53BP1: guarding the genome with a novel liquid weapon

In this Comment, Naveen Tangudu and Katherine Aird discuss recent findings showing that 53BP1 regulates heterochromatin through liquid-liquid phase separation.

heterochromatin organization and maintenance. Further analysis of the 53BP1 protein identified a core component containing the oligomerization domain (OD) that is critical for puncta formation (Fig. 1). Additionally, using genome wide ChIP-Seq, the authors correlated enrichment of 53BP1 and H3K9me3 at multiple loci, strengthening the observation that 53BP1 localizes to heterochromatin. While 53BP1 is known to be recruited to DSBs in heterochromatin 11 , this is the first report demonstrating recruitment of 53BP1 to heterochromatin under unperturbed conditions. One key function for heterochromatin in maintaining genome stability is to repress tandem repetitive DNAs or repetitive elements 12 . Upon loss of heterochromatin, these elements can cause genomic instability by affecting chromosome segregation, inducing replication stress, increasing transposon hopping, or impairing correct DSB repair 2 . Indeed, the authors found that binding of 53BP1 at heterochromatin marked by H3K9me2/3 is required to repress the expression of repetitive elements. Therefore, it is interesting to speculate that 53BP1 may function in tandem with other histone modifications to epigenetically regulate multiple other biological processes outside of its role in DNA repair.
The most striking finding of this study is that 53BP1 forms nuclear puncta along with HP1α to regulate heterochromatin via liquid-liquid phase separation (LLPS) 1 . Liquid-liquid phase separation has emerged as a crucial mechanism to facilitate multiple biological processes including chromatin organization 13 , DNA repair 14 , and gene transcription 15 . LLPS has been used to explain the formation of known nuclear organelles such as nucleoli and promyelocytic leukemia nuclear bodies (PML NBs) as well as other membraneless condensates (e.g., nucleosome arrays, DNA damage foci, stress granules, proteasomes, autophagosomes, etc.) 16 . Recent studies in multiple organisms, including mammalian cells, found that core heterochromatin proteins including HP1α, SUV39H1, and TRIM28 can undergo LLPS when certain conditions are met, including specific protein-protein interactions and/or post-translational modifications 17 . While 53BP1 has been previously shown to undergo phase separation 6 , prior to this study, it was unknown what factors or proteins contribute to the regulation of heterochromatin by 53BP1 through LLPS. The authors found that HP1α is required for 53BP1 puncta formation and 53BP1's LLPS 1 . By re-expressing different truncated and mutated 53BP1 constructs, the authors further identified domains required for LLPS and puncta formation that are dispensable for 53BP1 foci formation at DNA DSBs (Fig. 1). Together, these data provide evidence that LLPS of 53BP1 and puncta formation at heterochromatin are distinct from the role of 53BP1 in DNA repair.
Finally, the authors set out to determine the contribution of 53BP1 puncta on functional readouts of genome stability. After treating cells with the DNA damage agent bleomycin, the truncated mutant 53BP1 that is still capable of puncta formation but incapable of forming foci at DSBs rescued both total DNA damage and survivial 1 . These data suggest that the LLPSmediated 53BP1 puncta formation at heterochromatin is indeed critical for some aspects of genome stability. Genome instability due to loss of heterochromatin can also lead to cellular senescence, a state of stable cell cycle arrest with characterized hallmarks of increased DNA damage response, senescence associated heterochromatic foci (SAHF), and the senescence associated secretory phenotype (SASP) 18 . The SASP is characterized by secretion of a wide range of cytokines, chemokines, matrix metalloproteinases, and growth factors 18 . Many labs, including ours, have shown that SASP components IL6 and CXCL8 (IL8) are highly transcriptionally upregulated in senescent cells [19][20][21] .
Here, the authors found that induction of IL6 and CXCL8 was rescued to some extent by 53BP1 constructs that are proficient in puncta formation. Interestingly, our previous work has shown that high-mobility group box 2 (HMGB2) protects SASP genes from heterochromatin to allow for their transcriptional upregulation 21 . Further, we recently found that the histone methyltransferase disruptor of telomeric silencing 1-like (DOT1L), which promotes the active histone marks H3K79me2/3, helps to promote SASP gene expression 19 . It would be interesting to determine whether 53BP1 antagonizes HMGB2 and/or DOT1L to regulate SASP genes. In addition to this, future studies to characterize other senescent markers besides the SASP, such as senescence-associated beta-galactosidase (SA-β-gal) activity, cell proliferation, cell size, and SAHF, would strengthen the notion of 53BP1's involvement in cellular senescence through this novel mechanism.
Understanding mechanisms related to heterochromatin formation and maintenance is critically important since its dysregulation can lead to genome instability, thereby promoting pathological consequences such as cancer or aging. The study from Zhang et al. advances our knowledge of how heterochromatin is maintained to promote genome stability through a novel LLPS-mediated function of 53BP1, although additional studies to further mechanistically delineate how 53BP1 promotes heterochromatin maintenance and what effects this has on cell fate are needed. Due to its known role in promoting HRmediated DNA repair, there has been some interest in identifying 53BP1 inhibitors to help with CRISPR/Cas9-mediated genome editing 22 . The data presented by Zhang et al. would suggest that caution is needed when designing these inhibitors. As they identified specific amino acids in the OD domain that are required for heterochromatin maintenance yet dispensable for binding of 53BP1 at DSBs, the observations presented in this study not only provide new insight into functions for 53BP1, but also provide the opportunity to target specific domains related only to DNA repair and NHEJ to inhibit this pathway while maintaining heterochromatin.