Modulation of NF κB activation by primate lentiviruses

The cellular transcription factor NF-κB regulates the expression of genes involved in inflammation and immunity, including interferon-stimulated genes that protect against viral pathogens. Yet, NF-κB is not only a key mediator of antiviral immune responses but also exploited by many viruses for transcription of viral genes. Replication of HIV, for example, depends on binding of NF-κB p65/p50 heterodimers to the viral LTR promoter.

We have previously shown that primate lentiviruses tightly regulate the activation of this transcription factor throughout their replication cycle. Whereas the early protein Nef of most primate lentiviruses boosts the activation of NF-κB to initiate LTR-dependent viral gene transcription, Vpu inhibits the activation of NF‑κB during later stages to reduce expression of interferon (IFN) and IFN-stimulated genes (Sauter et al., 2015; Heusinger and Kirchhoff, 2017).

Viruses that do not encode a vpu gene may employ alternative strategies to prevent immune activation after NF-κB-dependent initiation of viral gene expression. For example, the Nef proteins of most viruses lacking a vpu gene efficiently down-modulate the T cell receptor CD3 from the cell surface. This Nef function was lost in most vpu-expressing viruses suggesting that the acquisition of Vpu-mediated NF‑κB inhibition may have reduced the selection pressure for suppression of T cell activation by Nef. We have recently identified two viruses, SIVcol and SIVolc, which have evolved yet another strategy, and use their Vpr protein to suppress NF-κB-mediated immune activation (Hotter et al., PLOS Pathogens, 2017). Interestingly, these viruses lack a vpu gene and the CD3 down-modulation function of Nef. Thus, primate lentiviruses have evolved at least three independent mechanisms to inhibit the expression of cellular NF-κB target genes.

We are currently performing transcriptome analyses of infected primary cells to investigate the global impact of Vpu on cellular gene expression. Since chronic immune activation is a hallmark of AIDS, our findings will help to better understand the pathophysiology of HIV. Notably, NF-κB also plays also a key role in the reactivation of HIV from latency. Thus, understanding the molecular basis of NF-κB modulation by HIV and related lentiviruses may also lead to novel approaches to reactivate the virus for “kick and kill” cure strategies.

Profilbild von Jun.Prof. Daniel Sauter

Jun.Prof. Daniel Sauter

Group leader

Coevolution of tetherin with enveloped viruses

Tetherin is a cellular restriction factor that inhibits the release of a large variety of human and animal viruses by directly trapping them at the membranes of infected cells (Neil et al., 2008; Van Damme et al., 2008). Until recently, only mammalian orthologs of tetherin had been described and the deep evolutionary origins of this antiviral protein had remained obscure.

The antiviral activity of tetherin depends on an unusual topology comprising an N-terminal transmembrane domain, followed by an extracellular coiled-coil region and a C-terminal glycosylphosphatidylinositol (GPI) anchor. One of the two membrane anchors is inserted into assembling virions, while the other remains in the plasma membrane of the infected cell. Thus, tetherin entraps budding viruses by physically bridging viral and cellular membranes. In addition to its ability to restrict virus release, tetherin also acts as an innate sensor inducing NF‑κB-dependent expression of antiviral genes (Galão et al., 2012).

To better understand the coevolution of tetherin with primate lentiviruses and other enveloped viruses, we are currently investigating the evolutionary history of this fascinating protein. Characterizing tetherin orthologs from diverse vertebrate species, we have shown that this antiviral protein emerged more than 450 million years ago (Heusinger et al., 2015). Thus, not only mammals, but also reptiles, birds and even fish express this restriction factor. Since efficient tetherin antagonism has been suggested to be a prerequisite for successful spread of HIV-1 in the human population (Sauter et al., 2009; Sauter et al., 2010), we are currently defining the role of tetherin in the cross-species transmission of influenza A viruses, HIV-2, and other zoonotic viral pathogens.

Adaptation of HIV-1 groups M, N, O and P to human tetherin

One major research focus of our lab is the coevolution of primate lentiviruses with their respective host species. We are especially interested in the zoonotic transmissions of simian immunodeficiency viruses infecting chimpanzees (SIVcpz) and gorillas (SIVgor) to humans and their subsequent adaptations that contributed to the spread of HIV in the human population. Simian immunodeficiency viruses have been transmitted at least four times independently to humans giving rise to HIV-1 groups M, N, O and P. Interestingly, only HIV-1 group M strains spread world-wide and are almost entirely responsible for the current AIDS pandemic. In contrast, HIV-1 O is endemic with around 100,000 infected individuals in Western Central Africa, and groups N and P are very rare with only 17 and 2 known cases, respectively. In 2009, we identified the antiviral protein tetherin as a key player in the evolution and spread of HIV-1 (Sauter et al., 2009). Most simian immunodeficiency viruses – including the direct precursors of HIV-1 – use their accessory protein Nef to counteract tetherin in their respective host species. Human tetherin, however, is resistant against SIV Nef due to a protective deletion in its cytoplasmic tail and thus poses a significant barrier for successful zoonotic transmission of SIV to humans (Sauter et al., 2010).

Interestingly, the four groups of HIV-1 (M, N, O and P) have evolved different mechanisms to overcome this hurdle: Whereas pandemic HIV‑1 group M viruses switched from Nef to Vpu to counteract human tetherin, rare group P and N Vpus do not or only poorly antagonize this restriciton factor (Sauter et al., 2009; 2011). Furthermore, we could show that Nef proteins of epidemic HIV‑1 group O viruses evolved the ability to target a region adjacent to the deletion in human tetherin to increase virion release from infected CD4+ T cells (Kluge et al., 2015; Mack et al., 2017). Thus, our data strongly suggest that tetherin counteraction is a prerequisite for the efficient spread of lentiviruses in the human population. This hypothesis is further supported by two collaborative studies with the labs of Kei Sato (Kyoto) and Beatrice Hahn (Philadelphia), in which we analyzed the replicative fitness of HIV and SIV mutants that selectively fail to antagonize tetherin (Kmiec et al., 2016; Yamada et al., 2018). Experiments in primary CD4+ T cells and humanized mice demonstrated that human tetherin-specific adaptations confer a significant selection advantage to the virus as they increase IFN resistance and replication capacity during acute infection.

In summary, our results provide a first plausible explanation why only viruses of HIV‑1 groups M and O spread significantly in the human population, whereas those of groups N and P remain largely restricted to Western Central Africa.

Role of fusion proteins in HIV replication

Pandemic strains of HIV-1 (group M) encode a total of nine structural (gag, pol, env), regulatory (rev, tat) and accessory (vif, vpr, vpu, nef) genes. In addition to these canonical proteins, several studies have reported the existence of fusion proteins, most of which are the result of alternative splicing, when exons of regular and/or alternative open reading frames (ORFs) are brought together. Although some of these proteins are expressed at high levels, their role in the viral replication cycle remained poorly understood (Langer and Sauter, 2017).

Notably, alternative splicing is not the only mechanism that can generate unusual fusion proteins in HIV-1. In 2010, Kraus and colleagues reported an HIV-1 gene arrangement in which rev1 and vpu genes were present in the same reading frame without an intervening stop codon (Kraus et al., 2010). Analysis of the deduced protein sequence of this gene fusion suggests that it spans the plasma membrane like Vpu, but may contain an additional extracellular Rev-derived N-terminal domain.

Although this rev1-vpu gene fusion is present in a considerable fraction of HIV-1 strains, its functional significance had remained unclear. Using infectious molecular clones differing only in their ability to express this fusion protein, we could show that Rev1-Vpu does neither affect known Vpu and Rev functions nor enhance viral replication. Instead, the rev1-vpu fusion gene seems to have a neutral phenotype. We therefore hypothesize that this unusual polymorphism may be the epiphenomenon of other adaptive changes, such as mutations optimizing Env expression (Langer et al., 2015).

Antiviral activity of guanylate binding proteins

In a genome-wide screening for previously uncharacterized restriction factors, we have identified the IFN-inducible guanylate binding protein 5 (GBP5) as a cellular protein reducing the particle infectivity of HIV-1 and other retroviruses (Krapp and Hotter et al., 2016). Experiments in transfected HEK293T cells and primary macrophages revealed that GBP5 interferes with the glycosylation, proteolytic processing and virion incorporation of the viral envelope (Env) protein. Intriguingly, mutations in vpu reduced the sensitivity to GBP5-mediated restriction, since loss of Vpu increased total cellular levels of Env, which is expressed from the same bi-cistronic mRNA. In follow-up experiments, we could show that GBP2, another member of the same family of large GTPases exerts similar antiviral activity. While the GTPase activity is dispensable for the antiviral effect, efficient restriction by GBP2 and GBP5 requires the presence of a C-terminal isoprenylation site. Mutational analyses and interaction assays revealed that GBPs do not target the viral envelope protein, but interfere with the activity of a cellular factor involved for Env maturation. As the cellular target is required for full infectivity of diverse virus families, we are currently analyzing the breadth of GBP2- and GBP5-mediated restriction and further elucidating the underlying mechanisms.