Abstract:
The central nervous system, and in particular brain, controls how we perceive the world
around us. The brain processes information collected by the receptors and generates responses in the form of electrical (action potentials) and chemical (neurotransmitters) signals. Such information processing is possible due to neurons which are basic brain units. In order to execute their complex function, neurons have a unique molecular organization emerging from interactions between various ion channels, receptors, cell adhesion molecules, adaptors, and cytoskeletal elements. Voltage-gated sodium channels (NaV) are especially important for the generation of action potentials in neurons. Out of nine isoforms, the most abundant in the adult mammalian brain is NaV1.6. The NaV1.6 is clustered at high density at the axon initial segment (AIS) and the nodes of Ranvier, where it promotes the initiation and propagation of action potentials, respectively. Proper expression and maintenance of the NaV1.6 clusters at the AIS and nodes of Ranvier depend on various proteins (e.g., ankyrin G and cell adhesion molecules). One particularly important cell adhesion molecule that directly interacts with NaV1.6 is a 186 kDa neurofascin isoform (NF186). Since both proteins are indispensable for proper signaling in the brain, they are implicated in various neurological conditions. For example, mutations of NaV1.6 have been linked to epilepsy. The altered expression and localization of NaV1.6 have also been found in multiple sclerosis. The NF186 has also been linked to multiple sclerosis and other demyelinating diseases. In order to decipher how NaV1.6 and NF186 contribute to various diseases and design new therapeutic strategies, one must first understand the regulation of function, localization, and trafficking of these two proteins under physiological conditions. However, there are many open questions on the regulation of NaV1.6 and NF186 in developing and mature neurons. The reason is the lack of suitable labeling approaches that would allow studying nanoscale organization and
dynamics of the NF186 and NaV1.6 in their native environment with advanced super-resolution(SRM) and live-cell microscopy. The current SRM and live-cell microscopy-compatible labeling approaches mainly involve labeling with antibodies conjugated to fluorescent dyes or genetically encoded fluorescent proteins (FPs). However, fluorescent antibodies are large (100–200 kDa) and can introduce artefacts in live-cell and SRM studies. Also, in most cases, they cannot be used for live labeling of transmembrane components and lack specificity when labeling closely related protein isoforms, such as NaV. Genetically encoded FPs are smaller (~30 kDa) and provide specificity. However, they can be added only to N or C termini, potentially affecting the trafficking or localization of target proteins.
A combination of the genetic code expansion (GCE) and bioorthogonal click chemistry
reactions (referred to as click labeling) has emerged as a promising tool for site-specific
noninvasive fluorescent protein labeling in mammalian cells. GCE technology utilizes orthogonal translation machinery to site-specifically install unnatural amino acid (UAA) into target proteins. Subsequently, the incorporated UAA is labeled with a fluorescent dye via a bioorthogonal click reaction. Contrary to the bulky fluorescent antibodies and N- and C-terminal FP-fusions, the UAA-based tags are small (0.2–0.5 kDa) and can be introduced at any position into a protein of interest. Therefore, UAAs represent an ideal alternative for labeling of the AIS components,whose localization and function can be affected by the size of the labeling tags. Furthermore, the UAA can be labeled with various cell-permeable and cell-impermeable fluorescent dyes compatible with live-cell and SRM microscopy techniques.
The main aim of my Ph.D. was to establish highly efficient UAA-based click labeling of the
axon initial segment and node of Ranvier components, NaV1.6 and NF186, in living neurons that would allow quantitative, live-cell, and SRM imaging studies. By designing different plasmids encoding NF186, NaV1.6, and orthogonal translational machinery and by optimizing conventional transfection and labeling conditions, I established click labeling of NF186 and NaV1.6 in primary neurons without affecting their localization and function. This was the first time that the large, complex, spatially confined proteins, including ion channels, were labeled in living primary neurons with click labeling approach. In recent years, click labeling has emerged as a promising protein labeling tool. However, due to the low to moderate UAA incorporation efficiency in mammalian cells and the complexity of the method, click labeling has not been widely used, especially in neurons. Achieving higher efficiency of the UAA incorporation in a larger number of neurons compared to conventional transfection would allow quantitative, live-cell, and SRM imaging studies of neuronal proteins. To increase click labeling efficiency of the large AIS component, NaV1.6, I probed different viral based systems for the efficient delivery of the GCE components, such as baculoviruses and adeno-associated viruses (AAVs). The usage of the AAV-based vectors that bore GCE elements led to a threefold increase in GCE efficiency and expression of NaV1.6 in neurons compared to conventional transfection.
Finally, in addition to establishing a highly efficient click labeling of the NaV1.6 and NF186,
I demonstrated that this approach could be used for live-cell and direct stochastic optical
reconstruction microscopy (dSTORM) super-resolution imaging of the AIS. Furthermore, I showed that click labeling could be used for quantitative studies of the localization of pathogenic epilepsy-causing NaV1.6 variants. In order to optimize dSTORM imaging and allow reproducible SRM imaging of the AIS, I tested different imaging buffers, including a commonly used mounting medium Vectashield. This led to the discovery that the fluorescence of AF647 and its variant AF647 Plus are quenched in Vectashield without the effect on dSTORM imaging. These observations are important for both SRM and conventional microscopy since Vectashield is a commonly used medium for tissue mounting. Hence, these results will allow other laboratories to optimize their imaging conditions and produce reliable results.
In conclusion, I developed a new approach for the highly efficient labeling of neuronal
proteins, including large and complex ion channels. This new approach is compatible with live-cell and SRM imaging. Therefore, it will allow gaining novel insight into the regulation of NF186 and NaV1.6 in health and diseases at the single molecule level in their native environment. This approach will also allow studying additional pathogenic NaV1.6 variants and other NaV isoforms for which suitable labeling approaches do not exist. Furthermore, AAV-based delivery will allowefficient labeling of other complex proteins and ion channels in living and fixed neurons, as well as in more complex model systems, such as brain slices, organoids, and whole animals. Therefore, my Ph.D. work provides the basis for novel discoveries in the field of AIS and neurobiology.
microscopy
Print-on-Demand