Date of Award

1992

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College

College of Graduate Studies

First Advisor

Rosalie K. Crouch

Second Advisor

Daniel R. Knapp

Third Advisor

Perry V. Halushka

Fourth Advisor

Donald Menick

Fifth Advisor

Wes Corson

Sixth Advisor

Kevin J. Schey

Abstract

The objective of these studies was to further define the amino acids of bacteriorhodopsin in close proximity to the ring end of retinal by labeling them using a retinal analog modified to contain a photoactivatable group. The sites of labeling were to be identified using mass spectrometry and tandem mass spectrometry. Eight potentially photoactivatable retinal analogs were characterized and thoroughly examined as bacteriorhodopsin analog pigments thus identifying those analogs that were most similar to the native retinal. A versatile flash photolysis apparatus was compared to a continuous wave light source on the basis of total photon output and on efficiency of activation of a model photoactivatable compound. The flash apparatus activated the model compound four times more efficiently than the continuous wave source. The specially designed and constructed sample cell was efficient for photolysis studies. Molecular modeling of the retinal analogs and bacteriorhodopsin allowed enhanced visualization of the structures and predictions of potentially labeled amino acids. The hypothesis is that a photoactivatable retinal analog would label the amino acids predicted by these studies. The previously published reports of labeling of bacteriorhodopsin were evaluated suggesting that only one** of the retinal analogs was in the retinal binding site. Mass spectrometric analysis of cleavage fragments of bacteriorhodopsin identified 49% of the protein. Nine percent of the protein was sequenced by tandem mass spectrometry. An unusual protein cleavage site was identified and confirmed by tandem mass spectrometry. The identified peptides were the smallest of the predicted peptides from the cleavages. These initial studies indicated the utility of mass spectrometry in identification of labeled amino acids. Analysis of the three available tritiated, photoactivatable retinal analogs that formed acceptable bacteriorhodopsin analog pigments by two different techniques showed no significant labeling of bacteriorhodopsin. Parallel positive control experiments estimated the limit of detection of labeling at <5%. Several modifications are suggested that should increase the likelihood of success for photo affinity labeling of bacteriorhodopsin with subsequent analysis by mass spectrometry.

Rights

All rights reserved. Copyright is held by the author.

Download

Share

COinS