Fluorescence microscopy is the method of choice to monitor dynamic processes in living cells due to its non-invasive nature. A variety of different fluorophores and labeling systems are currently used to selectively visualise structures or biomolecules of interest. However, recent progress in the field of fluorescence microscopy towards super-resolution microscopy (SRM) has increased the requirements for fluorophores, as well as labeling strategies. In order to make these techniques available for routine live-cell imaging, brighter, cell-permeable fluorophores and adequate labeling strategies are necessary. This thesis explores methods to improve both fields through developing a new photoactivatable fluorophore, as well as optimising the self-labeling protein tag HaloTag with regard to brightness.
Photoactivatable fluorophores are important tools for single-molecule localisation based SRM as well as tracking experiments. In here, the development of a photoactivatable fluorophore based on silicon rhodamine (SiR) is described. This photoactivatble SiR (PA-SiR) activates via an unprecedented light induced protonation and forms a bright photoproduct. In contrast to other photoactivatable fluorophores, no caging groups are required, nor are there any undesired side-products released. PA-SiRs are environmentally sensitive and therefore allowed to create a HaloTag probe that was not only photoactivatable but also fluorogenic. This leads to high signal-to-background ratios in live-cell microscopy and makes the probe a powerful tool for SRM. Its use in both fixed-cell and live-cell single-molecule localisation microscopy (SMLM) was demonstrated and it became possible to follow the fast dynamics of mitochondria in live cells. Most excitingly, the unlabeled lumen of the mitochondria could be distinguished from their labeled outer membrane, showcasing the power of this probe in combination with SRM.
With most fluorophores HaloTag shows higher fluorogenicity when compared to the self-labeling protein tag SNAP-tag. To characterize the underlying reasons and to identify brighter HaloTag variants, a screening based approach was established to identify amino acid residues that affect fluorogenicity. HaloTag variants harbouring mutations on the surface in close proximity to the rhodamine binding site were found that showed both increased and decreased fluorescence intensities compared to parental HaloTag. Multiple rounds of directed evolution subsequently led to the isolation of a variant that showed increases in fluorescence intensity as high as 300% when combined with several fluorogenic rhodamines in vitro. Expression in mammalian cells and analysis by live-cell confocal microscopy revealed similar trends. Mechanistic studies revealed that all the variants that decreased the fluorescence intensity of rhodamine fluorophores did so due to quenching via the introduced tryptophans. In contrast, a clear explanation of how the isolated brighter variants increased fluorescence could not be identified. Regardless, these results highlight that the properties of fluorogenic fluorophores cannot only be tuned by synthetic strategies but also by modifications of the protein surface. It is expected that the identified variants will be beneficial to SRM as well as conventional live-cell microscopy as they both profit from brighter fluorescent probes.