Human cytomegalovirus is a type of herpes virus that rarely causes symptoms in healthy people but can cause serious complications in unborn babies and in people with compromised immune systems, such as transplant recipients.

The virus has found ways to successfully evade the immune system, and once infected, the body retains the virus for life. It deploys an arsenal of proteins that bind to antibodies, specialized proteins the immune system uses to flag virus-infected cells for destruction. This prevents certain cells of the immune system, the natural killer cells, from recognizing and destroying virus-infected cells.

These immune-evading proteins are called viral Fc-gamma receptors, or vFcγRs. While it has been previously shown that these receptors are able to evade the immune system, it remained unknown how exactly they prevent natural killer cells from recognizing infected cells.

Now, Kolb et al. show that the cytomegalovirus deploys two vFcγRs called gp34 and gp68, which work together to block natural killer cells. The latter reduces the ability of natural killer cells to bind to antibodies on cytomegalovirus-infected cells. This paves the way for gp34 to pull virus proteins from the surface of the infected cell, making them inaccessible to the immune system. Neither protein fully protects virus-infected cells on its own, but together they are highly effective.

The experiments reveal further details about how cytomegalovirus uses two defense mechanisms simultaneously to outmaneuver the immune system. Understanding this two-part viral evasion system may help scientists to develop vaccines or new treatments that can protect vulnerable people from diseases caused by the cytomegalovirus.

Human cytomegalovirus (HCMV) constitutes the prototypical human pathogenic β-herpesvirus found worldwide with high immunoglobulin G (IgG) sero-prevalence rates of 56–94% depending on the respective countries (Zuhair et al., 2019). A hallmark of cytomegalovirus infection is the establishment of a lifelong persistence with recurring phases of latency and reactivation of productive infection and horizontal spread in presence of adaptive immune responses. HCMV encodes the largest known transcriptome of human viruses (Stern-Ginossar et al., 2012), giving rise to an equally large antigenic proteome including a huge and varied arsenal of immunoevasins (Berry et al., 2020; Hengel et al., 1998) that counteract immune recognition of infected cells facilitating virus persistence, shedding, and superinfection of sero-positive hosts. While primary HCMV infection of healthy individuals usually remains undetected, it can cause severe symptoms in the immunocompromised. Besides antiviral therapy involving nucleotide analogs, concentrated HCMV-immune IgG preparations (e.g. Cytotect) are used to prevent infection of immunocompromised patients including the prevention of congenital infection in primary infected pregnant women with varying degrees of success (Kagan et al., 2019; Revello et al., 2014). Generally, HCMV is observed to withstand a humoral immunity even replicating and disseminating in the presence of highly neutralizing immune sera in vitro and clinical isolates of HCMV tend to disseminate via cell-to-cell spread, thus avoiding an encounter with neutralizing antibodies (Falk et al., 2018). This, however, cannot fully explain the resistance of HCMV to a humoral response leading to virus dissemination across several organs. Notably, HCMV encodes a set of Fcγ-binding glycoproteins (viral FcγRs, vFcγRs) that have been shown to antagonize host FcγR activation by immune IgG (Corrales-Aguilar et al., 2014b). In this previous study, we showed that HCMV gp34, gp68 and HSV-1 gE/gI efficiently antagonize the activation of human FcγRs. By attacking the conserved Fc part of IgG, HSV-1 and HCMV are able to counteract potent antiviral immune responses including antibody-dependent cellular cytotoxicity (ADCC). As it has become increasingly evident that Fcγ mediated immune control is essential not only for the effectiveness of non-neutralizing but also neutralizing IgG antibodies (DiLillo et al., 2014; Forthal et al., 2013; Horwitz et al., 2017; Van den Hoecke et al., 2017), a mechanistic and causal analysis of this evasion process is highly warranted in the pursuit of better antibody based intervention strategies targeting herpesviruses and HCMV in particular. While several herpesviruses encode vFcγRs, HCMV is the only virus known so far that encodes more than one individual molecule with the capacity to bind Fcγ. Specifically, HCMV encodes four distinct molecules which share this ability: gp68 (UL119-118), gp34 (RL11), gp95 (RL12), and gpRL13 (RL13) (Atalay et al., 2002; Corrales-Aguilar et al., 2014a; Cortese et al., 2012; Sprague et al., 2008). Turning to other species, mouse CMV (MCMV) encoded m138 has been shown to bind Fcγ, yet it has more prominently been associated with a variety of unrelated functions proving m138 not to be a strict homolog of the HCMV counterparts (Arapović et al., 2009; Lenac et al., 2006). We recently found Rhesus CMV (RhCMV) to encode an Fcγ-binding protein in the Rh05 gene (RL11 gene family) seemingly more closely related to its HCMV analog (Kolb et al., 2019). This is supported by the fact that gpRh05, as HCMV vFcγRs gp34 and gp68, is able to generically antagonize activation of all macaque FcγRs. While it is clear that by targeting the invariant part of the key molecule of the humoral immune response, vFcγRs have the potential to manipulate a multitude of antibody mediated immune functions, their role in vivo has yet to be determined. While the function of HCMV vFcγRs gp34 and gp68 as antagonists of host FcγRs has been established (Corrales-Aguilar et al., 2014b), the underlying mechanism(s) had not been addressed yet. In recent years it has been shown that gp68 and gp34 are able to engage in antibody bipolar bridging (ABB) forming ternary complexes consisting of antigen, antibody, and vFcγR (Corrales-Aguilar et al., 2014a; Corrales-Aguilar et al., 2014b; Sprague et al., 2008). Moreover, gp68 has been shown to bind IgG in a 2:1 ratio and has the ability to internalize and translocate IgG to lysosomal compartments, while gp34 has been shown to form predominantly homo-dimeric structures (Ndjamen et al., 2016; Sprague et al., 2008). However, no studies have yet been performed in the context of HCMV infection investigating the coincident disposition of gp34 and gp68 at the plasma membrane and their functional interaction during the early and late phase of HCMV replication. Here, we show gp34 and gp68 to antagonize host FcγR activation by distinct but highly cooperative modes of Fcγ targeting, leading to efficient evasion from antibody mediated immune control by division of labor.

gp34 and gp68 simultaneously bind to distinct regions on IgG. gp68 binding to IgG has been mapped to the CH2–CH3 interdomain region of Fcγ (Sprague et al., 2008). Accordingly, in a first experiment we set out to narrow down the contact site of gp34 on IgG utilizing a methodology previously used to characterize HSV-1 gE and HCMV gp68 (Sprague et al., 2004; Sprague et al., 2008). To this end we infected CV-1 cells with recombinant vaccinia viruses (rVACV) encoding either human FcγRIIA, FcγRI or HCMV vFcγRs gp34 and gp68 (Sprague et al., 2008). After metabolic [³⁵S]-Met/Cys labeling, Fcγ-binding proteins were precipitated from cell lysates using CNBr-Sepharose coupled with human IgG1-Fc in its wild-type form (wtFc) or as a mutated variant (nbFc) with a scrambled CH2–CH3 interdomain amino acid sequence designed and provided by P. Bjorkman (Caltech, California, USA) (Sprague et al., 2004; Sprague et al., 2008). Expectedly, gp68 was only able to bind wtFc but not nbFc whereas gp34, comparable to human FcγRI, retained binding to both wtFc and nbFc (Figure 1A). While the high affinity FcγRI does not require the CH2–CH3 region to bind to the lower hinge of IgG, FcγRIIA and FcγRIII show lower affinity to monomeric IgG and utilize additional distinct residues for binding to Fcγ, which include the CH2 region adjacent to the lower hinge region of IgG (Shields et al., 2001; Sprague et al., 2008; Wines et al., 2000). Consequently, FcγRII exhibited reduced binding to nbFc (Figure 1A). Since gp34 showed intact binding similar to host FcγRI but not FcγRIIA we surmised that gp34 binding involves the hinge region of monomeric IgG. To validate this assumption further gp34 binding to a variant form of hIgG1 with point mutations in the lower hinge region (Leu234Ala and Leu235Ala, named LALA) exhibiting deficient interaction with human FcγRs (Hessell et al., 2007) was analyzed (Figure 1B). Cell lysates were incubated with either B12 wild-type antibody recognizing human immunodeficiency virus (HIV)-encoded gp120 or a B12-LALA variant. PGS-precipitation of FcγRI/CD64, exclusively recognizing the hinge region on IgG, was almost abrogated by the mutations in the LALA variant. Similarly, gp34 was sensitive to the mutations, albeit to a lower extent than FcγRI/CD64. As expected, the interaction of gp68 with B12 was unaffected by the LALA point mutations. Next, as gp34 and gp68 recognize non-overlapping regions on IgG we set out to test, if both molecules are able to simultaneously associate with IgG. To this end, we generated C-terminal His- or strep-tagged variants of gp34 and gp68 that lack their respective transmembrane and cytosolic domains (soluble gp34 and gp68, or sgp34 and sgp68). These proteins were produced by transfected 293 T cells and secreted into the cell culture supernatant (Figure 1—figure supplement 1). Supernatants of sgp34 and sgp68 producing cells were mixed with humanized anti-hCD20 IgG1 (Rituximab, or Rtx) before being precipitated via Ni²⁺-NTA-Sepharose beads. A non-Fcγ-binding point mutant of sgp34 (sgp34mtrp, W65F mutation) was identified and used as a control molecule (Figure 1—figure supplement 2). When precipitating His-tagged sgp68 in the presence of monomeric IgG, we observed co-precipitation of sgp34 but no co-precipitation of sgp34mtrp (Figure 1C). To further substantiate this observation, we designed an ELISA based setup using the same supernatants as above and performed the assay as depicted schematically in Figure 1D. In brief, titrated amounts of strep-tagged sgp34 or sgp34mtrp were immobilized on a 96-well adsorption plate (1° vFcγR) via pre-coated biotin. 1° vFcγRs were then incubated with either Rtx (IgG1) or a Rtx IgA isotype and incubated with His6-tagged sgp68 followed by detection with an anti-His-HRP antibody. Here, we observed dose-dependent binding of sgp68 to titrated amounts of immobilized sgp34. sgp68 could only be detected in the presence of IgG1 but not IgA. Furthermore, sgp68 binding in the presence of IgG1 and sgp34mtrp in this setup does not exceed background levels. Taken together, we conclude that gp34 and gp68 are able to simultaneously bind to IgG with gp34 binding to the upper hinge region and gp68 binding to the CH2–CH3 interface domain on IgG.

Figure 1 with 2 supplements see all

Download asset Open asset

Figure 1—source data 1

Relates to Figure 1D bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig1-data1-v2.xlsx

Download elife-63877-fig1-data1-v2.xlsx

gp34 and gp68 exhibit synergistic antagonization of FcγR activation. gp34 (RL11) and gp68 (UL119-118) have previously been shown to antagonize FcγR activation independently (Corrales-Aguilar et al., 2014b). As gp34 and gp68 appear to have a redundant inhibitory function and are able to simultaneously bind to IgG we utilized a previously established FcγR activation assay (Corrales-Aguilar et al., 2013) to assess their ability to antagonize FcγR activation individually or in combination. In brief, MRC-5 or human foreskin fibroblasts (HFF) were infected with HCMV AD169-pBAC2-derived mutant viruses lacking different combinations of vFcγRs. The other vFcγRs gp95 (RL12) and gpRL13 (RL13) were deliberately excluded in this study. RL13 (Cortese et al., 2012), a potent inhibitor of HCMV replication (Stanton et al., 2010), is notoriously mutated or absent from the HCMV genome in cell culture passages (Dargan et al., 2010) but has not yet been described to antagonize FcR activation. RL12 is one of the most polymorphic genes of HCMV (Dolan et al., 2004) with only minor surface expression in the context of AD169 infection (Corrales-Aguilar et al., 2014b). Due to remarkably low conservation of RL12 across HCMV strains, we focused on the well-conserved and previously described vFcγRs gp34 and gp68. Infected cells were incubated with titrated amounts of a HCMV-IgG hyperimmunoglobulin preparation (Cytotect) and co-cultured with BW5147 reporter cells expressing human FcγRIIIA as a chimeric molecule providing the ectodomain of the FcγR fused to the transmembrane domain and cytosolic tail of mouse CD3ξ (Corrales-Aguilar et al., 2013). In response to FcγR activation, reporter cells secrete mIL-2 which was then quantified by ELISA (Corrales-Aguilar et al., 2013), assay schematically shown in Figure 2—figure supplement 1. In line with previously published observations (Corrales-Aguilar et al., 2014b) we found that FcγRIIIA activation is antagonized efficiently by a virus co-expressing both gp34 and gp68 while viruses expressing either gp34 or gp68 show only minor or no antagonistic potential when compared with a mutant lacking all vFcγRs (Figure 2A). The extent of FcγR antagonization is visualized by area under curve (AUC) comparisons (Figure 2A bottom panel). This showed no statistically significant effect for singular vFcγRs, but a strong and significant antagonization by both vFcγRs in combination. To further confirm this key finding we measured degranulation of primary NK cells in response to vFcγR expression to assess antagonization of ADCC. Peripheral Blood Mononuclear Cells (PBMCs) from three donors were incubated with HCMV-infected opsonized human fibroblasts generated as above and CD107a positivity of NK cells was measured after 6 hr of co-culture. In line with the reporter cell assay, primary NK cell activation in response to target cells decorated with anti-HCMV antibodies was antagonized by a virus expressing both gp34 and gp68 with high significance, while the mutant viruses lacking gp34 or gp68 showed drastically lower antagonistic potential (Figure 2B). From these observations we conclude that gp34 and gp68 antagonize FcγRs by synergizing modes of action, suggesting cooperation, at least in the absence of gp95 (RL12) and gpRL13 (RL13).

Figure 2 with 1 supplement see all

Download asset Open asset

HCMV gp34 enhances internalization of immune complexes

As we observed gp34 and gp68 to work in concert to maximize FcγR inhibition, we set out to elucidate the mechanisms by which the vFcγRs achieve this synergistic effect. When comparing the amino acid sequences of the cytosolic domains of gp34 and gp68 it is revealed that both molecules encode distinct sorting motifs (Figure 3A; Atalay et al., 2002) (source: AD169 pBAC-2 sequence). The cytosolic tail of gp68 harbors an YXXΦ motif seven aa downstream of the transmembrane domain. Such a motif has been described as a marker for lysosomal targeting (Bonifacino and Traub, 2003) and is in line with the lysosomal translocation of gp68 shown in a previous study (Sprague et al., 2008). In HCMV gB, a non-Fcγ-binding HCMV encoded envelope glycoprotein, the location of two such motifs is further away from the transmembrane domain indicating it being destined for general endocytosis rather than immediate degradation, which is shared by another HCMV vFcγR, gpRL13, found primarily in intracellular compartments as well as HSV-1 gE, forming a heterodimeric gE/gI vFcγR described to internalize IgG and recycle back to the cell surface (Figure 3—figure supplement 1; Bonifacino and Traub, 2003; Cortese et al., 2012; Ndjamen et al., 2016; Sprague et al., 2004). Conversely, the cytosolic tail of gp34 encodes a di-leucine [D/E]XXX[LL/LI] motif which we find to be unique among surface-resident vFcγRs. Therefore we suspected gp34 and gp68 to possess different dynamics regarding IC internalization. As internalization studies in the context of HCMV infection proved not to be feasible due to lacking gp34- and gp68-specific antibodies for tracing, we aimed to establish a gain-of-function cell-based experimental model. To this end, we characterized the surface dynamics of vFcγR expressing transfected 293 T cells stably expressing a model antigen (hCD20) recognized by Rituximab (RTX). In order to manipulate internalization, we chose to replace the transmembrane and cytosolic domains of gp34 and gp68 with the according sequences from human CD4 (hCD4) as it contains a non-canonical di-leucine-based sorting motif (SQIKRLL) in its C-terminal cytoplasmic tail (CD4-tailed). To be recognized by AP-2 the serine residue upstream of the di-leucine motif in hCD4 needs to be phosphorylated in order to mimic the negatively charged glutamate or aspartate residues of a classical di-leucine motif (Pitcher et al., 1999). In a first experiment we ensured equal expression of the transfected constructs utilizing the polycistronic expression of GFP from a pIRES_eGFP expression vector to gate on transfected cells (Figure 3B). GFP-positive cells were also detected for Fcγ binding by a surface stain using a PE-TexasRed conjugated human Fcγ fragment (Figure 3C). This revealed that upon recombinant expression, surface exposition of Fcγ binding gp68 is drastically increased when CD4-tailed although overall protein expression was comparable and all constructs bear original signal peptides (Figure 3B). This can be attributed to the abrogation of its internalization evidenced by the fact that internalization of complete IC was drastically reduced with CD4-tailed gp68 when tracking hCD20/Rtx ICs using pulse-chase flow cytometry (Figure 3D). Conversely, while gp34 showed only mild differences in surface exposition when altered in the same way (Figure 3C), it proved to rapidly internalize complete Rtx-CD20 IC only with its native cytosolic tail intact (Figure 3D). Given that both molecules spontaneously internalize non-immune IgG (Figure 3—figure supplement 2), this could hint at different routes of trafficking following internalization. Specifically, as it has been shown that gp68 internalizes IgG-Fc translocating to the lysosomal compartment for degradation (Ndjamen et al., 2016), the difference seen for gp34 here suggests a recycling route as described for HSV-1 gE/gI (Ndjamen et al., 2014). Further, while we also observe gp34 to more efficiently internalize even non-immune IgG, the still rapid internalization of non-immune IgG by gp68 seems not to translate to the internalization of IC regarding efficiency (Figure 3—figure supplement 2), again in line with a previously published observation (Ndjamen et al., 2016). As we observed significantly more efficient internalization of IC by gp34wt compared to gp68wt or HSV-1 gEwt (Figure 3D, Figure 3—figure supplement 2), we conclude that a major task of gp34 compared to gp68 is mediating the efficient internalization of IC, a feature likely linked to its particular di-leucine cytosolic motif.

Figure 3 with 2 supplements see all

Download asset Open asset

Figure 3—source data 1

Relates to Figure 3D bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig3-data1-v2.xlsx

Download elife-63877-fig3-data1-v2.xlsx

FcγRIII binding to opsonized cells is reduced in the presence of gp68 but not gp34

As previously reported (Atalay et al., 2002; Corrales-Aguilar et al., 2014b) and confirmed in this study, we find that in the context of HCMV infection gp68 is more surface resident compared to gp34 (Figure 6A). Therefore, we next set out to test if the antagonizing effect of membrane-resident gp68 on FcγR activation can be attributed to a block of host FcγR binding to cell surface IC rather than IC internalization. To test this hypothesis, we established a flow cytometry based assay using vFcγR transfected 293 T-CD20 cells opsonized with Rtx and assayed for FcγR binding using soluble His-tagged ectodomains of human FcγRs, which are sequence identical between FcγRs IIIA and IIIB (Figure 4B). In this assay we compared gp34 and gp68 regarding their ability to interfere with FcγR binding to cell surface IC. Advantageously, CD4-tailed vFcγRs, besides circumventing confounding effects of internalization on our binding assay, closely mimic the relative surface density of gp34 and gp68 found on the plasma membrane of an HCMV-infected cell judged by human Fcγ binding (Figure 3C, Figure 5A; Corrales-Aguilar et al., 2014b). To ensure equal density of antigen upon recombinant vFcγR expression, CD20 levels were directly measured and found only in the case of CD99 expression to be slightly reduced, which served as a non-Fcγ-binding control molecule (Figure 4A). We found gp68 but not gp34 to significantly reduce binding of FcγRIIIA to cell surface immune complexes compared to CD99 (Figure 4C). This finding can be explained with certain CH2–CH3 region residues of IgG playing a subordinate, yet significant role in FcγRIII binding to IgG (Shields et al., 2001). Additionally, we found the effect of gp68 to be dependent on the accessibility of the CH2–CH3 interface region on IgG as pre-incubation with Protein G prevented the inhibition by gp68 and fully restored FcγRIIIA binding to Rtx (Figure 4C). These observations further narrow down the binding region of gp68 to involve the above-mentioned residues as opposed to Protein G which has been shown to not interact with these residues (Sauer-Eriksson et al., 1995). Finally, this conclusion is further substantiated by our observation that FcγRI, which binds to IgG independently of the above-mentioned residues in the CH2–CH3 interdomain region (Shields et al., 2001), is not blocked by gp68 or gp34 (Figure 4—figure supplement 1).

Figure 4 with 1 supplement see all

Download asset Open asset

Figure 4—source data 1

Relates to Figure 4C bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig4-data1-v2.xlsx

Download elife-63877-fig4-data1-v2.xlsx

Figure 5 with 2 supplements see all

Download asset Open asset

Figure 5—source data 1

Relates to Figure 5D bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig5-data1-v2.xlsx

Download elife-63877-fig5-data1-v2.xlsx

Cooperative antagonization of FcγR activation is achieved by combining gp68-mediated block of FcγR binding and subsequent gp34 internalization

While we delineated distinct mechanisms by which gp34 or gp68 are able to counteract FcγRII/III activation, we did not observe efficient antagonization of FcγR function by gp34 or gp68 individually in the context of HCMV deletion mutant infection (Figure 2). Therefore, we next wanted to test the antagonistic potential of gp68 or gp34 individually in the absence of non-immune IgG and in a gain-of-function setting. To this end, we conducted an FcγR activation assay with Hela cells co-transfected with Her2 as a model antigen and the indicated vFcγRs followed by incubation with titrated amounts of anti-Her2 IgG1 mAb (herceptin, Hc) (Figure 6A). Using this approach, gp68 or gp34 individually confirmed to antagonize FcγRIII activation. However, the more membrane resident gp68-CD4 showed a markedly stronger antagonistic effect compared to gp68wt, indicating a more prominent membrane-residence of gp68 to be beneficial regarding evasion from FcγR recognition despite its lower capacity to internalize ICs (Figure 3D). Conversely, when comparing unaltered gp34 to its less internalized gp34-CD4 variant we observed an opposite effect indicating internalization to be a major condition of gp34 driven antagonization of FcγR activation. Next, in an approach mimicking HCMV immune sera we tested if the addition of non-antigen-specific IgG impairs the antagonization of FcγR activation by gp34 or gp68. To this end, we added 10 µg/ml TNFα-specific Infliximab (Ifx) IgG1 given concomitantly with the reporter cells and graded amounts of Hc IgG1 (Figure 6B). This showed that indeed the presence of an excess of non-immune IgG interferes with both gp68 and gp34 antagonization in a dose-dependent manner. As ultrapure monomeric IgG does not activate the reporter cells on its own, shown by the samples that were treated only with Ifx but not Hc, this effect can be attributed to displacement of immune IgG from the vFcγRs resulting in restoration of FcγRIII activation. Synergism between gp34 and gp68 was optimal at a 100:1 ratio of non-immune Ifx over Hc (Figure 6B, right panel) indicating that vFcγRs are particularly adapted to work in the presence of excess non-immune IgG, supporting our previous observation with human hyperimmunoglobulin Cytotect (see Figure 2). On the other hand, host FcγRs have been shown to bind IgG immune complexes with a higher affinity compared to monomeric IgG (Bruhns et al., 2009), limiting the efficacy of monomeric IgG in attenuating FcγR activation. Finally, we addressed cooperative antagonization by gp34 and gp68 in a loss-of-function approach in the context of HCMV infection. When using the humanized anti-HCMV mAb MSL-109 targeting HCMV gH, we found that low gH levels expressed on the surface of HCMV-AD169-infected cell are insufficient to detect FcγR triggering (Figure 6—figure supplement 1). To overcome this experimental problem, we generated Her2-expressing BJ fibroblasts exhibiting high levels of Her2 on the cell surface. Infected Her2 BJ fibroblasts were opsonized with titrated amounts of Herceptin before cells were analyzed by FcγRIIIA reporter cells (Figure 6C). Although not statistically significant, this approach clearly demonstrated that compared to our observations with hyperimmunoglobulin Cytotect (see Figure 2), natively expressed gp34 as well as gp68 are able to antagonize FcγRIIIA activation individually. When both vFcγRs were expressed by HCMV-infected cells, a markedly stronger antagonistic effect was seen (p=0.0252).

Figure 6 with 1 supplement see all

Download asset Open asset

Figure 6—source data 1

Relates to Figure 6B bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig6-data1-v2.xlsx

Download elife-63877-fig6-data1-v2.xlsx

Figure 6—source data 2

Relates to Figure 6C bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig6-data2-v2.xlsx

Download elife-63877-fig6-data2-v2.xlsx

Figure 6—source data 3

Relates to Figure 6D bar graph.

https://cdn.elifesciences.org/articles/63877/elife-63877-fig6-data3-v2.xlsx

Download elife-63877-fig6-data3-v2.xlsx

Following the reveal of synergistic modes of action, we next explored cooperativity between gp34 and gp68 in a setup of reduced complexity. Specifically, in order to elaborate on the cooperativity of gp34 and gp68 regarding the here highlighted main mechanisms of internalization (gp34) and host FcγR-binding blockade (gp68) we chose to co-express gp34 and Her2 antigen while adding soluble gp68 (sgp68, Figure 1—figure supplement 1), or a soluble functional deficient gp34 point mutant as a control (sgp34mtrp, Figure 1—figure supplement 1 and Figure 1—figure supplement 2), together with the reporter cells after the removal of unbound Hc. Here we observe that the ectodomain of gp68 is sufficient to enhance the antagonistic effect of gp34 (Figure 6C).

In summary, we conclude that (i) gp34 is designed for internalization of IC while gp68 blocks FcγR binding to IC; (ii) gp34 and gp68 are able to antagonize FcγR activation individually when faced with titrated amounts of immune IgG, but non-immune IgG interferes with this inhibition; (iii) gp34 and gp68 show cooperativity in attenuating FcγR activation particularly under conditions of high excess of non-immune IgG.

Here we delineate for the first time synergistic modes of action involving two independent viral factors to efficiently achieve evasion from antibody-mediated immune cell surveillance (Figure 7). The novelty of these mechanisms acting cooperatively on the surface of an infected cell has major implications for our perception of how viruses have co-evolved multifactorial strategies to compete in a molecular arms race with the host at the junction between adaptive (humoral) and innate immunity. In particular, this knowledge refines our understanding of how HCMV can efficiently evade from a fully developed, broadly polyclonal and affinity-matured IgG response using its large coding capacity to ensure its reactivation, shedding, spread, and eventually transmission.

Figure 7

Download asset Open asset

Recognizing the conserved Fc part of IgG and exhibiting a very similar FcγR antagonization profile in reductionistic gain-of-function approaches (Corrales-Aguilar et al., 2014b), the two membrane-resident glycoproteins gp34 (RL11) and gp68 (UL119-118) appeared at first glance merely redundant. This left the puzzling question as to why HCMV (and cytomegaloviruses in general) have developed a whole arsenal of antagonists, unlike other viral and microbial pathogens with only singular Fc-binding glycoproteins (Berry et al., 2020; Corrales-Aguilar et al., 2014a; Falugi et al., 2013). However, when carefully examining their effect in a loss-of-function approach based on opsonized cells infected with a complete set of targeted HCMV gene deletion mutants, we convinced ourselves that gp34 and gp68 actually depend on one another to efficiently antagonize NK cell degranulation (Figure 2). This is explained by a number of individual features of gp34 and gp68 that only in conjunction can amount to antagonistic efficiency.

All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for all pertinent Figures (Figures 1D, 3D, 4C, 5D, 6B, 6C, 6D, Figure 4—figure supplement 1).

1. 1. Arapović J
   2. Lenac Rovis T
   3. Reddy AB
   4. Krmpotić A
   5. Jonjić S

   (2009) Promiscuity of MCMV immunoevasin of NKG2D: m138/fcr-1 down-modulates RAE-1epsilon in addition to MULT-1 and H60

   Molecular Immunology 47:114–122.

   https://doi.org/10.1016/j.molimm.2009.02.010

   • PubMed
   • Google Scholar
2. 1. Atalay R
   2. Zimmermann A
   3. Wagner M
   4. Borst E
   5. Benz C
   6. Messerle M
   7. Hengel H

   (2002) Identification and expression of human Cytomegalovirus transcription units coding for two distinct fcgamma receptor homologs

   Journal of Virology 76:8596–8608.

   https://doi.org/10.1128/JVI.76.17.8596-8608.2002

   • PubMed
   • Google Scholar
3. 1. Berry R
   2. Watson GM
   3. Jonjic S
   4. Degli-Esposti MA
   5. Rossjohn J

   (2020) Modulation of innate and adaptive immunity by cytomegaloviruses

   Nature Reviews Immunology 20:113–127.

   https://doi.org/10.1038/s41577-019-0225-5

   • PubMed
   • Google Scholar
4. 1. Bonifacino JS
   2. Traub LM

   (2003) Signals for sorting of transmembrane proteins to endosomes and lysosomes

   Annual Review of Biochemistry 72:395–447.

   https://doi.org/10.1146/annurev.biochem.72.121801.161800

   • PubMed
   • Google Scholar
5. 1. Bruhns P
   2. Iannascoli B
   3. England P
   4. Mancardi DA
   5. Fernandez N
   6. Jorieux S
   7. Daëron M

   (2009) Specificity and affinity of human fcgamma receptors and their polymorphic variants for human IgG subclasses

   Blood 113:3716–3725.

   https://doi.org/10.1182/blood-2008-09-179754

   • PubMed
   • Google Scholar
6. 1. Corrales-Aguilar E
   2. Trilling M
   3. Reinhard H
   4. Mercé-Maldonado E
   5. Widera M
   6. Schaal H
   7. Zimmermann A
   8. Mandelboim O
   9. Hengel H

   (2013) A novel assay for detecting virus-specific antibodies triggering activation of fcγ receptors

   Journal of Immunological Methods 387:21–35.

   https://doi.org/10.1016/j.jim.2012.09.006

   • PubMed
   • Google Scholar
7. 1. Corrales-Aguilar E
   2. Hoffmann K
   3. Hengel H

   (2014a) CMV-encoded fcγ receptors: modulators at the interface of innate and adaptive immunity

   Seminars in Immunopathology 36:627–640.

   https://doi.org/10.1007/s00281-014-0448-2

   • PubMed
   • Google Scholar
8. 1. Corrales-Aguilar E
   2. Trilling M
   3. Hunold K
   4. Fiedler M
   5. Le VT
   6. Reinhard H
   7. Ehrhardt K
   8. Mercé-Maldonado E
   9. Aliyev E
   10. Zimmermann A
   11. Johnson DC
   12. Hengel H

   (2014b) Human Cytomegalovirus fcγ binding proteins gp34 and gp68 antagonize fcγ receptors I, II and III

   PLOS Pathogens 10:e1004131.

   https://doi.org/10.1371/journal.ppat.1004131

   • PubMed
   • Google Scholar
9. 1. Cortese M
   2. Calò S
   3. D'Aurizio R
   4. Lilja A
   5. Pacchiani N
   6. Merola M

   (2012) Recombinant human Cytomegalovirus (HCMV) RL13 binds human immunoglobulin G fc

   PLOS ONE 7:e50166.

   https://doi.org/10.1371/journal.pone.0050166

   • PubMed
   • Google Scholar
10. 1. Dargan DJ
    2. Douglas E
    3. Cunningham C
    4. Jamieson F
    5. Stanton RJ
    6. Baluchova K
    7. McSharry BP
    8. Tomasec P
    9. Emery VC
    10. Percivalle E
    11. Sarasini A
    12. Gerna G
    13. Wilkinson GWG
    14. Davison AJ

    (2010) Sequential mutations associated with adaptation of human Cytomegalovirus to growth in cell culture

    Journal of General Virology 91:1535–1546.

    https://doi.org/10.1099/vir.0.018994-0

    • Google Scholar
11. 1. DiLillo DJ
    2. Tan GS
    3. Palese P
    4. Ravetch JV

    (2014) Broadly neutralizing hemagglutinin stalk-specific antibodies require fcγr interactions for protection against influenza virus in vivo

    Nature Medicine 20:143–151.

    https://doi.org/10.1038/nm.3443

    • PubMed
    • Google Scholar
12. 1. Dolan A
    2. Cunningham C
    3. Hector RD
    4. Hassan-Walker AF
    5. Lee L
    6. Addison C
    7. Dargan DJ
    8. McGeoch DJ
    9. Gatherer D
    10. Emery VC
    11. Griffiths PD
    12. Sinzger C
    13. McSharry BP
    14. Wilkinson GWG
    15. Davison AJ

    (2004) Genetic content of wild-type human Cytomegalovirus

    Journal of General Virology 85:1301–1312.

    https://doi.org/10.1099/vir.0.79888-0

    • Google Scholar
13. 1. Falk J
    2. Winkelmann M
    3. Laib Sampaio K
    4. Paal C
    5. Schrezenmeier H
    6. Alt M
    7. Stanton R
    8. Krawczyk A
    9. Lotfi R
    10. Sinzger C

    (2018) Large-Scale screening of HCMV-Seropositive blood donors indicates that HCMV effectively escapes from antibodies by Cell-Associated spread

    Viruses 10:500.

    https://doi.org/10.3390/v10090500

    • Google Scholar
14. 1. Falugi F
    2. Kim HK
    3. Missiakas DM
    4. Schneewind O

    (2013) Role of protein A in the evasion of host adaptive immune responses by Staphylococcus aureus

    mBio 4:e00575-13.

    https://doi.org/10.1128/mBio.00575-13

    • PubMed
    • Google Scholar
15. 1. Forthal D
    2. Hope TJ
    3. Alter G

    (2013) New paradigms for functional HIV-specific nonneutralizing antibodies

    Current Opinion in HIV and AIDS 8:393–401.

    https://doi.org/10.1097/COH.0b013e328363d486

    • PubMed
    • Google Scholar
16. 1. Halenius A
    2. Hauka S
    3. Dölken L
    4. Stindt J
    5. Reinhard H
    6. Wiek C
    7. Hanenberg H
    8. Koszinowski UH
    9. Momburg F
    10. Hengel H

    (2011) Human Cytomegalovirus disrupts the major histocompatibility complex class I peptide-loading complex and inhibits tapasin gene transcription

    Journal of Virology 85:3473–3485.

    https://doi.org/10.1128/JVI.01923-10

    • PubMed
    • Google Scholar
17. 1. Hengel H
    2. Brune W
    3. Koszinowski UH

    (1998) Immune evasion by cytomegalovirus--survival strategies of a highly adapted opportunist

    Trends in Microbiology 6:190–197.

    https://doi.org/10.1016/S0966-842X(98)01255-4

    • PubMed
    • Google Scholar
18. 1. Hessell AJ
    2. Hangartner L
    3. Hunter M
    4. Havenith CE
    5. Beurskens FJ
    6. Bakker JM
    7. Lanigan CM
    8. Landucci G
    9. Forthal DN
    10. Parren PW
    11. Marx PA
    12. Burton DR

    (2007) Fc receptor but not complement binding is important in antibody protection against HIV

    Nature 449:101–104.

    https://doi.org/10.1038/nature06106

    • PubMed
    • Google Scholar
19. 1. Horwitz JA
    2. Bar-On Y
    3. Lu CL
    4. Fera D
    5. Lockhart AAK
    6. Lorenzi JCC
    7. Nogueira L
    8. Golijanin J
    9. Scheid JF
    10. Seaman MS
    11. Gazumyan A
    12. Zolla-Pazner S
    13. Nussenzweig MC

    (2017) Non-neutralizing antibodies alter the course of HIV-1 infection in Vivo

    Cell 170:637–648.

    https://doi.org/10.1016/j.cell.2017.06.048

    • PubMed
    • Google Scholar
20. 1. Kagan KO
    2. Enders M
    3. Schampera MS
    4. Baeumel E
    5. Hoopmann M
    6. Geipel A
    7. Berg C
    8. Goelz R
    9. De Catte L
    10. Wallwiener D
    11. Brucker S
    12. Adler SP
    13. Jahn G
    14. Hamprecht K

    (2019) Prevention of maternal-fetal transmission of cytomegalovirus after primary maternal infection in the first trimester by biweekly hyperimmunoglobulin administration

    Ultrasound in Obstetrics & Gynecology 53:383–389.

    https://doi.org/10.1002/uog.19164

    • PubMed
    • Google Scholar
21. 1. Kolb P
    2. Sijmons S
    3. McArdle MR
    4. Taher H
    5. Womack J
    6. Hughes C
    7. Ventura A
    8. Jarvis MA
    9. Stahl-Hennig C
    10. Hansen S
    11. Picker LJ
    12. Malouli D
    13. Hengel H
    14. Früh K

    (2019) Identification and functional characterization of a novel fc Gamma-Binding glycoprotein in rhesus Cytomegalovirus

    Journal of Virology 93:e02077-18.

    https://doi.org/10.1128/JVI.02077-18

    • PubMed
    • Google Scholar
22. 1. Le VT
    2. Trilling M
    3. Hengel H

    (2011) The cytomegaloviral protein pUL138 acts as potentiator of tumor necrosis factor (TNF) Receptor 1 surface density to enhance ULb'-Encoded modulation of TNF- Signaling

    Journal of Virology 85:13260–13270.

    https://doi.org/10.1128/JVI.06005-11

    • PubMed
    • Google Scholar
23. 1. Le-Trilling VTK
    2. Becker T
    3. Nachshon A
    4. Stern-Ginossar N
    5. Schöler L
    6. Voigt S
    7. Hengel H
    8. Trilling M

    (2020) The human Cytomegalovirus pUL145 isoforms act as viral DDB1-Cullin-Associated factors to instruct host protein degradation to impede innate immunity

    Cell Reports 30:2248–2260.

    https://doi.org/10.1016/j.celrep.2020.01.070

    • PubMed
    • Google Scholar
24. 1. Lenac T
    2. Budt M
    3. Arapovic J
    4. Hasan M
    5. Zimmermann A
    6. Simic H
    7. Krmpotic A
    8. Messerle M
    9. Ruzsics Z
    10. Koszinowski UH
    11. Hengel H
    12. Jonjic S

    (2006) The herpesviral fc receptor fcr-1 down-regulates the NKG2D ligands MULT-1 and H60

    Journal of Experimental Medicine 203:1843–1850.

    https://doi.org/10.1084/jem.20060514

    • Google Scholar
25. 1. Li P
    2. Jiang N
    3. Nagarajan S
    4. Wohlhueter R
    5. Selvaraj P
    6. Zhu C

    (2007) Affinity and kinetic analysis of fcγ receptor IIIa (CD16a) Binding to IgG ligands

    Journal of Biological Chemistry 282:6210–6221.

    https://doi.org/10.1074/jbc.M609064200

    • Google Scholar
26. 1. Morizono K
    2. Ku A
    3. Xie Y
    4. Harui A
    5. Kung SK
    6. Roth MD
    7. Lee B
    8. Chen IS

    (2010) Redirecting lentiviral vectors pseudotyped with sindbis virus-derived envelope proteins to DC-SIGN by modification of N-linked glycans of envelope proteins

    Journal of Virology 84:6923–6934.

    https://doi.org/10.1128/JVI.00435-10

    • PubMed
    • Google Scholar
27. 1. Ndjamen B
    2. Farley AH
    3. Lee T
    4. Fraser SE
    5. Bjorkman PJ

    (2014) The herpes virus fc receptor gE-gI mediates antibody bipolar bridging to clear viral antigens from the cell surface

    PLOS Pathogens 10:e1003961.

    https://doi.org/10.1371/journal.ppat.1003961

    • Google Scholar
28. 1. Ndjamen B
    2. Joshi DS
    3. Fraser SE
    4. Bjorkman PJ

    (2016) Characterization of antibody bipolar bridging mediated by the human Cytomegalovirus fc receptor gp68

    Journal of Virology 90:3262–3267.

    https://doi.org/10.1128/JVI.02855-15

    • PubMed
    • Google Scholar
29. 1. Pitcher C
    2. Höning S
    3. Fingerhut A
    4. Bowers K
    5. Marsh M

    (1999) Cluster of differentiation antigen 4 (CD4) endocytosis and adaptor complex binding require activation of the CD4 endocytosis signal by serine phosphorylation

    Molecular Biology of the Cell 10:677–691.

    https://doi.org/10.1091/mbc.10.3.677

    • PubMed
    • Google Scholar
30. 1. Reinhard HC

    (2010) Struktur-Funktions-Beziehung der HCMV-Kodierten Fcgamma-Rezeptoren Gp34 und Gp68

    Humboldt-Universität Zu Berlin, Mathematisch-Naturwissenschaftliche Fakultät I 5:16124.

    https://doi.org/10.18452/16124

    • Google Scholar
31. 1. Revello MG
    2. Lazzarotto T
    3. Guerra B
    4. Spinillo A
    5. Ferrazzi E
    6. Kustermann A
    7. Guaschino S
    8. Vergani P
    9. Todros T
    10. Frusca T
    11. Arossa A
    12. Furione M
    13. Rognoni V
    14. Rizzo N
    15. Gabrielli L
    16. Klersy C
    17. Gerna G

    (2014) A randomized trial of hyperimmune globulin to prevent congenital Cytomegalovirus

    New England Journal of Medicine 370:1316–1326.

    https://doi.org/10.1056/NEJMoa1310214

    • Google Scholar
32. 1. Sauer-Eriksson AE
    2. Kleywegt GJ
    3. Uhlén M
    4. Jones TA

    (1995) Crystal structure of the C2 fragment of streptococcal protein G in complex with the fc domain of human IgG

    Structure 3:265–278.

    https://doi.org/10.1016/S0969-2126(01)00157-5

    • PubMed
    • Google Scholar
33. 1. Shields RL
    2. Namenuk AK
    3. Hong K
    4. Meng YG
    5. Rae J
    6. Briggs J
    7. Xie D
    8. Lai J
    9. Stadlen A
    10. Li B
    11. Fox JA
    12. Presta LG

    (2001) High resolution mapping of the binding site on human IgG1 for FcγrI, FcγrII, FcγrIII, and FcRn and design of IgG1 variants with improved binding to the fcγr

    Journal of Biological Chemistry 276:6591–6604.

    https://doi.org/10.1074/jbc.M009483200

    • Google Scholar
34. 1. Sprague ER
    2. Martin WL
    3. Bjorkman PJ

    (2004) pH dependence and stoichiometry of binding to the fc region of IgG by the herpes simplex virus fc receptor gE-gI

    Journal of Biological Chemistry 279:14184–14193.

    https://doi.org/10.1074/jbc.M313281200

    • Google Scholar
35. 1. Sprague ER
    2. Reinhard H
    3. Cheung EJ
    4. Farley AH
    5. Trujillo RD
    6. Hengel H
    7. Bjorkman PJ

    (2008) The human Cytomegalovirus fc receptor gp68 binds the fc CH2-CH3 interface of immunoglobulin G

    Journal of Virology 82:3490–3499.

    https://doi.org/10.1128/JVI.01476-07

    • PubMed
    • Google Scholar
36. 1. Staib C
    2. Drexler I
    3. Sutter G

    (2004) Construction and isolation of recombinant MVA

    Methods in Molecular Biology 269:77–100.

    https://doi.org/10.1385/1-59259-789-0:077

    • PubMed
    • Google Scholar
37. 1. Stanton RJ
    2. Baluchova K
    3. Dargan DJ
    4. Cunningham C
    5. Sheehy O
    6. Seirafian S
    7. McSharry BP
    8. Neale ML
    9. Davies JA
    10. Tomasec P
    11. Davison AJ
    12. Wilkinson GWG

    (2010) Reconstruction of the complete human Cytomegalovirus genome in a BAC reveals RL13 to be a potent inhibitor of replication

    Journal of Clinical Investigation 120:3191–3208.

    https://doi.org/10.1172/JCI42955

    • Google Scholar
38. 1. Stern-Ginossar N
    2. Weisburd B
    3. Michalski A
    4. Le VT
    5. Hein MY
    6. Huang SX
    7. Ma M
    8. Shen B
    9. Qian SB
    10. Hengel H
    11. Mann M
    12. Ingolia NT
    13. Weissman JS

    (2012) Decoding human cytomegalovirus

    Science 338:1088–1093.

    https://doi.org/10.1126/science.1227919

    • PubMed
    • Google Scholar
39. 1. Thäle R
    2. Lucin P
    3. Schneider K
    4. Eggers M
    5. Koszinowski UH

    (1994) Identification and expression of a murine Cytomegalovirus early gene coding for an fc receptor

    Journal of Virology 68:7757–7765.

    https://doi.org/10.1128/JVI.68.12.7757-7765.1994

    • PubMed
    • Google Scholar
40. 1. Tischer BK
    2. von Einem J
    3. Kaufer B
    4. Osterrieder N

    (2006) Two-step red-mediated recombination for versatile high-efficiency markerless DNA manipulation in Escherichia coli

    BioTechniques 40:191–197.

    https://doi.org/10.2144/000112096

    • PubMed
    • Google Scholar
41. 1. Van den Hoecke S
    2. Ehrhardt K
    3. Kolpe A
    4. El Bakkouri K
    5. Deng L
    6. Grootaert H
    7. Schoonooghe S
    8. Smet A
    9. Bentahir M
    10. Roose K
    11. Schotsaert M
    12. Schepens B
    13. Callewaert N
    14. Nimmerjahn F
    15. Staeheli P
    16. Hengel H
    17. Saelens X

    (2017) Hierarchical and redundant roles of activating fcγrs in protection against influenza disease by M2e-Specific IgG1 and IgG2a antibodies

    Journal of Virology 91:e02500-16.

    https://doi.org/10.1128/JVI.02500-16

    • PubMed
    • Google Scholar
42. 1. Wagner M
    2. Ruzsics Z
    3. Koszinowski UH

    (2002) Herpesvirus genetics has come of age

    Trends in Microbiology 10:318–324.

    https://doi.org/10.1016/S0966-842X(02)02394-6

    • PubMed
    • Google Scholar
43. 1. Wines BD
    2. Powell MS
    3. Parren PW
    4. Barnes N
    5. Hogarth PM

    (2000) The IgG fc contains distinct fc receptor (FcR) binding sites: the leukocyte receptors fc gamma RI and fc gamma RIIa bind to a region in the fc distinct from that recognized by neonatal FcR and protein A

    The Journal of Immunology 164:5313–5318.

    https://doi.org/10.4049/jimmunol.164.10.5313

    • PubMed
    • Google Scholar
44. 1. Zuhair M
    2. Smit GSA
    3. Wallis G
    4. Jabbar F
    5. Smith C
    6. Devleesschauwer B
    7. Griffiths P

    (2019) Estimation of the worldwide seroprevalence of Cytomegalovirus: a systematic review and meta-analysis

    Reviews in Medical Virology 29:e2034.

    https://doi.org/10.1002/rmv.2034

    • PubMed
    • Google Scholar

Author details

1. Philipp Kolb

   1. Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany
   2. Institute of Virology, University Medical Center, Albert-Ludwigs-University Freiburg, Freiburg, Germany

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Writing - original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7935-217X

2. Katja Hoffmann

   1. Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany
   2. Institute of Virology, University Medical Center, Albert-Ludwigs-University Freiburg, Freiburg, Germany

   Contribution

   Data curation, Formal analysis, Investigation, Writing - review and editing

   Competing interests

   No competing interests declared

3. Annika Sievert

   1. Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany
   2. Institute of Virology, University Medical Center, Albert-Ludwigs-University Freiburg, Freiburg, Germany

   Contribution

   Data curation, Investigation

   Competing interests

   No competing interests declared

4. Henrike Reinhard

   Institute of Virology, University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany

   Contribution

   Data curation, Investigation

   Competing interests

   No competing interests declared

5. Eva Merce-Maldonado

   Institute of Virology, University Hospital Düsseldorf, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany

   Contribution

   Data curation, Investigation

   Competing interests

   No competing interests declared

6. Vu Thuy Khanh Le-Trilling

   Institute for Virology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany

   Contribution

   Resources, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

7. Anne Halenius

   1. Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany
   2. Institute of Virology, University Medical Center, Albert-Ludwigs-University Freiburg, Freiburg, Germany

   Contribution

   Resources

   Competing interests

   No competing interests declared

8. Dominique Gütle

   1. Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany
   2. Institute of Virology, University Medical Center, Albert-Ludwigs-University Freiburg, Freiburg, Germany

   Contribution

   Data curation, Formal analysis

   Competing interests

   No competing interests declared

9. Hartmut Hengel

   1. Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany
   2. Institute of Virology, University Medical Center, Albert-Ludwigs-University Freiburg, Freiburg, Germany

   Contribution

   Conceptualization, Supervision, Funding acquisition, Writing - original draft

   For correspondence

   hartmut.hengel@uniklinik-freiburg.de

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3482-816X

• 1,978

  views

• 241

  downloads

• 14

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.