Agenda

Recent advances in cryo electron microscopy (cryoEM) and computer processing have enabled us to obtain atomic models for the capsids and capsid-associated tegument complexes (CATC) of herpes simplex virus type 1 (HSV-1), human cytomegalovirus (HCMV) and Kaposi’s sarcoma-associated herpesvirus (KSHV), representative members of the α, β, γ-herpesvirus subfamilies of the Herpesviridae, respectively. For all these capsids, multiple levels of inter major capsid protein (MCP) interactions—six sets of stacked hairpins lining the hexon channel, disulfide bonds across channel and buttress domains in neighboring MCPs, and an interaction network forged by the N-lasso domain and secured by the dimerization domain—define a robust capsid that is amenable to the tremendous pressure exerted by the enclosed genome. The triplexes, each composed of two “embracing” Tri2 molecules and a “third-wheeling” Tri1 molecule, anchor to the capsid floor via Tri1 N-anchor to plug holes of the MCP network and rivet the capsid floor. Crowning every vertex of HSV capsid are five copies of hetero-pentameric CATC, each containing a camel-shaped pUL17 supporting the coiled-coil helix bundle of a pUL25 dimer and a pUL36 dimer, thus positioning their flexible head domains for axonic capsid transport. By contrast, HCMV, which has the largest genome, uses the β-herpesvirus specific tegument protein pp150 to form a helix bundle that secures the capsid through cysteine tetrad interaction with SCP, the smallest and least conserved capsid protein. Notwithstanding newly discovered fold conservation between triplex proteins and gpD of bacteriophage l and the previously recognized bacteriophage HK97-like fold in MCP, herpesvirus capsid proteins fashion extraordinary structural diversities in the forms of domain insertions and conformation polymorphism, not only for tegument interactions but also for DNA encapsulation. The significances of these molecular interactions are established by structure-guided mutagenesis and design of peptides mimicking and disrupting SCP-MCP interactions leading to inhibition of KSHV lytic replication.

Herpesviruses share a unique maturation route that entails the assembly of new capsids in the nucleus and their egress across the two nuclear envelopes by first budding into the inner nuclear membrane and then fusing with the outer nuclear membrane. The resulting unenvelopped viral particles subsequently acquire their final envelope from an intracellular compartment. While debated for a long time, this commonly accepted model lacks many molecular details. Our lab has tried to elucidate this egress route, focusing on HSV-1, by characterizing the sequence of acquisition of the highly complex viral tegument layer, deciphering the complete protein content of viral intermediates, identifying the site of final envelopment and probing the machinery required to ultimately reach the cell surface. In the process, we also probed host-pathogen interactions and identified some cellular constituents that modulate this intracellular HSV-1 journey. It also yielded innovative approaches to analyze HSV-1 such as an in vitro reconstitution of nuclear egress and analysis & isolation of single viral particles by flow virometry. In this presentation, our past and present findings aimed at molecularly clarifying the HSV-1 nuclear egress route will be presented.

The talk will give an overview of the research work of M.M.’s group and a number of other groups on the molecular mechanisms regulating HCMV nuclear egress.

It will pick up current hotspots of regulation playing a role in the nuclear egress of HCMV and additional herpesviruses, i.e. the nature of the nuclear egress complex (NEC), 3D structural properties of the core NEC pUL50-pUL53, the nuclear import and rim recruitment of NEC proteins. Specific focus will be given to the site-specific lamin A/C phosphorylation through viral pUL97, the induction of lamina-depleted areas, the role of lamin cis-trans isomerization and a proteomics-based investigation of the multicomponent NEC. Additionally, the questions of conservation of NEC functions between herpesviruses, putative ways of NEC-capsid docking and the "swiss army knife" of NEC-host interaction will be addressed. A discussion of the potential of NEC-based strategies of antiviral drug design will also be included. In essence, the presentation will stress the importance of higher-order viral-cellular complexes, particularly the regulatory power of the multicomponent NEC.

Using whole genome sequencing we find that Human Cytomegalovirus (HCMV) and Herpes simplex virus (HSV) to be  freely recombining with only local areas of linkage disequilibrium while Epstein Barr virus (EBV)and Varicella zoster virus (VZV) show apparent local and  long-range co-inheritance patterns. These differences provide insights into the different factors shaping human herpesvirus evolutionary history.    

We have also used genomic approaches to model the natural history in its human host, of VZV, a human-restricted alphaherpesvirus.  We demonstrated that all strains within the live attenuated vOka vaccine can establish latency and reactivate from human ganglia and that latency occurs as little as 2 days post inoculation.  Our data predict mathematically that no viral replication occurs during latency.  Using targeted RNAseq enrichment for VZV and HSV-1, we have discovered that VZV latent in trigeminal ganglia expresses a unique antisense VZV associated transcript the function of which may be to downregulate VZV transcriptional activation.  HSV-1 latent in the same ganglia expresses only the LATs and associated microRNAs.     Using mutation rates calculated from sequentially collected samples, we estimate that current VZV clade divergence occurred between 20,000 and 50,000 years ago, later than the human migrations from Africa. 

Applying genomic modelling to intahost HCMV infections, we have found that HCMV substitution rates are similar to those of other human herpesviruses and lower than those of persistent RNA viruses and that the high nucleotide diversity that characterises some samples is due to mixed infection with distinct viral haplotypes (strains).  By resolving individual haplotypes we have been able to identify the timing of superinfections in multiply infected hosts and to observe changes consistent with recombination between  independent viral haplotypes with selection for the recombinant strain.  We are now exploring how  HCMV variation may influence compartmentalisation within the host and in maternal fetal transmission

Herpes simplex virus 2 (HSV2) is a human herpesvirus found worldwide that causes genital lesions and more rarely causes encephalitis. This pathogen is most common in Africa, and particularly in central and east Africa, an area of particular significance for the evolution of modern humans. Unlike HSV1, HSV2 has not simply co-speciated with humans from their last common ancestor with primates. HSV2 jumped the species barrier between 1.4 and 3 MYA, most likely through intermediate but unknown hominin species. I will present data from a probability-based network analysis, used to determine the most probable transmission path between intermediate hosts of HSV2 from the ancestors of chimpanzees to the ancestors of modern humans, using paleo-environmental data on the distribution of African tropical rainforest over the last 3 million years and data on the age and distribution of fossil species of hominin present in Africa between 1.4 and 3 MYA. This model identifies Paranthropus boisei as the most likely intermediate host of HSV2, while Homo habilis may also have played a role in the initial transmission of HSV2 from the ancestors of chimpanzees to P. boisei.

Human herpesvirus 6A (HHV-6A) and HHV-6B have long double-stranded DNA genomes (~160kb) with terminal direct repeats (DRs). The DRs include arrays of (TTAGGG)n, which is also the vertebrate telomere sequence. Therefore HHV-6A and HHV-6B, alongside other herpesviruses that infect a wide variety of other vertebrates, have the capacity to integrated into telomeres, the essential capping structures of chromosomes. Infections caused by HHV-6 (particularly HHV-6B) are common in early childhood and in most people the viral genome persists in a latent state lifelong, Occasional ‘opportunistic’ reactivation can be associated with severe consequences in immune-compromised patients. It has been proposed that telomeric integration is a form of HHV-6 latency but telomeres are dynamic structures that play important roles in human ageing and cancer. Surprising about 1% of people worldwide are carriers of an inherited telomere-integrated copy of HHV-6, known as chromosomally integrated HHV-6 (ciHHV-6). Recently, we have used sequence analysis to show that many ciHHV-6 carriers appear to have inherited the viral genome from a small number of ancient common ancestors. Nevertheless most of the sequenced ciHHV-6 genomes are intact and therefore appear capable of reactivation. If telomeric integration is an effect form of latency for HHV-6 there must be an efficient mechanism for release or excision of the viral genome from the telomere. We have proposed that the excision mechanism is dependent on the normal processing of telomere-loop structures (t-loops) that are involved in telomere capping. We have developed methods to measure the partial or complete excision of ciHHV-6 genomes from telomeres and we are using these to characterise the dynamic relationship between integrated HHV-6 and the telomere.

Maternal antibodies play a major role in protection of the fetus and neonate from infection. In addition to Zika, other vertically transmitted TORCH infections, (toxoplasma, other, rubella, cytomegalovirus, and herpes simplex virus), lead to severe neurological sequelae with considerable morbidity and mortality in this setting. In a mouse model of neonatal herpes simplex virus (HSV) infection, we have demonstrated the ability of passively transferred polyclonal maternal antibody to localize to the nervous system and to protect from neonatal disease and death. Translating these observations to humans, we have also established that HSV-specific antibody can be detected in fetal trigeminal ganglia. Using a live-attenuated vaccine or through passive administration of a monoclonal antibody to female mice, we have shown that their offspring are completely resistant to HSV challenge immediately postpartum. Once these challenged neonates become adults they do not display the neurological sequelae of neonatal HSV infection as measured by behavioral studies. Collectively, these results point to a previously unappreciated role of passively-acquired polyclonal antibodies in protecting fetal and newborn nervous systems -- typically considered immune privileged -- against neurotropic infections.  They also suggest a prophylactic approach whereby vaccination, or passive administration of monoclonal antibodies to HSV-seronegative mothers before parturition could prevent or mitigate the sequelae of neonatal HSV infection. The severity of vertically acquired neurological infections, in the context of a well-defined period of temporal risk, makes this a particularly appealing approach.

Efforts to therapeutically target herpesvirus latency have been largely unsuccessful due to the limited viral gene expression.  Our approach has been to identify and target pathologic changes to the host cell during latent infection.  We found that during latent infection of human endothelial cells, KSHV dramatically alters the metabolism of the cell, including inducing glycolysis, glutaminolysis and fatty acid synthesis.  Inhibition of these metabolic pathways during KSHV latency leads to increased cell death in the infected cells but not their uninfected counterparts.  We recently broadened our approach to identify global requirements for the survival of latently infected cells and have identified a number of targets that in theory could be used to eliminate latently infected cells.

Epstein-Barr virus (EBV) transforms primary B-cells into continuously proliferating lymphoblastoid cell lines (LCL). Knowledge of how EBV latency oncoproteins remodel host metabolic pathways to support rapid B-cell outgrowth and to overcome redox stress remains incomplete. To gain insights into EBV-induced metabolic dependency factors, we constructed a proteomic map of EBV B-cell transformation. Primary B-cells from 12 human donors were profiled by tandem-mass-tag mass spectrometry at rest and at 9 timepoints after infection. This approach generated expression profiles of >6500 host and 17 viral proteins and highlighted EBV-induced pathways. Glycolysis, cholesterol and fatty acid biosynthesis pathways were highly upregulated cytosolic metabolic pathways, each if of which were found to be critical for EBV-driven B-cell outgrowth. Mitochondria were also highly remodeled following EBV infection, with the one-carbon (1C) metabolism pathway among the most highly EBV-induced. Mitochondrial 1C uses folate carriers to interconvert serine into glycine, formate and NADPH building blocks. EBNA2 and cMyc were critical for upregulation of key mitochondrial 1C enzymes including MTHFD2, which is highly expressed during embryogenesis but not in most adult tissues. Chemical and CRISPR genetic analysis underscored MTHFD2 and 1C pathway roles in EBV-driven B-cell growth and survival. MTHFD2 was important for intramitochondrial NADPH generation, and compartment-specific perturbation of mitochondrial NADPH levels diminished LCL growth and survival. Isotope tracing studies further supported EBV-induced 1C pathway roles in purine nucleotide synthesis and NADPH production. To fuel 1C metabolism, EBV upregulated serine import and de novo synthesis, each of which were found to be important dependency factors for B-cell outgrowth. 1C-derived glycine also supported glutathione synthesis, which together with NADPH exerted key redox defense roles.  These results highlight mitochondrial 1C as a key EBV-induced metabolic dependency factor and potential therapeutic target.

During lytic replication of Kaposi’s sarcoma-associated herpesvirus (KSHV) and other gammaherpesviruses, the gene expression landscape of a cell is remodeled to evade the immune response and create an environment favorable to viral replication. A major driver of these gene expression changes is a virally encoded, messenger RNA (mRNA)-specific endonuclease termed SOX in KSHV. The majority of mRNAs are susceptible to SOX cleavage, but they are cleaved with varying efficiency, and the mechanistic basis underlying this selectivity has remained largely unknown.   Using a reconstituted cleavage system, we reveal how specific RNA sequence and structure proximal to the cleavage site direct SOX binding and cleavage efficiency. This provides key insight into sequence specificity underlying targeting of this viral endonuclease.  Using tandem mass tag labeling with mass spectrometry, we also show that cytoplasmic mRNA degradation in infected cells that is coordinated by SOX and the host exonuclease Xrn1 causes widespread redistribution of RNA binding proteins between the cytoplasm and the nucleus.  RNA decay-driven translocation of select proteins to the nucleus leads to a global repression of RNA polymerase II occupancy on host but not viral promoters.  Collectively, our findings reveal surprising interplay between mRNA decay and RNAPII transcription in mammalian cells, which can be actively controlled by gammaherpesviral endonucleases.

The control of mRNA translation has emerged as a critical regulatory step in viral gene expression. Herpesviruses such as human cytomegalovirus (HCMV) must circumvent host defenses that limit translation, while effectively competing with cellular mRNAs for access to host ribosomes. The major goals of our work are to understand how HCMV counteracts host defenses that limit translation, and define factors that facilitate ribosome recruitment to viral mRNAs. Our recent work has uncovered a further level of regulation in the control of viral gene expression, namely the complex interplay between viral transcript diversity and viral protein synthesis. We find that alternative transcription start site (TSS) usage significantly expands viral transcript diversity. In many cases, alternative TSS usage generates viral mRNAs that encode the same viral protein, but differ in their 5’ untranslated regions (5’UTRs). As the 5’UTR regulates mRNA translation efficiency, alternative TSS usage can significantly impact the levels of viral protein expression. Using the HCMV major immediate early transcription unit as an example, we highlight how transcriptional diversity affects translational control, leading to different outcomes of HCMV infection.

Unlike many viruses that suppress cellular protein synthesis, host mRNA translation and polyribosome formation are stimulated by human cytomegalovirus (HCMV). Using polysome profiling, we previously established that HCMV selectively controls which cellular mRNAs are translated and that the landscape of translationally-regulated host mRNAs regulates HCMV reproduction. Examples from among the more than 1,000 host mRNAs whose translation is stimulated by HCMV infection will be presented to illustrate how select host factors impact HCMV productive replication. 

During HSV-1 latency, lytic gene expression is repressed and abundant RNA expression occurs primarily from the long repeat regions (RLs) of the genome. The HSV-1 LATs, transcribed from the RLs, have long been considered a hallmark or HSV-1 latency, though it has become clear that only a subset of neurons that contain HSV DNA produce detectable LATs.  In addition, at least 8 miRNAs, and recently additional lncRNAs, have been mapped to the LAT region. There are a number of biological phenotypes that have been mapped to the LAT region including functions that facilitate the establishment, maintenance and reactivation from latency as well as facilitating neuronal survival and anti-apoptosis. A central difficulty in dissecting the specific genetic functions responsible for these phenotypes is related to the genetic complexity of the RL region and the tendency of deletion mutants to alter multiple genetic elements. In this talk I will describe recent results using ribozymes targeting specific RNAs in vivousing AAV vectors as well as the use of an in vitrohuman neuronal culture system to dissect LAT and miRNA functions and characterize different profiles of lncRNA expression during latency.

The human gammaherpesviruses Epstein-Barr Virus (EBV) and Kaposi’s Sacroma-Associated Herpesvirus (KSHV) persist as chromatinized episomes in latently infected cells and viral-associated tumors.  Genome persistence is linked to viral gene expression and viral carcinogenesis.  EBV and KSHV share similar, but not identical, mechanisms of genome persistence involving viral encoded DNA binding proteins EBNA1 and LANA, respectively.  EBNA1 and LANA have similar DNA binding domains (related also to papillomavirus E2) that bind with sequence specificity to a repetitive element in their respective viral genomes.  Both viral proteins confer origin of DNA replication activity when tested in plasmid assays and are essential for episome maintenance in the context of the larger viral genomes.  Here, we describe recent studies on the mechanisms of EBNA1 and LANA in episome maintenance. We find that EBNA1 binds with sequence specificity to viral, as well as cellular genomic loci.  The function of EBNA1 at cellular genomic loci has been investigated.  In some locations, EBNA1 regulates cellular gene expression for a small set of genes important for B-cell development, cell survival, and tumorigenesis.  Circular chromatin conformation capture (4C) methods suggest that sequence-specific EBNA1 binding sites in the cellular genome are frequent sites for viral episome tethering.  In contrast, LANA does not bind with similar sequence specific DNA binding affinity to sites in the host genome, but rather appears to bind chromatin structures.  LANA binds with highest affinity at the KSHV terminal repeats (TR).  We found that mutations in the LANA oligomerization interface limit the ability of LANA to bind stably to the KSHV TR.  These mutations also lead to the loss of KSHV episome maintenance, as well as loss of KSHV genome integrity, with loss of viral DNA at regions distal to the TR.   Loss of LANA oligomerization also correlates with the loss of binding by the Origin Recognition Complex (ORC) to the TR.  Furthermore, LANA oligomerization is required for higher-order genomic and chromatin structures that can be detected by 3C methods, as well as by confocal life cell microscopy with fluorescent LANA proteins bound to viral episomes.  Our findings suggest that LANA must form higher order oligomeric complexes to bind efficiently at the KSHV TR, and this nucleates a genomic architecture enabling stable replication and maintenance of the genome during cellular proliferation.  These findings have implications for molecular “phase-change” processes in the regulation of genome maintenance and viral persistence.

The ability of human cytomegalovirus (HCMV) to reactivate from latent infection of hematopoietic progenitor cells (HPCs) is intimately linked to cellular differentiation. Although several HCMV genes have been implicated in HCMV latency and reactivation, the mechanisms involved in these events are poorly understood at the molecular level. HCMV encodes UL7 that we have shown is secreted from infected cells and induces angiogenesis. Recently, we found that UL7 is a ligand for Fms-like tyrosine kinase 3 receptor (Flt-3R), a well-known critical factor in HPC differentiation. We observed that UL7 directly binds Flt-3R and induces downstream signaling cascades, including phosphatidylinositol 3-kinase (PI3K)/Akt and mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK) pathways. Importantly, UL7 protein induces differentiation of both CD34+ HPCs and CD14+ monocytes. Lastly, an HCMV mutant lacking UL7 fails to reactivate in CD34+ HPCs in vitro as well as in humanized mice. These observations define the first virally encoded differentiation factor with significant implications not only for HCMV reactivation but also for alteration of the hematopoietic compartment in transplant patients.

Herpesviruses establish lifelong infection in their hosts. To attain this goal, they must modulate host immune responses. Of particular importance are natural killer (NK) cells, innate immune effectors capable of sensing and eliminating stressed cells. NK cells constitute the body’s primary line of defence against Herpesviruses. Over the past two decades, diverse mechanisms that inhibit NK cell recognition of herpesvirus-infected cells have been discovered. These include both viral proteins and miRNAs that target multiple points along cellular stress response pathways. These discoveries enable us to better understand the interplay between virus and host. But more than that, they shed light on the evolutionary pressures that shaped human immune ligand repertoires, and uncovered fundamental aspects of tumorigenesis and cell biology.

Following primary infection at peripheral mucosal sites, herpes simplex virus (HSV) is transported via innervating sensory neurons to replicate in the sensory ganglia and establishes life-long latency there.  The virus then reactivates periodically to cause lesions at or near the primary infection site.  This HSV life-cycle, which repeatedly primes the host immune system, increases the potential for a host immune response to eradicate the virus.  Therefore, HSV must have evolved multiple mechanisms to evade immune detection and clearance.  However, limited information is available regarding HSV immune evasion mechanisms that have been shown to be effective in vivo. In this talk, I will explore and discuss novel mechanisms of HSV-1 immune evasion with respect to inflammasome activation and cytotoxic T lymphocyte infiltration, which we have recently revealed as significant contributors to viral replication and/or pathogenesis in vivo.

Misincorporation of uracil or spontaneous cytidine deamination are common mutagenic insults to DNA. Herpesviruses encode a viral uracil DNA glycosylase (vUNG) and a viral dUTPase (vDUT), each with enzymatic and non-enzymatic functions. However, the coordinated roles of these activities in promoting gammaherpesvirus pathogenesis and genomic stability of the virus have not been defined. In addition, the potential compensation by the host UNG has not been examined in vivo. The genetic tractability of the murine gammaherpesvirus 68 (MHV68) pathogen system has enabled us to delineate the contribution of host and viral factors to the prevention of uracilated viral DNA. We find that MHV68 lacking vUNG (ORF46.stop) was not further impaired for replication in the lungs of UNG-/- mice compared to WT mice, indicating the vUNG provides non-overlapping functions from host UNG. Next, we investigated the separate and combinatorial consequences of mutating the catalytic residues of the vUNG (ORF46.CM) and vDUT (ORF54.CM). ORF46.CM was not impaired for replication, while ORF54.CM led to a slight transient defect in lung replication. However, disabling both vUNG and vDUT enzymatic activities had a significant impact on acute expansion in the lungs, followed by impaired establishment of latency in the spleen. Upon serial passage of the double ORF46.CM/ORF54.CM mutant in fibroblasts or the lungs of mice, we noted rapid loss of the non-essential YFP reporter gene from the viral genome, consistent with an increase in recombination. We are currently examining genomic mutations or rearrangements by whole-genome sequencing. Taken together, our data indicate that the vUNG and vDUT coordinate to promote viral genomic stability and enable viral expansion prior to colonization of latent reservoirs. We propose that the vUNG and vDUT are required to prevent mutagenic insults initiated by uracil incorporation. DNA repair of the uracilated viral genome is likely detrimental to herpesvirus fitness in the host.

Herpesvirus infections are ubiquitous in human populations, and relatively few therapeutics have been developed.  Initial development of antiviral drugs, in the absence of availability of alternatives, used clinical disease as an endpoint.  However, the development of subsequent drugs is more challenging as the comparator is an active drug that prevents most clinical disease.  The HIV field has advanced an approach that relies on surrogate marker of plasma HIV RNA for efficacy studies.  Herpesvirus drug development is also moving in that direction with the use of molecular detection of viral DNA as an endpoint in early clinical studies.  Studies using CMV and HSV drug development will be used as an example. 

The field of HSV- 2 genomics is evolving at a rapid pace due to the availability of next generation sequencing (NGS) technologies which provide the ability to sequence full-length and near full-length HSV-2 genomes.  Initial NGS strategies relied on cultured virus, but with the use of oligonucleotide enrichment, DNA swabs with low quantities of virus can also be used to generate high quality sequence.   HSV-2 genomic sequences have provided insight into the evolutionary origins and geographic variability of the virus, which are critically important questions for HSV-2 vaccine development.  HSV-2 has been found to be highly conserved worldwide, with limited genomic distance between strains.  Unexpectedly, HSV-1 recombination events in UL29, UL30, and UL39 have been found to be circulating throughout the world, supporting the hypothesis that recombination events that occurred prior to human emigration from Africa.  In addition, HSV-2 sequencing approaches have allowed for development of genotyping assays based on population prevalent single nucleotide polymorphisms.  Such assays have provided robust estimation of the prevalence of HSV-2 dual-strain infection, another important parameter for HSV-2 vaccine development.  Dual-strain infection is estimated to occur rarely, but is associated with HIV infection.  Initial studies of HSV-2 evolution within host over time show a small number of SNPs between primary and recurrent infection, suggesting that HSV-2 slowly evolves within a host.  These recent studies will be discussed in the context of HSV-2 pathogenesis, and outstanding questions that can be addressed by leveraging HSV-2 genomic studies will explored.