UVR8-based optogenetics and new model systems to study the DNA damage response

In optogenetics, light is used as a minimally invasive cue to control the activity of cellular processes. At a molecular level, light induces conformational rearrangements in photosensor proteins to facilitate or dislodge protein-protein interactions or to mask or unmask functional domains. We envisaged the potential of optogenetics as a tool in the interrogation of the DNA damage response and cell cycle regulation. Commonly used photosensors are the red light sensitive phytochrome PhyB with binding partner PIF3 or PIF6, the blue light sensitive cryptochrome CRY2 with interacting partner CIB1, and a broad family of proteins containing blue light sensitive LOV domains. Arabidopsis thaliana UVR8 was recently characterized as an ultraviolet-B photosensor (Rizzini et al., 2011) that interacts with COP1 after exposure to ultraviolet-B. We developed a new optogenetic system based on UVR8 and COP1 and demonstrate its performance in mammalian cells. We were able to import a protein of interest into the nucleus, induce gene expression from a plasmid with GAL4-UAS, or, by outfitting our microscope with a custom-built collimator accessory module, to recruit proteins of interest to subsections of the chromatin, all under the control of ultraviolet-B light.

We then developed a system to initiate DNA double strand breaks using this optogenetic system. The study of DNA double breaks is important, as they are a prominent source of chromosomal instability leading to cancer. Using H2B as a platform for optogenetic recruitment, we recruited and activated FokI nuclease domains in chromatin, both in the entire nucleus, or confined to a 4 µm wide line. Both procedures led to fast onset of apoptosis, likely in response to highly digested DNA.

Recruitment of nuclease activity to a specified genomic region can be achieved by using deactivated Cas9 (dCas9) fused to UVR8. This system has the advantages that knowing the location of the induced double strand break facilitates later analysis, while with recruitment of FokI nuclease limited to several loci, toxicity due to excessive DNA damage is alleviated. As expected, use of this system in combination with guide RNA targeting the telomeric repeats, enabled ultraviolet-B induced recruitment of FokI to the telomeres. Cells still entered apoptosis, but with delayed onset. When we assayed for DNA damage response, we noticed that even in cells lacking FokI constructs or ultraviolet-B induction, there was recruitment of RPA to the dCas9 bound telomeric foci. Also most of the cells transfected with the dCas9-UVR8 fusion protein were in the G1 phase of the cell cycle. This phenotype, resembling a DNA damage response, depended solely on the recruitment of dCas9 to the telomeric repeats. While this complicated our studies of ultraviolet-B induced DNA double strand breaks, it also led to a new research direction. We reasoned that dCas9 arrays in the genome induce regions of single stranded DNA (ssDNA), which are then bound by RPA.

Indeed, upon further investigation we confirmed that dCas9 induces regions of ssDNA when targeted to the telomeric repeats. This ssDNA then recruited DNA repair and checkpoint proteins, such as RPA, ATRIP, BLM and Rad51, at the telomeres. Interestingly, targeting all these proteins to telomeric ssDNA was observed even in cells that were in the G1 phase of the cell cycle. Therefore, this system has the potential to serve as a platform for further investigation of DNA replication stress responses at specific loci in the human genome and in all phases of the cell cycle.