A heterochromatin inducing protein differentially recognizes self versus foreign genomes

Abstract

Krüppel-associated box-domain zinc finger protein (KRAB-ZFP) transcriptional repressors recruit TRIM28/KAP1 to heterochromatinize the mammalian genome while also guarding the host by silencing invading foreign genomes. However, how a KRAB-ZFP recognizes target sequences in the natural context of its own or foreign genomes is unclear. Our studies on B-lymphocytes permanently harboring the cancer-causing Epstein-Barr virus (EBV) have shown that SZF1, a KRAB-ZFP, binds to several lytic/replicative phase genes to silence them, thereby promoting the latent/quiescent phase of the virus. As a result, unless SZF1 and its binding partners are displaced from target regions on the viral genome, EBV remains dormant, i.e. refractory to lytic phase-inducing triggers. As SZF1 also heterochromatinizes the cellular genome, we performed in situ-footprint mapping on both viral and host genomes in physically separated B-lymphocytes bearing latent or replicative/active EBV genomes. By analyzing footprints, we learned that SZF1 recognizes the host genome through a repeat sequence-bearing motif near centromeres. Remarkably, SZF1 does not use this motif to recognize the EBV genome. Instead, it uses distinct binding sites that lack obvious similarities to each other or the above motif, to silence the viral genome. Virus mutagenesis studies show that these distinct binding sites are not only key to maintaining the established latent phase but also silencing the lytic phase in newly-infected cells, thus enabling the virus to establish latency and transform cells. Notably, these binding sites on the viral genome, when also present on the human genome, are not used by SZF1 to silence host genes during latency. This differential approach towards target site recognition may reflect a strategy by which the host silences and regulates genomes of persistent invaders without jeopardizing its own homeostasis.

Three-Dimensional Chromatin Structure of the EBV Genome: A Crucial Factor in Viral Infection

“…Interestingly, during latency EZH2 complex binds the EBV genome at the promoter region of lytic genes, but upon reactivation of EBV, the association of EZH2 with the viral genome is lost, and H3K27me3 mark is erased around lytic genes [ 59 , 60 , 61 ]. Moreover, recent studies revealed that the TRIM28/KRAB-ZFP/SZF1 complex plays an essential role in repressing EBV lytic gene expression by promoting the deposition of the heterochromatin mark at the promoter of lytic genes, facilitating the establishment of EBV latent infection [ 62 , 63 , 64 , 65 ]. Thus, the stable EBV gene expression patterns observed during latency depend on histone deposition and DNA methylation across the viral genome.…”

Inflammasome, the Constitutive Heterochromatin Machinery, and Replication of an Oncogenic Herpesvirus

“…This suggested that epigenetic mechanisms govern the fated decision to respond to a lytic trigger or remain latent and refractory to such stimuli. Subsequently, genome-, transcriptome-, and proteome-wide experiments, using lytic and refractory sorted cells, have revealed several insights [ 16 , 17 , 18 ], some of which will be described in the following sections. Notably, (i) exposure of latently infected cells to disparate lytic triggers reshapes the cellular transcriptome into a ‘pro-lytic’ state upstream of the expression of ZEBRA [ 19 ]; and (ii) an ancient cellular epigenetic silencing machinery that represses foreign genomes, such as those of endogenous retroviruses, also silences herpesvirus lytic gene expression.…”

“…It turns out that SZF1, a KRAB–ZFP that recruits KAP1 to host and gammaherpesvirus DNA, recognizes self, i.e., pericentromeric regions, using a repeat sequence-bearing motif, while targeting EBV lytic genes via non-consensus binding sites. Remarkably, SZF1 does not use the motif to recognize the EBV genome nor the binding sites on the viral genome to silence host genes during latency [ 16 ]. This differential approach towards target site recognition reflects a strategy by which the host (a) rapidly silences newly arrived foreign genomes and (b) regulates genomes of persistent invaders without jeopardizing its own genomic integrity.…”

An Ancestral Retrovirus Envelope Protein Regulates Persistent Gammaherpesvirus Lifecycles

“…KAP1 then recruits histone modifying enzymes and DNA methyl transferases to epigenetically silence select regions of the genome ( Groner et al, 2010 ; Cheng et al, 2014 ; Jang et al, 2018 ; Bhaduri-McIntosh and McIntosh, 2021 ). Previously, we and others have demonstrated that KAP1 also silences the genomes of EBV and KSHV, as well as the betaherpesvirus cytomegalovirus (CMV) during latency ( King et al, 2015 ; Rauwel et al, 2015 ; Li et al, 2017 , 2018 , 2019b ; Burton et al, 2020 , 2021 ; Bhaduri-McIntosh and McIntosh, 2021 ). In response to lytic cycle triggers/cues, suppression of cellular KAP1 by depletion triggers the lytic switch protein ZEBRA or via phosphorylation enhances expression of ZEBRA, leading to activation of lytic replication from latency ( King et al, 2015 ; Li et al, 2017 , 2019b ; Burton et al, 2020 , 2021 ; Bhaduri-McIntosh and McIntosh, 2021 ).…”

“…Previously, we and others have demonstrated that KAP1 also silences the genomes of EBV and KSHV, as well as the betaherpesvirus cytomegalovirus (CMV) during latency ( King et al, 2015 ; Rauwel et al, 2015 ; Li et al, 2017 , 2018 , 2019b ; Burton et al, 2020 , 2021 ; Bhaduri-McIntosh and McIntosh, 2021 ). In response to lytic cycle triggers/cues, suppression of cellular KAP1 by depletion triggers the lytic switch protein ZEBRA or via phosphorylation enhances expression of ZEBRA, leading to activation of lytic replication from latency ( King et al, 2015 ; Li et al, 2017 , 2019b ; Burton et al, 2020 , 2021 ; Bhaduri-McIntosh and McIntosh, 2021 ). Bam HI Z EBV replication activator (ZEBRA) and replication and transcription activator (RTA) in EBV, or RTA in KSHV, are immediate early viral transcription factors, also known as latent-to-lytic switch proteins, that initiate transcription of the vast majority of viral genes in a cascade that proceeds kinetically with early lytic genes (pre viral DNA replication) to late lytic genes (post viral DNA replication) ( Kenney and Mertz, 2014 ; Broussard and Damania, 2020 ).…”