Advertisement

Effect of variants in LGP2 on MDA5-mediated activation of interferon response and suppression of hepatitis D virus replication

  • Author Footnotes
    † These authors have equal contribution.
    Nadine Gillich
    Footnotes
    † These authors have equal contribution.
    Affiliations
    Department of Infectious Diseases, Molecular Virology, Heidelberg University, Heidelberg, Germany

    Division of Virus-Associated Carcinogenesis, German Cancer Research Center (DKFZ), Heidelberg, Germany
    Search for articles by this author
  • Author Footnotes
    † These authors have equal contribution.
    Zhenfeng Zhang
    Footnotes
    † These authors have equal contribution.
    Affiliations
    Department of Infectious Diseases, Molecular Virology, Heidelberg University, Heidelberg, Germany
    Search for articles by this author
  • Marco Binder
    Affiliations
    Research Group “Dynamics of Early Viral Infection and the Innate Antiviral Response,” Division Virus-Associated Carcinogenesis, German Cancer Research Center (DKFZ), Heidelberg, Germany
    Search for articles by this author
  • Stephan Urban
    Correspondence
    Tel.: +49-(0)6221-56-4902; fax: +49-(0)6221-56-1946
    Affiliations
    Department of Infectious Diseases, Molecular Virology, Heidelberg University, Heidelberg, Germany

    German Center for Infection Research (DZIF) – Heidelberg Partner Site, Heidelberg, Germany
    Search for articles by this author
  • Ralf Bartenschlager
    Correspondence
    Corresponding authors. Addresses: INF 344, 1st. Floor, 69120 Heidelberg, Germany. Tel.: +49-(0)6221-56-4569; fax: +49-(0)6221-56-4570
    Affiliations
    Department of Infectious Diseases, Molecular Virology, Heidelberg University, Heidelberg, Germany

    Division of Virus-Associated Carcinogenesis, German Cancer Research Center (DKFZ), Heidelberg, Germany

    German Center for Infection Research (DZIF) – Heidelberg Partner Site, Heidelberg, Germany
    Search for articles by this author
  • Author Footnotes
    † These authors have equal contribution.
Open AccessPublished:September 21, 2022DOI:https://doi.org/10.1016/j.jhep.2022.08.041

      Highlights

      • LGP2 binds HDV RNA.
      • LGP2 RNA binding and ATPase functions are essential for full MDA5-mediated IFN activation by HDV and viral suppression.
      • LGP2 Q425R predominates in Africans, mediating higher basal and faster HDV-induced IFN responses.
      • LGP2 Q425R leads to stronger HDV suppression and could explain the attenuated course of hepatitis D in African populations.
      • LGP2 Q425R promotes formation of stable MDA5–RNA complexes.

      Background & Aims

      Retinoic acid inducible gene I (RIG-I)-like receptors (RLRs), including RIG-I, melanoma differentiation-associated protein 5 (MDA5), and laboratory of genetics and physiology 2 (LGP2), sense viral RNA to induce the antiviral interferon (IFN) response. LGP2, unable to activate the IFN response itself, modulates RIG-I and MDA5 signalling. HDV, a small RNA virus causing the most severe form of viral hepatitis, is sensed by MDA5. The mechanism underlying IFN induction and its effect on HDV replication is unclear. Here, we aimed to unveil the role of LGP2 and clinically relevant variants thereof in these processes.

      Methods

      RLRs were depleted in HDV susceptible HepaRGNTCP cells and primary human hepatocytes. Cells were reconstituted to express different LGP2 versions. HDV and IFN markers were quantified in a time-resolved manner. Interaction studies among LGP2, MDA5, and RNA were performed by pull-down assays.

      Results

      LGP2 is essential for the MDA5-mediated IFN response induced upon HDV infection. This induction requires both RNA binding and ATPase activities of LGP2. The IFN response only moderately reduced HDV replication in resting cells but profoundly suppressed cell division-mediated HDV spread. An LGP2 variant (Q425R), predominating in Africans who develop less severe chronic hepatitis D, mediated detectably higher basal and faster HDV-induced IFN response as well as stronger HDV suppression. Mechanistically, LGP2 RNA binding was a prerequisite for the formation of stable MDA5–RNA complexes. MDA5 binding to RNA was enhanced by the Q425R LGP2 variant.

      Conclusions

      LGP2 is essential to mount an antiviral IFN response induced by HDV and stabilises MDA5–RNA interaction required for downstream signalling. The natural Q425R LGP2 is a gain-of-function variant and might contribute to an attenuated course of hepatitis D.

      Impact and implications

      HDV is the causative pathogen of chronic hepatitis D, a severe form of viral hepatitis that can lead to cirrhosis and hepatocellular carcinoma. Upon infection, the human immune system senses HDV and mounts an antiviral interferon (IFN) response. Here, we demonstrate that the immune sensor LGP2 cooperates with MDA5 to mount an IFN response that represses HDV replication. We mapped LGP2 determinants required for IFN system activation and characterised several natural genetic variants of LGP2. One of them reported to predominate in sub-Saharan Africans can accelerate HDV-induced IFN responses, arguing that genetic determinants, possibly including LGP2, might contribute to slower disease progression in this population. Our results will hopefully prompt further studies on genetic variations in LGP2 and other components of the innate immune sensing system, including assessments of their possible impact on the course of viral infection.

      Graphical abstract

      Keywords

      Introduction

      RIG-I-like receptors (RLRs), including retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA5), and laboratory of genetics and physiology 2 (LGP2), sense non-self, double-stranded (ds) RNA to mount a powerful antiviral response by induction of interferons (IFNs) and IFN-stimulated genes (ISGs). RLRs comprise (i) a DECH box helicase domain necessary for the coordination of dsRNA binding and ATP hydrolysis to induce conformational changes upon RNA recognition and (ii) a C-terminal domain mediating initial RNA binding and conferring substrate specificity (Fig. 1A).
      • Bruns A.M.
      • Horvath C.M.
      LGP2 synergy with MDA5 in RLR-mediated RNA recognition and antiviral signaling.
      ,
      • Lassig C.
      • Hopfner K.P.
      Discrimination of cytosolic self and non-self RNA by RIG-I-like receptors.
      In addition, MDA5 and RIG-I contain N-terminal tandem repeats of caspase activation and recruitment domains (CARDs) for the initiation of downstream signalling events. Upon dsRNA binding, RLRs undergo conformational changes resulting in MDA5 and RIG-I oligomerisation and release of their CARDs to allow binding of mitochondrial antiviral-signalling protein (MAVS) for subsequent signalling. As LGP2 lacks the tandem CARDs, it is unable to mount an IFN response. However, LGP2 supports MDA5 signalling by promoting MDA5 filament formation on RNA to expose MDA5’s CARDs.
      • Bruns A.M.
      • Leser G.P.
      • Lamb R.A.
      • Horvath C.M.
      The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5-RNA interaction and filament assembly.
      ,
      • Duic I.
      • Tadakuma H.
      • Harada Y.
      • Yamaue R.
      • Deguchi K.
      • Suzuki Y.
      • et al.
      Viral RNA recognition by LGP2 and MDA5, and activation of signaling through step-by-step conformational changes.
      Interaction between the CARDs of MDA5/RIG-I and the CARD of MAVS activates the transcription factors NF-κB and IFN regulatory factors 3 and 7 (IRF3 and IRF7, respectively). These induce the expression and secretion of type I and III IFNs that bind to their cognate receptor inducing the expression of hundreds of antiviral ISGs in both infected and noninfected bystander cells.
      Figure thumbnail gr1
      Fig. 1Both MDA5 and LGP2 are required for IFN response activation and HDV suppression in HepaRGNTCP cells.
      (A) Schematic representation of the RLRs and their domains modified from Onomoto et al.
      • Onomoto K.
      • Onoguchi K.
      • Yoneyama M.
      Regulation of RIG-I-like receptor-mediated signaling: interaction between host and viral factors.
      (B) HepaRGNTCP cell pools with given stable KO or NT guide control cells were stimulated for 24 h with 200 IU/ml IFN-α before harvesting for Western blot analysis. (C–E) Cell lines in (B) were infected with HDV. Intracellular IFN-β, IFN-λ1, RSAD2, and HDV RNA levels at 3 days p.i. were measured using RT-qPCR (n = 3) (C). HDV-positive cells at 5 and 17 days p.i. were visualised by IF of HDAg and quantified using ImageJ (Rockville, USA) (4 images per sample) (n = 3) (D). Representative images used for the quantification in (D) are depicted in (E). Scale bar: 200 μm. Circles indicate 3 biological replicates. Statistical analysis was performed using ordinary 1-way ANOVA of log-transformed data and Dunnett’s multiple-comparisons test with single pooled variance. ∗p ≤0.05; ∗∗p ≤0.01; ∗∗∗∗p ≤0.0001 (respective KO compared with NT control). CARD, caspase activation and recruitment domain; CTD, C-terminal domain; HDAg, hepatitis delta antigen; Hel, helicase domain; IF, immunofluorescence; IFN, interferon; KO, knockout; LGP2, laboratory of genetics and physiology 2; MDA5, melanoma differentiation-associated protein 5; NT, non-targeting; p.i., post infection; RIG-I, retinoic acid inducible gene I; RLR, RIG-I-like receptor; RT-qPCR, real-time quantitative PCR. (This figure appears in color on the web.)
      Because IFN response is critical in controlling viral infections, genetic variations of involved signalling molecules can profoundly affect the outcomes of infections as shown, for example, for MDA5 in HCV,
      • Hoffmann F.S.
      • Schmidt A.
      • Dittmann Chevillotte M.
      • Wisskirchen C.
      • Hellmuth J.
      • Willms S.
      • et al.
      Polymorphisms in melanoma differentiation-associated gene 5 link protein function to clearance of hepatitis C virus.
      IRF7 in SARS-CoV-2,6 or Mx1 in influenza virus infection.
      • Chen Y.
      • Graf L.
      • Chen T.
      • Liao Q.
      • Bai T.
      • Petric P.P.
      • et al.
      Rare variant MX1 alleles increase human susceptibility to zoonotic H7N9 influenza virus.
      Several genetic LGP2 variants have also been identified, with varying frequency in different human populations,
      • Vasseur E.
      • Patin E.
      • Laval G.
      • Pajon S.
      • Fornarino S.
      • Crouau-Roy B.
      • et al.
      The selective footprints of viral pressures at the human RIG-I-like receptor family.
      but their function in IFN activation and impact on viral replication have not been investigated.
      HDV has a single-stranded (ss) circular RNA genome (∼1,700 nucleotides) that forms a rod-like quasi dsRNA structure owing to extensive intramolecular base-pairing.
      • Wang K.S.
      • Choo Q.L.
      • Weiner A.J.
      • Ou J.H.
      • Najarian R.C.
      • Thayer R.M.
      • et al.
      Structure, sequence and expression of the hepatitis delta (delta) viral genome.
      HDV encodes only 1 protein in 2 isoforms: small and large hepatitis delta antigen (S- and L-HDAg) required for viral replication and virion assembly, respectively.
      • Kuo M.Y.
      • Chao M.
      • Taylor J.
      Initiation of replication of the human hepatitis delta virus genome from cloned DNA: role of delta antigen.
      • Chang F.L.
      • Chen P.J.
      • Tu S.J.
      • Wang C.J.
      • Chen D.S.
      The large form of hepatitis delta antigen is crucial for assembly of hepatitis delta virus.
      • Modahl L.E.
      • Lai M.M.
      The large delta antigen of hepatitis delta virus potently inhibits genomic but not antigenomic RNA synthesis: a mechanism enabling initiation of viral replication.
      • Glenn J.S.
      • Watson J.A.
      • Havel C.M.
      • White J.M.
      Identification of a prenylation site in delta virus large antigen.
      HDV is a satellite virus that needs HBV envelope proteins for assembly of virions that enter hepatocytes via the shared entry receptor sodium taurocholate cotransporting polypeptide (NTCP).
      • Yan H.
      • Zhong G.
      • Xu G.
      • He W.
      • Jing Z.
      • Gao Z.
      • et al.
      Sodium taurocholate cotransporting polypeptide is a functional receptor for human hepatitis B and D virus.
      HDV replicates in the nucleus,
      • Tavanez J.P.
      • Cunha C.
      • Silva M.C.
      • David E.
      • Monjardino J.
      • Carmo-Fonseca M.
      Hepatitis delta virus ribonucleoproteins shuttle between the nucleus and the cytoplasm.
      and progeny genomes form ribonucleoprotein (RNP) complexes with HDAg. These are exported into the cytoplasm for subsequent budding at HBV envelope protein-containing membranes.
      Owing to large gaps in epidemiologic data, the number of HDV-infected individuals is poorly defined.
      • Chen H.Y.
      • Shen D.T.
      • Ji D.Z.
      • Han P.C.
      • Zhang W.M.
      • Ma J.F.
      • et al.
      Prevalence and burden of hepatitis D virus infection in the global population: a systematic review and meta-analysis.
      • Miao Z.
      • Zhang S.
      • Ou X.
      • Li S.
      • Ma Z.
      • Wang W.
      • et al.
      Estimating the global prevalence, disease progression, and clinical outcome of hepatitis delta virus infection.
      • Stockdale A.J.
      • Kreuels B.
      • Henrion M.Y.R.
      • Giorgi E.
      • Kyomuhangi I.
      • de Martel C.
      • et al.
      The global prevalence of hepatitis D virus infection: systematic review and meta-analysis.
      It is estimated that at least 12 million HBV carriers are coinfected with HDV. Until recently, the only treatment option for chronic hepatitis D (CHD) was pegylated IFN-α, but this therapy has limited responses and high relapse rates.
      • Heidrich B.
      • Yurdaydin C.
      • Kabaçam G.
      • Ratsch B.A.
      • Zachou K.
      • Bremer B.
      • et al.
      Late HDV RNA relapse after peginterferon alpha-based therapy of chronic hepatitis delta.
      Although HDV-targeting antivirals such as lonafarnib are under clinical investigation, the situation has changed with the recent conditional approval of Hepcludex/bulevirtide (formerly Myrcludex B), an entry inhibitor blocking virus binding to NTCP.
      • Urban S.
      • Bartenschlager R.
      • Kubitz R.
      • Zoulim F.
      Strategies to inhibit entry of HBV and HDV into hepatocytes.
      This drug causes strong reduction of HDV viral load under monotherapy and a much faster decline of HDV in combination therapy with pegylated IFN-α.
      • Wedemeyer H.
      • Schöneweis K.
      • Bogomolov P.O.
      • Voronkova N.
      • Chulanov V.P.
      • Stepanova T.M.
      • et al.
      GS-13 – final results of a multicenter, open-label phase 2 clinical trial (MYR203) to assess safety and efficacy of myrcludex B in cwith PEG-interferon alpha 2a in patients with chronic HBV/HDV co-infection.
      The underlying mechanism of this synergy is unknown.
      Aiming to decipher how HDV activates, and is controlled by, the IFN response, we previously reported that HDV is sensed by MDA5.
      • Zhang Z.
      • Filzmayer C.
      • Ni Y.
      • Sültmann H.
      • Mutz P.
      • Hiet M.S.
      • et al.
      Hepatitis D virus replication is sensed by MDA5 and induces IFN-beta/lambda responses in hepatocytes.
      Here, we show that LGP2 is essential to mount the MDA5-triggered IFN response by HDV. We dissected the LGP2 determinants required for this induction and report a natural gain-of-function LGP2 variant predominating in Sub-Saharan Africans, enabling faster IFN activation and stronger HDV suppression.

      Materials and methods

      Cell lines and virus stocks

      For the generation of stable knockout (KO) cell pools, HepaRG cells stably expressing NTCP (HepaRGNTCP) were transduced with lentiviruses encoding individual guide RNAs and Cas9. Stable knockin cell pools were generated analogously but using lentiviruses encoding different LGP2 variants or mutants. Cryopreserved primary human hepatocytes (PHHs) were obtained from BioIVT (West Sussex, United Kingdom) and Biopredic (Saint Grégoire, France). HDV stocks were produced by cotransfection of HuH7 cells with pJC126
      • Gudima S.
      • Chang J.
      • Moraleda G.
      • Azvolinsky A.
      • Taylor J.
      Parameters of human hepatitis delta virus genome replication: the quantity, quality, and intracellular distribution of viral proteins and RNA.
      encoding HDV antigenome and pT7HB2.7
      • Sureau C.
      • Guerra B.
      • Lee H.
      The middle hepatitis B virus envelope protein is not necessary for infectivity of hepatitis delta virus.
      encoding HBV envelope proteins. Stocks were purified by heparin affinity chromatography as described earlier.
      • Lempp F.A.
      • Mutz P.
      • Lipps C.
      • Wirth D.
      • Bartenschlager R.
      • Urban S.
      Evidence that hepatitis B virus replication in mouse cells is limited by the lack of a host cell dependency factor.
      Further details related to materials and methods are provided in the supplementary information and supplementary CTAT table.

      Results

      Both MDA5 and LGP2 are required for IFN activation and HDV suppression in HepaRGNTCP cells and PHHs

      To study the role of LGP2 and MDA5 in the context of HDV infection, we used HepaRGNTCP cells expressing RLRs and, ectopically, the viral receptor NTCP, rendering these cells highly susceptible to infection (Fig. 1B). To determine the contribution of each RLR to IFN activation induced by HDV, we generated HepaRGNTCP KO cell pools lacking RIG-I, MDA5, or LGP2 (Fig. 1B). These cell lines were infected with HDV to monitor (i) the IFN response and (ii) HDV replication. Real-time quantitative PCR (RT-qPCR) quantification of IFN-β, IFN-λ1, and the ISG radical S-adenosyl methionine domain containing 2 (RSAD2) indicated that KO of either LGP2 or MDA5, but not RIG-I, suppressed innate immune activation profoundly (Fig. 1C; mock treatment shown in Fig. S1A). Consistently, quantification of hepatitis delta antigen (HDAg) positive cells at 5 and 17 days post infection (p.i.) demonstrated elevated HDV replication in LGP2KO and MDA5KO cells, especially at 17 days p.i. (Fig. 1D and E).
      To verify these results in PHHs, we transiently knocked down RLR expression by transduction with lentiviruses encoding RLR-specific short hairpin RNAs (shRNAs) (2 shRNAs for each RLR) (Fig. 2A). Also in these cells, depletion of LGP2 and MDA5 decreased HDV-induced IFN response profoundly, whereas depletion of RIG-I had a much weaker effect (Fig. 2B; mock treatment shown in Fig. S1B). Although the effect on HDV replication was very moderate, it was best visible for MDA5 knockdown 11 days p.i. as determined by immunofluorescence (IF) (Fig. 2C and D). This moderate effect on viral replication might be because later time points such as those used with HepaRGNTCP cells could not be analysed in PHHs owing to cell death. Nevertheless, similar impairment of HDV-induced IFN response was observed with PHHs from 2 additional donors (Fig. S2), arguing against a donor-specific phenotype. In conclusion, these results identify LGP2 as an important factor for mounting an HDV-induced IFN response that impairs viral replication.
      Figure thumbnail gr2
      Fig. 2Critical role of LGP2 for IFN activation by HDV in PHHs.
      During seeding, cells were transduced with purified lentiviruses encoding an NT control shRNA (shNT), or two different shRNAs targeting each RLR. Cells were infected with HDV or mock infected 24 h after lentiviral transduction. (A) Knockdown efficiency was determined by RT-qPCR at 6 days post transduction. (B) Intracellular IFN-β, IFN-λ1, RSAD2, and HDV RNA levels at 5 days p.i. were measured using RT-qPCR. (C) HDV-positive cells at 5 and 11 days p.i. were visualised by IF of HDAg and quantified using ImageJ (3 images per sample). (D) Representative images used for quantification in (C). Scale bar: 200 μm. Circles indicate 2 technical replicates. HDAg, hepatitis delta antigen; IF, immunofluorescence; IFN, interferon; LGP2, laboratory of genetics and physiology 2; MDA5, melanoma differentiation-associated protein 5; NT, non-targeting; p.i., post infection; PHH, primary human hepatocyte; RIG-I, retinoic acid inducible gene I; RLR, RIG-I-like receptor; RSAD2, radical S-adenosyl methionine domain containing 2; RT-qPCR, real-time quantitative PCR. (This figure appears in color on the web.)

      Both ATPase and RNA binding functions of LGP2 are critical for induction of the IFN response and HDV suppression in HepaRGNTCP cells

      The ATPase and RNA binding functions of LGP2 have been described as being critical for MDA5 signalling enhancement.
      • Bruns A.M.
      • Leser G.P.
      • Lamb R.A.
      • Horvath C.M.
      The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5-RNA interaction and filament assembly.
      Therefore, we investigated the role of these 2 activities in the context of HDV infection. We generated a single (K634E; ΔRB∗single) and a triple (K605E/K634E/K651E; ΔRB∗triple) LGP2 mutant with alterations in amino acids reported to be involved in LGP2 RNA binding (Fig. 3A and B).
      • Li X.
      • Ranjith-Kumar C.T.
      • Brooks M.T.
      • Dharmaiah S.
      • Herr A.B.
      • Kao C.
      • et al.
      The RIG-I-like receptor LGP2 recognizes the termini of double-stranded RNA.
      ,
      • Pippig D.A.
      • Hellmuth J.C.
      • Cui S.
      • Kirchhofer A.
      • Lammens K.
      • Lammens A.
      • et al.
      The regulatory domain of the RIG-I family ATPase LGP2 senses double-stranded RNA.
      Regarding LGP2 ATPase function, we targeted residue K30 in the Walker A motif (K30A, ΔATPase) required for ATP hydrolysis.
      • Bamming D.
      • Horvath C.M.
      Regulation of signal transduction by enzymatically inactive antiviral RNA helicase proteins MDA5, RIG-I, and LGP2.
      We reconstituted HepaRGNTCP-LGP2KO cells with either HA-LGP2 wild type (WT) or individual mutants thereof containing additional silent nucleotide changes in the guide RNA-targeting sequence to avoid cleavage by Cas9. We assessed RNA binding of HA-LGP2 WT and mutants using 2 different approaches. First, cell lysates were incubated with biotin-labelled high-molecular-weight (HMW) poly(I:C) coated to streptavidin beads, and poly(I:C)-bound LGP2 was quantified by Western blot (Fig. S3A). Second, cells were infected with HDV and lysed 5 days later, and hemagglutinin (HA) -tagged LGP2 was enriched by HA-specific immunoprecipitation (IP). The amount of cocaptured HDV RNA was quantified by RT-qPCR (Fig. S3B). Both assays confirmed almost complete or partial loss of RNA binding for the ΔRB∗triple mutant and the ΔRB∗single mutant, respectively, whereas the ATPase mutant retained RNA binding (Fig. S3).
      Figure thumbnail gr3
      Fig. 3Both ATPase and RNA binding functions of LGP2 are critical for IFN activation and HDV suppression in HepaRGNTCP cells.
      (A) Schematic representation of LGP2 and positions of introduced mutations. (B) HepaRGNTCP-LGP2KO cells were stably reconstituted with HA-tagged LGP2 WT, an LGP2 lacking ATPase activity (K30A; ΔATPase), or having reduced RNA binding because of a single (K634E; ΔRB∗single) or triple mutation (K605E/K634E/K651E; ΔRB∗triple). Cells transduced with an empty vector (empty) served as reference. Expression of HA-tagged LGP2 WT and mutants was determined by Western blot using HA-specific antibody. (C–F) Cell lines in (B) and HepaRGNTCP expressing a NT guide RNA (NT control) were infected with HDV. RNA levels of IFN-β, IFN-λ1, and RSAD2 at 2, 3, 5, 11, and 17 days p.i. were quantified by RT-qPCR (n = 3). ‘m’ refers to mock-infected cells harvested 2 days after mock inoculation. n.d., not detected (C). HDV RNA levels were determined in the same way (D). HDV-positive cells at 5 and 17 days p.i. were analysed by IF of HDAg and quantified using ImageJ (4 images per sample) (n = 3) (E). Representative images used for quantification are depicted in (F). Scale bar: 200 μm. Circles indicate 3 biological replicates. Statistical analysis was performed using ordinary 1-way ANOVA of log-transformed data and Dunnett’s multiple-comparisons test with a single pooled variance. ∗p ≤0.05; ∗∗p ≤0.01; ∗∗∗p ≤0.001; ∗∗∗∗p ≤0.0001 (cell lines compared with LGP2 WT at the respective time point). CTD, C-terminal domain; HDAg, hepatitis delta antigen; Hel, helicase domain; IF, immunofluorescence; IFN, interferon; LGP2, laboratory of genetics and physiology 2; NT, non-targeting; p.i., post infection; RSAD2, radical S-adenosyl methionine domain containing 2; RT-qPCR, real-time quantitative PCR; WT, wild type. (This figure appears in color on the web.)
      Next, we infected these cell lines with HDV and measured IFN responses by quantifying the amounts of IFN-β, IFN-λ1, and RSAD2 mRNAs up to 17 days p.i. (Fig. 3C). For these experiments we included a non-targeting (NT) guide RNA-expressing HepaRGNTCP cell line. These cells express only endogenous LGP2 corresponding to around 1% of the LGP2 level detected in cells expressing this protein exogenously (Fig. S4). Throughout the observation period, the ΔRB∗triple mutant did not induce an IFN response, underlining the importance of the RNA binding function of LGP2 to enhance MDA5 signalling. Likewise, loss of the ATP hydrolysis function of LGP2 impaired the upregulation of the innate signalling pathway upon stimulation with HDV, although a residual IFN response was still detectable (Fig. 3C).
      Analysis of HDV RNA replication by RT-qPCR revealed the strongest suppression in cells stably expressing LGP2 WT (Fig. 3D). In contrast, HDV RNA levels and the number of HDAg-positive cells were much less suppressed in cells expressing the LGP2 mutants (Fig. 3D–F). We note that cells expressing LGP2 WT exogenously showed faster IFN response and stronger HDV repression than NT control cells (Fig. 3C–F) likely because endogenous LGP2 and MDA5 need to be upregulated upon HDV infection to allow full signalling (Fig. S4). Taken together, these results reveal the importance of LGP2 ATPase and especially RNA binding function for promoting MDA5-dependent IFN response and suppression of HDV replication.

      LGP2-induced IFN response efficiently suppresses cell division-mediated HDV spread

      Our previous study has demonstrated that intracellular HDV RNA replication in resting cells is poorly sensitive to virus-induced IFN response.
      • Zhang Z.
      • Filzmayer C.
      • Ni Y.
      • Sültmann H.
      • Mutz P.
      • Hiet M.S.
      • et al.
      Hepatitis D virus replication is sensed by MDA5 and induces IFN-beta/lambda responses in hepatocytes.
      This might explain why LGP2-mediated IFN response only moderately inhibited HDV replication (Fig. 1C–E and Fig. 2B–D). However, we recently found that the IFN response profoundly suppresses cell division-mediated HDV spread.
      • Zhang Z.
      • Ni Y.
      • Lempp F.A.
      • Walter L.
      • Mutz P.
      • Bartenschlager R.
      • et al.
      Hepatitis D virus-induced interferon response and administered interferons control cell division-mediated virus spread.
      Therefore, we tested the role of LGP2 in suppressing HDV spread via this pathway (Fig. 4A). To allow cell division-mediated HDV spread, infected cells were split at 5 days p.i. and cultivated for further 6 days followed by IF staining of HDAg at 5 days p.i. and 6 days post passage (Fig. 4A–C). As expected, in cells with intact IFN activation, the number of HDV-positive cells declined dramatically after cell division. This reduction was significantly alleviated in LGP2KO cells and MDA5KO cells (Fig. 4B and C).
      Figure thumbnail gr4
      Fig. 4Efficient block of cell division-mediated HDV spread by LGP2 and dependency on LGP2 RNA binding and ATPase activity.
      (A) Schematic of the experimental set-up to investigate cell division-mediated HDV spread. Cells were infected with HDV and split 1:8 at 5 days p.i. to allow cell division-mediated HDV spread. At 5 days p.i. (P0) or 6 days post passage (P1), HDV-positive cells were visualised by HDAg IF and quantified using ImageJ (6 images per sample). (B and C) HepaRGNTCP cells with given stable KO or containing NT guide RNA described in were used (n = 4). (D, E) HepaRGNTCP-LGP2KO cells reconstituted with HA-tagged LGP2 WT, or LGP2 mutants were used. Quantification of HDAg-positive cells and fold reduction compared with P0 are shown in (B) and (D) (n = 4). Representative images used for quantifications are shown in (C) and (E). Scale bars: 200 μm. Circles indicate 4 biological replicates. Statistical analysis was performed using ordinary 1-way ANOVA of log-transformed P0 to P1 fold changes of the respective cell lines and Dunnett’s multiple-comparisons test with a single pooled variance. ∗∗∗p ≤0.001; ∗∗∗∗p ≤0.0001 (respective cell lines compared with NT control cells (B) or LGP2 WT (D)). HDAg, hepatitis delta antigen; IF, immunofluorescence; KO, knockout; LGP2, laboratory of genetics and physiology 2; MDA5, melanoma differentiation-associated protein 5; NT, non-targeting; p.i., post infection; RIG-I, retinoic acid inducible gene I; WT, wild type. (This figure appears in color on the web.)
      To corroborate the role of RNA binding and the ATPase function of LGP2 in suppressing cell division-mediated HDV spread, we used the same approach with HepaRGNTCP-LGP2KO cells reconstituted with LGP2 WT, ΔATPase, or ΔRB∗triple mutants. As shown in Fig. 4D and E, LGP2 WT overexpression exhibited even stronger inhibition than cells expressing endogenous LGP2 (compare Fig. 4B NT control with Fig. 4D LGP2 WT) similar to the results in resting cells (Fig. 3D–F). Moreover, the loss of HDV-positive cells was similar in HepaRGNTCP-LGP2KO cells and cells overexpressing ΔRB∗triple mutant (compare Fig. 4B LGP2KO with Fig. 4C ΔRB∗triple). Notably, the ATPase mutant mediating some IFN response upon HDV infection (Fig. 3C) exhibited stronger inhibition of cell division-mediated HDV spread than the ΔRB∗triple mutant (Fig. 4D). These results indicate that cell division-mediated HDV spread is highly sensitive to the LGP2/MDA5-mediated IFN response.

      Higher suppression of HDV replication and increased IFN response by the natural LGP2 variant Q425R

      Recently, HDV-infected Africans were found to less often develop cirrhosis than Europeans,
      • Roulot D.
      • Brichler S.
      • Layese R.
      • BenAbdesselam Z.
      • Zoulim F.
      • Thibault V.
      • et al.
      Origin, HDV genotype and persistent viremia determine outcome and treatment response in patients with chronic hepatitis delta.
      indicating that severity of CHD might be influenced by the patient’s ethnic origin. Several naturally occurring single nucleotide polymorphisms (SNPs) in genes encoding RLRs were found with different frequencies in healthy African, Asian, and European individuals.
      • Vasseur E.
      • Patin E.
      • Laval G.
      • Pajon S.
      • Fornarino S.
      • Crouau-Roy B.
      • et al.
      The selective footprints of viral pressures at the human RIG-I-like receptor family.
      These include 3 abundant, non-synonymous LGP2 SNPs resulting in an amino acid exchange either within the helicase (Q425R or N461S) or within the pincer domain (R523Q) of LGP2 (Fig. 5A and B, upper panel).
      Figure thumbnail gr5
      Fig. 5Evidence for enhanced HDV suppression by the natural LGP2 variant Q425R.
      (A) Table depicting 3 abundant, non-synonymous SNPs within the LGP2 coding region, corresponding amino acid changes in the LGP2 protein and SNP frequency in different human populations according to Vasseur et al.
      • Vasseur E.
      • Patin E.
      • Laval G.
      • Pajon S.
      • Fornarino S.
      • Crouau-Roy B.
      • et al.
      The selective footprints of viral pressures at the human RIG-I-like receptor family.
      (B) Schematic representation of the amino acid changes within LGP2 caused by the SNPs is shown on the top. HepaRGNTCP-LGP2KO cells stably expressing either HA-tagged LGP2 WT, 1 of the polymorphic variants, or an empty control vector were generated. Expression of HA-tagged LGP2 variants was determined by Western blot. (C–E) Cell lines in (B) were infected with HDV. RNA levels of intracellular HDV at 2, 3, 5, 11, and 17 days p.i. were quantified by RT-qPCR (n = 3). (C). HDV-positive cells at 5 and 17 days p.i. were analysed by HDAg IF and quantified using ImageJ (6 images per sample) (n = 3) (D). Representative images are depicted in (E). Scale bar: 200 μm. Circles indicate 3 biological replicates. Statistical analysis was performed using ordinary 1-way ANOVA and Dunnett’s multiple-comparisons test of log-transformed data with a single pooled variance. ∗p ≤0.05; ∗∗p ≤0.01; ∗∗∗p ≤0.001; ∗∗∗∗p ≤0.0001 (respective cell lines compared with LGP2 WT). CTD, C-terminal domain; HDAg, hepatitis delta antigen; Hel, helicase domain; IF, immunofluorescence; LGP2, laboratory of genetics and physiology 2; p.i., post infection; RT-qPCR, real-time quantitative PCR; SNP, single nucleotide polymorphism; WT, wild type. (This figure appears in color on the web.)
      To examine the effect of these SNPs on LGP2 function and control of HDV replication, we generated stable cell pools expressing N-terminally HA-tagged LGP2 WT or any of the 3 variants in the context of HepaRGNTCP-LGP2KO cells (Fig. 5B, lower panel). These cells were infected with HDV, and viral replication was determined by quantifying HDV RNA and the number of HDAg-positive cells (Fig. 5C–E). The highest viral replication was observed with cells lacking LGP2 (empty vector). Of note, we observed the strongest decrease of HDV replication with cells expressing the natural LGP2 variant Q425R, which was best detectable in the IF-based assay (Fig. 5D–E), and also when measuring HDV RNA, although there the effect was only found at early time points p.i. (Fig. 5C). HDV suppression by LGP2 Q425R was better visible when low multiply of infection was applied (Fig. S5).
      Because LGP2 enhances MDA5-induced IFN response, we compared induction of IFN-β, IFN-λ1, and RSAD2 expression by HDV in HepaRGNTCP-LGP2KO cells reconstituted with LGP2 WT or LGP2 variants (Fig. 6). Of note, already in uninfected cells expressing LGP2 Q425R, we observed a trend towards elevated expression of IFNs and RSAD2 (Fig. 6). This increase was not attributable to different LGP2 expression levels that were well comparable among the different variants (Fig. 5B, lower panel). Although also at 2 days p.i, a slightly higher IFN response was observed in LGP2 Q425R expressing cells, this effect vanished at later time points between LGP2 WT and the 3 variants (Fig. 6).
      Figure thumbnail gr6
      Fig. 6LGP2 Q425R increases basal and HDV-induced IFN response.
      HepaRGNTCP-LGP2KO cells stably reconstituted with HA-tagged LGP2 WT, variants, or an empty control vector were infected with HDV. Intracellular RNA levels of IFN-β, IFN-λ1, and RSAD2 at 2, 3, 5, 11, and 17 days p.i. were quantified by RT-qPCR (n = 3). Uninfected cells were harvested 2 days after mock inoculation. Circles indicate 3 biological replicates. Statistical analysis was performed using ordinary 1-way ANOVA and Dunnett’s multiple-comparisons test of log-transformed data with a single pooled variance. ∗p ≤0.05; ∗∗p ≤0.01; ∗∗∗p ≤0.001; ∗∗∗∗p ≤0.0001 (respective cell lines compared with LGP2 WT). IFN, interferon; LGP2, laboratory of genetics and physiology 2; p.i., post infection; RSAD2, radical S-adenosyl methionine domain containing 2; RT-qPCR, real-time quantitative PCR; WT, wild type.
      To exclude a possible cell type-specific effect, we validated our results in a second, well-defined surrogate cell culture system, that is, A549 cells, that are capable of producing high amounts of IFN in response to RLR stimulation. We investigated MDA5-specific responses by KO of RIG-I and, as A549 cells produce very little, if any, LGP2, reconstituted these RIG-IKO cells with WT or Q425R LGP2 (Fig. S6A and B). Upon stimulation with HMW poly(I:C), a commonly used inducer of MDA5 signalling, we observed a massive enhancement of the IFN response in the presence of LGP2 WT, especially with low doses of stimulus, demonstrating a profound sensitisation of MDA5 signalling by LGP2 (Fig. S6C). This response was further enhanced by LGP2 Q425R, especially at basal level and with the lowest amount of RNA used (Fig. S6C). Taken together, these results demonstrate that LGP2 Q425R is a gain-of-function variant sensitising IFN induction by MDA5.

      Elevated basal IFN response by LGP2 Q425R is mediated by MDA5-dependent IFN induction

      Earlier studies have shown that endogenous RNA can be sensed by MDA5 and that such erroneous IFN induction, leading to interferonopathies, is prevented by the RNA-editing enzyme adenosine deaminase acting on RNA 1 (ADAR1).
      • Pestal K.
      • Funk C.C.
      • Snyder J.M.
      • Price N.D.
      • Treuting P.M.
      • Stetson D.B.
      Isoforms of RNA-editing enzyme ADAR1 independently control nucleic acid sensor MDA5-driven autoimmunity and multi-organ development.
      ,
      • Liddicoat B.J.
      • Piskol R.
      • Chalk A.M.
      • Ramaswami G.
      • Higuchi M.
      • Hartner J.C.
      • et al.
      RNA editing by ADAR1 prevents MDA5 sensing of endogenous dsRNA as nonself.
      Hypothesising that LGP2 Q425R increases basal IFN response via MDA5, we transiently knocked down MDA5 by lentiviral shRNA transduction in cells expressing LGP2 WT or Q425R and analysed basal IFN-λ1 and RSAD2 mRNA levels (Fig. S7A and C). Cells transduced with NT shRNA or expressing the LGP2 ΔRB∗triple mutant served as controls. In shNT control cells, LGP2 Q425R significantly enhanced mRNA levels of IFN-λ1 and RSAD2 relative to LGP2 WT, whereas knockdown of MDA5 blunted this response in all cell lines. This result corroborated the dependency on MDA5 for basal IFN response and indicated higher activity of the LGP2 Q425R variant (Fig. S7C).
      To investigate further a potential contribution of endogenous RNAs to basal LGP2/MDA5-dependent IFN response, we augmented the unedited levels of endogenous RNAs by transient knockdown of ADAR1 in these cell lines to enable increased sensing by MDA5 (Fig. S7B). As expected, ADAR1 knockdown significantly elevated the abundance of IFN-λ1 and RSAD2 mRNAs in cells expressing LGP2 WT or Q425R (>100-fold; Fig. S7D). Of note, LGP2 Q425R enhanced basal IFN response even more than did LGP2 WT. Interestingly, in cells expressing the LGP2 ΔRB∗triple mutant, no basal IFN was detected even upon ADAR1 depletion, indicating that unedited, endogenous RNA sensing by MDA5 also depends on LGP2, similar to a very recent report.
      • Stok J.E.
      • Oosenbrug T.
      • Ter Haar L.R.
      • Gravekamp D.
      • Bromley C.P.
      • Zelenay S.
      • et al.
      RNA sensing via the RIG-I-like receptor LGP2 is essential for the induction of a type I IFN response in ADAR1 deficiency.
      In summary, these results suggest that LGP2 Q425R likely enhances basal IFN response via MDA5-dependent RNA sensing.

      Natural LGP2 variants bind poly(I:C) and HDV RNA similar to LGP2 WT

      To determine the molecular mechanism how LGP2 enhances MDA5-mediated basal and HDV-induced IFN response and how this might be affected by the LGP2 Q425R variant, we first evaluated LGP2 binding efficiency to HDV RNA. As we had found that proper LGP2 RNA binding function was important for MDA5 activation by HDV, we hypothesised that altered LGP2 RNA binding activity might influence the MDA5 response. To address this assumption, we used pull-down assays using biotinylated HMW poly(I:C) and HA-LGP2–HDV RNA co-IP (5 days p.i.) as described above to determine binding of LGP2 WT and variants to poly(I:C) and HDV RNA, respectively. Binding of all the natural LGP2 variants to both RNA species was well comparable and similar to that of LGP2 WT (Fig. S8). These results suggest that enhanced IFN response induced by the natural LGP2 variant Q425R is unlikely owing to altered RNA binding activity, at least for those substrates used.

      LGP2 variant Q425R but not N461S and R523Q enhances recruitment of MDA5 to RNA ligands

      It has been reported that LGP2 assists MDA5 filament formation on RNA, at least in vitro.
      • Bruns A.M.
      • Leser G.P.
      • Lamb R.A.
      • Horvath C.M.
      The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5-RNA interaction and filament assembly.
      ,
      • Duic I.
      • Tadakuma H.
      • Harada Y.
      • Yamaue R.
      • Deguchi K.
      • Suzuki Y.
      • et al.
      Viral RNA recognition by LGP2 and MDA5, and activation of signaling through step-by-step conformational changes.
      Given that we found comparable RNA binding of LGP2 Q425R and WT, we assumed that the stronger IFN response of this variant might be attributable to enhanced recruitment of MDA5 to RNA ligands. To test this assumption, we investigated MDA5–RNA interaction in the presence of LGP2 WT, Q425R, or mutants that alter MDA5 signalling (ΔATPase and ΔRB∗triple). Respective cell lines were prestimulated with IFN to enhance endogenous MDA5 levels and used for MDA5 pull-down. However, capture of MDA5 was below the detection limit (data not shown), possibly because of the low affinity of available MDA5-specific antibody or HDV RNA being a poor MDA5 ligand. As an alternative, we used biotinylated HDV RNA to pull down MDA5, along with LGP2, similar to the poly(I:C) pull-down described above (Fig. S3). Although LGP2 WT, Q425R, and ΔATPase bound genomic and antigenomic HDV RNA, no MDA5 binding was detectable (Fig. S9). Assuming that HDV RNA amounts were not sufficient to detect MDA5 binding, we used HMW poly(I:C) as a surrogate substrate and quantified MDA5 amounts bound to poly(I:C) in the presence of LGP2 WT, Q425R, ΔATPase, and ΔRB∗triple mutant (Fig. 7). MDA5 was coprecipitated with LGP2 WT and ΔATPase, but not with the ΔRB∗triple mutant, demonstrating that RNA binding of LGP2 is required for efficient recruitment of MDA5 to poly(I:C). Compared with WT, the LGP2 Q425R variant significantly enhanced binding of MDA5 to poly(I:C) (∼3-fold) (Fig. 7B, lower panel) but did not affect LGP2 binding to RNA (Fig. 7B, upper panel), consistent with the results described above (Fig. S8).
      Figure thumbnail gr7
      Fig. 7The natural LGP2 variant Q425R enhances MDA5 binding to RNA.
      (A and B) HepaRGNTCP-LGP2KO cells expressing HA-tagged LGP2 WT, the natural variant Q425R, or given LGP2 mutants were stimulated with 200 IU/ml IFN-α to induce endogenous MDA5 expression. After 24 h, cells were lysed, and cell lysates were incubated with biotin-coupled poly(I:C) captured on streptavidin-coated beads. Levels of RNA-bound MDA5 and HA-tagged LGP2 in the respective eluates as well as their abundance in lysates were determined by Western blot (A), and band intensities were quantified using the LabImage software package (Leipzig, Germany) (B). Values in (B) correspond to the ratio of captured protein amounts normalised to protein abundance in the respective lysates (n = 4). Circles indicate 4 biological replicates. Statistical analysis was performed using ordinary 1-way ANOVA and Dunnett’s multiple-comparisons test of log-transformed data with a single pooled variance. ∗p ≤0.05; ∗∗p ≤0.01; ∗∗∗p ≤0.001 (respective cell lines compared with LGP2 WT). IFN, interferon; LGP2, laboratory of genetics and physiology 2; MDA5, melanoma differentiation-associated protein 5; WT, wild type.
      In conclusion, these results underline the overall weak RNA binding efficiency of MDA5 relative to that of LGP2. Moreover, our data suggest that the natural LGP2 Q425R variant enhances the IFN response by elevating the recruitment of MDA5 to RNA ligands.

      Discussion

      In this study we evaluated the role of LGP2 and naturally occurring variants thereof in IFN activation and HDV suppression. Consistent with our earlier report,
      • Zhang Z.
      • Filzmayer C.
      • Ni Y.
      • Sültmann H.
      • Mutz P.
      • Hiet M.S.
      • et al.
      Hepatitis D virus replication is sensed by MDA5 and induces IFN-beta/lambda responses in hepatocytes.
      we found that MDA5, but not RIG-I, plays a central role in the induction of the IFN response by HDV. Notably, we demonstrate that MDA5 is unable to mount an IFN response in the absence of LGP2, similar to what has been described for HCV infection.
      • Hei L.
      • Zhong J.
      Laboratory of genetics and physiology 2 (LGP2) plays an essential role in hepatitis C virus infection-induced interferon responses.
      Thus, LGP2 is an essential cofactor of MDA5 with induction of the IFN system requiring both ATPase and RNA binding functions of LGP2. The weak RNA binding efficiency of MDA5 together with the fact that HDV RNA appears to be a poor RLR substrate did not allow direct detection of MDA5 HDV interaction in our cell system. Nevertheless, it underlines the importance of LGP2 in sensing HDV RNA to initiate MDA5 signalling. How LGP2 senses the circular HDV RNA genome remains to be determined. It is also unclear where in the cell HDV RNA is sensed, because the virus replicates in the nucleus, whereas LGP2 and MDA5 reside in the cytoplasm. However, a small fraction of these pattern recognition receptors (PRRs) may localise to the nucleus as shown for RIG-I
      • Liu G.
      • Lu Y.
      • Thulasi Raman S.N.
      • Xu F.
      • Wu Q.
      • Li Z.
      • et al.
      Nuclear-resident RIG-I senses viral replication inducing antiviral immunity.
      that is thought to activate MAVS at distinct perinuclear regions.
      The LGP2/MDA5-mediated antiviral response induced by HDV had rather moderate effects on viral replication in resting hepatocytes. Consistent with the reported enhanced IFN sensitivity of HDV in dividing cells,
      • Zhang Z.
      • Ni Y.
      • Lempp F.A.
      • Walter L.
      • Mutz P.
      • Bartenschlager R.
      • et al.
      Hepatitis D virus-induced interferon response and administered interferons control cell division-mediated virus spread.
      its suppression was more pronounced in the context of cell division-mediated viral spread.
      • Giersch K.
      • Bhadra O.D.
      • Volz T.
      • Allweiss L.
      • Riecken K.
      • Fehse B.
      • et al.
      Hepatitis delta virus persists during liver regeneration and is amplified through cell division both in vitro and in vivo.
      In this respect, the MDA5–LGP2-induced IFN response might be of particular importance to reduce HDV propagation by this route, although the exact mechanism remains elusive. As the effect of IFN is strongest during cell division,
      • Zhang Z.
      • Ni Y.
      • Lempp F.A.
      • Walter L.
      • Mutz P.
      • Bartenschlager R.
      • et al.
      Hepatitis D virus-induced interferon response and administered interferons control cell division-mediated virus spread.
      we assume that nucleases might be involved in HDV RNA degradation. Altered ADAR1-mediated editing by induction of the IFN-inducible p150 ADAR1 isoform or by proteins affecting ADAR1 activity such as Z-DNA binding protein 1 (ZBP1), oligoadenylate synthetase (OAS), and protein kinase RNA-activated (PKR) could also be of relevance. ISGs with broad antiviral activity and being upregulated upon HDV infection, such as II27, IFI44, ISG15, MX1, and RSAD2,
      • Zhang Z.
      • Filzmayer C.
      • Ni Y.
      • Sültmann H.
      • Mutz P.
      • Hiet M.S.
      • et al.
      Hepatitis D virus replication is sensed by MDA5 and induces IFN-beta/lambda responses in hepatocytes.
      might contribute to suppression of HDV cell division-mediated spread as well. Further studies are required to clarify this interesting aspect of HDV biology.
      Most individuals with CHD treated with IFN only mount a partial response, and in the majority of cases, HDV rebounds after cessation of therapy, indicating poor long-term efficacy of IFN treatment. However, in combination with the entry inhibitor bulevirtide, IFN might be of special importance to limit cell division-mediated HDV spread in already infected hepatocytes, which is not addressed by bulevirtide. Thus, the combination of bulevirtide, which blocks reinfection and spread to new hepatocytes, with simultaneous IFN treatment that limits cell division-mediated spread might offer a new avenue towards curative therapy.
      Our study provides evidence that the Q425R variant of LGP2 might be a natural gain-of-function variant. The observation that more MDA5 interacted with poly(I:C) in the presence of Q425R LGP2 argues for enhanced MDA5 recruitment or stronger RNA interaction as compared with LGP2 WT or both. Although discrimination between these possibilities requires sophisticated biochemical studies, we note that faster and more stable MDA5 recruitment likely sensitises the IFN system, resulting in faster IFN response, especially at low initial levels of MDA5, LGP2, and HDV RNA, which might be less relevant later in infection when expression of ISGs, including LGP2 and MDA5, is upregulated. Moreover, the kinetics of activation of the IFN response might also be critical to determine the outcome of infection with rapid induction, allowing early control of virus infection, which reduces the risk of chronicity.
      • Hoffmann F.S.
      • Schmidt A.
      • Dittmann Chevillotte M.
      • Wisskirchen C.
      • Hellmuth J.
      • Willms S.
      • et al.
      Polymorphisms in melanoma differentiation-associated gene 5 link protein function to clearance of hepatitis C virus.
      ,
      • Zhang Z.
      • Filzmayer C.
      • Ni Y.
      • Sültmann H.
      • Mutz P.
      • Hiet M.S.
      • et al.
      Hepatitis D virus replication is sensed by MDA5 and induces IFN-beta/lambda responses in hepatocytes.
      ,
      • Han Z.
      • Nogusa S.
      • Nicolas E.
      • Balachandran S.
      • Taylor J.
      Interferon impedes an early step of hepatitis delta virus infection.
      ,
      • Loske J.
      • Röhmel J.
      • Lukassen S.
      • Stricker S.
      • Magalhães V.G.
      • Liebig J.
      • et al.
      Pre-activated antiviral innate immunity in the upper airways controls early SARS-CoV-2 infection in children.
      Even though the IFN response and HDV repression mediated by LGP2 Q425R were only slightly higher than those by LGP2 WT, we note that our cell culture system supports only single-round infection. We assume that in an infected liver where continuous HDV reinfection occurs, subtle differences in the onset and magnitude of the IFN response might accumulate, giving rise to a more pronounced impact on HDV propagation and titres in infected individuals and disease severity, the latter correlating with higher viral load.
      • Yamashiro T.
      • Nagayama K.
      • Enomoto N.
      • Watanabe H.
      • Miyagi T.
      • Nakasone H.
      • et al.
      Quantitation of the level of hepatitis delta virus RNA in serum, by real-time polymerase chain reaction – and its possible correlation with the clinical stage of liver disease.
      A recent study analysing HDV-infected individuals showed that Africans seem to have a better prognosis and a less severe course of CHD, independent of the HDV genotype, as compared with Europeans.
      • Roulot D.
      • Brichler S.
      • Layese R.
      • BenAbdesselam Z.
      • Zoulim F.
      • Thibault V.
      • et al.
      Origin, HDV genotype and persistent viremia determine outcome and treatment response in patients with chronic hepatitis delta.
      Additionally, HDV-infected Africans (having predominantly LGP2 Q425R) have overall lower viral load and higher portion with undetectable viraemia than other ethnic populations,
      • Roulot D.
      • Brichler S.
      • Layese R.
      • BenAbdesselam Z.
      • Zoulim F.
      • Thibault V.
      • et al.
      Origin, HDV genotype and persistent viremia determine outcome and treatment response in patients with chronic hepatitis delta.
      ,
      • Spaan M.
      • Carey I.
      • Bruce M.
      • Shang D.
      • Horner M.
      • Dusheiko G.
      • et al.
      Hepatitis delta genotype 5 is associated with favourable disease outcome and better response to treatment compared to genotype 1.
      in line with the lower HDV replication in cells expressing LGP2 Q425R compared with LGP2 WT. These observations argue for a potential link between the LGP2 Q425R polymorphism and less severe outcome of CHD. Although this is an attractive possibility, another study linked less severe CHD to HDV genotype 5 compared with genotype 1, although all genotype 5-infected patients in this study were Africans and genotype 1-infected patients were mostly from Europe.
      • Spaan M.
      • Carey I.
      • Bruce M.
      • Shang D.
      • Horner M.
      • Dusheiko G.
      • et al.
      Hepatitis delta genotype 5 is associated with favourable disease outcome and better response to treatment compared to genotype 1.
      It remains to be determined to which extent genetic and environmental factors as well as HDV genotypes contribute to less severe outcome of CHD. Moreover, further studies are required to clarify whether polymorphisms in LGP2 or other relevant immune factor genes (as reported earlier for acute viral infections such as influenza
      • Chen Y.
      • Graf L.
      • Chen T.
      • Liao Q.
      • Bai T.
      • Petric P.P.
      • et al.
      Rare variant MX1 alleles increase human susceptibility to zoonotic H7N9 influenza virus.
      and COVID-19
      • Zhang Q.
      • Bastard P.
      • Liu Z.
      • Le Pen J.
      • Moncada-Velez M.
      • Chen J.
      • et al.
      Inborn errors of type I IFN immunity in patients with life-threatening COVID-19.
      ) might account for disease progression in HDV-infected patients.

      Abbreviations

      ADAR1, adenosine deaminase acting on RNA 1; CARD, caspase activation and recruitment domain; CHD, chronic hepatitis D; ds, double-stranded; HA, hemagglutinin; HDAg, hepatitis delta antigen; HMW, high molecular weight; IF, immunofluorescence; IFN, interferon; IP, immunoprecipitation; IRF3, IFN regulatory factor 3; IRF7, IFN regulatory factor 7; ISG, IFN-stimulated gene; KO, knockout; LGP2, laboratory of genetics and physiology 2; L-HDAg, large hepatitis delta antigen; MAVS, mitochondrial antiviral-signalling protein; MDA5, melanoma differentiation-associated protein 5; NT, non-targeting; NTCP, sodium-taurocholate cotransporting polypeptide; OAS, oligoadenylate synthetase; PHH, primary human hepatocyte; p.i., post infection; PKR, protein kinase RNA-activated; PRR, pattern recognition receptor; RIG-I, retinoic acid inducible gene I; RLR, RIG-I-like receptor; RNP, ribonucleoprotein; RSAD2, radical S-adenosyl methionine domain containing 2; RT-qPCR, real-time quantitative PCR; S-HDAg, small hepatitis delta antigen; shRNA, short hairpin RNA; SNP, single nucleotide polymorphism; ss, single-stranded; WT, wild type; ZBP1, Z-DNA binding protein 1.

      Financial support

      This work was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – Projektnummer 272983813 – TRR 179, TP9 to RB, TP11 to MB, and TP15 to SU. RB and SU were also supported by the Forum Gesundheitsstandort Baden-Württemberg, TP5 (Az 32-5400/58/2).

      Conflicts of interest

      RB holds patents in hepatitis C virus replicon technology. SU holds patents in bulevirtide/Hepcludex/Myrcludex. Other authors who have taken part in this study declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.
      Please refer to the accompanying ICMJE disclosure forms for further details.

      Authors’ contributions

      Conception and design of the study: NG, ZZ, SU, RB.
      Generation, collection, assembly, analysis, and interpretation of data: NG, ZZ, MB, SU, RB.
      Drafting of the manuscript: NG, ZZ, RB.
      Critical revision and approval of the final version of the manuscript: NG, ZZ, MB, SU, RB.

      Data availability statement

      Further details related to materials and methods are provided in the Supplementary information and Supplementary CTAT methods table.

      Acknowledgements

      We are grateful to Lisa Walter, Talisa Richardt, Sandra Wüst, Joschka Willemsen, Pascal Mutz, and Antje Reuter for experimental support and provision of materials. We are also indebted to Volker Lohmann and Alessia Ruggieri for continuous intellectual input.

      Supplementary data

      The following are the supplementary data to this article:

      References

        • Bruns A.M.
        • Horvath C.M.
        LGP2 synergy with MDA5 in RLR-mediated RNA recognition and antiviral signaling.
        Cytokine. 2015; 74: 198-206
        • Lassig C.
        • Hopfner K.P.
        Discrimination of cytosolic self and non-self RNA by RIG-I-like receptors.
        J Biol Chem. 2017; 292: 9000-9009
        • Bruns A.M.
        • Leser G.P.
        • Lamb R.A.
        • Horvath C.M.
        The innate immune sensor LGP2 activates antiviral signaling by regulating MDA5-RNA interaction and filament assembly.
        Mol Cell. 2014; 55: 771-781
        • Duic I.
        • Tadakuma H.
        • Harada Y.
        • Yamaue R.
        • Deguchi K.
        • Suzuki Y.
        • et al.
        Viral RNA recognition by LGP2 and MDA5, and activation of signaling through step-by-step conformational changes.
        Nucleic Acids Res. 2020; 48: 11664-11674
        • Hoffmann F.S.
        • Schmidt A.
        • Dittmann Chevillotte M.
        • Wisskirchen C.
        • Hellmuth J.
        • Willms S.
        • et al.
        Polymorphisms in melanoma differentiation-associated gene 5 link protein function to clearance of hepatitis C virus.
        Hepatology. 2015; 61: 460-470
        • Zhang Q.
        • Bastard P.
        • Liu Z.
        • Le Pen J.
        • Moncada-Velez M.
        • Chen J.
        • et al.
        Inborn errors of type I IFN immunity in patients with life-threatening COVID-19.
        Science. 2020; 370eabd4570
        • Chen Y.
        • Graf L.
        • Chen T.
        • Liao Q.
        • Bai T.
        • Petric P.P.
        • et al.
        Rare variant MX1 alleles increase human susceptibility to zoonotic H7N9 influenza virus.
        Science. 2021; 373: 918-922
        • Vasseur E.
        • Patin E.
        • Laval G.
        • Pajon S.
        • Fornarino S.
        • Crouau-Roy B.
        • et al.
        The selective footprints of viral pressures at the human RIG-I-like receptor family.
        Hum Mol Genet. 2011; 20: 4462-4474
        • Wang K.S.
        • Choo Q.L.
        • Weiner A.J.
        • Ou J.H.
        • Najarian R.C.
        • Thayer R.M.
        • et al.
        Structure, sequence and expression of the hepatitis delta (delta) viral genome.
        Nature. 1986; 323: 508-514
        • Kuo M.Y.
        • Chao M.
        • Taylor J.
        Initiation of replication of the human hepatitis delta virus genome from cloned DNA: role of delta antigen.
        J Virol. 1989; 63: 1945-1950
        • Chang F.L.
        • Chen P.J.
        • Tu S.J.
        • Wang C.J.
        • Chen D.S.
        The large form of hepatitis delta antigen is crucial for assembly of hepatitis delta virus.
        Proc Natl Acad Sci U S A. 1991; 88: 8490-8494
        • Modahl L.E.
        • Lai M.M.
        The large delta antigen of hepatitis delta virus potently inhibits genomic but not antigenomic RNA synthesis: a mechanism enabling initiation of viral replication.
        J Virol. 2000; 74: 7375-7380
        • Glenn J.S.
        • Watson J.A.
        • Havel C.M.
        • White J.M.
        Identification of a prenylation site in delta virus large antigen.
        Science. 1992; 256: 1331-1333
        • Yan H.
        • Zhong G.
        • Xu G.
        • He W.
        • Jing Z.
        • Gao Z.
        • et al.
        Sodium taurocholate cotransporting polypeptide is a functional receptor for human hepatitis B and D virus.
        Elife. 2012; 1e00049
        • Tavanez J.P.
        • Cunha C.
        • Silva M.C.
        • David E.
        • Monjardino J.
        • Carmo-Fonseca M.
        Hepatitis delta virus ribonucleoproteins shuttle between the nucleus and the cytoplasm.
        RNA. 2002; 8: 637-646
        • Chen H.Y.
        • Shen D.T.
        • Ji D.Z.
        • Han P.C.
        • Zhang W.M.
        • Ma J.F.
        • et al.
        Prevalence and burden of hepatitis D virus infection in the global population: a systematic review and meta-analysis.
        Gut. 2019; 68: 512-521
        • Miao Z.
        • Zhang S.
        • Ou X.
        • Li S.
        • Ma Z.
        • Wang W.
        • et al.
        Estimating the global prevalence, disease progression, and clinical outcome of hepatitis delta virus infection.
        J Infect Dis. 2020; 221: 1677-1687
        • Stockdale A.J.
        • Kreuels B.
        • Henrion M.Y.R.
        • Giorgi E.
        • Kyomuhangi I.
        • de Martel C.
        • et al.
        The global prevalence of hepatitis D virus infection: systematic review and meta-analysis.
        J Hepatol. 2020; 73: 523-532
        • Heidrich B.
        • Yurdaydin C.
        • Kabaçam G.
        • Ratsch B.A.
        • Zachou K.
        • Bremer B.
        • et al.
        Late HDV RNA relapse after peginterferon alpha-based therapy of chronic hepatitis delta.
        Hepatology. 2014; 60: 87-97
        • Urban S.
        • Bartenschlager R.
        • Kubitz R.
        • Zoulim F.
        Strategies to inhibit entry of HBV and HDV into hepatocytes.
        Gastroenterology. 2014; 147: 48-64
        • Wedemeyer H.
        • Schöneweis K.
        • Bogomolov P.O.
        • Voronkova N.
        • Chulanov V.P.
        • Stepanova T.M.
        • et al.
        GS-13 – final results of a multicenter, open-label phase 2 clinical trial (MYR203) to assess safety and efficacy of myrcludex B in cwith PEG-interferon alpha 2a in patients with chronic HBV/HDV co-infection.
        J Hepatol. 2019; 70: e81
        • Zhang Z.
        • Filzmayer C.
        • Ni Y.
        • Sültmann H.
        • Mutz P.
        • Hiet M.S.
        • et al.
        Hepatitis D virus replication is sensed by MDA5 and induces IFN-beta/lambda responses in hepatocytes.
        J Hepatol. 2018; 69: 25-35
        • Gudima S.
        • Chang J.
        • Moraleda G.
        • Azvolinsky A.
        • Taylor J.
        Parameters of human hepatitis delta virus genome replication: the quantity, quality, and intracellular distribution of viral proteins and RNA.
        J Virol. 2002; 76: 3709-3719
        • Sureau C.
        • Guerra B.
        • Lee H.
        The middle hepatitis B virus envelope protein is not necessary for infectivity of hepatitis delta virus.
        J Virol. 1994; 68: 4063-4066
        • Lempp F.A.
        • Mutz P.
        • Lipps C.
        • Wirth D.
        • Bartenschlager R.
        • Urban S.
        Evidence that hepatitis B virus replication in mouse cells is limited by the lack of a host cell dependency factor.
        J Hepatol. 2016; 64: 556-564
        • Li X.
        • Ranjith-Kumar C.T.
        • Brooks M.T.
        • Dharmaiah S.
        • Herr A.B.
        • Kao C.
        • et al.
        The RIG-I-like receptor LGP2 recognizes the termini of double-stranded RNA.
        J Biol Chem. 2009; 284: 13881-13891
        • Pippig D.A.
        • Hellmuth J.C.
        • Cui S.
        • Kirchhofer A.
        • Lammens K.
        • Lammens A.
        • et al.
        The regulatory domain of the RIG-I family ATPase LGP2 senses double-stranded RNA.
        Nucleic Acids Res. 2009; 37: 2014-2025
        • Bamming D.
        • Horvath C.M.
        Regulation of signal transduction by enzymatically inactive antiviral RNA helicase proteins MDA5, RIG-I, and LGP2.
        J Biol Chem. 2009; 284: 9700-9712
        • Zhang Z.
        • Ni Y.
        • Lempp F.A.
        • Walter L.
        • Mutz P.
        • Bartenschlager R.
        • et al.
        Hepatitis D virus-induced interferon response and administered interferons control cell division-mediated virus spread.
        J Hepatol. 2022; 77: 957-966
        • Roulot D.
        • Brichler S.
        • Layese R.
        • BenAbdesselam Z.
        • Zoulim F.
        • Thibault V.
        • et al.
        Origin, HDV genotype and persistent viremia determine outcome and treatment response in patients with chronic hepatitis delta.
        J Hepatol. 2020; 73: 1046-1062
        • Pestal K.
        • Funk C.C.
        • Snyder J.M.
        • Price N.D.
        • Treuting P.M.
        • Stetson D.B.
        Isoforms of RNA-editing enzyme ADAR1 independently control nucleic acid sensor MDA5-driven autoimmunity and multi-organ development.
        Immunity. 2015; 43: 933-944
        • Liddicoat B.J.
        • Piskol R.
        • Chalk A.M.
        • Ramaswami G.
        • Higuchi M.
        • Hartner J.C.
        • et al.
        RNA editing by ADAR1 prevents MDA5 sensing of endogenous dsRNA as nonself.
        Science. 2015; 349: 1115-1120
        • Stok J.E.
        • Oosenbrug T.
        • Ter Haar L.R.
        • Gravekamp D.
        • Bromley C.P.
        • Zelenay S.
        • et al.
        RNA sensing via the RIG-I-like receptor LGP2 is essential for the induction of a type I IFN response in ADAR1 deficiency.
        EMBO J. 2022; 41e109760
        • Hei L.
        • Zhong J.
        Laboratory of genetics and physiology 2 (LGP2) plays an essential role in hepatitis C virus infection-induced interferon responses.
        Hepatology. 2017; 65: 1478-1491
        • Liu G.
        • Lu Y.
        • Thulasi Raman S.N.
        • Xu F.
        • Wu Q.
        • Li Z.
        • et al.
        Nuclear-resident RIG-I senses viral replication inducing antiviral immunity.
        Nat Commun. 2018; 9: 3199
        • Giersch K.
        • Bhadra O.D.
        • Volz T.
        • Allweiss L.
        • Riecken K.
        • Fehse B.
        • et al.
        Hepatitis delta virus persists during liver regeneration and is amplified through cell division both in vitro and in vivo.
        Gut. 2019; 68: 150-157
        • Han Z.
        • Nogusa S.
        • Nicolas E.
        • Balachandran S.
        • Taylor J.
        Interferon impedes an early step of hepatitis delta virus infection.
        PLoS One. 2011; 6e22415
        • Loske J.
        • Röhmel J.
        • Lukassen S.
        • Stricker S.
        • Magalhães V.G.
        • Liebig J.
        • et al.
        Pre-activated antiviral innate immunity in the upper airways controls early SARS-CoV-2 infection in children.
        Nat Biotechnol. 2021; 40: 319-324
        • Yamashiro T.
        • Nagayama K.
        • Enomoto N.
        • Watanabe H.
        • Miyagi T.
        • Nakasone H.
        • et al.
        Quantitation of the level of hepatitis delta virus RNA in serum, by real-time polymerase chain reaction – and its possible correlation with the clinical stage of liver disease.
        J Infect Dis. 2004; 189: 1151-1157
        • Spaan M.
        • Carey I.
        • Bruce M.
        • Shang D.
        • Horner M.
        • Dusheiko G.
        • et al.
        Hepatitis delta genotype 5 is associated with favourable disease outcome and better response to treatment compared to genotype 1.
        J Hepatol. 2020; 72: 1097-1104
        • Onomoto K.
        • Onoguchi K.
        • Yoneyama M.
        Regulation of RIG-I-like receptor-mediated signaling: interaction between host and viral factors.
        Cell Mol Immunol. 2021; 18: 539-555