Innate immunity and HBV persistence

Abstract

Hepatitis B virus (HBV) causes chronic infections that are associated with immune dysfunction. Though T cell impairment is perhaps the most prominent immune change contributing to viral persistence, HBV interaction with the innate immune system is also likely key, as the lack of effective innate immunity has functional consequences that promote chronic infection. In addition to an intrinsic ability to fight viral infections, the innate immune system also impacts T cell responses and other adaptive immune mechanisms critical for HBV control. Therefore, it is essential to understand the relationships between HBV and innate immunity, as these interactions may be useful immunotherapeutic targets to manage the infection.

Introduction

The interactions between HBV and the immune system are important for the virus’ ability to persist, but the specific host and viral factors that impact whether an acute HBV infection is eliminated or progresses to chronic disease are poorly understood. Innate immunity plays crucial roles in the early response to viral infection and promotes adaptive immunity. Most viruses, including HBV, have evolved strategies to evade this response; however, the relationships between HBV and the cytokine and cellular components of innate immunity remain incompletely characterized. The pathways that drive innate immune activation represent promising therapeutic targets to overcome immune dysfunction and promote HBV cure. Here, we review the interactions between HBV and the interferon (IFN)- and cell-mediated mechanisms of innate immunity and discuss the therapeutic approaches that seek to engineer an effective innate immune response to the virus.

Interferon activation

HBV is often referred to as a “stealth” virus as it establishes infection without being detected by the innate immune system and thus avoids the activation of antiviral pathways [1]. Unlike other viruses that actively block innate responses, the lack of IFN induction by HBV in hepatocytes is thought to be a passive process [2-4]. This characteristic of HBV is debated, and multiple studies have aimed to investigate whether the virus stimulates innate immunity. In vivo analysis identified a lack of early antiviral gene expression in the chimpanzee liver after experimental HBV infection [5], and recent work using the woodchuck hepatitis virus (WHV) model of hepadnavirus infection also yielded findings consistent with the idea of HBV stealth [6]. Several mechanistic in vitro cell culture studies demonstrated that HBV passively avoids IFN activation [2,3], and recent findings further supporting this concept were made utilizing a novel scalable hepatic coculture system [7], which found no significant activation of the innate immune response by HBV [8]. Despite this evidence, some studies suggest that HBV stimulates or suppresses innate immunity [9-11], including recent work showing that activation of glycolysis by HBV inhibits RIG-I signaling [12]. Another recent study found that HBV infection induces TLR2-mediated expression of IL-6, IL-1β, and TNF-α in primary human hepatocytes without activating IFN or IFN-stimulated genes (ISG) [13]. Although these results seem to challenge the stealth virus concept, the high multiplicities of infection often required to reach this activation in vitro may not recapitulate a natural infection. Amongst the various experimental approaches used to evaluate HBV innate immune activation, the most physiologically relevant evidence to support HBV stealth perhaps comes from experiments performed using ex vivo human liver biopsy samples from infected patients [4]. These results showed a lack of ISG expression in the biopsies and no block of the IFN response, as hepatocytes from HBV-infected patients had a similar capacity to produce IFN as those from uninfected patients when exogenously stimulated with polyIC or Sendai virus [4].

Interferon antiviral activity

Although HBV appears to avoid activating innate immunity, IFN can nevertheless reduce viral replication [14]. There are three main types of IFN (α/β, γ, and λ) that signal through unique receptors with cell-type dependent expression patterns, and non-cytopathically inhibit HBV by downregulating HBV mRNA expression and viral pre-genomic RNA encapsidation [15-17].

IFN-α

IFN-α is the only immunomodulator currently approved for HBV therapy. In some cases, patients display prolonged viral suppression and HBsAg seroconversion following pegylated (PEG)-IFN-α treatment, but the reasons for why only certain patients respond to therapy remain unknown. While the precise therapeutic actions are unclear, recent studies have revealed mechanisms of ISG-mediated HBV suppression [18,19]. Interestingly, a recent study also found that activation of IFN-α signaling in the liver of HBV transgenic mice promotes the functional differentiation of intrahepatically primed CD8⁺ T cells [20]. After PEG-IFN-α treatment in humans, patients exhibit few changes in CD8⁺ T cells but increased proliferation and activation of CD56^bright NK cells, which would suggest a role for these cells in HBV control [21]. However, additional studies characterizing treatment-induced immunological changes are needed to better understand IFN-α-mediated HBV inhibition mechanisms.

IFN-γ

Cytotoxic CD8⁺ T cells restrict HBV by killing infected hepatocytes and suppressing viral replication via cytokine-mediated noncytopathic mechanisms. Analysis of chimpanzees and transgenic mice demonstrated that IFN-γ produced by CD8⁺ T cells activates ISGs in the liver and suppresses HBV replication [22,23]. Despite its crucial role in HBV control, clinical trials that tested therapeutic IFN-γ intervention failed to demonstrate efficacy against chronic HBV infection [24]. Although not effective as a standalone therapy, IFN-γ remains an important marker of T cell functionality for other HBV immunotherapies due to its essential role in viral clearance.

IFN-λ

Unlike the more universally expressed IFN-α/β receptor, the IFN-λs utilize a unique receptor that is restricted to barrier surfaces, including epithelial cells of the intestinal, respiratory and reproductive tracts, and hepatocytes [25]. Notwithstanding differential receptor expression and signaling kinetics, the ISG expression profile of IFN-λ-driven antiviral responses is similar to that activated by IFN-α/β [26]. Due to preferential induction in hepatocytes, IFN-λ may be more relevant than IFN-α/β for HBV control [27], and its powerful and localized antiviral role suggested that this cytokine might be effective therapeutically. However, although PEG-IFN-λ1 therapy reduced viral loads in chronic HBV patients, the reduction achieved during the treatment duration was inferior post-treatment compared to PEG-IFN-α [28].

Innate immune cells

HBV interactions with the various subsets of innate immune cells may impact the establishment of chronic infection. Similar to the capacity of HBV to initiate IFN responses, the relationship between HBV and several innate cell types is debated. The diversity of results observed may be explained by these cells’ unique roles in different stages of chronic HBV disease (HBeAg-positive chronic infection, HBeAg-positive chronic hepatitis, HBeAg-negative chronic infection, HBeAg-negative chronic hepatitis) and the tissue (liver or blood) analyzed.

Dendritic cells (DC)

The exact role of DCs in chronic HBV infection remains unclear. Several studies reported distinct findings regarding altered DC number or function [29-34], which is likely due to diverse patient study populations and the use of different experimental approaches [35]. Recent analyses of plasmacytoid DC (pDC) function showed impaired maturation and function in chronic HBV patients [33,36]. pDCs efficiently produce IFN upon recognizing viral proteins or nucleic acids, so their dysfunction might play a vital role in HBV persistence. Additionally, HBV-exposed DCs and BDCA1⁺ DCs from infected patients were recently found to be poor stimulators of IFN-γ production by natural killer (NK) cells, indicating a link between DC impairment and NK cell function in chronic HBV [37].

NK cells

NK cells constitute up to 40-50% of human liver lymphocytes and can have two opposing functions. On the one hand, they have cytotoxic capacity independent of antigen recognition and can produce IFN-γ that down-regulates HBV replication. However, these cells also negatively regulate adaptive immune responses by inhibiting or deleting T cells [38]. This negative regulation occurs through inhibitory cytokines, cytolytic enzymes, and signaling via death receptor ligands such as TRAIL [38]. Liver-resident NK cells show impaired IFN-γ production in chronic HBV patients, implying dysfunctional antiviral activity [39]. Dysregulated expression or function of NK cell-activating (NKG2D) or inhibitory (NKG2A) receptors may further contribute to HBV persistence [40-43]. Additionally, HBV-specific CD8⁺ T cells from chronic HBV patients display an abnormal expression of the apoptosis-inducing receptor TRAIL-R2, rendering them sensitive to depletion by TRAIL-expressing NK cells [44]. Blocking this receptor on hepatic T cells increases HBV-specific T cell responses ex vivo [44], which further indicates that the immune suppressive function of NK cells might predominate over their antiviral activities in chronic hepatitis B.

Macrophages

Liver macrophages can be divided into monocyte-derived infiltrating macrophages and Kupffer cells, which are tissue-resident macrophages in the liver sinusoids. Kupffer cells have crucial phagocytic and scavenger functions and represent the first line of immune defense in the liver [45]. These cells are also important regulators of inflammation and liver damage, as Kupffer cell scavenger function reduces inflammation and immunopathology in an HBV mouse model [46]. Several studies have shown that HBV can alter macrophage function and cytokine expression in a manner that may impact viral infection or persistence [2,47-49]. Recent work also demonstrated that liver [50] and lymphatic [51] macrophages promote experimental HBV therapeutic vaccine efficacy, further supporting a role for these cells in linking innate and adaptive immunity to HBV.

Myeloid-derived suppressor cells (MDSCs)

MDSCs are immature myeloid cells that consist of two principal subsets, monocytic (mMDSCs) and granulocytic (gMDSCs) [52]. Increased MDSC number in the peripheral blood of patients with chronic HBV was initially described in two reports [53,54], and in both, the increased MDSC frequency was associated with lower liver immunopathology. CD14⁺HLA-DR^−/low mMDSCs produce IL-10, which is responsible for their suppression of T cell function [53]. The CD14⁻ gMDSC population inhibits both HBV-specific and bystander T cells in a partially arginase-1-dependent manner [54]. Importantly, it was recently found that HBeAg induces mMDSC expansion, which reduces T cell function through the indoleamine 2,3-dioxygenase pathway [55]. Further investigation of these findings could lead to a combination therapy that blocks MDSC-mediated immunosuppression and therefore favors T cell antiviral immune functions.

Mucosal-associated invariant T (MAIT) cells

MAIT cells are T cells with an invariant TCR alpha chain and restricted beta chains [56]. These cells can be activated independently of their TCR and are considered innate-like cells [57]. MAIT cells are highly abundant in the human liver and represent up to 40% of T cells [58], but their interaction with HBV has only recently been investigated and is not fully understood. While one study reported normal MAIT cell numbers in the blood that had increased activation levels [59], others found decreased MAIT cells in the blood [60-62], which were functionally impaired and had elevated PD-1 levels [61,62]. More recently, Liu et al. examined MAIT cells in liver tissue and blood and found reduced numbers in both compartments, and the cells displayed MR1-dependent anti-HBV activity but altered activation and TCR-mediated expansion [63].

Innate immunity therapeutics

The failure of HBV to activate innate immunity and evidence of HBV IFN sensitivity has led to therapeutic interventions targeting IFN system activation. Two main treatment strategies to overcome the lack of innate immune stimulation have emerged: exogenous IFN and pattern recognition receptor (PRR) agonists. Various clinical trials studying IFN-stimulating PRR agonists are ongoing or were recently completed (Table 1). PRRs such as RIG-I-like receptors (RLR) and Toll-like receptors (TLR) are critical for activating innate immunity as they recognize and respond to infections by binding pathogen-associated molecular patterns and propagating antiviral signal transduction pathways. In theory, agonist-mediated PRR stimulation could overcome the lack of an innate response to HBV infection by activating IFN production, and has the potential to be more targeted, effective, and safe compared to exogenous IFN administration. Therapies that aim to activate IFN responses could be particularly beneficial as the passive failure of HBV-mediated IFN induction suggests that viral interference with innate immune-stimulating therapeutic interventions will be unlikely.

Table I.

Recent ongoing and completed PRR agonist clinical trials

Drug	Target	Research Phase	NCT Number*	Status	Company/Sponsor
inarigivir	RIG-I	Phase II Phase II	NCT02751996 NCT03932513	completed (02/10/2020) terminated (12/21/2019)	F-star Therapeutics, Inc.
vesatolimod	TLR7	Phase II Phase II	NCT02166047 NCT02579382	completed (10/20/2016) completed (05/03/2019)	Gilead Sciences
AL-034	TLR7	Phase I	NCT03285620	completed (11/14/2018)	Alios Biopharma Inc.
RO7020531	TLR7	Phase I Phase I	NCT03530917 NCT02956850	completed (05/15/2019) recruiting (expected 05/22/2021)	Hoffmann-La Roche
TQ-A3334	TLR7	Phase II	NCT04180150	recruiting (expected 12/31/2021)	Chia Tai Tianqing Pharmaceutical Group Co., Ltd.
selgantolimod	TLR8	Phase II Phase II	NCT03491553 NCT03615066	completed (8/10/2020) completed (05/28/2020)	Gilead Sciences
HepTcell/IC31	TLR9	Phase II	NCT04684914	recruiting (expected 02/2023)	Altimmune, Inc.
RO7020531 + RO7049389	TLR7 + CpAM	Phase II	NCT04225715	recruiting (expected 08/19/2023)	Hoffmann-La Roche
TQ-A3334 + TQ-B2450	TLR7 + PD-L1	Phase II	NCT04202653	not yet recruiting (expected 06/30/2021)	Tongji Hospital

^*https://clinicaltrials.gov

RIG-I

RLR agonists have been considered for chronic HBV therapies due to the cytoplasmic expression of their receptors in hepatocytes. Interestingly, a recent study found that RIG-I activation inhibits the formation and accelerates the decay of HBV covalently closed circular DNA (cccDNA) in hepatocytes [64]. The oral RIG-I/NOD2 agonist inarigivir (SB9200) reduced serum HBV DNA and antigens in the WHV woodchuck model [65], and was evaluated in combination with tenofovir in a completed Phase II clinical trial (Table 1). In 2020 however, inarigivir studies were discontinued following unexpected adverse events in a Phase IIb trial.

TLR7

Multiple TLR7 agonists have been investigated as the receptor is highly expressed by pDCs, and these cells potently produce IFN-α upon TLR7 stimulation [66]. Vesatolimod (GS-9620) yielded promising anti-HBV activity in chimpanzees and woodchucks [67,68]. Despite these results in animal models, vesatolimod treatment in humans showed no significant HBsAg decline either alone or in patients virally suppressed with nucleo(s/t)ide analogues [69-71]. AL-034 completed Phase I trials in 2018 and awaits further study (Table 1). RO7020531 completed a Phase I trial in chronic HBV patients in 2019 and is currently being further evaluated for safety and pharmacodynamics in another Phase I study, which is expected to finish in 2021. Another TLR7 agonist, TQ-A3334, is recruiting for a Phase II trial scheduled to be completed in 2021.

TLR8

Compared to TLR7 stimulation, TLR8 agonists preferentially induce the expression of IL-12 and other inflammatory cytokines from monocytes and myeloid DCs [66]. These cytokines may further stimulate IFN-γ production by other immune cells, which as discussed earlier, has important anti-HBV activity. Selgantolimod (GS-9688) is an oral TLR8 agonist that recently showed promising results in the WHV woodchuck model [72], and has been evaluated in two completed Phase II clinical trials alone and in combination with other antiviral therapies (Table 1). Recently published results have shown selgantolimod to be well-tolerated and to induce dose-dependent increases in serum cytokines [73,74]. Interestingly, human PBMCs from healthy and chronic HBV patients treated with selgantolimod in vitro exhibit enhanced DC and mononuclear phagocyte IL-12 production and increased frequencies of activated NK cells, MAIT cells, CD4⁺ follicular helper T cells, and HBV-specific CD8⁺ T cells expressing IFN-γ [75].

TLR9

CpG oligodeoxynucleotides (CpGs) have TLR9 agonist activity and were recently evaluated as adjuvants for HBV therapeutic vaccines [76]. The immunostimulatory properties of CpGs in the liver may in part derive from the fact that they promote the formation of structures known as iMATEs (intrahepatic myeloid cell aggregates associated with T cell expansion), which stimulate the population expansion of cytotoxic T lymphocytes in liver infection [77]. In the WHV woodchuck model, combined treatment with CpG and entecavir efficiently suppressed viral load and synergistically reduced serum HBsAg [78]. In late 2020, Altimmune, Inc. began a Phase II trial investigating the treatment of chronic HBV patients with an HBV peptide-based immunotherapeutic adjuvanted by the TLR9 agonist IC31 (Table 1).

Combination therapies

Despite mixed results in completed clinical trials, innate immunity therapeutic research has continued to advance (Table 1). Combination therapy with the TLR7 agonist RO7020531 and an HBV capsid allosteric modulator RO7049389 (CpAM) was effective in a preclinical AAV-based mouse model of HBV infection [79], and a phase II trial studying RO7020531 combined with RO7049389 began in July 2020 and is expected to be completed in March 2023. In a study not yet recruiting, the TLR7 agonist TQ-A3334 will be tested in combination with TQ-B2450, a humanized monoclonal antibody targeting programmed cell death ligand-1 (PD-L1). Checkpoint inhibitors targeting PD-1 or PD-L1 may restore impaired T cell function in chronically infected patients, and although their potential efficacy is currently unclear, a pilot study has provided promising results [80]. Combination approaches targeting innate and adaptive immunity may improve TLR agonist function; for example, treatment of human PBMCs with the TLR8 agonist selgantolimod led to a selection of MDSCs expressing higher levels of PD-L1 despite a beneficial reduction in MDSC frequency in vitro [75]. In January 2021, Gilead Sciences announced intentions to initiate Phase II trials investigating a combination of selgantolimod, antiviral siRNA VIR-2218 (Vir Biotechnology), and a commercially sourced PD-1 inhibitor. Preclinical studies of dual-acting agonists, such as the TLR2/7 agonist CL413 and the TLR7/8 agonist R848, have shown broader proinflammatory cytokine induction and enhanced efficacy compared to the single-acting TLR7 agonist CL264 or TLR9 agonist CpG-B in HBV-infected primary human hepatocytes [81]. It may be that the simultaneous activation of multiple PRRs yields a more robust cytokine spectrum capable of controlling HBV [81].

Conclusions

The passive evasion of IFN activation by HBV implies that therapeutic approaches aiming to stimulate these pathways will not be restrained by mechanisms of viral evasion or resistance, yet, the success of these approaches clinically has been thus far limited. Though PEG-IFN-α treatment can lead to significant sustained reductions in viremia and serum HBsAg in some cases, further work is needed to understand the mechanisms by which these positive clinical outcomes are achieved. This phenomenon nevertheless motivates future therapeutic advancements utilizing innate immune stimulation. To continue to improve upon previous work, it may be necessary to focus more specifically on the interplay of the innate and adaptive immune systems (Figure 1). It is possible that cell-specific targeting of DCs, NK cells, or MDSCs might reactivate the virus-specific T cells that these cells regulate [39,44,54]. Altering macrophage function to enhance T cell activation or to limit immunopathology is another tactic to consider [48]. Further, it is clear that cytokines such as IFN-α/β, -γ, and -λ can reduce HBV replication or promote adaptive immunity [82], so strategies that aim to activate innate immune cell production of these and other cytokines warrant continued evaluation. Combining these approaches with other methods to activate adaptive immunity, such as therapeutic vaccines, may also increase clinical efficacy. While an HBV cure has remained elusive, recent advancements nevertheless call for optimism and the need for continued basic and translational efforts focused on innate immune stimulation.

Figure 1. Therapeutic impacts of innate immune cells and cytokines on HBV replication.

Stimulating IFN-α/β or IFN-λ production by hepatocytes or DCs with RLR or TLR agonists, or activating IFN-γ production by T cells, NK cells, or MAIT cells may directly inhibit HBV replication in hepatocytes. In addition to activating innate antiviral mechanisms, IFN-α/β production may also stimulate virus-specific intrahepatic T cell responses. These T cell responses may be further enhanced by inhibiting the negative regulatory activities of MDSCs and NK cells or by altering macrophage function, which may also reduce immunopathology.