Skip to main content
PMC home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2008 Mar 17;105(12):4892–4897. doi: 10.1073/pnas.0800628105

Ischemic preconditioning blocks BAD translocation, Bcl-xL cleavage, and large channel activity in mitochondria of postischemic hippocampal neurons

Takahiro Miyawaki *,†, Toshihiro Mashiko †, Dimitry Ofengeim *, Richard J Flannery ‡, Kyung-Min Noh *, Sho Fujisawa *, Laura Bonanni §, Michael V L Bennett *,, R Suzanne Zukin *,, Elizabeth A Jonas ‡
PMCID: PMC2290755  PMID: 18347331

Abstract

Transient forebrain or global ischemia induces delayed neuronal death in vulnerable CA1 pyramidal cells with many features of apoptosis. A brief period of ischemia, i.e., ischemic preconditioning, affords robust protection of CA1 neurons against a subsequent more prolonged ischemic challenge. Here we show that preconditioning acts via PI3K/Akt signaling to block the ischemia-induced cascade involving mitochondrial translocation of Bad, assembly of Bad with Bcl-xL, cleavage of Bcl-xL to form its prodeath fragment, ΔN-Bcl-xL, activation of large-conductance channels in the mitochondrial outer membrane, mitochondrial release of cytochrome c and Smac/DIABLO (second mitochondria-derived activator of caspases/direct IAP-binding protein with low pI), caspase activation, and neuronal death. These findings show how preconditioning acts to prevent the release of cytochrome c and Smac/DIABLO from mitochondria and to preserve the integrity of the mitochondrial membrane. The specific PI3K inhibitor LY294002 administered in vivo 1 h before or immediately after ischemia or up to 120 h later significantly reverses preconditioning-induced protection, indicating a requirement for sustained PI3K signaling in ischemic tolerance. These findings implicate PI3K/Akt signaling in maintenance of the integrity of the mitochondrial outer membrane.

Keywords: Akt, ischemic tolerance, PI3K, postischemic neurodegeneration

Transient global or forebrain ischemia arising in humans as a consequence of cardiac arrest or cardiac surgery, or induced experimentally in animals, leads to selective and delayed neuronal death of hippocampal CA1 neurons and cognitive deficits (1–4). Injurious stimuli such as ischemia disrupt the integrity of the mitochondrial membrane, leading to the release of cytochrome c and activation of cysteine proteases including caspase-9, a critical “initiator” caspase, and caspase-3, a “terminator” caspase implicated in the execution step of apoptosis (5–7). Caspase-3 activation is critical to delayed neurodegeneration after ischemia (8). Global ischemia promotes expression of the prosurvival inhibitor-of-apoptosis (IAP) family member cIAP but causes release of Smac/DIABLO (second mitochondria-derived activator of caspases/direct IAP-binding protein with low pI) from mitochondria, a factor that neutralizes the protective actions of IAPs and promotes neuronal apoptosis (1–4).

The Bcl-2 family member Bcl-xL is a potent inhibitor of programmed cell death and acts by heterodimerization with other Bcl-2 family members to inhibit caspase activation, presumably by blocking release of cytochrome c (9, 10). Bcl-xL is abundantly expressed in adult neurons (11, 12) and localizes to the outer mitochondrial membrane (13). N-terminal cleavage of Bcl-xL to form its proapoptotic fragment, ΔN-Bcl-xL, leads to appearance of large-conductance channels in the mitochondrial outer membrane (14–16). In hippocampal neurons these channels appear within 1 h of insult (16). Activation of large-conductance channels correlates with appearance of ΔN-Bcl-xL in the mitochondria, is mimicked by introduction of recombinant ΔN-Bcl-xL, and is blocked by introduction of a Bcl-xL antibody via the patch pipette (16). These findings implicate ΔN-Bcl-xL in the large-conductance channels observed in the mitochondrial outer membrane of insulted hippocampal CA1 neurons.

Ischemic tolerance is a well established phenomenon in which a brief ischemic insult (or preconditioning) protects CA1 neurons against a subsequent more prolonged ischemic challenge (17, 18). This protective action involves inhibition of activated caspase-3 (19). Neuroprotective strategies such as ischemic preconditioning promote phosphorylation of Akt at Ser-473 and phosphorylation/inactivation of Akt downstream targets; PI3K/Akt signaling is required for preconditioning-induced neuroprotection (20–22). Preconditioning intervenes downstream of proteolytic processing of caspase-3 to its active form, but caspase-3 activity is inhibited by endogenous inhibitors, and one of its targets, caspase-activated DNase (CAD, a DNase that catalyzes DNA fragmentation) is unaffected (19). Preconditioning blocks the mitochondrial release of Smac/DIABLO but not the ischemia-induced up-regulation of IAPs (19). In the absence of Smac/DIABLO, cIAP halts the caspase death cascade and arrests neuronal death.

The present study was undertaken to examine the molecular mechanisms by which ischemic preconditioning prevents the ischemia-induced increase in cytosolic cytochrome c and Smac/DIABLO. We show that ischemic preconditioning acts via PI3K signaling to prevent mitochondrial translocation of Bad, binding of Bad to Bcl-xL, cleavage of Bcl-xL to generate ΔN-Bcl-xL, activation of large-conductance channels in the mitochondrial outer membrane, and activation of the caspase cascade leading to neuronal death. These findings give new insight into the molecular mechanisms of neuroprotection by ischemic preconditioning.

Results

Preconditioning Promotes Assembly of Akt with Bad.

Akt promotes cell survival via phosphorylation and reduction in cytoplasm to mitochondrial translocation of downstream targets such as the proapoptotic Bcl-2 family member Bad (23). To examine whether PI3K is involved in ischemic tolerance, rats were subjected to sham operation, preconditioning [4-min four-vessel occlusion (4-VO); PC], ischemia (10-min 4-VO; Isch), preconditioning followed by ischemia (4-min 4-VO, followed 48 h later by 10-min 4-VO; PC + Isch), or LY294002 injection (i.c.v.) just after sham operation (LY), 48 h after PC (PC + LY), + LY, or just after Isch in PC + Isch animals (PC + Isch + LY) (see details in supporting information (SI)). Preconditioning altered assembly/association of Akt with Bad, as demonstrated by coimmunoprecipitation with an antibody directed against Bad and probed for Akt. Ischemia did not detectably alter association of Akt with Bad, assayed at 1 h after reperfusion (Fig. 1a). Preconditioning itself induced a modest increase in assembly of Bad and Akt (to ≈170%, measured at 49 h after reperfusion) and markedly enhanced Bad/Akt interaction in animals subjected to ischemia 48 h after preconditioning (PC + Isch) (to ≈290% measured at 1 h after Isch; Fig. 1a). LY294002 did not itself alter assembly of Bad with Akt after 1 h but reduced the impact of preconditioning on Bad–Akt assembly (1 h after PC + Isch + LY, to ≈150%; Fig. 1a). These findings indicate that preconditioning promotes assembly of Akt with Bad and suggest a role for PI3K signaling in Bad phosphorylation.

Fig. 1.

Fig. 1.

Open in a new tab

Preconditioning promotes assembly of Akt with Bad, Akt-mediated Bad phosphorylation, and cytoplasmic retention of Bad. (a) Representative coimmunoprecipitation of Bad–Akt (Upper) and summary data (Lower). Preconditioning (PC) and preconditioning followed by ischemia (PC + Isch) promote assembly of Akt with Bad. The PI3K inhibitor LY294002 reverses the effect of PC. (b) Representative Western blot (Top) and summary data for p-Ser136-Bad (Middle) and total Bad (Bottom) in the cytosol. Preconditioning promotes Bad phosphorylation. n = 3–8 per group. (c) Western of total Bad in the mitochondria. n = 5–8 per treatment group. Cytochrome c oxidase IV (COX IV) served as loading control. Error bars indicate SEM. In all figures, significance between experimental and sham animals is indicated by *, and significance between PC + Isch and Isch groups is indicated by #. * and #, P < 0.05; ** and ##, P < 0.01.

Preconditioning Acts via PI3K Signaling to Prevent Mitochondrial Translocation of Bad.

Bcl-2 family members such as Bad shuttle between the cytosol and mitchondria and, when associated with mitochondria, can promote apoptosis (5, 6, 9). PI3K/Akt-mediated phosphorylation of Bad promotes its binding to the cytoplasmic retention factor 14-3-3 and prevents translocation of Bad to mitochondria (23). To examine the impact of preconditioning on phosphorylation of Bad at Ser-136, we probed Western blots of cytosolic and mitochondrial fractions with antibodies against total and p-Bad (1 h after ischemia and reperfusion; Fig. 1 b and c). Ischemia alone did not detectably alter the phosphorylation status of Bad (Fig. 1b). Preconditioning also did not alter the phosphorylation status of Bad (measured 49 h after preconditioning; data not illustrated) but modestly enhanced Bad phosphorylation induced by subsequent ischemia (to 130 ± 4% of control; P < 0.01; Fig. 1b Top). LY294002 (1 h after i.c.v. injection) did not itself alter the phosphorylation status of Bad, but it significantly reduced the increase in phosphorylation induced in PC + Isch animals [1 h after LY294002 injection immediately after reperfusion; P < 0.05 for (PC + Isch) vs. (PC + Isch + LY); Fig. 1b]. Bad abundance in the cytosol was unaltered by all treatments (Fig. 1b Bottom).

To examine the impact of preconditioning on Bad translocation, we examined Bad abundance in the mitochondrial fraction 1 h after Isch. Isch promoted Bad translocation (to 328 ± 26% of control; Fig. 1c); preconditioning did not itself affect Bad distribution (measured at 49 h; data not illustrated) but markedly attenuated ischemia-induced translocation of Bad (to 163 ± 27% of control; P < 0.01 vs. Isch). LY294002 itself did not significantly affect the amount of Bad in the mitochondria but reduced the effect of preconditioning on Bad translocation after ischemia [mitochondrial accumulation of Bad to 232 ± 29% of control; P < 0.01 for (PC + Isch + LY) vs. (PC + Isch) and P < 0.01 vs. sham; Fig. 1c]. These findings indicate that preconditioning acts via PI3K signaling to promote assembly of Akt with Bad, phosphorylation of Bad, and its cytosolic retention, presumably as p-Bad, in the face of ischemia, consistent with a role in neuroprotection by preconditioning. Because PC + Isch shows only a small increase in the p-Bad/Bad ratio and total cytosolic Bad shows little change, it is unlikely that cytosolic Bad is reduced enough to account for the large reduction in mitochondrial Bad in PC + Isch animals compared with Isch animals. We hypothesize that only a small fraction of the Bad we measure by Western blotting is available for translocation to the mitochondria and that this fraction is greatly reduced by preconditioning and Bad phosphorylation. p-Bad was not detectable in the mitochondrial fraction from any treatment group (data not illustrated), in confirmation of others (24).

Ischemia and PI3K Blockade Promote Assembly of Bad with Bcl-xL.

Upon translocation to the mitochondria, Bad forms homodimers and heterodimers with other Bcl-2 family members and promotes apoptosis (25). To examine the impact of ischemia on assembly of Bad with Bcl-xL, we performed coimmunoprecipitation with an anti-Bcl-xL antibody and probed for Bad. Ischemia promoted assembly of Bad and Bcl-xL in naïve animals assessed at 1 h after the last surgery (Fig. 2a). Moreover, Bad-Bcl-xL association was higher in LY and in PC + Isch + LY animals (Fig. 2a). These findings are consistent with the possibility that preconditioning acts via PI3K signaling to promote Bad phosphorylation and thereby prevent mitochondrial translocation of Bad and assembly of Bad with Bcl-xL. However, values for PC and for PC + Isch animals were too variable for reliable interpretation.

Fig. 2.

Fig. 2.

Open in a new tab

Preconditioning acts via PI3K signaling to promote Bad/Bcl-xL assembly and Bcl-xL cleavage to generate ΔN-Bcl-xL. (a) Coimmunoprecipitation of Bad/Bcl-xL was increased over sham by LY, Isch, and PC + Isch + LY. (b) Western blot of mitochondrial Bcl-xL and ΔN-Bcl-xL (Top) and summary data ΔN-Bcl-xL/Bcl-xL (Middle) and full-length Bcl-xL (Bottom). n = 6 per group. (c) Western blot of cytosolic Bcl-xL and ΔN-Bcl-xL (Top) and summary data (Middle and Bottom). Animals were killed 1 h after the last surgery or 49 h after PC. Values of n are indicated on the histogram bars. Isch promoted (and PC prevented) Bcl-xL cleavage. LY294002 (LY) attenuated the effects of preconditioning.

Preconditioning Acts via PI3K Signaling to Maintain Full-Length Bcl-xL.

Ischemia promotes cleavage of Bcl-xL to generate its prodeath fragment, ΔN-Bcl-xL, which in turn activates large-conductance channel activity in the mitochondrial outer membrane (16). To examine the impact of preconditioning on production of ΔN-Bcl-xL, we subjected animals to treatments as above, killed them at 1 h after sham operation or reperfusion, and analyzed samples of microdissected CA1 by Western blotting. Protein samples were sequentially probed for full-length Bcl-xL followed by ΔN61-Bcl-xL. Ischemia promoted cleavage of Bcl-xL to generate its proapoptotic counterpart, ΔN61-Bcl-xL, evident in mitochondria (211 ± 11% of control, P < 0.01; Fig. 2b Middle; see SI Fig. 6) and cytosolic fractions (207 ± 33% of control, P < 0.01; Fig. 2c Middle). Preconditioning did not itself alter the cleavage of Bcl-xL (data not shown) but prevented ischemia-induced generation of ΔN61-Bcl-xL in the mitochondria (95 ± 19% of control, P > 0.05; Fig. 2b Middle) and cytosol (97 ± 11% of control, P > 0.05; Fig. 2c Middle). LY294002 itself did not significantly alter the status of Bcl-xL, but it reduced the action of preconditioning to prevent the ischemia-induced rise in ΔN61-Bcl-xL in mitochondria (to 158 ± 15% of control, P < 0.01 vs. PC + Isch; Fig. 2b Top) and cytosol (to 235 ± 37% of control, P < 0.01 vs. PC + Isch; Fig. 2c Middle). The abundance of full-length Bcl-xL was unaltered by all treatments, except for LY294002, which moderately reduced full-length Bcl-xL abundance in the cytosol (to 76 ± 3% of control, P < 0.01; Fig. 2c Bottom). Abundance of full-length Bcl-xL in the total cell lysate was unchanged by all treatments (see SI Fig. 7). Thus, preconditioning acts via PI3K signaling to promote Bad phosphorylation and prevent mitochondrial translocation of Bad and cleavage of Bcl-xL to generate its prodeath fragment, ΔN61-Bcl-xL.

Preconditioning Acts via PI3K Signaling to Prevent Ischemia-Induced Large-Conductance Channel Activity.

Global ischemia promotes appearance of large-conductance (>750 pS) channel activity in the outer membranes of mitochondria isolated from postischemic hippocampus 1 h after reperfusion (16). Similar activity is induced in control mitochondria by introduction of recombinant ΔN61-Bcl-xL via the patch pipette (data not shown), and the activity in postischemic mitochondria is blocked by an antibody to Bcl-xL, consistent with a role for ΔN61-Bcl-xL in channel formation (16). To examine the impact of preconditioning on large-conductance channel activity, we recorded by patch clamping mitochondria isolated from the whole hippocampus of control and experimental animals (at times corresponding to those above). Mitochondria from the hippocampus of sham-treated animals exhibited primarily small channel openings (up to 180 pS; Fig. 3a) at low frequency and a few intermediate conductance openings and were predominantly in the closed state (Fig. 3 a and b). In contrast, mitochondria from the hippocampus of PC or LY animals exhibited more frequent openings of intermediate conductance (180–750 pS; Fig. 3 a and b) and few large-conductance openings (> 750 pS; Fig. 3c). Mitochondria from postischemic hippocampus showed markedly increased large-conductance activity in confirmation of earlier work (16) (Fig. 3 a and c). However, intermediate- and large-channel activity after PC + Isch was near that after PC alone (Fig. 3 a–c). In mitochondria from PC + Isch + LY animals, large-channel activity was increased relative to PC + Isch; i.e., the protective effect of PC was reduced, although not significantly (Fig. 3 a–c).

Fig. 3.

Fig. 3.

Open in a new tab

Preconditioning acts via PI3K signaling to prevent ischemia-induced formation of large-conductance channels in mitochondrial outer membrane and mitochondrial release of cytochrome c and Smac/DIABLO. (a) Sample patch recordings from the outer membrane of mitochondria isolated from the hippocampus of indicated groups. The lowest level for the PC + Isch + LY trace is 45 pA. (b and c) Histograms of percent recording time in which a patch showed open intermediate-conductance (180–750 pS, stippled lines) or large-conductance (>750 pS, black) channels for recordings like those in a; n values are indicated. (d) Current–voltage relations for a representative large-conductance channel and an inactive patch recorded from mitochondria from postischemic and control animals, respectively. (e) Current–voltage relations for a large-conductance channel recorded from a PC + Isch + LY and low-conductance channels from LY and PC + Isch animals. (f–h) Western blot of cytosolic cytochrome c at 1, 6, and 12 h (plus 48 h for PC). (i and j) Western blot of cytosolic Smac/DIABLO at 6 and 12 h. n values are indicated on bars.

The current–voltage relations of the large-conductance channels were linear with a reversal potential near zero (Fig. 3d). The current–voltage relations of the large-conductance activity of the mitochondria from PC + Isch + LY animals were also linear with a reversal potential near zero (Fig. 3e). These findings indicate that preconditioning acts via PI3K signaling to prevent the appearance of large-conductance channel activity in the mitochondrial outer membrane.

Preconditioning Acts via PI3K Signaling to Prevent Mitochondrial Release of Proapoptotic Proteins.

Injurious stimuli that are sufficiently potent promote the mitochondrial release of cytochrome c and Smac/DIABLO (5–7). To examine a possible role for PI3K signaling in protection of the mitochondrial outer membrane by preconditioning, we subjected animals to the procedures described above and examined the subcellular distribution of cytochrome c and Smac/DIABLO by Western blot analysis of mitochondrial and cytosolic fractions at 1, 6, and 12 h after reperfusion (or corresponding times in sham-operated, PC, and LY animals). Ischemia increased the abundance of cytochrome c in the cytosol of CA1, evident at 1 h (to 160 ± 18% of control; Fig. 3f), 6 h (to 196 ± 27% of control; Fig. 3g), and 12 h (to 314 ± 55% of control; Fig. 3h) after reperfusion, as assessed by Western blot analysis. Ischemia also increased the abundance of Smac/DIABLO in the cytosol at 6 h (to 191 ± 22% of control, P < 0.01) and to a lesser extent at 12 h (to 144 ± 9% of control, P < 0.05; Fig. 3 i and j). These data are in confirmation of previous studies (refs. 19 and 26 but see ref. 27). Preconditioning alone had little effect on cytosolic cytochrome c, but it blocked the ischemia-induced increases in cytosolic cytochrome c at all times of measurement (Fig. 3 f–j: * and **, P < 0.05 and 0.01 vs. sham; # and ##, P < 0.05 and 0.01 vs. ischemia alone) and Smac/DIABLO at 6 h (Fig. 3i; P < 0.05). Neither Isch nor PC caused a significant change in the mitochondrial fraction (at 6 h and 12 h; data not illustrated, but see ref. 19). Together, these findings indicate that preconditioning prevents the mitochondrial release of cytochrome c and Smac/DIABLO in the face of ischemic insults (19). LY294002 alone did not significantly alter cytosolic levels of cytochrome c at 1 and 12 h (Fig. 3 f and h) or Smac/DIABLO at 6 and 12 h (Fig. 3 i and j) but transiently increased cytochrome c release at 6 h (to 231 ± 23% of control; Fig. 3g, P < 0.01). Moreover, in PC + Isch + LY animals the protective effects of preconditioning were reduced; ischemia-induced cytochrome c release was greater than in PC + Isch animals [at 1 h to 151 ± 17% of control (Fig. 3f), at 6 h to 418 ± 56% of control (Fig. 3g), and at 12 h to 277 ± 75% of control (Fig. 3h); **, P < 0.01 for all times). Collectively, these data indicate that PI3K activity is required for the protective effect of preconditioning on the mitochondrial outer membrane.

Preconditioning Acts via PI3K Signaling to Block Caspase-3 Activity.

Cytochrome c release triggers formation of the apoptosome, which proteolytically processes procaspase-9 to generate the active caspase-9 (28, 29), which in turn cleaves procaspase-3 to form the active caspase-3, a terminator protease implicated in the execution step of apoptosis (30, 31). To directly measure caspase-3 activation, we used FAM-DEVD-FMK, a fluorescein-tagged analog of zDEVD-FMK. FAM-DEVD-FMK is a potent and cell-permanent inhibitor of caspase-3, which enters cells and binds irreversibly to the active site of caspase-3, and thus provides a fluorescent indicator of the abundance of activated caspase-3 (while inactivating it). In sections of control brain, abundance of activated caspase-3 was low (Fig. 4 a and c). Global ischemia induced a dramatic increase in activated caspase-3 in CA1, determined here at 24 h (see also ref. 19) (P < 0.01; Fig. 4 a and c). The increase in activated caspase-3 was region-specific in that it was not observed in the resistant CA3 or dentate gyrus. Preconditioning largely prevented the ischemic-induced increase in activated caspase-3 in CA1 pyramidal neurons (P < 0.01; Fig. 4 a and c). LY294002 alone did not significantly induce caspase-3 activity, but it reduced the protective effect of preconditioning on ischemia-induced formation of activated caspase-3 (P < 0.01; Fig. 4 a and c). These results confirm that preconditioning prevents the ischemia-induced increase in activated caspase-3 and indicate that PI3K signaling is required for this effect.

Fig. 4.

Fig. 4.

Open in a new tab

Preconditioning acts via PI3K signaling to reduce caspase-3 activity and neuronal death in postischemic hippocampus. (a) Caspase-3 activity at 24 h. (b) TUNEL positivity (green) and DAPI staining (blue) of coronal brain sections at the level of the dorsal hippocampus at 72 h. Treatment rows and magnifications are as indicated. (c) Summary data for caspase activity, fluorescence in arbitrary units (AU). (d) Quantification of TUNEL-positive cells.

Preconditioning Acts via PI3K Signaling to Prevent Neuronal Apoptosis.

Caspase-3 cleaves the inhibitory subunit of caspase-activated DNase (ICAD) of the CAD-ICAD heterodimer to generate activated CAD (5–7). CAD or DNA fragmentation factor (DFF) catalyzes internucleosomal cleavage of genomic DNA (19). If preconditioning indeed acts upstream of CAD activation, it would be expected to block DNA fragmentation in CA1 neurons destined to die. TUNEL is a labeling method in which a terminal transferase “tags” nicked ends present in fragmented DNA. To examine DNA fragmentation in neurons undergoing apoptosis, we labeled brain sections from control and experimental animals at 72 h after ischemia with TUNEL to assess DNA fragmentation and DAPI to assess cell numbers and possible alterations in nuclear morphology. In sections from control brain, TUNEL labeling was undetectable in CA1 (Fig. 4 b, a′, and d) and other subfields of the hippocampus (data not illustrated). Global ischemia induced a marked increase in the incidence of TUNEL-positive CA1 neurons, evident at 72 h after ischemia [P < 0.01; Fig. 4 b, b′ (higher magnification in Right), and d]. Preconditioning completely blocked ischemia-induced DNA fragmentation indicated by TUNEL (P < 0.01; Fig. 4 b, c′, and d). LY294002 alone did not elicit TUNEL labeling, but it greatly reduced the protective effect of preconditioning on ischemia-induced TUNEL positivity (P < 0.01; Fig. 4 b, e′, and d). These results indicate that preconditioning acts via PI3K signaling to prevent ischemia-induced neuronal apoptosis and that, whereas blockade of PI3K elicits transient cytochrome c release, it does not elicit caspase-3 activation, DNA fragmentation, or neuronal death.

Preconditioning Acts via PI3K Signaling to Protect Insulted CA1 Neurons.

To examine whether PI3K is involved in the neuroprotection provided by preconditioning, animals were subjected to sham operation, PC, Isch, PC + Isch, LY, or PC + Isch + LY. Neuronal death in CA1 was assessed histologically in animals terminated 7 days after sham operation, LY294002 injection alone, ischemia alone, or ischemia in preconditioned animals with LY294002 injection at times from 1 h before to 120 h after reperfusion. Ischemia caused extensive and selective death of pyramidal cells in the CA1 layer (P < 0.01 vs. sham; Fig. 5a, compare c′ with a′), in confirmation of other work (1, 4). These findings were quantified by neuronal counts (Fig. 5b). Preconditioning did not itself induce neuronal damage (Fig. 5 a, b′), but it afforded robust protection of CA1 neurons against the subsequent ischemic insult (Fig. 5 a, d′). LY294002 by itself did not induce neuronal death (Fig. 5 a, e′), but it largely reversed preconditioning-induced neuroprotection when given at 4 h after ischemia (Fig. 5a, f′). LY294002 reduced preconditioning-induced ischemic tolerance when given 1 h before ischemia or even 120 h after ischemia (P < 0.01; Fig. 5b). These results suggest that continued PI3K activity is necessary for preconditioning-induced neuroprotection and that ischemia causes a change lasting at least 5 days that can induce cell death if PI3K is inhibited. Maintained PI3K inhibition with multiple injections of LY294002 might lead to greater block of ischemic tolerance.

Fig. 5.

Fig. 5.

Open in a new tab

The PI3K inhibitor LY294002 reduces ischemic tolerance as late as 120 h after ischemia. (a) Toluidine blue-stained coronal brain sections at the level of the dorsal hippocampus. Treatments are as indicated, with higher magnification in Right. Animals were terminated 9 days after preconditioning, 7 days after LY294002 alone, and 7 days after ischemia in PC + Isch and PC + Isch +LY experiments. (f′) LY294002 (single injection) was administered at 4 h after PC + Isch. (b) Neuronal counts. PC and LY alone were no different from sham. LY (single injection at the indicated times relative to Isch) decreased the PC-induced protection against neurodegeneration in animals terminated 7 days after Isch. n values are shown for each bar.

Discussion

Ischemic preconditioning affords robust protection of CA1 neurons against a subsequent severe ischemic challenge. Although the molecular mechanisms underlying ischemic tolerance are only partially understood, a crucial role for PI3K/Akt signaling is clear. Here we show that preconditioning acts via PI3K/Akt signaling to block ischemia-induced mitochondrial translocation of Bad, cleavage of Bcl-xL to form its prodeath fragment, ΔN-Bcl-xL, activation of large-conductance channels in the mitochondrial outer membrane, mitochondrial release of cytochrome c and Smac/DIABLO, caspase activation, DNA fragmentation, and neuronal death. The specific PI3K inhibitor LY294002 administered in vivo 1 h before or up to 120 h after ischemia significantly reverses preconditioning-induced protection, indicating a requirement for sustained PI3K signaling in this protection. A caveat concerning the mitochondrial channel recordings is that the mitochondria were not specifically from CA1. Subsequent experiments show that ischemia-induced large-conductance channels are more frequent in CA1 than in CA3 mitochondria, but effects of preconditioning and LY294002 have not yet been evaluated.

Our finding that PI3K activity is required for preconditioning-induced neuroprotection is consistent with findings of others that wortmannin administered before preconditioning reduces ischemic tolerance and Akt phosphorylation in global (20) and focal (32) ischemia. Others showed that Akt phosphorylation is increased by preconditioning (21) and that LY294002 increases death in CA1 after a moderate ischemic insult (33). Moreover, in a very different model, neonatal rat brain slices, PI3K blockade during and immediately after hypoxia reduces Akt phosphorylation and promotes mitochondrial release of cytochrome c and the onset of neuronal apoptosis (34).

We observed that LY294002 alone, in the absence of a neuronal insult, activates intermediate-conductance (but few large-conductance) channels in the mitochondrial outer membrane, whereas ischemia activates both intermediate- and large-conductance channels. These findings implicate large-conductance channels in neuronal death. LY294002 alone also promotes transient release of cytochrome c, which does not lead to caspase-3 activity or neuronal death. That neurons can survive cytochrome c release was unexpected, because a rise in cytochrome c was thought to lead inexorably to postischemic cell death (5–7). A more recent study shows that microinjection of cytochrome c into the cytosol of neurons is not sufficient to cause cell death (35). Whereas ischemia promotes mitochondrial release of cytochrome c and Smac/DIABLO, PI3K blockade promotes transient release of cytochrome c only. The enhanced cytosolic levels of cytochrome c are evident at 6 h but not at 1 h or 12 h. Moreover, LY294002 does not promote Bcl-xL cleavage, caspase-3 activity, DNA fragmentation, or cell loss. These findings are consistent with a quantitative model in which cell death requires that sufficient cytochrome c must be present for sufficiently long; neurons may survive if the integrity of the mitochondrial outer membrane is not fully breached, if affected mitochondria seal up, or if they are removed by autophagy.

We previously showed that preconditioning causes substantial caspase-3 activation from 12 to 48 h afterward (earlier than the measurements here), but without cell death (19). Caspase-3 is still activated by ischemia in preconditioned animals, but the activation is transient, and the caspase-3 activity is inhibited. We did not determine whether preconditioning elevates cytochrome c in the cytosol before caspase-3 activation. This study also showed that preconditioning protects the integrity of the mitochondrial outer membrane against a later severe ischemic insult in that cytochrome c and Smac/DIABLO are not released (19). The present study confirms and extends these findings by indicating mediation by the PI3K pathway.

It was surprising that LY294002 reduces protection even 120 h after ischemia. We presume that removal of PI3K activity reveals ischemia-induced changes that persist in the presence of PI3K activity and that can still cause cell death when PI3K is inhibited.

Our study has further dissected the mitochondrial cell death signaling pathway in CA1 after ischemia. Although the neuroprotective mechanisms after preconditioning have not been fully elucidated, an inference from the present study is that preconditioning leads to preservation of the integrity of the mitochondrial membrane in the face of ischemic insults and thereby enables CA1 neurons to survive. Diverse protective strategies may intervene at different steps in the caspase cascade. An understanding of the molecular mechanisms underlying ischemic tolerance should help in the design of novel neuroprotective strategies for intervention in the neuronal death associated with stroke, cardiac arrest, and various neurodegenerative disorders.

Materials and Methods

Preconditioning, Global Ischemia, and Drug Injection.

Male Sprague–Dawley rats (100–150 g) were subjected to preconditioning, global ischemia, or preconditioning followed by ischemia, by four-vessel occlusion (preconditioning, 4 min; ischemia, 10 min), followed by reperfusion, as described (36, 37). LY294002 (12.5 mM in 5 μl of 25% DMSO) was administered by a single injection into the right lateral ventricle at a flow rate of 1 μl/min immediately after the last surgery. PC, Isch, and PC + Isch animals in the same series were injected with vehicle only.

Mitochondrial Preparation.

Mitochondria were isolated from control and experimental rat hippocampus and purified by centrifugation over a discontinuous Ficoll gradient (38). The mitochondrial fraction was frozen on dry ice and stored at − 80°C until use.

Caspase-3 Activity, TUNEL, and Histology.

Caspase-3 activity assays were performed on sections of fresh-frozen rat brain using the FAM-DEVD-FMK caspase detection kit according to the manufacturer's instructions (APO LOGIX; Cell Technology). To detect DNA fragmentation, animals were killed at 72 h after the last surgery and coronal sections (18 μm) of brain were processed for TUNEL and DAPI-labeled with an in situ cell death detection kit as per the manufacturer's instructions (Roche Molecular Biochemicals) as described (19).

Western Blot Analyses and Coimmunoprecipitation.

For Western blots, protein samples of CA1 were prepared and run on SDS/PAGE as described (39). For coimmunoprecipitation experiments, samples were precipitated with the indicated antibody (IP), precipitates were resuspended and run on SDS/PAGE, and membranes were blotted with a second antibody (IB).

Additional Details.

For more details see SI Materials and Methods.

Supplementary Material

Supporting Information
pnas_0800628105_index.html (14.2KB, html)

Acknowledgments.

This work was supported by National Institutes of Health Grants NS46742 and NS45693 (to R.S.Z.) and NS37402 and NS45287 (to M.V.L.B.) and by a generous grant from the F. M. Kirby Foundation Program in Neural Repair and Neuroprotection. M.V.L.B. is the Sylvia and Robert S. Olnick Professor of Neuroscience and Distinguished Professor of Albert Einstein College of Medicine.

Footnotes

The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0800628105/DC1.

References

  • 1.Lo EH, Dalkara T, Moskowitz MA. Mechanisms, challenges and opportunities in stroke. Nat Rev Neurosci. 2003;4:399–415. doi: 10.1038/nrn1106. [DOI] [PubMed] [Google Scholar]
  • 2.MacManus JP, Buchan AM. Apoptosis after experimental stroke: Fact or fashion? J Neurotrauma. 2000;17:899–914. doi: 10.1089/neu.2000.17.899. [DOI] [PubMed] [Google Scholar]
  • 3.Graham SH, Chen J. Programmed cell death in cerebral ischemia. J Cereb Blood Flow Metab. 2001;21:99–109. doi: 10.1097/00004647-200102000-00001. [DOI] [PubMed] [Google Scholar]
  • 4.Zukin RS. In: Stroke. Mohr J, Choi D, Grotta J, Weir B, Wolf P, editors. Philadelphia: Churchill Livingstone; 2004. pp. 829–854. [Google Scholar]
  • 5.Taylor RC, Cullen SP, Martin SJ. Apoptosis: Controlled demolition at the cellular level. Nat Rev Mol Cell Biol. 2008;9:231–241. doi: 10.1038/nrm2312. [DOI] [PubMed] [Google Scholar]
  • 6.Riedl SJ, Salvesen GS. The apoptosome: Signalling platform of cell death. Nat Rev Mol Cell Biol. 2007;8:405–413. doi: 10.1038/nrm2153. [DOI] [PubMed] [Google Scholar]
  • 7.Hengartner MO. The biochemistry of apoptosis. Nature. 2000;407:770–776. doi: 10.1038/35037710. [DOI] [PubMed] [Google Scholar]
  • 8.Chen J, et al. Induction of caspase-3-like protease may mediate delayed neuronal death in the hippocampus after transient cerebral ischemia. J Neurosci. 1998;18:4914–4928. doi: 10.1523/JNEUROSCI.18-13-04914.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Antignani A, Youle RJ. How do Bax and Bak lead to permeabilization of the outer mitochondrial membrane? Curr Opin Cell Biol. 2006;18:685–689. doi: 10.1016/j.ceb.2006.10.004. [DOI] [PubMed] [Google Scholar]
  • 10.Jonas E. BCL-xL regulates synaptic plasticity. Mol Interv. 2006;6:208–222. doi: 10.1124/mi.6.4.7. [DOI] [PubMed] [Google Scholar]
  • 11.Boise LH, et al. bcl-x, a bcl-2-related gene that functions as a dominant regulator of apoptotic cell death. Cell. 1993;74:597–608. doi: 10.1016/0092-8674(93)90508-n. [DOI] [PubMed] [Google Scholar]
  • 12.Krajewski S, et al. Immunohistochemical analysis of in vivo patterns of Bcl-X expression. Cancer Res. 1994;54:5501–5507. [PubMed] [Google Scholar]
  • 13.Kaufmann T, et al. Characterization of the signal that directs Bcl-x(L), but not Bcl-2, to the mitochondrial outer membrane. J Cell Biol. 2003;160:53–64. doi: 10.1083/jcb.200210084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Clem RJ, et al. Modulation of cell death by Bcl-XL through caspase interaction. Proc Natl Acad Sci USA. 1998;95:554–559. doi: 10.1073/pnas.95.2.554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Fujita N, et al. Acceleration of apoptotic cell death after the cleavage of Bcl-XL protein by caspase-3-like proteases. Oncogene. 1998;17:1295–1304. doi: 10.1038/sj.onc.1202065. [DOI] [PubMed] [Google Scholar]
  • 16.Bonanni L, et al. Zinc-dependent multi-conductance channel activity in mitochondria isolated from ischemic brain. J Neurosci. 2006;26:6851–6862. doi: 10.1523/JNEUROSCI.5444-05.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Gidday JM. Cerebral preconditioning and ischaemic tolerance. Nat Rev Neurosci. 2006;7:437–448. doi: 10.1038/nrn1927. [DOI] [PubMed] [Google Scholar]
  • 18.Kirino T. Ischemic tolerance. J Cereb Blood Flow Metab. 2002;22:1283–1296. doi: 10.1097/01.WCB.0000040942.89393.88. [DOI] [PubMed] [Google Scholar]
  • 19.Tanaka H, et al. Ischemic preconditioning: Neuronal survival in the face of caspase-3 activation. J Neurosci. 2004;24:2750–2759. doi: 10.1523/JNEUROSCI.5475-03.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Yano S, et al. Activation of Akt/protein kinase B contributes to induction of ischemic tolerance in the CA1 subfield of gerbil hippocampus. J Cereb Blood Flow Metab. 2001;21:351–360. doi: 10.1097/00004647-200104000-00004. [DOI] [PubMed] [Google Scholar]
  • 21.Nakajima T, et al. Preconditioning prevents ischemia-induced neuronal death through persistent Akt activation in the penumbra region of the rat brain. J Vet Med Sci. 2004;66:521–527. doi: 10.1292/jvms.66.521. [DOI] [PubMed] [Google Scholar]
  • 22.Hashiguchi A, et al. Up-regulation of endothelial nitric oxide synthase via phosphatidylinositol 3-kinase pathway contributes to ischemic tolerance in the CA1 subfield of gerbil hippocampus. J Cereb Blood Flow Metab. 2004;24:271–279. doi: 10.1097/01.WCB.0000110539.96047.FC. [DOI] [PubMed] [Google Scholar]
  • 23.Datta SR, et al. Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell. 1997;91:231–241. doi: 10.1016/s0092-8674(00)80405-5. [DOI] [PubMed] [Google Scholar]
  • 24.Abe T, et al. Altered Bad localization and interaction between Bad and Bcl-xL in the hippocampus after transient global ischemia. Brain Res. 2004;1009:159–168. doi: 10.1016/j.brainres.2004.03.003. [DOI] [PubMed] [Google Scholar]
  • 25.Cheng EH, et al. BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and BAK-mediated mitochondrial apoptosis. Mol Cell. 2001;8:705–711. doi: 10.1016/s1097-2765(01)00320-3. [DOI] [PubMed] [Google Scholar]
  • 26.Sugawara T, et al. Mitochondrial release of cytochrome c corresponds to the selective vulnerability of hippocampal CA1 neurons in rats after transient global cerebral ischemia. J Neurosci. 1999;19:RC39. doi: 10.1523/JNEUROSCI.19-22-j0002.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Colbourne F, Sutherland GR, Auer RN. Electron microscopic evidence against apoptosis as the mechanism of neuronal death in global ischemia. J Neurosci. 1999;19:4200–4210. doi: 10.1523/JNEUROSCI.19-11-04200.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Li P, et al. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell. 1997;91:479–489. doi: 10.1016/s0092-8674(00)80434-1. [DOI] [PubMed] [Google Scholar]
  • 29.Zou H, et al. Apaf-1, a human protein homologous to C. elegans CED-4, participates in cytochrome c-dependent activation of caspase-3. Cell. 1997;90:405–413. doi: 10.1016/s0092-8674(00)80501-2. [DOI] [PubMed] [Google Scholar]
  • 30.Cohen GM. Caspases: The executioners of apoptosis. Biochem J. 1997;326:1–16. doi: 10.1042/bj3260001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Nicholson DW, Thornberry NA. Caspases: Killer proteases. Trends Biochem Sci. 1997;22:299–306. doi: 10.1016/s0968-0004(97)01085-2. [DOI] [PubMed] [Google Scholar]
  • 32.Noshita N, Lewen A, Sugawara T, Chan P-H. Evidence of phosphorylation of Akt and neuronal survival after transient focal cerebral ischemia in mice. J Cereb Blood Flow Metab. 2001;21:1442–1450. doi: 10.1097/00004647-200112000-00009. [DOI] [PubMed] [Google Scholar]
  • 33.Endo H, et al. Activation of the Akt/GSK3beta signaling pathway mediates survival of vulnerable hippocampal neurons after transient global cerebral ischemia in rats. J Cereb Blood Flow Metab. 2006;26:1479–1489. doi: 10.1038/sj.jcbfm.9600303. [DOI] [PubMed] [Google Scholar]
  • 34.Hirai K, et al. PI3K inhibition in neonatal rat brain slices during and after hypoxia reduces phospho-Akt and increases cytosolic cytochrome c and apoptosis. Brain Res Mol Brain Res. 2004;124:51–61. doi: 10.1016/j.molbrainres.2004.02.009. [DOI] [PubMed] [Google Scholar]
  • 35.Deshmukh M, Johnson EM., Jr Evidence of a novel event during neuronal death: Development of competence-to-die in response to cytoplasmic cytochrome c. Neuron. 1998;21:695–705. doi: 10.1016/s0896-6273(00)80587-5. [DOI] [PubMed] [Google Scholar]
  • 36.Calderone A, et al. Ischemic insults derepress the gene silencer REST in neurons destined to die. J Neurosci. 2003;23:2112–2121. doi: 10.1523/JNEUROSCI.23-06-02112.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kinouchi H, et al. Induction of heat shock hsp70 mRNA and HSP70 kDa protein in neurons in the ‘penumbra’ following focal cerebral ischemia in the rat. Brain Res. 1993;619:334–338. doi: 10.1016/0006-8993(93)91630-b. [DOI] [PubMed] [Google Scholar]
  • 38.Sullivan PG, et al. Intrinsic differences in brain and spinal cord mitochondria: Implication for therapeutic interventions. J Comp Neurol. 2004;474:524–534. doi: 10.1002/cne.20130. [DOI] [PubMed] [Google Scholar]
  • 39.Opitz T, Grooms SY, Bennett MVL, Zukin RS. Remodeling of alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid receptor subunit composition in hippocampal neurons after global ischemia. Proc Natl Acad Sci USA. 2000;97:13360–13365. doi: 10.1073/pnas.97.24.13360. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
pnas_0800628105_index.html (14.2KB, html)
pnas_0800628105_1.pdf (85.9KB, pdf)
pnas_0800628105_2.pdf (85.9KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES