Brief, repeated, oxygen-glucose deprivation episodes protect neurotransmission from a longer ischemic episode in the in vitro hippocampus: role of adenosine receptors

Abstract

1. Ischemic preconditioning in the brain consists of reducing the sensitivity of neuronal tissue to further, more severe, ischemic insults. We recorded field epsps (fepsps) extracellularly from hippocampal slices to develop a model of in vitro ischemic preconditioning and to evaluate the role of A1, A2A and A3 adenosine receptors in this phenomenon. 2. The application of an ischemic insult, obtained by glucose and oxygen deprivation for 7 min, produced an irreversible depression of synaptic transmission. Ischemic preconditioning was induced by four ischemic insults (2 min each) separated by 13 min of normoxic conditions. After 30 min, an ischemic insult of 7 min was applied. This protocol substantially protected the tissue from the irreversible depression of synaptic activity. 3. The selective adenosine A1 receptor antagonist, 8-cyclopentyl-1,3-dipropylxanthine (DPCPX, 100 nm), completely prevented the protective effect of preconditioning. The selective adenosine A2A receptor antagonist 4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino]ethyl)phenol (ZM 241385, 100 nm) did not modify the magnitude of fepsp recovery compared to control slices. The selective A3 adenosine receptor antagonists, 3-propyl-6-ethyl-5[ethyl(thio)carbonyl]-2-phenyl-4-propyl-3-pyridinecarboxylate (MRS 1523, 100 nm) significantly improved the recovery of fepsps after 7 min of ischemia. 4. Our results show that in vitro ischemic preconditioning allows CA1 hippocampal neurons to become resistant to prolonged exposure to ischemia. Adenosine, by stimulating A1 receptors, plays a crucial role in eliciting the cell mechanisms underlying preconditioning; A2A receptors are not involved in this phenomenon, whereas A3 receptor activation is harmful to ischemic preconditioning.

Figure 1

Effects of ischemia or hypoxia on fepsps evoked by electrical stimulation of the stratum radiatum in the CA1 hippocampal region. (a) The graphs show the time course of fepsps amplitude, before, during different time durations of OGD and after reperfusion in normal oxygenated aCSF. Data, expressed as percent of baseline values, are means±s.e.m., n=20 for 2 and 5 min of OGD and n=50 for 7 min of OGD. Bars indicate the time duration of OGD. Note the transient reappearance of synaptic potential after superfusion in normal oxygenated aCSF only after 5-min OGD (arrowhead). Upper panel, left side: fepsps recorded from hippocampal slices in a typical 7-min OGD experiment: (a) control, (b) at the end of 7-min OGD, (c) after a 20-min washout in normal oxygenated aCSF. A 7-min OGD abolishes both fepsp and afferent volley amplitude. After reperfusion in normal oxygenated aCSF, only the recovery of afferent volley was recorded. Calibration bars: 0.5 mV, 5 ms. (b) Upper panel, right side: Values (mean±s.e.m.) in the graphs represent pO₂ levels (expressed in percentage) during 2- or 7-min OGD and immediately after reperfusion in aCSF. The lowest pO₂ values are obtained starting from the fifth minute of the 7-min OGD. (b) The graphs show the time-course of fepsps amplitude before, during OGD or OD application and after reperfusion in oxygenated aCSF. Data, expressed as percent of baseline values, are means±s.e.m. No electrical activity was recorded after 15 min of OGD (n=7), whereas a complete recovery of neurotransmission after 5 min of OGD, followed by 10-min anoxia, was observed (n=5). Closed and open bars indicate time duration of OGD and OD, respectively.

Figure 2

Ischemic preconditioning induces tolerance against the depression of synaptic transmission induced by 7 min of OGD in the CA1 hippocampal region. (a) Traces are fepsps recorded from CA1 stratum radiatum at representative times of the corresponding experiment shown in (b) Synaptic potentials recorded in normoxic aCSF and at the point of maximal depression are superimposed to show the maximal effect of ischemic episodes. Calibration bars: 0.5 mV, 5 ms. (b) The graph shows the time course of fepsp amplitude, expressed as percent of baseline responses, in a typical experiment, during the four preconditioning 2-min OGD episodes and the 7 min of OGD ‘test' application. Bars indicate the time duration of OGD episodes. Note the complete recovery of synaptic potential, after 7-min OGD, occurring within 10 min of superfusion in normal oxygenated aCSF and the transient reappearance of synaptic potentials at the end of the 7 min of OGD (arrowhead).

Figure 3

Ischemic preconditioning protects hippocampal slices from the fepsp depression induced by the 7 min of ischemia. The graphs show the time course of fepsp amplitude, expressed as percent of baseline, 10 min before, during 7 min of OGD and superfusion in normal oxygenated aCSF in the absence (mean±s.e.m., n=50) and after ischemic preconditioning (mean±s.e.m., n=22). A transient reappearance of synaptic potentials was observed in preconditioned slices at the end of 7-min OGD (arrowhead). Bars indicate the time duration of OGD episodes.

Figure 4

DPCPX prevents the protective effect of ischemic preconditioning on fepsp depression induced by 7 min of ischemia. The graph shows the time course of fepsp amplitude recorded from a typical experiment carried out in the presence of DPCPX (100 nM, open bar), an A₁ adenosine selective antagonist, applied 25 min before and during the preconditioning OGD episodes. (Inset) The time course of fepsp depression caused by 7-min OGD in nonconditioned (control) slices and in slices in which preconditioning was performed in the presence of DPCPX. Given are mean±s.e.m. values of fepsp amplitude (not conditioned: n=50; conditioned n=12). Note that the entity of the fepsp depression recorded at 2.5 min of the 7-min OGD application in slices preconditioned in the presence of DPCPX was similar (arrowhead) to that observed in controls (see also Table 1). Bars in the figure and in the inset indicate the time duration of OGD episode.

Figure 5

Role of A₁, A_2A and A₃ receptor activation in ischemic preconditioning in rat hippocampal slices. (a) Graphs show the time course of fepsp amplitude 10 min before, during 7 min of OGD and superfusion in normal oxygenated aCSF, after preconditioning in the presence of different adenosine receptor antagonists (DPCPX, 100 nM, n=12; ZM 241385, 100 nM, n=8; MRS 1523, 100 nM, n=10). Data, expressed as percent of baseline values, are means±s.e.m. of fepsp amplitude. Bar indicates the time duration of ischemic episodes. (b) Column bars in the graph summarize the average recovery (mean±s.e.m.) of fepsp after 7 min of OGD, recorded in hippocampal slices at 15 min reperfusion in normal, oxygenated aCSF in control and preconditioned slices. *P<0.05 versus control; ^§P<0.05 versus all experimental conditions. The selective adenosine antagonists, DPCPX, ZM 241385, MRS 1523, were applied during ischemic preconditioning at a concentration of 100 nM. Differences between data were analyzed by one way ANOVA, followed by Newman–Keuls multiple comparison post hoc test.