Investigation of the fusion machinery reconstituted into novel pore-spanning membranes
by Merve Sari
Date of Examination:2024-05-15
Date of issue:2024-09-06
Advisor:Prof. Dr. Claudia Steinem
Referee:Prof. Dr. Claudia Steinem
Referee:Prof. Dr. Michael Meinecke
Files in this item
Name:Dissertation_Sari_Merve.pdf
Size:9.93Mb
Format:PDF
This file will be freely accessible after 2025-05-14.
Abstract
English
The release of neurotransmitter into the synaptic cleft is achieved by fusion of synaptic vesicles with the presynaptic membrane. This essential process in the neuronal signal transmission is driven by the soluble N-ethylmaleimide-sensitive-factor attachment receptor (SNARE) proteins to overcome the repulsive forces of the membranes. To analyze the fusion process in vitro, pore-spanning membranes (PSMs) are an appropriate model system to mimic the presynaptic membrane and incorporate the target-SNAREs. PSMs have a mobile freestanding part with an underlying compartment for content release. Therefore, it is crucial to achieve a high protein reconstitution efficiency (here ΔN49 complex) in giant unilamellar vesicles (GUVs) used as precursors of PSMs. Two different methods were used to reconstitute the ΔN49 complex in this work: lipid-based microfluidics and electroformation. In lipid-based microfluidics, droplet-stabilized GUVs (dsGUVs) are formed within a surfactant shell. These dsGUVs are then destabilized by a de-emulsifier and released into a physiological surrounding. In this study, three fusion-related proteins varying in structural complexity (ΔN49 complex, synaptobrevin 2, and the transmembrane domain of syntaxin 1A (syx 1A-TMD)) were successfully reconstituted in dsGUV, as demonstrated by lipid and protein mobility determined by fluorescence recovery after photobleaching (FRAP) measurements. The results showed that the ΔN49 complex and syx 1A-TMD had a heterogeneous distribution in the membrane, with a reconstitution efficiency of ~50% for both. This thesis suggests that during the release process, the surfactant layer and de-emulsifier electrostatically interact with lipid and protein molecules leading to lipid and protein leakage. The syx 1A-TMD containing GUVs were successfully released and homogeneously distributed in the lipid membrane due to the absence of an extracellular part to interact with. Confocal microscopy detected the successful formation of PSMs containing Syx 1A-TMD. Thus, a novel reconstitution process was established that starts with proteoliposomes and ends with proteo-PSMs under physiological conditions. In a second context, reliable GUV spreading was established on porous substrates with closed cavities, forming osmolarity-dependent curved PSMs. However, a tight membrane arrangement of the PSMs was demonstrated due to dye entrapment in the underlying compartment. The PSMs were doped with a labeled ΔN49 complex and the mobility of lipid and protein was characterized. Single vesicle fusion assays were performed by adding content-labeled (sulforhodamine B) synaptobrevin 2-doped LUVs to these partially curved PSMs to gain further insight into the different stages of the fusion pathway. First, the diffusion behavior of docked vesicles on PSMs was investigated. The simultaneous observation of lipid mixing and content release allowed the identification of flickering fusion pore opening and closing. This allowed conclusions to be drawn about the different stability and regression kinetics of the Ω-shaped post-fusion structure. Finally, the influence of curved PSMs on the release kinetics was determined.
Keywords: PSMs; GUVs; SNARE proteins; Microfluidics; Membrane fusion