Hepatitis B Virus Transgenic Mice Cytotoxic T Lymphocyte Tolerance ...

Dendritic Cell Immunization Breaks Cytotoxic T Lymphocyte Tolerance in Hepatitis B Virus Transgenic Mice1 Yukihiro Shimizu, Luca G. Guidotti, Patricia Fowler, and Francis V. Chisari2 Hepatitis B virus (HBV) transgenic mice that are immunologically tolerant to HBV-encoded Ags represent a model of chronic HBV infection suitable for the development of therapeutic immunization strategies before testing in humans. Five lineages of HBV transgenic mice were immunized with plasmid DNA that encodes hepatitis B surface Ag (HBsAg) or with cytokine-activated bone marrow-derived dendritic cells (DCs) in an attempt to break tolerance to HBsAg at the B and T cell levels. DNA immunization stimulated an Ab response but not a cytotoxic T lymphocyte response to HBsAg in two of the five transgenic lineages studied. In contrast, infusion of activated transgenic or nontransgenic DCs stimulated a splenic CTL response in all three transgenic lineages immunized in this manner at precursor frequencies comparable to those in nontransgenic mice, indicating that DC function is normal, and HBsAg-specific CTLs are present but functionally silent in these transgenic animals. Importantly, none of the animals developed hepatitis or displayed suppressed viral gene expression or replication following either DNA immunization or DC administration even in the presence of anti-hepatitis B surface (HBs) Abs and HBs-specific CTLs. These results indicate that Ag presentation by cytokine-activated DCs can break tolerance and trigger an anti-viral CTL response in HBV transgenic mice, and they suggest that this strategy is more efficient than DNA immunization in this setting. Nonetheless, more efficient immunization strategies are needed to stimulate an immune response of sufficient quality and magnitude to achieve an immunotherapeutic antiviral effect. The Journal of Immunology, 1998, 161: 4520 – 4529.

C

ytotoxic T Lymphocytes are thought to contribute to hepatitis B virus (HBV)3 clearance by killing infected hepatocytes and by secreting antiviral cytokines (1–3). Acutely infected patients characteristically produce a vigorous, polyclonal, and multispecific CTL response that is usually sufficient to clear the infection, while persistently infected patients produce a weak or undetectable HBV-specific CTL response (1, 2). Based on these observations, therapeutic enhancement of the T cell response to HBV has the potential to terminate chronic HBV infection. The immunologic basis for impaired T cell responsiveness to HBV in chronically infected patients must be defined, however, for rational immunotherapeutic approaches to be designed. While thymic deletion is likely to be operative in neonatal HBV infection, the precise mechanisms responsible for diminished T cell responsiveness to HBV in persistently infected neonates and adults are not understood. The availability of transgenic mice that express HBV envelope proteins (3) and of mice that express all the viral gene products and replicate HBV in their hepatocytes at levels comparable to those in chronically infected patients (4) affords the opportunity to study Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA 92037 Received for publication April 20, 1998. Accepted for publication June 23, 1998. 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 R37CA40489 and AI40696 from the National Institutes of Health and in part by the Japanese Society for the Promotion of Science (to Y.S.). This is manuscript 11517-MEM from The Scripps Research Institute. 2 Address correspondence and reprint requests to Dr. Francis V. Chisari, Department of Molecular and Experimental Medicine, The Scripps Research Institute, 10550 North Torrey Pines Rd., La Jolla, CA 92037. E-mail address: [email protected] 3 Abbreviations used in this paper: HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HBs, hepatitis B surface; DC, dendritic cell; GM-CSF, granulocytemacrophage colony-stimulatine factor; CTL-p, cytotoxic T lymphocyte precursors; HBeAg, hepatitis B e antigen; ALT, alanine aminotransferase; HBcAg, hepatitis B core antigen.

Copyright © 1998 by The American Association of Immunologists

the mechanisms responsible for neonatal tolerance to HBV and to develop immunotherapeutic approaches that have the potential to terminate persistent infection. We have previously shown that transgenic mice from a lineage that expresses the HBV envelope proteins under the control of the mouse albumin promoter (lineage 107-5) are immunologically tolerant to hepatitis B surface Ag (HBsAg) at the T cell level despite immunization with recombinant vaccinia viruses that routinely induce HBsAg-specific proliferative and CTL responses in nontransgenic littermates (5). The current study was undertaken to examine the ability of other modes of immunization to break tolerance at the CTL level in these and other lineages of transgenic mice. Plasmid DNA immunization has recently been shown to induce HBsAg-specific Ab and CTL responses in normal mice and chimpanzees (6 – 8), and it has been reported to induce anti-hepatitis B surface (HBs) production and to inhibit HBV gene expression in a unique lineage of HBV envelope transgenic mice in the absence of histologic evidence of hepatitis (9). Dendritic cells (DCs) are professional APCs (10, 11) that are distributed throughout the body and play a central role in Ag presentation to CD41 and CD81 T cells. DCs can be expanded from murine bone marrow cells with granulocyte-macrophage CSF (GM-CSF) and IL-4 (12, 13) and, when pulsed with Ag protein or peptide, can induce specific Ab and CTLs in vivo (14 –20). Accordingly, activated DC immunotherapy has recently been studied in patients with melanoma (21) and malignant lymphoma (22). Since activated DCs express high levels of costimulatory molecules and secrete inflammatory cytokines, they have the potential to activate anergic T cells. This idea is supported by the observation that simultaneous expression of B7-1 and a transgene-coded viral protein on pancreatic islet cells could break tolerance and induce a CTL response against the viral (self) proteins with islet destruction after infection with the corresponding virus (23, 24). Although the thymus can eliminate self-reactive T cells in fetal and neonatal life (25), the foregoing reports and other observations (26 –29) raise the possibility that peripheral tolerance might play 0022-1767/98/$02.00

The Journal of Immunology an important role in the establishment and/or maintenance of immunologic tolerance. If so, HBsAg-specific CTLs might be present in these HBV transgenic mice (and possibly in persistently infected patients) but be anergized. In the present study we attempted to break tolerance to HBsAg in several lineages of HBV transgenic mice by plasmid DNA immunization or by the administration of in vitro cytokine-activated, bone marrow-derived DCs. The results demonstrate that DC administration can induce a quantitatively normal HBsAg-specific CTL response in these transgenic mice, but DNA immunization cannot. Apparently, HBsAg (self)-specific CTL precursors have escaped negative selection in the thymus and are present, but unresponsive, in the transgenic mice, yet they can be activated by appropriate stimulation. Nonetheless, the CTLs failed to induce hepatitis or to suppress viral gene expression or replication, suggesting that although CTL tolerance can be broken by DC immunization more efficient strategies must be developed to generate a functionally effective CTL response in these animals and, by extension, in humans.

Materials and Methods Transgenic mice Five HBV transgenic mouse strains were used in this study. Lineage 107-5 (official designation, Tg[Alb-1, HBV] Bri66; inbred B10.D2, H-2d) contains the entire HBV envelope open reading frame under transcriptional control of the mouse albumin promoter and produces the large, middle, and small envelope polypeptides, all of which contain HBsAg (30). Lineage MUP-Env1 (Tg[MUP, HBV]Chi1; C57BL/6 3 B10.D2 F1) contains the entire HBV open reading frame downstream of the mouse major urinary promoter; however, transcription is controlled by the internal HBs promoter, resulting in the production of the middle and small HBV envelope polypeptides, both of which contain HBsAg. Lineage pFC80-219 (Tg[4HBV]Chi219; C57BL/6 3 B10.D2 F1) contains four complete HBV genomes (ayw subtype) linked in a tandem head to tail orientation at the HBV EcoRI site of pBR322, but only the HBs promoter is active, resulting in expression of the middle and small HBV envelope polypeptides (30 – 32). Lineages 1.3.46 (Tg [HBV 1.3 genome] Chi46; inbred B10.D2) and 1.3.32 (Tg [HBV 1.3 genome] Chi32; C57BL/6 3 BALB F1) express all the HBV proteins and replicate the virus in the liver at levels comparable to those found in patients chronically infected by HBV (4). Age- and sexmatched nontransgenic B10.D2, C57BL/6 3 B10.D2 F1, and C57BL/6 3 BALB F1 mice were included as controls.

DC cultures and immunization Bone marrow cells were collected from the femurs and tibiae of transgenic and syngeneic nontransgenic mice and suspended in RPMI 1640 supplemented with 10% FBS, 2 mM L-glutamine, 1 mM sodium pyruvate, 0.1 mM MEM nonessential amino acids, 50 U/ml penicillin, 50 mg/ml streptomycin (all from Life Technologies, Gaithersburg, MD), and 50 mM 2-ME (Sigma, St. Louis, MO). DCs were expanded according to the method of Lu et al. (13) with minor modifications. Briefly, the cells (2 3 106/well) were seeded in 24-well plates (Corning, Corning, NY) in the presence of 5 ng/ml of recombinant GM-CSF (PeproTech, Rocky Hill, NJ) and cultured to allow activated DC or DC progenitors to aggregate and lightly attach to the wells. After overnight incubation, nonadherent single cells were gently removed by swirling the plates and aspirating the medium. New medium, supplemented with 5 ng/ml of GM-CSF and 50 U/ml of rIL-4 (PharMingen, San Diego, CA), was added to the wells, and fourfifths of the medium was changed every 2 days. After 2 to 3 days of culture, clusters of expanding DC were observed, and each time the medium was changed, nonadherent small cells were removed by the procedure described above. On day 5 of culture, 1000 U/ml of rTNF-a (Genentech, South San Francisco, CA) was added in addition to GM-CSF and IL-4, and the cells were used on day 7 for in vivo or in vitro analysis. The activated DCs were either injected without further manipulation or were pulsed with 10 mg/ml of the immunodominant, Ld-restricted HBs28 –39 peptide plus 10 mg/ml of human b2m (Sigma) for 2 h at 37°C. Alternatively, the DCs were pulsed with recombinant preS2.S particles produced in stably transfected Chinese hamster ovary cells (obtained from Pasteur Merieux Institute, Val de Revil, France) at a concentration of 1 mg/ml on day 5 and further pulsed with the same Ag (100 mg/ml) for 2 h at 37°C immediately before injection. For both preparations, the cells were

4521 washed three times, and 1 3 106 DCs were i.v. injected into the tail vein of each mouse.

Phenotypic analysis of DC surface markers Surface markers of fresh bone marrow cells and cultured DCs were analyzed by flow cytometry on a FACScan (Becton Dickinson, Mountain View, CA). Staining with anti-CD80, anti-CD86, anti-Ld, anti-I-Ad, antiCD1d (all were purchased from PharMingen, San Diego, CA), and NLDC145 (anti-DEC-205, provided by Dr. Ralph M. Steinman, Rockefeller University, New York, NY) was followed by incubation with anti-rat IgG2a or anti-rat IgG2b (PharMingen), and incubation with anti-Ly-6C was followed by phycoerythrin-labeled streptavidin (Caltag, San Francisco, CA). Phycoerythrin-labeled anti-CD11a, anti-CD11c, anti-CD40, anti-CD44, antiCD49d, anti-CD49f, anti-LPAM-1, and anti-CD54 (all from PharMingen) Abs were also used and analyzed by single-parameter fluorescence histograms. To block the Fc receptor-mediated binding of the Abs, antiCD16/32 Abs (1 mg/106 cells; PharMingen) were included in all reactions.

DC presentation of an immunodominant HBsAg epitope to CTLs Splenic DC-enriched fractions were isolated by the method described by Saikh et al. (33). Briefly, spleen cells were cultured without cytokines for 2 h, and nonadherent cells were removed by gentle washing with RPMI medium. The adherent cells were cultured further for 18 h to allow DCs to detach from the dish, and low density cells were collected by density gradient centrifugation with Histo-Paque (sp. gr., 1.077; Sigma). To examine the ability of DCs to present an immunodominant HBsAg epitope to CTLs, cytokine-activated bone marrow-derived DCs or freshly isolated splenic DCs from nontransgenic and transgenic mice were labeled with 51 Cr and incubated with an HBs28 –39-specific, Ld-restricted CTL clone (6C2) (34) in a 4-h 51Cr release assay. The DCs were untreated or pulsed for 2 h with HBsAg particles (100 mg/ml) or peptide (10 mg/ml) before exposure to the CTL.

CTL activity after immunization Spleen cells were cultured with irradiated (20,000 rad) P815/preS1 transfectants and 3% EL-4 supernatant as a source of IL-2 for 5 to 7 days, and CTL activity was measured in a 4-h 51Cr release assay using P815 cells, P815/S transfectants, or peptide (HBs28 –39)-pulsed P815 cells as target cells as previously described (35). The percent cytotoxicity was calculated as [(experimental release 2 spontaneous release)/(total release 2 spontaneous release)] 3 100, and total and spontaneous release from wells containing 4% (v/v) Triton X-100 or medium, respectively, were determined. The percent specific cytotoxicity shown in this paper reflects the HBsAgspecific cytotoxic activity minus the Ag-nonspecific cytotoxic activity observed against the P815 control cells. Transgenic mouse spleen cells that had been primed in vivo were depleted of Thy1.21, CD41, or CD81 cells in vitro by magnetic bead-conjugated Ab (Dynal, Lake Success, NY) and analyzed for the ability to kill P815/S targets or HBs28 –39 peptide-pulsed P815 cells in a 4-h 51Cr release assay.

Limiting dilution analysis to determine the frequency of HBsAg-specific CTL precursor in spleen cells Quantitative analysis of HBsAg-specific CTL precursors (CTL-p) was performed by seeding 1,000, 3,000, 10,000 and 30,000 spleen cells/well in a 96-well round-bottom plate with irradiated (20,000 rad) P815/preS1 transfectants (1 3 104/well), irradiated (3,000 rad) syngeneic nontransfectant spleen cells (1 3 105/well), and 3% EL-4 supernatant in 24 replicates in 200 ml of RPMI 1640 plus 10% heat-inactivated FBS. After 8 days of culture the cytotoxic activity of each well was examined against P815 cells and P815/S transfectants, and the wells in which HBsAg-specific cytolytic activity exceeded the mean 1 3 SD of unimmunized spleen cell cultures seeded at the corresponding cell numbers per well were scored as positive. The CTL-p frequencies were calculated as described by Taswell (36).

Passive transfer of DC-immunized spleen cells Lineage 107-5 mice were immunized with HBs28 –39 peptide-pulsed nontransgenic bone marrow-derived DCs expanded in the presence of GMCSF, IL-4, and TNF-a, and their spleen cells were harvested 7 days later. The cells were cultured with irradiated (20,000 rad) P815/preS1 cells and 3% EL-4 supernatant for 7 days, and 5 3 106 cells/mouse were i.v. injected into transgenic mice from the same lineage and into syngeneic nontransgenic controls.

DNA immunization Fifty micrograms of plasmid pCMV-S2/S that contains the preS2.S coding region of HBV under the transcriptional control of the CMV immediate

4522

DENDRITIC CELL FUNCTION IN HBV TRANSGENIC MICE

FIGURE 1. Developmental expression of HBV RNA. HBV gene expression in the liver and thymus in fetus (18 days gestation), newborn (day 1), and adult (6 wk) animals of five lineages of HBV transgenic mice analyzed by Northern blot. L, liver; T, thymus.

early promoter (provided by Drs. R. Whalen and H. Davis) was injected into regenerating tibialis anterior muscles (100 mg/mouse) of transgenic mice and syngeneic nontransgenic controls 5 days after injection of cardiotoxin, exactly as previously described (37). This specific plasmid has been shown by Davis et al. (37) and Reimann et al. (38) to induce anti-HBs Ab and HBsAg-specific CTLs in nontransgenic mice and by Mancini et al. to inhibit HBV gene expression in an independent lineage of transgenic mice (9).

Serologic and biochemical analysis Serum was examined for HBsAg, hepatitis B e Ag (HBeAg), and anti-HBs titers using commercially available reagents (AUSRIA II-125, Abbott Laboratories (Abbott Park, IL), and EBK 125I RIA kit, Incstar (Stillwater, MN), respectively) and HBsAg and HBeAg were quantitated by comparison with known standards, as previously described (39). Hepatocellular injury was monitored at various time points after immunization by measuring serum alanine aminotransferase (ALT) activity (40).

Histologic and immunohistochemical analysis Tissue samples were fixed in 10% zinc-buffered formalin (Anatek, Battle Creek, MI), embedded in paraffin, sectioned (3 mm), and stained with hematoxylin and eosin as previously described (40). The intracellular distribution of hepatitis B core Ag (HBcAg) was assessed by the labeled avidinbiotin detection procedure and counterstained with Mayer’s hematoxylin before mounting (41).

DNA isolation and Southern blot Southern blot analysis was performed on total liver DNA from lineages 1.3.32 and 1.3.46 by agarose gel electrophoresis of 20 mg of restricted genomic DNA as previously described (4). Before electrophoresis, all DNA samples were digested with RNase A (Boehringer Mannheim, Indianapolis, IN) at 10 mg/ml for 1 h at 37°C. Nylon filters were hybridized with an HBV-specific 32P-radiolabeled DNA probe as previously described (42).

Northern blot analysis Frozen tissues were mechanically pulverized and extracted by the acidguanidium phenol-chloroform method (52). Total RNA (20 mg) was analyzed for HBV, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and albumin expression by Northern blot as previously described (42).

Results Developmental expression of HBV mRNA in liver and thymus of HBV transgenic mice Total cellular RNA isolated from the liver and thymus of fetal (18 days gestation), newborn (day of birth), and 6-wk-old adult trans-

genic mice from five transgenic mouse lineages was analyzed for HBV envelope and pregenomic RNA by Northern blot analysis. As shown in Figure 1, the adult liver from all lineages expressed high levels of the appropriate HBV transcripts, while the thymus of the adult animals was uniformly negative. Specifically, the adult livers from lineages 1.3.32 and 1.3.46 displayed the 3.5-kb pregenomic RNA and the 2.1-kb HBs RNA, while lineage 107-5 expressed the 2.4-kb preS1 and 2.1-kb HBs transcripts, and lineages 219 and ENV-1 expressed only the 2.1-kb HBs RNA. Importantly, the 2.4-kb and/or the 2.1-kb envelope RNA species were expressed in the fetal and/or newborn liver and thymus in all lineages except 1.3.32. Furthermore, the 3.5-kb RNA was detectable in fetal liver and/or thymus in lineages 1.3.32 and 1.3.46, and it was also detectable in the newborn thymus of lineage 1.3.46. Assuming that these transcripts were translated during fetal life or immediately after birth, neonatal tolerance to HBV envelope Ags would be expected in all lineages except 1.3.32. DC immunization Activation and phenotypic analysis of bone marrow and splenic DCs. Lineages 107-5 and 1.3.46 (inbred B10.D2) and lineage 1.3.32 (C57BL/6 3 BALB F1) were immunized with cytokineactivated transgenic and nontransgenic bone marrow-derived DCs. Between 8 and 10 3 106 DCs were routinely expanded from the bone marrow after 7 days of in vitro culture in the presence of GM-CSF, IL-4, and TNF-a, and 6 to 8 3 105 freshly isolated splenic DCs were recovered from lineages B10.D2, 107-5, and 1.3.46. The expression of CD11c, DEC-205, I-Ad, CD86, and CD40 was up-regulated with those cytokines (Fig. 2A), and both the bone marrow-derived (Fig. 2A) and splenic (Fig. 2B) DCs expressed similar levels of those markers as well as CD1d, CD11a, CD44, CD49d, CD49f, CD54, LPAM-1, and Ly-6C (not shown) regardless of their transgenic or nontransgenic origin. Thus, it appears that HBV gene expression in these transgenic animals does not affect the expansion or expression of activation markers by their splenic or bone marrow-derived DCs. HBsAg presentation by bone marrow and splenic DCs. The ability of cytokine-activated, bone marrow-derived DCs and freshly isolated splenic DCs from transgenic and nontransgenic mice to process HBsAg and to present an immunodominant HBs peptide was examined by measuring the susceptibility of HBsAg-pulsed

The Journal of Immunology

4523

FIGURE 2. Phenotypic analysis of bone marrow and splenic DCs. Flow cytometric analysis of phenotypic markers on A) bone marrow-derived nontransgenic DCs expanded in the presence of 5 ng/ml GM-CSF alone, 5 ng/ml GM-CSF plus 50 U/ml IL-4 or 5 ng/ml GM-CSF, 50 U/ml IL-4, plus 1000 U/ml TNF-a and transgenic (lineages 107-5 and 1.3.46) DCs expanded with 5 ng/ml GM-CSF, 50 U/ml IL-4, plus 1000 U/ml TNF-a, and on B) freshly isolated splenic DCs from nontransgenic lineage B10.D2 and transgenic lineages 107-5 and 1.3.46.

DCs to cytolysis by an HBs28–39-specific, Ld-restricted CTL clone. Initially, we showed that nontransgenic B10.D2 DCs process and present HBsAg to CTLs in a time- and cytokine-dependent manner, becoming susceptible to lysis within 30 min of Ag pulsing, with lysis increasing thereafter (Fig. 3A) and displaying optimal lysis when they were expanded in the presence of GM-CSF, IL-4, and especially TNF-a (Fig. 3B). Importantly, HBsAg-pulsed bone marrow-derived and splenic DCs from lineages 107-5 and 1.3.46 were as susceptible to lysis as B10.D2 DCs (Fig. 4), indicating that the Ag processing and presenting functions of the transgenic DCs are normal. Furthermore, we showed that freshly isolated splenic

DCs from transgenic lineage 107-5 were killed by the CTL without HBsAg pulsing (Fig. 4), indicating that the splenic DCs in this lineage internalize, process, and present HBsAg in vivo. In contrast, HBsAg pulsing was required for lysis of splenic DCs from lineage 1.3.46 (Fig. 4). The basis for this difference is not clear, although it could be due to the differential expression of the HBV large envelope polypeptide in the two lineages, which generates long, branching, filamentous HBsAg particles in lineage 107-5 that may be more susceptible to uptake from thecirculation than the small spherical HBsAg particles produced by lineage 1.3.46. In contrast to freshly isolated splenic DCs from 107-5 mice, cytokine-activated,

4524


FIGURE 3. Ag presentation by in vitro HBsAg-pulsed nontransgenic DCs. HBsAg presentation of bone marrow-derived DCs to an HBs28 –39-specific CTL clone was assessed by measuring its ability to lyse 51Cr-labeled DCs with or without pulsing with HBsAg in vitro. A, Time dependence of Ag presentation. DCs, expanded from B10.D2 bone marrow cells with 5 ng/ml GM-CSF, 50 U/ml IL-4, and 1000 U/ml TNF-a, were pulsed with HBsAg particles (100 mg/ml) for 30, 60, or 120 min or with HBs28 –39 peptide (10 mg/ml) for 120 min and used as targets for an HBs28 –39-specific CTL clone in a 4-h 51Cr release assay. Untreated DCs (open circles); DCs pulsed with HBsAg for 30 min (closed triangles), 60 min (closed diamonds), or 120 min (closed squares); and DCs pulsed with HBs28 –39 peptide for 120 min as positive controls of the assay (closed circles) are shown. B, Cytokine dependence of Ag presentation. DCs were expanded from B10.D2 bone marrow cells with various cytokine mixtures and used as targets for an HBs28 –39-specific CTL clone in a 4-h 51Cr release assay. Untreated DCs (open circles); HBsAg-pulsed DCs expanded with 5 ng/ml GM-CSF and 50 U/ml IL-4 (closed triangles), 5 ng/ml GM-CSF plus 500 U/ml IL-4 (closed diamonds), 5 ng/ml GM-CSF, 50 U/ml IL-4, plus 1000 U/ml TNF-a (closed squares); and DCs expanded with 5 ng/ml GM-CSF, 50 U/ml IL-4, plus 1000 U/ml TNF-a and pulsed with HBs28 –39 peptide for 120 min as positive controls of the assay (closed circles) are shown.

bone marrow-derived DCs from the same lineage were not killed without continuous HBsAg pulsing (Fig. 4), presumably because the half-life of MHC class I molecules is relatively short (43), and the HBsAg peptide-MHC complexes formed in vivo could turn over in the absence of Ag after 7 days of culture in vitro. HBsAg-specific CTL induction. The ability of DCs to induce HBsAg-specific CTLs was examined 1 wk after i.v. injection of one million bone marrow-derived nontransgenic DCs into transgenic (lineages 107-5 and 1.3.46) and syngeneic nontransgenic (B10.D2) mice. HBsAg-specific CTL activity of harvested spleen cells was examined after 1 wk of in vitro restimulation with P815/ preS1 transfectants. As shown in Figure 5, CTL activity against P815/S transfectants and HBs28–39 peptide-pulsed P815 target cells was induced in all three groups of DC-immunized mice, while unimmunized spleen cells showed no CTL activity, indicating that HBsAg-specific CTL activity detected in the immunized

animals reflects in vivo priming. Interestingly, prior Ag pulsing was not necessary for activated DCs to induce CTL responses in the transgenic mice (Fig. 5), suggesting that the DCs can internalize and present the HBsAg present in vivo to indigenous, class I-restricted T cells in these transgenic animals. In blocking experiments we demonstrated that the HBsAg-specific CTL activity of the DC-induced transgenic and nontransgenic CTLs was mediated by Thy1.21, CD81 T cells (not shown). Interestingly, by limiting dilution analysis we showed that the HBsAg-specific CTL-p frequencies in both lineages of transgenic mice were comparable to those in nontransgenic mice (Fig. 5), indicating that CTL precursors are not deleted in these animals despite the likely expression of HBsAg before or immediately after birth (Fig. 1). Furthermore, we showed that CTL responses can be induced in transgenic recipients by injection of as few as 100 bone marrowderived nontransgenic DCs expanded with GM-CSF and IL-4

FIGURE 4. Spontaneous Ag presentation by transgenic DCs. HBsAg presentation by bone marrow-derived or freshly isolated splenic DCs from HBV transgenic lineages 107-5 and 1.3.46 and nontransgenic B10.D2 mice. DCs are expanded from the bone marrow in the presence of 5 ng/ml GM-CSF, 50 U/ml IL-4, and 1000 U/ml TNF-a or are isolated from the spleen cells and used as targets for an HBs28 –39-specific CTL clone with or without pulsing (2 h) with HBsAg particles (100 mg/ml) in a 4-h 51Cr release assay. Untreated bone marrow-derived DCs (closed circles), HBsAg particle-pulsed bone marrow-derived DCs (open circles), untreated splenic DCs (closed triangles), and HBsAg particle-pulsed splenic DCs (open triangles) are shown. The experiments were repeated three times for each DC population, and a representative experiment is shown.


4525

FIGURE 5. Immunogenicity of DC in transgenic and nontransgenic mice. The cytotoxic activity of spleen cell cultures from DC-immunized mice and unimmunized mice was determined. One million bone marrow-derived nontransgenic DCs, expanded with 5 ng/ml GM-CSF, 50 U/ml IL-4, and 1000 U/ml TNF-a, were incubated with HBs28 –39 peptide (10 mg/ml), HBsAg particles (100 mg/ml), or medium alone for 2 h, after which they were washed three times and i.v. injected into HBV transgenic mice lineages 107-5 and 1.3.46 and nontransgenic B10.D2 mice. Spleen cells harvested 1 wk after injection of HBs28 –39 peptide, HBsAg particle-pulsed DCs, untreated DCs or unimmunized spleen cells were cultured with irradiated (20,000 rad) P815/preS1 and 3% EL-4 supernatant as a source of IL-2 for 7 days. The cytotoxic activity of those cells was examined against those of P815, P815/S transfectant, and P815 plus HBs28 –39 in a 4-h 51Cr release assay, and HBsAg-specific cytotoxic activities are shown after subtracting the Ag-nonspecific CTL activity against P815 targets at the corresponding E:T cell ratios. The results represent the mean 6 1 SEM from three mice per group. The numbers in each box represent HBsAg-specific CTL precursor frequencies determined by limiting dilution analysis using P815 and P815/S target cells.

and that transgenic recipients respond more strongly than nontransgenic recipients to small numbers of DCs (Fig. 6A). Importantly, we showed that cytokine-activated bone marrow-derived transgenic DCs can induce CTL responses as efficiently as non-

transgenic DCs in nontransgenic and transgenic recipients (Fig. 6, B and C), indicating that transgenic DCs can function normally in vivo, in keeping with their phenotypic (Fig. 2, A and B) and Agpresenting functional integrity (Fig. 4) in vitro. As shown in Figure

FIGURE 6. Relative immunogenicity of DCs from transgenic and nontransgenic donors in homologous and heterologous recipients. A, Varying numbers of bone marrow-derived DCs, expanded from nontransgenic B10.D2 mice with 5 ng/ml GM-CSF and 50 U/ml IL-4 and pulsed with HBs28 –39 peptide (100 mg/ml), were i.v. injected into lineages 107-5 and B10.D2. Spleen cells harvested 2 wk after DC immunization were cultured with irradiated (20,000 rad) P815/preS1 transfectants for 5 days, and CTL activity was measured in a 4-h 51Cr release assay using P815 cells and P815/S transfectants. The data are the means for two mice per group and are shown as HBsAg-specific cytotoxic activity at an E:T cell ratio of 25 by subtracting the Ag-nonspecific activity against P815 cells at the same E:T cell ratio. B, Transgenic DCs induce HBsAg-specific CTLs in B10.D2 nontransgenic (B) and 107-5 (C) recipients. One million bone marrow-derived DCs, expanded from lineages 107-5 and nontransgenic B10.D2 with 5 ng/ml GM-CSF and 50 U/ml IL-4, and pulsed with HBs28 –39 peptide (10 mg/ml), were injected i.v. into both groups of mice, and cytotoxic activity was examined after 5 days of in vitro stimulation with irradiated P815/preS1 transfectants. The data are the mean of two mice per group and are shown as HBsAg-specific cytotoxic activity by subtracting the Ag nonspecific cytotoxic activity against P815 targets at the corresponding E:T cell ratios.

4526


FIGURE 8. Failure of DNA immunization to induce HBsAg specific CTLs in HBV transgenic mice. CTL induction in HBV transgenic and nontransgenic mice after DNA immunization. Spleen cells harvested 4 wk after DNA immunization were cultured with irradiated (20,000 rad) P815/ preS1 transfectants and 3% EL-4 supernatant as a source of IL-2 for 7 days, and CTL activity was measured in a 4-h 51Cr release assay using P815 cells and P815/S transfectants as target cells. The data are the mean 6 1 SD from three mice and are shown as HBsAg-specific cytotoxic activity by subtracting the Ag-nonspecific activity against P815 cells at the corresponding E:T cell ratios.

FIGURE 7. Transience of the CTL response to DC immunization. Kinetics of CTL induction after immunization with HBs28 –39 peptide-pulsed DCs (A) and HBsAg particle-pulsed DCs (B). One million bone marrowderived DCs from B10.B2 mice, expanded with 5 ng/ml GM-CSF, 50 U/ml IL-4, and 1000 U/ml TNF-a and pulsed with either HBs28 –39 (10 mg/ml) or HBsAg particles (100 mg/ml), were i.v. injected into lineages 107-5, 1.3.46, and nontransgenic B10.D2 mice. Spleen cells harvested 1, 2 and 4, or 5 wk after DC immunization were cultured with irradiated (20,000 rad) P815/preS1 transfectants and 3% EL-4 supernatant as a source of IL-2 for 7 days, and CTL activity was measured in a 4-h 51Cr release assay using P815 cells and P815/S transfectants. The data are the mean 6 1 SEM from three to five mice per each group and are shown as HBsAg-specific cytotoxic activity by subtracting the Ag nonspecific activity against P815 cells at the corresponding E:T cell ratios.

7, however, the DC-induced CTL response in all three groups was transient, being strongest 1 to 2 wk after DC immunization and waning rapidly thereafter.

Table I. Passive transfer of DC-immunized spleen cells into HBV transgenic and nontransgenic mice a Serum ALT Donor

Recipient

107-5

107-5 B10.D2

Day 0

Day 1

Day 2

80 80 65 70

2750 2730 110 350

2200 1920 80 100

a HBV transgenic mice were immunized with 1 million HBs28 –39 peptide-pulsed nontransgenic DCs, and the spleen cells were harvested 7 days later. The cells were cultured with irradiated P815/PreS1 cells and 3% EL-4 supernatant for 7 days, and 5 3 106 spleen cells were intravenously injected into unimmunized transgenic and nontransgenic B10.D2 mice (2 mice per group). Serum ALT was monitored, and all mice were sacrificed on day 2 for histologic analysis.

Serum HBsAg and HBeAg levels, liver histology, and HBV gene expression or replication Serum HBsAg levels were monitored in 58 DC-immunized animals from lineages 107-5 (19 mice), 1.3.46 (36 mice), and 1.3.32 (three mice), and serum HBeAg was monitored in 19 DC-immunized animals from lineages 1.3.46 (16 mice) and 1.3.32(three mice). Serum levels of HBsAg in the lineage 107-5 mice did not change compared with those in 18 unimmunized transgenic controls (not shown). Interestingly, serum HBsAg levels fell in 10 of 39 DC-immunized mice from lineages 1.3.46 and 1.3.32, some of which also produced anti-HBs Abs (not shown). Since age- and sex-matched mice from these lineages spontaneously produce anti-HBs, however, the role of DC immunization in the induction of anti-HBs in these animals is unclear. In contrast, serum HBeAg levels did not decrease in any of the transgenic mice (not shown). Despite the ability of DC immunization to break CTL tolerance to HBsAg in these animals, histologic analysis of the liver 1, 2, or 4 wk after immunization displayed only a very slight focal inflammatory infiltrate in the hepatic parenchyma in occasional mice from lineage 107-5 mice, and no histologic abnormalities or elevated serum ALT activity were detected in lineages 1.3.46 or 1.3.32. Similarly, hepatic HBsAg and HBcAg expression (by immunohistochemical analysis), HBV RNA (by Northern blot analysis), and (in lineages 1.3.46 and 1.3.32) HBV DNA replicative intermediates indicative of viral replication (by Southern blot analysis) were unchanged when examined 1, 2, and 5 wk after DC immunization. A total of 17 DC-immunized and nine unimmunized animals from lineages 107-5 (eight and three mice, respectively), 1.3.46 (six and three mice, respectively), and 1.3.32 (three and three mice, respectively) were examined for down-regulation of viral gene expression and replication. No differences were observed between the immunized and unimmunized animals with respect to any of these parameters (not shown).


4527

FIGURE 9. Relative immunogenicity of DC and DNA immunization in nontransgenic and transgenic mice. Spleen cells from nontransgenic B6/BALB F1 mice (A) and transgenic lineage 1.3.32 B6/BALB mice (B) were harvested 2 wk after immunization with either HBsAg particle (100 mg/ml)-pulsed bone marrow-derived DCs (open circles) or plasmid pCMV-S2/S (100 mg/ml/mouse; closed circles) and cultured with irradiated (20,000 rad) P815/preS1 transfectants and 3% EL-4 supernatant as a source of IL-2 for 7 days. CTL activity was measured in a 4-h 51Cr release assay using P815 cells and P815/S transfectants. The data are shown as HBsAg-specific cytotoxic activity by subtracting the Ag-nonspecific activity against P815 cells at the corresponding E:T cell ratios. The results represent the mean 6 1 SEM from three mice per group.

Passive transfer of DC-immunized spleen cells Spleen cells from cytokine-activated, bone marrow-derived, DCimmunized mice from lineages 107-5, 1.3.32, and 1.3.46 were cultured in vitro with P815/preS1 and 3% EL-4 supernatant for 7 days, after which 5 3 106 cells were i.v. injected into the same lineages of transgenic mice and syngeneic nontransgenic mice. As shown in Table I, DC-immunized spleen cells from lineage 107-5 mice induced more than a 30-fold increase in serum ALT activity after transfer into homologous 107-5 animals but not into syngeneic nontransgenic B10.D2 recipients. Histologically, the liver demonstrated marked mononuclear cell infiltrates, apoptotic hepatocytes, and necroinflammatory foci (not shown). These results demonstrate that CTLs induced in DC-immunized transgenic mice are cytopathic in vivo, suggesting that the absence of hepatitis in these animals is probably due to insufficient numbers of CTLs rather than to a functional defect in homing, Ag recognition, or effector activity. DNA immunization Groups of at least six animals from each of the five HBV transgenic lineages described in Figure 1 plus groups of three to five syngeneic nontransgenic controls were immunized one or more times with 100 mg of pCMV-S2/S in their regenerating tibialis anterior muscles 5 days after cardiotoxin injection. Serum HBsAg, anti-HBs, and ALT levels were monitored before and at weekly intervals thereafter, and mice were sacrificed for analysis of splenic CTL activity and hepatic HBV gene expression and replication 4 and 12 wk later. A single injection of pCMV-S2/S induced a rapid and vigorous Ab response to HBsAg in nontransgenic B10.D2 mice (not shown). In contrast, anti-HBs Ab responses were not detected in any of the transgenic mice until 4 wk after immunization, when one or two mice from lineages 1.3.32 and 1.3.46 displayed low titer Abs, and serum HBsAg levels fell to undetectable or nearly undetectable levels in the same two lineages. As indicated earlier, unimmunized mice from these two lineages spontaneously produce anti-HBs and clear HBsAg from the serum at this age (not shown), suggesting that the Ab response in these mice may not have been induced by DNA immunization. Importantly, a rapid, reproducible, and vigorous HBsAg-specific splenic CTL response was detectable in all the nontransgenic

B10.D2 controls but in none of the transgenic animals when they were examined 4 wk after a single DNA immunization (Fig. 8) or at 12 wk, i.e., 4 wk after boosting performed 8 wk after the primary injection (not shown). In keeping with these results, we did not observe any histologic evidence of liver disease or changes in the hepatic content of HBcAg, HBV RNA, or replicative HBV DNA intermediates in a total of 87 mice (five from lineage 1.3.46, six each from lineages 107-5 and ENV-1, 27 from lineage 219, and 43 from lineage 1.3.32) analyzed at 4 and 12 wk after immunization (not shown). Relative immunogenicity of DC and DNA immunization The relative abilities of DC and DNA immunization to induce HBsAg-specific CTLs were simultaneously assessed 2 wk after immunization in lineage 1.3.32 animals and in syngeneic nontransgenic controls. As shown in Figure 9, DC immunization induced a stronger CTL response than DNA immunization in nontransgenic mice. Importantly, HBsAg-specific CTLs were only induced in HBV transgenic mice by DC immunization.

Discussion The mechanisms responsible for diminished immune responsiveness to HBV in chronically infected patients must be elucidated to understand the pathogenesis of chronic hepatitis and to develop immunotherapeutic strategies to terminate the infection. Since HBV has not been shown to inhibit Ag processing or presentation, cytokine production or responsiveness, or T cell function, immunologic tolerance to HBV is probably a multifactorial, host-determined event involving negative selection, anergy, exhaustion, ignorance, or defective APC function depending on the immunologic status of the host and the dose and route of the infection. In this study transgenic mice that express HBV in the liver and/or thymus during fetal or early life were used as a model of neonatal HBV infection, the commonest cause of HBV persistence worldwide (44). Using this system, we showed that the mice fail to mount a CTL response to HBsAg after DNA immunization but they produce normal numbers of HBsAg-specific CTLs after the infusion of ex vivo activated DCs, even transgenic DCs. This implies that CTL precursors are present in normal numbers in these animals, which is surprising because they do not produce a CTL

4528


response to HBsAg either spontaneously or after DNA immunization. This is especially puzzling because the bone marrow-derived and splenic DCs from the transgenic mice express normal levels of adhesion molecules, homing receptors, and activation markers; display normal Ag processing and presenting activities; and can break tolerance and induce normal numbers of HBsAg-specific CTLs when as few as 100 cytokine-activated transgenic DCs are injected into syngeneic transgenic recipients. It is important to note that the activated DCs did not require prior pulsing with HBsAg to induce HBsAg-specific CTLs in the transgenic mice, implying that the activated transgenic DCs can internalize, process, and present HBsAg to CTL precursors in vivo. This is consistent with evidence that exogenous HBsAg can enter the class I processing pathway and induce class I-restricted CTLs in vivo in nontransgenic animals (45, 46). Indeed, in vitro pulsing of DCs with HBsAg supported this idea, since cytokine-activated DCs pulsed with HBsAg in vitro for only 30 min could be killed by HBs28 –39-specific CTLs, suggesting that cytokine-activated DCs have the ability to internalize HBsAg present in the circulation, and process and present the HBs immunodominant peptide rapidly after injection into HBV transgenic mice. These results suggest that the indigenous resting DCs in the transgenic mice are not only phenotypically and functionally normal but, when activated, they have the potential to process and present circulating HBsAg to class I-restricted T cell precursors that are present in normal numbers in these animals. Collectively, these results suggest that CTL tolerance to HBsAg in our HBV transgenic mice reflects anergy or ignorance rather than T cell exhaustion or deletion. The results also suggest that to become activated the transgenic CTLs must receive signals that are not required by nontransgenic CTLs, and they suggest that although these signals are provided by the ex vivo-activated DCs, they are not provided by the resting indigenous DCs in these animals even after DNA immunization remains. It is possible that the transgenic CTLs require more efficient help or more effective costimulation than the nontransgenic CTLs, as might occur if HBsAg-specific TCR expression or activation is down-regulated in the transgenic mice due to constant interaction with Ag presented by nonprofessional APCs (e.g., hepatocytes). Indeed, the up-regulation of costimulatory molecules on the DCs might be crucial, since they deliver not only costimulatory signals but also survival signals to CTLs through up-regulation of Bcl-XL (47). Furthermore, costimulation of T cells is reported to lower the activation threshold of T cells (48) and decreases the time of commitment for T cell activation (49). It is also possible that the ex vivo activated DCs deliver Ag more efficiently to the lymphoid tissues for CTL induction than occurs either spontaneously or after DNA immunization. Additional studies will be required to test these various hypotheses. It will also be interesting to determine whether pharmacologic activation of transgenic DCs in vivo will be sufficient to induce a CTL response to HBV in these transgenic mice, because if similar mechanisms are operative in chronically infected patients, the current results suggest that cytokine-activated DCs from chronically infected patients might have immunotherapeutic value. The immunogenicity of this approach must be improved, however, because despite the fact that HBsAg-specific CTLs were induced in the transgenic mice by DC immunization, they did not cause liver disease or down-regulate HBV RNA expression or HBV DNA replication (50). This could be due to the induction of insufficient numbers of CTLs, the brief duration of the CTL response, defective trafficking of the CTLs to the liver, or the activation of low avidity CTLs. We favor the first two alternatives because viral clearance during the acute phase of many viral infections is asso-

ciated with a sustained CTL response with splenic CTL-p frequencies of .1 in 100 T cells (23) while the CTL response to HBsAg induced by DC immunization was transient and only reached 1 in 6,000 –20,000 at its peak. Functional CTL deficiencies, including their trafficking and homing potential, are unlikely, since DC-immunized transgenic spleen cells were able to cause hepatitis when injected into transgenic recipients. Therefore, we consider it most likely that a higher number of HBsAg-specific CTLs must be present for a longer period of time to cause hepatitis and to inhibit HBV gene expression and replication in this system. Although additional experiments are required to confirm this hypothesis, our results suggest that Ag presentation by activated DCs can trigger an anti-viral CTL response in HBV transgenic mice and that this strategy is more efficient than DNA immunization in this setting. Additional experiments are clearly required to understand why CTL tolerance in these transgenic mice can be broken by DC immunization but not by plasmid DNA or recombinant vaccinia virus (5), both of which induce HBsAg-specific CTLs in syngeneic nontransgenic animals very efficiently (35, 51). Additional studies are also needed to understand the discrepancy between our results and a report by Mancini et al. (9) that DNA immunization can suppress hepatic HBsAg-specific RNA content in an independent lineage of HBsAg transgenic mice in the absence of evidence of a CTL response or inflammatory liver disease. Since DNA immunization did not reduce the hepatic HBV RNA content in the five HBV transgenic lineages we studied, it is possible that lineage-specific effects (e.g., transgene structure or integration site) unique to the transgenic mice used in their experiments rather than the virusspecific effects (i.e., inhibition of HBV gene expression by plasmid DNA-induced cytokine production) proposed by Mancini et al. (9), which should be shared by all of the lineages, may be responsible for the conflicting results. This idea is also supported by the fact that the genetic backgrounds of two of our lineages (219 and 1.3.32) and the lineage used by Mancini et al. are similar or identical and because the same plasmid (pCMV-S2S) was used under the same immunization conditions in their studies and ours. Until the basis for these conflicting results is resolved, generalizations regarding the value of DNA immunization for the treatment of chronic HBV infection (9) may be premature.

Acknowledgments We thank Dr. R. Steinman for providing the dendritic cell-specific Ab NLDC-145, Drs. R. G. Whalen and H. L. Davis for providing plasmid pCMV-S2/S, and Violeta Alvarado, Brent Matzke, Heike McClary, and Amy Brown for excellent technical assistance.

References 1. Nayersina, R., P. Fowler, S. Guilhot, G. Missale, A. Cerny, H.-J. Schlicht, A. Vitiello, R. Chesnut, J. L. Person, A. G. Redeker, et al. 1993. HLA A2 restricted cytotoxic T lymphocyte responses to multiple hepatitis B surface antigen epitopes during hepatitis B virus infection. J. Immunol. 150:4659. 2. Rehermann, B., P. Fowler, J. Sidney, J. Person, A. Redeker, M. Brown, B. Moss, A. Sette, and F. V. Chisari. 1995. The cytotoxic T lymphocyte response to multiple hepatitis B virus polymerase epitopes during and after acute viral hepatitis. J. Exp. Med. 181:1047. 3. Chisari, F. V., C. A. Pinkert, D. R. Milich, P. Filippi, A. McLachlan, R. D. Palmiter, and R. L. Brinster. 1985. A transgenic mouse model of the chronic hepatitis B surface antigen carrier state. Science 230:1157. 4. Guidotti, L. G., B. Matzke, H. Schaller, and F. V. Chisari. 1995. High level hepatitis B virus replication in transgenic mice. J. Virol. 69:6158. 5. Wirth, S., L. G. Guidotti, K. Ando, H. J. Schlicht, and F. V. Chisari. 1995. Breaking tolerance leads to autoantibody production but not autoimmune liver disease in HBV envelope transgenic mice. J. Immunol. 154:2504. 6. Pardoll, D. M., and A. M. Beckerleg. 1995. Exposing the immunology of naked DNA vaccines. Immunity 3:165. 7. Kuhrober, A., H.-P. Pudollek, K. Reifenberg, F. V. Chisari, H.-J. Schlicht, J. Reimann, and R. Schirmbeck. 1996. DNA immunization induces antibody and cytotoxic T cell response to hepatitis B core antigen to H-2b mice. J. Immunol. 156:3687.

The Journal of Immunology 8. Davis, H. L., M. J. McCluskie, J. L. Gerin, and R. H. Purcell. 1996. DNA vaccine for hepatitis B: evidence for immunogenicity in chimpanzees and comparison with other vaccines. Proc. Natl. Acad. Sci. USA 93:7213. 9. Mancini, M., M. Hadchouel, H. L. Davis, R. G. Whalen, P. Tiollais, and M.-L. Michel. 1996. DNA-mediated immunization in a transgenic mouse model for the hepatitis B surface antigen chronic carrier state. Proc. Natl. Acad. Sci. USA 93: 12496. 10. Steinman, R. M. 1991. The dendritic cell system and its role in immunogenicity. Annu. Rev. Immunol. 9:271. 11. Steinman, R. M. 1996. Dendritic cells and immune-based therapies. Exp. Hematol. 24:859. 12. Myordomo, J. I., T. Zorina, W. J. Storkus, L. Zitvogel, C. Celluzzi, L. D. Falo, C. J. Melief, S. T. Ildstad, W. M. Kast, A. B. DeLeo, et al. 1995. Bone marrowderived dendritic cells pulsed with synthetic tumour peptides elicit protective and therapeutic antitumor immunity. Nat. Med. 1:1297. 13. Lu, L., D. McCaslin, T. E. Starzl, and A. W. Thomson. 1995. Bone marrowderived dendritic cell progenitors (NLDC 1451, MHC class II1, B7-1dim, B7-22) induce alloantigen-specific hyporesponsiveness in murine T lymphocytes. Transplantation 60:1539. 14. Inaba, K., J. P. Metlay, M. T. Crowley, and R. M. Steinman. 1990. Dendritic cells pulsed with protein antigens in vitro can prime antigen-specific, MHC-restricted T cells in situ. J. Exp. Med. 172:631. 15. Sornasse, T., V. Flamand, G. De Becker, H. Bazin, F. Tielemans, K. Thielemans, J. Urbain, O. Leo, and M. Moser. 1992. Antigen-pulsed dendritic cells can efficiently induce an antibody response in vivo. J. Exp. Med. 175:15. 16. Celluzzi, C. M., J. I. Mayordomo, W. J. Storkus, M. T. Lotze, and L. D. J. Falo. 1996. Peptide-pulsed dendritic cells induce antigen-specific, CTL-mediated protective tumor immunity. J. Exp. Med. 183:283. 17. Young, J. W., and K. Inaba. 1996. Dendritic cells as adjuvants for class I major histocompatibility complex-restricted antitumor immunity. J. Exp. Med. 183:7. 18. Zitvogel, L., J. I. Mayordomo, T. Tjandrawan, A. B. DeLeo, M. R. Clarke, M. T. Lotze, and W. J. Storkus. 1996. Therapy of murine tumors with tumor peptide-pulsed dendritic cells: dependence on T cells, B7 costimulation, and T helper cell 1-associated cytokines. J. Exp. Med. 183:87. 19. Paglia, P., C. Chiodoni, M. Rodolfo, and M. P. Colombo. 1996. Murine dendritic cells loaded in vitro with soluble protein prime cytotoxic T lymphocytes against tumor antigen in vitro. J. Exp. Med. 183:317. 20. Porgador, A., and E. Gilboa. 1995. Bone marrow-generated dendritic cells pulsed with a class I-restricted peptide are potent inducers of cytotoxic T lymphocytes. J. Exp. Med. 182:255. 21. Hu, X., N. G. Chakraborty, J. R. Sporn, S. H. Kurtzman, M. T. Ergin, and B. Mukherji. 1996. Enhancement of cytolytic T lymphocyte precursor frequency in melanoma patients following immunization with the MAGE-1 peptide loaded antigen presenting cell-based vaccine. Cancer Res. 56:2479. 22. Caux, C., Y.-J. Liu, and J. Banchereau. 1995. Recent advances in the study of dendritic cells and follicular dendritic cells. Immunol. Today 16:2. 23. von Herrath, M. G., S. Guerder, H. Lewicki, R. A. Flavell, and M. B. A. Oldstone. 1995. Coexpression of B7-1 and viral (“self”) transgenes in pancreatic b cells can break peripheral ignorance and lead to spontaneous autoimmune diabetes. Immunity 36:727. 24. Harlan, D. M., H. Hengartner, M. L. Huang, Y. H. Kang, R. Abe, R. W. Moreadith, H. Pircher, G. S. Gray, P. S. Ohashi, and G. J. Freeman. 1994. Mice expressing both B7-1 and viral glycoprotein on pancreatic beta cells along with glycoprotein-specific transgenic T cells develop diabetes due to a breakdown of T-lymphocyte unresponsiveness. Proc. Natl. Acad. Sci. USA 91:3137. 25. Sprent, J., D. Lo, E.-K. Gao, and Y. Ron. 1988. T cell selection in the thymus. Immunol. Rev. 101:173. 26. Rocha, B., C. Tanchot, and H. von Boehmer. 1993. Clonal anergy blocks in vivo growth of mature T cells and can be reversed in the absence of antigen. J. Exp. Med. 177:1517. 27. Van Parijs, L., A. Ibraghimov, and A. K. Abbas. 1996. The roles of costimulation and Fas in T cell apoptosis and peripheral tolerance. Immunity 4:321. 28. Tivol, E. A., F. Borriello, A. N. Schweitzer, W. P. Lynch, J. A. Bluestone, and A. H. Sharpe. 1995. Loss of CTLA-4 leads to massive lymphoproliferation and fetal multiorgan tissue destruction, revealing a critical negative regulatory role for CTLA-4. Immunity 3:541. 29. Bluestone, J. A. 1997. Is CTLA-4 a master switch for peripheral T cell tolerance? J. Immunol. 158:1989. 30. Chisari, F. V., P. Filippi, J. Buras, A. McLachlan, H. Popper, C. A. Pinkert, R. D. Palmiter, and R. L. Brinster. 1987. Structural and pathological effects of

4529

31.

32.

33.

34.

35.

36. 37.

38.

39.

40.

41.

42.

43.

44. 45.

46. 47.

48. 49. 50.

51.

52.

synthesis of hepatitis B virus large envelope polypeptide in transgenic mice. Proc. Natl. Acad. Sci. USA 84:6909. Gilles, P. N., G. Fey, and F. V. Chisari. 1992. Tumor necrosis factor-alpha negatively regulates hepatitis B virus gene expression in transgenic mice. J. Virol. 66:3955. Gilles, P. N., D. L. Guerrette, R. J. Ulevitch, R. D. Schreiber, and F. V. Chisari. 1992. Hepatitis B surface antigen retention sensitizes the hepatocyte to injury by physiologic concentrations of g interferon. Hepatology 16:655. Saikh, K. U., A. H. Nishikawa, and S. B. Dillon. 1996. Detection of enhanced cytotoxic T lymphocyte response with splenic dendritic cells enriched by a simple sequential adherence method. Viral. Immunol. 9:121. Ando, K., T. Moriyama, L. G. Guidotti, S. Wirth, R. D. Schreiber, H. J. Schlicht, S. Huang, and F. V. Chisari. 1993. Mechanisms of class I restricted immunopathology: a transgenic mouse model of fulminant hepatitis. J. Exp. Med. 178:1541. Moriyama, T., S. Guilhot, K. Klopchin, B. Moss, C. A. Pinkert, R. D. Palmiter, R. L. Brinster, O. Kanagawa, and F. V. Chisari. 1990. Immunobiology and pathogenesis of hepatocellular injury in hepatitis B virus transgenic mice. Science 248:361. Taswell, C. 1981. Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis. J. Immunol. 126:1614. Michel, M.-L., H. L. Davis, M. Schleef, M. Mancini, P. Tiollais, and R. G. Whalen. 1995. DNA-mediated immunization to hepatitis B surface antigen in mice: aspects of the humoral response mimic hepatitis B viral infection in humans. Proc. Natl. Acad. Sci. USA 92:5307. Davis, H. L., R. Shirmbeck, J. Reimann, and R. G. Whalen. 1995. DNA-mediated immunization in mice induces a potent MHC class I-restricted cytotoxic T lymphocyte response to the hepatitis B envelope protein. Hum. Gene Ther. 6:1447. Chisari, F. V., P. Filippi, A. McLachlan, D. R. Milich, M. Riggs, S. Lee, R. D. Palmiter, C. A. Pinkert, and R. L. Brinster. 1986. Expression of hepatitis B virus large envelope polypeptide inhibits hepatitis B surface antigen secretion in transgenic mice. J. Virol. 60:880. Chisari, F. V., K. Klopchin, T. Moriyama, C. Pasquinelli, H. A. Dunsford, S. Sell, C. A. Pinkert, R. L. Brinster, and R. D. Palmiter. 1989. Molecular pathogenesis of hepatocellular carcinoma in hepatitis B virus transgenic mice. Cell 59:1145. Guidotti, L. G., V. Martinez, Y. T. Loh, C. E. Rogler, and F. V. Chisari. 1994. Hepatitis B virus nucleocapsid particles do not cross the hepatocyte nuclear membrane in transgenic mice. J. Virol. 68:5469. Guidotti, L. G., S. Guilhot, and F. V. Chisari. 1994. Interleukin 2 and interferon alpha/beta downregulate hepatitis B virus gene expression in vivo by tumor necrosis factor dependent and independent pathways. J. Virol. 68:1265. Cella, M., E. A., V. Pinet, J. Pieters, and A. Lanzavecchia. 1998. Inflammatory stimuli induce accumulation of MHC class II complexes on dendritic cells. Nature 388:782. Chisari, F. V., and C. Ferrari. 1995. Hepatitis B virus immunopathogenesis. Annu. Rev. Immunol. 13:29. Bo¨hm, W., R. Schirmbeck, A. Elbe, K. Melber, D. Diminky, G. Kraal, N. van Rooijen, Y. Barenholz, and J. Reimann. 1995. Exogenous hepatitis B surface antigen particles processed by dendritic cells or macrophages prime murine MHC class I-restricted cytotoxic T lymphocytes in vivo. J. Immunol. 155: 3313. Jondal, M., R. Schirmbeck, and J. Reimann. 1996. MHC class I-restricted CTL responses to exogenous antigens. Immunity 5:295. Boise, L. H., A. J. Minn, P. J. Noel, C. H. June, M. A. Accavitti, T. Lindsten, and C. B. Thompson. 1995. CD28 costimulation can promote T cell survival by enhancing the expression of Bcl-xL. Immunity 3:87. Viola, A., and A. Lanzavecchia. 1998. T cell activation determined by T cell receptor number and tunable thresholds. Science 273:104. Iezzi, G., K. Karjalainen, and A. Lanzavecchia. 1996. The duration of antigenic stimulation determines the fate of naive and effector T cells. Immunity 8:89. Guidotti, L. G., T. Ishikawa, M. V. Hobbs, B. Matzke, R. Schreiber, and F. V. Chisari. 1996. Intracellular inactivation of the hepatitis B virus by cytotoxic T lymphocytes. Immunity 4:25. Schirmbeck, R., W. Bo¨hm, K. Ando, F. V. Chisari, and J. Reimann. 1995. Nucleic acid vaccination primes hepatitis B surface antigen specific cytotoxic T lymphocytes in nonresponder mice. J. Virol. 69:5929. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162: 156.