Phosphatidylinositol metabolism during in vitro hypoxia

Differential alteration of catecholamine release during chemical hypoxia is correlated with cell toxicity and is blocked by protein kinase C inhibitors in PC12 cells

Release of neurotransmitters, including dopamine and glutamate, has been implicated in hypoxia/ischemia-induced alterations in neuronal function and in subsequent tissue damage. Although extensive studies have been done on the mechanism underlying the changes in glutamate release, few have examined the mechanism that is responsible for the changes in catecholamines. Rat pheochromocytoma-12 (PC12) cells synthesize, store, and release catecholamines including DA and NE. Therefore, we used HPLC and ED to evaluate extracellular DA and NE concentrations in a medium during chemical hypoxia in PC12 cells. Chemical hypoxia produced by KCN induced differential release of DA and NE. Under normal glucose conditions, KCN induced release of NE, but not DA. Under glucose-free conditions, KCN-induced release of DA was elevated transiently, whereas the release of NE increased progressively. Under parallel conditions, KCN biphasically elevated the level of cytosolic free calcium ([CA(2+)](i)) in glucose-free DMEM, peaking at 95 +/- 18 nM at 1,107 +/- 151 s, followed by a new plateau level at 249 +/- 24 nM sustained from 4,243 +/- 466 to 5,263 +/- 440 s. Cell toxicity, as measured by LDH release, was increased significantly by KCN in glucose-free DMEM but was diminished in the presence of glucose, and was correlated with DA release by chemical hypoxia. The protein kinase C (PKC) inhibitor GO6976 or staurosporine inhibited KCN-induced LDH release as well as the release of NE and DA. Taken together, selective activation of DA but not NE was correlated with the LDH release by chemical hypoxia, and was diminished with GO6976 or staurosporine. These results suggest that selective activation of PKC isoforms is involved in the chemical hypoxia-induced DA release, which may lead to neuronal cell toxicity.

Phospholipid composition and levels are not altered in fibroblasts bearing presenilin-1 mutations

Lipid alterations have been reported in brain regions affected by Alzheimer disease (AD). The mechanisms causing these changes are poorly understood because it is difficult to study dynamic, biochemical processes in post-mortem brain. Fibroblasts derived from AD patients offer an alternative model to study disease-related alterations in lipid metabolism. Therefore, we measured the phospholipid levels and composition of fibroblasts from individuals bearing two different presenilin-1 mutations and compared these values to appropriate control fibroblasts. There were no differences between groups in phospholipid composition or in individual phospholipid levels, including the plasmalogens. Cholesterol levels and the cholesterol/phospholipid ratio were not different between presenilin-1 mutation bearing and control fibroblasts. Although these presenilin-1 mutation bearing fibroblasts have a number of biochemical changes related to AD, the absence of a change in phospholipid levels suggests that under these conditions, these cells are not useful in studying the mechanisms underlying the alterations in brain phospholipid levels associated with AD. However, these results do not preclude the possible use of other fibroblasts bearing AD-related mutations, e.g., APP mutations, to examine AD-related changes in brain lipid metabolism, or of these fibroblasts under different conditions.

Selective subcellular redistributions of protein kinase C isoforms by chemical hypoxia

The mechanisms of neuronal degeneration following hypoxia/ischemia remain undefined, but the processes include increases in neurotransmitter release, elevation of cytosolic-free calcium concentration, and changes in signal transduction pathways. Activation of the multigene family of protein kinase C (PKC) has been associated with the release of neurotransmitter and the survival of neurons. Therefore, to understand which PKC isozymes are involved in hypoxia/ischemia-induced neuronal degeneration, we examined PKC isozymes after chemical hypoxia (i.e., KCN exposure) in PC12 cells. Cell toxicity, as measured by lactate dehydrogenase (LDH) release, was increased significantly by KCN in glucose-free DMEM and was exaggerated by acute 12-O-tetradecanoyl phorbol-13-acetate (TPA) pretreatment. Under parallel conditions, KCN elevated cytosolic-free calcium ([Ca2+]i) in glucose-free but not in glucose containing DMEM, and TPA pretreatment did not exaggerate KCN's effect on [Ca2+]i. Thus, increases in [Ca2+]i are not sufficient for the synergistic toxic effect of KCN and TPA. In the glucose-free DMEM, selective PKC isozyme inhibitor Go 6976 at 10 nM completely inhibited KCN-induced LDH release and at higher concentrations (1 microM) inhibited the basal levels of LDH release. The protein levels of PKCs in the nuclear, membrane, and cytosolic fractions were measured by Western blot analysis using antibodies against specific isoforms. Two Ca2+-dependent (-alpha, -gamma) and four Ca2+-independent (-delta, -epsilon, -zeta, and -lambda) isozymes were identified and two isozymes (-beta and -theta) were not detected in the subcellular fractions of PC12 cells. Treatment of the cells with TPA significantly activated translocation of conventional PKC-gamma from the cytosol to the membrane and nuclear fractions and other PKC isozymes (-alpha, -delta, and -epsilon) from the cytosol to the membrane, but not atypical PKC-zeta and -lambda. Although only the levels in the nuclear PKC-gamma but not other PKC isozymes were increased significantly following KCN, the levels of cPKC-alpha and -gamma in the membrane mainly- and those and PKC-epsilon in the nucleus-were increased when KCN was combined with TPA. In addition, this condition (TPA + KCN) did not affect the TPA insensitive atypical isozymes, PKC-zeta or -lambda. Taking the results together, differential activation/translocation of PKC isozymes by KCN and TPA is important in the regulation of chemical hypoxia-induced cell injury in PC12 cells.

Mechanisms of cerebral injury in perinatal asphyxia and strategies for prevention

We have investigated the mechanisms of hypoxic brain cell injury in the immature animal by examining (1) the role of excitatory amino acid neurotransmitter receptors, (2) the receptor-mediated increase in intracellular Ca2+, and (3) the generation of oxygen free radicals. We examined the effect of brain tissue hypoxia on the NMDA receptor-ion channel complex including the glutamate, Mg2+, spermine, CPP, and the non-NMDA receptor kainate sites. Brain tissue hypoxia resulted in modification of the NMDA receptor ion channel and its modulatory sites. Hypoxia increased the affinity of both the ion channel and the glutamate recognition site. Pretreatment of animals with the glutamate antagonist CPP prevented hypoxia-induced modification of the channel. Similarly, pretreatment of animals with Mg2+, a blocker of the NMDA receptor ion channel, prevented the hypoxia-induced modification of the receptor. In addition, an increased agonist-dependent entry of Ca2+ into synaptosomes was observed in hypoxic animals compared with normoxic animals. Increased free radical generation in the cerebral cortex during hypoxia was demonstrated using spin labeling technique and electron spin resonance spectroscopy. We conclude that hypoxia-induced modification of the NMDA receptor-ion channel complex leads to increased intracellular Ca2+ potentiating free radical generation and resulting in hypoxic cell injury.

A protein kinase C inhibitor attenuates cyanide toxicity in vivo

We have examined the effect of pretreatment with a potent protein kinase C (PKC) inhibitor, 1-(5-isoquinoline-sulfonyl)-2-methylpiperazine (H-7), against metabolic alterations induced by sodium cyanide (NaCN), 4.2 mg/kg, in brain of anesthetized male micropigs (6-10 kg). Brain high energy phosphates were analyzed using a 31P nuclear magnetic resonance (NMR) spectroscopic surface coil in a 4.7 Telsa horizontal bore magnet. H-7, 1 mg/kg, was given intravenously (i.v.) 30 min before NaCN challenge (H-7 + CN-). Prior to NaCN, H-7, or H-7 + CN- administration, baseline 31P resonance spectra of 1-min duration were acquired for 5-10 min, and continued for an additional 60 min following i.v. NaCN injection, each animal serving as its own control. Peaks were identified as phosphomonoester (PME), inorganic phosphate (Pi), phosphodiester (PDE), phosphocreatine (PCr) and adenosine triphosphate (ATP), based on their respective chemical shifts. Without H-7 pretreatment, NaCN effects were marked by a rising Pi and a declining PCr peak 2 min after injection, with only 2/5 of the animals surviving the 60 min experiment. Through a pretreatment period of 30 min, H-7 did not affect baseline cell energy profile as reflected by the 31P-NMR spectra, but in its presence, those changes (i.e. diminishing PCr and rising Pi peaks) elicited by NaCN were markedly blunted; 4/5 of the animals in this group survived the NaCN challenge. It is proposed that H-7, a pharmacologic inhibitor of PKC, may be useful in CN- antagonism, underscoring the role of PKC in cyanide intoxication.

Cytosolic free calcium and gene expression during chemical hypoxia

Understanding the cellular response to hypoxia may help elucidate the role of altered oxidation in neuronal death or abnormal cell function. In PC12 cells, 30 min of chemical hypoxia (i.e., KCN) reduced ATP concentrations by 92%, but diminished viability by only 10%. Ten minutes of hypoxia increased cytosolic free calcium ([Ca2+]i) 2.5-fold above control, but after 30 min of hypoxia, [Ca2+]i was slightly below that of nonhypoxic cells. Short periods of hypoxia also exaggerated the K(+)-induced elevation of [Ca2+]i, but by 30 min these ATP-depleted cells reestablished a calcium gradient that was equal to nonhypoxic, K(+)-depolarized cells. Thus, 30 min of severe ATP depletion left [Ca2+]i and viability relatively unaffected. Nerve growth factor caused slight, but significant, improvements in ATP and viability of hypoxic cells, but had no effect on [Ca2+]i. Although [Ca2+]i was equivalent in control and hypoxic cells after 30 or 60 min, hypoxia abolished the K(+)-stimulated elevation of [Ca2+]i. The nerve growth factor induction of c-fos, an indicator of the genomic response, was diminished by approximately 80%. Thus, hypoxic PC12 cells with greatly reduced ATP stores maintained normal [Ca2+]i, but their ability to respond to external stimulation was impaired. Further, the reduced oxidation that occurs in the brain in a variety of pathological conditions may interfere with the cellular response to stimulation and growth factors.

Mechanisms of asphyxial brain damage, and possible pharmacologic interventions, in the fetus

An examination of the cellular and molecular mechanisms of neuronal cell damage may lead to the design of pharmacologic interventions during presumed or actual fetal asphyxia. Hypoxia-ischemia in its severest form results in insufficient adenosine 5'-triphosphate production. The most important effect of this is failure of adenosine 5'-triphosphate-dependent membrane functions, which maintain ionic homeostasis, that is, ionic pumping. There is K+ efflux and Na+ influx across the cell membrane, depolarization of the cell membrane, opening of the voltage-dependent calcium channels, and entrance of Ca++ into the cell. Cytosolic Ca++ is also increased by Ca++ efflux from the mitochondria and the sarcoplasmic reticulum. Ca++ is a toxin in high cytosolic concentrations; it activates phospholipases A and C, which cause membrane breakdown and release of free fatty acids, including arachidonic acid. The membrane is damaged, lysis occurs, and the neuron dies. High cytosolic Ca++ also causes release of excitatory amino acids (especially glutamate), which overwhelm the suppressant neurotransmitters, causing seizures, increased metabolism, and aggravation of the insufficient adenosine 5'-triphosphate availability. Thromboxane A2 is generated from arachidonic acid, increasing smooth muscle tone and thereby worsening the ischemia. Cyclooxygenase activity also results in formation of oxygen-free radicals that contribute to cell membrane damage, lysis, and death. Possibilities for pharmacologic interventions include (1) calcium channel blockers and antagonists, (2) excitatory neurotransmitter blockers, (3) oxygen-free radical scavengers (e.g., superoxide dismutase), (4) cyclooxygenase or prostaglandin synthesis inhibitors, and (5) seizure suppressants (e.g., phenobarbital). Some of these treatments have been shown experimentally to limit neuronal death in the adult and fetus, and after more investigative work they may be applicable to clinical practice.

Cytosolic free calcium concentrations in synaptosomes during histotoxic hypoxia

Altered cytosolic free calcium concentrations ([Ca2+]i) accompany impaired brain metabolism and may mediate subsequent effects on brain function and cell death. The current experiments examined whether hypoxia-induced elevations in [Ca2+]i are from external or internal sources. In the absence of external calcium, neither KCl depolarization, histotoxic hypoxia (KCN), nor the combination changed [Ca2+]i. However, with external CaCl2 concentrations as small as 13 microM, KCl depolarization increased [Ca2+]i instantaneously while hypoxia gradually raised [Ca2+]i. The combination of KCN and KCl was additive. Increasing external calcium concentrations up to 2.6 mM exaggerated the effects of K+ and KCN on [Ca2+]i, but raising medium calcium to 5.2 mM did not further augment the rise. Diminishing the sodium in the media, which alters the activity and perhaps the direction of the Na/Ca exchanger, reduced the increase in [Ca2+]i due to hypoxia, but enhanced the KCl response. The changes in ATP following K+ depolarization, KCN or their combination in the presence of physiological calcium concentrations did not parallel alterations in [Ca2+]i, which suggests that diminished activity of the calcium dependent ATPase does not underlie the elevation in [Ca2+]i. Valinomycin, an ionophore which reduces the mitochondrial membrane potential, elevated [Ca2+]i and the effects were additive with K+ depolarization in a calcium dependent manner that paralleled the effects of hypoxia. Together these results suggest that hypoxia-induced elevations of synaptosomal [Ca2+]i are due to an inability of the synaptosome to buffer entering calcium.

Phosphoinositide breakdown in rat hippocampal slices: sensitivity to glutamate induced by in vitro anoxia

We examined the effects of in vitro anoxia on phosphoinositide (PI) breakdown in rat hippocampal slices stimulated by glutamate and quisqualate. In addition to assays of accumulations of 3H-inositol phosphates (3H-IPs) degraded from prelabeled PI, we adopted direct assay procedures of inositol 1,4,5-triphosphate (1,4,5-IP3) using 1,4,5-IP3-specific binding protein to determine the formation of 1,4,5-IP3. The first effect, observed with anoxic incubation by itself, was the diminished quisqualate (10(-5) M)-stimulated accumulation of 3H-IPs degraded from prelabeled PI under prolonged anoxia. Quisqualate caused a transient increase in 1,4,5-IP3 formation in the early phase of anoxia, similar to that under oxygenated conditions. Glutamate (10(-5) M), under normal conditions, influenced neither the accumulation of 3H-IPs nor the formation of 1,4,5-IP3. Also, the accumulation of 3H-IPs under prolonged anoxia was unaffected. The same concentration of glutamate, however, gave rise to a transient increase in 1,4,5-IP3 content in the early phase of anoxia, similar to that caused by quisqualate. The second effect, observed by oxygenation following anoxia, was the induction of glutamate-stimulated accumulation of 3H-IPs. When the hippocampal slices were oxygenated following a sufficiently long (greater than 30-min) exposure to anoxia, glutamate (10(-5) M) caused a significant increase in accumulation of 3H-IPs degraded from prelabeled PI. Quisqualate-stimulated accumulation of 3H-IPs under oxygenated incubations was also increased by prior exposure of slices to anoxia. These results support the hypothesis that an exposure of hippocampal slices to anoxia induces a sensitivity of the PI breakdown pathway to glutamate and that, given an oxygen supply following sufficiently long exposure to anoxia, the slices maintain their sensitivity to glutamate with an apparent increase in the accumulation of 3H-IPs.

Effects of in vitro hypoxia on depolarization-stimulated accumulation of inositol phosphates in synaptosomes

The effects of potassium and in vitro histotoxic hypoxia (i.e. KCN) on phosphatidylinositol turnover in rat cortical synaptosomes were determined. [2-3H] Inositol prelabelled rat synaptosomes were prepared from cerebral cortex slices that had been incubated with [2-3H] inositol. Depolarization with 60 mM KCl increased [2-3H] inositol phosphates in a time dependent manner. Depolarization with 60 mM KCl increased [2-3H] inositol trisphosphate transiently at 5 s. K+ induced rapid formation of [2-3H]-inositol bisphosphate and maintained an elevated level for at least 5 min. K+ stimulated gradual formation of [2-3H] inositol monophosphate with time. One minute of hypoxia enhanced potassium-stimulated [2-3H] inositol bisphosphate formation. However, 30 min of hypoxia impaired potassium-stimulated accumulation of [2-3H] inositol phosphates. The effects of histotoxic hypoxia were all dependent upon calcium in the medium and on K+-depolarization. Thus, hypoxia altered the K+-induced accumulation of inositol phosphates in prelabelled synaptosomes in a time dependent, biphasic manner that was calcium dependent.