Background and Purpose Treatment with acetylsalicylic acid (ASA) is established for secondary stroke prevention. Recent studies showed neuroprotection of ASA against glutamatergic excitants. The goal of this study was to investigate the time course of neuroprotection of ASA against indirect excitotoxicity by hypoxic hypoxia and chemical hypoxia.
Methods Population spike amplitude (PSA) and ATP content were measured in hippocampal slices from untreated control animals (c-slices) and slices prepared from animals pretreated in vivo with a single intraperitoneal injection of 20 mg/kg body wt ASA 1 to 48 hours before slice preparation (p-slices).
Results Posthypoxic recovery of PSA was 30% in c-slices (15 minutes of hypoxia, 45 minutes of recovery). When c-slices were treated in vitro for 15 minutes with 20 mg/L ASA 30 minutes before hypoxia, posthypoxic recovery improved to 82±4% (mean±SE, P<.01). In p-slices, posthypoxic recovery of PSA improved in a time-dependent manner. With a time interval of 1 hour between in vivo pretreatment with ASA and slice preparation, posthypoxic recovery of PSA was 64±16% (P<.05). With time intervals of 6 hours, 24 hours, and 48 hours, posthypoxic recovery of PSA was 87±19% (P<.01), 59±12%, and 40±9%, respectively. Pretreatment with ASA in vitro or in vivo decreased the decline of ATP content during hypoxic hypoxia and chemical hypoxia (inhibition of succinic dehydrogenase by 3-nitropropionic acid). When extracellular glucose was reduced to 4 mmol/L, no difference was observed between c-slices and p-slices.
Conclusions We conclude that ASA is neuroprotective against hypoxic hypoxia and chemical hypoxia and delays the decline of intracellular ATP content.
Acetylsalicylic acid (ASA) is an established prophylactic treatment in situations with increased risk for cerebral ischemia.1 Recently it was shown experimentally that ASA protects against direct excitotoxicity by application of glutamatergic excitants.2 This work was done under blood-free conditions in which the inhibition of platelet aggregation by ASA cannot be the reason for its neuroprotective action.
It is well established that cell damage in cerebral hypoxia and ischemia arises from indirect excitotoxicity.3 Glutamate antagonists repeatedly were shown to ameliorate damage due to hypoxia/ischemia.4 5 6 However, the reason for the toxicity of glutamate agonists is not potentiated excitation due to relief of the magnesium block of the N-methyl-d-aspartate receptor but the decline of cellular energy metabolism, which causes a failure of neuronal ion exchange and impairment of repolarization.7 It therefore is important to identify therapies that delay the breakdown of cellular energy metabolism.
Recently it was shown that mild inhibition of oxidative phosphorylation with inhibition of succinic dehydrogenase slows down the ATP decrease upon severe chemical hypoxia.8 This is associated with a decrease in posthypoxic free radical production, improved posthypoxic morphology, and preserved neuronal function.8 9 10
For many years it has been known that ASA interferes with mitochondrial function.11 12 13 14 Therefore, the goal of the present study was to determine whether ASA delays depletion of intracellular energy stores and to investigate the time dependency of a possible neuroprotective action of ASA against indirect excitotoxicity by hypoxic hypoxia and chemical hypoxia.
Materials and Methods
Male Wistar rats (weight, 150 to 200 g) were killed by cervical dislocation. Preparation of slices along the long axis of the hippocampus was performed as previously described.15 Before further treatment, slices were incubated for at least 2 hours at 35°C in Ringer’s solution containing the following (mmol/L): NaCl 126, KCl 5, KH2PO4 1.3, MgSO4 1.3, CaCl2 2.4, NaHCO3 26, dextrose 10, bubbled with 95% O2 and 5% CO2.
Further treatment for investigation of hypoxic hypoxia and chemical hypoxia was performed as described previously.8 9 10 Electrophysiological measurements were performed in the recording chamber (see below). Further treatment for biochemical analysis was performed by transferring the slices into a static bath bubbled with 95% N2 and 5% CO2 for experiments with hypoxic hypoxia or into a static bath bubbled with 95% O2 and 5% CO2 containing 1 mmol/L 3-nitropropionic acid (for experiments with chemical hypoxia). Treatment groups are shown in Tables 1 through 3⇓⇓⇓.
All experiments were performed in accordance with institutional guidelines.
Recording of population spike amplitude (PSA) was performed as previously described.9 10 In brief, slices were transferred to a recording chamber after preincubation. The recording chamber was perfused with Ringer’s solution at 6 mL/min through a peristaltic pump. Slices with PSA less than 2 mV were not included in the analysis. Upon stabilization of PSA, Ringer’s solution was made hypoxic by bubbling with 95% N2 and 5% CO2 (Po2 in the recording chamber was below 10 kPa after 5 minutes of perfusion with hypoxic Ringer’s solution; for detailed time course of Po2 in the recording chamber, see Reference 1616 ). After 15 minutes of hypoxia, slices were superfused with oxygenated Ringer’s solution until the end of the experiment. Schaffer collaterals were synaptically activated at 0.1 Hz by bipolar electrodes placed in hippocampal region CA3. Population spikes were recorded in hippocampal region CA1, and the analysis was performed as described previously.9 10
ATP content was determined and analyzed as described previously.8 Upon termination of treatment, hippocampal slices were placed in 5% perchloric acid, homogenized, and analyzed by reversed-phase high-performance liquid chromatography with the use of the Supelcosil LC18 column and elution with 10 mmol/L phosphate buffer (pH 6.7) containing 2 mmol/L tert-butylammonium phosphate. Concentrations of adenine phosphates were calculated as nanomoles per milligrams protein and normalized to ATP levels in slices prepared from untreated control animals (13.2±1.1 nmol/mg protein).
Each experiment was conducted with four to six slices from two to three animals. Statistical testing was performed by ANOVA and Tukey’s protected t test. Statistical significance was accepted at P<.05.
Recovery of PSA During In Vitro Treatment With ASA
The PSA upon hypoxia and recovery was determined in slices prepared from untreated control animals and animals pretreated with a single intraperitoneal injection of 20 mg/kg body wt ASA at different times before slice preparation. A typical example of PSA during hypoxia and recovery is shown in Fig 1⇓. Recovery of slices from control animals was reported previously10 ; PSA was approximately 10% at the end of hypoxia and 30% after 45 minutes of recovery. When slices from control animals were treated for 15 minutes with 20 mg/L ASA, PSAP at the end of 15 minutes of hypoxia was 5±1% (Fig 2⇓, Table 1⇑). The PSA was 73±4% (P<.01) after 15 minutes of recovery, 82±5% (P<.01) after 30 minutes of recovery, and 82±4% (P<.01) after 45 minutes of recovery from 15 minutes of hypoxia.
Recovery of PSA During In Vivo Treatment With ASA
The PSA at the end of hypoxia in slices prepared from animals with in vivo treatment 1 hour before slice preparation was 8±1% (Fig 3⇓, Table 1⇑), 13±2% with 6 hours, 10±2% with 24 hours, and 11±3% with 48 hours. The PSA at 45 minutes of recovery from 15 minutes of hypoxia was 64±16% (P<.05) in slices prepared from animals with in vivo treatment 1 hour before slice preparation (Fig 3⇓), 87±19% (P<.01) with 6 hours, 59±12% with 24 hours, and 40±9% with 48 hours.
Recovery of PSA After Reduction of Extracellular Glucose
Similar experiments were performed in 4 mmol/L glucose. In slices from untreated control animals, the PSA was 8±2% (mean±SE; Fig 4⇓, Table 1⇑) at the end of hypoxia and 3±2% after 45 minutes of recovery. Similarly, the PSA at the end of hypoxia in slices prepared from animals with in vivo treatment 6 hours before slice preparation was 6±1% (Fig 4⇓, Table 1⇑). The amplitude of the population spike upon 45 minutes of recovery from 15 minutes of hypoxia was 5±1% (P=NS versus control) in slices prepared from animals with in vivo treatment 6 hours before slice preparation (Fig 1⇑).
ATP Content After Treatment of Hippocampal Slices In Vitro
ATP content of hippocampal slices at the end of hypoxia was 26±2% (mean±SE; P<.01; Table 2⇑). Upon recovery from hypoxia, ATP levels increased to 80±3% (P<.01) but remained below control level (P<.01). In animals pretreated in vivo with 20 mg/kg body wt ASA, ATP content was 95±4% of control (P=NS versus control) before further treatment. Upon 15 minutes of hypoxia, ATP levels decreased to 22±2% (P<.01). Upon 60 minutes of recovery, ATP levels increased to 91±2% (P<.01) but remained lower than control levels (P<.05).
To investigate whether ASA has an effect on ATP levels of hippocampal slices, we treated slices from control animals in a static bath in vitro with 20 mg/L ASA. ATP levels were 91±5% (P=NS versus control) after 60 minutes of treatment and 99±6 (P=NS versus control) after 90 minutes of treatment. When similar experiments were performed in 4 mmol/L glucose, ATP levels were 97±10% (P=NS versus control) after 60 minutes of treatment and 105±6% (P=NS versus control) after 90 minutes of treatment.
ATP Content During Chemical Hypoxia After Pretreatment In Vivo
Treatment with ASA also protects against chemical inhibition of oxidative phosphorylation. Slices from control animals were continuously superfused in vitro with 1 mmol/L of 3-nitropropionic acid, an inhibitor of succinic dehydrogenase. After 60 minutes of superfusion, ATP levels declined to 74±6% of control levels (P<.01) and to 42±2% (P<.01) of levels after 90 minutes of superfusion. When slices were prepared from animals pretreated in vivo with a single injection with 20 mg/kg body wt ASA, ATP levels declined to 95±2% after 60 minutes (P=NS versus control; P<.01 versus 3-nitropropionic acid–treated slices from control animals) of 3-nitropropionic acid superfusion in vitro and to 77±4% (P<.01 versus control; P<.01 versus 3-nitropropionic acid–treated slices from control animals) after 90 minutes of superfusion with 3-nitropropionic acid.
When similar experiments were performed in 4 mmol/L glucose, no significant difference was observed between untreated controls and preconditioned animals. Slices from control animals were continuously superfused in vitro with 1 mmol/L of 3-nitropropionic acid. After 60 minutes of superfusion, ATP levels declined to 85±6% of control levels (P<.01 versus control; P=NS versus 10 mmol/L glucose) and to 52±4% (P<.01 versus control; P=NS versus 10 mmol/L glucose) of levels after 90 minutes of superfusion. When slices were prepared from animals pretreated in vivo with a single injection of 20 mg/kg body wt ASA, ATP levels declined to 75±2% after 60 minutes of 3-nitropropionic acid superfusion in vitro and to 47±2% after 90 minutes of superfusion with 3-nitropropionic acid.
Treatment with ASA is established for secondary stroke prevention.1 However, the mechanism of its action remained unclear. Thus far, it has been argued that ASA exerts its effect by inhibition of platelet aggregation.17 However, it was shown recently that ASA is neuroprotective against direct excitotoxicity by inhibition of transcription factor nuclear factor-κB2 under blood-free conditions. However, one of the remaining issues has been the time course of action. Clinically, it is generally accepted that dosage once a day is sufficient. However, none of the mechanisms suspected thus far has shown a time course of action in which the protective effect is related to a mechanism that covers exactly this time period. Single dosage with ASA has an effect on platelet aggregation for up to 48 hours in rats18 and up to 72 hours in humans.19
The above data show that ASA is neuroprotective against indirect excitotoxicity by hypoxic hypoxia and chemical hypoxia. Neuronal function, which is a good measure of the integrity of the slice,20 is well preserved after pretreatment with ASA in vivo and in vitro. The protection by pretreatment in vitro shows again that ASA is neuroprotective independent of its action on platelets. The biochemical analysis of high-energy phosphates shows that pretreatment with ASA slows down the decay of intracellular ATP upon hypoxia or chemical hypoxia. Since failure of neuronal ion exchange as a result of decay of intracellular ATP is an important mechanism of indirect excitotoxicity,7 this effect may also be important for the clinical actions of ASA.
In recent years it has been demonstrated repeatedly that a short ischemic or hypoxic episode before lethal ischemia ameliorates tissue damage, which is known as ischemic preconditioning.21 Recently it was demonstrated that not only short ischemic or hypoxic episodes but also mild chemical hypoxia can induce increased hypoxic tolerance.9 10 This has been demonstrated for inhibition of succinic dehydrogenase with 3-nitropropionic acid9 10 and inhibition of mitochondrial complex I with haloperidol.22 It has been known for a long time that ASA is an inhibitor and uncoupler of oxidative phosphorylation.11 13 23 We hypothesize that the protective effect of ASA against hypoxia and chemical hypoxia at least in part may be due to a similar increase of endogenous hypoxic tolerance. Pretreatment with ASA delays the decline of energy metabolism upon lethal hypoxic or chemical hypoxia, similar to other methods of induction of chemical preconditioning.
No increase in hypoxic tolerance with reduced glucose was observed in this study. This could indicate that damage by hypoxia in situations with reduced glucose is too severe to be prevented by prior increase of hypoxic tolerance with ASA. Alternatively, whether a glucose-dependent mechanism mediates increase of hypoxic tolerance needs to be investigated.
In summary, this study shows that hypoxic hypoxia and chemical hypoxia are associated with a decrease in intracellular ATP content and that this decrease is reduced and/or delayed by ASA. Posthypoxic recovery of neuronal function is improved upon pretreatment with ASA.
This study was supported by a grant (Ri 583/2-1) from the Deutsche Forschungsgemeinschaft to MWR.
- Received May 1, 1997.
- Revision received June 19, 1997.
- Accepted July 7, 1997.
- Copyright © 1997 by American Heart Association
Sivenius J, Laakso M, Penttila IM, Smets P, Lowenthal A, Riekkinen PJ. The European Stroke prevention study: results according to sex. Neurology. 1991;41:1189-1192.
Grilli M, Pizzi M, Memo M, Spano P. Neuroprotection by aspirin and sodium salicylate through blockade of NF-κB activation. Science. 1996;274:1383-1385.
Simon RP, Swan JH, Griffiths T, Meldrum BS. Blockade of N-methyl-D-aspartate receptors may protect against ischemic damage in the brain. Science. 1984;226:850-852.
Lutwak-Mann C. The effect of salicylate and cinchophen on enzymes and metabolic processes. Biochem J. 1942;36:706-728.
Lee TK, Chen YC, Lien IN, Liu MC, Huang ZS. Inhibitory effect of acetylsalicylic acid on platelet function in patients with completed stroke or reversible ischemic neurologic deficit. Stroke. 1988;19:566-570.
Brody TM. Action of sodium salicylate and related compounds on tissue metabolism in vitro. J Pharmacol Exp Ther. 1956;117:39-51.