltrating memory CD8+ T cells mediate the second wave of expansion and the persistence of GVHD in TBI-conditioned recipients.

The second wave of donor CD8 + T cell expansion took place in GVHD target tissues in recipients with lymphopenia in secondary lymphoid organs

We observed that only small percentage and low numbers of donor T cells were present in the spleen and lymph nodes during the second wave of donor CD8+ T cell expansion in TBI-conditioned recipients. Most of these cells were derived from the injected donor CD8+ T cells, and few had developed de novo from the thymus. In contrast, donor T cells and the injected donor CD8+ T cells had expanded in GVHD target tissues. In vivo BrdU-labeling assay showed that the injected donor CD8+ T cells proliferated much more vigorously in GVHD target tissues such as liver, gut, and skin as compared to the spleen and lymph nodes. This observation is consistent with a recent report that memory CD8+ T cells were able to proliferate in non-lymphoid tissues (34). These observations may help to explain why animal recipients and patients can have persistent clinical GVHD despite severe lymphopenia.

Down-regulated expression of B7-H1 in parenchymal tissues was associated with chronic tissue inflammation and increased production of IL-6

We previously showed that IFN-γ produced by donor T cells induces expression of B7-H1 by parenchymal tissues in the recipients with GVHD (23). In the current study, we observed that hepatocytes expressed similar levels of IFN-γR and that infiltrating mononuclear cells produced similar levels of IFN-γ in recipients with or without down-regulated expression of B7-H1. We also observed that down-regulated expression of B7-H1 by hepatocytes in recipients with GVHD could be reversed by in vivo administration of IFN-γ. These observations indicate that the down-regulation of B7-H1 expression of hepatocytes of GVHD recipients does not result from an intrinsic defect or from a direct reduction of IFN-γ/IFN-γR signaling, since enhanced IFN-γ/IFN-γR signaling could overcome the extrinsic factor and restore B7-H1 expression. The mononuclear cells infiltrating the liver of the recipients with hepatocyte down-regulation of B7-H1 expression produced 5 times more IL-6 than those from recipients that maintained hepatocyte expression of B7-H1. IL-6 is usually produced by cells and tissues involved in chronic inflammation (31), and IL-6/IL-6R signaling can upregulate suppressor of cytokine signaling 1 (SOCS1) that down-regulates IFN-γ/IFN-γR signaling (33). Thus, we speculate that a high level of IL-6 signaling accounts for the loss of B7-H1 expression by hepatocytes in recipients with GVHD.

The second wave expansion of injected donor CD8 + T cells in GVHD target tissues had a critical role in the pathogenesis of GVHD

We observed that infiltrating donor CD8+ T cells in the liver tissues of recipients without GVHD appeared to be exhausted, with upregulated expression of PD-1 and down-regulated expression of IL-7Rα. In contrast, the infiltrating donor CD8+ T cells in the liver tissues of recipients with GVHD down-regulated expression of PD-1 and up-regulated expression of IL-7Rα, and these cells actively proliferated in GVHD target tissues. We also observed that the loss of parenchymal tissue expression of B7-H1 in GVHD recipients resulted in a second wave of donor CD8+ T cell expansion and lethal GVHD. Therefore, we theorize that, while alloreactive T cell interaction with host APCs in lymphoid tissues has a critical role in the initial activation of donor T cells (35-37), maintenance of B7-H1 expression by parenchymal cells in GVHD target tissues has a critical role in tolerizing infiltrating donor T cells, thereby preventing the second wave of donor T cell expansion and preventing the persistence of GVHD. Our results suggest that therapies designed to preserve or restore B7-H1 expression by parenchymal tissues could prevent or ameliorate GVHD in humans.

Supplementary Material

01

Acknowledgments

This work was supported by the Nesvig Lymphoma Foundation and a private donation from DePasquale family. We thank Lucy Brown and her staff at the COH Flow Cytometry Facility, and Sofia Loera and her staff at the COH Anatomic Pathology Laboratory for their excellent technical assistance. We thank Aton Merck Pharmaceutical and the National Cancer Institute for kindly provided us with the vorinostat.

Footnotes

Contributions: X. Li designed and performed research as well as wrote the manuscript; R. Deng, W. He, C. Liu, J. Young, M. Wang, Z. Meng, and C. Du assisted in experiments; W. Huang provide advice on induction of tissue expression of B7-H1; L. Chen provided critical reagents and critically reviewed the manuscript; P. Martin provided advice on experimental design and critically reviewed the manuscript; S. Forman supported the project and reviewed the manuscript. D. Zeng designed the research, wrote the manuscript, and supervised the research progress.

Conflict of interest: The authors declare no conflict of interests.

References

1. Zhang C, Lou J, Li N, Todorov I, Lin CL, Cao YA, Contag CH, Kandeel F, Forman S, Zeng D. 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. 2007;178:838–850. [PubMed]
2. Zhang Y, Joe G, Hexner E, Zhu J, Emerson SG. Alloreactive memory T cells are responsible for the persistence of graft-versus-host disease. J Immunol. 2005;174:3051–3058. [PubMed]
3. Zhang Y, Joe G, Hexner E, Zhu J, Emerson SG. Host-reactive CD8+ memory stem cells in graft-versus-host disease. Nat Med. 2005;11:1299–1305. [PubMed]
4. Agata Y, Kawasaki A, Nishimura H, Ishida Y, Tsubata T, Yagita H, Honjo T. Expression of the PD-1 antigen on the surface of stimulated mouse T and B lymphocytes. International immunology. 1996;8:765–772. [PubMed]
5. Chen L. Co-inhibitory molecules of the B7-CD28 family in the control of T-cell immunity. Nat Rev Immunol. 2004;4:336–347. [PubMed]
6. Yamazaki T, Akiba H, Iwai H, Matsuda H, Aoki M, Tanno Y, Shin T, Tsuchiya H, Pardoll DM, Okumura K, Azuma M, Yagita H. Expression of programmed death 1 ligands by murine T cells and APC. J Immunol. 2002;169:5538–5545. [PubMed]
7. Ishida M, Iwai Y, Tanaka Y, Okazaki T, Freeman GJ, Minato N, Honjo T. Differential expression of PD-L1 and PD-L2, ligands for an inhibitory receptor PD-1, in the cells of lymphohematopoietic tissues. Immunology letters. 2002;84:57–62. [PubMed]
8. Schoop R, Wahl P, Le Hir M, Heemann U, Wang M, Wuthrich RP. Suppressed T-cell activation by IFN-gamma-induced expression of PD-L1 on renal tubular epithelial cells. Nephrol Dial Transplant. 2004;19:2713–2720. [PubMed]
9. Nakazawa A, Dotan I, Brimnes J, Allez M, Shao L, Tsushima F, Azuma M, Mayer L. The expression and function of costimulatory molecules B7H and B7-H1 on colonic epithelial cells. Gastroenterology. 2004;126:1347–1357. [PubMed]
10. Youngnak-Piboonratanakit P, Tsushima F, Otsuki N, Igarashi H, Machida U, Iwai H, Takahashi Y, Omura K, Yokozeki H, Azuma M. The expression of B7-H1 on keratinocytes in chronic inflammatory mucocutaneous disease and its regulatory role. Immunology letters. 2004;94:215–222. [PubMed]
11. Muhlbauer M, Fleck M, Schutz C, Weiss T, Froh M, Blank C, Scholmerich J, Hellerbrand C. PD-L1 is induced in hepatocytes by viral infection and by interferon- alpha and -gamma and mediates T cell apoptosis. Journal of hepatology. 2006;45:520–528. [PubMed]
12. Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands in tolerance and immunity. Annu Rev Immunol. 2008;26:677–704. [PubMed]
13. Blazar BR, Carreno BM, Panoskaltsis-Mortari A, Carter L, Iwai Y, Yagita H, Nishimura H, Taylor PA. Blockade of programmed death-1 engagement accelerates graft-versus-host disease lethality by an IFN-gamma-dependent mechanism. J Immunol. 2003;171:1272–1277. [PubMed]
14. Yi T, Chen Y, Wang L, Du G, Huang D, Zhao D, Johnston H, Young J, Todorov I, Umetsu DT, Chen L, Iwakura Y, Kandeel F, Forman S, Zeng D. Reciprocal differentiation and tissue-specific pathogenesis of Th1, Th2, and Th17 cells in graft-versus-host disease. Blood. 2009;114:3101–3112. [PubMed]
15. Asakura S, Hashimoto D, Takashima S, Sugiyama H, Maeda Y, Akashi K, Tanimoto M, Teshima T. Alloantigen expression on non-hematopoietic cells reduces graft-versus-leukemia effects in mice. J Clin Invest. 120:2370–2378. [PMC free article] [PubMed]
16. Flutter B, Edwards N, Fallah-Arani F, Henderson S, Chai JG, Sivakumaran S, Ghorashian S, Bennett CL, Freeman GJ, Sykes M, Chakraverty R. Nonhematopoietic antigen blocks memory programming of alloreactive CD8+ T cells and drives their eventual exhaustion in mouse models of bone marrow transplantation. J Clin Invest. 120:3855–3868. [PMC free article] [PubMed]
17. Koestner W, Hapke M, Herbst J, Klein C, Welte K, Fruehauf J, Flatley A, Vignali DA, Hardtke-Wolenski M, Jaeckel E, Blazar BR, Sauer MG. PD-L1 blockade effectively restores strong graft-versus-leukemia effects without graft-versus-host disease after delayed adoptive transfer of T-cell receptor gene-engineered allogeneic CD8+ T cells. Blood. 2011;117:1030–1041. [PubMed]
18. Dulude G, Roy DC, Perreault C. The effect of graft-versus-host disease on T cell production and homeostasis. J Exp Med. 1999;189:1329–1342. [PMC free article] [PubMed]
19. Cao YA, Wagers AJ, Beilhack A, Dusich J, Bachmann MH, Negrin RS, Weissman IL, Contag CH. Shifting foci of hematopoiesis during reconstitution from single stem cells. Proc Natl Acad Sci U S A. 2004;101:221–226. [PubMed]
20. Dong H, Zhu G, Tamada K, Flies DB, van Deursen JM, Chen L. B7-H1 determines accumulation and deletion of intrahepatic CD8(+) T lymphocytes. Immunity. 2004;20:327–336. [PubMed]
21. Dong H, Strome SE, Salomao DR, Tamura H, Hirano F, Flies DB, Roche PC, Lu J, Zhu G, Tamada K, Lennon VA, Celis E, Chen L. Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat Med. 2002;8:793–800. [PubMed]
22. Zhao D, Young JS, Chen YH, Shen E, Yi T, Todorov I, Chu PG, Forman SJ, Zeng D. Alloimmune response results in expansion of autoreactive donor CD4+ T cells in transplants that can mediate chronic graft-versus-host disease. J Immunol. 2011;186:856–868. [PMC free article] [PubMed]
23. Yi T, Chen Y, Wang L, Du G, Huang D, Zhao D, Johnston H, Young J, Todorov I, Umetsu D, Chen L, Iwakura Y, Kandee F, Forman S, Zeng D. Reciprocal differentiation and tissue-specific pathogenesis of Th1, Th2, and Th17 cells in graft versus host disease. Blood. 2009 In press. [PubMed]
24. Li N, Chen Y, He W, Yi T, Zhao D, Zhang C, Lin CL, Todorov I, Kandeel F, Forman S, Zeng D. Anti-CD3 preconditioning separates GVL from GVHD via modulating host dendritic cell and donor T-cell migration in recipients conditioned with TBI. Blood. 2009;113:953–962. [PubMed]
25. Liang Y, Huang T, Zhang C, Todorov I, Atkinson M, Kandeel F, Forman S, Zeng D. Donor CD8+ T cells facilitate induction of chimerism and tolerance without GVHD in autoimmune NOD mice conditioned with anti-CD3 mAb. Blood. 2005;105:2180–2188. [PubMed]
26. Li N, Zhao D, Kirschbaum M, Zhang C, Lin CL, Todorov I, Kandeel F, Forman S, Zeng D. HDAC inhibitor reduces cytokine storm and facilitates induction of chimerism that reverses lupus in anti-CD3 conditioning regimen. Proc Natl Acad Sci U S A. 2008;105:4796–4801. [PubMed]
27. Zhang C, Todorov I, Lin CL, Atkinson M, Kandeel F, Forman S, Zeng D. Elimination of insulitis and augmentation of islet beta cell regeneration via induction of chimerism in overtly diabetic NOD mice. Proc Natl Acad Sci U S A. 2007;104:2337–2342. [PubMed]
28. Zhao D, Zhang C, Yi T, Lin CL, Todorov I, Kandeel F, Forman S, Zeng D. In vivo-activated CD103+CD4+ regulatory T cells ameliorate ongoing chronic graft-versus-host disease. Blood. 2008;112:2129–2138. [PubMed]
29. Howard RB, Christensen AK, Gibbs FA, Pesch LA. The enzymatic preparation of isolated intact parenchymal cells from rat liver. J Cell Biol. 1967;35:675–684. [PMC free article] [PubMed]
30. Liu F, Song Y, Liu D. Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther. 1999;6:1258–1266. [PubMed]
31. Neurath MF, Finotto S. IL-6 signaling in autoimmunity, chronic inflammation and inflammation-associated cancer. Cytokine Growth Factor Rev. 2011;22:83–89. [PubMed]
32. Chen X, Das R, Komorowski R, Beres A, Hessner MJ, Mihara M, Drobyski WR. Blockade of interleukin-6 signaling augments regulatory T-cell reconstitution and attenuates the severity of graft-versus-host disease. Blood. 2009;114:891–900. [PubMed]
33. Diehl S, Anguita J, Hoffmeyer A, Zapton T, Ihle JN, Fikrig E, Rincon M. Inhibition of Th1 differentiation by IL-6 is mediated by SOCS1. Immunity. 2000;13:805–815. [PubMed]
34. Wakim LM, Waithman J, van Rooijen N, Heath WR, Carbone FR. Dendritic cell-induced memory T cell activation in nonlymphoid tissues. Science. 2008;319:198–202. [PubMed]
35. Ferrara JL, Reddy P. Pathophysiology of graft-versus-host disease. Semin Hematol. 2006;43:3–10. [PubMed]
36. Welniak LA, Blazar BR, Murphy WJ. Immunobiology of allogeneic hematopoietic stem cell transplantation. Annu Rev Immunol. 2007;25:139–170. [PubMed]
37. Shlomchik WD. Graft-versus-host disease. Nat Rev Immunol. 2007;7:340–352. [PubMed]