Uncovering the molecular mechanisms of human cytomegalovirus immunoevasins US2 and US11 using genetic screens

During millions of years, the evolutionary arms race between viruses and their hosts has resulted in mutual adaptation. The host has equipped itself with an extensive arsenal of antiviral mechanisms to defend itself against these intruders, while viruses have developed strategies to counter, evade and even exploit host immune responses. RNA viruses, such as myxoviruses, picornaviruses, and retroviruses, evade immune responses mostly through antigenic variation. This may be related due to their limited coding capacity. DNA viruses, such as poxviruses and herpesviruses, have a larger coding capacity and can afford to encode proteins dedicated to manipulating the hosts immune system. Scope of this thesis Most of our current knowledge on ERAD stems from early studies performed in Saccharomyces cerevisiae. This is related to the difficulty of gene manipulation in mammalian cells compared to yeast. In addition, the ERAD pathway in mammalians is considerably more intricate than in yeast, probably because the ERAD pathway has seen a high diversification and specialization due to the emergence of complex life and the expansion of the proteome. While many of the findings in yeast can be translated to mammalians due to the high conservation among eukaryotes, our current understanding of ERAD in mammalian cells is still far from complete. Recent advances in mammalian gene manipulation have created new ways for extensive interrogation, such as the generation of lentivirus-based high-throughput wholegenome shRNA libraries for conducting RNAi screens. Additionally, in early 2013, a groundbreaking new gene manipulation technique for use in mammalian cells emerged: CRISPR/Cas9-mediated genome engineering. This technique utilizes the bacterial nuclease protein Cas9 in combination with a guide RNA (gRNA) to promote programmed editi ng of DNA within mammalian cells. By matching the gRNA sequence to a desired sequence present in the human genome, the Cas9 molecule can be programmed to cleave any given site in the human genome, thereby producing a double-stranded break (DSB). The DSB is recognized and repaired by the non-homologous end joining (NHEJ) system, but due to the error-prone nature of NHEJ, the introduction of mutations is inevitable. These mutaƟ ons may result in gene disruption, ultimately producing a target gene knockout. The CRISPR/Cas9 genome-engineering technique was quickly adopted to effi ciently produce gene knockouts and knockins, in both small and large scale setups. After a publication in early 2013, in which the CRISPR/Cas9-tool was first used to modify the mammalian genetic code, the amount of publications using this technique for gene editing has skyrocketed. Herpesvirus immunoevasins not only demonstrate how viruses evade the immune system, but also represent valuable tools to further unravel the MHC-I anti gen presentation pathway and general cellular processes, including ER-associated protein degradation. We decided to study mammalian ERAD using the HMCV immunoevasins US2 and US11 as a model; the latest RNAi library screening and CRISPR/Cas9 genome-engineering techniques were employed to identify novel players in MHC I degradation. In Chapter 2, we review the interactions of herpesviruses with the MHC-I anti gen presentation pathway. In Chapter 3, we employ a pooled lentiviral genome-wide shRNA library to screen for host factors essential for US11-mediated MHC-I downregulation. We identify the previously uncharacterized protein TMEM129 as the elusive E3 ubiquitin ligase and UBE2J2 and UBE2K as the E2 ubiquitinconjuga ting enzymes exploited by US11. We also find evidence that TMEM129 might play a more general role in mammalian ERAD. In Chapter 4, we explore the topology of TMEM129. Many of the E3 proteins involved in ERAD are multi-spanning transmembrane proteins. We show that TMEM129 is an ER-localized tri-spanning transmembrane protein with a C-terminal cytosolic RING domain. In Chapter 5, we generate a CRISPR/Cas9 library focused on known human p97 co-factors to find the ones that are essential for US11- mediated MHC-I degradation. We identify UBXD8 as a p97 co-factor that is essential for MHC-I degradation by US11. In Chapter 6, we establish a CRISPR/Cas9 library targeting all known E2 ubiquitin-conjugating enzymes. Using this library, we screen for E2 enzymes essential for US2-induced degradation of MHC-I. We identify UBE2G2 as an E2 ubiquitin-conjugati ng enzyme that is critically involved in US2-mediated MHC-I degradation. Surprisingly, we find UBE2J2 to counteract degradation of MHC-I by US2. Chapter 7 summarizes the findings of this thesis and discusses the potential of combining various cuting-edge techniques to study mammalian ERAD