BRAIN M I T O C H O N D R I A II. The Relationship of Brain Mitochondria to Glyeolysis

A mitochondrial fraction prepared from calf brain cortex possessed negligible glycolytic activity in the absence of the enzymes of the high speed supernatant fraction. When mitochondria were added to a supernatant system supplemented with optimal amounts of crystalline hexokinase, a 20 per cent stimulation of glycolysis was observed. The supernatant fraction produced minimal amounts of lactate in the absence of exogenous hexokinase; the addition of mitochondria doubled the lactate production. The substitution of glycolytic intermediates for glucose as substrates as well as the addition of exogenous glycolytic enzymes to the supernatant fraction or supernatant fraction plus mitochondria indicated that the mitochondria contributed mainly hexokinase and phosphofructokinase. By direct assay of all of the enzymes of the glycolytic pathway, only hexokinase and phosphofructokinase were shown to be concentrated in the mitochondrial fraction. All other glycolytic enzymes were found to exhibit higher total and specific activities in the supernatant fraction. Although it is generally agreed that brain mitochondria, like mitochondria from other tissues, carry out citric acid cycle oxidations and oxidative phosphorylation, there are varying reports in the literature concerning the relationship of the enzymes of glycolysis to the mitochondrial fraction of brain. Hesselbach and Du Buy (1, 2) showed that brain mitochondria convert glucose and glycolytic intermediates to lactate both aerobically and anaerobically. Similarly, Gallagher et al. (3) demonstrated the complete oxidation of glucose and glucose-6-phosphate to carbon dioxide and water by brain mitochondria. On the other hand, BalSzs and Lagnado (4) reported that only 10 per cent of the total glycolytic activity of rat brain was associated with their rnitochondrial preparation. Johnson (5) and Brunngraber and Abood (6) have indicated that most of the glycolytic enzymes are found in the soluble portion of homogenates of brain but that hexokinase is concentrated in the mitochondria. Brunngraber and Abood (6) also reported that the addition of mitochondria to a supernatant fraction in the presence of optimal amounts of exogenous hexokinase caused a twofold stimulation of lactate formation. The mitochondria alone glycolyzedat 25 per cent of the rate of the supernatant fraction. Contrary to the above reports, Brody and Bain (7) and Aldridge (8) concluded that the brain mitochondrial preparations used in their studies possessed no glycolytic activity. Whittaker (9, 10) discovered that the crude mitochondrial fraction from brain contained particles derived from the pinching-off of nerve endings (NEPs) which were easily separable from the mitochondria by density gradient centrifugation. The presence of NEPs which must contain entrapped soluble cytoplasm may explain the 7 to 28 per cent of soluble glycolytic enzymes which Johnson (5) observed in the 309 on Jauary 7, 2018 jcb.rress.org D ow nladed fom mitochondrial fraction. The association of lactic dehydrogenase with the NEPs has been confirmed (10). The preparat ion of beef brain mitochondria described in the previous paper ( l l ) was shown to be relatively free of contamination by nonmitochondrial particulate material and contained mitochondria which exhibited enzymic and morphological properties comparable to those of mitochondrial preparations from other tissues. Studics on the glycolytic capacity of this mitochondrial fraction are reported here. Some aspects of the work have been reported (12, 13). M A T E R I A L S AND M E T I I O D S SOURCE OF MITOCHONDRIA AND SOLUBLE ENZYMES: Brain mitochondria were prepared by Method I of the previous paper (l I). The supernatant fraction from a 30-minute centrifugation of the homogenate at ]05,000 g was used as the source of glycolytic enzymes. Type III hexokinase (150,000 KM units/gin), crystalline hexokinasc (1.3 X I06 KM units/gm), crystalline aldolase, 3-phosphoglycerate kinase, and triosephosphate dehydrogenase were obtained from Sigma Chemical Co., St. Louis, Missouri, Glucose6-phosphate dehydrogenase, triosephosphate isomerase, ot-glycerophosphate dehydrogenase, enolase, pyruvate kinase, and lactic dehydrogenase were obtained from California Corporation for Biochemical Research, Inc., Los Angeles, Phosphofructokinase was partially purified (specific activity: 2.37 ~molc/ minute/mg protein) from rabbit muscle by the method of Ling et al. (14). ENZYMIC ASSAYS: Glycolysis was determined manometrically with 95 per cent N2-5 per cent CO2 as the gas phase in the following medium: ATP', 1 /~mole; MgC12, 20 #mole; KH2PO4, 20 #mole; nicotinamide, 60 #mole; DPN, 1 /zmole; KHCO3, 50 #mole; glucose, 25 #mole; mitochondrial and supernatant fractions and hexokinase (see Tables I, II, IV, and V); final volume, 2.5 ml, pH 7.4. The flasks were gassed for 10 minutes at 30°C and the reaction initiated by the addition of glucose and hexokinase from the side arm. The CO2 evolved as glycolysis progressed was used as an estimate of lactate formation. After 60 minutes, 0.2 ml of 75 per cent TCA was added to each flask and the proteinfree filtrate assayed for lactate by the method of Barker and Summerson (15). a The abbreviations used are: ATP, adenosine triphosphate; DPN, diphosphopyridine nucleotide; TPN and TPNH, oxidized and reduced triphosphopyridine nucleotide, respectively; EDTA, ethylenediaminetetraacetate; TCA, trichloroacetic acid; and NEPs, nerve ending particles. Hexokinase activity was measured spectrophotometrically (Zeiss model M4Q II spectrophotometer) in a system in which glucose-6-phosphate production was coupled to TPNH formation in the presence of glucose-6-phosphate dehydrogenase. The glucose-6phosphate dehydrogenase preparation was diluted to 1 mg per ml in 0.15 M potassium glycylglycine buffer, pH 8.0, and stored at --20°C. The reaction cuvette at 30°C contained in a final volume of 1 ml: 15 /zmole ATP, 25 /zmole glucose, 80 /zmole Tris, pH 8.0, 3 /~mole EDTA, 19 /zmole MgC12, 1 pmole TPN, and 5/zg of glucose-6-phosphate dehydrogenase. The reaction was started by the addition of 0.1 ml of a tissue sample diluted with 10 -~ M EDTA, pH 7.0. A similar assay procedure has been developed by Bennet et al. (16). The remaining glycolytic enzymes were assayed spectrophotometrically (Cary model 14 spectrophotometer) by measuring the absorbancy change at 340 m/z due to oxidation or reduction of pyridine nucleotide. When it was necessary to couple the enzyme being assayed to a reaction involving pyridine nucleotide, purified coupling enzymes were added in excess. Preliminary control experiments without tissue were performed to insure that the coupling enzymes were not contaminated with the particular enzyme being assayed. As an additional control, the change in absorbancy at 340 m/z was determined for each enzyme in the absence of substrate and subtracted from that obtained in its presence. Tissue suspensions were stored at 0°C and diluted prior to addition to the assay medium. The activities were expressed as #mole of substrate consumed per minute per mg of protein at 27°C. Phosphofructokinase, aldolase, triosephosphate isomerase, and oz-glycerophosphate dehydrogenase were assayed by the method of Wu and Racker (17). Triosephosphate dehydrogenase was assayed in the forward reaction by the method of Wu and Racker (17), and in the reverse reaction by the method of Johnson (5). Glucose-6-phosphate isomerase, phosphoglycerate kinase, phosphoglycerate mutase, enolase, and pyruvate kinase were assayed by the method of Johnson (5), and lactic dehydrogenase by the method of Kornberg (18). All substrates obtained as the barium salts were converted to potassium salts by addition of an equivalent amount of sulfate ion and adjusted to pH 7.5 with KOH. Protein was determined by the method previously described (11).