Activating δPKC antagonizes the protective effect of ERK1/2 inhibition against stroke in rats

doi:10.1016/j.brainres.2008.11.051

Brain Res. Author manuscript; available in PMC 2009 September 18.

Published in final edited form as:

Brain Res. 2009 January 28; 1251: 256–261.

Published online 2008 November 27. doi: 10.1016/j.brainres.2008.11.051

PMCID: PMC2746701

NIHMSID: NIHMS123643

Activating δPKC antagonizes the protective effect of ERK1/2 inhibition against stroke in rats

Dora Castañeda,^a Heng Zhao,^a^c Daria Mochly-Rosen,^a^b^c and Gary K. Steinberg^a^c^*

^aDepartment of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA

^bMolecular Pharmacology, Stanford University School of Medicine, Stanford, CA, USA

^cStanford Stroke Center, Stanford University School of Medicine, Stanford, CA, USA

^*Corresponding author. Department of Neurosurgery, Stanford University, School of Medicine, 300 Pasteur Drive R200, Stanford, CA 94305-5327, USA. Fax: +1 650 723 2815. Email: gsteinberg/at/stanford.edu (G.K. Steinberg).

The publisher's final edited version of this article is available at Brain Res

Abstract

Two pathways that have been shown to mediate cerebral ischemic damage are the MEK/ERK cascade and the pro-apoptotic δPKC pathway. We investigated the relationship between these pathways in a rat model of focal ischemia by observing and modifying the activation state of each pathway. The ERK1/2 inhibitor, U0126, injected at ischemia onset, attenuated the increase in phosphorylated ERK1/2 (P-ERK1/2) after reperfusion. The δPKC inhibitor, δV1-1, delivered at reperfusion, did not significantly change P-ERK1/2 levels. In contrast, the δPKC activator, ψδRACK, injected at reperfusion, reduced ERK1/2 phosphorylation measured 4 h after reperfusion. Additionally, U0126 pretreatment at ischemia onset reduced infarct size compared with vehicle, but U0126 injected at the onset of reperfusion had no protection. Finally, combination of U0126 injection at ischemia onset plus δV1-1 injection at reperfusion further reduced infarct size, while combination of U0126 delivered at ischemia onset with ψδRACK injected at reperfusion increased infarct size compared with U0126 alone. In conclusion, we find that inhibiting both the MEK/ERK and the δPKC pathways offers greater protection than either alone, indicating they likely act independently.

Keywords: Cerebral ischemia, MEK/ERK cascade, δPKC, ERK1/2

1. Introduction

The mitogen-activated protein kinase (MAPK) and the protein kinase C (PKC) pathways are implicated in neuronal death and survival after stroke [4,5,23]. The best characterized members in the MAPK pathways are extracellular signal-regulated kinase 1 and 2 (ERK1/2), whose activity is regulated by a three tiered kinase module, Raf—MEK—ERK, and is increased when ERK1/2 are phosphorylated [25]. ERK1/2 activity is reported to be involved in ischemic injury: levels of phosphorylated ERK1/2 (P-ERK1/2) increase after stroke [23]. But previous studies have shown that increases in P-ERK1/2 are associated with both neuronal survival and death [16,23], and whether ERK1/2 activity exacerbates or attenuates ischemic injury is controversial [27].

The PKC family consists of 10 isozymes of serine/threonine kinases with distinct roles for neuronal survival [3]. For instance, δPKC, one member of the PKC family, is pro-apoptotic [4]. We and others have shown that δPKC activity increases after stroke [32]; it is cleaved, and translocates to the particulate membrane after stroke [24,32]. In addition, the δPKC-specific inhibitor peptide, δV1-1, reduces ischemic injury by blocking δPKC subcellular translocation [4] and the ψδRACK, the δPKC-selective activator, inhibits protection induced by lowering body temperature after stroke [32].

ERK1/2 interacts with the PKC pathways in non-neuronal systems [1,9,13]. PKC directly activates the MAPK pathway by phosphorylating Raf, a signaling molecule which functions upstream of ERK1/2 [6,17]. In addition, VEGF-induced ERK1/2 activation requires δPKC activation and translocation [18], and PKC has positive feedback on the MAPK pathway [2]. Conversely, activation of ERK1/2 during apoptosis induced by DNA damage involves δPKC [1]. Nevertheless, whether ERK1/2 interacts with δPKC after stroke, and how such interplay affects ischemic injury has not been studied.

In this study, we first examined whether inhibiting ERK1/2 by U0126, a MEK1/2 inhibitor, reduces infarct size after stroke, then clarified how the δPKC inhibitor, δV1-1, affects the levels of phosphorylated ERK1/2, and finally determined whether δPKC and ERK1/2 signaling pathways interact in the response to stroke.

2. Results

The effects of the ERK1/2 inhibitor (U0126), the δPKC inhibitor (δV1-1), and the δPKC activator (ψδRACK) on the levels of phosphorylated ERK1/2 (P-ERK1/2) were evaluated/determined. The results of Western blot analysis showed that P-ERK1/2 levels transiently increased from 1 to 4 h after reperfusion in the ischemic cortex (Fig. 1), which is consistent with previous reports [23]. As expected, the ERK1/2 inhibitor, U0126, injected at ischemia onset, attenuated increases in P-ERK1/2 levels at 4 h (Fig. 2). Interestingly, the δPKC-selective inhibitor, δV1-1, did not change P-ERK1/2 levels compared with the vehicle. In contrast, the δPKC-selective activator, ψδRACK, reduced ERK1/2 phosphorylation when measured 4 h after reperfusion (Fig. 2). Total levels of ERK did not change (data not shown).

Fig. 1

P-ERK1/2 levels transiently increased in the ischemic penumbra after reperfusion. (A) Representative Western blot analysis with anti-P-ERK1/2 and anti-ERK1/2 of samples from the ischemic cortex (penumbra) at 0 min, 15 min, 1 h, 2 h, 4 h, 24 h and 48 h (more ...)

Fig. 2

The effect of δV1-1, U0126 and ψδRACK on ERK1/2 phosphorylation at 4 h after reperfusion. U0126 was delivered at ischemia onset while δV1-1 and ψδRACK were injected at reperfusion. Sham rats received sham (more ...)

Then we examined the effect of U0126, δV1-1 and ψδRACK, as well as some combinations of these drugs on infarct size. U0126 pretreatment at ischemia onset reduced infarct size compared with vehicle, but U0126 delivered at the onset of reperfusion had no protection (Fig. 3). We have previously reported that delivery of δV1-1 at reperfusion reduced infarct size in the same model used in this study [4]. In the current study, we found that combination treatment with the δPKC inhibitor δV1-1 (delivered at the onset of reperfusion) together with U0126 at the onset of ischemia tended to decrease infarct size further, but it did not reach significant difference. Interestingly, combination treatment with the δPKC activator ψδRACK (delivered at the onset of reperfusion) together with U0126 (delivered at ischemia onset) inhibited the protective effect of U0126; the infarct size with both drugs was larger as compared with U0126 alone (P < 0.001) (Fig. 3).

Fig. 3

The effects of ERK1/2 inhibitior and δPKC inhibitor or activator on infarct size. U0126 at ischemia, U0126 was delivered at ischemia onset; U0126 at reperfusion, U0126 was injected at reperfusion; U0126+δV1-1, U0126 (injected at ischemia (more ...)

3. Discussion

In this study we confirmed that P-ERK1/2 levels transiently increase after reperfusion [23], and the ERK1/2 inhibitor, U0126, given at the onset of stroke, reduces infarction by decreasing P-ERK1/2 levels [12,29], suggesting a detrimental role of ERK1/2 activity in stroke. In addition, we found that the δPKC inhibitor, δV1-1, delivered at the onset of reperfusion tended to enhance the protective effect of U0126, although it increased P-ERK1/2 levels. Thus, the protective effect of δv1-1 was not achieved by reducing P-ERK1/2 levels. Conversely, treatment with the δPKC activator, ψδRACK, which decreased ERK phosphorylation, partly abolished the protection induced by U0126, a MEK inhibitor. These data suggest that the detrimental effect of δPKC activation by ψδRACK is independent of the ERK1/2 activity and phosphorylation.

P-ERK1/2 is implicated in ischemic damage or neuronal survival after stroke. Previous studies generally agree that P-ERK1/2 transiently increases after reperfusion in both global and focal ischemia [10,11]. Our current study is consistent with these previous reports [11]; we showed that P-ERK1/2 levels transiently increased from 1 to 4 h after reperfusion, and then returned to baseline at 24 and 48 h.

Whether increases in P-ERK1/2 levels are detrimental or protective is controversial, as an increase in P-ERK1/2 after stroke is linked with both detrimental [12] and protective effects [22]. Several lines of evidence support the protective effects of P-ERK1/2. First, P-ERK1/2 is expressed in the ischemic penumbra after focal ischemia [8,14]. Second, P-ERK1/2 is expressed in the ischemic resistant region of dentate gyrus in the hippocampus after transient global ischemia [10,30]. Third, P-ERK1/2 is enhanced by a number of neuro-protectants, including various growth factors; in addition, inhibition of the blood brain barrier permeability augments P-ERK1/2 [16]. Alternatively, some previous studies also strongly support the detrimental effects of ERK1/2 activity. For instance, ERK1/2 inhibition reduces ischemic damage [12] and P-ERK1/2 levels are increased by free radical products [23], inflammatory response [20,26] and hyperglycemia [7], which exacerbate ischemic damage.

Neverthless, in the current study we found that injection of the ERK1/2 inhibitor, U0126, at ischemia onset robustly reduced infarct size while it had no effect when injected at reperfusion, suggesting a limited therapeutic time window. The underlying mechanisms responsible for such difference are not known, but our pilot study showed that ERK1/2 phosphorylation was increased as early as 10 min after ischemia onset (data not shown), indicating that ERK activity might have been triggered before reperfusion. U0126 delivered at reperfusion, thus, might not be able to attenuate infarct size as long as ERK activity started, which, however, needs further study.

Our current study using the δPKC inhibitor and agonist reinforces the complicated role of ERK1/2 after stroke. The interplay between δPKC and ERK/2 has been studied in the non-neuronal systems, in which δPKC activity increases P-ERK1/2 [1,9,13]. Thus, one would expect that activating δPKC would increase P-ERK1/2 and inhibiting δPKC would decrease P-ERK1/2. Surprisingly, the δPKC activator, ψδRACK, which increases δPKC activity, actually attenuated protein levels of P-ERK1/2, and inhibiting δPKC did not block increases in P-ERK1/2 in our study. Thus, our results are not consistent with previous reports regarding interaction between δPKC activity and P-ERK1/2 levels [18]. The ERK1/2 inhibitor, U0126, reduced infarct size and protein levels of P-ERK1/2, suggesting that reduction in P-ERK1/2 may be necessary for the protective effect of U0126. However, the δPKC inhibitor δV1-1, which we previously reported to block ischemic damage [4,5], did not reduce P-ERK1/2 levels. Furthermore, the δPKC activator, ψδRACK, inhibited P-ERK1/2 levels, yet it abolished the protective effect of the ERK1/2 inhibitor, U0126. Thus, protein levels of P-ERK1/2 itself may not be as critical for determining the protective or detrimental effect of ERK1/2; some other factors may be responsible for the effects of ERK1/2. It is well known that stroke causes changes in a myriad of genes and proteins [15]. Although U0126 is an ERK1/2 inhibitor, it may also affect many other proteins simultaneously. Likewise, P-ERK1/2 is not the only protein altered by the δPKC activator or inhibitor. Thus, P-ERK1/2 appears to have a complex role in mediating injury or survival of brain after stroke.

In conclusion, we demonstrated for the first time that inhibition of both ERK1/2 and the δPKC pathways offers greater protection than does inhibition of either alone, suggesting they act in parallel. Our results imply that the protein level of P-ERK1/2 is not a simple marker for cerebral neuroprotection or worsened injury after stroke.

4. Experimental procedures

Procedures using laboratory animals were approved by the Stanford University Administrative Panel on Laboratory Animal Care.

4.1. Focal ischemia model

Focal ischemia was induced by occluding the left middle cerebral artery (MCA) in male Sprague Dawley rats (250–300 g) as described previously [19,34]. Briefly, anesthesia was induced by a dose of 5% isoflurane and maintained by 2% isoflurane through a face mask during all surgical procedures. Physiological parameters, including respiration rate, were monitored. Body temperature was maintained at 37 °C using a heating blanket. The left common carotid artery was exposed, and incised. An uncoated 30 mm long segment of 3-0 nylon monofilament suture was inserted to occlude the ostium of the MCA. The suture was left in place for 2 h during which the rat was removed from the isoflurane. At the end of the ischemic period, the rat was re-anesthetized, the suture was removed, and the rat was allowed to recover.

4.2. Drug injection

The MEK-1 inhibitor, 1,4-diamino-2,3-dicyano-1,4-bis[2-aminophenylthio]butadiene (U0126) (Cell Signaling Technology, Danvers, MA) was dissolved in 2 ml DMSO (100%), and 1 ml of U0126 (10 mg/kg) was injected intraperitoneal right before ischemia onset or at onset of reperfusion. The same volume of vehicle (1 ml 100% DMSO) was injected at the onset of ischemia. The dosage was chosen based on the past studies that have shown that doses from 0.5 mg/kg [28] to 200 mg/kg [21] of U0126 reduced infarct size after focal ischemia. The δPKC-specific activator, ψδRACK (0.2 mg/kg), and the δPKC-specific inhibitor, δV1-1 (0.2 mg/kg) [4], were injected intraperitoneally at the onset of reperfusion. Peptide regulators of δPKC were designed and synthesized as described previously [4].

4.3. Assessment of brain infarct size

Rats were sacrificed after 24 h of reperfusion by overdoses of Isoflurane. Brains were quickly removed and sliced into 3 mm coronal sections, resulting in five slices. The slices were stained using 2% triphenyl tetrazolium chloride (TTC), and both faces of each slice were digitally scanned (EPSON®). Stroke regions of the five slices were measured and normalized to the contralateral hemisphere and expressed as percentages [31].

4.4. Western blot analysis [32,33]

Rat brains were harvested at various time points after ischemia for Western blot. The brains were quickly sliced. Since ERK1/2 phosphorylation was reported to occur in the ischemic cortex [23], the ischemic and non-ischemic cortex from each central slice were dissected and frozen until homogenization for detecting ERK1/2 phosphorylation. The tissues were added with a 7 times volume of 1x cell lysis buffer (Cell Signaling Technology, Danvers, MA) containing a protease inhibitor (Sigma) and phosphatase inhibitor cocktail I and II (Sigma), and homogenized by 35 strokes in 1 ml vessels then incubated in ice for 5 min and centrifuged at 14,000 ×g for 10 min at 4 °C, as described. Protein concentration was assessed using Bradford reagent (BioRad). In each lane 25 μg protein was subjected to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) using 4–15% Ready Gel (Cat# L050505A2, BIO-RAD Laboratories, USA) for 1.5 h. Proteins were transferred for 60 min to a nitrocellulose membrane (Amersham). The membrane was blocked in PBS solution containing 3% milk and 0.1% Tween 20 (Sigma), and then incubated in the solution containing a rabbit anti-phospho-ERK1/2(Thr202/Tyr204) antibody (1:1000; Cell Signaling Technology, Danvers, MA) overnight at 4 °C. Then the membrane was incubated in the solution containing a secondary anti-rabbit IgG HRP-linked antibody (1:2000, Cell Signaling Technology, Danvers, MA) for 1 h. Protein bands were detected using an enhanced chemiluminescence system (Amersham Biosciences, Arlington heights, IL) and exposed to hyperfilm. To determine even-loading of proteins, membranes were stripped and re-probed with a β-actin antibody (1:10,000; Cat#A3854 Sigma). Densities of protein bands from Western blot were measured by NIH Image J program.

Statistics

Data were expressed as the mean±SEM. All statistical analyses were performed using ANOVA followed by Bonferroni’s post hoc test.

Acknowledgments

This work was supported by NIH NINDS grants R01 NS27292 (GKS) and P01NS37520 (GKS), AHA grant SDG 0730113N (HZ), Stanford Medical Scholars Program (DC) and NS044350 (DM-R). The authors thank Ms. Elizabeth Hoyte for figure preparation.

Footnotes

Conflict of interest Dr. Mochly-Rosen is the founder of KAI Pharmaceuticals, Inc, a company that plans to bring PKC regulators to the clinic. However, none of the work described here is based on or supported by the company.

REFERENCES

Basu A, Tu H. Activation of ERK during DNA damage-induced apoptosis involves protein kinase Cdelta. Biochem. Biophys. Res. Commun. 2005;334:1068–1073. [PubMed]
Bhalla US, Ram PT, Iyengar R. MAP kinase phosphatase as a locus of flexibility in a mitogen-activated protein kinase signaling network. Science. 2002;297:1018–1023. [PubMed]
Mochly-Rosen D, Bright R. The role of protein kinase C in cerebral ischemia and reperfusion injury. Stroke. 2005;36:2781–2790. [PubMed]
RA, Bright R, Dembner JM, Perez-Pinzon MA, Steinberg G, Yenari M, Mochly-Rosen D. Protein kinase C delta mediates cerebral reperfusion injury in vivo. J. Neurosci. 2004;24:6880–6888. [PubMed]
Bright R, Steinberg GK, Mochly-Rosen D. DeltaPKC mediates microcerebrovascular dysfunction in acute ischemia and in chronic hypertensive stress in vivo. Brain Res. 2007;1144:146–155. [PubMed]
Carroll MP, May WS. Protein kinase C-mediated serine phosphorylation directly activates Raf-1 in murine hematopoietic cells. J. Biol. Chem. 1994;269:1249–1256. [PubMed]
Farrokhnia N, Roos MW, Terent A, Lennmyr F. Differential early mitogen-activated protein kinase activation in hyperglycemic ischemic brain injury in the rat. Eur. J. Clin. Invest. 2005;35:457–463. [PubMed]
Ferrer I, Friguls B, Dalfo E, Planas AM. Early modifications in the expression of mitogen-activated protein kinase (MAPK/ERK), stress-activated kinases SAPK/JNK and p38, and their phosphorylated substrates following focal cerebral ischemia. Acta Neuropathol. (Berl.) 2003;105:425–437. [PubMed]
Ginnan R, Guikema BJ, Singer HA, Jourd’heuil D. PKC-delta mediates activation of ERK1/2 and induction of iNOS by IL-1beta in vascular smooth muscle cells. Am. J. Physiol. Cell Physiol. 2006;290:C1583–1591. [PubMed]
Gu Z, Jiang Q, Zhang G. Extracellular signal-regulated kinase 1/2 activation in hippocampus after cerebral ischemia may not interfere with postischemic cell death. Brain Res. 2001;901:79–84. [PubMed]
Hasegawa Y, Morioka M, Hasegawa S, Matsumoto J, Kawano T, Kai Y, Yano S, Fukunaga K, Kuratsu J. Therapeutic time window and dose dependence of neuroprotective effects of sodium orthovanadate following transient middle cerebral artery occlusion in rats. J. Pharmacol. Exp. Ther. 2006;317:875–881. [PubMed]
Henriksson M, Stenman E, Vikman P, Edvinsson L. MEK1/2 inhibition attenuates vascular ETA and ETB receptor alterations after cerebral ischaemia. Exp. Brain. Res. 2007;178:470–476. [PubMed]
Hornberg J, Tijssen M, Lankelma J. Synergistic activation of signalling to extracellular signal-regulated kinases 1 and 2 by epidermal growth factor and 4 beta-phorbol 12-myristate 12-acetate. Eur. J. Biochem. 2004;271:3905–3913. [PubMed]
Irving EA, Barone FC, Reith AD, Hadingham SJ, Parsons AA. Differential activation of MAPK/ERK and p38/SAPK in neurones and glia following focal cerebral ischaemia in the rat. Brain Res. Mol. Brain Res. 2000;77:65–75. [PubMed]
Jin K, Mao XO, Eshoo MW, Nagayama T, Minami M, Simon RP, Greenberg DA. Microarray analysis of hippocampal gene expression in global cerebral ischemia. Ann. Neurol. 2001;50:93–103. [PubMed]
Kilic E, Kilic U, Soliz J, Bassetti CL, Gassmann M, Hermann DM. Brain-derived erythropoietin protects from focal cerebral ischemia by dual activation of ERK-1/-2 and Akt pathways. FASEB J. 2005;19:2026–2028. [PubMed]
Kolch W, Heidecker G, Kochs G, Hummel R, Vahidi H, Mischak H, Finkenzeller G, Marme D, Rapp U. Protein kinase C alpha activates RAF-1 by direct phosphorylation. Nature. 1993;364:249–252. [PubMed]
Kuriyama M, Taniguchi T, Shirai Y, Sasaki A, Yoshimura A, Saito N. Activation and translocation of PKCdelta is necessary for VEGF-induced ERK activation through KDR in HEK293T cells. Biochem. Biophys. Res. Commun. 2004;325:843–851. [PubMed]
Maier C, Sun G, Kunis D, Yenari M, Steinberg G. Delayed induction and long-term effects of mild hypothermia in a focal model of transient cerebral ischemia: neurological outcome and infarct size. J. Neurosurg. 2001;94:90–96. [PubMed]
Molina-Holgado E, Ortiz S, Molina-Holgado F, Guaza C. Induction of COX-2 and PGE(2) biosynthesis by IL-1beta is mediated by PKC and mitogen-activated protein kinases in murine astrocytes. Br. J. Pharmacol. 2000;131:152–159. [PubMed]
Namura S, Iihara K, Takami S, Nagata I, Kikuchi H, Matsushita K, Moskowitz MA, Bonventre JV, Alessandrini A. Intravenous administration of MEK inhibitor U0126 affords brain protection against forebrain ischemia and focal cerebral ischemia. Proc. Natl. Acad. Sci. U. S. A. 2001;98:11569–11574. [PubMed]
Niimura M, Takagi N, Takagi K, Funakoshi H, Nakamura T, Takeo S. Effects of hepatocyte growth factor on phosphorylation of extracellular signal-regulated kinase and hippocampal cell death in rats with transient forebrain ischemia. Eur. J. Pharmacol. 2006;535:114–124. [PubMed]
Noshita N, Sugawara T, Hayashi T, Lewen A, Omar G, Chan PH. Copper/zinc superoxide dismutase attenuates neuronal cell death by preventing extracellular signal-regulated kinase activation after transient focal cerebral ischemia in mice. J. Neurosci. 2002;22:7923–7930. [PubMed]
Raval AP, Dave KR, Prado R, Katz LM, Busto R, Sick TJ, Ginsberg MD, Mochly-Rosen D, Perez-Pinzon MA. Protein kinase C delta cleavage initiates an aberrant signal transduction pathway after cardiac arrest and oxygen glucose deprivation. J. Cereb. Blood Flow Metab. 2005;25:730–741. [PubMed]
Roux PP, Blenis J. ERK and p38 MAPK-activated protein kinases: a family of protein kinases with diverse biological functions. Microbiol. Mol. Biol. Rev. 2004;68:320–344. [PMC free article] [PubMed]
Saklatvala J, Davis W, Guesdon F. Interleukin 1 (IL1) and tumour necrosis factor (TNF) signal transduction. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 1996;351:151–157. [PubMed]
Sawe N, Steinberg G, Zhao H. Dual roles of the MAPK/ERK1/2 cell signaling pathway after stroke. J. Neurosci. Res. 2008 [PubMed]
Wang ZQ, Wu DC, Huang FP, Yang GY. Inhibition of MEK/ERK 1/2 pathway reduces pro-inflammatory cytokine interleukin-1 expression in focal cerebral ischemia. Brain Res. 2004;996:55–66. [PubMed]
Xu X, Chua CC, Gao J, Hamdy RC, Chua BH. Humanin is a novel neuroprotective agent against stroke. Stroke. 2006;37:2613–2619. [PubMed]
Zablocka B, Dluzniewska J, Zajac H, Domanska-Janik K. Opposite reaction of ERK and JNK in ischemia vulnerable and resistant regions of hippocampus: involvement of mitochondria. Brain Res. Mol. Brain Res. 2003;110:245–252. [PubMed]
Zhao H, Sapolsky RM, Steinberg GK. Interrupting reperfusion as a stroke therapy: ischemic postconditioning reduces infarct size after focal ischemia in rats. J. Cereb. Blood Flow Metab. 2006;26:1114–1121. [PubMed]
Zhao H, Shimohata T, Wang JQ, Sun G, Schaal DW, Sapolsky RM, Steinberg GK. Akt contributes to neuroprotection by hypothermia against cerebral ischemia in rats. J. Neurosci. 2005;25:9794–9806. [PubMed]
Zhao H, Yenari MA, Cheng D, Sapolsky RM, Steinberg GK. Biphasic cytochrome c release after transient global ischemia and its inhibition by hypothermia. J. Cereb. Blood Flow Metab. 2005;25:1119–1129. [PubMed]
Zhao H, Yenari MA, Sapolsky RM, Steinberg GK. Mild postischemic hypothermia prolongs the time window for gene therapy by inhibiting cytochrome C release. Stroke. 2004;35:572–577. [PubMed]