Brain Metabolism, “Ferrin,” and Non‐Carbohydrate Oxidation

B. Libet

DOI: https://doi.org/10.1096/fasebj.7.6.8472897

1993-04-01

The FASEB Journal

Abstract:K.A.C. Elliott and I submitted a paper (1) to the 1943 Federation Meeting in Cleveland, subsequently cancelled because of wartime exigencies. It was one of a series of studies of brain metabolism by Elliott, Libet, and McNair Scott in the early b940s. I was then a young postdoctoral physiologist with Elliott at the Institute of the Pennsylvania Hospital, before he went to McGill University to spend the next 30-44) years of his distinguished career as a neurochemist. After some war-related research on clothing and survival problems I resumed (in 1945) my primary interests in neuroand electrophysiology (begun earlier with R. W. Gerard); but my time with Elliott gave me invaluable experience in neurochemistry and in rigorous, carefully controlled experimentation. We had demonstrated that homogenized suspensions of rat brain in isotonic media respired at rates comparable to those of brain slices (2). That led to the use of such suspensions for quantitative studies of the oxidation of endogenous and added substrates by brain tissue (3). Among other things, additon of glucose or pyruvate largely suppressed the oxidation of endogenous non-carbohydrate substrates. In theabsence of added glucose, noncarbohydrate oxidation supported respiratory rates at about 50% those with glucose; even these rates fell off sharply after the first 1-2 hours, presumably as available substrate was depleted or as some inhibitory end product accumulated. Added glycogen also stimulated brain respiration, but later on we found that more purified glycogen did not do this and pointed to an impurity asthe agent. That impurity turned out to be iron,in a form bound to a protein obtainable from liver(4).We named it “ferrin.” In a partially purified form it contained b5 % of iron soluble at neutral or even alkaline pH. Ferrin differed from femtin (5) in its absence from horse spleen (the richest source of ferritin) and in certain solubility characteristics. Ido not know whether a physiological role for ferrin was ever established, but it seemed to have potential interest asa form of soluble liver iron and even as a contaminant in certain liver products (as we found with glycogen). The substrates and catalysts in noncarbohydrate oxidation by brain tissue were of considerable interest, especially for events during hypoglycemic shock and seizures induced by insulin. Panimon et al. (6) had reported that iron salts increased the rateof oxygen uptake by brain and other tissues, probably via the oxidation of phospholipids and proteins. We then found that the combined addition of very small amounts of iron and ascorbic acid (b mM) produced a large increase in oxygen uptake by isotonic brain (or liver) suspensions (b, 7). This was accompanied by no change in CO2 evolution. With purified mixed phospholipid from brain or liver, ascorbic acid plus iron compounds produced a very rapid oxidation. Only the cephalic fraction of brain phospholipd was oxidized. Ascorbate or iron alone were only weakly active; other reducing agents could not replace ascorbate; inorganic iron salts were as effective as ferrim, but the latter had the advantage of being soluble at body pH values. Epinephrine inhibited the effectsof iron and ascorbic acid, while certain amino acids facilitated them. All this led to the speculative possibility that iron-protein compounds together with ascorbate may have a role in normal phospholipid metabolism and could thereby affect membrane permeability characteristics. The additional finding, that the oxygen uptake by /ypotonic brain suspensions was greatly stimulated by iron plus ascorbate so as to rise to values almost equal to that of isotonic suspensions, seemed to support the notion of membrane-phospholipid oxidation. That is, disruption of membranes of the cells and their subcellular structures by the hypotonic medium would make such phospholipid-contasning structures more accessible to the iron-ascorbate action. It was also interesting that this action was heat-resistant and did not require the enzymatic catalyses that mediate carbohydrate oxidation (see also ref 6). If