Neuronal Enhancers are Hotspots For DNA Single-Strand Break Repair

Background

Our cells have evolved various DNA repair pathways to deal with DNA lesions that constantly challenge the stability of our genome. Defects in DNA repair may lead to the development of cancer and immunological or neurological disorders (Jackson and Bartek, 2009). Repair of various DNA lesions, including single-strand breaks (SSBs), involves a gap-filling step, which incorporates one or several nucleotides into the damaged DNA using the undamaged DNA strand as a template.

Although neurons face different kinds of DNA damaging insults, defects in the SSB repair pathway seem to impact predominantly their function (Caldecott, 2008; Tubbs A, 2017). Indeed, hereditary mutations in genes encoding for SSB repair proteins, like XRCC1, are mainly associated with neurological phenotypes.

However, the extent and genome-wide distribution of SSB repair signatures in neurons are not clear. Thus, to gauge the genome-wide distribution of DNA repair signatures in post-mitotic neurons, the authors of this preprint developed a novel technique called SAR-seq and report that neurons accumulate SSBs in enhancer regions of the genome.

 

Key findings

1. SAR-seq identifies sites of unscheduled DNA synthesis

Unscheduled DNA synthesis occurs during DNA repair to fill the gaps generated by excised or missing nucleotides (also known as gap filling). To map sites of DNA repair synthesis genome-wide, the authors developed a new method: synthesis-associated with repair sequencing (SAR-seq, fig 1a). For this purpose, they labelled iPSC-derived neurons (i³ neurons) with the thymidine analogue EdU, which is incorporated at the sites of DNA synthesis. The labelled DNA was biotinylated by click chemistry, sheared, and isolated for high-throughput sequencing. They identified >55,000 peaks at recurrent genomic locations around 200-2,000 bp in width, possibly reflecting clustered DNA repair loci. SAR peaks were also observed in rat neurons but not in other post-mitotic cells like G0-arrested pre-B cells or iPSC-derived skeletal muscles. Notably, SAR peaks were unaffected by inhibitors of the replicative DNA polymerases (α, δ, and ϵ), reinforcing that i³ neurons are not replicating and the SAR signal is from unscheduled DNA synthesis.

 

2. Neuronal enhancers are hotspots for unscheduled DNA synthesis

They found that neuronal SAR peaks were enriched in intragenic regions and associated with expressed genes. In particular, ~50% of the top-5000 peaks harboured a transcription factor motif – ONECUT – that is known to promote chromatin accessibility. Furthermore, these peaks also coincided with ATAC-seq peaks (a readout of open chromatin), suggesting that an open chromatin structure may influence the extent of DNA synthesis. Intriguingly, a majority of SAR peaks were enriched at cis-regulating enhancer elements (rather than promoters), as measured by enhancer histone modifications (H3K4me1, H3K27ac) and Lysine Methyltransferase 2B (MLL4) (ChIP-seq, fig 1b). Adding, most of the SAR peaks occupied super-enhancers that are active in differentiated neurons (fig 1c). Gene Ontology (GO) analysis also revealed that genes containing SAR peaks were related to neuronal functions (like neuronal migration, development, axon formation, and synapse assembly).

3. Neuronal DNA synthesis coincides with SSB and PARP activation loci

DNA double-strand breaks (DSBs) induced by Topoisomerase II (TOP2) can promote the expression of early response genes in neurons (Madabhushi, 2015). However, DNA synthesis associated with TOP2-induced lesions did not overlap with the SAR peaks observed in untreated neurons (reiterating that SSBs mostly occur in untreated i³ neurons). Additionally, no DSBs were detected in untreated neurons, as monitored by immunofluorescence for DSB markers or END-seq, a technique that maps DSBs (Canela, 2016).

Furthermore, they found that SAR peaks corresponded to sites of PARylation and XRCC1 recruitment, key events in SSB repair (assayed using ChIP-seq, fig 1d). In SSB repair either a single nucleotide or longer DNA patches are replaced respectively by DNA polymerase β (short-patch SSB) or DNA polymerase δ, ϵ or l (long-patch SSB). The authors observed that PARP1, XRCC1, or DNA polymerase β depletion increased SAR peaks, suggesting reliance on long-patch repair when the short-patch repair is challenged (fig 2).

 

Conclusion and perspective

In summary, this study reports that in human post-mitotic neurons, SSBs and unscheduled DNA synthesis occurs within enhancers. The presence of neurodegenerative phenotypes in patients with mutations in XRCC1 suggests a key role for short-patch SSB repair in neuronal viability (fig 2). The authors speculate that loss of XRCC1 function could lead to neuronal dysfunction by aberrant PARP1 activity or by the accumulation of mutations that could result in aberrant transcription and neurodegeneration.

Future work could reveal how different non-replicating cells deploy different DNA repair pathways and how endogenous DNA damage supports the normal physiology of different kinds of cells.

 

Acknowledgments: We are thankful to all the authors for their support and for taking the time to comment on the preLight.

 

References

• Caldecott, K. W. Single-strand break repair and genetic disease. Nat Rev Genet (2008).
• Canela, A.et al., DNA Breaks and End Resection Measured Genome-wide by End Sequencing, Mol Cell (2016).
• Jackson, S. P. & Bartek, J. The DNA-damage response in human biology and disease, Nature (2009).
• Madabhushi, R.et al., Activity-Induced DNA Breaks Govern the Expression of Neuronal Early-Response Genes, Cell (2015).
• Mellén M et al., 5-hydroxymethylcytosine accumulation in postmitotic neurons results in functional demethylation of expressed genes, Proc Natl Acad Sci U S A. (2017).
• Puc J,et al. Ligand-dependent enhancer activation regulated by topoisomerase-I activity, Cell (2015)
• Tubbs A, & Nussenzweig A. Endogenous DNA Damage as a Source of Genomic Instability in Cancer, Cell (2017).

Tags: enhancers, neurons, parp1, single-strand breaks, xrcc1

doi: https://doi.org/10.1242/prelights.26737

Read preprint (No Ratings Yet)