Fluorescent protein design for superresolution microscopy. Exploring the power of protein engineering

A profound insight into life can only be obtained by studying living systems with high spatiotemporal resolution. Until now, the most powerful method for doing this is light microscopy. Light microscopy allows us to study living systems, be it cells or complete organisms, with a submicrometer spatial and subsecond temporal resolution. To study specific molecules or reactions amidst the multitude of processes going on, one typically labels one specific molecule or process with a fluorescent marker, and images the system with fluorescence microscopy. Traditionally, this is done using small organic fluorophores or fluorescent proteins. Fluorescent proteins (FPs) are proteins that contain a fluorophore that is autocatalytically formed and absorbs and emits in the visible wavelength region. Being genetically encoded, they are ubiquitously used as reporter genes and as highly specific markers for fluorescence imaging. After the initial discovery of green fluorescent proteins, many variants with modified and improved properties were made. For instance, while the first FP was a green FP, the first variants had altered excitation and emission spectra. Nowadays, FPs spanning almost the entire visible range are available. One interesting subtype of fluorescent proteins are what we call “photophysically smart labels”, the photophysical behavior of which is dependent on the light with which they have been irradiated. These labels’ emissive properties depend on the light they have encountered before and are of crucial importance in diffraction-unlimited fluorescence microscopy. We call this class of FPs the phototransformable FPs. Examples of phototransformable fluorescent proteins are reversibly photoswitchable FPs and irreversibly photoconvertible FPs. In this dissertation, I introduce some basic concepts and techniques regarding the work that follows. Then, I describe in two publications my contribution to the repertoire of phototransformable FPs: in Chapter 2, I describe how I could rationally design a FP that is both reversibly photoswitchable from a bright to a dark state as well as irreversibly photoconvertible from a green to a red state. I did this by introducing photochromic behavior into Dendra2, a photoconvertible FP. In Chapter 3, I went the other way around. Using rational and random mutagenesis, I could introduce green-to-red photoconversion behavior into the green photochromic FP Dronpa. These studies have led to two new FPs, namely NijiFP (based on Dendra2) and pcDronpa2 (based on Dronpa). I showed that these labels can be used in advanced microscopy applications, including diffraction-unlimited fluorescence microscopy. In the last two chapters, the focus is on the microscopy, more specifically photochromic stochastic optical fluctuation imaging (pcSOFI). pcSOFI is a technique that allows an improvement of spatial resolution by making use of the intrinsic flickering of fluorophores. Chapter 4 is a reprint of a book chapter in which I first describe reversibly photoswitchable FPs and their applications in diffraction-unlimited fluorescence microscopy. In a second part, I describe how the reversibly photoswitchable FP Dronpa can be used to do pcSOFI. In Chapter 5 then, I tested a number of FPs as to their performance in pcSOFI microscopy. From this study, it was found that EGFP, the most widely used FP, typically seen as a “non-smart fluorophore”, is an ideal label for pcSOFI. The results that I obtained and describe in this dissertation have contributed to a broader understanding of FPs at an atomic level. Concretely, they have shown how particular residues influence particular photophysical properties. Next to these fundamental insights, I also made several new FPs, the most important of which are NijiFP, ffDronpa and pcDronpa2. In the second, microscopy-oriented part of this dissertation, I focused on pcSOFI microscopy. Via the step-by-step guide and the testing and scoring of different labels, I hope to have broadened the application area of pcSOFI and hope to have brought this simple and robust technique to a non-specialized public.

Jaar van publicatie:2015

Toegankelijkheid:Open