The continuing life of the Enzyme Club of New York City: the growth of American biochemistry from 1942 to 1982

The Heuristic of Form: Mitochondrial Morphology and the Explanation of Oxidative Phosphorylation

In the 1950s and 1960s, the search for the mechanism of oxidative phosphorylation by biochemists paralleled the description of mitochondrial form by George Palade and Fritiof Sjöstrand using electron microscopy. This paper explores the extent to which biochemists studying oxidative phosphorylation took mitochondrial form into account in the formulation of hypotheses, design of experiments, and interpretation of results. By examining experimental approaches employed by the biochemists studying oxidative phosphorylation, and their interactions with Palade, I suggest that use of mitochondrial form as a guide to experimentation and interpretation varied considerably among investigators. Most notably, Peter Mitchell, whose chemiosmotic hypothesis was ultimately the basis of the correct mechanism of oxidative phosphorylation, incorporated crucial aspects of mitochondrial form into his model that others failed to recognize. I discuss these historical observations in terms of the background and training of the biochemists, as well as a proposed heuristic of form, whose use may increase the possibility that biologically meaningful molecular mechanisms will be discovered.

Three proton pumps, morphology and movements

The diameter of F1 coupling factor and the distance it protrudes from the membrane of bovine heart submitochondrial particles were measured quantitatively using horse spleen ferritin as a standard. Employing the freeze-etch technique, particles of similar size were found on membranes of submitochondrial particles and on membranes of particles first depleted by F1, then reconstituted by addition of F1. The extramembranous size of F1 is 9.7 nm and F1 protrudes from the membrane surface by about 13.6 nm. Bacteriorhodopsin and cytochrome oxidase were incorporated into lipids derived from membranes of extremely thermoacidophilic microorganisms by the octylglucoside dilution method. The bacteriorhodopsin pump was fully functional provided high concentrations of valinomycin were added. With decanoyl-N-methylglucamide as detergent the pump was very active in the absence of valinomycin. Concentrations of gramicidin that collapsed the delta pH in bacteriorhodopsin liposomes prepared with soybean phospholipid had little or no effect on these rigid proteoliposomes. Very high concentrations (30 micrograms per ml) were partially effective, suggesting a mechanism other than formation of a gramicidin dimer channel. Cytochrome oxidase lost virtually all activity when incorporated into these rigid liposomes but was fully reactivated on addition of suitable detergents.