Whatever you do in one part of the molecule affects the molecule as a whole." M. F. Perutz (9 1)
INTRODUCTIONThe control of oxygen binding by hemoglobin (Hb) with pH is of profound importance in facilitating gas exchange in blood. This modulation, known as the Bohr effect, reflects the fact that protons are released upon Hb oxygena tion at physiological pH (the alkaline Bohr effect), and protons are taken up upon oxygenation at low pH (the acid or reversed Bohr effect). Reciprocally, changes in pH modulate oxygen affinity. These proton exchanges arise because conformational changes in the Hb associated with ligand binding at the heme result in pK changes in certain acid groups that are distant from the heme. Changes in proton binding also result from differential interaction of buffer ions with oxy-and deoxy-Hb (3,18). The binding of salt ions can be considered a special kind of Bohr effect (3). Resonance Raman spectroscopy shows that the iron-proximal histidine stretching motion is exquisitely sensi tive to amino acid substitutions distant from the heme (34). This review examines recent experiments to determine which groups are responsible for the Bohr effect and how these ligand-linked processes are modulated by other allosteric factors, e.g. buffer ions, organic phosphates, CO2, and chloride, all of which lower the oxygen affinity of Hb by preferential binding to deoxy-Hb. A central problem has been to determine the relative quantitative roles of different ionizable groups responsible for in the Bohr effect. What kinds and numbers of groups are involved? Attempts have been made to calculate pK values by using electrostatic theory. How do such electrostatic calculations compare with other methods for estimating the contributions of different groups? 181 0066-4278/88/0315-0181$02.00Annu. Rev. Physiol. 1988.50:181-204. Downloaded from www.annualreviews.org by North Dakota INBRE on 09/15/13. For personal use only. Quick links to online content Further ANNUAL REVIEWS 182 RIGGS Perutz (89) proposed a stereochemical model that provided a molecular explanation of both the Bohr effect and of cooperativity of oxygen binding. The model rested initially on differences between the x-ray structures de termined for horse deoxy-and oxy-like metHb. The model has been modified and refined as a result of many subsequent experiments and higher resolution x-ray analyses on human deoxy-Hb, HbCO, and Hb02 (8, 33,96,112). Although the original proposal depended on the structure of metHb, which was found to be isomorphous with Hb02, some differences between the met, CO, and oxy derivatives do exist (33,112). This is shown most simply by the pronounced differences in solubility: human HbCO is over 40% more soluble than Hb02 in phosphate buffers at 10°C and 67% more soluble than metHb; deoxy-Hb is only one tenth as soluble as HbCO (50). The differences in surface topology responsible for these solubility effects are unknown. How ever, the distal histidine of the a subunit is much closer to the heme axis in Hb02 than in HbCO (112). Presu...