The chemiosmotic hypothesis, formulated by Mitchell (1961), ascribes a central role in energy transduction to a transmembrane electrochemical gradient of protons. One way to test the validity of this hypothesis is to measure the energetic content of the proton gradient and compare it with that of the reactions generating (or consuming) the gradient, Such a comparison will lead to an estimate of the maximum (or minimum) number of protons involved in the transduction process (Weichmann et al., 1975). In rat liver mitochondria, we calculated the APH+ (electrochemical proton-activity difference) across the inner membrane from the distribution of a weak acid 5,5-dimethyl-2,4-oxazoIidedione and of K+ in the presence of valinomycin by using the silicone-centrifugation method, under different conditions (Wiechmann et al., 1975; cf. Padan & Rottenberg, 1973; Nicholls, 1974). In the same experiments the energetic content of the ATPase (adenosine triphosphatase) reaction was calculated from the measured concentrations of ATP, ADP and phosphate by using the AGO values given by Rosing & Slater (1972). Table 1 lists the results of a number of such experiments. The ratio of AGp/A&+ (where AGp is phosphate potential), indicating the minimum number of protons that is energetically required to synthesize one molecule of ATP, is greater than two under all conditions tested. This ‘apparent stoicheiometry’ is independent of the substrate used and of the pH between 6 and 7.5. If the mitochondria are uncoupled, the appparent stoicheiometry increases. This could be due to an increasing underestimation of Aj*+ with increasing degree of uncoupling of the system; otherwise, this variability is difficult to reconcile with a mechanism in which a fixed number of protons passes through the mitochondria1 ATPase complex to synthesize one molecule of ATP. The extramitochondrial AGp is greater than the intramitochondrial AGp, suggesting that there is an extra input of energy in the transport of adenine nucleotides across the membrane (cf. Klingenberg, 1972). In our experiments this extra energy input is equivalent to about four H+ ions per ATP. Even if we take the intramitochondrial pool of adenine nucleotides as the primary acceptor of the energy contained in A/iH+, the apparent stoicheiometry always exceeds the value of two. If ATP hydrolysis (in the absence of oxidative substrate) is used to generate ApH+, the results of Expt. V in Table 1 are found. In this case the ratio AGp/AJ,+ indicates the maximum number of protons that can be transported across the membrane during the hydrolysis of one molecule of ATP. Interestingly, this number is similar to the numbers found in the other experiments, suggesting a situation close to reversibility on the ATPase complex with approximately three protons transversing the membrane per hydrolysis or synthesis of one ATP molecule. Whereas in mitochondria the generator for A&+ is electron transport in Halobacterium halobium the chromoprotein bacteriorhodopsin is thought to function as a light-driven proton pump (Oesterhelt & Stoeckenius, 1973). Although one should expect that in membranes containing bacteriorhodopsin the proton-translocating activity would be rapidly limited by a membrane potential, little effect of valinomycin plus K+ could be demonstrated. However, we were able to show that in an artificial membrane system, containing both bacteriorhodopsin and ox heart cytochrome c oxidase, the oxidation of ascorbate by oxygen can be stimulated by light (Hellingwerf et al., 1976). In such a system the cytochrome c oxidase catalyses a transmembrane electron transport, leading to the development of a transmembrane proton gradient that is actively dissipated by bacteriorhodopsin in the light. The experiments above are compatible with a chemiosmotic mechanism of energy transductions with a central role for protons in the transduction process. At the same time
Read full abstract