A variety of experimental evidence indicates that the oxidation of fatty acids in mitochondria is closely related to the process of oxidative phosphorylation. Three main points may be considered. First, the same mitochondrial preparations which carry out fatty acid oxidation also catalyze phosphorylation coupled to oxidations of the Krebs cycle (27). Thus enzymes of oxidative phosphorylation are always present in mitochondrial preparations during fatty acid oxidation, and both system show roughly parallel degrees of lability toward relatively mild manipulations. Secondly, the requirement of the fatty acid oxidation system for both inorganic phosphate and adenine nucleotide strongly suggests the participation of a phos-phorylating mechanism. Finally, the fatty acid oxidation system in mitochondria has been shown by workers in Green's laboratory to be strikingly sensitive to uncoupling agents such as 2,4-dinitrophenol and gramicidin, which are known to prevent the uptake of inorganic phosphate into ATP linkages, while allowing most oxidative reactions to proceed unimpaired. Every effective uncoupling agent so far tested has proved to be a potent inhibitor of fatty acid oxidation in mitochondria.

Activation by oxidation of DPNH. In most experiments in which the activation of fatty acid oxidation has been studied, intermediates of the Krebs cycle have been used as priming substrates. It is not possible to conclude from these experiments whether it is the functioning of the Krebs cycle itself, or merely the accompanying oxidative phosphorylation which is necessary for the activating effect. Experiments which decide this point have been carried out using chemically reduced DPNH as primer (29). Work by Friedkin and Lehninger (15) had demonstrated that the oxidation of chemically reduced DPNH in washed, cell-free liver preparations was coupled to phosphorylation of ATP, and Lehninger and Smith (43,41) later established a P/O ratio of 3.0 for this oxidation. When small amounts of DPNH were tested as priming co-substrate an efficient activation of fatty acid oxidation was noted, in the complete absence of added Krebs cycle intermediates. The priming effect was abolished by the addition of dinitrophenol, and was not shown by comparable amounts of DPN in the oxidized form. A striking increase in the yield of acetoacetate from the fatty acid was noted with DPNH as the primer, since no oxalacetate precursors were added in these experiments. Palmitic acid, for example, when tested in the presence of low concentrations of succinate showed little tendency to form acetoacetate, being preferentially oxidized to carbon dioxide and water. When the succinate was replaced by DPNH as the primer, then the yields of acetoacetate from palmitate approached the theoretical, based on oxygen uptake. Similar results were noted with octanoate, although in the latter case there was a very considerable accumulation of acetoacetate even in the presence of succinate.

It is of interest that, once initiated, fatty acid oxidation may be " self-priming." In the presence of excess adenylic acid and DPNH as primer, fatty acid oxidation continues long after the DPNH has been completely oxidized. This observation is in accord with the finding that the oxidation of fatty acids itself brings about esterifi-cation of inorganic phosphate, as demonstrated (42) in experiments with P32. The maximum P/O ratio is not easily measured because fluoride, usually employed to inhibit phosphatase action in such tests, is a very potent inhibitor of fatty acid oxidation. However, Johnson and Lardy (23) have observed P/O ratios as high as 1.7 in the absence of fluoride ion.

Nature of activating reaction. On the basis of the significant correlation between fatty acid oxidation and oxidative phosphorylation it was at first concluded (35, 38) that the priming reaction was necessary only to generate ATP by phosphorylations coupled to the oxidation of the added Krebs cycle intermediates. However, it developed in later studies (30, 29) that both ATP and small amounts of co-substrate were needed. The presence of ATP may be necessary but not of itself sufficient for the activation of fatty acid oxidation in mitochondria. In contrast to these findings there is now evidence that this is not the case when fatty acid oxidation is studied in soluble extracts. 2,4-dinitrophenol has no inhibitory effect on oxidation of fatty acids in extracts of C. kluyveri (24), and the recent report of Drysdale (14) on the oxidation of fatty acids in soluble extracts of acetone powders of mitochondria also states that dinitrophenol is not inhibitory in this system. It is nevertheless important to note that in both these cases the presence of a high-energy activator may be demonstrated to be necessary, acetyl phosphate in the case of the C. kluyveri system (56) or adenine nucleotide in Drysdale's experiments. It is possible that continuing oxidative phosphorylation is needed to maintain the structural integrity of isolated mitochondria, even in the presence of ATP, and that fatty acid oxidation is in some secondary way dependent upon the intact structure of the particle.

To account for the experimental findings obtained in studies of fatty acid oxidation and synthesis in C. kluyveri, Kennedy and Barker (25) suggested that the first step in fatty acid oxidation in the bacterial enzyme system consists in a reaction between the fatty acid and coenzyme A. On the assumption that the oxidation of fatty acids in the bacterial extracts and in animal tissues is fundamentally similar, and that the activation in the latter case is at the expense of ATP, this reaction may be tentatively formulated in the following equation:

It would seem reasonable to suppose that this reaction is completely analogous to the activation of acetate by ATP to form acetyl-CoA, in which case an intermediate formation of a CoA pyrophosphate may be involved, as recently shown for the ATP-acetate reaction by Lipmann and his collaborators (45). At the time that this activating reaction was originally suggested, no direct evidence for the formation of long-chain fatty acid complexes with CoA was at hand. Very recent work by Kornberg (31) and by Drysdale (14) has now demonstrated the presence of enzymes in mammalian liver tissue which catalyze the formation of hydroxamic acids of long-chain fatty acids, when tested in the presence of CoA, ATP, and hydroxylamine, thus providing important evidence for activation of long-chain fatty acids by a mechanism essentially similar to the ATP + acetate reaction. As Lynen, Reichert, and Rueff (48) have pointed out, the activation of fatty acid molecules by such a reaction with CoA may also be expected to be required for triglyceride and phospholipid synthesis, as well as for fatty acid oxidation.