Fluorescent sensing methodologies for dynamic nanoscale events

Promovendus/a: Toon Van Thillo

Promotor(en): Prof. dr. Peter Dedecker

Life is dynamic, defined by the ability to maintain homeostasis in ever-changing environments. On a cellular level, this requires an intricate signalling network to regulate activities and reactivities in response to external stimuli. This complex network can be investigated with fluorescent indicators, remarkable molecules whose fluorescence emission depends on the presence of a specific stimulus, such as a signalling molecule. We use fluorescence microscopy to observe these molecules and to study their distribution throughout the cell. These studies have suggested that the subcellular architecture plays a major role in the regulation of this signalling network, even down to the nanoscale. However, quantitative observation of dynamic events at this nanoscale remains challenging.

In this thesis, we describe novel methodologies that take advantage of these fluorescent indicators to observe and to manipulate dynamic activities at the nanoscale. We discuss our perspectives on the potential and pitfalls of genetically-encoded biosensors, a subclass of fluorescent indicators which can be produced in living cells. We develop a genetically-encoded biosensor that is well-suited to the conditions encountered when imaging at the nanoscale. However, imaging artefacts can arise when observing these genetically-encoded biosensors at high resolution. To mitigate these artefacts, we implement a correction based on the optical transfer function, an intrinsic property of every microscope. This approach is explored using computer simulations and experimental characterization. We then combine our new genetically-encoded biosensor with this analysis strategy to demonstrate how these new tools can be used for high-resolution biosensing.

However, cellular signalling occurs in a highly complex environment, and an artificial system that reduces this complexity could simplify its experimental characterization. Therefore we also implement a model system that can mimic nanoscale signalling events. We use bacterial proteins that form nanochannels in lipid membranes to rapidly create such artificial nanodomains. These so-called biological nanopores allow for the passage of small molecules and ions, and this passage can be manipulated by the application of an external electric field. We theorize that we can leverage this electrical manipulation to create short-lived nanodomains near the nanopore aperture. We select calcium to characterize our model system, and observe its nanodomain formation with a fluorescent calcium indicator. We confirm that the nanodomain volume can be accurately controlled and provide evidence for sub-millisecond control of the nanopore, thereby generating the smallest rapidly controllable reaction volume presented to date.