TRAIL-Transduced Dendritic Cells Protect Mice from Acute Graft-versus-Host Disease and Leukemia Relapse
http://www.100md.com
免疫学杂志 2005年第7期
Abstract
TRAIL preferentially induces apoptotic cell death in a wide variety of transformed cells, whereas it induces no apoptosis, but inhibits activation of Ag-specific T cells via blockade of cell cycle progression. Although accumulating results suggest that TRAIL is involved in the maintenance of immunological homeostasis under steady state conditions as well as in the initiation and progression of immunopathologies, the potential regulatory effect of TRAIL on immune responses and its therapeutic potential in immunological diseases remains unclear. We report in this study the potential usefulness of TRAIL-transduced dendritic cells (DCs) for the treatment of lethal acute graft-vs-host disease (GVHD) and leukemia relapse. DCs genetically modified to express TRAIL showed potent cytotoxicity against both alloreactive T cells and leukemic cells through the induction of apoptosis. In addition, treatment with genetically modified DCs expressing TRAIL of allogeneic BM transplants recipients with leukemia was effective for protection against acute GVHD and leukemia relapse. Thus, gene transfer of TRAIL to DCs is a novel modality for the treatment of acute GVHD and leukemia relapse by selective targeting of pathogenic T cells and leukemic cells.
Introduction
Tumor necrosis factor-related apoptosis-inducing ligand, also known as Apo2 ligand, is a type II transmembrane protein belonging to the TNF family (1). TRAIL can potentially interact with five different receptors. These include death receptor (DR43; TRAIL-R1), DR5 (TRAIL-R2), decoy receptor (DcR1; TRAIL-R3), DcR2 (TRAIL-R4), and a soluble receptor called osteoprotegerin (1). Receptors for TRAIL are constitutively expressed in a variety of cell types (1). In contrast, the constitutive expression of TRAIL was observed in liver NK cells, whereas the levels of TRAIL expression in T cells as well as NK cells can be markedly up-regulated after cell activation (2, 3, 4, 5). In addition, TRAIL preferentially induces apoptotic cell death in a wide variety of transformed cells, whereas it induces no apoptosis, but inhibits activation of Ag-specific T cells via blockade of cell cycle progression (6, 7).
The presence of multiple receptors for TRAIL strongly suggests that TRAIL is involved in the maintenance of immunological homeostasis under steady state conditions as well as in the initiation and progression of immunopathologies. Previous studies have shown that TRAIL plays a crucial role in the surveillance of tumor initiation and metastasis in mice (2). Although the role of TRAIL in the negative selection of thymocytes remains controversial (8, 9), TRAIL plays a crucial role in the regulation of autoimmune diseases (6, 8, 10). However, the potential regulatory effect of TRAIL on immune responses and its therapeutic potential in immunological diseases are unknown.
Dendritic cells (DCs) are APC that consist of heterogeneous subsets with different lineages and maturity; they not only initiate immunity, but are also involved in the induction of tolerance in vivo (11, 12, 13). Therefore, in addition to their original application in the therapy of cancer and infectious diseases, strategies using immunoregulatory DCs are expected to be effective for the prevention and treatment of autoimmune diseases, allergic diseases, and allograft rejection.
Genetic modification of DCs with genes encoding immunoregulatory molecules provides a potential approach for Ag-specific regulation of T cell-mediated immunity by selectively targeting Ag-specific T cells. The use of these genetically modified DCs was reportedly effective for the prevention of experimental autoimmune and allergic diseases as well as allograft rejection in animals through the down-regulation of Ag-specific T cell responses (14, 15, 16, 17).
Allogeneic bone marrow (BM) transplantation (BMT) is an effective treatment for hematologic malignancies as well as genetic disorders (18, 19, 20, 21). However, acute graft-vs-host disease (GVHD), which is caused by alloreactive T cells in donor BM inocula, is a major cause of morbidity and mortality in patients undergoing allogeneic BMT (18, 19, 20, 21). Although the incidence and severity of acute GVHD can be dramatically improved by T cell depletion or the combination of immunosuppressive agents, the risk of leukemia relapse may be increased in turn, possibly due to the lack of antileukemia effect of allogeneic T cells infused, so-called graft-vs-leukemia (GVL) effect (19, 20, 21). Therefore, there is an increasing interest in the development of strategies that suppress acute GVHD but enhance the GVL effect.
In this study we report that genetically modified TRAIL-expressing DCs induce apoptotic cell death in alloreactive T cells and ameliorate acute GVHD while exerting an antileukemic effect.
Materials and Methods
Media and reagents
The medium used throughout was RPMI 1640 (Sigma-Aldrich) or DMEM (Sigma-Aldrich) supplemented with antibiotic-antimycotic (Invitrogen Life Technologies) and 10% heat inactivated FCS (Invitrogen Life Technologies). GM-CSF, IL-2, IL-4, IFN-, and soluble TRAIL were purchased from PeproTech.
Cell preparations
Human immature DCs (iDCs) were obtained by culturing peripheral blood monocytes with GM-CSF (50 ng/ml) and IL-4 (50 ng/ml) for 7 days (22). For the preparation of mature DCs (mDCs), cells were subsequently cultured with LPS (1 μg/ml; Sigma-Aldrich) for another 4 days (22). Murine iDCs were prepared by culturing BM cells obtained from female BALB/c mice (H-2d) or C57BL/6 mice (H-2b; all from Charles River Laboratories) with murine GM-CSF (20 ng/ml) for 8 days, and mDCs were obtained from culture of iDCs with LPS (1 μg/ml) for 4 days (21). Human T cells were purified from PBMC with a T cell negative isolation kit (Dynal Biotech), and CD4+ T cells were then negatively selected from T cells with anti-CD8 mAb (BD Biosciences) plus goat anti-mouse IgG Ab-conjugated immunomagnetic beads (Dynal Biotech) (22). Murine T cells were negatively selected from splenic mononuclear cells (MNC) obtained from C57BL/6 mice with mAbs to Ly-76, B220, Ly-6G, and I-A/I-E (all from BD Biosciences) plus sheep anti-rat IgG Ab-conjugated immunomagnetic beads (Dynal Biotech) (21). Subsequently, CD4+ T cells were negatively selected from T cells with anti-CD8 mAb (BD Biosciences) in combination with sheep anti-rat IgG Ab-conjugated immunomagnetic beads (21). Con A blasts were obtained from the culture of human or murine T cells with Con A (2.5 μg/ml; Sigma-Aldrich) for 3 days.
Production of adenovirus encoding the TRAIL genes
The full-length human TRAIL (hTRAIL) cDNA (884 bp) was prepared by RT-PCR amplification of total RNA from Con A blasts with the following oligonucleotide primers: 5'-CAG CAG TCA GAC TCT GAC AG-3' and 5'-TCT TTC CAG GTC AGT TAG CC-3'. The PCR product was subcloned into pCR2.1 vector using TA Cloning kit (Invitrogen Life Technologies), and the nucleotide sequence was confirmed using a 373A automated sequencer (Applied Biosystems) and the fluoresceinated dye terminator cycle sequencing method. The full-length murine TRAIL (mTRAIL) cDNA was prepared from mTRAIL/pMKITNeo expression vector (23). After XhoI and NotI digestion, the 850 bp of mTRAIL cDNA was obtained, and the nucleotide sequence was confirmed as described above.
A replication-deficient adenovirus vector (Ad) expressed from the CAG promotor was generated using an Adenovirus Expression Vector kit (TaKaRa Shuzo), in which an adenoviral cosmid, pAxCAwt, was included. The cosmid pAxCAwt consisted of E1- and E3-deficient Ad type 5 (Ad5) sequences, and the CAG promoter and rabbit globin poly-A were inserted at the former E1 site in reverse orientation with respect to the Ad5 sequences. Briefly, the entire coding sequence of hTRAIL or mTRAIL was blunted using a DNA Blunting kit (TaKaRa Shuzo) and was then subcloned into the SwaI site of pAxCAwt. The resulting cosmids were named hTRAIL/pAxCAwt and mTRAIL/pAxCAwt, respectively. Transfection of human embryonic kidney 293 cells (RIKEN Cell Bank) with these cosmid vectors and Ad backbone sequences (DNA-TPC) that had the E1 and E3 genes deleted was performed according to the manufacturer’s instructions to produce replication-incompetent, E1- and E3-deficient, Ad-expressing hTRAIL or mTRAIL (hTRAIL-Ad or mTRAIL-Ad). The viruses were then prepared by expansion of a single clone generated in 293 cells, which were purified by limiting dilution, and viral particles were isolated and amplified for analysis of hTRAIL or mTRAIL expression by flow cytometry. Recombinant adenoviruses generated from the homologous recombination of pAxCAwt and DNA-TPC were used as virus controls (control-Ad). Recombinant adenovirus titers were determined by plaque assays on 293 cells. These adenoviruses were suspended in culture medium, adjusted to 2 x 108 PFU/ml, and stored at –80°C until use.
Adenoviral infection
For Ad-mediated gene transfer into human DCs by centrifugal transduction (24), 500 μl of cells (106 cells) were mixed with 500 μl of adenoviral vector (multiplicity of infection (MOI) of 10 or 50), and 1 ml of the mixture was poured into a polypropylene tube (BD Biosciences). The tubes were centrifuged at 2000 x g at 37°C for 2 h. After the centrifugal transduction, the cells were washed twice in PBS. DCs were resuspended in culture medium under various culture conditions and cultured for the indicated periods in tissue culture dishes (BD Biosciences). For adenoviral infection of murine DCs, 500 μl of iDCs (106 cells) were mixed with 500 μl of adenoviral vector (MOI of 50) in a polypropylene tube, and 1 ml of the mixture was incubated at 37°C. After a 2-h incubation, culture medium was added to the cells, then the cultures were incubated with LPS (1 μg/ml) for 4 days in tissue culture dishes.
Flow cytometry
Cells were stained with the following mAbs to human and murine markers: CD3, CD4, CD11c, CD40, CD80, CD86, HLA-A/B/C, HLA-DR, H-2Kb, H-2Kd, I-A/I-E, and isotype-matched control IgG (all from BD Biosciences); CD83 (Coulter Immunology); and FITC-conjugated goat anti-rat IgG Ab (Santa Cruz Biotechnology). The purified mAbs to hTRAIL (RIK-2), mTRAIL (N2B2), human DR4 (hDR4; DJR1), human DR5 (hDR5; DJR2), human DcR1 (hDcR1; DJR3), human DcR2 (hDcR2; DJR4), and murine DR5 (mDR5; MD5–1) were prepared as described previously (5, 25, 26). Fluorescent staining was analyzed on a FACScan flow cytometer using CellQuest software (BD Biosciences), and the data are expressed as the mean fluorescence intensity (MFI).
Cytotoxicity assay
Adenoviral gene-transduced untransduced DCs were cultured with Na251CrO4 (NEN Life Science Products)-labeled Jurkat cells, L929 cells, P815 cells (104; all from RIKEN Cell Bank), or Con A blasts (104) for 4 h at various E:T cell ratios in the presence or the absence of 10 μg/ml anti-hTRAIL mAb, anti-hDR5 mAb, anti-mTRAIL mAb, or control IgG. As a control, soluble hTRAIL was added to the target cells at the indicated concentrations (10–1000 ng/ml). The radioactivity of the supernatants was measured, and the percent-specific lysis was calculated (21, 22). Spontaneous release was <10% of total release.
Cell proliferation assay
Human CD4+ T cells (105) were stimulated with or without plate-bound anti-human CD3 mAb (BD Biosciences) plus soluble anti-human CD28 mAb (BD Biosciences) in the presence or the absence of soluble hTRAIL (1000 ng/ml), anti-hTRAIL mAb, or control Ig (each 10 μg/ml). For human and murine allogeneic MLR, CD4+ T cells (105) were cultured in 96-well plates (BD Biosciences) with various numbers of irradiated (15 Gy from a 137Cs source; MBR-1505R2; Hitachi Medical) allogeneic DCs in the presence or the absence of soluble hTRAIL (1000 ng/ml), anti-hTRAIL mAb (10 μg/ml), anti-mTRAIL mAb (10 μg/ml), anti-hDR5 mAb (10 μg/ml), or control Ig (10 μg/ml). [3H]Thymidine incorporation was measured on day 5 for the last 18 h (21, 22).
In vitro analysis of T cell responses
For measuring cell division, CD4+ T cells were labeled with CFSE (Molecular Probes) according to the manufacturer’s instructions. CFSE-labeled CD4+ T cells (5 x 106) were cultured with irradiated (15 Gy) allogeneic DCs (5 x 105) for 3 days, then T cells were negatively selected with anti-human CD11c mAb (BD Biosciences) plus goat anti-mouse IgG Ab-conjugated immunomagnetic beads (22). These T cell preparations contained <0.1% CD11c+ cells as assessed by FACS analysis. The CFSE-positive cells were then analyzed by flow cytometry. Apoptosis in allogeneic DC-stimulated CD4+ T cells was measured by flow cytometry using an Annexin VFITC apoptosis detection kit (R&D Systems). For cell cycle analysis, the stimulated CD4+ T cells were incubated with BrdU (BD Biosciences) at 10 μM for 1 h at 37°C. Staining of incorporated BrdU was performed using a BrdU Flow kit (BD Bioscience) according to the manufacturer’s instructions. Cells were stained with FITC-conjugated anti-BrdU Ab for 20 min at room temperature. 7-Amino-actinomycin D (20 μg/ml) was added to the cell suspension before flow cytometric analysis.
Acute GVHD model
BALB/c recipient mice (five animals in each group) received lethal total body irradiation (TBI; 10 Gy) and then a single i.v. injection of C57BL/6 BM cells (1.5 x 107/mouse) plus splenic MNC (1.5 x 107/mouse) through the tail vein (21). The day of transplantation was designated day 0. The i.v. injection of adenoviral gene-transduced or untransduced DCs from BALB/c mice or C57BL/6 mice (106 to 5 x 106/mouse) 2 days after transplantation. For the in vivo blockade experiments, recipients were i.p. injected with control Ig or anti-mTRAIL mAb (1 mg/mouse) before i.v. injection of DCs. The recipients were monitored every day for survival. Some recipients were killed 5 days after transplantation to obtain serum and splenic MNCs.
Leukemia relapse model
BALB/c recipients (five animals in each group) were inoculated i.v. with P815 cells (2 x 105/mouse) 2 days before TBI (10 Gy) and i.v. transplantation with C57BL/6 BM cells (1.5 x 107/mouse) (21). The transplanted recipients received a single i.v. injection of adenoviral gene-transduced or untransduced DCs from BALB/c mice (5 x 106/mouse) 2 days after transplantation. Recipients were monitored every day for survival. Hepatosplenomegaly due to tumor burden in the dead mice was confirmed by laboratory.
Statistical analyses
Statistically significant differences were determined by Student’s paired t test or Mann-Whitney’s U test. A value of p < 0.01 was considered significant.
Results
Regulatory function of human DCs genetically modified to express TRAIL
To test the potential use of TRAIL-expressing DCs for selectively targeting Ag-specific T cells, we examined the conditions for the generation of human DCs genetically engineered to express TRAIL using Ad. Although stimulation of iDCs with IFN- or LPS induced low levels of TRAIL expression (27, 28), adenoviral gene transduction of hTRAIL (hTRAIL-Ad) into DCs resulted in higher expression of TRAIL. Of note, introduction of hTRAIL-Ad into iDCs followed by stimulation with LPS (Fig. 1, A and B) or TNF- (data not shown) resulted in the generation of mDCs with the highest level of TRAIL expression. Adenoviral infection had little or no effect on the expression of MHC and costimulatory molecules (Fig. 1C). The hTRAIL-Ad-infected mDCs (hTRAIL-Ad/mDCs) showed a more potent killing activity against hTRAIL-sensitive Jurkat cells than soluble hTRAIL, mDCs, and control-Ad-infected mDCs (control-Ad/mDCs; Fig. 1D). In addition, the cytotoxicity of hTRAIL-Ad/mDCs against Jurkat cells was blocked by anti-hTRAIL mAb and anti-hDR5 mAb (Fig. 1, D and E). These results indicate that hTRAIL was functionally expressed on the hTRAIL-Ad/mDCs. In contrast, Con A blasts (Fig. 1C) and DCs (data not shown) were relatively resistant to TRAIL-mediated cytotoxicity.
FIGURE 1. Generation of genetically modified human DCs expressing TRAIL. A and B, The iDCs were not infected or were infected with control-Ad or hTRAIL-Ad at MOI of 10 (A) and 50 (B). In some experiments, uninfected DCs or DCs infected with hTRAIL-Ad were stimulated with IFN- (A) or LPS (A and B) before or after infection. The expression of hTRAIL on DCs was analyzed by flow cytometry at the indicated time points. The expression level of TRAIL was expressed as the mean MFI ± SD of four individual experiments. The results are representative of two experiments with similar results. C, The iDCs were not infected or were infected with control-Ad or hTRAIL-Ad at an MOI of 50, followed by stimulation with LPS for generation of mDCs. Subsequently, mDCs were stained with the indicated mAbs, and cell surface expression was analyzed by flow cytometry. Data are represented by a dot plot for the expression of MHC and costimulatory molecules or by a histogram in which cells were stained with anti-hTRAIL mAb (thick lines) or isotype-matched control Ig (thin lines). The results are representative of four experiments with similar results. D, The cytotoxicity of soluble hTRAIL, uninfected DCs, or DCs infected with control-Ad or hTRAIL-Ad against Jurkat cells or Con A blasts at various E:T cell ratios in the presence or the absence of control Ig or anti-hTRAIL mAb was analyzed by the 4-h 51Cr release assay. Data were expressed as the mean ± SD of triplicate samples, and the results are representative of four experiments with similar results. E, The cytotoxicity of soluble hTRAIL, uninfected DCs, or DCs infected with control-Ad or hTRAIL-Ad against Jurkat cells at an E:T cell ratio of 10 in the presence or the absence of control Ig, anti-hTRAIL mAb, or anti-hDR5 mAb was analyzed by the 4-h 51Cr release assay. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with control Ig, by Student’s paired t test.
We also examined the expression levels of TRAIL receptors on various cell types (Fig. 2A). The iDCs constitutively expressed DR4, DR5, DcR1, and DcR2 at similar levels, and the expression of these receptors was slightly reduced after maturation. Unlike Jurkat cells, which predominantly expressed DR5, little or no expression of these receptors was observed on unstimulated CD4+ T cells and Con A blasts. Interestingly, stimulation of CD4+ T cells with allogeneic mDCs induced specific up-regulation of DR5, whereas stimulation with anti-CD3 and anti-CD28 mAbs up-regulated DR5 to a lesser degree (data not shown).
FIGURE 2. Regulatory function of human DCs genetically modified to express TRAIL. A, The expression of TRAIL receptors on the indicated cells was analyzed by flow cytometry. Values were expressed as the ratio of MFI with respective mAb compared with the MFI with control Ig, and data are expressed as the mean ± SD of four individual experiments. *, p < 0.01 compared with control Ig, by Student’s paired t test. B, Human CD4+ T cells (105) were stimulated with or without plate-bound anti-human CD3 plus soluble anti-human CD28 mAb in the presence or the absence of soluble hTRAIL and anti-hTRAIL mAb or control Ig. In another experiment, CD4+ T cells (105) were cultured with uninfected or control-Ad or hTRAIL-Ad-infected allogeneic mDCs (104) in the presence or the absence of soluble hTRAIL, anti-hTRAIL mAb, anti-hDR5 mAb, or control Ig. The proliferative response was measured by [3H]thymidine uptake on day 5. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01, by Student’s paired t test. C, CD4+ T cells (105) were cultured with various numbers of uninfected or control-Ad- or hTRAIL-Ad-infected allogeneic mDCs, and the proliferative response was measured by [3H]thymidine uptake on day 5. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with uninfected DCs, by Student’s paired t test. D, CFSE-labeled CD4+ T cells (5 x 106) were primed with uninfected or control-Ad- or hTRAIL-Ad-infected allogeneic mDCs (5 x 105) for 3 days, and CFSE levels were analyzed by flow cytometry. Data are represented by a histogram. The data shown are representative of four experiments with similar results. E and F, CD4+ T cells (5 x 106) were primed with uninfected or control-Ad- or hTRAIL-Ad-infected allogeneic mDCs (5 x 105) for 3 days, then CD4+ T cells were analyzed for apoptosis (E) or cell cycle (F) by flow cytometry. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with T cells primed with mDCs (E) or unprimed T cells (F), by Student’s paired t test.
We also examined the T cell regulatory function of hTRAIL-Ad/DCs. Soluble TRAIL showed a minimal inhibition of proliferation when CD4+ T cells were stimulated with anti-CD3 and anti-CD28 mAbs or allogeneic mDCs (Fig. 2B). In contrast, hTRAIL-Ad/DCs, but not control-Ad/mDCs, displayed a potent suppressive effect on the proliferation of alloreactive CD4+ T cells (Fig. 2, B and C). This suppression was abrogated by both anti-hTRAIL mAb and anti-hDR5 mAb, but not by control Ig (Fig. 2B). These results indicate that DCs genetically modified to express TRAIL could inhibit the proliferation of allogeneic CD4+ T cells through the interaction with DR5.
Previous studies have shown that soluble TRAIL did not induce apoptosis, but inhibited the proliferation of T cells through blockage of cell cycle progression (6, 7). To clarify the mechanism underlying the T cell regulatory function of TRAIL-transduced DCs, we characterized CD4+ T cells primed with hTRAIL-Ad/DCs. The CFSE labeling assay showed that the proportion of dividing cells was significantly reduced in allogeneic CD4+ T cells primed with hTRAIL-Ad/DCs compared with those primed with mDCs and control-Ad/mDCs (Fig. 2D). In contrast, hTRAIL-Ad/DCs induced significantly more apoptosis in allogeneic CD4+ T cells than mDCs and control-Ad/mDCs (Fig. 2E). Also in the cell cycle analysis, numerous apoptotic cells with sub-G0/G1 DNA content were detected in allogeneic CD4+ T cells primed with hTRAIL-Ad/DCs, whereas the proliferating cells in S phase and G2+M phase were increased in allogeneic CD4+ T cells primed with mDCs or control-Ad/mDCs (Fig. 2F). These results indicate that the DCs genetically modified to express TRAIL suppress the proliferation and cell division of allogeneic CD4+ T cells through the induction of apoptosis rather than cell cycle arrest.
TRAIL-transduced DCs ameliorate murine acute GVHD
The above results indicated that mDCs genetically modified to express hTRAIL could efficiently suppress the proliferation of alloreactive CD4+ T cells through the induction of apoptosis. We therefore tested the in vivo suppressive function of murine DCs genetically modified to express mouse TRAIL (mTRAIL-Ad/DCs). Similar to hTRAIL-Ad/mDCs, mTRAIL-Ad/DCs showed the functional expression of mTRAIL (Fig. 3, A and B). In addition, mTRAIL-Ad/DCs impaired the proliferation of alloreactive CD4+ T cells (Fig. 3C), and they induced a higher rate of apoptosis in allogeneic CD4+ T cells than control-Ad/DCs (data not shown).
FIGURE 3. Regulatory function of murine DCs genetically modified to express TRAIL. A, Murine DCs were uninfected or infected with control-Ad or mTRAIL-Ad at an MOI of 50, followed by stimulation with LPS. Subsequently, DCs were stained with the indicated mAbs, and cell surface expression was analyzed by flow cytometry. Data are represented by a dot plot for the expression of MHC and costimulatory molecules or by a histogram in which cells were stained with anti-hTRAIL mAb (thick lines) or isotype-matched control Ig (thin lines). The data shown are representative of four experiments with similar results. B, The cytotoxicity of uninfected DCs or DCs infected with control-Ad or mTRAIL-Ad against L929 cells or Con A blasts was analyzed in the presence or the absence of control Ig or anti-mTRAIL mAb. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. C, CD4+ T cells (105) from naive C57BL/6 mice were cultured with various numbers of uninfected or control-Ad- or mTRAIL-Ad-infected mDCs from BALB/c mice, and the proliferative response was measured on day 5 by [3H]thymidine uptake. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with uninfected DCs, by Student’s paired t test. D, The expression of DR5 on the indicated murine cells was analyzed by flow cytometry. Values were expressed as the ratio of MFI with anti-DR5 mAb compared with the MFI with control Ig, and data are expressed as the mean ± SD of four individual experiments. *, p < 0.01 compared with control Ig, by Student’s paired t test.
We also examined the expression level of mDR5 on various cell types (Fig. 3D). Flow cytometric analysis showed that mTRAIL-sensitive L929 cells as well as P815 cells expressed high level of mDR5. In contrast, mDCs exhibited slightly lower expression of mDR5 than mDCs. We also observed that DCs expressed the transcripts of all TRAIL receptors at similar levels, and these transcriptional expressions were reduced after maturation (data not shown). In addition, the primed CD4+ T cells with allogeneic mDCs showed higher expression of mDR5 than unstimulated CD4+ T cells and Con A blasts.
We then tested the therapeutic efficacy of mTRAIL-Ad/DCs against acute GVHD. All BALB/c recipients died on day 8 after transplantation of C57BL/6 BM and spleen cells (Fig. 4A). In these mice, clinical symptoms of acute GVHD, such as hair ruffling, lower mobility, and weight loss, became apparent within 6 days. In contrast, all BALB/c recipients of syngeneic BM and spleen cells survived >60 days without apparent acute GVHD (data not shown). A single injection of BALB/c-derived DCs or control-Ad/DCs 2 days after transplantation of C57BL6 BM and spleen cells to BALB/c recipients did not significantly affect the lethality caused by acute GVHD (Fig. 4A). In contrast, BALB/c-derived mTRAIL-Ad/DCs ameliorated acute GVHD in a dose-dependent fashion (Fig. 4A), whereas C57BL/6-derived mTRAIL-Ad/DCs showed a minimal inhibitory effect (Fig. 4B). In addition, in vivo blockade of mTRAIL with anti-mTRAIL mAb abrogated the therapeutic effect of mTRAIL-Ad/DCs (Fig. 4C).
FIGURE 4. Suppressive effect of TRAIL-transduced murine DCs on acute GVHD in murine allogeneic BMT. A, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without uninfected or control-Ad- or mTRAIL-Ad-infected BALB/c DCs (106 to 5 x 106/mouse) on 2 days after BMT. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with untreated mice, by Mann-Whitney’s U test. B, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without uninfected or mTRAIL-Ad-infected BALB/c or C57BL/6 DCs (5 x 106/mouse) on 2 days after BMT. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with untreated mice, by Mann-Whitney’s U test. C, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without mTRAIL-Ad-infected BALB/c DCs (5 x 106/mouse) and control Ig or anti-mTRAIL mAb on 2 days after BMT. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with control Ig-treated mice, by Mann-Whitney’s U test.
We also examined the in vivo regulatory effect of mTRAIL-Ad/DCs in the recipients of allogeneic transplantation. The administration of mTRAIL-Ad/DCs significantly reduced the number of total splenocytes (Fig. 5A) and inhibited the expansion of donor-derived CD4+ and CD8+ T cells in the spleen (Fig. 5B), whereas that of untransduced DCs increased the number of these cells. Moreover, donor-derived CD4+ T cells from the mTRAIL-Ad/DC-treated recipients showed a significantly reduced proliferative response to the host-type mDCs, whereas those from the DC-treated recipients showed an enhanced response (Fig. 5C). Furthermore, donor-derived CD8+ T cells from the mTRAIL-Ad/DC-treated recipients showed markedly impaired CTL activity against P815 cells expressing the host-type alloantigen, whereas those from the DC-treated recipients showed enhanced CTL activity (Fig. 5D). In addition, mTRAIL-Ad/DC-treated recipients showed a significant reduction of production of serum IFN-, TNF-, and IL-12p40 compared with untreated and control-Ad/DC-treated recipients (Fig. 5E). These results indicate that mTRAIL-Ad/DCs is could efficiently ameliorate acute GVHD through the suppression of alloreactive T cell responses.
FIGURE 5. Suppressive effect of TRAIL-transduced murine DCs on allogeneic responses of donor-derived T cells in transplanted mice. A–E, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without uninfected or mTRAIL-Ad-infected BALB/c DCs (5 x 106/mouse) on 2 days after BMT. On 5 days after BMT, splenic MNC and sera were obtained from indicated mice. A, The number of donor-derived (H-2b+) splenic MNC was analyzed by flow cytometry. Data were expressed as the mean ± SD of five mice in each group. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. B, The constitution of donor-derived (H-2b+) T cell subsets was analyzed by flow cytometry. Data are expressed as the mean ± SD of five mice in each group. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. C, Donor-derived CD4+ T cells (105) were cultured with various numbers of irradiated BALB/c DCs, and the proliferative response was measured by [3H]thymidine uptake on day 5. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. D, Donor-derived CD8+ T cells were subjected to CTL assay against P815 cells at various E:T cell ratios. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. E, Concentrations of IFN-, TNF-, and IL-12p40 in serum were evaluated by ELISA. Data are expressed as the mean ± SD of five mice in each group. *, p < 0.01 compared with untreated recipients, by Student’s paired t test.
Protection from leukemia relapse by mTRAIL-transduced DCs
TRAIL has been implicated in the GVL effect associated with allogeneic BMT (20). Therefore, we next examined the antileukemic effect of TRAIL-transduced DCs. The recipient mice were i.v. inoculated with or without P815 leukemia cells 2 days before TBI and transplantation of allogeneic BM cells. Consistent with previous reports (20, 21), the mice transplanted with allogeneic BM cells alone survived >60 days without apparent acute GVHD. All leukemia-bearing mice that received TBI alone died within 14 days with marked hepatosplenomegaly. In contrast, the leukemia-bearing recipients of allogeneic BM cells died within 26 days after transplantation, indicating that alloreactive T cells in BM cells exhibited an insufficient GVL effect. We then examined the therapeutic effect of mTRAIL-transduced DCs against leukemia relapse in this model. The mTRAIL-Ad/DCs showed potent cytotoxicity against P815 cells in vitro (Fig. 6A). A single injection of recipient-type mTRAIL-Ad/DCs, but not DCs or control-Ad/DCs, 2 days after transplantation markedly prolonged the survival of leukemia-bearing mice (Fig. 6B). These results suggested that TRAIL-transduced DCs are useful not only to ameliorate acute GVHD but also to suppress leukemia relapse.
FIGURE 6. Protective effect of TRAIL-transduced murine DCs against leukemia relapse. A, The cytotoxicity of uninfected or control-Ad- or mTRAIL-Ad-infected DCs against P815 cells was analyzed by the 4-h 51Cr release assay. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with uninfected DCs, by Student’s paired t test. B, BALB/c recipient mice were inoculated i.v. with P815 cells (2 x 105) on day –2 and received TBI and i.v. transplantation with C57BL/6 BM cells on day 0. The recipient mice were i.v. injected with or without uninfected or control-Ad- or mTRAIL-Ad-infected BALB/c DCs (5 x 106/mouse) on day 2. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with untreated mice, by Mann-Whitney’s U test.
Discussion
Gene transfer of immunoregulatory molecules to Ag-presenting DCs provides an attractive approach for the treatment of immunological diseases. Our present findings demonstrate the efficacy of TRAIL gene-transduced DCs for the therapy of lethal acute GVHD and leukemia relapse via elimination of the pathogenic T cells and leukemia cells.
Efficient expression of the introduced gene in target cells is crucial for regulating their function. We showed that the gene transfer of TRAIL into iDCs followed by inflammatory stimulation caused maximal expression of TRAIL. The precise molecular mechanism by which inflammatory stimulation up-regulated the expression of transduced TRAIL gene in DCs remains unknown, but inflammatory stimulation might activate certain transcriptional factors, leading to transactivation of the transduced TRAIL gene. In contrast, mDCs show more potent interaction with Ag-specific T cells than iDCs due to their higher expression of MHC and adhesion/costimulatory molecules. Thus, the maturation of iDCs genetically engineered to express TRAIL by inflammatory stimulation provides advantages for TRAIL expression as well as Ag-specific interaction with T cells.
We showed that human and murine TRAIL-sensitive cell types, including CD4+ T cells primed with allogeneic DCs as well as Jurkat cells and L929 cells, exhibited high expression of DR5, indicating that DR5 is crucial for TRAIL-mediated cytotoxicity. In contrast, analysis of the cell surface expression of TRAIL receptors as well as their transcriptional expression (data not shown) revealed that human and murine DCs showed similar expression levels of these TRAIL receptors. Therefore, suppression of binding of TRAIL to DRs by DcRs may contribute to the low sensitivity of these cell types to TRAIL-mediated cytotoxicity. Collectively, our results suggest that the balance of the expression of DRs and DcRs may determine the sensitivity to TRAIL-mediated cytotoxicity.
The weak suppressive effect of recombinant soluble TRAIL on alloreactive T cell proliferation might be due to a blockage of cell cycle progression as previously reported (6, 7), whereas mDCs genetically modified to express TRAIL could effectively induce apoptosis in alloreactive T cells. Although the precise molecular mechanism underling the difference in the biological function between soluble TRAIL and membrane-bound transduced TRAIL on DCs remains unknown, this phenomenon might reflect the strength of intracellular signaling via DRs. Additional study will be needed to test this possibility.
Although host-matched or -mismatched mTRAIL-transduced DCs may lyse donor-derived T cells that overexpresses mDRs regardless of histocompatibility differences, host-matched mTRAIL-transduced DCs could more effectively lyse host-reactive, donor-derived T cells than host-mismatched, mTRAIL-transduced DCs, because they show more potent reactivity to host-matched, mTRAIL-transduced DCs than host-mismatched, mTRAIL-transduced DCs. Indeed, BALB/c-derived, mTRAIL-Ad/DCs, but not C57BL/6-derived, mTRAIL-Ad/DCs, ameliorated acute GVHD in BALB/c recipients transplanted with C57BL6 BM and spleen cells. Therefore, host-type, TRAIL-transduced DCs are useful for the protection of recipients from lethality induced by acute GVHD in allogeneic BMT.
The expansion of donor-derived CD4+ and CD8+ T cells was markedly suppressed by a single injection of host-type, mTRAIL-transduced DCs. These results suggest that interaction of alloreactive CD4+ and CD8+ T cells with the infused TRAIL-transduced mDCs caused their apoptosis via up-regulation of mDR5, resulting in their selective elimination in vivo. In contrast, donor-derived CD4+ T cells obtained from these protected recipients showed a reduced proliferative response to host-type mDCs. Our results imply that the intracellular signaling events involving TRAIL/DR5 induce not only apoptosis but also functional deficiency in the targeted CD4+ T cells. We also showed that the CTL activity of donor-derived CD8+ T cells was greatly impaired in these protected recipients. Although the precise mechanism remains unclear, the functional impairment of donor-derived CD4+ T cells may contribute to the reduced response of donor-derived CD8+ T cells.
The deletion of donor-derived alloreactive T cells by mTRAIL-transduced DCs can ameliorate acute GVHD, but also may impair their GVL effect. However, our results showed that the TRAIL-transduced DCs could protect BMT recipients from leukemia relapse. This seems likely to be mediated by antileukemic effect of TRAIL expressed on DCs, because the TRAIL-transduced DCs exhibited potent cytotoxic activity in vitro. The ex vivo removal of donor T cell proportions from the marrow graft not only reduced the incidence and severity of acute GVHD, but also increased the risk of leukemia relapse due to lack of GVL effect (19, 20, 21). Therefore, our results imply that the use of TRAIL-transduced DCs and T cell-depleted BM inocula might be an alternative strategy for the prevention of acute GVHD and leukemia relapse in human leukemia patients undergoing allogeneic BMT.
The role of TRAIL in hepatic cell death is controversial (29, 30). We observed that a single injection of mTRAIL-Ad/DCs did not induce any apparent liver cytotoxicity in normal mice or recipients of allogeneic BMT (data not shown). Additional study will be needed to determine the safety and potential side effects of TRAIL-transduced DCs in vivo.
In the present study we demonstrated a potent effect of TRAIL-transduced DCs to ameliorate acute GVHD. A similar strategy may be also useful for Ag-specific immunosuppression in allogeneic organ transplantation, autoimmune diseases, and allergic diseases. Studies are now underway to address this possibility.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank K. Sato, Y. Sato, A. Takeuchi, and M. Yamamoto for excellent assistance.
Footnotes
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 Japan Rheumatism Foundation (to K.S.), the Uehara Memorial Foundation (to K.S.), and the Kanzawa Medical Research Foundation (to K.S.).
2 Address correspondence and reprint requests to Dr. Katsuaki Sato, Laboratory for Dendritic Cell Immunobiology, Research Center for Allergy and Immunology, RIKEN Yokohama Institute, Suehiro-cho 1-7-22, Tsurumi, Yokohama, Kanagawa 230-0045 Japan. E-mail address: katsuaki@rcai.riken.jp
3 Abbreviations used in this paper: DR, death receptor; Ad, adenovirus vector; BM, bone marrow; BMT, BM transplantation; DC, dendritic cell; DcR, decoy receptor; GVHD, acute graft-vs-host diseases; GVL, graft-vs-leukemia; hTRAIL, human TRAIL; iDC, immature DC; mDC, mature DC; MFI, mean fluorescence intensity; MNC, mononuclear cell; MOI, multiplicity of infection; mTRAIL, murine TRAIL; TBI, total body irradiation; TRAIL-Ad, Ad expressing TRAIL.
Received for publication October 19, 2004. Accepted for publication January 16, 2005.
References
Lahm, A., A. Paradisi, D. R. Green, G. Melino. 2003. Death fold domain interaction in apoptosis. Cell Death Differ. 10:10
Smyth, M. J., K. Takeda, Y. Hayakawa, J. J. Peschon, M. R. van den Brink, H. Yagita. 2003. Nature’s TRAIL: on a path to cancer immunotherapy. Immunity 18:1.
Smyth, M. J., E. Cretney, K. Takeda, R. H. Wiltrout, L. M. Sedger, N. Kayagaki, H. Yagita, K. Okumura. 2001. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) contributes to interferon -dependent natural killer cell protection from tumor metastasis. J. Exp. Med. 193:661.
Takeda, K., Y. Hayakawa, M. J. Smyth, N. Kayagaki, N. Yamaguchi, S. Kakuta, Y. Iwakura, H. Yagita, K. Okumura. 2001. Involvement of tumor necrosis factor-related apoptosis-inducing ligand in surveillance of tumor metastasis by liver natural killer cells. Nat. Med. 7:94.
Kayagaki, N., N. Yamaguchi, M. Nakayama, A. Kawasaki, H. Akiba, K. Okumura, H. Yagita. 1999. Involvement of TNF-related apoptosis-inducing ligand in human CD4+ T cell-mediated cytotoxicity. J. Immunol. 162:2639.
Song, K., Y. Chen, R. Goke, A. Wilmen, C. Seidel, A. Goke, B. Hilliard, Y. Chen. 2000. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is an inhibitor of autoimmune inflammation and cell cycle progression. J. Exp. Med. 191:1095.
Lunemann, J. D., S. Waiczies, S. Ehrlich, U. Wendling, B. Seeger, T. Kamradt, F. Zipp. 2002. Death ligand TRAIL induces no apoptosis but inhibits activation of human (auto)antigen-specific T cells. J. Immunol. 168:4881.
Lamhamedi-Cherradi, S. E., S. J. Zheng, K. A. Maguschak, J. Peschon, Y. H. Chen. 2003. Defective thymocyte apoptosis and accelerated autoimmune diseases in TRAIL–/– mice. Nat. Immunol. 4:255
Cretney, E., A. P. Uldrich, S. P. Berzins, A. Strasser, D. I. Godfrey, M. J. Smyth. 2003. Normal thymocyte negative selection in TRAIL-deficient mice. J. Exp. Med. 198:491.
Hilliard, B., A. Wilmen, C. Seidel, T. S. Liu, R. Goke, Y. Chen. 2001. Roles of TNF-related apoptosis-inducing ligand in experimental autoimmune encephalomyelitis. J. Immunol. 166:1314.
Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendra, K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767.
Steinman, R. M., M. C. Nussenzweig. 2002. Avoiding horror autotoxicus: the importance of dendritic cells in peripheral T cell tolerance. Proc. Natl. Acad. Sci. USA 99:351.
Morelli, A. E., A. W. Thomson. 2003. Dendritic cells: regulators of alloimmunity and opportunities for tolerance induction. Immunol. Rev. 196:125.
Matsue, H., K. Matsue, M. Walters, K. Okumura, H. Yagita, A. Takashima. 1999. Induction of antigen-specific immunosuppression by CD95L cDNA-transfected ‘killer’ dendritic cells. Nat. Med. 5:930.
Lu, L., W. C. Lee, T. Takayama, S. Qian, A. Gambotto, P. D. Robbins, A. W. Thomson. 1999. Genetic engineering of dendritic cells to express immunosuppressive molecules (viral IL-10, TGF-, and CTLA4Ig). J. Leukocyte Biol. 66:293.(Katsuaki Sato, Takashi Na)
TRAIL preferentially induces apoptotic cell death in a wide variety of transformed cells, whereas it induces no apoptosis, but inhibits activation of Ag-specific T cells via blockade of cell cycle progression. Although accumulating results suggest that TRAIL is involved in the maintenance of immunological homeostasis under steady state conditions as well as in the initiation and progression of immunopathologies, the potential regulatory effect of TRAIL on immune responses and its therapeutic potential in immunological diseases remains unclear. We report in this study the potential usefulness of TRAIL-transduced dendritic cells (DCs) for the treatment of lethal acute graft-vs-host disease (GVHD) and leukemia relapse. DCs genetically modified to express TRAIL showed potent cytotoxicity against both alloreactive T cells and leukemic cells through the induction of apoptosis. In addition, treatment with genetically modified DCs expressing TRAIL of allogeneic BM transplants recipients with leukemia was effective for protection against acute GVHD and leukemia relapse. Thus, gene transfer of TRAIL to DCs is a novel modality for the treatment of acute GVHD and leukemia relapse by selective targeting of pathogenic T cells and leukemic cells.
Introduction
Tumor necrosis factor-related apoptosis-inducing ligand, also known as Apo2 ligand, is a type II transmembrane protein belonging to the TNF family (1). TRAIL can potentially interact with five different receptors. These include death receptor (DR43; TRAIL-R1), DR5 (TRAIL-R2), decoy receptor (DcR1; TRAIL-R3), DcR2 (TRAIL-R4), and a soluble receptor called osteoprotegerin (1). Receptors for TRAIL are constitutively expressed in a variety of cell types (1). In contrast, the constitutive expression of TRAIL was observed in liver NK cells, whereas the levels of TRAIL expression in T cells as well as NK cells can be markedly up-regulated after cell activation (2, 3, 4, 5). In addition, TRAIL preferentially induces apoptotic cell death in a wide variety of transformed cells, whereas it induces no apoptosis, but inhibits activation of Ag-specific T cells via blockade of cell cycle progression (6, 7).
The presence of multiple receptors for TRAIL strongly suggests that TRAIL is involved in the maintenance of immunological homeostasis under steady state conditions as well as in the initiation and progression of immunopathologies. Previous studies have shown that TRAIL plays a crucial role in the surveillance of tumor initiation and metastasis in mice (2). Although the role of TRAIL in the negative selection of thymocytes remains controversial (8, 9), TRAIL plays a crucial role in the regulation of autoimmune diseases (6, 8, 10). However, the potential regulatory effect of TRAIL on immune responses and its therapeutic potential in immunological diseases are unknown.
Dendritic cells (DCs) are APC that consist of heterogeneous subsets with different lineages and maturity; they not only initiate immunity, but are also involved in the induction of tolerance in vivo (11, 12, 13). Therefore, in addition to their original application in the therapy of cancer and infectious diseases, strategies using immunoregulatory DCs are expected to be effective for the prevention and treatment of autoimmune diseases, allergic diseases, and allograft rejection.
Genetic modification of DCs with genes encoding immunoregulatory molecules provides a potential approach for Ag-specific regulation of T cell-mediated immunity by selectively targeting Ag-specific T cells. The use of these genetically modified DCs was reportedly effective for the prevention of experimental autoimmune and allergic diseases as well as allograft rejection in animals through the down-regulation of Ag-specific T cell responses (14, 15, 16, 17).
Allogeneic bone marrow (BM) transplantation (BMT) is an effective treatment for hematologic malignancies as well as genetic disorders (18, 19, 20, 21). However, acute graft-vs-host disease (GVHD), which is caused by alloreactive T cells in donor BM inocula, is a major cause of morbidity and mortality in patients undergoing allogeneic BMT (18, 19, 20, 21). Although the incidence and severity of acute GVHD can be dramatically improved by T cell depletion or the combination of immunosuppressive agents, the risk of leukemia relapse may be increased in turn, possibly due to the lack of antileukemia effect of allogeneic T cells infused, so-called graft-vs-leukemia (GVL) effect (19, 20, 21). Therefore, there is an increasing interest in the development of strategies that suppress acute GVHD but enhance the GVL effect.
In this study we report that genetically modified TRAIL-expressing DCs induce apoptotic cell death in alloreactive T cells and ameliorate acute GVHD while exerting an antileukemic effect.
Materials and Methods
Media and reagents
The medium used throughout was RPMI 1640 (Sigma-Aldrich) or DMEM (Sigma-Aldrich) supplemented with antibiotic-antimycotic (Invitrogen Life Technologies) and 10% heat inactivated FCS (Invitrogen Life Technologies). GM-CSF, IL-2, IL-4, IFN-, and soluble TRAIL were purchased from PeproTech.
Cell preparations
Human immature DCs (iDCs) were obtained by culturing peripheral blood monocytes with GM-CSF (50 ng/ml) and IL-4 (50 ng/ml) for 7 days (22). For the preparation of mature DCs (mDCs), cells were subsequently cultured with LPS (1 μg/ml; Sigma-Aldrich) for another 4 days (22). Murine iDCs were prepared by culturing BM cells obtained from female BALB/c mice (H-2d) or C57BL/6 mice (H-2b; all from Charles River Laboratories) with murine GM-CSF (20 ng/ml) for 8 days, and mDCs were obtained from culture of iDCs with LPS (1 μg/ml) for 4 days (21). Human T cells were purified from PBMC with a T cell negative isolation kit (Dynal Biotech), and CD4+ T cells were then negatively selected from T cells with anti-CD8 mAb (BD Biosciences) plus goat anti-mouse IgG Ab-conjugated immunomagnetic beads (Dynal Biotech) (22). Murine T cells were negatively selected from splenic mononuclear cells (MNC) obtained from C57BL/6 mice with mAbs to Ly-76, B220, Ly-6G, and I-A/I-E (all from BD Biosciences) plus sheep anti-rat IgG Ab-conjugated immunomagnetic beads (Dynal Biotech) (21). Subsequently, CD4+ T cells were negatively selected from T cells with anti-CD8 mAb (BD Biosciences) in combination with sheep anti-rat IgG Ab-conjugated immunomagnetic beads (21). Con A blasts were obtained from the culture of human or murine T cells with Con A (2.5 μg/ml; Sigma-Aldrich) for 3 days.
Production of adenovirus encoding the TRAIL genes
The full-length human TRAIL (hTRAIL) cDNA (884 bp) was prepared by RT-PCR amplification of total RNA from Con A blasts with the following oligonucleotide primers: 5'-CAG CAG TCA GAC TCT GAC AG-3' and 5'-TCT TTC CAG GTC AGT TAG CC-3'. The PCR product was subcloned into pCR2.1 vector using TA Cloning kit (Invitrogen Life Technologies), and the nucleotide sequence was confirmed using a 373A automated sequencer (Applied Biosystems) and the fluoresceinated dye terminator cycle sequencing method. The full-length murine TRAIL (mTRAIL) cDNA was prepared from mTRAIL/pMKITNeo expression vector (23). After XhoI and NotI digestion, the 850 bp of mTRAIL cDNA was obtained, and the nucleotide sequence was confirmed as described above.
A replication-deficient adenovirus vector (Ad) expressed from the CAG promotor was generated using an Adenovirus Expression Vector kit (TaKaRa Shuzo), in which an adenoviral cosmid, pAxCAwt, was included. The cosmid pAxCAwt consisted of E1- and E3-deficient Ad type 5 (Ad5) sequences, and the CAG promoter and rabbit globin poly-A were inserted at the former E1 site in reverse orientation with respect to the Ad5 sequences. Briefly, the entire coding sequence of hTRAIL or mTRAIL was blunted using a DNA Blunting kit (TaKaRa Shuzo) and was then subcloned into the SwaI site of pAxCAwt. The resulting cosmids were named hTRAIL/pAxCAwt and mTRAIL/pAxCAwt, respectively. Transfection of human embryonic kidney 293 cells (RIKEN Cell Bank) with these cosmid vectors and Ad backbone sequences (DNA-TPC) that had the E1 and E3 genes deleted was performed according to the manufacturer’s instructions to produce replication-incompetent, E1- and E3-deficient, Ad-expressing hTRAIL or mTRAIL (hTRAIL-Ad or mTRAIL-Ad). The viruses were then prepared by expansion of a single clone generated in 293 cells, which were purified by limiting dilution, and viral particles were isolated and amplified for analysis of hTRAIL or mTRAIL expression by flow cytometry. Recombinant adenoviruses generated from the homologous recombination of pAxCAwt and DNA-TPC were used as virus controls (control-Ad). Recombinant adenovirus titers were determined by plaque assays on 293 cells. These adenoviruses were suspended in culture medium, adjusted to 2 x 108 PFU/ml, and stored at –80°C until use.
Adenoviral infection
For Ad-mediated gene transfer into human DCs by centrifugal transduction (24), 500 μl of cells (106 cells) were mixed with 500 μl of adenoviral vector (multiplicity of infection (MOI) of 10 or 50), and 1 ml of the mixture was poured into a polypropylene tube (BD Biosciences). The tubes were centrifuged at 2000 x g at 37°C for 2 h. After the centrifugal transduction, the cells were washed twice in PBS. DCs were resuspended in culture medium under various culture conditions and cultured for the indicated periods in tissue culture dishes (BD Biosciences). For adenoviral infection of murine DCs, 500 μl of iDCs (106 cells) were mixed with 500 μl of adenoviral vector (MOI of 50) in a polypropylene tube, and 1 ml of the mixture was incubated at 37°C. After a 2-h incubation, culture medium was added to the cells, then the cultures were incubated with LPS (1 μg/ml) for 4 days in tissue culture dishes.
Flow cytometry
Cells were stained with the following mAbs to human and murine markers: CD3, CD4, CD11c, CD40, CD80, CD86, HLA-A/B/C, HLA-DR, H-2Kb, H-2Kd, I-A/I-E, and isotype-matched control IgG (all from BD Biosciences); CD83 (Coulter Immunology); and FITC-conjugated goat anti-rat IgG Ab (Santa Cruz Biotechnology). The purified mAbs to hTRAIL (RIK-2), mTRAIL (N2B2), human DR4 (hDR4; DJR1), human DR5 (hDR5; DJR2), human DcR1 (hDcR1; DJR3), human DcR2 (hDcR2; DJR4), and murine DR5 (mDR5; MD5–1) were prepared as described previously (5, 25, 26). Fluorescent staining was analyzed on a FACScan flow cytometer using CellQuest software (BD Biosciences), and the data are expressed as the mean fluorescence intensity (MFI).
Cytotoxicity assay
Adenoviral gene-transduced untransduced DCs were cultured with Na251CrO4 (NEN Life Science Products)-labeled Jurkat cells, L929 cells, P815 cells (104; all from RIKEN Cell Bank), or Con A blasts (104) for 4 h at various E:T cell ratios in the presence or the absence of 10 μg/ml anti-hTRAIL mAb, anti-hDR5 mAb, anti-mTRAIL mAb, or control IgG. As a control, soluble hTRAIL was added to the target cells at the indicated concentrations (10–1000 ng/ml). The radioactivity of the supernatants was measured, and the percent-specific lysis was calculated (21, 22). Spontaneous release was <10% of total release.
Cell proliferation assay
Human CD4+ T cells (105) were stimulated with or without plate-bound anti-human CD3 mAb (BD Biosciences) plus soluble anti-human CD28 mAb (BD Biosciences) in the presence or the absence of soluble hTRAIL (1000 ng/ml), anti-hTRAIL mAb, or control Ig (each 10 μg/ml). For human and murine allogeneic MLR, CD4+ T cells (105) were cultured in 96-well plates (BD Biosciences) with various numbers of irradiated (15 Gy from a 137Cs source; MBR-1505R2; Hitachi Medical) allogeneic DCs in the presence or the absence of soluble hTRAIL (1000 ng/ml), anti-hTRAIL mAb (10 μg/ml), anti-mTRAIL mAb (10 μg/ml), anti-hDR5 mAb (10 μg/ml), or control Ig (10 μg/ml). [3H]Thymidine incorporation was measured on day 5 for the last 18 h (21, 22).
In vitro analysis of T cell responses
For measuring cell division, CD4+ T cells were labeled with CFSE (Molecular Probes) according to the manufacturer’s instructions. CFSE-labeled CD4+ T cells (5 x 106) were cultured with irradiated (15 Gy) allogeneic DCs (5 x 105) for 3 days, then T cells were negatively selected with anti-human CD11c mAb (BD Biosciences) plus goat anti-mouse IgG Ab-conjugated immunomagnetic beads (22). These T cell preparations contained <0.1% CD11c+ cells as assessed by FACS analysis. The CFSE-positive cells were then analyzed by flow cytometry. Apoptosis in allogeneic DC-stimulated CD4+ T cells was measured by flow cytometry using an Annexin VFITC apoptosis detection kit (R&D Systems). For cell cycle analysis, the stimulated CD4+ T cells were incubated with BrdU (BD Biosciences) at 10 μM for 1 h at 37°C. Staining of incorporated BrdU was performed using a BrdU Flow kit (BD Bioscience) according to the manufacturer’s instructions. Cells were stained with FITC-conjugated anti-BrdU Ab for 20 min at room temperature. 7-Amino-actinomycin D (20 μg/ml) was added to the cell suspension before flow cytometric analysis.
Acute GVHD model
BALB/c recipient mice (five animals in each group) received lethal total body irradiation (TBI; 10 Gy) and then a single i.v. injection of C57BL/6 BM cells (1.5 x 107/mouse) plus splenic MNC (1.5 x 107/mouse) through the tail vein (21). The day of transplantation was designated day 0. The i.v. injection of adenoviral gene-transduced or untransduced DCs from BALB/c mice or C57BL/6 mice (106 to 5 x 106/mouse) 2 days after transplantation. For the in vivo blockade experiments, recipients were i.p. injected with control Ig or anti-mTRAIL mAb (1 mg/mouse) before i.v. injection of DCs. The recipients were monitored every day for survival. Some recipients were killed 5 days after transplantation to obtain serum and splenic MNCs.
Leukemia relapse model
BALB/c recipients (five animals in each group) were inoculated i.v. with P815 cells (2 x 105/mouse) 2 days before TBI (10 Gy) and i.v. transplantation with C57BL/6 BM cells (1.5 x 107/mouse) (21). The transplanted recipients received a single i.v. injection of adenoviral gene-transduced or untransduced DCs from BALB/c mice (5 x 106/mouse) 2 days after transplantation. Recipients were monitored every day for survival. Hepatosplenomegaly due to tumor burden in the dead mice was confirmed by laboratory.
Statistical analyses
Statistically significant differences were determined by Student’s paired t test or Mann-Whitney’s U test. A value of p < 0.01 was considered significant.
Results
Regulatory function of human DCs genetically modified to express TRAIL
To test the potential use of TRAIL-expressing DCs for selectively targeting Ag-specific T cells, we examined the conditions for the generation of human DCs genetically engineered to express TRAIL using Ad. Although stimulation of iDCs with IFN- or LPS induced low levels of TRAIL expression (27, 28), adenoviral gene transduction of hTRAIL (hTRAIL-Ad) into DCs resulted in higher expression of TRAIL. Of note, introduction of hTRAIL-Ad into iDCs followed by stimulation with LPS (Fig. 1, A and B) or TNF- (data not shown) resulted in the generation of mDCs with the highest level of TRAIL expression. Adenoviral infection had little or no effect on the expression of MHC and costimulatory molecules (Fig. 1C). The hTRAIL-Ad-infected mDCs (hTRAIL-Ad/mDCs) showed a more potent killing activity against hTRAIL-sensitive Jurkat cells than soluble hTRAIL, mDCs, and control-Ad-infected mDCs (control-Ad/mDCs; Fig. 1D). In addition, the cytotoxicity of hTRAIL-Ad/mDCs against Jurkat cells was blocked by anti-hTRAIL mAb and anti-hDR5 mAb (Fig. 1, D and E). These results indicate that hTRAIL was functionally expressed on the hTRAIL-Ad/mDCs. In contrast, Con A blasts (Fig. 1C) and DCs (data not shown) were relatively resistant to TRAIL-mediated cytotoxicity.
FIGURE 1. Generation of genetically modified human DCs expressing TRAIL. A and B, The iDCs were not infected or were infected with control-Ad or hTRAIL-Ad at MOI of 10 (A) and 50 (B). In some experiments, uninfected DCs or DCs infected with hTRAIL-Ad were stimulated with IFN- (A) or LPS (A and B) before or after infection. The expression of hTRAIL on DCs was analyzed by flow cytometry at the indicated time points. The expression level of TRAIL was expressed as the mean MFI ± SD of four individual experiments. The results are representative of two experiments with similar results. C, The iDCs were not infected or were infected with control-Ad or hTRAIL-Ad at an MOI of 50, followed by stimulation with LPS for generation of mDCs. Subsequently, mDCs were stained with the indicated mAbs, and cell surface expression was analyzed by flow cytometry. Data are represented by a dot plot for the expression of MHC and costimulatory molecules or by a histogram in which cells were stained with anti-hTRAIL mAb (thick lines) or isotype-matched control Ig (thin lines). The results are representative of four experiments with similar results. D, The cytotoxicity of soluble hTRAIL, uninfected DCs, or DCs infected with control-Ad or hTRAIL-Ad against Jurkat cells or Con A blasts at various E:T cell ratios in the presence or the absence of control Ig or anti-hTRAIL mAb was analyzed by the 4-h 51Cr release assay. Data were expressed as the mean ± SD of triplicate samples, and the results are representative of four experiments with similar results. E, The cytotoxicity of soluble hTRAIL, uninfected DCs, or DCs infected with control-Ad or hTRAIL-Ad against Jurkat cells at an E:T cell ratio of 10 in the presence or the absence of control Ig, anti-hTRAIL mAb, or anti-hDR5 mAb was analyzed by the 4-h 51Cr release assay. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with control Ig, by Student’s paired t test.
We also examined the expression levels of TRAIL receptors on various cell types (Fig. 2A). The iDCs constitutively expressed DR4, DR5, DcR1, and DcR2 at similar levels, and the expression of these receptors was slightly reduced after maturation. Unlike Jurkat cells, which predominantly expressed DR5, little or no expression of these receptors was observed on unstimulated CD4+ T cells and Con A blasts. Interestingly, stimulation of CD4+ T cells with allogeneic mDCs induced specific up-regulation of DR5, whereas stimulation with anti-CD3 and anti-CD28 mAbs up-regulated DR5 to a lesser degree (data not shown).
FIGURE 2. Regulatory function of human DCs genetically modified to express TRAIL. A, The expression of TRAIL receptors on the indicated cells was analyzed by flow cytometry. Values were expressed as the ratio of MFI with respective mAb compared with the MFI with control Ig, and data are expressed as the mean ± SD of four individual experiments. *, p < 0.01 compared with control Ig, by Student’s paired t test. B, Human CD4+ T cells (105) were stimulated with or without plate-bound anti-human CD3 plus soluble anti-human CD28 mAb in the presence or the absence of soluble hTRAIL and anti-hTRAIL mAb or control Ig. In another experiment, CD4+ T cells (105) were cultured with uninfected or control-Ad or hTRAIL-Ad-infected allogeneic mDCs (104) in the presence or the absence of soluble hTRAIL, anti-hTRAIL mAb, anti-hDR5 mAb, or control Ig. The proliferative response was measured by [3H]thymidine uptake on day 5. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01, by Student’s paired t test. C, CD4+ T cells (105) were cultured with various numbers of uninfected or control-Ad- or hTRAIL-Ad-infected allogeneic mDCs, and the proliferative response was measured by [3H]thymidine uptake on day 5. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with uninfected DCs, by Student’s paired t test. D, CFSE-labeled CD4+ T cells (5 x 106) were primed with uninfected or control-Ad- or hTRAIL-Ad-infected allogeneic mDCs (5 x 105) for 3 days, and CFSE levels were analyzed by flow cytometry. Data are represented by a histogram. The data shown are representative of four experiments with similar results. E and F, CD4+ T cells (5 x 106) were primed with uninfected or control-Ad- or hTRAIL-Ad-infected allogeneic mDCs (5 x 105) for 3 days, then CD4+ T cells were analyzed for apoptosis (E) or cell cycle (F) by flow cytometry. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with T cells primed with mDCs (E) or unprimed T cells (F), by Student’s paired t test.
We also examined the T cell regulatory function of hTRAIL-Ad/DCs. Soluble TRAIL showed a minimal inhibition of proliferation when CD4+ T cells were stimulated with anti-CD3 and anti-CD28 mAbs or allogeneic mDCs (Fig. 2B). In contrast, hTRAIL-Ad/DCs, but not control-Ad/mDCs, displayed a potent suppressive effect on the proliferation of alloreactive CD4+ T cells (Fig. 2, B and C). This suppression was abrogated by both anti-hTRAIL mAb and anti-hDR5 mAb, but not by control Ig (Fig. 2B). These results indicate that DCs genetically modified to express TRAIL could inhibit the proliferation of allogeneic CD4+ T cells through the interaction with DR5.
Previous studies have shown that soluble TRAIL did not induce apoptosis, but inhibited the proliferation of T cells through blockage of cell cycle progression (6, 7). To clarify the mechanism underlying the T cell regulatory function of TRAIL-transduced DCs, we characterized CD4+ T cells primed with hTRAIL-Ad/DCs. The CFSE labeling assay showed that the proportion of dividing cells was significantly reduced in allogeneic CD4+ T cells primed with hTRAIL-Ad/DCs compared with those primed with mDCs and control-Ad/mDCs (Fig. 2D). In contrast, hTRAIL-Ad/DCs induced significantly more apoptosis in allogeneic CD4+ T cells than mDCs and control-Ad/mDCs (Fig. 2E). Also in the cell cycle analysis, numerous apoptotic cells with sub-G0/G1 DNA content were detected in allogeneic CD4+ T cells primed with hTRAIL-Ad/DCs, whereas the proliferating cells in S phase and G2+M phase were increased in allogeneic CD4+ T cells primed with mDCs or control-Ad/mDCs (Fig. 2F). These results indicate that the DCs genetically modified to express TRAIL suppress the proliferation and cell division of allogeneic CD4+ T cells through the induction of apoptosis rather than cell cycle arrest.
TRAIL-transduced DCs ameliorate murine acute GVHD
The above results indicated that mDCs genetically modified to express hTRAIL could efficiently suppress the proliferation of alloreactive CD4+ T cells through the induction of apoptosis. We therefore tested the in vivo suppressive function of murine DCs genetically modified to express mouse TRAIL (mTRAIL-Ad/DCs). Similar to hTRAIL-Ad/mDCs, mTRAIL-Ad/DCs showed the functional expression of mTRAIL (Fig. 3, A and B). In addition, mTRAIL-Ad/DCs impaired the proliferation of alloreactive CD4+ T cells (Fig. 3C), and they induced a higher rate of apoptosis in allogeneic CD4+ T cells than control-Ad/DCs (data not shown).
FIGURE 3. Regulatory function of murine DCs genetically modified to express TRAIL. A, Murine DCs were uninfected or infected with control-Ad or mTRAIL-Ad at an MOI of 50, followed by stimulation with LPS. Subsequently, DCs were stained with the indicated mAbs, and cell surface expression was analyzed by flow cytometry. Data are represented by a dot plot for the expression of MHC and costimulatory molecules or by a histogram in which cells were stained with anti-hTRAIL mAb (thick lines) or isotype-matched control Ig (thin lines). The data shown are representative of four experiments with similar results. B, The cytotoxicity of uninfected DCs or DCs infected with control-Ad or mTRAIL-Ad against L929 cells or Con A blasts was analyzed in the presence or the absence of control Ig or anti-mTRAIL mAb. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. C, CD4+ T cells (105) from naive C57BL/6 mice were cultured with various numbers of uninfected or control-Ad- or mTRAIL-Ad-infected mDCs from BALB/c mice, and the proliferative response was measured on day 5 by [3H]thymidine uptake. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with uninfected DCs, by Student’s paired t test. D, The expression of DR5 on the indicated murine cells was analyzed by flow cytometry. Values were expressed as the ratio of MFI with anti-DR5 mAb compared with the MFI with control Ig, and data are expressed as the mean ± SD of four individual experiments. *, p < 0.01 compared with control Ig, by Student’s paired t test.
We also examined the expression level of mDR5 on various cell types (Fig. 3D). Flow cytometric analysis showed that mTRAIL-sensitive L929 cells as well as P815 cells expressed high level of mDR5. In contrast, mDCs exhibited slightly lower expression of mDR5 than mDCs. We also observed that DCs expressed the transcripts of all TRAIL receptors at similar levels, and these transcriptional expressions were reduced after maturation (data not shown). In addition, the primed CD4+ T cells with allogeneic mDCs showed higher expression of mDR5 than unstimulated CD4+ T cells and Con A blasts.
We then tested the therapeutic efficacy of mTRAIL-Ad/DCs against acute GVHD. All BALB/c recipients died on day 8 after transplantation of C57BL/6 BM and spleen cells (Fig. 4A). In these mice, clinical symptoms of acute GVHD, such as hair ruffling, lower mobility, and weight loss, became apparent within 6 days. In contrast, all BALB/c recipients of syngeneic BM and spleen cells survived >60 days without apparent acute GVHD (data not shown). A single injection of BALB/c-derived DCs or control-Ad/DCs 2 days after transplantation of C57BL6 BM and spleen cells to BALB/c recipients did not significantly affect the lethality caused by acute GVHD (Fig. 4A). In contrast, BALB/c-derived mTRAIL-Ad/DCs ameliorated acute GVHD in a dose-dependent fashion (Fig. 4A), whereas C57BL/6-derived mTRAIL-Ad/DCs showed a minimal inhibitory effect (Fig. 4B). In addition, in vivo blockade of mTRAIL with anti-mTRAIL mAb abrogated the therapeutic effect of mTRAIL-Ad/DCs (Fig. 4C).
FIGURE 4. Suppressive effect of TRAIL-transduced murine DCs on acute GVHD in murine allogeneic BMT. A, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without uninfected or control-Ad- or mTRAIL-Ad-infected BALB/c DCs (106 to 5 x 106/mouse) on 2 days after BMT. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with untreated mice, by Mann-Whitney’s U test. B, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without uninfected or mTRAIL-Ad-infected BALB/c or C57BL/6 DCs (5 x 106/mouse) on 2 days after BMT. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with untreated mice, by Mann-Whitney’s U test. C, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without mTRAIL-Ad-infected BALB/c DCs (5 x 106/mouse) and control Ig or anti-mTRAIL mAb on 2 days after BMT. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with control Ig-treated mice, by Mann-Whitney’s U test.
We also examined the in vivo regulatory effect of mTRAIL-Ad/DCs in the recipients of allogeneic transplantation. The administration of mTRAIL-Ad/DCs significantly reduced the number of total splenocytes (Fig. 5A) and inhibited the expansion of donor-derived CD4+ and CD8+ T cells in the spleen (Fig. 5B), whereas that of untransduced DCs increased the number of these cells. Moreover, donor-derived CD4+ T cells from the mTRAIL-Ad/DC-treated recipients showed a significantly reduced proliferative response to the host-type mDCs, whereas those from the DC-treated recipients showed an enhanced response (Fig. 5C). Furthermore, donor-derived CD8+ T cells from the mTRAIL-Ad/DC-treated recipients showed markedly impaired CTL activity against P815 cells expressing the host-type alloantigen, whereas those from the DC-treated recipients showed enhanced CTL activity (Fig. 5D). In addition, mTRAIL-Ad/DC-treated recipients showed a significant reduction of production of serum IFN-, TNF-, and IL-12p40 compared with untreated and control-Ad/DC-treated recipients (Fig. 5E). These results indicate that mTRAIL-Ad/DCs is could efficiently ameliorate acute GVHD through the suppression of alloreactive T cell responses.
FIGURE 5. Suppressive effect of TRAIL-transduced murine DCs on allogeneic responses of donor-derived T cells in transplanted mice. A–E, BALC/c recipient mice of C57BL/6 BMT were i.v. injected with or without uninfected or mTRAIL-Ad-infected BALB/c DCs (5 x 106/mouse) on 2 days after BMT. On 5 days after BMT, splenic MNC and sera were obtained from indicated mice. A, The number of donor-derived (H-2b+) splenic MNC was analyzed by flow cytometry. Data were expressed as the mean ± SD of five mice in each group. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. B, The constitution of donor-derived (H-2b+) T cell subsets was analyzed by flow cytometry. Data are expressed as the mean ± SD of five mice in each group. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. C, Donor-derived CD4+ T cells (105) were cultured with various numbers of irradiated BALB/c DCs, and the proliferative response was measured by [3H]thymidine uptake on day 5. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. D, Donor-derived CD8+ T cells were subjected to CTL assay against P815 cells at various E:T cell ratios. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with untreated recipients, by Student’s paired t test. E, Concentrations of IFN-, TNF-, and IL-12p40 in serum were evaluated by ELISA. Data are expressed as the mean ± SD of five mice in each group. *, p < 0.01 compared with untreated recipients, by Student’s paired t test.
Protection from leukemia relapse by mTRAIL-transduced DCs
TRAIL has been implicated in the GVL effect associated with allogeneic BMT (20). Therefore, we next examined the antileukemic effect of TRAIL-transduced DCs. The recipient mice were i.v. inoculated with or without P815 leukemia cells 2 days before TBI and transplantation of allogeneic BM cells. Consistent with previous reports (20, 21), the mice transplanted with allogeneic BM cells alone survived >60 days without apparent acute GVHD. All leukemia-bearing mice that received TBI alone died within 14 days with marked hepatosplenomegaly. In contrast, the leukemia-bearing recipients of allogeneic BM cells died within 26 days after transplantation, indicating that alloreactive T cells in BM cells exhibited an insufficient GVL effect. We then examined the therapeutic effect of mTRAIL-transduced DCs against leukemia relapse in this model. The mTRAIL-Ad/DCs showed potent cytotoxicity against P815 cells in vitro (Fig. 6A). A single injection of recipient-type mTRAIL-Ad/DCs, but not DCs or control-Ad/DCs, 2 days after transplantation markedly prolonged the survival of leukemia-bearing mice (Fig. 6B). These results suggested that TRAIL-transduced DCs are useful not only to ameliorate acute GVHD but also to suppress leukemia relapse.
FIGURE 6. Protective effect of TRAIL-transduced murine DCs against leukemia relapse. A, The cytotoxicity of uninfected or control-Ad- or mTRAIL-Ad-infected DCs against P815 cells was analyzed by the 4-h 51Cr release assay. Data were expressed as the mean ± SD of triplicate samples, and the values shown are representative of four experiments with similar results. *, p < 0.01 compared with uninfected DCs, by Student’s paired t test. B, BALB/c recipient mice were inoculated i.v. with P815 cells (2 x 105) on day –2 and received TBI and i.v. transplantation with C57BL/6 BM cells on day 0. The recipient mice were i.v. injected with or without uninfected or control-Ad- or mTRAIL-Ad-infected BALB/c DCs (5 x 106/mouse) on day 2. Mice were monitored daily for survival. The results are representative of two individual experiments with similar results. *, p < 0.01 compared with untreated mice, by Mann-Whitney’s U test.
Discussion
Gene transfer of immunoregulatory molecules to Ag-presenting DCs provides an attractive approach for the treatment of immunological diseases. Our present findings demonstrate the efficacy of TRAIL gene-transduced DCs for the therapy of lethal acute GVHD and leukemia relapse via elimination of the pathogenic T cells and leukemia cells.
Efficient expression of the introduced gene in target cells is crucial for regulating their function. We showed that the gene transfer of TRAIL into iDCs followed by inflammatory stimulation caused maximal expression of TRAIL. The precise molecular mechanism by which inflammatory stimulation up-regulated the expression of transduced TRAIL gene in DCs remains unknown, but inflammatory stimulation might activate certain transcriptional factors, leading to transactivation of the transduced TRAIL gene. In contrast, mDCs show more potent interaction with Ag-specific T cells than iDCs due to their higher expression of MHC and adhesion/costimulatory molecules. Thus, the maturation of iDCs genetically engineered to express TRAIL by inflammatory stimulation provides advantages for TRAIL expression as well as Ag-specific interaction with T cells.
We showed that human and murine TRAIL-sensitive cell types, including CD4+ T cells primed with allogeneic DCs as well as Jurkat cells and L929 cells, exhibited high expression of DR5, indicating that DR5 is crucial for TRAIL-mediated cytotoxicity. In contrast, analysis of the cell surface expression of TRAIL receptors as well as their transcriptional expression (data not shown) revealed that human and murine DCs showed similar expression levels of these TRAIL receptors. Therefore, suppression of binding of TRAIL to DRs by DcRs may contribute to the low sensitivity of these cell types to TRAIL-mediated cytotoxicity. Collectively, our results suggest that the balance of the expression of DRs and DcRs may determine the sensitivity to TRAIL-mediated cytotoxicity.
The weak suppressive effect of recombinant soluble TRAIL on alloreactive T cell proliferation might be due to a blockage of cell cycle progression as previously reported (6, 7), whereas mDCs genetically modified to express TRAIL could effectively induce apoptosis in alloreactive T cells. Although the precise molecular mechanism underling the difference in the biological function between soluble TRAIL and membrane-bound transduced TRAIL on DCs remains unknown, this phenomenon might reflect the strength of intracellular signaling via DRs. Additional study will be needed to test this possibility.
Although host-matched or -mismatched mTRAIL-transduced DCs may lyse donor-derived T cells that overexpresses mDRs regardless of histocompatibility differences, host-matched mTRAIL-transduced DCs could more effectively lyse host-reactive, donor-derived T cells than host-mismatched, mTRAIL-transduced DCs, because they show more potent reactivity to host-matched, mTRAIL-transduced DCs than host-mismatched, mTRAIL-transduced DCs. Indeed, BALB/c-derived, mTRAIL-Ad/DCs, but not C57BL/6-derived, mTRAIL-Ad/DCs, ameliorated acute GVHD in BALB/c recipients transplanted with C57BL6 BM and spleen cells. Therefore, host-type, TRAIL-transduced DCs are useful for the protection of recipients from lethality induced by acute GVHD in allogeneic BMT.
The expansion of donor-derived CD4+ and CD8+ T cells was markedly suppressed by a single injection of host-type, mTRAIL-transduced DCs. These results suggest that interaction of alloreactive CD4+ and CD8+ T cells with the infused TRAIL-transduced mDCs caused their apoptosis via up-regulation of mDR5, resulting in their selective elimination in vivo. In contrast, donor-derived CD4+ T cells obtained from these protected recipients showed a reduced proliferative response to host-type mDCs. Our results imply that the intracellular signaling events involving TRAIL/DR5 induce not only apoptosis but also functional deficiency in the targeted CD4+ T cells. We also showed that the CTL activity of donor-derived CD8+ T cells was greatly impaired in these protected recipients. Although the precise mechanism remains unclear, the functional impairment of donor-derived CD4+ T cells may contribute to the reduced response of donor-derived CD8+ T cells.
The deletion of donor-derived alloreactive T cells by mTRAIL-transduced DCs can ameliorate acute GVHD, but also may impair their GVL effect. However, our results showed that the TRAIL-transduced DCs could protect BMT recipients from leukemia relapse. This seems likely to be mediated by antileukemic effect of TRAIL expressed on DCs, because the TRAIL-transduced DCs exhibited potent cytotoxic activity in vitro. The ex vivo removal of donor T cell proportions from the marrow graft not only reduced the incidence and severity of acute GVHD, but also increased the risk of leukemia relapse due to lack of GVL effect (19, 20, 21). Therefore, our results imply that the use of TRAIL-transduced DCs and T cell-depleted BM inocula might be an alternative strategy for the prevention of acute GVHD and leukemia relapse in human leukemia patients undergoing allogeneic BMT.
The role of TRAIL in hepatic cell death is controversial (29, 30). We observed that a single injection of mTRAIL-Ad/DCs did not induce any apparent liver cytotoxicity in normal mice or recipients of allogeneic BMT (data not shown). Additional study will be needed to determine the safety and potential side effects of TRAIL-transduced DCs in vivo.
In the present study we demonstrated a potent effect of TRAIL-transduced DCs to ameliorate acute GVHD. A similar strategy may be also useful for Ag-specific immunosuppression in allogeneic organ transplantation, autoimmune diseases, and allergic diseases. Studies are now underway to address this possibility.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We thank K. Sato, Y. Sato, A. Takeuchi, and M. Yamamoto for excellent assistance.
Footnotes
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 Japan Rheumatism Foundation (to K.S.), the Uehara Memorial Foundation (to K.S.), and the Kanzawa Medical Research Foundation (to K.S.).
2 Address correspondence and reprint requests to Dr. Katsuaki Sato, Laboratory for Dendritic Cell Immunobiology, Research Center for Allergy and Immunology, RIKEN Yokohama Institute, Suehiro-cho 1-7-22, Tsurumi, Yokohama, Kanagawa 230-0045 Japan. E-mail address: katsuaki@rcai.riken.jp
3 Abbreviations used in this paper: DR, death receptor; Ad, adenovirus vector; BM, bone marrow; BMT, BM transplantation; DC, dendritic cell; DcR, decoy receptor; GVHD, acute graft-vs-host diseases; GVL, graft-vs-leukemia; hTRAIL, human TRAIL; iDC, immature DC; mDC, mature DC; MFI, mean fluorescence intensity; MNC, mononuclear cell; MOI, multiplicity of infection; mTRAIL, murine TRAIL; TBI, total body irradiation; TRAIL-Ad, Ad expressing TRAIL.
Received for publication October 19, 2004. Accepted for publication January 16, 2005.
References
Lahm, A., A. Paradisi, D. R. Green, G. Melino. 2003. Death fold domain interaction in apoptosis. Cell Death Differ. 10:10
Smyth, M. J., K. Takeda, Y. Hayakawa, J. J. Peschon, M. R. van den Brink, H. Yagita. 2003. Nature’s TRAIL: on a path to cancer immunotherapy. Immunity 18:1.
Smyth, M. J., E. Cretney, K. Takeda, R. H. Wiltrout, L. M. Sedger, N. Kayagaki, H. Yagita, K. Okumura. 2001. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) contributes to interferon -dependent natural killer cell protection from tumor metastasis. J. Exp. Med. 193:661.
Takeda, K., Y. Hayakawa, M. J. Smyth, N. Kayagaki, N. Yamaguchi, S. Kakuta, Y. Iwakura, H. Yagita, K. Okumura. 2001. Involvement of tumor necrosis factor-related apoptosis-inducing ligand in surveillance of tumor metastasis by liver natural killer cells. Nat. Med. 7:94.
Kayagaki, N., N. Yamaguchi, M. Nakayama, A. Kawasaki, H. Akiba, K. Okumura, H. Yagita. 1999. Involvement of TNF-related apoptosis-inducing ligand in human CD4+ T cell-mediated cytotoxicity. J. Immunol. 162:2639.
Song, K., Y. Chen, R. Goke, A. Wilmen, C. Seidel, A. Goke, B. Hilliard, Y. Chen. 2000. Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is an inhibitor of autoimmune inflammation and cell cycle progression. J. Exp. Med. 191:1095.
Lunemann, J. D., S. Waiczies, S. Ehrlich, U. Wendling, B. Seeger, T. Kamradt, F. Zipp. 2002. Death ligand TRAIL induces no apoptosis but inhibits activation of human (auto)antigen-specific T cells. J. Immunol. 168:4881.
Lamhamedi-Cherradi, S. E., S. J. Zheng, K. A. Maguschak, J. Peschon, Y. H. Chen. 2003. Defective thymocyte apoptosis and accelerated autoimmune diseases in TRAIL–/– mice. Nat. Immunol. 4:255
Cretney, E., A. P. Uldrich, S. P. Berzins, A. Strasser, D. I. Godfrey, M. J. Smyth. 2003. Normal thymocyte negative selection in TRAIL-deficient mice. J. Exp. Med. 198:491.
Hilliard, B., A. Wilmen, C. Seidel, T. S. Liu, R. Goke, Y. Chen. 2001. Roles of TNF-related apoptosis-inducing ligand in experimental autoimmune encephalomyelitis. J. Immunol. 166:1314.
Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendra, K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767.
Steinman, R. M., M. C. Nussenzweig. 2002. Avoiding horror autotoxicus: the importance of dendritic cells in peripheral T cell tolerance. Proc. Natl. Acad. Sci. USA 99:351.
Morelli, A. E., A. W. Thomson. 2003. Dendritic cells: regulators of alloimmunity and opportunities for tolerance induction. Immunol. Rev. 196:125.
Matsue, H., K. Matsue, M. Walters, K. Okumura, H. Yagita, A. Takashima. 1999. Induction of antigen-specific immunosuppression by CD95L cDNA-transfected ‘killer’ dendritic cells. Nat. Med. 5:930.
Lu, L., W. C. Lee, T. Takayama, S. Qian, A. Gambotto, P. D. Robbins, A. W. Thomson. 1999. Genetic engineering of dendritic cells to express immunosuppressive molecules (viral IL-10, TGF-, and CTLA4Ig). J. Leukocyte Biol. 66:293.(Katsuaki Sato, Takashi Na)