t-Butylhydroperoxide and gliotoxin stimulate Ca2+ release from rat skeletal muscle mitochondria

Summary

Rat liver mitochondria have a specific Ca²⁺ release pathway which operates when NAD⁺ is hydrolysed to nicotinamide and ADPribose. NAD⁺ hydrolysis is Ca²⁺-dependent and inhibited by cyclosporine A (CSA). Mitochondrial Ca²⁺ release can be activated by the prooxidant t-butylhydroperoxide (tbh) or by gliotoxin (GT), a fungal metabolite of the epipolythiodioxopiperazine group. Tbh oxidizes NADH to NAD⁺ through an enzyme cascade consisting of glutathione peroxidase, glutathione reductase, and the energy linked transhydrogenase, whereas GT oxidizes some vicinal thiols to the disulfide form, a prerequisite for NAD⁺ hydrolysis. We report now that rat skeletal muscle mitochondria also contain a specific Ca²⁺ release pathway activated by both tbh and GT. Ca²⁺ release increases with the mitochondrial Ca²⁺ load, is completely inhibited in the presence of CSA, and is paralleled by pyridine nucleotide oxidation. In the presence of tbh and GT, mitochondria do not lose their membrane potential and do not swell, provided continuous release and re-uptake of Ca²⁺ (‘Ca²⁺ cycling’) is prevented. These data support the notion that both tbh- and GT-induced Ca²⁺ release are not the consequence of an unspecific increase of the inner membrane permeability (‘pore’ formation). Tbh induces Ca²⁺ release from rat skeletal muscle less efficiently than from liver mitochondria indicating that the coupling between tbh and NADH oxidation is much weaker in skeletal muscle mitochondria. This conclusion is corroborated by a much lower glutathione peroxidase activity in skeletal muscle than in liver mitochondria. The prooxidant-dependent pathway promotes, under drastic conditions (high mitochondrial Ca²⁺ loads and high tbh concentrations), Ca²⁺ release to about the same extent and rate as the Na⁺/Ca²⁺ exchanger. This renders the prooxidant-dependent pathway relevant in the pathophysiology of mitochondrial myopathies where its activation by an increased generation of reactive oxygen species probably results in excessive Ca²⁺ cycling and damage to mitochondria.     

Effect of nitric oxide donors on Ca2+-uptake in the rat heart and liver mitochondria

The influence of NO donors, nitroglycerin (NG) and sodium nitroprusside (SNP), on Ca2+- uptake in rat heart and liver mitochondria is studied. It is shown that in vivo NG causes a rapid dose-dependent increase of Ca2+-uptake in rat heart mitochondria most pronounced at 0,5-1,0 mg/kg weight NG. This sharp increase of Ca2+-uptake is not accounted for by changes in membrane potential of mitochondria (deltapsim) because deltapsim is not influenced by less than 1,0 mg/kg NG, and moreover, decrease by approximately 30% is observed at 1,0-1,5 mg/kg NG. In vitro, on the contrary, a concentration-dependent decrease in Ca2+-uptake caused by NG as well as SNP is observed together with simultaneous decrease of deltapsim and concentration-dependent release of Ca2+ from mitochondria via Ca2+-uniporter as the result of partial depolarisation of mitochondrial inner membrane. The data obtained give an evidence that increase in Ca2+-uptake caused by NO donor in vivo takes place independently of changes in deltapsim and also is not resulted from a direct action of NO on Ca2+-uniporter. These observations allow us to suppose that activation of mitochondrial Ca2+-uptake in vivo and corresponding decrease in cytosolic Ca2+ concentration could be involved in vasodilatory action of nitric oxide.     

Isolation of a fraction with Ca2+ ionophore properties from rat liver mitochondria

Isolation of a small protein with properties of a Ca2+ ionophore from calf heart mitochondria has recently been reported [A. Y. Jeng and A. E. Shamoo, 1980, J. Biol. Chem. 255, 6897, 6904]. We have isolated a fraction with similar physical and chemical properties from rat liver mitochondria. In particular, the hepatic preparation is able to bind Ca2+ with high affinity in such a fashion that the resultant complex is soluble in a hydrophobic phase. It will also transport Ca2+ through a stirred organic phase (Pressman cell). Interaction of the liver preparation with Ca2+ is sensitive to inhibitors of mitochondrial Ca2+ uptake. The hepatic preparation contains both protein and lipid components. The phospholipid components were identified and the behavior of a similar mixture of commercially available phospholipids was compared to that of the ionophore fraction from rat liver mitochondria. All of the Ca2+ binding properties of the rat liver preparation could be mimicked by the lipids. In a preliminary experiment, reduction of the phospholipid content of the preparation to less than one lipid phosphate per protein molecule (assuming a molecular weight of 3000 by analogy with the calf heart case) resulted in a protein that was unable to bind Ca2+. We, therefore, suggest that the ability of the preparation to interact with Ca2+ is due to the constituent phospholipids. Measurements of phospholipid-Ca2+ interactions in the model systems and under the conditions of low (microM) Ca2+ and phospholipid concentration utilized here demonstrated an affinity for Ca2+ (Ks approximately 1 microM) and a cation selectivity that have not previously been reported.     

Characteristics of sarcoplasmic reticulum from normal and denervated rat skeletal muscle.

Denervation of rat skeletal muscle produces after 14 days a decrease in Ca2+ uptake of a heterogeneous population of sarcoplasmic-reticulum vesicles, when measured in the presence of oxalate. The Mg2+-dependent ATPase (Ca2+-independent) activity increased after the same period and the Ca2+ + Mg2+-dependent ATPase activity decreased. Concomitant with these changes, there was an increase in vesicle size and calcium content. The observations are discussed in terms of changes in altered membrane structure, manifested in the shift of the equilibrium of the ATPase from an enzyme involved in calcium transport to a phosphoenzyme giving rise to an increase in the Mg2+-dependent ATPase activity.     

1,25-Dihydroxyvitamin D3 stimulates 45Ca2+-uptake by cultured vascular smooth muscle cells derived from rat aorta

We have recently shown the presence of receptors for 1,25-dihydroxyvitamin D3 and that 1,25-dihydroxyvitamin D3 stimulates Ca-ATPase in vascular smooth muscle cells presumably via receptor mediated mechanism. These data suggest that the sterol may directly be involved in the regulation of cellular calcium homeostasis. To further define action of vitamin D in smooth muscle cells, we studied effect of the sterol on cellular uptake of calcium. 1,25-dihydroxyvitamin D3 stimulated 45Ca2+ uptake by cultured cells, A7r5, derived from fetal rat aorta, when the cells were incubated with the sterol for 18 hr. The effect was dose-dependent at 10(-10) to 10(-9) M, and three orders of magnitude higher concentration of 25-hydroxyvitamin D3 or 24,25-dihydroxyvitamin D3 was needed to obtain similar effects. Furthermore, the effect of 1,25-dihydroxyvitamin D3 was abolished by cycloheximide (10(-5) M), a protein synthesis inhibitor. These data clearly suggest that 1,25-dihydroxyvitamin D3 may directly regulate cellular calcium homeostasis in vascular smooth muscle cells presumably via receptor mediated mechanism.     

Fiber type populations and Ca2+-activation properties of single fibers in soleus muscles from SHR and WKY rats

Electrophoretic analyses of muscle proteins in whole musclehomogenates and single muscle fiber segments were used to examine myosin heavy chain (MHC) and myosin light chain 2 (MLC2) isoform composition and fiber type populations in soleus muscles from spontaneously hypertensive rats (SHRs) and their age-matchednormotensive controls [Wistar-Kyoto (WKY) rats], at threestages in the development of high blood pressure (4 wk, 16 wk, and 24 wk of age). Demembranated (chemically skinned with 2% Triton X-100),single fiber preparations were used to determine the maximumCa²⁺-activated force percross-sectional area, calcium sensitivity, and degree of cooperativityof the contractile apparatus andCa²⁺-regulatory system withrespect to Ca²⁺. The results showthat, at all ages examined, 1) SHRsoleus contained a lower proportion of MHCI and MLC2 slow (MLC2s) and ahigher proportion of MHCIIa, MHCIId/x, and MLC2 fast (MLC2f )isoforms than the age-matched controls;2) random dissection of single fibers from SHR and WKY soleus produced four populations of fibers: type I (expressing MHCI), type IIA (expressing MHCIIa), hybrid typeI+IIA (coexpressing MHCI and MHCIIa), and hybrid type IIA+IID (coexpressing MHCIIa and MHCIId/x); and3) single fiber dissection from SHRsoleus yielded a lower proportion of type I fibers, a higher proportionof fast-twitch fibers (types IIA and IIA+IID), and a higher proportionof hybrid fibers (types I+IIA and IIA+IID) than the homologous musclesfrom the age-matched WKY rats. Because the presence of hybrid fibers isviewed as a marker of muscle transformation, these data suggest thatSHR soleus undergoes transformation well into adulthood. Our data showalso that, for a given fiber type, there are no significant differencesbetween SHR and WKY soleus muscles with respect to any of theCa²⁺-activation propertiesexamined. This finding indicates that the lower specific tensionsreported in the literature for SHR soleus muscles are not due tostrain- or hypertension-related differences in the function of thecontractile apparatus or regulatory system.     

Alterations of nPKC distribution,but normal Akt/PKB activation in denervated rat soleus muscle

Denervation has been shown to impair the ability of insulin to stimulate glycogen synthesis and, to a lesser extent, glucose transport in rat skeletal muscle. Insulin binding to its receptor, activation of the receptor tyrosine kinase and phosphatidylinositol 3'-kinase do not appear to be involved. On the other hand, it has been shown that denervation causes an increase in the total diacylglycerol (DAG) content and membrane-associated protein kinase C (PKC) activity. In this study, we further characterize these changes in PKC and assess other possible signaling abnormalities that might be related to the decrease of glycogen synthesis. The results reveal that PKC-epsilon and -theta;, but not -alpha or -zeta, are increased in the membrane fraction 24 h after denervation and that the timing of these changes parallels the impaired ability of insulin to stimulate glycogen synthesis. At 24 h, these changes were associated with a 65% decrease in glycogen synthase (GS) activity ratio and decreased electrophoretic mobility, indicative of phosphorylation in GS in muscles incubated in the absence of insulin. Incubation of the denervated soleus with insulin for 30 min minimally increased glucose incorporation into glycogen; however, it increased GS activity threefold, to a value still less than that of control muscle, and it eliminated the gel shift. In addition, insulin increased the apparent abundance of GS kinase (GSK)-3 and protein phosphatase (PP)1 alpha in the supernatant fraction of muscle homogenate to control values, and it caused the same increases in GSK-3 and Akt/protein kinase B (PKB) phosphorylation and Akt/PKB activity that it did in nondenervated muscle. No alterations in hexokinase I or II activity were observed after denervation; however, in agreement with a previous report, glucose 6-phosphate levels were diminished in 24-h-denervated soleus, and they did not increase after insulin stimulation. These results indicate that alterations in the distribution of PKC-epsilon and -theta; accompany the impairment of glycogen synthesis in the 24-h-denervated soleus. They also indicate that the basal rate of glycogen synthesis and its stimulation by insulin in these muscles are diminished despite a normal activation of Akt/PKB and phosphorylation of GSK-3. The significance of the observed alterations to GSK-3 and PP1 alpha distribution remain to be determined.     

Respiratory activities of subsarcolemmal and intermyofibrillar mitochondrial populations isolated from denervated and control rat soleus muscles

Ultraturrax and Nagarse released populations of mitochondria isolated from control and day 21 denervated rat soleus muscle were characterized with respect to their oxidative phosphorylation, ADP translocase and ATPase activities. Both Ultraturrax and Nagarse released mitochondrial populations displayed lower capacities for oxidative phosphorylation; lower ADP translocase activities and higher Mg2+ stimulated ATPase activities than their corresponding controls. For both the denervated and control states, the Nagarse-released mitochondrial populations displayed significantly higher respiratory activities than the Ultraturrax released fractions. The significance of these findings is discussed with regard to the process of mitochondrial respiratory control. In addition the role of mitochondrial dysfunction in denervation muscular atrophy is assessed.     

Satellite cells on isolated myofibers from normal and denervated adult rat muscle.

Satellite cells (SCs) in normal adult muscle are quiescent. They can enter the mitotic program when stimulated with growth factors such as basic FGF. Short-term denervation stimulates SC to enter the mitotic cycle in vivo, whereas long-term denervation depletes the SC pool. The molecular basis for the neural influence on SCs has not been established. We studied the phenotype and the proliferative capacity of SCs from muscle that had been denervated before being cultured in vitro. The expression of PCNA, myogenin, and muscle (M)-cadherin in SCs of normal and denervated muscle fibers was examined at the single-cell level by immunolabeling in a culture system of isolated rat muscle fibers with attached SCs. Immediately after plating (Day 0), neither PCNA nor myogenin was present on normal muscle fibers, but we detected an average of 0.5 M-cadherin(+) SCs per muscle fiber. The number of these M-cadherin(+) cells (which are negative for PCNA and myogenin) increased over the time course examined. A larger fraction of cells negative for M-cadherin underwent mitosis and expressed PCNA, followed by myogenin. The kinetics of SCs from muscle fibers denervated for 4 days before culturing were similar to those of normal controls. Denervation from 1 to 32 weeks before plating, however, suppressed PCNA and myogenin expression almost completely. The fraction of M-cadherin(+) (PCNA(-)/myogenin(-)) SCs was decreased after 1 week of denervation, increased above normal after denervation for 4 or 8 weeks, and decreased again after denervation for 16 or 32 weeks. We suggest that the M-cadherin(+) cells are nondividing SCs because they co-express neither PCNA or myogenin, whereas the cells positive for PCNA or myogenin (and negative for M-cadherin) have entered the mitotic cycle. SCs from denervated muscle were different from normal controls when denervated for 1 week or longer. The effect of denervation on the phenotypic modulation of SCs includes resistance to recruitment into the mitotic cycle under the conditions studied here and a robust extension of the nonproliferative compartment. These characteristics of SCs deprived of neural influence may account for the failure of denervated muscle to fully regenerate. (J Histochem Cytochem 47:1375-1383, 1999)     

Kinetics of Ca2+ carrier in rat liver mitochondria.

The rate of aerobic Ca2+ transport is limited by the rate of the H+ pump rather than by the Ca2+ carrier. The kinetics of the Ca2+ carrier has therefore been studied by using the K+ diffusion potential as the driving force. The apparent Vmax of the Ca2+ carrier is, at 20 degrees C, about 900 nmol (mg of protein)-1 min-1, more than twice the rate of the H+ pump. The apparent Vmax is depressed by Mg2+ and Li+. This supports the view that the electrolytes act as noncompetitive inhibitors of the Ca2+ carrier. The degree of sigmoidicity of the kinetics of Ca2+ transport increases with the lowering of the temperature and proportionally with the concentration of impermeant electrolytes such as Mg2+ and Li+ but not choline. The effects of temperature and of electrolyte do not support the view that the sigmoidicity is due to modifications of the surface potential. Rather, they suggest that Ca2+ transport occurs through a multisubunit carrier, where cooperative phenomena are the result of ligand-induced conformational changes due to the interaction of several allosteric effectors with the carrier subunits. In contrast with La3+ which acts as a competitive inhibitor, Ruthenium Red affects the kinetics by inducing phenomena both of positive and of negative cooperativity. The Ruthenium Red induced kinetics has been reproduced through curve-fitting procedures by applying the Koshland sequential interaction hypothesis to a four-subunit Ca2+ carrier model.     

Quinone imine-induced Ca2+ release from isolated rat liver mitochondria

Incubation of Ca2(+)-loaded rat liver mitochondria with N-acetyl-p-benzoquinone imine (NAPQI) or its two dimethylated analogues resulted in a concentration dependent Ca2+ release, with the following order of potency: 2,6-(Me)2-NAPQI greater than NAPQI greater than 3,5-(Me)2-NAPQI. The quinone imine-induced Ca2+ release was associated with NAD(P)H oxidation and was prevented when NAD(P)+ reduction was stimulated by the addition of 3-hydroxybutyrate. Mitochondrial glutathione was completely depleted within 0.5 min by all three quinone imines, even at low concentrations that did not result in Ca2+ release. Depletion of mitochondrial GSH by pretreatment with 1-chloro-2,4-dinitrobenzene enhanced quinone imine-induced NAD(P)H oxidation and Ca2+ release. However, 3-hydroxybutyrate protected from quinone imine-induced Ca2+ release in GSH-depleted mitochondria. Mitochondrial membrane potential was lost after the addition of quinone imines at concentrations that caused rapid Ca2+ release; however, subsequent addition of EGTA led to the complete recovery of the transmembrane potential. In the absence of Ca2+, the quinone imines caused only a small and transient loss of the transmembrane potential. Taken together, our results suggests that the quinone imine-induced Ca2+ release from mitochondria is a consequence of NAD(P)H oxidation rather than GSH depletion, GSSG formation, or mitochondrial inner membrane damage.     

Effect of pH and Ca2+ on the retention of Ca2+ by rat liver mitochondria.

A transient Ca²⁺ release from preloaded mitochondria can be induced by a sudden decrease in the pH of the outer medium from 8.0 or 7.4 to 6.8. In the presence of inorganic phosphate the released Ca²⁺ is not taken up again. Upon Ca²⁺ addition to respiring mitochondria the mitochondrial membrane potential (Δ♀) decreases to a new resting level. A further decrease in Δ♀ occurs after the decrease in pH from 7.4 to 6.8, concomitant with the reuptake phase of the Ca²⁺ release. Phosphate, EGTA, and ruthenium red restore Δ♀ to its initial level. If phosphate is present initially, only transient changes in Δ♀ occur upon addition of Ca²⁺ or H⁺ ions. Only a small transient change in Δ♀ upon H⁺ ion addition is seen in the absence of accumulated Ca²⁺. La³⁺, a competitive inhibitor of Ca²⁺ transport, prevents the H⁺ ion-induced Ca²⁺ efflux, whereas this is not the case in the presence of the noncompetitive inhibitor ruthenium red. Ruthenium red, however, prevents the reuptake phase. Mg²⁺, an inhibitor of the surface binding of Ca²⁺, has no or only a slight effect on the H⁺ ion-induced Ca²⁺ release. Mitochondria preloaded with Ca²⁺ release a small fraction of Ca²⁺ during the subsequent uptake of another pulse of Ca²⁺. The results indicate that at least one pool of mitochondrial Ca²⁺ exists in a mobile state. The possible existence of a $\text{H}{}^{\mathrm{+}}\text{Ca}{}^{\mathrm{2+}}$ exchanger in the mitochondrial membrane is discussed.     

Intracellular Ca2+ transients in mouse soleus muscle after hindlimb unloading and reloading.

The objective of this study was to determine whether altered intracellular Ca(2+) handling contributes to the specific force loss in the soleus muscle after unloading and/or subsequent reloading of mouse hindlimbs. Three groups of female ICR mice were studied: 1) unloaded mice (n = 11) that were hindlimb suspended for 14 days, 2) reloaded mice (n = 10) that were returned to their cages for 1 day after 14 days of hindlimb suspension, and 3) control mice (n = 10) that had normal cage activity. Maximum isometric tetanic force (P(o)) was determined in the soleus muscle from the left hindlimb, and resting free cytosolic Ca(2+) concentration ([Ca(2+)](i)), tetanic [Ca(2+)](i), and 4-chloro-m-cresol-induced [Ca(2+)](i) were measured in the contralateral soleus muscle by confocal laser scanning microscopy. Unloading and reloading increased resting [Ca(2+)](i) above control by 36% and 24%, respectively. Although unloading reduced P(o) and specific force by 58% and 24%, respectively, compared with control mice, there was no difference in tetanic [Ca(2+)](i). P(o), specific force, and tetanic [Ca(2+)](i) were reduced by 58%, 23%, and 23%, respectively, in the reloaded animals compared with control mice; however, tetanic [Ca(2+)](i) was not different between unloaded and reloaded mice. These data indicate that although hindlimb suspension results in disturbed intracellular Ca(2+) homeostasis, changes in tetanic [Ca(2+)](i) do not contribute to force deficits. Compared with unloading, 24 h of physiological reloading in the mouse do not result in further changes in maximal strength or tetanic [Ca(2+)](i).     

Modulation of Ca2+ efflux and rebounding Ca2+ transport in rat liver mitochondria

The independent pathway for Ca2+ efflux of rat liver mitochondria exhibits a sharp temperature and pH dependence. The Arrhenius plot displays a break at 18 degrees C, activation energy being about 117 kJ/mol below 18 degrees C and 59 kJ/mol above 18 degrees C. The pH profile is bell-shaped, with a broad optimum at pH 7.0. These properties of the efflux pathway, together with the membrane potential modulation recently described (Bernardi, P. and Azzone, G.F. (1983) Eur. J. Biochem. 134, 377-383), suggest an explanation for the phenomenon of rebounding Ca2+ transport. Addition of a Ca2+ pulse to respiring mitochondria causes (i) a phase of rapid Ca2+ uptake, leading to a decrease of extramitochondrial free Ca2+ to a lower level with respect to that maintained before Ca2+ addition, and (ii) a slower phase of net Ca2+ efflux, leading to restoration of the steady-state extramitochondrial free Ca2+ preceeding Ca2+ addition. Evidence is provided that the excess Ca2+ uptake is linked to transient inactivation of the efflux pathway due to membrane depolarization. Conversely, the efflux phase is linked to reactivation of the efflux pathway upon repolarization. The efflux component of the rebound cycle and the isolated efflux pathway exhibit similar dependence on temperature, pH and membrane potential.     

Modulation of Ca2+ efflux from heart mitochondria.

The efflux of Ca2+ from rat heart mitochondria has been examined by using Ruthenium Red to inhibit active uptake after predetermined loadings with Ca2+. The efflux is proportional to the internal Ca2+ load; it is increased by Na+ applied when the mitochondria are respiring and this effect is inhibited by oligomycin. The efflux of Ca2+ is diminished by ATP and by ADP, with the latter the more effective. Both active uptake and efflux of Ca2+ are slowed by bongkrekic acid; this action has a time lag. The lower efflux found with the nucleotides and with bongkrekic acid seems to correspond to the more condensed state seen in the electron microscope when these agents are applied [Stoner & Sirak (1973) J. Cell Biol. 56, 51-64, 65-73]. The results are discussed in relation to the less-permeable state being contingent upon nucleotide binding to the membrane.