Development of a novel bioimaging and analysis platform to study DNA damage-induced nucleolar–nucleoplasmic protein shuttling in living cells

Abstract

The nucleolus, mainly known for its function in ribosome biogenesis, has recently emerged as a central hub regulating numerous non-canonical functions, including the DNA damage response. Several DNA repair factors have been identified to reside in nucleoli under non-stress conditions, and at least some have been reported to translocate from nucleoli into the nucleoplasm in response to DNA damage. Yet, the molecular mechanisms underlying nucleolar–nucleoplasmic protein shuttling remain poorly understood. Dynamic post-translational modifications (PTMs), such as phosphorylation or acetylation, are assumed to mediate protein translocation upon induction of genotoxic stress. Additionally, protein–protein interactions and liquid–liquid demixing are likely to contribute to these dynamics. In this thesis, the contribution of poly(ADP-ribose) polymerase 1 (PARP1) and poly(ADP-ribosyl)ation (PARylation) to the nucleolar–nucleoplasmic shuttling of two key genome maintenance factors — the base excision repair (BER) scaffold protein XRCC1 and the RecQ helicase WRN — was investigated. Immunofluorescence experiments were conducted in HeLa wild type (WT) and PARP1 knockout (KO) cells to analyze the role of PARP1, while the clinical PARP inhibitor (PARPi) olaparib was used to assess the contribution of PARylation. Our study revealed that after treatment with the oxidizing agent hydrogen peroxide (H₂O₂) and the alkylating agent 2-chloroethyl ethyl sulfide (CEES), but not the topoisomerase inhibitor camptothecin (CPT). This translocation required the presence of PARP1 but was independent of its enzymatic activity. Interestingly, XRCC1 remained in the nucleoplasm after H₂O₂ treatment in HeLa WT cells, whereas in PARP1 KO cells it was only transiently released and rapidly re-localized to nucleoli. Together with the observation that pharmacological PARP inhibition modulates XRCC1 redistribution, these findings indicate that nucleolar release of XRCC1 is PARP1-independent, whereas its nucleoplasmic retention requires auto-PARylated PARP1 as a loading platform. Furthermore, PARP1 itself was released from nucleoli after H₂O₂ treatment, indicating that PARP1 may serve as a shuttle for other proteins. Intriguingly, gossypol, which has been reported in previous studies to disrupt PARP1–protein interactions, fully abolished nucleolar–nucleoplasmic shuttling of WRN, XRCC1 and PARP1 itself, suggesting the involvement of upstream regulatory factors. Collectively, these findings demonstrate that nucleolar–nucleoplasmic protein shuttling is regulated in a toxicant and protein-specific manner (Veith S., Schink A., Engbrecht M. et al., 2019). To provide a detailed overview of this emerging field, current knowledge on nucleolar and PARP1 biology, as well as their crosstalk is summarized in a comprehensive review. In addition, potential implications of nucleolar PARP1 functions for cancer biology and therapy were discussed (Engbrecht M. and Mangerich A., 2020). Finally, this thesis presents a new image acquisition and analysis platform to investigate nucleolar–nucleoplasmic protein shuttling dynamics in living cells. The platform combines high-content microscopy with a newly developed KNIME-based analysis workflow implementing state-of-the-art segmentation tools, enabling robust analysis even in challenging conditions (e.g., clustered cells). This system was validated in cells transiently transfected with different GFP-tagged nucleolar proteins previously shown to translocate into the nucleoplasm upon genotoxic stress, including PARP1, the macrodomain-containing protein TARG1 and the BER protein APE1. Moreover, a HeLa cell line stably expressing PARP1-eGFP and the nucleolar marker NPM1-mCherry was developed. This new reporter cell line was subsequently used to perform detailed analyses of nucleolar–nucleoplasmic PARP1 shuttling after treatment with H₂O₂ alone or in combination with pharmacological inhibitors. In this context, we tested a panel of small-molecule inhibitors targeting different PTM-mediating enzymes. These included inhibitors of ATM, ATR, sirtuin 7 (SIRT7), the MRN (Mre11–Rad50–Nbs1) complex inhibitor (Z)-mirin, and the broad-spectrum kinase inhibitor staurosporine. While inhibition of ATM, ATR, the MRN complex, or kinases did not affect PARP1 shuttling, SIRT7 inhibition led to loss of nucleolar PARP1 localization independent of H₂O₂ treatment. Furthermore, the five clinical PARPi veliparib, niraparib, olaparib, rucaparib and talazoparib were tested for their potential effects on H₂O₂-induced PARP1 shuttling. All PARPi reduced nucleoplasmic PARP1 retention compared to H₂O₂ treatment alone, but with distinct effects on nucleolar–nucleoplasmic shuttling. While upon pre-incubation with veliparib, olaparib and rucaparib, PARP1 quickly relocated to nucleoli, niraparib and talazoparib prolonged nucleoplasmic retention. Veliparib, at a concentration of 1 µM, was the only inhibitor that significantly inhibited nucleolar PARP1 release. We conclude that several overlapping mechanisms, including differences in inhibitory and trapping potencies, as well as distinct binding interactions of individual PARPi, likely contribute to the observed compound-specific effects. Interestingly, after H₂O₂ treatment, PARP1 accumulated at the nucleolar periphery, which contains heterochromatic domains, indicating a potential yet unknown function of PARP1 in this region under genotoxic stress. In addition, an increase in silicon-rhodamine (SiR)-Hoechst fluorescence was observed after H₂O₂ treatment, which was completely abolished by PARP inhibition. Since PARP1-dependent PARylation mediates chromatin relaxation at DNA damage sites, facilitating access of DNA repair factors, the observed increase in SiR-Hoechst intensity likely reflects PAR-dependent chromatin decondensation (Engbrecht M. et al., 2023, Manuscript in preparation). In summary, the imaging and analysis platform established in this thesis enables detailed investigation of rapid nucleolar–nucleoplasmic protein dynamics in living cells. While many mechanistic questions remain open, this work provides new insights into the regulation of DNA damage-induced nucleolar–nucleoplasmic protein shuttling.