Deutsch Intern
    Institute for Virology and Immunobiology


    Systems biology of herpesvirus infections

    Herpesviruses are large DNA viruses, which co-evolved with their human and animal hosts for millions of years. Humans are infected with eight different herpesviruses causing a broad spectrum of disease ranging from the common cold sores to cancer. During these millions of years of co-evolution hosts, herpesviruses learned to both comprehensively modulate their host cell environment and efficiently evade the immune system. Besides representing important pathogens to human health, they thus also represent interesting tools to study fundamental aspects of cell biology and immunology.

    Within their large DNA genomes, herpesviruses encode hundreds of viral proteins and peptides, many of which do not only regulate a single gene, DNA or protein but interfere with complex cellular signal networks. To see beyond the tip of the iceberg of this regulation a systems level approach is required. Our lab employs a broad range of system biology methodology and analysis tools including microarrays, next-generation sequencing and quantitative proteomics to study host cell modulation and immune evasion in various herpesvirus models.

    During lytic infection, cellular gene expression is subject to rapid alterations induced by both viral and antiviral mechanisms. Standard approaches quantifying changes in both total RNA and protein levels do not provide the necessary temporal resolution to elucidate the underlying molecular mechanisms. Our lab pioneered metabolic labelling of newly transcribed RNA using 4-thiouridine (4sU-tagging) to analyse short-term changes in RNA synthesis, processing and decay with superior resolution (Dölken et al., RNA 2008; Friedel et al., NAR 2009; Windhager et al., Genome Research 2012; Rutkowski et al., Nature Communications 2015). Moreover, ribosome profiling depicts short-term changes in translational activity. It is based on large-scale sequencing of ribosome-protected RNA fragments and provides both a quantitative estimate of translational activity as well as a detailed picture of the translated RNA sequences at any given time of interest.

    Combining data obtained by 4sU-tagging, ribosome profiling and quantitative proteomics, we can now comprehensively record changes in RNA synthesis, processing and decay as well as their impact on protein production in a single experimental setting (Rutkowski et al., Nature communications 2015). Combined with cell-based and biochemical assays as well as additional high-throughput assays, we aim to understand the coordinated regulation of gene expression by viral proteins, microRNAs and RNA-binding proteins as well as their functional relevance in infection. We apply this approach to study host cell modulation by HSV-1, MCMV and HCMV.

    This work requires intensive bioinformatics analysis. These are performed in collaboration with the group of Prof. Florian Erhard at our institute as well as the groups of Prof. Ralf Zimmer and Prof. Caroline Friedel at the Ludwig-Maximilians-University Munich.

    Manipulation of RNA metabolism by herpes simplex virus 1 (HSV-1)

    In the frame of the ERC Consolidator Award “HERPES” (Herpesvirus Effectors of RNA Synthesis, Processing, Export and Stability), we study how HSV-1 manipulates the transcriptional machinery during productive infection. We previously reported that HSV-1 triggers widespread, host-specific disruption of transcription termination (Rutkowski et al., Nature Communications 2015). This results in extensive transcriptional activity for tens-of-thousands of nucleotides and into downstream genes. Recently, we observed that ongoing transcription downstream of the affected poly(A) sites leads to a massive increase in chromatin accessibility downstream of the affected genes. Work is ongoing to elucidate the underlying molecular mechanism employing a complementary approach of systems biology methodology and reverse virus genetics.

    Functional reannotation of herpesvirus genomes

    Application of systems biology methodology like RNA-seq and ribosome profiling revealed herpesvirus gene expression to be substantially more complex than previously thought. Within their 165-230kb genomes they encode hundreds of novel transcripts and open reading frames (ORFs). Based on a broad range of systems biology data including transcription and translation start site profiling, ribosome profiling and quantitative proteomics, we re-annotate the genomes of HSV-1 as well as murine and human cytomegalovirus (MCMV/HCMV). We also developed a new nomenclature to incorporate the novel gene products into the existing nomenclature. The fully revised HSV-1 genome annotation can be found at:  https://www.biorxiv.org/content/10.1101/603654v1. Work for HCMV and MCMV is ongoing.

    A particular interesting finding from ribosome profiling experiments are hundreds of novel small herpesvirus ORFs (sORFs). The majority of these are expressed upstream of previously annotated larger ORFs that represent so called upstream open reading frames (uORFs). Cellular uORFs are prevalent in eukaryotic genomes and constitute an important, yet poorly understood regulatory network governing gene expression at the level of translation. We hypothesize that viral uORFs allow these viruses to adapt viral gene expression to cell type, stress and inflammation.

    Transcriptional regulation at single cell level

    Single cell RNA sequencing (scRNA-seq) has highlighted the important role of intercellular heterogeneity that contributes to phenotype variability in both health and disease. Gene expression is a stochastic process, with intrinsic and extrinsic noise in transcription and translation contributing to intercellular heterogeneity in both mRNA and protein levels. However, this inherent characteristic cannot be resolved using current scRNA-seq approaches. A further key limitation of all existing approaches is that the RNA profile of each individual cell can only be analyzed once.

    We combined metabolic RNA labeling using 4-thiouridine with chemical nucleoside conversion and scRNA-seq to develop thiol-(SH)-linked nucleotide conversion sequencing (scSLAM-seq) (Erhard et al., Nature 2019 ). In addition, we developed the computational approach GRAND-SLAM (Global Refined Analysis of Newly transcribed RNA and Decay rates using SLAM-seq) to to absolutely quantify the new-to-total RNA ratio (NTR) for each gene in all cells, and to estimate quantification uncertainty, i.e. for genes with few reads (Jürges et al., Bioinformatics 2018 ).


    Our approach allows direct recording of transcriptional activity by differentiating newly synthesized from pre-existing RNA for thousands of genes in a single cell. scSLAM-seq recovers the earliest virus-induced changes in cytomegalovirus infection. It depicts transcriptional bursting kinetics and demonstrates extensive gene-specific differences that correlate with gene-/promoter-intrinsic features (Tbp-TATA-box interactions and DNA methylation). Gene and not cell-specific features thus explain the heterogeneity in transcriptomes between individual cells and the transcriptional response to perturbations. Moreover, “on-off” rather than “up-down” kinetics shape the intrinsic cellular response to infection with widespread implications for infection control and autoimmunity. We are now applying scSLAM-seq to other model systems.

    Functional role of cytomegalovirus miRNAs

    Although usually asymptomatic in healthy individuals, human cytomegalovirus (HCMV) is the major cause of morbidity in immunocompromised patients and allogeneic bone-marrow or organ-transplant recipients. As such, it poses an important risk factor for graft failure following heart and kidney transplantation, resulting in patient death or the need for re-transplantation. In addition, it is the leading agent of birth defects among congenitally transmitted infections affecting about 1:1,000 new-borns. During primary infection and reactivation, HCMV encounters an array of innate and adaptive immune responses.

    MicroRNAs (miRNAs) represent a novel entity of viral factors counteracting these defences requiring no protein expression to exert their function. This makes them ideal, non-immunogenic tools for these viruses to regulate their own as well as host gene expression during latency and reactivation thereof. To date, more than 80 miRNAs have been identified in six human herpesviruses – at least 11 pre-miRNAs are expressed by HCMV. We established its murine model to study the biology and function of cytomegalovirus miRNAs (Dölken et al., J Virol 2007) and identified two MCMV miRNAs to be required for efficient virus persistence and host to host spread (Dölken et al., PLoS Pathogens 2010).

    We employ a broad range of methodology including reverse virus genetics and new high-throughput approaches to study the function of herpesvirus miRNAs and answer whether they may serve as readily accessible targets for novel antiviral agents.