The protein actin polymerizes to produce filaments that form crosslinked networks in the cytoplasm of cells. These networks support many fundamental biological processes — such as cell movement and division and the intracellular trafficking of molecules. Reports that actin also has functions in the cell nucleus remain controversial1,2, partly because of the challenges of performing experiments that exclusively perturb the nuclear actin pool without also perturbing actin in the cytoplasm. Two studies3,4 in Nature now provide the most compelling evidence so far that polymerized actin has roles in cell nuclei in which DNA has been damaged, and that it could be essential for maintaining genome stability.
In cell nuclei, DNA is packaged with proteins to form a material called chromatin, which is subdivided into different domains. One of these domains, known as heterochromatin, is a tightly packed form of DNA characterized by large segments of repeated DNA sequences. These are highly prone to an aberrant form of genome shuffling called ectopic recombination, and therefore present a serious threat to genomic integrity. In fruit flies and mice, heterochromatic DNA that contains double-strand breaks (DSBs) is moved outside this domain to prevent such aberrant recombination5–7, but the mechanism involved has not been known. Actin filaments form in the nuclei of mammalian cells in response to DNA damage8,9, but their function in DNA repair has also been unclear.
Caridi et al.3 show that a subset of DSBs in fruit-fly cells move from heterochromatin towards the nuclear periphery in long-lasting, directed motions, rather than along random paths. Such directed motion has previously been observed only for a handful of cases in the nuclei of mammalian cells10–12. Crucially, the authors provide insight into how this re-location occurs. They find that nuclear actin polymerizes to form filaments at hetero-chromatic DSB-repair sites, in a process that requires the presence of the protein complex Arp2/3 and its activators (the Scar and Wash proteins). They observe that Arp2/3 promotes the formation of actin filaments that grow from hetero-chromatic DSBs towards the nuclear periphery. The nuclear motor proteins myosin I and myosin V then ‘walk’ the repair sites along the actin filaments (Fig. 1a). This walk is triggered by the myosin-activating protein Unc45 after filament formation, indicating that relocation is regulated both in space and in time.
An important finding of Caridi and colleagues is that recruitment of Arp2/3 and myosins at heterochromatic DSBs requires the DSB-repair protein Mre11 and a heterochromatin component, the HP1α protein. This suggests that DSB detection or processing is required for the recruitment of the motor proteins to the damaged sites and reveals why only heterochromatic DSBs are relocated to the nuclear periphery. Notably, the authors demonstrate that, when nuclear actin and myosins are defective in both fruit-fly and mouse cells, the integrity of the genome in hetero-chromatin is impaired and cells become less able to survive DNA damage, suggesting that actin and myosin have key roles in maintaining genome stability in the nucleus.
DSB repair can proceed by two distinct mechanisms: the error-prone non-homologous end-joining pathway (NHEJ) and the usually error-free homologous recombination (HR) pathway. During NHEJ, broken ends of DNA are simply re-stitched together, whereas HR requires that DNA ends are first processed to generate stretches of single-stranded DNA, which then search for appropriate (homologous) DNA sequences to use as a template for repair. Schrank et al.4 report that polymerization of nuclear actin in human cells is specifically required for the efficient repair of DSBs by HR.
Using a cellular system that controllably induces DSBs mainly within euchromatin13, the authors find that — as in heterochromatic breaks — Arp2/3 is recruited to the damaged genomic sites. However, the Arp2/3 activators recruited to these breaks are different from the ones shown by Caridi et al. to be recruited at heterochromatic breaks. This raises the possibility that different regulatory mechanisms promote actin-filament formation in different genomic and chromatin contexts.
Intriguingly, Schrank and colleagues find that Arp2/3 is recruited only at DSBs undergoing HR, whereas the Wiskott–Aldrich syndrome protein (WASP, an Arp2/3 activator) is recruited at breaks repaired by both HR and NHEJ. This observation indicates the existence of a currently unclear regulation mechanism that promotes actin polymerization so that it occurs specifically at breaks undergoing HR. The researchers also show that inhibition of nuclear-actin polymerization leads to a decrease in the processing of DNA ends and in the efficiency of HR repair, placing nuclear actin at the core of the cellular response to DNA damage (Fig. 1b).
Schrank and co-workers’ efforts might turn out to be relevant to disease. Wiskott–Aldrich syndrome (WAS) is a disorder characterized by severe immunodeficiency and predisposition to cancers caused by mutations in the WAS gene14 (which encodes WASP). How WAS mutations cause malignancies is largely unknown at the molecular level. Schrank et al. report that inhibition of WASP leads to defects in HR repair, and that immune cells (lymphocytes) from people with WAS show evidence of defects in DNA-end processing and have higher sensitivity to DNA-damaging agents than do lymphocytes from healthy people. Future studies should investigate whether the repair defects associated with WASP malfunction in the nucleus contribute to the malignancies in people with WAS.
The authors further show that perturbations of actin polymerization in the nucleus lead to a decrease in the mobility of DSBs, alongside the repair defects. This reduced motion contributes to a reduction in the ability of a subset of DSBs to form clusters within the nucleus. The authors propose that this reduced clustering limits HR efficiency by decreasing the local concentration of factors that process DNA ends. Given that relatively few DSBs form clusters, another plausible scenario is that actin-dependent DSB movement and DNA-end processing are parallel events, and are not functionally linked.
The two new studies provide a framework for further investigations into the role of nuclear polymerized actin in the cell’s response to DNA damage. Although the key biological players that support the nuclear functions of actin have been identified, the physical connections that attach damaged chromatin to nuclear actin structures and to motor proteins are unknown. The next step should be to use super-resolution microscopy and other imaging tools to fully characterize actin structures in the nucleus and to discover how the architecture of chromatin and the nucleus affects the formation and remodelling of actin polymerization. Further studies are also required to unravel the mechanism by which polymerized actin facilitates chromatin mobility and DNA-end processing. Given that genomic instability is a major contributor to the development of cancer, working out how cells use actin to safeguard their genome will have implications for our understanding of the basic principles of cancer aetiology.
Nature 559, 35-37 (2018)