Search

Mitochondria must maintain volume homeostasis in order to carry out oxidative phosphorylation. It has been postulated that the concentration of free Mg(2+) ([Mg(2+)]) serves as the sensor of matrix volume and regulates a K(+)-extruding K(+)/H(+) antiport (K. D. Garlid. J. Biol. Chem. 255: 11273-11279, 1980). To test this hypothesis, the fluorescent probe furaptra was used to monitor [Mg(2+)] and free Ca(2+) concentration ([Ca(2+)]) in the matrix of isolated beef heart mitochondria, and K(+)/H(+) antiport activity was measured by passive swelling in potassium acetate. Concentrations that result in 50% inhibition of maximum activity of 92 microM matrix [Mg(2+)] and 2.2 microM [Ca(2+)] were determined for the K(+)/H(+) antiport. Untreated mitochondria average 670 microM matrix [Mg(2+)], a value that would permit <1% of maximum K(+)/H(+) antiport activity. Hypotonic swelling results in large decreases in matrix [Mg(2+)], but swelling due to accumulation of acetate salts does not alter [Mg(2+)]. Swelling in phosphate salts decreases matrix [Mg(2+)], but not to levels that permit appreciable antiport activity. We conclude that 1) it is unlikely that matrix [Mg(2+)] serves as the mitochondrial volume sensor, 2) if K(+)/H(+) antiport functions as a volume control transporter, it is probably regulated by factors other than [Mg(2+)], and 3) alternative mechanisms for mitochondrial volume control should be considered.

The matrix free magnesium ion concentration, [Mg2+]m, estimated using the fluorescent probe furaptra, averaged 0.67 mM in 15 preparations of beef heart mitochondria containing an average of 21 nmol total Mg2+ per mg protein. [Mg2+]m was compared with total Mg2+ during respiration-dependent uptake and efflux of Mg2+ and during osmotic swelling. In the absence of external Pi these mitochondria contain about 32 nmol/mg non-diffusible Mg-binding sites with an apparent Kd of 0.34 mM. [Mg2+]m depends on both the size of the total Mg2+ pool and the ability of matrix anions to provide Mg-ligands. Pi interacts strongly with Mg2+ to decrease [Mg2+]m and, in the absence of external Mg2+, promotes respiration-dependent Mg2+ efflux and a decrease in [Mg2+]m to very low levels. The uptake of Pi by respiring mitochondria converts delta pH to membrane potential (delta psi) and provides additional Mg-binding sites. This permits large accumulations of Mg2+ and Pi with little change in [Mg2+]m. Nigericin also converts delta pH to delta psi in respiring mitochondria and induces a large and rapid increase in both total Mg2+ and [Mg2+]m. Mersalyl increases the permeability of the mitochondrial membrane to cations and this also induces a marked increase in both total Mg2+ and [Mg2+]m. These results suggest that mitochondria take up Mg2+ by electrophoretic flux through membrane leak pathways, rather than via a specific Mg2+ transporter. Mitochondria swollen by respiration dependent uptake of potassium phosphate show decreased [Mg2+]m, whereas those swollen to the same extent in potassium acetate do not. This suggests that [Mg2+]m is well-buffered during osmotic volume changes unless there is also a change in ligand availability.

The fluorescent probe furaptra shows increases and decreases in the concentration of free magnesium ion, [Mg2+], in the mitochondrial matrix with changes in total Mg2+ and ligand availability. The factors involved in the calibration of these fluorescence changes in terms of absolute [Mg2+] have been investigated. The affinity of furaptra for Mg2+ is highly dependent on both temperature and ionic strength. The Kd for Mg-furaptra in solution in 100 mM KCl was found to be 2.1 +/- 0.1 mM at 25 degrees C. The use of this Kd to calculate matrix [Mg2+] is more reliable than in situ Kd measurements because ionophores, such as BrA23187 and ionomycin, do not equilibrate external Mg2+ with the matrix in an acceptable way. Furaptra is present at high concentrations (up to 500 microM) in the matrix when introduced by hydrolysis of the acetoxymethyl ester. However, absorbance spectra of aqueous solutions show no evidence of dimerization of the probe or other changes in properties at these concentrations. Fluorescence intensity at 340 nmex is strongly attenuated for matrix-sequestered furaptra, mag-fura-5, and mag-indo-1. This appears to result in part from preferential binding of the Mg-probe to mitochondrial proteins. The fluorescence of uncomplexed furaptra at 375-380 nmex seems unaffected by protein binding, however, and changes in intensity in this region of the spectrum can be used in conjunction with the Kd found in aqueous solution to estimate matrix [Mg2+]. The presence of secondary equilibria, such as protein binding, and possible changes in ionic strength may undermine exact quantitation by this method. However, values for matrix [Mg2+] obtained in this way (0.5 to 0.7 mM) correspond well to estimates by other available methods and each of these methods suffers from comparable uncertainties.

Despite its recognition as the most prevalent HIV associated cancer, speculation still abounds regarding the pathogenesis of AIDS-related Kaposi's sarcoma (AIDS-KS). However, it has been established that both cytokines, e.g. IL-6, and HIV-associated products, e.g., Tat, are integral in AIDS-KS cellular proliferation. Further, both experimental and clinical evidence is accumulating to link reactive oxygen intermediates (ROI) with both cytokine induction (primarily via nuclear factor-kappa B[NF-kappa B] dependent routes) as well as the subsequent cytokine, tumor necrosis factor alpha (TNF alpha) stimulation of HIV replication. Features of AIDS-KS patients, such as retention of phagocytes, presence of sustained immunostimulation, and a frequent history of KS lesions arising at traumatized sites, make oxidant stress a viable clinical factor in AIDS-KS development. Time course nucleotide profile analyses show that AIDS-KS cells have an inherent, statistically significant, biochemical deficit, even prior to oxidant stress, due to 1) a more glycolytic bioenergetic profile, resulting in lower levels of high energy phosphates (impairing capacity for glutathione [GSH] synthesis and DNA repair); 2) lower levels of NADPH (compromising the activities of GSSG reductase and peroxidase function of catalase); and 3) reduced levels of GSH (impeding both GSH peroxidase and GSH-S-transferases). Following exposure to physiologically relevant levels of H2O2, only the human microvascular endothelial cells (a putative AIDS-KS progenitor cell) responded with bioenergetic adaptations that reflected co-ordination of energy generating and cytoprotective pathways, e.g., retention of the cellular energy charge, increased NAD+, and an accentuation of the ATP, NADPH, and total adenine nucleotide differences relative to AIDS-KS cells. Also, some of the AIDS-KS strains retained intracellular GSSG subsequent to oxidant challenge, inviting the formation of deleterious protein mixed disulfides. While the results of our study address some AIDS-KS issues, they also raise an etiological question, i.e., Does the inability to tolerate oxidant stress arise in conjunction with AIDS-KS neoplastic development, or is it pre-existing in the population at risk? Regardless, use of antioxidant therapy (low risk/ potentially high benefit) in both the "at risk" population as well as in those individuals with active disease may prove a useful preventative and/or treatment modality.

Kaposi sarcoma, the most common AIDS-associated malignancy, affects 10-30% of all AIDS patients. To date, research into the biological characteristics of AIDS-related Kaposi sarcoma (AIDS-KS) derived cell lines has been based on cultures established from skin explants or pleural effusions/peritoneal fluids. We have established several AIDS-KS lines from biopsy confirmed oral mucosal and epidermal AIDS-KS lesions and have found a correlation between AIDS-KS lesional grade and in vitro cellular growth characteristics. In comparison to epidermal AIDS-KS lesions, mucosal AIDS-KS lesions frequently possessed both a more advanced histologic grade and demonstrated a greater capacity to proliferate in minimal medium. We report the ability of AIDS-KS isolates from high grade lesions to sustain proliferation (greater than 60 population doubling levels) in medium not supplemented with endothelial cell growth supplement and/or cytokine rich conditioned medium. These findings indicate that AIDS-KS cells isolated from high grade lesions have reduced requirements for exogenously provided growth supplements, and suggest that increased autologous cytokine production accompanies AIDS-KS lesional progression.

Kaposi's sarcoma (KS) is both an AIDS-defining disease and the most common HIV-associated malignancy. A cytokine-mediated pathogenesis for AIDS-KS is implicated because AIDS-KS-derived cell strains both respond to and express a variety of cytokines. We have reported the establishment of several (n = 18) AIDS-KS cell strains and determined that reduced exogenous growth factors are necessary to sustain proliferation in isolates from high histologic grade KS lesions. This current investigation explored the possibility that there are histologic grade-associated differences in either the qualitative and/or quantitative constitutive release of AIDS-KS growth stimulatory cytokines. Our findings showed that the incorporation of HTLV-II cytokine-rich conditioned media induced both qualitative and significant quantitative cytokine release, suggesting that exogenous growth promoters stimulate constitutive cytokine release. ELISA of our AIDS-KS cell strains demonstrated constitutive release of IL-6 (seven of seven), FGF-2 (five of seven), GM-CSF (three of seven), and IL-1 beta (one of seven). None of our AIDS-KS cell strains constitutively released detectable levels of Onco-M, IL-4, PDGF, TNF-alpha, or TNF-beta. In addition, we report that the method of cytokine result quantitation significantly affects reported cytokine levels. We determined that there was no significant histologic grade-dependent difference in the constitutive release of soluble cytokines by in vitro grown cultures of AIDS-KS cells. The presence of HIV influenced the sera cytokine profiles by elevating IL-6 and decreasing PDGF concentrations of HIV+ individuals relative to HIV- healthy controls. However, the presence of KS was not associated with unique serum cytokine profiles, because no differences were noted in comparisons of HIV+/KS+ versus HIV+/KS- individuals. Our findings suggest that the local environment is key in modulating AIDS-KS cytokine expression and that KS growth-promoting factors function at the local or paracrine, not the systemic, level. In conclusion, our previous results demonstrated a histologic grade-associated difference in the in vitro growth capacity of AIDS-KS cells; with high histologic grade isolates displaying a marked growth advantage during culture in minimally supplemented media. Findings from this current study reveal that although the potential for a constitutive growth loop exists in the high-grade isolates, it is not reflected in the free levels of soluble cytokines secreted into the culture medium.

Heart mitochondria contain a nNa+/Ca2+ antiport that participates in the regulation of matrix [Ca2+]. Based largely on a single study (Brand, M. D. (1985) Biochem. J. 229, 161-166), there has been a consensus that this antiport promotes the electroneutral exchange of two Na+ for one Ca2+. However, a recent study in our laboratory (Baysal, K., Jung, D. W., Gunter, K. K., Gunter, T. P., and Brierley, G. P. (1994) Am. J. Physiol. 266, C800-C808) has shown that the Na(+)-dependent efflux of Ca2+ from heart mitochondria has more energy available to it than can be supplied by a passive 2Na+/Ca2+ exchange. We have therefore re-examined Brand's protocols using fluorescent probes to monitor matrix pH and free [Ca2+]. Respiring heart mitochondria, suspended in KCl and treated with ruthenium red to block Ca2+ influx, extrude Ca2+ and establish a large [Ca2+]out:[Ca2+]matrix gradient. The extrusion of Ca2+ under these conditions is Na(+)-dependent and diltiazem-sensitive and can be attributed to the nNa+/Ca2+ antiport. Addition of nigericin increases the membrane potential (delta psi) and decreases delta pH to 0.1 or less, but has virtually no effect on the magnitude of the [Ca2+] gradient. Under these conditions a gradient maintained by electroneutral 2Na+/Ca2+ antiport should be abolished because the mitochondrial Na+/H+ antiport keeps the [Na+] gradient equivalent to the [H+] gradient. The [Ca2+] gradient is abolished, however, when an uncoupler is added to dissipate delta psi or when the exogenous electroneutral antiport BrA23187 is added. In addition, [Ca2+] influx via the nNa+/Ca2+ antiport in nonrespiring mitochondria is enhanced when delta psi is abolished. These results are consistent with Ca2+ extrusion by an electrophoretic antiport that can respond to delta psi but not with an electroneutral antiport.

We introduce two methods, both of which are based on cellular-extracellular matrix interaction, which will facilitate the study of human microvascular endothelial cells. One method describes the means to obtain a G1 population baseline in human microvascular endothelial cells. Because of the contribution of the extracellular matrix in endothelial cell growth, synchronization in G1 was possible only after the incorporation of angiostatic levels of heparin and hydrocortisone into the extracellular matrix. In the second method, we demonstrate that selective perturbation of human microvascular endothelial cell-extracellular matrix interactions results in the induction of a transitional growth state, between proliferative and differentiated growth states, in human microvascular endothelial cells. In the functional, microtubule formation assays, transitional growth state endothelial cells display rates that are indermediate between those obtained from differentiated and proliferative endothelial cells. Our results demonstrate the importance of the human microvascular endothelial cell-extracellular matrix interaction in the determination of cellular growth state. Our findings also imply that responsiveness of microvascular endothelial cells to their cellular-extracellular matrix environs is highest during the differentiated growth state.

Features of AIDS-related Kaposi's sarcoma (AIDS-KS), such as the multifocal presentation at mucosal and epidermal sites subjected to trauma, suggest that AIDS-KS is initially a reactive hyperplasia that subsequently progresses to a neoplasia. It is recognized that there is an association between sustained inflammatory states and the subsequent development of neoplasia (e.g., ulcerative colitis/colonic adenocarcinoma). Furthermore, patients who develop AIDS-KS experience both a constant immune stimulation due to sustained high levels of virus-induced cytokines and, because of a sparing effect on their phagocytic cells, retention of the phagocytic inflammatory response. A component of phagocytic activation is the initiation of the oxidative burst, resulting in the generation of reactive oxygen species (ROS), which can be mutagenic to host cells if released beyond the phagolysosome and not inactivated. Our results demonstrate that cultured AIDS-KS cells possess decreased cytoprotective capabilities. Relative to either dermal fibroblasts, or human microvascular endothelial cells (HMECs), AIDS-KS cells contained significantly lower levels of glutathione, a tripeptide integral in both cytoprotection and maintenance of cellular thiol status. While HMECs increased catalase activity during culture in the cytokine-rich KS milieu (control medium supplemented with conditioned medium from MOT, an HTLV II-infected cell line), AIDS-KS cells demonstrated reduced catalase function under these conditions. Furthermore, HMEC cultures showed an inherent biochemical responsiveness, by increasing catalase activity following exposure to exogenous H2O2. In contrast, the catalase activity of AIDS-KS cells decreased following H2O2 challenge. Our results show that an inherent deficiency in cellular cytoprotection is present in AIDS-KS cells and suggest that oxidant stress may function in the development and progression of AIDS-KS.

The pathways for the uptake and extrusion of Mg2+ by mitochondria are now well defined, the present evidence suggests that uptake occurs by nonspecific diffusive pathways in response to elevated membrane potential. There is disagreement as to some of the properties of Mg2+ efflux from mitochondria, but the reaction resembles K+ efflux in many ways and may occur in exchange for H+. Matrix free magnesium ion concentration, [Mg2+], can be measured using fluorescent probes and is set very close to cytosol [Mg2+] by a balance between influx and efflux and by the availability of ligands, such as Pi. There are indications that matrix [Mg2+] may be under hormonal control and that it contributes to the regulation of mitochondrial metabolism and transport reactions.

It is now well established that mitochondria contain three antiporters that transport monovalent cations. A latent, allosterically regulated K+/H+ antiport appears to serve as a cation-extruding device that helps maintain mitochondrial volume homeostasis. An apparently unregulated Na+/H+ antiport keeps matrix [Na+] low and the Na(+)-gradient equal to the H(+)-gradient. A Na+/Ca2+ antiport provides a Ca(2+)-extruding mechanism that permits the mitochondrion to regulate matrix [Ca2+] by balancing Ca2+ efflux against influx on the Ca(2+)-uniport. All three antiports have well-defined physiological roles and their molecular properties and regulatory features are now being determined. Mitochondria also contain monovalent cation uniports, such as the recently described ATP- and glibenclamide-sensitive K+ channel and ruthenium red-sensitive uniports for Na+ and K+. A physiological role of such uniports has not been established and their properties are just beginning to be defined.

Regulation of the mitochondrial permeability transition pore has been investigated following the release of matrix solutes which normally participate in pore regulation. Under these conditions, neither cyclosporin A nor ADP induces pore closure, as judged by restoration of delta psi, unless Mg2+ is also added. Mg2+ alone is ineffective. In liver mitochondria, the Mg2+ effect is expressed over a 0 to 0.5 mM concentration range with higher concentrations inhibiting repolarization. In heart mitochondria, the inhibitory action of high Mg2+ is not seen and it can be shown that the Mg2+ effect on repolarization increases progressively up to a concentration of 5 mM. In liver mitochondria, when the pore is closed by maximally effective concentrations of Mg2+ plus cyclosporin A or Mg2+ plus ADP, reopening occurs upon the addition of carboxyatractyloside. The latter compound, however, fails to reopen the pore when Mg2+, cyclosporin A, and ADP are present simultaneously. In heart mitochondria, where higher Mg2+ concentrations can be employed, Mg2+ plus cyclosporin A or Mg2+ plus ADP produces pore closure in a carboxyatractyloside insensitive manner. Titration experiments support the adenine nucleotide translocase as the site at which carboxyatractyloside acts to regulate the pore. However, the action of ADP appears to involve a translocase-independent site. In intact mitochondria the action of carboxyatractyloside on pore opening is counteracted by oligomycin, apparently through inhibition of the F1F0 ATP synthase, with a consequent increase in the matrix space ADP/ATP ratio. It is concluded that the permeability transition pore induced by Ca2+ plus P(i) is not formed from the adenine nucleotide translocase although the translocase conformation is one of several factors which regulate the pore. The matrix Mg2+ concentration is also one of these factors. Formation of the pore by a Ca2+ and ADP binding protein is one model which is consistent with the present data.

Net Ca2+ flux across the inner membrane of respiring heart mitochondria was evaluated under conditions in which virtually all Ca2+ movement can be attributed to the Na+/Ca2+ antiport. If this antiport promotes a passive electroneutral exchange of Ca2+ for 2Na+, the Ca2+ gradient should be equal to the square of the Na+ gradient at equilibrium. Because the mitochondrial Na+/H+ antiport equilibrates the Na+ and H+ gradients, the Ca2+ gradient should also equal the square of the H+ gradient. In a series of > 20 determinations at different matrix [Ca2+], different delta pH, and varying membrane potential, it was found that Ca2+ is transported out of the mitochondrion against gradients from 15- to 100-fold greater than the value predicted for passive electroneutral exchange. It is concluded that the observed gradients are too large to be sustained by passive Ca2+/2Na+ exchange. The observed gradients are compatible with an electrogenic Ca2+/3Na+ exchange. Alternatively another source of energy is available to support these gradients.

Many features of AIDS-related Kaposi sarcoma (AIDS-KS), e.g., multifocal lesional presentation at sites perfused by the microvasculature, suggest that AIDS-KS is initially a hyperplasia that subsequently progresses to a neoplasia. We propose that the unique AIDS environment, which contains high levels of circulating factors such as viral cytokines, is key in initiating the KS lesion. Further, we maintain that due to their physiological function, human microvascular endothelial cells (HMECs) are both likely target cells for the AIDS-related cytokines, and are putative AIDS-KS progenitor cells. Previously, we have shown that as a component of HMEC transition between proliferative and differentiated growth, HMECs modulate their nucleotide and glutathione levels. After attaining contact inhibition, HMECs enter a state of differentiation, which is characterized by cellular entrance into a G0, quiescent growth state, a decrease in cellular bioenergetic profiles, and spontaneous formation of microtubules. In contrast, when cultured in a "KS milieu", HMECs fail to differentiate. Instead, the "KS milieu" cultured cells assume a "growth relaxed" phenotype and demonstrate a lack of contact inhibition, loss of anchorage dependence, and retention of a "proliferative" bioenergetic profile despite culture confluence. Our results imply both that HMECs are responsive to AIDS-related cytokines, and that the local environment is key to instigating a relaxation of cellular growth controls.

We investigated the hypotheses that norepinephrine stimulates Mg2+ efflux from intact isolated adult rat ventricular cardiomyocytes and that adenosine 3',5'-cyclic monophosphate stimulates Mg2+ efflux from permeabilized myocytes and isolated mitochondria. Norepinephrine stimulation of Mg2+ release from cardiac myocytes was observed only when cells at approximately 20 mg protein/ml in Mg(2+)-containing buffer were diluted 50- to 60-fold into an Mg(2+)-free medium. Under these conditions, > 30% of total cellular lactic acid dehydrogenase activity was also released, indicating that a significant portion of the cells had died. In other protocols, where Mg2+ efflux from myocytes was not observed, extracellular Mg2+ removal and administration of 10 microM norepinephrine increased 45Ca2+ accumulation by cells in suspension. In single myocytes, Mg2+ removal and norepinephrine administration increased intracellular free [Ca2+] as measured by fura-2 fluorescence microscopy, and this was accompanied by vigorous spontaneous contractile activity followed by Ca2+ overload hypercontracture. With permeabilized myocytes and isolated mitochondria from a variety of sources, adenosine 3',5'-cyclic monophosphate did not stimulate Mg2+ efflux. These results suggest that recent evidence for direct hormonal regulation of myocardial Mg2+ homeostasis may need to be reevaluated.

During angiogenesis, formerly differentiated human microvascular endothelial cells (HMECs) return to a proliferative growth state. Many fundamental questions regarding HMEC function, such as how HMECs adapt to changes in bioenergetic requirements upon return to proliferative growth, remained unanswered. In this study, we evaluated whether modifications in HMEC bioenergetic profiles and glutathione (GSH) levels accompanied the cellular transition between differentiated and proliferative growth. To provide insight into the continuum of cellular adaptations that occur during this transition, we used a method recently developed in our laboratory that induces a state of morphological and functional predifferentiation in HMECs. Cellular morphology, in conjunction with flow cytometric DNA analyses and HMEC functional assays (the directed migration and intercellular association involved in microtubule formation) were employed to validate the HMEC culture state of growth. Analysis of the HPLC nucleotide profiles disclosed several findings common to all culture growth states. These uniform findings, e.g., cellular energy charges > 0.90, and highly reduced redox states, revealed that cultured HMECs maintain high rates of oxidative metabolism. However, there were also significant, culture growth state related differences in the nucleotide profiles. Proliferative HMECs were shown to possess significantly higher (relative to both large vessel endothelial cells, and differentiated HMECs) levels of GSH and specific nucleotides which were related with a return to the active cell cycle-ATP, GTP, UTP, and CTP, and NADPH. Further, the nucleotide profiles and GSH levels of the predifferentiated HMECs were determined to be intermediate between levels obtained for the proliferative and differentiated HMECs. The results of this study demonstrate that the capacity to modulate their cellular bioenergetic status during growth state transitions is one of the adaptations that enable HMECs to retain a growth state reciprocity. In addition, our findings also show that HMECs, especially during the proliferative growth state, are biochemically distinct from endothelial cells harvested from large vessels, and therefore suggest that HMECs are the cells of choice to employ when studying diseases that affect the human microvasculature.

Heart mitochondria respiring in a sucrose medium containing P(i) show a permeability transition when challenged with Ca2+ and an oxidant such as cumene hydroperoxide. The transition results from the opening of a Ca(2+)-dependent pore and is evidenced by loss of membrane potential (delta psi) and osmotic swelling due to uptake of sucrose and other solutes. In the absence of oxidant, high concentrations of Ca2+ (100-150 microM) are necessary to induce loss of delta psi and initiate swelling. Cyclosporin A delays the loss of delta psi but enhances swelling under these conditions, apparently by promoting better retention of accumulated Ca2+. Cyclosporin A and ADP together restore delta psi in respiring mitochondria that have undergone the permeability transition at levels that are not effective when either is added alone. When the state of the Ca(2+)-dependent pore is assessed using passive osmotic contraction in response to polyethylene glycol (Haworth, R. A., and Hunter, D. R. (1979) Arch. Biochem. Biophys. 195, 460-467), cyclosporin A is found to be a partial inhibitor of solute flow through the open pore. Cyclosporin A decreases the Vmax of passive contraction and increases the Km for Ca2+ without affecting the Hill slope. ADP in the presence of carboxyatractyloside closes the pore almost completely even in the presence of 40 microM Ca2+. ADP shows mixed type inhibition of the Ca(2+)-dependent pore, and cyclosporin A increases the affinity of the pore for ADP. It is concluded that cyclosporin A and ADP act synergistically to close the Ca(2+)-dependent pore of the mitochondrion and that the pore is probably not formed directly from the adenine nucleotide transporter.

Exchangeable intracellular Ca2+ as measured by 45Ca2+ uptake more than doubled when isolated adult rat ventricular cardiomyocytes were incubated 30 min with 8 microM cyclosporin; nevertheless the cells retained a normal rod-shaped morphology. High concentrations of ouabain caused a similar increase in 45Ca2+ uptake, but in this case the Ca2+ overload caused nearly all cells to hypercontract into a round disorganized form. The response to cyclosporin was concentration dependent with an apparent half-maximal effective concentration of 0.5 microM for enhancement of net 45Ca2+ accumulation. Verapamil (1 microM) could not inhibit this cyclosporin effect, but it was abolished by a 5-min preincubation with 12 microM crude ruthenium red. Cyclosporin also decreased the rate of 45Ca2+ efflux from prelabeled myocytes into Ca(2+)-containing and Ca(2+)-free media. These data are consistent with inhibition of mitochondrial 45Ca2+ efflux through the cyclosporin-sensitive mitochondrial inner membrane pore. It would appear that periodic transient increases in mitochondrial inner membrane permeability provide a pathway for mitochondrial Ca2+ extrusion under relatively normal conditions in isolated adult rat heart cells.

The concentration of free Na+ in the matrix of isolated pig heart mitochondria has been monitored using the fluorescent probe sodium-binding benzofuran isophthalate (SBFI) developed by Minta and Tsien (J. Biol. Chem. 264: 19449-19457, 1989). SBFI was sequestered in the matrix by hydrolysis of the permeant acetoxymethyl ester. The sequestered probe showed altered quantum efficiency and excitation spectra in the presence and absence of Na+ when compared with SBFI free acid in solution. Fluorescence was calibrated in situ by using ionophores to equilibrate matrix [Na+] with external [Na+]. SBFI fluorescence showed that matrix [Na+] increased linearly as external [Na+] was increased to 95 mM in the presence or absence of respiration. Respiring mitochondria maintained a Na+ gradient (Na+ out greater than Na+in) of approximately 8.0. The corresponding gradient in nonrespiring mitochondria was approximately 2.0. The Na+ gradient was nearly equivalent to the H+ gradient in the presence or absence of respiration. The uptake of Pi by respiring mitochondria decreased matrix pH and increased matrix [Na+]. It is concluded that isolated mitochondria maintain a Na+ gradient across the inner membrane as a result of the activity of the endogenous Na(+)-H+ antiport.

The effect of matrix pH (pHi) on the activity of the mitochondrial Na+/Ca2+ antiport has been studied using the fluorescence of SNARF-1 to monitor pHi and Na(+)-dependent efflux of accumulated Ca2+ to follow antiport activity. Heart mitochondria respiring in a KCl medium maintain a large delta pH (interior alkaline) and show optimal Na+/Ca2+ antiport only when the pH of the medium (pH0) is acid. Addition of nigericin to these mitochondria decreases delta pH and increases the membrane potential (delta psi). Nigericin strongly activates Na+/Ca2+ antiport at values of pH0 near 7.4 but inhibits antiport activity at acid pH0. When pHi is evaluated in these protocols, a sharp optimum in Na+/Ca2+ antiport activity is seen near pHi 7.6 in the presence or absence of nigericin. Activity falls off rapidly at more alkaline values of pHi. The effects of nigericin on Na+/Ca2+ antiport are duplicated by 20 mM acetate and by 3 mM phosphate. In each case the optimum rate of Na+/Ca2+ antiport is obtained at pHi 7.5 to 7.6 and changes in antiport activity do not correlate with changes in components of the driving force of the reaction (i.e., delta psi, delta pH, or the steady-state Na+ gradient). It is concluded that the Na+/Ca2+ antiport of heart mitochondria is very sensitive to matrix [H+] and that changes in pHi may contribute to the regulation of matrix Ca2+ levels.

Cyclosporin A prevents the opening of a nonspecific pore in the inner membrane of liver mitochondria when added prior to Ca2+. In the presence of 10 microM Ca2+ cyclosporin is unable to close the pore and restore the original permeability unless ADP is also added. ADP acts at a high-affinity site (Km 5 microM), corresponding to the adenine nucleotide transporter. This effect of ADP is prevented and reversed by carboxyatractyloside. In the presence of carboxyatractyloside, cyclosporin added with higher concentrations of ADP (Km 70 microM) also can close the pore. This suggests that a lower-affinity ADP-binding component as well as cyclophilin and the adenine nucleotide transporter can modulate the sensitivity of the pore to cyclosporin.

The effect of matrix pH (pHi) on the activity of the mitochondrial K+/H+ antiport has been studied using the fluorescence of 2,7-biscarboxyethyl-5(6)-carboxyfluorescein (BCECF) to monitor pHi and passive swelling in K+ acetate to follow antiport activity. Heart mitochondria suspended in hypotonic K+ acetate in the absence of respiration show an initial delta pH of -0.4 (interior acid) that decays slowly. Addition of A23187 to deplete matrix Mg2+ results in a further acid shift in pHi followed by equilibration of delta pH. This equilibration appears to depend on K+/H+ antiport and is slow at acid pHi but very rapid when the matrix is alkaline. Swelling of Mg(2+)-depleted mitochondria in K+ acetate is multiphasic with a slow initial rate, a period of maximum swelling, and a final period in which the rate declines. At constant external pH (pH0), the initial rate of swelling is faster with increasing pHi and the time to the onset of the maximum swelling rate decreases. The maximum swelling rate is initiated at pHi 7.4 when pH0 is 7.8 and at pHi 7.1 when pH0 is 7.4. The maximum rate of swelling increases linearly with increasing pH0 in the range from 7.0 to 8.2. This rate also shows a linear relationship to the value of pHi at the time the maximum rate is attained. Dixon plots of the reciprocal of the maximum swelling rate vs [H+]0 suggest that external [H+] is a noncompetitive inhibitor of K+ entry on the antiport. It is concluded that K+/H+ antiport in Mg(2+)-depleted heart mitochondria can be regulated by matrix [H+] (see Beavis, A. D., and Garlid, K. D. (1990) J. Biol. Chem. 265, 2538-2545), but that this antiport is also sensitive to external [H+] or to delta pH when it acts in the direction of K+ uptake.

The concentration of free Mg2+ in the matrix of isolated heart mitochondria has been monitored using the fluorescent probe furaptra. The techniques used for loading, calibrating and using furaptra fluorescence to monitor matrix free Mg2+ are described. Furaptra is loaded as the membrane-permeable acetoxymethyl ester which is hydrolyzed to the impermeant acid form by nonspecific esterases. Loading is sufficient in a 20-min period at 25 degrees C to give signals approximately 5-fold greater than nonloaded mitochondria. Uncertainties concerning the apparent dissociation constant value for Mg2+ for entrapped furaptra and the ability of ionophores to equilibrate Mg2+ across the mitochondrial membrane are discussed. We conclude that our best estimate for matrix Mg2+ in isolated mitochondria is 0.5 mM and that it can change significantly with changing Mg2+ ligand availability in the matrix.

We have developed a model for characterizing calcium handling by the intact cardiac sarcoplasmic reticulum (SR) that yields data consistent with both mathematical simulations of in situ SR Ca2+ uptake and deduced behavior of the Ca2(+)-induced Ca2+ efflux channels in mechanically skinned single cardiac cells. In Na(+)-based media (37 degrees C, pH 7.2, 50 mM Pi, 10 mM MgATP, pMg 3.3, 10 mM phosphocreatine), SR 45Ca2+ uptake by digitonin-lysed rat myocytes as a function of free [Ca2+] peaked at pCa 6.2, declined until pCa 5.6 and increased again at lower pCa. When Ca2(+)-induced Ca2+ efflux was inhibited with 30 microM ruthenium red and 10 mM procaine, uptake was saturable with a Vmax of 160 +/- 5 nmol.min-1.mg-1, K0.5 of 500 nM free [Ca2+] and slope factor of 1.6. In K(+)-based media, maximum Pi- and oxalate-supported uptake increased to 220 and 260 nmol.min-1.mg-1, respectively. Without phosphocreatine, 45Ca2+ uptake declined under all conditions; this was correlated with a decrease in ATP/ADP. Vmax for 45Ca2+ uptake was increased 20% in hyperthyroid myocytes but depressed 30% in myocytes from heart failure-prone rats. In canine myocytes, Vmax was the same as in normal rat cells, but K0.5 was 830 nM. Without efflux inhibitors, ryanodine caused a concentration-dependent decline in net Pi-supported 45Ca2+ uptake at pCa 6.3 (K0.5 = 1 microM), while 10 microM ryanodine depressed uptake at all pCa between 7.2 and 5.6. Ruthenium red/procaine fully reversed this effect.

The concentration of free Mg2+ in the matrix of isolated heart mitochondria has been monitored by using the fluorescent probe furaptra (mag-fura-2). Beef heart mitochondria respiring in a KCl medium in the absence of external Mg2+ maintain free matrix Mg2+ near 0.50 mM. Addition of Pi under these conditions decreases free Mg2+ by 0.12-0.17 mM depending on the substrate. This decrease in free Mg2+ appears to reflect changing ligand availability in the matrix. The decrease is prevented when the Pi transporter is blocked by mersalyl. Addition of ADP to initiate state 3 respiration causes a marked increase in free matrix Mg2+ (0.1-0.2 mM) that persists as long as ATP formation is taking place; free Mg2+ then returns to the base level. This cyclic change is blocked by oligomycin and carboxyatractyloside and appears to reflect to a large extent the decrease in matrix Pi that accompanies oxidative phosphorylation. Exchange of external ADP for matrix ATP may also contribute to the increase in free matrix Mg2+. Addition of an uncoupler promotes anion efflux and increases free matrix Mg2+. Similar changes in free Mg2+ on addition of Pi, ADP, or uncoupler are seen when extramitochondrial Mg2+ is buffered from 0.5 to 2 mM, but the basal free matrix Mg2+ increases as external Mg2+ concentration increases in this range. Free matrix Mg2+ also increases when total mitochondrial Mg2+ is increased by respiration-dependent uptake in the presence of Pi. It is concluded that matrix free Mg2+ changes significantly with changing ligand availability and that such changes may contribute to the regulation of Mg2(+)-sensitive matrix enzymes and membrane transporters.

Freshly isolated adult rat heart myocytes contain total glutathione and reduced glutathione (GSH) at levels quite comparable to those in intact rat heart. Total glutathione can be depleted from 11 to 1 nmol/mg protein or less by treatment with cyclohex-2-ene-1-one without effect on either cellular ATP, rod-cell morphology or the integrity of the sarcolemma. Glutathione levels and redox state are not altered significantly when the Ca-tolerant, quiescent cells are subjected to a period of anoxia followed by reoxygenation. This oxygen paradox protocol results in irreversible hypercontracture of the contractile elements into an amorphous mass in the bulk of the cells, but little loss of sarcolemmal integrity. When the myocytes are subjected to an externally applied oxidant stress by the addition of either diamide or t-butylhydroperoxide, GSH is rapidly depleted with accumulation of oxidized glutathione (GSSG. On continued aerobic incubation both of these reagents promote a slower depletion of cellular ATP and a parallel hypercontracture. Cells treated with t-butylhydroperoxide, but not those with diamide, also generate increasing amounts of thiobarbituric acid reactive species as an indication of lipid peroxidation and show a parallel loss of sarcolemmal integrity. It is concluded that respiring myocytes and those subjected to the oxygen paradox do not produce oxygen radicals in sufficient amounts to displace the GSH/GSSG redox poise and depletion of myocyte glutathione per se is not detrimental to the short term survival of the cells. In addition, aerobic myocytes subjected to external oxidant stress can be damaged irreversibly by two pathways, a hypercontracture that correlates with depletion of ATP and a loss of sarcolemmal integrity that correlates with lipid peroxidation.

The fluorescence of 2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein (BCECF) has been used to follow a K(+)-dependent alkaline shift in the matrix pH (pHi) of isolated heart mitochondria. The K(+)-dependent pHi change has properties consistent with an inward exchange of K+ for matrix H+ on the K+/H+ antiport of the mitochondrion. The reaction is activated by depletion of matrix Mg2+ with A23187 and by an alkaline external pH (pHo) and hypotonic conditions. The exchange is inhibited by quinine, dicyclohexylcarbodiimide, and exogenous Mg2+, but not by Li+. The rate of K+/H+ antiport measured in this way increases with increasing pHo to a maximum near pHo 9. The rate is a hyperbolic function of [K+] at pHo values above 8.3 with an apparent Km of 30 mM at pHo 8.4 and 14 mM at pHo 8.8. External H+ acts as a mixed-type inhibitor of the K+/H+ antiport under these conditions with a Ki equivalent to pHo 8.6-8.8. When pHo is kept constant, the reaction is relatively insensitive to matrix pH (pHi) in the range from 7.0 to 7.5. Above this pHi, the K(+)-dependent H+ extrusion shows a hyperbolic dependence on [H+]i with an apparent Km equivalent to pHi 8.1. The activated antiport shows an affinity sequence of Li+ greater than K+ = Rb+ greater than Cs+. The inward antiport of K+ is inhibited noncompetitively by NH4+ and is also sensitive to benzamil and to 5-N-substituted amiloride analogues with I50 values near 20 microM. Both NH4+ and the amiloride analogues increase pHi at constant pHo and appear to be concentrated in the matrix under these conditions.

Mitochondria isolated from potato tubers fail to undergo passive osmotic swelling when suspended in isotonic Na(+) acetate or phosphate, in NaCl following addition of tripropyltin, or in Na(+) nitrate following addition of an uncoupler. Swelling under each of these conditions in mitochondria from other sources has been attributed to the inward movement of Na(+) on an endogenous Na(+)/H(+) exchanger. Such a monovalent cation/H(+) exchanger has also been implicated in respiration-dependent cation extrusion and contraction of swollen mitochondria. Potato mitochondria swollen in chloride and nitrate salts extrude ions and contract when respiration is initiated. The contraction reaction is slower and less efficient than that in beef heart mitochondria, but like the latter, is sensitive to uncouplers and stimulated by nigericin, butacaine, and Mg(2+). These comparative studies suggest that a cation(+)/H(+) exchanger is present in potato tuber mitochondria, but that it functions exclusively as a cation-extruding mechanism. They further suggest that cation(+)/H(+) exchange activity is not identical in mitochondria from different sources and that these exchange components may have a directionality and regulatory features which differ with the metabolic needs of the source tissue.

A rapid deenergization procedure was used to probe the regulation of in situ adenylate deaminase and 5'-nucleotidase in isolated adult rat heart cells. In cells depleted of ATP, the rate of ionosine monophosphate (IMP) production was fourfold greater in cells that had been respiring prior to deenergization than in cells that had been maintaining ATP stores through anaerobic glycolysis. This effect of respiratory inhibition was fully reversed by reaeration. When phenylephrine was present during preincubation, IMP production during a subsequent 5-minute rapid deenergization was increased by 70% in respiring cells and by 88% in those that had not been respiring. These effects of phenylephrine were abolished by prazosin. Adenosine production by cells without ATP was inversely related to that of IMP, whereas it was positively correlated with the amount of AMP remaining in cells after 5 minutes. We conclude from these data that rat heart adenylate deaminase is regulated by a product(s) of anaerobic glycolysis and by alpha 1-adrenergic stimulation. The production of intracellular adenosine in cells without ATP, on the other hand, is governed primarily by the concentration of AMP and appears to be catalyzed by the cytosolic type I 5'-nucleotidase.

Respiring heart mitochondria exchange matrix 42K+ with extramitochondrial K+ at a rapid rate in the presence of Pi (Chávez, E., Jung, D. W., and Brierley, G. P. (1977) Arch. Biochem. Biophys. 183, 460-470, 1977). This exchange reaction is strongly inhibited by uncouplers. However, under two rather similar sets of conditions, the addition of an uncoupler results in a rapid, transient increase in the exchange of matrix 42K+ with external K+ when the mitochondria are suspended in KCl or, alternatively, in a net loss of matrix 42K+ from mitochondria suspended in K+-free media. These conditions are: (a) the addition of an uncoupler to respiring mitochondria after the accumulation of a small amount of phosphate salt, and (b) the presence of a Ca2+-chelator or ruthenium red with uncoupler. Loss of 42K+ under these conditions occurs with all substrates tested, is completely blocked by rotenone, and is accompanied by an almost complete oxidation of both NADH and NADPH. In the presence of rotenone and acetoacetate, only NADH is oxidized and 42K+ efflux does not occur. It is concluded that simply dissipating the mitochondrial protonmotive force by addition of an uncoupler is not sufficient to induce release of mitochondrial K+. Uncoupler-induced oxidation of mitochondrial NADPH, in conjunction with elevated internal Pi, opens a rather nonspecific pathway for K+ loss which can be inhibited by ADP and enhanced by Ca2+. The more specific loss of K+ which occurs in the absence of elevated internal Pi when uncoupler and EGTA or ruthenium red are present suggests that K+ efflux is related to the Ca2+-uniporter. Loss of K+ by either of these pathways can be differentiated from efflux of K+ on the endogenous K+/H+ exchanger which functions without dissipation of the mitochondrial membrane potential.

Heart mitochondria swollen passively in nitrate salts contract in a respiration-dependent reaction which can be attributed to an endogenous cation/H+ exchange component (or components). The rate of contraction increases with increased extent of passive swelling in both Na+ and K+ salts. Since nearly constant internal cation concentrations are maintained during osmotic swelling, this result suggests that both Na+/H+ and K+/H+ exchange is enhanced by increased matrix volume. Endogenous Mg2+ is also lost with increased matrix volume, and this observation, in conjunction with other evidence available in the literature, suggests that monovalent cation/H+ exchanges may be regulated by divalent cations. Passive exchange of Na+/K+, 42K+/K+, and 24Na+/Na+ can be readily demonstrated in mitochondria swollen in nitrate. All these exchanges are low or not detectable in unswollen control mitochondria, and it appears that they are manifestations of the activated cation/H+ component (or components) functioning in the absence of delta pH.

The Na+-induced release of accumulated Ca2+ from heart mitochondria is inhibited by amiloride, benzamil and several other amiloride analogues. These drugs do not affect uptake or release of Ca2+ mediated by the ruthenium red-sensitive uniporter and their effects, like those of diltiazem and other Ca2+-antagonists, appear to be localized principally at the Na+/Ca2+ antiporter of the mitochondrion. Benzamil inhibits Na+/Ca2+ antiport non-competitively with respect to [Na+] with a Ki of 167 microM. In the presence of 1.5 mM Pi the Ki for benzamil inhibition of this reaction is decreased to 87 microM.

1. Ionophore A23187 depletes the endogenous Mg2+ of beef heart mitochondria to 4 ng ions . mg-1 (or less) and Ca2+ to less than 2 ng ions . mg-1 in a reaction which does not require metabolic energy. Ionophore A23187 also promotes the near quantitative extrusion of endogenous K+ from heart mitochondria to a K+-free medium, but this reaction requires respiration and is sensitive to uncouplers. Respiring mitochondria which extrude K+ in the presence of A23187 maintain high membrane potentials and do not swell. The K+ depletion has the properties of a K+/H+ exchange reaction occurring at the expense of delta pH of respiration. 2. The rate and extent of K+ efflux are dependent on the removal of endogenous divalent cations by A23187, but are otherwise independent of the amount of A23187 present. It is concluded that K+ efflux results from activation of an endogenous K+/H+ exchange component and that A23187 does not function directly as a K+/H+ exchanger in mitochondria.

The uncoupler-induced release of accumulated Ca2+ from heart mitochondria can be separated into two components, one sensitive and one insensitive to ruthenium red. In mitochondria maintaining reduced NAD(P)H pools and adequate levels of endogenous adenine nucleotides, the release of Ca2+ following addition of an uncoupler is virtually all inhibited by ruthenium red and can be presumed to occur via reversal of the Ca2+ uniporter. When ruthenium red is added to block efflux via this pathway, high rates of Ca2+ efflux can still be induced by an uncoupler, provided either NADH is oxidized or mitochondrial adenine nucleotide pools are depleted by prior treatment. This ruthenium red-insensitive Ca2+-efflux pathway is dependent on the level of Ca2+ accumulated and is accompanied by swelling of the mitochondria and loss of endogenous Mg2+. Loss of Ca2+ by this relatively nonspecific pathway is strongly inhibited by Sr2+ and by nupercaine, as well as by oligomycin and exogenous adenine nucleotides. The loss of Ca2+ from uncoupled heart mitochondria occurs via a combination of these two mechanisms except under conditions chosen specifically to limit efflux to one or the other pathway.

The permeability of rat heart myocytes to Na increases when extracellular Ca (Ca0) is decreased. This increased permeability is reflected in elevated Na/K ratios in nonenergized myocytes and in increased ouabain-sensitive lactate production in anaerobic myocytes supplemented with glucose. Myocytes treated with ethylene glycol bis(beta-aminoethyl ether)N,N'-tetraacetic acid (1 mM) maintain low Na/K ratios, but expend considerable glycolytic ATP in ouabain-sensitive cation cycling. The data suggest that Ca0 bound to the sarcolemma can restrict transmembrane movement of Na via pathways that are not yet defined. The lack of significant net accumulation of Ca argues against the explanation that Ca0 maintains low internal Na levels as a result of Na-Ca exchange. Both the increased uptake of Na and increased utilization of ATP in the absence of Ca0 may be relevant to the phenomenon of "Ca-paradox" in situ.

Isolated beef heart mitochondria were treated with A23187 in the presence of different concentrations of Mg2+ or EDTA to establish varying levels of total mitochondrial Mg2+. The Mg2+ content was related to the rate of passive swelling of the mitochondria in potassium acetate and other potassium salts in which swelling is presumed to depend on K+ entry via an endogenous K+/H+ antiport. Swelling in these salts does not commence until Mg2+ has been depleted from an initial value of 36 nmol X mg-1 of protein to 8 nmol/mg-1, or less. Below this level, swelling increases linearly with decreasing Mg2+ to a maximum rate at 2 nmol of Mg2+ X mg-1. Rotenone-treated heart mitochondria suspended in 75 mM potassium acetate at pH 7.80 show no delta pH by 5,5-dimethyl-2,4-oxazolidinedione distribution. Distribution of methylamine also shows essentially no delta pH under these conditions when allowance is made for binding of [14C]methylamine by mitochondrial membranes under these conditions. Addition of A23187 results in a small and transient delta pH (delta pH less than 0.14, acid interior) as measured by methylamine distribution. Estimation of the maximum matrix free Mg2+ concentration from the maximum delta pH observed and the external free Mg2+ concentration at equilibrium with A23187 shows that swelling is not initiated until matrix free Mg2+ is decreased to below 150 microM. An independent estimate of free Mg2+ using a null-point procedure gives a lower, but quite similar value (50 microM) for maximum matrix free Mg2+ when swelling commences. The large depletion of total and free Mg2+ that is required to activate swelling in potassium acetate (and presumably K+/H+ antiport activity) does not appear to be compatible with previous indications that free Mg2+ acts as a "carrier brake" to regulate K+ extrusion from the mitochondrion on such an antiport (Garlid, K. D. (1980) J. Biol. Chem. 255, 11273-11279). The removal of a tightly bound component of mitochondrial Mg2+ is closely related to increased K+ permeability and increased passive swelling in potassium salts. This Mg2+ appears to play a role in the maintenance of mitochondrial membrane structure and integrity.

Responses of isolated adult rat heart cells to conditions that emphasize various aspects of ischemia have been evaluated. Cells maintained in hypoxic media with limited substrate deteriorate more rapidly than aerobic controls supplemented with glucose. Two distinct irreversible pathways for cell alteration can be distinguished as follows: (1) continued anaerobic aging in the absence of glucose results in the production of large numbers of cells which retain the rod-shaped morphology of heart cells in situ, but which have lost sarcolemmal integrity, and (2) after a period of anaerobic aging, reaeration of the cells produces large numbers of rounded cells in irreversible contracture. These cells maintain an intact sarcolemma and are indistinguishable from those produced by addition of 1 mM Ca2+ to Na+-loaded, aerobic cells. Contracture of isolated cells on re-aeration is at least superficially analogous to the oxygen paradox in situ, but since the isolated cells maintain an intact sarcolemma, there is no loss of creatine phosphokinase or other components of the cytosol. Incubation of isolated heart cells at acid pH (pH 6.8 to 6.2) largely prevents both Ca2+-dependent contracture and a Ca2+- dependent loss of respiratory capacity. The acidic conditions virtually eliminate the net influx of 45Ca2+ into isolated cells that occurs at neutral pH, and the inhibition appears to be localized at the sarcolemma.

Mitochondria contain a latent K+/H+ antiporter that is activated by Mg2+-depletion and shows optimal activity in alkaline, hypotonic suspending media. This K+/H+ antiport activity appears responsible for a respiration-dependent extrusion of endogenous K+, for passive swelling in K+ acetate and other media, for a passive exchange of matrix 42K+ against external K+, Na+, or Li+, and for the respiration-dependent ion extrusion and osmotic contraction of mitochondria swollen passively in K+ nitrate. K+/H+ antiport is inhibited by quinine and by dicyclohexylcarbodiimide when this reagent is reacted with Mg2+-depleted mitochondria. There is good suggestive evidence that the K+/H+ antiport may serve as the endogenous K+-extruding device of the mitochondrion. There is also considerable experimental support for the concept that the K+/H+ antiport is regulated to prevent futile influx-efflux cycling of K+. However, it is not yet clear whether such regulation depends on matrix free Mg2+, on membrane conformational changes, or other as yet unknown factors.

Myocyte hypercontracture can be produced by adding Ca2+ to calcium-intolerant myocytes. A similar morphologic change occurs in Ca2+-free media when anoxic, ATP-depleted myocytes are reoxygenated or when respiring myocytes are lysed with digitonin. Hypercontracture in Ca2+-free media is abolished by rotenone, an inhibitor of NADH-linked respiration. Rotenone-treated, digitonin-permeabilized myocytes were used to examine the effects of MgATP, pCa, and respiration on hypercontracture. In the absence of Ca2+ (pCa 8.5), hypercontracture occurred at low MgATP but not when ATP was increased above 1 mM. At high MgATP (1-10 mM), hypercontracture was Ca2+-dependent. Succinate did not cause hypercontracture in the absence of added MgATP, but it shifted the concentration dependence for Ca2+-independent hypercontracture to lower values by regenerating ATP.

Isolated heart mitochondria hydrolyze the acetoxymethyl esters of the Ca2+-sensitive fluorescent probe fura-2 and the pH-sensitive 2',7'-biscarboxyethyl-5(6)-carboxyfluorescein (BCECF). The resulting charged forms of the probes are retained in the mitochondrial matrix and appear well-suited for the estimation of pCa and pH in this compartment. The mitochondria esterase activity is stimulated by Ca2+, inhibited by butacaine and quinine, and shows an alkaline pH optimum. The esterase has a similar affinity for the two probes (Km about 1.5 microM) and a somewhat higher Vmax for BCECF. Intramitochondrial pH can be determined by recording the ratio of the fluorescence of matrix BCECF at its excitation maximum of 509 nm to that at 450 nm, an excitation wavelength that is unresponsive to pH. A calibration plot relating the fluorescence ratio to pH is constructed using detergent-lysed mitochondria and the excitation maximum of 500 nm for BCECF in aqueous solution. Estimates of matrix pH by BCECF fluorescence in its useful range (pH 6 to 8) agree well with values obtained using the distribution of 5,5-dimethyl-2,4-oxazolidenedione. In protocols in which the fluorescence with excitation at 450 nm does not vary, a direct recording of BCECF fluorescence with excitation at 509 nm can be used to follow the kinetics of matrix pH changes.

Progesterone stimulates the rate and efficiency of respiration-dependent contraction of heart mitochondria in nitrate salts at alkaline pH. Ion extrusion under these conditions is normally slow and inefficient due to the elevated permeability of the membrane to monovalent cations through a putative uniport pathway. Progesterone also inhibits passive swelling under these conditions and appears to act by restricting cation influx through the uniport pathway.

A one hour hypoxic incubation causes the release of a small but significant amount of cytosolic lactic dehydrogenase from glucose-deprived isolated adult rat heart myocytes. However, enzymes associated with the mitochondria are not liberated, and there is no increase in the number of hypercontracted cells. These observations led Piper et al. (Life Sciences 35, 127-134 [1984]) to conclude that reversibly injured myocytes can release cytosolic proteins. This conclusion was based on the supposition that irreversibly hypoxic injury must cause mitochondrial enzyme efflux and hypercontracture. The present study establishes that this supposition is invalid.

Isolated heart mitochondria retain matrix K+ in a K+-free medium and exchange matrix 42K+ with external K+ only slowly. This low permeability to K+ is maintained when the proton motive force is dissipated by addition of an uncoupler, but can be increased markedly in uncoupled mitochondria when (a) NADPH becomes oxidized and (b) a Ca2+ chelator or ruthenium red is added (Jung, D. W., and Brierley , G. P. (1981) J. Biol. Chem. 256, 10490-10496). This latter requirement suggests that decreased Ca2+ binding or alteration of the Ca2+ uniporter may be involved in the induction of permeability to K+ in these mitochondria. The present studies establish that La3+ (k0.5 = 1.8 nmol X mg-1 of protein) also induces K+ permeability in uncoupled mitochondria in which NADPH has been oxidized. The amount of net K+ loss or passive 42K+/K+ exchange induced by La3+ corresponds to that produced by ruthenium red or EGTA and appears to vary from preparation to preparation as a function of the endogenous adenine nucleotide (AN) content of the mitochondria. The permeability to K+ induced by all three reagents is increased by depletion of endogenous AN with PPi and strongly inhibited by low levels of exogenous AN. The optimum passive permeability to K+ develops at pH 7.5, is inhibited by Nupercaine , quinine, and dicyclohexylcarbodiimide, and is increased in a sucrose, as opposed to a KCl medium. The increased permeability to K+ appears to result from the opening of one or more K+-conducting uniport pathways, rather than K+/H+ exchange. Since Ca2+ efflux remains sensitive to ruthenium red when K+ efflux is induced, it seems unlikely that the Ca2+ uniporter itself can provide a pathway for K+ flux. The presence of such latent pathways for passive K+ permeability must be considered when defining the properties of the putative K+/H+ antiporter and during isolation and reconstitution protocols involving mitochondrial K+ transport components.

The release of lactic dehydrogenase, creatine phosphokinase, and aspartate aminotransferase from initially viable, metabolically competent adult rat heart myocytes has been examined. Freshly isolated cells contain levels of total lactic dehydrogenase, creatine phosphokinase, and aspartate aminotransferase, as well as lactic dehydrogenase and creatine phosphokinase isoenzyme profiles that are quite comparable to those of intact heart tissue. When the cells are lysed with digitonin, 89% of total lactic dehydrogenase, but only 58% of creatine phosphokinase and 27% of aspartate aminotransferase are released. The retention of creatine phosphokinase by the digitonin-lysed cells is accounted for by complete retention of mitochondrial creatine phosphokinase and 20% of MM-creatine phosphokinase. When intact myocytes are incubated under anoxic, substrate-deprived conditions, there is a gradual loss of the three enzymes to the suspending medium and a parallel increase in the fraction of the cells permeable to trypan blue. The fraction of freely soluble cytoplasmic enzymes lost was equivalent to the fraction of the cells permeable to the dye over a wide range of viability (17-95% viable by dye exclusion criteria), but permeable cells retained mitochondrial creatine phosphokinase and particulate aspartate aminotransferase. These results suggest that simultaneous and complete release of soluble cytoplasmic enzymes occurs as each individual cell sustains sarcolemmal damage.