Mouse Model of Hepatitis B Virus Carrier

The Journal of Immunology

IL-12–Based Vaccination Therapy Reverses Liver-Induced Systemic Tolerance in a Mouse Model of Hepatitis B Virus Carrier

Zhutian Zeng,* Xiaohui Kong,* Fenglei Li,* Haiming Wei,*,† Rui Sun,*,† and Zhigang Tian*,†

Liver-induced systemic immune tolerance that occurs during chronic hepadnavirus infection is the biggest obstacle for effective viral clearance. Immunotherapeutic reversal of this tolerance is a promising strategy in the clinic but remains to be explored. In this study, using a hepatitis B virus (HBV)-carrier mouse model, we report that IL-12–based vaccination therapy can efficiently reverse systemic tolerance toward HBV. HBV-carrier mice lost responsiveness to hepatitis B surface Ag (HBsAg) vaccination, and IL-12 alone could not reverse this liver-induced immune tolerance. However, after IL-12–based vaccination therapy, the majority of treated mice became HBsAg2 in serum; hepatitis B core Ag was also undetectable in hepatocytes. HBV clearance was dependent on HBsAg vaccine-induced anti-HBV immunity. Further results showed that IL-12–based vaccination therapy strongly enhanced hepatic HBV-specific CD8+ T cell responses, including proliferation and IFN-g secretion. Systemic HBV-specific CD4+ T cell responses were also restored in HBV-carrier mice, leading to the arousal of HBsAg-specific follicular Th–germinal center B cell responses and anti–hepatitis B surface Ag Ab production. Recovery of HBsAg-specific responses also correlated with both reduced CD4+Foxp3+ regulatory T cell frequency and an enhanced capacity of effector T cells to overcome inhibition by regulatory T cells. In conclusion, IL-12–based vaccination therapy may reverse liver-induced immune tolerance toward HBV by restoring systemic HBV-specific CD4+ T cell responses, eliciting robust hepatic HBV-specific CD8+ T cell responses, and facilitating the generation of HBsAg-specific humoral immunity; thus, this therapy may become a viable approach to treating patients with chronic hepatitis B. The Journal of Immunology, 2013, 191: 4184–4193.

he liver favors induction of immune tolerance rather than Therapeutic vaccination against HBV is a promising approach immunity during chronic infection with hepatotropic patho- to reverse HBV tolerance in chronically infected individuals (6, 7). T gens (1), such as hepatitis B virus (HBV). HBV infection Previous animal studies revealed that immunization with a hepa- by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. is one of the most serious health concerns worldwide, because titis B surface Ag (HBsAg) vaccine resulted in decreased HBV chronically infected individuals are predisposed to development of titers and the generation of anti-HBs in HBV-transgenic (tg) mice severe liver cirrhosis and hepatocellular carcinoma (2, 3). Lifelong (8, 9). However, clinical studies using these strategies revealed viral persistence in chronic hepatitis B (CHB) carriers is due to poor therapeutic effects in CHB patients in an immunotolerant liver-induced immune tolerance toward HBV, which is character- phase (10, 11). The divergence in therapeutic outcome may be ized by defective HBV-speciﬁc T cell responses and undetectable due, in part, to differences in the type of tolerance induced be- anti–hepatitis B surface Ag Ab (anti-HBs) levels (4). Reversing this tween HBV-tg mice and CHB patients (12, 13). Therefore, a immune tolerance will therefore be a critical step toward completely mouse model that more accurately mimics HBV-induced “ac- eliminating HBV from the host (5, 6). quired” tolerance during CHB infection could aid in developing http://classic.jimmunol.org immunotherapeutic strategies against HBV. Recently, an immunocompetent HBV-carrier mouse model was established using *Department of Immunology, School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China; and †Hefei National Laboratory hepatic gene transfer of an HBV-containing plasmid by hydro- for Physical Sciences at Microscale, Hefei, Anhui 230027, China dynamic injection. Because these mice are completely unrespon- Received for publication December 19, 2012. Accepted for publication August 11, sive to peripheral HBsAg vaccination because of liver-induced 2013. systemic tolerance, they provide a better mouse model with which

Downloaded from This work was supported by the Ministry of Science and Technology of China of to study liver tolerance during HBV persistence (12). the People’s Republic of China (973 Basic Science Project 2012CB519004, The transient and insufficient effects of vaccine therapy alone in 2009CB522403), the National Natural Science Foundation of China (91029303, 31021061), and National Science and Technology Major Projects (2012ZX10002- clinical studies also suggest the need for combination therapy (5, 014, 2012ZX10002006). 6). On one hand, combining HBsAg vaccine with an immunosti- Address correspondence and reprint requests to Dr. Zhigang Tian, School of Life mulator, such as CpG, exhibited improved efficacy to break HBV Sciences, University of Science and Technology of China, 443 Huang-Shan Road, tolerance via Th1-biased mechanisms (14). On the other hand, Hefei, Anhui 230027, China. E-mail address: [email protected] combining HBsAg vaccine with an antiviral agent, including lami- The online version of this article contains supplemental material. vudine, was favored in other trials, because HBV-specific T cell Abbreviations used in this article: alum, aluminum hydroxide; anti-HBs, anti–hepatitis B surface Ag Ab; CHB, chronic hepatitis B; dLN, draining lymph node; GC, germinal hyporesponsiveness during CHB infection always correlated with center; HBcAg, hepatitis B core protein; HBeAg, hepatitis B e Ag; HBsAg, hepatitis B high viral load (15). However, neither approach was completely surface Ag; HBV, hepatitis B virus; HBV-tg, HBV-transgenic; SI, stimulation index; effective in the clinic, suggesting the possibility that an agent pos- TdR, thymidine; Treg, regulatory T cell. sessing both immunostimulatory and antiviral effects could optimize Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 vaccination therapy in combination with a vaccine.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1203449 The Journal of Immunology 4185

IL-12 is a typical proinflammatory cytokine that plays a critical (University of Science and Technology of China). All mice were main- role in host defense against pathogens, including HBV (16). Im- tained under specific pathogen-free conditions and used in accordance with paired IL-12 secretion by dendritic cells is implicated in con- the Guide for the Care and Use of Laboratory Animals. tributing to the chronicity of HBV infection (17). Importantly, IL- HBV-carrier mouse model 12 can also inhibit HBV replication in HBV-tg mice by inducing The pAAV-HBV1.2 plasmid, containing the full-length HBV DNA, was IFN-g production (18). However, the antiviral effect of IL-12 kindly provided by Dr. Ding-Shinn Chen (National Taiwan University alone does not appear to be more effective than other traditional College of Medicine, Taipei, Taiwan). To establish the HBV-carrier mouse drugs in clinical trials (19). In this study, we used an HBV-carrier model, we injected C57BL/6 mice (5-wk-old male mice) with 5 mg pAAV- mouse model to study the effect of IL-12 on overcoming liver- HBV1.2 plasmid using a hydrodynamic tail-vein injection approach. Six induced systemic tolerance. Although neither IL-12 nor HBV weeks later, mice with serum HBsAg levels of at least 200 ng/ml were considered to be HBV-carrier mice. vaccine alone could efficiently eliminate HBV, a combination treatment using both HBsAg vaccine and IL-12 led to effective HBV IL-12 treatment and immunization clearance. HBV-carrier mice were screened out 6 wk after HBV plasmid injection and s.c. injected with 100 ng IL-12 (in 200 ml total volume of PBS; Peprotech, Materials and Methods Rocky Hill, NJ) daily for 3 consecutive days, followed by s.c. immuni- Mice zation with 2 mg recombinant HBsAg vaccine (adw subtype; Biokangtai Company, Shenzhen, China) in the presence of 100 ng IL-12 as a coad- Five-week-old male C57BL/6 mice were purchased from SLAC Laboratory juvant. This treatment regimen was repeated weekly for 3 wk. In some Animal Corporation (Shanghai, China). HBV-tg mice were purchased from experiments, HBsAg vaccine was replaced by recombinant HBsAg protein the Infectious Disease Center of the No. 458 Hospital (Guangzhou, China). (HyTest, Turku, Finland) emulsified in aluminum hydroxide (alum) (Imject 2 2 Rag1 / mice were purchased from the Model Animal Research Center Alum, Pierce, IL). Blood samples were obtained weekly by tail bleeding, 2 2 (Nanjing, China). CD4 / mice were a kind gift from Dr. Zhexiong Lian and sera were kept at 220˚C until use. by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 1. IL-12–based vaccination therapy efficiently eliminated HBV in HBV-carrier mice. (A) HBV-carrier mice were screened out 6 wk after HBV plasmid injection and were treated with IL-12–based vaccination therapy (as shown), HBsAg vaccination alone, or no treatment. (B) HBsAg levels were monitored in PBS+HBsAg vaccine–treated, IL-12+HBsAg vaccine–treated, and untreated control mice. Data represent the pooled results from three separate experiments (each point represents one mouse). (C) The ratio of HBsAg-seropositive mice was analyzed. Mice exhibiting HBsAg levels ,2 ng/ml were considered HBsAg2.(D) Serum HBeAg levels and (E) HBV DNA copies were measured in each group 3 wk after treatment initiation. The cutoff value was 1 national clinical unit per milliliter (NCU/ml). (F) Immunohistochemical staining of HBcAg in liver sections from HBV-carrier mice treated with HBsAg vaccination alone or IL-12–based vaccination therapy. Micrographs (original magnification 3200) are representative liver sections from each group at week 4 after treatment initiation. A liver section from an HBV-carrier mouse that was screened out 6 wk after HBV injection and before treatment initiation was used as the control. (G) Serum ALT levels were monitored during treatment. ALT levels ,28 IU/l were considered physiologically normal. Data are representative results from three independent experiments for (D), (E), and (G), with each group comprising 7–10 mice, and are shown as the mean 6 SEM. *p , 0.05, **p , 0.01, ***p , 0.001. N.S., No significant difference. 4186 IL-12 REVERSES HBV TOLERANCE

Radioimmunoassay Cell isolation Serum HBsAg, hepatitis B e Ag (HBeAg), anti-HBs, and anti-HBe levels Hepatic mononuclear cells and splenocytes were separated as previously were measured using the corresponding radioimmunoassay kits (North described (20). In brief, the murine liver was removed, cut into small Institute of Biological Technology, Beijing, China) according to the manu- pieces, and passed through a 200-gauge mesh. After a 5-min incubation at facturer’s instructions. 4˚C, the cells were pelleted, resuspended in 3 ml of a 40% Percoll solution, and added onto 2 ml of a 70% Percoll solution. The cells were then HBV DNA detection centrifuged at 1260 3 g for 30 min. The hepatic mononuclear cells at the interface between the two Percoll gradients were collected and washed HBV DNA copies were detected by probe-based quantitative PCR using twice at 400 3 g for 5 min. The spleen was removed and passed through a diagnostic kit for HBV DNA (KHB, Shanghai, China) according to the a 200-gauge mesh. Splenocytes were then harvested after RBC lysis and manufacturer’s instructions. washing. For CD4+CD25+ and CD4+CD252 cell isolation, splenocytes were sorted by MACS separation using a CD4+CD25+ regulatory T cell Immunohistochemistry (Treg) isolation kit (Miltenyi Biotech, Auburn, CA) according to the manufacturer’s instructions. The purities of the isolated CD4+CD25+ and m + 2 Paraffin-embedded, 6- m-thick liver sections were stained with rat anti– CD4 CD25 cells were .90 and 85%, respectively. In a parallel experi- hepatitis B core protein (anti-HBcAg; Dako, Glostrup, Denmark) and vi- + + + 2 ment, both CD4 CD25 and CD4 CD25 cells were sorted using FACS to sualized using a diaminobenzidine peroxidase substrate kit (Vector, Bur- + achieve 95% purity. Total CD4 T cell isolation was performed by MACS lingame, CA). separation using a mouse CD4+ T cell isolation kit (Miltenyi Biotech). For isolating CD11ahiCD8alo and CD11aloCD8ahi cells, splenocytes were la- ELISPOT assay beled with PerCP–Cy5.5–anti-CD8a, PE–Cy7–anti-CD11a, and allophycocyanin–anti-CD3, and then the labeled cells were sorted by FACS. HBsAg-specific IgG ELISPOT detection was performed using a mouse IgG ELISPOT kit (MABTech, Stockholm, Sweden). In brief, prewetted poly- Flow cytometry vinylidene difluoride plates were coated with 5 mg HBsAg in sterile PBS overnight at 4˚C. The plates were blocked by adding 200 ml/well RPMI Cells in a single-cell suspension were incubated with anti-CD16/32 (2.4G2; 1640 medium containing 10% FCS for at least 1 h. Then, 5 3 105 BD Biosciences, San Jose, CA) to block any nonspecific staining before splenocytes in suspension (prepared under sterile conditions) were added staining with the specific Abs. Foxp3 and Ki67 were stained using a Foxp3 to the wells and incubated for an additional 24 h at 37˚C without any move- staining kit (eBioscience, San Diego, CA). In brief, cells were fixed, ment of the plate. After incubation, detection was performed according to permeabilized, and then stained with Alexa 647–anti-Foxp3 or Alexa 647– the manufacturer’s instructions. HBsAg- or HBcAg-specific IFN-g–secreting anti-Ki67 (eBioscience). For intracellular IFN-g and TNF-a staining, cells CD4+ T cells were detected using a mouse IFN-g ELISPOT kit (Dakewe, were stimulated with 30 ng/ml PMA (Sigma, St. Louis, MO) and 1 mg/ml Shenzhen, China). In brief, 1 3 106 irradiated (20 Gy) syngeneic spleno- ionomycin (Sigma) for 4 h. Monensin (2 mg/ml; Sigma) was added at the cytes were seeded into 96-well plates precoated with anti–IFN-g in the beginning of stimulation to prevent cytokine secretion. For staining the presence of 5 mg/ml HBsAg or 2 mg/ml HBcAg (ID LABS, London, Ag-specific CD8+ T cells, HBsAg (ILSPFLPLL) and HBcAg (MGLKFRQL) Canada) for 1 h at 37˚C. Then, 1 3 105 purified splenic CD4+ T cells were peptide-specific H2-Kb pentamers (Proimmune, Oxford, U.K.) were used added to the wells and incubated for 24 h. ELISPOT detection was then according to the manufacturer’s instructions. The following fluorescently performed according to the manufacturer’s instructions. All ELISPOT data conjugated mAbs were used in this study: FITC–anti-GL7, FITC–anti-PD1, were obtained and imaged by the plate-reading service provided by the FITC–anti-CD25, PE–anti–TNF-a, PE–anti–IFN-g, PE–anti-Fas, PE–anti- Dakewe Company. Tim3, PE–anti-2B4, PE–anti-Lag3, PE–anti-CD49d, PerCP–Cy5.5–anti-CD8,

by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. FIGURE 2. IL-12–based vaccination therapy reversed HBV tolerance by inducing anti-HBV adaptive immunity. (A) Serum HBsAg levels and (B) the ratios of HBsAg-seropositive mice were analyzed in HBV-carrier mice after IL-12–based vaccination therapy, IL-12 monotherapy, HBV vaccine therapy, or no treatment. HBsAg emulsiﬁed with alum replaced http://classic.jimmunol.org HBsAg vaccine, and alum was also added to the IL-12 monotherapy group. (C) HBV-carrier mice that received IL-12–based vaccination therapy, IL- 12 monotherapy, or HBsAg vaccine were reinjected with pAAV/HBV1.2 plasmid (5 mg) 6 wk after treatment Downloaded from initiation. Serum HBsAg levels were detected 1 wk after injection. (D)A total of 1 3 107 splenocytes from HBV-carrier mice that received either IL-12–based vaccination therapy or HBsAg vaccine were adoptively transferred into HBV-carrier Rag12/2 mice 6 wk after treatment initiation. Serum HBsAg levels in recipient mice were measured after 2 wk. (A)Dataare shown as the mean 6 SEM (unpaired t test). (C and D) Each point represents one mouse. All experiments were repeated twice with similar results. N.S., No signiﬁcant difference. The Journal of Immunology 4187

PerCP–Cy5.5–anti-B220, PE–Cy7–anti-NK1.1, PE–Cy7–anti-CD11a, allo- formed to analyze the differences between two groups. For analyzing the phycocyanin–anti-CD3, allophycocyanin-streptavidin, allophycocyanin– ratio of HBsAg-seropositive mice, the log rank test was used to determine Cy7–anti-CD4, allophycocyanin–Cy7–anti-CD19, Biotin-CXCR5 (all from statistical differences. Differences achieving a p value ,0.05 were con- BD Biosciences), and PE-Helios (Biolegend). All data were collected sidered statistically significant (*p , 0.05, **p , 0.01, ***p , 0.001). using a BD LSR II flow cytometer (BD Biosciences) and analyzed using FlowJo 7.6 software. Results [3H]thymidine incorporation and Treg induction IL-12–based vaccination therapy efficiently eliminated HBV in HBV-tolerant mice For evaluating HBsAg-specific proliferation, mouse splenocytes were separated and cultured (2 3 105 cells/well) in a 96-well U-bottom plate Previous studies showed that mice hydrodynamically injected with 22 with 200 ml RPMI 1640 (supplemented with 10% FCS, 5 3 10 mM 2- HBV plasmid represented a model of chronic HBV infection (12). ME, 2 mM L-glutamine, and 1% penicillin/streptomycin) for 72 h in the Approximately 30% of these mice showed very slow spontaneous presence of 5 mg/ml HBsAg. An unstimulated control was incubated under the same conditions in the absence of Ag. [3H]thymidine (TdR) (1 mCi/well; viral clearance and exhibited both persistent HBV replication in Amersham Pharmacia Biotech, Amersham, U.K.) was added during the the liver and HBV surface antigenemia in the serum for .4 mo; last 16 h for optimal incorporation. Data were collected using b-liquid these mice could be screened out 6 wk after plasmid injection scintillation analyzer (Perkin Elmer, Norwalk, CT) and expressed as a stim- based on the high levels of HBsAg in serum (Supplemental Fig. ulation index (SI = cpm of HBsAg-stimulated cells/cpm of the unstimulated control). To evaluate the suppressive effect of Tregs, we cocultured 1 3 105 1A–D). More importantly, these mice lost the ability to produce CD4+CD252 cells with CD4+CD25+ cells at a ratio of 4:1 or 2:1 in the anti-HBs after conventional HBV vaccination (Supplemental Fig. presence of anti-CD3 (2 mg/ml) and anti-CD28 (1 mg/ml) for 72 h. [3H]TdR 1E, 1F), indicating that they were systemically tolerant to HBV incorporation was determined as described earlier. For Treg induction stimulation. Thus, we called them HBV-carrier mice to mimic the 3 5 + 2 in vitro, a total of 2 10 CD4 CD25 splenocytes was cultured in the acquired tolerant state of human chronic HBV carriers. We then presence of anti-CD3 (2 mg/ml), anti-CD28 (1 mg/ml), TGF-b (5 ng/ml; Peprotech), and IL-2 (100 U/ml; Huaxin, Shanghai, China) for 96 h. tested whether combined treatment with IL-12 could reverse the tolerant state toward peripheral HBV vaccination (termed IL-12– Statistical analysis based vaccination therapy; Fig. 1A). As shown in Fig. 1B, serum Statistical differences were first determined using a one-way ANOVA test to HBsAg levels dramatically decreased in HBV-tolerant mice after 3 compare more than two groups. Nonpaired Student t tests were then per- wk of IL-12–based vaccination therapy. Eight weeks after IL-12– by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 3. IL-12–based vaccination therapy reversed the tolerant HBsAg-specific humoral immunity. HBV-carrier mice were treated with IL-12 monotherapy, IL-12–based vaccination therapy, or vaccine alone. Three weeks later, (A) follicular Th cells (CXCR5+PD-1+) and (B) GC B cells (B220+GL- 7+Fas+) were identified in the dLN (inguinal lymph node) of HBV-carrier mice, gated on CD44hiCD4+ T and B220+CD19+ B cells, respectively. Experiments were repeated three times with similar results. (C) HBsAg-specific IgG Ab-forming cell (AFC) frequency was detected using an HBsAg- specific IgG ELISPOT 4 wk after IL-12–based vaccination therapy initiation. Data represent three mice per group, and experiments were repeated twice with similar results. (D) Serum anti-HBs levels were evaluated at week 8. Data represent the pooled results from two separate experiments; each group comprised five to eight mice. (E) Splenocytes were isolated from HBV-carrier mice at week 4 after the indicated treatment. HBsAg-specific proliferation was determined using [3H]TdR incorporation. Data represent the pooled results from three independent experiments and are expressed as the SI, where SI = CPM of HBsAg-stimulated cells/CPM of unstimulated cells. N.S., No significant difference. 4188 IL-12 REVERSES HBV TOLERANCE

based vaccination therapy, almost all treated mice (.90%) be- vaccination-treated mice adoptively transferred into immunodefi- came negative for the presence of serum HBsAg, whereas this cient HBV-carrier Rag12/2 recipients eliminated HBV (Fig. 2D), complete loss of HBsAg occurred in only a few control mice indicating that a recall HBV-specific immune response was suc- (,20%; Fig. 1C). High HBeAg levels and HBV DNA titers in the cessfully established in HBV-tolerant mice after IL-12–based serum are generally considered to be markers of active HBV vaccination therapy. replication in the liver (2, 3). In our model, we found that IL-12– based vaccination therapy led to remarkably decreased levels of IL-12–based vaccination therapy reversed the tolerant state serum HBeAg (Fig. 1D) and HBV DNA (Fig. 1E), indicating a dra- toward peripheral HBsAg vaccination matic inhibition of active HBV replication. Furthermore, hepatocyte- The dominant contribution of HBsAg vaccine to HBV clearance expressed HBcAg disappeared after IL-12–based vaccination ther- during IL-12–based vaccination therapy prompted us to evaluate apy (Fig. 1F). In addition, the relatively low IL-12 dose used for this the immune responses against HBsAg vaccine in HBV-carrier therapeutic approach did not induce any obvious liver injury or mice. We observed that IL-12–based vaccination therapy reversed toxicity in any other organ (Fig. 1G and data not shown), sug- tolerance to peripheral HBsAg vaccination by promoting HBV- gesting that this therapy was noncytotoxic and safe. Intriguingly, specific CXCR5+PD-1+ follicular helper T cell accumulation in IL-12–based vaccination therapy preferentially exerted a stronger the draining lymph node (dLN) (Fig. 3A), which was essential for ability to reduce serum HBsAg levels than combination therapy germinal center (GC) formation (21). Consistent with this finding, with IL-2, IL-15, or IFN-g (Supplemental Fig. 2). Taken together, GC B cells proliferated significantly more in both the dLN and IL-12–based vaccination therapy successfully and safely reversed spleen (Fig. 3B and data not shown) in response to IL-12–based HBV tolerance in HBV-carrier mice. vaccination therapy compared with HBV vaccine or IL-12 alone. Because GC B cell amplification usually causes Ag-specific Ab IL-12–based vaccination therapy eliminated HBV by inducing production and memory responses (21), we examined serum anti- anti-HBV adaptive immunity HBs levels in HBV-carrier mice. As shown in Fig. 3C, HBsAg- We observed in our experiments that although IL-12 treatment specific, IgG-producing cells were dramatically enhanced after alone (termed IL-12 monotherapy) induced a rapid decrease in IL-12–based vaccination therapy, and most mice receiving IL-12– serum HBsAg levels during the first 1–2 wk, it did not induce loss based vaccination therapy became anti-HBs+ in serum at week 8 of serum HBsAg thereafter in HBV-tolerant mice (Supplemental (Fig. 3D). To confirm that HBsAg-specific T cell responses were Fig. 3A, 3B). Also, neither HBsAg nor IL-12 administered with recovered in HBV-carrier mice after IL-12–based vaccination alum were able to achieve similar therapeutic effects as IL-12– therapy, we performed an in vitro HBsAg-specific [3H]TdR in- based vaccination therapy (with alum; Fig. 2A, 2B), indicating corporation experiment. Only splenocytes from HBV-carrier mice that HBsAg rather than the alum adjuvant played a critical role in treated with IL-12–based vaccination therapy proliferated in response HBV clearance from tolerant mice. In accordance with this, IL- to HBsAg stimulation, although this proliferation was relatively 12–based vaccination therapy, but not IL-12 monotherapy or weak (Fig. 3E). Taken together, these findings clearly indicated HBsAg vaccine alone, successfully prevented HBV reconstitution that IL-12–based vaccination therapy could restore host respon- in HBV-carrier mice challenged with a second HBV plasmid in- siveness toward peripheral HBsAg vaccination in HBV-tolerant

FIGURE 4. Adaptive immune cells were indispensable for HBV clearance in HBV-carrier mice during IL-12– based vaccination therapy. (A and B) Wild-type HBV-carrier mice were trea-

http://classic.jimmunol.org ted with IL-12–based vaccination therapy, vaccine alone, or no treatment. Three weeks later, hepatic mononuclear cells were harvested. (A) Frequency and (B) absolute number of hepatic CD32 NK1.1+ cells (NK), CD3+NK1.1+ cells (NKT), B220+ cells (B), CD3+NK1.12 + + + 2 Downloaded from CD4 cells (CD4 T), and CD3 NK1.1 CD8+ cells (CD8+ T) were calculated. (C)Rag12/2 and (D)CD42/2 mice were injected with 5 mg pAAV/HBV1.2 plasmid and then treated with IL-12–based vaccination therapy, IL-12 monotherapy, or vaccine control 6 wk later. Serum HBsAg levels were monitored. Data are shown as the mean 6 SEM. (A and B) Experiments were repeated 3 times, with each group comprising 7–10 mice; (C and D) experiments were repeated twice, with each group comprising 3–5 mice. *p , 0.05, **p , 0.01, ***p , 0.001. N.S., No signiﬁcant difference. The Journal of Immunology 4189

IL-12–based vaccination therapy reinvigorated the systemic CD11ahi markers to label and monitor Ag-specific CD8+ or CD4+ HBV-specific CD4+ T cell response T cells, respectively, during viral infection (22, 23). Indeed, we observed that the kinetics of both peripheral CD11ahiCD8aloCD8+ In our study, IL-12–based vaccination therapy induced systemic hi hi + lymphocytosis, especially in the livers of HBV-carrier mice. In- T cells and CD49d CD11a CD4 T cells after high-dose HBV vestigating each specific subset individually, we found that the plasmid injection (mimicking acute infection) were similar to frequency and absolute number of hepatic adaptive immune cells, classical Ag-specific T cell kinetics during viral infection (Sup- including CD4+ T, CD8+ T, and B cells, strikingly increased plemental Fig. 4A, 4B) (24). Notably, low-dose HBV plasmid compared with mice treated with HBV vaccine alone or mice that injection (mimicking HBV tolerance) did not result in such sig- did not receive treatment, which was in contrast with the observed nificantly upregulated integrin markers on T cells from the PBL, decrease in predominant liver-resident innate immune cells, in- implying that the HBV-carrier mice lost the ability to elicit strong cluding NK and NKT cells (Fig. 4A, 4B). These results suggested HBV-specific T cell responses in the periphery (data not shown). + that T cell responses might play a critical role in reversing HBV We sorted CD8 T cells from HBV-carrier mice, separated them hi lo lo hi tolerance. Consistent with these findings, IL-12–based vaccination into CD11a CD8a and CD11a CD8a subsets, and transferred therapy lost the ability to eliminate HBV in HBV-carrier Rag12/2 them separately into a naive recipient. As expected, only the mice deficient in both T and B cells, but it still reduced HBV to CD11ahiCD8alo donor cells robustly proliferated in response to a similar level as IL-12 monotherapy, further suggesting that HBV rechallenge (Supplemental Fig. 4C, 4D), indicating that the adaptive immunity was required to eradicate HBV, whereas HBV CD11ahiCD8alo cells were HBV specific. Therefore, CD11ahi inhibition mediated by IL-12 alone seemed to be independent of CD8alo or CD49dhiCD11ahi markers could be used to monitor adaptive immunity (Fig. 4C). A similar phenomenon was observed HBV-specific CD8+ or CD4+ T cells in our model. in HBV-carrier CD42/2 mice (Fig. 4D), indicating that CD4+ CD4+ T cell responses to HBV are usually very weak in chronic T cells were indispensable for complete HBV clearance during IL- HBV-infected individuals and are considered to impede induction 12–based vaccination therapy. and maintenance of HBV-specific CD8+ T cell and Ab responses To further evaluate HBV-specific T cell responses, we used (3, 4). However, in our mouse model, HBV-specific CD4+ T cells a recently reported approach tracking CD11ahiCD8alo or CD49dhi from the PBL (Fig. 5A), liver, and spleen (Fig. 5B) dramatically by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 5. IL-12–based vaccination therapy restored systemic HBV-specific CD4+ T cell responses in HBV-carrier mice. HBV-carrier mice were treated with IL-12–based vaccination therapy, vaccine alone, or no treatment. Three weeks later, hepatic mononuclear cells and splenocytes were harvested.(A)CD49dhi CD11ahi frequency among CD4+ T cells in PBL was monitored weekly; (B) these cells were also identified in liver (left)andspleen(right), gated on CD4+ T cells. (C) Ki67 staining of splenic CD49dloCD11aloCD4+ T (Non-HBV) or CD49dhiCD11ahiCD4+ T cells (HBV) are shown. (D)KLRG1and(E) CD62L staining of splenic CD49dhiCD11ahi CD4+ T (HBV-specific) are shown. Intracellular (F)IFN-g and (G)TNF-a were examined using flow cytometry, gated on splenic CD49dhiCD11ahiCD4+ T (HBV-specific) and CD49dloCD11aloCD4+ T cells (HBV-nonspecific). Splenic CD4+ T cells from mice in all three groups were separated by MACS at week 3, and (H)HBcAg-and(I) HBsAg-specific IFN-g ELISPOT assays were performed. Results are expressed as the numbers of Ag- specific IFN-g–secreting spot-forming cells (SFC), where Ag-specific SFC numbers = SFCs in the presence of Ag stimulation 2 SFCs in the absence of Ag stimulation. Data are representative of three independent experiments. (A, H,andI)Dataareshownasthemean6 SEM. **p , 0.01, ***p , 0.001. 4190 IL-12 REVERSES HBV TOLERANCE

increased3wkafterIL-12–based vaccination therapy, mainly to be the induced Tregs (26, 27), were reduced after IL-12–based because of the selective expansion of HBV-specific CD4+ T cells vaccination therapy (Fig. 6C). In addition, CD4+CD252 T cells (Fig. 5C). Furthermore, HBV-specific CD4+ T cells from mice from HBV-carrier mice treated with IL-12–based vaccination ther- treated with IL-12–based vaccination therapy expressed higher apy were refractory to TGF-b–induced Foxp3 expression in vitro KLRG1 and lower CD62L levels (Fig. 5D, 5E), a phenotype (Fig. 6D); moreover, we also found that these CD4+CD252 effector representing effector/memory T cells (25). In line with this phe- T cells proliferated to a greater extent than cells from vaccine- notype, splenic HBV-specific CD4+ T cell effector function, assessed treated control mice in the presence of Tregs (Fig. 6E), indicating by IFN-g and TNF-a secretion, was greatly enhanced after IL-12– that IL-12–based vaccination therapy induced effector T cells to based vaccination therapy as compared with HBV vaccine alone overcome Treg-mediated inhibition. Consistent with these find- or untreated controls (Fig. 5F, 5G). Specifically, both HBcAg- and ings, splenocytes isolated from HBV-carrier mice completely in- HBsAg-specific IFN-g–secreting CD4+ T cells were greatly inhibited HBsAg-specific proliferation of splenocytes from HBsAg- creased after IL-12–based vaccination therapy (Fig. 5H, 5I), indi- vaccinated naive mice in vitro, whereas splenocytes from IL-12– cating that a multispecific and robust anti-HBV CD4+ T cell response based vaccination-treated HBV-carrier mice did not inhibit the was recovered. Taken together, these data showed that IL-12–based Ag-specific proliferation of these cells (Fig. 6F). These results vaccination therapy could reinvigorate systemic HBV-specific CD4+ suggest that IL-12–based vaccination therapy diminishes the fre- T cell responses in HBV-tolerant mice. quency of Ag-specific inhibitory cells that normally exist in HBV- carrier mice. IL-12–based vaccination therapy inhibited Treg generation + + IL-12–based vaccination therapy induced robust hepatic The balance between effector/memory CD4 and regulatory CD4 + T cells may be critical for the outcome of immune responses. As HBV-specific CD8 T cell responses expected, IL-12–based vaccination therapy reduced the frequency A robust HBV-specific CD8+ T cell response has previously been of CD4+Foxp3+ or CD4+CD25+ Tregs in both the liver and spleen reported to be critical for HBV elimination in the clinic (3). In our of HBV-carrier mice (Fig. 6A, 6B). We supposed that the reduced study, HBV-specific CD8+ T cells actually responded more quickly Treg frequency was due to the inhibition of Treg induction. In- and robustly than CD4+ T cells. HBV-carrier mice receiving IL-12– deed, the Helios2 cells among the Foxp3+ Tregs, which appeared based vaccination therapy showed rapidly increased CD11ahi by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 6. IL-12–based vaccination therapy inhibited Treg generation. HBV-carrier mice were treated with IL-12–based vaccination therapy or vaccine alone. Three weeks later, CD4+Foxp3+ (A) and CD4+CD25+ (B) frequencies among CD3+NK1.12 cells in spleen (left) and liver (right) were detected by ﬂow cytometry. Data are representative of three independent experiments. (C) The frequency of Helios2 cells among CD4+Foxp3+ cells in liver and spleen are shown. (D) CD4+CD252 T cells from the spleens of mice treated with IL-12–based vaccination therapy or vaccine alone were cultured in the presence or absence of TGF-b and IL-2 to induce Tregs. Cells were harvested 96 h later, and Foxp3 expression was determined. (E) CD4+CD25+ cells from untreated HBV-carrier mice were considered regulatory cells and cocultured with CD4+CD252 effector cells from HBV-carrier mice treated with either IL-12–based vaccination therapy or vaccine alone in the presence of anti-CD3 and anti-CD28. Proliferative responses were measured using [3H]TdR incorporation. (F) Naive C57BL/6 mice were vaccinated twice with HBsAg vaccine (5 mg) at a 2-wk interval. Splenocytes from HBV-vaccinated mice (anti-HBs) were harvested as effector cells 3 d after the second immunization and cocultured for 72 h with splenocytes from HBV-carrier mice, which had been treated with either IL-12–based vaccination therapy or vaccine alone, in the presence or absence of 5 mg/ml HBsAg. Proliferation was measured using [3H]TdR incorporation. Data are shown as Dcpm, where Dcpm = cpm of HBsAg-stimulated cells 2 cpm of unstimulated cells. (D–F) Experiments were repeated twice with similar results. The Journal of Immunology 4191

CD8aloCD8+ T cells in PBL, liver, and spleen, reaching a peak at world. In previous studies, several vaccination strategies attemp- week 3 (Fig. 7A, 7B), which coincided with a rapid decline in ted to break systemic HBV tolerance in an HBV-tg mouse model HBV (Fig. 1). Among these HBV-specific effector/memory cells, (8, 14, 30, 31); however, because HBV DNA integrates into the both HBsAg-specific (208–216) and HBcAg-specific (93–100) genome of mouse embryos in this model, their immune systems CD8+ T cells were dramatically increased after IL-12–based vacci- become nonresponsive (i.e., tolerant) to HBV because of central nation therapy (Fig. 7C, 7D). Moreover, similar to CD4+ T cells clonal deletion of HBV-specific T and B cells, which does not (Fig. 5), IL-12–based vaccination therapy also selectively induced reflect the acquired tolerant state in most human CHB patients. proliferation (Fig. 7E) and activation of HBV-specific CD11ahi Because HBV is unable to infect new hepatocytes in our HBV- CD8aloCD8+ T cells rather than nonspecific CD8+ T cells (Fig. carrier mouse model, no immunopathology or innate immune 7F, 7G). Interestingly, this CD8+ T cell stimulation effect only signals can occur, and low HBV titers are observed in HBV-carrier partially relied on IL-12, because HBV-carrier mice treated with mice; these aspects of the model more closely resemble human IL-12 alone exhibited a lower increase of HBV-specific effector/ asymptomatic chronic HBV carriers, who usually do not require memory CD8+ T cells than those treated with IL-12–based vacci- treatment, than CHB patients (2, 3). However, the Ag-specific nation therapy (Supplemental Fig. 3C–E). This result also indicated systemic tolerant state is similar in both human HBV carriers that a synergistic effect of IL-12 and HBsAg vaccine in stimulating and our HBV-carrier mice, which is mainly induced by persistent HBV-specific CTL cells. Notably, IL-12–based vaccination therapy HBV expression in liver and mediated by the “liver tolerance triggered IFN-g secretion from both hepatic HBV-specific and non– effects,” such as induction of Ag-specific regulatory cells, pe- HBV-specific CD8+ T cells (data not shown), resulting in an overall ripheral T cell clonal deletion, and anergy (32). Thus, our findings 5- to 8-fold increase of hepatic IFN-g–secreting CTL cells over that IL-12–based vaccination therapy can reverse liver-induced HBV vaccine alone (Fig. 7H). Given that CTLs are the major effec- acquired HBV tolerance may be instructive for therapeutic vac- tor cells for HBV clearance through IFN-g–mediated noncytotoxic cine design against CHB carriers. effects (28, 29), we speculated that the robust hepatic CD8+ T cell Although IL-12 monotherapy can significantly reduce HBV in responses induced by IL-12–based vaccination therapy dramati- HBV-carrier mice, these mice still persistently express HBV after cally reduced HBV titers, which could be beneficial for reversing IL-12 withdrawal and fail to rapidly eradicate virus upon “recar- systemic tolerance and eradicating HBV. rying” HBV (Fig. 2A–C, Supplemental Fig. 3A, 3B). These data suggest that the therapeutic effect of IL-12 monotherapy mainly Discussion relies on the transient and nonspecific viral inhibition effect of IL- Liver-induced systemic tolerance upon chronic infection poses 12 (18) rather than on triggering long-term anti-HBV–specific im- a great challenge to the development of therapeutic vaccine, es- munity through reversing intrinsic immune tolerance. In contrast pecially for CHB, which affects .350 million people around the with IL-12 monotherapy, vaccination therapy was originally de- by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from

FIGURE 7. IL-12–based vaccination therapy induced robust hepatic HBV-specific CD8+ T cell responses in HBV-carrier mice. HBV-carrier mice were treated with IL-12–based vaccination therapy, vaccine alone, or no treatment. Three weeks later, hepatic mononuclear cells and splenocytes were harvested. (A) CD11ahiCD8alo cell frequency among CD8+ T cells in PBL was monitored weekly. (B) CD11ahiCD8alo cells were identified in liver (left) and spleen (right), gated on CD8+ T cells. The frequencies of (C) HBcAg- and (D) HBsAg-specific CD8+ T cells among splenic CD192NK1.12CD3+CD8+ cells were determined by staining with HBcAg or HBsAg peptide–coupled H2kb pentamers. (E) Ki67 staining of splenic CD11ahiCD8alo cells (right, HBV-specific) and CD11aloCD8ahi cells (left, non–HBV-specific) is shown. (F) KLRG1 and (G) CD62L staining of splenic CD11ahiCD8ahiCD8+ T cells (HBV-specific) are shown. (H) Hepatic lymphocytes were stimulated with PMA and ionomycin in vitro for 4 h in the presence of monensin, and intracellular IFN-g was examined by flow cytometry. The numbers of total IFN-g+ hepatic CD8+ T cells were calculated. (A) Data are shown as the mean 6 SEM. (B–H) Each plot represents one mouse. Experiments were repeated three times with similar results. *p , 0.05, **p , 0.01, ***p , 0.001. 4192 IL-12 REVERSES HBV TOLERANCE

signed to induce long-term Ag-specific immunity to completely tion of Ag-specific T cells, which exhibit upregulated Tim3, PD1, control the virus, but vaccine monotherapy was inefficient in Lag3, and 2B4 expression, as well as downregulated effector HBV-carrier mice because it was not strong enough to reverse the molecule expression (40). This could be the fundamental reason to systemic tolerance already induced in the host by HBV. However, explain the failure of IL-12 monotherapy in clinic. However, in during IL-12–based vaccination therapy, a series of host factors our model, IL-12 was administered at a relatively low dose for altered by IL-12, such as reduced viral titers, extensive immune breaking tolerance to peripheral HBsAg vaccination, and none of activation, and proinflammatory cytokine milieu, facilitated the the earlier-mentioned inhibitory receptors, except 2B4, were up- vaccine to become more effective in mounting robust anti-HBV regulated on HBV-specific T cells during IL-12 therapy (data not immunity by reversing tolerance, subsequently leading to the shown), although the functional importance of this 2B4 expression elimination of HBV from the host. Notably, although a previous on T cells remains to be defined. Therefore, we propose that using study showed that a DNA vaccine containing a mutant IL-12 and IL-12 to “help” rather than “kill” T cell responses may be a more HBV expression plasmids could induce long-term HBV-specific rational and efficient use of IL-12 when applying IL-12–based CD4+ effector/memory cells in some lamivudine-treated CHB immunotherapy. patients, the effect of IL-12 in this study was obscure (33), be- Overall, our results demonstrate that the synergistic effect of IL- cause serum IL-12 and IFN-g levels were undetectable during IL- 12 and HBV vaccine can overcome liver-induced systemic immune 12/HBV vaccine combination therapy (33). Thus, our study pro- tolerance in HBV-carrier mice, leading to HBV clearance. Thus, vides more mechanistic insights into how the IL-12 protein this approach might be a promising immunotherapeutic strategy for functions synergistically with the vaccine to break the acquired CHB therapy. Furthermore, to our knowledge, this is the first study tolerance state in HBV carriers. Interestingly, even in HBV-tg to explore the role of IL-12 in breaking hepatic Ag-induced sys- mice that had central tolerance toward HBV, IL-12–based vacci- temic immune tolerance, and the therapeutic implications for this nation therapy could still induce more anti-HBs production and role of IL-12 can be extended to provide immunotherapy for other further reduce HBV DNA than vaccine or IL-12 alone (data not hepatic chronic infections or tumors when combined with a vaccine. shown), further demonstrating the unexpected effect of IL-12 in reversing T cell tolerance when used in combination with vaccine. Disclosures An essential role of IL-12 in our model is the induction of robust The authors have no financial conflicts of interest. T cell responses, as our experimental evidence shows that both HBV-carrier Rag12/2 and HBV-carrier CD42/2 mice cannot eradicate HBV after IL-12–based vaccination therapy (Fig. 4C, 4D), and References + that hepatic HBV-specific CD8 T cell responses are dramatically 1. Crispe, I. N. 2009. The liver as a lymphoid organ. Annu. Rev. Immunol. 27: 147– enhanced during this therapy (Fig. 7, Supplemental Fig. 3D, 3E). 163. 2. Ganem, D., and A. M. Prince. 2004. Hepatitis B virus infection—natural history These results are consistent with previous reports demonstrating and clinical consequences. N. Engl. J. Med. 350: 1118–1129. + that CD8 T cells are the main effectors of IFN-g–mediated 3. Chisari, F. V., M. Isogawa, and S. F. Wieland. 2010. Pathogenesis of hepatitis B noncytolytic effects during HBV clearance in both mouse models virus infection. Pathol. Biol. (Paris) 58: 258–266. + 4. Rehermann, B., and M. Nascimbeni. 2005. Immunology of hepatitis B virus and and HBV-infected chimpanzees (28, 29). HBV-specific CD4 hepatitis C virus infection. Nat. Rev. Immunol. 5: 215–229.

by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. T cells were functionally impaired in HBV-carrier mice (Z. Zeng, 5. Pol, S., and M. L. Michel. 2006. Therapeutic vaccination in chronic hepatitis B H. Wei, R. Sun, and Z. Tian, unpublished observations), suggesting virus carriers. Expert Rev. Vaccines 5: 707–716. + 6. Michel, M. L., and P. Tiollais. 2010. Hepatitis B vaccines: protective efficacy and that systemic HBV-specific CD4 Th responses might need to re- therapeutic potential. Pathol. Biol. (Paris) 58: 288–295. cover to restore responses to HBsAg vaccination in HBV carriers. 7. Deńy, P., and F. Zoulim. 2010. Hepatitis B virus: from diagnosis to treatment. Pathol. Biol. (Paris) 58: 245–253. Indeed, both the number and effector function of HBV-specific 8. Mancini, M., M. Hadchouel, P. Tiollais, C. Pourcel, and M. L. Michel. 1993. + CD4 T cells were significantly augmented during IL-12–based Induction of anti-hepatitis B surface antigen (HBsAg) antibodies in HBsAg vaccination therapy (Fig. 5). This strong T cell stimulation capacity producing transgenic mice: a possible way of circumventing “nonresponse” to HBsAg. J. Med. Virol. 39: 67–74. of IL-12 could be beneficial for reversing the liver-induced clonal 9. Akbar, S. M., K. Kajino, K. Tanimoto, K. Kurose, T. Masumoto, K. Michitaka, deletion/anergy of Ag-specific T cells. In addition, this therapy N. Horiike, and M. Onji. 1997. Placebo-controlled trial of vaccination with + + hepatitis B virus surface antigen in hepatitis B virus transgenic mice. J. Hepatol. http://classic.jimmunol.org reduced not only hepatic (34), but also splenic CD4 Foxp3 Treg 26: 131–137. frequency (Fig. 6), which was previously shown to partially 10. Pol, S., B. Nalpas, F. Driss, M. L. Michel, P. Tiollais, J. Denis, and C. Brećho; contribute to HBV-induced tolerance within CHB patients and Multicenter study group. 2001. Efficacy and limitations of a specific immunotherapy in chronic hepatitis B. J. Hepatol. 34: 917–921. asymptomatic carriers (35, 36). Meanwhile, in accordance with 11. Dikici, B., M. Bosnak, H. Ucmak, A. Dagli, A. Ece, and K. Haspolat. 2003. + a previous in vitro study (37), CD4 T cells from mice treated with Failure of therapeutic vaccination using hepatitis B surface antigen vaccine in the IL-12–based vaccination therapy seemed to be more resistant to immunotolerant phase of children with chronic hepatitis B infection. J. Gas- troenterol. Hepatol. 18: 218–222.

Downloaded from Treg-mediated suppression, perhaps providing another mechanism 12. Huang, L. R., H. L. Wu, P. J. Chen, and D. S. Chen. 2006. An immunocompetent for the efficient reversal of immune tolerance by IL-12–based mouse model for the tolerance of human chronic hepatitis B virus infection. vaccination therapy. Proc. Natl. Acad. Sci. USA 103: 17862–17867. 13. Takehara, T., T. Suzuki, K. Ohkawa, A. Hosui, M. Jinushi, T. Miyagi, T. Tatsumi, We speculate that the IL-12–induced cytokine milieu at a pe- Y. Kanazawa, and N. Hayashi. 2006. Viral covalently closed circular DNA in ripheral site could also be beneficial for reversing tolerance. On a non-transgenic mouse model for chronic hepatitis B virus replication. J. Hepatol. 44: 267–274. one hand, IL-12–induced Th1 cytokines, including high IFN-g 14. Malanche`re-Bre`s, E., P. J. Payette, M. Mancini, P. Tiollais, H. L. Davis, and and TNF-a levels (data not shown), might overcome suppression M. L. Michel. 2001. CpG oligodeoxynucleotides with hepatitis B surface antigen mediated by inhibitory cytokines, such as IL-10 and TGF-b.On (HBsAg) for vaccination in HBsAg-transgenic mice. J. Virol. 75: 6482–6491. 15. Vandepapelie`re, P., G. K. Lau, G. Leroux-Roels, Y. Horsmans, E. Gane, the other hand, IL-12 therapy also induced significantly increased T. Tawandee, M. I. Merican, K. M. Win, C. Trepo, G. Cooksley, et al; Thera- levels of serum IL-6, a cytokine essential for follicular Th cell peutic HBV Vaccine Group of Investigators. 2007. Therapeutic vaccination of differentiation in mice (38) and viral control during chronic chronic hepatitis B patients with virus suppression by antiviral therapy: a ran- domized, controlled study of co-administration of HBsAg/AS02 candidate LCMV infection (39), especially in the later phase of therapy (e.g., vaccine and lamivudine. Vaccine 25: 8585–8597. at week 3, data not shown), which likely promoted Ab production 16. Trinchieri, G. 2003. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat. Rev. Immunol. 3: 133–146. in response to peripheral vaccination. Recent studies show that 17. Beckebaum, S., V. R. Cicinnati, X. Zhang, S. Ferencik, A. Frilling, H. Grosse- continuous strong IL-12 stimulation results in functional exhaus- Wilde, C. E. Broelsch, and G. Gerken. 2003. Hepatitis B virus-induced defect of The Journal of Immunology 4193

monocyte-derived dendritic cells leads to impaired T helper type 1 response 31. Schirmbeck, R., N. Dikopoulos, M. Kwissa, F. Leithaüser, K. Lamberth, S. Buus, in vitro: mechanisms for viral immune escape. Immunology 109: 487–495. K. Melber, and J. Reimann. 2003. Breaking tolerance in hepatitis B surface 18. Cavanaugh, V. J., L. G. Guidotti, and F. V. Chisari. 1997. Interleukin-12 inhibits antigen (HBsAg) transgenic mice by vaccination with cross-reactive, natural hepatitis B virus replication in transgenic mice. J. Virol. 71: 3236–3243. HBsAg variants. Eur. J. Immunol. 33: 3342–3352. 19. Carrenõ, V., S. Zeuzem, U. Hopf, P. Marcellin, W. G. Cooksley, J. Fevery, M. Diago, 32. Tiegs, G., and A. W. Lohse. 2010. Immune tolerance: what is unique about the R. Reddy, M. Peters, K. Rittweger, et al. 2000. A phase I/II study of recombinant liver. J. Autoimmun. 34: 1–6. human interleukin-12 in patients with chronic hepatitis B. J. Hepatol. 32: 317–324. 33. Yang, S. H., C. G. Lee, S. H. Park, S. J. Im, Y. M. Kim, J. M. Son, J. S. Wang, 20. Dong, Z., H. Wei, R. Sun, Z. Hu, B. Gao, and Z. Tian. 2004. Involvement of S. K. Yoon, M. K. Song, A. Ambrozaitis, et al. 2006. Correlation of antiviral T- natural killer cells in PolyI:C-induced liver injury. J. Hepatol. 41: 966–973. cell responses with suppression of viral rebound in chronic hepatitis B carriers: 21. Nutt, S. L., and D. M. Tarlinton. 2011. Germinal center B and follicular helper a proof-of-concept study. Gene Ther. 13: 1110–1117. T cells: siblings, cousins or just good friends? Nat. Immunol. 12: 472–477. 34. Crettaz, J., I. Otano, L. Ochoa-Callejero, A. Benito, A. Paneda, I. Aurrekoetxea, 22. McDermott, D. S., and S. M. Varga. 2011. Quantifying antigen-specific CD4 P. Berraondo, J. R. Rodrı´guez-Madoz, A. Astudillo, F. Kreppel, et al. 2009. T cells during a viral infection: CD4 T cell responses are larger than we think. J. Treatment of chronic viral hepatitis in woodchucks by prolonged intrahepatic Immunol. 187: 5568–5576. expression of interleukin-12. [Published erratum appears in 2011 J Virol. 85: 23. Rai, D., N. L. Pham, J. T. Harty, and V. P. Badovinac. 2009. Tracking the total 13475.] J. Virol. 83: 2663–2674. CD8 T cell response to infection reveals substantial discordance in magnitude 35. Stoop, J. N., R. G. van der Molen, C. C. Baan, L. J. van der Laan, E. J. Kuipers, and kinetics between inbred and outbred hosts. J. Immunol. 183: 7672–7681. J. G. Kusters, and H. L. Janssen. 2005. Regulatory T cells contribute to the 24. Kaech, S. M., E. J. Wherry, and R. Ahmed. 2002. Effector and memory T-cell dif- impaired immune response in patients with chronic hepatitis B virus infection. ferentiation: implications for vaccine development. Nat. Rev. Immunol. 2: 251–262. Hepatology 41: 771–778. 25. Zhang, N., and M. J. Bevan. 2012. TGF-b signaling to T cells inhibits auto- 36. Xu, D., J. Fu, L. Jin, H. Zhang, C. Zhou, Z. Zou, J. M. Zhao, B. Zhang, M. Shi, immunity during lymphopenia-driven proliferation. Nat. Immunol. 13: 667–673. X. Ding, et al. 2006. Circulating and liver resident CD4+CD25+ regulatory 26. Thornton, A. M., P. E. Korty, D. Q. Tran, E. A. Wohlfert, P. E. Murray, T cells actively influence the antiviral immune response and disease progression Y. Belkaid, and E. M. Shevach. 2010. Expression of Helios, an Ikaros tran- in patients with hepatitis B. J. Immunol. 177: 739–747. scription factor family member, differentiates thymic-derived from peripherally 37. King, I. L., and B. M. Segal. 2005. Cutting edge: IL-12 induces CD4+CD25- induced Foxp3+ T regulatory cells. J. Immunol. 184: 3433–3441. T cell activation in the presence of T regulatory cells. J. Immunol. 175: 27. Gottschalk, R. A., E. Corse, and J. P. Allison. 2012. Expression of Helios in 641–645. peripherally induced Foxp3+ regulatory T cells. J. Immunol. 188: 976–980. 38. Eto, D., C. Lao, D. DiToro, B. Barnett, T. C. Escobar, R. Kageyama, I. Yusuf, 28. Guidotti, L. G., R. Rochford, J. Chung, M. Shapiro, R. Purcell, and F. V. Chisari. and S. Crotty. 2011. IL-21 and IL-6 are critical for different aspects of B cell 1999. Viral clearance without destruction of infected cells during acute HBV immunity and redundantly induce optimal follicular helper CD4 T cell (Tfh) infection. Science 284: 825–829. differentiation. PLoS ONE 6: e17739. 29. Thimme, R., S. Wieland, C. Steiger, J. Ghrayeb, K. A. Reimann, R. H. Purcell, 39. Harker, J. A., G. M. Lewis, L. Mack, and E. I. Zuniga. 2011. Late interleukin-6 and F. V. Chisari. 2003. CD8(+) T cells mediate viral clearance and disease escalates T follicular helper cell responses and controls a chronic viral infection. pathogenesis during acute hepatitis B virus infection. J. Virol. 77: 68–76. Science 334: 825–829. 30. Sette, A. D., C. Oseroff, J. Sidney, J. Alexander, R. W. Chesnut, K. Kakimi, 40. Yang, Z. Z., D. M. Grote, S. C. Ziesmer, T. Niki, M. Hirashima, A. J. Novak, L. G. Guidotti, and F. V. Chisari. 2001. Overcoming T cell tolerance to the T. E. Witzig, and S. M. Ansell. 2012. IL-12 upregulates TIM-3 expression and hepatitis B virus surface antigen in hepatitis B virus-transgenic mice. J. Immu- induces T cell exhaustion in patients with follicular B cell non-Hodgkin lym- nol. 166: 1389–1397. phoma. J. Clin. Invest. 122: 1271–1282. by guest on October 1, 2021. Copyright 2013 Pageant Media Ltd. http://classic.jimmunol.org Downloaded from 1 Interleukin-12-based vaccination therapy reverses liver-induced systemic tolerance in a mouse model of hepatitis B virus carrier

Zhutian Zeng†, Xiaohui Kong†, Fenglei Li†, Haiming Wei†, §, Rui Sun†, § and Zhigang Tian†, § †Department of Immunology, School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China; §Hefei National Laboratory for Physical Sciences at Microscale, Hefei, Anhui 230027, China

Supplemental Figures & Figure legends

Supplemental Figure 1. Hydrodynamic injection of HBV plasmid induced HBV persistence and tolerance to HBV vaccination in mice. Hydrodynamic injection of pAAV/HBV1.2 into 5-week-old C57BL/6 mice led to HBV persistence. (A) Serum HBsAg levels were detected at different time points after injection. (B) The ratio of serum HBsAg-positive mice was monitored. The pooled results from 4 independent experiments (n=44) are shown. (C) HBV-carrier mice were screened out 6 weeks after HBV plasmid injection and compared with the remaining non-HBV-carrier mice. (D) The ratio of serum HBsAg-positive mice was monitored in both HBV-carrier and non-HBV-carrier mice. HBV-carrier mice were immunized weekly with 2 μg of HBsAg vaccine. Serum (E) HBsAg and (F) HBeAg levels were detected after HBsAg vaccination (n=7) or PBS treatment (n=7) as a control. Data are shown as the mean±SEM. N.S. indicates no significant difference (t-test). Experiments were repeated more than 5 times with similar results.

Supplemental Figure 2. IL-12-based, but not IL2-, IL-15-, or IFNJ-based vaccination therapy preferentially accelerated HBV clearance in HBV-tolerant mice. HBV-carrier mice were treated with different cytokines in combination with HBsAg vaccine. In brief, HBV-carrier mice received PBS, rhIL-2 (5000 U or 25000 U), rhIL-15 (100 ng or 500 ng), rmIFNJ (1 μg), or rmIL-12 (100 ng) daily, followed by immunization with 2 Pg of HBsAg vaccine; treatment was repeated weekly for 3 weeks. (A) Serum HBsAg levels were detected weekly in groups of mice treated with rhIL-2 (5000 U)-, rhIL-15 (100 ng)-, rmIL-12 (100 ng)-, or PBS-based vaccination therapy. (B) Serum HBsAg levels were compared among groups of mice treated with high-dose IL-2 (25000 U)-, low-dose IL-2 (5000 U)-, IL-12 (100 ng)-, or PBS-based vaccination therapy at week 3. (C) Serum HBsAg levels were compared among groups of mice treated with high-dose IL-15 (500 ng)-, low-dose IL-15 (100 ng)-, IL-12 (100 ng)-, or PBS-based vaccination therapy at week 3. (D) Serum HBsAg levels were detected weekly in groups of mice treated with rmIFNJ (1 μg)-, rmIL-12 (100 ng)-, or PBS-based vaccination therapy. Data are shown as the mean±SEM (unpaired t-test). Each group was composed of 6–8 mice. Experiments were repeated with similar results.

Supplemental Figure 3. IL-12 treatment alone induced suboptimal CD8+ T cell responses and did not eliminate HBV in HBV-tolerant mice. (A) Serum HBsAg levels were monitored in HBV-carrier mice treated with IL-12 monotherapy, vaccine alone, or IL-12-based vaccination therapy during and after treatment. Data are shown as the mean±SEM: *p<0.05 (unpaired t-test). (B) HBsAg seroconversion was analyzed in all 3 groups during and after treatment. Statistical comparisons were performed using the Log-rank test. **p<0.01, N.S. no significant difference. Experiments were repeated twice with similar results, with each group comprising 7–10 mice. (C) & (D) HBV-carrier mice were treated with IL-12-based vaccination therapy, IL-12 monotherapy, vaccine alone, or without treatment, and the frequencies of CD11ahiCD49dhiCD4+ T cells and CD11ahiCD8ĮloCD8+ T cells in PBL were monitored during therapy. Data are shown as the mean±SEM: **p<0.01, N.S. no significant difference. Experiments were repeated twice with similar results, with each group comprising 4–6 mice. (E) HBV-carrier mice described above were sacrificed at week 3, splenocytes were harvested, and HBcAg-specific CD8+ T cells were determined by HBcAg pentamer staining (gated on CD19-NK1.1-CD3+CD8+ cells). Data shown are the representative results from 4–6 mice per group.

Supplemental Figure 4. Method to label HBV-specific T cells in HBV-carrier mice. C57BL/6 mice (CD45.2) were injected with 20 ȝg pAAV/HBV1.2 plasmid or pAAV vector. The frequency of (A) CD11ahiCD8alo cells among CD8+ T cells and of (B) CD11ahiCD49dhi cells among CD4+ T cells in PBL were monitored weekly. Data are shown as the mean±SEM, with each group comprising 10–12 mice. (C) Experimental design of (D): Splenic CD8+ T cells from pAAV/HBV1.2-injected mice were sorted into 2 subsets––CD11ahiCD8alo and CD11aloCD8ahi cells––at day 21. After CFSE labeling, 2 × 106 cells from each subset were transferred intravenously into CD45.1 congenic mice. Recipient mice were then hydrodynamically injected with 20 ȝg pAAV/HBV1.2 or control pAAV 12 hours later. (D) CD45.1 recipient mice were sacrificed on day 5 after HBV or vector challenge. CD45.2+ donor cells in liver MNC were gated and analyzed for CFSE dilution. Data represents 3 mice per group.