Studying Protein Organization in Cellular Membranes by High-Resolution Microscopy

by Sinem Saka Kırlı
Doctoral thesis
Date of Examination:2013-10-29
Date of issue:2014-10-09
Advisor:Prof. Dr. Silvio Rizzoli
Referee:Prof. Dr. Michael Kessel
Referee:Prof. Dr. Mikael Simons
Referee:Prof. Dr. Stefan Hell
Referee:Prof. Dr. Detlef Doenecke
Referee:Prof. Dr. Michael Thumm
Persistent Address: http://dx.doi.org/10.53846/goediss-4715

 

 

Files in this item

Name:Sinem Saka Kirli_Dissertation.pdf
Size:6.65Mb
Format:PDF
Description:Dissertation

The following license files are associated with this item:

Abstract

English

Most membrane proteins are found in clusters in the plasma membrane. For abundant proteins it is apparent that the clusters also tend to have patterned distributions, rather than being randomly scattered. To reveal the mechanism behind this observation, I have investigated all proteins in the plasma membrane simultaneously. This was achieved by large-scale metabolic labeling of proteins through incorporation of a non-canonical amino acid analogue and fluorescent tagging by click chemistry. Plasma membrane sheets were obtained from labeled PC12 and COS-7 cells and imaged by super-resolution stimulated emission depletion (STED) microscopy. Membrane proteins were found to form a mosaic like pattern. Heterogeneous domains, highly enriched in proteins, were distributed on a protein-poor background. I have termed these high-abundance domains “protein clouds” and assayed the contribution of different factors to their formation and maintenance. The protein cloud pattern was surprisingly robust and resisted to a variety of manipulations including changes in ionic composition and strength, decrease in protein density, disruption of cytoskeletal elements, and hydrolysis of phospholipids and sphingomyelin. Although actin disruption did not abolish the clouds, it resulted in formation of larger domains. Accordingly, actin was identified as a secondary factor that borders the clouds and prevents their coalescence. Depletion of cholesterol almost completely eliminated the protein clouds, in reversible fashion, suggesting that cholesterol is the major factor for the cloud patterning.  The cloud pattern was confirmed by two additional, independent techniques: first, label-free secondary ion mass spectrometry (SIMS) and second, STED fluorescence correlation spectroscopy (STED-FCS). For reliable application of SIMS, a novel experimental scheme, named correlated optical isotopic nanoscopy (COIN), was developed and verified.  To find out the relevance of clouds to specific proteins, distributions of different classes of proteins were investigated. All of the specific proteins analyzed were enriched in the protein clouds, but displayed differential enrichment profiles. Many proteins were preferentially located in particular areas, such as the edges or centers of the clouds. Some functional partners were observed to show similar profiles.  I conclude that enrichment of proteins in the clouds can act as a basic, low-hierarchy principle of membrane patterning, underlying the distributions of specific proteins.
Keywords: Protein Domains; Nanodomains; Multi-Protein Assemblies; Protein Clouds; Membrane Organization; Correlated Optical Isotopic Nanoscopy; COIN; Secondary Isotope Mass Spectrometry
 
Statistik