17 16 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 0 / 2 1 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 0 / 2 1 Macromolecular Crystallography Group Beamlines I02, I03 and I04-1 Howwe HUSH jumping genes and viruses Related publication: Douse C. H.,Tchasovnikarova I. A.,Timms R.T., Protasio A.V, SeczynskaM., Prigozhin D. M., Albecka A.,Wagstaff J.,Williamson J. C., Freund S. M.V, Lehner P. J. &ModisY.TASOR is a pseudo-PARP that directs HUSH complex assembly and epigenetic transposon control. Nat. Commun. 11 , 4940 (2020). DOI: 10.1038/s41467-020-18761-6 Publication keywords: Chromatin;Transposons; Epigenetics; Genome stability;Viral restriction T hehumansilencinghub(HUSH)isacomplexofthreeproteinsdiscoveredin2015.Itplaysaroleinrepressingthereplicationofretroviruses suchasHIVand 'jumpinggenes' (transposons) that canmovewithin thegenome. HUSHworks by regulating chromatin, thematerial that packages our genomes, in particular ways. However, until recently little was known about the molecular and structural mechanisms underpinning HUSH function. TASORisthekeyscaffoldproteinoftheHUSHcomplex.AninternationalteamofresearchersinvestigatedhowTASORfeaturesinHUSHassembly and regulates chromatin structure over transposons. They found that one end of the TASORmolecule contains a domain that is not needed for assembly but is critical for HUSH function. They used Diamond Light Source’s Macromolecular Crystallography (MX) beamlines I02 and I03 to screendatasets of native and SeMet crystals of this domain.Their resulting structure revealed that this part ofTASOR resembles a domain from enzymes called PARPs that maintain genome stability. Close examination of the structure revealed that TASOR’s PARP domain is catalytically inactive – in other words, TASOR is a 'pseudo-PARP'. Although we now have anti-retroviral treatments for HIV that are very efficient at inhibiting actively replicating viruses, we are left with the challenge of clearing latent (dormant) viruses. Inhibiting or degrading HUSHmight reactivate latent viruses, allowing existing treatments to clear the infection – a ‘shock and kill’ strategy. Knowing the structure of functionally critical domains of HUSH, like the TASOR pseudo-PARP domain, will help develop HUSH inhibitors. Mammalian genomes are under constant threat of colonisation by infectious retroviruses and 'jumping genes' called transposons. The danger that retroviruses can pose to humans is well known – HIV being perhaps the most famous example – but transposons also play important roles in health and disease. Unchecked mobilisation of such genetic elements could cause genome instability, which is a hallmark of cancer. On the flip side, transposons harbour great potential to spread regulatory genetic networks, so are powerful drivers of evolution. Since transposon sequences are now thought to account for more than half of all human DNA, there is tremendous biological interest in understanding how such elements are controlled. One of the principal ways in which viruses and transposons alike are kept in check in human cells is through a process called epigenetic silencing. Through specific modifications to chromatin – the substance into which our genomes are packaged – expression of the underlying DNA can be tuned up or down. The ' epi' prefix refers to the notion that epigenetic modifications are a layer of information in addition to the primary genetic code of DNA. Epigenetic modifications include chemical modification (e.g. methylation) of the DNA itself, or of molecules called histones that package DNA into chromatin. The Human Silencing Hub (HUSH) complex has recently emerged as a central player in the epigenetic repression of exogenous viruses, including HIV, and endogenous transposons called long interspersed nuclear elements (LINEs). HUSH regulates methylation of histones in a particular pattern that is correlated with low conversion of the underlying DNA to its corresponding messenger RNA (Fig. 1). The latter is a process called transcription, which is the first step by which retroviruses replicate and most transposons jump to new positions. It has been known since HUSH was discovered in 2015 that the complex is composed of three proteins, called TASOR, Periphilin and MPP8, which team up to repress transcription through changes to chromatin structure 1 . However, almost nothing is known of the structure of these proteins, how they assemble together to form a functional complex, or what their key molecular activities are. This highlighted research set out to answer some of these questions and focused on HUSH subunitTASOR, which was completely uncharacterised at the start of the project. TASOR (Transgene Activation SuppressOR) is the central scaffold of HUSH: the other two subunits cannot interact with one another in TASOR’s absence. Careful biochemical delineation of the interaction sites of HUSH subunits allowed purification of binary complexes containing TASOR and either Periphilin or MPP8. Indeed, using Diamond’s beamline I04-1, the structure of the TASOR-Periphilin subcomplex was successfully solved 2 . In the course of research figuring out HUSH interactions, one end of the TASOR molecule was found to be critical for function but not assembly of the complex. Notably, at the start of the project this was actually the only portion of TASOR to have any annotation at all: a cryptic sounding 'domain of unknown function' (DUF). TASOR’s DUF could be purified in large quantities but initially failed to crystallise in more than two thousand tested conditions. However, spectroscopic measurements on the protein solution suggested the presence of a short highly flexible part protruding from its core structure. Such disorder is often incompatible with forming well-ordered protein crystals. Sure enough, upon targeted removal of this internal loop, beautiful single prismatic crystals could be grown (Fig. 2a). Using data collected at Diamond’s I02 and I03 beamlines, the high-resolution structure of the TASOR DUF could be solved. The X-ray structure revealed that TASOR DUF closely resembled a poly- ADP-ribose polymerase (PARP) domain (Fig. 2b). PARPs are enzymes that maintain genome stability by stimulating a signalling cascade in response to DNA damage. This was interesting given the putative role of HUSH in protecting the genome against potential damage caused by unrestricted transposon jumping or viral activity. However, upon aligning the structures and zooming into their active sites (Fig. 2c), it was clear that the physical shape and chemistry that form a functional PARP active site were not present in TASOR. This is most clear from the observations that key amino acids involved in PARP activity are mutated. Indeed, TASOR more closely resembled PARP family members such as PARP13 which is thought to be catalytically inactive. Biochemical experiments confirmed that TASOR lacked canonical cofactor binding and enzymatic activity – in other words, TASOR was identified as a 'pseudo-PARP' 3 . Despite the lack of catalytic activity, all known HUSH activities – targeted histone methylation, restricting the expression of target LINE transposons and of integrating viruses – were found to be dependent on the TASOR pseudo-PARP domain. Remarkably, it was found that substitution of just a single conserved amino acid residue on the surface of this domain was sufficient to wipe out all these complex functions. Interestingly, mice bearing mutation at another site, that we now know is a structurally important residue of TASOR, die early in embryonic development. More work is required to fully understand the order of events in HUSH assembly and activity – but taken together, these observations imply that genome protection by TASOR and HUSH is critical for healthy human development. The work highlighted here is just the start: figuring out more about how HUSH works will continue to be an intense area of study for researchers for many years to come. References: 1. Tchasovnikarova I. A. et al. Epigenetic silencing by the HUSH complex mediates position-effect variegation in human cells. Science. 348 , 1481–1485 (2015). DOI: 10.1126/science.aaa7227 2. Prigozhin D. M. et al. Periphilin self-association underpins epigenetic silencing by the HUSH complex. Nucleic Acids Res. 48 , 10313–10328 (2020). DOI: 10.1093/nar/gkaa785 3. Douse C. H. et al. TASOR is a pseudo-PARP that directs HUSH complex assembly and epigenetic transposon control. Nat. Commun. 11 , 4940 (2020). DOI: 10.1038/s41467-020-18761-6 Funding acknowledgement: This work was supported byWellcome Trust Senior Research Fellowships 101908/Z/13/Z and 217191/Z/19/Z to Y.M., aWellcome Trust Principal Research Fellowship 101835/Z/13/Z to P.J.L., Swedish Society for Medical Research grant S19-0100 to C.H.D., and a BBSRC Future Leader Fellowship BB/N011791/1 to C.H.D. Corresponding author: Dr Christopher H. Douse, Lund University, Sweden,
[email protected] Figure 1: Schematic model for HUSH-dependent repression of integrated retroviral and transposon DNA (blue). Targeted histone methylation (red) depends on HUSH complex assembly at the chromatinised sequence. Figure 2: (a) Optimisation of crystal growth of TASOR domain of unknown function (DUF); (b) Overall structures of TASOR DUF (green) compared to active (PARP1, red) and inactive (PARP13, blue) members of the PARP superfamily; (c) Zoomed-in views of enzymatic sites show that inactive PARPs have closed, empty and mutated sites compared to PARP1, which can be seen bound to its cofactor.