Van Vugt lab: DNA damage response in cancer

Already in the 1950s, movies of cells that were irradiated during mitosis revealed that these cells continued mitosis without repairing their broken chromosomes. Only decades later, the mechanistic underpinning of the silenced DNA damage response during mitosis was uncovered. The mitotic kinases CDK1 and PLK1 inactivate key DNA break repair components upon mitosis entry, and thereby suppress canonical DNA repair. However, the response to DNA damage is not entirely inactivated during mitosis. Initial detection of DNA breaks still occurs, whereas actual repair of DNA lesions appears blocked. Intriguingly, we previously showed that specific DNA repair components, including RIF1, perform alternative functions in genome maintenance during mitosis, suggesting that DNA damage processing is re-wired rather than inactivated.

Whereas the mitotic response to DNA breaks is blunted, cancer cells are apparently able to deal with incompletely replicated DNA in mitosis, and must have dedicated pathways to do this. Although many details are unclear, several mechanisms to resolve incompletely replicated DNA in mitosis have been demonstrated.

1. DNA replication can be completed during mitosis by ‘mitotic DNA synthesis’ (MiDAS), which involves a dedicated DNA polymerase complex, and resembles RAD52-dependent break-induced replication (BIR). Unclear is how newly initiated replication forks can be initiated and sustained in a mitotic environment, and whether MiDAS reflects events in mitosis or in late G2-phase.
2. Several DNA endonucleases are activated during mitosis (e.g. SLX1, MUS81, XPF), or get access to chromosomes upon mitotic break-down of the nuclear membrane (i.e. GEN1). These endonucleases can separate aberrantly connected DNA molecules, allowing faithful chromosome segregation and completion of mitosis. How the resulting chromosome fragments are repaired, when canonical DSB repair pathways are inactivated during mitosis remains unclear.
3. Incompletely replicated DNA fragments that remain unresolved early during mitosis can form ‘ultrafine DNA bridges’ (UFBs) during anaphase. UFBs arise at sites of incomplete DNA replication, but can also results from unresolved recombination intermediates and persistent centromeric DNA catenanes. UFBs are marked by the recruitment of the PICH DNA translocase. UFBs originating from incompletely replicated DNA are converted into single-stranded DNA bridges, which likely facilitates UFB breakage and allows chromosome segregation at the cost of genome instability.

The molecular regulation of the above-mentioned mechanisms remains largely elusive. It is unclear whether these pathways represent parallel, consecutive or integrated mechanisms, and whether additional mechanisms exist. It is poorly understood how these mechanisms operate on cancer-relevant sources of DNA damage in mitosis.

Genomically instable cancers rely on mechanisms to survive their high levels of replication stress. These mechanisms therefore constitute potentially attractive therapeutic targets. We study the molecular wiring of these mitotic DNA repair mechanisms, and want to explore the therapeutic impact of targeting these mechanisms in cancer cells.

Genomic instability and inflammation are intricately connected hallmark features of cancer. A consequence of genomic and chromosomal instability is the leakage of DNA from the nucleus into the cytoplasm, either directly or through the formation and subsequent rupture of micronuclei. Cytoplasmic DNA subsequently activates cytoplasmic DNA sensors (including cGAS and RIG-!), triggering downstream pathways, including a type I interferon response. We recently showed that DNA repair defects, for instance due to BRCA1/2 mutation, indeed instigate immune signaling through the cGAS/STING pathway (Heijink et al, Nature communications, 2019). The subsequent inflammatory signaling provides both tumor-suppressive as well as tumor-promoting traits. To prevent clearance by the immune system, genomically instable cancer cells need to adapt to escape immune surveillance. Currently, it is unclear how genomically unstable cancers are rewired to escape immune clearance.

We study extend our studies on the mechanisms by which genomic instability trigger inflammatory signaling and want to understand the adaptive mechanisms by which cancer cells can 'fly under the radar' of the immune system. Ultimately, this knowledge should lead to starting points of how we can therapeutically activate the immune system to improve the treatment of genomically instable cancers.

Precise cell cycle control is critical for proliferating cells, especially under conditions of genomic stress. Modulation of the cell cycle checkpoint machinery is therefore often proposed as a therapeutic strategy to potentiate anticancer therapy. Therapeutic inhibition of checkpoint kinases can deregulate cell cycle control and improperly force cell cycle progression, even in the presence of DNA damage. Chemical inhibitors for several cell cycle checkpoint kinases have been developed (including for ATR, CHK1 and WEE1). Preclinical research has shown that the efficacy of therapeutic checkpoint inhibition is context-sensitive and depends on the genetic make-up of an individual cancer. Therefore, to optimally implement such novel inhibitors in the clinic, the molecular characteristics that determine inhibitor activity need to be identified to select eligible patients and to anticipate on mechanisms of acquired resistance.

Inhibition of ATR, CHK1 and WEE1 was initially considered an attractive anticancer therapy for TP53 mutant tumors. However, additional factors besides p53 inactivation may determine checkpoint inhibitor sensitivity. Using focused approaches and genetic screening approaches, we want to uncover the genetic factors that determine sensitivity to cell cycle checkpoint inhibition.

Cancers harboring oncogene-induced replication stress have been suggested to be particularly dependent on cell cycle control. In order to find better treatments for tumors with oncogene-induced replication stress, it could be of great clinical interest to target pathways that allow tumors to deal with replication stress. In addition, we previously showed that tumor cells with genome instability due to defective homologous recombination also depend on the ATR and WEE1 replication checkpoint kinases for their survival.

We additionally searched for genes that determine WEE1 inhibitor response using unbiased functional genetic screening. We discovered that the mutational status of several S-phase genes, including CDK2, determines the cytotoxicity induced by Wee1 inhibition (Heijink et al, PNAS, 2015). Thus, inactivation of nonessential S-phase genes can overcome Wee1 inhibitor resistance, while allowing the survival of genomically instable cancer cells. We are currently extending these approaches to ovarian cancer models in vitro and in vivo.

Even under physiological conditions, DNA replication is continuously challenged. For instance, some genomic regions are intrinsically difficult to replicate due to repetitive sequences (e.g. common fragile sites or contain alternative DNA structures (e.g. DNA/RNA hybrids and G-quadruplex DNA). Conversely, DNA can be chemically modified by cellular metabolites. In cancer cells, however, many additional processes lead to perturbed DNA replication, with a central role for oncogene activation.

We study how cancer cells deal with DNA lesions that arise upon oncogene overexpression. Amplification of the CCNE1 or MYC proto-oncogenes, for instance, leads to uncoordinated firing of replication origins, causing collisions between the translation and replication machinery and depletion of the nucleotide pool.

When replication is persistently perturbed, replication forks can stall and collapse, leading to DNA double-strand breaks (DSBs). For the repair of these highly toxic DNA lesions, cells depend on homologous recombination (HR), which is only active in S/G2 phase of the cell-cycle, and uses the newly formed sister chromatids as the repair template. Repair of these DNA lesions is achieved in coordination with cell-cycle checkpoints. These checkpoints integrate kinase-driven and transcriptional signaling, allowing for both rapid and sustained checkpoint activation to prevent entry into mitosis. These combined mechanisms prevent the initiation of mitosis in the presence of unrepaired DNA damage or incompletely replicated DNA.

We study how replication-born DNA lesions are processed in the cell cycle. Remarkably, we observed that DNA damage in Cyclin E1-overexpression cells frequently is transmitted into mitosis. Using isogenic models and analysis of patient samples, we aim to understand how oncogene-overexpressing cells deal with DNA damage, and if these mechanisms can be therapeutically targeted.