Hemoglobin and the origins of the concept of allosterism.

Bohr, Hasselbalch, and Krogh (1904) observed both what we now call the cooperative homotropic character of the binding of oxygen by hemoglobin and the heterotropic control exerted by CO2 in diminishing the oxygen affinity. Ten years later Christiansen, Douglas, and Haldane discovered the converse effect of oxygenation in diminishing CO2 uptake. It was then generally believed that hemoglobin contains only a single heme: A. V. Hill, to explain cooperative phenomena, postulated reversible aggregation of these monomer units (1910). After 1924, Adair and Svedberg independently showed that the molecule contained four hemes, and Adair's intermediate compound hypothesis, with four binding constants suitably chosen, could formally explain cooperative binding. Pauling proposed a simple model, involving only two constants, that fitted available data well. Haurowitz's demonstration that crystal structure changed on oxygenation (1938) gave the first evidence clearly pointing to a conformation change; in 1951 Wyman and Allen elaborated the idea in thermodynamic terms, and Perutz's crystallographic studies later revealed in molecular detail the nature of the change associated with ligand binding. The important heterotropic interactions that influence the binding of oxygen, necessarily with reciprocal interactions between oxygen binding and the uptake of the heterotropic ligands, are of three kinds: 1) proton binding by the "Bohr groups," 2) direct binding of CO2 as carbamate, and 3) binding of organic phosphate anions, such as diphosphoglycerate. The last of these, although fully as important as the first two, was not discovered for about half a century after the early work. Some major discoverers in the unraveling of these complicated relations were D. D. Van Slyke, F. J. W. Roughton, Linus Pauling, J. Wyman, and later Ruth and Reinhold Benesch, L. Rossi-Bernardi, and J. V. Kilmartin. All these, and numerous others, contributed to our understanding of both homogropic and heterotropic interactions. Brief final comments relate to the evolution of the concept of reversible conformational transitions as the basis for both homotropic and heterotropic interactions in allosteric proteins.