Rate of ATP synthesis in the perfused rat liver by 31P cryo-NMR

ATP concentrations in the perfused rat liver during normoxic perfusion, transient ischemia, and recovery from transient ischemia were measured using the modified 31P cryo-NMR method (Chance, B., Nakase, Y., Bond, M., Leigh, J. S., Jr., and McDonald, G. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 4925-4929). Transient ischemia was induced in the perfused livers of starved rats, and multiple freeze-trapped tissue samples were taken from each liver at short intervals (15-30 s) during ischemia or following reperfusion. The freeze-trapped tissue was pulverized together with an antifreezing agent and high energy metabolites were measured by 31P NMR at 243 K after thawing. By using the cryo-NMR technique, a biochemical time resolution of 2 s could be achieved. Absolute metabolite concentrations were calculated by comparing the peak areas with internal standards mixed into the samples. Good time resolution and reliable concentration measurements provided by the cryo-NMR method enable us to estimate the ATP synthesis rate in the perfused liver during reperfusion following transient ischemia. The rate of ATP synthesis in the normoxic perfusion was 1.95 mumol/min/g wet weight; the maximal ATP synthesis rate during the recovery phase from ischemia was 5.75 mumol/min/g wet weight.     

Visibility of ATP and ADP in freeze-trapped tissue from perfused rat liver during normoxia and ischemia using 31P-cryo-NMR

The visibility of ATP and ADP to NMR was studied by comparing simultaneous measurements of freeze-trapped tissue sections from perfused rat liver under normoxia and ischemia using a modified 31P-cryo-NMR method and biochemical assay. The 31P-cryo-NMR method provides good time resolution and allows the quantitation of absolute metabolite concentrations. Prior to 31P-cryo-NMR measurements, freeze-trapped tissues were thawed in the presence of cryoprotectant and EDTA. With this sample preparation procedure, the integrity of the plasma and mitochondrial membranes was not maintained, inducing homogeneous microviscosity and chelation of intracellular divalent cations, thereby increasing the visibility of metabolites compared to the in vivo NMR measurement. With ischemic stress, total cellular ATP concentration decreased significantly (P less than 0.001). While ADP concentrations measured by cryo-NMR and biochemical analysis were consistent during normoxia and ischemia, ATP concentrations measured by cryo-NMR were significantly lower (P less than 0.05) than those obtained by biochemical analysis. The amount of invisible ATP (0.42 +/- 0.10 mumol/g wet weight: mean +/- S.E.) did not change after the induction of ischemia. The results of this study suggest that ATP invisibility to cryo-NMR is not due to compartmentation into regions of high paramagnetic ion concentrations or high microviscosity, but is influenced by other factors.     

ATP extracellular concentrations are increased in the rat striatum during in vivo ischemia

Interest is growing in the role of adenosine triphosphate (ATP) on P2 receptors during hypoxic/ischemic events in the brain. However, there is no direct evidence of an increase in extracellular ATP levels during cerebral ischemia in vivo. The aim of the present study was to evaluate ATP outflow from the rat striatum by the microdialysis technique associated with focal cerebral ischemia in vivo by intraluminal occlusion of the right middle cerebral artery (MCA). Between 1 and 4h after ischemia, rats showed a clear turning behavior contralateral to the ischemic side. Twenty-four hour after MCA occlusion, ischemic rats had definite neurological deficit and striatal and cortical damage. The ATP concentration (mean+/-S.E.M.) in the striatum of normoxic rats (n = 8) was 3.10+/-0.34 nM. During 220 min after MCA occlusion, the extracellular ATP levels significantly increased two-fold, being 5.90+/-0.61 nM (p < 0.01 versus normoxic level). ATP outflow showed a tendency to increase over time during the 220 min of ischemia. Since extracellular ATP is rapidly metabolized to adenosine, we also assessed ATP outflow in the presence of the ecto-5'-nucleotidase inhibitor, alpha,beta-methylene-adenosine diphosphate (AOPCP, 1 mM) directly perfused into the striatum. The ATP concentration in normoxic rats (n = 8) was increased three-fold in the presence of the ecto-5'-nucleotidase inhibitor (9.57+/-0.26 nM). During 220 min of ischemia, extracellular ATP levels significantly increased 1.3-fold in AOPCP-treated rats (12.62+/-0.65 nM, p < 0.01 versus normoxic level). The present study confirms that ATP is continuously released in the brain and demonstrates for the first time that ATP outflow increases during ischemia in vivo. These results confirm that ATP may be an important mediator in brain ischemia.     

87Rb, 23Na and 31P nuclear magnetic resonance spectroscopy of the perfused rat kidney

87Rb, 23Na and 31P nuclear magnetic resonance (NMR) were used to monitor changes in renal cations and energetics during the induction of hypoxia in the isolated perfused rat kidney. The NMR-determined unidirectional Rb+ flux in normoxic kidneys was shown to be a good measure of net intracellular K+ influx in the perfused rat kidney model. The changes in 87Rb, 23Na and 31P spectra following the induction of hypoxia are consistent with hypoxic depletion of intracellular adenosine triphosphate (ATP) and a subsequent decrease in Na-K-ATPase transport activity. The exponential rate constant for 87Rb+ efflux measured during Rb+ uptake in normoxic kidneys (0.12 +/- 0.01 min-1) was not significantly different to the rate constant for 87Rb+ efflux during the induction of hypoxia (0.16 +/- 0.07 min-1). We conclude that there is no direct effect of hypoxia on renal cellular membrane integrity and that renal cell sensitivity to hypoxia is due to an inability to sustain cellular ion gradients following depletion of intracellular ATP.     

Sustained postischemic cardiodepression following magnesium-diltiazem cardioplegia

Magnesium-diltiazem cardioplegia was evaluated in the intact, perfused rat heart to determine whether the joint administration of these agents would adversely affect myocardial contractile and high-energy phosphate recovery following intermittent, normothermic global ischemic arrest. Sequential metabolic and functional analyses were performed on isolated perfused rat hearts during each phase of the experimental protocol: control (10 min), normoxic cardioplegia (10 min), intermittent global ischemic arrest (two 15-min periods separated by 2 min infusion of the normoxic cardioplegic perfusate), and normoxic postischemic control reperfusion (60 min). Four different cardioplegic solutions were evaluated: 30 mM KCl, 30 mM KCl with 2 mg diltiazem/liter, 20 mM MgCl2, and 20 mM MgCl2 with 2 mg diltiazem/liter. Myocardial phosphatic metabolite levels and intracellular pH were analyzed nondestructively in the intact hearts by phosphorus-31 NMR spectroscopy. Corresponding measurements of peak left intraventricular pressure, rate of peak pressure development (dP/dt), and contraction frequency were performed at the midpoint during each 5-min interval of 31P NMR signal averaging. Magnesium plus diltiazem-treated hearts were distinguished from all other groups by a marked delay in postischemic functional recovery consisting of a prolonged depression in contractility (34% of control, P less than 0.01) that persisted throughout the first 50 min of postischemic reperfusion. Diltiazem in combination with magnesium cardioplegia was detrimental to postischemic functional recovery, despite a rapid restoration of high-energy phosphate stores. The apparent adverse interactive effects of excess magnesium and diltiazem suggest that elective ischemic arrest with magnesium cardioplegia in combination with diltiazem may be contraindicated clinically. The mechanistic basis and drug specificity of this response require further clarification. The present findings appear to exclude ATP and PCr production, and structural causes as the basis for the observed aberrant functional recovery from global ischemia of magnesium plus diltiazem-arrested hearts.     

Bioenergetic targeting during organ preservation: (31)P magnetic resonance spectroscopy investigations into the use of fructose to sustain hepatic ATP turnover during cold hypoxia in porcine livers

During liver preservation, ATP supplies become depleted, leading to loss of cellular homeostatic controls and a cascade of ensuing harmful changes. Anaerobic glycolysis is unable to prolong ATP production for a significant period because of metabolic blockade. Our aim was to promote glycolysis during liver cold hypoxia by supplying fructose as an additional substrate, compared to supplementation with an equivalent concentration of glucose. Porcine livers (two groups; n = 5 in each) were retrieved by clinical harvesting techniques and subjected to two cycles of cold hypoxia and oxygenated hypothermic reperfusion. In the second cycle of reperfusion, the perfusate was supplemented with either 10 mmol/L glucose (Group 1) or 10 mmol/L fructose (Group 2). During reperfusion in both groups, similar levels of ATP were detected by phosphorus magnetic resonance spectroscopy ((31)P MRS). However, during subsequent hypoxia, ATP was detected for much longer periods in the fructose-perfused group. The rate of ATP loss was sevenfold slower during hypoxia in the presence of fructose than in the presence of glucose (ATP consumption of -7.2 x 10(-3)% total (31)P for Group 1 versus -1.0 x 10(-3)% total (31)P for Group 2; P < 0. 001). The changes in ATP were mirrored by differences in other MRS-detectable intermediates; e.g., inorganic phosphate was significantly higher during subsequent hypoxia in Group 1 (45.7 +/- 2.7% total (31)P) than in Group 2 (33.7 +/- 1.1% total (31)P; P < 0. 01). High-resolution MRS of liver tissue extracts demonstrated that fructose was metabolized mainly via fructose 1-phosphate. We conclude that fructose supplied by brief hypothermic perfusion may improve the bioenergetic status of cold hypoxic livers by sustaining anaerobic glycolysis via a point of entry into the pathway that is different from that for glucose.     

Phosphorus-31 nuclear-magnetic-resonance study of phosphorylated metabolites compartmentation, intracellular pH and phosphorylation state during normoxia, hypoxia and ethanol perfusion, in the perfused rat liver

A quantitative analysis of the phosphorus-31 NMR spectra of excised perfused rat liver has been carried out at 80.9 MHz using a 30-mm sample cell. The results indicate that in liver from fed rats, all intracellular ATP is detected by NMR. In contrast, only the cytosolic fractions of Pi and ADP can be observed as indicated by careful analysis of spectra obtained from perchloric acid liver extracts and intact liver under valinomycin perfusion. In well-oxygenated perfused liver the ATP concentration is 7.4 mM. Values of 5.3 mM and 0.9 mM are found respectively for Pi and ADP concentrations in the cytosolic compartment. Cytosolic pH value (pHi) is 7.25 +/- 0.05 and free magnesium concentration 0.5 mM. Addition of 70 mM (0.4%) ethanol to the perfusate of a fed rat liver induces 25% and 38% reduction of ATP and Pi levels, respectively. A large amount of sn-glycerol 3-phosphate is synthesized (up to 11 mM) in the cytosol. After ethanol withdrawal, a large overshoot in cytosolic Pi is observed, which is indicative of a net uptake of Pi across the plasma membrane that occurred during ethanol oxidation. No significant pH variation is observed during ethanol infusion. In perfused liver of rats subjected to 48-h fasts, the concentrations of cytosolic phosphorylated metabolites are 5.3 mM, 0.8 mM and 11.5 mM for ATP, ADP and Pi, respectively. The perfusion of the liver with 70 mM ethanol does not change the adenine nucleotide levels, while the Pi content is decreased by 10%. During a 4-min hypoxia, induced by reducing the perfusion flow rate from 12 ml to 3 ml min-1 (100 g body weight)-1, ATP concentration decreases to 5.8 mM in the fed rat liver. Cytosolic Pi and ADP increase to 8.7 mM and 1.6 mM, respectively. The cytosolic pH evolves to more acidic values and reaches 7.02 +/- 0.05 at the end of the 4-min hypoxic period.     

Cytosolic free magnesium levels in ischemic rat heart

Changes in cytosolic free magnesium ion concentration (Mgi) during myocardial ischemia were measured by 19F NMR in perfused rat hearts loaded with fluorine-labeled derivatives of the magnesium chelator o-aminophenol-N,N,O-triacetate. The perfused rat hearts were loaded intracellularly with the appropriate magnesium indicator by perfusion with 200-400 ml of Krebs-Henseleit buffer containing 5 microM acetoxymethyl ester of the indicator. Basal Mgi concentrations measured by three different indicators averaged 0.85 +/- 0.10 mM (n = 9) and showed no correlation with the KD of the indicator used. 31P NMR measurements of the magnesium-dependent shift between alpha- and beta-phosphates of ATP demonstrate that there is no measurable lowering of Mgi during loading with fluorinated o-aminophenol-N,N,O-triacetate. Between 10 and 15 min of ischemia, Mgi rose nearly 3-fold to 2.1 +/- 0.4 mM. This increase in Mgi occurred over the same time course as the decrease in ATP. After 20 min of reperfusion with Krebs-Henseleit buffer, Mgi declined to 1.5 +/- 0.5 mM. This sustained elevation of Mgi above basal levels may inhibit calcium release from sarcoplasmic reticulum, thereby contributing to the well documented impairment of mechanical function that occurs after a reversible period of ischemia.     

Effects of fructose on the energy metabolism and acid-base status of the perfused starved-rat liver. A 31phosphorus nuclear magnetic resonance study.

Fructose metabolism has been studied with 31P n.m.r. in perfused livers from rats starved for 48h. The time course of changes in liver ATP, Pi and sugar phosphate (fructose l-phosphate) concentrations, and intracellular pH were followed in each perfusion after infusion of fructose to give an initial concentration of either 5mM or 10mM. Rapid falls in the concentrations of ATP and Pi and intracellular pH occurred after infusion of fructose, reaching a minimum after 4-5 min, which was lower in the 10mM group than in the 5mM group. These changes were accompanied by a rapid rise in fructose 1-phosphate, reaching a plateau also after 4-5 min. At both concentrations of fructose, after the early falls, some recovery of ATP, Pi and intracellular pH occurred; this was complete for Pi and intracellular pH in the 5mM-fructose experiments (within 12-30 min). Complete restoration of ATP to the pre-fructose value was not achieved in either the 5mM of 10mM groups. Measurements of the uptake of lactate by the liver indicated that the fall in intracellular pH was caused primarily by production of protons accompanying the formation of lactate from fructose with possibly a transient contribution generated during the rise in fructose 1-phosphate.     

Fructose protects rat hepatocytes from anoxic injury. Effect on intracellular ATP, Ca2+i, Mg2+i, Na+i, and pHi.

The effects of fructose on the intracellular ionic changes evoked by anoxia were studied in freshly isolated rat hepatocytes maintained in agarose gel threads and perfused with Krebs-Henseleit bicarbonate buffer (KHB). Cytosolic free calcium (Ca2+i) was measured with aequorin, intracellular sodium (Na+i) with sodium-binding benzofuran isophthalate, intracellular pH (pHi) with 2'-7'-bis(carboxyethyl)-5,6-carboxyfluorescein, lactic dehydrogenase (LDH) by the increase in NADH absorbance during lactate oxidation to pyruvate, and viability by trypan blue exclusion. ATP, Pi, phosphomonoesters, and the cell phosphorylation potential assessed by the reciprocal of the Pi/ATP ratio were measured by 31P NMR spectroscopy in real time. Intracellular free Mg2+ (Mg2+i) was calculated from the chemical shift of beta-ATP relative to alpha-ATP in the NMR spectra. Anoxia was induced by perfusing the cells with KHB saturated with 95% N2, 5% CO2. When the perfusate contained 5 mM glucose as substrate, anoxia caused a fall in ATP, a rise in Pi, and in the Pi/ATP ratio, a biphasic increase in Ca2+i that reached 1.45 +/- 0.42 microM and a 6-fold increase in LDH. When 15 mM fructose was used as substrate during the anoxic period, intracellular ATP decreased much faster than with glucose, Pi did not increase, and the concentration of phosphomonoesters increased 2.5-fold. During the first hour of anoxia, the Pi/ATP ratio was higher in the fructose than in the glucose group indicating that the hepatocyte phosphorylation potential and ATP decreased faster and to lower levels with fructose than with glucose. On the other hand, ATP and the phosphorylation potential of the fructose group increased during the second hour of anoxia, in contrast to their continuous decline in the glucose group. The major surge in Ca2+i was depressed 52% when glucose was replaced by fructose: Ca2+i reached only 0.7 +/- 0.2 microM instead of 1.45 +/- 0.42 microM (p less than 0.01). Anoxia also caused an increase in Na+i and an intracellular acidosis. The rise in Na+i was significantly greater with fructose than with glucose. Na+i rose from a control value of 15.9 +/- 2.4 to 32.2 +/- 0.4 mM with glucose and to 48.7 +/- 0.7 mM with fructose (p less than 0.001). The decrease in pHi from a control value of 7.43 +/- 0.03 was consistently greater and faster with fructose than with glucose: 6.59 +/- 0.03 and 7.04 +/- 0.01, respectively. At the same time, fructose completely suppressed LDH release and reduced the loss of viability produced by anoxia from 27.7 +/- 2.9 to 14 +/- 3.1% (p less than 0.05).     

Free fatty acids, but not ketone bodies, protect diabetic rat hearts during low-flow ischemia

To determine whether the effects of fatty acids on the diabetic heart during ischemia involve altered glycolytic ATP and proton production, we measured energetics and intracellular pH (pH(i)) by using (31)P NMR spectroscopy plus [2-(3)H]glucose uptake in isolated rat hearts. Hearts from 7-wk streptozotocin diabetic and control rats, perfused with buffer containing 11 mM glucose, with or without 1.2 mM palmitate or the ketone bodies, 4 mM beta-hydroxybutyrate plus 1 mM acetoacetate, were subjected to 32 min of low-flow (0.3 ml x g wet wt(-1) x min(-1)) ischemia, followed by 32 min of reperfusion. In control rat hearts, neither palmitate nor ketone bodies altered the recovery of contractile function. Diabetic rat hearts perfused with glucose alone or with ketone bodies, had functional recoveries 50% lower than those of the control hearts, but palmitate restored recovery to control levels. In a parallel group with the functional recoveries, palmitate prevented the 54% faster loss of ATP in the diabetic, glucose-perfused rat hearts during ischemia, but had no effect on the rate of ATP depletion in control hearts. Palmitate decreased total glucose uptake in control rat hearts during low-flow ischemia, from 106 +/- 17 to 52 +/- 12 micromol/g wet wt, but did not alter the total glucose uptake in the diabetic rat hearts, which was 42 +/- 5 micromol/g wet wt. Recovery of contractile function was unrelated to pH(i) during ischemia; the glucose-perfused control and palmitate-perfused diabetic hearts had end-ischemic pH(i) values that were significantly different at 6.36 +/- 0.04 and 6.60 +/- 0.02, respectively, but had similar functional recoveries, whereas the glucose-perfused diabetic hearts had significantly lower functional recoveries, but their pH(i) was 6.49 +/- 0.04. We conclude that fatty acids, but not ketone bodies, protect the diabetic heart by decreasing ATP depletion, with neither having detrimental effects on the normal rat heart during low-flow ischemia.     

Gluconeogenesis from fructose predominates in periportal regions of the liver lobule

Gluconeogenesis from fructose was studied in periportal and pericentral regions of the liver lobule in perfused livers from fasted, phenobarbital-treated rats. When fructose was infused in increasing concentrations from 0.25 to 4 mM, corresponding stepwise increases in glucose formation by the perfused liver were observed as expected. Rates of glucose and lactate production from 4 mM fructose were around 100 and 75 mumol/g/h, respectively. Rates of fructose uptake were around 190 mumol/g/h when 4 mM fructose was infused. 3-Mercaptopicolinate, an inhibitor of phosphoenolpyruvate carboxykinase, decreased glucose formation from fructose maximally by 20% suggesting that a fraction of the lactate formed from fructose is used for glucose synthesis. A good correlation (r = 0.92) between extra oxygen consumed and glucose produced from fructose was observed. At low fructose concentrations (less than 0.5 mM), the extra oxygen uptake was much greater than could be accounted for by glucose synthesis possibly reflecting fructose 1-phosphate accumulation. Furthermore, fructose diminished ATP/ADP ratios from about 4.0 to 2.0 in periportal and pericentral regions of the liver lobule indicating that the initial phosphorylation of fructose via fructokinase occurs in both regions of the liver lobule. Basal rates of oxygen uptake measured with miniature oxygen electrodes were 2- to 3-fold higher in periportal than in pericentral regions of the liver lobule during perfusions in the anterograde direction. Infusion of fructose increased oxygen uptake by 65 mumol/g/h in periportal areas but had no effect in pericentral regions of the liver lobule indicating higher local rates of gluconeogenesis in hepatocytes located around the portal vein. When perfusion was in the retrograde direction, however, glucose was synthesized nearly exclusively from fructose in upstream, pericentral regions. Thus, gluconeogenesis from fructose is confined to oxygen-rich upstream regions of the liver lobule in the perfused liver.     

Cardioprotection in chronically hypoxic rabbits persists on exposure to normoxia: role of NOS and KATP channels

Hypoxia from birth increases resistance to myocardial ischemia in infant rabbits. We hypothesized that increased cardioprotection in hearts chronically hypoxic from birth persists following development in a normoxic environment and involves increased activation of nitric oxide synthase (NOS) and ATP-dependent K (K(ATP)) channels. Resistance to myocardial ischemia was determined in rabbits raised from birth to 10 days of age in a normoxic (Fi(O(2)) = 0.21) or hypoxic (Fi(O(2)) = 0.12) environment and subsequently exposed to normoxia for up to 60 days of age. Isolated hearts (n = 8/group) were subjected to 30 min of global ischemia followed by 35 min of reperfusion. At 10 days of age, resistance to myocardial ischemia (percent recovery postischemic recovery left ventricular developed pressure) was higher in chronically hypoxic hearts (68 +/- 4%) than normoxic controls (43 +/- 4%). At 10 days of age, N(G)-nitro-L-arginine methyl ester (200 microM) and glibenclamide (3 microM) abolished the cardioprotective effects of chronic hypoxia (45 +/- 4% and 46 +/- 5%, respectively) but had no effect on normoxic hearts. At 30 days of age resistance to ischemia in normoxic hearts declined (36 +/- 5%). However, in hearts subjected to chronic hypoxia from birth to 10 days and then exposed to normoxia until 30 days of age, resistance to ischemia persisted (63 +/- 4%). L-NAME or glibenclamide abolished cardioprotection in previously hypoxic hearts (37 +/- 4% and 39 +/- 5%, respectively) but had no effect on normoxic hearts. Increased cardioprotection was lost by 60 days. We conclude that cardioprotection conferred by adaptation to hypoxia from birth persists on subsequent exposure to normoxia and is associated with enhanced NOS activity and activation of K(ATP) channels.     

Formation of activated oxygen in the hypoxic rat liver

The biliary GSSG efflux rate of normoxic perfused rat liver was 1.5 +/- 0.2 nmol/min/g liver wet weight. The GSSG efflux rate as indicator for the flux through the glutathione peroxidase reaction and, therefore, for an oxidative loading increased with the extent of hypoxia. 2.6 +/- 0.5 nmol/min/g were released from the severely hypoxic liver. The hydroxyl radical scavenger formate as well as the xanthine oxidase inhibitor allopurinol reduced the efflux rate of GSSG. GSH was released from the perfused liver at a rate of 15.5 nmol/min/g which was nearly unchanged in severe hypoxia. The high rate of glucose liberation from the hypoxic liver declined to almost that of the normoxic organ in the presence of formate. There is an 'oxidative stress' during hypoxic liver perfusion which probably originates from increased generation of activated oxygen species in the degradation of purine nucleotides.     

31P nuclear magnetic resonance of metabolic changes associated with cyanide intoxication in the perfused rat liver.

Metabolic changes associated with cyanide intoxication were observed for the first time in perfused rat liver using ³¹P nuclear magnetic resonance (NMR) at 60.7 MHz. Well-oxygenated control livers showed strong ATP peaks and little discernable internal orthophosphate (Pi). Perfusion with 2 mM cyanide eliminated the observable ATP peaks and caused internal Pi to increase. Despite clear evidence for ATP hydrolysis, resonances from cytoplasmic ADP were conspicuously absent. Resumption of perfusion with cyanide-free buffer caused a dramatic return of the ATP peaks with a concomitant fall in internal Pi. These metabolic changes are consistent with reversible binding of cyanide to mitochondrial cytochromes and their observation by ³¹P NMR indicates the potential of this method for studying metabolism in whole, perfused rat liver under physiologic conditions.     

Role of intracellular buffering power on the mitochondria-cytosol pH gradient in the rat liver perfused at 4 degrees C

The factors regulating the amplitude and the pH gradient between cytosol and mitochondria (DeltapHmito-cyt) were investigated in the isolated rat liver perfused at 4 degrees C. Liver ATP content, pH, and buffering power of cytosolic and mitochondrial compartments were evaluated in situ using phosphorus-31 nuclear magnetic resonance spectroscopy. No DeltapHmito-cyt was detected in the liver perfused without bicarbonate. Permeant weak acid in the perfusate (H2CO3, 25 mM, or isobutyric acid, 25, 50, or 100 mM) acidified both cytosol and mitochondria and revealed a DeltapHmito-cyt from 0.06 to 0.31 pH unit. Nevertheless, the manipulations of the DeltapHmito-cyt were more effective under bicarbonate-free conditions, due to the absence of buffering by H2CO3/HCO-3. In the absence of bicarbonate, the intracellular buffering power was threefold higher in the mitochondria (110 mmol/pH unit at pHmito 7.16) than in the cytosol (44 mmol/pH unit at pHcyt 7.30) and dependent on the matrix and cytosol pH, respectively. These buffering powers were almost double in the presence of bicarbonate. In the bicarbonate-free perfused liver, the respiratory activity was 0.08 +/- 0.02 micromol O2/min. g liver wet weight and the ATP turnover was only 40 +/- 7 nmol/min. g liver wet weight, indicating the weak activity of liver mitochondria when DeltapHmito-cyt was <0.05 pH unit. The ATP turnover during a 50 mM isobutyric acid load was 35 +/- 4 nmol/min. g liver wet weight whereas DeltapHmito-cyt rose to 0.26 +/- 0.02 pH unit and pHmito remained alkaline. Hence, although DeltapHmito-cyt was increased the ATP turnover remained unchanged. This work is the first evaluation of the mitochondrial buffering power in the isolated liver. The DeltapHmito-cyt observed within various acid loads reflected the differential titration of cytosol and mitochondria containing proteins and H2CO3/HCO-3 buffering systems. Moreover, no direct relationship between DeltapHmito-cyt and ATP turnover could be shown.     

NMR-invisible ATP in rat heart and its change in ischemia

The subcellular compartmentalization of adenosine 5'-triphosphate (ATP) in isolated perfused rat heart and its relation to energy depletion in ischemia were examined by 31P nuclear magnetic resonance (31P-NMR) spectroscopy and chemical analyses. The signal intensities of the beta-phosphate of ATP and creatine phosphate in the 31P-NMR were standardized by the intracellular volume ratio measured with 23Na-NMR to determine the actual content of each. During aerobic perfusion the ATP content determined by NMR (13.7 +/- 2.2 mumol/g dry weight) was significantly lower than that found by chemical analysis (22.4 +/- 0.7 mumol/g dry weight), while the creatine phosphate contents determined by the two methods were the same. During ischemia at 33 degrees C, the signal of the beta-phosphate of ATP in the 31P-NMR spectrum decreased progressively, disappearing completely after 16 min. But at this time 5.7 +/- 1.7 mumol/g dry weight of myocardial ATP was still detected by chemical analysis. These results indicated that there were two different compartments of intracellular ATP in the heart, only one of which is detectable by 31P-NMR spectroscopy, and that during ischemia the ATP that is detectable, which seems to be the free ATP in the cytosol, decreased more rapidly than the ATP in the other compartment.     

Stabilization of hsp70 mRNA on prolonged cell exposure to hypertonicity

The question arises as to the effect of ethanol on the actual yield of oxidative phosphorylation in the whole liver because of contradictory results reported in isolated hepatic mitochondria.The adenosine triphosphate (ATP) content of liver isolated from fed rats and perfused in the presence (10 mM) and absence of ethanol was continuously evaluated using 31P Nuclear Magnetic Resonance (NMR). An accurate estimation of mitochondrial ATP synthesis in the whole organ was obtained by subtracting the glycolytic ATP supply from the total ATP production. Simultaneously, the respiratory activity was assessed using O(2) Clark electrodes.The data indicate that ethanol enhanced the net consumption of ATP, leading to a new steady state of the ATP content. ATP synthesis was also found higher under ethanol [1.86+/-0.02 micromol/min g wet weight (min g ww)] than in control [1.44+/-0.18 micromol/min g ww]. However, mitochondrial respiration remained unchanged [2.20+/-0.13 micromol/min g ww] and, consequently, the in situ mitochondrial ATP/O ratio increased from 0.33+/-0.035 (control) to 0.42+/-0.015 (ethanol).The increase of the oxidative phosphorylation yield in the whole liver may be linked to the decrease in cytochrome oxidase activity induced by ethanol [FEBS Lett. 468 (2000) 239]. The significant raise (27%) of the ATP/O ratio was not sufficient to maintain the ATP level following ethanol-increased ATP consumption.     

Measurement of the dissociation constant of MgATP at physiological nucleotide levels by a combination of 31P NMR and optical absorbance spectroscopy

To resolve a controversy in the literature concerning the affinity of Mg++ for ATP to be used in our noninvasive 31P NMR procedure for the determination of free Mg++ in living cells, we have reinvestigated the apparent dissociation constant of MgATP under physiologic ionic conditions and over the cellular range of ATP concentrations by a combination of NMR and optical absorbance techniques. The new combination method utilizes 31P NMR chemical shifts to determine the degree of Mg++ chelation of ATP in a solution containing free ATP and MgATP, and uses a properly calibrated indicator dye, antipyrylazo III, for optical measurement of free Mg++ in the same solution. The data yield an average value of 50 +/- 10 microM for the apparent dissociation constant of MgATP which indicates low levels of free Mg++ (less than 1 mM) in several different types of tissues, including perfused heart muscle, contrary to a recent report in the literature.     

Phosphorus-31 nuclear magnetic resonance of C6 glioma cells and rat astrocytes. Evidence for a modification of the longitudinal relaxation time of ATP and Pi during glucose starvation

31P-NMR spectroscopy has been used to study the energy metabolism and the NMR visibility of ATP and intracellular Pi of the C6 glioma cell line and rat astrocyte grown on microcarrier beads with the following results. 1. In vivo NMR spectra of C6 glioma cells and rat astrocytes indicate that these cells were able to maintain their level of ATP resonances during a long anoxic period (more than an hour). Both cell types were sensitive to ischemia which induced a loss of ATP resonances within 40 min. Glucose starvation induced by 40% decrease in ATP resonances correlated to a 50% increase in the intensity of the Pi signal. These changes corresponded to a new steady state which could be reversed by reperfusing the cells with a glucose-containing medium. 2. In contrast to in vivo data, 31P-NMR analyses of perchloric acid extracts of cells incubated in a glucose-free medium showed that their ATP and Pi contents were unchanged during starvation. The changes of NMR visibility of the metabolites in living C6 cells were correlated to modifications of their macroscopic longitudinal relaxation times, evolving from 0.30 +/- 0.08 s and 6.6 +/- 1.5 s in the presence of glucose to 0.68 +/- 0.26 s and 3.2 +/- 0.9 s in the absence of glucose for ATP and Pi, respectively. The changes of the NMR detectability of ATP and Pi indicate that changes in their microenvironment occur during glucose starvation, suggesting the existence of different pools of these metabolites within the cells. 3. Under various experimental conditions, i.e. anoxia, ischemia and glucose starvation, rat astrocytes in primary culture showed a very similar behavior to that of C6 cells, suggesting a similar adaptability to the nature of the energy supply for both the normal and the malignant cell.