(1) Glycoproteins and development in the cellular slime mold Dictyostelium discoideum. (2) Separation of cells using isopycnic centrifugation in linear density gradients of colloidal silica

Citation

West, Christopher Mark (1978) (1) Glycoproteins and development in the cellular slime mold Dictyostelium discoideum. (2) Separation of cells using isopycnic centrifugation in linear density gradients of colloidal silica. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/NM0K-3F42. https://resolver.caltech.edu/CaltechETD:etd-08082006-082425

Abstract

I: Two methods were adapted for the study of glycoproteins in Dictyostelium discoideum. Both techniques relied upon prior separation of glycoproteins on one- or two-dimensional SDS-polyacrylamide gels. The first method was a modified form of crossed immunoelectrophoresis which substituted the carbohydrate-binding protein (lectin) concanavalin A (Con A) as the precipitating agent. In the second method, polyacrylamide gels were simply fixed and incubated in the presence of fluorescein-tagged lectins. After washing away free lectin, the gels were photographed and stained for protein. Identified glycoproteins were defined by their apparent molecular weights, identity and, anomeric linkage of some of their monosaccharides and in two-dimensional gels, their apparent isoelectric points. The ability of these techniques to specifically identify authentic glycoproteins was confirmed using a defined membrane system, the erythrocyte ghost, other known proteins, and hapten inhibitors. Of the two techniques, lectin diffusion was capable of higher resolution but did not give information about receptor multivalency. Both methods were more sensitive than the commonly-used periodic acid-schiff's base stain.

More than fifty different glycoproteins were detected in vegetative cells on the basis of their labeling with Con A or wheat germ agglutinin (WGA) and their sensitivity to proteolysis. These glycoproteins were distributed throughout the cell, giving each of several subfractions, including the plasma membrane, its own profile of glycoproteins. WGA receptors were apparently membrane bound and predominantly localized in the plasma membrane. A small number of glycoproteins were also detected with a galactose-binding protein, but these glycoproteins were not present in the plasma membrane or on the cell surface as determined by an independent technique. Receptors for L-fucose binding proteins were also absent from the plasma membrane of these cells. In contrast, another eukaryotic cell plasma membrane, the erythrocyte ghost, contained receptors to all lectins tested.

During the formation of the pseudoplasmodium from vegetative cells, many glycoproteins were lost, modified, or many new ones were synthesized in all cell subfractions, including the plasma membrane. In addition, there was an asymmetry of distribution of glycoproteins in the pseudoplasmodium. There were three prestalk-cell specific glycoconjugates, which were all restricted to the plasma membrane, and two prespore-specific glycoproteins, which were present in plasma membranes as well as in a potential plasma membrane precursor. These plasma membrane differences occurred prior to overt differentiation of any stalk or spore cells. The region-specific glycoproteins were also present in the plasma membranes of cells which had not yet formed pseudoplasmodia, but were absent from vegetative cells.

In order to reinforce the temporal evidence that these region-specific glycoproteins were involved in the creation of the pseudoplasmodium, dissociated pseudoplasmodial cells were treated with the lectin which originally identified the region-specific molecules (WGA). Although dissociated cells typically reformed pseudoplasmodia, in the presence of WGA, they did not. This effect of WGA was blocked by a hapten inhibitor.

II: In other work, separation of cells with different densities was attempted. When cells of Dictyostelium discoideum were centrifuged to density equilibrium in linear gradients of colloidal silica (Ludox), approximately 40 discrete bands appeared. A similar result was found for formalinized red blood cells and plastic beads. Isolated bands of cells rebanded faithfully in new gradients and band spacing depended upon gradient steepness. It was found that cell bands resulted from microscopic discontinuities in the linear gradients caused by centrifugation. When the gradients were analyzed in the analytical ultracentrifuge, absorbance scans revealed that cell bands coincided with "bands" of Ludox, which formed even without cells. Evidence ruling out other possible causes for cell bands is presented and procedures which avoid this condition are described.

Thesis Files