Currently 247 million people are living with chronic hepatitis B virus infection (CHB), and the development of novel curative treatments is urgently needed. Immunotherapy is an attractive approach to treat CHB, yet therapeutic approaches to augment the endogenous hepatitis B virus (HBV)–specific T cell response in CHB patients have demonstrated little success. In this study, we show that strain 68-1 rhesus macaque (RM) CMV vaccine vectors expressing HBV Ags engender HBV-specific CD8+ T cells unconventionally restricted by MHC class II and the nonclassical MHC-E molecule in RM. Surface staining of human donor and RM primary hepatocytes (PH) ex vivo revealed the majority of PH expressed MHC-E but not MHC class II. HBV-specific, MHC-E–restricted CD8+ T cells from RM vaccinated with RM CMV vaccine vectors expressing HBV Ags recognized HBV-infected PH from both human donor and RM. These results provide proof-of-concept that MHC-E–restricted CD8+ T cells could be harnessed for the treatment of CHB, either through therapeutic vaccination or adoptive immunotherapy.

This article is featured in In This Issue, p.2009

Chronic hepatitis B virus infection (CHB) is a major global health concern, affecting 247 million individuals worldwide and causing 887,000 deaths annually (1). Although there is an effective prophylactic vaccine available, 10–15% of individuals do not respond adequately to vaccination and are not protected against hepatitis B virus (HBV) infection (2). CHB can lead to progressive liver dysfunction, cirrhosis, and in some cases hepatocellular carcinoma. There are multiple treatment options for CHB, including pegylated–IFN-α and reverse-transcriptase inhibitors (3), but these treatments are rarely curative (4). Thus, there is an urgent global need to develop curative therapies for HBV.

Developing cellular immunotherapeutic strategies for CHB is supported by the fact that 90–95% of acutely HBV-infected adults mount broad, highly functional HBV-specific T cell responses and subsequently clear infection (57). In contrast, patients progressing to CHB exhibit narrowly focused, low-frequency, functionally exhausted HBV-specific T cell responses (810). Therefore, many immunotherapeutic strategies currently in development focus on augmenting HBV-specific T cell immunity.

T cell–based immunotherapies for CHB must overcome or circumvent T cell exhaustion and provide sustained viral suppression or clearance. Given the difficulty in reversing the immunological tolerance of established HBV-specific T cells in CHB patients, the most effective way to augment HBV-specific T cell immunity may be to engender or impart a completely unique set of HBV-specific T cell responses through therapeutic vaccination or adoptive T cell therapy. We recently discovered that strain 68-1 rhesus macaques CMV (RhCMV) vectors engineered to express antigenic targets elicit broad, effector-memory CD8+ T cell responses exclusively restricted by either MHC class II (MHC-II) or by MHC-E, a highly conserved, monomorphic, “nonclassical” MHC class Ib molecule (11). Importantly, RhCMV-based vaccines eliciting such unconventionally restricted CD8+ T cells have been shown to protect rhesus macaques (RM) against SIV, tuberculosis, and malaria (1214).

MHC-E (HLA-E in humans) is an MHC class Ib molecule that binds a conserved peptide (VMAPRTL[L/V/I]L, VL9) encoded in the signal sequence of polymorphic MHC class Ia molecules (15). The MHC-E/VL9 complex then serves as a ligand for the NK cell receptors NKG2A and NKG2C (16). Because MHC-E protects against NK cell activity by engaging the inhibitory receptor NKG2A, it is often upregulated by chronic pathogens or in cancer (1719). The unique ability of CMV-based vectors to elicit MHC-E–restricted CD8+ T cells enables the specific targeting of pathogens or cancer cells via MHC-E, thus potentially turning NK cell evasion into a T cell vulnerability.

In this study, we explored the possibility that HBV-infected primary hepatocytes (PH) could be targeted by MHC-E–restricted CD8+ T cells. This innovative strategy is particularly attractive for HBV because such T cells are not known to be naturally induced by HBV and as such would not be subjected to immunological tolerance. Moreover, the high conservation of MHC-E could allow for the use of universal TCR-based therapies, overcoming the major limitation of disparate HLA genotypes in conventional T cell therapies. However, very little is known about MHC-E expression on HBV-infected PH. Moreover, it is unknown whether HBV Ag-derived peptides would be accessible to MHC-E–restricted CD8+ T cells. To address these questions, we elicited HBV-specific, MHC-E–restricted CD8+ T cells in RM using RhCMV-based vectors expressing HBV Ags. Importantly, we demonstrate that the majority of rhesus and human HBV-infected PH express MHC-E and can be recognized by HBV-specific MHC-E–restricted CD8+ T cells. These data thus suggest a new paradigm for the treatment of HBV infection that theoretically could be universally applied to all patients given the extreme conservation of MHC-E across humans and macaques (11).

Genotype D, serotype ayw HBV core, polymerase, and S Ag gene fragments were isolated by PCR from previously described plasmids (kindly provided by Frank Chisari, Scripps Research Institute). The N-terminal 333 aa of polymerase obtained from plasmid pCDNA3-POL/ENV (20) were C-terminally HA-epitope tagged and fused by PCR-mediated mutagenesis to the C-terminal 228 aa of the S Ag obtained from plasmid pCMV-S2/S (21) to generate fusion S/PolN (left forward primer: 5′-CATCGAGCTAGCACCATGGAGAACATCACATCAGG-3′, left reverse primer: 5′-GTGTTGATAGGATAGGGGAATGTATACCCAAAGAC-3′; right forward primer: 5′-GTCTTTGGGTATACATTCCCCTATCCTATCAACAC-3′, right reverse primer: 5′-GGAATCGTCGACTCAAGCGTAATCTGGAACATCGTATGGGTAAAGATTGACGATAAGGGAGAGGCAG-3′). The final PCR product was blunt-end cloned into either pJet vector (Thermo Fisher Scientific) to be a template for bacterial artificial chromosome (BAC) recombineering or into pORI to evaluate expression. The C-terminal 416 aa of polymerase obtained from plasmid pCDNA3-POL/ENV (20) was HA-epitope tagged by PCR-mediated mutagenesis and inserted into pORI (forward primer: 5′-GTGGTACCCTCGAGGATTGGGGACCCTGCGCTGAACATGGAG-3′, reverse primer: 5′-TCAGTCGACCTAAGCGTAATCTGGAACATCGTATGGGTAC-3′). The gene encoding Core was PCR amplified from plasmid pCDNA-CORE (22) and inserted into pORI (forward primer: 5′-CTGCTAGCATGGACATTGACCCTTATAAAGAATTTGG-3′, reverse primer: 5′-CTAGGTACCACATTGAGATTCCCGAGATTGAG-3′). The C-terminal polymerase fragment was then inserted downstream of Core using KpnI and SalI to generate fusion protein HBV core and the C terminus of polymerase (Core/PolC). The KpnI site adds a 2-aa (GT) linker between the two proteins.

To generate 68-1 RhCMV/Core/PolC and 68-1 RhCMV/S/PolN, we replaced the pp71-encoding Rh110 gene in the 68-1 RhCMV BAC (23) using a modified galactokinase (galK) selection system, a two-step method that allows DNA modification without introducing unwanted heterologous sequences (24). We recently demonstrated that replacement of Rh110 can be used to elicit robust Ag responses while attenuating the 68-1 RhCMV vector (25). To delete Rh110, competent SW105 bacteria containing the 68-1 RhCMV BAC were electroporated with a PCR product containing a galK/kanamycin cassette with 50-bp flanking homology to Rh110. The bacteria were plated on kanamycin/chloramphenicol Luria Bertani agar at 30°C for positive selection. To replace the galK/kanamycin cassette with the HBV fusion genes, a PCR product containing the HBV S–PolN fusion or HBV Core–PolC fusion with the same flanking homology to Rh110 was electroporated, and the bacteria were plated on 2-deoxy-galactose (DOG) chloramphenicol minimal media plates with glycerol as the carbon source for negative selection. PCR primers for homologous recombination were as follows: Rh110 S/PolN forward: 5′-GATCACGTCATTGACACCGGCCTCCCACCAGCTCTCACATTCTCCGCATCACCATGGAGAACATCACATCAGGAT-3′, Rh110 S/PolN reverse: 5′-CAAAATATTATTACATGGTACGCAATTTATTGTCTATTTTCGTTATTTGTTTATTCAAGCGTAATCTGGAACATCGTAT-3′ and Rh110 Core/PolC forward: 5′-GATCACGTCATTGACACCGGCCTCCCACCAGCTCTCACATTCTCCGCATCACCATGGACATTGACCCTTATAAAGAAT-3′, Rh110 Core/PolC reverse: 5′-CAAAATATTATTACATGGTACGCAATTTATTGTCTATTTTCGTTATTTGTTTATCTAAGCGTAATCTGGAACATCGTAT-3′. To generate 68-1 RhCMV/Core, we amplified the HBV core gene from pCDNA-CORE and introduced an N-terminal FLAG-tag by PCR (forward primer: 5′-CTGCTAGCATGGATTACAAGGATGACAAGGACATCGACCCTTATAAAGAATTTGG-3′; reverse primer: 5′-CTAGTCGACACATTGAGATTCCCGAGATTGAG-3′). The amplified product was cloned into pORI downstream of the EF1α promoter. This expression cassette was inserted into Rh211 region of 68-1 RhCMV together with a Kan resistance cassette flanked by flippase recognition target sites by homologous recombination using primers containing 50-bp homology to regions of Rh211 (forward primer: 5′-GGGAAATCACGTCATCAGGCTGGGTAGTCAACATGGGCATACGAAACTTGCCCGAATAGATGCTCTCACTTAACGGCTGACATG-3′, reverse primer: 5′-CCAGAATGTGCTCTACTTTTTGGCCAGCGGGTTGGATGATTTCGCGCGTCATGGACTGCTTCACTGTAGCTTAGTACGTTAAAC-3′). The PCR fragment was electroporated into EL250 bacteria containing the RhCMV 68-1 BAC for in vivo recombination and recombinants selected for Kan resistance. The Kan resistance cassette was removed by temperature-inducible flippase recombination.

The resulting BACs were analyzed by restriction digest, PCR analysis of recombination sites, and next-generation sequencing on an Illumina MiSeq sequencer. This sequence analysis revealed two point mutations in S/PolN that were introduced during PCR amplification resulting in amino acid exchanges A118T and T125M in the S Ag.

BAC DNA was purified using alkaline lysis, phenol/chloroform extraction, and isopropanol precipitation, and virus was reconstituted by transfection of BAC DNA using Lipofectamine 2000 (following manufacturer’s protocol; Thermo Fisher Scientific) of telomerized pp71 expressing rhesus fibroblasts (25) or primary rhesus fibroblasts.

Expression of HBV Ags was confirmed by infecting telomerized RM fibroblasts with 68-1 RhCMV/Surface/PolN or RhCMV/Core/PolC. Cells were harvested at full cytopathic effect and lysed in SDS sample buffer. 293T cells transfected (Lipofectamine 2000) with the pORI expression plasmids containing the HA-tagged HBV proteins served as positive controls. After electrophoretic separation, immunoblots were performed with anti-HA Ab MMS-101P (Covance MMS).

A total of eight Indian RM (RM1-F-8yr, RM2-F-3yr, RM3-F-7yr, RM4-F-13yr, RM5-M-1yr, RM6-F-2yr, RM7-M-5yr, RM8-M-5yr) were inoculated s.c. with 1 × 107 PFU of each RhCMV expressing HBV Ag (RhCMV/HBV). RM are naturally resistant to HBV infection and therefore were not screened for preexisting anti-HBV immune responses (26). At assignment, all study RM were free of cercopithecine herpesvirus 1, D-type simian retrovirus, simian T-lymphotropic virus type 1, and Mycobacterium tuberculosis, but all were naturally RhCMV infected. All study RM were housed at the Oregon National Primate Research Center (ONPRC) in animal biosafety level 2 rooms with autonomously controlled temperature, humidity, and lighting. RM were fed commercially prepared primate chow twice daily and received supplemental fresh fruit or vegetables daily. Fresh, potable water was provided via automatic water systems. Physical examinations including body weight and complete blood counts were performed at all protocol time points. RM care and all experimental protocols and procedures were approved by the ONPRC Institutional Animal Care and Use Committee. The ONPRC is a Category I facility. The Laboratory Animal Care and Use Program at the ONPRC is fully accredited by the American Association for Accreditation of Laboratory Animal Care and has an approved assurance (no. A3304-01) for the care and use of animals on file with the National Institutes of Health Office for Protection from Research Risks. The Institutional Animal Care and Use Committee adheres to national guidelines established in the Animal Welfare Act (7 U.S. Code, sections 2131–2159) and the Guide for the Care and Use of Laboratory Animals, Eighth Edition as mandated by the U.S. Public Health Service Policy.

A single lobe of RM liver was perfused with 200 ml of preperfusion media (0.5 mM EGTA [catalog no. 40120128-1; Bio-World], 10 IU/ml heparin [catalog no. C504730; Fresenius Kabi], HBSS with calcium and magnesium [catalog no. 24-020-117; Fisher Scientific]) followed by 200 ml of HBSS without calcium and magnesium (catalog no. SH3003103; Fisher Scientific) to remove remaining EGTA. Next, 100 ml of collagenase media (DMEM/F12 [catalog no. 11320-082; Life Technologies], 1 mM calcium chloride [catalog no. C5670-100G; Sigma-Aldrich], 20 mM HEPES [catalog no. SH30237.01; HyClone], 1 mg/ml collagenase IV [catalog no. C9722-50MG; Sigma-Aldrich]) warmed to 42°C was perfused into the lobe and discarded. This was followed by recirculation of 150 ml of collagenase media through the liver lobe at 42°C for 30 min to 1 h using a rate of 75–150 ml/min, depending on the size of the liver lobe. Following collagenase perfusion, the liver was filleted with scalpels and washed over with remaining collagenase media, and media was filtered through a tea strainer. PH were washed three times in wash media (DMEM/F12, 2% bovine growth serum [catalog no. SH3054103; HyClone], 23 mM HEPES buffer, 0.6 mg/ml glucose, 2 mM l-glutamine [catalog no. SH3003401; HyClone], 1× antibiotic/antimycotic [catalog no. SV3007901; HyClone], and 0.1 mg/ml gentamicin [catalog no. 15750-060; Life Technologies]) at room temperature, with centrifugation between each wash at 50 × g for 3 min. Prior to the third wash spin, PH were passed through a 70-μM filter to ensure single-cell suspension. PH were then suspended in 20 ml of 36% isotonic Percoll (catalog no. 17-0891-01; GE Healthcare) in a 50-ml conical using PH media as a diluent (DMEM/F12, 10% bovine growth serum, 23 mM HEPES buffer, 0.6 mg/ml glucose, 2 mM l-glutamine, 1× antibiotic/antimycotic, and 0.1 mg/ml gentamicin) and centrifuged at 200 × g for 7 min. The purified PH pellet was then resuspended in room temperature PH media and counted. Collagenized plates for the hepatocytes were prepared using 0.2 mg/ml collagen R in 0.01% acetic acid (catalog no. 47254; Serva), left on the plate for at least 20 min prior to washing with 1 ml of HBSS and immediately prior to plating at 2 × 105 PH per well in a 12-well plate. Plates were placed at 37°C, 5% CO2. The next day, wells were washed twice with HBSS and cultured in 1 ml of PH media supplemented with 1.8% DMSO (PH-DMSO) for the remainder of the experiment. Human donor (HD) PH were isolated from murine humanized livers and purchased from Yecuris humanized murine donors that were generated with cryopreserved PH collected from deceased patients with the following demographics: HD1 (13 y old, female, Hispanic, HBV naive), HD2 (13 y old, female, white, HBV naive), and HD3 (27 y old, male, white, HBV naive).

One day after plating RM PH (see 5Isolation of PH), replication-incompetent adenovirus serotype five expressing human NTCP (multiplicity of infection [MOI]: 10) under the liver-specific TTR promoter was added to the culture for 2 d. On the second day, cells were refed with 1 ml of PH-DMSO media. On the fourth day following adenovirus transduction, PH were washed twice in 1 ml of HBSS and overlaid with HBV-containing media at an MOI of 100 (PH-DMSO containing 4% PEG6000, catalog no. 81253-250G; Sigma-Aldrich) and incubated overnight. The next morning, wells were washed three times with 1 ml of HBSS and then cultured in 1 ml of PH-DMSO for the remainder of the experiment.

One day after plating, HD PH were overlaid with HBV-containing media at an MOI of 100 (PH-DMSO containing 4% PEG6000, catalog no. 81253-250G; Sigma-Aldrich) and incubated overnight. The next morning, wells were washed three times with 1 ml of HBSS and then cultured in 1 ml of PH-DMSO for the remainder of the experiment.

Prior to HBV infection, one well of PH was collected and stained as a baseline. Starting on day two postinfection, one well each of HBV-infected and HBV-naive PH was collected with 0.5% trypsin-EDTA (catalog no. SH30236.01; Fisher Scientific) and washed twice in ice-cold FACS buffer (PBS, catalog no. SH30256FS; Fisher Scientific; with 10% FBS). Cells were incubated with anti–HLA-E Ab (clone: 4D12, catalog no. LS-C179742; Origene) for 30 min at 4°C, washed twice in ice-cold FACS buffer, and incubated with F(ab)2-goat anti-mouse IgG (H L)-allophycocyanin (catalog no. A10539; Invitrogen) for 30 min at 4°C. Cells were then washed twice in ice-cold PBS and incubated with pan–MHC class I (MHC-I)–PerCP–Cy5.5 (clone: W6/32, catalog no. 311419; BioLegend), anti-HLA-DR Alexa 700 (clone: L243, catalog no. 560743; BD Biosciences), and Live/Dead fixable yellow (catalog no. L-34959; Invitrogen) for 30 min at 4°C. Cells were washed in FACS buffer and fixed using Foxp3/Transcription Factor Staining Buffer Set (catalog no. 00-5523-00; eBioscience) for 1 h at room temperature. Prior to fixation, all wash spins were performed at 350 × g for 3 min. After fixation, cells were suspended in permeabilization buffer (catalog no. 00-8333-56; eBioscience). All wash spins after fixation were performed at 830 × g for 3 min. PH were incubated for 1 h at 4°C with HBV Core Ag (HBcAg) Ab (clone: 13A9, catalog no. MA1-7606; Fisher Scientific) conjugated to R-PE using the Lightning-Link R-PE Kit (catalog no. 703-0005; Innova Biosciences). Cells were washed three times in permeabilization buffer and then collected on a Becton-Dickenson LSR-II. Analysis was performed on FlowJo X (TreeStar). In all analyses, gating on the light scatter signature of large, complex PH was followed by assessment of specific MHC and HBV markers.

HBV-specific CD8+ T cell responses were measured in mononuclear cell preparations from the peripheral blood (longitudinal and deconvolution) or spleens (MHC restriction and HBV + PH recognition) of RhCMV/HBV-vaccinated RM by flow cytometric intracellular cytokine staining (ICS), as previously described (27, 28). Briefly, splenocytes (E:T of 10:1) or isolated CD8β+ T cells (E:T of 5:1) were incubated with HBV-infected or HBV-naive PH targets and the costimulatory molecules CD28 and CD49d (BD Biosciences) for 1 h, followed by addition of brefeldin A (Sigma-Aldrich) for an additional 8 h. CD8β+ T cells were isolated by staining with anti–CD8β-PE (Beckman Coulter) and then secondarily staining with anti-PE beads before running over a positive selection magnetic column according to manufacturer’s instructions (Miltenyi Biotec). Costimulation without PH target coculture served as a background control. The MHC restriction of a response was determined by the following: 1) preincubating PH targets for 1 h at room temperature in the presence of pan anti–MHC-I Ab (25 μg/ml; clone W6-32), VL9 peptide (20 μM), CLIP peptide (MHC-II–associated invariant chain, aa 89–100; 10 μg/ml), or anti–HLA-DR Ab (10 μg/ml; clone L243) before coculturing with effector cells or 2) preincubating cell lines (K562 or 0.221 derived) expressing a single MHC molecule with HBV peptides for 1 h at room temperature prior to coculturing with effector cells (E:T of 10:1). K562 cell lines expressing HLA-E*01:03 or Mamu-E*02:20 were generated, cultured, and used in our assays, as previously described (11). The 0.221 cell lines expressing HLA-A*02:01 or HLA-C*06:02 were generated using lentiviral transduction as described in (29). HLA fragments were cloned into the modified pLVX-EF1α-IRES-Puro (Clontech) vector, in which EF1α was replaced with the spleen focus-forming virus promoter. The expression cassette encoded ZsGreen linked via self-cleaved P2A peptide to HLA with a FLAG-tag at its N terminus. HLA-positive cells were selected using 0.25 μg/ml puromycin. Stimulated T cells were fixed, permeabilized, and stained as described (27, 28), and flow cytometric analysis was performed on an LSR-II instrument (BD Biosciences). Analysis was done using FlowJo X software (Tree Star). In all analyses, gating on the light scatter signature of small lymphocytes was followed by progressive gating on the CD3+ population and then the CD4/CD8+ T cell subset. Longitudinal Ag-specific response frequencies for CD8+ T cell populations were routinely determined by intracellular expression of IFN-γ following stimulation of PBMC with pools of overlapping 15-mer peptides corresponding to the HBV core open reading frame (Genscript; >70% purity). For all ICS results shown, we collected >1 × 105 CD3+ T cell events to ensure the accuracy of our conclusions. CD8+ T cell responses were memory corrected (naive cells excluded) as previously described (28).

To elicit MHC-II and MHC-E–restricted HBV-specific CD8+ T cell responses in RM, we constructed two recombinant strain 68-1 RhCMV vectors expressing HBV genotype D serotype ayw core, surface, and polymerase Ags. Using BAC recombineering, we replaced the gene Rh110, which encodes for the RhCMV homolog of HCMV pp71 with either an HA-tagged fusion protein of Core/PolC or an HA-tagged fusion protein of S and the N terminus of/S and the N terminus of polymerase (Surface/PolN) (Fig. 1A). Upon reconstitution of recombinant RhCMV, we confirmed fusion protein expression by immunoblot using HA-epitope–specific Abs (Fig. 1B). We inoculated two RM (RM1, RM2) with both 68-1 RhCMV/Surface/PolN and 68-1 RhCMV/Core/PolC vectors and longitudinally monitored the CD8+ T cell response against each of the Ags by ICS, using pools of overlapping 15-mer peptides corresponding to each Ag. We found robust CD8+ T cell responses to all three HBV Ags in both animals (Fig. 1C). In addition, we inoculated two additional RM (RM3, RM4) with a 68-1 RhCMV–based vector that expressed HBV Core under the EF1α promoter (Fig. 2A) and observed longitudinal HBcAg-specific CD8+ T cell responses in the blood of both RM (Fig. 2B). To verify that the CD8+ T cells elicited by the 68-1 RhCMV/Core/PolC and 68-1 RhCMV/Core vector were restricted by MHC-II and MHC-E, we characterized the breadth and MHC restriction of HBcAg-specific CD8+ T cell responses in all four animals via ICS by using individual peptides and reagents that specifically block presentation by MHC-I, MHC-II, and MHC-E, as previously described (Supplemental Fig. 1) (27, 28). We observed that, similar to strain 68-1 RhCMV vectors expressing SIV Ags under the endogenous Rh110 promoter (25) or EF1α promoter (27, 28), the HBcAg-specific CD8+ T cells targeted a broad array of HBcAg peptides that were presented either by MHC-II or MHC-E (Fig. 2C). However, given that many of these responses are to consecutive 15 mers, it is likely that the true number of targeted HBV Ags is smaller. We next further confirmed the MHC-E restriction of the RhCMV/HBV engendered CD8+ T cell responses by performing MHC restriction assays using splenocytes from RM3 as effector cells cocultured with K562 cells (MHC null) expressing either a single human (HLA) or RM (Macaca mulatta [Mamu]) MHC-E allele and pulsed with one of three individual HBcAg 15-mer peptides identified as MHC-E restricted via blocking in Fig. 2C. We found that these HBcAg-specific CD8+ T cells recognized their cognate Ag presented in the context of both HLA-E and Mamu-E (Fig. 2D). Of note, these HBcAg-derived 15-mer peptides do not contain amino acid sequences matching the canonical leader sequence-like binding motif of MHC-E, in line with the alternate MHC-E binding repertoire that we, and others, have recently described (27, 3032). These results demonstrate the presence of MHC-E–restricted, HBV-specific CD8+ T cells in 68-1 RhCMV/HBV–inoculated RM and further support the high functional conservation of primate MHC-E molecules as we have shown previously (11, 27). Because the surface and polymerase Ags are expressed by the same vector backbone in RM1 and RM2, it is highly likely that surface and polymerase-specific CD8+ T cells are similarly restricted by MHC-E and MHC-II.

We next set out to determine if PH express MHC-E. To this end, we isolated PH from three unrelated RM and three unrelated HD and examined surface expression of bulk MHC-I using Ab W6/32, MHC-II using Ab HLA-DR, and MHC-E using Ab 4D12 (33). We previously demonstrated that in addition to HLA-E, 4D12 specifically stains Mamu-E but not classical Mamu-Ia molecules (27). All three HD shared one HLA-A and one HLA-C allele (Table I). Thus, before proceeding, we confirmed that the MHC-E–specific 4D12 clone stains only HLA-E and not the HLA-A or -C molecules shared between the three HD (Fig. 3A). We then stained for expression of MHC-E and demonstrated that the majority of PH from both species expressed MHC-E (Fig. 3B, 3C). To determine if HBV infection influences the expression of MHC-E on the surface of PH, we collected PH from the same three HD and infected these cells 1 d after plating with an MOI of 100 of HBV genotype D serotype ayw. HBV infection of HD PH was confirmed by measuring the level of HBV envelope Ag in the supernatant prior to staining of the cells (Supplemental Fig. 2A). We costained for MHC markers (MHC-I, MHC-E, and MHC-II) along with intracellular HBcAg on day 4 post–HBV infection because this was the first time point where intracellular HBcAg was detectable (Supplemental Fig. 2B). We observed staining with the 4D12 Ab on both HBV-infected and HBV-naive PH, indicating moderate levels of MHC-E surface expression (Fig. 3D, 3E, Supplemental Fig. 3). In contrast, HLA-DR expression was minimal or absent in all three HD PH samples, in line with previously published results (34). It is possible that our ex vivo manipulation of the PH prior to assessing surface MHC levels induced a fraction of these cells to lose MHC-E expression. Nevertheless, taken together, these results showed that HBV-infected PH express MHC-E and that MHC-E could represent a potential HBV-specific CD8+ T cell restriction element.

Our MHC study revealed MHC-E expression on HD and RM PH, regardless of HBV infection. However, given the dominance of VL9 peptide binding to MHC-E, it was unclear whether hepatocyte MHC-E would present HBV-derived peptides and whether these noncanonical MHC-E/peptide complexes could be recognized by RhCMV/HBV vector-induced, MHC-E-restricted, HBV-specific CD8+ T cells. We thus determined whether HBcAg-specific CD8+ T cells from 68-1 RhCMV/Core–inoculated RM would recognize allogeneic, HBV-infected RM PH taking advantage of the high functional conservation of MHC-E in primates (11). Strikingly, we found that CD8+ T cells (bulk splenocytes and purified CD8β+ T cells) from RM3 and RM4 recognized HBV-infected PH from two unrelated RM donors but did not respond to uninfected PH (Fig. 4A).

To more comprehensively determine the MHC restriction of CD8+ T cells recognizing HBV-infected PH targets, we performed a series of recognition experiments using the same MHC-specific blocking reagents used in Fig. 2C and as we have previously described (27, 28). HBV-infected or HBV-naive targets were collected at day 6 post–HBV infection (MOI = 100); incubated with the blocking agents W6/32 Ab (pan MHC-I), VL9 peptide (MHC-E), CLIP (MHC-II), or HLA-DR Ab (MHC-II); and then cocultured with splenocytes or isolated CD8β+ T cells overnight. Following coculture, CD8+ T cells were stained intracellularly for IFN-γ and TNF-α to assess recognition of the targets. We discovered that CD8+ T cell recognition of HBV-infected RM PH was blocked with W6/32 Ab and VL9 peptide, but not with CLIP or HLA-DR, indicating that the entirety of the response to HBV-infected targets was MHC-E restricted (Fig. 4A).

Because MHC-E is functionally conserved across primates (11), and because we demonstrated that the HBV-specific RM CD8+ T cells recognized HLA-E expressing K562 cells (Fig. 2D), we hypothesized that CD8+ T cells from 68-1 RhCMV/Core–inoculated RM would also recognize HBV-infected HD PH. To test this hypothesis, we performed similar recognition experiments incubating splenocytes and purified CD8β+ T cells from the same 68-1 RhCMV/Core–inoculated macaques (RM3, RM4) with HBV-infected HD PH. As hypothesized, these CD8+ T cells recognized HBV-infected, xenogeneic HD PH (Fig. 4B). As described above for RM PH target coculture experiments, the CD8+ T cell recognition of HBV-infected HD PH was completely blocked by the MHC-E–binding VL9 peptide. Taken together, these results definitively show that HBV-infected PH present HBV Ag in the context of MHC-E, indicating that this pathway can be exploited to target CD8+ T cells to HBV-infected cells.

Immunotherapies currently under investigation are designed to harness the immune system to better target HBV-infected hepatocytes and include immune stimulation with pattern recognition receptor agonists, check point inhibitor blockades, therapeutic vaccines, and adoptive T cell therapy (35). A common hurdle facing HBV immunotherapies is T cell immunotolerance (3638). The initial triggers of immunotolerance, which distinguish patients that successfully clear acute HBV viremia from those that do not, are not completely understood. However, it is likely in part a consequence of the immunotolerant environment of the liver. Thus, to successfully clear CHB via immunotherapy, T cell immunotolerance must be overcome. Unfortunately, no immunotherapies to date have consistently achieved this goal, and this reality has been exacerbated by the lack of physiologically relevant animal models of CHB.

In this study, we report MHC-E–restricted CD8+ T cell responses against HBV, which paves the way for development of innovative HBV therapies. Whereas MHC-E–restricted CD8+ T cell responses have been occasionally identified in natural viral infections with CMV, EBV, tuberculosis, and HCV (39), no reports of MHC-E–restricted CD8+ T cells responses against HBV have been published. Moreover, it was unclear whether HBV-infected hepatocytes would be able to present HBV Ag in the context of MHC-E. We show in this study that MHC-E does present HBV Ags on the surface of HBV-infected hepatocytes and that CD8+ T cells from a completely distinct primate species can recognize these MHC-E:peptide complexes. Because we previously showed that MHC-E–restricted CD8+ T cells can recognize CD4+ T cells infected with SIV (27), these results suggest that endogenous peptide presentation by MHC-E seems to be a common occurrence in virally infected cells. Thus, although 68-1 RhCMV is currently the only vector that can be “programmed” to elicit MHC-E–restricted CD8+ T cells to any given target Ag, these T cells seem to be able to recognize, and possibly eliminate, pathogens via presentation of endogenous peptides by MHC-E.

Outside of representing a completely unique type of CD8+ T cell response against HBV, the breadth of epitopes targeted within HBcAg indicates that therapeutic vaccination with CMV/HBV vectors would elicit broadly targeted CD8+ T cell responses. Although this broad targeting has been shown previously against SIV (27), M. tuberculosis (40), and malaria (12), it may be particularly efficacious against HBV because the vast majority of the HBV genome is composed almost exclusively of sequence-constrained overlapping reading frames. In this study, we present evidence that MHC-E–restricted CD8+ T cells can be harnessed for the treatment of CHB, either through therapeutic vaccination or adoptive immunotherapy. These results warrant the continued study of these unconventional CD8+ T cell responses in terms of in vivo HBV targeting and suppression.

We thank Mary N. Carrington for providing single MHC expressing cell lines and Daniel E. Geraghty for the MHC-E mAb 4D12 hybridoma. We thank Frank Chisari for providing plasmids POL/ENV, pCMV-S2/S, and pCDNA-CORE. We thank Ulla Protzer for providing cell culture–derived HBV genotype D serotype ayw.

This work was supported by Vir Biotechnology (SRA-17-111, to B.J.B.); the Oregon Nanoscience and Microtechnologies Institute (SRA-13-053-B, to K.F.); the National Institutes of Health (NIH)/National Institute of Allergy and Infectious Diseases under Awards R01 AI144008 (to B.J.B.) and R01 AI117802, R01 AI129703, and R01 AI140888 (to J.B.S.); and the NIH Office of the Director (P51OD011092, to the Oregon National Primate Research Center).

The online version of this article contains supplemental material.

Abbreviations used in this article:

BAC

bacterial artificial chromosome

CHB

chronic hepatitis B virus infection

Core/PolC

HBV core and the C terminus of polymerase

galK

galactokinase

HBcAg

HBV Core Ag

HBV

hepatitis B virus

HD

human donor

ICS

intracellular cytokine staining

Mamu

Macaca mulatta

MHC-I

MHC class I

MHC-II

MHC class II

MOI

multiplicity of infection

ONPRC

Oregon National Primate Research Center

PH

primary hepatocyte

RhCMV

rhesus macaque CMV

RhCMV/HBV

RhCMV expressing HBV Ag

RM

rhesus macaque

Surface/PolN

preS2/S and the N terminus of polymerase.

1
World Health Organization
. Hepatitis B. Available at: http://www.who.int/mediacentre/factsheets/fs204/en/. Accessed: November 13, 2019.
2
Joint Committee on Vaccination and Immunisation/Great Britain Department of Health. 2006. Hepatitis B. In Immunisation Against Infectious Disease, 3rd Ed. D. Salisbury, M. Ramsay, K. Noakes, eds. The Stationery Office, London, p. 468–492.
3
Bhattacharya
,
D.
,
C. L.
Thio
.
2010
.
Review of hepatitis B therapeutics.
Clin. Infect. Dis.
51
:
1201
1208
.
4
Zhang
,
Y.
,
B.
Chen
,
L.
Wang
,
J.
Chi
,
S.
Song
,
M.
Liu
,
Z.
Zhao
.
2016
.
HBsAg seroclearance or seroconversion induced by peg-interferon alpha and lamivudine or adefovir combination therapy in chronic hepatitis B treatment: a meta-analysis and systematic review.
Rev. Esp. Enferm. Dig.
108
:
263
270
.
5
Maini
,
M. K.
,
C.
Boni
,
G. S.
Ogg
,
A. S.
King
,
S.
Reignat
,
C. K.
Lee
,
J. R.
Larrubia
,
G. J.
Webster
,
A. J.
McMichael
,
C.
Ferrari
, et al
.
1999
.
Direct ex vivo analysis of hepatitis B virus-specific CD8(+) T cells associated with the control of infection.
Gastroenterology
117
:
1386
1396
.
6
Phillips
,
S.
,
S.
Chokshi
,
A.
Riva
,
A.
Evans
,
R.
Williams
,
N. V.
Naoumov
.
2010
.
CD8(+) T cell control of hepatitis B virus replication: direct comparison between cytolytic and noncytolytic functions.
J. Immunol.
184
:
287
295
.
7
Fisicaro
,
P.
,
C.
Valdatta
,
C.
Boni
,
M.
Massari
,
C.
Mori
,
A.
Zerbini
,
A.
Orlandini
,
L.
Sacchelli
,
G.
Missale
,
C.
Ferrari
.
2009
.
Early kinetics of innate and adaptive immune responses during hepatitis B virus infection.
Gut
58
:
974
982
.
8
Bertoletti
,
A.
,
C.
Ferrari
.
2016
.
Adaptive immunity in HBV infection.
J. Hepatol.
64
(
1
Suppl.
):
S71
S83
.
9
Rehermann
,
B.
,
A.
Bertoletti
.
2015
.
Immunological aspects of antiviral therapy of chronic hepatitis B virus and hepatitis C virus infections.
Hepatology
61
:
712
721
.
10
Kurktschiev
,
P. D.
,
B.
Raziorrouh
,
W.
Schraut
,
M.
Backmund
,
M.
Wächtler
,
C.-M.
Wendtner
,
B.
Bengsch
,
R.
Thimme
,
G.
Denk
,
R.
Zachoval
, et al
.
2014
.
Dysfunctional CD8+ T cells in hepatitis B and C are characterized by a lack of antigen-specific T-bet induction.
J. Exp. Med.
211
:
2047
2059
.
11
Wu
,
H. L.
,
R. W.
Wiseman
,
C. M.
Hughes
,
G. M.
Webb
,
S. A.
Abdulhaqq
,
B. N.
Bimber
,
K. B.
Hammond
,
J. S.
Reed
,
L.
Gao
,
B. J.
Burwitz
, et al
.
2018
.
The role of MHC-E in T cell immunity is conserved among humans, rhesus macaques, and cynomolgus macaques.
J. Immunol.
200
:
49
60
.
12
Hansen
,
S. G.
,
J.
Womack
,
I.
Scholz
,
A.
Renner
,
K. A.
Edgel
,
G.
Xu
,
J. C.
Ford
,
M.
Grey
,
B.
St Laurent
,
J. M.
Turner
, et al
.
2019
.
Cytomegalovirus vectors expressing Plasmodium knowlesi antigens induce immune responses that delay parasitemia upon sporozoite challenge.
PLoS One
14
: e0210252.
13
Hansen
,
S. G.
,
M.
Piatak
Jr.
,
A. B.
Ventura
,
C. M.
Hughes
,
R. M.
Gilbride
,
J. C.
Ford
,
K.
Oswald
,
R.
Shoemaker
,
Y.
Li
,
M. S.
Lewis
, et al
.
2013
.
Immune clearance of highly pathogenic SIV infection. [Published erratum appears in 2014 Nature 514: 654.]
Nature
502
:
100
104
.
14
Hansen
,
S. G.
,
C.
Vieville
,
N.
Whizin
,
L.
Coyne-Johnson
,
D. C.
Siess
,
D. D.
Drummond
,
A. W.
Legasse
,
M. K.
Axthelm
,
K.
Oswald
,
C. M.
Trubey
, et al
.
2009
.
Effector memory T cell responses are associated with protection of rhesus monkeys from mucosal simian immunodeficiency virus challenge. [Published erratum appears in 2009 Nat. Med. 15: 462.]; [Published erratum appears in 2011 Nat. Med. 17: 1692.]
Nat. Med.
15
:
293
299
.
15
Braud
,
V.
,
E. Y.
Jones
,
A.
McMichael
.
1997
.
The human major histocompatibility complex class Ib molecule HLA-E binds signal sequence-derived peptides with primary anchor residues at positions 2 and 9.
Eur. J. Immunol.
27
:
1164
1169
.
16
Lee
,
N.
,
M.
Llano
,
M.
Carretero
,
A.
Ishitani
,
F.
Navarro
,
M.
López-Botet
,
D. E.
Geraghty
.
1998
.
HLA-E is a major ligand for the natural killer inhibitory receptor CD94/NKG2A.
Proc. Natl. Acad. Sci. USA
95
:
5199
5204
.
17
Chen
,
A.
,
Y.
Shen
,
M.
Xia
,
L.
Xu
,
N.
Pan
,
Y.
Yin
,
F.
Miao
,
C.
Shen
,
W.
Xie
,
J.
Zhang
.
2011
.
Expression of the nonclassical HLA class I and MICA/B molecules in human hepatocellular carcinoma.
Neoplasma
58
:
371
376
.
18
Pietra
,
G.
,
C.
Romagnani
,
P.
Mazzarino
,
M.
Falco
,
E.
Millo
,
A.
Moretta
,
L.
Moretta
,
M. C.
Mingari
.
2003
.
HLA-E-restricted recognition of cytomegalovirus-derived peptides by human CD8+ cytolytic T lymphocytes.
Proc. Natl. Acad. Sci. USA
100
:
10896
10901
.
19
Jinushi
,
M.
,
T.
Takehara
,
T.
Tatsumi
,
T.
Kanto
,
T.
Miyagi
,
T.
Suzuki
,
Y.
Kanazawa
,
N.
Hiramatsu
,
N.
Hayashi
.
2004
.
Negative regulation of NK cell activities by inhibitory receptor CD94/NKG2A leads to altered NK cell-induced modulation of dendritic cell functions in chronic hepatitis C virus infection.
J. Immunol.
173
:
6072
6081
.
20
Kakimi
,
K.
,
M.
Isogawa
,
J.
Chung
,
A.
Sette
,
F. V.
Chisari
.
2002
.
Immunogenicity and tolerogenicity of hepatitis B virus structural and nonstructural proteins: implications for immunotherapy of persistent viral infections.
J. Virol.
76
:
8609
8620
.
21
Michel
,
M. L.
,
H. L.
Davis
,
M.
Schleef
,
M.
Mancini
,
P.
Tiollais
,
R. G.
Whalen
.
1995
.
DNA-mediated immunization to the 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
5311
.
22
Isogawa
,
M.
,
Y.
Furuichi
,
F. V.
Chisari
.
2005
.
Oscillating CD8(+) T cell effector functions after antigen recognition in the liver.
Immunity
23
:
53
63
.
23
Chang
,
W. L. W.
,
P. A.
Barry
.
2003
.
Cloning of the full-length rhesus cytomegalovirus genome as an infectious and self-excisable bacterial artificial chromosome for analysis of viral pathogenesis.
J. Virol.
77
:
5073
5083
.
24
Warming
,
S.
,
N.
Costantino
,
D. L.
Court
,
N. A.
Jenkins
,
N. G.
Copeland
.
2005
.
Simple and highly efficient BAC recombineering using galK selection.
Nucleic Acids Res.
33
: e36.
25
Marshall, E. E., D. Malouli, S. G. Hansen, R. M. Gilbride, C. M. Hughes, A. B. Ventura, E. Ainslie, A. N. Selseth, J. C. Ford, D. Burke, et al. 2019. Enhancing safety of cytomegalovirus-based vaccine vectors by engaging host intrinsic immunity. Sci. Transl. Med. 11: eaaw2603 10.1126/scitranslmed.aaw2603
26
Burwitz
,
B. J.
,
J. M.
Wettengel
,
M. A.
Mück-Häusl
,
M.
Ringelhan
,
C.
Ko
,
M. M.
Festag
,
K. B.
Hammond
,
M.
Northrup
,
B. N.
Bimber
,
T.
Jacob
, et al
.
2017
.
Hepatocytic expression of human sodium-taurocholate cotransporting polypeptide enables hepatitis B virus infection of macaques.
Nat. Commun.
8
:
2146
.
27
Hansen
,
S. G.
,
H. L.
Wu
,
B. J.
Burwitz
,
C. M.
Hughes
,
K. B.
Hammond
,
A. B.
Ventura
,
J. S.
Reed
,
R. M.
Gilbride
,
E.
Ainslie
,
D. W.
Morrow
, et al
.
2016
.
Broadly targeted CD8+ T cell responses restricted by major histocompatibility complex E.
Science
351
:
714
720
.
28
Hansen
,
S. G.
,
J. B.
Sacha
,
C. M.
Hughes
,
J. C.
Ford
,
B. J.
Burwitz
,
I.
Scholz
,
R. M.
Gilbride
,
M. S.
Lewis
,
A. N.
Gilliam
,
A. B.
Ventura
, et al
.
2013
.
Cytomegalovirus vectors violate CD8+ T cell epitope recognition paradigms.
Science
340
: 1237874.
29
Garcia-Beltran
,
W. F.
,
A.
Hölzemer
,
G.
Martrus
,
A. W.
Chung
,
Y.
Pacheco
,
C. R.
Simoneau
,
M.
Rucevic
,
P. A.
Lamothe-Molina
,
T.
Pertel
,
T.-E.
Kim
, et al
.
2016
.
Open conformers of HLA-F are high-affinity ligands of the activating NK-cell receptor KIR3DS1.
Nat. Immunol.
17
:
1067
1074
.
30
McMurtrey
,
C.
,
M. J.
Harriff
,
G. M.
Swarbrick
,
A.
Duncan
,
M.
Cansler
,
M.
Null
,
W.
Bardet
,
K. W.
Jackson
,
D. A.
Lewinsohn
,
W.
Hildebrand
,
D. M.
Lewinsohn
.
2017
.
T cell recognition of Mycobacterium tuberculosis peptides presented by HLA-E derived from infected human cells.
PLoS One
12
: e0188288.
31
Lampen
,
M. H.
,
C.
Hassan
,
M.
Sluijter
,
A.
Geluk
,
K.
Dijkman
,
J. M.
Tjon
,
A. H.
de Ru
,
S. H.
van der Burg
,
P. A.
van Veelen
,
T.
van Hall
.
2013
.
Alternative peptide repertoire of HLA-E reveals a binding motif that is strikingly similar to HLA-A2.
Mol. Immunol.
53
:
126
131
.
32
Walters
,
L. C.
,
K.
Harlos
,
S.
Brackenridge
,
D.
Rozbesky
,
J. R.
Barrett
,
V.
Jain
,
T. S.
Walter
,
C. A.
O’Callaghan
,
P.
Borrow
,
M.
Toebes
, et al
.
2018
.
Pathogen-derived HLA-E bound epitopes reveal broad primary anchor pocket tolerability and conformationally malleable peptide binding. [Published erratum appears in 2018 Nat. Commun. 9: 4833.]
Nat. Commun.
9
:
3137
.
33
Haynes
,
B. F.
,
E. G.
Reisner
,
M. E.
Hemler
,
J. L.
Strominger
,
G. S.
Eisenbarth
.
1982
.
Description of monoclonal antibody defining an HLA allotypic determinant that includes specificities within the B5 cross-reacting group.
Hum. Immunol.
4
:
273
285
.
34
Senaldi
,
G.
,
A.
Lobo-Yeo
,
A. P.
Mowat
,
G.
Mieli-Vergani
,
D.
Vergani
.
1991
.
Class I and class II major histocompatibility complex antigens on hepatocytes: importance of the method of detection and expression in histologically normal and diseased livers.
J. Clin. Pathol.
44
:
107
114
.
35
Gill
,
U. S.
,
P. T. F.
Kennedy
.
2017
.
Current therapeutic approaches for HBV infected patients.
J. Hepatol.
67
:
412
414
.
36
Zong
,
L.
,
H.
Peng
,
C.
Sun
,
F.
Li
,
M.
Zheng
,
Y.
Chen
,
H.
Wei
,
R.
Sun
,
Z.
Tian
.
2019
.
Breakdown of adaptive immunotolerance induces hepatocellular carcinoma in HBsAg-tg mice.
Nat. Commun.
10
:
221
.
37
Kong
,
X.
,
R.
Sun
,
Y.
Chen
,
H.
Wei
,
Z.
Tian
.
2014
.
γδT cells drive myeloid-derived suppressor cell-mediated CD8+ T cell exhaustion in hepatitis B virus-induced immunotolerance.
J. Immunol.
193
:
1645
1653
.
38
Milich
,
D. R.
2016
.
The concept of immune tolerance in chronic hepatitis B virus infection is alive and well.
Gastroenterology
151
:
801
804
.
39
Joosten
,
S. A.
,
L. C.
Sullivan
,
T. H. M.
Ottenhoff
.
2016
.
Characteristics of HLA-E restricted T-cell responses and their role in infectious diseases.
J. Immunol. Res.
2016
: 2695396.
40
Hansen
,
S. G.
,
D. E.
Zak
,
G.
Xu
,
J. C.
Ford
,
E. E.
Marshall
,
D.
Malouli
,
R. M.
Gilbride
,
C. M.
Hughes
,
A. B.
Ventura
,
E.
Ainslie
, et al
.
2018
.
Prevention of tuberculosis in rhesus macaques by a cytomegalovirus-based vaccine.
Nat. Med.
24
:
130
143
.

Oregon Health & Science University and Drs. Picker, Hansen, Sacha, Burwitz, and Früh have a significant financial interest in Vir Biotechnology, Inc., a company that may have a commercial interest in the results of this research and technology. The potential individual and institutional conflicts of interest have been reviewed and managed by Oregon Health & Science University. The other authors have no financial conflicts of interest.

Supplementary data