Our research focuses on the molecular characterization of DNA replication stress and its contribution to genome instability. We aim to understand the mechanistic basis of genome rearrangements arising during perturbed DNA replication in stem and somatic cells, affecting various aspects of human disease and most specifically cancer. These studies take advantage of an established technological platform, ranging from standard molecular and cell biology methods, to specialized single-molecule in vivo analysis of replication intermediates.

Replication fork remodeling upon cancer chemotherapeutic treatments

DNA replication interference is one of the most common strategies employed in cancer chemotherapy. Although many of these drugs have been used in the clinics for decades, their molecular mechanism of action is often poorly understood, preventing the informed selection of appropriate chemotherapeutic regimens for different tumors and the development of potent combinatorial treatments. Combining classical cell and molecular biology with specialized single-molecule approaches on replication intermediates (Zellweger and Lopes, Meth Mol Biol 2018), we have uncovered surprising alterations of replication fork architecture upon several chemotherapeutic treatments. Most notably we reported the conversion of a high number of replication forks into four way junctions (Figure 1A), a process also known as replication fork reversal (Ray Chaudhuri et al., Nature Str Mol Biol 2012; Zellweger et al., JCB 2015; Neelsen and Lopes, Nature Rev Mol Cell Biol 2015). We have provided several lines of evidence that these transactions at replication forks are transient and genome protective, making them attractive targets for combinatorial chemotherapy. We have thus recently started to elucidate the cellular factors playing a role in the formation and resolution of these structures, and uncovered the central recombinase RAD51, PCNA polyubiquitination and the DNA translocase ZRANB3 as strictly required for drug induced fork reversal (Zellweger et al., JCB 2015; Vujanovic et al., Mol Cell 2017). We are now expanding these studies to understand how DNA damage signaling events and DNA repair proteins collaborate to orchestrate fork progression, remodeling and protection, contributing to genome integrity. Despite their genome maintenance function, we have shown that reversed forks can also become entry points for extensive, nuclease-mediated degradation of newly replicated DNA (Mijic et al., Nature Comms 2017), which was recently reported as a crucial molecular determinant of the exquisite sensitivity to cancer chemotherapeutics observed in BRCA-defective tumors (Figure 1B-C).


Figure 1. Replication fork reversal triggers clinically-relevant fork degradation in BRCA2-defective cells.

A. Representative electron micrograph of a replication fork converted into a 4-way junction (magnified in the inset) by replication fork reversal. P, parental duplex. D, daughter duplexes. R, regressed arm.

B. Frequency of reversed forks visualized by EM in U2OS cells, in the indicated conditions. Reversed forks are induced by nucleotide depletion (HU), but degraded in a MRE11-dependent manner (Mirin=MRE11 inhibitor) in BRCA2-defective cells1.

C. Graphical model for the role of fork reversal in the clinically relevant degradation of stalled forks upon BRCA2-defects. RAD51/ZRANB3-induced fork reversal leads to ds-ends at the regressed arms: these ends are targeted by controlled re- section in wild type cells, to mediate fork restart, but are subjected to pathological fork degradation in BRCA2-defective cells, leading to chemosensitivity of BRCA2-defective tumours. Genetic impairment of this fork degradation restores fork integrity and chemoresistance in BRCA2-defective cells. See Mijic et al., Nature Communications 2017.


Most recently we have investigated the peculiar transactions that allow replication forks to bypass and repair inter-strand crosslinks. These are particularly toxic adducts – induced by byproducts of cell metabolism, as well as several chemotherapeutic drugs – which were long considered roadblocks to the replication process, but are in fact mysteriously bypassed at high efficiency during genome replication of human cells (Mutreja et al., Cell Reports 2018). Given the crucial role of replication fork remodeling in the response to chemotherapeutic treatments, we are now attempting to build specific read-outs for these molecular transactions, in order to use them for genome-wide screens and to reveal novel cellular factors modulating chemo-sensitivity or chemo-resistance in specific types of cancer. We also plan to expand these studies from cancer cell lines into animal models and clinical samples, in order to exploit these specialized approaches as prognostic tools, prospectively contributing to cancer personalized medicine.


DNA replication stress in stem cells

Embryonic stem cells (ESCs) have the unique ability to self-renew and are capable of differentiating into multiple cell types. In contrast, adult stem cells – such as hematopoietic stem cells (HSCs) - repopulate specific tissues. The exhaustion of adult stem cells has been linked to ageing, but the underlying molecular mechanisms are still largely unknown. Several knockout-mouse models have uncovered a role for numerous DNA repair factors in ageing and cancer. Besides well-known repair activities, conditional deletion of the ATR gene - which is a central factor activated in response to DNA replication stress - causes depletion of the stem cell niche, suggesting that stem cells need to protect their genomes during active proliferation. We are investigating the intriguing connection between replication stress and aging, applying some of our most revealing approaches to different populations of stem cells. We have recently shown in cultured ESCs and mouse embryos that H2AX phosphorylation is dependent on ATR and is associated with chromatin loading of the ssDNA-binding proteins RPA and RAD51 (Figure 2A; Ahuja et al., Nature Comms 2016).


Figure 2. Replication stress markers in embryonic stem cells and their alternative, replication-coupled strategy for genome maintenance.

A. Immunofluorescence staining for the DNA damage-marker γH2AX, and for chromatin binding of the ssDNA-binding proteins RPA32 and RAD51 in E3.5 blastocysts. The results indicate activation of DNA damage response and accumulation of ssDNA in unperturbed mouse embryonic stem cells (ESCs) within their natural environment.

B. A model depicting differential control of genome stability in ESCs and proliferating somatic cells. Under-replicated regions and residual DNA damage are unavoidably present at the end of each S phase in both ESCs and somatic cells. Differentiated cells have prolonged gap phases, assemble 53BP1 nuclear bodies and repair most of these lesions prior to S phase entry. Conversely, owing to the brief gap phases, ESCs channel a high number of these lesions into the following S phase and protect genome integrity by extensive fork reversal and replication-coupled repair. See Ahuja et al., Nature Communications 2016.


Single-molecule analysis of replication intermediates reveals massive ssDNA gap accumulation, reduced fork speed and frequent fork reversal. All these marks of replication stress – which surprisingly do not impair the mitotic process - are rapidly lost at the onset of differentiation and result from the rapid transition through the G1 phase, which is strictly required to maintain pluripotency. In this context, when cell cycle checkpoints are mostly inactive and numerous DNA lesions are channeled into replication, fork slowing and reversal are strictly required to avoid chromosomal breakage and represent an effective alternative strategy of genome maintenance, compatible with the high proliferation rate of these cells (Figure 2B; Ahuja et al. Nature Comms 2016). In light of the recently proposed links between adult stem cell proliferation, DNA damage and cancer, we are now expanding these studies to investigate whether activation of quiescent adult stem cells leads to detectable replication stress phenotypes.


Replication stress during unperturbed S-phase and upon oncogene activation

Besides the established role of reversed forks in response to chemotherapeutic treatments (see above), we have provided solid evidence that these unusual intermediates accumulate also during unperturbed S phase, at regions that are intrinsically difficult to replicate, such as repetitive sequences (Follonier et al., NSMB 2013; Ray Chaudhuri et al., MCB 2015). Intriguingly, in light of the remarkable similarity of regressed arms to double-stranded breaks (DSB), we have uncovered – in collaboration with the Penengo lab here at the IMCR – a crucial role for classical DSB-response factors in promoting reversed fork stability and restart, mediating efficient fork progression during unperturbed S phase (Figure 3; Schmid et al., Mol Cell 2018).


Figure 3. DNA damage response factors control remodeling and progression of replication factors at endogenous obstacles.

A. DNA fiber analysis of replication fork progression in living human cells, by incorporation of nucleotide analogues and labelling of newly replicated tracts. Defects in the canonical DNA damage response (DDR) lead to frequent fork stalling and asymmetric progression of sister forks during unperturbed S phase

B. Agarose bidimensional electrophoresis of replication intermediates from SV40-derived plasmids replicating in human cells and containing expanded GAA/TTC repeats (known to challenge replication fork progression). Besides expected signals for moving and pausing replication forks, DDR-defective cells display accumulation of reversed replication forks at the repetitive sequences, suggesting that canonical DDR factors are required to restart replication forks that pause and reverse at endogenous difficult-to-replicate sequences. See Schmid et al., Molecular Cell 2018.


We are also currently investigating another type of endogenous replication stress, linked to replication-transcription interference and to accumulation of DNA-RNA hybrids, known as R-loops. We are attempting to provide direct visualization of these events in human cells and to understand the molecular mechanism of interference that these events are known to exert on the replication process.

We have also contributed to uncover the molecular determinants of oncogene-induced replication stress, which is known as one of the earliest causative events in tumorigenesis. We have applied our specialized investigation platform to cell culture models, studying specific genetic alterations frequently associated with early tumorigenesis, such as CyclinE- and CDC25A overexpression, or partial DNA re-replication (Neelsen et al., JCB 2013; Neelsen et al., G&D 2013). We are currently extending our studies to additional oncogenes and to more clinically relevant experimental systems, including 3D cultures of cancer cells and simultaneous alterations of oncogenes and tumor suppressors. We aim to investigate how multiple alterations can concur to fuel genomic instability in early tumorigenesis, uncovering the specific contributions of R-loops and replication/transcription interference in these phenomena.



Funding of our research is currently provided by University of Zurich (Forschungskredit), Swiss National Science Foundation, European Research Council, Krebsliga Schweiz and European Molecular Biology Organization.