Secondary Lymphoid Tissue Chemokine in Enhancement of Antitumor ...

The Journal of Immunology

Blocking Programmed Death-1 Ligand-PD-1 Interactions by Local Gene Therapy Results in Enhancement of Antitumor Effect of Secondary Lymphoid Tissue Chemokine1 Yu-Fei He, Gui-Mei Zhang, Xiao-Hong Wang, Hui Zhang, Ye Yuan, Dong Li, and Zuo-Hua Feng2 The negative signal provided by interactions of programmed death-1 (PD-1) and its ligands, costimulatory molecules PD-L1 (also B7-H1) and PD-L2 (also B7-DC), is involved in the mechanisms of tumor immune evasion. In this study, we found that this negative signal was also involved in immune evasion in tumor immunotherapy. When we used different doses of a constructed eukaryotic expression plasmid, pSLC, which expresses functional murine secondary lymphoid tissue chemokine (SLC, CCL21), to treat BALB/c mice inoculated with H22 murine hepatoma cells, the inhibitory effect was enhanced along with the increase of pSLC dosage. Unexpectedly, however, the best complete inhibition rate of tumor was reached when pSLC was used at the dosage of 50 ␮g but not 100 or 200 ␮g. RT-PCR and real-time PCR revealed that both PD-L1 and PD-L2 genes were expressed in tumor and vicinal muscle tissues of tumor-bearing mice and the expression level was significantly increased if a higher dosage of pSLC was administered. We then constructed a eukaryotic expression plasmid (pPD-1A) that expresses the extracellular domain of murine PD-1 (sPD-1). sPD-1 could bind PD-1 ligands, block PD-Ls-PD-1 interactions, and enhance the cytotoxicity of tumorspecific CTL. Local gene transfer by injection of pPD-1A mediated antitumor effect and improved SLC-mediated antitumor immunity. The combined gene therapy with SLC plus sPD-1 did not induce remarkable autoimmune manifestations. Our findings provide a potent method of improving the antitumor effects of SLC and possibly other immunotherapeutic methods by local blockade of negative costimulatory molecules. The Journal of Immunology, 2004, 173: 4919 – 4928.

E

ffective antitumor responses require both APC and lymphocyte effectors (1). Given the chemoattractive and promotive properties of chemokines on the effector function of different immune cells in vitro and in vivo, numerous studies on the effects of chemokines in the antitumor responses have recently been initiated, including secondary lymphoid tissue chemokine (SLC;3 CCL21) (2). SLC, also referred to as exodus 2, TCA4, or 6Ckine, is mainly expressed in lymphoid tissues (3– 6) and is selective in its recruitment of the activated and naive T lymphocytes (7–10), and dendritic cells (DCs) (11–13). SLC is believed to play an important role in the initiation of an immune response. Injection of rSLC into the tumor or lymph node region evidenced T cell-dependent antitumor responses (14 –16). The colon carcinoma tumor cell line

Department of Biochemistry and Molecular Biology, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, The People’s Republic of China Received for publication March 31, 2004. Accepted for publication August 16, 2004. 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 National Development Program (973) for Key Basic Research (2002CB513100), China. 2 Address correspondence and reprint requests to Dr. Zuo-Hua Feng, Department of Biochemistry and Molecular Biology, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, The People’s Republic of China. E-mail address: [email protected] 3 Abbreviations used in this paper: SLC, secondary lymphoid tissue chemokine; DC, dendritic cell; PD-1, programmed death-1; PD-L, programmed death-1 ligand; pSLC, plasmid encoding SLC; pPD-1A, plasmid encoding extracellular domain of PD-1; HSP70, heat shock protein 70; pPD-1B, plasmid encoding extracellular domain of PD-1 and GFP; sPD-1, extracellular domain of PD-1; sPD-1-GFP, expression product of pPD-1B.

Copyright © 2004 by The American Association of Immunologists, Inc.

transduced with a cDNA encoding mouse SLC showed reduced tumorigenicity in immunocompetent or nude mice (17). However, chemokines alone seem to show limited antitumor efficacy (2). Ag-induced activation and proliferation of antitumor T cells are regulated by both positive and negative costimulatory molecules. The interactions of negative costimulatory molecules with their receptors, including CTLA-4 and programmed death-1 (PD-1), function to limit, terminate, and/or attenuate T cell responses, and appear to be particularly important for regulating T cell tolerance (18). CTLA-4 shares the B7 ligands with CD28, and plays a critical role in down-regulation of T cell responses (19). PD-1, in contrast, interacts with distinct ligands, PD-L1 (also B7-H1) and PD-L2 (also B7-DC), which are expressed on not only APCs but also other cell types in nonlymphoid tissues such as heart, lung, liver, kidney, and skeletal muscle (20 –23). It is speculated that PD-Ls expressed in these tissues may function in part as “veto” molecules against the potentially autoreactive effector T cells (24, 25). Furthermore, PD-Ls are expressed on DCs (26, 27) and tumors including carcinomas of lung, ovary and colon, and melanomas (28), also squamous cell carcinomas of the head and neck (29), and glioma (30) and are involved in tumor immune evasion. It was reported that CTLA-4 blockade in vivo led to suppression or even rejection of a number of potentially lethal transplanted tumors (19). Growth of the myeloma cells in normal syngeneic mice was inhibited significantly albeit transiently by injection i.p. of anti-PD-L1 Ab (25). These results indicate that weak immune responses against the tumors can be converted into more potent ones by inhibiting the effects of negative immunoreceptors. Because the antitumor effects of SLC mainly depend on its recruitment of T lymphocytes and DCs, and PD-Ls-PD-1 interactions are involved in the mechanisms of tumor immune evasion, we supposed that blockade of the PD-Ls-PD-1 interactions could 0022-1767/04/$02.00

4920

BLOCKING PD-Ls/PD-1 ENHANCES ANTITUMOR EFFECT OF SLC

enhance the antitumor effects by augmentation of the immune responses against tumors when more immune cells were chemoattracted to tumor tissues by SLC. In this study, we constructed two recombinant eukaryotic expression plasmids (pSLC and pPD-1A) encoding murine SLC and the extracellular domain of murine PD-1 (sPD-1), respectively. The expressed sPD-1 could block the PD-Ls-PD-1 interactions, and local gene transfer of sPD-1 in tumor inoculation sites significantly improved the therapeutic effect of pSLC.

Materials and Methods Mice Female BALB/c (H-2d) mice, 6 to 8 wk old, were purchased from Center of Medical Experimental Animals of Hubei Province (Wuhan, China) and maintained in our facilities under standard conditions. All studies involving mice were approved by the institute’s Animal Care and Use Committee.

Cells and Abs The BHK cell line (baby hamster kidney cells) and the H22 cell line (murine hepatoma) were obtained from the China Type Culture Collection (Wuhan, China) and cultured in DMEM supplemented with 10% FCS, 2 mM L-glutamine, 100 U/ml streptomycin, and 100 mg/ml penicillin (complete medium). All cell culture reagents were obtained from Invitrogen Life Technologies (Grand Island, NY). The H22 cell line was also passaged in vivo with ascitic fluid. Spleen cells from BALB/c mice were treated with hypotonic buffer at room temperature for 2 min to remove erythrocytes and washed three times with the complete RPMI 1640 medium. For stimulation, the resultant cells were cultured at the concentration of 2 ⫻ 106 cells/ml in the presence of 2 ␮g/ml Con A (Sigma-Aldrich, St. Louis, MO) in a 24-well culture plate for 2 days (31). For chemotaxis assay, T cells from spleens were enriched by plastic adherence and passage through nylon wool columns. The cells were cultured at 37°C in a humidified incubator with 5% CO2. Rat anti-mouse SLC mAb (clone 59106) and goat anti-mouse PD-1 polyclonal Ab were obtained from R&D Systems (Minneapolis, MN). Rat anti-mouse PD-L1 (MIH5) and PD-L2 (TY25) mAbs were obtained from eBioscience (San Diego, CA). Mouse anti-human CD3 (PC3/188A) mAb was purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Spleen cells stimulated by the HSP70-peptide complex The heat shock protein 70 (HSP70)-peptide complex was prepared as described previously (32). For preparation of the peptide mixture, H22 tumor cells from the peritoneal cavity of tumor-bearing mice were washed with PBS and resuspended in PBS to a final concentration of 5 ⫻ 107/ml. After two freeze-and-thaw cycles, the cell suspension was mixed with a 2-fold volume of distilled water and vortexed for 30 s followed by centrifugation at 14000 ⫻ g for 15 min. The supernatant was incubated in a boiling water bath for 10 min, then ice bath for 30 min and centrifugated. The pH value of the supernatant was adjusted to 3.0 with HCl followed by another centrifugation and the pH value of the supernatant was adjusted to 7.0 with NaOH solution. Recombinant human HSP70 was prepared from engineered bacteria carrying an expression plasmid pMSH70 of human HSP70 (kindly provided by Drs. R. Morimoto and J. Z. Cheng, University of Texas Medical Branch, Galveston, TX) and the purity reached 90% (SDS-PAGE identification). HSP70 and the peptide mixture were mixed to bind each other as described (33). Briefly, peptides and HSP70, at the concentrations of 75 and 250 ␮g/ml, respectively, were mixed and incubated at 37°C for 2 h in the presence of 1 mM ADP and 1 mM MgCl2. For Ag-specific stimulation, the splenocytes were cultured at the concentration of 1 ⫻ 107/ml in RPMI 1640 supplemented with 20 U/ml IL-2 (PeproTech, London, U.K.) in a 96-well culture plate in the presence of 0.75 ␮g/ml HSP70-peptide complex. The cells were passaged 48 h later and cultured for another 5 days, then used for cytotoxicity assay. Our HSP70-H22 peptide-stimulated polyclonal T cell populations encompassed approximately equal numbers of CD4⫹ and CD8⫹ T cells (32).

Construction of expression plasmid vectors All enzymes used for the construction of recombinant plasmids were products of Promega (Madison, WI). Total RNA was isolated from spleen cells using TRIzol reagent (Invitrogen Life Technologies) and reverse transcribed. The 857-bp cDNA of murine SLC (6Ckine-Ser) was produced by PCR using the primers sense 5⬘-GGCATCCTAAGCTTAACTTGCGGCTGTCCA-3⬘ and antisense 5⬘-GGGTACATGGATCCTTTGAGGCTGGATCAC-3⬘ and

inserted into compatible enzyme restriction sites of pcDNA3.1 (Invitrogen Life Technologies, Carlsbad, CA) to generate murine SLC eukaryotic expression plasmid (pSLC). To construct the expression plasmid pPD-1A, carrying a cDNA encoding the extracellular domain of murine PD-1 (sPD-1), specific primers were used as follows: sense 5⬘-GGTTCATAAAGCTTCTGAAGGCGACACTGCC-3⬘ and antisense 5⬘-CCTGGTGAATTCATTGAAACCGGCCTTCTGG-3⬘, containing a termination codon. The amplified fragment by RT-PCR was digested by restriction enzymes HindIII and EcoRI and inserted into compatible enzyme restriction sites of pcDNA3.1. Another plasmid, pPD-1B, carrying a cDNA encoding a fusion protein of sPD-1 (N-terminal) and a GFP fragment (C-terminal), was constructed using different primers except that the sense primer for PD-1 is the same as that used for pPD-1A. Another primer (antisense) used was 5⬘-CCTGGTGAATTCTTTGAAACCGGCCTTCTGG3⬘, containing an EcoRI restriction site. To amplify the GFP fragment from the pTracer plasmid (Invitrogen Life Technologies), specific primers were used: sense 5⬘-GGTTGAGAATTCTGGCTAGCAAAGGAGAAG-3⬘ and antisense 5⬘-GGCACTGGTTCTAGATCAATCCATGCCATGTGT-3⬘ containing a termination codon. The two amplified fragments were digested by restriction enzymes HindIII and EcoRI, EcoRI and XbaI, respectively, and recombinated with pcDNA3.1 digested with HindIII and XbaI. All cDNAs inserted into the constructed plasmids were sequenced and identical to the published sequences (4, 6, 34).

Cell transfection The transfection of cells with plasmid was performed in a 24-well culture plate with 60 – 80% confluence of BHK cells using Dosper liposomal transfection reagent (Boehringer Mannheim, Mannheim, Germany) following the manufacturer’s instructions. The ratio of liposome to plasmid was 4 ␮l:1 ␮g. After transfection of 48 h, cells transfected with pTracer and pPD-1B were observed and images were obtained using an inverted fluorescence microscope (Olympus IMT-2; Olympus Optical, Tokyo, Japan), and supernatants of cells transfected with pPD-1B or different doses of pSLC were used for FACS or chemotaxis analysis, respectively. For stable transfection with pPD-1A, the transfected BHK cells were selected on the basis of G418 resistance (800 ␮g/ml) for 3 wk.

Chemotaxis assays Mouse T lymphocytes obtained from spleen as described above were resuspended in RPMI 1640 medium containing 10% FCS, incubated for 1 h at 37°C, and subjected to a chemotaxis assay. For the assay, 106 cells in 100 ␮l of medium were added to the top chamber of a 6.5-mm diameter, 5-␮m pore polycarbonate Transwell insert (Costar, Cambridge, MA) and incubated with the supernatants of BHK cells nontransfected or transfected with pcDNA3.1 or different doses of pSLC (or plus mAb anti-mouse SLC, 2 ␮g/ml, as indicated) in the bottom chamber for 4 h. Cells migrating to the bottom chamber were counted with a FACSCalibur (BD Biosciences, San Jose, CA) for 50 s. The results were expressed as the fold increase in cells migrating in response to chemoattractant vs the medium control (chemotaxis index). All assays were done in triplicate. For chemotaxis assay in vivo, 100 ␮g of pcDNA3.1 plasmid, or 25, 50, 100, 200 ␮g of pSLC plasmid (in 100 ␮l of saline) were injected into the hind thigh muscle of four groups of BALB/c mice (three mice per group) twice with a 1 day interval. After 72 h of the second injection, the muscle tissue of injection site was analyzed for SLC protein expression by ELISA, or fixed by 4% paraformaldehyde and used for paraffin sections analyzed by immunohistochemistry.

Gene therapy and tumor surveillance in vivo Plasmid DNA used for gene therapy was prepared by the alkaline lysis method and purified by polyethylene glycol, followed by selective compaction with spermine (Sigma-Aldrich, St. Louis, MO) as described (35). LPS concentration in the plasmid DNA preparation was determined by the Limulus amebocyte lysate (LAL) assay and was ⬍1.5 endotoxin units/␮g. All plasmid preparations for i.m. injections were resuspended in sterile 0.9% saline. Spectrophotometric analysis revealed 260/280 nm ratios ⱖ1.80. Purity and conformation of the prepared DNA were confirmed by agarose gel electrophoresis. For experiments in vivo, BALB/c mice were inoculated with ascitic H22 hepatoma cells by the injection of 1 ⫻ 105 cells in 100 ␮l of PBS into the right hind thigh muscle. Two days after inoculation, the naked expression plasmids were injected, every other day for six times, into the inoculation site at the indicated doses in 100 ␮l of sterile saline. Mice of the control group received an equal volume of saline or an equal amount of pcDNA3.1 plasmid. Tumor size was measured using calipers fitted with a Vernier scale when tumor could be palpated. The tumor diameter was calculated using the formula (a ⫹ b)/2, with a as the larger diameter and b as the

The Journal of Immunology smaller diameter. Mice were sacrificed and tumors were dissected and weighed on the indicated days after inoculation or for ethical reasons when animals showed severe distress or when tumors exceeded 1.5 cm in diameter. The average inhibition rate percentage ⫽ ((average tumor weight of pcDNA3.1 plasmid control group ⫺ average tumor weight of treated group)/average tumor weight of pcDNA3.1 plasmid control group) ⫻ 100%. The complete inhibition rate percentage ⫽ (mice without palpable tumor on indicated day/all treated mice in a group) ⫻ 100%.

Detection of protein expression in vivo After plasmid injections, equal amounts of muscle tissues of injection site (6 – 8 mm diameter) were excised at the indicated time, minced, and cultured in a 12-well dish with 1 ml of medium (DMEM with 10% FCS and 2% chick embryo extract). After incubation in a CO2 incubator for 48 h, the supernatants were subjected to specific protein analysis. SLC protein was detected using an SLC-specific ELISA (R&D Systems) following the manufacturer’s guidelines. To detect sPD-1 protein expression, Western blot was used. Briefly, 48-h culture supernatants were concentrated using the Freeze Dry System (Labconco, Kansas, MO) and then separated by SDSPAGE followed by transfer onto nitrocellulose membranes (Millipore, Bedford, MA). Membranes were blocked in PBS containing 5% milk and 0.1% Tween 20 and probed with the anti-mouse PD-1 polyclonal Ab at a concentration of 0.2 ␮g/ml. After incubation with the secondary anti-goat Ab conjugated with HRP (Amersham Pharmacia Biotech, Piscataway, NJ), membranes were extensively washed and the reaction was visualized by ECL according to the manufacturer’s protocol (Amersham Pharmacia Biotech).

4921 Cytotoxicity assay Standard 51Cr-release assays were performed. Briefly, target cells, either untreated, or treated with mAbs anti-PD-L1 and anti-PD-L2 (each 3 ␮g/ml) and/or the supernatants of BHK cells transfected with pPD-1A for 1 h, were labeled with Na51CrO4 (0.1 ␮Ci/106 cells; Amersham Pharmacia Biotech) at 37°C for 1 h. After extensive washing, target cells were incubated with effectors at different E:T ratios in triplicate for 4 h at 37°C, and 51Cr release (cpm) into the supernatants was measured in a gamma counter to calculate percentage-specific release. The percent-specific lysis was calculated by the formula: percentage lysis ⫽ 100 ⫻ (experimental cpm ⫺ spontaneous release)/(maximum release ⫺ spontaneous release).

Histology and immunohistochemistry Mouse muscle tissues of injection sites from treated mice were surgically excised, fixed for 12–24 h in 4% formamint, embedded in paraffin, and sectioned. Sections were stained with H&E. For indirect immunostaining, sections were incubated overnight at 4°C with a mouse anti-CD3 mAb diluted at 1/100. Biotinylated Abs to mouse IgG were used as secondary Abs, followed by peroxidase-conjugated streptavidin in the third step.

Data analysis Data were expressed as mean ⫾ SEM and results were interpreted using one-way ANOVA (followed by Newman-Keuls test). Differences were considered to be statistically significant when p ⬍ 0.05.

Results

Analysis of gene expression by RT-PCR and real-time PCR

Dose-chemotaxis relationship of pSLC transfection in vitro

Total RNAs were isolated using TRIzol reagent (Invitrogen Life Technologies) from H22 cells or muscle tissues of normal mice, or half part tumor tissues together with half part vicinal muscle tissues of tumor-bearing mice after treatment with different doses of pSLC. After being treated with DNase I (Promega), equal amounts of RNAs were reverse-transcribed and the cDNAs were amplified by the PCR with gene-specific primers at 94°C for denaturing, 54°C for annealing, and 72°C for extension. ␤-Actin mRNA was used as an internal control and coamplified. Previous experiments determined 26 cycles to be optimal. Specific primers used were: PD-L1, sense 5⬘-GTGAAACCCT GAGTCTTATCC-3⬘, antisense 5⬘-GACCATTCTGAGACAATTCC-3⬘; PDL2, sense 5⬘-CATCGCTTTGATCTTCCTGG-3⬘, antisense 5⬘-CCTGAA AGTCATTAGGAGCC-3⬘; ␤-actin, sense 5⬘-ATGGGTCAGAAGGAC TCCTATG-3⬘, antisense 5⬘-ATCTCCTGCTCGAAGTCTAGAG-3⬘. The amplified products were analyzed by electrophoresis on 1.5% agarose gels and stained with ethidium bromide. Relative quantitative real-time PCR was performed using the TaqMan probe on the ABI Prism 7700 (Applied Biosystems, Foster City, CA) to confirm the results obtained using RT-PCR above, following the manufacturer’s instructions. The primer and probe sequences for ␤-actin were available commercially and others were as follows: PD-L1, sense 5⬘-GG AATTGTCTCAGAATGGTC-3⬘, antisense 5⬘-GTAGTTGCTTCTAGGAA GGAG-3⬘, and probe 5⬘-CACCAAACCAGCTCTATTCCCTCAGCCTAT3⬘; PD-L2: sense 5⬘-AAGACTGACAATCTTCCCTC-3⬘, antisense 5⬘CCTGAAAGTCATTAGGAGCC-3⬘, and probe 5⬘-TCAAGACTCT CTGAACAGCAAGACCCCAAT-3⬘. To detect sPD-1 expression in mouse muscle after the pPD-1A plasmid injections, the primer and probe sequences used were: sense 5⬘-CAGGTACCCTGGTCATTCACT-3⬘, antisense 5⬘CCATTGGGGACCTCTAGAAG-3⬘, and probe 5⬘-TGTGCTGCAGTT GAGCTGGCAATCA-3⬘. cDNA samples (100 ng each, except for ␤-actin (10 ng)) reverse-transcribed from tissue RNA were mixed with primers and TaqMan Universal PCR Master Mix in a total volume of 50 ␮l. The PCR was conducted using the following parameters: 50°C for 2 min, 95°C for 10 min, and 40 cycles at 95°C for 15 s and 60°C for 1 min. Each gene was normalized to the housekeeping gene (␤-actin) from the same sample before fold change was calculated (using crossing point values) to account for variations between different samples. The mRNA level in normal muscle tissue was used as the calibrator. All PCR assays were performed in duplicate and results are represented by the mean values.

Although the sequence of SLC cDNA in pSLC was identical to that published, whether SLC expressed by pSLC was functional was unknown. To this end, BHK cells were transfected at different doses and the culture supernatants were harvested 48 h later. Chemotactic activity for spleen T cells was assessed by quantifying T cell migration across 5-␮m pore polycarbonate membranes in a Transwell cell culture chamber assay to measure the response to chemotactic stimuli placed into the lower chamber. As shown in Fig. 1, supernatants from pSLC-transfected cells attracted spleen T cells above control levels. Such an activity was increased when cells were transfected at higher doses but decreased obviously at 0.75 ␮g. The chemotactic activity was abrogated by the addition of a neutralizing anti-SLC mAb and it was not present in the supernatants of pcDNA3.1-transfected BHK cells. Dose-effect relationship of pSLC plasmid on inhibition of mouse H22 hepatoma in vivo To explore whether the transfection with pSLC had a therapeutic effect against tumors, we first observed whether the product expressed by pSLC in vivo had chemotactic activity. Different doses

FACS analysis H22 cells or spleen cells stimulated with Con A for 48 h were washed with PBS and incubated with supernatants of BHK cells transfected with pTracer (control) or pPD-1B at 37°C for 1 h. In blocking experiment, PD-Ls on spleen cells or H22 cells were blocked using mAbs anti-PD-L1 and anti-PD-L2 (each 3 ␮g/ml) before incubation with the supernatants. The cells were then washed twice and parameters were acquired on a FACSCalibur flow cytometer (BD Biosciences) and analyzed using CellQuest software (BD Biosciences).

FIGURE 1. Chemotactic attraction of spleen T lymphocytes by pSLCencoded SLC. Chemotactic activity of culture supernatants of BHK cells transfected with pcDNA3.1, pSLC at dose of 0.125, 0.25, 0.5 (with or without neutralizing Ab, 2 ␮g/ml), 0.75 ␮g, measured as the fold increase in spleen T cells migrating in response to chemoattractant vs the medium control (chemotaxis index). Cells migrating to the bottom chamber were counted with a FACSCalibur for 50 s. ⴱⴱ, Significant differences at p ⬍ 0.01 according to one-way ANOVA.

4922


of pSLC were injected into the hind thigh muscle, and the muscle tissue of the injection site was analyzed by immunohistochemistry. As shown in Fig. 2, A and B, in the detected range, along with the increase of plasmid dosage, SLC protein expression levels were increased and chemotactic response on T lymphocytes was enhanced correspondingly. This was different from the results ob-

tained in vitro that the chemotactic response was decreased at a high dose (Fig. 1). Because the transfection with a high dosage of pSLC could result in the enhanced chemotactic response of immune cells, we reasoned that the local transfection with pSLC would have an antitumor effect and the increased dosage of pSLC would result in

FIGURE 2. Dose-effect relationship of pSLC plasmid on mouse H22 hepatoma in vivo. A and B, Dose-chemotaxis relationship of pSLC transfection in vivo. BALB/c mice received 100 ␮g of pcDNA3.1 plasmid, or 25, 50, 100, 200 ␮g of pSLC plasmid (a, b, c, d, and e, respectively) by injection into the hind thigh muscle twice. A, SLC protein expression levels were determined by ELISA as described in Materials and Methods. B, Immunohistochemical staining was performed with anti-CD3 mAb of paraffin sections of muscle tissue from the injection site surgically removed 72 h after the second injection. Images are representative of multiple microscopic fields observed in three mice per group. C, Mice inoculated with 1 ⫻ 105 H22 ascitic hepatoma cells in the right hind thigh muscle were treated, starting from day 3 after inoculation, with i.m. injection of 25, 50, 100, 200 ␮g of naked pSLC or 100 ␮g of pcDNA3.1 or saline six times. Individual tumor weight on day 20 after tumor inoculation was measured and the mean tumor weight was significantly decreased along with the increase of pSLC dosage (left, p ⬍ 0.01). Two independent experiments represented by open or filled circles have been pooled. Also, the presence of palpable tumor was checked three times a week (right).

The Journal of Immunology augmentation of antitumor efficacy. To study this issue, BALB/c mice were inoculated with H22 ascitic cells by injection of cells into the right hind thigh muscle. We inoculated animals with ascitic tumor cells because such cells had adapted to the environment in vivo. Mice received pSLC at different doses every other day for six times and were sacrificed 20 days after inoculation. As shown in Fig. 2C, transfection with pSLC induced significant inhibition of malignant tumors compared with saline- or pcDNA3.1-treated control mice and showed a dose-effect relationship. The tumor inhibition efficacy was enhanced along with the increase of pSLC dosage. Injection of 200 ␮g of pSLC reached the best tumor inhibition rate, 92.2%. Transfection with a lower dosage of pSLC reached a higher complete inhibition rate of tumors Transfection with 25 ␮g of pSLC achieved a minor average inhibition rate of tumors (55.4%), and also a minor complete inhibition rate (21.4%). When pSLC was administered at 50 ␮g, both the average and complete inhibition rates of tumors were increased (70.6 and 50%, respectively, Fig. 2C). The complete inhibition rate could be increased at a higher dosage of pSLC according to this tendency. Unexpectedly, however, it was decreased at the dosage of 100 ␮g (35.7%) or even lower at 200 ␮g (21.4%), the same as that at the dosage of 25 ␮g (Fig. 2C, right), suggesting that although more immune cells chemoattracted into tumor sites could result in stronger inhibition of tumor growth, the ability of immune cells to eliminate the tumor cells was declined. We supposed that the mechanisms of tumor escape from immune attack had been developed. Up-regulated expression of PD-L1 and PD-L2 genes in residual pSLC-treated tumors The negative costimulatory molecules such as PD-L1 play a critical role in tumor immune evasion (28, 30). To evaluate whether PD-Ls-PD-1 pathways were involved in the mechanism of tumor evasion in pSLC treatment, RT-PCR was performed to detect the expression level of PD-Ls genes in H22 cells and tumor tissues.

4923 Fig. 3A shows that PD-L1 but not PD-L2 mRNA was expressed in H22 hepatoma cells. We also detected PD-Ls mRNAs in muscle tissues of normal mice or mice bearing tumors after pSLC treatment. Both PD-L1 and PD-L2 mRNAs, as expected, were expressed in detected tissues of tumor-bearing mice to a higher level as compared with normal mice (Fig. 3A). The expression level was significantly increased along with the increase of the dosage of pSLC administered. The quantitative technique, real-time PCR, was used to confirm the results. Fig. 3B shows an analysis of these data normalized for the levels of a housekeeping gene (mouse ␤-actin). The data provided evidence that more PD-Ls mRNA was expressed in tumor and vicinal muscle tissues in the face of more immune cells chemoattracted to tumor sites, and H22 tumors could evade immune attack through PD-Ls-PD-1 pathways. We reasoned that blocking these pathways would enhance the immune response against tumors, and thus improve SLC-mediated antitumor effects. To this end, we constructed a plasmid vector, pPD-1A, carrying a cDNA encoding the extracellular domain of murine PD-1 (sPD-1). Expression products of plasmids carrying the cDNA coding for the extracellular domain of murine PD-1 could bind to PD-1 ligands In theory, sPD-1, the extracellular domain of PD-1, should be capable of binding to both PD-L1 and PD-L2, thereby blocking interactions of either PD-L1-PD-1 or PD-L2-PD-1. To verify whether sPD-1 could be secreted out from transfected cells, we constructed an expression plasmid pPD-1B, carrying a cDNA encoding the fusion protein sPD-1-GFP. BHK cells were transfected with either plasmid pPD-1B or pTracer under the same conditions and checked for expression 48 h later. As shown in Fig. 4A, sPD1-GFP produced by pPD-1B in BHK cells was mostly secreted out as compared with GFP produced by pTracer, which was constructed based on the similar vector of pPD-1B. The binding of sPD-1-GFP to PD-Ls on either H22 cells or spleen cells stimulated by Con A was determined by FACS analysis. Fig. 4B shows that fluorescence intensity on the cells incubated with sPD-1-GFP was significantly increased as compared with control. When H22 and

FIGURE 3. Expression of PD-L1 and PD-L2 genes in H22 cells and tumors. A, RT-PCR analysis as described in Materials and Methods of RNA isolated from the indicated cells or tissues. Lane 1, H22 cells; lane 2, normal muscle tissue; lanes 3–7, half part tumor tissues together with half part vicinal muscle tissues from tumor-bearing mice after treatment with 100 ␮g of pcDNA3.1, or 25, 50, 100, 200 ␮g of pSLC as in Fig. 2. B, Relative quantitative real-time PCR analysis of PD-L1 and PD-L2 transcription as described in Materials and Methods. Each RNA sample was analyzed in triplicate at two RNA concentrations. The expression level of both PD-L1 and PD-L2 transcripts was significantly increased if a higher dosage of pSLC was administered (p ⬍ 0.01). Data are representative of three individual experiments.

4924


FIGURE 4. sPD-1 expressed from pPD-1A could block PD-Ls-PD-1 interactions. pPD-1A, carrying cDNA encoding for sPD-1, and pPD-1B, carrying a cDNA encoding sPD-1-GFP were constructed as described in Materials and Methods. A, BHK cells were transfected with either pPD-1B or pTracer under the same conditions using the Dosper liposomal reagent and observed 48 h later. Images are representative of multiple microscopic fields observed in two individual experiments. B, H22 cells (upper) or spleen cells stimulated by Con A (lower) were incubated with the expression product of pPD-1B (48 h-tissue culture supernatants of BHK cells transfected with pPD-1B) and analyzed by FACS. Fluorescence intensity on the cells was significantly increased as compared with control. C, Cytotoxic activity displayed by spleen lymphocytes (control A) or HSP70-peptide complex-stimulated spleen lymphocytes against H22 cell lines (control B). In some experiments, H22 cells were incubated before the 4-h 51Cr release assay with the expression product of pPD-1A (48 h-tissue culture supernatants of BHK cells stably transfected with pPD-1A) or 3 ␮g/ml anti-PD-L1 and anti-PD-L2 mAbs. Data represent the specific lysis SEM at different E:T ratios from three independent experiments.

spleen cells were pretreated with mAbs anti-PD-L1 and antiPD-L2 before incubation with sPD-1-GFP, however, the fluorescence intensity decreased nearly to the control level (data not shown). These data indicated that sPD-1-GFP was specific and bound efficiently to PD-Ls. Blocking PD-Ls-PD-1 interactions mediated the enhanced lysis of H22 cells by HSP70-peptide complex-stimulated spleen cells To evaluate whether sPD-1 could block PD-Ls-PD-1 interactions, H22 cells, either untreated or pretreated with sPD-1, were incubated with HSP70-peptide complex-stimulated spleen cells at different E:T ratios. Fig. 4C shows that pretreatment with sPD-1 enhanced lysis of H22 cells, which was consistent with the result when H22 cells were pretreated with anti-PD-L1 and anti-PD-L2 mAbs. To determine whether the increased tumor lysis was mediated by blockade of PD-Ls-PD-1 interactions though sPD-1, a blocking assay was performed in which H22 cells were pretreated with sPD-1 plus anti-PD-L1 and anti-PD-L2 mAbs. Under such conditions, as expected, tumor lysis was not enhanced further (data not shown). These results further reinforced the hypothesis that sPD-1 was capable of binding to PD-Ls, thereby blocking PD-LsPD-1 pathways.

Local gene transfer of sPD-1 mediated antitumor effect To further evaluate whether sPD-1 has antitumor activity, 2 days after inoculation of H22 hepatoma cells, mice received 100 ␮g of pPD-1A by local injection into the inoculation site six times. Realtime PCR revealed that the sPD-1 mRNA expression levels became relatively stable from day 12 after inoculation. Western blot confirmed the sPD-1 protein expression (Fig. 5A). Injection with pPD-1A significantly inhibited tumor growth in 24 days. The tumor growth in 36.4% (4 of 11) of the recipients was completely inhibited (Fig. 5B), similar to the inhibitory effect of i.p.-injected anti-PD-L1 mAb on tumor (25). Meanwhile, all mice (8 of 8) in the control groups developed tumors. These findings, taken together with the above results from experiments in vitro, suggest that local expression of sPD-1 expression plasmid in vivo provides an effective approach for blocking PD-Ls-PD-1 interactions and thereby enhancing antitumor immune responses.

sPD-1 improved SLC-mediated antitumor immunity Local gene transfer of SLC and sPD-1 was conducted after tumor inoculation, and the individual diameter of the tumor nodules in 20

The Journal of Immunology

4925

FIGURE 5. Antitumor effect of injection of pPD-1A. BALB/c mice were inoculated i.m. with 1 ⫻ 105 H22 ascitic cells per mouse and treated, starting from day 2 after inoculation, by i.m. injection with pPD-1A (100 ␮g), pcDNA3.1 (100 ␮g), or saline as indicated. A, sPD-1 mRNA expression levels (left) and sPD-1 protein expression before and after pPD-1A injections (right) were detected by relative quantitative real-time PCR (three mice per point) or Western blot. B, Individual follow-up of local tumor growth was monitored.

days as well as the survival rates were investigated. For comparison with the pSLC plus pPD-1A group, pcDNA3.1 was used for supplementation of the amount of plasmid DNA in each of the pSLC and pPD-A1 groups. As shown in Fig. 6, treatment by injection with either pSLC or pPD-1A showed mild inhibition on the average tumor growth, and the tumor growth in 36.4% of the recipients (4 of 11) in both groups was completely suppressed. Treatment by injection with pSLC plus pPD-1A reached much better efficacy: the tumor growth in 57.1% of mice (8 of 14) was completely suppressed, and the rest of the mice showed prolonged survival as compared with those that received pSLC or pPD-1A alone (Fig. 6B). Meanwhile, 100% of the mice (7 of 7) in control groups that received saline or pcDNA3.1 showed progressive tu-

mor growth and died within 7 wk (Fig. 6B). These results proved a significant synergistic effect of local gene transfer of SLC in combination with sPD-1. Local combinatorial gene transfer of SLC and sPD-1 did not induce remarkable autoimmune manifestations Because blockade of CTLA-4 by i.v. administration of antiCTLA-4 Ab induced various autoimmune manifestations (36), and no reports are currently available for the autoimmune reaction induced by anti-PD-Ls Abs, the possible autoimmune manifestations induced by local combinatorial gene transfer of SLC and sPD-1 were investigated.

FIGURE 6. Treatment with pSLC in combination with pPD-1A. Mice inoculated with 1 ⫻ 105 H22 ascitic hepatoma cells in the right hind thigh muscle were treated, starting from day 3 after tumor cell inoculation, with i.m. injection of 100 ␮g of the indicated plasmids and 100 ␮g of pcDNA3.1 or 200 ␮g of pcDNA3.1 or saline. In the indicated group, mice were also injected with pSLC plus pPD-1A, 100 ␮g of each. A, Individual tumor diameter on day 20 after tumor inoculation. Statistical significance (p ⬍ 0.01) was found when comparing mice treated with pSLC plus pPD-1A with each of other groups. B, Long-term survival follow-up of these groups.

4926


FIGURE 7. Transfection with pPD1A did not induce remarkable autoimmune manifestations. Upper panels, H&E staining of muscle tissue sections from pPD-1A- and pSLC plus pPD-1A-treated mice. Lower panels, Immunohistochemical studies with mAb to CD3 showed predominance of CD3⫹ cells in the infiltrating cells of the tissues. Images are representative of multiple microscopic fields observed in at least four mice per group.

Because pPD-1A was locally used in muscle tissues that express PD-Ls, histopathologic changes of muscle tissues where pPD-1A was injected were examined. As shown in Fig. 7, H&E staining of muscle tissue sections from pPD-1A-treated mice showed infiltration of few lymphocytes into the interstitial region. Tissues from mice that received pSLC plus pPD-1A showed a prominent infiltration of lymphocytes, which may be chemoattracted by SLC. Immunohistochemical studies revealed that the majority of cells in these tissues were CD3⫹ cells. However, all of the examined tissues did not show muscle fiber degeneration, necrosis, as well as atrophy. These results indicated that no remarkable autoimmune inflammation in local tissues was developed during immunotherapy using pPD-1A alone or pSLC plus pPD-1A. Serial liver pathological sections of treated mice also did not show inflammation changes (data not shown), suggesting that sPD-1 expressed by pPD-1A played a role mainly in sites of gene transfer, and no marked autoimmune responses occurred in liver although sPD-1 mediated the enhanced immune attack to liver cancer cells. In addition, serum levels of creatine kinase, lactate dehydrogenase, glutamic oxaloacetic transaminase, and glutamic pyruvic transaminase were not significantly increased in mice that received pPD-1A or pSLC plus pPD-1A (data not shown). Taken together, these results indicate that local combinatorial gene transfer of SLC and sPD-1 did not induce remarkable autoimmune manifestations.

Discussion One hallmark of cancers is that they could not be eliminated by immune system and continue to grow. Cancer progression has been attributed to a variety of immune evasion strategies. These include down-regulation of cell surface MHC class I molecules, secretion of immunosuppressive factors (for example, TGF-␤), lack of T cell costimulation, and, the programmed cell death (apoptosis) of T cells when encountering tumor (37, 38), for example, through PD-L1-PD-1 interaction (28). In this study, we found that

the negative signal provided by PD-Ls-PD-1 interactions was also involved in the immune evasion in tumor immunotherapy. In theory, any chemokine capable of inducing the migration of T, NK cells, DCs, and/or macrophages could promote the regression or even eradication of a tumor mass by boosting the immune response against the tumor (39). Various approaches have been used to deliver chemokines in vivo including chemokine-transfected tumors, DCs or stromal cells, protein injection, and adenoviral vectors (2, 39). When we used naked plasmid to deliver SLC into tumor inoculation sites, similar antitumor effects were obtained as compared with other methods used. However, plasmids have different advantages with regard to efficiency of gene transfer, ease of production, safety, and sustained expression (40). SLC might mediate antitumor activity through at least two possibly related mechanisms: 1) the recruitment of immune cells such as T lymphocytes and DCs into local tissue, and 2) inhibition of tumor angiogenesis. In our study, SLC mediated significant inhibition of malignant tumors and when it was locally expressed in vivo by the injection of pSLC below the dose of 200 ␮g, the tumor inhibition efficacy was enhanced along with the increase of pSLC dosage. However, our results could not reach the conclusion that a better therapeutic effect would be obtained if a higher dosage of pSLC was used, because chemotaxis depends on the concentration gradient of chemokine and a high concentration of chemokine would lead to significantly decreased chemotactic response and toxic side effects (2). This could be supported by the experiment that transgenic overexpression of SLC disrupted T cell migration in mice (41). In contrast, although more immune cells chemoattracted to tumor sites by gene transfer of a high dosage of pSLC could result in higher average tumor inhibition rate, the complete inhibition rate was decreased as compared with that when a lower dosage of pSLC was used for local gene transfer, suggesting that immune

The Journal of Immunology evasion of tumors was developed and would not be attenuated in the face of increased immune cells. Then, why do tumors continue to grow in the face of seemingly potent cellular antitumor reactions? We suppose that the mechanisms that account for this can come from 1) the tumor itself, and 2) immune responses that occur during treatment, which may lead to more potent immune evasion regardless of increased immune cells. Altogether, this indicates that the tumor microenvironment favors the recruitment and survival of leukocyte subsets that induce anergy or tolerance toward tumor Ags (2). Our results showed that PD-L1 was also expressed in mouse H22 hepatoma cells. More importantly, the increased expression of PD-Ls was observed in pSLC-treated residual tumors and vicinal muscle tissues, and the expression level was significantly increased if the higher dosage of pSLC was administered (Fig. 3). These results suggest that immune tolerance through PD-Ls-PD-1 pathways may serve as a potent mechanism for residual pSLC-treated tumors to escape from seemingly enhanced immune responses. This is supported by the results that the local gene transfer of SLC in combination with sPD-1, a product that could block PD-LsPD-1 interactions, induced marked antitumor synergy. Although we did not observe that SLC directly induced PD-L1 or PD-L2 expression in H22 hepatoma cells in vitro (data not shown), some reasons may account for the increased expression of PD-Ls in vivo after the local gene transfer of SLC: 1) IFN-␥, which can up-regulate expression of PD-L1 on tumor cells (28, 30), as well as PD-L1 and PD-L2 on endothelial cells (42), resident macrophages (43), and DCs (44). IFN-␥ can be released from many different types of immune cells including some cells that can be chemoattracted to SLC. 2) T lymphocytes and DCs that are chemoattracted to SLC express PD-Ls. The recruitment of more such cells into tissues would result in the increase of a total level of PD-Ls expression in the microenvironment. 3) The continuous growth of tumor by evasion from immune attack could result in the increase of an absolute number of tumor cells expressing PD-Ls so that a vicious cycle occurred. Because PD-Ls are expressed widely in tumors and involved in peripheral tolerance, the PD-Ls-PD-1 pathways may be involved in immune evasion in most, if not all, tumors, especially in residual tumors after immunotherapy. We have found that the expression of PD-Ls was also up-regulated in tumors treated with the HSP70peptide complex, and treatment by the HSP70-peptide complex in combination with sPD-1 also mediated marked antitumor synergy (X.-H. Wang, G.-M. Zhang, Y.-F. He, and Z.-H. Feng, unpublished observations). Therefore, it is important to consider the possibility that tumors evade from immune attack in immunotherapy through PD-Ls, which are expressed in tumor cells as well as other cells such as DCs, because PD-Ls themselves are expressed in most tumor cells and could be up-regulated in tumor cells and immune cells by such molecules as IFN-␥ during immunotherapy. IFN-␥ is a cytokine involved in the antitumor mechanisms of many other cytokines used for immunotherapy, such as IL-12 (45) and SLC (14). Taken together, immune evasion involving PD-Ls-PD-1 pathways may be the “side effect” of many cytokines used in tumor immunotherapy, and the use of such cytokines in combination with the blockade of PD-Ls-PD-1 pathways may improve the antitumor effect. Our results showed that although the combinational use of sPD-1 and SLC produced marked antitumor synergy, there were yet a portion of tumors that could not be suppressed completely, suggesting that the blockade of PD-Ls-PD-1 interactions alone could not completely brake the tolerance to such tumors. In fact, there are other negative costimulatory receptors on T cells except for PD-1, such as CTLA-4 and B and T lymphocyte attenuator, a

4927 newly discovered inhibitory receptor with similarities to CTLA-4 and PD-1 (46). Whether these receptors are involved in immune evasion of tumors treated with pSLC plus pPD-1A, and whether blocking these inhibitory pathways could result in stronger antitumor immunity or even complete eradication of the tumors are two very interesting issues and worth further study. Abrogation of inhibitory costimulatory activity, by administration of mAbs and soluble receptors (often called immunoadhesins) that neutralize costimulatory molecules, can result in the enhanced cellular and humoral immune responses. However, an important issue to consider is whether such blockade could induce the side effect of autoimmune reaction. This concern is heightened by experiments showing that CTLA-4 blockade allowed the breaking of peripheral T cell tolerance to a normal tissue-specific differentiation Ag (19) and a clinical trial showing that blockade of CTLA-4 induced grade III/IV autoimmune manifestations in 43% of patients (36). In contrast, PD-1, whose deficiency in mice causes autoimmune diseases including autoimmune-dilated cardiomyopathy, is considered to play a crucial role in peripheral tolerance through negative immunoregulatory signals delivered by PD-LsPD-1 pathways (18, 24, 47). Fortunately, however, we did not find remarkable autoimmune manifestations in sites of local gene transfer as well as liver when we treated H22 hepatoma models by local gene transfer of sPD-1. This may be explained by the possibility that immune cells chemoattracted by SLC induced immune response mainly to tumor Ags, and sPD-1 was expressed mainly in transfection sites and thus blocked the PD-Ls-PD-1 pathways mainly concerning antitumor T cells. This is different from the methods using mAbs and immunoadhesins that may play a broad role. Together, although we did not detect histopathologic changes of other tissues except for liver and muscle, and the issue of whether sPD-1 used locally with a higher dose of expression plasmid could induce marked autoimmune manifestations needs to be further studied, the blockade of inhibitory costimulatory molecules in tumor sites is a worthwhile therapeutic method in tumor immunotherapy. To conclude, the results of this study indicate that the expression of SLC and blockade of PD-Ls-PD-1 pathways by local gene transfer could result in marked antitumor synergy. The combination of chemokines with other immunostimulatory cytokines provides enhanced and long-term antitumor immunity and represents a way to reduce toxic side effects (2). Therefore, our findings provide a potent method of improving antitumor effects of SLC and possibly other immunotherapeutic methods by local blockade of negative costimulatory molecules.

Acknowledgments We thank Hui-Fen Zhu for FACS analysis, Mu-Lan Yang for immunohistochemistry analysis, and Prof. Guan-Xin Shen for his help and suggestion.

References 1. Huang, A. Y. C., P. Golumbek, M. Ahmadzadeh, E. Jaffee, D. Pardoll, and H. Levitsky. 1994. Role of bone marrow-derived cells in presenting MHC class I-restricted tumor antigens. Science 264:961. 2. Homey, B., A. Muller, and A. Zlotnik. 2002. Chemokines: agents for the immunotherapy of cancer? Nat. Rev. Immunol. 2:175. 3. Nagira, M., T. Imai, K. Hieshima, J. Kusuda, M. Ridanpaa, S. Takagi, M. Nishimura, M. Kakizaki, H. Nomiyama, and O. Yoshie. 1997. Molecular cloning of a novel human CC chemokine secondary lymphoid-tissue chemokine that is a potent chemoattractant for lymphocytes and mapped to chromosome 9p13. J. Biol. Chem. 272:19518. 4. Hedrick, J. A., and A. Zlotnik. 1997. Identification and characterization of a novel ␤ chemokine containing six conserved cysteines. J. Immunol. 159:1589. 5. Hromas, R., C. Kim, M. Klemsz, M. Krathwohl, K. Fife, S. Cooper, C. Schnizlein-Bick, and H. E. Broxmeyer. 1997. Isolation and characterization of exodus-2, a novel C-C chemokine with a unique 37-amino acid carboxyl-terminal extension. J. Immunol. 159:2554. 6. Tanabe, S., Z. Lu, Y. Luo, E. J. Quackenbush, M. A. Berman, L. A. Collins-Racie, S. Mi, C. Reilly, D. Lo, K. A. Jacobs, and M. E. Dorf. 1997.

4928

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18. 19. 20.

21.

22.

23.

24. 25.

26.


Identification of a new mouse ␤-chemokine, thymus-derived chemotactic agent 4, with activity on T lymphocytes and mesangial cells. J. Immunol. 159:5671. Lo, J. C., R. K. Chin, Y. Lee, H. S. Kang, Y. Wang, J. V. Weinstock, T. Banks, C. F. Ware, G. Franzoso, and Y. X. Fu. 2003. Differential regulation of CCL21 in lymphoid/nonlymphoid tissues for effectively attracting T cells to peripheral tissues. J. Clin. Invest. 112:1495. Nagira, M., T. Imai, R. Yoshida, S. Takagi, M. Iwasaki, M. Baba, Y. Tabira, J. Akagi, H. Nomiyama, and O. Yoshie. 1998. A lymphocyte-specific CC chemokine, secondary lymphoid tissue chemokine (SLC), is a highly efficient chemoattractant for B cells and activated T cells. Eur. J. Immunol. 28:1516. Willimann, K., D. F. Legler, M. Loetscher, R. S. Roos, M. B. Delgado, I. Clark-Lewis, M. Baggiolini, and B. Moser. 1998. The chemokine SLC is expressed in T cell areas of lymph nodes and mucosal lymphoid tissues and attracts activated T cells via CCR7. Eur. J. Immunol. 28:2025. Gunn, M. D., K. Tangemann, C. Tam, J. G. Cyster, S. D. Rosen, and L. T. Williams. 1998. A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc. Natl. Acad. Sci. USA 95:258. Chan, V. W., S. Kothakota, M. C. Rohan, L. Panganiban-Lustan, J. P. Gardner, M. S. Wachowicz, J. A. Winter, and L.T. Williams. 1999. Secondary lymphoidtissue chemokine (SLC) is chemotactic for mature dendritic cells. Blood 93:3610. Kellermann, S. A., S. Hudak, E. R. Oldham, Y. J. Liu, and L. M. McEvoy. 1999. The CC chemokine receptor-7 ligands 6Ckine and macrophage inflammatory protein-3␤ are potent chemoattractants for in vitro- and in vivo-derived dendritic cells. J. Immunol. 162:3859. Saeki, H., A. M. Moore, M. J. Brown, and S. T. Hwang. 1999. Secondary lymphoid-tissue chemokine (SLC) and CC chemokine receptor 7 (CCR7) participate in the emigration pathway of mature dendritic cells from the skin to regional lymph nodes. J. Immunol. 162:2472. Sharma, S., M. Stolina, J. Luo, R. M. Strieter, M. Burdick, L. X. Zhu, R. K. Batra, and S. M. Dubinett. 2000. Secondary lymphoid tissue chemokine mediates T cell-dependent antitumor responses in vivo. J. Immunol. 164:4558. Sharma, S, M. Stolina, L. Zhu, Y. Lin, R. Batra, M. Huang, R. Strieter, and S. M. Dubinett. 2001. Secondary lymphoid organ chemokine reduces pulmonary tumor burden in spontaneous murine bronchoalveolar cell carcinoma. Cancer Res. 61:6406. Kirk, C. J., D. Hartigan-O’Connor, B. J. Nickoloff, J. S. Chamberlain, M. Giedlin, L. Aukerman, and J. J. Mule. 2001. T cell-dependent antitumor immunity mediated by secondary lymphoid tissue chemokine: augmentation of dendritic cellbased immunotherapy. Cancer Res. 61:2062. Vicari, A. P., S. Ait-Yahia, K. Chemin, A. Mueller, A. Zlotnik, and C. Caux. 2000. Antitumor effects of the mouse chemokine 6Ckine/SLC through angiostatic and immunological mechanisms. J. Immunol. 165:1992. Greenwald, R. J., Y. E. Latchman, and A. H. Sharpe. 2002. Negative co-receptors on lymphocytes. Curr. Opin. Immunol. 14:391. Egen, J. G., M. S. Kuhns, and J. P. Allison. 2002. CTLA-4: new insights into its biological function and use in tumor immunotherapy. Nat. Immunol. 3:611. Dong, H., G. Zhu, K. Tamada, and L. Chen. 1999. B7–H1, a third member of the B7 family, co-stimulates T-cell proliferation and interleukin-10 secretion. Nat. Med. 5:1365. Freeman, G. J., A. J. Long, Y. Iwai, K. Bourque, T. Chernova, H. Nishimura, L. J. Fitz, N. Malenkovich, T. Okazaki, M. C. Byrne, et al. 2000. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J. Exp. Med. 192:1027. Tamura, H., H. Dong, G. Zhu, G. L. Sica, D. B. Flies, K. Tamada, and L. Chen. 2001. B7–H1 costimulation preferentially enhances CD28-independent T-helper cell function. Blood 97:1809. Latchman, Y., C. R. Wood, T. Chernova, D. Chaudhary, M. Borde, I. Chernova, Y. Iwai, A. J. Long, J. A. Brown, R. Nunes, et al. 2001. PD-L2 is a second ligand for PD-1 and inhibits T cell activation. Nat. Immunol. 2:261. Nishimura, H., and T. Honjo. 2001. PD-1: an inhibitory immunoreceptor involved in peripheral tolerance. Trends Immunol. 22:265. Iwai, Y., M. Ishida, Y. Tanaka, T. Okazaki, T. Honjo, and N. Minato. 2002. Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade. Proc. Natl. Acad. Sci. USA 99:12293. Curiel, T. J., S. Wei, H. Dong, X. Alvarez, P. Cheng, P. Mottram, R. Krzysiek, K. L. Knutson, B. Daniel, M. C. Zimmermann, et al. 2003. Blockade of B7–H1 improves myeloid dendritic cell-mediated antitumor immunity. Nat. Med. 9:562.

27. Brown, J. A., D. M. Dorfman, F. R. Ma, E. L. Sullivan, O. Munoz, C. R. Wood, E. A. Greenfield, and G. J. Freeman. 2003. Blockade of programmed death-1 ligands on dendritic cells enhances T cell activation and cytokine production. J. Immunol. 170:1257. 28. Dong, H., S. E. Strome, D. R. Salomao, H. Tamura, F. Hirano, D. B. Flies, P. C. Roche, J. Lu, G. Zhu, K. Tamada, et al. 2002. Tumor-associated B7–H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat. Med. 8:793. 29. Strome, S. E., H. Dong, H. Tamura, S. G. Voss, D. B. Flies, K. Tamada, D. Salomao, J. Cheville, F. Hirano, W. Lin, et al. 2003. B7–H1 blockade augments adoptive T-cell immunotherapy for squamous cell carcinoma. Cancer Res. 63:6501. 30. Wintterle, S., B. Schreiner, M. Mitsdoerffer, D. Schneider, L. Chen, R. Meyermann, M. Weller, and H. Wiendl. 2003. Expression of the B7-related molecule B7–H1 by glioma cells: a potential mechanism of immune paralysis. Cancer Res. 63:7462. 31. Ishida, M., Y. Iwai, Y. Tanaka, T. Okazaki, G. J. Freeman, N. Minato, and T. Honjo. 2002. Differential expression of PD-L1 and PD-L2, ligands for an inhibitory receptor PD-1, in the cells of lymphohematopoietic tissues. Immunol. Lett. 84:57. 32. Feng, Z. H., B. Huang, G. M. Zhang, D. Li, and H. T. Wang. 2002. Investigation on the effect of peptides mixture from tumor cells inducing anti-tumor specific immune response. Sci. China Ser. C 45:361. 33. Blachere, N. E., Z. H. Li, R. Y. Chandauarkar, R. Suto, N. S. Jaikaria, S. Basu, H. Udono, and P. K. Srivastava. 1997. Heat shock protein-peptide complexes reconstituted in vitro, elicit peptide-specific cytotoxic T lymphocyte response and tumor immunity. J. Exp. Med. 186:1315. 34. Ishida, Y., Y. Agata, K. Shibahara, and T. Honjo. 1992. Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death. EMBO J. 11:3887. 35. Mourich, D. V., M. W. Munks, J. C. Murphy, R. C. Willson, and A. B. Hill. 2003. Spermine compaction is an efficient and economical method of producing vaccination-grade DNA. J. Immunol. Methods 274:257. 36. Phan, G. Q., J. C. Yang, R. M. Sherry, P. Hwu, S. L. Topalian, D. J. Schwartzentruber, N. P. Restifo, L. R. Haworth, C. A. Seipp, L. J. Freezer, et al. 2003. Cancer regression and autoimmunity induced by cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma. Proc. Natl. Acad. Sci. USA 100:8372. 37. Rosenberg, S. A. 2001. Progress in human tumour immunology and immunotherapy. Nature 411:380. 38. Smyth, M. J., D. I. Godfrey, and J. A. Trapani. 2001. A fresh look at tumor immunosurveillance and immunotherapy. Nat. Immunol. 2:293. 39. Rossi, D., and A. Zlotnik. 2000. The biology of chemokines and their receptors. Annu. Rev. Immunol. 18:217. 40. Herweijer, H., and J. A. Wolff. 2003. Progress and prospects: naked DNA gene transfer and therapy. Gene Ther. 10:453. 41. Christopherson, K. W., 2nd, J. J. Campbell, and R. A. Hromas. 2001. Transgenic overexpression of the CC chemokine CCL21 disrupts T-cell migration. Blood 98:3562. 42. Rodig, N., T. Ryan, J. A. Allen, H. Pang, N. Grabie, T. Chernova, E. A. Greenfield, S. C. Liang, A. H. Sharpe, A. H. Lichtman, and G. J. Freeman. 2003. Endothelial expression of PD-L1 and PD-L2 down-regulates CD8⫹ T cell activation and cytolysis. Eur. J. Immunol. 33:3117. 43. Loke, P., and J. P. Allison. 2003. PD-L1 and PD-L2 are differentially regulated by Th1 and Th2 cells. Proc. Natl. Acad. Sci. USA 100:5336. 44. Yamazaki, T., H. Akiba, H. Iwai, H. Matsuda, M. Aoki, Y. Tanno, T. Shin, H. Tsuchiya, D. M. Pardoll, K. Okumura, et al. 2002. Expression of programmed death 1 ligands by murine T cells and APC. J. Immunol. 169:5538. 45. Nastala, C. L., H. D. Edington, T. G. McKinney, H. Tahara, M. A. Nalesnik, M. J. Brunda, M. K. Gately, S. F. Wolf, R. D. Schreiber, W. J. Storkus, et al. 1994. Recombinant IL-12 administration induces tumor regression in association with IFN-␥ production. J. Immunol. 153:1697. 46. Watanabe, N., M. Gavrieli, J. R. Sedy, J. Yang, F. Fallarino, S. K. Loftin, M. A. Hurchla, N. Zimmerman, J. Sim, X. Zang, et al. 2003. BTLA is a lymphocyte inhibitory receptor with similarities to CTLA-4 and PD-1. Nat. Immunol. 4:670. 47. Nishimura, H., T. Okazaki, Y. Tanaka, K. Nakatani, M. Hara, A. Matsumori, S. Sasayama, A. Mizoguchi, H. Hiai, N. Minato, and T. Honjo. 2001. Autoimmune dilated cardiomyopathy in PD-1 receptor-deficient mice. Science 291:319.