Erwin Ilegems
In this thesis, two different fluorescent labeling techniques for in vivo investigations on the 5-HT3 receptor (5-HT3R) functions are presented. This plasma membrane protein contains five subunits surrounding an ion channel that opens after binding of a 5-HT3-specific neurotransmitter. The first technique, described in the first part of this report, focuses on receptor labeling via genetic fusion to spectral variants of the green fluorescent protein. I found that the resulting chimeras containing one fluorescent protein per subunit exhibit preserved ligand binding and channel activity, opening a wide range of biological research applications. Among these, I present the possibility to follow the 5-HT3R trafficking and localization during its entire life cycle by multicolor imaging in live cells, starting with its cytoplasmic biogenesis and ending with its ligand-induced internalization. The intracellular subunit assembly is shown to occur in the endoplasmic reticulum, and the importance of the cytoskeleton microtubules for proper membrane targeting is unraveled. The utility of bioluminescent 5-HT3R contructs was also demonstrated in another approach using the chemical disruption of cellular actin filaments to produce vesicular fractions of cells, in the order of 0.1 to a few micrometers in diameter. These so-called native vesicles, containing the labeled receptors in their membrane, were shown to be suitable for measurements using fluorescence confocal microscopy of ligand binding and of ion influx, opening new possibilities for miniaturized bioanalytics. In a third approach, I demonstrated that after detergent-solubilization of the receptor, the green fluorescent protein (GFP) inserted into the receptor sequence permitted to monitor ligand binding via fluorescence resonance energy transfer (FRET). Furthermore, I could observe a spatial reorientation of the receptor GFPs upon binding an agonist to the receptor. In the second part of this thesis, I adapted the mis-acylated suppressor tRNA technology to mammalian cells, permitting the introduction of unnatural amino acids at specific positions in the protein of interest. I achieved an efficiency of amino acid incorporation using in vitro aminoacylated suppressor tRNA in the order of 15% in CHO cells. A novel methodology for the quantification of background natural nonsense codon readthrough in different cell lines was also developed, permitting to select the most suitable codon-anticodon pair for this suppressor tRNA technique in various cell lines. Finally, I present a general strategy to increase the aforementioned artificial incorporation efficiency by down-regulating the competing eukaryotic release factor 1 (eRF1) using small interfering RNAs.
EPFL2004
